Merge branch 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6

Pull crypto update from Herbert Xu: "API: - Try to catch hash output overrun in testmgr - Introduce walksize attribute for batched walking - Make crypto_xor() and crypto_inc() alignment agnostic Algorithms: - Add time-invariant AES algorithm - Add standalone CBCMAC algorithm Drivers: - Add NEON acclerated chacha20 on ARM/ARM64 - Expose AES-CTR as synchronous skcipher on ARM64 - Add scalar AES implementation on ARM64 - Improve scalar AES implementation on ARM - Improve NEON AES implementation on ARM/ARM64 - Merge CRC32 and PMULL instruction based drivers on ARM64 - Add NEON acclerated CBCMAC/CMAC/XCBC AES on ARM64 - Add IPsec AUTHENC implementation in atmel - Add Support for Octeon-tx CPT Engine - Add Broadcom SPU driver - Add MediaTek driver" * 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (142 commits) crypto: xts - Add ECB dependency crypto: cavium - switch to pci_alloc_irq_vectors crypto: cavium - switch to pci_alloc_irq_vectors crypto: cavium - remove dead MSI-X related define crypto: brcm - Avoid double free in ahash_finup() crypto: cavium - fix Kconfig dependencies crypto: cavium - cpt_bind_vq_to_grp could return an error code crypto: doc - fix typo hwrng: omap - update Kconfig help description crypto: ccm - drop unnecessary minimum 32-bit alignment crypto: ccm - honour alignmask of subordinate MAC cipher crypto: caam - fix state buffer DMA (un)mapping crypto: caam - abstract ahash request double buffering crypto: caam - fix error path for ctx_dma mapping failure crypto: caam - fix DMA API leaks for multiple setkey() calls crypto: caam - don't dma_map key for hash algorithms crypto: caam - use dma_map_sg() return code crypto: caam - replace sg_count() with sg_nents_for_len() crypto: caam - check sg_count() return value crypto: caam - fix HW S/G in ablkcipher_giv_edesc_alloc() ..
author: Linus Torvalds <torvalds@linux-foundation.org> 2017-02-23 09:54:19 -0800
committer: Linus Torvalds <torvalds@linux-foundation.org> 2017-02-23 09:54:19 -0800
commit: 5bcbe22ca47da04cda3a858cef67f55b550c1d13 (patch)
tree: 49bd61e32eb2d652085a49182436322a3e0e9840 /arch
parent: 1db934a5b77a9e37c4742c704fde6af233187a98 (diff)
parent: 12cb3a1c4184f891d965d1f39f8cfcc9ef617647 (diff)
68 files changed, 5337 insertions, 7272 deletions
diff --git a/arch/arm/crypto/Kconfig b/arch/arm/crypto/Kconfig
index 13f1b4c289d4..a8fce93137fb 100644
--- a/arch/arm/crypto/Kconfig
+++ b/arch/arm/crypto/Kconfig
@@ -62,35 +62,18 @@ config CRYPTO_SHA512_ARM
 	  using optimized ARM assembler and NEON, when available.
 
 config CRYPTO_AES_ARM
-	tristate "AES cipher algorithms (ARM-asm)"
-	depends on ARM
+	tristate "Scalar AES cipher for ARM"
 	select CRYPTO_ALGAPI
 	select CRYPTO_AES
 	help
 	  Use optimized AES assembler routines for ARM platforms.
 
-	  AES cipher algorithms (FIPS-197). AES uses the Rijndael
-	  algorithm.
-
-	  Rijndael appears to be consistently a very good performer in
-	  both hardware and software across a wide range of computing
-	  environments regardless of its use in feedback or non-feedback
-	  modes. Its key setup time is excellent, and its key agility is
-	  good. Rijndael's very low memory requirements make it very well
-	  suited for restricted-space environments, in which it also
-	  demonstrates excellent performance. Rijndael's operations are
-	  among the easiest to defend against power and timing attacks.
-
-	  The AES specifies three key sizes: 128, 192 and 256 bits
-
-	  See <http://csrc.nist.gov/encryption/aes/> for more information.
-
 config CRYPTO_AES_ARM_BS
 	tristate "Bit sliced AES using NEON instructions"
 	depends on KERNEL_MODE_NEON
-	select CRYPTO_AES_ARM
 	select CRYPTO_BLKCIPHER
 	select CRYPTO_SIMD
+	select CRYPTO_AES_ARM
 	help
 	  Use a faster and more secure NEON based implementation of AES in CBC,
 	  CTR and XTS modes
@@ -130,4 +113,10 @@ config CRYPTO_CRC32_ARM_CE
 	depends on KERNEL_MODE_NEON && CRC32
 	select CRYPTO_HASH
 
+config CRYPTO_CHACHA20_NEON
+	tristate "NEON accelerated ChaCha20 symmetric cipher"
+	depends on KERNEL_MODE_NEON
+	select CRYPTO_BLKCIPHER
+	select CRYPTO_CHACHA20
+
 endif
diff --git a/arch/arm/crypto/Makefile b/arch/arm/crypto/Makefile
index b578a1820ab1..1822c4697278 100644
--- a/arch/arm/crypto/Makefile
+++ b/arch/arm/crypto/Makefile
@@ -8,6 +8,7 @@ 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) += sha512-arm.o
+obj-$(CONFIG_CRYPTO_CHACHA20_NEON) += chacha20-neon.o
 
 ce-obj-$(CONFIG_CRYPTO_AES_ARM_CE) += aes-arm-ce.o
 ce-obj-$(CONFIG_CRYPTO_SHA1_ARM_CE) += sha1-arm-ce.o
@@ -26,8 +27,8 @@ $(warning $(ce-obj-y) $(ce-obj-m))
 endif
 endif
 
-aes-arm-y	:= aes-armv4.o aes_glue.o
-aes-arm-bs-y	:= aesbs-core.o aesbs-glue.o
+aes-arm-y	:= aes-cipher-core.o aes-cipher-glue.o
+aes-arm-bs-y	:= aes-neonbs-core.o aes-neonbs-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
@@ -40,17 +41,15 @@ aes-arm-ce-y	:= aes-ce-core.o aes-ce-glue.o
 ghash-arm-ce-y	:= ghash-ce-core.o ghash-ce-glue.o
 crct10dif-arm-ce-y	:= crct10dif-ce-core.o crct10dif-ce-glue.o
 crc32-arm-ce-y:= crc32-ce-core.o crc32-ce-glue.o
+chacha20-neon-y := chacha20-neon-core.o chacha20-neon-glue.o
 
 quiet_cmd_perl = PERL    $@
       cmd_perl = $(PERL) $(<) > $(@)
 
-$(src)/aesbs-core.S_shipped: $(src)/bsaes-armv7.pl
-	$(call cmd,perl)
-
 $(src)/sha256-core.S_shipped: $(src)/sha256-armv4.pl
 	$(call cmd,perl)
 
 $(src)/sha512-core.S_shipped: $(src)/sha512-armv4.pl
 	$(call cmd,perl)
 
-.PRECIOUS: $(obj)/aesbs-core.S $(obj)/sha256-core.S $(obj)/sha512-core.S
+.PRECIOUS: $(obj)/sha256-core.S $(obj)/sha512-core.S
diff --git a/arch/arm/crypto/aes-armv4.S b/arch/arm/crypto/aes-armv4.S
deleted file mode 100644
index ebb9761fb572..000000000000
--- a/arch/arm/crypto/aes-armv4.S
+++ /dev/null
@@ -1,1089 +0,0 @@
-#define __ARM_ARCH__ __LINUX_ARM_ARCH__
-@ ====================================================================
-@ Written by Andy Polyakov <appro@fy.chalmers.se> for the OpenSSL
-@ project. The module is, however, dual licensed under OpenSSL and
-@ CRYPTOGAMS licenses depending on where you obtain it. For further
-@ details see http://www.openssl.org/~appro/cryptogams/.
-@ ====================================================================
-
-@ AES for ARMv4
-
-@ January 2007.
-@
-@ Code uses single 1K S-box and is >2 times faster than code generated
-@ by gcc-3.4.1. This is thanks to unique feature of ARMv4 ISA, which
-@ allows to merge logical or arithmetic operation with shift or rotate
-@ in one instruction and emit combined result every cycle. The module
-@ is endian-neutral. The performance is ~42 cycles/byte for 128-bit
-@ key [on single-issue Xscale PXA250 core].
-
-@ May 2007.
-@
-@ AES_set_[en|de]crypt_key is added.
-
-@ July 2010.
-@
-@ Rescheduling for dual-issue pipeline resulted in 12% improvement on
-@ Cortex A8 core and ~25 cycles per byte processed with 128-bit key.
-
-@ February 2011.
-@
-@ Profiler-assisted and platform-specific optimization resulted in 16%
-@ improvement on Cortex A8 core and ~21.5 cycles per byte.
-
-@ A little glue here to select the correct code below for the ARM CPU
-@ that is being targetted.
-
-#include <linux/linkage.h>
-#include <asm/assembler.h>
-
-.text
-
-.type	AES_Te,%object
-.align	5
-AES_Te:
-.word	0xc66363a5, 0xf87c7c84, 0xee777799, 0xf67b7b8d
-.word	0xfff2f20d, 0xd66b6bbd, 0xde6f6fb1, 0x91c5c554
-.word	0x60303050, 0x02010103, 0xce6767a9, 0x562b2b7d
-.word	0xe7fefe19, 0xb5d7d762, 0x4dababe6, 0xec76769a
-.word	0x8fcaca45, 0x1f82829d, 0x89c9c940, 0xfa7d7d87
-.word	0xeffafa15, 0xb25959eb, 0x8e4747c9, 0xfbf0f00b
-.word	0x41adadec, 0xb3d4d467, 0x5fa2a2fd, 0x45afafea
-.word	0x239c9cbf, 0x53a4a4f7, 0xe4727296, 0x9bc0c05b
-.word	0x75b7b7c2, 0xe1fdfd1c, 0x3d9393ae, 0x4c26266a
-.word	0x6c36365a, 0x7e3f3f41, 0xf5f7f702, 0x83cccc4f
-.word	0x6834345c, 0x51a5a5f4, 0xd1e5e534, 0xf9f1f108
-.word	0xe2717193, 0xabd8d873, 0x62313153, 0x2a15153f
-.word	0x0804040c, 0x95c7c752, 0x46232365, 0x9dc3c35e
-.word	0x30181828, 0x379696a1, 0x0a05050f, 0x2f9a9ab5
-.word	0x0e070709, 0x24121236, 0x1b80809b, 0xdfe2e23d
-.word	0xcdebeb26, 0x4e272769, 0x7fb2b2cd, 0xea75759f
-.word	0x1209091b, 0x1d83839e, 0x582c2c74, 0x341a1a2e
-.word	0x361b1b2d, 0xdc6e6eb2, 0xb45a5aee, 0x5ba0a0fb
-.word	0xa45252f6, 0x763b3b4d, 0xb7d6d661, 0x7db3b3ce
-.word	0x5229297b, 0xdde3e33e, 0x5e2f2f71, 0x13848497
-.word	0xa65353f5, 0xb9d1d168, 0x00000000, 0xc1eded2c
-.word	0x40202060, 0xe3fcfc1f, 0x79b1b1c8, 0xb65b5bed
-.word	0xd46a6abe, 0x8dcbcb46, 0x67bebed9, 0x7239394b
-.word	0x944a4ade, 0x984c4cd4, 0xb05858e8, 0x85cfcf4a
-.word	0xbbd0d06b, 0xc5efef2a, 0x4faaaae5, 0xedfbfb16
-.word	0x864343c5, 0x9a4d4dd7, 0x66333355, 0x11858594
-.word	0x8a4545cf, 0xe9f9f910, 0x04020206, 0xfe7f7f81
-.word	0xa05050f0, 0x783c3c44, 0x259f9fba, 0x4ba8a8e3
-.word	0xa25151f3, 0x5da3a3fe, 0x804040c0, 0x058f8f8a
-.word	0x3f9292ad, 0x219d9dbc, 0x70383848, 0xf1f5f504
-.word	0x63bcbcdf, 0x77b6b6c1, 0xafdada75, 0x42212163
-.word	0x20101030, 0xe5ffff1a, 0xfdf3f30e, 0xbfd2d26d
-.word	0x81cdcd4c, 0x180c0c14, 0x26131335, 0xc3ecec2f
-.word	0xbe5f5fe1, 0x359797a2, 0x884444cc, 0x2e171739
-.word	0x93c4c457, 0x55a7a7f2, 0xfc7e7e82, 0x7a3d3d47
-.word	0xc86464ac, 0xba5d5de7, 0x3219192b, 0xe6737395
-.word	0xc06060a0, 0x19818198, 0x9e4f4fd1, 0xa3dcdc7f
-.word	0x44222266, 0x542a2a7e, 0x3b9090ab, 0x0b888883
-.word	0x8c4646ca, 0xc7eeee29, 0x6bb8b8d3, 0x2814143c
-.word	0xa7dede79, 0xbc5e5ee2, 0x160b0b1d, 0xaddbdb76
-.word	0xdbe0e03b, 0x64323256, 0x743a3a4e, 0x140a0a1e
-.word	0x924949db, 0x0c06060a, 0x4824246c, 0xb85c5ce4
-.word	0x9fc2c25d, 0xbdd3d36e, 0x43acacef, 0xc46262a6
-.word	0x399191a8, 0x319595a4, 0xd3e4e437, 0xf279798b
-.word	0xd5e7e732, 0x8bc8c843, 0x6e373759, 0xda6d6db7
-.word	0x018d8d8c, 0xb1d5d564, 0x9c4e4ed2, 0x49a9a9e0
-.word	0xd86c6cb4, 0xac5656fa, 0xf3f4f407, 0xcfeaea25
-.word	0xca6565af, 0xf47a7a8e, 0x47aeaee9, 0x10080818
-.word	0x6fbabad5, 0xf0787888, 0x4a25256f, 0x5c2e2e72
-.word	0x381c1c24, 0x57a6a6f1, 0x73b4b4c7, 0x97c6c651
-.word	0xcbe8e823, 0xa1dddd7c, 0xe874749c, 0x3e1f1f21
-.word	0x964b4bdd, 0x61bdbddc, 0x0d8b8b86, 0x0f8a8a85
-.word	0xe0707090, 0x7c3e3e42, 0x71b5b5c4, 0xcc6666aa
-.word	0x904848d8, 0x06030305, 0xf7f6f601, 0x1c0e0e12
-.word	0xc26161a3, 0x6a35355f, 0xae5757f9, 0x69b9b9d0
-.word	0x17868691, 0x99c1c158, 0x3a1d1d27, 0x279e9eb9
-.word	0xd9e1e138, 0xebf8f813, 0x2b9898b3, 0x22111133
-.word	0xd26969bb, 0xa9d9d970, 0x078e8e89, 0x339494a7
-.word	0x2d9b9bb6, 0x3c1e1e22, 0x15878792, 0xc9e9e920
-.word	0x87cece49, 0xaa5555ff, 0x50282878, 0xa5dfdf7a
-.word	0x038c8c8f, 0x59a1a1f8, 0x09898980, 0x1a0d0d17
-.word	0x65bfbfda, 0xd7e6e631, 0x844242c6, 0xd06868b8
-.word	0x824141c3, 0x299999b0, 0x5a2d2d77, 0x1e0f0f11
-.word	0x7bb0b0cb, 0xa85454fc, 0x6dbbbbd6, 0x2c16163a
-@ Te4[256]
-.byte	0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5
-.byte	0x30, 0x01, 0x67, 0x2b, 0xfe, 0xd7, 0xab, 0x76
-.byte	0xca, 0x82, 0xc9, 0x7d, 0xfa, 0x59, 0x47, 0xf0
-.byte	0xad, 0xd4, 0xa2, 0xaf, 0x9c, 0xa4, 0x72, 0xc0
-.byte	0xb7, 0xfd, 0x93, 0x26, 0x36, 0x3f, 0xf7, 0xcc
-.byte	0x34, 0xa5, 0xe5, 0xf1, 0x71, 0xd8, 0x31, 0x15
-.byte	0x04, 0xc7, 0x23, 0xc3, 0x18, 0x96, 0x05, 0x9a
-.byte	0x07, 0x12, 0x80, 0xe2, 0xeb, 0x27, 0xb2, 0x75
-.byte	0x09, 0x83, 0x2c, 0x1a, 0x1b, 0x6e, 0x5a, 0xa0
-.byte	0x52, 0x3b, 0xd6, 0xb3, 0x29, 0xe3, 0x2f, 0x84
-.byte	0x53, 0xd1, 0x00, 0xed, 0x20, 0xfc, 0xb1, 0x5b
-.byte	0x6a, 0xcb, 0xbe, 0x39, 0x4a, 0x4c, 0x58, 0xcf
-.byte	0xd0, 0xef, 0xaa, 0xfb, 0x43, 0x4d, 0x33, 0x85
-.byte	0x45, 0xf9, 0x02, 0x7f, 0x50, 0x3c, 0x9f, 0xa8
-.byte	0x51, 0xa3, 0x40, 0x8f, 0x92, 0x9d, 0x38, 0xf5
-.byte	0xbc, 0xb6, 0xda, 0x21, 0x10, 0xff, 0xf3, 0xd2
-.byte	0xcd, 0x0c, 0x13, 0xec, 0x5f, 0x97, 0x44, 0x17
-.byte	0xc4, 0xa7, 0x7e, 0x3d, 0x64, 0x5d, 0x19, 0x73
-.byte	0x60, 0x81, 0x4f, 0xdc, 0x22, 0x2a, 0x90, 0x88
-.byte	0x46, 0xee, 0xb8, 0x14, 0xde, 0x5e, 0x0b, 0xdb
-.byte	0xe0, 0x32, 0x3a, 0x0a, 0x49, 0x06, 0x24, 0x5c
-.byte	0xc2, 0xd3, 0xac, 0x62, 0x91, 0x95, 0xe4, 0x79
-.byte	0xe7, 0xc8, 0x37, 0x6d, 0x8d, 0xd5, 0x4e, 0xa9
-.byte	0x6c, 0x56, 0xf4, 0xea, 0x65, 0x7a, 0xae, 0x08
-.byte	0xba, 0x78, 0x25, 0x2e, 0x1c, 0xa6, 0xb4, 0xc6
-.byte	0xe8, 0xdd, 0x74, 0x1f, 0x4b, 0xbd, 0x8b, 0x8a
-.byte	0x70, 0x3e, 0xb5, 0x66, 0x48, 0x03, 0xf6, 0x0e
-.byte	0x61, 0x35, 0x57, 0xb9, 0x86, 0xc1, 0x1d, 0x9e
-.byte	0xe1, 0xf8, 0x98, 0x11, 0x69, 0xd9, 0x8e, 0x94
-.byte	0x9b, 0x1e, 0x87, 0xe9, 0xce, 0x55, 0x28, 0xdf
-.byte	0x8c, 0xa1, 0x89, 0x0d, 0xbf, 0xe6, 0x42, 0x68
-.byte	0x41, 0x99, 0x2d, 0x0f, 0xb0, 0x54, 0xbb, 0x16
-@ rcon[]
-.word	0x01000000, 0x02000000, 0x04000000, 0x08000000
-.word	0x10000000, 0x20000000, 0x40000000, 0x80000000
-.word	0x1B000000, 0x36000000, 0, 0, 0, 0, 0, 0
-.size	AES_Te,.-AES_Te
-
-@ void AES_encrypt(const unsigned char *in, unsigned char *out,
-@ 		 const AES_KEY *key) {
-.align	5
-ENTRY(AES_encrypt)
-	adr	r3,AES_encrypt
-	stmdb   sp!,{r1,r4-r12,lr}
-	mov	r12,r0		@ inp
-	mov	r11,r2
-	sub	r10,r3,#AES_encrypt-AES_Te	@ Te
-#if __ARM_ARCH__<7
-	ldrb	r0,[r12,#3]	@ load input data in endian-neutral
-	ldrb	r4,[r12,#2]	@ manner...
-	ldrb	r5,[r12,#1]
-	ldrb	r6,[r12,#0]
-	orr	r0,r0,r4,lsl#8
-	ldrb	r1,[r12,#7]
-	orr	r0,r0,r5,lsl#16
-	ldrb	r4,[r12,#6]
-	orr	r0,r0,r6,lsl#24
-	ldrb	r5,[r12,#5]
-	ldrb	r6,[r12,#4]
-	orr	r1,r1,r4,lsl#8
-	ldrb	r2,[r12,#11]
-	orr	r1,r1,r5,lsl#16
-	ldrb	r4,[r12,#10]
-	orr	r1,r1,r6,lsl#24
-	ldrb	r5,[r12,#9]
-	ldrb	r6,[r12,#8]
-	orr	r2,r2,r4,lsl#8
-	ldrb	r3,[r12,#15]
-	orr	r2,r2,r5,lsl#16
-	ldrb	r4,[r12,#14]
-	orr	r2,r2,r6,lsl#24
-	ldrb	r5,[r12,#13]
-	ldrb	r6,[r12,#12]
-	orr	r3,r3,r4,lsl#8
-	orr	r3,r3,r5,lsl#16
-	orr	r3,r3,r6,lsl#24
-#else
-	ldr	r0,[r12,#0]
-	ldr	r1,[r12,#4]
-	ldr	r2,[r12,#8]
-	ldr	r3,[r12,#12]
-#ifdef __ARMEL__
-	rev	r0,r0
-	rev	r1,r1
-	rev	r2,r2
-	rev	r3,r3
-#endif
-#endif
-	bl	_armv4_AES_encrypt
-
-	ldr	r12,[sp],#4		@ pop out
-#if __ARM_ARCH__>=7
-#ifdef __ARMEL__
-	rev	r0,r0
-	rev	r1,r1
-	rev	r2,r2
-	rev	r3,r3
-#endif
-	str	r0,[r12,#0]
-	str	r1,[r12,#4]
-	str	r2,[r12,#8]
-	str	r3,[r12,#12]
-#else
-	mov	r4,r0,lsr#24		@ write output in endian-neutral
-	mov	r5,r0,lsr#16		@ manner...
-	mov	r6,r0,lsr#8
-	strb	r4,[r12,#0]
-	strb	r5,[r12,#1]
-	mov	r4,r1,lsr#24
-	strb	r6,[r12,#2]
-	mov	r5,r1,lsr#16
-	strb	r0,[r12,#3]
-	mov	r6,r1,lsr#8
-	strb	r4,[r12,#4]
-	strb	r5,[r12,#5]
-	mov	r4,r2,lsr#24
-	strb	r6,[r12,#6]
-	mov	r5,r2,lsr#16
-	strb	r1,[r12,#7]
-	mov	r6,r2,lsr#8
-	strb	r4,[r12,#8]
-	strb	r5,[r12,#9]
-	mov	r4,r3,lsr#24
-	strb	r6,[r12,#10]
-	mov	r5,r3,lsr#16
-	strb	r2,[r12,#11]
-	mov	r6,r3,lsr#8
-	strb	r4,[r12,#12]
-	strb	r5,[r12,#13]
-	strb	r6,[r12,#14]
-	strb	r3,[r12,#15]
-#endif
-	ldmia	sp!,{r4-r12,pc}
-ENDPROC(AES_encrypt)
-
-.type   _armv4_AES_encrypt,%function
-.align	2
-_armv4_AES_encrypt:
-	str	lr,[sp,#-4]!		@ push lr
-	ldmia	r11!,{r4-r7}
-	eor	r0,r0,r4
-	ldr	r12,[r11,#240-16]
-	eor	r1,r1,r5
-	eor	r2,r2,r6
-	eor	r3,r3,r7
-	sub	r12,r12,#1
-	mov	lr,#255
-
-	and	r7,lr,r0
-	and	r8,lr,r0,lsr#8
-	and	r9,lr,r0,lsr#16
-	mov	r0,r0,lsr#24
-.Lenc_loop:
-	ldr	r4,[r10,r7,lsl#2]	@ Te3[s0>>0]
-	and	r7,lr,r1,lsr#16	@ i0
-	ldr	r5,[r10,r8,lsl#2]	@ Te2[s0>>8]
-	and	r8,lr,r1
-	ldr	r6,[r10,r9,lsl#2]	@ Te1[s0>>16]
-	and	r9,lr,r1,lsr#8
-	ldr	r0,[r10,r0,lsl#2]	@ Te0[s0>>24]
-	mov	r1,r1,lsr#24
-
-	ldr	r7,[r10,r7,lsl#2]	@ Te1[s1>>16]
-	ldr	r8,[r10,r8,lsl#2]	@ Te3[s1>>0]
-	ldr	r9,[r10,r9,lsl#2]	@ Te2[s1>>8]
-	eor	r0,r0,r7,ror#8
-	ldr	r1,[r10,r1,lsl#2]	@ Te0[s1>>24]
-	and	r7,lr,r2,lsr#8	@ i0
-	eor	r5,r5,r8,ror#8
-	and	r8,lr,r2,lsr#16	@ i1
-	eor	r6,r6,r9,ror#8
-	and	r9,lr,r2
-	ldr	r7,[r10,r7,lsl#2]	@ Te2[s2>>8]
-	eor	r1,r1,r4,ror#24
-	ldr	r8,[r10,r8,lsl#2]	@ Te1[s2>>16]
-	mov	r2,r2,lsr#24
-
-	ldr	r9,[r10,r9,lsl#2]	@ Te3[s2>>0]
-	eor	r0,r0,r7,ror#16
-	ldr	r2,[r10,r2,lsl#2]	@ Te0[s2>>24]
-	and	r7,lr,r3		@ i0
-	eor	r1,r1,r8,ror#8
-	and	r8,lr,r3,lsr#8	@ i1
-	eor	r6,r6,r9,ror#16
-	and	r9,lr,r3,lsr#16	@ i2
-	ldr	r7,[r10,r7,lsl#2]	@ Te3[s3>>0]
-	eor	r2,r2,r5,ror#16
-	ldr	r8,[r10,r8,lsl#2]	@ Te2[s3>>8]
-	mov	r3,r3,lsr#24
-
-	ldr	r9,[r10,r9,lsl#2]	@ Te1[s3>>16]
-	eor	r0,r0,r7,ror#24
-	ldr	r7,[r11],#16
-	eor	r1,r1,r8,ror#16
-	ldr	r3,[r10,r3,lsl#2]	@ Te0[s3>>24]
-	eor	r2,r2,r9,ror#8
-	ldr	r4,[r11,#-12]
-	eor	r3,r3,r6,ror#8
-
-	ldr	r5,[r11,#-8]
-	eor	r0,r0,r7
-	ldr	r6,[r11,#-4]
-	and	r7,lr,r0
-	eor	r1,r1,r4
-	and	r8,lr,r0,lsr#8
-	eor	r2,r2,r5
-	and	r9,lr,r0,lsr#16
-	eor	r3,r3,r6
-	mov	r0,r0,lsr#24
-
-	subs	r12,r12,#1
-	bne	.Lenc_loop
-
-	add	r10,r10,#2
-
-	ldrb	r4,[r10,r7,lsl#2]	@ Te4[s0>>0]
-	and	r7,lr,r1,lsr#16	@ i0
-	ldrb	r5,[r10,r8,lsl#2]	@ Te4[s0>>8]
-	and	r8,lr,r1
-	ldrb	r6,[r10,r9,lsl#2]	@ Te4[s0>>16]
-	and	r9,lr,r1,lsr#8
-	ldrb	r0,[r10,r0,lsl#2]	@ Te4[s0>>24]
-	mov	r1,r1,lsr#24
-
-	ldrb	r7,[r10,r7,lsl#2]	@ Te4[s1>>16]
-	ldrb	r8,[r10,r8,lsl#2]	@ Te4[s1>>0]
-	ldrb	r9,[r10,r9,lsl#2]	@ Te4[s1>>8]
-	eor	r0,r7,r0,lsl#8
-	ldrb	r1,[r10,r1,lsl#2]	@ Te4[s1>>24]
-	and	r7,lr,r2,lsr#8	@ i0
-	eor	r5,r8,r5,lsl#8
-	and	r8,lr,r2,lsr#16	@ i1
-	eor	r6,r9,r6,lsl#8
-	and	r9,lr,r2
-	ldrb	r7,[r10,r7,lsl#2]	@ Te4[s2>>8]
-	eor	r1,r4,r1,lsl#24
-	ldrb	r8,[r10,r8,lsl#2]	@ Te4[s2>>16]
-	mov	r2,r2,lsr#24
-
-	ldrb	r9,[r10,r9,lsl#2]	@ Te4[s2>>0]
-	eor	r0,r7,r0,lsl#8
-	ldrb	r2,[r10,r2,lsl#2]	@ Te4[s2>>24]
-	and	r7,lr,r3		@ i0
-	eor	r1,r1,r8,lsl#16
-	and	r8,lr,r3,lsr#8	@ i1
-	eor	r6,r9,r6,lsl#8
-	and	r9,lr,r3,lsr#16	@ i2
-	ldrb	r7,[r10,r7,lsl#2]	@ Te4[s3>>0]
-	eor	r2,r5,r2,lsl#24
-	ldrb	r8,[r10,r8,lsl#2]	@ Te4[s3>>8]
-	mov	r3,r3,lsr#24
-
-	ldrb	r9,[r10,r9,lsl#2]	@ Te4[s3>>16]
-	eor	r0,r7,r0,lsl#8
-	ldr	r7,[r11,#0]
-	ldrb	r3,[r10,r3,lsl#2]	@ Te4[s3>>24]
-	eor	r1,r1,r8,lsl#8
-	ldr	r4,[r11,#4]
-	eor	r2,r2,r9,lsl#16
-	ldr	r5,[r11,#8]
-	eor	r3,r6,r3,lsl#24
-	ldr	r6,[r11,#12]
-
-	eor	r0,r0,r7
-	eor	r1,r1,r4
-	eor	r2,r2,r5
-	eor	r3,r3,r6
-
-	sub	r10,r10,#2
-	ldr	pc,[sp],#4		@ pop and return
-.size	_armv4_AES_encrypt,.-_armv4_AES_encrypt
-
-.align	5
-ENTRY(private_AES_set_encrypt_key)
-_armv4_AES_set_encrypt_key:
-	adr	r3,_armv4_AES_set_encrypt_key
-	teq	r0,#0
-	moveq	r0,#-1
-	beq	.Labrt
-	teq	r2,#0
-	moveq	r0,#-1
-	beq	.Labrt
-
-	teq	r1,#128
-	beq	.Lok
-	teq	r1,#192
-	beq	.Lok
-	teq	r1,#256
-	movne	r0,#-1
-	bne	.Labrt
-
-.Lok:	stmdb   sp!,{r4-r12,lr}
-	sub	r10,r3,#_armv4_AES_set_encrypt_key-AES_Te-1024	@ Te4
-
-	mov	r12,r0		@ inp
-	mov	lr,r1			@ bits
-	mov	r11,r2			@ key
-
-#if __ARM_ARCH__<7
-	ldrb	r0,[r12,#3]	@ load input data in endian-neutral
-	ldrb	r4,[r12,#2]	@ manner...
-	ldrb	r5,[r12,#1]
-	ldrb	r6,[r12,#0]
-	orr	r0,r0,r4,lsl#8
-	ldrb	r1,[r12,#7]
-	orr	r0,r0,r5,lsl#16
-	ldrb	r4,[r12,#6]
-	orr	r0,r0,r6,lsl#24
-	ldrb	r5,[r12,#5]
-	ldrb	r6,[r12,#4]
-	orr	r1,r1,r4,lsl#8
-	ldrb	r2,[r12,#11]
-	orr	r1,r1,r5,lsl#16
-	ldrb	r4,[r12,#10]
-	orr	r1,r1,r6,lsl#24
-	ldrb	r5,[r12,#9]
-	ldrb	r6,[r12,#8]
-	orr	r2,r2,r4,lsl#8
-	ldrb	r3,[r12,#15]
-	orr	r2,r2,r5,lsl#16
-	ldrb	r4,[r12,#14]
-	orr	r2,r2,r6,lsl#24
-	ldrb	r5,[r12,#13]
-	ldrb	r6,[r12,#12]
-	orr	r3,r3,r4,lsl#8
-	str	r0,[r11],#16
-	orr	r3,r3,r5,lsl#16
-	str	r1,[r11,#-12]
-	orr	r3,r3,r6,lsl#24
-	str	r2,[r11,#-8]
-	str	r3,[r11,#-4]
-#else
-	ldr	r0,[r12,#0]
-	ldr	r1,[r12,#4]
-	ldr	r2,[r12,#8]
-	ldr	r3,[r12,#12]
-#ifdef __ARMEL__
-	rev	r0,r0
-	rev	r1,r1
-	rev	r2,r2
-	rev	r3,r3
-#endif
-	str	r0,[r11],#16
-	str	r1,[r11,#-12]
-	str	r2,[r11,#-8]
-	str	r3,[r11,#-4]
-#endif
-
-	teq	lr,#128
-	bne	.Lnot128
-	mov	r12,#10
-	str	r12,[r11,#240-16]
-	add	r6,r10,#256			@ rcon
-	mov	lr,#255
-
-.L128_loop:
-	and	r5,lr,r3,lsr#24
-	and	r7,lr,r3,lsr#16
-	ldrb	r5,[r10,r5]
-	and	r8,lr,r3,lsr#8
-	ldrb	r7,[r10,r7]
-	and	r9,lr,r3
-	ldrb	r8,[r10,r8]
-	orr	r5,r5,r7,lsl#24
-	ldrb	r9,[r10,r9]
-	orr	r5,r5,r8,lsl#16
-	ldr	r4,[r6],#4			@ rcon[i++]
-	orr	r5,r5,r9,lsl#8
-	eor	r5,r5,r4
-	eor	r0,r0,r5			@ rk[4]=rk[0]^...
-	eor	r1,r1,r0			@ rk[5]=rk[1]^rk[4]
-	str	r0,[r11],#16
-	eor	r2,r2,r1			@ rk[6]=rk[2]^rk[5]
-	str	r1,[r11,#-12]
-	eor	r3,r3,r2			@ rk[7]=rk[3]^rk[6]
-	str	r2,[r11,#-8]
-	subs	r12,r12,#1
-	str	r3,[r11,#-4]
-	bne	.L128_loop
-	sub	r2,r11,#176
-	b	.Ldone
-
-.Lnot128:
-#if __ARM_ARCH__<7
-	ldrb	r8,[r12,#19]
-	ldrb	r4,[r12,#18]
-	ldrb	r5,[r12,#17]
-	ldrb	r6,[r12,#16]
-	orr	r8,r8,r4,lsl#8
-	ldrb	r9,[r12,#23]
-	orr	r8,r8,r5,lsl#16
-	ldrb	r4,[r12,#22]
-	orr	r8,r8,r6,lsl#24
-	ldrb	r5,[r12,#21]
-	ldrb	r6,[r12,#20]
-	orr	r9,r9,r4,lsl#8
-	orr	r9,r9,r5,lsl#16
-	str	r8,[r11],#8
-	orr	r9,r9,r6,lsl#24
-	str	r9,[r11,#-4]
-#else
-	ldr	r8,[r12,#16]
-	ldr	r9,[r12,#20]
-#ifdef __ARMEL__
-	rev	r8,r8
-	rev	r9,r9
-#endif
-	str	r8,[r11],#8
-	str	r9,[r11,#-4]
-#endif
-
-	teq	lr,#192
-	bne	.Lnot192
-	mov	r12,#12
-	str	r12,[r11,#240-24]
-	add	r6,r10,#256			@ rcon
-	mov	lr,#255
-	mov	r12,#8
-
-.L192_loop:
-	and	r5,lr,r9,lsr#24
-	and	r7,lr,r9,lsr#16
-	ldrb	r5,[r10,r5]
-	and	r8,lr,r9,lsr#8
-	ldrb	r7,[r10,r7]
-	and	r9,lr,r9
-	ldrb	r8,[r10,r8]
-	orr	r5,r5,r7,lsl#24
-	ldrb	r9,[r10,r9]
-	orr	r5,r5,r8,lsl#16
-	ldr	r4,[r6],#4			@ rcon[i++]
-	orr	r5,r5,r9,lsl#8
-	eor	r9,r5,r4
-	eor	r0,r0,r9			@ rk[6]=rk[0]^...
-	eor	r1,r1,r0			@ rk[7]=rk[1]^rk[6]
-	str	r0,[r11],#24
-	eor	r2,r2,r1			@ rk[8]=rk[2]^rk[7]
-	str	r1,[r11,#-20]
-	eor	r3,r3,r2			@ rk[9]=rk[3]^rk[8]
-	str	r2,[r11,#-16]
-	subs	r12,r12,#1
-	str	r3,[r11,#-12]
-	subeq	r2,r11,#216
-	beq	.Ldone
-
-	ldr	r7,[r11,#-32]
-	ldr	r8,[r11,#-28]
-	eor	r7,r7,r3			@ rk[10]=rk[4]^rk[9]
-	eor	r9,r8,r7			@ rk[11]=rk[5]^rk[10]
-	str	r7,[r11,#-8]
-	str	r9,[r11,#-4]
-	b	.L192_loop
-
-.Lnot192:
-#if __ARM_ARCH__<7
-	ldrb	r8,[r12,#27]
-	ldrb	r4,[r12,#26]
-	ldrb	r5,[r12,#25]
-	ldrb	r6,[r12,#24]
-	orr	r8,r8,r4,lsl#8
-	ldrb	r9,[r12,#31]
-	orr	r8,r8,r5,lsl#16
-	ldrb	r4,[r12,#30]
-	orr	r8,r8,r6,lsl#24
-	ldrb	r5,[r12,#29]
-	ldrb	r6,[r12,#28]
-	orr	r9,r9,r4,lsl#8
-	orr	r9,r9,r5,lsl#16
-	str	r8,[r11],#8
-	orr	r9,r9,r6,lsl#24
-	str	r9,[r11,#-4]
-#else
-	ldr	r8,[r12,#24]
-	ldr	r9,[r12,#28]
-#ifdef __ARMEL__
-	rev	r8,r8
-	rev	r9,r9
-#endif
-	str	r8,[r11],#8
-	str	r9,[r11,#-4]
-#endif
-
-	mov	r12,#14
-	str	r12,[r11,#240-32]
-	add	r6,r10,#256			@ rcon
-	mov	lr,#255
-	mov	r12,#7
-
-.L256_loop:
-	and	r5,lr,r9,lsr#24
-	and	r7,lr,r9,lsr#16
-	ldrb	r5,[r10,r5]
-	and	r8,lr,r9,lsr#8
-	ldrb	r7,[r10,r7]
-	and	r9,lr,r9
-	ldrb	r8,[r10,r8]
-	orr	r5,r5,r7,lsl#24
-	ldrb	r9,[r10,r9]
-	orr	r5,r5,r8,lsl#16
-	ldr	r4,[r6],#4			@ rcon[i++]
-	orr	r5,r5,r9,lsl#8
-	eor	r9,r5,r4
-	eor	r0,r0,r9			@ rk[8]=rk[0]^...
-	eor	r1,r1,r0			@ rk[9]=rk[1]^rk[8]
-	str	r0,[r11],#32
-	eor	r2,r2,r1			@ rk[10]=rk[2]^rk[9]
-	str	r1,[r11,#-28]
-	eor	r3,r3,r2			@ rk[11]=rk[3]^rk[10]
-	str	r2,[r11,#-24]
-	subs	r12,r12,#1
-	str	r3,[r11,#-20]
-	subeq	r2,r11,#256
-	beq	.Ldone
-
-	and	r5,lr,r3
-	and	r7,lr,r3,lsr#8
-	ldrb	r5,[r10,r5]
-	and	r8,lr,r3,lsr#16
-	ldrb	r7,[r10,r7]
-	and	r9,lr,r3,lsr#24
-	ldrb	r8,[r10,r8]
-	orr	r5,r5,r7,lsl#8
-	ldrb	r9,[r10,r9]
-	orr	r5,r5,r8,lsl#16
-	ldr	r4,[r11,#-48]
-	orr	r5,r5,r9,lsl#24
-
-	ldr	r7,[r11,#-44]
-	ldr	r8,[r11,#-40]
-	eor	r4,r4,r5			@ rk[12]=rk[4]^...
-	ldr	r9,[r11,#-36]
-	eor	r7,r7,r4			@ rk[13]=rk[5]^rk[12]
-	str	r4,[r11,#-16]
-	eor	r8,r8,r7			@ rk[14]=rk[6]^rk[13]
-	str	r7,[r11,#-12]
-	eor	r9,r9,r8			@ rk[15]=rk[7]^rk[14]
-	str	r8,[r11,#-8]
-	str	r9,[r11,#-4]
-	b	.L256_loop
-
-.Ldone:	mov	r0,#0
-	ldmia   sp!,{r4-r12,lr}
-.Labrt:	ret	lr
-ENDPROC(private_AES_set_encrypt_key)
-
-.align	5
-ENTRY(private_AES_set_decrypt_key)
-	str	lr,[sp,#-4]!            @ push lr
-#if 0
-	@ kernel does both of these in setkey so optimise this bit out by
-	@ expecting the key to already have the enc_key work done (see aes_glue.c)
-	bl	_armv4_AES_set_encrypt_key
-#else
-	mov	r0,#0
-#endif
-	teq	r0,#0
-	ldrne	lr,[sp],#4              @ pop lr
-	bne	.Labrt
-
-	stmdb   sp!,{r4-r12}
-
-	ldr	r12,[r2,#240]	@ AES_set_encrypt_key preserves r2,
-	mov	r11,r2			@ which is AES_KEY *key
-	mov	r7,r2
-	add	r8,r2,r12,lsl#4
-
-.Linv:	ldr	r0,[r7]
-	ldr	r1,[r7,#4]
-	ldr	r2,[r7,#8]
-	ldr	r3,[r7,#12]
-	ldr	r4,[r8]
-	ldr	r5,[r8,#4]
-	ldr	r6,[r8,#8]
-	ldr	r9,[r8,#12]
-	str	r0,[r8],#-16
-	str	r1,[r8,#16+4]
-	str	r2,[r8,#16+8]
-	str	r3,[r8,#16+12]
-	str	r4,[r7],#16
-	str	r5,[r7,#-12]
-	str	r6,[r7,#-8]
-	str	r9,[r7,#-4]
-	teq	r7,r8
-	bne	.Linv
-	ldr	r0,[r11,#16]!		@ prefetch tp1
-	mov	r7,#0x80
-	mov	r8,#0x1b
-	orr	r7,r7,#0x8000
-	orr	r8,r8,#0x1b00
-	orr	r7,r7,r7,lsl#16
-	orr	r8,r8,r8,lsl#16
-	sub	r12,r12,#1
-	mvn	r9,r7
-	mov	r12,r12,lsl#2	@ (rounds-1)*4
-
-.Lmix:	and	r4,r0,r7
-	and	r1,r0,r9
-	sub	r4,r4,r4,lsr#7
-	and	r4,r4,r8
-	eor	r1,r4,r1,lsl#1	@ tp2
-
-	and	r4,r1,r7
-	and	r2,r1,r9
-	sub	r4,r4,r4,lsr#7
-	and	r4,r4,r8
-	eor	r2,r4,r2,lsl#1	@ tp4
-
-	and	r4,r2,r7
-	and	r3,r2,r9
-	sub	r4,r4,r4,lsr#7
-	and	r4,r4,r8
-	eor	r3,r4,r3,lsl#1	@ tp8
-
-	eor	r4,r1,r2
-	eor	r5,r0,r3		@ tp9
-	eor	r4,r4,r3		@ tpe
-	eor	r4,r4,r1,ror#24
-	eor	r4,r4,r5,ror#24	@ ^= ROTATE(tpb=tp9^tp2,8)
-	eor	r4,r4,r2,ror#16
-	eor	r4,r4,r5,ror#16	@ ^= ROTATE(tpd=tp9^tp4,16)
-	eor	r4,r4,r5,ror#8	@ ^= ROTATE(tp9,24)
-
-	ldr	r0,[r11,#4]		@ prefetch tp1
-	str	r4,[r11],#4
-	subs	r12,r12,#1
-	bne	.Lmix
-
-	mov	r0,#0
-	ldmia	sp!,{r4-r12,pc}
-ENDPROC(private_AES_set_decrypt_key)
-
-.type	AES_Td,%object
-.align	5
-AES_Td:
-.word	0x51f4a750, 0x7e416553, 0x1a17a4c3, 0x3a275e96
-.word	0x3bab6bcb, 0x1f9d45f1, 0xacfa58ab, 0x4be30393
-.word	0x2030fa55, 0xad766df6, 0x88cc7691, 0xf5024c25
-.word	0x4fe5d7fc, 0xc52acbd7, 0x26354480, 0xb562a38f
-.word	0xdeb15a49, 0x25ba1b67, 0x45ea0e98, 0x5dfec0e1
-.word	0xc32f7502, 0x814cf012, 0x8d4697a3, 0x6bd3f9c6
-.word	0x038f5fe7, 0x15929c95, 0xbf6d7aeb, 0x955259da
-.word	0xd4be832d, 0x587421d3, 0x49e06929, 0x8ec9c844
-.word	0x75c2896a, 0xf48e7978, 0x99583e6b, 0x27b971dd
-.word	0xbee14fb6, 0xf088ad17, 0xc920ac66, 0x7dce3ab4
-.word	0x63df4a18, 0xe51a3182, 0x97513360, 0x62537f45
-.word	0xb16477e0, 0xbb6bae84, 0xfe81a01c, 0xf9082b94
-.word	0x70486858, 0x8f45fd19, 0x94de6c87, 0x527bf8b7
-.word	0xab73d323, 0x724b02e2, 0xe31f8f57, 0x6655ab2a
-.word	0xb2eb2807, 0x2fb5c203, 0x86c57b9a, 0xd33708a5
-.word	0x302887f2, 0x23bfa5b2, 0x02036aba, 0xed16825c
-.word	0x8acf1c2b, 0xa779b492, 0xf307f2f0, 0x4e69e2a1
-.word	0x65daf4cd, 0x0605bed5, 0xd134621f, 0xc4a6fe8a
-.word	0x342e539d, 0xa2f355a0, 0x058ae132, 0xa4f6eb75
-.word	0x0b83ec39, 0x4060efaa, 0x5e719f06, 0xbd6e1051
-.word	0x3e218af9, 0x96dd063d, 0xdd3e05ae, 0x4de6bd46
-.word	0x91548db5, 0x71c45d05, 0x0406d46f, 0x605015ff
-.word	0x1998fb24, 0xd6bde997, 0x894043cc, 0x67d99e77
-.word	0xb0e842bd, 0x07898b88, 0xe7195b38, 0x79c8eedb
-.word	0xa17c0a47, 0x7c420fe9, 0xf8841ec9, 0x00000000
-.word	0x09808683, 0x322bed48, 0x1e1170ac, 0x6c5a724e
-.word	0xfd0efffb, 0x0f853856, 0x3daed51e, 0x362d3927
-.word	0x0a0fd964, 0x685ca621, 0x9b5b54d1, 0x24362e3a
-.word	0x0c0a67b1, 0x9357e70f, 0xb4ee96d2, 0x1b9b919e
-.word	0x80c0c54f, 0x61dc20a2, 0x5a774b69, 0x1c121a16
-.word	0xe293ba0a, 0xc0a02ae5, 0x3c22e043, 0x121b171d
-.word	0x0e090d0b, 0xf28bc7ad, 0x2db6a8b9, 0x141ea9c8
-.word	0x57f11985, 0xaf75074c, 0xee99ddbb, 0xa37f60fd
-.word	0xf701269f, 0x5c72f5bc, 0x44663bc5, 0x5bfb7e34
-.word	0x8b432976, 0xcb23c6dc, 0xb6edfc68, 0xb8e4f163
-.word	0xd731dcca, 0x42638510, 0x13972240, 0x84c61120
-.word	0x854a247d, 0xd2bb3df8, 0xaef93211, 0xc729a16d
-.word	0x1d9e2f4b, 0xdcb230f3, 0x0d8652ec, 0x77c1e3d0
-.word	0x2bb3166c, 0xa970b999, 0x119448fa, 0x47e96422
-.word	0xa8fc8cc4, 0xa0f03f1a, 0x567d2cd8, 0x223390ef
-.word	0x87494ec7, 0xd938d1c1, 0x8ccaa2fe, 0x98d40b36
-.word	0xa6f581cf, 0xa57ade28, 0xdab78e26, 0x3fadbfa4
-.word	0x2c3a9de4, 0x5078920d, 0x6a5fcc9b, 0x547e4662
-.word	0xf68d13c2, 0x90d8b8e8, 0x2e39f75e, 0x82c3aff5
-.word	0x9f5d80be, 0x69d0937c, 0x6fd52da9, 0xcf2512b3
-.word	0xc8ac993b, 0x10187da7, 0xe89c636e, 0xdb3bbb7b
-.word	0xcd267809, 0x6e5918f4, 0xec9ab701, 0x834f9aa8
-.word	0xe6956e65, 0xaaffe67e, 0x21bccf08, 0xef15e8e6
-.word	0xbae79bd9, 0x4a6f36ce, 0xea9f09d4, 0x29b07cd6
-.word	0x31a4b2af, 0x2a3f2331, 0xc6a59430, 0x35a266c0
-.word	0x744ebc37, 0xfc82caa6, 0xe090d0b0, 0x33a7d815
-.word	0xf104984a, 0x41ecdaf7, 0x7fcd500e, 0x1791f62f
-.word	0x764dd68d, 0x43efb04d, 0xccaa4d54, 0xe49604df
-.word	0x9ed1b5e3, 0x4c6a881b, 0xc12c1fb8, 0x4665517f
-.word	0x9d5eea04, 0x018c355d, 0xfa877473, 0xfb0b412e
-.word	0xb3671d5a, 0x92dbd252, 0xe9105633, 0x6dd64713
-.word	0x9ad7618c, 0x37a10c7a, 0x59f8148e, 0xeb133c89
-.word	0xcea927ee, 0xb761c935, 0xe11ce5ed, 0x7a47b13c
-.word	0x9cd2df59, 0x55f2733f, 0x1814ce79, 0x73c737bf
-.word	0x53f7cdea, 0x5ffdaa5b, 0xdf3d6f14, 0x7844db86
-.word	0xcaaff381, 0xb968c43e, 0x3824342c, 0xc2a3405f
-.word	0x161dc372, 0xbce2250c, 0x283c498b, 0xff0d9541
-.word	0x39a80171, 0x080cb3de, 0xd8b4e49c, 0x6456c190
-.word	0x7bcb8461, 0xd532b670, 0x486c5c74, 0xd0b85742
-@ Td4[256]
-.byte	0x52, 0x09, 0x6a, 0xd5, 0x30, 0x36, 0xa5, 0x38
-.byte	0xbf, 0x40, 0xa3, 0x9e, 0x81, 0xf3, 0xd7, 0xfb
-.byte	0x7c, 0xe3, 0x39, 0x82, 0x9b, 0x2f, 0xff, 0x87
-.byte	0x34, 0x8e, 0x43, 0x44, 0xc4, 0xde, 0xe9, 0xcb
-.byte	0x54, 0x7b, 0x94, 0x32, 0xa6, 0xc2, 0x23, 0x3d
-.byte	0xee, 0x4c, 0x95, 0x0b, 0x42, 0xfa, 0xc3, 0x4e
-.byte	0x08, 0x2e, 0xa1, 0x66, 0x28, 0xd9, 0x24, 0xb2
-.byte	0x76, 0x5b, 0xa2, 0x49, 0x6d, 0x8b, 0xd1, 0x25
-.byte	0x72, 0xf8, 0xf6, 0x64, 0x86, 0x68, 0x98, 0x16
-.byte	0xd4, 0xa4, 0x5c, 0xcc, 0x5d, 0x65, 0xb6, 0x92
-.byte	0x6c, 0x70, 0x48, 0x50, 0xfd, 0xed, 0xb9, 0xda
-.byte	0x5e, 0x15, 0x46, 0x57, 0xa7, 0x8d, 0x9d, 0x84
-.byte	0x90, 0xd8, 0xab, 0x00, 0x8c, 0xbc, 0xd3, 0x0a
-.byte	0xf7, 0xe4, 0x58, 0x05, 0xb8, 0xb3, 0x45, 0x06
-.byte	0xd0, 0x2c, 0x1e, 0x8f, 0xca, 0x3f, 0x0f, 0x02
-.byte	0xc1, 0xaf, 0xbd, 0x03, 0x01, 0x13, 0x8a, 0x6b
-.byte	0x3a, 0x91, 0x11, 0x41, 0x4f, 0x67, 0xdc, 0xea
-.byte	0x97, 0xf2, 0xcf, 0xce, 0xf0, 0xb4, 0xe6, 0x73
-.byte	0x96, 0xac, 0x74, 0x22, 0xe7, 0xad, 0x35, 0x85
-.byte	0xe2, 0xf9, 0x37, 0xe8, 0x1c, 0x75, 0xdf, 0x6e
-.byte	0x47, 0xf1, 0x1a, 0x71, 0x1d, 0x29, 0xc5, 0x89
-.byte	0x6f, 0xb7, 0x62, 0x0e, 0xaa, 0x18, 0xbe, 0x1b
-.byte	0xfc, 0x56, 0x3e, 0x4b, 0xc6, 0xd2, 0x79, 0x20
-.byte	0x9a, 0xdb, 0xc0, 0xfe, 0x78, 0xcd, 0x5a, 0xf4
-.byte	0x1f, 0xdd, 0xa8, 0x33, 0x88, 0x07, 0xc7, 0x31
-.byte	0xb1, 0x12, 0x10, 0x59, 0x27, 0x80, 0xec, 0x5f
-.byte	0x60, 0x51, 0x7f, 0xa9, 0x19, 0xb5, 0x4a, 0x0d
-.byte	0x2d, 0xe5, 0x7a, 0x9f, 0x93, 0xc9, 0x9c, 0xef
-.byte	0xa0, 0xe0, 0x3b, 0x4d, 0xae, 0x2a, 0xf5, 0xb0
-.byte	0xc8, 0xeb, 0xbb, 0x3c, 0x83, 0x53, 0x99, 0x61
-.byte	0x17, 0x2b, 0x04, 0x7e, 0xba, 0x77, 0xd6, 0x26
-.byte	0xe1, 0x69, 0x14, 0x63, 0x55, 0x21, 0x0c, 0x7d
-.size	AES_Td,.-AES_Td
-
-@ void AES_decrypt(const unsigned char *in, unsigned char *out,
-@ 		 const AES_KEY *key) {
-.align	5
-ENTRY(AES_decrypt)
-	adr	r3,AES_decrypt
-	stmdb   sp!,{r1,r4-r12,lr}
-	mov	r12,r0		@ inp
-	mov	r11,r2
-	sub	r10,r3,#AES_decrypt-AES_Td		@ Td
-#if __ARM_ARCH__<7
-	ldrb	r0,[r12,#3]	@ load input data in endian-neutral
-	ldrb	r4,[r12,#2]	@ manner...
-	ldrb	r5,[r12,#1]
-	ldrb	r6,[r12,#0]
-	orr	r0,r0,r4,lsl#8
-	ldrb	r1,[r12,#7]
-	orr	r0,r0,r5,lsl#16
-	ldrb	r4,[r12,#6]
-	orr	r0,r0,r6,lsl#24
-	ldrb	r5,[r12,#5]
-	ldrb	r6,[r12,#4]
-	orr	r1,r1,r4,lsl#8
-	ldrb	r2,[r12,#11]
-	orr	r1,r1,r5,lsl#16
-	ldrb	r4,[r12,#10]
-	orr	r1,r1,r6,lsl#24
-	ldrb	r5,[r12,#9]
-	ldrb	r6,[r12,#8]
-	orr	r2,r2,r4,lsl#8
-	ldrb	r3,[r12,#15]
-	orr	r2,r2,r5,lsl#16
-	ldrb	r4,[r12,#14]
-	orr	r2,r2,r6,lsl#24
-	ldrb	r5,[r12,#13]
-	ldrb	r6,[r12,#12]
-	orr	r3,r3,r4,lsl#8
-	orr	r3,r3,r5,lsl#16
-	orr	r3,r3,r6,lsl#24
-#else
-	ldr	r0,[r12,#0]
-	ldr	r1,[r12,#4]
-	ldr	r2,[r12,#8]
-	ldr	r3,[r12,#12]
-#ifdef __ARMEL__
-	rev	r0,r0
-	rev	r1,r1
-	rev	r2,r2
-	rev	r3,r3
-#endif
-#endif
-	bl	_armv4_AES_decrypt
-
-	ldr	r12,[sp],#4		@ pop out
-#if __ARM_ARCH__>=7
-#ifdef __ARMEL__
-	rev	r0,r0
-	rev	r1,r1
-	rev	r2,r2
-	rev	r3,r3
-#endif
-	str	r0,[r12,#0]
-	str	r1,[r12,#4]
-	str	r2,[r12,#8]
-	str	r3,[r12,#12]
-#else
-	mov	r4,r0,lsr#24		@ write output in endian-neutral
-	mov	r5,r0,lsr#16		@ manner...
-	mov	r6,r0,lsr#8
-	strb	r4,[r12,#0]
-	strb	r5,[r12,#1]
-	mov	r4,r1,lsr#24
-	strb	r6,[r12,#2]
-	mov	r5,r1,lsr#16
-	strb	r0,[r12,#3]
-	mov	r6,r1,lsr#8
-	strb	r4,[r12,#4]
-	strb	r5,[r12,#5]
-	mov	r4,r2,lsr#24
-	strb	r6,[r12,#6]
-	mov	r5,r2,lsr#16
-	strb	r1,[r12,#7]
-	mov	r6,r2,lsr#8
-	strb	r4,[r12,#8]
-	strb	r5,[r12,#9]
-	mov	r4,r3,lsr#24
-	strb	r6,[r12,#10]
-	mov	r5,r3,lsr#16
-	strb	r2,[r12,#11]
-	mov	r6,r3,lsr#8
-	strb	r4,[r12,#12]
-	strb	r5,[r12,#13]
-	strb	r6,[r12,#14]
-	strb	r3,[r12,#15]
-#endif
-	ldmia	sp!,{r4-r12,pc}
-ENDPROC(AES_decrypt)
-
-.type   _armv4_AES_decrypt,%function
-.align	2
-_armv4_AES_decrypt:
-	str	lr,[sp,#-4]!		@ push lr
-	ldmia	r11!,{r4-r7}
-	eor	r0,r0,r4
-	ldr	r12,[r11,#240-16]
-	eor	r1,r1,r5
-	eor	r2,r2,r6
-	eor	r3,r3,r7
-	sub	r12,r12,#1
-	mov	lr,#255
-
-	and	r7,lr,r0,lsr#16
-	and	r8,lr,r0,lsr#8
-	and	r9,lr,r0
-	mov	r0,r0,lsr#24
-.Ldec_loop:
-	ldr	r4,[r10,r7,lsl#2]	@ Td1[s0>>16]
-	and	r7,lr,r1		@ i0
-	ldr	r5,[r10,r8,lsl#2]	@ Td2[s0>>8]
-	and	r8,lr,r1,lsr#16
-	ldr	r6,[r10,r9,lsl#2]	@ Td3[s0>>0]
-	and	r9,lr,r1,lsr#8
-	ldr	r0,[r10,r0,lsl#2]	@ Td0[s0>>24]
-	mov	r1,r1,lsr#24
-
-	ldr	r7,[r10,r7,lsl#2]	@ Td3[s1>>0]
-	ldr	r8,[r10,r8,lsl#2]	@ Td1[s1>>16]
-	ldr	r9,[r10,r9,lsl#2]	@ Td2[s1>>8]
-	eor	r0,r0,r7,ror#24
-	ldr	r1,[r10,r1,lsl#2]	@ Td0[s1>>24]
-	and	r7,lr,r2,lsr#8	@ i0
-	eor	r5,r8,r5,ror#8
-	and	r8,lr,r2		@ i1
-	eor	r6,r9,r6,ror#8
-	and	r9,lr,r2,lsr#16
-	ldr	r7,[r10,r7,lsl#2]	@ Td2[s2>>8]
-	eor	r1,r1,r4,ror#8
-	ldr	r8,[r10,r8,lsl#2]	@ Td3[s2>>0]
-	mov	r2,r2,lsr#24
-
-	ldr	r9,[r10,r9,lsl#2]	@ Td1[s2>>16]
-	eor	r0,r0,r7,ror#16
-	ldr	r2,[r10,r2,lsl#2]	@ Td0[s2>>24]
-	and	r7,lr,r3,lsr#16	@ i0
-	eor	r1,r1,r8,ror#24
-	and	r8,lr,r3,lsr#8	@ i1
-	eor	r6,r9,r6,ror#8
-	and	r9,lr,r3		@ i2
-	ldr	r7,[r10,r7,lsl#2]	@ Td1[s3>>16]
-	eor	r2,r2,r5,ror#8
-	ldr	r8,[r10,r8,lsl#2]	@ Td2[s3>>8]
-	mov	r3,r3,lsr#24
-
-	ldr	r9,[r10,r9,lsl#2]	@ Td3[s3>>0]
-	eor	r0,r0,r7,ror#8
-	ldr	r7,[r11],#16
-	eor	r1,r1,r8,ror#16
-	ldr	r3,[r10,r3,lsl#2]	@ Td0[s3>>24]
-	eor	r2,r2,r9,ror#24
-
-	ldr	r4,[r11,#-12]
-	eor	r0,r0,r7
-	ldr	r5,[r11,#-8]
-	eor	r3,r3,r6,ror#8
-	ldr	r6,[r11,#-4]
-	and	r7,lr,r0,lsr#16
-	eor	r1,r1,r4
-	and	r8,lr,r0,lsr#8
-	eor	r2,r2,r5
-	and	r9,lr,r0
-	eor	r3,r3,r6
-	mov	r0,r0,lsr#24
-
-	subs	r12,r12,#1
-	bne	.Ldec_loop
-
-	add	r10,r10,#1024
-
-	ldr	r5,[r10,#0]		@ prefetch Td4
-	ldr	r6,[r10,#32]
-	ldr	r4,[r10,#64]
-	ldr	r5,[r10,#96]
-	ldr	r6,[r10,#128]
-	ldr	r4,[r10,#160]
-	ldr	r5,[r10,#192]
-	ldr	r6,[r10,#224]
-
-	ldrb	r0,[r10,r0]		@ Td4[s0>>24]
-	ldrb	r4,[r10,r7]		@ Td4[s0>>16]
-	and	r7,lr,r1		@ i0
-	ldrb	r5,[r10,r8]		@ Td4[s0>>8]
-	and	r8,lr,r1,lsr#16
-	ldrb	r6,[r10,r9]		@ Td4[s0>>0]
-	and	r9,lr,r1,lsr#8
-
-	ldrb	r7,[r10,r7]		@ Td4[s1>>0]
- ARM(	ldrb	r1,[r10,r1,lsr#24]  )	@ Td4[s1>>24]
- THUMB(	add	r1,r10,r1,lsr#24    ) 	@ Td4[s1>>24]
- THUMB(	ldrb	r1,[r1]		    )
-	ldrb	r8,[r10,r8]		@ Td4[s1>>16]
-	eor	r0,r7,r0,lsl#24
-	ldrb	r9,[r10,r9]		@ Td4[s1>>8]
-	eor	r1,r4,r1,lsl#8
-	and	r7,lr,r2,lsr#8	@ i0
-	eor	r5,r5,r8,lsl#8
-	and	r8,lr,r2		@ i1
-	ldrb	r7,[r10,r7]		@ Td4[s2>>8]
-	eor	r6,r6,r9,lsl#8
-	ldrb	r8,[r10,r8]		@ Td4[s2>>0]
-	and	r9,lr,r2,lsr#16
-
- ARM(	ldrb	r2,[r10,r2,lsr#24]  )	@ Td4[s2>>24]
- THUMB(	add	r2,r10,r2,lsr#24    )	@ Td4[s2>>24]
- THUMB(	ldrb	r2,[r2]		    )
-	eor	r0,r0,r7,lsl#8
-	ldrb	r9,[r10,r9]		@ Td4[s2>>16]
-	eor	r1,r8,r1,lsl#16
-	and	r7,lr,r3,lsr#16	@ i0
-	eor	r2,r5,r2,lsl#16
-	and	r8,lr,r3,lsr#8	@ i1
-	ldrb	r7,[r10,r7]		@ Td4[s3>>16]
-	eor	r6,r6,r9,lsl#16
-	ldrb	r8,[r10,r8]		@ Td4[s3>>8]
-	and	r9,lr,r3		@ i2
-
-	ldrb	r9,[r10,r9]		@ Td4[s3>>0]
- ARM(	ldrb	r3,[r10,r3,lsr#24]  )	@ Td4[s3>>24]
- THUMB(	add	r3,r10,r3,lsr#24    )	@ Td4[s3>>24]
- THUMB(	ldrb	r3,[r3]		    )
-	eor	r0,r0,r7,lsl#16
-	ldr	r7,[r11,#0]
-	eor	r1,r1,r8,lsl#8
-	ldr	r4,[r11,#4]
-	eor	r2,r9,r2,lsl#8
-	ldr	r5,[r11,#8]
-	eor	r3,r6,r3,lsl#24
-	ldr	r6,[r11,#12]
-
-	eor	r0,r0,r7
-	eor	r1,r1,r4
-	eor	r2,r2,r5
-	eor	r3,r3,r6
-
-	sub	r10,r10,#1024
-	ldr	pc,[sp],#4		@ pop and return
-.size	_armv4_AES_decrypt,.-_armv4_AES_decrypt
-.asciz	"AES for ARMv4, CRYPTOGAMS by <appro@openssl.org>"
-.align	2
diff --git a/arch/arm/crypto/aes-ce-core.S b/arch/arm/crypto/aes-ce-core.S
index 987aa632c9f0..ba8e6a32fdc9 100644
--- a/arch/arm/crypto/aes-ce-core.S
+++ b/arch/arm/crypto/aes-ce-core.S
@@ -169,19 +169,19 @@ ENTRY(ce_aes_ecb_encrypt)
 .Lecbencloop3x:
 	subs		r4, r4, #3
 	bmi		.Lecbenc1x
-	vld1.8		{q0-q1}, [r1, :64]!
-	vld1.8		{q2}, [r1, :64]!
+	vld1.8		{q0-q1}, [r1]!
+	vld1.8		{q2}, [r1]!
 	bl		aes_encrypt_3x
-	vst1.8		{q0-q1}, [r0, :64]!
-	vst1.8		{q2}, [r0, :64]!
+	vst1.8		{q0-q1}, [r0]!
+	vst1.8		{q2}, [r0]!
 	b		.Lecbencloop3x
 .Lecbenc1x:
 	adds		r4, r4, #3
 	beq		.Lecbencout
 .Lecbencloop:
-	vld1.8		{q0}, [r1, :64]!
+	vld1.8		{q0}, [r1]!
 	bl		aes_encrypt
-	vst1.8		{q0}, [r0, :64]!
+	vst1.8		{q0}, [r0]!
 	subs		r4, r4, #1
 	bne		.Lecbencloop
 .Lecbencout:
@@ -195,19 +195,19 @@ ENTRY(ce_aes_ecb_decrypt)
 .Lecbdecloop3x:
 	subs		r4, r4, #3
 	bmi		.Lecbdec1x
-	vld1.8		{q0-q1}, [r1, :64]!
-	vld1.8		{q2}, [r1, :64]!
+	vld1.8		{q0-q1}, [r1]!
+	vld1.8		{q2}, [r1]!
 	bl		aes_decrypt_3x
-	vst1.8		{q0-q1}, [r0, :64]!
-	vst1.8		{q2}, [r0, :64]!
+	vst1.8		{q0-q1}, [r0]!
+	vst1.8		{q2}, [r0]!
 	b		.Lecbdecloop3x
 .Lecbdec1x:
 	adds		r4, r4, #3
 	beq		.Lecbdecout
 .Lecbdecloop:
-	vld1.8		{q0}, [r1, :64]!
+	vld1.8		{q0}, [r1]!
 	bl		aes_decrypt
-	vst1.8		{q0}, [r0, :64]!
+	vst1.8		{q0}, [r0]!
 	subs		r4, r4, #1
 	bne		.Lecbdecloop
 .Lecbdecout:
@@ -226,10 +226,10 @@ ENTRY(ce_aes_cbc_encrypt)
 	vld1.8		{q0}, [r5]
 	prepare_key	r2, r3
 .Lcbcencloop:
-	vld1.8		{q1}, [r1, :64]!	@ get next pt block
+	vld1.8		{q1}, [r1]!		@ get next pt block
 	veor		q0, q0, q1		@ ..and xor with iv
 	bl		aes_encrypt
-	vst1.8		{q0}, [r0, :64]!
+	vst1.8		{q0}, [r0]!
 	subs		r4, r4, #1
 	bne		.Lcbcencloop
 	vst1.8		{q0}, [r5]
@@ -244,8 +244,8 @@ ENTRY(ce_aes_cbc_decrypt)
 .Lcbcdecloop3x:
 	subs		r4, r4, #3
 	bmi		.Lcbcdec1x
-	vld1.8		{q0-q1}, [r1, :64]!
-	vld1.8		{q2}, [r1, :64]!
+	vld1.8		{q0-q1}, [r1]!
+	vld1.8		{q2}, [r1]!
 	vmov		q3, q0
 	vmov		q4, q1
 	vmov		q5, q2
@@ -254,19 +254,19 @@ ENTRY(ce_aes_cbc_decrypt)
 	veor		q1, q1, q3
 	veor		q2, q2, q4
 	vmov		q6, q5
-	vst1.8		{q0-q1}, [r0, :64]!
-	vst1.8		{q2}, [r0, :64]!
+	vst1.8		{q0-q1}, [r0]!
+	vst1.8		{q2}, [r0]!
 	b		.Lcbcdecloop3x
 .Lcbcdec1x:
 	adds		r4, r4, #3
 	beq		.Lcbcdecout
 	vmov		q15, q14		@ preserve last round key
 .Lcbcdecloop:
-	vld1.8		{q0}, [r1, :64]!	@ get next ct block
+	vld1.8		{q0}, [r1]!		@ get next ct block
 	veor		q14, q15, q6		@ combine prev ct with last key
 	vmov		q6, q0
 	bl		aes_decrypt
-	vst1.8		{q0}, [r0, :64]!
+	vst1.8		{q0}, [r0]!
 	subs		r4, r4, #1
 	bne		.Lcbcdecloop
 .Lcbcdecout:
@@ -300,15 +300,15 @@ ENTRY(ce_aes_ctr_encrypt)
 	rev		ip, r6
 	add		r6, r6, #1
 	vmov		s11, ip
-	vld1.8		{q3-q4}, [r1, :64]!
-	vld1.8		{q5}, [r1, :64]!
+	vld1.8		{q3-q4}, [r1]!
+	vld1.8		{q5}, [r1]!
 	bl		aes_encrypt_3x
 	veor		q0, q0, q3
 	veor		q1, q1, q4
 	veor		q2, q2, q5
 	rev		ip, r6
-	vst1.8		{q0-q1}, [r0, :64]!
-	vst1.8		{q2}, [r0, :64]!
+	vst1.8		{q0-q1}, [r0]!
+	vst1.8		{q2}, [r0]!
 	vmov		s27, ip
 	b		.Lctrloop3x
 .Lctr1x:
@@ -318,10 +318,10 @@ ENTRY(ce_aes_ctr_encrypt)
 	vmov		q0, q6
 	bl		aes_encrypt
 	subs		r4, r4, #1
-	bmi		.Lctrhalfblock		@ blocks < 0 means 1/2 block
-	vld1.8		{q3}, [r1, :64]!
+	bmi		.Lctrtailblock		@ blocks < 0 means tail block
+	vld1.8		{q3}, [r1]!
 	veor		q3, q0, q3
-	vst1.8		{q3}, [r0, :64]!
+	vst1.8		{q3}, [r0]!
 
