crypto: arm/sha256 - Add optimized SHA-256/224

Add Andy Polyakov's optimized assembly and NEON implementations for
SHA-256/224.

The sha256-armv4.pl script for generating the assembly code is from
OpenSSL commit 51f8d095562f36cdaa6893597b5c609e943b0565.

Compared to sha256-generic these implementations have the following
tcrypt speed improvements on Motorola Nexus 6 (Snapdragon 805):

  bs    b/u      sha256-neon  sha256-asm
  16    16       x1.32        x1.19
  64    16       x1.27        x1.15
  64    64       x1.36        x1.20
  256   16       x1.22        x1.11
  256   64       x1.36        x1.19
  256   256      x1.59        x1.23
  1024  16       x1.21        x1.10
  1024  256      x1.65        x1.23
  1024  1024     x1.76        x1.25
  2048  16       x1.21        x1.10
  2048  256      x1.66        x1.23
  2048  1024     x1.78        x1.25
  2048  2048     x1.79        x1.25
  4096  16       x1.20        x1.09
  4096  256      x1.66        x1.23
  4096  1024     x1.79        x1.26
  4096  4096     x1.82        x1.26
  8192  16       x1.20        x1.09
  8192  256      x1.67        x1.23
  8192  1024     x1.80        x1.26
  8192  4096     x1.85        x1.28
  8192  8192     x1.85        x1.27

Where bs refers to block size and b/u to bytes per update.
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Cc: Andy Polyakov <appro@openssl.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

arch/arm/crypto/Kconfig

@@ -46,6 +46,13 @@ config CRYPTO_SHA2_ARM_CE
 	  SHA-256 secure hash standard (DFIPS 180-2) implemented
 	  using special ARMv8 Crypto Extensions.
 
+config CRYPTO_SHA256_ARM
+	tristate "SHA-224/256 digest algorithm (ARM-asm and NEON)"
+	select CRYPTO_HASH
+	help
+	  SHA-256 secure hash standard (DFIPS 180-2) implemented
+	  using optimized ARM assembler and NEON, when available.
+
 config CRYPTO_SHA512_ARM_NEON
 	tristate "SHA384 and SHA512 digest algorithm (ARM NEON)"
 	depends on KERNEL_MODE_NEON

arch/arm/crypto/Makefile

@@ -7,6 +7,7 @@ obj-$(CONFIG_CRYPTO_AES_ARM_BS) += aes-arm-bs.o
 obj-$(CONFIG_CRYPTO_AES_ARM_CE) += aes-arm-ce.o
 obj-$(CONFIG_CRYPTO_SHA1_ARM) += sha1-arm.o
 obj-$(CONFIG_CRYPTO_SHA1_ARM_NEON) += sha1-arm-neon.o
+obj-$(CONFIG_CRYPTO_SHA256_ARM) += sha256-arm.o
 obj-$(CONFIG_CRYPTO_SHA512_ARM_NEON) += sha512-arm-neon.o
 obj-$(CONFIG_CRYPTO_SHA1_ARM_CE) += sha1-arm-ce.o
 obj-$(CONFIG_CRYPTO_SHA2_ARM_CE) += sha2-arm-ce.o
@@ -16,6 +17,8 @@ aes-arm-y	:= aes-armv4.o aes_glue.o
 aes-arm-bs-y	:= aesbs-core.o aesbs-glue.o
 sha1-arm-y	:= sha1-armv4-large.o sha1_glue.o
 sha1-arm-neon-y	:= sha1-armv7-neon.o sha1_neon_glue.o
+sha256-arm-neon-$(CONFIG_KERNEL_MODE_NEON) := sha256_neon_glue.o
+sha256-arm-y	:= sha256-core.o sha256_glue.o $(sha256-arm-neon-y)
 sha512-arm-neon-y := sha512-armv7-neon.o sha512_neon_glue.o
 sha1-arm-ce-y	:= sha1-ce-core.o sha1-ce-glue.o
 sha2-arm-ce-y	:= sha2-ce-core.o sha2-ce-glue.o
@@ -28,4 +31,7 @@ quiet_cmd_perl = PERL    $@
 $(src)/aesbs-core.S_shipped: $(src)/bsaes-armv7.pl
 	$(call cmd,perl)
 
-.PRECIOUS: $(obj)/aesbs-core.S
+$(src)/sha256-core.S_shipped: $(src)/sha256-armv4.pl
+	$(call cmd,perl)
+
+.PRECIOUS: $(obj)/aesbs-core.S $(obj)/sha256-core.S

arch/arm/crypto/sha2-ce-glue.c

@@ -163,7 +163,7 @@ static struct shash_alg algs[] = { {
 	.base			= {
 		.cra_name		= "sha224",
 		.cra_driver_name	= "sha224-ce",
-		.cra_priority		= 200,
+		.cra_priority		= 300,
 		.cra_flags		= CRYPTO_ALG_TYPE_SHASH,
 		.cra_blocksize		= SHA256_BLOCK_SIZE,
 		.cra_module		= THIS_MODULE,
@@ -180,7 +180,7 @@ static struct shash_alg algs[] = { {
 	.base			= {
 		.cra_name		= "sha256",
 		.cra_driver_name	= "sha256-ce",
-		.cra_priority		= 200,
+		.cra_priority		= 300,
 		.cra_flags		= CRYPTO_ALG_TYPE_SHASH,
 		.cra_blocksize		= SHA256_BLOCK_SIZE,
 		.cra_module		= THIS_MODULE,

