feedc0de/wolfssl
1
0
Fork 0
forked from wolfSSL/wolfssl
Code Pull Requests Activity

ARMv7a SHA256: explicit size on vmov

Browse Source 
For SHA256 using NEON on ARM32, change vmov instruction that is moving
from scalar to general-purpose register to have explicit size (32 bits).
May be needed by some compilers.
This commit is contained in:
Sean Parkinson
2023-10-17 10:26:26 +10:00
parent 1abaa94120
commit 4ac795961b
2 changed files with 64 additions and 64 deletions
Download Patch File Download Diff File Expand all files Collapse all files

64
wolfcrypt/src/port/arm/armv8-32-sha256-asm.S
View File

@ -1897,7 +1897,7 @@ L_SHA256_transform_neon_len_begin:
# Start of 16 rounds
L_SHA256_transform_neon_len_start:
# Round 0
vmov r10, d0[0]
vmov.32 r10, d0[0]
ror r0, r6, #6
eor r1, r7, r8
eor r0, r0, r6, ror #11
@ -1920,7 +1920,7 @@ L_SHA256_transform_neon_len_start:
add r9, r9, r0
add r9, r9, r1
# Round 1
vmov r10, d0[1]
vmov.32 r10, d0[1]
# Calc new W[0]-W[1]
vext.8 d10, d0, d1, #4
ror r0, r5, #6
@ -1963,7 +1963,7 @@ L_SHA256_transform_neon_len_start:
add r8, r8, r0
add r8, r8, r1
# Round 2
vmov r10, d1[0]
vmov.32 r10, d1[0]
ror r0, r4, #6
eor r1, r5, r6
eor r0, r0, r4, ror #11
@ -1986,7 +1986,7 @@ L_SHA256_transform_neon_len_start:
add r7, r7, r0
add r7, r7, r1
# Round 3
vmov r10, d1[1]
vmov.32 r10, d1[1]
# Calc new W[2]-W[3]
vext.8 d10, d1, d2, #4
ror r0, r3, #6
@ -2029,7 +2029,7 @@ L_SHA256_transform_neon_len_start:
add r6, r6, r0
add r6, r6, r1
# Round 4
vmov r10, d2[0]
vmov.32 r10, d2[0]
ror r0, r2, #6
eor r1, r3, r4
eor r0, r0, r2, ror #11
@ -2052,7 +2052,7 @@ L_SHA256_transform_neon_len_start:
add r5, r5, r0
add r5, r5, r1
# Round 5
vmov r10, d2[1]
vmov.32 r10, d2[1]
# Calc new W[4]-W[5]
vext.8 d10, d2, d3, #4
ror r0, r9, #6
@ -2095,7 +2095,7 @@ L_SHA256_transform_neon_len_start:
add r4, r4, r0
add r4, r4, r1
# Round 6
vmov r10, d3[0]
vmov.32 r10, d3[0]
ror r0, r8, #6
eor r1, r9, r2
eor r0, r0, r8, ror #11
@ -2118,7 +2118,7 @@ L_SHA256_transform_neon_len_start:
add r3, r3, r0
add r3, r3, r1
# Round 7
vmov r10, d3[1]
vmov.32 r10, d3[1]
# Calc new W[6]-W[7]
vext.8 d10, d3, d4, #4
ror r0, r7, #6
@ -2161,7 +2161,7 @@ L_SHA256_transform_neon_len_start:
add r2, r2, r0
add r2, r2, r1
# Round 8
vmov r10, d4[0]
vmov.32 r10, d4[0]
ror r0, r6, #6
eor r1, r7, r8
eor r0, r0, r6, ror #11
@ -2184,7 +2184,7 @@ L_SHA256_transform_neon_len_start:
add r9, r9, r0
add r9, r9, r1
# Round 9
vmov r10, d4[1]
vmov.32 r10, d4[1]
# Calc new W[8]-W[9]
vext.8 d10, d4, d5, #4
ror r0, r5, #6
@ -2227,7 +2227,7 @@ L_SHA256_transform_neon_len_start:
add r8, r8, r0
add r8, r8, r1
