X-Git-Url: https://git.cryptolib.org/?a=blobdiff_plain;f=threefish1024_enc_asm.S;fp=threefish1024_enc_asm.S;h=bb75f04b9b40228354808d306c4a3017f70e029a;hb=f215d8ed649fd5bf5f85eb34f851a1539302013d;hp=0000000000000000000000000000000000000000;hpb=ba4ac1b4a8020026fb6f1a45a1e04e16017262f7;p=avr-crypto-lib.git diff --git a/threefish1024_enc_asm.S b/threefish1024_enc_asm.S new file mode 100644 index 0000000..bb75f04 --- /dev/null +++ b/threefish1024_enc_asm.S @@ -0,0 +1,605 @@ +/* threefish1024_enc_asm.S */ +/* + This file is part of the AVR-Crypto-Lib. + Copyright (C) 2009 Daniel Otte (daniel.otte@rub.de) + + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . +*/ +/* + * \author Daniel Otte + * \email daniel.otte@rub.de + * \date 2009-03-24 + * \license GPLv3 or later + */ + +#include "avr-asm-macros.S" + +/******************************************************************************/ +A0 = 14 +A1 = 15 +A2 = 16 +A3 = 17 +A4 = 18 +A5 = 19 +A6 = 20 +A7 = 21 +/* +#define THREEFISH_KEY_CONST 0x5555.5555.5555.5555.LL / * 2**64/3 * / + +#define K(s) (((uint64_t*)key)[(s)]) +#define T(s) (((uint64_t*)tweak)[(s)]) +void threefish1024_init(const void* key, const void* tweak, threefish512_ctx_t* ctx){ + memcpy(ctx->k, key, 16*8); + memcpy(ctx->t, tweak, 2*8); + uint8_t i; + ctx->k[16] = THREEFISH_KEY_CONST; + for(i=0; i<8; ++i){ + ctx->k[16] ^= K(i); + } + ctx->t[2] = T(0) ^ T(1); +} +*/ +/* + * param key: r24:r25 + * param tweak: r22:r23 + * param ctx: r20:r21 + */ +.global threefish1024_init +threefish1024_init: + push_range 14, 17 + movw r30, r20 + movw r26, r24 + ldi r24, 16 + ldi A7, 0x55 + mov A6, A7 + movw A4, A6 + movw A2, A6 + movw A0, A6 +1: + ld r0, X+ + st Z+, r0 + eor A0, r0 + ld r0, X+ + st Z+, r0 + eor A1, r0 + ld r0, X+ + st Z+, r0 + eor A2, r0 + ld r0, X+ + st Z+, r0 + eor A3, r0 + ld r0, X+ + st Z+, r0 + eor A4, r0 + ld r0, X+ + st Z+, r0 + eor A5, r0 + ld r0, X+ + st Z+, r0 + eor A6, r0 + ld r0, X+ + st Z+, r0 + eor A7, r0 + dec r24 + brne 1b + st Z+, A0 + st Z+, A1 + st Z+, A2 + st Z+, A3 + st Z+, A4 + st Z+, A5 + st Z+, A6 + st Z+, A7 + /* now the tweak */ + movw r26, r22 + ld A0, X+ + ld A1, X+ + ld A2, X+ + ld A3, X+ + ld A4, X+ + ld A5, X+ + ld A6, X+ + ld A7, X+ + st Z+, A0 + st Z+, A1 + st Z+, A2 + st Z+, A3 + st Z+, A4 + st Z+, A5 + st Z+, A6 + st Z+, A7 + ld r0, X+ + eor A0, r0 + st Z+, r0 + ld r0, X+ + eor A1, r0 + st Z+, r0 + ld r0, X+ + eor A2, r0 + st Z+, r0 + ld r0, X+ + eor A3, r0 + st Z+, r0 + ld r0, X+ + eor A4, r0 + st Z+, r0 + ld r0, X+ + eor A5, r0 + st Z+, r0 + ld r0, X+ + eor A6, r0 + st Z+, r0 + ld r0, X+ + eor A7, r0 + st Z+, r0 + st Z+, A0 + st Z+, A1 + st Z+, A2 + st Z+, A3 + st Z+, A4 + st Z+, A5 + st Z+, A6 + st Z+, A7 + pop_range 14, 17 + ret + +/******************************************************************************/ +/* +#define X(a) (((uint64_t*)data)[(a)]) +void permute_16(void* data){ + uint64_t t; + t = X(1); + X(1) = X(9); + X(9) = X(7); + X(7) = X(15); + X(15) = t; + t = X(3); + X(3) = X(13); + X(13) = X(5); + X(5) = X(11); + X(11) = t; + t = X(4); + X(4) = X(6); + X(6) = t; + t = X(8); + X(8) = X(10); + X(10) = X(12); + X(12) = X(14); + X(14) = t; +} +void add_key_16(void* data, const threefish1024_ctx_t* ctx, uint8_t s){ + uint8_t i; + for(i=0; i<13; ++i){ + X(i) += ctx->k[(s+i)%17]; + } + X(13) += ctx->k[(s+13)%17] + ctx->t[s%3]; + X(14) += ctx->k[(s+14)%17] + ctx->t[(s+1)%3]; + X(15) += ctx->k[(s+15)%17] + s; +} +void threefish1024_enc(void* data, const threefish1024_ctx_t* ctx){ + uint8_t i=0,s=0; + uint8_t r0[8] = {55, 25, 33, 34, 28, 17, 58, 47}; + uint8_t r1[8] = {43, 25, 8, 43, 7, 6, 7, 49}; + uint8_t r2[8] = {37, 46, 18, 25, 47, 18, 32, 27}; + uint8_t r3[8] = {40, 13, 57, 60, 48, 25, 45, 58}; + uint8_t r4[8] = {16, 14, 21, 44, 51, 43, 19, 37}; + uint8_t r5[8] = {22, 13, 12, 9, 9, 42, 18, 48}; + uint8_t r6[8] = {38, 52, 32, 59, 35, 40, 2, 53}; + uint8_t r7[8] = {12, 57, 54, 34, 41, 15, 56, 56}; + do{ + if(i%4==0){ + add_key_16(data, ctx, s); + ++s; + } + threefish_mix((uint8_t*)data + 0, r0[i%8]); + threefish_mix((uint8_t*)data + 16, r1[i%8]); + threefish_mix((uint8_t*)data + 32, r2[i%8]); + threefish_mix((uint8_t*)data + 48, r3[i%8]); + threefish_mix((uint8_t*)data + 64, r4[i%8]); + threefish_mix((uint8_t*)data + 80, r5[i%8]); + threefish_mix((uint8_t*)data + 96, r6[i%8]); + threefish_mix((uint8_t*)data +112, r7[i%8]); + permute_16(data); + ++i; + }while(i!=80); + add_key_16(data, ctx, s); +} +*/ +I = 2 +S = 3 +DATA0 = 4 +DATA1 = 5 +CTX0 = 6 +CTX1 = 7 +IDX0 = 8 +IDX1 = 9 +IDX2 = 10 +IDX3 = 11 +IDX4 = 12 +IDX5 = 13 +IDX6 = 14 +IDX7 = 15 + +/* + * param data: r24:r25 + * param ctx: r22:r23 + */ +.global threefish1024_enc +threefish1024_enc: + push r28 + push r29 + push_range 2, 17 + movw DATA0, r24 + movw CTX0, r22 + clr I + clr S +1: + mov r30, I + andi r30, 0x03 + breq 2f + rjmp 4f +2: + ldi r30, lo8(threefish1024_slut17) + ldi r31, hi8(threefish1024_slut17) + add r30, S + adc r31, r1 + lpm IDX0, Z+ + lpm IDX1, Z+ + lpm IDX2, Z+ + lpm IDX3, Z+ + lpm IDX4, Z+ + lpm IDX5, Z+ + lpm IDX6, Z+ + lpm IDX7, Z + movw r30, CTX0 + movw r26, DATA0 + add r30, IDX0 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX1 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX2 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX3 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX4 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX5 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX6 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX7 + adc r31, r1 + rcall add_z_to_x8 + /* second half */ + ldi r30, lo8(threefish1024_slut17) + ldi