diff options
author | Haochen Jiang <haochen.jiang@intel.com> | 2023-03-17 14:42:32 +0800 |
---|---|---|
committer | liuhongt <hongtao.liu@intel.com> | 2023-09-22 10:30:42 +0800 |
commit | 36a88e3d464d238c481c9e716d72c5d3a7a21869 (patch) | |
tree | 9ffafc6bb6c8b5109a3a53da06465f429da5f6ab | |
parent | 4a53b3fd1fac091aa44c835cf2759c314f78866e (diff) | |
download | gcc-36a88e3d464d238c481c9e716d72c5d3a7a21869.zip gcc-36a88e3d464d238c481c9e716d72c5d3a7a21869.tar.gz gcc-36a88e3d464d238c481c9e716d72c5d3a7a21869.tar.bz2 |
Push evex512 target for 512 bit intrins
gcc/Changelog:
* config/i386/avx512fp16intrin.h: Add evex512 target for 512 bit
intrins.
Co-authored-by: Hu, Lin1 <lin1.hu@intel.com>
-rw-r--r-- | gcc/config/i386/avx512fp16intrin.h | 5383 |
1 files changed, 2705 insertions, 2678 deletions
diff --git a/gcc/config/i386/avx512fp16intrin.h b/gcc/config/i386/avx512fp16intrin.h index dd083e5..92c0c24e9 100644 --- a/gcc/config/i386/avx512fp16intrin.h +++ b/gcc/config/i386/avx512fp16intrin.h @@ -25,8 +25,8 @@ #error "Never use <avx512fp16intrin.h> directly; include <immintrin.h> instead." #endif -#ifndef __AVX512FP16INTRIN_H_INCLUDED -#define __AVX512FP16INTRIN_H_INCLUDED +#ifndef _AVX512FP16INTRIN_H_INCLUDED +#define _AVX512FP16INTRIN_H_INCLUDED #ifndef __AVX512FP16__ #pragma GCC push_options @@ -37,21 +37,17 @@ /* Internal data types for implementing the intrinsics. */ typedef _Float16 __v8hf __attribute__ ((__vector_size__ (16))); typedef _Float16 __v16hf __attribute__ ((__vector_size__ (32))); -typedef _Float16 __v32hf __attribute__ ((__vector_size__ (64))); /* The Intel API is flexible enough that we must allow aliasing with other vector types, and their scalar components. */ typedef _Float16 __m128h __attribute__ ((__vector_size__ (16), __may_alias__)); typedef _Float16 __m256h __attribute__ ((__vector_size__ (32), __may_alias__)); -typedef _Float16 __m512h __attribute__ ((__vector_size__ (64), __may_alias__)); /* Unaligned version of the same type. */ typedef _Float16 __m128h_u __attribute__ ((__vector_size__ (16), \ __may_alias__, __aligned__ (1))); typedef _Float16 __m256h_u __attribute__ ((__vector_size__ (32), \ __may_alias__, __aligned__ (1))); -typedef _Float16 __m512h_u __attribute__ ((__vector_size__ (64), \ - __may_alias__, __aligned__ (1))); extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) @@ -78,33 +74,8 @@ _mm256_set_ph (_Float16 __A15, _Float16 __A14, _Float16 __A13, __A12, __A13, __A14, __A15 }; } -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_set_ph (_Float16 __A31, _Float16 __A30, _Float16 __A29, - _Float16 __A28, _Float16 __A27, _Float16 __A26, - _Float16 __A25, _Float16 __A24, _Float16 __A23, - _Float16 __A22, _Float16 __A21, _Float16 __A20, - _Float16 __A19, _Float16 __A18, _Float16 __A17, - _Float16 __A16, _Float16 __A15, _Float16 __A14, - _Float16 __A13, _Float16 __A12, _Float16 __A11, - _Float16 __A10, _Float16 __A9, _Float16 __A8, - _Float16 __A7, _Float16 __A6, _Float16 __A5, - _Float16 __A4, _Float16 __A3, _Float16 __A2, - _Float16 __A1, _Float16 __A0) -{ - return __extension__ (__m512h)(__v32hf){ __A0, __A1, __A2, __A3, - __A4, __A5, __A6, __A7, - __A8, __A9, __A10, __A11, - __A12, __A13, __A14, __A15, - __A16, __A17, __A18, __A19, - __A20, __A21, __A22, __A23, - __A24, __A25, __A26, __A27, - __A28, __A29, __A30, __A31 }; -} - -/* Create vectors of elements in the reversed order from _mm_set_ph, - _mm256_set_ph and _mm512_set_ph functions. */ - +/* Create vectors of elements in the reversed order from _mm_set_ph + and _mm256_set_ph functions. */ extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) _mm_setr_ph (_Float16 __A0, _Float16 __A1, _Float16 __A2, @@ -128,30 +99,7 @@ _mm256_setr_ph (_Float16 __A0, _Float16 __A1, _Float16 __A2, __A0); } -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_setr_ph (_Float16 __A0, _Float16 __A1, _Float16 __A2, - _Float16 __A3, _Float16 __A4, _Float16 __A5, - _Float16 __A6, _Float16 __A7, _Float16 __A8, - _Float16 __A9, _Float16 __A10, _Float16 __A11, - _Float16 __A12, _Float16 __A13, _Float16 __A14, - _Float16 __A15, _Float16 __A16, _Float16 __A17, - _Float16 __A18, _Float16 __A19, _Float16 __A20, - _Float16 __A21, _Float16 __A22, _Float16 __A23, - _Float16 __A24, _Float16 __A25, _Float16 __A26, - _Float16 __A27, _Float16 __A28, _Float16 __A29, - _Float16 __A30, _Float16 __A31) - -{ - return _mm512_set_ph (__A31, __A30, __A29, __A28, __A27, __A26, __A25, - __A24, __A23, __A22, __A21, __A20, __A19, __A18, - __A17, __A16, __A15, __A14, __A13, __A12, __A11, - __A10, __A9, __A8, __A7, __A6, __A5, __A4, __A3, - __A2, __A1, __A0); -} - /* Broadcast _Float16 to vector. */ - extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) _mm_set1_ph (_Float16 __A) @@ -167,18 +115,7 @@ _mm256_set1_ph (_Float16 __A) __A, __A, __A, __A, __A, __A, __A, __A); } -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_set1_ph (_Float16 __A) -{ - return _mm512_set_ph (__A, __A, __A, __A, __A, __A, __A, __A, - __A, __A, __A, __A, __A, __A, __A, __A, - __A, __A, __A, __A, __A, __A, __A, __A, - __A, __A, __A, __A, __A, __A, __A, __A); -} - /* Create a vector with all zeros. */ - extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) _mm_setzero_ph (void) @@ -193,13 +130,6 @@ _mm256_setzero_ph (void) return _mm256_set1_ph (0.0f16); } -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_setzero_ph (void) -{ - return _mm512_set1_ph (0.0f16); -} - extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) _mm_undefined_ph (void) @@ -222,24 +152,6 @@ _mm256_undefined_ph (void) return __Y; } -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_undefined_ph (void) -{ -#pragma GCC diagnostic push -#pragma GCC diagnostic ignored "-Winit-self" - __m512h __Y = __Y; -#pragma GCC diagnostic pop - return __Y; -} - -extern __inline _Float16 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtsh_h (__m128h __A) -{ - return __A[0]; -} - extern __inline _Float16 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) _mm256_cvtsh_h (__m256h __A) @@ -247,146 +159,6 @@ _mm256_cvtsh_h (__m256h __A) return __A[0]; } -extern __inline _Float16 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_cvtsh_h (__m512h __A) -{ - return __A[0]; -} - -extern __inline __m512 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_castph_ps (__m512h __a) -{ - return (__m512) __a; -} - -extern __inline __m512d -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_castph_pd (__m512h __a) -{ - return (__m512d) __a; -} - -extern __inline __m512i -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_castph_si512 (__m512h __a) -{ - return (__m512i) __a; -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_castph512_ph128 (__m512h __A) -{ - union - { - __m128h __a[4]; - __m512h __v; - } __u = { .__v = __A }; - return __u.__a[0]; -} - -extern __inline __m256h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_castph512_ph256 (__m512h __A) -{ - union - { - __m256h __a[2]; - __m512h __v; - } __u = { .__v = __A }; - return __u.__a[0]; -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_castph128_ph512 (__m128h __A) -{ - union - { - __m128h __a[4]; - __m512h __v; - } __u; - __u.__a[0] = __A; - return __u.__v; -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_castph256_ph512 (__m256h __A) -{ - union - { - __m256h __a[2]; - __m512h __v; - } __u; - __u.__a[0] = __A; - return __u.__v; -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_zextph128_ph512 (__m128h __A) -{ - return (__m512h) _mm512_insertf32x4 (_mm512_setzero_ps (), - (__m128) __A, 0); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_zextph256_ph512 (__m256h __A) -{ - return (__m512h) _mm512_insertf64x4 (_mm512_setzero_pd (), - (__m256d) __A, 0); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_castps_ph (__m512 __a) -{ - return (__m512h) __a; -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_castpd_ph (__m512d __a) -{ - return (__m512h) __a; -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_castsi512_ph (__m512i __a) -{ - return (__m512h) __a; -} - -/* Create a vector with element 0 as F and the rest zero. */ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_set_sh (_Float16 __F) -{ - return _mm_set_ph (0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, - __F); -} - -/* Create a vector with element 0 as *P and the rest zero. */ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_load_sh (void const *__P) -{ - return _mm_set_ph (0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, - *(_Float16 const *) __P); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_load_ph (void const *__P) -{ - return *(const __m512h *) __P; -} - extern __inline __m256h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) _mm256_load_ph (void const *__P) @@ -401,13 +173,6 @@ _mm_load_ph (void const *__P) return *(const __m128h *) __P; } -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_loadu_ph (void const *__P) -{ - return *(const __m512h_u *) __P; -} - extern __inline __m256h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) _mm256_loadu_ph (void const *__P) @@ -422,21 +187,6 @@ _mm_loadu_ph (void const *__P) return *(const __m128h_u *) __P; } -/* Stores the lower _Float16 value. */ -extern __inline void -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_store_sh (void *__P, __m128h __A) -{ - *(_Float16 *) __P = ((__v8hf)__A)[0]; -} - -extern __inline void -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_store_ph (void *__P, __m512h __A) -{ - *(__m512h *) __P = __A; -} - extern __inline void __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) _mm256_store_ph (void *__P, __m256h __A) @@ -453,13 +203,6 @@ _mm_store_ph (void *__P, __m128h __A) extern __inline void __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_storeu_ph (void *__P, __m512h __A) -{ - *(__m512h_u *) __P = __A; -} - -extern __inline void -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) _mm256_storeu_ph (void *__P, __m256h __A) { *(__m256h_u *) __P = __A; @@ -472,290 +215,30 @@ _mm_storeu_ph (void *__P, __m128h __A) *(__m128h_u *) __P = __A; } -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_abs_ph (__m512h __A) -{ - return (__m512h) _mm512_and_epi32 ( _mm512_set1_epi32 (0x7FFF7FFF), - (__m512i) __A); -} - -/* Intrinsics v[add,sub,mul,div]ph. */ -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_add_ph (__m512h __A, __m512h __B) -{ - return (__m512h) ((__v32hf) __A + (__v32hf) __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_add_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) -{ - return __builtin_ia32_addph512_mask (__C, __D, __A, __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_add_ph (__mmask32 __A, __m512h __B, __m512h __C) -{ - return __builtin_ia32_addph512_mask (__B, __C, - _mm512_setzero_ph (), __A); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_sub_ph (__m512h __A, __m512h __B) -{ - return (__m512h) ((__v32hf) __A - (__v32hf) __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_sub_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) -{ - return __builtin_ia32_subph512_mask (__C, __D, __A, __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_sub_ph (__mmask32 __A, __m512h __B, __m512h __C) -{ - return __builtin_ia32_subph512_mask (__B, __C, - _mm512_setzero_ph (), __A); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mul_ph (__m512h __A, __m512h __B) -{ - return (__m512h) ((__v32hf) __A * (__v32hf) __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_mul_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) -{ - return __builtin_ia32_mulph512_mask (__C, __D, __A, __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_mul_ph (__mmask32 __A, __m512h __B, __m512h __C) -{ - return __builtin_ia32_mulph512_mask (__B, __C, - _mm512_setzero_ph (), __A); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_div_ph (__m512h __A, __m512h __B) -{ - return (__m512h) ((__v32hf) __A / (__v32hf) __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_div_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) -{ - return __builtin_ia32_divph512_mask (__C, __D, __A, __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_div_ph (__mmask32 __A, __m512h __B, __m512h __C) -{ - return __builtin_ia32_divph512_mask (__B, __C, - _mm512_setzero_ph (), __A); -} - -#ifdef __OPTIMIZE__ -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_add_round_ph (__m512h __A, __m512h __B, const int __C) -{ - return __builtin_ia32_addph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, __C); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_add_round_ph (__m512h __A, __mmask32 __B, __m512h __C, - __m512h __D, const int __E) -{ - return __builtin_ia32_addph512_mask_round (__C, __D, __A, __B, __E); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_add_round_ph (__mmask32 __A, __m512h __B, __m512h __C, - const int __D) -{ - return __builtin_ia32_addph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, __D); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_sub_round_ph (__m512h __A, __m512h __B, const int __C) -{ - return __builtin_ia32_subph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, __C); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_sub_round_ph (__m512h __A, __mmask32 __B, __m512h __C, - __m512h __D, const int __E) -{ - return __builtin_ia32_subph512_mask_round (__C, __D, __A, __B, __E); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_sub_round_ph (__mmask32 __A, __m512h __B, __m512h __C, - const int __D) -{ - return __builtin_ia32_subph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, __D); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mul_round_ph (__m512h __A, __m512h __B, const int __C) -{ - return __builtin_ia32_mulph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, __C); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_mul_round_ph (__m512h __A, __mmask32 __B, __m512h __C, - __m512h __D, const int __E) -{ - return __builtin_ia32_mulph512_mask_round (__C, __D, __A, __B, __E); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_mul_round_ph (__mmask32 __A, __m512h __B, __m512h __C, - const int __D) -{ - return __builtin_ia32_mulph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, __D); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_div_round_ph (__m512h __A, __m512h __B, const int __C) -{ - return __builtin_ia32_divph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, __C); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_div_round_ph (__m512h __A, __mmask32 __B, __m512h __C, - __m512h __D, const int __E) -{ - return __builtin_ia32_divph512_mask_round (__C, __D, __A, __B, __E); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_div_round_ph (__mmask32 __A, __m512h __B, __m512h __C, - const int __D) -{ - return __builtin_ia32_divph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, __D); -} -#else -#define _mm512_add_round_ph(A, B, C) \ - ((__m512h)__builtin_ia32_addph512_mask_round((A), (B), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, (C))) - -#define _mm512_mask_add_round_ph(A, B, C, D, E) \ - ((__m512h)__builtin_ia32_addph512_mask_round((C), (D), (A), (B), (E))) - -#define _mm512_maskz_add_round_ph(A, B, C, D) \ - ((__m512h)__builtin_ia32_addph512_mask_round((B), (C), \ - _mm512_setzero_ph (), \ - (A), (D))) - -#define _mm512_sub_round_ph(A, B, C) \ - ((__m512h)__builtin_ia32_subph512_mask_round((A), (B), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, (C))) - -#define _mm512_mask_sub_round_ph(A, B, C, D, E) \ - ((__m512h)__builtin_ia32_subph512_mask_round((C), (D), (A), (B), (E))) - -#define _mm512_maskz_sub_round_ph(A, B, C, D) \ - ((__m512h)__builtin_ia32_subph512_mask_round((B), (C), \ - _mm512_setzero_ph (), \ - (A), (D))) - -#define _mm512_mul_round_ph(A, B, C) \ - ((__m512h)__builtin_ia32_mulph512_mask_round((A), (B), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, (C))) - -#define _mm512_mask_mul_round_ph(A, B, C, D, E) \ - ((__m512h)__builtin_ia32_mulph512_mask_round((C), (D), (A), (B), (E))) - -#define _mm512_maskz_mul_round_ph(A, B, C, D) \ - ((__m512h)__builtin_ia32_mulph512_mask_round((B), (C), \ - _mm512_setzero_ph (), \ - (A), (D))) - -#define _mm512_div_round_ph(A, B, C) \ - ((__m512h)__builtin_ia32_divph512_mask_round((A), (B), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, (C))) - -#define _mm512_mask_div_round_ph(A, B, C, D, E) \ - ((__m512h)__builtin_ia32_divph512_mask_round((C), (D), (A), (B), (E))) - -#define _mm512_maskz_div_round_ph(A, B, C, D) \ - ((__m512h)__builtin_ia32_divph512_mask_round((B), (C), \ - _mm512_setzero_ph (), \ - (A), (D))) -#endif /* __OPTIMIZE__ */ - -extern __inline __m512h +/* Create a vector with element 0 as F and the rest zero. */ +extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_conj_pch (__m512h __A) +_mm_set_sh (_Float16 __F) { - return (__m512h) _mm512_xor_epi32 ((__m512i) __A, _mm512_set1_epi32 (1<<31)); + return _mm_set_ph (0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, + __F); } -extern __inline __m512h +/* Create a vector with element 