/freebsd/contrib/llvm-project/clang/lib/Headers/ |
H A D | xmmintrin.h | 80 return (__m128)((__v4sf)__a + (__v4sf)__b); in _mm_add_ps() 123 return (__m128)((__v4sf)__a - (__v4sf)__b); in _mm_sub_ps() 165 return (__m128)((__v4sf)__a * (__v4sf)__b); in _mm_mul_ps() 206 return (__m128)((__v4sf)__a / (__v4sf)__b); in _mm_div_ps() 335 return __builtin_ia32_minss((__v4sf)__a, (__v4sf)__b); in _mm_min_ss() 354 return __builtin_ia32_minps((__v4sf)__a, (__v4sf)__b); in _mm_min_ps() 377 return __builtin_ia32_maxss((__v4sf)__a, (__v4sf)__b); in _mm_max_ss() 396 return __builtin_ia32_maxps((__v4sf)__a, (__v4sf)__b); in _mm_max_ps() 622 (__v4sf)__builtin_ia32_cmpltss((__v4sf)__b, (__v4sf)__a), in _mm_cmpgt_ss() 667 (__v4sf)__builtin_ia32_cmpless((__v4sf)__b, (__v4sf)__a), in _mm_cmpge_ss() [all …]
|
H A D | fma4intrin.h | 26 return (__m128)__builtin_ia32_vfmaddps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C); in _mm_macc_ps() 38 return (__m128)__builtin_ia32_vfmaddss((__v4sf)__A, (__v4sf)__B, (__v4sf)__C); in _mm_macc_ss() 50 return (__m128)__builtin_ia32_vfmaddps((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C); in _mm_msub_ps() 62 return (__m128)__builtin_ia32_vfmaddss((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C); in _mm_msub_ss() 74 return (__m128)__builtin_ia32_vfmaddps(-(__v4sf)__A, (__v4sf)__B, (__v4sf)__C); in _mm_nmacc_ps() 86 return (__m128)__builtin_ia32_vfmaddss(-(__v4sf)__A, (__v4sf)__B, (__v4sf)__C); in _mm_nmacc_ss() 98 return (__m128)__builtin_ia32_vfmaddps(-(__v4sf)__A, (__v4sf)__B, -(__v4sf)__C); in _mm_nmsub_ps() 110 return (__m128)__builtin_ia32_vfmaddss(-(__v4sf)__A, (__v4sf)__B, -(__v4sf)__C); in _mm_nmsub_ss() 122 return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C); in _mm_maddsub_ps() 134 return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C); in _mm_msubadd_ps()
|
H A D | fmaintrin.h | 38 return (__m128)__builtin_ia32_vfmaddps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C); in _mm_fmadd_ps() 86 return (__m128)__builtin_ia32_vfmaddss3((__v4sf)__A, (__v4sf)__B, (__v4sf)__C); in _mm_fmadd_ss() 134 return (__m128)__builtin_ia32_vfmaddps((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C); in _mm_fmsub_ps() 182 return (__m128)__builtin_ia32_vfmaddss3((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C); in _mm_fmsub_ss() 230 return (__m128)__builtin_ia32_vfmaddps(-(__v4sf)__A, (__v4sf)__B, (__v4sf)__C); in _mm_fnmadd_ps() 278 return (__m128)__builtin_ia32_vfmaddss3((__v4sf)__A, -(__v4sf)__B, (__v4sf)__C); in _mm_fnmadd_ss() 326 return (__m128)__builtin_ia32_vfmaddps(-(__v4sf)__A, (__v4sf)__B, -(__v4sf)__C); in _mm_fnmsub_ps() 374 return (__m128)__builtin_ia32_vfmaddss3((__v4sf)__A, -(__v4sf)__B, -(__v4sf)__C); in _mm_fnmsub_ss() 428 return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C); in _mm_fmaddsub_ps() 478 return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C); in _mm_fmsubadd_ps()
