/external/clang/lib/Headers/ |
D | xmmintrin.h | 31 typedef float __m128 __attribute__((__vector_size__(16))); typedef 42 static __inline__ __m128 __DEFAULT_FN_ATTRS 43 _mm_add_ss(__m128 __a, __m128 __b) in _mm_add_ss() 49 static __inline__ __m128 __DEFAULT_FN_ATTRS 50 _mm_add_ps(__m128 __a, __m128 __b) in _mm_add_ps() 55 static __inline__ __m128 __DEFAULT_FN_ATTRS 56 _mm_sub_ss(__m128 __a, __m128 __b) in _mm_sub_ss() 62 static __inline__ __m128 __DEFAULT_FN_ATTRS 63 _mm_sub_ps(__m128 __a, __m128 __b) in _mm_sub_ps() 68 static __inline__ __m128 __DEFAULT_FN_ATTRS [all …]
|
D | fma4intrin.h | 36 static __inline__ __m128 __DEFAULT_FN_ATTRS 37 _mm_macc_ps(__m128 __A, __m128 __B, __m128 __C) in _mm_macc_ps() 39 return (__m128)__builtin_ia32_vfmaddps(__A, __B, __C); in _mm_macc_ps() 48 static __inline__ __m128 __DEFAULT_FN_ATTRS 49 _mm_macc_ss(__m128 __A, __m128 __B, __m128 __C) in _mm_macc_ss() 51 return (__m128)__builtin_ia32_vfmaddss(__A, __B, __C); in _mm_macc_ss() 60 static __inline__ __m128 __DEFAULT_FN_ATTRS 61 _mm_msub_ps(__m128 __A, __m128 __B, __m128 __C) in _mm_msub_ps() 63 return (__m128)__builtin_ia32_vfmsubps(__A, __B, __C); in _mm_msub_ps() 72 static __inline__ __m128 __DEFAULT_FN_ATTRS [all …]
|
D | fmaintrin.h | 34 static __inline__ __m128 __DEFAULT_FN_ATTRS 35 _mm_fmadd_ps(__m128 __A, __m128 __B, __m128 __C) in _mm_fmadd_ps() 37 return (__m128)__builtin_ia32_vfmaddps(__A, __B, __C); in _mm_fmadd_ps() 46 static __inline__ __m128 __DEFAULT_FN_ATTRS 47 _mm_fmadd_ss(__m128 __A, __m128 __B, __m128 __C) in _mm_fmadd_ss() 49 return (__m128)__builtin_ia32_vfmaddss(__A, __B, __C); in _mm_fmadd_ss() 58 static __inline__ __m128 __DEFAULT_FN_ATTRS 59 _mm_fmsub_ps(__m128 __A, __m128 __B, __m128 __C) in _mm_fmsub_ps() 61 return (__m128)__builtin_ia32_vfmsubps(__A, __B, __C); in _mm_fmsub_ps() 70 static __inline__ __m128 __DEFAULT_FN_ATTRS [all …]
|
D | avx512vldqintrin.h | 129 static __inline__ __m128 __DEFAULT_FN_ATTRS 130 _mm_mask_andnot_ps (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) { in _mm_mask_andnot_ps() 131 return (__m128) __builtin_ia32_andnps128_mask ((__v4sf) __A, in _mm_mask_andnot_ps() 137 static __inline__ __m128 __DEFAULT_FN_ATTRS 138 _mm_maskz_andnot_ps (__mmask8 __U, __m128 __A, __m128 __B) { in _mm_maskz_andnot_ps() 139 return (__m128) __builtin_ia32_andnps128_mask ((__v4sf) __A, in _mm_maskz_andnot_ps() 197 static __inline__ __m128 __DEFAULT_FN_ATTRS 198 _mm_mask_and_ps (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) { in _mm_mask_and_ps() 199 return (__m128) __builtin_ia32_andps128_mask ((__v4sf) __A, in _mm_mask_and_ps() 205 static __inline__ __m128 __DEFAULT_FN_ATTRS [all …]
|