 	adds		r6, r6, #1		@ increment BE ctr
 	rev		ip, r6
@@ -333,10 +333,8 @@ ENTRY(ce_aes_ctr_encrypt)
 	vst1.8		{q6}, [r5]
 	pop		{r4-r6, pc}
 
-.Lctrhalfblock:
-	vld1.8		{d1}, [r1, :64]
-	veor		d0, d0, d1
-	vst1.8		{d0}, [r0, :64]
+.Lctrtailblock:
+	vst1.8		{q0}, [r0, :64]		@ return just the key stream
 	pop		{r4-r6, pc}
 
 .Lctrcarry:
@@ -405,8 +403,8 @@ ENTRY(ce_aes_xts_encrypt)
 .Lxtsenc3x:
 	subs		r4, r4, #3
 	bmi		.Lxtsenc1x
-	vld1.8		{q0-q1}, [r1, :64]!	@ get 3 pt blocks
-	vld1.8		{q2}, [r1, :64]!
+	vld1.8		{q0-q1}, [r1]!		@ get 3 pt blocks
+	vld1.8		{q2}, [r1]!
 	next_tweak	q4, q3, q7, q6
 	veor		q0, q0, q3
 	next_tweak	q5, q4, q7, q6
@@ -416,8 +414,8 @@ ENTRY(ce_aes_xts_encrypt)
 	veor		q0, q0, q3
 	veor		q1, q1, q4
 	veor		q2, q2, q5
-	vst1.8		{q0-q1}, [r0, :64]!	@ write 3 ct blocks
-	vst1.8		{q2}, [r0, :64]!
+	vst1.8		{q0-q1}, [r0]!		@ write 3 ct blocks
+	vst1.8		{q2}, [r0]!
 	vmov		q3, q5
 	teq		r4, #0
 	beq		.Lxtsencout
@@ -426,11 +424,11 @@ ENTRY(ce_aes_xts_encrypt)
 	adds		r4, r4, #3
 	beq		.Lxtsencout
 .Lxtsencloop:
-	vld1.8		{q0}, [r1, :64]!
+	vld1.8		{q0}, [r1]!
 	veor		q0, q0, q3
 	bl		aes_encrypt
 	veor		q0, q0, q3
-	vst1.8		{q0}, [r0, :64]!
+	vst1.8		{q0}, [r0]!
 	subs		r4, r4, #1
 	beq		.Lxtsencout
 	next_tweak	q3, q3, q7, q6
@@ -456,8 +454,8 @@ ENTRY(ce_aes_xts_decrypt)
 .Lxtsdec3x:
 	subs		r4, r4, #3
 	bmi		.Lxtsdec1x
-	vld1.8		{q0-q1}, [r1, :64]!	@ get 3 ct blocks
-	vld1.8		{q2}, [r1, :64]!
+	vld1.8		{q0-q1}, [r1]!		@ get 3 ct blocks
+	vld1.8		{q2}, [r1]!
 	next_tweak	q4, q3, q7, q6
 	veor		q0, q0, q3
 	next_tweak	q5, q4, q7, q6
@@ -467,8 +465,8 @@ ENTRY(ce_aes_xts_decrypt)
 	veor		q0, q0, q3
 	veor		q1, q1, q4
 	veor		q2, q2, q5
-	vst1.8		{q0-q1}, [r0, :64]!	@ write 3 pt blocks
-	vst1.8		{q2}, [r0, :64]!
+	vst1.8		{q0-q1}, [r0]!		@ write 3 pt blocks
+	vst1.8		{q2}, [r0]!
 	vmov		q3, q5
 	teq		r4, #0
 	beq		.Lxtsdecout
@@ -477,12 +475,12 @@ ENTRY(ce_aes_xts_decrypt)
 	adds		r4, r4, #3
 	beq		.Lxtsdecout
 .Lxtsdecloop:
-	vld1.8		{q0}, [r1, :64]!
+	vld1.8		{q0}, [r1]!
 	veor		q0, q0, q3
 	add		ip, r2, #32		@ 3rd round key
 	bl		aes_decrypt
 	veor		q0, q0, q3
-	vst1.8		{q0}, [r0, :64]!
+	vst1.8		{q0}, [r0]!
 	subs		r4, r4, #1
 	beq		.Lxtsdecout
 	next_tweak	q3, q3, q7, q6
diff --git a/arch/arm/crypto/aes-ce-glue.c b/arch/arm/crypto/aes-ce-glue.c
index 8857531915bf..883b84d828c5 100644
--- a/arch/arm/crypto/aes-ce-glue.c
+++ b/arch/arm/crypto/aes-ce-glue.c
@@ -278,14 +278,15 @@ static int ctr_encrypt(struct skcipher_request *req)
 		u8 *tsrc = walk.src.virt.addr;
 
 		/*
-		 * Minimum alignment is 8 bytes, so if nbytes is <= 8, we need
-		 * to tell aes_ctr_encrypt() to only read half a block.
+		 * Tell aes_ctr_encrypt() to process a tail block.
 		 */
-		blocks = (nbytes <= 8) ? -1 : 1;
+		blocks = -1;
 
