Skip to content

GitLab

Unverified Commit 41cdd380 authored by Chao Liu's avatar Chao Liu Committed by GitHub
Browse files

GEMM/Conv+BiasAdd+ReLU+Add (#55) 

* gemm+activation

* move C pointwise operation into threadwise copy

* add pointwise operation to A/B matrix

* update ckProfiler

* adding bias add

* adding bias add

* adding bias add

* added bias add; worked around compiler issues

* clean up

* clean up

* Update README.md

* Update README.md

* Update README.md

* clean up

* add conv_xdl example

* adding conv_xdl_bias_relu_add example

* add conv+bias+relu+add, but has register spill issue

* tweak

* tweak

* refactor

* Update README.md

update readme for example/2_gemm_xdl_bias_relu_add

* clean up

* Update README.md

update readme for example/3_conv_xdl

* Update README.md
parent d7a0a3f9
Expand all Show whitespace changes
Inline Side-by-side
Showing with 1540 additions and 187 deletions
composable_kernel/include/tensor_operation/blockwise_tensor_slice_transfer.hpp
View file @ 41cdd380
......@@ -14,6 +14,7 @@ namespace ck {
// 2. ThreadwiseTensorSliceTransfer_v3 does not keep reference to tensor descriptor
// 3. ThreadwiseTensorSliceTransfer_v3::Run() does not construct new tensor coordinate
template <index_t BlockSize,
typename SrcElementwiseOperation,
InMemoryDataOperationEnum_t DstInMemOp,
typename BlockSliceLengths,
typename ThreadSliceLengths,
......@@ -39,12 +40,17 @@ struct BlockwiseTensorSliceTransfer_v4
using Index = MultiIndex<nDim>;
__device__ constexpr BlockwiseTensorSliceTransfer_v4(const SrcDesc& src_desc,
__device__ constexpr BlockwiseTensorSliceTransfer_v4(
const SrcDesc& src_desc,
const Index& src_block_slice_origin,
const DstDesc& dst_desc,
const Index& dst_block_slice_origin)
: threadwise_transfer_(
src_desc, make_zero_multi_index<nDim>(), dst_desc, make_zero_multi_index<nDim>())
const Index& dst_block_slice_origin,
const SrcElementwiseOperation& src_element_op)
: threadwise_transfer_(src_desc,
make_zero_multi_index<nDim>(),
dst_desc,
make_zero_multi_index<nDim>(),
src_element_op)
{
static_assert(nDim == remove_reference_t<remove_cv_t<SrcDesc>>::GetNumOfDimension() &&
......@@ -147,6 +153,7 @@ struct BlockwiseTensorSliceTransfer_v4
using ThreadwiseTransfer =
ThreadwiseTensorSliceTransfer_v3r2<ThreadSliceLengths,
SrcElementwiseOperation,
DstInMemOp,
SrcData,
DstData,
......
composable_kernel/include/tensor_operation/gridwise_gemm_xdlops_v2r3.hpp
View file @ 41cdd380
......@@ -19,6 +19,9 @@ template <typename GridwiseGemm,
typename AGridDesc_K0_M_K1,
typename BGridDesc_K0_N_K1,
typename CGridDesc_M0_N0_M1_N1_M2_M3_M4_N2,
typename AElementwiseOperation,
typename BElementwiseOperation,
typename CElementwiseOperation,
typename Block2CTileMap,
bool HasMainKBlockLoop>
__global__ void
......@@ -32,6 +35,9 @@ __global__ void
const AGridDesc_K0_M_K1 a_grid_desc_k0_m_k1,
const BGridDesc_K0_N_K1 b_grid_desc_k0_n_k1,
const CGridDesc_M0_N0_M1_N1_M2_M3_M4_N2 c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2,
const AElementwiseOperation a_element_op,
const BElementwiseOperation b_element_op,
const CElementwiseOperation c_element_op,
const Block2CTileMap block_2_ctile_map)
{
constexpr index_t shared_block_size =
......@@ -46,6 +52,9 @@ __global__ void
a_grid_desc_k0_m_k1,
b_grid_desc_k0_n_k1,
c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2,
a_element_op,
b_element_op,
c_element_op,
block_2_ctile_map);
}
#elif CK_EXPERIMENTAL_PASS_TENSOR_DESCRIPTOR_BY_VOID_POINTER
......@@ -55,6 +64,9 @@ template <typename GridwiseGemm,
typename AGridDesc_K0_M_K1,
typename BGridDesc_K0_N_K1,
typename CGridDesc_M0_N0_M1_N1_M2_M3_M4_N2,
typename AElementwiseOperation,
typename BElementwiseOperation,
typename CElementwiseOperation,
typename Block2CTileMap>
__global__ void
#if CK_USE_LAUNCH_BOUNDS
......@@ -66,6 +78,9 @@ __global__ void
const void CONSTANT* p_a_grid_desc_k0_m_k1,
const void CONSTANT* p_b_grid_desc_k0_n_k1,
const void CONSTANT* p_c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2,
const void CONSTANT* p_a_element_op,
const void CONSTANT* p_b_element_op,
const void CONSTANT* p_c_element_op,
const void CONSTANT* p_block_2_ctile_map)
{
constexpr index_t shared_block_size =
......@@ -80,6 +95,12 @@ __global__ void
cast_pointer_to_generic_address_space(p_c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2));
const auto block_2_ctile_map = *reinterpret_cast<const Block2CTileMap*>(
cast_pointer_to_generic_address_space(p_block_2_ctile_map));
const auto a_element_op = *reinterpret_cast<const AElementwiseOperation*>(
cast_pointer_to_generic_address_space(p_a_element_op));
const auto b_element_op = *reinterpret_cast<const BElementwiseOperation*>(
cast_pointer_to_generic_address_space(p_b_element_op));
const auto c_element_op = *reinterpret_cast<const CElementwiseOperation*>(
cast_pointer_to_generic_address_space(p_c_element_op));
__shared__ FloatAB p_shared_block[shared_block_size];
......@@ -90,6 +111,9 @@ __global__ void
a_grid_desc_k0_m_k1,
b_grid_desc_k0_n_k1,
c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2,
a_element_op,
b_element_op,
c_element_op,
block_2_ctile_map);
}
#endif
......@@ -102,6 +126,9 @@ template <index_t BlockSize,
typename AGridDesc_K0_M_K1,
typename BGridDesc_K0_N_K1,
typename CGridDesc_M_N,
typename AElementwiseOperation,
typename BElementwiseOperation,
typename CElementwiseOperation,
index_t MPerBlock,
index_t NPerBlock,
index_t K0PerBlock,
......@@ -353,6 +380,9 @@ struct GridwiseGemm_k0mk1_k0nk1_mn_xdlops_v2r3
const AGridDesc_K0_M_K1& a_grid_desc_k0_m_k1,
const BGridDesc_K0_N_K1& b_grid_desc_k0_n_k1,
const CGridDesc_M0_N0_M1_N1_M2_M3_M4_N2& c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2,
const AElementwiseOperation& a_element_op,
const BElementwiseOperation& b_element_op,
const CElementwiseOperation& c_element_op,
const Block2CTileMap& block_2_ctile_map)
{
const auto a_grid_buf = make_dynamic_buffer<AddressSpaceEnum_t::Global>(
......@@ -411,6 +441,7 @@ struct GridwiseGemm_k0mk1_k0nk1_mn_xdlops_v2r3
// A matrix blockwise copy
auto a_blockwise_copy =
BlockwiseTensorSliceTransfer_v4<BlockSize,
AElementwiseOperation,
InMemoryDataOperationEnum_t::Set,
Sequence<K0PerBlock, MPerBlock, K1>,
ABlockTransferThreadSliceLengths_K0_M_K1,
......@@ -432,11 +463,13 @@ struct GridwiseGemm_k0mk1_k0nk1_mn_xdlops_v2r3
true>(a_grid_desc_k0_m_k1,