r31, hi8(threefish1024_slut17) + add r30, S + adc r31, r1 + adiw r30, 8 + lpm IDX0, Z+ + lpm IDX1, Z+ + lpm IDX2, Z+ + lpm IDX3, Z+ + lpm IDX4, Z+ + lpm IDX5, Z+ + lpm IDX6, Z+ + lpm IDX7, Z + movw r30, CTX0 + add r30, IDX0 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX1 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX2 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX3 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX4 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX5 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX6 + adc r31, r1 + rcall add_z_to_x8 + movw r30, CTX0 + add r30, IDX7 + adc r31, r1 + rcall add_z_to_x8 + /* now the remaining key */ + sbiw r26, 3*8 + ldi r30, lo8(threefish1024_slut3) + ldi r31, hi8(threefish1024_slut3) + add r30, S + adc r31, r1 + lpm IDX0, Z+ + lpm IDX1, Z + movw r30, CTX0 + adiw r30, 7*8 /* make Z pointing to (extended tweak) */ + adiw r30, 7*8 + adiw r30, 3*8 + movw IDX2, r30 + add r30, IDX0 + adc r31, r1 + rcall add_z_to_x8 + movw r30, IDX2 + add r30, IDX1 + adc r31, r1 + rcall add_z_to_x8 + ld r0, X + add r0, S + st X+, r0 + ld r0, X + adc r0, r1 + st X+, r0 + ld r0, X + adc r0, r1 + st X+, r0 + ld r0, X + adc r0, r1 + st X+, r0 + ld r0, X + adc r0, r1 + st X+, r0 + ld r0, X + adc r0, r1 + st X+, r0 + ld r0, X + adc r0, r1 + st X+, r0 + ld r0, X + adc r0, r1 + st X+, r0 + inc S + mov r26, S + cpi r26, 21 + brmi 4f +exit: + pop_range 2, 17 + pop r29 + pop r28 + ret +4: + /* call mix */ + ldi r30, lo8(threefish1024_rc0) + ldi r31, hi8(threefish1024_rc0) + mov r26, I + andi r26, 0x07 + add r30, r26 + adc r31, r1 + lpm r22, Z + adiw r30, 8 + lpm IDX0, Z + adiw r30, 8 + lpm IDX1, Z + adiw r30, 8 + lpm IDX2, Z + adiw r30, 8 + lpm IDX3, Z + adiw r30, 8 + lpm IDX4, Z + adiw r30, 8 + lpm IDX5, Z + adiw r30, 8 + lpm IDX6, Z + push IDX6 + push IDX5 + push IDX4 + push IDX3 + push IDX2 + + movw r24, DATA0 + call threefish_mix_asm /* no rcall? */ + movw r24, DATA0 + adiw r24, 16 + mov r22, IDX0 + call threefish_mix_asm /* no rcall? */ + movw r24, DATA0 + adiw r24, 32 + mov r22, IDX1 + call threefish_mix_asm /* no rcall? */ + movw r24, DATA0 + adiw r24, 48 + pop r22 + call threefish_mix_asm /* no rcall? */ + movw r24, DATA0 + adiw r24, 63 + adiw r24, 1 + pop r22 + call threefish_mix_asm /* no rcall? */ + movw r24, DATA0 + adiw r24, 63 + adiw r24, 17 + pop r22 + call threefish_mix_asm /* no rcall? */ + movw r24, DATA0 + adiw r24, 63 + adiw r24, 33 + pop r22 + call threefish_mix_asm /* no rcall? */ + movw r24, DATA0 + adiw r24, 63 + adiw r24, 49 + pop r22 + call threefish_mix_asm /* no rcall? */ + /* now the permutation */ + movw r26, DATA0 /* X1 <-> X15 */ + adiw r26, 1*8 + movw r30, DATA0 + adiw r30, 7*8+4 + adiw r30, 7*8+4 + rcall xchg_zx8 + movw r26, DATA0 /* X1 <-> X9 */ + adiw r26, 1*8 + movw r30, DATA0 + adiw r30, 7*8 + adiw r30, 2*8 + rcall xchg_zx8 + movw r26, DATA0 /* X9 <-> X7 */ + adiw r26, 7*8 + adiw r26, 2*8 + movw r30, DATA0 + adiw r30, 7*8 + rcall xchg_zx8 + /* --- */ + movw r26, DATA0 /* X3 <-> X11 */ + adiw r26, 3*8 + movw r30, DATA0 + adiw r30, 7*8 + adiw r30, 4*8 + rcall xchg_zx8 + movw r26, DATA0 /* X3 <-> X13 */ + adiw r26, 3*8 + movw r30, DATA0 + adiw r30, 7*8 + adiw r30, 6*8 + rcall xchg_zx8 + movw r26, DATA0 /* X13 <-> X5 */ + adiw r26, 7*8 + adiw r26, 6*8 + movw r30, DATA0 + adiw r30, 5*8 + rcall xchg_zx8 + /* --- */ + movw r26, DATA0 /* X8 <-> X14 */ + adiw r26, 7*8 + adiw r26, 1*8 + movw r30, DATA0 + adiw r30, 7*8 + adiw r30, 7*8 + rcall xchg_zx8 + movw r26, DATA0 /* X8 <-> X10 */ + adiw r26, 7*8 + adiw r26, 1*8 + movw r30, DATA0 + adiw r30, 7*8 + adiw r30, 3*8 + rcall xchg_zx8 + movw r26, DATA0 /* X10 <-> X12 */ + adiw r26, 7*8 + adiw r26, 3*8 + movw r30, DATA0 + adiw r30, 7*8 + adiw r30, 5*8 + rcall xchg_zx8 + /* --- */ + movw r26, DATA0 /* X4 <-> X6 */ + adiw r26, 4*8 + movw r30, DATA0 + adiw r30, 6*8 + rcall xchg_zx8 + + inc I +; mov r26, I +; cpi r26, 5 +; brne 9f +; rjmp exit +9: + rjmp 1b + +threefish1024_slut17: + .byte 0x00, 0x08, 0x10, 0x18, 0x20, 0x28, 0x30, 0x38 + .byte 0x40, 0x48, 0x50, 0x58, 0x60, 0x68, 0x70, 0x78 + .byte 0x80, 0x00, 0x08, 0x10, 0x18, 0x20, 0x28, 0x30 + .byte 0x38, 0x40, 0x48, 0x50, 0x58, 0x60, 0x68, 0x70 + .byte 0x78, 0x80, 0x00, 0x08, 0x10 +threefish1024_slut3: + .byte 0x00, 0x08, 0x10, 0x00, 0x08, 0x10, 0x00, 0x08 + .byte 0x10, 0x00, 0x08, 0x10, 0x00, 0x08, 0x10, 0x00 + .byte 0x08, 0x10, 0x00, 0x08, 0x10, 0x00, 0x08, 0x10 + .byte 0x00 +threefish1024_rc0: .byte 0x79, 0x31, 0x41, 0x42, 0x34, 0x21, 0x72, 0x69 +threefish1024_rc1: .byte 0x53, 0x31, 0x10, 0x53, 0x19, 0x1a, 0x19, 0x61 +threefish1024_rc2: .byte 0x5b, 0x6a, 0x22, 0x31, 0x69, 0x22, 0x40, 0x33 +threefish1024_rc3: .byte 0x50, 0x2b, 0x71, 0x74, 0x60, 0x31, 0x6b, 0x72 +threefish1024_rc4: .byte 0x20, 0x2a, 0x3b, 0x54, 0x63, 0x53, 0x23, 0x5b +threefish1024_rc5: .byte 0x3a, 0x2b, 0x14, 0x11, 0x11, 0x52, 0x22, 0x60 +threefish1024_rc6: .byte 0x5a, 0x64, 0x40, 0x73, 0x43, 0x50, 0x02, 0x7b +threefish1024_rc7: .byte 0x14, 0x71, 0x7a, 0x42, 0x51, 0x29, 0x70, 0x70 + +add_z_to_x8: + ld r0, Z+ + ld r1, X + add r1, r0 + st X+, r1 + ld r0, Z+ + ld r1, X + adc r1, r0 + st X+, r1 + ld r0, Z+ + ld r1, X + adc r1, r0 + st X+, r1 + ld r0, Z+ + ld r1, X + adc r1, r0 + st X+, r1 + ld r0, Z+ + ld r1, X + adc r1, r0 + st X+, r1 + ld r0, Z+ + ld r1, X + adc r1, r0 + st X+, r1 + ld r0, Z+ + ld r1, X + adc r1, r0 + st X+, r1 + ld r0, Z+ + ld r1, X + adc r1, r0 + st X+, r1 + clr r1 + ret + +T0 = IDX0 +T1 = 0 +CNT = 24 +xchg_zx8: + ldi CNT, 8 +1: ld T0, X + ld T1, Z + st X+, T1 + st Z+, T0 + dec CNT + brne 1b + ret + + +