Merge branch 'develop' into release-0.5.2

1 files changed, 1998 insertions, 0 deletions

diff --git a/app/openssl/crypto/sha/asm/sha256-mips.S b/app/openssl/crypto/sha/asm/sha256-mips.S
new file mode 100644
index 00000000..2bd728e9
--- /dev/null
+++ b/app/openssl/crypto/sha/asm/sha256-mips.S
@@ -0,0 +1,1998 @@
+#ifdef OPENSSL_FIPSCANISTER
+# include <openssl/fipssyms.h>
+#endif
+
+.text
+.set	noat
+#if !defined(__vxworks) || defined(__pic__)
+.option	pic2
+#endif
+
+.align	5
+.globl	sha256_block_data_order
+.ent	sha256_block_data_order
+sha256_block_data_order:
+	.frame	$29,128,$31
+	.mask	3237937152,-4
+	.set	noreorder
+	.cpload	$25
+	sub $29,128
+	sw	$31,128-1*4($29)
+	sw	$30,128-2*4($29)
+	sw	$23,128-3*4($29)
+	sw	$22,128-4*4($29)
+	sw	$21,128-5*4($29)
+	sw	$20,128-6*4($29)
+	sw	$19,128-7*4($29)
+	sw	$18,128-8*4($29)
+	sw	$17,128-9*4($29)
+	sw	$16,128-10*4($29)
+	sll $23,$6,6
+	.set	reorder
+	la	$6,K256		# PIC-ified 'load address'
+
+	lw	$1,0*4($4)		# load context
+	lw	$2,1*4($4)
+	lw	$3,2*4($4)
+	lw	$7,3*4($4)
+	lw	$24,4*4($4)
+	lw	$25,5*4($4)
+	lw	$30,6*4($4)
+	lw	$31,7*4($4)
+
+	add $23,$5		# pointer to the end of input
+	sw	$23,16*4($29)
+	b	.Loop
+
+.align	5
+.Loop:
+	lwl	$8,3($5)
+	lwr	$8,0($5)
+	lwl	$9,7($5)
+	lwr	$9,4($5)
+	srl	$13,$8,24		# byte swap(0)
+	srl	$14,$8,8
+	andi	$15,$8,0xFF00
+	sll	$8,$8,24
+	andi	$14,0xFF00
+	sll	$15,$15,8
+	or	$8,$13
+	or	$14,$15
+	or	$8,$14
+	addu	$12,$8,$31			# 0
+	srl	$31,$24,6
+	xor	$15,$25,$30
+	sll	$14,$24,7
+	and	$15,$24
+	srl	$13,$24,11
+	xor	$31,$14
+	sll	$14,$24,21
+	xor	$31,$13
+	srl	$13,$24,25
+	xor	$31,$14
+	sll	$14,$24,26
+	xor	$31,$13
+	xor	$15,$30			# Ch(e,f,g)
+	xor	$13,$14,$31			# Sigma1(e)
+
+	srl	$31,$1,2
+	addu	$12,$15
+	lw	$15,0($6)		# K[0]
+	sll	$14,$1,10
+	addu	$12,$13
+	srl	$13,$1,13
+	xor	$31,$14
+	sll	$14,$1,19
+	xor	$31,$13
+	srl	$13,$1,22
+	xor	$31,$14
+	sll	$14,$1,30
+	xor	$31,$13
+	sw	$8,0($29)	# offload to ring buffer
+	xor	$31,$14			# Sigma0(a)
+
+	or	$13,$1,$2
+	and	$14,$1,$2
+	and	$13,$3
+	or	$14,$13			# Maj(a,b,c)
+	addu	$12,$15			# +=K[0]
+	addu	$31,$14
+
+	addu	$7,$12
+	addu	$31,$12
+	lwl	$10,11($5)
+	lwr	$10,8($5)
+	srl	$14,$9,24		# byte swap(1)
+	srl	$15,$9,8
+	andi	$16,$9,0xFF00
+	sll	$9,$9,24
+	andi	$15,0xFF00
+	sll	$16,$16,8
+	or	$9,$14
+	or	$15,$16
+	or	$9,$15
+	addu	$13,$9,$30			# 1
+	srl	$30,$7,6
+	xor	$16,$24,$25
+	sll	$15,$7,7
+	and	$16,$7
+	srl	$14,$7,11
+	xor	$30,$15
+	sll	$15,$7,21
+	xor	$30,$14
+	srl	$14,$7,25
+	xor	$30,$15
+	sll	$15,$7,26
+	xor	$30,$14
+	xor	$16,$25			# Ch(e,f,g)
+	xor	$14,$15,$30			# Sigma1(e)
+
+	srl	$30,$31,2
+	addu	$13,$16
+	lw	$16,4($6)		# K[1]
+	sll	$15,$31,10
+	addu	$13,$14
+	srl	$14,$31,13
+	xor	$30,$15
+	sll	$15,$31,19
+	xor	$30,$14
+	srl	$14,$31,22
+	xor	$30,$15
+	sll	$15,$31,30
+	xor	$30,$14
+	sw	$9,4($29)	# offload to ring buffer
+	xor	$30,$15			# Sigma0(a)
+
+	or	$14,$31,$1
+	and	$15,$31,$1
+	and	$14,$2
+	or	$15,$14			# Maj(a,b,c)
+	addu	$13,$16			# +=K[1]
+	addu	$30,$15
+
+	addu	$3,$13
+	addu	$30,$13
+	lwl	$11,15($5)
+	lwr	$11,12($5)
+	srl	$15,$10,24		# byte swap(2)
+	srl	$16,$10,8
+	andi	$17,$10,0xFF00
+	sll	$10,$10,24
+	andi	$16,0xFF00
+	sll	$17,$17,8
+	or	$10,$15
+	or	$16,$17
+	or	$10,$16
+	addu	$14,$10,$25			# 2
+	srl	$25,$3,6
+	xor	$17,$7,$24
+	sll	$16,$3,7
+	and	$17,$3
+	srl	$15,$3,11
+	xor	$25,$16
+	sll	$16,$3,21
+	xor	$25,$15
+	srl	$15,$3,25
+	xor	$25,$16
+	sll	$16,$3,26
+	xor	$25,$15
+	xor	$17,$24			# Ch(e,f,g)
+	xor	$15,$16,$25			# Sigma1(e)
+
+	srl	$25,$30,2
+	addu	$14,$17
+	lw	$17,8($6)		# K[2]
+	sll	$16,$30,10
+	addu	$14,$15
+	srl	$15,$30,13
+	xor	$25,$16
+	sll	$16,$30,19
+	xor	$25,$15
+	srl	$15,$30,22
+	xor	$25,$16
+	sll	$16,$30,30
+	xor	$25,$15
+	sw	$10,8($29)	# offload to ring buffer
+	xor	$25,$16			# Sigma0(a)
+
+	or	$15,$30,$31
+	and	$16,$30,$31
+	and	$15,$1
+	or	$16,$15			# Maj(a,b,c)
+	addu	$14,$17			# +=K[2]
+	addu	$25,$16
+
+	addu	$2,$14
+	addu	$25,$14
+	lwl	$12,19($5)
+	lwr	$12,16($5)
+	srl	$16,$11,24		# byte swap(3)
+	srl	$17,$11,8
+	andi	$18,$11,0xFF00
+	sll	$11,$11,24
+	andi	$17,0xFF00
+	sll	$18,$18,8
+	or	$11,$16
+	or	$17,$18
+	or	$11,$17
+	addu	$15,$11,$24			# 3
+	srl	$24,$2,6
+	xor	$18,$3,$7
+	sll	$17,$2,7
+	and	$18,$2
+	srl	$16,$2,11
+	xor	$24,$17
+	sll	$17,$2,21
+	xor	$24,$16
+	srl	$16,$2,25
+	xor	$24,$17
+	sll	$17,$2,26
+	xor	$24,$16
