Home
last modified time | relevance | path

Searched refs:__m128 (Results 1 – 25 of 70) sorted by relevance

123

/external/clang/lib/Headers/
Dxmmintrin.h35 typedef float __m128 __attribute__((__vector_size__(16))); typedef
43 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
44 _mm_add_ss(__m128 __a, __m128 __b) in _mm_add_ss()
50 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
51 _mm_add_ps(__m128 __a, __m128 __b) in _mm_add_ps()
56 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
57 _mm_sub_ss(__m128 __a, __m128 __b) in _mm_sub_ss()
63 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
64 _mm_sub_ps(__m128 __a, __m128 __b) in _mm_sub_ps()
69 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
[all …]
Dfma4intrin.h37 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
38 _mm_macc_ps(__m128 __A, __m128 __B, __m128 __C) in _mm_macc_ps()
40 return (__m128)__builtin_ia32_vfmaddps(__A, __B, __C); in _mm_macc_ps()
49 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
50 _mm_macc_ss(__m128 __A, __m128 __B, __m128 __C) in _mm_macc_ss()
52 return (__m128)__builtin_ia32_vfmaddss(__A, __B, __C); in _mm_macc_ss()
61 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
62 _mm_msub_ps(__m128 __A, __m128 __B, __m128 __C) in _mm_msub_ps()
64 return (__m128)__builtin_ia32_vfmsubps(__A, __B, __C); in _mm_msub_ps()
73 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
[all …]
Dfmaintrin.h35 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
36 _mm_fmadd_ps(__m128 __A, __m128 __B, __m128 __C) in _mm_fmadd_ps()
38 return (__m128)__builtin_ia32_vfmaddps(__A, __B, __C); in _mm_fmadd_ps()
47 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
48 _mm_fmadd_ss(__m128 __A, __m128 __B, __m128 __C) in _mm_fmadd_ss()
50 return (__m128)__builtin_ia32_vfmaddss(__A, __B, __C); in _mm_fmadd_ss()
59 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
60 _mm_fmsub_ps(__m128 __A, __m128 __B, __m128 __C) in _mm_fmsub_ps()
62 return (__m128)__builtin_ia32_vfmsubps(__A, __B, __C); in _mm_fmsub_ps()
71 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
[all …]
Dpmmintrin.h39 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
40 _mm_addsub_ps(__m128 __a, __m128 __b) in _mm_addsub_ps()
45 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
46 _mm_hadd_ps(__m128 __a, __m128 __b) in _mm_hadd_ps()
51 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
52 _mm_hsub_ps(__m128 __a, __m128 __b) in _mm_hsub_ps()
57 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
58 _mm_movehdup_ps(__m128 __a) in _mm_movehdup_ps()
63 static __inline__ __m128 __attribute__((__always_inline__, __nodebug__))
64 _mm_moveldup_ps(__m128 __a) in _mm_moveldup_ps()
Davx512erintrin.h130 (__m128)__builtin_ia32_rsqrt28ss_mask((__v4sf)(__m128)(A), \
131 (__v4sf)(__m128)(B), \
136 (__m128)__builtin_ia32_rsqrt28ss_mask((__v4sf)(__m128)(A), \
137 (__v4sf)(__m128)(B), \
138 (__v4sf)(__m128)(S), \
142 (__m128)__builtin_ia32_rsqrt28ss_mask((__v4sf)(__m128)(A), \
143 (__v4sf)(__m128)(B), \
233 (__m128)__builtin_ia32_rcp28ss_mask((__v4sf)(__m128)(A), \
234 (__v4sf)(__m128)(B), \
239 (__m128)__builtin_ia32_rcp28ss_mask((__v4sf)(__m128)(A), \
[all …]
/external/clang/test/CodeGen/
Dsse-builtins.c7 __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 …]
Dfma4-builtins.c8 __m128 test_mm_macc_ps(__m128 a, __m128 b, __m128 c) { in test_mm_macc_ps()
