| /src/crypto/openssl/crypto/modes/asm/ |
| H A D | ghash-riscv64.pl | 148 my ($z0,$z1,$z2,$z3,$t0,$t1,$polymod) = ("t0","t1","t2","t3","t4","t5","t6");
158 @{[brev8_rv64i $x0, $z0, $z1, $z2]}
159 @{[brev8_rv64i $x1, $z0, $z1, $z2]}
171 @{[clmul $z2, $x1, $y1]}
174 xor $z2, $z2, $t1
177 xor $z2, $z2, $t1
186 xor $z2, $z2, $t1
188 @{[clmulh $t1, $z2, $polymod]}
189 @{[clmul $t0, $z2, $polymod]}
194 @{[brev8_rv64i $x0, $z0, $z1, $z2]}
[all …]
|
| /src/sys/contrib/libsodium/src/libsodium/crypto_scalarmult/curve25519/sandy2x/ |
| H A D | curve25519_sandy2x.c | 22 #define z2 var[2] macro
45 z_51.v[0] = (z2[1] << 26) + z2[0]; in crypto_scalarmult_curve25519_sandy2x()
46 z_51.v[1] = (z2[3] << 26) + z2[2]; in crypto_scalarmult_curve25519_sandy2x()
47 z_51.v[2] = (z2[5] << 26) + z2[4]; in crypto_scalarmult_curve25519_sandy2x()
48 z_51.v[3] = (z2[7] << 26) + z2[6]; in crypto_scalarmult_curve25519_sandy2x()
49 z_51.v[4] = (z2[9] << 26) + z2[8]; in crypto_scalarmult_curve25519_sandy2x()
65 #undef z2
68 #define z2 var[1] macro
89 z_51.v[0] = (z2[1] << 26) + z2[0]; in crypto_scalarmult_curve25519_sandy2x_base()
90 z_51.v[1] = (z2[3] << 26) + z2[2]; in crypto_scalarmult_curve25519_sandy2x_base()
[all …]
|
| /src/sys/contrib/libsodium/src/libsodium/crypto_scalarmult/curve25519/ref10/ |
| H A D | x25519_ref10.c | 60 fe25519 z2; in crypto_scalarmult_curve25519_ref10() local
80 fe25519_0(z2); in crypto_scalarmult_curve25519_ref10()
90 fe25519_cswap(z2, z3, swap); in crypto_scalarmult_curve25519_ref10()
93 fe25519_sub(tmp1, x2, z2); in crypto_scalarmult_curve25519_ref10()
94 fe25519_add(x2, x2, z2); in crypto_scalarmult_curve25519_ref10()
95 fe25519_add(z2, x3, z3); in crypto_scalarmult_curve25519_ref10()
97 fe25519_mul(z2, z2, tmp1); in crypto_scalarmult_curve25519_ref10()
100 fe25519_add(x3, z3, z2); in crypto_scalarmult_curve25519_ref10()
101 fe25519_sub(z2, z3, z2); in crypto_scalarmult_curve25519_ref10()
104 fe25519_sq(z2, z2); in crypto_scalarmult_curve25519_ref10()
[all …]
|
| /src/contrib/bearssl/src/ec/ |
| H A D | ec_c25519_i31.c | 179 uint32_t x1[10], x2[10], x3[10], z2[10], z3[10]; in api_mul() local
228 br_i31_zero(z2, C255_P[0]); in api_mul()
229 memcpy(x2, z2, sizeof z2); in api_mul()
253 cswap(z2, z3, swap); in api_mul()
263 c255_add(a, x2, z2); in api_mul()
265 c255_sub(b, x2, z2); in api_mul()
291 c255_mul(z2, C255_A24, e); in api_mul()
292 c255_add(z2, z2, aa); in api_mul()
293 c255_mul(z2, e, z2); in api_mul()
303 cswap(z2, z3, swap); in api_mul()
[all …]
|
| H A D | ec_c25519_i15.c | 188 uint16_t x1[18], x2[18], x3[18], z2[18], z3[18]; in api_mul() local
237 br_i15_zero(z2, C255_P[0]); in api_mul()
238 memcpy(x2, z2, ILEN); in api_mul()
259 cswap(z2, z3, swap); in api_mul()
269 c255_add(a, x2, z2); in api_mul()
271 c255_sub(b, x2, z2); in api_mul()
297 c255_mul(z2, C255_A24, e); in api_mul()
298 c255_add(z2, z2, aa); in api_mul()
299 c255_mul(z2, e, z2); in api_mul()
309 cswap(z2, z3, swap); in api_mul()
[all …]
|
