10#error "Never use <avx512fintrin.h> directly; include <immintrin.h> instead."
13#ifndef __AVX512FINTRIN_H
14#define __AVX512FINTRIN_H
24typedef unsigned char __v64qu
__attribute__((__vector_size__(64)));
25typedef unsigned short __v32hu
__attribute__((__vector_size__(64)));
26typedef unsigned long long __v8du
__attribute__((__vector_size__(64)));
27typedef unsigned int __v16su
__attribute__((__vector_size__(64)));
31typedef signed char __v64qs
__attribute__((__vector_size__(64)));
33typedef float __m512
__attribute__((__vector_size__(64), __aligned__(64)));
34typedef double __m512d
__attribute__((__vector_size__(64), __aligned__(64)));
35typedef long long __m512i
__attribute__((__vector_size__(64), __aligned__(64)));
37typedef float __m512_u
__attribute__((__vector_size__(64), __aligned__(1)));
38typedef double __m512d_u
__attribute__((__vector_size__(64), __aligned__(1)));
39typedef long long __m512i_u
__attribute__((__vector_size__(64), __aligned__(1)));
45#define _MM_FROUND_TO_NEAREST_INT 0x00
46#define _MM_FROUND_TO_NEG_INF 0x01
47#define _MM_FROUND_TO_POS_INF 0x02
48#define _MM_FROUND_TO_ZERO 0x03
49#define _MM_FROUND_CUR_DIRECTION 0x04
59#define _MM_CMPINT_GE _MM_CMPINT_NLT
61#define _MM_CMPINT_GT _MM_CMPINT_NLE
170#define __DEFAULT_FN_ATTRS512 __attribute__((__always_inline__, __nodebug__, __target__("avx512f,evex512"), __min_vector_width__(512)))
171#define __DEFAULT_FN_ATTRS128 \
172 __attribute__((__always_inline__, __nodebug__, \
173 __target__("avx512f,no-evex512"), __min_vector_width__(128)))
174#define __DEFAULT_FN_ATTRS \
175 __attribute__((__always_inline__, __nodebug__, \
176 __target__("avx512f,no-evex512")))
183 return __extension__ (__m512i)(__v8di){ 0, 0, 0, 0, 0, 0, 0, 0 };
186#define _mm512_setzero_epi32 _mm512_setzero_si512
191 return (__m512d)__builtin_ia32_undef512();
197 return (__m512)__builtin_ia32_undef512();
203 return (__m512)__builtin_ia32_undef512();
209 return (__m512i)__builtin_ia32_undef512();
215 return (__m512i)__builtin_shufflevector((__v4si) __A, (__v4si) __A,
216 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
222 return (__m512i)__builtin_ia32_selectd_512(__M,
230 return (__m512i)__builtin_ia32_selectd_512(__M,
238 return (__m512i)__builtin_shufflevector((__v2di) __A, (__v2di) __A,
239 0, 0, 0, 0, 0, 0, 0, 0);
245 return (__m512i)__builtin_ia32_selectq_512(__M,
254 return (__m512i)__builtin_ia32_selectq_512(__M,
263 return __extension__ (__m512){ 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f,
264 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f };
267#define _mm512_setzero _mm512_setzero_ps
272 return __extension__ (__m512d){ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 };
278 return __extension__ (__m512){ __w, __w, __w, __w, __w, __w, __w, __w,
279 __w, __w, __w, __w, __w, __w, __w, __w };
285 return __extension__ (__m512d){ __w, __w, __w, __w, __w, __w, __w, __w };
291 return __extension__ (__m512i)(__v64qi){
292 __w, __w, __w, __w, __w, __w, __w, __w,
293 __w, __w, __w, __w, __w, __w, __w, __w,
294 __w, __w, __w, __w, __w, __w, __w, __w,
295 __w, __w, __w, __w, __w, __w, __w, __w,
296 __w, __w, __w, __w, __w, __w, __w, __w,
297 __w, __w, __w, __w, __w, __w, __w, __w,
298 __w, __w, __w, __w, __w, __w, __w, __w,
299 __w, __w, __w, __w, __w, __w, __w, __w };
305 return __extension__ (__m512i)(__v32hi){
306 __w, __w, __w, __w, __w, __w, __w, __w,
307 __w, __w, __w, __w, __w, __w, __w, __w,
308 __w, __w, __w, __w, __w, __w, __w, __w,
309 __w, __w, __w, __w, __w, __w, __w, __w };
315 return __extension__ (__m512i)(__v16si){
316 __s, __s, __s, __s, __s, __s, __s, __s,
317 __s, __s, __s, __s, __s, __s, __s, __s };
323 return (__m512i)__builtin_ia32_selectd_512(__M,
331 return __extension__(__m512i)(__v8di){ __d, __d, __d, __d, __d, __d, __d, __d };
337 return (__m512i)__builtin_ia32_selectq_512(__M,
345 return (__m512)__builtin_shufflevector((__v4sf) __A, (__v4sf) __A,
346 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
352 return __extension__ (__m512i)(__v16si)
353 {
__D, __C, __B, __A,
__D, __C, __B, __A,
354 __D, __C, __B, __A,
__D, __C, __B, __A };
361 return __extension__ (__m512i) (__v8di)
362 {
__D, __C, __B, __A,
__D, __C, __B, __A };
368 return __extension__ (__m512d)
369 {
__D, __C, __B, __A,
__D, __C, __B, __A };
375 return __extension__ (__m512)
376 {
__D, __C, __B, __A,
__D, __C, __B, __A,
377 __D, __C, __B, __A,
__D, __C, __B, __A };
380#define _mm512_setr4_epi32(e0,e1,e2,e3) \
381 _mm512_set4_epi32((e3),(e2),(e1),(e0))
383#define _mm512_setr4_epi64(e0,e1,e2,e3) \
384 _mm512_set4_epi64((e3),(e2),(e1),(e0))
386#define _mm512_setr4_pd(e0,e1,e2,e3) \
387 _mm512_set4_pd((e3),(e2),(e1),(e0))
389#define _mm512_setr4_ps(e0,e1,e2,e3) \
390 _mm512_set4_ps((e3),(e2),(e1),(e0))
395 return (__m512d)__builtin_shufflevector((__v2df) __A, (__v2df) __A,
396 0, 0, 0, 0, 0, 0, 0, 0);
404 return __builtin_shufflevector(
__a, __builtin_nondeterministic_value(
__a), 0,
405 1, 2, 3, 4, 5, 6, 7);
411 return __builtin_shufflevector(
__a, __builtin_nondeterministic_value(
__a), 0,
412 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
418 return __builtin_shufflevector(
__a,
__a, 0, 1);
424 return __builtin_shufflevector(__A, __A, 0, 1, 2, 3);
430 return __builtin_shufflevector(
__a,
__a, 0, 1, 2, 3);
436 return __builtin_shufflevector(__A, __A, 0, 1, 2, 3, 4, 5, 6, 7);
442 return (__m512) (__A);
448 return (__m512i) (__A);
454 __m256d __B = __builtin_nondeterministic_value(__B);
455 return __builtin_shufflevector(
456 __builtin_shufflevector(__A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3),
457 __B, 0, 1, 2, 3, 4, 5, 6, 7);
463 return (__m512d) (__A);
469 return (__m512i) (__A);
475 __m256 __B = __builtin_nondeterministic_value(__B);
476 return __builtin_shufflevector(
477 __builtin_shufflevector(__A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3, 4, 5, 6, 7),
478 __B, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
484 __m256i __B = __builtin_nondeterministic_value(__B);
485 return __builtin_shufflevector(
486 __builtin_shufflevector(__A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3),
487 __B, 0, 1, 2, 3, 4, 5, 6, 7);
493 return __builtin_shufflevector( __A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3, 4, 5, 6, 7);
499 return (__m512) (__A);
505 return (__m512d) (__A);
511 return (__m128i)__builtin_shufflevector(__A, __A , 0, 1);
517 return (__m256i)__builtin_shufflevector(__A, __A , 0, 1, 2, 3);
548 return __builtin_shufflevector((__v2df)
__a, (__v2df)
_mm_setzero_pd(), 0, 1, 2, 3, 2, 3, 2, 3);
567 return __builtin_shufflevector((__v4df)
__a, (__v4df)
_mm256_setzero_pd(), 0, 1, 2, 3, 4, 5, 6, 7);
585 return __builtin_shufflevector((__v4sf)
__a, (__v4sf)
_mm_setzero_ps(), 0, 1, 2, 3, 4, 5, 6, 7, 4, 5, 6, 7, 4, 5, 6, 7);
603 return __builtin_shufflevector((__v8sf)
__a, (__v8sf)
_mm256_setzero_ps(), 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
621 return __builtin_shufflevector((__v2di)
__a, (__v2di)
_mm_setzero_si128(), 0, 1, 2, 3, 2, 3, 2, 3);
646 return (__m512i)((__v16su)
__a & (__v16su)
__b);
652 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__k,
667 return (__m512i)((__v8du)
__a & (__v8du)
__b);
673 return (__m512i) __builtin_ia32_selectq_512 ((
__mmask8) __k,
688 return (__m512i)(~(__v8du)__A & (__v8du)__B);
694 return (__m512i)(~(__v16su)__A & (__v16su)__B);
700 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
715 return (__m512i)(~(__v8du)__A & (__v8du)__B);
721 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
736 return (__m512i)((__v16su)
__a | (__v16su)
__b);
742 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__k,
756 return (__m512i)((__v8du)
__a | (__v8du)
__b);
762 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__k,
776 return (__m512i)((__v16su)
__a ^ (__v16su)
__b);
782 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__k,
796 return (__m512i)((__v8du)
__a ^ (__v8du)
__b);
802 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__k,
816 return (__m512i)((__v8du)
__a & (__v8du)
__b);
822 return (__m512i)((__v8du)
__a | (__v8du)
__b);
828 return (__m512i)((__v8du)
__a ^ (__v8du)
__b);
836 return (__m512d)((__v8df)
__a + (__v8df)
__b);
842 return (__m512)((__v16sf)
__a + (__v16sf)
__b);
848 return (__m512d)((__v8df)
__a * (__v8df)
__b);
854 return (__m512)((__v16sf)
__a * (__v16sf)
__b);
860 return (__m512d)((__v8df)
__a - (__v8df)
__b);
866 return (__m512)((__v16sf)
__a - (__v16sf)
__b);
872 return (__m512i) ((__v8du) __A + (__v8du) __B);
878 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
886 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
894 return (__m512i) ((__v8du) __A - (__v8du) __B);
900 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
908 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
916 return (__m512i) ((__v16su) __A + (__v16su) __B);
922 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
930 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
938 return (__m512i) ((__v16su) __A - (__v16su) __B);
944 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
952 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
957#define _mm512_max_round_pd(A, B, R) \
958 ((__m512d)__builtin_ia32_maxpd512((__v8df)(__m512d)(A), \
959 (__v8df)(__m512d)(B), (int)(R)))
961#define _mm512_mask_max_round_pd(W, U, A, B, R) \
962 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
963 (__v8df)_mm512_max_round_pd((A), (B), (R)), \
966#define _mm512_maskz_max_round_pd(U, A, B, R) \
967 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
968 (__v8df)_mm512_max_round_pd((A), (B), (R)), \
969 (__v8df)_mm512_setzero_pd()))
974 return (__m512d) __builtin_ia32_maxpd512((__v8df) __A, (__v8df) __B,
981 return (__m512d)__builtin_ia32_selectpd_512(__U,
989 return (__m512d)__builtin_ia32_selectpd_512(__U,
994#define _mm512_max_round_ps(A, B, R) \
995 ((__m512)__builtin_ia32_maxps512((__v16sf)(__m512)(A), \
996 (__v16sf)(__m512)(B), (int)(R)))
998#define _mm512_mask_max_round_ps(W, U, A, B, R) \
999 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
1000 (__v16sf)_mm512_max_round_ps((A), (B), (R)), \
1003#define _mm512_maskz_max_round_ps(U, A, B, R) \
1004 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
1005 (__v16sf)_mm512_max_round_ps((A), (B), (R)), \
1006 (__v16sf)_mm512_setzero_ps()))
1011 return (__m512) __builtin_ia32_maxps512((__v16sf) __A, (__v16sf) __B,
1018 return (__m512)__builtin_ia32_selectps_512(__U,
1026 return (__m512)__builtin_ia32_selectps_512(__U,
1033 return (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A,
1042 return (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A,
1049#define _mm_max_round_ss(A, B, R) \
1050 ((__m128)__builtin_ia32_maxss_round_mask((__v4sf)(__m128)(A), \
1051 (__v4sf)(__m128)(B), \
1052 (__v4sf)_mm_setzero_ps(), \
1053 (__mmask8)-1, (int)(R)))
1055#define _mm_mask_max_round_ss(W, U, A, B, R) \
1056 ((__m128)__builtin_ia32_maxss_round_mask((__v4sf)(__m128)(A), \
1057 (__v4sf)(__m128)(B), \
1058 (__v4sf)(__m128)(W), (__mmask8)(U), \
1061#define _mm_maskz_max_round_ss(U, A, B, R) \
1062 ((__m128)__builtin_ia32_maxss_round_mask((__v4sf)(__m128)(A), \
1063 (__v4sf)(__m128)(B), \
1064 (__v4sf)_mm_setzero_ps(), \
1065 (__mmask8)(U), (int)(R)))
1069 return (__m128d) __builtin_ia32_maxsd_round_mask ((__v2df) __A,
1078 return (__m128d) __builtin_ia32_maxsd_round_mask ((__v2df) __A,
1085#define _mm_max_round_sd(A, B, R) \
1086 ((__m128d)__builtin_ia32_maxsd_round_mask((__v2df)(__m128d)(A), \
1087 (__v2df)(__m128d)(B), \
1088 (__v2df)_mm_setzero_pd(), \
1089 (__mmask8)-1, (int)(R)))
1091#define _mm_mask_max_round_sd(W, U, A, B, R) \
1092 ((__m128d)__builtin_ia32_maxsd_round_mask((__v2df)(__m128d)(A), \
1093 (__v2df)(__m128d)(B), \
1094 (__v2df)(__m128d)(W), \
1095 (__mmask8)(U), (int)(R)))
1097#define _mm_maskz_max_round_sd(U, A, B, R) \
1098 ((__m128d)__builtin_ia32_maxsd_round_mask((__v2df)(__m128d)(A), \
1099 (__v2df)(__m128d)(B), \
1100 (__v2df)_mm_setzero_pd(), \
1101 (__mmask8)(U), (int)(R)))
1103static __inline __m512i
1107 return (__m512i)__builtin_elementwise_max((__v16si)__A, (__v16si)__B);
1113 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
1121 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
1129 return (__m512i)__builtin_elementwise_max((__v16su)__A, (__v16su)__B);
1135 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
1143 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
1151 return (__m512i)__builtin_elementwise_max((__v8di)__A, (__v8di)__B);
1157 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1165 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1173 return (__m512i)__builtin_elementwise_max((__v8du)__A, (__v8du)__B);
1179 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1187 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1192#define _mm512_min_round_pd(A, B, R) \
1193 ((__m512d)__builtin_ia32_minpd512((__v8df)(__m512d)(A), \
1194 (__v8df)(__m512d)(B), (int)(R)))
1196#define _mm512_mask_min_round_pd(W, U, A, B, R) \
1197 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
1198 (__v8df)_mm512_min_round_pd((A), (B), (R)), \
1201#define _mm512_maskz_min_round_pd(U, A, B, R) \
1202 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
1203 (__v8df)_mm512_min_round_pd((A), (B), (R)), \
1204 (__v8df)_mm512_setzero_pd()))
1209 return (__m512d) __builtin_ia32_minpd512((__v8df) __A, (__v8df) __B,
1216 return (__m512d)__builtin_ia32_selectpd_512(__U,
1224 return (__m512d)__builtin_ia32_selectpd_512(__U,
1229#define _mm512_min_round_ps(A, B, R) \
1230 ((__m512)__builtin_ia32_minps512((__v16sf)(__m512)(A), \
1231 (__v16sf)(__m512)(B), (int)(R)))
1233#define _mm512_mask_min_round_ps(W, U, A, B, R) \
1234 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
1235 (__v16sf)_mm512_min_round_ps((A), (B), (R)), \
1238#define _mm512_maskz_min_round_ps(U, A, B, R) \
1239 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
1240 (__v16sf)_mm512_min_round_ps((A), (B), (R)), \
1241 (__v16sf)_mm512_setzero_ps()))
1246 return (__m512) __builtin_ia32_minps512((__v16sf) __A, (__v16sf) __B,
1253 return (__m512)__builtin_ia32_selectps_512(__U,
1261 return (__m512)__builtin_ia32_selectps_512(__U,
1268 return (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A,
1277 return (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A,
1284#define _mm_min_round_ss(A, B, R) \
1285 ((__m128)__builtin_ia32_minss_round_mask((__v4sf)(__m128)(A), \
1286 (__v4sf)(__m128)(B), \
1287 (__v4sf)_mm_setzero_ps(), \
1288 (__mmask8)-1, (int)(R)))
1290#define _mm_mask_min_round_ss(W, U, A, B, R) \
1291 ((__m128)__builtin_ia32_minss_round_mask((__v4sf)(__m128)(A), \
1292 (__v4sf)(__m128)(B), \
1293 (__v4sf)(__m128)(W), (__mmask8)(U), \
1296#define _mm_maskz_min_round_ss(U, A, B, R) \
1297 ((__m128)__builtin_ia32_minss_round_mask((__v4sf)(__m128)(A), \
1298 (__v4sf)(__m128)(B), \
1299 (__v4sf)_mm_setzero_ps(), \
1300 (__mmask8)(U), (int)(R)))
1304 return (__m128d) __builtin_ia32_minsd_round_mask ((__v2df) __A,
1313 return (__m128d) __builtin_ia32_minsd_round_mask ((__v2df) __A,
1320#define _mm_min_round_sd(A, B, R) \
1321 ((__m128d)__builtin_ia32_minsd_round_mask((__v2df)(__m128d)(A), \
1322 (__v2df)(__m128d)(B), \
1323 (__v2df)_mm_setzero_pd(), \
1324 (__mmask8)-1, (int)(R)))
1326#define _mm_mask_min_round_sd(W, U, A, B, R) \
1327 ((__m128d)__builtin_ia32_minsd_round_mask((__v2df)(__m128d)(A), \
1328 (__v2df)(__m128d)(B), \
1329 (__v2df)(__m128d)(W), \
1330 (__mmask8)(U), (int)(R)))
1332#define _mm_maskz_min_round_sd(U, A, B, R) \
1333 ((__m128d)__builtin_ia32_minsd_round_mask((__v2df)(__m128d)(A), \
1334 (__v2df)(__m128d)(B), \
1335 (__v2df)_mm_setzero_pd(), \
1336 (__mmask8)(U), (int)(R)))
1338static __inline __m512i
1342 return (__m512i)__builtin_elementwise_min((__v16si)__A, (__v16si)__B);
1348 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
1356 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
1364 return (__m512i)__builtin_elementwise_min((__v16su)__A, (__v16su)__B);
1370 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
1378 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
1386 return (__m512i)__builtin_elementwise_min((__v8di)__A, (__v8di)__B);
1392 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1400 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1408 return (__m512i)__builtin_elementwise_min((__v8du)__A, (__v8du)__B);
1414 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1422 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1430 return (__m512i)__builtin_ia32_pmuldq512((__v16si)__X, (__v16si)
__Y);
1436 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1444 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1452 return (__m512i)__builtin_ia32_pmuludq512((__v16si)__X, (__v16si)
__Y);
1458 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1466 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
1474 return (__m512i) ((__v16su) __A * (__v16su) __B);
1480 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
1488 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
1495 return (__m512i) ((__v8du) __A * (__v8du) __B);
1500 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
1505#define _mm512_sqrt_round_pd(A, R) \
1506 ((__m512d)__builtin_ia32_sqrtpd512((__v8df)(__m512d)(A), (int)(R)))
1508#define _mm512_mask_sqrt_round_pd(W, U, A, R) \
1509 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
1510 (__v8df)_mm512_sqrt_round_pd((A), (R)), \
1511 (__v8df)(__m512d)(W)))
1513#define _mm512_maskz_sqrt_round_pd(U, A, R) \
1514 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
1515 (__v8df)_mm512_sqrt_round_pd((A), (R)), \
1516 (__v8df)_mm512_setzero_pd()))
1521 return (__m512d)__builtin_ia32_sqrtpd512((__v8df)__A,
1528 return (__m512d)__builtin_ia32_selectpd_512(__U,
1536 return (__m512d)__builtin_ia32_selectpd_512(__U,
1541#define _mm512_sqrt_round_ps(A, R) \
1542 ((__m512)__builtin_ia32_sqrtps512((__v16sf)(__m512)(A), (int)(R)))
1544#define _mm512_mask_sqrt_round_ps(W, U, A, R) \
1545 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
1546 (__v16sf)_mm512_sqrt_round_ps((A), (R)), \
1547 (__v16sf)(__m512)(W)))
1549#define _mm512_maskz_sqrt_round_ps(U, A, R) \
1550 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
1551 (__v16sf)_mm512_sqrt_round_ps((A), (R)), \
1552 (__v16sf)_mm512_setzero_ps()))
1557 return (__m512)__builtin_ia32_sqrtps512((__v16sf)__A,
1564 return (__m512)__builtin_ia32_selectps_512(__U,
1572 return (__m512)__builtin_ia32_selectps_512(__U,
1580 return (__m512d) __builtin_ia32_rsqrt14pd512_mask ((__v8df) __A,
1588 return (__m512d) __builtin_ia32_rsqrt14pd512_mask ((__v8df) __A,
1596 return (__m512d) __builtin_ia32_rsqrt14pd512_mask ((__v8df) __A,
1605 return (__m512) __builtin_ia32_rsqrt14ps512_mask ((__v16sf) __A,
1614 return (__m512) __builtin_ia32_rsqrt14ps512_mask ((__v16sf) __A,
1622 return (__m512) __builtin_ia32_rsqrt14ps512_mask ((__v16sf) __A,
1631 return (__m128) __builtin_ia32_rsqrt14ss_mask ((__v4sf) __A,
1641 return (__m128) __builtin_ia32_rsqrt14ss_mask ((__v4sf) __A,
1650 return (__m128) __builtin_ia32_rsqrt14ss_mask ((__v4sf) __A,
1659 return (__m128d) __builtin_ia32_rsqrt14sd_mask ((__v2df) __A,
1669 return (__m128d) __builtin_ia32_rsqrt14sd_mask ( (__v2df) __A,
1678 return (__m128d) __builtin_ia32_rsqrt14sd_mask ( (__v2df) __A,
1687 return (__m512d) __builtin_ia32_rcp14pd512_mask ((__v8df) __A,
1696 return (__m512d) __builtin_ia32_rcp14pd512_mask ((__v8df) __A,
1704 return (__m512d) __builtin_ia32_rcp14pd512_mask ((__v8df) __A,
1713 return (__m512) __builtin_ia32_rcp14ps512_mask ((__v16sf) __A,
1722 return (__m512) __builtin_ia32_rcp14ps512_mask ((__v16sf) __A,
1730 return (__m512) __builtin_ia32_rcp14ps512_mask ((__v16sf) __A,
1739 return (__m128) __builtin_ia32_rcp14ss_mask ((__v4sf) __A,
1749 return (__m128) __builtin_ia32_rcp14ss_mask ((__v4sf) __A,
1758 return (__m128) __builtin_ia32_rcp14ss_mask ((__v4sf) __A,
1767 return (__m128d) __builtin_ia32_rcp14sd_mask ((__v2df) __A,
1777 return (__m128d) __builtin_ia32_rcp14sd_mask ( (__v2df) __A,
1786 return (__m128d) __builtin_ia32_rcp14sd_mask ( (__v2df) __A,
1795 return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
1797 (__v16sf) __A, (
unsigned short)-1,
1804 return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
1813 return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
1815 (__v8df) __A, (
unsigned char)-1,
1822 return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
1831 return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
1840 return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
1842 (__v16sf) __A, (
unsigned short)-1,
1849 return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
1851 (__v8df) __A, (
unsigned char)-1,
1858 return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
1867 return (__m512i)__builtin_elementwise_abs((__v8di)__A);
1873 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
1881 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
1889 return (__m512i)__builtin_elementwise_abs((__v16si) __A);
1895 return (__m512i)__builtin_ia32_selectd_512(__U,
1903 return (__m512i)__builtin_ia32_selectd_512(__U,
1911 return __builtin_ia32_selectss_128(__U, __A, __W);
1920#define _mm_add_round_ss(A, B, R) \
1921 ((__m128)__builtin_ia32_addss_round_mask((__v4sf)(__m128)(A), \
1922 (__v4sf)(__m128)(B), \
1923 (__v4sf)_mm_setzero_ps(), \
1924 (__mmask8)-1, (int)(R)))
1926#define _mm_mask_add_round_ss(W, U, A, B, R) \
1927 ((__m128)__builtin_ia32_addss_round_mask((__v4sf)(__m128)(A), \
1928 (__v4sf)(__m128)(B), \
1929 (__v4sf)(__m128)(W), (__mmask8)(U), \
1932#define _mm_maskz_add_round_ss(U, A, B, R) \
1933 ((__m128)__builtin_ia32_addss_round_mask((__v4sf)(__m128)(A), \
1934 (__v4sf)(__m128)(B), \
1935 (__v4sf)_mm_setzero_ps(), \
1936 (__mmask8)(U), (int)(R)))
1941 return __builtin_ia32_selectsd_128(__U, __A, __W);
1949#define _mm_add_round_sd(A, B, R) \
1950 ((__m128d)__builtin_ia32_addsd_round_mask((__v2df)(__m128d)(A), \
1951 (__v2df)(__m128d)(B), \
1952 (__v2df)_mm_setzero_pd(), \
1953 (__mmask8)-1, (int)(R)))
1955#define _mm_mask_add_round_sd(W, U, A, B, R) \
1956 ((__m128d)__builtin_ia32_addsd_round_mask((__v2df)(__m128d)(A), \
1957 (__v2df)(__m128d)(B), \
1958 (__v2df)(__m128d)(W), \
1959 (__mmask8)(U), (int)(R)))
1961#define _mm_maskz_add_round_sd(U, A, B, R) \
1962 ((__m128d)__builtin_ia32_addsd_round_mask((__v2df)(__m128d)(A), \
1963 (__v2df)(__m128d)(B), \
1964 (__v2df)_mm_setzero_pd(), \
1965 (__mmask8)(U), (int)(R)))
1969 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
1976 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
1983 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
1990 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
1995#define _mm512_add_round_pd(A, B, R) \
1996 ((__m512d)__builtin_ia32_addpd512((__v8df)(__m512d)(A), \
1997 (__v8df)(__m512d)(B), (int)(R)))
1999#define _mm512_mask_add_round_pd(W, U, A, B, R) \
2000 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
2001 (__v8df)_mm512_add_round_pd((A), (B), (R)), \
2002 (__v8df)(__m512d)(W)))
2004#define _mm512_maskz_add_round_pd(U, A, B, R) \
2005 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
2006 (__v8df)_mm512_add_round_pd((A), (B), (R)), \
2007 (__v8df)_mm512_setzero_pd()))
2009#define _mm512_add_round_ps(A, B, R) \
2010 ((__m512)__builtin_ia32_addps512((__v16sf)(__m512)(A), \
2011 (__v16sf)(__m512)(B), (int)(R)))
2013#define _mm512_mask_add_round_ps(W, U, A, B, R) \
2014 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
2015 (__v16sf)_mm512_add_round_ps((A), (B), (R)), \
2016 (__v16sf)(__m512)(W)))
2018#define _mm512_maskz_add_round_ps(U, A, B, R) \
2019 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
2020 (__v16sf)_mm512_add_round_ps((A), (B), (R)), \
2021 (__v16sf)_mm512_setzero_ps()))
2026 return __builtin_ia32_selectss_128(__U, __A, __W);
2034#define _mm_sub_round_ss(A, B, R) \
2035 ((__m128)__builtin_ia32_subss_round_mask((__v4sf)(__m128)(A), \
2036 (__v4sf)(__m128)(B), \
2037 (__v4sf)_mm_setzero_ps(), \
2038 (__mmask8)-1, (int)(R)))
2040#define _mm_mask_sub_round_ss(W, U, A, B, R) \
2041 ((__m128)__builtin_ia32_subss_round_mask((__v4sf)(__m128)(A), \
2042 (__v4sf)(__m128)(B), \
2043 (__v4sf)(__m128)(W), (__mmask8)(U), \
2046#define _mm_maskz_sub_round_ss(U, A, B, R) \
2047 ((__m128)__builtin_ia32_subss_round_mask((__v4sf)(__m128)(A), \
2048 (__v4sf)(__m128)(B), \
2049 (__v4sf)_mm_setzero_ps(), \
2050 (__mmask8)(U), (int)(R)))
2055 return __builtin_ia32_selectsd_128(__U, __A, __W);
2064#define _mm_sub_round_sd(A, B, R) \
2065 ((__m128d)__builtin_ia32_subsd_round_mask((__v2df)(__m128d)(A), \
2066 (__v2df)(__m128d)(B), \
2067 (__v2df)_mm_setzero_pd(), \
2068 (__mmask8)-1, (int)(R)))
2070#define _mm_mask_sub_round_sd(W, U, A, B, R) \
2071 ((__m128d)__builtin_ia32_subsd_round_mask((__v2df)(__m128d)(A), \
2072 (__v2df)(__m128d)(B), \
2073 (__v2df)(__m128d)(W), \
2074 (__mmask8)(U), (int)(R)))
2076#define _mm_maskz_sub_round_sd(U, A, B, R) \
2077 ((__m128d)__builtin_ia32_subsd_round_mask((__v2df)(__m128d)(A), \
2078 (__v2df)(__m128d)(B), \
2079 (__v2df)_mm_setzero_pd(), \
2080 (__mmask8)(U), (int)(R)))
2084 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
2091 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
2098 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
2105 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
2110#define _mm512_sub_round_pd(A, B, R) \
2111 ((__m512d)__builtin_ia32_subpd512((__v8df)(__m512d)(A), \
2112 (__v8df)(__m512d)(B), (int)(R)))
2114#define _mm512_mask_sub_round_pd(W, U, A, B, R) \
2115 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
2116 (__v8df)_mm512_sub_round_pd((A), (B), (R)), \
2117 (__v8df)(__m512d)(W)))
2119#define _mm512_maskz_sub_round_pd(U, A, B, R) \
2120 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
2121 (__v8df)_mm512_sub_round_pd((A), (B), (R)), \
2122 (__v8df)_mm512_setzero_pd()))
2124#define _mm512_sub_round_ps(A, B, R) \
2125 ((__m512)__builtin_ia32_subps512((__v16sf)(__m512)(A), \
2126 (__v16sf)(__m512)(B), (int)(R)))
2128#define _mm512_mask_sub_round_ps(W, U, A, B, R) \
2129 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
2130 (__v16sf)_mm512_sub_round_ps((A), (B), (R)), \
2131 (__v16sf)(__m512)(W)))
2133#define _mm512_maskz_sub_round_ps(U, A, B, R) \
2134 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
2135 (__v16sf)_mm512_sub_round_ps((A), (B), (R)), \
2136 (__v16sf)_mm512_setzero_ps()))
2141 return __builtin_ia32_selectss_128(__U, __A, __W);
2149#define _mm_mul_round_ss(A, B, R) \
2150 ((__m128)__builtin_ia32_mulss_round_mask((__v4sf)(__m128)(A), \
2151 (__v4sf)(__m128)(B), \
2152 (__v4sf)_mm_setzero_ps(), \
2153 (__mmask8)-1, (int)(R)))
2155#define _mm_mask_mul_round_ss(W, U, A, B, R) \
2156 ((__m128)__builtin_ia32_mulss_round_mask((__v4sf)(__m128)(A), \
2157 (__v4sf)(__m128)(B), \
2158 (__v4sf)(__m128)(W), (__mmask8)(U), \
2161#define _mm_maskz_mul_round_ss(U, A, B, R) \
2162 ((__m128)__builtin_ia32_mulss_round_mask((__v4sf)(__m128)(A), \
2163 (__v4sf)(__m128)(B), \
2164 (__v4sf)_mm_setzero_ps(), \
2165 (__mmask8)(U), (int)(R)))
2170 return __builtin_ia32_selectsd_128(__U, __A, __W);
2179#define _mm_mul_round_sd(A, B, R) \
2180 ((__m128d)__builtin_ia32_mulsd_round_mask((__v2df)(__m128d)(A), \
2181 (__v2df)(__m128d)(B), \
2182 (__v2df)_mm_setzero_pd(), \
2183 (__mmask8)-1, (int)(R)))
2185#define _mm_mask_mul_round_sd(W, U, A, B, R) \
2186 ((__m128d)__builtin_ia32_mulsd_round_mask((__v2df)(__m128d)(A), \
2187 (__v2df)(__m128d)(B), \
2188 (__v2df)(__m128d)(W), \
2189 (__mmask8)(U), (int)(R)))
2191#define _mm_maskz_mul_round_sd(U, A, B, R) \
2192 ((__m128d)__builtin_ia32_mulsd_round_mask((__v2df)(__m128d)(A), \
2193 (__v2df)(__m128d)(B), \
2194 (__v2df)_mm_setzero_pd(), \
2195 (__mmask8)(U), (int)(R)))
2199 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
2206 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
2213 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
2220 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
2225#define _mm512_mul_round_pd(A, B, R) \
2226 ((__m512d)__builtin_ia32_mulpd512((__v8df)(__m512d)(A), \
2227 (__v8df)(__m512d)(B), (int)(R)))
2229#define _mm512_mask_mul_round_pd(W, U, A, B, R) \
2230 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
2231 (__v8df)_mm512_mul_round_pd((A), (B), (R)), \
2232 (__v8df)(__m512d)(W)))
2234#define _mm512_maskz_mul_round_pd(U, A, B, R) \
