mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-04-20 06:49:15 +00:00
Implement grouped gemm tile loop for RDNA4 (#3304)
* feat: grouped gemm tile loop support for RDNA4 * fix: removed extra parameter from grouped gemm example instance * fix: FP8 check incorrectly enabling FP8 on RDNA3
This commit is contained in:
Erwin Terpstra
@@ -6,20 +6,9 @@
|
||||
#include <iomanip>
|
||||
|
||||
#include "ck/ck.hpp"
|
||||
#include "ck/utility/env.hpp"
|
||||
#include "ck/host_utility/hip_check_error.hpp"
|
||||
#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
|
||||
#include "ck/tensor_operation/gpu/device/device_grouped_gemm_tile_loop.hpp"
|
||||
#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
|
||||
|
||||
#include "ck/library/tensor_operation_instance/gpu/grouped_gemm_tile_loop_multiply.hpp"
|
||||
|
||||
#include "ck/library/utility/check_err.hpp"
|
||||
#include "ck/library/utility/device_memory.hpp"
|
||||
#include "ck/library/utility/host_tensor.hpp"
|
||||
#include "ck/library/utility/literals.hpp"
|
||||
#include "ck/library/utility/fill.hpp"
|
||||
#include "ck/library/reference_tensor_operation/cpu/reference_gemm.hpp"
|
||||
#include "profile_grouped_gemm_tile_loop_generic_impl.hpp"
|
||||
|
||||
namespace ck {
|
||||
namespace profiler {
|
||||
@@ -47,300 +36,36 @@ bool profile_grouped_gemm_multiply_tile_loop_impl(int do_verification,
|
||||
int n_warmup = 10,
|
||||
int n_iter = 50)
|
||||
{
|
||||
using CDataType = EDataType;
|
||||
bool pass = true;
|
||||
|
||||
auto f_host_tensor_descriptor =
|
||||
[](std::size_t row, std::size_t col, std::size_t stride, auto layout) {
|
||||
using namespace ck::literals;
|
||||
|
||||
if(is_same<decltype(layout), tensor_layout::gemm::RowMajor>::value)
|
||||
{
|
||||
return HostTensorDescriptor({row, col}, {stride, 1_uz});
|
||||
}
|
||||
else
|
||||
{
|
||||
return HostTensorDescriptor({row, col}, {1_uz, stride});
|
||||
}
|
||||
};
|
||||
|
||||
std::size_t group_count = Ms.size();
|
||||
|
||||
if(!(group_count == Ns.size() && group_count == Ks.size() && group_count == StrideAs.size() &&
|
||||
group_count == StrideBs.size() && group_count == StrideEs.size()))
|
||||
std::vector<std::array<int, 1>> stride_ds;
|
||||
for(size_t i = 0; i < StrideDs.size(); ++i)
|
||||
{
|
||||
throw std::runtime_error("wrong! inconsistent M/N/Ks, StrideA/B/Cs size\n");
|
||||
stride_ds.emplace_back(std::array<int, 1>{StrideDs[i]});
|
||||
}
|
||||
|
||||
std::vector<Tensor<ADataType>> a_m_k;
|
||||
std::vector<Tensor<BDataType>> b_k_n;
|
||||
std::vector<Tensor<DDataType>> d_m_n;
|
||||
std::vector<Tensor<CDataType>> e_m_n_host_results;
|
||||
std::vector<Tensor<CDataType>> e_m_n_device_results;
|
||||
|
||||
for(std::size_t i = 0; i < group_count; i++)
|
||||
{
|
||||
a_m_k.push_back(
|
||||
Tensor<ADataType>(f_host_tensor_descriptor(Ms[i], Ks[i], StrideAs[i], ALayout{})));
|
||||
b_k_n.push_back(
|
||||
Tensor<BDataType>(f_host_tensor_descriptor(Ks[i], Ns[i], StrideBs[i], BLayout{})));
|
||||
d_m_n.push_back(
|
||||
Tensor<DDataType>(f_host_tensor_descriptor(Ms[i], Ns[i], StrideDs[i], DLayout{})));
|
||||
e_m_n_device_results.push_back(
|
||||
Tensor<CDataType>(f_host_tensor_descriptor(Ms[i], Ns[i], StrideEs[i], ELayout{})));
|
||||
e_m_n_host_results.push_back(
|
||||
Tensor<CDataType>(f_host_tensor_descriptor(Ms[i], Ns[i], StrideEs[i], ELayout{})));
|
||||
if(ck::EnvIsEnabled(CK_ENV(CK_LOGGING)))
|
||||
{
|
||||
std::cout << "group: " << i << " a_m_k[" << i << "]:" << a_m_k[i].mDesc << ", b_k_n["
|
||||
<< i << "]:" << b_k_n[i].mDesc << ", e_m_n_device_results[" << i
|
||||
<< "]:" << e_m_n_device_results[i].mDesc << std::endl;
|
||||
}
|
||||
switch(init_method)
|
||||
{
|
||||
case 0: break;
|
||||
case 1:
|
||||
ck::utils::FillUniformDistributionIntegerValue<ADataType>{-5, 5}(a_m_k[i]);
|
||||
ck::utils::FillUniformDistributionIntegerValue<BDataType>{-5, 5}(b_k_n[i]);
|
||||
ck::utils::FillUniformDistributionIntegerValue<DDataType>{-5, 5}(d_m_n[i]);
|
||||
break;
|
||||
case 2:
|
||||
ck::utils::FillUniformDistribution<ADataType>{.0, 1.}(a_m_k[i]);
|
||||
ck::utils::FillUniformDistribution<BDataType>{-0.5, 0.5}(b_k_n[i]);
|
||||
ck::utils::FillUniformDistribution<DDataType>{-0.5, 0.5}(d_m_n[i]);
|
||||
break;
|
||||
default:
|
||||
ck::utils::FillConstant<ADataType>{1}(a_m_k[i]);
|
||||
ck::utils::FillConstant<BDataType>{1}(b_k_n[i]);
|
||||
ck::utils::FillConstant<DDataType>{1}(d_m_n[i]);
|
||||
}
|
||||
}
|
||||
|
||||
using AElementOp = ck::tensor_operation::element_wise::PassThrough;
|
||||
using BElementOp = ck::tensor_operation::element_wise::PassThrough;
|
||||
using CElementOp = ck::tensor_operation::element_wise::PassThrough;
|
||||
using CDEElementOp = ck::tensor_operation::element_wise::Multiply;
|
||||
|
||||
const auto a_element_op = AElementOp{};
|
||||
const auto b_element_op = BElementOp{};
|
||||
const auto c_element_op = CElementOp{};
|
||||
const auto cde_element_op = CDEElementOp{};
|
||||
|
||||