D | pmmintrin.h | 38 static __inline__ __m128 __DEFAULT_FN_ATTRS 39 _mm_addsub_ps(__m128 __a, __m128 __b) in _mm_addsub_ps() 44 static __inline__ __m128 __DEFAULT_FN_ATTRS 45 _mm_hadd_ps(__m128 __a, __m128 __b) in _mm_hadd_ps() 50 static __inline__ __m128 __DEFAULT_FN_ATTRS 51 _mm_hsub_ps(__m128 __a, __m128 __b) in _mm_hsub_ps() 56 static __inline__ __m128 __DEFAULT_FN_ATTRS 57 _mm_movehdup_ps(__m128 __a) in _mm_movehdup_ps() 62 static __inline__ __m128 __DEFAULT_FN_ATTRS 63 _mm_moveldup_ps(__m128 __a) in _mm_moveldup_ps()
|
/external/vulkan-validation-layers/libs/glm/detail/ |
D | intrinsic_common.inl | 52 static const __m128 GLM_VAR_USED zero = _mm_setzero_ps(); 53 static const __m128 GLM_VAR_USED one = _mm_set_ps1(1.0f); 54 static const __m128 GLM_VAR_USED minus_one = _mm_set_ps1(-1.0f); 55 static const __m128 GLM_VAR_USED two = _mm_set_ps1(2.0f); 56 static const __m128 GLM_VAR_USED three = _mm_set_ps1(3.0f); 57 static const __m128 GLM_VAR_USED pi = _mm_set_ps1(3.1415926535897932384626433832795f); 58 static const __m128 GLM_VAR_USED hundred_eighty = _mm_set_ps1(180.f); 59 …static const __m128 GLM_VAR_USED pi_over_hundred_eighty = _mm_set_ps1(0.01745329251994329576923690… 60 …static const __m128 GLM_VAR_USED hundred_eighty_over_pi = _mm_set_ps1(57.2957795130823208767981548… 63 static const __m128 GLM_VAR_USED abs4Mask = _mm_set_ps1(absMask.f); [all …]
|
D | intrinsic_common.hpp | 41 __m128 sse_abs_ps(__m128 x); 43 __m128 sse_sgn_ps(__m128 x); 46 __m128 sse_flr_ps(__m128 v); 49 __m128 sse_trc_ps(__m128 v); 52 __m128 sse_nd_ps(__m128 v); 55 __m128 sse_rde_ps(__m128 v); 57 __m128 sse_rnd_ps(__m128 x); 59 __m128 sse_ceil_ps(__m128 v); 61 __m128 sse_frc_ps(__m128 x); 63 __m128 sse_mod_ps(__m128 x, __m128 y); [all …]
|
D | intrinsic_matrix.inl | 32 static const __m128 GLM_VAR_USED _m128_rad_ps = _mm_set_ps1(3.141592653589793238462643383279f / 180… 33 static const __m128 GLM_VAR_USED _m128_deg_ps = _mm_set_ps1(180.f / 3.14159265358979323846264338327… 38 __m128 const in1[4], 39 __m128 const in2[4], 40 __m128 out[4] 49 GLM_FUNC_QUALIFIER void sse_add_ps(__m128 const in1[4], __m128 const in2[4], __m128 out[4]) 59 GLM_FUNC_QUALIFIER void sse_sub_ps(__m128 const in1[4], __m128 const in2[4], __m128 out[4]) 69 GLM_FUNC_QUALIFIER __m128 sse_mul_ps(__m128 const m[4], __m128 v) 71 __m128 v0 = _mm_shuffle_ps(v, v, _MM_SHUFFLE(0, 0, 0, 0)); 72 __m128 v1 = _mm_shuffle_ps(v, v, _MM_SHUFFLE(1, 1, 1, 1)); [all …]
|
D | intrinsic_geometric.inl | 33 GLM_FUNC_QUALIFIER __m128 sse_len_ps(__m128 x) 35 __m128 dot0 = sse_dot_ps(x, x); 36 __m128 sqt0 = _mm_sqrt_ps(dot0); 41 GLM_FUNC_QUALIFIER __m128 sse_dst_ps(__m128 p0, __m128 p1) 43 __m128 sub0 = _mm_sub_ps(p0, p1); 44 __m128 len0 = sse_len_ps(sub0); 49 GLM_FUNC_QUALIFIER __m128 sse_dot_ps(__m128 v1, __m128 v2) 54 __m128 mul0 = _mm_mul_ps(v1, v2); 55 __m128 swp0 = _mm_shuffle_ps(mul0, mul0, _MM_SHUFFLE(2, 3, 0, 1)); 56 __m128 add0 = _mm_add_ps(mul0, swp0); [all …]
|
