/**

 * This program is free software, you can redistribute it and/or modify it.

 * Copyright (c) 2025 Huawei Technologies Co., Ltd.

 * This file is a part of the CANN Open Software.

 * Licensed under CANN Open Software License Agreement Version 2.0 (the "License").

 * Please refer to the License for details. You may not use this file except in compliance with the License.

 * THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING

 * BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.

 * See LICENSE in the root of the software repository for the full text of the License.

 */



#include <iostream>

#include <vector>

#include "acl/acl.h"

#include "aclnn_pow2.h"



#define CHECK_RET(cond, return_expr) \

    do {                             \

        if (!(cond)) {               \

            return_expr;             \

        }                            \

    } while (0)

#define LOG_PRINT(message, ...)         \

    do {                                \

        printf(message, ##__VA_ARGS__); \

    } while (0)

int64_t GetShapeSize(const std::vector<int64_t>& shape)

{

    int64_t shapeSize = 1;

    for (auto i : shape) {

        shapeSize *= i;

    }

    return shapeSize;

}

int Init(int32_t deviceId, aclrtStream* stream)

{

    // 固定写法，资源初始化

    auto ret = aclInit(nullptr);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclInit failed. ERROR: %d\n", ret); return ret);

    ret = aclrtSetDevice(deviceId);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSetDevice failed. ERROR: %d\n", ret); return ret);

    ret = aclrtCreateStream(stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtCreateStream failed. ERROR: %d\n", ret); return ret);

    return 0;

}



template <typename T>

int CreateAclTensor(

    const std::vector<T>& hostData, const std::vector<int64_t>& shape, void** deviceAddr, aclDataType dataType,

    aclTensor** tensor)

{

    auto size = GetShapeSize(shape) * sizeof(T);

    // 调用aclrtMalloc申请device侧内存

    auto ret = aclrtMalloc(deviceAddr, size, ACL_MEM_MALLOC_HUGE_FIRST);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMalloc failed. ERROR: %d\n", ret); return ret);

    // 调用aclrtMemcpy将host侧数据拷贝到device侧内存上

    ret = aclrtMemcpy(*deviceAddr, size, hostData.data(), size, ACL_MEMCPY_HOST_TO_DEVICE);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMemcpy failed. ERROR: %d\n", ret); return ret);

    // 计算连续tensor的strides

    std::vector<int64_t> strides(shape.size(), 1);

    for (int64_t i = shape.size() - 2; i >= 0; i--) {

        strides[i] = shape[i + 1] * strides[i + 1];

    }

    // 调用aclCreateTensor接口创建aclTensor

    *tensor = aclCreateTensor(

        shape.data(), shape.size(), dataType, strides.data(), 0, aclFormat::ACL_FORMAT_ND, shape.data(), shape.size(),

        *deviceAddr);

    return 0;

}



int testFp32(){

    LOG_PRINT("Test for Fp32\n");

    // 1. （固定写法）device/stream初始化，参考acl API手册

    // 根据自己的实际device填写deviceId

    int32_t deviceId = 0;

    aclrtStream stream;

    auto ret = Init(deviceId, &stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

    // 2. 构造输入与输出，需要根据API的接口自定义构造

    std::vector<int64_t> selfShape = {10240};

    std::vector<int64_t> exponentShape = {10240};

    std::vector<int64_t> outShape = {10240};

    void* selfDeviceAddr = nullptr;

    void* exponentDeviceAddr = nullptr;

    void* outDeviceAddr = nullptr;

    aclTensor* self = nullptr;

    aclTensor* exponent = nullptr;

    aclTensor* out = nullptr;

    std::vector<float> selfHostData = {99, -1, -2, 0};

    std::vector<float> exponentHostData = {0, 1, 2, 0};

    std::vector<float> outHostData = {0, 0, 0, 0};

    for(int i = 4; i < 10240; i++){

        selfHostData.push_back(i);

        exponentHostData.push_back(2);

        outHostData.push_back(0);

    }

    // 创建self aclTensor

    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建threshold aclScalar

    ret = CreateAclTensor(exponentHostData, exponentShape, &exponentDeviceAddr, aclDataType::ACL_FLOAT, &exponent);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建out aclTensor

    ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 3. 调用CANN算子库API，需要修改为具体的API名称

    // aclnnPow2接口调用示例

    uint64_t workspaceSize = 0;

    aclOpExecutor* executor;

