| #include <openssl/arm_arch.h> |
| |
| .text |
| #if defined(__thumb2__) |
| .syntax unified |
| .thumb |
| #else |
| .code 32 |
| #endif |
| |
| .globl _sha1_block_data_order |
| .private_extern _sha1_block_data_order |
| #ifdef __thumb2__ |
| .thumb_func _sha1_block_data_order |
| #endif |
| |
| .align 5 |
| _sha1_block_data_order: |
| #if __ARM_MAX_ARCH__>=7 |
| Lsha1_block: |
| adr r3,Lsha1_block |
| ldr r12,LOPENSSL_armcap |
| ldr r12,[r3,r12] @ OPENSSL_armcap_P |
| #ifdef __APPLE__ |
| ldr r12,[r12] |
| #endif |
| tst r12,#ARMV8_SHA1 |
| bne LARMv8 |
| tst r12,#ARMV7_NEON |
| bne LNEON |
| #endif |
| stmdb sp!,{r4,r5,r6,r7,r8,r9,r10,r11,r12,lr} |
| add r2,r1,r2,lsl#6 @ r2 to point at the end of r1 |
| ldmia r0,{r3,r4,r5,r6,r7} |
| Lloop: |
| ldr r8,LK_00_19 |
| mov r14,sp |
| sub sp,sp,#15*4 |
| mov r5,r5,ror#30 |
| mov r6,r6,ror#30 |
| mov r7,r7,ror#30 @ [6] |
| L_00_15: |
| #if __ARM_ARCH__<7 |
| ldrb r10,[r1,#2] |
| ldrb r9,[r1,#3] |
| ldrb r11,[r1,#1] |
| add r7,r8,r7,ror#2 @ E+=K_00_19 |
| ldrb r12,[r1],#4 |
| orr r9,r9,r10,lsl#8 |
| eor r10,r5,r6 @ F_xx_xx |
| orr r9,r9,r11,lsl#16 |
| add r7,r7,r3,ror#27 @ E+=ROR(A,27) |
| orr r9,r9,r12,lsl#24 |
| #else |
| ldr r9,[r1],#4 @ handles unaligned |
| add r7,r8,r7,ror#2 @ E+=K_00_19 |
| eor r10,r5,r6 @ F_xx_xx |
| add r7,r7,r3,ror#27 @ E+=ROR(A,27) |
| #ifdef __ARMEL__ |
| rev r9,r9 @ byte swap |
| #endif |
| #endif |
| and r10,r4,r10,ror#2 |
| add r7,r7,r9 @ E+=X[i] |
| eor r10,r10,r6,ror#2 @ F_00_19(B,C,D) |
| str r9,[r14,#-4]! |
| add r7,r7,r10 @ E+=F_00_19(B,C,D) |
| #if __ARM_ARCH__<7 |
| ldrb r10,[r1,#2] |
| ldrb r9,[r1,#3] |
| ldrb r11,[r1,#1] |
| add r6,r8,r6,ror#2 @ E+=K_00_19 |
| ldrb r12,[r1],#4 |
| orr r9,r9,r10,lsl#8 |
| eor r10,r4,r5 @ F_xx_xx |
| orr r9,r9,r11,lsl#16 |
| add r6,r6,r7,ror#27 @ E+=ROR(A,27) |
| orr r9,r9,r12,lsl#24 |
| #else |
| ldr r9,[r1],#4 @ handles unaligned |
| add r6,r8,r6,ror#2 @ E+=K_00_19 |
| eor r10,r4,r5 @ F_xx_xx |
| add r6,r6,r7,ror#27 @ E+=ROR(A,27) |
| #ifdef __ARMEL__ |
| rev r9,r9 @ byte swap |
| #endif |
| #endif |
| and r10,r3,r10,ror#2 |
| add r6,r6,r9 @ E+=X[i] |
| eor r10,r10,r5,ror#2 @ F_00_19(B,C,D) |
| str r9,[r14,#-4]! |
| add r6,r6,r10 @ E+=F_00_19(B,C,D) |
| #if __ARM_ARCH__<7 |
| ldrb r10,[r1,#2] |
| ldrb r9,[r1,#3] |
| ldrb r11,[r1,#1] |
| add r5,r8,r5,ror#2 @ E+=K_00_19 |
| ldrb r12,[r1],#4 |
| orr r9,r9,r10,lsl#8 |
| eor r10,r3,r4 @ F_xx_xx |
| orr r9,r9,r11,lsl#16 |
| add r5,r5,r6,ror#27 @ E+=ROR(A,27) |
| orr r9,r9,r12,lsl#24 |
| #else |
| ldr r9,[r1],#4 @ handles unaligned |
| add r5,r8,r5,ror#2 @ E+=K_00_19 |
| eor r10,r3,r4 @ F_xx_xx |
| add r5,r5,r6,ror#27 @ E+=ROR(A,27) |
| #ifdef __ARMEL__ |
| rev r9,r9 @ byte swap |
| #endif |
| #endif |
| and r10,r7,r10,ror#2 |
| add r5,r5,r9 @ E+=X[i] |
| eor r10,r10,r4,ror#2 @ F_00_19(B,C,D) |
| str r9,[r14,#-4]! |
| add r5,r5,r10 @ E+=F_00_19(B,C,D) |
| #if __ARM_ARCH__<7 |
| ldrb r10,[r1,#2] |
| ldrb r9,[r1,#3] |
| ldrb r11,[r1,#1] |
| add r4,r8,r4,ror#2 @ E+=K_00_19 |
| ldrb r12,[r1],#4 |
| orr r9,r9,r10,lsl#8 |
| eor r10,r7,r3 @ F_xx_xx |
| orr r9,r9,r11,lsl#16 |
| add r4,r4,r5,ror#27 @ E+=ROR(A,27) |
| orr r9,r9,r12,lsl#24 |
| #else |
| ldr r9,[r1],#4 @ handles unaligned |
| add r4,r8,r4,ror#2 @ E+=K_00_19 |
| eor r10,r7,r3 @ F_xx_xx |
| add r4,r4,r5,ror#27 @ E+=ROR(A,27) |
| #ifdef __ARMEL__ |
| rev r9,r9 @ byte swap |
| #endif |
| #endif |
| and r10,r6,r10,ror#2 |
| add r4,r4,r9 @ E+=X[i] |
| eor r10,r10,r3,ror#2 @ F_00_19(B,C,D) |
| str r9,[r14,#-4]! |
| add r4,r4,r10 @ E+=F_00_19(B,C,D) |
| #if __ARM_ARCH__<7 |
| ldrb r10,[r1,#2] |
| ldrb r9,[r1,#3] |
| ldrb r11,[r1,#1] |
| add r3,r8,r3,ror#2 @ E+=K_00_19 |
| ldrb r12,[r1],#4 |
| orr r9,r9,r10,lsl#8 |
| eor r10,r6,r7 @ F_xx_xx |
| orr r9,r9,r11,lsl#16 |
| add r3,r3,r4,ror#27 @ E+=ROR(A,27) |
| orr r9,r9,r12,lsl#24 |
| #else |
| ldr r9,[r1],#4 @ handles unaligned |
| add r3,r8,r3,ror#2 @ E+=K_00_19 |
| eor r10,r6,r7 @ F_xx_xx |
| add r3,r3,r4,ror#27 @ E+=ROR(A,27) |
| #ifdef __ARMEL__ |
| rev r9,r9 @ byte swap |
| #endif |
| #endif |
| and r10,r5,r10,ror#2 |
| add r3,r3,r9 @ E+=X[i] |
| eor r10,r10,r7,ror#2 @ F_00_19(B,C,D) |
| str r9,[r14,#-4]! |
| add r3,r3,r10 @ E+=F_00_19(B,C,D) |
| #if defined(__thumb2__) |
| mov r12,sp |
| teq r14,r12 |
| #else |
| teq r14,sp |
| #endif |
| bne L_00_15 @ [((11+4)*5+2)*3] |
| sub sp,sp,#25*4 |
| #if __ARM_ARCH__<7 |
| ldrb r10,[r1,#2] |
| ldrb r9,[r1,#3] |
| ldrb r11,[r1,#1] |
| add r7,r8,r7,ror#2 @ E+=K_00_19 |