-		ce_aes_ctr_encrypt(tail, tsrc, (u8 *)ctx->key_enc,
+		ce_aes_ctr_encrypt(tail, NULL, (u8 *)ctx->key_enc,
 				   num_rounds(ctx), blocks, walk.iv);
-		memcpy(tdst, tail, nbytes);
+		if (tdst != tsrc)
+			memcpy(tdst, tsrc, nbytes);
+		crypto_xor(tdst, tail, nbytes);
 		err = skcipher_walk_done(&walk, 0);
 	}
 	kernel_neon_end();
@@ -345,7 +346,6 @@ static struct skcipher_alg aes_algs[] = { {
 		.cra_flags		= CRYPTO_ALG_INTERNAL,
 		.cra_blocksize		= AES_BLOCK_SIZE,
 		.cra_ctxsize		= sizeof(struct crypto_aes_ctx),
-		.cra_alignmask		= 7,
 		.cra_module		= THIS_MODULE,
 	},
 	.min_keysize	= AES_MIN_KEY_SIZE,
@@ -361,7 +361,6 @@ static struct skcipher_alg aes_algs[] = { {
 		.cra_flags		= CRYPTO_ALG_INTERNAL,
 		.cra_blocksize		= AES_BLOCK_SIZE,
 		.cra_ctxsize		= sizeof(struct crypto_aes_ctx),
-		.cra_alignmask		= 7,
 		.cra_module		= THIS_MODULE,
 	},
 	.min_keysize	= AES_MIN_KEY_SIZE,
@@ -378,7 +377,6 @@ static struct skcipher_alg aes_algs[] = { {
 		.cra_flags		= CRYPTO_ALG_INTERNAL,
 		.cra_blocksize		= 1,
 		.cra_ctxsize		= sizeof(struct crypto_aes_ctx),
-		.cra_alignmask		= 7,
 		.cra_module		= THIS_MODULE,
 	},
 	.min_keysize	= AES_MIN_KEY_SIZE,
@@ -396,7 +394,6 @@ static struct skcipher_alg aes_algs[] = { {
 		.cra_flags		= CRYPTO_ALG_INTERNAL,
 		.cra_blocksize		= AES_BLOCK_SIZE,
 		.cra_ctxsize		= sizeof(struct crypto_aes_xts_ctx),
-		.cra_alignmask		= 7,
 		.cra_module		= THIS_MODULE,
 	},
 	.min_keysize	= 2 * AES_MIN_KEY_SIZE,
diff --git a/arch/arm/crypto/aes-cipher-core.S b/arch/arm/crypto/aes-cipher-core.S
new file mode 100644
index 000000000000..c817a86c4ca8
--- /dev/null
+++ b/arch/arm/crypto/aes-cipher-core.S
@@ -0,0 +1,179 @@
+/*
+ * Scalar AES core transform
+ *
+ * Copyright (C) 2017 Linaro Ltd.
+ * Author: Ard Biesheuvel <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+
+	.text
+	.align		5
+
+	rk		.req	r0
+	rounds		.req	r1
+	in		.req	r2
+	out		.req	r3
+	ttab		.req	ip
+
+	t0		.req	lr
+	t1		.req	r2
+	t2		.req	r3
+
+	.macro		__select, out, in, idx
+	.if		__LINUX_ARM_ARCH__ < 7
+	and		\out, \in, #0xff << (8 * \idx)
+	.else
+	ubfx		\out, \in, #(8 * \idx), #8
+	.endif
+	.endm
+
+	.macro		__load, out, in, idx
+	.if		__LINUX_ARM_ARCH__ < 7 && \idx > 0
+	ldr		\out, [ttab, \in, lsr #(8 * \idx) - 2]
+	.else
+	ldr		\out, [ttab, \in, lsl #2]
+	.endif
+	.endm
+
+	.macro		__hround, out0, out1, in0, in1, in2, in3, t3, t4, enc
+	__select	\out0, \in0, 0
+	__select	t0, \in1, 1
+	__load		\out0, \out0, 0
+	__load		t0, t0, 1
+
+	.if		\enc
+	__select	\out1, \in1, 0
+	__select	t1, \in2, 1
+	.else
+	__select	\out1, \in3, 0
+	__select	t1, \in0, 1
+	.endif
+	__load		\out1, \out1, 0
+	__select	t2, \in2, 2
+	__load		t1, t1, 1
+	__load		t2, t2, 2
+
+	eor		\out0, \out0, t0, ror #24
+
+	__select	t0, \in3, 3
+	.if		\enc
+	__select	\t3, \in3, 2
+	__select	\t4, \in0, 3
+	.else
+	__select	\t3, \in1, 2
+	__select	\t4, \in2, 3
+	.endif
+	__load		\t3, \t3, 2
+	__load		t0, t0, 3
+	__load		\t4, \t4, 3
+
+	eor		\out1, \out1, t1, ror #24
+	eor		\out0, \out0, t2, ror #16
+	ldm		rk!, {t1, t2}
+	eor		\out1, \out1, \t3, ror #16
+	eor		\out0, \out0, t0, ror #8
+	eor		\out1, \out1, \t4, ror #8
+	eor		\out0, \out0, t1
+	eor		\out1, \out1, t2
+	.endm
+
+	.macro		fround, out0, out1, out2, out3, in0, in1, in2, in3
+	__hround	\out0, \out1, \in0, \in1, \in2, \in3, \out2, \out3, 1
+	__hround	\out2, \out3, \in2, \in3, \in0, \in1, \in1, \in2, 1
+	.endm
+
+	.macro		iround, out0, out1, out2, out3, in0, in1, in2, in3
+	__hround	\out0, \out1, \in0, \in3, \in2, \in1, \out2, \out3, 0
+	__hround	\out2, \out3, \in2, \in1, \in0, \in3, \in1, \in0, 0
+	.endm
+
+	.macro		__rev, out, in
+	.if		__LINUX_ARM_ARCH__ < 6
+	lsl		t0, \in, #24
+	and		t1, \in, #0xff00
+	and		t2, \in, #0xff0000
+	orr		\out, t0, \in, lsr #24
+	orr		\out, \out, t1, lsl #8
+	orr		\out, \out, t2, lsr #8
+	.else
+	rev		\out, \in
+	.endif
+	.endm
+
+	.macro		__adrl, out, sym, c
+	.if		__LINUX_ARM_ARCH__ < 7
+	ldr\c		\out, =\sym
+	.else
+	movw\c		\out, #:lower16:\sym
+	movt\c		\out, #:upper16:\sym
+	.endif
+	.endm
+
+	.macro		do_crypt, round, ttab, ltab
+	push		{r3-r11, lr}
+
+	ldr		r4, [in]
+	ldr		r5, [in, #4]
+	ldr		r6, [in, #8]
+	ldr		r7, [in, #12]
+
+	ldm		rk!, {r8-r11}
+
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	__rev		r4, r4
+	__rev		r5, r5
+	__rev		r6, r6
+	__rev		r7, r7
+#endif
+
+	eor		r4, r4, r8
+	eor		r5, r5, r9
+	eor		r6, r6, r10
+	eor		r7, r7, r11
+
+	__adrl		ttab, \ttab
+
+	tst		rounds, #2
+	bne		1f
+
+0:	\round		r8, r9, r10, r11, r4, r5, r6, r7
+	\round		r4, r5, r6, r7, r8, r9, r10, r11
+
+1:	subs		rounds, rounds, #4
+	\round		r8, r9, r10, r11, r4, r5, r6, r7
+	__adrl		ttab, \ltab, ls
+	\round		r4, r5, r6, r7, r8, r9, r10, r11
+	bhi		0b
+
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	__rev		r4, r4
+	__rev		r5, r5
+	__rev		r6, r6
+	__rev		r7, r7
+#endif
+
+	ldr		out, [sp]
+
+	str		r4, [out]
+	str		r5, [out, #4]
+	str		r6, [out, #8]
+	str		r7, [out, #12]
+
+	pop		{r3-r11, pc}
+
+	.align		3
+	.ltorg
+	.endm
+
+ENTRY(__aes_arm_encrypt)
+	do_crypt	fround, crypto_ft_tab, crypto_fl_tab
+ENDPROC(__aes_arm_encrypt)
+
+ENTRY(__aes_arm_decrypt)
+	do_crypt	iround, crypto_it_tab, crypto_il_tab
+ENDPROC(__aes_arm_decrypt)
diff --git a/arch/arm/crypto/aes-cipher-glue.c b/arch/arm/crypto/aes-cipher-glue.c
new file mode 100644
index 000000000000..c222f6e072ad
--- /dev/null
+++ b/arch/arm/crypto/aes-cipher-glue.c
@@ -0,0 +1,74 @@
+/*
+ * Scalar AES core transform
+ *
+ * Copyright (C) 2017 Linaro Ltd.
+ * Author: Ard Biesheuvel <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <crypto/aes.h>
+#include <linux/crypto.h>
+#include <linux/module.h>
+
+asmlinkage void __aes_arm_encrypt(u32 *rk, int rounds, const u8 *in, u8 *out);
+EXPORT_SYMBOL(__aes_arm_encrypt);
+
+asmlinkage void __aes_arm_decrypt(u32 *rk, int rounds, const u8 *in, u8 *out);
+EXPORT_SYMBOL(__aes_arm_decrypt);
+
+static void aes_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+	struct crypto_aes_ctx *ctx = crypto_tfm_ctx(tfm);
+	int rounds = 6 + ctx->key_length / 4;
+
+	__aes_arm_encrypt(ctx->key_enc, rounds, in, out);
+}
+
+static void aes_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+	struct crypto_aes_ctx *ctx = crypto_tfm_ctx(tfm);
+	int rounds = 6 + ctx->key_length / 4;
+
+	__aes_arm_decrypt(ctx->key_dec, rounds, in, out);
+}
+
+static struct crypto_alg aes_alg = {
+	.cra_name			= "aes",
+	.cra_driver_name		= "aes-arm",
+	.cra_priority			= 200,
+	.cra_flags			= CRYPTO_ALG_TYPE_CIPHER,
+	.cra_blocksize			= AES_BLOCK_SIZE,
+	.cra_ctxsize			= sizeof(struct crypto_aes_ctx),
+	.cra_module			= THIS_MODULE,
+
+	.cra_cipher.cia_min_keysize	= AES_MIN_KEY_SIZE,
+	.cra_cipher.cia_max_keysize	= AES_MAX_KEY_SIZE,
+	.cra_cipher.cia_setkey		= crypto_aes_set_key,
+	.cra_cipher.cia_encrypt		= aes_encrypt,
+	.cra_cipher.cia_decrypt		= aes_decrypt,
+
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+	.cra_alignmask			= 3,
+#endif
+};
+
+static int __init aes_init(void)
+{
+	return crypto_register_alg(&aes_alg);
+}
+
+static void __exit aes_fini(void)
+{
+	crypto_unregister_alg(&aes_alg);
+}
+
+module_init(aes_init);
+module_exit(aes_fini);
+
+MODULE_DESCRIPTION("Scalar AES cipher for ARM");
+MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS_CRYPTO("aes");
diff --git a/arch/arm/crypto/aes-neonbs-core.S b/arch/arm/crypto/aes-neonbs-core.S
new file mode 100644
index 000000000000..2b625c6d4712
--- /dev/null
+++ b/arch/arm/crypto/aes-neonbs-core.S
@@ -0,0 +1,1023 @@
+/*
+ * Bit sliced AES using NEON instructions
+ *
+ * Copyright (C) 2017 Linaro Ltd.
+ * Author: Ard Biesheuvel <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/*
+ * The algorithm implemented here is described in detail by the paper
+ * 'Faster and Timing-Attack Resistant AES-GCM' by Emilia Kaesper and
+ * Peter Schwabe (https://eprint.iacr.org/2009/129.pdf)
+ *
+ * This implementation is based primarily on the OpenSSL implementation
+ * for 32-bit ARM written by Andy Polyakov <appro@openssl.org>
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+	.text
+	.fpu		neon
+
+	rounds		.req	ip
+	bskey		.req	r4
+
+	q0l		.req	d0
+	q0h		.req	d1
+	q1l		.req	d2
+	q1h		.req	d3
+	q2l		.req	d4
+	q2h		.req	d5
+	q3l		.req	d6
+	q3h		.req	d7
+	q4l		.req	d8
+	q4h		.req	d9
+	q5l		.req	d10
+	q5h		.req	d11
+	q6l		.req	d12
+	q6h		.req	d13
+	q7l		.req	d14
+	q7h		.req	d15
+	q8l		.req	d16
+	q8h		.req	d17
+	q9l		.req	d18
+	q9h		.req	d19
+	q10l		.req	d20
+	q10h		.req	d21
+	q11l		.req	d22
+	q11h		.req	d23
+	q12l		.req	d24
+	q12h		.req	d25
+	q13l		.req	d26
+	q13h		.req	d27
+	q14l		.req	d28
+	q14h		.req	d29
+	q15l		.req	d30
+	q15h		.req	d31
+
+	.macro		__tbl, out, tbl, in, tmp
+	.ifc		\out, \tbl
+	.ifb		\tmp
+	.error		__tbl needs temp register if out == tbl
+	.endif
+	vmov		\tmp, \out
+	.endif
+	vtbl.8		\out\()l, {\tbl}, \in\()l
+	.ifc		\out, \tbl
+	vtbl.8		\out\()h, {\tmp}, \in\()h
+	.else
+	vtbl.8		\out\()h, {\tbl}, \in\()h
+	.endif
+	.endm
+
+	.macro		__ldr, out, sym
+	vldr		\out\()l, \sym
+	vldr		\out\()h, \sym + 8
+	.endm
+
+	.macro		__adr, reg, lbl
+	adr		\reg, \lbl
+THUMB(	orr		\reg, \reg, #1		)
+	.endm
+
+	.macro		in_bs_ch, b0, b1, b2, b3, b4, b5, b6, b7
+	veor		\b2, \b2, \b1
+	veor		\b5, \b5, \b6
+	veor		\b3, \b3, \b0
+	veor		\b6, \b6, \b2
+	veor		\b5, \b5, \b0
+	veor		\b6, \b6, \b3
+	veor		\b3, \b3, \b7
+	veor		\b7, \b7, \b5
+	veor		\b3, \b3, \b4
+	veor		\b4, \b4, \b5
+	veor		\b2, \b2, \b7
+	veor		\b3, \b3, \b1
+	veor		\b1, \b1, \b5
+	.endm
+
+	.macro		out_bs_ch, b0, b1, b2, b3, b4, b5, b6, b7
+	veor		\b0, \b0, \b6
+	veor		\b1, \b1, \b4
+	veor		\b4, \b4, \b6
+	veor		\b2, \b2, \b0
+	veor		\b6, \b6, \b1
+	veor		\b1, \b1, \b5
+	veor		\b5, \b5, \b3
+	veor		\b3, \b3, \b7
+	veor		\b7, \b7, \b5
+	veor		\b2, \b2, \b5
+	veor		\b4, \b4, \b7
+	.endm
+
+	.macro		inv_in_bs_ch, b6, b1, b2, b4, b7, b0, b3, b5
+	veor		\b1, \b1, \b7
+	veor		\b4, \b4, \b7
+	veor		\b7, \b7, \b5
+	veor		\b1, \b1, \b3
+	veor		\b2, \b2, \b5
+	veor		\b3, \b3, \b7
+	veor		\b6, \b6, \b1
+	veor		\b2, \b2, \b0
+	veor		\b5, \b5, \b3
+	veor		\b4, \b4, \b6
+	veor		\b0, \b0, \b6
+	veor		\b1, \b1, \b4
+	.endm
+
+	.macro		inv_out_bs_ch, b6, b5, b0, b3, b7, b1, b4, b2
+	veor		\b1, \b1, \b5
+	veor		\b2, \b2, \b7
+	veor		\b3, \b3, \b1
+	veor		\b4, \b4, \b5
+	veor		\b7, \b7, \b5
+	veor		\b3, \b3, \b4
+	veor 		\b5, \b5, \b0
+	veor		\b3, \b3, \b7
+	veor		\b6, \b6, \b2
+	veor		\b2, \b2, \b1
+	veor		\b6, \b6, \b3
+	veor		\b3, \b3, \b0
+	veor		\b5, \b5, \b6
+	.endm
+
+	.macro		mul_gf4, x0, x1, y0, y1, t0, t1
+	veor 		\t0, \y0, \y1
+	vand		\t0, \t0, \x0
+	veor		\x0, \x0, \x1
+	vand		\t1, \x1, \y0
+	vand		\x0, \x0, \y1
+	veor		\x1, \t1, \t0
+	veor		\x0, \x0, \t1
+	.endm
+
+	.macro		mul_gf4_n_gf4, x0, x1, y0, y1, t0, x2, x3, y2, y3, t1
+	veor		\t0, \y0, \y1
+	veor 		\t1, \y2, \y3
+	vand		\t0, \t0, \x0
+	vand		\t1, \t1, \x2
+	veor		\x0, \x0, \x1
+	veor		\x2, \x2, \x3
+	vand		\x1, \x1, \y0
+	vand		\x3, \x3, \y2
+	vand		\x0, \x0, \y1
+	vand		\x2, \x2, \y3
+	veor		\x1, \x1, \x0
+	veor		\x2, \x2, \x3
+	veor		\x0, \x0, \t0
+	veor		\x3, \x3, \t1
+	.endm
+
+	.macro		mul_gf16_2, x0, x1, x2, x3, x4, x5, x6, x7, \
+				    y0, y1, y2, y3, t0, t1, t2, t3
+	veor		\t0, \x0, \x2
+	veor		\t1, \x1, \x3
+	mul_gf4  	\x0, \x1, \y0, \y1, \t2, \t3
+	veor		\y0, \y0, \y2
+	veor		\y1, \y1, \y3
+	mul_gf4_n_gf4	\t0, \t1, \y0, \y1, \t3, \x2, \x3, \y2, \y3, \t2
+	veor		\x0, \x0, \t0
+	veor		\x2, \x2, \t0
+	veor		\x1, \x1, \t1
+	veor		\x3, \x3, \t1
+	veor		\t0, \x4, \x6
+	veor		\t1, \x5, \x7
+	mul_gf4_n_gf4	\t0, \t1, \y0, \y1, \t3, \x6, \x7, \y2, \y3, \t2
+	veor		\y0, \y0, \y2
+	veor		\y1, \y1, \y3
+	mul_gf4  	\x4, \x5, \y0, \y1, \t2, \t3
+	veor		\x4, \x4, \t0
+	veor		\x6, \x6, \t0
+	veor		\x5, \x5, \t1
+	veor		\x7, \x7, \t1
+	.endm
+
+	.macro		inv_gf256, x0, x1, x2, x3, x4, x5, x6, x7, \
+				   t0, t1, t2, t3, s0, s1, s2, s3
+	veor		\t3, \x4, \x6
+	veor		\t0, \x5, \x7
+	veor		\t1, \x1, \x3
+	veor		\s1, \x7, \x6
+	veor		\s0, \x0, \x2
+	veor		\s3, \t3, \t0
+	vorr		\t2, \t0, \t1
+	vand		\s2, \t3, \s0
+	vorr		\t3, \t3, \s0
+	veor		\s0, \s0, \t1
+	vand		\t0, \t0, \t1
+	veor		\t1, \x3, \x2
+	vand		\s3, \s3, \s0
+	vand		\s1, \s1, \t1
+	veor		\t1, \x4, \x5
+	veor		\s0, \x1, \x0
+	veor		\t3, \t3, \s1
+	veor		\t2, \t2, \s1
+	vand		\s1, \t1, \s0
+	vorr		\t1, \t1, \s0
+	veor		\t3, \t3, \s3
+	veor		\t0, \t0, \s1
+	veor		\t2, \t2, \s2
+	veor		\t1, \t1, \s3
+	veor		\t0, \t0, \s2
+	vand		\s0, \x7, \x3
+	veor		\t1, \t1, \s2
+	vand		\s1, \x6, \x2
+	vand		\s2, \x5, \x1
+	vorr		\s3, \x4, \x0
+	veor		\t3, \t3, \s0
+	veor		\t1, \t1, \s2
+	veor		\s0, \t0, \s3
+	veor		\t2, \t2, \s1
+	vand		\s2, \t3, \t1
+	veor		\s1, \t2, \s2
+	veor		\s3, \s0, \s2
+	vbsl		\s1, \t1, \s0
+	vmvn		\t0, \s0
+	vbsl		\s0, \s1, \s3
+	vbsl		\t0, \s1, \s3
+	vbsl		\s3, \t3, \t2
+	veor		\t3, \t3, \t2
+	vand		\s2, \s0, \s3
+	veor		\t1, \t1, \t0
+	veor		\s2, \s2, \t3
+	mul_gf16_2	\x0, \x1, \x2, \x3, \x4, \x5, \x6, \x7, \
+			\s3, \s2, \s1, \t1, \s0, \t0, \t2, \t3
+	.endm
+
+	.macro		sbox, b0, b1, b2, b3, b4, b5, b6, b7, \
+			      t0, t1, t2, t3, s0, s1, s2, s3
+	in_bs_ch	\b0, \b1, \b2, \b3, \b4, \b5, \b6, \b7
+	inv_gf256	\b6, \b5, \b0, \b3, \b7, \b1, \b4, \b2, \
+			\t0, \t1, \t2, \t3, \s0, \s1, \s2, \s3
+	out_bs_ch	\b7, \b1, \b4, \b2, \b6, \b5, \b0, \b3
+	.endm
+
+	.macro		inv_sbox, b0, b1, b2, b3, b4, b5, b6, b7, \
+				  t0, t1, t2, t3, s0, s1, s2, s3
+	inv_in_bs_ch	\b0, \b1, \b2, \b3, \b4, \b5, \b6, \b7
+	inv_gf256	\b5, \b1, \b2, \b6, \b3, \b7, \b0, \b4, \
+			\t0, \t1, \t2, \t3, \s0, \s1, \s2, \s3
+	inv_out_bs_ch	\b3, \b7, \b0, \b4, \b5, \b1, \b2, \b6
+	.endm
+
+	.macro		shift_rows, x0, x1, x2, x3, x4, x5, x6, x7, \
+				    t0, t1, t2, t3, mask
+	vld1.8		{\t0-\t1}, [bskey, :256]!
+	veor		\t0, \t0, \x0
+	vld1.8		{\t2-\t3}, [bskey, :256]!
+	veor		\t1, \t1, \x1
+	__tbl		\x0, \t0, \mask
+	veor		\t2, \t2, \x2
+	__tbl		\x1, \t1, \mask
+	vld1.8		{\t0-\t1}, [bskey, :256]!
+	veor		\t3, \t3, \x3
+	__tbl		\x2, \t2, \mask
+	__tbl		\x3, \t3, \mask
+	vld1.8		{\t2-\t3}, [bskey, :256]!
+	veor		\t0, \t0, \x4
+	veor		\t1, \t1, \x5
+	__tbl		\x4, \t0, \mask
+	veor		\t2, \t2, \x6
+	__tbl		\x5, \t1, \mask
+	veor		\t3, \t3, \x7
+	__tbl		\x6, \t2, \mask
+	__tbl		\x7, \t3, \mask
+	.endm
+
+	.macro		inv_shift_rows, x0, x1, x2, x3, x4, x5, x6, x7, \
+					t0, t1, t2, t3, mask
+	__tbl		\x0, \x0, \mask, \t0
+	__tbl		\x1, \x1, \mask, \t1
+	__tbl		\x2, \x2, \mask, \t2
+	__tbl		\x3, \x3, \mask, \t3
+	__tbl		\x4, \x4, \mask, \t0
+	__tbl		\x5, \x5, \mask, \t1
+	__tbl		\x6, \x6, \mask, \t2
+	__tbl		\x7, \x7, \mask, \t3
+	.endm
+
+	.macro		mix_cols, x0, x1, x2, x3, x4, x5, x6, x7, \
+				  t0, t1, t2, t3, t4, t5, t6, t7, inv
+	vext.8		\t0, \x0, \x0, #12
+	vext.8		\t1, \x1, \x1, #12
+	veor		\x0, \x0, \t0
+	vext.8		\t2, \x2, \x2, #12
+	veor		\x1, \x1, \t1
+	vext.8		\t3, \x3, \x3, #12
+	veor		\x2, \x2, \t2
+	vext.8		\t4, \x4, \x4, #12
+	veor		\x3, \x3, \t3
+	vext.8		\t5, \x5, \x5, #12
+	veor		\x4, \x4, \t4
+	vext.8		\t6, \x6, \x6, #12
+	veor		\x5, \x5, \t5
+	vext.8		\t7, \x7, \x7, #12
+	veor		\x6, \x6, \t6
+	veor		\t1, \t1, \x0
+	veor.8		\x7, \x7, \t7
+	vext.8		\x0, \x0, \x0, #8
+	veor		\t2, \t2, \x1
+	veor		\t0, \t0, \x7
+	veor		\t1, \t1, \x7
+	vext.8		\x1, \x1, \x1, #8
+	veor		\t5, \t5, \x4
+	veor		\x0, \x0, \t0
+	veor		\t6, \t6, \x5
+	veor		\x1, \x1, \t1
+	vext.8		\t0, \x4, \x4, #8
+	veor		\t4, \t4, \x3
+	vext.8		\t1, \x5, \x5, #8
+	veor		\t7, \t7, \x6
+	vext.8		\x4, \x3, \x3, #8
+	veor		\t3, \t3, \x2
+	vext.8		\x5, \x7, \x7, #8
+	veor		\t4, \t4, \x7
+	vext.8		\x3, \x6, \x6, #8
+	veor		\t3, \t3, \x7
+	vext.8		\x6, \x2, \x2, #8
+	veor		\x7, \t1, \t5
+	.ifb		\inv
+	veor		\x2, \t0, \t4
+	veor		\x4, \x4, \t3
+	veor		\x5, \x5, \t7
+	veor		\x3, \x3, \t6
+	veor		\x6, \x6, \t2
+	.else
+	veor		\t3, \t3, \x4
+	veor		\x5, \x5, \t7
+	veor		\x2, \x3, \t6
+	veor		\x3, \t0, \t4
+	veor		\x4, \x6, \t2
+	vmov		\x6, \t3
+	.endif
+	.endm
+
+	.macro		inv_mix_cols, x0, x1, x2, x3, x4, x5, x6, x7, \
+				      t0, t1, t2, t3, t4, t5, t6, t7
+	vld1.8		{\t0-\t1}, [bskey, :256]!
+	veor		\x0, \x0, \t0
+	vld1.8		{\t2-\t3}, [bskey, :256]!
+	veor		\x1, \x1, \t1
+	vld1.8		{\t4-\t5}, [bskey, :256]!
+	veor		\x2, \x2, \t2
+	vld1.8		{\t6-\t7}, [bskey, :256]
+	sub		bskey, bskey, #224
+	veor		\x3, \x3, \t3
+	veor		\x4, \x4, \t4
+	veor		\x5, \x5, \t5
+	veor		\x6, \x6, \t6
+	veor		\x7, \x7, \t7
+	vext.8		\t0, \x0, \x0, #8
+	vext.8		\t6, \x6, \x6, #8
+	vext.8		\t7, \x7, \x7, #8
+	veor		\t0, \t0, \x0
+	vext.8		\t1, \x1, \x1, #8
+	veor		\t6, \t6, \x6
+	vext.8		\t2, \x2, \x2, #8
+	veor		\t7, \t7, \x7
+	vext.8		\t3, \x3, \x3, #8
+	veor		\t1, \t1, \x1
+	vext.8		\t4, \x4, \x4, #8
+	veor		\t2, \t2, \x2
+	vext.8		\t5, \x5, \x5, #8
+	veor		\t3, \t3, \x3
+	veor		\t4, \t4, \x4
+	veor		\t5, \t5, \x5
+	veor		\x0, \x0, \t6
+	veor		\x1, \x1, \t6
+	veor		\x2, \x2, \t0
+	veor		\x4, \x4, \t2
+	veor		\x3, \x3, \t1
+	veor		\x1, \x1, \t7
+	veor		\x2, \x2, \t7
+	veor		\x4, \x4, \t6
+	veor		\x5, \x5, \t3
+	veor		\x3, \x3, \t6
+	veor		\x6, \x6, \t4
+	veor		\x4, \x4, \t7
+	veor		\x5, \x5, \t7
+	veor		\x7, \x7, \t5
+	mix_cols	\x0, \x1, \x2, \x3, \x4, \x5, \x6, \x7, \
+			\t0, \t1, \t2, \t3, \t4, \t5, \t6, \t7, 1
+	.endm
+
+	.macro		swapmove_2x, a0, b0, a1, b1, n, mask, t0, t1
+	vshr.u64	\t0, \b0, #\n
+	vshr.u64	\t1, \b1, #\n
+	veor		\t0, \t0, \a0
+	veor		\t1, \t1, \a1
+	vand		\t0, \t0, \mask
+	vand		\t1, \t1, \mask
+	veor		\a0, \a0, \t0
+	vshl.s64	\t0, \t0, #\n
+	veor		\a1, \a1, \t1
+	vshl.s64	\t1, \t1, #\n
+	veor		\b0, \b0, \t0
+	veor		\b1, \b1, \t1
+	.endm
+
+	.macro		bitslice, x7, x6, x5, x4, x3, x2, x1, x0, t0, t1, t2, t3
+	vmov.i8		\t0, #0x55
+	vmov.i8		\t1, #0x33
+	swapmove_2x	\x0, \x1, \x2, \x3, 1, \t0, \t2, \t3
+	swapmove_2x	\x4, \x5, \x6, \x7, 1, \t0, \t2, \t3
+	vmov.i8		\t0, #0x0f
+	swapmove_2x	\x0, \x2, \x1, \x3, 2, \t1, \t2, \t3
+	swapmove_2x	\x4, \x6, \x5, \x7, 2, \t1, \t2, \t3
+	swapmove_2x	\x0, \x4, \x1, \x5, 4, \t0, \t2, \t3
+	swapmove_2x	\x2, \x6, \x3, \x7, 4, \t0, \t2, \t3
+	.endm
+
+	.align		4
+M0:	.quad		0x02060a0e03070b0f, 0x0004080c0105090d
+
+	/*
+	 * void aesbs_convert_key(u8 out[], u32 const rk[], int rounds)
+	 */
+ENTRY(aesbs_convert_key)
+	vld1.32		{q7}, [r1]!		// load round 0 key
+	vld1.32		{q15}, [r1]!		// load round 1 key
+
+	vmov.i8		q8,  #0x01		// bit masks
+	vmov.i8		q9,  #0x02
+	vmov.i8		q10, #0x04
+	vmov.i8		q11, #0x08
+	vmov.i8		q12, #0x10
+	vmov.i8		q13, #0x20
+	__ldr		q14, M0
+
+	sub		r2, r2, #1
+	vst1.8		{q7}, [r0, :128]!	// save round 0 key
+
+.Lkey_loop:
+	__tbl		q7, q15, q14
+	vmov.i8		q6, #0x40
+	vmov.i8		q15, #0x80
+
+	vtst.8		q0, q7, q8
+	vtst.8		q1, q7, q9
+	vtst.8		q2, q7, q10
+	vtst.8		q3, q7, q11
+	vtst.8		q4, q7, q12
+	vtst.8		q5, q7, q13
+	vtst.8		q6, q7, q6
+	vtst.8		q7, q7, q15
+	vld1.32		{q15}, [r1]!		// load next round key
+	vmvn		q0, q0
+	vmvn		q1, q1
+	vmvn		q5, q5
+	vmvn		q6, q6
+
+	subs		r2, r2, #1
+	vst1.8		{q0-q1}, [r0, :256]!
+	vst1.8		{q2-q3}, [r0, :256]!
+	vst1.8		{q4-q5}, [r0, :256]!
+	vst1.8		{q6-q7}, [r0, :256]!
+	bne		.Lkey_loop
+
+	vmov.i8		q7, #0x63		// compose .L63
+	veor		q15, q15, q7
+	vst1.8		{q15}, [r0, :128]
+	bx		lr
+ENDPROC(aesbs_convert_key)
+
+	.align		4
+M0SR:	.quad		0x0a0e02060f03070b, 0x0004080c05090d01
+
+aesbs_encrypt8:
+	vld1.8		{q9}, [bskey, :128]!	// round 0 key
+	__ldr		q8, M0SR
+
+	veor		q10, q0, q9		// xor with round0 key
+	veor		q11, q1, q9
+	__tbl		q0, q10, q8
+	veor		q12, q2, q9
+	__tbl		q1, q11, q8
+	veor		q13, q3, q9
+	__tbl		q2, q12, q8
+	veor		q14, q4, q9
+	__tbl		q3, q13, q8
+	veor		q15, q5, q9
+	__tbl		q4, q14, q8
+	veor		q10, q6, q9
+	__tbl		q5, q15, q8
+	veor		q11, q7, q9
+	__tbl		q6, q10, q8
+	__tbl		q7, q11, q8
+
+	bitslice	q0, q1, q2, q3, q4, q5, q6, q7, q8, q9, q10, q11
+
+	sub		rounds, rounds, #1
+	b		.Lenc_sbox
+
+	.align		5
+SR:	.quad		0x0504070600030201, 0x0f0e0d0c0a09080b
+SRM0:	.quad		0x0304090e00050a0f, 0x01060b0c0207080d
+
+.Lenc_last:
+	__ldr		q12, SRM0
+.Lenc_loop:
+	shift_rows	q0, q1, q2, q3, q4, q5, q6, q7, q8, q9, q10, q11, q12
+.Lenc_sbox:
+	sbox		q0, q1, q2, q3, q4, q5, q6, q7, q8, q9, q10, q11, q12, \
+								q13, q14, q15
+	subs		rounds, rounds, #1
+	bcc		.Lenc_done
+
+	mix_cols	q0, q1, q4, q6, q3, q7, q2, q5, q8, q9, q10, q11, q12, \
+								q13, q14, q15
+
+	beq		.Lenc_last
+	__ldr		q12, SR
+	b		.Lenc_loop
+
+.Lenc_done:
+	vld1.8		{q12}, [bskey, :128]	// last round key
+
+	bitslice	q0, q1, q4, q6, q3, q7, q2, q5, q8, q9, q10, q11
+
+	veor		q0, q0, q12
+	veor		q1, q1, q12
+	veor		q4, q4, q12
+	veor		q6, q6, q12
+	veor		q3, q3, q12
+	veor		q7, q7, q12
+	veor		q2, q2, q12
+	veor		q5, q5, q12
+	bx		lr
+ENDPROC(aesbs_encrypt8)
+
+	.align		4
+M0ISR:	.quad		0x0a0e0206070b0f03, 0x0004080c0d010509
+
+aesbs_decrypt8:
+	add		bskey, bskey, rounds, lsl #7
+	sub		bskey, bskey, #112
+	vld1.8		{q9}, [bskey, :128]	// round 0 key
+	sub		bskey, bskey, #128
+	__ldr		q8, M0ISR
+
+	veor		q10, q0, q9		// xor with round0 key
+	veor		q11, q1, q9
+	__tbl		q0, q10, q8
+	veor		q12, q2, q9
+	__tbl		q1, q11, q8
+	veor		q13, q3, q9
+	__tbl		q2, q12, q8
+	veor		q14, q4, q9
+	__tbl		q3, q13, q8
+	veor		q15, q5, q9
+	__tbl		q4, q14, q8
+	veor		q10, q6, q9
+	__tbl		q5, q15, q8
+	veor		q11, q7, q9
+	__tbl		q6, q10, q8
+	__tbl		q7, q11, q8
+
+	bitslice	q0, q1, q2, q3, q4, q5, q6, q7, q8, q9, q10, q11
+
+	sub		rounds, rounds, #1
+	b		.Ldec_sbox
+
+	.align		5
+ISR:	.quad		0x0504070602010003, 0x0f0e0d0c080b0a09
+ISRM0:	.quad		0x01040b0e0205080f, 0x0306090c00070a0d
+
+.Ldec_last:
+	__ldr		q12, ISRM0
+.Ldec_loop:
+	inv_shift_rows	q0, q1, q2, q3, q4, q5, q6, q7, q8, q9, q10, q11, q12
+.Ldec_sbox:
+	inv_sbox	q0, q1, q2, q3, q4, q5, q6, q7, q8, q9, q10, q11, q12, \
+								q13, q14, q15
+	subs		rounds, rounds, #1
+	bcc		.Ldec_done
+
+	inv_mix_cols	q0, q1, q6, q4, q2, q7, q3, q5, q8, q9, q10, q11, q12, \
+								q13, q14, q15
+
+	beq		.Ldec_last
+	__ldr		q12, ISR
+	b		.Ldec_loop
+
+.Ldec_done:
+	add		bskey, bskey, #112
+	vld1.8		{q12}, [bskey, :128]	// last round key
+
+	bitslice	q0, q1, q6, q4, q2, q7, q3, q5, q8, q9, q10, q11
+
+	veor		q0, q0, q12
+	veor		q1, q1, q12
+	veor		q6, q6, q12
+	veor		q4, q4, q12
+	veor		q2, q2, q12
+	veor		q7, q7, q12
+	veor		q3, q3, q12
+	veor		q5, q5, q12
+	bx		lr
+ENDPROC(aesbs_decrypt8)
+
+	/*
+	 * aesbs_ecb_encrypt(u8 out[], u8 const in[], u8 const rk[], int rounds,
+	 *		     int blocks)
+	 * aesbs_ecb_decrypt(u8 out[], u8 const in[], u8 const rk[], int rounds,
+	 *		     int blocks)
+	 */
+	.macro		__ecb_crypt, do8, o0, o1, o2, o3, o4, o5, o6, o7
+	push		{r4-r6, lr}
+	ldr		r5, [sp, #16]		// number of blocks
+
+99:	__adr		ip, 0f
+	and		lr, r5, #7
+	cmp		r5, #8
+	sub		ip, ip, lr, lsl #2
+	bxlt		ip			// computed goto if blocks < 8
+
+	vld1.8		{q0}, [r1]!
+	vld1.8		{q1}, [r1]!
+	vld1.8		{q2}, [r1]!
+	vld1.8		{q3}, [r1]!
+	vld1.8		{q4}, [r1]!
+	vld1.8		{q5}, [r1]!
+	vld1.8		{q6}, [r1]!
+	vld1.8		{q7}, [r1]!
+
+0:	mov		bskey, r2
+	mov		rounds, r3
+	bl		\do8
+
+	__adr		ip, 1f
+	and		lr, r5, #7
+	cmp		r5, #8
+	sub		ip, ip, lr, lsl #2
+	bxlt		ip			// computed goto if blocks < 8
+
+	vst1.8		{\o0}, [r0]!
+	vst1.8		{\o1}, [r0]!
+	vst1.8		{\o2}, [r0]!
+	vst1.8		{\o3}, [r0]!
+	vst1.8		{\o4}, [r0]!
+	vst1.8		{\o5}, [r0]!
+	vst1.8		{\o6}, [r0]!
+	vst1.8		{\o7}, [r0]!
+
+1:	subs		r5, r5, #8
+	bgt		99b
+
+	pop		{r4-r6, pc}
+	.endm
+
+	.align		4
+ENTRY(aesbs_ecb_encrypt)
+	__ecb_crypt	aesbs_encrypt8, q0, q1, q4, q6, q3, q7, q2, q5
+ENDPROC(aesbs_ecb_encrypt)
+
+	.align		4
+ENTRY(aesbs_ecb_decrypt)
+	__ecb_crypt	aesbs_decrypt8, q0, q1, q6, q4, q2, q7, q3, q5
+ENDPROC(aesbs_ecb_decrypt)
+
+	/*
+	 * aesbs_cbc_decrypt(u8 out[], u8 const in[], u8 const rk[],
+	 *		     int rounds, int blocks, u8 iv[])
+	 */
+	.align		4
+ENTRY(aesbs_cbc_decrypt)
+	mov		ip, sp
+	push		{r4-r6, lr}
+	ldm		ip, {r5-r6}		// load args 4-5
+
+99:	__adr		ip, 0f
+	and		lr, r5, #7
+	cmp		r5, #8
+	sub		ip, ip, lr, lsl #2
+	mov		lr, r1
+	bxlt		ip			// computed goto if blocks < 8
+
+	vld1.8		{q0}, [lr]!
+	vld1.8		{q1}, [lr]!
+	vld1.8		{q2}, [lr]!
+	vld1.8		{q3}, [lr]!
+	vld1.8		{q4}, [lr]!
+	vld1.8		{q5}, [lr]!
+	vld1.8		{q6}, [lr]!
+	vld1.8		{q7}, [lr]
+
+0:	mov		bskey, r2
+	mov		rounds, r3
+	bl		aesbs_decrypt8
+
+	vld1.8		{q8}, [r6]
+	vmov		q9, q8
+	vmov		q10, q8
+	vmov		q11, q8
+	vmov		q12, q8
+	vmov		q13, q8
+	vmov		q14, q8
+	vmov		q15, q8
+
+	__adr		ip, 1f
+	and		lr, r5, #7
+	cmp		r5, #8
+	sub		ip, ip, lr, lsl #2
+	bxlt		ip			// computed goto if blocks < 8
+
+	vld1.8		{q9}, [r1]!
+	vld1.8		{q10}, [r1]!
+	vld1.8		{q11}, [r1]!
+	vld1.8		{q12}, [r1]!
+	vld1.8		{q13}, [r1]!
+	vld1.8		{q14}, [r1]!
+	vld1.8		{q15}, [r1]!
+	W(nop)
+
+1:	__adr		ip, 2f
+	sub		ip, ip, lr, lsl #3
+	bxlt		ip			// computed goto if blocks < 8
+
+	veor		q0, q0, q8
+	vst1.8		{q0}, [r0]!
+	veor		q1, q1, q9
+	vst1.8		{q1}, [r0]!
+	veor		q6, q6, q10
+	vst1.8		{q6}, [r0]!
+	veor		q4, q4, q11
+	vst1.8		{q4}, [r0]!
+	veor		q2, q2, q12
+	vst1.8		{q2}, [r0]!
+	veor		q7, q7, q13
+	vst1.8		{q7}, [r0]!
+	veor		q3, q3, q14
+	vst1.8		{q3}, [r0]!
+	veor		q5, q5, q15
+	vld1.8		{q8}, [r1]!		// load next round's iv
+2:	vst1.8		{q5}, [r0]!
+
+	subs		r5, r5, #8
+	vst1.8		{q8}, [r6]		// store next round's iv
+	bgt		99b
+
+	pop		{r4-r6, pc}
+ENDPROC(aesbs_cbc_decrypt)
+
+	.macro		next_ctr, q
+	vmov.32		\q\()h[1], r10
+	adds		r10, r10, #1
+	vmov.32		\q\()h[0], r9
+	adcs		r9, r9, #0
+	vmov.32		\q\()l[1], r8
+	adcs		r8, r8, #0
+	vmov.32		\q\()l[0], r7
+	adc		r7, r7, #0
+	vrev32.8	\q, \q
+	.endm
+
+	/*
+	 * aesbs_ctr_encrypt(u8 out[], u8 const in[], u8 const rk[],
+	 *		     int rounds, int blocks, u8 ctr[], u8 final[])
+	 */
+ENTRY(aesbs_ctr_encrypt)
+	mov		ip, sp
+	push		{r4-r10, lr}
+
+	ldm		ip, {r5-r7}		// load args 4-6
+	teq		r7, #0
+	addne		r5, r5, #1		// one extra block if final != 0
+
+	vld1.8		{q0}, [r6]		// load counter
+	vrev32.8	q1, q0
+	vmov		r9, r10, d3
+	vmov		r7, r8, d2
+
+	adds		r10, r10, #1
+	adcs		r9, r9, #0
+	adcs		r8, r8, #0
+	adc		r7, r7, #0
+
+99:	vmov		q1, q0
+	vmov		q2, q0
+	vmov		q3, q0
+	vmov		q4, q0
+	vmov		q5, q0
+	vmov		q6, q0
+	vmov		q7, q0
+
+	__adr		ip, 0f
+	sub		lr, r5, #1
+	and		lr, lr, #7
+	cmp		r5, #8
+	sub		ip, ip, lr, lsl #5
+	sub		ip, ip, lr, lsl #2
+	bxlt		ip			// computed goto if blocks < 8
+
+	next_ctr	q1
+	next_ctr	q2
+	next_ctr	q3
+	next_ctr	q4
+	next_ctr	q5
+	next_ctr	q6
+	next_ctr	q7
+
+0:	mov		bskey, r2
+	mov		rounds, r3
+	bl		aesbs_encrypt8
+
+	__adr		ip, 1f
+	and		lr, r5, #7
+	cmp		r5, #8
+	movgt		r4, #0
+	ldrle		r4, [sp, #40]		// load final in the last round
+	sub		ip, ip, lr, lsl #2
+	bxlt		ip			// computed goto if blocks < 8
+
+	vld1.8		{q8}, [r1]!
+	vld1.8		{q9}, [r1]!
+	vld1.8		{q10}, [r1]!
+	vld1.8		{q11}, [r1]!
+	vld1.8		{q12}, [r1]!
+	vld1.8		{q13}, [r1]!
+	vld1.8		{q14}, [r1]!
+	teq		r4, #0			// skip last block if 'final'
+1:	bne		2f
+	vld1.8		{q15}, [r1]!
+
+2:	__adr		ip, 3f
+	cmp		r5, #8
+	sub		ip, ip, lr, lsl #3
+	bxlt		ip			// computed goto if blocks < 8
+
+	veor		q0, q0, q8
+	vst1.8		{q0}, [r0]!
+	veor		q1, q1, q9
+	vst1.8		{q1}, [r0]!
+	veor		q4, q4, q10
+	vst1.8		{q4}, [r0]!
+	veor		q6, q6, q11
+	vst1.8		{q6}, [r0]!
+	veor		q3, q3, q12
+	vst1.8		{q3}, [r0]!
+	veor		q7, q7, q13
+	vst1.8		{q7}, [r0]!
+	veor		q2, q2, q14
+	vst1.8		{q2}, [r0]!
+	teq		r4, #0			// skip last block if 'final'
+	W(bne)		5f
+3:	veor		q5, q5, q15
+	vst1.8		{q5}, [r0]!
+
+4:	next_ctr	q0
+
+	subs		r5, r5, #8
+	bgt		99b
+
+	vst1.8		{q0}, [r6]
+	pop		{r4-r10, pc}
+
+5:	vst1.8		{q5}, [r4]
+	b		4b
+ENDPROC(aesbs_ctr_encrypt)
+
+	.macro		next_tweak, out, in, const, tmp
+	vshr.s64	\tmp, \in, #63
+	vand		\tmp, \tmp, \const
+	vadd.u64	\out, \in, \in
+	vext.8		\tmp, \tmp, \tmp, #8
+	veor		\out, \out, \tmp
+	.endm
+
+	.align		4
+.Lxts_mul_x:
+	.quad		1, 0x87
+
+	/*
+	 * aesbs_xts_encrypt(u8 out[], u8 const in[], u8 const rk[], int rounds,
+	 *		     int blocks, u8 iv[])
+	 * aesbs_xts_decrypt(u8 out[], u8 const in[], u8 const rk[], int rounds,
+	 *		     int blocks, u8 iv[])
+	 */
+__xts_prepare8:
+	vld1.8		{q14}, [r7]		// load iv
+	__ldr		q15, .Lxts_mul_x	// load tweak mask
+	vmov		q12, q14
+
+	__adr		ip, 0f
+	and		r4, r6, #7
+	cmp		r6, #8
+	sub		ip, ip, r4, lsl #5
+	mov		r4, sp
+	bxlt		ip			// computed goto if blocks < 8
+
+	vld1.8		{q0}, [r1]!
+	next_tweak	q12, q14, q15, q13
+	veor		q0, q0, q14
+	vst1.8		{q14}, [r4, :128]!
+
+	vld1.8		{q1}, [r1]!
+	next_tweak	q14, q12, q15, q13
+	veor		q1, q1, q12
+	vst1.8		{q12}, [r4, :128]!
+
+	vld1.8		{q2}, [r1]!
+	next_tweak	q12, q14, q15, q13
+	veor		q2, q2, q14
+	vst1.8		{q14}, [r4, :128]!
+
+	vld1.8		{q3}, [r1]!
+	next_tweak	q14, q12, q15, q13
+	veor		q3, q3, q12
+	vst1.8		{q12}, [r4, :128]!
+
+	vld1.8		{q4}, [r1]!
+	next_tweak	q12, q14, q15, q13
+	veor		q4, q4, q14
+	vst1.8		{q14}, [r4, :128]!
+
+	vld1.8		{q5}, [r1]!
+	next_tweak	q14, q12, q15, q13
+	veor		q5, q5, q12
+	vst1.8		{q12}, [r4, :128]!
+
+	vld1.8		{q6}, [r1]!
+	next_tweak	q12, q14, q15, q13
+	veor		q6, q6, q14
+	vst1.8		{q14}, [r4, :128]!
+
+	vld1.8		{q7}, [r1]!
+	next_tweak	q14, q12, q15, q13
+	veor		q7, q7, q12
+	vst1.8		{q12}, [r4, :128]
+
+0:	vst1.8		{q14}, [r7]		// store next iv
+	bx		lr
+ENDPROC(__xts_prepare8)
+
+	.macro		__xts_crypt, do8, o0, o1, o2, o3, o4, o5, o6, o7
+	push		{r4-r8, lr}
+	mov		r5, sp			// preserve sp
+	ldrd		r6, r7, [sp, #24]	// get blocks and iv args
+	sub		ip, sp, #128		// make room for 8x tweak
+	bic		ip, ip, #0xf		// align sp to 16 bytes
+	mov		sp, ip
+
+99:	bl		__xts_prepare8
+
+	mov		bskey, r2
+	mov		rounds, r3
+	bl		\do8
+
+	__adr		ip, 0f
+	and		lr, r6, #7
+	cmp		r6, #8
+	sub		ip, ip, lr, lsl #2
+	mov		r4, sp
+	bxlt		ip			// computed goto if blocks < 8
+
+	vld1.8		{q8}, [r4, :128]!
+	vld1.8		{q9}, [r4, :128]!
+	vld1.8		{q10}, [r4, :128]!
+	vld1.8		{q11}, [r4, :128]!
+	vld1.8		{q12}, [r4, :128]!
+	vld1.8		{q13}, [r4, :128]!
+	vld1.8		{q14}, [r4, :128]!
+	vld1.8		{q15}, [r4, :128]
+
+0:	__adr		ip, 1f
+	sub		ip, ip, lr, lsl #3
+	bxlt		ip			// computed goto if blocks < 8
+
+	veor		\o0, \o0, q8
+	vst1.8		{\o0}, [r0]!
+	veor		\o1, \o1, q9
+	vst1.8		{\o1}, [r0]!
+	veor		\o2, \o2, q10
+	vst1.8		{\o2}, [r0]!
+	veor		\o3, \o3, q11
+	vst1.8		{\o3}, [r0]!
+	veor		\o4, \o4, q12
+	vst1.8		{\o4}, [r0]!
+	veor		\o5, \o5, q13
+	vst1.8		{\o5}, [r0]!
+	veor		\o6, \o6, q14
+	vst1.8		{\o6}, [r0]!
+	veor		\o7, \o7, q15
+	vst1.8		{\o7}, [r0]!
+
+1:	subs		r6, r6, #8
+	bgt		99b
+
+	mov		sp, r5
+	pop		{r4-r8, pc}
+	.endm
+
+ENTRY(aesbs_xts_encrypt)
+	__xts_crypt	aesbs_encrypt8, q0, q1, q4, q6, q3, q7, q2, q5
+ENDPROC(aesbs_xts_encrypt)
+
+ENTRY(aesbs_xts_decrypt)
+	__xts_crypt	aesbs_decrypt8, q0, q1, q6, q4, q2, q7, q3, q5
+ENDPROC(aesbs_xts_decrypt)
diff --git a/arch/arm/crypto/aes-neonbs-glue.c b/arch/arm/crypto/aes-neonbs-glue.c
new file mode 100644
index 000000000000..2920b96dbd36
--- /dev/null
+++ b/arch/arm/crypto/aes-neonbs-glue.c
@@ -0,0 +1,406 @@
+/*
+ * Bit sliced AES using NEON instructions
+ *
+ * Copyright (C) 2017 Linaro Ltd <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <asm/neon.h>
+#include <crypto/aes.h>
+#include <crypto/cbc.h>
+#include <crypto/internal/simd.h>
+#include <crypto/internal/skcipher.h>
+#include <crypto/xts.h>
+#include <linux/module.h>
+
+MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
+MODULE_LICENSE("GPL v2");
+
+MODULE_ALIAS_CRYPTO("ecb(aes)");
+MODULE_ALIAS_CRYPTO("cbc(aes)");
+MODULE_ALIAS_CRYPTO("ctr(aes)");
+MODULE_ALIAS_CRYPTO("xts(aes)");
+
+asmlinkage void aesbs_convert_key(u8 out[], u32 const rk[], int rounds);
+
+asmlinkage void aesbs_ecb_encrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks);
+asmlinkage void aesbs_ecb_decrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks);
+
+asmlinkage void aesbs_cbc_decrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks, u8 iv[]);
+
+asmlinkage void aesbs_ctr_encrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks, u8 ctr[], u8 final[]);
+
+asmlinkage void aesbs_xts_encrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks, u8 iv[]);
+asmlinkage void aesbs_xts_decrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks, u8 iv[]);
+
+asmlinkage void __aes_arm_encrypt(const u32 rk[], int rounds, const u8 in[],
+				  u8 out[]);
+
+struct aesbs_ctx {
+	int	rounds;
+	u8	rk[13 * (8 * AES_BLOCK_SIZE) + 32] __aligned(AES_BLOCK_SIZE);
+};
+
+struct aesbs_cbc_ctx {
+	struct aesbs_ctx	key;
+	u32			enc[AES_MAX_KEYLENGTH_U32];
+};
+
+struct aesbs_xts_ctx {
+	struct aesbs_ctx	key;
+	u32			twkey[AES_MAX_KEYLENGTH_U32];
+};
+
+static int aesbs_setkey(struct crypto_skcipher *tfm, const u8 *in_key,
+			unsigned int key_len)
+{
+	struct aesbs_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct crypto_aes_ctx rk;
+	int err;
+
+	err = crypto_aes_expand_key(&rk, in_key, key_len);
+	if (err)
+		return err;
+
+	ctx->rounds = 6 + key_len / 4;
+
+	kernel_neon_begin();
+	aesbs_convert_key(ctx->rk, rk.key_enc, ctx->rounds);
+	kernel_neon_end();
+
+	return 0;
+}
+
+static int __ecb_crypt(struct skcipher_request *req,
+		       void (*fn)(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks))
+{
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct aesbs_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_walk walk;
+	int err;
+
+	err = skcipher_walk_virt(&walk, req, true);
+
+	kernel_neon_begin();
+	while (walk.nbytes >= AES_BLOCK_SIZE) {
+		unsigned int blocks = walk.nbytes / AES_BLOCK_SIZE;
+
+		if (walk.nbytes < walk.total)
+			blocks = round_down(blocks,
+					    walk.stride / AES_BLOCK_SIZE);
+
+		fn(walk.dst.virt.addr, walk.src.virt.addr, ctx->rk,
+		   ctx->rounds, blocks);
+		err = skcipher_walk_done(&walk,
+					 walk.nbytes - blocks * AES_BLOCK_SIZE);
+	}
+	kernel_neon_end();
+
+	return err;
+}
+
+static int ecb_encrypt(struct skcipher_request *req)
+{
+	return __ecb_crypt(req, aesbs_ecb_encrypt);
+}
+
+static int ecb_decrypt(struct skcipher_request *req)
+{
+	return __ecb_crypt(req, aesbs_ecb_decrypt);
+}
+
+static int aesbs_cbc_setkey(struct crypto_skcipher *tfm, const u8 *in_key,
+			    unsigned int key_len)
+{
+	struct aesbs_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct crypto_aes_ctx rk;
+	int err;
+
+	err = crypto_aes_expand_key(&rk, in_key, key_len);
+	if (err)
+		return err;
+
+	ctx->key.rounds = 6 + key_len / 4;
+
+	memcpy(ctx->enc, rk.key_enc, sizeof(ctx->enc));
+
+	kernel_neon_begin();
+	aesbs_convert_key(ctx->key.rk, rk.key_enc, ctx->key.rounds);
+	kernel_neon_end();
+
+	return 0;
+}
+
+static void cbc_encrypt_one(struct crypto_skcipher *tfm, const u8 *src, u8 *dst)
+{
+	struct aesbs_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
+
+	__aes_arm_encrypt(ctx->enc, ctx->key.rounds, src, dst);
+}
+
+static int cbc_encrypt(struct skcipher_request *req)
+{
+	return crypto_cbc_encrypt_walk(req, cbc_encrypt_one);
+}
+
+static int cbc_decrypt(struct skcipher_request *req)
+{
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct aesbs_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_walk walk;
+	int err;
+
+	err = skcipher_walk_virt(&walk, req, true);
+
+	kernel_neon_begin();
+	while (walk.nbytes >= AES_BLOCK_SIZE) {
+		unsigned int blocks = walk.nbytes / AES_BLOCK_SIZE;
+
+		if (walk.nbytes < walk.total)
+			blocks = round_down(blocks,
+					    walk.stride / AES_BLOCK_SIZE);
+
+		aesbs_cbc_decrypt(walk.dst.virt.addr, walk.src.virt.addr,
+				  ctx->key.rk, ctx->key.rounds, blocks,
+				  walk.iv);
+		err = skcipher_walk_done(&walk,
+					 walk.nbytes - blocks * AES_BLOCK_SIZE);
+	}
+	kernel_neon_end();
+
+	return err;
+}
+
+static int ctr_encrypt(struct skcipher_request *req)
+{
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct aesbs_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_walk walk;
+	u8 buf[AES_BLOCK_SIZE];
+	int err;
+
+	err = skcipher_walk_virt(&walk, req, true);
+
+	kernel_neon_begin();
+	while (walk.nbytes > 0) {
+		unsigned int blocks = walk.nbytes / AES_BLOCK_SIZE;
+		u8 *final = (walk.total % AES_BLOCK_SIZE) ? buf : NULL;
+
+		if (walk.nbytes < walk.total) {
+			blocks = round_down(blocks,
+					    walk.stride / AES_BLOCK_SIZE);
+			final = NULL;
+		}
+
+		aesbs_ctr_encrypt(walk.dst.virt.addr, walk.src.virt.addr,
+				  ctx->rk, ctx->rounds, blocks, walk.iv, final);
+
+		if (final) {
+			u8 *dst = walk.dst.virt.addr + blocks * AES_BLOCK_SIZE;
+			u8 *src = walk.src.virt.addr + blocks * AES_BLOCK_SIZE;
+
+			if (dst != src)
+				memcpy(dst, src, walk.total % AES_BLOCK_SIZE);
+			crypto_xor(dst, final, walk.total % AES_BLOCK_SIZE);
+
+			err = skcipher_walk_done(&walk, 0);
+			break;
+		}
+		err = skcipher_walk_done(&walk,
+					 walk.nbytes - blocks * AES_BLOCK_SIZE);
+	}
+	kernel_neon_end();
+
+	return err;
+}
+
+static int aesbs_xts_setkey(struct crypto_skcipher *tfm, const u8 *in_key,
+			    unsigned int key_len)
+{
+	struct aesbs_xts_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct crypto_aes_ctx rk;
+	int err;
+
+	err = xts_verify_key(tfm, in_key, key_len);
+	if (err)
+		return err;
+
+	key_len /= 2;
+	err = crypto_aes_expand_key(&rk, in_key + key_len, key_len);
+	if (err)
+		return err;
+
+	memcpy(ctx->twkey, rk.key_enc, sizeof(ctx->twkey));
+
+	return aesbs_setkey(tfm, in_key, key_len);
+}
+
+static int __xts_crypt(struct skcipher_request *req,
+		       void (*fn)(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks, u8 iv[]))
+{
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct aesbs_xts_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_walk walk;
+	int err;
+
+	err = skcipher_walk_virt(&walk, req, true);
+
+	__aes_arm_encrypt(ctx->twkey, ctx->key.rounds, walk.iv, walk.iv);
+
+	kernel_neon_begin();
+	while (walk.nbytes >= AES_BLOCK_SIZE) {
+		unsigned int blocks = walk.nbytes / AES_BLOCK_SIZE;
+
+		if (walk.nbytes < walk.total)
+			blocks = round_down(blocks,
+					    walk.stride / AES_BLOCK_SIZE);
+
+		fn(walk.dst.virt.addr, walk.src.virt.addr, ctx->key.rk,
+		   ctx->key.rounds, blocks, walk.iv);
+		err = skcipher_walk_done(&walk,
+					 walk.nbytes - blocks * AES_BLOCK_SIZE);
+	}
+	kernel_neon_end();
+
+	return err;
+}
+
+static int xts_encrypt(struct skcipher_request *req)
+{
+	return __xts_crypt(req, aesbs_xts_encrypt);
+}
+
+static int xts_decrypt(struct skcipher_request *req)
+{
+	return __xts_crypt(req, aesbs_xts_decrypt);
+}
+
+static struct skcipher_alg aes_algs[] = { {
+	.base.cra_name		= "__ecb(aes)",
+	.base.cra_driver_name	= "__ecb-aes-neonbs",
+	.base.cra_priority	= 250,
+	.base.cra_blocksize	= AES_BLOCK_SIZE,
+	.base.cra_ctxsize	= sizeof(struct aesbs_ctx),
+	.base.cra_module	= THIS_MODULE,
+	.base.cra_flags		= CRYPTO_ALG_INTERNAL,
+
+	.min_keysize		= AES_MIN_KEY_SIZE,
+	.max_keysize		= AES_MAX_KEY_SIZE,
+	.walksize		= 8 * AES_BLOCK_SIZE,
+	.setkey			= aesbs_setkey,
+	.encrypt		= ecb_encrypt,
+	.decrypt		= ecb_decrypt,
+}, {
+	.base.cra_name		= "__cbc(aes)",
+	.base.cra_driver_name	= "__cbc-aes-neonbs",
+	.base.cra_priority	= 250,
+	.base.cra_blocksize	= AES_BLOCK_SIZE,
+	.base.cra_ctxsize	= sizeof(struct aesbs_cbc_ctx),
+	.base.cra_module	= THIS_MODULE,
+	.base.cra_flags		= CRYPTO_ALG_INTERNAL,
+
+	.min_keysize		= AES_MIN_KEY_SIZE,
+	.max_keysize		= AES_MAX_KEY_SIZE,
+	.walksize		= 8 * AES_BLOCK_SIZE,
+	.ivsize			= AES_BLOCK_SIZE,
+	.setkey			= aesbs_cbc_setkey,
+	.encrypt		= cbc_encrypt,
+	.decrypt		= cbc_decrypt,
+}, {
+	.base.cra_name		= "__ctr(aes)",
+	.base.cra_driver_name	= "__ctr-aes-neonbs",
+	.base.cra_priority	= 250,
+	.base.cra_blocksize	= 1,
+	.base.cra_ctxsize	= sizeof(struct aesbs_ctx),
+	.base.cra_module	= THIS_MODULE,
+	.base.cra_flags		= CRYPTO_ALG_INTERNAL,
+
+	.min_keysize		= AES_MIN_KEY_SIZE,
+	.max_keysize		= AES_MAX_KEY_SIZE,
+	.chunksize		= AES_BLOCK_SIZE,
+	.walksize		= 8 * AES_BLOCK_SIZE,
+	.ivsize			= AES_BLOCK_SIZE,
+	.setkey			= aesbs_setkey,
+	.encrypt		= ctr_encrypt,
+	.decrypt		= ctr_encrypt,
+}, {
+	.base.cra_name		= "__xts(aes)",
+	.base.cra_driver_name	= "__xts-aes-neonbs",
+	.base.cra_priority	= 250,
+	.base.cra_blocksize	= AES_BLOCK_SIZE,
+	.base.cra_ctxsize	= sizeof(struct aesbs_xts_ctx),
+	.base.cra_module	= THIS_MODULE,
+	.base.cra_flags		= CRYPTO_ALG_INTERNAL,
+
+	.min_keysize		= 2 * AES_MIN_KEY_SIZE,
+	.max_keysize		= 2 * AES_MAX_KEY_SIZE,
+	.walksize		= 8 * AES_BLOCK_SIZE,
+	.ivsize			= AES_BLOCK_SIZE,
+	.setkey			= aesbs_xts_setkey,
+	.encrypt		= xts_encrypt,
+	.decrypt		= xts_decrypt,
+} };
+
+static struct simd_skcipher_alg *aes_simd_algs[ARRAY_SIZE(aes_algs)];
+
+static void aes_exit(void)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(aes_simd_algs); i++)
+		if (aes_simd_algs[i])
+			simd_skcipher_free(aes_simd_algs[i]);
+
+	crypto_unregister_skciphers(aes_algs, ARRAY_SIZE(aes_algs));
+}
+
+static int __init aes_init(void)
+{
+	struct simd_skcipher_alg *simd;
+	const char *basename;
+	const char *algname;
+	const char *drvname;
+	int err;
+	int i;
+
+	if (!(elf_hwcap & HWCAP_NEON))
+		return -ENODEV;
+
+	err = crypto_register_skciphers(aes_algs, ARRAY_SIZE(aes_algs));
+	if (err)
+		return err;
+
+	for (i = 0; i < ARRAY_SIZE(aes_algs); i++) {
+		if (!(aes_algs[i].base.cra_flags & CRYPTO_ALG_INTERNAL))
+			continue;
+
+		algname = aes_algs[i].base.cra_name + 2;
+		drvname = aes_algs[i].base.cra_driver_name + 2;
+		basename = aes_algs[i].base.cra_driver_name;
+		simd = simd_skcipher_create_compat(algname, drvname, basename);
+		err = PTR_ERR(simd);
+		if (IS_ERR(simd))
+			goto unregister_simds;
+
+		aes_simd_algs[i] = simd;
+	}
+	return 0;
+
+unregister_simds:
+	aes_exit();
+	return err;
+}
+
+module_init(aes_init);
+module_exit(aes_exit);
diff --git a/arch/arm/crypto/aes_glue.c b/arch/arm/crypto/aes_glue.c
deleted file mode 100644
index 0409b8f89782..000000000000
--- a/arch/arm/crypto/aes_glue.c
+++ /dev/null
@@ -1,98 +0,0 @@
-/*
- * Glue Code for the asm optimized version of the AES Cipher Algorithm
- */
-
-#include <linux/module.h>
-#include <linux/crypto.h>
-#include <crypto/aes.h>
-
-#include "aes_glue.h"
-
-EXPORT_SYMBOL(AES_encrypt);
-EXPORT_SYMBOL(AES_decrypt);
-EXPORT_SYMBOL(private_AES_set_encrypt_key);
-EXPORT_SYMBOL(private_AES_set_decrypt_key);
-
-static void aes_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
-{
-	struct AES_CTX *ctx = crypto_tfm_ctx(tfm);
-	AES_encrypt(src, dst, &ctx->enc_key);
-}
-
-static void aes_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
-{
-	struct AES_CTX *ctx = crypto_tfm_ctx(tfm);
-	AES_decrypt(src, dst, &ctx->dec_key);
-}
-
-static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
-		unsigned int key_len)
-{
-	struct AES_CTX *ctx = crypto_tfm_ctx(tfm);
-
-	switch (key_len) {
-	case AES_KEYSIZE_128:
-		key_len = 128;
-		break;
-	case AES_KEYSIZE_192:
-		key_len = 192;
-		break;
-	case AES_KEYSIZE_256:
-		key_len = 256;
-		break;
-	default:
-		tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
-		return -EINVAL;
-	}
-
-	if (private_AES_set_encrypt_key(in_key, key_len, &ctx->enc_key) == -1) {
-		tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
-		return -EINVAL;
-	}
-	/* private_AES_set_decrypt_key expects an encryption key as input */
-	ctx->dec_key = ctx->enc_key;
-	if (private_AES_set_decrypt_key(in_key, key_len, &ctx->dec_key) == -1) {
-		tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN;
-		return -EINVAL;
-	}
-	return 0;
-}
-
-static struct crypto_alg aes_alg = {
-	.cra_name		= "aes",
-	.cra_driver_name	= "aes-asm",
-	.cra_priority		= 200,
-	.cra_flags		= CRYPTO_ALG_TYPE_CIPHER,
-	.cra_blocksize		= AES_BLOCK_SIZE,
-	.cra_ctxsize		= sizeof(struct AES_CTX),
-	.cra_module		= THIS_MODULE,
-	.cra_list		= LIST_HEAD_INIT(aes_alg.cra_list),
-	.cra_u	= {
-		.cipher	= {
-			.cia_min_keysize	= AES_MIN_KEY_SIZE,
-			.cia_max_keysize	= AES_MAX_KEY_SIZE,
-			.cia_setkey		= aes_set_key,
-			.cia_encrypt		= aes_encrypt,
-			.cia_decrypt		= aes_decrypt
-		}
-	}
-};
-
-static int __init aes_init(void)
-{
-	return crypto_register_alg(&aes_alg);
-}
-
-static void __exit aes_fini(void)
-{
-	crypto_unregister_alg(&aes_alg);
-}
-
-module_init(aes_init);
-module_exit(aes_fini);
-
-MODULE_DESCRIPTION("Rijndael (AES) Cipher Algorithm (ASM)");
-MODULE_LICENSE("GPL");
-MODULE_ALIAS_CRYPTO("aes");
-MODULE_ALIAS_CRYPTO("aes-asm");
-MODULE_AUTHOR("David McCullough <ucdevel@gmail.com>");
diff --git a/arch/arm/crypto/aes_glue.h b/arch/arm/crypto/aes_glue.h
deleted file mode 100644
index cca3e51eb606..000000000000
--- a/arch/arm/crypto/aes_glue.h
+++ /dev/null
@@ -1,19 +0,0 @@
-
-#define AES_MAXNR 14
-
-struct AES_KEY {
-	unsigned int rd_key[4 * (AES_MAXNR + 1)];
-	int rounds;
-};
-
-struct AES_CTX {
-	struct AES_KEY enc_key;
-	struct AES_KEY dec_key;
-};
-
-asmlinkage void AES_encrypt(const u8 *in, u8 *out, struct AES_KEY *ctx);
-asmlinkage void AES_decrypt(const u8 *in, u8 *out, struct AES_KEY *ctx);
-asmlinkage int private_AES_set_decrypt_key(const unsigned char *userKey,
-					   const int bits, struct AES_KEY *key);
-asmlinkage int private_AES_set_encrypt_key(const unsigned char *userKey,
-					   const int bits, struct AES_KEY *key);
diff --git a/arch/arm/crypto/aesbs-core.S_shipped b/arch/arm/crypto/aesbs-core.S_shipped
deleted file mode 100644
index 1d1800f71c5b..000000000000
--- a/arch/arm/crypto/aesbs-core.S_shipped
+++ /dev/null
@@ -1,2548 +0,0 @@
-
-@ ====================================================================
-@ 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/.
-@
-@ Specific modes and adaptation for Linux kernel by Ard Biesheuvel
-@ <ard.biesheuvel@linaro.org>. Permission to use under GPL terms is
-@ granted.
-@ ====================================================================
-
-@ Bit-sliced AES for ARM NEON
-@
-@ February 2012.
-@
-@ This implementation is direct adaptation of bsaes-x86_64 module for
-@ ARM NEON. Except that this module is endian-neutral [in sense that
-@ it can be compiled for either endianness] by courtesy of vld1.8's
-@ neutrality. Initial version doesn't implement interface to OpenSSL,
-@ only low-level primitives and unsupported entry points, just enough
-@ to collect performance results, which for Cortex-A8 core are:
-@
-@ encrypt	19.5 cycles per byte processed with 128-bit key
-@ decrypt	22.1 cycles per byte processed with 128-bit key
-@ key conv.	440  cycles per 128-bit key/0.18 of 8x block
-@
-@ Snapdragon S4 encrypts byte in 17.6 cycles and decrypts in 19.7,
-@ which is [much] worse than anticipated (for further details see
-@ http://www.openssl.org/~appro/Snapdragon-S4.html).
-@
-@ Cortex-A15 manages in 14.2/16.1 cycles [when integer-only code
-@ manages in 20.0 cycles].
-@
-@ When comparing to x86_64 results keep in mind that NEON unit is
-@ [mostly] single-issue and thus can't [fully] benefit from
-@ instruction-level parallelism. And when comparing to aes-armv4
-@ results keep in mind key schedule conversion overhead (see
-@ bsaes-x86_64.pl for further details)...
-@
-@						<appro@openssl.org>
-
-@ April-August 2013
-@
-@ Add CBC, CTR and XTS subroutines, adapt for kernel use.
-@
-@					<ard.biesheuvel@linaro.org>
-
-#ifndef __KERNEL__
-# include "arm_arch.h"
-
-# define VFP_ABI_PUSH	vstmdb	sp!,{d8-d15}
-# define VFP_ABI_POP	vldmia	sp!,{d8-d15}
-# define VFP_ABI_FRAME	0x40
-#else
-# define VFP_ABI_PUSH
-# define VFP_ABI_POP
-# define VFP_ABI_FRAME	0
-# define BSAES_ASM_EXTENDED_KEY
-# define XTS_CHAIN_TWEAK
-# define __ARM_ARCH__ __LINUX_ARM_ARCH__
-# define __ARM_MAX_ARCH__ 7
-#endif
-
-#ifdef __thumb__
-# define adrl adr
-#endif
-
-#if __ARM_MAX_ARCH__>=7
-.arch	armv7-a
-.fpu	neon
-
-.text
-.syntax	unified 	@ ARMv7-capable assembler is expected to handle this
-#ifdef __thumb2__
-.thumb
-#else
-.code   32
-#endif
-
-.type	_bsaes_decrypt8,%function
-.align	4
-_bsaes_decrypt8:
-	adr	r6,_bsaes_decrypt8
-	vldmia	r4!, {q9}		@ round 0 key
-	add	r6,r6,#.LM0ISR-_bsaes_decrypt8
-
-	vldmia	r6!, {q8}		@ .LM0ISR
-	veor	q10, q0, q9	@ xor with round0 key
-	veor	q11, q1, q9
-	 vtbl.8	d0, {q10}, d16
-	 vtbl.8	d1, {q10}, d17
-	veor	q12, q2, q9
-	 vtbl.8	d2, {q11}, d16
-	 vtbl.8	d3, {q11}, d17
-	veor	q13, q3, q9
-	 vtbl.8	d4, {q12}, d16
-	 vtbl.8	d5, {q12}, d17
-	veor	q14, q4, q9
-	 vtbl.8	d6, {q13}, d16
-	 vtbl.8	d7, {q13}, d17
-	veor	q15, q5, q9
-	 vtbl.8	d8, {q14}, d16
-	 vtbl.8	d9, {q14}, d17
-	veor	q10, q6, q9
-	 vtbl.8	d10, {q15}, d16
-	 vtbl.8	d11, {q15}, d17
-	veor	q11, q7, q9
-	 vtbl.8	d12, {q10}, d16
-	 vtbl.8	d13, {q10}, d17
-	 vtbl.8	d14, {q11}, d16
-	 vtbl.8	d15, {q11}, d17
-	vmov.i8	q8,#0x55			@ compose .LBS0
-	vmov.i8	q9,#0x33			@ compose .LBS1
-	vshr.u64	q10, q6, #1
-	 vshr.u64	q11, q4, #1
-	veor		q10, q10, q7
-	 veor		q11, q11, q5
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q7, q7, q10
-	vshl.u64	q10, q10, #1
-	 veor		q5, q5, q11
-	 vshl.u64	q11, q11, #1
-	veor		q6, q6, q10
-	 veor		q4, q4, q11
-	vshr.u64	q10, q2, #1
-	 vshr.u64	q11, q0, #1
-	veor		q10, q10, q3
-	 veor		q11, q11, q1
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q3, q3, q10
-	vshl.u64	q10, q10, #1
-	 veor		q1, q1, q11
-	 vshl.u64	q11, q11, #1
-	veor		q2, q2, q10
-	 veor		q0, q0, q11
-	vmov.i8	q8,#0x0f			@ compose .LBS2
-	vshr.u64	q10, q5, #2
-	 vshr.u64	q11, q4, #2
-	veor		q10, q10, q7
-	 veor		q11, q11, q6
-	vand		q10, q10, q9
-	 vand		q11, q11, q9
-	veor		q7, q7, q10
-	vshl.u64	q10, q10, #2
-	 veor		q6, q6, q11
-	 vshl.u64	q11, q11, #2
-	veor		q5, q5, q10
-	 veor		q4, q4, q11
-	vshr.u64	q10, q1, #2
-	 vshr.u64	q11, q0, #2
-	veor		q10, q10, q3
-	 veor		q11, q11, q2
-	vand		q10, q10, q9
-	 vand		q11, q11, q9
-	veor		q3, q3, q10
-	vshl.u64	q10, q10, #2
-	 veor		q2, q2, q11
-	 vshl.u64	q11, q11, #2
-	veor		q1, q1, q10
-	 veor		q0, q0, q11
-	vshr.u64	q10, q3, #4
-	 vshr.u64	q11, q2, #4
-	veor		q10, q10, q7
-	 veor		q11, q11, q6
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q7, q7, q10
-	vshl.u64	q10, q10, #4
-	 veor		q6, q6, q11
-	 vshl.u64	q11, q11, #4
-	veor		q3, q3, q10
-	 veor		q2, q2, q11
-	vshr.u64	q10, q1, #4
-	 vshr.u64	q11, q0, #4
-	veor		q10, q10, q5
-	 veor		q11, q11, q4
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q5, q5, q10
-	vshl.u64	q10, q10, #4
-	 veor		q4, q4, q11
-	 vshl.u64	q11, q11, #4
-	veor		q1, q1, q10
-	 veor		q0, q0, q11
-	sub	r5,r5,#1
-	b	.Ldec_sbox
-.align	4
-.Ldec_loop:
-	vldmia	r4!, {q8-q11}
-	veor	q8, q8, q0
-	veor	q9, q9, q1
-	vtbl.8	d0, {q8}, d24
-	vtbl.8	d1, {q8}, d25
-	vldmia	r4!, {q8}
-	veor	q10, q10, q2
-	vtbl.8	d2, {q9}, d24
-	vtbl.8	d3, {q9}, d25
-	vldmia	r4!, {q9}
-	veor	q11, q11, q3
-	vtbl.8	d4, {q10}, d24
-	vtbl.8	d5, {q10}, d25
-	vldmia	r4!, {q10}
-	vtbl.8	d6, {q11}, d24
-	vtbl.8	d7, {q11}, d25
-	vldmia	r4!, {q11}
-	veor	q8, q8, q4
-	veor	q9, q9, q5
-	vtbl.8	d8, {q8}, d24
-	vtbl.8	d9, {q8}, d25
-	veor	q10, q10, q6
-	vtbl.8	d10, {q9}, d24
-	vtbl.8	d11, {q9}, d25
-	veor	q11, q11, q7
-	vtbl.8	d12, {q10}, d24
-	vtbl.8	d13, {q10}, d25
-	vtbl.8	d14, {q11}, d24
-	vtbl.8	d15, {q11}, d25
-.Ldec_sbox:
-	 veor	q1, q1, q4
-	veor	q3, q3, q4
-
-	veor	q4, q4, q7
-	 veor	q1, q1, q6
-	veor	q2, q2, q7
-	veor	q6, q6, q4
-
-	veor	q0, q0, q1
-	veor	q2, q2, q5
-	 veor	q7, q7, q6
-	veor	q3, q3, q0
-	veor	q5, q5, q0
-	veor	q1, q1, q3
-	veor	q11, q3, q0
-	veor	q10, q7, q4
-	veor	q9, q1, q6
-	veor	q13, q4, q0
-	 vmov	q8, q10
-	veor	q12, q5, q2
-
-	vorr	q10, q10, q9
-	veor	q15, q11, q8
-	vand	q14, q11, q12
-	vorr	q11, q11, q12
-	veor	q12, q12, q9
-	vand	q8, q8, q9
-	veor	q9, q6, q2
-	vand	q15, q15, q12
-	vand	q13, q13, q9
-	veor	q9, q3, q7
-	veor	q12, q1, q5
-	veor	q11, q11, q13
-	veor	q10, q10, q13
-	vand	q13, q9, q12
-	vorr	q9, q9, q12
-	veor	q11, q11, q15
-	veor	q8, q8, q13
-	veor	q10, q10, q14
-	veor	q9, q9, q15
-	veor	q8, q8, q14
-	vand	q12, q4, q6
-	veor	q9, q9, q14
-	vand	q13, q0, q2
-	vand	q14, q7, q1
-	vorr	q15, q3, q5
-	veor	q11, q11, q12
-	veor	q9, q9, q14
-	veor	q8, q8, q15
-	veor	q10, q10, q13
-
-	@ Inv_GF16 	0, 	1, 	2, 	3, s0, s1, s2, s3
-
-	@ new smaller inversion
-
-	vand	q14, q11, q9
-	vmov	q12, q8
-
-	veor	q13, q10, q14
-	veor	q15, q8, q14
-	veor	q14, q8, q14	@ q14=q15
-
-	vbsl	q13, q9, q8
-	vbsl	q15, q11, q10
-	veor	q11, q11, q10
-
-	vbsl	q12, q13, q14
-	vbsl	q8, q14, q13
-
-	vand	q14, q12, q15
-	veor	q9, q9, q8
-
-	veor	q14, q14, q11
-	veor	q12, q5, q2
-	veor	q8, q1, q6
-	veor 	q10, q15, q14
-	vand	q10, q10, q5
-	veor	q5, q5, q1
-	vand	q11, q1, q15
-	vand	q5, q5, q14
-	veor	q1, q11, q10
-	veor	q5, q5, q11
-	veor	q15, q15, q13
-	veor	q14, q14, q9
-	veor	q11, q15, q14
-	 veor 	q10, q13, q9
-	vand	q11, q11, q12
-	 vand	q10, q10, q2
-	veor	q12, q12, q8
-	 veor	q2, q2, q6
-	vand	q8, q8, q15
-	 vand	q6, q6, q13
-	vand	q12, q12, q14
-	 vand	q2, q2, q9
-	veor	q8, q8, q12
-	 veor	q2, q2, q6
-	veor	q12, q12, q11
-	 veor	q6, q6, q10
-	veor	q5, q5, q12
-	veor	q2, q2, q12
-	veor	q1, q1, q8
-	veor	q6, q6, q8
-
-	veor	q12, q3, q0
-	veor	q8, q7, q4
-	veor	q11, q15, q14
-	 veor 	q10, q13, q9
-	vand	q11, q11, q12
-	 vand	q10, q10, q0
-	veor	q12, q12, q8
-	 veor	q0, q0, q4
-	vand	q8, q8, q15
-	 vand	q4, q4, q13
-	vand	q12, q12, q14
-	 vand	q0, q0, q9
-	veor	q8, q8, q12
-	 veor	q0, q0, q4
-	veor	q12, q12, q11
-	 veor	q4, q4, q10
-	veor	q15, q15, q13
-	veor	q14, q14, q9
-	veor 	q10, q15, q14
-	vand	q10, q10, q3
-	veor	q3, q3, q7
-	vand	q11, q7, q15
-	vand	q3, q3, q14
-	veor	q7, q11, q10
-	veor	q3, q3, q11
-	veor	q3, q3, q12
-	veor	q0, q0, q12
-	veor	q7, q7, q8
-	veor	q4, q4, q8
-	veor	q1, q1, q7
-	veor	q6, q6, q5
-
-	veor	q4, q4, q1
-	veor	q2, q2, q7
-	veor	q5, q5, q7
-	veor	q4, q4, q2
-	 veor 	q7, q7, q0
-	veor	q4, q4, q5
-	 veor	q3, q3, q6
-	 veor	q6, q6, q1
-	veor	q3, q3, q4
-
-	veor	q4, q4, q0
-	veor	q7, q7, q3
-	subs	r5,r5,#1
-	bcc	.Ldec_done
-	@ multiplication by 0x05-0x00-0x04-0x00
-	vext.8	q8, q0, q0, #8
-	vext.8	q14, q3, q3, #8
-	vext.8	q15, q5, q5, #8
-	veor	q8, q8, q0
-	vext.8	q9, q1, q1, #8
-	veor	q14, q14, q3
-	vext.8	q10, q6, q6, #8
-	veor	q15, q15, q5
-	vext.8	q11, q4, q4, #8
-	veor	q9, q9, q1
-	vext.8	q12, q2, q2, #8
-	veor	q10, q10, q6
-	vext.8	q13, q7, q7, #8
-	veor	q11, q11, q4
-	veor	q12, q12, q2
-	veor	q13, q13, q7
-
-	 veor	q0, q0, q14
-	 veor	q1, q1, q14
-	 veor	q6, q6, q8
-	 veor	q2, q2, q10
-	 veor	q4, q4, q9
-	 veor	q1, q1, q15
-	 veor	q6, q6, q15
-	 veor	q2, q2, q14
-	 veor	q7, q7, q11
-	 veor	q4, q4, q14
-	 veor	q3, q3, q12
-	 veor	q2, q2, q15
-	 veor	q7, q7, q15
-	 veor	q5, q5, q13
-	vext.8	q8, q0, q0, #12	@ x0 <<< 32
-	vext.8	q9, q1, q1, #12
-	 veor	q0, q0, q8		@ x0 ^ (x0 <<< 32)
-	vext.8	q10, q6, q6, #12
-	 veor	q1, q1, q9
-	vext.8	q11, q4, q4, #12
-	 veor	q6, q6, q10
-	vext.8	q12, q2, q2, #12
-	 veor	q4, q4, q11
-	vext.8	q13, q7, q7, #12
-	 veor	q2, q2, q12
-	vext.8	q14, q3, q3, #12
-	 veor	q7, q7, q13
-	vext.8	q15, q5, q5, #12
-	 veor	q3, q3, q14
-
-	veor	q9, q9, q0
-	 veor	q5, q5, q15
-	 vext.8	q0, q0, q0, #8		@ (x0 ^ (x0 <<< 32)) <<< 64)
-	veor	q10, q10, q1
-	veor	q8, q8, q5
-	veor	q9, q9, q5
-	 vext.8	q1, q1, q1, #8
-	veor	q13, q13, q2
-	 veor	q0, q0, q8
-	veor	q14, q14, q7
-	 veor	q1, q1, q9
-	 vext.8	q8, q2, q2, #8
-	veor	q12, q12, q4
-	 vext.8	q9, q7, q7, #8
-	veor	q15, q15, q3
-	 vext.8	q2, q4, q4, #8
-	veor	q11, q11, q6
-	 vext.8	q7, q5, q5, #8
-	veor	q12, q12, q5
-	 vext.8	q4, q3, q3, #8
-	veor	q11, q11, q5
-	 vext.8	q3, q6, q6, #8
-	veor	q5, q9, q13
-	veor	q11, q11, q2
-	veor	q7, q7, q15
-	veor	q6, q4, q14
-	veor	q4, q8, q12
-	veor	q2, q3, q10
-	vmov	q3, q11
-	 @ vmov	q5, q9
-	vldmia	r6, {q12}		@ .LISR
-	ite	eq				@ Thumb2 thing, sanity check in ARM
-	addeq	r6,r6,#0x10
-	bne	.Ldec_loop
-	vldmia	r6, {q12}		@ .LISRM0
-	b	.Ldec_loop
-.align	4
-.Ldec_done:
-	vmov.i8	q8,#0x55			@ compose .LBS0
-	vmov.i8	q9,#0x33			@ compose .LBS1
-	vshr.u64	q10, q3, #1
-	 vshr.u64	q11, q2, #1
-	veor		q10, q10, q5
-	 veor		q11, q11, q7
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q5, q5, q10
-	vshl.u64	q10, q10, #1
-	 veor		q7, q7, q11
-	 vshl.u64	q11, q11, #1
-	veor		q3, q3, q10
-	 veor		q2, q2, q11
-	vshr.u64	q10, q6, #1
-	 vshr.u64	q11, q0, #1
-	veor		q10, q10, q4
-	 veor		q11, q11, q1
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q4, q4, q10
-	vshl.u64	q10, q10, #1
-	 veor		q1, q1, q11
-	 vshl.u64	q11, q11, #1
-	veor		q6, q6, q10
-	 veor		q0, q0, q11
-	vmov.i8	q8,#0x0f			@ compose .LBS2
-	vshr.u64	q10, q7, #2
-	 vshr.u64	q11, q2, #2
-	veor		q10, q10, q5
-	 veor		q11, q11, q3
-	vand		q10, q10, q9
-	 vand		q11, q11, q9
-	veor		q5, q5, q10
-	vshl.u64	q10, q10, #2
-	 veor		q3, q3, q11
-	 vshl.u64	q11, q11, #2
-	veor		q7, q7, q10
-	 veor		q2, q2, q11
-	vshr.u64	q10, q1, #2
-	 vshr.u64	q11, q0, #2
-	veor		q10, q10, q4
-	 veor		q11, q11, q6
-	vand		q10, q10, q9
-	 vand		q11, q11, q9
-	veor		q4, q4, q10
-	vshl.u64	q10, q10, #2
-	 veor		q6, q6, q11
-	 vshl.u64	q11, q11, #2
-	veor		q1, q1, q10
-	 veor		q0, q0, q11
-	vshr.u64	q10, q4, #4
-	 vshr.u64	q11, q6, #4
-	veor		q10, q10, q5
-	 veor		q11, q11, q3
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q5, q5, q10
-	vshl.u64	q10, q10, #4
-	 veor		q3, q3, q11
-	 vshl.u64	q11, q11, #4
-	veor		q4, q4, q10
-	 veor		q6, q6, q11
-	vshr.u64	q10, q1, #4
-	 vshr.u64	q11, q0, #4
-	veor		q10, q10, q7
-	 veor		q11, q11, q2
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q7, q7, q10
-	vshl.u64	q10, q10, #4
-	 veor		q2, q2, q11
-	 vshl.u64	q11, q11, #4
-	veor		q1, q1, q10
-	 veor		q0, q0, q11
-	vldmia	r4, {q8}			@ last round key
-	veor	q6, q6, q8
-	veor	q4, q4, q8
-	veor	q2, q2, q8
-	veor	q7, q7, q8
-	veor	q3, q3, q8
-	veor	q5, q5, q8
-	veor	q0, q0, q8
-	veor	q1, q1, q8
-	bx	lr
-.size	_bsaes_decrypt8,.-_bsaes_decrypt8
-
-.type	_bsaes_const,%object
-.align	6
-_bsaes_const:
-.LM0ISR:	@ InvShiftRows constants
-	.quad	0x0a0e0206070b0f03, 0x0004080c0d010509
-.LISR:
-	.quad	0x0504070602010003, 0x0f0e0d0c080b0a09
-.LISRM0:
-	.quad	0x01040b0e0205080f, 0x0306090c00070a0d
-.LM0SR:		@ ShiftRows constants
-	.quad	0x0a0e02060f03070b, 0x0004080c05090d01
-.LSR:
-	.quad	0x0504070600030201, 0x0f0e0d0c0a09080b
-.LSRM0:
-	.quad	0x0304090e00050a0f, 0x01060b0c0207080d
-.LM0:
-	.quad	0x02060a0e03070b0f, 0x0004080c0105090d
-.LREVM0SR:
-	.quad	0x090d01050c000408, 0x03070b0f060a0e02
-.asciz	"Bit-sliced AES for NEON, CRYPTOGAMS by <appro@openssl.org>"
-.align	6
-.size	_bsaes_const,.-_bsaes_const
-
-.type	_bsaes_encrypt8,%function
-.align	4
-_bsaes_encrypt8:
-	adr	r6,_bsaes_encrypt8
-	vldmia	r4!, {q9}		@ round 0 key
-	sub	r6,r6,#_bsaes_encrypt8-.LM0SR
-
-	vldmia	r6!, {q8}		@ .LM0SR
-_bsaes_encrypt8_alt:
-	veor	q10, q0, q9	@ xor with round0 key
-	veor	q11, q1, q9
-	 vtbl.8	d0, {q10}, d16
-	 vtbl.8	d1, {q10}, d17
-	veor	q12, q2, q9
-	 vtbl.8	d2, {q11}, d16
-	 vtbl.8	d3, {q11}, d17
-	veor	q13, q3, q9
-	 vtbl.8	d4, {q12}, d16
-	 vtbl.8	d5, {q12}, d17
-	veor	q14, q4, q9
-	 vtbl.8	d6, {q13}, d16
-	 vtbl.8	d7, {q13}, d17
-	veor	q15, q5, q9
-	 vtbl.8	d8, {q14}, d16
-	 vtbl.8	d9, {q14}, d17
-	veor	q10, q6, q9
-	 vtbl.8	d10, {q15}, d16
-	 vtbl.8	d11, {q15}, d17
-	veor	q11, q7, q9
-	 vtbl.8	d12, {q10}, d16
-	 vtbl.8	d13, {q10}, d17
-	 vtbl.8	d14, {q11}, d16
-	 vtbl.8	d15, {q11}, d17
-_bsaes_encrypt8_bitslice:
-	vmov.i8	q8,#0x55			@ compose .LBS0
-	vmov.i8	q9,#0x33			@ compose .LBS1
-	vshr.u64	q10, q6, #1
-	 vshr.u64	q11, q4, #1
-	veor		q10, q10, q7
-	 veor		q11, q11, q5
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q7, q7, q10
-	vshl.u64	q10, q10, #1
-	 veor		q5, q5, q11
-	 vshl.u64	q11, q11, #1
-	veor		q6, q6, q10
-	 veor		q4, q4, q11
-	vshr.u64	q10, q2, #1
-	 vshr.u64	q11, q0, #1
-	veor		q10, q10, q3
-	 veor		q11, q11, q1
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q3, q3, q10
-	vshl.u64	q10, q10, #1
-	 veor		q1, q1, q11
-	 vshl.u64	q11, q11, #1
-	veor		q2, q2, q10
-	 veor		q0, q0, q11
-	vmov.i8	q8,#0x0f			@ compose .LBS2
-	vshr.u64	q10, q5, #2
-	 vshr.u64	q11, q4, #2
-	veor		q10, q10, q7
-	 veor		q11, q11, q6
-	vand		q10, q10, q9
-	 vand		q11, q11, q9
-	veor		q7, q7, q10
-	vshl.u64	q10, q10, #2
-	 veor		q6, q6, q11
-	 vshl.u64	q11, q11, #2
-	veor		q5, q5, q10
-	 veor		q4, q4, q11
-	vshr.u64	q10, q1, #2
-	 vshr.u64	q11, q0, #2
-	veor		q10, q10, q3
-	 veor		q11, q11, q2
-	vand		q10, q10, q9
-	 vand		q11, q11, q9
-	veor		q3, q3, q10
-	vshl.u64	q10, q10, #2
-	 veor		q2, q2, q11
-	 vshl.u64	q11, q11, #2
-	veor		q1, q1, q10
-	 veor		q0, q0, q11
-	vshr.u64	q10, q3, #4
-	 vshr.u64	q11, q2, #4
-	veor		q10, q10, q7
-	 veor		q11, q11, q6
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q7, q7, q10
-	vshl.u64	q10, q10, #4
-	 veor		q6, q6, q11
-	 vshl.u64	q11, q11, #4
-	veor		q3, q3, q10
-	 veor		q2, q2, q11
-	vshr.u64	q10, q1, #4
-	 vshr.u64	q11, q0, #4
-	veor		q10, q10, q5
-	 veor		q11, q11, q4
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q5, q5, q10
-	vshl.u64	q10, q10, #4
-	 veor		q4, q4, q11
-	 vshl.u64	q11, q11, #4
-	veor		q1, q1, q10
-	 veor		q0, q0, q11
-	sub	r5,r5,#1
-	b	.Lenc_sbox
-.align	4
-.Lenc_loop:
-	vldmia	r4!, {q8-q11}
-	veor	q8, q8, q0
-	veor	q9, q9, q1
-	vtbl.8	d0, {q8}, d24
-	vtbl.8	d1, {q8}, d25
-	vldmia	r4!, {q8}
-	veor	q10, q10, q2
-	vtbl.8	d2, {q9}, d24
-	vtbl.8	d3, {q9}, d25
-	vldmia	r4!, {q9}
-	veor	q11, q11, q3
-	vtbl.8	d4, {q10}, d24
-	vtbl.8	d5, {q10}, d25
-	vldmia	r4!, {q10}
-	vtbl.8	d6, {q11}, d24
-	vtbl.8	d7, {q11}, d25
-	vldmia	r4!, {q11}
-	veor	q8, q8, q4
-	veor	q9, q9, q5
-	vtbl.8	d8, {q8}, d24
-	vtbl.8	d9, {q8}, d25
-	veor	q10, q10, q6
-	vtbl.8	d10, {q9}, d24
-	vtbl.8	d11, {q9}, d25
-	veor	q11, q11, q7
-	vtbl.8	d12, {q10}, d24
-	vtbl.8	d13, {q10}, d25
-	vtbl.8	d14, {q11}, d24
-	vtbl.8	d15, {q11}, d25
-.Lenc_sbox:
-	veor	q2, q2, q1
-	veor	q5, q5, q6
-	veor	q3, q3, q0
-	veor	q6, q6, q2
-	veor	q5, q5, q0
-
-	veor	q6, q6, q3
-	veor	q3, q3, q7
-	veor	q7, q7, q5
-	veor	q3, q3, q4
-	veor	q4, q4, q5
-
-	veor	q2, q2, q7
-	veor	q3, q3, q1
-	veor	q1, q1, q5
-	veor	q11, q7, q4
-	veor	q10, q1, q2
-	veor	q9, q5, q3
-	veor	q13, q2, q4
-	 vmov	q8, q10
-	veor	q12, q6, q0
-
-	vorr	q10, q10, q9
-	veor	q15, q11, q8
-	vand	q14, q11, q12
-	vorr	q11, q11, q12
-	veor	q12, q12, q9
-	vand	q8, q8, q9
-	veor	q9, q3, q0
-	vand	q15, q15, q12
-	vand	q13, q13, q9
-	veor	q9, q7, q1
-	veor	q12, q5, q6
-	veor	q11, q11, q13
-	veor	q10, q10, q13
-	vand	q13, q9, q12
-	vorr	q9, q9, q12
-	veor	q11, q11, q15
-	veor	q8, q8, q13
-	veor	q10, q10, q14
-	veor	q9, q9, q15
-	veor	q8, q8, q14
-	vand	q12, q2, q3
-	veor	q9, q9, q14
-	vand	q13, q4, q0
-	vand	q14, q1, q5
-	vorr	q15, q7, q6
-	veor	q11, q11, q12
-	veor	q9, q9, q14
-	veor	q8, q8, q15
-	veor	q10, q10, q13
-
-	@ Inv_GF16 	0, 	1, 	2, 	3, s0, s1, s2, s3
-
-	@ new smaller inversion
-
-	vand	q14, q11, q9
-	vmov	q12, q8
-
-	veor	q13, q10, q14
-	veor	q15, q8, q14
-	veor	q14, q8, q14	@ q14=q15
-
-	vbsl	q13, q9, q8
-	vbsl	q15, q11, q10
-	veor	q11, q11, q10
-
-	vbsl	q12, q13, q14
-	vbsl	q8, q14, q13
-
-	vand	q14, q12, q15
-	veor	q9, q9, q8
-
-	veor	q14, q14, q11
-	veor	q12, q6, q0
-	veor	q8, q5, q3
-	veor 	q10, q15, q14
-	vand	q10, q10, q6
-	veor	q6, q6, q5
-	vand	q11, q5, q15
-	vand	q6, q6, q14
-	veor	q5, q11, q10
-	veor	q6, q6, q11
-	veor	q15, q15, q13
-	veor	q14, q14, q9
-	veor	q11, q15, q14
-	 veor 	q10, q13, q9
-	vand	q11, q11, q12
-	 vand	q10, q10, q0
-	veor	q12, q12, q8
-	 veor	q0, q0, q3
-	vand	q8, q8, q15
-	 vand	q3, q3, q13
-	vand	q12, q12, q14
-	 vand	q0, q0, q9
-	veor	q8, q8, q12
-	 veor	q0, q0, q3
-	veor	q12, q12, q11
-	 veor	q3, q3, q10
-	veor	q6, q6, q12
-	veor	q0, q0, q12
-	veor	q5, q5, q8
-	veor	q3, q3, q8
-
-	veor	q12, q7, q4
-	veor	q8, q1, q2
-	veor	q11, q15, q14
-	 veor 	q10, q13, q9
-	vand	q11, q11, q12
-	 vand	q10, q10, q4
-	veor	q12, q12, q8
-	 veor	q4, q4, q2
-	vand	q8, q8, q15
