/external/clang/test/CodeGen/ |
D | avx512vlbw-builtins.c | 22 __mmask16 test_mm_cmpeq_epi8_mask(__m128i __a, __m128i __b) { in test_mm_cmpeq_epi8_mask() 25 return (__mmask16)_mm_cmpeq_epi8_mask(__a, __b); in test_mm_cmpeq_epi8_mask() 28 __mmask16 test_mm_mask_cmpeq_epi8_mask(__mmask16 __u, __m128i __a, __m128i __b) { in test_mm_mask_cmpeq_epi8_mask() 32 return (__mmask16)_mm_mask_cmpeq_epi8_mask(__u, __a, __b); in test_mm_mask_cmpeq_epi8_mask() 35 __mmask16 test_mm256_cmpeq_epi16_mask(__m256i __a, __m256i __b) { in test_mm256_cmpeq_epi16_mask() 38 return (__mmask16)_mm256_cmpeq_epi16_mask(__a, __b); in test_mm256_cmpeq_epi16_mask() 41 __mmask16 test_mm256_mask_cmpeq_epi16_mask(__mmask16 __u, __m256i __a, __m256i __b) { in test_mm256_mask_cmpeq_epi16_mask() 45 return (__mmask16)_mm256_mask_cmpeq_epi16_mask(__u, __a, __b); in test_mm256_mask_cmpeq_epi16_mask() 74 __mmask16 test_mm_cmpgt_epi8_mask(__m128i __a, __m128i __b) { in test_mm_cmpgt_epi8_mask() 77 return (__mmask16)_mm_cmpgt_epi8_mask(__a, __b); in test_mm_cmpgt_epi8_mask() [all …]
|
D | avx512f-builtins.c | 57 __m512 test_mm512_mask_sqrt_ps(__m512 __W, __mmask16 __U, __m512 __A) in test_mm512_mask_sqrt_ps() 64 __m512 test_mm512_maskz_sqrt_ps( __mmask16 __U, __m512 __A) in test_mm512_maskz_sqrt_ps() 71 __m512 test_mm512_mask_sqrt_round_ps(__m512 __W,__mmask16 __U,__m512 __A) in test_mm512_mask_sqrt_round_ps() 78 __m512 test_mm512_maskz_sqrt_round_ps(__mmask16 __U,__m512 __A) in test_mm512_maskz_sqrt_round_ps() 120 __m512 test_mm512_mask_rsqrt14_ps (__m512 __W, __mmask16 __U, __m512 __A) in test_mm512_mask_rsqrt14_ps() 127 __m512 test_mm512_maskz_rsqrt14_ps (__mmask16 __U, __m512 __A) in test_mm512_maskz_rsqrt14_ps() 186 void test_mm512_mask_store_ps(void *p, __m512 a, __mmask16 m) in test_mm512_mask_store_ps() 244 void test_mm512_mask_storeu_epi32(void *__P, __mmask16 __U, __m512i __A) { in test_mm512_mask_storeu_epi32() 263 __m512i test_mm512_mask_loadu_epi32 (__m512i __W, __mmask16 __U, void *__P) in test_mm512_mask_loadu_epi32() 284 __m512 test_mm512_mask_loadu_ps (__m512 __W, __mmask16 __U, void *__P) in test_mm512_mask_loadu_ps() [all …]
|
D | avx512er-builtins.c | 50 __m512 test_mm512_mask_rsqrt28_round_ps(__m512 s, __mmask16 m, __m512 a) { in test_mm512_mask_rsqrt28_round_ps() 56 __m512 test_mm512_maskz_rsqrt28_round_ps(__mmask16 m, __m512 a) { in test_mm512_maskz_rsqrt28_round_ps() 68 __m512 test_mm512_mask_rsqrt28_ps(__m512 s, __mmask16 m, __m512 a) { in test_mm512_mask_rsqrt28_ps() 74 __m512 test_mm512_maskz_rsqrt28_ps(__mmask16 m, __m512 a) { in test_mm512_maskz_rsqrt28_ps() 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() 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() 176 __m512 test_mm512_mask_rcp28_round_ps(__m512 s, __mmask16 m, __m512 a) { in test_mm512_mask_rcp28_round_ps() 182 __m512 test_mm512_maskz_rcp28_round_ps(__mmask16 m, __m512 a) { in test_mm512_maskz_rcp28_round_ps() [all …]
|