+	xor	$18,$7			# Ch(e,f,g)
+	xor	$16,$17,$24			# Sigma1(e)
+
+	srl	$24,$25,2
+	addu	$15,$18
+	lw	$18,12($6)		# K[3]
+	sll	$17,$25,10
+	addu	$15,$16
+	srl	$16,$25,13
+	xor	$24,$17
+	sll	$17,$25,19
+	xor	$24,$16
+	srl	$16,$25,22
+	xor	$24,$17
+	sll	$17,$25,30
+	xor	$24,$16
+	sw	$11,12($29)	# offload to ring buffer
+	xor	$24,$17			# Sigma0(a)
+
+	or	$16,$25,$30
+	and	$17,$25,$30
+	and	$16,$31
+	or	$17,$16			# Maj(a,b,c)
+	addu	$15,$18			# +=K[3]
+	addu	$24,$17
+
+	addu	$1,$15
+	addu	$24,$15
+	lwl	$13,23($5)
+	lwr	$13,20($5)
+	srl	$17,$12,24		# byte swap(4)
+	srl	$18,$12,8
+	andi	$19,$12,0xFF00
+	sll	$12,$12,24
+	andi	$18,0xFF00
+	sll	$19,$19,8
+	or	$12,$17
+	or	$18,$19
+	or	$12,$18
+	addu	$16,$12,$7			# 4
+	srl	$7,$1,6
+	xor	$19,$2,$3
+	sll	$18,$1,7
+	and	$19,$1
+	srl	$17,$1,11
+	xor	$7,$18
+	sll	$18,$1,21
+	xor	$7,$17
+	srl	$17,$1,25
+	xor	$7,$18
+	sll	$18,$1,26
+	xor	$7,$17
+	xor	$19,$3			# Ch(e,f,g)
+	xor	$17,$18,$7			# Sigma1(e)
+
+	srl	$7,$24,2
+	addu	$16,$19
+	lw	$19,16($6)		# K[4]
+	sll	$18,$24,10
+	addu	$16,$17
+	srl	$17,$24,13
+	xor	$7,$18
+	sll	$18,$24,19
+	xor	$7,$17
+	srl	$17,$24,22
+	xor	$7,$18
+	sll	$18,$24,30
+	xor	$7,$17
+	sw	$12,16($29)	# offload to ring buffer
+	xor	$7,$18			# Sigma0(a)
+
+	or	$17,$24,$25
+	and	$18,$24,$25
+	and	$17,$30
+	or	$18,$17			# Maj(a,b,c)
+	addu	$16,$19			# +=K[4]
+	addu	$7,$18
+
+	addu	$31,$16
+	addu	$7,$16
+	lwl	$14,27($5)
+	lwr	$14,24($5)
+	srl	$18,$13,24		# byte swap(5)
+	srl	$19,$13,8
+	andi	$20,$13,0xFF00
+	sll	$13,$13,24
+	andi	$19,0xFF00
+	sll	$20,$20,8
+	or	$13,$18
+	or	$19,$20
+	or	$13,$19
+	addu	$17,$13,$3			# 5
+	srl	$3,$31,6
+	xor	$20,$1,$2
+	sll	$19,$31,7
+	and	$20,$31
+	srl	$18,$31,11
+	xor	$3,$19
+	sll	$19,$31,21
+	xor	$3,$18
+	srl	$18,$31,25
+	xor	$3,$19
+	sll	$19,$31,26
+	xor	$3,$18
+	xor	$20,$2			# Ch(e,f,g)
+	xor	$18,$19,$3			# Sigma1(e)
+
+	srl	$3,$7,2
+	addu	$17,$20
+	lw	$20,20($6)		# K[5]
+	sll	$19,$7,10
+	addu	$17,$18
+	srl	$18,$7,13
+	xor	$3,$19
+	sll	$19,$7,19
+	xor	$3,$18
+	srl	$18,$7,22
+	xor	$3,$19
+	sll	$19,$7,30
+	xor	$3,$18
+	sw	$13,20($29)	# offload to ring buffer
+	xor	$3,$19			# Sigma0(a)
+
+	or	$18,$7,$24
+	and	$19,$7,$24
+	and	$18,$25
+	or	$19,$18			# Maj(a,b,c)
+	addu	$17,$20			# +=K[5]
+	addu	$3,$19
+
+	addu	$30,$17
+	addu	$3,$17
+	lwl	$15,31($5)
+	lwr	$15,28($5)
+	srl	$19,$14,24		# byte swap(6)
+	srl	$20,$14,8
+	andi	$21,$14,0xFF00
+	sll	$14,$14,24
+	andi	$20,0xFF00
+	sll	$21,$21,8
+	or	$14,$19
+	or	$20,$21
+	or	$14,$20
+	addu	$18,$14,$2			# 6
+	srl	$2,$30,6
+	xor	$21,$31,$1
+	sll	$20,$30,7
+	and	$21,$30
+	srl	$19,$30,11
+	xor	$2,$20
+	sll	$20,$30,21
+	xor	$2,$19
+	srl	$19,$30,25
+	xor	$2,$20
+	sll	$20,$30,26
+	xor	$2,$19
+	xor	$21,$1			# Ch(e,f,g)
+	xor	$19,$20,$2			# Sigma1(e)
+
+	srl	$2,$3,2
+	addu	$18,$21
+	lw	$21,24($6)		# K[6]
+	sll	$20,$3,10
+	addu	$18,$19
+	srl	$19,$3,13
+	xor	$2,$20
+	sll	$20,$3,19
+	xor	$2,$19
+	srl	$19,$3,22
+	xor	$2,$20
+	sll	$20,$3,30
+	xor	$2,$19
+	sw	$14,24($29)	# offload to ring buffer
+	xor	$2,$20			# Sigma0(a)
+
+	or	$19,$3,$7
+	and	$20,$3,$7
+	and	$19,$24
+	or	$20,$19			# Maj(a,b,c)
+	addu	$18,$21			# +=K[6]
+	addu	$2,$20
+
+	addu	$25,$18
+	addu	$2,$18
+	lwl	$16,35($5)
+	lwr	$16,32($5)
+	srl	$20,$15,24		# byte swap(7)
+	srl	$21,$15,8
+	andi	$22,$15,0xFF00
+	sll	$15,$15,24
+	andi	$21,0xFF00
+	sll	$22,$22,8
+	or	$15,$20
+	or	$21,$22
+	or	$15,$21
+	addu	$19,$15,$1			# 7
+	srl	$1,$25,6
+	xor	$22,$30,$31
+	sll	$21,$25,7
+	and	$22,$25
+	srl	$20,$25,11
+	xor	$1,$21
+	sll	$21,$25,21
+	xor	$1,$20
+	srl	$20,$25,25
+	xor	$1,$21
+	sll	$21,$25,26
+	xor	$1,$20
+	xor	$22,$31			# Ch(e,f,g)
+	xor	$20,$21,$1			# Sigma1(e)
+
+	srl	$1,$2,2
+	addu	$19,$22
+	lw	$22,28($6)		# K[7]
+	sll	$21,$2,10
+	addu	$19,$20
+	srl	$20,$2,13
+	xor	$1,$21
+	sll	$21,$2,19
+	xor	$1,$20
+	srl	$20,$2,22
+	xor	$1,$21
+	sll	$21,$2,30
+	xor	$1,$20
+	sw	$15,28($29)	# offload to ring buffer
+	xor	$1,$21			# Sigma0(a)
+
+	or	$20,$2,$3
+	and	$21,$2,$3
+	and	$20,$7
+	or	$21,$20			# Maj(a,b,c)
+	addu	$19,$22			# +=K[7]
+	addu	$1,$21
+
+	addu	$24,$19
+	addu	$1,$19
+	lwl	$17,39($5)
+	lwr	$17,36($5)
+	srl	$21,$16,24		# byte swap(8)
+	srl	$22,$16,8
+	andi	$23,$16,0xFF00
+	sll	$16,$16,24
+	andi	$22,0xFF00
+	sll	$23,$23,8
+	or	$16,$21
+	or	$22,$23
+	or	$16,$22
+	addu	$20,$16,$31			# 8
+	srl	$31,$24,6
+	xor	$23,$25,$30
+	sll	$22,$24,7
+	and	$23,$24