2235 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
2236 (__v8df)_mm512_mul_round_pd((A), (B), (R)), \
2237 (__v8df)_mm512_setzero_pd()))
2239#define _mm512_mul_round_ps(A, B, R) \
2240 ((__m512)__builtin_ia32_mulps512((__v16sf)(__m512)(A), \
2241 (__v16sf)(__m512)(B), (int)(R)))
2243#define _mm512_mask_mul_round_ps(W, U, A, B, R) \
2244 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
2245 (__v16sf)_mm512_mul_round_ps((A), (B), (R)), \
2246 (__v16sf)(__m512)(W)))
2248#define _mm512_maskz_mul_round_ps(U, A, B, R) \
2249 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
2250 (__v16sf)_mm512_mul_round_ps((A), (B), (R)), \
2251 (__v16sf)_mm512_setzero_ps()))
2256 return __builtin_ia32_selectss_128(__U, __A, __W);
2265#define _mm_div_round_ss(A, B, R) \
2266 ((__m128)__builtin_ia32_divss_round_mask((__v4sf)(__m128)(A), \
2267 (__v4sf)(__m128)(B), \
2268 (__v4sf)_mm_setzero_ps(), \
2269 (__mmask8)-1, (int)(R)))
2271#define _mm_mask_div_round_ss(W, U, A, B, R) \
2272 ((__m128)__builtin_ia32_divss_round_mask((__v4sf)(__m128)(A), \
2273 (__v4sf)(__m128)(B), \
2274 (__v4sf)(__m128)(W), (__mmask8)(U), \
2277#define _mm_maskz_div_round_ss(U, A, B, R) \
2278 ((__m128)__builtin_ia32_divss_round_mask((__v4sf)(__m128)(A), \
2279 (__v4sf)(__m128)(B), \
2280 (__v4sf)_mm_setzero_ps(), \
2281 (__mmask8)(U), (int)(R)))
2286 return __builtin_ia32_selectsd_128(__U, __A, __W);
2295#define _mm_div_round_sd(A, B, R) \
2296 ((__m128d)__builtin_ia32_divsd_round_mask((__v2df)(__m128d)(A), \
2297 (__v2df)(__m128d)(B), \
2298 (__v2df)_mm_setzero_pd(), \
2299 (__mmask8)-1, (int)(R)))
2301#define _mm_mask_div_round_sd(W, U, A, B, R) \
2302 ((__m128d)__builtin_ia32_divsd_round_mask((__v2df)(__m128d)(A), \
2303 (__v2df)(__m128d)(B), \
2304 (__v2df)(__m128d)(W), \
2305 (__mmask8)(U), (int)(R)))
2307#define _mm_maskz_div_round_sd(U, A, B, R) \
2308 ((__m128d)__builtin_ia32_divsd_round_mask((__v2df)(__m128d)(A), \
2309 (__v2df)(__m128d)(B), \
2310 (__v2df)_mm_setzero_pd(), \
2311 (__mmask8)(U), (int)(R)))
2316 return (__m512d)((__v8df)
__a/(__v8df)
__b);
2321 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
2328 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
2336 return (__m512)((__v16sf)
__a/(__v16sf)
__b);
2341 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
2348 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
2353#define _mm512_div_round_pd(A, B, R) \
2354 ((__m512d)__builtin_ia32_divpd512((__v8df)(__m512d)(A), \
2355 (__v8df)(__m512d)(B), (int)(R)))
2357#define _mm512_mask_div_round_pd(W, U, A, B, R) \
2358 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
2359 (__v8df)_mm512_div_round_pd((A), (B), (R)), \
2360 (__v8df)(__m512d)(W)))
2362#define _mm512_maskz_div_round_pd(U, A, B, R) \
2363 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
2364 (__v8df)_mm512_div_round_pd((A), (B), (R)), \
2365 (__v8df)_mm512_setzero_pd()))
2367#define _mm512_div_round_ps(A, B, R) \
2368 ((__m512)__builtin_ia32_divps512((__v16sf)(__m512)(A), \
2369 (__v16sf)(__m512)(B), (int)(R)))
2371#define _mm512_mask_div_round_ps(W, U, A, B, R) \
2372 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
2373 (__v16sf)_mm512_div_round_ps((A), (B), (R)), \
2374 (__v16sf)(__m512)(W)))
2376#define _mm512_maskz_div_round_ps(U, A, B, R) \
2377 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
2378 (__v16sf)_mm512_div_round_ps((A), (B), (R)), \
2379 (__v16sf)_mm512_setzero_ps()))
2381#define _mm512_roundscale_ps(A, B) \
2382 ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(A), (int)(B), \
2383 (__v16sf)_mm512_undefined_ps(), \
2385 _MM_FROUND_CUR_DIRECTION))
2387#define _mm512_mask_roundscale_ps(A, B, C, imm) \
2388 ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(C), (int)(imm), \
2389 (__v16sf)(__m512)(A), (__mmask16)(B), \
2390 _MM_FROUND_CUR_DIRECTION))
2392#define _mm512_maskz_roundscale_ps(A, B, imm) \
2393 ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(B), (int)(imm), \
2394 (__v16sf)_mm512_setzero_ps(), \
2396 _MM_FROUND_CUR_DIRECTION))
2398#define _mm512_mask_roundscale_round_ps(A, B, C, imm, R) \
2399 ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(C), (int)(imm), \
2400 (__v16sf)(__m512)(A), (__mmask16)(B), \
2403#define _mm512_maskz_roundscale_round_ps(A, B, imm, R) \
2404 ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(B), (int)(imm), \
2405 (__v16sf)_mm512_setzero_ps(), \
2406 (__mmask16)(A), (int)(R)))
2408#define _mm512_roundscale_round_ps(A, imm, R) \
2409 ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(A), (int)(imm), \
2410 (__v16sf)_mm512_undefined_ps(), \
2411 (__mmask16)-1, (int)(R)))
2413#define _mm512_roundscale_pd(A, B) \
2414 ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(A), (int)(B), \
2415 (__v8df)_mm512_undefined_pd(), \
2417 _MM_FROUND_CUR_DIRECTION))
2419#define _mm512_mask_roundscale_pd(A, B, C, imm) \
2420 ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(C), (int)(imm), \
2421 (__v8df)(__m512d)(A), (__mmask8)(B), \
2422 _MM_FROUND_CUR_DIRECTION))
2424#define _mm512_maskz_roundscale_pd(A, B, imm) \
2425 ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(B), (int)(imm), \
2426 (__v8df)_mm512_setzero_pd(), \
2428 _MM_FROUND_CUR_DIRECTION))
2430#define _mm512_mask_roundscale_round_pd(A, B, C, imm, R) \
2431 ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(C), (int)(imm), \
2432 (__v8df)(__m512d)(A), (__mmask8)(B), \
2435#define _mm512_maskz_roundscale_round_pd(A, B, imm, R) \
2436 ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(B), (int)(imm), \
2437 (__v8df)_mm512_setzero_pd(), \
2438 (__mmask8)(A), (int)(R)))
2440#define _mm512_roundscale_round_pd(A, imm, R) \
2441 ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(A), (int)(imm), \
2442 (__v8df)_mm512_undefined_pd(), \
2443 (__mmask8)-1, (int)(R)))
2445#define _mm512_fmadd_round_pd(A, B, C, R) \
2446 ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
2447 (__v8df)(__m512d)(B), \
2448 (__v8df)(__m512d)(C), \
2449 (__mmask8)-1, (int)(R)))
2452#define _mm512_mask_fmadd_round_pd(A, U, B, C, R) \
2453 ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
2454 (__v8df)(__m512d)(B), \
2455 (__v8df)(__m512d)(C), \
2456 (__mmask8)(U), (int)(R)))
2459#define _mm512_mask3_fmadd_round_pd(A, B, C, U, R) \
2460 ((__m512d)__builtin_ia32_vfmaddpd512_mask3((__v8df)(__m512d)(A), \
2461 (__v8df)(__m512d)(B), \
2462 (__v8df)(__m512d)(C), \
2463 (__mmask8)(U), (int)(R)))
2466#define _mm512_maskz_fmadd_round_pd(U, A, B, C, R) \
2467 ((__m512d)__builtin_ia32_vfmaddpd512_maskz((__v8df)(__m512d)(A), \
2468 (__v8df)(__m512d)(B), \
2469 (__v8df)(__m512d)(C), \
2470 (__mmask8)(U), (int)(R)))
2473#define _mm512_fmsub_round_pd(A, B, C, R) \
2474 ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
2475 (__v8df)(__m512d)(B), \
2476 -(__v8df)(__m512d)(C), \
2477 (__mmask8)-1, (int)(R)))
2480#define _mm512_mask_fmsub_round_pd(A, U, B, C, R) \
2481 ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
2482 (__v8df)(__m512d)(B), \
2483 -(__v8df)(__m512d)(C), \
2484 (__mmask8)(U), (int)(R)))
2487#define _mm512_maskz_fmsub_round_pd(U, A, B, C, R) \
2488 ((__m512d)__builtin_ia32_vfmaddpd512_maskz((__v8df)(__m512d)(A), \
2489 (__v8df)(__m512d)(B), \
2490 -(__v8df)(__m512d)(C), \
2491 (__mmask8)(U), (int)(R)))
2494#define _mm512_fnmadd_round_pd(A, B, C, R) \
2495 ((__m512d)__builtin_ia32_vfmaddpd512_mask(-(__v8df)(__m512d)(A), \
2496 (__v8df)(__m512d)(B), \
2497 (__v8df)(__m512d)(C), \
2498 (__mmask8)-1, (int)(R)))
2501#define _mm512_mask3_fnmadd_round_pd(A, B, C, U, R) \
2502 ((__m512d)__builtin_ia32_vfmaddpd512_mask3(-(__v8df)(__m512d)(A), \
2503 (__v8df)(__m512d)(B), \
2504 (__v8df)(__m512d)(C), \
2505 (__mmask8)(U), (int)(R)))
2508#define _mm512_maskz_fnmadd_round_pd(U, A, B, C, R) \
2509 ((__m512d)__builtin_ia32_vfmaddpd512_maskz(-(__v8df)(__m512d)(A), \
2510 (__v8df)(__m512d)(B), \
2511 (__v8df)(__m512d)(C), \
2512 (__mmask8)(U), (int)(R)))
2515#define _mm512_fnmsub_round_pd(A, B, C, R) \
2516 ((__m512d)__builtin_ia32_vfmaddpd512_mask(-(__v8df)(__m512d)(A), \
2517 (__v8df)(__m512d)(B), \
2518 -(__v8df)(__m512d)(C), \
2519 (__mmask8)-1, (int)(R)))
2522#define _mm512_maskz_fnmsub_round_pd(U, A, B, C, R) \
2523 ((__m512d)__builtin_ia32_vfmaddpd512_maskz(-(__v8df)(__m512d)(A), \
2524 (__v8df)(__m512d)(B), \
2525 -(__v8df)(__m512d)(C), \
2526 (__mmask8)(U), (int)(R)))
2532 return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
2542 return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
2552 return (__m512d) __builtin_ia32_vfmaddpd512_mask3 ((__v8df) __A,
2562 return (__m512d) __builtin_ia32_vfmaddpd512_maskz ((__v8df) __A,
2572 return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
2582 return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
2592 return (__m512d) __builtin_ia32_vfmaddpd512_maskz ((__v8df) __A,
2602 return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
2612 return (__m512d) __builtin_ia32_vfmaddpd512_mask3 (-(__v8df) __A,
2622 return (__m512d) __builtin_ia32_vfmaddpd512_maskz (-(__v8df) __A,
2632 return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
2642 return (__m512d) __builtin_ia32_vfmaddpd512_maskz (-(__v8df) __A,
2649#define _mm512_fmadd_round_ps(A, B, C, R) \
2650 ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
2651 (__v16sf)(__m512)(B), \
2652 (__v16sf)(__m512)(C), \
2653 (__mmask16)-1, (int)(R)))
2656#define _mm512_mask_fmadd_round_ps(A, U, B, C, R) \
2657 ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
2658 (__v16sf)(__m512)(B), \
2659 (__v16sf)(__m512)(C), \
2660 (__mmask16)(U), (int)(R)))
2663#define _mm512_mask3_fmadd_round_ps(A, B, C, U, R) \
2664 ((__m512)__builtin_ia32_vfmaddps512_mask3((__v16sf)(__m512)(A), \
2665 (__v16sf)(__m512)(B), \
2666 (__v16sf)(__m512)(C), \
2667 (__mmask16)(U), (int)(R)))
2670#define _mm512_maskz_fmadd_round_ps(U, A, B, C, R) \
2671 ((__m512)__builtin_ia32_vfmaddps512_maskz((__v16sf)(__m512)(A), \
2672 (__v16sf)(__m512)(B), \
2673 (__v16sf)(__m512)(C), \
2674 (__mmask16)(U), (int)(R)))
2677#define _mm512_fmsub_round_ps(A, B, C, R) \
2678 ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
2679 (__v16sf)(__m512)(B), \
2680 -(__v16sf)(__m512)(C), \
2681 (__mmask16)-1, (int)(R)))
2684#define _mm512_mask_fmsub_round_ps(A, U, B, C, R) \
2685 ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
2686 (__v16sf)(__m512)(B), \
2687 -(__v16sf)(__m512)(C), \
2688 (__mmask16)(U), (int)(R)))
2691#define _mm512_maskz_fmsub_round_ps(U, A, B, C, R) \
2692 ((__m512)__builtin_ia32_vfmaddps512_maskz((__v16sf)(__m512)(A), \
2693 (__v16sf)(__m512)(B), \
2694 -(__v16sf)(__m512)(C), \
2695 (__mmask16)(U), (int)(R)))
2698#define _mm512_fnmadd_round_ps(A, B, C, R) \
2699 ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
2700 -(__v16sf)(__m512)(B), \
2701 (__v16sf)(__m512)(C), \
2702 (__mmask16)-1, (int)(R)))
2705#define _mm512_mask3_fnmadd_round_ps(A, B, C, U, R) \
2706 ((__m512)__builtin_ia32_vfmaddps512_mask3(-(__v16sf)(__m512)(A), \
2707 (__v16sf)(__m512)(B), \
2708 (__v16sf)(__m512)(C), \
2709 (__mmask16)(U), (int)(R)))
2712#define _mm512_maskz_fnmadd_round_ps(U, A, B, C, R) \
2713 ((__m512)__builtin_ia32_vfmaddps512_maskz(-(__v16sf)(__m512)(A), \
2714 (__v16sf)(__m512)(B), \
2715 (__v16sf)(__m512)(C), \
2716 (__mmask16)(U), (int)(R)))
2719#define _mm512_fnmsub_round_ps(A, B, C, R) \
2720 ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
2721 -(__v16sf)(__m512)(B), \
2722 -(__v16sf)(__m512)(C), \
2723 (__mmask16)-1, (int)(R)))
2726#define _mm512_maskz_fnmsub_round_ps(U, A, B, C, R) \
2727 ((__m512)__builtin_ia32_vfmaddps512_maskz(-(__v16sf)(__m512)(A), \
2728 (__v16sf)(__m512)(B), \
2729 -(__v16sf)(__m512)(C), \
2730 (__mmask16)(U), (int)(R)))
2736 return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
2746 return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
2756 return (__m512) __builtin_ia32_vfmaddps512_mask3 ((__v16sf) __A,
2766 return (__m512) __builtin_ia32_vfmaddps512_maskz ((__v16sf) __A,
2776 return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
2786 return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
2796 return (__m512) __builtin_ia32_vfmaddps512_maskz ((__v16sf) __A,
2806 return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
2816 return (__m512) __builtin_ia32_vfmaddps512_mask3 (-(__v16sf) __A,
2826 return (__m512) __builtin_ia32_vfmaddps512_maskz (-(__v16sf) __A,
2836 return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
2846 return (__m512) __builtin_ia32_vfmaddps512_maskz (-(__v16sf) __A,
2853#define _mm512_fmaddsub_round_pd(A, B, C, R) \
2854 ((__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
2855 (__v8df)(__m512d)(B), \
2856 (__v8df)(__m512d)(C), \
2857 (__mmask8)-1, (int)(R)))
2860#define _mm512_mask_fmaddsub_round_pd(A, U, B, C, R) \
2861 ((__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
2862 (__v8df)(__m512d)(B), \
2863 (__v8df)(__m512d)(C), \
2864 (__mmask8)(U), (int)(R)))
2867#define _mm512_mask3_fmaddsub_round_pd(A, B, C, U, R) \
2868 ((__m512d)__builtin_ia32_vfmaddsubpd512_mask3((__v8df)(__m512d)(A), \
2869 (__v8df)(__m512d)(B), \
2870 (__v8df)(__m512d)(C), \
2871 (__mmask8)(U), (int)(R)))
2874#define _mm512_maskz_fmaddsub_round_pd(U, A, B, C, R) \
2875 ((__m512d)__builtin_ia32_vfmaddsubpd512_maskz((__v8df)(__m512d)(A), \
2876 (__v8df)(__m512d)(B), \
2877 (__v8df)(__m512d)(C), \
2878 (__mmask8)(U), (int)(R)))
2881#define _mm512_fmsubadd_round_pd(A, B, C, R) \
2882 ((__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
2883 (__v8df)(__m512d)(B), \
2884 -(__v8df)(__m512d)(C), \
2885 (__mmask8)-1, (int)(R)))
2888#define _mm512_mask_fmsubadd_round_pd(A, U, B, C, R) \
2889 ((__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
2890 (__v8df)(__m512d)(B), \
2891 -(__v8df)(__m512d)(C), \
2892 (__mmask8)(U), (int)(R)))
2895#define _mm512_maskz_fmsubadd_round_pd(U, A, B, C, R) \
2896 ((__m512d)__builtin_ia32_vfmaddsubpd512_maskz((__v8df)(__m512d)(A), \
2897 (__v8df)(__m512d)(B), \
2898 -(__v8df)(__m512d)(C), \
2899 (__mmask8)(U), (int)(R)))
2905 return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
2915 return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
2925 return (__m512d) __builtin_ia32_vfmaddsubpd512_mask3 ((__v8df) __A,
2935 return (__m512d) __builtin_ia32_vfmaddsubpd512_maskz ((__v8df) __A,
2945 return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
2955 return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
2965 return (__m512d) __builtin_ia32_vfmaddsubpd512_maskz ((__v8df) __A,
2972#define _mm512_fmaddsub_round_ps(A, B, C, R) \
2973 ((__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
2974 (__v16sf)(__m512)(B), \
2975 (__v16sf)(__m512)(C), \
2976 (__mmask16)-1, (int)(R)))
2979#define _mm512_mask_fmaddsub_round_ps(A, U, B, C, R) \
2980 ((__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
2981 (__v16sf)(__m512)(B), \
2982 (__v16sf)(__m512)(C), \
2983 (__mmask16)(U), (int)(R)))
2986#define _mm512_mask3_fmaddsub_round_ps(A, B, C, U, R) \
2987 ((__m512)__builtin_ia32_vfmaddsubps512_mask3((__v16sf)(__m512)(A), \
2988 (__v16sf)(__m512)(B), \
2989 (__v16sf)(__m512)(C), \
2990 (__mmask16)(U), (int)(R)))
2993#define _mm512_maskz_fmaddsub_round_ps(U, A, B, C, R) \
2994 ((__m512)__builtin_ia32_vfmaddsubps512_maskz((__v16sf)(__m512)(A), \
2995 (__v16sf)(__m512)(B), \
2996 (__v16sf)(__m512)(C), \
2997 (__mmask16)(U), (int)(R)))
3000#define _mm512_fmsubadd_round_ps(A, B, C, R) \
3001 ((__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
3002 (__v16sf)(__m512)(B), \
3003 -(__v16sf)(__m512)(C), \
3004 (__mmask16)-1, (int)(R)))
3007#define _mm512_mask_fmsubadd_round_ps(A, U, B, C, R) \
3008 ((__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
3009 (__v16sf)(__m512)(B), \
3010 -(__v16sf)(__m512)(C), \
3011 (__mmask16)(U), (int)(R)))
3014#define _mm512_maskz_fmsubadd_round_ps(U, A, B, C, R) \
3015 ((__m512)__builtin_ia32_vfmaddsubps512_maskz((__v16sf)(__m512)(A), \
3016 (__v16sf)(__m512)(B), \
3017 -(__v16sf)(__m512)(C), \
3018 (__mmask16)(U), (int)(R)))
3024 return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
3034 return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
3044 return (__m512) __builtin_ia32_vfmaddsubps512_mask3 ((__v16sf) __A,
3054 return (__m512) __builtin_ia32_vfmaddsubps512_maskz ((__v16sf) __A,
3064 return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
3074 return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
3084 return (__m512) __builtin_ia32_vfmaddsubps512_maskz ((__v16sf) __A,
3091#define _mm512_mask3_fmsub_round_pd(A, B, C, U, R) \
3092 ((__m512d)__builtin_ia32_vfmsubpd512_mask3((__v8df)(__m512d)(A), \
3093 (__v8df)(__m512d)(B), \
3094 (__v8df)(__m512d)(C), \
3095 (__mmask8)(U), (int)(R)))
3101 return (__m512d)__builtin_ia32_vfmsubpd512_mask3 ((__v8df) __A,
3108#define _mm512_mask3_fmsub_round_ps(A, B, C, U, R) \
3109 ((__m512)__builtin_ia32_vfmsubps512_mask3((__v16sf)(__m512)(A), \
3110 (__v16sf)(__m512)(B), \
3111 (__v16sf)(__m512)(C), \
3112 (__mmask16)(U), (int)(R)))
3117 return (__m512)__builtin_ia32_vfmsubps512_mask3 ((__v16sf) __A,
3124#define _mm512_mask3_fmsubadd_round_pd(A, B, C, U, R) \
3125 ((__m512d)__builtin_ia32_vfmsubaddpd512_mask3((__v8df)(__m512d)(A), \
3126 (__v8df)(__m512d)(B), \
3127 (__v8df)(__m512d)(C), \
3128 (__mmask8)(U), (int)(R)))
3134 return (__m512d)__builtin_ia32_vfmsubaddpd512_mask3 ((__v8df) __A,
3141#define _mm512_mask3_fmsubadd_round_ps(A, B, C, U, R) \
3142 ((__m512)__builtin_ia32_vfmsubaddps512_mask3((__v16sf)(__m512)(A), \
3143 (__v16sf)(__m512)(B), \
3144 (__v16sf)(__m512)(C), \
3145 (__mmask16)(U), (int)(R)))
3151 return (__m512)__builtin_ia32_vfmsubaddps512_mask3 ((__v16sf) __A,
3158#define _mm512_mask_fnmadd_round_pd(A, U, B, C, R) \
3159 ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
3160 -(__v8df)(__m512d)(B), \
3161 (__v8df)(__m512d)(C), \
3162 (__mmask8)(U), (int)(R)))
3168 return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
3175#define _mm512_mask_fnmadd_round_ps(A, U, B, C, R) \
3176 ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
3177 -(__v16sf)(__m512)(B), \
3178 (__v16sf)(__m512)(C), \
3179 (__mmask16)(U), (int)(R)))
3185 return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
3192#define _mm512_mask_fnmsub_round_pd(A, U, B, C, R) \
3193 ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
3194 -(__v8df)(__m512d)(B), \
3195 -(__v8df)(__m512d)(C), \
3196 (__mmask8)(U), (int)(R)))
3199#define _mm512_mask3_fnmsub_round_pd(A, B, C, U, R) \
3200 ((__m512d)__builtin_ia32_vfmsubpd512_mask3(-(__v8df)(__m512d)(A), \
3201 (__v8df)(__m512d)(B), \
3202 (__v8df)(__m512d)(C), \
3203 (__mmask8)(U), (int)(R)))
3209 return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
3219 return (__m512d) __builtin_ia32_vfmsubpd512_mask3 (-(__v8df) __A,
3226#define _mm512_mask_fnmsub_round_ps(A, U, B, C, R) \
3227 ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
3228 -(__v16sf)(__m512)(B), \
3229 -(__v16sf)(__m512)(C), \
3230 (__mmask16)(U), (int)(R)))
3233#define _mm512_mask3_fnmsub_round_ps(A, B, C, U, R) \
3234 ((__m512)__builtin_ia32_vfmsubps512_mask3(-(__v16sf)(__m512)(A), \
3235 (__v16sf)(__m512)(B), \
3236 (__v16sf)(__m512)(C), \
3237 (__mmask16)(U), (int)(R)))
3243 return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
3253 return (__m512) __builtin_ia32_vfmsubps512_mask3 (-(__v16sf) __A,
3267 return (__m512i)__builtin_ia32_vpermi2vard512((__v16si)__A, (__v16si) __I,
3275 return (__m512i)__builtin_ia32_selectd_512(__U,
3284 return (__m512i)__builtin_ia32_selectd_512(__U,
3293 return (__m512i)__builtin_ia32_selectd_512(__U,
3301 return (__m512i)__builtin_ia32_vpermi2varq512((__v8di)__A, (__v8di) __I,
3309 return (__m512i)__builtin_ia32_selectq_512(__U,
3318 return (__m512i)__builtin_ia32_selectq_512(__U,
3327 return (__m512i)__builtin_ia32_selectq_512(__U,
3332#define _mm512_alignr_epi64(A, B, I) \
3333 ((__m512i)__builtin_ia32_alignq512((__v8di)(__m512i)(A), \
3334 (__v8di)(__m512i)(B), (int)(I)))
3336#define _mm512_mask_alignr_epi64(W, U, A, B, imm) \
3337 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
3338 (__v8di)_mm512_alignr_epi64((A), (B), (imm)), \
3339 (__v8di)(__m512i)(W)))
3341#define _mm512_maskz_alignr_epi64(U, A, B, imm) \
3342 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
3343 (__v8di)_mm512_alignr_epi64((A), (B), (imm)), \
3344 (__v8di)_mm512_setzero_si512()))
3346#define _mm512_alignr_epi32(A, B, I) \
3347 ((__m512i)__builtin_ia32_alignd512((__v16si)(__m512i)(A), \
3348 (__v16si)(__m512i)(B), (int)(I)))
3350#define _mm512_mask_alignr_epi32(W, U, A, B, imm) \
3351 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
3352 (__v16si)_mm512_alignr_epi32((A), (B), (imm)), \
3353 (__v16si)(__m512i)(W)))
3355#define _mm512_maskz_alignr_epi32(U, A, B, imm) \
3356 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
3357 (__v16si)_mm512_alignr_epi32((A), (B), (imm)), \
3358 (__v16si)_mm512_setzero_si512()))
3361#define _mm512_extractf64x4_pd(A, I) \
3362 ((__m256d)__builtin_ia32_extractf64x4_mask((__v8df)(__m512d)(A), (int)(I), \
3363 (__v4df)_mm256_undefined_pd(), \
3366#define _mm512_mask_extractf64x4_pd(W, U, A, imm) \
3367 ((__m256d)__builtin_ia32_extractf64x4_mask((__v8df)(__m512d)(A), (int)(imm), \
3368 (__v4df)(__m256d)(W), \
3371#define _mm512_maskz_extractf64x4_pd(U, A, imm) \
3372 ((__m256d)__builtin_ia32_extractf64x4_mask((__v8df)(__m512d)(A), (int)(imm), \
3373 (__v4df)_mm256_setzero_pd(), \
3376#define _mm512_extractf32x4_ps(A, I) \
3377 ((__m128)__builtin_ia32_extractf32x4_mask((__v16sf)(__m512)(A), (int)(I), \
3378 (__v4sf)_mm_undefined_ps(), \
3381#define _mm512_mask_extractf32x4_ps(W, U, A, imm) \
3382 ((__m128)__builtin_ia32_extractf32x4_mask((__v16sf)(__m512)(A), (int)(imm), \
3383 (__v4sf)(__m128)(W), \
3386#define _mm512_maskz_extractf32x4_ps(U, A, imm) \
3387 ((__m128)__builtin_ia32_extractf32x4_mask((__v16sf)(__m512)(A), (int)(imm), \
3388 (__v4sf)_mm_setzero_ps(), \
3396 return (__m512d) __builtin_ia32_selectpd_512 ((
__mmask8) __U,
3404 return (__m512) __builtin_ia32_selectps_512 ((
__mmask16) __U,
3412 return (__m512i) __builtin_ia32_selectq_512 ((
__mmask8) __U,
3420 return (__m512i) __builtin_ia32_selectd_512 ((
__mmask16) __U,
3427#define _mm512_cmp_round_ps_mask(A, B, P, R) \
3428 ((__mmask16)__builtin_ia32_cmpps512_mask((__v16sf)(__m512)(A), \
3429 (__v16sf)(__m512)(B), (int)(P), \
3430 (__mmask16)-1, (int)(R)))
3432#define _mm512_mask_cmp_round_ps_mask(U, A, B, P, R) \
3433 ((__mmask16)__builtin_ia32_cmpps512_mask((__v16sf)(__m512)(A), \
3434 (__v16sf)(__m512)(B), (int)(P), \
3435 (__mmask16)(U), (int)(R)))
3437#define _mm512_cmp_ps_mask(A, B, P) \
3438 _mm512_cmp_round_ps_mask((A), (B), (P), _MM_FROUND_CUR_DIRECTION)
3439#define _mm512_mask_cmp_ps_mask(U, A, B, P) \
3440 _mm512_mask_cmp_round_ps_mask((U), (A), (B), (P), _MM_FROUND_CUR_DIRECTION)
3442#define _mm512_cmpeq_ps_mask(A, B) \
3443 _mm512_cmp_ps_mask((A), (B), _CMP_EQ_OQ)
3444#define _mm512_mask_cmpeq_ps_mask(k, A, B) \
3445 _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_EQ_OQ)
3447#define _mm512_cmplt_ps_mask(A, B) \
3448 _mm512_cmp_ps_mask((A), (B), _CMP_LT_OS)
3449#define _mm512_mask_cmplt_ps_mask(k, A, B) \
3450 _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_LT_OS)
3452#define _mm512_cmple_ps_mask(A, B) \
3453 _mm512_cmp_ps_mask((A), (B), _CMP_LE_OS)
3454#define _mm512_mask_cmple_ps_mask(k, A, B) \
3455 _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_LE_OS)
3457#define _mm512_cmpunord_ps_mask(A, B) \
3458 _mm512_cmp_ps_mask((A), (B), _CMP_UNORD_Q)
3459#define _mm512_mask_cmpunord_ps_mask(k, A, B) \
3460 _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_UNORD_Q)
3462#define _mm512_cmpneq_ps_mask(A, B) \
3463 _mm512_cmp_ps_mask((A), (B), _CMP_NEQ_UQ)
3464#define _mm512_mask_cmpneq_ps_mask(k, A, B) \
3465 _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_NEQ_UQ)
3467#define _mm512_cmpnlt_ps_mask(A, B) \
3468 _mm512_cmp_ps_mask((A), (B), _CMP_NLT_US)
3469#define _mm512_mask_cmpnlt_ps_mask(k, A, B) \
3470 _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_NLT_US)
3472#define _mm512_cmpnle_ps_mask(A, B) \
3473 _mm512_cmp_ps_mask((A), (B), _CMP_NLE_US)
3474#define _mm512_mask_cmpnle_ps_mask(k, A, B) \
3475 _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_NLE_US)
3477#define _mm512_cmpord_ps_mask(A, B) \
3478 _mm512_cmp_ps_mask((A), (B), _CMP_ORD_Q)
3479#define _mm512_mask_cmpord_ps_mask(k, A, B) \
3480 _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_ORD_Q)
3482#define _mm512_cmp_round_pd_mask(A, B, P, R) \
3483 ((__mmask8)__builtin_ia32_cmppd512_mask((__v8df)(__m512d)(A), \
3484 (__v8df)(__m512d)(B), (int)(P), \
3485 (__mmask8)-1, (int)(R)))
3487#define _mm512_mask_cmp_round_pd_mask(U, A, B, P, R) \
3488 ((__mmask8)__builtin_ia32_cmppd512_mask((__v8df)(__m512d)(A), \
3489 (__v8df)(__m512d)(B), (int)(P), \
3490 (__mmask8)(U), (int)(R)))
3492#define _mm512_cmp_pd_mask(A, B, P) \
3493 _mm512_cmp_round_pd_mask((A), (B), (P), _MM_FROUND_CUR_DIRECTION)
3494#define _mm512_mask_cmp_pd_mask(U, A, B, P) \
3495 _mm512_mask_cmp_round_pd_mask((U), (A), (B), (P), _MM_FROUND_CUR_DIRECTION)
3497#define _mm512_cmpeq_pd_mask(A, B) \
3498 _mm512_cmp_pd_mask((A), (B), _CMP_EQ_OQ)
3499#define _mm512_mask_cmpeq_pd_mask(k, A, B) \
3500 _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_EQ_OQ)
3502#define _mm512_cmplt_pd_mask(A, B) \
3503 _mm512_cmp_pd_mask((A), (B), _CMP_LT_OS)
3504#define _mm512_mask_cmplt_pd_mask(k, A, B) \
3505 _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_LT_OS)
3507#define _mm512_cmple_pd_mask(A, B) \
3508 _mm512_cmp_pd_mask((A), (B), _CMP_LE_OS)
3509#define _mm512_mask_cmple_pd_mask(k, A, B) \
3510 _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_LE_OS)
3512#define _mm512_cmpunord_pd_mask(A, B) \
3513 _mm512_cmp_pd_mask((A), (B), _CMP_UNORD_Q)
3514#define _mm512_mask_cmpunord_pd_mask(k, A, B) \
3515 _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_UNORD_Q)
3517#define _mm512_cmpneq_pd_mask(A, B) \
3518 _mm512_cmp_pd_mask((A), (B), _CMP_NEQ_UQ)
3519#define _mm512_mask_cmpneq_pd_mask(k, A, B) \
3520 _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_NEQ_UQ)
3522#define _mm512_cmpnlt_pd_mask(A, B) \
3523 _mm512_cmp_pd_mask((A), (B), _CMP_NLT_US)
3524#define _mm512_mask_cmpnlt_pd_mask(k, A, B) \
3525 _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_NLT_US)
3527#define _mm512_cmpnle_pd_mask(A, B) \
3528 _mm512_cmp_pd_mask((A), (B), _CMP_NLE_US)
3529#define _mm512_mask_cmpnle_pd_mask(k, A, B) \
3530 _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_NLE_US)
3532#define _mm512_cmpord_pd_mask(A, B) \
3533 _mm512_cmp_pd_mask((A), (B), _CMP_ORD_Q)
3534#define _mm512_mask_cmpord_pd_mask(k, A, B) \
3535 _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_ORD_Q)
3539#define _mm512_cvtt_roundps_epu32(A, R) \
3540 ((__m512i)__builtin_ia32_cvttps2udq512_mask((__v16sf)(__m512)(A), \
3541 (__v16si)_mm512_undefined_epi32(), \
3542 (__mmask16)-1, (int)(R)))
3544#define _mm512_mask_cvtt_roundps_epu32(W, U, A, R) \
3545 ((__m512i)__builtin_ia32_cvttps2udq512_mask((__v16sf)(__m512)(A), \
3546 (__v16si)(__m512i)(W), \
3547 (__mmask16)(U), (int)(R)))
3549#define _mm512_maskz_cvtt_roundps_epu32(U, A, R) \
3550 ((__m512i)__builtin_ia32_cvttps2udq512_mask((__v16sf)(__m512)(A), \
3551 (__v16si)_mm512_setzero_si512(), \
3552 (__mmask16)(U), (int)(R)))
3558 return (__m512i) __builtin_ia32_cvttps2udq512_mask ((__v16sf) __A,
3568 return (__m512i) __builtin_ia32_cvttps2udq512_mask ((__v16sf) __A,
3577 return (__m512i) __builtin_ia32_cvttps2udq512_mask ((__v16sf) __A,
3583#define _mm512_cvt_roundepi32_ps(A, R) \
3584 ((__m512)__builtin_ia32_cvtdq2ps512_mask((__v16si)(__m512i)(A), \
3585 (__v16sf)_mm512_setzero_ps(), \
3586 (__mmask16)-1, (int)(R)))
3588#define _mm512_mask_cvt_roundepi32_ps(W, U, A, R) \
3589 ((__m512)__builtin_ia32_cvtdq2ps512_mask((__v16si)(__m512i)(A), \
3590 (__v16sf)(__m512)(W), \
3591 (__mmask16)(U), (int)(R)))
3593#define _mm512_maskz_cvt_roundepi32_ps(U, A, R) \
3594 ((__m512)__builtin_ia32_cvtdq2ps512_mask((__v16si)(__m512i)(A), \
3595 (__v16sf)_mm512_setzero_ps(), \
3596 (__mmask16)(U), (int)(R)))
3598#define _mm512_cvt_roundepu32_ps(A, R) \
3599 ((__m512)__builtin_ia32_cvtudq2ps512_mask((__v16si)(__m512i)(A), \
3600 (__v16sf)_mm512_setzero_ps(), \
3601 (__mmask16)-1, (int)(R)))
3603#define _mm512_mask_cvt_roundepu32_ps(W, U, A, R) \
3604 ((__m512)__builtin_ia32_cvtudq2ps512_mask((__v16si)(__m512i)(A), \