arch/arm/crypto/sha256-armv4.pl

+#!/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/.
+#
+# Permission to use under GPL terms is granted.
+# ====================================================================
+
+# SHA256 block procedure for ARMv4. May 2007.
+
+# Performance is ~2x better than gcc 3.4 generated code and in "abso-
+# lute" terms is ~2250 cycles per 64-byte block or ~35 cycles per
+# byte [on single-issue Xscale PXA250 core].
+
+# July 2010.
+#
+# Rescheduling for dual-issue pipeline resulted in 22% improvement on
+# Cortex A8 core and ~20 cycles per processed byte.
+
+# February 2011.
+#
+# Profiler-assisted and platform-specific optimization resulted in 16%
+# improvement on Cortex A8 core and ~15.4 cycles per processed byte.
+
+# September 2013.
+#
+# Add NEON implementation. On Cortex A8 it was measured to process one
+# byte in 12.5 cycles or 23% faster than integer-only code. Snapdragon
+# S4 does it in 12.5 cycles too, but it's 50% faster than integer-only
+# code (meaning that latter performs sub-optimally, nothing was done
+# about it).
+
+# May 2014.
+#
+# Add ARMv8 code path performing at 2.0 cpb on Apple A7.
+
+while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {}
+open STDOUT,">$output";
+
+$ctx="r0";	$t0="r0";
+$inp="r1";	$t4="r1";
+$len="r2";	$t1="r2";
+$T1="r3";	$t3="r3";
+$A="r4";
+$B="r5";
+$C="r6";
+$D="r7";
+$E="r8";
+$F="r9";
+$G="r10";
+$H="r11";
+@V=($A,$B,$C,$D,$E,$F,$G,$H);
+$t2="r12";
+$Ktbl="r14";
+
+@Sigma0=( 2,13,22);
+@Sigma1=( 6,11,25);
+@sigma0=( 7,18, 3);
+@sigma1=(17,19,10);
+
+sub BODY_00_15 {
+my ($i,$a,$b,$c,$d,$e,$f,$g,$h) = @_;
+
+$code.=<<___ if ($i<16);
+#if __ARM_ARCH__>=7
+	@ ldr	$t1,[$inp],#4			@ $i
+# if $i==15
+	str	$inp,[sp,#17*4]			@ make room for $t4
+# endif
+	eor	$t0,$e,$e,ror#`$Sigma1[1]-$Sigma1[0]`
+	add	$a,$a,$t2			@ h+=Maj(a,b,c) from the past
+	eor	$t0,$t0,$e,ror#`$Sigma1[2]-$Sigma1[0]`	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	$t1,$t1
+# endif
+#else
+	@ ldrb	$t1,[$inp,#3]			@ $i
+	add	$a,$a,$t2			@ h+=Maj(a,b,c) from the past
+	ldrb	$t2,[$inp,#2]
+	ldrb	$t0,[$inp,#1]
+	orr	$t1,$t1,$t2,lsl#8
+	ldrb	$t2,[$inp],#4
+	orr	$t1,$t1,$t0,lsl#16
+# if $i==15
+	str	$inp,[sp,#17*4]			@ make room for $t4
+# endif
+	eor	$t0,$e,$e,ror#`$Sigma1[1]-$Sigma1[0]`
+	orr	$t1,$t1,$t2,lsl#24
+	eor	$t0,$t0,$e,ror#`$Sigma1[2]-$Sigma1[0]`	@ Sigma1(e)
+#endif
+___
+$code.=<<___;
+	ldr	$t2,[$Ktbl],#4			@ *K256++
+	add	$h,$h,$t1			@ h+=X[i]
+	str	$t1,[sp,#`$i%16`*4]
+	eor	$t1,$f,$g
+	add	$h,$h,$t0,ror#$Sigma1[0]	@ h+=Sigma1(e)
+	and	$t1,$t1,$e
+	add	$h,$h,$t2			@ h+=K256[i]
+	eor	$t1,$t1,$g			@ Ch(e,f,g)
+	eor	$t0,$a,$a,ror#`$Sigma0[1]-$Sigma0[0]`
+	add	$h,$h,$t1			@ h+=Ch(e,f,g)
+#if $i==31
+	and	$t2,$t2,#0xff
+	cmp	$t2,#0xf2			@ done?
+#endif
+#if $i<15
+# if __ARM_ARCH__>=7
+	ldr	$t1,[$inp],#4			@ prefetch
+# else
+	ldrb	$t1,[$inp,#3]
+# endif
+	eor	$t2,$a,$b			@ a^b, b^c in next round
+#else
+	ldr	$t1,[sp,#`($i+2)%16`*4]		@ from future BODY_16_xx
+	eor	$t2,$a,$b			@ a^b, b^c in next round
+	ldr	$t4,[sp,#`($i+15)%16`*4]	@ from future BODY_16_xx
+#endif
+	eor	$t0,$t0,$a,ror#`$Sigma0[2]-$Sigma0[0]`	@ Sigma0(a)
+	and	$t3,$t3,$t2			@ (b^c)&=(a^b)
+	add	$d,$d,$h			@ d+=h
+	eor	$t3,$t3,$b			@ Maj(a,b,c)
+	add	$h,$h,$t0,ror#$Sigma0[0]	@ h+=Sigma0(a)
+	@ add	$h,$h,$t3			@ h+=Maj(a,b,c)
+___
+	($t2,$t3)=($t3,$t2);
+}
+
+sub BODY_16_XX {
+my ($i,$a,$b,$c,$d,$e,$f,$g,$h) = @_;
+
+$code.=<<___;
+	@ ldr	$t1,[sp,#`($i+1)%16`*4]		@ $i
+	@ ldr	$t4,[sp,#`($i+14)%16`*4]
+	mov	$t0,$t1,ror#$sigma0[0]
+	add	$a,$a,$t2			@ h+=Maj(a,b,c) from the past
+	mov	$t2,$t4,ror#$sigma1[0]
+	eor	$t0,$t0,$t1,ror#$sigma0[1]
+	eor	$t2,$t2,$t4,ror#$sigma1[1]
+	eor	$t0,$t0,$t1,lsr#$sigma0[2]	@ sigma0(X[i+1])
+	ldr	$t1,[sp,#`($i+0)%16`*4]
+	eor	$t2,$t2,$t4,lsr#$sigma1[2]	@ sigma1(X[i+14])
+	ldr	$t4,[sp,#`($i+9)%16`*4]
+
+	add	$t2,$t2,$t0
+	eor	$t0,$e,$e,ror#`$Sigma1[1]-$Sigma1[0]`	@ from BODY_00_15
+	add	$t1,$t1,$t2
+	eor	$t0,$t0,$e,ror#`$Sigma1[2]-$Sigma1[0]`	@ Sigma1(e)
+	add	$t1,$t1,$t4			@ X[i]
+___
+	&BODY_00_15(@_);
+}
+
+$code=<<___;
+#ifndef __KERNEL__
+# include "arm_arch.h"
+#else
+# define __ARM_ARCH__ __LINUX_ARM_ARCH__
+# define __ARM_MAX_ARCH__ 7
+#endif
+
+.text
+#if __ARM_ARCH__<7
+.code	32
+#else
+.syntax unified
+# ifdef __thumb2__
+#  define adrl adr
+.thumb
+# else
+.code   32
+# endif
+#endif
+
+.type	K256,%object
+.align	5
+K256:
+.word	0x428a2f98,0x71374491,0xb5c0fbcf,0xe9b5dba5
+.word	0x3956c25b,0x59f111f1,0x923f82a4,0xab1c5ed5
+.word	0xd807aa98,0x12835b01,0x243185be,0x550c7dc3
+.word	0x72be5d74,0x80deb1fe,0x9bdc06a7,0xc19bf174
+.word	0xe49b69c1,0xefbe4786,0x0fc19dc6,0x240ca1cc
+.word	0x2de92c6f,0x4a7484aa,0x5cb0a9dc,0x76f988da
+.word	0x983e5152,0xa831c66d,0xb00327c8,0xbf597fc7
+.word	0xc6e00bf3,0xd5a79147,0x06ca6351,0x14292967
+.word	0x27b70a85,0x2e1b2138,0x4d2c6dfc,0x53380d13
+.word	0x650a7354,0x766a0abb,0x81c2c92e,0x92722c85
+.word	0xa2bfe8a1,0xa81a664b,0xc24b8b70,0xc76c51a3
+.word	0xd192e819,0xd6990624,0xf40e3585,0x106aa070
+.word	0x19a4c116,0x1e376c08,0x2748774c,0x34b0bcb5
+.word	0x391c0cb3,0x4ed8aa4a,0x5b9cca4f,0x682e6ff3
+.word	0x748f82ee,0x78a5636f,0x84c87814,0x8cc70208
+.word	0x90befffa,0xa4506ceb,0xbef9a3f7,0xc67178f2
+.size	K256,.-K256
+.word	0				@ terminator
+#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__)
+.LOPENSSL_armcap:
+.word	OPENSSL_armcap_P-sha256_block_data_order
+#endif
+.align	5
+
+.global	sha256_block_data_order
+.type	sha256_block_data_order,%function
+sha256_block_data_order:
+#if __ARM_ARCH__<7
+	sub	r3,pc,#8		@ sha256_block_data_order
+#else
+	adr	r3,sha256_block_data_order
+#endif
+#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__)
+	ldr	r12,.LOPENSSL_armcap
+	ldr	r12,[r3,r12]		@ OPENSSL_armcap_P
+	tst	r12,#ARMV8_SHA256
+	bne	.LARMv8
+	tst	r12,#ARMV7_NEON
+	bne	.LNEON
+#endif
+	add	$len,$inp,$len,lsl#6	@ len to point at the end of inp
+	stmdb	sp!,{$ctx,$inp,$len,r4-r11,lr}
+	ldmia	$ctx,{$A,$B,$C,$D,$E,$F,$G,$H}
+	sub	$Ktbl,r3,#256+32	@ K256
+	sub	sp,sp,#16*4		@ alloca(X[16])
+.Loop:
+# if __ARM_ARCH__>=7
+	ldr	$t1,[$inp],#4
+# else
+	ldrb	$t1,[$inp,#3]
+# endif
+	eor	$t3,$B,$C		@ magic
+	eor	$t2,$t2,$t2
+___
+for($i=0;$i<16;$i++)	{ &BODY_00_15($i,@V); unshift(@V,pop(@V)); }
+$code.=".Lrounds_16_xx:\n";
+for (;$i<32;$i++)	{ &BODY_16_XX($i,@V); unshift(@V,pop(@V)); }
+$code.=<<___;
+#if __ARM_ARCH__>=7
+	ite	eq			@ Thumb2 thing, sanity check in ARM
+#endif
+	ldreq	$t3,[sp,#16*4]		@ pull ctx
+	bne	.Lrounds_16_xx
+
+	add	$A,$A,$t2		@ h+=Maj(a,b,c) from the past
+	ldr	$t0,[$t3,#0]
+	ldr	$t1,[$t3,#4]
+	ldr	$t2,[$t3,#8]
+	add	$A,$A,$t0
+	ldr	$t0,[$t3,#12]
+	add	$B,$B,$t1
+	ldr	$t1,[$t3,#16]
+	add	$C,$C,$t2
+	ldr	$t2,[$t3,#20]
+	add	$D,$D,$t0
+	ldr	$t0,[$t3,#24]
+	add	$E,$E,$t1
+	ldr	$t1,[$t3,#28]
+	add	$F,$F,$t2
+	ldr	$inp,[sp,#17*4]		@ pull inp
+	ldr	$t2,[sp,#18*4]		@ pull inp+len
+	add	$G,$G,$t0
+	add	$H,$H,$t1
+	stmia	$t3,{$A,$B,$C,$D,$E,$F,$G,$H}
+	cmp	$inp,$t2
+	sub	$Ktbl,$Ktbl,#256	@ rewind Ktbl
+	bne	.Loop
+
+	add	sp,sp,#`16+3`*4	@ destroy frame
+#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
+	bx	lr			@ interoperable with Thumb ISA:-)
+#endif
+.size	sha256_block_data_order,.-sha256_block_data_order
+___
+######################################################################
+# NEON stuff
+#
+{{{
+my @X=map("q$_",(0..3));
+my ($T0,$T1,$T2,$T3,$T4,$T5)=("q8","q9","q10","q11","d24","d25");
+my $Xfer=$t4;
+my $j=0;
+
+sub Dlo()   { shift=~m|q([1]?[0-9])|?"d".($1*2):"";     }
+sub Dhi()   { shift=~m|q([1]?[0-9])|?"d".($1*2+1):"";   }
+
+sub AUTOLOAD()          # thunk [simplified] x86-style perlasm
+{ my $opcode = $AUTOLOAD; $opcode =~ s/.*:://; $opcode =~ s/_/\./;
+  my $arg = pop;
+    $arg = "#$arg" if ($arg*1 eq $arg);
+    $code .= "\t$opcode\t".join(',',@_,$arg)."\n";
+}
+
+sub Xupdate()
+{ use integer;
+  my $body = shift;
+  my @insns = (&$body,&$body,&$body,&$body);
+  my ($a,$b,$c,$d,$e,$f,$g,$h);
+
+	&vext_8		($T0,@X[0],@X[1],4);	# X[1..4]
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vext_8		($T1,@X[2],@X[3],4);	# X[9..12]
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vshr_u32	($T2,$T0,$sigma0[0]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vadd_i32	(@X[0],@X[0],$T1);	# X[0..3] += X[9..12]
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vshr_u32	($T1,$T0,$sigma0[2]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vsli_32	($T2,$T0,32-$sigma0[0]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vshr_u32	($T3,$T0,$sigma0[1]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&veor		($T1,$T1,$T2);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vsli_32	($T3,$T0,32-$sigma0[1]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &vshr_u32	($T4,&Dhi(@X[3]),$sigma1[0]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&veor		($T1,$T1,$T3);		# sigma0(X[1..4])
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &vsli_32	($T4,&Dhi(@X[3]),32-$sigma1[0]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &vshr_u32	($T5,&Dhi(@X[3]),$sigma1[2]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vadd_i32	(@X[0],@X[0],$T1);	# X[0..3] += sigma0(X[1..4])
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &veor		($T5,$T5,$T4);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &vshr_u32	($T4,&Dhi(@X[3]),$sigma1[1]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &vsli_32	($T4,&Dhi(@X[3]),32-$sigma1[1]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &veor		($T5,$T5,$T4);		# sigma1(X[14..15])
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vadd_i32	(&Dlo(@X[0]),&Dlo(@X[0]),$T5);# X[0..1] += sigma1(X[14..15])
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &vshr_u32	($T4,&Dlo(@X[0]),$sigma1[0]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &vsli_32	($T4,&Dlo(@X[0]),32-$sigma1[0]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &vshr_u32	($T5,&Dlo(@X[0]),$sigma1[2]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &veor		($T5,$T5,$T4);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &vshr_u32	($T4,&Dlo(@X[0]),$sigma1[1]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vld1_32	("{$T0}","[$Ktbl,:128]!");
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &vsli_32	($T4,&Dlo(@X[0]),32-$sigma1[1]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	  &veor		($T5,$T5,$T4);		# sigma1(X[16..17])
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vadd_i32	(&Dhi(@X[0]),&Dhi(@X[0]),$T5);# X[2..3] += sigma1(X[16..17])
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vadd_i32	($T0,$T0,@X[0]);
+	 while($#insns>=2) { eval(shift(@insns)); }
+	&vst1_32	("{$T0}","[$Xfer,:128]!");
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+
+	push(@X,shift(@X));		# "rotate" X[]
+}
+
+sub Xpreload()
+{ use integer;
+  my $body = shift;
+  my @insns = (&$body,&$body,&$body,&$body);
+  my ($a,$b,$c,$d,$e,$f,$g,$h);
+
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vld1_32	("{$T0}","[$Ktbl,:128]!");
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vrev32_8	(@X[0],@X[0]);
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	 eval(shift(@insns));
+	&vadd_i32	($T0,$T0,@X[0]);
+	 foreach (@insns) { eval; }	# remaining instructions
+	&vst1_32	("{$T0}","[$Xfer,:128]!");
+
+	push(@X,shift(@X));		# "rotate" X[]
+}
+
+sub body_00_15 () {
+	(
+	'($a,$b,$c,$d,$e,$f,$g,$h)=@V;'.
+	'&add	($h,$h,$t1)',			# h+=X[i]+K[i]
+	'&eor	($t1,$f,$g)',
+	'&eor	($t0,$e,$e,"ror#".($Sigma1[1]-$Sigma1[0]))',
+	'&add	($a,$a,$t2)',			# h+=Maj(a,b,c) from the past
+	'&and	($t1,$t1,$e)',
+	'&eor	($t2,$t0,$e,"ror#".($Sigma1[2]-$Sigma1[0]))',	# Sigma1(e)
+	'&eor	($t0,$a,$a,"ror#".($Sigma0[1]-$Sigma0[0]))',
+	'&eor	($t1,$t1,$g)',			# Ch(e,f,g)
+	'&add	($h,$h,$t2,"ror#$Sigma1[0]")',	# h+=Sigma1(e)
+	'&eor	($t2,$a,$b)',			# a^b, b^c in next round
+	'&eor	($t0,$t0,$a,"ror#".($Sigma0[2]-$Sigma0[0]))',	# Sigma0(a)
+	'&add	($h,$h,$t1)',			# h+=Ch(e,f,g)
+	'&ldr	($t1,sprintf "[sp,#%d]",4*(($j+1)&15))	if (($j&15)!=15);'.
+	'&ldr	($t1,"[$Ktbl]")				if ($j==15);'.
+	'&ldr	($t1,"[sp,#64]")			if ($j==31)',
+	'&and	($t3,$t3,$t2)',			# (b^c)&=(a^b)
+	'&add	($d,$d,$h)',			# d+=h
+	'&add	($h,$h,$t0,"ror#$Sigma0[0]");'.	# h+=Sigma0(a)
+	'&eor	($t3,$t3,$b)',			# Maj(a,b,c)
+	'$j++;	unshift(@V,pop(@V)); ($t2,$t3)=($t3,$t2);'
+	)
+}
+
+$code.=<<___;
+#if __ARM_MAX_ARCH__>=7
+.arch	armv7-a
+.fpu	neon
+
+.global	sha256_block_data_order_neon
+.type	sha256_block_data_order_neon,%function
+.align	4
+sha256_block_data_order_neon:
+.LNEON:
+	stmdb	sp!,{r4-r12,lr}
+
+	sub	$H,sp,#16*4+16
+	adrl	$Ktbl,K256
+	bic	$H,$H,#15		@ align for 128-bit stores
+	mov	$t2,sp
+	mov	sp,$H			@ alloca
+	add	$len,$inp,$len,lsl#6	@ len to point at the end of inp
+
+	vld1.8		{@X[0]},[$inp]!
+	vld1.8		{@X[1]},[$inp]!
+	vld1.8		{@X[2]},[$inp]!
+	vld1.8		{@X[3]},[$inp]!
+	vld1.32		{$T0},[$Ktbl,:128]!
+	vld1.32		{$T1},[$Ktbl,:128]!
+	vld1.32		{$T2},[$Ktbl,:128]!
+	vld1.32		{$T3},[$Ktbl,:128]!
+	vrev32.8	@X[0],@X[0]		@ yes, even on
+	str		$ctx,[sp,#64]
+	vrev32.8	@X[1],@X[1]		@ big-endian
+	str		$inp,[sp,#68]
+	mov		$Xfer,sp
+	vrev32.8	@X[2],@X[2]
+	str		$len,[sp,#72]
+	vrev32.8	@X[3],@X[3]
+	str		$t2,[sp,#76]		@ save original sp
+	vadd.i32	$T0,$T0,@X[0]
+	vadd.i32	$T1,$T1,@X[1]
+	vst1.32		{$T0},[$Xfer,:128]!
+	vadd.i32	$T2,$T2,@X[2]
+	vst1.32		{$T1},[$Xfer,:128]!
+	vadd.i32	$T3,$T3,@X[3]
+	vst1.32		{$T2},[$Xfer,:128]!
+	vst1.32		{$T3},[$Xfer,:128]!
+
+	ldmia		$ctx,{$A-$H}
+	sub		$Xfer,$Xfer,#64
+	ldr		$t1,[sp,#0]
+	eor		$t2,$t2,$t2
+	eor		$t3,$B,$C
+	b		.L_00_48
+
+.align	4
+.L_00_48:
+___
+	&Xupdate(\&body_00_15);
+	&Xupdate(\&body_00_15);
+	&Xupdate(\&body_00_15);
+	&Xupdate(\&body_00_15);
+$code.=<<___;
+	teq	$t1,#0				@ check for K256 terminator
+	ldr	$t1,[sp,#0]
+	sub	$Xfer,$Xfer,#64
+	bne	.L_00_48
+
+	ldr		$inp,[sp,#68]
+	ldr		$t0,[sp,#72]
+	sub		$Ktbl,$Ktbl,#256	@ rewind $Ktbl
+	teq		$inp,$t0
+	it		eq
+	subeq		$inp,$inp,#64		@ avoid SEGV
+	vld1.8		{@X[0]},[$inp]!		@ load next input block
+	vld1.8		{@X[1]},[$inp]!
+	vld1.8		{@X[2]},[$inp]!
+	vld1.8		{@X[3]},[$inp]!
+	it		ne
+	strne		$inp,[sp,#68]
+	mov		$Xfer,sp
+___
+	&Xpreload(\&body_00_15);
+	&Xpreload(\&body_00_15);
+	&Xpreload(\&body_00_15);
+	&Xpreload(\&body_00_15);
+$code.=<<___;
+	ldr	$t0,[$t1,#0]
+	add	$A,$A,$t2			@ h+=Maj(a,b,c) from the past
+	ldr	$t2,[$t1,#4]
+	ldr	$t3,[$t1,#8]
+	ldr	$t4,[$t1,#12]
+	add	$A,$A,$t0			@ accumulate
+	ldr	$t0,[$t1,#16]
+	add	$B,$B,$t2
+	ldr	$t2,[$t1,#20]
+	add	$C,$C,$t3
+	ldr	$t3,[$t1,#24]
+	add	$D,$D,$t4
+	ldr	$t4,[$t1,#28]
+	add	$E,$E,$t0
+	str	$A,[$t1],#4
+	add	$F,$F,$t2
+	str	$B,[$t1],#4
+	add	$G,$G,$t3
+	str	$C,[$t1],#4
+	add	$H,$H,$t4
+	str	$D,[$t1],#4
+	stmia	$t1,{$E-$H}
+
+	ittte	ne
+	movne	$Xfer,sp
+	ldrne	$t1,[sp,#0]
+	eorne	$t2,$t2,$t2
+	ldreq	sp,[sp,#76]			@ restore original sp
+	itt	ne
+	eorne	$t3,$B,$C
+	bne	.L_00_48
+
+	ldmia	sp!,{r4-r12,pc}
+.size	sha256_block_data_order_neon,.-sha256_block_data_order_neon
+#endif
+___
+}}}
+######################################################################
+# ARMv8 stuff
+#
+{{{
+my ($ABCD,$EFGH,$abcd)=map("q$_",(0..2));
+my @MSG=map("q$_",(8..11));
+my ($W0,$W1,$ABCD_SAVE,$EFGH_SAVE)=map("q$_",(12..15));
+my $Ktbl="r3";
+
+$code.=<<___;
+#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__)
+
+# ifdef __thumb2__
+#  define INST(a,b,c,d)	.byte	c,d|0xc,a,b
+# else
+#  define INST(a,b,c,d)	.byte	a,b,c,d
+# endif
+
+.type	sha256_block_data_order_armv8,%function
+.align	5
+sha256_block_data_order_armv8:
+.LARMv8:
+	vld1.32	{$ABCD,$EFGH},[$ctx]
+# ifdef __thumb2__
+	adr	$Ktbl,.LARMv8
+	sub	$Ktbl,$Ktbl,#.LARMv8-K256
+# else
+	adrl	$Ktbl,K256
+# endif
+	add	$len,$inp,$len,lsl#6	@ len to point at the end of inp
+
+.Loop_v8:
+	vld1.8		{@MSG[0]-@MSG[1]},[$inp]!
+	vld1.8		{@MSG[2]-@MSG[3]},[$inp]!
+	vld1.32		{$W0},[$Ktbl]!
+	vrev32.8	@MSG[0],@MSG[0]
+	vrev32.8	@MSG[1],@MSG[1]
+	vrev32.8	@MSG[2],@MSG[2]
+	vrev32.8	@MSG[3],@MSG[3]
+	vmov		$ABCD_SAVE,$ABCD	@ offload
+	vmov		$EFGH_SAVE,$EFGH
+	teq		$inp,$len
+___
+for($i=0;$i<12;$i++) {
+$code.=<<___;
+	vld1.32		{$W1},[$Ktbl]!
+	vadd.i32	$W0,$W0,@MSG[0]
+	sha256su0	@MSG[0],@MSG[1]
+	vmov		$abcd,$ABCD
+	sha256h		$ABCD,$EFGH,$W0
+	sha256h2	$EFGH,$abcd,$W0
+	sha256su1	@MSG[0],@MSG[2],@MSG[3]
+___
+	($W0,$W1)=($W1,$W0);	push(@MSG,shift(@MSG));
+}
+$code.=<<___;
+	vld1.32		{$W1},[$Ktbl]!
+	vadd.i32	$W0,$W0,@MSG[0]
+	vmov		$abcd,$ABCD
+	sha256h		$ABCD,$EFGH,$W0
+	sha256h2	$EFGH,$abcd,$W0
+
+	vld1.32		{$W0},[$Ktbl]!
+	vadd.i32	$W1,$W1,@MSG[1]
+	vmov		$abcd,$ABCD
+	sha256h		$ABCD,$EFGH,$W1
+	sha256h2	$EFGH,$abcd,$W1
+
+	vld1.32		{$W1},[$Ktbl]
+	vadd.i32	$W0,$W0,@MSG[2]
+	sub		$Ktbl,$Ktbl,#256-16	@ rewind
+	vmov		$abcd,$ABCD
+	sha256h		$ABCD,$EFGH,$W0
+	sha256h2	$EFGH,$abcd,$W0
+
+	vadd.i32	$W1,$W1,@MSG[3]
+	vmov		$abcd,$ABCD
+	sha256h		$ABCD,$EFGH,$W1
+	sha256h2	$EFGH,$abcd,$W1
+
+	vadd.i32	$ABCD,$ABCD,$ABCD_SAVE
+	vadd.i32	$EFGH,$EFGH,$EFGH_SAVE
+	it		ne
+	bne		.Loop_v8
+
+	vst1.32		{$ABCD,$EFGH},[$ctx]
+
+	ret		@ bx lr
+.size	sha256_block_data_order_armv8,.-sha256_block_data_order_armv8
+#endif
+___
+}}}
+$code.=<<___;
+.asciz  "SHA256 block transform for ARMv4/NEON/ARMv8, CRYPTOGAMS by <appro\@openssl.org>"
+.align	2
+#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__)
+.comm   OPENSSL_armcap_P,4,4
+#endif
+___
+
+open SELF,$0;
+while(<SELF>) {
+	next if (/^#!/);
+	last if (!s/^#/@/ and !/^$/);
+	print;
+}
+close SELF;
+
+{   my  %opcode = (
+	"sha256h"	=> 0xf3000c40,	"sha256h2"	=> 0xf3100c40,
+	"sha256su0"	=> 0xf3ba03c0,	"sha256su1"	=> 0xf3200c40	);
+
+    sub unsha256 {
+	my ($mnemonic,$arg)=@_;
+
+	if ($arg =~ m/q([0-9]+)(?:,\s*q([0-9]+))?,\s*q([0-9]+)/o) {
+	    my $word = $opcode{$mnemonic}|(($1&7)<<13)|(($1&8)<<19)
+					 |(($2&7)<<17)|(($2&8)<<4)
+					 |(($3&7)<<1) |(($3&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 "INST(0x%02x,0x%02x,0x%02x,0x%02x)\t@ %s %s",
+			$word&0xff,($word>>8)&0xff,
+			($word>>16)&0xff,($word>>24)&0xff,
+			$mnemonic,$arg;
+	}
+    }
+}
+
+foreach (split($/,$code)) {
+
+	s/\`([^\`]*)\`/eval $1/geo;
+
+	s/\b(sha256\w+)\s+(q.*)/unsha256($1,$2)/geo;
+
+	s/\bret\b/bx	lr/go		or
+	s/\bbx\s+lr\b/.word\t0xe12fff1e/go;	# make it possible to compile with -march=armv4
+
+	print $_,"\n";
+}
+
+close STDOUT; # enforce flush

