c - C中的for循环优化

Question

如何优化这个 for 循环:

for (l = 1; l <= loop; l++) {
    for (i = 1; i < n; i++) {
        x[i] = z[i] * (y[i] - x[i - 1]);
    }
}

如何通过 OpenMp 并行它的原始版本和优化版本？

Answer 1

假设您想要并行化内部循环

for ( i = 1; i < n; ++i ) {
    x[i] = z[i] * ( y[i] - x[i - 1] );
}

我建议预先计算不依赖于前一个循环的部分。这更容易并行化。

double preComps [n];
#pragma omp parallel for
for( i = 1; i < n ; ++i ) {
    preComps[i] = z[i] * y[i];
}

// this loop is difficult to parallelize because of the data dependency on what was computed in the previous loop
for( i = 1; i < n ; ++i ) {
    x[i] = preComps[i] - z[i] * x[i - 1];
}