You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
1174 lines
22 KiB
1174 lines
22 KiB
9 years ago
|
/*
|
||
|
|
* Copyright 2012 pooler@litecoinpool.org
|
||
|
|
*
|
||
|
|
* 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 2 of the License, or (at your option)
|
||
|
|
* any later version. See COPYING for more details.
|
||
|
|
*/
|
||
|
|
|
||
|
|
#include "cpuminer-config.h"
|
||
|
|
|
||
|
|
#if defined(__arm__) && defined(__APCS_32__)
|
||
|
|
|
||
|
|
#if defined(__ARM_ARCH_5E__) || defined(__ARM_ARCH_5TE__) || \
|
||
|
|
defined(__ARM_ARCH_5TEJ__) || defined(__ARM_ARCH_6__) || \
|
||
|
|
defined(__ARM_ARCH_6J__) || defined(__ARM_ARCH_6K__) || \
|
||
|
|
defined(__ARM_ARCH_6M__) || defined(__ARM_ARCH_6T2__) || \
|
||
|
|
defined(__ARM_ARCH_6Z__) || defined(__ARM_ARCH_6ZK__)
|
||
|
|
#define __ARM_ARCH_5E_OR_6__
|
||
|
|
#endif
|
||
|
|
|
||
|
|
#if defined(__ARM_ARCH_5E_OR_6__) || defined(__ARM_ARCH_7__) || \
|
||
|
|
defined(__ARM_ARCH_7A__) || defined(__ARM_ARCH_7R__) || \
|
||
|
|
defined(__ARM_ARCH_7M__) || defined(__ARM_ARCH_7EM__)
|
||
|
|
#define __ARM_ARCH_5E_OR_6_OR_7__
|
||
|
|
#endif
|
||
|
|
|
||
|
|
#ifdef __ARM_ARCH_5E_OR_6__
|
||
|
|
|
||
|
|
.macro scrypt_shuffle
|
||
|
|
add lr, r0, #9*4
|
||
|
|
ldmia r0, {r2-r7}
|
||
|
|
ldmia lr, {r2, r8-r12, lr}
|
||
|
|
str r3, [r0, #5*4]
|
||
|
|
str r5, [r0, #15*4]
|
||
|
|
str r6, [r0, #12*4]
|
||
|
|
str r7, [r0, #1*4]
|
||
|
|
ldr r5, [r0, #7*4]
|
||
|
|
str r2, [r0, #13*4]
|
||
|
|
str r8, [r0, #2*4]
|
||
|
|
strd r4, [r0, #10*4]
|
||
|
|
str r9, [r0, #7*4]
|
||
|
|
str r10, [r0, #4*4]
|
||
|
|
str r11, [r0, #9*4]
|
||
|
|
str lr, [r0, #3*4]
|
||
|
|
|
||
|
|
add r2, r0, #64+0*4
|
||
|
|
add lr, r0, #64+9*4
|
||
|
|
ldmia r2, {r2-r7}
|
||
|
|
ldmia lr, {r2, r8-r12, lr}
|
||
|
|
str r3, [r0, #64+5*4]
|
||
|
|
str r5, [r0, #64+15*4]
|
||
|
|
str r6, [r0, #64+12*4]
|
||
|
|
str r7, [r0, #64+1*4]
|
||
|
|
ldr r5, [r0, #64+7*4]
|
||
|
|
str r2, [r0, #64+13*4]
|
||
|
|
str r8, [r0, #64+2*4]
|
||
|
|
strd r4, [r0, #64+10*4]
|
||
|
|
str r9, [r0, #64+7*4]
|
||
|
|
str r10, [r0, #64+4*4]
|
||
|
|
str r11, [r0, #64+9*4]
|
||
|
|
str lr, [r0, #64+3*4]
|
||
|
|
.endm
|
||
|
|
|
||
|
|
.macro salsa8_core_doubleround_body
|
||
|
|
add r6, r2, r6
|
||
|
|
add r7, r3, r7
|
||
|
|
eor r10, r10, r6, ror #25
|
||
|
|
add r6, r0, r4
|
||
|
|
eor r11, r11, r7, ror #25
|
||
|
|
add r7, r1, r5
|
||
|
|
strd r10, [sp, #14*4]
|
||
|
|
eor r12, r12, r6, ror #25
|
||
|
|
eor lr, lr, r7, ror #25
|
||
|
|
|
||
|
|
ldrd r6, [sp, #10*4]
|
||
|
|
add r2, r10, r2
|
||
|
|
add r3, r11, r3
|
||
|
|
eor r6, r6, r2, ror #23
|
||
|
|
add r2, r12, r0
|
||
|
|
eor r7, r7, r3, ror #23
|
||
|
|
add r3, lr, r1
|
||
|
|
strd r6, [sp, #10*4]
|
||
|
|
eor r8, r8, r2, ror #23
|
||
|
|
eor r9, r9, r3, ror #23
|
||
|
|
|
||
|
|
ldrd r2, [sp, #6*4]
|
||
|
|
add r10, r6, r10
|
||
|
|
add r11, r7, r11
|
||
|
|
eor r2, r2, r10, ror #19
|
||
|
|
add r10, r8, r12
|
||
|
|
eor r3, r3, r11, ror #19
|
||
|
|
add r11, r9, lr
|
||
|
|
eor r4, r4, r10, ror #19
|
||
|
|
eor r5, r5, r11, ror #19
|
||
|
|
|
||
|
|
ldrd r10, [sp, #2*4]
|
||
|
|
add r6, r2, r6
|
||
|
|
add r7, r3, r7
|
||
|
|
eor r10, r10, r6, ror #14
|
||
|
|
add r6, r4, r8
|
||
|
|
eor r11, r11, r7, ror #14
|
||
|
|
add r7, r5, r9
|
||
|
|
eor r0, r0, r6, ror #14
|
||
|
|
eor r1, r1, r7, ror #14
|
||
|
|
|
||
|
|
|
||
|
|
ldrd r6, [sp, #14*4]
|
||
|
|
strd r2, [sp, #6*4]
|
||
|
|
strd r10, [sp, #2*4]
|
||
|
|
add r6, r11, r6
|
||
|
|
add r7, r0, r7
|
||
|
|
eor r4, r4, r6, ror #25
|
||
|
|
add r6, r1, r12
|
||
|
|
eor r5, r5, r7, ror #25
|
||
|
|
add r7, r10, lr
|
||
|
|
eor r2, r2, r6, ror #25
|
||
|
|
eor r3, r3, r7, ror #25
|
||
|
|
strd r2, [sp, #6*4]
|
||
|
|
|
||
|
|
add r10, r3, r10
|
||
|
|
ldrd r6, [sp, #10*4]
|
||
|
|
add r11, r4, r11
|
||
|
|
eor r8, r8, r10, ror #23
|
||
|
|
add r10, r5, r0
|
||
|
|
eor r9, r9, r11, ror #23
|
||
|
|
add r11, r2, r1
|
||
|
|
eor r6, r6, r10, ror #23
|
||
|
|
eor r7, r7, r11, ror #23
|
||