+	srl	$21,$24,11
+	xor	$31,$22
+	sll	$22,$24,21
+	xor	$31,$21
+	srl	$21,$24,25
+	xor	$31,$22
+	sll	$22,$24,26
+	xor	$31,$21
+	xor	$23,$30			# Ch(e,f,g)
+	xor	$21,$22,$31			# Sigma1(e)
+
+	srl	$31,$1,2
+	addu	$20,$23
+	lw	$23,32($6)		# K[8]
+	sll	$22,$1,10
+	addu	$20,$21
+	srl	$21,$1,13
+	xor	$31,$22
+	sll	$22,$1,19
+	xor	$31,$21
+	srl	$21,$1,22
+	xor	$31,$22
+	sll	$22,$1,30
+	xor	$31,$21
+	sw	$16,32($29)	# offload to ring buffer
+	xor	$31,$22			# Sigma0(a)
+
+	or	$21,$1,$2
+	and	$22,$1,$2
+	and	$21,$3
+	or	$22,$21			# Maj(a,b,c)
+	addu	$20,$23			# +=K[8]
+	addu	$31,$22
+
+	addu	$7,$20
+	addu	$31,$20
+	lwl	$18,43($5)
+	lwr	$18,40($5)
+	srl	$22,$17,24		# byte swap(9)
+	srl	$23,$17,8
+	andi	$8,$17,0xFF00
+	sll	$17,$17,24
+	andi	$23,0xFF00
+	sll	$8,$8,8
+	or	$17,$22
+	or	$23,$8
+	or	$17,$23
+	addu	$21,$17,$30			# 9
+	srl	$30,$7,6
+	xor	$8,$24,$25
+	sll	$23,$7,7
+	and	$8,$7
+	srl	$22,$7,11
+	xor	$30,$23
+	sll	$23,$7,21
+	xor	$30,$22
+	srl	$22,$7,25
+	xor	$30,$23
+	sll	$23,$7,26
+	xor	$30,$22
+	xor	$8,$25			# Ch(e,f,g)
+	xor	$22,$23,$30			# Sigma1(e)
+
+	srl	$30,$31,2
+	addu	$21,$8
+	lw	$8,36($6)		# K[9]
+	sll	$23,$31,10
+	addu	$21,$22
+	srl	$22,$31,13
+	xor	$30,$23
+	sll	$23,$31,19
+	xor	$30,$22
+	srl	$22,$31,22
+	xor	$30,$23
+	sll	$23,$31,30
+	xor	$30,$22
+	sw	$17,36($29)	# offload to ring buffer
+	xor	$30,$23			# Sigma0(a)
+
+	or	$22,$31,$1
+	and	$23,$31,$1
+	and	$22,$2
+	or	$23,$22			# Maj(a,b,c)
+	addu	$21,$8			# +=K[9]
+	addu	$30,$23
+
+	addu	$3,$21
+	addu	$30,$21
+	lwl	$19,47($5)
+	lwr	$19,44($5)
+	srl	$23,$18,24		# byte swap(10)
+	srl	$8,$18,8
+	andi	$9,$18,0xFF00
+	sll	$18,$18,24
+	andi	$8,0xFF00
+	sll	$9,$9,8
+	or	$18,$23
+	or	$8,$9
+	or	$18,$8
+	addu	$22,$18,$25			# 10
+	srl	$25,$3,6
+	xor	$9,$7,$24
+	sll	$8,$3,7
+	and	$9,$3
+	srl	$23,$3,11
+	xor	$25,$8
+	sll	$8,$3,21
+	xor	$25,$23
+	srl	$23,$3,25
+	xor	$25,$8
+	sll	$8,$3,26
+	xor	$25,$23
+	xor	$9,$24			# Ch(e,f,g)
+	xor	$23,$8,$25			# Sigma1(e)
+
+	srl	$25,$30,2
+	addu	$22,$9
+	lw	$9,40($6)		# K[10]
+	sll	$8,$30,10
+	addu	$22,$23
+	srl	$23,$30,13
+	xor	$25,$8
+	sll	$8,$30,19
+	xor	$25,$23
+	srl	$23,$30,22
+	xor	$25,$8
+	sll	$8,$30,30
+	xor	$25,$23
+	sw	$18,40($29)	# offload to ring buffer
+	xor	$25,$8			# Sigma0(a)
+
+	or	$23,$30,$31
+	and	$8,$30,$31
+	and	$23,$1
+	or	$8,$23			# Maj(a,b,c)
+	addu	$22,$9			# +=K[10]
+	addu	$25,$8
+
+	addu	$2,$22
+	addu	$25,$22
+	lwl	$20,51($5)
+	lwr	$20,48($5)
+	srl	$8,$19,24		# byte swap(11)
+	srl	$9,$19,8
+	andi	$10,$19,0xFF00
+	sll	$19,$19,24
+	andi	$9,0xFF00
+	sll	$10,$10,8
+	or	$19,$8
+	or	$9,$10
+	or	$19,$9
+	addu	$23,$19,$24			# 11
+	srl	$24,$2,6
+	xor	$10,$3,$7
+	sll	$9,$2,7
+	and	$10,$2
+	srl	$8,$2,11
+	xor	$24,$9
+	sll	$9,$2,21
+	xor	$24,$8
+	srl	$8,$2,25
+	xor	$24,$9
+	sll	$9,$2,26
+	xor	$24,$8
+	xor	$10,$7			# Ch(e,f,g)
+	xor	$8,$9,$24			# Sigma1(e)
+
+	srl	$24,$25,2
+	addu	$23,$10
+	lw	$10,44($6)		# K[11]
+	sll	$9,$25,10
+	addu	$23,$8
+	srl	$8,$25,13
+	xor	$24,$9
+	sll	$9,$25,19
+	xor	$24,$8
+	srl	$8,$25,22
+	xor	$24,$9
+	sll	$9,$25,30
+	xor	$24,$8
+	sw	$19,44($29)	# offload to ring buffer
+	xor	$24,$9			# Sigma0(a)
+
+	or	$8,$25,$30
+	and	$9,$25,$30
+	and	$8,$31
+	or	$9,$8			# Maj(a,b,c)
+	addu	$23,$10			# +=K[11]
+	addu	$24,$9
+
+	addu	$1,$23
+	addu	$24,$23
+	lwl	$21,55($5)
+	lwr	$21,52($5)
+	srl	$9,$20,24		# byte swap(12)
+	srl	$10,$20,8
+	andi	$11,$20,0xFF00
+	sll	$20,$20,24
+	andi	$10,0xFF00
+	sll	$11,$11,8
+	or	$20,$9
+	or	$10,$11
+	or	$20,$10
+	addu	$8,$20,$7			# 12
+	srl	$7,$1,6
+	xor	$11,$2,$3
+	sll	$10,$1,7
+	and	$11,$1
+	srl	$9,$1,11
+	xor	$7,$10
+	sll	$10,$1,21
+	xor	$7,$9
+	srl	$9,$1,25
+	xor	$7,$10
+	sll	$10,$1,26
+	xor	$7,$9
+	xor	$11,$3			# Ch(e,f,g)
+	xor	$9,$10,$7			# Sigma1(e)
+
+	srl	$7,$24,2
+	addu	$8,$11
+	lw	$11,48($6)		# K[12]
+	sll	$10,$24,10
+	addu	$8,$9
+	srl	$9,$24,13
+	xor	$7,$10
+	sll	$10,$24,19
+	xor	$7,$9
+	srl	$9,$24,22
+	xor	$7,$10
+	sll	$10,$24,30
+	xor	$7,$9
+	sw	$20,48($29)	# offload to ring buffer
+	xor	$7,$10			# Sigma0(a)
+
+	or	$9,$24,$25
+	and	$10,$24,$25
+	and	$9,$30
+	or	$10,$9			# Maj(a,b,c)
+	addu	$8,$11			# +=K[12]
+	addu	$7,$10
+
+	addu	$31,$8
+	addu	$7,$8
+	lwl	$22,59($5)
+	lwr	$22,56($5)
+	srl	$10,$21,24		# byte swap(13)
+	srl	$11,$21,8
+	andi	$12,$21,0xFF00
+	sll	$21,$21,24
+	andi	$11,0xFF00
+	sll	$12,$12,8
+	or	$21,$10
+	or	$11,$12
+	or	$21,$11