# Round 10
vmov r10, d5[0]
vmov.32 r10, d5[0]
ror r0, r4, #6
eor r1, r5, r6
eor r0, r0, r4, ror #11
@ -2250,7 +2250,7 @@ L_SHA256_transform_neon_len_start:
add r7, r7, r0
add r7, r7, r1
# Round 11
vmov r10, d5[1]
vmov.32 r10, d5[1]
# Calc new W[10]-W[11]
vext.8 d10, d5, d6, #4
ror r0, r3, #6
@ -2293,7 +2293,7 @@ L_SHA256_transform_neon_len_start:
add r6, r6, r0
add r6, r6, r1
# Round 12
vmov r10, d6[0]
vmov.32 r10, d6[0]
ror r0, r2, #6
eor r1, r3, r4
eor r0, r0, r2, ror #11
@ -2316,7 +2316,7 @@ L_SHA256_transform_neon_len_start:
add r5, r5, r0
add r5, r5, r1
# Round 13
vmov r10, d6[1]
vmov.32 r10, d6[1]
# Calc new W[12]-W[13]
vext.8 d10, d6, d7, #4
ror r0, r9, #6
@ -2359,7 +2359,7 @@ L_SHA256_transform_neon_len_start:
add r4, r4, r0
add r4, r4, r1
# Round 14
vmov r10, d7[0]
vmov.32 r10, d7[0]
ror r0, r8, #6
eor r1, r9, r2
eor r0, r0, r8, ror #11
@ -2382,7 +2382,7 @@ L_SHA256_transform_neon_len_start:
add r3, r3, r0
add r3, r3, r1
# Round 15
vmov r10, d7[1]
vmov.32 r10, d7[1]
# Calc new W[14]-W[15]
vext.8 d10, d7, d0, #4
ror r0, r7, #6
@ -2428,7 +2428,7 @@ L_SHA256_transform_neon_len_start:
subs lr, lr, #1
bne L_SHA256_transform_neon_len_start
# Round 0
vmov r10, d0[0]
vmov.32 r10, d0[0]
ror r0, r6, #6
eor r1, r7, r8
eor r0, r0, r6, ror #11
@ -2451,7 +2451,7 @@ L_SHA256_transform_neon_len_start:
add r9, r9, r0
add r9, r9, r1
# Round 1
vmov r10, d0[1]
vmov.32 r10, d0[1]
ror r0, r5, #6
eor r1, r6, r7
eor r0, r0, r5, ror #11
@ -2474,7 +2474,7 @@ L_SHA256_transform_neon_len_start:
add r8, r8, r0
add r8, r8, r1
# Round 2
vmov r10, d1[0]
vmov.32 r10, d1[0]
ror r0, r4, #6
eor r1, r5, r6
eor r0, r0, r4, ror #11
@ -2497,7 +2497,7 @@ L_SHA256_transform_neon_len_start:
add r7, r7, r0
add r7, r7, r1
# Round 3
vmov r10, d1[1]
vmov.32 r10, d1[1]
ror r0, r3, #6
eor r1, r4, r5
eor r0, r0, r3, ror #11
@ -2520,7 +2520,7 @@ L_SHA256_transform_neon_len_start:
add r6, r6, r0
add r6, r6, r1
# Round 4
vmov r10, d2[0]
vmov.32 r10, d2[0]
ror r0, r2, #6
eor r1, r3, r4
eor r0, r0, r2, ror #11
@ -2543,7 +2543,7 @@ L_SHA256_transform_neon_len_start:
add r5, r5, r0
add r5, r5, r1
# Round 5
vmov r10, d2[1]
vmov.32 r10, d2[1]
ror r0, r9, #6
eor r1, r2, r3
eor r0, r0, r9, ror #11
@ -2566,7 +2566,7 @@ L_SHA256_transform_neon_len_start:
add r4, r4, r0
add r4, r4, r1
# Round 6
vmov r10, d3[0]
vmov.32 r10, d3[0]
ror r0, r8, #6
eor r1, r9, r2
eor r0, r0, r8, ror #11
@ -2589,7 +2589,7 @@ L_SHA256_transform_neon_len_start:
add r3, r3, r0
add r3, r3, r1
# Round 7
vmov r10, d3[1]
vmov.32 r10, d3[1]
ror r0, r7, #6
eor r1, r8, r9
eor r0, r0, r7, ror #11
@ -2612,7 +2612,7 @@ L_SHA256_transform_neon_len_start:
add r2, r2, r0
add r2, r2, r1
# Round 8
vmov r10, d4[0]
vmov.32 r10, d4[0]
ror r0, r6, #6
eor r1, r7, r8
eor r0, r0, r6, ror #11
@ -2635,7 +2635,7 @@ L_SHA256_transform_neon_len_start:
add r9, r9, r0
add r9, r9, r1
# Round 9
vmov r10, d4[1]
vmov.32 r10, d4[1]
ror r0, r5, #6
eor r1, r6, r7
eor r0, r0, r5, ror #11
@ -2658,7 +2658,7 @@ L_SHA256_transform_neon_len_start:
add r8, r8, r0
add r8, r8, r1
# Round 10
vmov r10, d5[0]
vmov.32 r10, d5[0]
ror r0, r4, #6
eor r1, r5, r6
eor r0, r0, r4, ror #11
@ -2681,7 +2681,7 @@ L_SHA256_transform_neon_len_start:
add r7, r7, r0
add r7, r7, r1
# Round 11
vmov r10, d5[1]
vmov.32 r10, d5[1]
ror r0, r3, #6
eor r1, r4, r5
eor r0, r0, r3, ror #11
@ -2704,7 +2704,7 @@ L_SHA256_transform_neon_len_start:
add r6, r6, r0
add r6, r6, r1
# Round 12
vmov r10, d6[0]
vmov.32 r10, d6[0]
ror r0, r2, #6
eor r1, r3, r4
eor r0, r0, r2, ror #11
@ -2727,7 +2727,7 @@ L_SHA256_transform_neon_len_start:
add r5, r5, r0
add r5, r5, r1
# Round 13
vmov r10, d6[1]
vmov.32 r10, d6[1]
ror r0, r9, #6
eor r1, r2, r3
eor r0, r0, r9, ror #11
@ -2750,7 +2750,7 @@ L_SHA256_transform_neon_len_start:
add r4, r4, r0
add r4, r4, r1
# Round 14
vmov r10, d7[0]
vmov.32 r10, d7[0]
ror r0, r8, #6
eor r1, r9, r2
eor r0, r0, r8, ror #11
@ -2773,7 +2773,7 @@ L_SHA256_transform_neon_len_start:
add r3, r3, r0
add r3, r3, r1
# Round 15
vmov r10, d7[1]
vmov.32 r10, d7[1]
ror r0, r7, #6
eor r1, r8, r9
eor r0, r0, r7, ror #11

64
wolfcrypt/src/port/arm/armv8-32-sha256-asm_c.c
View File

@ -1831,7 +1831,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"\n"
"L_SHA256_transform_neon_len_start_%=: \n\t"
/* Round 0 */
"vmov r10, d0[0]\n\t"
"vmov.32 r10, d0[0]\n\t"
"ror %[sha256], r6, #6\n\t"
"eor %[data], r7, r8\n\t"
"eor %[sha256], %[sha256], r6, ror #11\n\t"
@ -1854,7 +1854,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r9, r9, %[sha256]\n\t"
"add r9, r9, %[data]\n\t"
/* Round 1 */
"vmov r10, d0[1]\n\t"
"vmov.32 r10, d0[1]\n\t"
/* Calc new W[0]-W[1] */
"vext.8 d10, d0, d1, #4\n\t"
"ror %[sha256], r5, #6\n\t"
@ -1897,7 +1897,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r8, r8, %[sha256]\n\t"
"add r8, r8, %[data]\n\t"
/* Round 2 */
"vmov r10, d1[0]\n\t"
"vmov.32 r10, d1[0]\n\t"
"ror %[sha256], r4, #6\n\t"
"eor %[data], r5, r6\n\t"
"eor %[sha256], %[sha256], r4, ror #11\n\t"
@ -1920,7 +1920,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r7, r7, %[sha256]\n\t"
"add r7, r7, %[data]\n\t"
/* Round 3 */
"vmov r10, d1[1]\n\t"
"vmov.32 r10, d1[1]\n\t"
/* Calc new W[2]-W[3] */
"vext.8 d10, d1, d2, #4\n\t"
"ror %[sha256], r3, #6\n\t"