using DeviceMemPtr = std::unique_ptr<DeviceMem>;
|
||||
std::vector<DeviceMemPtr> a_device_buf, b_device_buf, d_device_buf, e_device_buf;
|
||||
|
||||
a_device_buf.reserve(group_count);
|
||||
b_device_buf.reserve(group_count);
|
||||
d_device_buf.reserve(group_count);
|
||||
e_device_buf.reserve(group_count);
|
||||
|
||||
std::vector<const void*> p_a, p_b, p_d;
|
||||
constexpr ck::index_t NumDTensor = 1;
|
||||
auto p_ds = std::vector<std::array<const void*, NumDTensor>>{};
|
||||
std::vector<void*> p_e;
|
||||
|
||||
p_a.reserve(group_count);
|
||||
p_b.reserve(group_count);
|
||||
p_ds.reserve(group_count);
|
||||
p_e.reserve(group_count);
|
||||
|
||||
using KernelArguments = ck::tensor_operation::device::GroupedGemmKernelArgument<NumDTensor>;
|
||||
|
||||
std::vector<ck::tensor_operation::device::GemmDesc> gemm_descs;
|
||||
std::vector<KernelArguments> gemm_kargs;
|
||||
|
||||
gemm_descs.reserve(group_count);
|
||||
gemm_kargs.reserve(group_count);
|
||||
|
||||
for(std::size_t i = 0; i < group_count; i++)
|
||||
{
|
||||
a_device_buf.emplace_back(
|
||||
std::make_unique<DeviceMem>(sizeof(ADataType) * a_m_k[i].mDesc.GetElementSpaceSize()));
|
||||
b_device_buf.emplace_back(
|
||||
std::make_unique<DeviceMem>(sizeof(BDataType) * b_k_n[i].mDesc.GetElementSpaceSize()));
|
||||
d_device_buf.emplace_back(
|
||||
std::make_unique<DeviceMem>(sizeof(DDataType) * d_m_n[i].mDesc.GetElementSpaceSize()));
|
||||
e_device_buf.emplace_back(std::make_unique<DeviceMem>(
|
||||
sizeof(CDataType) * e_m_n_device_results[i].mDesc.GetElementSpaceSize()));
|
||||
|
||||
a_device_buf[i]->ToDevice(a_m_k[i].mData.data());
|
||||
b_device_buf[i]->ToDevice(b_k_n[i].mData.data());
|
||||
d_device_buf[i]->ToDevice(d_m_n[i].mData.data());
|
||||
e_device_buf[i]->SetZero();
|
||||
|
||||
p_a.push_back(a_device_buf[i]->GetDeviceBuffer());
|
||||
p_b.push_back(b_device_buf[i]->GetDeviceBuffer());
|
||||
p_ds.push_back({d_device_buf[i]->GetDeviceBuffer()});
|
||||
p_e.push_back(e_device_buf[i]->GetDeviceBuffer());
|
||||
|
||||
gemm_descs.push_back(
|
||||
{0, Ns[i], Ks[i], StrideAs[i], StrideBs[i], StrideEs[i], {StrideDs[i]}});
|
||||
gemm_kargs.push_back({a_device_buf[i]->GetDeviceBuffer(),
|
||||
b_device_buf[i]->GetDeviceBuffer(),
|
||||
{d_device_buf[i]->GetDeviceBuffer()},
|
||||
e_device_buf[i]->GetDeviceBuffer(),
|
||||
Ms[i],
|
||||
Ns[i],
|
||||
Ks[i],
|
||||
StrideAs[i],
|
||||
StrideBs[i],
|
||||
{StrideDs[i]},
|
||||
StrideEs[i]});
|
||||
}
|
||||
|
||||
using DeviceOp = ck::tensor_operation::device::DeviceGroupedGemmTileLoop<ALayout,
|
||||
BLayout,
|
||||
ck::Tuple<DLayout>,
|
||||
ELayout,
|
||||
ADataType,
|
||||
BDataType,
|
||||
ck::Tuple<DDataType>,
|
||||
EDataType,
|
||||
AElementOp,
|
||||
BElementOp,
|
||||
CDEElementOp>;
|
||||
|
||||
const auto op_ptrs = ck::tensor_operation::device::instance::DeviceOperationInstanceFactory<
|
||||
DeviceOp>::GetInstances();
|
||||
|
||||
if(op_ptrs.size() <= 0)
|
||||
{
|
||||
throw std::runtime_error("wrong! no device GEMM instance found");
|
||||
}
|
||||
|
||||
std::string best_gemm_name;
|
||||
float best_ave_time = 0;
|
||||
float best_tflops = 0;
|
||||
float best_gb_per_sec = 0;
|
||||
|
||||
if(do_verification)
|
||||
{
|
||||
for(std::size_t i = 0; i < gemm_descs.size(); i++)
|
||||
{
|
||||
Tensor<CDataType> c_m_n({Ms[i], Ns[i]});
|
||||
|
||||
using ReferenceGemmInstance = ck::tensor_operation::host::ReferenceGemm<ADataType,
|
||||
BDataType,
|
||||
CDataType,
|
||||
AccDataType,
|
||||
AElementOp,
|
||||
BElementOp,
|
||||
CElementOp>;
|
||||
|
||||
auto ref_gemm = ReferenceGemmInstance{};
|
||||
auto ref_invoker = ref_gemm.MakeInvoker();
|
||||
auto ref_argument = ref_gemm.MakeArgument(
|
||||
a_m_k[i], b_k_n[i], c_m_n, a_element_op, b_element_op, c_element_op);
|
||||
ref_invoker.Run(ref_argument);
|
||||
|
||||
for(int m = 0; m < Ms[i]; ++m)
|
||||
{
|
||||
for(int n = 0; n < Ns[i]; ++n)
|
||||
{
|
||||
cde_element_op(e_m_n_host_results[i](m, n), c_m_n(m, n), d_m_n[i](m, n));
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// profile device GEMM instances
|
||||
for(auto& gemm_ptr : op_ptrs)
|
||||
{
|
||||
auto argument_ptr =
|
||||
gemm_ptr->MakeArgumentPointer(p_a,
|
||||
p_b,
|
||||
p_ds,
|
||||
p_e,
|
||||
gemm_descs,
|
||||
ck::tensor_operation::element_wise::PassThrough{},
|
||||
ck::tensor_operation::element_wise::PassThrough{},
|
||||
cde_element_op);
|
||||
auto invoker_ptr = gemm_ptr->MakeInvokerPointer();
|
||||
std::string gemm_name = gemm_ptr->GetTypeString();
|
||||
|
||||
DeviceMem gemm_arg_dev_mem(gemm_ptr->GetDeviceKernelArgSize(argument_ptr.get()));
|
||||
hip_check_error(hipMemcpy(gemm_arg_dev_mem.GetDeviceBuffer(),
|
||||
gemm_kargs.data(),
|
||||
gemm_ptr->GetDeviceKernelArgSize(argument_ptr.get()),
|
||||
hipMemcpyHostToDevice));
|
||||
gemm_ptr->SetDeviceKernelArgs(argument_ptr.get(), gemm_arg_dev_mem.GetDeviceBuffer());
|
||||
|
||||
if(gemm_ptr->IsSupportedArgument(argument_ptr.get()))
|
||||
{
|
||||