|
H A D | pmmintrin.h | 60 return __builtin_ia32_addsubps((__v4sf)__a, (__v4sf)__b); in _mm_addsub_ps() 83 return __builtin_ia32_haddps((__v4sf)__a, (__v4sf)__b); in _mm_hadd_ps() 106 return __builtin_ia32_hsubps((__v4sf)__a, (__v4sf)__b); in _mm_hsub_ps() 128 return __builtin_shufflevector((__v4sf)__a, (__v4sf)__a, 1, 1, 3, 3); in _mm_movehdup_ps() 149 return __builtin_shufflevector((__v4sf)__a, (__v4sf)__a, 0, 0, 2, 2); in _mm_moveldup_ps()
|
H A D | avx512vldqintrin.h | 111 (__v4sf)__W); in _mm_mask_andnot_ps() 167 (__v4sf)__W); in _mm_mask_and_ps() 223 (__v4sf)__W); in _mm_mask_xor_ps() 279 (__v4sf)__W); in _mm_mask_or_ps() 498 (__v4sf) _mm_setzero_ps(), in _mm_cvtepi64_ps() 505 (__v4sf) __W, in _mm_mask_cvtepi64_ps() 512 (__v4sf) _mm_setzero_ps(), in _mm_maskz_cvtepi64_ps() 744 (__v4sf) _mm_setzero_ps(), in _mm_cvtepu64_ps() 751 (__v4sf) __W, in _mm_mask_cvtepu64_ps() 758 (__v4sf) _mm_setzero_ps(), in _mm_maskz_cvtepu64_ps() [all …]
|
H A D | avx512erintrin.h | 115 ((__m128)__builtin_ia32_rsqrt28ss_round_mask((__v4sf)(__m128)(A), \ 116 (__v4sf)(__m128)(B), \ 121 ((__m128)__builtin_ia32_rsqrt28ss_round_mask((__v4sf)(__m128)(A), \ 122 (__v4sf)(__m128)(B), \ 218 ((__m128)__builtin_ia32_rcp28ss_round_mask((__v4sf)(__m128)(A), \ 219 (__v4sf)(__m128)(B), \ 224 ((__m128)__builtin_ia32_rcp28ss_round_mask((__v4sf)(__m128)(A), \ 225 (__v4sf)(__m128)(B), \ 226 (__v4sf)(__m128)(S), \ 230 ((__m128)__builtin_ia32_rcp28ss_round_mask((__v4sf)(__m128)(A), \ [all …]
|
H A D | avx512vlbf16intrin.h | 41 return (__m128bh)__builtin_ia32_cvtne2ps2bf16_128((__v4sf) __A, in _mm_cvtne2ps_pbh() 42 (__v4sf) __B); in _mm_cvtne2ps_pbh() 166 ((__m128bh)__builtin_ia32_vcvtneps2bf16128((__v4sf)(A))) 185 return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A, in _mm_mask_cvtneps_pbh() 205 return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A, in _mm_maskz_cvtneps_pbh() 278 return (__m128)__builtin_ia32_dpbf16ps_128((__v4sf)__D, in _mm_dpbf16_ps() 303 (__v4sf)_mm_dpbf16_ps(__D, __A, __B), in _mm_mask_dpbf16_ps() 304 (__v4sf)__D); in _mm_mask_dpbf16_ps() 328 (__v4sf)_mm_setzero_si128()); in _mm_maskz_dpbf16_ps() 411 __v4sf __V = {__A, 0, 0, 0}; in _mm_cvtness_sbh() [all …]
|
H A D | avx512fintrin.h | 345 return (__m512)__builtin_shufflevector((__v4sf) __A, (__v4sf) __A, in _mm512_broadcastss_ps() 1633 (__v4sf) in _mm_rsqrt14_ss() 1642 (__v4sf) __B, in _mm_mask_rsqrt14_ss() 1643 (__v4sf) __W, in _mm_mask_rsqrt14_ss() 1651 (__v4sf) __B, in _mm_maskz_rsqrt14_ss() 1750 (__v4sf) __B, in _mm_mask_rcp14_ss() 1751 (__v4sf) __W, in _mm_mask_rcp14_ss() 1759 (__v4sf) __B, in _mm_maskz_rcp14_ss() 6582 (__v4sf)( __B), (__v4sf) _mm_setzero_ps(), in _mm_scalef_ss() 6828 return (__m512)__builtin_shufflevector((__v4sf)__A, (__v4sf)__A, in _mm512_broadcast_f32x4() [all …]
