Update OpenSSL to aosp/master

--HG--
extra : amend_source : 1d737bc919a18f072b1038b548a2cbad4cf76e22

5 files changed, 2752 insertions, 55 deletions

diff --git a/main/openssl/crypto/aes/asm/aes-armv4.pl b/main/openssl/crypto/aes/asm/aes-armv4.pl
index 86b86c4a..4f891708 100644
--- a/main/openssl/crypto/aes/asm/aes-armv4.pl
+++ b/main/openssl/crypto/aes/asm/aes-armv4.pl
@@ -1,7 +1,7 @@
 #!/usr/bin/env perl
 
 # ====================================================================
-# Written by Andy Polyakov <appro@fy.chalmers.se> for the OpenSSL
+# Written by Andy Polyakov <appro@openssl.org> for the OpenSSL
 # project. The module is, however, dual licensed under OpenSSL and
 # CRYPTOGAMS licenses depending on where you obtain it. For further
 # details see http://www.openssl.org/~appro/cryptogams/.
@@ -51,9 +51,23 @@ $key="r11";
 $rounds="r12";
 
 $code=<<___;
-#include "arm_arch.h"
+#ifndef __KERNEL__
+# include "arm_arch.h"
+#else
+# define __ARM_ARCH__ __LINUX_ARM_ARCH__
+#endif
+
 .text
+#if __ARM_ARCH__<7
+.code	32
+#else
+.syntax	unified
+# ifdef __thumb2__
+.thumb
+# else
 .code	32
+# endif
+#endif
 
 .type	AES_Te,%object
 .align	5
@@ -167,7 +181,11 @@ AES_Te:
 .type   AES_encrypt,%function
 .align	5
 AES_encrypt:
+#if __ARM_ARCH__<7
 	sub	r3,pc,#8		@ AES_encrypt
+#else
+	adr	r3,AES_encrypt
+#endif
 	stmdb   sp!,{r1,r4-r12,lr}
 	mov	$rounds,r0		@ inp
 	mov	$key,r2
@@ -409,11 +427,21 @@ _armv4_AES_encrypt:
 .align	5
 private_AES_set_encrypt_key:
 _armv4_AES_set_encrypt_key:
+#if __ARM_ARCH__<7
 	sub	r3,pc,#8		@ AES_set_encrypt_key
+#else
+	adr	r3,private_AES_set_encrypt_key
+#endif
 	teq	r0,#0
+#if __ARM_ARCH__>=7
+	itt	eq			@ Thumb2 thing, sanity check in ARM
+#endif
 	moveq	r0,#-1
 	beq	.Labrt
 	teq	r2,#0
+#if __ARM_ARCH__>=7
+	itt	eq			@ Thumb2 thing, sanity check in ARM
+#endif
 	moveq	r0,#-1
 	beq	.Labrt
 
@@ -422,6 +450,9 @@ _armv4_AES_set_encrypt_key:
 	teq	r1,#192
 	beq	.Lok
 	teq	r1,#256
+#if __ARM_ARCH__>=7
+	itt	ne			@ Thumb2 thing, sanity check in ARM
+#endif
 	movne	r0,#-1
 	bne	.Labrt
 
@@ -576,6 +607,9 @@ _armv4_AES_set_encrypt_key:
 	str	$s2,[$key,#-16]
 	subs	$rounds,$rounds,#1
 	str	$s3,[$key,#-12]
+#if __ARM_ARCH__>=7
+	itt	eq				@ Thumb2 thing, sanity check in ARM
+#endif
 	subeq	r2,$key,#216
 	beq	.Ldone
 
@@ -645,6 +679,9 @@ _armv4_AES_set_encrypt_key:
 	str	$s2,[$key,#-24]
 	subs	$rounds,$rounds,#1
 	str	$s3,[$key,#-20]
+#if __ARM_ARCH__>=7
+	itt	eq				@ Thumb2 thing, sanity check in ARM
+#endif
 	subeq	r2,$key,#256
 	beq	.Ldone
 
@@ -674,11 +711,17 @@ _armv4_AES_set_encrypt_key:
 	str	$i3,[$key,#-4]
 	b	.L256_loop
 
+.align	2
 .Ldone:	mov	r0,#0
 	ldmia   sp!,{r4-r12,lr}
-.Labrt:	tst	lr,#1
+.Labrt:
+#if __ARM_ARCH__>=5
+	ret				@ bx lr
+#else
+	tst	lr,#1
 	moveq	pc,lr			@ be binary compatible with V4, yet
 	bx	lr			@ interoperable with Thumb ISA:-)
+#endif
 .size	private_AES_set_encrypt_key,.-private_AES_set_encrypt_key
 
 .global private_AES_set_decrypt_key
@@ -688,34 +731,57 @@ private_AES_set_decrypt_key:
 	str	lr,[sp,#-4]!            @ push lr
 	bl	_armv4_AES_set_encrypt_key
 	teq	r0,#0
-	ldrne	lr,[sp],#4              @ pop lr
+	ldr	lr,[sp],#4              @ pop lr
 	bne	.Labrt
 
-	stmdb   sp!,{r4-r12}
+	mov	r0,r2			@ AES_set_encrypt_key preserves r2,
+	mov	r1,r2			@ which is AES_KEY *key
+	b	_armv4_AES_set_enc2dec_key
+.size	private_AES_set_decrypt_key,.-private_AES_set_decrypt_key
 
-	ldr	$rounds,[r2,#240]	@ AES_set_encrypt_key preserves r2,
-	mov	$key,r2			@ which is AES_KEY *key
-	mov	$i1,r2
-	add	$i2,r2,$rounds,lsl#4
+@ void AES_set_enc2dec_key(const AES_KEY *inp,AES_KEY *out)
+.global	AES_set_enc2dec_key
+.type	AES_set_enc2dec_key,%function
+.align	5
+AES_set_enc2dec_key:
+_armv4_AES_set_enc2dec_key:
+	stmdb   sp!,{r4-r12,lr}
+
+	ldr	$rounds,[r0,#240]
+	mov	$i1,r0			@ input
+	add	$i2,r0,$rounds,lsl#4
+	mov	$key,r1			@ ouput
+	add	$tbl,r1,$rounds,lsl#4
+	str	$rounds,[r1,#240]
+
+.Linv:	ldr	$s0,[$i1],#16
+	ldr	$s1,[$i1,#-12]
+	ldr	$s2,[$i1,#-8]
+	ldr	$s3,[$i1,#-4]
+	ldr	$t1,[$i2],#-16
+	ldr	$t2,[$i2,#16+4]
+	ldr	$t3,[$i2,#16+8]
+	ldr	$i3,[$i2,#16+12]
+	str	$s0,[$tbl],#-16
+	str	$s1,[$tbl,#16+4]
+	str	$s2,[$tbl,#16+8]
+	str	$s3,[$tbl,#16+12]
+	str	$t1,[$key],#16
+	str	$t2,[$key,#-12]
+	str	$t3,[$key,#-8]
+	str	$i3,[$key,#-4]
+	teq	$i1,$i2
+	bne	.Linv
 
-.Linv:	ldr	$s0,[$i1]
+	ldr	$s0,[$i1]
 	ldr	$s1,[$i1,#4]
 	ldr	$s2,[$i1,#8]
 	ldr	$s3,[$i1,#12]
-	ldr	$t1,[$i2]
-	ldr	$t2,[$i2,#4]
-	ldr	$t3,[$i2,#8]
-	ldr	$i3,[$i2,#12]
-	str	$s0,[$i2],#-16
-	str	$s1,[$i2,#16+4]
-	str	$s2,[$i2,#16+8]
-	str	$s3,[$i2,#16+12]
-	str	$t1,[$i1],#16
-	str	$t2,[$i1,#-12]
-	str	$t3,[$i1,#-8]
-	str	$i3,[$i1,#-4]
-	teq	$i1,$i2
-	bne	.Linv
+	str	$s0,[$key]
+	str	$s1,[$key,#4]
+	str	$s2,[$key,#8]
+	str	$s3,[$key,#12]
+	sub	$key,$key,$rounds,lsl#3
 ___
 $mask80=$i1;
 $mask1b=$i2;
@@ -773,7 +839,7 @@ $code.=<<___;
 	moveq	pc,lr			@ be binary compatible with V4, yet
 	bx	lr			@ interoperable with Thumb ISA:-)
 #endif
-.size	private_AES_set_decrypt_key,.-private_AES_set_decrypt_key
+.size	AES_set_enc2dec_key,.-AES_set_enc2dec_key
 
 .type	AES_Td,%object
 .align	5
@@ -883,7 +949,11 @@ AES_Td:
 .type   AES_decrypt,%function
 .align	5
 AES_decrypt:
+#if __ARM_ARCH__<7
 	sub	r3,pc,#8		@ AES_decrypt
+#else
+	adr	r3,AES_decrypt
+#endif
 	stmdb   sp!,{r1,r4-r12,lr}
 	mov	$rounds,r0		@ inp
 	mov	$key,r2
@@ -1080,8 +1150,9 @@ _armv4_AES_decrypt:
 	ldrb	$t3,[$tbl,$i3]		@ Td4[s0>>0]
 	and	$i3,lr,$s1,lsr#8
 
+	add	$s1,$tbl,$s1,lsr#24
 	ldrb	$i1,[$tbl,$i1]		@ Td4[s1>>0]
-	ldrb	$s1,[$tbl,$s1,lsr#24]	@ Td4[s1>>24]
+	ldrb	$s1,[$s1]		@ Td4[s1>>24]
 	ldrb	$i2,[$tbl,$i2]		@ Td4[s1>>16]
 	eor	$s0,$i1,$s0,lsl#24
 	ldrb	$i3,[$tbl,$i3]		@ Td4[s1>>8]
@@ -1094,7 +1165,8 @@ _armv4_AES_decrypt:
 	ldrb	$i2,[$tbl,$i2]		@ Td4[s2>>0]
 	and	$i3,lr,$s2,lsr#16
 
-	ldrb	$s2,[$tbl,$s2,lsr#24]	@ Td4[s2>>24]
+	add	$s2,$tbl,$s2,lsr#24
+	ldrb	$s2,[$s2]		@ Td4[s2>>24]
 	eor	$s0,$s0,$i1,lsl#8
 	ldrb	$i3,[$tbl,$i3]		@ Td4[s2>>16]
 	eor	$s1,$i2,$s1,lsl#16
@@ -1106,8 +1178,9 @@ _armv4_AES_decrypt:
 	ldrb	$i2,[$tbl,$i2]		@ Td4[s3>>8]
 	and	$i3,lr,$s3		@ i2
 
+	add	$s3,$tbl,$s3,lsr#24
 	ldrb	$i3,[$tbl,$i3]		@ Td4[s3>>0]
-	ldrb	$s3,[$tbl,$s3,lsr#24]	@ Td4[s3>>24]
+	ldrb	$s3,[$s3]		@ Td4[s3>>24]
 	eor	$s0,$s0,$i1,lsl#16
 	ldr	$i1,[$key,#0]
 	eor	$s1,$s1,$i2,lsl#8
@@ -1130,5 +1203,15 @@ _armv4_AES_decrypt:
 ___
 
 $code =~ s/\bbx\s+lr\b/.word\t0xe12fff1e/gm;	# make it possible to compile with -march=armv4
+$code =~ s/\bret\b/bx\tlr/gm;
+
+open SELF,$0;
+while(<SELF>) {
+	next if (/^#!/);
+	last if (!s/^#/@/ and !/^$/);
+	print;
+}
+close SELF;
+
 print $code;
 close STDOUT;	# enforce flush
diff --git a/main/openssl/crypto/aes/asm/aes-armv4.s b/main/openssl/crypto/aes/asm/aes-armv4.s
index 2697d4ce..333a5227 100644
--- a/main/openssl/crypto/aes/asm/aes-armv4.s
+++ b/main/openssl/crypto/aes/asm/aes-armv4.s
@@ -1,6 +1,53 @@
-#include "arm_arch.h"
+
+@ ====================================================================
+@ Written by Andy Polyakov <appro@openssl.org> for the OpenSSL
+@ project. The module is, however, dual licensed under OpenSSL and
+@ CRYPTOGAMS licenses depending on where you obtain it. For further
+@ details see http://www.openssl.org/~appro/cryptogams/.
+@ ====================================================================
+
+@ AES for ARMv4
+
+@ January 2007.
+@
+@ Code uses single 1K S-box and is >2 times faster than code generated
+@ by gcc-3.4.1. This is thanks to unique feature of ARMv4 ISA, which
+@ allows to merge logical or arithmetic operation with shift or rotate
+@ in one instruction and emit combined result every cycle. The module
+@ is endian-neutral. The performance is ~42 cycles/byte for 128-bit
+@ key [on single-issue Xscale PXA250 core].
+
+@ May 2007.
+@
+@ AES_set_[en|de]crypt_key is added.
+
+@ July 2010.
+@
+@ Rescheduling for dual-issue pipeline resulted in 12% improvement on
+@ Cortex A8 core and ~25 cycles per byte processed with 128-bit key.
+
+@ February 2011.
+@
+@ Profiler-assisted and platform-specific optimization resulted in 16%
+@ improvement on Cortex A8 core and ~21.5 cycles per byte.
+
+#ifndef __KERNEL__
+# include "arm_arch.h"
+#else
+# define __ARM_ARCH__ __LINUX_ARM_ARCH__
+#endif
+
 .text
+#if __ARM_ARCH__<7
+.code	32
+#else
+.syntax	unified
+# ifdef __thumb2__
+.thumb
+# else
 .code	32
+# endif
+#endif
 
 .type	AES_Te,%object
 .align	5
@@ -114,7 +161,11 @@ AES_Te:
 .type   AES_encrypt,%function
 .align	5
 AES_encrypt:
+#if __ARM_ARCH__<7
 	sub	r3,pc,#8		@ AES_encrypt
+#else
+	adr	r3,AES_encrypt
+#endif
 	stmdb   sp!,{r1,r4-r12,lr}
 	mov	r12,r0		@ inp
 	mov	r11,r2
@@ -356,11 +407,21 @@ _armv4_AES_encrypt:
 .align	5
 private_AES_set_encrypt_key:
 _armv4_AES_set_encrypt_key:
+#if __ARM_ARCH__<7
 	sub	r3,pc,#8		@ AES_set_encrypt_key
+#else
+	adr	r3,private_AES_set_encrypt_key
+#endif
 	teq	r0,#0
+#if __ARM_ARCH__>=7
+	itt	eq			@ Thumb2 thing, sanity check in ARM
+#endif
 	moveq	r0,#-1
 	beq	.Labrt
 	teq	r2,#0
+#if __ARM_ARCH__>=7
+	itt	eq			@ Thumb2 thing, sanity check in ARM
+#endif
 	moveq	r0,#-1
 	beq	.Labrt
 
@@ -369,6 +430,9 @@ _armv4_AES_set_encrypt_key:
 	teq	r1,#192
 	beq	.Lok
 	teq	r1,#256
+#if __ARM_ARCH__>=7
+	itt	ne			@ Thumb2 thing, sanity check in ARM
+#endif
 	movne	r0,#-1
 	bne	.Labrt
 