18 __m128 test_mm_macc_ss(__m128 a, __m128 b, __m128 c) { in test_mm_macc_ss()
28 __m128 test_mm_msub_ps(__m128 a, __m128 b, __m128 c) { in test_mm_msub_ps()
38 __m128 test_mm_msub_ss(__m128 a, __m128 b, __m128 c) { in test_mm_msub_ss()
48 __m128 test_mm_nmacc_ps(__m128 a, __m128 b, __m128 c) { in test_mm_nmacc_ps()
58 __m128 test_mm_nmacc_ss(__m128 a, __m128 b, __m128 c) { in test_mm_nmacc_ss()
68 __m128 test_mm_nmsub_ps(__m128 a, __m128 b, __m128 c) { in test_mm_nmsub_ps()
78 __m128 test_mm_nmsub_ss(__m128 a, __m128 b, __m128 c) { in test_mm_nmsub_ss()
88 __m128 test_mm_maddsub_ps(__m128 a, __m128 b, __m128 c) { in test_mm_maddsub_ps()
98 __m128 test_mm_msubadd_ps(__m128 a, __m128 b, __m128 c) { in test_mm_msubadd_ps()
Dfma-builtins.c8 __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()
Davx512er-builtins.c77 __m128 test_mm_rsqrt28_round_ss(__m128 a, __m128 b) { in test_mm_rsqrt28_round_ss()
83 __m128 test_mm_mask_rsqrt28_round_ss(__m128 s, __mmask16 m, __m128 a, __m128 b) { in test_mm_mask_rsqrt28_round_ss()
89 __m128 test_mm_maskz_rsqrt28_round_ss(__mmask16 m, __m128 a, __m128 b) { in test_mm_maskz_rsqrt28_round_ss()
95 __m128 test_mm_rsqrt28_ss(__m128 a, __m128 b) { in test_mm_rsqrt28_ss()
101 __m128 test_mm_mask_rsqrt28_ss(__m128 s, __mmask16 m, __m128 a, __m128 b) { in test_mm_mask_rsqrt28_ss()
107 __m128 test_mm_maskz_rsqrt28_ss(__mmask16 m, __m128 a, __m128 b) { in test_mm_maskz_rsqrt28_ss()
203 __m128 test_mm_rcp28_round_ss(__m128 a, __m128 b) { in test_mm_rcp28_round_ss()
209 __m128 test_mm_mask_rcp28_round_ss(__m128 s, __mmask16 m, __m128 a, __m128 b) { in test_mm_mask_rcp28_round_ss()
215 __m128 test_mm_maskz_rcp28_round_ss(__mmask16 m, __m128 a, __m128 b) { in test_mm_maskz_rcp28_round_ss()
221 __m128 test_mm_rcp28_ss(__m128 a, __m128 b) { in test_mm_rcp28_ss()
[all …]
Dsse.c6 __m128 test_mm_slli_si128(__m128 a) { in test_mm_slli_si128()
12 __m128 test_mm_slli_si128_0(__m128 a) { in test_mm_slli_si128_0()
18 __m128 test_mm_slli_si128_16(__m128 a) { in test_mm_slli_si128_16()
24 __m128 test_mm_srli_si128(__m128 a) { in test_mm_srli_si128()
30 __m128 test_mm_srli_si128_0(__m128 a) { in test_mm_srli_si128_0()
36 __m128 test_mm_srli_si128_16(__m128 a) { in test_mm_srli_si128_16()
Davx-cmp-builtins.c18 __m128d test_cmp_ps(__m128 a, __m128 b) { in test_cmp_ps()
42 __m128d test_cmp_ss(__m128 a, __m128 b) { in test_cmp_ss()
48 __m128 test_cmpgt_ss(__m128 a, __m128 b) { in test_cmpgt_ss()
54 __m128 test_cmpge_ss(__m128 a, __m128 b) { in test_cmpge_ss()
60 __m128 test_cmpngt_ss(__m128 a, __m128 b) { in test_cmpngt_ss()
66 __m128 test_cmpnge_ss(__m128 a, __m128 b) { in test_cmpnge_ss()
Davx-shuffle-builtins.c30 __m128 test_mm_permute_ps(__m128 a) { in test_mm_permute_ps()
37 __m128 test_mm_permute_ps2(__m128 a) { in test_mm_permute_ps2()
67 __m128
103 __m256 test_mm256_insertf128_ps_0(__m256 a, __m128 b) { in test_mm256_insertf128_ps_0()
121 __m256 test_mm256_insertf128_ps_1(__m256 a, __m128 b) { in test_mm256_insertf128_ps_1()
141 __m128 test_mm256_extractf128_ps_0(__m256 a) { in test_mm256_extractf128_ps_0()
159 __m128 test_mm256_extractf128_ps_1(__m256 a) { in test_mm256_extractf128_ps_1()
/external/webrtc/src/modules/audio_processing/aec/
Daec_rdft_sse2.c22 const __m128 mm_swap_sign = _mm_load_ps(k_swap_sign); in cft1st_128_SSE2()