3605 (__v16sf)(__m512)(W), \
3606 (__mmask16)(U), (int)(R)))
3608#define _mm512_maskz_cvt_roundepu32_ps(U, A, R) \
3609 ((__m512)__builtin_ia32_cvtudq2ps512_mask((__v16si)(__m512i)(A), \
3610 (__v16sf)_mm512_setzero_ps(), \
3611 (__mmask16)(U), (int)(R)))
3616 return (__m512)__builtin_convertvector((__v16su)__A, __v16sf);
3622 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
3630 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
3638 return (__m512d)__builtin_convertvector((__v8si)__A, __v8df);
3644 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8) __U,
3652 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8) __U,
3672 return (__m512)__builtin_convertvector((__v16si)__A, __v16sf);
3678 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
3686 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
3694 return (__m512d)__builtin_convertvector((__v8su)__A, __v8df);
3700 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8) __U,
3708 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8) __U,
3725#define _mm512_cvt_roundpd_ps(A, R) \
3726 ((__m256)__builtin_ia32_cvtpd2ps512_mask((__v8df)(__m512d)(A), \
3727 (__v8sf)_mm256_setzero_ps(), \
3728 (__mmask8)-1, (int)(R)))
3730#define _mm512_mask_cvt_roundpd_ps(W, U, A, R) \
3731 ((__m256)__builtin_ia32_cvtpd2ps512_mask((__v8df)(__m512d)(A), \
3732 (__v8sf)(__m256)(W), (__mmask8)(U), \
3735#define _mm512_maskz_cvt_roundpd_ps(U, A, R) \
3736 ((__m256)__builtin_ia32_cvtpd2ps512_mask((__v8df)(__m512d)(A), \
3737 (__v8sf)_mm256_setzero_ps(), \
3738 (__mmask8)(U), (int)(R)))
3743 return (__m256) __builtin_ia32_cvtpd2ps512_mask ((__v8df) __A,
3752 return (__m256) __builtin_ia32_cvtpd2ps512_mask ((__v8df) __A,
3761 return (__m256) __builtin_ia32_cvtpd2ps512_mask ((__v8df) __A,
3770 return (__m512) __builtin_shufflevector((__v8sf)
_mm512_cvtpd_ps(__A),
3772 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
3778 return (__m512) __builtin_shufflevector (
3782 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
3785#define _mm512_cvt_roundps_ph(A, I) \
3786 ((__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
3787 (__v16hi)_mm256_undefined_si256(), \
3790#define _mm512_mask_cvt_roundps_ph(U, W, A, I) \
3791 ((__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
3792 (__v16hi)(__m256i)(U), \
3795#define _mm512_maskz_cvt_roundps_ph(W, A, I) \
3796 ((__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
3797 (__v16hi)_mm256_setzero_si256(), \
3800#define _mm512_cvtps_ph _mm512_cvt_roundps_ph
3801#define _mm512_mask_cvtps_ph _mm512_mask_cvt_roundps_ph
3802#define _mm512_maskz_cvtps_ph _mm512_maskz_cvt_roundps_ph
3804#define _mm512_cvt_roundph_ps(A, R) \
3805 ((__m512)__builtin_ia32_vcvtph2ps512_mask((__v16hi)(__m256i)(A), \
3806 (__v16sf)_mm512_undefined_ps(), \
3807 (__mmask16)-1, (int)(R)))
3809#define _mm512_mask_cvt_roundph_ps(W, U, A, R) \
3810 ((__m512)__builtin_ia32_vcvtph2ps512_mask((__v16hi)(__m256i)(A), \
3811 (__v16sf)(__m512)(W), \
3812 (__mmask16)(U), (int)(R)))
3814#define _mm512_maskz_cvt_roundph_ps(U, A, R) \
3815 ((__m512)__builtin_ia32_vcvtph2ps512_mask((__v16hi)(__m256i)(A), \
3816 (__v16sf)_mm512_setzero_ps(), \
3817 (__mmask16)(U), (int)(R)))
3823 return (__m512) __builtin_ia32_vcvtph2ps512_mask ((__v16hi) __A,
3833 return (__m512) __builtin_ia32_vcvtph2ps512_mask ((__v16hi) __A,
3842 return (__m512) __builtin_ia32_vcvtph2ps512_mask ((__v16hi) __A,
3848#define _mm512_cvtt_roundpd_epi32(A, R) \
3849 ((__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df)(__m512d)(A), \
3850 (__v8si)_mm256_setzero_si256(), \
3851 (__mmask8)-1, (int)(R)))
3853#define _mm512_mask_cvtt_roundpd_epi32(W, U, A, R) \
3854 ((__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df)(__m512d)(A), \
3855 (__v8si)(__m256i)(W), \
3856 (__mmask8)(U), (int)(R)))
3858#define _mm512_maskz_cvtt_roundpd_epi32(U, A, R) \
3859 ((__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df)(__m512d)(A), \
3860 (__v8si)_mm256_setzero_si256(), \
3861 (__mmask8)(U), (int)(R)))
3866 return (__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df)
__a,
3875 return (__m256i) __builtin_ia32_cvttpd2dq512_mask ((__v8df) __A,
3884 return (__m256i) __builtin_ia32_cvttpd2dq512_mask ((__v8df) __A,
3890#define _mm512_cvtt_roundps_epi32(A, R) \
3891 ((__m512i)__builtin_ia32_cvttps2dq512_mask((__v16sf)(__m512)(A), \
3892 (__v16si)_mm512_setzero_si512(), \
3893 (__mmask16)-1, (int)(R)))
3895#define _mm512_mask_cvtt_roundps_epi32(W, U, A, R) \
3896 ((__m512i)__builtin_ia32_cvttps2dq512_mask((__v16sf)(__m512)(A), \
3897 (__v16si)(__m512i)(W), \
3898 (__mmask16)(U), (int)(R)))
3900#define _mm512_maskz_cvtt_roundps_epi32(U, A, R) \
3901 ((__m512i)__builtin_ia32_cvttps2dq512_mask((__v16sf)(__m512)(A), \
3902 (__v16si)_mm512_setzero_si512(), \
3903 (__mmask16)(U), (int)(R)))
3909 __builtin_ia32_cvttps2dq512_mask((__v16sf)
__a,
3917 return (__m512i) __builtin_ia32_cvttps2dq512_mask ((__v16sf) __A,
3926 return (__m512i) __builtin_ia32_cvttps2dq512_mask ((__v16sf) __A,
3932#define _mm512_cvt_roundps_epi32(A, R) \
3933 ((__m512i)__builtin_ia32_cvtps2dq512_mask((__v16sf)(__m512)(A), \
3934 (__v16si)_mm512_setzero_si512(), \
3935 (__mmask16)-1, (int)(R)))
3937#define _mm512_mask_cvt_roundps_epi32(W, U, A, R) \
3938 ((__m512i)__builtin_ia32_cvtps2dq512_mask((__v16sf)(__m512)(A), \
3939 (__v16si)(__m512i)(W), \
3940 (__mmask16)(U), (int)(R)))
3942#define _mm512_maskz_cvt_roundps_epi32(U, A, R) \
3943 ((__m512i)__builtin_ia32_cvtps2dq512_mask((__v16sf)(__m512)(A), \
3944 (__v16si)_mm512_setzero_si512(), \
3945 (__mmask16)(U), (int)(R)))
3950 return (__m512i) __builtin_ia32_cvtps2dq512_mask ((__v16sf) __A,
3959 return (__m512i) __builtin_ia32_cvtps2dq512_mask ((__v16sf) __A,
3968 return (__m512i) __builtin_ia32_cvtps2dq512_mask ((__v16sf) __A,
3975#define _mm512_cvt_roundpd_epi32(A, R) \
3976 ((__m256i)__builtin_ia32_cvtpd2dq512_mask((__v8df)(__m512d)(A), \
3977 (__v8si)_mm256_setzero_si256(), \
3978 (__mmask8)-1, (int)(R)))
3980#define _mm512_mask_cvt_roundpd_epi32(W, U, A, R) \
3981 ((__m256i)__builtin_ia32_cvtpd2dq512_mask((__v8df)(__m512d)(A), \
3982 (__v8si)(__m256i)(W), \
3983 (__mmask8)(U), (int)(R)))
3985#define _mm512_maskz_cvt_roundpd_epi32(U, A, R) \
3986 ((__m256i)__builtin_ia32_cvtpd2dq512_mask((__v8df)(__m512d)(A), \
3987 (__v8si)_mm256_setzero_si256(), \
3988 (__mmask8)(U), (int)(R)))
3993 return (__m256i) __builtin_ia32_cvtpd2dq512_mask ((__v8df) __A,
4003 return (__m256i) __builtin_ia32_cvtpd2dq512_mask ((__v8df) __A,
4012 return (__m256i) __builtin_ia32_cvtpd2dq512_mask ((__v8df) __A,
4019#define _mm512_cvt_roundps_epu32(A, R) \
4020 ((__m512i)__builtin_ia32_cvtps2udq512_mask((__v16sf)(__m512)(A), \
4021 (__v16si)_mm512_setzero_si512(), \
4022 (__mmask16)-1, (int)(R)))
4024#define _mm512_mask_cvt_roundps_epu32(W, U, A, R) \
4025 ((__m512i)__builtin_ia32_cvtps2udq512_mask((__v16sf)(__m512)(A), \
4026 (__v16si)(__m512i)(W), \
4027 (__mmask16)(U), (int)(R)))
4029#define _mm512_maskz_cvt_roundps_epu32(U, A, R) \
4030 ((__m512i)__builtin_ia32_cvtps2udq512_mask((__v16sf)(__m512)(A), \
4031 (__v16si)_mm512_setzero_si512(), \
4032 (__mmask16)(U), (int)(R)))
4037 return (__m512i) __builtin_ia32_cvtps2udq512_mask ((__v16sf) __A,\
4047 return (__m512i) __builtin_ia32_cvtps2udq512_mask ((__v16sf) __A,
4056 return (__m512i) __builtin_ia32_cvtps2udq512_mask ((__v16sf) __A,
4063#define _mm512_cvt_roundpd_epu32(A, R) \
4064 ((__m256i)__builtin_ia32_cvtpd2udq512_mask((__v8df)(__m512d)(A), \
4065 (__v8si)_mm256_setzero_si256(), \
4066 (__mmask8)-1, (int)(R)))
4068#define _mm512_mask_cvt_roundpd_epu32(W, U, A, R) \
4069 ((__m256i)__builtin_ia32_cvtpd2udq512_mask((__v8df)(__m512d)(A), \
4070 (__v8si)(__m256i)(W), \
4071 (__mmask8)(U), (int)(R)))
4073#define _mm512_maskz_cvt_roundpd_epu32(U, A, R) \
4074 ((__m256i)__builtin_ia32_cvtpd2udq512_mask((__v8df)(__m512d)(A), \
4075 (__v8si)_mm256_setzero_si256(), \
4076 (__mmask8)(U), (int)(R)))
4081 return (__m256i) __builtin_ia32_cvtpd2udq512_mask ((__v8df) __A,
4091 return (__m256i) __builtin_ia32_cvtpd2udq512_mask ((__v8df) __A,
4100 return (__m256i) __builtin_ia32_cvtpd2udq512_mask ((__v8df) __A,
4124 return (__m512d)__builtin_shufflevector((__v8df)
__a, (__v8df)
__b,
4125 1, 9, 1+2, 9+2, 1+4, 9+4, 1+6, 9+6);
4131 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8) __U,
4139 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8) __U,
4147 return (__m512d)__builtin_shufflevector((__v8df)
__a, (__v8df)
__b,
4148 0, 8, 0+2, 8+2, 0+4, 8+4, 0+6, 8+6);
4154 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8) __U,
4162 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8) __U,
4170 return (__m512)__builtin_shufflevector((__v16sf)
__a, (__v16sf)
__b,
4172 2+4, 18+4, 3+4, 19+4,
4173 2+8, 18+8, 3+8, 19+8,
4174 2+12, 18+12, 3+12, 19+12);
4180 return (__m512)__builtin_ia32_selectps_512((
__mmask16) __U,
4188 return (__m512)__builtin_ia32_selectps_512((
__mmask16) __U,
4196 return (__m512)__builtin_shufflevector((__v16sf)
__a, (__v16sf)
__b,
4198 0+4, 16+4, 1+4, 17+4,
4199 0+8, 16+8, 1+8, 17+8,
4200 0+12, 16+12, 1+12, 17+12);
4206 return (__m512)__builtin_ia32_selectps_512((
__mmask16) __U,
4214 return (__m512)__builtin_ia32_selectps_512((
__mmask16) __U,
4222 return (__m512i)__builtin_shufflevector((__v16si)__A, (__v16si)__B,
4224 2+4, 18+4, 3+4, 19+4,
4225 2+8, 18+8, 3+8, 19+8,
4226 2+12, 18+12, 3+12, 19+12);
4232 return (__m512i)__builtin_ia32_selectd_512((
__mmask16) __U,
4240 return (__m512i)__builtin_ia32_selectd_512((
__mmask16) __U,
4248 return (__m512i)__builtin_shufflevector((__v16si)__A, (__v16si)__B,
4250 0+4, 16+4, 1+4, 17+4,
4251 0+8, 16+8, 1+8, 17+8,
4252 0+12, 16+12, 1+12, 17+12);
4258 return (__m512i)__builtin_ia32_selectd_512((
__mmask16) __U,
4266 return (__m512i)__builtin_ia32_selectd_512((
__mmask16) __U,
4274 return (__m512i)__builtin_shufflevector((__v8di)__A, (__v8di)__B,
4275 1, 9, 1+2, 9+2, 1+4, 9+4, 1+6, 9+6);
4281 return (__m512i)__builtin_ia32_selectq_512((
__mmask8) __U,
4289 return (__m512i)__builtin_ia32_selectq_512((
__mmask8) __U,
4297 return (__m512i)__builtin_shufflevector((__v8di)__A, (__v8di)__B,
4298 0, 8, 0+2, 8+2, 0+4, 8+4, 0+6, 8+6);
4304 return (__m512i)__builtin_ia32_selectq_512((
__mmask8) __U,
4312 return (__m512i)__builtin_ia32_selectq_512((
__mmask8) __U,
4323 struct __loadu_si512 {
4326 return ((
const struct __loadu_si512*)
__P)->__v;
4332 struct __loadu_epi32 {
4335 return ((
const struct __loadu_epi32*)
__P)->__v;
4341 return (__m512i) __builtin_ia32_loaddqusi512_mask ((
const int *)
__P,
4350 return (__m512i) __builtin_ia32_loaddqusi512_mask ((
const int *)
__P,
4359 struct __loadu_epi64 {
4362 return ((
const struct __loadu_epi64*)
__P)->__v;
4368 return (__m512i) __builtin_ia32_loaddqudi512_mask ((
const long long *)
__P,
4376 return (__m512i) __builtin_ia32_loaddqudi512_mask ((
const long long *)
__P,
4385 return (__m512) __builtin_ia32_loadups512_mask ((
const float *)
__P,
4393 return (__m512) __builtin_ia32_loadups512_mask ((
const float *)
__P,
4402 return (__m512d) __builtin_ia32_loadupd512_mask ((
const double *)
__P,
4410 return (__m512d) __builtin_ia32_loadupd512_mask ((
const double *)
__P,
4422 return ((
const struct __loadu_pd*)
__p)->__v;
4431 return ((
const struct __loadu_ps*)
__p)->__v;
4437 return *(
const __m512*)
__p;
4443 return (__m512) __builtin_ia32_loadaps512_mask ((
const __v16sf *)
__P,
4451 return (__m512) __builtin_ia32_loadaps512_mask ((
const __v16sf *)
__P,
4460 return *(
const __m512d*)
__p;
4466 return (__m512d) __builtin_ia32_loadapd512_mask ((
const __v8df *)
__P,
4474 return (__m512d) __builtin_ia32_loadapd512_mask ((
const __v8df *)
__P,
4483 return *(
const __m512i *)
__P;
4489 return *(
const __m512i *)
__P;
4495 return *(
const __m512i *)
__P;
4503 struct __storeu_epi64 {
4506 ((
struct __storeu_epi64*)
__P)->__v = __A;
4512 __builtin_ia32_storedqudi512_mask ((
long long *)
__P, (__v8di) __A,
4519 struct __storeu_si512 {
4522 ((
struct __storeu_si512*)
__P)->__v = __A;
4528 struct __storeu_epi32 {
4531 ((
struct __storeu_epi32*)
__P)->__v = __A;
4537 __builtin_ia32_storedqusi512_mask ((
int *)
__P, (__v16si) __A,
4544 __builtin_ia32_storeupd512_mask ((
double *)
__P, (__v8df) __A, (
__mmask8) __U);
4550 struct __storeu_pd {
4553 ((
struct __storeu_pd*)
__P)->__v = __A;
4559 __builtin_ia32_storeups512_mask ((
float *)
__P, (__v16sf) __A,
4566 struct __storeu_ps {
4569 ((
struct __storeu_ps*)
__P)->__v = __A;
4575 __builtin_ia32_storeapd512_mask ((__v8df *)
__P, (__v8df) __A, (
__mmask8) __U);
4581 *(__m512d*)
__P = __A;
4587 __builtin_ia32_storeaps512_mask ((__v16sf *)
__P, (__v16sf) __A,
4594 *(__m512*)
__P = __A;
4600 *(__m512i *)
__P = __A;
4606 *(__m512i *)
__P = __A;
4612 *(__m512i *)
__P = __A;
4620 return __builtin_ia32_knothi(__M);
4625#define _mm512_cmpeq_epi32_mask(A, B) \
4626 _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_EQ)
4627#define _mm512_mask_cmpeq_epi32_mask(k, A, B) \
4628 _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_EQ)
4629#define _mm512_cmpge_epi32_mask(A, B) \
4630 _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_GE)
4631#define _mm512_mask_cmpge_epi32_mask(k, A, B) \
4632 _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_GE)
4633#define _mm512_cmpgt_epi32_mask(A, B) \
4634 _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_GT)
4635#define _mm512_mask_cmpgt_epi32_mask(k, A, B) \
4636 _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_GT)
4637#define _mm512_cmple_epi32_mask(A, B) \
4638 _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_LE)
4639#define _mm512_mask_cmple_epi32_mask(k, A, B) \
4640 _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_LE)
4641#define _mm512_cmplt_epi32_mask(A, B) \
4642 _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_LT)
4643#define _mm512_mask_cmplt_epi32_mask(k, A, B) \
4644 _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_LT)
4645#define _mm512_cmpneq_epi32_mask(A, B) \
4646 _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_NE)
4647#define _mm512_mask_cmpneq_epi32_mask(k, A, B) \
4648 _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_NE)
4650#define _mm512_cmpeq_epu32_mask(A, B) \
4651 _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_EQ)
4652#define _mm512_mask_cmpeq_epu32_mask(k, A, B) \
4653 _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_EQ)
4654#define _mm512_cmpge_epu32_mask(A, B) \
4655 _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_GE)
4656#define _mm512_mask_cmpge_epu32_mask(k, A, B) \
4657 _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_GE)
4658#define _mm512_cmpgt_epu32_mask(A, B) \
4659 _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_GT)
4660#define _mm512_mask_cmpgt_epu32_mask(k, A, B) \
4661 _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_GT)
4662#define _mm512_cmple_epu32_mask(A, B) \
4663 _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_LE)
4664#define _mm512_mask_cmple_epu32_mask(k, A, B) \
4665 _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_LE)
4666#define _mm512_cmplt_epu32_mask(A, B) \
4667 _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_LT)
4668#define _mm512_mask_cmplt_epu32_mask(k, A, B) \
4669 _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_LT)
4670#define _mm512_cmpneq_epu32_mask(A, B) \
4671 _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_NE)
4672#define _mm512_mask_cmpneq_epu32_mask(k, A, B) \
4673 _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_NE)
4675#define _mm512_cmpeq_epi64_mask(A, B) \
4676 _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_EQ)
4677#define _mm512_mask_cmpeq_epi64_mask(k, A, B) \
4678 _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_EQ)
4679#define _mm512_cmpge_epi64_mask(A, B) \
4680 _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_GE)
4681#define _mm512_mask_cmpge_epi64_mask(k, A, B) \
4682 _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_GE)
4683#define _mm512_cmpgt_epi64_mask(A, B) \
4684 _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_GT)
4685#define _mm512_mask_cmpgt_epi64_mask(k, A, B) \
4686 _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_GT)
4687#define _mm512_cmple_epi64_mask(A, B) \
4688 _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_LE)
4689#define _mm512_mask_cmple_epi64_mask(k, A, B) \
4690 _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_LE)
4691#define _mm512_cmplt_epi64_mask(A, B) \
4692 _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_LT)
4693#define _mm512_mask_cmplt_epi64_mask(k, A, B) \
4694 _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_LT)
4695#define _mm512_cmpneq_epi64_mask(A, B) \
4696 _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_NE)
4697#define _mm512_mask_cmpneq_epi64_mask(k, A, B) \
4698 _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_NE)
4700#define _mm512_cmpeq_epu64_mask(A, B) \
4701 _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_EQ)
4702#define _mm512_mask_cmpeq_epu64_mask(k, A, B) \
4703 _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_EQ)
4704#define _mm512_cmpge_epu64_mask(A, B) \
4705 _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_GE)
4706#define _mm512_mask_cmpge_epu64_mask(k, A, B) \
4707 _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_GE)
4708#define _mm512_cmpgt_epu64_mask(A, B) \
4709 _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_GT)
4710#define _mm512_mask_cmpgt_epu64_mask(k, A, B) \
4711 _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_GT)
4712#define _mm512_cmple_epu64_mask(A, B) \
4713 _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_LE)
4714#define _mm512_mask_cmple_epu64_mask(k, A, B) \
4715 _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_LE)
4716#define _mm512_cmplt_epu64_mask(A, B) \
4717 _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_LT)
4718#define _mm512_mask_cmplt_epu64_mask(k, A, B) \
4719 _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_LT)
4720#define _mm512_cmpneq_epu64_mask(A, B) \
4721 _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_NE)
4722#define _mm512_mask_cmpneq_epu64_mask(k, A, B) \
4723 _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_NE)
4730 return (__m512i)__builtin_convertvector((__v16qs)__A, __v16si);
4736 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
4744 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
4754 return (__m512i)__builtin_convertvector(__builtin_shufflevector((__v16qs)__A, (__v16qs)__A, 0, 1, 2, 3, 4, 5, 6, 7), __v8di);
4760 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4768 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4776 return (__m512i)__builtin_convertvector((__v8si)__X, __v8di);
4782 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4790 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4798 return (__m512i)__builtin_convertvector((__v16hi)__A, __v16si);
4804 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
4812 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
4820 return (__m512i)__builtin_convertvector((__v8hi)__A, __v8di);
4826 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4834 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4842 return (__m512i)__builtin_convertvector((__v16qu)__A, __v16si);
4848 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
4856 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
4864 return (__m512i)__builtin_convertvector(__builtin_shufflevector((__v16qu)__A, (__v16qu)__A, 0, 1, 2, 3, 4, 5, 6, 7), __v8di);
4870 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4878 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4886 return (__m512i)__builtin_convertvector((__v8su)__X, __v8di);
4892 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4900 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4908 return (__m512i)__builtin_convertvector((__v16hu)__A, __v16si);
4914 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
4922 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
4930 return (__m512i)__builtin_convertvector((__v8hu)__A, __v8di);
4936 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4944 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
4952 return (__m512i)__builtin_ia32_prorvd512((__v16si)__A, (__v16si)__B);
4958 return (__m512i)__builtin_ia32_selectd_512(__U,
4966 return (__m512i)__builtin_ia32_selectd_512(__U,
4974 return (__m512i)__builtin_ia32_prorvq512((__v8di)__A, (__v8di)__B);
4980 return (__m512i)__builtin_ia32_selectq_512(__U,
4988 return (__m512i)__builtin_ia32_selectq_512(__U,
4995#define _mm512_cmp_epi32_mask(a, b, p) \
4996 ((__mmask16)__builtin_ia32_cmpd512_mask((__v16si)(__m512i)(a), \
4997 (__v16si)(__m512i)(b), (int)(p), \
5000#define _mm512_cmp_epu32_mask(a, b, p) \
5001 ((__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)(__m512i)(a), \
5002 (__v16si)(__m512i)(b), (int)(p), \
5005#define _mm512_cmp_epi64_mask(a, b, p) \
5006 ((__mmask8)__builtin_ia32_cmpq512_mask((__v8di)(__m512i)(a), \
5007 (__v8di)(__m512i)(b), (int)(p), \
5010#define _mm512_cmp_epu64_mask(a, b, p) \
5011 ((__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)(__m512i)(a), \
5012 (__v8di)(__m512i)(b), (int)(p), \
5015#define _mm512_mask_cmp_epi32_mask(m, a, b, p) \
5016 ((__mmask16)__builtin_ia32_cmpd512_mask((__v16si)(__m512i)(a), \
5017 (__v16si)(__m512i)(b), (int)(p), \
5020#define _mm512_mask_cmp_epu32_mask(m, a, b, p) \
5021 ((__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)(__m512i)(a), \
5022 (__v16si)(__m512i)(b), (int)(p), \
5025#define _mm512_mask_cmp_epi64_mask(m, a, b, p) \
5026 ((__mmask8)__builtin_ia32_cmpq512_mask((__v8di)(__m512i)(a), \
5027 (__v8di)(__m512i)(b), (int)(p), \
5030#define _mm512_mask_cmp_epu64_mask(m, a, b, p) \
5031 ((__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)(__m512i)(a), \
5032 (__v8di)(__m512i)(b), (int)(p), \
5035#define _mm512_rol_epi32(a, b) \
5036 ((__m512i)__builtin_ia32_prold512((__v16si)(__m512i)(a), (int)(b)))
5038#define _mm512_mask_rol_epi32(W, U, a, b) \
5039 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
5040 (__v16si)_mm512_rol_epi32((a), (b)), \
5041 (__v16si)(__m512i)(W)))
5043#define _mm512_maskz_rol_epi32(U, a, b) \
5044 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
5045 (__v16si)_mm512_rol_epi32((a), (b)), \
5046 (__v16si)_mm512_setzero_si512()))
5048#define _mm512_rol_epi64(a, b) \
5049 ((__m512i)__builtin_ia32_prolq512((__v8di)(__m512i)(a), (int)(b)))
5051#define _mm512_mask_rol_epi64(W, U, a, b) \
5052 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
5053 (__v8di)_mm512_rol_epi64((a), (b)), \
5054 (__v8di)(__m512i)(W)))
5056#define _mm512_maskz_rol_epi64(U, a, b) \
5057 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
5058 (__v8di)_mm512_rol_epi64((a), (b)), \
5059 (__v8di)_mm512_setzero_si512()))
5064 return (__m512i)__builtin_ia32_prolvd512((__v16si)__A, (__v16si)__B);
5070 return (__m512i)__builtin_ia32_selectd_512(__U,
5078 return (__m512i)__builtin_ia32_selectd_512(__U,
5086 return (__m512i)__builtin_ia32_prolvq512((__v8di)__A, (__v8di)__B);
5092 return (__m512i)__builtin_ia32_selectq_512(__U,
5100 return (__m512i)__builtin_ia32_selectq_512(__U,
5105#define _mm512_ror_epi32(A, B) \
5106 ((__m512i)__builtin_ia32_prord512((__v16si)(__m512i)(A), (int)(B)))
5108#define _mm512_mask_ror_epi32(W, U, A, B) \
5109 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
5110 (__v16si)_mm512_ror_epi32((A), (B)), \
5111 (__v16si)(__m512i)(W)))
5113#define _mm512_maskz_ror_epi32(U, A, B) \
5114 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
5115 (__v16si)_mm512_ror_epi32((A), (B)), \
5116 (__v16si)_mm512_setzero_si512()))
5118#define _mm512_ror_epi64(A, B) \
5119 ((__m512i)__builtin_ia32_prorq512((__v8di)(__m512i)(A), (int)(B)))
5121#define _mm512_mask_ror_epi64(W, U, A, B) \
5122 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
5123 (__v8di)_mm512_ror_epi64((A), (B)), \
5124 (__v8di)(__m512i)(W)))
5126#define _mm512_maskz_ror_epi64(U, A, B) \
5127 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
5128 (__v8di)_mm512_ror_epi64((A), (B)), \
5129 (__v8di)_mm512_setzero_si512()))
5134 return (__m512i)__builtin_ia32_pslldi512((__v16si)__A, (
int)__B);
5141 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5148 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5156 return (__m512i)__builtin_ia32_psllqi512((__v8di)__A, (
int)__B);
5162 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5170 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5178 return (__m512i)__builtin_ia32_psrldi512((__v16si)__A, (
int)__B);
5185 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5192 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5200 return (__m512i)__builtin_ia32_psrlqi512((__v8di)__A, (
int)__B);
5207 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5216 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5224 return (__m512i) __builtin_ia32_movdqa32load512_mask ((
const __v16si *)
__P,
5232 return (__m512i) __builtin_ia32_movdqa32load512_mask ((
const __v16si *)
__P,
5241 __builtin_ia32_movdqa32store512_mask ((__v16si *)
__P, (__v16si) __A,
5248 return (__m512i) __builtin_ia32_selectd_512 ((
__mmask16) __U,
5256 return (__m512i) __builtin_ia32_selectd_512 ((
__mmask16) __U,
5264 return (__m512i) __builtin_ia32_selectq_512 ((
__mmask8) __U,
5272 return (__m512i) __builtin_ia32_selectq_512 ((
__mmask8) __U,
5280 return (__m512i) __builtin_ia32_movdqa64load512_mask ((
const __v8di *)
__P,
5288 return (__m512i) __builtin_ia32_movdqa64load512_mask ((
const __v8di *)
__P,
5297 __builtin_ia32_movdqa64store512_mask ((__v8di *)
__P, (__v8di) __A,
5304 return (__m512d)__builtin_shufflevector((__v8df)__A, (__v8df)__A,
5305 0, 0, 2, 2, 4, 4, 6, 6);
5311 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
5319 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
5324#define _mm512_fixupimm_round_pd(A, B, C, imm, R) \
5325 ((__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
5326 (__v8df)(__m512d)(B), \
5327 (__v8di)(__m512i)(C), (int)(imm), \
5328 (__mmask8)-1, (int)(R)))
5330#define _mm512_mask_fixupimm_round_pd(A, U, B, C, imm, R) \
5331 ((__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
5332 (__v8df)(__m512d)(B), \
5333 (__v8di)(__m512i)(C), (int)(imm), \
5334 (__mmask8)(U), (int)(R)))
5336#define _mm512_fixupimm_pd(A, B, C, imm) \
5337 ((__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
5338 (__v8df)(__m512d)(B), \
5339 (__v8di)(__m512i)(C), (int)(imm), \
5341 _MM_FROUND_CUR_DIRECTION))
5343#define _mm512_mask_fixupimm_pd(A, U, B, C, imm) \
5344 ((__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
5345 (__v8df)(__m512d)(B), \
5346 (__v8di)(__m512i)(C), (int)(imm), \
5348 _MM_FROUND_CUR_DIRECTION))
5350#define _mm512_maskz_fixupimm_round_pd(U, A, B, C, imm, R) \
5351 ((__m512d)__builtin_ia32_fixupimmpd512_maskz((__v8df)(__m512d)(A), \
5352 (__v8df)(__m512d)(B), \
5353 (__v8di)(__m512i)(C), \
5354 (int)(imm), (__mmask8)(U), \
5357#define _mm512_maskz_fixupimm_pd(U, A, B, C, imm) \
5358 ((__m512d)__builtin_ia32_fixupimmpd512_maskz((__v8df)(__m512d)(A), \
5359 (__v8df)(__m512d)(B), \
5360 (__v8di)(__m512i)(C), \
5361 (int)(imm), (__mmask8)(U), \
5362 _MM_FROUND_CUR_DIRECTION))
5364#define _mm512_fixupimm_round_ps(A, B, C, imm, R) \
5365 ((__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
5366 (__v16sf)(__m512)(B), \
5367 (__v16si)(__m512i)(C), (int)(imm), \
5368 (__mmask16)-1, (int)(R)))
5370#define _mm512_mask_fixupimm_round_ps(A, U, B, C, imm, R) \
5371 ((__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
5372 (__v16sf)(__m512)(B), \
5373 (__v16si)(__m512i)(C), (int)(imm), \
5374 (__mmask16)(U), (int)(R)))
5376#define _mm512_fixupimm_ps(A, B, C, imm) \
5377 ((__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
5378 (__v16sf)(__m512)(B), \
5379 (__v16si)(__m512i)(C), (int)(imm), \
5381 _MM_FROUND_CUR_DIRECTION))
5383#define _mm512_mask_fixupimm_ps(A, U, B, C, imm) \
5384 ((__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
5385 (__v16sf)(__m512)(B), \
5386 (__v16si)(__m512i)(C), (int)(imm), \
5388 _MM_FROUND_CUR_DIRECTION))
5390#define _mm512_maskz_fixupimm_round_ps(U, A, B, C, imm, R) \
5391 ((__m512)__builtin_ia32_fixupimmps512_maskz((__v16sf)(__m512)(A), \
5392 (__v16sf)(__m512)(B), \
5393 (__v16si)(__m512i)(C), \
5394 (int)(imm), (__mmask16)(U), \
5397#define _mm512_maskz_fixupimm_ps(U, A, B, C, imm) \
5398 ((__m512)__builtin_ia32_fixupimmps512_maskz((__v16sf)(__m512)(A), \