@@ -523,6 +587,9 @@ _armv4_AES_set_encrypt_key:
 	str	r2,[r11,#-16]
 	subs	r12,r12,#1
 	str	r3,[r11,#-12]
+#if __ARM_ARCH__>=7
+	itt	eq				@ Thumb2 thing, sanity check in ARM
+#endif
 	subeq	r2,r11,#216
 	beq	.Ldone
 
@@ -592,6 +659,9 @@ _armv4_AES_set_encrypt_key:
 	str	r2,[r11,#-24]
 	subs	r12,r12,#1
 	str	r3,[r11,#-20]
+#if __ARM_ARCH__>=7
+	itt	eq				@ Thumb2 thing, sanity check in ARM
+#endif
 	subeq	r2,r11,#256
 	beq	.Ldone
 
@@ -621,11 +691,17 @@ _armv4_AES_set_encrypt_key:
 	str	r9,[r11,#-4]
 	b	.L256_loop
 
+.align	2
 .Ldone:	mov	r0,#0
 	ldmia   sp!,{r4-r12,lr}
-.Labrt:	tst	lr,#1
+.Labrt:
+#if __ARM_ARCH__>=5
+	bx	lr				@ .word	0xe12fff1e
+#else
+	tst	lr,#1
 	moveq	pc,lr			@ be binary compatible with V4, yet
 	.word	0xe12fff1e			@ interoperable with Thumb ISA:-)
+#endif
 .size	private_AES_set_encrypt_key,.-private_AES_set_encrypt_key
 
 .global private_AES_set_decrypt_key
@@ -635,34 +711,57 @@ private_AES_set_decrypt_key:
 	str	lr,[sp,#-4]!            @ push lr
 	bl	_armv4_AES_set_encrypt_key
 	teq	r0,#0
-	ldrne	lr,[sp],#4              @ pop lr
+	ldr	lr,[sp],#4              @ pop lr
 	bne	.Labrt
 
-	stmdb   sp!,{r4-r12}
+	mov	r0,r2			@ AES_set_encrypt_key preserves r2,
+	mov	r1,r2			@ which is AES_KEY *key
+	b	_armv4_AES_set_enc2dec_key
+.size	private_AES_set_decrypt_key,.-private_AES_set_decrypt_key
+
+@ void AES_set_enc2dec_key(const AES_KEY *inp,AES_KEY *out)
+.global	AES_set_enc2dec_key
+.type	AES_set_enc2dec_key,%function
+.align	5
+AES_set_enc2dec_key:
+_armv4_AES_set_enc2dec_key:
+	stmdb   sp!,{r4-r12,lr}
+
+	ldr	r12,[r0,#240]
+	mov	r7,r0			@ input
+	add	r8,r0,r12,lsl#4
+	mov	r11,r1			@ ouput
+	add	r10,r1,r12,lsl#4
+	str	r12,[r1,#240]
 
-	ldr	r12,[r2,#240]	@ AES_set_encrypt_key preserves r2,
-	mov	r11,r2			@ which is AES_KEY *key
-	mov	r7,r2
-	add	r8,r2,r12,lsl#4
+.Linv:	ldr	r0,[r7],#16
+	ldr	r1,[r7,#-12]
+	ldr	r2,[r7,#-8]
+	ldr	r3,[r7,#-4]
+	ldr	r4,[r8],#-16
+	ldr	r5,[r8,#16+4]
+	ldr	r6,[r8,#16+8]
+	ldr	r9,[r8,#16+12]
+	str	r0,[r10],#-16
+	str	r1,[r10,#16+4]
+	str	r2,[r10,#16+8]
+	str	r3,[r10,#16+12]
+	str	r4,[r11],#16
+	str	r5,[r11,#-12]
+	str	r6,[r11,#-8]
+	str	r9,[r11,#-4]
+	teq	r7,r8
+	bne	.Linv
 
-.Linv:	ldr	r0,[r7]
+	ldr	r0,[r7]
 	ldr	r1,[r7,#4]
 	ldr	r2,[r7,#8]
 	ldr	r3,[r7,#12]
-	ldr	r4,[r8]
-	ldr	r5,[r8,#4]
-	ldr	r6,[r8,#8]
-	ldr	r9,[r8,#12]
-	str	r0,[r8],#-16
-	str	r1,[r8,#16+4]
-	str	r2,[r8,#16+8]
-	str	r3,[r8,#16+12]
-	str	r4,[r7],#16
-	str	r5,[r7,#-12]
-	str	r6,[r7,#-8]
-	str	r9,[r7,#-4]
-	teq	r7,r8
-	bne	.Linv
+	str	r0,[r11]
+	str	r1,[r11,#4]
+	str	r2,[r11,#8]
+	str	r3,[r11,#12]
+	sub	r11,r11,r12,lsl#3
 	ldr	r0,[r11,#16]!		@ prefetch tp1
 	mov	r7,#0x80
 	mov	r8,#0x1b
@@ -715,7 +814,7 @@ private_AES_set_decrypt_key:
 	moveq	pc,lr			@ be binary compatible with V4, yet
 	.word	0xe12fff1e			@ interoperable with Thumb ISA:-)
 #endif
-.size	private_AES_set_decrypt_key,.-private_AES_set_decrypt_key
+.size	AES_set_enc2dec_key,.-AES_set_enc2dec_key
 
 .type	AES_Td,%object
 .align	5
@@ -825,7 +924,11 @@ AES_Td:
 .type   AES_decrypt,%function
 .align	5
 AES_decrypt:
+#if __ARM_ARCH__<7
 	sub	r3,pc,#8		@ AES_decrypt
+#else
+	adr	r3,AES_decrypt
+#endif
 	stmdb   sp!,{r1,r4-r12,lr}
 	mov	r12,r0		@ inp
 	mov	r11,r2
@@ -1022,8 +1125,9 @@ _armv4_AES_decrypt:
 	ldrb	r6,[r10,r9]		@ Td4[s0>>0]
 	and	r9,lr,r1,lsr#8
 
+	add	r1,r10,r1,lsr#24
 	ldrb	r7,[r10,r7]		@ Td4[s1>>0]
-	ldrb	r1,[r10,r1,lsr#24]	@ Td4[s1>>24]
+	ldrb	r1,[r1]		@ Td4[s1>>24]
 	ldrb	r8,[r10,r8]		@ Td4[s1>>16]
 	eor	r0,r7,r0,lsl#24
 	ldrb	r9,[r10,r9]		@ Td4[s1>>8]
@@ -1036,7 +1140,8 @@ _armv4_AES_decrypt:
 	ldrb	r8,[r10,r8]		@ Td4[s2>>0]
 	and	r9,lr,r2,lsr#16
 
-	ldrb	r2,[r10,r2,lsr#24]	@ Td4[s2>>24]
+	add	r2,r10,r2,lsr#24
+	ldrb	r2,[r2]		@ Td4[s2>>24]
 	eor	r0,r0,r7,lsl#8
 	ldrb	r9,[r10,r9]		@ Td4[s2>>16]
 	eor	r1,r8,r1,lsl#16
@@ -1048,8 +1153,9 @@ _armv4_AES_decrypt:
 	ldrb	r8,[r10,r8]		@ Td4[s3>>8]
 	and	r9,lr,r3		@ i2
 