D | intrinsic_geometric.hpp | 44 __m128 sse_len_ps(__m128 x); 47 __m128 sse_dst_ps(__m128 p0, __m128 p1); 50 __m128 sse_dot_ps(__m128 v1, __m128 v2); 53 __m128 sse_dot_ss(__m128 v1, __m128 v2); 56 __m128 sse_xpd_ps(__m128 v1, __m128 v2); 59 __m128 sse_nrm_ps(__m128 v); 62 __m128 sse_ffd_ps(__m128 N, __m128 I, __m128 Nref); 65 __m128 sse_rfe_ps(__m128 I, __m128 N); 68 __m128 sse_rfa_ps(__m128 I, __m128 N, __m128 eta);
|
D | intrinsic_matrix.hpp | 43 void sse_add_ps(__m128 in1[4], __m128 in2[4], __m128 out[4]); 45 void sse_sub_ps(__m128 in1[4], __m128 in2[4], __m128 out[4]); 47 __m128 sse_mul_ps(__m128 m[4], __m128 v); 49 __m128 sse_mul_ps(__m128 v, __m128 m[4]); 51 void sse_mul_ps(__m128 const in1[4], __m128 const in2[4], __m128 out[4]); 53 void sse_transpose_ps(__m128 const in[4], __m128 out[4]); 55 void sse_inverse_ps(__m128 const in[4], __m128 out[4]); 57 void sse_rotate_ps(__m128 const in[4], float Angle, float const v[3], __m128 out[4]); 59 __m128 sse_det_ps(__m128 const m[4]); 61 __m128 sse_slow_det_ps(__m128 const m[4]);
|
/external/clang/test/CodeGen/ |
D | sse-builtins.c | 7 __m128 test_rsqrt_ss(__m128 x) { in test_rsqrt_ss() 17 __m128 test_rcp_ss(__m128 x) { in test_rcp_ss() 27 __m128 test_sqrt_ss(__m128 x) { in test_sqrt_ss() 37 __m128 test_loadl_pi(__m128 x, void* y) { in test_loadl_pi() 45 __m128 test_loadh_pi(__m128 x, void* y) { in test_loadh_pi() 53 __m128 test_load_ss(void* y) { in test_load_ss() 59 __m128 test_load1_ps(void* y) { in test_load1_ps() 65 void test_store_ss(__m128 x, void* y) { in test_store_ss() 71 __m128d test_load1_pd(__m128 x, void* y) { in test_load1_pd() 77 __m128d test_loadr_pd(__m128 x, void* y) { in test_loadr_pd() [all …]
|
D | fma4-builtins.c | 8 __m128 test_mm_macc_ps(__m128 a, __m128 b, __m128 c) { in test_mm_macc_ps() 20 __m128 test_mm_macc_ss(__m128 a, __m128 b, __m128 c) { in test_mm_macc_ss() 32 __m128 test_mm_msub_ps(__m128 a, __m128 b, __m128 c) { in test_mm_msub_ps() 44 __m128 test_mm_msub_ss(__m128 a, __m128 b, __m128 c) { in test_mm_msub_ss() 56 __m128 test_mm_nmacc_ps(__m128 a, __m128 b, __m128 c) { in test_mm_nmacc_ps() 68 __m128 test_mm_nmacc_ss(__m128 a, __m128 b, __m128 c) { in test_mm_nmacc_ss() 80 __m128 test_mm_nmsub_ps(__m128 a, __m128 b, __m128 c) { in test_mm_nmsub_ps() 92 __m128 test_mm_nmsub_ss(__m128 a, __m128 b, __m128 c) { in test_mm_nmsub_ss() 104 __m128 test_mm_maddsub_ps(__m128 a, __m128 b, __m128 c) { in test_mm_maddsub_ps() 116 __m128 test_mm_msubadd_ps(__m128 a, __m128 b, __m128 c) { in test_mm_msubadd_ps()
|
D | fma-builtins.c | 8 __m128 test_mm_fmadd_ps(__m128 a, __m128 b, __m128 c) { in test_mm_fmadd_ps() 18 __m128 test_mm_fmadd_ss(__m128 a, __m128 b, __m128 c) { in test_mm_fmadd_ss() 28 __m128 test_mm_fmsub_ps(__m128 a, __m128 b, __m128 c) { in test_mm_fmsub_ps() 38 __m128 test_mm_fmsub_ss(__m128 a, __m128 b, __m128 c) { in test_mm_fmsub_ss() 48 __m128 test_mm_fnmadd_ps(__m128 a, __m128 b, __m128 c) { in test_mm_fnmadd_ps() 58 __m128 test_mm_fnmadd_ss(__m128 a, __m128 b, __m128 c) { in test_mm_fnmadd_ss() 68 __m128 test_mm_fnmsub_ps(__m128 a, __m128 b, __m128 c) { in test_mm_fnmsub_ps() 78 __m128 test_mm_fnmsub_ss(__m128 a, __m128 b, __m128 c) { in test_mm_fnmsub_ss() 88 __m128 test_mm_fmaddsub_ps(__m128 a, __m128 b, __m128 c) { in test_mm_fmaddsub_ps() 98 __m128 test_mm_fmsubadd_ps(__m128 a, __m128 b, __m128 c) { in test_mm_fmsubadd_ps()