make_multi_index(0, m_block_data_idx_on_grid, 0),
a_block_desc_k0_m_k1,
make_multi_index(0, 0, 0));
make_multi_index(0, 0, 0),
a_element_op);
// B matrix blockwise copy
auto b_blockwise_copy =
BlockwiseTensorSliceTransfer_v4<BlockSize,
BElementwiseOperation,
InMemoryDataOperationEnum_t::Set,
Sequence<K0PerBlock, NPerBlock, K1>,
BBlockTransferThreadSliceLengths_K0_N_K1,
......@@ -458,7 +491,8 @@ struct GridwiseGemm_k0mk1_k0nk1_mn_xdlops_v2r3
true>(b_grid_desc_k0_n_k1,
make_multi_index(0, n_block_data_idx_on_grid, 0),
b_block_desc_k0_n_k1,
make_multi_index(0, 0, 0));
make_multi_index(0, 0, 0),
b_element_op);
// GEMM definition
// c_mtx += transpose(a_mtx) * b_mtx
......@@ -611,6 +645,7 @@ struct GridwiseGemm_k0mk1_k0nk1_mn_xdlops_v2r3
FloatC,
decltype(c_thread_desc_m0_n0_m1_n1_m2_m3_m4_n2),
decltype(c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2),
CElementwiseOperation,
Sequence<M0, N0, I1, I1, M2, I1, M4, I1>,
CThreadTransferSrcDstAccessOrder,
CThreadTransferSrcDstVectorDim,
......@@ -618,7 +653,6 @@ struct GridwiseGemm_k0mk1_k0nk1_mn_xdlops_v2r3
CGlobalMemoryDataOperation,
1,
true>{
c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2,
make_multi_index(m_thread_data_on_grid_idx[I0],
n_thread_data_on_grid_idx[I0],
......@@ -627,7 +661,8 @@ struct GridwiseGemm_k0mk1_k0nk1_mn_xdlops_v2r3
m_thread_data_on_grid_idx[I2],
m_thread_data_on_grid_idx[I3],
m_thread_data_on_grid_idx[I4],
n_thread_data_on_grid_idx[I2])};
n_thread_data_on_grid_idx[I2]),
c_element_op};
c_thread_copy.Run(c_thread_desc_m0_n0_m1_n1_m2_m3_m4_n2,
make_tuple(I0, I0, I0, I0, I0, I0, I0, I0),
......
composable_kernel/include/tensor_operation/threadwise_tensor_slice_transfer.hpp
View file @ 41cdd380
......@@ -50,6 +50,7 @@ template <typename SrcData,
typename DstData,
typename SrcDesc,
typename DstDesc,
typename DstElementwiseOperation,
typename SliceLengths,
typename DimAccessOrder,
index_t DstVectorDim,
......@@ -68,9 +69,12 @@ struct ThreadwiseTensorSliceTransfer_v1r3
using DstCoordStep = decltype(make_tensor_coordinate_step(DstDesc{}, Index{}));
__device__ constexpr ThreadwiseTensorSliceTransfer_v1r3(const DstDesc& dst_desc,
const Index& dst_slice_origin_idx)
: dst_coord_(make_tensor_coordinate(dst_desc, dst_slice_origin_idx))
__device__ constexpr ThreadwiseTensorSliceTransfer_v1r3(
const DstDesc& dst_desc,
const Index& dst_slice_origin_idx,
const DstElementwiseOperation& dst_element_op)
: dst_coord_(make_tensor_coordinate(dst_desc, dst_slice_origin_idx)),
dst_element_op_{dst_element_op}
{
static_assert(SrcDesc::IsKnownAtCompileTime(),
"wrong! SrcDesc need to known at compile-time");
......@@ -195,8 +199,9 @@ struct ThreadwiseTensorSliceTransfer_v1r3
constexpr index_t src_offset = src_desc.CalculateOffset(
src_slice_origin_idx + dst_data_idx + i * dst_scalar_step_in_vector);
// apply element-wise operation and type convert
dst_vector.template AsType<DstData>()(i) =
type_convert<DstData>(src_buf[Number<src_offset>{}]);
type_convert<DstData>(dst_element_op_(src_buf[Number<src_offset>{}]));
});
const bool is_dst_valid =
......@@ -373,6 +378,7 @@ struct ThreadwiseTensorSliceTransfer_v1r3
private:
DstCoord dst_coord_;
const DstElementwiseOperation dst_element_op_;
}; // namespace ck
// Assume:
......
composable_kernel/include/tensor_operation/threadwise_tensor_slice_transfer_v3r2.hpp
View file @ 41cdd380
......@@ -46,6 +46,7 @@ struct lambda_scalar_per_access_for_src_and_dst
// 3. src_slice_origin and dst_slice_origin are not known at compile-time,
// 4. Use thread buffer
template <typename SliceLengths,
typename SrcElementwiseOperation,
InMemoryDataOperationEnum_t DstInMemOp,
typename SrcData,
typename DstData,
......@@ -76,12 +77,15 @@ struct ThreadwiseTensorSliceTransfer_v3r2
using SrcCoordStep = decltype(make_tensor_coordinate_step(SrcDesc{}, Index{}));
using DstCoordStep = decltype(make_tensor_coordinate_step(DstDesc{}, Index{}));
__device__ constexpr ThreadwiseTensorSliceTransfer_v3r2(const SrcDesc& src_desc,
__device__ constexpr ThreadwiseTensorSliceTransfer_v3r2(
const SrcDesc& src_desc,
const Index& src_slice_origin,
const DstDesc& dst_desc,
const Index& dst_slice_origin)
const Index& dst_slice_origin,
const SrcElementwiseOperation& src_element_op)
: src_coord_(make_tensor_coordinate(src_desc, src_slice_origin)),
dst_coord_(make_tensor_coordinate(dst_desc, dst_slice_origin))
dst_coord_(make_tensor_coordinate(dst_desc, dst_slice_origin)),
src_element_op_(src_element_op)
{
}
......@@ -191,12 +195,22 @@ struct ThreadwiseTensorSliceTransfer_v3r2
const bool is_src_valid =
coordinate_has_valid_offset_assuming_visible_index_is_valid(src_desc, src_coord_);
using src_vector_t = typename vector_type_maker_t<SrcData, SrcScalarPerVector>::type;
using src_vector_type = vector_type_maker_t<SrcData, SrcScalarPerVector>;
using src_vector_t = typename src_vector_type::type;
// copy data from src_buf to src_thread_scratch_
// copy data from src_buf into src_vector_container
auto src_vector_container = src_vector_type{
src_buf.template Get<src_vector_t>(src_coord_.GetOffset(), is_src_valid)};
// apply SrcElementwiseOperation on src_vector_container
static_for<0, SrcScalarPerVector, 1>{}([&](auto i) {
src_vector_container.template AsType<SrcData>()(i) =
src_element_op_(src_vector_container.template AsType<SrcData>()[i]);
});
// copy data from src_vector_container into src_thread_scratch_
src_thread_scratch_.template SetAsType<src_vector_t>(
src_data_idx_seq,
src_buf.template Get<src_vector_t>(src_coord_.GetOffset(), is_src_valid));
src_data_idx_seq, src_vector_container.template AsType<src_vector_t>()[I0]);
constexpr auto move_on_dim = [&]() constexpr
{
......@@ -796,6 +810,7 @@ struct ThreadwiseTensorSliceTransfer_v3r2
SrcCoord src_coord_;
DstCoord dst_coord_;
const SrcElementwiseOperation src_element_op_;
};
} // namespace ck
......
composable_kernel/include/utility/config.hpp
View file @ 41cdd380
......@@ -136,6 +136,11 @@
#define CK_WORKAROUND_SWDEV_XXXXXX_INT8_DS_WRITE_ISSUE 1
#endif
// workaround for register spill due to compiler issue, when casting type between fp32 and fp16
#ifndef CK_WORKAROUND_SWDEV_XXXXXX_THREAD_WISE_COPY_V1R4_TYPE_CONVERT_ISSUE
#define CK_WORKAROUND_SWDEV_XXXXXX_THREAD_WISE_COPY_V1R4_TYPE_CONVERT_ISSUE 1
#endif
namespace ck {
enum InMemoryDataOperationEnum_t
......
device_operation/device_conv_xdl_instance_f16_f16_f16_nhwc_kyxc_nhwk.cpp
