cmpccxaddintrin.h amxfp16intrin.h prfchiintrin.h
raointintrin.h amxcomplexintrin.h avxvnniint16intrin.h
sm3intrin.h sha512intrin.h sm4intrin.h usermsrintrin.h
- avx10_2mediaintrin.h avx10_2-512mediaintrin.h
- avx10_2convertintrin.h avx10_2-512convertintrin.h
- avx10_2bf16intrin.h avx10_2-512bf16intrin.h
- avx10_2satcvtintrin.h avx10_2-512satcvtintrin.h
- avx10_2minmaxintrin.h avx10_2-512minmaxintrin.h
- avx10_2copyintrin.h amxavx512intrin.h amxtf32intrin.h
- amxtransposeintrin.h amxfp8intrin.h movrsintrin.h
- amxmovrsintrin.h"
+ avx10_2mediaintrin.h avx10_2convertintrin.h
+ avx10_2bf16intrin.h avx10_2satcvtintrin.h
+ avx10_2minmaxintrin.h avx10_2copyintrin.h
+ amxavx512intrin.h amxtf32intrin.h amxtransposeintrin.h
+ amxfp8intrin.h movrsintrin.h amxmovrsintrin.h"
;;
ia64-*-*)
extra_headers=ia64intrin.h
+++ /dev/null
-/* Copyright (C) 2024-2025 Free Software Foundation, Inc.
-
- This file is part of GCC.
-
- GCC is free software; you can redistribute it and/or modify
- it under the terms of the GNU General Public License as published by
- the Free Software Foundation; either version 3, or (at your option)
- any later version.
-
- GCC is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- GNU General Public License for more details.
-
- Under Section 7 of GPL version 3, you are granted additional
- permissions described in the GCC Runtime Library Exception, version
- 3.1, as published by the Free Software Foundation.
-
- You should have received a copy of the GNU General Public License and
- a copy of the GCC Runtime Library Exception along with this program;
- see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
- <http://www.gnu.org/licenses/>. */
-
-#ifndef _IMMINTRIN_H_INCLUDED
-#error "Never use <avx10_2-512bf16intrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef _AVX10_2_512BF16INTRIN_H_INCLUDED
-#define _AVX10_2_512BF16INTRIN_H_INCLUDED
-
-#if !defined (__AVX10_2__)
-#pragma GCC push_options
-#pragma GCC target("avx10.2")
-#define __DISABLE_AVX10_2__
-#endif /* __AVX10_2__ */
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_add_pbh (__m512bh __A, __m512bh __B)
-{
- return (__m512bh) __builtin_ia32_addbf16512 (__A, __B);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_add_pbh (__m512bh __W, __mmask32 __U,
- __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_addbf16512_mask (__A, __B, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_add_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_addbf16512_mask (__A, __B,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_sub_pbh (__m512bh __A, __m512bh __B)
-{
- return (__m512bh) __builtin_ia32_subbf16512 (__A, __B);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_sub_pbh (__m512bh __W, __mmask32 __U,
- __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_subbf16512_mask (__A, __B, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_sub_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_subbf16512_mask (__A, __B,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mul_pbh (__m512bh __A, __m512bh __B)
-{
- return (__m512bh) __builtin_ia32_mulbf16512 (__A, __B);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_mul_pbh (__m512bh __W, __mmask32 __U,
- __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_mulbf16512_mask (__A, __B, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_mul_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_mulbf16512_mask (__A, __B,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_div_pbh (__m512bh __A, __m512bh __B)
-{
- return (__m512bh) __builtin_ia32_divbf16512 (__A, __B);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_div_pbh (__m512bh __W, __mmask32 __U,
- __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_divbf16512_mask (__A, __B, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_div_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_divbf16512_mask (__A, __B,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_max_pbh (__m512bh __A, __m512bh __B)
-{
- return (__m512bh) __builtin_ia32_maxbf16512 (__A, __B);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_max_pbh (__m512bh __W, __mmask32 __U,
- __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_maxbf16512_mask (__A, __B, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_max_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_maxbf16512_mask (__A, __B,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_min_pbh (__m512bh __A, __m512bh __B)
-{
- return (__m512bh) __builtin_ia32_minbf16512 (__A, __B);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_min_pbh (__m512bh __W, __mmask32 __U,
- __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_minbf16512_mask (__A, __B, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_min_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_minbf16512_mask (__A, __B,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_scalef_pbh (__m512bh __A, __m512bh __B)
-{
- return (__m512bh) __builtin_ia32_scalefbf16512 (__A, __B);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_scalef_pbh (__m512bh __W, __mmask32 __U,
- __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_scalefbf16512_mask (__A, __B, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_scalef_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
-{
- return (__m512bh)
- __builtin_ia32_scalefbf16512_mask (__A, __B,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_fmadd_pbh (__m512bh __A, __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fmaddbf16512_mask (__A, __B, __C, (__mmask32) -1);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_fmadd_pbh (__m512bh __A, __mmask32 __U,
- __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fmaddbf16512_mask (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask3_fmadd_pbh (__m512bh __A, __m512bh __B,
- __m512bh __C, __mmask32 __U)
-{
- return (__m512bh)
- __builtin_ia32_fmaddbf16512_mask3 (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_fmadd_pbh (__mmask32 __U, __m512bh __A,
- __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fmaddbf16512_maskz (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_fmsub_pbh (__m512bh __A, __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fmsubbf16512_mask (__A, __B, __C, (__mmask32) -1);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_fmsub_pbh (__m512bh __A, __mmask32 __U,
- __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fmsubbf16512_mask (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask3_fmsub_pbh (__m512bh __A, __m512bh __B,
- __m512bh __C, __mmask32 __U)
-{
- return (__m512bh)
- __builtin_ia32_fmsubbf16512_mask3 (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_fmsub_pbh (__mmask32 __U, __m512bh __A,
- __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fmsubbf16512_maskz (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_fnmadd_pbh (__m512bh __A, __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fnmaddbf16512_mask (__A, __B, __C, (__mmask32) -1);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_fnmadd_pbh (__m512bh __A, __mmask32 __U,
- __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fnmaddbf16512_mask (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask3_fnmadd_pbh (__m512bh __A, __m512bh __B,
- __m512bh __C, __mmask32 __U)
-{
- return (__m512bh)
- __builtin_ia32_fnmaddbf16512_mask3 (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_fnmadd_pbh (__mmask32 __U, __m512bh __A,
- __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fnmaddbf16512_maskz (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_fnmsub_pbh (__m512bh __A, __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fnmsubbf16512_mask (__A, __B, __C, (__mmask32) -1);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_fnmsub_pbh (__m512bh __A, __mmask32 __U,
- __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fnmsubbf16512_mask (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask3_fnmsub_pbh (__m512bh __A, __m512bh __B,
- __m512bh __C, __mmask32 __U)
-{
- return (__m512bh)
- __builtin_ia32_fnmsubbf16512_mask3 (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_fnmsub_pbh (__mmask32 __U, __m512bh __A,
- __m512bh __B, __m512bh __C)
-{
- return (__m512bh)
- __builtin_ia32_fnmsubbf16512_maskz (__A, __B, __C, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_rsqrt_pbh (__m512bh __A)
-{
- return (__m512bh)
- __builtin_ia32_rsqrtbf16512_mask (__A,
- (__v32bf) _mm512_setzero_si512 (),
- (__mmask32) -1);
-
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_rsqrt_pbh (__m512bh __W, __mmask32 __U, __m512bh __A)
-{
- return (__m512bh)
- __builtin_ia32_rsqrtbf16512_mask (__A, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_rsqrt_pbh (__mmask32 __U, __m512bh __A)
-{
- return (__m512bh)
- __builtin_ia32_rsqrtbf16512_mask (__A,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_sqrt_pbh (__m512bh __A)
-{
- return (__m512bh)
- __builtin_ia32_sqrtbf16512_mask (__A,
- (__v32bf) _mm512_setzero_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_sqrt_pbh (__m512bh __W, __mmask32 __U, __m512bh __A)
-{
- return (__m512bh)
- __builtin_ia32_sqrtbf16512_mask (__A, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_sqrt_pbh (__mmask32 __U, __m512bh __A)
-{
- return (__m512bh)
- __builtin_ia32_sqrtbf16512_mask (__A,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_rcp_pbh (__m512bh __A)
-{
- return (__m512bh)
- __builtin_ia32_rcpbf16512_mask (__A,
- (__v32bf) _mm512_setzero_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_rcp_pbh (__m512bh __W, __mmask32 __U, __m512bh __A)
-{
- return (__m512bh)
- __builtin_ia32_rcpbf16512_mask (__A, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_rcp_pbh (__mmask32 __U, __m512bh __A)
-{
- return (__m512bh)
- __builtin_ia32_rcpbf16512_mask (__A,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_getexp_pbh (__m512bh __A)
-{
- return (__m512bh)
- __builtin_ia32_getexpbf16512_mask (__A,
- (__v32bf) _mm512_setzero_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_getexp_pbh (__m512bh __W, __mmask32 __U, __m512bh __A)
-{
- return (__m512bh) __builtin_ia32_getexpbf16512_mask (__A, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_getexp_pbh (__mmask32 __U, __m512bh __A)
-{
- return (__m512bh)
- __builtin_ia32_getexpbf16512_mask (__A,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-/* Intrinsics vrndscalebf16. */
-#ifdef __OPTIMIZE__
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_roundscale_pbh (__m512bh __A, int B)
-{
- return (__m512bh)
- __builtin_ia32_rndscalebf16512_mask (__A, B,
- (__v32bf) _mm512_setzero_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_roundscale_pbh (__m512bh __W, __mmask32 __U, __m512bh __A, int B)
-{
- return (__m512bh)
- __builtin_ia32_rndscalebf16512_mask (__A, B, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_roundscale_pbh (__mmask32 __U, __m512bh __A, int B)
-{
- return (__m512bh)
- __builtin_ia32_rndscalebf16512_mask (__A, B,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-#else
-#define _mm512_roundscale_pbh(A, B) \
- (__builtin_ia32_rndscalebf16512_mask ((A), (B), \
- (__v32bf) _mm512_setzero_si512 (), \
- (__mmask32) -1))
-
-#define _mm512_mask_roundscale_pbh(A, B, C, D) \
- (__builtin_ia32_rndscalebf16512_mask ((C), (D), (A), (B)))
-
-#define _mm512_maskz_roundscale_pbh(A, B, C) \
- (__builtin_ia32_rndscalebf16512_mask ((B), (C), \
- (__v32bf) _mm512_setzero_si512 (), \
- (A)))
-
-#endif /* __OPTIMIZE__ */
-
-/* Intrinsics vreducebf16. */
-#ifdef __OPTIMIZE__
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_reduce_pbh (__m512bh __A, int B)
-{
- return (__m512bh)
- __builtin_ia32_reducebf16512_mask (__A, B,
- (__v32bf) _mm512_setzero_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_reduce_pbh (__m512bh __W, __mmask32 __U,
- __m512bh __A, int B)
-{
- return (__m512bh)
- __builtin_ia32_reducebf16512_mask (__A, B, __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_reduce_pbh (__mmask32 __U, __m512bh __A, int B)
-{
- return (__m512bh)
- __builtin_ia32_reducebf16512_mask (__A, B,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-#else
-#define _mm512_reduce_pbh(A, B) \
- (__builtin_ia32_reducebf16512_mask ((A), (B), \
- (__v32bf) _mm512_setzero_si512 (), \
- (__mmask32) -1))
-
-#define _mm512_mask_reduce_pbh(A, B, C, D) \
- (__builtin_ia32_reducebf16512_mask ((C), (D), (A), (B)))
-
-#define _mm512_maskz_reduce_pbh(A, B, C) \
- (__builtin_ia32_reducebf16512_mask ((B), (C), \
- (__v32bf) _mm512_setzero_si512 (), \
- (A)))
-
-#endif /* __OPTIMIZE__ */
-
-/* Intrinsics vgetmantbf16. */
-#ifdef __OPTIMIZE__
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_getmant_pbh (__m512bh __A, _MM_MANTISSA_NORM_ENUM __B,
- _MM_MANTISSA_SIGN_ENUM __C)
-{
- return (__m512bh)
- __builtin_ia32_getmantbf16512_mask (__A, (int) (__C << 2) | __B,
- (__v32bf) _mm512_setzero_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_getmant_pbh (__m512bh __W, __mmask32 __U, __m512bh __A,
- _MM_MANTISSA_NORM_ENUM __B,
- _MM_MANTISSA_SIGN_ENUM __C)
-{
- return (__m512bh)
- __builtin_ia32_getmantbf16512_mask (__A, (int) (__C << 2) | __B,
- __W, __U);
-}
-
-extern __inline__ __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_getmant_pbh (__mmask32 __U, __m512bh __A,
- _MM_MANTISSA_NORM_ENUM __B,
- _MM_MANTISSA_SIGN_ENUM __C)
-{
- return (__m512bh)
- __builtin_ia32_getmantbf16512_mask (__A, (int) (__C << 2) | __B,
- (__v32bf) _mm512_setzero_si512 (),
- __U);
-}
-
-#else
-#define _mm512_getmant_pbh(A, B, C) \
- (__builtin_ia32_getmantbf16512_mask ((A), (int)(((C)<<2) | (B)), \
- (__v32bf) _mm512_setzero_si512 (), \
- (__mmask32) -1))
-
-#define _mm512_mask_getmant_pbh(A, B, C, D, E) \
- (__builtin_ia32_getmantbf16512_mask ((C), (int)(((D)<<2) | (E)), (A), (B)))
-
-#define _mm512_maskz_getmant_pbh(A, B, C, D) \
- (__builtin_ia32_getmantbf16512_mask ((B), (int)(((C)<<2) | (D)), \
- (__v32bf) _mm512_setzero_si512 (), \
- (A)))
-
-#endif /* __OPTIMIZE__ */
-
-/* Intrinsics vfpclassbf16. */
-#ifdef __OPTIMIZE__
-extern __inline __mmask32
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_fpclass_pbh_mask (__mmask32 __U, __m512bh __A,
- const int __imm)
-{
- return (__mmask32)
- __builtin_ia32_fpclassbf16512_mask (__A, __imm, __U);
-}
-
-extern __inline __mmask32
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_fpclass_pbh_mask (__m512bh __A, const int __imm)
-{
- return (__mmask32)
- __builtin_ia32_fpclassbf16512_mask (__A, __imm,
- (__mmask32) -1);
-}
-
-#else
-#define _mm512_mask_fpclass_pbh_mask(U, X, C) \
- ((__mmask32) __builtin_ia32_fpclassbf16512_mask ( \
- (__v32bf) (__m512bh) (X), (int) (C), (__mmask32) (U)))
-
-#define _mm512_fpclass_pbh_mask(X, C) \
- ((__mmask32) __builtin_ia32_fpclassbf16512_mask ( \
- (__v32bf) (__m512bh) (X), (int) (C), (__mmask32) (-1)))
-#endif /* __OPIMTIZE__ */
-
-
-/* Intrinsics vcmpbf16. */
-#ifdef __OPTIMIZE__
-extern __inline __mmask32
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cmp_pbh_mask (__mmask32 __U, __m512bh __A, __m512bh __B,
- const int __imm)
-{
- return (__mmask32)
- __builtin_ia32_cmpbf16512_mask (__A, __B, __imm, __U);
-}
-
-extern __inline __mmask32
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cmp_pbh_mask (__m512bh __A, __m512bh __B, const int __imm)
-{
- return (__mmask32)
- __builtin_ia32_cmpbf16512_mask (__A, __B, __imm,
- (__mmask32) -1);
-}
-
-#else
-#define _mm512_mask_cmp_pbh_mask(A, B, C, D) \
- ((__mmask32) __builtin_ia32_cmpbf16512_mask ((B), (C), (D), (A)))
-
-#define _mm512_cmp_pbh_mask(A, B, C) \
- ((__mmask32) __builtin_ia32_cmpbf16512_mask ((A), (B), (C), (-1)))
-
-#endif /* __OPIMTIZE__ */
-
-#ifdef __DISABLE_AVX10_2__
-#undef __DISABLE_AVX10_2__
-#pragma GCC pop_options
-#endif /* __DISABLE_AVX10_2__ */
-
-#endif /* _AVX10_2_512BF16INTRIN_H_INCLUDED */
+++ /dev/null
-/* Copyright (C) 2024-2025 Free Software Foundation, Inc.
-
- This file is part of GCC.
-
- GCC is free software; you can redistribute it and/or modify
- it under the terms of the GNU General Public License as published by
- the Free Software Foundation; either version 3, or (at your option)
- any later version.
-
- GCC is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- GNU General Public License for more details.
-
- Under Section 7 of GPL version 3, you are granted additional
- permissions described in the GCC Runtime Library Exception, version
- 3.1, as published by the Free Software Foundation.
-
- You should have received a copy of the GNU General Public License and
- a copy of the GCC Runtime Library Exception along with this program;
- see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
- <http://www.gnu.org/licenses/>. */
-
-#ifndef _IMMINTRIN_H_INCLUDED
-#error "Never use <avx10_2-512convertintrin.h> directly; include <immintrin.h> instead."
-#endif // _IMMINTRIN_H_INCLUDED
-
-#ifndef __AVX10_2_512CONVERTINTRIN_H_INCLUDED
-#define __AVX10_2_512CONVERTINTRIN_H_INCLUDED
-
-#ifndef __AVX10_2__
-#pragma GCC push_options
-#pragma GCC target("avx10.2")
-#define __DISABLE_AVX10_2__
-#endif /* __AVX10_2__ */
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtx2ps_ph (__m512 __A, __m512 __B)
-{
- return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- (__v32hf)
- _mm512_setzero_ph (),
- (__mmask32) -1,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtx2ps_ph (__m512h __W, __mmask32 __U, __m512 __A,
- __m512 __B)
-{
- return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- (__v32hf) __W,
- (__mmask32) __U,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtx2ps_ph (__mmask32 __U, __m512 __A, __m512 __B)
-{
- return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- (__v32hf)
- _mm512_setzero_ph (),
- (__mmask32) __U,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-#ifdef __OPTIMIZE__
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtx_round2ps_ph (__m512 __A, __m512 __B, const int __R)
-{
- return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- (__v32hf)
- _mm512_setzero_ph (),
- (__mmask32) -1,
- __R);
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtx_round2ps_ph (__m512h __W, __mmask32 __U, __m512 __A,
- __m512 __B, const int __R)
-{
- return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- (__v32hf) __W,
- (__mmask32) __U,
- __R);
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtx_round2ps_ph (__mmask32 __U, __m512 __A,
- __m512 __B, const int __R)
-{
- return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- (__v32hf)
- _mm512_setzero_ph (),
- (__mmask32) __U,
- __R);
-}
-
-#else
-#define _mm512_cvtx_round2ps_ph(A, B, R) \
- ((__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) (A), \
- (__v16sf) (B), \
- (__v32hf) \
- (_mm512_setzero_ph ()), \
- (__mmask32) (-1), \
- (R)))
-#define _mm512_mask_cvtx_round2ps_ph(W, U, A, B, R) \
- ((__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) (A), \
- (__v16sf) (B), \
- (__v32hf) (W), \
- (__mmask32) (U), \
- (R)))
-#define _mm512_maskz_cvtx_round2ps_ph(U, A, B, R) \
- ((__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) (A), \
- (__v16sf) (B), \
- (__v32hf) \
- (_mm512_setzero_ph ()), \
- (__mmask32) (U), \
- (R)))
-#endif /* __OPTIMIZE__ */
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtbiasph_bf8 (__m512i __A, __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2bf8512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i)
- _mm256_undefined_si256 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtbiasph_bf8 (__m256i __W, __mmask32 __U,
- __m512i __A, __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2bf8512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i) __W,
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtbiasph_bf8 (__mmask32 __U, __m512i __A, __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2bf8512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i)
- _mm256_setzero_si256 (),
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvts_biasph_bf8 (__m512i __A, __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2bf8s512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i)
- _mm256_undefined_si256 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvts_biasph_bf8 (__m256i __W, __mmask32 __U,
- __m512i __A, __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2bf8s512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i) __W,
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvts_biasph_bf8 (__mmask32 __U, __m512i __A, __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2bf8s512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i)
- _mm256_setzero_si256 (),
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtbiasph_hf8 (__m512i __A, __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2hf8512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i)
- _mm256_undefined_si256 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtbiasph_hf8 (__m256i __W, __mmask32 __U, __m512i __A,
- __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2hf8512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i) __W,
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtbiasph_hf8 (__mmask32 __U, __m512i __A, __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2hf8512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i)
- _mm256_setzero_si256 (),
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvts_biasph_hf8 (__m512i __A, __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2hf8s512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i)
- _mm256_undefined_si256 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvts_biasph_hf8 (__m256i __W, __mmask32 __U,
- __m512i __A, __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2hf8s512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i) __W,
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvts_biasph_hf8 (__mmask32 __U, __m512i __A, __m512h __B)
-{
- return (__m256i) __builtin_ia32_vcvtbiasph2hf8s512_mask ((__v64qi) __A,
- (__v32hf) __B,
- (__v32qi)(__m256i)
- _mm256_setzero_si256 (),
- (__mmask32) __U);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvt2ph_bf8 (__m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2bf8512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi)
- _mm512_setzero_si512 (),
- (__mmask64) -1);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvt2ph_bf8 (__m512i __W, __mmask64 __U,
- __m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2bf8512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi) __W,
- (__mmask64) __U);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvt2ph_bf8 (__mmask64 __U, __m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2bf8512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi)
- _mm512_setzero_si512 (),
- (__mmask64) __U);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvts_2ph_bf8 (__m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2bf8s512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi)
- _mm512_setzero_si512 (),
- (__mmask64) -1);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvts_2ph_bf8 (__m512i __W, __mmask64 __U,
- __m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2bf8s512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi) __W,
- (__mmask64) __U);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvts_2ph_bf8 (__mmask64 __U, __m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2bf8s512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi)
- _mm512_setzero_si512 (),
- (__mmask64) __U);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvt2ph_hf8 (__m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2hf8512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi)
- _mm512_setzero_si512 (),
- (__mmask64) -1);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvt2ph_hf8 (__m512i __W, __mmask64 __U,
- __m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2hf8512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi) __W,
- (__mmask64) __U);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvt2ph_hf8 (__mmask64 __U, __m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2hf8512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi)
- _mm512_setzero_si512 (),
- (__mmask64) __U);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvts_2ph_hf8 (__m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2hf8s512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi)
- _mm512_setzero_si512 (),
- (__mmask64) -1);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvts_2ph_hf8 (__m512i __W, __mmask64 __U,
- __m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2hf8s512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi) __W,
- (__mmask64) __U);
-}
-
-extern __inline__ __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvts_2ph_hf8 (__mmask64 __U, __m512h __A, __m512h __B)
-{
- return (__m512i) __builtin_ia32_vcvt2ph2hf8s512_mask ((__v32hf) __A,
- (__v32hf) __B,
- (__v64qi)
- _mm512_setzero_si512 (),
- (__mmask64) __U);
-}
-
-extern __inline__ __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvthf8_ph (__m256i __A)
-{
- return (__m512h) __builtin_ia32_vcvthf82ph512_mask ((__v32qi) __A,
- (__v32hf) (__m512h)
- _mm512_undefined_ph (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvthf8_ph (__m512h __W, __mmask32 __U, __m256i __A)
-{
- return (__m512h) __builtin_ia32_vcvthf82ph512_mask ((__v32qi) __A,
- (__v32hf) (__m512h) __W,
- (__mmask32) __U);
-}
-
-extern __inline__ __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvthf8_ph (__mmask32 __U, __m256i __A)
-{
- return (__m512h) __builtin_ia32_vcvthf82ph512_mask ((__v32qi) __A,
- (__v32hf) (__m512h)
- _mm512_setzero_ph (),
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtph_bf8 (__m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2bf8512_mask ((__v32hf) __A,
- (__v32qi) (__m256i)
- _mm256_undefined_si256 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtph_bf8 (__m256i __W, __mmask32 __U, __m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2bf8512_mask ((__v32hf) __A,
- (__v32qi) (__m256i) __W,
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtph_bf8 (__mmask32 __U, __m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2bf8512_mask ((__v32hf) __A,
- (__v32qi) (__m256i)
- _mm256_setzero_si256 (),
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvts_ph_bf8 (__m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2bf8s512_mask ((__v32hf) __A,
- (__v32qi) (__m256i)
- _mm256_undefined_si256 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvts_ph_bf8 (__m256i __W, __mmask32 __U, __m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2bf8s512_mask ((__v32hf) __A,
- (__v32qi) (__m256i) __W,
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvts_ph_bf8 (__mmask32 __U, __m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2bf8s512_mask ((__v32hf) __A,
- (__v32qi) (__m256i)
- _mm256_setzero_si256 (),
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtph_hf8 (__m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2hf8512_mask ((__v32hf) __A,
- (__v32qi) (__m256i)
- _mm256_undefined_si256 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtph_hf8 (__m256i __W, __mmask32 __U, __m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2hf8512_mask ((__v32hf) __A,
- (__v32qi)(__m256i) __W,
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtph_hf8 (__mmask32 __U, __m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2hf8512_mask ((__v32hf) __A,
- (__v32qi) (__m256i)
- _mm256_setzero_si256 (),
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvts_ph_hf8 (__m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2hf8s512_mask ((__v32hf) __A,
- (__v32qi) (__m256i)
- _mm256_undefined_si256 (),
- (__mmask32) -1);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvts_ph_hf8 (__m256i __W, __mmask32 __U, __m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2hf8s512_mask ((__v32hf) __A,
- (__v32qi) (__m256i) __W,
- (__mmask32) __U);
-}
-
-extern __inline__ __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvts_ph_hf8 (__mmask32 __U, __m512h __A)
-{
- return (__m256i) __builtin_ia32_vcvtph2hf8s512_mask ((__v32hf) __A,
- (__v32qi) (__m256i)
- _mm256_setzero_si256 (),
- (__mmask32) __U);
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtbf8_ph (__m256i __A)
-{
- return (__m512h) _mm512_castsi512_ph ((__m512i) _mm512_slli_epi16 (
- (__m512i) _mm512_cvtepi8_epi16 (__A), 8));
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtbf8_ph (__m512h __S, __mmask32 __U, __m256i __A)
-{
- return (__m512h) _mm512_castsi512_ph ((__m512i) _mm512_mask_slli_epi16 (
- (__m512i) __S, __U, (__m512i) _mm512_cvtepi8_epi16 (__A), 8));
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtbf8_ph (__mmask32 __U, __m256i __A)
-{
- return (__m512h) _mm512_castsi512_ph ((__m512i) _mm512_slli_epi16 (
- (__m512i) _mm512_maskz_cvtepi8_epi16 (__U, __A), 8));
-}
-
-#ifdef __DISABLE_AVX10_2__
-#undef __DISABLE_AVX10_2__
-#pragma GCC pop_options
-#endif /* __DISABLE_AVX10_2__ */
-
-#endif /* __AVX10_2_512CONVERTINTRIN_H_INCLUDED */
+++ /dev/null
-/* Copyright (C) 2024-2025 Free Software Foundation, Inc.
