/* ************************************************************************
 * Copyright 2016-2021 Advanced Micro Devices, Inc.
 *
 * ************************************************************************ */

#include <fstream>
#include <iostream>
#include <stdlib.h>
#include <vector>

#include "testing_common.hpp"

using namespace std;

template <typename T>
hipblasStatus_t testing_geqrf_strided_batched(const Arguments& argus)
{
    using U      = real_t<T>;
    bool FORTRAN = argus.fortran;
    auto hipblasGeqrfStridedBatchedFn
        = FORTRAN ? hipblasGeqrfStridedBatched<T, true> : hipblasGeqrfStridedBatched<T, false>;

    int    M            = argus.M;
    int    N            = argus.N;
    int    lda          = argus.lda;
    int    batch_count  = argus.batch_count;
    double stride_scale = argus.stride_scale;

    hipblasStride strideA   = lda * N * stride_scale;
    hipblasStride strideP   = min(M, N) * stride_scale;
    int           A_size    = strideA * batch_count;
    int           Ipiv_size = strideP * batch_count;

    // Check to prevent memory allocation error
    if(M < 0 || N < 0 || lda < M || batch_count < 0)
    {
        return HIPBLAS_STATUS_INVALID_VALUE;
    }
    if(batch_count == 0)
    {
        return HIPBLAS_STATUS_SUCCESS;
    }

    // Naming: dK is in GPU (device) memory. hK is in CPU (host) memory
    host_vector<T> hA(A_size);
    host_vector<T> hA1(A_size);
    host_vector<T> hIpiv(Ipiv_size);
    host_vector<T> hIpiv1(Ipiv_size);
    int            info;

    device_vector<T> dA(A_size);
    device_vector<T> dIpiv(Ipiv_size);

    double             gpu_time_used, hipblas_error;
    hipblasLocalHandle handle(argus);

    // Initial hA on CPU
    srand(1);
    for(int b = 0; b < batch_count; b++)
    {
        T* hAb = hA.data() + b * strideA;

        hipblas_init<T>(hAb, M, N, lda);

        // scale A to avoid singularities
        for(int i = 0; i < M; i++)
        {
            for(int j = 0; j < N; j++)
            {
                if(i == j)
                    hAb[i + j * lda] += 400;
                else
                    hAb[i + j * lda] -= 4;
            }
        }
    }

    // Copy data from CPU to device
    CHECK_HIP_ERROR(hipMemcpy(dA, hA.data(), A_size * sizeof(T), hipMemcpyHostToDevice));
    CHECK_HIP_ERROR(hipMemset(dIpiv, 0, Ipiv_size * sizeof(T)));

    /* =====================================================================
           HIPBLAS
    =================================================================== */
    CHECK_HIPBLAS_ERROR(hipblasGeqrfStridedBatchedFn(
        handle, M, N, dA, lda, strideA, dIpiv, strideP, &info, batch_count));

    // Copy output from device to CPU
    CHECK_HIP_ERROR(hipMemcpy(hA1.data(), dA, A_size * sizeof(T), hipMemcpyDeviceToHost));
    CHECK_HIP_ERROR(hipMemcpy(hIpiv1.data(), dIpiv, Ipiv_size * sizeof(T), hipMemcpyDeviceToHost));

    if(argus.unit_check || argus.norm_check)
    {
        /* =====================================================================
           CPU LAPACK
        =================================================================== */

        // Workspace query
        host_vector<T> work(1);
        cblas_geqrf(M, N, hA.data(), lda, hIpiv.data(), work.data(), -1);
        int lwork = type2int(work[0]);

        // Perform factorization
        work = host_vector<T>(lwork);
        for(int b = 0; b < batch_count; b++)
        {
            cblas_geqrf(
                M, N, hA.data() + b * strideA, lda, hIpiv.data() + b * strideP, work.data(), N);
        }

        double e1 = norm_check_general<T>('F', M, N, lda, strideA, hA, hA1, batch_count);
        double e2 = norm_check_general<T>(
            'F', min(M, N), 1, min(M, N), strideP, hIpiv, hIpiv1, batch_count);
        hipblas_error = e1 + e2;

        if(argus.unit_check)
        {
            U      eps       = std::numeric_limits<U>::epsilon();
            double tolerance = eps * 2000;

            unit_check_error(e1, tolerance);
            unit_check_error(e2, tolerance);
        }
    }

    if(argus.timing)
    {
        hipStream_t stream;
        CHECK_HIPBLAS_ERROR(hipblasGetStream(handle, &stream));

        int runs = argus.cold_iters + argus.iters;
        for(int iter = 0; iter < runs; iter++)
        {
            if(iter == argus.cold_iters)
                gpu_time_used = get_time_us_sync(stream);

            CHECK_HIPBLAS_ERROR(hipblasGeqrfStridedBatchedFn(
                handle, M, N, dA, lda, strideA, dIpiv, strideP, &info, batch_count));
        }
        gpu_time_used = get_time_us_sync(stream) - gpu_time_used;

        ArgumentModel<e_M, e_N, e_lda, e_stride_a, e_batch_count>{}.log_args<T>(
            std::cout,
            argus,
            gpu_time_used,
            geqrf_gflop_count<T>(N, M),
            ArgumentLogging::NA_value,
            hipblas_error);
    }

    return HIPBLAS_STATUS_SUCCESS;
}