|
D | avx512vldq-builtins.c | 80 __m128 test_mm_mask_andnot_ps (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) { in test_mm_mask_andnot_ps() 83 return (__m128) _mm_mask_andnot_ps ( __W, __U, __A, __B); in test_mm_mask_andnot_ps() 86 __m128 test_mm_maskz_andnot_ps (__mmask8 __U, __m128 __A, __m128 __B) { in test_mm_maskz_andnot_ps() 89 return (__m128) _mm_maskz_andnot_ps (__U, __A, __B); in test_mm_maskz_andnot_ps() 128 __m128 test_mm_mask_and_ps (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) { in test_mm_mask_and_ps() 131 return (__m128) _mm_mask_and_ps ( __W, __U, __A, __B); in test_mm_mask_and_ps() 134 __m128 test_mm_maskz_and_ps (__mmask8 __U, __m128 __A, __m128 __B) { in test_mm_maskz_and_ps() 137 return (__m128) _mm_maskz_and_ps (__U, __A, __B); in test_mm_maskz_and_ps() 176 __m128 test_mm_mask_xor_ps (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B) { in test_mm_mask_xor_ps() 179 return (__m128) _mm_mask_xor_ps ( __W, __U, __A, __B); in test_mm_mask_xor_ps() [all …]
|
D | avx512er-builtins.c | 80 __m128 test_mm_rsqrt28_round_ss(__m128 a, __m128 b) { in test_mm_rsqrt28_round_ss() 86 __m128 test_mm_mask_rsqrt28_round_ss(__m128 s, __mmask16 m, __m128 a, __m128 b) { in test_mm_mask_rsqrt28_round_ss() 92 __m128 test_mm_maskz_rsqrt28_round_ss(__mmask16 m, __m128 a, __m128 b) { in test_mm_maskz_rsqrt28_round_ss() 98 __m128 test_mm_rsqrt28_ss(__m128 a, __m128 b) { in test_mm_rsqrt28_ss() 104 __m128 test_mm_mask_rsqrt28_ss(__m128 s, __mmask16 m, __m128 a, __m128 b) { in test_mm_mask_rsqrt28_ss() 110 __m128 test_mm_maskz_rsqrt28_ss(__mmask16 m, __m128 a, __m128 b) { in test_mm_maskz_rsqrt28_ss() 206 __m128 test_mm_rcp28_round_ss(__m128 a, __m128 b) { in test_mm_rcp28_round_ss() 212 __m128 test_mm_mask_rcp28_round_ss(__m128 s, __mmask16 m, __m128 a, __m128 b) { in test_mm_mask_rcp28_round_ss() 218 __m128 test_mm_maskz_rcp28_round_ss(__mmask16 m, __m128 a, __m128 b) { in test_mm_maskz_rcp28_round_ss() 224 __m128 test_mm_rcp28_ss(__m128 a, __m128 b) { in test_mm_rcp28_ss() [all …]
|
D | avx-cmp-builtins.c | 19 __m128d test_cmp_ps(__m128 a, __m128 b) { in test_cmp_ps() 43 __m128d test_cmp_ss(__m128 a, __m128 b) { in test_cmp_ss() 49 __m128 test_cmpgt_ss(__m128 a, __m128 b) { in test_cmpgt_ss() 55 __m128 test_cmpge_ss(__m128 a, __m128 b) { in test_cmpge_ss() 61 __m128 test_cmpngt_ss(__m128 a, __m128 b) { in test_cmpngt_ss() 67 __m128 test_cmpnge_ss(__m128 a, __m128 b) { in test_cmpnge_ss()
|