5399 (__v16sf)(__m512)(B), \
5400 (__v16si)(__m512i)(C), \
5401 (int)(imm), (__mmask16)(U), \
5402 _MM_FROUND_CUR_DIRECTION))
5404#define _mm_fixupimm_round_sd(A, B, C, imm, R) \
5405 ((__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
5406 (__v2df)(__m128d)(B), \
5407 (__v2di)(__m128i)(C), (int)(imm), \
5408 (__mmask8)-1, (int)(R)))
5410#define _mm_mask_fixupimm_round_sd(A, U, B, C, imm, R) \
5411 ((__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
5412 (__v2df)(__m128d)(B), \
5413 (__v2di)(__m128i)(C), (int)(imm), \
5414 (__mmask8)(U), (int)(R)))
5416#define _mm_fixupimm_sd(A, B, C, imm) \
5417 ((__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
5418 (__v2df)(__m128d)(B), \
5419 (__v2di)(__m128i)(C), (int)(imm), \
5421 _MM_FROUND_CUR_DIRECTION))
5423#define _mm_mask_fixupimm_sd(A, U, B, C, imm) \
5424 ((__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
5425 (__v2df)(__m128d)(B), \
5426 (__v2di)(__m128i)(C), (int)(imm), \
5428 _MM_FROUND_CUR_DIRECTION))
5430#define _mm_maskz_fixupimm_round_sd(U, A, B, C, imm, R) \
5431 ((__m128d)__builtin_ia32_fixupimmsd_maskz((__v2df)(__m128d)(A), \
5432 (__v2df)(__m128d)(B), \
5433 (__v2di)(__m128i)(C), (int)(imm), \
5434 (__mmask8)(U), (int)(R)))
5436#define _mm_maskz_fixupimm_sd(U, A, B, C, imm) \
5437 ((__m128d)__builtin_ia32_fixupimmsd_maskz((__v2df)(__m128d)(A), \
5438 (__v2df)(__m128d)(B), \
5439 (__v2di)(__m128i)(C), (int)(imm), \
5441 _MM_FROUND_CUR_DIRECTION))
5443#define _mm_fixupimm_round_ss(A, B, C, imm, R) \
5444 ((__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
5445 (__v4sf)(__m128)(B), \
5446 (__v4si)(__m128i)(C), (int)(imm), \
5447 (__mmask8)-1, (int)(R)))
5449#define _mm_mask_fixupimm_round_ss(A, U, B, C, imm, R) \
5450 ((__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
5451 (__v4sf)(__m128)(B), \
5452 (__v4si)(__m128i)(C), (int)(imm), \
5453 (__mmask8)(U), (int)(R)))
5455#define _mm_fixupimm_ss(A, B, C, imm) \
5456 ((__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
5457 (__v4sf)(__m128)(B), \
5458 (__v4si)(__m128i)(C), (int)(imm), \
5460 _MM_FROUND_CUR_DIRECTION))
5462#define _mm_mask_fixupimm_ss(A, U, B, C, imm) \
5463 ((__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
5464 (__v4sf)(__m128)(B), \
5465 (__v4si)(__m128i)(C), (int)(imm), \
5467 _MM_FROUND_CUR_DIRECTION))
5469#define _mm_maskz_fixupimm_round_ss(U, A, B, C, imm, R) \
5470 ((__m128)__builtin_ia32_fixupimmss_maskz((__v4sf)(__m128)(A), \
5471 (__v4sf)(__m128)(B), \
5472 (__v4si)(__m128i)(C), (int)(imm), \
5473 (__mmask8)(U), (int)(R)))
5475#define _mm_maskz_fixupimm_ss(U, A, B, C, imm) \
5476 ((__m128)__builtin_ia32_fixupimmss_maskz((__v4sf)(__m128)(A), \
5477 (__v4sf)(__m128)(B), \
5478 (__v4si)(__m128i)(C), (int)(imm), \
5480 _MM_FROUND_CUR_DIRECTION))
5482#define _mm_getexp_round_sd(A, B, R) \
5483 ((__m128d)__builtin_ia32_getexpsd128_round_mask((__v2df)(__m128d)(A), \
5484 (__v2df)(__m128d)(B), \
5485 (__v2df)_mm_setzero_pd(), \
5486 (__mmask8)-1, (int)(R)))
5492 return (__m128d) __builtin_ia32_getexpsd128_round_mask ((__v2df) __A,
5499 return (__m128d) __builtin_ia32_getexpsd128_round_mask ( (__v2df) __A,
5506#define _mm_mask_getexp_round_sd(W, U, A, B, R) \
5507 ((__m128d)__builtin_ia32_getexpsd128_round_mask((__v2df)(__m128d)(A), \
5508 (__v2df)(__m128d)(B), \
5509 (__v2df)(__m128d)(W), \
5510 (__mmask8)(U), (int)(R)))
5515 return (__m128d) __builtin_ia32_getexpsd128_round_mask ( (__v2df) __A,
5522#define _mm_maskz_getexp_round_sd(U, A, B, R) \
5523 ((__m128d)__builtin_ia32_getexpsd128_round_mask((__v2df)(__m128d)(A), \
5524 (__v2df)(__m128d)(B), \
5525 (__v2df)_mm_setzero_pd(), \
5526 (__mmask8)(U), (int)(R)))
5528#define _mm_getexp_round_ss(A, B, R) \
5529 ((__m128)__builtin_ia32_getexpss128_round_mask((__v4sf)(__m128)(A), \
5530 (__v4sf)(__m128)(B), \
5531 (__v4sf)_mm_setzero_ps(), \
5532 (__mmask8)-1, (int)(R)))
5537 return (__m128) __builtin_ia32_getexpss128_round_mask ((__v4sf) __A,
5544 return (__m128) __builtin_ia32_getexpss128_round_mask ((__v4sf) __A,
5551#define _mm_mask_getexp_round_ss(W, U, A, B, R) \
5552 ((__m128)__builtin_ia32_getexpss128_round_mask((__v4sf)(__m128)(A), \
5553 (__v4sf)(__m128)(B), \
5554 (__v4sf)(__m128)(W), \
5555 (__mmask8)(U), (int)(R)))
5560 return (__m128) __builtin_ia32_getexpss128_round_mask ((__v4sf) __A,
5567#define _mm_maskz_getexp_round_ss(U, A, B, R) \
5568 ((__m128)__builtin_ia32_getexpss128_round_mask((__v4sf)(__m128)(A), \
5569 (__v4sf)(__m128)(B), \
5570 (__v4sf)_mm_setzero_ps(), \
5571 (__mmask8)(U), (int)(R)))
5573#define _mm_getmant_round_sd(A, B, C, D, R) \
5574 ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
5575 (__v2df)(__m128d)(B), \
5576 (int)(((D)<<2) | (C)), \
5577 (__v2df)_mm_setzero_pd(), \
5578 (__mmask8)-1, (int)(R)))
5580#define _mm_getmant_sd(A, B, C, D) \
5581 ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
5582 (__v2df)(__m128d)(B), \
5583 (int)(((D)<<2) | (C)), \
5584 (__v2df)_mm_setzero_pd(), \
5586 _MM_FROUND_CUR_DIRECTION))
5588#define _mm_mask_getmant_sd(W, U, A, B, C, D) \
5589 ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
5590 (__v2df)(__m128d)(B), \
5591 (int)(((D)<<2) | (C)), \
5592 (__v2df)(__m128d)(W), \
5594 _MM_FROUND_CUR_DIRECTION))
5596#define _mm_mask_getmant_round_sd(W, U, A, B, C, D, R) \
5597 ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
5598 (__v2df)(__m128d)(B), \
5599 (int)(((D)<<2) | (C)), \
5600 (__v2df)(__m128d)(W), \
5601 (__mmask8)(U), (int)(R)))
5603#define _mm_maskz_getmant_sd(U, A, B, C, D) \
5604 ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
5605 (__v2df)(__m128d)(B), \
5606 (int)(((D)<<2) | (C)), \
5607 (__v2df)_mm_setzero_pd(), \
5609 _MM_FROUND_CUR_DIRECTION))
5611#define _mm_maskz_getmant_round_sd(U, A, B, C, D, R) \
5612 ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
5613 (__v2df)(__m128d)(B), \
5614 (int)(((D)<<2) | (C)), \
5615 (__v2df)_mm_setzero_pd(), \
5616 (__mmask8)(U), (int)(R)))
5618#define _mm_getmant_round_ss(A, B, C, D, R) \
5619 ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
5620 (__v4sf)(__m128)(B), \
5621 (int)(((D)<<2) | (C)), \
5622 (__v4sf)_mm_setzero_ps(), \
5623 (__mmask8)-1, (int)(R)))
5625#define _mm_getmant_ss(A, B, C, D) \
5626 ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
5627 (__v4sf)(__m128)(B), \
5628 (int)(((D)<<2) | (C)), \
5629 (__v4sf)_mm_setzero_ps(), \
5631 _MM_FROUND_CUR_DIRECTION))
5633#define _mm_mask_getmant_ss(W, U, A, B, C, D) \
5634 ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
5635 (__v4sf)(__m128)(B), \
5636 (int)(((D)<<2) | (C)), \
5637 (__v4sf)(__m128)(W), \
5639 _MM_FROUND_CUR_DIRECTION))
5641#define _mm_mask_getmant_round_ss(W, U, A, B, C, D, R) \
5642 ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
5643 (__v4sf)(__m128)(B), \
5644 (int)(((D)<<2) | (C)), \
5645 (__v4sf)(__m128)(W), \
5646 (__mmask8)(U), (int)(R)))
5648#define _mm_maskz_getmant_ss(U, A, B, C, D) \
5649 ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
5650 (__v4sf)(__m128)(B), \
5651 (int)(((D)<<2) | (C)), \
5652 (__v4sf)_mm_setzero_ps(), \
5654 _MM_FROUND_CUR_DIRECTION))
5656#define _mm_maskz_getmant_round_ss(U, A, B, C, D, R) \
5657 ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
5658 (__v4sf)(__m128)(B), \
5659 (int)(((D)<<2) | (C)), \
5660 (__v4sf)_mm_setzero_ps(), \
5661 (__mmask8)(U), (int)(R)))
5669#define _mm_comi_round_sd(A, B, P, R) \
5670 ((int)__builtin_ia32_vcomisd((__v2df)(__m128d)(A), (__v2df)(__m128d)(B), \
5671 (int)(P), (int)(R)))
5673#define _mm_comi_round_ss(A, B, P, R) \
5674 ((int)__builtin_ia32_vcomiss((__v4sf)(__m128)(A), (__v4sf)(__m128)(B), \
5675 (int)(P), (int)(R)))
5678#define _mm_cvt_roundsd_si64(A, R) \
5679 ((long long)__builtin_ia32_vcvtsd2si64((__v2df)(__m128d)(A), (int)(R)))
5685 return (__m512i)__builtin_ia32_pslld512((__v16si) __A, (__v4si)__B);
5691 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5699 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5707 return (__m512i)__builtin_ia32_psllq512((__v8di)__A, (__v2di)__B);
5713 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5721 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5729 return (__m512i)__builtin_ia32_psllv16si((__v16si)__X, (__v16si)
__Y);
5735 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5743 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5751 return (__m512i)__builtin_ia32_psllv8di((__v8di)__X, (__v8di)
__Y);
5757 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5765 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5773 return (__m512i)__builtin_ia32_psrad512((__v16si) __A, (__v4si)__B);
5779 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5787 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5795 return (__m512i)__builtin_ia32_psraq512((__v8di)__A, (__v2di)__B);
5801 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5809 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5817 return (__m512i)__builtin_ia32_psrav16si((__v16si)__X, (__v16si)
__Y);
5823 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5831 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5839 return (__m512i)__builtin_ia32_psrav8di((__v8di)__X, (__v8di)
__Y);
5845 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5853 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5861 return (__m512i)__builtin_ia32_psrld512((__v16si) __A, (__v4si)__B);
5867 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5875 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5883 return (__m512i)__builtin_ia32_psrlq512((__v8di)__A, (__v2di)__B);
5889 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5897 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5905 return (__m512i)__builtin_ia32_psrlv16si((__v16si)__X, (__v16si)
__Y);
5911 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5919 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
5927 return (__m512i)__builtin_ia32_psrlv8di((__v8di)__X, (__v8di)
__Y);
5933 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5941 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
5955#define _mm512_ternarylogic_epi32(A, B, C, imm) \
5956 ((__m512i)__builtin_ia32_pternlogd512_mask( \
5957 (__v16si)(__m512i)(A), (__v16si)(__m512i)(B), (__v16si)(__m512i)(C), \
5958 (unsigned char)(imm), (__mmask16)-1))
5960#define _mm512_mask_ternarylogic_epi32(A, U, B, C, imm) \
5961 ((__m512i)__builtin_ia32_pternlogd512_mask( \
5962 (__v16si)(__m512i)(A), (__v16si)(__m512i)(B), (__v16si)(__m512i)(C), \
5963 (unsigned char)(imm), (__mmask16)(U)))
5965#define _mm512_maskz_ternarylogic_epi32(U, A, B, C, imm) \
5966 ((__m512i)__builtin_ia32_pternlogd512_maskz( \
5967 (__v16si)(__m512i)(A), (__v16si)(__m512i)(B), (__v16si)(__m512i)(C), \
5968 (unsigned char)(imm), (__mmask16)(U)))
5970#define _mm512_ternarylogic_epi64(A, B, C, imm) \
5971 ((__m512i)__builtin_ia32_pternlogq512_mask( \
5972 (__v8di)(__m512i)(A), (__v8di)(__m512i)(B), (__v8di)(__m512i)(C), \
5973 (unsigned char)(imm), (__mmask8)-1))
5975#define _mm512_mask_ternarylogic_epi64(A, U, B, C, imm) \
5976 ((__m512i)__builtin_ia32_pternlogq512_mask( \
5977 (__v8di)(__m512i)(A), (__v8di)(__m512i)(B), (__v8di)(__m512i)(C), \
5978 (unsigned char)(imm), (__mmask8)(U)))
5980#define _mm512_maskz_ternarylogic_epi64(U, A, B, C, imm) \
5981 ((__m512i)__builtin_ia32_pternlogq512_maskz( \
5982 (__v8di)(__m512i)(A), (__v8di)(__m512i)(B), (__v8di)(__m512i)(C), \
5983 (unsigned char)(imm), (__mmask8)(U)))
5986#define _mm_cvt_roundsd_i64(A, R) \
5987 ((long long)__builtin_ia32_vcvtsd2si64((__v2df)(__m128d)(A), (int)(R)))
5990#define _mm_cvt_roundsd_si32(A, R) \
5991 ((int)__builtin_ia32_vcvtsd2si32((__v2df)(__m128d)(A), (int)(R)))
5993#define _mm_cvt_roundsd_i32(A, R) \
5994 ((int)__builtin_ia32_vcvtsd2si32((__v2df)(__m128d)(A), (int)(R)))
5996#define _mm_cvt_roundsd_u32(A, R) \
5997 ((unsigned int)__builtin_ia32_vcvtsd2usi32((__v2df)(__m128d)(A), (int)(R)))
6002 return (
unsigned) __builtin_ia32_vcvtsd2usi32 ((__v2df) __A,
6007#define _mm_cvt_roundsd_u64(A, R) \
6008 ((unsigned long long)__builtin_ia32_vcvtsd2usi64((__v2df)(__m128d)(A), \
6012_mm_cvtsd_u64 (__m128d __A)
6014 return (
unsigned long long) __builtin_ia32_vcvtsd2usi64 ((__v2df)
6020#define _mm_cvt_roundss_si32(A, R) \
6021 ((int)__builtin_ia32_vcvtss2si32((__v4sf)(__m128)(A), (int)(R)))
6023#define _mm_cvt_roundss_i32(A, R) \
6024 ((int)__builtin_ia32_vcvtss2si32((__v4sf)(__m128)(A), (int)(R)))
6027#define _mm_cvt_roundss_si64(A, R) \
6028 ((long long)__builtin_ia32_vcvtss2si64((__v4sf)(__m128)(A), (int)(R)))
6030#define _mm_cvt_roundss_i64(A, R) \
6031 ((long long)__builtin_ia32_vcvtss2si64((__v4sf)(__m128)(A), (int)(R)))
6034#define _mm_cvt_roundss_u32(A, R) \
6035 ((unsigned int)__builtin_ia32_vcvtss2usi32((__v4sf)(__m128)(A), (int)(R)))
6040 return (
unsigned) __builtin_ia32_vcvtss2usi32 ((__v4sf) __A,
6045#define _mm_cvt_roundss_u64(A, R) \
6046 ((unsigned long long)__builtin_ia32_vcvtss2usi64((__v4sf)(__m128)(A), \
6050_mm_cvtss_u64 (__m128 __A)
6052 return (
unsigned long long) __builtin_ia32_vcvtss2usi64 ((__v4sf)
6058#define _mm_cvtt_roundsd_i32(A, R) \
6059 ((int)__builtin_ia32_vcvttsd2si32((__v2df)(__m128d)(A), (int)(R)))
6061#define _mm_cvtt_roundsd_si32(A, R) \
6062 ((int)__builtin_ia32_vcvttsd2si32((__v2df)(__m128d)(A), (int)(R)))
6067 return (
int) __builtin_ia32_vcvttsd2si32 ((__v2df) __A,
6072#define _mm_cvtt_roundsd_si64(A, R) \
6073 ((long long)__builtin_ia32_vcvttsd2si64((__v2df)(__m128d)(A), (int)(R)))
6075#define _mm_cvtt_roundsd_i64(A, R) \
6076 ((long long)__builtin_ia32_vcvttsd2si64((__v2df)(__m128d)(A), (int)(R)))
6079_mm_cvttsd_i64 (__m128d __A)
6081 return (
long long) __builtin_ia32_vcvttsd2si64 ((__v2df) __A,
6086#define _mm_cvtt_roundsd_u32(A, R) \
6087 ((unsigned int)__builtin_ia32_vcvttsd2usi32((__v2df)(__m128d)(A), (int)(R)))
6092 return (
unsigned) __builtin_ia32_vcvttsd2usi32 ((__v2df) __A,
6097#define _mm_cvtt_roundsd_u64(A, R) \
6098 ((unsigned long long)__builtin_ia32_vcvttsd2usi64((__v2df)(__m128d)(A), \
6102_mm_cvttsd_u64 (__m128d __A)
6104 return (
unsigned long long) __builtin_ia32_vcvttsd2usi64 ((__v2df)
6110#define _mm_cvtt_roundss_i32(A, R) \
6111 ((int)__builtin_ia32_vcvttss2si32((__v4sf)(__m128)(A), (int)(R)))
6113#define _mm_cvtt_roundss_si32(A, R) \
6114 ((int)__builtin_ia32_vcvttss2si32((__v4sf)(__m128)(A), (int)(R)))
6119 return (
int) __builtin_ia32_vcvttss2si32 ((__v4sf) __A,
6124#define _mm_cvtt_roundss_i64(A, R) \
6125 ((long long)__builtin_ia32_vcvttss2si64((__v4sf)(__m128)(A), (int)(R)))
6127#define _mm_cvtt_roundss_si64(A, R) \
6128 ((long long)__builtin_ia32_vcvttss2si64((__v4sf)(__m128)(A), (int)(R)))
6131_mm_cvttss_i64 (__m128 __A)
6133 return (
long long) __builtin_ia32_vcvttss2si64 ((__v4sf) __A,
6138#define _mm_cvtt_roundss_u32(A, R) \
6139 ((unsigned int)__builtin_ia32_vcvttss2usi32((__v4sf)(__m128)(A), (int)(R)))
6144 return (
unsigned) __builtin_ia32_vcvttss2usi32 ((__v4sf) __A,
6149#define _mm_cvtt_roundss_u64(A, R) \
6150 ((unsigned long long)__builtin_ia32_vcvttss2usi64((__v4sf)(__m128)(A), \
6154_mm_cvttss_u64 (__m128 __A)
6156 return (
unsigned long long) __builtin_ia32_vcvttss2usi64 ((__v4sf)
6162#define _mm512_permute_pd(X, C) \
6163 ((__m512d)__builtin_ia32_vpermilpd512((__v8df)(__m512d)(X), (int)(C)))
6165#define _mm512_mask_permute_pd(W, U, X, C) \
6166 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
6167 (__v8df)_mm512_permute_pd((X), (C)), \
6168 (__v8df)(__m512d)(W)))
6170#define _mm512_maskz_permute_pd(U, X, C) \
6171 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
6172 (__v8df)_mm512_permute_pd((X), (C)), \
6173 (__v8df)_mm512_setzero_pd()))
6175#define _mm512_permute_ps(X, C) \
6176 ((__m512)__builtin_ia32_vpermilps512((__v16sf)(__m512)(X), (int)(C)))
6178#define _mm512_mask_permute_ps(W, U, X, C) \
6179 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
6180 (__v16sf)_mm512_permute_ps((X), (C)), \
6181 (__v16sf)(__m512)(W)))
6183#define _mm512_maskz_permute_ps(U, X, C) \
6184 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
6185 (__v16sf)_mm512_permute_ps((X), (C)), \
6186 (__v16sf)_mm512_setzero_ps()))
6191 return (__m512d)__builtin_ia32_vpermilvarpd512((__v8df)__A, (__v8di)__C);
6197 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
6205 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
6213 return (__m512)__builtin_ia32_vpermilvarps512((__v16sf)__A, (__v16si)__C);
6219 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
6227 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
6235 return (__m512d)__builtin_ia32_vpermi2varpd512((__v8df)__A, (__v8di)__I,
6242 return (__m512d)__builtin_ia32_selectpd_512(__U,
6251 return (__m512d)__builtin_ia32_selectpd_512(__U,
6253 (__v8df)(__m512d)__I);
6260 return (__m512d)__builtin_ia32_selectpd_512(__U,
6268 return (__m512)__builtin_ia32_vpermi2varps512((__v16sf)__A, (__v16si)__I,
6275 return (__m512)__builtin_ia32_selectps_512(__U,
6283 return (__m512)__builtin_ia32_selectps_512(__U,
6285 (__v16sf)(__m512)__I);
6291 return (__m512)__builtin_ia32_selectps_512(__U,
6297#define _mm512_cvtt_roundpd_epu32(A, R) \
6298 ((__m256i)__builtin_ia32_cvttpd2udq512_mask((__v8df)(__m512d)(A), \
6299 (__v8si)_mm256_undefined_si256(), \
6300 (__mmask8)-1, (int)(R)))
6302#define _mm512_mask_cvtt_roundpd_epu32(W, U, A, R) \
6303 ((__m256i)__builtin_ia32_cvttpd2udq512_mask((__v8df)(__m512d)(A), \
6304 (__v8si)(__m256i)(W), \
6305 (__mmask8)(U), (int)(R)))
6307#define _mm512_maskz_cvtt_roundpd_epu32(U, A, R) \
6308 ((__m256i)__builtin_ia32_cvttpd2udq512_mask((__v8df)(__m512d)(A), \
6309 (__v8si)_mm256_setzero_si256(), \
6310 (__mmask8)(U), (int)(R)))
6315 return (__m256i) __builtin_ia32_cvttpd2udq512_mask ((__v8df) __A,
6325 return (__m256i) __builtin_ia32_cvttpd2udq512_mask ((__v8df) __A,
6334 return (__m256i) __builtin_ia32_cvttpd2udq512_mask ((__v8df) __A,
6341#define _mm_roundscale_round_sd(A, B, imm, R) \
6342 ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
6343 (__v2df)(__m128d)(B), \
6344 (__v2df)_mm_setzero_pd(), \
6345 (__mmask8)-1, (int)(imm), \
6348#define _mm_roundscale_sd(A, B, imm) \
6349 ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
6350 (__v2df)(__m128d)(B), \
6351 (__v2df)_mm_setzero_pd(), \
6352 (__mmask8)-1, (int)(imm), \
6353 _MM_FROUND_CUR_DIRECTION))
6355#define _mm_mask_roundscale_sd(W, U, A, B, imm) \
6356 ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
6357 (__v2df)(__m128d)(B), \
6358 (__v2df)(__m128d)(W), \
6359 (__mmask8)(U), (int)(imm), \
6360 _MM_FROUND_CUR_DIRECTION))
6362#define _mm_mask_roundscale_round_sd(W, U, A, B, I, R) \
6363 ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
6364 (__v2df)(__m128d)(B), \
6365 (__v2df)(__m128d)(W), \
6366 (__mmask8)(U), (int)(I), \
6369#define _mm_maskz_roundscale_sd(U, A, B, I) \
6370 ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
6371 (__v2df)(__m128d)(B), \
6372 (__v2df)_mm_setzero_pd(), \
6373 (__mmask8)(U), (int)(I), \
6374 _MM_FROUND_CUR_DIRECTION))
6376#define _mm_maskz_roundscale_round_sd(U, A, B, I, R) \
6377 ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
6378 (__v2df)(__m128d)(B), \
6379 (__v2df)_mm_setzero_pd(), \
6380 (__mmask8)(U), (int)(I), \
6383#define _mm_roundscale_round_ss(A, B, imm, R) \
6384 ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
6385 (__v4sf)(__m128)(B), \
6386 (__v4sf)_mm_setzero_ps(), \
6387 (__mmask8)-1, (int)(imm), \
6390#define _mm_roundscale_ss(A, B, imm) \
6391 ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
6392 (__v4sf)(__m128)(B), \
6393 (__v4sf)_mm_setzero_ps(), \
6394 (__mmask8)-1, (int)(imm), \
6395 _MM_FROUND_CUR_DIRECTION))
6397#define _mm_mask_roundscale_ss(W, U, A, B, I) \
6398 ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
6399 (__v4sf)(__m128)(B), \
6400 (__v4sf)(__m128)(W), \
6401 (__mmask8)(U), (int)(I), \
6402 _MM_FROUND_CUR_DIRECTION))
6404#define _mm_mask_roundscale_round_ss(W, U, A, B, I, R) \
6405 ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
6406 (__v4sf)(__m128)(B), \
6407 (__v4sf)(__m128)(W), \
6408 (__mmask8)(U), (int)(I), \
6411#define _mm_maskz_roundscale_ss(U, A, B, I) \
6412 ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
6413 (__v4sf)(__m128)(B), \
6414 (__v4sf)_mm_setzero_ps(), \
6415 (__mmask8)(U), (int)(I), \
6416 _MM_FROUND_CUR_DIRECTION))
6418#define _mm_maskz_roundscale_round_ss(U, A, B, I, R) \
6419 ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
6420 (__v4sf)(__m128)(B), \
6421 (__v4sf)_mm_setzero_ps(), \
6422 (__mmask8)(U), (int)(I), \
6425#define _mm512_scalef_round_pd(A, B, R) \
6426 ((__m512d)__builtin_ia32_scalefpd512_mask((__v8df)(__m512d)(A), \
6427 (__v8df)(__m512d)(B), \
6428 (__v8df)_mm512_undefined_pd(), \
6429 (__mmask8)-1, (int)(R)))
6431#define _mm512_mask_scalef_round_pd(W, U, A, B, R) \
6432 ((__m512d)__builtin_ia32_scalefpd512_mask((__v8df)(__m512d)(A), \
6433 (__v8df)(__m512d)(B), \
6434 (__v8df)(__m512d)(W), \
6435 (__mmask8)(U), (int)(R)))
6437#define _mm512_maskz_scalef_round_pd(U, A, B, R) \
6438 ((__m512d)__builtin_ia32_scalefpd512_mask((__v8df)(__m512d)(A), \
6439 (__v8df)(__m512d)(B), \
6440 (__v8df)_mm512_setzero_pd(), \
6441 (__mmask8)(U), (int)(R)))
6446 return (__m512d) __builtin_ia32_scalefpd512_mask ((__v8df) __A,
6457 return (__m512d) __builtin_ia32_scalefpd512_mask ((__v8df) __A,
6467 return (__m512d) __builtin_ia32_scalefpd512_mask ((__v8df) __A,
6475#define _mm512_scalef_round_ps(A, B, R) \
6476 ((__m512)__builtin_ia32_scalefps512_mask((__v16sf)(__m512)(A), \
6477 (__v16sf)(__m512)(B), \
6478 (__v16sf)_mm512_undefined_ps(), \
6479 (__mmask16)-1, (int)(R)))
6481#define _mm512_mask_scalef_round_ps(W, U, A, B, R) \
6482 ((__m512)__builtin_ia32_scalefps512_mask((__v16sf)(__m512)(A), \
6483 (__v16sf)(__m512)(B), \
6484 (__v16sf)(__m512)(W), \
6485 (__mmask16)(U), (int)(R)))
6487#define _mm512_maskz_scalef_round_ps(U, A, B, R) \
6488 ((__m512)__builtin_ia32_scalefps512_mask((__v16sf)(__m512)(A), \
6489 (__v16sf)(__m512)(B), \
6490 (__v16sf)_mm512_setzero_ps(), \
6491 (__mmask16)(U), (int)(R)))
6496 return (__m512) __builtin_ia32_scalefps512_mask ((__v16sf) __A,
6507 return (__m512) __builtin_ia32_scalefps512_mask ((__v16sf) __A,
6517 return (__m512) __builtin_ia32_scalefps512_mask ((__v16sf) __A,
6525#define _mm_scalef_round_sd(A, B, R) \
6526 ((__m128d)__builtin_ia32_scalefsd_round_mask((__v2df)(__m128d)(A), \
6527 (__v2df)(__m128d)(B), \
6528 (__v2df)_mm_setzero_pd(), \
6529 (__mmask8)-1, (int)(R)))
6534 return (__m128d) __builtin_ia32_scalefsd_round_mask ((__v2df) __A,
6543 return (__m128d) __builtin_ia32_scalefsd_round_mask ( (__v2df) __A,
6550#define _mm_mask_scalef_round_sd(W, U, A, B, R) \
6551 ((__m128d)__builtin_ia32_scalefsd_round_mask((__v2df)(__m128d)(A), \
6552 (__v2df)(__m128d)(B), \
6553 (__v2df)(__m128d)(W), \
6554 (__mmask8)(U), (int)(R)))
6559 return (__m128d) __builtin_ia32_scalefsd_round_mask ( (__v2df) __A,
6566#define _mm_maskz_scalef_round_sd(U, A, B, R) \
6567 ((__m128d)__builtin_ia32_scalefsd_round_mask((__v2df)(__m128d)(A), \
6568 (__v2df)(__m128d)(B), \
6569 (__v2df)_mm_setzero_pd(), \
6570 (__mmask8)(U), (int)(R)))
6572#define _mm_scalef_round_ss(A, B, R) \
6573 ((__m128)__builtin_ia32_scalefss_round_mask((__v4sf)(__m128)(A), \
6574 (__v4sf)(__m128)(B), \
6575 (__v4sf)_mm_setzero_ps(), \
6576 (__mmask8)-1, (int)(R)))
6581 return (__m128) __builtin_ia32_scalefss_round_mask ((__v4sf) __A,
6590 return (__m128) __builtin_ia32_scalefss_round_mask ( (__v4sf) __A,
6597#define _mm_mask_scalef_round_ss(W, U, A, B, R) \
6598 ((__m128)__builtin_ia32_scalefss_round_mask((__v4sf)(__m128)(A), \
6599 (__v4sf)(__m128)(B), \
6600 (__v4sf)(__m128)(W), \
6601 (__mmask8)(U), (int)(R)))
6606 return (__m128) __builtin_ia32_scalefss_round_mask ( (__v4sf) __A,
6613#define _mm_maskz_scalef_round_ss(U, A, B, R) \
6614 ((__m128)__builtin_ia32_scalefss_round_mask((__v4sf)(__m128)(A), \
6615 (__v4sf)(__m128)(B), \
6616 (__v4sf)_mm_setzero_ps(), \
6623 return (__m512i)__builtin_ia32_psradi512((__v16si)__A, (
int)__B);
6630 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
6638 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__U,
6646 return (__m512i)__builtin_ia32_psraqi512((__v8di)__A, (
int)__B);
6652 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
6660 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__U,
6665#define _mm512_shuffle_f32x4(A, B, imm) \
6666 ((__m512)__builtin_ia32_shuf_f32x4((__v16sf)(__m512)(A), \
6667 (__v16sf)(__m512)(B), (int)(imm)))
6669#define _mm512_mask_shuffle_f32x4(W, U, A, B, imm) \
6670 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
6671 (__v16sf)_mm512_shuffle_f32x4((A), (B), (imm)), \
6672 (__v16sf)(__m512)(W)))
6674#define _mm512_maskz_shuffle_f32x4(U, A, B, imm) \
6675 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
6676 (__v16sf)_mm512_shuffle_f32x4((A), (B), (imm)), \
6677 (__v16sf)_mm512_setzero_ps()))
6679#define _mm512_shuffle_f64x2(A, B, imm) \
6680 ((__m512d)__builtin_ia32_shuf_f64x2((__v8df)(__m512d)(A), \
6681 (__v8df)(__m512d)(B), (int)(imm)))
6683#define _mm512_mask_shuffle_f64x2(W, U, A, B, imm) \
6684 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
6685 (__v8df)_mm512_shuffle_f64x2((A), (B), (imm)), \
6686 (__v8df)(__m512d)(W)))
6688#define _mm512_maskz_shuffle_f64x2(U, A, B, imm) \
6689 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
6690 (__v8df)_mm512_shuffle_f64x2((A), (B), (imm)), \
6691 (__v8df)_mm512_setzero_pd()))
6693#define _mm512_shuffle_i32x4(A, B, imm) \
6694 ((__m512i)__builtin_ia32_shuf_i32x4((__v16si)(__m512i)(A), \
6695 (__v16si)(__m512i)(B), (int)(imm)))
6697#define _mm512_mask_shuffle_i32x4(W, U, A, B, imm) \
6698 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
6699 (__v16si)_mm512_shuffle_i32x4((A), (B), (imm)), \
6700 (__v16si)(__m512i)(W)))
6702#define _mm512_maskz_shuffle_i32x4(U, A, B, imm) \
6703 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
6704 (__v16si)_mm512_shuffle_i32x4((A), (B), (imm)), \
6705 (__v16si)_mm512_setzero_si512()))
6707#define _mm512_shuffle_i64x2(A, B, imm) \
6708 ((__m512i)__builtin_ia32_shuf_i64x2((__v8di)(__m512i)(A), \
6709 (__v8di)(__m512i)(B), (int)(imm)))
6711#define _mm512_mask_shuffle_i64x2(W, U, A, B, imm) \
6712 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
6713 (__v8di)_mm512_shuffle_i64x2((A), (B), (imm)), \
6714 (__v8di)(__m512i)(W)))
6716#define _mm512_maskz_shuffle_i64x2(U, A, B, imm) \
6717 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
6718 (__v8di)_mm512_shuffle_i64x2((A), (B), (imm)), \
6719 (__v8di)_mm512_setzero_si512()))
6721#define _mm512_shuffle_pd(A, B, M) \
6722 ((__m512d)__builtin_ia32_shufpd512((__v8df)(__m512d)(A), \
6723 (__v8df)(__m512d)(B), (int)(M)))
6725#define _mm512_mask_shuffle_pd(W, U, A, B, M) \
6726 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
6727 (__v8df)_mm512_shuffle_pd((A), (B), (M)), \
6728 (__v8df)(__m512d)(W)))
6730#define _mm512_maskz_shuffle_pd(U, A, B, M) \
6731 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
6732 (__v8df)_mm512_shuffle_pd((A), (B), (M)), \
6733 (__v8df)_mm512_setzero_pd()))
6735#define _mm512_shuffle_ps(A, B, M) \
6736 ((__m512)__builtin_ia32_shufps512((__v16sf)(__m512)(A), \
6737 (__v16sf)(__m512)(B), (int)(M)))
6739#define _mm512_mask_shuffle_ps(W, U, A, B, M) \
6740 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
6741 (__v16sf)_mm512_shuffle_ps((A), (B), (M)), \
6742 (__v16sf)(__m512)(W)))
6744#define _mm512_maskz_shuffle_ps(U, A, B, M) \
6745 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
6746 (__v16sf)_mm512_shuffle_ps((A), (B), (M)), \
6747 (__v16sf)_mm512_setzero_ps()))
6749#define _mm_sqrt_round_sd(A, B, R) \
6750 ((__m128d)__builtin_ia32_sqrtsd_round_mask((__v2df)(__m128d)(A), \
6751 (__v2df)(__m128d)(B), \
6752 (__v2df)_mm_setzero_pd(), \
6753 (__mmask8)-1, (int)(R)))
6758 return (__m128d) __builtin_ia32_sqrtsd_round_mask ( (__v2df) __A,
6765#define _mm_mask_sqrt_round_sd(W, U, A, B, R) \
6766 ((__m128d)__builtin_ia32_sqrtsd_round_mask((__v2df)(__m128d)(A), \
6767 (__v2df)(__m128d)(B), \
6768 (__v2df)(__m128d)(W), \
6769 (__mmask8)(U), (int)(R)))
6774 return (__m128d) __builtin_ia32_sqrtsd_round_mask ( (__v2df) __A,
6781#define _mm_maskz_sqrt_round_sd(U, A, B, R) \
6782 ((__m128d)__builtin_ia32_sqrtsd_round_mask((__v2df)(__m128d)(A), \
6783 (__v2df)(__m128d)(B), \
6784 (__v2df)_mm_setzero_pd(), \
6785 (__mmask8)(U), (int)(R)))
6787#define _mm_sqrt_round_ss(A, B, R) \