-
- This file is part of GCC.
-
- GCC is free software; you can redistribute it and/or modify
- it under the terms of the GNU General Public License as published by
- the Free Software Foundation; either version 3, or (at your option)
- any later version.
-
- GCC is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- GNU General Public License for more details.
-
- Under Section 7 of GPL version 3, you are granted additional
- permissions described in the GCC Runtime Library Exception, version
- 3.1, as published by the Free Software Foundation.
-
- You should have received a copy of the GNU General Public License and
- a copy of the GCC Runtime Library Exception along with this program;
- see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
- <http://www.gnu.org/licenses/>. */
-
-#if !defined _IMMINTRIN_H_INCLUDED
-#error "Never use <avx10_2-512mediaintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef _AVX10_2_512MEDIAINTRIN_H_INCLUDED
-#define _AVX10_2_512MEDIAINTRIN_H_INCLUDED
-
-#if !defined(__AVX10_2__)
-#pragma GCC push_options
-#pragma GCC target("avx10.2")
-#define __DISABLE_AVX10_2__
-#endif /* __AVX10_2__ */
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpbssd_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbssd512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpbssd_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbssd_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpbssd_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbssd_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpbssds_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbssds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpbssds_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbssds_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpbssds_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbssds_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpbsud_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbsud512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpbsud_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbsud_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpbsud_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbsud_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpbsuds_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbsuds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpbsuds_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbsuds_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpbsuds_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbsuds_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpbuud_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbuud512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpbuud_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbuud_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpbuud_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbuud_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpbuuds_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbuuds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpbuuds_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbuuds_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpbuuds_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpbuuds_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpwsud_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwsud512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpwsud_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwsud_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpwsud_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwsud_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpwsuds_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwsuds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpwsuds_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwsuds_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpwsuds_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwsuds_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpwusd_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwusd512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpwusd_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwusd_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpwusd_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwusd_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpwusds_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwusds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpwusds_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwusds_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpwusds_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwusds_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpwuud_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwuud512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpwuud_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwuud_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpwuud_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwuud_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpwuuds_epi32 (__m512i __W, __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwuuds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpwuuds_epi32 (__m512i __W, __mmask16 __U,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwuuds_v16si_mask ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpwuuds_epi32 (__mmask16 __U, __m512i __W,
- __m512i __A, __m512i __B)
-{
- return (__m512i)
- __builtin_ia32_vpdpwuuds_v16si_maskz ((__v16si) __W,
- (__v16si) __A,
- (__v16si) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_dpph_ps (__m512 __W, __m512h __A, __m512h __B)
-{
- return (__m512)
- __builtin_ia32_vdpphps512_mask ((__v16sf) __W,
- (__v16sf) __A,
- (__v16sf) __B,
- (__mmask16) -1);
-}
-
-extern __inline __m512
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_dpph_ps (__m512 __W, __mmask16 __U, __m512h __A,
- __m512h __B)
-{
- return (__m512)
- __builtin_ia32_vdpphps512_mask ((__v16sf) __W,
- (__v16sf) __A,
- (__v16sf) __B,
- (__mmask16) __U);
-}
-
-extern __inline __m512
-__attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_dpph_ps (__mmask16 __U, __m512 __W, __m512h __A,
- __m512h __B)
-{
- return (__m512)
- __builtin_ia32_vdpphps512_maskz ((__v16sf) __W,
- (__v16sf) __A,
- (__v16sf) __B,
- (__mmask16) __U);
-}
-
-#ifdef __OPTIMIZE__
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mpsadbw_epu8 (__m512i __X, __m512i __Y, const int __M)
-{
- return (__m512i) __builtin_ia32_mpsadbw512 ((__v64qi) __X,
- (__v64qi) __Y,
- __M);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_mpsadbw_epu8 (__m512i __W, __mmask32 __U, __m512i __X,
- __m512i __Y, const int __M)
-{
- return (__m512i) __builtin_ia32_mpsadbw512_mask ((__v64qi) __X,
- (__v64qi) __Y,
- __M,
- (__v32hi) __W,
- __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_mpsadbw_epu8 (__mmask32 __U, __m512i __X,
- __m512i __Y, const int __M)
-{
- return (__m512i) __builtin_ia32_mpsadbw512_mask ((__v64qi) __X,
- (__v64qi) __Y,
- __M,
- (__v32hi) _mm512_setzero_epi32 (),
- __U);
-}
-#else
-#define _mm512_mpsadbw_epu8(X, Y, M) \
- (__m512i) __builtin_ia32_mpsadbw512 ((__v64qi)(__m512i)(X), \
- (__v64qi)(__m512i)(Y), (int)(M))
-
-#define _mm512_mask_mpsadbw_epu8(W, U, X, Y, M) \
- (__m512i) __builtin_ia32_mpsadbw512_mask ((__v64qi)(__m512i)(X), \
- (__v64qi)(__m512i)(Y), \
- (int)(M), \
- (__v32hi)(__m512i)(W), \
- (__mmask32)(U))
-
-#define _mm512_maskz_mpsadbw_epu8(U, X, Y, M) \
- (__m512i) __builtin_ia32_mpsadbw512_mask ((__v64qi)(__m512i)(X), \
- (__v64qi)(__m512i)(Y), \
- (int)(M), \
- (__v32hi) _mm512_setzero_epi32 (), \
- (__mmask32)(U))
-#endif
-
-#ifdef __DISABLE_AVX10_2__
-#undef __DISABLE_AVX10_2__
-#pragma GCC pop_options
-#endif /* __DISABLE_AVX10_2__ */
-
-#endif /* __AVX10_2_512MEDIAINTRIN_H_INCLUDED */
+++ /dev/null
-/* Copyright (C) 2024-2025 Free Software Foundation, Inc.
- This file is part of GCC.
- GCC is free software; you can redistribute it and/or modify
- it under the terms of the GNU General Public License as published by
- the Free Software Foundation; either version 3, or (at your option)
- any later version.
- GCC is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- GNU General Public License for more details.
- Under Section 7 of GPL version 3, you are granted additional
- permissions described in the GCC Runtime Library Exception, version
- 3.1, as published by the Free Software Foundation.
- You should have received a copy of the GNU General Public License and
- a copy of the GCC Runtime Library Exception along with this program;
- see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
- <http://www.gnu.org/licenses/>. */
-
-#if !defined _IMMINTRIN_H_INCLUDED
-#error "Never use <avx10_2-512minmaxintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef _AVX10_2_512MINMAXINTRIN_H_INCLUDED
-#define _AVX10_2_512MINMAXINTRIN_H_INCLUDED
-
-#if !defined (__AVX10_2__)
-#pragma GCC push_options
-#pragma GCC target("avx10.2")
-#define __DISABLE_AVX10_2__
-#endif /* __AVX10_2__ */
-
-#ifdef __OPTIMIZE__
-extern __inline __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_minmax_pbh (__m512bh __A, __m512bh __B, const int __C)
-{
- return (__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) __A,
- (__v32bf) __B,
- __C,
- (__v32bf)(__m512bh)
- _mm512_setzero_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_minmax_pbh (__m512bh __W, __mmask32 __U,
- __m512bh __A, __m512bh __B, const int __C)
-{
- return (__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) __A,
- (__v32bf) __B,
- __C,
- (__v32bf) __W,
- (__mmask32) __U);
-}
-
-extern __inline __m512bh
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_minmax_pbh (__mmask32 __U, __m512bh __A,
- __m512bh __B, const int __C)
-{
- return (__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) __A,
- (__v32bf) __B,
- __C,
- (__v32bf)(__m512bh)
- _mm512_setzero_si512 (),
- (__mmask32) __U);
-}
-
-extern __inline __m512d
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_minmax_pd (__m512d __A, __m512d __B, const int __C)
-{
- return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
- (__v8df) __B,
- __C,
- (__v8df)
- _mm512_undefined_pd (),
- (__mmask8) -1,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-extern __inline __m512d
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_minmax_pd (__m512d __W, __mmask8 __U, __m512d __A,
- __m512d __B, const int __C)
-{
- return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
- (__v8df) __B,
- __C,
- (__v8df) __W,
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-extern __inline __m512d
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_minmax_pd (__mmask8 __U, __m512d __A, __m512d __B,
- const int __C)
-{
- return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
- (__v8df) __B,
- __C,
- (__v8df)
- _mm512_setzero_pd (),
- (__mmask8) __U,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-extern __inline __m512d
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_minmax_round_pd (__m512d __A, __m512d __B, const int __C,
- const int __R)
-{
- return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
- (__v8df) __B,
- __C,
- (__v8df)
- _mm512_undefined_pd (),
- (__mmask8) -1, __R);
-}
-
-extern __inline __m512d
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_minmax_round_pd (__m512d __W, __mmask8 __U, __m512d __A,
- __m512d __B, const int __C, const int __R)
-{
- return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
- (__v8df) __B,
- __C,
- (__v8df) __W,
- (__mmask8) __U, __R);
-}
-
-extern __inline __m512d
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_minmax_round_pd (__mmask8 __U, __m512d __A, __m512d __B,
- const int __C, const int __R)
-{
- return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
- (__v8df) __B,
- __C,
- (__v8df)
- _mm512_setzero_pd (),
- (__mmask8) __U, __R);
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_minmax_ph (__m512h __A, __m512h __B, const int __C)
-{
- return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
- (__v32hf) __B,
- __C,
- (__v32hf)
- _mm512_undefined_ph (),
- (__mmask32) -1,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_minmax_ph (__m512h __W, __mmask32 __U, __m512h __A,
- __m512h __B, const int __C)
-{
- return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
- (__v32hf) __B,
- __C,
- (__v32hf) __W,
- (__mmask32) __U,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_minmax_ph (__mmask32 __U, __m512h __A, __m512h __B,
- const int __C)
-{
- return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
- (__v32hf) __B,
- __C,
- (__v32hf)
- _mm512_setzero_ph (),
- (__mmask32) __U,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_minmax_round_ph (__m512h __A, __m512h __B, const int __C, const int __R)
-{
- return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
- (__v32hf) __B,
- __C,
- (__v32hf)
- _mm512_undefined_ph (),
- (__mmask32) -1, __R);
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_minmax_round_ph (__m512h __W, __mmask32 __U, __m512h __A,
- __m512h __B, const int __C, const int __R)
-{
- return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
- (__v32hf) __B,
- __C,
- (__v32hf) __W,
- (__mmask32) __U, __R);
-}
-
-extern __inline __m512h
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_minmax_round_ph (__mmask32 __U, __m512h __A, __m512h __B,
- const int __C, const int __R)
-{
- return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
- (__v32hf) __B,
- __C,
- (__v32hf)
- _mm512_setzero_ph (),
- (__mmask32) __U, __R);
-}
-
-extern __inline __m512
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_minmax_ps (__m512 __A, __m512 __B, const int __C)
-{
- return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- __C,
- (__v16sf)
- _mm512_undefined_ps (),
- (__mmask16) -1,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-extern __inline __m512
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_minmax_ps (__m512 __W, __mmask16 __U, __m512 __A,
- __m512 __B, const int __C)
-{
- return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- __C,
- (__v16sf) __W,
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-extern __inline __m512
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_minmax_ps (__mmask16 __U, __m512 __A, __m512 __B,
- const int __C)
-{
- return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- __C,
- (__v16sf)
- _mm512_setzero_ps (),
- (__mmask16) __U,
- _MM_FROUND_CUR_DIRECTION);
-}
-
-extern __inline __m512
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_minmax_round_ps (__m512 __A, __m512 __B, const int __C, const int __R)
-{
- return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- __C,
- (__v16sf)
- _mm512_undefined_ps (),
- (__mmask16) -1, __R);
-}
-
-extern __inline __m512
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_minmax_round_ps (__m512 __W, __mmask16 __U, __m512 __A,
- __m512 __B, const int __C, const int __R)
-{
- return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- __C,
- (__v16sf) __W,
- (__mmask16) __U, __R);
-}
-
-extern __inline __m512
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_minmax_round_ps (__mmask16 __U, __m512 __A, __m512 __B,
- const int __C, const int __R)
-{
- return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
- (__v16sf) __B,
- __C,
- (__v16sf)
- _mm512_setzero_ps (),
- (__mmask16) __U, __R);
-}
-
-#else
-#define _mm512_minmax_pbh(A, B, C) \
- ((__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) (A), \
- (__v32bf) (B), \
- (int) (C), \
- (__v32bf) (__m512bh) \
- _mm512_setzero_si512 (), \
- (__mmask32) (-1)))
-
-#define _mm512_mask_minmax_pbh(W, U, A, B, C) \
- ((__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) (A), \
- (__v32bf) (B), \
- (int) (C), \
- (__v32bf) (__m512bh) (W), \
- (__mmask32) (U)))
-
-#define _mm512_maskz_minmax_pbh(U, A, B, C) \
- ((__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) (A), \
- (__v32bf) (B), \
- (int) (C), \
- (__v32bf) (__m512bh) \
- _mm512_setzero_si512 (), \
- (__mmask32) (U)))
-
-#define _mm512_minmax_round_pd(A, B, C, R) \
- ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
- (__v8df) (B), \
- (int) (C), \
- (__v8df) (__m512d) \
- _mm512_undefined_pd (), \
- (__mmask8) (-1), \
- (int) (R)))
-
-#define _mm512_mask_minmax_round_pd(W, U, A, B, C, R) \
- ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
- (__v8df) (B), \
- (int) (C), \
- (__v8df) (__m512d) (W), \
- (__mmask8) (U), \
- (int) (R)))
-
-#define _mm512_maskz_minmax_round_pd(U, A, B, C, R) \
- ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
- (__v8df) (B), \
- (int) (C), \
- (__v8df) (__m512d) \
- _mm512_setzero_pd (), \
- (__mmask8) (U), \
- (int) (R)))
-
-#define _mm512_minmax_round_ph(A, B, C, R) \
- ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
- (__v32hf) (B), \
- (int) (C), \
- (__v32hf) (__m512h) \
- _mm512_undefined_ph (), \
- (__mmask32) (-1), \
- (int) (R)))
-
-#define _mm512_mask_minmax_round_ph(W, U, A, B, C, R) \
- ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
- (__v32hf) (B), \
- (int) (C), \
- (__v32hf) (__m512h) (W), \
- (__mmask32) (U), \
- (int) (R)))
-
-#define _mm512_maskz_minmax_round_ph(U, A, B, C, R) \
- ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
- (__v32hf) (B), \
- (int) (C), \
- (__v32hf) (__m512h) \
- _mm512_setzero_ph (), \
- (__mmask32) (U), \
- (int) (R)))
-
-#define _mm512_minmax_round_ps(A, B, C, R) \
- ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
- (__v16sf) (B), \
- (int) (C), \
- (__v16sf) (__m512) \
- _mm512_undefined_ps (), \
- (__mmask16) (-1), \
- (int) (R)))
-
-#define _mm512_mask_minmax_round_ps(W, U, A, B, C, R) \
- ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
- (__v16sf) (B), \
- (int) (C), \
- (__v16sf) (__m512) (W), \
- (__mmask16) (U), \
- (int) (R)))
-
-#define _mm512_maskz_minmax_round_ps(U, A, B, C, R) \
- ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
- (__v16sf) (B), \
- (int) (C), \
- (__v16sf) (__m512) \
- _mm512_setzero_ps (), \
- (__mmask16) (U), \
- (int) (R)))
-
-#define _mm512_minmax_pd(A, B, C) \
- ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
- (__v8df) (B), \
- (int) (C), \
- (__v8df) (__m512d) \
- _mm512_undefined_pd (), \
- (__mmask8) (-1), \
- _MM_FROUND_CUR_DIRECTION))
-
-#define _mm512_mask_minmax_pd(W, U, A, B, C) \
- ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
- (__v8df) (B), \
- (int) (C), \
- (__v8df) (__m512d) (W), \
- (__mmask8) (U), \
- _MM_FROUND_CUR_DIRECTION))
-
-#define _mm512_maskz_minmax_pd(U, A, B, C) \
- ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
- (__v8df) (B), \
- (int) (C), \
- (__v8df) (__m512d) \
- _mm512_setzero_pd (), \
- (__mmask8) (U), \
- _MM_FROUND_CUR_DIRECTION))
-
-#define _mm512_minmax_ph(A, B, C) \
- ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
- (__v32hf) (B), \
- (int) (C), \
- (__v32hf) (__m512h) \
- _mm512_undefined_ph (), \
- (__mmask32) (-1), \
- _MM_FROUND_CUR_DIRECTION))
-
-#define _mm512_mask_minmax_ph(W, U, A, B, C) \
- ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
- (__v32hf) (B), \
- (int) (C), \
- (__v32hf) (__m512h) (W), \
- (__mmask32) (U), \
- _MM_FROUND_CUR_DIRECTION))
-
-#define _mm512_maskz_minmax_ph(U, A, B, C) \
- ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
- (__v32hf) (B), \
- (int) (C), \
- (__v32hf) (__m512h) \
- _mm512_setzero_ph (), \
- (__mmask32) (U), \
- _MM_FROUND_CUR_DIRECTION))
-
-#define _mm512_minmax_ps(A, B, C) \
- ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
- (__v16sf) (B), \
- (int) (C), \
- (__v16sf) (__m512) \
- _mm512_undefined_ps (), \
- (__mmask16) (-1), \
- _MM_FROUND_CUR_DIRECTION))
-
-#define _mm512_mask_minmax_ps(W, U, A, B, C) \
- ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
- (__v16sf) (B), \
- (int) (C), \
- (__v16sf) (__m512) (W), \
- (__mmask16) (U), \
- _MM_FROUND_CUR_DIRECTION))
-
-#define _mm512_maskz_minmax_ps(U, A, B, C) \
- ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
- (__v16sf) (B), \
- (int) (C), \
- (__v16sf) (__m512) \
- _mm512_setzero_ps (), \
- (__mmask16) (U), \
- _MM_FROUND_CUR_DIRECTION))
-
-#endif
-
-#ifdef __DISABLE_AVX10_2__
-#undef __DISABLE_AVX10_2__
-#pragma GCC pop_options
-#endif /* __DISABLE_AVX10_2__ */
-
-#endif /* _AVX10_2_512MINMAXINTRIN_H_INCLUDED */
+++ /dev/null
-/* Copyright (C) 2024-2025 Free Software Foundation, Inc.
-
- This file is part of GCC.
-
- GCC is free software; you can redistribute it and/or modify
- it under the terms of the GNU General Public License as published by
- the Free Software Foundation; either version 3, or (at your option)
- any later version.
-
- GCC is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- GNU General Public License for more details.
-
- Under Section 7 of GPL version 3, you are granted additional
- permissions described in the GCC Runtime Library Exception, version
- 3.1, as published by the Free Software Foundation.
-
- You should have received a copy of the GNU General Public License and
- a copy of the GCC Runtime Library Exception along with this program;
- see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
- <http://www.gnu.org/licenses/>. */
-
-#if !defined _IMMINTRIN_H_INCLUDED
-#error "Never use <avx10_2-512satcvtintrin.h> directly; include <immintrin.h> instead."
