| /dports/devel/hyperscan/hyperscan-5.4.0/src/rose/ |
| H A D | validate_shufti.h | 47 int validateShuftiMask16x16(const m256 data, const m256 hi_mask, in validateShuftiMask16x16()
48 const m256 lo_mask, const m256 and_mask, in validateShuftiMask16x16()
101 int validateShuftiMask32x8(const m256 data, const m256 hi_mask, in validateShuftiMask32x8()
102 const m256 lo_mask, const m256 and_mask, in validateShuftiMask32x8()
131 const m256 hi_mask_1, const m256 hi_mask_2, in validateShuftiMask32x16()
132 const m256 lo_mask_1, const m256 lo_mask_2, in validateShuftiMask32x16()
296 const m256 hi_mask, const m256 lo_mask, in validateMultipathShuftiMask32x8()
318 const m256 hi_mask_1, const m256 hi_mask_2, in validateMultipathShuftiMask32x16()
319 const m256 lo_mask_1, const m256 lo_mask_2, in validateMultipathShuftiMask32x16()
345 int validateMultipathShuftiMask64(const m256 data_1, const m256 data_2, in validateMultipathShuftiMask64()
[all …]
|
| /dports/graphics/librsvg2-rust/librsvg-2.52.8/vendor/safe_arch/src/x86_x64/ |
| H A D | avx.rs | 23 pub fn add_m256(a: m256, b: m256) -> m256 { in add_m256() argument
45 pub fn addsub_m256(a: m256, b: m256) -> m256 { in addsub_m256() argument
67 pub fn bitand_m256(a: m256, b: m256) -> m256 { in bitand_m256() argument
89 pub fn bitandnot_m256(a: m256, b: m256) -> m256 { in bitandnot_m256() argument
139 pub fn blend_varying_m256(a: m256, b: m256, mask: m256) -> m256 { in blend_varying_m256() argument
646 pub fn div_m256(a: m256, b: m256) -> m256 { in div_m256() argument
1086 pub fn max_m256(a: m256, b: m256) -> m256 { in max_m256() argument
1108 pub fn min_m256(a: m256, b: m256) -> m256 { in min_m256() argument
1185 pub fn mul_m256(a: m256, b: m256) -> m256 { in mul_m256() argument
1207 pub fn bitor_m256(a: m256, b: m256) -> m256 { in bitor_m256() argument
[all …]
|
| H A D | m256_.rs | 16 pub struct m256(pub __m256); struct
25 impl m256 { impl
64 impl Clone for m256 { implementation
71 impl Copy for m256 {} implementation
73 impl Default for m256 { implementation
105 impl Debug for m256 { implementation
124 impl Display for m256 { implementation
143 impl Binary for m256 { implementation
162 impl LowerExp for m256 { implementation
181 impl UpperExp for m256 { implementation
[all …]
|
| H A D | fma.rs | 116 pub fn fused_mul_add_m256(a: m256, b: m256, c: m256) -> m256 { in fused_mul_add_m256() argument
117 m256(unsafe { _mm256_fmadd_ps(a.0, b.0, c.0) }) in fused_mul_add_m256()
246 pub fn fused_mul_sub_m256(a: m256, b: m256, c: m256) -> m256 { in fused_mul_sub_m256() argument
247 m256(unsafe { _mm256_fmsub_ps(a.0, b.0, c.0) }) in fused_mul_sub_m256()
376 pub fn fused_mul_neg_add_m256(a: m256, b: m256, c: m256) -> m256 { in fused_mul_neg_add_m256() argument
377 m256(unsafe { _mm256_fnmadd_ps(a.0, b.0, c.0) }) in fused_mul_neg_add_m256()
506 pub fn fused_mul_neg_sub_m256(a: m256, b: m256, c: m256) -> m256 { in fused_mul_neg_sub_m256() argument
507 m256(unsafe { _mm256_fnmsub_ps(a.0, b.0, c.0) }) in fused_mul_neg_sub_m256()
594 pub fn fused_mul_addsub_m256(a: m256, b: m256, c: m256) -> m256 { in fused_mul_addsub_m256() argument