arch/arm/crypto/sha256-core.S_shipped

+
+@ ====================================================================
+@ 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/.
+@
+@ Permission to use under GPL terms is granted.
+@ ====================================================================
+
+@ SHA256 block procedure for ARMv4. May 2007.
+
+@ Performance is ~2x better than gcc 3.4 generated code and in "abso-
+@ lute" terms is ~2250 cycles per 64-byte block or ~35 cycles per
+@ byte [on single-issue Xscale PXA250 core].
+
+@ July 2010.
+@
+@ Rescheduling for dual-issue pipeline resulted in 22% improvement on
+@ Cortex A8 core and ~20 cycles per processed byte.
+
+@ February 2011.
+@
+@ Profiler-assisted and platform-specific optimization resulted in 16%
+@ improvement on Cortex A8 core and ~15.4 cycles per processed byte.
+
+@ September 2013.
+@
+@ Add NEON implementation. On Cortex A8 it was measured to process one
+@ byte in 12.5 cycles or 23% faster than integer-only code. Snapdragon
+@ S4 does it in 12.5 cycles too, but it's 50% faster than integer-only
+@ code (meaning that latter performs sub-optimally, nothing was done
+@ about it).
+
+@ May 2014.
+@
+@ Add ARMv8 code path performing at 2.0 cpb on Apple A7.
+
+#ifndef __KERNEL__
+# include "arm_arch.h"
+#else
+# define __ARM_ARCH__ __LINUX_ARM_ARCH__
+# define __ARM_MAX_ARCH__ 7
+#endif
+
+.text
+#if __ARM_ARCH__<7
+.code	32
+#else
+.syntax unified
+# ifdef __thumb2__
+#  define adrl adr
+.thumb
+# else
+.code   32
+# endif
+#endif
+
+.type	K256,%object
+.align	5
+K256:
+.word	0x428a2f98,0x71374491,0xb5c0fbcf,0xe9b5dba5
+.word	0x3956c25b,0x59f111f1,0x923f82a4,0xab1c5ed5
+.word	0xd807aa98,0x12835b01,0x243185be,0x550c7dc3
+.word	0x72be5d74,0x80deb1fe,0x9bdc06a7,0xc19bf174
+.word	0xe49b69c1,0xefbe4786,0x0fc19dc6,0x240ca1cc
+.word	0x2de92c6f,0x4a7484aa,0x5cb0a9dc,0x76f988da
+.word	0x983e5152,0xa831c66d,0xb00327c8,0xbf597fc7
+.word	0xc6e00bf3,0xd5a79147,0x06ca6351,0x14292967
+.word	0x27b70a85,0x2e1b2138,0x4d2c6dfc,0x53380d13
+.word	0x650a7354,0x766a0abb,0x81c2c92e,0x92722c85
+.word	0xa2bfe8a1,0xa81a664b,0xc24b8b70,0xc76c51a3
+.word	0xd192e819,0xd6990624,0xf40e3585,0x106aa070
+.word	0x19a4c116,0x1e376c08,0x2748774c,0x34b0bcb5
+.word	0x391c0cb3,0x4ed8aa4a,0x5b9cca4f,0x682e6ff3
+.word	0x748f82ee,0x78a5636f,0x84c87814,0x8cc70208
+.word	0x90befffa,0xa4506ceb,0xbef9a3f7,0xc67178f2
+.size	K256,.-K256
+.word	0				@ terminator
+#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__)
+.LOPENSSL_armcap:
+.word	OPENSSL_armcap_P-sha256_block_data_order
+#endif
+.align	5
+
+.global	sha256_block_data_order
+.type	sha256_block_data_order,%function
+sha256_block_data_order:
+#if __ARM_ARCH__<7
+	sub	r3,pc,#8		@ sha256_block_data_order
+#else
+	adr	r3,sha256_block_data_order
+#endif
+#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__)
+	ldr	r12,.LOPENSSL_armcap
+	ldr	r12,[r3,r12]		@ OPENSSL_armcap_P
+	tst	r12,#ARMV8_SHA256
+	bne	.LARMv8
+	tst	r12,#ARMV7_NEON
+	bne	.LNEON
+#endif
+	add	r2,r1,r2,lsl#6	@ len to point at the end of inp
+	stmdb	sp!,{r0,r1,r2,r4-r11,lr}
+	ldmia	r0,{r4,r5,r6,r7,r8,r9,r10,r11}
+	sub	r14,r3,#256+32	@ K256
+	sub	sp,sp,#16*4		@ alloca(X[16])
+.Loop:
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r5,r6		@ magic
+	eor	r12,r12,r12
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 0
+# if 0==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r8,r8,ror#5
+	add	r4,r4,r12			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r8,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 0
+	add	r4,r4,r12			@ h+=Maj(a,b,c) from the past
+	ldrb	r12,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r12,lsl#8
+	ldrb	r12,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 0==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r8,r8,ror#5
+	orr	r2,r2,r12,lsl#24
+	eor	r0,r0,r8,ror#19	@ Sigma1(e)
+#endif
+	ldr	r12,[r14],#4			@ *K256++
+	add	r11,r11,r2			@ h+=X[i]
+	str	r2,[sp,#0*4]
+	eor	r2,r9,r10
+	add	r11,r11,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r8
+	add	r11,r11,r12			@ h+=K256[i]
+	eor	r2,r2,r10			@ Ch(e,f,g)
+	eor	r0,r4,r4,ror#11
+	add	r11,r11,r2			@ h+=Ch(e,f,g)
+#if 0==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 0<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r4,r5			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#2*4]		@ from future BODY_16_xx
+	eor	r12,r4,r5			@ a^b, b^c in next round
+	ldr	r1,[sp,#15*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r4,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r7,r7,r11			@ d+=h
+	eor	r3,r3,r5			@ Maj(a,b,c)
+	add	r11,r11,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r11,r11,r3			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 1
+# if 1==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r7,r7,ror#5
+	add	r11,r11,r3			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r7,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 1
+	add	r11,r11,r3			@ h+=Maj(a,b,c) from the past
+	ldrb	r3,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r3,lsl#8
+	ldrb	r3,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 1==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r7,r7,ror#5
+	orr	r2,r2,r3,lsl#24
+	eor	r0,r0,r7,ror#19	@ Sigma1(e)
+#endif
+	ldr	r3,[r14],#4			@ *K256++
+	add	r10,r10,r2			@ h+=X[i]
+	str	r2,[sp,#1*4]
+	eor	r2,r8,r9
+	add	r10,r10,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r7
+	add	r10,r10,r3			@ h+=K256[i]
+	eor	r2,r2,r9			@ Ch(e,f,g)
+	eor	r0,r11,r11,ror#11
+	add	r10,r10,r2			@ h+=Ch(e,f,g)
+#if 1==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 1<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r11,r4			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#3*4]		@ from future BODY_16_xx
+	eor	r3,r11,r4			@ a^b, b^c in next round
+	ldr	r1,[sp,#0*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r11,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r6,r6,r10			@ d+=h
+	eor	r12,r12,r4			@ Maj(a,b,c)
+	add	r10,r10,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r10,r10,r12			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 2
+# if 2==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r6,r6,ror#5
+	add	r10,r10,r12			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r6,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 2
+	add	r10,r10,r12			@ h+=Maj(a,b,c) from the past
+	ldrb	r12,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r12,lsl#8
+	ldrb	r12,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 2==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r6,r6,ror#5
+	orr	r2,r2,r12,lsl#24
+	eor	r0,r0,r6,ror#19	@ Sigma1(e)
+#endif
+	ldr	r12,[r14],#4			@ *K256++
+	add	r9,r9,r2			@ h+=X[i]
+	str	r2,[sp,#2*4]
+	eor	r2,r7,r8
+	add	r9,r9,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r6
+	add	r9,r9,r12			@ h+=K256[i]
+	eor	r2,r2,r8			@ Ch(e,f,g)
+	eor	r0,r10,r10,ror#11
+	add	r9,r9,r2			@ h+=Ch(e,f,g)
+#if 2==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 2<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r10,r11			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#4*4]		@ from future BODY_16_xx
+	eor	r12,r10,r11			@ a^b, b^c in next round
+	ldr	r1,[sp,#1*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r10,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r5,r5,r9			@ d+=h
+	eor	r3,r3,r11			@ Maj(a,b,c)
+	add	r9,r9,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r9,r9,r3			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 3
+# if 3==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r5,r5,ror#5
+	add	r9,r9,r3			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r5,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 3
+	add	r9,r9,r3			@ h+=Maj(a,b,c) from the past
+	ldrb	r3,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r3,lsl#8
+	ldrb	r3,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 3==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r5,r5,ror#5
+	orr	r2,r2,r3,lsl#24
+	eor	r0,r0,r5,ror#19	@ Sigma1(e)
+#endif
+	ldr	r3,[r14],#4			@ *K256++
+	add	r8,r8,r2			@ h+=X[i]
+	str	r2,[sp,#3*4]
+	eor	r2,r6,r7
+	add	r8,r8,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r5
+	add	r8,r8,r3			@ h+=K256[i]
+	eor	r2,r2,r7			@ Ch(e,f,g)
+	eor	r0,r9,r9,ror#11
+	add	r8,r8,r2			@ h+=Ch(e,f,g)
+#if 3==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 3<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r9,r10			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#5*4]		@ from future BODY_16_xx
+	eor	r3,r9,r10			@ a^b, b^c in next round
+	ldr	r1,[sp,#2*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r9,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r4,r4,r8			@ d+=h
+	eor	r12,r12,r10			@ Maj(a,b,c)
+	add	r8,r8,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r8,r8,r12			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 4
+# if 4==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r4,r4,ror#5
+	add	r8,r8,r12			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r4,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 4
+	add	r8,r8,r12			@ h+=Maj(a,b,c) from the past
+	ldrb	r12,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r12,lsl#8
+	ldrb	r12,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 4==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r4,r4,ror#5
+	orr	r2,r2,r12,lsl#24
+	eor	r0,r0,r4,ror#19	@ Sigma1(e)
+#endif
+	ldr	r12,[r14],#4			@ *K256++
+	add	r7,r7,r2			@ h+=X[i]
+	str	r2,[sp,#4*4]
+	eor	r2,r5,r6
+	add	r7,r7,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r4
+	add	r7,r7,r12			@ h+=K256[i]
+	eor	r2,r2,r6			@ Ch(e,f,g)
+	eor	r0,r8,r8,ror#11
+	add	r7,r7,r2			@ h+=Ch(e,f,g)
+#if 4==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 4<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r8,r9			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#6*4]		@ from future BODY_16_xx
+	eor	r12,r8,r9			@ a^b, b^c in next round
+	ldr	r1,[sp,#3*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r8,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r11,r11,r7			@ d+=h
+	eor	r3,r3,r9			@ Maj(a,b,c)
+	add	r7,r7,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r7,r7,r3			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 5
+# if 5==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r11,r11,ror#5
+	add	r7,r7,r3			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r11,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 5
+	add	r7,r7,r3			@ h+=Maj(a,b,c) from the past
+	ldrb	r3,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r3,lsl#8
+	ldrb	r3,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 5==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r11,r11,ror#5
+	orr	r2,r2,r3,lsl#24
+	eor	r0,r0,r11,ror#19	@ Sigma1(e)
+#endif
+	ldr	r3,[r14],#4			@ *K256++
+	add	r6,r6,r2			@ h+=X[i]
+	str	r2,[sp,#5*4]
+	eor	r2,r4,r5
+	add	r6,r6,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r11
+	add	r6,r6,r3			@ h+=K256[i]
+	eor	r2,r2,r5			@ Ch(e,f,g)
+	eor	r0,r7,r7,ror#11
+	add	r6,r6,r2			@ h+=Ch(e,f,g)
+#if 5==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 5<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r7,r8			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#7*4]		@ from future BODY_16_xx
+	eor	r3,r7,r8			@ a^b, b^c in next round
+	ldr	r1,[sp,#4*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r7,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r10,r10,r6			@ d+=h
+	eor	r12,r12,r8			@ Maj(a,b,c)
+	add	r6,r6,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r6,r6,r12			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 6
+# if 6==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r10,r10,ror#5
+	add	r6,r6,r12			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r10,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 6
+	add	r6,r6,r12			@ h+=Maj(a,b,c) from the past
+	ldrb	r12,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r12,lsl#8
+	ldrb	r12,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 6==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r10,r10,ror#5
+	orr	r2,r2,r12,lsl#24
+	eor	r0,r0,r10,ror#19	@ Sigma1(e)
+#endif
+	ldr	r12,[r14],#4			@ *K256++
+	add	r5,r5,r2			@ h+=X[i]
+	str	r2,[sp,#6*4]
+	eor	r2,r11,r4
+	add	r5,r5,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r10
+	add	r5,r5,r12			@ h+=K256[i]
+	eor	r2,r2,r4			@ Ch(e,f,g)
+	eor	r0,r6,r6,ror#11
+	add	r5,r5,r2			@ h+=Ch(e,f,g)
+#if 6==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 6<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r6,r7			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#8*4]		@ from future BODY_16_xx
+	eor	r12,r6,r7			@ a^b, b^c in next round
+	ldr	r1,[sp,#5*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r6,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r9,r9,r5			@ d+=h
+	eor	r3,r3,r7			@ Maj(a,b,c)
+	add	r5,r5,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r5,r5,r3			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 7
+# if 7==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r9,r9,ror#5
+	add	r5,r5,r3			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r9,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 7
+	add	r5,r5,r3			@ h+=Maj(a,b,c) from the past
+	ldrb	r3,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r3,lsl#8