6788 ((__m128)__builtin_ia32_sqrtss_round_mask((__v4sf)(__m128)(A), \
6789 (__v4sf)(__m128)(B), \
6790 (__v4sf)_mm_setzero_ps(), \
6791 (__mmask8)-1, (int)(R)))
6796 return (__m128) __builtin_ia32_sqrtss_round_mask ( (__v4sf) __A,
6803#define _mm_mask_sqrt_round_ss(W, U, A, B, R) \
6804 ((__m128)__builtin_ia32_sqrtss_round_mask((__v4sf)(__m128)(A), \
6805 (__v4sf)(__m128)(B), \
6806 (__v4sf)(__m128)(W), (__mmask8)(U), \
6812 return (__m128) __builtin_ia32_sqrtss_round_mask ( (__v4sf) __A,
6819#define _mm_maskz_sqrt_round_ss(U, A, B, R) \
6820 ((__m128)__builtin_ia32_sqrtss_round_mask((__v4sf)(__m128)(A), \
6821 (__v4sf)(__m128)(B), \
6822 (__v4sf)_mm_setzero_ps(), \
6823 (__mmask8)(U), (int)(R)))
6828 return (__m512)__builtin_shufflevector((__v4sf)__A, (__v4sf)__A,
6829 0, 1, 2, 3, 0, 1, 2, 3,
6830 0, 1, 2, 3, 0, 1, 2, 3);
6836 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__M,
6844 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__M,
6852 return (__m512d)__builtin_shufflevector((__v4df)__A, (__v4df)__A,
6853 0, 1, 2, 3, 0, 1, 2, 3);
6859 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__M,
6867 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__M,
6875 return (__m512i)__builtin_shufflevector((__v4si)__A, (__v4si)__A,
6876 0, 1, 2, 3, 0, 1, 2, 3,
6877 0, 1, 2, 3, 0, 1, 2, 3);
6883 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
6891 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
6899 return (__m512i)__builtin_shufflevector((__v4di)__A, (__v4di)__A,
6900 0, 1, 2, 3, 0, 1, 2, 3);
6906 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
6914 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
6922 return (__m512d)__builtin_ia32_selectpd_512(__M,
6930 return (__m512d)__builtin_ia32_selectpd_512(__M,
6938 return (__m512)__builtin_ia32_selectps_512(__M,
6946 return (__m512)__builtin_ia32_selectps_512(__M,
6954 return (__m128i) __builtin_ia32_pmovsdb512_mask ((__v16si) __A,
6962 return (__m128i) __builtin_ia32_pmovsdb512_mask ((__v16si) __A,
6963 (__v16qi) __O, __M);
6969 return (__m128i) __builtin_ia32_pmovsdb512_mask ((__v16si) __A,
6977 __builtin_ia32_pmovsdb512mem_mask ((__v16qi *)
__P, (__v16si) __A, __M);
6983 return (__m256i) __builtin_ia32_pmovsdw512_mask ((__v16si) __A,
6991 return (__m256i) __builtin_ia32_pmovsdw512_mask ((__v16si) __A,
6992 (__v16hi) __O, __M);
6998 return (__m256i) __builtin_ia32_pmovsdw512_mask ((__v16si) __A,
7006 __builtin_ia32_pmovsdw512mem_mask ((__v16hi*)
__P, (__v16si) __A, __M);
7012 return (__m128i) __builtin_ia32_pmovsqb512_mask ((__v8di) __A,
7020 return (__m128i) __builtin_ia32_pmovsqb512_mask ((__v8di) __A,
7021 (__v16qi) __O, __M);
7027 return (__m128i) __builtin_ia32_pmovsqb512_mask ((__v8di) __A,
7035 __builtin_ia32_pmovsqb512mem_mask ((__v16qi *)
__P, (__v8di) __A, __M);
7041 return (__m256i) __builtin_ia32_pmovsqd512_mask ((__v8di) __A,
7049 return (__m256i) __builtin_ia32_pmovsqd512_mask ((__v8di) __A,
7056 return (__m256i) __builtin_ia32_pmovsqd512_mask ((__v8di) __A,
7064 __builtin_ia32_pmovsqd512mem_mask ((__v8si *)
__P, (__v8di) __A, __M);
7070 return (__m128i) __builtin_ia32_pmovsqw512_mask ((__v8di) __A,
7078 return (__m128i) __builtin_ia32_pmovsqw512_mask ((__v8di) __A,
7085 return (__m128i) __builtin_ia32_pmovsqw512_mask ((__v8di) __A,
7093 __builtin_ia32_pmovsqw512mem_mask ((__v8hi *)
__P, (__v8di) __A, __M);
7099 return (__m128i) __builtin_ia32_pmovusdb512_mask ((__v16si) __A,
7107 return (__m128i) __builtin_ia32_pmovusdb512_mask ((__v16si) __A,
7115 return (__m128i) __builtin_ia32_pmovusdb512_mask ((__v16si) __A,
7123 __builtin_ia32_pmovusdb512mem_mask ((__v16qi *)
__P, (__v16si) __A, __M);
7129 return (__m256i) __builtin_ia32_pmovusdw512_mask ((__v16si) __A,
7137 return (__m256i) __builtin_ia32_pmovusdw512_mask ((__v16si) __A,
7145 return (__m256i) __builtin_ia32_pmovusdw512_mask ((__v16si) __A,
7153 __builtin_ia32_pmovusdw512mem_mask ((__v16hi*)
__P, (__v16si) __A, __M);
7159 return (__m128i) __builtin_ia32_pmovusqb512_mask ((__v8di) __A,
7167 return (__m128i) __builtin_ia32_pmovusqb512_mask ((__v8di) __A,
7175 return (__m128i) __builtin_ia32_pmovusqb512_mask ((__v8di) __A,
7183 __builtin_ia32_pmovusqb512mem_mask ((__v16qi *)
__P, (__v8di) __A, __M);
7189 return (__m256i) __builtin_ia32_pmovusqd512_mask ((__v8di) __A,
7197 return (__m256i) __builtin_ia32_pmovusqd512_mask ((__v8di) __A,
7204 return (__m256i) __builtin_ia32_pmovusqd512_mask ((__v8di) __A,
7212 __builtin_ia32_pmovusqd512mem_mask ((__v8si*)
__P, (__v8di) __A, __M);
7218 return (__m128i) __builtin_ia32_pmovusqw512_mask ((__v8di) __A,
7226 return (__m128i) __builtin_ia32_pmovusqw512_mask ((__v8di) __A,
7233 return (__m128i) __builtin_ia32_pmovusqw512_mask ((__v8di) __A,
7241 __builtin_ia32_pmovusqw512mem_mask ((__v8hi*)
__P, (__v8di) __A, __M);
7247 return (__m128i) __builtin_ia32_pmovdb512_mask ((__v16si) __A,
7255 return (__m128i) __builtin_ia32_pmovdb512_mask ((__v16si) __A,
7256 (__v16qi) __O, __M);
7262 return (__m128i) __builtin_ia32_pmovdb512_mask ((__v16si) __A,
7270 __builtin_ia32_pmovdb512mem_mask ((__v16qi *)
__P, (__v16si) __A, __M);
7276 return (__m256i) __builtin_ia32_pmovdw512_mask ((__v16si) __A,
7284 return (__m256i) __builtin_ia32_pmovdw512_mask ((__v16si) __A,
7285 (__v16hi) __O, __M);
7291 return (__m256i) __builtin_ia32_pmovdw512_mask ((__v16si) __A,
7299 __builtin_ia32_pmovdw512mem_mask ((__v16hi *)
__P, (__v16si) __A, __M);
7305 return (__m128i) __builtin_ia32_pmovqb512_mask ((__v8di) __A,
7313 return (__m128i) __builtin_ia32_pmovqb512_mask ((__v8di) __A,
7314 (__v16qi) __O, __M);
7320 return (__m128i) __builtin_ia32_pmovqb512_mask ((__v8di) __A,
7328 __builtin_ia32_pmovqb512mem_mask ((__v16qi *)
__P, (__v8di) __A, __M);
7334 return (__m256i) __builtin_ia32_pmovqd512_mask ((__v8di) __A,
7342 return (__m256i) __builtin_ia32_pmovqd512_mask ((__v8di) __A,
7349 return (__m256i) __builtin_ia32_pmovqd512_mask ((__v8di) __A,
7357 __builtin_ia32_pmovqd512mem_mask ((__v8si *)
__P, (__v8di) __A, __M);
7363 return (__m128i) __builtin_ia32_pmovqw512_mask ((__v8di) __A,
7371 return (__m128i) __builtin_ia32_pmovqw512_mask ((__v8di) __A,
7378 return (__m128i) __builtin_ia32_pmovqw512_mask ((__v8di) __A,
7386 __builtin_ia32_pmovqw512mem_mask ((__v8hi *)
__P, (__v8di) __A, __M);
7389#define _mm512_extracti32x4_epi32(A, imm) \
7390 ((__m128i)__builtin_ia32_extracti32x4_mask((__v16si)(__m512i)(A), (int)(imm), \
7391 (__v4si)_mm_undefined_si128(), \
7394#define _mm512_mask_extracti32x4_epi32(W, U, A, imm) \
7395 ((__m128i)__builtin_ia32_extracti32x4_mask((__v16si)(__m512i)(A), (int)(imm), \
7396 (__v4si)(__m128i)(W), \
7399#define _mm512_maskz_extracti32x4_epi32(U, A, imm) \
7400 ((__m128i)__builtin_ia32_extracti32x4_mask((__v16si)(__m512i)(A), (int)(imm), \
7401 (__v4si)_mm_setzero_si128(), \
7404#define _mm512_extracti64x4_epi64(A, imm) \
7405 ((__m256i)__builtin_ia32_extracti64x4_mask((__v8di)(__m512i)(A), (int)(imm), \
7406 (__v4di)_mm256_undefined_si256(), \
7409#define _mm512_mask_extracti64x4_epi64(W, U, A, imm) \
7410 ((__m256i)__builtin_ia32_extracti64x4_mask((__v8di)(__m512i)(A), (int)(imm), \
7411 (__v4di)(__m256i)(W), \
7414#define _mm512_maskz_extracti64x4_epi64(U, A, imm) \
7415 ((__m256i)__builtin_ia32_extracti64x4_mask((__v8di)(__m512i)(A), (int)(imm), \
7416 (__v4di)_mm256_setzero_si256(), \
7419#define _mm512_insertf64x4(A, B, imm) \
7420 ((__m512d)__builtin_ia32_insertf64x4((__v8df)(__m512d)(A), \
7421 (__v4df)(__m256d)(B), (int)(imm)))
7423#define _mm512_mask_insertf64x4(W, U, A, B, imm) \
7424 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
7425 (__v8df)_mm512_insertf64x4((A), (B), (imm)), \
7426 (__v8df)(__m512d)(W)))
7428#define _mm512_maskz_insertf64x4(U, A, B, imm) \
7429 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
7430 (__v8df)_mm512_insertf64x4((A), (B), (imm)), \
7431 (__v8df)_mm512_setzero_pd()))
7433#define _mm512_inserti64x4(A, B, imm) \
7434 ((__m512i)__builtin_ia32_inserti64x4((__v8di)(__m512i)(A), \
7435 (__v4di)(__m256i)(B), (int)(imm)))
7437#define _mm512_mask_inserti64x4(W, U, A, B, imm) \
7438 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
7439 (__v8di)_mm512_inserti64x4((A), (B), (imm)), \
7440 (__v8di)(__m512i)(W)))
7442#define _mm512_maskz_inserti64x4(U, A, B, imm) \
7443 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
7444 (__v8di)_mm512_inserti64x4((A), (B), (imm)), \
7445 (__v8di)_mm512_setzero_si512()))
7447#define _mm512_insertf32x4(A, B, imm) \
7448 ((__m512)__builtin_ia32_insertf32x4((__v16sf)(__m512)(A), \
7449 (__v4sf)(__m128)(B), (int)(imm)))
7451#define _mm512_mask_insertf32x4(W, U, A, B, imm) \
7452 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
7453 (__v16sf)_mm512_insertf32x4((A), (B), (imm)), \
7454 (__v16sf)(__m512)(W)))
7456#define _mm512_maskz_insertf32x4(U, A, B, imm) \
7457 ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
7458 (__v16sf)_mm512_insertf32x4((A), (B), (imm)), \
7459 (__v16sf)_mm512_setzero_ps()))
7461#define _mm512_inserti32x4(A, B, imm) \
7462 ((__m512i)__builtin_ia32_inserti32x4((__v16si)(__m512i)(A), \
7463 (__v4si)(__m128i)(B), (int)(imm)))
7465#define _mm512_mask_inserti32x4(W, U, A, B, imm) \
7466 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
7467 (__v16si)_mm512_inserti32x4((A), (B), (imm)), \
7468 (__v16si)(__m512i)(W)))
7470#define _mm512_maskz_inserti32x4(U, A, B, imm) \
7471 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
7472 (__v16si)_mm512_inserti32x4((A), (B), (imm)), \
7473 (__v16si)_mm512_setzero_si512()))
7475#define _mm512_getmant_round_pd(A, B, C, R) \
7476 ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
7477 (int)(((C)<<2) | (B)), \
7478 (__v8df)_mm512_undefined_pd(), \
7479 (__mmask8)-1, (int)(R)))
7481#define _mm512_mask_getmant_round_pd(W, U, A, B, C, R) \
7482 ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
7483 (int)(((C)<<2) | (B)), \
7484 (__v8df)(__m512d)(W), \
7485 (__mmask8)(U), (int)(R)))
7487#define _mm512_maskz_getmant_round_pd(U, A, B, C, R) \
7488 ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
7489 (int)(((C)<<2) | (B)), \
7490 (__v8df)_mm512_setzero_pd(), \
7491 (__mmask8)(U), (int)(R)))
7493#define _mm512_getmant_pd(A, B, C) \
7494 ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
7495 (int)(((C)<<2) | (B)), \
7496 (__v8df)_mm512_setzero_pd(), \
7498 _MM_FROUND_CUR_DIRECTION))
7500#define _mm512_mask_getmant_pd(W, U, A, B, C) \
7501 ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
7502 (int)(((C)<<2) | (B)), \
7503 (__v8df)(__m512d)(W), \
7505 _MM_FROUND_CUR_DIRECTION))
7507#define _mm512_maskz_getmant_pd(U, A, B, C) \
7508 ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
7509 (int)(((C)<<2) | (B)), \
7510 (__v8df)_mm512_setzero_pd(), \
7512 _MM_FROUND_CUR_DIRECTION))
7514#define _mm512_getmant_round_ps(A, B, C, R) \
7515 ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
7516 (int)(((C)<<2) | (B)), \
7517 (__v16sf)_mm512_undefined_ps(), \
7518 (__mmask16)-1, (int)(R)))
7520#define _mm512_mask_getmant_round_ps(W, U, A, B, C, R) \
7521 ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
7522 (int)(((C)<<2) | (B)), \
7523 (__v16sf)(__m512)(W), \
7524 (__mmask16)(U), (int)(R)))
7526#define _mm512_maskz_getmant_round_ps(U, A, B, C, R) \
7527 ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
7528 (int)(((C)<<2) | (B)), \
7529 (__v16sf)_mm512_setzero_ps(), \
7530 (__mmask16)(U), (int)(R)))
7532#define _mm512_getmant_ps(A, B, C) \
7533 ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
7534 (int)(((C)<<2)|(B)), \
7535 (__v16sf)_mm512_undefined_ps(), \
7537 _MM_FROUND_CUR_DIRECTION))
7539#define _mm512_mask_getmant_ps(W, U, A, B, C) \
7540 ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
7541 (int)(((C)<<2)|(B)), \
7542 (__v16sf)(__m512)(W), \
7544 _MM_FROUND_CUR_DIRECTION))
7546#define _mm512_maskz_getmant_ps(U, A, B, C) \
7547 ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
7548 (int)(((C)<<2)|(B)), \
7549 (__v16sf)_mm512_setzero_ps(), \
7551 _MM_FROUND_CUR_DIRECTION))
7553#define _mm512_getexp_round_pd(A, R) \
7554 ((__m512d)__builtin_ia32_getexppd512_mask((__v8df)(__m512d)(A), \
7555 (__v8df)_mm512_undefined_pd(), \
7556 (__mmask8)-1, (int)(R)))
7558#define _mm512_mask_getexp_round_pd(W, U, A, R) \
7559 ((__m512d)__builtin_ia32_getexppd512_mask((__v8df)(__m512d)(A), \
7560 (__v8df)(__m512d)(W), \
7561 (__mmask8)(U), (int)(R)))
7563#define _mm512_maskz_getexp_round_pd(U, A, R) \
7564 ((__m512d)__builtin_ia32_getexppd512_mask((__v8df)(__m512d)(A), \
7565 (__v8df)_mm512_setzero_pd(), \
7566 (__mmask8)(U), (int)(R)))
7571 return (__m512d) __builtin_ia32_getexppd512_mask ((__v8df) __A,
7580 return (__m512d) __builtin_ia32_getexppd512_mask ((__v8df) __A,
7589 return (__m512d) __builtin_ia32_getexppd512_mask ((__v8df) __A,
7595#define _mm512_getexp_round_ps(A, R) \
7596 ((__m512)__builtin_ia32_getexpps512_mask((__v16sf)(__m512)(A), \
7597 (__v16sf)_mm512_undefined_ps(), \
7598 (__mmask16)-1, (int)(R)))
7600#define _mm512_mask_getexp_round_ps(W, U, A, R) \
7601 ((__m512)__builtin_ia32_getexpps512_mask((__v16sf)(__m512)(A), \
7602 (__v16sf)(__m512)(W), \
7603 (__mmask16)(U), (int)(R)))
7605#define _mm512_maskz_getexp_round_ps(U, A, R) \
7606 ((__m512)__builtin_ia32_getexpps512_mask((__v16sf)(__m512)(A), \
7607 (__v16sf)_mm512_setzero_ps(), \
7608 (__mmask16)(U), (int)(R)))
7613 return (__m512) __builtin_ia32_getexpps512_mask ((__v16sf) __A,
7622 return (__m512) __builtin_ia32_getexpps512_mask ((__v16sf) __A,
7631 return (__m512) __builtin_ia32_getexpps512_mask ((__v16sf) __A,
7637#define _mm512_i64gather_ps(index, addr, scale) \
7638 ((__m256)__builtin_ia32_gatherdiv16sf((__v8sf)_mm256_undefined_ps(), \
7639 (void const *)(addr), \
7640 (__v8di)(__m512i)(index), (__mmask8)-1, \
7643#define _mm512_mask_i64gather_ps(v1_old, mask, index, addr, scale) \
7644 ((__m256)__builtin_ia32_gatherdiv16sf((__v8sf)(__m256)(v1_old),\
7645 (void const *)(addr), \
7646 (__v8di)(__m512i)(index), \
7647 (__mmask8)(mask), (int)(scale)))
7649#define _mm512_i64gather_epi32(index, addr, scale) \
7650 ((__m256i)__builtin_ia32_gatherdiv16si((__v8si)_mm256_undefined_si256(), \
7651 (void const *)(addr), \
7652 (__v8di)(__m512i)(index), \
7653 (__mmask8)-1, (int)(scale)))
7655#define _mm512_mask_i64gather_epi32(v1_old, mask, index, addr, scale) \
7656 ((__m256i)__builtin_ia32_gatherdiv16si((__v8si)(__m256i)(v1_old), \
7657 (void const *)(addr), \
7658 (__v8di)(__m512i)(index), \
7659 (__mmask8)(mask), (int)(scale)))
7661#define _mm512_i64gather_pd(index, addr, scale) \
7662 ((__m512d)__builtin_ia32_gatherdiv8df((__v8df)_mm512_undefined_pd(), \
7663 (void const *)(addr), \
7664 (__v8di)(__m512i)(index), (__mmask8)-1, \
7667#define _mm512_mask_i64gather_pd(v1_old, mask, index, addr, scale) \
7668 ((__m512d)__builtin_ia32_gatherdiv8df((__v8df)(__m512d)(v1_old), \
7669 (void const *)(addr), \
7670 (__v8di)(__m512i)(index), \
7671 (__mmask8)(mask), (int)(scale)))
7673#define _mm512_i64gather_epi64(index, addr, scale) \
7674 ((__m512i)__builtin_ia32_gatherdiv8di((__v8di)_mm512_undefined_epi32(), \
7675 (void const *)(addr), \
7676 (__v8di)(__m512i)(index), (__mmask8)-1, \
7679#define _mm512_mask_i64gather_epi64(v1_old, mask, index, addr, scale) \
7680 ((__m512i)__builtin_ia32_gatherdiv8di((__v8di)(__m512i)(v1_old), \
7681 (void const *)(addr), \
7682 (__v8di)(__m512i)(index), \
7683 (__mmask8)(mask), (int)(scale)))
7685#define _mm512_i32gather_ps(index, addr, scale) \
7686 ((__m512)__builtin_ia32_gathersiv16sf((__v16sf)_mm512_undefined_ps(), \
7687 (void const *)(addr), \
7688 (__v16si)(__m512)(index), \
7689 (__mmask16)-1, (int)(scale)))
7691#define _mm512_mask_i32gather_ps(v1_old, mask, index, addr, scale) \
7692 ((__m512)__builtin_ia32_gathersiv16sf((__v16sf)(__m512)(v1_old), \
7693 (void const *)(addr), \
7694 (__v16si)(__m512)(index), \
7695 (__mmask16)(mask), (int)(scale)))
7697#define _mm512_i32gather_epi32(index, addr, scale) \
7698 ((__m512i)__builtin_ia32_gathersiv16si((__v16si)_mm512_undefined_epi32(), \
7699 (void const *)(addr), \
7700 (__v16si)(__m512i)(index), \
7701 (__mmask16)-1, (int)(scale)))
7703#define _mm512_mask_i32gather_epi32(v1_old, mask, index, addr, scale) \
7704 ((__m512i)__builtin_ia32_gathersiv16si((__v16si)(__m512i)(v1_old), \
7705 (void const *)(addr), \
7706 (__v16si)(__m512i)(index), \
7707 (__mmask16)(mask), (int)(scale)))
7709#define _mm512_i32gather_pd(index, addr, scale) \
7710 ((__m512d)__builtin_ia32_gathersiv8df((__v8df)_mm512_undefined_pd(), \
7711 (void const *)(addr), \
7712 (__v8si)(__m256i)(index), (__mmask8)-1, \
7715#define _mm512_mask_i32gather_pd(v1_old, mask, index, addr, scale) \
7716 ((__m512d)__builtin_ia32_gathersiv8df((__v8df)(__m512d)(v1_old), \
7717 (void const *)(addr), \
7718 (__v8si)(__m256i)(index), \
7719 (__mmask8)(mask), (int)(scale)))
7721#define _mm512_i32gather_epi64(index, addr, scale) \
7722 ((__m512i)__builtin_ia32_gathersiv8di((__v8di)_mm512_undefined_epi32(), \
7723 (void const *)(addr), \
7724 (__v8si)(__m256i)(index), (__mmask8)-1, \
7727#define _mm512_mask_i32gather_epi64(v1_old, mask, index, addr, scale) \
7728 ((__m512i)__builtin_ia32_gathersiv8di((__v8di)(__m512i)(v1_old), \
7729 (void const *)(addr), \
7730 (__v8si)(__m256i)(index), \
7731 (__mmask8)(mask), (int)(scale)))
7733#define _mm512_i64scatter_ps(addr, index, v1, scale) \
7734 __builtin_ia32_scatterdiv16sf((void *)(addr), (__mmask8)-1, \
7735 (__v8di)(__m512i)(index), \
7736 (__v8sf)(__m256)(v1), (int)(scale))
7738#define _mm512_mask_i64scatter_ps(addr, mask, index, v1, scale) \
7739 __builtin_ia32_scatterdiv16sf((void *)(addr), (__mmask8)(mask), \
7740 (__v8di)(__m512i)(index), \
7741 (__v8sf)(__m256)(v1), (int)(scale))
7743#define _mm512_i64scatter_epi32(addr, index, v1, scale) \
7744 __builtin_ia32_scatterdiv16si((void *)(addr), (__mmask8)-1, \
7745 (__v8di)(__m512i)(index), \
7746 (__v8si)(__m256i)(v1), (int)(scale))
7748#define _mm512_mask_i64scatter_epi32(addr, mask, index, v1, scale) \
7749 __builtin_ia32_scatterdiv16si((void *)(addr), (__mmask8)(mask), \
7750 (__v8di)(__m512i)(index), \
7751 (__v8si)(__m256i)(v1), (int)(scale))
7753#define _mm512_i64scatter_pd(addr, index, v1, scale) \
7754 __builtin_ia32_scatterdiv8df((void *)(addr), (__mmask8)-1, \
7755 (__v8di)(__m512i)(index), \
7756 (__v8df)(__m512d)(v1), (int)(scale))
7758#define _mm512_mask_i64scatter_pd(addr, mask, index, v1, scale) \
7759 __builtin_ia32_scatterdiv8df((void *)(addr), (__mmask8)(mask), \
7760 (__v8di)(__m512i)(index), \
7761 (__v8df)(__m512d)(v1), (int)(scale))
7763#define _mm512_i64scatter_epi64(addr, index, v1, scale) \
7764 __builtin_ia32_scatterdiv8di((void *)(addr), (__mmask8)-1, \
7765 (__v8di)(__m512i)(index), \
7766 (__v8di)(__m512i)(v1), (int)(scale))
7768#define _mm512_mask_i64scatter_epi64(addr, mask, index, v1, scale) \
7769 __builtin_ia32_scatterdiv8di((void *)(addr), (__mmask8)(mask), \
7770 (__v8di)(__m512i)(index), \
7771 (__v8di)(__m512i)(v1), (int)(scale))
7773#define _mm512_i32scatter_ps(addr, index, v1, scale) \
7774 __builtin_ia32_scattersiv16sf((void *)(addr), (__mmask16)-1, \
7775 (__v16si)(__m512i)(index), \
7776 (__v16sf)(__m512)(v1), (int)(scale))
7778#define _mm512_mask_i32scatter_ps(addr, mask, index, v1, scale) \
7779 __builtin_ia32_scattersiv16sf((void *)(addr), (__mmask16)(mask), \
7780 (__v16si)(__m512i)(index), \
7781 (__v16sf)(__m512)(v1), (int)(scale))
7783#define _mm512_i32scatter_epi32(addr, index, v1, scale) \
7784 __builtin_ia32_scattersiv16si((void *)(addr), (__mmask16)-1, \
7785 (__v16si)(__m512i)(index), \
7786 (__v16si)(__m512i)(v1), (int)(scale))
7788#define _mm512_mask_i32scatter_epi32(addr, mask, index, v1, scale) \
7789 __builtin_ia32_scattersiv16si((void *)(addr), (__mmask16)(mask), \
7790 (__v16si)(__m512i)(index), \
7791 (__v16si)(__m512i)(v1), (int)(scale))
7793#define _mm512_i32scatter_pd(addr, index, v1, scale) \
7794 __builtin_ia32_scattersiv8df((void *)(addr), (__mmask8)-1, \
7795 (__v8si)(__m256i)(index), \
7796 (__v8df)(__m512d)(v1), (int)(scale))
7798#define _mm512_mask_i32scatter_pd(addr, mask, index, v1, scale) \
7799 __builtin_ia32_scattersiv8df((void *)(addr), (__mmask8)(mask), \
7800 (__v8si)(__m256i)(index), \
7801 (__v8df)(__m512d)(v1), (int)(scale))
7803#define _mm512_i32scatter_epi64(addr, index, v1, scale) \
7804 __builtin_ia32_scattersiv8di((void *)(addr), (__mmask8)-1, \
7805 (__v8si)(__m256i)(index), \
7806 (__v8di)(__m512i)(v1), (int)(scale))
7808#define _mm512_mask_i32scatter_epi64(addr, mask, index, v1, scale) \
7809 __builtin_ia32_scattersiv8di((void *)(addr), (__mmask8)(mask), \
7810 (__v8si)(__m256i)(index), \
7811 (__v8di)(__m512i)(v1), (int)(scale))
7816 return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
7823#define _mm_fmadd_round_ss(A, B, C, R) \
7824 ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
7825 (__v4sf)(__m128)(B), \
7826 (__v4sf)(__m128)(C), (__mmask8)-1, \
7829#define _mm_mask_fmadd_round_ss(W, U, A, B, R) \
7830 ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
7831 (__v4sf)(__m128)(A), \
7832 (__v4sf)(__m128)(B), (__mmask8)(U), \
7838 return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
7845#define _mm_maskz_fmadd_round_ss(U, A, B, C, R) \
7846 ((__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
7847 (__v4sf)(__m128)(B), \
7848 (__v4sf)(__m128)(C), (__mmask8)(U), \
7854 return __builtin_ia32_vfmaddss3_mask3((__v4sf)__W,
7861#define _mm_mask3_fmadd_round_ss(W, X, Y, U, R) \
7862 ((__m128)__builtin_ia32_vfmaddss3_mask3((__v4sf)(__m128)(W), \
7863 (__v4sf)(__m128)(X), \
7864 (__v4sf)(__m128)(Y), (__mmask8)(U), \
7870 return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
7877#define _mm_fmsub_round_ss(A, B, C, R) \
7878 ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
7879 (__v4sf)(__m128)(B), \
7880 -(__v4sf)(__m128)(C), (__mmask8)-1, \
7883#define _mm_mask_fmsub_round_ss(W, U, A, B, R) \
7884 ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
7885 (__v4sf)(__m128)(A), \
7886 -(__v4sf)(__m128)(B), (__mmask8)(U), \
7892 return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
7899#define _mm_maskz_fmsub_round_ss(U, A, B, C, R) \
7900 ((__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
7901 (__v4sf)(__m128)(B), \
7902 -(__v4sf)(__m128)(C), (__mmask8)(U), \
7908 return __builtin_ia32_vfmsubss3_mask3((__v4sf)__W,
7915#define _mm_mask3_fmsub_round_ss(W, X, Y, U, R) \
7916 ((__m128)__builtin_ia32_vfmsubss3_mask3((__v4sf)(__m128)(W), \
7917 (__v4sf)(__m128)(X), \
7918 (__v4sf)(__m128)(Y), (__mmask8)(U), \
7924 return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
7931#define _mm_fnmadd_round_ss(A, B, C, R) \
7932 ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
7933 -(__v4sf)(__m128)(B), \
7934 (__v4sf)(__m128)(C), (__mmask8)-1, \
7937#define _mm_mask_fnmadd_round_ss(W, U, A, B, R) \
7938 ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
7939 -(__v4sf)(__m128)(A), \
7940 (__v4sf)(__m128)(B), (__mmask8)(U), \
7946 return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
7953#define _mm_maskz_fnmadd_round_ss(U, A, B, C, R) \
7954 ((__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
7955 -(__v4sf)(__m128)(B), \
7956 (__v4sf)(__m128)(C), (__mmask8)(U), \
7962 return __builtin_ia32_vfmaddss3_mask3((__v4sf)__W,
7969#define _mm_mask3_fnmadd_round_ss(W, X, Y, U, R) \
7970 ((__m128)__builtin_ia32_vfmaddss3_mask3((__v4sf)(__m128)(W), \
7971 -(__v4sf)(__m128)(X), \
7972 (__v4sf)(__m128)(Y), (__mmask8)(U), \
7978 return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
7985#define _mm_fnmsub_round_ss(A, B, C, R) \
7986 ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
7987 -(__v4sf)(__m128)(B), \
7988 -(__v4sf)(__m128)(C), (__mmask8)-1, \
7991#define _mm_mask_fnmsub_round_ss(W, U, A, B, R) \
7992 ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
7993 -(__v4sf)(__m128)(A), \
7994 -(__v4sf)(__m128)(B), (__mmask8)(U), \
8000 return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
8007#define _mm_maskz_fnmsub_round_ss(U, A, B, C, R) \
8008 ((__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
8009 -(__v4sf)(__m128)(B), \
8010 -(__v4sf)(__m128)(C), (__mmask8)(U), \
8016 return __builtin_ia32_vfmsubss3_mask3((__v4sf)__W,
8023#define _mm_mask3_fnmsub_round_ss(W, X, Y, U, R) \
8024 ((__m128)__builtin_ia32_vfmsubss3_mask3((__v4sf)(__m128)(W), \
8025 -(__v4sf)(__m128)(X), \
8026 (__v4sf)(__m128)(Y), (__mmask8)(U), \
8032 return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
8039#define _mm_fmadd_round_sd(A, B, C, R) \
8040 ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
8041 (__v2df)(__m128d)(B), \
8042 (__v2df)(__m128d)(C), (__mmask8)-1, \
8045#define _mm_mask_fmadd_round_sd(W, U, A, B, R) \
8046 ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
8047 (__v2df)(__m128d)(A), \
8048 (__v2df)(__m128d)(B), (__mmask8)(U), \
8054 return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
8061#define _mm_maskz_fmadd_round_sd(U, A, B, C, R) \
8062 ((__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
8063 (__v2df)(__m128d)(B), \
8064 (__v2df)(__m128d)(C), (__mmask8)(U), \
8070 return __builtin_ia32_vfmaddsd3_mask3((__v2df)__W,
8077#define _mm_mask3_fmadd_round_sd(W, X, Y, U, R) \
8078 ((__m128d)__builtin_ia32_vfmaddsd3_mask3((__v2df)(__m128d)(W), \
8079 (__v2df)(__m128d)(X), \
8080 (__v2df)(__m128d)(Y), (__mmask8)(U), \
8086 return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
8093#define _mm_fmsub_round_sd(A, B, C, R) \
8094 ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
8095 (__v2df)(__m128d)(B), \
8096 -(__v2df)(__m128d)(C), (__mmask8)-1, \
8099#define _mm_mask_fmsub_round_sd(W, U, A, B, R) \
8100 ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
8101 (__v2df)(__m128d)(A), \
8102 -(__v2df)(__m128d)(B), (__mmask8)(U), \
8108 return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
8115#define _mm_maskz_fmsub_round_sd(U, A, B, C, R) \
8116 ((__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
8117 (__v2df)(__m128d)(B), \
8118 -(__v2df)(__m128d)(C), \
8119 (__mmask8)(U), (int)(R)))
8124 return __builtin_ia32_vfmsubsd3_mask3((__v2df)__W,
8131#define _mm_mask3_fmsub_round_sd(W, X, Y, U, R) \
8132 ((__m128d)__builtin_ia32_vfmsubsd3_mask3((__v2df)(__m128d)(W), \
8133 (__v2df)(__m128d)(X), \
8134 (__v2df)(__m128d)(Y), \
8135 (__mmask8)(U), (int)(R)))
8140 return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
8147#define _mm_fnmadd_round_sd(A, B, C, R) \
8148 ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
8149 -(__v2df)(__m128d)(B), \
8150 (__v2df)(__m128d)(C), (__mmask8)-1, \
8153#define _mm_mask_fnmadd_round_sd(W, U, A, B, R) \
8154 ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
8155 -(__v2df)(__m128d)(A), \
8156 (__v2df)(__m128d)(B), (__mmask8)(U), \
8162 return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
8169#define _mm_maskz_fnmadd_round_sd(U, A, B, C, R) \
8170 ((__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
8171 -(__v2df)(__m128d)(B), \
8172 (__v2df)(__m128d)(C), (__mmask8)(U), \
8178 return __builtin_ia32_vfmaddsd3_mask3((__v2df)__W,
8185#define _mm_mask3_fnmadd_round_sd(W, X, Y, U, R) \
8186 ((__m128d)__builtin_ia32_vfmaddsd3_mask3((__v2df)(__m128d)(W), \
8187 -(__v2df)(__m128d)(X), \
8188 (__v2df)(__m128d)(Y), (__mmask8)(U), \
8194 return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
8201#define _mm_fnmsub_round_sd(A, B, C, R) \
8202 ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
8203 -(__v2df)(__m128d)(B), \
8204 -(__v2df)(__m128d)(C), (__mmask8)-1, \
8207#define _mm_mask_fnmsub_round_sd(W, U, A, B, R) \
8208 ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
8209 -(__v2df)(__m128d)(A), \
8210 -(__v2df)(__m128d)(B), (__mmask8)(U), \
8216 return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
8223#define _mm_maskz_fnmsub_round_sd(U, A, B, C, R) \
8224 ((__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
8225 -(__v2df)(__m128d)(B), \
8226 -(__v2df)(__m128d)(C), \
8233 return __builtin_ia32_vfmsubsd3_mask3((__v2df)__W,
8240#define _mm_mask3_fnmsub_round_sd(W, X, Y, U, R) \
8241 ((__m128d)__builtin_ia32_vfmsubsd3_mask3((__v2df)(__m128d)(W), \
8242 -(__v2df)(__m128d)(X), \
8243 (__v2df)(__m128d)(Y), \
8244 (__mmask8)(U), (int)(R)))
8246#define _mm512_permutex_pd(X, C) \
8247 ((__m512d)__builtin_ia32_permdf512((__v8df)(__m512d)(X), (int)(C)))
8249#define _mm512_mask_permutex_pd(W, U, X, C) \
8250 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
8251 (__v8df)_mm512_permutex_pd((X), (C)), \
