1  /* { dg-do compile } */
       2  /* { dg-options "-mavx512fp16 -O2" } */
       3  /* { dg-final { scan-assembler-times "vfmaddcph\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
       4  /* { dg-final { scan-assembler-times "vfmaddcph\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\{%k\[0-9\]\}\[^\{\n\r]*(?:\n|\[ \\t\]+#)" 2 } } */
       5  /* { dg-final { scan-assembler-times "vfmaddcph\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\{%k\[0-9\]\}\{z\}\[^\n\r]*(?:\n|\[ \\t\]+#)" 1 } } */
       6  /* { dg-final { scan-assembler-times "vfmaddcph\[ \\t\]+\{rn-sae\}\[^\{\n\]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
       7  /* { dg-final { scan-assembler-times "vfmaddcph\[ \\t\]+\{rn-sae\}\[^\{\n\]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\{%k\[0-9\]\}\[^\n\r]*(?:\n|\[ \\t\]+#)" 2 } } */
       8  /* { dg-final { scan-assembler-times "vfmaddcph\[ \\t\]+\{rz-sae\}\[^\{\n\]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\{%k\[0-9\]\}\{z\}\[^\n\r]*(?:\n|\[ \\t\]+#)" 1 } } */
       9  /* { dg-final { scan-assembler-times "(?:vblendmps\[ \\t\]+%zmm\[0-9\]+|vmovaps\[ \\t\]+)\[^\n\r]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\{%k\[0-9\]\}(?:\n|\[ \\t\]+#)" 2 } } */
      10  
      11  #include <immintrin.h>
      12  
      13  volatile __m512h res, res1, res2;
      14  volatile __m512h x1, x2, x3;
      15  volatile __mmask16 m16;
      16  
      17  void extern
      18  avx512f_test (void)
      19  {
      20    res = _mm512_fmadd_pch (x1, x2, x3);
      21    res1 = _mm512_mask_fmadd_pch (res1, m16, x1, x2);
      22    res1 = _mm512_mask3_fmadd_pch (res1, x1, x2, m16);
      23    res2 = _mm512_maskz_fmadd_pch (m16, x1, x2, x3);
      24    res = _mm512_fmadd_round_pch (x1, x2, x3, 8);
      25    res1 = _mm512_mask_fmadd_round_pch (res1, m16, x1, x2, 8);
      26    res1 = _mm512_mask3_fmadd_round_pch (res1, x1, x2, m16, 8);
      27    res2 = _mm512_maskz_fmadd_round_pch (m16, x1, x2, x3, 11);
      28  }