* Copyright (c) 2025 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.
*/
#ifndef AICPU_KERNELS_NORMALIZED_CROSS_H_
#define AICPU_KERNELS_NORMALIZED_CROSS_H_
#include "cpu_kernel.h"
#include "utils/status.h"
#include "utils/bcast.h"
namespace aicpu {
class CrossCpuKernel : public CpuKernel {
public:
CrossCpuKernel() = default;
~CrossCpuKernel() override = default;
uint32_t Compute(CpuKernelContext &ctx) override;
private:
template <typename T1>
uint32_t CrossBcastCompute(const CpuKernelContext &ctx);
template <typename T1>
uint32_t CrossCompute(const CpuKernelContext &ctx, const BCalcInfo &calc_info);
KernelStatus GetDimAndCheck(const CpuKernelContext &ctx, const BCalcInfo &calc_info);
std::vector<int64_t> StridesCompute(int64_t stride_length,
std::vector<int64_t> data_dims);
void InitStride(const BCalcInfo &calc_info);
template <typename T1>
void SetResult(const BCalcInfo &calc_info, int64_t data_start_bcast) const;
template <typename T1>
void ParallelCompute(const BCalcInfo &calc_info, int64_t start, int64_t end);
int64_t dim_ = 0;
std::vector<int64_t> stride_;
int64_t data_stride_ = 0;
};
}
#endif