D | sse.c | 10 __m128 test_mm_slli_si128(__m128 a) { in test_mm_slli_si128() 16 __m128 test_mm_slli_si128_0(__m128 a) { in test_mm_slli_si128_0() 22 __m128 test_mm_slli_si128_16(__m128 a) { in test_mm_slli_si128_16() 28 __m128 test_mm_srli_si128(__m128 a) { in test_mm_srli_si128() 34 __m128 test_mm_srli_si128_0(__m128 a) { in test_mm_srli_si128_0() 40 __m128 test_mm_srli_si128_16(__m128 a) { in test_mm_srli_si128_16()
|
D | sse41-builtins.c | 21 __m128 test_mm_blend_ps(__m128 V1, __m128 V2) { in test_mm_blend_ps() 39 __m128 test_mm_blendv_ps(__m128 V1, __m128 V2, __m128 V3) { in test_mm_blendv_ps() 51 __m128 test_mm_ceil_ps(__m128 x) { in test_mm_ceil_ps() 63 __m128 test_mm_ceil_ss(__m128 x, __m128 y) { in test_mm_ceil_ss() 153 __m128 test_mm_dp_ps(__m128 x, __m128 y) { in test_mm_dp_ps() 188 __m128 test_mm_floor_ps(__m128 x) { in test_mm_floor_ps() 200 __m128 test_mm_floor_ss(__m128 x, __m128 y) { in test_mm_floor_ss() 224 __m128 test_mm_insert_ps(__m128 x, __m128 y) { in test_mm_insert_ps() 314 __m128 test_mm_round_ps(__m128 x) { in test_mm_round_ps() 326 __m128 test_mm_round_ss(__m128 x, __m128 y) { in test_mm_round_ss()
|
D | sse3-builtins.c | 14 __m128 test_mm_addsub_ps(__m128 A, __m128 B) { in test_mm_addsub_ps() 26 __m128 test_mm_hadd_ps(__m128 A, __m128 B) { in test_mm_hadd_ps() 38 __m128 test_mm_hsub_ps(__m128 A, __m128 B) { in test_mm_hsub_ps() 62 __m128 test_mm_movehdup_ps(__m128 A) { in test_mm_movehdup_ps() 68 __m128 test_mm_moveldup_ps(__m128 A) { in test_mm_moveldup_ps()
|
D | target-builtin-noerror.c | 34 __m128 __attribute__((target("fma"))) fma_1(__m128 a, __m128 b, __m128 c) { in fma_1() 38 __m128 __attribute__((target("fma4"))) fma_2(__m128 a, __m128 b, __m128 c) { in fma_2() 42 __m128 __attribute__((target("fma,fma4"))) fma_3(__m128 a, __m128 b, __m128 c) { in fma_3()
|
D | avx512vl-builtins.c | 996 __mmask8 test_mm128_cmp_ps_mask(__m128 __A, __m128 __B) { in test_mm128_cmp_ps_mask() 1002 __mmask8 test_mm128_mask_cmp_ps_mask(__mmask8 m, __m128 __A, __m128 __B) { in test_mm128_mask_cmp_ps_mask() 1131 __m128 test_mm_mask_fmadd_ps(__m128 __A, __mmask8 __U, __m128 __B, __m128 __C) { in test_mm_mask_fmadd_ps() 1137 __m128 test_mm_mask_fmsub_ps(__m128 __A, __mmask8 __U, __m128 __B, __m128 __C) { in test_mm_mask_fmsub_ps() 1143 __m128 test_mm_mask3_fmadd_ps(__m128 __A, __m128 __B, __m128 __C, __mmask8 __U) { in test_mm_mask3_fmadd_ps() 1149 __m128 test_mm_mask3_fnmadd_ps(__m128 __A, __m128 __B, __m128 __C, __mmask8 __U) { in test_mm_mask3_fnmadd_ps() 1155 __m128 test_mm_maskz_fmadd_ps(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C) { in test_mm_maskz_fmadd_ps() 1161 __m128 test_mm_maskz_fmsub_ps(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C) { in test_mm_maskz_fmsub_ps() 1167 __m128 test_mm_maskz_fnmadd_ps(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C) { in test_mm_maskz_fnmadd_ps() 1173 __m128 test_mm_maskz_fnmsub_ps(__mmask8 __U, __m128 __A, __m128 __B, __m128 __C) { in test_mm_maskz_fnmsub_ps() [all …]