invoker_ptr->Run(argument_ptr.get(), StreamConfig{nullptr, false, 0, n_warmup, n_iter});
|
||||
if(do_verification)
|
||||
{
|
||||
bool instance_pass = true;
|
||||
for(std::size_t i = 0; i < gemm_descs.size(); i++)
|
||||
{
|
||||
e_device_buf[i]->FromDevice(e_m_n_device_results[i].mData.data());
|
||||
instance_pass = instance_pass && ck::utils::check_err(e_m_n_device_results[i],
|
||||
e_m_n_host_results[i]);
|
||||
|
||||
if(do_log)
|
||||
{
|
||||
LogRangeAsType<float>(std::cout << "a : ", a_m_k[i].mData, ",")
|
||||
<< std::endl;
|
||||
LogRangeAsType<float>(std::cout << "b: ", b_k_n[i].mData, ",") << std::endl;
|
||||
LogRangeAsType<float>(
|
||||
std::cout << "e_device: ", e_m_n_device_results[i].mData, ",")
|
||||
<< std::endl;
|
||||
LogRangeAsType<float>(
|
||||
std::cout << "e_host : ", e_m_n_host_results[i].mData, ",")
|
||||
<< std::endl;
|
||||
}
|
||||
}
|
||||
|
||||
std::cout << "Instance: " << gemm_name << " verification "
|
||||
<< (instance_pass ? "SUCCEED" : "FAILED") << std::endl;
|
||||
|
||||
pass = pass && instance_pass;
|
||||
}
|
||||
|
||||
if(time_kernel)
|
||||
{
|
||||
float ave_time = invoker_ptr->Run(
|
||||
argument_ptr.get(), StreamConfig{nullptr, time_kernel, 0, n_warmup, n_iter});
|
||||
|
||||
std::size_t flop = 0, num_btype = 0;
|
||||
for(std::size_t i = 0; i < gemm_descs.size(); i++)
|
||||
{
|
||||
flop += std::size_t(2) * Ms[i] * Ns[i] * Ks[i];
|
||||
|
||||
num_btype += sizeof(ADataType) * Ms[i] * Ks[i] +
|
||||
sizeof(BDataType) * Ks[i] * Ns[i] +
|
||||
sizeof(EDataType) * Ms[i] * Ns[i] + // D matrix
|
||||
sizeof(EDataType) * Ms[i] * Ns[i];
|
||||
}
|
||||
|
||||
float tflops = static_cast<float>(flop) / 1.E9 / ave_time;
|
||||
float gb_per_sec = num_btype / 1.E6 / ave_time;
|
||||
std::cout << "Perf: " << std::setw(10) << ave_time << " ms, " << tflops
|
||||
<< " TFlops, " << gb_per_sec << " GB/s, " << gemm_name << std::endl;
|
||||
|
||||
if(tflops > best_tflops)
|
||||
{
|
||||
best_gemm_name = gemm_name;
|
||||
best_tflops = tflops;
|
||||
best_ave_time = ave_time;
|
||||
best_gb_per_sec = gb_per_sec;
|
||||
}
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
std::cout << "Instance: " << gemm_name << ", does not support this GEMM problem"
|
||||
<< std::endl;
|
||||
}
|
||||
}
|
||||
|
||||
if(time_kernel)
|
||||
{
|
||||
std::cout << "Best Perf: " << best_ave_time << " ms, " << best_tflops << " TFlops, "
|
||||
<< best_gb_per_sec << " GB/s, " << best_gemm_name << std::endl;
|
||||
}
|
||||
|
||||
return pass;
|
||||
return profile_grouped_gemm_tile_loop_generic_impl<
|
||||
ADataType,
|
||||
BDataType,
|
||||
Tuple<DDataType>,
|
||||
EDataType,
|
||||
ALayout,
|
||||
BLayout,
|
||||
Tuple<DLayout>,
|
||||
ELayout,
|
||||
PassThrough,
|
||||
PassThrough,
|
||||
ck::tensor_operation::element_wise::Multiply>(do_verification,
|
||||
init_method,
|
||||
do_log,
|
||||
time_kernel,
|
||||
Ms,
|
||||
Ns,
|
||||
Ks,
|
||||
StrideAs,
|
||||
StrideBs,
|
||||
stride_ds,
|
||||
StrideEs,
|
||||
n_warmup,
|
||||
n_iter);
|
||||
}
|
||||
|
||||
} // namespace profiler
|
||||
|
||||
@@ -0,0 +1,436 @@
|
||||
// Copyright (c) Advanced Micro Devices, Inc., or its affiliates.
|
||||
// SPDX-License-Identifier: MIT
|
||||
|
||||
#pragma once
|
||||
|
||||
#include <iomanip>
|
||||
#include <type_traits>
|
||||
|
||||
#include "ck/ck.hpp"
|
||||
#include "ck/utility/env.hpp"
|
||||
#include "ck/host_utility/hip_check_error.hpp"
|
||||
#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
|
||||
#include "ck/tensor_operation/gpu/device/device_grouped_gemm_tile_loop.hpp"
|
||||
#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
|
||||
|
||||
#include "ck/library/tensor_operation_instance/gpu/grouped_gemm_tile_loop.hpp"
|
||||
#include "ck/library/tensor_operation_instance/gpu/grouped_gemm_tile_loop_multiply.hpp"
|
||||
|
||||
#include "ck/library/utility/check_err.hpp"
|
||||
#include "ck/library/utility/device_memory.hpp"
|
||||
#include "ck/library/utility/host_tensor.hpp"
|
||||
#include "ck/library/utility/literals.hpp"
|
||||
#include "ck/library/utility/fill.hpp"
|
||||
#include "ck/library/utility/host_tensor_generator.hpp"
|
||||
#include "ck/library/reference_tensor_operation/cpu/reference_gemm.hpp"
|
||||
#include "ck/library/reference_tensor_operation/cpu/reference_gemm_multiple_d.hpp"
|
||||
#include "ck/utility/integral_constant.hpp"
|
||||
#include "ck/utility/tuple.hpp"
|
||||
#include "ck/utility/tuple_helper.hpp"
|
||||
|
||||
namespace ck {
|
||||
namespace profiler {
|
||||
|
||||
using PassThrough = ck::tensor_operation::element_wise::PassThrough;
|
||||
|
||||
template <class F, std::size_t... I>
|
||||
constexpr auto make_array_from_fn_impl(F&& f, std::index_sequence<I...>)
|
||||
{
|
||||
using T = std::decay_t<decltype(f(std::integral_constant<std::size_t, 0>{}))>;
|
||||
return std::array<T, sizeof...(I)>{f(std::integral_constant<std::size_t, I>{})...};
|
||||
}
|
||||
|
||||
template <std::size_t N, class F>
|
||||
constexpr auto make_array_from_fn(F&& f)
