#!/usr/bin/python3

# coding=utf-8

# --------------------------------------------------------------------------------

# 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.

# --------------------------------------------------------------------------------



import os

import numpy as np

from utils import NumExt

np.random.seed(19)



def gen_golden_data_tmax(case_name, param):

    dtype = param.dtype



    H, W = [param.tile_row, param.tile_col]

    h_valid, w_valid = [param.valid_row, param.valid_col]



    # Generate random input arrays

    input1 = NumExt.astype(np.random.randint(1, 10, size=[H, W]), dtype)

    input2 = NumExt.astype(np.random.randint(1, 10, size=[H, W]), dtype)



    # Perform the addbtraction

    golden = NumExt.astype(np.maximum(input1, input2), dtype)



    # Apply valid region constraints

    output = NumExt.zeros([H, W], dtype)

    for h in range(H):

        for w in range(W):

            if h >= h_valid or w >= w_valid:

                golden[h][w] = output[h][w]



    # Save the input and golden data to binary files

    NumExt.write_array("input1.bin", input1, dtype)

    NumExt.write_array("input2.bin", input2, dtype)

    NumExt.write_array("golden.bin", golden, dtype)



    return output, input1, input2, golden



class tmaxParams:

    def __init__(self, dtype, global_row, global_col, tile_row, tile_col, valid_row, valid_col):

        self.dtype = dtype

        self.global_row = global_row

        self.global_col = global_col

        self.tile_row = tile_row

        self.tile_col = tile_col

        self.valid_row = valid_row

        self.valid_col = valid_col



def generate_case_name(param):

    dtype_str = NumExt.get_short_type_name(param.dtype)

    return f"TMAXTest.case_{dtype_str}_{param.global_row}x{param.global_col}_{param.tile_row}x{param.tile_col}_{param.valid_row}x{param.valid_col}"



if __name__ == "__main__":

    # Get the absolute path of the script

    script_dir = os.path.dirname(os.path.abspath(__file__))

    testcases_dir = os.path.join(script_dir, "testcases")



    # Ensure the testcases directory exists

    if not os.path.exists(testcases_dir):

        os.makedirs(testcases_dir)



    case_params_list = [

        tmaxParams(np.float32, 64, 64, 64, 64, 64, 64),

        tmaxParams(np.int32, 64, 64, 64, 64, 64, 64),

        tmaxParams(np.int16, 64, 64, 64, 64, 64, 64),

        tmaxParams(np.float16, 16, 256, 16, 256, 16, 256),

    ]

    if os.getenv("PTO_CPU_SIM_ENABLE_BF16") == "1":

        case_params_list.append(tmaxParams(NumExt.bf16, 16, 256, 16, 256, 16, 256))



    for i, param in enumerate(case_params_list):

        case_name = generate_case_name(param)

        if not os.path.exists(case_name):

            os.makedirs(case_name)

        original_dir = os.getcwd()

        os.chdir(case_name)

        gen_golden_data_tmax(case_name, param)

        os.chdir(original_dir)