-#endif
-
-#ifndef _AVX10_2_512SATCVTINTRIN_H_INCLUDED
-#define _AVX10_2_512SATCVTINTRIN_H_INCLUDED
-
-#if !defined (__AVX10_2__)
-#pragma GCC push_options
-#pragma GCC target("avx10.2")
-#define __DISABLE_AVX10_2__
-#endif /* __AVX10_2__ */
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvts_bf16_epi8 (__m512bh __A)
-{
- return
- (__m512i) __builtin_ia32_cvtbf162ibs512_mask ((__v32bf) __A,
- (__v32hi)
- _mm512_undefined_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvts_bf16_epi8 (__m512i __W, __mmask32 __U, __m512bh __A)
-{
- return (__m512i) __builtin_ia32_cvtbf162ibs512_mask ((__v32bf) __A,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvts_bf16_epi8 (__mmask32 __U, __m512bh __A)
-{
- return
- (__m512i) __builtin_ia32_cvtbf162ibs512_mask ((__v32bf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvts_bf16_epu8 (__m512bh __A)
-{
- return
- (__m512i) __builtin_ia32_cvtbf162iubs512_mask ((__v32bf) __A,
- (__v32hi)
- _mm512_undefined_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvts_bf16_epu8 (__m512i __W, __mmask32 __U, __m512bh __A)
-{
- return (__m512i) __builtin_ia32_cvtbf162iubs512_mask ((__v32bf) __A,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvts_bf16_epu8 (__mmask32 __U, __m512bh __A)
-{
- return
- (__m512i) __builtin_ia32_cvtbf162iubs512_mask ((__v32bf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvtts_bf16_epi8 (__m512bh __A)
-{
- return
- (__m512i) __builtin_ia32_cvttbf162ibs512_mask ((__v32bf) __A,
- (__v32hi)
- _mm512_undefined_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvtts_bf16_epi8 (__m512i __W, __mmask32 __U, __m512bh __A)
-{
- return (__m512i) __builtin_ia32_cvttbf162ibs512_mask ((__v32bf) __A,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvtts_bf16_epi8 (__mmask32 __U, __m512bh __A)
-{
- return
- (__m512i) __builtin_ia32_cvttbf162ibs512_mask ((__v32bf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvtts_bf16_epu8 (__m512bh __A)
-{
- return (__m512i)
- __builtin_ia32_cvttbf162iubs512_mask ((__v32bf) __A,
- (__v32hi) _mm512_undefined_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvtts_bf16_epu8 (__m512i __W, __mmask32 __U, __m512bh __A)
-{
- return (__m512i) __builtin_ia32_cvttbf162iubs512_mask ((__v32bf) __A,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvtts_bf16_epu8 (__mmask32 __U, __m512bh __A)
-{
- return (__m512i)
- __builtin_ia32_cvttbf162iubs512_mask ((__v32bf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvts_ph_epi8 (__m512h __A)
-{
- return
- (__m512i) __builtin_ia32_cvtph2ibs512_mask ((__v32hf) __A,
- (__v32hi)
- _mm512_undefined_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvts_ph_epi8 (__m512i __W, __mmask32 __U, __m512h __A)
-{
- return (__m512i) __builtin_ia32_cvtph2ibs512_mask ((__v32hf) __A,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvts_ph_epi8 (__mmask32 __U, __m512h __A)
-{
- return
- (__m512i) __builtin_ia32_cvtph2ibs512_mask ((__v32hf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvts_ph_epu8 (__m512h __A)
-{
- return
- (__m512i) __builtin_ia32_cvtph2iubs512_mask ((__v32hf) __A,
- (__v32hi)
- _mm512_undefined_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvts_ph_epu8 (__m512i __W, __mmask32 __U, __m512h __A)
-{
- return (__m512i) __builtin_ia32_cvtph2iubs512_mask ((__v32hf) __A,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvts_ph_epu8 (__mmask32 __U, __m512h __A)
-{
- return
- (__m512i) __builtin_ia32_cvtph2iubs512_mask ((__v32hf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvts_ps_epi8 (__m512 __A)
-{
- return
- (__m512i) __builtin_ia32_cvtps2ibs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvts_ps_epi8 (__m512i __W, __mmask16 __U, __m512 __A)
-{
- return (__m512i) __builtin_ia32_cvtps2ibs512_mask ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvts_ps_epi8 (__mmask16 __U, __m512 __A)
-{
- return
- (__m512i) __builtin_ia32_cvtps2ibs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvts_ps_epu8 (__m512 __A)
-{
- return
- (__m512i) __builtin_ia32_cvtps2iubs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvts_ps_epu8 (__m512i __W, __mmask16 __U, __m512 __A)
-{
- return (__m512i) __builtin_ia32_cvtps2iubs512_mask ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvts_ps_epu8 (__mmask16 __U, __m512 __A)
-{
- return
- (__m512i) __builtin_ia32_cvtps2iubs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvtts_ph_epi8 (__m512h __A)
-{
- return (__m512i)
- __builtin_ia32_cvttph2ibs512_mask ((__v32hf) __A,
- (__v32hi)
- _mm512_undefined_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvtts_ph_epi8 (__m512i __W, __mmask32 __U, __m512h __A)
-{
- return (__m512i) __builtin_ia32_cvttph2ibs512_mask ((__v32hf) __A,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvtts_ph_epi8 (__mmask32 __U, __m512h __A)
-{
- return
- (__m512i) __builtin_ia32_cvttph2ibs512_mask ((__v32hf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvtts_ph_epu8 (__m512h __A)
-{
- return (__m512i)
- __builtin_ia32_cvttph2iubs512_mask ((__v32hf) __A,
- (__v32hi)
- _mm512_undefined_si512 (),
- (__mmask32) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvtts_ph_epu8 (__m512i __W, __mmask32 __U, __m512h __A)
-{
- return (__m512i) __builtin_ia32_cvttph2iubs512_mask ((__v32hf) __A,
- (__v32hi) __W,
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvtts_ph_epu8 (__mmask32 __U, __m512h __A)
-{
- return (__m512i)
- __builtin_ia32_cvttph2iubs512_mask ((__v32hf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvtts_ps_epi8 (__m512 __A)
-{
- return (__m512i)
- __builtin_ia32_cvttps2ibs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvtts_ps_epi8 (__m512i __W, __mmask16 __U, __m512 __A)
-{
- return (__m512i) __builtin_ia32_cvttps2ibs512_mask ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvtts_ps_epi8 (__mmask16 __U, __m512 __A)
-{
- return (__m512i)
- __builtin_ia32_cvttps2ibs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvtts_ps_epu8 (__m512 __A)
-{
- return (__m512i)
- __builtin_ia32_cvttps2iubs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvtts_ps_epu8 (__m512i __W, __mmask16 __U, __m512 __A)
-{
- return (__m512i) __builtin_ia32_cvttps2iubs512_mask ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvtts_ps_epu8 (__mmask16 __U, __m512 __A)
-{
- return (__m512i)
- __builtin_ia32_cvttps2iubs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_pd_epi32 (__m512d __A)
-{
- return (__m256i)
- __builtin_ia32_cvttpd2dqs512_mask ((__v8df) __A,
- (__v8si)
- _mm256_undefined_si256 (),
- (__mmask8) -1);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_pd_epi32 (__m256i __W, __mmask8 __U, __m512d __A)
-{
- return (__m256i) __builtin_ia32_cvttpd2dqs512_mask ((__v8df) __A,
- (__v8si) __W,
- (__mmask8) __U);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_pd_epi32 (__mmask8 __U, __m512d __A)
-{
- return
- (__m256i) __builtin_ia32_cvttpd2dqs512_mask ((__v8df) __A,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_pd_epi64 (__m512d __A)
-{
- return (__m512i)
- __builtin_ia32_cvttpd2qqs512_mask ((__v8df) __A,
- (__v8di)
- _mm512_undefined_si512 (),
- (__mmask8) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_pd_epi64 (__m512i __W, __mmask8 __U, __m512d __A)
-{
- return (__m512i) __builtin_ia32_cvttpd2qqs512_mask ((__v8df) __A,
- (__v8di) __W,
- (__mmask8) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_pd_epi64 (__mmask8 __U, __m512d __A)
-{
- return
- (__m512i) __builtin_ia32_cvttpd2qqs512_mask ((__v8df) __A,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) __U);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_pd_epu32 (__m512d __A)
-{
- return (__m256i)
- __builtin_ia32_cvttpd2udqs512_mask ((__v8df) __A,
- (__v8si)
- _mm256_undefined_si256 (),
- (__mmask8) -1);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_pd_epu32 (__m256i __W, __mmask8 __U, __m512d __A)
-{
- return (__m256i) __builtin_ia32_cvttpd2udqs512_mask ((__v8df) __A,
- (__v8si) __W,
- (__mmask8) __U);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_pd_epu32 (__mmask8 __U, __m512d __A)
-{
- return
- (__m256i) __builtin_ia32_cvttpd2udqs512_mask ((__v8df) __A,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_pd_epu64 (__m512d __A)
-{
- return (__m512i)
- __builtin_ia32_cvttpd2uqqs512_mask ((__v8df) __A,
- (__v8di)
- _mm512_undefined_si512 (),
- (__mmask8) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_pd_epu64 (__m512i __W, __mmask8 __U, __m512d __A)
-{
- return (__m512i) __builtin_ia32_cvttpd2uqqs512_mask ((__v8df) __A,
- (__v8di) __W,
- (__mmask8) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_pd_epu64 (__mmask8 __U, __m512d __A)
-{
- return (__m512i)
- __builtin_ia32_cvttpd2uqqs512_mask ((__v8df) __A,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_ps_epi32 (__m512 __A)
-{
- return (__m512i)
- __builtin_ia32_cvttps2dqs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_ps_epi32 (__m512i __W, __mmask16 __U, __m512 __A)
-{
- return (__m512i) __builtin_ia32_cvttps2dqs512_mask ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_ps_epi32 (__mmask16 __U, __m512 __A)
-{
- return
- (__m512i) __builtin_ia32_cvttps2dqs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_ps_epi64 (__m256 __A)
-{
- return (__m512i)
- __builtin_ia32_cvttps2qqs512_mask ((__v8sf) __A,
- (__v8di)
- _mm512_undefined_si512 (),
- (__mmask8) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_ps_epi64 (__m512i __W, __mmask8 __U, __m256 __A)
-{
- return (__m512i) __builtin_ia32_cvttps2qqs512_mask ((__v8sf) __A,
- (__v8di) __W,
- (__mmask8) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_ps_epi64 (__mmask8 __U, __m256 __A)
-{
- return
- (__m512i) __builtin_ia32_cvttps2qqs512_mask ((__v8sf) __A,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_ps_epu32 (__m512 __A)
-{
- return (__m512i)
- __builtin_ia32_cvttps2udqs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_ps_epu32 (__m512i __W, __mmask16 __U, __m512 __A)
-{
- return (__m512i) __builtin_ia32_cvttps2udqs512_mask ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_ps_epu32 (__mmask16 __U, __m512 __A)
-{
- return (__m512i)
- __builtin_ia32_cvttps2udqs512_mask ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_ps_epu64 (__m256 __A)
-{
- return (__m512i)
- __builtin_ia32_cvttps2uqqs512_mask ((__v8sf) __A,
- (__v8di)
- _mm512_undefined_si512 (),
- (__mmask8) -1);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_ps_epu64 (__m512i __W, __mmask8 __U, __m256 __A)
-{
- return (__m512i) __builtin_ia32_cvttps2uqqs512_mask ((__v8sf) __A,
- (__v8di) __W,
- (__mmask8) __U);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_ps_epu64 (__mmask8 __U, __m256 __A)
-{
- return
- (__m512i) __builtin_ia32_cvttps2uqqs512_mask ((__v8sf) __A,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) __U);
-}
-
-#ifdef __OPTIMIZE__
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvts_roundph_epi8 (__m512h __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) __A,
- (__v32hi)
- _mm512_undefined_si512 (),
- (__mmask32) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvts_roundph_epi8 (__m512i __W, __mmask32 __U, __m512h __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) __A,
- (__v32hi) __W,
- (__mmask32) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvts_roundph_epi8 (__mmask32 __U, __m512h __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvts_roundph_epu8 (__m512h __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) __A,
- (__v32hi)
- _mm512_undefined_si512 (),
- (__mmask32) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvts_roundph_epu8 (__m512i __W, __mmask32 __U, __m512h __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) __A,
- (__v32hi) __W,
- (__mmask32) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvts_roundph_epu8 (__mmask32 __U, __m512h __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvts_roundps_epi8 (__m512 __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvts_roundps_epi8 (__m512i __W, __mmask16 __U, __m512 __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvts_roundps_epi8 (__mmask16 __U, __m512 __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvts_roundps_epu8 (__m512 __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvts_roundps_epu8 (__m512i __W, __mmask16 __U, __m512 __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvts_roundps_epu8 (__mmask16 __U, __m512 __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvtts_roundph_epi8 (__m512h __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) __A,
- (__v32hi)
- _mm512_undefined_si512 (),
- (__mmask32) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvtts_roundph_epi8 (__m512i __W, __mmask32 __U, __m512h __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) __A,
- (__v32hi) __W,
- (__mmask32) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvtts_roundph_epi8 (__mmask32 __U, __m512h __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvtts_roundph_epu8 (__m512h __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) __A,
- (__v32hi)
- _mm512_undefined_si512 (),
- (__mmask32) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvtts_roundph_epu8 (__m512i __W, __mmask32 __U, __m512h __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) __A,
- (__v32hi) __W,
- (__mmask32) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvtts_roundph_epu8 (__mmask32 __U, __m512h __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) __A,
- (__v32hi)
- _mm512_setzero_si512 (),
- (__mmask32) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvtts_roundps_epi8 (__m512 __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvtts_roundps_epi8 (__m512i __W, __mmask16 __U, __m512 __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvtts_roundps_epi8 (__mmask16 __U, __m512 __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_ipcvtts_roundps_epu8 (__m512 __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_ipcvtts_roundps_epu8 (__m512i __W, __mmask16 __U, __m512 __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_ipcvtts_roundps_epu8 (__mmask16 __U, __m512 __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_roundpd_epi32 (__m512d __A, const int __R)
-{
- return (__m256i)
- __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) __A,
- (__v8si)
- _mm256_undefined_si256 (),
- (__mmask8) -1,
- __R);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_roundpd_epi32 (__m256i __W, __mmask8 __U, __m512d __A,
- const int __R)
-{
- return (__m256i) __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) __A,
- (__v8si) __W,
- (__mmask8) __U,
- __R);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_roundpd_epi32 (__mmask8 __U, __m512d __A, const int __R)
-{
- return
- (__m256i) __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) __A,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_roundpd_epi64 (__m512d __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) __A,
- (__v8di)
- _mm512_undefined_si512 (),
- (__mmask8) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_roundpd_epi64 (__m512i __W, __mmask8 __U, __m512d __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) __A,
- (__v8di) __W,
- (__mmask8) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_roundpd_epi64 (__mmask8 __U, __m512d __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) __A,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) __U,
- __R);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_roundpd_epu32 (__m512d __A, const int __R)
-{
- return (__m256i)
- __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) __A,
- (__v8si)
- _mm256_undefined_si256 (),
- (__mmask8) -1,
- __R);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_roundpd_epu32 (__m256i __W, __mmask8 __U, __m512d __A,
- const int __R)
-{
- return (__m256i) __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) __A,
- (__v8si) __W,
- (__mmask8) __U,
- __R);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_roundpd_epu32 (__mmask8 __U, __m512d __A, const int __R)
-{
- return
- (__m256i) __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) __A,
- (__v8si)
- _mm256_setzero_si256 (),
- (__mmask8) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_roundpd_epu64 (__m512d __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) __A,
- (__v8di)
- _mm512_undefined_si512 (),
- (__mmask8) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_roundpd_epu64 (__m512i __W, __mmask8 __U, __m512d __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) __A,
- (__v8di) __W,
- (__mmask8) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_roundpd_epu64 (__mmask8 __U, __m512d __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) __A,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_roundps_epi32 (__m512 __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_roundps_epi32 (__m512i __W, __mmask16 __U, __m512 __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_roundps_epi32 (__mmask16 __U, __m512 __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_roundps_epi64 (__m256 __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) __A,
- (__v8di)
- _mm512_undefined_si512 (),
- (__mmask8) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_roundps_epi64 (__m512i __W, __mmask8 __U, __m256 __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) __A,
- (__v8di) __W,
- (__mmask8) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_roundps_epi64 (__mmask8 __U, __m256 __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) __A,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_roundps_epu32 (__m512 __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_undefined_si512 (),
- (__mmask16) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_roundps_epu32 (__m512i __W, __mmask16 __U, __m512 __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) __A,
- (__v16si) __W,
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_roundps_epu32 (__mmask16 __U, __m512 __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) __A,
- (__v16si)
- _mm512_setzero_si512 (),
- (__mmask16) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_cvtts_roundps_epu64 (__m256 __A, const int __R)
-{
- return (__m512i)
- __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) __A,
- (__v8di)
- _mm512_undefined_si512 (),
- (__mmask8) -1,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_mask_cvtts_roundps_epu64 (__m512i __W, __mmask8 __U, __m256 __A,
- const int __R)
-{
- return (__m512i) __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) __A,
- (__v8di) __W,
- (__mmask8) __U,
- __R);
-}
-
-extern __inline __m512i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm512_maskz_cvtts_roundps_epu64 (__mmask8 __U, __m256 __A, const int __R)
-{
- return
- (__m512i) __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) __A,
- (__v8di)
- _mm512_setzero_si512 (),
- (__mmask8) __U,
- __R);
-}
-#else
-#define _mm512_ipcvts_roundph_epi8(A, R) \
- ((__m512i) \
- __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) (A), \
- (__v32hi) \
- (_mm512_undefined_si512 ()), \
- (__mmask32) (-1), \
- (R)))
-
-#define _mm512_mask_ipcvts_roundph_epi8(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) (A), \
- (__v32hi) (W), \
- (__mmask32) (U), \
- (R)))
-
-#define _mm512_maskz_ipcvts_roundph_epi8(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) (A), \
- (__v32hi) \
- (_mm512_setzero_si512 ()), \
- (__mmask32) (U), \
- (R)))
-
-#define _mm512_ipcvts_roundph_epu8(A, R) \
- ((__m512i) \
- __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) (A), \
- (__v32hi) \
- (_mm512_undefined_si512 ()), \
- (__mmask32) (-1), \
- (R)))
-
-#define _mm512_mask_ipcvts_roundph_epu8(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) (A), \
- (__v32hi) (W), \
- (__mmask32) (U), \
- (R)))
-
-#define _mm512_maskz_ipcvts_roundph_epu8(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) (A), \
- (__v32hi) \
- (_mm512_setzero_si512 ()), \
- (__mmask32) (U), \
- (R)))
-
-#define _mm512_ipcvts_roundps_epi8(A, R) \
- ((__m512i) \
- __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_undefined_si512 ()), \
- (__mmask16) (-1), \
- (R)))
-
-#define _mm512_mask_ipcvts_roundps_epi8(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) (A), \
- (__v16si) (W), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_maskz_ipcvts_roundps_epi8(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_setzero_si512 ()), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_ipcvts_roundps_epu8(A, R) \
- ((__m512i) \
- __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_undefined_si512 ()), \
- (__mmask16) (-1), \
- (R)))
-
-#define _mm512_mask_ipcvts_roundps_epu8(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) (A), \
- (__v16si) (W), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_maskz_ipcvts_roundps_epu8(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_setzero_si512 ()), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_ipcvtts_roundph_epi8(A, R) \
- ((__m512i) \
- __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) (A), \
- (__v32hi) \
- (_mm512_undefined_si512 ()), \
- (__mmask32) (-1), \
- (R)))
-
-#define _mm512_mask_ipcvtts_roundph_epi8(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) (A), \
- (__v32hi) (W), \
- (__mmask32) (U), \
- (R)))
-
-#define _mm512_maskz_ipcvtts_roundph_epi8(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) (A), \
- (__v32hi) \
- (_mm512_setzero_si512 ()), \
- (__mmask32) (U), \
- (R)))
-
-#define _mm512_ipcvtts_roundph_epu8(A, R) \
- ((__m512i) \
- __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) (A), \
- (__v32hi) \
- (_mm512_undefined_si512 ()), \
- (__mmask32) (-1), \
- (R)))
-
-#define _mm512_mask_ipcvtts_roundph_epu8(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) (A), \
- (__v32hi) (W), \
- (__mmask32) (U), \
- (R)))
-
-#define _mm512_maskz_ipcvtts_roundph_epu8(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) (A), \
- (__v32hi) \
- (_mm512_setzero_si512 ()), \
- (__mmask32) (U), \
- (R)))
-
-#define _mm512_ipcvtts_roundps_epi8(A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_undefined_si512 ()), \
- (__mmask16) (-1), \
- (R)))
-
-#define _mm512_mask_ipcvtts_roundps_epi8(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) (A), \
- (__v16si) (W), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_maskz_ipcvtts_roundps_epi8(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_setzero_si512 ()), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_ipcvtts_roundps_epu8(A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_undefined_si512 ()), \
- (__mmask16) (-1), \
- (R)))
-
-#define _mm512_mask_ipcvtts_roundps_epu8(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) (A), \
- (__v16si) (W), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_maskz_ipcvtts_roundps_epu8(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_setzero_si512 ()), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_cvtts_roundpd_epi32(A, R) \
- ((__m256i) \
- __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) (A), \
- (__v8si) \
- (_mm256_undefined_si256 ()), \
- (__mmask8) (-1), \
- (R)))
-
-#define _mm512_mask_cvtts_roundpd_epi32(W, U, A, R) \
- ((__m256i) __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) (A), \
- (__v8si) (W), \
- (__mmask8) (U), \
- (R)))
-
-#define _mm512_maskz_cvtts_roundpd_epi32(U, A, R) \
- ((__m256i) \
- __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) (A), \
- (__v8si) \
- (_mm256_setzero_si256 ()), \
- (__mmask8) (U), \
- (R)))
-
-#define _mm512_cvtts_roundpd_epi64(A, R) \
- ((__m512i) \
- __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) (A), \
- (__v8di) \
- (_mm512_undefined_si512 ()), \
- (__mmask8) (-1), \
- (R)))
-
-#define _mm512_mask_cvtts_roundpd_epi64(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) (A), \
- (__v8di) (W), \
- (__mmask8) (U), \
- (R)))
-
-#define _mm512_maskz_cvtts_roundpd_epi64(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) (A), \
- (__v8di) \
- (_mm512_setzero_si512 ()), \
- (__mmask8) (U), \
- (R)))
-
-#define _mm512_cvtts_roundpd_epu32(A, R) \
- ((__m256i) \
- __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) (A), \
- (__v8si) \
- (_mm256_undefined_si256 ()), \
- (__mmask8) (-1), \
- (R)))
-
-#define _mm512_mask_cvtts_roundpd_epu32(W, U, A, R) \
- ((__m256i) __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) (A), \
- (__v8si) (W), \
- (__mmask8) (U), \
- (R)))
-
-#define _mm512_maskz_cvtts_roundpd_epu32(U, A, R) \
- ((__m256i) \
- __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) (A), \
- (__v8si) \
- (_mm256_setzero_si256 ()), \
- (__mmask8) (U), \
- (R)))
-
-#define _mm512_cvtts_roundpd_epu64(A, R) \
- ((__m512i) \
- __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) (A), \
- (__v8di) \
- (_mm512_undefined_si512 ()), \
- (__mmask8) (-1), \
- (R)))
-
-#define _mm512_mask_cvtts_roundpd_epu64(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) (A), \
- (__v8di) (W), \
- (__mmask8) (U), \
- (R)))
-