|
|
strd r6, [sp, #10*4]
|
||
|
|
|
||
|
|
add r2, r7, r2
|
||
|
|
ldrd r10, [sp, #14*4]
|
||
|
|
add r3, r8, r3
|
||
|
|
eor r12, r12, r2, ror #19
|
||
|
|
add r2, r9, r4
|
||
|
|
eor lr, lr, r3, ror #19
|
||
|
|
add r3, r6, r5
|
||
|
|
eor r10, r10, r2, ror #19
|
||
|
|
eor r11, r11, r3, ror #19
|
||
|
|
|
||
|
|
ldrd r2, [sp, #2*4]
|
||
|
|
add r6, r11, r6
|
||
|
|
add r7, r12, r7
|
||
|
|
eor r0, r0, r6, ror #14
|
||
|
|
add r6, lr, r8
|
||
|
|
eor r1, r1, r7, ror #14
|
||
|
|
add r7, r10, r9
|
||
|
|
eor r2, r2, r6, ror #14
|
||
|
|
eor r3, r3, r7, ror #14
|
||
|
|
.endm
|
||
|
|
|
||
|
|
.macro salsa8_core
|
||
|
|
ldmia sp, {r0-r12, lr}
|
||
|
|
|
||
|
|
ldrd r10, [sp, #14*4]
|
||
|
|
salsa8_core_doubleround_body
|
||
|
|
ldrd r6, [sp, #6*4]
|
||
|
|
strd r2, [sp, #2*4]
|
||
|
|
strd r10, [sp, #14*4]
|
||
|
|
salsa8_core_doubleround_body
|
||
|
|
ldrd r6, [sp, #6*4]
|
||
|
|
strd r2, [sp, #2*4]
|
||
|
|
strd r10, [sp, #14*4]
|
||
|
|
salsa8_core_doubleround_body
|
||
|
|
ldrd r6, [sp, #6*4]
|
||
|
|
strd r2, [sp, #2*4]
|
||
|
|
strd r10, [sp, #14*4]
|
||
|
|
salsa8_core_doubleround_body
|
||
|
|
|
||
|
|
stmia sp, {r0-r5}
|
||
|
|
strd r8, [sp, #8*4]
|
||
|
|
str r12, [sp, #12*4]
|
||
|
|
str lr, [sp, #13*4]
|
||
|
|
strd r10, [sp, #14*4]
|
||
|
|
.endm
|
||
|
|
|
||
|
|
#else
|
||
|
|
|
||
|
|
.macro scrypt_shuffle
|
||
|
|
.endm
|
||
|
|
|
||
|
|
.macro salsa8_core_doubleround_body
|
||
|
|
ldr r8, [sp, #8*4]
|
||
|
|
add r11, r11, r10
|
||
|
|
ldr lr, [sp, #13*4]
|
||
|
|
add r12, r12, r3
|
||
|
|
eor r2, r2, r11, ror #23
|
||
|
|
add r11, r4, r0
|
||
|
|
eor r7, r7, r12, ror #23
|
||
|
|
add r12, r9, r5
|
||
|
|
str r9, [sp, #9*4]
|
||
|
|
eor r8, r8, r11, ror #23
|
||
|
|
str r10, [sp, #14*4]
|
||
|
|
eor lr, lr, r12, ror #23
|
||
|
|
|
||
|
|
ldr r11, [sp, #11*4]
|
||
|
|
add r9, lr, r9
|
||
|
|
ldr r12, [sp, #12*4]
|
||
|
|
add r10, r2, r10
|
||
|
|
eor r1, r1, r9, ror #19
|
||
|
|
add r9, r7, r3
|
||
|
|
eor r6, r6, r10, ror #19
|
||
|
|
add r10, r8, r4
|
||
|
|
str r8, [sp, #8*4]
|
||
|
|
eor r11, r11, r9, ror #19
|
||
|
|
str lr, [sp, #13*4]
|
||
|
|
eor r12, r12, r10, ror #19
|
||
|
|
|
||
|
|
ldr r9, [sp, #10*4]
|
||
|
|
add r8, r12, r8
|
||
|
|
ldr r10, [sp, #15*4]
|
||
|
|
add lr, r1, lr
|
||
|
|
eor r0, r0, r8, ror #14
|
||
|
|
add r8, r6, r2
|
||
|
|
eor r5, r5, lr, ror #14
|
||
|
|
add lr, r11, r7
|
||
|
|
eor r9, r9, r8, ror #14
|
||
|
|
ldr r8, [sp, #9*4]
|
||
|
|
eor r10, r10, lr, ror #14
|
||
|
|
ldr lr, [sp, #14*4]
|
||
|
|
|
||
|
|
|
||
|
|
add r8, r9, r8
|
||
|
|
str r9, [sp, #10*4]
|
||
|
|
add lr, r10, lr
|
||
|
|
str r10, [sp, #15*4]
|
||
|
|
eor r11, r11, r8, ror #25
|
||
|
|
add r8, r0, r3
|
||
|
|
eor r12, r12, lr, ror #25
|
||
|
|
add lr, r5, r4
|
||
|
|
eor r1, r1, r8, ror #25
|
||
|
|
ldr r8, [sp, #8*4]
|
||
|
|
eor r6, r6, lr, ror #25
|
||
|
|
|
||
|
|
add r9, r11, r9
|
||
|
|
ldr lr, [sp, #13*4]
|
||
|
|
add r10, r12, r10
|
||
|
|
eor r8, r8, r9, ror #23
|
||
|
|
add r9, r1, r0
|
||
|
|
eor lr, lr, r10, ror #23
|
||
|
|
add r10, r6, r5
|
||
|
|
str r11, [sp, #11*4]
|
||
|
|
eor r2, r2, r9, ror #23
|
||
|
|
str r12, [sp, #12*4]
|
||
|
|
eor r7, r7, r10, ror #23
|
||
|
|
|
||
|
|
ldr r9, [sp, #9*4]
|
||
|
|
add r11, r8, r11
|
||
|
|
ldr r10, [sp, #14*4]
|
||
|
|
add r12, lr, r12
|
||
|
|
eor r9, r9, r11, ror #19
|
||
|
|
add r11, r2, r1
|
||
|
|
eor r10, r10, r12, ror #19
|
||
|
|
add r12, r7, r6
|
||
|
|
str r8, [sp, #8*4]
|
||
|
|
eor r3, r3, r11, ror #19
|
||
|
|
str lr, [sp, #13*4]
|
||
|
|
eor r4, r4, r12, ror #19
|
||
|
|
.endm
|
||
|
|
|
||
|
|
.macro salsa8_core
|
||
|
|
ldmia sp, {r0-r7}
|
||
|
|
|
||
|
|
ldr r12, [sp, #15*4]
|
||
|
|
ldr r8, [sp, #11*4]
|
||
|
|
ldr lr, [sp, #12*4]
|
||
|
|
|
||
|
|
ldr r9, [sp, #9*4]
|
||
|
|
add r8, r8, r12
|
||
|
|
ldr r11, [sp, #10*4]
|
||
|
|
add lr, lr, r0
|
||
|
|
eor r3, r3, r8, ror #25
|
||
|
|
add r8, r5, r1
|
||
|
|
ldr r10, [sp, #14*4]
|
||
|
|
eor r4, r4, lr, ror #25
|
||
|
|
add lr, r11, r6
|
||
|
|
eor r9, r9, r8, ror #25
