1/*	$NetBSD: aes_ct64_dec.c,v 1.1 2025/11/23 22:44:14 riastradh Exp $	*/
2
3/*
4 * Copyright (c) 2016 Thomas Pornin <pornin@bolet.org>
5 *
6 * Permission is hereby granted, free of charge, to any person obtaining
7 * a copy of this software and associated documentation files (the
8 * "Software"), to deal in the Software without restriction, including
9 * without limitation the rights to use, copy, modify, merge, publish,
10 * distribute, sublicense, and/or sell copies of the Software, and to
11 * permit persons to whom the Software is furnished to do so, subject to
12 * the following conditions:
13 *
14 * The above copyright notice and this permission notice shall be
15 * included in all copies or substantial portions of the Software.
16 *
17 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
18 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
19 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
20 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
21 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
22 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
23 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
24 * SOFTWARE.
25 */
26
27#include <sys/cdefs.h>
28__KERNEL_RCSID(1, "$NetBSD: aes_ct64_dec.c,v 1.1 2025/11/23 22:44:14 riastradh Exp $");
29
30#include <sys/types.h>
31
32#include <crypto/aes/aes_bear64.h>
33
34/* see inner.h */
35void
36br_aes_ct64_bitslice_invSbox(uint64_t q[static 8])
37{
38	/*
39	 * See br_aes_ct_bitslice_invSbox(). This is the natural extension
40	 * to 64-bit registers.
41	 */
42	uint64_t q0, q1, q2, q3, q4, q5, q6, q7;
43
44	q0 = ~q[0];
45	q1 = ~q[1];
46	q2 = q[2];
47	q3 = q[3];
48	q4 = q[4];
49	q5 = ~q[5];
50	q6 = ~q[6];
51	q7 = q[7];
52	q[7] = q1 ^ q4 ^ q6;
53	q[6] = q0 ^ q3 ^ q5;
54	q[5] = q7 ^ q2 ^ q4;
55	q[4] = q6 ^ q1 ^ q3;
56	q[3] = q5 ^ q0 ^ q2;
57	q[2] = q4 ^ q7 ^ q1;
58	q[1] = q3 ^ q6 ^ q0;
59	q[0] = q2 ^ q5 ^ q7;
60
61	br_aes_ct64_bitslice_Sbox(q);
62
63	q0 = ~q[0];
64	q1 = ~q[1];
65	q2 = q[2];
66	q3 = q[3];
67	q4 = q[4];
68	q5 = ~q[5];
69	q6 = ~q[6];
70	q7 = q[7];
71	q[7] = q1 ^ q4 ^ q6;
72	q[6] = q0 ^ q3 ^ q5;
73	q[5] = q7 ^ q2 ^ q4;
74	q[4] = q6 ^ q1 ^ q3;
75	q[3] = q5 ^ q0 ^ q2;
76	q[2] = q4 ^ q7 ^ q1;
77	q[1] = q3 ^ q6 ^ q0;
78	q[0] = q2 ^ q5 ^ q7;
79}
80
81static void
82add_round_key(uint64_t q[static 8], const uint64_t sk[static 8])
83{
84	int i;
85
86	for (i = 0; i < 8; i ++) {
87		q[i] ^= sk[i];
88	}
89}
90
91static void
92inv_shift_rows(uint64_t q[static 8])
93{
94	int i;