+	ldrb	r3,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 7==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r9,r9,ror#5
+	orr	r2,r2,r3,lsl#24
+	eor	r0,r0,r9,ror#19	@ Sigma1(e)
+#endif
+	ldr	r3,[r14],#4			@ *K256++
+	add	r4,r4,r2			@ h+=X[i]
+	str	r2,[sp,#7*4]
+	eor	r2,r10,r11
+	add	r4,r4,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r9
+	add	r4,r4,r3			@ h+=K256[i]
+	eor	r2,r2,r11			@ Ch(e,f,g)
+	eor	r0,r5,r5,ror#11
+	add	r4,r4,r2			@ h+=Ch(e,f,g)
+#if 7==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 7<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r5,r6			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#9*4]		@ from future BODY_16_xx
+	eor	r3,r5,r6			@ a^b, b^c in next round
+	ldr	r1,[sp,#6*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r5,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r8,r8,r4			@ d+=h
+	eor	r12,r12,r6			@ Maj(a,b,c)
+	add	r4,r4,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r4,r4,r12			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 8
+# if 8==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r8,r8,ror#5
+	add	r4,r4,r12			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r8,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 8
+	add	r4,r4,r12			@ h+=Maj(a,b,c) from the past
+	ldrb	r12,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r12,lsl#8
+	ldrb	r12,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 8==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r8,r8,ror#5
+	orr	r2,r2,r12,lsl#24
+	eor	r0,r0,r8,ror#19	@ Sigma1(e)
+#endif
+	ldr	r12,[r14],#4			@ *K256++
+	add	r11,r11,r2			@ h+=X[i]
+	str	r2,[sp,#8*4]
+	eor	r2,r9,r10
+	add	r11,r11,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r8
+	add	r11,r11,r12			@ h+=K256[i]
+	eor	r2,r2,r10			@ Ch(e,f,g)
+	eor	r0,r4,r4,ror#11
+	add	r11,r11,r2			@ h+=Ch(e,f,g)
+#if 8==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 8<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r4,r5			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#10*4]		@ from future BODY_16_xx
+	eor	r12,r4,r5			@ a^b, b^c in next round
+	ldr	r1,[sp,#7*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r4,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r7,r7,r11			@ d+=h
+	eor	r3,r3,r5			@ Maj(a,b,c)
+	add	r11,r11,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r11,r11,r3			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 9
+# if 9==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r7,r7,ror#5
+	add	r11,r11,r3			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r7,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 9
+	add	r11,r11,r3			@ h+=Maj(a,b,c) from the past
+	ldrb	r3,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r3,lsl#8
+	ldrb	r3,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 9==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r7,r7,ror#5
+	orr	r2,r2,r3,lsl#24
+	eor	r0,r0,r7,ror#19	@ Sigma1(e)
+#endif
+	ldr	r3,[r14],#4			@ *K256++
+	add	r10,r10,r2			@ h+=X[i]
+	str	r2,[sp,#9*4]
+	eor	r2,r8,r9
+	add	r10,r10,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r7
+	add	r10,r10,r3			@ h+=K256[i]
+	eor	r2,r2,r9			@ Ch(e,f,g)
+	eor	r0,r11,r11,ror#11
+	add	r10,r10,r2			@ h+=Ch(e,f,g)
+#if 9==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 9<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r11,r4			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#11*4]		@ from future BODY_16_xx
+	eor	r3,r11,r4			@ a^b, b^c in next round
+	ldr	r1,[sp,#8*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r11,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r6,r6,r10			@ d+=h
+	eor	r12,r12,r4			@ Maj(a,b,c)
+	add	r10,r10,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r10,r10,r12			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 10
+# if 10==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r6,r6,ror#5
+	add	r10,r10,r12			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r6,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 10
+	add	r10,r10,r12			@ h+=Maj(a,b,c) from the past
+	ldrb	r12,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r12,lsl#8
+	ldrb	r12,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 10==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r6,r6,ror#5
+	orr	r2,r2,r12,lsl#24
+	eor	r0,r0,r6,ror#19	@ Sigma1(e)
+#endif
+	ldr	r12,[r14],#4			@ *K256++
+	add	r9,r9,r2			@ h+=X[i]
+	str	r2,[sp,#10*4]
+	eor	r2,r7,r8
+	add	r9,r9,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r6
+	add	r9,r9,r12			@ h+=K256[i]
+	eor	r2,r2,r8			@ Ch(e,f,g)
+	eor	r0,r10,r10,ror#11
+	add	r9,r9,r2			@ h+=Ch(e,f,g)
+#if 10==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 10<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r10,r11			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#12*4]		@ from future BODY_16_xx
+	eor	r12,r10,r11			@ a^b, b^c in next round
+	ldr	r1,[sp,#9*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r10,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r5,r5,r9			@ d+=h
+	eor	r3,r3,r11			@ Maj(a,b,c)
+	add	r9,r9,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r9,r9,r3			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 11
+# if 11==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r5,r5,ror#5
+	add	r9,r9,r3			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r5,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 11
+	add	r9,r9,r3			@ h+=Maj(a,b,c) from the past
+	ldrb	r3,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r3,lsl#8
+	ldrb	r3,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 11==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r5,r5,ror#5
+	orr	r2,r2,r3,lsl#24
+	eor	r0,r0,r5,ror#19	@ Sigma1(e)
+#endif
+	ldr	r3,[r14],#4			@ *K256++
+	add	r8,r8,r2			@ h+=X[i]
+	str	r2,[sp,#11*4]
+	eor	r2,r6,r7
+	add	r8,r8,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r5
+	add	r8,r8,r3			@ h+=K256[i]
+	eor	r2,r2,r7			@ Ch(e,f,g)
+	eor	r0,r9,r9,ror#11
+	add	r8,r8,r2			@ h+=Ch(e,f,g)
+#if 11==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 11<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r9,r10			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#13*4]		@ from future BODY_16_xx
+	eor	r3,r9,r10			@ a^b, b^c in next round
+	ldr	r1,[sp,#10*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r9,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r4,r4,r8			@ d+=h
+	eor	r12,r12,r10			@ Maj(a,b,c)
+	add	r8,r8,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r8,r8,r12			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 12
+# if 12==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r4,r4,ror#5
+	add	r8,r8,r12			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r4,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 12
+	add	r8,r8,r12			@ h+=Maj(a,b,c) from the past
+	ldrb	r12,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r12,lsl#8
+	ldrb	r12,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 12==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r4,r4,ror#5
+	orr	r2,r2,r12,lsl#24
+	eor	r0,r0,r4,ror#19	@ Sigma1(e)
+#endif
+	ldr	r12,[r14],#4			@ *K256++
+	add	r7,r7,r2			@ h+=X[i]
+	str	r2,[sp,#12*4]
+	eor	r2,r5,r6
+	add	r7,r7,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r4
+	add	r7,r7,r12			@ h+=K256[i]
+	eor	r2,r2,r6			@ Ch(e,f,g)
+	eor	r0,r8,r8,ror#11
+	add	r7,r7,r2			@ h+=Ch(e,f,g)
+#if 12==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 12<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r8,r9			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#14*4]		@ from future BODY_16_xx
+	eor	r12,r8,r9			@ a^b, b^c in next round
+	ldr	r1,[sp,#11*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r8,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r11,r11,r7			@ d+=h
+	eor	r3,r3,r9			@ Maj(a,b,c)
+	add	r7,r7,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r7,r7,r3			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 13
+# if 13==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r11,r11,ror#5
+	add	r7,r7,r3			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r11,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 13
+	add	r7,r7,r3			@ h+=Maj(a,b,c) from the past
+	ldrb	r3,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r3,lsl#8
+	ldrb	r3,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 13==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r11,r11,ror#5
+	orr	r2,r2,r3,lsl#24
+	eor	r0,r0,r11,ror#19	@ Sigma1(e)
+#endif
+	ldr	r3,[r14],#4			@ *K256++
+	add	r6,r6,r2			@ h+=X[i]
+	str	r2,[sp,#13*4]
+	eor	r2,r4,r5
+	add	r6,r6,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r11
+	add	r6,r6,r3			@ h+=K256[i]
+	eor	r2,r2,r5			@ Ch(e,f,g)
+	eor	r0,r7,r7,ror#11
+	add	r6,r6,r2			@ h+=Ch(e,f,g)
+#if 13==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 13<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r7,r8			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#15*4]		@ from future BODY_16_xx
+	eor	r3,r7,r8			@ a^b, b^c in next round
+	ldr	r1,[sp,#12*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r7,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r10,r10,r6			@ d+=h
+	eor	r12,r12,r8			@ Maj(a,b,c)
+	add	r6,r6,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r6,r6,r12			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 14
+# if 14==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r10,r10,ror#5
+	add	r6,r6,r12			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r10,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 14
+	add	r6,r6,r12			@ h+=Maj(a,b,c) from the past
+	ldrb	r12,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r12,lsl#8
+	ldrb	r12,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 14==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r10,r10,ror#5
+	orr	r2,r2,r12,lsl#24
+	eor	r0,r0,r10,ror#19	@ Sigma1(e)
+#endif
+	ldr	r12,[r14],#4			@ *K256++
+	add	r5,r5,r2			@ h+=X[i]
+	str	r2,[sp,#14*4]
+	eor	r2,r11,r4
+	add	r5,r5,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r10
+	add	r5,r5,r12			@ h+=K256[i]
+	eor	r2,r2,r4			@ Ch(e,f,g)
+	eor	r0,r6,r6,ror#11
+	add	r5,r5,r2			@ h+=Ch(e,f,g)
+#if 14==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 14<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r6,r7			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#0*4]		@ from future BODY_16_xx
+	eor	r12,r6,r7			@ a^b, b^c in next round
+	ldr	r1,[sp,#13*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r6,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r9,r9,r5			@ d+=h
+	eor	r3,r3,r7			@ Maj(a,b,c)
+	add	r5,r5,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r5,r5,r3			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	@ ldr	r2,[r1],#4			@ 15
+# if 15==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r9,r9,ror#5
+	add	r5,r5,r3			@ h+=Maj(a,b,c) from the past
+	eor	r0,r0,r9,ror#19	@ Sigma1(e)
+# ifndef __ARMEB__
+	rev	r2,r2
+# endif
+#else
+	@ ldrb	r2,[r1,#3]			@ 15
+	add	r5,r5,r3			@ h+=Maj(a,b,c) from the past
+	ldrb	r3,[r1,#2]
+	ldrb	r0,[r1,#1]
+	orr	r2,r2,r3,lsl#8
+	ldrb	r3,[r1],#4
+	orr	r2,r2,r0,lsl#16
+# if 15==15
+	str	r1,[sp,#17*4]			@ make room for r1
+# endif
+	eor	r0,r9,r9,ror#5
+	orr	r2,r2,r3,lsl#24
+	eor	r0,r0,r9,ror#19	@ Sigma1(e)
+#endif
+	ldr	r3,[r14],#4			@ *K256++
+	add	r4,r4,r2			@ h+=X[i]
+	str	r2,[sp,#15*4]
+	eor	r2,r10,r11
+	add	r4,r4,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r9
+	add	r4,r4,r3			@ h+=K256[i]
+	eor	r2,r2,r11			@ Ch(e,f,g)
+	eor	r0,r5,r5,ror#11
+	add	r4,r4,r2			@ h+=Ch(e,f,g)
+#if 15==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 15<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r5,r6			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#1*4]		@ from future BODY_16_xx
+	eor	r3,r5,r6			@ a^b, b^c in next round
+	ldr	r1,[sp,#14*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r5,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r8,r8,r4			@ d+=h
+	eor	r12,r12,r6			@ Maj(a,b,c)
+	add	r4,r4,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r4,r4,r12			@ h+=Maj(a,b,c)
+.Lrounds_16_xx:
+	@ ldr	r2,[sp,#1*4]		@ 16
+	@ ldr	r1,[sp,#14*4]
+	mov	r0,r2,ror#7
+	add	r4,r4,r12			@ h+=Maj(a,b,c) from the past
+	mov	r12,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r12,r12,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#0*4]
+	eor	r12,r12,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#9*4]
+
+	add	r12,r12,r0
+	eor	r0,r8,r8,ror#5	@ from BODY_00_15
+	add	r2,r2,r12
+	eor	r0,r0,r8,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r12,[r14],#4			@ *K256++
+	add	r11,r11,r2			@ h+=X[i]
+	str	r2,[sp,#0*4]
+	eor	r2,r9,r10
+	add	r11,r11,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r8
+	add	r11,r11,r12			@ h+=K256[i]
+	eor	r2,r2,r10			@ Ch(e,f,g)
+	eor	r0,r4,r4,ror#11
+	add	r11,r11,r2			@ h+=Ch(e,f,g)
+#if 16==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 16<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r4,r5			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#2*4]		@ from future BODY_16_xx
+	eor	r12,r4,r5			@ a^b, b^c in next round
+	ldr	r1,[sp,#15*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r4,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r7,r7,r11			@ d+=h
+	eor	r3,r3,r5			@ Maj(a,b,c)
+	add	r11,r11,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r11,r11,r3			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#2*4]		@ 17
