| 1 |
/* { dg-do compile } */ |
| 2 |
/* { dg-options "-mavx512bw -mavx512vl -O2" } */ |
| 3 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*%zmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 4 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 5 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 6 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*%zmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 7 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 8 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 9 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*\\)\[^\n\]*%zmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 10 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*\\)\[^\n\]*%ymm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 11 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*\\)\[^\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 12 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*\\)\[^\n\]*%zmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 13 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*\\)\[^\n\]*%ymm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 14 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*\\)\[^\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 15 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*\\)\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 16 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*\\)\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 17 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*\\)\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ |
| 18 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*\\)\[^\n\]*%zmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */ |
| 19 |
/* { dg-final { scan-assembler-times "(?:vmovdqu16|vinserti128)\[ \\t\]+\[^\{\n\]*\\)\[^\n\]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */ |
| 20 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*\\)\[^\n\]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */ |
| 21 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*\\)(?:\n|\[ \\t\]+#)" 1 } } */ |
| 22 |
/* { dg-final { scan-assembler-times "(?:vmovdqu16|vextracti128)\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*\\)(?:\n|\[ \\t\]+#)" 1 } } */ |
| 23 |
/* { dg-final { scan-assembler-times "vmovdqu16\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*\\)(?:\n|\[ \\t\]+#)" 1 } } */ |
| 24 |
|
| 25 |
#include <immintrin.h> |
| 26 |
|
| 27 |
short *p, *p1, *p2, *p3, *p4, *p5, *p6; |
| 28 |
volatile __m512i x1, yy, zzz; |
| 29 |
volatile __m256i x2, y2, yyy; |
| 30 |
volatile __m128i x3, y3, xxx; |
| 31 |
volatile __mmask32 m32; |
| 32 |
volatile __mmask16 m16; |
| 33 |
volatile __mmask8 m8; |
| 34 |
|
| 35 |
void extern |
| 36 |
avx512bw_test (void) |
| 37 |
{ |
| 38 |
x1 = _mm512_mask_mov_epi16 (x1, m32, yy); |
| 39 |
x2 = _mm256_mask_mov_epi16 (x2, m16, y2); |
| 40 |
x3 = _mm_mask_mov_epi16 (x3, m8, y3); |
| 41 |
|
| 42 |
x1 = _mm512_maskz_mov_epi16 (m32, yy); |
| 43 |
x2 = _mm256_maskz_mov_epi16 (m16, y2); |
| 44 |
x3 = _mm_maskz_mov_epi16 (m8, y3); |
| 45 |
|
| 46 |
x1 = _mm512_mask_loadu_epi16 (x1, m32, p); |
| 47 |
x2 = _mm256_mask_loadu_epi16 (x2, m16, p); |
| 48 |
x3 = _mm_mask_loadu_epi16 (x3, m8, p); |
| 49 |
|
| 50 |
x1 = _mm512_maskz_loadu_epi16 (m32, p); |
| 51 |
x2 = _mm256_maskz_loadu_epi16 (m16, p); |
| 52 |
x3 = _mm_maskz_loadu_epi16 (m8, p); |
| 53 |
|
| 54 |
zzz = _mm512_loadu_epi16 (p5); |
| 55 |
yyy = _mm256_loadu_epi16 (p3); |
| 56 |
xxx = _mm_loadu_epi16 (p1); |
| 57 |
|
| 58 |
_mm512_mask_storeu_epi16 (p, m32, x1); |
| 59 |
_mm256_mask_storeu_epi16 (p, m16, x2); |
| 60 |
_mm_mask_storeu_epi16 (p, m8, x3); |
| 61 |
|
| 62 |
_mm512_storeu_epi16 (p6, zzz); |
| 63 |
_mm256_storeu_epi16 (p4, yyy); |
| 64 |
_mm_storeu_epi16 (p2, xxx); |
| 65 |
} |