|
||||
{
|
||||
return make_array_from_fn_impl(std::forward<F>(f), std::make_index_sequence<N>{});
|
||||
}
|
||||
|
||||
template <typename ADataType,
|
||||
typename BDataType,
|
||||
typename DsDataType,
|
||||
typename EDataType,
|
||||
typename ALayout,
|
||||
typename BLayout,
|
||||
typename DsLayout,
|
||||
typename ELayout,
|
||||
typename AElementOp = PassThrough,
|
||||
typename BElementOp = PassThrough,
|
||||
typename CDEElementOp = PassThrough>
|
||||
bool profile_grouped_gemm_tile_loop_generic_impl(
|
||||
int do_verification,
|
||||
int init_method,
|
||||
bool do_log,
|
||||
bool time_kernel,
|
||||
const std::vector<int>& Ms,
|
||||
const std::vector<int>& Ns,
|
||||
const std::vector<int>& Ks,
|
||||
const std::vector<int>& StrideAs,
|
||||
const std::vector<int>& StrideBs,
|
||||
const std::vector<std::array<int, DsDataType::Size()>>& StrideDs,
|
||||
const std::vector<int>& StrideEs,
|
||||
int n_warmup = 10,
|
||||
int n_iter = 50)
|
||||
{
|
||||
using AccDataType = float;
|
||||
constexpr ck::index_t NumDTensor = DsDataType::Size();
|
||||
|
||||
static_assert(DsLayout::Size() == DsDataType::Size(), "wrong! inconsistent NumDTensor");
|
||||
|
||||
bool pass = true;
|
||||
|
||||
auto f_host_tensor_descriptor =
|
||||
[](std::size_t row, std::size_t col, std::size_t stride, auto layout) {
|
||||
using namespace ck::literals;
|
||||
|
||||
if(is_same<decltype(layout), tensor_layout::gemm::RowMajor>::value)
|
||||
{
|
||||
return HostTensorDescriptor({row, col}, {stride, 1_uz});
|
||||
}
|
||||
else
|
||||
{
|
||||
return HostTensorDescriptor({row, col}, {1_uz, stride});
|
||||
}
|
||||
};
|
||||
|
||||
std::size_t group_count = Ms.size();
|
||||
|
||||
if(!(group_count == Ns.size() && group_count == Ks.size() && group_count == StrideAs.size() &&
|
||||
group_count == StrideBs.size() &&
|
||||
((StrideDs.size() == 0 && NumDTensor == 0) || group_count == StrideDs.size()) &&
|
||||
group_count == StrideEs.size()))
|
||||
{
|
||||
throw std::runtime_error("wrong! inconsistent M/N/Ks, StrideA/B/D/Es size\n");
|
||||
}
|
||||
|
||||
std::vector<Tensor<ADataType>> a_m_k;
|
||||
std::vector<Tensor<BDataType>> b_k_n;
|
||||
std::vector<tuple_map_t<Tensor, DsDataType>> d_m_n;
|
||||
std::vector<Tensor<EDataType>> e_m_n_host_results;
|
||||
std::vector<Tensor<EDataType>> e_m_n_device_results;
|
||||
|
||||
for(std::size_t i = 0; i < group_count; i++)
|
||||
{
|
||||
a_m_k.push_back(
|
||||
Tensor<ADataType>(f_host_tensor_descriptor(Ms[i], Ks[i], StrideAs[i], ALayout{})));
|
||||
b_k_n.push_back(
|
||||
Tensor<BDataType>(f_host_tensor_descriptor(Ks[i], Ns[i], StrideBs[i], BLayout{})));
|
||||
|
||||
auto d_tensors = ck::generate_tuple(
|
||||
[&](auto j) {
|
||||
using DDataType = tuple_element_t<j, DsDataType>;
|
||||
|
||||
return Tensor<DDataType>(f_host_tensor_descriptor(
|
||||
Ms[i], Ns[i], StrideDs[i][j], tuple_element_t<j, DsLayout>{}));
|
||||
},
|
||||
Number<NumDTensor>{});
|
||||
d_m_n.emplace_back(d_tensors);
|
||||
|
||||
e_m_n_device_results.push_back(
|
||||
Tensor<EDataType>(f_host_tensor_descriptor(Ms[i], Ns[i], StrideEs[i], ELayout{})));
|
||||
e_m_n_host_results.push_back(
|
||||
Tensor<EDataType>(f_host_tensor_descriptor(Ms[i], Ns[i], StrideEs[i], ELayout{})));
|
||||
if(ck::EnvIsEnabled(CK_ENV(CK_LOGGING)))
|
||||
{
|
||||
std::cout << "group: " << i << " a_m_k[" << i << "]:" << a_m_k[i].mDesc << ", b_k_n["
|
||||
<< i << "]:" << b_k_n[i].mDesc << ", e_m_n_device_results[" << i
|
||||
<< "]:" << e_m_n_device_results[i].mDesc << std::endl;
|
||||
}
|
||||
switch(init_method)
|
||||
{
|
||||
case 0: break;
|
||||
case 1:
|
||||
a_m_k[i].GenerateTensorValue(GeneratorTensor_2<ADataType>{-5, 5});
|
||||
b_k_n[i].GenerateTensorValue(GeneratorTensor_2<BDataType>{-5, 5});
|
||||
static_for<0, NumDTensor, 1>{}([&](auto j) -> void {
|
||||
d_m_n[i](j).GenerateTensorValue(
|
||||
GeneratorTensor_2<tuple_element_t<j, DsDataType>>{-5, 5});
|
||||
});
|
||||
break;
|
||||
case 2:
|
||||
a_m_k[i].GenerateTensorValue(GeneratorTensor_3<ADataType>{0.0, 1.0});
|
||||
b_k_n[i].GenerateTensorValue(GeneratorTensor_3<BDataType>{-0.5, 0.5});
|
||||
static_for<0, NumDTensor, 1>{}([&](auto j) -> void {
|
||||
d_m_n[i](j).GenerateTensorValue(
|
||||
GeneratorTensor_3<tuple_element_t<j, DsDataType>>{-0.5, 0.5});
|
||||
});
|
||||
break;
|
||||
default:
|
||||
ck::utils::FillConstant<ADataType>{1}(a_m_k[i]);
|
||||
ck::utils::FillConstant<BDataType>{1}(b_k_n[i]);
|
||||
static_for<0, NumDTensor, 1>{}([&](auto j) -> void {
|
||||
ck::utils::FillConstant<tuple_element_t<j, DsDataType>>{1}(d_m_n[i](j));
|
||||
});
|
||||
}
|
||||
}
|
||||
const auto a_element_op = AElementOp{};
|
||||
const auto b_element_op = BElementOp{};
|
||||
const auto cde_element_op = CDEElementOp{};
|
||||
|
||||
using DeviceMemPtr = std::unique_ptr<DeviceMem>;
|
||||
std::vector<DeviceMemPtr> a_device_buf, b_device_buf, e_device_buf;
|
||||
std::vector<std::array<DeviceMemPtr, NumDTensor>> d_device_bufs;
|
||||
|