595 m256(unsafe { _mm256_fmaddsub_ps(a.0, b.0, c.0) }) in fused_mul_addsub_m256()
[all …]
|
| /dports/graphics/blender/blender-2.91.0/intern/cycles/util/ |
| H A D | util_avxf.h | 40 m256 = other.m256; in avxf()
44 m256 = other.m256;
159 return _mm256_add_ps(a.m256, b.m256);
172 return _mm256_sub_ps(a.m256, b.m256);
185 return _mm256_mul_ps(a.m256, b.m256);
198 return _mm256_div_ps(a.m256, b.m256);
211 return _mm256_or_ps(a.m256, b.m256);
216 return _mm256_xor_ps(a.m256, b.m256);
221 return _mm256_and_ps(a.m256, b.m256);
226 return _mm256_max_ps(a.m256, b.m256); in max()
[all …]
|
| H A D | util_avxi.h | 46 m256 = a.m256; in avxi()
50 m256 = a.m256;
199 return _mm256_add_epi32(a.m256, b.m256);
219 return _mm256_sub_epi32(a.m256, b.m256);
239 return _mm256_mullo_epi32(a.m256, b.m256);
259 return _mm256_and_si256(a.m256, b.m256);
279 return _mm256_or_si256(a.m256, b.m256);
299 return _mm256_xor_si256(a.m256, b.m256);
357 return _mm256_min_epi32(a.m256, b.m256); in min()
377 return _mm256_max_epi32(a.m256, b.m256); in max()
[all …]
|
| H A D | util_avxb.h | 32 __m256 m256; member
45 m256 = other.m256; in avxb()
49 m256 = other.m256;
53 __forceinline avxb(const __m256 input) : m256(input) in avxb()
57 : m256(_mm256_insertf128_ps(_mm256_castps128_ps256(a), b, 1)) in avxb()
62 return m256;
66 return _mm256_castps_si256(m256); in __m256i()
70 return _mm256_castps_pd(m256); in __m256d()
77 __forceinline avxb(FalseTy) : m256(_mm256_setzero_ps()) in avxb()
80 __forceinline avxb(TrueTy) : m256(_mm256_castsi256_ps(_mm256_set1_epi32(-1))) in avxb()
[all …]
|
| H A D | util_types_float8_impl.h | 48 __forceinline float8::float8(const float8 &f) : m256(f.m256) in float8()
52 __forceinline float8::float8(const __m256 &f) : m256(f) in float8()
58 return m256;
63 return m256;
68 m256 = f.m256;
|
| /dports/devel/hyperscan/hyperscan-5.4.0/src/fdr/ |
| H A D | teddy_avx2.c | 496 m256 vectoredLoad2x128(m256 *p_mask, const u8 *ptr, const size_t start_offset, in vectoredLoad2x128()
508 m256 prep_conf_fat_teddy_m1(const m256 *maskBase, m256 val) { in prep_conf_fat_teddy_m1()
509 m256 mask = set32x8(0xf); in prep_conf_fat_teddy_m1()
510 m256 lo = and256(val, mask); in prep_conf_fat_teddy_m1()
517 m256 prep_conf_fat_teddy_m2(const m256 *maskBase, m256 *old_1, m256 val) { in prep_conf_fat_teddy_m2()
518 m256 mask = set32x8(0xf); in prep_conf_fat_teddy_m2()
531 m256 prep_conf_fat_teddy_m3(const m256 *maskBase, m256 *old_1, m256 *old_2, in prep_conf_fat_teddy_m3()
533 m256 mask = set32x8(0xf); in prep_conf_fat_teddy_m3()
546 m256 prep_conf_fat_teddy_m4(const m256 *maskBase, m256 *old_1, m256 *old_2, in prep_conf_fat_teddy_m4()
547 m256 *old_3, m256 val) { in prep_conf_fat_teddy_m4()
[all …]
|
| /dports/devel/hyperscan/hyperscan-5.4.0/src/util/ |
| H A D | simd_utils.h | 338 m256 pshufb_m256(m256 a, m256 b) { in pshufb_m256()
434 m256 lshift64_m256(m256 a, int b) { in lshift64_m256()
457 m256 eq256(m256 a, m256 b) { in eq256()
497 static really_inline m256 and256(m256 a, m256 b) { in and256()
501 static really_inline m256 and256(m256 a, m256 b) { in and256()
510 static really_inline m256 or256(m256 a, m256 b) { in or256()
514 static really_inline m256 or256(m256 a, m256 b) { in or256()
