hom*_*omm 8 c performance 64-bit gcc
我注意到我的代码在64位Linux上运行比在32位Linux或64位Window或64位Mac上慢得多.这是最小的测试用例.
#include <stdlib.h>
typedef unsigned char UINT8;
void
stretch(UINT8 * lineOut, UINT8 * lineIn, int xsize, float *kk)
{
int xx, x;
for (xx = 0; xx < xsize; xx++) {
float ss = 0.0;
for (x = 0; x < xsize; x++) {
ss += lineIn[x] * kk[x];
}
lineOut[xx] = (UINT8) ss;
}
}
int
main( int argc, char** argv )
{
int i;
int xsize = 2048;
UINT8 *lineIn = calloc(xsize, sizeof(UINT8));
UINT8 *lineOut = calloc(xsize, sizeof(UINT8));
float *kk = calloc(xsize, sizeof(float));
for (i = 0; i < 1024; i++) {
stretch(lineOut, lineIn, xsize, kk);
}
return 0;
}
Run Code Online (Sandbox Code Playgroud)
它有如何运行:
$ cc --version
cc (Ubuntu 4.8.2-19ubuntu1) 4.8.2
$ cc -O2 -Wall -m64 ./tt.c -o ./tt && time ./tt
user 14.166s
$ cc -O2 -Wall -m32 ./tt.c -o ./tt && time ./tt
user 5.018s
Run Code Online (Sandbox Code Playgroud)
正如您所看到的,32位版本的运行速度快了近3倍(我在32位和64位Ubuntu上测试过,结果相同).更奇怪的是性能取决于C标准:
$ cc -O2 -Wall -std=c99 -m32 ./tt.c -o ./tt && time ./tt
user 15.825s
$ cc -O2 -Wall -std=gnu99 -m32 ./tt.c -o ./tt && time ./tt
user 5.090s
Run Code Online (Sandbox Code Playgroud)
怎么会这样?我该如何解决这个问题来加速GCC生成的64位版本.
更新1
我比较了快速32位(默认和gnu99)和慢速(c99)生成的汇编程序,发现如下:
.L5:
movzbl (%ebx,%eax), %edx # MEM[base: lineIn_10(D), index: _72, offset: 0B], D.1543
movl %edx, (%esp) # D.1543,
fildl (%esp) #
fmuls (%esi,%eax,4) # MEM[base: kk_18(D), index: _72, step: 4, offset: 0B]
addl $1, %eax #, x
cmpl %ecx, %eax # xsize, x
faddp %st, %st(1) #,
fstps 12(%esp) #
flds 12(%esp) #
jne .L5 #,
Run Code Online (Sandbox Code Playgroud)
在快速的情况下没有fstps和flds命令.所以GCC在每一步都存储并加载内存中的值.我尝试过register float类型,但这没有用.
更新2
我已经在gcc-4.9上进行了测试,看起来它为64位生成了最佳代码.并且-ffast-math(由@jch建议)修复-m32 -std=c99了两个GCC版本.我仍然在gcc-4.8上寻找64位的解决方案,因为它现在更常见的是4.9.
旧版本的GCC生成的代码中存在部分依赖性停顿.
movzbl (%rsi,%rax), %r8d
cvtsi2ss %r8d, %xmm0 ;; all upper bits in %xmm0 are false dependency
Run Code Online (Sandbox Code Playgroud)
依赖性可以被打破xorps.
#ifdef __SSE__
float __attribute__((always_inline)) i2f(int v) {
float x;
__asm__("xorps %0, %0; cvtsi2ss %1, %0" : "=x"(x) : "r"(v) );
return x;
}
#else
float __attribute__((always_inline)) i2f(int v) { return (float) v; }
#endif
void stretch(UINT8* lineOut, UINT8* lineIn, int xsize, float *kk)
{
int xx, x;
for (xx = 0; xx < xsize; xx++) {
float ss = 0.0;
for (x = 0; x < xsize; x++) {
ss += i2f(lineIn[x]) * kk[x];
}
lineOut[xx] = (UINT8) ss;
}
}
Run Code Online (Sandbox Code Playgroud)
结果
$ cc -O2 -Wall -m64 ./test.c -o ./test64 && time ./test64
./test64 4.07s user 0.00s system 99% cpu 4.070 total
$ cc -O2 -Wall -m32 ./test.c -o ./test32 && time ./test32
./test32 3.94s user 0.00s system 99% cpu 3.938 total
Run Code Online (Sandbox Code Playgroud)