-#define _mm512_maskz_cvtts_roundpd_epu64(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) (A), \
- (__v8di) \
- (_mm512_setzero_si512 ()), \
- (__mmask8) (U), \
- (R)))
-
-#define _mm512_cvtts_roundps_epi32(A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_undefined_si512 ()), \
- (__mmask16) (-1), \
- (R)))
-
-#define _mm512_mask_cvtts_roundps_epi32(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) (A), \
- (__v16si) (W), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_maskz_cvtts_roundps_epi32(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_setzero_si512 ()), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_cvtts_roundps_epi64(A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) (A), \
- (__v8di) \
- (_mm512_undefined_si512 ()), \
- (__mmask8) (-1), \
- (R)))
-
-#define _mm512_mask_cvtts_roundps_epi64(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) (A), \
- (__v8di) (W), \
- (__mmask8) (U), \
- (R)))
-
-#define _mm512_maskz_cvtts_roundps_epi64(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) (A), \
- (__v8di) \
- (_mm512_setzero_si512 ()), \
- (__mmask8) (U), \
- (R)))
-
-#define _mm512_cvtts_roundps_epu32(A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_undefined_si512 ()), \
- (__mmask16) (-1), \
- (R)))
-
-#define _mm512_mask_cvtts_roundps_epu32(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) (A), \
- (__v16si) (W), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_maskz_cvtts_roundps_epu32(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) (A), \
- (__v16si) \
- (_mm512_setzero_si512 ()), \
- (__mmask16) (U), \
- (R)))
-
-#define _mm512_cvtts_roundps_epu64(A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) (A), \
- (__v8di) \
- (_mm512_undefined_si512 ()), \
- (__mmask8) (-1), \
- (R)))
-
-#define _mm512_mask_cvtts_roundps_epu64(W, U, A, R) \
- ((__m512i) __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) (A), \
- (__v8di) (W), \
- (__mmask8) (U), \
- (R)))
-
-#define _mm512_maskz_cvtts_roundps_epu64(U, A, R) \
- ((__m512i) \
- __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) (A), \
- (__v8di) \
- (_mm512_setzero_si512 ()), \
- (__mmask8) (U), \
- (R)))
-#endif
-
-#ifdef __DISABLE_AVX10_2__
-#undef __DISABLE_AVX10_2__
-#pragma GCC pop_options
-#endif /* __DISABLE_AVX10_2__ */
-
-#endif /* _AVX10_2_512SATCVTINTRIN_H_INCLUDED */
#define __DISABLE_AVX10_2__
#endif /* __AVX10_2__ */
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_add_pbh (__m512bh __A, __m512bh __B)
+{
+ return (__m512bh) __builtin_ia32_addbf16512 (__A, __B);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_add_pbh (__m512bh __W, __mmask32 __U,
+ __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_addbf16512_mask (__A, __B, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_add_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_addbf16512_mask (__A, __B,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_add_pbh (__m256bh __A, __m256bh __B)
__U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_sub_pbh (__m512bh __A, __m512bh __B)
+{
+ return (__m512bh) __builtin_ia32_subbf16512 (__A, __B);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_sub_pbh (__m512bh __W, __mmask32 __U,
+ __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_subbf16512_mask (__A, __B, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_sub_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_subbf16512_mask (__A, __B,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_sub_pbh (__m256bh __A, __m256bh __B)
__U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mul_pbh (__m512bh __A, __m512bh __B)
+{
+ return (__m512bh) __builtin_ia32_mulbf16512 (__A, __B);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_mul_pbh (__m512bh __W, __mmask32 __U,
+ __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_mulbf16512_mask (__A, __B, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_mul_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_mulbf16512_mask (__A, __B,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_mul_pbh (__m256bh __A, __m256bh __B)
__U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_div_pbh (__m512bh __A, __m512bh __B)
+{
+ return (__m512bh) __builtin_ia32_divbf16512 (__A, __B);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_div_pbh (__m512bh __W, __mmask32 __U,
+ __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_divbf16512_mask (__A, __B, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_div_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_divbf16512_mask (__A, __B,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_div_pbh (__m256bh __A, __m256bh __B)
__U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_max_pbh (__m512bh __A, __m512bh __B)
+{
+ return (__m512bh) __builtin_ia32_maxbf16512 (__A, __B);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_max_pbh (__m512bh __W, __mmask32 __U,
+ __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_maxbf16512_mask (__A, __B, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_max_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_maxbf16512_mask (__A, __B,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_max_pbh (__m256bh __A, __m256bh __B)
__U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_min_pbh (__m512bh __A, __m512bh __B)
+{
+ return (__m512bh) __builtin_ia32_minbf16512 (__A, __B);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_min_pbh (__m512bh __W, __mmask32 __U,
+ __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_minbf16512_mask (__A, __B, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_min_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_minbf16512_mask (__A, __B,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_min_pbh (__m256bh __A, __m256bh __B)
__U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_scalef_pbh (__m512bh __A, __m512bh __B)
+{
+ return (__m512bh) __builtin_ia32_scalefbf16512 (__A, __B);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_scalef_pbh (__m512bh __W, __mmask32 __U,
+ __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_scalefbf16512_mask (__A, __B, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_scalef_pbh (__mmask32 __U, __m512bh __A, __m512bh __B)
+{
+ return (__m512bh)
+ __builtin_ia32_scalefbf16512_mask (__A, __B,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_scalef_pbh (__m256bh __A, __m256bh __B)
__U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmadd_pbh (__m512bh __A, __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fmaddbf16512_mask (__A, __B, __C, (__mmask32) -1);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmadd_pbh (__m512bh __A, __mmask32 __U,
+ __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fmaddbf16512_mask (__A, __B, __C, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmadd_pbh (__m512bh __A, __m512bh __B,
+ __m512bh __C, __mmask32 __U)
+{
+ return (__m512bh)
+ __builtin_ia32_fmaddbf16512_mask3 (__A, __B, __C, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmadd_pbh (__mmask32 __U, __m512bh __A,
+ __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fmaddbf16512_maskz (__A, __B, __C, __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_fmadd_pbh (__m256bh __A, __m256bh __B, __m256bh __C)
__builtin_ia32_fmaddbf16128_maskz (__A, __B, __C, __U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmsub_pbh (__m512bh __A, __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fmsubbf16512_mask (__A, __B, __C, (__mmask32) -1);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmsub_pbh (__m512bh __A, __mmask32 __U,
+ __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fmsubbf16512_mask (__A, __B, __C, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmsub_pbh (__m512bh __A, __m512bh __B,
+ __m512bh __C, __mmask32 __U)
+{
+ return (__m512bh)
+ __builtin_ia32_fmsubbf16512_mask3 (__A, __B, __C, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmsub_pbh (__mmask32 __U, __m512bh __A,
+ __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fmsubbf16512_maskz (__A, __B, __C, __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_fmsub_pbh (__m256bh __A, __m256bh __B, __m256bh __C)
__builtin_ia32_fmsubbf16128_maskz (__A, __B, __C, __U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fnmadd_pbh (__m512bh __A, __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fnmaddbf16512_mask (__A, __B, __C, (__mmask32) -1);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fnmadd_pbh (__m512bh __A, __mmask32 __U,
+ __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fnmaddbf16512_mask (__A, __B, __C, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fnmadd_pbh (__m512bh __A, __m512bh __B,
+ __m512bh __C, __mmask32 __U)
+{
+ return (__m512bh)
+ __builtin_ia32_fnmaddbf16512_mask3 (__A, __B, __C, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fnmadd_pbh (__mmask32 __U, __m512bh __A,
+ __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fnmaddbf16512_maskz (__A, __B, __C, __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_fnmadd_pbh (__m256bh __A, __m256bh __B, __m256bh __C)
__builtin_ia32_fnmaddbf16128_maskz (__A, __B, __C, __U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fnmsub_pbh (__m512bh __A, __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fnmsubbf16512_mask (__A, __B, __C, (__mmask32) -1);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fnmsub_pbh (__m512bh __A, __mmask32 __U,
+ __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fnmsubbf16512_mask (__A, __B, __C, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fnmsub_pbh (__m512bh __A, __m512bh __B,
+ __m512bh __C, __mmask32 __U)
+{
+ return (__m512bh)
+ __builtin_ia32_fnmsubbf16512_mask3 (__A, __B, __C, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fnmsub_pbh (__mmask32 __U, __m512bh __A,
+ __m512bh __B, __m512bh __C)
+{
+ return (__m512bh)
+ __builtin_ia32_fnmsubbf16512_maskz (__A, __B, __C, __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_fnmsub_pbh (__m256bh __A, __m256bh __B, __m256bh __C)
__builtin_ia32_fnmsubbf16128_maskz (__A, __B, __C, __U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_rsqrt_pbh (__m512bh __A)
+{
+ return (__m512bh)
+ __builtin_ia32_rsqrtbf16512_mask (__A,
+ (__v32bf) _mm512_setzero_si512 (),
+ (__mmask32) -1);
+
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_rsqrt_pbh (__m512bh __W, __mmask32 __U, __m512bh __A)
+{
+ return (__m512bh)
+ __builtin_ia32_rsqrtbf16512_mask (__A, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_rsqrt_pbh (__mmask32 __U, __m512bh __A)
+{
+ return (__m512bh)
+ __builtin_ia32_rsqrtbf16512_mask (__A,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_rsqrt_pbh (__m256bh __A)
__U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_sqrt_pbh (__m512bh __A)
+{
+ return (__m512bh)
+ __builtin_ia32_sqrtbf16512_mask (__A,
+ (__v32bf) _mm512_setzero_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_sqrt_pbh (__m512bh __W, __mmask32 __U, __m512bh __A)
+{
+ return (__m512bh)
+ __builtin_ia32_sqrtbf16512_mask (__A, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_sqrt_pbh (__mmask32 __U, __m512bh __A)
+{
+ return (__m512bh)
+ __builtin_ia32_sqrtbf16512_mask (__A,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_sqrt_pbh (__m256bh __A)
__U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_rcp_pbh (__m512bh __A)
+{
+ return (__m512bh)
+ __builtin_ia32_rcpbf16512_mask (__A,
+ (__v32bf) _mm512_setzero_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_rcp_pbh (__m512bh __W, __mmask32 __U, __m512bh __A)
+{
+ return (__m512bh)
+ __builtin_ia32_rcpbf16512_mask (__A, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_rcp_pbh (__mmask32 __U, __m512bh __A)
+{
+ return (__m512bh)
+ __builtin_ia32_rcpbf16512_mask (__A,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_rcp_pbh (__m256bh __A)
__U);
}
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_getexp_pbh (__m512bh __A)
+{
+ return (__m512bh)
+ __builtin_ia32_getexpbf16512_mask (__A,
+ (__v32bf) _mm512_setzero_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_getexp_pbh (__m512bh __W, __mmask32 __U, __m512bh __A)
+{
+ return (__m512bh) __builtin_ia32_getexpbf16512_mask (__A, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_getexp_pbh (__mmask32 __U, __m512bh __A)
+{
+ return (__m512bh)
+ __builtin_ia32_getexpbf16512_mask (__A,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_getexp_pbh (__m256bh __A)
/* Intrinsics vrndscalebf16. */
#ifdef __OPTIMIZE__
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_roundscale_pbh (__m512bh __A, int B)
+{
+ return (__m512bh)
+ __builtin_ia32_rndscalebf16512_mask (__A, B,
+ (__v32bf) _mm512_setzero_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_roundscale_pbh (__m512bh __W, __mmask32 __U, __m512bh __A, int B)
+{
+ return (__m512bh)
+ __builtin_ia32_rndscalebf16512_mask (__A, B, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_roundscale_pbh (__mmask32 __U, __m512bh __A, int B)
+{
+ return (__m512bh)
+ __builtin_ia32_rndscalebf16512_mask (__A, B,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_roundscale_pbh (__m256bh __A, int B)
}
#else
+#define _mm512_roundscale_pbh(A, B) \
+ (__builtin_ia32_rndscalebf16512_mask ((A), (B), \
+ (__v32bf) _mm512_setzero_si512 (), \
+ (__mmask32) -1))
+
+#define _mm512_mask_roundscale_pbh(A, B, C, D) \
+ (__builtin_ia32_rndscalebf16512_mask ((C), (D), (A), (B)))
+
+#define _mm512_maskz_roundscale_pbh(A, B, C) \
+ (__builtin_ia32_rndscalebf16512_mask ((B), (C), \
+ (__v32bf) _mm512_setzero_si512 (), \
+ (A)))
+
#define _mm256_roundscale_pbh(A, B) \
(__builtin_ia32_rndscalebf16256_mask ((A), (B), \
(__v16bf) _mm256_setzero_si256 (), \
/* Intrinsics vreducebf16. */
#ifdef __OPTIMIZE__
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_reduce_pbh (__m512bh __A, int B)
+{
+ return (__m512bh)
+ __builtin_ia32_reducebf16512_mask (__A, B,
+ (__v32bf) _mm512_setzero_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_reduce_pbh (__m512bh __W, __mmask32 __U,
+ __m512bh __A, int B)
+{
+ return (__m512bh)
+ __builtin_ia32_reducebf16512_mask (__A, B, __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_reduce_pbh (__mmask32 __U, __m512bh __A, int B)
+{
+ return (__m512bh)
+ __builtin_ia32_reducebf16512_mask (__A, B,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_reduce_pbh (__m256bh __A, int B)
}
#else
+#define _mm512_reduce_pbh(A, B) \
+ (__builtin_ia32_reducebf16512_mask ((A), (B), \
+ (__v32bf) _mm512_setzero_si512 (), \
+ (__mmask32) -1))
+
+#define _mm512_mask_reduce_pbh(A, B, C, D) \
+ (__builtin_ia32_reducebf16512_mask ((C), (D), (A), (B)))
+
+#define _mm512_maskz_reduce_pbh(A, B, C) \
+ (__builtin_ia32_reducebf16512_mask ((B), (C), \
+ (__v32bf) _mm512_setzero_si512 (), \
+ (A)))
+
#define _mm256_reduce_pbh(A, B) \
(__builtin_ia32_reducebf16256_mask ((A), (B), \
(__v16bf) _mm256_setzero_si256 (), \
/* Intrinsics vgetmantbf16. */
#ifdef __OPTIMIZE__
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_getmant_pbh (__m512bh __A, _MM_MANTISSA_NORM_ENUM __B,
+ _MM_MANTISSA_SIGN_ENUM __C)
+{
+ return (__m512bh)
+ __builtin_ia32_getmantbf16512_mask (__A, (int) (__C << 2) | __B,
+ (__v32bf) _mm512_setzero_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_getmant_pbh (__m512bh __W, __mmask32 __U, __m512bh __A,
+ _MM_MANTISSA_NORM_ENUM __B,
+ _MM_MANTISSA_SIGN_ENUM __C)
+{
+ return (__m512bh)
+ __builtin_ia32_getmantbf16512_mask (__A, (int) (__C << 2) | __B,
+ __W, __U);
+}
+
+extern __inline__ __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_getmant_pbh (__mmask32 __U, __m512bh __A,
+ _MM_MANTISSA_NORM_ENUM __B,
+ _MM_MANTISSA_SIGN_ENUM __C)
+{
+ return (__m512bh)
+ __builtin_ia32_getmantbf16512_mask (__A, (int) (__C << 2) | __B,
+ (__v32bf) _mm512_setzero_si512 (),
+ __U);
+}
+
extern __inline__ __m256bh
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_getmant_pbh (__m256bh __A, _MM_MANTISSA_NORM_ENUM __B,
}
#else
+#define _mm512_getmant_pbh(A, B, C) \
+ (__builtin_ia32_getmantbf16512_mask ((A), (int)(((C)<<2) | (B)), \
+ (__v32bf) _mm512_setzero_si512 (), \
+ (__mmask32) -1))
+
+#define _mm512_mask_getmant_pbh(A, B, C, D, E) \
+ (__builtin_ia32_getmantbf16512_mask ((C), (int)(((D)<<2) | (E)), (A), (B)))
+
+#define _mm512_maskz_getmant_pbh(A, B, C, D) \
+ (__builtin_ia32_getmantbf16512_mask ((B), (int)(((C)<<2) | (D)), \
+ (__v32bf) _mm512_setzero_si512 (), \
+ (A)))
+
#define _mm256_getmant_pbh(A, B, C) \
(__builtin_ia32_getmantbf16256_mask ((A), (int)(((C)<<2) | (B)), \
(__v16bf) _mm256_setzero_si256 (), \
/* Intrinsics vfpclassbf16. */
#ifdef __OPTIMIZE__
+extern __inline __mmask32
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fpclass_pbh_mask (__mmask32 __U, __m512bh __A,
+ const int __imm)
+{
+ return (__mmask32)
+ __builtin_ia32_fpclassbf16512_mask (__A, __imm, __U);
+}
+
+extern __inline __mmask32
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fpclass_pbh_mask (__m512bh __A, const int __imm)
+{
+ return (__mmask32)
+ __builtin_ia32_fpclassbf16512_mask (__A, __imm,
+ (__mmask32) -1);
+}
+
extern __inline __mmask16
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_mask_fpclass_pbh_mask (__mmask16 __U, __m256bh __A,
}
#else
+#define _mm512_mask_fpclass_pbh_mask(U, X, C) \
+ ((__mmask32) __builtin_ia32_fpclassbf16512_mask ( \
+ (__v32bf) (__m512bh) (X), (int) (C), (__mmask32) (U)))
+
+#define _mm512_fpclass_pbh_mask(X, C) \
+ ((__mmask32) __builtin_ia32_fpclassbf16512_mask ( \
+ (__v32bf) (__m512bh) (X), (int) (C), (__mmask32) (-1)))
+
#define _mm256_mask_fpclass_pbh_mask(U, A, B) \
((__mmask16) __builtin_ia32_fpclassbf16256_mask ((A), (B), (U)))
/* Intrinsics vcmpbf16. */
#ifdef __OPTIMIZE__
+extern __inline __mmask32
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cmp_pbh_mask (__mmask32 __U, __m512bh __A, __m512bh __B,
+ const int __imm)
+{
+ return (__mmask32)
+ __builtin_ia32_cmpbf16512_mask (__A, __B, __imm, __U);
+}
+
+extern __inline __mmask32
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cmp_pbh_mask (__m512bh __A, __m512bh __B, const int __imm)
+{
+ return (__mmask32)
+ __builtin_ia32_cmpbf16512_mask (__A, __B, __imm,
+ (__mmask32) -1);
+}
+
extern __inline __mmask16
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_mask_cmp_pbh_mask (__mmask16 __U, __m256bh __A,
}
#else
+#define _mm512_mask_cmp_pbh_mask(A, B, C, D) \
+ ((__mmask32) __builtin_ia32_cmpbf16512_mask ((B), (C), (D), (A)))
+
+#define _mm512_cmp_pbh_mask(A, B, C) \
+ ((__mmask32) __builtin_ia32_cmpbf16512_mask ((A), (B), (C), (-1)))
+
#define _mm256_mask_cmp_pbh_mask(A, B, C, D) \
((__mmask16) __builtin_ia32_cmpbf16256_mask ((B), (C), (D), (A)))
(__mmask16) __U);
}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtx2ps_ph (__m512 __A, __m512 __B)
+{
+ return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ (__v32hf)
+ _mm512_setzero_ph (),
+ (__mmask32) -1,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtx2ps_ph (__m512h __W, __mmask32 __U, __m512 __A,
+ __m512 __B)
+{
+ return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ (__v32hf) __W,
+ (__mmask32) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtx2ps_ph (__mmask32 __U, __m512 __A, __m512 __B)
+{
+ return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ (__v32hf)
+ _mm512_setzero_ph (),
+ (__mmask32) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtx_round2ps_ph (__m512 __A, __m512 __B, const int __R)
+{
+ return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ (__v32hf)
+ _mm512_setzero_ph (),
+ (__mmask32) -1,
+ __R);
+}
+
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtx_round2ps_ph (__m512h __W, __mmask32 __U, __m512 __A,
+ __m512 __B, const int __R)
+{
+ return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ (__v32hf) __W,
+ (__mmask32) __U,
+ __R);
+}
+
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtx_round2ps_ph (__mmask32 __U, __m512 __A,
+ __m512 __B, const int __R)
+{
+ return (__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ (__v32hf)
+ _mm512_setzero_ph (),
+ (__mmask32) __U,
+ __R);
+}
+
+#else
+#define _mm512_cvtx_round2ps_ph(A, B, R) \
+ ((__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) (A), \
+ (__v16sf) (B), \
+ (__v32hf) \
+ (_mm512_setzero_ph ()), \
+ (__mmask32) (-1), \
+ (R)))
+#define _mm512_mask_cvtx_round2ps_ph(W, U, A, B, R) \
+ ((__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) (A), \
+ (__v16sf) (B), \
+ (__v32hf) (W), \
+ (__mmask32) (U), \
+ (R)))
+#define _mm512_maskz_cvtx_round2ps_ph(U, A, B, R) \
+ ((__m512h) __builtin_ia32_vcvt2ps2phx512_mask_round ((__v16sf) (A), \
+ (__v16sf) (B), \
+ (__v32hf) \
+ (_mm512_setzero_ph ()), \
+ (__mmask32) (U), \
+ (R)))
+#endif /* __OPTIMIZE__ */
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtbiasph_bf8 (__m128i __A, __m128h __B)
(__mmask16) __U);
}
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtbiasph_bf8 (__m512i __A, __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2bf8512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i)
+ _mm256_undefined_si256 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtbiasph_bf8 (__m256i __W, __mmask32 __U,
+ __m512i __A, __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2bf8512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i) __W,
+ (__mmask32) __U);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtbiasph_bf8 (__mmask32 __U, __m512i __A, __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2bf8512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i)
+ _mm256_setzero_si256 (),
+ (__mmask32) __U);
+}
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvts_biasph_bf8 (__m128i __A, __m128h __B)
(__mmask16) __U);
}
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvts_biasph_bf8 (__m512i __A, __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2bf8s512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i)
+ _mm256_undefined_si256 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvts_biasph_bf8 (__m256i __W, __mmask32 __U,
+ __m512i __A, __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2bf8s512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i) __W,
+ (__mmask32) __U);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvts_biasph_bf8 (__mmask32 __U, __m512i __A, __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2bf8s512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i)
+ _mm256_setzero_si256 (),
+ (__mmask32) __U);
+}
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtbiasph_hf8 (__m128i __A, __m128h __B)
(__mmask16) __U);
}
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtbiasph_hf8 (__m512i __A, __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2hf8512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i)
+ _mm256_undefined_si256 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtbiasph_hf8 (__m256i __W, __mmask32 __U, __m512i __A,
+ __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2hf8512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i) __W,
+ (__mmask32) __U);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtbiasph_hf8 (__mmask32 __U, __m512i __A, __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2hf8512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i)
+ _mm256_setzero_si256 (),
+ (__mmask32) __U);
+}
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvts_biasph_hf8 (__m128i __A, __m128h __B)
(__mmask16) __U);
}
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvts_biasph_hf8 (__m512i __A, __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2hf8s512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i)
+ _mm256_undefined_si256 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvts_biasph_hf8 (__m256i __W, __mmask32 __U,
+ __m512i __A, __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2hf8s512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i) __W,
+ (__mmask32) __U);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvts_biasph_hf8 (__mmask32 __U, __m512i __A, __m512h __B)
+{
+ return (__m256i) __builtin_ia32_vcvtbiasph2hf8s512_mask ((__v64qi) __A,
+ (__v32hf) __B,
+ (__v32qi)(__m256i)
+ _mm256_setzero_si256 (),
+ (__mmask32) __U);
+}
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvt2ph_bf8 (__m128h __A, __m128h __B)
(__mmask32) __U);
}
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt2ph_bf8 (__m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2bf8512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi)
+ _mm512_setzero_si512 (),
+ (__mmask64) -1);
+}
+
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt2ph_bf8 (__m512i __W, __mmask64 __U,
+ __m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2bf8512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi) __W,
+ (__mmask64) __U);
+}
+
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt2ph_bf8 (__mmask64 __U, __m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2bf8512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi)
+ _mm512_setzero_si512 (),
+ (__mmask64) __U);
+}