View file @ 41cdd380
......@@ -2,6 +2,7 @@
#include "config.hpp"
#include "device_conv_fwd_xdl_nhwc_kyxc_nhwk.hpp"
#include "device_conv_instance.hpp"
#include "element_wise_operation.hpp"
namespace ck {
namespace tensor_operation {
......@@ -18,32 +19,34 @@ using NHWK = ck::tensor_layout::convolution::NHWK;
template <ck::index_t... Is>
using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// Compilation parameters for in[n, hi, wi, c] * wei[k, y, x, c] = out[n, ho, wo, k]
using device_conv_fwd_xdl_instances_f16_f16_f16_nhwc_kyxc_nhwk = std::tuple<
// clang-format off
//##############| NDim| InData| WeiData| OutData| AccData| In| Wei| Out| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##############| Spatial| Type| Type| Type| Type| Layout| Layout| Layout| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##############| | | | | | | | | | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##############| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 256, 256, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 256, 128, 256, 4, 8, 32, 32, 2, 4, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 128, 128, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 256, 128, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 128, 128, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 128, 64, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 64, 64, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 256, 128, 64, 4, 8, 32, 32, 2, 1, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 256, 64, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 128, 128, 32, 4, 8, 32, 32, 2, 1, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 1, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 128, 32, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 64, 64, 32, 4, 8, 32, 32, 2, 1, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, 64, 32, 64, 4, 8, 32, 32, 1, 2, S<1, 2, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>
//##############| NDim| InData| WeiData| OutData| AccData| In| Wei| Out| A| B| C| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##############| Spatial| Type| Type| Type| Type| Layout| Layout| Layout| Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##############| | | | | | | | | Operation| Operation| Operation| | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##############| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 256, 256, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 256, 128, 256, 4, 8, 32, 32, 2, 4, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 128, 128, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 256, 128, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 128, 128, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 128, 64, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 64, 64, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 256, 128, 64, 4, 8, 32, 32, 2, 1, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 256, 64, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 128, 128, 32, 4, 8, 32, 32, 2, 1, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 1, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 128, 32, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 64, 64, 32, 4, 8, 32, 32, 2, 1, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F16, F16, F16, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 64, 32, 64, 4, 8, 32, 32, 1, 2, S<1, 2, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>
// clang-format on
>;
template <>
void add_device_conv_fwd_instance<2, F16, F16, F16, NHWC, KYXC, NHWK>(
std::vector<DeviceConvFwdPtr>& device_conv_instances)
std::vector<DeviceConvFwdPtr<PassThrough, PassThrough, PassThrough>>& device_conv_instances)
{
using DeviceConvs = device_conv_fwd_xdl_instances_f16_f16_f16_nhwc_kyxc_nhwk;
......
device_operation/device_conv_xdl_instance_f32_f32_f32_nhwc_kyxc_nhwk.cpp
View file @ 41cdd380
......@@ -2,6 +2,7 @@
#include "config.hpp"
#include "device_conv_fwd_xdl_nhwc_kyxc_nhwk.hpp"
#include "device_conv_instance.hpp"
#include "element_wise_operation.hpp"
namespace ck {
namespace tensor_operation {
......@@ -18,32 +19,34 @@ using NHWK = ck::tensor_layout::convolution::NHWK;
template <ck::index_t... Is>
using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// Compilation parameters for in[n, hi, wi, c] * wei[k, y, x, c] = out[n, ho, wo, k]
using device_conv_fwd_xdl_instances_f32_f32_f32_nhwc_kyxc_nhwk = std::tuple<
// clang-format off
//##############| NDim| InData| WeiData| OutData| AccData| In| Wei| Out| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##############| Spatial| Type| Type| Type| Type| Layout| Layout| Layout| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##############| | | | | | | | | | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##############| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 256, 256, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 256, 128, 256, 4, 4, 32, 32, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 128, 128, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 256, 128, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 128, 128, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 128, 64, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 64, 64, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 256, 128, 64, 4, 4, 32, 32, 2, 1, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 256, 64, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 128, 128, 32, 4, 4, 32, 32, 2, 1, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 1, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 128, 32, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 64, 64, 32, 4, 4, 32, 32, 2, 1, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, 64, 32, 64, 4, 4, 32, 32, 1, 2, S<1, 2, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>
