/**
 * 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.
 */

/*!
 * \file dynamic_mla.h
 * \brief
 */

#pragma once
#ifndef MLA_DYNAMIC
#define MLA_DYNAMIC

#include "interface/inner/pre_def.h"
#include "tilefwk/tilefwk.h"
#include "interface/inner/tilefwk.h"

namespace npu::tile_fwk {

struct MlaQuantInputs {
    Tensor dequantScaleX;
    Tensor dequantScaleWDq;
    Tensor dequantScaleWUqQr;
    Tensor dequantScaleWDkvKr;
    Tensor quantScaleCkv;
    Tensor quantScaleCkr;
    Tensor smoothScalesCq;
};

struct SimpleParams {
    int b;
    int s;
    int s2;
    int d;
    int m;
    int k;
    int n;
    int n2;
    int right;
    int h;
    int q_lora_rank;
    int kv_lora_rank;
    int qk_rope_head_dim;
    int qk_nope_head_dim;
    int q_head_dim;
    std::string cacheMode;
    int blockSize;
    std::vector<int> vecTile;
    std::vector<int> cubeMTile;
    std::vector<int> cubeKTile;
    std::vector<int> cubeNTile;
    int tileB;
    static SimpleParams getCommonParams()
    {
        SimpleParams params;
        params.n2 = 1;
        params.s = 1;
        params.h = 7168;               // 7168
        params.q_lora_rank = 1536;     // 1536
        params.kv_lora_rank = 512;     // 512
        params.qk_rope_head_dim = 64;  // 64
        params.qk_nope_head_dim = 128; // 128
        params.q_head_dim = params.qk_rope_head_dim + params.qk_nope_head_dim;
        params.cacheMode = "BNSD";
        params.blockSize = 128; // 128
        return params;
    }

    static SimpleParams getLowParams()
    {
        SimpleParams params = getCommonParams();
        params.b = 4;    // 4
        params.n = 32;   // 32
        params.s2 = 256; // 256
        return params;
    }

    static SimpleParams getHighParams()
    {
        SimpleParams params = getCommonParams();
        params.b = 32;    // 32
        params.n = 128;   // 128
        params.s2 = 4096; // 4096
        return params;
    }
};

std::vector<Tensor> mlaPre(
    const Tensor& tokenX, const Tensor& wDq, const Tensor& wUqQr, const Tensor& wDkvKr, const Tensor& gammaCq,
    float epsilonCq, const MlaQuantInputs& quantInputs, bool splitK = false, bool isSmooth = true);

void MlaProlog(
    const Tensor& tokenX, const Tensor& wDq, const Tensor& wUqQr, const Tensor& wUk, const Tensor& wDkvKr,
    const Tensor& gammaCq, const Tensor& gammaCkv, const Tensor& sin, const Tensor& cos, const Tensor& cacheIndex,
    Tensor& kvCache, Tensor& krCache, const MlaQuantInputs& quantInputs, const RoPETileShapeConfigNew& ropeConfig,
    Tensor& queryOut, Tensor& queryRopeOut, Tensor& kvCacheOut, Tensor& krCacheOut, float epsilonCq = 1e-5f,
    float epsilonCkv = 1e-5f, std::string cacheMode = "BNSD", bool splitK = false, bool isSmooth = true);

// tile config
struct MlaTileConfig {
    int tileB = 8; // tileB is 8
    int tileS = 1;
};

std::vector<Tensor> PreCompute(
    const Tensor& tokenX, const Tensor& wDq, const Tensor& wUqQr, const Tensor& wDkvKr, const Tensor& gammaCq,
    float epsilonCq, const MlaQuantInputs& quantInputs);

void MlaPrologCompute(
    const Tensor& tokenX, const Tensor& wDq, const Tensor& wUqQr, const Tensor& wUk, const Tensor& wDkvKr,
    const Tensor& gammaCq, const Tensor& gammaCkv, const Tensor& sin, const Tensor& cos, const Tensor& cacheIndex,
    Tensor& kvCache, Tensor& krCache, const MlaQuantInputs& quantInputs, const MlaTileConfig& tileConfig,
    Tensor& queryOut, Tensor& queryRopeOut, Tensor& kvCacheOut, Tensor& krCacheOut, float epsilonCq, float epsilonCkv,
    std::string cacheMode);

void MlaProlog(
    const Tensor& tokenX, const Tensor& wDq, const Tensor& wUqQr, const Tensor& wUk, const Tensor& wDkvKr,
    const Tensor& gammaCq, const Tensor& gammaCkv, const Tensor& sin, const Tensor& cos, const Tensor& cacheIndex,
    Tensor& kvCache, Tensor& krCache, const MlaQuantInputs& quantInputs, const MlaTileConfig& tileConfig,
    Tensor& queryOut, Tensor& queryRopeOut, Tensor& kvCacheOut, Tensor& krCacheOut, float epsilonCq = 1e-5f,
    float epsilonCkv = 1e-5f, std::string cacheMode = "PA_NZ");

} // namespace npu::tile_fwk

#endif // MLA_DYNAMIC