|
||
|
|
eor r10, r10, lr, ror #25
|
||
|
|
|
||
|
|
salsa8_core_doubleround_body
|
||
|
|
|
||
|
|
ldr r11, [sp, #10*4]
|
||
|
|
add r8, r9, r8
|
||
|
|
ldr r12, [sp, #15*4]
|
||
|
|
add lr, r10, lr
|
||
|
|
eor r11, r11, r8, ror #14
|
||
|
|
add r8, r3, r2
|
||
|
|
eor r12, r12, lr, ror #14
|
||
|
|
add lr, r4, r7
|
||
|
|
eor r0, r0, r8, ror #14
|
||
|
|
ldr r8, [sp, #11*4]
|
||
|
|
eor r5, r5, lr, ror #14
|
||
|
|
ldr lr, [sp, #12*4]
|
||
|
|
|
||
|
|
add r8, r8, r12
|
||
|
|
str r11, [sp, #10*4]
|
||
|
|
add lr, lr, r0
|
||
|
|
str r12, [sp, #15*4]
|
||
|
|
eor r3, r3, r8, ror #25
|
||
|
|
add r8, r5, r1
|
||
|
|
eor r4, r4, lr, ror #25
|
||
|
|
add lr, r11, r6
|
||
|
|
str r9, [sp, #9*4]
|
||
|
|
eor r9, r9, r8, ror #25
|
||
|
|
str r10, [sp, #14*4]
|
||
|
|
eor r10, r10, lr, ror #25
|
||
|
|
|
||
|
|
salsa8_core_doubleround_body
|
||
|
|
|
||
|
|
ldr r11, [sp, #10*4]
|
||
|
|
add r8, r9, r8
|
||
|
|
ldr r12, [sp, #15*4]
|
||
|
|
add lr, r10, lr
|
||
|
|
eor r11, r11, r8, ror #14
|
||
|
|
add r8, r3, r2
|
||
|
|
eor r12, r12, lr, ror #14
|
||
|
|
add lr, r4, r7
|
||
|
|
eor r0, r0, r8, ror #14
|
||
|
|
ldr r8, [sp, #11*4]
|
||
|
|
eor r5, r5, lr, ror #14
|
||
|
|
ldr lr, [sp, #12*4]
|
||
|
|
|
||
|
|
add r8, r8, r12
|
||
|
|
str r11, [sp, #10*4]
|
||
|
|
add lr, lr, r0
|
||
|
|
str r12, [sp, #15*4]
|
||
|
|
eor r3, r3, r8, ror #25
|
||
|
|
add r8, r5, r1
|
||
|
|
eor r4, r4, lr, ror #25
|
||
|
|
add lr, r11, r6
|
||
|
|
str r9, [sp, #9*4]
|
||
|
|
eor r9, r9, r8, ror #25
|
||
|
|
str r10, [sp, #14*4]
|
||
|
|
eor r10, r10, lr, ror #25
|
||
|
|
|
||
|
|
salsa8_core_doubleround_body
|
||
|
|
|
||
|
|
ldr r11, [sp, #10*4]
|
||
|
|
add r8, r9, r8
|
||
|
|
ldr r12, [sp, #15*4]
|
||
|
|
add lr, r10, lr
|
||
|
|
eor r11, r11, r8, ror #14
|
||
|
|
add r8, r3, r2
|
||
|
|
eor r12, r12, lr, ror #14
|
||
|
|
add lr, r4, r7
|
||
|
|
eor r0, r0, r8, ror #14
|
||
|
|
ldr r8, [sp, #11*4]
|
||
|
|
eor r5, r5, lr, ror #14
|
||
|
|
ldr lr, [sp, #12*4]
|
||
|
|
|
||
|
|
add r8, r8, r12
|
||
|
|
str r11, [sp, #10*4]
|
||
|
|
add lr, lr, r0
|
||
|
|
str r12, [sp, #15*4]
|
||
|
|
eor r3, r3, r8, ror #25
|
||
|
|
add r8, r5, r1
|
||
|
|
eor r4, r4, lr, ror #25
|
||
|
|
add lr, r11, r6
|
||
|
|
str r9, [sp, #9*4]
|
||
|
|
eor r9, r9, r8, ror #25
|
||
|
|
str r10, [sp, #14*4]
|
||
|
|
eor r10, r10, lr, ror #25
|
||
|
|
|
||
|
|
salsa8_core_doubleround_body
|
||
|
|
|
||
|
|
ldr r11, [sp, #10*4]
|
||
|
|
add r8, r9, r8
|
||
|
|
ldr r12, [sp, #15*4]
|
||
|
|
add lr, r10, lr
|
||
|
|
str r9, [sp, #9*4]
|
||
|
|
eor r11, r11, r8, ror #14
|
||
|
|
eor r12, r12, lr, ror #14
|
||
|
|
add r8, r3, r2
|
||
|
|
str r10, [sp, #14*4]
|
||
|
|
add lr, r4, r7
|
||
|
|
str r11, [sp, #10*4]
|
||
|
|
eor r0, r0, r8, ror #14
|
||
|
|
str r12, [sp, #15*4]
|
||
|
|
eor r5, r5, lr, ror #14
|
||
|
|
|
||
|
|
stmia sp, {r0-r7}
|
||
|
|
.endm
|
||
|
|
|
||
|
|
#endif
|
||
|
|
|
||
|
|
|
||
|
|
.macro scrypt_core_macro1a_x4
|
||
|
|
ldmia r0, {r4-r7}
|
||
|
|
ldmia lr!, {r8-r11}
|
||
|
|
stmia r1!, {r4-r7}
|
||
|
|
stmia r3!, {r8-r11}
|
||
|
|
eor r4, r4, r8
|
||
|
|
eor r5, r5, r9
|
||
|
|
eor r6, r6, r10
|
||
|
|
eor r7, r7, r11
|
||
|
|
stmia r0!, {r4-r7}
|
||
|
|
stmia r12!, {r4-r7}
|
||
|
|
.endm
|
||
|
|
|
||
|
|
.macro scrypt_core_macro1b_x4
|
||
|
|
ldmia r3!, {r8-r11}
|
||
|
|
ldmia r2, {r4-r7}
|
||
|
|
eor r8, r8, r4
|
||
|
|
eor r9, r9, r5
|
||
|
|
eor r10, r10, r6
|
||
|
|
eor r11, r11, r7
|
||
|
|
ldmia r0, {r4-r7}
|
||
|
|
stmia r2!, {r8-r11}
|
||
|
|
eor r4, r4, r8
|
||
|
|
eor r5, r5, r9
|
||
|
|
eor r6, r6, r10
|
||
|
|
eor r7, r7, r11
|
||
|
|
ldmia r1!, {r8-r11}
|
||
|
|
eor r4, r4, r8
|
||
|
|
eor r5, r5, r9
|
||
|
|
eor r6, r6, r10
|
||
|
|
eor r7, r7, r11
|
||
|
|
stmia r0!, {r4-r7}
|
||
|
|
stmia r12!, {r4-r7}
|
||
|
|
.endm
|
||
|
|
|
||
|
|
.macro scrypt_core_macro2_x4
|
||
|
|
ldmia r12, {r4-r7}
|
||
|
|
ldmia r0, {r8-r11}
|
||
|
|
add r4, r4, r8
|
||
|
|
add r5, r5, r9
|
||
|
|
add r6, r6, r10
|
||
|
|
add r7, r7, r11
|
||