| ldrb r12,[r1],#4 |
| orr r9,r9,r10,lsl#8 |
| eor r10,r5,r6 @ F_xx_xx |
| orr r9,r9,r11,lsl#16 |
| add r7,r7,r3,ror#27 @ E+=ROR(A,27) |
| orr r9,r9,r12,lsl#24 |
| #else |
| ldr r9,[r1],#4 @ handles unaligned |
| add r7,r8,r7,ror#2 @ E+=K_00_19 |
| eor r10,r5,r6 @ F_xx_xx |
| add r7,r7,r3,ror#27 @ E+=ROR(A,27) |
| #ifdef __ARMEL__ |
| rev r9,r9 @ byte swap |
| #endif |
| #endif |
| and r10,r4,r10,ror#2 |
| add r7,r7,r9 @ E+=X[i] |
| eor r10,r10,r6,ror#2 @ F_00_19(B,C,D) |
| str r9,[r14,#-4]! |
| add r7,r7,r10 @ E+=F_00_19(B,C,D) |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r6,r8,r6,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r4,r5 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r6,r6,r7,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| and r10,r3,r10,ror#2 @ F_xx_xx |
| @ F_xx_xx |
| add r6,r6,r9 @ E+=X[i] |
| eor r10,r10,r5,ror#2 @ F_00_19(B,C,D) |
| add r6,r6,r10 @ E+=F_00_19(B,C,D) |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r5,r8,r5,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r3,r4 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r5,r5,r6,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| and r10,r7,r10,ror#2 @ F_xx_xx |
| @ F_xx_xx |
| add r5,r5,r9 @ E+=X[i] |
| eor r10,r10,r4,ror#2 @ F_00_19(B,C,D) |
| add r5,r5,r10 @ E+=F_00_19(B,C,D) |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r4,r8,r4,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r7,r3 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r4,r4,r5,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| and r10,r6,r10,ror#2 @ F_xx_xx |
| @ F_xx_xx |
| add r4,r4,r9 @ E+=X[i] |
| eor r10,r10,r3,ror#2 @ F_00_19(B,C,D) |
| add r4,r4,r10 @ E+=F_00_19(B,C,D) |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r3,r8,r3,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r6,r7 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r3,r3,r4,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| and r10,r5,r10,ror#2 @ F_xx_xx |
| @ F_xx_xx |
| add r3,r3,r9 @ E+=X[i] |
| eor r10,r10,r7,ror#2 @ F_00_19(B,C,D) |
| add r3,r3,r10 @ E+=F_00_19(B,C,D) |
| |
| ldr r8,LK_20_39 @ [+15+16*4] |
| cmn sp,#0 @ [+3], clear carry to denote 20_39 |
| L_20_39_or_60_79: |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r7,r8,r7,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r5,r6 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r7,r7,r3,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| eor r10,r4,r10,ror#2 @ F_xx_xx |
| @ F_xx_xx |
| add r7,r7,r9 @ E+=X[i] |
| add r7,r7,r10 @ E+=F_20_39(B,C,D) |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r6,r8,r6,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r4,r5 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r6,r6,r7,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| eor r10,r3,r10,ror#2 @ F_xx_xx |
| @ F_xx_xx |
| add r6,r6,r9 @ E+=X[i] |
| add r6,r6,r10 @ E+=F_20_39(B,C,D) |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r5,r8,r5,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r3,r4 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r5,r5,r6,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| eor r10,r7,r10,ror#2 @ F_xx_xx |
| @ F_xx_xx |
| add r5,r5,r9 @ E+=X[i] |
| add r5,r5,r10 @ E+=F_20_39(B,C,D) |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r4,r8,r4,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r7,r3 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r4,r4,r5,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| eor r10,r6,r10,ror#2 @ F_xx_xx |
| @ F_xx_xx |
| add r4,r4,r9 @ E+=X[i] |
| add r4,r4,r10 @ E+=F_20_39(B,C,D) |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r3,r8,r3,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r6,r7 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r3,r3,r4,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| eor r10,r5,r10,ror#2 @ F_xx_xx |
| @ F_xx_xx |
| add r3,r3,r9 @ E+=X[i] |
| add r3,r3,r10 @ E+=F_20_39(B,C,D) |
| #if defined(__thumb2__) |
| mov r12,sp |
| teq r14,r12 |
| #else |
| teq r14,sp @ preserve carry |
| #endif |
| bne L_20_39_or_60_79 @ [+((12+3)*5+2)*4] |
| bcs L_done @ [+((12+3)*5+2)*4], spare 300 bytes |
| |
| ldr r8,LK_40_59 |
| sub sp,sp,#20*4 @ [+2] |
| L_40_59: |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r7,r8,r7,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r5,r6 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r7,r7,r3,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| and r10,r4,r10,ror#2 @ F_xx_xx |