+	add	r3,r10,r3,lsr#24
 	ldrb	r9,[r10,r9]		@ Td4[s3>>0]
-	ldrb	r3,[r10,r3,lsr#24]	@ Td4[s3>>24]
+	ldrb	r3,[r3]		@ Td4[s3>>24]
 	eor	r0,r0,r7,lsl#16
 	ldr	r7,[r11,#0]
 	eor	r1,r1,r8,lsl#8
diff --git a/main/openssl/crypto/aes/asm/aesv8-armx-64.S b/main/openssl/crypto/aes/asm/aesv8-armx-64.S
new file mode 100644
index 00000000..be0a13df
--- /dev/null
+++ b/main/openssl/crypto/aes/asm/aesv8-armx-64.S
@@ -0,0 +1,761 @@
+#include "arm_arch.h"
+
+#if __ARM_ARCH__>=7
+.text
+.arch	armv8-a+crypto
+.align	5
+rcon:
+.long	0x01,0x01,0x01,0x01
+.long	0x0c0f0e0d,0x0c0f0e0d,0x0c0f0e0d,0x0c0f0e0d	// rotate-n-splat
+.long	0x1b,0x1b,0x1b,0x1b
+
+.globl	aes_v8_set_encrypt_key
+.type	aes_v8_set_encrypt_key,%function
+.align	5
+aes_v8_set_encrypt_key:
+.Lenc_key:
+	stp	x29,x30,[sp,#-16]!
+	add	x29,sp,#0
+	adr	x3,rcon
+	cmp	w1,#192
+
+	eor	v0.16b,v0.16b,v0.16b
+	ld1	{v3.16b},[x0],#16
+	mov	w1,#8		// reuse w1
+	ld1	{v1.4s,v2.4s},[x3],#32
+
+	b.lt	.Loop128
+	b.eq	.L192
+	b	.L256
+
+.align	4
+.Loop128:
+	tbl	v6.16b,{v3.16b},v2.16b
+	ext	v5.16b,v0.16b,v3.16b,#12
+	st1	{v3.4s},[x2],#16
+	aese	v6.16b,v0.16b
+	subs	w1,w1,#1
+
+	eor	v3.16b,v3.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	eor	v3.16b,v3.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	 eor	v6.16b,v6.16b,v1.16b
+	eor	v3.16b,v3.16b,v5.16b
+	shl	v1.16b,v1.16b,#1
+	eor	v3.16b,v3.16b,v6.16b
+	b.ne	.Loop128
+
+	ld1	{v1.4s},[x3]
+
+	tbl	v6.16b,{v3.16b},v2.16b
+	ext	v5.16b,v0.16b,v3.16b,#12
+	st1	{v3.4s},[x2],#16
+	aese	v6.16b,v0.16b
+
+	eor	v3.16b,v3.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	eor	v3.16b,v3.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	 eor	v6.16b,v6.16b,v1.16b
+	eor	v3.16b,v3.16b,v5.16b
+	shl	v1.16b,v1.16b,#1
+	eor	v3.16b,v3.16b,v6.16b
+
+	tbl	v6.16b,{v3.16b},v2.16b
+	ext	v5.16b,v0.16b,v3.16b,#12
+	st1	{v3.4s},[x2],#16
+	aese	v6.16b,v0.16b
+
+	eor	v3.16b,v3.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	eor	v3.16b,v3.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	 eor	v6.16b,v6.16b,v1.16b
+	eor	v3.16b,v3.16b,v5.16b
+	eor	v3.16b,v3.16b,v6.16b
+	st1	{v3.4s},[x2]
+	add	x2,x2,#0x50
+
+	mov	w12,#10
+	b	.Ldone
+
+.align	4
+.L192:
+	ld1	{v4.8b},[x0],#8
+	movi	v6.16b,#8			// borrow v6.16b
+	st1	{v3.4s},[x2],#16
+	sub	v2.16b,v2.16b,v6.16b	// adjust the mask
+
+.Loop192:
+	tbl	v6.16b,{v4.16b},v2.16b
+	ext	v5.16b,v0.16b,v3.16b,#12
+	st1	{v4.8b},[x2],#8
+	aese	v6.16b,v0.16b
+	subs	w1,w1,#1
+
+	eor	v3.16b,v3.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	eor	v3.16b,v3.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	eor	v3.16b,v3.16b,v5.16b
+
+	dup	v5.4s,v3.s[3]
+	eor	v5.16b,v5.16b,v4.16b
+	 eor	v6.16b,v6.16b,v1.16b
+	ext	v4.16b,v0.16b,v4.16b,#12
+	shl	v1.16b,v1.16b,#1
+	eor	v4.16b,v4.16b,v5.16b
+	eor	v3.16b,v3.16b,v6.16b
+	eor	v4.16b,v4.16b,v6.16b
+	st1	{v3.4s},[x2],#16
+	b.ne	.Loop192
+
+	mov	w12,#12
+	add	x2,x2,#0x20
+	b	.Ldone
+
+.align	4
+.L256:
+	ld1	{v4.16b},[x0]
+	mov	w1,#7
+	mov	w12,#14
+	st1	{v3.4s},[x2],#16
+
+.Loop256:
+	tbl	v6.16b,{v4.16b},v2.16b
+	ext	v5.16b,v0.16b,v3.16b,#12
+	st1	{v4.4s},[x2],#16
+	aese	v6.16b,v0.16b
+	subs	w1,w1,#1
+
+	eor	v3.16b,v3.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	eor	v3.16b,v3.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	 eor	v6.16b,v6.16b,v1.16b
+	eor	v3.16b,v3.16b,v5.16b
+	shl	v1.16b,v1.16b,#1
+	eor	v3.16b,v3.16b,v6.16b
+	st1	{v3.4s},[x2],#16
+	b.eq	.Ldone
+
+	dup	v6.4s,v3.s[3]		// just splat
+	ext	v5.16b,v0.16b,v4.16b,#12
+	aese	v6.16b,v0.16b
+
+	eor	v4.16b,v4.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	eor	v4.16b,v4.16b,v5.16b
+	ext	v5.16b,v0.16b,v5.16b,#12
+	eor	v4.16b,v4.16b,v5.16b
+
+	eor	v4.16b,v4.16b,v6.16b
+	b	.Loop256
+
+.Ldone:
+	str	w12,[x2]
+
+	eor	x0,x0,x0		// return value
+	ldr	x29,[sp],#16
+	ret
+.size	aes_v8_set_encrypt_key,.-aes_v8_set_encrypt_key
+
+.globl	aes_v8_set_decrypt_key
+.type	aes_v8_set_decrypt_key,%function
+.align	5
+aes_v8_set_decrypt_key:
+	stp	x29,x30,[sp,#-16]!
+	add	x29,sp,#0
+	bl	.Lenc_key
+
+	sub	x2,x2,#240		// restore original x2
+	mov	x4,#-16
+	add	x0,x2,x12,lsl#4	// end of key schedule
+
+	ld1	{v0.4s},[x2]
+	ld1	{v1.4s},[x0]
+	st1	{v0.4s},[x0],x4
+	st1	{v1.4s},[x2],#16
+
+.Loop_imc:
+	ld1	{v0.4s},[x2]
+	ld1	{v1.4s},[x0]
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	st1	{v0.4s},[x0],x4
+	st1	{v1.4s},[x2],#16
+	cmp	x0,x2
+	b.hi	.Loop_imc
+
+	ld1	{v0.4s},[x2]
+	aesimc	v0.16b,v0.16b
+	st1	{v0.4s},[x0]
+
+	eor	x0,x0,x0		// return value
+	ldp	x29,x30,[sp],#16
+	ret
+.size	aes_v8_set_decrypt_key,.-aes_v8_set_decrypt_key
+.globl	aes_v8_encrypt
+.type	aes_v8_encrypt,%function
+.align	5
+aes_v8_encrypt:
+	ldr	w3,[x2,#240]
+	ld1	{v0.4s},[x2],#16
+	ld1	{v2.16b},[x0]
+	sub	w3,w3,#2
+	ld1	{v1.4s},[x2],#16
+
+.Loop_enc:
+	aese	v2.16b,v0.16b
+	ld1	{v0.4s},[x2],#16
+	aesmc	v2.16b,v2.16b
+	subs	w3,w3,#2
+	aese	v2.16b,v1.16b
+	ld1	{v1.4s},[x2],#16
+	aesmc	v2.16b,v2.16b
+	b.gt	.Loop_enc
+
+	aese	v2.16b,v0.16b
+	ld1	{v0.4s},[x2]
+	aesmc	v2.16b,v2.16b
+	aese	v2.16b,v1.16b
+	eor	v2.16b,v2.16b,v0.16b
+
+	st1	{v2.16b},[x1]
+	ret
+.size	aes_v8_encrypt,.-aes_v8_encrypt
+.globl	aes_v8_decrypt
+.type	aes_v8_decrypt,%function
+.align	5
+aes_v8_decrypt:
+	ldr	w3,[x2,#240]
+	ld1	{v0.4s},[x2],#16
+	ld1	{v2.16b},[x0]
+	sub	w3,w3,#2
+	ld1	{v1.4s},[x2],#16
+
+.Loop_dec:
+	aesd	v2.16b,v0.16b
+	ld1	{v0.4s},[x2],#16
+	aesimc	v2.16b,v2.16b
+	subs	w3,w3,#2
+	aesd	v2.16b,v1.16b
+	ld1	{v1.4s},[x2],#16
+	aesimc	v2.16b,v2.16b
+	b.gt	.Loop_dec
+
+	aesd	v2.16b,v0.16b
+	ld1	{v0.4s},[x2]
+	aesimc	v2.16b,v2.16b
+	aesd	v2.16b,v1.16b
+	eor	v2.16b,v2.16b,v0.16b
+
+	st1	{v2.16b},[x1]
+	ret
+.size	aes_v8_decrypt,.-aes_v8_decrypt
+.globl	aes_v8_cbc_encrypt
+.type	aes_v8_cbc_encrypt,%function
+.align	5
+aes_v8_cbc_encrypt:
+	stp	x29,x30,[sp,#-16]!
+	add	x29,sp,#0
+	subs	x2,x2,#16
+	mov	x8,#16
+	b.lo	.Lcbc_abort
+	csel	x8,xzr,x8,eq
+
+	cmp	w5,#0			// en- or decrypting?
+	ldr	w5,[x3,#240]
+	and	x2,x2,#-16
+	ld1	{v6.16b},[x4]
+	ld1	{v0.16b},[x0],x8
+
+	ld1	{v16.4s-v17.4s},[x3]		// load key schedule...
+	sub	w5,w5,#6
+	add	x7,x3,x5,lsl#4	// pointer to last 7 round keys
+	sub	w5,w5,#2
+	ld1	{v18.4s-v19.4s},[x7],#32
+	ld1	{v20.4s-v21.4s},[x7],#32
+	ld1	{v22.4s-v23.4s},[x7],#32
+	ld1	{v7.4s},[x7]
+
+	add	x7,x3,#32
+	mov	w6,w5
+	b.eq	.Lcbc_dec
+
+	cmp	w5,#2
+	eor	v0.16b,v0.16b,v6.16b
+	eor	v5.16b,v16.16b,v7.16b
+	b.eq	.Lcbc_enc128
+
+.Loop_cbc_enc:
+	aese	v0.16b,v16.16b
+	ld1	{v16.4s},[x7],#16
+	aesmc	v0.16b,v0.16b
+	subs	w6,w6,#2
+	aese	v0.16b,v17.16b
+	ld1	{v17.4s},[x7],#16
+	aesmc	v0.16b,v0.16b
+	b.gt	.Loop_cbc_enc
+
+	aese	v0.16b,v16.16b
+	aesmc	v0.16b,v0.16b
+	 subs	x2,x2,#16
+	aese	v0.16b,v17.16b
+	aesmc	v0.16b,v0.16b
+	 csel	x8,xzr,x8,eq
+	aese	v0.16b,v18.16b
+	aesmc	v0.16b,v0.16b
+	 add	x7,x3,#16
+	aese	v0.16b,v19.16b
+	aesmc	v0.16b,v0.16b
+	 ld1	{v16.16b},[x0],x8
+	aese	v0.16b,v20.16b
+	aesmc	v0.16b,v0.16b
+	 eor	v16.16b,v16.16b,v5.16b
+	aese	v0.16b,v21.16b
+	aesmc	v0.16b,v0.16b
+	 ld1 {v17.4s},[x7],#16	// re-pre-load rndkey[1]
+	aese	v0.16b,v22.16b
+	aesmc	v0.16b,v0.16b
+	aese	v0.16b,v23.16b
+
+	 mov	w6,w5
+	eor	v6.16b,v0.16b,v7.16b
+	st1	{v6.16b},[x1],#16
+	b.hs	.Loop_cbc_enc
+
+	b	.Lcbc_done
+
+.align	5
+.Lcbc_enc128:
+	ld1	{v2.4s-v3.4s},[x7]
+	aese	v0.16b,v16.16b
+	aesmc	v0.16b,v0.16b
+	b	.Lenter_cbc_enc128
+.Loop_cbc_enc128:
+	aese	v0.16b,v16.16b
+	aesmc	v0.16b,v0.16b
+	 st1	{v6.16b},[x1],#16
+.Lenter_cbc_enc128:
+	aese	v0.16b,v17.16b
+	aesmc	v0.16b,v0.16b
+	 subs	x2,x2,#16
+	aese	v0.16b,v2.16b
+	aesmc	v0.16b,v0.16b
+	 csel	x8,xzr,x8,eq
+	aese	v0.16b,v3.16b
+	aesmc	v0.16b,v0.16b
+	aese	v0.16b,v18.16b
+	aesmc	v0.16b,v0.16b
+	aese	v0.16b,v19.16b
+	aesmc	v0.16b,v0.16b
+	 ld1	{v16.16b},[x0],x8
+	aese	v0.16b,v20.16b
+	aesmc	v0.16b,v0.16b
+	aese	v0.16b,v21.16b
+	aesmc	v0.16b,v0.16b
+	aese	v0.16b,v22.16b
+	aesmc	v0.16b,v0.16b
+	 eor	v16.16b,v16.16b,v5.16b
+	aese	v0.16b,v23.16b
+	eor	v6.16b,v0.16b,v7.16b
+	b.hs	.Loop_cbc_enc128
+
+	st1	{v6.16b},[x1],#16
+	b	.Lcbc_done
+
+.align	5
+.Lcbc_dec128:
+	ld1	{v4.4s-v5.4s},[x7]
+	eor	v6.16b,v6.16b,v7.16b
+	eor	v2.16b,v0.16b,v7.16b
+	mov	x12,x8
+
+.Loop2x_cbc_dec128:
+	aesd	v0.16b,v16.16b
+	aesd	v1.16b,v16.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	 subs	x2,x2,#32
+	aesd	v0.16b,v17.16b
+	aesd	v1.16b,v17.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	 csel	x8,xzr,x8,lo
+	aesd	v0.16b,v4.16b
+	aesd	v1.16b,v4.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	 csel	x12,xzr,x12,ls
+	aesd	v0.16b,v5.16b
+	aesd	v1.16b,v5.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	aesd	v0.16b,v18.16b
+	aesd	v1.16b,v18.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	aesd	v0.16b,v19.16b
+	aesd	v1.16b,v19.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	aesd	v0.16b,v20.16b
+	aesd	v1.16b,v20.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	aesd	v0.16b,v21.16b
+	aesd	v1.16b,v21.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	aesd	v0.16b,v22.16b
+	aesd	v1.16b,v22.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	aesd	v0.16b,v23.16b
+	aesd	v1.16b,v23.16b
+
+	eor	v6.16b,v6.16b,v0.16b
+	ld1	{v0.16b},[x0],x8
+	eor	v2.16b,v2.16b,v1.16b
+	ld1	{v1.16b},[x0],x12
+	st1	{v6.16b},[x1],#16
+	eor	v6.16b,v3.16b,v7.16b
+	st1	{v2.16b},[x1],#16
+	eor	v2.16b,v0.16b,v7.16b
+	orr	v3.16b,v1.16b,v1.16b
+	b.hs	.Loop2x_cbc_dec128
+
+	adds	x2,x2,#32
+	eor	v6.16b,v6.16b,v7.16b
+	b.eq	.Lcbc_done
+	eor	v2.16b,v2.16b,v7.16b
+	b	.Lcbc_dec_tail
+
+.align	5
+.Lcbc_dec:
+	subs	x2,x2,#16
+	orr	v2.16b,v0.16b,v0.16b
+	b.lo	.Lcbc_dec_tail
+
+	csel	x8,xzr,x8,eq
+	cmp	w5,#2
+	ld1	{v1.16b},[x0],x8
+	orr	v3.16b,v1.16b,v1.16b
+	b.eq	.Lcbc_dec128
+
+.Loop2x_cbc_dec:
+	aesd	v0.16b,v16.16b
+	aesd	v1.16b,v16.16b
+	ld1	{v16.4s},[x7],#16
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	subs	w6,w6,#2
+	aesd	v0.16b,v17.16b
+	aesd	v1.16b,v17.16b
+	ld1	{v17.4s},[x7],#16
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	b.gt	.Loop2x_cbc_dec
+
+	aesd	v0.16b,v16.16b
+	aesd	v1.16b,v16.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	 eor	v4.16b,v6.16b,v7.16b
+	 eor	v5.16b,v2.16b,v7.16b
+	aesd	v0.16b,v17.16b
+	aesd	v1.16b,v17.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	 orr	v6.16b,v3.16b,v3.16b
+	 subs	x2,x2,#32
+	aesd	v0.16b,v18.16b
+	aesd	v1.16b,v18.16b
+	aesimc	v0.16b,v0.16b
+	 csel	x8,xzr,x8,lo
+	aesimc	v1.16b,v1.16b
+	 mov	x7,x3
+	aesd	v0.16b,v19.16b
+	aesd	v1.16b,v19.16b
+	aesimc	v0.16b,v0.16b
+	 ld1	{v2.16b},[x0],x8
+	aesimc	v1.16b,v1.16b
+	 csel	x8,xzr,x8,ls
+	aesd	v0.16b,v20.16b
+	aesd	v1.16b,v20.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	 ld1	{v3.16b},[x0],x8
+	aesd	v0.16b,v21.16b
+	aesd	v1.16b,v21.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	 ld1 {v16.4s},[x7],#16	// re-pre-load rndkey[0]
+	aesd	v0.16b,v22.16b
+	aesd	v1.16b,v22.16b
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	 ld1 {v17.4s},[x7],#16	// re-pre-load rndkey[1]
+	aesd	v0.16b,v23.16b
+	aesd	v1.16b,v23.16b
+
+	 mov	w6,w5
+	eor	v4.16b,v4.16b,v0.16b
+	eor	v5.16b,v5.16b,v1.16b
+	 orr	v0.16b,v2.16b,v2.16b
+	st1	{v4.16b},[x1],#16
+	 orr	v1.16b,v3.16b,v3.16b
+	st1	{v5.16b},[x1],#16
+	b.hs	.Loop2x_cbc_dec
+
+	adds	x2,x2,#32
+	b.eq	.Lcbc_done
+
+.Lcbc_dec_tail:
+	aesd	v0.16b,v16.16b
+	ld1	{v16.4s},[x7],#16
+	aesimc	v0.16b,v0.16b
+	subs	w6,w6,#2
+	aesd	v0.16b,v17.16b
+	ld1	{v17.4s},[x7],#16
+	aesimc	v0.16b,v0.16b
+	b.gt	.Lcbc_dec_tail
+
+	aesd	v0.16b,v16.16b
+	aesimc	v0.16b,v0.16b
+	aesd	v0.16b,v17.16b
+	aesimc	v0.16b,v0.16b
+	 eor	v4.16b,v6.16b,v7.16b
+	aesd	v0.16b,v18.16b
+	aesimc	v0.16b,v0.16b
+	 orr	v6.16b,v2.16b,v2.16b
+	aesd	v0.16b,v19.16b
+	aesimc	v0.16b,v0.16b
+	aesd	v0.16b,v20.16b
+	aesimc	v0.16b,v0.16b
+	aesd	v0.16b,v21.16b
+	aesimc	v0.16b,v0.16b
+	aesd	v0.16b,v22.16b
+	aesimc	v0.16b,v0.16b
+	aesd	v0.16b,v23.16b
+
+	eor	v4.16b,v4.16b,v0.16b
+	st1	{v4.16b},[x1],#16
+
+.Lcbc_done:
+	st1	{v6.16b},[x4]
+.Lcbc_abort:
+	ldr	x29,[sp],#16
+	ret
+.size	aes_v8_cbc_encrypt,.-aes_v8_cbc_encrypt
+.globl	aes_v8_ctr32_encrypt_blocks
+.type	aes_v8_ctr32_encrypt_blocks,%function
+.align	5
+aes_v8_ctr32_encrypt_blocks:
+	stp		x29,x30,[sp,#-16]!
+	add		x29,sp,#0
+	ldr		w5,[x3,#240]
+
+	ldr		w8, [x4, #12]
+	ld1		{v0.4s},[x4]
+
+	ld1		{v16.4s-v17.4s},[x3]		// load key schedule...
+	sub		w5,w5,#6
+	add		x7,x3,x5,lsl#4	// pointer to last 7 round keys
+	sub		w5,w5,#2
+	ld1		{v18.4s-v19.4s},[x7],#32
+	ld1		{v20.4s-v21.4s},[x7],#32
+	ld1		{v22.4s-v23.4s},[x7],#32
+	ld1		{v7.4s},[x7]
+
+	add		x7,x3,#32
+	mov		w6,w5
+
+	subs		x2,x2,#2
+	b.lo		.Lctr32_tail
+
+#ifndef __ARMEB__
+	rev		w8, w8
+#endif
+	orr		v1.16b,v0.16b,v0.16b
+	add		w8, w8, #1
+	orr		v6.16b,v0.16b,v0.16b
+	rev		w10, w8
+	cmp		w5,#2
+	mov		v1.s[3],w10
+	b.eq		.Lctr32_128
+
+.Loop2x_ctr32:
+	aese		v0.16b,v16.16b
+	aese		v1.16b,v16.16b
+	ld1		{v16.4s},[x7],#16
+	aesmc		v0.16b,v0.16b
+	aesmc		v1.16b,v1.16b
+	subs		w6,w6,#2
+	aese		v0.16b,v17.16b
+	aese		v1.16b,v17.16b
+	ld1		{v17.4s},[x7],#16
+	aesmc		v0.16b,v0.16b
+	aesmc		v1.16b,v1.16b
+	b.gt		.Loop2x_ctr32
+
+	aese		v0.16b,v16.16b
+	aese		v1.16b,v16.16b
+	aesmc		v4.16b,v0.16b
+	 orr		v0.16b,v6.16b,v6.16b
+	aesmc		v5.16b,v1.16b
+	 orr		v1.16b,v6.16b,v6.16b
+	aese		v4.16b,v17.16b
+	aese		v5.16b,v17.16b
+	 ld1		{v2.16b},[x0],#16
+	aesmc		v4.16b,v4.16b
+	 ld1		{v3.16b},[x0],#16
+	aesmc		v5.16b,v5.16b
+	 add		w8,w8,#1
+	aese		v4.16b,v18.16b
+	aese		v5.16b,v18.16b
+	 rev		w9,w8
+	aesmc		v4.16b,v4.16b
+	aesmc		v5.16b,v5.16b
+	 add		w8,w8,#1
+	aese		v4.16b,v19.16b
+	aese		v5.16b,v19.16b
+	 eor		v2.16b,v2.16b,v7.16b
+	 rev		w10,w8
+	aesmc		v4.16b,v4.16b
+	aesmc		v5.16b,v5.16b
+	 eor		v3.16b,v3.16b,v7.16b
+	 mov		x7,x3
+	aese		v4.16b,v20.16b
+	aese		v5.16b,v20.16b
+	 subs		x2,x2,#2
+	aesmc		v4.16b,v4.16b
+	aesmc		v5.16b,v5.16b
+	 ld1	 {v16.4s-v17.4s},[x7],#32	// re-pre-load rndkey[0-1]
+	aese		v4.16b,v21.16b
+	aese		v5.16b,v21.16b
+	aesmc		v4.16b,v4.16b
+	aesmc		v5.16b,v5.16b
+	aese		v4.16b,v22.16b
+	aese		v5.16b,v22.16b
+	 mov	v0.s[3], w9
+	aesmc		v4.16b,v4.16b
+	 mov	v1.s[3], w10
+	aesmc		v5.16b,v5.16b
+	aese		v4.16b,v23.16b
+	aese		v5.16b,v23.16b
+
+	 mov		w6,w5
+	eor		v2.16b,v2.16b,v4.16b
+	eor		v3.16b,v3.16b,v5.16b
+	st1		{v2.16b},[x1],#16
+	st1		{v3.16b},[x1],#16
+	b.hs		.Loop2x_ctr32
+
+	adds		x2,x2,#2
+	b.eq		.Lctr32_done
+	b		.Lctr32_tail
+
+.Lctr32_128:
+	ld1		{v4.4s-v5.4s},[x7]
+
+.Loop2x_ctr32_128:
+	aese		v0.16b,v16.16b
+	aese		v1.16b,v16.16b
+	aesmc		v0.16b,v0.16b
+	 ld1		{v2.16b},[x0],#16
+	aesmc		v1.16b,v1.16b
+	 ld1		{v3.16b},[x0],#16
+	aese		v0.16b,v17.16b
+	aese		v1.16b,v17.16b
+	 add		w8,w8,#1
+	aesmc		v0.16b,v0.16b
+	aesmc		v1.16b,v1.16b
+	 rev		w9,w8
+	aese		v0.16b,v4.16b
+	aese		v1.16b,v4.16b
+	 add		w8,w8,#1
+	aesmc		v0.16b,v0.16b
+	aesmc		v1.16b,v1.16b
+	 rev		w10,w8
+	aese		v0.16b,v5.16b
+	aese		v1.16b,v5.16b
+	 subs		x2,x2,#2
+	aesmc		v0.16b,v0.16b
+	aesmc		v1.16b,v1.16b
+	aese		v0.16b,v18.16b
+	aese		v1.16b,v18.16b
+	aesmc		v0.16b,v0.16b
+	aesmc		v1.16b,v1.16b
+	aese		v0.16b,v19.16b
+	aese		v1.16b,v19.16b
+	aesmc		v0.16b,v0.16b
+	aesmc		v1.16b,v1.16b
+	aese		v0.16b,v20.16b
+	aese		v1.16b,v20.16b
+	aesmc		v0.16b,v0.16b
+	aesmc		v1.16b,v1.16b
+	aese		v0.16b,v21.16b
+	aese		v1.16b,v21.16b
+	aesmc		v0.16b,v0.16b
+	aesmc		v1.16b,v1.16b
+	aese		v0.16b,v22.16b
+	aese		v1.16b,v22.16b
+	aesmc		v0.16b,v0.16b
+	aesmc		v1.16b,v1.16b
+	 eor		v2.16b,v2.16b,v7.16b
+	aese		v0.16b,v23.16b
+	 eor		v3.16b,v3.16b,v7.16b
+	aese		v1.16b,v23.16b
+
+	eor		v2.16b,v2.16b,v0.16b
+	orr		v0.16b,v6.16b,v6.16b
+	eor		v3.16b,v3.16b,v1.16b
+	orr		v1.16b,v6.16b,v6.16b
+	st1		{v2.16b},[x1],#16
+	mov		v0.s[3], w9
+	st1		{v3.16b},[x1],#16
+	mov		v1.s[3], w10
+	b.hs		.Loop2x_ctr32_128
+
+	adds		x2,x2,#2
+	b.eq		.Lctr32_done
+
+.Lctr32_tail:
+	aese		v0.16b,v16.16b
+	ld1		{v16.4s},[x7],#16
+	aesmc		v0.16b,v0.16b
+	subs		w6,w6,#2
+	aese		v0.16b,v17.16b
+	ld1		{v17.4s},[x7],#16
+	aesmc		v0.16b,v0.16b
+	b.gt		.Lctr32_tail
+
+	aese		v0.16b,v16.16b
+	aesmc		v0.16b,v0.16b
+	aese		v0.16b,v17.16b
+	aesmc		v0.16b,v0.16b
+	 ld1		{v2.16b},[x0]
+	aese		v0.16b,v18.16b
+	aesmc		v0.16b,v0.16b
+	aese		v0.16b,v19.16b
+	aesmc		v0.16b,v0.16b
+	aese		v0.16b,v20.16b
+	aesmc		v0.16b,v0.16b
+	aese		v0.16b,v21.16b
+	aesmc		v0.16b,v0.16b
+	aese		v0.16b,v22.16b
+	aesmc		v0.16b,v0.16b
+	 eor		v2.16b,v2.16b,v7.16b
+	aese		v0.16b,v23.16b
+
+	eor		v2.16b,v2.16b,v0.16b
+	st1		{v2.16b},[x1]
+
+.Lctr32_done:
+	ldr		x29,[sp],#16
+	ret
+.size	aes_v8_ctr32_encrypt_blocks,.-aes_v8_ctr32_encrypt_blocks
+#endif
diff --git a/main/openssl/crypto/aes/asm/aesv8-armx.S b/main/openssl/crypto/aes/asm/aesv8-armx.S
new file mode 100644
index 00000000..1637e4d4
--- /dev/null
+++ b/main/openssl/crypto/aes/asm/aesv8-armx.S
@@ -0,0 +1,767 @@
+#include "arm_arch.h"
+
+#if __ARM_ARCH__>=7
+.text
+.fpu	neon
+.code	32
+.align	5
+rcon:
+.long	0x01,0x01,0x01,0x01
+.long	0x0c0f0e0d,0x0c0f0e0d,0x0c0f0e0d,0x0c0f0e0d	@ rotate-n-splat
+.long	0x1b,0x1b,0x1b,0x1b
+
+.globl	aes_v8_set_encrypt_key
+.type	aes_v8_set_encrypt_key,%function
+.align	5
+aes_v8_set_encrypt_key:
+.Lenc_key:
+	adr	r3,rcon
+	cmp	r1,#192
+
+	veor	q0,q0,q0
+	vld1.8	{q3},[r0]!
+	mov	r1,#8		@ reuse r1
+	vld1.32	{q1,q2},[r3]!
+
+	blt	.Loop128
+	beq	.L192
+	b	.L256
+
+.align	4
+.Loop128:
+	vtbl.8	d20,{q3},d4
+	vtbl.8	d21,{q3},d5
+	vext.8	q9,q0,q3,#12
+	vst1.32	{q3},[r2]!
+	.byte	0x00,0x43,0xf0,0xf3	@ aese q10,q0
+	subs	r1,r1,#1
+
+	veor	q3,q3,q9
+	vext.8	q9,q0,q9,#12
+	veor	q3,q3,q9
+	vext.8	q9,q0,q9,#12
+	 veor	q10,q10,q1
+	veor	q3,q3,q9
+	vshl.u8	q1,q1,#1
+	veor	q3,q3,q10
+	bne	.Loop128
+
+	vld1.32	{q1},[r3]
+
+	vtbl.8	d20,{q3},d4
+	vtbl.8	d21,{q3},d5
+	vext.8	q9,q0,q3,#12
+	vst1.32	{q3},[r2]!
+	.byte	0x00,0x43,0xf0,0xf3	@ aese q10,q0
+
+	veor	q3,q3,q9
+	vext.8	q9,q0,q9,#12
+	veor	q3,q3,q9
+	vext.8	q9,q0,q9,#12
+	 veor	q10,q10,q1
+	veor	q3,q3,q9
+	vshl.u8	q1,q1,#1
+	veor	q3,q3,q10
+
+	vtbl.8	d20,{q3},d4
+	vtbl.8	d21,{q3},d5
+	vext.8	q9,q0,q3,#12
+	vst1.32	{q3},[r2]!
+	.byte	0x00,0x43,0xf0,0xf3	@ aese q10,q0
+
+	veor	q3,q3,q9
+	vext.8	q9,q0,q9,#12
+	veor	q3,q3,q9
+	vext.8	q9,q0,q9,#12
+	 veor	q10,q10,q1
+	veor	q3,q3,q9
+	veor	q3,q3,q10
+	vst1.32	{q3},[r2]
+	add	r2,r2,#0x50
+
+	mov	r12,#10
+	b	.Ldone
+
+.align	4
+.L192:
+	vld1.8	{d16},[r0]!
+	vmov.i8	q10,#8			@ borrow q10
+	vst1.32	{q3},[r2]!
+	vsub.i8	q2,q2,q10	@ adjust the mask
+
+.Loop192:
+	vtbl.8	d20,{q8},d4
+	vtbl.8	d21,{q8},d5
+	vext.8	q9,q0,q3,#12
+	vst1.32	{d16},[r2]!
+	.byte	0x00,0x43,0xf0,0xf3	@ aese q10,q0
+	subs	r1,r1,#1
+
+	veor	q3,q3,q9
+	vext.8	q9,q0,q9,#12
+	veor	q3,q3,q9
+	vext.8	q9,q0,q9,#12
+	veor	q3,q3,q9
+
+	vdup.32	q9,d7[1]
+	veor	q9,q9,q8
+	 veor	q10,q10,q1
+	vext.8	q8,q0,q8,#12
+	vshl.u8	q1,q1,#1
+	veor	q8,q8,q9
+	veor	q3,q3,q10
+	veor	q8,q8,q10
+	vst1.32	{q3},[r2]!
+	bne	.Loop192
+
+	mov	r12,#12
+	add	r2,r2,#0x20
+	b	.Ldone
+
+.align	4
+.L256:
+	vld1.8	{q8},[r0]
+	mov	r1,#7
+	mov	r12,#14
+	vst1.32	{q3},[r2]!
+
+.Loop256:
+	vtbl.8	d20,{q8},d4
+	vtbl.8	d21,{q8},d5
+	vext.8	q9,q0,q3,#12
+	vst1.32	{q8},[r2]!
+	.byte	0x00,0x43,0xf0,0xf3	@ aese q10,q0
+	subs	r1,r1,#1
+
+	veor	q3,q3,q9
+	vext.8	q9,q0,q9,#12
+	veor	q3,q3,q9
+	vext.8	q9,q0,q9,#12
+	 veor	q10,q10,q1
+	veor	q3,q3,q9
+	vshl.u8	q1,q1,#1
+	veor	q3,q3,q10
+	vst1.32	{q3},[r2]!
+	beq	.Ldone
+
+	vdup.32	q10,d7[1]
+	vext.8	q9,q0,q8,#12
+	.byte	0x00,0x43,0xf0,0xf3	@ aese q10,q0
+
+	veor	q8,q8,q9
+	vext.8	q9,q0,q9,#12
+	veor	q8,q8,q9
+	vext.8	q9,q0,q9,#12
+	veor	q8,q8,q9
+
+	veor	q8,q8,q10
+	b	.Loop256
+
+.Ldone:
+	str	r12,[r2]
+
+	eor	r0,r0,r0		@ return value
+	
+	bx	lr
+.size	aes_v8_set_encrypt_key,.-aes_v8_set_encrypt_key
+
+.globl	aes_v8_set_decrypt_key
+.type	aes_v8_set_decrypt_key,%function
+.align	5
+aes_v8_set_decrypt_key:
+	stmdb	sp!,{r4,lr}
+	bl	.Lenc_key
+
+	sub	r2,r2,#240		@ restore original r2
+	mov	r4,#-16
+	add	r0,r2,r12,lsl#4	@ end of key schedule
+
+	vld1.32	{q0},[r2]
+	vld1.32	{q1},[r0]
+	vst1.32	{q0},[r0],r4
+	vst1.32	{q1},[r2]!
+
+.Loop_imc:
+	vld1.32	{q0},[r2]
+	vld1.32	{q1},[r0]
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	vst1.32	{q0},[r0],r4
+	vst1.32	{q1},[r2]!
+	cmp	r0,r2
+	bhi	.Loop_imc
+
+	vld1.32	{q0},[r2]
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	vst1.32	{q0},[r0]
+
+	eor	r0,r0,r0		@ return value
+	ldmia	sp!,{r4,pc}
+.size	aes_v8_set_decrypt_key,.-aes_v8_set_decrypt_key
+.globl	aes_v8_encrypt
+.type	aes_v8_encrypt,%function
+.align	5
+aes_v8_encrypt:
+	ldr	r3,[r2,#240]
+	vld1.32	{q0},[r2]!
+	vld1.8	{q2},[r0]
+	sub	r3,r3,#2
+	vld1.32	{q1},[r2]!
+
+.Loop_enc:
+	.byte	0x00,0x43,0xb0,0xf3	@ aese q2,q0
+	vld1.32	{q0},[r2]!
+	.byte	0x84,0x43,0xb0,0xf3	@ aesmc q2,q2
+	subs	r3,r3,#2
+	.byte	0x02,0x43,0xb0,0xf3	@ aese q2,q1
+	vld1.32	{q1},[r2]!
+	.byte	0x84,0x43,0xb0,0xf3	@ aesmc q2,q2
+	bgt	.Loop_enc
+
+	.byte	0x00,0x43,0xb0,0xf3	@ aese q2,q0
+	vld1.32	{q0},[r2]
+	.byte	0x84,0x43,0xb0,0xf3	@ aesmc q2,q2
+	.byte	0x02,0x43,0xb0,0xf3	@ aese q2,q1
+	veor	q2,q2,q0
+
+	vst1.8	{q2},[r1]
+	bx	lr
+.size	aes_v8_encrypt,.-aes_v8_encrypt
+.globl	aes_v8_decrypt
+.type	aes_v8_decrypt,%function
+.align	5
+aes_v8_decrypt:
+	ldr	r3,[r2,#240]
+	vld1.32	{q0},[r2]!
+	vld1.8	{q2},[r0]
+	sub	r3,r3,#2
+	vld1.32	{q1},[r2]!
+
+.Loop_dec:
+	.byte	0x40,0x43,0xb0,0xf3	@ aesd q2,q0
+	vld1.32	{q0},[r2]!
+	.byte	0xc4,0x43,0xb0,0xf3	@ aesimc q2,q2
+	subs	r3,r3,#2
+	.byte	0x42,0x43,0xb0,0xf3	@ aesd q2,q1
+	vld1.32	{q1},[r2]!
+	.byte	0xc4,0x43,0xb0,0xf3	@ aesimc q2,q2
+	bgt	.Loop_dec
+
+	.byte	0x40,0x43,0xb0,0xf3	@ aesd q2,q0
+	vld1.32	{q0},[r2]
+	.byte	0xc4,0x43,0xb0,0xf3	@ aesimc q2,q2
+	.byte	0x42,0x43,0xb0,0xf3	@ aesd q2,q1
+	veor	q2,q2,q0
+
+	vst1.8	{q2},[r1]
+	bx	lr
+.size	aes_v8_decrypt,.-aes_v8_decrypt
+.globl	aes_v8_cbc_encrypt
+.type	aes_v8_cbc_encrypt,%function
+.align	5
+aes_v8_cbc_encrypt:
+	mov	ip,sp
+	stmdb	sp!,{r4-r8,lr}
+	vstmdb	sp!,{d8-d15}            @ ABI specification says so
+	ldmia	ip,{r4-r5}		@ load remaining args
+	subs	r2,r2,#16
+	mov	r8,#16
+	blo	.Lcbc_abort
+	moveq	r8,#0
+
+	cmp	r5,#0			@ en- or decrypting?
+	ldr	r5,[r3,#240]
+	and	r2,r2,#-16
+	vld1.8	{q6},[r4]
+	vld1.8	{q0},[r0],r8
+
+	vld1.32	{q8-q9},[r3]		@ load key schedule...
+	sub	r5,r5,#6
+	add	r7,r3,r5,lsl#4	@ pointer to last 7 round keys
+	sub	r5,r5,#2
+	vld1.32	{q10-q11},[r7]!
+	vld1.32	{q12-q13},[r7]!
+	vld1.32	{q14-q15},[r7]!
+	vld1.32	{q7},[r7]
+
+	add	r7,r3,#32
+	mov	r6,r5
+	beq	.Lcbc_dec
+
+	cmp	r5,#2
+	veor	q0,q0,q6
+	veor	q5,q8,q7
+	beq	.Lcbc_enc128
+
+.Loop_cbc_enc:
+	.byte	0x20,0x03,0xb0,0xf3	@ aese q0,q8
+	vld1.32	{q8},[r7]!
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	subs	r6,r6,#2
+	.byte	0x22,0x03,0xb0,0xf3	@ aese q0,q9
+	vld1.32	{q9},[r7]!
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	bgt	.Loop_cbc_enc
+
+	.byte	0x20,0x03,0xb0,0xf3	@ aese q0,q8
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 subs	r2,r2,#16
+	.byte	0x22,0x03,0xb0,0xf3	@ aese q0,q9
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 moveq	r8,#0
+	.byte	0x24,0x03,0xb0,0xf3	@ aese q0,q10
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 add	r7,r3,#16
+	.byte	0x26,0x03,0xb0,0xf3	@ aese q0,q11
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 vld1.8	{q8},[r0],r8
+	.byte	0x28,0x03,0xb0,0xf3	@ aese q0,q12
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 veor	q8,q8,q5
+	.byte	0x2a,0x03,0xb0,0xf3	@ aese q0,q13
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 vld1.32 {q9},[r7]!	@ re-pre-load rndkey[1]
+	.byte	0x2c,0x03,0xb0,0xf3	@ aese q0,q14
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x2e,0x03,0xb0,0xf3	@ aese q0,q15
+
+	 mov	r6,r5
+	veor	q6,q0,q7
+	vst1.8	{q6},[r1]!
+	bhs	.Loop_cbc_enc
+
+	b	.Lcbc_done
+
+.align	5
+.Lcbc_enc128:
+	vld1.32	{q2-q3},[r7]
+	.byte	0x20,0x03,0xb0,0xf3	@ aese q0,q8
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	b	.Lenter_cbc_enc128
+.Loop_cbc_enc128:
+	.byte	0x20,0x03,0xb0,0xf3	@ aese q0,q8
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 vst1.8	{q6},[r1]!
+.Lenter_cbc_enc128:
+	.byte	0x22,0x03,0xb0,0xf3	@ aese q0,q9
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 subs	r2,r2,#16
+	.byte	0x04,0x03,0xb0,0xf3	@ aese q0,q2
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 moveq	r8,#0
+	.byte	0x06,0x03,0xb0,0xf3	@ aese q0,q3
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x24,0x03,0xb0,0xf3	@ aese q0,q10
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x26,0x03,0xb0,0xf3	@ aese q0,q11
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 vld1.8	{q8},[r0],r8
+	.byte	0x28,0x03,0xb0,0xf3	@ aese q0,q12
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x2a,0x03,0xb0,0xf3	@ aese q0,q13
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x2c,0x03,0xb0,0xf3	@ aese q0,q14
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 veor	q8,q8,q5
+	.byte	0x2e,0x03,0xb0,0xf3	@ aese q0,q15
+	veor	q6,q0,q7
+	bhs	.Loop_cbc_enc128
+
+	vst1.8	{q6},[r1]!
+	b	.Lcbc_done
+
+.align	5
+.Lcbc_dec128:
+	vld1.32	{q4-q5},[r7]
+	veor	q6,q6,q7
+	veor	q2,q0,q7
+	mov	r12,r8
+
+.Loop2x_cbc_dec128:
+	.byte	0x60,0x03,0xb0,0xf3	@ aesd q0,q8
+	.byte	0x60,0x23,0xb0,0xf3	@ aesd q1,q8
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	 subs	r2,r2,#32
+	.byte	0x62,0x03,0xb0,0xf3	@ aesd q0,q9
+	.byte	0x62,0x23,0xb0,0xf3	@ aesd q1,q9
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	 movlo	r8,#0
+	.byte	0x48,0x03,0xb0,0xf3	@ aesd q0,q4
+	.byte	0x48,0x23,0xb0,0xf3	@ aesd q1,q4
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	 movls	r12,#0
+	.byte	0x4a,0x03,0xb0,0xf3	@ aesd q0,q5
+	.byte	0x4a,0x23,0xb0,0xf3	@ aesd q1,q5
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	.byte	0x64,0x03,0xb0,0xf3	@ aesd q0,q10
+	.byte	0x64,0x23,0xb0,0xf3	@ aesd q1,q10
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	.byte	0x66,0x03,0xb0,0xf3	@ aesd q0,q11
+	.byte	0x66,0x23,0xb0,0xf3	@ aesd q1,q11
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	.byte	0x68,0x03,0xb0,0xf3	@ aesd q0,q12
+	.byte	0x68,0x23,0xb0,0xf3	@ aesd q1,q12
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	.byte	0x6a,0x03,0xb0,0xf3	@ aesd q0,q13
+	.byte	0x6a,0x23,0xb0,0xf3	@ aesd q1,q13
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	.byte	0x6c,0x03,0xb0,0xf3	@ aesd q0,q14
+	.byte	0x6c,0x23,0xb0,0xf3	@ aesd q1,q14
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	.byte	0x6e,0x03,0xb0,0xf3	@ aesd q0,q15
+	.byte	0x6e,0x23,0xb0,0xf3	@ aesd q1,q15
+
+	veor	q6,q6,q0
+	vld1.8	{q0},[r0],r8
+	veor	q2,q2,q1
+	vld1.8	{q1},[r0],r12
+	vst1.8	{q6},[r1]!
+	veor	q6,q3,q7
+	vst1.8	{q2},[r1]!
+	veor	q2,q0,q7
+	vorr	q3,q1,q1
+	bhs	.Loop2x_cbc_dec128
+
+	adds	r2,r2,#32
+	veor	q6,q6,q7
+	beq	.Lcbc_done
+	veor	q2,q2,q7
+	b	.Lcbc_dec_tail
+
+.align	5
+.Lcbc_dec:
+	subs	r2,r2,#16
+	vorr	q2,q0,q0
+	blo	.Lcbc_dec_tail
+
+	moveq	r8,#0
+	cmp	r5,#2
+	vld1.8	{q1},[r0],r8
+	vorr	q3,q1,q1
+	beq	.Lcbc_dec128
+
+.Loop2x_cbc_dec:
+	.byte	0x60,0x03,0xb0,0xf3	@ aesd q0,q8
+	.byte	0x60,0x23,0xb0,0xf3	@ aesd q1,q8
+	vld1.32	{q8},[r7]!
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	subs	r6,r6,#2
+	.byte	0x62,0x03,0xb0,0xf3	@ aesd q0,q9
+	.byte	0x62,0x23,0xb0,0xf3	@ aesd q1,q9
+	vld1.32	{q9},[r7]!
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	bgt	.Loop2x_cbc_dec
+
+	.byte	0x60,0x03,0xb0,0xf3	@ aesd q0,q8
+	.byte	0x60,0x23,0xb0,0xf3	@ aesd q1,q8
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	 veor	q4,q6,q7
+	 veor	q5,q2,q7
+	.byte	0x62,0x03,0xb0,0xf3	@ aesd q0,q9
+	.byte	0x62,0x23,0xb0,0xf3	@ aesd q1,q9
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	 vorr	q6,q3,q3
+	 subs	r2,r2,#32
+	.byte	0x64,0x03,0xb0,0xf3	@ aesd q0,q10
+	.byte	0x64,0x23,0xb0,0xf3	@ aesd q1,q10
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	 movlo	r8,#0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	 mov	r7,r3
+	.byte	0x66,0x03,0xb0,0xf3	@ aesd q0,q11
+	.byte	0x66,0x23,0xb0,0xf3	@ aesd q1,q11
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	 vld1.8	{q2},[r0],r8
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	 movls	r8,#0
+	.byte	0x68,0x03,0xb0,0xf3	@ aesd q0,q12
+	.byte	0x68,0x23,0xb0,0xf3	@ aesd q1,q12
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	 vld1.8	{q3},[r0],r8
+	.byte	0x6a,0x03,0xb0,0xf3	@ aesd q0,q13
+	.byte	0x6a,0x23,0xb0,0xf3	@ aesd q1,q13
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	 vld1.32 {q8},[r7]!	@ re-pre-load rndkey[0]
+	.byte	0x6c,0x03,0xb0,0xf3	@ aesd q0,q14
+	.byte	0x6c,0x23,0xb0,0xf3	@ aesd q1,q14
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0xc2,0x23,0xb0,0xf3	@ aesimc q1,q1
+	 vld1.32 {q9},[r7]!	@ re-pre-load rndkey[1]
+	.byte	0x6e,0x03,0xb0,0xf3	@ aesd q0,q15
+	.byte	0x6e,0x23,0xb0,0xf3	@ aesd q1,q15
+
+	 mov	r6,r5
+	veor	q4,q4,q0
+	veor	q5,q5,q1
+	 vorr	q0,q2,q2
+	vst1.8	{q4},[r1]!
+	 vorr	q1,q3,q3
+	vst1.8	{q5},[r1]!
+	bhs	.Loop2x_cbc_dec
+
+	adds	r2,r2,#32
+	beq	.Lcbc_done
+
+.Lcbc_dec_tail:
+	.byte	0x60,0x03,0xb0,0xf3	@ aesd q0,q8
+	vld1.32	{q8},[r7]!
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	subs	r6,r6,#2
+	.byte	0x62,0x03,0xb0,0xf3	@ aesd q0,q9
+	vld1.32	{q9},[r7]!
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	bgt	.Lcbc_dec_tail
+
+	.byte	0x60,0x03,0xb0,0xf3	@ aesd q0,q8
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0x62,0x03,0xb0,0xf3	@ aesd q0,q9
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	 veor	q4,q6,q7
+	.byte	0x64,0x03,0xb0,0xf3	@ aesd q0,q10
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	 vorr	q6,q2,q2
+	.byte	0x66,0x03,0xb0,0xf3	@ aesd q0,q11
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0x68,0x03,0xb0,0xf3	@ aesd q0,q12
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0x6a,0x03,0xb0,0xf3	@ aesd q0,q13
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0x6c,0x03,0xb0,0xf3	@ aesd q0,q14
+	.byte	0xc0,0x03,0xb0,0xf3	@ aesimc q0,q0
+	.byte	0x6e,0x03,0xb0,0xf3	@ aesd q0,q15
+
+	veor	q4,q4,q0
+	vst1.8	{q4},[r1]!
+
+.Lcbc_done:
+	vst1.8	{q6},[r4]
+.Lcbc_abort:
+	vldmia	sp!,{d8-d15}
+	ldmia	sp!,{r4-r8,pc}
+.size	aes_v8_cbc_encrypt,.-aes_v8_cbc_encrypt
+.globl	aes_v8_ctr32_encrypt_blocks
+.type	aes_v8_ctr32_encrypt_blocks,%function
+.align	5
+aes_v8_ctr32_encrypt_blocks:
+	mov		ip,sp
+	stmdb		sp!,{r4-r10,lr}
+	vstmdb		sp!,{d8-d15}            @ ABI specification says so
+	ldr		r4, [ip]		@ load remaining arg
+	ldr		r5,[r3,#240]
+
+	ldr		r8, [r4, #12]
+	vld1.32		{q0},[r4]
+
+	vld1.32		{q8-q9},[r3]		@ load key schedule...
+	sub		r5,r5,#6
+	add		r7,r3,r5,lsl#4	@ pointer to last 7 round keys
+	sub		r5,r5,#2
+	vld1.32		{q10-q11},[r7]!
+	vld1.32		{q12-q13},[r7]!
+	vld1.32		{q14-q15},[r7]!
+	vld1.32		{q7},[r7]
+
+	add		r7,r3,#32
+	mov		r6,r5
+
+	subs		r2,r2,#2
+	blo		.Lctr32_tail
+
+#ifndef __ARMEB__
+	rev		r8, r8
+#endif
+	vorr		q1,q0,q0
+	add		r8, r8, #1
+	vorr		q6,q0,q0
+	rev		r10, r8
+	cmp		r5,#2
+	vmov.32	d3[1],r10
+	beq		.Lctr32_128
+
+.Loop2x_ctr32:
+	.byte	0x20,0x03,0xb0,0xf3	@ aese q0,q8
+	.byte	0x20,0x23,0xb0,0xf3	@ aese q1,q8
+	vld1.32		{q8},[r7]!
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x82,0x23,0xb0,0xf3	@ aesmc q1,q1
+	subs		r6,r6,#2
+	.byte	0x22,0x03,0xb0,0xf3	@ aese q0,q9
+	.byte	0x22,0x23,0xb0,0xf3	@ aese q1,q9
+	vld1.32		{q9},[r7]!
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x82,0x23,0xb0,0xf3	@ aesmc q1,q1
+	bgt		.Loop2x_ctr32
+
+	.byte	0x20,0x03,0xb0,0xf3	@ aese q0,q8
+	.byte	0x20,0x23,0xb0,0xf3	@ aese q1,q8
+	.byte	0x80,0x83,0xb0,0xf3	@ aesmc q4,q0
+	 vorr		q0,q6,q6
+	.byte	0x82,0xa3,0xb0,0xf3	@ aesmc q5,q1
+	 vorr		q1,q6,q6
+	.byte	0x22,0x83,0xb0,0xf3	@ aese q4,q9
+	.byte	0x22,0xa3,0xb0,0xf3	@ aese q5,q9
+	 vld1.8		{q2},[r0]!
+	.byte	0x88,0x83,0xb0,0xf3	@ aesmc q4,q4
+	 vld1.8		{q3},[r0]!
+	.byte	0x8a,0xa3,0xb0,0xf3	@ aesmc q5,q5
+	 add		r8,r8,#1
+	.byte	0x24,0x83,0xb0,0xf3	@ aese q4,q10
+	.byte	0x24,0xa3,0xb0,0xf3	@ aese q5,q10
+	 rev		r9,r8
+	.byte	0x88,0x83,0xb0,0xf3	@ aesmc q4,q4
+	.byte	0x8a,0xa3,0xb0,0xf3	@ aesmc q5,q5
+	 add		r8,r8,#1
+	.byte	0x26,0x83,0xb0,0xf3	@ aese q4,q11
+	.byte	0x26,0xa3,0xb0,0xf3	@ aese q5,q11
+	 veor		q2,q2,q7
+	 rev		r10,r8
+	.byte	0x88,0x83,0xb0,0xf3	@ aesmc q4,q4
+	.byte	0x8a,0xa3,0xb0,0xf3	@ aesmc q5,q5
+	 veor		q3,q3,q7
+	 mov		r7,r3
+	.byte	0x28,0x83,0xb0,0xf3	@ aese q4,q12
+	.byte	0x28,0xa3,0xb0,0xf3	@ aese q5,q12
+	 subs		r2,r2,#2
+	.byte	0x88,0x83,0xb0,0xf3	@ aesmc q4,q4
+	.byte	0x8a,0xa3,0xb0,0xf3	@ aesmc q5,q5
+	 vld1.32	 {q8-q9},[r7]!	@ re-pre-load rndkey[0-1]
+	.byte	0x2a,0x83,0xb0,0xf3	@ aese q4,q13
+	.byte	0x2a,0xa3,0xb0,0xf3	@ aese q5,q13
+	.byte	0x88,0x83,0xb0,0xf3	@ aesmc q4,q4
+	.byte	0x8a,0xa3,0xb0,0xf3	@ aesmc q5,q5
+	.byte	0x2c,0x83,0xb0,0xf3	@ aese q4,q14
+	.byte	0x2c,0xa3,0xb0,0xf3	@ aese q5,q14
+	 vmov.32	d1[1], r9
+	.byte	0x88,0x83,0xb0,0xf3	@ aesmc q4,q4
+	 vmov.32	d3[1], r10
+	.byte	0x8a,0xa3,0xb0,0xf3	@ aesmc q5,q5
+	.byte	0x2e,0x83,0xb0,0xf3	@ aese q4,q15
+	.byte	0x2e,0xa3,0xb0,0xf3	@ aese q5,q15
+
+	 mov		r6,r5
+	veor		q2,q2,q4
+	veor		q3,q3,q5
+	vst1.8		{q2},[r1]!
+	vst1.8		{q3},[r1]!
+	bhs		.Loop2x_ctr32
+
+	adds		r2,r2,#2
+	beq		.Lctr32_done
+	b		.Lctr32_tail
+
+.Lctr32_128:
+	vld1.32		{q4-q5},[r7]
+
+.Loop2x_ctr32_128:
+	.byte	0x20,0x03,0xb0,0xf3	@ aese q0,q8
+	.byte	0x20,0x23,0xb0,0xf3	@ aese q1,q8
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 vld1.8		{q2},[r0]!
+	.byte	0x82,0x23,0xb0,0xf3	@ aesmc q1,q1
+	 vld1.8		{q3},[r0]!
+	.byte	0x22,0x03,0xb0,0xf3	@ aese q0,q9
+	.byte	0x22,0x23,0xb0,0xf3	@ aese q1,q9
+	 add		r8,r8,#1
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x82,0x23,0xb0,0xf3	@ aesmc q1,q1
+	 rev		r9,r8
+	.byte	0x08,0x03,0xb0,0xf3	@ aese q0,q4
+	.byte	0x08,0x23,0xb0,0xf3	@ aese q1,q4
+	 add		r8,r8,#1
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x82,0x23,0xb0,0xf3	@ aesmc q1,q1
+	 rev		r10,r8
+	.byte	0x0a,0x03,0xb0,0xf3	@ aese q0,q5
+	.byte	0x0a,0x23,0xb0,0xf3	@ aese q1,q5
+	 subs		r2,r2,#2
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x82,0x23,0xb0,0xf3	@ aesmc q1,q1
+	.byte	0x24,0x03,0xb0,0xf3	@ aese q0,q10
+	.byte	0x24,0x23,0xb0,0xf3	@ aese q1,q10
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x82,0x23,0xb0,0xf3	@ aesmc q1,q1
+	.byte	0x26,0x03,0xb0,0xf3	@ aese q0,q11
+	.byte	0x26,0x23,0xb0,0xf3	@ aese q1,q11
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x82,0x23,0xb0,0xf3	@ aesmc q1,q1
+	.byte	0x28,0x03,0xb0,0xf3	@ aese q0,q12
+	.byte	0x28,0x23,0xb0,0xf3	@ aese q1,q12
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x82,0x23,0xb0,0xf3	@ aesmc q1,q1
+	.byte	0x2a,0x03,0xb0,0xf3	@ aese q0,q13
+	.byte	0x2a,0x23,0xb0,0xf3	@ aese q1,q13
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x82,0x23,0xb0,0xf3	@ aesmc q1,q1
+	.byte	0x2c,0x03,0xb0,0xf3	@ aese q0,q14
+	.byte	0x2c,0x23,0xb0,0xf3	@ aese q1,q14
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x82,0x23,0xb0,0xf3	@ aesmc q1,q1
+	 veor		q2,q2,q7
+	.byte	0x2e,0x03,0xb0,0xf3	@ aese q0,q15
+	 veor		q3,q3,q7
+	.byte	0x2e,0x23,0xb0,0xf3	@ aese q1,q15
+
+	veor		q2,q2,q0
+	vorr		q0,q6,q6
+	veor		q3,q3,q1
+	vorr		q1,q6,q6
+	vst1.8		{q2},[r1]!
+	vmov.32	d1[1], r9
+	vst1.8		{q3},[r1]!
+	vmov.32	d3[1], r10
+	bhs		.Loop2x_ctr32_128
+
+	adds		r2,r2,#2
+	beq		.Lctr32_done
+
+.Lctr32_tail:
+	.byte	0x20,0x03,0xb0,0xf3	@ aese q0,q8
+	vld1.32		{q8},[r7]!
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	subs		r6,r6,#2
+	.byte	0x22,0x03,0xb0,0xf3	@ aese q0,q9
+	vld1.32		{q9},[r7]!
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	bgt		.Lctr32_tail
+
+	.byte	0x20,0x03,0xb0,0xf3	@ aese q0,q8
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x22,0x03,0xb0,0xf3	@ aese q0,q9
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 vld1.8		{q2},[r0]
+	.byte	0x24,0x03,0xb0,0xf3	@ aese q0,q10
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x26,0x03,0xb0,0xf3	@ aese q0,q11
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x28,0x03,0xb0,0xf3	@ aese q0,q12
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x2a,0x03,0xb0,0xf3	@ aese q0,q13
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	.byte	0x2c,0x03,0xb0,0xf3	@ aese q0,q14
+	.byte	0x80,0x03,0xb0,0xf3	@ aesmc q0,q0
+	 veor		q2,q2,q7
+	.byte	0x2e,0x03,0xb0,0xf3	@ aese q0,q15
+
+	veor		q2,q2,q0
+	vst1.8		{q2},[r1]
+
+.Lctr32_done:
+	vldmia		sp!,{d8-d15}
+	ldmia		sp!,{r4-r10,pc}
+.size	aes_v8_ctr32_encrypt_blocks,.-aes_v8_ctr32_encrypt_blocks
+#endif
diff --git a/main/openssl/crypto/aes/asm/aesv8-armx.pl b/main/openssl/crypto/aes/asm/aesv8-armx.pl
new file mode 100644
index 00000000..415dc04a
--- /dev/null
+++ b/main/openssl/crypto/aes/asm/aesv8-armx.pl
@@ -0,0 +1,980 @@
+#!/usr/bin/env perl
+#
+# ====================================================================
+# Written by Andy Polyakov <appro@openssl.org> for the OpenSSL
+# project. The module is, however, dual licensed under OpenSSL and
+# CRYPTOGAMS licenses depending on where you obtain it. For further
+# details see http://www.openssl.org/~appro/cryptogams/.
+# ====================================================================
+#
+# This module implements support for ARMv8 AES instructions. The
+# module is endian-agnostic in sense that it supports both big- and
+# little-endian cases. As does it support both 32- and 64-bit modes
+# of operation. Latter is achieved by limiting amount of utilized
+# registers to 16, which implies additional instructions. This has
+# no effect on mighty Apple A7, as results are literally equal to
+# the theoretical estimates based on instruction latencies and issue
+# rate. It remains to be seen how does it affect other platforms...
+#
+# Performance in cycles per byte processed with 128-bit key:
+#
+#		CBC enc		CBC dec		CTR
+# Apple A7	2.39		1.20		1.20
+# Cortex-A5x	n/a		n/a		n/a
+
+$flavour = shift;
+open STDOUT,">".shift;
+
+$prefix="aes_v8";
+
+$code=<<___;
+#include "arm_arch.h"
+
+#if __ARM_ARCH__>=7
+.text
+___
+$code.=".arch	armv8-a+crypto\n"	if ($flavour =~ /64/);
+$code.=".fpu	neon\n.code	32\n"	if ($flavour !~ /64/);
+
+# Assembler mnemonics are an eclectic mix of 32- and 64-bit syntax,
+# NEON is mostly 32-bit mnemonics, integer - mostly 64. Goal is to
+# maintain both 32- and 64-bit codes within single module and
+# transliterate common code to either flavour with regex vodoo.
+#
+{{{
+my ($inp,$bits,$out,$ptr,$rounds)=("x0","w1","x2","x3","w12");
+my ($zero,$rcon,$mask,$in0,$in1,$tmp,$key)=
+	$flavour=~/64/? map("q$_",(0..6)) : map("q$_",(0..3,8..10));
+
+
+$code.=<<___;
+.align	5
+rcon:
+.long	0x01,0x01,0x01,0x01
+.long	0x0c0f0e0d,0x0c0f0e0d,0x0c0f0e0d,0x0c0f0e0d	// rotate-n-splat
+.long	0x1b,0x1b,0x1b,0x1b
+
+.globl	${prefix}_set_encrypt_key
+.type	${prefix}_set_encrypt_key,%function
+.align	5
+${prefix}_set_encrypt_key:
+.Lenc_key:
+___
+$code.=<<___	if ($flavour =~ /64/);
+	stp	x29,x30,[sp,#-16]!
+	add	x29,sp,#0
+___
+$code.=<<___;
+	adr	$ptr,rcon
+	cmp	$bits,#192
+
+	veor	$zero,$zero,$zero
+	vld1.8	{$in0},[$inp],#16
+	mov	$bits,#8		// reuse $bits
+	vld1.32	{$rcon,$mask},[$ptr],#32
+
+	b.lt	.Loop128
+	b.eq	.L192
+	b	.L256
+
+.align	4
+.Loop128:
+	vtbl.8	$key,{$in0},$mask
+	vext.8	$tmp,$zero,$in0,#12
+	vst1.32	{$in0},[$out],#16
+	aese	$key,$zero
+	subs	$bits,$bits,#1
+
+	veor	$in0,$in0,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	veor	$in0,$in0,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	 veor	$key,$key,$rcon
+	veor	$in0,$in0,$tmp
+	vshl.u8	$rcon,$rcon,#1
+	veor	$in0,$in0,$key
+	b.ne	.Loop128
+
+	vld1.32	{$rcon},[$ptr]
+
+	vtbl.8	$key,{$in0},$mask
+	vext.8	$tmp,$zero,$in0,#12
+	vst1.32	{$in0},[$out],#16
+	aese	$key,$zero
+
+	veor	$in0,$in0,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	veor	$in0,$in0,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	 veor	$key,$key,$rcon
+	veor	$in0,$in0,$tmp
+	vshl.u8	$rcon,$rcon,#1
+	veor	$in0,$in0,$key
+
+	vtbl.8	$key,{$in0},$mask
+	vext.8	$tmp,$zero,$in0,#12
+	vst1.32	{$in0},[$out],#16
+	aese	$key,$zero
+
+	veor	$in0,$in0,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	veor	$in0,$in0,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	 veor	$key,$key,$rcon
+	veor	$in0,$in0,$tmp
+	veor	$in0,$in0,$key
+	vst1.32	{$in0},[$out]
+	add	$out,$out,#0x50
+
+	mov	$rounds,#10
+	b	.Ldone
+
+.align	4
+.L192:
+	vld1.8	{$in1},[$inp],#8
+	vmov.i8	$key,#8			// borrow $key
+	vst1.32	{$in0},[$out],#16
+	vsub.i8	$mask,$mask,$key	// adjust the mask
+
+.Loop192:
+	vtbl.8	$key,{$in1},$mask
+	vext.8	$tmp,$zero,$in0,#12
+	vst1.32	{$in1},[$out],#8
+	aese	$key,$zero
+	subs	$bits,$bits,#1
+
+	veor	$in0,$in0,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	veor	$in0,$in0,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	veor	$in0,$in0,$tmp
+
+	vdup.32	$tmp,${in0}[3]
+	veor	$tmp,$tmp,$in1
+	 veor	$key,$key,$rcon
+	vext.8	$in1,$zero,$in1,#12
+	vshl.u8	$rcon,$rcon,#1
+	veor	$in1,$in1,$tmp
+	veor	$in0,$in0,$key
+	veor	$in1,$in1,$key
+	vst1.32	{$in0},[$out],#16
+	b.ne	.Loop192
+
+	mov	$rounds,#12
+	add	$out,$out,#0x20
+	b	.Ldone
+
+.align	4
+.L256:
+	vld1.8	{$in1},[$inp]
+	mov	$bits,#7
+	mov	$rounds,#14
+	vst1.32	{$in0},[$out],#16
+
+.Loop256:
+	vtbl.8	$key,{$in1},$mask
+	vext.8	$tmp,$zero,$in0,#12
+	vst1.32	{$in1},[$out],#16
+	aese	$key,$zero
+	subs	$bits,$bits,#1
+
+	veor	$in0,$in0,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	veor	$in0,$in0,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	 veor	$key,$key,$rcon
+	veor	$in0,$in0,$tmp
+	vshl.u8	$rcon,$rcon,#1
+	veor	$in0,$in0,$key
+	vst1.32	{$in0},[$out],#16
+	b.eq	.Ldone
+
+	vdup.32	$key,${in0}[3]		// just splat
+	vext.8	$tmp,$zero,$in1,#12
+	aese	$key,$zero
+
+	veor	$in1,$in1,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	veor	$in1,$in1,$tmp
+	vext.8	$tmp,$zero,$tmp,#12
+	veor	$in1,$in1,$tmp
+
+	veor	$in1,$in1,$key
+	b	.Loop256
+
+.Ldone:
+	str	$rounds,[$out]
+
+	eor	x0,x0,x0		// return value
+	`"ldr	x29,[sp],#16"		if ($flavour =~ /64/)`
+	ret
+.size	${prefix}_set_encrypt_key,.-${prefix}_set_encrypt_key
+
+.globl	${prefix}_set_decrypt_key
+.type	${prefix}_set_decrypt_key,%function
+.align	5
+${prefix}_set_decrypt_key:
+___
+$code.=<<___	if ($flavour =~ /64/);
+	stp	x29,x30,[sp,#-16]!
+	add	x29,sp,#0
+___
+$code.=<<___	if ($flavour !~ /64/);
+	stmdb	sp!,{r4,lr}
+___
+$code.=<<___;
+	bl	.Lenc_key
+
+	sub	$out,$out,#240		// restore original $out
+	mov	x4,#-16
+	add	$inp,$out,x12,lsl#4	// end of key schedule
+
+	vld1.32	{v0.16b},[$out]
+	vld1.32	{v1.16b},[$inp]
+	vst1.32	{v0.16b},[$inp],x4
+	vst1.32	{v1.16b},[$out],#16
+
+.Loop_imc:
+	vld1.32	{v0.16b},[$out]
+	vld1.32	{v1.16b},[$inp]
+	aesimc	v0.16b,v0.16b
+	aesimc	v1.16b,v1.16b
+	vst1.32	{v0.16b},[$inp],x4
+	vst1.32	{v1.16b},[$out],#16
+	cmp	$inp,$out
+	b.hi	.Loop_imc
+
+	vld1.32	{v0.16b},[$out]
+	aesimc	v0.16b,v0.16b
+	vst1.32	{v0.16b},[$inp]
+
+	eor	x0,x0,x0		// return value
+___
+$code.=<<___	if ($flavour !~ /64/);
+	ldmia	sp!,{r4,pc}
+___
+$code.=<<___	if ($flavour =~ /64/);
+	ldp	x29,x30,[sp],#16
+	ret
+___
+$code.=<<___;
+.size	${prefix}_set_decrypt_key,.-${prefix}_set_decrypt_key
+___
+}}}
+{{{
+sub gen_block () {
+my $dir = shift;
+my ($e,$mc) = $dir eq "en" ? ("e","mc") : ("d","imc");
+my ($inp,$out,$key)=map("x$_",(0..2));
+my $rounds="w3";
+my ($rndkey0,$rndkey1,$inout)=map("q$_",(0..3));
+
+$code.=<<___;
+.globl	${prefix}_${dir}crypt
+.type	${prefix}_${dir}crypt,%function
+.align	5
+${prefix}_${dir}crypt:
+	ldr	$rounds,[$key,#240]
+	vld1.32	{$rndkey0},[$key],#16
+	vld1.8	{$inout},[$inp]
+	sub	$rounds,$rounds,#2
+	vld1.32	{$rndkey1},[$key],#16
+
+.Loop_${dir}c:
+	aes$e	$inout,$rndkey0
+	vld1.32	{$rndkey0},[$key],#16
+	aes$mc	$inout,$inout
+	subs	$rounds,$rounds,#2
+	aes$e	$inout,$rndkey1
+	vld1.32	{$rndkey1},[$key],#16
+	aes$mc	$inout,$inout
+	b.gt	.Loop_${dir}c
+
+	aes$e	$inout,$rndkey0
+	vld1.32	{$rndkey0},[$key]
+	aes$mc	$inout,$inout
+	aes$e	$inout,$rndkey1
+	veor	$inout,$inout,$rndkey0
+
+	vst1.8	{$inout},[$out]
+	ret
+.size	${prefix}_${dir}crypt,.-${prefix}_${dir}crypt
+___
+}
+&gen_block("en");
+&gen_block("de");
+}}}
+{{{
+my ($inp,$out,$len,$key,$ivp)=map("x$_",(0..4)); my $enc="w5";
+my ($rounds,$cnt,$key_,$step,$step1)=($enc,"w6","x7","x8","x12");
+my ($dat0,$dat1,$in0,$in1,$tmp0,$tmp1,$ivec,$rndlast)=map("q$_",(0..7));
+
+my ($dat,$tmp,$rndzero_n_last)=($dat0,$tmp0,$tmp1);
+
+### q8-q15	preloaded key schedule
+
+$code.=<<___;
+.globl	${prefix}_cbc_encrypt
+.type	${prefix}_cbc_encrypt,%function
+.align	5
+${prefix}_cbc_encrypt:
+___
+$code.=<<___	if ($flavour =~ /64/);
+	stp	x29,x30,[sp,#-16]!
+	add	x29,sp,#0
+___
+$code.=<<___	if ($flavour !~ /64/);
+	mov	ip,sp
+	stmdb	sp!,{r4-r8,lr}
+	vstmdb	sp!,{d8-d15}            @ ABI specification says so
+	ldmia	ip,{r4-r5}		@ load remaining args
+___
+$code.=<<___;
+	subs	$len,$len,#16
+	mov	$step,#16
+	b.lo	.Lcbc_abort
+	cclr	$step,eq
+
+	cmp	$enc,#0			// en- or decrypting?
+	ldr	$rounds,[$key,#240]
+	and	$len,$len,#-16
+	vld1.8	{$ivec},[$ivp]
+	vld1.8	{$dat},[$inp],$step
+
+	vld1.32	{q8-q9},[$key]		// load key schedule...
+	sub	$rounds,$rounds,#6
+	add	$key_,$key,x5,lsl#4	// pointer to last 7 round keys
+	sub	$rounds,$rounds,#2
+	vld1.32	{q10-q11},[$key_],#32
+	vld1.32	{q12-q13},[$key_],#32
+	vld1.32	{q14-q15},[$key_],#32
+	vld1.32	{$rndlast},[$key_]
+
+	add	$key_,$key,#32
+	mov	$cnt,$rounds
+	b.eq	.Lcbc_dec
+
+	cmp	$rounds,#2
+	veor	$dat,$dat,$ivec
+	veor	$rndzero_n_last,q8,$rndlast
+	b.eq	.Lcbc_enc128
+
+.Loop_cbc_enc:
+	aese	$dat,q8
+	vld1.32	{q8},[$key_],#16
+	aesmc	$dat,$dat
+	subs	$cnt,$cnt,#2
+	aese	$dat,q9
+	vld1.32	{q9},[$key_],#16
+	aesmc	$dat,$dat
+	b.gt	.Loop_cbc_enc
+
+	aese	$dat,q8
+	aesmc	$dat,$dat
+	 subs	$len,$len,#16
+	aese	$dat,q9
+	aesmc	$dat,$dat
+	 cclr	$step,eq
+	aese	$dat,q10
+	aesmc	$dat,$dat
+	 add	$key_,$key,#16
+	aese	$dat,q11
+	aesmc	$dat,$dat
+	 vld1.8	{q8},[$inp],$step
+	aese	$dat,q12
+	aesmc	$dat,$dat
+	 veor	q8,q8,$rndzero_n_last
+	aese	$dat,q13
+	aesmc	$dat,$dat
+	 vld1.32 {q9},[$key_],#16	// re-pre-load rndkey[1]
+	aese	$dat,q14
+	aesmc	$dat,$dat
+	aese	$dat,q15
+
+	 mov	$cnt,$rounds
+	veor	$ivec,$dat,$rndlast
+	vst1.8	{$ivec},[$out],#16
+	b.hs	.Loop_cbc_enc
+
+	b	.Lcbc_done
+
+.align	5
+.Lcbc_enc128:
+	vld1.32	{$in0-$in1},[$key_]
+	aese	$dat,q8
+	aesmc	$dat,$dat
+	b	.Lenter_cbc_enc128
+.Loop_cbc_enc128:
+	aese	$dat,q8
+	aesmc	$dat,$dat
+	 vst1.8	{$ivec},[$out],#16
+.Lenter_cbc_enc128:
+	aese	$dat,q9
+	aesmc	$dat,$dat
+	 subs	$len,$len,#16
+	aese	$dat,$in0
+	aesmc	$dat,$dat
+	 cclr	$step,eq
+	aese	$dat,$in1
+	aesmc	$dat,$dat
+	aese	$dat,q10
+	aesmc	$dat,$dat
+	aese	$dat,q11
+	aesmc	$dat,$dat
+	 vld1.8	{q8},[$inp],$step
+	aese	$dat,q12
+	aesmc	$dat,$dat
+	aese	$dat,q13
+	aesmc	$dat,$dat
+	aese	$dat,q14
+	aesmc	$dat,$dat
+	 veor	q8,q8,$rndzero_n_last
+	aese	$dat,q15
+	veor	$ivec,$dat,$rndlast
+	b.hs	.Loop_cbc_enc128
+
+	vst1.8	{$ivec},[$out],#16
+	b	.Lcbc_done
+
+.align	5
+.Lcbc_dec128:
+	vld1.32	{$tmp0-$tmp1},[$key_]
+	veor	$ivec,$ivec,$rndlast
+	veor	$in0,$dat0,$rndlast
+	mov	$step1,$step
+
+.Loop2x_cbc_dec128:
+	aesd	$dat0,q8
+	aesd	$dat1,q8
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	 subs	$len,$len,#32
+	aesd	$dat0,q9
+	aesd	$dat1,q9
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	 cclr	$step,lo
+	aesd	$dat0,$tmp0
+	aesd	$dat1,$tmp0
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	 cclr	$step1,ls
+	aesd	$dat0,$tmp1
+	aesd	$dat1,$tmp1
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	aesd	$dat0,q10
+	aesd	$dat1,q10
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	aesd	$dat0,q11
+	aesd	$dat1,q11
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	aesd	$dat0,q12
+	aesd	$dat1,q12
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	aesd	$dat0,q13
+	aesd	$dat1,q13
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	aesd	$dat0,q14
+	aesd	$dat1,q14
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	aesd	$dat0,q15
+	aesd	$dat1,q15
+
+	veor	$ivec,$ivec,$dat0
+	vld1.8	{$dat0},[$inp],$step
+	veor	$in0,$in0,$dat1
+	vld1.8	{$dat1},[$inp],$step1
+	vst1.8	{$ivec},[$out],#16
+	veor	$ivec,$in1,$rndlast
+	vst1.8	{$in0},[$out],#16
+	veor	$in0,$dat0,$rndlast
+	vorr	$in1,$dat1,$dat1
+	b.hs	.Loop2x_cbc_dec128
+
+	adds	$len,$len,#32
+	veor	$ivec,$ivec,$rndlast
+	b.eq	.Lcbc_done
+	veor	$in0,$in0,$rndlast
+	b	.Lcbc_dec_tail
+
+.align	5
+.Lcbc_dec:
+	subs	$len,$len,#16
+	vorr	$in0,$dat,$dat
+	b.lo	.Lcbc_dec_tail
+
+	cclr	$step,eq
+	cmp	$rounds,#2
+	vld1.8	{$dat1},[$inp],$step
+	vorr	$in1,$dat1,$dat1
+	b.eq	.Lcbc_dec128
+
+.Loop2x_cbc_dec:
+	aesd	$dat0,q8
+	aesd	$dat1,q8
+	vld1.32	{q8},[$key_],#16
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	subs	$cnt,$cnt,#2
+	aesd	$dat0,q9
+	aesd	$dat1,q9
+	vld1.32	{q9},[$key_],#16
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	b.gt	.Loop2x_cbc_dec
+
+	aesd	$dat0,q8
+	aesd	$dat1,q8
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	 veor	$tmp0,$ivec,$rndlast
+	 veor	$tmp1,$in0,$rndlast
+	aesd	$dat0,q9
+	aesd	$dat1,q9
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	 vorr	$ivec,$in1,$in1
+	 subs	$len,$len,#32
+	aesd	$dat0,q10
+	aesd	$dat1,q10
+	aesimc	$dat0,$dat0
+	 cclr	$step,lo
+	aesimc	$dat1,$dat1
+	 mov	$key_,$key
+	aesd	$dat0,q11
+	aesd	$dat1,q11
+	aesimc	$dat0,$dat0
+	 vld1.8	{$in0},[$inp],$step