-	 vand	q2, q2, q13
-	vand	q12, q12, q14
-	 vand	q4, q4, q9
-	veor	q8, q8, q12
-	 veor	q4, q4, q2
-	veor	q12, q12, q11
-	 veor	q2, q2, q10
-	veor	q15, q15, q13
-	veor	q14, q14, q9
-	veor 	q10, q15, q14
-	vand	q10, q10, q7
-	veor	q7, q7, q1
-	vand	q11, q1, q15
-	vand	q7, q7, q14
-	veor	q1, q11, q10
-	veor	q7, q7, q11
-	veor	q7, q7, q12
-	veor	q4, q4, q12
-	veor	q1, q1, q8
-	veor	q2, q2, q8
-	veor	q7, q7, q0
-	veor	q1, q1, q6
-	veor	q6, q6, q0
-	veor	q4, q4, q7
-	veor	q0, q0, q1
-
-	veor	q1, q1, q5
-	veor	q5, q5, q2
-	veor	q2, q2, q3
-	veor	q3, q3, q5
-	veor	q4, q4, q5
-
-	veor	q6, q6, q3
-	subs	r5,r5,#1
-	bcc	.Lenc_done
-	vext.8	q8, q0, q0, #12	@ x0 <<< 32
-	vext.8	q9, q1, q1, #12
-	 veor	q0, q0, q8		@ x0 ^ (x0 <<< 32)
-	vext.8	q10, q4, q4, #12
-	 veor	q1, q1, q9
-	vext.8	q11, q6, q6, #12
-	 veor	q4, q4, q10
-	vext.8	q12, q3, q3, #12
-	 veor	q6, q6, q11
-	vext.8	q13, q7, q7, #12
-	 veor	q3, q3, q12
-	vext.8	q14, q2, q2, #12
-	 veor	q7, q7, q13
-	vext.8	q15, q5, q5, #12
-	 veor	q2, q2, q14
-
-	veor	q9, q9, q0
-	 veor	q5, q5, q15
-	 vext.8	q0, q0, q0, #8		@ (x0 ^ (x0 <<< 32)) <<< 64)
-	veor	q10, q10, q1
-	veor	q8, q8, q5
-	veor	q9, q9, q5
-	 vext.8	q1, q1, q1, #8
-	veor	q13, q13, q3
-	 veor	q0, q0, q8
-	veor	q14, q14, q7
-	 veor	q1, q1, q9
-	 vext.8	q8, q3, q3, #8
-	veor	q12, q12, q6
-	 vext.8	q9, q7, q7, #8
-	veor	q15, q15, q2
-	 vext.8	q3, q6, q6, #8
-	veor	q11, q11, q4
-	 vext.8	q7, q5, q5, #8
-	veor	q12, q12, q5
-	 vext.8	q6, q2, q2, #8
-	veor	q11, q11, q5
-	 vext.8	q2, q4, q4, #8
-	veor	q5, q9, q13
-	veor	q4, q8, q12
-	veor	q3, q3, q11
-	veor	q7, q7, q15
-	veor	q6, q6, q14
-	 @ vmov	q4, q8
-	veor	q2, q2, q10
-	 @ vmov	q5, q9
-	vldmia	r6, {q12}		@ .LSR
-	ite	eq				@ Thumb2 thing, samity check in ARM
-	addeq	r6,r6,#0x10
-	bne	.Lenc_loop
-	vldmia	r6, {q12}		@ .LSRM0
-	b	.Lenc_loop
-.align	4
-.Lenc_done:
-	vmov.i8	q8,#0x55			@ compose .LBS0
-	vmov.i8	q9,#0x33			@ compose .LBS1
-	vshr.u64	q10, q2, #1
-	 vshr.u64	q11, q3, #1
-	veor		q10, q10, q5
-	 veor		q11, q11, q7
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q5, q5, q10
-	vshl.u64	q10, q10, #1
-	 veor		q7, q7, q11
-	 vshl.u64	q11, q11, #1
-	veor		q2, q2, q10
-	 veor		q3, q3, q11
-	vshr.u64	q10, q4, #1
-	 vshr.u64	q11, q0, #1
-	veor		q10, q10, q6
-	 veor		q11, q11, q1
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q6, q6, q10
-	vshl.u64	q10, q10, #1
-	 veor		q1, q1, q11
-	 vshl.u64	q11, q11, #1
-	veor		q4, q4, q10
-	 veor		q0, q0, q11
-	vmov.i8	q8,#0x0f			@ compose .LBS2
-	vshr.u64	q10, q7, #2
-	 vshr.u64	q11, q3, #2
-	veor		q10, q10, q5
-	 veor		q11, q11, q2
-	vand		q10, q10, q9
-	 vand		q11, q11, q9
-	veor		q5, q5, q10
-	vshl.u64	q10, q10, #2
-	 veor		q2, q2, q11
-	 vshl.u64	q11, q11, #2
-	veor		q7, q7, q10
-	 veor		q3, q3, q11
-	vshr.u64	q10, q1, #2
-	 vshr.u64	q11, q0, #2
-	veor		q10, q10, q6
-	 veor		q11, q11, q4
-	vand		q10, q10, q9
-	 vand		q11, q11, q9
-	veor		q6, q6, q10
-	vshl.u64	q10, q10, #2
-	 veor		q4, q4, q11
-	 vshl.u64	q11, q11, #2
-	veor		q1, q1, q10
-	 veor		q0, q0, q11
-	vshr.u64	q10, q6, #4
-	 vshr.u64	q11, q4, #4
-	veor		q10, q10, q5
-	 veor		q11, q11, q2
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q5, q5, q10
-	vshl.u64	q10, q10, #4
-	 veor		q2, q2, q11
-	 vshl.u64	q11, q11, #4
-	veor		q6, q6, q10
-	 veor		q4, q4, q11
-	vshr.u64	q10, q1, #4
-	 vshr.u64	q11, q0, #4
-	veor		q10, q10, q7
-	 veor		q11, q11, q3
-	vand		q10, q10, q8
-	 vand		q11, q11, q8
-	veor		q7, q7, q10
-	vshl.u64	q10, q10, #4
-	 veor		q3, q3, q11
-	 vshl.u64	q11, q11, #4
-	veor		q1, q1, q10
-	 veor		q0, q0, q11
-	vldmia	r4, {q8}			@ last round key
-	veor	q4, q4, q8
-	veor	q6, q6, q8
-	veor	q3, q3, q8
-	veor	q7, q7, q8
-	veor	q2, q2, q8
-	veor	q5, q5, q8
-	veor	q0, q0, q8
-	veor	q1, q1, q8
-	bx	lr
-.size	_bsaes_encrypt8,.-_bsaes_encrypt8
-.type	_bsaes_key_convert,%function
-.align	4
-_bsaes_key_convert:
-	adr	r6,_bsaes_key_convert
-	vld1.8	{q7},  [r4]!		@ load round 0 key
-	sub	r6,r6,#_bsaes_key_convert-.LM0
-	vld1.8	{q15}, [r4]!		@ load round 1 key
-
-	vmov.i8	q8,  #0x01			@ bit masks
-	vmov.i8	q9,  #0x02
-	vmov.i8	q10, #0x04
-	vmov.i8	q11, #0x08
-	vmov.i8	q12, #0x10
-	vmov.i8	q13, #0x20
-	vldmia	r6, {q14}		@ .LM0
-
-#ifdef __ARMEL__
-	vrev32.8	q7,  q7
-	vrev32.8	q15, q15
-#endif
-	sub	r5,r5,#1
-	vstmia	r12!, {q7}		@ save round 0 key
-	b	.Lkey_loop
-
-.align	4
-.Lkey_loop:
-	vtbl.8	d14,{q15},d28
-	vtbl.8	d15,{q15},d29
-	vmov.i8	q6,  #0x40
-	vmov.i8	q15, #0x80
-
-	vtst.8	q0, q7, q8
-	vtst.8	q1, q7, q9
-	vtst.8	q2, q7, q10
-	vtst.8	q3, q7, q11
-	vtst.8	q4, q7, q12
-	vtst.8	q5, q7, q13
-	vtst.8	q6, q7, q6
-	vtst.8	q7, q7, q15
-	vld1.8	{q15}, [r4]!		@ load next round key
-	vmvn	q0, q0		@ "pnot"
-	vmvn	q1, q1
-	vmvn	q5, q5
-	vmvn	q6, q6
-#ifdef __ARMEL__
-	vrev32.8	q15, q15
-#endif
-	subs	r5,r5,#1
-	vstmia	r12!,{q0-q7}		@ write bit-sliced round key
-	bne	.Lkey_loop
-
-	vmov.i8	q7,#0x63			@ compose .L63
-	@ don't save last round key
-	bx	lr
-.size	_bsaes_key_convert,.-_bsaes_key_convert
-.extern AES_cbc_encrypt
-.extern AES_decrypt
-
-.global	bsaes_cbc_encrypt
-.type	bsaes_cbc_encrypt,%function
-.align	5
-bsaes_cbc_encrypt:
-#ifndef	__KERNEL__
-	cmp	r2, #128
-#ifndef	__thumb__
-	blo	AES_cbc_encrypt
-#else
-	bhs	1f
-	b	AES_cbc_encrypt
-1:
-#endif
-#endif
-
-	@ it is up to the caller to make sure we are called with enc == 0
-
-	mov	ip, sp
-	stmdb	sp!, {r4-r10, lr}
-	VFP_ABI_PUSH
-	ldr	r8, [ip]			@ IV is 1st arg on the stack
-	mov	r2, r2, lsr#4		@ len in 16 byte blocks
-	sub	sp, #0x10			@ scratch space to carry over the IV
-	mov	r9, sp				@ save sp
-
-	ldr	r10, [r3, #240]		@ get # of rounds
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	@ allocate the key schedule on the stack
-	sub	r12, sp, r10, lsl#7		@ 128 bytes per inner round key
-	add	r12, #96			@ sifze of bit-slices key schedule
-
-	@ populate the key schedule
-	mov	r4, r3			@ pass key
-	mov	r5, r10			@ pass # of rounds
-	mov	sp, r12				@ sp is sp
-	bl	_bsaes_key_convert
-	vldmia	sp, {q6}
-	vstmia	r12,  {q15}		@ save last round key
-	veor	q7, q7, q6	@ fix up round 0 key
-	vstmia	sp, {q7}
-#else
-	ldr	r12, [r3, #244]
-	eors	r12, #1
-	beq	0f
-
-	@ populate the key schedule
-	str	r12, [r3, #244]
-	mov	r4, r3			@ pass key
-	mov	r5, r10			@ pass # of rounds
-	add	r12, r3, #248			@ pass key schedule
-	bl	_bsaes_key_convert
-	add	r4, r3, #248
-	vldmia	r4, {q6}
-	vstmia	r12, {q15}			@ save last round key
-	veor	q7, q7, q6	@ fix up round 0 key
-	vstmia	r4, {q7}
-
-.align	2
-0:
-#endif
-
-	vld1.8	{q15}, [r8]		@ load IV
-	b	.Lcbc_dec_loop
-
-.align	4
-.Lcbc_dec_loop:
-	subs	r2, r2, #0x8
-	bmi	.Lcbc_dec_loop_finish
-
-	vld1.8	{q0-q1}, [r0]!	@ load input
-	vld1.8	{q2-q3}, [r0]!
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	mov	r4, sp			@ pass the key
-#else
-	add	r4, r3, #248
-#endif
-	vld1.8	{q4-q5}, [r0]!
-	mov	r5, r10
-	vld1.8	{q6-q7}, [r0]
-	sub	r0, r0, #0x60
-	vstmia	r9, {q15}			@ put aside IV
-
-	bl	_bsaes_decrypt8
-
-	vldmia	r9, {q14}			@ reload IV
-	vld1.8	{q8-q9}, [r0]!	@ reload input
-	veor	q0, q0, q14	@ ^= IV
-	vld1.8	{q10-q11}, [r0]!
-	veor	q1, q1, q8
-	veor	q6, q6, q9
-	vld1.8	{q12-q13}, [r0]!
-	veor	q4, q4, q10
-	veor	q2, q2, q11
-	vld1.8	{q14-q15}, [r0]!
-	veor	q7, q7, q12
-	vst1.8	{q0-q1}, [r1]!	@ write output
-	veor	q3, q3, q13
-	vst1.8	{q6}, [r1]!
-	veor	q5, q5, q14
-	vst1.8	{q4}, [r1]!
-	vst1.8	{q2}, [r1]!
-	vst1.8	{q7}, [r1]!
-	vst1.8	{q3}, [r1]!
-	vst1.8	{q5}, [r1]!
-
-	b	.Lcbc_dec_loop
-
-.Lcbc_dec_loop_finish:
-	adds	r2, r2, #8
-	beq	.Lcbc_dec_done
-
-	vld1.8	{q0}, [r0]!		@ load input
-	cmp	r2, #2
-	blo	.Lcbc_dec_one
-	vld1.8	{q1}, [r0]!
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	mov	r4, sp			@ pass the key
-#else
-	add	r4, r3, #248
-#endif
-	mov	r5, r10
-	vstmia	r9, {q15}			@ put aside IV
-	beq	.Lcbc_dec_two
-	vld1.8	{q2}, [r0]!
-	cmp	r2, #4
-	blo	.Lcbc_dec_three
-	vld1.8	{q3}, [r0]!
-	beq	.Lcbc_dec_four
-	vld1.8	{q4}, [r0]!
-	cmp	r2, #6
-	blo	.Lcbc_dec_five
-	vld1.8	{q5}, [r0]!
-	beq	.Lcbc_dec_six
-	vld1.8	{q6}, [r0]!
-	sub	r0, r0, #0x70
-
-	bl	_bsaes_decrypt8
-
-	vldmia	r9, {q14}			@ reload IV
-	vld1.8	{q8-q9}, [r0]!	@ reload input
-	veor	q0, q0, q14	@ ^= IV
-	vld1.8	{q10-q11}, [r0]!
-	veor	q1, q1, q8
-	veor	q6, q6, q9
-	vld1.8	{q12-q13}, [r0]!
-	veor	q4, q4, q10
-	veor	q2, q2, q11
-	vld1.8	{q15}, [r0]!
-	veor	q7, q7, q12
-	vst1.8	{q0-q1}, [r1]!	@ write output
-	veor	q3, q3, q13
-	vst1.8	{q6}, [r1]!
-	vst1.8	{q4}, [r1]!
-	vst1.8	{q2}, [r1]!
-	vst1.8	{q7}, [r1]!
-	vst1.8	{q3}, [r1]!
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_six:
-	sub	r0, r0, #0x60
-	bl	_bsaes_decrypt8
-	vldmia	r9,{q14}			@ reload IV
-	vld1.8	{q8-q9}, [r0]!	@ reload input
-	veor	q0, q0, q14	@ ^= IV
-	vld1.8	{q10-q11}, [r0]!
-	veor	q1, q1, q8
-	veor	q6, q6, q9
-	vld1.8	{q12}, [r0]!
-	veor	q4, q4, q10
-	veor	q2, q2, q11
-	vld1.8	{q15}, [r0]!
-	veor	q7, q7, q12
-	vst1.8	{q0-q1}, [r1]!	@ write output
-	vst1.8	{q6}, [r1]!
-	vst1.8	{q4}, [r1]!
-	vst1.8	{q2}, [r1]!
-	vst1.8	{q7}, [r1]!
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_five:
-	sub	r0, r0, #0x50
-	bl	_bsaes_decrypt8
-	vldmia	r9, {q14}			@ reload IV
-	vld1.8	{q8-q9}, [r0]!	@ reload input
-	veor	q0, q0, q14	@ ^= IV
-	vld1.8	{q10-q11}, [r0]!
-	veor	q1, q1, q8
-	veor	q6, q6, q9
-	vld1.8	{q15}, [r0]!
-	veor	q4, q4, q10
-	vst1.8	{q0-q1}, [r1]!	@ write output
-	veor	q2, q2, q11
-	vst1.8	{q6}, [r1]!
-	vst1.8	{q4}, [r1]!
-	vst1.8	{q2}, [r1]!
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_four:
-	sub	r0, r0, #0x40
-	bl	_bsaes_decrypt8
-	vldmia	r9, {q14}			@ reload IV
-	vld1.8	{q8-q9}, [r0]!	@ reload input
-	veor	q0, q0, q14	@ ^= IV
-	vld1.8	{q10}, [r0]!
-	veor	q1, q1, q8
-	veor	q6, q6, q9
-	vld1.8	{q15}, [r0]!
-	veor	q4, q4, q10
-	vst1.8	{q0-q1}, [r1]!	@ write output
-	vst1.8	{q6}, [r1]!
-	vst1.8	{q4}, [r1]!
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_three:
-	sub	r0, r0, #0x30
-	bl	_bsaes_decrypt8
-	vldmia	r9, {q14}			@ reload IV
-	vld1.8	{q8-q9}, [r0]!	@ reload input
-	veor	q0, q0, q14	@ ^= IV
-	vld1.8	{q15}, [r0]!
-	veor	q1, q1, q8
-	veor	q6, q6, q9
-	vst1.8	{q0-q1}, [r1]!	@ write output
-	vst1.8	{q6}, [r1]!
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_two:
-	sub	r0, r0, #0x20
-	bl	_bsaes_decrypt8
-	vldmia	r9, {q14}			@ reload IV
-	vld1.8	{q8}, [r0]!		@ reload input
-	veor	q0, q0, q14	@ ^= IV
-	vld1.8	{q15}, [r0]!		@ reload input
-	veor	q1, q1, q8
-	vst1.8	{q0-q1}, [r1]!	@ write output
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_one:
-	sub	r0, r0, #0x10
-	mov	r10, r1			@ save original out pointer
-	mov	r1, r9			@ use the iv scratch space as out buffer
-	mov	r2, r3
-	vmov	q4,q15		@ just in case ensure that IV
-	vmov	q5,q0			@ and input are preserved
-	bl	AES_decrypt
-	vld1.8	{q0}, [r9,:64]		@ load result
-	veor	q0, q0, q4	@ ^= IV
-	vmov	q15, q5		@ q5 holds input
-	vst1.8	{q0}, [r10]		@ write output
-
-.Lcbc_dec_done:
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	vmov.i32	q0, #0
-	vmov.i32	q1, #0
-.Lcbc_dec_bzero:				@ wipe key schedule [if any]
-	vstmia		sp!, {q0-q1}
-	cmp		sp, r9
-	bne		.Lcbc_dec_bzero
-#endif
-
-	mov	sp, r9
-	add	sp, #0x10			@ add sp,r9,#0x10 is no good for thumb
-	vst1.8	{q15}, [r8]		@ return IV
-	VFP_ABI_POP
-	ldmia	sp!, {r4-r10, pc}
-.size	bsaes_cbc_encrypt,.-bsaes_cbc_encrypt
-.extern	AES_encrypt
-.global	bsaes_ctr32_encrypt_blocks
-.type	bsaes_ctr32_encrypt_blocks,%function
-.align	5
-bsaes_ctr32_encrypt_blocks:
-	cmp	r2, #8			@ use plain AES for
-	blo	.Lctr_enc_short			@ small sizes
-
-	mov	ip, sp
-	stmdb	sp!, {r4-r10, lr}
-	VFP_ABI_PUSH
-	ldr	r8, [ip]			@ ctr is 1st arg on the stack
-	sub	sp, sp, #0x10			@ scratch space to carry over the ctr
-	mov	r9, sp				@ save sp
-
-	ldr	r10, [r3, #240]		@ get # of rounds
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	@ allocate the key schedule on the stack
-	sub	r12, sp, r10, lsl#7		@ 128 bytes per inner round key
-	add	r12, #96			@ size of bit-sliced key schedule
-
-	@ populate the key schedule
-	mov	r4, r3			@ pass key
-	mov	r5, r10			@ pass # of rounds
-	mov	sp, r12				@ sp is sp
-	bl	_bsaes_key_convert
-	veor	q7,q7,q15	@ fix up last round key
-	vstmia	r12, {q7}			@ save last round key
-
-	vld1.8	{q0}, [r8]		@ load counter
-	add	r8, r6, #.LREVM0SR-.LM0	@ borrow r8
-	vldmia	sp, {q4}		@ load round0 key
-#else
-	ldr	r12, [r3, #244]
-	eors	r12, #1
-	beq	0f
-
-	@ populate the key schedule
-	str	r12, [r3, #244]
-	mov	r4, r3			@ pass key
-	mov	r5, r10			@ pass # of rounds
-	add	r12, r3, #248			@ pass key schedule
-	bl	_bsaes_key_convert
-	veor	q7,q7,q15	@ fix up last round key
-	vstmia	r12, {q7}			@ save last round key
-
-.align	2
-0:	add	r12, r3, #248
-	vld1.8	{q0}, [r8]		@ load counter
-	adrl	r8, .LREVM0SR			@ borrow r8
-	vldmia	r12, {q4}			@ load round0 key
-	sub	sp, #0x10			@ place for adjusted round0 key
-#endif
-
-	vmov.i32	q8,#1		@ compose 1<<96
-	veor		q9,q9,q9
-	vrev32.8	q0,q0
-	vext.8		q8,q9,q8,#4
-	vrev32.8	q4,q4
-	vadd.u32	q9,q8,q8	@ compose 2<<96
-	vstmia	sp, {q4}		@ save adjusted round0 key
-	b	.Lctr_enc_loop
-
-.align	4
-.Lctr_enc_loop:
-	vadd.u32	q10, q8, q9	@ compose 3<<96
-	vadd.u32	q1, q0, q8	@ +1
-	vadd.u32	q2, q0, q9	@ +2
-	vadd.u32	q3, q0, q10	@ +3
-	vadd.u32	q4, q1, q10
-	vadd.u32	q5, q2, q10
-	vadd.u32	q6, q3, q10
-	vadd.u32	q7, q4, q10
-	vadd.u32	q10, q5, q10	@ next counter
-
-	@ Borrow prologue from _bsaes_encrypt8 to use the opportunity
-	@ to flip byte order in 32-bit counter
-
-	vldmia		sp, {q9}		@ load round0 key
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x10		@ pass next round key
-#else
-	add		r4, r3, #264
-#endif
-	vldmia		r8, {q8}			@ .LREVM0SR
-	mov		r5, r10			@ pass rounds
-	vstmia		r9, {q10}			@ save next counter
-	sub		r6, r8, #.LREVM0SR-.LSR	@ pass constants
-
-	bl		_bsaes_encrypt8_alt
-
-	subs		r2, r2, #8
-	blo		.Lctr_enc_loop_done
-
-	vld1.8		{q8-q9}, [r0]!	@ load input
-	vld1.8		{q10-q11}, [r0]!
-	veor		q0, q8
-	veor		q1, q9
-	vld1.8		{q12-q13}, [r0]!
-	veor		q4, q10
-	veor		q6, q11
-	vld1.8		{q14-q15}, [r0]!
-	veor		q3, q12
-	vst1.8		{q0-q1}, [r1]!	@ write output
-	veor		q7, q13
-	veor		q2, q14
-	vst1.8		{q4}, [r1]!
-	veor		q5, q15
-	vst1.8		{q6}, [r1]!
-	vmov.i32	q8, #1			@ compose 1<<96
-	vst1.8		{q3}, [r1]!
-	veor		q9, q9, q9
-	vst1.8		{q7}, [r1]!
-	vext.8		q8, q9, q8, #4
-	vst1.8		{q2}, [r1]!
-	vadd.u32	q9,q8,q8		@ compose 2<<96
-	vst1.8		{q5}, [r1]!
-	vldmia		r9, {q0}			@ load counter
-
-	bne		.Lctr_enc_loop
-	b		.Lctr_enc_done
-
-.align	4
-.Lctr_enc_loop_done:
-	add		r2, r2, #8
-	vld1.8		{q8}, [r0]!	@ load input
-	veor		q0, q8
-	vst1.8		{q0}, [r1]!	@ write output
-	cmp		r2, #2
-	blo		.Lctr_enc_done
-	vld1.8		{q9}, [r0]!
-	veor		q1, q9
-	vst1.8		{q1}, [r1]!
-	beq		.Lctr_enc_done
-	vld1.8		{q10}, [r0]!
-	veor		q4, q10
-	vst1.8		{q4}, [r1]!
-	cmp		r2, #4
-	blo		.Lctr_enc_done
-	vld1.8		{q11}, [r0]!
-	veor		q6, q11
-	vst1.8		{q6}, [r1]!
-	beq		.Lctr_enc_done
-	vld1.8		{q12}, [r0]!
-	veor		q3, q12
-	vst1.8		{q3}, [r1]!
-	cmp		r2, #6
-	blo		.Lctr_enc_done
-	vld1.8		{q13}, [r0]!
-	veor		q7, q13
-	vst1.8		{q7}, [r1]!
-	beq		.Lctr_enc_done
-	vld1.8		{q14}, [r0]
-	veor		q2, q14
-	vst1.8		{q2}, [r1]!
-
-.Lctr_enc_done:
-	vmov.i32	q0, #0
-	vmov.i32	q1, #0
-#ifndef	BSAES_ASM_EXTENDED_KEY
-.Lctr_enc_bzero:			@ wipe key schedule [if any]
-	vstmia		sp!, {q0-q1}
-	cmp		sp, r9
-	bne		.Lctr_enc_bzero
-#else
-	vstmia		sp, {q0-q1}
-#endif
-
-	mov	sp, r9
-	add	sp, #0x10		@ add sp,r9,#0x10 is no good for thumb
-	VFP_ABI_POP
-	ldmia	sp!, {r4-r10, pc}	@ return
-
-.align	4
-.Lctr_enc_short:
-	ldr	ip, [sp]		@ ctr pointer is passed on stack
-	stmdb	sp!, {r4-r8, lr}
-
-	mov	r4, r0		@ copy arguments
-	mov	r5, r1
-	mov	r6, r2
-	mov	r7, r3
-	ldr	r8, [ip, #12]		@ load counter LSW
-	vld1.8	{q1}, [ip]		@ load whole counter value
-#ifdef __ARMEL__
-	rev	r8, r8
-#endif
-	sub	sp, sp, #0x10
-	vst1.8	{q1}, [sp,:64]	@ copy counter value
-	sub	sp, sp, #0x10
-
-.Lctr_enc_short_loop:
-	add	r0, sp, #0x10		@ input counter value
-	mov	r1, sp			@ output on the stack
-	mov	r2, r7			@ key
-
-	bl	AES_encrypt
-
-	vld1.8	{q0}, [r4]!	@ load input
-	vld1.8	{q1}, [sp,:64]	@ load encrypted counter
-	add	r8, r8, #1
-#ifdef __ARMEL__
-	rev	r0, r8
-	str	r0, [sp, #0x1c]		@ next counter value
-#else
-	str	r8, [sp, #0x1c]		@ next counter value
-#endif
-	veor	q0,q0,q1
-	vst1.8	{q0}, [r5]!	@ store output
-	subs	r6, r6, #1
-	bne	.Lctr_enc_short_loop
-
-	vmov.i32	q0, #0
-	vmov.i32	q1, #0
-	vstmia		sp!, {q0-q1}
-
-	ldmia	sp!, {r4-r8, pc}
-.size	bsaes_ctr32_encrypt_blocks,.-bsaes_ctr32_encrypt_blocks
-.globl	bsaes_xts_encrypt
-.type	bsaes_xts_encrypt,%function
-.align	4
-bsaes_xts_encrypt:
-	mov	ip, sp
-	stmdb	sp!, {r4-r10, lr}		@ 0x20
-	VFP_ABI_PUSH
-	mov	r6, sp				@ future r3
-
-	mov	r7, r0
-	mov	r8, r1
-	mov	r9, r2
-	mov	r10, r3
-
-	sub	r0, sp, #0x10			@ 0x10
-	bic	r0, #0xf			@ align at 16 bytes
-	mov	sp, r0
-
-#ifdef	XTS_CHAIN_TWEAK
-	ldr	r0, [ip]			@ pointer to input tweak
-#else
-	@ generate initial tweak
-	ldr	r0, [ip, #4]			@ iv[]
-	mov	r1, sp
-	ldr	r2, [ip, #0]			@ key2
-	bl	AES_encrypt
-	mov	r0,sp				@ pointer to initial tweak
-#endif
-
-	ldr	r1, [r10, #240]		@ get # of rounds
-	mov	r3, r6
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	@ allocate the key schedule on the stack
-	sub	r12, sp, r1, lsl#7		@ 128 bytes per inner round key
-	@ add	r12, #96			@ size of bit-sliced key schedule
-	sub	r12, #48			@ place for tweak[9]
-
-	@ populate the key schedule
-	mov	r4, r10			@ pass key
-	mov	r5, r1			@ pass # of rounds
-	mov	sp, r12
-	add	r12, #0x90			@ pass key schedule
-	bl	_bsaes_key_convert
-	veor	q7, q7, q15	@ fix up last round key
-	vstmia	r12, {q7}			@ save last round key
-#else
-	ldr	r12, [r10, #244]
-	eors	r12, #1
-	beq	0f
-
-	str	r12, [r10, #244]
-	mov	r4, r10			@ pass key
-	mov	r5, r1			@ pass # of rounds
-	add	r12, r10, #248			@ pass key schedule
-	bl	_bsaes_key_convert
-	veor	q7, q7, q15	@ fix up last round key
-	vstmia	r12, {q7}
-
-.align	2
-0:	sub	sp, #0x90			@ place for tweak[9]
-#endif
-
-	vld1.8	{q8}, [r0]			@ initial tweak
-	adr	r2, .Lxts_magic
-
-	subs	r9, #0x80
-	blo	.Lxts_enc_short
-	b	.Lxts_enc_loop
-
-.align	4
-.Lxts_enc_loop:
-	vldmia		r2, {q5}	@ load XTS magic
-	vshr.s64	q6, q8, #63
-	mov		r0, sp
-	vand		q6, q6, q5
-	vadd.u64	q9, q8, q8
-	vst1.64		{q8}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q9, #63
-	veor		q9, q9, q6
-	vand		q7, q7, q5
-	vadd.u64	q10, q9, q9
-	vst1.64		{q9}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q10, #63
-	veor		q10, q10, q7
-	vand		q6, q6, q5
-	vld1.8		{q0}, [r7]!
-	vadd.u64	q11, q10, q10
-	vst1.64		{q10}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q11, #63
-	veor		q11, q11, q6
-	vand		q7, q7, q5
-	vld1.8		{q1}, [r7]!
-	veor		q0, q0, q8
-	vadd.u64	q12, q11, q11
-	vst1.64		{q11}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q12, #63
-	veor		q12, q12, q7
-	vand		q6, q6, q5
-	vld1.8		{q2}, [r7]!
-	veor		q1, q1, q9
-	vadd.u64	q13, q12, q12
-	vst1.64		{q12}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q13, #63
-	veor		q13, q13, q6
-	vand		q7, q7, q5
-	vld1.8		{q3}, [r7]!
-	veor		q2, q2, q10
-	vadd.u64	q14, q13, q13
-	vst1.64		{q13}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q14, #63
-	veor		q14, q14, q7
-	vand		q6, q6, q5
-	vld1.8		{q4}, [r7]!
-	veor		q3, q3, q11
-	vadd.u64	q15, q14, q14
-	vst1.64		{q14}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q15, #63
-	veor		q15, q15, q6
-	vand		q7, q7, q5
-	vld1.8		{q5}, [r7]!
-	veor		q4, q4, q12
-	vadd.u64	q8, q15, q15
-	vst1.64		{q15}, [r0,:128]!
-	vswp		d15,d14
-	veor		q8, q8, q7
-	vst1.64		{q8}, [r0,:128]		@ next round tweak
-
-	vld1.8		{q6-q7}, [r7]!
-	veor		q5, q5, q13
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q6, q6, q14
-	mov		r5, r1			@ pass rounds
-	veor		q7, q7, q15
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10-q11}, [r0,:128]!
-	veor		q0, q0, q8
-	vld1.64		{q12-q13}, [r0,:128]!
-	veor		q1, q1, q9
-	veor		q8, q4, q10
-	vst1.8		{q0-q1}, [r8]!
-	veor		q9, q6, q11
-	vld1.64		{q14-q15}, [r0,:128]!
-	veor		q10, q3, q12
-	vst1.8		{q8-q9}, [r8]!
-	veor		q11, q7, q13
-	veor		q12, q2, q14
-	vst1.8		{q10-q11}, [r8]!
-	veor		q13, q5, q15
-	vst1.8		{q12-q13}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-
-	subs		r9, #0x80
-	bpl		.Lxts_enc_loop
-
-.Lxts_enc_short:
-	adds		r9, #0x70
-	bmi		.Lxts_enc_done
-
-	vldmia		r2, {q5}	@ load XTS magic
-	vshr.s64	q7, q8, #63
-	mov		r0, sp
-	vand		q7, q7, q5
-	vadd.u64	q9, q8, q8
-	vst1.64		{q8}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q9, #63
-	veor		q9, q9, q7
-	vand		q6, q6, q5
-	vadd.u64	q10, q9, q9
-	vst1.64		{q9}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q10, #63
-	veor		q10, q10, q6
-	vand		q7, q7, q5
-	vld1.8		{q0}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_enc_1
-	vadd.u64	q11, q10, q10
-	vst1.64		{q10}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q11, #63
-	veor		q11, q11, q7
-	vand		q6, q6, q5
-	vld1.8		{q1}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_enc_2
-	veor		q0, q0, q8
-	vadd.u64	q12, q11, q11
-	vst1.64		{q11}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q12, #63
-	veor		q12, q12, q6
-	vand		q7, q7, q5
-	vld1.8		{q2}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_enc_3
-	veor		q1, q1, q9
-	vadd.u64	q13, q12, q12
-	vst1.64		{q12}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q13, #63
-	veor		q13, q13, q7
-	vand		q6, q6, q5
-	vld1.8		{q3}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_enc_4
-	veor		q2, q2, q10
-	vadd.u64	q14, q13, q13
-	vst1.64		{q13}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q14, #63
-	veor		q14, q14, q6
-	vand		q7, q7, q5
-	vld1.8		{q4}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_enc_5
-	veor		q3, q3, q11
-	vadd.u64	q15, q14, q14
-	vst1.64		{q14}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q15, #63
-	veor		q15, q15, q7
-	vand		q6, q6, q5
-	vld1.8		{q5}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_enc_6
-	veor		q4, q4, q12
-	sub		r9, #0x10
-	vst1.64		{q15}, [r0,:128]		@ next round tweak
-
-	vld1.8		{q6}, [r7]!
-	veor		q5, q5, q13
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q6, q6, q14
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10-q11}, [r0,:128]!
-	veor		q0, q0, q8
-	vld1.64		{q12-q13}, [r0,:128]!
-	veor		q1, q1, q9
-	veor		q8, q4, q10
-	vst1.8		{q0-q1}, [r8]!
-	veor		q9, q6, q11
-	vld1.64		{q14}, [r0,:128]!
-	veor		q10, q3, q12
-	vst1.8		{q8-q9}, [r8]!
-	veor		q11, q7, q13
-	veor		q12, q2, q14
-	vst1.8		{q10-q11}, [r8]!
-	vst1.8		{q12}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-.align	4
-.Lxts_enc_6:
-	vst1.64		{q14}, [r0,:128]		@ next round tweak
-
-	veor		q4, q4, q12
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q5, q5, q13
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10-q11}, [r0,:128]!
-	veor		q0, q0, q8
-	vld1.64		{q12-q13}, [r0,:128]!
-	veor		q1, q1, q9
-	veor		q8, q4, q10
-	vst1.8		{q0-q1}, [r8]!
-	veor		q9, q6, q11
-	veor		q10, q3, q12
-	vst1.8		{q8-q9}, [r8]!
-	veor		q11, q7, q13
-	vst1.8		{q10-q11}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-
-@ put this in range for both ARM and Thumb mode adr instructions
-.align	5
-.Lxts_magic:
-	.quad	1, 0x87
-
-.align	5
-.Lxts_enc_5:
-	vst1.64		{q13}, [r0,:128]		@ next round tweak
-
-	veor		q3, q3, q11
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q4, q4, q12
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10-q11}, [r0,:128]!
-	veor		q0, q0, q8
-	vld1.64		{q12}, [r0,:128]!
-	veor		q1, q1, q9
-	veor		q8, q4, q10
-	vst1.8		{q0-q1}, [r8]!
-	veor		q9, q6, q11
-	veor		q10, q3, q12
-	vst1.8		{q8-q9}, [r8]!
-	vst1.8		{q10}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-.align	4
-.Lxts_enc_4:
-	vst1.64		{q12}, [r0,:128]		@ next round tweak
-
-	veor		q2, q2, q10
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q3, q3, q11
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10-q11}, [r0,:128]!
-	veor		q0, q0, q8
-	veor		q1, q1, q9
-	veor		q8, q4, q10
-	vst1.8		{q0-q1}, [r8]!
-	veor		q9, q6, q11
-	vst1.8		{q8-q9}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-.align	4
-.Lxts_enc_3:
-	vst1.64		{q11}, [r0,:128]		@ next round tweak
-
-	veor		q1, q1, q9
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q2, q2, q10
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10}, [r0,:128]!
-	veor		q0, q0, q8
-	veor		q1, q1, q9
-	veor		q8, q4, q10
-	vst1.8		{q0-q1}, [r8]!
-	vst1.8		{q8}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-.align	4
-.Lxts_enc_2:
-	vst1.64		{q10}, [r0,:128]		@ next round tweak
-
-	veor		q0, q0, q8
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q1, q1, q9
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	veor		q0, q0, q8
-	veor		q1, q1, q9
-	vst1.8		{q0-q1}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-.align	4
-.Lxts_enc_1:
-	mov		r0, sp
-	veor		q0, q8
-	mov		r1, sp
-	vst1.8		{q0}, [sp,:128]
-	mov		r2, r10
-	mov		r4, r3				@ preserve fp
-
-	bl		AES_encrypt
-
-	vld1.8		{q0}, [sp,:128]
-	veor		q0, q0, q8
-	vst1.8		{q0}, [r8]!
-	mov		r3, r4
-
-	vmov		q8, q9		@ next round tweak
-
-.Lxts_enc_done:
-#ifndef	XTS_CHAIN_TWEAK
-	adds		r9, #0x10
-	beq		.Lxts_enc_ret
-	sub		r6, r8, #0x10
-
-.Lxts_enc_steal:
-	ldrb		r0, [r7], #1
-	ldrb		r1, [r8, #-0x10]
-	strb		r0, [r8, #-0x10]
-	strb		r1, [r8], #1
-
-	subs		r9, #1
-	bhi		.Lxts_enc_steal
-
-	vld1.8		{q0}, [r6]
-	mov		r0, sp
-	veor		q0, q0, q8
-	mov		r1, sp
-	vst1.8		{q0}, [sp,:128]
-	mov		r2, r10
-	mov		r4, r3			@ preserve fp
-
-	bl		AES_encrypt
-
-	vld1.8		{q0}, [sp,:128]
-	veor		q0, q0, q8
-	vst1.8		{q0}, [r6]
-	mov		r3, r4
-#endif
-
-.Lxts_enc_ret:
-	bic		r0, r3, #0xf
-	vmov.i32	q0, #0
-	vmov.i32	q1, #0
-#ifdef	XTS_CHAIN_TWEAK
-	ldr		r1, [r3, #0x20+VFP_ABI_FRAME]	@ chain tweak
-#endif
-.Lxts_enc_bzero:				@ wipe key schedule [if any]
-	vstmia		sp!, {q0-q1}
-	cmp		sp, r0
-	bne		.Lxts_enc_bzero
-
-	mov		sp, r3
-#ifdef	XTS_CHAIN_TWEAK
-	vst1.8		{q8}, [r1]
-#endif
-	VFP_ABI_POP
-	ldmia		sp!, {r4-r10, pc}	@ return
-
-.size	bsaes_xts_encrypt,.-bsaes_xts_encrypt
-
-.globl	bsaes_xts_decrypt
-.type	bsaes_xts_decrypt,%function
-.align	4
-bsaes_xts_decrypt:
-	mov	ip, sp
-	stmdb	sp!, {r4-r10, lr}		@ 0x20
-	VFP_ABI_PUSH
-	mov	r6, sp				@ future r3
-
-	mov	r7, r0
-	mov	r8, r1
-	mov	r9, r2
-	mov	r10, r3
-
-	sub	r0, sp, #0x10			@ 0x10
-	bic	r0, #0xf			@ align at 16 bytes
-	mov	sp, r0
-
-#ifdef	XTS_CHAIN_TWEAK
-	ldr	r0, [ip]			@ pointer to input tweak
-#else
-	@ generate initial tweak
-	ldr	r0, [ip, #4]			@ iv[]
-	mov	r1, sp
-	ldr	r2, [ip, #0]			@ key2
-	bl	AES_encrypt
-	mov	r0, sp				@ pointer to initial tweak
-#endif
-
-	ldr	r1, [r10, #240]		@ get # of rounds
-	mov	r3, r6
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	@ allocate the key schedule on the stack
-	sub	r12, sp, r1, lsl#7		@ 128 bytes per inner round key
-	@ add	r12, #96			@ size of bit-sliced key schedule
-	sub	r12, #48			@ place for tweak[9]
-
-	@ populate the key schedule
-	mov	r4, r10			@ pass key
-	mov	r5, r1			@ pass # of rounds
-	mov	sp, r12
-	add	r12, #0x90			@ pass key schedule
-	bl	_bsaes_key_convert
-	add	r4, sp, #0x90
-	vldmia	r4, {q6}
-	vstmia	r12,  {q15}		@ save last round key
-	veor	q7, q7, q6	@ fix up round 0 key
-	vstmia	r4, {q7}
-#else
-	ldr	r12, [r10, #244]
-	eors	r12, #1
-	beq	0f
-
-	str	r12, [r10, #244]
-	mov	r4, r10			@ pass key
-	mov	r5, r1			@ pass # of rounds
-	add	r12, r10, #248			@ pass key schedule
-	bl	_bsaes_key_convert
-	add	r4, r10, #248
-	vldmia	r4, {q6}
-	vstmia	r12,  {q15}		@ save last round key
-	veor	q7, q7, q6	@ fix up round 0 key
-	vstmia	r4, {q7}
-
-.align	2
-0:	sub	sp, #0x90			@ place for tweak[9]
-#endif
-	vld1.8	{q8}, [r0]			@ initial tweak
-	adr	r2, .Lxts_magic
-
-#ifndef	XTS_CHAIN_TWEAK
-	tst	r9, #0xf			@ if not multiple of 16
-	it	ne				@ Thumb2 thing, sanity check in ARM
-	subne	r9, #0x10			@ subtract another 16 bytes
-#endif
-	subs	r9, #0x80
-
-	blo	.Lxts_dec_short
-	b	.Lxts_dec_loop
-
-.align	4
-.Lxts_dec_loop:
-	vldmia		r2, {q5}	@ load XTS magic
-	vshr.s64	q6, q8, #63
-	mov		r0, sp
-	vand		q6, q6, q5
-	vadd.u64	q9, q8, q8
-	vst1.64		{q8}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q9, #63
-	veor		q9, q9, q6
-	vand		q7, q7, q5
-	vadd.u64	q10, q9, q9
-	vst1.64		{q9}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q10, #63
-	veor		q10, q10, q7
-	vand		q6, q6, q5
-	vld1.8		{q0}, [r7]!
-	vadd.u64	q11, q10, q10
-	vst1.64		{q10}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q11, #63
-	veor		q11, q11, q6
-	vand		q7, q7, q5
-	vld1.8		{q1}, [r7]!
-	veor		q0, q0, q8
-	vadd.u64	q12, q11, q11
-	vst1.64		{q11}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q12, #63
-	veor		q12, q12, q7
-	vand		q6, q6, q5
-	vld1.8		{q2}, [r7]!
-	veor		q1, q1, q9
-	vadd.u64	q13, q12, q12
-	vst1.64		{q12}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q13, #63
-	veor		q13, q13, q6
-	vand		q7, q7, q5
-	vld1.8		{q3}, [r7]!
-	veor		q2, q2, q10
-	vadd.u64	q14, q13, q13
-	vst1.64		{q13}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q14, #63
-	veor		q14, q14, q7
-	vand		q6, q6, q5
-	vld1.8		{q4}, [r7]!
-	veor		q3, q3, q11
-	vadd.u64	q15, q14, q14
-	vst1.64		{q14}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q15, #63
-	veor		q15, q15, q6
-	vand		q7, q7, q5
-	vld1.8		{q5}, [r7]!
-	veor		q4, q4, q12
-	vadd.u64	q8, q15, q15
-	vst1.64		{q15}, [r0,:128]!
-	vswp		d15,d14
-	veor		q8, q8, q7
-	vst1.64		{q8}, [r0,:128]		@ next round tweak
-
-	vld1.8		{q6-q7}, [r7]!
-	veor		q5, q5, q13
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q6, q6, q14
-	mov		r5, r1			@ pass rounds
-	veor		q7, q7, q15
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10-q11}, [r0,:128]!
-	veor		q0, q0, q8
-	vld1.64		{q12-q13}, [r0,:128]!
-	veor		q1, q1, q9
-	veor		q8, q6, q10
-	vst1.8		{q0-q1}, [r8]!
-	veor		q9, q4, q11
-	vld1.64		{q14-q15}, [r0,:128]!
-	veor		q10, q2, q12
-	vst1.8		{q8-q9}, [r8]!
-	veor		q11, q7, q13
-	veor		q12, q3, q14
-	vst1.8		{q10-q11}, [r8]!
-	veor		q13, q5, q15
-	vst1.8		{q12-q13}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-
-	subs		r9, #0x80
-	bpl		.Lxts_dec_loop
-
-.Lxts_dec_short:
-	adds		r9, #0x70
-	bmi		.Lxts_dec_done
-
-	vldmia		r2, {q5}	@ load XTS magic
-	vshr.s64	q7, q8, #63
-	mov		r0, sp
-	vand		q7, q7, q5
-	vadd.u64	q9, q8, q8
-	vst1.64		{q8}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q9, #63
-	veor		q9, q9, q7
-	vand		q6, q6, q5
-	vadd.u64	q10, q9, q9
-	vst1.64		{q9}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q10, #63
-	veor		q10, q10, q6
-	vand		q7, q7, q5
-	vld1.8		{q0}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_dec_1
-	vadd.u64	q11, q10, q10
-	vst1.64		{q10}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q11, #63
-	veor		q11, q11, q7
-	vand		q6, q6, q5
-	vld1.8		{q1}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_dec_2
-	veor		q0, q0, q8
-	vadd.u64	q12, q11, q11
-	vst1.64		{q11}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q12, #63
-	veor		q12, q12, q6
-	vand		q7, q7, q5
-	vld1.8		{q2}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_dec_3
-	veor		q1, q1, q9
-	vadd.u64	q13, q12, q12
-	vst1.64		{q12}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q13, #63
-	veor		q13, q13, q7
-	vand		q6, q6, q5
-	vld1.8		{q3}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_dec_4
-	veor		q2, q2, q10
-	vadd.u64	q14, q13, q13
-	vst1.64		{q13}, [r0,:128]!
-	vswp		d13,d12
-	vshr.s64	q7, q14, #63
-	veor		q14, q14, q6
-	vand		q7, q7, q5
-	vld1.8		{q4}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_dec_5
-	veor		q3, q3, q11
-	vadd.u64	q15, q14, q14
-	vst1.64		{q14}, [r0,:128]!
-	vswp		d15,d14
-	vshr.s64	q6, q15, #63
-	veor		q15, q15, q7
-	vand		q6, q6, q5
-	vld1.8		{q5}, [r7]!
-	subs		r9, #0x10
-	bmi		.Lxts_dec_6
-	veor		q4, q4, q12
-	sub		r9, #0x10
-	vst1.64		{q15}, [r0,:128]		@ next round tweak
-
-	vld1.8		{q6}, [r7]!
-	veor		q5, q5, q13
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q6, q6, q14
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10-q11}, [r0,:128]!
-	veor		q0, q0, q8
-	vld1.64		{q12-q13}, [r0,:128]!
-	veor		q1, q1, q9
-	veor		q8, q6, q10
-	vst1.8		{q0-q1}, [r8]!
-	veor		q9, q4, q11
-	vld1.64		{q14}, [r0,:128]!
-	veor		q10, q2, q12
-	vst1.8		{q8-q9}, [r8]!
-	veor		q11, q7, q13
-	veor		q12, q3, q14
-	vst1.8		{q10-q11}, [r8]!
-	vst1.8		{q12}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_6:
-	vst1.64		{q14}, [r0,:128]		@ next round tweak
-
-	veor		q4, q4, q12
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q5, q5, q13
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10-q11}, [r0,:128]!
-	veor		q0, q0, q8
-	vld1.64		{q12-q13}, [r0,:128]!
-	veor		q1, q1, q9
-	veor		q8, q6, q10
-	vst1.8		{q0-q1}, [r8]!
-	veor		q9, q4, q11
-	veor		q10, q2, q12
-	vst1.8		{q8-q9}, [r8]!
-	veor		q11, q7, q13
-	vst1.8		{q10-q11}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_5:
-	vst1.64		{q13}, [r0,:128]		@ next round tweak
-
-	veor		q3, q3, q11
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q4, q4, q12
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10-q11}, [r0,:128]!
-	veor		q0, q0, q8
-	vld1.64		{q12}, [r0,:128]!
-	veor		q1, q1, q9
-	veor		q8, q6, q10
-	vst1.8		{q0-q1}, [r8]!
-	veor		q9, q4, q11
-	veor		q10, q2, q12
-	vst1.8		{q8-q9}, [r8]!
-	vst1.8		{q10}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_4:
-	vst1.64		{q12}, [r0,:128]		@ next round tweak
-
-	veor		q2, q2, q10
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q3, q3, q11
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10-q11}, [r0,:128]!
-	veor		q0, q0, q8
-	veor		q1, q1, q9
-	veor		q8, q6, q10
-	vst1.8		{q0-q1}, [r8]!
-	veor		q9, q4, q11
-	vst1.8		{q8-q9}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_3:
-	vst1.64		{q11}, [r0,:128]		@ next round tweak
-
-	veor		q1, q1, q9
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q2, q2, q10
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	vld1.64		{q10}, [r0,:128]!
-	veor		q0, q0, q8
-	veor		q1, q1, q9
-	veor		q8, q6, q10
-	vst1.8		{q0-q1}, [r8]!
-	vst1.8		{q8}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_2:
-	vst1.64		{q10}, [r0,:128]		@ next round tweak
-
-	veor		q0, q0, q8
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, r10, #248			@ pass key schedule
-#endif
-	veor		q1, q1, q9
-	mov		r5, r1			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{q8-q9}, [r0,:128]!
-	veor		q0, q0, q8
-	veor		q1, q1, q9
-	vst1.8		{q0-q1}, [r8]!
-
-	vld1.64		{q8}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_1:
-	mov		r0, sp
-	veor		q0, q8
-	mov		r1, sp
-	vst1.8		{q0}, [sp,:128]
-	mov		r2, r10
-	mov		r4, r3				@ preserve fp
-	mov		r5, r2			@ preserve magic
-
-	bl		AES_decrypt
-
-	vld1.8		{q0}, [sp,:128]
-	veor		q0, q0, q8
-	vst1.8		{q0}, [r8]!
-	mov		r3, r4
-	mov		r2, r5
-
-	vmov		q8, q9		@ next round tweak
-
-.Lxts_dec_done:
-#ifndef	XTS_CHAIN_TWEAK
-	adds		r9, #0x10
-	beq		.Lxts_dec_ret
-
-	@ calculate one round of extra tweak for the stolen ciphertext
-	vldmia		r2, {q5}
-	vshr.s64	q6, q8, #63
-	vand		q6, q6, q5
-	vadd.u64	q9, q8, q8
-	vswp		d13,d12
-	veor		q9, q9, q6
-
-	@ perform the final decryption with the last tweak value
-	vld1.8		{q0}, [r7]!
-	mov		r0, sp
-	veor		q0, q0, q9
-	mov		r1, sp
-	vst1.8		{q0}, [sp,:128]
-	mov		r2, r10
-	mov		r4, r3			@ preserve fp
-
-	bl		AES_decrypt
-
-	vld1.8		{q0}, [sp,:128]
-	veor		q0, q0, q9
-	vst1.8		{q0}, [r8]
-
-	mov		r6, r8
-.Lxts_dec_steal:
-	ldrb		r1, [r8]
-	ldrb		r0, [r7], #1
-	strb		r1, [r8, #0x10]
-	strb		r0, [r8], #1
-
-	subs		r9, #1
-	bhi		.Lxts_dec_steal
-
-	vld1.8		{q0}, [r6]
-	mov		r0, sp
-	veor		q0, q8
-	mov		r1, sp
-	vst1.8		{q0}, [sp,:128]
-	mov		r2, r10
-
-	bl		AES_decrypt
-
-	vld1.8		{q0}, [sp,:128]
-	veor		q0, q0, q8
-	vst1.8		{q0}, [r6]
-	mov		r3, r4
-#endif
-
-.Lxts_dec_ret:
-	bic		r0, r3, #0xf
-	vmov.i32	q0, #0
-	vmov.i32	q1, #0
-#ifdef	XTS_CHAIN_TWEAK
-	ldr		r1, [r3, #0x20+VFP_ABI_FRAME]	@ chain tweak
-#endif
-.Lxts_dec_bzero:				@ wipe key schedule [if any]
-	vstmia		sp!, {q0-q1}
-	cmp		sp, r0
-	bne		.Lxts_dec_bzero
-
-	mov		sp, r3
-#ifdef	XTS_CHAIN_TWEAK
-	vst1.8		{q8}, [r1]
-#endif
-	VFP_ABI_POP
-	ldmia		sp!, {r4-r10, pc}	@ return
-
-.size	bsaes_xts_decrypt,.-bsaes_xts_decrypt
-#endif
diff --git a/arch/arm/crypto/aesbs-glue.c b/arch/arm/crypto/aesbs-glue.c
deleted file mode 100644
index d8e06de72ef3..000000000000
--- a/arch/arm/crypto/aesbs-glue.c
+++ /dev/null
@@ -1,367 +0,0 @@
-/*
- * linux/arch/arm/crypto/aesbs-glue.c - glue code for NEON bit sliced AES
- *
- * Copyright (C) 2013 Linaro Ltd <ard.biesheuvel@linaro.org>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#include <asm/neon.h>
-#include <crypto/aes.h>
-#include <crypto/cbc.h>
-#include <crypto/internal/simd.h>
-#include <crypto/internal/skcipher.h>
-#include <linux/module.h>
-#include <crypto/xts.h>
-
-#include "aes_glue.h"
-
-#define BIT_SLICED_KEY_MAXSIZE	(128 * (AES_MAXNR - 1) + 2 * AES_BLOCK_SIZE)
-
-struct BS_KEY {
-	struct AES_KEY	rk;
-	int		converted;
-	u8 __aligned(8)	bs[BIT_SLICED_KEY_MAXSIZE];
-} __aligned(8);
-
-asmlinkage void bsaes_enc_key_convert(u8 out[], struct AES_KEY const *in);
-asmlinkage void bsaes_dec_key_convert(u8 out[], struct AES_KEY const *in);
-
-asmlinkage void bsaes_cbc_encrypt(u8 const in[], u8 out[], u32 bytes,
-				  struct BS_KEY *key, u8 iv[]);
-
-asmlinkage void bsaes_ctr32_encrypt_blocks(u8 const in[], u8 out[], u32 blocks,
-					   struct BS_KEY *key, u8 const iv[]);
-
-asmlinkage void bsaes_xts_encrypt(u8 const in[], u8 out[], u32 bytes,
-				  struct BS_KEY *key, u8 tweak[]);
-
-asmlinkage void bsaes_xts_decrypt(u8 const in[], u8 out[], u32 bytes,
-				  struct BS_KEY *key, u8 tweak[]);
-
-struct aesbs_cbc_ctx {
-	struct AES_KEY	enc;
-	struct BS_KEY	dec;
-};
-
-struct aesbs_ctr_ctx {
-	struct BS_KEY	enc;
-};
-
-struct aesbs_xts_ctx {
-	struct BS_KEY	enc;
-	struct BS_KEY	dec;
-	struct AES_KEY	twkey;
-};
-
-static int aesbs_cbc_set_key(struct crypto_skcipher *tfm, const u8 *in_key,
-			     unsigned int key_len)
-{
-	struct aesbs_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
-	int bits = key_len * 8;
-
-	if (private_AES_set_encrypt_key(in_key, bits, &ctx->enc)) {
-		crypto_skcipher_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
-		return -EINVAL;
-	}
-	ctx->dec.rk = ctx->enc;
-	private_AES_set_decrypt_key(in_key, bits, &ctx->dec.rk);
-	ctx->dec.converted = 0;
-	return 0;
-}
-
-static int aesbs_ctr_set_key(struct crypto_skcipher *tfm, const u8 *in_key,
-			     unsigned int key_len)
-{
-	struct aesbs_ctr_ctx *ctx = crypto_skcipher_ctx(tfm);
-	int bits = key_len * 8;
-
-	if (private_AES_set_encrypt_key(in_key, bits, &ctx->enc.rk)) {
-		crypto_skcipher_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
-		return -EINVAL;
-	}
-	ctx->enc.converted = 0;
-	return 0;
-}
-
-static int aesbs_xts_set_key(struct crypto_skcipher *tfm, const u8 *in_key,
-			     unsigned int key_len)
-{
-	struct aesbs_xts_ctx *ctx = crypto_skcipher_ctx(tfm);
-	int bits = key_len * 4;
-	int err;
-
-	err = xts_verify_key(tfm, in_key, key_len);
-	if (err)
-		return err;
-
-	if (private_AES_set_encrypt_key(in_key, bits, &ctx->enc.rk)) {
-		crypto_skcipher_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
-		return -EINVAL;
-	}
-	ctx->dec.rk = ctx->enc.rk;
-	private_AES_set_decrypt_key(in_key, bits, &ctx->dec.rk);
-	private_AES_set_encrypt_key(in_key + key_len / 2, bits, &ctx->twkey);
-	ctx->enc.converted = ctx->dec.converted = 0;
-	return 0;
-}
-
-static inline void aesbs_encrypt_one(struct crypto_skcipher *tfm,
-				     const u8 *src, u8 *dst)
-{
-	struct aesbs_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
-
-	AES_encrypt(src, dst, &ctx->enc);
-}
-
-static int aesbs_cbc_encrypt(struct skcipher_request *req)
-{
-	return crypto_cbc_encrypt_walk(req, aesbs_encrypt_one);
-}
-
-static inline void aesbs_decrypt_one(struct crypto_skcipher *tfm,
-				     const u8 *src, u8 *dst)
-{
-	struct aesbs_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
-
-	AES_decrypt(src, dst, &ctx->dec.rk);
-}
-
-static int aesbs_cbc_decrypt(struct skcipher_request *req)
-{
-	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
-	struct aesbs_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
-	struct skcipher_walk walk;
-	unsigned int nbytes;
-	int err;
-
-	for (err = skcipher_walk_virt(&walk, req, false);
-	     (nbytes = walk.nbytes); err = skcipher_walk_done(&walk, nbytes)) {
-		u32 blocks = nbytes / AES_BLOCK_SIZE;
-		u8 *dst = walk.dst.virt.addr;
-		u8 *src = walk.src.virt.addr;
-		u8 *iv = walk.iv;
-
-		if (blocks >= 8) {
-			kernel_neon_begin();
-			bsaes_cbc_encrypt(src, dst, nbytes, &ctx->dec, iv);
-			kernel_neon_end();
-			nbytes %= AES_BLOCK_SIZE;
-			continue;
-		}
-
-		nbytes = crypto_cbc_decrypt_blocks(&walk, tfm,
-						   aesbs_decrypt_one);
-	}
-	return err;
-}
-
-static void inc_be128_ctr(__be32 ctr[], u32 addend)
-{
-	int i;
-
-	for (i = 3; i >= 0; i--, addend = 1) {
-		u32 n = be32_to_cpu(ctr[i]) + addend;
-
-		ctr[i] = cpu_to_be32(n);
-		if (n >= addend)
-			break;
-	}
-}
-
-static int aesbs_ctr_encrypt(struct skcipher_request *req)
-{
-	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
-	struct aesbs_ctr_ctx *ctx = crypto_skcipher_ctx(tfm);
-	struct skcipher_walk walk;
-	u32 blocks;
-	int err;
-
-	err = skcipher_walk_virt(&walk, req, false);
-
-	while ((blocks = walk.nbytes / AES_BLOCK_SIZE)) {
-		u32 tail = walk.nbytes % AES_BLOCK_SIZE;
-		__be32 *ctr = (__be32 *)walk.iv;
-		u32 headroom = UINT_MAX - be32_to_cpu(ctr[3]);
-
-		/* avoid 32 bit counter overflow in the NEON code */
-		if (unlikely(headroom < blocks)) {
-			blocks = headroom + 1;
-			tail = walk.nbytes - blocks * AES_BLOCK_SIZE;
-		}
-		kernel_neon_begin();
-		bsaes_ctr32_encrypt_blocks(walk.src.virt.addr,
-					   walk.dst.virt.addr, blocks,
-					   &ctx->enc, walk.iv);
-		kernel_neon_end();
-		inc_be128_ctr(ctr, blocks);
-
-		err = skcipher_walk_done(&walk, tail);
-	}
-	if (walk.nbytes) {
-		u8 *tdst = walk.dst.virt.addr + blocks * AES_BLOCK_SIZE;
-		u8 *tsrc = walk.src.virt.addr + blocks * AES_BLOCK_SIZE;
-		u8 ks[AES_BLOCK_SIZE];
-
-		AES_encrypt(walk.iv, ks, &ctx->enc.rk);
-		if (tdst != tsrc)
-			memcpy(tdst, tsrc, walk.nbytes);
-		crypto_xor(tdst, ks, walk.nbytes);
-		err = skcipher_walk_done(&walk, 0);
-	}
-	return err;
-}
-
-static int aesbs_xts_encrypt(struct skcipher_request *req)
-{
-	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
-	struct aesbs_xts_ctx *ctx = crypto_skcipher_ctx(tfm);
-	struct skcipher_walk walk;
-	int err;
-
-	err = skcipher_walk_virt(&walk, req, false);
-
-	/* generate the initial tweak */
-	AES_encrypt(walk.iv, walk.iv, &ctx->twkey);
-
-	while (walk.nbytes) {
-		kernel_neon_begin();
-		bsaes_xts_encrypt(walk.src.virt.addr, walk.dst.virt.addr,
-				  walk.nbytes, &ctx->enc, walk.iv);
-		kernel_neon_end();
-		err = skcipher_walk_done(&walk, walk.nbytes % AES_BLOCK_SIZE);
-	}
-	return err;
-}
-
-static int aesbs_xts_decrypt(struct skcipher_request *req)
-{
-	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
-	struct aesbs_xts_ctx *ctx = crypto_skcipher_ctx(tfm);
-	struct skcipher_walk walk;
-	int err;
-
-	err = skcipher_walk_virt(&walk, req, false);
-
-	/* generate the initial tweak */
-	AES_encrypt(walk.iv, walk.iv, &ctx->twkey);
-
-	while (walk.nbytes) {
-		kernel_neon_begin();
-		bsaes_xts_decrypt(walk.src.virt.addr, walk.dst.virt.addr,
-				  walk.nbytes, &ctx->dec, walk.iv);
-		kernel_neon_end();
-		err = skcipher_walk_done(&walk, walk.nbytes % AES_BLOCK_SIZE);
-	}
-	return err;
-}
-
-static struct skcipher_alg aesbs_algs[] = { {
-	.base = {
-		.cra_name		= "__cbc(aes)",
-		.cra_driver_name	= "__cbc-aes-neonbs",
-		.cra_priority		= 300,
-		.cra_flags		= CRYPTO_ALG_INTERNAL,
-		.cra_blocksize		= AES_BLOCK_SIZE,
-		.cra_ctxsize		= sizeof(struct aesbs_cbc_ctx),
-		.cra_alignmask		= 7,
-		.cra_module		= THIS_MODULE,
-	},
-	.min_keysize	= AES_MIN_KEY_SIZE,
-	.max_keysize	= AES_MAX_KEY_SIZE,
-	.ivsize		= AES_BLOCK_SIZE,
-	.setkey		= aesbs_cbc_set_key,
-	.encrypt	= aesbs_cbc_encrypt,
-	.decrypt	= aesbs_cbc_decrypt,
-}, {
-	.base = {
-		.cra_name		= "__ctr(aes)",
-		.cra_driver_name	= "__ctr-aes-neonbs",
-		.cra_priority		= 300,
-		.cra_flags		= CRYPTO_ALG_INTERNAL,
-		.cra_blocksize		= 1,
-		.cra_ctxsize		= sizeof(struct aesbs_ctr_ctx),
-		.cra_alignmask		= 7,
-		.cra_module		= THIS_MODULE,
-	},
-	.min_keysize	= AES_MIN_KEY_SIZE,
-	.max_keysize	= AES_MAX_KEY_SIZE,
-	.ivsize		= AES_BLOCK_SIZE,
-	.chunksize	= AES_BLOCK_SIZE,
-	.setkey		= aesbs_ctr_set_key,
-	.encrypt	= aesbs_ctr_encrypt,
-	.decrypt	= aesbs_ctr_encrypt,
-}, {
-	.base = {
-		.cra_name		= "__xts(aes)",
-		.cra_driver_name	= "__xts-aes-neonbs",
-		.cra_priority		= 300,
-		.cra_flags		= CRYPTO_ALG_INTERNAL,
-		.cra_blocksize		= AES_BLOCK_SIZE,
-		.cra_ctxsize		= sizeof(struct aesbs_xts_ctx),
-		.cra_alignmask		= 7,
-		.cra_module		= THIS_MODULE,
-	},
-	.min_keysize	= 2 * AES_MIN_KEY_SIZE,
-	.max_keysize	= 2 * AES_MAX_KEY_SIZE,
-	.ivsize		= AES_BLOCK_SIZE,
-	.setkey		= aesbs_xts_set_key,
-	.encrypt	= aesbs_xts_encrypt,
-	.decrypt	= aesbs_xts_decrypt,
-} };
-
-struct simd_skcipher_alg *aesbs_simd_algs[ARRAY_SIZE(aesbs_algs)];
-
-static void aesbs_mod_exit(void)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(aesbs_simd_algs) && aesbs_simd_algs[i]; i++)
-		simd_skcipher_free(aesbs_simd_algs[i]);
-
-	crypto_unregister_skciphers(aesbs_algs, ARRAY_SIZE(aesbs_algs));
-}
-
-static int __init aesbs_mod_init(void)
-{
-	struct simd_skcipher_alg *simd;
-	const char *basename;
-	const char *algname;
-	const char *drvname;
-	int err;
-	int i;
-
-	if (!cpu_has_neon())
-		return -ENODEV;
-
-	err = crypto_register_skciphers(aesbs_algs, ARRAY_SIZE(aesbs_algs));
-	if (err)
-		return err;
-
-	for (i = 0; i < ARRAY_SIZE(aesbs_algs); i++) {
-		algname = aesbs_algs[i].base.cra_name + 2;
-		drvname = aesbs_algs[i].base.cra_driver_name + 2;
-		basename = aesbs_algs[i].base.cra_driver_name;
-		simd = simd_skcipher_create_compat(algname, drvname, basename);
-		err = PTR_ERR(simd);
-		if (IS_ERR(simd))
-			goto unregister_simds;
-
-		aesbs_simd_algs[i] = simd;
-	}
-
-	return 0;
-
-unregister_simds:
-	aesbs_mod_exit();
-	return err;
-}
-
-module_init(aesbs_mod_init);
-module_exit(aesbs_mod_exit);
-
-MODULE_DESCRIPTION("Bit sliced AES in CBC/CTR/XTS modes using NEON");
-MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
-MODULE_LICENSE("GPL");
diff --git a/arch/arm/crypto/bsaes-armv7.pl b/arch/arm/crypto/bsaes-armv7.pl
deleted file mode 100644
index a4d3856e7d24..000000000000
--- a/arch/arm/crypto/bsaes-armv7.pl
+++ /dev/null
@@ -1,2471 +0,0 @@
-#!/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/.
-#
-# Specific modes and adaptation for Linux kernel by Ard Biesheuvel
-# <ard.biesheuvel@linaro.org>. Permission to use under GPL terms is
-# granted.
-# ====================================================================
-
-# Bit-sliced AES for ARM NEON
-#
-# February 2012.
-#
-# This implementation is direct adaptation of bsaes-x86_64 module for
-# ARM NEON. Except that this module is endian-neutral [in sense that
-# it can be compiled for either endianness] by courtesy of vld1.8's
-# neutrality. Initial version doesn't implement interface to OpenSSL,
-# only low-level primitives and unsupported entry points, just enough
-# to collect performance results, which for Cortex-A8 core are:
-#
-# encrypt	19.5 cycles per byte processed with 128-bit key
-# decrypt	22.1 cycles per byte processed with 128-bit key
-# key conv.	440  cycles per 128-bit key/0.18 of 8x block
-#
-# Snapdragon S4 encrypts byte in 17.6 cycles and decrypts in 19.7,
-# which is [much] worse than anticipated (for further details see
-# http://www.openssl.org/~appro/Snapdragon-S4.html).
-#
-# Cortex-A15 manages in 14.2/16.1 cycles [when integer-only code
-# manages in 20.0 cycles].
-#
-# When comparing to x86_64 results keep in mind that NEON unit is
-# [mostly] single-issue and thus can't [fully] benefit from
-# instruction-level parallelism. And when comparing to aes-armv4
-# results keep in mind key schedule conversion overhead (see
-# bsaes-x86_64.pl for further details)...
-#
-#						<appro@openssl.org>
-
-# April-August 2013
-#
-# Add CBC, CTR and XTS subroutines, adapt for kernel use.
-#
-#					<ard.biesheuvel@linaro.org>
-
-while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {}
-open STDOUT,">$output";
-
-my ($inp,$out,$len,$key)=("r0","r1","r2","r3");
-my @XMM=map("q$_",(0..15));
-
-{
-my ($key,$rounds,$const)=("r4","r5","r6");
-
-sub Dlo()   { shift=~m|q([1]?[0-9])|?"d".($1*2):"";     }
-sub Dhi()   { shift=~m|q([1]?[0-9])|?"d".($1*2+1):"";   }
-
-sub Sbox {
-# input in  lsb > [b0, b1, b2, b3, b4, b5, b6, b7] < msb
-# output in lsb > [b0, b1, b4, b6, b3, b7, b2, b5] < msb
-my @b=@_[0..7];
-my @t=@_[8..11];
-my @s=@_[12..15];
-	&InBasisChange	(@b);
-	&Inv_GF256	(@b[6,5,0,3,7,1,4,2],@t,@s);
-	&OutBasisChange	(@b[7,1,4,2,6,5,0,3]);
-}
-
-sub InBasisChange {
-# input in  lsb > [b0, b1, b2, b3, b4, b5, b6, b7] < msb
-# output in lsb > [b6, b5, b0, b3, b7, b1, b4, b2] < msb 
-my @b=@_[0..7];
-$code.=<<___;
-	veor	@b[2], @b[2], @b[1]
-	veor	@b[5], @b[5], @b[6]
-	veor	@b[3], @b[3], @b[0]
-	veor	@b[6], @b[6], @b[2]
-	veor	@b[5], @b[5], @b[0]
-
-	veor	@b[6], @b[6], @b[3]
-	veor	@b[3], @b[3], @b[7]
-	veor	@b[7], @b[7], @b[5]
-	veor	@b[3], @b[3], @b[4]
-	veor	@b[4], @b[4], @b[5]
-
-	veor	@b[2], @b[2], @b[7]
-	veor	@b[3], @b[3], @b[1]
-	veor	@b[1], @b[1], @b[5]
-___
-}
-
-sub OutBasisChange {
-# input in  lsb > [b0, b1, b2, b3, b4, b5, b6, b7] < msb
-# output in lsb > [b6, b1, b2, b4, b7, b0, b3, b5] < msb
-my @b=@_[0..7];
-$code.=<<___;
-	veor	@b[0], @b[0], @b[6]
-	veor	@b[1], @b[1], @b[4]
-	veor	@b[4], @b[4], @b[6]
-	veor	@b[2], @b[2], @b[0]
-	veor	@b[6], @b[6], @b[1]
-
-	veor	@b[1], @b[1], @b[5]
-	veor	@b[5], @b[5], @b[3]
-	veor	@b[3], @b[3], @b[7]
-	veor	@b[7], @b[7], @b[5]
-	veor	@b[2], @b[2], @b[5]
-
-	veor	@b[4], @b[4], @b[7]
-___
-}
-
-sub InvSbox {
-# input in lsb 	> [b0, b1, b2, b3, b4, b5, b6, b7] < msb
-# output in lsb	> [b0, b1, b6, b4, b2, b7, b3, b5] < msb
-my @b=@_[0..7];
-my @t=@_[8..11];
-my @s=@_[12..15];
-	&InvInBasisChange	(@b);
-	&Inv_GF256		(@b[5,1,2,6,3,7,0,4],@t,@s);
-	&InvOutBasisChange	(@b[3,7,0,4,5,1,2,6]);
-}
-
-sub InvInBasisChange {		# OutBasisChange in reverse (with twist)
-my @b=@_[5,1,2,6,3,7,0,4];
-$code.=<<___
-	 veor	@b[1], @b[1], @b[7]
-	veor	@b[4], @b[4], @b[7]
-
-	veor	@b[7], @b[7], @b[5]
-	 veor	@b[1], @b[1], @b[3]
-	veor	@b[2], @b[2], @b[5]
-	veor	@b[3], @b[3], @b[7]
-
-	veor	@b[6], @b[6], @b[1]
-	veor	@b[2], @b[2], @b[0]
-	 veor	@b[5], @b[5], @b[3]
-	veor	@b[4], @b[4], @b[6]
-	veor	@b[0], @b[0], @b[6]
-	veor	@b[1], @b[1], @b[4]
-___
-}
-
-sub InvOutBasisChange {		# InBasisChange in reverse
-my @b=@_[2,5,7,3,6,1,0,4];
-$code.=<<___;
-	veor	@b[1], @b[1], @b[5]
-	veor	@b[2], @b[2], @b[7]
-
-	veor	@b[3], @b[3], @b[1]
-	veor	@b[4], @b[4], @b[5]
-	veor	@b[7], @b[7], @b[5]
-	veor	@b[3], @b[3], @b[4]
-	 veor 	@b[5], @b[5], @b[0]
-	veor	@b[3], @b[3], @b[7]
-	 veor	@b[6], @b[6], @b[2]
-	 veor	@b[2], @b[2], @b[1]
-	veor	@b[6], @b[6], @b[3]
-
-	veor	@b[3], @b[3], @b[0]
-	veor	@b[5], @b[5], @b[6]
-___
-}
-
-sub Mul_GF4 {
-#;*************************************************************
-#;* Mul_GF4: Input x0-x1,y0-y1 Output x0-x1 Temp t0 (8) *
-#;*************************************************************
-my ($x0,$x1,$y0,$y1,$t0,$t1)=@_;
-$code.=<<___;
-	veor 	$t0, $y0, $y1
-	vand	$t0, $t0, $x0
-	veor	$x0, $x0, $x1
-	vand	$t1, $x1, $y0
-	vand	$x0, $x0, $y1
-	veor	$x1, $t1, $t0
-	veor	$x0, $x0, $t1
-___
-}
-
-sub Mul_GF4_N {				# not used, see next subroutine
-# multiply and scale by N
-my ($x0,$x1,$y0,$y1,$t0)=@_;
-$code.=<<___;
-	veor	$t0, $y0, $y1
-	vand	$t0, $t0, $x0
-	veor	$x0, $x0, $x1
-	vand	$x1, $x1, $y0
-	vand	$x0, $x0, $y1
-	veor	$x1, $x1, $x0
-	veor	$x0, $x0, $t0
-___
-}
-
-sub Mul_GF4_N_GF4 {
-# interleaved Mul_GF4_N and Mul_GF4
-my ($x0,$x1,$y0,$y1,$t0,
-    $x2,$x3,$y2,$y3,$t1)=@_;
-$code.=<<___;
-	veor	$t0, $y0, $y1
-	 veor 	$t1, $y2, $y3
-	vand	$t0, $t0, $x0
-	 vand	$t1, $t1, $x2
-	veor	$x0, $x0, $x1
-	 veor	$x2, $x2, $x3
-	vand	$x1, $x1, $y0
-	 vand	$x3, $x3, $y2
-	vand	$x0, $x0, $y1
-	 vand	$x2, $x2, $y3
-	veor	$x1, $x1, $x0
-	 veor	$x2, $x2, $x3
-	veor	$x0, $x0, $t0
-	 veor	$x3, $x3, $t1
-___
-}
-sub Mul_GF16_2 {
-my @x=@_[0..7];
-my @y=@_[8..11];
-my @t=@_[12..15];
-$code.=<<___;
-	veor	@t[0], @x[0], @x[2]
-	veor	@t[1], @x[1], @x[3]
-___
-	&Mul_GF4  	(@x[0], @x[1], @y[0], @y[1], @t[2..3]);
-$code.=<<___;
-	veor	@y[0], @y[0], @y[2]
-	veor	@y[1], @y[1], @y[3]
-___
-	Mul_GF4_N_GF4	(@t[0], @t[1], @y[0], @y[1], @t[3],
-			 @x[2], @x[3], @y[2], @y[3], @t[2]);
-$code.=<<___;
-	veor	@x[0], @x[0], @t[0]
-	veor	@x[2], @x[2], @t[0]
-	veor	@x[1], @x[1], @t[1]
-	veor	@x[3], @x[3], @t[1]
-
-	veor	@t[0], @x[4], @x[6]
-	veor	@t[1], @x[5], @x[7]
-___
-	&Mul_GF4_N_GF4	(@t[0], @t[1], @y[0], @y[1], @t[3],
-			 @x[6], @x[7], @y[2], @y[3], @t[2]);
-$code.=<<___;
-	veor	@y[0], @y[0], @y[2]
-	veor	@y[1], @y[1], @y[3]
-___
-	&Mul_GF4  	(@x[4], @x[5], @y[0], @y[1], @t[2..3]);
-$code.=<<___;
-	veor	@x[4], @x[4], @t[0]
-	veor	@x[6], @x[6], @t[0]
-	veor	@x[5], @x[5], @t[1]
-	veor	@x[7], @x[7], @t[1]
-___
-}
-sub Inv_GF256 {
-#;********************************************************************
-#;* Inv_GF256: Input x0-x7 Output x0-x7 Temp t0-t3,s0-s3 (144)       *
-#;********************************************************************
-my @x=@_[0..7];
-my @t=@_[8..11];
-my @s=@_[12..15];
-# direct optimizations from hardware
-$code.=<<___;
-	veor	@t[3], @x[4], @x[6]
-	veor	@t[2], @x[5], @x[7]
-	veor	@t[1], @x[1], @x[3]
-	veor	@s[1], @x[7], @x[6]
-	 vmov	@t[0], @t[2]
-	veor	@s[0], @x[0], @x[2]
-
-	vorr	@t[2], @t[2], @t[1]
-	veor	@s[3], @t[3], @t[0]
-	vand	@s[2], @t[3], @s[0]
-	vorr	@t[3], @t[3], @s[0]
-	veor	@s[0], @s[0], @t[1]
-	vand	@t[0], @t[0], @t[1]
-	veor	@t[1], @x[3], @x[2]
-	vand	@s[3], @s[3], @s[0]
-	vand	@s[1], @s[1], @t[1]
-	veor	@t[1], @x[4], @x[5]
-	veor	@s[0], @x[1], @x[0]
-	veor	@t[3], @t[3], @s[1]
-	veor	@t[2], @t[2], @s[1]
-	vand	@s[1], @t[1], @s[0]
-	vorr	@t[1], @t[1], @s[0]
-	veor	@t[3], @t[3], @s[3]
-	veor	@t[0], @t[0], @s[1]
-	veor	@t[2], @t[2], @s[2]
-	veor	@t[1], @t[1], @s[3]
-	veor	@t[0], @t[0], @s[2]
-	vand	@s[0], @x[7], @x[3]
-	veor	@t[1], @t[1], @s[2]
-	vand	@s[1], @x[6], @x[2]
-	vand	@s[2], @x[5], @x[1]
-	vorr	@s[3], @x[4], @x[0]
-	veor	@t[3], @t[3], @s[0]
-	veor	@t[1], @t[1], @s[2]
-	veor	@t[0], @t[0], @s[3]
-	veor	@t[2], @t[2], @s[1]
-
-	@ Inv_GF16 \t0, \t1, \t2, \t3, \s0, \s1, \s2, \s3
-
-	@ new smaller inversion
-
-	vand	@s[2], @t[3], @t[1]
-	vmov	@s[0], @t[0]
-
-	veor	@s[1], @t[2], @s[2]
-	veor	@s[3], @t[0], @s[2]
-	veor	@s[2], @t[0], @s[2]	@ @s[2]=@s[3]
-
-	vbsl	@s[1], @t[1], @t[0]
-	vbsl	@s[3], @t[3], @t[2]
-	veor	@t[3], @t[3], @t[2]
-
-	vbsl	@s[0], @s[1], @s[2]
-	vbsl	@t[0], @s[2], @s[1]
-
-	vand	@s[2], @s[0], @s[3]
-	veor	@t[1], @t[1], @t[0]
-
-	veor	@s[2], @s[2], @t[3]
-___
-# output in s3, s2, s1, t1
-
-# Mul_GF16_2 \x0, \x1, \x2, \x3, \x4, \x5, \x6, \x7, \t2, \t3, \t0, \t1, \s0, \s1, \s2, \s3
-
-# Mul_GF16_2 \x0, \x1, \x2, \x3, \x4, \x5, \x6, \x7, \s3, \s2, \s1, \t1, \s0, \t0, \t2, \t3
-	&Mul_GF16_2(@x,@s[3,2,1],@t[1],@s[0],@t[0,2,3]);
-
-### output msb > [x3,x2,x1,x0,x7,x6,x5,x4] < lsb
-}
-
-# AES linear components
-
-sub ShiftRows {
-my @x=@_[0..7];
-my @t=@_[8..11];
-my $mask=pop;
-$code.=<<___;
-	vldmia	$key!, {@t[0]-@t[3]}
-	veor	@t[0], @t[0], @x[0]
-	veor	@t[1], @t[1], @x[1]
-	vtbl.8	`&Dlo(@x[0])`, {@t[0]}, `&Dlo($mask)`
-	vtbl.8	`&Dhi(@x[0])`, {@t[0]}, `&Dhi($mask)`
-	vldmia	$key!, {@t[0]}
-	veor	@t[2], @t[2], @x[2]
-	vtbl.8	`&Dlo(@x[1])`, {@t[1]}, `&Dlo($mask)`
-	vtbl.8	`&Dhi(@x[1])`, {@t[1]}, `&Dhi($mask)`
-	vldmia	$key!, {@t[1]}
-	veor	@t[3], @t[3], @x[3]
-	vtbl.8	`&Dlo(@x[2])`, {@t[2]}, `&Dlo($mask)`
-	vtbl.8	`&Dhi(@x[2])`, {@t[2]}, `&Dhi($mask)`
-	vldmia	$key!, {@t[2]}
-	vtbl.8	`&Dlo(@x[3])`, {@t[3]}, `&Dlo($mask)`
-	vtbl.8	`&Dhi(@x[3])`, {@t[3]}, `&Dhi($mask)`
-	vldmia	$key!, {@t[3]}
-	veor	@t[0], @t[0], @x[4]
-	veor	@t[1], @t[1], @x[5]
-	vtbl.8	`&Dlo(@x[4])`, {@t[0]}, `&Dlo($mask)`
-	vtbl.8	`&Dhi(@x[4])`, {@t[0]}, `&Dhi($mask)`
-	veor	@t[2], @t[2], @x[6]
-	vtbl.8	`&Dlo(@x[5])`, {@t[1]}, `&Dlo($mask)`
-	vtbl.8	`&Dhi(@x[5])`, {@t[1]}, `&Dhi($mask)`
-	veor	@t[3], @t[3], @x[7]
-	vtbl.8	`&Dlo(@x[6])`, {@t[2]}, `&Dlo($mask)`
-	vtbl.8	`&Dhi(@x[6])`, {@t[2]}, `&Dhi($mask)`
-	vtbl.8	`&Dlo(@x[7])`, {@t[3]}, `&Dlo($mask)`
-	vtbl.8	`&Dhi(@x[7])`, {@t[3]}, `&Dhi($mask)`
-___
-}
-
-sub MixColumns {
-# modified to emit output in order suitable for feeding back to aesenc[last]
-my @x=@_[0..7];
-my @t=@_[8..15];
-my $inv=@_[16];	# optional
-$code.=<<___;
-	vext.8	@t[0], @x[0], @x[0], #12	@ x0 <<< 32
-	vext.8	@t[1], @x[1], @x[1], #12
-	 veor	@x[0], @x[0], @t[0]		@ x0 ^ (x0 <<< 32)
-	vext.8	@t[2], @x[2], @x[2], #12
-	 veor	@x[1], @x[1], @t[1]
-	vext.8	@t[3], @x[3], @x[3], #12
-	 veor	@x[2], @x[2], @t[2]
-	vext.8	@t[4], @x[4], @x[4], #12
-	 veor	@x[3], @x[3], @t[3]
-	vext.8	@t[5], @x[5], @x[5], #12
-	 veor	@x[4], @x[4], @t[4]
-	vext.8	@t[6], @x[6], @x[6], #12
-	 veor	@x[5], @x[5], @t[5]
-	vext.8	@t[7], @x[7], @x[7], #12
-	 veor	@x[6], @x[6], @t[6]
-
-	veor	@t[1], @t[1], @x[0]
-	 veor	@x[7], @x[7], @t[7]
-	 vext.8	@x[0], @x[0], @x[0], #8		@ (x0 ^ (x0 <<< 32)) <<< 64)
-	veor	@t[2], @t[2], @x[1]
-	veor	@t[0], @t[0], @x[7]
-	veor	@t[1], @t[1], @x[7]
-	 vext.8	@x[1], @x[1], @x[1], #8
-	veor	@t[5], @t[5], @x[4]
-	 veor	@x[0], @x[0], @t[0]
-	veor	@t[6], @t[6], @x[5]
-	 veor	@x[1], @x[1], @t[1]
-	 vext.8	@t[0], @x[4], @x[4], #8
-	veor	@t[4], @t[4], @x[3]
-	 vext.8	@t[1], @x[5], @x[5], #8
-	veor	@t[7], @t[7], @x[6]
-	 vext.8	@x[4], @x[3], @x[3], #8
-	veor	@t[3], @t[3], @x[2]
-	 vext.8	@x[5], @x[7], @x[7], #8
-	veor	@t[4], @t[4], @x[7]
-	 vext.8	@x[3], @x[6], @x[6], #8
-	veor	@t[3], @t[3], @x[7]
-	 vext.8	@x[6], @x[2], @x[2], #8
-	veor	@x[7], @t[1], @t[5]
-___
-$code.=<<___ if (!$inv);
-	veor	@x[2], @t[0], @t[4]
-	veor	@x[4], @x[4], @t[3]
-	veor	@x[5], @x[5], @t[7]
-	veor	@x[3], @x[3], @t[6]
-	 @ vmov	@x[2], @t[0]
-	veor	@x[6], @x[6], @t[2]
-	 @ vmov	@x[7], @t[1]
-___
-$code.=<<___ if ($inv);
-	veor	@t[3], @t[3], @x[4]
-	veor	@x[5], @x[5], @t[7]
-	veor	@x[2], @x[3], @t[6]
-	veor	@x[3], @t[0], @t[4]
-	veor	@x[4], @x[6], @t[2]
-	vmov	@x[6], @t[3]
-	 @ vmov	@x[7], @t[1]
-___
-}
-
-sub InvMixColumns_orig {
-my @x=@_[0..7];
-my @t=@_[8..15];
-
-$code.=<<___;
-	@ multiplication by 0x0e
-	vext.8	@t[7], @x[7], @x[7], #12
-	vmov	@t[2], @x[2]
-	veor	@x[2], @x[2], @x[5]		@ 2 5
-	veor	@x[7], @x[7], @x[5]		@ 7 5
-	vext.8	@t[0], @x[0], @x[0], #12
-	vmov	@t[5], @x[5]
-	veor	@x[5], @x[5], @x[0]		@ 5 0		[1]
-	veor	@x[0], @x[0], @x[1]		@ 0 1
-	vext.8	@t[1], @x[1], @x[1], #12
-	veor	@x[1], @x[1], @x[2]		@ 1 25
-	veor	@x[0], @x[0], @x[6]		@ 01 6		[2]
-	vext.8	@t[3], @x[3], @x[3], #12
-	veor	@x[1], @x[1], @x[3]		@ 125 3		[4]
-	veor	@x[2], @x[2], @x[0]		@ 25 016	[3]
-	veor	@x[3], @x[3], @x[7]		@ 3 75
-	veor	@x[7], @x[7], @x[6]		@ 75 6		[0]
-	vext.8	@t[6], @x[6], @x[6], #12
-	vmov	@t[4], @x[4]
-	veor	@x[6], @x[6], @x[4]		@ 6 4
-	veor	@x[4], @x[4], @x[3]		@ 4 375		[6]
-	veor	@x[3], @x[3], @x[7]		@ 375 756=36
-	veor	@x[6], @x[6], @t[5]		@ 64 5		[7]
-	veor	@x[3], @x[3], @t[2]		@ 36 2
-	vext.8	@t[5], @t[5], @t[5], #12
-	veor	@x[3], @x[3], @t[4]		@ 362 4		[5]
-___
-					my @y = @x[7,5,0,2,1,3,4,6];
-$code.=<<___;
-	@ multiplication by 0x0b
-	veor	@y[1], @y[1], @y[0]
-	veor	@y[0], @y[0], @t[0]
-	vext.8	@t[2], @t[2], @t[2], #12
-	veor	@y[1], @y[1], @t[1]
-	veor	@y[0], @y[0], @t[5]
-	vext.8	@t[4], @t[4], @t[4], #12
-	veor	@y[1], @y[1], @t[6]
-	veor	@y[0], @y[0], @t[7]
-	veor	@t[7], @t[7], @t[6]		@ clobber t[7]
-
-	veor	@y[3], @y[3], @t[0]
-	 veor	@y[1], @y[1], @y[0]
-	vext.8	@t[0], @t[0], @t[0], #12
-	veor	@y[2], @y[2], @t[1]
-	veor	@y[4], @y[4], @t[1]
-	vext.8	@t[1], @t[1], @t[1], #12
-	veor	@y[2], @y[2], @t[2]
-	veor	@y[3], @y[3], @t[2]
-	veor	@y[5], @y[5], @t[2]
-	veor	@y[2], @y[2], @t[7]
-	vext.8	@t[2], @t[2], @t[2], #12
-	veor	@y[3], @y[3], @t[3]
-	veor	@y[6], @y[6], @t[3]
-	veor	@y[4], @y[4], @t[3]
-	veor	@y[7], @y[7], @t[4]
-	vext.8	@t[3], @t[3], @t[3], #12
-	veor	@y[5], @y[5], @t[4]
-	veor	@y[7], @y[7], @t[7]
-	veor	@t[7], @t[7], @t[5]		@ clobber t[7] even more
-	veor	@y[3], @y[3], @t[5]
-	veor	@y[4], @y[4], @t[4]
-
-	veor	@y[5], @y[5], @t[7]
-	vext.8	@t[4], @t[4], @t[4], #12
-	veor	@y[6], @y[6], @t[7]
-	veor	@y[4], @y[4], @t[7]
-
-	veor	@t[7], @t[7], @t[5]
-	vext.8	@t[5], @t[5], @t[5], #12
-
-	@ multiplication by 0x0d
-	veor	@y[4], @y[4], @y[7]
-	 veor	@t[7], @t[7], @t[6]		@ restore t[7]
-	veor	@y[7], @y[7], @t[4]
-	vext.8	@t[6], @t[6], @t[6], #12
-	veor	@y[2], @y[2], @t[0]
-	veor	@y[7], @y[7], @t[5]
-	vext.8	@t[7], @t[7], @t[7], #12
-	veor	@y[2], @y[2], @t[2]
-
-	veor	@y[3], @y[3], @y[1]
-	veor	@y[1], @y[1], @t[1]
-	veor	@y[0], @y[0], @t[0]
-	veor	@y[3], @y[3], @t[0]
-	veor	@y[1], @y[1], @t[5]
-	veor	@y[0], @y[0], @t[5]
-	vext.8	@t[0], @t[0], @t[0], #12
-	veor	@y[1], @y[1], @t[7]
-	veor	@y[0], @y[0], @t[6]
-	veor	@y[3], @y[3], @y[1]
-	veor	@y[4], @y[4], @t[1]
-	vext.8	@t[1], @t[1], @t[1], #12
-
-	veor	@y[7], @y[7], @t[7]
-	veor	@y[4], @y[4], @t[2]
-	veor	@y[5], @y[5], @t[2]
-	veor	@y[2], @y[2], @t[6]
-	veor	@t[6], @t[6], @t[3]		@ clobber t[6]
-	vext.8	@t[2], @t[2], @t[2], #12
-	veor	@y[4], @y[4], @y[7]
-	veor	@y[3], @y[3], @t[6]
-
-	veor	@y[6], @y[6], @t[6]
-	veor	@y[5], @y[5], @t[5]
-	vext.8	@t[5], @t[5], @t[5], #12
-	veor	@y[6], @y[6], @t[4]
-	vext.8	@t[4], @t[4], @t[4], #12
-	veor	@y[5], @y[5], @t[6]
-	veor	@y[6], @y[6], @t[7]
-	vext.8	@t[7], @t[7], @t[7], #12
-	veor	@t[6], @t[6], @t[3]		@ restore t[6]
-	vext.8	@t[3], @t[3], @t[3], #12
-
-	@ multiplication by 0x09
-	veor	@y[4], @y[4], @y[1]
-	veor	@t[1], @t[1], @y[1]		@ t[1]=y[1]
-	veor	@t[0], @t[0], @t[5]		@ clobber t[0]
-	vext.8	@t[6], @t[6], @t[6], #12
-	veor	@t[1], @t[1], @t[5]
-	veor	@y[3], @y[3], @t[0]
-	veor	@t[0], @t[0], @y[0]		@ t[0]=y[0]
-	veor	@t[1], @t[1], @t[6]
-	veor	@t[6], @t[6], @t[7]		@ clobber t[6]
-	veor	@y[4], @y[4], @t[1]
-	veor	@y[7], @y[7], @t[4]
-	veor	@y[6], @y[6], @t[3]
-	veor	@y[5], @y[5], @t[2]
-	veor	@t[4], @t[4], @y[4]		@ t[4]=y[4]
-	veor	@t[3], @t[3], @y[3]		@ t[3]=y[3]
-	veor	@t[5], @t[5], @y[5]		@ t[5]=y[5]
-	veor	@t[2], @t[2], @y[2]		@ t[2]=y[2]
-	veor	@t[3], @t[3], @t[7]
-	veor	@XMM[5], @t[5], @t[6]
-	veor	@XMM[6], @t[6], @y[6]		@ t[6]=y[6]
-	veor	@XMM[2], @t[2], @t[6]
-	veor	@XMM[7], @t[7], @y[7]		@ t[7]=y[7]
-
-	vmov	@XMM[0], @t[0]
-	vmov	@XMM[1], @t[1]
-	@ vmov	@XMM[2], @t[2]
-	vmov	@XMM[3], @t[3]
-	vmov	@XMM[4], @t[4]
-	@ vmov	@XMM[5], @t[5]
-	@ vmov	@XMM[6], @t[6]
-	@ vmov	@XMM[7], @t[7]
-___
-}
-
-sub InvMixColumns {
-my @x=@_[0..7];
-my @t=@_[8..15];
-
-# Thanks to Jussi Kivilinna for providing pointer to
-#
-# | 0e 0b 0d 09 |   | 02 03 01 01 |   | 05 00 04 00 |
-# | 09 0e 0b 0d | = | 01 02 03 01 | x | 00 05 00 04 |
-# | 0d 09 0e 0b |   | 01 01 02 03 |   | 04 00 05 00 |
-# | 0b 0d 09 0e |   | 03 01 01 02 |   | 00 04 00 05 |
-
-$code.=<<___;
-	@ multiplication by 0x05-0x00-0x04-0x00
-	vext.8	@t[0], @x[0], @x[0], #8
-	vext.8	@t[6], @x[6], @x[6], #8
-	vext.8	@t[7], @x[7], @x[7], #8
-	veor	@t[0], @t[0], @x[0]
-	vext.8	@t[1], @x[1], @x[1], #8
-	veor	@t[6], @t[6], @x[6]
-	vext.8	@t[2], @x[2], @x[2], #8
-	veor	@t[7], @t[7], @x[7]
-	vext.8	@t[3], @x[3], @x[3], #8
-	veor	@t[1], @t[1], @x[1]
-	vext.8	@t[4], @x[4], @x[4], #8
-	veor	@t[2], @t[2], @x[2]
-	vext.8	@t[5], @x[5], @x[5], #8
-	veor	@t[3], @t[3], @x[3]
-	veor	@t[4], @t[4], @x[4]
-	veor	@t[5], @t[5], @x[5]
-
-	 veor	@x[0], @x[0], @t[6]
-	 veor	@x[1], @x[1], @t[6]
-	 veor	@x[2], @x[2], @t[0]
-	 veor	@x[4], @x[4], @t[2]
-	 veor	@x[3], @x[3], @t[1]
-	 veor	@x[1], @x[1], @t[7]
-	 veor	@x[2], @x[2], @t[7]
-	 veor	@x[4], @x[4], @t[6]
-	 veor	@x[5], @x[5], @t[3]
-	 veor	@x[3], @x[3], @t[6]
-	 veor	@x[6], @x[6], @t[4]
-	 veor	@x[4], @x[4], @t[7]
-	 veor	@x[5], @x[5], @t[7]
-	 veor	@x[7], @x[7], @t[5]
-___
-	&MixColumns	(@x,@t,1);	# flipped 2<->3 and 4<->6
-}
-
-sub swapmove {
-my ($a,$b,$n,$mask,$t)=@_;
-$code.=<<___;
-	vshr.u64	$t, $b, #$n
-	veor		$t, $t, $a
-	vand		$t, $t, $mask
-	veor		$a, $a, $t
-	vshl.u64	$t, $t, #$n
-	veor		$b, $b, $t
-___
-}
-sub swapmove2x {
-my ($a0,$b0,$a1,$b1,$n,$mask,$t0,$t1)=@_;
-$code.=<<___;
-	vshr.u64	$t0, $b0, #$n
-	 vshr.u64	$t1, $b1, #$n
-	veor		$t0, $t0, $a0
-	 veor		$t1, $t1, $a1
-	vand		$t0, $t0, $mask
-	 vand		$t1, $t1, $mask
-	veor		$a0, $a0, $t0
-	vshl.u64	$t0, $t0, #$n
-	 veor		$a1, $a1, $t1
-	 vshl.u64	$t1, $t1, #$n
-	veor		$b0, $b0, $t0
-	 veor		$b1, $b1, $t1
-___
-}
-
-sub bitslice {
-my @x=reverse(@_[0..7]);
-my ($t0,$t1,$t2,$t3)=@_[8..11];
-$code.=<<___;
-	vmov.i8	$t0,#0x55			@ compose .LBS0
-	vmov.i8	$t1,#0x33			@ compose .LBS1
-___
-	&swapmove2x(@x[0,1,2,3],1,$t0,$t2,$t3);
-	&swapmove2x(@x[4,5,6,7],1,$t0,$t2,$t3);
-$code.=<<___;
-	vmov.i8	$t0,#0x0f			@ compose .LBS2
-___
-	&swapmove2x(@x[0,2,1,3],2,$t1,$t2,$t3);
-	&swapmove2x(@x[4,6,5,7],2,$t1,$t2,$t3);
-
-	&swapmove2x(@x[0,4,1,5],4,$t0,$t2,$t3);
-	&swapmove2x(@x[2,6,3,7],4,$t0,$t2,$t3);
-}
-
-$code.=<<___;
-#ifndef __KERNEL__
-# include "arm_arch.h"
-
-# define VFP_ABI_PUSH	vstmdb	sp!,{d8-d15}
-# define VFP_ABI_POP	vldmia	sp!,{d8-d15}
-# define VFP_ABI_FRAME	0x40
-#else
-# define VFP_ABI_PUSH
-# define VFP_ABI_POP
-# define VFP_ABI_FRAME	0
-# define BSAES_ASM_EXTENDED_KEY
-# define XTS_CHAIN_TWEAK
-# define __ARM_ARCH__ __LINUX_ARM_ARCH__
-# define __ARM_MAX_ARCH__ 7
-#endif
-
-#ifdef __thumb__
-# define adrl adr
-#endif
-
-#if __ARM_MAX_ARCH__>=7
-.arch	armv7-a
-.fpu	neon
-
-.text
-.syntax	unified 	@ ARMv7-capable assembler is expected to handle this
-#ifdef __thumb2__
-.thumb
-#else
-.code   32
-#endif
-
-.type	_bsaes_decrypt8,%function
-.align	4
-_bsaes_decrypt8:
-	adr	$const,_bsaes_decrypt8
-	vldmia	$key!, {@XMM[9]}		@ round 0 key
-	add	$const,$const,#.LM0ISR-_bsaes_decrypt8
-
-	vldmia	$const!, {@XMM[8]}		@ .LM0ISR
-	veor	@XMM[10], @XMM[0], @XMM[9]	@ xor with round0 key
-	veor	@XMM[11], @XMM[1], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[0])`, {@XMM[10]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[0])`, {@XMM[10]}, `&Dhi(@XMM[8])`
-	veor	@XMM[12], @XMM[2], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[1])`, {@XMM[11]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[1])`, {@XMM[11]}, `&Dhi(@XMM[8])`
-	veor	@XMM[13], @XMM[3], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[2])`, {@XMM[12]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[2])`, {@XMM[12]}, `&Dhi(@XMM[8])`
-	veor	@XMM[14], @XMM[4], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[3])`, {@XMM[13]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[3])`, {@XMM[13]}, `&Dhi(@XMM[8])`
-	veor	@XMM[15], @XMM[5], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[4])`, {@XMM[14]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[4])`, {@XMM[14]}, `&Dhi(@XMM[8])`
-	veor	@XMM[10], @XMM[6], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[5])`, {@XMM[15]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[5])`, {@XMM[15]}, `&Dhi(@XMM[8])`
-	veor	@XMM[11], @XMM[7], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[6])`, {@XMM[10]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[6])`, {@XMM[10]}, `&Dhi(@XMM[8])`
-	 vtbl.8	`&Dlo(@XMM[7])`, {@XMM[11]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[7])`, {@XMM[11]}, `&Dhi(@XMM[8])`
-___
-	&bitslice	(@XMM[0..7, 8..11]);
-$code.=<<___;
-	sub	$rounds,$rounds,#1
-	b	.Ldec_sbox
-.align	4
-.Ldec_loop:
-___
-	&ShiftRows	(@XMM[0..7, 8..12]);
-$code.=".Ldec_sbox:\n";
-	&InvSbox	(@XMM[0..7, 8..15]);
-$code.=<<___;
-	subs	$rounds,$rounds,#1
-	bcc	.Ldec_done
-___
-	&InvMixColumns	(@XMM[0,1,6,4,2,7,3,5, 8..15]);
-$code.=<<___;
-	vldmia	$const, {@XMM[12]}		@ .LISR
-	ite	eq				@ Thumb2 thing, sanity check in ARM
-	addeq	$const,$const,#0x10
-	bne	.Ldec_loop
-	vldmia	$const, {@XMM[12]}		@ .LISRM0
-	b	.Ldec_loop
-.align	4
-.Ldec_done:
-___
-	&bitslice	(@XMM[0,1,6,4,2,7,3,5, 8..11]);
-$code.=<<___;
-	vldmia	$key, {@XMM[8]}			@ last round key
-	veor	@XMM[6], @XMM[6], @XMM[8]
-	veor	@XMM[4], @XMM[4], @XMM[8]
-	veor	@XMM[2], @XMM[2], @XMM[8]
-	veor	@XMM[7], @XMM[7], @XMM[8]
-	veor	@XMM[3], @XMM[3], @XMM[8]
-	veor	@XMM[5], @XMM[5], @XMM[8]
-	veor	@XMM[0], @XMM[0], @XMM[8]
-	veor	@XMM[1], @XMM[1], @XMM[8]
-	bx	lr
-.size	_bsaes_decrypt8,.-_bsaes_decrypt8
-
-.type	_bsaes_const,%object
-.align	6
-_bsaes_const:
-.LM0ISR:	@ InvShiftRows constants
-	.quad	0x0a0e0206070b0f03, 0x0004080c0d010509
-.LISR:
-	.quad	0x0504070602010003, 0x0f0e0d0c080b0a09
-.LISRM0:
-	.quad	0x01040b0e0205080f, 0x0306090c00070a0d
-.LM0SR:		@ ShiftRows constants
-	.quad	0x0a0e02060f03070b, 0x0004080c05090d01
-.LSR:
-	.quad	0x0504070600030201, 0x0f0e0d0c0a09080b
-.LSRM0:
-	.quad	0x0304090e00050a0f, 0x01060b0c0207080d
-.LM0:
-	.quad	0x02060a0e03070b0f, 0x0004080c0105090d
-.LREVM0SR:
-	.quad	0x090d01050c000408, 0x03070b0f060a0e02
-.asciz	"Bit-sliced AES for NEON, CRYPTOGAMS by <appro\@openssl.org>"
-.align	6
-.size	_bsaes_const,.-_bsaes_const
-
-.type	_bsaes_encrypt8,%function
-.align	4
-_bsaes_encrypt8:
-	adr	$const,_bsaes_encrypt8
-	vldmia	$key!, {@XMM[9]}		@ round 0 key
-	sub	$const,$const,#_bsaes_encrypt8-.LM0SR
-
-	vldmia	$const!, {@XMM[8]}		@ .LM0SR
-_bsaes_encrypt8_alt:
-	veor	@XMM[10], @XMM[0], @XMM[9]	@ xor with round0 key
-	veor	@XMM[11], @XMM[1], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[0])`, {@XMM[10]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[0])`, {@XMM[10]}, `&Dhi(@XMM[8])`
-	veor	@XMM[12], @XMM[2], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[1])`, {@XMM[11]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[1])`, {@XMM[11]}, `&Dhi(@XMM[8])`
-	veor	@XMM[13], @XMM[3], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[2])`, {@XMM[12]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[2])`, {@XMM[12]}, `&Dhi(@XMM[8])`
-	veor	@XMM[14], @XMM[4], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[3])`, {@XMM[13]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[3])`, {@XMM[13]}, `&Dhi(@XMM[8])`
-	veor	@XMM[15], @XMM[5], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[4])`, {@XMM[14]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[4])`, {@XMM[14]}, `&Dhi(@XMM[8])`
-	veor	@XMM[10], @XMM[6], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[5])`, {@XMM[15]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[5])`, {@XMM[15]}, `&Dhi(@XMM[8])`
-	veor	@XMM[11], @XMM[7], @XMM[9]
-	 vtbl.8	`&Dlo(@XMM[6])`, {@XMM[10]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[6])`, {@XMM[10]}, `&Dhi(@XMM[8])`
-	 vtbl.8	`&Dlo(@XMM[7])`, {@XMM[11]}, `&Dlo(@XMM[8])`
-	 vtbl.8	`&Dhi(@XMM[7])`, {@XMM[11]}, `&Dhi(@XMM[8])`
-_bsaes_encrypt8_bitslice:
-___
-	&bitslice	(@XMM[0..7, 8..11]);
-$code.=<<___;
-	sub	$rounds,$rounds,#1
-	b	.Lenc_sbox
-.align	4
-.Lenc_loop:
-___
-	&ShiftRows	(@XMM[0..7, 8..12]);
-$code.=".Lenc_sbox:\n";
-	&Sbox		(@XMM[0..7, 8..15]);
-$code.=<<___;
-	subs	$rounds,$rounds,#1
-	bcc	.Lenc_done
-___
-	&MixColumns	(@XMM[0,1,4,6,3,7,2,5, 8..15]);
-$code.=<<___;
-	vldmia	$const, {@XMM[12]}		@ .LSR
-	ite	eq				@ Thumb2 thing, samity check in ARM
-	addeq	$const,$const,#0x10
-	bne	.Lenc_loop
-	vldmia	$const, {@XMM[12]}		@ .LSRM0
-	b	.Lenc_loop
-.align	4
-.Lenc_done:
-___
-	# output in lsb > [t0, t1, t4, t6, t3, t7, t2, t5] < msb
-	&bitslice	(@XMM[0,1,4,6,3,7,2,5, 8..11]);
-$code.=<<___;
-	vldmia	$key, {@XMM[8]}			@ last round key
-	veor	@XMM[4], @XMM[4], @XMM[8]
-	veor	@XMM[6], @XMM[6], @XMM[8]
-	veor	@XMM[3], @XMM[3], @XMM[8]
-	veor	@XMM[7], @XMM[7], @XMM[8]
-	veor	@XMM[2], @XMM[2], @XMM[8]
-	veor	@XMM[5], @XMM[5], @XMM[8]
-	veor	@XMM[0], @XMM[0], @XMM[8]
-	veor	@XMM[1], @XMM[1], @XMM[8]
-	bx	lr
-.size	_bsaes_encrypt8,.-_bsaes_encrypt8
-___
-}
-{
-my ($out,$inp,$rounds,$const)=("r12","r4","r5","r6");
-
-sub bitslice_key {
-my @x=reverse(@_[0..7]);
-my ($bs0,$bs1,$bs2,$t2,$t3)=@_[8..12];
-
-	&swapmove	(@x[0,1],1,$bs0,$t2,$t3);
-$code.=<<___;
-	@ &swapmove(@x[2,3],1,$t0,$t2,$t3);
-	vmov	@x[2], @x[0]
-	vmov	@x[3], @x[1]
-___
-	#&swapmove2x(@x[4,5,6,7],1,$t0,$t2,$t3);
-
-	&swapmove2x	(@x[0,2,1,3],2,$bs1,$t2,$t3);
-$code.=<<___;
-	@ &swapmove2x(@x[4,6,5,7],2,$t1,$t2,$t3);
-	vmov	@x[4], @x[0]
-	vmov	@x[6], @x[2]
-	vmov	@x[5], @x[1]
-	vmov	@x[7], @x[3]
-___
-	&swapmove2x	(@x[0,4,1,5],4,$bs2,$t2,$t3);
-	&swapmove2x	(@x[2,6,3,7],4,$bs2,$t2,$t3);
-}
-
-$code.=<<___;
-.type	_bsaes_key_convert,%function
-.align	4
-_bsaes_key_convert:
-	adr	$const,_bsaes_key_convert
-	vld1.8	{@XMM[7]},  [$inp]!		@ load round 0 key
-	sub	$const,$const,#_bsaes_key_convert-.LM0
-	vld1.8	{@XMM[15]}, [$inp]!		@ load round 1 key
-
-	vmov.i8	@XMM[8],  #0x01			@ bit masks
-	vmov.i8	@XMM[9],  #0x02
-	vmov.i8	@XMM[10], #0x04
-	vmov.i8	@XMM[11], #0x08
-	vmov.i8	@XMM[12], #0x10
-	vmov.i8	@XMM[13], #0x20
-	vldmia	$const, {@XMM[14]}		@ .LM0
-
-#ifdef __ARMEL__
-	vrev32.8	@XMM[7],  @XMM[7]
-	vrev32.8	@XMM[15], @XMM[15]
-#endif
-	sub	$rounds,$rounds,#1
-	vstmia	$out!, {@XMM[7]}		@ save round 0 key
-	b	.Lkey_loop
-
-.align	4
-.Lkey_loop:
-	vtbl.8	`&Dlo(@XMM[7])`,{@XMM[15]},`&Dlo(@XMM[14])`
-	vtbl.8	`&Dhi(@XMM[7])`,{@XMM[15]},`&Dhi(@XMM[14])`
-	vmov.i8	@XMM[6],  #0x40
-	vmov.i8	@XMM[15], #0x80
-
-	vtst.8	@XMM[0], @XMM[7], @XMM[8]
-	vtst.8	@XMM[1], @XMM[7], @XMM[9]
-	vtst.8	@XMM[2], @XMM[7], @XMM[10]
-	vtst.8	@XMM[3], @XMM[7], @XMM[11]
-	vtst.8	@XMM[4], @XMM[7], @XMM[12]
-	vtst.8	@XMM[5], @XMM[7], @XMM[13]
-	vtst.8	@XMM[6], @XMM[7], @XMM[6]
-	vtst.8	@XMM[7], @XMM[7], @XMM[15]
-	vld1.8	{@XMM[15]}, [$inp]!		@ load next round key
-	vmvn	@XMM[0], @XMM[0]		@ "pnot"
-	vmvn	@XMM[1], @XMM[1]
-	vmvn	@XMM[5], @XMM[5]
-	vmvn	@XMM[6], @XMM[6]
-#ifdef __ARMEL__
-	vrev32.8	@XMM[15], @XMM[15]
-#endif
-	subs	$rounds,$rounds,#1
-	vstmia	$out!,{@XMM[0]-@XMM[7]}		@ write bit-sliced round key
-	bne	.Lkey_loop
-
-	vmov.i8	@XMM[7],#0x63			@ compose .L63
-	@ don't save last round key
-	bx	lr
-.size	_bsaes_key_convert,.-_bsaes_key_convert
-___
-}
-
-if (0) {		# following four functions are unsupported interface
-			# used for benchmarking...
-$code.=<<___;
-.globl	bsaes_enc_key_convert
-.type	bsaes_enc_key_convert,%function
-.align	4
-bsaes_enc_key_convert:
-	stmdb	sp!,{r4-r6,lr}
-	vstmdb	sp!,{d8-d15}		@ ABI specification says so
-
-	ldr	r5,[$inp,#240]			@ pass rounds
-	mov	r4,$inp				@ pass key
-	mov	r12,$out			@ pass key schedule
-	bl	_bsaes_key_convert
-	veor	@XMM[7],@XMM[7],@XMM[15]	@ fix up last round key
-	vstmia	r12, {@XMM[7]}			@ save last round key
-
-	vldmia	sp!,{d8-d15}
-	ldmia	sp!,{r4-r6,pc}
-.size	bsaes_enc_key_convert,.-bsaes_enc_key_convert
-
-.globl	bsaes_encrypt_128
-.type	bsaes_encrypt_128,%function
-.align	4
-bsaes_encrypt_128:
-	stmdb	sp!,{r4-r6,lr}
-	vstmdb	sp!,{d8-d15}		@ ABI specification says so
-.Lenc128_loop:
-	vld1.8	{@XMM[0]-@XMM[1]}, [$inp]!	@ load input
-	vld1.8	{@XMM[2]-@XMM[3]}, [$inp]!
-	mov	r4,$key				@ pass the key
-	vld1.8	{@XMM[4]-@XMM[5]}, [$inp]!
-	mov	r5,#10				@ pass rounds
-	vld1.8	{@XMM[6]-@XMM[7]}, [$inp]!
-
-	bl	_bsaes_encrypt8
-
-	vst1.8	{@XMM[0]-@XMM[1]}, [$out]!	@ write output
-	vst1.8	{@XMM[4]}, [$out]!
-	vst1.8	{@XMM[6]}, [$out]!
-	vst1.8	{@XMM[3]}, [$out]!
-	vst1.8	{@XMM[7]}, [$out]!
-	vst1.8	{@XMM[2]}, [$out]!
-	subs	$len,$len,#0x80
-	vst1.8	{@XMM[5]}, [$out]!
-	bhi	.Lenc128_loop
-
-	vldmia	sp!,{d8-d15}
-	ldmia	sp!,{r4-r6,pc}
-.size	bsaes_encrypt_128,.-bsaes_encrypt_128
-
-.globl	bsaes_dec_key_convert
-.type	bsaes_dec_key_convert,%function
-.align	4
-bsaes_dec_key_convert:
-	stmdb	sp!,{r4-r6,lr}
-	vstmdb	sp!,{d8-d15}		@ ABI specification says so
-
-	ldr	r5,[$inp,#240]			@ pass rounds
-	mov	r4,$inp				@ pass key
-	mov	r12,$out			@ pass key schedule
-	bl	_bsaes_key_convert
-	vldmia	$out, {@XMM[6]}
-	vstmia	r12,  {@XMM[15]}		@ save last round key
-	veor	@XMM[7], @XMM[7], @XMM[6]	@ fix up round 0 key
-	vstmia	$out, {@XMM[7]}
-
-	vldmia	sp!,{d8-d15}
-	ldmia	sp!,{r4-r6,pc}
-.size	bsaes_dec_key_convert,.-bsaes_dec_key_convert
-
-.globl	bsaes_decrypt_128
-.type	bsaes_decrypt_128,%function
-.align	4
-bsaes_decrypt_128:
-	stmdb	sp!,{r4-r6,lr}
-	vstmdb	sp!,{d8-d15}		@ ABI specification says so
-.Ldec128_loop:
-	vld1.8	{@XMM[0]-@XMM[1]}, [$inp]!	@ load input
-	vld1.8	{@XMM[2]-@XMM[3]}, [$inp]!
-	mov	r4,$key				@ pass the key
-	vld1.8	{@XMM[4]-@XMM[5]}, [$inp]!
-	mov	r5,#10				@ pass rounds
-	vld1.8	{@XMM[6]-@XMM[7]}, [$inp]!
-
-	bl	_bsaes_decrypt8
-
-	vst1.8	{@XMM[0]-@XMM[1]}, [$out]!	@ write output
-	vst1.8	{@XMM[6]}, [$out]!
-	vst1.8	{@XMM[4]}, [$out]!
-	vst1.8	{@XMM[2]}, [$out]!
-	vst1.8	{@XMM[7]}, [$out]!
-	vst1.8	{@XMM[3]}, [$out]!
-	subs	$len,$len,#0x80
-	vst1.8	{@XMM[5]}, [$out]!
-	bhi	.Ldec128_loop
-
-	vldmia	sp!,{d8-d15}
-	ldmia	sp!,{r4-r6,pc}
-.size	bsaes_decrypt_128,.-bsaes_decrypt_128
-___
-}
-{
-my ($inp,$out,$len,$key, $ivp,$fp,$rounds)=map("r$_",(0..3,8..10));
-my ($keysched)=("sp");
-
-$code.=<<___;
-.extern AES_cbc_encrypt
-.extern AES_decrypt
-
-.global	bsaes_cbc_encrypt
-.type	bsaes_cbc_encrypt,%function
-.align	5
-bsaes_cbc_encrypt:
-#ifndef	__KERNEL__
-	cmp	$len, #128
-#ifndef	__thumb__
-	blo	AES_cbc_encrypt
-#else
-	bhs	1f
-	b	AES_cbc_encrypt
-1:
-#endif
-#endif
-
-	@ it is up to the caller to make sure we are called with enc == 0
-
-	mov	ip, sp
-	stmdb	sp!, {r4-r10, lr}
-	VFP_ABI_PUSH
-	ldr	$ivp, [ip]			@ IV is 1st arg on the stack
-	mov	$len, $len, lsr#4		@ len in 16 byte blocks
-	sub	sp, #0x10			@ scratch space to carry over the IV
-	mov	$fp, sp				@ save sp
-
-	ldr	$rounds, [$key, #240]		@ get # of rounds
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	@ allocate the key schedule on the stack
-	sub	r12, sp, $rounds, lsl#7		@ 128 bytes per inner round key
-	add	r12, #`128-32`			@ sifze of bit-slices key schedule
-
-	@ populate the key schedule
-	mov	r4, $key			@ pass key
-	mov	r5, $rounds			@ pass # of rounds
-	mov	sp, r12				@ sp is $keysched
-	bl	_bsaes_key_convert
-	vldmia	$keysched, {@XMM[6]}
-	vstmia	r12,  {@XMM[15]}		@ save last round key
-	veor	@XMM[7], @XMM[7], @XMM[6]	@ fix up round 0 key
-	vstmia	$keysched, {@XMM[7]}
-#else
-	ldr	r12, [$key, #244]
-	eors	r12, #1
-	beq	0f
-
-	@ populate the key schedule
-	str	r12, [$key, #244]
-	mov	r4, $key			@ pass key
-	mov	r5, $rounds			@ pass # of rounds
-	add	r12, $key, #248			@ pass key schedule
-	bl	_bsaes_key_convert
-	add	r4, $key, #248
-	vldmia	r4, {@XMM[6]}
-	vstmia	r12, {@XMM[15]}			@ save last round key
-	veor	@XMM[7], @XMM[7], @XMM[6]	@ fix up round 0 key
-	vstmia	r4, {@XMM[7]}
-
-.align	2
-0:
-#endif
-
-	vld1.8	{@XMM[15]}, [$ivp]		@ load IV
-	b	.Lcbc_dec_loop
-
-.align	4
-.Lcbc_dec_loop:
-	subs	$len, $len, #0x8
-	bmi	.Lcbc_dec_loop_finish
-
-	vld1.8	{@XMM[0]-@XMM[1]}, [$inp]!	@ load input
-	vld1.8	{@XMM[2]-@XMM[3]}, [$inp]!
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	mov	r4, $keysched			@ pass the key
-#else
-	add	r4, $key, #248
-#endif
-	vld1.8	{@XMM[4]-@XMM[5]}, [$inp]!
-	mov	r5, $rounds
-	vld1.8	{@XMM[6]-@XMM[7]}, [$inp]
-	sub	$inp, $inp, #0x60
-	vstmia	$fp, {@XMM[15]}			@ put aside IV
-
-	bl	_bsaes_decrypt8
-
-	vldmia	$fp, {@XMM[14]}			@ reload IV
-	vld1.8	{@XMM[8]-@XMM[9]}, [$inp]!	@ reload input
-	veor	@XMM[0], @XMM[0], @XMM[14]	@ ^= IV
-	vld1.8	{@XMM[10]-@XMM[11]}, [$inp]!
-	veor	@XMM[1], @XMM[1], @XMM[8]
-	veor	@XMM[6], @XMM[6], @XMM[9]
-	vld1.8	{@XMM[12]-@XMM[13]}, [$inp]!
-	veor	@XMM[4], @XMM[4], @XMM[10]
-	veor	@XMM[2], @XMM[2], @XMM[11]
-	vld1.8	{@XMM[14]-@XMM[15]}, [$inp]!
-	veor	@XMM[7], @XMM[7], @XMM[12]
-	vst1.8	{@XMM[0]-@XMM[1]}, [$out]!	@ write output
-	veor	@XMM[3], @XMM[3], @XMM[13]
-	vst1.8	{@XMM[6]}, [$out]!
-	veor	@XMM[5], @XMM[5], @XMM[14]
-	vst1.8	{@XMM[4]}, [$out]!
-	vst1.8	{@XMM[2]}, [$out]!
-	vst1.8	{@XMM[7]}, [$out]!
-	vst1.8	{@XMM[3]}, [$out]!
-	vst1.8	{@XMM[5]}, [$out]!
-
-	b	.Lcbc_dec_loop
-
-.Lcbc_dec_loop_finish:
-	adds	$len, $len, #8
-	beq	.Lcbc_dec_done
-
-	vld1.8	{@XMM[0]}, [$inp]!		@ load input
-	cmp	$len, #2
-	blo	.Lcbc_dec_one
-	vld1.8	{@XMM[1]}, [$inp]!
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	mov	r4, $keysched			@ pass the key
-#else
-	add	r4, $key, #248
-#endif
-	mov	r5, $rounds
-	vstmia	$fp, {@XMM[15]}			@ put aside IV
-	beq	.Lcbc_dec_two
-	vld1.8	{@XMM[2]}, [$inp]!
-	cmp	$len, #4
-	blo	.Lcbc_dec_three
-	vld1.8	{@XMM[3]}, [$inp]!
-	beq	.Lcbc_dec_four
-	vld1.8	{@XMM[4]}, [$inp]!
-	cmp	$len, #6
-	blo	.Lcbc_dec_five
-	vld1.8	{@XMM[5]}, [$inp]!
-	beq	.Lcbc_dec_six
-	vld1.8	{@XMM[6]}, [$inp]!
-	sub	$inp, $inp, #0x70
-
-	bl	_bsaes_decrypt8
-
-	vldmia	$fp, {@XMM[14]}			@ reload IV
-	vld1.8	{@XMM[8]-@XMM[9]}, [$inp]!	@ reload input
-	veor	@XMM[0], @XMM[0], @XMM[14]	@ ^= IV
-	vld1.8	{@XMM[10]-@XMM[11]}, [$inp]!
-	veor	@XMM[1], @XMM[1], @XMM[8]
-	veor	@XMM[6], @XMM[6], @XMM[9]
-	vld1.8	{@XMM[12]-@XMM[13]}, [$inp]!
-	veor	@XMM[4], @XMM[4], @XMM[10]
-	veor	@XMM[2], @XMM[2], @XMM[11]
-	vld1.8	{@XMM[15]}, [$inp]!
-	veor	@XMM[7], @XMM[7], @XMM[12]
-	vst1.8	{@XMM[0]-@XMM[1]}, [$out]!	@ write output
-	veor	@XMM[3], @XMM[3], @XMM[13]
-	vst1.8	{@XMM[6]}, [$out]!
-	vst1.8	{@XMM[4]}, [$out]!
-	vst1.8	{@XMM[2]}, [$out]!
-	vst1.8	{@XMM[7]}, [$out]!
-	vst1.8	{@XMM[3]}, [$out]!
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_six:
-	sub	$inp, $inp, #0x60
-	bl	_bsaes_decrypt8
-	vldmia	$fp,{@XMM[14]}			@ reload IV
-	vld1.8	{@XMM[8]-@XMM[9]}, [$inp]!	@ reload input
-	veor	@XMM[0], @XMM[0], @XMM[14]	@ ^= IV
-	vld1.8	{@XMM[10]-@XMM[11]}, [$inp]!
-	veor	@XMM[1], @XMM[1], @XMM[8]
-	veor	@XMM[6], @XMM[6], @XMM[9]
-	vld1.8	{@XMM[12]}, [$inp]!
-	veor	@XMM[4], @XMM[4], @XMM[10]
-	veor	@XMM[2], @XMM[2], @XMM[11]
-	vld1.8	{@XMM[15]}, [$inp]!
-	veor	@XMM[7], @XMM[7], @XMM[12]
-	vst1.8	{@XMM[0]-@XMM[1]}, [$out]!	@ write output
-	vst1.8	{@XMM[6]}, [$out]!
-	vst1.8	{@XMM[4]}, [$out]!
-	vst1.8	{@XMM[2]}, [$out]!
-	vst1.8	{@XMM[7]}, [$out]!
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_five:
-	sub	$inp, $inp, #0x50
-	bl	_bsaes_decrypt8
-	vldmia	$fp, {@XMM[14]}			@ reload IV
-	vld1.8	{@XMM[8]-@XMM[9]}, [$inp]!	@ reload input
-	veor	@XMM[0], @XMM[0], @XMM[14]	@ ^= IV
-	vld1.8	{@XMM[10]-@XMM[11]}, [$inp]!
-	veor	@XMM[1], @XMM[1], @XMM[8]
-	veor	@XMM[6], @XMM[6], @XMM[9]
-	vld1.8	{@XMM[15]}, [$inp]!
-	veor	@XMM[4], @XMM[4], @XMM[10]
-	vst1.8	{@XMM[0]-@XMM[1]}, [$out]!	@ write output
-	veor	@XMM[2], @XMM[2], @XMM[11]
-	vst1.8	{@XMM[6]}, [$out]!
-	vst1.8	{@XMM[4]}, [$out]!
-	vst1.8	{@XMM[2]}, [$out]!
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_four:
-	sub	$inp, $inp, #0x40
-	bl	_bsaes_decrypt8
-	vldmia	$fp, {@XMM[14]}			@ reload IV
-	vld1.8	{@XMM[8]-@XMM[9]}, [$inp]!	@ reload input
-	veor	@XMM[0], @XMM[0], @XMM[14]	@ ^= IV
-	vld1.8	{@XMM[10]}, [$inp]!
-	veor	@XMM[1], @XMM[1], @XMM[8]
-	veor	@XMM[6], @XMM[6], @XMM[9]
-	vld1.8	{@XMM[15]}, [$inp]!
-	veor	@XMM[4], @XMM[4], @XMM[10]
-	vst1.8	{@XMM[0]-@XMM[1]}, [$out]!	@ write output
-	vst1.8	{@XMM[6]}, [$out]!
-	vst1.8	{@XMM[4]}, [$out]!
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_three:
-	sub	$inp, $inp, #0x30
-	bl	_bsaes_decrypt8
-	vldmia	$fp, {@XMM[14]}			@ reload IV
-	vld1.8	{@XMM[8]-@XMM[9]}, [$inp]!	@ reload input
-	veor	@XMM[0], @XMM[0], @XMM[14]	@ ^= IV
-	vld1.8	{@XMM[15]}, [$inp]!
-	veor	@XMM[1], @XMM[1], @XMM[8]
-	veor	@XMM[6], @XMM[6], @XMM[9]
-	vst1.8	{@XMM[0]-@XMM[1]}, [$out]!	@ write output
-	vst1.8	{@XMM[6]}, [$out]!
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_two:
-	sub	$inp, $inp, #0x20
-	bl	_bsaes_decrypt8
-	vldmia	$fp, {@XMM[14]}			@ reload IV
-	vld1.8	{@XMM[8]}, [$inp]!		@ reload input
-	veor	@XMM[0], @XMM[0], @XMM[14]	@ ^= IV
-	vld1.8	{@XMM[15]}, [$inp]!		@ reload input
-	veor	@XMM[1], @XMM[1], @XMM[8]
-	vst1.8	{@XMM[0]-@XMM[1]}, [$out]!	@ write output
-	b	.Lcbc_dec_done
-.align	4
-.Lcbc_dec_one:
-	sub	$inp, $inp, #0x10
-	mov	$rounds, $out			@ save original out pointer
-	mov	$out, $fp			@ use the iv scratch space as out buffer
-	mov	r2, $key
-	vmov	@XMM[4],@XMM[15]		@ just in case ensure that IV
-	vmov	@XMM[5],@XMM[0]			@ and input are preserved
-	bl	AES_decrypt
-	vld1.8	{@XMM[0]}, [$fp,:64]		@ load result
-	veor	@XMM[0], @XMM[0], @XMM[4]	@ ^= IV
-	vmov	@XMM[15], @XMM[5]		@ @XMM[5] holds input
-	vst1.8	{@XMM[0]}, [$rounds]		@ write output
-
-.Lcbc_dec_done:
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	vmov.i32	q0, #0
-	vmov.i32	q1, #0
-.Lcbc_dec_bzero:				@ wipe key schedule [if any]
-	vstmia		$keysched!, {q0-q1}
-	cmp		$keysched, $fp
-	bne		.Lcbc_dec_bzero
-#endif
-
-	mov	sp, $fp
-	add	sp, #0x10			@ add sp,$fp,#0x10 is no good for thumb
-	vst1.8	{@XMM[15]}, [$ivp]		@ return IV
-	VFP_ABI_POP
-	ldmia	sp!, {r4-r10, pc}
-.size	bsaes_cbc_encrypt,.-bsaes_cbc_encrypt
-___
-}
-{
-my ($inp,$out,$len,$key, $ctr,$fp,$rounds)=(map("r$_",(0..3,8..10)));
-my $const = "r6";	# shared with _bsaes_encrypt8_alt
-my $keysched = "sp";
-
-$code.=<<___;
-.extern	AES_encrypt
-.global	bsaes_ctr32_encrypt_blocks
-.type	bsaes_ctr32_encrypt_blocks,%function
-.align	5
-bsaes_ctr32_encrypt_blocks:
-	cmp	$len, #8			@ use plain AES for
-	blo	.Lctr_enc_short			@ small sizes
-
-	mov	ip, sp
-	stmdb	sp!, {r4-r10, lr}
-	VFP_ABI_PUSH
-	ldr	$ctr, [ip]			@ ctr is 1st arg on the stack
-	sub	sp, sp, #0x10			@ scratch space to carry over the ctr
-	mov	$fp, sp				@ save sp
-
-	ldr	$rounds, [$key, #240]		@ get # of rounds
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	@ allocate the key schedule on the stack
-	sub	r12, sp, $rounds, lsl#7		@ 128 bytes per inner round key
-	add	r12, #`128-32`			@ size of bit-sliced key schedule
-
-	@ populate the key schedule
-	mov	r4, $key			@ pass key
-	mov	r5, $rounds			@ pass # of rounds
-	mov	sp, r12				@ sp is $keysched
-	bl	_bsaes_key_convert
-	veor	@XMM[7],@XMM[7],@XMM[15]	@ fix up last round key
-	vstmia	r12, {@XMM[7]}			@ save last round key
-
-	vld1.8	{@XMM[0]}, [$ctr]		@ load counter
-	add	$ctr, $const, #.LREVM0SR-.LM0	@ borrow $ctr
-	vldmia	$keysched, {@XMM[4]}		@ load round0 key
-#else
-	ldr	r12, [$key, #244]
-	eors	r12, #1
-	beq	0f
-
-	@ populate the key schedule
-	str	r12, [$key, #244]
-	mov	r4, $key			@ pass key
-	mov	r5, $rounds			@ pass # of rounds
-	add	r12, $key, #248			@ pass key schedule
-	bl	_bsaes_key_convert
-	veor	@XMM[7],@XMM[7],@XMM[15]	@ fix up last round key
-	vstmia	r12, {@XMM[7]}			@ save last round key
-
-.align	2
-0:	add	r12, $key, #248
-	vld1.8	{@XMM[0]}, [$ctr]		@ load counter
-	adrl	$ctr, .LREVM0SR			@ borrow $ctr
-	vldmia	r12, {@XMM[4]}			@ load round0 key
-	sub	sp, #0x10			@ place for adjusted round0 key
-#endif
-
-	vmov.i32	@XMM[8],#1		@ compose 1<<96
-	veor		@XMM[9],@XMM[9],@XMM[9]
-	vrev32.8	@XMM[0],@XMM[0]
-	vext.8		@XMM[8],@XMM[9],@XMM[8],#4
-	vrev32.8	@XMM[4],@XMM[4]
-	vadd.u32	@XMM[9],@XMM[8],@XMM[8]	@ compose 2<<96
-	vstmia	$keysched, {@XMM[4]}		@ save adjusted round0 key
-	b	.Lctr_enc_loop
-
-.align	4
-.Lctr_enc_loop:
-	vadd.u32	@XMM[10], @XMM[8], @XMM[9]	@ compose 3<<96
-	vadd.u32	@XMM[1], @XMM[0], @XMM[8]	@ +1
-	vadd.u32	@XMM[2], @XMM[0], @XMM[9]	@ +2
-	vadd.u32	@XMM[3], @XMM[0], @XMM[10]	@ +3
-	vadd.u32	@XMM[4], @XMM[1], @XMM[10]
-	vadd.u32	@XMM[5], @XMM[2], @XMM[10]
-	vadd.u32	@XMM[6], @XMM[3], @XMM[10]
-	vadd.u32	@XMM[7], @XMM[4], @XMM[10]
-	vadd.u32	@XMM[10], @XMM[5], @XMM[10]	@ next counter
-
-	@ Borrow prologue from _bsaes_encrypt8 to use the opportunity
-	@ to flip byte order in 32-bit counter
-
-	vldmia		$keysched, {@XMM[9]}		@ load round0 key
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, $keysched, #0x10		@ pass next round key
-#else
-	add		r4, $key, #`248+16`
-#endif
-	vldmia		$ctr, {@XMM[8]}			@ .LREVM0SR
-	mov		r5, $rounds			@ pass rounds
-	vstmia		$fp, {@XMM[10]}			@ save next counter
-	sub		$const, $ctr, #.LREVM0SR-.LSR	@ pass constants
-
-	bl		_bsaes_encrypt8_alt
-
-	subs		$len, $len, #8
-	blo		.Lctr_enc_loop_done
-
-	vld1.8		{@XMM[8]-@XMM[9]}, [$inp]!	@ load input
-	vld1.8		{@XMM[10]-@XMM[11]}, [$inp]!
-	veor		@XMM[0], @XMM[8]
-	veor		@XMM[1], @XMM[9]
-	vld1.8		{@XMM[12]-@XMM[13]}, [$inp]!
-	veor		@XMM[4], @XMM[10]
-	veor		@XMM[6], @XMM[11]
-	vld1.8		{@XMM[14]-@XMM[15]}, [$inp]!
-	veor		@XMM[3], @XMM[12]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!	@ write output
-	veor		@XMM[7], @XMM[13]
-	veor		@XMM[2], @XMM[14]
-	vst1.8		{@XMM[4]}, [$out]!
-	veor		@XMM[5], @XMM[15]
-	vst1.8		{@XMM[6]}, [$out]!
-	vmov.i32	@XMM[8], #1			@ compose 1<<96
-	vst1.8		{@XMM[3]}, [$out]!
-	veor		@XMM[9], @XMM[9], @XMM[9]
-	vst1.8		{@XMM[7]}, [$out]!
-	vext.8		@XMM[8], @XMM[9], @XMM[8], #4
-	vst1.8		{@XMM[2]}, [$out]!
-	vadd.u32	@XMM[9],@XMM[8],@XMM[8]		@ compose 2<<96
-	vst1.8		{@XMM[5]}, [$out]!
-	vldmia		$fp, {@XMM[0]}			@ load counter
-
-	bne		.Lctr_enc_loop
-	b		.Lctr_enc_done
-
-.align	4
-.Lctr_enc_loop_done:
-	add		$len, $len, #8
-	vld1.8		{@XMM[8]}, [$inp]!	@ load input
-	veor		@XMM[0], @XMM[8]
-	vst1.8		{@XMM[0]}, [$out]!	@ write output
-	cmp		$len, #2
-	blo		.Lctr_enc_done
-	vld1.8		{@XMM[9]}, [$inp]!
-	veor		@XMM[1], @XMM[9]
-	vst1.8		{@XMM[1]}, [$out]!
-	beq		.Lctr_enc_done
-	vld1.8		{@XMM[10]}, [$inp]!
-	veor		@XMM[4], @XMM[10]
-	vst1.8		{@XMM[4]}, [$out]!
-	cmp		$len, #4
-	blo		.Lctr_enc_done
-	vld1.8		{@XMM[11]}, [$inp]!
-	veor		@XMM[6], @XMM[11]
-	vst1.8		{@XMM[6]}, [$out]!
-	beq		.Lctr_enc_done
-	vld1.8		{@XMM[12]}, [$inp]!
-	veor		@XMM[3], @XMM[12]
-	vst1.8		{@XMM[3]}, [$out]!
-	cmp		$len, #6
-	blo		.Lctr_enc_done
-	vld1.8		{@XMM[13]}, [$inp]!
-	veor		@XMM[7], @XMM[13]
-	vst1.8		{@XMM[7]}, [$out]!
-	beq		.Lctr_enc_done
-	vld1.8		{@XMM[14]}, [$inp]
-	veor		@XMM[2], @XMM[14]
-	vst1.8		{@XMM[2]}, [$out]!
-
-.Lctr_enc_done:
-	vmov.i32	q0, #0
-	vmov.i32	q1, #0
-#ifndef	BSAES_ASM_EXTENDED_KEY
-.Lctr_enc_bzero:			@ wipe key schedule [if any]
-	vstmia		$keysched!, {q0-q1}
-	cmp		$keysched, $fp
-	bne		.Lctr_enc_bzero
-#else
-	vstmia		$keysched, {q0-q1}
-#endif
-
-	mov	sp, $fp
-	add	sp, #0x10		@ add sp,$fp,#0x10 is no good for thumb
-	VFP_ABI_POP
-	ldmia	sp!, {r4-r10, pc}	@ return
-
-.align	4
-.Lctr_enc_short:
-	ldr	ip, [sp]		@ ctr pointer is passed on stack
-	stmdb	sp!, {r4-r8, lr}
-
-	mov	r4, $inp		@ copy arguments
-	mov	r5, $out
-	mov	r6, $len
-	mov	r7, $key
-	ldr	r8, [ip, #12]		@ load counter LSW
-	vld1.8	{@XMM[1]}, [ip]		@ load whole counter value
-#ifdef __ARMEL__
-	rev	r8, r8
-#endif
-	sub	sp, sp, #0x10
-	vst1.8	{@XMM[1]}, [sp,:64]	@ copy counter value
-	sub	sp, sp, #0x10
-
-.Lctr_enc_short_loop:
-	add	r0, sp, #0x10		@ input counter value
-	mov	r1, sp			@ output on the stack
-	mov	r2, r7			@ key
-
-	bl	AES_encrypt
-
-	vld1.8	{@XMM[0]}, [r4]!	@ load input
-	vld1.8	{@XMM[1]}, [sp,:64]	@ load encrypted counter
-	add	r8, r8, #1
-#ifdef __ARMEL__
-	rev	r0, r8
-	str	r0, [sp, #0x1c]		@ next counter value
-#else
-	str	r8, [sp, #0x1c]		@ next counter value
-#endif
-	veor	@XMM[0],@XMM[0],@XMM[1]
-	vst1.8	{@XMM[0]}, [r5]!	@ store output
-	subs	r6, r6, #1
-	bne	.Lctr_enc_short_loop
-
-	vmov.i32	q0, #0
-	vmov.i32	q1, #0
-	vstmia		sp!, {q0-q1}
-
-	ldmia	sp!, {r4-r8, pc}
-.size	bsaes_ctr32_encrypt_blocks,.-bsaes_ctr32_encrypt_blocks
-___
-}
-{
-######################################################################
-# void bsaes_xts_[en|de]crypt(const char *inp,char *out,size_t len,
-#	const AES_KEY *key1, const AES_KEY *key2,
-#	const unsigned char iv[16]);
-#
-my ($inp,$out,$len,$key,$rounds,$magic,$fp)=(map("r$_",(7..10,1..3)));
-my $const="r6";		# returned by _bsaes_key_convert
-my $twmask=@XMM[5];
-my @T=@XMM[6..7];
-
-$code.=<<___;
-.globl	bsaes_xts_encrypt
-.type	bsaes_xts_encrypt,%function
-.align	4
-bsaes_xts_encrypt:
-	mov	ip, sp
-	stmdb	sp!, {r4-r10, lr}		@ 0x20
-	VFP_ABI_PUSH
-	mov	r6, sp				@ future $fp
-
-	mov	$inp, r0
-	mov	$out, r1
-	mov	$len, r2
-	mov	$key, r3
-
-	sub	r0, sp, #0x10			@ 0x10
-	bic	r0, #0xf			@ align at 16 bytes
-	mov	sp, r0
-
-#ifdef	XTS_CHAIN_TWEAK
-	ldr	r0, [ip]			@ pointer to input tweak
-#else
-	@ generate initial tweak
-	ldr	r0, [ip, #4]			@ iv[]
-	mov	r1, sp
-	ldr	r2, [ip, #0]			@ key2
-	bl	AES_encrypt
-	mov	r0,sp				@ pointer to initial tweak
-#endif
-
-	ldr	$rounds, [$key, #240]		@ get # of rounds
-	mov	$fp, r6
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	@ allocate the key schedule on the stack
-	sub	r12, sp, $rounds, lsl#7		@ 128 bytes per inner round key
-	@ add	r12, #`128-32`			@ size of bit-sliced key schedule
-	sub	r12, #`32+16`			@ place for tweak[9]
-
-	@ populate the key schedule
-	mov	r4, $key			@ pass key
-	mov	r5, $rounds			@ pass # of rounds
-	mov	sp, r12
-	add	r12, #0x90			@ pass key schedule
-	bl	_bsaes_key_convert
-	veor	@XMM[7], @XMM[7], @XMM[15]	@ fix up last round key
-	vstmia	r12, {@XMM[7]}			@ save last round key
-#else
-	ldr	r12, [$key, #244]
-	eors	r12, #1
-	beq	0f
-
-	str	r12, [$key, #244]
-	mov	r4, $key			@ pass key
-	mov	r5, $rounds			@ pass # of rounds
-	add	r12, $key, #248			@ pass key schedule
-	bl	_bsaes_key_convert
-	veor	@XMM[7], @XMM[7], @XMM[15]	@ fix up last round key
-	vstmia	r12, {@XMM[7]}
-
-.align	2
-0:	sub	sp, #0x90			@ place for tweak[9]
-#endif
-
-	vld1.8	{@XMM[8]}, [r0]			@ initial tweak
-	adr	$magic, .Lxts_magic
-
-	subs	$len, #0x80
-	blo	.Lxts_enc_short
-	b	.Lxts_enc_loop
-
-.align	4
-.Lxts_enc_loop:
-	vldmia		$magic, {$twmask}	@ load XTS magic
-	vshr.s64	@T[0], @XMM[8], #63
-	mov		r0, sp
-	vand		@T[0], @T[0], $twmask
-___
-for($i=9;$i<16;$i++) {
-$code.=<<___;
-	vadd.u64	@XMM[$i], @XMM[$i-1], @XMM[$i-1]
-	vst1.64		{@XMM[$i-1]}, [r0,:128]!
-	vswp		`&Dhi("@T[0]")`,`&Dlo("@T[0]")`
-	vshr.s64	@T[1], @XMM[$i], #63
-	veor		@XMM[$i], @XMM[$i], @T[0]
-	vand		@T[1], @T[1], $twmask
-___
-	@T=reverse(@T);
-
-$code.=<<___ if ($i>=10);
-	vld1.8		{@XMM[$i-10]}, [$inp]!
-___
-$code.=<<___ if ($i>=11);
-	veor		@XMM[$i-11], @XMM[$i-11], @XMM[$i-3]
-___
-}
-$code.=<<___;
-	vadd.u64	@XMM[8], @XMM[15], @XMM[15]
-	vst1.64		{@XMM[15]}, [r0,:128]!
-	vswp		`&Dhi("@T[0]")`,`&Dlo("@T[0]")`
-	veor		@XMM[8], @XMM[8], @T[0]
-	vst1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-
-	vld1.8		{@XMM[6]-@XMM[7]}, [$inp]!
-	veor		@XMM[5], @XMM[5], @XMM[13]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[6], @XMM[6], @XMM[14]
-	mov		r5, $rounds			@ pass rounds
-	veor		@XMM[7], @XMM[7], @XMM[15]
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{@XMM[ 8]-@XMM[ 9]}, [r0,:128]!
-	vld1.64		{@XMM[10]-@XMM[11]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	vld1.64		{@XMM[12]-@XMM[13]}, [r0,:128]!
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[4], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	veor		@XMM[9], @XMM[6], @XMM[11]
-	vld1.64		{@XMM[14]-@XMM[15]}, [r0,:128]!
-	veor		@XMM[10], @XMM[3], @XMM[12]
-	vst1.8		{@XMM[8]-@XMM[9]}, [$out]!
-	veor		@XMM[11], @XMM[7], @XMM[13]
-	veor		@XMM[12], @XMM[2], @XMM[14]
-	vst1.8		{@XMM[10]-@XMM[11]}, [$out]!
-	veor		@XMM[13], @XMM[5], @XMM[15]
-	vst1.8		{@XMM[12]-@XMM[13]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-
-	subs		$len, #0x80
-	bpl		.Lxts_enc_loop
-
-.Lxts_enc_short:
-	adds		$len, #0x70
-	bmi		.Lxts_enc_done
-
-	vldmia		$magic, {$twmask}	@ load XTS magic
-	vshr.s64	@T[0], @XMM[8], #63
-	mov		r0, sp
-	vand		@T[0], @T[0], $twmask
-___
-for($i=9;$i<16;$i++) {
-$code.=<<___;
-	vadd.u64	@XMM[$i], @XMM[$i-1], @XMM[$i-1]
-	vst1.64		{@XMM[$i-1]}, [r0,:128]!
-	vswp		`&Dhi("@T[0]")`,`&Dlo("@T[0]")`
-	vshr.s64	@T[1], @XMM[$i], #63
-	veor		@XMM[$i], @XMM[$i], @T[0]
-	vand		@T[1], @T[1], $twmask
-___
-	@T=reverse(@T);
-
-$code.=<<___ if ($i>=10);
-	vld1.8		{@XMM[$i-10]}, [$inp]!
-	subs		$len, #0x10
-	bmi		.Lxts_enc_`$i-9`
-___
-$code.=<<___ if ($i>=11);
-	veor		@XMM[$i-11], @XMM[$i-11], @XMM[$i-3]
-___
-}
-$code.=<<___;
-	sub		$len, #0x10
-	vst1.64		{@XMM[15]}, [r0,:128]		@ next round tweak
-
-	vld1.8		{@XMM[6]}, [$inp]!
-	veor		@XMM[5], @XMM[5], @XMM[13]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[6], @XMM[6], @XMM[14]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{@XMM[ 8]-@XMM[ 9]}, [r0,:128]!
-	vld1.64		{@XMM[10]-@XMM[11]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	vld1.64		{@XMM[12]-@XMM[13]}, [r0,:128]!
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[4], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	veor		@XMM[9], @XMM[6], @XMM[11]
-	vld1.64		{@XMM[14]}, [r0,:128]!
-	veor		@XMM[10], @XMM[3], @XMM[12]
-	vst1.8		{@XMM[8]-@XMM[9]}, [$out]!
-	veor		@XMM[11], @XMM[7], @XMM[13]
-	veor		@XMM[12], @XMM[2], @XMM[14]
-	vst1.8		{@XMM[10]-@XMM[11]}, [$out]!
-	vst1.8		{@XMM[12]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-.align	4
-.Lxts_enc_6:
-	vst1.64		{@XMM[14]}, [r0,:128]		@ next round tweak
-
-	veor		@XMM[4], @XMM[4], @XMM[12]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[5], @XMM[5], @XMM[13]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{@XMM[ 8]-@XMM[ 9]}, [r0,:128]!
-	vld1.64		{@XMM[10]-@XMM[11]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	vld1.64		{@XMM[12]-@XMM[13]}, [r0,:128]!
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[4], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	veor		@XMM[9], @XMM[6], @XMM[11]
-	veor		@XMM[10], @XMM[3], @XMM[12]
-	vst1.8		{@XMM[8]-@XMM[9]}, [$out]!
-	veor		@XMM[11], @XMM[7], @XMM[13]
-	vst1.8		{@XMM[10]-@XMM[11]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-
-@ put this in range for both ARM and Thumb mode adr instructions
-.align	5
-.Lxts_magic:
-	.quad	1, 0x87
-
-.align	5
-.Lxts_enc_5:
-	vst1.64		{@XMM[13]}, [r0,:128]		@ next round tweak
-
-	veor		@XMM[3], @XMM[3], @XMM[11]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[4], @XMM[4], @XMM[12]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{@XMM[ 8]-@XMM[ 9]}, [r0,:128]!
-	vld1.64		{@XMM[10]-@XMM[11]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	vld1.64		{@XMM[12]}, [r0,:128]!
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[4], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	veor		@XMM[9], @XMM[6], @XMM[11]
-	veor		@XMM[10], @XMM[3], @XMM[12]
-	vst1.8		{@XMM[8]-@XMM[9]}, [$out]!
-	vst1.8		{@XMM[10]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-.align	4
-.Lxts_enc_4:
-	vst1.64		{@XMM[12]}, [r0,:128]		@ next round tweak
-
-	veor		@XMM[2], @XMM[2], @XMM[10]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[3], @XMM[3], @XMM[11]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{@XMM[ 8]-@XMM[ 9]}, [r0,:128]!
-	vld1.64		{@XMM[10]-@XMM[11]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[4], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	veor		@XMM[9], @XMM[6], @XMM[11]
-	vst1.8		{@XMM[8]-@XMM[9]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-.align	4
-.Lxts_enc_3:
-	vst1.64		{@XMM[11]}, [r0,:128]		@ next round tweak
-
-	veor		@XMM[1], @XMM[1], @XMM[9]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[2], @XMM[2], @XMM[10]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{@XMM[8]-@XMM[9]}, [r0,:128]!
-	vld1.64		{@XMM[10]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[4], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	vst1.8		{@XMM[8]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-.align	4
-.Lxts_enc_2:
-	vst1.64		{@XMM[10]}, [r0,:128]		@ next round tweak
-
-	veor		@XMM[0], @XMM[0], @XMM[8]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[1], @XMM[1], @XMM[9]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_encrypt8
-
-	vld1.64		{@XMM[8]-@XMM[9]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_enc_done
-.align	4
-.Lxts_enc_1:
-	mov		r0, sp
-	veor		@XMM[0], @XMM[8]
-	mov		r1, sp
-	vst1.8		{@XMM[0]}, [sp,:128]
-	mov		r2, $key
-	mov		r4, $fp				@ preserve fp
-
-	bl		AES_encrypt
-
-	vld1.8		{@XMM[0]}, [sp,:128]
-	veor		@XMM[0], @XMM[0], @XMM[8]
-	vst1.8		{@XMM[0]}, [$out]!
-	mov		$fp, r4
-
-	vmov		@XMM[8], @XMM[9]		@ next round tweak
-
-.Lxts_enc_done:
-#ifndef	XTS_CHAIN_TWEAK
-	adds		$len, #0x10
-	beq		.Lxts_enc_ret
-	sub		r6, $out, #0x10
-
-.Lxts_enc_steal:
-	ldrb		r0, [$inp], #1
-	ldrb		r1, [$out, #-0x10]
-	strb		r0, [$out, #-0x10]
-	strb		r1, [$out], #1
-
-	subs		$len, #1
-	bhi		.Lxts_enc_steal
-
-	vld1.8		{@XMM[0]}, [r6]
-	mov		r0, sp
-	veor		@XMM[0], @XMM[0], @XMM[8]
-	mov		r1, sp
-	vst1.8		{@XMM[0]}, [sp,:128]
-	mov		r2, $key
-	mov		r4, $fp			@ preserve fp
-
-	bl		AES_encrypt
-
-	vld1.8		{@XMM[0]}, [sp,:128]
-	veor		@XMM[0], @XMM[0], @XMM[8]
-	vst1.8		{@XMM[0]}, [r6]
-	mov		$fp, r4
-#endif
-
-.Lxts_enc_ret:
-	bic		r0, $fp, #0xf
-	vmov.i32	q0, #0
-	vmov.i32	q1, #0
-#ifdef	XTS_CHAIN_TWEAK
-	ldr		r1, [$fp, #0x20+VFP_ABI_FRAME]	@ chain tweak
-#endif
-.Lxts_enc_bzero:				@ wipe key schedule [if any]
-	vstmia		sp!, {q0-q1}
-	cmp		sp, r0
-	bne		.Lxts_enc_bzero
-
-	mov		sp, $fp
-#ifdef	XTS_CHAIN_TWEAK
-	vst1.8		{@XMM[8]}, [r1]
-#endif
-	VFP_ABI_POP
-	ldmia		sp!, {r4-r10, pc}	@ return
-
-.size	bsaes_xts_encrypt,.-bsaes_xts_encrypt
-
-.globl	bsaes_xts_decrypt
-.type	bsaes_xts_decrypt,%function
-.align	4
-bsaes_xts_decrypt:
-	mov	ip, sp
-	stmdb	sp!, {r4-r10, lr}		@ 0x20
-	VFP_ABI_PUSH
-	mov	r6, sp				@ future $fp
-
-	mov	$inp, r0
-	mov	$out, r1
-	mov	$len, r2
-	mov	$key, r3
-
-	sub	r0, sp, #0x10			@ 0x10
-	bic	r0, #0xf			@ align at 16 bytes
-	mov	sp, r0
-
-#ifdef	XTS_CHAIN_TWEAK
-	ldr	r0, [ip]			@ pointer to input tweak
-#else
-	@ generate initial tweak
-	ldr	r0, [ip, #4]			@ iv[]
-	mov	r1, sp
-	ldr	r2, [ip, #0]			@ key2
-	bl	AES_encrypt
-	mov	r0, sp				@ pointer to initial tweak
-#endif
-
-	ldr	$rounds, [$key, #240]		@ get # of rounds
-	mov	$fp, r6
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	@ allocate the key schedule on the stack
-	sub	r12, sp, $rounds, lsl#7		@ 128 bytes per inner round key
-	@ add	r12, #`128-32`			@ size of bit-sliced key schedule
-	sub	r12, #`32+16`			@ place for tweak[9]
-
-	@ populate the key schedule
-	mov	r4, $key			@ pass key
-	mov	r5, $rounds			@ pass # of rounds
-	mov	sp, r12
-	add	r12, #0x90			@ pass key schedule
-	bl	_bsaes_key_convert
-	add	r4, sp, #0x90
-	vldmia	r4, {@XMM[6]}
-	vstmia	r12,  {@XMM[15]}		@ save last round key
-	veor	@XMM[7], @XMM[7], @XMM[6]	@ fix up round 0 key
-	vstmia	r4, {@XMM[7]}
-#else
-	ldr	r12, [$key, #244]
-	eors	r12, #1
-	beq	0f
-
-	str	r12, [$key, #244]
-	mov	r4, $key			@ pass key
-	mov	r5, $rounds			@ pass # of rounds
-	add	r12, $key, #248			@ pass key schedule
-	bl	_bsaes_key_convert
-	add	r4, $key, #248
-	vldmia	r4, {@XMM[6]}
-	vstmia	r12,  {@XMM[15]}		@ save last round key
-	veor	@XMM[7], @XMM[7], @XMM[6]	@ fix up round 0 key
-	vstmia	r4, {@XMM[7]}
-
-.align	2
-0:	sub	sp, #0x90			@ place for tweak[9]
-#endif
-	vld1.8	{@XMM[8]}, [r0]			@ initial tweak
-	adr	$magic, .Lxts_magic
-
-#ifndef	XTS_CHAIN_TWEAK
-	tst	$len, #0xf			@ if not multiple of 16
-	it	ne				@ Thumb2 thing, sanity check in ARM
-	subne	$len, #0x10			@ subtract another 16 bytes
-#endif
-	subs	$len, #0x80
-
-	blo	.Lxts_dec_short
-	b	.Lxts_dec_loop
-
-.align	4
-.Lxts_dec_loop:
-	vldmia		$magic, {$twmask}	@ load XTS magic
-	vshr.s64	@T[0], @XMM[8], #63
-	mov		r0, sp
-	vand		@T[0], @T[0], $twmask
-___
-for($i=9;$i<16;$i++) {
-$code.=<<___;
-	vadd.u64	@XMM[$i], @XMM[$i-1], @XMM[$i-1]
-	vst1.64		{@XMM[$i-1]}, [r0,:128]!
-	vswp		`&Dhi("@T[0]")`,`&Dlo("@T[0]")`
-	vshr.s64	@T[1], @XMM[$i], #63
-	veor		@XMM[$i], @XMM[$i], @T[0]
-	vand		@T[1], @T[1], $twmask
-___
-	@T=reverse(@T);
-
-$code.=<<___ if ($i>=10);
-	vld1.8		{@XMM[$i-10]}, [$inp]!
-___
-$code.=<<___ if ($i>=11);
-	veor		@XMM[$i-11], @XMM[$i-11], @XMM[$i-3]
-___
-}
-$code.=<<___;
-	vadd.u64	@XMM[8], @XMM[15], @XMM[15]
-	vst1.64		{@XMM[15]}, [r0,:128]!
-	vswp		`&Dhi("@T[0]")`,`&Dlo("@T[0]")`
-	veor		@XMM[8], @XMM[8], @T[0]
-	vst1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-
-	vld1.8		{@XMM[6]-@XMM[7]}, [$inp]!
-	veor		@XMM[5], @XMM[5], @XMM[13]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[6], @XMM[6], @XMM[14]
-	mov		r5, $rounds			@ pass rounds
-	veor		@XMM[7], @XMM[7], @XMM[15]
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{@XMM[ 8]-@XMM[ 9]}, [r0,:128]!
-	vld1.64		{@XMM[10]-@XMM[11]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	vld1.64		{@XMM[12]-@XMM[13]}, [r0,:128]!
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[6], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	veor		@XMM[9], @XMM[4], @XMM[11]
-	vld1.64		{@XMM[14]-@XMM[15]}, [r0,:128]!
-	veor		@XMM[10], @XMM[2], @XMM[12]
-	vst1.8		{@XMM[8]-@XMM[9]}, [$out]!
-	veor		@XMM[11], @XMM[7], @XMM[13]
-	veor		@XMM[12], @XMM[3], @XMM[14]
-	vst1.8		{@XMM[10]-@XMM[11]}, [$out]!
-	veor		@XMM[13], @XMM[5], @XMM[15]
-	vst1.8		{@XMM[12]-@XMM[13]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-
-	subs		$len, #0x80
-	bpl		.Lxts_dec_loop
-
-.Lxts_dec_short:
-	adds		$len, #0x70
-	bmi		.Lxts_dec_done
-
-	vldmia		$magic, {$twmask}	@ load XTS magic
-	vshr.s64	@T[0], @XMM[8], #63
-	mov		r0, sp
-	vand		@T[0], @T[0], $twmask
-___
-for($i=9;$i<16;$i++) {
-$code.=<<___;
-	vadd.u64	@XMM[$i], @XMM[$i-1], @XMM[$i-1]
-	vst1.64		{@XMM[$i-1]}, [r0,:128]!
-	vswp		`&Dhi("@T[0]")`,`&Dlo("@T[0]")`
-	vshr.s64	@T[1], @XMM[$i], #63
-	veor		@XMM[$i], @XMM[$i], @T[0]
-	vand		@T[1], @T[1], $twmask
-___
-	@T=reverse(@T);
-
-$code.=<<___ if ($i>=10);
-	vld1.8		{@XMM[$i-10]}, [$inp]!
-	subs		$len, #0x10
-	bmi		.Lxts_dec_`$i-9`
-___
-$code.=<<___ if ($i>=11);
-	veor		@XMM[$i-11], @XMM[$i-11], @XMM[$i-3]
-___
-}
-$code.=<<___;
-	sub		$len, #0x10
-	vst1.64		{@XMM[15]}, [r0,:128]		@ next round tweak
-
-	vld1.8		{@XMM[6]}, [$inp]!
-	veor		@XMM[5], @XMM[5], @XMM[13]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[6], @XMM[6], @XMM[14]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{@XMM[ 8]-@XMM[ 9]}, [r0,:128]!
-	vld1.64		{@XMM[10]-@XMM[11]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	vld1.64		{@XMM[12]-@XMM[13]}, [r0,:128]!
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[6], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	veor		@XMM[9], @XMM[4], @XMM[11]
-	vld1.64		{@XMM[14]}, [r0,:128]!
-	veor		@XMM[10], @XMM[2], @XMM[12]
-	vst1.8		{@XMM[8]-@XMM[9]}, [$out]!
-	veor		@XMM[11], @XMM[7], @XMM[13]
-	veor		@XMM[12], @XMM[3], @XMM[14]
-	vst1.8		{@XMM[10]-@XMM[11]}, [$out]!
-	vst1.8		{@XMM[12]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_6:
-	vst1.64		{@XMM[14]}, [r0,:128]		@ next round tweak
-
-	veor		@XMM[4], @XMM[4], @XMM[12]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[5], @XMM[5], @XMM[13]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{@XMM[ 8]-@XMM[ 9]}, [r0,:128]!
-	vld1.64		{@XMM[10]-@XMM[11]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	vld1.64		{@XMM[12]-@XMM[13]}, [r0,:128]!
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[6], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	veor		@XMM[9], @XMM[4], @XMM[11]
-	veor		@XMM[10], @XMM[2], @XMM[12]
-	vst1.8		{@XMM[8]-@XMM[9]}, [$out]!
-	veor		@XMM[11], @XMM[7], @XMM[13]
-	vst1.8		{@XMM[10]-@XMM[11]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_5:
-	vst1.64		{@XMM[13]}, [r0,:128]		@ next round tweak
-
-	veor		@XMM[3], @XMM[3], @XMM[11]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[4], @XMM[4], @XMM[12]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{@XMM[ 8]-@XMM[ 9]}, [r0,:128]!
-	vld1.64		{@XMM[10]-@XMM[11]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	vld1.64		{@XMM[12]}, [r0,:128]!
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[6], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	veor		@XMM[9], @XMM[4], @XMM[11]
-	veor		@XMM[10], @XMM[2], @XMM[12]
-	vst1.8		{@XMM[8]-@XMM[9]}, [$out]!
-	vst1.8		{@XMM[10]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_4:
-	vst1.64		{@XMM[12]}, [r0,:128]		@ next round tweak
-
-	veor		@XMM[2], @XMM[2], @XMM[10]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[3], @XMM[3], @XMM[11]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{@XMM[ 8]-@XMM[ 9]}, [r0,:128]!
-	vld1.64		{@XMM[10]-@XMM[11]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[6], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	veor		@XMM[9], @XMM[4], @XMM[11]
-	vst1.8		{@XMM[8]-@XMM[9]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_3:
-	vst1.64		{@XMM[11]}, [r0,:128]		@ next round tweak
-
-	veor		@XMM[1], @XMM[1], @XMM[9]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[2], @XMM[2], @XMM[10]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{@XMM[8]-@XMM[9]}, [r0,:128]!
-	vld1.64		{@XMM[10]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	veor		@XMM[8], @XMM[6], @XMM[10]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-	vst1.8		{@XMM[8]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_2:
-	vst1.64		{@XMM[10]}, [r0,:128]		@ next round tweak
-
-	veor		@XMM[0], @XMM[0], @XMM[8]
-#ifndef	BSAES_ASM_EXTENDED_KEY
-	add		r4, sp, #0x90			@ pass key schedule
-#else
-	add		r4, $key, #248			@ pass key schedule
-#endif
-	veor		@XMM[1], @XMM[1], @XMM[9]
-	mov		r5, $rounds			@ pass rounds
-	mov		r0, sp
-
-	bl		_bsaes_decrypt8
-
-	vld1.64		{@XMM[8]-@XMM[9]}, [r0,:128]!
-	veor		@XMM[0], @XMM[0], @XMM[ 8]
-	veor		@XMM[1], @XMM[1], @XMM[ 9]
-	vst1.8		{@XMM[0]-@XMM[1]}, [$out]!
-
-	vld1.64		{@XMM[8]}, [r0,:128]		@ next round tweak
-	b		.Lxts_dec_done
-.align	4
-.Lxts_dec_1:
-	mov		r0, sp
-	veor		@XMM[0], @XMM[8]
-	mov		r1, sp
-	vst1.8		{@XMM[0]}, [sp,:128]
-	mov		r2, $key
-	mov		r4, $fp				@ preserve fp
-	mov		r5, $magic			@ preserve magic
-
-	bl		AES_decrypt
-
-	vld1.8		{@XMM[0]}, [sp,:128]
-	veor		@XMM[0], @XMM[0], @XMM[8]
-	vst1.8		{@XMM[0]}, [$out]!
-	mov		$fp, r4
-	mov		$magic, r5
-
-	vmov		@XMM[8], @XMM[9]		@ next round tweak
-
-.Lxts_dec_done:
-#ifndef	XTS_CHAIN_TWEAK
-	adds		$len, #0x10
-	beq		.Lxts_dec_ret
-
-	@ calculate one round of extra tweak for the stolen ciphertext
-	vldmia		$magic, {$twmask}
-	vshr.s64	@XMM[6], @XMM[8], #63
-	vand		@XMM[6], @XMM[6], $twmask
-	vadd.u64	@XMM[9], @XMM[8], @XMM[8]
-	vswp		`&Dhi("@XMM[6]")`,`&Dlo("@XMM[6]")`
-	veor		@XMM[9], @XMM[9], @XMM[6]
-
-	@ perform the final decryption with the last tweak value
-	vld1.8		{@XMM[0]}, [$inp]!
-	mov		r0, sp
-	veor		@XMM[0], @XMM[0], @XMM[9]
-	mov		r1, sp
-	vst1.8		{@XMM[0]}, [sp,:128]
-	mov		r2, $key
-	mov		r4, $fp			@ preserve fp
-
-	bl		AES_decrypt
-
-	vld1.8		{@XMM[0]}, [sp,:128]
-	veor		@XMM[0], @XMM[0], @XMM[9]
-	vst1.8		{@XMM[0]}, [$out]
-
-	mov		r6, $out
-.Lxts_dec_steal:
-	ldrb		r1, [$out]
-	ldrb		r0, [$inp], #1
-	strb		r1, [$out, #0x10]
-	strb		r0, [$out], #1
-
-	subs		$len, #1
-	bhi		.Lxts_dec_steal
-
-	vld1.8		{@XMM[0]}, [r6]
-	mov		r0, sp
-	veor		@XMM[0], @XMM[8]
-	mov		r1, sp
-	vst1.8		{@XMM[0]}, [sp,:128]
-	mov		r2, $key
-
-	bl		AES_decrypt
-
-	vld1.8		{@XMM[0]}, [sp,:128]
-	veor		@XMM[0], @XMM[0], @XMM[8]
-	vst1.8		{@XMM[0]}, [r6]
-	mov		$fp, r4
-#endif
-
-.Lxts_dec_ret:
-	bic		r0, $fp, #0xf
-	vmov.i32	q0, #0
-	vmov.i32	q1, #0
-#ifdef	XTS_CHAIN_TWEAK
-	ldr		r1, [$fp, #0x20+VFP_ABI_FRAME]	@ chain tweak
-#endif
-.Lxts_dec_bzero:				@ wipe key schedule [if any]
-	vstmia		sp!, {q0-q1}
-	cmp		sp, r0
-	bne		.Lxts_dec_bzero
-
-	mov		sp, $fp
-#ifdef	XTS_CHAIN_TWEAK
-	vst1.8		{@XMM[8]}, [r1]
-#endif
-	VFP_ABI_POP
-	ldmia		sp!, {r4-r10, pc}	@ return
-
-.size	bsaes_xts_decrypt,.-bsaes_xts_decrypt
-___
-}
-$code.=<<___;
-#endif
-___
-
-$code =~ s/\`([^\`]*)\`/eval($1)/gem;
-
-open SELF,$0;
-while(<SELF>) {
-	next if (/^#!/);
-        last if (!s/^#/@/ and !/^$/);
-        print;
-}
-close SELF;
-
-print $code;
-
-close STDOUT;
diff --git a/arch/arm/crypto/chacha20-neon-core.S b/arch/arm/crypto/chacha20-neon-core.S
new file mode 100644
index 000000000000..3fecb2124c35
--- /dev/null
+++ b/arch/arm/crypto/chacha20-neon-core.S
@@ -0,0 +1,523 @@
+/*
+ * ChaCha20 256-bit cipher algorithm, RFC7539, ARM NEON functions
+ *
+ * Copyright (C) 2016 Linaro, Ltd. <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Based on:
+ * ChaCha20 256-bit cipher algorithm, RFC7539, x64 SSE3 functions
+ *
+ * Copyright (C) 2015 Martin Willi
+ *
+ * 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 <linux/linkage.h>
+
+	.text
+	.fpu		neon
+	.align		5
+
+ENTRY(chacha20_block_xor_neon)
+	// r0: Input state matrix, s
+	// r1: 1 data block output, o
+	// r2: 1 data block input, i
+
+	//
+	// This function encrypts one ChaCha20 block by loading the state matrix
+	// in four NEON registers. It performs matrix operation on four words in
+	// parallel, but requireds shuffling to rearrange the words after each
+	// round.
+	//
+
+	// x0..3 = s0..3
+	add		ip, r0, #0x20
+	vld1.32		{q0-q1}, [r0]
+	vld1.32		{q2-q3}, [ip]
+
+	vmov		q8, q0
+	vmov		q9, q1
+	vmov		q10, q2
+	vmov		q11, q3
+
+	mov		r3, #10
+
+.Ldoubleround:
+	// x0 += x1, x3 = rotl32(x3 ^ x0, 16)
+	vadd.i32	q0, q0, q1
+	veor		q4, q3, q0
+	vshl.u32	q3, q4, #16
+	vsri.u32	q3, q4, #16
+
+	// x2 += x3, x1 = rotl32(x1 ^ x2, 12)
+	vadd.i32	q2, q2, q3
+	veor		q4, q1, q2
+	vshl.u32	q1, q4, #12
+	vsri.u32	q1, q4, #20
+
+	// x0 += x1, x3 = rotl32(x3 ^ x0, 8)
+	vadd.i32	q0, q0, q1
+	veor		q4, q3, q0
+	vshl.u32	q3, q4, #8
+	vsri.u32	q3, q4, #24
+
+	// x2 += x3, x1 = rotl32(x1 ^ x2, 7)
+	vadd.i32	q2, q2, q3
+	veor		q4, q1, q2
+	vshl.u32	q1, q4, #7
+	vsri.u32	q1, q4, #25
+
+	// x1 = shuffle32(x1, MASK(0, 3, 2, 1))
+	vext.8		q1, q1, q1, #4
+	// x2 = shuffle32(x2, MASK(1, 0, 3, 2))
+	vext.8		q2, q2, q2, #8
+	// x3 = shuffle32(x3, MASK(2, 1, 0, 3))
+	vext.8		q3, q3, q3, #12
+
+	// x0 += x1, x3 = rotl32(x3 ^ x0, 16)
+	vadd.i32	q0, q0, q1
+	veor		q4, q3, q0
+	vshl.u32	q3, q4, #16
+	vsri.u32	q3, q4, #16
+
+	// x2 += x3, x1 = rotl32(x1 ^ x2, 12)
+	vadd.i32	q2, q2, q3
+	veor		q4, q1, q2
+	vshl.u32	q1, q4, #12
+	vsri.u32	q1, q4, #20
+
+	// x0 += x1, x3 = rotl32(x3 ^ x0, 8)
+	vadd.i32	q0, q0, q1
+	veor		q4, q3, q0
+	vshl.u32	q3, q4, #8
+	vsri.u32	q3, q4, #24
+
+	// x2 += x3, x1 = rotl32(x1 ^ x2, 7)
+	vadd.i32	q2, q2, q3
+	veor		q4, q1, q2
+	vshl.u32	q1, q4, #7
+	vsri.u32	q1, q4, #25
+
+	// x1 = shuffle32(x1, MASK(2, 1, 0, 3))
+	vext.8		q1, q1, q1, #12
+	// x2 = shuffle32(x2, MASK(1, 0, 3, 2))
+	vext.8		q2, q2, q2, #8
+	// x3 = shuffle32(x3, MASK(0, 3, 2, 1))
+	vext.8		q3, q3, q3, #4
+
+	subs		r3, r3, #1
+	bne		.Ldoubleround
+
+	add		ip, r2, #0x20
+	vld1.8		{q4-q5}, [r2]
+	vld1.8		{q6-q7}, [ip]
+
+	// o0 = i0 ^ (x0 + s0)
+	vadd.i32	q0, q0, q8
+	veor		q0, q0, q4
+
+	// o1 = i1 ^ (x1 + s1)
+	vadd.i32	q1, q1, q9
+	veor		q1, q1, q5
+
+	// o2 = i2 ^ (x2 + s2)
+	vadd.i32	q2, q2, q10
+	veor		q2, q2, q6
+
+	// o3 = i3 ^ (x3 + s3)
+	vadd.i32	q3, q3, q11
+	veor		q3, q3, q7
+
+	add		ip, r1, #0x20
+	vst1.8		{q0-q1}, [r1]
+	vst1.8		{q2-q3}, [ip]
+
+	bx		lr
+ENDPROC(chacha20_block_xor_neon)
+
+	.align		5
+ENTRY(chacha20_4block_xor_neon)
+	push		{r4-r6, lr}
+	mov		ip, sp			// preserve the stack pointer
+	sub		r3, sp, #0x20		// allocate a 32 byte buffer
+	bic		r3, r3, #0x1f		// aligned to 32 bytes
+	mov		sp, r3
+
+	// r0: Input state matrix, s
+	// r1: 4 data blocks output, o
+	// r2: 4 data blocks input, i
+
+	//
+	// This function encrypts four consecutive ChaCha20 blocks by loading
+	// the state matrix in NEON registers four times. The algorithm performs
+	// each operation on the corresponding word of each state matrix, hence
+	// requires no word shuffling. For final XORing step we transpose the
+	// matrix by interleaving 32- and then 64-bit words, which allows us to
+	// do XOR in NEON registers.
+	//
+
+	// x0..15[0-3] = s0..3[0..3]
+	add		r3, r0, #0x20
+	vld1.32		{q0-q1}, [r0]
+	vld1.32		{q2-q3}, [r3]
+
+	adr		r3, CTRINC
+	vdup.32		q15, d7[1]
+	vdup.32		q14, d7[0]
+	vld1.32		{q11}, [r3, :128]
+	vdup.32		q13, d6[1]
+	vdup.32		q12, d6[0]
+	vadd.i32	q12, q12, q11		// x12 += counter values 0-3
+	vdup.32		q11, d5[1]
+	vdup.32		q10, d5[0]
+	vdup.32		q9, d4[1]
+	vdup.32		q8, d4[0]
+	vdup.32		q7, d3[1]
+	vdup.32		q6, d3[0]
+	vdup.32		q5, d2[1]
+	vdup.32		q4, d2[0]
+	vdup.32		q3, d1[1]
+	vdup.32		q2, d1[0]
+	vdup.32		q1, d0[1]
+	vdup.32		q0, d0[0]
+
+	mov		r3, #10
+
+.Ldoubleround4:
+	// x0 += x4, x12 = rotl32(x12 ^ x0, 16)
+	// x1 += x5, x13 = rotl32(x13 ^ x1, 16)
+	// x2 += x6, x14 = rotl32(x14 ^ x2, 16)
+	// x3 += x7, x15 = rotl32(x15 ^ x3, 16)
+	vadd.i32	q0, q0, q4
+	vadd.i32	q1, q1, q5
+	vadd.i32	q2, q2, q6
+	vadd.i32	q3, q3, q7
+
+	veor		q12, q12, q0
+	veor		q13, q13, q1
+	veor		q14, q14, q2
+	veor		q15, q15, q3
+
+	vrev32.16	q12, q12
+	vrev32.16	q13, q13
+	vrev32.16	q14, q14
+	vrev32.16	q15, q15
+
+	// x8 += x12, x4 = rotl32(x4 ^ x8, 12)
+	// x9 += x13, x5 = rotl32(x5 ^ x9, 12)
+	// x10 += x14, x6 = rotl32(x6 ^ x10, 12)
+	// x11 += x15, x7 = rotl32(x7 ^ x11, 12)
+	vadd.i32	q8, q8, q12
+	vadd.i32	q9, q9, q13
+	vadd.i32	q10, q10, q14
+	vadd.i32	q11, q11, q15
+
+	vst1.32		{q8-q9}, [sp, :256]
+
+	veor		q8, q4, q8
+	veor		q9, q5, q9
+	vshl.u32	q4, q8, #12
+	vshl.u32	q5, q9, #12
+	vsri.u32	q4, q8, #20
+	vsri.u32	q5, q9, #20
+
+	veor		q8, q6, q10
+	veor		q9, q7, q11
+	vshl.u32	q6, q8, #12
+	vshl.u32	q7, q9, #12
+	vsri.u32	q6, q8, #20
+	vsri.u32	q7, q9, #20
+
+	// x0 += x4, x12 = rotl32(x12 ^ x0, 8)
+	// x1 += x5, x13 = rotl32(x13 ^ x1, 8)
+	// x2 += x6, x14 = rotl32(x14 ^ x2, 8)
+	// x3 += x7, x15 = rotl32(x15 ^ x3, 8)
+	vadd.i32	q0, q0, q4
+	vadd.i32	q1, q1, q5
+	vadd.i32	q2, q2, q6
+	vadd.i32	q3, q3, q7
+
+	veor		q8, q12, q0
+	veor		q9, q13, q1
+	vshl.u32	q12, q8, #8
+	vshl.u32	q13, q9, #8
+	vsri.u32	q12, q8, #24
+	vsri.u32	q13, q9, #24
+
+	veor		q8, q14, q2
+	veor		q9, q15, q3
+	vshl.u32	q14, q8, #8
+	vshl.u32	q15, q9, #8
+	vsri.u32	q14, q8, #24
+	vsri.u32	q15, q9, #24
+
+	vld1.32		{q8-q9}, [sp, :256]
+
+	// x8 += x12, x4 = rotl32(x4 ^ x8, 7)
+	// x9 += x13, x5 = rotl32(x5 ^ x9, 7)
+	// x10 += x14, x6 = rotl32(x6 ^ x10, 7)
+	// x11 += x15, x7 = rotl32(x7 ^ x11, 7)
+	vadd.i32	q8, q8, q12
+	vadd.i32	q9, q9, q13
+	vadd.i32	q10, q10, q14
+	vadd.i32	q11, q11, q15
+
+	vst1.32		{q8-q9}, [sp, :256]
+
+	veor		q8, q4, q8
+	veor		q9, q5, q9
+	vshl.u32	q4, q8, #7
+	vshl.u32	q5, q9, #7
+	vsri.u32	q4, q8, #25
+	vsri.u32	q5, q9, #25
+
+	veor		q8, q6, q10
+	veor		q9, q7, q11
+	vshl.u32	q6, q8, #7
+	vshl.u32	q7, q9, #7
+	vsri.u32	q6, q8, #25
+	vsri.u32	q7, q9, #25
+
+	vld1.32		{q8-q9}, [sp, :256]
+
+	// x0 += x5, x15 = rotl32(x15 ^ x0, 16)
+	// x1 += x6, x12 = rotl32(x12 ^ x1, 16)
+	// x2 += x7, x13 = rotl32(x13 ^ x2, 16)
+	// x3 += x4, x14 = rotl32(x14 ^ x3, 16)
+	vadd.i32	q0, q0, q5
+	vadd.i32	q1, q1, q6
+	vadd.i32	q2, q2, q7
+	vadd.i32	q3, q3, q4
+
+	veor		q15, q15, q0
+	veor		q12, q12, q1
+	veor		q13, q13, q2
+	veor		q14, q14, q3
+
+	vrev32.16	q15, q15
+	vrev32.16	q12, q12
+	vrev32.16	q13, q13
+	vrev32.16	q14, q14
+
+	// x10 += x15, x5 = rotl32(x5 ^ x10, 12)
+	// x11 += x12, x6 = rotl32(x6 ^ x11, 12)
+	// x8 += x13, x7 = rotl32(x7 ^ x8, 12)
+	// x9 += x14, x4 = rotl32(x4 ^ x9, 12)
+	vadd.i32	q10, q10, q15
+	vadd.i32	q11, q11, q12
+	vadd.i32	q8, q8, q13
+	vadd.i32	q9, q9, q14
+
+	vst1.32		{q8-q9}, [sp, :256]
+
+	veor		q8, q7, q8
+	veor		q9, q4, q9
+	vshl.u32	q7, q8, #12
+	vshl.u32	q4, q9, #12
+	vsri.u32	q7, q8, #20
+	vsri.u32	q4, q9, #20
+
+	veor		q8, q5, q10
+	veor		q9, q6, q11
+	vshl.u32	q5, q8, #12
+	vshl.u32	q6, q9, #12
+	vsri.u32	q5, q8, #20
+	vsri.u32	q6, q9, #20
+
+	// x0 += x5, x15 = rotl32(x15 ^ x0, 8)
+	// x1 += x6, x12 = rotl32(x12 ^ x1, 8)
+	// x2 += x7, x13 = rotl32(x13 ^ x2, 8)
+	// x3 += x4, x14 = rotl32(x14 ^ x3, 8)
+	vadd.i32	q0, q0, q5
+	vadd.i32	q1, q1, q6
+	vadd.i32	q2, q2, q7
+	vadd.i32	q3, q3, q4
+
+	veor		q8, q15, q0
+	veor		q9, q12, q1
+	vshl.u32	q15, q8, #8
+	vshl.u32	q12, q9, #8
+	vsri.u32	q15, q8, #24
+	vsri.u32	q12, q9, #24
+
+	veor		q8, q13, q2
+	veor		q9, q14, q3
+	vshl.u32	q13, q8, #8
+	vshl.u32	q14, q9, #8
+	vsri.u32	q13, q8, #24
+	vsri.u32	q14, q9, #24
+
+	vld1.32		{q8-q9}, [sp, :256]
+
+	// x10 += x15, x5 = rotl32(x5 ^ x10, 7)
+	// x11 += x12, x6 = rotl32(x6 ^ x11, 7)
+	// x8 += x13, x7 = rotl32(x7 ^ x8, 7)
+	// x9 += x14, x4 = rotl32(x4 ^ x9, 7)
+	vadd.i32	q10, q10, q15
+	vadd.i32	q11, q11, q12
+	vadd.i32	q8, q8, q13
+	vadd.i32	q9, q9, q14
+
+	vst1.32		{q8-q9}, [sp, :256]
+
+	veor		q8, q7, q8
+	veor		q9, q4, q9
+	vshl.u32	q7, q8, #7
+	vshl.u32	q4, q9, #7
+	vsri.u32	q7, q8, #25
+	vsri.u32	q4, q9, #25
+
+	veor		q8, q5, q10
+	veor		q9, q6, q11
+	vshl.u32	q5, q8, #7
+	vshl.u32	q6, q9, #7
+	vsri.u32	q5, q8, #25
+	vsri.u32	q6, q9, #25
+
+	subs		r3, r3, #1
+	beq		0f
+
+	vld1.32		{q8-q9}, [sp, :256]
+	b		.Ldoubleround4
+
+	// x0[0-3] += s0[0]
+	// x1[0-3] += s0[1]
+	// x2[0-3] += s0[2]
+	// x3[0-3] += s0[3]
+0:	ldmia		r0!, {r3-r6}
+	vdup.32		q8, r3
+	vdup.32		q9, r4
+	vadd.i32	q0, q0, q8
+	vadd.i32	q1, q1, q9
+	vdup.32		q8, r5
+	vdup.32		q9, r6
+	vadd.i32	q2, q2, q8
+	vadd.i32	q3, q3, q9
+
+	// x4[0-3] += s1[0]
+	// x5[0-3] += s1[1]
+	// x6[0-3] += s1[2]
+	// x7[0-3] += s1[3]
+	ldmia		r0!, {r3-r6}
+	vdup.32		q8, r3
+	vdup.32		q9, r4
+	vadd.i32	q4, q4, q8
+	vadd.i32	q5, q5, q9
+	vdup.32		q8, r5
+	vdup.32		q9, r6
+	vadd.i32	q6, q6, q8
+	vadd.i32	q7, q7, q9
+
+	// interleave 32-bit words in state n, n+1
+	vzip.32		q0, q1
+	vzip.32		q2, q3
+	vzip.32		q4, q5
+	vzip.32		q6, q7
+
+	// interleave 64-bit words in state n, n+2
+	vswp		d1, d4
+	vswp		d3, d6
+	vswp		d9, d12
+	vswp		d11, d14
+
+	// xor with corresponding input, write to output
+	vld1.8		{q8-q9}, [r2]!
+	veor		q8, q8, q0
+	veor		q9, q9, q4
+	vst1.8		{q8-q9}, [r1]!
+
+	vld1.32		{q8-q9}, [sp, :256]
+
+	// x8[0-3] += s2[0]
+	// x9[0-3] += s2[1]
+	// x10[0-3] += s2[2]
+	// x11[0-3] += s2[3]
+	ldmia		r0!, {r3-r6}
+	vdup.32		q0, r3
+	vdup.32		q4, r4
+	vadd.i32	q8, q8, q0
+	vadd.i32	q9, q9, q4
+	vdup.32		q0, r5
+	vdup.32		q4, r6
+	vadd.i32	q10, q10, q0
+	vadd.i32	q11, q11, q4
+
+	// x12[0-3] += s3[0]
+	// x13[0-3] += s3[1]
+	// x14[0-3] += s3[2]
+	// x15[0-3] += s3[3]
+	ldmia		r0!, {r3-r6}
+	vdup.32		q0, r3
+	vdup.32		q4, r4
+	adr		r3, CTRINC
+	vadd.i32	q12, q12, q0
+	vld1.32		{q0}, [r3, :128]
+	vadd.i32	q13, q13, q4
+	vadd.i32	q12, q12, q0		// x12 += counter values 0-3
+
+	vdup.32		q0, r5
+	vdup.32		q4, r6
+	vadd.i32	q14, q14, q0
+	vadd.i32	q15, q15, q4
+
+	// interleave 32-bit words in state n, n+1
+	vzip.32		q8, q9
+	vzip.32		q10, q11
+	vzip.32		q12, q13
+	vzip.32		q14, q15
+
+	// interleave 64-bit words in state n, n+2
+	vswp		d17, d20
+	vswp		d19, d22
+	vswp		d25, d28
+	vswp		d27, d30
+
+	vmov		q4, q1
+
+	vld1.8		{q0-q1}, [r2]!
+	veor		q0, q0, q8
+	veor		q1, q1, q12
+	vst1.8		{q0-q1}, [r1]!
+
+	vld1.8		{q0-q1}, [r2]!
+	veor		q0, q0, q2
+	veor		q1, q1, q6
+	vst1.8		{q0-q1}, [r1]!
+
+	vld1.8		{q0-q1}, [r2]!
+	veor		q0, q0, q10
+	veor		q1, q1, q14
+	vst1.8		{q0-q1}, [r1]!
+
+	vld1.8		{q0-q1}, [r2]!
+	veor		q0, q0, q4
+	veor		q1, q1, q5
+	vst1.8		{q0-q1}, [r1]!
+
+	vld1.8		{q0-q1}, [r2]!
+	veor		q0, q0, q9
+	veor		q1, q1, q13
+	vst1.8		{q0-q1}, [r1]!
+
+	vld1.8		{q0-q1}, [r2]!
+	veor		q0, q0, q3
+	veor		q1, q1, q7
+	vst1.8		{q0-q1}, [r1]!
+
+	vld1.8		{q0-q1}, [r2]
+	veor		q0, q0, q11
+	veor		q1, q1, q15
+	vst1.8		{q0-q1}, [r1]
+
+	mov		sp, ip
+	pop		{r4-r6, pc}
+ENDPROC(chacha20_4block_xor_neon)
+
+	.align		4
+CTRINC:	.word		0, 1, 2, 3
diff --git a/arch/arm/crypto/chacha20-neon-glue.c b/arch/arm/crypto/chacha20-neon-glue.c
new file mode 100644
index 000000000000..59a7be08e80c
--- /dev/null
+++ b/arch/arm/crypto/chacha20-neon-glue.c
@@ -0,0 +1,127 @@
+/*
+ * ChaCha20 256-bit cipher algorithm, RFC7539, ARM NEON functions
+ *
+ * Copyright (C) 2016 Linaro, Ltd. <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Based on:
+ * ChaCha20 256-bit cipher algorithm, RFC7539, SIMD glue code
+ *
+ * Copyright (C) 2015 Martin Willi
+ *
+ * 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/algapi.h>
+#include <crypto/chacha20.h>
+#include <crypto/internal/skcipher.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+#include <asm/hwcap.h>
+#include <asm/neon.h>
+#include <asm/simd.h>
+
+asmlinkage void chacha20_block_xor_neon(u32 *state, u8 *dst, const u8 *src);
+asmlinkage void chacha20_4block_xor_neon(u32 *state, u8 *dst, const u8 *src);
+
+static void chacha20_doneon(u32 *state, u8 *dst, const u8 *src,
+			    unsigned int bytes)
+{
+	u8 buf[CHACHA20_BLOCK_SIZE];
+
+	while (bytes >= CHACHA20_BLOCK_SIZE * 4) {
+		chacha20_4block_xor_neon(state, dst, src);
+		bytes -= CHACHA20_BLOCK_SIZE * 4;
+		src += CHACHA20_BLOCK_SIZE * 4;
+		dst += CHACHA20_BLOCK_SIZE * 4;
+		state[12] += 4;
+	}
+	while (bytes >= CHACHA20_BLOCK_SIZE) {
+		chacha20_block_xor_neon(state, dst, src);
+		bytes -= CHACHA20_BLOCK_SIZE;
+		src += CHACHA20_BLOCK_SIZE;
+		dst += CHACHA20_BLOCK_SIZE;
+		state[12]++;
+	}
+	if (bytes) {
+		memcpy(buf, src, bytes);
+		chacha20_block_xor_neon(state, buf, buf);
+		memcpy(dst, buf, bytes);
+	}
+}
+
+static int chacha20_neon(struct skcipher_request *req)
+{
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct chacha20_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_walk walk;
+	u32 state[16];
+	int err;
+
+	if (req->cryptlen <= CHACHA20_BLOCK_SIZE || !may_use_simd())
+		return crypto_chacha20_crypt(req);
+
+	err = skcipher_walk_virt(&walk, req, true);
+
+	crypto_chacha20_init(state, ctx, walk.iv);
+
+	kernel_neon_begin();
+	while (walk.nbytes > 0) {
+		unsigned int nbytes = walk.nbytes;
+
+		if (nbytes < walk.total)
+			nbytes = round_down(nbytes, walk.stride);
+
+		chacha20_doneon(state, walk.dst.virt.addr, walk.src.virt.addr,
+				nbytes);
+		err = skcipher_walk_done(&walk, walk.nbytes - nbytes);
+	}
+	kernel_neon_end();
+
+	return err;
+}
+
+static struct skcipher_alg alg = {
+	.base.cra_name		= "chacha20",
+	.base.cra_driver_name	= "chacha20-neon",
+	.base.cra_priority	= 300,
+	.base.cra_blocksize	= 1,
+	.base.cra_ctxsize	= sizeof(struct chacha20_ctx),
+	.base.cra_module	= THIS_MODULE,
+
+	.min_keysize		= CHACHA20_KEY_SIZE,
+	.max_keysize		= CHACHA20_KEY_SIZE,
+	.ivsize			= CHACHA20_IV_SIZE,
+	.chunksize		= CHACHA20_BLOCK_SIZE,
+	.walksize		= 4 * CHACHA20_BLOCK_SIZE,
+	.setkey			= crypto_chacha20_setkey,
+	.encrypt		= chacha20_neon,
+	.decrypt		= chacha20_neon,
+};
+
+static int __init chacha20_simd_mod_init(void)
+{
+	if (!(elf_hwcap & HWCAP_NEON))
+		return -ENODEV;
+
+	return crypto_register_skcipher(&alg);
+}
+
+static void __exit chacha20_simd_mod_fini(void)
+{
+	crypto_unregister_skcipher(&alg);
+}
+
+module_init(chacha20_simd_mod_init);
+module_exit(chacha20_simd_mod_fini);
+
+MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS_CRYPTO("chacha20");
diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig
index 33b744d54739..6fc6f5a2a6e5 100644
--- a/arch/arm64/configs/defconfig
+++ b/arch/arm64/configs/defconfig
@@ -516,4 +516,3 @@ CONFIG_CRYPTO_GHASH_ARM64_CE=y
 CONFIG_CRYPTO_AES_ARM64_CE_CCM=y
 CONFIG_CRYPTO_AES_ARM64_CE_BLK=y
 # CONFIG_CRYPTO_AES_ARM64_NEON_BLK is not set
-CONFIG_CRYPTO_CRC32_ARM64=y
diff --git a/arch/arm64/crypto/Kconfig b/arch/arm64/crypto/Kconfig
index 450a85df041a..d92293747d63 100644
--- a/arch/arm64/crypto/Kconfig
+++ b/arch/arm64/crypto/Kconfig
@@ -37,10 +37,14 @@ config CRYPTO_CRCT10DIF_ARM64_CE
 	select CRYPTO_HASH
 