|
H A D | avx512vlintrin.h | 2003 (__v4sf) in _mm_maskz_cvtpd_ps() 2591 (__v4sf) in _mm_maskz_expandloadu_ps() 2756 (__v4sf) in _mm_getexp_ps() 2771 (__v4sf) in _mm_maskz_getexp_ps() 3429 (__v4sf) in _mm_scalef_ps() 3446 (__v4sf) in _mm_maskz_scalef_ps() 5482 (__v4sf) in _mm_maskz_load_ps() 5654 (__v4sf) in _mm_maskz_loadu_ps() 5993 (__v4sf) in _mm_rcp14_ps() 6798 return (__m256)__builtin_shufflevector((__v4sf)__A, (__v4sf)__A, in _mm256_broadcast_f32x4() [all …]
|
H A D | avx512fp16intrin.h | 2947 (__v4sf)__A, (__v4sf)(__B), (__v4sf)(__C), __U, _MM_FROUND_CUR_DIRECTION); in _mm_mask_fcmadd_sch() 2960 (__v4sf)__A, (__v4sf)__B, (__v4sf)__C, __U, _MM_FROUND_CUR_DIRECTION); in _mm_mask3_fcmadd_sch() 2965 (__v4sf)(__m128h)(A), (__v4sf)(__m128h)(B), (__v4sf)(__m128h)(C), \ 2970 (__v4sf)(__m128h)(A), (__v4sf)(__m128h)(B), (__v4sf)(__m128h)(C), \ 2975 (__v4sf)(__m128h)(A), (__v4sf)(__m128h)(B), (__v4sf)(__m128h)(C), \ 2980 (__v4sf)(__m128h)(A), (__v4sf)(__m128h)(B), (__v4sf)(__m128h)(C), \ 3007 (__v4sf)__A, (__v4sf)__B, (__v4sf)__C, __U, _MM_FROUND_CUR_DIRECTION); in _mm_mask3_fmadd_sch() 3033 (__v4sf)__A, (__v4sf)__B, (__v4sf)_mm_undefined_ph(), (__mmask8)-1, in _mm_fcmul_sch() 3047 (__v4sf)__A, (__v4sf)__B, (__v4sf)_mm_setzero_ph(), (__mmask8)__U, in _mm_maskz_fcmul_sch() 3069 (__v4sf)__A, (__v4sf)__B, (__v4sf)_mm_undefined_ph(), (__mmask8)-1, in _mm_fmul_sch() [all …]
|
H A D | smmintrin.h | 240 ((__m128)__builtin_ia32_roundps((__v4sf)(__m128)(X), (M))) 281 ((__m128)__builtin_ia32_roundss((__v4sf)(__m128)(X), (__v4sf)(__m128)(Y), \ 413 ((__m128)__builtin_ia32_blendps((__v4sf)(__m128)(V1), (__v4sf)(__m128)(V2), \ 466 return (__m128)__builtin_ia32_blendvps((__v4sf)__V1, (__v4sf)__V2, in _mm_blendv_ps() 467 (__v4sf)__M); in _mm_blendv_ps() 597 ((__m128)__builtin_ia32_dpps((__v4sf)(__m128)(X), (__v4sf)(__m128)(Y), (M))) 864 int, __builtin_ia32_vec_ext_v4sf((__v4sf)(__m128)(X), (int)(N))) 870 (D) = __builtin_ia32_vec_ext_v4sf((__v4sf)(__m128)(X), (int)(N)); \
|
H A D | f16cintrin.h | 42 __v4sf __r = __builtin_ia32_vcvtph2ps(__v); in _cvtsh_ss() 69 (unsigned short)(((__v8hi)__builtin_ia32_vcvtps2ph((__v4sf){a, 0, 0, 0}, \ 96 ((__m128i)__builtin_ia32_vcvtps2ph((__v4sf)(__m128)(a), (imm)))
|
H A D | avx512vlfp16intrin.h | 350 (__mmask8)__U, (__v4sf)_mm_conj_pch(__A), (__v4sf)__W); in _mm_mask_conj_pch() 356 (__mmask8)__U, (__v4sf)_mm_conj_pch(__A), (__v4sf)_mm_setzero_ps()); in _mm_maskz_conj_pch() 1794 (__v4sf)__A, (__v4sf)__B, (__v4sf)_mm_undefined_ph(), (__mmask8)-1); in _mm_fcmul_pch() 1806 (__v4sf)__A, (__v4sf)__B, (__v4sf)_mm_setzero_ph(), (__mmask8)__U); in _mm_maskz_fcmul_pch() 1838 __builtin_ia32_vfcmaddcph128_mask((__v4sf)__A, (__v4sf)(__m128h)__B, in _mm_mask_fcmadd_pch() 1840 (__v4sf)__A); in _mm_mask_fcmadd_pch() 1852 (__v4sf)__A, (__v4sf)__B, (__v4sf)__C, (__mmask8)__U); in _mm_maskz_fcmadd_pch() 1886 (__v4sf)__A, (__v4sf)__B, (__v4sf)_mm_undefined_ph(), (__mmask8)-1); in _mm_fmul_pch() 1900 (__v4sf)__A, (__v4sf)__B, (__v4sf)_mm_setzero_ph(), (__mmask8)__U); in _mm_maskz_fmul_pch() 1932 __builtin_ia32_vfmaddcph128_mask((__v4sf)__A, (__v4sf)__B, (__v4sf)__C, in _mm_mask_fmadd_pch() [all …]
