1
2
3
4
5
6
7
8
9
10
11
12#ifndef _SKEIN_BLOCK_H_
13#define _SKEIN_BLOCK_H_
14
15#include "skein_base.h"
16
17#ifndef SKEIN_USE_ASM
18#define SKEIN_USE_ASM (0)
19#endif
20
21#ifndef SKEIN_LOOP
22#define SKEIN_LOOP 001
23#endif
24
25#define BLK_BITS (WCNT * 64)
26#define KW_TWK_BASE (0)
27#define KW_KEY_BASE (3)
28#define ks (kw + KW_KEY_BASE)
29#define ts (kw + KW_TWK_BASE)
30
31#ifdef SKEIN_DEBUG
32#define debug_save_tweak(ctx) \
33{ \
34 ctx->h.tweak[0] = ts[0]; \
35 ctx->h.tweak[1] = ts[1]; \
36}
37#else
38#define debug_save_tweak(ctx)
39#endif
40
41#if !(SKEIN_USE_ASM & 256)
42#undef RCNT
43#define RCNT (SKEIN_256_ROUNDS_TOTAL / 8)
44#ifdef SKEIN_LOOP
45#define SKEIN_UNROLL_256 (((SKEIN_LOOP) / 100) % 10)
46#else
47#define SKEIN_UNROLL_256 (0)
48#endif
49
50#if SKEIN_UNROLL_256
51#if (RCNT % SKEIN_UNROLL_256)
52#error "Invalid SKEIN_UNROLL_256"
53#endif
54#endif
55#define ROUND256(p0, p1, p2, p3, ROT, r_num) \
56 do { \
57 X##p0 += X##p1; \
58 X##p1 = rol64(X##p1, ROT##_0); \
59 X##p1 ^= X##p0; \
60 X##p2 += X##p3; \
61 X##p3 = rol64(X##p3, ROT##_1); \
62 X##p3 ^= X##p2; \
63 } while (0)
64
65#if SKEIN_UNROLL_256 == 0
66#define R256(p0, p1, p2, p3, ROT, r_num) \
67 ROUND256(p0, p1, p2, p3, ROT, r_num)
68
69#define I256(R) \
70 do { \
71 \
72 X0 += ks[((R) + 1) % 5]; \
73 X1 += ks[((R) + 2) % 5] + ts[((R) + 1) % 3]; \
74 X2 += ks[((R) + 3) % 5] + ts[((R) + 2) % 3]; \
75 X3 += ks[((R) + 4) % 5] + (R) + 1; \
76 } while (0)
77#else
78
79#define R256(p0, p1, p2, p3, ROT, r_num) ROUND256(p0, p1, p2, p3, ROT, r_num)
80
81#define I256(R) \
82 do { \
83 \
84 X0 += ks[r + (R) + 0]; \
85 X1 += ks[r + (R) + 1] + ts[r + (R) + 0];\
86 X2 += ks[r + (R) + 2] + ts[r + (R) + 1];\
87 X3 += ks[r + (R) + 3] + r + (R); \
88 \
89 ks[r + (R) + 4] = ks[r + (R) - 1]; \
90 ts[r + (R) + 2] = ts[r + (R) - 1]; \
91 } while (0)
92#endif
93#define R256_8_ROUNDS(R) \
94 do { \
95 R256(0, 1, 2, 3, R_256_0, 8 * (R) + 1); \
96 R256(0, 3, 2, 1, R_256_1, 8 * (R) + 2); \
97 R256(0, 1, 2, 3, R_256_2, 8 * (R) + 3); \
98 R256(0, 3, 2, 1, R_256_3, 8 * (R) + 4); \
99 I256(2 * (R)); \
100 R256(0, 1, 2, 3, R_256_4, 8 * (R) + 5); \
101 R256(0, 3, 2, 1, R_256_5, 8 * (R) + 6); \
102 R256(0, 1, 2, 3, R_256_6, 8 * (R) + 7); \
103 R256(0, 3, 2, 1, R_256_7, 8 * (R) + 8); \
104 I256(2 * (R) + 1); \
105 } while (0)
106
107#define R256_UNROLL_R(NN) \
108 ((SKEIN_UNROLL_256 == 0 && \
109 SKEIN_256_ROUNDS_TOTAL / 8 > (NN)) || \