+	@ ldr	r1,[sp,#15*4]
+	mov	r0,r2,ror#7
+	add	r11,r11,r3			@ h+=Maj(a,b,c) from the past
+	mov	r3,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r3,r3,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#1*4]
+	eor	r3,r3,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#10*4]
+
+	add	r3,r3,r0
+	eor	r0,r7,r7,ror#5	@ from BODY_00_15
+	add	r2,r2,r3
+	eor	r0,r0,r7,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r3,[r14],#4			@ *K256++
+	add	r10,r10,r2			@ h+=X[i]
+	str	r2,[sp,#1*4]
+	eor	r2,r8,r9
+	add	r10,r10,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r7
+	add	r10,r10,r3			@ h+=K256[i]
+	eor	r2,r2,r9			@ Ch(e,f,g)
+	eor	r0,r11,r11,ror#11
+	add	r10,r10,r2			@ h+=Ch(e,f,g)
+#if 17==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 17<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r11,r4			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#3*4]		@ from future BODY_16_xx
+	eor	r3,r11,r4			@ a^b, b^c in next round
+	ldr	r1,[sp,#0*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r11,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r6,r6,r10			@ d+=h
+	eor	r12,r12,r4			@ Maj(a,b,c)
+	add	r10,r10,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r10,r10,r12			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#3*4]		@ 18
+	@ ldr	r1,[sp,#0*4]
+	mov	r0,r2,ror#7
+	add	r10,r10,r12			@ h+=Maj(a,b,c) from the past
+	mov	r12,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r12,r12,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#2*4]
+	eor	r12,r12,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#11*4]
+
+	add	r12,r12,r0
+	eor	r0,r6,r6,ror#5	@ from BODY_00_15
+	add	r2,r2,r12
+	eor	r0,r0,r6,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r12,[r14],#4			@ *K256++
+	add	r9,r9,r2			@ h+=X[i]
+	str	r2,[sp,#2*4]
+	eor	r2,r7,r8
+	add	r9,r9,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r6
+	add	r9,r9,r12			@ h+=K256[i]
+	eor	r2,r2,r8			@ Ch(e,f,g)
+	eor	r0,r10,r10,ror#11
+	add	r9,r9,r2			@ h+=Ch(e,f,g)
+#if 18==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 18<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r10,r11			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#4*4]		@ from future BODY_16_xx
+	eor	r12,r10,r11			@ a^b, b^c in next round
+	ldr	r1,[sp,#1*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r10,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r5,r5,r9			@ d+=h
+	eor	r3,r3,r11			@ Maj(a,b,c)
+	add	r9,r9,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r9,r9,r3			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#4*4]		@ 19
+	@ ldr	r1,[sp,#1*4]
+	mov	r0,r2,ror#7
+	add	r9,r9,r3			@ h+=Maj(a,b,c) from the past
+	mov	r3,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r3,r3,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#3*4]
+	eor	r3,r3,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#12*4]
+
+	add	r3,r3,r0
+	eor	r0,r5,r5,ror#5	@ from BODY_00_15
+	add	r2,r2,r3
+	eor	r0,r0,r5,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r3,[r14],#4			@ *K256++
+	add	r8,r8,r2			@ h+=X[i]
+	str	r2,[sp,#3*4]
+	eor	r2,r6,r7
+	add	r8,r8,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r5
+	add	r8,r8,r3			@ h+=K256[i]
+	eor	r2,r2,r7			@ Ch(e,f,g)
+	eor	r0,r9,r9,ror#11
+	add	r8,r8,r2			@ h+=Ch(e,f,g)
+#if 19==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 19<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r9,r10			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#5*4]		@ from future BODY_16_xx
+	eor	r3,r9,r10			@ a^b, b^c in next round
+	ldr	r1,[sp,#2*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r9,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r4,r4,r8			@ d+=h
+	eor	r12,r12,r10			@ Maj(a,b,c)
+	add	r8,r8,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r8,r8,r12			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#5*4]		@ 20
+	@ ldr	r1,[sp,#2*4]
+	mov	r0,r2,ror#7
+	add	r8,r8,r12			@ h+=Maj(a,b,c) from the past
+	mov	r12,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r12,r12,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#4*4]
+	eor	r12,r12,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#13*4]
+
+	add	r12,r12,r0
+	eor	r0,r4,r4,ror#5	@ from BODY_00_15
+	add	r2,r2,r12
+	eor	r0,r0,r4,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r12,[r14],#4			@ *K256++
+	add	r7,r7,r2			@ h+=X[i]
+	str	r2,[sp,#4*4]
+	eor	r2,r5,r6
+	add	r7,r7,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r4
+	add	r7,r7,r12			@ h+=K256[i]
+	eor	r2,r2,r6			@ Ch(e,f,g)
+	eor	r0,r8,r8,ror#11
+	add	r7,r7,r2			@ h+=Ch(e,f,g)
+#if 20==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 20<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r8,r9			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#6*4]		@ from future BODY_16_xx
+	eor	r12,r8,r9			@ a^b, b^c in next round
+	ldr	r1,[sp,#3*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r8,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r11,r11,r7			@ d+=h
+	eor	r3,r3,r9			@ Maj(a,b,c)
+	add	r7,r7,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r7,r7,r3			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#6*4]		@ 21
+	@ ldr	r1,[sp,#3*4]
+	mov	r0,r2,ror#7
+	add	r7,r7,r3			@ h+=Maj(a,b,c) from the past
+	mov	r3,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r3,r3,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#5*4]
+	eor	r3,r3,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#14*4]
+
+	add	r3,r3,r0
+	eor	r0,r11,r11,ror#5	@ from BODY_00_15
+	add	r2,r2,r3
+	eor	r0,r0,r11,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r3,[r14],#4			@ *K256++
+	add	r6,r6,r2			@ h+=X[i]
+	str	r2,[sp,#5*4]
+	eor	r2,r4,r5
+	add	r6,r6,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r11
+	add	r6,r6,r3			@ h+=K256[i]
+	eor	r2,r2,r5			@ Ch(e,f,g)
+	eor	r0,r7,r7,ror#11
+	add	r6,r6,r2			@ h+=Ch(e,f,g)
+#if 21==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 21<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r7,r8			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#7*4]		@ from future BODY_16_xx
+	eor	r3,r7,r8			@ a^b, b^c in next round
+	ldr	r1,[sp,#4*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r7,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r10,r10,r6			@ d+=h
+	eor	r12,r12,r8			@ Maj(a,b,c)
+	add	r6,r6,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r6,r6,r12			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#7*4]		@ 22
+	@ ldr	r1,[sp,#4*4]
+	mov	r0,r2,ror#7
+	add	r6,r6,r12			@ h+=Maj(a,b,c) from the past
+	mov	r12,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r12,r12,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#6*4]
+	eor	r12,r12,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#15*4]
+
+	add	r12,r12,r0
+	eor	r0,r10,r10,ror#5	@ from BODY_00_15
+	add	r2,r2,r12
+	eor	r0,r0,r10,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r12,[r14],#4			@ *K256++
+	add	r5,r5,r2			@ h+=X[i]
+	str	r2,[sp,#6*4]
+	eor	r2,r11,r4
+	add	r5,r5,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r10
+	add	r5,r5,r12			@ h+=K256[i]
+	eor	r2,r2,r4			@ Ch(e,f,g)
+	eor	r0,r6,r6,ror#11
+	add	r5,r5,r2			@ h+=Ch(e,f,g)
+#if 22==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 22<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r6,r7			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#8*4]		@ from future BODY_16_xx
+	eor	r12,r6,r7			@ a^b, b^c in next round
+	ldr	r1,[sp,#5*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r6,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r9,r9,r5			@ d+=h
+	eor	r3,r3,r7			@ Maj(a,b,c)
+	add	r5,r5,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r5,r5,r3			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#8*4]		@ 23
+	@ ldr	r1,[sp,#5*4]
+	mov	r0,r2,ror#7
+	add	r5,r5,r3			@ h+=Maj(a,b,c) from the past
+	mov	r3,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r3,r3,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#7*4]
+	eor	r3,r3,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#0*4]
+
+	add	r3,r3,r0
+	eor	r0,r9,r9,ror#5	@ from BODY_00_15
+	add	r2,r2,r3
+	eor	r0,r0,r9,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r3,[r14],#4			@ *K256++
+	add	r4,r4,r2			@ h+=X[i]
+	str	r2,[sp,#7*4]
+	eor	r2,r10,r11
+	add	r4,r4,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r9
+	add	r4,r4,r3			@ h+=K256[i]
+	eor	r2,r2,r11			@ Ch(e,f,g)
+	eor	r0,r5,r5,ror#11
+	add	r4,r4,r2			@ h+=Ch(e,f,g)
+#if 23==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 23<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r5,r6			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#9*4]		@ from future BODY_16_xx
+	eor	r3,r5,r6			@ a^b, b^c in next round
+	ldr	r1,[sp,#6*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r5,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r8,r8,r4			@ d+=h
+	eor	r12,r12,r6			@ Maj(a,b,c)
+	add	r4,r4,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r4,r4,r12			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#9*4]		@ 24
+	@ ldr	r1,[sp,#6*4]
+	mov	r0,r2,ror#7
+	add	r4,r4,r12			@ h+=Maj(a,b,c) from the past
+	mov	r12,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r12,r12,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#8*4]
+	eor	r12,r12,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#1*4]
+
+	add	r12,r12,r0
+	eor	r0,r8,r8,ror#5	@ from BODY_00_15
+	add	r2,r2,r12
+	eor	r0,r0,r8,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r12,[r14],#4			@ *K256++
+	add	r11,r11,r2			@ h+=X[i]
+	str	r2,[sp,#8*4]
+	eor	r2,r9,r10
+	add	r11,r11,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r8
+	add	r11,r11,r12			@ h+=K256[i]
+	eor	r2,r2,r10			@ Ch(e,f,g)
+	eor	r0,r4,r4,ror#11
+	add	r11,r11,r2			@ h+=Ch(e,f,g)
+#if 24==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 24<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r4,r5			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#10*4]		@ from future BODY_16_xx
+	eor	r12,r4,r5			@ a^b, b^c in next round
+	ldr	r1,[sp,#7*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r4,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r7,r7,r11			@ d+=h
+	eor	r3,r3,r5			@ Maj(a,b,c)
+	add	r11,r11,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r11,r11,r3			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#10*4]		@ 25
+	@ ldr	r1,[sp,#7*4]
+	mov	r0,r2,ror#7
+	add	r11,r11,r3			@ h+=Maj(a,b,c) from the past
+	mov	r3,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r3,r3,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#9*4]
+	eor	r3,r3,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#2*4]
+
+	add	r3,r3,r0
+	eor	r0,r7,r7,ror#5	@ from BODY_00_15
+	add	r2,r2,r3
+	eor	r0,r0,r7,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r3,[r14],#4			@ *K256++
+	add	r10,r10,r2			@ h+=X[i]
+	str	r2,[sp,#9*4]
+	eor	r2,r8,r9
+	add	r10,r10,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r7
+	add	r10,r10,r3			@ h+=K256[i]
+	eor	r2,r2,r9			@ Ch(e,f,g)
+	eor	r0,r11,r11,ror#11
+	add	r10,r10,r2			@ h+=Ch(e,f,g)
+#if 25==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 25<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r11,r4			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#11*4]		@ from future BODY_16_xx
+	eor	r3,r11,r4			@ a^b, b^c in next round
+	ldr	r1,[sp,#8*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r11,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r6,r6,r10			@ d+=h
+	eor	r12,r12,r4			@ Maj(a,b,c)
+	add	r10,r10,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r10,r10,r12			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#11*4]		@ 26
+	@ ldr	r1,[sp,#8*4]
+	mov	r0,r2,ror#7
+	add	r10,r10,r12			@ h+=Maj(a,b,c) from the past
+	mov	r12,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r12,r12,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#10*4]
+	eor	r12,r12,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#3*4]
+
+	add	r12,r12,r0
+	eor	r0,r6,r6,ror#5	@ from BODY_00_15
+	add	r2,r2,r12
+	eor	r0,r0,r6,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r12,[r14],#4			@ *K256++
+	add	r9,r9,r2			@ h+=X[i]
+	str	r2,[sp,#10*4]
+	eor	r2,r7,r8
+	add	r9,r9,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r6
+	add	r9,r9,r12			@ h+=K256[i]
+	eor	r2,r2,r8			@ Ch(e,f,g)
+	eor	r0,r10,r10,ror#11
+	add	r9,r9,r2			@ h+=Ch(e,f,g)
+#if 26==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 26<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r10,r11			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#12*4]		@ from future BODY_16_xx
+	eor	r12,r10,r11			@ a^b, b^c in next round
+	ldr	r1,[sp,#9*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r10,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r5,r5,r9			@ d+=h
+	eor	r3,r3,r11			@ Maj(a,b,c)
+	add	r9,r9,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r9,r9,r3			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#12*4]		@ 27
+	@ ldr	r1,[sp,#9*4]
+	mov	r0,r2,ror#7
+	add	r9,r9,r3			@ h+=Maj(a,b,c) from the past
+	mov	r3,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r3,r3,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#11*4]
+	eor	r3,r3,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#4*4]
+
+	add	r3,r3,r0
+	eor	r0,r5,r5,ror#5	@ from BODY_00_15
+	add	r2,r2,r3
+	eor	r0,r0,r5,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r3,[r14],#4			@ *K256++