26 __m128 a00v = _mm_loadu_ps(&a[j + 0]); in cft1st_128_SSE2()
27 __m128 a04v = _mm_loadu_ps(&a[j + 4]); in cft1st_128_SSE2()
28 __m128 a08v = _mm_loadu_ps(&a[j + 8]); in cft1st_128_SSE2()
29 __m128 a12v = _mm_loadu_ps(&a[j + 12]); in cft1st_128_SSE2()
30 __m128 a01v = _mm_shuffle_ps(a00v, a08v, _MM_SHUFFLE(1, 0, 1 ,0)); in cft1st_128_SSE2()
31 __m128 a23v = _mm_shuffle_ps(a00v, a08v, _MM_SHUFFLE(3, 2, 3 ,2)); in cft1st_128_SSE2()
32 __m128 a45v = _mm_shuffle_ps(a04v, a12v, _MM_SHUFFLE(1, 0, 1 ,0)); in cft1st_128_SSE2()
33 __m128 a67v = _mm_shuffle_ps(a04v, a12v, _MM_SHUFFLE(3, 2, 3 ,2)); in cft1st_128_SSE2()
35 const __m128 wk1rv = _mm_load_ps(&rdft_wk1r[k2]); in cft1st_128_SSE2()
[all …]
Daec_core_sse2.c48 const __m128 xfBuf_re = _mm_loadu_ps(&aec->xfBuf[0][xPos + j]); in FilterFarSSE2()
49 const __m128 xfBuf_im = _mm_loadu_ps(&aec->xfBuf[1][xPos + j]); in FilterFarSSE2()
50 const __m128 wfBuf_re = _mm_loadu_ps(&aec->wfBuf[0][pos + j]); in FilterFarSSE2()
51 const __m128 wfBuf_im = _mm_loadu_ps(&aec->wfBuf[1][pos + j]); in FilterFarSSE2()
52 const __m128 yf_re = _mm_loadu_ps(&yf[0][j]); in FilterFarSSE2()
53 const __m128 yf_im = _mm_loadu_ps(&yf[1][j]); in FilterFarSSE2()
54 const __m128 a = _mm_mul_ps(xfBuf_re, wfBuf_re); in FilterFarSSE2()
55 const __m128 b = _mm_mul_ps(xfBuf_im, wfBuf_im); in FilterFarSSE2()
56 const __m128 c = _mm_mul_ps(xfBuf_re, wfBuf_im); in FilterFarSSE2()
57 const __m128 d = _mm_mul_ps(xfBuf_im, wfBuf_re); in FilterFarSSE2()
[all …]
/external/speex/libspeex/
Dvq_sse.h36 void vq_nbest(spx_word16_t *_in, const __m128 *codebook, int len, int entries, __m128 *E, int N, in… in vq_nbest()
40 VARDECL(__m128 *in); in vq_nbest()
41 __m128 half; in vq_nbest()
45 ALLOC(in, len, __m128); in vq_nbest()
50 __m128 d = _mm_mul_ps(E[i], half); in vq_nbest()
75 void vq_nbest_sign(spx_word16_t *_in, const __m128 *codebook, int len, int entries, __m128 *E, int … in vq_nbest_sign()
79 VARDECL(__m128 *in); in vq_nbest_sign()
80 __m128 half; in vq_nbest_sign()
84 ALLOC(in, len, __m128); in vq_nbest_sign()
89 __m128 d = _mm_setzero_ps(); in vq_nbest_sign()
Dcb_search_sse.h37 static inline void _spx_mm_getr_ps (__m128 U, float *__Z, float *__Y, float *__X, float *__W) in _spx_mm_getr_ps()
41 __m128 __v; in _spx_mm_getr_ps()
54 …ok(const signed char *shape_cb, const spx_sig_t *_r, float *resp, __m128 *resp2, __m128 *E, int sh… in compute_weighted_codebook()
57 __m128 resj, EE; in compute_weighted_codebook()
58 VARDECL(__m128 *r); in compute_weighted_codebook()
59 VARDECL(__m128 *shape); in compute_weighted_codebook()
60 ALLOC(r, subvect_size, __m128); in compute_weighted_codebook()
61 ALLOC(shape, subvect_size, __m128); in compute_weighted_codebook()
Dfilters_sse.h39 __m128 num[3], den[3], mem[3]; in filter_mem16_10()
56 __m128 xx; in filter_mem16_10()
57 __m128 yy; in filter_mem16_10()
92 __m128 num[2], den[2], mem[2]; in filter_mem16_8()
106 __m128 xx; in filter_mem16_8()
107 __m128 yy; in filter_mem16_8()
146 __m128 den[3], mem[3]; in iir_mem16_10()
161 __m128 xx; in iir_mem16_10()
162 __m128 yy; in iir_mem16_10()
195 __m128 den[2], mem[2]; in iir_mem16_8()
[all …]
Dresample_sse.h44 __m128 sum = _mm_setzero_ps(); in inner_product_single()
60 __m128 sum = _mm_setzero_ps(); in interpolate_product_single()
61 __m128 f = _mm_loadu_ps(frac); in interpolate_product_single()