@ -1963,7 +1963,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r6, r6, %[sha256]\n\t"
"add r6, r6, %[data]\n\t"
/* Round 4 */
"vmov r10, d2[0]\n\t"
"vmov.32 r10, d2[0]\n\t"
"ror %[sha256], %[len], #6\n\t"
"eor %[data], r3, r4\n\t"
"eor %[sha256], %[sha256], %[len], ror #11\n\t"
@ -1986,7 +1986,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r5, r5, %[sha256]\n\t"
"add r5, r5, %[data]\n\t"
/* Round 5 */
"vmov r10, d2[1]\n\t"
"vmov.32 r10, d2[1]\n\t"
/* Calc new W[4]-W[5] */
"vext.8 d10, d2, d3, #4\n\t"
"ror %[sha256], r9, #6\n\t"
@ -2029,7 +2029,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r4, r4, %[sha256]\n\t"
"add r4, r4, %[data]\n\t"
/* Round 6 */
"vmov r10, d3[0]\n\t"
"vmov.32 r10, d3[0]\n\t"
"ror %[sha256], r8, #6\n\t"
"eor %[data], r9, %[len]\n\t"
"eor %[sha256], %[sha256], r8, ror #11\n\t"
@ -2052,7 +2052,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r3, r3, %[sha256]\n\t"
"add r3, r3, %[data]\n\t"
/* Round 7 */
"vmov r10, d3[1]\n\t"
"vmov.32 r10, d3[1]\n\t"
/* Calc new W[6]-W[7] */
"vext.8 d10, d3, d4, #4\n\t"
"ror %[sha256], r7, #6\n\t"
@ -2095,7 +2095,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add %[len], %[len], %[sha256]\n\t"
"add %[len], %[len], %[data]\n\t"
/* Round 8 */
"vmov r10, d4[0]\n\t"
"vmov.32 r10, d4[0]\n\t"
"ror %[sha256], r6, #6\n\t"
"eor %[data], r7, r8\n\t"
"eor %[sha256], %[sha256], r6, ror #11\n\t"
@ -2118,7 +2118,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r9, r9, %[sha256]\n\t"
"add r9, r9, %[data]\n\t"
/* Round 9 */
"vmov r10, d4[1]\n\t"
"vmov.32 r10, d4[1]\n\t"
/* Calc new W[8]-W[9] */
"vext.8 d10, d4, d5, #4\n\t"
"ror %[sha256], r5, #6\n\t"
@ -2161,7 +2161,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r8, r8, %[sha256]\n\t"
"add r8, r8, %[data]\n\t"
/* Round 10 */
"vmov r10, d5[0]\n\t"
"vmov.32 r10, d5[0]\n\t"
"ror %[sha256], r4, #6\n\t"
"eor %[data], r5, r6\n\t"
"eor %[sha256], %[sha256], r4, ror #11\n\t"
@ -2184,7 +2184,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r7, r7, %[sha256]\n\t"
"add r7, r7, %[data]\n\t"
/* Round 11 */
"vmov r10, d5[1]\n\t"
"vmov.32 r10, d5[1]\n\t"
/* Calc new W[10]-W[11] */
"vext.8 d10, d5, d6, #4\n\t"
"ror %[sha256], r3, #6\n\t"
@ -2227,7 +2227,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r6, r6, %[sha256]\n\t"
"add r6, r6, %[data]\n\t"
/* Round 12 */
"vmov r10, d6[0]\n\t"
"vmov.32 r10, d6[0]\n\t"
"ror %[sha256], %[len], #6\n\t"
"eor %[data], r3, r4\n\t"
"eor %[sha256], %[sha256], %[len], ror #11\n\t"
@ -2250,7 +2250,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r5, r5, %[sha256]\n\t"
"add r5, r5, %[data]\n\t"
/* Round 13 */
"vmov r10, d6[1]\n\t"