|
H A D | avxintrin.h | 1727 ((__m128)__builtin_ia32_cmpps((__v4sf)(__m128)(a), \ 1728 (__v4sf)(__m128)(b), (c))) 1965 ((__m128)__builtin_ia32_cmpss((__v4sf)(__m128)(a), \ 2631 return __builtin_ia32_vtestzps((__v4sf)__a, (__v4sf)__b); in _mm_testz_ps() 2660 return __builtin_ia32_vtestcps((__v4sf)__a, (__v4sf)__b); in _mm_testc_ps() 2690 return __builtin_ia32_vtestnzcps((__v4sf)__a, (__v4sf)__b); in _mm_testnzc_ps() 3111 return (__m256)__builtin_shufflevector((__v4sf)__b, (__v4sf)__b, in _mm256_broadcast_ps() 3552 __builtin_ia32_maskstoreps((__v4sf *)__p, (__v4si)__m, (__v4sf)__a); in _mm_maskstore_ps() 4537 return __builtin_shufflevector((__v4sf)__a, in _mm256_castps128_ps256() 4596 return __builtin_shufflevector((__v4sf)__a, (__v4sf)_mm_setzero_ps(), 0, 1, 2, 3, 4, 5, 6, 7); in _mm256_zextps128_ps256() [all …]
|
H A D | avx512dqintrin.h | 981 ((__m128)__builtin_ia32_reducess_mask((__v4sf)(__m128)(A), \ 982 (__v4sf)(__m128)(B), \ 987 ((__m128)__builtin_ia32_reducess_mask((__v4sf)(__m128)(A), \ 988 (__v4sf)(__m128)(B), \ 993 ((__m128)__builtin_ia32_reducess_mask((__v4sf)(__m128)(A), \ 994 (__v4sf)(__m128)(B), \ 1000 ((__m128)__builtin_ia32_reducess_mask((__v4sf)(__m128)(A), \ 1001 (__v4sf)(__m128)(B), \ 1006 ((__m128)__builtin_ia32_reducess_mask((__v4sf)(__m128)(A), \ 1007 (__v4sf)(__m128)(B), \ [all …]
|
H A D | avx2intrin.h | 3006 return (__m128)__builtin_shufflevector((__v4sf)__X, (__v4sf)__X, 0, 0, 0, 0); in _mm_broadcastss_ps() 3040 return (__m256)__builtin_shufflevector((__v4sf)__X, (__v4sf)__X, 0, 0, 0, 0, 0, 0, 0, 0); in _mm256_broadcastss_ps() 4192 ((__m128)__builtin_ia32_gatherd_ps((__v4sf)(__m128)(a), \ 4195 (__v4sf)(__m128)(mask), (s))) 4291 ((__m128)__builtin_ia32_gatherq_ps((__v4sf)(__m128)(a), \ 4294 (__v4sf)(__m128)(mask), (s))) 4339 ((__m128)__builtin_ia32_gatherq_ps256((__v4sf)(__m128)(a), \ 4342 (__v4sf)(__m128)(mask), (s))) 4903 ((__m128)__builtin_ia32_gatherd_ps((__v4sf)_mm_undefined_ps(), \ 4976 ((__m128)__builtin_ia32_gatherq_ps((__v4sf)_mm_undefined_ps(), \ [all …]
|
H A D | ammintrin.h | 178 __builtin_ia32_movntss((float *)__p, (__v4sf)__a); in _mm_stream_ss()
|
H A D | avxneconvertintrin.h | 449 return (__m128bh)__builtin_ia32_vcvtneps2bf16128((__v4sf)__A); in _mm_cvtneps_avx_pbh()
|
H A D | xopintrin.h | 723 ((__m128)__builtin_ia32_vpermil2ps((__v4sf)(__m128)(X), (__v4sf)(__m128)(Y), \ 734 return (__m128)__builtin_ia32_vfrczss((__v4sf)__A); in _mm_frcz_ss() 746 return (__m128)__builtin_ia32_vfrczps((__v4sf)__A); in _mm_frcz_ps()