83 __m128 t; in inner_product_double()
94 sum = _mm_add_sd(sum, (__m128d) _mm_movehl_ps((__m128) sum, (__m128) sum)); in inner_product_double()
106 __m128 f = _mm_loadu_ps(frac); in interpolate_product_double()
109 __m128 t; in interpolate_product_double()
123 sum = _mm_add_sd(sum, (__m128d) _mm_movehl_ps((__m128) sum, (__m128) sum)); in interpolate_product_double()
Dltp_sse.h42 __m128 sum = _mm_setzero_ps(); in inner_prod()
60 VARDECL(__m128 *x); in pitch_xcorr()
61 VARDECL(__m128 *y); in pitch_xcorr()
65 ALLOC(x, N, __m128); in pitch_xcorr()
66 ALLOC(y, N+L, __m128); in pitch_xcorr()
76 __m128 sum, *xx, *yy; in pitch_xcorr()
/external/clang/test/Sema/
Duninit-variables-vectors.c4 typedef float __m128 __attribute__((__vector_size__(16))); typedef
5 __m128 _mm_xor_ps(__m128 a, __m128 b);
6 __m128 _mm_loadu_ps(const float *p);
9 …__m128 x, y, z, w, X; // expected-note {{variable 'x' is declared here}} expected-note {{variable … in test1()
Dbuiltins-x86.c4 typedef float __m128 __attribute__((__vector_size__(16))); typedef
13 __m128 test__builtin_ia32_cmpps(__m128 __a, __m128 __b) { in test__builtin_ia32_cmpps()
21 __m128 test__builtin_ia32_cmpss(__m128 __a, __m128 __b) { in test__builtin_ia32_cmpss()
/external/libvpx/libvpx/vpx_ports/
Demmintrin_compat.h18 extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
21 return (__m128) __A; in _mm_castpd_ps()
31 _mm_castps_pd(__m128 __A) in _mm_castps_pd()
37 _mm_castps_si128(__m128 __A) in _mm_castps_si128()
42 extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
45 return (__m128) __A; in _mm_castsi128_ps()
/external/libopus/celt/x86/
Dpitch_sse.h42 __m128 xsum1, xsum2; in xcorr_kernel()
48 __m128 x0 = _mm_loadu_ps(x+j); in xcorr_kernel()
49 __m128 yj = _mm_loadu_ps(y+j); in xcorr_kernel()
50 __m128 y3 = _mm_loadu_ps(y+j+3); in xcorr_kernel()
79 __m128 xsum1, xsum2; in dual_inner_prod()
84 __m128 xi = _mm_loadu_ps(x+i); in dual_inner_prod()
85 __m128 y1i = _mm_loadu_ps(y01+i); in dual_inner_prod()
86 __m128 y2i = _mm_loadu_ps(y02+i); in dual_inner_prod()
109 __m128 x0v; in comb_filter_const()
110 __m128 g10v, g11v, g12v; in comb_filter_const()
[all …]
/external/eigen/Eigen/src/Geometry/arch/
DGeometry_SSE.h23 const __m128 mask = _mm_castsi128_ps(_mm_setr_epi32(0,0,0,0x80000000));
25 __m128 a = _a.coeffs().template packet<Aligned>(0);
26 __m128 b = _b.coeffs().template packet<Aligned>(0);
27 __m128 flip1 = _mm_xor_ps(_mm_mul_ps(vec4f_swizzle1(a,1,2,0,2),
29 __m128 flip2 = _mm_xor_ps(_mm_mul_ps(vec4f_swizzle1(a,3,3,3,1),
46 __m128 a = lhs.template packet<VectorLhs::Flags&AlignedBit ? Aligned : Unaligned>(0);
47 __m128 b = rhs.template packet<VectorRhs::Flags&AlignedBit ? Aligned : Unaligned>(0);
48 __m128 mul1=_mm_mul_ps(vec4f_swizzle1(a,1,2,0,3),vec4f_swizzle1(b,2,0,1,3));
49 __m128 mul2=_mm_mul_ps(vec4f_swizzle1(a,2,0,1,3),vec4f_swizzle1(b,1,2,0,3));
/external/libmpeg2/common/x86/
Dicv_sad_ssse3.c98 __m128 src_r0, src_r1; in icv_sad_8x4_ssse3()
99 __m128 ref_r0, ref_r1; in icv_sad_8x4_ssse3()
108 src_r0 = (__m128)_mm_loadl_epi64((__m128i *) (pu1_src)); in icv_sad_8x4_ssse3()
111 src_r1 = (__m128)_mm_loadl_epi64((__m128i *) (pu1_src)); in icv_sad_8x4_ssse3()
122 ref_r0 = (__m128)_mm_loadl_epi64((__m128i *) (pu1_ref)); in icv_sad_8x4_ssse3()
125 ref_r1 = (__m128)_mm_loadl_epi64((__m128i *) (pu1_ref)); in icv_sad_8x4_ssse3()

123