D | avx512dq-builtins.c | 50 __m512 test_mm512_mask_xor_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) { in test_mm512_mask_xor_ps() 56 __m512 test_mm512_maskz_xor_ps (__mmask16 __U, __m512 __A, __m512 __B) { in test_mm512_maskz_xor_ps() 86 __m512 test_mm512_mask_or_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) { in test_mm512_mask_or_ps() 92 __m512 test_mm512_maskz_or_ps (__mmask16 __U, __m512 __A, __m512 __B) { in test_mm512_maskz_or_ps() 122 __m512 test_mm512_mask_and_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) { in test_mm512_mask_and_ps() 128 __m512 test_mm512_maskz_and_ps (__mmask16 __U, __m512 __A, __m512 __B) { in test_mm512_maskz_and_ps() 158 __m512 test_mm512_mask_andnot_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) { in test_mm512_mask_andnot_ps() 164 __m512 test_mm512_maskz_andnot_ps (__mmask16 __U, __m512 __A, __m512 __B) { in test_mm512_maskz_andnot_ps() 716 __m512 test_mm512_mask_range_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) { in test_mm512_mask_range_ps() 722 __m512 test_mm512_maskz_range_ps(__mmask16 __U, __m512 __A, __m512 __B) { in test_mm512_maskz_range_ps() [all …]
|
D | avx512vbmivl-builtin.c | 14 __m128i test_mm_maskz_permutexvar_epi8(__mmask16 __M, __m128i __A, __m128i __B) { in test_mm_maskz_permutexvar_epi8() 20 __m128i test_mm_mask_permutexvar_epi8(__m128i __W, __mmask16 __M, __m128i __A, __m128i __B) { in test_mm_mask_permutexvar_epi8() 44 __m128i test_mm_mask2_permutex2var_epi8(__m128i __A, __m128i __I, __mmask16 __U, __m128i __B) { in test_mm_mask2_permutex2var_epi8() 62 __m128i test_mm_mask_permutex2var_epi8(__m128i __A, __mmask16 __U, __m128i __I, __m128i __B) { in test_mm_mask_permutex2var_epi8() 68 __m128i test_mm_maskz_permutex2var_epi8(__mmask16 __U, __m128i __A, __m128i __I, __m128i __B) { in test_mm_maskz_permutex2var_epi8() 92 __m128i test_mm_mask_multishift_epi64_epi8(__m128i __W, __mmask16 __M, __m128i __X, __m128i __Y) { in test_mm_mask_multishift_epi64_epi8() 98 __m128i test_mm_maskz_multishift_epi64_epi8(__mmask16 __M, __m128i __X, __m128i __Y) { in test_mm_maskz_multishift_epi64_epi8()
|
D | avx512cdintrin.c | 28 __m512i test_mm512_mask_conflict_epi32(__m512i __W, __mmask16 __U, __m512i __A) { in test_mm512_mask_conflict_epi32() 33 __m512i test_mm512_maskz_conflict_epi32(__mmask16 __U, __m512i __A) { in test_mm512_maskz_conflict_epi32() 43 __m512i test_mm512_mask_lzcnt_epi32(__m512i __W, __mmask16 __U, __m512i __A) { in test_mm512_mask_lzcnt_epi32() 49 __m512i test_mm512_maskz_lzcnt_epi32(__mmask16 __U, __m512i __A) { in test_mm512_maskz_lzcnt_epi32() 79 __m512i test_mm512_broadcastmw_epi32(__mmask16 __A) { in test_mm512_broadcastmw_epi32()
|
D | avx512pf-builtins.c | 20 void test_mm512_mask_prefetch_i32gather_ps(__m512i index, __mmask16 mask, void const *addr, int hin… in test_mm512_mask_prefetch_i32gather_ps() 74 void test_mm512_mask_prefetch_i32scatter_ps(void *addr, __mmask16 mask, __m512i index) { in test_mm512_mask_prefetch_i32scatter_ps() 86 void test_mm512_mask_prefetch_i64scatter_pd(void *addr, __mmask16 mask, __m512i index) { in test_mm512_mask_prefetch_i64scatter_pd() 98 void test_mm512_mask_prefetch_i64scatter_ps(void *addr, __mmask16 mask, __m512i index) { in test_mm512_mask_prefetch_i64scatter_ps()
|
/external/clang/lib/Headers/ |
D | avx512vlbwintrin.h | 41 static __inline__ __mmask16 __DEFAULT_FN_ATTRS 43 return (__mmask16)__builtin_ia32_pcmpeqb128_mask((__v16qi)__a, (__v16qi)__b, in _mm_cmpeq_epi8_mask() 44 (__mmask16)-1); in _mm_cmpeq_epi8_mask() 47 static __inline__ __mmask16 __DEFAULT_FN_ATTRS 48 _mm_mask_cmpeq_epi8_mask(__mmask16 __u, __m128i __a, __m128i __b) { in _mm_mask_cmpeq_epi8_mask() 49 return (__mmask16)__builtin_ia32_pcmpeqb128_mask((__v16qi)__a, (__v16qi)__b, in _mm_mask_cmpeq_epi8_mask() 53 static __inline__ __mmask16 __DEFAULT_FN_ATTRS 55 return (__mmask16)__builtin_ia32_ucmpb128_mask((__v16qi)__a, (__v16qi)__b, 0, in _mm_cmpeq_epu8_mask() 56 (__mmask16)-1); in _mm_cmpeq_epu8_mask() 59 static __inline__ __mmask16 __DEFAULT_FN_ATTRS [all …]
|
D | avx512fintrin.h | 48 typedef unsigned short __mmask16; typedef 208 _mm512_mask_broadcastd_epi32 (__m512i __O, __mmask16 __M, __m128i __A) in _mm512_mask_broadcastd_epi32() 216 _mm512_maskz_broadcastd_epi32 (__mmask16 __M, __m128i __A) in _mm512_maskz_broadcastd_epi32() 249 _mm512_maskz_set1_epi32(__mmask16 __M, int __A) in _mm512_maskz_set1_epi32() 514 _mm512_mask_and_epi32(__m512i __src, __mmask16 __k, __m512i __a, __m512i __b) in _mm512_mask_and_epi32() 516 return (__m512i)__builtin_ia32_selectd_512((__mmask16)__k, in _mm512_mask_and_epi32() 522 _mm512_maskz_and_epi32(__mmask16 __k, __m512i __a, __m512i __b) in _mm512_maskz_and_epi32() 562 _mm512_mask_andnot_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B) in _mm512_mask_andnot_epi32() 564 return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U, in _mm512_mask_andnot_epi32() 570 _mm512_maskz_andnot_epi32(__mmask16 __U, __m512i __A, __m512i __B) in _mm512_maskz_andnot_epi32() [all …]
|
D | avx512vbmivlintrin.h | 36 _mm_mask2_permutex2var_epi8 (__m128i __A, __m128i __I, __mmask16 __U, in _mm_mask2_permutex2var_epi8() 43 (__mmask16) in _mm_mask2_permutex2var_epi8() 66 (__mmask16) - in _mm_permutex2var_epi8() 71 _mm_mask_permutex2var_epi8 (__m128i __A, __mmask16 __U, __m128i __I, in _mm_mask_permutex2var_epi8() 78 (__mmask16) in _mm_mask_permutex2var_epi8() 83 _mm_maskz_permutex2var_epi8 (__mmask16 __U, __m128i __A, __m128i __I, in _mm_maskz_permutex2var_epi8() 90 (__mmask16) in _mm_maskz_permutex2var_epi8() 135 (__mmask16) -1); in _mm_permutexvar_epi8() 139 _mm_maskz_permutexvar_epi8 (__mmask16 __M, __m128i __A, __m128i __B) in _mm_maskz_permutexvar_epi8() 144 (__mmask16) __M); in _mm_maskz_permutexvar_epi8() [all …]
|
D | avx512cdintrin.h | 63 (__mmask16) -1); in _mm512_conflict_epi32() 67 _mm512_mask_conflict_epi32 (__m512i __W, __mmask16 __U, __m512i __A) in _mm512_mask_conflict_epi32() 71 (__mmask16) __U); in _mm512_mask_conflict_epi32() 75 _mm512_maskz_conflict_epi32 (__mmask16 __U, __m512i __A) in _mm512_maskz_conflict_epi32() 79 (__mmask16) __U); in _mm512_maskz_conflict_epi32() 87 (__mmask16) -1); in _mm512_lzcnt_epi32() 91 _mm512_mask_lzcnt_epi32 (__m512i __W, __mmask16 __U, __m512i __A) in _mm512_mask_lzcnt_epi32() 95 (__mmask16) __U); in _mm512_mask_lzcnt_epi32() 99 _mm512_maskz_lzcnt_epi32 (__mmask16 __U, __m512i __A) in _mm512_maskz_lzcnt_epi32() 103 (__mmask16) __U); in _mm512_maskz_lzcnt_epi32() [all …]
|
D | avx512dqintrin.h | 84 _mm512_mask_xor_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) { in _mm512_mask_xor_ps() 88 (__mmask16) __U); in _mm512_mask_xor_ps() 92 _mm512_maskz_xor_ps (__mmask16 __U, __m512 __A, __m512 __B) { in _mm512_maskz_xor_ps() 97 (__mmask16) __U); in _mm512_maskz_xor_ps() 128 _mm512_mask_or_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) { in _mm512_mask_or_ps() 132 (__mmask16) __U); in _mm512_mask_or_ps() 136 _mm512_maskz_or_ps (__mmask16 __U, __m512 __A, __m512 __B) { in _mm512_maskz_or_ps() 141 (__mmask16) __U); in _mm512_maskz_or_ps() 172 _mm512_mask_and_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) { in _mm512_mask_and_ps() 176 (__mmask16) __U); in _mm512_mask_and_ps() [all …]
|
D | avx512erintrin.h | 58 (__mmask16)-1, (int)(R)); }) 62 (__v16sf)(__m512)(S), (__mmask16)(M), \ 68 (__mmask16)(M), (int)(R)); }) 107 (__mmask16)-1, (int)(R)); }) 111 (__v16sf)(__m512)(S), (__mmask16)(M), \ 117 (__mmask16)(M), (int)(R)); }) 210 (__mmask16)-1, (int)(R)); }) 214 (__v16sf)(__m512)(S), (__mmask16)(M), \ 220 (__mmask16)(M), (int)(R)); })
|
/external/XNNPACK/src/f32-prelu/gen/ |
D | avx512f-2x32.c | 61 …const __mmask16 vsign0x0123456789ABCDEF = _mm512_cmp_ps_mask(vi0x0123456789ABCDEF, vzero, _CMP_LT_… in xnn_f32_prelu_ukernel__avx512f_2x32() 63 …const __mmask16 vsign0xGHIJKLMNOPQRSTUV = _mm512_cmp_ps_mask(vi0xGHIJKLMNOPQRSTUV, vzero, _CMP_LT_… in xnn_f32_prelu_ukernel__avx512f_2x32() 65 …const __mmask16 vsign1x0123456789ABCDEF = _mm512_cmp_ps_mask(vi1x0123456789ABCDEF, vzero, _CMP_LT_… in xnn_f32_prelu_ukernel__avx512f_2x32() 67 …const __mmask16 vsign1xGHIJKLMNOPQRSTUV = _mm512_cmp_ps_mask(vi1xGHIJKLMNOPQRSTUV, vzero, _CMP_LT_… in xnn_f32_prelu_ukernel__avx512f_2x32() 86 const __mmask16 vsign0 = _mm512_cmp_ps_mask(vi0, vzero, _CMP_LT_OQ); in xnn_f32_prelu_ukernel__avx512f_2x32() 88 const __mmask16 vsign1 = _mm512_cmp_ps_mask(vi1, vzero, _CMP_LT_OQ); in xnn_f32_prelu_ukernel__avx512f_2x32() 100 …const __mmask16 vmask = _cvtu32_mask16((uint16_t) ((uint32_t) (UINT32_C(1) << (c >> 2 /* log2(size… in xnn_f32_prelu_ukernel__avx512f_2x32() 109 const __mmask16 vsign0 = _mm512_cmp_ps_mask(vi0, vzero, _CMP_LT_OQ); in xnn_f32_prelu_ukernel__avx512f_2x32() 111 const __mmask16 vsign1 = _mm512_cmp_ps_mask(vi1, vzero, _CMP_LT_OQ); in xnn_f32_prelu_ukernel__avx512f_2x32()
|
D | avx512f-2x16.c | 58 …const __mmask16 vsign0x0123456789ABCDEF = _mm512_cmp_ps_mask(vi0x0123456789ABCDEF, vzero, _CMP_LT_… in xnn_f32_prelu_ukernel__avx512f_2x16() 60 …const __mmask16 vsign1x0123456789ABCDEF = _mm512_cmp_ps_mask(vi1x0123456789ABCDEF, vzero, _CMP_LT_… in xnn_f32_prelu_ukernel__avx512f_2x16() 72 …const __mmask16 vmask = _cvtu32_mask16((uint16_t) ((uint32_t) (UINT32_C(1) << (c >> 2 /* log2(size… in xnn_f32_prelu_ukernel__avx512f_2x16() 81 const __mmask16 vsign0 = _mm512_cmp_ps_mask(vi0, vzero, _CMP_LT_OQ); in xnn_f32_prelu_ukernel__avx512f_2x16() 83 const __mmask16 vsign1 = _mm512_cmp_ps_mask(vi1, vzero, _CMP_LT_OQ); in xnn_f32_prelu_ukernel__avx512f_2x16()
|
/external/mesa3d/src/gallium/drivers/swr/rasterizer/common/ |
D | simdlib_128_avx512.inl | 68 #define SIMD_WRAPPER_1(op) SIMD_WRAPPER_1_(op, op, __mmask16(0xf)) 76 #define SIMD_WRAPPER_1I(op) SIMD_WRAPPER_1I_(op, op, __mmask16(0xf)) 83 #define SIMD_WRAPPER_2(op) SIMD_WRAPPER_2_(op, op, __mmask16(0xf)) 97 #define SIMD_WRAPPER_3(op) SIMD_WRAPPER_3_(op, op, __mmask16(0xf)) 111 #define SIMD_IWRAPPER_1_32(op) SIMD_IWRAPPER_1_(op, op, __mmask16(0xf)) 119 #define SIMD_IWRAPPER_1I_32(op) SIMD_IWRAPPER_1I_(op, op, __mmask16(0xf)) 126 #define SIMD_IWRAPPER_2_32(op) SIMD_IWRAPPER_2_(op, op, __mmask16(0xf)) 145 SIMD_WRAPPER_1_(rcp_ps, rcp14_ps, __mmask16(0xf)); // return 1.0f / a 146 SIMD_WRAPPER_1_(rsqrt_ps, rsqrt14_ps, __mmask16(0xf)); // return 1.0f / sqrt(a) 176 SIMD_IWRAPPER_2_(and_si, and_epi32, __mmask16(0xf)); // return a & b (int) [all …]
|
D | simdlib_256_avx512.inl | 68 #define SIMD_WRAPPER_1(op) SIMD_WRAPPER_1_(op, op, __mmask16(0xff)) 76 #define SIMD_WRAPPER_1I(op) SIMD_WRAPPER_1I_(op, op, __mmask16(0xff)) 83 #define SIMD_WRAPPER_2(op) SIMD_WRAPPER_2_(op, op, __mmask16(0xff)) 97 #define SIMD_WRAPPER_3(op) SIMD_WRAPPER_3_(op, op, __mmask16(0xff)) 111 #define SIMD_IWRAPPER_1_32(op) SIMD_IWRAPPER_1_(op, op, __mmask16(0xff)) 119 #define SIMD_IWRAPPER_1I_32(op) SIMD_IWRAPPER_1I_(op, op, __mmask16(0xff)) 126 #define SIMD_IWRAPPER_2_32(op) SIMD_IWRAPPER_2_(op, op, __mmask16(0xff)) 145 SIMD_WRAPPER_1_(rcp_ps, rcp14_ps, __mmask16(0xff)); // return 1.0f / a 146 SIMD_WRAPPER_1_(rsqrt_ps, rsqrt14_ps, __mmask16(0xff)); // return 1.0f / sqrt(a) 176 SIMD_IWRAPPER_2_(and_si, and_epi32, __mmask16(0xff)); // return a & b (int) [all …]
|
/external/XNNPACK/src/f32-vlrelu/gen/ |
D | vlrelu-avx512f-x32.c | 36 …const __mmask16 vsign0123456789ABCDEF = _mm512_cmp_ps_mask(vacc0123456789ABCDEF, vzero, _CMP_LT_OQ… in xnn_f32_vlrelu_ukernel__avx512f_x32() 37 …const __mmask16 vsignGHIJKLMNOPQRSTUV = _mm512_cmp_ps_mask(vaccGHIJKLMNOPQRSTUV, vzero, _CMP_LT_OQ… in xnn_f32_vlrelu_ukernel__avx512f_x32() 49 const __mmask16 vsign = _mm512_cmp_ps_mask(vacc, vzero, _CMP_LT_OQ); in xnn_f32_vlrelu_ukernel__avx512f_x32() 59 … const __mmask16 vmask = _cvtu32_mask16((uint16_t) ((uint32_t) (UINT32_C(1) << n) - UINT32_C(1))); in xnn_f32_vlrelu_ukernel__avx512f_x32() 62 const __mmask16 vsign = _mm512_mask_cmp_ps_mask(vmask, vacc, vzero, _CMP_LT_OQ); in xnn_f32_vlrelu_ukernel__avx512f_x32()
|
D | vlrelu-avx512f-x16.c | 35 …const __mmask16 vsign0123456789ABCDEF = _mm512_cmp_ps_mask(vacc0123456789ABCDEF, vzero, _CMP_LT_OQ… in xnn_f32_vlrelu_ukernel__avx512f_x16() 47 … const __mmask16 vmask = _cvtu32_mask16((uint16_t) ((uint32_t) (UINT32_C(1) << n) - UINT32_C(1))); in xnn_f32_vlrelu_ukernel__avx512f_x16() 50 const __mmask16 vsign = _mm512_mask_cmp_ps_mask(vmask, vacc, vzero, _CMP_LT_OQ); in xnn_f32_vlrelu_ukernel__avx512f_x16()
|
/external/XNNPACK/src/f32-velu/gen/ |
D | velu-avx512f-rr1-lut16-p3-perm-x112.c | 144 const __mmask16 vsign0 = _mm512_cmp_ps_mask(vx0, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x112() 146 const __mmask16 vsign1 = _mm512_cmp_ps_mask(vx1, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x112() 148 const __mmask16 vsign2 = _mm512_cmp_ps_mask(vx2, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x112() 150 const __mmask16 vsign3 = _mm512_cmp_ps_mask(vx3, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x112() 152 const __mmask16 vsign4 = _mm512_cmp_ps_mask(vx4, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x112() 154 const __mmask16 vsign5 = _mm512_cmp_ps_mask(vx5, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x112() 156 const __mmask16 vsign6 = _mm512_cmp_ps_mask(vx6, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x112() 180 const __mmask16 vsign = _mm512_cmp_ps_mask(vx, _mm512_setzero_ps(), _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x112() 208 … const __mmask16 vmask = _cvtu32_mask16((uint16_t) ((uint32_t) (UINT32_C(1) << n) - UINT32_C(1))); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x112() 213 const __mmask16 vsign = _mm512_cmp_ps_mask(vx, _mm512_setzero_ps(), _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x112()
|
D | velu-avx512f-rr1-lut16-p3-perm-x128.c | 157 const __mmask16 vsign0 = _mm512_cmp_ps_mask(vx0, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x128() 159 const __mmask16 vsign1 = _mm512_cmp_ps_mask(vx1, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x128() 161 const __mmask16 vsign2 = _mm512_cmp_ps_mask(vx2, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x128() 163 const __mmask16 vsign3 = _mm512_cmp_ps_mask(vx3, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x128() 165 const __mmask16 vsign4 = _mm512_cmp_ps_mask(vx4, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x128() 167 const __mmask16 vsign5 = _mm512_cmp_ps_mask(vx5, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x128() 169 const __mmask16 vsign6 = _mm512_cmp_ps_mask(vx6, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x128() 171 const __mmask16 vsign7 = _mm512_cmp_ps_mask(vx7, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x128() 197 const __mmask16 vsign = _mm512_cmp_ps_mask(vx, _mm512_setzero_ps(), _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x128() 225 … const __mmask16 vmask = _cvtu32_mask16((uint16_t) ((uint32_t) (UINT32_C(1) << n) - UINT32_C(1))); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x128() [all …]
|
D | velu-avx512f-rr1-p6-x128.c | 169 const __mmask16 vsign0 = _mm512_cmp_ps_mask(vx0, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x128() 171 const __mmask16 vsign1 = _mm512_cmp_ps_mask(vx1, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x128() 173 const __mmask16 vsign2 = _mm512_cmp_ps_mask(vx2, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x128() 175 const __mmask16 vsign3 = _mm512_cmp_ps_mask(vx3, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x128() 177 const __mmask16 vsign4 = _mm512_cmp_ps_mask(vx4, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x128() 179 const __mmask16 vsign5 = _mm512_cmp_ps_mask(vx5, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x128() 181 const __mmask16 vsign6 = _mm512_cmp_ps_mask(vx6, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x128() 183 const __mmask16 vsign7 = _mm512_cmp_ps_mask(vx7, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x128() 209 const __mmask16 vsign = _mm512_cmp_ps_mask(vx, _mm512_setzero_ps(), _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x128() 238 … const __mmask16 vmask = _cvtu32_mask16((uint16_t) ((uint32_t) (UINT32_C(1) << n) - UINT32_C(1))); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x128() [all …]
|
