3 This file is part of the ARM-Crypto-Lib.
4 Copyright (C) 2006-2010 Daniel Otte (daniel.otte@rub.de)
6 This program is free software: you can redistribute it and/or modify
7 it under the terms of the GNU General Public License as published by
8 the Free Software Foundation, either version 3 of the License, or
9 (at your option) any later version.
11 This program is distributed in the hope that it will be useful,
12 but WITHOUT ANY WARRANTY; without even the implied warranty of
13 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 GNU General Public License for more details.
16 You should have received a copy of the GNU General Public License
17 along with this program. If not, see <http://www.gnu.org/licenses/>.
22 * \email daniel.otte@rub.de
24 * \license GPLv3 or later
30 #include "bmw_small.h"
32 #define SHL32(a,n) ((a)<<(n))
33 #define SHR32(a,n) ((a)>>(n))
34 #define ROTL32(a,n) (((a)<<(n))|((a)>>(32-(n))))
35 #define ROTR32(a,n) (((a)>>(n))|((a)<<(32-(n))))
44 void ctx_dump(const bmw_small_ctx_t* ctx){
46 cli_putstr("\r\n==== ctx dump ====");
48 cli_putstr("\r\n h[");
51 cli_hexdump_rev(&(ctx->h[i]), 4);
53 cli_putstr("\r\n counter = ");
54 cli_hexdump(&(ctx->counter), 4);
57 void dump_x(const uint32_t* q, uint8_t elements, char x){
59 cli_putstr("\r\n==== ");
61 cli_putstr(" dump ====");
62 for(i=0; i<elements;++i){
68 cli_hexdump_rev(&(q[i]), 4);
76 #define S32_0(x) ( (SHR32((x), 1)) ^ \
81 #define S32_1(x) ( (SHR32((x), 1)) ^ \
86 #define S32_2(x) ( (SHR32((x), 2)) ^ \
91 #define S32_3(x) ( (SHR32((x), 2)) ^ \
96 #define S32_4(x) ( (SHR32((x), 1)) ^ (x))
98 #define S32_5(x) ( (SHR32((x), 2)) ^ (x))
100 #define R32_1(x) (ROTL32((x), 3))
101 #define R32_2(x) (ROTL32((x), 7))
102 #define R32_3(x) (ROTL32((x), 13))
103 #define R32_4(x) (ROTL32((x), 16))
104 #define R32_5(x) (ROTR32((x), 13))
105 #define R32_6(x) (ROTR32((x), 9))
106 #define R32_7(x) (ROTR32((x), 5))
109 #include "f1_autogen.i"
112 void bmw_small_f0(uint32_t* q, uint32_t* h, const uint32_t* m){
114 uint32_t tr0, tr1, tr2, tmp;
115 t[ 0] = h[ 0] ^ m[ 0];
116 t[ 1] = h[ 1] ^ m[ 1];
117 t[ 2] = h[ 2] ^ m[ 2];
118 t[ 3] = h[ 3] ^ m[ 3];
119 t[ 4] = h[ 4] ^ m[ 4];
120 t[ 6] = h[ 6] ^ m[ 6];
121 t[ 8] = h[ 8] ^ m[ 8];
122 t[ 9] = h[ 9] ^ m[ 9];
123 t[11] = h[11] ^ m[11];
124 t[12] = h[12] ^ m[12];
125 t[15] = h[15] ^ m[15];
126 t[ 5] = h[ 5] ^ m[ 5];
127 t[10] = h[10] ^ m[10];
128 t[13] = h[13] ^ m[13];
129 t[ 7] = h[ 7] ^ m[ 7];
130 t[14] = h[14] ^ m[14];
134 q[ 0] = (t[ 5] - t[ 7] + t[10] + t[13] + t[14]);
135 q[ 3] = (t[ 0] - t[ 1] + t[ 8] - t[10] + t[13]);
136 q[ 6] = (t[ 4] - t[ 0] - t[ 3] - t[11] + t[13]);
137 q[ 9] = (t[ 0] - t[ 3] + t[ 6] - t[ 7] + t[14]);
138 q[12] = (t[ 1] + t[ 3] - t[ 6] - t[ 9] + t[10]);
139 q[15] = (t[12] - t[ 4] - t[ 6] - t[ 9] + t[13]);
140 q[ 2] = (t[ 0] + t[ 7] + t[ 9] - t[12] + t[15]);
141 q[ 5] = (t[ 3] - t[ 2] + t[10] - t[12] + t[15]);
142 q[ 8] = (t[ 2] - t[ 5] - t[ 6] + t[13] - t[15]);