| and r11,r5,r6 @ F_xx_xx |
| add r7,r7,r9 @ E+=X[i] |
| add r7,r7,r10 @ E+=F_40_59(B,C,D) |
| add r7,r7,r11,ror#2 |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r6,r8,r6,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r4,r5 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r6,r6,r7,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| and r10,r3,r10,ror#2 @ F_xx_xx |
| and r11,r4,r5 @ F_xx_xx |
| add r6,r6,r9 @ E+=X[i] |
| add r6,r6,r10 @ E+=F_40_59(B,C,D) |
| add r6,r6,r11,ror#2 |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r5,r8,r5,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r3,r4 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r5,r5,r6,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| and r10,r7,r10,ror#2 @ F_xx_xx |
| and r11,r3,r4 @ F_xx_xx |
| add r5,r5,r9 @ E+=X[i] |
| add r5,r5,r10 @ E+=F_40_59(B,C,D) |
| add r5,r5,r11,ror#2 |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r4,r8,r4,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r7,r3 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r4,r4,r5,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| and r10,r6,r10,ror#2 @ F_xx_xx |
| and r11,r7,r3 @ F_xx_xx |
| add r4,r4,r9 @ E+=X[i] |
| add r4,r4,r10 @ E+=F_40_59(B,C,D) |
| add r4,r4,r11,ror#2 |
| ldr r9,[r14,#15*4] |
| ldr r10,[r14,#13*4] |
| ldr r11,[r14,#7*4] |
| add r3,r8,r3,ror#2 @ E+=K_xx_xx |
| ldr r12,[r14,#2*4] |
| eor r9,r9,r10 |
| eor r11,r11,r12 @ 1 cycle stall |
| eor r10,r6,r7 @ F_xx_xx |
| mov r9,r9,ror#31 |
| add r3,r3,r4,ror#27 @ E+=ROR(A,27) |
| eor r9,r9,r11,ror#31 |
| str r9,[r14,#-4]! |
| and r10,r5,r10,ror#2 @ F_xx_xx |
| and r11,r6,r7 @ F_xx_xx |
| add r3,r3,r9 @ E+=X[i] |
| add r3,r3,r10 @ E+=F_40_59(B,C,D) |
| add r3,r3,r11,ror#2 |
| #if defined(__thumb2__) |
| mov r12,sp |
| teq r14,r12 |
| #else |
| teq r14,sp |
| #endif |
| bne L_40_59 @ [+((12+5)*5+2)*4] |
| |
| ldr r8,LK_60_79 |
| sub sp,sp,#20*4 |
| cmp sp,#0 @ set carry to denote 60_79 |
| b L_20_39_or_60_79 @ [+4], spare 300 bytes |
| L_done: |
| add sp,sp,#80*4 @ "deallocate" stack frame |
| ldmia r0,{r8,r9,r10,r11,r12} |
| add r3,r8,r3 |
| add r4,r9,r4 |
| add r5,r10,r5,ror#2 |
| add r6,r11,r6,ror#2 |
| add r7,r12,r7,ror#2 |
| stmia r0,{r3,r4,r5,r6,r7} |
| teq r1,r2 |
| bne Lloop @ [+18], total 1307 |
| |
| #if __ARM_ARCH__>=5 |
| ldmia sp!,{r4,r5,r6,r7,r8,r9,r10,r11,r12,pc} |
| #else |
| ldmia sp!,{r4,r5,r6,r7,r8,r9,r10,r11,r12,lr} |
| tst lr,#1 |
| moveq pc,lr @ be binary compatible with V4, yet |
| .word 0xe12fff1e @ interoperable with Thumb ISA:-) |
| #endif |
| |
| |
| .align 5 |
| LK_00_19:.word 0x5a827999 |
| LK_20_39:.word 0x6ed9eba1 |
| LK_40_59:.word 0x8f1bbcdc |
| LK_60_79:.word 0xca62c1d6 |
| #if __ARM_MAX_ARCH__>=7 |
| LOPENSSL_armcap: |
| .word OPENSSL_armcap_P-Lsha1_block |
| #endif |
| .byte 83,72,65,49,32,98,108,111,99,107,32,116,114,97,110,115,102,111,114,109,32,102,111,114,32,65,82,77,118,52,47,78,69,79,78,47,65,82,77,118,56,44,32,67,82,89,80,84,79,71,65,77,83,32,98,121,32,60,97,112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103,62,0 |
| .align 2 |
| .align 5 |
| #if __ARM_MAX_ARCH__>=7 |
| |
| |
| |
| #ifdef __thumb2__ |
| .thumb_func sha1_block_data_order_neon |
| #endif |
| .align 4 |
| sha1_block_data_order_neon: |
| LNEON: |
| stmdb sp!,{r4,r5,r6,r7,r8,r9,r10,r11,r12,lr} |
| add r2,r1,r2,lsl#6 @ r2 to point at the end of r1 |
| @ dmb @ errata #451034 on early Cortex A8 |
| @ vstmdb sp!,{d8-d15} @ ABI specification says so |
| mov r14,sp |
| sub r12,sp,#64 |
| adr r8,LK_00_19 |
| bic r12,r12,#15 @ align for 128-bit stores |
| |
| ldmia r0,{r3,r4,r5,r6,r7} @ load context |
| mov sp,r12 @ alloca |
| |
| vld1.8 {q0,q1},[r1]! @ handles unaligned |
| veor q15,q15,q15 |
| vld1.8 {q2,q3},[r1]! |
| vld1.32 {d28[],d29[]},[r8,:32]! @ load K_00_19 |
| vrev32.8 q0,q0 @ yes, even on |
| vrev32.8 q1,q1 @ big-endian... |
| vrev32.8 q2,q2 |
| vadd.i32 q8,q0,q14 |
| vrev32.8 q3,q3 |
| vadd.i32 q9,q1,q14 |
| vst1.32 {q8},[r12,:128]! |
| vadd.i32 q10,q2,q14 |
| vst1.32 {q9},[r12,:128]! |
| vst1.32 {q10},[r12,:128]! |
| ldr r9,[sp] @ big RAW stall |
| |
| Loop_neon: |
| vext.8 q8,q0,q1,#8 |
| bic r10,r6,r4 |
| add r7,r7,r9 |
| and r11,r5,r4 |
| vadd.i32 q13,q3,q14 |
| ldr r9,[sp,#4] |
| add r7,r7,r3,ror#27 |
| vext.8 q12,q3,q15,#4 |
| eor r11,r11,r10 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| veor q8,q8,q0 |