 config CRYPTO_CRC32_ARM64_CE
-	tristate "CRC32 and CRC32C digest algorithms using PMULL instructions"
-	depends on KERNEL_MODE_NEON && CRC32
+	tristate "CRC32 and CRC32C digest algorithms using ARMv8 extensions"
+	depends on CRC32
 	select CRYPTO_HASH
 
+config CRYPTO_AES_ARM64
+	tristate "AES core cipher using scalar instructions"
+	select CRYPTO_AES
+
 config CRYPTO_AES_ARM64_CE
 	tristate "AES core cipher using ARMv8 Crypto Extensions"
 	depends on ARM64 && KERNEL_MODE_NEON
@@ -67,9 +71,17 @@ config CRYPTO_AES_ARM64_NEON_BLK
 	select CRYPTO_AES
 	select CRYPTO_SIMD
 
-config CRYPTO_CRC32_ARM64
-	tristate "CRC32 and CRC32C using optional ARMv8 instructions"
-	depends on ARM64
-	select CRYPTO_HASH
+config CRYPTO_CHACHA20_NEON
+	tristate "NEON accelerated ChaCha20 symmetric cipher"
+	depends on KERNEL_MODE_NEON
+	select CRYPTO_BLKCIPHER
+	select CRYPTO_CHACHA20
+
+config CRYPTO_AES_ARM64_BS
+	tristate "AES in ECB/CBC/CTR/XTS modes using bit-sliced NEON algorithm"
+	depends on KERNEL_MODE_NEON
+	select CRYPTO_BLKCIPHER
+	select CRYPTO_AES_ARM64_NEON_BLK
+	select CRYPTO_SIMD
 
 endif
diff --git a/arch/arm64/crypto/Makefile b/arch/arm64/crypto/Makefile
index aa8888d7b744..b5edc5918c28 100644
--- a/arch/arm64/crypto/Makefile
+++ b/arch/arm64/crypto/Makefile
@@ -41,15 +41,20 @@ sha256-arm64-y := sha256-glue.o sha256-core.o
 obj-$(CONFIG_CRYPTO_SHA512_ARM64) += sha512-arm64.o
 sha512-arm64-y := sha512-glue.o sha512-core.o
 
+obj-$(CONFIG_CRYPTO_CHACHA20_NEON) += chacha20-neon.o
+chacha20-neon-y := chacha20-neon-core.o chacha20-neon-glue.o
+
+obj-$(CONFIG_CRYPTO_AES_ARM64) += aes-arm64.o
+aes-arm64-y := aes-cipher-core.o aes-cipher-glue.o
+
+obj-$(CONFIG_CRYPTO_AES_ARM64_BS) += aes-neon-bs.o
+aes-neon-bs-y := aes-neonbs-core.o aes-neonbs-glue.o
+
 AFLAGS_aes-ce.o		:= -DINTERLEAVE=4
 AFLAGS_aes-neon.o	:= -DINTERLEAVE=4
 
 CFLAGS_aes-glue-ce.o	:= -DUSE_V8_CRYPTO_EXTENSIONS
 
-obj-$(CONFIG_CRYPTO_CRC32_ARM64) += crc32-arm64.o
-
-CFLAGS_crc32-arm64.o	:= -mcpu=generic+crc
-
 $(obj)/aes-glue-%.o: $(src)/aes-glue.c FORCE
 	$(call if_changed_rule,cc_o_c)
 
diff --git a/arch/arm64/crypto/aes-ce-ccm-glue.c b/arch/arm64/crypto/aes-ce-ccm-glue.c
index cc5515dac74a..6a7dbc7c83a6 100644
--- a/arch/arm64/crypto/aes-ce-ccm-glue.c
+++ b/arch/arm64/crypto/aes-ce-ccm-glue.c
@@ -258,7 +258,6 @@ static struct aead_alg ccm_aes_alg = {
 		.cra_priority		= 300,
 		.cra_blocksize		= 1,
 		.cra_ctxsize		= sizeof(struct crypto_aes_ctx),
-		.cra_alignmask		= 7,
 		.cra_module		= THIS_MODULE,
 	},
 	.ivsize		= AES_BLOCK_SIZE,
diff --git a/arch/arm64/crypto/aes-cipher-core.S b/arch/arm64/crypto/aes-cipher-core.S
new file mode 100644
index 000000000000..f2f9cc519309
--- /dev/null
+++ b/arch/arm64/crypto/aes-cipher-core.S
@@ -0,0 +1,110 @@
+/*
+ * Scalar AES core transform
+ *
+ * Copyright (C) 2017 Linaro Ltd <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+	.text
+
+	rk		.req	x0
+	out		.req	x1
+	in		.req	x2
+	rounds		.req	x3
+	tt		.req	x4
+	lt		.req	x2
+
+	.macro		__pair, enc, reg0, reg1, in0, in1e, in1d, shift
+	ubfx		\reg0, \in0, #\shift, #8
+	.if		\enc
+	ubfx		\reg1, \in1e, #\shift, #8
+	.else
+	ubfx		\reg1, \in1d, #\shift, #8
+	.endif
+	ldr		\reg0, [tt, \reg0, uxtw #2]
+	ldr		\reg1, [tt, \reg1, uxtw #2]
+	.endm
+
+	.macro		__hround, out0, out1, in0, in1, in2, in3, t0, t1, enc
+	ldp		\out0, \out1, [rk], #8
+
+	__pair		\enc, w13, w14, \in0, \in1, \in3, 0
+	__pair		\enc, w15, w16, \in1, \in2, \in0, 8
+	__pair		\enc, w17, w18, \in2, \in3, \in1, 16
+	__pair		\enc, \t0, \t1, \in3, \in0, \in2, 24
+
+	eor		\out0, \out0, w13
+	eor		\out1, \out1, w14
+	eor		\out0, \out0, w15, ror #24
+	eor		\out1, \out1, w16, ror #24
+	eor		\out0, \out0, w17, ror #16
+	eor		\out1, \out1, w18, ror #16
+	eor		\out0, \out0, \t0, ror #8
+	eor		\out1, \out1, \t1, ror #8
+	.endm
+
+	.macro		fround, out0, out1, out2, out3, in0, in1, in2, in3
+	__hround	\out0, \out1, \in0, \in1, \in2, \in3, \out2, \out3, 1
+	__hround	\out2, \out3, \in2, \in3, \in0, \in1, \in1, \in2, 1
+	.endm
+
+	.macro		iround, out0, out1, out2, out3, in0, in1, in2, in3
+	__hround	\out0, \out1, \in0, \in3, \in2, \in1, \out2, \out3, 0
+	__hround	\out2, \out3, \in2, \in1, \in0, \in3, \in1, \in0, 0
+	.endm
+
+	.macro		do_crypt, round, ttab, ltab
+	ldp		w5, w6, [in]
+	ldp		w7, w8, [in, #8]
+	ldp		w9, w10, [rk], #16
+	ldp		w11, w12, [rk, #-8]
+
+CPU_BE(	rev		w5, w5		)
+CPU_BE(	rev		w6, w6		)
+CPU_BE(	rev		w7, w7		)
+CPU_BE(	rev		w8, w8		)
+
+	eor		w5, w5, w9
+	eor		w6, w6, w10
+	eor		w7, w7, w11
+	eor		w8, w8, w12
+
+	adr_l		tt, \ttab
+	adr_l		lt, \ltab
+
+	tbnz		rounds, #1, 1f
+
+0:	\round		w9, w10, w11, w12, w5, w6, w7, w8
+	\round		w5, w6, w7, w8, w9, w10, w11, w12
+
+1:	subs		rounds, rounds, #4
+	\round		w9, w10, w11, w12, w5, w6, w7, w8
+	csel		tt, tt, lt, hi
+	\round		w5, w6, w7, w8, w9, w10, w11, w12
+	b.hi		0b
+
+CPU_BE(	rev		w5, w5		)
+CPU_BE(	rev		w6, w6		)
+CPU_BE(	rev		w7, w7		)
+CPU_BE(	rev		w8, w8		)
+
+	stp		w5, w6, [out]
+	stp		w7, w8, [out, #8]
+	ret
+	.endm
+
+	.align		5
+ENTRY(__aes_arm64_encrypt)
+	do_crypt	fround, crypto_ft_tab, crypto_fl_tab
+ENDPROC(__aes_arm64_encrypt)
+
+	.align		5
+ENTRY(__aes_arm64_decrypt)
+	do_crypt	iround, crypto_it_tab, crypto_il_tab
+ENDPROC(__aes_arm64_decrypt)
diff --git a/arch/arm64/crypto/aes-cipher-glue.c b/arch/arm64/crypto/aes-cipher-glue.c
new file mode 100644
index 000000000000..7288e7cbebff
--- /dev/null
+++ b/arch/arm64/crypto/aes-cipher-glue.c
@@ -0,0 +1,69 @@
+/*
+ * Scalar AES core transform
+ *
+ * Copyright (C) 2017 Linaro Ltd <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <crypto/aes.h>
+#include <linux/crypto.h>
+#include <linux/module.h>
+
+asmlinkage void __aes_arm64_encrypt(u32 *rk, u8 *out, const u8 *in, int rounds);
+EXPORT_SYMBOL(__aes_arm64_encrypt);
+
+asmlinkage void __aes_arm64_decrypt(u32 *rk, u8 *out, const u8 *in, int rounds);
+EXPORT_SYMBOL(__aes_arm64_decrypt);
+
+static void aes_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+	struct crypto_aes_ctx *ctx = crypto_tfm_ctx(tfm);
+	int rounds = 6 + ctx->key_length / 4;
+
+	__aes_arm64_encrypt(ctx->key_enc, out, in, rounds);
+}
+
+static void aes_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
+{
+	struct crypto_aes_ctx *ctx = crypto_tfm_ctx(tfm);
+	int rounds = 6 + ctx->key_length / 4;
+
+	__aes_arm64_decrypt(ctx->key_dec, out, in, rounds);
+}
+
+static struct crypto_alg aes_alg = {
+	.cra_name			= "aes",
+	.cra_driver_name		= "aes-arm64",
+	.cra_priority			= 200,
+	.cra_flags			= CRYPTO_ALG_TYPE_CIPHER,
+	.cra_blocksize			= AES_BLOCK_SIZE,
+	.cra_ctxsize			= sizeof(struct crypto_aes_ctx),
+	.cra_module			= THIS_MODULE,
+
+	.cra_cipher.cia_min_keysize	= AES_MIN_KEY_SIZE,
+	.cra_cipher.cia_max_keysize	= AES_MAX_KEY_SIZE,
+	.cra_cipher.cia_setkey		= crypto_aes_set_key,
+	.cra_cipher.cia_encrypt		= aes_encrypt,
+	.cra_cipher.cia_decrypt		= aes_decrypt
+};
+
+static int __init aes_init(void)
+{
+	return crypto_register_alg(&aes_alg);
+}
+
+static void __exit aes_fini(void)
+{
+	crypto_unregister_alg(&aes_alg);
+}
+
+module_init(aes_init);
+module_exit(aes_fini);
+
+MODULE_DESCRIPTION("Scalar AES cipher for arm64");
+MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS_CRYPTO("aes");
diff --git a/arch/arm64/crypto/aes-glue.c b/arch/arm64/crypto/aes-glue.c
index 4e3f8adb1793..bcf596b0197e 100644
--- a/arch/arm64/crypto/aes-glue.c
+++ b/arch/arm64/crypto/aes-glue.c
@@ -1,7 +1,7 @@
 /*
  * linux/arch/arm64/crypto/aes-glue.c - wrapper code for ARMv8 AES
  *
- * Copyright (C) 2013 Linaro Ltd <ard.biesheuvel@linaro.org>
+ * Copyright (C) 2013 - 2017 Linaro Ltd <ard.biesheuvel@linaro.org>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -11,6 +11,7 @@
 #include <asm/neon.h>
 #include <asm/hwcap.h>
 #include <crypto/aes.h>
+#include <crypto/internal/hash.h>
 #include <crypto/internal/simd.h>
 #include <crypto/internal/skcipher.h>
 #include <linux/module.h>
@@ -31,6 +32,7 @@
 #define aes_ctr_encrypt		ce_aes_ctr_encrypt
 #define aes_xts_encrypt		ce_aes_xts_encrypt
 #define aes_xts_decrypt		ce_aes_xts_decrypt
+#define aes_mac_update		ce_aes_mac_update
 MODULE_DESCRIPTION("AES-ECB/CBC/CTR/XTS using ARMv8 Crypto Extensions");
 #else
 #define MODE			"neon"
@@ -44,11 +46,15 @@ MODULE_DESCRIPTION("AES-ECB/CBC/CTR/XTS using ARMv8 Crypto Extensions");
 #define aes_ctr_encrypt		neon_aes_ctr_encrypt
 #define aes_xts_encrypt		neon_aes_xts_encrypt
 #define aes_xts_decrypt		neon_aes_xts_decrypt
+#define aes_mac_update		neon_aes_mac_update
 MODULE_DESCRIPTION("AES-ECB/CBC/CTR/XTS using ARMv8 NEON");
 MODULE_ALIAS_CRYPTO("ecb(aes)");
 MODULE_ALIAS_CRYPTO("cbc(aes)");
 MODULE_ALIAS_CRYPTO("ctr(aes)");
 MODULE_ALIAS_CRYPTO("xts(aes)");
+MODULE_ALIAS_CRYPTO("cmac(aes)");
+MODULE_ALIAS_CRYPTO("xcbc(aes)");
+MODULE_ALIAS_CRYPTO("cbcmac(aes)");
 #endif
 
 MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
@@ -75,11 +81,25 @@ asmlinkage void aes_xts_decrypt(u8 out[], u8 const in[], u8 const rk1[],
 				int rounds, int blocks, u8 const rk2[], u8 iv[],
 				int first);
 
+asmlinkage void aes_mac_update(u8 const in[], u32 const rk[], int rounds,
+			       int blocks, u8 dg[], int enc_before,
+			       int enc_after);
+
 struct crypto_aes_xts_ctx {
 	struct crypto_aes_ctx key1;
 	struct crypto_aes_ctx __aligned(8) key2;
 };
 
+struct mac_tfm_ctx {
+	struct crypto_aes_ctx key;
+	u8 __aligned(8) consts[];
+};
+
+struct mac_desc_ctx {
+	unsigned int len;
+	u8 dg[AES_BLOCK_SIZE];
+};
+
 static int skcipher_aes_setkey(struct crypto_skcipher *tfm, const u8 *in_key,
 			       unsigned int key_len)
 {
@@ -215,14 +235,15 @@ static int ctr_encrypt(struct skcipher_request *req)
 		u8 *tsrc = walk.src.virt.addr;
 
 		/*
-		 * Minimum alignment is 8 bytes, so if nbytes is <= 8, we need
-		 * to tell aes_ctr_encrypt() to only read half a block.
+		 * Tell aes_ctr_encrypt() to process a tail block.
 		 */
-		blocks = (nbytes <= 8) ? -1 : 1;
+		blocks = -1;
 
-		aes_ctr_encrypt(tail, tsrc, (u8 *)ctx->key_enc, rounds,
+		aes_ctr_encrypt(tail, NULL, (u8 *)ctx->key_enc, rounds,
 				blocks, walk.iv, first);
-		memcpy(tdst, tail, nbytes);
+		if (tdst != tsrc)
+			memcpy(tdst, tsrc, nbytes);
+		crypto_xor(tdst, tail, nbytes);
 		err = skcipher_walk_done(&walk, 0);
 	}
 	kernel_neon_end();
@@ -282,7 +303,6 @@ static struct skcipher_alg aes_algs[] = { {
 		.cra_flags		= CRYPTO_ALG_INTERNAL,
 		.cra_blocksize		= AES_BLOCK_SIZE,
 		.cra_ctxsize		= sizeof(struct crypto_aes_ctx),
-		.cra_alignmask		= 7,
 		.cra_module		= THIS_MODULE,
 	},
 	.min_keysize	= AES_MIN_KEY_SIZE,
@@ -298,7 +318,6 @@ static struct skcipher_alg aes_algs[] = { {
 		.cra_flags		= CRYPTO_ALG_INTERNAL,
 		.cra_blocksize		= AES_BLOCK_SIZE,
 		.cra_ctxsize		= sizeof(struct crypto_aes_ctx),
-		.cra_alignmask		= 7,
 		.cra_module		= THIS_MODULE,
 	},
 	.min_keysize	= AES_MIN_KEY_SIZE,
@@ -315,7 +334,22 @@ static struct skcipher_alg aes_algs[] = { {
 		.cra_flags		= CRYPTO_ALG_INTERNAL,
 		.cra_blocksize		= 1,
 		.cra_ctxsize		= sizeof(struct crypto_aes_ctx),
-		.cra_alignmask		= 7,
+		.cra_module		= THIS_MODULE,
+	},
+	.min_keysize	= AES_MIN_KEY_SIZE,
+	.max_keysize	= AES_MAX_KEY_SIZE,
+	.ivsize		= AES_BLOCK_SIZE,
+	.chunksize	= AES_BLOCK_SIZE,
+	.setkey		= skcipher_aes_setkey,
+	.encrypt	= ctr_encrypt,
+	.decrypt	= ctr_encrypt,
+}, {
+	.base = {
+		.cra_name		= "ctr(aes)",
+		.cra_driver_name	= "ctr-aes-" MODE,
+		.cra_priority		= PRIO - 1,
+		.cra_blocksize		= 1,
+		.cra_ctxsize		= sizeof(struct crypto_aes_ctx),
 		.cra_module		= THIS_MODULE,
 	},
 	.min_keysize	= AES_MIN_KEY_SIZE,
@@ -333,7 +367,6 @@ static struct skcipher_alg aes_algs[] = { {
 		.cra_flags		= CRYPTO_ALG_INTERNAL,
 		.cra_blocksize		= AES_BLOCK_SIZE,
 		.cra_ctxsize		= sizeof(struct crypto_aes_xts_ctx),
-		.cra_alignmask		= 7,
 		.cra_module		= THIS_MODULE,
 	},
 	.min_keysize	= 2 * AES_MIN_KEY_SIZE,
@@ -344,15 +377,228 @@ static struct skcipher_alg aes_algs[] = { {
 	.decrypt	= xts_decrypt,
 } };
 
+static int cbcmac_setkey(struct crypto_shash *tfm, const u8 *in_key,
+			 unsigned int key_len)
+{
+	struct mac_tfm_ctx *ctx = crypto_shash_ctx(tfm);
+	int err;
+
+	err = aes_expandkey(&ctx->key, in_key, key_len);
+	if (err)
+		crypto_shash_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
+
+	return err;
+}
+
+static void cmac_gf128_mul_by_x(be128 *y, const be128 *x)
+{
+	u64 a = be64_to_cpu(x->a);
+	u64 b = be64_to_cpu(x->b);
+
+	y->a = cpu_to_be64((a << 1) | (b >> 63));
+	y->b = cpu_to_be64((b << 1) ^ ((a >> 63) ? 0x87 : 0));
+}
+
+static int cmac_setkey(struct crypto_shash *tfm, const u8 *in_key,
+		       unsigned int key_len)
+{
+	struct mac_tfm_ctx *ctx = crypto_shash_ctx(tfm);
+	be128 *consts = (be128 *)ctx->consts;
+	u8 *rk = (u8 *)ctx->key.key_enc;
+	int rounds = 6 + key_len / 4;
+	int err;
+
+	err = cbcmac_setkey(tfm, in_key, key_len);
+	if (err)
+		return err;
+
+	/* encrypt the zero vector */
+	kernel_neon_begin();
+	aes_ecb_encrypt(ctx->consts, (u8[AES_BLOCK_SIZE]){}, rk, rounds, 1, 1);
+	kernel_neon_end();
+
+	cmac_gf128_mul_by_x(consts, consts);
+	cmac_gf128_mul_by_x(consts + 1, consts);
+
+	return 0;
+}
+
+static int xcbc_setkey(struct crypto_shash *tfm, const u8 *in_key,
+		       unsigned int key_len)
+{
+	static u8 const ks[3][AES_BLOCK_SIZE] = {
+		{ [0 ... AES_BLOCK_SIZE - 1] = 0x1 },
+		{ [0 ... AES_BLOCK_SIZE - 1] = 0x2 },
+		{ [0 ... AES_BLOCK_SIZE - 1] = 0x3 },
+	};
+
+	struct mac_tfm_ctx *ctx = crypto_shash_ctx(tfm);
+	u8 *rk = (u8 *)ctx->key.key_enc;
+	int rounds = 6 + key_len / 4;
+	u8 key[AES_BLOCK_SIZE];
+	int err;
+
+	err = cbcmac_setkey(tfm, in_key, key_len);
+	if (err)
+		return err;
+
+	kernel_neon_begin();
+	aes_ecb_encrypt(key, ks[0], rk, rounds, 1, 1);
+	aes_ecb_encrypt(ctx->consts, ks[1], rk, rounds, 2, 0);
+	kernel_neon_end();
+
+	return cbcmac_setkey(tfm, key, sizeof(key));
+}
+
+static int mac_init(struct shash_desc *desc)
+{
+	struct mac_desc_ctx *ctx = shash_desc_ctx(desc);
+
+	memset(ctx->dg, 0, AES_BLOCK_SIZE);
+	ctx->len = 0;
+
+	return 0;
+}
+
+static int mac_update(struct shash_desc *desc, const u8 *p, unsigned int len)
+{
+	struct mac_tfm_ctx *tctx = crypto_shash_ctx(desc->tfm);
+	struct mac_desc_ctx *ctx = shash_desc_ctx(desc);
+	int rounds = 6 + tctx->key.key_length / 4;
+
+	while (len > 0) {
+		unsigned int l;
+
+		if ((ctx->len % AES_BLOCK_SIZE) == 0 &&
+		    (ctx->len + len) > AES_BLOCK_SIZE) {
+
+			int blocks = len / AES_BLOCK_SIZE;
+
+			len %= AES_BLOCK_SIZE;
+
+			kernel_neon_begin();
+			aes_mac_update(p, tctx->key.key_enc, rounds, blocks,
+				       ctx->dg, (ctx->len != 0), (len != 0));
+			kernel_neon_end();
+
+			p += blocks * AES_BLOCK_SIZE;
+
+			if (!len) {
+				ctx->len = AES_BLOCK_SIZE;
+				break;
+			}
+			ctx->len = 0;
+		}
+
+		l = min(len, AES_BLOCK_SIZE - ctx->len);
+
+		if (l <= AES_BLOCK_SIZE) {
+			crypto_xor(ctx->dg + ctx->len, p, l);
+			ctx->len += l;
+			len -= l;
+			p += l;
+		}
+	}
+
+	return 0;
+}
+
+static int cbcmac_final(struct shash_desc *desc, u8 *out)
+{
+	struct mac_tfm_ctx *tctx = crypto_shash_ctx(desc->tfm);
+	struct mac_desc_ctx *ctx = shash_desc_ctx(desc);
+	int rounds = 6 + tctx->key.key_length / 4;
+
+	kernel_neon_begin();
+	aes_mac_update(NULL, tctx->key.key_enc, rounds, 0, ctx->dg, 1, 0);
+	kernel_neon_end();
+
+	memcpy(out, ctx->dg, AES_BLOCK_SIZE);
+
+	return 0;
+}
+
+static int cmac_final(struct shash_desc *desc, u8 *out)
+{
+	struct mac_tfm_ctx *tctx = crypto_shash_ctx(desc->tfm);
+	struct mac_desc_ctx *ctx = shash_desc_ctx(desc);
+	int rounds = 6 + tctx->key.key_length / 4;
+	u8 *consts = tctx->consts;
+
+	if (ctx->len != AES_BLOCK_SIZE) {
+		ctx->dg[ctx->len] ^= 0x80;
+		consts += AES_BLOCK_SIZE;
+	}
+
+	kernel_neon_begin();
+	aes_mac_update(consts, tctx->key.key_enc, rounds, 1, ctx->dg, 0, 1);
+	kernel_neon_end();
+
+	memcpy(out, ctx->dg, AES_BLOCK_SIZE);
+
+	return 0;
+}
+
+static struct shash_alg mac_algs[] = { {
+	.base.cra_name		= "cmac(aes)",
+	.base.cra_driver_name	= "cmac-aes-" MODE,
+	.base.cra_priority	= PRIO,
+	.base.cra_flags		= CRYPTO_ALG_TYPE_SHASH,
+	.base.cra_blocksize	= AES_BLOCK_SIZE,
+	.base.cra_ctxsize	= sizeof(struct mac_tfm_ctx) +
+				  2 * AES_BLOCK_SIZE,
+	.base.cra_module	= THIS_MODULE,
+
+	.digestsize		= AES_BLOCK_SIZE,
+	.init			= mac_init,
+	.update			= mac_update,
+	.final			= cmac_final,
+	.setkey			= cmac_setkey,
+	.descsize		= sizeof(struct mac_desc_ctx),
+}, {
+	.base.cra_name		= "xcbc(aes)",
+	.base.cra_driver_name	= "xcbc-aes-" MODE,
+	.base.cra_priority	= PRIO,
+	.base.cra_flags		= CRYPTO_ALG_TYPE_SHASH,
+	.base.cra_blocksize	= AES_BLOCK_SIZE,
+	.base.cra_ctxsize	= sizeof(struct mac_tfm_ctx) +
+				  2 * AES_BLOCK_SIZE,
+	.base.cra_module	= THIS_MODULE,
+
+	.digestsize		= AES_BLOCK_SIZE,
+	.init			= mac_init,
+	.update			= mac_update,
+	.final			= cmac_final,
+	.setkey			= xcbc_setkey,
+	.descsize		= sizeof(struct mac_desc_ctx),
+}, {
+	.base.cra_name		= "cbcmac(aes)",
+	.base.cra_driver_name	= "cbcmac-aes-" MODE,
+	.base.cra_priority	= PRIO,
+	.base.cra_flags		= CRYPTO_ALG_TYPE_SHASH,
+	.base.cra_blocksize	= 1,
+	.base.cra_ctxsize	= sizeof(struct mac_tfm_ctx),
+	.base.cra_module	= THIS_MODULE,
+
+	.digestsize		= AES_BLOCK_SIZE,
+	.init			= mac_init,
+	.update			= mac_update,
+	.final			= cbcmac_final,
+	.setkey			= cbcmac_setkey,
+	.descsize		= sizeof(struct mac_desc_ctx),
+} };
+
 static struct simd_skcipher_alg *aes_simd_algs[ARRAY_SIZE(aes_algs)];
 
 static void aes_exit(void)
 {
 	int i;
 
-	for (i = 0; i < ARRAY_SIZE(aes_simd_algs) && aes_simd_algs[i]; i++)
-		simd_skcipher_free(aes_simd_algs[i]);
+	for (i = 0; i < ARRAY_SIZE(aes_simd_algs); i++)
+		if (aes_simd_algs[i])
+			simd_skcipher_free(aes_simd_algs[i]);
 
+	crypto_unregister_shashes(mac_algs, ARRAY_SIZE(mac_algs));
 	crypto_unregister_skciphers(aes_algs, ARRAY_SIZE(aes_algs));
 }
 
@@ -369,7 +615,14 @@ static int __init aes_init(void)
 	if (err)
 		return err;
 
+	err = crypto_register_shashes(mac_algs, ARRAY_SIZE(mac_algs));
+	if (err)
+		goto unregister_ciphers;
+
 	for (i = 0; i < ARRAY_SIZE(aes_algs); i++) {
+		if (!(aes_algs[i].base.cra_flags & CRYPTO_ALG_INTERNAL))
+			continue;
+
 		algname = aes_algs[i].base.cra_name + 2;
 		drvname = aes_algs[i].base.cra_driver_name + 2;
 		basename = aes_algs[i].base.cra_driver_name;
@@ -385,6 +638,8 @@ static int __init aes_init(void)
 
 unregister_simds:
 	aes_exit();
+unregister_ciphers:
+	crypto_unregister_skciphers(aes_algs, ARRAY_SIZE(aes_algs));
 	return err;
 }
 
@@ -392,5 +647,7 @@ unregister_simds:
 module_cpu_feature_match(AES, aes_init);
 #else
 module_init(aes_init);
+EXPORT_SYMBOL(neon_aes_ecb_encrypt);
+EXPORT_SYMBOL(neon_aes_cbc_encrypt);
 #endif
 module_exit(aes_exit);
diff --git a/arch/arm64/crypto/aes-modes.S b/arch/arm64/crypto/aes-modes.S
index 838dad5c209f..2674d43d1384 100644
--- a/arch/arm64/crypto/aes-modes.S
+++ b/arch/arm64/crypto/aes-modes.S
@@ -1,7 +1,7 @@
 /*
  * linux/arch/arm64/crypto/aes-modes.S - chaining mode wrappers for AES
  *
- * Copyright (C) 2013 Linaro Ltd <ard.biesheuvel@linaro.org>
+ * Copyright (C) 2013 - 2017 Linaro Ltd <ard.biesheuvel@linaro.org>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -337,7 +337,7 @@ AES_ENTRY(aes_ctr_encrypt)
 
 .Lctrcarrydone:
 	subs		w4, w4, #1
-	bmi		.Lctrhalfblock		/* blocks < 0 means 1/2 block */
+	bmi		.Lctrtailblock		/* blocks <0 means tail block */
 	ld1		{v3.16b}, [x1], #16
 	eor		v3.16b, v0.16b, v3.16b
 	st1		{v3.16b}, [x0], #16
@@ -348,10 +348,8 @@ AES_ENTRY(aes_ctr_encrypt)
 	FRAME_POP
 	ret
 
-.Lctrhalfblock:
-	ld1		{v3.8b}, [x1]
-	eor		v3.8b, v0.8b, v3.8b
-	st1		{v3.8b}, [x0]
+.Lctrtailblock:
+	st1		{v0.16b}, [x0]
 	FRAME_POP
 	ret
 
@@ -527,3 +525,30 @@ AES_ENTRY(aes_xts_decrypt)
 	FRAME_POP
 	ret
 AES_ENDPROC(aes_xts_decrypt)
+
+	/*
+	 * aes_mac_update(u8 const in[], u32 const rk[], int rounds,
+	 *		  int blocks, u8 dg[], int enc_before, int enc_after)
+	 */
+AES_ENTRY(aes_mac_update)
+	ld1		{v0.16b}, [x4]			/* get dg */
+	enc_prepare	w2, x1, x7
+	cbnz		w5, .Lmacenc
+
+.Lmacloop:
+	cbz		w3, .Lmacout
+	ld1		{v1.16b}, [x0], #16		/* get next pt block */
+	eor		v0.16b, v0.16b, v1.16b		/* ..and xor with dg */
+
+	subs		w3, w3, #1
+	csinv		x5, x6, xzr, eq
+	cbz		w5, .Lmacout
+
+.Lmacenc:
+	encrypt_block	v0, w2, x1, x7, w8
+	b		.Lmacloop
+
+.Lmacout:
+	st1		{v0.16b}, [x4]			/* return dg */
+	ret
+AES_ENDPROC(aes_mac_update)
diff --git a/arch/arm64/crypto/aes-neon.S b/arch/arm64/crypto/aes-neon.S
index 85f07ead7c5c..f1e3aa2732f9 100644
--- a/arch/arm64/crypto/aes-neon.S
+++ b/arch/arm64/crypto/aes-neon.S
@@ -1,7 +1,7 @@
 /*
  * linux/arch/arm64/crypto/aes-neon.S - AES cipher for ARMv8 NEON
  *
- * Copyright (C) 2013 Linaro Ltd <ard.biesheuvel@linaro.org>
+ * Copyright (C) 2013 - 2017 Linaro Ltd. <ard.biesheuvel@linaro.org>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -17,17 +17,25 @@
 	/* multiply by polynomial 'x' in GF(2^8) */
 	.macro		mul_by_x, out, in, temp, const
 	sshr		\temp, \in, #7
-	add		\out, \in, \in
+	shl		\out, \in, #1
 	and		\temp, \temp, \const
 	eor		\out, \out, \temp
 	.endm
 
+	/* multiply by polynomial 'x^2' in GF(2^8) */
+	.macro		mul_by_x2, out, in, temp, const
+	ushr		\temp, \in, #6
+	shl		\out, \in, #2
+	pmul		\temp, \temp, \const
+	eor		\out, \out, \temp
+	.endm
+
 	/* preload the entire Sbox */
 	.macro		prepare, sbox, shiftrows, temp
 	adr		\temp, \sbox
-	movi		v12.16b, #0x40
+	movi		v12.16b, #0x1b
 	ldr		q13, \shiftrows
-	movi		v14.16b, #0x1b
+	ldr		q14, .Lror32by8
 	ld1		{v16.16b-v19.16b}, [\temp], #64
 	ld1		{v20.16b-v23.16b}, [\temp], #64
 	ld1		{v24.16b-v27.16b}, [\temp], #64
@@ -50,37 +58,31 @@
 
 	/* apply SubBytes transformation using the the preloaded Sbox */
 	.macro		sub_bytes, in
-	sub		v9.16b, \in\().16b, v12.16b
+	sub		v9.16b, \in\().16b, v15.16b
 	tbl		\in\().16b, {v16.16b-v19.16b}, \in\().16b
-	sub		v10.16b, v9.16b, v12.16b
+	sub		v10.16b, v9.16b, v15.16b
 	tbx		\in\().16b, {v20.16b-v23.16b}, v9.16b
-	sub		v11.16b, v10.16b, v12.16b
+	sub		v11.16b, v10.16b, v15.16b
 	tbx		\in\().16b, {v24.16b-v27.16b}, v10.16b
 	tbx		\in\().16b, {v28.16b-v31.16b}, v11.16b
 	.endm
 
 	/* apply MixColumns transformation */
-	.macro		mix_columns, in
-	mul_by_x	v10.16b, \in\().16b, v9.16b, v14.16b
-	rev32		v8.8h, \in\().8h
-	eor		\in\().16b, v10.16b, \in\().16b
-	shl		v9.4s, v8.4s, #24
-	shl		v11.4s, \in\().4s, #24
-	sri		v9.4s, v8.4s, #8
-	sri		v11.4s, \in\().4s, #8
-	eor		v9.16b, v9.16b, v8.16b
-	eor		v10.16b, v10.16b, v9.16b
-	eor		\in\().16b, v10.16b, v11.16b
-	.endm
-
+	.macro		mix_columns, in, enc
+	.if		\enc == 0
 	/* Inverse MixColumns: pre-multiply by { 5, 0, 4, 0 } */
-	.macro		inv_mix_columns, in
-	mul_by_x	v11.16b, \in\().16b, v10.16b, v14.16b
-	mul_by_x	v11.16b, v11.16b, v10.16b, v14.16b
-	eor		\in\().16b, \in\().16b, v11.16b
-	rev32		v11.8h, v11.8h
-	eor		\in\().16b, \in\().16b, v11.16b
-	mix_columns	\in
+	mul_by_x2	v8.16b, \in\().16b, v9.16b, v12.16b
+	eor		\in\().16b, \in\().16b, v8.16b
+	rev32		v8.8h, v8.8h
+	eor		\in\().16b, \in\().16b, v8.16b
+	.endif
+
+	mul_by_x	v9.16b, \in\().16b, v8.16b, v12.16b
+	rev32		v8.8h, \in\().8h
+	eor		v8.16b, v8.16b, v9.16b
+	eor		\in\().16b, \in\().16b, v8.16b
+	tbl		\in\().16b, {\in\().16b}, v14.16b
+	eor		\in\().16b, \in\().16b, v8.16b
 	.endm
 
 	.macro		do_block, enc, in, rounds, rk, rkp, i
@@ -88,16 +90,13 @@
 	add		\rkp, \rk, #16
 	mov		\i, \rounds
 1111:	eor		\in\().16b, \in\().16b, v15.16b		/* ^round key */
+	movi		v15.16b, #0x40
 	tbl		\in\().16b, {\in\().16b}, v13.16b	/* ShiftRows */
 	sub_bytes	\in
-	ld1		{v15.4s}, [\rkp], #16
 	subs		\i, \i, #1
+	ld1		{v15.4s}, [\rkp], #16
 	beq		2222f
-	.if		\enc == 1
-	mix_columns	\in
-	.else
-	inv_mix_columns	\in
-	.endif
+	mix_columns	\in, \enc
 	b		1111b
 2222:	eor		\in\().16b, \in\().16b, v15.16b		/* ^round key */
 	.endm
@@ -116,139 +115,114 @@
 	 */
 
 	.macro		sub_bytes_2x, in0, in1
-	sub		v8.16b, \in0\().16b, v12.16b
-	sub		v9.16b, \in1\().16b, v12.16b
+	sub		v8.16b, \in0\().16b, v15.16b
 	tbl		\in0\().16b, {v16.16b-v19.16b}, \in0\().16b
+	sub		v9.16b, \in1\().16b, v15.16b
 	tbl		\in1\().16b, {v16.16b-v19.16b}, \in1\().16b
-	sub		v10.16b, v8.16b, v12.16b
-	sub		v11.16b, v9.16b, v12.16b
+	sub		v10.16b, v8.16b, v15.16b
 	tbx		\in0\().16b, {v20.16b-v23.16b}, v8.16b
+	sub		v11.16b, v9.16b, v15.16b
 	tbx		\in1\().16b, {v20.16b-v23.16b}, v9.16b
-	sub		v8.16b, v10.16b, v12.16b
-	sub		v9.16b, v11.16b, v12.16b
+	sub		v8.16b, v10.16b, v15.16b
 	tbx		\in0\().16b, {v24.16b-v27.16b}, v10.16b
+	sub		v9.16b, v11.16b, v15.16b
 	tbx		\in1\().16b, {v24.16b-v27.16b}, v11.16b
 	tbx		\in0\().16b, {v28.16b-v31.16b}, v8.16b
 	tbx		\in1\().16b, {v28.16b-v31.16b}, v9.16b
 	.endm
 
 	.macro		sub_bytes_4x, in0, in1, in2, in3
-	sub		v8.16b, \in0\().16b, v12.16b
+	sub		v8.16b, \in0\().16b, v15.16b
 	tbl		\in0\().16b, {v16.16b-v19.16b}, \in0\().16b
-	sub		v9.16b, \in1\().16b, v12.16b
+	sub		v9.16b, \in1\().16b, v15.16b
 	tbl		\in1\().16b, {v16.16b-v19.16b}, \in1\().16b
-	sub		v10.16b, \in2\().16b, v12.16b
+	sub		v10.16b, \in2\().16b, v15.16b
 	tbl		\in2\().16b, {v16.16b-v19.16b}, \in2\().16b
-	sub		v11.16b, \in3\().16b, v12.16b
+	sub		v11.16b, \in3\().16b, v15.16b
 	tbl		\in3\().16b, {v16.16b-v19.16b}, \in3\().16b
 	tbx		\in0\().16b, {v20.16b-v23.16b}, v8.16b
 	tbx		\in1\().16b, {v20.16b-v23.16b}, v9.16b
-	sub		v8.16b, v8.16b, v12.16b
+	sub		v8.16b, v8.16b, v15.16b
 	tbx		\in2\().16b, {v20.16b-v23.16b}, v10.16b
-	sub		v9.16b, v9.16b, v12.16b
+	sub		v9.16b, v9.16b, v15.16b
 	tbx		\in3\().16b, {v20.16b-v23.16b}, v11.16b
-	sub		v10.16b, v10.16b, v12.16b
+	sub		v10.16b, v10.16b, v15.16b
 	tbx		\in0\().16b, {v24.16b-v27.16b}, v8.16b
-	sub		v11.16b, v11.16b, v12.16b
+	sub		v11.16b, v11.16b, v15.16b
 	tbx		\in1\().16b, {v24.16b-v27.16b}, v9.16b
-	sub		v8.16b, v8.16b, v12.16b
+	sub		v8.16b, v8.16b, v15.16b
 	tbx		\in2\().16b, {v24.16b-v27.16b}, v10.16b
-	sub		v9.16b, v9.16b, v12.16b
+	sub		v9.16b, v9.16b, v15.16b
 	tbx		\in3\().16b, {v24.16b-v27.16b}, v11.16b
-	sub		v10.16b, v10.16b, v12.16b
+	sub		v10.16b, v10.16b, v15.16b
 	tbx		\in0\().16b, {v28.16b-v31.16b}, v8.16b
-	sub		v11.16b, v11.16b, v12.16b
+	sub		v11.16b, v11.16b, v15.16b
 	tbx		\in1\().16b, {v28.16b-v31.16b}, v9.16b
 	tbx		\in2\().16b, {v28.16b-v31.16b}, v10.16b
 	tbx		\in3\().16b, {v28.16b-v31.16b}, v11.16b
 	.endm
 
 	.macro		mul_by_x_2x, out0, out1, in0, in1, tmp0, tmp1, const
-	sshr		\tmp0\().16b, \in0\().16b,  #7
-	add		\out0\().16b, \in0\().16b,  \in0\().16b
-	sshr		\tmp1\().16b, \in1\().16b,  #7
+	sshr		\tmp0\().16b, \in0\().16b, #7
+	shl		\out0\().16b, \in0\().16b, #1
+	sshr		\tmp1\().16b, \in1\().16b, #7
 	and		\tmp0\().16b, \tmp0\().16b, \const\().16b
-	add		\out1\().16b, \in1\().16b,  \in1\().16b
+	shl		\out1\().16b, \in1\().16b, #1
 	and		\tmp1\().16b, \tmp1\().16b, \const\().16b
 	eor		\out0\().16b, \out0\().16b, \tmp0\().16b
 	eor		\out1\().16b, \out1\().16b, \tmp1\().16b
 	.endm
 
-	.macro		mix_columns_2x, in0, in1
-	mul_by_x_2x	v8, v9, \in0, \in1, v10, v11, v14
-	rev32		v10.8h, \in0\().8h
-	rev32		v11.8h, \in1\().8h
-	eor		\in0\().16b, v8.16b, \in0\().16b
-	eor		\in1\().16b, v9.16b, \in1\().16b
-	shl		v12.4s, v10.4s, #24
-	shl		v13.4s, v11.4s, #24
-	eor		v8.16b, v8.16b, v10.16b
-	sri		v12.4s, v10.4s, #8
-	shl		v10.4s, \in0\().4s, #24
-	eor		v9.16b, v9.16b, v11.16b
-	sri		v13.4s, v11.4s, #8
-	shl		v11.4s, \in1\().4s, #24
-	sri		v10.4s, \in0\().4s, #8
-	eor		\in0\().16b, v8.16b, v12.16b
-	sri		v11.4s, \in1\().4s, #8
-	eor		\in1\().16b, v9.16b, v13.16b
-	eor		\in0\().16b, v10.16b, \in0\().16b
-	eor		\in1\().16b, v11.16b, \in1\().16b
+	.macro		mul_by_x2_2x, out0, out1, in0, in1, tmp0, tmp1, const
+	ushr		\tmp0\().16b, \in0\().16b, #6
+	shl		\out0\().16b, \in0\().16b, #2
+	ushr		\tmp1\().16b, \in1\().16b, #6
+	pmul		\tmp0\().16b, \tmp0\().16b, \const\().16b
+	shl		\out1\().16b, \in1\().16b, #2
+	pmul		\tmp1\().16b, \tmp1\().16b, \const\().16b
+	eor		\out0\().16b, \out0\().16b, \tmp0\().16b
+	eor		\out1\().16b, \out1\().16b, \tmp1\().16b
 	.endm
 
-	.macro		inv_mix_cols_2x, in0, in1
-	mul_by_x_2x	v8, v9, \in0, \in1, v10, v11, v14
-	mul_by_x_2x	v8, v9, v8, v9, v10, v11, v14
+	.macro		mix_columns_2x, in0, in1, enc
+	.if		\enc == 0
+	/* Inverse MixColumns: pre-multiply by { 5, 0, 4, 0 } */
+	mul_by_x2_2x	v8, v9, \in0, \in1, v10, v11, v12
 	eor		\in0\().16b, \in0\().16b, v8.16b
-	eor		\in1\().16b, \in1\().16b, v9.16b
 	rev32		v8.8h, v8.8h
-	rev32		v9.8h, v9.8h
-	eor		\in0\().16b, \in0\().16b, v8.16b
-	eor		\in1\().16b, \in1\().16b, v9.16b
-	mix_columns_2x	\in0, \in1
-	.endm
-
-	.macro		inv_mix_cols_4x, in0, in1, in2, in3
-	mul_by_x_2x	v8, v9, \in0, \in1, v10, v11, v14
-	mul_by_x_2x	v10, v11, \in2, \in3, v12, v13, v14
-	mul_by_x_2x	v8, v9, v8, v9, v12, v13, v14
-	mul_by_x_2x	v10, v11, v10, v11, v12, v13, v14
-	eor		\in0\().16b, \in0\().16b, v8.16b
 	eor		\in1\().16b, \in1\().16b, v9.16b
-	eor		\in2\().16b, \in2\().16b, v10.16b
-	eor		\in3\().16b, \in3\().16b, v11.16b
-	rev32		v8.8h, v8.8h
 	rev32		v9.8h, v9.8h
-	rev32		v10.8h, v10.8h
-	rev32		v11.8h, v11.8h
 	eor		\in0\().16b, \in0\().16b, v8.16b
 	eor		\in1\().16b, \in1\().16b, v9.16b
-	eor		\in2\().16b, \in2\().16b, v10.16b
-	eor		\in3\().16b, \in3\().16b, v11.16b
-	mix_columns_2x	\in0, \in1
-	mix_columns_2x	\in2, \in3
+	.endif
+
+	mul_by_x_2x	v8, v9, \in0, \in1, v10, v11, v12
+	rev32		v10.8h, \in0\().8h
+	rev32		v11.8h, \in1\().8h
+	eor		v10.16b, v10.16b, v8.16b
+	eor		v11.16b, v11.16b, v9.16b
+	eor		\in0\().16b, \in0\().16b, v10.16b
+	eor		\in1\().16b, \in1\().16b, v11.16b
+	tbl		\in0\().16b, {\in0\().16b}, v14.16b
+	tbl		\in1\().16b, {\in1\().16b}, v14.16b
+	eor		\in0\().16b, \in0\().16b, v10.16b
+	eor		\in1\().16b, \in1\().16b, v11.16b
 	.endm
 