+	aesimc	$dat1,$dat1
+	 cclr	$step,ls
+	aesd	$dat0,q12
+	aesd	$dat1,q12
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	 vld1.8	{$in1},[$inp],$step
+	aesd	$dat0,q13
+	aesd	$dat1,q13
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	 vld1.32 {q8},[$key_],#16	// re-pre-load rndkey[0]
+	aesd	$dat0,q14
+	aesd	$dat1,q14
+	aesimc	$dat0,$dat0
+	aesimc	$dat1,$dat1
+	 vld1.32 {q9},[$key_],#16	// re-pre-load rndkey[1]
+	aesd	$dat0,q15
+	aesd	$dat1,q15
+
+	 mov	$cnt,$rounds
+	veor	$tmp0,$tmp0,$dat0
+	veor	$tmp1,$tmp1,$dat1
+	 vorr	$dat0,$in0,$in0
+	vst1.8	{$tmp0},[$out],#16
+	 vorr	$dat1,$in1,$in1
+	vst1.8	{$tmp1},[$out],#16
+	b.hs	.Loop2x_cbc_dec
+
+	adds	$len,$len,#32
+	b.eq	.Lcbc_done
+
+.Lcbc_dec_tail:
+	aesd	$dat,q8
+	vld1.32	{q8},[$key_],#16
+	aesimc	$dat,$dat
+	subs	$cnt,$cnt,#2
+	aesd	$dat,q9
+	vld1.32	{q9},[$key_],#16
+	aesimc	$dat,$dat
+	b.gt	.Lcbc_dec_tail
+
+	aesd	$dat,q8
+	aesimc	$dat,$dat
+	aesd	$dat,q9
+	aesimc	$dat,$dat
+	 veor	$tmp,$ivec,$rndlast
+	aesd	$dat,q10
+	aesimc	$dat,$dat
+	 vorr	$ivec,$in0,$in0
+	aesd	$dat,q11
+	aesimc	$dat,$dat
+	aesd	$dat,q12
+	aesimc	$dat,$dat
+	aesd	$dat,q13
+	aesimc	$dat,$dat
+	aesd	$dat,q14
+	aesimc	$dat,$dat
+	aesd	$dat,q15
+
+	veor	$tmp,$tmp,$dat
+	vst1.8	{$tmp},[$out],#16
+
+.Lcbc_done:
+	vst1.8	{$ivec},[$ivp]
+.Lcbc_abort:
+___
+$code.=<<___	if ($flavour !~ /64/);
+	vldmia	sp!,{d8-d15}
+	ldmia	sp!,{r4-r8,pc}
+___
+$code.=<<___	if ($flavour =~ /64/);
+	ldr	x29,[sp],#16
+	ret
+___
+$code.=<<___;
+.size	${prefix}_cbc_encrypt,.-${prefix}_cbc_encrypt
+___
+}}}
+{{{
+my ($inp,$out,$len,$key,$ivp)=map("x$_",(0..4));
+my ($rounds,$cnt,$key_,$ctr,$tctr,$tctr1)=("w5","w6","x7","w8","w9","w10");
+my ($dat0,$dat1,$in0,$in1,$tmp0,$tmp1,$ivec,$rndlast)=map("q$_",(0..7));
+
+my ($dat,$tmp)=($dat0,$tmp0);
+
+### q8-q15	preloaded key schedule
+
+$code.=<<___;
+.globl	${prefix}_ctr32_encrypt_blocks
+.type	${prefix}_ctr32_encrypt_blocks,%function
+.align	5
+${prefix}_ctr32_encrypt_blocks:
+___
+$code.=<<___	if ($flavour =~ /64/);
+	stp		x29,x30,[sp,#-16]!
+	add		x29,sp,#0
+___
+$code.=<<___	if ($flavour !~ /64/);
+	mov		ip,sp
+	stmdb		sp!,{r4-r10,lr}
+	vstmdb		sp!,{d8-d15}            @ ABI specification says so
+	ldr		r4, [ip]		@ load remaining arg
+___
+$code.=<<___;
+	ldr		$rounds,[$key,#240]
+
+	ldr		$ctr, [$ivp, #12]
+	vld1.32		{$dat0},[$ivp]
+
+	vld1.32		{q8-q9},[$key]		// load key schedule...
+	sub		$rounds,$rounds,#6
+	add		$key_,$key,x5,lsl#4	// pointer to last 7 round keys
+	sub		$rounds,$rounds,#2
+	vld1.32		{q10-q11},[$key_],#32
+	vld1.32		{q12-q13},[$key_],#32
+	vld1.32		{q14-q15},[$key_],#32
+	vld1.32		{$rndlast},[$key_]
+
+	add		$key_,$key,#32
+	mov		$cnt,$rounds
+
+	subs		$len,$len,#2
+	b.lo		.Lctr32_tail
+
+#ifndef __ARMEB__
+	rev		$ctr, $ctr
+#endif
+	vorr		$dat1,$dat0,$dat0
+	add		$ctr, $ctr, #1
+	vorr		$ivec,$dat0,$dat0
+	rev		$tctr1, $ctr
+	cmp		$rounds,#2
+	vmov.32		${dat1}[3],$tctr1
+	b.eq		.Lctr32_128
+
+.Loop2x_ctr32:
+	aese		$dat0,q8
+	aese		$dat1,q8
+	vld1.32		{q8},[$key_],#16
+	aesmc		$dat0,$dat0
+	aesmc		$dat1,$dat1
+	subs		$cnt,$cnt,#2
+	aese		$dat0,q9
+	aese		$dat1,q9
+	vld1.32		{q9},[$key_],#16
+	aesmc		$dat0,$dat0
+	aesmc		$dat1,$dat1
+	b.gt		.Loop2x_ctr32
+
+	aese		$dat0,q8
+	aese		$dat1,q8
+	aesmc		$tmp0,$dat0
+	 vorr		$dat0,$ivec,$ivec
+	aesmc		$tmp1,$dat1
+	 vorr		$dat1,$ivec,$ivec
+	aese		$tmp0,q9
+	aese		$tmp1,q9
+	 vld1.8		{$in0},[$inp],#16
+	aesmc		$tmp0,$tmp0
+	 vld1.8		{$in1},[$inp],#16
+	aesmc		$tmp1,$tmp1
+	 add		$ctr,$ctr,#1
+	aese		$tmp0,q10
+	aese		$tmp1,q10
+	 rev		$tctr,$ctr
+	aesmc		$tmp0,$tmp0
+	aesmc		$tmp1,$tmp1
+	 add		$ctr,$ctr,#1
+	aese		$tmp0,q11
+	aese		$tmp1,q11
+	 veor		$in0,$in0,$rndlast
+	 rev		$tctr1,$ctr
+	aesmc		$tmp0,$tmp0
+	aesmc		$tmp1,$tmp1
+	 veor		$in1,$in1,$rndlast
+	 mov		$key_,$key
+	aese		$tmp0,q12
+	aese		$tmp1,q12
+	 subs		$len,$len,#2
+	aesmc		$tmp0,$tmp0
+	aesmc		$tmp1,$tmp1
+	 vld1.32	 {q8-q9},[$key_],#32	// re-pre-load rndkey[0-1]
+	aese		$tmp0,q13
+	aese		$tmp1,q13
+	aesmc		$tmp0,$tmp0
+	aesmc		$tmp1,$tmp1
+	aese		$tmp0,q14
+	aese		$tmp1,q14
+	 vmov.32	${dat0}[3], $tctr
+	aesmc		$tmp0,$tmp0
+	 vmov.32	${dat1}[3], $tctr1
+	aesmc		$tmp1,$tmp1
+	aese		$tmp0,q15
+	aese		$tmp1,q15
+
+	 mov		$cnt,$rounds
+	veor		$in0,$in0,$tmp0
+	veor		$in1,$in1,$tmp1
+	vst1.8		{$in0},[$out],#16
+	vst1.8		{$in1},[$out],#16
+	b.hs		.Loop2x_ctr32
+
+	adds		$len,$len,#2
+	b.eq		.Lctr32_done
+	b		.Lctr32_tail
+
+.Lctr32_128:
+	vld1.32		{$tmp0-$tmp1},[$key_]
+
+.Loop2x_ctr32_128:
+	aese		$dat0,q8
+	aese		$dat1,q8
+	aesmc		$dat0,$dat0
+	 vld1.8		{$in0},[$inp],#16
+	aesmc		$dat1,$dat1
+	 vld1.8		{$in1},[$inp],#16
+	aese		$dat0,q9
+	aese		$dat1,q9
+	 add		$ctr,$ctr,#1
+	aesmc		$dat0,$dat0
+	aesmc		$dat1,$dat1
+	 rev		$tctr,$ctr
+	aese		$dat0,$tmp0
+	aese		$dat1,$tmp0
+	 add		$ctr,$ctr,#1
+	aesmc		$dat0,$dat0
+	aesmc		$dat1,$dat1
+	 rev		$tctr1,$ctr
+	aese		$dat0,$tmp1
+	aese		$dat1,$tmp1
+	 subs		$len,$len,#2
+	aesmc		$dat0,$dat0
+	aesmc		$dat1,$dat1
+	aese		$dat0,q10
+	aese		$dat1,q10
+	aesmc		$dat0,$dat0
+	aesmc		$dat1,$dat1
+	aese		$dat0,q11
+	aese		$dat1,q11
+	aesmc		$dat0,$dat0
+	aesmc		$dat1,$dat1
+	aese		$dat0,q12
+	aese		$dat1,q12
+	aesmc		$dat0,$dat0
+	aesmc		$dat1,$dat1
+	aese		$dat0,q13
+	aese		$dat1,q13
+	aesmc		$dat0,$dat0
+	aesmc		$dat1,$dat1
+	aese		$dat0,q14
+	aese		$dat1,q14
+	aesmc		$dat0,$dat0
+	aesmc		$dat1,$dat1
+	 veor		$in0,$in0,$rndlast
+	aese		$dat0,q15
+	 veor		$in1,$in1,$rndlast
+	aese		$dat1,q15
+
+	veor		$in0,$in0,$dat0
+	vorr		$dat0,$ivec,$ivec
+	veor		$in1,$in1,$dat1
+	vorr		$dat1,$ivec,$ivec
+	vst1.8		{$in0},[$out],#16
+	vmov.32		${dat0}[3], $tctr
+	vst1.8		{$in1},[$out],#16
+	vmov.32		${dat1}[3], $tctr1
+	b.hs		.Loop2x_ctr32_128
+
+	adds		$len,$len,#2
+	b.eq		.Lctr32_done
+
+.Lctr32_tail:
+	aese		$dat,q8
+	vld1.32		{q8},[$key_],#16
+	aesmc		$dat,$dat
+	subs		$cnt,$cnt,#2
+	aese		$dat,q9
+	vld1.32		{q9},[$key_],#16
+	aesmc		$dat,$dat
+	b.gt		.Lctr32_tail
+
+	aese		$dat,q8
+	aesmc		$dat,$dat
+	aese		$dat,q9
+	aesmc		$dat,$dat
+	 vld1.8		{$in0},[$inp]
+	aese		$dat,q10
+	aesmc		$dat,$dat
+	aese		$dat,q11
+	aesmc		$dat,$dat
+	aese		$dat,q12
+	aesmc		$dat,$dat
+	aese		$dat,q13
+	aesmc		$dat,$dat
+	aese		$dat,q14
+	aesmc		$dat,$dat
+	 veor		$in0,$in0,$rndlast
+	aese		$dat,q15
+
+	veor		$in0,$in0,$dat
+	vst1.8		{$in0},[$out]
+
+.Lctr32_done:
+___
+$code.=<<___	if ($flavour !~ /64/);
+	vldmia		sp!,{d8-d15}
+	ldmia		sp!,{r4-r10,pc}
+___
+$code.=<<___	if ($flavour =~ /64/);
+	ldr		x29,[sp],#16
+	ret
+___
+$code.=<<___;
+.size	${prefix}_ctr32_encrypt_blocks,.-${prefix}_ctr32_encrypt_blocks
+___
+}}}
+$code.=<<___;
+#endif
+___
+########################################
+if ($flavour =~ /64/) {			######## 64-bit code
+    my %opcode = (
+	"aesd"	=>	0x4e285800,	"aese"	=>	0x4e284800,
+	"aesimc"=>	0x4e287800,	"aesmc"	=>	0x4e286800	);
+
+    local *unaes = sub {
+	my ($mnemonic,$arg)=@_;
+
+	$arg =~ m/[qv]([0-9]+)[^,]*,\s*[qv]([0-9]+)/o	&&
+	sprintf ".inst\t0x%08x\t//%s %s",
+			$opcode{$mnemonic}|$1|($2<<5),
+			$mnemonic,$arg;
+    };
+
+    foreach(split("\n",$code)) {
+        s/\`([^\`]*)\`/eval($1)/geo;
+
+	s/\bq([0-9]+)\b/"v".($1<8?$1:$1+8).".16b"/geo;	# old->new registers
+        s/@\s/\/\//o;			# old->new style commentary
+
+	#s/[v]?(aes\w+)\s+([qv].*)/unaes($1,$2)/geo	or
+	s/cclr\s+([wx])([^,]+),\s*([a-z]+)/csel	$1$2,$1zr,$1$2,$3/o	or
+        s/vmov\.i8/movi/o	or	# fix up legacy mnemonics
+        s/vext\.8/ext/o		or
+        s/vrev32\.8/rev32/o	or
+        s/vtst\.8/cmtst/o	or
+        s/vshr/ushr/o		or
+        s/^(\s+)v/$1/o		or	# strip off v prefix
+	s/\bbx\s+lr\b/ret/o;
+
+	# fix up remainig legacy suffixes
+	s/\.[ui]?8//o;
+	m/\],#8/o and s/\.16b/\.8b/go;
+        s/\.[ui]?32//o and s/\.16b/\.4s/go;
+        s/\.[ui]?64//o and s/\.16b/\.2d/go;
+	s/\.[42]([sd])\[([0-3])\]/\.$1\[$2\]/o;
+
+        print $_,"\n";
+    }
+} else {				######## 32-bit code
+    my %opcode = (
+	"aesd"	=>	0xf3b00340,	"aese"	=>	0xf3b00300,
+	"aesimc"=>	0xf3b003c0,	"aesmc"	=>	0xf3b00380	);
+
+    local *unaes = sub {
+	my ($mnemonic,$arg)=@_;
+
+	if ($arg =~ m/[qv]([0-9]+)[^,]*,\s*[qv]([0-9]+)/o) {
+	    my $word = $opcode{$mnemonic}|(($1&7)<<13)|(($1&8)<<19)
+					 |(($2&7)<<1) |(($2&8)<<2);
+	    # since ARMv7 instructions are always encoded little-endian.
+	    # correct solution is to use .inst directive, but older
+	    # assemblers don't implement it:-(
+	    sprintf ".byte\t0x%02x,0x%02x,0x%02x,0x%02x\t@ %s %s",
+			$word&0xff,($word>>8)&0xff,
+			($word>>16)&0xff,($word>>24)&0xff,
+			$mnemonic,$arg;
+	}
+    };
+
+    sub unvtbl {
+	my $arg=shift;
+
+	$arg =~ m/q([0-9]+),\s*\{q([0-9]+)\},\s*q([0-9]+)/o &&
+	sprintf	"vtbl.8	d%d,{q%d},d%d\n\t".
+		"vtbl.8	d%d,{q%d},d%d", 2*$1,$2,2*$3, 2*$1+1,$2,2*$3+1;	
+    }
+
+    sub unvdup32 {
+	my $arg=shift;
+
+	$arg =~ m/q([0-9]+),\s*q([0-9]+)\[([0-3])\]/o &&
+	sprintf	"vdup.32	q%d,d%d[%d]",$1,2*$2+($3>>1),$3&1;	
+    }
+
+    sub unvmov32 {
+	my $arg=shift;
+
+	$arg =~ m/q([0-9]+)\[([0-3])\],(.*)/o &&
+	sprintf	"vmov.32	d%d[%d],%s",2*$1+($2>>1),$2&1,$3;	
+    }
+
+    foreach(split("\n",$code)) {
+        s/\`([^\`]*)\`/eval($1)/geo;
+
+	s/\b[wx]([0-9]+)\b/r$1/go;		# new->old registers
+	s/\bv([0-9])\.[12468]+[bsd]\b/q$1/go;	# new->old registers
+        s/\/\/\s?/@ /o;				# new->old style commentary
+
+	# fix up remainig new-style suffixes
+	s/\{q([0-9]+)\},\s*\[(.+)\],#8/sprintf "{d%d},[$2]!",2*$1/eo	or
+	s/\],#[0-9]+/]!/o;
+
+	s/[v]?(aes\w+)\s+([qv].*)/unaes($1,$2)/geo	or
+	s/cclr\s+([^,]+),\s*([a-z]+)/mov$2	$1,#0/o	or
+	s/vtbl\.8\s+(.*)/unvtbl($1)/geo			or
+	s/vdup\.32\s+(.*)/unvdup32($1)/geo		or
+	s/vmov\.32\s+(.*)/unvmov32($1)/geo		or
+	s/^(\s+)b\./$1b/o				or
+	s/^(\s+)ret/$1bx\tlr/o;
+
+        print $_,"\n";
+    }
+}
+
+close STDOUT;