//##############| NDim| InData| WeiData| OutData| AccData| In| Wei| Out| A| B| C| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##############| Spatial| Type| Type| Type| Type| Layout| Layout| Layout| Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##############| | | | | | | | | Operation| Operation| Operation| | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##############| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 256, 256, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 256, 128, 256, 4, 4, 32, 32, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 128, 128, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 256, 128, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 128, 128, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 128, 64, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 64, 64, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 256, 128, 64, 4, 4, 32, 32, 2, 1, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 256, 64, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 128, 128, 32, 4, 4, 32, 32, 2, 1, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 1, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 128, 32, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 64, 64, 32, 4, 4, 32, 32, 2, 1, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceConvFwdXdl< 2, F32, F32, F32, F32, NHWC, KYXC, NHWK, PassThrough, PassThrough, PassThrough, 64, 32, 64, 4, 4, 32, 32, 1, 2, S<1, 2, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>
// clang-format on
>;
template <>
void add_device_conv_fwd_instance<2, F32, F32, F32, NHWC, KYXC, NHWK>(
std::vector<DeviceConvFwdPtr>& device_conv_instances)
std::vector<DeviceConvFwdPtr<PassThrough, PassThrough, PassThrough>>& device_conv_instances)
{
using DeviceConvs = device_conv_fwd_xdl_instances_f32_f32_f32_nhwc_kyxc_nhwk;
......
device_operation/device_gemm_xdl_instance_f16_f16_f16_km_kn_mn.cpp
View file @ 41cdd380
......@@ -2,6 +2,7 @@
#include "config.hpp"
#include "device_gemm_xdl.hpp"
#include "device_gemm_instance.hpp"
#include "element_wise_operation.hpp"
namespace ck {
namespace tensor_operation {
......@@ -17,27 +18,29 @@ using Col = ck::tensor_layout::gemm::ColumnMajor;
template <ck::index_t... Is>
using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// Compilation parameters for a[k, m] * b[k, n] = c[m, n]
using device_gemm_xdl_instance_f16_f16_f16_km_kn_mn = std::tuple<
// clang-format off
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, 256, 256, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, 256, 128, 256, 4, 8, 32, 32, 2, 4, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 4, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, 128, 128, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, 256, 128, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, 128, 128, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 2, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, 128, 64, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, 256, 128, 64, 4, 8, 32, 32, 2, 1, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 1, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, 256, 64, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| A| B| C| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | Operation| Operation| Operation| | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 256, 256, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 256, 4, 8, 32, 32, 2, 4, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 4, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 128, 128, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 128, 128, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 2, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 128, 64, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 64, 4, 8, 32, 32, 2, 1, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 1, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 256, 64, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>
// clang-format on
>;
template <>
void add_device_gemm_instance<F16, F16, F16, Col, Row, Row>(
std::vector<DeviceGemmPtr>& device_op_instances)
std::vector<DeviceGemmPtr<PassThrough, PassThrough, PassThrough>>& device_op_instances)
{
using DeviceGemms = device_gemm_instance::device_gemm_xdl_instance_f16_f16_f16_km_kn_mn;
......
device_operation/device_gemm_xdl_instance_f16_f16_f16_km_nk_mn.cpp
View file @ 41cdd380
......@@ -2,6 +2,7 @@
#include "config.hpp"
#include "device_gemm_xdl.hpp"
#include "device_gemm_instance.hpp"
#include "element_wise_operation.hpp"
namespace ck {
namespace tensor_operation {
......@@ -17,27 +18,29 @@ using Col = ck::tensor_layout::gemm::ColumnMajor;
template <ck::index_t... Is>
using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// Compilation parameters for a[k, m] * b[n, k] = c[m, n]
using device_gemm_xdl_instance_f16_f16_f16_km_nk_mn = std::tuple<
// clang-format off
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, 256, 256, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, 256, 128, 256, 4, 8, 32, 32, 2, 4, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, 128, 128, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, 256, 128, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, 128, 128, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, 128, 64, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, 256, 128, 64, 4, 8, 32, 32, 2, 1, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, 256, 64, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| A| B| C| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | Operation| Operation| Operation| | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 256, 256, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 256, 4, 8, 32, 32, 2, 4, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 128, 128, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 128, 128, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 128, 64, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 64, 4, 8, 32, 32, 2, 1, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 256, 64, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>
// clang-format on
>;
template <>
void add_device_gemm_instance<F16, F16, F16, Col, Col, Row>(
std::vector<DeviceGemmPtr>& device_op_instances)
std::vector<DeviceGemmPtr<PassThrough, PassThrough, PassThrough>>& device_op_instances)
{
using DeviceGemms = device_gemm_instance::device_gemm_xdl_instance_f16_f16_f16_km_nk_mn;
......
device_operation/device_gemm_xdl_instance_f16_f16_f16_mk_kn_mn.cpp
View file @ 41cdd380
......@@ -2,6 +2,7 @@
#include "config.hpp"
#include "device_gemm_xdl.hpp"
#include "device_gemm_instance.hpp"
#include "element_wise_operation.hpp"
namespace ck {