110 (SKEIN_UNROLL_256 > (NN)))
111
112#if (SKEIN_UNROLL_256 > 14)
113#error "need more unrolling in skein_256_process_block"
114#endif
115#endif
116
117#if !(SKEIN_USE_ASM & 512)
118#undef RCNT
119#define RCNT (SKEIN_512_ROUNDS_TOTAL / 8)
120
121#ifdef SKEIN_LOOP
122#define SKEIN_UNROLL_512 (((SKEIN_LOOP) / 10) % 10)
123#else
124#define SKEIN_UNROLL_512 (0)
125#endif
126
127#if SKEIN_UNROLL_512
128#if (RCNT % SKEIN_UNROLL_512)
129#error "Invalid SKEIN_UNROLL_512"
130#endif
131#endif
132#define ROUND512(p0, p1, p2, p3, p4, p5, p6, p7, ROT, r_num) \
133 do { \
134 X##p0 += X##p1; \
135 X##p1 = rol64(X##p1, ROT##_0); \
136 X##p1 ^= X##p0; \
137 X##p2 += X##p3; \
138 X##p3 = rol64(X##p3, ROT##_1); \
139 X##p3 ^= X##p2; \
140 X##p4 += X##p5; \
141 X##p5 = rol64(X##p5, ROT##_2); \
142 X##p5 ^= X##p4; \
143 X##p6 += X##p7; \
144 X##p7 = rol64(X##p7, ROT##_3); \
145 X##p7 ^= X##p6; \
146 } while (0)
147
148#if SKEIN_UNROLL_512 == 0
149#define R512(p0, p1, p2, p3, p4, p5, p6, p7, ROT, r_num) \
150 ROUND512(p0, p1, p2, p3, p4, p5, p6, p7, ROT, r_num)
151
152#define I512(R) \
153 do { \
154 \
155 X0 += ks[((R) + 1) % 9]; \
156 X1 += ks[((R) + 2) % 9]; \
157 X2 += ks[((R) + 3) % 9]; \
158 X3 += ks[((R) + 4) % 9]; \
159 X4 += ks[((R) + 5) % 9]; \
160 X5 += ks[((R) + 6) % 9] + ts[((R) + 1) % 3]; \
161 X6 += ks[((R) + 7) % 9] + ts[((R) + 2) % 3]; \
162 X7 += ks[((R) + 8) % 9] + (R) + 1; \
163 } while (0)
164
165#else
166#define R512(p0, p1, p2, p3, p4, p5, p6, p7, ROT, r_num) \
167 ROUND512(p0, p1, p2, p3, p4, p5, p6, p7, ROT, r_num) \
168
169#define I512(R) \
170 do { \
171 \
172 X0 += ks[r + (R) + 0]; \
173 X1 += ks[r + (R) + 1]; \
174 X2 += ks[r + (R) + 2]; \
175 X3 += ks[r + (R) + 3]; \
176 X4 += ks[r + (R) + 4]; \
177 X5 += ks[r + (R) + 5] + ts[r + (R) + 0]; \
178 X6 += ks[r + (R) + 6] + ts[r + (R) + 1]; \
179 X7 += ks[r + (R) + 7] + r + (R); \
180 \
181 ks[r + (R) + 8] = ks[r + (R) - 1]; \
182 ts[r + (R) + 2] = ts[r + (R) - 1]; \
183 } while (0)
184#endif
185#define R512_8_ROUNDS(R) \
186 do { \
187 R512(0, 1, 2, 3, 4, 5, 6, 7, R_512_0, 8 * (R) + 1); \
188 R512(2, 1, 4, 7, 6, 5, 0, 3, R_512_1, 8 * (R) + 2); \
189 R512(4, 1, 6, 3, 0, 5, 2, 7, R_512_2, 8 * (R) + 3); \
190 R512(6, 1, 0, 7, 2, 5, 4, 3, R_512_3, 8 * (R) + 4); \
191 I512(2 * (R)); \
192 R512(0, 1, 2, 3, 4, 5, 6, 7, R_512_4, 8 * (R) + 5); \
193 R512(2, 1, 4, 7, 6, 5, 0, 3, R_512_5, 8 * (R) + 6); \
194 R512(4, 1, 6, 3, 0, 5, 2, 7, R_512_6, 8 * (R) + 7); \
