如何重写这个code上证所内部函数 [英] How to rewrite this code to sse intrinsics

问题描述

林在上证所内部函数新会AP preciate在使用这种9as这一些提​​示，帮助尚未大雾给我）

Im new in sse intrinsics and would appreciate some hints assistance in using this 9as this is yet foggy to me)

我得到了这样的code

I got such code

for(int k=0; k<=n-4; k+=4) 
 { 

  int xc0 = 512 + ((idx + k*iddx)>>6); 
  int yc0 = 512 + ((idy + k*iddy)>>6); 

  int xc1 = 512 + ((idx + (k+1)*iddx)>>6); 
  int yc1 = 512 + ((idy + (k+1)*iddy)>>6); 

  int xc2 = 512 + ((idx + (k+2)*iddx)>>6); 
  int yc2 = 512 + ((idy + (k+2)*iddy)>>6); 

  int xc3 = 512 + ((idx + (k+3)*iddx)>>6); 
  int yc3 = 512 + ((idy + (k+3)*iddy)>>6); 

  unsigned color0 =  working_buffer[yc0*working_buffer_size_x + xc0]; 
  unsigned color1 =  working_buffer[yc1*working_buffer_size_x + xc1]; 
  unsigned color2 =  working_buffer[yc2*working_buffer_size_x + xc2]; 
  unsigned color3 =  working_buffer[yc3*working_buffer_size_x + xc3]; 

  int adr = base_adr + k; 

  frame_bitmap[adr]  = color0; 
  frame_bitmap[adr+1]= color1; 
  frame_bitmap[adr+2]= color2; 
  frame_bitmap[adr+3]= color3; 
 } 

都在这里为int /符号，这是循环的重要组成部分，不知道是否整数上证所将有助于在这里的速度，但不知道它是否会在所有的工作？可以someopne与此帮助？

all here is int/unsigned, this is critical part of the loop, not sure if integer sse would help here in speed but wonder if it would work at all? could someopne help with this?

（即时通讯使用的mingw32）

(im using mingw32)

推荐答案

我的SSE是有点生疏，但你应该做的是：

My sse is a bit rusty, but what you should do is:

xmm0: [k, k+1, k+2, k+3] //xc0, xc1,....
xmm1: [k, k+1, k+2, k+3] //yc0, yc1,....
//initialize before the loop
xmm2: [512, 512, 512, 512]
xmm3: [idx, idx, idx, idx]
xmm4: [iddx, iddx, iddx, iddx]
xmm5: [idy, idy, idy, idy]
xmm6: [iddy, iddy, iddy, iddy]
xmm7: [working_buffer_size_x, working_buffer_size_x, working_buffer_size_x, working_buffer_size_x]

计算：

xmm0 * xmm4
xmm0 + xmm3
xmm0 >> 6
xmm0 + xmm2

xmm0: [xc0, xc1, xc2, xc3]
///////////////////////////////

xmm1 * xmm6
xmm1 + xmm5
xmm1 >> 6
xmm1 + xmm2

xmm1: [yc0, yc1, yc2, yc3]

xmm1 * xmm7
xmm1 + xmm0

现在将xmm1 是：

xmm1: [yc0*working_buffer_size_x + xc0, yc1*working_buffer_size_x + xc1, yc2*working_buffer_size_x + xc2, yc3*working_buffer_size_x + xc3]

您正在阅读，并在每个循环写入存储器（working_buffer，frame_bitmap阵列），即比计算本身太慢的操作，所以当你有望成为速度的提升将不会这么大。

You are reading and writing memory in each loop (working_buffer, frame_bitmap arrays), operations that are way too slower than the calculations itself, so the speed improvement won't be as much as you expected to be.

修改

您需要的 working_buffer和frame_bitmap 的对齐阵列和的 SSE4.1 的：

You need working_buffer and frame_bitmap arrays to be aligned and SSE4.1:

#include <emmintrin.h>
#include <smmintrin.h> //SSE4.1

int a[4] __attribute__((aligned(16)));
__m128i xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7;

xmm2 = _mm_set1_epi32(512);
xmm3 = _mm_set1_epi32(idx);
xmm4 = _mm_set1_epi32(iddx);
xmm5 = _mm_set1_epi32(idy);
xmm6 = _mm_set1_epi32(iddy);
xmm7 = _mm_set1_epi32(working_buffer_size_x);

for(k = 0; k <= n - 4; k +=4){
    xmm0 = _mm_set_epi32(k + 3, k + 2, k + 1, k);
    xmm1 = _mm_set_epi32(k + 3, k + 2, k + 1, k);

    //xmm0 * xmm4
    xmm0 = _mm_mullo_epi32(xmm0, xmm4);

    //xmm0 + xmm3
    xmm0 = _mm_add_epi32(xmm0, xmm3);

    //xmm0 >> 6
    xmm0 = _mm_srai_epi32(xmm0, 6);

    //xmm0 + xmm2
    xmm0 = _mm_add_epi32(xmm0, xmm2);



    //xmm1 * xmm6
    xmm1 = _mm_mullo_epi32(xmm1, xmm6);

    //xmm1 + xmm5
    xmm1 = _mm_add_epi32(xmm1, xmm5);

    //xmm1 >> 6
    xmm1 = _mm_srai_epi32(xmm1, 6);

    //xmm1 + xmm2
    xmm1 = _mm_add_epi32(xmm1, xmm2);


    //xmm1 * xmm7
    xmm1 = _mm_mullo_epi32(xmm1, xmm7);
    //xmm1 + xmm0
    xmm1 = _mm_add_epi32(xmm1, xmm0);


    //a[0] = yc0*working_buffer_size_x + xc0
    //a[1] = yc1*working_buffer_size_x + xc1
    //a[2] = yc2*working_buffer_size_x + xc2
    //a[3] = yc3*working_buffer_size_x + xc3
    _mm_store_si128((__m128i *)&a[0], xmm1);

    unsigned color0 =  working_buffer[ a[0] ]; 
    unsigned color1 =  working_buffer[ a[1] ]; 
    unsigned color2 =  working_buffer[ a[2] ]; 
    unsigned color3 =  working_buffer[ a[3] ]; 

    int adr = base_adr + k; 

    frame_bitmap[adr]  = color0; 
    frame_bitmap[adr+1]= color1; 
    frame_bitmap[adr+2]= color2; 
    frame_bitmap[adr+3]= color3; 
}

您更可以通过避免 _mm_store_si128优化它（（__ m128i *）及一个[0]，将xmm1）; 或 INT ADR = base_adr + K; 使用汇编的内存直接操作

You can optimize it even more by avoiding the _mm_store_si128((__m128i *)&a[0], xmm1); or the int adr = base_adr + k; using assembly with direct manipulation of memory.

这篇关于如何重写这个code上证所内部函数的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！