namespace tensor_operation {
......@@ -17,27 +18,29 @@ using Col = ck::tensor_layout::gemm::ColumnMajor;
template <ck::index_t... Is>
using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// Compilation parameters for a[m, k] * b[k, n] = c[m, n]
using device_gemm_xdl_instance_f16_f16_f16_mk_kn_mn = std::tuple<
// clang-format off
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, 256, 256, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, 256, 128, 256, 4, 8, 32, 32, 2, 4, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, 128, 128, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, 256, 128, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, 128, 128, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, 128, 64, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, 256, 128, 64, 4, 8, 32, 32, 2, 1, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 1, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, 256, 64, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| A| B| C| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | Operation| Operation| Operation| | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 256, 256, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 256, 4, 8, 32, 32, 2, 4, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 128, 128, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 128, 128, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 128, 64, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 64, 4, 8, 32, 32, 2, 1, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 1, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 256, 64, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 8, 7, 1, true, true>
// clang-format on
>;
template <>
void add_device_gemm_instance<F16, F16, F16, Row, Row, Row>(
std::vector<DeviceGemmPtr>& device_op_instances)
std::vector<DeviceGemmPtr<PassThrough, PassThrough, PassThrough>>& device_op_instances)
{
using DeviceGemms = device_gemm_instance::device_gemm_xdl_instance_f16_f16_f16_mk_kn_mn;
......
device_operation/device_gemm_xdl_instance_f16_f16_f16_mk_nk_mn.cpp
View file @ 41cdd380
......@@ -2,6 +2,7 @@
#include "config.hpp"
#include "device_gemm_xdl.hpp"
#include "device_gemm_instance.hpp"
#include "element_wise_operation.hpp"
namespace ck {
namespace tensor_operation {
......@@ -17,32 +18,34 @@ using Col = ck::tensor_layout::gemm::ColumnMajor;
template <ck::index_t... Is>
using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// Compilation parameters for a[m, k] * b[n, k] = c[m, n]
using device_gemm_xdl_instance_f16_f16_f16_mk_nk_mn = std::tuple<
// clang-format off
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 256, 256, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 256, 128, 256, 4, 8, 32, 32, 2, 4, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 128, 128, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 256, 128, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 128, 128, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 128, 64, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 64, 64, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 256, 128, 64, 4, 8, 32, 32, 2, 1, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 256, 64, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 128, 128, 32, 4, 8, 32, 32, 2, 1, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 1, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 128, 32, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 64, 64, 32, 4, 8, 32, 32, 2, 1, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, 64, 32, 64, 4, 8, 32, 32, 1, 2, S<1, 2, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| A| B| C| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | Operation| Operation| Operation| | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 256, 256, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 256, 4, 8, 32, 32, 2, 4, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 128, 128, 128, 4, 8, 32, 32, 4, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 128, 128, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 128, 64, 128, 4, 8, 32, 32, 2, 2, S<1, 2, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 64, 64, 64, 4, 8, 32, 32, 2, 2, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 64, 4, 8, 32, 32, 2, 1, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 256, 64, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 128, 128, 32, 4, 8, 32, 32, 2, 1, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 1, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 128, 32, 128, 4, 8, 32, 32, 1, 2, S<1, 1, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 64, 64, 32, 4, 8, 32, 32, 2, 1, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 2, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>,
DeviceGemmXdl< F16, F16, F16, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 64, 32, 64, 4, 8, 32, 32, 1, 2, S<1, 2, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, S<1, 4, 8>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 7, 1, true, true>
// clang-format on
>;
template <>
void add_device_gemm_instance<F16, F16, F16, Row, Col, Row>(
std::vector<DeviceGemmPtr>& device_op_instances)
std::vector<DeviceGemmPtr<PassThrough, PassThrough, PassThrough>>& device_op_instances)
{
using DeviceGemms = device_gemm_instance::device_gemm_xdl_instance_f16_f16_f16_mk_nk_mn;
......
device_operation/device_gemm_xdl_instance_f32_f32_f32_km_kn_mn.cpp
View file @ 41cdd380
......@@ -2,6 +2,7 @@
#include "config.hpp"
#include "device_gemm_xdl.hpp"
#include "device_gemm_instance.hpp"
#include "element_wise_operation.hpp"
namespace ck {
namespace tensor_operation {
......@@ -17,27 +18,29 @@ using Col = ck::tensor_layout::gemm::ColumnMajor;
template <ck::index_t... Is>
using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// Compilation parameters for a[k, m] * b[k, n] = c[m, n]
using device_gemm_xdl_instance_f32_f32_f32_km_kn_mn = std::tuple<
// clang-format off