8252 (__v8df)(__m512d)(W)))
8254#define _mm512_maskz_permutex_pd(U, X, C) \
8255 ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
8256 (__v8df)_mm512_permutex_pd((X), (C)), \
8257 (__v8df)_mm512_setzero_pd()))
8259#define _mm512_permutex_epi64(X, C) \
8260 ((__m512i)__builtin_ia32_permdi512((__v8di)(__m512i)(X), (int)(C)))
8262#define _mm512_mask_permutex_epi64(W, U, X, C) \
8263 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
8264 (__v8di)_mm512_permutex_epi64((X), (C)), \
8265 (__v8di)(__m512i)(W)))
8267#define _mm512_maskz_permutex_epi64(U, X, C) \
8268 ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
8269 (__v8di)_mm512_permutex_epi64((X), (C)), \
8270 (__v8di)_mm512_setzero_si512()))
8275 return (__m512d)__builtin_ia32_permvardf512((__v8df)
__Y, (__v8di) __X);
8281 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
8289 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
8297 return (__m512i)__builtin_ia32_permvardi512((__v8di)
__Y, (__v8di)__X);
8303 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
8312 return (__m512i)__builtin_ia32_selectq_512((
__mmask8)__M,
8320 return (__m512)__builtin_ia32_permvarsf512((__v16sf)
__Y, (__v16si)__X);
8326 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
8334 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
8342 return (__m512i)__builtin_ia32_permvarsi512((__v16si)
__Y, (__v16si)__X);
8345#define _mm512_permutevar_epi32 _mm512_permutexvar_epi32
8350 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
8359 return (__m512i)__builtin_ia32_selectd_512((
__mmask16)__M,
8364#define _mm512_mask_permutevar_epi32 _mm512_mask_permutexvar_epi32
8399 return (
unsigned char)__builtin_ia32_kortestchi(__A, __B);
8405 return (
unsigned char)__builtin_ia32_kortestzhi(__A, __B);
8410 *__C = (
unsigned char)__builtin_ia32_kortestchi(__A, __B);
8411 return (
unsigned char)__builtin_ia32_kortestzhi(__A, __B);
8432#define _kand_mask16 _mm512_kand
8433#define _kandn_mask16 _mm512_kandn
8434#define _knot_mask16 _mm512_knot
8435#define _kor_mask16 _mm512_kor
8436#define _kxnor_mask16 _mm512_kxnor
8437#define _kxor_mask16 _mm512_kxor
8439#define _kshiftli_mask16(A, I) \
8440 ((__mmask16)__builtin_ia32_kshiftlihi((__mmask16)(A), (unsigned int)(I)))
8442#define _kshiftri_mask16(A, I) \
8443 ((__mmask16)__builtin_ia32_kshiftrihi((__mmask16)(A), (unsigned int)(I)))
8447 return (
unsigned int)__builtin_ia32_kmovw((
__mmask16)__A);
8469 __builtin_nontemporal_store((__v8di_aligned)__A, (__v8di_aligned*)
__P);
8476 return (__m512i) __builtin_nontemporal_load((
const __v8di_aligned *)
__P);
8483 __builtin_nontemporal_store((__v8df_aligned)__A, (__v8df_aligned*)
__P);
8489 typedef __v16sf __v16sf_aligned
__attribute__((aligned(64)));
8490 __builtin_nontemporal_store((__v16sf_aligned)__A, (__v16sf_aligned*)
__P);
8496 return (__m512d) __builtin_ia32_compressdf512_mask ((__v8df) __A,
8504 return (__m512d) __builtin_ia32_compressdf512_mask ((__v8df) __A,
8513 return (__m512i) __builtin_ia32_compressdi512_mask ((__v8di) __A,
8521 return (__m512i) __builtin_ia32_compressdi512_mask ((__v8di) __A,
8530 return (__m512) __builtin_ia32_compresssf512_mask ((__v16sf) __A,
8538 return (__m512) __builtin_ia32_compresssf512_mask ((__v16sf) __A,
8547 return (__m512i) __builtin_ia32_compresssi512_mask ((__v16si) __A,
8555 return (__m512i) __builtin_ia32_compresssi512_mask ((__v16si) __A,
8561#define _mm_cmp_round_ss_mask(X, Y, P, R) \
8562 ((__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
8563 (__v4sf)(__m128)(Y), (int)(P), \
8564 (__mmask8)-1, (int)(R)))
8566#define _mm_mask_cmp_round_ss_mask(M, X, Y, P, R) \
8567 ((__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
8568 (__v4sf)(__m128)(Y), (int)(P), \
8569 (__mmask8)(M), (int)(R)))
8571#define _mm_cmp_ss_mask(X, Y, P) \
8572 ((__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
8573 (__v4sf)(__m128)(Y), (int)(P), \
8575 _MM_FROUND_CUR_DIRECTION))
8577#define _mm_mask_cmp_ss_mask(M, X, Y, P) \
8578 ((__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
8579 (__v4sf)(__m128)(Y), (int)(P), \
8581 _MM_FROUND_CUR_DIRECTION))
8583#define _mm_cmp_round_sd_mask(X, Y, P, R) \
8584 ((__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
8585 (__v2df)(__m128d)(Y), (int)(P), \
8586 (__mmask8)-1, (int)(R)))
8588#define _mm_mask_cmp_round_sd_mask(M, X, Y, P, R) \
8589 ((__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
8590 (__v2df)(__m128d)(Y), (int)(P), \
8591 (__mmask8)(M), (int)(R)))
8593#define _mm_cmp_sd_mask(X, Y, P) \
8594 ((__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
8595 (__v2df)(__m128d)(Y), (int)(P), \
8597 _MM_FROUND_CUR_DIRECTION))
8599#define _mm_mask_cmp_sd_mask(M, X, Y, P) \
8600 ((__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
8601 (__v2df)(__m128d)(Y), (int)(P), \
8603 _MM_FROUND_CUR_DIRECTION))
8666 return (__m512)__builtin_shufflevector((__v16sf)__A, (__v16sf)__A,
8667 1, 1, 3, 3, 5, 5, 7, 7, 9, 9, 11, 11, 13, 13, 15, 15);
8673 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
8681 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
8689 return (__m512)__builtin_shufflevector((__v16sf)__A, (__v16sf)__A,
8690 0, 0, 2, 2, 4, 4, 6, 6, 8, 8, 10, 10, 12, 12, 14, 14);
8696 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
8704 return (__m512)__builtin_ia32_selectps_512((
__mmask16)__U,
8712 return __builtin_ia32_selectss_128(__U,
_mm_move_ss(__A, __B), __W);
8718 return __builtin_ia32_selectss_128(__U,
_mm_move_ss(__A, __B),
8725 return __builtin_ia32_selectsd_128(__U,
_mm_move_sd(__A, __B), __W);
8731 return __builtin_ia32_selectsd_128(__U,
_mm_move_sd(__A, __B),
8738 __builtin_ia32_storess128_mask ((__v4sf *)__W, __A, __U & 1);
8744 __builtin_ia32_storesd128_mask ((__v2df *)__W, __A, __U & 1);
8750 __m128 src = (__v4sf) __builtin_shufflevector((__v4sf) __W,
8754 return (__m128) __builtin_ia32_loadss128_mask ((
const __v4sf *) __A, src, __U & 1);
8760 return (__m128)__builtin_ia32_loadss128_mask ((
const __v4sf *) __A,
8768 __m128d src = (__v2df) __builtin_shufflevector((__v2df) __W,
8772 return (__m128d) __builtin_ia32_loadsd128_mask ((
const __v2df *) __A, src, __U & 1);
8778 return (__m128d) __builtin_ia32_loadsd128_mask ((
const __v2df *) __A,
8783#define _mm512_shuffle_epi32(A, I) \
8784 ((__m512i)__builtin_ia32_pshufd512((__v16si)(__m512i)(A), (int)(I)))
8786#define _mm512_mask_shuffle_epi32(W, U, A, I) \
8787 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
8788 (__v16si)_mm512_shuffle_epi32((A), (I)), \
8789 (__v16si)(__m512i)(W)))
8791#define _mm512_maskz_shuffle_epi32(U, A, I) \
8792 ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
8793 (__v16si)_mm512_shuffle_epi32((A), (I)), \
8794 (__v16si)_mm512_setzero_si512()))
8799 return (__m512d) __builtin_ia32_expanddf512_mask ((__v8df) __A,
8807 return (__m512d) __builtin_ia32_expanddf512_mask ((__v8df) __A,
8815 return (__m512i) __builtin_ia32_expanddi512_mask ((__v8di) __A,
8823 return (__m512i) __builtin_ia32_expanddi512_mask ((__v8di) __A,
8831 return (__m512d) __builtin_ia32_expandloaddf512_mask ((
const __v8df *)
__P,
8839 return (__m512d) __builtin_ia32_expandloaddf512_mask ((
const __v8df *)
__P,
8847 return (__m512i) __builtin_ia32_expandloaddi512_mask ((
const __v8di *)
__P,
8855 return (__m512i) __builtin_ia32_expandloaddi512_mask ((
const __v8di *)
__P,
8863 return (__m512) __builtin_ia32_expandloadsf512_mask ((
const __v16sf *)
__P,
8871 return (__m512) __builtin_ia32_expandloadsf512_mask ((
const __v16sf *)
__P,
8879 return (__m512i) __builtin_ia32_expandloadsi512_mask ((
const __v16si *)
__P,
8887 return (__m512i) __builtin_ia32_expandloadsi512_mask ((
const __v16si *)
__P,
8895 return (__m512) __builtin_ia32_expandsf512_mask ((__v16sf) __A,
8903 return (__m512) __builtin_ia32_expandsf512_mask ((__v16sf) __A,
8911 return (__m512i) __builtin_ia32_expandsi512_mask ((__v16si) __A,
8919 return (__m512i) __builtin_ia32_expandsi512_mask ((__v16si) __A,
8924#define _mm512_cvt_roundps_pd(A, R) \
8925 ((__m512d)__builtin_ia32_cvtps2pd512_mask((__v8sf)(__m256)(A), \
8926 (__v8df)_mm512_undefined_pd(), \
8927 (__mmask8)-1, (int)(R)))
8929#define _mm512_mask_cvt_roundps_pd(W, U, A, R) \
8930 ((__m512d)__builtin_ia32_cvtps2pd512_mask((__v8sf)(__m256)(A), \
8931 (__v8df)(__m512d)(W), \
8932 (__mmask8)(U), (int)(R)))
8934#define _mm512_maskz_cvt_roundps_pd(U, A, R) \
8935 ((__m512d)__builtin_ia32_cvtps2pd512_mask((__v8sf)(__m256)(A), \
8936 (__v8df)_mm512_setzero_pd(), \
8937 (__mmask8)(U), (int)(R)))
8942 return (__m512d) __builtin_convertvector((__v8sf)__A, __v8df);
8948 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
8956 return (__m512d)__builtin_ia32_selectpd_512((
__mmask8)__U,
8976 return (__m512d) __builtin_ia32_selectpd_512 ((
__mmask8) __U,
8984 return (__m512d) __builtin_ia32_selectpd_512 ((
__mmask8) __U,
8992 return (__m512) __builtin_ia32_selectps_512 ((
__mmask16) __U,
9000 return (__m512) __builtin_ia32_selectps_512 ((
__mmask16) __U,
9008 __builtin_ia32_compressstoredf512_mask ((__v8df *)
__P, (__v8df) __A,
9015 __builtin_ia32_compressstoredi512_mask ((__v8di *)
__P, (__v8di) __A,
9022 __builtin_ia32_compressstoresf512_mask ((__v16sf *)
__P, (__v16sf) __A,
9029 __builtin_ia32_compressstoresi512_mask ((__v16si *)
__P, (__v16si) __A,
9033#define _mm_cvt_roundsd_ss(A, B, R) \
9034 ((__m128)__builtin_ia32_cvtsd2ss_round_mask((__v4sf)(__m128)(A), \
9035 (__v2df)(__m128d)(B), \
9036 (__v4sf)_mm_undefined_ps(), \
9037 (__mmask8)-1, (int)(R)))
9039#define _mm_mask_cvt_roundsd_ss(W, U, A, B, R) \
9040 ((__m128)__builtin_ia32_cvtsd2ss_round_mask((__v4sf)(__m128)(A), \
9041 (__v2df)(__m128d)(B), \
9042 (__v4sf)(__m128)(W), \
9043 (__mmask8)(U), (int)(R)))
9045#define _mm_maskz_cvt_roundsd_ss(U, A, B, R) \
9046 ((__m128)__builtin_ia32_cvtsd2ss_round_mask((__v4sf)(__m128)(A), \
9047 (__v2df)(__m128d)(B), \
9048 (__v4sf)_mm_setzero_ps(), \
9049 (__mmask8)(U), (int)(R)))
9054 return __builtin_ia32_cvtsd2ss_round_mask ((__v4sf)__A,
9063 return __builtin_ia32_cvtsd2ss_round_mask ((__v4sf)__A,
9069#define _mm_cvtss_i32 _mm_cvtss_si32
9070#define _mm_cvtsd_i32 _mm_cvtsd_si32
9071#define _mm_cvti32_sd _mm_cvtsi32_sd
9072#define _mm_cvti32_ss _mm_cvtsi32_ss
9074#define _mm_cvtss_i64 _mm_cvtss_si64
9075#define _mm_cvtsd_i64 _mm_cvtsd_si64
9076#define _mm_cvti64_sd _mm_cvtsi64_sd
9077#define _mm_cvti64_ss _mm_cvtsi64_ss
9081#define _mm_cvt_roundi64_sd(A, B, R) \
9082 ((__m128d)__builtin_ia32_cvtsi2sd64((__v2df)(__m128d)(A), (long long)(B), \
9085#define _mm_cvt_roundsi64_sd(A, B, R) \
9086 ((__m128d)__builtin_ia32_cvtsi2sd64((__v2df)(__m128d)(A), (long long)(B), \
9090#define _mm_cvt_roundsi32_ss(A, B, R) \
9091 ((__m128)__builtin_ia32_cvtsi2ss32((__v4sf)(__m128)(A), (int)(B), (int)(R)))
9093#define _mm_cvt_roundi32_ss(A, B, R) \
9094 ((__m128)__builtin_ia32_cvtsi2ss32((__v4sf)(__m128)(A), (int)(B), (int)(R)))
9097#define _mm_cvt_roundsi64_ss(A, B, R) \
9098 ((__m128)__builtin_ia32_cvtsi2ss64((__v4sf)(__m128)(A), (long long)(B), \
9101#define _mm_cvt_roundi64_ss(A, B, R) \
9102 ((__m128)__builtin_ia32_cvtsi2ss64((__v4sf)(__m128)(A), (long long)(B), \
9106#define _mm_cvt_roundss_sd(A, B, R) \
9107 ((__m128d)__builtin_ia32_cvtss2sd_round_mask((__v2df)(__m128d)(A), \
9108 (__v4sf)(__m128)(B), \
9109 (__v2df)_mm_undefined_pd(), \
9110 (__mmask8)-1, (int)(R)))
9112#define _mm_mask_cvt_roundss_sd(W, U, A, B, R) \
9113 ((__m128d)__builtin_ia32_cvtss2sd_round_mask((__v2df)(__m128d)(A), \
9114 (__v4sf)(__m128)(B), \
9115 (__v2df)(__m128d)(W), \
9116 (__mmask8)(U), (int)(R)))
9118#define _mm_maskz_cvt_roundss_sd(U, A, B, R) \
9119 ((__m128d)__builtin_ia32_cvtss2sd_round_mask((__v2df)(__m128d)(A), \
9120 (__v4sf)(__m128)(B), \
9121 (__v2df)_mm_setzero_pd(), \
9122 (__mmask8)(U), (int)(R)))
9127 return __builtin_ia32_cvtss2sd_round_mask((__v2df)__A,
9136 return __builtin_ia32_cvtss2sd_round_mask((__v2df)__A,
9150#define _mm_cvt_roundu64_sd(A, B, R) \
9151 ((__m128d)__builtin_ia32_cvtusi2sd64((__v2df)(__m128d)(A), \
9152 (unsigned long long)(B), (int)(R)))
9155_mm_cvtu64_sd (__m128d __A,
unsigned long long __B)
9162#define _mm_cvt_roundu32_ss(A, B, R) \
9163 ((__m128)__builtin_ia32_cvtusi2ss32((__v4sf)(__m128)(A), (unsigned int)(B), \
9174#define _mm_cvt_roundu64_ss(A, B, R) \
9175 ((__m128)__builtin_ia32_cvtusi2ss64((__v4sf)(__m128)(A), \
9176 (unsigned long long)(B), (int)(R)))
9179_mm_cvtu64_ss (__m128 __A,
unsigned long long __B)
9189 return (__m512i) __builtin_ia32_selectd_512(__M,
9197 return (__m512i) __builtin_ia32_selectq_512(__M,
9204 char __e58,
char __e57,
char __e56,
char __e55,
char __e54,
char __e53,
9205 char __e52,
char __e51,
char __e50,
char __e49,
char __e48,
char __e47,
9206 char __e46,
char __e45,
char __e44,
char __e43,
char __e42,
char __e41,
9207 char __e40,
char __e39,
char __e38,
char __e37,
char __e36,
char __e35,
9208 char __e34,
char __e33,
char __e32,
char __e31,
char __e30,
char __e29,
9209 char __e28,
char __e27,
char __e26,
char __e25,
char __e24,
char __e23,
9210 char __e22,
char __e21,
char __e20,
char __e19,
char __e18,
char __e17,
9211 char __e16,
char __e15,
char __e14,
char __e13,
char __e12,
char __e11,
9212 char __e10,
char __e9,
char __e8,
char __e7,
char __e6,
char __e5,
9213 char __e4,
char __e3,
char __e2,
char __e1,
char __e0) {
9215 return __extension__ (__m512i)(__v64qi)
9216 {__e0, __e1, __e2, __e3, __e4, __e5, __e6, __e7,
9217 __e8, __e9, __e10, __e11, __e12, __e13, __e14, __e15,
9218 __e16, __e17, __e18, __e19, __e20, __e21, __e22, __e23,
9219 __e24, __e25, __e26, __e27, __e28, __e29, __e30, __e31,
9220 __e32, __e33, __e34, __e35, __e36, __e37, __e38, __e39,
9221 __e40, __e41, __e42, __e43, __e44, __e45, __e46, __e47,
9222 __e48, __e49, __e50, __e51, __e52, __e53, __e54, __e55,
9223 __e56, __e57, __e58, __e59, __e60, __e61, __e62, __e63};
9228 short __e27,
short __e26,
short __e25,
short __e24,
short __e23,
9229 short __e22,
short __e21,
short __e20,
short __e19,
short __e18,
9230 short __e17,
short __e16,
short __e15,
short __e14,
short __e13,
9231 short __e12,
short __e11,
short __e10,
short __e9,
short __e8,
9232 short __e7,
short __e6,
short __e5,
short __e4,
short __e3,
9233 short __e2,
short __e1,
short __e0) {
9234 return __extension__ (__m512i)(__v32hi)
9235 {__e0, __e1, __e2, __e3, __e4, __e5, __e6, __e7,
9236 __e8, __e9, __e10, __e11, __e12, __e13, __e14, __e15,
9237 __e16, __e17, __e18, __e19, __e20, __e21, __e22, __e23,
9238 __e24, __e25, __e26, __e27, __e28, __e29, __e30, __e31 };
9243 int __E,
int __F,
int __G,
int __H,
9244 int __I,
int __J,
int __K,
int __L,
9245 int __M,
int __N,
int __O,
int __P)
9247 return __extension__ (__m512i)(__v16si)
9248 {
__P, __O, __N, __M,
__L, __K, __J, __I,
9249 __H, __G, __F, __E,
__D, __C, __B, __A };
9252#define _mm512_setr_epi32(e0,e1,e2,e3,e4,e5,e6,e7, \
9253 e8,e9,e10,e11,e12,e13,e14,e15) \
9254 _mm512_set_epi32((e15),(e14),(e13),(e12),(e11),(e10),(e9),(e8),(e7),(e6), \
9255 (e5),(e4),(e3),(e2),(e1),(e0))
9259 long long __D,
long long __E,
long long __F,
9260 long long __G,
long long __H)
9262 return __extension__ (__m512i) (__v8di)
9263 { __H, __G, __F, __E,
__D, __C, __B, __A };
9266#define _mm512_setr_epi64(e0,e1,e2,e3,e4,e5,e6,e7) \
9267 _mm512_set_epi64((e7),(e6),(e5),(e4),(e3),(e2),(e1),(e0))
9271 double __E,
double __F,
double __G,
double __H)
9273 return __extension__ (__m512d)
9274 { __H, __G, __F, __E,
__D, __C, __B, __A };
9277#define _mm512_setr_pd(e0,e1,e2,e3,e4,e5,e6,e7) \
9278 _mm512_set_pd((e7),(e6),(e5),(e4),(e3),(e2),(e1),(e0))
9282 float __E,
float __F,
float __G,
float __H,
9283 float __I,
float __J,
float __K,
float __L,
9284 float __M,
float __N,
float __O,
float __P)
9286 return __extension__ (__m512)
9287 {
__P, __O, __N, __M,
__L, __K, __J, __I,
9288 __H, __G, __F, __E,
__D, __C, __B, __A };
9291#define _mm512_setr_ps(e0,e1,e2,e3,e4,e5,e6,e7,e8,e9,e10,e11,e12,e13,e14,e15) \
9292 _mm512_set_ps((e15),(e14),(e13),(e12),(e11),(e10),(e9),(e8),(e7),(e6),(e5), \
9293 (e4),(e3),(e2),(e1),(e0))
9336 return __builtin_reduce_add((__v8di)__W);
9340 return __builtin_reduce_mul((__v8di)__W);
9344 return __builtin_reduce_and((__v8di)__W);
9348 return __builtin_reduce_or((__v8di)__W);
9354 return __builtin_reduce_add((__v8di)__W);
9360 return __builtin_reduce_mul((__v8di)__W);
9366 return __builtin_reduce_and((__v8di)__W);
9372 return __builtin_reduce_or((__v8di)__W);
9379 return __builtin_ia32_reduce_fadd_pd512(-0.0, __W);
9383 return __builtin_ia32_reduce_fmul_pd512(1.0, __W);
9389 return __builtin_ia32_reduce_fadd_pd512(-0.0, __W);
9395 return __builtin_ia32_reduce_fmul_pd512(1.0, __W);
9400 return __builtin_reduce_add((__v16si)__W);
9405 return __builtin_reduce_mul((__v16si)__W);
9410 return __builtin_reduce_and((__v16si)__W);
9415 return __builtin_reduce_or((__v16si)__W);
9421 return __builtin_reduce_add((__v16si)__W);
9427 return __builtin_reduce_mul((__v16si)__W);
9433 return __builtin_reduce_and((__v16si)__W);
9439 return __builtin_reduce_or((__v16si)__W);
9444 return __builtin_ia32_reduce_fadd_ps512(-0.0f, __W);
9449 return __builtin_ia32_reduce_fmul_ps512(1.0f, __W);
9455 return __builtin_ia32_reduce_fadd_ps512(-0.0f, __W);
9461 return __builtin_ia32_reduce_fmul_ps512(1.0f, __W);
9466 return __builtin_reduce_max((__v8di)__V);
9471 return __builtin_reduce_max((__v8du)__V);
9476 return __builtin_reduce_min((__v8di)__V);
9481 return __builtin_reduce_min((__v8du)__V);
9487 return __builtin_reduce_max((__v8di)__V);
9493 return __builtin_reduce_max((__v8du)__V);
9499 return __builtin_reduce_min((__v8di)__V);
9505 return __builtin_reduce_min((__v8du)__V);
9509 return __builtin_reduce_max((__v16si)__V);
9514 return __builtin_reduce_max((__v16su)__V);
9519 return __builtin_reduce_min((__v16si)__V);
9524 return __builtin_reduce_min((__v16su)__V);
9530 return __builtin_reduce_max((__v16si)__V);
9536 return __builtin_reduce_max((__v16su)__V);
9542 return __builtin_reduce_min((__v16si)__V);
9548 return __builtin_reduce_min((__v16su)__V);
9553 return __builtin_ia32_reduce_fmax_pd512(__V);
9558 return __builtin_ia32_reduce_fmin_pd512(__V);
9564 return __builtin_ia32_reduce_fmax_pd512(__V);
9570 return __builtin_ia32_reduce_fmin_pd512(__V);
9575 return __builtin_ia32_reduce_fmax_ps512(__V);
9580 return __builtin_ia32_reduce_fmin_ps512(__V);
9586 return __builtin_ia32_reduce_fmax_ps512(__V);
9592 return __builtin_ia32_reduce_fmin_ps512(__V);
9608 __v16si
__b = (__v16si)__A;
9627#define _mm512_i32logather_pd(vindex, base_addr, scale) \
9628 _mm512_i32gather_pd(_mm512_castsi512_si256(vindex), (base_addr), (scale))
9651#define _mm512_mask_i32logather_pd(src, mask, vindex, base_addr, scale) \
9652 _mm512_mask_i32gather_pd((src), (mask), _mm512_castsi512_si256(vindex), \
9653 (base_addr), (scale))
9670#define _mm512_i32logather_epi64(vindex, base_addr, scale) \
9671 _mm512_i32gather_epi64(_mm512_castsi512_si256(vindex), (base_addr), (scale))
9693#define _mm512_mask_i32logather_epi64(src, mask, vindex, base_addr, scale) \
9694 _mm512_mask_i32gather_epi64((src), (mask), _mm512_castsi512_si256(vindex), \
9695 (base_addr), (scale))
9711#define _mm512_i32loscatter_pd(base_addr, vindex, v1, scale) \
9712 _mm512_i32scatter_pd((base_addr), _mm512_castsi512_si256(vindex), (v1), (scale))
9732#define _mm512_mask_i32loscatter_pd(base_addr, mask, vindex, v1, scale) \
9733 _mm512_mask_i32scatter_pd((base_addr), (mask), \
9734 _mm512_castsi512_si256(vindex), (v1), (scale))
9750#define _mm512_i32loscatter_epi64(base_addr, vindex, v1, scale) \
9751 _mm512_i32scatter_epi64((base_addr), \
9752 _mm512_castsi512_si256(vindex), (v1), (scale))
9771#define _mm512_mask_i32loscatter_epi64(base_addr, mask, vindex, v1, scale) \
9772 _mm512_mask_i32scatter_epi64((base_addr), (mask), \
9773 _mm512_castsi512_si256(vindex), (v1), (scale))
9775#undef __DEFAULT_FN_ATTRS512
9776#undef __DEFAULT_FN_ATTRS128
9777#undef __DEFAULT_FN_ATTRS
static __inline__ vector float vector float __b
static __inline__ uint32_t volatile uint32_t * __p
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask3_fmadd_sd(__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_sll_epi64(__mmask8 __U, __m512i __A, __m128i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_slli_epi32(__mmask16 __U, __m512i __A, unsigned int __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_blend_epi32(__mmask16 __U, __m512i __A, __m512i __W)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_or_si512(__m512i __a, __m512i __b)
static __inline__ unsigned __DEFAULT_FN_ATTRS128 _mm_cvttsd_u32(__m128d __A)
static __inline__ unsigned __DEFAULT_FN_ATTRS128 _mm_cvtsd_u32(__m128d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_sub_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_min_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_setzero_pd(void)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_undefined(void)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_broadcast_i32x4(__m512i __O, __mmask16 __M, __m128i __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtpd_epu32(__mmask8 __U, __m512d __A)
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_mul_epi64(__mmask8 __M, __m512i __W)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask3_fmsub_ss(__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_mov_epi32(__mmask16 __U, __m512i __A)
static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_mul_pd(__mmask8 __M, __m512d __W)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_add_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_mul_epi64(__m512i __W)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_load_epi32(__m512i __W, __mmask16 __U, void const *__P)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtepi32_epi16(__m512i __A)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi64_epi8(__m128i __O, __mmask8 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_min_epu64(__mmask8 __M, __m512i __A, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_sub_pd(__mmask8 __U, __m512d __A, __m512d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_permutexvar_epi64(__m512i __X, __m512i __Y)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvttps_epu32(__m512i __W, __mmask16 __U, __m512 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_fmsubadd_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_load_ps(void const *__p)
#define _mm512_mask_cmpeq_epi64_mask(k, A, B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_sll_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m128i __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_add_ps(__mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_expandloadu_ps(__m512 __W, __mmask16 __U, void const *__P)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_max_epu32(__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_getexp_ss(__m128 __A, __m128 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_add_epi32(__mmask16 __U, __m512i __A, __m512i __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_min_epu64(__m512i __A, __m512i __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_fnmadd_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ float __DEFAULT_FN_ATTRS512 _mm512_reduce_mul_ps(__m512 __W)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_castpd_ps(__m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_andnot_epi64(__mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_slli_epi64(__mmask8 __U, __m512i __A, unsigned int __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_sll_epi32(__m512i __A, __m128i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_andnot_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
static __inline__ int __DEFAULT_FN_ATTRS128 _mm_cvttss_i32(__m128 __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_loadu_si512(void const *__P)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_loadu_epi64(__m512i __W, __mmask8 __U, void const *__P)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_fmsubadd_pd(__m512d __A, __m512d __B, __m512d __C)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_set1_epi32(int __s)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_sub_epi32(__m512i __A, __m512i __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_scalef_ps(__mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_andnot_si512(__m512i __A, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_rcp14_pd(__m512d __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_cvttps_epi32(__m512 __a)
static __inline__ unsigned long long __DEFAULT_FN_ATTRS512 _mm512_reduce_max_epu64(__m512i __V)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_max_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_permutex2var_ps(__mmask16 __U, __m512 __A, __m512i __I, __m512 __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_castps_si512(__m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_xor_epi32(__m512i __a, __m512i __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_broadcastd_epi32(__m128i __A)
static __inline__ unsigned int __DEFAULT_FN_ATTRS512 _mm512_reduce_min_epu32(__m512i __V)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_mullox_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_xor_epi32(__m512i __src, __mmask16 __k, __m512i __a, __m512i __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepu16_epi32(__m512i __W, __mmask16 __U, __m256i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_andnot_epi32(__mmask16 __U, __m512i __A, __m512i __B)
static __inline__ float __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_max_ps(__mmask16 __M, __m512 __V)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_sllv_epi32(__m512i __W, __mmask16 __U, __m512i __X, __m512i __Y)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_moveldup_ps(__mmask16 __U, __m512 __A)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_cvtsepi64_epi8(__m512i __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_set1_epi16(short __w)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_cvtss_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_sub_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_loadu_pd(__m512d __W, __mmask8 __U, void const *__P)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_permutex2var_epi64(__m512i __A, __m512i __I, __m512i __B)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_compressstoreu_epi32(void *__P, __mmask16 __U, __m512i __A)
static __inline__ unsigned char __DEFAULT_FN_ATTRS _kortest_mask16_u8(__mmask16 __A, __mmask16 __B, unsigned char *__C)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_fnmadd_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_broadcast_i64x4(__m256i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_srli_epi32(__m512i __W, __mmask16 __U, __m512i __A, unsigned int __B)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvttpd_epu32(__m512d __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_getexp_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_min_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask3_fmadd_ss(__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepu32_ps(__mmask16 __U, __m512i __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_fnmadd_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ unsigned int __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_max_epu32(__mmask16 __M, __m512i __V)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_movedup_pd(__m512d __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_min_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_rsqrt14_pd(__m512d __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_rsqrt14_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi64_storeu_epi8(void *__P, __mmask8 __M, __m512i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_fmsub_ps(__m512 __A, __m512 __B, __m512 __C)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_scalef_pd(__m512d __A, __m512d __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask3_fnmsub_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_setzero_ps(void)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_mul_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtps_epi32(__m512i __W, __mmask16 __U, __m512 __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_mul_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_unpacklo_epi64(__m512i __A, __m512i __B)