+	addu	$9,$21,$3			# 13
+	srl	$3,$31,6
+	xor	$12,$1,$2
+	sll	$11,$31,7
+	and	$12,$31
+	srl	$10,$31,11
+	xor	$3,$11
+	sll	$11,$31,21
+	xor	$3,$10
+	srl	$10,$31,25
+	xor	$3,$11
+	sll	$11,$31,26
+	xor	$3,$10
+	xor	$12,$2			# Ch(e,f,g)
+	xor	$10,$11,$3			# Sigma1(e)
+
+	srl	$3,$7,2
+	addu	$9,$12
+	lw	$12,52($6)		# K[13]
+	sll	$11,$7,10
+	addu	$9,$10
+	srl	$10,$7,13
+	xor	$3,$11
+	sll	$11,$7,19
+	xor	$3,$10
+	srl	$10,$7,22
+	xor	$3,$11
+	sll	$11,$7,30
+	xor	$3,$10
+	sw	$21,52($29)	# offload to ring buffer
+	xor	$3,$11			# Sigma0(a)
+
+	or	$10,$7,$24
+	and	$11,$7,$24
+	and	$10,$25
+	or	$11,$10			# Maj(a,b,c)
+	addu	$9,$12			# +=K[13]
+	addu	$3,$11
+
+	addu	$30,$9
+	addu	$3,$9
+	lw	$8,0($29)	# prefetch from ring buffer
+	lwl	$23,63($5)
+	lwr	$23,60($5)
+	srl	$11,$22,24		# byte swap(14)
+	srl	$12,$22,8
+	andi	$13,$22,0xFF00
+	sll	$22,$22,24
+	andi	$12,0xFF00
+	sll	$13,$13,8
+	or	$22,$11
+	or	$12,$13
+	or	$22,$12
+	addu	$10,$22,$2			# 14
+	srl	$2,$30,6
+	xor	$13,$31,$1
+	sll	$12,$30,7
+	and	$13,$30
+	srl	$11,$30,11
+	xor	$2,$12
+	sll	$12,$30,21
+	xor	$2,$11
+	srl	$11,$30,25
+	xor	$2,$12
+	sll	$12,$30,26
+	xor	$2,$11
+	xor	$13,$1			# Ch(e,f,g)
+	xor	$11,$12,$2			# Sigma1(e)
+
+	srl	$2,$3,2
+	addu	$10,$13
+	lw	$13,56($6)		# K[14]
+	sll	$12,$3,10
+	addu	$10,$11
+	srl	$11,$3,13
+	xor	$2,$12
+	sll	$12,$3,19
+	xor	$2,$11
+	srl	$11,$3,22
+	xor	$2,$12
+	sll	$12,$3,30
+	xor	$2,$11
+	sw	$22,56($29)	# offload to ring buffer
+	xor	$2,$12			# Sigma0(a)
+
+	or	$11,$3,$7
+	and	$12,$3,$7
+	and	$11,$24
+	or	$12,$11			# Maj(a,b,c)
+	addu	$10,$13			# +=K[14]
+	addu	$2,$12
+
+	addu	$25,$10
+	addu	$2,$10
+	lw	$9,4($29)	# prefetch from ring buffer
+	srl	$12,$23,24		# byte swap(15)
+	srl	$13,$23,8
+	andi	$14,$23,0xFF00
+	sll	$23,$23,24
+	andi	$13,0xFF00
+	sll	$14,$14,8
+	or	$23,$12
+	or	$13,$14
+	or	$23,$13
+	addu	$11,$23,$1			# 15
+	srl	$1,$25,6
+	xor	$14,$30,$31
+	sll	$13,$25,7
+	and	$14,$25
+	srl	$12,$25,11
+	xor	$1,$13
+	sll	$13,$25,21
+	xor	$1,$12
+	srl	$12,$25,25
+	xor	$1,$13
+	sll	$13,$25,26
+	xor	$1,$12
+	xor	$14,$31			# Ch(e,f,g)
+	xor	$12,$13,$1			# Sigma1(e)
+
+	srl	$1,$2,2
+	addu	$11,$14
+	lw	$14,60($6)		# K[15]
+	sll	$13,$2,10
+	addu	$11,$12
+	srl	$12,$2,13
+	xor	$1,$13
+	sll	$13,$2,19
+	xor	$1,$12
+	srl	$12,$2,22
+	xor	$1,$13
+	sll	$13,$2,30
+	xor	$1,$12
+	sw	$23,60($29)	# offload to ring buffer
+	xor	$1,$13			# Sigma0(a)
+
+	or	$12,$2,$3
+	and	$13,$2,$3
+	and	$12,$7
+	or	$13,$12			# Maj(a,b,c)
+	addu	$11,$14			# +=K[15]
+	addu	$1,$13
+
+	addu	$24,$11
+	addu	$1,$11
+	lw	$10,8($29)	# prefetch from ring buffer
+	b	.L16_xx
+.align	4
+.L16_xx:
+	srl	$14,$9,3		# Xupdate(16)
+	addu	$8,$17			# +=X[i+9]
+	sll	$13,$9,14
+	srl	$12,$9,7
+	xor	$14,$13
+	sll	$13,11
+	xor	$14,$12
+	srl	$12,$9,18
+	xor	$14,$13
+
+	srl	$15,$22,10
+	xor	$14,$12			# sigma0(X[i+1])
+	sll	$13,$22,13
+	addu	$8,$14
+	srl	$12,$22,17
+	xor	$15,$13
+	sll	$13,2
+	xor	$15,$12
+	srl	$12,$22,19
+	xor	$15,$13
+
+	xor	$15,$12			# sigma1(X[i+14])
+	addu	$8,$15
+	addu	$12,$8,$31			# 16
+	srl	$31,$24,6
+	xor	$15,$25,$30
+	sll	$14,$24,7
+	and	$15,$24
+	srl	$13,$24,11
+	xor	$31,$14
+	sll	$14,$24,21
+	xor	$31,$13
+	srl	$13,$24,25
+	xor	$31,$14
+	sll	$14,$24,26
+	xor	$31,$13
+	xor	$15,$30			# Ch(e,f,g)
+	xor	$13,$14,$31			# Sigma1(e)
+
+	srl	$31,$1,2
+	addu	$12,$15
+	lw	$15,64($6)		# K[16]
+	sll	$14,$1,10
+	addu	$12,$13
+	srl	$13,$1,13
+	xor	$31,$14
+	sll	$14,$1,19
+	xor	$31,$13
+	srl	$13,$1,22
+	xor	$31,$14
+	sll	$14,$1,30
+	xor	$31,$13
+	sw	$8,0($29)	# offload to ring buffer
+	xor	$31,$14			# Sigma0(a)
+
+	or	$13,$1,$2
+	and	$14,$1,$2
+	and	$13,$3
+	or	$14,$13			# Maj(a,b,c)
+	addu	$12,$15			# +=K[16]
+	addu	$31,$14
+
+	addu	$7,$12
+	addu	$31,$12
+	lw	$11,12($29)	# prefetch from ring buffer
+	srl	$15,$10,3		# Xupdate(17)
+	addu	$9,$18			# +=X[i+9]
+	sll	$14,$10,14
+	srl	$13,$10,7
+	xor	$15,$14
+	sll	$14,11
+	xor	$15,$13
+	srl	$13,$10,18
+	xor	$15,$14
+
+	srl	$16,$23,10
+	xor	$15,$13			# sigma0(X[i+1])
+	sll	$14,$23,13
+	addu	$9,$15
+	srl	$13,$23,17
+	xor	$16,$14
+	sll	$14,2
+	xor	$16,$13
+	srl	$13,$23,19
+	xor	$16,$14
+
+	xor	$16,$13			# sigma1(X[i+14])
+	addu	$9,$16
+	addu	$13,$9,$30			# 17
+	srl	$30,$7,6
+	xor	$16,$24,$25
+	sll	$15,$7,7
+	and	$16,$7
+	srl	$14,$7,11
+	xor	$30,$15
+	sll	$15,$7,21
+	xor	$30,$14
+	srl	$14,$7,25
+	xor	$30,$15
+	sll	$15,$7,26
+	xor	$30,$14
