/*
 * Copyright (c) 2024 Huawei Technologies Co., Ltd.
 * This program is free software, you can redistribute it and/or modify it under the terms and conditions of
 * 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 <gtest/gtest.h>
#include <ATen/ATen.h>
#include <torch/torch.h>
#include "test_common.h"
#include <mki/utils/log/log.h>
#include "asdops/params/elewise.h"
#include "test_utils/float_util.h"
#include "test_utils/op_test.h"
#include "test_utils/golden.h"
#include "op_desc_json.h"

using namespace AsdOps;
using namespace Mki;

namespace {
constexpr float ATOL = 0.001;
constexpr float RTOL = 0.001;
constexpr float HALF_FLOAT_MIN = 6.2E-5;
constexpr float HALF_FLOAT_MAX = 65504;

Status NegF16Compare(float atol, float rtol, const Mki::Test::GoldenContext &context)
{
    const Tensor &inTensor = context.hostInTensors.at(0);
    at::Tensor atInRefTensor = at::from_blob(inTensor.data, ToIntArrayRef(inTensor.desc.dims), at::kHalf);
    at::Tensor atInRefTensorFloat = atInRefTensor.to(at::kFloat);

    at::Tensor atOutRefTensor = at::neg(atInRefTensorFloat);

    const Tensor &outTensor = context.hostOutTensors.at(0);
    at::Tensor atOutTensor = at::from_blob(outTensor.data, ToIntArrayRef(outTensor.desc.dims), at::kHalf);
    at::Tensor atOutTensorFloat = atOutTensor.to(at::kFloat);

    float *atOutArray = (float *)atOutTensorFloat.storage().data_ptr().get();
    float *atOutRefArray = (float *)atOutRefTensor.storage().data_ptr().get();

    for (int64_t i = 0; i < outTensor.Numel(); i++) {
        float expect = atOutRefArray[i];
        float result = atOutArray[i];
        if (!Mki::Test::FloatUtil::FloatJudgeEqual(expect, result, atol, rtol)) {
            return Status::FailStatus(-1, "float judge not equal");
        }
    }
    return Status::OkStatus();
}
}

TEST(TestOpElewiseNeg, TestNegF16)
{
    Mki::Test::MkiOpTest opTest;
    opTest.Golden(std::bind(&NegF16Compare, ATOL, RTOL, std::placeholders::_1));
    opTest.FloatRand(HALF_FLOAT_MIN, HALF_FLOAT_MAX);
    OpParam::Elewise opParam = {OpParam::Elewise::ELEWISE_NEG};
    Mki::Test::UtOpDesc opDesc = {"ElewiseOperation", opParam};

    Status status = opTest.Run(opDesc, {TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {1000}});
    ASSERT_EQ(status.Ok(), true);

    status = opTest.Run(opDesc, {TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {10000000}});
    ASSERT_EQ(status.Ok(), true);

    status = opTest.Run(opDesc, {TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {2, 3}});
    ASSERT_EQ(status.Ok(), true);

    status = opTest.Run(opDesc, {TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {2, 1000000}});
    ASSERT_EQ(status.Ok(), true);

    status = opTest.Run(opDesc, {TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {2, 3, 4, 5, 6, 7}});
    ASSERT_EQ(status.Ok(), true);

    status = opTest.Run(opDesc, {TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {2, 3, 4, 5, 6, 7, 9, 10}});
    ASSERT_EQ(status.Ok(), true);

    status = opTest.Run(opDesc, {TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {12, 1, 32, 32}});
    ASSERT_EQ(status.Ok(), true);
}

/**
 * @brief ok
 */
TEST(TestOpElewiseNeg, TestCanSupport0)
{
    LaunchParam launchParam;
    launchParam.AddInTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    launchParam.AddOutTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    OpParam::Elewise opParam = {};
    opParam.elewiseType = OpParam::Elewise::ELEWISE_NEG;
    Mki::Test::UtOpDesc opDesc = {"ElewiseOperation", opParam};
    launchParam.SetParam(opDesc.specificParam);
    Mki::Operation *op = Mki::AutoGen::GetOpByName(opDesc.opName);
    auto kernel = std::unique_ptr<Mki::Kernel>(op->GetKernelByName("NegF16Kernel"));
    ASSERT_NE(kernel, nullptr);
    ASSERT_EQ(kernel->CanSupport(launchParam), true);
}

/**
 * @brief elewiseType wrong
 */
TEST(TestOpElewiseNeg, TestCanSupport1)
{
    LaunchParam launchParam;
    launchParam.AddInTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    launchParam.AddOutTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    OpParam::Elewise opParam = {};
    opParam.elewiseType = OpParam::Elewise::ELEWISE_LOGICAL_NOT;
    Mki::Test::UtOpDesc opDesc = {"ElewiseOperation", opParam};
    launchParam.SetParam(opDesc.specificParam);
    Mki::Operation *op = Mki::AutoGen::GetOpByName(opDesc.opName);
    auto kernel = std::unique_ptr<Mki::Kernel>(op->GetKernelByName("NegF16Kernel"));
    ASSERT_NE(kernel, nullptr);
    ASSERT_EQ(kernel->CanSupport(launchParam), false);
}