| bic r10,r5,r3 |
| add r6,r6,r9 |
| veor q12,q12,q2 |
| and r11,r4,r3 |
| ldr r9,[sp,#8] |
| veor q12,q12,q8 |
| add r6,r6,r7,ror#27 |
| eor r11,r11,r10 |
| vst1.32 {q13},[r12,:128]! |
| sub r12,r12,#64 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| vext.8 q13,q15,q12,#4 |
| bic r10,r4,r7 |
| add r5,r5,r9 |
| vadd.i32 q8,q12,q12 |
| and r11,r3,r7 |
| ldr r9,[sp,#12] |
| vsri.32 q8,q12,#31 |
| add r5,r5,r6,ror#27 |
| eor r11,r11,r10 |
| mov r7,r7,ror#2 |
| vshr.u32 q12,q13,#30 |
| add r5,r5,r11 |
| bic r10,r3,r6 |
| vshl.u32 q13,q13,#2 |
| add r4,r4,r9 |
| and r11,r7,r6 |
| veor q8,q8,q12 |
| ldr r9,[sp,#16] |
| add r4,r4,r5,ror#27 |
| veor q8,q8,q13 |
| eor r11,r11,r10 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| vext.8 q9,q1,q2,#8 |
| bic r10,r7,r5 |
| add r3,r3,r9 |
| and r11,r6,r5 |
| vadd.i32 q13,q8,q14 |
| ldr r9,[sp,#20] |
| vld1.32 {d28[],d29[]},[r8,:32]! |
| add r3,r3,r4,ror#27 |
| vext.8 q12,q8,q15,#4 |
| eor r11,r11,r10 |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| veor q9,q9,q1 |
| bic r10,r6,r4 |
| add r7,r7,r9 |
| veor q12,q12,q3 |
| and r11,r5,r4 |
| ldr r9,[sp,#24] |
| veor q12,q12,q9 |
| add r7,r7,r3,ror#27 |
| eor r11,r11,r10 |
| vst1.32 {q13},[r12,:128]! |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| vext.8 q13,q15,q12,#4 |
| bic r10,r5,r3 |
| add r6,r6,r9 |
| vadd.i32 q9,q12,q12 |
| and r11,r4,r3 |
| ldr r9,[sp,#28] |
| vsri.32 q9,q12,#31 |
| add r6,r6,r7,ror#27 |
| eor r11,r11,r10 |
| mov r3,r3,ror#2 |
| vshr.u32 q12,q13,#30 |
| add r6,r6,r11 |
| bic r10,r4,r7 |
| vshl.u32 q13,q13,#2 |
| add r5,r5,r9 |
| and r11,r3,r7 |
| veor q9,q9,q12 |
| ldr r9,[sp,#32] |
| add r5,r5,r6,ror#27 |
| veor q9,q9,q13 |
| eor r11,r11,r10 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| vext.8 q10,q2,q3,#8 |
| bic r10,r3,r6 |
| add r4,r4,r9 |
| and r11,r7,r6 |
| vadd.i32 q13,q9,q14 |
| ldr r9,[sp,#36] |
| add r4,r4,r5,ror#27 |
| vext.8 q12,q9,q15,#4 |
| eor r11,r11,r10 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| veor q10,q10,q2 |
| bic r10,r7,r5 |
| add r3,r3,r9 |
| veor q12,q12,q8 |
| and r11,r6,r5 |
| ldr r9,[sp,#40] |
| veor q12,q12,q10 |
| add r3,r3,r4,ror#27 |
| eor r11,r11,r10 |
| vst1.32 {q13},[r12,:128]! |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| vext.8 q13,q15,q12,#4 |
| bic r10,r6,r4 |
| add r7,r7,r9 |
| vadd.i32 q10,q12,q12 |
| and r11,r5,r4 |
| ldr r9,[sp,#44] |
| vsri.32 q10,q12,#31 |
| add r7,r7,r3,ror#27 |
| eor r11,r11,r10 |
| mov r4,r4,ror#2 |
| vshr.u32 q12,q13,#30 |
| add r7,r7,r11 |
| bic r10,r5,r3 |
| vshl.u32 q13,q13,#2 |
| add r6,r6,r9 |
| and r11,r4,r3 |
| veor q10,q10,q12 |
| ldr r9,[sp,#48] |
| add r6,r6,r7,ror#27 |
| veor q10,q10,q13 |
| eor r11,r11,r10 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| vext.8 q11,q3,q8,#8 |
| bic r10,r4,r7 |
| add r5,r5,r9 |
| and r11,r3,r7 |
| vadd.i32 q13,q10,q14 |
| ldr r9,[sp,#52] |
| add r5,r5,r6,ror#27 |
| vext.8 q12,q10,q15,#4 |
| eor r11,r11,r10 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| veor q11,q11,q3 |
| bic r10,r3,r6 |
| add r4,r4,r9 |
| veor q12,q12,q9 |
| and r11,r7,r6 |
| ldr r9,[sp,#56] |
| veor q12,q12,q11 |
| add r4,r4,r5,ror#27 |
| eor r11,r11,r10 |
| vst1.32 {q13},[r12,:128]! |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| vext.8 q13,q15,q12,#4 |
| bic r10,r7,r5 |
| add r3,r3,r9 |
| vadd.i32 q11,q12,q12 |
| and r11,r6,r5 |
| ldr r9,[sp,#60] |
| vsri.32 q11,q12,#31 |
| add r3,r3,r4,ror#27 |
| eor r11,r11,r10 |
| mov r5,r5,ror#2 |
| vshr.u32 q12,q13,#30 |
| add r3,r3,r11 |
| bic r10,r6,r4 |
| vshl.u32 q13,q13,#2 |
| add r7,r7,r9 |
| and r11,r5,r4 |
| veor q11,q11,q12 |
| ldr r9,[sp,#0] |
| add r7,r7,r3,ror#27 |
| veor q11,q11,q13 |
| eor r11,r11,r10 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| vext.8 q12,q10,q11,#8 |
| bic r10,r5,r3 |
| add r6,r6,r9 |
| and r11,r4,r3 |
| veor q0,q0,q8 |
| ldr r9,[sp,#4] |
| add r6,r6,r7,ror#27 |
| veor q0,q0,q1 |
| eor r11,r11,r10 |
| mov r3,r3,ror#2 |
| vadd.i32 q13,q11,q14 |
| add r6,r6,r11 |
| bic r10,r4,r7 |
| veor q12,q12,q0 |
| add r5,r5,r9 |
| and r11,r3,r7 |
| vshr.u32 q0,q12,#30 |
| ldr r9,[sp,#8] |
| add r5,r5,r6,ror#27 |
| vst1.32 {q13},[r12,:128]! |
| sub r12,r12,#64 |
| eor r11,r11,r10 |
| mov r7,r7,ror#2 |
| vsli.32 q0,q12,#2 |
| add r5,r5,r11 |
| bic r10,r3,r6 |
| add r4,r4,r9 |
| and r11,r7,r6 |
| ldr r9,[sp,#12] |
| add r4,r4,r5,ror#27 |
| eor r11,r11,r10 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| bic r10,r7,r5 |