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvts_2ph_bf8 (__m128h __A, __m128h __B)
(__mmask32) __U);
}
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvts_2ph_bf8 (__m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2bf8s512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi)
+ _mm512_setzero_si512 (),
+ (__mmask64) -1);
+}
+
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvts_2ph_bf8 (__m512i __W, __mmask64 __U,
+ __m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2bf8s512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi) __W,
+ (__mmask64) __U);
+}
+
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvts_2ph_bf8 (__mmask64 __U, __m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2bf8s512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi)
+ _mm512_setzero_si512 (),
+ (__mmask64) __U);
+}
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvt2ph_hf8 (__m128h __A, __m128h __B)
(__mmask32) __U);
}
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt2ph_hf8 (__m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2hf8512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi)
+ _mm512_setzero_si512 (),
+ (__mmask64) -1);
+}
+
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt2ph_hf8 (__m512i __W, __mmask64 __U,
+ __m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2hf8512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi) __W,
+ (__mmask64) __U);
+}
+
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt2ph_hf8 (__mmask64 __U, __m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2hf8512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi)
+ _mm512_setzero_si512 (),
+ (__mmask64) __U);
+}
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvts_2ph_hf8 (__m128h __A, __m128h __B)
(__mmask32) __U);
}
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvts_2ph_hf8 (__m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2hf8s512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi)
+ _mm512_setzero_si512 (),
+ (__mmask64) -1);
+}
+
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvts_2ph_hf8 (__m512i __W, __mmask64 __U,
+ __m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2hf8s512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi) __W,
+ (__mmask64) __U);
+}
+
+extern __inline__ __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvts_2ph_hf8 (__mmask64 __U, __m512h __A, __m512h __B)
+{
+ return (__m512i) __builtin_ia32_vcvt2ph2hf8s512_mask ((__v32hf) __A,
+ (__v32hf) __B,
+ (__v64qi)
+ _mm512_setzero_si512 (),
+ (__mmask64) __U);
+}
+
extern __inline__ __m128h
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvthf8_ph (__m128i __A)
(__mmask16) __U);
}
+extern __inline__ __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvthf8_ph (__m256i __A)
+{
+ return (__m512h) __builtin_ia32_vcvthf82ph512_mask ((__v32qi) __A,
+ (__v32hf) (__m512h)
+ _mm512_undefined_ph (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvthf8_ph (__m512h __W, __mmask32 __U, __m256i __A)
+{
+ return (__m512h) __builtin_ia32_vcvthf82ph512_mask ((__v32qi) __A,
+ (__v32hf) (__m512h) __W,
+ (__mmask32) __U);
+}
+
+extern __inline__ __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvthf8_ph (__mmask32 __U, __m256i __A)
+{
+ return (__m512h) __builtin_ia32_vcvthf82ph512_mask ((__v32qi) __A,
+ (__v32hf) (__m512h)
+ _mm512_setzero_ph (),
+ (__mmask32) __U);
+}
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtph_bf8 (__m128h __A)
(__mmask16) __U);
}
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtph_bf8 (__m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2bf8512_mask ((__v32hf) __A,
+ (__v32qi) (__m256i)
+ _mm256_undefined_si256 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtph_bf8 (__m256i __W, __mmask32 __U, __m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2bf8512_mask ((__v32hf) __A,
+ (__v32qi) (__m256i) __W,
+ (__mmask32) __U);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtph_bf8 (__mmask32 __U, __m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2bf8512_mask ((__v32hf) __A,
+ (__v32qi) (__m256i)
+ _mm256_setzero_si256 (),
+ (__mmask32) __U);
+}
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvts_ph_bf8 (__m128h __A)
(__mmask16) __U);
}
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvts_ph_bf8 (__m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2bf8s512_mask ((__v32hf) __A,
+ (__v32qi) (__m256i)
+ _mm256_undefined_si256 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvts_ph_bf8 (__m256i __W, __mmask32 __U, __m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2bf8s512_mask ((__v32hf) __A,
+ (__v32qi) (__m256i) __W,
+ (__mmask32) __U);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvts_ph_bf8 (__mmask32 __U, __m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2bf8s512_mask ((__v32hf) __A,
+ (__v32qi) (__m256i)
+ _mm256_setzero_si256 (),
+ (__mmask32) __U);
+}
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtph_hf8 (__m128h __A)
(__mmask16) __U);
}
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtph_hf8 (__m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2hf8512_mask ((__v32hf) __A,
+ (__v32qi) (__m256i)
+ _mm256_undefined_si256 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtph_hf8 (__m256i __W, __mmask32 __U, __m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2hf8512_mask ((__v32hf) __A,
+ (__v32qi)(__m256i) __W,
+ (__mmask32) __U);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtph_hf8 (__mmask32 __U, __m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2hf8512_mask ((__v32hf) __A,
+ (__v32qi) (__m256i)
+ _mm256_setzero_si256 (),
+ (__mmask32) __U);
+}
+
extern __inline__ __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvts_ph_hf8 (__m128h __A)
(__mmask16) __U);
}
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvts_ph_hf8 (__m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2hf8s512_mask ((__v32hf) __A,
+ (__v32qi) (__m256i)
+ _mm256_undefined_si256 (),
+ (__mmask32) -1);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvts_ph_hf8 (__m256i __W, __mmask32 __U, __m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2hf8s512_mask ((__v32hf) __A,
+ (__v32qi) (__m256i) __W,
+ (__mmask32) __U);
+}
+
+extern __inline__ __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvts_ph_hf8 (__mmask32 __U, __m512h __A)
+{
+ return (__m256i) __builtin_ia32_vcvtph2hf8s512_mask ((__v32hf) __A,
+ (__v32qi) (__m256i)
+ _mm256_setzero_si256 (),
+ (__mmask32) __U);
+}
+
extern __inline __m128h
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtbf8_ph (__m128i __A)
(__m256i) _mm256_maskz_cvtepi8_epi16 (__U, __A), 8));
}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtbf8_ph (__m256i __A)
+{
+ return (__m512h) _mm512_castsi512_ph ((__m512i) _mm512_slli_epi16 (
+ (__m512i) _mm512_cvtepi8_epi16 (__A), 8));
+}
+
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtbf8_ph (__m512h __S, __mmask32 __U, __m256i __A)
+{
+ return (__m512h) _mm512_castsi512_ph ((__m512i) _mm512_mask_slli_epi16 (
+ (__m512i) __S, __U, (__m512i) _mm512_cvtepi8_epi16 (__A), 8));
+}
+
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtbf8_ph (__mmask32 __U, __m256i __A)
+{
+ return (__m512h) _mm512_castsi512_ph ((__m512i) _mm512_slli_epi16 (
+ (__m512i) _mm512_maskz_cvtepi8_epi16 (__U, __A), 8));
+}
+
#ifdef __DISABLE_AVX10_2__
#undef __DISABLE_AVX10_2__
#pragma GCC pop_options
(__mmask8) __U);
}
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpbssd_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbssd512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpbssd_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbssd_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpbssd_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbssd_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpbssds_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbssds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpbssds_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbssds_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpbssds_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbssds_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpbsud_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbsud512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpbsud_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbsud_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpbsud_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbsud_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpbsuds_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbsuds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpbsuds_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbsuds_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpbsuds_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbsuds_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpbuud_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbuud512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpbuud_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbuud_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpbuud_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbuud_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpbuuds_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbuuds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpbuuds_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbuuds_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpbuuds_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpbuuds_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
extern __inline __m128i
__attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm_mask_dpwsud_epi32 (__m128i __W, __mmask8 __U,
(__mmask8) __U);
}
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpwsud_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwsud512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpwsud_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwsud_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpwsud_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwsud_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpwsuds_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwsuds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpwsuds_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwsuds_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpwsuds_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwsuds_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpwusd_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwusd512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpwusd_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwusd_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpwusd_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwusd_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpwusds_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwusds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpwusds_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwusds_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpwusds_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwusds_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpwuud_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwuud512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpwuud_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwuud_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpwuud_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwuud_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpwuuds_epi32 (__m512i __W, __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwuuds512 ((__v16si) __W, (__v16si) __A, (__v16si) __B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpwuuds_epi32 (__m512i __W, __mmask16 __U,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwuuds_v16si_mask ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpwuuds_epi32 (__mmask16 __U, __m512i __W,
+ __m512i __A, __m512i __B)
+{
+ return (__m512i)
+ __builtin_ia32_vpdpwuuds_v16si_maskz ((__v16si) __W,
+ (__v16si) __A,
+ (__v16si) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpph_ps (__m512 __W, __m512h __A, __m512h __B)
+{
+ return (__m512)
+ __builtin_ia32_vdpphps512_mask ((__v16sf) __W,
+ (__v16sf) __A,
+ (__v16sf) __B,
+ (__mmask16) -1);
+}
+
+extern __inline __m512
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpph_ps (__m512 __W, __mmask16 __U, __m512h __A,
+ __m512h __B)
+{
+ return (__m512)
+ __builtin_ia32_vdpphps512_mask ((__v16sf) __W,
+ (__v16sf) __A,
+ (__v16sf) __B,
+ (__mmask16) __U);
+}
+
+extern __inline __m512
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpph_ps (__mmask16 __U, __m512 __W, __m512h __A,
+ __m512h __B)
+{
+ return (__m512)
+ __builtin_ia32_vdpphps512_maskz ((__v16sf) __W,
+ (__v16sf) __A,
+ (__v16sf) __B,
+ (__mmask16) __U);
+}
+
extern __inline __m256
__attribute__((__gnu_inline__, __always_inline__, __artificial__))
_mm256_dpph_ps (__m256 __W, __m256h __A, __m256h __B)
(__v16hi) _mm256_setzero_si256 (),
__U);
}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mpsadbw_epu8 (__m512i __X, __m512i __Y, const int __M)
+{
+ return (__m512i) __builtin_ia32_mpsadbw512 ((__v64qi) __X,
+ (__v64qi) __Y,
+ __M);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_mpsadbw_epu8 (__m512i __W, __mmask32 __U, __m512i __X,
+ __m512i __Y, const int __M)
+{
+ return (__m512i) __builtin_ia32_mpsadbw512_mask ((__v64qi) __X,
+ (__v64qi) __Y,
+ __M,
+ (__v32hi) __W,
+ __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_mpsadbw_epu8 (__mmask32 __U, __m512i __X,
+ __m512i __Y, const int __M)
+{
+ return (__m512i) __builtin_ia32_mpsadbw512_mask ((__v64qi) __X,
+ (__v64qi) __Y,
+ __M,
+ (__v32hi) _mm512_setzero_epi32 (),
+ __U);
+}
#else
#define _mm_mask_mpsadbw_epu8(W, U, X, Y, M) \
(__m128i) __builtin_ia32_mpsadbw128_mask ((__v16qi)(__m128i)(X), \
(__v16hi) _mm256_setzero_si256 (), \
(__mmask16)(U))
+#define _mm512_mpsadbw_epu8(X, Y, M) \
+ (__m512i) __builtin_ia32_mpsadbw512 ((__v64qi)(__m512i)(X), \
+ (__v64qi)(__m512i)(Y), (int)(M))
+
+#define _mm512_mask_mpsadbw_epu8(W, U, X, Y, M) \
+ (__m512i) __builtin_ia32_mpsadbw512_mask ((__v64qi)(__m512i)(X), \
+ (__v64qi)(__m512i)(Y), \
+ (int)(M), \
+ (__v32hi)(__m512i)(W), \
+ (__mmask32)(U))
+
+#define _mm512_maskz_mpsadbw_epu8(U, X, Y, M) \
+ (__m512i) __builtin_ia32_mpsadbw512_mask ((__v64qi)(__m512i)(X), \
+ (__v64qi)(__m512i)(Y), \
+ (int)(M), \
+ (__v32hi) _mm512_setzero_epi32 (), \
+ (__mmask32)(U))
#endif
#ifdef __DISABLE_AVX10_2__
(__mmask16) __U);
}
+extern __inline __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_minmax_pbh (__m512bh __A, __m512bh __B, const int __C)
+{
+ return (__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) __A,
+ (__v32bf) __B,
+ __C,
+ (__v32bf)(__m512bh)
+ _mm512_setzero_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_minmax_pbh (__m512bh __W, __mmask32 __U,
+ __m512bh __A, __m512bh __B, const int __C)
+{
+ return (__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) __A,
+ (__v32bf) __B,
+ __C,
+ (__v32bf) __W,
+ (__mmask32) __U);
+}
+
+extern __inline __m512bh
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_minmax_pbh (__mmask32 __U, __m512bh __A,
+ __m512bh __B, const int __C)
+{
+ return (__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) __A,
+ (__v32bf) __B,
+ __C,
+ (__v32bf)(__m512bh)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U);
+}
+
extern __inline __m128d
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_minmax_pd (__m128d __A, __m128d __B, const int __C)
(__mmask8) __U);
}
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_minmax_pd (__m512d __A, __m512d __B, const int __C)
+{
+ return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
+ (__v8df) __B,
+ __C,
+ (__v8df)
+ _mm512_undefined_pd (),
+ (__mmask8) -1,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_minmax_pd (__m512d __W, __mmask8 __U, __m512d __A,
+ __m512d __B, const int __C)
+{
+ return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
+ (__v8df) __B,
+ __C,
+ (__v8df) __W,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_minmax_pd (__mmask8 __U, __m512d __A, __m512d __B,
+ const int __C)
+{
+ return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
+ (__v8df) __B,
+ __C,
+ (__v8df)
+ _mm512_setzero_pd (),
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_minmax_round_pd (__m512d __A, __m512d __B, const int __C,
+ const int __R)
+{
+ return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
+ (__v8df) __B,
+ __C,
+ (__v8df)
+ _mm512_undefined_pd (),
+ (__mmask8) -1, __R);
+}
+
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_minmax_round_pd (__m512d __W, __mmask8 __U, __m512d __A,
+ __m512d __B, const int __C, const int __R)
+{
+ return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
+ (__v8df) __B,
+ __C,
+ (__v8df) __W,
+ (__mmask8) __U, __R);
+}
+
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_minmax_round_pd (__mmask8 __U, __m512d __A, __m512d __B,
+ const int __C, const int __R)
+{
+ return (__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) __A,
+ (__v8df) __B,
+ __C,
+ (__v8df)
+ _mm512_setzero_pd (),
+ (__mmask8) __U, __R);
+}
+
extern __inline __m128h
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_minmax_ph (__m128h __A, __m128h __B, const int __C)
(__mmask16) __U);
}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_minmax_ph (__m512h __A, __m512h __B, const int __C)
+{
+ return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
+ (__v32hf) __B,
+ __C,
+ (__v32hf)
+ _mm512_undefined_ph (),
+ (__mmask32) -1,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_minmax_ph (__m512h __W, __mmask32 __U, __m512h __A,
+ __m512h __B, const int __C)
+{
+ return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
+ (__v32hf) __B,
+ __C,
+ (__v32hf) __W,
+ (__mmask32) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_minmax_ph (__mmask32 __U, __m512h __A, __m512h __B,
+ const int __C)
+{
+ return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
+ (__v32hf) __B,
+ __C,
+ (__v32hf)
+ _mm512_setzero_ph (),
+ (__mmask32) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_minmax_round_ph (__m512h __A, __m512h __B, const int __C, const int __R)
+{
+ return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
+ (__v32hf) __B,
+ __C,
+ (__v32hf)
+ _mm512_undefined_ph (),
+ (__mmask32) -1, __R);
+}
+
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_minmax_round_ph (__m512h __W, __mmask32 __U, __m512h __A,
+ __m512h __B, const int __C, const int __R)
+{
+ return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
+ (__v32hf) __B,
+ __C,
+ (__v32hf) __W,
+ (__mmask32) __U, __R);
+}
+
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_minmax_round_ph (__mmask32 __U, __m512h __A, __m512h __B,
+ const int __C, const int __R)
+{
+ return (__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) __A,
+ (__v32hf) __B,
+ __C,
+ (__v32hf)
+ _mm512_setzero_ph (),
+ (__mmask32) __U, __R);
+}
+
extern __inline __m128
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_minmax_ps (__m128 __A, __m128 __B, const int __C)
(__mmask8) __U);
}
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_minmax_ps (__m512 __A, __m512 __B, const int __C)
+{
+ return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ __C,
+ (__v16sf)
+ _mm512_undefined_ps (),
+ (__mmask16) -1,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_minmax_ps (__m512 __W, __mmask16 __U, __m512 __A,
+ __m512 __B, const int __C)
+{
+ return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ __C,
+ (__v16sf) __W,
+ (__mmask16) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_minmax_ps (__mmask16 __U, __m512 __A, __m512 __B,
+ const int __C)
+{
+ return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ __C,
+ (__v16sf)
+ _mm512_setzero_ps (),
+ (__mmask16) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_minmax_round_ps (__m512 __A, __m512 __B, const int __C, const int __R)
+{
+ return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ __C,
+ (__v16sf)
+ _mm512_undefined_ps (),
+ (__mmask16) -1, __R);
+}
+
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_minmax_round_ps (__m512 __W, __mmask16 __U, __m512 __A,
+ __m512 __B, const int __C, const int __R)
+{
+ return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ __C,
+ (__v16sf) __W,
+ (__mmask16) __U, __R);
+}
+
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_minmax_round_ps (__mmask16 __U, __m512 __A, __m512 __B,
+ const int __C, const int __R)
+{
+ return (__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) __A,
+ (__v16sf) __B,
+ __C,
+ (__v16sf)
+ _mm512_setzero_ps (),
+ (__mmask16) __U, __R);
+}
+
extern __inline __m128d
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_minmax_sd (__m128d __A, __m128d __B, const int __C)
_mm256_setzero_si256 (), \
(__mmask16) (U)))
+#define _mm512_minmax_pbh(A, B, C) \
+ ((__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) (A), \
+ (__v32bf) (B), \
+ (int) (C), \
+ (__v32bf) (__m512bh) \
+ _mm512_setzero_si512 (), \
+ (__mmask32) (-1)))
+
+#define _mm512_mask_minmax_pbh(W, U, A, B, C) \
+ ((__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) (A), \
+ (__v32bf) (B), \
+ (int) (C), \
+ (__v32bf) (__m512bh) (W), \
+ (__mmask32) (U)))
+
+#define _mm512_maskz_minmax_pbh(U, A, B, C) \
+ ((__m512bh) __builtin_ia32_minmaxbf16512_mask ((__v32bf) (A), \
+ (__v32bf) (B), \
+ (int) (C), \
+ (__v32bf) (__m512bh) \
+ _mm512_setzero_si512 (), \
+ (__mmask32) (U)))
+
#define _mm_minmax_pd(A, B, C) \
((__m128d) __builtin_ia32_minmaxpd128_mask ((__v2df) (A), \
(__v2df) (B), \
_mm256_setzero_pd (), \
(__mmask8) (U)))
+#define _mm512_minmax_pd(A, B, C) \
+ ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
+ (__v8df) (B), \
+ (int) (C), \
+ (__v8df) (__m512d) \
+ _mm512_undefined_pd (), \
+ (__mmask8) (-1), \
+ _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_mask_minmax_pd(W, U, A, B, C) \
+ ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
+ (__v8df) (B), \
+ (int) (C), \
+ (__v8df) (__m512d) (W), \
+ (__mmask8) (U), \
+ _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_maskz_minmax_pd(U, A, B, C) \
+ ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
+ (__v8df) (B), \
+ (int) (C), \
+ (__v8df) (__m512d) \
+ _mm512_setzero_pd (), \
+ (__mmask8) (U), \
+ _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_minmax_round_pd(A, B, C, R) \
+ ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
+ (__v8df) (B), \
+ (int) (C), \
+ (__v8df) (__m512d) \
+ _mm512_undefined_pd (), \
+ (__mmask8) (-1), \
+ (int) (R)))
+
+#define _mm512_mask_minmax_round_pd(W, U, A, B, C, R) \
+ ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
+ (__v8df) (B), \
+ (int) (C), \
+ (__v8df) (__m512d) (W), \
+ (__mmask8) (U), \
+ (int) (R)))
+
+#define _mm512_maskz_minmax_round_pd(U, A, B, C, R) \
+ ((__m512d) __builtin_ia32_minmaxpd512_mask_round ((__v8df) (A), \
+ (__v8df) (B), \
+ (int) (C), \
+ (__v8df) (__m512d) \
+ _mm512_setzero_pd (), \
+ (__mmask8) (U), \
+ (int) (R)))
+
#define _mm_minmax_ph(A, B, C) \
((__m128h) __builtin_ia32_minmaxph128_mask ((__v8hf) (A), \
(__v8hf) (B), \
_mm256_setzero_ph (), \
(__mmask16) (U)))
+#define _mm512_minmax_ph(A, B, C) \
+ ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
+ (__v32hf) (B), \
+ (int) (C), \
+ (__v32hf) (__m512h) \
+ _mm512_undefined_ph (), \
+ (__mmask32) (-1), \
+ _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_mask_minmax_ph(W, U, A, B, C) \
+ ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
+ (__v32hf) (B), \
+ (int) (C), \
+ (__v32hf) (__m512h) (W), \
+ (__mmask32) (U), \
+ _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_maskz_minmax_ph(U, A, B, C) \
+ ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
+ (__v32hf) (B), \
+ (int) (C), \
+ (__v32hf) (__m512h) \
+ _mm512_setzero_ph (), \
+ (__mmask32) (U), \
+ _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_minmax_round_ph(A, B, C, R) \
+ ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
+ (__v32hf) (B), \
+ (int) (C), \
+ (__v32hf) (__m512h) \
+ _mm512_undefined_ph (), \
+ (__mmask32) (-1), \
+ (int) (R)))
+
+#define _mm512_mask_minmax_round_ph(W, U, A, B, C, R) \
+ ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
+ (__v32hf) (B), \
+ (int) (C), \
+ (__v32hf) (__m512h) (W), \
+ (__mmask32) (U), \
+ (int) (R)))
+
+#define _mm512_maskz_minmax_round_ph(U, A, B, C, R) \
+ ((__m512h) __builtin_ia32_minmaxph512_mask_round ((__v32hf) (A), \
+ (__v32hf) (B), \
+ (int) (C), \
+ (__v32hf) (__m512h) \
+ _mm512_setzero_ph (), \
+ (__mmask32) (U), \
+ (int) (R)))
+
#define _mm_minmax_ps(A, B, C) \
((__m128) __builtin_ia32_minmaxps128_mask ((__v4sf) (A), \
(__v4sf) (B), \
_mm256_setzero_ps (), \
(__mmask8) (U)))
+#define _mm512_minmax_ps(A, B, C) \
+ ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
+ (__v16sf) (B), \
+ (int) (C), \
+ (__v16sf) (__m512) \