/**
 * @brief inPutNum wrong
 */
TEST(TestOpElewiseNeg, TestCanSupport2)
{
    LaunchParam launchParam;
    launchParam.AddInTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    launchParam.AddInTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    launchParam.AddOutTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    OpParam::Elewise opParam = {};
    opParam.elewiseType = OpParam::Elewise::ELEWISE_NEG;
    Mki::Test::UtOpDesc opDesc = {"ElewiseOperation", opParam};
    launchParam.SetParam(opDesc.specificParam);
    Mki::Operation *op = Mki::AutoGen::GetOpByName(opDesc.opName);
    auto kernel = std::unique_ptr<Mki::Kernel>(op->GetKernelByName("NegF16Kernel"));
    ASSERT_NE(kernel, nullptr);
    ASSERT_EQ(kernel->CanSupport(launchParam), false);
}

/**
 * @brief outPutNum wrong
 */
TEST(TestOpElewiseNeg, TestCanSupport3)
{
    LaunchParam launchParam;
    launchParam.AddInTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    launchParam.AddOutTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    launchParam.AddOutTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    OpParam::Elewise opParam = {};
    opParam.elewiseType = OpParam::Elewise::ELEWISE_NEG;
    Mki::Test::UtOpDesc opDesc = {"ElewiseOperation", opParam};
    launchParam.SetParam(opDesc.specificParam);
    Mki::Operation *op = Mki::AutoGen::GetOpByName(opDesc.opName);
    auto kernel = std::unique_ptr<Mki::Kernel>(op->GetKernelByName("NegF16Kernel"));
    ASSERT_NE(kernel, nullptr);
    ASSERT_EQ(kernel->CanSupport(launchParam), false);
}

/**
 * @brief inTensor dtype wrong
 */
TEST(TestOpElewiseNeg, TestCanSupport4)
{
    LaunchParam launchParam;
    launchParam.AddInTensor({{TENSOR_DTYPE_FLOAT, TENSOR_FORMAT_ND, {5, 5}}});
    launchParam.AddOutTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    OpParam::Elewise opParam = {};
    opParam.elewiseType = OpParam::Elewise::ELEWISE_NEG;
    Mki::Test::UtOpDesc opDesc = {"ElewiseOperation", opParam};
    launchParam.SetParam(opDesc.specificParam);
    Mki::Operation *op = Mki::AutoGen::GetOpByName(opDesc.opName);
    auto kernel = std::unique_ptr<Mki::Kernel>(op->GetKernelByName("NegF16Kernel"));
    ASSERT_NE(kernel, nullptr);
    ASSERT_EQ(kernel->CanSupport(launchParam), false);
}

/**
 * @brief outTensor dtype wrong
 */
TEST(TestOpElewiseNeg, TestCanSupport5)
{
    LaunchParam launchParam;
    launchParam.AddInTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    launchParam.AddOutTensor({{TENSOR_DTYPE_FLOAT, TENSOR_FORMAT_ND, {5, 5}}});
    OpParam::Elewise opParam = {};
    opParam.elewiseType = OpParam::Elewise::ELEWISE_NEG;
    Mki::Test::UtOpDesc opDesc = {"ElewiseOperation", opParam};
    launchParam.SetParam(opDesc.specificParam);
    Mki::Operation *op = Mki::AutoGen::GetOpByName(opDesc.opName);
    auto kernel = std::unique_ptr<Mki::Kernel>(op->GetKernelByName("NegF16Kernel"));
    ASSERT_NE(kernel, nullptr);
    ASSERT_EQ(kernel->CanSupport(launchParam), false);
}

/**
 * @brief ok
 */
TEST(TestOpElewiseNeg, TestGetBestKernelNeg0)
{
    LaunchParam launchParam;
    launchParam.AddInTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    launchParam.AddOutTensor({{TENSOR_DTYPE_FLOAT16, TENSOR_FORMAT_ND, {5, 5}}});
    OpParam::Elewise opParam = {};
    opParam.elewiseType = OpParam::Elewise::ELEWISE_NEG;
    Mki::Test::UtOpDesc opDesc = {"ElewiseOperation", opParam};
    launchParam.SetParam(opDesc.specificParam);
    Mki::Operation *op = Mki::AutoGen::GetOpByName(opDesc.opName);
    auto kernel = std::unique_ptr<Mki::Kernel>(op->GetBestKernel(launchParam));
    ASSERT_NE(kernel, nullptr);
}

/**
 * @brief inTensor dtype wrong
 */
TEST(TestOpElewiseNeg, TestGetBestKernelNeg1)
{
    LaunchParam launchParam;
    launchParam.AddInTensor({{TENSOR_DTYPE_FLOAT, TENSOR_FORMAT_ND, {5, 5}}});
    launchParam.AddOutTensor({{TENSOR_DTYPE_FLOAT, TENSOR_FORMAT_ND, {5, 5}}});
    OpParam::Elewise opParam = {};
    opParam.elewiseType = OpParam::Elewise::ELEWISE_NEG;
    Mki::Test::UtOpDesc opDesc = {"ElewiseOperation", opParam};
    launchParam.SetParam(opDesc.specificParam);
    Mki::Operation *op = Mki::AutoGen::GetOpByName(opDesc.opName);
    auto kernel = std::unique_ptr<Mki::Kernel>(op->GetBestKernel(launchParam));
    ASSERT_EQ(kernel, nullptr);
}