|
|
stmia r0!, {r4-r7}
|
||
|
|
ldmia r2, {r8-r11}
|
||
|
|
eor r4, r4, r8
|
||
|
|
eor r5, r5, r9
|
||
|
|
eor r6, r6, r10
|
||
|
|
eor r7, r7, r11
|
||
|
|
stmia r2!, {r4-r7}
|
||
|
|
stmia r12!, {r4-r7}
|
||
|
|
.endm
|
||
|
|
|
||
|
|
.macro scrypt_core_macro3_x4
|
||
|
|
ldmia r1!, {r4-r7}
|
||
|
|
ldmia r0, {r8-r11}
|
||
|
|
add r4, r4, r8
|
||
|
|
add r5, r5, r9
|
||
|
|
add r6, r6, r10
|
||
|
|
add r7, r7, r11
|
||
|
|
stmia r0!, {r4-r7}
|
||
|
|
.endm
|
||
|
|
|
||
|
|
.macro scrypt_core_macro3_x6
|
||
|
|
ldmia r1!, {r2-r7}
|
||
|
|
ldmia r0, {r8-r12, lr}
|
||
|
|
add r2, r2, r8
|
||
|
|
add r3, r3, r9
|
||
|
|
add r4, r4, r10
|
||
|
|
add r5, r5, r11
|
||
|
|
add r6, r6, r12
|
||
|
|
add r7, r7, lr
|
||
|
|
stmia r0!, {r2-r7}
|
||
|
|
.endm
|
||
|
|
|
||
|
|
|
||
|
|
.text
|
||
|
|
.code 32
|
||
|
|
.align 2
|
||
|
|
.globl scrypt_core
|
||
|
|
.globl _scrypt_core
|
||
|
|
#ifdef __ELF__
|
||
|
|
.type scrypt_core, %function
|
||
|
|
#endif
|
||
|
|
scrypt_core:
|
||
|
|
_scrypt_core:
|
||
|
|
stmfd sp!, {r4-r11, lr}
|
||
|
|
mov r12, sp
|
||
|
|
sub sp, sp, #21*4
|
||
|
|
bic sp, sp, #63
|
||
|
|
str r12, [sp, #20*4]
|
||
|
|
|
||
|
|
scrypt_shuffle
|
||
|
|
|
||
|
|
str r0, [sp, #16*4]
|
||
|
|
add r12, r1, #1024*32*4
|
||
|
|
str r12, [sp, #18*4]
|
||
|
|
scrypt_core_loop1:
|
||
|
|
add lr, r0, #16*4
|
||
|
|
add r3, r1, #16*4
|
||
|
|
mov r12, sp
|
||
|
|
scrypt_core_macro1a_x4
|
||
|
|
scrypt_core_macro1a_x4
|
||
|
|
scrypt_core_macro1a_x4
|
||
|
|
scrypt_core_macro1a_x4
|
||
|
|
str r1, [sp, #17*4]
|
||
|
|
|
||
|
|
salsa8_core
|
||
|
|
|
||
|
|
ldr r0, [sp, #16*4]
|
||
|
|
mov r12, sp
|
||
|
|
add r2, r0, #16*4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
|
||
|
|
salsa8_core
|
||
|
|
|
||
|
|
ldr r0, [sp, #16*4]
|
||
|
|
mov r1, sp
|
||
|
|
add r0, r0, #16*4
|
||
|
|
scrypt_core_macro3_x6
|
||
|
|
scrypt_core_macro3_x6
|
||
|
|
ldr r3, [sp, #17*4]
|
||
|
|
ldr r12, [sp, #18*4]
|
||
|
|
scrypt_core_macro3_x4
|
||
|
|
|
||
|
|
add r1, r3, #16*4
|
||
|
|
sub r0, r0, #32*4
|
||
|
|
cmp r1, r12
|
||
|
|
bne scrypt_core_loop1
|
||
|
|
|
||
|
|
ldr r4, [r0, #16*4]
|
||
|
|
sub r1, r1, #1024*32*4
|
||
|
|
str r1, [sp, #17*4]
|
||
|
|
mov r4, r4, lsl #32-10
|
||
|
|
mov r12, #1024
|
||
|
|
add r1, r1, r4, lsr #32-10-7
|
||
|
|
scrypt_core_loop2:
|
||
|
|
add r2, r0, #16*4
|
||
|
|
add r3, r1, #16*4
|
||
|
|
str r12, [sp, #18*4]
|
||
|
|
mov r12, sp
|
||
|
|
#ifdef __ARM_ARCH_5E_OR_6_OR_7__
|
||
|
|
pld [r1, #24*4]
|
||
|
|
pld [r1, #8*4]
|
||
|
|
#endif
|
||
|
|
scrypt_core_macro1b_x4
|
||
|
|
scrypt_core_macro1b_x4
|
||
|
|
scrypt_core_macro1b_x4
|
||
|
|
scrypt_core_macro1b_x4
|
||
|
|
|
||
|
|
salsa8_core
|
||
|
|
|
||
|
|
ldr r0, [sp, #16*4]
|
||
|
|
mov r12, sp
|
||
|
|
add r2, r0, #16*4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
|
||
|
|
salsa8_core
|
||
|
|
|
||
|
|
ldr r0, [sp, #16*4]
|
||
|
|
mov r1, sp
|
||
|
|
ldr r3, [sp, #17*4]
|
||
|
|
add r0, r0, #16*4
|
||
|
|
scrypt_core_macro3_x4
|
||
|
|
mov r4, r4, lsl #32-10
|
||
|
|
add r3, r3, r4, lsr #32-10-7
|
||
|
|
str r3, [sp, #19*4]
|
||
|
|
#ifdef __ARM_ARCH_5E_OR_6_OR_7__
|
||
|
|
pld [r3, #16*4]
|
||
|
|
pld [r3]
|
||
|
|
#endif
|
||
|
|
scrypt_core_macro3_x6
|
||
|
|
scrypt_core_macro3_x6
|
||
|
|
|
||
|
|
ldr r12, [sp, #18*4]
|
||
|
|
sub r0, r0, #32*4
|
||
|
|
ldr r1, [sp, #19*4]
|
||
|
|
subs r12, r12, #1
|
||
|
|
bne scrypt_core_loop2
|
||
|
|
|
||
|
|
scrypt_shuffle
|
||
|
|
|
||
|
|
ldr sp, [sp, #20*4]
|
||
|
|
#ifdef __thumb__
|
||
|
|
ldmfd sp!, {r4-r11, lr}
|
||
|
|
bx lr
|
||
|
|
#else
|
||
|
|
ldmfd sp!, {r4-r11, pc}
|
||
|
|
#endif
|
||
|
|
|
||
|
|
|
||
|
|
#ifdef __ARM_NEON__
|
||
|
|
|
||
|
|
.macro salsa8_core_3way_doubleround
|
||
|
|
ldrd r6, [sp, #6*4]
|
||
|
|
vadd.u32 q4, q0, q1
|
||
|
|
add r6, r2, r6
|
||
|
|
vadd.u32 q6, q8, q9
|
||
|
|
add r7, r3, r7
|
||
|
|
vshl.u32 q5, q4, #7
|
||
|
|
eor r10, r10, r6, ror #25
|
||
|
|
vshl.u32 q7, q6, #7
|
||
|
|
add r6, r0, r4
|
||
|
|
vshr.u32 q4, q4, #32-7