"vmov.32 r10, d6[1]\n\t"
/* Calc new W[12]-W[13] */
"vext.8 d10, d6, d7, #4\n\t"
"ror %[sha256], r9, #6\n\t"
@ -2293,7 +2293,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r4, r4, %[sha256]\n\t"
"add r4, r4, %[data]\n\t"
/* Round 14 */
"vmov r10, d7[0]\n\t"
"vmov.32 r10, d7[0]\n\t"
"ror %[sha256], r8, #6\n\t"
"eor %[data], r9, %[len]\n\t"
"eor %[sha256], %[sha256], r8, ror #11\n\t"
@ -2316,7 +2316,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r3, r3, %[sha256]\n\t"
"add r3, r3, %[data]\n\t"
/* Round 15 */
"vmov r10, d7[1]\n\t"
"vmov.32 r10, d7[1]\n\t"
/* Calc new W[14]-W[15] */
"vext.8 d10, d7, d0, #4\n\t"
"ror %[sha256], r7, #6\n\t"
@ -2362,7 +2362,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"subs lr, lr, #1\n\t"
"bne L_SHA256_transform_neon_len_start_%=\n\t"
/* Round 0 */
"vmov r10, d0[0]\n\t"
"vmov.32 r10, d0[0]\n\t"
"ror %[sha256], r6, #6\n\t"
"eor %[data], r7, r8\n\t"
"eor %[sha256], %[sha256], r6, ror #11\n\t"
@ -2385,7 +2385,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r9, r9, %[sha256]\n\t"
"add r9, r9, %[data]\n\t"
/* Round 1 */
"vmov r10, d0[1]\n\t"
"vmov.32 r10, d0[1]\n\t"
"ror %[sha256], r5, #6\n\t"
"eor %[data], r6, r7\n\t"
"eor %[sha256], %[sha256], r5, ror #11\n\t"
@ -2408,7 +2408,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r8, r8, %[sha256]\n\t"
"add r8, r8, %[data]\n\t"
/* Round 2 */
"vmov r10, d1[0]\n\t"
"vmov.32 r10, d1[0]\n\t"
"ror %[sha256], r4, #6\n\t"
"eor %[data], r5, r6\n\t"
"eor %[sha256], %[sha256], r4, ror #11\n\t"
@ -2431,7 +2431,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r7, r7, %[sha256]\n\t"
"add r7, r7, %[data]\n\t"
/* Round 3 */
"vmov r10, d1[1]\n\t"
"vmov.32 r10, d1[1]\n\t"
"ror %[sha256], r3, #6\n\t"
"eor %[data], r4, r5\n\t"
"eor %[sha256], %[sha256], r3, ror #11\n\t"
@ -2454,7 +2454,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r6, r6, %[sha256]\n\t"
"add r6, r6, %[data]\n\t"
/* Round 4 */
"vmov r10, d2[0]\n\t"
"vmov.32 r10, d2[0]\n\t"
"ror %[sha256], %[len], #6\n\t"
"eor %[data], r3, r4\n\t"
"eor %[sha256], %[sha256], %[len], ror #11\n\t"
@ -2477,7 +2477,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r5, r5, %[sha256]\n\t"
"add r5, r5, %[data]\n\t"
/* Round 5 */
"vmov r10, d2[1]\n\t"
"vmov.32 r10, d2[1]\n\t"
"ror %[sha256], r9, #6\n\t"
"eor %[data], %[len], r3\n\t"
"eor %[sha256], %[sha256], r9, ror #11\n\t"
@ -2500,7 +2500,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r4, r4, %[sha256]\n\t"
"add r4, r4, %[data]\n\t"
/* Round 6 */
"vmov r10, d3[0]\n\t"
"vmov.32 r10, d3[0]\n\t"
"ror %[sha256], r8, #6\n\t"
"eor %[data], r9, %[len]\n\t"