+ _mm512_undefined_ps (), \
+ (__mmask16) (-1), \
+ _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_mask_minmax_ps(W, U, A, B, C) \
+ ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
+ (__v16sf) (B), \
+ (int) (C), \
+ (__v16sf) (__m512) (W), \
+ (__mmask16) (U), \
+ _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_maskz_minmax_ps(U, A, B, C) \
+ ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
+ (__v16sf) (B), \
+ (int) (C), \
+ (__v16sf) (__m512) \
+ _mm512_setzero_ps (), \
+ (__mmask16) (U), \
+ _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_minmax_round_ps(A, B, C, R) \
+ ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
+ (__v16sf) (B), \
+ (int) (C), \
+ (__v16sf) (__m512) \
+ _mm512_undefined_ps (), \
+ (__mmask16) (-1), \
+ (int) (R)))
+
+#define _mm512_mask_minmax_round_ps(W, U, A, B, C, R) \
+ ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
+ (__v16sf) (B), \
+ (int) (C), \
+ (__v16sf) (__m512) (W), \
+ (__mmask16) (U), \
+ (int) (R)))
+
+#define _mm512_maskz_minmax_round_ps(U, A, B, C, R) \
+ ((__m512) __builtin_ia32_minmaxps512_mask_round ((__v16sf) (A), \
+ (__v16sf) (B), \
+ (int) (C), \
+ (__v16sf) (__m512) \
+ _mm512_setzero_ps (), \
+ (__mmask16) (U), \
+ (int) (R)))
+
#define _mm_minmax_round_sd(A, B, C, R) \
((__m128d) __builtin_ia32_minmaxsd_mask_round ((__v2df) (A), \
(__v2df) (B), \
(__mmask8) __U);
}
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm256_ipcvts_bf16_epi8 (__m256bh __A)
-{
- return
- (__m256i) __builtin_ia32_cvtbf162ibs256_mask ((__v16bf) __A,
- (__v16hi)
- _mm256_undefined_si256 (),
- (__mmask16) -1);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm256_mask_ipcvts_bf16_epi8 (__m256i __W, __mmask16 __U, __m256bh __A)
-{
- return (__m256i) __builtin_ia32_cvtbf162ibs256_mask ((__v16bf) __A,
- (__v16hi) __W,
- (__mmask16) __U);
-}
-
-extern __inline __m256i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm256_maskz_ipcvts_bf16_epi8 (__mmask16 __U, __m256bh __A)
-{
- return
- (__m256i) __builtin_ia32_cvtbf162ibs256_mask ((__v16bf) __A,
- (__v16hi)
- _mm256_setzero_si256 (),
- (__mmask16) __U);
-}
-
extern __inline __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_ipcvts_bf16_epu8 (__m128bh __A)
(__mmask8) __U);
}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_ipcvts_bf16_epi8 (__m256bh __A)
+{
+ return
+ (__m256i) __builtin_ia32_cvtbf162ibs256_mask ((__v16bf) __A,
+ (__v16hi)
+ _mm256_undefined_si256 (),
+ (__mmask16) -1);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_ipcvts_bf16_epi8 (__m256i __W, __mmask16 __U, __m256bh __A)
+{
+ return (__m256i) __builtin_ia32_cvtbf162ibs256_mask ((__v16bf) __A,
+ (__v16hi) __W,
+ (__mmask16) __U);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_ipcvts_bf16_epi8 (__mmask16 __U, __m256bh __A)
+{
+ return
+ (__m256i) __builtin_ia32_cvtbf162ibs256_mask ((__v16bf) __A,
+ (__v16hi)
+ _mm256_setzero_si256 (),
+ (__mmask16) __U);
+}
+
extern __inline __m256i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm256_ipcvts_bf16_epu8 (__m256bh __A)
(__mmask16) __U);
}
-extern __inline __m128i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_ipcvts_ph_epi8 (__m128h __A)
-{
- return (__m128i) __builtin_ia32_cvtph2ibs128_mask ((__v8hf) __A,
- (__v8hi)
- _mm_undefined_si128 (),
- (__mmask8) -1);
-}
-
-extern __inline __m128i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_mask_ipcvts_ph_epi8 (__m128i __W, __mmask8 __U, __m128h __A)
-{
- return (__m128i) __builtin_ia32_cvtph2ibs128_mask ((__v8hf) __A,
- (__v8hi) __W,
- (__mmask8) __U);
-}
-
-extern __inline __m128i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_maskz_ipcvts_ph_epi8 (__mmask8 __U, __m128h __A)
-{
- return (__m128i) __builtin_ia32_cvtph2ibs128_mask ((__v8hf) __A,
- (__v8hi)
- _mm_setzero_si128 (),
- (__mmask8) __U);
-}
-
-extern __inline __m128i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_ipcvts_ph_epu8 (__m128h __A)
-{
- return (__m128i) __builtin_ia32_cvtph2iubs128_mask ((__v8hf) __A,
- (__v8hi)
- _mm_undefined_si128 (),
- (__mmask8) -1);
-}
-
-extern __inline __m128i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_mask_ipcvts_ph_epu8 (__m128i __W, __mmask8 __U, __m128h __A)
-{
- return (__m128i) __builtin_ia32_cvtph2iubs128_mask ((__v8hf) __A,
- (__v8hi) __W,
- (__mmask8) __U);
-}
-
-extern __inline __m128i
-__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_maskz_ipcvts_ph_epu8 (__mmask8 __U, __m128h __A)
-{
- return (__m128i) __builtin_ia32_cvtph2iubs128_mask ((__v8hf) __A,
- (__v8hi)
- _mm_setzero_si128 (),
- (__mmask8) __U);
-}
-
-extern __inline __m128i
+extern __inline __m512i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_ipcvts_ps_epi8 (__m128 __A)
+_mm512_ipcvts_bf16_epi8 (__m512bh __A)
{
- return (__m128i) __builtin_ia32_cvtps2ibs128_mask ((__v4sf) __A,
- (__v4si)
- _mm_undefined_si128 (),
- (__mmask8) -1);
+ return
+ (__m512i) __builtin_ia32_cvtbf162ibs512_mask ((__v32bf) __A,
+ (__v32hi)
+ _mm512_undefined_si512 (),
+ (__mmask32) -1);
}
-extern __inline __m128i
+extern __inline __m512i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_mask_ipcvts_ps_epi8 (__m128i __W, __mmask8 __U, __m128 __A)
+_mm512_mask_ipcvts_bf16_epi8 (__m512i __W, __mmask32 __U, __m512bh __A)
{
- return (__m128i) __builtin_ia32_cvtps2ibs128_mask ((__v4sf) __A,
- (__v4si) __W,
- (__mmask8) __U);
+ return (__m512i) __builtin_ia32_cvtbf162ibs512_mask ((__v32bf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U);
}
-extern __inline __m128i
+extern __inline __m512i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_maskz_ipcvts_ps_epi8 (__mmask8 __U, __m128 __A)
+_mm512_maskz_ipcvts_bf16_epi8 (__mmask32 __U, __m512bh __A)
{
- return (__m128i) __builtin_ia32_cvtps2ibs128_mask ((__v4sf) __A,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return
+ (__m512i) __builtin_ia32_cvtbf162ibs512_mask ((__v32bf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U);
}
-extern __inline __m128i
+extern __inline __m512i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_ipcvts_ps_epu8 (__m128 __A)
+_mm512_ipcvts_bf16_epu8 (__m512bh __A)
{
- return (__m128i) __builtin_ia32_cvtps2iubs128_mask ((__v4sf) __A,
- (__v4si)
- _mm_undefined_si128 (),
- (__mmask8) -1);
+ return
+ (__m512i) __builtin_ia32_cvtbf162iubs512_mask ((__v32bf) __A,
+ (__v32hi)
+ _mm512_undefined_si512 (),
+ (__mmask32) -1);
}
-extern __inline __m128i
+extern __inline __m512i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_mask_ipcvts_ps_epu8 (__m128i __W, __mmask8 __U, __m128 __A)
+_mm512_mask_ipcvts_bf16_epu8 (__m512i __W, __mmask32 __U, __m512bh __A)
{
- return (__m128i) __builtin_ia32_cvtps2iubs128_mask ((__v4sf) __A,
- (__v4si) __W,
- (__mmask8) __U);
+ return (__m512i) __builtin_ia32_cvtbf162iubs512_mask ((__v32bf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U);
}
-extern __inline __m128i
+extern __inline __m512i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_maskz_ipcvts_ps_epu8 (__mmask8 __U, __m128 __A)
+_mm512_maskz_ipcvts_bf16_epu8 (__mmask32 __U, __m512bh __A)
{
- return (__m128i) __builtin_ia32_cvtps2iubs128_mask ((__v4sf) __A,
- (__v4si)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return
+ (__m512i) __builtin_ia32_cvtbf162iubs512_mask ((__v32bf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U);
}
extern __inline __m128i
(__mmask16) __U);
}
-extern __inline __m128i
+extern __inline __m512i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_ipcvtts_ph_epi8 (__m128h __A)
+_mm512_ipcvtts_bf16_epi8 (__m512bh __A)
{
- return (__m128i) __builtin_ia32_cvttph2ibs128_mask ((__v8hf) __A,
- (__v8hi)
- _mm_undefined_si128 (),
- (__mmask8) -1);
+ return
+ (__m512i) __builtin_ia32_cvttbf162ibs512_mask ((__v32bf) __A,
+ (__v32hi)
+ _mm512_undefined_si512 (),
+ (__mmask32) -1);
}
-extern __inline __m128i
+extern __inline __m512i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_mask_ipcvtts_ph_epi8 (__m128i __W, __mmask8 __U, __m128h __A)
+_mm512_mask_ipcvtts_bf16_epi8 (__m512i __W, __mmask32 __U, __m512bh __A)
{
- return (__m128i) __builtin_ia32_cvttph2ibs128_mask ((__v8hf) __A,
- (__v8hi) __W,
- (__mmask8) __U);
+ return (__m512i) __builtin_ia32_cvttbf162ibs512_mask ((__v32bf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U);
}
-extern __inline __m128i
+extern __inline __m512i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_maskz_ipcvtts_ph_epi8 (__mmask8 __U, __m128h __A)
+_mm512_maskz_ipcvtts_bf16_epi8 (__mmask32 __U, __m512bh __A)
{
- return (__m128i) __builtin_ia32_cvttph2ibs128_mask ((__v8hf) __A,
- (__v8hi)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return
+ (__m512i) __builtin_ia32_cvttbf162ibs512_mask ((__v32bf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvtts_bf16_epu8 (__m512bh __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttbf162iubs512_mask ((__v32bf) __A,
+ (__v32hi) _mm512_undefined_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvtts_bf16_epu8 (__m512i __W, __mmask32 __U, __m512bh __A)
+{
+ return (__m512i) __builtin_ia32_cvttbf162iubs512_mask ((__v32bf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvtts_bf16_epu8 (__mmask32 __U, __m512bh __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttbf162iubs512_mask ((__v32bf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U);
}
extern __inline __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_ipcvtts_ph_epu8 (__m128h __A)
+_mm_ipcvts_ph_epi8 (__m128h __A)
{
- return (__m128i) __builtin_ia32_cvttph2iubs128_mask ((__v8hf) __A,
- (__v8hi)
- _mm_undefined_si128 (),
- (__mmask8) -1);
+ return (__m128i) __builtin_ia32_cvtph2ibs128_mask ((__v8hf) __A,
+ (__v8hi)
+ _mm_undefined_si128 (),
+ (__mmask8) -1);
}
extern __inline __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_mask_ipcvtts_ph_epu8 (__m128i __W, __mmask8 __U, __m128h __A)
+_mm_mask_ipcvts_ph_epi8 (__m128i __W, __mmask8 __U, __m128h __A)
{
- return (__m128i) __builtin_ia32_cvttph2iubs128_mask ((__v8hf) __A,
- (__v8hi) __W,
- (__mmask8) __U);
+ return (__m128i) __builtin_ia32_cvtph2ibs128_mask ((__v8hf) __A,
+ (__v8hi) __W,
+ (__mmask8) __U);
}
extern __inline __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_maskz_ipcvtts_ph_epu8 (__mmask8 __U, __m128h __A)
+_mm_maskz_ipcvts_ph_epi8 (__mmask8 __U, __m128h __A)
{
- return (__m128i) __builtin_ia32_cvttph2iubs128_mask ((__v8hf) __A,
- (__v8hi)
- _mm_setzero_si128 (),
- (__mmask8) __U);
+ return (__m128i) __builtin_ia32_cvtph2ibs128_mask ((__v8hf) __A,
+ (__v8hi)
+ _mm_setzero_si128 (),
+ (__mmask8) __U);
}
extern __inline __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_ipcvtts_ps_epi8 (__m128 __A)
+_mm_ipcvts_ph_epu8 (__m128h __A)
{
- return (__m128i) __builtin_ia32_cvttps2ibs128_mask ((__v4sf) __A,
- (__v4si)
+ return (__m128i) __builtin_ia32_cvtph2iubs128_mask ((__v8hf) __A,
+ (__v8hi)
_mm_undefined_si128 (),
(__mmask8) -1);
}
extern __inline __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_mask_ipcvtts_ps_epi8 (__m128i __W, __mmask8 __U, __m128 __A)
+_mm_mask_ipcvts_ph_epu8 (__m128i __W, __mmask8 __U, __m128h __A)
{
- return (__m128i) __builtin_ia32_cvttps2ibs128_mask ((__v4sf) __A,
- (__v4si) __W,
+ return (__m128i) __builtin_ia32_cvtph2iubs128_mask ((__v8hf) __A,
+ (__v8hi) __W,
(__mmask8) __U);
}
extern __inline __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_maskz_ipcvtts_ps_epi8 (__mmask8 __U, __m128 __A)
+_mm_maskz_ipcvts_ph_epu8 (__mmask8 __U, __m128h __A)
{
- return (__m128i) __builtin_ia32_cvttps2ibs128_mask ((__v4sf) __A,
- (__v4si)
+ return (__m128i) __builtin_ia32_cvtph2iubs128_mask ((__v8hf) __A,
+ (__v8hi)
_mm_setzero_si128 (),
(__mmask8) __U);
}
extern __inline __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_ipcvtts_ps_epu8 (__m128 __A)
+_mm_ipcvts_ps_epi8 (__m128 __A)
{
- return (__m128i) __builtin_ia32_cvttps2iubs128_mask ((__v4sf) __A,
- (__v4si)
- _mm_undefined_si128 (),
- (__mmask8) -1);
+ return (__m128i) __builtin_ia32_cvtps2ibs128_mask ((__v4sf) __A,
+ (__v4si)
+ _mm_undefined_si128 (),
+ (__mmask8) -1);
}
extern __inline __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_mask_ipcvtts_ps_epu8 (__m128i __W, __mmask8 __U, __m128 __A)
+_mm_mask_ipcvts_ps_epi8 (__m128i __W, __mmask8 __U, __m128 __A)
{
- return (__m128i) __builtin_ia32_cvttps2iubs128_mask ((__v4sf) __A,
- (__v4si) __W,
- (__mmask8) __U);
+ return (__m128i) __builtin_ia32_cvtps2ibs128_mask ((__v4sf) __A,
+ (__v4si) __W,
+ (__mmask8) __U);
}
extern __inline __m128i
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
-_mm_maskz_ipcvtts_ps_epu8 (__mmask8 __U, __m128 __A)
+_mm_maskz_ipcvts_ps_epi8 (__mmask8 __U, __m128 __A)
{
- return (__m128i) __builtin_ia32_cvttps2iubs128_mask ((__v4sf) __A,
+ return (__m128i) __builtin_ia32_cvtps2ibs128_mask ((__v4sf) __A,
+ (__v4si)
+ _mm_setzero_si128 (),
+ (__mmask8) __U);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ipcvts_ps_epu8 (__m128 __A)
+{
+ return (__m128i) __builtin_ia32_cvtps2iubs128_mask ((__v4sf) __A,
+ (__v4si)
+ _mm_undefined_si128 (),
+ (__mmask8) -1);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_ipcvts_ps_epu8 (__m128i __W, __mmask8 __U, __m128 __A)
+{
+ return (__m128i) __builtin_ia32_cvtps2iubs128_mask ((__v4sf) __A,
+ (__v4si) __W,
+ (__mmask8) __U);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_ipcvts_ps_epu8 (__mmask8 __U, __m128 __A)
+{
+ return (__m128i) __builtin_ia32_cvtps2iubs128_mask ((__v4sf) __A,
+ (__v4si)
+ _mm_setzero_si128 (),
+ (__mmask8) __U);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ipcvtts_ph_epi8 (__m128h __A)
+{
+ return (__m128i) __builtin_ia32_cvttph2ibs128_mask ((__v8hf) __A,
+ (__v8hi)
+ _mm_undefined_si128 (),
+ (__mmask8) -1);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_ipcvtts_ph_epi8 (__m128i __W, __mmask8 __U, __m128h __A)
+{
+ return (__m128i) __builtin_ia32_cvttph2ibs128_mask ((__v8hf) __A,
+ (__v8hi) __W,
+ (__mmask8) __U);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_ipcvtts_ph_epi8 (__mmask8 __U, __m128h __A)
+{
+ return (__m128i) __builtin_ia32_cvttph2ibs128_mask ((__v8hf) __A,
+ (__v8hi)
+ _mm_setzero_si128 (),
+ (__mmask8) __U);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ipcvtts_ph_epu8 (__m128h __A)
+{
+ return (__m128i) __builtin_ia32_cvttph2iubs128_mask ((__v8hf) __A,
+ (__v8hi)
+ _mm_undefined_si128 (),
+ (__mmask8) -1);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_ipcvtts_ph_epu8 (__m128i __W, __mmask8 __U, __m128h __A)
+{
+ return (__m128i) __builtin_ia32_cvttph2iubs128_mask ((__v8hf) __A,
+ (__v8hi) __W,
+ (__mmask8) __U);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_ipcvtts_ph_epu8 (__mmask8 __U, __m128h __A)
+{
+ return (__m128i) __builtin_ia32_cvttph2iubs128_mask ((__v8hf) __A,
+ (__v8hi)
+ _mm_setzero_si128 (),
+ (__mmask8) __U);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ipcvtts_ps_epi8 (__m128 __A)
+{
+ return (__m128i) __builtin_ia32_cvttps2ibs128_mask ((__v4sf) __A,
+ (__v4si)
+ _mm_undefined_si128 (),
+ (__mmask8) -1);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_ipcvtts_ps_epi8 (__m128i __W, __mmask8 __U, __m128 __A)
+{
+ return (__m128i) __builtin_ia32_cvttps2ibs128_mask ((__v4sf) __A,
+ (__v4si) __W,
+ (__mmask8) __U);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_ipcvtts_ps_epi8 (__mmask8 __U, __m128 __A)
+{
+ return (__m128i) __builtin_ia32_cvttps2ibs128_mask ((__v4sf) __A,
+ (__v4si)
+ _mm_setzero_si128 (),
+ (__mmask8) __U);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ipcvtts_ps_epu8 (__m128 __A)
+{
+ return (__m128i) __builtin_ia32_cvttps2iubs128_mask ((__v4sf) __A,
+ (__v4si)
+ _mm_undefined_si128 (),
+ (__mmask8) -1);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_ipcvtts_ps_epu8 (__m128i __W, __mmask8 __U, __m128 __A)
+{
+ return (__m128i) __builtin_ia32_cvttps2iubs128_mask ((__v4sf) __A,
+ (__v4si) __W,
+ (__mmask8) __U);
+}
+
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_ipcvtts_ps_epu8 (__mmask8 __U, __m128 __A)
+{
+ return (__m128i) __builtin_ia32_cvttps2iubs128_mask ((__v4sf) __A,
(__v4si)
_mm_setzero_si128 (),
(__mmask8) __U);
(__mmask8) __U);
}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvts_ph_epi8 (__m512h __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvtph2ibs512_mask ((__v32hf) __A,
+ (__v32hi)
+ _mm512_undefined_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvts_ph_epi8 (__m512i __W, __mmask32 __U, __m512h __A)
+{
+ return (__m512i) __builtin_ia32_cvtph2ibs512_mask ((__v32hf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvts_ph_epi8 (__mmask32 __U, __m512h __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvtph2ibs512_mask ((__v32hf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvts_ph_epu8 (__m512h __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvtph2iubs512_mask ((__v32hf) __A,
+ (__v32hi)
+ _mm512_undefined_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvts_ph_epu8 (__m512i __W, __mmask32 __U, __m512h __A)
+{
+ return (__m512i) __builtin_ia32_cvtph2iubs512_mask ((__v32hf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvts_ph_epu8 (__mmask32 __U, __m512h __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvtph2iubs512_mask ((__v32hf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvts_ps_epi8 (__m512 __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvtps2ibs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvts_ps_epi8 (__m512i __W, __mmask16 __U, __m512 __A)
+{
+ return (__m512i) __builtin_ia32_cvtps2ibs512_mask ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvts_ps_epi8 (__mmask16 __U, __m512 __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvtps2ibs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvts_ps_epu8 (__m512 __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvtps2iubs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvts_ps_epu8 (__m512i __W, __mmask16 __U, __m512 __A)
+{
+ return (__m512i) __builtin_ia32_cvtps2iubs512_mask ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvts_ps_epu8 (__mmask16 __U, __m512 __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvtps2iubs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvtts_ph_epi8 (__m512h __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttph2ibs512_mask ((__v32hf) __A,
+ (__v32hi)
+ _mm512_undefined_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvtts_ph_epi8 (__m512i __W, __mmask32 __U, __m512h __A)
+{
+ return (__m512i) __builtin_ia32_cvttph2ibs512_mask ((__v32hf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvtts_ph_epi8 (__mmask32 __U, __m512h __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvttph2ibs512_mask ((__v32hf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvtts_ph_epu8 (__m512h __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttph2iubs512_mask ((__v32hf) __A,
+ (__v32hi)
+ _mm512_undefined_si512 (),
+ (__mmask32) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvtts_ph_epu8 (__m512i __W, __mmask32 __U, __m512h __A)
+{
+ return (__m512i) __builtin_ia32_cvttph2iubs512_mask ((__v32hf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvtts_ph_epu8 (__mmask32 __U, __m512h __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttph2iubs512_mask ((__v32hf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvtts_ps_epi8 (__m512 __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2ibs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvtts_ps_epi8 (__m512i __W, __mmask16 __U, __m512 __A)
+{
+ return (__m512i) __builtin_ia32_cvttps2ibs512_mask ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvtts_ps_epi8 (__mmask16 __U, __m512 __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2ibs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvtts_ps_epu8 (__m512 __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2iubs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvtts_ps_epu8 (__m512i __W, __mmask16 __U, __m512 __A)
+{
+ return (__m512i) __builtin_ia32_cvttps2iubs512_mask ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvtts_ps_epu8 (__mmask16 __U, __m512 __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2iubs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_pd_epi32 (__m512d __A)
+{
+ return (__m256i)
+ __builtin_ia32_cvttpd2dqs512_mask ((__v8df) __A,
+ (__v8si)
+ _mm256_undefined_si256 (),
+ (__mmask8) -1);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_pd_epi32 (__m256i __W, __mmask8 __U, __m512d __A)
+{
+ return (__m256i) __builtin_ia32_cvttpd2dqs512_mask ((__v8df) __A,
+ (__v8si) __W,
+ (__mmask8) __U);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_pd_epi32 (__mmask8 __U, __m512d __A)
+{
+ return
+ (__m256i) __builtin_ia32_cvttpd2dqs512_mask ((__v8df) __A,
+ (__v8si)
+ _mm256_setzero_si256 (),
+ (__mmask8) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_pd_epi64 (__m512d __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttpd2qqs512_mask ((__v8df) __A,
+ (__v8di)
+ _mm512_undefined_si512 (),
+ (__mmask8) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_pd_epi64 (__m512i __W, __mmask8 __U, __m512d __A)
+{
+ return (__m512i) __builtin_ia32_cvttpd2qqs512_mask ((__v8df) __A,
+ (__v8di) __W,
+ (__mmask8) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_pd_epi64 (__mmask8 __U, __m512d __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvttpd2qqs512_mask ((__v8df) __A,
+ (__v8di)
+ _mm512_setzero_si512 (),
+ (__mmask8) __U);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_pd_epu32 (__m512d __A)
+{
+ return (__m256i)
+ __builtin_ia32_cvttpd2udqs512_mask ((__v8df) __A,
+ (__v8si)
+ _mm256_undefined_si256 (),
+ (__mmask8) -1);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_pd_epu32 (__m256i __W, __mmask8 __U, __m512d __A)
+{
+ return (__m256i) __builtin_ia32_cvttpd2udqs512_mask ((__v8df) __A,
+ (__v8si) __W,
+ (__mmask8) __U);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_pd_epu32 (__mmask8 __U, __m512d __A)
+{
+ return
+ (__m256i) __builtin_ia32_cvttpd2udqs512_mask ((__v8df) __A,
+ (__v8si)
+ _mm256_setzero_si256 (),
+ (__mmask8) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_pd_epu64 (__m512d __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttpd2uqqs512_mask ((__v8df) __A,
+ (__v8di)
+ _mm512_undefined_si512 (),
+ (__mmask8) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_pd_epu64 (__m512i __W, __mmask8 __U, __m512d __A)
+{
+ return (__m512i) __builtin_ia32_cvttpd2uqqs512_mask ((__v8df) __A,
+ (__v8di) __W,
+ (__mmask8) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_pd_epu64 (__mmask8 __U, __m512d __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttpd2uqqs512_mask ((__v8df) __A,
+ (__v8di)
+ _mm512_setzero_si512 (),
+ (__mmask8) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_ps_epi32 (__m512 __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2dqs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_ps_epi32 (__m512i __W, __mmask16 __U, __m512 __A)
+{
+ return (__m512i) __builtin_ia32_cvttps2dqs512_mask ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_ps_epi32 (__mmask16 __U, __m512 __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvttps2dqs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_ps_epi64 (__m256 __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2qqs512_mask ((__v8sf) __A,
+ (__v8di)
+ _mm512_undefined_si512 (),
+ (__mmask8) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_ps_epi64 (__m512i __W, __mmask8 __U, __m256 __A)
+{
+ return (__m512i) __builtin_ia32_cvttps2qqs512_mask ((__v8sf) __A,
+ (__v8di) __W,
+ (__mmask8) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_ps_epi64 (__mmask8 __U, __m256 __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvttps2qqs512_mask ((__v8sf) __A,
+ (__v8di)
+ _mm512_setzero_si512 (),
+ (__mmask8) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_ps_epu32 (__m512 __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2udqs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_ps_epu32 (__m512i __W, __mmask16 __U, __m512 __A)
+{
+ return (__m512i) __builtin_ia32_cvttps2udqs512_mask ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_ps_epu32 (__mmask16 __U, __m512 __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2udqs512_mask ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_ps_epu64 (__m256 __A)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2uqqs512_mask ((__v8sf) __A,
+ (__v8di)
+ _mm512_undefined_si512 (),
+ (__mmask8) -1);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_ps_epu64 (__m512i __W, __mmask8 __U, __m256 __A)
+{
+ return (__m512i) __builtin_ia32_cvttps2uqqs512_mask ((__v8sf) __A,
+ (__v8di) __W,
+ (__mmask8) __U);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_ps_epu64 (__mmask8 __U, __m256 __A)
+{
+ return
+ (__m512i) __builtin_ia32_cvttps2uqqs512_mask ((__v8sf) __A,
+ (__v8di)
+ _mm512_setzero_si512 (),
+ (__mmask8) __U);
+}
+
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvts_roundph_epi8 (__m512h __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) __A,
+ (__v32hi)
+ _mm512_undefined_si512 (),
