Lines Matching full:v
79 INLINE void round_fn4(uint32x4_t v[16], uint32x4_t m[16], size_t r) { in round_fn4()
80 v[0] = add_128(v[0], m[(size_t)MSG_SCHEDULE[r][0]]); in round_fn4()
81 v[1] = add_128(v[1], m[(size_t)MSG_SCHEDULE[r][2]]); in round_fn4()
82 v[2] = add_128(v[2], m[(size_t)MSG_SCHEDULE[r][4]]); in round_fn4()
83 v[3] = add_128(v[3], m[(size_t)MSG_SCHEDULE[r][6]]); in round_fn4()
84 v[0] = add_128(v[0], v[4]); in round_fn4()
85 v[1] = add_128(v[1], v[5]); in round_fn4()
86 v[2] = add_128(v[2], v[6]); in round_fn4()
87 v[3] = add_128(v[3], v[7]); in round_fn4()
88 v[12] = xor_128(v[12], v[0]); in round_fn4()
89 v[13] = xor_128(v[13], v[1]); in round_fn4()
90 v[14] = xor_128(v[14], v[2]); in round_fn4()
91 v[15] = xor_128(v[15], v[3]); in round_fn4()
92 v[12] = rot16_128(v[12]); in round_fn4()
93 v[13] = rot16_128(v[13]); in round_fn4()
94 v[14] = rot16_128(v[14]); in round_fn4()
95 v[15] = rot16_128(v[15]); in round_fn4()
96 v[8] = add_128(v[8], v[12]); in round_fn4()
97 v[9] = add_128(v[9], v[13]); in round_fn4()
98 v[10] = add_128(v[10], v[14]); in round_fn4()
99 v[11] = add_128(v[11], v[15]); in round_fn4()
100 v[4] = xor_128(v[4], v[8]); in round_fn4()
101 v[5] = xor_128(v[5], v[9]); in round_fn4()
102 v[6] = xor_128(v[6], v[10]); in round_fn4()
103 v[7] = xor_128(v[7], v[11]); in round_fn4()
104 v[4] = rot12_128(v[4]); in round_fn4()
105 v[5] = rot12_128(v[5]); in round_fn4()
106 v[6] = rot12_128(v[6]); in round_fn4()
107 v[7] = rot12_128(v[7]); in round_fn4()
108 v[0] = add_128(v[0], m[(size_t)MSG_SCHEDULE[r][1]]); in round_fn4()
109 v[1] = add_128(v[1], m[(size_t)MSG_SCHEDULE[r][3]]); in round_fn4()
110 v[2] = add_128(v[2], m[(size_t)MSG_SCHEDULE[r][5]]); in round_fn4()
111 v[3] = add_128(v[3], m[(size_t)MSG_SCHEDULE[r][7]]); in round_fn4()
112 v[0] = add_128(v[0], v[4]); in round_fn4()
113 v[1] = add_128(v[1], v[5]); in round_fn4()
114 v[2] = add_128(v[2], v[6]); in round_fn4()
115 v[3] = add_128(v[3], v[7]); in round_fn4()
116 v[12] = xor_128(v[12], v[0]); in round_fn4()
117 v[13] = xor_128(v[13], v[1]); in round_fn4()
118 v[14] = xor_128(v[14], v[2]); in round_fn4()
119 v[15] = xor_128(v[15], v[3]); in round_fn4()
120 v[12] = rot8_128(v[12]); in round_fn4()
121 v[13] = rot8_128(v[13]); in round_fn4()
122 v[14] = rot8_128(v[14]); in round_fn4()
123 v[15] = rot8_128(v[15]); in round_fn4()
124 v[8] = add_128(v[8], v[12]); in round_fn4()
125 v[9] = add_128(v[9], v[13]); in round_fn4()
126 v[10] = add_128(v[10], v[14]); in round_fn4()
127 v[11] = add_128(v[11], v[15]); in round_fn4()
128 v[4] = xor_128(v[4], v[8]); in round_fn4()
129 v[5] = xor_128(v[5], v[9]); in round_fn4()
130 v[6] = xor_128(v[6], v[10]); in round_fn4()
131 v[7] = xor_128(v[7], v[11]); in round_fn4()
132 v[4] = rot7_128(v[4]); in round_fn4()
133 v[5] = rot7_128(v[5]); in round_fn4()
134 v[6] = rot7_128(v[6]); in round_fn4()
135 v[7] = rot7_128(v[7]); in round_fn4()
137 v[0] = add_128(v[0], m[(size_t)MSG_SCHEDULE[r][8]]); in round_fn4()
138 v[1] = add_128(v[1], m[(size_t)MSG_SCHEDULE[r][10]]); in round_fn4()
139 v[2] = add_128(v[2], m[(size_t)MSG_SCHEDULE[r][12]]); in round_fn4()
140 v[3] = add_128(v[3], m[(size_t)MSG_SCHEDULE[r][14]]); in round_fn4()
141 v[0] = add_128(v[0], v[5]); in round_fn4()
142 v[1] = add_128(v[1], v[6]); in round_fn4()
143 v[2] = add_128(v[2], v[7]); in round_fn4()