-	.macro		do_block_2x, enc, in0, in1 rounds, rk, rkp, i
+	.macro		do_block_2x, enc, in0, in1, rounds, rk, rkp, i
 	ld1		{v15.4s}, [\rk]
 	add		\rkp, \rk, #16
 	mov		\i, \rounds
 1111:	eor		\in0\().16b, \in0\().16b, v15.16b	/* ^round key */
 	eor		\in1\().16b, \in1\().16b, v15.16b	/* ^round key */
-	sub_bytes_2x	\in0, \in1
+	movi		v15.16b, #0x40
 	tbl		\in0\().16b, {\in0\().16b}, v13.16b	/* ShiftRows */
 	tbl		\in1\().16b, {\in1\().16b}, v13.16b	/* ShiftRows */
-	ld1		{v15.4s}, [\rkp], #16
+	sub_bytes_2x	\in0, \in1
 	subs		\i, \i, #1
+	ld1		{v15.4s}, [\rkp], #16
 	beq		2222f
-	.if		\enc == 1
-	mix_columns_2x	\in0, \in1
-	ldr		q13, .LForward_ShiftRows
-	.else
-	inv_mix_cols_2x	\in0, \in1
-	ldr		q13, .LReverse_ShiftRows
-	.endif
-	movi		v12.16b, #0x40
+	mix_columns_2x	\in0, \in1, \enc
 	b		1111b
 2222:	eor		\in0\().16b, \in0\().16b, v15.16b	/* ^round key */
 	eor		\in1\().16b, \in1\().16b, v15.16b	/* ^round key */
@@ -262,23 +236,17 @@
 	eor		\in1\().16b, \in1\().16b, v15.16b	/* ^round key */
 	eor		\in2\().16b, \in2\().16b, v15.16b	/* ^round key */
 	eor		\in3\().16b, \in3\().16b, v15.16b	/* ^round key */
-	sub_bytes_4x	\in0, \in1, \in2, \in3
+	movi		v15.16b, #0x40
 	tbl		\in0\().16b, {\in0\().16b}, v13.16b	/* ShiftRows */
 	tbl		\in1\().16b, {\in1\().16b}, v13.16b	/* ShiftRows */
 	tbl		\in2\().16b, {\in2\().16b}, v13.16b	/* ShiftRows */
 	tbl		\in3\().16b, {\in3\().16b}, v13.16b	/* ShiftRows */
-	ld1		{v15.4s}, [\rkp], #16
+	sub_bytes_4x	\in0, \in1, \in2, \in3
 	subs		\i, \i, #1
+	ld1		{v15.4s}, [\rkp], #16
 	beq		2222f
-	.if		\enc == 1
-	mix_columns_2x	\in0, \in1
-	mix_columns_2x	\in2, \in3
-	ldr		q13, .LForward_ShiftRows
-	.else
-	inv_mix_cols_4x	\in0, \in1, \in2, \in3
-	ldr		q13, .LReverse_ShiftRows
-	.endif
-	movi		v12.16b, #0x40
+	mix_columns_2x	\in0, \in1, \enc
+	mix_columns_2x	\in2, \in3, \enc
 	b		1111b
 2222:	eor		\in0\().16b, \in0\().16b, v15.16b	/* ^round key */
 	eor		\in1\().16b, \in1\().16b, v15.16b	/* ^round key */
@@ -305,19 +273,7 @@
 #include "aes-modes.S"
 
 	.text
-	.align		4
-.LForward_ShiftRows:
-CPU_LE(	.byte		0x0, 0x5, 0xa, 0xf, 0x4, 0x9, 0xe, 0x3	)
-CPU_LE(	.byte		0x8, 0xd, 0x2, 0x7, 0xc, 0x1, 0x6, 0xb	)
-CPU_BE(	.byte		0xb, 0x6, 0x1, 0xc, 0x7, 0x2, 0xd, 0x8	)
-CPU_BE(	.byte		0x3, 0xe, 0x9, 0x4, 0xf, 0xa, 0x5, 0x0	)
-
-.LReverse_ShiftRows:
-CPU_LE(	.byte		0x0, 0xd, 0xa, 0x7, 0x4, 0x1, 0xe, 0xb	)
-CPU_LE(	.byte		0x8, 0x5, 0x2, 0xf, 0xc, 0x9, 0x6, 0x3	)
-CPU_BE(	.byte		0x3, 0x6, 0x9, 0xc, 0xf, 0x2, 0x5, 0x8	)
-CPU_BE(	.byte		0xb, 0xe, 0x1, 0x4, 0x7, 0xa, 0xd, 0x0	)
-
+	.align		6
 .LForward_Sbox:
 	.byte		0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5
 	.byte		0x30, 0x01, 0x67, 0x2b, 0xfe, 0xd7, 0xab, 0x76
@@ -385,3 +341,12 @@ CPU_BE(	.byte		0xb, 0xe, 0x1, 0x4, 0x7, 0xa, 0xd, 0x0	)
 	.byte		0xc8, 0xeb, 0xbb, 0x3c, 0x83, 0x53, 0x99, 0x61
 	.byte		0x17, 0x2b, 0x04, 0x7e, 0xba, 0x77, 0xd6, 0x26
 	.byte		0xe1, 0x69, 0x14, 0x63, 0x55, 0x21, 0x0c, 0x7d
+
+.LForward_ShiftRows:
+	.octa		0x0b06010c07020d08030e09040f0a0500
+
+.LReverse_ShiftRows:
+	.octa		0x0306090c0f0205080b0e0104070a0d00
+
+.Lror32by8:
+	.octa		0x0c0f0e0d080b0a090407060500030201
diff --git a/arch/arm64/crypto/aes-neonbs-core.S b/arch/arm64/crypto/aes-neonbs-core.S
new file mode 100644
index 000000000000..ca0472500433
--- /dev/null
+++ b/arch/arm64/crypto/aes-neonbs-core.S
@@ -0,0 +1,972 @@
+/*
+ * Bit sliced AES using NEON instructions
+ *
+ * Copyright (C) 2016 Linaro Ltd <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/*
+ * The algorithm implemented here is described in detail by the paper
+ * 'Faster and Timing-Attack Resistant AES-GCM' by Emilia Kaesper and
+ * Peter Schwabe (https://eprint.iacr.org/2009/129.pdf)
+ *
+ * This implementation is based primarily on the OpenSSL implementation
+ * for 32-bit ARM written by Andy Polyakov <appro@openssl.org>
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+	.text
+
+	rounds		.req	x11
+	bskey		.req	x12
+
+	.macro		in_bs_ch, b0, b1, b2, b3, b4, b5, b6, b7
+	eor		\b2, \b2, \b1
+	eor		\b5, \b5, \b6
+	eor		\b3, \b3, \b0
+	eor		\b6, \b6, \b2
+	eor		\b5, \b5, \b0
+	eor		\b6, \b6, \b3
+	eor		\b3, \b3, \b7
+	eor		\b7, \b7, \b5
+	eor		\b3, \b3, \b4
+	eor		\b4, \b4, \b5
+	eor		\b2, \b2, \b7
+	eor		\b3, \b3, \b1
+	eor		\b1, \b1, \b5
+	.endm
+
+	.macro		out_bs_ch, b0, b1, b2, b3, b4, b5, b6, b7
+	eor		\b0, \b0, \b6
+	eor		\b1, \b1, \b4
+	eor		\b4, \b4, \b6
+	eor		\b2, \b2, \b0
+	eor		\b6, \b6, \b1
+	eor		\b1, \b1, \b5
+	eor		\b5, \b5, \b3
+	eor		\b3, \b3, \b7
+	eor		\b7, \b7, \b5
+	eor		\b2, \b2, \b5
+	eor		\b4, \b4, \b7
+	.endm
+
+	.macro		inv_in_bs_ch, b6, b1, b2, b4, b7, b0, b3, b5
+	eor		\b1, \b1, \b7
+	eor		\b4, \b4, \b7
+	eor		\b7, \b7, \b5
+	eor		\b1, \b1, \b3
+	eor		\b2, \b2, \b5
+	eor		\b3, \b3, \b7
+	eor		\b6, \b6, \b1
+	eor		\b2, \b2, \b0
+	eor		\b5, \b5, \b3
+	eor		\b4, \b4, \b6
+	eor		\b0, \b0, \b6
+	eor		\b1, \b1, \b4
+	.endm
+
+	.macro		inv_out_bs_ch, b6, b5, b0, b3, b7, b1, b4, b2
+	eor		\b1, \b1, \b5
+	eor		\b2, \b2, \b7
+	eor		\b3, \b3, \b1
+	eor		\b4, \b4, \b5
+	eor		\b7, \b7, \b5
+	eor		\b3, \b3, \b4
+	eor 		\b5, \b5, \b0
+	eor		\b3, \b3, \b7
+	eor		\b6, \b6, \b2
+	eor		\b2, \b2, \b1
+	eor		\b6, \b6, \b3
+	eor		\b3, \b3, \b0
+	eor		\b5, \b5, \b6
+	.endm
+
+	.macro		mul_gf4, x0, x1, y0, y1, t0, t1
+	eor 		\t0, \y0, \y1
+	and		\t0, \t0, \x0
+	eor		\x0, \x0, \x1
+	and		\t1, \x1, \y0
+	and		\x0, \x0, \y1
+	eor		\x1, \t1, \t0
+	eor		\x0, \x0, \t1
+	.endm
+
+	.macro		mul_gf4_n_gf4, x0, x1, y0, y1, t0, x2, x3, y2, y3, t1
+	eor		\t0, \y0, \y1
+	eor 		\t1, \y2, \y3
+	and		\t0, \t0, \x0
+	and		\t1, \t1, \x2
+	eor		\x0, \x0, \x1
+	eor		\x2, \x2, \x3
+	and		\x1, \x1, \y0
+	and		\x3, \x3, \y2
+	and		\x0, \x0, \y1
+	and		\x2, \x2, \y3
+	eor		\x1, \x1, \x0
+	eor		\x2, \x2, \x3
+	eor		\x0, \x0, \t0
+	eor		\x3, \x3, \t1
+	.endm
+
+	.macro		mul_gf16_2, x0, x1, x2, x3, x4, x5, x6, x7, \
+				    y0, y1, y2, y3, t0, t1, t2, t3
+	eor		\t0, \x0, \x2
+	eor		\t1, \x1, \x3
+	mul_gf4  	\x0, \x1, \y0, \y1, \t2, \t3
+	eor		\y0, \y0, \y2
+	eor		\y1, \y1, \y3
+	mul_gf4_n_gf4	\t0, \t1, \y0, \y1, \t3, \x2, \x3, \y2, \y3, \t2
+	eor		\x0, \x0, \t0
+	eor		\x2, \x2, \t0
+	eor		\x1, \x1, \t1
+	eor		\x3, \x3, \t1
+	eor		\t0, \x4, \x6
+	eor		\t1, \x5, \x7
+	mul_gf4_n_gf4	\t0, \t1, \y0, \y1, \t3, \x6, \x7, \y2, \y3, \t2
+	eor		\y0, \y0, \y2
+	eor		\y1, \y1, \y3
+	mul_gf4  	\x4, \x5, \y0, \y1, \t2, \t3
+	eor		\x4, \x4, \t0
+	eor		\x6, \x6, \t0
+	eor		\x5, \x5, \t1
+	eor		\x7, \x7, \t1
+	.endm
+
+	.macro		inv_gf256, x0, x1, x2, x3, x4, x5, x6, x7, \
+				   t0, t1, t2, t3, s0, s1, s2, s3
+	eor		\t3, \x4, \x6
+	eor		\t0, \x5, \x7
+	eor		\t1, \x1, \x3
+	eor		\s1, \x7, \x6
+	eor		\s0, \x0, \x2
+	eor		\s3, \t3, \t0
+	orr		\t2, \t0, \t1
+	and		\s2, \t3, \s0
+	orr		\t3, \t3, \s0
+	eor		\s0, \s0, \t1
+	and		\t0, \t0, \t1
+	eor		\t1, \x3, \x2
+	and		\s3, \s3, \s0
+	and		\s1, \s1, \t1
+	eor		\t1, \x4, \x5
+	eor		\s0, \x1, \x0
+	eor		\t3, \t3, \s1
+	eor		\t2, \t2, \s1
+	and		\s1, \t1, \s0
+	orr		\t1, \t1, \s0
+	eor		\t3, \t3, \s3
+	eor		\t0, \t0, \s1
+	eor		\t2, \t2, \s2
+	eor		\t1, \t1, \s3
+	eor		\t0, \t0, \s2
+	and		\s0, \x7, \x3
+	eor		\t1, \t1, \s2
+	and		\s1, \x6, \x2
+	and		\s2, \x5, \x1
+	orr		\s3, \x4, \x0
+	eor		\t3, \t3, \s0
+	eor		\t1, \t1, \s2
+	eor		\s0, \t0, \s3
+	eor		\t2, \t2, \s1
+	and		\s2, \t3, \t1
+	eor		\s1, \t2, \s2
+	eor		\s3, \s0, \s2
+	bsl		\s1, \t1, \s0
+	not		\t0, \s0
+	bsl		\s0, \s1, \s3
+	bsl		\t0, \s1, \s3
+	bsl		\s3, \t3, \t2
+	eor		\t3, \t3, \t2
+	and		\s2, \s0, \s3
+	eor		\t1, \t1, \t0
+	eor		\s2, \s2, \t3
+	mul_gf16_2	\x0, \x1, \x2, \x3, \x4, \x5, \x6, \x7, \
+			\s3, \s2, \s1, \t1, \s0, \t0, \t2, \t3
+	.endm
+
+	.macro		sbox, b0, b1, b2, b3, b4, b5, b6, b7, \
+			      t0, t1, t2, t3, s0, s1, s2, s3
+	in_bs_ch	\b0\().16b, \b1\().16b, \b2\().16b, \b3\().16b, \
+			\b4\().16b, \b5\().16b, \b6\().16b, \b7\().16b
+	inv_gf256	\b6\().16b, \b5\().16b, \b0\().16b, \b3\().16b, \
+			\b7\().16b, \b1\().16b, \b4\().16b, \b2\().16b, \
+			\t0\().16b, \t1\().16b, \t2\().16b, \t3\().16b, \
+			\s0\().16b, \s1\().16b, \s2\().16b, \s3\().16b
+	out_bs_ch	\b7\().16b, \b1\().16b, \b4\().16b, \b2\().16b, \
+			\b6\().16b, \b5\().16b, \b0\().16b, \b3\().16b
+	.endm
+
+	.macro		inv_sbox, b0, b1, b2, b3, b4, b5, b6, b7, \
+				  t0, t1, t2, t3, s0, s1, s2, s3
+	inv_in_bs_ch	\b0\().16b, \b1\().16b, \b2\().16b, \b3\().16b, \
+			\b4\().16b, \b5\().16b, \b6\().16b, \b7\().16b
+	inv_gf256	\b5\().16b, \b1\().16b, \b2\().16b, \b6\().16b, \
+			\b3\().16b, \b7\().16b, \b0\().16b, \b4\().16b, \
+			\t0\().16b, \t1\().16b, \t2\().16b, \t3\().16b, \
+			\s0\().16b, \s1\().16b, \s2\().16b, \s3\().16b
+	inv_out_bs_ch	\b3\().16b, \b7\().16b, \b0\().16b, \b4\().16b, \
+			\b5\().16b, \b1\().16b, \b2\().16b, \b6\().16b
+	.endm
+
+	.macro		enc_next_rk
+	ldp		q16, q17, [bskey], #128
+	ldp		q18, q19, [bskey, #-96]
+	ldp		q20, q21, [bskey, #-64]
+	ldp		q22, q23, [bskey, #-32]
+	.endm
+
+	.macro		dec_next_rk
+	ldp		q16, q17, [bskey, #-128]!
+	ldp		q18, q19, [bskey, #32]
+	ldp		q20, q21, [bskey, #64]
+	ldp		q22, q23, [bskey, #96]
+	.endm
+
+	.macro		add_round_key, x0, x1, x2, x3, x4, x5, x6, x7
+	eor		\x0\().16b, \x0\().16b, v16.16b
+	eor		\x1\().16b, \x1\().16b, v17.16b
+	eor		\x2\().16b, \x2\().16b, v18.16b
+	eor		\x3\().16b, \x3\().16b, v19.16b
+	eor		\x4\().16b, \x4\().16b, v20.16b
+	eor		\x5\().16b, \x5\().16b, v21.16b
+	eor		\x6\().16b, \x6\().16b, v22.16b
+	eor		\x7\().16b, \x7\().16b, v23.16b
+	.endm
+
+	.macro		shift_rows, x0, x1, x2, x3, x4, x5, x6, x7, mask
+	tbl		\x0\().16b, {\x0\().16b}, \mask\().16b
+	tbl		\x1\().16b, {\x1\().16b}, \mask\().16b
+	tbl		\x2\().16b, {\x2\().16b}, \mask\().16b
+	tbl		\x3\().16b, {\x3\().16b}, \mask\().16b
+	tbl		\x4\().16b, {\x4\().16b}, \mask\().16b
+	tbl		\x5\().16b, {\x5\().16b}, \mask\().16b
+	tbl		\x6\().16b, {\x6\().16b}, \mask\().16b
+	tbl		\x7\().16b, {\x7\().16b}, \mask\().16b
+	.endm
+
+	.macro		mix_cols, x0, x1, x2, x3, x4, x5, x6, x7, \
+				  t0, t1, t2, t3, t4, t5, t6, t7, inv
+	ext		\t0\().16b, \x0\().16b, \x0\().16b, #12
+	ext		\t1\().16b, \x1\().16b, \x1\().16b, #12
+	eor		\x0\().16b, \x0\().16b, \t0\().16b
+	ext		\t2\().16b, \x2\().16b, \x2\().16b, #12
+	eor		\x1\().16b, \x1\().16b, \t1\().16b
+	ext		\t3\().16b, \x3\().16b, \x3\().16b, #12
+	eor		\x2\().16b, \x2\().16b, \t2\().16b
+	ext		\t4\().16b, \x4\().16b, \x4\().16b, #12
+	eor		\x3\().16b, \x3\().16b, \t3\().16b
+	ext		\t5\().16b, \x5\().16b, \x5\().16b, #12
+	eor		\x4\().16b, \x4\().16b, \t4\().16b
+	ext		\t6\().16b, \x6\().16b, \x6\().16b, #12
+	eor		\x5\().16b, \x5\().16b, \t5\().16b
+	ext		\t7\().16b, \x7\().16b, \x7\().16b, #12
+	eor		\x6\().16b, \x6\().16b, \t6\().16b
+	eor		\t1\().16b, \t1\().16b, \x0\().16b
+	eor		\x7\().16b, \x7\().16b, \t7\().16b
+	ext		\x0\().16b, \x0\().16b, \x0\().16b, #8
+	eor		\t2\().16b, \t2\().16b, \x1\().16b
+	eor		\t0\().16b, \t0\().16b, \x7\().16b
+	eor		\t1\().16b, \t1\().16b, \x7\().16b
+	ext		\x1\().16b, \x1\().16b, \x1\().16b, #8
+	eor		\t5\().16b, \t5\().16b, \x4\().16b
+	eor		\x0\().16b, \x0\().16b, \t0\().16b
+	eor		\t6\().16b, \t6\().16b, \x5\().16b
+	eor		\x1\().16b, \x1\().16b, \t1\().16b
+	ext		\t0\().16b, \x4\().16b, \x4\().16b, #8
+	eor		\t4\().16b, \t4\().16b, \x3\().16b
+	ext		\t1\().16b, \x5\().16b, \x5\().16b, #8
+	eor		\t7\().16b, \t7\().16b, \x6\().16b
+	ext		\x4\().16b, \x3\().16b, \x3\().16b, #8
+	eor		\t3\().16b, \t3\().16b, \x2\().16b
+	ext		\x5\().16b, \x7\().16b, \x7\().16b, #8
+	eor		\t4\().16b, \t4\().16b, \x7\().16b
+	ext		\x3\().16b, \x6\().16b, \x6\().16b, #8
+	eor		\t3\().16b, \t3\().16b, \x7\().16b
+	ext		\x6\().16b, \x2\().16b, \x2\().16b, #8
+	eor		\x7\().16b, \t1\().16b, \t5\().16b
+	.ifb		\inv
+	eor		\x2\().16b, \t0\().16b, \t4\().16b
+	eor		\x4\().16b, \x4\().16b, \t3\().16b
+	eor		\x5\().16b, \x5\().16b, \t7\().16b
+	eor		\x3\().16b, \x3\().16b, \t6\().16b
+	eor		\x6\().16b, \x6\().16b, \t2\().16b
+	.else
+	eor		\t3\().16b, \t3\().16b, \x4\().16b
+	eor		\x5\().16b, \x5\().16b, \t7\().16b
+	eor		\x2\().16b, \x3\().16b, \t6\().16b
+	eor		\x3\().16b, \t0\().16b, \t4\().16b
+	eor		\x4\().16b, \x6\().16b, \t2\().16b
+	mov		\x6\().16b, \t3\().16b
+	.endif
+	.endm
+
+	.macro		inv_mix_cols, x0, x1, x2, x3, x4, x5, x6, x7, \
+				      t0, t1, t2, t3, t4, t5, t6, t7
+	ext		\t0\().16b, \x0\().16b, \x0\().16b, #8
+	ext		\t6\().16b, \x6\().16b, \x6\().16b, #8
+	ext		\t7\().16b, \x7\().16b, \x7\().16b, #8
+	eor		\t0\().16b, \t0\().16b, \x0\().16b
+	ext		\t1\().16b, \x1\().16b, \x1\().16b, #8
+	eor		\t6\().16b, \t6\().16b, \x6\().16b
+	ext		\t2\().16b, \x2\().16b, \x2\().16b, #8
+	eor		\t7\().16b, \t7\().16b, \x7\().16b
+	ext		\t3\().16b, \x3\().16b, \x3\().16b, #8
+	eor		\t1\().16b, \t1\().16b, \x1\().16b
+	ext		\t4\().16b, \x4\().16b, \x4\().16b, #8
+	eor		\t2\().16b, \t2\().16b, \x2\().16b
+	ext		\t5\().16b, \x5\().16b, \x5\().16b, #8
+	eor		\t3\().16b, \t3\().16b, \x3\().16b
+	eor		\t4\().16b, \t4\().16b, \x4\().16b
+	eor		\t5\().16b, \t5\().16b, \x5\().16b
+	eor		\x0\().16b, \x0\().16b, \t6\().16b
+	eor		\x1\().16b, \x1\().16b, \t6\().16b
+	eor		\x2\().16b, \x2\().16b, \t0\().16b
+	eor		\x4\().16b, \x4\().16b, \t2\().16b
+	eor		\x3\().16b, \x3\().16b, \t1\().16b
+	eor		\x1\().16b, \x1\().16b, \t7\().16b
+	eor		\x2\().16b, \x2\().16b, \t7\().16b
+	eor		\x4\().16b, \x4\().16b, \t6\().16b
+	eor		\x5\().16b, \x5\().16b, \t3\().16b
+	eor		\x3\().16b, \x3\().16b, \t6\().16b
+	eor		\x6\().16b, \x6\().16b, \t4\().16b
+	eor		\x4\().16b, \x4\().16b, \t7\().16b
+	eor		\x5\().16b, \x5\().16b, \t7\().16b
+	eor		\x7\().16b, \x7\().16b, \t5\().16b
+	mix_cols	\x0, \x1, \x2, \x3, \x4, \x5, \x6, \x7, \
+			\t0, \t1, \t2, \t3, \t4, \t5, \t6, \t7, 1
+	.endm
+
+	.macro		swapmove_2x, a0, b0, a1, b1, n, mask, t0, t1
+	ushr		\t0\().2d, \b0\().2d, #\n
+	ushr		\t1\().2d, \b1\().2d, #\n
+	eor		\t0\().16b, \t0\().16b, \a0\().16b
+	eor		\t1\().16b, \t1\().16b, \a1\().16b
+	and		\t0\().16b, \t0\().16b, \mask\().16b
+	and		\t1\().16b, \t1\().16b, \mask\().16b
+	eor		\a0\().16b, \a0\().16b, \t0\().16b
+	shl		\t0\().2d, \t0\().2d, #\n
+	eor		\a1\().16b, \a1\().16b, \t1\().16b
+	shl		\t1\().2d, \t1\().2d, #\n
+	eor		\b0\().16b, \b0\().16b, \t0\().16b
+	eor		\b1\().16b, \b1\().16b, \t1\().16b
+	.endm
+
+	.macro		bitslice, x7, x6, x5, x4, x3, x2, x1, x0, t0, t1, t2, t3
+	movi		\t0\().16b, #0x55
+	movi		\t1\().16b, #0x33
+	swapmove_2x	\x0, \x1, \x2, \x3, 1, \t0, \t2, \t3
+	swapmove_2x	\x4, \x5, \x6, \x7, 1, \t0, \t2, \t3
+	movi		\t0\().16b, #0x0f
+	swapmove_2x	\x0, \x2, \x1, \x3, 2, \t1, \t2, \t3
+	swapmove_2x	\x4, \x6, \x5, \x7, 2, \t1, \t2, \t3
+	swapmove_2x	\x0, \x4, \x1, \x5, 4, \t0, \t2, \t3
+	swapmove_2x	\x2, \x6, \x3, \x7, 4, \t0, \t2, \t3
+	.endm
+
+
+	.align		6
+M0:	.octa		0x0004080c0105090d02060a0e03070b0f
+
+M0SR:	.octa		0x0004080c05090d010a0e02060f03070b
+SR:	.octa		0x0f0e0d0c0a09080b0504070600030201
+SRM0:	.octa		0x01060b0c0207080d0304090e00050a0f
+
+M0ISR:	.octa		0x0004080c0d0105090a0e0206070b0f03
+ISR:	.octa		0x0f0e0d0c080b0a090504070602010003
+ISRM0:	.octa		0x0306090c00070a0d01040b0e0205080f
+
+	/*
+	 * void aesbs_convert_key(u8 out[], u32 const rk[], int rounds)
+	 */
+ENTRY(aesbs_convert_key)
+	ld1		{v7.4s}, [x1], #16		// load round 0 key
+	ld1		{v17.4s}, [x1], #16		// load round 1 key
+
+	movi		v8.16b,  #0x01			// bit masks
+	movi		v9.16b,  #0x02
+	movi		v10.16b, #0x04
+	movi		v11.16b, #0x08
+	movi		v12.16b, #0x10
+	movi		v13.16b, #0x20
+	movi		v14.16b, #0x40
+	movi		v15.16b, #0x80
+	ldr		q16, M0
+
+	sub		x2, x2, #1
+	str		q7, [x0], #16		// save round 0 key
+
+.Lkey_loop:
+	tbl		v7.16b ,{v17.16b}, v16.16b
+	ld1		{v17.4s}, [x1], #16		// load next round key
+
+	cmtst		v0.16b, v7.16b, v8.16b
+	cmtst		v1.16b, v7.16b, v9.16b
+	cmtst		v2.16b, v7.16b, v10.16b
+	cmtst		v3.16b, v7.16b, v11.16b
+	cmtst		v4.16b, v7.16b, v12.16b
+	cmtst		v5.16b, v7.16b, v13.16b
+	cmtst		v6.16b, v7.16b, v14.16b
+	cmtst		v7.16b, v7.16b, v15.16b
+	not		v0.16b, v0.16b
+	not		v1.16b, v1.16b
+	not		v5.16b, v5.16b
+	not		v6.16b, v6.16b
+
+	subs		x2, x2, #1
+	stp		q0, q1, [x0], #128
+	stp		q2, q3, [x0, #-96]
+	stp		q4, q5, [x0, #-64]
+	stp		q6, q7, [x0, #-32]
+	b.ne		.Lkey_loop
+
+	movi		v7.16b, #0x63			// compose .L63
+	eor		v17.16b, v17.16b, v7.16b
+	str		q17, [x0]
+	ret
+ENDPROC(aesbs_convert_key)
+
+	.align		4
+aesbs_encrypt8:
+	ldr		q9, [bskey], #16		// round 0 key
+	ldr		q8, M0SR
+	ldr		q24, SR
+
+	eor		v10.16b, v0.16b, v9.16b		// xor with round0 key
+	eor		v11.16b, v1.16b, v9.16b
+	tbl		v0.16b, {v10.16b}, v8.16b
+	eor		v12.16b, v2.16b, v9.16b
+	tbl		v1.16b, {v11.16b}, v8.16b
+	eor		v13.16b, v3.16b, v9.16b
+	tbl		v2.16b, {v12.16b}, v8.16b
+	eor		v14.16b, v4.16b, v9.16b
+	tbl		v3.16b, {v13.16b}, v8.16b
+	eor		v15.16b, v5.16b, v9.16b
+	tbl		v4.16b, {v14.16b}, v8.16b
+	eor		v10.16b, v6.16b, v9.16b
+	tbl		v5.16b, {v15.16b}, v8.16b
+	eor		v11.16b, v7.16b, v9.16b
+	tbl		v6.16b, {v10.16b}, v8.16b
+	tbl		v7.16b, {v11.16b}, v8.16b
+
+	bitslice	v0, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11
+
+	sub		rounds, rounds, #1
+	b		.Lenc_sbox
+
+.Lenc_loop:
+	shift_rows	v0, v1, v2, v3, v4, v5, v6, v7, v24
+.Lenc_sbox:
+	sbox		v0, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11, v12, \
+								v13, v14, v15
+	subs		rounds, rounds, #1
+	b.cc		.Lenc_done
+
+	enc_next_rk
+
+	mix_cols	v0, v1, v4, v6, v3, v7, v2, v5, v8, v9, v10, v11, v12, \
+								v13, v14, v15
+
+	add_round_key	v0, v1, v2, v3, v4, v5, v6, v7
+
+	b.ne		.Lenc_loop
+	ldr		q24, SRM0
+	b		.Lenc_loop
+
+.Lenc_done:
+	ldr		q12, [bskey]			// last round key
+
+	bitslice	v0, v1, v4, v6, v3, v7, v2, v5, v8, v9, v10, v11
+
+	eor		v0.16b, v0.16b, v12.16b
+	eor		v1.16b, v1.16b, v12.16b
+	eor		v4.16b, v4.16b, v12.16b
+	eor		v6.16b, v6.16b, v12.16b
+	eor		v3.16b, v3.16b, v12.16b
+	eor		v7.16b, v7.16b, v12.16b
+	eor		v2.16b, v2.16b, v12.16b
+	eor		v5.16b, v5.16b, v12.16b
+	ret
+ENDPROC(aesbs_encrypt8)
+
+	.align		4
+aesbs_decrypt8:
+	lsl		x9, rounds, #7
+	add		bskey, bskey, x9
+
+	ldr		q9, [bskey, #-112]!		// round 0 key
+	ldr		q8, M0ISR
+	ldr		q24, ISR
+
+	eor		v10.16b, v0.16b, v9.16b		// xor with round0 key
+	eor		v11.16b, v1.16b, v9.16b
+	tbl		v0.16b, {v10.16b}, v8.16b
+	eor		v12.16b, v2.16b, v9.16b
+	tbl		v1.16b, {v11.16b}, v8.16b
+	eor		v13.16b, v3.16b, v9.16b
+	tbl		v2.16b, {v12.16b}, v8.16b
+	eor		v14.16b, v4.16b, v9.16b
+	tbl		v3.16b, {v13.16b}, v8.16b
+	eor		v15.16b, v5.16b, v9.16b
+	tbl		v4.16b, {v14.16b}, v8.16b
+	eor		v10.16b, v6.16b, v9.16b
+	tbl		v5.16b, {v15.16b}, v8.16b
+	eor		v11.16b, v7.16b, v9.16b
+	tbl		v6.16b, {v10.16b}, v8.16b
+	tbl		v7.16b, {v11.16b}, v8.16b
+
+	bitslice	v0, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11
+
+	sub		rounds, rounds, #1
+	b		.Ldec_sbox
+
+.Ldec_loop:
+	shift_rows	v0, v1, v2, v3, v4, v5, v6, v7, v24
+.Ldec_sbox:
+	inv_sbox	v0, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11, v12, \
+								v13, v14, v15
+	subs		rounds, rounds, #1
+	b.cc		.Ldec_done
+
+	dec_next_rk
+
+	add_round_key	v0, v1, v6, v4, v2, v7, v3, v5
+
+	inv_mix_cols	v0, v1, v6, v4, v2, v7, v3, v5, v8, v9, v10, v11, v12, \
+								v13, v14, v15
+
+	b.ne		.Ldec_loop
+	ldr		q24, ISRM0
+	b		.Ldec_loop
+.Ldec_done:
+	ldr		q12, [bskey, #-16]		// last round key
+
+	bitslice	v0, v1, v6, v4, v2, v7, v3, v5, v8, v9, v10, v11
+
+	eor		v0.16b, v0.16b, v12.16b
+	eor		v1.16b, v1.16b, v12.16b
+	eor		v6.16b, v6.16b, v12.16b
+	eor		v4.16b, v4.16b, v12.16b
+	eor		v2.16b, v2.16b, v12.16b
+	eor		v7.16b, v7.16b, v12.16b
+	eor		v3.16b, v3.16b, v12.16b
+	eor		v5.16b, v5.16b, v12.16b
+	ret
+ENDPROC(aesbs_decrypt8)
+
+	/*
+	 * aesbs_ecb_encrypt(u8 out[], u8 const in[], u8 const rk[], int rounds,
+	 *		     int blocks)
+	 * aesbs_ecb_decrypt(u8 out[], u8 const in[], u8 const rk[], int rounds,
+	 *		     int blocks)
+	 */
+	.macro		__ecb_crypt, do8, o0, o1, o2, o3, o4, o5, o6, o7
+	stp		x29, x30, [sp, #-16]!
+	mov		x29, sp
+
+99:	mov		x5, #1
+	lsl		x5, x5, x4
+	subs		w4, w4, #8
+	csel		x4, x4, xzr, pl
+	csel		x5, x5, xzr, mi
+
+	ld1		{v0.16b}, [x1], #16
+	tbnz		x5, #1, 0f
+	ld1		{v1.16b}, [x1], #16
+	tbnz		x5, #2, 0f
+	ld1		{v2.16b}, [x1], #16
+	tbnz		x5, #3, 0f
+	ld1		{v3.16b}, [x1], #16
+	tbnz		x5, #4, 0f
+	ld1		{v4.16b}, [x1], #16
+	tbnz		x5, #5, 0f
+	ld1		{v5.16b}, [x1], #16
+	tbnz		x5, #6, 0f
+	ld1		{v6.16b}, [x1], #16
+	tbnz		x5, #7, 0f
+	ld1		{v7.16b}, [x1], #16
+
+0:	mov		bskey, x2
+	mov		rounds, x3
+	bl		\do8
+
+	st1		{\o0\().16b}, [x0], #16
+	tbnz		x5, #1, 1f
+	st1		{\o1\().16b}, [x0], #16
+	tbnz		x5, #2, 1f
+	st1		{\o2\().16b}, [x0], #16
+	tbnz		x5, #3, 1f
+	st1		{\o3\().16b}, [x0], #16
+	tbnz		x5, #4, 1f
+	st1		{\o4\().16b}, [x0], #16
+	tbnz		x5, #5, 1f
+	st1		{\o5\().16b}, [x0], #16
+	tbnz		x5, #6, 1f
+	st1		{\o6\().16b}, [x0], #16
+	tbnz		x5, #7, 1f
+	st1		{\o7\().16b}, [x0], #16
+
+	cbnz		x4, 99b
+
+1:	ldp		x29, x30, [sp], #16
+	ret
+	.endm
+
+	.align		4
+ENTRY(aesbs_ecb_encrypt)
+	__ecb_crypt	aesbs_encrypt8, v0, v1, v4, v6, v3, v7, v2, v5
+ENDPROC(aesbs_ecb_encrypt)
+
+	.align		4
+ENTRY(aesbs_ecb_decrypt)
+	__ecb_crypt	aesbs_decrypt8, v0, v1, v6, v4, v2, v7, v3, v5
+ENDPROC(aesbs_ecb_decrypt)
+
+	/*
+	 * aesbs_cbc_decrypt(u8 out[], u8 const in[], u8 const rk[], int rounds,
+	 *		     int blocks, u8 iv[])
+	 */
+	.align		4
+ENTRY(aesbs_cbc_decrypt)
+	stp		x29, x30, [sp, #-16]!
+	mov		x29, sp
+
+99:	mov		x6, #1
+	lsl		x6, x6, x4
+	subs		w4, w4, #8
+	csel		x4, x4, xzr, pl
+	csel		x6, x6, xzr, mi
+
+	ld1		{v0.16b}, [x1], #16
+	mov		v25.16b, v0.16b
+	tbnz		x6, #1, 0f
+	ld1		{v1.16b}, [x1], #16
+	mov		v26.16b, v1.16b
+	tbnz		x6, #2, 0f
+	ld1		{v2.16b}, [x1], #16
+	mov		v27.16b, v2.16b
+	tbnz		x6, #3, 0f
+	ld1		{v3.16b}, [x1], #16
+	mov		v28.16b, v3.16b
+	tbnz		x6, #4, 0f
+	ld1		{v4.16b}, [x1], #16
+	mov		v29.16b, v4.16b
+	tbnz		x6, #5, 0f
+	ld1		{v5.16b}, [x1], #16
+	mov		v30.16b, v5.16b
+	tbnz		x6, #6, 0f
+	ld1		{v6.16b}, [x1], #16
+	mov		v31.16b, v6.16b
+	tbnz		x6, #7, 0f
+	ld1		{v7.16b}, [x1]
+
+0:	mov		bskey, x2
+	mov		rounds, x3
+	bl		aesbs_decrypt8
+
+	ld1		{v24.16b}, [x5]			// load IV
+
+	eor		v1.16b, v1.16b, v25.16b
+	eor		v6.16b, v6.16b, v26.16b
+	eor		v4.16b, v4.16b, v27.16b
+	eor		v2.16b, v2.16b, v28.16b
+	eor		v7.16b, v7.16b, v29.16b
+	eor		v0.16b, v0.16b, v24.16b
+	eor		v3.16b, v3.16b, v30.16b
+	eor		v5.16b, v5.16b, v31.16b
+
+	st1		{v0.16b}, [x0], #16
+	mov		v24.16b, v25.16b
+	tbnz		x6, #1, 1f
+	st1		{v1.16b}, [x0], #16
+	mov		v24.16b, v26.16b
+	tbnz		x6, #2, 1f
+	st1		{v6.16b}, [x0], #16
+	mov		v24.16b, v27.16b
+	tbnz		x6, #3, 1f
+	st1		{v4.16b}, [x0], #16
+	mov		v24.16b, v28.16b
+	tbnz		x6, #4, 1f
+	st1		{v2.16b}, [x0], #16
+	mov		v24.16b, v29.16b
+	tbnz		x6, #5, 1f
+	st1		{v7.16b}, [x0], #16
+	mov		v24.16b, v30.16b
+	tbnz		x6, #6, 1f
+	st1		{v3.16b}, [x0], #16
+	mov		v24.16b, v31.16b
+	tbnz		x6, #7, 1f
+	ld1		{v24.16b}, [x1], #16
+	st1		{v5.16b}, [x0], #16
+1:	st1		{v24.16b}, [x5]			// store IV
+
+	cbnz		x4, 99b
+
+	ldp		x29, x30, [sp], #16
+	ret
+ENDPROC(aesbs_cbc_decrypt)
+
+	.macro		next_tweak, out, in, const, tmp
+	sshr		\tmp\().2d,  \in\().2d,   #63
+	and		\tmp\().16b, \tmp\().16b, \const\().16b
+	add		\out\().2d,  \in\().2d,   \in\().2d
+	ext		\tmp\().16b, \tmp\().16b, \tmp\().16b, #8
+	eor		\out\().16b, \out\().16b, \tmp\().16b
+	.endm
+
+	.align		4
+.Lxts_mul_x:
+CPU_LE(	.quad		1, 0x87		)
+CPU_BE(	.quad		0x87, 1		)
+
+	/*
+	 * aesbs_xts_encrypt(u8 out[], u8 const in[], u8 const rk[], int rounds,
+	 *		     int blocks, u8 iv[])
+	 * aesbs_xts_decrypt(u8 out[], u8 const in[], u8 const rk[], int rounds,
+	 *		     int blocks, u8 iv[])
+	 */
+__xts_crypt8:
+	mov		x6, #1
+	lsl		x6, x6, x4
+	subs		w4, w4, #8
+	csel		x4, x4, xzr, pl
+	csel		x6, x6, xzr, mi
+
+	ld1		{v0.16b}, [x1], #16
+	next_tweak	v26, v25, v30, v31
+	eor		v0.16b, v0.16b, v25.16b
+	tbnz		x6, #1, 0f
+
+	ld1		{v1.16b}, [x1], #16
+	next_tweak	v27, v26, v30, v31
+	eor		v1.16b, v1.16b, v26.16b
+	tbnz		x6, #2, 0f
+
+	ld1		{v2.16b}, [x1], #16
+	next_tweak	v28, v27, v30, v31
+	eor		v2.16b, v2.16b, v27.16b
+	tbnz		x6, #3, 0f
+
+	ld1		{v3.16b}, [x1], #16
+	next_tweak	v29, v28, v30, v31
+	eor		v3.16b, v3.16b, v28.16b
+	tbnz		x6, #4, 0f
+
+	ld1		{v4.16b}, [x1], #16
+	str		q29, [sp, #16]
+	eor		v4.16b, v4.16b, v29.16b
+	next_tweak	v29, v29, v30, v31
+	tbnz		x6, #5, 0f
+
+	ld1		{v5.16b}, [x1], #16
+	str		q29, [sp, #32]
+	eor		v5.16b, v5.16b, v29.16b
+	next_tweak	v29, v29, v30, v31
+	tbnz		x6, #6, 0f
+
+	ld1		{v6.16b}, [x1], #16
+	str		q29, [sp, #48]
+	eor		v6.16b, v6.16b, v29.16b
+	next_tweak	v29, v29, v30, v31
+	tbnz		x6, #7, 0f
+
+	ld1		{v7.16b}, [x1], #16
+	str		q29, [sp, #64]
+	eor		v7.16b, v7.16b, v29.16b
+	next_tweak	v29, v29, v30, v31
+
+0:	mov		bskey, x2
+	mov		rounds, x3
+	br		x7
+ENDPROC(__xts_crypt8)
+
+	.macro		__xts_crypt, do8, o0, o1, o2, o3, o4, o5, o6, o7
+	stp		x29, x30, [sp, #-80]!
+	mov		x29, sp
+
+	ldr		q30, .Lxts_mul_x
+	ld1		{v25.16b}, [x5]
+
+99:	adr		x7, \do8
+	bl		__xts_crypt8
+
+	ldp		q16, q17, [sp, #16]
+	ldp		q18, q19, [sp, #48]
+
+	eor		\o0\().16b, \o0\().16b, v25.16b
+	eor		\o1\().16b, \o1\().16b, v26.16b
+	eor		\o2\().16b, \o2\().16b, v27.16b
+	eor		\o3\().16b, \o3\().16b, v28.16b
+
+	st1		{\o0\().16b}, [x0], #16
+	mov		v25.16b, v26.16b
+	tbnz		x6, #1, 1f
+	st1		{\o1\().16b}, [x0], #16
+	mov		v25.16b, v27.16b
+	tbnz		x6, #2, 1f
+	st1		{\o2\().16b}, [x0], #16
+	mov		v25.16b, v28.16b
+	tbnz		x6, #3, 1f
+	st1		{\o3\().16b}, [x0], #16
+	mov		v25.16b, v29.16b
+	tbnz		x6, #4, 1f
+
+	eor		\o4\().16b, \o4\().16b, v16.16b
+	eor		\o5\().16b, \o5\().16b, v17.16b
+	eor		\o6\().16b, \o6\().16b, v18.16b
+	eor		\o7\().16b, \o7\().16b, v19.16b
+
+	st1		{\o4\().16b}, [x0], #16
+	tbnz		x6, #5, 1f
+	st1		{\o5\().16b}, [x0], #16
+	tbnz		x6, #6, 1f
+	st1		{\o6\().16b}, [x0], #16
+	tbnz		x6, #7, 1f
+	st1		{\o7\().16b}, [x0], #16
+
+	cbnz		x4, 99b
+
+1:	st1		{v25.16b}, [x5]
+	ldp		x29, x30, [sp], #80
+	ret
+	.endm
+
+ENTRY(aesbs_xts_encrypt)
+	__xts_crypt	aesbs_encrypt8, v0, v1, v4, v6, v3, v7, v2, v5
+ENDPROC(aesbs_xts_encrypt)
+
+ENTRY(aesbs_xts_decrypt)
+	__xts_crypt	aesbs_decrypt8, v0, v1, v6, v4, v2, v7, v3, v5
+ENDPROC(aesbs_xts_decrypt)
+
+	.macro		next_ctr, v
+	mov		\v\().d[1], x8
+	adds		x8, x8, #1
+	mov		\v\().d[0], x7
+	adc		x7, x7, xzr
+	rev64		\v\().16b, \v\().16b
+	.endm
+
+	/*
+	 * aesbs_ctr_encrypt(u8 out[], u8 const in[], u8 const rk[],
+	 *		     int rounds, int blocks, u8 iv[], u8 final[])
+	 */
+ENTRY(aesbs_ctr_encrypt)
+	stp		x29, x30, [sp, #-16]!
+	mov		x29, sp
+
+	cmp		x6, #0
+	cset		x10, ne
+	add		x4, x4, x10		// do one extra block if final
+
+	ldp		x7, x8, [x5]
+	ld1		{v0.16b}, [x5]
+CPU_LE(	rev		x7, x7		)
+CPU_LE(	rev		x8, x8		)
+	adds		x8, x8, #1
+	adc		x7, x7, xzr
+
+99:	mov		x9, #1
+	lsl		x9, x9, x4
+	subs		w4, w4, #8
+	csel		x4, x4, xzr, pl
+	csel		x9, x9, xzr, le
+
+	tbnz		x9, #1, 0f
+	next_ctr	v1
+	tbnz		x9, #2, 0f
+	next_ctr	v2
+	tbnz		x9, #3, 0f
+	next_ctr	v3
+	tbnz		x9, #4, 0f
+	next_ctr	v4
+	tbnz		x9, #5, 0f
+	next_ctr	v5
+	tbnz		x9, #6, 0f
+	next_ctr	v6
+	tbnz		x9, #7, 0f
+	next_ctr	v7
+
+0:	mov		bskey, x2
+	mov		rounds, x3
+	bl		aesbs_encrypt8
+
+	lsr		x9, x9, x10		// disregard the extra block
+	tbnz		x9, #0, 0f
+
+	ld1		{v8.16b}, [x1], #16
+	eor		v0.16b, v0.16b, v8.16b
+	st1		{v0.16b}, [x0], #16
+	tbnz		x9, #1, 1f
+
+	ld1		{v9.16b}, [x1], #16
+	eor		v1.16b, v1.16b, v9.16b
+	st1		{v1.16b}, [x0], #16
+	tbnz		x9, #2, 2f
+
+	ld1		{v10.16b}, [x1], #16
+	eor		v4.16b, v4.16b, v10.16b
+	st1		{v4.16b}, [x0], #16
+	tbnz		x9, #3, 3f
+
+	ld1		{v11.16b}, [x1], #16
+	eor		v6.16b, v6.16b, v11.16b
+	st1		{v6.16b}, [x0], #16
+	tbnz		x9, #4, 4f
+
+	ld1		{v12.16b}, [x1], #16
+	eor		v3.16b, v3.16b, v12.16b
+	st1		{v3.16b}, [x0], #16
+	tbnz		x9, #5, 5f
+
+	ld1		{v13.16b}, [x1], #16
+	eor		v7.16b, v7.16b, v13.16b
+	st1		{v7.16b}, [x0], #16
+	tbnz		x9, #6, 6f
+
+	ld1		{v14.16b}, [x1], #16
+	eor		v2.16b, v2.16b, v14.16b
+	st1		{v2.16b}, [x0], #16
+	tbnz		x9, #7, 7f
+
+	ld1		{v15.16b}, [x1], #16
+	eor		v5.16b, v5.16b, v15.16b
+	st1		{v5.16b}, [x0], #16
+
+8:	next_ctr	v0
+	cbnz		x4, 99b
+
+0:	st1		{v0.16b}, [x5]
+	ldp		x29, x30, [sp], #16
+	ret
+
+	/*
+	 * If we are handling the tail of the input (x6 != NULL), return the
+	 * final keystream block back to the caller.
+	 */
+1:	cbz		x6, 8b
+	st1		{v1.16b}, [x6]
+	b		8b
+2:	cbz		x6, 8b
+	st1		{v4.16b}, [x6]
+	b		8b
+3:	cbz		x6, 8b
+	st1		{v6.16b}, [x6]
+	b		8b
+4:	cbz		x6, 8b
+	st1		{v3.16b}, [x6]
+	b		8b
+5:	cbz		x6, 8b
+	st1		{v7.16b}, [x6]
+	b		8b
+6:	cbz		x6, 8b
+	st1		{v2.16b}, [x6]
+	b		8b
+7:	cbz		x6, 8b
+	st1		{v5.16b}, [x6]
+	b		8b
+ENDPROC(aesbs_ctr_encrypt)
diff --git a/arch/arm64/crypto/aes-neonbs-glue.c b/arch/arm64/crypto/aes-neonbs-glue.c
new file mode 100644
index 000000000000..db2501d93550
--- /dev/null
+++ b/arch/arm64/crypto/aes-neonbs-glue.c
@@ -0,0 +1,439 @@
+/*
+ * Bit sliced AES using NEON instructions
+ *
+ * Copyright (C) 2016 Linaro Ltd <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <asm/neon.h>
+#include <crypto/aes.h>
+#include <crypto/internal/simd.h>
+#include <crypto/internal/skcipher.h>
+#include <crypto/xts.h>
+#include <linux/module.h>
+
+MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
+MODULE_LICENSE("GPL v2");
+
+MODULE_ALIAS_CRYPTO("ecb(aes)");
+MODULE_ALIAS_CRYPTO("cbc(aes)");
+MODULE_ALIAS_CRYPTO("ctr(aes)");
+MODULE_ALIAS_CRYPTO("xts(aes)");
+
+asmlinkage void aesbs_convert_key(u8 out[], u32 const rk[], int rounds);
+
+asmlinkage void aesbs_ecb_encrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks);
+asmlinkage void aesbs_ecb_decrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks);
+
+asmlinkage void aesbs_cbc_decrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks, u8 iv[]);
+
+asmlinkage void aesbs_ctr_encrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks, u8 iv[], u8 final[]);
+
+asmlinkage void aesbs_xts_encrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks, u8 iv[]);
+asmlinkage void aesbs_xts_decrypt(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks, u8 iv[]);
+
+/* borrowed from aes-neon-blk.ko */
+asmlinkage void neon_aes_ecb_encrypt(u8 out[], u8 const in[], u32 const rk[],
+				     int rounds, int blocks, int first);
+asmlinkage void neon_aes_cbc_encrypt(u8 out[], u8 const in[], u32 const rk[],
+				     int rounds, int blocks, u8 iv[],
+				     int first);
+
+struct aesbs_ctx {
+	u8	rk[13 * (8 * AES_BLOCK_SIZE) + 32];
+	int	rounds;
+} __aligned(AES_BLOCK_SIZE);
+
+struct aesbs_cbc_ctx {
+	struct aesbs_ctx	key;
+	u32			enc[AES_MAX_KEYLENGTH_U32];
+};
+
+struct aesbs_xts_ctx {
+	struct aesbs_ctx	key;
+	u32			twkey[AES_MAX_KEYLENGTH_U32];
+};
+
+static int aesbs_setkey(struct crypto_skcipher *tfm, const u8 *in_key,
+			unsigned int key_len)
+{
+	struct aesbs_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct crypto_aes_ctx rk;
+	int err;
+
+	err = crypto_aes_expand_key(&rk, in_key, key_len);
+	if (err)
+		return err;
+
+	ctx->rounds = 6 + key_len / 4;
+
+	kernel_neon_begin();
+	aesbs_convert_key(ctx->rk, rk.key_enc, ctx->rounds);
+	kernel_neon_end();
+
+	return 0;
+}
+
+static int __ecb_crypt(struct skcipher_request *req,
+		       void (*fn)(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks))
+{
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct aesbs_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_walk walk;
+	int err;
+
+	err = skcipher_walk_virt(&walk, req, true);
+
+	kernel_neon_begin();
+	while (walk.nbytes >= AES_BLOCK_SIZE) {
+		unsigned int blocks = walk.nbytes / AES_BLOCK_SIZE;
+
+		if (walk.nbytes < walk.total)
+			blocks = round_down(blocks,
+					    walk.stride / AES_BLOCK_SIZE);
+
+		fn(walk.dst.virt.addr, walk.src.virt.addr, ctx->rk,
+		   ctx->rounds, blocks);
+		err = skcipher_walk_done(&walk,
+					 walk.nbytes - blocks * AES_BLOCK_SIZE);
+	}
+	kernel_neon_end();
+
+	return err;
+}
+
+static int ecb_encrypt(struct skcipher_request *req)
+{
+	return __ecb_crypt(req, aesbs_ecb_encrypt);
+}
+
+static int ecb_decrypt(struct skcipher_request *req)
+{
+	return __ecb_crypt(req, aesbs_ecb_decrypt);
+}
+
+static int aesbs_cbc_setkey(struct crypto_skcipher *tfm, const u8 *in_key,
+			    unsigned int key_len)
+{
+	struct aesbs_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct crypto_aes_ctx rk;
+	int err;
+
+	err = crypto_aes_expand_key(&rk, in_key, key_len);
+	if (err)
+		return err;
+
+	ctx->key.rounds = 6 + key_len / 4;
+
+	memcpy(ctx->enc, rk.key_enc, sizeof(ctx->enc));
+
+	kernel_neon_begin();
+	aesbs_convert_key(ctx->key.rk, rk.key_enc, ctx->key.rounds);
+	kernel_neon_end();
+
+	return 0;
+}
+
+static int cbc_encrypt(struct skcipher_request *req)
+{
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct aesbs_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_walk walk;
+	int err, first = 1;
+
+	err = skcipher_walk_virt(&walk, req, true);
+
+	kernel_neon_begin();
+	while (walk.nbytes >= AES_BLOCK_SIZE) {
+		unsigned int blocks = walk.nbytes / AES_BLOCK_SIZE;
+
+		/* fall back to the non-bitsliced NEON implementation */
+		neon_aes_cbc_encrypt(walk.dst.virt.addr, walk.src.virt.addr,
+				     ctx->enc, ctx->key.rounds, blocks, walk.iv,
+				     first);
+		err = skcipher_walk_done(&walk, walk.nbytes % AES_BLOCK_SIZE);
+		first = 0;
+	}
+	kernel_neon_end();
+	return err;
+}
+
+static int cbc_decrypt(struct skcipher_request *req)
+{
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct aesbs_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_walk walk;
+	int err;
+
+	err = skcipher_walk_virt(&walk, req, true);
+
+	kernel_neon_begin();
+	while (walk.nbytes >= AES_BLOCK_SIZE) {
+		unsigned int blocks = walk.nbytes / AES_BLOCK_SIZE;
+
+		if (walk.nbytes < walk.total)
+			blocks = round_down(blocks,
+					    walk.stride / AES_BLOCK_SIZE);
+
+		aesbs_cbc_decrypt(walk.dst.virt.addr, walk.src.virt.addr,
+				  ctx->key.rk, ctx->key.rounds, blocks,
+				  walk.iv);
+		err = skcipher_walk_done(&walk,
+					 walk.nbytes - blocks * AES_BLOCK_SIZE);
+	}
+	kernel_neon_end();
+
+	return err;
+}
+
+static int ctr_encrypt(struct skcipher_request *req)
+{
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct aesbs_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_walk walk;
+	u8 buf[AES_BLOCK_SIZE];
+	int err;
+
+	err = skcipher_walk_virt(&walk, req, true);
+
+	kernel_neon_begin();
+	while (walk.nbytes > 0) {
+		unsigned int blocks = walk.nbytes / AES_BLOCK_SIZE;
+		u8 *final = (walk.total % AES_BLOCK_SIZE) ? buf : NULL;
+
+		if (walk.nbytes < walk.total) {
+			blocks = round_down(blocks,
+					    walk.stride / AES_BLOCK_SIZE);
+			final = NULL;
+		}
+
+		aesbs_ctr_encrypt(walk.dst.virt.addr, walk.src.virt.addr,
+				  ctx->rk, ctx->rounds, blocks, walk.iv, final);
+
+		if (final) {
+			u8 *dst = walk.dst.virt.addr + blocks * AES_BLOCK_SIZE;
+			u8 *src = walk.src.virt.addr + blocks * AES_BLOCK_SIZE;
+
+			if (dst != src)
+				memcpy(dst, src, walk.total % AES_BLOCK_SIZE);
+			crypto_xor(dst, final, walk.total % AES_BLOCK_SIZE);
+
+			err = skcipher_walk_done(&walk, 0);
+			break;
+		}
+		err = skcipher_walk_done(&walk,
+					 walk.nbytes - blocks * AES_BLOCK_SIZE);
+	}
+	kernel_neon_end();
+
+	return err;
+}
+
+static int aesbs_xts_setkey(struct crypto_skcipher *tfm, const u8 *in_key,
+			    unsigned int key_len)
+{
+	struct aesbs_xts_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct crypto_aes_ctx rk;
+	int err;
+
+	err = xts_verify_key(tfm, in_key, key_len);
+	if (err)
+		return err;
+
+	key_len /= 2;
+	err = crypto_aes_expand_key(&rk, in_key + key_len, key_len);
+	if (err)
+		return err;
+
+	memcpy(ctx->twkey, rk.key_enc, sizeof(ctx->twkey));
+
+	return aesbs_setkey(tfm, in_key, key_len);
+}
+
+static int __xts_crypt(struct skcipher_request *req,
+		       void (*fn)(u8 out[], u8 const in[], u8 const rk[],
+				  int rounds, int blocks, u8 iv[]))
+{
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct aesbs_xts_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_walk walk;
+	int err;
+
+	err = skcipher_walk_virt(&walk, req, true);
+
+	kernel_neon_begin();
+
+	neon_aes_ecb_encrypt(walk.iv, walk.iv, ctx->twkey,
+			     ctx->key.rounds, 1, 1);
+
+	while (walk.nbytes >= AES_BLOCK_SIZE) {
+		unsigned int blocks = walk.nbytes / AES_BLOCK_SIZE;
+
+		if (walk.nbytes < walk.total)
+			blocks = round_down(blocks,
+					    walk.stride / AES_BLOCK_SIZE);
+
+		fn(walk.dst.virt.addr, walk.src.virt.addr, ctx->key.rk,
+		   ctx->key.rounds, blocks, walk.iv);
+		err = skcipher_walk_done(&walk,
+					 walk.nbytes - blocks * AES_BLOCK_SIZE);
+	}
+	kernel_neon_end();
+
+	return err;
+}
+
+static int xts_encrypt(struct skcipher_request *req)
+{
+	return __xts_crypt(req, aesbs_xts_encrypt);
+}
+
+static int xts_decrypt(struct skcipher_request *req)
+{
+	return __xts_crypt(req, aesbs_xts_decrypt);
+}
+
+static struct skcipher_alg aes_algs[] = { {
+	.base.cra_name		= "__ecb(aes)",
+	.base.cra_driver_name	= "__ecb-aes-neonbs",
+	.base.cra_priority	= 250,
+	.base.cra_blocksize	= AES_BLOCK_SIZE,
+	.base.cra_ctxsize	= sizeof(struct aesbs_ctx),
+	.base.cra_module	= THIS_MODULE,
+	.base.cra_flags		= CRYPTO_ALG_INTERNAL,
+
+	.min_keysize		= AES_MIN_KEY_SIZE,
+	.max_keysize		= AES_MAX_KEY_SIZE,
+	.walksize		= 8 * AES_BLOCK_SIZE,
+	.setkey			= aesbs_setkey,
+	.encrypt		= ecb_encrypt,
+	.decrypt		= ecb_decrypt,
+}, {
+	.base.cra_name		= "__cbc(aes)",
+	.base.cra_driver_name	= "__cbc-aes-neonbs",
+	.base.cra_priority	= 250,
+	.base.cra_blocksize	= AES_BLOCK_SIZE,
+	.base.cra_ctxsize	= sizeof(struct aesbs_cbc_ctx),
+	.base.cra_module	= THIS_MODULE,
+	.base.cra_flags		= CRYPTO_ALG_INTERNAL,
+
+	.min_keysize		= AES_MIN_KEY_SIZE,
+	.max_keysize		= AES_MAX_KEY_SIZE,
+	.walksize		= 8 * AES_BLOCK_SIZE,
+	.ivsize			= AES_BLOCK_SIZE,
+	.setkey			= aesbs_cbc_setkey,
+	.encrypt		= cbc_encrypt,
+	.decrypt		= cbc_decrypt,
+}, {
+	.base.cra_name		= "__ctr(aes)",
+	.base.cra_driver_name	= "__ctr-aes-neonbs",
+	.base.cra_priority	= 250,
+	.base.cra_blocksize	= 1,
+	.base.cra_ctxsize	= sizeof(struct aesbs_ctx),
+	.base.cra_module	= THIS_MODULE,
+	.base.cra_flags		= CRYPTO_ALG_INTERNAL,
+
+	.min_keysize		= AES_MIN_KEY_SIZE,
+	.max_keysize		= AES_MAX_KEY_SIZE,
+	.chunksize		= AES_BLOCK_SIZE,
+	.walksize		= 8 * AES_BLOCK_SIZE,
+	.ivsize			= AES_BLOCK_SIZE,
+	.setkey			= aesbs_setkey,
+	.encrypt		= ctr_encrypt,
+	.decrypt		= ctr_encrypt,
+}, {
+	.base.cra_name		= "ctr(aes)",
+	.base.cra_driver_name	= "ctr-aes-neonbs",
+	.base.cra_priority	= 250 - 1,
+	.base.cra_blocksize	= 1,
+	.base.cra_ctxsize	= sizeof(struct aesbs_ctx),
+	.base.cra_module	= THIS_MODULE,
+
+	.min_keysize		= AES_MIN_KEY_SIZE,
+	.max_keysize		= AES_MAX_KEY_SIZE,
+	.chunksize		= AES_BLOCK_SIZE,
+	.walksize		= 8 * AES_BLOCK_SIZE,
+	.ivsize			= AES_BLOCK_SIZE,
+	.setkey			= aesbs_setkey,
+	.encrypt		= ctr_encrypt,
+	.decrypt		= ctr_encrypt,
+}, {
+	.base.cra_name		= "__xts(aes)",
+	.base.cra_driver_name	= "__xts-aes-neonbs",
+	.base.cra_priority	= 250,
+	.base.cra_blocksize	= AES_BLOCK_SIZE,
+	.base.cra_ctxsize	= sizeof(struct aesbs_xts_ctx),
+	.base.cra_module	= THIS_MODULE,
+	.base.cra_flags		= CRYPTO_ALG_INTERNAL,
+
+	.min_keysize		= 2 * AES_MIN_KEY_SIZE,
+	.max_keysize		= 2 * AES_MAX_KEY_SIZE,
+	.walksize		= 8 * AES_BLOCK_SIZE,
+	.ivsize			= AES_BLOCK_SIZE,
+	.setkey			= aesbs_xts_setkey,
+	.encrypt		= xts_encrypt,
+	.decrypt		= xts_decrypt,
+} };
+
+static struct simd_skcipher_alg *aes_simd_algs[ARRAY_SIZE(aes_algs)];
+
+static void aes_exit(void)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(aes_simd_algs); i++)
+		if (aes_simd_algs[i])
+			simd_skcipher_free(aes_simd_algs[i]);
+
+	crypto_unregister_skciphers(aes_algs, ARRAY_SIZE(aes_algs));
+}
+
+static int __init aes_init(void)
+{
+	struct simd_skcipher_alg *simd;
+	const char *basename;
+	const char *algname;
+	const char *drvname;
+	int err;
+	int i;
+
+	if (!(elf_hwcap & HWCAP_ASIMD))
+		return -ENODEV;
+
+	err = crypto_register_skciphers(aes_algs, ARRAY_SIZE(aes_algs));
+	if (err)
+		return err;
+
+	for (i = 0; i < ARRAY_SIZE(aes_algs); i++) {
+		if (!(aes_algs[i].base.cra_flags & CRYPTO_ALG_INTERNAL))
+			continue;
+
+		algname = aes_algs[i].base.cra_name + 2;
+		drvname = aes_algs[i].base.cra_driver_name + 2;
+		basename = aes_algs[i].base.cra_driver_name;
+		simd = simd_skcipher_create_compat(algname, drvname, basename);
+		err = PTR_ERR(simd);
+		if (IS_ERR(simd))
+			goto unregister_simds;
+
+		aes_simd_algs[i] = simd;
+	}
+	return 0;
+
+unregister_simds:
+	aes_exit();
+	return err;
+}
+
+module_init(aes_init);
+module_exit(aes_exit);
diff --git a/arch/arm64/crypto/chacha20-neon-core.S b/arch/arm64/crypto/chacha20-neon-core.S
new file mode 100644
index 000000000000..13c85e272c2a
--- /dev/null
+++ b/arch/arm64/crypto/chacha20-neon-core.S
@@ -0,0 +1,450 @@
+/*
+ * ChaCha20 256-bit cipher algorithm, RFC7539, arm64 NEON functions
+ *
+ * Copyright (C) 2016 Linaro, Ltd. <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Based on:
+ * ChaCha20 256-bit cipher algorithm, RFC7539, x64 SSSE3 functions
+ *
+ * Copyright (C) 2015 Martin Willi
+ *
+ * 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 <linux/linkage.h>
+
+	.text
+	.align		6
+
+ENTRY(chacha20_block_xor_neon)
+	// x0: Input state matrix, s
+	// x1: 1 data block output, o
+	// x2: 1 data block input, i
+
+	//
+	// This function encrypts one ChaCha20 block by loading the state matrix
+	// in four NEON registers. It performs matrix operation on four words in
+	// parallel, but requires shuffling to rearrange the words after each
+	// round.
+	//
+
+	// x0..3 = s0..3
+	adr		x3, ROT8
+	ld1		{v0.4s-v3.4s}, [x0]
+	ld1		{v8.4s-v11.4s}, [x0]
+	ld1		{v12.4s}, [x3]
+
+	mov		x3, #10
+
+.Ldoubleround:
+	// x0 += x1, x3 = rotl32(x3 ^ x0, 16)
+	add		v0.4s, v0.4s, v1.4s
+	eor		v3.16b, v3.16b, v0.16b
+	rev32		v3.8h, v3.8h
+
+	// x2 += x3, x1 = rotl32(x1 ^ x2, 12)
+	add		v2.4s, v2.4s, v3.4s
+	eor		v4.16b, v1.16b, v2.16b
+	shl		v1.4s, v4.4s, #12
+	sri		v1.4s, v4.4s, #20
+
+	// x0 += x1, x3 = rotl32(x3 ^ x0, 8)
+	add		v0.4s, v0.4s, v1.4s
+	eor		v3.16b, v3.16b, v0.16b
+	tbl		v3.16b, {v3.16b}, v12.16b
+
+	// x2 += x3, x1 = rotl32(x1 ^ x2, 7)
+	add		v2.4s, v2.4s, v3.4s
+	eor		v4.16b, v1.16b, v2.16b
+	shl		v1.4s, v4.4s, #7
+	sri		v1.4s, v4.4s, #25
+
+	// x1 = shuffle32(x1, MASK(0, 3, 2, 1))
+	ext		v1.16b, v1.16b, v1.16b, #4
+	// x2 = shuffle32(x2, MASK(1, 0, 3, 2))
+	ext		v2.16b, v2.16b, v2.16b, #8
+	// x3 = shuffle32(x3, MASK(2, 1, 0, 3))
+	ext		v3.16b, v3.16b, v3.16b, #12
+
+	// x0 += x1, x3 = rotl32(x3 ^ x0, 16)
+	add		v0.4s, v0.4s, v1.4s
+	eor		v3.16b, v3.16b, v0.16b
+	rev32		v3.8h, v3.8h
+
+	// x2 += x3, x1 = rotl32(x1 ^ x2, 12)
+	add		v2.4s, v2.4s, v3.4s
+	eor		v4.16b, v1.16b, v2.16b
+	shl		v1.4s, v4.4s, #12
+	sri		v1.4s, v4.4s, #20
+
+	// x0 += x1, x3 = rotl32(x3 ^ x0, 8)
+	add		v0.4s, v0.4s, v1.4s
+	eor		v3.16b, v3.16b, v0.16b
+	tbl		v3.16b, {v3.16b}, v12.16b
+
+	// x2 += x3, x1 = rotl32(x1 ^ x2, 7)
+	add		v2.4s, v2.4s, v3.4s
+	eor		v4.16b, v1.16b, v2.16b
+	shl		v1.4s, v4.4s, #7
+	sri		v1.4s, v4.4s, #25
+
+	// x1 = shuffle32(x1, MASK(2, 1, 0, 3))
+	ext		v1.16b, v1.16b, v1.16b, #12
+	// x2 = shuffle32(x2, MASK(1, 0, 3, 2))
+	ext		v2.16b, v2.16b, v2.16b, #8
+	// x3 = shuffle32(x3, MASK(0, 3, 2, 1))
+	ext		v3.16b, v3.16b, v3.16b, #4
+
+	subs		x3, x3, #1
+	b.ne		.Ldoubleround
+
+	ld1		{v4.16b-v7.16b}, [x2]
+
+	// o0 = i0 ^ (x0 + s0)
+	add		v0.4s, v0.4s, v8.4s
+	eor		v0.16b, v0.16b, v4.16b
+
+	// o1 = i1 ^ (x1 + s1)
+	add		v1.4s, v1.4s, v9.4s
+	eor		v1.16b, v1.16b, v5.16b
+
+	// o2 = i2 ^ (x2 + s2)
+	add		v2.4s, v2.4s, v10.4s
+	eor		v2.16b, v2.16b, v6.16b
+
+	// o3 = i3 ^ (x3 + s3)
+	add		v3.4s, v3.4s, v11.4s
+	eor		v3.16b, v3.16b, v7.16b
+
+	st1		{v0.16b-v3.16b}, [x1]
+
+	ret
+ENDPROC(chacha20_block_xor_neon)
+
+	.align		6
+ENTRY(chacha20_4block_xor_neon)
+	// x0: Input state matrix, s
+	// x1: 4 data blocks output, o
+	// x2: 4 data blocks input, i
+
+	//
+	// This function encrypts four consecutive ChaCha20 blocks by loading
+	// the state matrix in NEON registers four times. The algorithm performs
+	// each operation on the corresponding word of each state matrix, hence
+	// requires no word shuffling. For final XORing step we transpose the
+	// matrix by interleaving 32- and then 64-bit words, which allows us to
+	// do XOR in NEON registers.
+	//
+	adr		x3, CTRINC		// ... and ROT8
+	ld1		{v30.4s-v31.4s}, [x3]
+
+	// x0..15[0-3] = s0..3[0..3]
+	mov		x4, x0
+	ld4r		{ v0.4s- v3.4s}, [x4], #16
+	ld4r		{ v4.4s- v7.4s}, [x4], #16
+	ld4r		{ v8.4s-v11.4s}, [x4], #16
+	ld4r		{v12.4s-v15.4s}, [x4]
+
+	// x12 += counter values 0-3
+	add		v12.4s, v12.4s, v30.4s
+
+	mov		x3, #10
+
+.Ldoubleround4:
+	// x0 += x4, x12 = rotl32(x12 ^ x0, 16)
+	// x1 += x5, x13 = rotl32(x13 ^ x1, 16)
+	// x2 += x6, x14 = rotl32(x14 ^ x2, 16)
+	// x3 += x7, x15 = rotl32(x15 ^ x3, 16)
+	add		v0.4s, v0.4s, v4.4s
+	add		v1.4s, v1.4s, v5.4s
+	add		v2.4s, v2.4s, v6.4s
+	add		v3.4s, v3.4s, v7.4s
+
+	eor		v12.16b, v12.16b, v0.16b
+	eor		v13.16b, v13.16b, v1.16b
+	eor		v14.16b, v14.16b, v2.16b
+	eor		v15.16b, v15.16b, v3.16b
+
+	rev32		v12.8h, v12.8h
+	rev32		v13.8h, v13.8h
+	rev32		v14.8h, v14.8h
+	rev32		v15.8h, v15.8h
+
+	// x8 += x12, x4 = rotl32(x4 ^ x8, 12)
+	// x9 += x13, x5 = rotl32(x5 ^ x9, 12)
+	// x10 += x14, x6 = rotl32(x6 ^ x10, 12)
+	// x11 += x15, x7 = rotl32(x7 ^ x11, 12)
+	add		v8.4s, v8.4s, v12.4s
+	add		v9.4s, v9.4s, v13.4s
+	add		v10.4s, v10.4s, v14.4s
+	add		v11.4s, v11.4s, v15.4s
+
+	eor		v16.16b, v4.16b, v8.16b
+	eor		v17.16b, v5.16b, v9.16b
+	eor		v18.16b, v6.16b, v10.16b
+	eor		v19.16b, v7.16b, v11.16b
+
+	shl		v4.4s, v16.4s, #12
+	shl		v5.4s, v17.4s, #12
+	shl		v6.4s, v18.4s, #12
+	shl		v7.4s, v19.4s, #12
+
+	sri		v4.4s, v16.4s, #20
+	sri		v5.4s, v17.4s, #20
+	sri		v6.4s, v18.4s, #20
+	sri		v7.4s, v19.4s, #20
+
+	// x0 += x4, x12 = rotl32(x12 ^ x0, 8)
+	// x1 += x5, x13 = rotl32(x13 ^ x1, 8)
+	// x2 += x6, x14 = rotl32(x14 ^ x2, 8)
+	// x3 += x7, x15 = rotl32(x15 ^ x3, 8)
+	add		v0.4s, v0.4s, v4.4s
+	add		v1.4s, v1.4s, v5.4s
+	add		v2.4s, v2.4s, v6.4s
+	add		v3.4s, v3.4s, v7.4s
+
+	eor		v12.16b, v12.16b, v0.16b
+	eor		v13.16b, v13.16b, v1.16b
+	eor		v14.16b, v14.16b, v2.16b
+	eor		v15.16b, v15.16b, v3.16b
+
+	tbl		v12.16b, {v12.16b}, v31.16b
+	tbl		v13.16b, {v13.16b}, v31.16b
+	tbl		v14.16b, {v14.16b}, v31.16b
+	tbl		v15.16b, {v15.16b}, v31.16b
+
+	// x8 += x12, x4 = rotl32(x4 ^ x8, 7)
+	// x9 += x13, x5 = rotl32(x5 ^ x9, 7)
+	// x10 += x14, x6 = rotl32(x6 ^ x10, 7)
+	// x11 += x15, x7 = rotl32(x7 ^ x11, 7)
+	add		v8.4s, v8.4s, v12.4s
+	add		v9.4s, v9.4s, v13.4s
+	add		v10.4s, v10.4s, v14.4s
+	add		v11.4s, v11.4s, v15.4s
+
+	eor		v16.16b, v4.16b, v8.16b
+	eor		v17.16b, v5.16b, v9.16b
+	eor		v18.16b, v6.16b, v10.16b
+	eor		v19.16b, v7.16b, v11.16b
+
+	shl		v4.4s, v16.4s, #7
+	shl		v5.4s, v17.4s, #7
+	shl		v6.4s, v18.4s, #7
+	shl		v7.4s, v19.4s, #7
+
+	sri		v4.4s, v16.4s, #25
+	sri		v5.4s, v17.4s, #25
+	sri		v6.4s, v18.4s, #25
+	sri		v7.4s, v19.4s, #25
+
+	// x0 += x5, x15 = rotl32(x15 ^ x0, 16)
+	// x1 += x6, x12 = rotl32(x12 ^ x1, 16)
+	// x2 += x7, x13 = rotl32(x13 ^ x2, 16)
+	// x3 += x4, x14 = rotl32(x14 ^ x3, 16)
+	add		v0.4s, v0.4s, v5.4s
+	add		v1.4s, v1.4s, v6.4s
+	add		v2.4s, v2.4s, v7.4s
+	add		v3.4s, v3.4s, v4.4s
+
+	eor		v15.16b, v15.16b, v0.16b
+	eor		v12.16b, v12.16b, v1.16b
+	eor		v13.16b, v13.16b, v2.16b
+	eor		v14.16b, v14.16b, v3.16b
+
+	rev32		v15.8h, v15.8h
+	rev32		v12.8h, v12.8h
+	rev32		v13.8h, v13.8h
+	rev32		v14.8h, v14.8h
+
+	// x10 += x15, x5 = rotl32(x5 ^ x10, 12)
+	// x11 += x12, x6 = rotl32(x6 ^ x11, 12)
+	// x8 += x13, x7 = rotl32(x7 ^ x8, 12)
+	// x9 += x14, x4 = rotl32(x4 ^ x9, 12)
+	add		v10.4s, v10.4s, v15.4s
+	add		v11.4s, v11.4s, v12.4s
+	add		v8.4s, v8.4s, v13.4s
+	add		v9.4s, v9.4s, v14.4s
+
+	eor		v16.16b, v5.16b, v10.16b
+	eor		v17.16b, v6.16b, v11.16b
+	eor		v18.16b, v7.16b, v8.16b
+	eor		v19.16b, v4.16b, v9.16b
+
+	shl		v5.4s, v16.4s, #12
+	shl		v6.4s, v17.4s, #12
+	shl		v7.4s, v18.4s, #12
+	shl		v4.4s, v19.4s, #12
+
+	sri		v5.4s, v16.4s, #20
+	sri		v6.4s, v17.4s, #20
+	sri		v7.4s, v18.4s, #20
+	sri		v4.4s, v19.4s, #20
+
+	// x0 += x5, x15 = rotl32(x15 ^ x0, 8)
+	// x1 += x6, x12 = rotl32(x12 ^ x1, 8)
+	// x2 += x7, x13 = rotl32(x13 ^ x2, 8)
+	// x3 += x4, x14 = rotl32(x14 ^ x3, 8)
+	add		v0.4s, v0.4s, v5.4s
+	add		v1.4s, v1.4s, v6.4s
+	add		v2.4s, v2.4s, v7.4s
+	add		v3.4s, v3.4s, v4.4s
+
+	eor		v15.16b, v15.16b, v0.16b
+	eor		v12.16b, v12.16b, v1.16b
+	eor		v13.16b, v13.16b, v2.16b
+	eor		v14.16b, v14.16b, v3.16b
+
+	tbl		v15.16b, {v15.16b}, v31.16b
+	tbl		v12.16b, {v12.16b}, v31.16b
+	tbl		v13.16b, {v13.16b}, v31.16b
+	tbl		v14.16b, {v14.16b}, v31.16b
+
+	// x10 += x15, x5 = rotl32(x5 ^ x10, 7)
+	// x11 += x12, x6 = rotl32(x6 ^ x11, 7)
+	// x8 += x13, x7 = rotl32(x7 ^ x8, 7)
+	// x9 += x14, x4 = rotl32(x4 ^ x9, 7)
+	add		v10.4s, v10.4s, v15.4s
+	add		v11.4s, v11.4s, v12.4s
+	add		v8.4s, v8.4s, v13.4s
+	add		v9.4s, v9.4s, v14.4s
+
+	eor		v16.16b, v5.16b, v10.16b
+	eor		v17.16b, v6.16b, v11.16b
+	eor		v18.16b, v7.16b, v8.16b
+	eor		v19.16b, v4.16b, v9.16b
+
+	shl		v5.4s, v16.4s, #7
+	shl		v6.4s, v17.4s, #7
+	shl		v7.4s, v18.4s, #7
+	shl		v4.4s, v19.4s, #7
+
+	sri		v5.4s, v16.4s, #25
+	sri		v6.4s, v17.4s, #25
+	sri		v7.4s, v18.4s, #25
+	sri		v4.4s, v19.4s, #25
+
+	subs		x3, x3, #1
+	b.ne		.Ldoubleround4
+
+	ld4r		{v16.4s-v19.4s}, [x0], #16
+	ld4r		{v20.4s-v23.4s}, [x0], #16
+
+	// x12 += counter values 0-3
+	add		v12.4s, v12.4s, v30.4s
+
+	// x0[0-3] += s0[0]
+	// x1[0-3] += s0[1]
+	// x2[0-3] += s0[2]
+	// x3[0-3] += s0[3]
+	add		v0.4s, v0.4s, v16.4s
+	add		v1.4s, v1.4s, v17.4s
+	add		v2.4s, v2.4s, v18.4s
+	add		v3.4s, v3.4s, v19.4s
+
+	ld4r		{v24.4s-v27.4s}, [x0], #16
+	ld4r		{v28.4s-v31.4s}, [x0]
+
+	// x4[0-3] += s1[0]
+	// x5[0-3] += s1[1]
+	// x6[0-3] += s1[2]
+	// x7[0-3] += s1[3]
+	add		v4.4s, v4.4s, v20.4s
+	add		v5.4s, v5.4s, v21.4s
+	add		v6.4s, v6.4s, v22.4s
+	add		v7.4s, v7.4s, v23.4s
+
+	// x8[0-3] += s2[0]
+	// x9[0-3] += s2[1]
+	// x10[0-3] += s2[2]
+	// x11[0-3] += s2[3]
+	add		v8.4s, v8.4s, v24.4s
+	add		v9.4s, v9.4s, v25.4s
+	add		v10.4s, v10.4s, v26.4s
+	add		v11.4s, v11.4s, v27.4s
+
+	// x12[0-3] += s3[0]
+	// x13[0-3] += s3[1]
+	// x14[0-3] += s3[2]
+	// x15[0-3] += s3[3]
+	add		v12.4s, v12.4s, v28.4s
+	add		v13.4s, v13.4s, v29.4s
+	add		v14.4s, v14.4s, v30.4s
+	add		v15.4s, v15.4s, v31.4s
+
+	// interleave 32-bit words in state n, n+1
+	zip1		v16.4s, v0.4s, v1.4s
+	zip2		v17.4s, v0.4s, v1.4s
+	zip1		v18.4s, v2.4s, v3.4s
+	zip2		v19.4s, v2.4s, v3.4s
+	zip1		v20.4s, v4.4s, v5.4s
+	zip2		v21.4s, v4.4s, v5.4s
+	zip1		v22.4s, v6.4s, v7.4s
+	zip2		v23.4s, v6.4s, v7.4s
+	zip1		v24.4s, v8.4s, v9.4s
+	zip2		v25.4s, v8.4s, v9.4s
+	zip1		v26.4s, v10.4s, v11.4s
+	zip2		v27.4s, v10.4s, v11.4s
+	zip1		v28.4s, v12.4s, v13.4s
+	zip2		v29.4s, v12.4s, v13.4s
+	zip1		v30.4s, v14.4s, v15.4s
+	zip2		v31.4s, v14.4s, v15.4s
+
+	// interleave 64-bit words in state n, n+2
+	zip1		v0.2d, v16.2d, v18.2d
+	zip2		v4.2d, v16.2d, v18.2d
+	zip1		v8.2d, v17.2d, v19.2d
+	zip2		v12.2d, v17.2d, v19.2d
+	ld1		{v16.16b-v19.16b}, [x2], #64
+
+	zip1		v1.2d, v20.2d, v22.2d
+	zip2		v5.2d, v20.2d, v22.2d
+	zip1		v9.2d, v21.2d, v23.2d
+	zip2		v13.2d, v21.2d, v23.2d
+	ld1		{v20.16b-v23.16b}, [x2], #64
+
+	zip1		v2.2d, v24.2d, v26.2d
+	zip2		v6.2d, v24.2d, v26.2d
+	zip1		v10.2d, v25.2d, v27.2d
+	zip2		v14.2d, v25.2d, v27.2d
+	ld1		{v24.16b-v27.16b}, [x2], #64
+
+	zip1		v3.2d, v28.2d, v30.2d
+	zip2		v7.2d, v28.2d, v30.2d
+	zip1		v11.2d, v29.2d, v31.2d
+	zip2		v15.2d, v29.2d, v31.2d
+	ld1		{v28.16b-v31.16b}, [x2]
+
+	// xor with corresponding input, write to output
+	eor		v16.16b, v16.16b, v0.16b
+	eor		v17.16b, v17.16b, v1.16b
+	eor		v18.16b, v18.16b, v2.16b
+	eor		v19.16b, v19.16b, v3.16b
+	eor		v20.16b, v20.16b, v4.16b
+	eor		v21.16b, v21.16b, v5.16b
+	st1		{v16.16b-v19.16b}, [x1], #64
+	eor		v22.16b, v22.16b, v6.16b
+	eor		v23.16b, v23.16b, v7.16b
+	eor		v24.16b, v24.16b, v8.16b
+	eor		v25.16b, v25.16b, v9.16b
+	st1		{v20.16b-v23.16b}, [x1], #64
+	eor		v26.16b, v26.16b, v10.16b
+	eor		v27.16b, v27.16b, v11.16b
+	eor		v28.16b, v28.16b, v12.16b
+	st1		{v24.16b-v27.16b}, [x1], #64
+	eor		v29.16b, v29.16b, v13.16b
+	eor		v30.16b, v30.16b, v14.16b
+	eor		v31.16b, v31.16b, v15.16b
+	st1		{v28.16b-v31.16b}, [x1]
+
+	ret
+ENDPROC(chacha20_4block_xor_neon)
+
+CTRINC:	.word		0, 1, 2, 3
+ROT8:	.word		0x02010003, 0x06050407, 0x0a09080b, 0x0e0d0c0f
diff --git a/arch/arm64/crypto/chacha20-neon-glue.c b/arch/arm64/crypto/chacha20-neon-glue.c
new file mode 100644
index 000000000000..a7cd575ea223
--- /dev/null
+++ b/arch/arm64/crypto/chacha20-neon-glue.c
@@ -0,0 +1,126 @@
+/*
+ * ChaCha20 256-bit cipher algorithm, RFC7539, arm64 NEON functions
+ *
+ * Copyright (C) 2016 Linaro, Ltd. <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Based on:
+ * ChaCha20 256-bit cipher algorithm, RFC7539, SIMD glue code
+ *
+ * Copyright (C) 2015 Martin Willi
+ *
+ * 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/algapi.h>
+#include <crypto/chacha20.h>
+#include <crypto/internal/skcipher.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+#include <asm/hwcap.h>
+#include <asm/neon.h>
+
+asmlinkage void chacha20_block_xor_neon(u32 *state, u8 *dst, const u8 *src);
+asmlinkage void chacha20_4block_xor_neon(u32 *state, u8 *dst, const u8 *src);
+
+static void chacha20_doneon(u32 *state, u8 *dst, const u8 *src,
+			    unsigned int bytes)
+{
+	u8 buf[CHACHA20_BLOCK_SIZE];
+
+	while (bytes >= CHACHA20_BLOCK_SIZE * 4) {
+		chacha20_4block_xor_neon(state, dst, src);
+		bytes -= CHACHA20_BLOCK_SIZE * 4;
+		src += CHACHA20_BLOCK_SIZE * 4;
+		dst += CHACHA20_BLOCK_SIZE * 4;
+		state[12] += 4;
+	}
+	while (bytes >= CHACHA20_BLOCK_SIZE) {
+		chacha20_block_xor_neon(state, dst, src);
+		bytes -= CHACHA20_BLOCK_SIZE;
+		src += CHACHA20_BLOCK_SIZE;
+		dst += CHACHA20_BLOCK_SIZE;
+		state[12]++;
+	}
+	if (bytes) {
+		memcpy(buf, src, bytes);
+		chacha20_block_xor_neon(state, buf, buf);
+		memcpy(dst, buf, bytes);
+	}
+}
+
+static int chacha20_neon(struct skcipher_request *req)
+{
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct chacha20_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_walk walk;
+	u32 state[16];
+	int err;
+
+	if (req->cryptlen <= CHACHA20_BLOCK_SIZE)
+		return crypto_chacha20_crypt(req);
+
+	err = skcipher_walk_virt(&walk, req, true);
+
+	crypto_chacha20_init(state, ctx, walk.iv);
+
+	kernel_neon_begin();
+	while (walk.nbytes > 0) {
+		unsigned int nbytes = walk.nbytes;
+
+		if (nbytes < walk.total)
+			nbytes = round_down(nbytes, walk.stride);
+
+		chacha20_doneon(state, walk.dst.virt.addr, walk.src.virt.addr,
+				nbytes);
+		err = skcipher_walk_done(&walk, walk.nbytes - nbytes);
+	}
+	kernel_neon_end();
+
+	return err;
+}
+
+static struct skcipher_alg alg = {
+	.base.cra_name		= "chacha20",
+	.base.cra_driver_name	= "chacha20-neon",
+	.base.cra_priority	= 300,
+	.base.cra_blocksize	= 1,
+	.base.cra_ctxsize	= sizeof(struct chacha20_ctx),
+	.base.cra_module	= THIS_MODULE,
+
+	.min_keysize		= CHACHA20_KEY_SIZE,
+	.max_keysize		= CHACHA20_KEY_SIZE,
+	.ivsize			= CHACHA20_IV_SIZE,
+	.chunksize		= CHACHA20_BLOCK_SIZE,
+	.walksize		= 4 * CHACHA20_BLOCK_SIZE,
+	.setkey			= crypto_chacha20_setkey,
+	.encrypt		= chacha20_neon,
+	.decrypt		= chacha20_neon,
+};
+
+static int __init chacha20_simd_mod_init(void)
+{
+	if (!(elf_hwcap & HWCAP_ASIMD))
+		return -ENODEV;
+
+	return crypto_register_skcipher(&alg);
+}
+
+static void __exit chacha20_simd_mod_fini(void)
+{
+	crypto_unregister_skcipher(&alg);
+}
+
+module_init(chacha20_simd_mod_init);
+module_exit(chacha20_simd_mod_fini);
+
+MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS_CRYPTO("chacha20");
diff --git a/arch/arm64/crypto/crc32-arm64.c b/arch/arm64/crypto/crc32-arm64.c
deleted file mode 100644
index 6a37c3c6b11d..000000000000
--- a/arch/arm64/crypto/crc32-arm64.c
+++ /dev/null
@@ -1,290 +0,0 @@
-/*
- * crc32-arm64.c - CRC32 and CRC32C using optional ARMv8 instructions
- *
- * Module based on crypto/crc32c_generic.c
- *
- * CRC32 loop taken from Ed Nevill's Hadoop CRC patch
- * http://mail-archives.apache.org/mod_mbox/hadoop-common-dev/201406.mbox/%3C1403687030.3355.19.camel%40localhost.localdomain%3E
- *
- * Using inline assembly instead of intrinsics in order to be backwards
- * compatible with older compilers.
- *
- * Copyright (C) 2014 Linaro Ltd <yazen.ghannam@linaro.org>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#include <linux/unaligned/access_ok.h>
-#include <linux/cpufeature.h>
-#include <linux/init.h>
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/string.h>
-
-#include <crypto/internal/hash.h>
-
-MODULE_AUTHOR("Yazen Ghannam <yazen.ghannam@linaro.org>");
-MODULE_DESCRIPTION("CRC32 and CRC32C using optional ARMv8 instructions");
-MODULE_LICENSE("GPL v2");
-
-#define CRC32X(crc, value) __asm__("crc32x %w[c], %w[c], %x[v]":[c]"+r"(crc):[v]"r"(value))
-#define CRC32W(crc, value) __asm__("crc32w %w[c], %w[c], %w[v]":[c]"+r"(crc):[v]"r"(value))
-#define CRC32H(crc, value) __asm__("crc32h %w[c], %w[c], %w[v]":[c]"+r"(crc):[v]"r"(value))
-#define CRC32B(crc, value) __asm__("crc32b %w[c], %w[c], %w[v]":[c]"+r"(crc):[v]"r"(value))
-#define CRC32CX(crc, value) __asm__("crc32cx %w[c], %w[c], %x[v]":[c]"+r"(crc):[v]"r"(value))
-#define CRC32CW(crc, value) __asm__("crc32cw %w[c], %w[c], %w[v]":[c]"+r"(crc):[v]"r"(value))
-#define CRC32CH(crc, value) __asm__("crc32ch %w[c], %w[c], %w[v]":[c]"+r"(crc):[v]"r"(value))
-#define CRC32CB(crc, value) __asm__("crc32cb %w[c], %w[c], %w[v]":[c]"+r"(crc):[v]"r"(value))
-
-static u32 crc32_arm64_le_hw(u32 crc, const u8 *p, unsigned int len)
-{
-	s64 length = len;
-
-	while ((length -= sizeof(u64)) >= 0) {
-		CRC32X(crc, get_unaligned_le64(p));
-		p += sizeof(u64);
-	}
-
-	/* The following is more efficient than the straight loop */
-	if (length & sizeof(u32)) {
-		CRC32W(crc, get_unaligned_le32(p));
-		p += sizeof(u32);
-	}
-	if (length & sizeof(u16)) {
-		CRC32H(crc, get_unaligned_le16(p));
-		p += sizeof(u16);
-	}
-	if (length & sizeof(u8))
-		CRC32B(crc, *p);
-
-	return crc;
-}
-
-static u32 crc32c_arm64_le_hw(u32 crc, const u8 *p, unsigned int len)
-{
-	s64 length = len;
-
-	while ((length -= sizeof(u64)) >= 0) {
-		CRC32CX(crc, get_unaligned_le64(p));
-		p += sizeof(u64);
-	}
-
-	/* The following is more efficient than the straight loop */
-	if (length & sizeof(u32)) {
-		CRC32CW(crc, get_unaligned_le32(p));
-		p += sizeof(u32);
-	}
-	if (length & sizeof(u16)) {
-		CRC32CH(crc, get_unaligned_le16(p));
-		p += sizeof(u16);
-	}
-	if (length & sizeof(u8))
-		CRC32CB(crc, *p);
-
-	return crc;
-}
-
-#define CHKSUM_BLOCK_SIZE	1
-#define CHKSUM_DIGEST_SIZE	4
-
-struct chksum_ctx {
-	u32 key;
-};
-
-struct chksum_desc_ctx {
-	u32 crc;
-};
-
-static int chksum_init(struct shash_desc *desc)
-{
-	struct chksum_ctx *mctx = crypto_shash_ctx(desc->tfm);
-	struct chksum_desc_ctx *ctx = shash_desc_ctx(desc);
-
-	ctx->crc = mctx->key;
-
-	return 0;
-}
-
-/*
- * Setting the seed allows arbitrary accumulators and flexible XOR policy
- * If your algorithm starts with ~0, then XOR with ~0 before you set
- * the seed.
- */
-static int chksum_setkey(struct crypto_shash *tfm, const u8 *key,
-			 unsigned int keylen)
-{
-	struct chksum_ctx *mctx = crypto_shash_ctx(tfm);
-
-	if (keylen != sizeof(mctx->key)) {
-		crypto_shash_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
-		return -EINVAL;
-	}
-	mctx->key = get_unaligned_le32(key);
-	return 0;
-}
-
-static int chksum_update(struct shash_desc *desc, const u8 *data,
-			 unsigned int length)
-{
-	struct chksum_desc_ctx *ctx = shash_desc_ctx(desc);
-
-	ctx->crc = crc32_arm64_le_hw(ctx->crc, data, length);
-	return 0;
-}
-
-static int chksumc_update(struct shash_desc *desc, const u8 *data,
-			 unsigned int length)
-{
-	struct chksum_desc_ctx *ctx = shash_desc_ctx(desc);
-
-	ctx->crc = crc32c_arm64_le_hw(ctx->crc, data, length);
-	return 0;
-}
-
-static int chksum_final(struct shash_desc *desc, u8 *out)
-{
-	struct chksum_desc_ctx *ctx = shash_desc_ctx(desc);
-
-	put_unaligned_le32(ctx->crc, out);
-	return 0;
-}
-
-static int chksumc_final(struct shash_desc *desc, u8 *out)
-{
-	struct chksum_desc_ctx *ctx = shash_desc_ctx(desc);
-
-	put_unaligned_le32(~ctx->crc, out);
-	return 0;
-}
-
-static int __chksum_finup(u32 crc, const u8 *data, unsigned int len, u8 *out)
-{
-	put_unaligned_le32(crc32_arm64_le_hw(crc, data, len), out);
-	return 0;
-}
-
-static int __chksumc_finup(u32 crc, const u8 *data, unsigned int len, u8 *out)
-{
-	put_unaligned_le32(~crc32c_arm64_le_hw(crc, data, len), out);
-	return 0;
-}
-
-static int chksum_finup(struct shash_desc *desc, const u8 *data,
-			unsigned int len, u8 *out)
-{
-	struct chksum_desc_ctx *ctx = shash_desc_ctx(desc);
-
-	return __chksum_finup(ctx->crc, data, len, out);
-}
-
-static int chksumc_finup(struct shash_desc *desc, const u8 *data,
-			unsigned int len, u8 *out)
-{
-	struct chksum_desc_ctx *ctx = shash_desc_ctx(desc);
-
-	return __chksumc_finup(ctx->crc, data, len, out);
-}
-
-static int chksum_digest(struct shash_desc *desc, const u8 *data,
-			 unsigned int length, u8 *out)
-{
-	struct chksum_ctx *mctx = crypto_shash_ctx(desc->tfm);
-
-	return __chksum_finup(mctx->key, data, length, out);
-}
-
-static int chksumc_digest(struct shash_desc *desc, const u8 *data,
-			 unsigned int length, u8 *out)
-{
-	struct chksum_ctx *mctx = crypto_shash_ctx(desc->tfm);
-
-	return __chksumc_finup(mctx->key, data, length, out);
-}
-
-static int crc32_cra_init(struct crypto_tfm *tfm)
-{
-	struct chksum_ctx *mctx = crypto_tfm_ctx(tfm);
-
-	mctx->key = 0;
-	return 0;
-}
-
-static int crc32c_cra_init(struct crypto_tfm *tfm)
-{
-	struct chksum_ctx *mctx = crypto_tfm_ctx(tfm);
-
-	mctx->key = ~0;
-	return 0;
-}
-
-static struct shash_alg crc32_alg = {
-	.digestsize		=	CHKSUM_DIGEST_SIZE,
-	.setkey			=	chksum_setkey,
-	.init			=	chksum_init,
-	.update			=	chksum_update,
-	.final			=	chksum_final,
-	.finup			=	chksum_finup,
-	.digest			=	chksum_digest,
-	.descsize		=	sizeof(struct chksum_desc_ctx),
-	.base			=	{
-		.cra_name		=	"crc32",
-		.cra_driver_name	=	"crc32-arm64-hw",
-		.cra_priority		=	300,
-		.cra_blocksize		=	CHKSUM_BLOCK_SIZE,
-		.cra_alignmask		=	0,
-		.cra_ctxsize		=	sizeof(struct chksum_ctx),
-		.cra_module		=	THIS_MODULE,
-		.cra_init		=	crc32_cra_init,
-	}
-};
-
-static struct shash_alg crc32c_alg = {
-	.digestsize		=	CHKSUM_DIGEST_SIZE,
-	.setkey			=	chksum_setkey,
-	.init			=	chksum_init,
-	.update			=	chksumc_update,
-	.final			=	chksumc_final,
-	.finup			=	chksumc_finup,
-	.digest			=	chksumc_digest,
-	.descsize		=	sizeof(struct chksum_desc_ctx),
-	.base			=	{
-		.cra_name		=	"crc32c",
-		.cra_driver_name	=	"crc32c-arm64-hw",
-		.cra_priority		=	300,
-		.cra_blocksize		=	CHKSUM_BLOCK_SIZE,
-		.cra_alignmask		=	0,
-		.cra_ctxsize		=	sizeof(struct chksum_ctx),
-		.cra_module		=	THIS_MODULE,
-		.cra_init		=	crc32c_cra_init,
-	}
-};
-
-static int __init crc32_mod_init(void)
-{
-	int err;
-
-	err = crypto_register_shash(&crc32_alg);
-
-	if (err)
-		return err;
-
-	err = crypto_register_shash(&crc32c_alg);
-
-	if (err) {
-		crypto_unregister_shash(&crc32_alg);
-		return err;
-	}
-
-	return 0;
-}
-
-static void __exit crc32_mod_exit(void)
-{
-	crypto_unregister_shash(&crc32_alg);
-	crypto_unregister_shash(&crc32c_alg);
-}
-
-module_cpu_feature_match(CRC32, crc32_mod_init);
-module_exit(crc32_mod_exit);
diff --git a/arch/arm64/crypto/crc32-ce-glue.c b/arch/arm64/crypto/crc32-ce-glue.c
index 8594127d5e01..eccb1ae90064 100644
--- a/arch/arm64/crypto/crc32-ce-glue.c
+++ b/arch/arm64/crypto/crc32-ce-glue.c
@@ -72,6 +72,24 @@ static int crc32_pmull_init(struct shash_desc *desc)
 	return 0;
 }
 