|
H A D | emmintrin.h | 1278 __builtin_shufflevector((__v4sf)__a, (__v4sf)__a, 0, 1), __v2df); in _mm_cvtps_pd() 1355 return (__m128)__builtin_ia32_cvtsd2ss((__v4sf)__a, (__v2df)__b); in _mm_cvtsd_ss() 3251 return (__m128) __builtin_convertvector((__v4si)__a, __v4sf); in _mm_cvtepi32_ps() 3265 return (__m128i)__builtin_ia32_cvtps2dq((__v4sf)__a); in _mm_cvtps_epi32() 3280 return (__m128i)__builtin_ia32_cvttps2dq((__v4sf)__a); in _mm_cvttps_epi32()
|
/freebsd/contrib/llvm-project/clang/lib/Headers/ppc_wrappers/ |
H A D | xmmintrin.h | 150 vec_st((__v4sf)__A, 0, (__v4sf *)__P); in _mm_store_ps() 178 __v4sf __va = vec_splat((__v4sf)__A, 0); in _mm_store1_ps() 201 return (vec_sel((__v4sf)__A, (__v4sf)__B, __mask)); in _mm_move_ss() 330 return (__m128)((__v4sf)__A + (__v4sf)__B); in _mm_add_ps() 336 return (__m128)((__v4sf)__A - (__v4sf)__B); in _mm_sub_ps() 342 return (__m128)((__v4sf)__A * (__v4sf)__B); in _mm_mul_ps() 348 return (__m128)((__v4sf)__A / (__v4sf)__B); in _mm_div_ps() 453 return ((__m128)vec_and((__v4sf)__A, (__v4sf)__B)); in _mm_and_ps() 466 return ((__m128)vec_or((__v4sf)__A, (__v4sf)__B)); in _mm_or_ps() 472 return ((__m128)vec_xor((__v4sf)__A, (__v4sf)__B)); in _mm_xor_ps() [all …]
|
H A D | pmmintrin.h | 51 const __v4sf __even_n0 = {-0.0, 0.0, -0.0, 0.0}; in _mm_addsub_ps() 52 __v4sf __even_neg_Y = vec_xor(__Y, __even_n0); in _mm_addsub_ps() 73 return (__m128)vec_add(vec_perm((__v4sf)__X, (__v4sf)__Y, __xform2), in _mm_hadd_ps() 74 vec_perm((__v4sf)__X, (__v4sf)__Y, __xform1)); in _mm_hadd_ps() 86 return (__m128)vec_sub(vec_perm((__v4sf)__X, (__v4sf)__Y, __xform2), in _mm_hsub_ps() 87 vec_perm((__v4sf)__X, (__v4sf)__Y, __xform1)); in _mm_hsub_ps()
|
H A D | smmintrin.h | 159 __v4sf __r; in _mm_round_ps() 203 __r = vec_rint((__v4sf)__A); in _mm_round_ps() 215 __r = vec_floor((__v4sf)__A); in _mm_round_ps() 219 __r = vec_ceil((__v4sf)__A); in _mm_round_ps() 223 __r = vec_trunc((__v4sf)__A); in _mm_round_ps() 226 __r = vec_rint((__v4sf)__A); in _mm_round_ps() 253 __v4sf __r = (__v4sf)__A; in _mm_round_ss() 254 __r[0] = ((__v4sf)__B)[0]; in _mm_round_ss() 379 return (__m128)vec_blendv((__v4sf)__A, (__v4sf)__B, (__v4su)__mask); in _mm_blendv_ps()
|
H A D | emmintrin.h | 881 __v4sf __result; in _mm_cvtpd_ps() 967 __v4sf __rounded; in _mm_cvtps_epi32() 970 __rounded = vec_rint((__v4sf)__A); in _mm_cvtps_epi32() 980 __result = vec_cts((__v4sf)__A, 0); in _mm_cvttps_epi32() 989 return (__m128d)vec_doubleh((__v4sf)__A); in _mm_cvtps_pd() 993 __v4sf __a = (__v4sf)__A; in _mm_cvtps_pd() 994 __v4sf __temp; in _mm_cvtps_pd() 1067 __v4sf __result = (__v4sf)__A; in _mm_cvtsd_ss() 1070 __v4sf __temp_s; in _mm_cvtsd_ss() 1117 __v4sf __temp = vec_splat((__v4sf)__B, 0); in _mm_cvtss_sd() [all …]
|