| add r3,r3,r9 |
| and r11,r6,r5 |
| ldr r9,[sp,#16] |
| add r3,r3,r4,ror#27 |
| eor r11,r11,r10 |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| vext.8 q12,q11,q0,#8 |
| eor r10,r4,r6 |
| add r7,r7,r9 |
| ldr r9,[sp,#20] |
| veor q1,q1,q9 |
| eor r11,r10,r5 |
| add r7,r7,r3,ror#27 |
| veor q1,q1,q2 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| vadd.i32 q13,q0,q14 |
| eor r10,r3,r5 |
| add r6,r6,r9 |
| veor q12,q12,q1 |
| ldr r9,[sp,#24] |
| eor r11,r10,r4 |
| vshr.u32 q1,q12,#30 |
| add r6,r6,r7,ror#27 |
| mov r3,r3,ror#2 |
| vst1.32 {q13},[r12,:128]! |
| add r6,r6,r11 |
| eor r10,r7,r4 |
| vsli.32 q1,q12,#2 |
| add r5,r5,r9 |
| ldr r9,[sp,#28] |
| eor r11,r10,r3 |
| add r5,r5,r6,ror#27 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| eor r10,r6,r3 |
| add r4,r4,r9 |
| ldr r9,[sp,#32] |
| eor r11,r10,r7 |
| add r4,r4,r5,ror#27 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| vext.8 q12,q0,q1,#8 |
| eor r10,r5,r7 |
| add r3,r3,r9 |
| ldr r9,[sp,#36] |
| veor q2,q2,q10 |
| eor r11,r10,r6 |
| add r3,r3,r4,ror#27 |
| veor q2,q2,q3 |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| vadd.i32 q13,q1,q14 |
| eor r10,r4,r6 |
| vld1.32 {d28[],d29[]},[r8,:32]! |
| add r7,r7,r9 |
| veor q12,q12,q2 |
| ldr r9,[sp,#40] |
| eor r11,r10,r5 |
| vshr.u32 q2,q12,#30 |
| add r7,r7,r3,ror#27 |
| mov r4,r4,ror#2 |
| vst1.32 {q13},[r12,:128]! |
| add r7,r7,r11 |
| eor r10,r3,r5 |
| vsli.32 q2,q12,#2 |
| add r6,r6,r9 |
| ldr r9,[sp,#44] |
| eor r11,r10,r4 |
| add r6,r6,r7,ror#27 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| eor r10,r7,r4 |
| add r5,r5,r9 |
| ldr r9,[sp,#48] |
| eor r11,r10,r3 |
| add r5,r5,r6,ror#27 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| vext.8 q12,q1,q2,#8 |
| eor r10,r6,r3 |
| add r4,r4,r9 |
| ldr r9,[sp,#52] |
| veor q3,q3,q11 |
| eor r11,r10,r7 |
| add r4,r4,r5,ror#27 |
| veor q3,q3,q8 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| vadd.i32 q13,q2,q14 |
| eor r10,r5,r7 |
| add r3,r3,r9 |
| veor q12,q12,q3 |
| ldr r9,[sp,#56] |
| eor r11,r10,r6 |
| vshr.u32 q3,q12,#30 |
| add r3,r3,r4,ror#27 |
| mov r5,r5,ror#2 |
| vst1.32 {q13},[r12,:128]! |
| add r3,r3,r11 |
| eor r10,r4,r6 |
| vsli.32 q3,q12,#2 |
| add r7,r7,r9 |
| ldr r9,[sp,#60] |
| eor r11,r10,r5 |
| add r7,r7,r3,ror#27 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| eor r10,r3,r5 |
| add r6,r6,r9 |
| ldr r9,[sp,#0] |
| eor r11,r10,r4 |
| add r6,r6,r7,ror#27 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| vext.8 q12,q2,q3,#8 |
| eor r10,r7,r4 |
| add r5,r5,r9 |
| ldr r9,[sp,#4] |
| veor q8,q8,q0 |
| eor r11,r10,r3 |
| add r5,r5,r6,ror#27 |
| veor q8,q8,q9 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| vadd.i32 q13,q3,q14 |
| eor r10,r6,r3 |
| add r4,r4,r9 |
| veor q12,q12,q8 |
| ldr r9,[sp,#8] |
| eor r11,r10,r7 |
| vshr.u32 q8,q12,#30 |
| add r4,r4,r5,ror#27 |
| mov r6,r6,ror#2 |
| vst1.32 {q13},[r12,:128]! |
| sub r12,r12,#64 |
| add r4,r4,r11 |
| eor r10,r5,r7 |
| vsli.32 q8,q12,#2 |
| add r3,r3,r9 |
| ldr r9,[sp,#12] |
| eor r11,r10,r6 |
| add r3,r3,r4,ror#27 |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| eor r10,r4,r6 |
| add r7,r7,r9 |
| ldr r9,[sp,#16] |
| eor r11,r10,r5 |
| add r7,r7,r3,ror#27 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| vext.8 q12,q3,q8,#8 |
| eor r10,r3,r5 |
| add r6,r6,r9 |
| ldr r9,[sp,#20] |
| veor q9,q9,q1 |
| eor r11,r10,r4 |
| add r6,r6,r7,ror#27 |
| veor q9,q9,q10 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| vadd.i32 q13,q8,q14 |
| eor r10,r7,r4 |
| add r5,r5,r9 |
| veor q12,q12,q9 |
| ldr r9,[sp,#24] |
| eor r11,r10,r3 |
| vshr.u32 q9,q12,#30 |
| add r5,r5,r6,ror#27 |
| mov r7,r7,ror#2 |
| vst1.32 {q13},[r12,:128]! |
| add r5,r5,r11 |
| eor r10,r6,r3 |
| vsli.32 q9,q12,#2 |
| add r4,r4,r9 |
| ldr r9,[sp,#28] |
| eor r11,r10,r7 |
| add r4,r4,r5,ror#27 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| eor r10,r5,r7 |
| add r3,r3,r9 |
| ldr r9,[sp,#32] |
| eor r11,r10,r6 |
| add r3,r3,r4,ror#27 |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| vext.8 q12,q8,q9,#8 |
| add r7,r7,r9 |
| and r10,r5,r6 |
| ldr r9,[sp,#36] |
| veor q10,q10,q2 |
| add r7,r7,r3,ror#27 |
| eor r11,r5,r6 |
| veor q10,q10,q11 |
| add r7,r7,r10 |
| and r11,r11,r4 |
| vadd.i32 q13,q9,q14 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| veor q12,q12,q10 |
| add r6,r6,r9 |
| and r10,r4,r5 |
| vshr.u32 q10,q12,#30 |
| ldr r9,[sp,#40] |
| add r6,r6,r7,ror#27 |
| vst1.32 {q13},[r12,:128]! |