| H A D | ec_c25519_m62.c | 380 uint64_t x1[5], x2[5], z2[5], x3[5], z3[5]; in api_mul() local
412 memset(z2, 0, sizeof z2); in api_mul()
436 f255_cswap(z2, z3, swap); in api_mul()
451 f255_add(a, x2, z2); in api_mul()
457 f255_sub(b, x2, z2); in api_mul()
490 f255_mul_a24(z2, e); in api_mul()
491 f255_add(z2, aa, z2); in api_mul()
492 f255_mul(z2, e, z2); in api_mul()
496 f255_cswap(z2, z3, swap); in api_mul()
502 memcpy(x1, z2, sizeof z2); in api_mul()
[all …]
|
| H A D | ec_c25519_m31.c | 625 uint32_t x1[9], x2[9], x3[9], z2[9], z3[9]; in api_mul() local
651 memset(z2, 0, sizeof z2); in api_mul()
674 cswap(z2, z3, swap); in api_mul()
684 f255_add(a, x2, z2); in api_mul()
686 f255_sub(b, x2, z2); in api_mul()
712 f255_mul_a24(z2, e); in api_mul()
713 f255_add(z2, z2, aa); in api_mul()
714 f255_mul(z2, e, z2); in api_mul()
724 cswap(z2, z3, swap); in api_mul()
731 memcpy(a, z2, sizeof z2); in api_mul()
[all …]
|
| /src/contrib/arm-optimized-routines/math/aarch64/advsimd/ |
| H A D | atan.c | 84 float64x2_t z2 = vmulq_f64 (z, z); in atan() local
85 float64x2_t x2 = vmulq_f64 (z2, z2); in atan()
90 float64x2_t p01 = vfmaq_laneq_f64 (d->c0, z2, c13, 0); in atan()
91 float64x2_t p23 = vfmaq_laneq_f64 (d->c2, z2, c13, 1); in atan()
94 float64x2_t p45 = vfmaq_laneq_f64 (d->c4, z2, c57, 0); in atan()
95 float64x2_t p67 = vfmaq_laneq_f64 (d->c6, z2, c57, 1); in atan()
101 float64x2_t p89 = vfmaq_laneq_f64 (d->c8, z2, c911, 0); in atan()
102 float64x2_t p1011 = vfmaq_laneq_f64 (d->c10, z2, c911, 1); in atan()
105 float64x2_t p1213 = vfmaq_laneq_f64 (d->c12, z2, c1315, 0); in atan()
106 float64x2_t p1415 = vfmaq_laneq_f64 (d->c14, z2, c1315, 1); in atan()
[all …]
|
| H A D | asin.c | 82 float64x2_t z2 = vbslq_f64 (a_lt_half, vmulq_f64 (x, x), in asin() local
84 float64x2_t z = vbslq_f64 (a_lt_half, ax, vsqrtq_f64 (z2)); in asin()
87 float64x2_t z4 = vmulq_f64 (z2, z2); in asin()
96 float64x2_t p01 = vfmaq_laneq_f64 (d->c0, z2, c13, 0); in asin()
97 float64x2_t p23 = vfmaq_laneq_f64 (d->c2, z2, c13, 1); in asin()
100 float64x2_t p45 = vfmaq_laneq_f64 (d->c4, z2, c57, 0); in asin()
101 float64x2_t p67 = vfmaq_laneq_f64 (d->c6, z2, c57, 1); in asin()
104 float64x2_t p89 = vfmaq_laneq_f64 (d->c8, z2, c911, 0); in asin()
105 float64x2_t p1011 = vfmaq_laneq_f64 (d->c10, z2, c911, 1); in asin()
112 p = vfmaq_f64 (z, vmulq_f64 (z, z2), p); in asin()
|
| H A D | atan2.c | 108 float64x2_t z2 = vmulq_f64 (z, z); in atan2() local
109 float64x2_t x2 = vmulq_f64 (z2, z2); in atan2()
120 float64x2_t p01 = vfmaq_laneq_f64 (d->c0, z2, c13, 0); in atan2()
121 float64x2_t p23 = vfmaq_laneq_f64 (d->c2, z2, c13, 1); in atan2()
124 float64x2_t p45 = vfmaq_laneq_f64 (d->c4, z2, c57, 0); in atan2()
125 float64x2_t p67 = vfmaq_laneq_f64 (d->c6, z2, c57, 1); in atan2()
131 float64x2_t p89 = vfmaq_laneq_f64 (d->c8, z2, c911, 0); in atan2()
132 float64x2_t p1011 = vfmaq_laneq_f64 (d->c10, z2, c911, 1); in atan2()
135 float64x2_t p1213 = vfmaq_laneq_f64 (d->c12, z2, c1315, 0); in atan2()