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, 256, 256, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, 256, 128, 256, 4, 4, 32, 32, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 4, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, 128, 128, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, 256, 128, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, 128, 128, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 2, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, 128, 64, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, 256, 128, 64, 4, 4, 32, 32, 2, 1, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 1, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, 256, 64, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| A| B| C| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | Operation| Operation| Operation| | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 256, 256, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 256, 4, 4, 32, 32, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 4, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 128, 128, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 128, 128, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 2, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 128, 64, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 64, 4, 4, 32, 32, 2, 1, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 1, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Row, Row, PassThrough, PassThrough, PassThrough, 256, 64, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>
// clang-format on
>;
template <>
void add_device_gemm_instance<F32, F32, F32, Col, Row, Row>(
std::vector<DeviceGemmPtr>& device_op_instances)
std::vector<DeviceGemmPtr<PassThrough, PassThrough, PassThrough>>& device_op_instances)
{
using DeviceGemms = device_gemm_instance::device_gemm_xdl_instance_f32_f32_f32_km_kn_mn;
......
device_operation/device_gemm_xdl_instance_f32_f32_f32_km_nk_mn.cpp
View file @ 41cdd380
......@@ -2,6 +2,7 @@
#include "config.hpp"
#include "device_gemm_xdl.hpp"
#include "device_gemm_instance.hpp"
#include "element_wise_operation.hpp"
namespace ck {
namespace tensor_operation {
......@@ -17,27 +18,29 @@ using Col = ck::tensor_layout::gemm::ColumnMajor;
template <ck::index_t... Is>
using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// Compilation parameters for a[k, m] * b[n, k] = c[m, n]
using device_gemm_xdl_instance_f32_f32_f32_km_nk_mn = std::tuple<
// clang-format off
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, 256, 256, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, 256, 128, 256, 4, 4, 32, 32, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, 128, 128, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, 256, 128, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, 128, 128, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, 128, 64, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, 256, 128, 64, 4, 4, 32, 32, 2, 1, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, 256, 64, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| A| B| C| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | Operation| Operation| Operation| | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 256, 256, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 256, 4, 4, 32, 32, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 128, 128, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 128, 128, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 128, 64, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 64, 4, 4, 32, 32, 2, 1, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Col, Col, Row, PassThrough, PassThrough, PassThrough, 256, 64, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>
// clang-format on
>;
template <>
void add_device_gemm_instance<F32, F32, F32, Col, Col, Row>(
std::vector<DeviceGemmPtr>& device_op_instances)
std::vector<DeviceGemmPtr<PassThrough, PassThrough, PassThrough>>& device_op_instances)
{
using DeviceGemms = device_gemm_instance::device_gemm_xdl_instance_f32_f32_f32_km_nk_mn;
......
device_operation/device_gemm_xdl_instance_f32_f32_f32_mk_kn_mn.cpp
View file @ 41cdd380
......@@ -2,6 +2,7 @@
#include "config.hpp"
#include "device_gemm_xdl.hpp"
#include "device_gemm_instance.hpp"
#include "element_wise_operation.hpp"
namespace ck {
namespace tensor_operation {
......@@ -17,27 +18,29 @@ using Col = ck::tensor_layout::gemm::ColumnMajor;
template <ck::index_t... Is>
using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// Compilation parameters for a[m, k] * b[k, n] = c[m, n]
using device_gemm_xdl_instance_f32_f32_f32_mk_kn_mn = std::tuple<
// clang-format off
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, 256, 256, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, 256, 128, 256, 4, 4, 32, 32, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, 128, 128, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, 256, 128, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, 128, 128, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, 128, 64, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, 256, 128, 64, 4, 4, 32, 32, 2, 1, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 1, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, 256, 64, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| A| B| C| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | Operation| Operation| Operation| | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 256, 256, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 256, 4, 4, 32, 32, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 128, 128, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 128, 128, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 128, 64, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 256, 128, 64, 4, 4, 32, 32, 2, 1, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 1, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 1, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Row, Row, PassThrough, PassThrough, PassThrough, 256, 64, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>, 1, 2, 4, 7, 1, true, true>
// clang-format on
>;
template <>
void add_device_gemm_instance<F32, F32, F32, Row, Row, Row>(
std::vector<DeviceGemmPtr>& device_op_instances)
std::vector<DeviceGemmPtr<PassThrough, PassThrough, PassThrough>>& device_op_instances)
{
using DeviceGemms = device_gemm_instance::device_gemm_xdl_instance_f32_f32_f32_mk_kn_mn;
......
device_operation/device_gemm_xdl_instance_f32_f32_f32_mk_nk_mn.cpp
View file @ 41cdd380
......@@ -2,6 +2,7 @@
#include "config.hpp"
#include "device_gemm_xdl.hpp"
