wehub-resource-sync
84 lines
2.1 KiB
C++
84 lines
2.1 KiB
C++
// Copyright (c) 2021 CINN Authors. All Rights Reserved.
|
|
//
|
|
// Licensed under the Apache License, Version 2.0 (the "License");
|
|
// you may not use this file except in compliance with the License.
|
|
// You may obtain a copy of the License at
|
|
//
|
|
// http://www.apache.org/licenses/LICENSE-2.0
|
|
//
|
|
// Unless required by applicable law or agreed to in writing, software
|
|
// distributed under the License is distributed on an "AS IS" BASIS,
|
|
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
// See the License for the specific language governing permissions and
|
|
// limitations under the License.
|
|
|
|
#pragma once
|
|
#ifdef CINN_WITH_CUDA
|
|
|
|
#include <cuda.h>
|
|
#include <cuda_runtime.h>
|
|
|
|
#include <mutex> // NOLINT
|
|
#include <string>
|
|
#include <vector>
|
|
|
|
#include "paddle/cinn/backends/cuda_util.h"
|
|
#include "paddle/cinn/runtime/cuda/cuda_util.h"
|
|
|
|
namespace cinn {
|
|
namespace runtime {
|
|
namespace cuda {
|
|
|
|
/**
|
|
* The CUDA module, helps to compile CUDA codes and fetch symbols.
|
|
* Currently, it is a wrapper of NVRTC.
|
|
*/
|
|
class CUDAModule {
|
|
public:
|
|
enum class Kind {
|
|
PTX = 0,
|
|
CUBIN = 1,
|
|
};
|
|
|
|
CUDAModule(const std::string& data, Kind kind);
|
|
|
|
void LaunchKernel(int device_id,
|
|
const std::string& func_name,
|
|
dim3 gridDim,
|
|
dim3 blockDim,
|
|
void** args,
|
|
size_t share_memory_size = 0,
|
|
CUstream stream = nullptr);
|
|
|
|
//! Get a function.
|
|
CUfunction GetFunction(int device_id, const std::string& func_name);
|
|
|
|
//! Get a function by CudaGetDevice
|
|
CUfunction GetFunction(const std::string& func_name);
|
|
|
|
//! Get a global variable.
|
|
CUdeviceptr GetGlobal(int device_id, const std::string& name, size_t nbytes);
|
|
|
|
~CUDAModule();
|
|
|
|
private:
|
|
//! The input data.
|
|
std::string data_;
|
|
//! Kind of the input.
|
|
Kind kind_;
|
|
//! To make parallel, we prepare one module for each card.
|
|
std::vector<CUmodule> module_per_card_{kCUDAMaxCards, nullptr};
|
|
std::string cuda_source_;
|
|
std::mutex mutex_;
|
|
|
|
CUdevice device_;
|
|
CUcontext context_;
|
|
int num_devices_{0};
|
|
};
|
|
|
|
} // namespace cuda
|
|
} // namespace runtime
|
|
} // namespace cinn
|
|
|
|
#endif // CINN_WITH_CUDA
|