|
||
|
|
eor r11, r11, r7, ror #25
|
||
|
|
vshr.u32 q6, q6, #32-7
|
||
|
|
add r7, r1, r5
|
||
|
|
veor.u32 q3, q3, q5
|
||
|
|
strd r10, [sp, #14*4]
|
||
|
|
veor.u32 q11, q11, q7
|
||
|
|
eor r12, r12, r6, ror #25
|
||
|
|
veor.u32 q3, q3, q4
|
||
|
|
eor lr, lr, r7, ror #25
|
||
|
|
veor.u32 q11, q11, q6
|
||
|
|
|
||
|
|
ldrd r6, [sp, #10*4]
|
||
|
|
vadd.u32 q4, q3, q0
|
||
|
|
add r2, r10, r2
|
||
|
|
vadd.u32 q6, q11, q8
|
||
|
|
add r3, r11, r3
|
||
|
|
vshl.u32 q5, q4, #9
|
||
|
|
eor r6, r6, r2, ror #23
|
||
|
|
vshl.u32 q7, q6, #9
|
||
|
|
add r2, r12, r0
|
||
|
|
vshr.u32 q4, q4, #32-9
|
||
|
|
eor r7, r7, r3, ror #23
|
||
|
|
vshr.u32 q6, q6, #32-9
|
||
|
|
add r3, lr, r1
|
||
|
|
veor.u32 q2, q2, q5
|
||
|
|
strd r6, [sp, #10*4]
|
||
|
|
veor.u32 q10, q10, q7
|
||
|
|
eor r8, r8, r2, ror #23
|
||
|
|
veor.u32 q2, q2, q4
|
||
|
|
eor r9, r9, r3, ror #23
|
||
|
|
veor.u32 q10, q10, q6
|
||
|
|
|
||
|
|
ldrd r2, [sp, #6*4]
|
||
|
|
vadd.u32 q4, q2, q3
|
||
|
|
add r10, r6, r10
|
||
|
|
vadd.u32 q6, q10, q11
|
||
|
|
add r11, r7, r11
|
||
|
|
vext.u32 q3, q3, q3, #3
|
||
|
|
eor r2, r2, r10, ror #19
|
||
|
|
vshl.u32 q5, q4, #13
|
||
|
|
add r10, r8, r12
|
||
|
|
vext.u32 q11, q11, q11, #3
|
||
|
|
eor r3, r3, r11, ror #19
|
||
|
|
vshl.u32 q7, q6, #13
|
||
|
|
add r11, r9, lr
|
||
|
|
vshr.u32 q4, q4, #32-13
|
||
|
|
eor r4, r4, r10, ror #19
|
||
|
|
vshr.u32 q6, q6, #32-13
|
||
|
|
eor r5, r5, r11, ror #19
|
||
|
|
veor.u32 q1, q1, q5
|
||
|
|
veor.u32 q9, q9, q7
|
||
|
|
veor.u32 q1, q1, q4
|
||
|
|
veor.u32 q9, q9, q6
|
||
|
|
|
||
|
|
ldrd r10, [sp, #2*4]
|
||
|
|
vadd.u32 q4, q1, q2
|
||
|
|
add r6, r2, r6
|
||
|
|
vadd.u32 q6, q9, q10
|
||
|
|
add r7, r3, r7
|
||
|
|
vswp.u32 d4, d5
|
||
|
|
eor r10, r10, r6, ror #14
|
||
|
|
vshl.u32 q5, q4, #18
|
||
|
|
add r6, r4, r8
|
||
|
|
vswp.u32 d20, d21
|
||
|
|
eor r11, r11, r7, ror #14
|
||
|
|
vshl.u32 q7, q6, #18
|
||
|
|
add r7, r5, r9
|
||
|
|
vshr.u32 q4, q4, #32-18
|
||
|
|
eor r0, r0, r6, ror #14
|
||
|
|
vshr.u32 q6, q6, #32-18
|
||
|
|
eor r1, r1, r7, ror #14
|
||
|
|
veor.u32 q0, q0, q5
|
||
|
|
ldrd r6, [sp, #14*4]
|
||
|
|
veor.u32 q8, q8, q7
|
||
|
|
veor.u32 q0, q0, q4
|
||
|
|
veor.u32 q8, q8, q6
|
||
|
|
|
||
|
|
|
||
|
|
strd r2, [sp, #6*4]
|
||
|
|
vadd.u32 q4, q0, q3
|
||
|
|
strd r10, [sp, #2*4]
|
||
|
|
vadd.u32 q6, q8, q11
|
||
|
|
add r6, r11, r6
|
||
|
|
vext.u32 q1, q1, q1, #1
|
||
|
|
add r7, r0, r7
|
||
|
|
vshl.u32 q5, q4, #7
|
||
|
|
eor r4, r4, r6, ror #25
|
||
|
|
vext.u32 q9, q9, q9, #1
|
||
|
|
add r6, r1, r12
|
||
|
|
vshl.u32 q7, q6, #7
|
||
|
|
eor r5, r5, r7, ror #25
|
||
|
|
vshr.u32 q4, q4, #32-7
|
||
|
|
add r7, r10, lr
|
||
|
|
vshr.u32 q6, q6, #32-7
|
||
|
|
eor r2, r2, r6, ror #25
|
||
|
|
veor.u32 q1, q1, q5
|
||
|
|
eor r3, r3, r7, ror #25
|
||
|
|
veor.u32 q9, q9, q7
|
||
|
|
strd r2, [sp, #6*4]
|
||
|
|
veor.u32 q1, q1, q4
|
||
|
|
veor.u32 q9, q9, q6
|
||
|
|
|
||
|
|
add r10, r3, r10
|
||
|
|
vadd.u32 q4, q1, q0
|
||
|
|
ldrd r6, [sp, #10*4]
|
||
|
|
vadd.u32 q6, q9, q8
|
||
|
|
add r11, r4, r11
|
||
|
|
vshl.u32 q5, q4, #9
|
||
|
|
eor r8, r8, r10, ror #23
|
||
|
|
vshl.u32 q7, q6, #9
|
||
|
|
add r10, r5, r0
|
||
|
|
vshr.u32 q4, q4, #32-9
|
||
|
|
eor r9, r9, r11, ror #23
|
||
|
|
vshr.u32 q6, q6, #32-9
|
||
|
|
add r11, r2, r1
|
||
|
|
veor.u32 q2, q2, q5
|
||
|
|
eor r6, r6, r10, ror #23
|
||
|
|
veor.u32 q10, q10, q7
|
||
|
|
eor r7, r7, r11, ror #23
|
||
|
|
veor.u32 q2, q2, q4
|
||
|
|
strd r6, [sp, #10*4]
|
||
|
|
veor.u32 q10, q10, q6
|
||
|
|
|
||
|
|
add r2, r7, r2
|
||
|
|
vadd.u32 q4, q2, q1
|
||
|
|
ldrd r10, [sp, #14*4]
|
||
|
|
vadd.u32 q6, q10, q9
|
||
|
|
add r3, r8, r3
|
||
|
|
vext.u32 q1, q1, q1, #3
|
||
|
|
eor r12, r12, r2, ror #19
|
||
|
|
vshl.u32 q5, q4, #13
|
||
|
|
add r2, r9, r4
|
||
|
|
vext.u32 q9, q9, q9, #3
|
||
|
|
eor lr, lr, r3, ror #19
|
||
|
|
vshl.u32 q7, q6, #13
|
||
|
|
add r3, r6, r5
|
||
|
|
vshr.u32 q4, q4, #32-13
|
||
|
|
eor r10, r10, r2, ror #19
|
||
|
|