    // 调用aclnnPow2第一段接口

    ret = aclnnPow2GetWorkspaceSize(self, exponent, out, &workspaceSize, &executor);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2GetWorkspaceSize failed. ERROR: %d\n", ret);

              return ret);

    // 根据第一段接口计算出的workspaceSize申请device内存

    void* workspaceAddr = nullptr;

    if (workspaceSize > 0) {

        ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);

        CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);

    }

    // 调用aclnnPow2第二段接口

    ret = aclnnPow2(workspaceAddr, workspaceSize, executor, stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2 failed. ERROR: %d\n", ret); return ret);



    // 4. （固定写法）同步等待任务执行结束

    ret = aclrtSynchronizeStream(stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);



    // 5. 获取输出的值，将device侧内存上的结果拷贝至host侧，需要根据具体API的接口定义修改

    auto size = GetShapeSize(outShape);

    std::vector<float> resultData(size, 0);

    ret = aclrtMemcpy(

        resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * sizeof(resultData[0]),

        ACL_MEMCPY_DEVICE_TO_HOST);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);

    for (int64_t i = 0; i < 5; i++) {

        LOG_PRINT("aclnnPow2 result[%ld] is: %f\n", i, resultData[i]);

    }

    // 6. 释放aclTensor和aclScalar，需要根据具体API的接口定义修改

    aclDestroyTensor(self);

    aclDestroyTensor(exponent);

    aclDestroyTensor(out);



    // 7. 释放device资源，需要根据具体API的接口定义修改

    aclrtFree(selfDeviceAddr);

    aclrtFree(exponentDeviceAddr);

    aclrtFree(outDeviceAddr);

    if (workspaceSize > 0) {

        aclrtFree(workspaceAddr);

    }

    aclrtDestroyStream(stream);

    aclrtResetDevice(deviceId);

    aclFinalize();

}



int testUint8(){

    LOG_PRINT("Test for UINT8\n");

    int32_t deviceId = 0;

    aclrtStream stream;

    auto ret = Init(deviceId, &stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

    // 2. 构造输入与输出，需要根据API的接口自定义构造

    std::vector<int64_t> selfShape = {102400};

    std::vector<int64_t> exponentShape = {102400};

    std::vector<int64_t> outShape = {102400};

    void* selfDeviceAddr = nullptr;

    void* exponentDeviceAddr = nullptr;

    void* outDeviceAddr = nullptr;

    aclTensor* self = nullptr;

    aclTensor* exponent = nullptr;

    aclTensor* out = nullptr;

    std::vector<uint8_t> selfHostData = {99, 1, 0, 0};

    std::vector<uint8_t> exponentHostData = {0, 1, 2, 0};

    std::vector<uint8_t> outHostData = {0, 0, 0, 0};

    for(int i = 4; i < 102400; i++){

        selfHostData.push_back(3);

        exponentHostData.push_back(2);

        outHostData.push_back(0);

    }

    // 创建self aclTensor

    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_UINT8, &self);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建threshold aclScalar

    ret = CreateAclTensor(exponentHostData, exponentShape, &exponentDeviceAddr, aclDataType::ACL_UINT8, &exponent);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建out aclTensor

    ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_UINT8, &out);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 3. 调用CANN算子库API，需要修改为具体的API名称

    // aclnnPow2接口调用示例

    uint64_t workspaceSize = 0;

    aclOpExecutor* executor;

    // 调用aclnnPow2第一段接口

    ret = aclnnPow2GetWorkspaceSize(self, exponent, out, &workspaceSize, &executor);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2GetWorkspaceSize failed. ERROR: %d\n", ret);

              return ret);

    // 根据第一段接口计算出的workspaceSize申请device内存

    void* workspaceAddr = nullptr;

    if (workspaceSize > 0) {

        ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);

        CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);

    }

    // 调用aclnnPow2第二段接口

    ret = aclnnPow2(workspaceAddr, workspaceSize, executor, stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2 failed. ERROR: %d\n", ret); return ret);



    // 4. （固定写法）同步等待任务执行结束

    ret = aclrtSynchronizeStream(stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);



    // 5. 获取输出的值，将device侧内存上的结果拷贝至host侧，需要根据具体API的接口定义修改

    auto size = GetShapeSize(outShape);

    std::vector<uint8_t> resultData(size, 0);

    ret = aclrtMemcpy(

        resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * sizeof(resultData[0]),

        ACL_MEMCPY_DEVICE_TO_HOST);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);

    for (int64_t i = 0; i < 15; i++) {

        LOG_PRINT("aclnnPow2 result[%ld] is: %u\n", i, (uint32_t)resultData[i]);