#include "device_gemm_instance.hpp"
#include "element_wise_operation.hpp"
namespace ck {
namespace tensor_operation {
......@@ -17,32 +18,34 @@ using Col = ck::tensor_layout::gemm::ColumnMajor;
template <ck::index_t... Is>
using S = ck::Sequence<Is...>;
using PassThrough = ck::tensor_operation::element_wise::PassThrough;
// Compilation parameters for a[m, k] * b[n, k] = c[m, n]
using device_gemm_xdl_instance_f32_f32_f32_mk_nk_mn = std::tuple<
// clang-format off
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 256, 256, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 256, 128, 256, 4, 4, 32, 32, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 128, 128, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 256, 128, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 128, 128, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 128, 64, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 64, 64, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 256, 128, 64, 4, 4, 32, 32, 2, 1, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 256, 64, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 128, 128, 32, 4, 4, 32, 32, 2, 1, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 1, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 128, 32, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 64, 64, 32, 4, 4, 32, 32, 2, 1, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, 64, 32, 64, 4, 4, 32, 32, 1, 2, S<1, 2, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>
//##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout| A| B| C| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| CThreadTransfer| CThreadTransfer| ABlockLds| BBlockLds|
//##########| Type| Type| Type| Type| | | | Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| ThreadSlice| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| SrcDstVectorDim| DstScalar| AddExtraM| AddExtraN|
//##########| | | | | | | | Operation| Operation| Operation| | | | | | | | Wave| Wave| Lengths_K0_N_K1| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| Lengths_K0_N_K1| Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerVector| | |
//##########| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 256, 256, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 256, 4, 4, 32, 32, 2, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 128, 128, 128, 4, 4, 32, 32, 4, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 128, 128, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 128, 64, 128, 4, 4, 32, 32, 2, 2, S<1, 2, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 64, 64, 64, 4, 4, 32, 32, 2, 2, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 256, 128, 64, 4, 4, 32, 32, 2, 1, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 256, 64, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 128, 128, 32, 4, 4, 32, 32, 2, 1, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 1, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 128, 32, 128, 4, 4, 32, 32, 1, 2, S<1, 1, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 64, 64, 32, 4, 4, 32, 32, 2, 1, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 2, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>,
DeviceGemmXdl< F32, F32, F32, F32, Row, Col, Row, PassThrough, PassThrough, PassThrough, 64, 32, 64, 4, 4, 32, 32, 1, 2, S<1, 2, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, S<1, 4, 4>, S<4, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 4, 4, 7, 1, true, true>
// clang-format on
>;
template <>
void add_device_gemm_instance<F32, F32, F32, Row, Col, Row>(
std::vector<DeviceGemmPtr>& device_op_instances)
std::vector<DeviceGemmPtr<PassThrough, PassThrough, PassThrough>>& device_op_instances)
{
using DeviceGemms = device_gemm_instance::device_gemm_xdl_instance_f32_f32_f32_mk_nk_mn;
......
device_operation/include/device_conv.hpp
View file @ 41cdd380
......@@ -8,6 +8,9 @@ namespace ck {
namespace tensor_operation {
namespace device {
template <typename InElementwiseOperation,
typename WeiElementwiseOperation,
typename OutElementwiseOperation>
struct DeviceConvFwd : public BaseOperator
{
virtual std::unique_ptr<BaseArgument>
......@@ -23,11 +26,17 @@ struct DeviceConvFwd : public BaseOperator
std::vector<ck::index_t> conv_filter_strides,
std::vector<ck::index_t> conv_filter_dilations,
std::vector<ck::index_t> input_left_pads,
std::vector<ck::index_t> input_right_pads) = 0;
std::vector<ck::index_t> input_right_pads,
InElementwiseOperation in_element_op,
WeiElementwiseOperation wei_element_op,
OutElementwiseOperation out_element_op) = 0;
virtual std::unique_ptr<BaseInvoker> MakeInvokerPointer() = 0;
};
template <typename InElementwiseOperation,
typename WeiElementwiseOperation,
typename OutElementwiseOperation>
struct DeviceConvBwd : public BaseOperator
{
virtual std::unique_ptr<BaseArgument>
......@@ -43,11 +52,17 @@ struct DeviceConvBwd : public BaseOperator
std::vector<ck::index_t> conv_filter_strides,
std::vector<ck::index_t> conv_filter_dilations,
std::vector<ck::index_t> input_left_pads,
std::vector<ck::index_t> input_right_pads) = 0;
std::vector<ck::index_t> input_right_pads,
InElementwiseOperation in_element_op,
WeiElementwiseOperation wei_element_op,
OutElementwiseOperation out_element_op) = 0;
virtual std::unique_ptr<BaseInvoker> MakeInvokerPointer() = 0;
};
template <typename InElementwiseOperation,
typename WeiElementwiseOperation,
typename OutElementwiseOperation>
struct DeviceConvWrw : public BaseOperator
{
virtual std::unique_ptr<BaseArgument>
......@@ -63,14 +78,31 @@ struct DeviceConvWrw : public BaseOperator
std::vector<ck::index_t> conv_filter_strides,
std::vector<ck::index_t> conv_filter_dilations,
std::vector<ck::index_t> input_left_pads,
std::vector<ck::index_t> input_right_pads) = 0;
std::vector<ck::index_t> input_right_pads,
InElementwiseOperation in_element_op,
WeiElementwiseOperation wei_element_op,
OutElementwiseOperation out_element_op) = 0;
virtual std::unique_ptr<BaseInvoker> MakeInvokerPointer() = 0;
};
using DeviceConvFwdPtr = std::unique_ptr<DeviceConvFwd>;
using DeviceConvBwdPtr = std::unique_ptr<DeviceConvBwd>;
using DeviceConvWrwPtr = std::unique_ptr<DeviceConvWrw>;
template <typename InElementwiseOperation,
typename WeiElementwiseOperation,
typename OutElementwiseOperation>
using DeviceConvFwdPtr = std::unique_ptr<
