Merge branch 'feature/establish-an-upstream-relationship-with-ics-openvpn-codebase-#5381' into develop

1 files changed, 632 insertions, 226 deletions

diff --git a/app/openssl/crypto/sha/asm/sha256-armv4.s b/app/openssl/crypto/sha/asm/sha256-armv4.s
index ee903dc4..9c20a63c 100644
--- a/app/openssl/crypto/sha/asm/sha256-armv4.s
+++ b/app/openssl/crypto/sha/asm/sha256-armv4.s
@@ -1,3 +1,5 @@
+#include "arm_arch.h"
+
 .text
 .code	32
 
@@ -27,11 +29,14 @@ K256:
 sha256_block_data_order:
 	sub	r3,pc,#8		@ sha256_block_data_order
 	add	r2,r1,r2,lsl#6	@ len to point at the end of inp
-	stmdb	sp!,{r0,r1,r2,r4-r12,lr}
+	stmdb	sp!,{r0,r1,r2,r4-r11,lr}
 	ldmia	r0,{r4,r5,r6,r7,r8,r9,r10,r11}
 	sub	r14,r3,#256		@ K256
 	sub	sp,sp,#16*4		@ alloca(X[16])
 .Loop:
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 0
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -39,14 +44,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r8,ror#6
-	str	r3,[sp,#0*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r8,ror#11
 	eor	r2,r9,r10
+#if 0>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 0==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r8,ror#25	@ Sigma1(e)
 	and	r2,r2,r8
+	str	r3,[sp,#0*4]
 	add	r3,r3,r0
 	eor	r2,r2,r10			@ Ch(e,f,g)
 	add	r3,r3,r11
@@ -55,6 +68,9 @@ sha256_block_data_order:
 	eor	r11,r11,r4,ror#13
 	add	r3,r3,r12
 	eor	r11,r11,r4,ror#22		@ Sigma0(a)
+#if 0>=15
+	ldr	r1,[sp,#2*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r4,r5
 	and	r2,r4,r5
 	and	r0,r0,r6
@@ -62,6 +78,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r7,r7,r3
 	add	r11,r11,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 1
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -69,14 +88,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r7,ror#6
-	str	r3,[sp,#1*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r7,ror#11
 	eor	r2,r8,r9
+#if 1>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 1==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r7,ror#25	@ Sigma1(e)
 	and	r2,r2,r7
+	str	r3,[sp,#1*4]
 	add	r3,r3,r0
 	eor	r2,r2,r9			@ Ch(e,f,g)
 	add	r3,r3,r10
@@ -85,6 +112,9 @@ sha256_block_data_order:
 	eor	r10,r10,r11,ror#13
 	add	r3,r3,r12
 	eor	r10,r10,r11,ror#22		@ Sigma0(a)
+#if 1>=15
+	ldr	r1,[sp,#3*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r11,r4
 	and	r2,r11,r4
 	and	r0,r0,r5
@@ -92,6 +122,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r6,r6,r3
 	add	r10,r10,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 2
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -99,14 +132,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r6,ror#6
-	str	r3,[sp,#2*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r6,ror#11
 	eor	r2,r7,r8
+#if 2>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 2==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r6,ror#25	@ Sigma1(e)
 	and	r2,r2,r6
+	str	r3,[sp,#2*4]
 	add	r3,r3,r0
 	eor	r2,r2,r8			@ Ch(e,f,g)
 	add	r3,r3,r9
@@ -115,6 +156,9 @@ sha256_block_data_order:
 	eor	r9,r9,r10,ror#13
 	add	r3,r3,r12
 	eor	r9,r9,r10,ror#22		@ Sigma0(a)
+#if 2>=15
+	ldr	r1,[sp,#4*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r10,r11
 	and	r2,r10,r11
 	and	r0,r0,r4
@@ -122,6 +166,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r5,r5,r3
 	add	r9,r9,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 3
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -129,14 +176,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r5,ror#6
-	str	r3,[sp,#3*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r5,ror#11
 	eor	r2,r6,r7
+#if 3>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 3==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r5,ror#25	@ Sigma1(e)
 	and	r2,r2,r5
+	str	r3,[sp,#3*4]
 	add	r3,r3,r0
 	eor	r2,r2,r7			@ Ch(e,f,g)
 	add	r3,r3,r8
@@ -145,6 +200,9 @@ sha256_block_data_order:
 	eor	r8,r8,r9,ror#13
 	add	r3,r3,r12
 	eor	r8,r8,r9,ror#22		@ Sigma0(a)
+#if 3>=15
+	ldr	r1,[sp,#5*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r9,r10
 	and	r2,r9,r10
 	and	r0,r0,r11
@@ -152,6 +210,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r4,r4,r3
 	add	r8,r8,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 4
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -159,14 +220,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r4,ror#6
-	str	r3,[sp,#4*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r4,ror#11
 	eor	r2,r5,r6
+#if 4>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 4==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r4,ror#25	@ Sigma1(e)
 	and	r2,r2,r4
+	str	r3,[sp,#4*4]
 	add	r3,r3,r0
 	eor	r2,r2,r6			@ Ch(e,f,g)
 	add	r3,r3,r7
@@ -175,6 +244,9 @@ sha256_block_data_order:
 	eor	r7,r7,r8,ror#13
 	add	r3,r3,r12
 	eor	r7,r7,r8,ror#22		@ Sigma0(a)
+#if 4>=15
+	ldr	r1,[sp,#6*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r8,r9
 	and	r2,r8,r9
 	and	r0,r0,r10
@@ -182,6 +254,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r11,r11,r3
 	add	r7,r7,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 5
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -189,14 +264,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r11,ror#6
-	str	r3,[sp,#5*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r11,ror#11
 	eor	r2,r4,r5
+#if 5>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 5==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r11,ror#25	@ Sigma1(e)
 	and	r2,r2,r11
+	str	r3,[sp,#5*4]
 	add	r3,r3,r0
 	eor	r2,r2,r5			@ Ch(e,f,g)
 	add	r3,r3,r6
@@ -205,6 +288,9 @@ sha256_block_data_order:
 	eor	r6,r6,r7,ror#13
 	add	r3,r3,r12
 	eor	r6,r6,r7,ror#22		@ Sigma0(a)
+#if 5>=15
+	ldr	r1,[sp,#7*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r7,r8
 	and	r2,r7,r8
 	and	r0,r0,r9
@@ -212,6 +298,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r10,r10,r3
 	add	r6,r6,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 6
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -219,14 +308,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r10,ror#6
-	str	r3,[sp,#6*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r10,ror#11
 	eor	r2,r11,r4
+#if 6>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 6==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r10,ror#25	@ Sigma1(e)
 	and	r2,r2,r10
+	str	r3,[sp,#6*4]
 	add	r3,r3,r0
 	eor	r2,r2,r4			@ Ch(e,f,g)
 	add	r3,r3,r5
@@ -235,6 +332,9 @@ sha256_block_data_order:
 	eor	r5,r5,r6,ror#13
 	add	r3,r3,r12
 	eor	r5,r5,r6,ror#22		@ Sigma0(a)
+#if 6>=15
+	ldr	r1,[sp,#8*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r6,r7
 	and	r2,r6,r7
 	and	r0,r0,r8
@@ -242,6 +342,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r9,r9,r3
 	add	r5,r5,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 7
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -249,14 +352,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r9,ror#6
-	str	r3,[sp,#7*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r9,ror#11
 	eor	r2,r10,r11
+#if 7>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 7==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r9,ror#25	@ Sigma1(e)
 	and	r2,r2,r9
+	str	r3,[sp,#7*4]
 	add	r3,r3,r0
 	eor	r2,r2,r11			@ Ch(e,f,g)
 	add	r3,r3,r4
@@ -265,6 +376,9 @@ sha256_block_data_order:
 	eor	r4,r4,r5,ror#13
 	add	r3,r3,r12
 	eor	r4,r4,r5,ror#22		@ Sigma0(a)
+#if 7>=15
+	ldr	r1,[sp,#9*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r5,r6
 	and	r2,r5,r6
 	and	r0,r0,r7
@@ -272,6 +386,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r8,r8,r3
 	add	r4,r4,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 8
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -279,14 +396,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r8,ror#6
-	str	r3,[sp,#8*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r8,ror#11
 	eor	r2,r9,r10
+#if 8>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 8==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r8,ror#25	@ Sigma1(e)
 	and	r2,r2,r8
+	str	r3,[sp,#8*4]
 	add	r3,r3,r0
 	eor	r2,r2,r10			@ Ch(e,f,g)
 	add	r3,r3,r11
@@ -295,6 +420,9 @@ sha256_block_data_order:
 	eor	r11,r11,r4,ror#13
 	add	r3,r3,r12
 	eor	r11,r11,r4,ror#22		@ Sigma0(a)
+#if 8>=15
+	ldr	r1,[sp,#10*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r4,r5
 	and	r2,r4,r5
 	and	r0,r0,r6
@@ -302,6 +430,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r7,r7,r3
 	add	r11,r11,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 9
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -309,14 +440,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r7,ror#6
-	str	r3,[sp,#9*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r7,ror#11
 	eor	r2,r8,r9
+#if 9>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 9==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r7,ror#25	@ Sigma1(e)
 	and	r2,r2,r7
+	str	r3,[sp,#9*4]
 	add	r3,r3,r0
 	eor	r2,r2,r9			@ Ch(e,f,g)
 	add	r3,r3,r10
@@ -325,6 +464,9 @@ sha256_block_data_order:
 	eor	r10,r10,r11,ror#13
 	add	r3,r3,r12
 	eor	r10,r10,r11,ror#22		@ Sigma0(a)
+#if 9>=15
+	ldr	r1,[sp,#11*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r11,r4
 	and	r2,r11,r4
 	and	r0,r0,r5
@@ -332,6 +474,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r6,r6,r3
 	add	r10,r10,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 10
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -339,14 +484,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r6,ror#6
-	str	r3,[sp,#10*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r6,ror#11
 	eor	r2,r7,r8
+#if 10>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 10==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r6,ror#25	@ Sigma1(e)
 	and	r2,r2,r6
+	str	r3,[sp,#10*4]
 	add	r3,r3,r0
 	eor	r2,r2,r8			@ Ch(e,f,g)
 	add	r3,r3,r9
@@ -355,6 +508,9 @@ sha256_block_data_order:
 	eor	r9,r9,r10,ror#13
 	add	r3,r3,r12
 	eor	r9,r9,r10,ror#22		@ Sigma0(a)
+#if 10>=15
+	ldr	r1,[sp,#12*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r10,r11
 	and	r2,r10,r11
 	and	r0,r0,r4
@@ -362,6 +518,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r5,r5,r3
 	add	r9,r9,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 11
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -369,14 +528,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r5,ror#6
-	str	r3,[sp,#11*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r5,ror#11
 	eor	r2,r6,r7
+#if 11>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 11==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r5,ror#25	@ Sigma1(e)
 	and	r2,r2,r5
+	str	r3,[sp,#11*4]
 	add	r3,r3,r0
 	eor	r2,r2,r7			@ Ch(e,f,g)
 	add	r3,r3,r8
@@ -385,6 +552,9 @@ sha256_block_data_order:
 	eor	r8,r8,r9,ror#13
 	add	r3,r3,r12
 	eor	r8,r8,r9,ror#22		@ Sigma0(a)
+#if 11>=15
+	ldr	r1,[sp,#13*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r9,r10
 	and	r2,r9,r10
 	and	r0,r0,r11
@@ -392,6 +562,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r4,r4,r3
 	add	r8,r8,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 12
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -399,14 +572,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r4,ror#6
-	str	r3,[sp,#12*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r4,ror#11
 	eor	r2,r5,r6
+#if 12>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 12==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r4,ror#25	@ Sigma1(e)
 	and	r2,r2,r4
+	str	r3,[sp,#12*4]
 	add	r3,r3,r0
 	eor	r2,r2,r6			@ Ch(e,f,g)
 	add	r3,r3,r7
@@ -415,6 +596,9 @@ sha256_block_data_order:
 	eor	r7,r7,r8,ror#13
 	add	r3,r3,r12
 	eor	r7,r7,r8,ror#22		@ Sigma0(a)
+#if 12>=15
+	ldr	r1,[sp,#14*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r8,r9
 	and	r2,r8,r9
 	and	r0,r0,r10
@@ -422,6 +606,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r11,r11,r3
 	add	r7,r7,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 13
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -429,14 +616,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r11,ror#6
-	str	r3,[sp,#13*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r11,ror#11
 	eor	r2,r4,r5
+#if 13>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 13==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r11,ror#25	@ Sigma1(e)
 	and	r2,r2,r11
+	str	r3,[sp,#13*4]
 	add	r3,r3,r0
 	eor	r2,r2,r5			@ Ch(e,f,g)
 	add	r3,r3,r6
@@ -445,6 +640,9 @@ sha256_block_data_order:
 	eor	r6,r6,r7,ror#13
 	add	r3,r3,r12
 	eor	r6,r6,r7,ror#22		@ Sigma0(a)
+#if 13>=15
+	ldr	r1,[sp,#15*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r7,r8
 	and	r2,r7,r8
 	and	r0,r0,r9
@@ -452,6 +650,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r10,r10,r3
 	add	r6,r6,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 14
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -459,14 +660,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r10,ror#6
-	str	r3,[sp,#14*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r10,ror#11
 	eor	r2,r11,r4
+#if 14>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 14==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r10,ror#25	@ Sigma1(e)
 	and	r2,r2,r10
+	str	r3,[sp,#14*4]
 	add	r3,r3,r0
 	eor	r2,r2,r4			@ Ch(e,f,g)
 	add	r3,r3,r5
@@ -475,6 +684,9 @@ sha256_block_data_order:
 	eor	r5,r5,r6,ror#13
 	add	r3,r3,r12
 	eor	r5,r5,r6,ror#22		@ Sigma0(a)
+#if 14>=15
+	ldr	r1,[sp,#0*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r6,r7
 	and	r2,r6,r7
 	and	r0,r0,r8
@@ -482,6 +694,9 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r9,r9,r3
 	add	r5,r5,r0
+#if __ARM_ARCH__>=7
+	ldr	r3,[r1],#4
+#else
 	ldrb	r3,[r1,#3]			@ 15
 	ldrb	r12,[r1,#2]
 	ldrb	r2,[r1,#1]
@@ -489,14 +704,22 @@ sha256_block_data_order:
 	orr	r3,r3,r12,lsl#8
 	orr	r3,r3,r2,lsl#16
 	orr	r3,r3,r0,lsl#24
-	str	r1,[sp,#17*4]
-	ldr	r12,[r14],#4			@ *K256++
+#endif
 	mov	r0,r9,ror#6
-	str	r3,[sp,#15*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r9,ror#11
 	eor	r2,r10,r11
+#if 15>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 15==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r9,ror#25	@ Sigma1(e)
 	and	r2,r2,r9
+	str	r3,[sp,#15*4]
 	add	r3,r3,r0
 	eor	r2,r2,r11			@ Ch(e,f,g)
 	add	r3,r3,r4
@@ -505,6 +728,9 @@ sha256_block_data_order:
 	eor	r4,r4,r5,ror#13
 	add	r3,r3,r12
 	eor	r4,r4,r5,ror#22		@ Sigma0(a)
+#if 15>=15
+	ldr	r1,[sp,#1*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r5,r6
 	and	r2,r5,r6
 	and	r0,r0,r7
@@ -513,26 +739,34 @@ sha256_block_data_order:
 	add	r8,r8,r3
 	add	r4,r4,r0
 .Lrounds_16_xx:
-	ldr	r2,[sp,#1*4]		@ 16
+	@ ldr	r1,[sp,#1*4]		@ 16
 	ldr	r12,[sp,#14*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#0*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#9*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#9*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r8,ror#6
-	str	r3,[sp,#0*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r8,ror#11
 	eor	r2,r9,r10
+#if 16>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 16==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r8,ror#25	@ Sigma1(e)
 	and	r2,r2,r8
+	str	r3,[sp,#0*4]
 	add	r3,r3,r0
 	eor	r2,r2,r10			@ Ch(e,f,g)
 	add	r3,r3,r11
@@ -541,6 +775,9 @@ sha256_block_data_order:
 	eor	r11,r11,r4,ror#13
 	add	r3,r3,r12
 	eor	r11,r11,r4,ror#22		@ Sigma0(a)
+#if 16>=15
+	ldr	r1,[sp,#2*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r4,r5
 	and	r2,r4,r5
 	and	r0,r0,r6
@@ -548,26 +785,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r7,r7,r3
 	add	r11,r11,r0
-	ldr	r2,[sp,#2*4]		@ 17
+	@ ldr	r1,[sp,#2*4]		@ 17
 	ldr	r12,[sp,#15*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#1*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#10*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#10*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r7,ror#6
-	str	r3,[sp,#1*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r7,ror#11
 	eor	r2,r8,r9
+#if 17>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 17==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r7,ror#25	@ Sigma1(e)
 	and	r2,r2,r7
+	str	r3,[sp,#1*4]
 	add	r3,r3,r0
 	eor	r2,r2,r9			@ Ch(e,f,g)
 	add	r3,r3,r10
@@ -576,6 +821,9 @@ sha256_block_data_order:
 	eor	r10,r10,r11,ror#13
 	add	r3,r3,r12
 	eor	r10,r10,r11,ror#22		@ Sigma0(a)
+#if 17>=15
+	ldr	r1,[sp,#3*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r11,r4
 	and	r2,r11,r4
 	and	r0,r0,r5
@@ -583,26 +831,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r6,r6,r3
 	add	r10,r10,r0
-	ldr	r2,[sp,#3*4]		@ 18
+	@ ldr	r1,[sp,#3*4]		@ 18
 	ldr	r12,[sp,#0*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#2*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#11*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#11*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r6,ror#6
-	str	r3,[sp,#2*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r6,ror#11
 	eor	r2,r7,r8
+#if 18>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 18==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r6,ror#25	@ Sigma1(e)
 	and	r2,r2,r6
+	str	r3,[sp,#2*4]
 	add	r3,r3,r0
 	eor	r2,r2,r8			@ Ch(e,f,g)
 	add	r3,r3,r9
@@ -611,6 +867,9 @@ sha256_block_data_order:
 	eor	r9,r9,r10,ror#13
 	add	r3,r3,r12
 	eor	r9,r9,r10,ror#22		@ Sigma0(a)
+#if 18>=15
+	ldr	r1,[sp,#4*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r10,r11
 	and	r2,r10,r11
 	and	r0,r0,r4
@@ -618,26 +877,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r5,r5,r3
 	add	r9,r9,r0
-	ldr	r2,[sp,#4*4]		@ 19
+	@ ldr	r1,[sp,#4*4]		@ 19
 	ldr	r12,[sp,#1*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#3*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#12*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#12*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r5,ror#6
-	str	r3,[sp,#3*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r5,ror#11
 	eor	r2,r6,r7
+#if 19>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 19==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r5,ror#25	@ Sigma1(e)
 	and	r2,r2,r5
+	str	r3,[sp,#3*4]
 	add	r3,r3,r0
 	eor	r2,r2,r7			@ Ch(e,f,g)
 	add	r3,r3,r8
@@ -646,6 +913,9 @@ sha256_block_data_order:
 	eor	r8,r8,r9,ror#13
 	add	r3,r3,r12
 	eor	r8,r8,r9,ror#22		@ Sigma0(a)
+#if 19>=15
+	ldr	r1,[sp,#5*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r9,r10
 	and	r2,r9,r10
 	and	r0,r0,r11
@@ -653,26 +923,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r4,r4,r3
 	add	r8,r8,r0
-	ldr	r2,[sp,#5*4]		@ 20
+	@ ldr	r1,[sp,#5*4]		@ 20
 	ldr	r12,[sp,#2*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#4*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#13*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#13*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r4,ror#6
-	str	r3,[sp,#4*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r4,ror#11
 	eor	r2,r5,r6
+#if 20>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 20==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r4,ror#25	@ Sigma1(e)
 	and	r2,r2,r4
+	str	r3,[sp,#4*4]
 	add	r3,r3,r0
 	eor	r2,r2,r6			@ Ch(e,f,g)
 	add	r3,r3,r7
@@ -681,6 +959,9 @@ sha256_block_data_order:
 	eor	r7,r7,r8,ror#13
 	add	r3,r3,r12
 	eor	r7,r7,r8,ror#22		@ Sigma0(a)
+#if 20>=15
+	ldr	r1,[sp,#6*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r8,r9
 	and	r2,r8,r9
 	and	r0,r0,r10
@@ -688,26 +969,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r11,r11,r3
 	add	r7,r7,r0
-	ldr	r2,[sp,#6*4]		@ 21
+	@ ldr	r1,[sp,#6*4]		@ 21
 	ldr	r12,[sp,#3*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#5*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#14*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#14*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r11,ror#6
-	str	r3,[sp,#5*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r11,ror#11
 	eor	r2,r4,r5
+#if 21>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 21==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r11,ror#25	@ Sigma1(e)
 	and	r2,r2,r11
+	str	r3,[sp,#5*4]
 	add	r3,r3,r0
 	eor	r2,r2,r5			@ Ch(e,f,g)
 	add	r3,r3,r6
@@ -716,6 +1005,9 @@ sha256_block_data_order:
 	eor	r6,r6,r7,ror#13
 	add	r3,r3,r12
 	eor	r6,r6,r7,ror#22		@ Sigma0(a)
+#if 21>=15
+	ldr	r1,[sp,#7*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r7,r8
 	and	r2,r7,r8
 	and	r0,r0,r9
@@ -723,26 +1015,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r10,r10,r3
 	add	r6,r6,r0
-	ldr	r2,[sp,#7*4]		@ 22
+	@ ldr	r1,[sp,#7*4]		@ 22
 	ldr	r12,[sp,#4*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#6*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#15*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#15*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r10,ror#6
-	str	r3,[sp,#6*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r10,ror#11
 	eor	r2,r11,r4
+#if 22>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 22==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r10,ror#25	@ Sigma1(e)
 	and	r2,r2,r10
+	str	r3,[sp,#6*4]
 	add	r3,r3,r0
 	eor	r2,r2,r4			@ Ch(e,f,g)
 	add	r3,r3,r5
@@ -751,6 +1051,9 @@ sha256_block_data_order:
 	eor	r5,r5,r6,ror#13
 	add	r3,r3,r12
 	eor	r5,r5,r6,ror#22		@ Sigma0(a)
+#if 22>=15
+	ldr	r1,[sp,#8*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r6,r7
 	and	r2,r6,r7
 	and	r0,r0,r8
@@ -758,26 +1061,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r9,r9,r3
 	add	r5,r5,r0
-	ldr	r2,[sp,#8*4]		@ 23
+	@ ldr	r1,[sp,#8*4]		@ 23
 	ldr	r12,[sp,#5*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#7*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#0*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#0*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r9,ror#6
-	str	r3,[sp,#7*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r9,ror#11
 	eor	r2,r10,r11
+#if 23>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 23==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r9,ror#25	@ Sigma1(e)
 	and	r2,r2,r9
+	str	r3,[sp,#7*4]
 	add	r3,r3,r0
 	eor	r2,r2,r11			@ Ch(e,f,g)
 	add	r3,r3,r4
@@ -786,6 +1097,9 @@ sha256_block_data_order:
 	eor	r4,r4,r5,ror#13
 	add	r3,r3,r12
 	eor	r4,r4,r5,ror#22		@ Sigma0(a)
+#if 23>=15
+	ldr	r1,[sp,#9*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r5,r6
 	and	r2,r5,r6
 	and	r0,r0,r7
@@ -793,26 +1107,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r8,r8,r3
 	add	r4,r4,r0
-	ldr	r2,[sp,#9*4]		@ 24
+	@ ldr	r1,[sp,#9*4]		@ 24
 	ldr	r12,[sp,#6*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#8*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#1*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#1*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r8,ror#6
-	str	r3,[sp,#8*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r8,ror#11
 	eor	r2,r9,r10
+#if 24>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 24==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r8,ror#25	@ Sigma1(e)
 	and	r2,r2,r8
+	str	r3,[sp,#8*4]
 	add	r3,r3,r0
 	eor	r2,r2,r10			@ Ch(e,f,g)
 	add	r3,r3,r11
@@ -821,6 +1143,9 @@ sha256_block_data_order:
 	eor	r11,r11,r4,ror#13
 	add	r3,r3,r12
 	eor	r11,r11,r4,ror#22		@ Sigma0(a)
+#if 24>=15
+	ldr	r1,[sp,#10*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r4,r5
 	and	r2,r4,r5
 	and	r0,r0,r6
@@ -828,26 +1153,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r7,r7,r3
 	add	r11,r11,r0
-	ldr	r2,[sp,#10*4]		@ 25
+	@ ldr	r1,[sp,#10*4]		@ 25
 	ldr	r12,[sp,#7*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#9*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#2*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#2*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r7,ror#6
-	str	r3,[sp,#9*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r7,ror#11
 	eor	r2,r8,r9
+#if 25>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 25==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r7,ror#25	@ Sigma1(e)
 	and	r2,r2,r7
+	str	r3,[sp,#9*4]
 	add	r3,r3,r0
 	eor	r2,r2,r9			@ Ch(e,f,g)
 	add	r3,r3,r10
@@ -856,6 +1189,9 @@ sha256_block_data_order:
 	eor	r10,r10,r11,ror#13
 	add	r3,r3,r12
 	eor	r10,r10,r11,ror#22		@ Sigma0(a)
+#if 25>=15
+	ldr	r1,[sp,#11*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r11,r4
 	and	r2,r11,r4
 	and	r0,r0,r5
@@ -863,26 +1199,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r6,r6,r3
 	add	r10,r10,r0
-	ldr	r2,[sp,#11*4]		@ 26
+	@ ldr	r1,[sp,#11*4]		@ 26
 	ldr	r12,[sp,#8*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#10*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#3*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#3*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r6,ror#6
-	str	r3,[sp,#10*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r6,ror#11
 	eor	r2,r7,r8
+#if 26>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 26==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r6,ror#25	@ Sigma1(e)
 	and	r2,r2,r6
+	str	r3,[sp,#10*4]
 	add	r3,r3,r0
 	eor	r2,r2,r8			@ Ch(e,f,g)
 	add	r3,r3,r9
@@ -891,6 +1235,9 @@ sha256_block_data_order:
 	eor	r9,r9,r10,ror#13
 	add	r3,r3,r12
 	eor	r9,r9,r10,ror#22		@ Sigma0(a)
+#if 26>=15
+	ldr	r1,[sp,#12*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r10,r11
 	and	r2,r10,r11
 	and	r0,r0,r4
@@ -898,26 +1245,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r5,r5,r3
 	add	r9,r9,r0
-	ldr	r2,[sp,#12*4]		@ 27
+	@ ldr	r1,[sp,#12*4]		@ 27
 	ldr	r12,[sp,#9*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#11*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#4*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#4*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r5,ror#6
-	str	r3,[sp,#11*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r5,ror#11
 	eor	r2,r6,r7
+#if 27>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 27==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r5,ror#25	@ Sigma1(e)
 	and	r2,r2,r5
+	str	r3,[sp,#11*4]
 	add	r3,r3,r0
 	eor	r2,r2,r7			@ Ch(e,f,g)
 	add	r3,r3,r8
@@ -926,6 +1281,9 @@ sha256_block_data_order:
 	eor	r8,r8,r9,ror#13
 	add	r3,r3,r12
 	eor	r8,r8,r9,ror#22		@ Sigma0(a)
+#if 27>=15
+	ldr	r1,[sp,#13*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r9,r10
 	and	r2,r9,r10
 	and	r0,r0,r11
@@ -933,26 +1291,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r4,r4,r3
 	add	r8,r8,r0
-	ldr	r2,[sp,#13*4]		@ 28
+	@ ldr	r1,[sp,#13*4]		@ 28
 	ldr	r12,[sp,#10*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#12*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#5*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#5*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r4,ror#6
-	str	r3,[sp,#12*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r4,ror#11
 	eor	r2,r5,r6
+#if 28>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 28==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r4,ror#25	@ Sigma1(e)
 	and	r2,r2,r4
+	str	r3,[sp,#12*4]
 	add	r3,r3,r0
 	eor	r2,r2,r6			@ Ch(e,f,g)
 	add	r3,r3,r7
@@ -961,6 +1327,9 @@ sha256_block_data_order:
 	eor	r7,r7,r8,ror#13
 	add	r3,r3,r12
 	eor	r7,r7,r8,ror#22		@ Sigma0(a)
+#if 28>=15
+	ldr	r1,[sp,#14*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r8,r9
 	and	r2,r8,r9
 	and	r0,r0,r10
@@ -968,26 +1337,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r11,r11,r3
 	add	r7,r7,r0
-	ldr	r2,[sp,#14*4]		@ 29
+	@ ldr	r1,[sp,#14*4]		@ 29
 	ldr	r12,[sp,#11*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#13*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#6*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#6*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r11,ror#6
-	str	r3,[sp,#13*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r11,ror#11
 	eor	r2,r4,r5
+#if 29>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 29==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r11,ror#25	@ Sigma1(e)
 	and	r2,r2,r11
+	str	r3,[sp,#13*4]
 	add	r3,r3,r0
 	eor	r2,r2,r5			@ Ch(e,f,g)
 	add	r3,r3,r6
@@ -996,6 +1373,9 @@ sha256_block_data_order:
 	eor	r6,r6,r7,ror#13
 	add	r3,r3,r12
 	eor	r6,r6,r7,ror#22		@ Sigma0(a)
+#if 29>=15
+	ldr	r1,[sp,#15*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r7,r8
 	and	r2,r7,r8
 	and	r0,r0,r9
@@ -1003,26 +1383,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r10,r10,r3
 	add	r6,r6,r0
-	ldr	r2,[sp,#15*4]		@ 30
+	@ ldr	r1,[sp,#15*4]		@ 30
 	ldr	r12,[sp,#12*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#14*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#7*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#7*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r10,ror#6
-	str	r3,[sp,#14*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r10,ror#11
 	eor	r2,r11,r4
+#if 30>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 30==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r10,ror#25	@ Sigma1(e)
 	and	r2,r2,r10
+	str	r3,[sp,#14*4]
 	add	r3,r3,r0
 	eor	r2,r2,r4			@ Ch(e,f,g)
 	add	r3,r3,r5
@@ -1031,6 +1419,9 @@ sha256_block_data_order:
 	eor	r5,r5,r6,ror#13
 	add	r3,r3,r12
 	eor	r5,r5,r6,ror#22		@ Sigma0(a)
+#if 30>=15
+	ldr	r1,[sp,#0*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r6,r7
 	and	r2,r6,r7
 	and	r0,r0,r8
@@ -1038,26 +1429,34 @@ sha256_block_data_order:
 	orr	r0,r0,r2			@ Maj(a,b,c)
 	add	r9,r9,r3
 	add	r5,r5,r0
-	ldr	r2,[sp,#0*4]		@ 31
+	@ ldr	r1,[sp,#0*4]		@ 31
 	ldr	r12,[sp,#13*4]
+	mov	r0,r1,ror#7
 	ldr	r3,[sp,#15*4]
-	mov	r0,r2,ror#7
-	ldr	r1,[sp,#8*4]
-	eor	r0,r0,r2,ror#18
-	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
-	mov	r2,r12,ror#17
+	eor	r0,r0,r1,ror#18
+	ldr	r2,[sp,#8*4]
+	eor	r0,r0,r1,lsr#3	@ sigma0(X[i+1])
+	mov	r1,r12,ror#17
 	add	r3,r3,r0
-	eor	r2,r2,r12,ror#19
-	add	r3,r3,r1
-	eor	r2,r2,r12,lsr#10	@ sigma1(X[i+14])
+	eor	r1,r1,r12,ror#19
 	add	r3,r3,r2
-	ldr	r12,[r14],#4			@ *K256++
+	eor	r1,r1,r12,lsr#10	@ sigma1(X[i+14])
+	@ add	r3,r3,r1
 	mov	r0,r9,ror#6
-	str	r3,[sp,#15*4]
+	ldr	r12,[r14],#4			@ *K256++
 	eor	r0,r0,r9,ror#11
 	eor	r2,r10,r11
+#if 31>=16
+	add	r3,r3,r1			@ from BODY_16_xx
+#elif __ARM_ARCH__>=7 && defined(__ARMEL__)
+	rev	r3,r3
+#endif
+#if 31==15
+	str	r1,[sp,#17*4]			@ leave room for r1
+#endif
 	eor	r0,r0,r9,ror#25	@ Sigma1(e)
 	and	r2,r2,r9
+	str	r3,[sp,#15*4]
 	add	r3,r3,r0
 	eor	r2,r2,r11			@ Ch(e,f,g)
 	add	r3,r3,r4
@@ -1066,6 +1465,9 @@ sha256_block_data_order:
 	eor	r4,r4,r5,ror#13
 	add	r3,r3,r12
 	eor	r4,r4,r5,ror#22		@ Sigma0(a)
+#if 31>=15
+	ldr	r1,[sp,#1*4]		@ from BODY_16_xx
+#endif
 	orr	r0,r5,r6
 	and	r2,r5,r6
 	and	r0,r0,r7
@@ -1102,10 +1504,14 @@ sha256_block_data_order:
 	bne	.Loop
 
 	add	sp,sp,#19*4	@ destroy frame
-	ldmia	sp!,{r4-r12,lr}
+#if __ARM_ARCH__>=5
+	ldmia	sp!,{r4-r11,pc}
+#else
+	ldmia	sp!,{r4-r11,lr}
 	tst	lr,#1
 	moveq	pc,lr			@ be binary compatible with V4, yet
 	.word	0xe12fff1e			@ interoperable with Thumb ISA:-)
+#endif
 .size   sha256_block_data_order,.-sha256_block_data_order
 .asciz  "SHA256 block transform for ARMv4, CRYPTOGAMS by <appro@openssl.org>"
 .align	2