||||
a_device_buf.reserve(group_count);
|
||||
b_device_buf.reserve(group_count);
|
||||
d_device_bufs.reserve(group_count);
|
||||
e_device_buf.reserve(group_count);
|
||||
|
||||
std::vector<const void*> p_a, p_b;
|
||||
std::vector<std::array<const void*, NumDTensor>> p_ds;
|
||||
std::vector<void*> p_e;
|
||||
|
||||
p_a.reserve(group_count);
|
||||
p_b.reserve(group_count);
|
||||
p_ds.reserve(group_count);
|
||||
p_e.reserve(group_count);
|
||||
|
||||
using KernelArguments = ck::tensor_operation::device::GroupedGemmKernelArgument<NumDTensor>;
|
||||
|
||||
std::vector<ck::tensor_operation::device::GemmDesc> gemm_descs;
|
||||
std::vector<KernelArguments> gemm_kargs;
|
||||
|
||||
gemm_descs.reserve(group_count);
|
||||
gemm_kargs.reserve(group_count);
|
||||
|
||||
for(std::size_t i = 0; i < group_count; i++)
|
||||
{
|
||||
a_device_buf.emplace_back(
|
||||
std::make_unique<DeviceMem>(sizeof(ADataType) * a_m_k[i].mDesc.GetElementSpaceSize()));
|
||||
b_device_buf.emplace_back(
|
||||
std::make_unique<DeviceMem>(sizeof(BDataType) * b_k_n[i].mDesc.GetElementSpaceSize()));
|
||||
|
||||
if constexpr(NumDTensor > 0)
|
||||
{
|
||||
d_device_bufs.emplace_back(make_array_from_fn<NumDTensor>([&](auto j) {
|
||||
return std::make_unique<DeviceMem>(
|
||||
sizeof(tuple_element_t<j, DsDataType>) *
|
||||
d_m_n[i][ck::integral_constant<index_t, j>{}].mDesc.GetElementSpaceSize());
|
||||
}));
|
||||
}
|
||||
|
||||
e_device_buf.emplace_back(std::make_unique<DeviceMem>(
|
||||
sizeof(EDataType) * e_m_n_device_results[i].mDesc.GetElementSpaceSize()));
|
||||
|
||||
a_device_buf[i]->ToDevice(a_m_k[i].mData.data());
|
||||
b_device_buf[i]->ToDevice(b_k_n[i].mData.data());
|
||||
|
||||
static_for<0, NumDTensor, 1>{}(
|
||||
[&](auto j) -> void { d_device_bufs[i][j]->ToDevice(d_m_n[i][j].mData.data()); });
|
||||
|
||||
e_device_buf[i]->SetZero();
|
||||
|
||||
p_a.push_back(a_device_buf[i]->GetDeviceBuffer());
|
||||
p_b.push_back(b_device_buf[i]->GetDeviceBuffer());
|
||||
|
||||
std::array<const void*, NumDTensor> p_d;
|
||||
static_for<0, NumDTensor, 1>{}(
|
||||
[&](auto j) -> void { p_d[j] = d_device_bufs[i][j]->GetDeviceBuffer(); });
|
||||
|
||||
p_ds.push_back(p_d);
|
||||
|
||||
p_e.push_back(e_device_buf[i]->GetDeviceBuffer());
|
||||
|
||||
gemm_descs.push_back({Ms[i],
|
||||
Ns[i],
|
||||
Ks[i],
|
||||
StrideAs[i],
|
||||
StrideBs[i],
|
||||
StrideEs[i],
|
||||
std::vector<int>(StrideDs[i].begin(), StrideDs[i].end())});
|
||||
gemm_kargs.push_back({a_device_buf[i]->GetDeviceBuffer(),
|
||||
b_device_buf[i]->GetDeviceBuffer(),
|
||||
p_d,
|
||||
e_device_buf[i]->GetDeviceBuffer(),
|
||||
Ms[i],
|
||||
Ns[i],
|
||||
Ks[i],
|
||||
StrideAs[i],
|
||||
StrideBs[i],
|
||||
StrideDs[i],
|
||||
StrideEs[i]});
|
||||
}
|
||||
|
||||
using DeviceOp = ck::tensor_operation::device::DeviceGroupedGemmTileLoop<ALayout,
|
||||
BLayout,
|
||||
DsLayout,
|
||||
ELayout,
|
||||
ADataType,
|
||||
BDataType,
|
||||
DsDataType,
|
||||
EDataType,
|
||||
AElementOp,
|
||||
BElementOp,
|
||||
CDEElementOp>;
|
||||
|
||||
const auto op_ptrs = ck::tensor_operation::device::instance::DeviceOperationInstanceFactory<
|
||||
DeviceOp>::GetInstances();
|
||||
|
||||
if(op_ptrs.size() <= 0)
|
||||
{
|
||||
throw std::runtime_error("wrong! no device GEMM instance found");
|
||||
}
|
||||
|
||||
std::string best_gemm_name;
|
||||
float best_ave_time = 0;
|
||||
float best_tflops = 0;
|
||||
float best_gb_per_sec = 0;
|
||||
|
||||
if(do_verification)
|
||||
{
|
||||
for(std::size_t i = 0; i < gemm_descs.size(); i++)
|
||||
{
|
||||
if constexpr(NumDTensor > 0)
|
||||
{
|
||||
using ReferenceGemmInstance =
|
||||
ck::tensor_operation::host::ReferenceGemmMultipleD<ADataType,
|
||||
BDataType,
|
||||
DsDataType,
|
||||
EDataType,
|
||||
AccDataType,
|
||||
AElementOp,
|
||||
BElementOp,
|
||||
CDEElementOp>;
|
||||
|
||||
// HACK: reference GEMM expects D tensors as std::array
|
||||
// This limits D tensors to all have the same data type
|
||||
using DDataType = tuple_element_t<0, DsDataType>;
|
||||
std::array<Tensor<DDataType>, NumDTensor> d_tensors =
|
||||
make_array_from_fn<NumDTensor>(
|
||||
[&](auto j) { return d_m_n[i][ck::integral_constant<index_t, j>{}]; });
|
||||
|
||||
auto ref_gemm = ReferenceGemmInstance{};
|
||||
auto ref_invoker = ref_gemm.MakeInvoker();
|
||||
auto ref_argument = ref_gemm.MakeArgument(a_m_k[i],
|
||||
b_k_n[i],
|
||||
d_tensors,
|
||||
e_m_n_host_results[i],
|
||||
a_element_op,
|
||||
b_element_op,
|
||||
cde_element_op);
|
||||
ref_invoker.Run(ref_argument);
|
||||
}
|
||||
else
|
||||
{
|
||||
|
||||
using ReferenceGemmInstance =
|
||||
ck::tensor_operation::host::ReferenceGemm<ADataType,
|
||||
BDataType,
|
||||
EDataType,
|
||||
AccDataType,
|
||||
AElementOp,
|
||||
BElementOp,
|
||||
CDEElementOp>;
|
||||
|
||||
auto ref_gemm = ReferenceGemmInstance{};
|
||||
auto ref_invoker = ref_gemm.MakeInvoker();
|
||||
auto ref_argument = ref_gemm.MakeArgument(a_m_k[i],
|
||||
b_k_n[i],
|