vshr.u32 q6, q6, #32-13
|
||
|
|
eor r11, r11, r3, ror #19
|
||
|
|
veor.u32 q3, q3, q5
|
||
|
|
veor.u32 q11, q11, q7
|
||
|
|
veor.u32 q3, q3, q4
|
||
|
|
veor.u32 q11, q11, q6
|
||
|
|
|
||
|
|
ldrd r2, [sp, #2*4]
|
||
|
|
vadd.u32 q4, q3, q2
|
||
|
|
add r6, r11, r6
|
||
|
|
vadd.u32 q6, q11, q10
|
||
|
|
add r7, r12, r7
|
||
|
|
vswp.u32 d4, d5
|
||
|
|
eor r0, r0, r6, ror #14
|
||
|
|
vshl.u32 q5, q4, #18
|
||
|
|
add r6, lr, r8
|
||
|
|
vswp.u32 d20, d21
|
||
|
|
eor r1, r1, r7, ror #14
|
||
|
|
vshl.u32 q7, q6, #18
|
||
|
|
add r7, r10, r9
|
||
|
|
vext.u32 q3, q3, q3, #1
|
||
|
|
eor r2, r2, r6, ror #14
|
||
|
|
vshr.u32 q4, q4, #32-18
|
||
|
|
eor r3, r3, r7, ror #14
|
||
|
|
vshr.u32 q6, q6, #32-18
|
||
|
|
strd r2, [sp, #2*4]
|
||
|
|
vext.u32 q11, q11, q11, #1
|
||
|
|
strd r10, [sp, #14*4]
|
||
|
|
veor.u32 q0, q0, q5
|
||
|
|
veor.u32 q8, q8, q7
|
||
|
|
veor.u32 q0, q0, q4
|
||
|
|
veor.u32 q8, q8, q6
|
||
|
|
.endm
|
||
|
|
|
||
|
|
.macro salsa8_core_3way
|
||
|
|
ldmia sp, {r0-r12, lr}
|
||
|
|
ldrd r10, [sp, #14*4]
|
||
|
|
salsa8_core_3way_doubleround
|
||
|
|
salsa8_core_3way_doubleround
|
||
|
|
salsa8_core_3way_doubleround
|
||
|
|
salsa8_core_3way_doubleround
|
||
|
|
stmia sp, {r0-r5}
|
||
|
|
strd r8, [sp, #8*4]
|
||
|
|
str r12, [sp, #12*4]
|
||
|
|
str lr, [sp, #13*4]
|
||
|
|
.endm
|
||
|
|
|
||
|
|
.text
|
||
|
|
.code 32
|
||
|
|
.align 2
|
||
|
|
.globl scrypt_core_3way
|
||
|
|
.globl _scrypt_core_3way
|
||
|
|
#ifdef __ELF__
|
||
|
|
.type scrypt_core_3way, %function
|
||
|
|
#endif
|
||
|
|
scrypt_core_3way:
|
||
|
|
_scrypt_core_3way:
|
||
|
|
stmfd sp!, {r4-r11, lr}
|
||
|
|
vpush {q4-q7}
|
||
|
|
mov r12, sp
|
||
|
|
sub sp, sp, #24*16
|
||
|
|
bic sp, sp, #63
|
||
|
|
str r12, [sp, #4*16+3*4]
|
||
|
|
|
||
|
|
mov r2, r0
|
||
|
|
vldmia r2!, {q8-q15}
|
||
|
|
vmov.u64 q0, #0xffffffff
|
||
|
|
vmov.u32 q1, q8
|
||
|
|
vmov.u32 q2, q12
|
||
|
|
vbif.u32 q8, q9, q0
|
||
|
|
vbif.u32 q12, q13, q0
|
||
|
|
vbif.u32 q9, q10, q0
|
||
|
|
vbif.u32 q13, q14, q0
|
||
|
|
vbif.u32 q10, q11, q0
|
||
|
|
vbif.u32 q14, q15, q0
|
||
|
|
vbif.u32 q11, q1, q0
|
||
|
|
vbif.u32 q15, q2, q0
|
||
|
|
vldmia r2!, {q0-q7}
|
||
|
|
vswp.u32 d17, d21
|
||
|
|
vswp.u32 d25, d29
|
||
|
|
vswp.u32 d18, d22
|
||
|
|
vswp.u32 d26, d30
|
||
|
|
vstmia r0, {q8-q15}
|
||
|
|
vmov.u64 q8, #0xffffffff
|
||
|
|
vmov.u32 q9, q0
|
||
|
|
vmov.u32 q10, q4
|
||
|
|
vbif.u32 q0, q1, q8
|
||
|
|
vbif.u32 q4, q5, q8
|
||
|
|
vbif.u32 q1, q2, q8
|
||
|
|
vbif.u32 q5, q6, q8
|
||
|
|
vbif.u32 q2, q3, q8
|
||
|
|
vbif.u32 q6, q7, q8
|
||
|
|
vbif.u32 q3, q9, q8
|
||
|
|
vbif.u32 q7, q10, q8
|
||
|
|
vldmia r2, {q8-q15}
|
||
|
|
vswp.u32 d1, d5
|
||
|
|
vswp.u32 d9, d13
|
||
|
|
vswp.u32 d2, d6
|
||
|
|
vswp.u32 d10, d14
|
||
|
|
add r12, sp, #8*16
|
||
|
|
vstmia r12!, {q0-q7}
|
||
|
|
vmov.u64 q0, #0xffffffff
|
||
|
|
vmov.u32 q1, q8
|
||
|
|
vmov.u32 q2, q12
|
||
|
|
vbif.u32 q8, q9, q0
|
||
|
|
vbif.u32 q12, q13, q0
|
||
|
|
vbif.u32 q9, q10, q0
|
||
|
|
vbif.u32 q13, q14, q0
|
||
|
|
vbif.u32 q10, q11, q0
|
||
|
|
vbif.u32 q14, q15, q0
|
||
|
|
vbif.u32 q11, q1, q0
|
||
|
|
vbif.u32 q15, q2, q0
|
||
|
|
vswp.u32 d17, d21
|
||
|
|
vswp.u32 d25, d29
|
||
|
|
vswp.u32 d18, d22
|
||
|
|
vswp.u32 d26, d30
|
||
|
|
vstmia r12, {q8-q15}
|
||
|
|
|
||
|
|
add lr, sp, #128
|
||
|
|
vldmia lr, {q0-q7}
|
||
|
|
add r2, r1, #1024*32*4
|
||
|
|
str r0, [sp, #4*16+0*4]
|
||
|
|
str r2, [sp, #4*16+2*4]
|
||
|
|
scrypt_core_3way_loop1:
|
||
|
|
add lr, r0, #16*4
|
||
|
|
add r3, r1, #16*4
|
||
|
|
str r1, [sp, #4*16+1*4]
|
||
|
|
mov r12, sp
|
||
|
|
scrypt_core_macro1a_x4
|
||
|
|
scrypt_core_macro1a_x4
|
||
|
|
scrypt_core_macro1a_x4
|
||
|
|
scrypt_core_macro1a_x4
|
||
|
|
sub r1, r1, #4*16
|
||
|
|
|
||
|
|
add r1, r1, #1024*32*4
|
||
|
|
vstmia r1, {q0-q7}
|
||
|
|
add r3, r1, #1024*32*4
|
||
|
|
vstmia r3, {q8-q15}
|
||
|
|
|
||
|
|
add lr, sp, #128
|
||
|
|
veor.u32 q0, q0, q4
|
||
|
|
veor.u32 q1, q1, q5
|
||
|
|