+ (__mmask32) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvts_roundph_epi8 (__m512i __W, __mmask32 __U, __m512h __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvts_roundph_epi8 (__mmask32 __U, __m512h __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvts_roundph_epu8 (__m512h __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) __A,
+ (__v32hi)
+ _mm512_undefined_si512 (),
+ (__mmask32) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvts_roundph_epu8 (__m512i __W, __mmask32 __U, __m512h __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvts_roundph_epu8 (__mmask32 __U, __m512h __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvts_roundps_epi8 (__m512 __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvts_roundps_epi8 (__m512i __W, __mmask16 __U, __m512 __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvts_roundps_epi8 (__mmask16 __U, __m512 __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvts_roundps_epu8 (__m512 __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvts_roundps_epu8 (__m512i __W, __mmask16 __U, __m512 __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvts_roundps_epu8 (__mmask16 __U, __m512 __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvtts_roundph_epi8 (__m512h __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) __A,
+ (__v32hi)
+ _mm512_undefined_si512 (),
+ (__mmask32) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvtts_roundph_epi8 (__m512i __W, __mmask32 __U, __m512h __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvtts_roundph_epi8 (__mmask32 __U, __m512h __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvtts_roundph_epu8 (__m512h __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) __A,
+ (__v32hi)
+ _mm512_undefined_si512 (),
+ (__mmask32) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvtts_roundph_epu8 (__m512i __W, __mmask32 __U, __m512h __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) __A,
+ (__v32hi) __W,
+ (__mmask32) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvtts_roundph_epu8 (__mmask32 __U, __m512h __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) __A,
+ (__v32hi)
+ _mm512_setzero_si512 (),
+ (__mmask32) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvtts_roundps_epi8 (__m512 __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvtts_roundps_epi8 (__m512i __W, __mmask16 __U, __m512 __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvtts_roundps_epi8 (__mmask16 __U, __m512 __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_ipcvtts_roundps_epu8 (__m512 __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_ipcvtts_roundps_epu8 (__m512i __W, __mmask16 __U, __m512 __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_ipcvtts_roundps_epu8 (__mmask16 __U, __m512 __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_roundpd_epi32 (__m512d __A, const int __R)
+{
+ return (__m256i)
+ __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) __A,
+ (__v8si)
+ _mm256_undefined_si256 (),
+ (__mmask8) -1,
+ __R);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_roundpd_epi32 (__m256i __W, __mmask8 __U, __m512d __A,
+ const int __R)
+{
+ return (__m256i) __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) __A,
+ (__v8si) __W,
+ (__mmask8) __U,
+ __R);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_roundpd_epi32 (__mmask8 __U, __m512d __A, const int __R)
+{
+ return
+ (__m256i) __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) __A,
+ (__v8si)
+ _mm256_setzero_si256 (),
+ (__mmask8) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_roundpd_epi64 (__m512d __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) __A,
+ (__v8di)
+ _mm512_undefined_si512 (),
+ (__mmask8) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_roundpd_epi64 (__m512i __W, __mmask8 __U, __m512d __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) __A,
+ (__v8di) __W,
+ (__mmask8) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_roundpd_epi64 (__mmask8 __U, __m512d __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) __A,
+ (__v8di)
+ _mm512_setzero_si512 (),
+ (__mmask8) __U,
+ __R);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_roundpd_epu32 (__m512d __A, const int __R)
+{
+ return (__m256i)
+ __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) __A,
+ (__v8si)
+ _mm256_undefined_si256 (),
+ (__mmask8) -1,
+ __R);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_roundpd_epu32 (__m256i __W, __mmask8 __U, __m512d __A,
+ const int __R)
+{
+ return (__m256i) __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) __A,
+ (__v8si) __W,
+ (__mmask8) __U,
+ __R);
+}
+
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_roundpd_epu32 (__mmask8 __U, __m512d __A, const int __R)
+{
+ return
+ (__m256i) __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) __A,
+ (__v8si)
+ _mm256_setzero_si256 (),
+ (__mmask8) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_roundpd_epu64 (__m512d __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) __A,
+ (__v8di)
+ _mm512_undefined_si512 (),
+ (__mmask8) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_roundpd_epu64 (__m512i __W, __mmask8 __U, __m512d __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) __A,
+ (__v8di) __W,
+ (__mmask8) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_roundpd_epu64 (__mmask8 __U, __m512d __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) __A,
+ (__v8di)
+ _mm512_setzero_si512 (),
+ (__mmask8) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_roundps_epi32 (__m512 __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_roundps_epi32 (__m512i __W, __mmask16 __U, __m512 __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_roundps_epi32 (__mmask16 __U, __m512 __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_roundps_epi64 (__m256 __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) __A,
+ (__v8di)
+ _mm512_undefined_si512 (),
+ (__mmask8) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_roundps_epi64 (__m512i __W, __mmask8 __U, __m256 __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) __A,
+ (__v8di) __W,
+ (__mmask8) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_roundps_epi64 (__mmask8 __U, __m256 __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) __A,
+ (__v8di)
+ _mm512_setzero_si512 (),
+ (__mmask8) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_roundps_epu32 (__m512 __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_undefined_si512 (),
+ (__mmask16) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_roundps_epu32 (__m512i __W, __mmask16 __U, __m512 __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) __A,
+ (__v16si) __W,
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_roundps_epu32 (__mmask16 __U, __m512 __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) __A,
+ (__v16si)
+ _mm512_setzero_si512 (),
+ (__mmask16) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtts_roundps_epu64 (__m256 __A, const int __R)
+{
+ return (__m512i)
+ __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) __A,
+ (__v8di)
+ _mm512_undefined_si512 (),
+ (__mmask8) -1,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtts_roundps_epu64 (__m512i __W, __mmask8 __U, __m256 __A,
+ const int __R)
+{
+ return (__m512i) __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) __A,
+ (__v8di) __W,
+ (__mmask8) __U,
+ __R);
+}
+
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtts_roundps_epu64 (__mmask8 __U, __m256 __A, const int __R)
+{
+ return
+ (__m512i) __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) __A,
+ (__v8di)
+ _mm512_setzero_si512 (),
+ (__mmask8) __U,
+ __R);
+}
+#else
+#define _mm512_ipcvts_roundph_epi8(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) (A), \
+ (__v32hi) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask32) (-1), \
+ (R)))
+
+#define _mm512_mask_ipcvts_roundph_epi8(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) (A), \
+ (__v32hi) (W), \
+ (__mmask32) (U), \
+ (R)))
+
+#define _mm512_maskz_ipcvts_roundph_epi8(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvtph2ibs512_mask_round ((__v32hf) (A), \
+ (__v32hi) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask32) (U), \
+ (R)))
+
+#define _mm512_ipcvts_roundph_epu8(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) (A), \
+ (__v32hi) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask32) (-1), \
+ (R)))
+
+#define _mm512_mask_ipcvts_roundph_epu8(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) (A), \
+ (__v32hi) (W), \
+ (__mmask32) (U), \
+ (R)))
+
+#define _mm512_maskz_ipcvts_roundph_epu8(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvtph2iubs512_mask_round ((__v32hf) (A), \
+ (__v32hi) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask32) (U), \
+ (R)))
+
+#define _mm512_ipcvts_roundps_epi8(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask16) (-1), \
+ (R)))
+
+#define _mm512_mask_ipcvts_roundps_epi8(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) (A), \
+ (__v16si) (W), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_maskz_ipcvts_roundps_epi8(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvtps2ibs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_ipcvts_roundps_epu8(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask16) (-1), \
+ (R)))
+
+#define _mm512_mask_ipcvts_roundps_epu8(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) (A), \
+ (__v16si) (W), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_maskz_ipcvts_roundps_epu8(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvtps2iubs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_ipcvtts_roundph_epi8(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) (A), \
+ (__v32hi) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask32) (-1), \
+ (R)))
+
+#define _mm512_mask_ipcvtts_roundph_epi8(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) (A), \
+ (__v32hi) (W), \
+ (__mmask32) (U), \
+ (R)))
+
+#define _mm512_maskz_ipcvtts_roundph_epi8(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttph2ibs512_mask_round ((__v32hf) (A), \
+ (__v32hi) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask32) (U), \
+ (R)))
+
+#define _mm512_ipcvtts_roundph_epu8(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) (A), \
+ (__v32hi) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask32) (-1), \
+ (R)))
+
+#define _mm512_mask_ipcvtts_roundph_epu8(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) (A), \
+ (__v32hi) (W), \
+ (__mmask32) (U), \
+ (R)))
+
+#define _mm512_maskz_ipcvtts_roundph_epu8(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttph2iubs512_mask_round ((__v32hf) (A), \
+ (__v32hi) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask32) (U), \
+ (R)))
+
+#define _mm512_ipcvtts_roundps_epi8(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask16) (-1), \
+ (R)))
+
+#define _mm512_mask_ipcvtts_roundps_epi8(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) (A), \
+ (__v16si) (W), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_maskz_ipcvtts_roundps_epi8(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2ibs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_ipcvtts_roundps_epu8(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask16) (-1), \
+ (R)))
+
+#define _mm512_mask_ipcvtts_roundps_epu8(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) (A), \
+ (__v16si) (W), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_maskz_ipcvtts_roundps_epu8(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2iubs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_cvtts_roundpd_epi32(A, R) \
+ ((__m256i) \
+ __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) (A), \
+ (__v8si) \
+ (_mm256_undefined_si256 ()), \
+ (__mmask8) (-1), \
+ (R)))
+
+#define _mm512_mask_cvtts_roundpd_epi32(W, U, A, R) \
+ ((__m256i) __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) (A), \
+ (__v8si) (W), \
+ (__mmask8) (U), \
+ (R)))
+
+#define _mm512_maskz_cvtts_roundpd_epi32(U, A, R) \
+ ((__m256i) \
+ __builtin_ia32_cvttpd2dqs512_mask_round ((__v8df) (A), \
+ (__v8si) \
+ (_mm256_setzero_si256 ()), \
+ (__mmask8) (U), \
+ (R)))
+
+#define _mm512_cvtts_roundpd_epi64(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) (A), \
+ (__v8di) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask8) (-1), \
+ (R)))
+
+#define _mm512_mask_cvtts_roundpd_epi64(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) (A), \
+ (__v8di) (W), \
+ (__mmask8) (U), \
+ (R)))
+
+#define _mm512_maskz_cvtts_roundpd_epi64(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttpd2qqs512_mask_round ((__v8df) (A), \
+ (__v8di) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask8) (U), \
+ (R)))
+
+#define _mm512_cvtts_roundpd_epu32(A, R) \
+ ((__m256i) \
+ __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) (A), \
+ (__v8si) \
+ (_mm256_undefined_si256 ()), \
+ (__mmask8) (-1), \
+ (R)))
+
+#define _mm512_mask_cvtts_roundpd_epu32(W, U, A, R) \
+ ((__m256i) __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) (A), \
+ (__v8si) (W), \
+ (__mmask8) (U), \
+ (R)))
+
+#define _mm512_maskz_cvtts_roundpd_epu32(U, A, R) \
+ ((__m256i) \
+ __builtin_ia32_cvttpd2udqs512_mask_round ((__v8df) (A), \
+ (__v8si) \
+ (_mm256_setzero_si256 ()), \
+ (__mmask8) (U), \
+ (R)))
+
+#define _mm512_cvtts_roundpd_epu64(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) (A), \
+ (__v8di) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask8) (-1), \
+ (R)))
+
+#define _mm512_mask_cvtts_roundpd_epu64(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) (A), \
+ (__v8di) (W), \
+ (__mmask8) (U), \
+ (R)))
+
+#define _mm512_maskz_cvtts_roundpd_epu64(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttpd2uqqs512_mask_round ((__v8df) (A), \
+ (__v8di) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask8) (U), \
+ (R)))
+
+#define _mm512_cvtts_roundps_epi32(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask16) (-1), \
+ (R)))
+
+#define _mm512_mask_cvtts_roundps_epi32(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) (A), \
+ (__v16si) (W), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_maskz_cvtts_roundps_epi32(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2dqs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_cvtts_roundps_epi64(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) (A), \
+ (__v8di) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask8) (-1), \
+ (R)))
+
+#define _mm512_mask_cvtts_roundps_epi64(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) (A), \
+ (__v8di) (W), \
+ (__mmask8) (U), \
+ (R)))
+
+#define _mm512_maskz_cvtts_roundps_epi64(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2qqs512_mask_round ((__v8sf) (A), \
+ (__v8di) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask8) (U), \
+ (R)))
+
+#define _mm512_cvtts_roundps_epu32(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask16) (-1), \
+ (R)))
+
+#define _mm512_mask_cvtts_roundps_epu32(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) (A), \
+ (__v16si) (W), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_maskz_cvtts_roundps_epu32(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2udqs512_mask_round ((__v16sf) (A), \
+ (__v16si) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask16) (U), \
+ (R)))
+
+#define _mm512_cvtts_roundps_epu64(A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) (A), \
+ (__v8di) \
+ (_mm512_undefined_si512 ()), \
+ (__mmask8) (-1), \
+ (R)))
+
+#define _mm512_mask_cvtts_roundps_epu64(W, U, A, R) \
+ ((__m512i) __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) (A), \
+ (__v8di) (W), \
+ (__mmask8) (U), \
+ (R)))
+
+#define _mm512_maskz_cvtts_roundps_epu64(U, A, R) \
+ ((__m512i) \
+ __builtin_ia32_cvttps2uqqs512_mask_round ((__v8sf) (A), \
+ (__v8di) \
+ (_mm512_setzero_si512 ()), \
+ (__mmask8) (U), \
+ (R)))
+#endif
+
extern __inline int
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_cvtts_sd_epi32 (__m128d __A)
#include <avx10_2mediaintrin.h>
-#include <avx10_2-512mediaintrin.h>
-
#include <avx10_2convertintrin.h>
-#include <avx10_2-512convertintrin.h>
-
#include <avx10_2bf16intrin.h>
-#include <avx10_2-512bf16intrin.h>
-
#include <avx10_2satcvtintrin.h>
-#include <avx10_2-512satcvtintrin.h>
-
#include <avx10_2minmaxintrin.h>
-#include <avx10_2-512minmaxintrin.h>
-
#include <avx10_2copyintrin.h>
#include <movrsintrin.h>
/* sm3intrin.h */
#define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
-/* avx10_2-512mediaintrin.h */
+/* avx10_2mediaintrin.h */
#define __builtin_ia32_mpsadbw512(A, B, C) __builtin_ia32_mpsadbw512 (A, B, 1)
#define __builtin_ia32_mpsadbw512_mask(A, B, C, D, E) __builtin_ia32_mpsadbw512_mask (A, B, 1, D, E)
-
-/* avx10_2mediaintrin.h */
-#define __builtin_ia32_mpsadbw128_mask(A, B, C, D, E) __builtin_ia32_mpsadbw128_mask (A, B, 1, D, E)
#define __builtin_ia32_mpsadbw256_mask(A, B, C, D, E) __builtin_ia32_mpsadbw256_mask (A, B, 1, D, E)
+#define __builtin_ia32_mpsadbw128_mask(A, B, C, D, E) __builtin_ia32_mpsadbw128_mask (A, B, 1, D, E)
-/* avx10_2-512convertintrin.h */
+/* avx10_2convertintrin.h */
#define __builtin_ia32_vcvt2ps2phx512_mask_round(A, B, C, D, E) __builtin_ia32_vcvt2ps2phx512_mask_round(A, B, C, D, 8)
-/* avx10_2-512bf16intrin.h */
-#define __builtin_ia32_rndscalebf16512_mask(A, B, C, D) __builtin_ia32_rndscalebf16512_mask(A, 123, C, D)
-#define __builtin_ia32_reducebf16512_mask(A, B, C, D) __builtin_ia32_reducebf16512_mask(A, 123, C, D)
-#define __builtin_ia32_getmantbf16512_mask(A, B, C, D) __builtin_ia32_getmantbf16512_mask(A, 1, C, D)
-#define __builtin_ia32_fpclassbf16512_mask(A, B, C) __builtin_ia32_fpclassbf16512_mask(A, 1, C)
-#define __builtin_ia32_cmpbf16512_mask(A, B, C, D) __builtin_ia32_cmpbf16512_mask(A, B, 1, D)
-
/* avx10_2bf16intrin.h */
+#define __builtin_ia32_rndscalebf16512_mask(A, B, C, D) __builtin_ia32_rndscalebf16512_mask(A, 123, C, D)
#define __builtin_ia32_rndscalebf16256_mask(A, B, C, D) __builtin_ia32_rndscalebf16256_mask(A, 123, C, D)
#define __builtin_ia32_rndscalebf16128_mask(A, B, C, D) __builtin_ia32_rndscalebf16128_mask(A, 123, C, D)
+#define __builtin_ia32_reducebf16512_mask(A, B, C, D) __builtin_ia32_reducebf16512_mask(A, 123, C, D)
#define __builtin_ia32_reducebf16256_mask(A, B, C, D) __builtin_ia32_reducebf16256_mask(A, 123, C, D)
#define __builtin_ia32_reducebf16128_mask(A, B, C, D) __builtin_ia32_reducebf16128_mask(A, 123, C, D)
+#define __builtin_ia32_getmantbf16512_mask(A, B, C, D) __builtin_ia32_getmantbf16512_mask(A, 1, C, D)
#define __builtin_ia32_getmantbf16256_mask(A, B, C, D) __builtin_ia32_getmantbf16256_mask(A, 1, C, D)
#define __builtin_ia32_getmantbf16128_mask(A, B, C, D) __builtin_ia32_getmantbf16128_mask(A, 1, C, D)
+#define __builtin_ia32_fpclassbf16512_mask(A, B, C) __builtin_ia32_fpclassbf16512_mask(A, 1, C)
#define __builtin_ia32_fpclassbf16256_mask(A, B, C) __builtin_ia32_fpclassbf16256_mask(A, 1, C)
#define __builtin_ia32_fpclassbf16128_mask(A, B, C) __builtin_ia32_fpclassbf16128_mask(A, 1, C)
+#define __builtin_ia32_cmpbf16512_mask(A, B, C, D) __builtin_ia32_cmpbf16512_mask(A, B, 1, D)
#define __builtin_ia32_cmpbf16256_mask(A, B, C, D) __builtin_ia32_cmpbf16256_mask(A, B, 1, D)
#define __builtin_ia32_cmpbf16128_mask(A, B, C, D) __builtin_ia32_cmpbf16128_mask(A, B, 1, D)
-/* avx10_2-512satcvtintrin.h */
+/* avx10_2satcvtintrin.h */
#define __builtin_ia32_cvtph2ibs512_mask_round(A, B, C, D) __builtin_ia32_cvtph2ibs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvtph2iubs512_mask_round(A, B, C, D) __builtin_ia32_cvtph2iubs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvtps2ibs512_mask_round(A, B, C, D) __builtin_ia32_cvtps2ibs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvttps2qqs512_mask_round(A, B, C, D) __builtin_ia32_cvttps2qqs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvttps2udqs512_mask_round(A, B, C, D) __builtin_ia32_cvttps2udqs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvttps2uqqs512_mask_round(A, B, C, D) __builtin_ia32_cvttps2uqqs512_mask_round(A, B, C, 8)
-
-/* avx10_2satcvtintrin.h */
#define __builtin_ia32_cvttsd2sis32_round(A, B) __builtin_ia32_cvttsd2sis32_round(A, 8)
#define __builtin_ia32_cvttsd2usis32_round(A, B) __builtin_ia32_cvttsd2usis32_round(A, 8)
#define __builtin_ia32_cvttss2sis32_round(A, B) __builtin_ia32_cvttss2sis32_round(A, 8)
#define __builtin_ia32_cvttss2usis64_round(A, B) __builtin_ia32_cvttss2usis64_round(A, 8)
#endif
-/* avx10_2-512minmaxintrin.h */
+/* avx10_2minmaxintrin.h */
+#define __builtin_ia32_minmaxbf16512_mask(A, B, C, W, U) __builtin_ia32_minmaxbf16512_mask (A, B, 4, W, U)
+#define __builtin_ia32_minmaxbf16256_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16256_mask (A, B, 4, D, E)
+#define __builtin_ia32_minmaxbf16128_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16128_mask (A, B, 4, D, E)
#define __builtin_ia32_minmaxpd512_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxpd512_mask_round (A, B, 4, D, E, 4)
+#define __builtin_ia32_minmaxpd256_mask(A, B, C, D, E) __builtin_ia32_minmaxpd256_mask (A, B, 4, D, E)
+#define __builtin_ia32_minmaxpd128_mask(A, B, C, D, E) __builtin_ia32_minmaxpd128_mask (A, B, 4, D, E)
#define __builtin_ia32_minmaxph512_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxph512_mask_round (A, B, 4, D, E, 4)
+#define __builtin_ia32_minmaxph256_mask(A, B, C, D, E) __builtin_ia32_minmaxph256_mask (A, B, 4, D, E)
+#define __builtin_ia32_minmaxph128_mask(A, B, C, D, E) __builtin_ia32_minmaxph128_mask (A, B, 4, D, E)
#define __builtin_ia32_minmaxps512_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxps512_mask_round (A, B, 4, D, E, 4)
-#define __builtin_ia32_minmaxbf16512_mask(A, B, C, W, U) __builtin_ia32_minmaxbf16512_mask (A, B, 4, W, U)
-
-/* avx10_2minmaxintrin.h */
+#define __builtin_ia32_minmaxps256_mask(A, B, C, D, E) __builtin_ia32_minmaxps256_mask (A, B, 4, D, E)
+#define __builtin_ia32_minmaxps128_mask(A, B, C, D, E) __builtin_ia32_minmaxps128_mask (A, B, 4, D, E)
#define __builtin_ia32_minmaxsd_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxsd_mask_round (A, B, 4, D, E, 4)
#define __builtin_ia32_minmaxsh_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxsh_mask_round (A, B, 4, D, E, 4)
#define __builtin_ia32_minmaxss_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxss_mask_round (A, B, 4, D, E, 4)
-#define __builtin_ia32_minmaxbf16128_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16128_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxbf16256_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16256_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxpd128_mask(A, B, C, D, E) __builtin_ia32_minmaxpd128_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxpd256_mask(A, B, C, D, E) __builtin_ia32_minmaxpd256_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxph128_mask(A, B, C, D, E) __builtin_ia32_minmaxph128_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxph256_mask(A, B, C, D, E) __builtin_ia32_minmaxph256_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxps128_mask(A, B, C, D, E) __builtin_ia32_minmaxps128_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxps256_mask(A, B, C, D, E) __builtin_ia32_minmaxps256_mask (A, B, 4, D, E)
#include <wmmintrin.h>
#include <immintrin.h>
/* sm3intrin.h */
#define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
-/* avx10_2-512mediaintrin.h */
+/* avx10_2mediaintrin.h */
#define __builtin_ia32_mpsadbw512(A, B, C) __builtin_ia32_mpsadbw512 (A, B, 1)
#define __builtin_ia32_mpsadbw512_mask(A, B, C, D, E) __builtin_ia32_mpsadbw512_mask (A, B, 1, D, E)
-
-/* avx10_2mediaintrin.h */
-#define __builtin_ia32_mpsadbw128_mask(A, B, C, D, E) __builtin_ia32_mpsadbw128_mask (A, B, 1, D, E)
#define __builtin_ia32_mpsadbw256_mask(A, B, C, D, E) __builtin_ia32_mpsadbw256_mask (A, B, 1, D, E)
+#define __builtin_ia32_mpsadbw128_mask(A, B, C, D, E) __builtin_ia32_mpsadbw128_mask (A, B, 1, D, E)
-/* avx10_2-512convertintrin.h */
+/* avx10_2convertintrin.h */
#define __builtin_ia32_vcvt2ps2phx512_mask_round(A, B, C, D, E) __builtin_ia32_vcvt2ps2phx512_mask_round(A, B, C, D, 8)
-/* avx10_2-512bf16intrin.h */
-#define __builtin_ia32_rndscalebf16512_mask(A, B, C, D) __builtin_ia32_rndscalebf16512_mask(A, 123, C, D)
-#define __builtin_ia32_reducebf16512_mask(A, B, C, D) __builtin_ia32_reducebf16512_mask(A, 123, C, D)
-#define __builtin_ia32_getmantbf16512_mask(A, B, C, D) __builtin_ia32_getmantbf16512_mask(A, 1, C, D)
-#define __builtin_ia32_fpclassbf16512_mask(A, B, C) __builtin_ia32_fpclassbf16512_mask(A, 1, C)
-#define __builtin_ia32_cmpbf16512_mask(A, B, C, D) __builtin_ia32_cmpbf16512_mask(A, B, 1, D)
-
/* avx10_2bf16intrin.h */
+#define __builtin_ia32_rndscalebf16512_mask(A, B, C, D) __builtin_ia32_rndscalebf16512_mask(A, 123, C, D)
#define __builtin_ia32_rndscalebf16256_mask(A, B, C, D) __builtin_ia32_rndscalebf16256_mask(A, 123, C, D)