||||
e_m_n_host_results[i],
|
||||
a_element_op,
|
||||
b_element_op,
|
||||
cde_element_op);
|
||||
ref_invoker.Run(ref_argument);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// profile device GEMM instances
|
||||
for(auto& gemm_ptr : op_ptrs)
|
||||
{
|
||||
auto argument_ptr = gemm_ptr->MakeArgumentPointer(
|
||||
p_a, p_b, p_ds, p_e, gemm_descs, a_element_op, b_element_op, cde_element_op);
|
||||
auto invoker_ptr = gemm_ptr->MakeInvokerPointer();
|
||||
std::string gemm_name = gemm_ptr->GetTypeString();
|
||||
|
||||
DeviceMem gemm_arg_dev_mem(gemm_ptr->GetDeviceKernelArgSize(argument_ptr.get()));
|
||||
ck::hip_check_error(hipMemcpy(gemm_arg_dev_mem.GetDeviceBuffer(),
|
||||
gemm_kargs.data(),
|
||||
gemm_ptr->GetDeviceKernelArgSize(argument_ptr.get()),
|
||||
hipMemcpyHostToDevice));
|
||||
gemm_ptr->SetDeviceKernelArgs(argument_ptr.get(), gemm_arg_dev_mem.GetDeviceBuffer());
|
||||
|
||||
if(gemm_ptr->IsSupportedArgument(argument_ptr.get()))
|
||||
{
|
||||
invoker_ptr->Run(argument_ptr.get(), StreamConfig{nullptr, false, 0, n_warmup, n_iter});
|
||||
if(do_verification)
|
||||
{
|
||||
bool instance_pass = true;
|
||||
for(std::size_t i = 0; i < gemm_descs.size(); i++)
|
||||
{
|
||||
e_device_buf[i]->FromDevice(e_m_n_device_results[i].mData.data());
|
||||
instance_pass = instance_pass && ck::utils::check_err(e_m_n_device_results[i],
|
||||
e_m_n_host_results[i]);
|
||||
|
||||
if(do_log)
|
||||
{
|
||||
LogRangeAsType<float>(std::cout << "a : ", a_m_k[i].mData, ",")
|
||||
<< std::endl;
|
||||
LogRangeAsType<float>(std::cout << "b: ", b_k_n[i].mData, ",") << std::endl;
|
||||
LogRangeAsType<float>(
|
||||
std::cout << "e_device: ", e_m_n_device_results[i].mData, ",")
|
||||
<< std::endl;
|
||||
LogRangeAsType<float>(
|
||||
std::cout << "e_host : ", e_m_n_host_results[i].mData, ",")
|
||||
<< std::endl;
|
||||
}
|
||||
}
|
||||
|
||||
std::cout << "Instance: " << gemm_name << " verification "
|
||||
<< (instance_pass ? "SUCCEED" : "FAILED") << std::endl;
|
||||
|
||||
pass = pass && instance_pass;
|
||||
}
|
||||
|
||||
if(time_kernel)
|
||||
{
|
||||
float ave_time = invoker_ptr->Run(
|
||||
argument_ptr.get(), StreamConfig{nullptr, time_kernel, 0, n_warmup, n_iter});
|
||||
|
||||
std::size_t flop = 0, num_btype = 0;
|
||||
for(std::size_t i = 0; i < gemm_descs.size(); i++)
|
||||
{
|
||||
flop += std::size_t(2) * Ms[i] * Ns[i] * Ks[i];
|
||||
|
||||
num_btype += sizeof(ADataType) * Ms[i] * Ks[i] +
|
||||
sizeof(BDataType) * Ks[i] * Ns[i] +
|
||||
sizeof(EDataType) * Ms[i] * Ns[i];
|
||||
|
||||
static_for<0, NumDTensor, 1>{}([&](auto j) -> void {
|
||||
num_btype +=
|
||||
sizeof(tuple_element_t<j, DsDataType>) * Ms[i] * Ns[i]; // D matrix
|
||||
});
|
||||
}
|
||||
|
||||
float tflops = static_cast<float>(flop) / 1.E9 / ave_time;
|
||||
float gb_per_sec = num_btype / 1.E6 / ave_time;
|
||||
std::cout << "Perf: " << std::setw(10) << ave_time << " ms, " << tflops
|
||||
<< " TFlops, " << gb_per_sec << " GB/s, " << gemm_name << std::endl;
|
||||
|
||||
if(tflops > best_tflops)
|
||||
{
|
||||
best_gemm_name = gemm_name;
|
||||
best_tflops = tflops;
|
||||
best_ave_time = ave_time;
|
||||
best_gb_per_sec = gb_per_sec;
|
||||
}
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
std::cout << "Instance: " << gemm_name << ", does not support this GEMM problem"
|
||||
<< std::endl;
|
||||
}
|
||||
}
|
||||
|
||||
if(time_kernel)
|
||||
{
|
||||
std::cout << "Best Perf: " << best_ave_time << " ms, " << best_tflops << " TFlops, "
|
||||
<< best_gb_per_sec << " GB/s, " << best_gemm_name << std::endl;
|
||||
}
|
||||
|
||||
return pass;
|
||||
}
|
||||
|
||||
} // namespace profiler
|
||||
} // namespace ck
|
||||
@@ -6,20 +6,9 @@
|
||||
#include <iomanip>
|
||||
|
||||
#include "ck/ck.hpp"
|
||||
#include "ck/utility/env.hpp"
|
||||
#include "ck/host_utility/hip_check_error.hpp"
|
||||
#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
|
||||
#include "ck/tensor_operation/gpu/device/device_grouped_gemm_tile_loop.hpp"
|
||||
#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
|
||||
|
||||
#include "ck/library/tensor_operation_instance/gpu/grouped_gemm_tile_loop.hpp"
|
||||
|
||||
#include "ck/library/utility/check_err.hpp"
|
||||
#include "ck/library/utility/device_memory.hpp"
|
||||
#include "ck/library/utility/host_tensor.hpp"
|
||||
#include "ck/library/utility/literals.hpp"
|
||||
#include "ck/library/utility/fill.hpp"
|
||||
#include "ck/library/reference_tensor_operation/cpu/reference_gemm.hpp"
|
||||
#include "profile_grouped_gemm_tile_loop_generic_impl.hpp"
|
||||
|
||||
namespace ck {
|
||||
namespace profiler {
|
||||
@@ -44,277 +33,30 @@ bool profile_grouped_gemm_tile_loop_impl(int do_verification,
|
||||
int n_warmup = 10,
|
||||
int n_iter = 50)
|
||||
{
|
||||
bool pass = true;
|
||||
|
||||
auto f_host_tensor_descriptor =
|
||||
[](std::size_t row, std::size_t col, std::size_t stride, auto layout) {
|
||||
using namespace ck::literals;
|
||||
|
||||