+	xor	$16,$25			# Ch(e,f,g)
+	xor	$14,$15,$30			# Sigma1(e)
+
+	srl	$30,$31,2
+	addu	$13,$16
+	lw	$16,68($6)		# K[17]
+	sll	$15,$31,10
+	addu	$13,$14
+	srl	$14,$31,13
+	xor	$30,$15
+	sll	$15,$31,19
+	xor	$30,$14
+	srl	$14,$31,22
+	xor	$30,$15
+	sll	$15,$31,30
+	xor	$30,$14
+	sw	$9,4($29)	# offload to ring buffer
+	xor	$30,$15			# Sigma0(a)
+
+	or	$14,$31,$1
+	and	$15,$31,$1
+	and	$14,$2
+	or	$15,$14			# Maj(a,b,c)
+	addu	$13,$16			# +=K[17]
+	addu	$30,$15
+
+	addu	$3,$13
+	addu	$30,$13
+	lw	$12,16($29)	# prefetch from ring buffer
+	srl	$16,$11,3		# Xupdate(18)
+	addu	$10,$19			# +=X[i+9]
+	sll	$15,$11,14
+	srl	$14,$11,7
+	xor	$16,$15
+	sll	$15,11
+	xor	$16,$14
+	srl	$14,$11,18
+	xor	$16,$15
+
+	srl	$17,$8,10
+	xor	$16,$14			# sigma0(X[i+1])
+	sll	$15,$8,13
+	addu	$10,$16
+	srl	$14,$8,17
+	xor	$17,$15
+	sll	$15,2
+	xor	$17,$14
+	srl	$14,$8,19
+	xor	$17,$15
+
+	xor	$17,$14			# sigma1(X[i+14])
+	addu	$10,$17
+	addu	$14,$10,$25			# 18
+	srl	$25,$3,6
+	xor	$17,$7,$24
+	sll	$16,$3,7
+	and	$17,$3
+	srl	$15,$3,11
+	xor	$25,$16
+	sll	$16,$3,21
+	xor	$25,$15
+	srl	$15,$3,25
+	xor	$25,$16
+	sll	$16,$3,26
+	xor	$25,$15
+	xor	$17,$24			# Ch(e,f,g)
+	xor	$15,$16,$25			# Sigma1(e)
+
+	srl	$25,$30,2
+	addu	$14,$17
+	lw	$17,72($6)		# K[18]
+	sll	$16,$30,10
+	addu	$14,$15
+	srl	$15,$30,13
+	xor	$25,$16
+	sll	$16,$30,19
+	xor	$25,$15
+	srl	$15,$30,22
+	xor	$25,$16
+	sll	$16,$30,30
+	xor	$25,$15
+	sw	$10,8($29)	# offload to ring buffer
+	xor	$25,$16			# Sigma0(a)
+
+	or	$15,$30,$31
+	and	$16,$30,$31
+	and	$15,$1
+	or	$16,$15			# Maj(a,b,c)
+	addu	$14,$17			# +=K[18]
+	addu	$25,$16
+
+	addu	$2,$14
+	addu	$25,$14
+	lw	$13,20($29)	# prefetch from ring buffer
+	srl	$17,$12,3		# Xupdate(19)
+	addu	$11,$20			# +=X[i+9]
+	sll	$16,$12,14
+	srl	$15,$12,7
+	xor	$17,$16
+	sll	$16,11
+	xor	$17,$15
+	srl	$15,$12,18
+	xor	$17,$16
+
+	srl	$18,$9,10
+	xor	$17,$15			# sigma0(X[i+1])
+	sll	$16,$9,13
+	addu	$11,$17
+	srl	$15,$9,17
+	xor	$18,$16
+	sll	$16,2
+	xor	$18,$15
+	srl	$15,$9,19
+	xor	$18,$16
+
+	xor	$18,$15			# sigma1(X[i+14])
+	addu	$11,$18
+	addu	$15,$11,$24			# 19
+	srl	$24,$2,6
+	xor	$18,$3,$7
+	sll	$17,$2,7
+	and	$18,$2
+	srl	$16,$2,11
+	xor	$24,$17
+	sll	$17,$2,21
+	xor	$24,$16
+	srl	$16,$2,25
+	xor	$24,$17
+	sll	$17,$2,26
+	xor	$24,$16
+	xor	$18,$7			# Ch(e,f,g)
+	xor	$16,$17,$24			# Sigma1(e)
+
+	srl	$24,$25,2
+	addu	$15,$18
+	lw	$18,76($6)		# K[19]
+	sll	$17,$25,10
+	addu	$15,$16
+	srl	$16,$25,13
+	xor	$24,$17
+	sll	$17,$25,19
+	xor	$24,$16
+	srl	$16,$25,22
+	xor	$24,$17
+	sll	$17,$25,30
+	xor	$24,$16
+	sw	$11,12($29)	# offload to ring buffer
+	xor	$24,$17			# Sigma0(a)
+
+	or	$16,$25,$30
+	and	$17,$25,$30
+	and	$16,$31
+	or	$17,$16			# Maj(a,b,c)
+	addu	$15,$18			# +=K[19]
+	addu	$24,$17
+
+	addu	$1,$15
+	addu	$24,$15
+	lw	$14,24($29)	# prefetch from ring buffer
+	srl	$18,$13,3		# Xupdate(20)
+	addu	$12,$21			# +=X[i+9]
+	sll	$17,$13,14
+	srl	$16,$13,7
+	xor	$18,$17
+	sll	$17,11
+	xor	$18,$16
+	srl	$16,$13,18
+	xor	$18,$17
+
+	srl	$19,$10,10
+	xor	$18,$16			# sigma0(X[i+1])
+	sll	$17,$10,13
+	addu	$12,$18
+	srl	$16,$10,17
+	xor	$19,$17
+	sll	$17,2
+	xor	$19,$16
+	srl	$16,$10,19
+	xor	$19,$17
+
+	xor	$19,$16			# sigma1(X[i+14])
+	addu	$12,$19
+	addu	$16,$12,$7			# 20
+	srl	$7,$1,6
+	xor	$19,$2,$3
+	sll	$18,$1,7
+	and	$19,$1
+	srl	$17,$1,11
+	xor	$7,$18
+	sll	$18,$1,21
+	xor	$7,$17
+	srl	$17,$1,25
+	xor	$7,$18
+	sll	$18,$1,26
+	xor	$7,$17
+	xor	$19,$3			# Ch(e,f,g)
+	xor	$17,$18,$7			# Sigma1(e)
+
+	srl	$7,$24,2
+	addu	$16,$19
+	lw	$19,80($6)		# K[20]
+	sll	$18,$24,10
+	addu	$16,$17
+	srl	$17,$24,13
+	xor	$7,$18
+	sll	$18,$24,19
+	xor	$7,$17
+	srl	$17,$24,22
+	xor	$7,$18
+	sll	$18,$24,30
+	xor	$7,$17
+	sw	$12,16($29)	# offload to ring buffer
+	xor	$7,$18			# Sigma0(a)
+
+	or	$17,$24,$25
+	and	$18,$24,$25
+	and	$17,$30
+	or	$18,$17			# Maj(a,b,c)
+	addu	$16,$19			# +=K[20]
+	addu	$7,$18
+
+	addu	$31,$16
+	addu	$7,$16
+	lw	$15,28($29)	# prefetch from ring buffer
+	srl	$19,$14,3		# Xupdate(21)
+	addu	$13,$22			# +=X[i+9]
+	sll	$18,$14,14
+	srl	$17,$14,7
+	xor	$19,$18
+	sll	$18,11
+	xor	$19,$17
+	srl	$17,$14,18
+	xor	$19,$18
+
+	srl	$20,$11,10
+	xor	$19,$17			# sigma0(X[i+1])
+	sll	$18,$11,13
+	addu	$13,$19
+	srl	$17,$11,17
+	xor	$20,$18
+	sll	$18,2
+	xor	$20,$17
+	srl	$17,$11,19
+	xor	$20,$18
+
+	xor	$20,$17			# sigma1(X[i+14])