0 as *P and the rest zero. */ +extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_conj_pch (__m512h __W, __mmask16 __U, __m512h __A) +_mm_load_sh (void const *__P) { - return (__m512h) - __builtin_ia32_movaps512_mask ((__v16sf) _mm512_conj_pch (__A), - (__v16sf) __W, - (__mmask16) __U); + return _mm_set_ph (0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, 0.0f16, + *(_Float16 const *) __P); } -extern __inline __m512h +/* Stores the lower _Float16 value. */ +extern __inline void __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_conj_pch (__mmask16 __U, __m512h __A) +_mm_store_sh (void *__P, __m128h __A) { - return (__m512h) - __builtin_ia32_movaps512_mask ((__v16sf) _mm512_conj_pch (__A), - (__v16sf) _mm512_setzero_ps (), - (__mmask16) __U); + *(_Float16 *) __P = ((__v8hf)__A)[0]; } /* Intrinsics of v[add,sub,mul,div]sh. */ @@ -1012,138 +495,6 @@ _mm_maskz_div_round_sh (__mmask8 __A, __m128h __B, __m128h __C, (A), (D))) #endif /* __OPTIMIZE__ */ -/* Intrinsic vmaxph vminph. */ -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_max_ph (__m512h __A, __m512h __B) -{ - return __builtin_ia32_maxph512_mask (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_max_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) -{ - return __builtin_ia32_maxph512_mask (__C, __D, __A, __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_max_ph (__mmask32 __A, __m512h __B, __m512h __C) -{ - return __builtin_ia32_maxph512_mask (__B, __C, - _mm512_setzero_ph (), __A); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_min_ph (__m512h __A, __m512h __B) -{ - return __builtin_ia32_minph512_mask (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_min_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) -{ - return __builtin_ia32_minph512_mask (__C, __D, __A, __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_min_ph (__mmask32 __A, __m512h __B, __m512h __C) -{ - return __builtin_ia32_minph512_mask (__B, __C, - _mm512_setzero_ph (), __A); -} - -#ifdef __OPTIMIZE__ -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_max_round_ph (__m512h __A, __m512h __B, const int __C) -{ - return __builtin_ia32_maxph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, __C); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_max_round_ph (__m512h __A, __mmask32 __B, __m512h __C, - __m512h __D, const int __E) -{ - return __builtin_ia32_maxph512_mask_round (__C, __D, __A, __B, __E); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_max_round_ph (__mmask32 __A, __m512h __B, __m512h __C, - const int __D) -{ - return __builtin_ia32_maxph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, __D); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_min_round_ph (__m512h __A, __m512h __B, const int __C) -{ - return __builtin_ia32_minph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, __C); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_min_round_ph (__m512h __A, __mmask32 __B, __m512h __C, - __m512h __D, const int __E) -{ - return __builtin_ia32_minph512_mask_round (__C, __D, __A, __B, __E); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_min_round_ph (__mmask32 __A, __m512h __B, __m512h __C, - const int __D) -{ - return __builtin_ia32_minph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, __D); -} - -#else -#define _mm512_max_round_ph(A, B, C) \ - (__builtin_ia32_maxph512_mask_round ((A), (B), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, (C))) - -#define _mm512_mask_max_round_ph(A, B, C, D, E) \ - (__builtin_ia32_maxph512_mask_round ((C), (D), (A), (B), (E))) - -#define _mm512_maskz_max_round_ph(A, B, C, D) \ - (__builtin_ia32_maxph512_mask_round ((B), (C), \ - _mm512_setzero_ph (), \ - (A), (D))) - -#define _mm512_min_round_ph(A, B, C) \ - (__builtin_ia32_minph512_mask_round ((A), (B), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, (C))) - -#define _mm512_mask_min_round_ph(A, B, C, D, E) \ - (__builtin_ia32_minph512_mask_round ((C), (D), (A), (B), (E))) - -#define _mm512_maskz_min_round_ph(A, B, C, D) \ - (__builtin_ia32_minph512_mask_round ((B), (C), \ - _mm512_setzero_ph (), \ - (A), (D))) -#endif /* __OPTIMIZE__ */ - /* Intrinsic vmaxsh vminsh. */ extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) @@ -1275,60 +626,6 @@ _mm_maskz_min_round_sh (__mmask8 __A, __m128h __B, __m128h __C, #endif /* __OPTIMIZE__ */ -/* vcmpph */ -#ifdef __OPTIMIZE -extern __inline __mmask32 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_cmp_ph_mask (__m512h __A, __m512h __B, const int __C) -{ - return (__mmask32) __builtin_ia32_cmpph512_mask (__A, __B, __C, - (__mmask32) -1); -} - -extern __inline __mmask32 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_cmp_ph_mask (__mmask32 __A, __m512h __B, __m512h __C, - const int __D) -{ - return (__mmask32) __builtin_ia32_cmpph512_mask (__B, __C, __D, - __A); -} - -extern __inline __mmask32 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_cmp_round_ph_mask (__m512h __A, __m512h __B, const int __C, - const int __D) -{ - return (__mmask32) __builtin_ia32_cmpph512_mask_round (__A, __B, - __C, (__mmask32) -1, - __D); -} - -extern __inline __mmask32 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_cmp_round_ph_mask (__mmask32 __A, __m512h __B, __m512h __C, - const int __D, const int __E) -{ - return (__mmask32) __builtin_ia32_cmpph512_mask_round (__B, __C, - __D, __A, - __E); -} - -#else -#define _mm512_cmp_ph_mask(A, B, C) \ - (__builtin_ia32_cmpph512_mask ((A), (B), (C), (-1))) - -#define _mm512_mask_cmp_ph_mask(A, B, C, D) \ - (__builtin_ia32_cmpph512_mask ((B), (C), (D), (A))) - -#define _mm512_cmp_round_ph_mask(A, B, C, D) \ - (__builtin_ia32_cmpph512_mask_round ((A), (B), (C), (-1), (D))) - -#define _mm512_mask_cmp_round_ph_mask(A, B, C, D, E) \ - (__builtin_ia32_cmpph512_mask_round ((B), (C), (D), (A), (E))) - -#endif /* __OPTIMIZE__ */ - /* Intrinsics vcmpsh. */ #ifdef __OPTIMIZE__ extern __inline __mmask8 @@ -1525,126 +822,6 @@ _mm_comi_round_sh (__m128h __A, __m128h __B, const int __P, const int __R) #endif /* __OPTIMIZE__ */ -/* Intrinsics vsqrtph. */ -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_sqrt_ph (__m512h __A) -{ - return __builtin_ia32_sqrtph512_mask_round (__A, - _mm512_setzero_ph(), - (__mmask32) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_sqrt_ph (__m512h __A, __mmask32 __B, __m512h __C) -{ - return __builtin_ia32_sqrtph512_mask_round (__C, __A, __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_sqrt_ph (__mmask32 __A, __m512h __B) -{ - return __builtin_ia32_sqrtph512_mask_round (__B, - _mm512_setzero_ph (), - __A, - _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_sqrt_round_ph (__m512h __A, const int __B) -{ - return __builtin_ia32_sqrtph512_mask_round (__A, - _mm512_setzero_ph(), - (__mmask32) -1, __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_sqrt_round_ph (__m512h __A, __mmask32 __B, __m512h __C, - const int __D) -{ - return __builtin_ia32_sqrtph512_mask_round (__C, __A, __B, __D); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_sqrt_round_ph (__mmask32 __A, __m512h __B, const int __C) -{ - return __builtin_ia32_sqrtph512_mask_round (__B, - _mm512_setzero_ph (), - __A, __C); -} - -#else -#define _mm512_sqrt_round_ph(A, B) \ - (__builtin_ia32_sqrtph512_mask_round ((A), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, (B))) - -#define _mm512_mask_sqrt_round_ph(A, B, C, D) \ - (__builtin_ia32_sqrtph512_mask_round ((C), (A), (B), (D))) - -#define _mm512_maskz_sqrt_round_ph(A, B, C) \ - (__builtin_ia32_sqrtph512_mask_round ((B), \ - _mm512_setzero_ph (), \ - (A), (C))) - -#endif /* __OPTIMIZE__ */ - -/* Intrinsics vrsqrtph. */ -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_rsqrt_ph (__m512h __A) -{ - return __builtin_ia32_rsqrtph512_mask (__A, _mm512_setzero_ph (), - (__mmask32) -1); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_rsqrt_ph (__m512h __A, __mmask32 __B, __m512h __C) -{ - return __builtin_ia32_rsqrtph512_mask (__C, __A, __B); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_rsqrt_ph (__mmask32 __A, __m512h __B) -{ - return __builtin_ia32_rsqrtph512_mask (__B, _mm512_setzero_ph (), - __A); -} - -/* Intrinsics vrsqrtsh. */ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_rsqrt_sh (__m128h __A, __m128h __B) -{ - return __builtin_ia32_rsqrtsh_mask (__B, __A, _mm_setzero_ph (), - (__mmask8) -1); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_rsqrt_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D) -{ - return __builtin_ia32_rsqrtsh_mask (__D, __C, __A, __B); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_rsqrt_sh (__mmask8 __A, __m128h __B, __m128h __C) -{ - return __builtin_ia32_rsqrtsh_mask (__C, __B, _mm_setzero_ph (), - __A); -} - /* Intrinsics vsqrtsh. */ extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) @@ -1718,28 +895,28 @@ _mm_maskz_sqrt_round_sh (__mmask8 __A, __m128h __B, __m128h __C, #endif /* __OPTIMIZE__ */ -/* Intrinsics vrcpph. */ -extern __inline __m512h +/* Intrinsics vrsqrtsh. */ +extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_rcp_ph (__m512h __A) +_mm_rsqrt_sh (__m128h __A, __m128h __B) { - return __builtin_ia32_rcpph512_mask (__A, _mm512_setzero_ph (), - (__mmask32) -1); + return __builtin_ia32_rsqrtsh_mask (__B, __A, _mm_setzero_ph (), + (__mmask8) -1); } -extern __inline __m512h +extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_rcp_ph (__m512h __A, __mmask32 __B, __m512h __C) +_mm_mask_rsqrt_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D) { - return __builtin_ia32_rcpph512_mask (__C, __A, __B); + return __builtin_ia32_rsqrtsh_mask (__D, __C, __A, __B); } -extern __inline __m512h +extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_rcp_ph (__mmask32 __A, __m512h __B) +_mm_maskz_rsqrt_sh (__mmask8 __A, __m128h __B, __m128h __C) { - return __builtin_ia32_rcpph512_mask (__B, _mm512_setzero_ph (), - __A); + return __builtin_ia32_rsqrtsh_mask (__C, __B, _mm_setzero_ph (), + __A); } /* Intrinsics vrcpsh. */ @@ -1766,80 +943,6 @@ _mm_maskz_rcp_sh (__mmask32 __A, __m128h __B, __m128h __C) __A); } -/* Intrinsics vscalefph. */ -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_scalef_ph (__m512h __A, __m512h __B) -{ - return __builtin_ia32_scalefph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_scalef_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) -{ - return __builtin_ia32_scalefph512_mask_round (__C, __D, __A, __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_scalef_ph (__mmask32 __A, __m512h __B, __m512h __C) -{ - return __builtin_ia32_scalefph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, - _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_scalef_round_ph (__m512h __A, __m512h __B, const int __C) -{ - return __builtin_ia32_scalefph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, __C); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_scalef_round_ph (__m512h __A, __mmask32 __B, __m512h __C, - __m512h __D, const int __E) -{ - return __builtin_ia32_scalefph512_mask_round (__C, __D, __A, __B, - __E); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_scalef_round_ph (__mmask32 __A, __m512h __B, __m512h __C, - const int __D) -{ - return __builtin_ia32_scalefph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, __D); -} - -#else -#define _mm512_scalef_round_ph(A, B, C) \ - (__builtin_ia32_scalefph512_mask_round ((A), (B), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, (C))) - -#define _mm512_mask_scalef_round_ph(A, B, C, D, E) \ - (__builtin_ia32_scalefph512_mask_round ((C), (D), (A), (B), (E))) - -#define _mm512_maskz_scalef_round_ph(A, B, C, D) \ - (__builtin_ia32_scalefph512_mask_round ((B), (C), \ - _mm512_setzero_ph (), \ - (A), (D))) - -#endif /* __OPTIMIZE__ */ - /* Intrinsics vscalefsh. */ extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) @@ -1913,95 +1016,6 @@ _mm_maskz_scalef_round_sh (__mmask8 __A, __m128h __B, __m128h __C, #endif /* __OPTIMIZE__ */ -/* Intrinsics vreduceph. */ -#ifdef __OPTIMIZE__ -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_reduce_ph (__m512h __A, int __B) -{ - return __builtin_ia32_reduceph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_reduce_ph (__m512h __A, __mmask32 __B, __m512h __C, int __D) -{ - return __builtin_ia32_reduceph512_mask_round (__C, __D, __A, __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_reduce_ph (__mmask32 __A, __m512h __B, int __C) -{ - return __builtin_ia32_reduceph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_reduce_round_ph (__m512h __A, int __B, const int __C) -{ - return __builtin_ia32_reduceph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, __C); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_reduce_round_ph (__m512h __A, __mmask32 __B, __m512h __C, - int __D, const int __E) -{ - return __builtin_ia32_reduceph512_mask_round (__C, __D, __A, __B, - __E); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_reduce_round_ph (__mmask32 __A, __m512h __B, int __C, - const int __D) -{ - return __builtin_ia32_reduceph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, __D); -} - -#else -#define _mm512_reduce_ph(A, B) \ - (__builtin_ia32_reduceph512_mask_round ((A), (B), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, \ - _MM_FROUND_CUR_DIRECTION)) - -#define _mm512_mask_reduce_ph(A, B, C, D) \ - (__builtin_ia32_reduceph512_mask_round ((C), (D), (A), (B), \ - _MM_FROUND_CUR_DIRECTION)) - -#define _mm512_maskz_reduce_ph(A, B, C) \ - (__builtin_ia32_reduceph512_mask_round ((B), (C), \ - _mm512_setzero_ph (), \ - (A), _MM_FROUND_CUR_DIRECTION)) - -#define _mm512_reduce_round_ph(A, B, C) \ - (__builtin_ia32_reduceph512_mask_round ((A), (B), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, (C))) - -#define _mm512_mask_reduce_round_ph(A, B, C, D, E) \ - (__builtin_ia32_reduceph512_mask_round ((C), (D), (A), (B), (E))) - -#define _mm512_maskz_reduce_round_ph(A, B, C, D) \ - (__builtin_ia32_reduceph512_mask_round ((B), (C), \ - _mm512_setzero_ph (), \ - (A), (D))) - -#endif /* __OPTIMIZE__ */ - /* Intrinsics vreducesh. */ #ifdef __OPTIMIZE__ extern __inline __m128h @@ -2091,97 +1105,6 @@ _mm_maskz_reduce_round_sh (__mmask8 __A, __m128h __B, __m128h __C, #endif /* __OPTIMIZE__ */ -/* Intrinsics vrndscaleph. */ -#ifdef __OPTIMIZE__ -extern __inline __m512h - __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_roundscale_ph (__m512h __A, int __B) -{ - return __builtin_ia32_rndscaleph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_roundscale_ph (__m512h __A, __mmask32 __B, - __m512h __C, int __D) -{ - return __builtin_ia32_rndscaleph512_mask_round (__C, __D, __A, __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_roundscale_ph (__mmask32 __A, __m512h __B, int __C) -{ - return __builtin_ia32_rndscaleph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_roundscale_round_ph (__m512h __A, int __B, const int __C) -{ - return __builtin_ia32_rndscaleph512_mask_round (__A, __B, - _mm512_setzero_ph (), - (__mmask32) -1, - __C); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_roundscale_round_ph (__m512h __A, __mmask32 __B, - __m512h __C, int __D, const int __E) -{ - return __builtin_ia32_rndscaleph512_mask_round (__C, __D, __A, - __B, __E); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_roundscale_round_ph (__mmask32 __A, __m512h __B, int __C, - const int __D) -{ - return __builtin_ia32_rndscaleph512_mask_round (__B, __C, - _mm512_setzero_ph (), - __A, __D); -} - -#else -#define _mm512_roundscale_ph(A, B) \ - (__builtin_ia32_rndscaleph512_mask_round ((A), (B), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, \ - _MM_FROUND_CUR_DIRECTION)) - -#define _mm512_mask_roundscale_ph(A, B, C, D) \ - (__builtin_ia32_rndscaleph512_mask_round ((C), (D), (A), (B), \ - _MM_FROUND_CUR_DIRECTION)) - -#define _mm512_maskz_roundscale_ph(A, B, C) \ - (__builtin_ia32_rndscaleph512_mask_round ((B), (C), \ - _mm512_setzero_ph (), \ - (A), \ - _MM_FROUND_CUR_DIRECTION)) -#define _mm512_roundscale_round_ph(A, B, C) \ - (__builtin_ia32_rndscaleph512_mask_round ((A), (B), \ - _mm512_setzero_ph (), \ - (__mmask32)-1, (C))) - -#define _mm512_mask_roundscale_round_ph(A, B, C, D, E) \ - (__builtin_ia32_rndscaleph512_mask_round ((C), (D), (A), (B), (E))) - -#define _mm512_maskz_roundscale_round_ph(A, B, C, D) \ - (__builtin_ia32_rndscaleph512_mask_round ((B), (C), \ - _mm512_setzero_ph (), \ - (A), (D))) - -#endif /* __OPTIMIZE__ */ - /* Intrinsics vrndscalesh. */ #ifdef __OPTIMIZE__ extern __inline __m128h @@ -2297,39 +1220,10 @@ _mm_mask_fpclass_sh_mask (__mmask8 __U, __m128h __A, const int __imm) #define _mm_mask_fpclass_sh_mask(U, X, C) \ ((__mmask8) __builtin_ia32_fpclasssh_mask ((__v8hf) (__m128h) (X), \ (int) (C), (__mmask8) (U))) -#endif /* __OPTIMIZE__ */ - -/* Intrinsics vfpclassph. */ -#ifdef __OPTIMIZE__ -extern __inline __mmask32 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_fpclass_ph_mask (__mmask32 __U, __m512h __A, - const int __imm) -{ - return (__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) __A, - __imm, __U); -} - -extern __inline __mmask32 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_fpclass_ph_mask (__m512h __A, const int __imm) -{ - return (__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) __A, - __imm, - (__mmask32) -1); -} - -#else -#define _mm512_mask_fpclass_ph_mask(u, x, c) \ - ((__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) (__m512h) (x), \ - (int) (c),(__mmask8)(u))) -#define _mm512_fpclass_ph_mask(x, c) \ - ((__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) (__m512h) (x), \ - (int) (c),(__mmask8)-1)) -#endif /* __OPIMTIZE__ */ +#endif /* __OPTIMIZE__ */ -/* Intrinsics vgetexpph, vgetexpsh. */ +/* Intrinsics vgetexpsh. */ extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) _mm_getexp_sh (__m128h __A, __m128h __B) @@ -2362,35 +1256,6 @@ _mm_maskz_getexp_sh (__mmask8 __U, __m128h __A, __m128h __B) _MM_FROUND_CUR_DIRECTION); } -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_getexp_ph (__m512h __A) -{ - return (__m512h) - __builtin_ia32_getexpph512_mask ((__v32hf) __A, - (__v32hf) _mm512_setzero_ph (), - (__mmask32) -1, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_getexp_ph (__m512h __W, __mmask32 __U, __m512h __A) -{ - return (__m512h) - __builtin_ia32_getexpph512_mask ((__v32hf) __A, (__v32hf) __W, - (__mmask32) __U, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_getexp_ph (__mmask32 __U, __m512h __A) -{ - return (__m512h) - __builtin_ia32_getexpph512_mask ((__v32hf) __A, - (__v32hf) _mm512_setzero_ph (), - (__mmask32) __U, _MM_FROUND_CUR_DIRECTION); -} - #ifdef __OPTIMIZE__ extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) @@ -2426,36 +1291,6 @@ _mm_maskz_getexp_round_sh (__mmask8 __U, __m128h __A, __m128h __B, (__mmask8) __U, __R); } -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_getexp_round_ph (__m512h __A, const int __R) -{ - return (__m512h) __builtin_ia32_getexpph512_mask ((__v32hf) __A, - (__v32hf) - _mm512_setzero_ph (), - (__mmask32) -1, __R); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_getexp_round_ph (__m512h __W, __mmask32 __U, __m512h __A, - const int __R) -{ - return (__m512h) __builtin_ia32_getexpph512_mask ((__v32hf) __A, - (__v32hf) __W, - (__mmask32) __U, __R); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_getexp_round_ph (__mmask32 __U, __m512h __A, const int __R) -{ - return (__m512h) __builtin_ia32_getexpph512_mask ((__v32hf) __A, - (__v32hf) - _mm512_setzero_ph (), - (__mmask32) __U, __R); -} - #else #define _mm_getexp_round_sh(A, B, R) \ ((__m128h)__builtin_ia32_getexpsh_mask_round((__v8hf)(__m128h)(A), \ @@ -2471,21 +1306,9 @@ _mm512_maskz_getexp_round_ph (__mmask32 __U, __m512h __A, const int __R) (__v8hf)_mm_setzero_ph(), \ U, C) -#define _mm512_getexp_round_ph(A, R) \ - ((__m512h)__builtin_ia32_getexpph512_mask((__v32hf)(__m512h)(A), \ - (__v32hf)_mm512_setzero_ph(), (__mmask32)-1, R)) - -#define _mm512_mask_getexp_round_ph(W, U, A, R) \ - ((__m512h)__builtin_ia32_getexpph512_mask((__v32hf)(__m512h)(A), \ - (__v32hf)(__m512h)(W), (__mmask32)(U), R)) - -#define _mm512_maskz_getexp_round_ph(U, A, R) \ - ((__m512h)__builtin_ia32_getexpph512_mask((__v32hf)(__m512h)(A), \ - (__v32hf)_mm512_setzero_ph(), (__mmask32)(U), R)) - #endif /* __OPTIMIZE__ */ -/* Intrinsics vgetmantph, vgetmantsh. */ +/* Intrinsics vgetmantsh. */ #ifdef __OPTIMIZE__ extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) @@ -2525,44 +1348,6 @@ _mm_maskz_getmant_sh (__mmask8 __U, __m128h __A, __m128h __B, __U, _MM_FROUND_CUR_DIRECTION); } -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_getmant_ph (__m512h __A, _MM_MANTISSA_NORM_ENUM __B, - _MM_MANTISSA_SIGN_ENUM __C) -{ - return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, - (__C << 2) | __B, - _mm512_setzero_ph (), - (__mmask32) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_getmant_ph (__m512h __W, __mmask32 __U, __m512h __A, - _MM_MANTISSA_NORM_ENUM __B, - _MM_MANTISSA_SIGN_ENUM __C) -{ - return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, - (__C << 2) | __B, - (__v32hf) __W, __U, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_getmant_ph (__mmask32 __U, __m512h __A, - _MM_MANTISSA_NORM_ENUM __B, - _MM_MANTISSA_SIGN_ENUM __C) -{ - return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, - (__C << 2) | __B, - (__v32hf) - _mm512_setzero_ph (), - __U, - _MM_FROUND_CUR_DIRECTION); -} - extern __inline __m128h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) _mm_getmant_round_sh (__m128h __A, __m128h __B, @@ -2604,67 +1389,7 @@ _mm_maskz_getmant_round_sh (__mmask8 __U, __m128h __A, __m128h __B, __U, __R); } -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_getmant_round_ph (__m512h __A, _MM_MANTISSA_NORM_ENUM __B, - _MM_MANTISSA_SIGN_ENUM __C, const int __R) -{ - return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, - (__C << 2) | __B, - _mm512_setzero_ph (), - (__mmask32) -1, __R); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_mask_getmant_round_ph (__m512h __W, __mmask32 __U, __m512h __A, - _MM_MANTISSA_NORM_ENUM __B, - _MM_MANTISSA_SIGN_ENUM __C, const int __R) -{ - return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, - (__C << 2) | __B, - (__v32hf) __W, __U, - __R); -} - -extern __inline __m512h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm512_maskz_getmant_round_ph (__mmask32 __U, __m512h __A, - _MM_MANTISSA_NORM_ENUM __B, - _MM_MANTISSA_SIGN_ENUM __C, const int __R) -{ - return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, - (__C << 2) | __B, - (__v32hf) - _mm512_setzero_ph (), - __U, __R); -} - #else -#define _mm512_getmant_ph(X, B, C) \ - ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ - (int)(((C)<<2) | (B)), \ - (__v32hf)(__m512h) \ - _mm512_setzero_ph(), \ - (__mmask32)-1, \ - _MM_FROUND_CUR_DIRECTION)) - -#define _mm512_mask_getmant_ph(W, U, X, B, C) \ - ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ - (int)(((C)<<2) | (B)), \ - (__v32hf)(__m512h)(W), \ - (__mmask32)(U), \ - _MM_FROUND_CUR_DIRECTION)) - - -#define _mm512_maskz_getmant_ph(U, X, B, C) \ - ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ - (int)(((C)<<2) | (B)), \ - (__v32hf)(__m512h) \ - _mm512_setzero_ph(), \ - (__mmask32)(U), \ - _MM_FROUND_CUR_DIRECTION)) - #define _mm_getmant_sh(X, Y, C, D) \ ((__m128h)__builtin_ia32_getmantsh_mask_round ((__v8hf)(__m128h)(X), \ (__v8hf)(__m128h)(Y), \ @@ -2691,30 +1416,6 @@ _mm512_maskz_getmant_round_ph (__mmask32 __U, __m512h __A, (__mmask8)(U), \ _MM_FROUND_CUR_DIRECTION)) -#define _mm512_getmant_round_ph(X, B, C, R) \ - ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ - (int)(((C)<<2) | (B)), \ - (__v32hf)(__m512h) \ - _mm512_setzero_ph(), \ - (__mmask32)-1, \ - (R))) - -#define _mm512_mask_getmant_round_ph(W, U, X, B, C, R) \ - ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ - (int)(((C)<<2) | (B)), \ - (__v32hf)(__m512h)(W), \ - (__mmask32)(U), \ - (R))) - - -#define _mm512_maskz_getmant_round_ph(U, X, B, C, R) \ - ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ - (int)(((C)<<2) | (B)), \ - (__v32hf)(__m512h) \ - _mm512_setzero_ph(), \ - (__mmask32)(U), \ - (R))) - #define _mm_getmant_round_sh(X, Y, C, D, R) \ ((__m128h)__builtin_ia32_getmantsh_mask_round ((__v8hf)(__m128h)(X), \ (__v8hf)(__m128h)(Y), \ @@ -2802,6 +1503,2674 @@ _mm_maskz_move_sh (__mmask8 __A, __m128h __B, __m128h __C) return __builtin_ia32_vmovsh_mask (__B, __C, _mm_setzero_ph (), __A); } +/* Intrinsics vcvtsh2si, vcvtsh2us. */ +extern __inline int +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsh_i32 (__m128h __A) +{ + return (int) __builtin_ia32_vcvtsh2si32_round (__A, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline unsigned +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsh_u32 (__m128h __A) +{ + return (int) __builtin_ia32_vcvtsh2usi32_round (__A, + _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline int +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundsh_i32 (__m128h __A, const int __R) +{ + return (int) __builtin_ia32_vcvtsh2si32_round (__A, __R); +} + +extern __inline unsigned +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundsh_u32 (__m128h __A, const int __R) +{ + return (int) __builtin_ia32_vcvtsh2usi32_round (__A, __R); +} + +#else +#define _mm_cvt_roundsh_i32(A, B) \ + ((int)__builtin_ia32_vcvtsh2si32_round ((A), (B))) +#define _mm_cvt_roundsh_u32(A, B) \ + ((int)__builtin_ia32_vcvtsh2usi32_round ((A), (B))) + +#endif /* __OPTIMIZE__ */ + +#ifdef __x86_64__ +extern __inline long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsh_i64 (__m128h __A) +{ + return (long long) + __builtin_ia32_vcvtsh2si64_round (__A, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline unsigned long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsh_u64 (__m128h __A) +{ + return (long long) + __builtin_ia32_vcvtsh2usi64_round (__A, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundsh_i64 (__m128h __A, const int __R) +{ + return (long long) __builtin_ia32_vcvtsh2si64_round (__A, __R); +} + +extern __inline unsigned long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundsh_u64 (__m128h __A, const int __R) +{ + return (long long) __builtin_ia32_vcvtsh2usi64_round (__A, __R); +} + +#else +#define _mm_cvt_roundsh_i64(A, B) \ + ((long long)__builtin_ia32_vcvtsh2si64_round ((A), (B))) +#define _mm_cvt_roundsh_u64(A, B) \ + ((long long)__builtin_ia32_vcvtsh2usi64_round ((A), (B))) + +#endif /* __OPTIMIZE__ */ +#endif /* __x86_64__ */ + +/* Intrinsics vcvtsi2sh, vcvtusi2sh. */ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvti32_sh (__m128h __A, int __B) +{ + return __builtin_ia32_vcvtsi2sh32_round (__A, __B, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtu32_sh (__m128h __A, unsigned int __B) +{ + return __builtin_ia32_vcvtusi2sh32_round (__A, __B, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundi32_sh (__m128h __A, int __B, const int __R) +{ + return __builtin_ia32_vcvtsi2sh32_round (__A, __B, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundu32_sh (__m128h __A, unsigned int __B, const int __R) +{ + return __builtin_ia32_vcvtusi2sh32_round (__A, __B, __R); +} + +#else +#define _mm_cvt_roundi32_sh(A, B, C) \ + (__builtin_ia32_vcvtsi2sh32_round ((A), (B), (C))) +#define _mm_cvt_roundu32_sh(A, B, C) \ + (__builtin_ia32_vcvtusi2sh32_round ((A), (B), (C))) + +#endif /* __OPTIMIZE__ */ + +#ifdef __x86_64__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvti64_sh (__m128h __A, long long __B) +{ + return __builtin_ia32_vcvtsi2sh64_round (__A, __B, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtu64_sh (__m128h __A, unsigned long long __B) +{ + return __builtin_ia32_vcvtusi2sh64_round (__A, __B, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundi64_sh (__m128h __A, long long __B, const int __R) +{ + return __builtin_ia32_vcvtsi2sh64_round (__A, __B, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundu64_sh (__m128h __A, unsigned long long __B, const int __R) +{ + return __builtin_ia32_vcvtusi2sh64_round (__A, __B, __R); +} + +#else +#define _mm_cvt_roundi64_sh(A, B, C) \ + (__builtin_ia32_vcvtsi2sh64_round ((A), (B), (C))) +#define _mm_cvt_roundu64_sh(A, B, C) \ + (__builtin_ia32_vcvtusi2sh64_round ((A), (B), (C))) + +#endif /* __OPTIMIZE__ */ +#endif /* __x86_64__ */ + +/* Intrinsics vcvttsh2si, vcvttsh2us. */ +extern __inline int +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvttsh_i32 (__m128h __A) +{ + return (int) + __builtin_ia32_vcvttsh2si32_round (__A, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline unsigned +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvttsh_u32 (__m128h __A) +{ + return (int) + __builtin_ia32_vcvttsh2usi32_round (__A, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline int +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtt_roundsh_i32 (__m128h __A, const int __R) +{ + return (int) __builtin_ia32_vcvttsh2si32_round (__A, __R); +} + +extern __inline unsigned +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtt_roundsh_u32 (__m128h __A, const int __R) +{ + return (int) __builtin_ia32_vcvttsh2usi32_round (__A, __R); +} + +#else +#define _mm_cvtt_roundsh_i32(A, B) \ + ((int)__builtin_ia32_vcvttsh2si32_round ((A), (B))) +#define _mm_cvtt_roundsh_u32(A, B) \ + ((int)__builtin_ia32_vcvttsh2usi32_round ((A), (B))) + +#endif /* __OPTIMIZE__ */ + +#ifdef __x86_64__ +extern __inline long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvttsh_i64 (__m128h __A) +{ + return (long long) + __builtin_ia32_vcvttsh2si64_round (__A, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline unsigned long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvttsh_u64 (__m128h __A) +{ + return (long long) + __builtin_ia32_vcvttsh2usi64_round (__A, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtt_roundsh_i64 (__m128h __A, const int __R) +{ + return (long long) __builtin_ia32_vcvttsh2si64_round (__A, __R); +} + +extern __inline unsigned long long +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtt_roundsh_u64 (__m128h __A, const int __R) +{ + return (long long) __builtin_ia32_vcvttsh2usi64_round (__A, __R); +} + +#else +#define _mm_cvtt_roundsh_i64(A, B) \ + ((long long)__builtin_ia32_vcvttsh2si64_round ((A), (B))) +#define _mm_cvtt_roundsh_u64(A, B) \ + ((long long)__builtin_ia32_vcvttsh2usi64_round ((A), (B))) + +#endif /* __OPTIMIZE__ */ +#endif /* __x86_64__ */ + +/* Intrinsics vcvtsh2ss, vcvtsh2sd. */ +extern __inline __m128 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsh_ss (__m128 __A, __m128h __B) +{ + return __builtin_ia32_vcvtsh2ss_mask_round (__B, __A, + _mm_setzero_ps (), + (__mmask8) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_cvtsh_ss (__m128 __A, __mmask8 __B, __m128 __C, + __m128h __D) +{ + return __builtin_ia32_vcvtsh2ss_mask_round (__D, __C, __A, __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_cvtsh_ss (__mmask8 __A, __m128 __B, + __m128h __C) +{ + return __builtin_ia32_vcvtsh2ss_mask_round (__C, __B, + _mm_setzero_ps (), + __A, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128d +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsh_sd (__m128d __A, __m128h __B) +{ + return __builtin_ia32_vcvtsh2sd_mask_round (__B, __A, + _mm_setzero_pd (), + (__mmask8) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128d +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_cvtsh_sd (__m128d __A, __mmask8 __B, __m128d __C, + __m128h __D) +{ + return __builtin_ia32_vcvtsh2sd_mask_round (__D, __C, __A, __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128d +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_cvtsh_sd (__mmask8 __A, __m128d __B, __m128h __C) +{ + return __builtin_ia32_vcvtsh2sd_mask_round (__C, __B, + _mm_setzero_pd (), + __A, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline __m128 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundsh_ss (__m128 __A, __m128h __B, const int __R) +{ + return __builtin_ia32_vcvtsh2ss_mask_round (__B, __A, + _mm_setzero_ps (), + (__mmask8) -1, __R); +} + +extern __inline __m128 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_cvt_roundsh_ss (__m128 __A, __mmask8 __B, __m128 __C, + __m128h __D, const int __R) +{ + return __builtin_ia32_vcvtsh2ss_mask_round (__D, __C, __A, __B, __R); +} + +extern __inline __m128 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_cvt_roundsh_ss (__mmask8 __A, __m128 __B, + __m128h __C, const int __R) +{ + return __builtin_ia32_vcvtsh2ss_mask_round (__C, __B, + _mm_setzero_ps (), + __A, __R); +} + +extern __inline __m128d +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundsh_sd (__m128d __A, __m128h __B, const int __R) +{ + return __builtin_ia32_vcvtsh2sd_mask_round (__B, __A, + _mm_setzero_pd (), + (__mmask8) -1, __R); +} + +extern __inline __m128d +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_cvt_roundsh_sd (__m128d __A, __mmask8 __B, __m128d __C, + __m128h __D, const int __R) +{ + return __builtin_ia32_vcvtsh2sd_mask_round (__D, __C, __A, __B, __R); +} + +extern __inline __m128d +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_cvt_roundsh_sd (__mmask8 __A, __m128d __B, __m128h __C, const int __R) +{ + return __builtin_ia32_vcvtsh2sd_mask_round (__C, __B, + _mm_setzero_pd (), + __A, __R); +} + +#else +#define _mm_cvt_roundsh_ss(A, B, R) \ + (__builtin_ia32_vcvtsh2ss_mask_round ((B), (A), \ + _mm_setzero_ps (), \ + (__mmask8) -1, (R))) + +#define _mm_mask_cvt_roundsh_ss(A, B, C, D, R) \ + (__builtin_ia32_vcvtsh2ss_mask_round ((D), (C), (A), (B), (R))) + +#define _mm_maskz_cvt_roundsh_ss(A, B, C, R) \ + (__builtin_ia32_vcvtsh2ss_mask_round ((C), (B), \ + _mm_setzero_ps (), \ + (A), (R))) + +#define _mm_cvt_roundsh_sd(A, B, R) \ + (__builtin_ia32_vcvtsh2sd_mask_round ((B), (A), \ + _mm_setzero_pd (), \ + (__mmask8) -1, (R))) + +#define _mm_mask_cvt_roundsh_sd(A, B, C, D, R) \ + (__builtin_ia32_vcvtsh2sd_mask_round ((D), (C), (A), (B), (R))) + +#define _mm_maskz_cvt_roundsh_sd(A, B, C, R) \ + (__builtin_ia32_vcvtsh2sd_mask_round ((C), (B), \ + _mm_setzero_pd (), \ + (A), (R))) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vcvtss2sh, vcvtsd2sh. */ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtss_sh (__m128h __A, __m128 __B) +{ + return __builtin_ia32_vcvtss2sh_mask_round (__B, __A, + _mm_setzero_ph (), + (__mmask8) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_cvtss_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128 __D) +{ + return __builtin_ia32_vcvtss2sh_mask_round (__D, __C, __A, __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_cvtss_sh (__mmask8 __A, __m128h __B, __m128 __C) +{ + return __builtin_ia32_vcvtss2sh_mask_round (__C, __B, + _mm_setzero_ph (), + __A, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsd_sh (__m128h __A, __m128d __B) +{ + return __builtin_ia32_vcvtsd2sh_mask_round (__B, __A, + _mm_setzero_ph (), + (__mmask8) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_cvtsd_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128d __D) +{ + return __builtin_ia32_vcvtsd2sh_mask_round (__D, __C, __A, __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_cvtsd_sh (__mmask8 __A, __m128h __B, __m128d __C) +{ + return __builtin_ia32_vcvtsd2sh_mask_round (__C, __B, + _mm_setzero_ph (), + __A, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundss_sh (__m128h __A, __m128 __B, const int __R) +{ + return __builtin_ia32_vcvtss2sh_mask_round (__B, __A, + _mm_setzero_ph (), + (__mmask8) -1, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_cvt_roundss_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128 __D, + const int __R) +{ + return __builtin_ia32_vcvtss2sh_mask_round (__D, __C, __A, __B, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_cvt_roundss_sh (__mmask8 __A, __m128h __B, __m128 __C, + const int __R) +{ + return __builtin_ia32_vcvtss2sh_mask_round (__C, __B, + _mm_setzero_ph (), + __A, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvt_roundsd_sh (__m128h __A, __m128d __B, const int __R) +{ + return __builtin_ia32_vcvtsd2sh_mask_round (__B, __A, + _mm_setzero_ph (), + (__mmask8) -1, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_cvt_roundsd_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128d __D, + const int __R) +{ + return __builtin_ia32_vcvtsd2sh_mask_round (__D, __C, __A, __B, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_cvt_roundsd_sh (__mmask8 __A, __m128h __B, __m128d __C, + const int __R) +{ + return __builtin_ia32_vcvtsd2sh_mask_round (__C, __B, + _mm_setzero_ph (), + __A, __R); +} + +#else +#define _mm_cvt_roundss_sh(A, B, R) \ + (__builtin_ia32_vcvtss2sh_mask_round ((B), (A), \ + _mm_setzero_ph (), \ + (__mmask8) -1, R)) + +#define _mm_mask_cvt_roundss_sh(A, B, C, D, R) \ + (__builtin_ia32_vcvtss2sh_mask_round ((D), (C), (A), (B), (R))) + +#define _mm_maskz_cvt_roundss_sh(A, B, C, R) \ + (__builtin_ia32_vcvtss2sh_mask_round ((C), (B), \ + _mm_setzero_ph (), \ + A, R)) + +#define _mm_cvt_roundsd_sh(A, B, R) \ + (__builtin_ia32_vcvtsd2sh_mask_round ((B), (A), \ + _mm_setzero_ph (), \ + (__mmask8) -1, R)) + +#define _mm_mask_cvt_roundsd_sh(A, B, C, D, R) \ + (__builtin_ia32_vcvtsd2sh_mask_round ((D), (C), (A), (B), (R))) + +#define _mm_maskz_cvt_roundsd_sh(A, B, C, R) \ + (__builtin_ia32_vcvtsd2sh_mask_round ((C), (B), \ + _mm_setzero_ph (), \ + (A), (R))) + +#endif /* __OPTIMIZE__ */ + +extern __inline _Float16 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_cvtsh_h (__m128h __A) +{ + return __A[0]; +} + +/* Intrinsics vfmadd[132,213,231]sh. */ +extern __inline __m128h + __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fmadd_sh (__m128h __W, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fmadd_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fmadd_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask3 ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fmadd_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + + +#ifdef __OPTIMIZE__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) -1, + __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fmadd_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B, + const int __R) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U, + const int __R) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask3 ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fmadd_round_sh (__mmask8 __U, __m128h __W, __m128h __A, + __m128h __B, const int __R) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, __R); +} + +#else +#define _mm_fmadd_round_sh(A, B, C, R) \ + ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), (C), (-1), (R))) +#define _mm_mask_fmadd_round_sh(A, U, B, C, R) \ + ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), (C), (U), (R))) +#define _mm_mask3_fmadd_round_sh(A, B, C, U, R) \ + ((__m128h) __builtin_ia32_vfmaddsh3_mask3 ((A), (B), (C), (U), (R))) +#define _mm_maskz_fmadd_round_sh(U, A, B, C, R) \ + ((__m128h) __builtin_ia32_vfmaddsh3_maskz ((A), (B), (C), (U), (R))) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vfnmadd[132,213,231]sh. */ +extern __inline __m128h + __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fnmadd_sh (__m128h __W, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fnmadd_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fnmadd_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U) +{ + return (__m128h) __builtin_ia32_vfnmaddsh3_mask3 ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fnmadd_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfnmaddsh3_maskz ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + + +#ifdef __OPTIMIZE__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fnmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R) +{ + return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) -1, + __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fnmadd_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B, + const int __R) +{ + return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fnmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U, + const int __R) +{ + return (__m128h) __builtin_ia32_vfnmaddsh3_mask3 ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fnmadd_round_sh (__mmask8 __U, __m128h __W, __m128h __A, + __m128h __B, const int __R) +{ + return (__m128h) __builtin_ia32_vfnmaddsh3_maskz ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, __R); +} + +#else +#define _mm_fnmadd_round_sh(A, B, C, R) \ + ((__m128h) __builtin_ia32_vfnmaddsh3_mask ((A), (B), (C), (-1), (R))) +#define _mm_mask_fnmadd_round_sh(A, U, B, C, R) \ + ((__m128h) __builtin_ia32_vfnmaddsh3_mask ((A), (B), (C), (U), (R))) +#define _mm_mask3_fnmadd_round_sh(A, B, C, U, R) \ + ((__m128h) __builtin_ia32_vfnmaddsh3_mask3 ((A), (B), (C), (U), (R))) +#define _mm_maskz_fnmadd_round_sh(U, A, B, C, R) \ + ((__m128h) __builtin_ia32_vfnmaddsh3_maskz ((A), (B), (C), (U), (R))) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vfmsub[132,213,231]sh. */ +extern __inline __m128h + __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fmsub_sh (__m128h __W, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + -(__v8hf) __B, + (__mmask8) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fmsub_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + -(__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fmsub_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U) +{ + return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fmsub_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, + (__v8hf) __A, + -(__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + + +#ifdef __OPTIMIZE__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + -(__v8hf) __B, + (__mmask8) -1, + __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fmsub_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B, + const int __R) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + (__v8hf) __A, + -(__v8hf) __B, + (__mmask8) __U, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U, + const int __R) +{ + return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W, + (__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fmsub_round_sh (__mmask8 __U, __m128h __W, __m128h __A, + __m128h __B, const int __R) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, + (__v8hf) __A, + -(__v8hf) __B, + (__mmask8) __U, __R); +} + +#else +#define _mm_fmsub_round_sh(A, B, C, R) \ + ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), -(C), (-1), (R))) +#define _mm_mask_fmsub_round_sh(A, U, B, C, R) \ + ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), -(C), (U), (R))) +#define _mm_mask3_fmsub_round_sh(A, B, C, U, R) \ + ((__m128h) __builtin_ia32_vfmsubsh3_mask3 ((A), (B), (C), (U), (R))) +#define _mm_maskz_fmsub_round_sh(U, A, B, C, R) \ + ((__m128h) __builtin_ia32_vfmaddsh3_maskz ((A), (B), -(C), (U), (R))) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vfnmsub[132,213,231]sh. */ +extern __inline __m128h + __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fnmsub_sh (__m128h __W, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + -(__v8hf) __A, + -(__v8hf) __B, + (__mmask8) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fnmsub_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + -(__v8hf) __A, + -(__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fnmsub_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U) +{ + return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W, + -(__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fnmsub_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, + -(__v8hf) __A, + -(__v8hf) __B, + (__mmask8) __U, + _MM_FROUND_CUR_DIRECTION); +} + + +#ifdef __OPTIMIZE__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fnmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + -(__v8hf) __A, + -(__v8hf) __B, + (__mmask8) -1, + __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fnmsub_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B, + const int __R) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, + -(__v8hf) __A, + -(__v8hf) __B, + (__mmask8) __U, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fnmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U, + const int __R) +{ + return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W, + -(__v8hf) __A, + (__v8hf) __B, + (__mmask8) __U, __R); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fnmsub_round_sh (__mmask8 __U, __m128h __W, __m128h __A, + __m128h __B, const int __R) +{ + return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, + -(__v8hf) __A, + -(__v8hf) __B, + (__mmask8) __U, __R); +} + +#else +#define _mm_fnmsub_round_sh(A, B, C, R) \ + ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), -(B), -(C), (-1), (R))) +#define _mm_mask_fnmsub_round_sh(A, U, B, C, R) \ + ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), -(B), -(C), (U), (R))) +#define _mm_mask3_fnmsub_round_sh(A, B, C, U, R) \ + ((__m128h) __builtin_ia32_vfmsubsh3_mask3 ((A), -(B), (C), (U), (R))) +#define _mm_maskz_fnmsub_round_sh(U, A, B, C, R) \ + ((__m128h) __builtin_ia32_vfmaddsh3_maskz ((A), -(B), -(C), (U), (R))) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vf[,c]maddcsh. */ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fcmadd_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D) +{ + return (__m128h) + __builtin_ia32_vfcmaddcsh_mask_round ((__v8hf) __A, + (__v8hf) __C, + (__v8hf) __D, __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fcmadd_sch (__m128h __A, __m128h __B, __m128h __C, __mmask8 __D) +{ + return (__m128h) + __builtin_ia32_vfcmaddcsh_mask3_round ((__v8hf) __A, + (__v8hf) __B, + (__v8hf) __C, __D, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fcmadd_sch (__mmask8 __A, __m128h __B, __m128h __C, __m128h __D) +{ + return (__m128h) + __builtin_ia32_vfcmaddcsh_maskz_round ((__v8hf) __B, + (__v8hf) __C, + (__v8hf) __D, + __A, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fcmadd_sch (__m128h __A, __m128h __B, __m128h __C) +{ + return (__m128h) + __builtin_ia32_vfcmaddcsh_round ((__v8hf) __A, + (__v8hf) __B, + (__v8hf) __C, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fmadd_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D) +{ + return (__m128h) + __builtin_ia32_vfmaddcsh_mask_round ((__v8hf) __A, + (__v8hf) __C, + (__v8hf) __D, __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fmadd_sch (__m128h __A, __m128h __B, __m128h __C, __mmask8 __D) +{ + return (__m128h) + __builtin_ia32_vfmaddcsh_mask3_round ((__v8hf) __A, + (__v8hf) __B, + (__v8hf) __C, __D, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fmadd_sch (__mmask8 __A, __m128h __B, __m128h __C, __m128h __D) +{ + return (__m128h) + __builtin_ia32_vfmaddcsh_maskz_round ((__v8hf) __B, + (__v8hf) __C, + (__v8hf) __D, + __A, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fmadd_sch (__m128h __A, __m128h __B, __m128h __C) +{ + return (__m128h) + __builtin_ia32_vfmaddcsh_round ((__v8hf) __A, + (__v8hf) __B, + (__v8hf) __C, + _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fcmadd_round_sch (__m128h __A, __mmask8 __B, __m128h __C, + __m128h __D, const int __E) +{ + return (__m128h) + __builtin_ia32_vfcmaddcsh_mask_round ((__v8hf) __A, + (__v8hf) __C, + (__v8hf) __D, + __B, __E); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fcmadd_round_sch (__m128h __A, __m128h __B, __m128h __C, + __mmask8 __D, const int __E) +{ + return (__m128h) + __builtin_ia32_vfcmaddcsh_mask3_round ((__v8hf) __A, + (__v8hf) __B, + (__v8hf) __C, + __D, __E); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fcmadd_round_sch (__mmask8 __A, __m128h __B, __m128h __C, + __m128h __D, const int __E) +{ + return (__m128h) + __builtin_ia32_vfcmaddcsh_maskz_round ((__v8hf) __B, + (__v8hf) __C, + (__v8hf) __D, + __A, __E); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fcmadd_round_sch (__m128h __A, __m128h __B, __m128h __C, const int __D) +{ + return (__m128h) + __builtin_ia32_vfcmaddcsh_round ((__v8hf) __A, + (__v8hf) __B, + (__v8hf) __C, + __D); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fmadd_round_sch (__m128h __A, __mmask8 __B, __m128h __C, + __m128h __D, const int __E) +{ + return (__m128h) + __builtin_ia32_vfmaddcsh_mask_round ((__v8hf) __A, + (__v8hf) __C, + (__v8hf) __D, + __B, __E); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask3_fmadd_round_sch (__m128h __A, __m128h __B, __m128h __C, + __mmask8 __D, const int __E) +{ + return (__m128h) + __builtin_ia32_vfmaddcsh_mask3_round ((__v8hf) __A, + (__v8hf) __B, + (__v8hf) __C, + __D, __E); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fmadd_round_sch (__mmask8 __A, __m128h __B, __m128h __C, + __m128h __D, const int __E) +{ + return (__m128h) + __builtin_ia32_vfmaddcsh_maskz_round ((__v8hf) __B, + (__v8hf) __C, + (__v8hf) __D, + __A, __E); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fmadd_round_sch (__m128h __A, __m128h __B, __m128h __C, const int __D) +{ + return (__m128h) + __builtin_ia32_vfmaddcsh_round ((__v8hf) __A, + (__v8hf) __B, + (__v8hf) __C, + __D); +} +#else +#define _mm_mask_fcmadd_round_sch(A, B, C, D, E) \ + ((__m128h) \ + __builtin_ia32_vfcmaddcsh_mask_round ((__v8hf) (A), \ + (__v8hf) (C), \ + (__v8hf) (D), \ + (B), (E))) + + +#define _mm_mask3_fcmadd_round_sch(A, B, C, D, E) \ + ((__m128h) \ + __builtin_ia32_vfcmaddcsh_mask3_round ((__v8hf) (A), \ + (__v8hf) (B), \ + (__v8hf) (C), \ + (D), (E))) + +#define _mm_maskz_fcmadd_round_sch(A, B, C, D, E) \ + __builtin_ia32_vfcmaddcsh_maskz_round ((B), (C), (D), (A), (E)) + +#define _mm_fcmadd_round_sch(A, B, C, D) \ + __builtin_ia32_vfcmaddcsh_round ((A), (B), (C), (D)) + +#define _mm_mask_fmadd_round_sch(A, B, C, D, E) \ + ((__m128h) \ + __builtin_ia32_vfmaddcsh_mask_round ((__v8hf) (A), \ + (__v8hf) (C), \ + (__v8hf) (D), \ + (B), (E))) + +#define _mm_mask3_fmadd_round_sch(A, B, C, D, E) \ + ((__m128h) \ + __builtin_ia32_vfmaddcsh_mask3_round ((__v8hf) (A), \ + (__v8hf) (B), \ + (__v8hf) (C), \ + (D), (E))) + +#define _mm_maskz_fmadd_round_sch(A, B, C, D, E) \ + __builtin_ia32_vfmaddcsh_maskz_round ((B), (C), (D), (A), (E)) + +#define _mm_fmadd_round_sch(A, B, C, D) \ + __builtin_ia32_vfmaddcsh_round ((A), (B), (C), (D)) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vf[,c]mulcsh. */ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fcmul_sch (__m128h __A, __m128h __B) +{ + return (__m128h) + __builtin_ia32_vfcmulcsh_round ((__v8hf) __A, + (__v8hf) __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fcmul_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D) +{ + return (__m128h) + __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __C, + (__v8hf) __D, + (__v8hf) __A, + __B, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fcmul_sch (__mmask8 __A, __m128h __B, __m128h __C) +{ + return (__m128h) + __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __B, + (__v8hf) __C, + _mm_setzero_ph (), + __A, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fmul_sch (__m128h __A, __m128h __B) +{ + return (__m128h) + __builtin_ia32_vfmulcsh_round ((__v8hf) __A, + (__v8hf) __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fmul_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D) +{ + return (__m128h) + __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __C, + (__v8hf) __D, + (__v8hf) __A, + __B, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fmul_sch (__mmask8 __A, __m128h __B, __m128h __C) +{ + return (__m128h) + __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __B, + (__v8hf) __C, + _mm_setzero_ph (), + __A, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fcmul_round_sch (__m128h __A, __m128h __B, const int __D) +{ + return (__m128h) + __builtin_ia32_vfcmulcsh_round ((__v8hf) __A, + (__v8hf) __B, + __D); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fcmul_round_sch (__m128h __A, __mmask8 __B, __m128h __C, + __m128h __D, const int __E) +{ + return (__m128h) + __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __C, + (__v8hf) __D, + (__v8hf) __A, + __B, __E); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fcmul_round_sch (__mmask8 __A, __m128h __B, __m128h __C, + const int __E) +{ + return (__m128h) + __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __B, + (__v8hf) __C, + _mm_setzero_ph (), + __A, __E); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_fmul_round_sch (__m128h __A, __m128h __B, const int __D) +{ + return (__m128h) + __builtin_ia32_vfmulcsh_round ((__v8hf) __A, + (__v8hf) __B, __D); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_mask_fmul_round_sch (__m128h __A, __mmask8 __B, __m128h __C, + __m128h __D, const int __E) +{ + return (__m128h) + __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __C, + (__v8hf) __D, + (__v8hf) __A, + __B, __E); +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm_maskz_fmul_round_sch (__mmask8 __A, __m128h __B, __m128h __C, const int __E) +{ + return (__m128h) + __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __B, + (__v8hf) __C, + _mm_setzero_ph (), + __A, __E); +} + +#else +#define _mm_fcmul_round_sch(__A, __B, __D) \ + (__m128h) __builtin_ia32_vfcmulcsh_round ((__v8hf) __A, \ + (__v8hf) __B, __D) + +#define _mm_mask_fcmul_round_sch(__A, __B, __C, __D, __E) \ + (__m128h) __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __C, \ + (__v8hf) __D, \ + (__v8hf) __A, \ + __B, __E) + +#define _mm_maskz_fcmul_round_sch(__A, __B, __C, __E) \ + (__m128h) __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __B, \ + (__v8hf) __C, \ + _mm_setzero_ph (), \ + __A, __E) + +#define _mm_fmul_round_sch(__A, __B, __D) \ + (__m128h) __builtin_ia32_vfmulcsh_round ((__v8hf) __A, \ + (__v8hf) __B, __D) + +#define _mm_mask_fmul_round_sch(__A, __B, __C, __D, __E) \ + (__m128h) __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __C, \ + (__v8hf) __D, \ + (__v8hf) __A, \ + __B, __E) + +#define _mm_maskz_fmul_round_sch(__A, __B, __C, __E) \ + (__m128h) __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __B, \ + (__v8hf) __C, \ + _mm_setzero_ph (), \ + __A, __E) + +#endif /* __OPTIMIZE__ */ + +#define _mm_mul_sch(A, B) _mm_fmul_sch ((A), (B)) +#define _mm_mask_mul_sch(W, U, A, B) _mm_mask_fmul_sch ((W), (U), (A), (B)) +#define _mm_maskz_mul_sch(U, A, B) _mm_maskz_fmul_sch ((U), (A), (B)) +#define _mm_mul_round_sch(A, B, R) _mm_fmul_round_sch ((A), (B), (R)) +#define _mm_mask_mul_round_sch(W, U, A, B, R) \ + _mm_mask_fmul_round_sch ((W), (U), (A), (B), (R)) +#define _mm_maskz_mul_round_sch(U, A, B, R) \ + _mm_maskz_fmul_round_sch ((U), (A), (B), (R)) + +#define _mm_cmul_sch(A, B) _mm_fcmul_sch ((A), (B)) +#define _mm_mask_cmul_sch(W, U, A, B) _mm_mask_fcmul_sch ((W), (U), (A), (B)) +#define _mm_maskz_cmul_sch(U, A, B) _mm_maskz_fcmul_sch ((U), (A), (B)) +#define _mm_cmul_round_sch(A, B, R) _mm_fcmul_round_sch ((A), (B), (R)) +#define _mm_mask_cmul_round_sch(W, U, A, B, R) \ + _mm_mask_fcmul_round_sch ((W), (U), (A), (B), (R)) +#define _mm_maskz_cmul_round_sch(U, A, B, R) \ + _mm_maskz_fcmul_round_sch ((U), (A), (B), (R)) + +#ifdef __DISABLE_AVX512FP16__ +#undef __DISABLE_AVX512FP16__ +#pragma GCC pop_options +#endif /* __DISABLE_AVX512FP16__ */ + +#if !defined (__AVX512FP16__) || !defined (__EVEX512__) +#pragma GCC push_options +#pragma GCC target("avx512fp16,evex512") +#define __DISABLE_AVX512FP16_512__ +#endif /* __AVX512FP16_512__ */ + +typedef _Float16 __v32hf __attribute__ ((__vector_size__ (64))); +typedef _Float16 __m512h __attribute__ ((__vector_size__ (64), __may_alias__)); +typedef _Float16 __m512h_u __attribute__ ((__vector_size__ (64), \ + __may_alias__, __aligned__ (1))); + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_set_ph (_Float16 __A31, _Float16 __A30, _Float16 __A29, + _Float16 __A28, _Float16 __A27, _Float16 __A26, + _Float16 __A25, _Float16 __A24, _Float16 __A23, + _Float16 __A22, _Float16 __A21, _Float16 __A20, + _Float16 __A19, _Float16 __A18, _Float16 __A17, + _Float16 __A16, _Float16 __A15, _Float16 __A14, + _Float16 __A13, _Float16 __A12, _Float16 __A11, + _Float16 __A10, _Float16 __A9, _Float16 __A8, + _Float16 __A7, _Float16 __A6, _Float16 __A5, + _Float16 __A4, _Float16 __A3, _Float16 __A2, + _Float16 __A1, _Float16 __A0) +{ + return __extension__ (__m512h)(__v32hf){ __A0, __A1, __A2, __A3, + __A4, __A5, __A6, __A7, + __A8, __A9, __A10, __A11, + __A12, __A13, __A14, __A15, + __A16, __A17, __A18, __A19, + __A20, __A21, __A22, __A23, + __A24, __A25, __A26, __A27, + __A28, __A29, __A30, __A31 }; +} + +/* Create vectors of elements in the reversed order from + _mm512_set_ph functions. */ + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_setr_ph (_Float16 __A0, _Float16 __A1, _Float16 __A2, + _Float16 __A3, _Float16 __A4, _Float16 __A5, + _Float16 __A6, _Float16 __A7, _Float16 __A8, + _Float16 __A9, _Float16 __A10, _Float16 __A11, + _Float16 __A12, _Float16 __A13, _Float16 __A14, + _Float16 __A15, _Float16 __A16, _Float16 __A17, + _Float16 __A18, _Float16 __A19, _Float16 __A20, + _Float16 __A21, _Float16 __A22, _Float16 __A23, + _Float16 __A24, _Float16 __A25, _Float16 __A26, + _Float16 __A27, _Float16 __A28, _Float16 __A29, + _Float16 __A30, _Float16 __A31) + +{ + return _mm512_set_ph (__A31, __A30, __A29, __A28, __A27, __A26, __A25, + __A24, __A23, __A22, __A21, __A20, __A19, __A18, + __A17, __A16, __A15, __A14, __A13, __A12, __A11, + __A10, __A9, __A8, __A7, __A6, __A5, __A4, __A3, + __A2, __A1, __A0); +} + +/* Broadcast _Float16 to vector. */ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_set1_ph (_Float16 __A) +{ + return _mm512_set_ph (__A, __A, __A, __A, __A, __A, __A, __A, + __A, __A, __A, __A, __A, __A, __A, __A, + __A, __A, __A, __A, __A, __A, __A, __A, + __A, __A, __A, __A, __A, __A, __A, __A); +} + +/* Create a vector with all zeros. */ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_setzero_ph (void) +{ + return _mm512_set1_ph (0.0f16); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_undefined_ph (void) +{ +#pragma GCC diagnostic push +#pragma GCC diagnostic ignored "-Winit-self" + __m512h __Y = __Y; +#pragma GCC diagnostic pop + return __Y; +} + +extern __inline _Float16 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_cvtsh_h (__m512h __A) +{ + return __A[0]; +} + +extern __inline __m512 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_castph_ps (__m512h __a) +{ + return (__m512) __a; +} + +extern __inline __m512d +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_castph_pd (__m512h __a) +{ + return (__m512d) __a; +} + +extern __inline __m512i +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_castph_si512 (__m512h __a) +{ + return (__m512i) __a; +} + +extern __inline __m128h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_castph512_ph128 (__m512h __A) +{ + union + { + __m128h __a[4]; + __m512h __v; + } __u = { .__v = __A }; + return __u.__a[0]; +} + +extern __inline __m256h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_castph512_ph256 (__m512h __A) +{ + union + { + __m256h __a[2]; + __m512h __v; + } __u = { .__v = __A }; + return __u.__a[0]; +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_castph128_ph512 (__m128h __A) +{ + union + { + __m128h __a[4]; + __m512h __v; + } __u; + __u.__a[0] = __A; + return __u.__v; +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_castph256_ph512 (__m256h __A) +{ + union + { + __m256h __a[2]; + __m512h __v; + } __u; + __u.__a[0] = __A; + return __u.__v; +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_zextph128_ph512 (__m128h __A) +{ + return (__m512h) _mm512_insertf32x4 (_mm512_setzero_ps (), + (__m128) __A, 0); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_zextph256_ph512 (__m256h __A) +{ + return (__m512h) _mm512_insertf64x4 (_mm512_setzero_pd (), + (__m256d) __A, 0); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_castps_ph (__m512 __a) +{ + return (__m512h) __a; +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_castpd_ph (__m512d __a) +{ + return (__m512h) __a; +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_castsi512_ph (__m512i __a) +{ + return (__m512h) __a; +} + +/* Create a vector with element 0 as *P and the rest zero. */ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_load_ph (void const *__P) +{ + return *(const __m512h *) __P; +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_loadu_ph (void const *__P) +{ + return *(const __m512h_u *) __P; +} + +/* Stores the lower _Float16 value. */ +extern __inline void +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_store_ph (void *__P, __m512h __A) +{ + *(__m512h *) __P = __A; +} + +extern __inline void +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_storeu_ph (void *__P, __m512h __A) +{ + *(__m512h_u *) __P = __A; +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_abs_ph (__m512h __A) +{ + return (__m512h) _mm512_and_epi32 ( _mm512_set1_epi32 (0x7FFF7FFF), + (__m512i) __A); +} + +/* Intrinsics v[add,sub,mul,div]ph. */ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_add_ph (__m512h __A, __m512h __B) +{ + return (__m512h) ((__v32hf) __A + (__v32hf) __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_add_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) +{ + return __builtin_ia32_addph512_mask (__C, __D, __A, __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_add_ph (__mmask32 __A, __m512h __B, __m512h __C) +{ + return __builtin_ia32_addph512_mask (__B, __C, + _mm512_setzero_ph (), __A); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_sub_ph (__m512h __A, __m512h __B) +{ + return (__m512h) ((__v32hf) __A - (__v32hf) __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_sub_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) +{ + return __builtin_ia32_subph512_mask (__C, __D, __A, __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_sub_ph (__mmask32 __A, __m512h __B, __m512h __C) +{ + return __builtin_ia32_subph512_mask (__B, __C, + _mm512_setzero_ph (), __A); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mul_ph (__m512h __A, __m512h __B) +{ + return (__m512h) ((__v32hf) __A * (__v32hf) __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_mul_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) +{ + return __builtin_ia32_mulph512_mask (__C, __D, __A, __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_mul_ph (__mmask32 __A, __m512h __B, __m512h __C) +{ + return __builtin_ia32_mulph512_mask (__B, __C, + _mm512_setzero_ph (), __A); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_div_ph (__m512h __A, __m512h __B) +{ + return (__m512h) ((__v32hf) __A / (__v32hf) __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_div_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) +{ + return __builtin_ia32_divph512_mask (__C, __D, __A, __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_div_ph (__mmask32 __A, __m512h __B, __m512h __C) +{ + return __builtin_ia32_divph512_mask (__B, __C, + _mm512_setzero_ph (), __A); +} + +#ifdef __OPTIMIZE__ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_add_round_ph (__m512h __A, __m512h __B, const int __C) +{ + return __builtin_ia32_addph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, __C); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_add_round_ph (__m512h __A, __mmask32 __B, __m512h __C, + __m512h __D, const int __E) +{ + return __builtin_ia32_addph512_mask_round (__C, __D, __A, __B, __E); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_add_round_ph (__mmask32 __A, __m512h __B, __m512h __C, + const int __D) +{ + return __builtin_ia32_addph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, __D); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_sub_round_ph (__m512h __A, __m512h __B, const int __C) +{ + return __builtin_ia32_subph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, __C); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_sub_round_ph (__m512h __A, __mmask32 __B, __m512h __C, + __m512h __D, const int __E) +{ + return __builtin_ia32_subph512_mask_round (__C, __D, __A, __B, __E); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_sub_round_ph (__mmask32 __A, __m512h __B, __m512h __C, + const int __D) +{ + return __builtin_ia32_subph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, __D); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mul_round_ph (__m512h __A, __m512h __B, const int __C) +{ + return __builtin_ia32_mulph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, __C); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_mul_round_ph (__m512h __A, __mmask32 __B, __m512h __C, + __m512h __D, const int __E) +{ + return __builtin_ia32_mulph512_mask_round (__C, __D, __A, __B, __E); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_mul_round_ph (__mmask32 __A, __m512h __B, __m512h __C, + const int __D) +{ + return __builtin_ia32_mulph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, __D); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_div_round_ph (__m512h __A, __m512h __B, const int __C) +{ + return __builtin_ia32_divph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, __C); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_div_round_ph (__m512h __A, __mmask32 __B, __m512h __C, + __m512h __D, const int __E) +{ + return __builtin_ia32_divph512_mask_round (__C, __D, __A, __B, __E); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_div_round_ph (__mmask32 __A, __m512h __B, __m512h __C, + const int __D) +{ + return __builtin_ia32_divph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, __D); +} +#else +#define _mm512_add_round_ph(A, B, C) \ + ((__m512h)__builtin_ia32_addph512_mask_round((A), (B), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, (C))) + +#define _mm512_mask_add_round_ph(A, B, C, D, E) \ + ((__m512h)__builtin_ia32_addph512_mask_round((C), (D), (A), (B), (E))) + +#define _mm512_maskz_add_round_ph(A, B, C, D) \ + ((__m512h)__builtin_ia32_addph512_mask_round((B), (C), \ + _mm512_setzero_ph (), \ + (A), (D))) + +#define _mm512_sub_round_ph(A, B, C) \ + ((__m512h)__builtin_ia32_subph512_mask_round((A), (B), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, (C))) + +#define _mm512_mask_sub_round_ph(A, B, C, D, E) \ + ((__m512h)__builtin_ia32_subph512_mask_round((C), (D), (A), (B), (E))) + +#define _mm512_maskz_sub_round_ph(A, B, C, D) \ + ((__m512h)__builtin_ia32_subph512_mask_round((B), (C), \ + _mm512_setzero_ph (), \ + (A), (D))) + +#define _mm512_mul_round_ph(A, B, C) \ + ((__m512h)__builtin_ia32_mulph512_mask_round((A), (B), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, (C))) + +#define _mm512_mask_mul_round_ph(A, B, C, D, E) \ + ((__m512h)__builtin_ia32_mulph512_mask_round((C), (D), (A), (B), (E))) + +#define _mm512_maskz_mul_round_ph(A, B, C, D) \ + ((__m512h)__builtin_ia32_mulph512_mask_round((B), (C), \ + _mm512_setzero_ph (), \ + (A), (D))) + +#define _mm512_div_round_ph(A, B, C) \ + ((__m512h)__builtin_ia32_divph512_mask_round((A), (B), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, (C))) + +#define _mm512_mask_div_round_ph(A, B, C, D, E) \ + ((__m512h)__builtin_ia32_divph512_mask_round((C), (D), (A), (B), (E))) + +#define _mm512_maskz_div_round_ph(A, B, C, D) \ + ((__m512h)__builtin_ia32_divph512_mask_round((B), (C), \ + _mm512_setzero_ph (), \ + (A), (D))) +#endif /* __OPTIMIZE__ */ + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_conj_pch (__m512h __A) +{ + return (__m512h) _mm512_xor_epi32 ((__m512i) __A, _mm512_set1_epi32 (1<<31)); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_conj_pch (__m512h __W, __mmask16 __U, __m512h __A) +{ + return (__m512h) + __builtin_ia32_movaps512_mask ((__v16sf) _mm512_conj_pch (__A), + (__v16sf) __W, + (__mmask16) __U); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_conj_pch (__mmask16 __U, __m512h __A) +{ + return (__m512h) + __builtin_ia32_movaps512_mask ((__v16sf) _mm512_conj_pch (__A), + (__v16sf) _mm512_setzero_ps (), + (__mmask16) __U); +} + +/* Intrinsic vmaxph vminph. */ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_max_ph (__m512h __A, __m512h __B) +{ + return __builtin_ia32_maxph512_mask (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_max_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) +{ + return __builtin_ia32_maxph512_mask (__C, __D, __A, __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_max_ph (__mmask32 __A, __m512h __B, __m512h __C) +{ + return __builtin_ia32_maxph512_mask (__B, __C, + _mm512_setzero_ph (), __A); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_min_ph (__m512h __A, __m512h __B) +{ + return __builtin_ia32_minph512_mask (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_min_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) +{ + return __builtin_ia32_minph512_mask (__C, __D, __A, __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_min_ph (__mmask32 __A, __m512h __B, __m512h __C) +{ + return __builtin_ia32_minph512_mask (__B, __C, + _mm512_setzero_ph (), __A); +} + +#ifdef __OPTIMIZE__ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_max_round_ph (__m512h __A, __m512h __B, const int __C) +{ + return __builtin_ia32_maxph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, __C); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_max_round_ph (__m512h __A, __mmask32 __B, __m512h __C, + __m512h __D, const int __E) +{ + return __builtin_ia32_maxph512_mask_round (__C, __D, __A, __B, __E); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_max_round_ph (__mmask32 __A, __m512h __B, __m512h __C, + const int __D) +{ + return __builtin_ia32_maxph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, __D); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_min_round_ph (__m512h __A, __m512h __B, const int __C) +{ + return __builtin_ia32_minph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, __C); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_min_round_ph (__m512h __A, __mmask32 __B, __m512h __C, + __m512h __D, const int __E) +{ + return __builtin_ia32_minph512_mask_round (__C, __D, __A, __B, __E); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_min_round_ph (__mmask32 __A, __m512h __B, __m512h __C, + const int __D) +{ + return __builtin_ia32_minph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, __D); +} + +#else +#define _mm512_max_round_ph(A, B, C) \ + (__builtin_ia32_maxph512_mask_round ((A), (B), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, (C))) + +#define _mm512_mask_max_round_ph(A, B, C, D, E) \ + (__builtin_ia32_maxph512_mask_round ((C), (D), (A), (B), (E))) + +#define _mm512_maskz_max_round_ph(A, B, C, D) \ + (__builtin_ia32_maxph512_mask_round ((B), (C), \ + _mm512_setzero_ph (), \ + (A), (D))) + +#define _mm512_min_round_ph(A, B, C) \ + (__builtin_ia32_minph512_mask_round ((A), (B), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, (C))) + +#define _mm512_mask_min_round_ph(A, B, C, D, E) \ + (__builtin_ia32_minph512_mask_round ((C), (D), (A), (B), (E))) + +#define _mm512_maskz_min_round_ph(A, B, C, D) \ + (__builtin_ia32_minph512_mask_round ((B), (C), \ + _mm512_setzero_ph (), \ + (A), (D))) +#endif /* __OPTIMIZE__ */ + +/* vcmpph */ +#ifdef __OPTIMIZE +extern __inline __mmask32 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_cmp_ph_mask (__m512h __A, __m512h __B, const int __C) +{ + return (__mmask32) __builtin_ia32_cmpph512_mask (__A, __B, __C, + (__mmask32) -1); +} + +extern __inline __mmask32 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_cmp_ph_mask (__mmask32 __A, __m512h __B, __m512h __C, + const int __D) +{ + return (__mmask32) __builtin_ia32_cmpph512_mask (__B, __C, __D, + __A); +} + +extern __inline __mmask32 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_cmp_round_ph_mask (__m512h __A, __m512h __B, const int __C, + const int __D) +{ + return (__mmask32) __builtin_ia32_cmpph512_mask_round (__A, __B, + __C, (__mmask32) -1, + __D); +} + +extern __inline __mmask32 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_cmp_round_ph_mask (__mmask32 __A, __m512h __B, __m512h __C, + const int __D, const int __E) +{ + return (__mmask32) __builtin_ia32_cmpph512_mask_round (__B, __C, + __D, __A, + __E); +} + +#else +#define _mm512_cmp_ph_mask(A, B, C) \ + (__builtin_ia32_cmpph512_mask ((A), (B), (C), (-1))) + +#define _mm512_mask_cmp_ph_mask(A, B, C, D) \ + (__builtin_ia32_cmpph512_mask ((B), (C), (D), (A))) + +#define _mm512_cmp_round_ph_mask(A, B, C, D) \ + (__builtin_ia32_cmpph512_mask_round ((A), (B), (C), (-1), (D))) + +#define _mm512_mask_cmp_round_ph_mask(A, B, C, D, E) \ + (__builtin_ia32_cmpph512_mask_round ((B), (C), (D), (A), (E))) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vsqrtph. */ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_sqrt_ph (__m512h __A) +{ + return __builtin_ia32_sqrtph512_mask_round (__A, + _mm512_setzero_ph(), + (__mmask32) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_sqrt_ph (__m512h __A, __mmask32 __B, __m512h __C) +{ + return __builtin_ia32_sqrtph512_mask_round (__C, __A, __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_sqrt_ph (__mmask32 __A, __m512h __B) +{ + return __builtin_ia32_sqrtph512_mask_round (__B, + _mm512_setzero_ph (), + __A, + _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_sqrt_round_ph (__m512h __A, const int __B) +{ + return __builtin_ia32_sqrtph512_mask_round (__A, + _mm512_setzero_ph(), + (__mmask32) -1, __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_sqrt_round_ph (__m512h __A, __mmask32 __B, __m512h __C, + const int __D) +{ + return __builtin_ia32_sqrtph512_mask_round (__C, __A, __B, __D); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_sqrt_round_ph (__mmask32 __A, __m512h __B, const int __C) +{ + return __builtin_ia32_sqrtph512_mask_round (__B, + _mm512_setzero_ph (), + __A, __C); +} + +#else +#define _mm512_sqrt_round_ph(A, B) \ + (__builtin_ia32_sqrtph512_mask_round ((A), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, (B))) + +#define _mm512_mask_sqrt_round_ph(A, B, C, D) \ + (__builtin_ia32_sqrtph512_mask_round ((C), (A), (B), (D))) + +#define _mm512_maskz_sqrt_round_ph(A, B, C) \ + (__builtin_ia32_sqrtph512_mask_round ((B), \ + _mm512_setzero_ph (), \ + (A), (C))) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vrsqrtph. */ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_rsqrt_ph (__m512h __A) +{ + return __builtin_ia32_rsqrtph512_mask (__A, _mm512_setzero_ph (), + (__mmask32) -1); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_rsqrt_ph (__m512h __A, __mmask32 __B, __m512h __C) +{ + return __builtin_ia32_rsqrtph512_mask (__C, __A, __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_rsqrt_ph (__mmask32 __A, __m512h __B) +{ + return __builtin_ia32_rsqrtph512_mask (__B, _mm512_setzero_ph (), + __A); +} + +/* Intrinsics vrcpph. */ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_rcp_ph (__m512h __A) +{ + return __builtin_ia32_rcpph512_mask (__A, _mm512_setzero_ph (), + (__mmask32) -1); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_rcp_ph (__m512h __A, __mmask32 __B, __m512h __C) +{ + return __builtin_ia32_rcpph512_mask (__C, __A, __B); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_rcp_ph (__mmask32 __A, __m512h __B) +{ + return __builtin_ia32_rcpph512_mask (__B, _mm512_setzero_ph (), + __A); +} + +/* Intrinsics vscalefph. */ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_scalef_ph (__m512h __A, __m512h __B) +{ + return __builtin_ia32_scalefph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_scalef_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D) +{ + return __builtin_ia32_scalefph512_mask_round (__C, __D, __A, __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_scalef_ph (__mmask32 __A, __m512h __B, __m512h __C) +{ + return __builtin_ia32_scalefph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, + _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_scalef_round_ph (__m512h __A, __m512h __B, const int __C) +{ + return __builtin_ia32_scalefph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, __C); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_scalef_round_ph (__m512h __A, __mmask32 __B, __m512h __C, + __m512h __D, const int __E) +{ + return __builtin_ia32_scalefph512_mask_round (__C, __D, __A, __B, + __E); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_scalef_round_ph (__mmask32 __A, __m512h __B, __m512h __C, + const int __D) +{ + return __builtin_ia32_scalefph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, __D); +} + +#else +#define _mm512_scalef_round_ph(A, B, C) \ + (__builtin_ia32_scalefph512_mask_round ((A), (B), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, (C))) + +#define _mm512_mask_scalef_round_ph(A, B, C, D, E) \ + (__builtin_ia32_scalefph512_mask_round ((C), (D), (A), (B), (E))) + +#define _mm512_maskz_scalef_round_ph(A, B, C, D) \ + (__builtin_ia32_scalefph512_mask_round ((B), (C), \ + _mm512_setzero_ph (), \ + (A), (D))) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vreduceph. */ +#ifdef __OPTIMIZE__ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_reduce_ph (__m512h __A, int __B) +{ + return __builtin_ia32_reduceph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_reduce_ph (__m512h __A, __mmask32 __B, __m512h __C, int __D) +{ + return __builtin_ia32_reduceph512_mask_round (__C, __D, __A, __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_reduce_ph (__mmask32 __A, __m512h __B, int __C) +{ + return __builtin_ia32_reduceph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_reduce_round_ph (__m512h __A, int __B, const int __C) +{ + return __builtin_ia32_reduceph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, __C); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_reduce_round_ph (__m512h __A, __mmask32 __B, __m512h __C, + int __D, const int __E) +{ + return __builtin_ia32_reduceph512_mask_round (__C, __D, __A, __B, + __E); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_reduce_round_ph (__mmask32 __A, __m512h __B, int __C, + const int __D) +{ + return __builtin_ia32_reduceph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, __D); +} + +#else +#define _mm512_reduce_ph(A, B) \ + (__builtin_ia32_reduceph512_mask_round ((A), (B), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, \ + _MM_FROUND_CUR_DIRECTION)) + +#define _mm512_mask_reduce_ph(A, B, C, D) \ + (__builtin_ia32_reduceph512_mask_round ((C), (D), (A), (B), \ + _MM_FROUND_CUR_DIRECTION)) + +#define _mm512_maskz_reduce_ph(A, B, C) \ + (__builtin_ia32_reduceph512_mask_round ((B), (C), \ + _mm512_setzero_ph (), \ + (A), _MM_FROUND_CUR_DIRECTION)) + +#define _mm512_reduce_round_ph(A, B, C) \ + (__builtin_ia32_reduceph512_mask_round ((A), (B), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, (C))) + +#define _mm512_mask_reduce_round_ph(A, B, C, D, E) \ + (__builtin_ia32_reduceph512_mask_round ((C), (D), (A), (B), (E))) + +#define _mm512_maskz_reduce_round_ph(A, B, C, D) \ + (__builtin_ia32_reduceph512_mask_round ((B), (C), \ + _mm512_setzero_ph (), \ + (A), (D))) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vrndscaleph. */ +#ifdef __OPTIMIZE__ +extern __inline __m512h + __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_roundscale_ph (__m512h __A, int __B) +{ + return __builtin_ia32_rndscaleph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_roundscale_ph (__m512h __A, __mmask32 __B, + __m512h __C, int __D) +{ + return __builtin_ia32_rndscaleph512_mask_round (__C, __D, __A, __B, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_roundscale_ph (__mmask32 __A, __m512h __B, int __C) +{ + return __builtin_ia32_rndscaleph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_roundscale_round_ph (__m512h __A, int __B, const int __C) +{ + return __builtin_ia32_rndscaleph512_mask_round (__A, __B, + _mm512_setzero_ph (), + (__mmask32) -1, + __C); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_roundscale_round_ph (__m512h __A, __mmask32 __B, + __m512h __C, int __D, const int __E) +{ + return __builtin_ia32_rndscaleph512_mask_round (__C, __D, __A, + __B, __E); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_roundscale_round_ph (__mmask32 __A, __m512h __B, int __C, + const int __D) +{ + return __builtin_ia32_rndscaleph512_mask_round (__B, __C, + _mm512_setzero_ph (), + __A, __D); +} + +#else +#define _mm512_roundscale_ph(A, B) \ + (__builtin_ia32_rndscaleph512_mask_round ((A), (B), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, \ + _MM_FROUND_CUR_DIRECTION)) + +#define _mm512_mask_roundscale_ph(A, B, C, D) \ + (__builtin_ia32_rndscaleph512_mask_round ((C), (D), (A), (B), \ + _MM_FROUND_CUR_DIRECTION)) + +#define _mm512_maskz_roundscale_ph(A, B, C) \ + (__builtin_ia32_rndscaleph512_mask_round ((B), (C), \ + _mm512_setzero_ph (), \ + (A), \ + _MM_FROUND_CUR_DIRECTION)) +#define _mm512_roundscale_round_ph(A, B, C) \ + (__builtin_ia32_rndscaleph512_mask_round ((A), (B), \ + _mm512_setzero_ph (), \ + (__mmask32)-1, (C))) + +#define _mm512_mask_roundscale_round_ph(A, B, C, D, E) \ + (__builtin_ia32_rndscaleph512_mask_round ((C), (D), (A), (B), (E))) + +#define _mm512_maskz_roundscale_round_ph(A, B, C, D) \ + (__builtin_ia32_rndscaleph512_mask_round ((B), (C), \ + _mm512_setzero_ph (), \ + (A), (D))) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vfpclassph. */ +#ifdef __OPTIMIZE__ +extern __inline __mmask32 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_fpclass_ph_mask (__mmask32 __U, __m512h __A, + const int __imm) +{ + return (__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) __A, + __imm, __U); +} + +extern __inline __mmask32 +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_fpclass_ph_mask (__m512h __A, const int __imm) +{ + return (__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) __A, + __imm, + (__mmask32) -1); +} + +#else +#define _mm512_mask_fpclass_ph_mask(u, x, c) \ + ((__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) (__m512h) (x), \ + (int) (c),(__mmask8)(u))) + +#define _mm512_fpclass_ph_mask(x, c) \ + ((__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) (__m512h) (x), \ + (int) (c),(__mmask8)-1)) +#endif /* __OPIMTIZE__ */ + +/* Intrinsics vgetexpph. */ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_getexp_ph (__m512h __A) +{ + return (__m512h) + __builtin_ia32_getexpph512_mask ((__v32hf) __A, + (__v32hf) _mm512_setzero_ph (), + (__mmask32) -1, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_getexp_ph (__m512h __W, __mmask32 __U, __m512h __A) +{ + return (__m512h) + __builtin_ia32_getexpph512_mask ((__v32hf) __A, (__v32hf) __W, + (__mmask32) __U, _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_getexp_ph (__mmask32 __U, __m512h __A) +{ + return (__m512h) + __builtin_ia32_getexpph512_mask ((__v32hf) __A, + (__v32hf) _mm512_setzero_ph (), + (__mmask32) __U, _MM_FROUND_CUR_DIRECTION); +} + +#ifdef __OPTIMIZE__ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_getexp_round_ph (__m512h __A, const int __R) +{ + return (__m512h) __builtin_ia32_getexpph512_mask ((__v32hf) __A, + (__v32hf) + _mm512_setzero_ph (), + (__mmask32) -1, __R); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_getexp_round_ph (__m512h __W, __mmask32 __U, __m512h __A, + const int __R) +{ + return (__m512h) __builtin_ia32_getexpph512_mask ((__v32hf) __A, + (__v32hf) __W, + (__mmask32) __U, __R); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_getexp_round_ph (__mmask32 __U, __m512h __A, const int __R) +{ + return (__m512h) __builtin_ia32_getexpph512_mask ((__v32hf) __A, + (__v32hf) + _mm512_setzero_ph (), + (__mmask32) __U, __R); +} + +#else +#define _mm512_getexp_round_ph(A, R) \ + ((__m512h)__builtin_ia32_getexpph512_mask((__v32hf)(__m512h)(A), \ + (__v32hf)_mm512_setzero_ph(), (__mmask32)-1, R)) + +#define _mm512_mask_getexp_round_ph(W, U, A, R) \ + ((__m512h)__builtin_ia32_getexpph512_mask((__v32hf)(__m512h)(A), \ + (__v32hf)(__m512h)(W), (__mmask32)(U), R)) + +#define _mm512_maskz_getexp_round_ph(U, A, R) \ + ((__m512h)__builtin_ia32_getexpph512_mask((__v32hf)(__m512h)(A), \ + (__v32hf)_mm512_setzero_ph(), (__mmask32)(U), R)) + +#endif /* __OPTIMIZE__ */ + +/* Intrinsics vgetmantph. */ +#ifdef __OPTIMIZE__ +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_getmant_ph (__m512h __A, _MM_MANTISSA_NORM_ENUM __B, + _MM_MANTISSA_SIGN_ENUM __C) +{ + return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, + (__C << 2) | __B, + _mm512_setzero_ph (), + (__mmask32) -1, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_getmant_ph (__m512h __W, __mmask32 __U, __m512h __A, + _MM_MANTISSA_NORM_ENUM __B, + _MM_MANTISSA_SIGN_ENUM __C) +{ + return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, + (__C << 2) | __B, + (__v32hf) __W, __U, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_getmant_ph (__mmask32 __U, __m512h __A, + _MM_MANTISSA_NORM_ENUM __B, + _MM_MANTISSA_SIGN_ENUM __C) +{ + return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, + (__C << 2) | __B, + (__v32hf) + _mm512_setzero_ph (), + __U, + _MM_FROUND_CUR_DIRECTION); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_getmant_round_ph (__m512h __A, _MM_MANTISSA_NORM_ENUM __B, + _MM_MANTISSA_SIGN_ENUM __C, const int __R) +{ + return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, + (__C << 2) | __B, + _mm512_setzero_ph (), + (__mmask32) -1, __R); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_mask_getmant_round_ph (__m512h __W, __mmask32 __U, __m512h __A, + _MM_MANTISSA_NORM_ENUM __B, + _MM_MANTISSA_SIGN_ENUM __C, const int __R) +{ + return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, + (__C << 2) | __B, + (__v32hf) __W, __U, + __R); +} + +extern __inline __m512h +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) +_mm512_maskz_getmant_round_ph (__mmask32 __U, __m512h __A, + _MM_MANTISSA_NORM_ENUM __B, + _MM_MANTISSA_SIGN_ENUM __C, const int __R) +{ + return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A, + (__C << 2) | __B, + (__v32hf) + _mm512_setzero_ph (), + __U, __R); +} + +#else +#define _mm512_getmant_ph(X, B, C) \ + ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ + (int)(((C)<<2) | (B)), \ + (__v32hf)(__m512h) \ + _mm512_setzero_ph(), \ + (__mmask32)-1, \ + _MM_FROUND_CUR_DIRECTION)) + +#define _mm512_mask_getmant_ph(W, U, X, B, C) \ + ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ + (int)(((C)<<2) | (B)), \ + (__v32hf)(__m512h)(W), \ + (__mmask32)(U), \ + _MM_FROUND_CUR_DIRECTION)) + + +#define _mm512_maskz_getmant_ph(U, X, B, C) \ + ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ + (int)(((C)<<2) | (B)), \ + (__v32hf)(__m512h) \ + _mm512_setzero_ph(), \ + (__mmask32)(U), \ + _MM_FROUND_CUR_DIRECTION)) + +#define _mm512_getmant_round_ph(X, B, C, R) \ + ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ + (int)(((C)<<2) | (B)), \ + (__v32hf)(__m512h) \ + _mm512_setzero_ph(), \ + (__mmask32)-1, \ + (R))) + +#define _mm512_mask_getmant_round_ph(W, U, X, B, C, R) \ + ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ + (int)(((C)<<2) | (B)), \ + (__v32hf)(__m512h)(W), \ + (__mmask32)(U), \ + (R))) + + +#define _mm512_maskz_getmant_round_ph(U, X, B, C, R) \ + ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), \ + (int)(((C)<<2) | (B)), \ + (__v32hf)(__m512h) \ + _mm512_setzero_ph(), \ + (__mmask32)(U), \ + (R))) + +#endif /* __OPTIMIZE__ */ + /* Intrinsics vcvtph2dq. */ extern __inline __m512i __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) @@ -4364,244 +5733,6 @@ _mm512_maskz_cvt_roundepu16_ph (__mmask32 __A, __m512i __B, int __C) #endif /* __OPTIMIZE__ */ -/* Intrinsics vcvtsh2si, vcvtsh2us. */ -extern __inline int -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtsh_i32 (__m128h __A) -{ - return (int) __builtin_ia32_vcvtsh2si32_round (__A, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline unsigned -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtsh_u32 (__m128h __A) -{ - return (int) __builtin_ia32_vcvtsh2usi32_round (__A, - _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline int -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundsh_i32 (__m128h __A, const int __R) -{ - return (int) __builtin_ia32_vcvtsh2si32_round (__A, __R); -} - -extern __inline unsigned -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundsh_u32 (__m128h __A, const int __R) -{ - return (int) __builtin_ia32_vcvtsh2usi32_round (__A, __R); -} - -#else -#define _mm_cvt_roundsh_i32(A, B) \ - ((int)__builtin_ia32_vcvtsh2si32_round ((A), (B))) -#define _mm_cvt_roundsh_u32(A, B) \ - ((int)__builtin_ia32_vcvtsh2usi32_round ((A), (B))) - -#endif /* __OPTIMIZE__ */ - -#ifdef __x86_64__ -extern __inline long long -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtsh_i64 (__m128h __A) -{ - return (long long) - __builtin_ia32_vcvtsh2si64_round (__A, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline unsigned long long -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtsh_u64 (__m128h __A) -{ - return (long long) - __builtin_ia32_vcvtsh2usi64_round (__A, _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline long long -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundsh_i64 (__m128h __A, const int __R) -{ - return (long long) __builtin_ia32_vcvtsh2si64_round (__A, __R); -} - -extern __inline unsigned long long -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundsh_u64 (__m128h __A, const int __R) -{ - return (long long) __builtin_ia32_vcvtsh2usi64_round (__A, __R); -} - -#else -#define _mm_cvt_roundsh_i64(A, B) \ - ((long long)__builtin_ia32_vcvtsh2si64_round ((A), (B))) -#define _mm_cvt_roundsh_u64(A, B) \ - ((long long)__builtin_ia32_vcvtsh2usi64_round ((A), (B))) - -#endif /* __OPTIMIZE__ */ -#endif /* __x86_64__ */ - -/* Intrinsics vcvttsh2si, vcvttsh2us. */ -extern __inline int -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvttsh_i32 (__m128h __A) -{ - return (int) - __builtin_ia32_vcvttsh2si32_round (__A, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline unsigned -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvttsh_u32 (__m128h __A) -{ - return (int) - __builtin_ia32_vcvttsh2usi32_round (__A, _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline int -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtt_roundsh_i32 (__m128h __A, const int __R) -{ - return (int) __builtin_ia32_vcvttsh2si32_round (__A, __R); -} - -extern __inline unsigned -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtt_roundsh_u32 (__m128h __A, const int __R) -{ - return (int) __builtin_ia32_vcvttsh2usi32_round (__A, __R); -} - -#else -#define _mm_cvtt_roundsh_i32(A, B) \ - ((int)__builtin_ia32_vcvttsh2si32_round ((A), (B))) -#define _mm_cvtt_roundsh_u32(A, B) \ - ((int)__builtin_ia32_vcvttsh2usi32_round ((A), (B))) - -#endif /* __OPTIMIZE__ */ - -#ifdef __x86_64__ -extern __inline long long -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvttsh_i64 (__m128h __A) -{ - return (long long) - __builtin_ia32_vcvttsh2si64_round (__A, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline unsigned long long -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvttsh_u64 (__m128h __A) -{ - return (long long) - __builtin_ia32_vcvttsh2usi64_round (__A, _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline long long -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtt_roundsh_i64 (__m128h __A, const int __R) -{ - return (long long) __builtin_ia32_vcvttsh2si64_round (__A, __R); -} - -extern __inline unsigned long long -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtt_roundsh_u64 (__m128h __A, const int __R) -{ - return (long long) __builtin_ia32_vcvttsh2usi64_round (__A, __R); -} - -#else -#define _mm_cvtt_roundsh_i64(A, B) \ - ((long long)__builtin_ia32_vcvttsh2si64_round ((A), (B))) -#define _mm_cvtt_roundsh_u64(A, B) \ - ((long long)__builtin_ia32_vcvttsh2usi64_round ((A), (B))) - -#endif /* __OPTIMIZE__ */ -#endif /* __x86_64__ */ - -/* Intrinsics vcvtsi2sh, vcvtusi2sh. */ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvti32_sh (__m128h __A, int __B) -{ - return __builtin_ia32_vcvtsi2sh32_round (__A, __B, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtu32_sh (__m128h __A, unsigned int __B) -{ - return __builtin_ia32_vcvtusi2sh32_round (__A, __B, _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundi32_sh (__m128h __A, int __B, const int __R) -{ - return __builtin_ia32_vcvtsi2sh32_round (__A, __B, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundu32_sh (__m128h __A, unsigned int __B, const int __R) -{ - return __builtin_ia32_vcvtusi2sh32_round (__A, __B, __R); -} - -#else -#define _mm_cvt_roundi32_sh(A, B, C) \ - (__builtin_ia32_vcvtsi2sh32_round ((A), (B), (C))) -#define _mm_cvt_roundu32_sh(A, B, C) \ - (__builtin_ia32_vcvtusi2sh32_round ((A), (B), (C))) - -#endif /* __OPTIMIZE__ */ - -#ifdef __x86_64__ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvti64_sh (__m128h __A, long long __B) -{ - return __builtin_ia32_vcvtsi2sh64_round (__A, __B, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtu64_sh (__m128h __A, unsigned long long __B) -{ - return __builtin_ia32_vcvtusi2sh64_round (__A, __B, _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundi64_sh (__m128h __A, long long __B, const int __R) -{ - return __builtin_ia32_vcvtsi2sh64_round (__A, __B, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundu64_sh (__m128h __A, unsigned long long __B, const int __R) -{ - return __builtin_ia32_vcvtusi2sh64_round (__A, __B, __R); -} - -#else -#define _mm_cvt_roundi64_sh(A, B, C) \ - (__builtin_ia32_vcvtsi2sh64_round ((A), (B), (C))) -#define _mm_cvt_roundu64_sh(A, B, C) \ - (__builtin_ia32_vcvtusi2sh64_round ((A), (B), (C))) - -#endif /* __OPTIMIZE__ */ -#endif /* __x86_64__ */ - /* Intrinsics vcvtph2pd. */ extern __inline __m512d __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) @@ -4900,286 +6031,6 @@ _mm512_maskz_cvt_roundpd_ph (__mmask8 __A, __m512d __B, int __C) #endif /* __OPTIMIZE__ */ -/* Intrinsics vcvtsh2ss, vcvtsh2sd. */ -extern __inline __m128 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtsh_ss (__m128 __A, __m128h __B) -{ - return __builtin_ia32_vcvtsh2ss_mask_round (__B, __A, - _mm_setzero_ps (), - (__mmask8) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_cvtsh_ss (__m128 __A, __mmask8 __B, __m128 __C, - __m128h __D) -{ - return __builtin_ia32_vcvtsh2ss_mask_round (__D, __C, __A, __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_cvtsh_ss (__mmask8 __A, __m128 __B, - __m128h __C) -{ - return __builtin_ia32_vcvtsh2ss_mask_round (__C, __B, - _mm_setzero_ps (), - __A, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128d -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtsh_sd (__m128d __A, __m128h __B) -{ - return __builtin_ia32_vcvtsh2sd_mask_round (__B, __A, - _mm_setzero_pd (), - (__mmask8) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128d -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_cvtsh_sd (__m128d __A, __mmask8 __B, __m128d __C, - __m128h __D) -{ - return __builtin_ia32_vcvtsh2sd_mask_round (__D, __C, __A, __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128d -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_cvtsh_sd (__mmask8 __A, __m128d __B, __m128h __C) -{ - return __builtin_ia32_vcvtsh2sd_mask_round (__C, __B, - _mm_setzero_pd (), - __A, _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline __m128 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundsh_ss (__m128 __A, __m128h __B, const int __R) -{ - return __builtin_ia32_vcvtsh2ss_mask_round (__B, __A, - _mm_setzero_ps (), - (__mmask8) -1, __R); -} - -extern __inline __m128 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_cvt_roundsh_ss (__m128 __A, __mmask8 __B, __m128 __C, - __m128h __D, const int __R) -{ - return __builtin_ia32_vcvtsh2ss_mask_round (__D, __C, __A, __B, __R); -} - -extern __inline __m128 -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_cvt_roundsh_ss (__mmask8 __A, __m128 __B, - __m128h __C, const int __R) -{ - return __builtin_ia32_vcvtsh2ss_mask_round (__C, __B, - _mm_setzero_ps (), - __A, __R); -} - -extern __inline __m128d -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundsh_sd (__m128d __A, __m128h __B, const int __R) -{ - return __builtin_ia32_vcvtsh2sd_mask_round (__B, __A, - _mm_setzero_pd (), - (__mmask8) -1, __R); -} - -extern __inline __m128d -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_cvt_roundsh_sd (__m128d __A, __mmask8 __B, __m128d __C, - __m128h __D, const int __R) -{ - return __builtin_ia32_vcvtsh2sd_mask_round (__D, __C, __A, __B, __R); -} - -extern __inline __m128d -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_cvt_roundsh_sd (__mmask8 __A, __m128d __B, __m128h __C, const int __R) -{ - return __builtin_ia32_vcvtsh2sd_mask_round (__C, __B, - _mm_setzero_pd (), - __A, __R); -} - -#else -#define _mm_cvt_roundsh_ss(A, B, R) \ - (__builtin_ia32_vcvtsh2ss_mask_round ((B), (A), \ - _mm_setzero_ps (), \ - (__mmask8) -1, (R))) - -#define _mm_mask_cvt_roundsh_ss(A, B, C, D, R) \ - (__builtin_ia32_vcvtsh2ss_mask_round ((D), (C), (A), (B), (R))) - -#define _mm_maskz_cvt_roundsh_ss(A, B, C, R) \ - (__builtin_ia32_vcvtsh2ss_mask_round ((C), (B), \ - _mm_setzero_ps (), \ - (A), (R))) - -#define _mm_cvt_roundsh_sd(A, B, R) \ - (__builtin_ia32_vcvtsh2sd_mask_round ((B), (A), \ - _mm_setzero_pd (), \ - (__mmask8) -1, (R))) - -#define _mm_mask_cvt_roundsh_sd(A, B, C, D, R) \ - (__builtin_ia32_vcvtsh2sd_mask_round ((D), (C), (A), (B), (R))) - -#define _mm_maskz_cvt_roundsh_sd(A, B, C, R) \ - (__builtin_ia32_vcvtsh2sd_mask_round ((C), (B), \ - _mm_setzero_pd (), \ - (A), (R))) - -#endif /* __OPTIMIZE__ */ - -/* Intrinsics vcvtss2sh, vcvtsd2sh. */ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtss_sh (__m128h __A, __m128 __B) -{ - return __builtin_ia32_vcvtss2sh_mask_round (__B, __A, - _mm_setzero_ph (), - (__mmask8) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_cvtss_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128 __D) -{ - return __builtin_ia32_vcvtss2sh_mask_round (__D, __C, __A, __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_cvtss_sh (__mmask8 __A, __m128h __B, __m128 __C) -{ - return __builtin_ia32_vcvtss2sh_mask_round (__C, __B, - _mm_setzero_ph (), - __A, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvtsd_sh (__m128h __A, __m128d __B) -{ - return __builtin_ia32_vcvtsd2sh_mask_round (__B, __A, - _mm_setzero_ph (), - (__mmask8) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_cvtsd_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128d __D) -{ - return __builtin_ia32_vcvtsd2sh_mask_round (__D, __C, __A, __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_cvtsd_sh (__mmask8 __A, __m128h __B, __m128d __C) -{ - return __builtin_ia32_vcvtsd2sh_mask_round (__C, __B, - _mm_setzero_ph (), - __A, _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundss_sh (__m128h __A, __m128 __B, const int __R) -{ - return __builtin_ia32_vcvtss2sh_mask_round (__B, __A, - _mm_setzero_ph (), - (__mmask8) -1, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_cvt_roundss_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128 __D, - const int __R) -{ - return __builtin_ia32_vcvtss2sh_mask_round (__D, __C, __A, __B, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_cvt_roundss_sh (__mmask8 __A, __m128h __B, __m128 __C, - const int __R) -{ - return __builtin_ia32_vcvtss2sh_mask_round (__C, __B, - _mm_setzero_ph (), - __A, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_cvt_roundsd_sh (__m128h __A, __m128d __B, const int __R) -{ - return __builtin_ia32_vcvtsd2sh_mask_round (__B, __A, - _mm_setzero_ph (), - (__mmask8) -1, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_cvt_roundsd_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128d __D, - const int __R) -{ - return __builtin_ia32_vcvtsd2sh_mask_round (__D, __C, __A, __B, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_cvt_roundsd_sh (__mmask8 __A, __m128h __B, __m128d __C, - const int __R) -{ - return __builtin_ia32_vcvtsd2sh_mask_round (__C, __B, - _mm_setzero_ph (), - __A, __R); -} - -#else -#define _mm_cvt_roundss_sh(A, B, R) \ - (__builtin_ia32_vcvtss2sh_mask_round ((B), (A), \ - _mm_setzero_ph (), \ - (__mmask8) -1, R)) - -#define _mm_mask_cvt_roundss_sh(A, B, C, D, R) \ - (__builtin_ia32_vcvtss2sh_mask_round ((D), (C), (A), (B), (R))) - -#define _mm_maskz_cvt_roundss_sh(A, B, C, R) \ - (__builtin_ia32_vcvtss2sh_mask_round ((C), (B), \ - _mm_setzero_ph (), \ - A, R)) - -#define _mm_cvt_roundsd_sh(A, B, R) \ - (__builtin_ia32_vcvtsd2sh_mask_round ((B), (A), \ - _mm_setzero_ph (), \ - (__mmask8) -1, R)) - -#define _mm_mask_cvt_roundsd_sh(A, B, C, D, R) \ - (__builtin_ia32_vcvtsd2sh_mask_round ((D), (C), (A), (B), (R))) - -#define _mm_maskz_cvt_roundsd_sh(A, B, C, R) \ - (__builtin_ia32_vcvtsd2sh_mask_round ((C), (B), \ - _mm_setzero_ph (), \ - (A), (R))) - -#endif /* __OPTIMIZE__ */ - /* Intrinsics vfmaddsub[132,213,231]ph. */ extern __inline __m512h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) @@ -5840,418 +6691,6 @@ _mm512_maskz_fnmsub_round_ph (__mmask32 __U, __m512h __A, __m512h __B, #endif /* __OPTIMIZE__ */ -/* Intrinsics vfmadd[132,213,231]sh. */ -extern __inline __m128h - __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fmadd_sh (__m128h __W, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fmadd_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fmadd_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask3 ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fmadd_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - - -#ifdef __OPTIMIZE__ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) -1, - __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fmadd_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B, - const int __R) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U, - const int __R) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask3 ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fmadd_round_sh (__mmask8 __U, __m128h __W, __m128h __A, - __m128h __B, const int __R) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, __R); -} - -#else -#define _mm_fmadd_round_sh(A, B, C, R) \ - ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), (C), (-1), (R))) -#define _mm_mask_fmadd_round_sh(A, U, B, C, R) \ - ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), (C), (U), (R))) -#define _mm_mask3_fmadd_round_sh(A, B, C, U, R) \ - ((__m128h) __builtin_ia32_vfmaddsh3_mask3 ((A), (B), (C), (U), (R))) -#define _mm_maskz_fmadd_round_sh(U, A, B, C, R) \ - ((__m128h) __builtin_ia32_vfmaddsh3_maskz ((A), (B), (C), (U), (R))) - -#endif /* __OPTIMIZE__ */ - -/* Intrinsics vfnmadd[132,213,231]sh. */ -extern __inline __m128h - __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fnmadd_sh (__m128h __W, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fnmadd_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fnmadd_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U) -{ - return (__m128h) __builtin_ia32_vfnmaddsh3_mask3 ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fnmadd_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfnmaddsh3_maskz ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - - -#ifdef __OPTIMIZE__ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fnmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R) -{ - return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) -1, - __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fnmadd_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B, - const int __R) -{ - return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fnmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U, - const int __R) -{ - return (__m128h) __builtin_ia32_vfnmaddsh3_mask3 ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fnmadd_round_sh (__mmask8 __U, __m128h __W, __m128h __A, - __m128h __B, const int __R) -{ - return (__m128h) __builtin_ia32_vfnmaddsh3_maskz ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, __R); -} - -#else -#define _mm_fnmadd_round_sh(A, B, C, R) \ - ((__m128h) __builtin_ia32_vfnmaddsh3_mask ((A), (B), (C), (-1), (R))) -#define _mm_mask_fnmadd_round_sh(A, U, B, C, R) \ - ((__m128h) __builtin_ia32_vfnmaddsh3_mask ((A), (B), (C), (U), (R))) -#define _mm_mask3_fnmadd_round_sh(A, B, C, U, R) \ - ((__m128h) __builtin_ia32_vfnmaddsh3_mask3 ((A), (B), (C), (U), (R))) -#define _mm_maskz_fnmadd_round_sh(U, A, B, C, R) \ - ((__m128h) __builtin_ia32_vfnmaddsh3_maskz ((A), (B), (C), (U), (R))) - -#endif /* __OPTIMIZE__ */ - -/* Intrinsics vfmsub[132,213,231]sh. */ -extern __inline __m128h - __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fmsub_sh (__m128h __W, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - -(__v8hf) __B, - (__mmask8) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fmsub_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - -(__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fmsub_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U) -{ - return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fmsub_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, - (__v8hf) __A, - -(__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - - -#ifdef __OPTIMIZE__ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - -(__v8hf) __B, - (__mmask8) -1, - __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fmsub_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B, - const int __R) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - (__v8hf) __A, - -(__v8hf) __B, - (__mmask8) __U, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U, - const int __R) -{ - return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W, - (__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fmsub_round_sh (__mmask8 __U, __m128h __W, __m128h __A, - __m128h __B, const int __R) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, - (__v8hf) __A, - -(__v8hf) __B, - (__mmask8) __U, __R); -} - -#else -#define _mm_fmsub_round_sh(A, B, C, R) \ - ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), -(C), (-1), (R))) -#define _mm_mask_fmsub_round_sh(A, U, B, C, R) \ - ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), -(C), (U), (R))) -#define _mm_mask3_fmsub_round_sh(A, B, C, U, R) \ - ((__m128h) __builtin_ia32_vfmsubsh3_mask3 ((A), (B), (C), (U), (R))) -#define _mm_maskz_fmsub_round_sh(U, A, B, C, R) \ - ((__m128h) __builtin_ia32_vfmaddsh3_maskz ((A), (B), -(C), (U), (R))) - -#endif /* __OPTIMIZE__ */ - -/* Intrinsics vfnmsub[132,213,231]sh. */ -extern __inline __m128h - __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fnmsub_sh (__m128h __W, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - -(__v8hf) __A, - -(__v8hf) __B, - (__mmask8) -1, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fnmsub_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - -(__v8hf) __A, - -(__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fnmsub_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U) -{ - return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W, - -(__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fnmsub_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, - -(__v8hf) __A, - -(__v8hf) __B, - (__mmask8) __U, - _MM_FROUND_CUR_DIRECTION); -} - - -#ifdef __OPTIMIZE__ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fnmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - -(__v8hf) __A, - -(__v8hf) __B, - (__mmask8) -1, - __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fnmsub_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B, - const int __R) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W, - -(__v8hf) __A, - -(__v8hf) __B, - (__mmask8) __U, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fnmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U, - const int __R) -{ - return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W, - -(__v8hf) __A, - (__v8hf) __B, - (__mmask8) __U, __R); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fnmsub_round_sh (__mmask8 __U, __m128h __W, __m128h __A, - __m128h __B, const int __R) -{ - return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W, - -(__v8hf) __A, - -(__v8hf) __B, - (__mmask8) __U, __R); -} - -#else -#define _mm_fnmsub_round_sh(A, B, C, R) \ - ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), -(B), -(C), (-1), (R))) -#define _mm_mask_fnmsub_round_sh(A, U, B, C, R) \ - ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), -(B), -(C), (U), (R))) -#define _mm_mask3_fnmsub_round_sh(A, B, C, U, R) \ - ((__m128h) __builtin_ia32_vfmsubsh3_mask3 ((A), -(B), (C), (U), (R))) -#define _mm_maskz_fnmsub_round_sh(U, A, B, C, R) \ - ((__m128h) __builtin_ia32_vfmaddsh3_maskz ((A), -(B), -(C), (U), (R))) - -#endif /* __OPTIMIZE__ */ - /* Intrinsics vf[,c]maddcph. */ extern __inline __m512h __attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) @@ -6636,400 +7075,6 @@ _mm512_maskz_fmul_round_pch (__mmask16 __A, __m512h __B, #endif /* __OPTIMIZE__ */ -/* Intrinsics vf[,c]maddcsh. */ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fcmadd_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D) -{ - return (__m128h) - __builtin_ia32_vfcmaddcsh_mask_round ((__v8hf) __A, - (__v8hf) __C, - (__v8hf) __D, __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fcmadd_sch (__m128h __A, __m128h __B, __m128h __C, __mmask8 __D) -{ - return (__m128h) - __builtin_ia32_vfcmaddcsh_mask3_round ((__v8hf) __A, - (__v8hf) __B, - (__v8hf) __C, __D, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fcmadd_sch (__mmask8 __A, __m128h __B, __m128h __C, __m128h __D) -{ - return (__m128h) - __builtin_ia32_vfcmaddcsh_maskz_round ((__v8hf) __B, - (__v8hf) __C, - (__v8hf) __D, - __A, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fcmadd_sch (__m128h __A, __m128h __B, __m128h __C) -{ - return (__m128h) - __builtin_ia32_vfcmaddcsh_round ((__v8hf) __A, - (__v8hf) __B, - (__v8hf) __C, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fmadd_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D) -{ - return (__m128h) - __builtin_ia32_vfmaddcsh_mask_round ((__v8hf) __A, - (__v8hf) __C, - (__v8hf) __D, __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fmadd_sch (__m128h __A, __m128h __B, __m128h __C, __mmask8 __D) -{ - return (__m128h) - __builtin_ia32_vfmaddcsh_mask3_round ((__v8hf) __A, - (__v8hf) __B, - (__v8hf) __C, __D, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fmadd_sch (__mmask8 __A, __m128h __B, __m128h __C, __m128h __D) -{ - return (__m128h) - __builtin_ia32_vfmaddcsh_maskz_round ((__v8hf) __B, - (__v8hf) __C, - (__v8hf) __D, - __A, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fmadd_sch (__m128h __A, __m128h __B, __m128h __C) -{ - return (__m128h) - __builtin_ia32_vfmaddcsh_round ((__v8hf) __A, - (__v8hf) __B, - (__v8hf) __C, - _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fcmadd_round_sch (__m128h __A, __mmask8 __B, __m128h __C, - __m128h __D, const int __E) -{ - return (__m128h) - __builtin_ia32_vfcmaddcsh_mask_round ((__v8hf) __A, - (__v8hf) __C, - (__v8hf) __D, - __B, __E); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fcmadd_round_sch (__m128h __A, __m128h __B, __m128h __C, - __mmask8 __D, const int __E) -{ - return (__m128h) - __builtin_ia32_vfcmaddcsh_mask3_round ((__v8hf) __A, - (__v8hf) __B, - (__v8hf) __C, - __D, __E); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fcmadd_round_sch (__mmask8 __A, __m128h __B, __m128h __C, - __m128h __D, const int __E) -{ - return (__m128h) - __builtin_ia32_vfcmaddcsh_maskz_round ((__v8hf) __B, - (__v8hf) __C, - (__v8hf) __D, - __A, __E); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fcmadd_round_sch (__m128h __A, __m128h __B, __m128h __C, const int __D) -{ - return (__m128h) - __builtin_ia32_vfcmaddcsh_round ((__v8hf) __A, - (__v8hf) __B, - (__v8hf) __C, - __D); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fmadd_round_sch (__m128h __A, __mmask8 __B, __m128h __C, - __m128h __D, const int __E) -{ - return (__m128h) - __builtin_ia32_vfmaddcsh_mask_round ((__v8hf) __A, - (__v8hf) __C, - (__v8hf) __D, - __B, __E); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask3_fmadd_round_sch (__m128h __A, __m128h __B, __m128h __C, - __mmask8 __D, const int __E) -{ - return (__m128h) - __builtin_ia32_vfmaddcsh_mask3_round ((__v8hf) __A, - (__v8hf) __B, - (__v8hf) __C, - __D, __E); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fmadd_round_sch (__mmask8 __A, __m128h __B, __m128h __C, - __m128h __D, const int __E) -{ - return (__m128h) - __builtin_ia32_vfmaddcsh_maskz_round ((__v8hf) __B, - (__v8hf) __C, - (__v8hf) __D, - __A, __E); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fmadd_round_sch (__m128h __A, __m128h __B, __m128h __C, const int __D) -{ - return (__m128h) - __builtin_ia32_vfmaddcsh_round ((__v8hf) __A, - (__v8hf) __B, - (__v8hf) __C, - __D); -} -#else -#define _mm_mask_fcmadd_round_sch(A, B, C, D, E) \ - ((__m128h) \ - __builtin_ia32_vfcmaddcsh_mask_round ((__v8hf) (A), \ - (__v8hf) (C), \ - (__v8hf) (D), \ - (B), (E))) - - -#define _mm_mask3_fcmadd_round_sch(A, B, C, D, E) \ - ((__m128h) \ - __builtin_ia32_vfcmaddcsh_mask3_round ((__v8hf) (A), \ - (__v8hf) (B), \ - (__v8hf) (C), \ - (D), (E))) - -#define _mm_maskz_fcmadd_round_sch(A, B, C, D, E) \ - __builtin_ia32_vfcmaddcsh_maskz_round ((B), (C), (D), (A), (E)) - -#define _mm_fcmadd_round_sch(A, B, C, D) \ - __builtin_ia32_vfcmaddcsh_round ((A), (B), (C), (D)) - -#define _mm_mask_fmadd_round_sch(A, B, C, D, E) \ - ((__m128h) \ - __builtin_ia32_vfmaddcsh_mask_round ((__v8hf) (A), \ - (__v8hf) (C), \ - (__v8hf) (D), \ - (B), (E))) - -#define _mm_mask3_fmadd_round_sch(A, B, C, D, E) \ - ((__m128h) \ - __builtin_ia32_vfmaddcsh_mask3_round ((__v8hf) (A), \ - (__v8hf) (B), \ - (__v8hf) (C), \ - (D), (E))) - -#define _mm_maskz_fmadd_round_sch(A, B, C, D, E) \ - __builtin_ia32_vfmaddcsh_maskz_round ((B), (C), (D), (A), (E)) - -#define _mm_fmadd_round_sch(A, B, C, D) \ - __builtin_ia32_vfmaddcsh_round ((A), (B), (C), (D)) - -#endif /* __OPTIMIZE__ */ - -/* Intrinsics vf[,c]mulcsh. */ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fcmul_sch (__m128h __A, __m128h __B) -{ - return (__m128h) - __builtin_ia32_vfcmulcsh_round ((__v8hf) __A, - (__v8hf) __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fcmul_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D) -{ - return (__m128h) - __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __C, - (__v8hf) __D, - (__v8hf) __A, - __B, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fcmul_sch (__mmask8 __A, __m128h __B, __m128h __C) -{ - return (__m128h) - __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __B, - (__v8hf) __C, - _mm_setzero_ph (), - __A, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fmul_sch (__m128h __A, __m128h __B) -{ - return (__m128h) - __builtin_ia32_vfmulcsh_round ((__v8hf) __A, - (__v8hf) __B, - _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fmul_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D) -{ - return (__m128h) - __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __C, - (__v8hf) __D, - (__v8hf) __A, - __B, _MM_FROUND_CUR_DIRECTION); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fmul_sch (__mmask8 __A, __m128h __B, __m128h __C) -{ - return (__m128h) - __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __B, - (__v8hf) __C, - _mm_setzero_ph (), - __A, _MM_FROUND_CUR_DIRECTION); -} - -#ifdef __OPTIMIZE__ -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fcmul_round_sch (__m128h __A, __m128h __B, const int __D) -{ - return (__m128h) - __builtin_ia32_vfcmulcsh_round ((__v8hf) __A, - (__v8hf) __B, - __D); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fcmul_round_sch (__m128h __A, __mmask8 __B, __m128h __C, - __m128h __D, const int __E) -{ - return (__m128h) - __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __C, - (__v8hf) __D, - (__v8hf) __A, - __B, __E); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fcmul_round_sch (__mmask8 __A, __m128h __B, __m128h __C, - const int __E) -{ - return (__m128h) - __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __B, - (__v8hf) __C, - _mm_setzero_ph (), - __A, __E); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_fmul_round_sch (__m128h __A, __m128h __B, const int __D) -{ - return (__m128h) - __builtin_ia32_vfmulcsh_round ((__v8hf) __A, - (__v8hf) __B, __D); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_mask_fmul_round_sch (__m128h __A, __mmask8 __B, __m128h __C, - __m128h __D, const int __E) -{ - return (__m128h) - __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __C, - (__v8hf) __D, - (__v8hf) __A, - __B, __E); -} - -extern __inline __m128h -__attribute__ ((__gnu_inline__, __always_inline__, __artificial__)) -_mm_maskz_fmul_round_sch (__mmask8 __A, __m128h __B, __m128h __C, const int __E) -{ - return (__m128h) - __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __B, - (__v8hf) __C, - _mm_setzero_ph (), - __A, __E); -} - -#else -#define _mm_fcmul_round_sch(__A, __B, __D) \ - (__m128h) __builtin_ia32_vfcmulcsh_round ((__v8hf) __A, \ - (__v8hf) __B, __D) - -#define _mm_mask_fcmul_round_sch(__A, __B, __C, __D, __E) \ - (__m128h) __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __C, \ - (__v8hf) __D, \ - (__v8hf) __A, \ - __B, __E) - -#define _mm_maskz_fcmul_round_sch(__A, __B, __C, __E) \ - (__m128h) __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __B, \ - (__v8hf) __C, \ - _mm_setzero_ph (), \ - __A, __E) - -#define _mm_fmul_round_sch(__A, __B, __D) \ - (__m128h) __builtin_ia32_vfmulcsh_round ((__v8hf) __A, \ - (__v8hf) __B, __D) - -#define _mm_mask_fmul_round_sch(__A, __B, __C, __D, __E) \ - (__m128h) __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __C, \ - (__v8hf) __D, \ - (__v8hf) __A, \ - __B, __E) - -#define _mm_maskz_fmul_round_sch(__A, __B, __C, __E) \ - (__m128h) __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __B, \ - (__v8hf) __C, \ - _mm_setzero_ph (), \ - __A, __E) - -#endif /* __OPTIMIZE__ */ - #define _MM512_REDUCE_OP(op) \ __m256h __T1 = (__m256h) _mm512_extractf64x4_pd ((__m512d) __A, 0); \ __m256h __T2 = (__m256h) _mm512_extractf64x4_pd ((__m512d) __A, 1); \ @@ -7193,27 +7238,9 @@ _mm512_set1_pch (_Float16 _Complex __A) #define _mm512_maskz_cmul_round_pch(U, A, B, R) \ _mm512_maskz_fcmul_round_pch ((U), (A), (B), (R)) -#define _mm_mul_sch(A, B) _mm_fmul_sch ((A), (B)) -#define _mm_mask_mul_sch(W, U, A, B) _mm_mask_fmul_sch ((W), (U), (A), (B)) -#define _mm_maskz_mul_sch(U, A, B) _mm_maskz_fmul_sch ((U), (A), (B)) -#define _mm_mul_round_sch(A, B, R) _mm_fmul_round_sch ((A), (B), (R)) -#define _mm_mask_mul_round_sch(W, U, A, B, R) \ - _mm_mask_fmul_round_sch ((W), (U), (A), (B), (R)) -#define _mm_maskz_mul_round_sch(U, A, B, R) \ - _mm_maskz_fmul_round_sch ((U), (A), (B), (R)) - -#define _mm_cmul_sch(A, B) _mm_fcmul_sch ((A), (B)) -#define _mm_mask_cmul_sch(W, U, A, B) _mm_mask_fcmul_sch ((W), (U), (A), (B)) -#define _mm_maskz_cmul_sch(U, A, B) _mm_maskz_fcmul_sch ((U), (A), (B)) -#define _mm_cmul_round_sch(A, B, R) _mm_fcmul_round_sch ((A), (B), (R)) -#define _mm_mask_cmul_round_sch(W, U, A, B, R) \ - _mm_mask_fcmul_round_sch ((W), (U), (A), (B), (R)) -#define _mm_maskz_cmul_round_sch(U, A, B, R) \ - _mm_maskz_fcmul_round_sch ((U), (A), (B), (R)) - -#ifdef __DISABLE_AVX512FP16__ -#undef __DISABLE_AVX512FP16__ +#ifdef __DISABLE_AVX512FP16_512__ +#undef __DISABLE_AVX512FP16_512__ #pragma GCC pop_options -#endif /* __DISABLE_AVX512FP16__ */ +#endif /* __DISABLE_AVX512FP16_512__ */ -#endif /* __AVX512FP16INTRIN_H_INCLUDED */ +#endif /* _AVX512FP16INTRIN_H_INCLUDED */ |