/*

* Simd Library (http://ermig1979.github.io/Simd).

*

* Copyright (c) 2011-2024 Yermalayeu Ihar.

*

* Permission is hereby granted, free of charge, to any person obtaining a copy

* of this software and associated documentation files (the "Software"), to deal

* in the Software without restriction, including without limitation the rights

* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell

* copies of the Software, and to permit persons to whom the Software is

* furnished to do so, subject to the following conditions:

*

* The above copyright notice and this permission notice shall be included in

* all copies or substantial portions of the Software.

*

* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR

* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,

* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE

* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER

* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,

* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE

* SOFTWARE.

*/

#include "Simd/SimdSynetConvolution32f.h"

#include "Simd/SimdSynetConvolution32fCommon.h"

#include "Simd/SimdSynet.h"

#include "Simd/SimdBase.h"

#include "Simd/SimdCpu.h"

namespace Simd

{

#if defined(SIMD_SYNET_ENABLE)

    namespace Base

    {

        static void ConvolutionNhwcGroupedBlock1x2(const float* src, const ConvParam& p, const float* weight, const float* bias, const float* params, float* dst)

        {

            size_t dW = p.kernelY * p.kernelX * p.srcC, srcC = p.srcC;

            for (size_t dy = 0; dy < p.dstH; ++dy)

            {

                for (size_t dx = 0; dx < p.dstW; ++dx)

                {

                    memset(dst, 0, p.dstC * sizeof(float));

                    for (size_t ky = 0; ky < p.kernelY; ++ky)

                    {

                        size_t sy = dy * p.strideY + ky * p.dilationY - p.padY;

                        if (sy < p.srcH)

                        {

                            for (size_t kx = 0; kx < p.kernelX; ++kx)

                            {

                                size_t sx = dx * p.strideX + kx * p.dilationX - p.padX;

                                if (sx < p.srcW)

                                {

                                    const float* pw0 = weight + (ky * p.kernelX + kx) * srcC, *pw1 = pw0 + dW;

                                    const float* ps = src + (sy * p.srcW + sx) * p.srcC;

                                    float* pd = dst;

                                    for (size_t c = 0; c < srcC; ++c, pd += 2)

                                    {

                                        pd[0] += ps[c] * pw0[c];

                                        pd[1] += ps[c] * pw1[c];

                                    }

                                }

                            }

                        }

                    }

                    ConvolutionBiasAndActivation(bias, p.dstC, 1, p.activation, params, ::SimdTrue, dst);

                    dst += p.dstC;

                }

            }

        }

        SynetConvolution32fNhwcGroupedBlock1x2::SynetConvolution32fNhwcGroupedBlock1x2(const ConvParam& p)

            : SynetConvolution32f(p)

        {

            _batch = p.batch;

            _sizeS = p.srcC * p.srcH * p.srcW;

            _sizeD = p.dstC * p.dstH * p.dstW;

            _convolution = ConvolutionNhwcGroupedBlock1x2;

        }

        void SynetConvolution32fNhwcGroupedBlock1x2::SetParams(const float* weight, SimdBool* internal, const float* bias, const float* params)

        {

            SynetConvolution32f::SetParams(weight, internal, bias, params);

            const ConvParam& p = _param;

            size_t size = p.kernelY * p.kernelX * p.srcC;

            _rWeight.Resize(size * 2);

            const float* src = _weight;

            float* dst0 = _rWeight.data, *dst1 = dst0 + size;

            for (size_t i = 0; i < size; ++i)

            {

                dst0[i] = src[0];

                dst1[i] = src[1];

                src += 2;

            }

            _weight = _rWeight.data;

            if (_bias == NULL)

            {

                _rBias.Resize(p.dstC, true);

                _bias = _rBias.data;

            }

        }

        void SynetConvolution32fNhwcGroupedBlock1x2::Forward(const float* src, float* buf, float* dst)

        {

            for (size_t b = 0; b < _batch; ++b)

            {

                _convolution(src, _param, _weight, _bias, _params, dst);

                src += _sizeS;

                dst += _sizeD;

            }

        }

        bool SynetConvolution32fNhwcGroupedBlock1x2::Preferable(const ConvParam& p)

        {

            if (p.trans == 0 || p.group == 1 || p.IsDepthwise())

                return false;

            return p.group == p.srcC && p.dstC == 2 * p.srcC;

        }

    }

#endif

}