523 static really_inline m256 xor256(m256 a, m256 b) { in xor256()
527 static really_inline m256 xor256(m256 a, m256 b) { in xor256()
549 static really_inline m256 andnot256(m256 a, m256 b) { in andnot256()
[all …]
|
| H A D | masked_move.h | 49 static really_inline m256
53 m256 mask = load256(masky + 32); in _get_mm_mask_end()
64 static really_inline m256
68 m256 lmask = _get_mm_mask_end(len); in masked_move256_len()
71 m256 preshufend = _mm256_broadcastq_epi64(_mm_cvtsi32_si128(end)); in masked_move256_len()
72 m256 v = _mm256_maskload_epi32((const int *)buf, lmask); in masked_move256_len()
73 m256 shufend = pshufb_m256(preshufend, in masked_move256_len()
75 m256 target = or256(v, shufend); in masked_move256_len()
|
| /dports/devel/hyperscan/hyperscan-5.4.0/src/hwlm/ |
| H A D | noodle_engine_avx2.c | 36 static really_inline m256 getCaseMask(void) { in getCaseMask()
43 m256 caseMask, m256 mask1, in scanSingleUnaligned()
50 m256 v = loadu256(d); in scanSingleUnaligned()
72 m256 caseMask, m256 mask1, m256 mask2, in scanDoubleUnaligned()
79 m256 v = loadu256(d); in scanDoubleUnaligned()
105 size_t len, bool noCase, m256 caseMask, m256 mask1, in scanSingleShort()
116 m256 v; in scanSingleShort()
145 size_t len, bool noCase, m256 caseMask, m256 mask1, in scanDoubleShort()
154 m256 v; in scanDoubleShort()
186 size_t len, bool noCase, m256 caseMask, m256 mask1, in scanSingleFast()
[all …]
|
| /dports/graphics/librsvg2-rust/librsvg-2.52.8/vendor/safe_arch/tests/integration/ |
| H A D | avx_tests.rs | 13 let a = m256::from_array([1.0, 2.0, 3.0, 4.0, 20.0, 30.0, 40.0, 50.0]); in test_add_m256()
45 let a = m256::from_array([1.0, 0.0, 1.0, 0.0, 1.0, 0.0, 1.0, 0.0]); in test_bitand_m256()
46 let b = m256::from_array([1.0, 1.0, 0.0, 0.0, 1.0, 1.0, 0.0, 0.0]); in test_bitand_m256()
61 let a = m256::from_array([1.0, 0.0, 1.0, 0.0, 1.0, 0.0, 1.0, 0.0]); in test_bitandnot_m256()
160 let a = m256::from([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0]); in test_cast_to_m128_from_m256()
271 let a = m256::from([4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0]); in test_convert_to_i32_m256i_from_m256()
299 let a = m256::from([4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0]); in test_convert_to_f32_from_m256_s()
389 let a = m256::from([8.0, 7.0, 6.0, 5.0, 4.0, 3.0, 2.0, 1.0]); in test_add_horizontal_m256()
453 let a = m256::from([0.0; 8]); in test_insert_m128_to_m256()
567 let mut a = m256::default(); in test_store_masked_m256()
[all …]
|
| /dports/devel/hyperscan/hyperscan-5.4.0/src/nfa/ |
| H A D | shufti.c | 374 u32 block(m256 mask_lo, m256 mask_hi, m256 chars, const m256 low4bits, in block()
375 const m256 compare) { in block()
438 const u8 *fwdBlock(m256 mask_lo, m256 mask_hi, m256 chars, const u8 *buf, in fwdBlock()
439 const m256 low4bits, const m256 zeroes) { in fwdBlock()
473 m256 chars = loadu256(buf); in shuftiExec()
517 const u8 *revBlock(m256 mask_lo, m256 mask_hi, m256 chars, const u8 *buf, in revBlock()
518 const m256 low4bits, const m256 zeroes) { in revBlock()
627 const u8 *fwdBlock2(m256 mask1_lo, m256 mask1_hi, m256 mask2_lo, m256 mask2_hi, in fwdBlock2()