95
96	for (i = 0; i < 8; i ++) {
97		uint64_t x;
98
99		x = q[i];
100		q[i] = (x & (uint64_t)0x000000000000FFFF)
101			| ((x & (uint64_t)0x000000000FFF0000) << 4)
102			| ((x & (uint64_t)0x00000000F0000000) >> 12)
103			| ((x & (uint64_t)0x000000FF00000000) << 8)
104			| ((x & (uint64_t)0x0000FF0000000000) >> 8)
105			| ((x & (uint64_t)0x000F000000000000) << 12)
106			| ((x & (uint64_t)0xFFF0000000000000) >> 4);
107	}
108}
109
110static inline uint64_t
111rotr32(uint64_t x)
112{
113	return (x << 32) | (x >> 32);
114}
115
116static void
117inv_mix_columns(uint64_t q[static 8])
118{
119	uint64_t q0, q1, q2, q3, q4, q5, q6, q7;
120	uint64_t r0, r1, r2, r3, r4, r5, r6, r7;
121
122	q0 = q[0];
123	q1 = q[1];
124	q2 = q[2];
125	q3 = q[3];
126	q4 = q[4];
127	q5 = q[5];
128	q6 = q[6];
129	q7 = q[7];
130	r0 = (q0 >> 16) | (q0 << 48);
131	r1 = (q1 >> 16) | (q1 << 48);
132	r2 = (q2 >> 16) | (q2 << 48);
133	r3 = (q3 >> 16) | (q3 << 48);
134	r4 = (q4 >> 16) | (q4 << 48);
135	r5 = (q5 >> 16) | (q5 << 48);
136	r6 = (q6 >> 16) | (q6 << 48);
137	r7 = (q7 >> 16) | (q7 << 48);
138
139	q[0] = q5 ^ q6 ^ q7 ^ r0 ^ r5 ^ r7 ^ rotr32(q0 ^ q5 ^ q6 ^ r0 ^ r5);
140	q[1] = q0 ^ q5 ^ r0 ^ r1 ^ r5 ^ r6 ^ r7 ^ rotr32(q1 ^ q5 ^ q7 ^ r1 ^ r5 ^ r6);
141	q[2] = q0 ^ q1 ^ q6 ^ r1 ^ r2 ^ r6 ^ r7 ^ rotr32(q0 ^ q2 ^ q6 ^ r2 ^ r6 ^ r7);
142	q[3] = q0 ^ q1 ^ q2 ^ q5 ^ q6 ^ r0 ^ r2 ^ r3 ^ r5 ^ rotr32(q0 ^ q1 ^ q3 ^ q5 ^ q6 ^ q7 ^ r0 ^ r3 ^ r5 ^ r7);
143	q[4] = q1 ^ q2 ^ q3 ^ q5 ^ r1 ^ r3 ^ r4 ^ r5 ^ r6 ^ r7 ^ rotr32(q1 ^ q2 ^ q4 ^ q5 ^ q7 ^ r1 ^ r4 ^ r5 ^ r6);
144	q[5] = q2 ^ q3 ^ q4 ^ q6 ^ r2 ^ r4 ^ r5 ^ r6 ^ r7 ^ rotr32(q2 ^ q3 ^ q5 ^ q6 ^ r2 ^ r5 ^ r6 ^ r7);
145	q[6] = q3 ^ q4 ^ q5 ^ q7 ^ r3 ^ r5 ^ r6 ^ r7 ^ rotr32(q3 ^ q4 ^ q6 ^ q7 ^ r3 ^ r6 ^ r7);
146	q[7] = q4 ^ q5 ^ q6 ^ r4 ^ r6 ^ r7 ^ rotr32(q4 ^ q5 ^ q7 ^ r4 ^ r7);
147}
148
149/* see inner.h */
150void
151br_aes_ct64_bitslice_decrypt(unsigned num_rounds,
152	const uint64_t skey[static 120], uint64_t q[static 8])
153{
154	unsigned u;
155
156	add_round_key(q, skey + (num_rounds << 3));
157	for (u = num_rounds - 1; u > 0; u --) {
158		inv_shift_rows(q);
159		br_aes_ct64_bitslice_invSbox(q);
160		add_round_key(q, skey + (u << 3));
161		inv_mix_columns(q);
162	}
163	inv_shift_rows(q);
164	br_aes_ct64_bitslice_invSbox(q);
165	add_round_key(q, skey);
166}
167
168/* NetBSD addition, for generating compatible decryption keys */
169void
170br_aes_ct64_inv_mix_columns(uint64_t q[static 8])
171{
172
173	inv_mix_columns(q);
174}
175