+	add	r8,r8,r2			@ h+=X[i]
+	str	r2,[sp,#11*4]
+	eor	r2,r6,r7
+	add	r8,r8,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r5
+	add	r8,r8,r3			@ h+=K256[i]
+	eor	r2,r2,r7			@ Ch(e,f,g)
+	eor	r0,r9,r9,ror#11
+	add	r8,r8,r2			@ h+=Ch(e,f,g)
+#if 27==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 27<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r9,r10			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#13*4]		@ from future BODY_16_xx
+	eor	r3,r9,r10			@ a^b, b^c in next round
+	ldr	r1,[sp,#10*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r9,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r4,r4,r8			@ d+=h
+	eor	r12,r12,r10			@ Maj(a,b,c)
+	add	r8,r8,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r8,r8,r12			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#13*4]		@ 28
+	@ ldr	r1,[sp,#10*4]
+	mov	r0,r2,ror#7
+	add	r8,r8,r12			@ h+=Maj(a,b,c) from the past
+	mov	r12,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r12,r12,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#12*4]
+	eor	r12,r12,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#5*4]
+
+	add	r12,r12,r0
+	eor	r0,r4,r4,ror#5	@ from BODY_00_15
+	add	r2,r2,r12
+	eor	r0,r0,r4,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r12,[r14],#4			@ *K256++
+	add	r7,r7,r2			@ h+=X[i]
+	str	r2,[sp,#12*4]
+	eor	r2,r5,r6
+	add	r7,r7,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r4
+	add	r7,r7,r12			@ h+=K256[i]
+	eor	r2,r2,r6			@ Ch(e,f,g)
+	eor	r0,r8,r8,ror#11
+	add	r7,r7,r2			@ h+=Ch(e,f,g)
+#if 28==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 28<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r8,r9			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#14*4]		@ from future BODY_16_xx
+	eor	r12,r8,r9			@ a^b, b^c in next round
+	ldr	r1,[sp,#11*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r8,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r11,r11,r7			@ d+=h
+	eor	r3,r3,r9			@ Maj(a,b,c)
+	add	r7,r7,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r7,r7,r3			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#14*4]		@ 29
+	@ ldr	r1,[sp,#11*4]
+	mov	r0,r2,ror#7
+	add	r7,r7,r3			@ h+=Maj(a,b,c) from the past
+	mov	r3,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r3,r3,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#13*4]
+	eor	r3,r3,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#6*4]
+
+	add	r3,r3,r0
+	eor	r0,r11,r11,ror#5	@ from BODY_00_15
+	add	r2,r2,r3
+	eor	r0,r0,r11,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r3,[r14],#4			@ *K256++
+	add	r6,r6,r2			@ h+=X[i]
+	str	r2,[sp,#13*4]
+	eor	r2,r4,r5
+	add	r6,r6,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r11
+	add	r6,r6,r3			@ h+=K256[i]
+	eor	r2,r2,r5			@ Ch(e,f,g)
+	eor	r0,r7,r7,ror#11
+	add	r6,r6,r2			@ h+=Ch(e,f,g)
+#if 29==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 29<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r7,r8			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#15*4]		@ from future BODY_16_xx
+	eor	r3,r7,r8			@ a^b, b^c in next round
+	ldr	r1,[sp,#12*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r7,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r10,r10,r6			@ d+=h
+	eor	r12,r12,r8			@ Maj(a,b,c)
+	add	r6,r6,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r6,r6,r12			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#15*4]		@ 30
+	@ ldr	r1,[sp,#12*4]
+	mov	r0,r2,ror#7
+	add	r6,r6,r12			@ h+=Maj(a,b,c) from the past
+	mov	r12,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r12,r12,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#14*4]
+	eor	r12,r12,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#7*4]
+
+	add	r12,r12,r0
+	eor	r0,r10,r10,ror#5	@ from BODY_00_15
+	add	r2,r2,r12
+	eor	r0,r0,r10,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r12,[r14],#4			@ *K256++
+	add	r5,r5,r2			@ h+=X[i]
+	str	r2,[sp,#14*4]
+	eor	r2,r11,r4
+	add	r5,r5,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r10
+	add	r5,r5,r12			@ h+=K256[i]
+	eor	r2,r2,r4			@ Ch(e,f,g)
+	eor	r0,r6,r6,ror#11
+	add	r5,r5,r2			@ h+=Ch(e,f,g)
+#if 30==31
+	and	r12,r12,#0xff
+	cmp	r12,#0xf2			@ done?
+#endif
+#if 30<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r12,r6,r7			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#0*4]		@ from future BODY_16_xx
+	eor	r12,r6,r7			@ a^b, b^c in next round
+	ldr	r1,[sp,#13*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r6,ror#20	@ Sigma0(a)
+	and	r3,r3,r12			@ (b^c)&=(a^b)
+	add	r9,r9,r5			@ d+=h
+	eor	r3,r3,r7			@ Maj(a,b,c)
+	add	r5,r5,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r5,r5,r3			@ h+=Maj(a,b,c)
+	@ ldr	r2,[sp,#0*4]		@ 31
+	@ ldr	r1,[sp,#13*4]
+	mov	r0,r2,ror#7
+	add	r5,r5,r3			@ h+=Maj(a,b,c) from the past
+	mov	r3,r1,ror#17
+	eor	r0,r0,r2,ror#18
+	eor	r3,r3,r1,ror#19
+	eor	r0,r0,r2,lsr#3	@ sigma0(X[i+1])
+	ldr	r2,[sp,#15*4]
+	eor	r3,r3,r1,lsr#10	@ sigma1(X[i+14])
+	ldr	r1,[sp,#8*4]
+
+	add	r3,r3,r0
+	eor	r0,r9,r9,ror#5	@ from BODY_00_15
+	add	r2,r2,r3
+	eor	r0,r0,r9,ror#19	@ Sigma1(e)
+	add	r2,r2,r1			@ X[i]
+	ldr	r3,[r14],#4			@ *K256++
+	add	r4,r4,r2			@ h+=X[i]
+	str	r2,[sp,#15*4]
+	eor	r2,r10,r11
+	add	r4,r4,r0,ror#6	@ h+=Sigma1(e)
+	and	r2,r2,r9
+	add	r4,r4,r3			@ h+=K256[i]
+	eor	r2,r2,r11			@ Ch(e,f,g)
+	eor	r0,r5,r5,ror#11
+	add	r4,r4,r2			@ h+=Ch(e,f,g)
+#if 31==31
+	and	r3,r3,#0xff
+	cmp	r3,#0xf2			@ done?
+#endif
+#if 31<15
+# if __ARM_ARCH__>=7
+	ldr	r2,[r1],#4			@ prefetch
+# else
+	ldrb	r2,[r1,#3]
+# endif
+	eor	r3,r5,r6			@ a^b, b^c in next round
+#else
+	ldr	r2,[sp,#1*4]		@ from future BODY_16_xx
+	eor	r3,r5,r6			@ a^b, b^c in next round
+	ldr	r1,[sp,#14*4]	@ from future BODY_16_xx
+#endif
+	eor	r0,r0,r5,ror#20	@ Sigma0(a)
+	and	r12,r12,r3			@ (b^c)&=(a^b)
+	add	r8,r8,r4			@ d+=h
+	eor	r12,r12,r6			@ Maj(a,b,c)
+	add	r4,r4,r0,ror#2	@ h+=Sigma0(a)
+	@ add	r4,r4,r12			@ h+=Maj(a,b,c)
+#if __ARM_ARCH__>=7
+	ite	eq			@ Thumb2 thing, sanity check in ARM
+#endif
+	ldreq	r3,[sp,#16*4]		@ pull ctx
+	bne	.Lrounds_16_xx
+
+	add	r4,r4,r12		@ h+=Maj(a,b,c) from the past
+	ldr	r0,[r3,#0]
+	ldr	r2,[r3,#4]
+	ldr	r12,[r3,#8]
+	add	r4,r4,r0
+	ldr	r0,[r3,#12]
+	add	r5,r5,r2
+	ldr	r2,[r3,#16]
+	add	r6,r6,r12
+	ldr	r12,[r3,#20]
+	add	r7,r7,r0
+	ldr	r0,[r3,#24]
+	add	r8,r8,r2
+	ldr	r2,[r3,#28]
+	add	r9,r9,r12
+	ldr	r1,[sp,#17*4]		@ pull inp
+	ldr	r12,[sp,#18*4]		@ pull inp+len
+	add	r10,r10,r0
+	add	r11,r11,r2
+	stmia	r3,{r4,r5,r6,r7,r8,r9,r10,r11}
+	cmp	r1,r12
+	sub	r14,r14,#256	@ rewind Ktbl
+	bne	.Loop
+
+	add	sp,sp,#19*4	@ destroy frame
+#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
+#if __ARM_MAX_ARCH__>=7
+.arch	armv7-a
+.fpu	neon
+
+.global	sha256_block_data_order_neon
+.type	sha256_block_data_order_neon,%function
+.align	4
+sha256_block_data_order_neon:
+.LNEON:
+	stmdb	sp!,{r4-r12,lr}
+
+	sub	r11,sp,#16*4+16
+	adrl	r14,K256
+	bic	r11,r11,#15		@ align for 128-bit stores
+	mov	r12,sp
+	mov	sp,r11			@ alloca
+	add	r2,r1,r2,lsl#6	@ len to point at the end of inp
+
+	vld1.8		{q0},[r1]!
+	vld1.8		{q1},[r1]!
+	vld1.8		{q2},[r1]!
+	vld1.8		{q3},[r1]!
+	vld1.32		{q8},[r14,:128]!
+	vld1.32		{q9},[r14,:128]!
+	vld1.32		{q10},[r14,:128]!
+	vld1.32		{q11},[r14,:128]!
+	vrev32.8	q0,q0		@ yes, even on
+	str		r0,[sp,#64]
+	vrev32.8	q1,q1		@ big-endian
+	str		r1,[sp,#68]
+	mov		r1,sp
+	vrev32.8	q2,q2
+	str		r2,[sp,#72]
+	vrev32.8	q3,q3
+	str		r12,[sp,#76]		@ save original sp
+	vadd.i32	q8,q8,q0
+	vadd.i32	q9,q9,q1
+	vst1.32		{q8},[r1,:128]!
+	vadd.i32	q10,q10,q2
+	vst1.32		{q9},[r1,:128]!
+	vadd.i32	q11,q11,q3
+	vst1.32		{q10},[r1,:128]!
+	vst1.32		{q11},[r1,:128]!
+
+	ldmia		r0,{r4-r11}
+	sub		r1,r1,#64
+	ldr		r2,[sp,#0]
+	eor		r12,r12,r12
+	eor		r3,r5,r6
+	b		.L_00_48
+
+.align	4
+.L_00_48:
+	vext.8	q8,q0,q1,#4
+	add	r11,r11,r2
+	eor	r2,r9,r10
+	eor	r0,r8,r8,ror#5
+	vext.8	q9,q2,q3,#4
+	add	r4,r4,r12
+	and	r2,r2,r8
+	eor	r12,r0,r8,ror#19
+	vshr.u32	q10,q8,#7
+	eor	r0,r4,r4,ror#11
+	eor	r2,r2,r10
+	vadd.i32	q0,q0,q9
+	add	r11,r11,r12,ror#6
+	eor	r12,r4,r5
+	vshr.u32	q9,q8,#3
+	eor	r0,r0,r4,ror#20
+	add	r11,r11,r2
+	vsli.32	q10,q8,#25
+	ldr	r2,[sp,#4]
+	and	r3,r3,r12
+	vshr.u32	q11,q8,#18
+	add	r7,r7,r11
+	add	r11,r11,r0,ror#2
+	eor	r3,r3,r5
+	veor	q9,q9,q10
+	add	r10,r10,r2
+	vsli.32	q11,q8,#14
+	eor	r2,r8,r9
+	eor	r0,r7,r7,ror#5
+	vshr.u32	d24,d7,#17
+	add	r11,r11,r3
+	and	r2,r2,r7
+	veor	q9,q9,q11
+	eor	r3,r0,r7,ror#19
+	eor	r0,r11,r11,ror#11
+	vsli.32	d24,d7,#15
+	eor	r2,r2,r9
+	add	r10,r10,r3,ror#6
+	vshr.u32	d25,d7,#10
+	eor	r3,r11,r4
+	eor	r0,r0,r11,ror#20
+	vadd.i32	q0,q0,q9
+	add	r10,r10,r2
+	ldr	r2,[sp,#8]
+	veor	d25,d25,d24
+	and	r12,r12,r3
+	add	r6,r6,r10
+	vshr.u32	d24,d7,#19
+	add	r10,r10,r0,ror#2
+	eor	r12,r12,r4
+	vsli.32	d24,d7,#13
+	add	r9,r9,r2
+	eor	r2,r7,r8
+	veor	d25,d25,d24
+	eor	r0,r6,r6,ror#5
+	add	r10,r10,r12
+	vadd.i32	d0,d0,d25
+	and	r2,r2,r6
+	eor	r12,r0,r6,ror#19
+	vshr.u32	d24,d0,#17
+	eor	r0,r10,r10,ror#11
+	eor	r2,r2,r8
+	vsli.32	d24,d0,#15
+	add	r9,r9,r12,ror#6
+	eor	r12,r10,r11
+	vshr.u32	d25,d0,#10
+	eor	r0,r0,r10,ror#20
+	add	r9,r9,r2
+	veor	d25,d25,d24
+	ldr	r2,[sp,#12]
+	and	r3,r3,r12
+	vshr.u32	d24,d0,#19
+	add	r5,r5,r9
+	add	r9,r9,r0,ror#2
+	eor	r3,r3,r11
+	vld1.32	{q8},[r14,:128]!
+	add	r8,r8,r2
+	vsli.32	d24,d0,#13
+	eor	r2,r6,r7
+	eor	r0,r5,r5,ror#5
+	veor	d25,d25,d24
+	add	r9,r9,r3
+	and	r2,r2,r5
+	vadd.i32	d1,d1,d25
+	eor	r3,r0,r5,ror#19
+	eor	r0,r9,r9,ror#11
+	vadd.i32	q8,q8,q0
+	eor	r2,r2,r7
+	add	r8,r8,r3,ror#6
+	eor	r3,r9,r10
+	eor	r0,r0,r9,ror#20
+	add	r8,r8,r2
+	ldr	r2,[sp,#16]
+	and	r12,r12,r3
+	add	r4,r4,r8
+	vst1.32	{q8},[r1,:128]!
+	add	r8,r8,r0,ror#2
+	eor	r12,r12,r10
+	vext.8	q8,q1,q2,#4
+	add	r7,r7,r2
+	eor	r2,r5,r6
+	eor	r0,r4,r4,ror#5
+	vext.8	q9,q3,q0,#4
+	add	r8,r8,r12
+	and	r2,r2,r4
+	eor	r12,r0,r4,ror#19
+	vshr.u32	q10,q8,#7
+	eor	r0,r8,r8,ror#11
+	eor	r2,r2,r6
+	vadd.i32	q1,q1,q9
+	add	r7,r7,r12,ror#6
+	eor	r12,r8,r9
+	vshr.u32	q9,q8,#3
+	eor	r0,r0,r8,ror#20
+	add	r7,r7,r2
+	vsli.32	q10,q8,#25
+	ldr	r2,[sp,#20]
+	and	r3,r3,r12
+	vshr.u32	q11,q8,#18
+	add	r11,r11,r7
+	add	r7,r7,r0,ror#2
+	eor	r3,r3,r9
+	veor	q9,q9,q10
+	add	r6,r6,r2
+	vsli.32	q11,q8,#14
+	eor	r2,r4,r5
+	eor	r0,r11,r11,ror#5
+	vshr.u32	d24,d1,#17
+	add	r7,r7,r3
+	and	r2,r2,r11
+	veor	q9,q9,q11
+	eor	r3,r0,r11,ror#19
+	eor	r0,r7,r7,ror#11
+	vsli.32	d24,d1,#15
+	eor	r2,r2,r5
+	add	r6,r6,r3,ror#6
+	vshr.u32	d25,d1,#10
+	eor	r3,r7,r8
+	eor	r0,r0,r7,ror#20
+	vadd.i32	q1,q1,q9
+	add	r6,r6,r2
+	ldr	r2,[sp,#24]
+	veor	d25,d25,d24
+	and	r12,r12,r3
+	add	r10,r10,r6
+	vshr.u32	d24,d1,#19
+	add	r6,r6,r0,ror#2
+	eor	r12,r12,r8
+	vsli.32	d24,d1,#13
+	add	r5,r5,r2
+	eor	r2,r11,r4
+	veor	d25,d25,d24
+	eor	r0,r10,r10,ror#5
+	add	r6,r6,r12
+	vadd.i32	d2,d2,d25
+	and	r2,r2,r10
+	eor	r12,r0,r10,ror#19
+	vshr.u32	d24,d2,#17
+	eor	r0,r6,r6,ror#11
+	eor	r2,r2,r4
+	vsli.32	d24,d2,#15
+	add	r5,r5,r12,ror#6
+	eor	r12,r6,r7
+	vshr.u32	d25,d2,#10
+	eor	r0,r0,r6,ror#20
+	add	r5,r5,r2
+	veor	d25,d25,d24
+	ldr	r2,[sp,#28]
+	and	r3,r3,r12
+	vshr.u32	d24,d2,#19
+	add	r9,r9,r5
+	add	r5,r5,r0,ror#2
+	eor	r3,r3,r7
+	vld1.32	{q8},[r14,:128]!
+	add	r4,r4,r2
+	vsli.32	d24,d2,#13
+	eor	r2,r10,r11
+	eor	r0,r9,r9,ror#5
+	veor	d25,d25,d24
+	add	r5,r5,r3
+	and	r2,r2,r9
+	vadd.i32	d3,d3,d25
+	eor	r3,r0,r9,ror#19
+	eor	r0,r5,r5,ror#11
+	vadd.i32	q8,q8,q1
+	eor	r2,r2,r11
+	add	r4,r4,r3,ror#6
+	eor	r3,r5,r6
+	eor	r0,r0,r5,ror#20
+	add	r4,r4,r2
+	ldr	r2,[sp,#32]
+	and	r12,r12,r3
+	add	r8,r8,r4
+	vst1.32	{q8},[r1,:128]!
+	add	r4,r4,r0,ror#2
+	eor	r12,r12,r6
+	vext.8	q8,q2,q3,#4
+	add	r11,r11,r2
+	eor	r2,r9,r10
+	eor	r0,r8,r8,ror#5
+	vext.8	q9,q0,q1,#4
+	add	r4,r4,r12
+	and	r2,r2,r8
+	eor	r12,r0,r8,ror#19
+	vshr.u32	q10,q8,#7
+	eor	r0,r4,r4,ror#11
+	eor	r2,r2,r10
+	vadd.i32	q2,q2,q9
+	add	r11,r11,r12,ror#6
+	eor	r12,r4,r5
+	vshr.u32	q9,q8,#3
+	eor	r0,r0,r4,ror#20
+	add	r11,r11,r2
+	vsli.32	q10,q8,#25
+	ldr	r2,[sp,#36]
+	and	r3,r3,r12
+	vshr.u32	q11,q8,#18
+	add	r7,r7,r11
+	add	r11,r11,r0,ror#2
+	eor	r3,r3,r5
+	veor	q9,q9,q10
+	add	r10,r10,r2
+	vsli.32	q11,q8,#14
+	eor	r2,r8,r9
+	eor	r0,r7,r7,ror#5
+	vshr.u32	d24,d3,#17
+	add	r11,r11,r3
+	and	r2,r2,r7
+	veor	q9,q9,q11
+	eor	r3,r0,r7,ror#19
+	eor	r0,r11,r11,ror#11
+	vsli.32	d24,d3,#15
+	eor	r2,r2,r9
+	add	r10,r10,r3,ror#6
+	vshr.u32	d25,d3,#10
+	eor	r3,r11,r4
+	eor	r0,r0,r11,ror#20
+	vadd.i32	q2,q2,q9
+	add	r10,r10,r2
+	ldr	r2,[sp,#40]
+	veor	d25,d25,d24
+	and	r12,r12,r3
+	add	r6,r6,r10
+	vshr.u32	d24,d3,#19
+	add	r10,r10,r0,ror#2
+	eor	r12,r12,r4
+	vsli.32	d24,d3,#13
+	add	r9,r9,r2
+	eor	r2,r7,r8
+	veor	d25,d25,d24
+	eor	r0,r6,r6,ror#5
+	add	r10,r10,r12
+	vadd.i32	d4,d4,d25
+	and	r2,r2,r6
+	eor	r12,r0,r6,ror#19
+	vshr.u32	d24,d4,#17
+	eor	r0,r10,r10,ror#11
+	eor	r2,r2,r8
+	vsli.32	d24,d4,#15
+	add	r9,r9,r12,ror#6
+	eor	r12,r10,r11
+	vshr.u32	d25,d4,#10
+	eor	r0,r0,r10,ror#20
+	add	r9,r9,r2
+	veor	d25,d25,d24
+	ldr	r2,[sp,#44]
+	and	r3,r3,r12
+	vshr.u32	d24,d4,#19
+	add	r5,r5,r9
+	add	r9,r9,r0,ror#2
+	eor	r3,r3,r11
+	vld1.32	{q8},[r14,:128]!
+	add	r8,r8,r2
+	vsli.32	d24,d4,#13
+	eor	r2,r6,r7
+	eor	r0,r5,r5,ror#5
+	veor	d25,d25,d24
+	add	r9,r9,r3
+	and	r2,r2,r5
+	vadd.i32	d5,d5,d25
+	eor	r3,r0,r5,ror#19
+	eor	r0,r9,r9,ror#11
+	vadd.i32	q8,q8,q2
+	eor	r2,r2,r7
+	add	r8,r8,r3,ror#6
+	eor	r3,r9,r10
+	eor	r0,r0,r9,ror#20
+	add	r8,r8,r2
+	ldr	r2,[sp,#48]
+	and	r12,r12,r3
+	add	r4,r4,r8
+	vst1.32	{q8},[r1,:128]!
+	add	r8,r8,r0,ror#2
