| 1 |
/* { dg-do compile } */ |
| 2 |
/* { dg-options "-O2 -march=x86-64-v3 -mavx10.2" } */ |
| 3 |
/* { dg-final { scan-assembler-times "vminmaxbf16\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */ |
| 4 |
/* { dg-final { scan-assembler-times "vminmaxbf16\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 5 |
/* { dg-final { scan-assembler-times "vminmaxbf16\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 6 |
/* { dg-final { scan-assembler-times "vminmaxph\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+(?:\n|\[ \\t\]+#)" 2 } } */ |
| 7 |
/* { dg-final { scan-assembler-times "vminmaxph\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 2 } } */ |
| 8 |
/* { dg-final { scan-assembler-times "vminmaxph\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 2 } } */ |
| 9 |
/* { dg-final { scan-assembler-times "vminmaxps\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+(?:\n|\[ \\t\]+#)" 2 } } */ |
| 10 |
/* { dg-final { scan-assembler-times "vminmaxps\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 2 } } */ |
| 11 |
/* { dg-final { scan-assembler-times "vminmaxps\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 2 } } */ |
| 12 |
/* { dg-final { scan-assembler-times "vminmaxpd\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+(?:\n|\[ \\t\]+#)" 2 } } */ |
| 13 |
/* { dg-final { scan-assembler-times "vminmaxpd\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 2 } } */ |
| 14 |
/* { dg-final { scan-assembler-times "vminmaxpd\[ \\t\]+\[^\{\n\]*\[^\}\]%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 2 } } */ |
| 15 |
|
| 16 |
|
| 17 |
#include <immintrin.h> |
| 18 |
|
| 19 |
volatile __m512bh x1; |
| 20 |
volatile __m512h x2; |
| 21 |
volatile __m512 x3; |
| 22 |
volatile __m512d x4; |
| 23 |
volatile __mmask32 m32; |
| 24 |
volatile __mmask16 m16; |
| 25 |
volatile __mmask8 m8; |
| 26 |
|
| 27 |
void extern |
| 28 |
avx10_2_test (void) |
| 29 |
{ |
| 30 |
x1 = _mm512_minmax_pbh (x1, x1, 100); |
| 31 |
x1 = _mm512_mask_minmax_pbh (x1, m32, x1, x1, 100); |
| 32 |
x1 = _mm512_maskz_minmax_pbh (m32, x1, x1, 100); |
| 33 |
x2 = _mm512_minmax_ph (x2, x2, 1); |
| 34 |
x2 = _mm512_mask_minmax_ph (x2, m32, x2, x2, 1); |
| 35 |
x2 = _mm512_maskz_minmax_ph (m32, x2, x2, 1); |
| 36 |
x2 = _mm512_minmax_round_ph (x2, x2, 1, 4); |
| 37 |
x2 = _mm512_mask_minmax_round_ph (x2, m32, x2, x2, 1, 4); |
| 38 |
x2 = _mm512_maskz_minmax_round_ph (m32, x2, x2, 1, 4); |
| 39 |
x3 = _mm512_minmax_ps (x3, x3, 1); |
| 40 |
x3 = _mm512_mask_minmax_ps (x3, m16, x3, x3, 1); |
| 41 |
x3 = _mm512_maskz_minmax_ps (m16, x3, x3, 1); |
| 42 |
x3 = _mm512_minmax_round_ps (x3, x3, 1, 4); |
| 43 |
x3 = _mm512_mask_minmax_round_ps (x3, m16, x3, x3, 1, 4); |
| 44 |
x3 = _mm512_maskz_minmax_round_ps (m16, x3, x3, 1, 4); |
| 45 |
x4 = _mm512_minmax_pd (x4, x4, 100); |
| 46 |
x4 = _mm512_mask_minmax_pd (x4, m8, x4, x4, 100); |
| 47 |
x4 = _mm512_maskz_minmax_pd (m8, x4, x4, 100); |
| 48 |
x4 = _mm512_minmax_round_pd (x4, x4, 100, 4); |
| 49 |
x4 = _mm512_mask_minmax_round_pd (x4, m8, x4, x4, 100, 4); |
| 50 |
x4 = _mm512_maskz_minmax_round_pd (m8, x4, x4, 100, 4); |
| 51 |
} |