195 R512(6, 1, 0, 7, 2, 5, 4, 3, R_512_7, 8 * (R) + 8); \
196 I512(2 * (R) + 1); \
197 } while (0)
198#define R512_UNROLL_R(NN) \
199 ((SKEIN_UNROLL_512 == 0 && \
200 SKEIN_512_ROUNDS_TOTAL / 8 > (NN)) || \
201 (SKEIN_UNROLL_512 > (NN)))
202
203#if (SKEIN_UNROLL_512 > 14)
204#error "need more unrolling in skein_512_process_block"
205#endif
206#endif
207
208#if !(SKEIN_USE_ASM & 1024)
209#undef RCNT
210#define RCNT (SKEIN_1024_ROUNDS_TOTAL / 8)
211#ifdef SKEIN_LOOP
212#define SKEIN_UNROLL_1024 ((SKEIN_LOOP) % 10)
213#else
214#define SKEIN_UNROLL_1024 (0)
215#endif
216
217#if (SKEIN_UNROLL_1024 != 0)
218#if (RCNT % SKEIN_UNROLL_1024)
219#error "Invalid SKEIN_UNROLL_1024"
220#endif
221#endif
222#define ROUND1024(p0, p1, p2, p3, p4, p5, p6, p7, p8, p9, pA, pB, pC, pD, pE, \
223 pF, ROT, r_num) \
224 do { \
225 X##p0 += X##p1; \
226 X##p1 = rol64(X##p1, ROT##_0); \
227 X##p1 ^= X##p0; \
228 X##p2 += X##p3; \
229 X##p3 = rol64(X##p3, ROT##_1); \
230 X##p3 ^= X##p2; \
231 X##p4 += X##p5; \
232 X##p5 = rol64(X##p5, ROT##_2); \
233 X##p5 ^= X##p4; \
234 X##p6 += X##p7; \
235 X##p7 = rol64(X##p7, ROT##_3); \
236 X##p7 ^= X##p6; \
237 X##p8 += X##p9; \
238 X##p9 = rol64(X##p9, ROT##_4); \
239 X##p9 ^= X##p8; \
240 X##pA += X##pB; \
241 X##pB = rol64(X##pB, ROT##_5); \
242 X##pB ^= X##pA; \
243 X##pC += X##pD; \
244 X##pD = rol64(X##pD, ROT##_6); \
245 X##pD ^= X##pC; \
246 X##pE += X##pF; \
247 X##pF = rol64(X##pF, ROT##_7); \
248 X##pF ^= X##pE; \
249 } while (0)
250
251#if SKEIN_UNROLL_1024 == 0
252#define R1024(p0, p1, p2, p3, p4, p5, p6, p7, p8, p9, pA, pB, pC, pD, pE, pF, \
253 ROT, rn) \
254 ROUND1024(p0, p1, p2, p3, p4, p5, p6, p7, p8, p9, pA, pB, pC, pD, pE, \
255 pF, ROT, rn) \
256
257#define I1024(R) \
258 do { \
259 \
260 X00 += ks[((R) + 1) % 17]; \
261 X01 += ks[((R) + 2) % 17]; \
262 X02 += ks[((R) + 3) % 17]; \
263 X03 += ks[((R) + 4) % 17]; \
264 X04 += ks[((R) + 5) % 17]; \
265 X05 += ks[((R) + 6) % 17]; \
266 X06 += ks[((R) + 7) % 17]; \
267 X07 += ks[((R) + 8) % 17]; \
268 X08 += ks[((R) + 9) % 17]; \
269 X09 += ks[((R) + 10) % 17]; \
270 X10 += ks[((R) + 11) % 17]; \
271 X11 += ks[((R) + 12) % 17]; \
272 X12 += ks[((R) + 13) % 17]; \
273 X13 += ks[((R) + 14) % 17] + ts[((R) + 1) % 3]; \
274 X14 += ks[((R) + 15) % 17] + ts[((R) + 2) % 3]; \
275 X15 += ks[((R) + 16) % 17] + (R) + 1; \
276 } while (0)
277#else
278#define R1024(p0, p1, p2, p3, p4, p5, p6, p7, p8, p9, pA, pB, pC, pD, pE, pF, \