+	eor	r12,r12,r10
+	vext.8	q8,q3,q0,#4
+	add	r7,r7,r2
+	eor	r2,r5,r6
+	eor	r0,r4,r4,ror#5
+	vext.8	q9,q1,q2,#4
+	add	r8,r8,r12
+	and	r2,r2,r4
+	eor	r12,r0,r4,ror#19
+	vshr.u32	q10,q8,#7
+	eor	r0,r8,r8,ror#11
+	eor	r2,r2,r6
+	vadd.i32	q3,q3,q9
+	add	r7,r7,r12,ror#6
+	eor	r12,r8,r9
+	vshr.u32	q9,q8,#3
+	eor	r0,r0,r8,ror#20
+	add	r7,r7,r2
+	vsli.32	q10,q8,#25
+	ldr	r2,[sp,#52]
+	and	r3,r3,r12
+	vshr.u32	q11,q8,#18
+	add	r11,r11,r7
+	add	r7,r7,r0,ror#2
+	eor	r3,r3,r9
+	veor	q9,q9,q10
+	add	r6,r6,r2
+	vsli.32	q11,q8,#14
+	eor	r2,r4,r5
+	eor	r0,r11,r11,ror#5
+	vshr.u32	d24,d5,#17
+	add	r7,r7,r3
+	and	r2,r2,r11
+	veor	q9,q9,q11
+	eor	r3,r0,r11,ror#19
+	eor	r0,r7,r7,ror#11
+	vsli.32	d24,d5,#15
+	eor	r2,r2,r5
+	add	r6,r6,r3,ror#6
+	vshr.u32	d25,d5,#10
+	eor	r3,r7,r8
+	eor	r0,r0,r7,ror#20
+	vadd.i32	q3,q3,q9
+	add	r6,r6,r2
+	ldr	r2,[sp,#56]
+	veor	d25,d25,d24
+	and	r12,r12,r3
+	add	r10,r10,r6
+	vshr.u32	d24,d5,#19
+	add	r6,r6,r0,ror#2
+	eor	r12,r12,r8
+	vsli.32	d24,d5,#13
+	add	r5,r5,r2
+	eor	r2,r11,r4
+	veor	d25,d25,d24
+	eor	r0,r10,r10,ror#5
+	add	r6,r6,r12
+	vadd.i32	d6,d6,d25
+	and	r2,r2,r10
+	eor	r12,r0,r10,ror#19
+	vshr.u32	d24,d6,#17
+	eor	r0,r6,r6,ror#11
+	eor	r2,r2,r4
+	vsli.32	d24,d6,#15
+	add	r5,r5,r12,ror#6
+	eor	r12,r6,r7
+	vshr.u32	d25,d6,#10
+	eor	r0,r0,r6,ror#20
+	add	r5,r5,r2
+	veor	d25,d25,d24
+	ldr	r2,[sp,#60]
+	and	r3,r3,r12
+	vshr.u32	d24,d6,#19
+	add	r9,r9,r5
+	add	r5,r5,r0,ror#2
+	eor	r3,r3,r7
+	vld1.32	{q8},[r14,:128]!
+	add	r4,r4,r2
+	vsli.32	d24,d6,#13
+	eor	r2,r10,r11
+	eor	r0,r9,r9,ror#5
+	veor	d25,d25,d24
+	add	r5,r5,r3
+	and	r2,r2,r9
+	vadd.i32	d7,d7,d25
+	eor	r3,r0,r9,ror#19
+	eor	r0,r5,r5,ror#11
+	vadd.i32	q8,q8,q3
+	eor	r2,r2,r11
+	add	r4,r4,r3,ror#6
+	eor	r3,r5,r6
+	eor	r0,r0,r5,ror#20
+	add	r4,r4,r2
+	ldr	r2,[r14]
+	and	r12,r12,r3
+	add	r8,r8,r4
+	vst1.32	{q8},[r1,:128]!
+	add	r4,r4,r0,ror#2
+	eor	r12,r12,r6
+	teq	r2,#0				@ check for K256 terminator
+	ldr	r2,[sp,#0]
+	sub	r1,r1,#64
+	bne	.L_00_48
+
+	ldr		r1,[sp,#68]
+	ldr		r0,[sp,#72]
+	sub		r14,r14,#256	@ rewind r14
+	teq		r1,r0
+	it		eq
+	subeq		r1,r1,#64		@ avoid SEGV
+	vld1.8		{q0},[r1]!		@ load next input block
+	vld1.8		{q1},[r1]!
+	vld1.8		{q2},[r1]!
+	vld1.8		{q3},[r1]!
+	it		ne
+	strne		r1,[sp,#68]
+	mov		r1,sp
+	add	r11,r11,r2
+	eor	r2,r9,r10
+	eor	r0,r8,r8,ror#5
+	add	r4,r4,r12
+	vld1.32	{q8},[r14,:128]!
+	and	r2,r2,r8
+	eor	r12,r0,r8,ror#19
+	eor	r0,r4,r4,ror#11
+	eor	r2,r2,r10
+	vrev32.8	q0,q0
+	add	r11,r11,r12,ror#6
+	eor	r12,r4,r5
+	eor	r0,r0,r4,ror#20
+	add	r11,r11,r2
+	vadd.i32	q8,q8,q0
+	ldr	r2,[sp,#4]
+	and	r3,r3,r12
+	add	r7,r7,r11
+	add	r11,r11,r0,ror#2
+	eor	r3,r3,r5
+	add	r10,r10,r2
+	eor	r2,r8,r9
+	eor	r0,r7,r7,ror#5
+	add	r11,r11,r3
+	and	r2,r2,r7
+	eor	r3,r0,r7,ror#19
+	eor	r0,r11,r11,ror#11
+	eor	r2,r2,r9
+	add	r10,r10,r3,ror#6
+	eor	r3,r11,r4
+	eor	r0,r0,r11,ror#20
+	add	r10,r10,r2
+	ldr	r2,[sp,#8]
+	and	r12,r12,r3
+	add	r6,r6,r10
+	add	r10,r10,r0,ror#2
+	eor	r12,r12,r4
+	add	r9,r9,r2
+	eor	r2,r7,r8
+	eor	r0,r6,r6,ror#5
+	add	r10,r10,r12
+	and	r2,r2,r6
+	eor	r12,r0,r6,ror#19
+	eor	r0,r10,r10,ror#11
+	eor	r2,r2,r8
+	add	r9,r9,r12,ror#6
+	eor	r12,r10,r11
+	eor	r0,r0,r10,ror#20
+	add	r9,r9,r2
+	ldr	r2,[sp,#12]
+	and	r3,r3,r12
+	add	r5,r5,r9
+	add	r9,r9,r0,ror#2
+	eor	r3,r3,r11
+	add	r8,r8,r2
+	eor	r2,r6,r7
+	eor	r0,r5,r5,ror#5
+	add	r9,r9,r3
+	and	r2,r2,r5
+	eor	r3,r0,r5,ror#19
+	eor	r0,r9,r9,ror#11
+	eor	r2,r2,r7
+	add	r8,r8,r3,ror#6
+	eor	r3,r9,r10
+	eor	r0,r0,r9,ror#20
+	add	r8,r8,r2
+	ldr	r2,[sp,#16]
+	and	r12,r12,r3
+	add	r4,r4,r8
+	add	r8,r8,r0,ror#2
+	eor	r12,r12,r10
+	vst1.32	{q8},[r1,:128]!
+	add	r7,r7,r2
+	eor	r2,r5,r6
+	eor	r0,r4,r4,ror#5
+	add	r8,r8,r12
+	vld1.32	{q8},[r14,:128]!
+	and	r2,r2,r4
+	eor	r12,r0,r4,ror#19
+	eor	r0,r8,r8,ror#11
+	eor	r2,r2,r6
+	vrev32.8	q1,q1
+	add	r7,r7,r12,ror#6
+	eor	r12,r8,r9
+	eor	r0,r0,r8,ror#20
+	add	r7,r7,r2
+	vadd.i32	q8,q8,q1
+	ldr	r2,[sp,#20]
+	and	r3,r3,r12
+	add	r11,r11,r7
+	add	r7,r7,r0,ror#2
+	eor	r3,r3,r9
+	add	r6,r6,r2
+	eor	r2,r4,r5
+	eor	r0,r11,r11,ror#5
+	add	r7,r7,r3
+	and	r2,r2,r11
+	eor	r3,r0,r11,ror#19
+	eor	r0,r7,r7,ror#11
+	eor	r2,r2,r5
+	add	r6,r6,r3,ror#6
+	eor	r3,r7,r8
+	eor	r0,r0,r7,ror#20
+	add	r6,r6,r2
+	ldr	r2,[sp,#24]
+	and	r12,r12,r3
+	add	r10,r10,r6
+	add	r6,r6,r0,ror#2
+	eor	r12,r12,r8
+	add	r5,r5,r2
+	eor	r2,r11,r4
+	eor	r0,r10,r10,ror#5
+	add	r6,r6,r12
+	and	r2,r2,r10
+	eor	r12,r0,r10,ror#19
+	eor	r0,r6,r6,ror#11
+	eor	r2,r2,r4
+	add	r5,r5,r12,ror#6
+	eor	r12,r6,r7
+	eor	r0,r0,r6,ror#20
+	add	r5,r5,r2
+	ldr	r2,[sp,#28]
+	and	r3,r3,r12
+	add	r9,r9,r5
+	add	r5,r5,r0,ror#2
+	eor	r3,r3,r7
+	add	r4,r4,r2
+	eor	r2,r10,r11
+	eor	r0,r9,r9,ror#5
+	add	r5,r5,r3
+	and	r2,r2,r9
+	eor	r3,r0,r9,ror#19
+	eor	r0,r5,r5,ror#11
+	eor	r2,r2,r11
+	add	r4,r4,r3,ror#6
+	eor	r3,r5,r6
+	eor	r0,r0,r5,ror#20
+	add	r4,r4,r2
+	ldr	r2,[sp,#32]
+	and	r12,r12,r3
+	add	r8,r8,r4
+	add	r4,r4,r0,ror#2
+	eor	r12,r12,r6
+	vst1.32	{q8},[r1,:128]!
+	add	r11,r11,r2
+	eor	r2,r9,r10
+	eor	r0,r8,r8,ror#5
+	add	r4,r4,r12
+	vld1.32	{q8},[r14,:128]!
+	and	r2,r2,r8
+	eor	r12,r0,r8,ror#19
+	eor	r0,r4,r4,ror#11
+	eor	r2,r2,r10
+	vrev32.8	q2,q2
+	add	r11,r11,r12,ror#6
+	eor	r12,r4,r5
+	eor	r0,r0,r4,ror#20
+	add	r11,r11,r2
+	vadd.i32	q8,q8,q2
+	ldr	r2,[sp,#36]
+	and	r3,r3,r12
+	add	r7,r7,r11
+	add	r11,r11,r0,ror#2
+	eor	r3,r3,r5
+	add	r10,r10,r2
+	eor	r2,r8,r9
+	eor	r0,r7,r7,ror#5
+	add	r11,r11,r3
+	and	r2,r2,r7
+	eor	r3,r0,r7,ror#19
+	eor	r0,r11,r11,ror#11
+	eor	r2,r2,r9
+	add	r10,r10,r3,ror#6
+	eor	r3,r11,r4
+	eor	r0,r0,r11,ror#20
+	add	r10,r10,r2
+	ldr	r2,[sp,#40]
+	and	r12,r12,r3
+	add	r6,r6,r10
+	add	r10,r10,r0,ror#2
+	eor	r12,r12,r4
+	add	r9,r9,r2
+	eor	r2,r7,r8
+	eor	r0,r6,r6,ror#5
+	add	r10,r10,r12
+	and	r2,r2,r6
+	eor	r12,r0,r6,ror#19
+	eor	r0,r10,r10,ror#11
+	eor	r2,r2,r8
+	add	r9,r9,r12,ror#6
+	eor	r12,r10,r11
+	eor	r0,r0,r10,ror#20
+	add	r9,r9,r2
+	ldr	r2,[sp,#44]
+	and	r3,r3,r12
+	add	r5,r5,r9
+	add	r9,r9,r0,ror#2
+	eor	r3,r3,r11
+	add	r8,r8,r2
+	eor	r2,r6,r7
+	eor	r0,r5,r5,ror#5
+	add	r9,r9,r3
+	and	r2,r2,r5
+	eor	r3,r0,r5,ror#19
+	eor	r0,r9,r9,ror#11
+	eor	r2,r2,r7
+	add	r8,r8,r3,ror#6
+	eor	r3,r9,r10
+	eor	r0,r0,r9,ror#20
+	add	r8,r8,r2
+	ldr	r2,[sp,#48]
+	and	r12,r12,r3
+	add	r4,r4,r8
+	add	r8,r8,r0,ror#2
+	eor	r12,r12,r10
+	vst1.32	{q8},[r1,:128]!
+	add	r7,r7,r2
+	eor	r2,r5,r6
+	eor	r0,r4,r4,ror#5
+	add	r8,r8,r12
+	vld1.32	{q8},[r14,:128]!
+	and	r2,r2,r4
+	eor	r12,r0,r4,ror#19
+	eor	r0,r8,r8,ror#11
+	eor	r2,r2,r6
+	vrev32.8	q3,q3
+	add	r7,r7,r12,ror#6
+	eor	r12,r8,r9
+	eor	r0,r0,r8,ror#20
+	add	r7,r7,r2
+	vadd.i32	q8,q8,q3
+	ldr	r2,[sp,#52]
+	and	r3,r3,r12
+	add	r11,r11,r7
+	add	r7,r7,r0,ror#2
+	eor	r3,r3,r9
+	add	r6,r6,r2
+	eor	r2,r4,r5
+	eor	r0,r11,r11,ror#5
+	add	r7,r7,r3
+	and	r2,r2,r11
+	eor	r3,r0,r11,ror#19
+	eor	r0,r7,r7,ror#11
+	eor	r2,r2,r5
+	add	r6,r6,r3,ror#6
+	eor	r3,r7,r8
+	eor	r0,r0,r7,ror#20
+	add	r6,r6,r2
+	ldr	r2,[sp,#56]
+	and	r12,r12,r3
+	add	r10,r10,r6
+	add	r6,r6,r0,ror#2
+	eor	r12,r12,r8
+	add	r5,r5,r2
+	eor	r2,r11,r4
+	eor	r0,r10,r10,ror#5
+	add	r6,r6,r12
+	and	r2,r2,r10
+	eor	r12,r0,r10,ror#19
+	eor	r0,r6,r6,ror#11
+	eor	r2,r2,r4
+	add	r5,r5,r12,ror#6
+	eor	r12,r6,r7
+	eor	r0,r0,r6,ror#20
+	add	r5,r5,r2
+	ldr	r2,[sp,#60]
+	and	r3,r3,r12
+	add	r9,r9,r5
+	add	r5,r5,r0,ror#2
+	eor	r3,r3,r7
+	add	r4,r4,r2
+	eor	r2,r10,r11
+	eor	r0,r9,r9,ror#5
+	add	r5,r5,r3
+	and	r2,r2,r9
+	eor	r3,r0,r9,ror#19
+	eor	r0,r5,r5,ror#11
+	eor	r2,r2,r11
+	add	r4,r4,r3,ror#6
+	eor	r3,r5,r6
+	eor	r0,r0,r5,ror#20
+	add	r4,r4,r2
+	ldr	r2,[sp,#64]
+	and	r12,r12,r3
+	add	r8,r8,r4
+	add	r4,r4,r0,ror#2
+	eor	r12,r12,r6
+	vst1.32	{q8},[r1,:128]!
+	ldr	r0,[r2,#0]
+	add	r4,r4,r12			@ h+=Maj(a,b,c) from the past
+	ldr	r12,[r2,#4]
+	ldr	r3,[r2,#8]
+	ldr	r1,[r2,#12]
+	add	r4,r4,r0			@ accumulate
+	ldr	r0,[r2,#16]
+	add	r5,r5,r12
+	ldr	r12,[r2,#20]
+	add	r6,r6,r3
+	ldr	r3,[r2,#24]
+	add	r7,r7,r1
+	ldr	r1,[r2,#28]
+	add	r8,r8,r0
+	str	r4,[r2],#4
+	add	r9,r9,r12
+	str	r5,[r2],#4
+	add	r10,r10,r3
+	str	r6,[r2],#4
+	add	r11,r11,r1
+	str	r7,[r2],#4
+	stmia	r2,{r8-r11}
+
+	ittte	ne
+	movne	r1,sp
+	ldrne	r2,[sp,#0]
+	eorne	r12,r12,r12
+	ldreq	sp,[sp,#76]			@ restore original sp
+	itt	ne
+	eorne	r3,r5,r6
+	bne	.L_00_48
+
+	ldmia	sp!,{r4-r12,pc}
+.size	sha256_block_data_order_neon,.-sha256_block_data_order_neon
+#endif
+#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__)
+
+# ifdef __thumb2__
+#  define INST(a,b,c,d)	.byte	c,d|0xc,a,b
+# else
+#  define INST(a,b,c,d)	.byte	a,b,c,d
+# endif
+
+.type	sha256_block_data_order_armv8,%function
+.align	5
+sha256_block_data_order_armv8:
+.LARMv8:
+	vld1.32	{q0,q1},[r0]
+# ifdef __thumb2__
+	adr	r3,.LARMv8
+	sub	r3,r3,#.LARMv8-K256
+# else
+	adrl	r3,K256
+# endif
+	add	r2,r1,r2,lsl#6	@ len to point at the end of inp
+
+.Loop_v8:
+	vld1.8		{q8-q9},[r1]!
+	vld1.8		{q10-q11},[r1]!
+	vld1.32		{q12},[r3]!
+	vrev32.8	q8,q8
+	vrev32.8	q9,q9
+	vrev32.8	q10,q10
+	vrev32.8	q11,q11
+	vmov		q14,q0	@ offload
+	vmov		q15,q1
+	teq		r1,r2
+	vld1.32		{q13},[r3]!
+	vadd.i32	q12,q12,q8
+	INST(0xe2,0x03,0xfa,0xf3)	@ sha256su0 q8,q9
+	vmov		q2,q0
+	INST(0x68,0x0c,0x02,0xf3)	@ sha256h q0,q1,q12
+	INST(0x68,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q12
+	INST(0xe6,0x0c,0x64,0xf3)	@ sha256su1 q8,q10,q11
+	vld1.32		{q12},[r3]!
+	vadd.i32	q13,q13,q9
+	INST(0xe4,0x23,0xfa,0xf3)	@ sha256su0 q9,q10
+	vmov		q2,q0
+	INST(0x6a,0x0c,0x02,0xf3)	@ sha256h q0,q1,q13
+	INST(0x6a,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q13
+	INST(0xe0,0x2c,0x66,0xf3)	@ sha256su1 q9,q11,q8
+	vld1.32		{q13},[r3]!
+	vadd.i32	q12,q12,q10
+	INST(0xe6,0x43,0xfa,0xf3)	@ sha256su0 q10,q11
+	vmov		q2,q0
+	INST(0x68,0x0c,0x02,0xf3)	@ sha256h q0,q1,q12
+	INST(0x68,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q12
+	INST(0xe2,0x4c,0x60,0xf3)	@ sha256su1 q10,q8,q9
+	vld1.32		{q12},[r3]!
+	vadd.i32	q13,q13,q11
+	INST(0xe0,0x63,0xfa,0xf3)	@ sha256su0 q11,q8
+	vmov		q2,q0
+	INST(0x6a,0x0c,0x02,0xf3)	@ sha256h q0,q1,q13
+	INST(0x6a,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q13
+	INST(0xe4,0x6c,0x62,0xf3)	@ sha256su1 q11,q9,q10
+	vld1.32		{q13},[r3]!
+	vadd.i32	q12,q12,q8
+	INST(0xe2,0x03,0xfa,0xf3)	@ sha256su0 q8,q9
+	vmov		q2,q0
+	INST(0x68,0x0c,0x02,0xf3)	@ sha256h q0,q1,q12
+	INST(0x68,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q12
+	INST(0xe6,0x0c,0x64,0xf3)	@ sha256su1 q8,q10,q11
+	vld1.32		{q12},[r3]!
+	vadd.i32	q13,q13,q9
+	INST(0xe4,0x23,0xfa,0xf3)	@ sha256su0 q9,q10
+	vmov		q2,q0
+	INST(0x6a,0x0c,0x02,0xf3)	@ sha256h q0,q1,q13
+	INST(0x6a,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q13
+	INST(0xe0,0x2c,0x66,0xf3)	@ sha256su1 q9,q11,q8
+	vld1.32		{q13},[r3]!
+	vadd.i32	q12,q12,q10
+	INST(0xe6,0x43,0xfa,0xf3)	@ sha256su0 q10,q11
+	vmov		q2,q0
+	INST(0x68,0x0c,0x02,0xf3)	@ sha256h q0,q1,q12
+	INST(0x68,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q12
+	INST(0xe2,0x4c,0x60,0xf3)	@ sha256su1 q10,q8,q9
+	vld1.32		{q12},[r3]!
+	vadd.i32	q13,q13,q11
+	INST(0xe0,0x63,0xfa,0xf3)	@ sha256su0 q11,q8
+	vmov		q2,q0
+	INST(0x6a,0x0c,0x02,0xf3)	@ sha256h q0,q1,q13
+	INST(0x6a,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q13
+	INST(0xe4,0x6c,0x62,0xf3)	@ sha256su1 q11,q9,q10
+	vld1.32		{q13},[r3]!
+	vadd.i32	q12,q12,q8
+	INST(0xe2,0x03,0xfa,0xf3)	@ sha256su0 q8,q9
+	vmov		q2,q0
+	INST(0x68,0x0c,0x02,0xf3)	@ sha256h q0,q1,q12
+	INST(0x68,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q12
+	INST(0xe6,0x0c,0x64,0xf3)	@ sha256su1 q8,q10,q11
+	vld1.32		{q12},[r3]!
+	vadd.i32	q13,q13,q9
+	INST(0xe4,0x23,0xfa,0xf3)	@ sha256su0 q9,q10
+	vmov		q2,q0
+	INST(0x6a,0x0c,0x02,0xf3)	@ sha256h q0,q1,q13
+	INST(0x6a,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q13
+	INST(0xe0,0x2c,0x66,0xf3)	@ sha256su1 q9,q11,q8
+	vld1.32		{q13},[r3]!
+	vadd.i32	q12,q12,q10
+	INST(0xe6,0x43,0xfa,0xf3)	@ sha256su0 q10,q11
+	vmov		q2,q0
+	INST(0x68,0x0c,0x02,0xf3)	@ sha256h q0,q1,q12
+	INST(0x68,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q12
+	INST(0xe2,0x4c,0x60,0xf3)	@ sha256su1 q10,q8,q9
+	vld1.32		{q12},[r3]!
+	vadd.i32	q13,q13,q11
+	INST(0xe0,0x63,0xfa,0xf3)	@ sha256su0 q11,q8
+	vmov		q2,q0
+	INST(0x6a,0x0c,0x02,0xf3)	@ sha256h q0,q1,q13
+	INST(0x6a,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q13
+	INST(0xe4,0x6c,0x62,0xf3)	@ sha256su1 q11,q9,q10
+	vld1.32		{q13},[r3]!
+	vadd.i32	q12,q12,q8
+	vmov		q2,q0
+	INST(0x68,0x0c,0x02,0xf3)	@ sha256h q0,q1,q12
+	INST(0x68,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q12
+
+	vld1.32		{q12},[r3]!
+	vadd.i32	q13,q13,q9
+	vmov		q2,q0
+	INST(0x6a,0x0c,0x02,0xf3)	@ sha256h q0,q1,q13
+	INST(0x6a,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q13
+
+	vld1.32		{q13},[r3]
+	vadd.i32	q12,q12,q10
+	sub		r3,r3,#256-16	@ rewind
+	vmov		q2,q0
+	INST(0x68,0x0c,0x02,0xf3)	@ sha256h q0,q1,q12
+	INST(0x68,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q12
+
+	vadd.i32	q13,q13,q11
+	vmov		q2,q0
+	INST(0x6a,0x0c,0x02,0xf3)	@ sha256h q0,q1,q13
+	INST(0x6a,0x2c,0x14,0xf3)	@ sha256h2 q1,q2,q13
+
+	vadd.i32	q0,q0,q14
+	vadd.i32	q1,q1,q15
+	it		ne
+	bne		.Loop_v8
+
+	vst1.32		{q0,q1},[r0]
+
+	bx	lr		@ bx lr
+.size	sha256_block_data_order_armv8,.-sha256_block_data_order_armv8
+#endif
+.asciz  "SHA256 block transform for ARMv4/NEON/ARMv8, CRYPTOGAMS by <appro@openssl.org>"
+.align	2
+#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__)
+.comm   OPENSSL_armcap_P,4,4
+#endif