136 float64x2_t p1415 = vfmaq_laneq_f64 (d->c14, z2, c1315, 1); in atan2()
[all …]
|
| H A D | tanf.c | 49 float32x4_t z2 = vmulq_f32 (z, z); in eval_poly() local
57 z2 = vbslq_f32 (will_uflow, v_f32 (0), z2); in eval_poly()
59 float32x4_t z4 = vmulq_f32 (z2, z2); in eval_poly()
60 return v_estrin_5_f32 (z, z2, z4, d->poly); in eval_poly()
111 float32x4_t z2 = vmulq_f32 (r, r); in tan() local
112 float32x4_t p = eval_poly (z2, d); in tan()
113 float32x4_t y = vfmaq_f32 (z, vmulq_f32 (z, z2), p); in tan()
|
| H A D | atan2f.c | 91 float32x4_t z2 = vmulq_f32 (z, z); in atan2() local
92 float32x4_t z4 = vmulq_f32 (z2, z2); in atan2()
95 float32x4_t p01 = vfmaq_laneq_f32 (d->c0, z2, c1357, 0); in atan2()
96 float32x4_t p23 = vfmaq_laneq_f32 (d->c2, z2, c1357, 1); in atan2()
97 float32x4_t p45 = vfmaq_laneq_f32 (d->c4, z2, c1357, 2); in atan2()
98 float32x4_t p67 = vfmaq_laneq_f32 (d->c6, z2, c1357, 3); in atan2()
105 ret = vaddq_f32 (vfmaq_f32 (z, ret, vmulq_f32 (z2, z)), shift); in atan2()
|
| H A D | atanf.c | 85 float32x4_t z2 = vmulq_f32 (z, z); in atan() local
86 float32x4_t z4 = vmulq_f32 (z2, z2); in atan()
89 v_pairwise_poly_3_f32 (z2, z4, d->poly), z4, in atan()
90 vmulq_f32 (z4, v_pairwise_poly_3_f32 (z2, z4, d->poly + 4))); in atan()
93 y = vaddq_f32 (vfmaq_f32 (az, y, vmulq_f32 (z2, az)), shift); in atan()
|
| H A D | acos.c | 87 float64x2_t z2 = vbslq_f64 (a_le_half, vmulq_f64 (x, x), in acos() local
89 float64x2_t z = vbslq_f64 (a_le_half, ax, vsqrtq_f64 (z2)); in acos()
92 float64x2_t z4 = vmulq_f64 (z2, z2); in acos()
95 float64x2_t p = v_estrin_11_f64 (z2, z4, z8, z16, d->poly); in acos()
98 p = vfmaq_f64 (z, vmulq_f64 (z, z2), p); in acos()
|
| /src/contrib/arm-optimized-routines/math/aarch64/experimental/ |
| H A D | atanf_common.h | 26 float z2 = z * z; in eval_poly() local
27 float z4 = z2 * z2; in eval_poly()
31 z4, z4 * pairwise_poly_3_f32 (z2, z4, __atanf_poly_data.poly + 4), in eval_poly()
32 pairwise_poly_3_f32 (z2, z4, __atanf_poly_data.poly)); in eval_poly()
35 return fmaf (y, z2 * az, az) + shift; in eval_poly()
|
| H A D | atan_common.h | 19 double z2 = z * z; in eval_poly() local
20 double x2 = z2 * z2; in eval_poly()
23 double y = fma (estrin_11_f64 (z2, x2, x4, x8, __atan_poly_data.poly + 8), in eval_poly()
24 x8, estrin_7_f64 (z2, x2, x4, __atan_poly_data.poly)); in eval_poly()
27 y = fma (y, z2 * az, az); in eval_poly()
|
| H A D | asin_3u.c | 79 double z2 = ax < 0.5 ? x * x : fma (-0.5, ax, 0.5); in asin() local
80 double z = ax < 0.5 ? ax : sqrt (z2); in asin()
83 double z4 = z2 * z2; in asin()
86 double p = estrin_11_f64 (z2, z4, z8, z16, __asin_poly); in asin()
89 p = fma (z * z2, p, z); in asin()
|
| H A D | acos_2u.c | 72 double z2 = ax < 0.5 ? x * x : fma (-0.5, ax, 0.5); in acos() local
73 double z = ax < 0.5 ? ax : sqrt (z2); in acos()
76 double z4 = z2 * z2; in acos()
79 double p = estrin_11_f64 (z2, z4, z8, z16, __asin_poly); in acos()
82 p = fma (z * z2, p, z); in acos()