veor.u32 q2, q2, q6
|
||
|
|
veor.u32 q3, q3, q7
|
||
|
|
vstmia lr, {q0-q3}
|
||
|
|
veor.u32 q8, q8, q12
|
||
|
|
veor.u32 q9, q9, q13
|
||
|
|
veor.u32 q10, q10, q14
|
||
|
|
veor.u32 q11, q11, q15
|
||
|
|
add r12, sp, #256
|
||
|
|
vstmia r12, {q8-q11}
|
||
|
|
|
||
|
|
salsa8_core_3way
|
||
|
|
|
||
|
|
ldr r0, [sp, #4*16+0*4]
|
||
|
|
mov r12, sp
|
||
|
|
add r2, r0, #16*4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
|
||
|
|
add lr, sp, #128
|
||
|
|
vldmia lr, {q4-q7}
|
||
|
|
vadd.u32 q4, q4, q0
|
||
|
|
vadd.u32 q5, q5, q1
|
||
|
|
vadd.u32 q6, q6, q2
|
||
|
|
vadd.u32 q7, q7, q3
|
||
|
|
add r12, sp, #256
|
||
|
|
vldmia r12, {q0-q3}
|
||
|
|
vstmia lr, {q4-q7}
|
||
|
|
vadd.u32 q8, q8, q0
|
||
|
|
vadd.u32 q9, q9, q1
|
||
|
|
vadd.u32 q10, q10, q2
|
||
|
|
vadd.u32 q11, q11, q3
|
||
|
|
|
||
|
|
add r4, sp, #128+4*16
|
||
|
|
vldmia r4, {q0-q3}
|
||
|
|
vstmia r12, {q8-q11}
|
||
|
|
veor.u32 q0, q0, q4
|
||
|
|
veor.u32 q1, q1, q5
|
||
|
|
veor.u32 q2, q2, q6
|
||
|
|
veor.u32 q3, q3, q7
|
||
|
|
vstmia r4, {q0-q3}
|
||
|
|
veor.u32 q8, q8, q12
|
||
|
|
veor.u32 q9, q9, q13
|
||
|
|
veor.u32 q10, q10, q14
|
||
|
|
veor.u32 q11, q11, q15
|
||
|
|
vmov q12, q8
|
||
|
|
vmov q13, q9
|
||
|
|
vmov q14, q10
|
||
|
|
vmov q15, q11
|
||
|
|
|
||
|
|
salsa8_core_3way
|
||
|
|
|
||
|
|
ldr r0, [sp, #4*16+0*4]
|
||
|
|
mov r1, sp
|
||
|
|
add r0, r0, #16*4
|
||
|
|
scrypt_core_macro3_x6
|
||
|
|
scrypt_core_macro3_x6
|
||
|
|
scrypt_core_macro3_x4
|
||
|
|
sub r0, r0, #8*16
|
||
|
|
|
||
|
|
ldr r1, [sp, #4*16+1*4]
|
||
|
|
ldr r2, [sp, #4*16+2*4]
|
||
|
|
add lr, sp, #128
|
||
|
|
add r4, sp, #128+4*16
|
||
|
|
vldmia r4, {q4-q7}
|
||
|
|
vadd.u32 q4, q4, q0
|
||
|
|
vadd.u32 q5, q5, q1
|
||
|
|
vadd.u32 q6, q6, q2
|
||
|
|
vadd.u32 q7, q7, q3
|
||
|
|
vstmia r4, {q4-q7}
|
||
|
|
vldmia lr, {q0-q3}
|
||
|
|
vadd.u32 q12, q12, q8
|
||
|
|
vadd.u32 q13, q13, q9
|
||
|
|
vadd.u32 q14, q14, q10
|
||
|
|
vadd.u32 q15, q15, q11
|
||
|
|
add r12, sp, #256
|
||
|
|
vldmia r12, {q8-q11}
|
||
|
|
|
||
|
|
add r1, r1, #8*16
|
||
|
|
cmp r1, r2
|
||
|
|
bne scrypt_core_3way_loop1
|
||
|
|
|
||
|
|
add r5, sp, #256+4*16
|
||
|
|
vstmia r5, {q12-q15}
|
||
|
|
|
||
|
|
sub r1, r1, #1024*32*4
|
||
|
|
str r1, [sp, #4*16+1*4]
|
||
|
|
mov r2, #1024
|
||
|
|
scrypt_core_3way_loop2:
|
||
|
|
str r2, [sp, #4*16+2*4]
|
||
|
|
|
||
|
|
ldr r0, [sp, #4*16+0*4]
|
||
|
|
ldr r1, [sp, #4*16+1*4]
|
||
|
|
ldr r4, [r0, #16*4]
|
||
|
|
mov r4, r4, lsl #32-10
|
||
|
|
add r1, r1, r4, lsr #32-10-7
|
||
|
|
add r2, r0, #16*4
|
||
|
|
add r3, r1, #16*4
|
||
|
|
mov r12, sp
|
||
|
|
scrypt_core_macro1b_x4
|
||
|
|
scrypt_core_macro1b_x4
|
||
|
|
scrypt_core_macro1b_x4
|
||
|
|
scrypt_core_macro1b_x4
|
||
|
|
|
||
|
|
ldr r1, [sp, #4*16+1*4]
|
||
|
|
add r1, r1, #1024*32*4
|
||
|
|
add r3, r1, #1024*32*4
|
||
|
|
vmov r6, r7, d8
|
||
|
|
mov r6, r6, lsl #32-10
|
||
|
|
add r6, r1, r6, lsr #32-10-7
|
||
|
|
vmov r7, r8, d24
|
||
|
|
add lr, sp, #128
|
||
|
|
vldmia lr, {q0-q3}
|
||
|
|
pld [r6]
|
||
|
|
pld [r6, #8*4]
|
||
|
|
pld [r6, #16*4]
|
||
|
|
pld [r6, #24*4]
|
||
|
|
vldmia r6, {q8-q15}
|
||
|
|
mov r7, r7, lsl #32-10
|
||
|
|
add r7, r3, r7, lsr #32-10-7
|
||
|
|
veor.u32 q8, q8, q0
|
||
|
|
veor.u32 q9, q9, q1
|
||
|
|
veor.u32 q10, q10, q2
|
||
|
|
veor.u32 q11, q11, q3
|
||
|
|
pld [r7]
|
||
|
|
pld [r7, #8*4]
|
||
|
|
pld [r7, #16*4]
|
||
|
|
pld [r7, #24*4]
|
||
|
|
veor.u32 q12, q12, q4
|
||
|
|
veor.u32 q13, q13, q5
|
||
|
|
veor.u32 q14, q14, q6
|
||
|
|
veor.u32 q15, q15, q7
|
||
|
|
vldmia r7, {q0-q7}
|
||
|
|
vstmia lr, {q8-q15}
|
||
|
|
add r12, sp, #256
|
||
|
|
vldmia r12, {q8-q15}
|
||
|
|
veor.u32 q8, q8, q0
|
||
|
|
veor.u32 q9, q9, q1
|
||
|
|
veor.u32 q10, q10, q2
|
||
|
|
veor.u32 q11, q11, q3
|
||
|
|
veor.u32 q12, q12, q4
|
||
|
|
veor.u32 q13, q13, q5
|
||
|
|
veor.u32 q14, q14, q6
|
||
|
|
veor.u32 q15, q15, q7
|
||
|
|
|
||
|
|
vldmia lr, {q0-q7}
|
||
|
|
veor.u32 q0, q0, q4
|
||
|
|
veor.u32 q1, q1, q5
|
||
|
|