    }

    // 6. 释放aclTensor和aclScalar，需要根据具体API的接口定义修改

    aclDestroyTensor(self);

    aclDestroyTensor(exponent);

    aclDestroyTensor(out);



    // 7. 释放device资源，需要根据具体API的接口定义修改

    aclrtFree(selfDeviceAddr);

    aclrtFree(exponentDeviceAddr);

    aclrtFree(outDeviceAddr);

    if (workspaceSize > 0) {

        aclrtFree(workspaceAddr);

    }

    aclrtDestroyStream(stream);

    aclrtResetDevice(deviceId);

    aclFinalize();

    return 0;

}



int testInt8(){

    LOG_PRINT("Test for INT8\n");

    int32_t deviceId = 0;

    aclrtStream stream;

    auto ret = Init(deviceId, &stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

    // 2. 构造输入与输出，需要根据API的接口自定义构造

    std::vector<int64_t> selfShape = {102400};

    std::vector<int64_t> exponentShape = {102400};

    std::vector<int64_t> outShape = {102400};

    void* selfDeviceAddr = nullptr;

    void* exponentDeviceAddr = nullptr;

    void* outDeviceAddr = nullptr;

    aclTensor* self = nullptr;

    aclTensor* exponent = nullptr;

    aclTensor* out = nullptr;

    std::vector<int8_t> selfHostData = {99, -1, -2, 0};

    std::vector<int8_t> exponentHostData = {0, 1, 2, 0};

    std::vector<int8_t> outHostData = {0, 0, 0, 0};

    for(int i = 4; i < 102400; i++){

        selfHostData.push_back(3);

        exponentHostData.push_back(2);

        outHostData.push_back(0);

    }

    // 创建self aclTensor

    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_INT8, &self);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建threshold aclScalar

    ret = CreateAclTensor(exponentHostData, exponentShape, &exponentDeviceAddr, aclDataType::ACL_INT8, &exponent);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建out aclTensor

    ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_INT8, &out);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 3. 调用CANN算子库API，需要修改为具体的API名称

    // aclnnPow2接口调用示例

    uint64_t workspaceSize = 0;

    aclOpExecutor* executor;

    // 调用aclnnPow2第一段接口

    ret = aclnnPow2GetWorkspaceSize(self, exponent, out, &workspaceSize, &executor);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2GetWorkspaceSize failed. ERROR: %d\n", ret);

              return ret);

    // 根据第一段接口计算出的workspaceSize申请device内存

    void* workspaceAddr = nullptr;

    if (workspaceSize > 0) {

        ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);

        CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);

    }

    // 调用aclnnPow2第二段接口

    ret = aclnnPow2(workspaceAddr, workspaceSize, executor, stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2 failed. ERROR: %d\n", ret); return ret);



    // 4. （固定写法）同步等待任务执行结束

    ret = aclrtSynchronizeStream(stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);



    // 5. 获取输出的值，将device侧内存上的结果拷贝至host侧，需要根据具体API的接口定义修改

    auto size = GetShapeSize(outShape);

    std::vector<int8_t> resultData(size, 0);

    ret = aclrtMemcpy(

        resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * sizeof(resultData[0]),

        ACL_MEMCPY_DEVICE_TO_HOST);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);

    for (int64_t i = 0; i < 15; i++) {

        LOG_PRINT("aclnnPow2 result[%ld] is: %d\n", i, (int32_t)resultData[i]);

    }

    // 6. 释放aclTensor和aclScalar，需要根据具体API的接口定义修改

    aclDestroyTensor(self);

    aclDestroyTensor(exponent);

    aclDestroyTensor(out);



    // 7. 释放device资源，需要根据具体API的接口定义修改

    aclrtFree(selfDeviceAddr);

    aclrtFree(exponentDeviceAddr);

    aclrtFree(outDeviceAddr);

    if (workspaceSize > 0) {

        aclrtFree(workspaceAddr);

    }

    aclrtDestroyStream(stream);

    aclrtResetDevice(deviceId);

    aclFinalize();

    return 0;

}



int testInt16(){

    LOG_PRINT("Test for INT16\n");

    int32_t deviceId = 0;

    aclrtStream stream;

    auto ret = Init(deviceId, &stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

