| 1 |
/* { dg-do run } */ |
| 2 |
/* { dg-require-effective-target avx512f } */ |
| 3 |
/* { dg-options "-O3 -mfma -save-temps -mavx512f -mprefer-vector-width=512" } */ |
| 4 |
|
| 5 |
#include "fma-check.h" |
| 6 |
void __attribute__((noipa)) |
| 7 |
check_fmaddsub (double * __restrict a, double *b, double *c, int n) |
| 8 |
{ |
| 9 |
for (int i = 0; i < n; ++i) |
| 10 |
{ |
| 11 |
a[8*i + 0] = b[8*i + 0] * c[8*i + 0] - a[8*i + 0]; |
| 12 |
a[8*i + 1] = b[8*i + 1] * c[8*i + 1] + a[8*i + 1]; |
| 13 |
a[8*i + 2] = b[8*i + 2] * c[8*i + 2] - a[8*i + 2]; |
| 14 |
a[8*i + 3] = b[8*i + 3] * c[8*i + 3] + a[8*i + 3]; |
| 15 |
a[8*i + 4] = b[8*i + 4] * c[8*i + 4] - a[8*i + 4]; |
| 16 |
a[8*i + 5] = b[8*i + 5] * c[8*i + 5] + a[8*i + 5]; |
| 17 |
a[8*i + 6] = b[8*i + 6] * c[8*i + 6] - a[8*i + 6]; |
| 18 |
a[8*i + 7] = b[8*i + 7] * c[8*i + 7] + a[8*i + 7]; |
| 19 |
} |
| 20 |
} |
| 21 |
|
| 22 |
static void |
| 23 |
fma_test (void) |
| 24 |
{ |
| 25 |
if (!__builtin_cpu_supports ("avx512f")) |
| 26 |
return; |
| 27 |
double a[8], b[8], c[8]; |
| 28 |
for (int i = 0; i < 8; ++i) |
| 29 |
{ |
| 30 |
a[i] = i; |
| 31 |
b[i] = 3*i; |
| 32 |
c[i] = 7*i; |
| 33 |
} |
| 34 |
check_fmaddsub (a, b, c, 1); |
| 35 |
const double d[8] = { 0., 22., 82., 192., 332., 530., 750., 1036.}; |
| 36 |
for (int i = 0; i < 8; ++i) |
| 37 |
if (a[i] != d[i]) |
| 38 |
__builtin_abort (); |
| 39 |
} |
| 40 |
|
| 41 |
/* { dg-final { scan-assembler {(?n)fmaddsub...pd[ \t].*%zmm[0-9]} } } */ |