+	addu	$13,$20
+	addu	$17,$13,$3			# 21
+	srl	$3,$31,6
+	xor	$20,$1,$2
+	sll	$19,$31,7
+	and	$20,$31
+	srl	$18,$31,11
+	xor	$3,$19
+	sll	$19,$31,21
+	xor	$3,$18
+	srl	$18,$31,25
+	xor	$3,$19
+	sll	$19,$31,26
+	xor	$3,$18
+	xor	$20,$2			# Ch(e,f,g)
+	xor	$18,$19,$3			# Sigma1(e)
+
+	srl	$3,$7,2
+	addu	$17,$20
+	lw	$20,84($6)		# K[21]
+	sll	$19,$7,10
+	addu	$17,$18
+	srl	$18,$7,13
+	xor	$3,$19
+	sll	$19,$7,19
+	xor	$3,$18
+	srl	$18,$7,22
+	xor	$3,$19
+	sll	$19,$7,30
+	xor	$3,$18
+	sw	$13,20($29)	# offload to ring buffer
+	xor	$3,$19			# Sigma0(a)
+
+	or	$18,$7,$24
+	and	$19,$7,$24
+	and	$18,$25
+	or	$19,$18			# Maj(a,b,c)
+	addu	$17,$20			# +=K[21]
+	addu	$3,$19
+
+	addu	$30,$17
+	addu	$3,$17
+	lw	$16,32($29)	# prefetch from ring buffer
+	srl	$20,$15,3		# Xupdate(22)
+	addu	$14,$23			# +=X[i+9]
+	sll	$19,$15,14
+	srl	$18,$15,7
+	xor	$20,$19
+	sll	$19,11
+	xor	$20,$18
+	srl	$18,$15,18
+	xor	$20,$19
+
+	srl	$21,$12,10
+	xor	$20,$18			# sigma0(X[i+1])
+	sll	$19,$12,13
+	addu	$14,$20
+	srl	$18,$12,17
+	xor	$21,$19
+	sll	$19,2
+	xor	$21,$18
+	srl	$18,$12,19
+	xor	$21,$19
+
+	xor	$21,$18			# sigma1(X[i+14])
+	addu	$14,$21
+	addu	$18,$14,$2			# 22
+	srl	$2,$30,6
+	xor	$21,$31,$1
+	sll	$20,$30,7
+	and	$21,$30
+	srl	$19,$30,11
+	xor	$2,$20
+	sll	$20,$30,21
+	xor	$2,$19
+	srl	$19,$30,25
+	xor	$2,$20
+	sll	$20,$30,26
+	xor	$2,$19
+	xor	$21,$1			# Ch(e,f,g)
+	xor	$19,$20,$2			# Sigma1(e)
+
+	srl	$2,$3,2
+	addu	$18,$21
+	lw	$21,88($6)		# K[22]
+	sll	$20,$3,10
+	addu	$18,$19
+	srl	$19,$3,13
+	xor	$2,$20
+	sll	$20,$3,19
+	xor	$2,$19
+	srl	$19,$3,22
+	xor	$2,$20
+	sll	$20,$3,30
+	xor	$2,$19
+	sw	$14,24($29)	# offload to ring buffer
+	xor	$2,$20			# Sigma0(a)
+
+	or	$19,$3,$7
+	and	$20,$3,$7
+	and	$19,$24
+	or	$20,$19			# Maj(a,b,c)
+	addu	$18,$21			# +=K[22]
+	addu	$2,$20
+
+	addu	$25,$18
+	addu	$2,$18
+	lw	$17,36($29)	# prefetch from ring buffer
+	srl	$21,$16,3		# Xupdate(23)
+	addu	$15,$8			# +=X[i+9]
+	sll	$20,$16,14
+	srl	$19,$16,7
+	xor	$21,$20
+	sll	$20,11
+	xor	$21,$19
+	srl	$19,$16,18
+	xor	$21,$20
+
+	srl	$22,$13,10
+	xor	$21,$19			# sigma0(X[i+1])
+	sll	$20,$13,13
+	addu	$15,$21
+	srl	$19,$13,17
+	xor	$22,$20
+	sll	$20,2
+	xor	$22,$19
+	srl	$19,$13,19
+	xor	$22,$20
+
+	xor	$22,$19			# sigma1(X[i+14])
+	addu	$15,$22
+	addu	$19,$15,$1			# 23
+	srl	$1,$25,6
+	xor	$22,$30,$31
+	sll	$21,$25,7
+	and	$22,$25
+	srl	$20,$25,11
+	xor	$1,$21
+	sll	$21,$25,21
+	xor	$1,$20
+	srl	$20,$25,25
+	xor	$1,$21
+	sll	$21,$25,26
+	xor	$1,$20
+	xor	$22,$31			# Ch(e,f,g)
+	xor	$20,$21,$1			# Sigma1(e)
+
+	srl	$1,$2,2
+	addu	$19,$22
+	lw	$22,92($6)		# K[23]
+	sll	$21,$2,10
+	addu	$19,$20
+	srl	$20,$2,13
+	xor	$1,$21
+	sll	$21,$2,19
+	xor	$1,$20
+	srl	$20,$2,22
+	xor	$1,$21
+	sll	$21,$2,30
+	xor	$1,$20
+	sw	$15,28($29)	# offload to ring buffer
+	xor	$1,$21			# Sigma0(a)
+
+	or	$20,$2,$3
+	and	$21,$2,$3
+	and	$20,$7
+	or	$21,$20			# Maj(a,b,c)
+	addu	$19,$22			# +=K[23]
+	addu	$1,$21
+
+	addu	$24,$19
+	addu	$1,$19
+	lw	$18,40($29)	# prefetch from ring buffer
+	srl	$22,$17,3		# Xupdate(24)
+	addu	$16,$9			# +=X[i+9]
+	sll	$21,$17,14
+	srl	$20,$17,7
+	xor	$22,$21
+	sll	$21,11
+	xor	$22,$20
+	srl	$20,$17,18
+	xor	$22,$21
+
+	srl	$23,$14,10
+	xor	$22,$20			# sigma0(X[i+1])
+	sll	$21,$14,13
+	addu	$16,$22
+	srl	$20,$14,17
+	xor	$23,$21
+	sll	$21,2
+	xor	$23,$20
+	srl	$20,$14,19
+	xor	$23,$21
+
+	xor	$23,$20			# sigma1(X[i+14])
+	addu	$16,$23
+	addu	$20,$16,$31			# 24
+	srl	$31,$24,6
+	xor	$23,$25,$30
+	sll	$22,$24,7
+	and	$23,$24
+	srl	$21,$24,11
+	xor	$31,$22
+	sll	$22,$24,21
+	xor	$31,$21
+	srl	$21,$24,25
+	xor	$31,$22
+	sll	$22,$24,26
+	xor	$31,$21
+	xor	$23,$30			# Ch(e,f,g)
+	xor	$21,$22,$31			# Sigma1(e)
+
+	srl	$31,$1,2
+	addu	$20,$23
+	lw	$23,96($6)		# K[24]
+	sll	$22,$1,10
+	addu	$20,$21
+	srl	$21,$1,13
+	xor	$31,$22
+	sll	$22,$1,19
+	xor	$31,$21
+	srl	$21,$1,22
+	xor	$31,$22
+	sll	$22,$1,30
+	xor	$31,$21
+	sw	$16,32($29)	# offload to ring buffer
+	xor	$31,$22			# Sigma0(a)
+
+	or	$21,$1,$2
+	and	$22,$1,$2
+	and	$21,$3
+	or	$22,$21			# Maj(a,b,c)
+	addu	$20,$23			# +=K[24]
+	addu	$31,$22
+
+	addu	$7,$20
+	addu	$31,$20
+	lw	$19,44($29)	# prefetch from ring buffer
+	srl	$23,$18,3		# Xupdate(25)
+	addu	$17,$10			# +=X[i+9]
+	sll	$22,$18,14
+	srl	$21,$18,7
+	xor	$23,$22
+	sll	$22,11
+	xor	$23,$21
+	srl	$21,$18,18
+	xor	$23,$22
+
+	srl	$8,$15,10
