ncnn/src/layer/arm/batchnorm_arm.cpp

// Tencent is pleased to support the open source community by making ncnn available.
//
// Copyright (C) 2017 THL A29 Limited, a Tencent company. All rights reserved.
//
// Licensed under the BSD 3-Clause License (the "License"); you may not use this file except
// in compliance with the License. You may obtain a copy of the License at
//
// https://opensource.org/licenses/BSD-3-Clause
//
// Unless required by applicable law or agreed to in writing, software distributed
// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR
// CONDITIONS OF ANY KIND, either express or implied. See the License for the
// specific language governing permissions and limitations under the License.

#include "batchnorm_arm.h"

#if __ARM_NEON
#include <arm_neon.h>
#endif // __ARM_NEON

namespace ncnn {

DEFINE_LAYER_CREATOR(BatchNorm_arm)

int BatchNorm_arm::forward(const Mat& bottom_blob, Mat& top_blob) const
{
    // a = bias - slope * mean / sqrt(var)
    // b = slope / sqrt(var)
    // value = b * value + a

    int w = bottom_blob.w;
    int h = bottom_blob.h;
    int size = w * h;

    top_blob.create(w, h, channels);
    if (top_blob.empty())
        return -100;

    const float* a_data_ptr = a_data;
    const float* b_data_ptr = b_data;
    #pragma omp parallel for
    for (int q=0; q<channels; q++)
    {
        const float* ptr = bottom_blob.channel(q);
        float* outptr = top_blob.channel(q);

        float a = a_data_ptr[q];
        float b = b_data_ptr[q];

#if __ARM_NEON
        int nn = size >> 2;
        int remain = size - (nn << 2);
#else
        int remain = size;
#endif // __ARM_NEON

#if __ARM_NEON
#if __aarch64__
        float32x4_t _a = vdupq_n_f32(a);
        float32x4_t _b = vdupq_n_f32(b);
        for (; nn>0; nn--)
        {
            float32x4_t _p = vld1q_f32(ptr);
            float32x4_t _outp = _a;
            _outp = vfmaq_f32(_outp, _p, _b);
            vst1q_f32(outptr, _outp);

            ptr += 4;
            outptr += 4;
        }
#else
        if (nn > 0)
        {
        asm volatile(
            "vdup.f32   q1, %6              \n"
            "vdup.f32   q2, %7              \n"
            "0:                             \n"
            "pld        [%1, #128]          \n"
            "vld1.f32   {d0-d1}, [%1 :128]! \n"
            "vorr.32    q3, q1, q1          \n"
            "vmla.f32   q3, q0, q2          \n"
            "subs       %0, #1              \n"
            "vst1.f32   {d6-d7}, [%2 :128]! \n"
            "bne        0b                  \n"
            : "=r"(nn),     // %0
              "=r"(ptr),    // %1
              "=r"(outptr)  // %2
            : "0"(nn),
              "1"(ptr),
              "2"(outptr),
              "r"(a),       // %6
              "r"(b)        // %7
            : "cc", "memory", "q0", "q1", "q2", "q3"
        );
        }
#endif // __aarch64__
#endif // __ARM_NEON
        for (; remain>0; remain--)
        {
            *outptr = b * *ptr + a;

            ptr++;
            outptr++;
        }
    }

    return 0;
}

int BatchNorm_arm::forward_inplace(Mat& bottom_top_blob) const
{
    // a = bias - slope * mean / sqrt(var)
    // b = slope / sqrt(var)
    // value = b * value + a

    int w = bottom_top_blob.w;
    int h = bottom_top_blob.h;
    int size = w * h;

    const float* a_data_ptr = a_data;
    const float* b_data_ptr = b_data;
    #pragma omp parallel for
    for (int q=0; q<channels; q++)
    {
        float* ptr = bottom_top_blob.channel(q);

        float a = a_data_ptr[q];
        float b = b_data_ptr[q];

#if __ARM_NEON
        int nn = size >> 2;
        int remain = size - (nn << 2);
#else
        int remain = size;
#endif // __ARM_NEON

#if __ARM_NEON
#if __aarch64__
        float32x4_t _a = vdupq_n_f32(a);
        float32x4_t _b = vdupq_n_f32(b);
        for (; nn>0; nn--)
        {
            float32x4_t _p = vld1q_f32(ptr);
            float32x4_t _outp = _a;
            _outp = vfmaq_f32(_outp, _p, _b);
            vst1q_f32(ptr, _outp);

            ptr += 4;
        }
#else
        if (nn > 0)
        {
        asm volatile(
            "vdup.f32   q1, %4              \n"
            "vdup.f32   q2, %5              \n"
            "0:                             \n"
            "pld        [%1, #128]          \n"
            "vld1.f32   {d0-d1}, [%1 :128]  \n"
            "vorr.32    q3, q1, q1          \n"
            "vmla.f32   q3, q0, q2          \n"
            "subs       %0, #1              \n"
            "vst1.f32   {d6-d7}, [%1 :128]! \n"
            "bne        0b                  \n"
            : "=r"(nn),     // %0
              "=r"(ptr)     // %1
            : "0"(nn),
              "1"(ptr),
              "r"(a),       // %4
              "r"(b)        // %5
            : "cc", "memory", "q0", "q1", "q2", "q3"
        );
        }
#endif // __aarch64__
#endif // __ARM_NEON
        for (; remain>0; remain--)
        {
            *ptr = b * *ptr + a;

            ptr++;
        }
    }

    return 0;
}

} // namespace ncnn