279 ROT, rn) \
280 ROUND1024(p0, p1, p2, p3, p4, p5, p6, p7, p8, p9, pA, pB, pC, pD, pE, \
281 pF, ROT, rn) \
282
283#define I1024(R) \
284 do { \
285 \
286 X00 += ks[r + (R) + 0]; \
287 X01 += ks[r + (R) + 1]; \
288 X02 += ks[r + (R) + 2]; \
289 X03 += ks[r + (R) + 3]; \
290 X04 += ks[r + (R) + 4]; \
291 X05 += ks[r + (R) + 5]; \
292 X06 += ks[r + (R) + 6]; \
293 X07 += ks[r + (R) + 7]; \
294 X08 += ks[r + (R) + 8]; \
295 X09 += ks[r + (R) + 9]; \
296 X10 += ks[r + (R) + 10]; \
297 X11 += ks[r + (R) + 11]; \
298 X12 += ks[r + (R) + 12]; \
299 X13 += ks[r + (R) + 13] + ts[r + (R) + 0]; \
300 X14 += ks[r + (R) + 14] + ts[r + (R) + 1]; \
301 X15 += ks[r + (R) + 15] + r + (R); \
302 \
303 ks[r + (R) + 16] = ks[r + (R) - 1]; \
304 ts[r + (R) + 2] = ts[r + (R) - 1]; \
305 } while (0)
306
307#endif
308#define R1024_8_ROUNDS(R) \
309 do { \
310 R1024(00, 01, 02, 03, 04, 05, 06, 07, 08, 09, 10, 11, 12, \
311 13, 14, 15, R1024_0, 8 * (R) + 1); \
312 R1024(00, 09, 02, 13, 06, 11, 04, 15, 10, 07, 12, 03, 14, \
313 05, 08, 01, R1024_1, 8 * (R) + 2); \
314 R1024(00, 07, 02, 05, 04, 03, 06, 01, 12, 15, 14, 13, 08, \
315 11, 10, 09, R1024_2, 8 * (R) + 3); \
316 R1024(00, 15, 02, 11, 06, 13, 04, 09, 14, 01, 08, 05, 10, \
317 03, 12, 07, R1024_3, 8 * (R) + 4); \
318 I1024(2 * (R)); \
319 R1024(00, 01, 02, 03, 04, 05, 06, 07, 08, 09, 10, 11, 12, \
320 13, 14, 15, R1024_4, 8 * (R) + 5); \
321 R1024(00, 09, 02, 13, 06, 11, 04, 15, 10, 07, 12, 03, 14, \
322 05, 08, 01, R1024_5, 8 * (R) + 6); \
323 R1024(00, 07, 02, 05, 04, 03, 06, 01, 12, 15, 14, 13, 08, \
324 11, 10, 09, R1024_6, 8 * (R) + 7); \
325 R1024(00, 15, 02, 11, 06, 13, 04, 09, 14, 01, 08, 05, 10, \
326 03, 12, 07, R1024_7, 8 * (R) + 8); \
327 I1024(2 * (R) + 1); \
328 } while (0)
329
330#define R1024_UNROLL_R(NN) \
331 ((SKEIN_UNROLL_1024 == 0 && \
332 SKEIN_1024_ROUNDS_TOTAL / 8 > (NN)) || \
333 (SKEIN_UNROLL_1024 > (NN)))
334
335#if (SKEIN_UNROLL_1024 > 14)
336#error "need more unrolling in Skein_1024_Process_Block"
337#endif
338#endif
339
340void skein_256_process_block(struct skein_256_ctx *ctx, const u8 *blk_ptr,
341 size_t blk_cnt, size_t byte_cnt_add);
342void skein_512_process_block(struct skein_512_ctx *ctx, const u8 *blk_ptr,
343 size_t blk_cnt, size_t byte_cnt_add);
344void skein_1024_process_block(struct skein_1024_ctx *ctx, const u8 *blk_ptr,
345 size_t blk_cnt, size_t byte_cnt_add);
346
347#endif
348