143 q[11] = (t[ 8] - t[ 0] - t[ 2] - t[ 5] + t[ 9]);
144 q[14] = (t[ 3] - t[ 5] + t[ 8] - t[11] - t[12]);
145 q[ 1] = (t[ 6] - t[ 8] + t[11] + t[14] - t[15]);
146 q[ 4] = (t[ 1] + t[ 2] + t[ 9] - t[11] - t[14]);
147 q[ 7] = (t[ 1] - t[ 4] - t[ 5] - t[12] - t[14]);
148 q[10] = (t[ 8] - t[ 1] - t[ 4] - t[ 7] + t[15]);
149 q[13] = (t[ 2] + t[ 4] + t[ 7] + t[10] + t[11]);
151 tmp = +t[ 5] +t[10] +t[13] +(tr1=-t[ 7]+t[14]) ;
152 q[ 0] = S32_0(tmp) + h[ 1];
153 tmp = +t[ 8] +t[13] +t[ 0] -(tr2=+t[ 1]+t[10]) ;
154 q[ 3] = S32_3(tmp) + h[ 4];
155 tmp = -t[11] +t[13] -t[ 0] -t[ 3] +t[ 4] ;
156 q[ 6] = S32_1(tmp) + h[ 7];
157 tmp = +t[ 0] +(tr0=-t[ 3]+t[ 6]) +(tr1) ;
158 q[ 9] = S32_4(tmp) + h[10];
159 tmp = -t[ 9] -(tr0) +(tr2) ;
160 q[12] = S32_2(tmp) + h[13];
161 tmp = -t[ 4] +(tr0=-t[ 9]+t[12]) +(tr1=-t[ 6]+t[13]) ;
162 q[15] = S32_0(tmp) + h[ 0];
163 tmp = +t[ 7] +t[15] +t[ 0] -(tr0) ;
164 q[ 2] = S32_2(tmp) + h[ 3];
165 tmp = +t[10] +(tr0=-t[ 2]+t[15]) +(tr2=+t[ 3]-t[12]) ;
166 q[ 5] = S32_0(tmp) + h[ 6];
167 tmp = -t[ 5] -(tr0) +(tr1) ;
168 q[ 8] = S32_3(tmp) + h[ 9];
169 tmp = -t[ 0] -t[ 2] +t[ 9] +(tr0=-t[ 5]+t[ 8]) ;
170 q[11] = S32_1(tmp) + h[12];
171 tmp = -t[11] +(tr0) +(tr2) ;
172 q[14] = S32_4(tmp) + h[15];
173 tmp = +t[ 6] +(tr0=+t[11]+t[14]) -(tr1=+t[ 8]+t[15]) ;
174 q[ 1] = S32_1(tmp) + h[ 2];
175 tmp = +t[ 9] +t[ 1] +t[ 2] -(tr0) ;
176 q[ 4] = S32_4(tmp) + h[ 5];
177 tmp = -t[12] -t[14] +t[ 1] -t[ 4] -t[ 5] ;
178 q[ 7] = S32_2(tmp) + h[ 8];
179 tmp = -t[ 1] -(tr0=+t[ 4]+t[ 7]) +(tr1) ;
180 q[10] = S32_0(tmp) + h[11];
181 tmp = +t[ 2] +t[10] +t[11] +(tr0) ;
182 q[13] = S32_3(tmp) + h[14];
188 void bmw_small_f2(uint32_t* h, uint32_t* q, const uint32_t* m){
190 xl = q[16] ^ q[17] ^ q[18] ^ q[19] ^ q[20] ^ q[21] ^ q[22] ^ q[23];
191 xh = xl ^ q[24] ^ q[25] ^ q[26] ^ q[27] ^ q[28] ^ q[29] ^ q[30] ^ q[31];
193 cli_putstr("\r\n XL = ");
194 cli_hexdump_rev(&xl, 4);
195 cli_putstr("\r\n XH = ");
196 cli_hexdump_rev(&xh, 4);
199 h[0] = (SHL32(xh, 5) ^ SHR32(q[16], 5) ^ m[ 0]) + (xl ^ q[24] ^ q[ 0]);
200 h[1] = (SHR32(xh, 7) ^ SHL32(q[17], 8) ^ m[ 1]) + (xl ^ q[25] ^ q[ 1]);
201 h[2] = (SHR32(xh, 5) ^ SHL32(q[18], 5) ^ m[ 2]) + (xl ^ q[26] ^ q[ 2]);
202 h[3] = (SHR32(xh, 1) ^ SHL32(q[19], 5) ^ m[ 3]) + (xl ^ q[27] ^ q[ 3]);
203 h[4] = (SHR32(xh, 3) ^ q[20] ^ m[ 4]) + (xl ^ q[28] ^ q[ 4]);
204 h[5] = (SHL32(xh, 6) ^ SHR32(q[21], 6) ^ m[ 5]) + (xl ^ q[29] ^ q[ 5]);
205 h[6] = (SHR32(xh, 4) ^ SHL32(q[22], 6) ^ m[ 6]) + (xl ^ q[30] ^ q[ 6]);
206 h[7] = (SHR32(xh,11) ^ SHL32(q[23], 2) ^ m[ 7]) + (xl ^ q[31] ^ q[ 7]);
208 h[ 8] = ROTL32(h[4], 9) + (xh ^ q[24] ^ m[ 8]) + (SHL32(xl, 8) ^ q[23] ^ q[ 8]);
209 h[ 9] = ROTL32(h[5], 10) + (xh ^ q[25] ^ m[ 9]) + (SHR32(xl, 6) ^ q[16] ^ q[ 9]);
210 h[10] = ROTL32(h[6], 11) + (xh ^ q[26] ^ m[10]) + (SHL32(xl, 6) ^ q[17] ^ q[10]);
211 h[11] = ROTL32(h[7], 12) + (xh ^ q[27] ^ m[11]) + (SHL32(xl, 4) ^ q[18] ^ q[11]);
212 h[12] = ROTL32(h[0], 13) + (xh ^ q[28] ^ m[12]) + (SHR32(xl, 3) ^ q[19] ^ q[12]);