+	xor	$23,$21			# sigma0(X[i+1])
+	sll	$22,$15,13
+	addu	$17,$23
+	srl	$21,$15,17
+	xor	$8,$22
+	sll	$22,2
+	xor	$8,$21
+	srl	$21,$15,19
+	xor	$8,$22
+
+	xor	$8,$21			# sigma1(X[i+14])
+	addu	$17,$8
+	addu	$21,$17,$30			# 25
+	srl	$30,$7,6
+	xor	$8,$24,$25
+	sll	$23,$7,7
+	and	$8,$7
+	srl	$22,$7,11
+	xor	$30,$23
+	sll	$23,$7,21
+	xor	$30,$22
+	srl	$22,$7,25
+	xor	$30,$23
+	sll	$23,$7,26
+	xor	$30,$22
+	xor	$8,$25			# Ch(e,f,g)
+	xor	$22,$23,$30			# Sigma1(e)
+
+	srl	$30,$31,2
+	addu	$21,$8
+	lw	$8,100($6)		# K[25]
+	sll	$23,$31,10
+	addu	$21,$22
+	srl	$22,$31,13
+	xor	$30,$23
+	sll	$23,$31,19
+	xor	$30,$22
+	srl	$22,$31,22
+	xor	$30,$23
+	sll	$23,$31,30
+	xor	$30,$22
+	sw	$17,36($29)	# offload to ring buffer
+	xor	$30,$23			# Sigma0(a)
+
+	or	$22,$31,$1
+	and	$23,$31,$1
+	and	$22,$2
+	or	$23,$22			# Maj(a,b,c)
+	addu	$21,$8			# +=K[25]
+	addu	$30,$23
+
+	addu	$3,$21
+	addu	$30,$21
+	lw	$20,48($29)	# prefetch from ring buffer
+	srl	$8,$19,3		# Xupdate(26)
+	addu	$18,$11			# +=X[i+9]
+	sll	$23,$19,14
+	srl	$22,$19,7
+	xor	$8,$23
+	sll	$23,11
+	xor	$8,$22
+	srl	$22,$19,18
+	xor	$8,$23
+
+	srl	$9,$16,10
+	xor	$8,$22			# sigma0(X[i+1])
+	sll	$23,$16,13
+	addu	$18,$8
+	srl	$22,$16,17
+	xor	$9,$23
+	sll	$23,2
+	xor	$9,$22
+	srl	$22,$16,19
+	xor	$9,$23
+
+	xor	$9,$22			# sigma1(X[i+14])
+	addu	$18,$9
+	addu	$22,$18,$25			# 26
+	srl	$25,$3,6
+	xor	$9,$7,$24
+	sll	$8,$3,7
+	and	$9,$3
+	srl	$23,$3,11
+	xor	$25,$8
+	sll	$8,$3,21
+	xor	$25,$23
+	srl	$23,$3,25
+	xor	$25,$8
+	sll	$8,$3,26
+	xor	$25,$23
+	xor	$9,$24			# Ch(e,f,g)
+	xor	$23,$8,$25			# Sigma1(e)
+
+	srl	$25,$30,2
+	addu	$22,$9
+	lw	$9,104($6)		# K[26]
+	sll	$8,$30,10
+	addu	$22,$23
+	srl	$23,$30,13
+	xor	$25,$8
+	sll	$8,$30,19
+	xor	$25,$23
+	srl	$23,$30,22
+	xor	$25,$8
+	sll	$8,$30,30
+	xor	$25,$23
+	sw	$18,40($29)	# offload to ring buffer
+	xor	$25,$8			# Sigma0(a)
+
+	or	$23,$30,$31
+	and	$8,$30,$31
+	and	$23,$1
+	or	$8,$23			# Maj(a,b,c)
+	addu	$22,$9			# +=K[26]
+	addu	$25,$8
+
+	addu	$2,$22
+	addu	$25,$22
+	lw	$21,52($29)	# prefetch from ring buffer
+	srl	$9,$20,3		# Xupdate(27)
+	addu	$19,$12			# +=X[i+9]
+	sll	$8,$20,14
+	srl	$23,$20,7
+	xor	$9,$8
+	sll	$8,11
+	xor	$9,$23
+	srl	$23,$20,18
+	xor	$9,$8
+
+	srl	$10,$17,10
+	xor	$9,$23			# sigma0(X[i+1])
+	sll	$8,$17,13
+	addu	$19,$9
+	srl	$23,$17,17
+	xor	$10,$8
+	sll	$8,2
+	xor	$10,$23
+	srl	$23,$17,19
+	xor	$10,$8
+
+	xor	$10,$23			# sigma1(X[i+14])
+	addu	$19,$10
+	addu	$23,$19,$24			# 27
+	srl	$24,$2,6
+	xor	$10,$3,$7
+	sll	$9,$2,7
+	and	$10,$2
+	srl	$8,$2,11
+	xor	$24,$9
+	sll	$9,$2,21
+	xor	$24,$8
+	srl	$8,$2,25
+	xor	$24,$9
+	sll	$9,$2,26
+	xor	$24,$8
+	xor	$10,$7			# Ch(e,f,g)
+	xor	$8,$9,$24			# Sigma1(e)
+
+	srl	$24,$25,2
+	addu	$23,$10
+	lw	$10,108($6)		# K[27]
+	sll	$9,$25,10
+	addu	$23,$8
+	srl	$8,$25,13
+	xor	$24,$9
+	sll	$9,$25,19
+	xor	$24,$8
+	srl	$8,$25,22
+	xor	$24,$9
+	sll	$9,$25,30
+	xor	$24,$8
+	sw	$19,44($29)	# offload to ring buffer
+	xor	$24,$9			# Sigma0(a)
+
+	or	$8,$25,$30
+	and	$9,$25,$30
+	and	$8,$31
+	or	$9,$8			# Maj(a,b,c)
+	addu	$23,$10			# +=K[27]
+	addu	$24,$9
+
+	addu	$1,$23
+	addu	$24,$23
+	lw	$22,56($29)	# prefetch from ring buffer
+	srl	$10,$21,3		# Xupdate(28)
+	addu	$20,$13			# +=X[i+9]
+	sll	$9,$21,14
+	srl	$8,$21,7
+	xor	$10,$9
+	sll	$9,11
+	xor	$10,$8
+	srl	$8,$21,18
+	xor	$10,$9
+
+	srl	$11,$18,10
+	xor	$10,$8			# sigma0(X[i+1])
+	sll	$9,$18,13
+	addu	$20,$10
+	srl	$8,$18,17
+	xor	$11,$9
+	sll	$9,2
+	xor	$11,$8
+	srl	$8,$18,19
+	xor	$11,$9
+
+	xor	$11,$8			# sigma1(X[i+14])
+	addu	$20,$11
+	addu	$8,$20,$7			# 28
+	srl	$7,$1,6
+	xor	$11,$2,$3
+	sll	$10,$1,7
+	and	$11,$1
+	srl	$9,$1,11
+	xor	$7,$10
+	sll	$10,$1,21
+	xor	$7,$9
+	srl	$9,$1,25
+	xor	$7,$10
+	sll	$10,$1,26
+	xor	$7,$9
+	xor	$11,$3			# Ch(e,f,g)
+	xor	$9,$10,$7			# Sigma1(e)
+
+	srl	$7,$24,2
+	addu	$8,$11
+	lw	$11,112($6)		# K[28]
+	sll	$10,$24,10
+	addu	$8,$9
+	srl	$9,$24,13
+	xor	$7,$10
+	sll	$10,$24,19
+	xor	$7,$9
+	srl	$9,$24,22
+	xor	$7,$10
+	sll	$10,$24,30
+	xor	$7,$9
+	sw	$20,48($29)	# offload to ring buffer
+	xor	$7,$10			# Sigma0(a)
+
+	or	$9,$24,$25
+	and	$10,$24,$25
+	and	$9,$30
+	or	$10,$9			# Maj(a,b,c)
+	addu	$8,$11			# +=K[28]
+	addu	$7,$10
+
+	addu	$31,$8
+	addu	$7,$8
+	lw	$23,60($29)	# prefetch from ring buffer
+	srl	$11,$22,3		# Xupdate(29)
+	addu	$21,$14			# +=X[i+9]