| eor r11,r4,r5 |
| add r6,r6,r10 |
| vsli.32 q10,q12,#2 |
| and r11,r11,r3 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| add r5,r5,r9 |
| and r10,r3,r4 |
| ldr r9,[sp,#44] |
| add r5,r5,r6,ror#27 |
| eor r11,r3,r4 |
| add r5,r5,r10 |
| and r11,r11,r7 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| add r4,r4,r9 |
| and r10,r7,r3 |
| ldr r9,[sp,#48] |
| add r4,r4,r5,ror#27 |
| eor r11,r7,r3 |
| add r4,r4,r10 |
| and r11,r11,r6 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| vext.8 q12,q9,q10,#8 |
| add r3,r3,r9 |
| and r10,r6,r7 |
| ldr r9,[sp,#52] |
| veor q11,q11,q3 |
| add r3,r3,r4,ror#27 |
| eor r11,r6,r7 |
| veor q11,q11,q0 |
| add r3,r3,r10 |
| and r11,r11,r5 |
| vadd.i32 q13,q10,q14 |
| mov r5,r5,ror#2 |
| vld1.32 {d28[],d29[]},[r8,:32]! |
| add r3,r3,r11 |
| veor q12,q12,q11 |
| add r7,r7,r9 |
| and r10,r5,r6 |
| vshr.u32 q11,q12,#30 |
| ldr r9,[sp,#56] |
| add r7,r7,r3,ror#27 |
| vst1.32 {q13},[r12,:128]! |
| eor r11,r5,r6 |
| add r7,r7,r10 |
| vsli.32 q11,q12,#2 |
| and r11,r11,r4 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| add r6,r6,r9 |
| and r10,r4,r5 |
| ldr r9,[sp,#60] |
| add r6,r6,r7,ror#27 |
| eor r11,r4,r5 |
| add r6,r6,r10 |
| and r11,r11,r3 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| add r5,r5,r9 |
| and r10,r3,r4 |
| ldr r9,[sp,#0] |
| add r5,r5,r6,ror#27 |
| eor r11,r3,r4 |
| add r5,r5,r10 |
| and r11,r11,r7 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| vext.8 q12,q10,q11,#8 |
| add r4,r4,r9 |
| and r10,r7,r3 |
| ldr r9,[sp,#4] |
| veor q0,q0,q8 |
| add r4,r4,r5,ror#27 |
| eor r11,r7,r3 |
| veor q0,q0,q1 |
| add r4,r4,r10 |
| and r11,r11,r6 |
| vadd.i32 q13,q11,q14 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| veor q12,q12,q0 |
| add r3,r3,r9 |
| and r10,r6,r7 |
| vshr.u32 q0,q12,#30 |
| ldr r9,[sp,#8] |
| add r3,r3,r4,ror#27 |
| vst1.32 {q13},[r12,:128]! |
| sub r12,r12,#64 |
| eor r11,r6,r7 |
| add r3,r3,r10 |
| vsli.32 q0,q12,#2 |
| and r11,r11,r5 |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| add r7,r7,r9 |
| and r10,r5,r6 |
| ldr r9,[sp,#12] |
| add r7,r7,r3,ror#27 |
| eor r11,r5,r6 |
| add r7,r7,r10 |
| and r11,r11,r4 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| add r6,r6,r9 |
| and r10,r4,r5 |
| ldr r9,[sp,#16] |
| add r6,r6,r7,ror#27 |
| eor r11,r4,r5 |
| add r6,r6,r10 |
| and r11,r11,r3 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| vext.8 q12,q11,q0,#8 |
| add r5,r5,r9 |
| and r10,r3,r4 |
| ldr r9,[sp,#20] |
| veor q1,q1,q9 |
| add r5,r5,r6,ror#27 |
| eor r11,r3,r4 |
| veor q1,q1,q2 |
| add r5,r5,r10 |
| and r11,r11,r7 |
| vadd.i32 q13,q0,q14 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| veor q12,q12,q1 |
| add r4,r4,r9 |
| and r10,r7,r3 |
| vshr.u32 q1,q12,#30 |
| ldr r9,[sp,#24] |
| add r4,r4,r5,ror#27 |
| vst1.32 {q13},[r12,:128]! |
| eor r11,r7,r3 |
| add r4,r4,r10 |
| vsli.32 q1,q12,#2 |
| and r11,r11,r6 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| add r3,r3,r9 |
| and r10,r6,r7 |
| ldr r9,[sp,#28] |
| add r3,r3,r4,ror#27 |
| eor r11,r6,r7 |
| add r3,r3,r10 |
| and r11,r11,r5 |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| add r7,r7,r9 |
| and r10,r5,r6 |
| ldr r9,[sp,#32] |
| add r7,r7,r3,ror#27 |
| eor r11,r5,r6 |
| add r7,r7,r10 |
| and r11,r11,r4 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| vext.8 q12,q0,q1,#8 |
| add r6,r6,r9 |
| and r10,r4,r5 |
| ldr r9,[sp,#36] |
| veor q2,q2,q10 |
| add r6,r6,r7,ror#27 |
| eor r11,r4,r5 |
| veor q2,q2,q3 |
| add r6,r6,r10 |
| and r11,r11,r3 |
| vadd.i32 q13,q1,q14 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| veor q12,q12,q2 |
| add r5,r5,r9 |
| and r10,r3,r4 |
| vshr.u32 q2,q12,#30 |
| ldr r9,[sp,#40] |
| add r5,r5,r6,ror#27 |
| vst1.32 {q13},[r12,:128]! |
| eor r11,r3,r4 |
| add r5,r5,r10 |
| vsli.32 q2,q12,#2 |
| and r11,r11,r7 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| add r4,r4,r9 |
| and r10,r7,r3 |
| ldr r9,[sp,#44] |
| add r4,r4,r5,ror#27 |
| eor r11,r7,r3 |
| add r4,r4,r10 |
| and r11,r11,r6 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| add r3,r3,r9 |
| and r10,r6,r7 |
| ldr r9,[sp,#48] |
| add r3,r3,r4,ror#27 |
| eor r11,r6,r7 |
| add r3,r3,r10 |
| and r11,r11,r5 |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| vext.8 q12,q1,q2,#8 |
| eor r10,r4,r6 |
| add r7,r7,r9 |
| ldr r9,[sp,#52] |
| veor q3,q3,q11 |
| eor r11,r10,r5 |
| add r7,r7,r3,ror#27 |
| veor q3,q3,q8 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| vadd.i32 q13,q2,q14 |