628 m256 chars, const u8 *buf, const m256 low4bits, in fwdBlock2()
660 const u8 *fwdBlockShort2(m256 mask1, m256 mask2, m128 chars, const u8 *buf, in fwdBlockShort2()
[all …]
|
| H A D | truffle.c | 261 u32 block(m256 shuf_mask_lo_highclear, m256 shuf_mask_lo_highset, m256 v) { in block()
268 m256 t1 = xor256(v, highconst); in block()
273 m256 tmp2 = eq256(tmp, zeroes256()); in block()
280 const u8 *truffleMini(m256 shuf_mask_lo_highclear, m256 shuf_mask_lo_highset, in truffleMini()
285 m256 chars = zeroes256(); in truffleMini()
301 const u8 *fwdBlock(m256 shuf_mask_lo_highclear, m256 shuf_mask_lo_highset, in fwdBlock()
308 const u8 *revBlock(m256 shuf_mask_lo_highclear, m256 shuf_mask_lo_highset, in revBlock()
333 m256 chars = loadu256(buf); in truffleExec()
342 m256 lchars = load256(buf); in truffleExec()
368 m256 chars = zeroes256(); in truffleRevMini()
[all …]
|
| H A D | limex_shuffle.h | 55 u32 packedExtract256(m256 s, const m256 permute, const m256 compare) { in packedExtract256()
57 m256 shuffled = pshufb_m256(s, permute); in packedExtract256()
58 m256 compared = and256(shuffled, compare); in packedExtract256()
|
| /dports/devel/hyperscan/hyperscan-5.4.0/unit/internal/ |
| H A D | simd_utils.cpp | 79 bool simd_diff(const m256 &a, const m256 &b) { return !!diff256(a, b); } in simd_diff()
87 m256 simd_and(const m256 &a, const m256 &b) { return and256(a, b); } in simd_and()
91 m256 simd_or(const m256 &a, const m256 &b) { return or256(a, b); } in simd_or()
95 m256 simd_xor(const m256 &a, const m256 &b) { return xor256(a, b); } in simd_xor()
99 m256 simd_andnot(const m256 &a, const m256 &b) { return andnot256(a, b); } in simd_andnot()
103 m256 simd_not(const m256 &a) { return not256(a); } in simd_not()
119 u32 simd_diffrich(const m256 &a, const m256 &b) { return diffrich256(a, b); } in simd_diffrich()
224 m256 a = setbit<m256>(i); in TEST()
225 m256 b = setbit<m256>(i); in TEST()
231 m256 a = setbit<m256>(i); in TEST()
[all …]
|
| /dports/devel/simde/simde-0.7.2/simde/x86/avx512/ |
| H A D | div.h | 51 for (size_t i = 0 ; i < (sizeof(r_.m256) / sizeof(r_.m256[0])) ; i++) { in simde_mm512_div_ps()
52 r_.m256[i] = simde_mm256_div_ps(a_.m256[i], b_.m256[i]); in simde_mm512_div_ps()
58 for (size_t i = 0 ; i < (sizeof(r_.m256) / sizeof(r_.m256[0])) ; i++) { in simde_mm512_div_ps()
59 r_.m256[i] = simde_mm256_div_ps(a_.m256[i], b_.m256[i]); in simde_mm512_div_ps()
|
| /dports/biology/bowtie2/simde-no-tests-f6a0b3b/x86/avx512/ |
| H A D | div.h | 51 for (size_t i = 0 ; i < (sizeof(r_.m256) / sizeof(r_.m256[0])) ; i++) { in simde_mm512_div_ps()
52 r_.m256[i] = simde_mm256_div_ps(a_.m256[i], b_.m256[i]); in simde_mm512_div_ps()
58 for (size_t i = 0 ; i < (sizeof(r_.m256) / sizeof(r_.m256[0])) ; i++) { in simde_mm512_div_ps()
59 r_.m256[i] = simde_mm256_div_ps(a_.m256[i], b_.m256[i]); in simde_mm512_div_ps()
|
| /dports/biology/mmseqs2/MMseqs2-13-45111/lib/simde/simde/x86/avx512/ |
| H A D | div.h | 51 for (size_t i = 0 ; i < (sizeof(r_.m256) / sizeof(r_.m256[0])) ; i++) { in simde_mm512_div_ps()
52 r_.m256[i] = simde_mm256_div_ps(a_.m256[i], b_.m256[i]); in simde_mm512_div_ps()