    // 2. 构造输入与输出，需要根据API的接口自定义构造

    std::vector<int64_t> selfShape = {2, 2};

    std::vector<int64_t> exponentShape = {2, 2};

    std::vector<int64_t> outShape = {2, 2};

    void* selfDeviceAddr = nullptr;

    void* exponentDeviceAddr = nullptr;

    void* outDeviceAddr = nullptr;

    aclTensor* self = nullptr;

    aclTensor* exponent = nullptr;

    aclTensor* out = nullptr;

    std::vector<int16_t> selfHostData = {99, -1, -2, 0};

    std::vector<int16_t> exponentHostData = {0, 1, 2, 0};

    std::vector<int16_t> outHostData = {0, 0, 0, 0};

    // 创建self aclTensor

    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_INT16, &self);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建threshold aclScalar

    ret = CreateAclTensor(exponentHostData, exponentShape, &exponentDeviceAddr, aclDataType::ACL_INT16, &exponent);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建out aclTensor

    ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_INT16, &out);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 3. 调用CANN算子库API，需要修改为具体的API名称

    // aclnnPow2接口调用示例

    uint64_t workspaceSize = 0;

    aclOpExecutor* executor;

    // 调用aclnnPow2第一段接口

    ret = aclnnPow2GetWorkspaceSize(self, exponent, out, &workspaceSize, &executor);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2GetWorkspaceSize failed. ERROR: %d\n", ret);

              return ret);

    // 根据第一段接口计算出的workspaceSize申请device内存

    void* workspaceAddr = nullptr;

    if (workspaceSize > 0) {

        ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);

        CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);

    }

    // 调用aclnnPow2第二段接口

    ret = aclnnPow2(workspaceAddr, workspaceSize, executor, stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2 failed. ERROR: %d\n", ret); return ret);



    // 4. （固定写法）同步等待任务执行结束

    ret = aclrtSynchronizeStream(stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);



    // 5. 获取输出的值，将device侧内存上的结果拷贝至host侧，需要根据具体API的接口定义修改

    auto size = GetShapeSize(outShape);

    std::vector<int16_t> resultData(size, 0);

    ret = aclrtMemcpy(

        resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * sizeof(resultData[0]),

        ACL_MEMCPY_DEVICE_TO_HOST);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);

    for (int64_t i = 0; i < size; i++) {

        LOG_PRINT("aclnnPow2 result[%ld] is: %d\n", i, (int32_t)resultData[i]);

    }

    // 6. 释放aclTensor和aclScalar，需要根据具体API的接口定义修改

    aclDestroyTensor(self);

    aclDestroyTensor(exponent);

    aclDestroyTensor(out);



    // 7. 释放device资源，需要根据具体API的接口定义修改

    aclrtFree(selfDeviceAddr);

    aclrtFree(exponentDeviceAddr);

    aclrtFree(outDeviceAddr);

    if (workspaceSize > 0) {

        aclrtFree(workspaceAddr);

    }

    aclrtDestroyStream(stream);

    aclrtResetDevice(deviceId);

    aclFinalize();

    return 0;

}



int testInt32(){

    LOG_PRINT("Test for INT32\n");

    int32_t deviceId = 0;

    aclrtStream stream;

    auto ret = Init(deviceId, &stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

    // 2. 构造输入与输出，需要根据API的接口自定义构造

    std::vector<int64_t> selfShape = {10240};

    std::vector<int64_t> exponentShape = {10240};

    std::vector<int64_t> outShape = {10240};

    void* selfDeviceAddr = nullptr;

    void* exponentDeviceAddr = nullptr;

    void* outDeviceAddr = nullptr;

    aclTensor* self = nullptr;

    aclTensor* exponent = nullptr;

    aclTensor* out = nullptr;

    std::vector<int32_t> selfHostData = {99, -1, -2, 0};

    std::vector<int32_t> exponentHostData = {0, 1, 2, 0};

    std::vector<int32_t> outHostData = {0, 0, 0, 0};

    for(int i = 4; i < 10240; i++){

        selfHostData.push_back(i);

        exponentHostData.push_back(2);

        outHostData.push_back(0);

    }

    // 创建self aclTensor

    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_INT32, &self);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建threshold aclScalar

    ret = CreateAclTensor(exponentHostData, exponentShape, &exponentDeviceAddr, aclDataType::ACL_INT32, &exponent);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建out aclTensor

    ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_INT32, &out);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 3. 调用CANN算子库API，需要修改为具体的API名称

    // aclnnPow2接口调用示例

    uint64_t workspaceSize = 0;

    aclOpExecutor* executor;

    // 调用aclnnPow2第一段接口

    ret = aclnnPow2GetWorkspaceSize(self, exponent, out, &workspaceSize, &executor);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2GetWorkspaceSize failed. ERROR: %d\n", ret);

              return ret);

    // 根据第一段接口计算出的workspaceSize申请device内存

    void* workspaceAddr = nullptr;

    if (workspaceSize > 0) {

        ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);

        CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);

    }

    // 调用aclnnPow2第二段接口

    ret = aclnnPow2(workspaceAddr, workspaceSize, executor, stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2 failed. ERROR: %d\n", ret); return ret);



    // 4. （固定写法）同步等待任务执行结束

    ret = aclrtSynchronizeStream(stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);



    // 5. 获取输出的值，将device侧内存上的结果拷贝至host侧，需要根据具体API的接口定义修改

    auto size = GetShapeSize(outShape);

    std::vector<int32_t> resultData(size, 0);

    ret = aclrtMemcpy(

        resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * sizeof(resultData[0]),

        ACL_MEMCPY_DEVICE_TO_HOST);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);

    for (int64_t i = 0; i < 10; i++) {

        LOG_PRINT("aclnnPow2 result[%ld] is: %d\n", i, (int32_t)resultData[i]);

    }

    // 6. 释放aclTensor和aclScalar，需要根据具体API的接口定义修改

    aclDestroyTensor(self);

    aclDestroyTensor(exponent);

    aclDestroyTensor(out);



    // 7. 释放device资源，需要根据具体API的接口定义修改

    aclrtFree(selfDeviceAddr);

    aclrtFree(exponentDeviceAddr);

    aclrtFree(outDeviceAddr);

    if (workspaceSize > 0) {

        aclrtFree(workspaceAddr);

    }

    aclrtDestroyStream(stream);

    aclrtResetDevice(deviceId);

    aclFinalize();

    return 0;

}



int testInt8UInt8(){

    LOG_PRINT("Test for testInt8UInt8\n");

    const aclDataType dataTypeX1 = ACL_INT8;             // 数据类型：float

    const aclDataType dataTypeX2 = ACL_UINT8;  

    const aclDataType dataTypeY = ACL_INT16;  

    using DataX1 = int8_t;

    using DataX2 = uint8_t; 

    using DataY = int16_t;    

    int32_t deviceId = 0;

    aclrtStream stream;

    auto ret = Init(deviceId, &stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

    // 2. 构造输入与输出，需要根据API的接口自定义构造

    std::vector<int64_t> selfShape = {20};

    std::vector<int64_t> exponentShape = {20};

    std::vector<int64_t> outShape = {20};

    void* selfDeviceAddr = nullptr;

    void* exponentDeviceAddr = nullptr;

    void* outDeviceAddr = nullptr;

    aclTensor* self = nullptr;

    aclTensor* exponent = nullptr;

    aclTensor* out = nullptr;

    std::vector<DataX1> selfHostData = {99, -1, -2, 0};

    std::vector<DataX2> exponentHostData = {0, 1, 2, 0};

    std::vector<DataY> outHostData = {0, 0, 0, 0};

    for(int i = 4; i < 20; i++){

        selfHostData.push_back(i);

        exponentHostData.push_back(2);

        outHostData.push_back(0);

    }

    // 创建self aclTensor

    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, dataTypeX1, &self);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建threshold aclScalar

    ret = CreateAclTensor(exponentHostData, exponentShape, &exponentDeviceAddr, dataTypeX2, &exponent);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建out aclTensor

    ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, dataTypeY, &out);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 3. 调用CANN算子库API，需要修改为具体的API名称

    // aclnnPow2接口调用示例

    uint64_t workspaceSize = 0;

    aclOpExecutor* executor;

    // 调用aclnnPow2第一段接口

    ret = aclnnPow2GetWorkspaceSize(self, exponent, out, &workspaceSize, &executor);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2GetWorkspaceSize failed. ERROR: %d\n", ret);

              return ret);

    // 根据第一段接口计算出的workspaceSize申请device内存

    void* workspaceAddr = nullptr;

    if (workspaceSize > 0) {

        ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);

        CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);

