trying optimize doitgen

polybench-c-4.2.1-beta/linear-algebra/kernels/doitgen/doitgen_optimized.c

@@ -70,8 +70,8 @@ void kernel_doitgen(int nr, int nq, int np,
   int r, q, p, s;
 
 #pragma scop
-#pragma omp for
   for (q = 0; q < _PB_NQ; q++)
+#pragma omp for
     for (r = 0; r < _PB_NR; r++)
     {
       for (p = 0; p < _PB_NP; p++)