DeviceConvFwd<InElementwiseOperation, WeiElementwiseOperation, OutElementwiseOperation>>;
template <typename InElementwiseOperation,
typename WeiElementwiseOperation,
typename OutElementwiseOperation>
using DeviceConvBwdPtr = std::unique_ptr<
DeviceConvBwd<InElementwiseOperation, WeiElementwiseOperation, OutElementwiseOperation>>;
template <typename InElementwiseOperation,
typename WeiElementwiseOperation,
typename OutElementwiseOperation>
using DeviceConvWrwPtr = std::unique_ptr<
DeviceConvWrw<InElementwiseOperation, WeiElementwiseOperation, OutElementwiseOperation>>;
} // namespace device
} // namespace tensor_operation
......
device_operation/include/device_conv_fwd_xdl.hpp
View file @ 41cdd380
......@@ -23,6 +23,9 @@ template <ck::index_t NDimSpatial,
typename InLayout,
typename WeiLayout,
typename OutLayout,
typename InElementwiseOperation,
typename WeiElementwiseOperation,
typename OutElementwiseOperation,
ck::index_t BlockSize,
ck::index_t MPerBlock,
ck::index_t NPerBlock,
......
device_operation/include/device_conv_fwd_xdl_nhwc_kyxc_nhwk.hpp
View file @ 41cdd380
......@@ -22,6 +22,9 @@ template <typename InDataType,
typename WeiDataType,
typename OutDataType,
typename AccDataType,
typename InElementwiseOperation,
typename WeiElementwiseOperation,
typename OutElementwiseOperation,
ck::index_t BlockSize,
ck::index_t MPerBlock,
ck::index_t NPerBlock,
......@@ -58,6 +61,9 @@ struct DeviceConvFwdXdl<
ck::tensor_layout::convolution::NHWC, // typename InLayout,
ck::tensor_layout::convolution::KYXC, // typename WeiLayout,
ck::tensor_layout::convolution::NHWK, // typename OutLayout,
InElementwiseOperation, // typename InElementwiseOperation,
WeiElementwiseOperation, // typename WeiElementwiseOperation,
OutElementwiseOperation, // typename OutElementwiseOperation,
BlockSize, // ck::index_t BlockSize,
MPerBlock, // ck::index_t MPerBlock,
NPerBlock, // ck::index_t NPerBlock,
......@@ -87,7 +93,8 @@ struct DeviceConvFwdXdl<
CThreadTransferDstScalarPerVector, // ck::index_t CThreadTransferDstScalarPerVector,
ABlockLdsAddExtraM, // bool ABlockLdsAddExtraM,
BBlockLdsAddExtraN // bool BBlockLdsAddExtraN>
> : public DeviceConvFwd
>
: public DeviceConvFwd<InElementwiseOperation, WeiElementwiseOperation, OutElementwiseOperation>
{
using ADataType = InDataType;
using BDataType = WeiDataType;
......@@ -293,6 +300,9 @@ struct DeviceConvFwdXdl<
AGridDesc_K0_M_K1,
BGridDesc_K0_N_K1,
CGridDesc_M_N,
InElementwiseOperation,
WeiElementwiseOperation,
OutElementwiseOperation,
MPerBlock,
NPerBlock,
K0PerBlock,
......@@ -351,7 +361,10 @@ struct DeviceConvFwdXdl<
std::vector<ck::index_t> input_left_pads,
std::vector<ck::index_t> input_right_pads,
ck::index_t M01,
ck::index_t N01)
ck::index_t N01,
InElementwiseOperation in_element_op,
WeiElementwiseOperation wei_element_op,
OutElementwiseOperation out_element_op)
: p_a_grid_{p_in_grid},
p_b_grid_{p_wei_grid},
p_c_grid_{p_out_grid},
......@@ -361,7 +374,10 @@ struct DeviceConvFwdXdl<
c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2_{},
block_2_ctile_map_{},
M01_{M01},
N01_{N01}
N01_{N01},
in_element_op_{in_element_op},
wei_element_op_{wei_element_op},
out_element_op_{out_element_op}
{
const auto descs = DeviceConvFwdXdl::MakeABCGridDescriptor_A_K0_M_K1_B_K0_N_K1_C_M_N(
N,
......@@ -400,6 +416,9 @@ struct DeviceConvFwdXdl<
Block2CTileMap block_2_ctile_map_;
index_t M01_;
index_t N01_;
InElementwiseOperation in_element_op_;
WeiElementwiseOperation wei_element_op_;
OutElementwiseOperation out_element_op_;
};
// Invoker
......@@ -449,6 +468,9 @@ struct DeviceConvFwdXdl<
remove_reference_t<DeviceConvFwdXdl::AGridDesc_K0_M_K1>,
remove_reference_t<DeviceConvFwdXdl::BGridDesc_K0_N_K1>,
remove_reference_t<DeviceConvFwdXdl::CGridDesc_M0_N0_M1_N1_M2_M3_M4_N2>,
InElementwiseOperation,
WeiElementwiseOperation,
OutElementwiseOperation,
remove_reference_t<DeviceConvFwdXdl::Block2CTileMap>,
true>;
......@@ -463,6 +485,9 @@ struct DeviceConvFwdXdl<
arg.a_grid_desc_k0_m_k1_,
arg.b_grid_desc_k0_n_k1_,
arg.c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2_,
arg.in_element_op_,
arg.wei_element_op_,
arg.out_element_op_,
arg.block_2_ctile_map_);
}
else
......@@ -474,6 +499,9 @@ struct DeviceConvFwdXdl<
remove_reference_t<DeviceConvFwdXdl::AGridDesc_K0_M_K1>,
remove_reference_t<DeviceConvFwdXdl::BGridDesc_K0_N_K1>,
remove_reference_t<DeviceConvFwdXdl::CGridDesc_M0_N0_M1_N1_M2_M3_M4_N2>,
InElementwiseOperation,
WeiElementwiseOperation,
OutElementwiseOperation,
remove_reference_t<DeviceConvFwdXdl::Block2CTileMap>,
false>;
......@@ -488,6 +516,9 @@ struct DeviceConvFwdXdl<
arg.a_grid_desc_k0_m_k1_,
arg.b_grid_desc_k0_n_k1_,
arg.c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2_,
arg.in_element_op_,
arg.wei_element_op_,
arg.out_element_op_,
arg.block_2_ctile_map_);
}
......@@ -534,7 +565,10 @@ struct DeviceConvFwdXdl<
std::vector<ck::index_t> conv_filter_strides,
std::vector<ck::index_t> conv_filter_dilations,
std::vector<ck::index_t> input_left_pads,
std::vector<ck::index_t> input_right_pads)
std::vector<ck::index_t> input_right_pads,
InElementwiseOperation in_element_op,
WeiElementwiseOperation wei_element_op,
OutElementwiseOperation out_element_op)
{
return Argument{p_in_grid,
p_wei_grid,
......@@ -550,7 +584,10 @@ struct DeviceConvFwdXdl<
input_left_pads,
input_right_pads,
1,
1};
1,
in_element_op,
wei_element_op,
out_element_op};
}
static auto MakeInvoker() { return Invoker{}; }
......@@ -569,7 +606,10 @@ struct DeviceConvFwdXdl<
std::vector<ck::index_t> conv_filter_strides,
std::vector<ck::index_t> conv_filter_dilations,
std::vector<ck::index_t> input_left_pads,
std::vector<ck::index_t> input_right_pads) override
std::vector<ck::index_t> input_right_pads,
InElementwiseOperation in_element_op,
WeiElementwiseOperation wei_element_op,
OutElementwiseOperation out_element_op) override
{
return std::make_unique<Argument>(static_cast<const InDataType*>(p_in_grid),
static_cast<const WeiDataType*>(p_wei_grid),
......@@ -585,7 +625,10 @@ struct DeviceConvFwdXdl<
input_left_pads,
input_right_pads,
1,
1);
1,
in_element_op,
wei_element_op,
out_element_op);
}
// polymorphic
......@@ -593,7 +636,7 @@ struct DeviceConvFwdXdl<
{
return std::make_unique<Invoker>(Invoker{});
}
};
}; // namespace device
} // namespace device
} // namespace tensor_operation
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment