gcc/testsuite/gcc.dg/vect/vect-98-big-array.c

   1 /* { dg-require-effective-target vect_int } */
   2 /* { dg-additional-options "-fdump-tree-optimized-details-blocks" } */
   3
   4 #include <stdarg.h>
   5 #include "tree-vect.h"
   6
   7 #define N 16
   8 #define DOT16( a, b)   ( a[0]*b[0]   + a[1]*b[1]   + a[2]*b[2]   + a[3]*b[3] + \
   9                          a[4]*b[4]   + a[5]*b[5]   + a[6]*b[6]   + a[7]*b[7] + \
  10                          a[8]*b[8]   + a[9]*b[9]   + a[10]*b[10] + a[11]*b[11] + \
  11                          a[12]*b[12] + a[13]*b[13] + a[14]*b[14] + a[15]*b[15])
  12
  13 __attribute__ ((noinline))
  14 int main1 (int ia[][N])
  15 {
  16   int i, j;
  17   int ib[N] = {0,3,6,9};
  18   int ic[N][N];
  19
  20   for (i = 0; i < N; i++)
  21     {
  22         ic[0][i] = DOT16 (ia[i], ib);
  23     }
  24
  25   /* check results: */
  26 #pragma GCC novector
  27   for (i = 0; i < N; i++)
  28     {
  29        if (ic[0][i] != DOT16 (ia[i], ib))
  30            abort ();
  31     }
  32
  33   return 0;
  34 }
  35
  36 int main (void)
  37 {
  38   int ia[N][N];
  39   int i,j;
  40   for (i = 0; i < N; i++)
  41     for (j = 0; j < N; j++)
  42       {
  43         ia[i][j] = i + j + 1;
  44         asm volatile ("" ::: "memory");
  45       }
  46
  47   check_vect ();
  48
  49   return main1 (ia);
  50 }
  51
  52 /* Needs interleaving support.  */
  53 /* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" { target vect_strided4 } } } */
  54 /* { dg-final { scan-tree-dump-times "vectorized 1 loops" 0 "vect" { xfail  vect_strided4 } } } */
  55 /* { dg-final { scan-tree-dump-not "Invalid sum" "optimized" } } */