arch/arm/crypto/sha256_glue.c

+/*
+ * Glue code for the SHA256 Secure Hash Algorithm assembly implementation
+ * using optimized ARM assembler and NEON instructions.
+ *
+ * Copyright  2015 Google Inc.
+ *
+ * This file is based on sha256_ssse3_glue.c:
+ *   Copyright (C) 2013 Intel Corporation
+ *   Author: Tim Chen <tim.c.chen@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+
+#include <crypto/internal/hash.h>
+#include <linux/crypto.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/mm.h>
+#include <linux/cryptohash.h>
+#include <linux/types.h>
+#include <linux/string.h>
+#include <crypto/sha.h>
+#include <asm/byteorder.h>
+#include <asm/simd.h>
+#include <asm/neon.h>
+#include "sha256_glue.h"
+
+asmlinkage void sha256_block_data_order(u32 *digest, const void *data,
+				      unsigned int num_blks);
+
+
+int sha256_init(struct shash_desc *desc)
+{
+	struct sha256_state *sctx = shash_desc_ctx(desc);
+
+	sctx->state[0] = SHA256_H0;
+	sctx->state[1] = SHA256_H1;
+	sctx->state[2] = SHA256_H2;
+	sctx->state[3] = SHA256_H3;
+	sctx->state[4] = SHA256_H4;
+	sctx->state[5] = SHA256_H5;
+	sctx->state[6] = SHA256_H6;
+	sctx->state[7] = SHA256_H7;
+	sctx->count = 0;
+
+	return 0;
+}
+
+int sha224_init(struct shash_desc *desc)
+{
+	struct sha256_state *sctx = shash_desc_ctx(desc);
+
+	sctx->state[0] = SHA224_H0;
+	sctx->state[1] = SHA224_H1;
+	sctx->state[2] = SHA224_H2;
+	sctx->state[3] = SHA224_H3;
+	sctx->state[4] = SHA224_H4;
+	sctx->state[5] = SHA224_H5;
+	sctx->state[6] = SHA224_H6;
+	sctx->state[7] = SHA224_H7;
+	sctx->count = 0;
+
+	return 0;
+}
+
+int __sha256_update(struct shash_desc *desc, const u8 *data, unsigned int len,
+		    unsigned int partial)
+{
+	struct sha256_state *sctx = shash_desc_ctx(desc);
+	unsigned int done = 0;
+
+	sctx->count += len;
+
+	if (partial) {
+		done = SHA256_BLOCK_SIZE - partial;
+		memcpy(sctx->buf + partial, data, done);
+		sha256_block_data_order(sctx->state, sctx->buf, 1);
+	}
+
+	if (len - done >= SHA256_BLOCK_SIZE) {
+		const unsigned int rounds = (len - done) / SHA256_BLOCK_SIZE;
+
+		sha256_block_data_order(sctx->state, data + done, rounds);
+		done += rounds * SHA256_BLOCK_SIZE;
+	}
+
+	memcpy(sctx->buf, data + done, len - done);
+
+	return 0;
+}
+
+int sha256_update(struct shash_desc *desc, const u8 *data, unsigned int len)
+{
+	struct sha256_state *sctx = shash_desc_ctx(desc);
+	unsigned int partial = sctx->count % SHA256_BLOCK_SIZE;
+
+	/* Handle the fast case right here */
+	if (partial + len < SHA256_BLOCK_SIZE) {
+		sctx->count += len;
+		memcpy(sctx->buf + partial, data, len);
+
+		return 0;
+	}
+
+	return __sha256_update(desc, data, len, partial);
+}
+
+/* Add padding and return the message digest. */
+static int sha256_final(struct shash_desc *desc, u8 *out)
+{
+	struct sha256_state *sctx = shash_desc_ctx(desc);
+	unsigned int i, index, padlen;
+	__be32 *dst = (__be32 *)out;
+	__be64 bits;
+	static const u8 padding[SHA256_BLOCK_SIZE] = { 0x80, };
+
+	/* save number of bits */
+	bits = cpu_to_be64(sctx->count << 3);
+
+	/* Pad out to 56 mod 64 and append length */
+	index = sctx->count % SHA256_BLOCK_SIZE;
+	padlen = (index < 56) ? (56 - index) : ((SHA256_BLOCK_SIZE+56)-index);
+
+	/* We need to fill a whole block for __sha256_update */
+	if (padlen <= 56) {
+		sctx->count += padlen;
+		memcpy(sctx->buf + index, padding, padlen);
+	} else {
+		__sha256_update(desc, padding, padlen, index);
+	}
+	__sha256_update(desc, (const u8 *)&bits, sizeof(bits), 56);
+
+	/* Store state in digest */
+	for (i = 0; i < 8; i++)
+		dst[i] = cpu_to_be32(sctx->state[i]);
+
+	/* Wipe context */
+	memset(sctx, 0, sizeof(*sctx));
+
+	return 0;
+}
+
+static int sha224_final(struct shash_desc *desc, u8 *out)
+{
+	u8 D[SHA256_DIGEST_SIZE];
+
+	sha256_final(desc, D);
+
+	memcpy(out, D, SHA224_DIGEST_SIZE);
+	memzero_explicit(D, SHA256_DIGEST_SIZE);
+
+	return 0;
+}
+
+int sha256_export(struct shash_desc *desc, void *out)
+{
+	struct sha256_state *sctx = shash_desc_ctx(desc);
+
+	memcpy(out, sctx, sizeof(*sctx));
+
+	return 0;
+}
+
+int sha256_import(struct shash_desc *desc, const void *in)
+{
+	struct sha256_state *sctx = shash_desc_ctx(desc);
+
+	memcpy(sctx, in, sizeof(*sctx));
+
+	return 0;
+}
+
+static struct shash_alg algs[] = { {
+	.digestsize	=	SHA256_DIGEST_SIZE,
+	.init		=	sha256_init,
+	.update		=	sha256_update,
+	.final		=	sha256_final,
+	.export		=	sha256_export,
+	.import		=	sha256_import,
+	.descsize	=	sizeof(struct sha256_state),
+	.statesize	=	sizeof(struct sha256_state),
+	.base		=	{
+		.cra_name	=	"sha256",
+		.cra_driver_name =	"sha256-asm",
+		.cra_priority	=	150,
+		.cra_flags	=	CRYPTO_ALG_TYPE_SHASH,
+		.cra_blocksize	=	SHA256_BLOCK_SIZE,
+		.cra_module	=	THIS_MODULE,
+	}
+}, {
+	.digestsize	=	SHA224_DIGEST_SIZE,
+	.init		=	sha224_init,
+	.update		=	sha256_update,
+	.final		=	sha224_final,
+	.export		=	sha256_export,
+	.import		=	sha256_import,
+	.descsize	=	sizeof(struct sha256_state),
+	.statesize	=	sizeof(struct sha256_state),
+	.base		=	{
+		.cra_name	=	"sha224",
+		.cra_driver_name =	"sha224-asm",
+		.cra_priority	=	150,
+		.cra_flags	=	CRYPTO_ALG_TYPE_SHASH,
+		.cra_blocksize	=	SHA224_BLOCK_SIZE,
+		.cra_module	=	THIS_MODULE,
+	}
+} };
+
+static int __init sha256_mod_init(void)
+{
+	int res = crypto_register_shashes(algs, ARRAY_SIZE(algs));
+
+	if (res < 0)
+		return res;
+
+	if (IS_ENABLED(CONFIG_KERNEL_MODE_NEON) && cpu_has_neon()) {
+		res = crypto_register_shashes(sha256_neon_algs,
+					      ARRAY_SIZE(sha256_neon_algs));
+
+		if (res < 0)
+			crypto_unregister_shashes(algs, ARRAY_SIZE(algs));
+	}
+
+	return res;
+}
+
+static void __exit sha256_mod_fini(void)
+{
+	crypto_unregister_shashes(algs, ARRAY_SIZE(algs));
+
+	if (IS_ENABLED(CONFIG_KERNEL_MODE_NEON) && cpu_has_neon())
+		crypto_unregister_shashes(sha256_neon_algs,
+					  ARRAY_SIZE(sha256_neon_algs));
+}
+
+module_init(sha256_mod_init);
+module_exit(sha256_mod_fini);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("SHA256 Secure Hash Algorithm (ARM), including NEON");
+
+MODULE_ALIAS_CRYPTO("sha256");

arch/arm/crypto/sha256_glue.h

+#ifndef _CRYPTO_SHA256_GLUE_H
+#define _CRYPTO_SHA256_GLUE_H
+
+#include <linux/crypto.h>
+#include <crypto/sha.h>
+
+extern struct shash_alg sha256_neon_algs[2];
+
+extern int sha256_init(struct shash_desc *desc);
+
+extern int sha224_init(struct shash_desc *desc);
+
+extern int __sha256_update(struct shash_desc *desc, const u8 *data,
+			   unsigned int len, unsigned int partial);
+
+extern int sha256_update(struct shash_desc *desc, const u8 *data,
+			 unsigned int len);
+
+extern int sha256_export(struct shash_desc *desc, void *out);
+
+extern int sha256_import(struct shash_desc *desc, const void *in);
+
+#endif /* _CRYPTO_SHA256_GLUE_H */

arch/arm/crypto/sha256_neon_glue.c

+/*
+ * Glue code for the SHA256 Secure Hash Algorithm assembly implementation
+ * using NEON instructions.
+ *
+ * Copyright  2015 Google Inc.
+ *
+ * This file is based on sha512_neon_glue.c:
+ *   Copyright  2014 Jussi Kivilinna <jussi.kivilinna@iki.fi>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+
+#include <crypto/internal/hash.h>
+#include <linux/cryptohash.h>
+#include <linux/types.h>
+#include <linux/string.h>
+#include <crypto/sha.h>
+#include <asm/byteorder.h>
+#include <asm/simd.h>
+#include <asm/neon.h>
+#include "sha256_glue.h"
+
+asmlinkage void sha256_block_data_order_neon(u32 *digest, const void *data,
+				      unsigned int num_blks);
+
+
+static int __sha256_neon_update(struct shash_desc *desc, const u8 *data,
+				unsigned int len, unsigned int partial)
+{
+	struct sha256_state *sctx = shash_desc_ctx(desc);
+	unsigned int done = 0;
+
+	sctx->count += len;
+
+	if (partial) {
+		done = SHA256_BLOCK_SIZE - partial;
+		memcpy(sctx->buf + partial, data, done);
+		sha256_block_data_order_neon(sctx->state, sctx->buf, 1);
+	}
+
+	if (len - done >= SHA256_BLOCK_SIZE) {
+		const unsigned int rounds = (len - done) / SHA256_BLOCK_SIZE;
+
+		sha256_block_data_order_neon(sctx->state, data + done, rounds);
+		done += rounds * SHA256_BLOCK_SIZE;
+	}
+
+	memcpy(sctx->buf, data + done, len - done);
+
+	return 0;
+}
+
+static int sha256_neon_update(struct shash_desc *desc, const u8 *data,
+			      unsigned int len)
+{
+	struct sha256_state *sctx = shash_desc_ctx(desc);
+	unsigned int partial = sctx->count % SHA256_BLOCK_SIZE;
+	int res;
+
+	/* Handle the fast case right here */
+	if (partial + len < SHA256_BLOCK_SIZE) {
+		sctx->count += len;
+		memcpy(sctx->buf + partial, data, len);
+
+		return 0;
+	}
+
+	if (!may_use_simd()) {
+		res = __sha256_update(desc, data, len, partial);
+	} else {
+		kernel_neon_begin();
+		res = __sha256_neon_update(desc, data, len, partial);
+		kernel_neon_end();
+	}
+
+	return res;
+}
+
+/* Add padding and return the message digest. */
+static int sha256_neon_final(struct shash_desc *desc, u8 *out)
+{
+	struct sha256_state *sctx = shash_desc_ctx(desc);
+	unsigned int i, index, padlen;
+	__be32 *dst = (__be32 *)out;
+	__be64 bits;
+	static const u8 padding[SHA256_BLOCK_SIZE] = { 0x80, };
+
+	/* save number of bits */
+	bits = cpu_to_be64(sctx->count << 3);
+
+	/* Pad out to 56 mod 64 and append length */
+	index = sctx->count % SHA256_BLOCK_SIZE;
+	padlen = (index < 56) ? (56 - index) : ((SHA256_BLOCK_SIZE+56)-index);
+
+	if (!may_use_simd()) {
+		sha256_update(desc, padding, padlen);
+		sha256_update(desc, (const u8 *)&bits, sizeof(bits));
+	} else {
+		kernel_neon_begin();
+		/* We need to fill a whole block for __sha256_neon_update() */
+		if (padlen <= 56) {
+			sctx->count += padlen;
+			memcpy(sctx->buf + index, padding, padlen);
+		} else {
+			__sha256_neon_update(desc, padding, padlen, index);
+		}
+		__sha256_neon_update(desc, (const u8 *)&bits,
+					sizeof(bits), 56);
+		kernel_neon_end();
+	}
+
+	/* Store state in digest */
+	for (i = 0; i < 8; i++)
+		dst[i] = cpu_to_be32(sctx->state[i]);
+
+	/* Wipe context */
+	memzero_explicit(sctx, sizeof(*sctx));
+
+	return 0;
+}
+
+static int sha224_neon_final(struct shash_desc *desc, u8 *out)
+{
+	u8 D[SHA256_DIGEST_SIZE];
+
+	sha256_neon_final(desc, D);
+
+	memcpy(out, D, SHA224_DIGEST_SIZE);
+	memzero_explicit(D, SHA256_DIGEST_SIZE);
+
+	return 0;
+}
+
+struct shash_alg sha256_neon_algs[] = { {
+	.digestsize	=	SHA256_DIGEST_SIZE,
+	.init		=	sha256_init,
+	.update		=	sha256_neon_update,
+	.final		=	sha256_neon_final,
+	.export		=	sha256_export,
+	.import		=	sha256_import,
+	.descsize	=	sizeof(struct sha256_state),
+	.statesize	=	sizeof(struct sha256_state),
+	.base		=	{
+		.cra_name	=	"sha256",
+		.cra_driver_name =	"sha256-neon",
+		.cra_priority	=	250,
+		.cra_flags	=	CRYPTO_ALG_TYPE_SHASH,
+		.cra_blocksize	=	SHA256_BLOCK_SIZE,
+		.cra_module	=	THIS_MODULE,
+	}
+}, {
+	.digestsize	=	SHA224_DIGEST_SIZE,
+	.init		=	sha224_init,
+	.update		=	sha256_neon_update,
+	.final		=	sha224_neon_final,
+	.export		=	sha256_export,
+	.import		=	sha256_import,
+	.descsize	=	sizeof(struct sha256_state),
+	.statesize	=	sizeof(struct sha256_state),
+	.base		=	{
+		.cra_name	=	"sha224",
+		.cra_driver_name =	"sha224-neon",
+		.cra_priority	=	250,
+		.cra_flags	=	CRYPTO_ALG_TYPE_SHASH,
+		.cra_blocksize	=	SHA224_BLOCK_SIZE,
+		.cra_module	=	THIS_MODULE,
+	}
+} };