yarpgen support pragma omp simd

[huangcj123456]

I used the Yargen tool and found a bug in the tool.

wrong code：
void test(int *acc, int *inc) {
#pragma omp simd
for (int i = 0; i < 16; i++)
*acc += *inc;
}

OK code：
void test(int *acc, int *inc) {
#pragma omp simd reduction(+:acc[0])
for (int i = 0; i < 16; i++)
*acc += *inc;
}
The absence of the "reduction" keyword leads to undefined behavior in this test case.

Who knows how to solve this problem?

[dbabokin]

Could you please elaborate why is this UB? I'm not an expert in OpenMP, so pointers to standard would help.

[huangcj123456]

SIMD Reduction:

#pragma omp for simd reduction(+:local_sum) is applied to the loop. This tells the compiler to vectorize the loop and also to perform a reduction operation on local_sum across different iterations assigned to different threads.
The reduction is done with the addition operator +, which combines local_sum from all threads into the final result.

[dbabokin]

But why is it UB without reduction?