为什么 GCC 不会自动矢量化这个循环？

Question

我有以下 C 程序（我的实际用例的简化，它表现出相同的行为）

#include <stdlib.h>
#include <math.h>
int main(int argc, char ** argv) {
    const float * __restrict__ const input = malloc(20000*sizeof(float));
    float * __restrict__ const output = malloc(20000*sizeof(float));

    unsigned int pos=0;
    while(1) {
            unsigned int rest=100;
            for(unsigned int i=pos;i<pos+rest; i++) {
                    output[i] = input[i] * 0.1;
            }

            pos+=rest;            
            if(pos>10000) {
                    break;
            }
    }
}

当我用

编译时

 -O3 -g -Wall -ftree-vectorizer-verbose=5 -msse -msse2 -msse3 -march=native -mtune=native --std=c99 -fPIC -ffast-math

我得到了输出

main.c:10: note: not vectorized: unhandled data-ref 

其中 10 是内部 for 循环的行。当我查找它为什么会这样说时，它似乎是在说指针可以别名，但它们不能在我的代码中，因为我有 __restrict 关键字。他们还建议包括 -msse 标志，但他们似乎也没有做任何事情。有什么帮助吗？

Answer 1

这确实看起来像一个错误。在下面的等效函数中，foo() 是矢量化的，但 bar() 不是，在为 x86-64 目标编译时：

void foo(const float * restrict input, float * restrict output)
{
    unsigned int pos;
    for (pos = 0; pos < 10100; pos++)
        output[pos] = input[pos] * 0.1;
}

void bar(const float * restrict input, float * restrict output)
{
    unsigned int pos;
    unsigned int i;
    for (pos = 0; pos <= 10000; pos += 100)
        for (i = 0; i < 100; i++)
            output[pos + i] = input[pos + i] * 0.1;
}

添加 -m32 标志，改为针对 x86 目标进行编译，会导致两个函数都被矢量化。

Answer 2

它不喜欢阻止它理解内循环的外循环格式。如果我把它折叠成一个循环，我可以让它矢量化：

#include <stdlib.h>
#include <math.h>
int main(int argc, char ** argv) {
    const float * __restrict__ input = malloc(20000*sizeof(float));
    float * __restrict__ output = malloc(20000*sizeof(float));

    for(unsigned int i=0; i<=10100; i++) {
            output[i] = input[i] * 0.1f;
    }
}

（请注意，我并没有想太多关于如何正确地将pos+rest限制转换为单个for循环条件，可能是错误的）

您可以通过将简化的内部循环放入您使用指针和计数调用的函数中来利用这一点。即使它再次内联，它也可以正常工作。这是假设您删除了我刚刚简化但您需要保留的 while() 循环的部分内容。

Answer 3

尝试：

const float * __restrict__ input = ...;
float * __restrict__ output = ...;

通过改变周围的事物进行一些实验：

#include <stdlib.h>
#include <math.h>

int main(int argc, char ** argv) {

    const float * __restrict__ input = new float[20000];
    float * __restrict__  output = new float[20000];

    unsigned int pos=0;
    while(1) {
        unsigned int rest=100;
        output += pos;
        input += pos;
        for(unsigned int i=0;i<rest; ++i) {
            output[i] = input[i] * 0.1;
        }

        pos+=rest;
        if(pos>10000) {
            break;
        }
    }
}

g++ -O3 -g -Wall -ftree-vectorizer-verbose=7 -msse -msse2 -msse3 -c test.cpp

test.cpp:14: note: versioning for alias required: can't determine dependence between *D.4096_24 and *D.4095_21
test.cpp:14: note: mark for run-time aliasing test between *D.4096_24 and *D.4095_21
test.cpp:14: note: Alignment of access forced using versioning.
test.cpp:14: note: Vectorizing an unaligned access.
test.cpp:14: note: vect_model_load_cost: unaligned supported by hardware.
test.cpp:14: note: vect_model_load_cost: inside_cost = 2, outside_cost = 0 .
test.cpp:14: note: vect_model_simple_cost: inside_cost = 2, outside_cost = 0 .
test.cpp:14: note: vect_model_simple_cost: inside_cost = 2, outside_cost = 1 .
test.cpp:14: note: vect_model_simple_cost: inside_cost = 1, outside_cost = 0 .
test.cpp:14: note: vect_model_store_cost: inside_cost = 1, outside_cost = 0 .
test.cpp:14: note: cost model: Adding cost of checks for loop versioning to treat misalignment.

test.cpp:14: note: cost model: Adding cost of checks for loop versioning aliasing.

test.cpp:14: note: Cost model analysis:
  Vector inside of loop cost: 8
  Vector outside of loop cost: 6
  Scalar iteration cost: 5
  Scalar outside cost: 1
  prologue iterations: 0
  epilogue iterations: 0
  Calculated minimum iters for profitability: 2

test.cpp:14: note:   Profitability threshold = 3

test.cpp:14: note: Vectorization may not be profitable.
test.cpp:14: note: create runtime check for data references *D.4096_24 and *D.4095_21
test.cpp:14: note: created 1 versioning for alias checks.

test.cpp:14: note: LOOP VECTORIZED.
test.cpp:4: note: vectorized 1 loops in function.

Compilation finished at Wed Feb 16 19:17:59