veor.u32 q2, q2, q6
|
||
|
|
veor.u32 q3, q3, q7
|
||
|
|
vstmia lr, {q0-q3}
|
||
|
|
veor.u32 q8, q8, q12
|
||
|
|
veor.u32 q9, q9, q13
|
||
|
|
veor.u32 q10, q10, q14
|
||
|
|
veor.u32 q11, q11, q15
|
||
|
|
vstmia r12, {q8-q15}
|
||
|
|
|
||
|
|
salsa8_core_3way
|
||
|
|
|
||
|
|
ldr r0, [sp, #4*16+0*4]
|
||
|
|
mov r12, sp
|
||
|
|
add r2, r0, #16*4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
scrypt_core_macro2_x4
|
||
|
|
|
||
|
|
add lr, sp, #128
|
||
|
|
vldmia lr, {q4-q7}
|
||
|
|
vadd.u32 q4, q4, q0
|
||
|
|
vadd.u32 q5, q5, q1
|
||
|
|
vadd.u32 q6, q6, q2
|
||
|
|
vadd.u32 q7, q7, q3
|
||
|
|
add r12, sp, #256
|
||
|
|
vldmia r12, {q12-q15}
|
||
|
|
vstmia lr, {q4-q7}
|
||
|
|
vadd.u32 q12, q12, q8
|
||
|
|
vadd.u32 q13, q13, q9
|
||
|
|
vadd.u32 q14, q14, q10
|
||
|
|
vadd.u32 q15, q15, q11
|
||
|
|
|
||
|
|
add r4, sp, #128+4*16
|
||
|
|
vldmia r4, {q0-q3}
|
||
|
|
vstmia r12, {q12-q15}
|
||
|
|
veor.u32 q0, q0, q4
|
||
|
|
veor.u32 q1, q1, q5
|
||
|
|
veor.u32 q2, q2, q6
|
||
|
|
veor.u32 q3, q3, q7
|
||
|
|
add r5, sp, #256+4*16
|
||
|
|
vldmia r5, {q8-q11}
|
||
|
|
vstmia r4, {q0-q3}
|
||
|
|
veor.u32 q8, q8, q12
|
||
|
|
veor.u32 q9, q9, q13
|
||
|
|
veor.u32 q10, q10, q14
|
||
|
|
veor.u32 q11, q11, q15
|
||
|
|
vmov q12, q8
|
||
|
|
vmov q13, q9
|
||
|
|
vmov q14, q10
|
||
|
|
vmov q15, q11
|
||
|
|
|
||
|
|
salsa8_core_3way
|
||
|
|
|
||
|
|
ldr r0, [sp, #4*16+0*4]
|
||
|
|
ldr r3, [sp, #4*16+1*4]
|
||
|
|
mov r1, sp
|
||
|
|
add r0, r0, #16*4
|
||
|
|
scrypt_core_macro3_x4
|
||
|
|
mov r4, r4, lsl #32-10
|
||
|
|
add r3, r3, r4, lsr #32-10-7
|
||
|
|
pld [r3, #16*4]
|
||
|
|
pld [r3]
|
||
|
|
pld [r3, #24*4]
|
||
|
|
pld [r3, #8*4]
|
||
|
|
scrypt_core_macro3_x6
|
||
|
|
scrypt_core_macro3_x6
|
||
|
|
|
||
|
|
add lr, sp, #128
|
||
|
|
add r4, sp, #128+4*16
|
||
|
|
vldmia r4, {q4-q7}
|
||
|
|
vadd.u32 q4, q4, q0
|
||
|
|
vadd.u32 q5, q5, q1
|
||
|
|
vadd.u32 q6, q6, q2
|
||
|
|
vadd.u32 q7, q7, q3
|
||
|
|
vstmia r4, {q4-q7}
|
||
|
|
vadd.u32 q12, q12, q8
|
||
|
|
vadd.u32 q13, q13, q9
|
||
|
|
vadd.u32 q14, q14, q10
|
||
|
|
vadd.u32 q15, q15, q11
|
||
|
|
add r5, sp, #256+4*16
|
||
|
|
vstmia r5, {q12-q15}
|
||
|
|
|
||
|
|
ldr r2, [sp, #4*16+2*4]
|
||
|
|
subs r2, r2, #1
|
||
|
|
bne scrypt_core_3way_loop2
|
||
|
|
|
||
|
|
ldr r0, [sp, #4*16+0*4]
|
||
|
|
vldmia r0, {q8-q15}
|
||
|
|
vmov.u64 q0, #0xffffffff
|
||
|
|
vmov.u32 q1, q8
|
||
|
|
vmov.u32 q2, q12
|
||
|
|
vbif.u32 q8, q9, q0
|
||
|
|
vbif.u32 q12, q13, q0
|
||
|
|
vbif.u32 q9, q10, q0
|
||
|
|
vbif.u32 q13, q14, q0
|
||
|
|
vbif.u32 q10, q11, q0
|
||
|
|
vbif.u32 q14, q15, q0
|
||
|
|
vbif.u32 q11, q1, q0
|
||
|
|
vbif.u32 q15, q2, q0
|
||
|
|
add r12, sp, #8*16
|
||
|
|
vldmia r12!, {q0-q7}
|
||
|
|
vswp.u32 d17, d21
|
||
|
|
vswp.u32 d25, d29
|
||
|
|
vswp.u32 d18, d22
|
||
|
|
vswp.u32 d26, d30
|
||
|
|
vstmia r0!, {q8-q15}
|
||
|
|
vmov.u64 q8, #0xffffffff
|
||
|
|
vmov.u32 q9, q0
|
||
|
|
vmov.u32 q10, q4
|
||
|
|
vbif.u32 q0, q1, q8
|
||
|
|
vbif.u32 q4, q5, q8
|
||
|
|
vbif.u32 q1, q2, q8
|
||
|
|
vbif.u32 q5, q6, q8
|
||
|
|
vbif.u32 q2, q3, q8
|
||
|
|
vbif.u32 q6, q7, q8
|
||
|
|
vbif.u32 q3, q9, q8
|
||
|
|
vbif.u32 q7, q10, q8
|
||
|
|
vldmia r12, {q8-q15}
|
||
|
|
vswp.u32 d1, d5
|
||
|
|
vswp.u32 d9, d13
|
||
|
|
vswp.u32 d2, d6
|
||
|
|
vswp.u32 d10, d14
|
||
|
|
vstmia r0!, {q0-q7}
|
||
|
|
vmov.u64 q0, #0xffffffff
|
||
|
|
vmov.u32 q1, q8
|
||
|
|
vmov.u32 q2, q12
|
||
|
|
vbif.u32 q8, q9, q0
|
||
|
|
vbif.u32 q12, q13, q0
|
||
|
|
vbif.u32 q9, q10, q0
|
||
|
|
vbif.u32 q13, q14, q0
|
||
|
|
vbif.u32 q10, q11, q0
|
||
|
|
vbif.u32 q14, q15, q0
|
||
|
|
vbif.u32 q11, q1, q0
|
||
|
|
vbif.u32 q15, q2, q0
|
||
|
|
vswp.u32 d17, d21
|
||
|
|
vswp.u32 d25, d29
|
||
|
|
vswp.u32 d18, d22
|
||
|
|
vswp.u32 d26, d30
|
||
|
|
vstmia r0, {q8-q15}
|
||
|
|
|
||
|
|
ldr sp, [sp, #4*16+3*4]
|
||
|
|
vpop {q4-q7}
|
||
|
|
ldmfd sp!, {r4-r11, pc}
|
||
|
|
|
||
|
|
#endif /* __ARM_NEON__ */
|
||
|
|
|
||
|
|
#endif
|