D | velu-avx512f-rr1-lut16-p3-perm-x80.c | 118 const __mmask16 vsign0 = _mm512_cmp_ps_mask(vx0, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x80() 120 const __mmask16 vsign1 = _mm512_cmp_ps_mask(vx1, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x80() 122 const __mmask16 vsign2 = _mm512_cmp_ps_mask(vx2, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x80() 124 const __mmask16 vsign3 = _mm512_cmp_ps_mask(vx3, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x80() 126 const __mmask16 vsign4 = _mm512_cmp_ps_mask(vx4, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x80() 146 const __mmask16 vsign = _mm512_cmp_ps_mask(vx, _mm512_setzero_ps(), _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x80() 174 … const __mmask16 vmask = _cvtu32_mask16((uint16_t) ((uint32_t) (UINT32_C(1) << n) - UINT32_C(1))); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x80() 179 const __mmask16 vsign = _mm512_cmp_ps_mask(vx, _mm512_setzero_ps(), _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x80()
|
D | velu-avx512f-rr1-lut16-p3-perm-x96.c | 131 const __mmask16 vsign0 = _mm512_cmp_ps_mask(vx0, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x96() 133 const __mmask16 vsign1 = _mm512_cmp_ps_mask(vx1, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x96() 135 const __mmask16 vsign2 = _mm512_cmp_ps_mask(vx2, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x96() 137 const __mmask16 vsign3 = _mm512_cmp_ps_mask(vx3, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x96() 139 const __mmask16 vsign4 = _mm512_cmp_ps_mask(vx4, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x96() 141 const __mmask16 vsign5 = _mm512_cmp_ps_mask(vx5, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x96() 163 const __mmask16 vsign = _mm512_cmp_ps_mask(vx, _mm512_setzero_ps(), _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x96() 191 … const __mmask16 vmask = _cvtu32_mask16((uint16_t) ((uint32_t) (UINT32_C(1) << n) - UINT32_C(1))); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x96() 196 const __mmask16 vsign = _mm512_cmp_ps_mask(vx, _mm512_setzero_ps(), _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x96()
|
D | velu-avx512f-rr1-p6-x112.c | 155 const __mmask16 vsign0 = _mm512_cmp_ps_mask(vx0, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x112() 157 const __mmask16 vsign1 = _mm512_cmp_ps_mask(vx1, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x112() 159 const __mmask16 vsign2 = _mm512_cmp_ps_mask(vx2, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x112() 161 const __mmask16 vsign3 = _mm512_cmp_ps_mask(vx3, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x112() 163 const __mmask16 vsign4 = _mm512_cmp_ps_mask(vx4, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x112() 165 const __mmask16 vsign5 = _mm512_cmp_ps_mask(vx5, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x112() 167 const __mmask16 vsign6 = _mm512_cmp_ps_mask(vx6, vzero, _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x112() 191 const __mmask16 vsign = _mm512_cmp_ps_mask(vx, _mm512_setzero_ps(), _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x112() 220 … const __mmask16 vmask = _cvtu32_mask16((uint16_t) ((uint32_t) (UINT32_C(1) << n) - UINT32_C(1))); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x112() 225 const __mmask16 vsign = _mm512_cmp_ps_mask(vx, _mm512_setzero_ps(), _CMP_NLT_US); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x112()
|