if(is_same<decltype(layout), tensor_layout::gemm::RowMajor>::value)
|
||||
{
|
||||
return HostTensorDescriptor({row, col}, {stride, 1_uz});
|
||||
}
|
||||
else
|
||||
{
|
||||
return HostTensorDescriptor({row, col}, {1_uz, stride});
|
||||
}
|
||||
};
|
||||
|
||||
std::size_t group_count = Ms.size();
|
||||
|
||||
if(!(group_count == Ns.size() && group_count == Ks.size() && group_count == StrideAs.size() &&
|
||||
group_count == StrideBs.size() && group_count == StrideCs.size()))
|
||||
{
|
||||
throw std::runtime_error("wrong! inconsistent M/N/Ks, StrideA/B/Cs size\n");
|
||||
}
|
||||
|
||||
std::vector<Tensor<ADataType>> a_m_k;
|
||||
std::vector<Tensor<BDataType>> b_k_n;
|
||||
std::vector<Tensor<CDataType>> c_m_n_host_results;
|
||||
std::vector<Tensor<CDataType>> c_m_n_device_results;
|
||||
|
||||
for(std::size_t i = 0; i < group_count; i++)
|
||||
{
|
||||
a_m_k.push_back(
|
||||
Tensor<ADataType>(f_host_tensor_descriptor(Ms[i], Ks[i], StrideAs[i], ALayout{})));
|
||||
b_k_n.push_back(
|
||||
Tensor<BDataType>(f_host_tensor_descriptor(Ks[i], Ns[i], StrideBs[i], BLayout{})));
|
||||
c_m_n_device_results.push_back(
|
||||
Tensor<CDataType>(f_host_tensor_descriptor(Ms[i], Ns[i], StrideCs[i], CLayout{})));
|
||||
c_m_n_host_results.push_back(
|
||||
Tensor<CDataType>(f_host_tensor_descriptor(Ms[i], Ns[i], StrideCs[i], CLayout{})));
|
||||
if(ck::EnvIsEnabled(CK_ENV(CK_LOGGING)))
|
||||
{
|
||||
std::cout << "group: " << i << " a_m_k[" << i << "]:" << a_m_k[i].mDesc << ", b_k_n["
|
||||
<< i << "]:" << b_k_n[i].mDesc << ", c_m_n_device_results[" << i
|
||||
<< "]:" << c_m_n_device_results[i].mDesc << std::endl;
|
||||
}
|
||||
switch(init_method)
|
||||
{
|
||||
case 0: break;
|
||||
case 1:
|
||||
ck::utils::FillUniformDistributionIntegerValue<ADataType>{-5, 5}(a_m_k[i]);
|
||||
ck::utils::FillUniformDistributionIntegerValue<BDataType>{-5, 5}(b_k_n[i]);
|
||||
break;
|
||||
case 2:
|
||||
ck::utils::FillUniformDistribution<ADataType>{.0, 1.}(a_m_k[i]);
|
||||
ck::utils::FillUniformDistribution<BDataType>{-0.5, 0.5}(b_k_n[i]);
|
||||
break;
|
||||
default:
|
||||
ck::utils::FillConstant<ADataType>{1}(a_m_k[i]);
|
||||
ck::utils::FillConstant<BDataType>{1}(b_k_n[i]);
|
||||
}
|
||||
}
|
||||
|
||||
using AElementOp = ck::tensor_operation::element_wise::PassThrough;
|
||||
using BElementOp = ck::tensor_operation::element_wise::PassThrough;
|
||||
using CElementOp = ck::tensor_operation::element_wise::PassThrough;
|
||||
|
||||
const auto a_element_op = AElementOp{};
|
||||
const auto b_element_op = BElementOp{};
|
||||
const auto c_element_op = CElementOp{};
|
||||
|
||||
using DeviceMemPtr = std::unique_ptr<DeviceMem>;
|
||||
std::vector<DeviceMemPtr> a_device_buf, b_device_buf, c_device_buf;
|
||||
|
||||
a_device_buf.reserve(group_count);
|
||||
b_device_buf.reserve(group_count);
|
||||
c_device_buf.reserve(group_count);
|
||||
|
||||
std::vector<const void*> p_a, p_b;
|
||||
std::vector<void*> p_c;
|
||||
|
||||
p_a.reserve(group_count);
|
||||
p_b.reserve(group_count);
|
||||
p_c.reserve(group_count);
|
||||
|
||||
using KernelArguments = ck::tensor_operation::device::GroupedGemmKernelArgument<>;
|
||||
|
||||
std::vector<ck::tensor_operation::device::GemmDesc> gemm_descs;
|
||||
std::vector<KernelArguments> gemm_kargs;
|
||||
|
||||
gemm_descs.reserve(group_count);
|
||||
gemm_kargs.reserve(group_count);
|
||||
|
||||
for(std::size_t i = 0; i < group_count; i++)
|
||||
{
|
||||
a_device_buf.emplace_back(
|
||||
std::make_unique<DeviceMem>(sizeof(ADataType) * a_m_k[i].mDesc.GetElementSpaceSize()));
|
||||
b_device_buf.emplace_back(
|
||||
std::make_unique<DeviceMem>(sizeof(BDataType) * b_k_n[i].mDesc.GetElementSpaceSize()));
|
||||
c_device_buf.emplace_back(std::make_unique<DeviceMem>(
|
||||
sizeof(CDataType) * c_m_n_device_results[i].mDesc.GetElementSpaceSize()));
|
||||
|
||||
a_device_buf[i]->ToDevice(a_m_k[i].mData.data());
|
||||
b_device_buf[i]->ToDevice(b_k_n[i].mData.data());
|
||||
c_device_buf[i]->SetZero();
|
||||
|
||||
p_a.push_back(a_device_buf[i]->GetDeviceBuffer());
|
||||
p_b.push_back(b_device_buf[i]->GetDeviceBuffer());
|
||||
p_c.push_back(c_device_buf[i]->GetDeviceBuffer());
|
||||
|
||||
gemm_descs.push_back({0, Ns[i], Ks[i], StrideAs[i], StrideBs[i], StrideCs[i], {}});
|
||||
gemm_kargs.push_back({a_device_buf[i]->GetDeviceBuffer(),
|
||||
b_device_buf[i]->GetDeviceBuffer(),
|
||||
{},
|
||||
c_device_buf[i]->GetDeviceBuffer(),
|
||||
Ms[i],
|
||||
Ns[i],
|
||||
Ks[i],
|
||||
StrideAs[i],
|
||||
StrideBs[i],
|
||||
{},
|
||||
StrideCs[i]});
|
||||
}
|
||||
|
||||
using DeviceOp = ck::tensor_operation::device::DeviceGroupedGemmTileLoop<ALayout,
|
||||
BLayout,
|
||||
ck::Tuple<>,
|
||||
CLayout,
|
||||
ADataType,
|
||||
BDataType,
|
||||
ck::Tuple<>,
|
||||
CDataType,
|
||||
AElementOp,
|
||||
BElementOp,
|
||||
CElementOp>;
|
||||
|
||||
const auto op_ptrs = ck::tensor_operation::device::instance::DeviceOperationInstanceFactory<
|
||||
DeviceOp>::GetInstances();
|