| eor r10,r3,r5 |
| add r6,r6,r9 |
| veor q12,q12,q3 |
| ldr r9,[sp,#56] |
| eor r11,r10,r4 |
| vshr.u32 q3,q12,#30 |
| add r6,r6,r7,ror#27 |
| mov r3,r3,ror#2 |
| vst1.32 {q13},[r12,:128]! |
| add r6,r6,r11 |
| eor r10,r7,r4 |
| vsli.32 q3,q12,#2 |
| add r5,r5,r9 |
| ldr r9,[sp,#60] |
| eor r11,r10,r3 |
| add r5,r5,r6,ror#27 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| eor r10,r6,r3 |
| add r4,r4,r9 |
| ldr r9,[sp,#0] |
| eor r11,r10,r7 |
| add r4,r4,r5,ror#27 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| vadd.i32 q13,q3,q14 |
| eor r10,r5,r7 |
| add r3,r3,r9 |
| vst1.32 {q13},[r12,:128]! |
| sub r12,r12,#64 |
| teq r1,r2 |
| sub r8,r8,#16 |
| it eq |
| subeq r1,r1,#64 |
| vld1.8 {q0,q1},[r1]! |
| ldr r9,[sp,#4] |
| eor r11,r10,r6 |
| vld1.8 {q2,q3},[r1]! |
| add r3,r3,r4,ror#27 |
| mov r5,r5,ror#2 |
| vld1.32 {d28[],d29[]},[r8,:32]! |
| add r3,r3,r11 |
| eor r10,r4,r6 |
| vrev32.8 q0,q0 |
| add r7,r7,r9 |
| ldr r9,[sp,#8] |
| eor r11,r10,r5 |
| add r7,r7,r3,ror#27 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| eor r10,r3,r5 |
| add r6,r6,r9 |
| ldr r9,[sp,#12] |
| eor r11,r10,r4 |
| add r6,r6,r7,ror#27 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| eor r10,r7,r4 |
| add r5,r5,r9 |
| ldr r9,[sp,#16] |
| eor r11,r10,r3 |
| add r5,r5,r6,ror#27 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| vrev32.8 q1,q1 |
| eor r10,r6,r3 |
| add r4,r4,r9 |
| vadd.i32 q8,q0,q14 |
| ldr r9,[sp,#20] |
| eor r11,r10,r7 |
| vst1.32 {q8},[r12,:128]! |
| add r4,r4,r5,ror#27 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| eor r10,r5,r7 |
| add r3,r3,r9 |
| ldr r9,[sp,#24] |
| eor r11,r10,r6 |
| add r3,r3,r4,ror#27 |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| eor r10,r4,r6 |
| add r7,r7,r9 |
| ldr r9,[sp,#28] |
| eor r11,r10,r5 |
| add r7,r7,r3,ror#27 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| eor r10,r3,r5 |
| add r6,r6,r9 |
| ldr r9,[sp,#32] |
| eor r11,r10,r4 |
| add r6,r6,r7,ror#27 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| vrev32.8 q2,q2 |
| eor r10,r7,r4 |
| add r5,r5,r9 |
| vadd.i32 q9,q1,q14 |
| ldr r9,[sp,#36] |
| eor r11,r10,r3 |
| vst1.32 {q9},[r12,:128]! |
| add r5,r5,r6,ror#27 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| eor r10,r6,r3 |
| add r4,r4,r9 |
| ldr r9,[sp,#40] |
| eor r11,r10,r7 |
| add r4,r4,r5,ror#27 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| eor r10,r5,r7 |
| add r3,r3,r9 |
| ldr r9,[sp,#44] |
| eor r11,r10,r6 |
| add r3,r3,r4,ror#27 |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| eor r10,r4,r6 |
| add r7,r7,r9 |
| ldr r9,[sp,#48] |
| eor r11,r10,r5 |
| add r7,r7,r3,ror#27 |
| mov r4,r4,ror#2 |
| add r7,r7,r11 |
| vrev32.8 q3,q3 |
| eor r10,r3,r5 |
| add r6,r6,r9 |
| vadd.i32 q10,q2,q14 |
| ldr r9,[sp,#52] |
| eor r11,r10,r4 |
| vst1.32 {q10},[r12,:128]! |
| add r6,r6,r7,ror#27 |
| mov r3,r3,ror#2 |
| add r6,r6,r11 |
| eor r10,r7,r4 |
| add r5,r5,r9 |
| ldr r9,[sp,#56] |
| eor r11,r10,r3 |
| add r5,r5,r6,ror#27 |
| mov r7,r7,ror#2 |
| add r5,r5,r11 |
| eor r10,r6,r3 |
| add r4,r4,r9 |
| ldr r9,[sp,#60] |
| eor r11,r10,r7 |
| add r4,r4,r5,ror#27 |
| mov r6,r6,ror#2 |
| add r4,r4,r11 |
| eor r10,r5,r7 |
| add r3,r3,r9 |
| eor r11,r10,r6 |
| add r3,r3,r4,ror#27 |
| mov r5,r5,ror#2 |
| add r3,r3,r11 |
| ldmia r0,{r9,r10,r11,r12} @ accumulate context |
| add r3,r3,r9 |
| ldr r9,[r0,#16] |
| add r4,r4,r10 |
| add r5,r5,r11 |
| add r6,r6,r12 |
| it eq |
| moveq sp,r14 |
| add r7,r7,r9 |
| it ne |
| ldrne r9,[sp] |
| stmia r0,{r3,r4,r5,r6,r7} |
| itt ne |
| addne r12,sp,#3*16 |
| bne Loop_neon |
| |
| @ vldmia sp!,{d8-d15} |
| ldmia sp!,{r4,r5,r6,r7,r8,r9,r10,r11,r12,pc} |
| |
| #endif |
| #if __ARM_MAX_ARCH__>=7 |
| |
| # if defined(__thumb2__) |
| # define INST(a,b,c,d) .byte c,d|0xf,a,b |
| # else |
| # define INST(a,b,c,d) .byte a,b,c,d|0x10 |
| # endif |
| |
| #ifdef __thumb2__ |
| .thumb_func sha1_block_data_order_armv8 |
| #endif |
| .align 5 |
| sha1_block_data_order_armv8: |
| LARMv8: |
| vstmdb sp!,{d8,d9,d10,d11,d12,d13,d14,d15} @ ABI specification says so |
| |
| veor q1,q1,q1 |
| adr r3,LK_00_19 |
| vld1.32 {q0},[r0]! |
| vld1.32 {d2[0]},[r0] |
| sub r0,r0,#16 |
| vld1.32 {d16[],d17[]},[r3,:32]! |
| vld1.32 {d18[],d19[]},[r3,:32]! |
| vld1.32 {d20[],d21[]},[r3,:32]! |
| vld1.32 {d22[],d23[]},[r3,:32] |
| |
| Loop_v8: |
| vld1.8 {q4,q5},[r1]! |
| vld1.8 {q6,q7},[r1]! |
| vrev32.8 q4,q4 |