|
| /src/contrib/arm-optimized-routines/math/aarch64/sve/ |
| H A D | asin.c | 58 svfloat64_t z2 = svsel (a_ge_half, svmls_x (pg, sv_f64 (0.5), ax, 0.5), in asin() local
60 svfloat64_t z = svsqrt_m (ax, a_ge_half, z2); in asin()
63 svfloat64_t z4 = svmul_x (pg, z2, z2); in asin()
66 svfloat64_t p = sv_estrin_11_f64_x (pg, z2, z4, z8, z16, d->poly); in asin()
68 p = svmla_x (pg, z, svmul_x (pg, z, z2), p); in asin()
|
| H A D | atan.c | 60 svfloat64_t z2 = svmul_x (pg, z, z); in atan() local
61 svfloat64_t x2 = svmul_x (pg, z2, z2); in atan()
66 = svmla_x (pg, sv_estrin_7_f64_x (pg, z2, x2, x4, d->poly), in atan()
67 sv_estrin_11_f64_x (pg, z2, x2, x4, x8, d->poly + 8), x8); in atan()
70 svfloat64_t z3 = svmul_x (pg, z2, az); in atan()
|
| H A D | acos.c | 58 svfloat64_t z2 = svsel (a_gt_half, svmls_x (pg, sv_f64 (0.5), ax, 0.5), in acos() local
60 svfloat64_t z = svsqrt_m (ax, a_gt_half, z2); in acos()
63 svfloat64_t z4 = svmul_x (pg, z2, z2); in acos()
66 svfloat64_t p = sv_estrin_11_f64_x (pg, z2, z4, z8, z16, d->poly); in acos()
69 p = svmla_x (pg, z, svmul_x (pg, z, z2), p); in acos()
|
| H A D | tanf.c | 81 svfloat32_t z2 = svmul_x (svptrue_b32 (), r, r); in tan() local
82 svfloat32_t p01 = svmla_lane (sv_f32 (d->c0), z2, odd_coeffs, 0); in tan()
83 svfloat32_t p23 = svmla_lane (sv_f32 (d->c2), z2, odd_coeffs, 1); in tan()
84 svfloat32_t p45 = svmla_lane (sv_f32 (d->c4), z2, odd_coeffs, 2); in tan()
86 svfloat32_t z4 = svmul_x (pg, z2, z2); in tan()
92 svfloat32_t y = svmla_x (pg, z, p, svmul_x (pg, z, z2)); in tan()
|
| H A D | atanf.c | 54 svfloat32_t z2 = svmul_x (pg, z, z); in atan() local
55 svfloat32_t z4 = svmul_x (pg, z2, z2); in atan()
58 svfloat32_t y = sv_estrin_7_f32_x (pg, z2, z4, z8, d->poly); in atan()
61 svfloat32_t z3 = svmul_x (pg, z2, az); in atan()
|
| /src/contrib/bearssl/src/hash/ |
| H A D | ghash_ctmul64.c | 37 uint64_t z0, z1, z2, z3; in bmul64() local
49 z2 = (x0 * y2) ^ (x1 * y1) ^ (x2 * y0) ^ (x3 * y3); in bmul64()
53 z2 &= (uint64_t)0x4444444444444444; in bmul64()
55 return z0 | z1 | z2 | z3; in bmul64()
100 uint64_t z0, z1, z2, z0h, z1h, z2h; in br_ghash_ctmul64() local
123 z2 = bmul64(y2, h2); in br_ghash_ctmul64()
127 z2 ^= z0 ^ z1; in br_ghash_ctmul64()
134 v1 = z0h ^ z2; in br_ghash_ctmul64()
|
| /src/crypto/openssl/crypto/ec/curve448/ |
| H A D | curve448.c | 384 gf x1, x2, z2, x3, z3, t1, t2; in ossl_x448_int() local
391 gf_copy(z2, ZERO); in ossl_x448_int()
410 gf_cond_swap(z2, z3, swap); in ossl_x448_int()
418 gf_add_nr(t1, x2, z2); /* A = x2 + z2 */ /* 2+e */ in ossl_x448_int()
419 gf_sub_nr(t2, x2, z2); /* B = x2 - z2 */ /* 3+e */ in ossl_x448_int()
420 gf_sub_nr(z2, x3, z3); /* D = x3 - z3 */ /* 3+e */ in ossl_x448_int()
421 ossl_gf_mul(x2, t1, z2); /* DA */ in ossl_x448_int()
422 gf_add_nr(z2, z3, x3); /* C = x3 + z3 */ /* 2+e */ in ossl_x448_int()
423 ossl_gf_mul(x3, t2, z2); /* CB */ in ossl_x448_int()
425 ossl_gf_sqr(z2, z3); /* (DA-CB)^2 */ in ossl_x448_int()
[all …]
|