||||
|
||||
if(op_ptrs.size() <= 0)
|
||||
{
|
||||
throw std::runtime_error("wrong! no device GEMM instance found");
|
||||
}
|
||||
|
||||
std::string best_gemm_name;
|
||||
float best_ave_time = 0;
|
||||
float best_tflops = 0;
|
||||
float best_gb_per_sec = 0;
|
||||
|
||||
auto p_ds = std::vector<std::array<const void*, 0>>{};
|
||||
|
||||
if(do_verification)
|
||||
{
|
||||
for(std::size_t i = 0; i < gemm_descs.size(); i++)
|
||||
{
|
||||
using ReferenceGemmInstance = ck::tensor_operation::host::ReferenceGemm<ADataType,
|
||||
BDataType,
|
||||
CDataType,
|
||||
AccDataType,
|
||||
AElementOp,
|
||||
BElementOp,
|
||||
CElementOp>;
|
||||
|
||||
auto ref_gemm = ReferenceGemmInstance{};
|
||||
auto ref_invoker = ref_gemm.MakeInvoker();
|
||||
auto ref_argument = ref_gemm.MakeArgument(a_m_k[i],
|
||||
b_k_n[i],
|
||||
c_m_n_host_results[i],
|
||||
a_element_op,
|
||||
b_element_op,
|
||||
c_element_op);
|
||||
ref_invoker.Run(ref_argument);
|
||||
}
|
||||
}
|
||||
|
||||
// profile device GEMM instances
|
||||
for(auto& gemm_ptr : op_ptrs)
|
||||
{
|
||||
auto argument_ptr =
|
||||
gemm_ptr->MakeArgumentPointer(p_a,
|
||||
p_b,
|
||||
p_ds,
|
||||
p_c,
|
||||
gemm_descs,
|
||||
ck::tensor_operation::element_wise::PassThrough{},
|
||||
ck::tensor_operation::element_wise::PassThrough{},
|
||||
ck::tensor_operation::element_wise::PassThrough{});
|
||||
auto invoker_ptr = gemm_ptr->MakeInvokerPointer();
|
||||
std::string gemm_name = gemm_ptr->GetTypeString();
|
||||
|
||||
DeviceMem gemm_arg_dev_mem(gemm_ptr->GetDeviceKernelArgSize(argument_ptr.get()));
|
||||
hip_check_error(hipMemcpy(gemm_arg_dev_mem.GetDeviceBuffer(),
|
||||
gemm_kargs.data(),
|
||||
gemm_ptr->GetDeviceKernelArgSize(argument_ptr.get()),
|
||||
hipMemcpyHostToDevice));
|
||||
gemm_ptr->SetDeviceKernelArgs(argument_ptr.get(), gemm_arg_dev_mem.GetDeviceBuffer());
|
||||
|
||||
if(gemm_ptr->IsSupportedArgument(argument_ptr.get()))
|
||||
{
|
||||
invoker_ptr->Run(argument_ptr.get(), StreamConfig{nullptr, false, 0, n_warmup, n_iter});
|
||||
if(do_verification)
|
||||
{
|
||||
bool instance_pass = true;
|
||||
for(std::size_t i = 0; i < gemm_descs.size(); i++)
|
||||
{
|
||||
c_device_buf[i]->FromDevice(c_m_n_device_results[i].mData.data());
|
||||
instance_pass = instance_pass && ck::utils::check_err(c_m_n_device_results[i],
|
||||
c_m_n_host_results[i]);
|
||||
|
||||
if(do_log)
|
||||
{
|
||||
LogRangeAsType<float>(std::cout << "a : ", a_m_k[i].mData, ",")
|
||||
<< std::endl;
|
||||
LogRangeAsType<float>(std::cout << "b: ", b_k_n[i].mData, ",") << std::endl;
|
||||
LogRangeAsType<float>(
|
||||
std::cout << "c_device: ", c_m_n_device_results[i].mData, ",")
|
||||
<< std::endl;
|
||||
LogRangeAsType<float>(
|
||||
std::cout << "c_host : ", c_m_n_host_results[i].mData, ",")
|
||||
<< std::endl;
|
||||
}
|
||||
}
|
||||
|
||||
std::cout << "Instance: " << gemm_name << " verification "
|
||||
<< (instance_pass ? "SUCCEED" : "FAILED") << std::endl;
|
||||
|
||||
pass = pass && instance_pass;
|
||||
}
|
||||
|
||||
if(time_kernel)
|
||||
{
|
||||
float ave_time = invoker_ptr->Run(
|
||||
argument_ptr.get(), StreamConfig{nullptr, time_kernel, 0, n_warmup, n_iter});
|
||||
|
||||
std::size_t flop = 0, num_btype = 0;
|
||||
for(std::size_t i = 0; i < gemm_descs.size(); i++)
|
||||
{
|
||||
flop += std::size_t(2) * Ms[i] * Ns[i] * Ks[i];
|
||||
|
||||
num_btype += sizeof(ADataType) * Ms[i] * Ks[i] +
|
||||
sizeof(BDataType) * Ks[i] * Ns[i] +
|
||||
sizeof(CDataType) * Ms[i] * Ns[i];
|
||||
}
|
||||
|
||||
float tflops = static_cast<float>(flop) / 1.E9 / ave_time;
|
||||
float gb_per_sec = num_btype / 1.E6 / ave_time;
|
||||
std::cout << "Perf: " << std::setw(10) << ave_time << " ms, " << tflops
|
||||
<< " TFlops, " << gb_per_sec << " GB/s, " << gemm_name << std::endl;
|
||||
|
||||
if(tflops > best_tflops)
|
||||
{
|
||||
best_gemm_name = gemm_name;
|
||||
best_tflops = tflops;
|
||||
best_ave_time = ave_time;
|
||||
best_gb_per_sec = gb_per_sec;
|
||||
}
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
std::cout << "Instance: " << gemm_name << ", does not support this GEMM problem"
|
||||
<< std::endl;
|
||||
}
|
||||
}
|
||||
|
||||
if(time_kernel)
|
||||
{
|
||||
std::cout << "Best Perf: " << best_ave_time << " ms, " << best_tflops << " TFlops, "
|
||||
<< best_gb_per_sec << " GB/s, " << best_gemm_name << std::endl;
|
||||
}
|
||||
|
||||
return pass;
|
||||
return profile_grouped_gemm_tile_loop_generic_impl<ADataType,
|
||||
BDataType,
|
||||
Tuple<>,
|
||||
CDataType,
|
||||
ALayout,
|
||||
BLayout,
|
||||
Tuple<>,
|
||||
CLayout,
|
||||
PassThrough,
|
||||
PassThrough,
|
||||
PassThrough>(
|
||||
do_verification,
|
||||
init_method,
|
||||
do_log,
|
||||
time_kernel,
|
||||
Ms,
|
||||
Ns,
|
||||
Ks,
|
||||
StrideAs,
|
||||
StrideBs,
|
||||
std::vector<std::array<int, 0>>{},
|
||||
StrideCs,
|
||||
n_warmup,
|
||||
n_iter);
|
||||
}
|
||||
|
||||
} // namespace profiler
|
||||
|
||||
Reference in New Issue
Block a user