| vrev32.8 q5,q5 |
| |
| vadd.i32 q12,q8,q4 |
| vrev32.8 q6,q6 |
| vmov q14,q0 @ offload |
| subs r2,r2,#1 |
| |
| vadd.i32 q13,q8,q5 |
| vrev32.8 q7,q7 |
| INST(0xc0,0x62,0xb9,0xf3) @ sha1h q3,q0 @ 0 |
| INST(0x68,0x0c,0x02,0xe2) @ sha1c q0,q1,q12 |
| vadd.i32 q12,q8,q6 |
| INST(0x4c,0x8c,0x3a,0xe2) @ sha1su0 q4,q5,q6 |
| INST(0xc0,0x42,0xb9,0xf3) @ sha1h q2,q0 @ 1 |
| INST(0x6a,0x0c,0x06,0xe2) @ sha1c q0,q3,q13 |
| vadd.i32 q13,q8,q7 |
| INST(0x8e,0x83,0xba,0xf3) @ sha1su1 q4,q7 |
| INST(0x4e,0xac,0x3c,0xe2) @ sha1su0 q5,q6,q7 |
| INST(0xc0,0x62,0xb9,0xf3) @ sha1h q3,q0 @ 2 |
| INST(0x68,0x0c,0x04,0xe2) @ sha1c q0,q2,q12 |
| vadd.i32 q12,q8,q4 |
| INST(0x88,0xa3,0xba,0xf3) @ sha1su1 q5,q4 |
| INST(0x48,0xcc,0x3e,0xe2) @ sha1su0 q6,q7,q4 |
| INST(0xc0,0x42,0xb9,0xf3) @ sha1h q2,q0 @ 3 |
| INST(0x6a,0x0c,0x06,0xe2) @ sha1c q0,q3,q13 |
| vadd.i32 q13,q9,q5 |
| INST(0x8a,0xc3,0xba,0xf3) @ sha1su1 q6,q5 |
| INST(0x4a,0xec,0x38,0xe2) @ sha1su0 q7,q4,q5 |
| INST(0xc0,0x62,0xb9,0xf3) @ sha1h q3,q0 @ 4 |
| INST(0x68,0x0c,0x04,0xe2) @ sha1c q0,q2,q12 |
| vadd.i32 q12,q9,q6 |
| INST(0x8c,0xe3,0xba,0xf3) @ sha1su1 q7,q6 |
| INST(0x4c,0x8c,0x3a,0xe2) @ sha1su0 q4,q5,q6 |
| INST(0xc0,0x42,0xb9,0xf3) @ sha1h q2,q0 @ 5 |
| INST(0x6a,0x0c,0x16,0xe2) @ sha1p q0,q3,q13 |
| vadd.i32 q13,q9,q7 |
| INST(0x8e,0x83,0xba,0xf3) @ sha1su1 q4,q7 |
| INST(0x4e,0xac,0x3c,0xe2) @ sha1su0 q5,q6,q7 |
| INST(0xc0,0x62,0xb9,0xf3) @ sha1h q3,q0 @ 6 |
| INST(0x68,0x0c,0x14,0xe2) @ sha1p q0,q2,q12 |
| vadd.i32 q12,q9,q4 |
| INST(0x88,0xa3,0xba,0xf3) @ sha1su1 q5,q4 |
| INST(0x48,0xcc,0x3e,0xe2) @ sha1su0 q6,q7,q4 |
| INST(0xc0,0x42,0xb9,0xf3) @ sha1h q2,q0 @ 7 |
| INST(0x6a,0x0c,0x16,0xe2) @ sha1p q0,q3,q13 |
| vadd.i32 q13,q9,q5 |
| INST(0x8a,0xc3,0xba,0xf3) @ sha1su1 q6,q5 |
| INST(0x4a,0xec,0x38,0xe2) @ sha1su0 q7,q4,q5 |
| INST(0xc0,0x62,0xb9,0xf3) @ sha1h q3,q0 @ 8 |
| INST(0x68,0x0c,0x14,0xe2) @ sha1p q0,q2,q12 |
| vadd.i32 q12,q10,q6 |
| INST(0x8c,0xe3,0xba,0xf3) @ sha1su1 q7,q6 |
| INST(0x4c,0x8c,0x3a,0xe2) @ sha1su0 q4,q5,q6 |
| INST(0xc0,0x42,0xb9,0xf3) @ sha1h q2,q0 @ 9 |
| INST(0x6a,0x0c,0x16,0xe2) @ sha1p q0,q3,q13 |
| vadd.i32 q13,q10,q7 |
| INST(0x8e,0x83,0xba,0xf3) @ sha1su1 q4,q7 |
| INST(0x4e,0xac,0x3c,0xe2) @ sha1su0 q5,q6,q7 |
| INST(0xc0,0x62,0xb9,0xf3) @ sha1h q3,q0 @ 10 |
| INST(0x68,0x0c,0x24,0xe2) @ sha1m q0,q2,q12 |
| vadd.i32 q12,q10,q4 |
| INST(0x88,0xa3,0xba,0xf3) @ sha1su1 q5,q4 |
| INST(0x48,0xcc,0x3e,0xe2) @ sha1su0 q6,q7,q4 |
| INST(0xc0,0x42,0xb9,0xf3) @ sha1h q2,q0 @ 11 |
| INST(0x6a,0x0c,0x26,0xe2) @ sha1m q0,q3,q13 |
| vadd.i32 q13,q10,q5 |
| INST(0x8a,0xc3,0xba,0xf3) @ sha1su1 q6,q5 |
| INST(0x4a,0xec,0x38,0xe2) @ sha1su0 q7,q4,q5 |
| INST(0xc0,0x62,0xb9,0xf3) @ sha1h q3,q0 @ 12 |
| INST(0x68,0x0c,0x24,0xe2) @ sha1m q0,q2,q12 |
| vadd.i32 q12,q10,q6 |
| INST(0x8c,0xe3,0xba,0xf3) @ sha1su1 q7,q6 |
| INST(0x4c,0x8c,0x3a,0xe2) @ sha1su0 q4,q5,q6 |
| INST(0xc0,0x42,0xb9,0xf3) @ sha1h q2,q0 @ 13 |
| INST(0x6a,0x0c,0x26,0xe2) @ sha1m q0,q3,q13 |
| vadd.i32 q13,q11,q7 |
| INST(0x8e,0x83,0xba,0xf3) @ sha1su1 q4,q7 |
| INST(0x4e,0xac,0x3c,0xe2) @ sha1su0 q5,q6,q7 |
| INST(0xc0,0x62,0xb9,0xf3) @ sha1h q3,q0 @ 14 |
| INST(0x68,0x0c,0x24,0xe2) @ sha1m q0,q2,q12 |
| vadd.i32 q12,q11,q4 |
| INST(0x88,0xa3,0xba,0xf3) @ sha1su1 q5,q4 |
| INST(0x48,0xcc,0x3e,0xe2) @ sha1su0 q6,q7,q4 |
| INST(0xc0,0x42,0xb9,0xf3) @ sha1h q2,q0 @ 15 |
| INST(0x6a,0x0c,0x16,0xe2) @ sha1p q0,q3,q13 |
| vadd.i32 q13,q11,q5 |
| INST(0x8a,0xc3,0xba,0xf3) @ sha1su1 q6,q5 |
| INST(0x4a,0xec,0x38,0xe2) @ sha1su0 q7,q4,q5 |
| INST(0xc0,0x62,0xb9,0xf3) @ sha1h q3,q0 @ 16 |
| INST(0x68,0x0c,0x14,0xe2) @ sha1p q0,q2,q12 |
| vadd.i32 q12,q11,q6 |
| INST(0x8c,0xe3,0xba,0xf3) @ sha1su1 q7,q6 |
| INST(0xc0,0x42,0xb9,0xf3) @ sha1h q2,q0 @ 17 |
| INST(0x6a,0x0c,0x16,0xe2) @ sha1p q0,q3,q13 |
| vadd.i32 q13,q11,q7 |
| |
| INST(0xc0,0x62,0xb9,0xf3) @ sha1h q3,q0 @ 18 |
| INST(0x68,0x0c,0x14,0xe2) @ sha1p q0,q2,q12 |
| |
| INST(0xc0,0x42,0xb9,0xf3) @ sha1h q2,q0 @ 19 |
| INST(0x6a,0x0c,0x16,0xe2) @ sha1p q0,q3,q13 |
| |
| vadd.i32 q1,q1,q2 |
| vadd.i32 q0,q0,q14 |
| bne Loop_v8 |
| |
| vst1.32 {q0},[r0]! |
| vst1.32 {d2[0]},[r0] |
| |
| vldmia sp!,{d8,d9,d10,d11,d12,d13,d14,d15} |
| bx lr @ bx lr |
| |
| #endif |
| #if __ARM_MAX_ARCH__>=7 |
| .comm _OPENSSL_armcap_P,4 |
| .non_lazy_symbol_pointer |
| OPENSSL_armcap_P: |
| .indirect_symbol _OPENSSL_armcap_P |
| .long 0 |
| .private_extern _OPENSSL_armcap_P |
| #endif |