static __inline__ unsigned __DEFAULT_FN_ATTRS128 _mm_cvttss_u32(__m128 __A)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS _mm512_kmov(__mmask16 __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtusepi32_epi16(__m256i __O, __mmask16 __M, __m512i __A)
static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_min_pd(__mmask8 __M, __m512d __V)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_cvtsi512_si32(__m512i __A)
Moves the least significant 32 bits of a vector of [16 x i32] to a 32-bit signed integer value.
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_movedup_pd(__mmask8 __U, __m512d __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_add_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_compressstoreu_ps(void *__P, __mmask16 __U, __m512 __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_cvttps_epu32(__m512 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_fmaddsub_pd(__m512d __A, __m512d __B, __m512d __C)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_fnmadd_pd(__m512d __A, __m512d __B, __m512d __C)
#define _mm512_cmpeq_epi32_mask(A, B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_div_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_broadcastd_epi32(__m512i __O, __mmask16 __M, __m128i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_abs_epi64(__m512i __W, __mmask8 __U, __m512i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_add_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_sll_epi32(__mmask16 __U, __m512i __A, __m128i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_max_epu32(__mmask16 __M, __m512i __A, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_max_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_abs_epi32(__m512i __W, __mmask16 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi8_epi32(__m512i __W, __mmask16 __U, __m128i __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_fnmadd_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtps_epu32(__m512i __W, __mmask16 __U, __m512 __A)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_reduce_add_epi32(__m512i __W)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_scalef_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
static __inline __mmask8 __DEFAULT_FN_ATTRS512 _mm512_test_epi64_mask(__m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtepu32_epi64(__m256i __X)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_min_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_expandloadu_epi32(__m512i __W, __mmask16 __U, void const *__P)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_min_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepu32_epi64(__mmask8 __U, __m256i __X)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_set1_epi32(__mmask16 __M, int __A)
static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_cvtsd_f64(__m512d __a)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi64_storeu_epi16(void *__P, __mmask8 __M, __m512i __A)
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_add_epi64(__mmask8 __M, __m512i __W)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepu8_epi32(__mmask16 __U, __m128i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_rolv_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS _mm512_kor(__mmask16 __A, __mmask16 __B)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_floor_ps(__m512 __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_castps128_ps512(__m128 __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_getexp_ps(__m512 __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_rsqrt14_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_fnmsub_pd(__m512d __A, __m512d __B, __m512d __C)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_rcp14_ss(__m128 __A, __m128 __B)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvttpd_epi32(__mmask8 __U, __m512d __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_mul_epu32(__m512i __X, __m512i __Y)
static __inline __m256i __DEFAULT_FN_ATTRS512 _mm512_cvttpd_epi32(__m512d __a)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_cvtpd_pslo(__m512 __W, __mmask8 __U, __m512d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_abs_pd(__m512d __W, __mmask8 __K, __m512d __A)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_load_ps(__m512 __W, __mmask16 __U, void const *__P)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_rolv_epi64(__mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_fmadd_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_max_epi32(__mmask16 __M, __m512i __V)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_srli_epi32(__m512i __A, unsigned int __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask3_fmsubadd_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi32_epi8(__mmask16 __M, __m512i __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_fnmsub_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_min_epi64(__mmask8 __M, __m512i __A, __m512i __B)
static __inline__ unsigned long long __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_min_epu64(__mmask8 __M, __m512i __V)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_cvtepu32_ps(__m512i __A)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_div_pd(__m512d __a, __m512d __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_castsi128_si512(__m128i __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_rsqrt14_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ void __DEFAULT_FN_ATTRS _store_mask16(__mmask16 *__A, __mmask16 __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_abs_ps(__m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_set1_epi32(__m512i __O, __mmask16 __M, int __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_or_epi32(__m512i __src, __mmask16 __k, __m512i __a, __m512i __b)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_sqrt_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_unpackhi_epi32(__mmask16 __U, __m512i __A, __m512i __B)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS _mm512_kand(__mmask16 __A, __mmask16 __B)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_stream_pd(void *__P, __m512d __A)
static __inline__ __mmask8 __DEFAULT_FN_ATTRS512 _mm512_testn_epi64_mask(__m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_xor_epi64(__m512i __src, __mmask8 __k, __m512i __a, __m512i __b)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_movehdup_ps(__m512 __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_set1_epi8(char __w)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_load_ss(__mmask8 __U, const float *__A)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_cvtsepi32_epi8(__m512i __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask3_fmaddsub_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_fmadd_ss(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask3_fmsub_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_abs_epi32(__m512i __A)
static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_reduce_mul_pd(__m512d __W)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_fnmadd_ps(__m512 __A, __m512 __B, __m512 __C)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask2_permutex2var_pd(__m512d __A, __m512i __I, __mmask8 __U, __m512d __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_set_epi32(int __A, int __B, int __C, int __D, int __E, int __F, int __G, int __H, int __I, int __J, int __K, int __L, int __M, int __N, int __O, int __P)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_max_epi64(__m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtps_epu32(__m512 __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi32_epi16(__mmask16 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_unpackhi_epi64(__mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_expand_ps(__m512 __W, __mmask16 __U, __m512 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_rcp14_pd(__m512d __W, __mmask8 __U, __m512d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_set_pd(double __A, double __B, double __C, double __D, double __E, double __F, double __G, double __H)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_min_epu32(__mmask16 __M, __m512i __A, __m512i __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_fnmsub_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtusepi64_epi32(__m256i __O, __mmask8 __M, __m512i __A)
static __inline __m256d __DEFAULT_FN_ATTRS512 _mm512_castpd512_pd256(__m512d __A)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtusepi64_epi16(__m128i __O, __mmask8 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtepi8_epi32(__m128i __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_set_epi8(char __e63, char __e62, char __e61, char __e60, char __e59, char __e58, char __e57, char __e56, char __e55, char __e54, char __e53, char __e52, char __e51, char __e50, char __e49, char __e48, char __e47, char __e46, char __e45, char __e44, char __e43, char __e42, char __e41, char __e40, char __e39, char __e38, char __e37, char __e36, char __e35, char __e34, char __e33, char __e32, char __e31, char __e30, char __e29, char __e28, char __e27, char __e26, char __e25, char __e24, char __e23, char __e22, char __e21, char __e20, char __e19, char __e18, char __e17, char __e16, char __e15, char __e14, char __e13, char __e12, char __e11, char __e10, char __e9, char __e8, char __e7, char __e6, char __e5, char __e4, char __e3, char __e2, char __e1, char __e0)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_srai_epi64(__m512i __W, __mmask8 __U, __m512i __A, unsigned int __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_srai_epi32(__mmask16 __U, __m512i __A, unsigned int __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_unpacklo_pd(__mmask8 __U, __m512d __A, __m512d __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_mullo_epi32(__m512i __A, __m512i __B)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_store_epi64(void *__P, __m512i __A)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_cvtph_ps(__m256i __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_load_si512(void const *__P)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_set_epi64(long long __A, long long __B, long long __C, long long __D, long long __E, long long __F, long long __G, long long __H)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_min_epi32(__mmask16 __M, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi16_epi32(__m512i __W, __mmask16 __U, __m256i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_fmaddsub_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtsepi32_epi8(__mmask16 __M, __m512i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_fmaddsub_ps(__m512 __A, __m512 __B, __m512 __C)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_rcp14_sd(__m128d __A, __m128d __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_sqrt_pd(__mmask8 __U, __m512d __A)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_compressstoreu_pd(void *__P, __mmask8 __U, __m512d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_fnmadd_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_getexp_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
#define __DEFAULT_FN_ATTRS512
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_zextpd256_pd512(__m256d __a)
Constructs a 512-bit floating-point vector of [8 x double] from a 256-bit floating-point vector of [4...
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_expand_epi64(__m512i __W, __mmask8 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_srli_epi64(__m512i __A, unsigned int __B)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_mask_store_pd(void *__P, __mmask8 __U, __m512d __A)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_load_ps(__mmask16 __U, void const *__P)
static __inline__ void __DEFAULT_FN_ATTRS128 _mm_mask_store_sd(double *__W, __mmask8 __U, __m128d __A)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS512 _mm512_mask_testn_epi32_mask(__mmask16 __U, __m512i __A, __m512i __B)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_fnmsub_sd(__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepu32_ps(__m512 __W, __mmask16 __U, __m512i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_set_ps(float __A, float __B, float __C, float __D, float __E, float __F, float __G, float __H, float __I, float __J, float __K, float __L, float __M, float __N, float __O, float __P)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_expand_epi64(__mmask8 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_broadcastq_epi64(__m128i __A)
static __inline__ int __DEFAULT_FN_ATTRS _mm512_kortestz(__mmask16 __A, __mmask16 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_srlv_epi32(__m512i __X, __m512i __Y)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_andnot_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_mul_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline__ int __DEFAULT_FN_ATTRS _mm512_mask2int(__mmask16 __a)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_fmadd_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_zextsi256_si512(__m256i __a)
Constructs a 512-bit integer vector from a 256-bit integer vector.
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_broadcastq_epi64(__mmask8 __M, __m128i __A)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtusepi32_storeu_epi8(void *__P, __mmask16 __M, __m512i __A)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_load_pd(__m512d __W, __mmask8 __U, void const *__P)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtepi16_epi64(__m128i __A)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_unpacklo_pd(__m512d __a, __m512d __b)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_add_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_move_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtps_epi32(__m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepu16_epi64(__mmask8 __U, __m128i __A)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_set1_pd(double __w)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_permutevar_ps(__m512 __A, __m512i __C)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_stream_si512(void *__P, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_srl_epi64(__mmask8 __U, __m512i __A, __m128i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_sqrt_pd(__m512d __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_loadu_epi32(__mmask16 __U, void const *__P)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask3_fmsubadd_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_sub_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_add_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask3_fnmsub_ss(__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_sllv_epi64(__mmask8 __U, __m512i __X, __m512i __Y)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_fmadd_sd(__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
#define __DEFAULT_FN_ATTRS
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_broadcastss_ps(__mmask16 __M, __m128 __A)
#define _mm512_cmpneq_epi64_mask(A, B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_permutexvar_pd(__m512i __X, __m512d __Y)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_scalef_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_fmadd_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_mul_epu32(__mmask8 __M, __m512i __X, __m512i __Y)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtpd_epu32(__m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask2_permutex2var_epi32(__m512i __A, __m512i __I, __mmask16 __U, __m512i __B)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_unpacklo_ps(__m512 __a, __m512 __b)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_mul_ps(__m512 __a, __m512 __b)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtsepi32_epi16(__mmask16 __M, __m512i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_broadcast_f32x4(__mmask16 __M, __m128 __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtsepi64_epi32(__mmask8 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_sra_epi64(__mmask8 __U, __m512i __A, __m128i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_abs_epi64(__mmask8 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_unpackhi_epi32(__m512i __A, __m512i __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_div_ps(__mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_slli_epi64(__m512i __A, unsigned int __B)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_loadu_ps(void const *__p)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_unpacklo_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_permutexvar_epi64(__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y)
static __inline__ unsigned char __DEFAULT_FN_ATTRS _kortestz_mask16_u8(__mmask16 __A, __mmask16 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi8_epi64(__mmask8 __U, __m128i __A)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_mask_storeu_ps(void *__P, __mmask16 __U, __m512 __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_broadcastss_ps(__m128 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_sub_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_cvtpslo_pd(__m512d __W, __mmask8 __U, __m512 __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_sqrt_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_srli_epi64(__mmask8 __U, __m512i __A, unsigned int __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_getexp_pd(__mmask8 __U, __m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_min_epu32(__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_broadcast_i32x4(__m128i __A)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_reduce_mul_epi32(__m512i __W)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_cvtsd_ss(__mmask8 __U, __m128 __A, __m128d __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_mul_epu32(__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi64_epi8(__mmask8 __M, __m512i __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_permutex2var_epi32(__m512i __A, __m512i __I, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_unpackhi_epi64(__m512i __A, __m512i __B)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_mask_storeu_epi32(void *__P, __mmask16 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_rolv_epi32(__m512i __A, __m512i __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_set1_epi64(long long __d)
static __inline__ __mmask8 __DEFAULT_FN_ATTRS512 _mm512_mask_testn_epi64_mask(__mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_scalef_ss(__m128 __A, __m128 __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_fmaddsub_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_cvtusepi64_epi16(__m512i __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_div_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepu32_epi64(__m512i __W, __mmask8 __U, __m256i __X)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_max_epu64(__m512i __A, __m512i __B)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvttpd_epi32(__m256i __W, __mmask8 __U, __m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_expandloadu_epi32(__mmask16 __U, void const *__P)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_load_epi64(__mmask8 __U, void const *__P)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi64_epi16(__m128i __O, __mmask8 __M, __m512i __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_rsqrt14_pd(__m512d __W, __mmask8 __U, __m512d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_fmaddsub_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_cvtpslo_pd(__m512 __A)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_cvtusepi32_epi8(__m512i __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_fnmsub_ss(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_max_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_castps_pd(__m512 __A)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_mask_storeu_pd(void *__P, __mmask8 __U, __m512d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask3_fnmsub_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_max_epu32(__m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_sllv_epi64(__m512i __W, __mmask8 __U, __m512i __X, __m512i __Y)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi32_ps(__m512 __W, __mmask16 __U, __m512i __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_permutexvar_pd(__m512d __W, __mmask8 __U, __m512i __X, __m512d __Y)
static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_reduce_min_pd(__m512d __V)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepu32_pd(__m512d __W, __mmask8 __U, __m256i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_fmadd_ps(__m512 __A, __m512 __B, __m512 __C)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_compress_pd(__m512d __W, __mmask8 __U, __m512d __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_abs_epi64(__m512i __A)
static __inline__ unsigned long long __DEFAULT_FN_ATTRS512 _mm512_reduce_min_epu64(__m512i __V)
static __inline__ unsigned long long __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_max_epu64(__mmask8 __M, __m512i __V)
#define _mm512_mask_cmpeq_epi32_mask(k, A, B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_permutevar_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512i __C)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtusepi64_epi32(__m512i __A)
static __inline __mmask16 __DEFAULT_FN_ATTRS512 _mm512_test_epi32_mask(__m512i __A, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_broadcastsd_pd(__m512d __O, __mmask8 __M, __m128d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_scalef_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_max_epi32(__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
static __inline__ __m256 __DEFAULT_FN_ATTRS512 _mm512_mask_cvtpd_ps(__m256 __W, __mmask8 __U, __m512d __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_sqrt_ps(__mmask16 __U, __m512 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_min_pd(__mmask8 __U, __m512d __A, __m512d __B)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_getexp_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS512 _mm512_mask_test_epi32_mask(__mmask16 __U, __m512i __A, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_mov_pd(__mmask8 __U, __m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask2_permutex2var_epi64(__m512i __A, __m512i __I, __mmask8 __U, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_permutevar_pd(__m512d __A, __m512i __C)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_cvtpd_pslo(__m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_expand_epi32(__m512i __W, __mmask16 __U, __m512i __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_sqrt_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_undefined_pd(void)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_srl_epi32(__mmask16 __U, __m512i __A, __m128i __B)
#define _mm512_cmpneq_epi32_mask(A, B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_cvtph_ps(__m512 __W, __mmask16 __U, __m256i __A)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_reduce_or_epi32(__m512i __W)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_cvtu32_sd(__m128d __A, unsigned __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_mov_ps(__mmask16 __U, __m512 __A)
static __inline __m128i __DEFAULT_FN_ATTRS512 _mm512_castsi512_si128(__m512i __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_rsqrt14_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_floor_pd(__m512d __W, __mmask8 __U, __m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_sll_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m128i __B)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_storeu_si512(void *__P, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_add_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_rsqrt14_ps(__mmask16 __U, __m512 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_mul_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_unpackhi_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtepu8_epi64(__m128i __A)
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_add_epi64(__m512i __W)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_permutexvar_ps(__mmask16 __U, __m512i __X, __m512 __Y)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_broadcastss_ps(__m512 __O, __mmask16 __M, __m128 __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_mullo_epi32(__mmask16 __M, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_xor_epi64(__m512i __a, __m512i __b)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_mul_ps(__mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_unpacklo_epi64(__mmask8 __U, __m512i __A, __m512i __B)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_ceil_ps(__m512 __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtpd_epu32(__m256i __W, __mmask8 __U, __m512d __A)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtusepi64_storeu_epi32(void *__P, __mmask8 __M, __m512i __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_mul_epi32(__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_div_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask3_fnmadd_ss(__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_mullo_epi32(__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_srav_epi32(__m512i __X, __m512i __Y)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_sra_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m128i __B)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_cvtepi64_epi8(__m512i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_movehdup_ps(__mmask16 __U, __m512 __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_expandloadu_ps(__mmask16 __U, void const *__P)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_cvtepu32lo_pd(__m512i __A)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_storeu_pd(void *__P, __m512d __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_sqrt_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_cvtusepi64_epi8(__m512i __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_scalef_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
#define _mm512_mask_cmpneq_epi32_mask(k, A, B)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_fmadd_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_loadu_epi64(__mmask8 __U, void const *__P)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_store_epi64(void *__P, __mmask8 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_compress_epi64(__mmask8 __U, __m512i __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtusepi32_epi16(__mmask16 __M, __m512i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_min_ps(__m512 __A, __m512 __B)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_div_ps(__m512 __a, __m512 __b)
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_and_epi64(__mmask8 __M, __m512i __W)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_min_epu64(__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_fmsub_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvttpd_epu32(__mmask8 __U, __m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_srlv_epi64(__m512i __W, __mmask8 __U, __m512i __X, __m512i __Y)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_fmsub_sd(__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_reduce_max_epi32(__m512i __V)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_blend_ps(__mmask16 __U, __m512 __A, __m512 __W)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_and_epi64(__m512i __a, __m512i __b)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_fmsubadd_ps(__m512 __A, __m512 __B, __m512 __C)
char __v64qi __attribute__((__vector_size__(64)))
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi16_epi64(__m512i __W, __mmask8 __U, __m128i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepu8_epi64(__mmask8 __U, __m128i __A)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_set4_ps(float __A, float __B, float __C, float __D)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_cvtepi32_epi8(__m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_srav_epi64(__m512i __W, __mmask8 __U, __m512i __X, __m512i __Y)
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_min_epi64(__mmask8 __M, __m512i __V)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_zextps256_ps512(__m256 __a)
Constructs a 512-bit floating-point vector of [16 x float] from a 256-bit floating-point vector of [8...