213 h[13] = ROTL32(h[1], 14) + (xh ^ q[29] ^ m[13]) + (SHR32(xl, 4) ^ q[20] ^ q[13]);
214 h[14] = ROTL32(h[2], 15) + (xh ^ q[30] ^ m[14]) + (SHR32(xl, 7) ^ q[21] ^ q[14]);
215 h[15] = ROTL32(h[3], 16) + (xh ^ q[31] ^ m[15]) + (SHR32(xl, 2) ^ q[22] ^ q[15]);
218 void bmw_small_nextBlock(bmw_small_ctx_t* ctx, const void* block){
220 dump_x(block, 16, 'M');
221 bmw_small_f0(q, ctx->h, block);
223 bmw_small_f1(q, block, ctx->h);
224 dump_x(q+16, 16, 'Q');
225 bmw_small_f2(ctx->h, q, block);
230 void bmw_small_lastBlock(bmw_small_ctx_t* ctx, const void* block, uint16_t length_b){
232 while(length_b >= BMW_SMALL_BLOCKSIZE){
233 bmw_small_nextBlock(ctx, block);
234 length_b -= BMW_SMALL_BLOCKSIZE;
235 block = (uint8_t*)block + BMW_SMALL_BLOCKSIZE_B;
237 memset(buffer, 0, 64);
238 memcpy(buffer, block, (length_b+7)/8);
239 buffer[length_b>>3] |= 0x80 >> (length_b&0x07);
240 if(length_b+1>64*8-64){
241 bmw_small_nextBlock(ctx, buffer);
242 memset(buffer, 0, 64-8);
245 *((uint64_t*)&(buffer[64-8])) = (uint64_t)(ctx->counter*512LL)+(uint64_t)length_b;
246 bmw_small_nextBlock(ctx, buffer);
249 memset(buffer, 0xaa, 64);
251 buffer[i*4] = i+0xa0;
253 // dump_x(buffer, 16, 'A');
254 dump_x(ctx->h, 16, 'M');
255 bmw_small_f0(q, (uint32_t*)buffer, ctx->h);
256 dump_x(buffer, 16, 'a');
258 bmw_small_f1(q, ctx->h, (uint32_t*)buffer);
260 bmw_small_f2((uint32_t*)buffer, q, ctx->h);
261 memcpy(ctx->h, buffer, 64);
264 void bmw224_init(bmw224_ctx_t* ctx){
266 ctx->h[0] = 0x00010203;
268 ctx->h[i] = ctx->h[i-1]+ 0x04040404;
274 void bmw256_init(bmw256_ctx_t* ctx){
276 ctx->h[0] = 0x40414243;
278 ctx->h[i] = ctx->h[i-1]+ 0x04040404;
284 void bmw224_nextBlock(bmw224_ctx_t* ctx, const void* block){
285 bmw_small_nextBlock(ctx, block);
288 void bmw256_nextBlock(bmw256_ctx_t* ctx, const void* block){
289 bmw_small_nextBlock(ctx, block);
292 void bmw224_lastBlock(bmw224_ctx_t* ctx, const void* block, uint16_t length_b){
293 bmw_small_lastBlock(ctx, block, length_b);
296 void bmw256_lastBlock(bmw256_ctx_t* ctx, const void* block, uint16_t length_b){
297 bmw_small_lastBlock(ctx, block, length_b);
300 void bmw224_ctx2hash(void* dest, const bmw224_ctx_t* ctx){
301 memcpy(dest, &(ctx->h[9]), 224/8);
304 void bmw256_ctx2hash(void* dest, const bmw256_ctx_t* ctx){
305 memcpy(dest, &(ctx->h[8]), 256/8);
308 void bmw224(void* dest, const void* msg, uint32_t length_b){
311 while(length_b>=BMW_SMALL_BLOCKSIZE){
312 bmw_small_nextBlock(&ctx, msg);
313 length_b -= BMW_SMALL_BLOCKSIZE;
314 msg = (uint8_t*)msg + BMW_SMALL_BLOCKSIZE_B;
316 bmw_small_lastBlock(&ctx, msg, length_b);
317 bmw224_ctx2hash(dest, &ctx);
320 void bmw256(void* dest, const void* msg, uint32_t length_b){
323 while(length_b>=BMW_SMALL_BLOCKSIZE){
324 bmw_small_nextBlock(&ctx, msg);
325 length_b -= BMW_SMALL_BLOCKSIZE;
326 msg = (uint8_t*)msg + BMW_SMALL_BLOCKSIZE_B;
328 bmw_small_lastBlock(&ctx, msg, length_b);
329 bmw256_ctx2hash(dest, &ctx);