    }

    // 调用aclnnPow2第二段接口

    ret = aclnnPow2(workspaceAddr, workspaceSize, executor, stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2 failed. ERROR: %d\n", ret); return ret);



    // 4. （固定写法）同步等待任务执行结束

    ret = aclrtSynchronizeStream(stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);



    // 5. 获取输出的值，将device侧内存上的结果拷贝至host侧，需要根据具体API的接口定义修改

    auto size = GetShapeSize(outShape);

    std::vector<int32_t> resultData(size, 0);

    ret = aclrtMemcpy(

        resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * sizeof(resultData[0]),

        ACL_MEMCPY_DEVICE_TO_HOST);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);

    for (int64_t i = 0; i < 10; i++) {

        LOG_PRINT("aclnnPow2 result[%ld] is: %d\n", i, (int32_t)resultData[i]);

    }

    // 6. 释放aclTensor和aclScalar，需要根据具体API的接口定义修改

    aclDestroyTensor(self);

    aclDestroyTensor(exponent);

    aclDestroyTensor(out);



    // 7. 释放device资源，需要根据具体API的接口定义修改

    aclrtFree(selfDeviceAddr);

    aclrtFree(exponentDeviceAddr);

    aclrtFree(outDeviceAddr);

    if (workspaceSize > 0) {

        aclrtFree(workspaceAddr);

    }

    aclrtDestroyStream(stream);

    aclrtResetDevice(deviceId);

    aclFinalize();

    return 0;

}



int testInt8Fp32(){

    LOG_PRINT("Test for testInt8Fp32\n");

    const aclDataType dataTypeX1 = ACL_INT8;             // 数据类型：float

    const aclDataType dataTypeX2 = ACL_FLOAT;  

    const aclDataType dataTypeY = ACL_FLOAT;  

    using DataX1 = int8_t;

    using DataX2 = float; 

    using DataY = float;    

    int32_t deviceId = 0;

    aclrtStream stream;

    auto ret = Init(deviceId, &stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

    // 2. 构造输入与输出，需要根据API的接口自定义构造

    std::vector<int64_t> selfShape = {10240};

    std::vector<int64_t> exponentShape = {10240};

    std::vector<int64_t> outShape = {10240};

    void* selfDeviceAddr = nullptr;

    void* exponentDeviceAddr = nullptr;

    void* outDeviceAddr = nullptr;

    aclTensor* self = nullptr;

    aclTensor* exponent = nullptr;

    aclTensor* out = nullptr;

    std::vector<DataX1> selfHostData = {99, -1, -2, 0};

    std::vector<DataX2> exponentHostData = {0, 1, 2, 0};

    std::vector<DataY> outHostData = {0, 0, 0, 0};

    for(int i = 4; i < 32; i++){

        selfHostData.push_back(i);

        exponentHostData.push_back(2);

        outHostData.push_back(0);

    }

    // 创建self aclTensor

    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, dataTypeX1, &self);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建threshold aclScalar

    ret = CreateAclTensor(exponentHostData, exponentShape, &exponentDeviceAddr, dataTypeX2, &exponent);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建out aclTensor

    ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, dataTypeY, &out);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 3. 调用CANN算子库API，需要修改为具体的API名称

    // aclnnPow2接口调用示例

    uint64_t workspaceSize = 0;

    aclOpExecutor* executor;

    // 调用aclnnPow2第一段接口

    ret = aclnnPow2GetWorkspaceSize(self, exponent, out, &workspaceSize, &executor);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2GetWorkspaceSize failed. ERROR: %d\n", ret);

              return ret);

    // 根据第一段接口计算出的workspaceSize申请device内存

    void* workspaceAddr = nullptr;

    if (workspaceSize > 0) {

        ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);

        CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);

    }

    // 调用aclnnPow2第二段接口

    ret = aclnnPow2(workspaceAddr, workspaceSize, executor, stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2 failed. ERROR: %d\n", ret); return ret);



    // 4. （固定写法）同步等待任务执行结束

    ret = aclrtSynchronizeStream(stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);



    // 5. 获取输出的值，将device侧内存上的结果拷贝至host侧，需要根据具体API的接口定义修改

    auto size = GetShapeSize(outShape);

    std::vector<int32_t> resultData(size, 0);

    ret = aclrtMemcpy(

        resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * sizeof(resultData[0]),

        ACL_MEMCPY_DEVICE_TO_HOST);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);

    for (int64_t i = 0; i < 5; i++) {

            LOG_PRINT("aclnnPow2 result[%ld] is: %f\n", i, resultData[i]);