+	sll	$10,$22,14
+	srl	$9,$22,7
+	xor	$11,$10
+	sll	$10,11
+	xor	$11,$9
+	srl	$9,$22,18
+	xor	$11,$10
+
+	srl	$12,$19,10
+	xor	$11,$9			# sigma0(X[i+1])
+	sll	$10,$19,13
+	addu	$21,$11
+	srl	$9,$19,17
+	xor	$12,$10
+	sll	$10,2
+	xor	$12,$9
+	srl	$9,$19,19
+	xor	$12,$10
+
+	xor	$12,$9			# sigma1(X[i+14])
+	addu	$21,$12
+	addu	$9,$21,$3			# 29
+	srl	$3,$31,6
+	xor	$12,$1,$2
+	sll	$11,$31,7
+	and	$12,$31
+	srl	$10,$31,11
+	xor	$3,$11
+	sll	$11,$31,21
+	xor	$3,$10
+	srl	$10,$31,25
+	xor	$3,$11
+	sll	$11,$31,26
+	xor	$3,$10
+	xor	$12,$2			# Ch(e,f,g)
+	xor	$10,$11,$3			# Sigma1(e)
+
+	srl	$3,$7,2
+	addu	$9,$12
+	lw	$12,116($6)		# K[29]
+	sll	$11,$7,10
+	addu	$9,$10
+	srl	$10,$7,13
+	xor	$3,$11
+	sll	$11,$7,19
+	xor	$3,$10
+	srl	$10,$7,22
+	xor	$3,$11
+	sll	$11,$7,30
+	xor	$3,$10
+	sw	$21,52($29)	# offload to ring buffer
+	xor	$3,$11			# Sigma0(a)
+
+	or	$10,$7,$24
+	and	$11,$7,$24
+	and	$10,$25
+	or	$11,$10			# Maj(a,b,c)
+	addu	$9,$12			# +=K[29]
+	addu	$3,$11
+
+	addu	$30,$9
+	addu	$3,$9
+	lw	$8,0($29)	# prefetch from ring buffer
+	srl	$12,$23,3		# Xupdate(30)
+	addu	$22,$15			# +=X[i+9]
+	sll	$11,$23,14
+	srl	$10,$23,7
+	xor	$12,$11
+	sll	$11,11
+	xor	$12,$10
+	srl	$10,$23,18
+	xor	$12,$11
+
+	srl	$13,$20,10
+	xor	$12,$10			# sigma0(X[i+1])
+	sll	$11,$20,13
+	addu	$22,$12
+	srl	$10,$20,17
+	xor	$13,$11
+	sll	$11,2
+	xor	$13,$10
+	srl	$10,$20,19
+	xor	$13,$11
+
+	xor	$13,$10			# sigma1(X[i+14])
+	addu	$22,$13
+	addu	$10,$22,$2			# 30
+	srl	$2,$30,6
+	xor	$13,$31,$1
+	sll	$12,$30,7
+	and	$13,$30
+	srl	$11,$30,11
+	xor	$2,$12
+	sll	$12,$30,21
+	xor	$2,$11
+	srl	$11,$30,25
+	xor	$2,$12
+	sll	$12,$30,26
+	xor	$2,$11
+	xor	$13,$1			# Ch(e,f,g)
+	xor	$11,$12,$2			# Sigma1(e)
+
+	srl	$2,$3,2
+	addu	$10,$13
+	lw	$13,120($6)		# K[30]
+	sll	$12,$3,10
+	addu	$10,$11
+	srl	$11,$3,13
+	xor	$2,$12
+	sll	$12,$3,19
+	xor	$2,$11
+	srl	$11,$3,22
+	xor	$2,$12
+	sll	$12,$3,30
+	xor	$2,$11
+	sw	$22,56($29)	# offload to ring buffer
+	xor	$2,$12			# Sigma0(a)
+
+	or	$11,$3,$7
+	and	$12,$3,$7
+	and	$11,$24
+	or	$12,$11			# Maj(a,b,c)
+	addu	$10,$13			# +=K[30]
+	addu	$2,$12
+
+	addu	$25,$10
+	addu	$2,$10
+	lw	$9,4($29)	# prefetch from ring buffer
+	srl	$13,$8,3		# Xupdate(31)
+	addu	$23,$16			# +=X[i+9]
+	sll	$12,$8,14
+	srl	$11,$8,7
+	xor	$13,$12
+	sll	$12,11
+	xor	$13,$11
+	srl	$11,$8,18
+	xor	$13,$12
+
+	srl	$14,$21,10
+	xor	$13,$11			# sigma0(X[i+1])
+	sll	$12,$21,13
+	addu	$23,$13
+	srl	$11,$21,17
+	xor	$14,$12
+	sll	$12,2
+	xor	$14,$11
+	srl	$11,$21,19
+	xor	$14,$12
+
+	xor	$14,$11			# sigma1(X[i+14])
+	addu	$23,$14
+	addu	$11,$23,$1			# 31
+	srl	$1,$25,6
+	xor	$14,$30,$31
+	sll	$13,$25,7
+	and	$14,$25
+	srl	$12,$25,11
+	xor	$1,$13
+	sll	$13,$25,21
+	xor	$1,$12
+	srl	$12,$25,25
+	xor	$1,$13
+	sll	$13,$25,26
+	xor	$1,$12
+	xor	$14,$31			# Ch(e,f,g)
+	xor	$12,$13,$1			# Sigma1(e)
+
+	srl	$1,$2,2
+	addu	$11,$14
+	lw	$14,124($6)		# K[31]
+	sll	$13,$2,10
+	addu	$11,$12
+	srl	$12,$2,13
+	xor	$1,$13
+	sll	$13,$2,19
+	xor	$1,$12
+	srl	$12,$2,22
+	xor	$1,$13
+	sll	$13,$2,30
+	xor	$1,$12
+	sw	$23,60($29)	# offload to ring buffer
+	xor	$1,$13			# Sigma0(a)
+
+	or	$12,$2,$3
+	and	$13,$2,$3
+	and	$12,$7
+	or	$13,$12			# Maj(a,b,c)
+	addu	$11,$14			# +=K[31]
+	addu	$1,$13
+
+	addu	$24,$11
+	addu	$1,$11
+	lw	$10,8($29)	# prefetch from ring buffer
+	and	$14,0xfff
+	li	$15,2290
+	.set	noreorder
+	bne	$14,$15,.L16_xx
+	add $6,16*4		# Ktbl+=16
+
+	lw	$23,16*4($29)	# restore pointer to the end of input
+	lw	$8,0*4($4)
+	lw	$9,1*4($4)
+	lw	$10,2*4($4)
+	add $5,16*4
+	lw	$11,3*4($4)
+	addu	$1,$8
+	lw	$12,4*4($4)
+	addu	$2,$9
+	lw	$13,5*4($4)
+	addu	$3,$10
+	lw	$14,6*4($4)
+	addu	$7,$11
+	lw	$15,7*4($4)
+	addu	$24,$12
+	sw	$1,0*4($4)
+	addu	$25,$13
+	sw	$2,1*4($4)
+	addu	$30,$14
+	sw	$3,2*4($4)
+	addu	$31,$15
+	sw	$7,3*4($4)
+	sw	$24,4*4($4)
+	sw	$25,5*4($4)
+	sw	$30,6*4($4)
+	sw	$31,7*4($4)
+
+	bne	$5,$23,.Loop
+	sub $6,192	# rewind $6
+
+	lw	$31,128-1*4($29)
+	lw	$30,128-2*4($29)
+	lw	$23,128-3*4($29)
+	lw	$22,128-4*4($29)
+	lw	$21,128-5*4($29)
+	lw	$20,128-6*4($29)
+	lw	$19,128-7*4($29)
+	lw	$18,128-8*4($29)
+	lw	$17,128-9*4($29)
+	lw	$16,128-10*4($29)
+	jr	$31
+	add $29,128
+.end	sha256_block_data_order
+
+.rdata
+.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
+.asciiz	"SHA256 for MIPS, CRYPTOGAMS by <appro@openssl.org>"
+.align	5
+