"eor %[sha256], %[sha256], r8, ror #11\n\t"
@ -2523,7 +2523,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r3, r3, %[sha256]\n\t"
"add r3, r3, %[data]\n\t"
/* Round 7 */
"vmov r10, d3[1]\n\t"
"vmov.32 r10, d3[1]\n\t"
"ror %[sha256], r7, #6\n\t"
"eor %[data], r8, r9\n\t"
"eor %[sha256], %[sha256], r7, ror #11\n\t"
@ -2546,7 +2546,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add %[len], %[len], %[sha256]\n\t"
"add %[len], %[len], %[data]\n\t"
/* Round 8 */
"vmov r10, d4[0]\n\t"
"vmov.32 r10, d4[0]\n\t"
"ror %[sha256], r6, #6\n\t"
"eor %[data], r7, r8\n\t"
"eor %[sha256], %[sha256], r6, ror #11\n\t"
@ -2569,7 +2569,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r9, r9, %[sha256]\n\t"
"add r9, r9, %[data]\n\t"
/* Round 9 */
"vmov r10, d4[1]\n\t"
"vmov.32 r10, d4[1]\n\t"
"ror %[sha256], r5, #6\n\t"
"eor %[data], r6, r7\n\t"
"eor %[sha256], %[sha256], r5, ror #11\n\t"
@ -2592,7 +2592,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r8, r8, %[sha256]\n\t"
"add r8, r8, %[data]\n\t"
/* Round 10 */
"vmov r10, d5[0]\n\t"
"vmov.32 r10, d5[0]\n\t"
"ror %[sha256], r4, #6\n\t"
"eor %[data], r5, r6\n\t"
"eor %[sha256], %[sha256], r4, ror #11\n\t"
@ -2615,7 +2615,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r7, r7, %[sha256]\n\t"
"add r7, r7, %[data]\n\t"
/* Round 11 */
"vmov r10, d5[1]\n\t"
"vmov.32 r10, d5[1]\n\t"
"ror %[sha256], r3, #6\n\t"
"eor %[data], r4, r5\n\t"
"eor %[sha256], %[sha256], r3, ror #11\n\t"
@ -2638,7 +2638,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r6, r6, %[sha256]\n\t"
"add r6, r6, %[data]\n\t"
/* Round 12 */
"vmov r10, d6[0]\n\t"
"vmov.32 r10, d6[0]\n\t"
"ror %[sha256], %[len], #6\n\t"
"eor %[data], r3, r4\n\t"
"eor %[sha256], %[sha256], %[len], ror #11\n\t"
@ -2661,7 +2661,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r5, r5, %[sha256]\n\t"
"add r5, r5, %[data]\n\t"
/* Round 13 */
"vmov r10, d6[1]\n\t"
"vmov.32 r10, d6[1]\n\t"
"ror %[sha256], r9, #6\n\t"
"eor %[data], %[len], r3\n\t"
"eor %[sha256], %[sha256], r9, ror #11\n\t"
@ -2684,7 +2684,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r4, r4, %[sha256]\n\t"
"add r4, r4, %[data]\n\t"
/* Round 14 */
"vmov r10, d7[0]\n\t"
"vmov.32 r10, d7[0]\n\t"
"ror %[sha256], r8, #6\n\t"
"eor %[data], r9, %[len]\n\t"
"eor %[sha256], %[sha256], r8, ror #11\n\t"
@ -2707,7 +2707,7 @@ void Transform_Sha256_Len(wc_Sha256* sha256_p, const byte* data_p, word32 len_p)
"add r3, r3, %[sha256]\n\t"
"add r3, r3, %[data]\n\t"
/* Round 15 */
"vmov r10, d7[1]\n\t"
"vmov.32 r10, d7[1]\n\t"
"ror %[sha256], r7, #6\n\t"
"eor %[data], r8, r9\n\t"
"eor %[sha256], %[sha256], r7, ror #11\n\t"