|
/external/webrtc/webrtc/modules/audio_processing/aec/ |
D | aec_rdft_sse2.c | 19 const __m128 mm_swap_sign = _mm_load_ps(k_swap_sign); in cft1st_128_SSE2() 23 __m128 a00v = _mm_loadu_ps(&a[j + 0]); in cft1st_128_SSE2() 24 __m128 a04v = _mm_loadu_ps(&a[j + 4]); in cft1st_128_SSE2() 25 __m128 a08v = _mm_loadu_ps(&a[j + 8]); in cft1st_128_SSE2() 26 __m128 a12v = _mm_loadu_ps(&a[j + 12]); in cft1st_128_SSE2() 27 __m128 a01v = _mm_shuffle_ps(a00v, a08v, _MM_SHUFFLE(1, 0, 1, 0)); in cft1st_128_SSE2() 28 __m128 a23v = _mm_shuffle_ps(a00v, a08v, _MM_SHUFFLE(3, 2, 3, 2)); in cft1st_128_SSE2() 29 __m128 a45v = _mm_shuffle_ps(a04v, a12v, _MM_SHUFFLE(1, 0, 1, 0)); in cft1st_128_SSE2() 30 __m128 a67v = _mm_shuffle_ps(a04v, a12v, _MM_SHUFFLE(3, 2, 3, 2)); in cft1st_128_SSE2() 32 const __m128 wk1rv = _mm_load_ps(&rdft_wk1r[k2]); in cft1st_128_SSE2() [all …]
|
D | aec_core_sse2.c | 51 const __m128 x_fft_buf_re = _mm_loadu_ps(&x_fft_buf[0][xPos + j]); in FilterFarSSE2() 52 const __m128 x_fft_buf_im = _mm_loadu_ps(&x_fft_buf[1][xPos + j]); in FilterFarSSE2() 53 const __m128 h_fft_buf_re = _mm_loadu_ps(&h_fft_buf[0][pos + j]); in FilterFarSSE2() 54 const __m128 h_fft_buf_im = _mm_loadu_ps(&h_fft_buf[1][pos + j]); in FilterFarSSE2() 55 const __m128 y_fft_re = _mm_loadu_ps(&y_fft[0][j]); in FilterFarSSE2() 56 const __m128 y_fft_im = _mm_loadu_ps(&y_fft[1][j]); in FilterFarSSE2() 57 const __m128 a = _mm_mul_ps(x_fft_buf_re, h_fft_buf_re); in FilterFarSSE2() 58 const __m128 b = _mm_mul_ps(x_fft_buf_im, h_fft_buf_im); in FilterFarSSE2() 59 const __m128 c = _mm_mul_ps(x_fft_buf_re, h_fft_buf_im); in FilterFarSSE2() 60 const __m128 d = _mm_mul_ps(x_fft_buf_im, h_fft_buf_re); in FilterFarSSE2() [all …]
|
/external/opencv3/modules/core/include/opencv2/core/ |
D | sse_utils.hpp | 489 inline void _mm_deinterleave_ps(__m128 & v_r0, __m128 & v_r1, __m128 & v_g0, __m128 & v_g1) in _mm_deinterleave_ps() 491 __m128 layer1_chunk0 = _mm_unpacklo_ps(v_r0, v_g0); in _mm_deinterleave_ps() 492 __m128 layer1_chunk1 = _mm_unpackhi_ps(v_r0, v_g0); in _mm_deinterleave_ps() 493 __m128 layer1_chunk2 = _mm_unpacklo_ps(v_r1, v_g1); in _mm_deinterleave_ps() 494 __m128 layer1_chunk3 = _mm_unpackhi_ps(v_r1, v_g1); in _mm_deinterleave_ps() 496 __m128 layer2_chunk0 = _mm_unpacklo_ps(layer1_chunk0, layer1_chunk2); in _mm_deinterleave_ps() 497 __m128 layer2_chunk1 = _mm_unpackhi_ps(layer1_chunk0, layer1_chunk2); in _mm_deinterleave_ps() 498 __m128 layer2_chunk2 = _mm_unpacklo_ps(layer1_chunk1, layer1_chunk3); in _mm_deinterleave_ps() 499 __m128 layer2_chunk3 = _mm_unpackhi_ps(layer1_chunk1, layer1_chunk3); in _mm_deinterleave_ps() 507 inline void _mm_deinterleave_ps(__m128 & v_r0, __m128 & v_r1, __m128 & v_g0, in _mm_deinterleave_ps() [all …]
|