#define __builtin_ia32_rndscalebf16128_mask(A, B, C, D) __builtin_ia32_rndscalebf16128_mask(A, 123, C, D)
+#define __builtin_ia32_reducebf16512_mask(A, B, C, D) __builtin_ia32_reducebf16512_mask(A, 123, C, D)
#define __builtin_ia32_reducebf16256_mask(A, B, C, D) __builtin_ia32_reducebf16256_mask(A, 123, C, D)
#define __builtin_ia32_reducebf16128_mask(A, B, C, D) __builtin_ia32_reducebf16128_mask(A, 123, C, D)
+#define __builtin_ia32_getmantbf16512_mask(A, B, C, D) __builtin_ia32_getmantbf16512_mask(A, 1, C, D)
#define __builtin_ia32_getmantbf16256_mask(A, B, C, D) __builtin_ia32_getmantbf16256_mask(A, 1, C, D)
#define __builtin_ia32_getmantbf16128_mask(A, B, C, D) __builtin_ia32_getmantbf16128_mask(A, 1, C, D)
+#define __builtin_ia32_fpclassbf16512_mask(A, B, C) __builtin_ia32_fpclassbf16512_mask(A, 1, C)
#define __builtin_ia32_fpclassbf16256_mask(A, B, C) __builtin_ia32_fpclassbf16256_mask(A, 1, C)
#define __builtin_ia32_fpclassbf16128_mask(A, B, C) __builtin_ia32_fpclassbf16128_mask(A, 1, C)
+#define __builtin_ia32_cmpbf16512_mask(A, B, C, D) __builtin_ia32_cmpbf16512_mask(A, B, 1, D)
#define __builtin_ia32_cmpbf16256_mask(A, B, C, D) __builtin_ia32_cmpbf16256_mask(A, B, 1, D)
#define __builtin_ia32_cmpbf16128_mask(A, B, C, D) __builtin_ia32_cmpbf16128_mask(A, B, 1, D)
-/* avx10_2-512satcvtintrin.h */
+/* avx10_2satcvtintrin.h */
#define __builtin_ia32_cvtph2ibs512_mask_round(A, B, C, D) __builtin_ia32_cvtph2ibs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvtph2iubs512_mask_round(A, B, C, D) __builtin_ia32_cvtph2iubs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvtps2ibs512_mask_round(A, B, C, D) __builtin_ia32_cvtps2ibs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvttps2qqs512_mask_round(A, B, C, D) __builtin_ia32_cvttps2qqs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvttps2udqs512_mask_round(A, B, C, D) __builtin_ia32_cvttps2udqs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvttps2uqqs512_mask_round(A, B, C, D) __builtin_ia32_cvttps2uqqs512_mask_round(A, B, C, 8)
-
-/* avx10_2satcvtintrin.h */
#define __builtin_ia32_cvttsd2sis32_round(A, B) __builtin_ia32_cvttsd2sis32_round(A, 8)
#define __builtin_ia32_cvttsd2usis32_round(A, B) __builtin_ia32_cvttsd2usis32_round(A, 8)
#define __builtin_ia32_cvttss2sis32_round(A, B) __builtin_ia32_cvttss2sis32_round(A, 8)
#define __builtin_ia32_cvttss2usis64_round(A, B) __builtin_ia32_cvttss2usis64_round(A, 8)
#endif
-/* avx10_2-512minmaxintrin.h */
+/* avx10_2minmaxintrin.h */
+#define __builtin_ia32_minmaxbf16512_mask(A, B, C, W, U) __builtin_ia32_minmaxbf16512_mask (A, B, 4, W, U)
+#define __builtin_ia32_minmaxbf16256_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16256_mask (A, B, 4, D, E)
+#define __builtin_ia32_minmaxbf16128_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16128_mask (A, B, 4, D, E)
#define __builtin_ia32_minmaxpd512_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxpd512_mask_round (A, B, 4, D, E, 4)
+#define __builtin_ia32_minmaxpd256_mask(A, B, C, D, E) __builtin_ia32_minmaxpd256_mask (A, B, 4, D, E)
+#define __builtin_ia32_minmaxpd128_mask(A, B, C, D, E) __builtin_ia32_minmaxpd128_mask (A, B, 4, D, E)
#define __builtin_ia32_minmaxph512_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxph512_mask_round (A, B, 4, D, E, 4)
+#define __builtin_ia32_minmaxph256_mask(A, B, C, D, E) __builtin_ia32_minmaxph256_mask (A, B, 4, D, E)
+#define __builtin_ia32_minmaxph128_mask(A, B, C, D, E) __builtin_ia32_minmaxph128_mask (A, B, 4, D, E)
#define __builtin_ia32_minmaxps512_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxps512_mask_round (A, B, 4, D, E, 4)
-#define __builtin_ia32_minmaxbf16512_mask(A, B, C, W, U) __builtin_ia32_minmaxbf16512_mask (A, B, 4, W, U)
-
-/* avx10_2minmaxintrin.h */
+#define __builtin_ia32_minmaxps256_mask(A, B, C, D, E) __builtin_ia32_minmaxps256_mask (A, B, 4, D, E)
+#define __builtin_ia32_minmaxps128_mask(A, B, C, D, E) __builtin_ia32_minmaxps128_mask (A, B, 4, D, E)
#define __builtin_ia32_minmaxsd_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxsd_mask_round (A, B, 4, D, E, 4)
#define __builtin_ia32_minmaxsh_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxsh_mask_round (A, B, 4, D, E, 4)
#define __builtin_ia32_minmaxss_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxss_mask_round (A, B, 4, D, E, 4)
-#define __builtin_ia32_minmaxbf16128_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16128_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxbf16256_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16256_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxpd128_mask(A, B, C, D, E) __builtin_ia32_minmaxpd128_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxpd256_mask(A, B, C, D, E) __builtin_ia32_minmaxpd256_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxph128_mask(A, B, C, D, E) __builtin_ia32_minmaxph128_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxph256_mask(A, B, C, D, E) __builtin_ia32_minmaxph256_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxps128_mask(A, B, C, D, E) __builtin_ia32_minmaxps128_mask (A, B, 4, D, E)
-#define __builtin_ia32_minmaxps256_mask(A, B, C, D, E) __builtin_ia32_minmaxps256_mask (A, B, 4, D, E)
#include <x86intrin.h>
/* sm3intrin.h */
test_3 (_mm_sm3rnds2_epi32, __m128i, __m128i, __m128i, __m128i, 1)
-/* avx10_2-512mediaintrin.h */
+/* avx10_2mediaintrin.h */
test_2 (_mm512_mpsadbw_epu8, __m512i, __m512i, __m512i, 1)
test_3 (_mm512_maskz_mpsadbw_epu8, __m512i, __mmask32, __m512i, __m512i, 1)
-test_4 (_mm512_mask_mpsadbw_epu8, __m512i, __m512i, __mmask32, __m512i, __m512i, 1)
-
-/* avx10_2mediaintrin.h */
-test_3 (_mm_maskz_mpsadbw_epu8, __m128i, __mmask8, __m128i, __m128i, 1)
test_3 (_mm256_maskz_mpsadbw_epu8, __m256i, __mmask16, __m256i, __m256i, 1)
-test_4 (_mm_mask_mpsadbw_epu8, __m128i, __m128i, __mmask8, __m128i, __m128i, 1)
+test_3 (_mm_maskz_mpsadbw_epu8, __m128i, __mmask8, __m128i, __m128i, 1)
+test_4 (_mm512_mask_mpsadbw_epu8, __m512i, __m512i, __mmask32, __m512i, __m512i, 1)
test_4 (_mm256_mask_mpsadbw_epu8, __m256i, __m256i, __mmask16, __m256i, __m256i, 1)
+test_4 (_mm_mask_mpsadbw_epu8, __m128i, __m128i, __mmask8, __m128i, __m128i, 1)
-/* avx10_2-512convertintrin.h */
+/* avx10_2convertintrin.h */
test_2 (_mm512_cvtx_round2ps_ph, __m512h, __m512, __m512, 4)
-/* avx10_2-512bf16intrin.h */
-test_1 (_mm512_roundscale_pbh, __m512bh, __m512bh, 123)
-test_2 (_mm512_maskz_roundscale_pbh, __m512bh, __mmask32, __m512bh, 123)
-test_3 (_mm512_mask_roundscale_pbh, __m512bh, __m512bh, __mmask32, __m512bh, 123)
-test_1 (_mm512_reduce_pbh, __m512bh, __m512bh, 123)
-test_2 (_mm512_maskz_reduce_pbh, __m512bh, __mmask32, __m512bh, 123)
-test_3 (_mm512_mask_reduce_pbh, __m512bh, __m512bh, __mmask32, __m512bh, 123)
-test_1x (_mm512_getmant_pbh, __m512bh, __m512bh, 1, 1)
-test_2x (_mm512_maskz_getmant_pbh, __m512bh, __mmask32,__m512bh, 1, 1)
-test_3x (_mm512_mask_getmant_pbh, __m512bh, __m512bh, __mmask32,__m512bh, 1, 1)
-test_1 (_mm512_fpclass_pbh_mask, __mmask32, __m512bh, 13)
-test_2 (_mm512_mask_fpclass_pbh_mask, __mmask32, __mmask32, __m512bh, 13)
-test_2 (_mm512_cmp_pbh_mask, __mmask32, __m512bh, __m512bh, 1)
-test_3 (_mm512_mask_cmp_pbh_mask, __mmask32, __mmask32,__m512bh, __m512bh, 1)
-
/* avx10_2bf16intrin.h */
+test_1 (_mm512_roundscale_pbh, __m512bh, __m512bh, 123)
test_1 (_mm256_roundscale_pbh, __m256bh, __m256bh, 123)
test_1 (_mm_roundscale_pbh, __m128bh, __m128bh, 123)
+test_2 (_mm512_maskz_roundscale_pbh, __m512bh, __mmask32, __m512bh, 123)
test_2 (_mm256_maskz_roundscale_pbh, __m256bh, __mmask16, __m256bh, 123)
test_2 (_mm_maskz_roundscale_pbh, __m128bh, __mmask8, __m128bh, 123)
+test_3 (_mm512_mask_roundscale_pbh, __m512bh, __m512bh, __mmask32, __m512bh, 123)
test_3 (_mm256_mask_roundscale_pbh, __m256bh, __m256bh, __mmask16, __m256bh, 123)
test_3 (_mm_mask_roundscale_pbh, __m128bh, __m128bh, __mmask8, __m128bh, 123)
+test_1 (_mm512_reduce_pbh, __m512bh, __m512bh, 123)
test_1 (_mm256_reduce_pbh, __m256bh, __m256bh, 123)
test_1 (_mm_reduce_pbh, __m128bh, __m128bh, 123)
+test_2 (_mm512_maskz_reduce_pbh, __m512bh, __mmask32, __m512bh, 123)
test_2 (_mm256_maskz_reduce_pbh, __m256bh, __mmask16, __m256bh, 123)
test_2 (_mm_maskz_reduce_pbh, __m128bh, __mmask8, __m128bh, 123)
+test_3 (_mm512_mask_reduce_pbh, __m512bh, __m512bh, __mmask32, __m512bh, 123)
test_3 (_mm256_mask_reduce_pbh, __m256bh, __m256bh, __mmask16, __m256bh, 123)
test_3 (_mm_mask_reduce_pbh, __m128bh, __m128bh, __mmask8, __m128bh, 123)
+test_1x (_mm512_getmant_pbh, __m512bh, __m512bh, 1, 1)
test_1x (_mm256_getmant_pbh, __m256bh, __m256bh, 1, 1)
test_1x (_mm_getmant_pbh, __m128bh, __m128bh, 1, 1)
+test_2x (_mm512_maskz_getmant_pbh, __m512bh, __mmask32,__m512bh, 1, 1)
test_2x (_mm256_maskz_getmant_pbh, __m256bh, __mmask16,__m256bh, 1, 1)
test_2x (_mm_maskz_getmant_pbh, __m128bh, __mmask8, __m128bh, 1, 1)
+test_3x (_mm512_mask_getmant_pbh, __m512bh, __m512bh, __mmask32,__m512bh, 1, 1)
test_3x (_mm256_mask_getmant_pbh, __m256bh, __m256bh, __mmask16,__m256bh, 1, 1)
test_3x (_mm_mask_getmant_pbh, __m128bh, __m128bh, __mmask8, __m128bh, 1, 1)
+test_1 (_mm512_fpclass_pbh_mask, __mmask32, __m512bh, 13)
test_1 (_mm256_fpclass_pbh_mask, __mmask16, __m256bh, 13)
test_1 (_mm_fpclass_pbh_mask, __mmask8, __m128bh, 13)
+test_2 (_mm512_mask_fpclass_pbh_mask, __mmask32, __mmask32, __m512bh, 13)
test_2 (_mm256_mask_fpclass_pbh_mask, __mmask16, __mmask16, __m256bh, 13)
test_2 (_mm_mask_fpclass_pbh_mask, __mmask8, __mmask8, __m128bh, 13)
+test_2 (_mm512_cmp_pbh_mask, __mmask32, __m512bh, __m512bh, 1)
test_2 (_mm256_cmp_pbh_mask, __mmask16, __m256bh, __m256bh, 1)
test_2 (_mm_cmp_pbh_mask, __mmask8, __m128bh, __m128bh, 1)
+test_3 (_mm512_mask_cmp_pbh_mask, __mmask32, __mmask32,__m512bh, __m512bh, 1)
test_3 (_mm256_mask_cmp_pbh_mask, __mmask16, __mmask16, __m256bh, __m256bh, 1)
test_3 (_mm_mask_cmp_pbh_mask, __mmask8, __mmask8, __m128bh, __m128bh, 1)
-/* avx10_2-512satcvtintrin.h */
+/* avx10_2satcvtintrin.h */
test_1 (_mm512_ipcvts_roundph_epi8, __m512i, __m512h, 8)
test_1 (_mm512_ipcvts_roundph_epu8, __m512i, __m512h, 8)
test_1 (_mm512_ipcvts_roundps_epi8, __m512i, __m512, 8)
test_1 (_mm512_cvtts_roundps_epu64, __m512i, __m256, 8)
test_2 (_mm512_maskz_cvtts_roundps_epu64, __m512i, __mmask8, __m256, 8)
test_3 (_mm512_mask_cvtts_roundps_epu64, __m512i, __m512i, __mmask8, __m256, 8)
-
-/* avx10_2satcvtintrin.h */
test_1 (_mm_cvtts_roundsd_epi32, int, __m128d, 8)
test_1 (_mm_cvtts_roundsd_epu32, unsigned int, __m128d, 8)
test_1 (_mm_cvtts_roundss_epi32, int, __m128, 8)
test_1 (_mm_cvtts_roundss_epu64, unsigned long long, __m128, 8)
#endif
-/* avx10_2-512minmaxintrin.h */
+/* avx10_2minmaxintrin.h */
test_2 (_mm512_minmax_pbh, __m512bh, __m512bh, __m512bh, 100)
test_3 (_mm512_maskz_minmax_pbh, __m512bh, __mmask32, __m512bh, __m512bh, 100)
test_4 (_mm512_mask_minmax_pbh, __m512bh, __m512bh, __mmask32, __m512bh, __m512bh, 100)
test_2 (_mm512_minmax_ph, __m512h, __m512h, __m512h, 100)
test_3 (_mm512_maskz_minmax_ph, __m512h, __mmask32, __m512h, __m512h, 100)
test_4 (_mm512_mask_minmax_ph, __m512h, __m512h, __mmask32, __m512h, __m512h, 100)
-
-/* avx10_2minmaxintrin.h */
test_2 (_mm256_minmax_pbh, __m256bh, __m256bh, __m256bh, 100)
test_3 (_mm256_maskz_minmax_pbh, __m256bh, __mmask16, __m256bh, __m256bh, 100)
test_4 (_mm256_mask_minmax_pbh, __m256bh, __m256bh, __mmask16, __m256bh, __m256bh, 100)
/* sm3intrin.h */
test_3 (_mm_sm3rnds2_epi32, __m128i, __m128i, __m128i, __m128i, 1)
-/* avx10_2-512mediaintrin.h */
+/* avx10_2mediaintrin.h */
test_2 (_mm512_mpsadbw_epu8, __m512i, __m512i, __m512i, 1)
test_3 (_mm512_maskz_mpsadbw_epu8, __m512i, __mmask32, __m512i, __m512i, 1)
-test_4 (_mm512_mask_mpsadbw_epu8, __m512i, __m512i, __mmask32, __m512i, __m512i, 1)
-
-/* avx10_2mediaintrin.h */
-test_3 (_mm_maskz_mpsadbw_epu8, __m128i, __mmask8, __m128i, __m128i, 1)
test_3 (_mm256_maskz_mpsadbw_epu8, __m256i, __mmask16, __m256i, __m256i, 1)
-test_4 (_mm_mask_mpsadbw_epu8, __m128i, __m128i, __mmask8, __m128i, __m128i, 1)
+test_3 (_mm_maskz_mpsadbw_epu8, __m128i, __mmask8, __m128i, __m128i, 1)
+test_4 (_mm512_mask_mpsadbw_epu8, __m512i, __m512i, __mmask32, __m512i, __m512i, 1)
test_4 (_mm256_mask_mpsadbw_epu8, __m256i, __m256i, __mmask16, __m256i, __m256i, 1)
+test_4 (_mm_mask_mpsadbw_epu8, __m128i, __m128i, __mmask8, __m128i, __m128i, 1)
-/* avx10_2-512convertintrin.h */
+/* avx10_2convertintrin.h */
test_2 (_mm512_cvtx_round2ps_ph, __m512h, __m512, __m512, 4)
-/* avx10_2-512bf16intrin.h */
-test_1 (_mm512_roundscale_pbh, __m512bh, __m512bh, 123)
-test_2 (_mm512_maskz_roundscale_pbh, __m512bh, __mmask32, __m512bh, 123)
-test_3 (_mm512_mask_roundscale_pbh, __m512bh, __m512bh, __mmask32, __m512bh, 123)
-test_1 (_mm512_reduce_pbh, __m512bh, __m512bh, 123)
-test_2 (_mm512_maskz_reduce_pbh, __m512bh, __mmask32, __m512bh, 123)
-test_3 (_mm512_mask_reduce_pbh, __m512bh, __m512bh, __mmask32, __m512bh, 123)
-test_1x (_mm512_getmant_pbh, __m512bh, __m512bh, 1, 1)
-test_2x (_mm512_maskz_getmant_pbh, __m512bh, __mmask32,__m512bh, 1, 1)
-test_3x (_mm512_mask_getmant_pbh, __m512bh, __m512bh, __mmask32,__m512bh, 1, 1)
-test_1 (_mm512_fpclass_pbh_mask, __mmask32, __m512bh, 13)
-test_2 (_mm512_mask_fpclass_pbh_mask, __mmask32, __mmask32, __m512bh, 13)
-test_2 (_mm512_cmp_pbh_mask, __mmask32, __m512bh, __m512bh, 1)
-test_3 (_mm512_mask_cmp_pbh_mask, __mmask32, __mmask32,__m512bh, __m512bh, 1)
-
/* avx10_2bf16intrin.h */
+test_1 (_mm512_roundscale_pbh, __m512bh, __m512bh, 123)
test_1 (_mm256_roundscale_pbh, __m256bh, __m256bh, 123)
test_1 (_mm_roundscale_pbh, __m128bh, __m128bh, 123)
+test_2 (_mm512_maskz_roundscale_pbh, __m512bh, __mmask32, __m512bh, 123)
test_2 (_mm256_maskz_roundscale_pbh, __m256bh, __mmask16, __m256bh, 123)
test_2 (_mm_maskz_roundscale_pbh, __m128bh, __mmask8, __m128bh, 123)
+test_3 (_mm512_mask_roundscale_pbh, __m512bh, __m512bh, __mmask32, __m512bh, 123)
test_3 (_mm256_mask_roundscale_pbh, __m256bh, __m256bh, __mmask16, __m256bh, 123)
test_3 (_mm_mask_roundscale_pbh, __m128bh, __m128bh, __mmask8, __m128bh, 123)
+test_1 (_mm512_reduce_pbh, __m512bh, __m512bh, 123)
test_1 (_mm256_reduce_pbh, __m256bh, __m256bh, 123)
test_1 (_mm_reduce_pbh, __m128bh, __m128bh, 123)
+test_2 (_mm512_maskz_reduce_pbh, __m512bh, __mmask32, __m512bh, 123)
test_2 (_mm256_maskz_reduce_pbh, __m256bh, __mmask16, __m256bh, 123)
test_2 (_mm_maskz_reduce_pbh, __m128bh, __mmask8, __m128bh, 123)
+test_3 (_mm512_mask_reduce_pbh, __m512bh, __m512bh, __mmask32, __m512bh, 123)
test_3 (_mm256_mask_reduce_pbh, __m256bh, __m256bh, __mmask16, __m256bh, 123)
test_3 (_mm_mask_reduce_pbh, __m128bh, __m128bh, __mmask8, __m128bh, 123)
+test_1x (_mm512_getmant_pbh, __m512bh, __m512bh, 1, 1)
test_1x (_mm256_getmant_pbh, __m256bh, __m256bh, 1, 1)
test_1x (_mm_getmant_pbh, __m128bh, __m128bh, 1, 1)
+test_2x (_mm512_maskz_getmant_pbh, __m512bh, __mmask32,__m512bh, 1, 1)
test_2x (_mm256_maskz_getmant_pbh, __m256bh, __mmask16,__m256bh, 1, 1)
test_2x (_mm_maskz_getmant_pbh, __m128bh, __mmask8, __m128bh, 1, 1)
+test_3x (_mm512_mask_getmant_pbh, __m512bh, __m512bh, __mmask32,__m512bh, 1, 1)
test_3x (_mm256_mask_getmant_pbh, __m256bh, __m256bh, __mmask16,__m256bh, 1, 1)
test_3x (_mm_mask_getmant_pbh, __m128bh, __m128bh, __mmask8, __m128bh, 1, 1)
+test_1 (_mm512_fpclass_pbh_mask, __mmask32, __m512bh, 13)
test_1 (_mm256_fpclass_pbh_mask, __mmask16, __m256bh, 13)
test_1 (_mm_fpclass_pbh_mask, __mmask8, __m128bh, 13)
+test_2 (_mm512_mask_fpclass_pbh_mask, __mmask32, __mmask32, __m512bh, 13)
test_2 (_mm256_mask_fpclass_pbh_mask, __mmask16, __mmask16, __m256bh, 13)
test_2 (_mm_mask_fpclass_pbh_mask, __mmask8, __mmask8, __m128bh, 13)
+test_2 (_mm512_cmp_pbh_mask, __mmask32, __m512bh, __m512bh, 1)
test_2 (_mm256_cmp_pbh_mask, __mmask16, __m256bh, __m256bh, 1)
test_2 (_mm_cmp_pbh_mask, __mmask8, __m128bh, __m128bh, 1)
+test_3 (_mm512_mask_cmp_pbh_mask, __mmask32, __mmask32,__m512bh, __m512bh, 1)
test_3 (_mm256_mask_cmp_pbh_mask, __mmask16, __mmask16, __m256bh, __m256bh, 1)
test_3 (_mm_mask_cmp_pbh_mask, __mmask8, __mmask8, __m128bh, __m128bh, 1)
-/* avx10_2-512satcvtintrin.h */
+/* avx10_2satcvtintrin.h */
test_1 (_mm512_ipcvts_roundph_epi8, __m512i, __m512h, 8)
test_1 (_mm512_ipcvts_roundph_epu8, __m512i, __m512h, 8)
test_1 (_mm512_ipcvts_roundps_epi8, __m512i, __m512, 8)
test_1 (_mm512_cvtts_roundps_epu64, __m512i, __m256, 8)
test_2 (_mm512_maskz_cvtts_roundps_epu64, __m512i, __mmask8, __m256, 8)
test_3 (_mm512_mask_cvtts_roundps_epu64, __m512i, __m512i, __mmask8, __m256, 8)
-
-/* avx10_2satcvtintrin.h */
test_1 (_mm_cvtts_roundsd_epi32, int, __m128d, 8)
test_1 (_mm_cvtts_roundsd_epu32, unsigned int, __m128d, 8)
test_1 (_mm_cvtts_roundss_epi32, int, __m128, 8)
test_1 (_mm_cvtts_roundss_epu64, unsigned long long, __m128, 8)
#endif
-/* avx10_2-512minmaxintrin.h */
+/* avx10_2minmaxintrin.h */
test_2 (_mm512_minmax_pbh, __m512bh, __m512bh, __m512bh, 100)
test_3 (_mm512_maskz_minmax_pbh, __m512bh, __mmask32, __m512bh, __m512bh, 100)
test_4 (_mm512_mask_minmax_pbh, __m512bh, __m512bh, __mmask32, __m512bh, __m512bh, 100)
test_2 (_mm512_minmax_ph, __m512h, __m512h, __m512h, 100)
test_3 (_mm512_maskz_minmax_ph, __m512h, __mmask32, __m512h, __m512h, 100)
test_4 (_mm512_mask_minmax_ph, __m512h, __m512h, __mmask32, __m512h, __m512h, 100)
-
-/* avx10_2minmaxintrin.h */
test_2 (_mm256_minmax_pbh, __m256bh, __m256bh, __m256bh, 100)
test_3 (_mm256_maskz_minmax_pbh, __m256bh, __mmask16, __m256bh, __m256bh, 100)
test_4 (_mm256_mask_minmax_pbh, __m256bh, __m256bh, __mmask16, __m256bh, __m256bh, 100)
/* sm3intrin.h */
#define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
-/* avx10_2-512mediaintrin.h */
+/* avx10_2-mediaintrin.h */
#define __builtin_ia32_mpsadbw512(A, B, C) __builtin_ia32_mpsadbw512 (A, B, 1)
#define __builtin_ia32_mpsadbw512_mask(A, B, C, D, E) __builtin_ia32_mpsadbw512_mask (A, B, 1, D, E)
-
-/* avx10_2-mediaintrin.h */
-#define __builtin_ia32_mpsadbw128_mask(A, B, C, D, E) __builtin_ia32_mpsadbw128_mask (A, B, 1, D, E)
#define __builtin_ia32_mpsadbw256_mask(A, B, C, D, E) __builtin_ia32_mpsadbw256_mask (A, B, 1, D, E)
+#define __builtin_ia32_mpsadbw128_mask(A, B, C, D, E) __builtin_ia32_mpsadbw128_mask (A, B, 1, D, E)
-/* avx10_2-512convertintrin.h */
+/* avx10_2convertintrin.h */
#define __builtin_ia32_vcvt2ps2phx512_mask_round(A, B, C, D, E) __builtin_ia32_vcvt2ps2phx512_mask_round(A, B, C, D, 8)
-/* avx10_2-512bf16intrin.h */
-#define __builtin_ia32_rndscalebf16512_mask(A, B, C, D) __builtin_ia32_rndscalebf16512_mask(A, 123, C, D)
-#define __builtin_ia32_reducebf16512_mask(A, B, C, D) __builtin_ia32_reducebf16512_mask(A, 123, C, D)
-#define __builtin_ia32_getmantbf16512_mask(A, B, C, D) __builtin_ia32_getmantbf16512_mask(A, 1, C, D)
-#define __builtin_ia32_fpclassbf16512_mask(A, B, C) __builtin_ia32_fpclassbf16512_mask(A, 1, C)
-#define __builtin_ia32_cmpbf16512_mask(A, B, C, D) __builtin_ia32_cmpbf16512_mask(A, B, 1, D)
-
/* avx10_2bf16intrin.h */
+#define __builtin_ia32_rndscalebf16512_mask(A, B, C, D) __builtin_ia32_rndscalebf16512_mask(A, 123, C, D)
#define __builtin_ia32_rndscalebf16256_mask(A, B, C, D) __builtin_ia32_rndscalebf16256_mask(A, 123, C, D)
#define __builtin_ia32_rndscalebf16128_mask(A, B, C, D) __builtin_ia32_rndscalebf16128_mask(A, 123, C, D)
+#define __builtin_ia32_reducebf16512_mask(A, B, C, D) __builtin_ia32_reducebf16512_mask(A, 123, C, D)
#define __builtin_ia32_reducebf16256_mask(A, B, C, D) __builtin_ia32_reducebf16256_mask(A, 123, C, D)
#define __builtin_ia32_reducebf16128_mask(A, B, C, D) __builtin_ia32_reducebf16128_mask(A, 123, C, D)
+#define __builtin_ia32_getmantbf16512_mask(A, B, C, D) __builtin_ia32_getmantbf16512_mask(A, 1, C, D)
#define __builtin_ia32_getmantbf16256_mask(A, B, C, D) __builtin_ia32_getmantbf16256_mask(A, 1, C, D)
#define __builtin_ia32_getmantbf16128_mask(A, B, C, D) __builtin_ia32_getmantbf16128_mask(A, 1, C, D)
+#define __builtin_ia32_fpclassbf16512_mask(A, B, C) __builtin_ia32_fpclassbf16512_mask(A, 1, C)
#define __builtin_ia32_fpclassbf16256_mask(A, B, C) __builtin_ia32_fpclassbf16256_mask(A, 1, C)
#define __builtin_ia32_fpclassbf16128_mask(A, B, C) __builtin_ia32_fpclassbf16128_mask(A, 1, C)
+#define __builtin_ia32_cmpbf16512_mask(A, B, C, D) __builtin_ia32_cmpbf16512_mask(A, B, 1, D)
#define __builtin_ia32_cmpbf16256_mask(A, B, C, D) __builtin_ia32_cmpbf16256_mask(A, B, 1, D)
#define __builtin_ia32_cmpbf16128_mask(A, B, C, D) __builtin_ia32_cmpbf16128_mask(A, B, 1, D)
-/* avx10_2-512satcvtintrin.h */
+/* avx10_2satcvtintrin.h */
#define __builtin_ia32_cvtph2ibs512_mask_round(A, B, C, D) __builtin_ia32_cvtph2ibs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvtph2iubs512_mask_round(A, B, C, D) __builtin_ia32_cvtph2iubs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvtps2ibs512_mask_round(A, B, C, D) __builtin_ia32_cvtps2ibs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvttps2qqs512_mask_round(A, B, C, D) __builtin_ia32_cvttps2qqs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvttps2udqs512_mask_round(A, B, C, D) __builtin_ia32_cvttps2udqs512_mask_round(A, B, C, 8)
#define __builtin_ia32_cvttps2uqqs512_mask_round(A, B, C, D) __builtin_ia32_cvttps2uqqs512_mask_round(A, B, C, 8)
-
-/* avx10_2satcvtintrin.h */
#define __builtin_ia32_cvttsd2sis32_round(A, B) __builtin_ia32_cvttsd2sis32_round(A, 8)
#define __builtin_ia32_cvttsd2usis32_round(A, B) __builtin_ia32_cvttsd2usis32_round(A, 8)
#define __builtin_ia32_cvttss2sis32_round(A, B) __builtin_ia32_cvttss2sis32_round(A, 8)
#define __builtin_ia32_cvttss2usis64_round(A, B) __builtin_ia32_cvttss2usis64_round(A, 8)
#endif
-/* avx10_2-512minmaxintrin.h */
+/* avx10_2-minmaxintrin.h */
+#define __builtin_ia32_minmaxbf16512_mask(A, B, C, W, U) __builtin_ia32_minmaxbf16512_mask (A, B, 100, W, U)
+#define __builtin_ia32_minmaxbf16256_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16256_mask (A, B, 100, D, E)
+#define __builtin_ia32_minmaxbf16128_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16128_mask (A, B, 100, D, E)
#define __builtin_ia32_minmaxpd512_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxpd512_mask_round (A, B, 100, D, E, 4)
+#define __builtin_ia32_minmaxpd256_mask(A, B, C, D, E) __builtin_ia32_minmaxpd256_mask (A, B, 100, D, E)
+#define __builtin_ia32_minmaxpd128_mask(A, B, C, D, E) __builtin_ia32_minmaxpd128_mask (A, B, 100, D, E)
#define __builtin_ia32_minmaxph512_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxph512_mask_round (A, B, 100, D, E, 4)
+#define __builtin_ia32_minmaxph256_mask(A, B, C, D, E) __builtin_ia32_minmaxph256_mask (A, B, 100, D, E)
+#define __builtin_ia32_minmaxph128_mask(A, B, C, D, E) __builtin_ia32_minmaxph128_mask (A, B, 100, D, E)
#define __builtin_ia32_minmaxps512_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxps512_mask_round (A, B, 100, D, E, 4)
-#define __builtin_ia32_minmaxbf16512_mask(A, B, C, W, U) __builtin_ia32_minmaxbf16512_mask (A, B, 100, W, U)
-
-/* avx10_2-minmaxintrin.h */
+#define __builtin_ia32_minmaxps256_mask(A, B, C, D, E) __builtin_ia32_minmaxps256_mask (A, B, 100, D, E)
+#define __builtin_ia32_minmaxps128_mask(A, B, C, D, E) __builtin_ia32_minmaxps128_mask (A, B, 100, D, E)
#define __builtin_ia32_minmaxsd_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxsd_mask_round (A, B, 100, D, E, 4)
#define __builtin_ia32_minmaxsh_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxsh_mask_round (A, B, 100, D, E, 4)
#define __builtin_ia32_minmaxss_mask_round(A, B, C, D, E, F) __builtin_ia32_minmaxss_mask_round (A, B, 100, D, E, 4)
-#define __builtin_ia32_minmaxbf16128_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16128_mask (A, B, 100, D, E)
-#define __builtin_ia32_minmaxbf16256_mask(A, B, C, D, E) __builtin_ia32_minmaxbf16256_mask (A, B, 100, D, E)
-#define __builtin_ia32_minmaxpd128_mask(A, B, C, D, E) __builtin_ia32_minmaxpd128_mask (A, B, 100, D, E)
-#define __builtin_ia32_minmaxpd256_mask(A, B, C, D, E) __builtin_ia32_minmaxpd256_mask (A, B, 100, D, E)
-#define __builtin_ia32_minmaxph128_mask(A, B, C, D, E) __builtin_ia32_minmaxph128_mask (A, B, 100, D, E)
-#define __builtin_ia32_minmaxph256_mask(A, B, C, D, E) __builtin_ia32_minmaxph256_mask (A, B, 100, D, E)
-#define __builtin_ia32_minmaxps128_mask(A, B, C, D, E) __builtin_ia32_minmaxps128_mask (A, B, 100, D, E)
-#define __builtin_ia32_minmaxps256_mask(A, B, C, D, E) __builtin_ia32_minmaxps256_mask (A, B, 100, D, E)
#pragma GCC target ("sse4a,3dnow,avx,avx2,fma4,xop,aes,pclmul,popcnt,abm,lzcnt,bmi,bmi2,tbm,lwp,fsgsbase,rdrnd,f16c,fma,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,sha,xsavec,xsaves,clflushopt,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,vpclmulqdq,pconfig,wbnoinvd,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16,sm3,sha512,sm4,avx10.2,amx-avx512,amx-tf32,amx-transpose,amx-fp8,movrs,amx-movrs")
projects / thirdparty / gcc.git / commitdiff