58 for (size_t i = 0 ; i < (sizeof(r_.m256) / sizeof(r_.m256[0])) ; i++) { in simde_mm512_div_ps()
59 r_.m256[i] = simde_mm256_div_ps(a_.m256[i], b_.m256[i]); in simde_mm512_div_ps()
|
| /dports/devel/vc/Vc-1.4.2/Vc/avx/ |
| H A D | deinterleave.tcc | 52 m256 tmp0 = Mem::blend<X0, Y1, X2, X3, Y4, X5, X6, Y7>( ac0, b.data()); in deinterleave()
54 m256 tmp1 = Mem::blend<X0, X1, Y2, X3, X4, Y5, X6, X7>( ac0, b.data()); in deinterleave()
56 m256 tmp2 = Mem::blend<Y0, X1, X2, Y3, X4, X5, Y6, X7>( ac0, b.data()); in deinterleave()
115 const m256 tmp2 = _mm256_unpacklo_ps(tmp0, tmp1); // b5 b1 b4 b0 a5 a1 a4 a0 in deinterleave()
116 const m256 tmp3 = _mm256_unpackhi_ps(tmp0, tmp1); // b7 b3 b6 b2 a7 a3 a6 a2 in deinterleave()
204 const m256 tmp0 = avx_cast<m256>(Mem::shuffle128<Vc::X0, Vc::Y0>(a.data(), b.data())); in deinterleave()
205 const m256 tmp1 = avx_cast<m256>(Mem::shuffle128<Vc::X1, Vc::Y1>(a.data(), b.data())); in deinterleave()
207 const m256 tmp2 = _mm256_unpacklo_ps(tmp0, tmp1); // b5 b1 b4 b0 a5 a1 a4 a0 in deinterleave()
208 const m256 tmp3 = _mm256_unpackhi_ps(tmp0, tmp1); // b7 b3 b6 b2 a7 a3 a6 a2 in deinterleave()
235 const m256 tmp0 = avx_cast<m256>(Mem::shuffle128<Vc::X0, Vc::Y0>(a.data(), b.data())); in deinterleave()
[all …]
|
| /dports/security/snowflake-tor/snowflake-ead5a960d7fa19dc890ccbfc0765c5ab6629eaa9/vendor/github.com/mmcloughlin/avo/internal/data/ |
| H A D | x86.v0.2.csv | 1756 "VLDDQU ymm1, m256","VLDDQU m256, ymm1","vlddqu m256, ymm1","VEX.256.F2.0F.WIG F0 /r","V","V","AVX"…
1781 "VMOVAPD ymm1, ymm2/m256","VMOVAPD ymm2/m256, ymm1","vmovapd ymm2/m256, ymm1","VEX.256.66.0F.WIG 28…
1782 "VMOVAPD ymm2/m256, ymm1","VMOVAPD ymm1, ymm2/m256","vmovapd ymm1, ymm2/m256","VEX.256.66.0F.WIG 29…
1785 "VMOVAPS ymm1, ymm2/m256","VMOVAPS ymm2/m256, ymm1","vmovaps ymm2/m256, ymm1","VEX.256.0F.WIG 28 /r…
1786 "VMOVAPS ymm2/m256, ymm1","VMOVAPS ymm1, ymm2/m256","vmovaps ymm1, ymm2/m256","VEX.256.0F.WIG 29 /r…
1820 "VMOVNTPS m256, ymm1","VMOVNTPS ymm1, m256","vmovntps ymm1, m256","VEX.256.0F.WIG 2B /r","V","V","A…
1844 "VMOVUPS ymm1, ymm2/m256","VMOVUPS ymm2/m256, ymm1","vmovups ymm2/m256, ymm1","VEX.256.0F.WIG 10 /r…
1845 "VMOVUPS ymm2/m256, ymm1","VMOVUPS ymm1, ymm2/m256","vmovups ymm1, ymm2/m256","VEX.256.0F.WIG 11 /r…
2118 "VPTEST ymm1, ymm2/m256","VPTEST ymm2/m256, ymm1","vptest ymm2/m256, ymm1","VEX.256.66.0F38.WIG 17 …
2138 "VRCPPS ymm1, ymm2/m256","VRCPPS ymm2/m256, ymm1","vrcpps ymm2/m256, ymm1","VEX.256.0F.WIG 53 /r","…
[all …]
|
| /dports/lang/gcc9/gcc-9.4.0/gcc/testsuite/g++.target/i386/ |
| H A D | pr94046-1.C | 11 S (m256) in S()
22 f2 (m256 src, float const *base, m256i idx, m256 mask) in f2()
52 f7 (m256 x, m256 y, m256i c) in f7()
|
| /dports/lang/gcc11-devel/gcc-11-20211009/gcc/testsuite/g++.target/i386/ |
| H A D | pr94046-1.C | 11 S (m256) in S()
22 f2 (m256 src, float const *base, m256i idx, m256 mask) in f2()
52 f7 (m256 x, m256 y, m256i c) in f7()
|
| /dports/lang/gcc11/gcc-11.2.0/gcc/testsuite/g++.target/i386/ |
| H A D | pr94046-1.C | 11 S (m256) in S()
22 f2 (m256 src, float const *base, m256i idx, m256 mask) in f2()
52 f7 (m256 x, m256 y, m256i c) in f7()
|