static __inline void __DEFAULT_FN_ATTRS512 _mm512_store_si512(void *__P, __m512i __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtusepi64_epi32(__mmask8 __M, __m512i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_broadcast_f32x4(__m128 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_broadcastsd_pd(__m128d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_broadcastq_epi64(__m512i __O, __mmask8 __M, __m128i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_unpacklo_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask2_permutex2var_ps(__m512 __A, __m512i __I, __mmask16 __U, __m512 __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_min_epu32(__m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtepu16_epi32(__m256i __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_fmsub_pd(__m512d __A, __m512d __B, __m512d __C)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_permutex2var_pd(__mmask8 __U, __m512d __A, __m512i __I, __m512d __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_expandloadu_pd(__m512d __W, __mmask8 __U, void const *__P)
static __inline__ unsigned int __DEFAULT_FN_ATTRS512 _mm512_reduce_max_epu32(__m512i __V)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_xor_epi64(__mmask8 __k, __m512i __a, __m512i __b)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtusepi64_epi16(__mmask8 __M, __m512i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_max_ps(__m512 __A, __m512 __B)
static __inline __m256 __DEFAULT_FN_ATTRS512 _mm512_castps512_ps256(__m512 __A)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_store_pd(void *__P, __m512d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_min_pd(__m512d __A, __m512d __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask3_fmadd_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_ceil_ps(__m512 __W, __mmask16 __U, __m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepu16_epi32(__mmask16 __U, __m256i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_srav_epi64(__m512i __X, __m512i __Y)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtpd_epi32(__m512d __A)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtusepi64_epi8(__m128i __O, __mmask8 __M, __m512i __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_fnmsub_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
static __inline__ float __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_add_ps(__mmask16 __M, __m512 __W)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_expand_ps(__mmask16 __U, __m512 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_fmadd_pd(__m512d __A, __m512d __B, __m512d __C)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_rsqrt14_ss(__m128 __A, __m128 __B)
#define _mm512_mask_cmpneq_epi64_mask(k, A, B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_max_epi32(__m512i __A, __m512i __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_min_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi32_pd(__m512d __W, __mmask8 __U, __m256i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_sllv_epi64(__m512i __X, __m512i __Y)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_min_epi64(__m512i __A, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_cvtps_pd(__m256 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_rolv_epi32(__mmask16 __U, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_srav_epi32(__m512i __W, __mmask16 __U, __m512i __X, __m512i __Y)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_add_epi32(__m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_min_epi64(__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtepu8_epi32(__m128i __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_sqrt_pd(__m512d __W, __mmask8 __U, __m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_max_epi64(__mmask8 __M, __m512i __A, __m512i __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_mul_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_compressstoreu_epi64(void *__P, __mmask8 __U, __m512i __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_fnmsub_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtusepi32_storeu_epi16(void *__P, __mmask16 __M, __m512i __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_getexp_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_xor_epi32(__mmask16 __k, __m512i __a, __m512i __b)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_sqrt_ps(__m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi16_epi32(__mmask16 __U, __m256i __A)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_storeu_epi64(void *__P, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtepi32_epi64(__m256i __X)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtsepi64_epi16(__mmask8 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_max_epu64(__mmask8 __M, __m512i __A, __m512i __B)
static __inline__ __m256 __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtpd_ps(__mmask8 __U, __m512d __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_min_ps(__mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtusepi32_epi16(__m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_set1_epi64(__m512i __O, __mmask8 __M, long long __A)
static __inline__ float __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_min_ps(__mmask16 __M, __m512 __V)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_fmsub_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS _load_mask16(__mmask16 *__A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_unpackhi_ps(__mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_cvtsd_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128d __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_movehdup_ps(__m512 __W, __mmask16 __U, __m512 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi32lo_pd(__m512d __W, __mmask8 __U, __m512i __A)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_store_epi32(void *__P, __mmask16 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_srl_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m128i __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_permutexvar_ps(__m512 __W, __mmask16 __U, __m512i __X, __m512 __Y)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvttps_epu32(__mmask16 __U, __m512 __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi32_ps(__mmask16 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_and_epi32(__m512i __src, __mmask16 __k, __m512i __a, __m512i __b)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_max_ps(__mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_fmsub_ss(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtpd_epi32(__m256i __W, __mmask8 __U, __m512d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_ceil_pd(__m512d __W, __mmask8 __U, __m512d __A)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_permutex2var_ps(__m512 __A, __m512i __I, __m512 __B)
static __inline__ float __DEFAULT_FN_ATTRS512 _mm512_cvtss_f32(__m512 __a)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtusepi32_epi8(__mmask16 __M, __m512i __A)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtsepi32_epi8(__m128i __O, __mmask16 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_srlv_epi32(__m512i __W, __mmask16 __U, __m512i __X, __m512i __Y)
static __inline__ float __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_mul_ps(__mmask16 __M, __m512 __W)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS _mm512_kxor(__mmask16 __A, __mmask16 __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_fmsub_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_zextsi128_si512(__m128i __a)
Constructs a 512-bit integer vector from a 128-bit integer vector.
static __inline__ int __DEFAULT_FN_ATTRS128 _mm_cvttsd_i32(__m128d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_max_pd(__mmask8 __U, __m512d __A, __m512d __B)
_MM_TERNLOG_ENUM
A helper to represent the ternary logic operations among vector A, B and C.
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_set4_pd(double __A, double __B, double __C, double __D)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_mask_store_ps(void *__P, __mmask16 __U, __m512 __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_sub_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_sra_epi32(__mmask16 __U, __m512i __A, __m128i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_broadcast_i64x4(__mmask8 __M, __m256i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_fmsubadd_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_mov_epi64(__m512i __W, __mmask8 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_or_epi64(__mmask8 __k, __m512i __a, __m512i __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_and_epi32(__m512i __a, __m512i __b)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS _mm512_kandn(__mmask16 __A, __mmask16 __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_div_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_rsqrt14_pd(__mmask8 __U, __m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_unpacklo_epi32(__mmask16 __U, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_compress_epi64(__m512i __W, __mmask8 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_max_epi64(__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_undefined_ps(void)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_permutexvar_epi32(__mmask16 __M, __m512i __X, __m512i __Y)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_compress_epi32(__mmask16 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_unpacklo_epi32(__m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi8_epi32(__mmask16 __U, __m128i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_broadcast_i32x4(__mmask16 __M, __m128i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_abs_epi32(__mmask16 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_srlv_epi32(__mmask16 __U, __m512i __X, __m512i __Y)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_andnot_epi64(__m512i __A, __m512i __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_broadcast_f32x4(__m512 __O, __mmask16 __M, __m128 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_unpacklo_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_stream_ps(void *__P, __m512 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_getexp_pd(__m512d __W, __mmask8 __U, __m512d __A)
static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_reduce_max_pd(__m512d __V)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_loadu_pd(__mmask8 __U, void const *__P)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtps_epi32(__mmask16 __U, __m512 __A)
static __inline__ unsigned char __DEFAULT_FN_ATTRS _kortestc_mask16_u8(__mmask16 __A, __mmask16 __B)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_loadu_ps(__m512 __W, __mmask16 __U, void const *__P)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_expand_pd(__m512d __W, __mmask8 __U, __m512d __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_unpackhi_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask3_fnmadd_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_srlv_epi64(__mmask8 __U, __m512i __X, __m512i __Y)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_compress_ps(__mmask16 __U, __m512 __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_max_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_abs_ps(__m512 __W, __mmask16 __K, __m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_permutex2var_epi32(__m512i __A, __mmask16 __U, __m512i __I, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_load_epi32(__mmask16 __U, void const *__P)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_mov_ps(__m512 __W, __mmask16 __U, __m512 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_broadcast_f64x4(__mmask8 __M, __m256d __A)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS _mm512_int2mask(int __a)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtps_epu32(__mmask16 __U, __m512 __A)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_store_ps(void *__P, __m512 __A)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi64_storeu_epi32(void *__P, __mmask8 __M, __m512i __A)
#define __DEFAULT_FN_ATTRS128
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_min_epi64(__m512i __V)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_movedup_pd(__m512d __W, __mmask8 __U, __m512d __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_scalef_ps(__m512 __A, __m512 __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask3_fmaddsub_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_load_ss(__m128 __W, __mmask8 __U, const float *__A)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_floor_pd(__m512d __A)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_loadu_pd(void const *__p)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_reduce_and_epi32(__m512i __W)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtsepi64_storeu_epi32(void *__P, __mmask8 __M, __m512i __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_min_epi32(__m512i __A, __m512i __B)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtsepi32_epi16(__m256i __O, __mmask16 __M, __m512i __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_compress_pd(__mmask8 __U, __m512d __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_castpd_si512(__m512d __A)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtsepi64_epi8(__mmask8 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtepu16_epi64(__m128i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_fmaddsub_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_cvtsepi64_epi16(__m512i __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi32_epi16(__m256i __O, __mmask16 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi16_epi64(__mmask8 __U, __m128i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask3_fnmadd_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_stream_load_si512(void const *__P)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_mul_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
static __inline__ float __DEFAULT_FN_ATTRS512 _mm512_reduce_min_ps(__m512 __V)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_sub_epi64(__mmask8 __U, __m512i __A, __m512i __B)
static __inline__ float __DEFAULT_FN_ATTRS512 _mm512_reduce_max_ps(__m512 __V)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtsepi64_epi16(__m128i __O, __mmask8 __M, __m512i __A)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_min_epi32(__mmask16 __M, __m512i __V)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_fmsubadd_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_srai_epi64(__m512i __A, unsigned int __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_rcp14_ps(__m512 __W, __mmask16 __U, __m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_srli_epi32(__mmask16 __U, __m512i __A, unsigned int __B)
static __inline__ int __DEFAULT_FN_ATTRS _mm512_kortestc(__mmask16 __A, __mmask16 __B)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_move_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_cvtepi32lo_pd(__m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_undefined_epi32(void)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepu32_pd(__mmask8 __U, __m256i __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtepi64_epi32(__m512i __A)
static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_reduce_add_pd(__m512d __W)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_loadu_epi32(void const *__P)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_castpd128_pd512(__m128d __A)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_mul_pd(__m512d __a, __m512d __b)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask3_fnmadd_sd(__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_permutevar_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512i __C)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_srl_epi64(__m512i __A, __m128i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_permutex2var_epi32(__mmask16 __U, __m512i __A, __m512i __I, __m512i __B)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_set1_ps(float __w)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_or_epi32(__mmask16 __k, __m512i __a, __m512i __b)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_sub_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_castsi512_ps(__m512i __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtpd_epi32(__mmask8 __U, __m512d __A)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi32_storeu_epi16(void *__P, __mmask16 __M, __m512i __A)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_loadu_ps(__mmask16 __U, void const *__P)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_add_epi64(__m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_max_epi32(__mmask16 __M, __m512i __A, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi32_pd(__mmask8 __U, __m256i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_expand_epi32(__mmask16 __U, __m512i __A)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_castps256_ps512(__m256 __a)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_add_ps(__m512 __a, __m512 __b)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtusepi32_epi8(__m128i __O, __mmask16 __M, __m512i __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi64_epi32(__m256i __O, __mmask8 __M, __m512i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_getexp_ps(__mmask16 __U, __m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_rorv_epi32(__m512i __A, __m512i __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_load_epi32(void const *__P)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_rcp14_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_set4_epi64(long long __A, long long __B, long long __C, long long __D)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_add_pd(__m512d __a, __m512d __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_and_epi32(__mmask16 __k, __m512i __a, __m512i __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_sllv_epi32(__m512i __X, __m512i __Y)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_or_epi64(__m512i __a, __m512i __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtepi8_epi64(__m128i __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_load_sd(__mmask8 __U, const double *__A)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi64_epi16(__mmask8 __M, __m512i __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_scalef_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_mov_epi32(__m512i __W, __mmask16 __U, __m512i __A)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_storeu_ps(void *__P, __m512 __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_mul_epi32(__m512i __X, __m512i __Y)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask3_fmadd_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepu16_epi64(__m512i __W, __mmask8 __U, __m128i __A)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtsepi64_storeu_epi8(void *__P, __mmask8 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_rorv_epi32(__mmask16 __U, __m512i __A, __m512i __B)
static __inline__ float __DEFAULT_FN_ATTRS512 _mm512_reduce_add_ps(__m512 __W)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_broadcastsd_pd(__mmask8 __M, __m128d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvttps_epi32(__m512i __W, __mmask16 __U, __m512 __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtsepi32_epi16(__m512i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_unpacklo_ps(__mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepu8_epi32(__m512i __W, __mmask16 __U, __m128i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_permutexvar_epi32(__m512i __X, __m512i __Y)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_add_epi32(__mmask16 __M, __m512i __W)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_add_pd(__mmask8 __U, __m512d __A, __m512d __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_rcp14_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_add_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_cvtss_sd(__mmask8 __U, __m128d __A, __m128 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_permutex2var_epi64(__m512i __A, __mmask8 __U, __m512i __I, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_scalef_pd(__mmask8 __U, __m512d __A, __m512d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_and_epi64(__mmask8 __k, __m512i __a, __m512i __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_srli_epi64(__m512i __W, __mmask8 __U, __m512i __A, unsigned int __B)
static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_add_pd(__mmask8 __M, __m512d __W)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_srai_epi64(__mmask8 __U, __m512i __A, unsigned int __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_srl_epi32(__m512i __A, __m128i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_andnot_epi32(__m512i __A, __m512i __B)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_load_pd(void const *__p)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_getexp_pd(__m512d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_abs_pd(__m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_rorv_epi64(__m512i __A, __m512i __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_load_epi64(void const *__P)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_ceil_pd(__m512d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_div_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_and_epi64(__m512i __W)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi8_epi64(__m512i __W, __mmask8 __U, __m128i __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_fnmsub_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_store_epi32(void *__P, __m512i __A)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_reduce_min_epi32(__m512i __V)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_unpackhi_pd(__m512d __a, __m512d __b)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_sub_ps(__mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_permutevar_ps(__mmask16 __U, __m512 __A, __m512i __C)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_sub_epi32(__mmask16 __U, __m512i __A, __m512i __B)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_and_epi32(__mmask16 __M, __m512i __W)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_cvtepi32_ps(__m512i __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_mask_move_ss(__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_slli_epi32(__m512i __W, __mmask16 __U, __m512i __A, unsigned int __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_unpacklo_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS512 _mm512_testn_epi32_mask(__m512i __A, __m512i __B)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtsepi32_storeu_epi8(void *__P, __mmask16 __M, __m512i __A)
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_max_epi64(__m512i __V)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_rcp14_ps(__m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_sra_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m128i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_rorv_epi64(__mmask8 __U, __m512i __A, __m512i __B)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_cvtepi32_pd(__m256i __A)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtsepi64_epi8(__m128i __O, __mmask8 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_rolv_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask3_fnmsub_sd(__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_expand_pd(__mmask8 __U, __m512d __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_permutexvar_ps(__m512i __X, __m512 __Y)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_max_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS _cvtu32_mask16(unsigned int __A)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS _mm512_kunpackb(__mmask16 __A, __mmask16 __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_compress_ps(__m512 __W, __mmask16 __U, __m512 __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_load_sd(__m128d __W, __mmask8 __U, const double *__A)
static __inline__ __m256 __DEFAULT_FN_ATTRS512 _mm512_cvtpd_ps(__m512d __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_fnmadd_ss(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_max_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_castpd256_pd512(__m256d __a)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_mask_storeu_epi64(void *__P, __mmask8 __U, __m512i __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_getexp_sd(__m128d __A, __m128d __B)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_scalef_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_sra_epi64(__m512i __A, __m128i __B)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_rcp14_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_set_epi16(short __e31, short __e30, short __e29, short __e28, short __e27, short __e26, short __e25, short __e24, short __e23, short __e22, short __e21, short __e20, short __e19, short __e18, short __e17, short __e16, short __e15, short __e14, short __e13, short __e12, short __e11, short __e10, short __e9, short __e8, short __e7, short __e6, short __e5, short __e4, short __e3, short __e2, short __e1, short __e0)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_expandloadu_pd(__mmask8 __U, void const *__P)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_broadcast_f64x4(__m256d __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_fmsub_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_cvtepi64_epi16(__m512i __A)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_sub_ps(__m512 __a, __m512 __b)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_mul_pd(__mmask8 __U, __m512d __A, __m512d __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_fmadd_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvttpd_epu32(__m256i __W, __mmask8 __U, __m512d __A)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtsepi32_storeu_epi16(void *__P, __mmask16 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_min_epi32(__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
static __inline__ void __DEFAULT_FN_ATTRS128 _mm_mask_store_ss(float *__W, __mmask8 __U, __m128 __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_fnmsub_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtps_pd(__mmask8 __U, __m256 __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_blend_epi64(__mmask8 __U, __m512i __A, __m512i __W)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_permutex2var_epi64(__mmask8 __U, __m512i __A, __m512i __I, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_permutex2var_pd(__m512d __A, __mmask8 __U, __m512i __I, __m512d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_expandloadu_epi64(__mmask8 __U, void const *__P)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_fnmadd_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_max_epi64(__mmask8 __M, __m512i __V)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_getexp_ps(__m512 __W, __mmask16 __U, __m512 __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_moveldup_ps(__m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_load_epi64(__m512i __W, __mmask8 __U, void const *__P)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_sra_epi32(__m512i __A, __m128i __B)
static __inline__ unsigned __DEFAULT_FN_ATTRS128 _mm_cvtss_u32(__m128 __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_add_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline void __DEFAULT_FN_ATTRS512 _mm512_storeu_epi32(void *__P, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_and_epi64(__m512i __src, __mmask8 __k, __m512i __a, __m512i __b)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi64_epi32(__mmask8 __M, __m512i __A)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_zextps128_ps512(__m128 __a)
Constructs a 512-bit floating-point vector of [16 x float] from a 128-bit floating-point vector of [4...
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_mov_pd(__m512d __W, __mmask8 __U, __m512d __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_srlv_epi64(__m512i __X, __m512i __Y)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtusepi64_storeu_epi8(void *__P, __mmask8 __M, __m512i __A)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_cvtepu32_pd(__m256i __A)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_load_pd(__mmask8 __U, void const *__P)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtepi32_epi64(__mmask8 __U, __m256i __X)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_rcp14_ps(__mmask16 __U, __m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_unpackhi_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_fmadd_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_sub_epi64(__m512i __A, __m512i __B)
static __inline __m256i __DEFAULT_FN_ATTRS512 _mm512_castsi512_si256(__m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepu8_epi64(__m512i __W, __mmask8 __U, __m128i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_xor_si512(__m512i __a, __m512i __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_and_si512(__m512i __a, __m512i __b)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_or_epi32(__mmask16 __M, __m512i __W)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_permutevar_pd(__mmask8 __U, __m512d __A, __m512i __C)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtph_ps(__mmask16 __U, __m256i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_srav_epi64(__mmask8 __U, __m512i __X, __m512i __Y)
#define _MM_FROUND_CUR_DIRECTION
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtusepi64_storeu_epi16(void *__P, __mmask8 __M, __m512i __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_mul_epi32(__mmask8 __M, __m512i __X, __m512i __Y)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_max_pd(__m512d __A, __m512d __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_floor_ps(__m512 __W, __mmask16 __U, __m512 __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_div_pd(__mmask8 __U, __m512d __A, __m512d __B)
static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_max_pd(__mmask8 __M, __m512d __V)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_fnmadd_sd(__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_unpackhi_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtsepi64_storeu_epi16(void *__P, __mmask8 __M, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_permutexvar_epi32(__m512i __W, __mmask16 __M, __m512i __X, __m512i __Y)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_loadu_epi32(__m512i __W, __mmask16 __U, void const *__P)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_maskz_fmsub_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_fnmsub_ps(__m512 __A, __m512 __B, __m512 __C)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_sub_pd(__m512d __a, __m512d __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_rorv_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_broadcast_i64x4(__m512i __O, __mmask8 __M, __m256i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_broadcastd_epi32(__mmask16 __M, __m128i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_permutexvar_epi64(__mmask8 __M, __m512i __X, __m512i __Y)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_rcp14_pd(__mmask8 __U, __m512d __A)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_fmsubadd_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
static __inline__ int __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_mul_epi32(__mmask16 __M, __m512i __W)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi32_epi8(__m128i __O, __mmask16 __M, __m512i __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_rsqrt14_sd(__m128d __A, __m128d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_srai_epi32(__m512i __W, __mmask16 __U, __m512i __A, unsigned int __B)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_div_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __mmask16 __DEFAULT_FN_ATTRS _mm512_kxnor(__mmask16 __A, __mmask16 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_slli_epi64(__m512i __W, __mmask8 __U, __m512i __A, unsigned int __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepu32lo_pd(__m512d __W, __mmask8 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_srav_epi32(__mmask16 __U, __m512i __X, __m512i __Y)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_set4_epi32(int __A, int __B, int __C, int __D)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_fmsub_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_max_epu64(__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_slli_epi32(__m512i __A, unsigned int __B)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_set1_epi64(__mmask8 __M, long long __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_srl_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m128i __B)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_permutex2var_pd(__m512d __A, __m512i __I, __m512d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_add_epi64(__mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_or_epi32(__m512i __a, __m512i __b)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_rsqrt14_ps(__m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi32_epi64(__m512i __W, __mmask8 __U, __m256i __X)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvtepi16_epi32(__m256i __A)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtsepi64_epi32(__m512i __A)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_zextpd128_pd512(__m128d __a)
Constructs a 512-bit floating-point vector of [8 x double] from a 128-bit floating-point vector of [2...
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_broadcast_f64x4(__m512d __O, __mmask8 __M, __m256d __A)
#define _mm512_cmpeq_epi64_mask(A, B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_moveldup_ps(__m512 __W, __mmask16 __U, __m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_sllv_epi32(__mmask16 __U, __m512i __X, __m512i __Y)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_sub_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_or_epi64(__mmask8 __M, __m512i __W)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_unpackhi_pd(__mmask8 __U, __m512d __A, __m512d __B)
static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtsepi64_epi32(__m256i __O, __mmask8 __M, __m512i __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_loadu_epi64(void const *__P)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_rorv_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_rolv_epi64(__m512i __A, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_cvtps_pd(__m512d __W, __mmask8 __U, __m256 __A)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_maskz_move_ss(__mmask8 __U, __m128 __A, __m128 __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvttps_epi32(__mmask16 __U, __m512 __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_or_epi64(__m512i __src, __mmask8 __k, __m512i __a, __m512i __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_castsi256_si512(__m256i __A)
static __inline __mmask16 __DEFAULT_FN_ATTRS _mm512_knot(__mmask16 __M)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mullox_epi64(__m512i __A, __m512i __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_maskz_permutexvar_pd(__mmask8 __U, __m512i __X, __m512d __Y)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_rsqrt14_ps(__m512 __W, __mmask16 __U, __m512 __A)
static __inline __m512i __DEFAULT_FN_ATTRS512 _mm512_setzero_si512(void)
static __inline __m128d __DEFAULT_FN_ATTRS512 _mm512_castpd512_pd128(__m512d __a)
static __inline__ unsigned int __DEFAULT_FN_ATTRS _cvtmask16_u32(__mmask16 __A)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_permutex2var_ps(__m512 __A, __mmask16 __U, __m512i __I, __m512 __B)
static __inline__ __m512 __DEFAULT_FN_ATTRS512 _mm512_mask_sqrt_ps(__m512 __W, __mmask16 __U, __m512 __A)
static __inline __m512 __DEFAULT_FN_ATTRS512 _mm512_unpackhi_ps(__m512 __a, __m512 __b)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_expandloadu_epi64(__m512i __W, __mmask8 __U, void const *__P)
static __inline __m128 __DEFAULT_FN_ATTRS512 _mm512_castps512_ps128(__m512 __a)
static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_cvtu32_ss(__m128 __A, unsigned __B)
static __inline__ __m512d __DEFAULT_FN_ATTRS512 _mm512_mask3_fmsub_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_maskz_sub_sd(__mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_srai_epi32(__m512i __A, unsigned int __B)
static __inline__ __mmask8 __DEFAULT_FN_ATTRS512 _mm512_mask_test_epi64_mask(__mmask8 __U, __m512i __A, __m512i __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_maskz_mov_epi64(__mmask8 __U, __m512i __A)
static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_or_epi64(__m512i __W)
static __inline__ __m128i __DEFAULT_FN_ATTRS512 _mm512_maskz_cvtusepi64_epi8(__mmask8 __M, __m512i __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask3_fmsub_sd(__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_castsi512_pd(__m512i __A)
static __inline__ unsigned int __DEFAULT_FN_ATTRS512 _mm512_mask_reduce_min_epu32(__mmask16 __M, __m512i __V)
static __inline__ void __DEFAULT_FN_ATTRS512 _mm512_mask_cvtepi32_storeu_epi8(void *__P, __mmask16 __M, __m512i __A)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_scalef_sd(__m128d __A, __m128d __B)
static __inline __m512d __DEFAULT_FN_ATTRS512 _mm512_mask_blend_pd(__mmask8 __U, __m512d __A, __m512d __W)
static __inline__ __m128d __DEFAULT_FN_ATTRS128 _mm_mask_rcp14_sd(__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_mask_compress_epi32(__m512i __W, __mmask16 __U, __m512i __A)
static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_sll_epi64(__m512i __A, __m128i __B)
static __inline __m256 __DEFAULT_FN_ATTRS _mm256_setzero_ps(void)
Constructs a 256-bit floating-point vector of [8 x float] with all vector elements initialized to zer...
static __inline__ __m256 __DEFAULT_FN_ATTRS _mm256_undefined_ps(void)
Create a 256-bit vector of [8 x float] with undefined values.
static __inline__ __m256i __DEFAULT_FN_ATTRS _mm256_undefined_si256(void)
Create a 256-bit integer vector with undefined values.
static __inline __m256d __DEFAULT_FN_ATTRS _mm256_setzero_pd(void)
Constructs a 256-bit floating-point vector of [4 x double] with all vector elements initialized to ze...
static __inline __m256i __DEFAULT_FN_ATTRS _mm256_setzero_si256(void)
Constructs a 256-bit integer vector initialized to zero.
static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_add_sd(__m128d __a, __m128d __b)
Adds lower double-precision values in both operands and returns the sum in the lower 64 bits of the r...
static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_div_sd(__m128d __a, __m128d __b)
Divides the lower double-precision value of the first operand by the lower double-precision value of ...
static __inline__ void int __a
static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_mul_sd(__m128d __a, __m128d __b)
Multiplies lower double-precision values in both operands and returns the product in the lower 64 bit...
static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_move_sd(__m128d __a, __m128d __b)
Constructs a 128-bit floating-point vector of [2 x double].
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_undefined_si128(void)
Generates a 128-bit vector of [4 x i32] with unspecified content.
static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_sub_sd(__m128d __a, __m128d __b)
Subtracts the lower double-precision value of the second operand from the lower double-precision valu...
static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_setzero_pd(void)
Constructs a 128-bit floating-point vector of [2 x double] initialized to zero.
static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_setzero_si128(void)
Creates a 128-bit integer vector initialized to zero.
static __inline__ void short __D
struct __storeu_i16 *__P __v
static __inline__ void const void * __src
__inline unsigned int unsigned int unsigned int * __P
__inline unsigned int unsigned int __Y
static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_setzero_ps(void)
Constructs a 128-bit floating-point vector of [4 x float] initialized to zero.
static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_sub_ss(__m128 __a, __m128 __b)
Subtracts the 32-bit float value in the low-order bits of the second operand from the corresponding v...
static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_move_ss(__m128 __a, __m128 __b)
Constructs a 128-bit floating-point vector of [4 x float].
static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_mul_ss(__m128 __a, __m128 __b)
Multiplies two 32-bit float values in the low-order bits of the operands.
static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_add_ss(__m128 __a, __m128 __b)
Adds the 32-bit float values in the low-order bits of the operands.
static __inline__ __m128 __DEFAULT_FN_ATTRS _mm_div_ss(__m128 __a, __m128 __b)
Divides the value in the low-order 32 bits of the first operand by the corresponding value in the sec...