        }

    // 6. 释放aclTensor和aclScalar，需要根据具体API的接口定义修改

    aclDestroyTensor(self);

    aclDestroyTensor(exponent);

    aclDestroyTensor(out);



    // 7. 释放device资源，需要根据具体API的接口定义修改

    aclrtFree(selfDeviceAddr);

    aclrtFree(exponentDeviceAddr);

    aclrtFree(outDeviceAddr);

    if (workspaceSize > 0) {

        aclrtFree(workspaceAddr);

    }

    aclrtDestroyStream(stream);

    aclrtResetDevice(deviceId);

    aclFinalize();

    return 0;

}



int testBroadcast(){

    LOG_PRINT("Test for Broadcast\n");

    // 1. （固定写法）device/stream初始化，参考acl API手册

    // 根据自己的实际device填写deviceId

    int32_t deviceId = 0;

    aclrtStream stream;

    auto ret = Init(deviceId, &stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

    // 2. 构造输入与输出，需要根据API的接口自定义构造

    std::vector<int64_t> selfShape = {2 , 1 ,9};

    std::vector<int64_t> exponentShape = {1 ,4 ,1};

    std::vector<int64_t> outShape = {2, 4, 9};

    void* selfDeviceAddr = nullptr;

    void* exponentDeviceAddr = nullptr;

    void* outDeviceAddr = nullptr;

    aclTensor* self = nullptr;

    aclTensor* exponent = nullptr;

    aclTensor* out = nullptr;

    std::vector<float> selfHostData = {2, 1, -2, 0};

    std::vector<float> exponentHostData = {0, 1, 2, 0};

    std::vector<float> outHostData (72, 0);

    for(int i = 4; i < 18; i++){

        selfHostData.push_back(i);

    }

    // 创建self aclTensor

    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建threshold aclScalar

    ret = CreateAclTensor(exponentHostData, exponentShape, &exponentDeviceAddr, aclDataType::ACL_FLOAT, &exponent);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 创建out aclTensor

    ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);

    CHECK_RET(ret == ACL_SUCCESS, return ret);

    // 3. 调用CANN算子库API，需要修改为具体的API名称

    // aclnnPow2接口调用示例

    uint64_t workspaceSize = 0;

    aclOpExecutor* executor;

    // 调用aclnnPow2第一段接口

    ret = aclnnPow2GetWorkspaceSize(self, exponent, out, &workspaceSize, &executor);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2GetWorkspaceSize failed. ERROR: %d\n", ret);

              return ret);

    // 根据第一段接口计算出的workspaceSize申请device内存

    void* workspaceAddr = nullptr;

    if (workspaceSize > 0) {

        ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);

        CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);

    }

    // 调用aclnnPow2第二段接口

    ret = aclnnPow2(workspaceAddr, workspaceSize, executor, stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnPow2 failed. ERROR: %d\n", ret); return ret);



    // 4. （固定写法）同步等待任务执行结束

    ret = aclrtSynchronizeStream(stream);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);



    // 5. 获取输出的值，将device侧内存上的结果拷贝至host侧，需要根据具体API的接口定义修改

    auto size = GetShapeSize(outShape);

    std::vector<float> resultData(size, 0);

    ret = aclrtMemcpy(

        resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * sizeof(resultData[0]),

        ACL_MEMCPY_DEVICE_TO_HOST);

    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);

    for (int64_t i = 0; i < 36; i++) {

        LOG_PRINT("aclnnPow2 result[%ld] is: %f\n", i, resultData[i]);

    }

    // 6. 释放aclTensor和aclScalar，需要根据具体API的接口定义修改

    aclDestroyTensor(self);

    aclDestroyTensor(exponent);

    aclDestroyTensor(out);



    // 7. 释放device资源，需要根据具体API的接口定义修改

    aclrtFree(selfDeviceAddr);

    aclrtFree(exponentDeviceAddr);

    aclrtFree(outDeviceAddr);

    if (workspaceSize > 0) {

        aclrtFree(workspaceAddr);

    }

    aclrtDestroyStream(stream);

    aclrtResetDevice(deviceId);

    aclFinalize();

}



int main()

{

    // testFp32();

    // testInt32();

    // testInt16();

    // testInt8();

    // testUint8();

    testInt8UInt8();

    // testInt8Fp32();

    // testBroadcast();

    return 0;

}