uint8 使用 SIMD Neon 內在函數浮動

Question

我正在嘗試優化將灰度圖像轉換為在 Neon A64/v8 上運行的浮動圖像的代碼。

當前的實現使用 OpenCV 的convertTo() （為 android 編譯convertTo()相當快，但這仍然是我們的瓶頸。

所以我想出了以下代碼，並想聽聽可能的改進。

如果有幫助，圖像高度和寬度是 16 的系數。

我正在運行for循環：

static void u8_2_f(unsigned char* in, float* out)
{
    //1 u8x8->u16x8
    uint8x8_t u8x8src = vld1_u8(in);
    uint16x8_t u16x8src = vmovl_u8(u8x8src);

    //2 u16x8 -> u32x4high, u32x4low
    uint32x4_t u32x4srch = vmovl_u16(vget_high_u16(u16x8src));
    uint32x4_t u32x4srcl = vmovl_u16(vget_low_u16(u16x8src));

    //3 u32x4high, u32x4low -> f32x4high, f32x4low
    vst1q_f32(out, vcvtq_f32_u32(u32x4srch));
    vst1q_f32(out+4, vcvtq_f32_u32(u32x4srcl));
}

Answer 1

為了可能的改進，嘗試用這個函數替換vcvtq_f32_u32 。 它是 2 條指令而不是 1 條指令，但它們在某些 CPU 上可能更快。

// Convert bytes to float, assuming the input is within [ 0 .. 0xFF ] interval
inline float32x4_t byteToFloat( uint32x4_t u32 )
{
    // Floats have 23 bits of mantissa.
    // We want least significant 8 bits to be shifted to [ 0 .. 255 ], therefore need to add 2^23
    // See this page for details: https://www.h-schmidt.net/FloatConverter/IEEE754.html
    // If you want output floats in [ 0 .. 255.0 / 256.0 ] interval, change into 2^15 = 0x47000000
    constexpr uint32_t offsetValue = 0x4b000000;
    // Check disassembly & verify your compiler has moved this initialization outside the loop
    const uint32x4_t offsetInt = vdupq_n_u32( offsetValue );
    // Bitwise is probably slightly faster than addition, delivers same results for our input
    u32 = vorrq_u32( u32, offsetInt );
    // The only FP operation required is subtraction, hopefully faster than UCVTF
    return vsubq_f32( vreinterpretq_f32_u32( u32 ), vreinterpretq_f32_u32( offsetInt ) );
}