144 v[3] = add_128(v[3], v[4]); in round_fn4()
145 v[15] = xor_128(v[15], v[0]); in round_fn4()
146 v[12] = xor_128(v[12], v[1]); in round_fn4()
147 v[13] = xor_128(v[13], v[2]); in round_fn4()
148 v[14] = xor_128(v[14], v[3]); in round_fn4()
149 v[15] = rot16_128(v[15]); in round_fn4()
150 v[12] = rot16_128(v[12]); in round_fn4()
151 v[13] = rot16_128(v[13]); in round_fn4()
152 v[14] = rot16_128(v[14]); in round_fn4()
153 v[10] = add_128(v[10], v[15]); in round_fn4()
154 v[11] = add_128(v[11], v[12]); in round_fn4()
155 v[8] = add_128(v[8], v[13]); in round_fn4()
156 v[9] = add_128(v[9], v[14]); in round_fn4()
157 v[5] = xor_128(v[5], v[10]); in round_fn4()
158 v[6] = xor_128(v[6], v[11]); in round_fn4()
159 v[7] = xor_128(v[7], v[8]); in round_fn4()
160 v[4] = xor_128(v[4], v[9]); in round_fn4()
161 v[5] = rot12_128(v[5]); in round_fn4()
162 v[6] = rot12_128(v[6]); in round_fn4()
163 v[7] = rot12_128(v[7]); in round_fn4()
164 v[4] = rot12_128(v[4]); in round_fn4()
165 v[0] = add_128(v[0], m[(size_t)MSG_SCHEDULE[r][9]]); in round_fn4()
166 v[1] = add_128(v[1], m[(size_t)MSG_SCHEDULE[r][11]]); in round_fn4()
167 v[2] = add_128(v[2], m[(size_t)MSG_SCHEDULE[r][13]]); in round_fn4()
168 v[3] = add_128(v[3], m[(size_t)MSG_SCHEDULE[r][15]]); in round_fn4()
169 v[0] = add_128(v[0], v[5]); in round_fn4()
170 v[1] = add_128(v[1], v[6]); in round_fn4()
171 v[2] = add_128(v[2], v[7]); in round_fn4()
172 v[3] = add_128(v[3], v[4]); in round_fn4()
173 v[15] = xor_128(v[15], v[0]); in round_fn4()
174 v[12] = xor_128(v[12], v[1]); in round_fn4()
175 v[13] = xor_128(v[13], v[2]); in round_fn4()
176 v[14] = xor_128(v[14], v[3]); in round_fn4()
177 v[15] = rot8_128(v[15]); in round_fn4()
178 v[12] = rot8_128(v[12]); in round_fn4()
179 v[13] = rot8_128(v[13]); in round_fn4()
180 v[14] = rot8_128(v[14]); in round_fn4()
181 v[10] = add_128(v[10], v[15]); in round_fn4()
182 v[11] = add_128(v[11], v[12]); in round_fn4()
183 v[8] = add_128(v[8], v[13]); in round_fn4()
184 v[9] = add_128(v[9], v[14]); in round_fn4()
185 v[5] = xor_128(v[5], v[10]); in round_fn4()
186 v[6] = xor_128(v[6], v[11]); in round_fn4()
187 v[7] = xor_128(v[7], v[8]); in round_fn4()
188 v[4] = xor_128(v[4], v[9]); in round_fn4()
189 v[5] = rot7_128(v[5]); in round_fn4()
190 v[6] = rot7_128(v[6]); in round_fn4()
191 v[7] = rot7_128(v[7]); in round_fn4()
192 v[4] = rot7_128(v[4]); in round_fn4()
268 uint32x4_t v[16] = { in blake3_hash4_neon() local
274 round_fn4(v, msg_vecs, 0); in blake3_hash4_neon()
275 round_fn4(v, msg_vecs, 1); in blake3_hash4_neon()
276 round_fn4(v, msg_vecs, 2); in blake3_hash4_neon()
277 round_fn4(v, msg_vecs, 3); in blake3_hash4_neon()
278 round_fn4(v, msg_vecs, 4); in blake3_hash4_neon()
279 round_fn4(v, msg_vecs, 5); in blake3_hash4_neon()
280 round_fn4(v, msg_vecs, 6); in blake3_hash4_neon()
281 h_vecs[0] = xor_128(v[0], v[8]); in blake3_hash4_neon()
282 h_vecs[1] = xor_128(v[1], v[9]); in blake3_hash4_neon()
283 h_vecs[2] = xor_128(v[2], v[10]); in blake3_hash4_neon()
284 h_vecs[3] = xor_128(v[3], v[11]); in blake3_hash4_neon()
285 h_vecs[4] = xor_128(v[4], v[12]); in blake3_hash4_neon()
286 h_vecs[5] = xor_128(v[5], v[13]); in blake3_hash4_neon()
287 h_vecs[6] = xor_128(v[6], v[14]); in blake3_hash4_neon()
288 h_vecs[7] = xor_128(v[7], v[15]); in blake3_hash4_neon()