+static int crc32_update(struct shash_desc *desc, const u8 *data,
+			unsigned int length)
+{
+	u32 *crc = shash_desc_ctx(desc);
+
+	*crc = crc32_armv8_le(*crc, data, length);
+	return 0;
+}
+
+static int crc32c_update(struct shash_desc *desc, const u8 *data,
+			 unsigned int length)
+{
+	u32 *crc = shash_desc_ctx(desc);
+
+	*crc = crc32c_armv8_le(*crc, data, length);
+	return 0;
+}
+
 static int crc32_pmull_update(struct shash_desc *desc, const u8 *data,
 			 unsigned int length)
 {
@@ -156,7 +174,7 @@ static int crc32c_pmull_final(struct shash_desc *desc, u8 *out)
 static struct shash_alg crc32_pmull_algs[] = { {
 	.setkey			= crc32_pmull_setkey,
 	.init			= crc32_pmull_init,
-	.update			= crc32_pmull_update,
+	.update			= crc32_update,
 	.final			= crc32_pmull_final,
 	.descsize		= sizeof(u32),
 	.digestsize		= sizeof(u32),
@@ -171,7 +189,7 @@ static struct shash_alg crc32_pmull_algs[] = { {
 }, {
 	.setkey			= crc32_pmull_setkey,
 	.init			= crc32_pmull_init,
-	.update			= crc32c_pmull_update,
+	.update			= crc32c_update,
 	.final			= crc32c_pmull_final,
 	.descsize		= sizeof(u32),
 	.digestsize		= sizeof(u32),
@@ -187,14 +205,20 @@ static struct shash_alg crc32_pmull_algs[] = { {
 
 static int __init crc32_pmull_mod_init(void)
 {
-	if (elf_hwcap & HWCAP_CRC32) {
-		fallback_crc32 = crc32_armv8_le;
-		fallback_crc32c = crc32c_armv8_le;
-	} else {
-		fallback_crc32 = crc32_le;
-		fallback_crc32c = __crc32c_le;
+	if (IS_ENABLED(CONFIG_KERNEL_MODE_NEON) && (elf_hwcap & HWCAP_PMULL)) {
+		crc32_pmull_algs[0].update = crc32_pmull_update;
+		crc32_pmull_algs[1].update = crc32c_pmull_update;
+
+		if (elf_hwcap & HWCAP_CRC32) {
+			fallback_crc32 = crc32_armv8_le;
+			fallback_crc32c = crc32c_armv8_le;
+		} else {
+			fallback_crc32 = crc32_le;
+			fallback_crc32c = __crc32c_le;
+		}
+	} else if (!(elf_hwcap & HWCAP_CRC32)) {
+		return -ENODEV;
 	}
-
 	return crypto_register_shashes(crc32_pmull_algs,
 				       ARRAY_SIZE(crc32_pmull_algs));
 }
@@ -205,7 +229,12 @@ static void __exit crc32_pmull_mod_exit(void)
 				  ARRAY_SIZE(crc32_pmull_algs));
 }
 
-module_cpu_feature_match(PMULL, crc32_pmull_mod_init);
+static const struct cpu_feature crc32_cpu_feature[] = {
+	{ cpu_feature(CRC32) }, { cpu_feature(PMULL) }, { }
+};
+MODULE_DEVICE_TABLE(cpu, crc32_cpu_feature);
+
+module_init(crc32_pmull_mod_init);
 module_exit(crc32_pmull_mod_exit);
 
 MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S
index 383a6f84a060..3c465184ff8a 100644
--- a/arch/x86/crypto/aesni-intel_asm.S
+++ b/arch/x86/crypto/aesni-intel_asm.S
@@ -46,28 +46,49 @@
 
 #ifdef __x86_64__
 
-.data
+# constants in mergeable sections, linker can reorder and merge
+.section	.rodata.cst16.gf128mul_x_ble_mask, "aM", @progbits, 16
 .align 16
 .Lgf128mul_x_ble_mask:
 	.octa 0x00000000000000010000000000000087
+.section	.rodata.cst16.POLY, "aM", @progbits, 16
+.align 16
 POLY:   .octa 0xC2000000000000000000000000000001
+.section	.rodata.cst16.TWOONE, "aM", @progbits, 16
+.align 16
 TWOONE: .octa 0x00000001000000000000000000000001
 
-# order of these constants should not change.
-# more specifically, ALL_F should follow SHIFT_MASK,
-# and ZERO should follow ALL_F
-
+.section	.rodata.cst16.SHUF_MASK, "aM", @progbits, 16
+.align 16
 SHUF_MASK:  .octa 0x000102030405060708090A0B0C0D0E0F
+.section	.rodata.cst16.MASK1, "aM", @progbits, 16
+.align 16
 MASK1:      .octa 0x0000000000000000ffffffffffffffff
+.section	.rodata.cst16.MASK2, "aM", @progbits, 16
+.align 16
 MASK2:      .octa 0xffffffffffffffff0000000000000000
-SHIFT_MASK: .octa 0x0f0e0d0c0b0a09080706050403020100
-ALL_F:      .octa 0xffffffffffffffffffffffffffffffff
-ZERO:       .octa 0x00000000000000000000000000000000
+.section	.rodata.cst16.ONE, "aM", @progbits, 16
+.align 16
 ONE:        .octa 0x00000000000000000000000000000001
+.section	.rodata.cst16.F_MIN_MASK, "aM", @progbits, 16
+.align 16
 F_MIN_MASK: .octa 0xf1f2f3f4f5f6f7f8f9fafbfcfdfeff0
+.section	.rodata.cst16.dec, "aM", @progbits, 16
+.align 16
 dec:        .octa 0x1
+.section	.rodata.cst16.enc, "aM", @progbits, 16
+.align 16
 enc:        .octa 0x2
 
+# order of these constants should not change.
+# more specifically, ALL_F should follow SHIFT_MASK,
+# and zero should follow ALL_F
+.section	.rodata, "a", @progbits
+.align 16
+SHIFT_MASK: .octa 0x0f0e0d0c0b0a09080706050403020100
+ALL_F:      .octa 0xffffffffffffffffffffffffffffffff
+            .octa 0x00000000000000000000000000000000
+
 
 .text
 
diff --git a/arch/x86/crypto/aesni-intel_avx-x86_64.S b/arch/x86/crypto/aesni-intel_avx-x86_64.S
index 522ab68d1c88..d664382c6e56 100644
--- a/arch/x86/crypto/aesni-intel_avx-x86_64.S
+++ b/arch/x86/crypto/aesni-intel_avx-x86_64.S
@@ -122,23 +122,39 @@
 #include <linux/linkage.h>
 #include <asm/inst.h>
 
-.data
+# constants in mergeable sections, linker can reorder and merge
+.section	.rodata.cst16.POLY, "aM", @progbits, 16
 .align 16
-
 POLY:            .octa     0xC2000000000000000000000000000001
+
+.section	.rodata.cst16.POLY2, "aM", @progbits, 16
+.align 16
 POLY2:           .octa     0xC20000000000000000000001C2000000
-TWOONE:          .octa     0x00000001000000000000000000000001
 
-# order of these constants should not change.
-# more specifically, ALL_F should follow SHIFT_MASK, and ZERO should follow ALL_F
+.section	.rodata.cst16.TWOONE, "aM", @progbits, 16
+.align 16
+TWOONE:          .octa     0x00000001000000000000000000000001
 
+.section	.rodata.cst16.SHUF_MASK, "aM", @progbits, 16
+.align 16
 SHUF_MASK:       .octa     0x000102030405060708090A0B0C0D0E0F
-SHIFT_MASK:      .octa     0x0f0e0d0c0b0a09080706050403020100
-ALL_F:           .octa     0xffffffffffffffffffffffffffffffff
-ZERO:            .octa     0x00000000000000000000000000000000
+
+.section	.rodata.cst16.ONE, "aM", @progbits, 16
+.align 16
 ONE:             .octa     0x00000000000000000000000000000001
+
+.section	.rodata.cst16.ONEf, "aM", @progbits, 16
+.align 16
 ONEf:            .octa     0x01000000000000000000000000000000
 
+# order of these constants should not change.
+# more specifically, ALL_F should follow SHIFT_MASK, and zero should follow ALL_F
+.section	.rodata, "a", @progbits
+.align 16
+SHIFT_MASK:      .octa     0x0f0e0d0c0b0a09080706050403020100
+ALL_F:           .octa     0xffffffffffffffffffffffffffffffff
+                 .octa     0x00000000000000000000000000000000
+
 .text
 
 
diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c
index 7ff1b0c86a8e..93de8ea51548 100644
--- a/arch/x86/crypto/aesni-intel_glue.c
+++ b/arch/x86/crypto/aesni-intel_glue.c
@@ -740,9 +740,11 @@ static int helper_rfc4106_encrypt(struct aead_request *req)
 	*((__be32 *)(iv+12)) = counter;
 
 	if (sg_is_last(req->src) &&
-	    req->src->offset + req->src->length <= PAGE_SIZE &&
+	    (!PageHighMem(sg_page(req->src)) ||
+	    req->src->offset + req->src->length <= PAGE_SIZE) &&
 	    sg_is_last(req->dst) &&
-	    req->dst->offset + req->dst->length <= PAGE_SIZE) {
+	    (!PageHighMem(sg_page(req->dst)) ||
+	    req->dst->offset + req->dst->length <= PAGE_SIZE)) {
 		one_entry_in_sg = 1;
 		scatterwalk_start(&src_sg_walk, req->src);
 		assoc = scatterwalk_map(&src_sg_walk);
@@ -822,9 +824,11 @@ static int helper_rfc4106_decrypt(struct aead_request *req)
 	*((__be32 *)(iv+12)) = counter;
 
 	if (sg_is_last(req->src) &&
-	    req->src->offset + req->src->length <= PAGE_SIZE &&
+	    (!PageHighMem(sg_page(req->src)) ||
+	    req->src->offset + req->src->length <= PAGE_SIZE) &&
 	    sg_is_last(req->dst) &&
-	    req->dst->offset + req->dst->length <= PAGE_SIZE) {
+	    (!PageHighMem(sg_page(req->dst)) ||
+	    req->dst->offset + req->dst->length <= PAGE_SIZE)) {
 		one_entry_in_sg = 1;
 		scatterwalk_start(&src_sg_walk, req->src);
 		assoc = scatterwalk_map(&src_sg_walk);
diff --git a/arch/x86/crypto/camellia-aesni-avx-asm_64.S b/arch/x86/crypto/camellia-aesni-avx-asm_64.S
index aa9e8bd163f6..f7c495e2863c 100644
--- a/arch/x86/crypto/camellia-aesni-avx-asm_64.S
+++ b/arch/x86/crypto/camellia-aesni-avx-asm_64.S
@@ -571,7 +571,9 @@ ENDPROC(roundsm16_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 	vmovdqu y6, 14 * 16(rio); \
 	vmovdqu y7, 15 * 16(rio);
 
-.data
+
+/* NB: section is mergeable, all elements must be aligned 16-byte blocks */
+.section	.rodata.cst16, "aM", @progbits, 16
 .align 16
 
 #define SHUFB_BYTES(idx) \
@@ -711,6 +713,7 @@ ENDPROC(roundsm16_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 	.byte 0x08, 0x05, 0x02, 0x0f, 0x0c, 0x09, 0x06, 0x03
 
 /* 4-bit mask */
+.section	.rodata.cst4.L0f0f0f0f, "aM", @progbits, 4
 .align 4
 .L0f0f0f0f:
 	.long 0x0f0f0f0f
diff --git a/arch/x86/crypto/camellia-aesni-avx2-asm_64.S b/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
index 16186c18656d..eee5b3982cfd 100644
--- a/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
+++ b/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
@@ -610,20 +610,25 @@ ENDPROC(roundsm32_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 	vmovdqu y6, 14 * 32(rio); \
 	vmovdqu y7, 15 * 32(rio);
 
-.data
-.align 32
 
+.section	.rodata.cst32.shufb_16x16b, "aM", @progbits, 32
+.align 32
 #define SHUFB_BYTES(idx) \
 	0 + (idx), 4 + (idx), 8 + (idx), 12 + (idx)
-
 .Lshufb_16x16b:
 	.byte SHUFB_BYTES(0), SHUFB_BYTES(1), SHUFB_BYTES(2), SHUFB_BYTES(3)
 	.byte SHUFB_BYTES(0), SHUFB_BYTES(1), SHUFB_BYTES(2), SHUFB_BYTES(3)
 
+.section	.rodata.cst32.pack_bswap, "aM", @progbits, 32
+.align 32
 .Lpack_bswap:
 	.long 0x00010203, 0x04050607, 0x80808080, 0x80808080
 	.long 0x00010203, 0x04050607, 0x80808080, 0x80808080
 
+/* NB: section is mergeable, all elements must be aligned 16-byte blocks */
+.section	.rodata.cst16, "aM", @progbits, 16
+.align 16
+
 /* For CTR-mode IV byteswap */
 .Lbswap128_mask:
 	.byte 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0
@@ -750,6 +755,7 @@ ENDPROC(roundsm32_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 	.byte 0x00, 0x0d, 0x0a, 0x07, 0x04, 0x01, 0x0e, 0x0b
 	.byte 0x08, 0x05, 0x02, 0x0f, 0x0c, 0x09, 0x06, 0x03
 
+.section	.rodata.cst4.L0f0f0f0f, "aM", @progbits, 4
 .align 4
 /* 4-bit mask */
 .L0f0f0f0f:
diff --git a/arch/x86/crypto/cast5-avx-x86_64-asm_64.S b/arch/x86/crypto/cast5-avx-x86_64-asm_64.S
index 14fa1966bf01..b4a8806234ea 100644
--- a/arch/x86/crypto/cast5-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/cast5-avx-x86_64-asm_64.S
@@ -195,19 +195,29 @@
 	vpshufb rmask,	x0, x0;           \
 	vpshufb rmask,	x1, x1;
 
-.data
-
+.section	.rodata.cst16.bswap_mask, "aM", @progbits, 16
 .align 16
 .Lbswap_mask:
 	.byte 3, 2, 1, 0, 7, 6, 5, 4, 11, 10, 9, 8, 15, 14, 13, 12
+.section	.rodata.cst16.bswap128_mask, "aM", @progbits, 16
+.align 16
 .Lbswap128_mask:
 	.byte 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0
+.section	.rodata.cst16.bswap_iv_mask, "aM", @progbits, 16
+.align 16
 .Lbswap_iv_mask:
 	.byte 7, 6, 5, 4, 3, 2, 1, 0, 7, 6, 5, 4, 3, 2, 1, 0
+
+.section	.rodata.cst4.16_mask, "aM", @progbits, 4
+.align 4
 .L16_mask:
 	.byte 16, 16, 16, 16
+.section	.rodata.cst4.32_mask, "aM", @progbits, 4
+.align 4
 .L32_mask:
 	.byte 32, 0, 0, 0
+.section	.rodata.cst4.first_mask, "aM", @progbits, 4
+.align 4
 .Lfirst_mask:
 	.byte 0x1f, 0, 0, 0
 
diff --git a/arch/x86/crypto/cast6-avx-x86_64-asm_64.S b/arch/x86/crypto/cast6-avx-x86_64-asm_64.S
index c419389889cd..952d3156a933 100644
--- a/arch/x86/crypto/cast6-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/cast6-avx-x86_64-asm_64.S
@@ -225,8 +225,7 @@
 	vpshufb rmask,		x2, x2;       \
 	vpshufb rmask,		x3, x3;
 
-.data
-
+.section	.rodata.cst16, "aM", @progbits, 16
 .align 16
 .Lxts_gf128mul_and_shl1_mask:
 	.byte 0x87, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0
@@ -244,10 +243,19 @@
 	.byte 12, 13, 14, 15, 8, 9, 10, 11, 7, 6, 5, 4, 3, 2, 1, 0
 .Lrkr_dec_QBAR_QBAR_QBAR_QBAR:
 	.byte 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0
+
+.section	.rodata.cst4.L16_mask, "aM", @progbits, 4
+.align 4
 .L16_mask:
 	.byte 16, 16, 16, 16
+
+.section	.rodata.cst4.L32_mask, "aM", @progbits, 4
+.align 4
 .L32_mask:
 	.byte 32, 0, 0, 0
+
+.section	.rodata.cst4.first_mask, "aM", @progbits, 4
+.align 4
 .Lfirst_mask:
 	.byte 0x1f, 0, 0, 0
 
diff --git a/arch/x86/crypto/chacha20-avx2-x86_64.S b/arch/x86/crypto/chacha20-avx2-x86_64.S
index 16694e625f77..3a2dc3dc6cac 100644
--- a/arch/x86/crypto/chacha20-avx2-x86_64.S
+++ b/arch/x86/crypto/chacha20-avx2-x86_64.S
@@ -11,13 +11,18 @@
 
 #include <linux/linkage.h>
 
-.data
+.section	.rodata.cst32.ROT8, "aM", @progbits, 32
 .align 32
-
 ROT8:	.octa 0x0e0d0c0f0a09080b0605040702010003
 	.octa 0x0e0d0c0f0a09080b0605040702010003
+
+.section	.rodata.cst32.ROT16, "aM", @progbits, 32
+.align 32
 ROT16:	.octa 0x0d0c0f0e09080b0a0504070601000302
 	.octa 0x0d0c0f0e09080b0a0504070601000302
+
+.section	.rodata.cst32.CTRINC, "aM", @progbits, 32
+.align 32
 CTRINC:	.octa 0x00000003000000020000000100000000
 	.octa 0x00000007000000060000000500000004
 
diff --git a/arch/x86/crypto/chacha20-ssse3-x86_64.S b/arch/x86/crypto/chacha20-ssse3-x86_64.S
index 3a33124e9112..3f511a7d73b8 100644
--- a/arch/x86/crypto/chacha20-ssse3-x86_64.S
+++ b/arch/x86/crypto/chacha20-ssse3-x86_64.S
@@ -11,11 +11,14 @@
 
 #include <linux/linkage.h>
 
-.data
+.section	.rodata.cst16.ROT8, "aM", @progbits, 16
 .align 16
-
 ROT8:	.octa 0x0e0d0c0f0a09080b0605040702010003
+.section	.rodata.cst16.ROT16, "aM", @progbits, 16
+.align 16
 ROT16:	.octa 0x0d0c0f0e09080b0a0504070601000302
+.section	.rodata.cst16.CTRINC, "aM", @progbits, 16
+.align 16
 CTRINC:	.octa 0x00000003000000020000000100000000
 
 .text
diff --git a/arch/x86/crypto/chacha20_glue.c b/arch/x86/crypto/chacha20_glue.c
index f910d1d449f0..1e6af1b35f7b 100644
--- a/arch/x86/crypto/chacha20_glue.c
+++ b/arch/x86/crypto/chacha20_glue.c
@@ -11,7 +11,7 @@
 
 #include <crypto/algapi.h>
 #include <crypto/chacha20.h>
-#include <linux/crypto.h>
+#include <crypto/internal/skcipher.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <asm/fpu/api.h>
@@ -63,36 +63,37 @@ static void chacha20_dosimd(u32 *state, u8 *dst, const u8 *src,
 	}
 }
 
-static int chacha20_simd(struct blkcipher_desc *desc, struct scatterlist *dst,
-			 struct scatterlist *src, unsigned int nbytes)
+static int chacha20_simd(struct skcipher_request *req)
 {
-	u32 *state, state_buf[16 + (CHACHA20_STATE_ALIGN / sizeof(u32)) - 1];
-	struct blkcipher_walk walk;
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct chacha20_ctx *ctx = crypto_skcipher_ctx(tfm);
+	u32 *state, state_buf[16 + 2] __aligned(8);
+	struct skcipher_walk walk;
 	int err;
 
-	if (nbytes <= CHACHA20_BLOCK_SIZE || !may_use_simd())
-		return crypto_chacha20_crypt(desc, dst, src, nbytes);
+	BUILD_BUG_ON(CHACHA20_STATE_ALIGN != 16);
+	state = PTR_ALIGN(state_buf + 0, CHACHA20_STATE_ALIGN);
 
-	state = (u32 *)roundup((uintptr_t)state_buf, CHACHA20_STATE_ALIGN);
+	if (req->cryptlen <= CHACHA20_BLOCK_SIZE || !may_use_simd())
+		return crypto_chacha20_crypt(req);
 
-	blkcipher_walk_init(&walk, dst, src, nbytes);
-	err = blkcipher_walk_virt_block(desc, &walk, CHACHA20_BLOCK_SIZE);
+	err = skcipher_walk_virt(&walk, req, true);
 
-	crypto_chacha20_init(state, crypto_blkcipher_ctx(desc->tfm), walk.iv);
+	crypto_chacha20_init(state, ctx, walk.iv);
 
 	kernel_fpu_begin();
 
 	while (walk.nbytes >= CHACHA20_BLOCK_SIZE) {
 		chacha20_dosimd(state, walk.dst.virt.addr, walk.src.virt.addr,
 				rounddown(walk.nbytes, CHACHA20_BLOCK_SIZE));
-		err = blkcipher_walk_done(desc, &walk,
-					  walk.nbytes % CHACHA20_BLOCK_SIZE);
+		err = skcipher_walk_done(&walk,
+					 walk.nbytes % CHACHA20_BLOCK_SIZE);
 	}
 
 	if (walk.nbytes) {
 		chacha20_dosimd(state, walk.dst.virt.addr, walk.src.virt.addr,
 				walk.nbytes);
-		err = blkcipher_walk_done(desc, &walk, 0);
+		err = skcipher_walk_done(&walk, 0);
 	}
 
 	kernel_fpu_end();
@@ -100,27 +101,22 @@ static int chacha20_simd(struct blkcipher_desc *desc, struct scatterlist *dst,
 	return err;
 }
 
-static struct crypto_alg alg = {
-	.cra_name		= "chacha20",
-	.cra_driver_name	= "chacha20-simd",
-	.cra_priority		= 300,
-	.cra_flags		= CRYPTO_ALG_TYPE_BLKCIPHER,
-	.cra_blocksize		= 1,
-	.cra_type		= &crypto_blkcipher_type,
-	.cra_ctxsize		= sizeof(struct chacha20_ctx),
-	.cra_alignmask		= sizeof(u32) - 1,
-	.cra_module		= THIS_MODULE,
-	.cra_u			= {
-		.blkcipher = {
-			.min_keysize	= CHACHA20_KEY_SIZE,
-			.max_keysize	= CHACHA20_KEY_SIZE,
-			.ivsize		= CHACHA20_IV_SIZE,
-			.geniv		= "seqiv",
-			.setkey		= crypto_chacha20_setkey,
-			.encrypt	= chacha20_simd,
-			.decrypt	= chacha20_simd,
-		},
-	},
+static struct skcipher_alg alg = {
+	.base.cra_name		= "chacha20",
+	.base.cra_driver_name	= "chacha20-simd",
+	.base.cra_priority	= 300,
+	.base.cra_blocksize	= 1,
+	.base.cra_ctxsize	= sizeof(struct chacha20_ctx),
+	.base.cra_alignmask	= sizeof(u32) - 1,
+	.base.cra_module	= THIS_MODULE,
+
+	.min_keysize		= CHACHA20_KEY_SIZE,
+	.max_keysize		= CHACHA20_KEY_SIZE,
+	.ivsize			= CHACHA20_IV_SIZE,
+	.chunksize		= CHACHA20_BLOCK_SIZE,
+	.setkey			= crypto_chacha20_setkey,
+	.encrypt		= chacha20_simd,
+	.decrypt		= chacha20_simd,
 };
 
 static int __init chacha20_simd_mod_init(void)
@@ -133,12 +129,12 @@ static int __init chacha20_simd_mod_init(void)
 			    boot_cpu_has(X86_FEATURE_AVX2) &&
 			    cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM, NULL);
 #endif
-	return crypto_register_alg(&alg);
+	return crypto_register_skcipher(&alg);
 }
 
 static void __exit chacha20_simd_mod_fini(void)
 {
-	crypto_unregister_alg(&alg);
+	crypto_unregister_skcipher(&alg);
 }
 
 module_init(chacha20_simd_mod_init);
diff --git a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S
index dc05f010ca9b..7a7de27c6f41 100644
--- a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S
+++ b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S
@@ -312,7 +312,7 @@ do_return:
         ret
 ENDPROC(crc_pcl)
 
-.section	.rodata, "a", %progbits
+.section	.rodata, "a", @progbits
         ################################################################
         ## jump table        Table is 129 entries x 2 bytes each
         ################################################################
diff --git a/arch/x86/crypto/crct10dif-pcl-asm_64.S b/arch/x86/crypto/crct10dif-pcl-asm_64.S
index 35e97569d05f..de04d3e98d8d 100644
--- a/arch/x86/crypto/crct10dif-pcl-asm_64.S
+++ b/arch/x86/crypto/crct10dif-pcl-asm_64.S
@@ -554,12 +554,11 @@ _only_less_than_2:
 
 ENDPROC(crc_t10dif_pcl)
 
-.data
-
+.section	.rodata, "a", @progbits
+.align 16
 # precomputed constants
 # these constants are precomputed from the poly:
 # 0x8bb70000 (0x8bb7 scaled to 32 bits)
-.align 16
 # Q = 0x18BB70000
 # rk1 = 2^(32*3) mod Q << 32
 # rk2 = 2^(32*5) mod Q << 32
@@ -613,14 +612,23 @@ rk20:
 
 
 
+.section	.rodata.cst16.mask1, "aM", @progbits, 16
+.align 16
 mask1:
 .octa 0x80808080808080808080808080808080
+
+.section	.rodata.cst16.mask2, "aM", @progbits, 16
+.align 16
 mask2:
 .octa 0x00000000FFFFFFFFFFFFFFFFFFFFFFFF
 
+.section	.rodata.cst16.SHUF_MASK, "aM", @progbits, 16
+.align 16
 SHUF_MASK:
 .octa 0x000102030405060708090A0B0C0D0E0F
 
+.section	.rodata.cst32.pshufb_shf_table, "aM", @progbits, 32
+.align 32
 pshufb_shf_table:
 # use these values for shift constants for the pshufb instruction
 # different alignments result in values as shown:
diff --git a/arch/x86/crypto/des3_ede-asm_64.S b/arch/x86/crypto/des3_ede-asm_64.S
index 038f6ae87c5e..f3e91647ca27 100644
--- a/arch/x86/crypto/des3_ede-asm_64.S
+++ b/arch/x86/crypto/des3_ede-asm_64.S
@@ -537,7 +537,7 @@ ENTRY(des3_ede_x86_64_crypt_blk_3way)
 	ret;
 ENDPROC(des3_ede_x86_64_crypt_blk_3way)
 
-.data
+.section	.rodata, "a", @progbits
 .align 16
 .L_s1:
 	.quad 0x0010100001010400, 0x0000000000000000
diff --git a/arch/x86/crypto/ghash-clmulni-intel_asm.S b/arch/x86/crypto/ghash-clmulni-intel_asm.S
index eed55c8cca4f..f94375a8dcd1 100644
--- a/arch/x86/crypto/ghash-clmulni-intel_asm.S
+++ b/arch/x86/crypto/ghash-clmulni-intel_asm.S
@@ -20,8 +20,7 @@
 #include <asm/inst.h>
 #include <asm/frame.h>
 
-.data
-
+.section	.rodata.cst16.bswap_mask, "aM", @progbits, 16
 .align 16
 .Lbswap_mask:
 	.octa 0x000102030405060708090a0b0c0d0e0f
diff --git a/arch/x86/crypto/poly1305-avx2-x86_64.S b/arch/x86/crypto/poly1305-avx2-x86_64.S
index eff2f414e22b..3b6e70d085da 100644
--- a/arch/x86/crypto/poly1305-avx2-x86_64.S
+++ b/arch/x86/crypto/poly1305-avx2-x86_64.S
@@ -11,11 +11,13 @@
 
 #include <linux/linkage.h>
 
-.data
+.section	.rodata.cst32.ANMASK, "aM", @progbits, 32
 .align 32
-
 ANMASK:	.octa 0x0000000003ffffff0000000003ffffff
 	.octa 0x0000000003ffffff0000000003ffffff
+
+.section	.rodata.cst32.ORMASK, "aM", @progbits, 32
+.align 32
 ORMASK:	.octa 0x00000000010000000000000001000000
 	.octa 0x00000000010000000000000001000000
 
diff --git a/arch/x86/crypto/poly1305-sse2-x86_64.S b/arch/x86/crypto/poly1305-sse2-x86_64.S
index 338c748054ed..c88c670cb5fc 100644
--- a/arch/x86/crypto/poly1305-sse2-x86_64.S
+++ b/arch/x86/crypto/poly1305-sse2-x86_64.S
@@ -11,10 +11,12 @@
 
 #include <linux/linkage.h>
 
-.data
+.section	.rodata.cst16.ANMASK, "aM", @progbits, 16
 .align 16
-
 ANMASK:	.octa 0x0000000003ffffff0000000003ffffff
+
+.section	.rodata.cst16.ORMASK, "aM", @progbits, 16
+.align 16
 ORMASK:	.octa 0x00000000010000000000000001000000
 
 .text
diff --git a/arch/x86/crypto/serpent-avx-x86_64-asm_64.S b/arch/x86/crypto/serpent-avx-x86_64-asm_64.S
index 8be571808342..2925077f8c6a 100644
--- a/arch/x86/crypto/serpent-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/serpent-avx-x86_64-asm_64.S
@@ -29,11 +29,12 @@
 
 .file "serpent-avx-x86_64-asm_64.S"
 
-.data
+.section	.rodata.cst16.bswap128_mask, "aM", @progbits, 16
 .align 16
-
 .Lbswap128_mask:
 	.byte 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0
+.section	.rodata.cst16.xts_gf128mul_and_shl1_mask, "aM", @progbits, 16
+.align 16
 .Lxts_gf128mul_and_shl1_mask:
 	.byte 0x87, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0
 
diff --git a/arch/x86/crypto/serpent-avx2-asm_64.S b/arch/x86/crypto/serpent-avx2-asm_64.S
index 97c48add33ed..d67888f2a52a 100644
--- a/arch/x86/crypto/serpent-avx2-asm_64.S
+++ b/arch/x86/crypto/serpent-avx2-asm_64.S
@@ -20,13 +20,18 @@
 
 .file "serpent-avx2-asm_64.S"
 
-.data
+.section	.rodata.cst16.bswap128_mask, "aM", @progbits, 16
 .align 16
-
 .Lbswap128_mask:
 	.byte 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0
+
+.section	.rodata.cst16.xts_gf128mul_and_shl1_mask_0, "aM", @progbits, 16
+.align 16
 .Lxts_gf128mul_and_shl1_mask_0:
 	.byte 0x87, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0
+
+.section	.rodata.cst16.xts_gf128mul_and_shl1_mask_1, "aM", @progbits, 16
+.align 16
 .Lxts_gf128mul_and_shl1_mask_1:
 	.byte 0x0e, 1, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0
 
diff --git a/arch/x86/crypto/sha1-mb/sha1_mb_mgr_flush_avx2.S b/arch/x86/crypto/sha1-mb/sha1_mb_mgr_flush_avx2.S
index 96df6a39d7e2..93b945597ecf 100644
--- a/arch/x86/crypto/sha1-mb/sha1_mb_mgr_flush_avx2.S
+++ b/arch/x86/crypto/sha1-mb/sha1_mb_mgr_flush_avx2.S
@@ -281,11 +281,13 @@ ENTRY(sha1_mb_mgr_get_comp_job_avx2)
 	ret
 ENDPROC(sha1_mb_mgr_get_comp_job_avx2)
 
-.data
-
+.section	.rodata.cst16.clear_low_nibble, "aM", @progbits, 16
 .align 16
 clear_low_nibble:
 .octa	0x000000000000000000000000FFFFFFF0
+
+.section	.rodata.cst8, "aM", @progbits, 8
+.align 8
 one:
 .quad  1
 two:
diff --git a/arch/x86/crypto/sha1-mb/sha1_mb_mgr_submit_avx2.S b/arch/x86/crypto/sha1-mb/sha1_mb_mgr_submit_avx2.S
index 63a0d9c8e31f..7a93b1c0d69a 100644
--- a/arch/x86/crypto/sha1-mb/sha1_mb_mgr_submit_avx2.S
+++ b/arch/x86/crypto/sha1-mb/sha1_mb_mgr_submit_avx2.S
@@ -203,8 +203,7 @@ return_null:
 
 ENDPROC(sha1_mb_mgr_submit_avx2)
 
-.data
-
+.section	.rodata.cst16.clear_low_nibble, "aM", @progbits, 16
 .align 16
 clear_low_nibble:
 	.octa	0x000000000000000000000000FFFFFFF0
diff --git a/arch/x86/crypto/sha1-mb/sha1_x8_avx2.S b/arch/x86/crypto/sha1-mb/sha1_x8_avx2.S
index c9dae1cd2919..20f77aa633de 100644
--- a/arch/x86/crypto/sha1-mb/sha1_x8_avx2.S
+++ b/arch/x86/crypto/sha1-mb/sha1_x8_avx2.S
@@ -461,21 +461,32 @@ lloop:
 ENDPROC(sha1_x8_avx2)
 
 
-.data
-
+.section	.rodata.cst32.K00_19, "aM", @progbits, 32
 .align 32
 K00_19:
 .octa 0x5A8279995A8279995A8279995A827999
 .octa 0x5A8279995A8279995A8279995A827999
+
+.section	.rodata.cst32.K20_39, "aM", @progbits, 32
+.align 32
 K20_39:
 .octa 0x6ED9EBA16ED9EBA16ED9EBA16ED9EBA1
 .octa 0x6ED9EBA16ED9EBA16ED9EBA16ED9EBA1
+
+.section	.rodata.cst32.K40_59, "aM", @progbits, 32
+.align 32
 K40_59:
 .octa 0x8F1BBCDC8F1BBCDC8F1BBCDC8F1BBCDC
 .octa 0x8F1BBCDC8F1BBCDC8F1BBCDC8F1BBCDC
+
+.section	.rodata.cst32.K60_79, "aM", @progbits, 32
+.align 32
 K60_79:
 .octa 0xCA62C1D6CA62C1D6CA62C1D6CA62C1D6
 .octa 0xCA62C1D6CA62C1D6CA62C1D6CA62C1D6
+
+.section	.rodata.cst32.PSHUFFLE_BYTE_FLIP_MASK, "aM", @progbits, 32
+.align 32
 PSHUFFLE_BYTE_FLIP_MASK:
 .octa 0x0c0d0e0f08090a0b0405060700010203
 .octa 0x0c0d0e0f08090a0b0405060700010203
diff --git a/arch/x86/crypto/sha1_ni_asm.S b/arch/x86/crypto/sha1_ni_asm.S
index 874a651b9e7d..ebbdba72ae07 100644
--- a/arch/x86/crypto/sha1_ni_asm.S
+++ b/arch/x86/crypto/sha1_ni_asm.S
@@ -293,10 +293,12 @@ ENTRY(sha1_ni_transform)
 	ret
 ENDPROC(sha1_ni_transform)
 
-.data
-
-.align 64
+.section	.rodata.cst16.PSHUFFLE_BYTE_FLIP_MASK, "aM", @progbits, 16
+.align 16
 PSHUFFLE_BYTE_FLIP_MASK:
 	.octa 0x000102030405060708090a0b0c0d0e0f
+
+.section	.rodata.cst16.UPPER_WORD_MASK, "aM", @progbits, 16
+.align 16
 UPPER_WORD_MASK:
 	.octa 0xFFFFFFFF000000000000000000000000
diff --git a/arch/x86/crypto/sha256-avx-asm.S b/arch/x86/crypto/sha256-avx-asm.S
index 92b3b5d75ba9..e08888a1a5f2 100644
--- a/arch/x86/crypto/sha256-avx-asm.S
+++ b/arch/x86/crypto/sha256-avx-asm.S
@@ -463,7 +463,7 @@ done_hash:
 	ret
 ENDPROC(sha256_transform_avx)
 
-.data
+.section	.rodata.cst256.K256, "aM", @progbits, 256
 .align 64
 K256:
 	.long 0x428a2f98,0x71374491,0xb5c0fbcf,0xe9b5dba5
@@ -483,14 +483,21 @@ K256:
 	.long 0x748f82ee,0x78a5636f,0x84c87814,0x8cc70208
 	.long 0x90befffa,0xa4506ceb,0xbef9a3f7,0xc67178f2
 
+.section	.rodata.cst16.PSHUFFLE_BYTE_FLIP_MASK, "aM", @progbits, 16
+.align 16
 PSHUFFLE_BYTE_FLIP_MASK:
 	.octa 0x0c0d0e0f08090a0b0405060700010203
 
+.section	.rodata.cst16._SHUF_00BA, "aM", @progbits, 16
+.align 16
 # shuffle xBxA -> 00BA
 _SHUF_00BA:
 	.octa 0xFFFFFFFFFFFFFFFF0b0a090803020100
 
+.section	.rodata.cst16._SHUF_DC00, "aM", @progbits, 16
+.align 16
 # shuffle xDxC -> DC00
 _SHUF_DC00:
 	.octa 0x0b0a090803020100FFFFFFFFFFFFFFFF
+
 #endif
diff --git a/arch/x86/crypto/sha256-avx2-asm.S b/arch/x86/crypto/sha256-avx2-asm.S
index 570ec5ec62d7..89c8f09787d2 100644
--- a/arch/x86/crypto/sha256-avx2-asm.S
+++ b/arch/x86/crypto/sha256-avx2-asm.S
@@ -723,7 +723,7 @@ done_hash:
 	ret
 ENDPROC(sha256_transform_rorx)
 
-.data
+.section	.rodata.cst512.K256, "aM", @progbits, 512
 .align 64
 K256:
 	.long	0x428a2f98,0x71374491,0xb5c0fbcf,0xe9b5dba5
@@ -759,14 +759,21 @@ K256:
 	.long	0x90befffa,0xa4506ceb,0xbef9a3f7,0xc67178f2
 	.long	0x90befffa,0xa4506ceb,0xbef9a3f7,0xc67178f2
 
+.section	.rodata.cst32.PSHUFFLE_BYTE_FLIP_MASK, "aM", @progbits, 32
+.align 32
 PSHUFFLE_BYTE_FLIP_MASK:
 	.octa 0x0c0d0e0f08090a0b0405060700010203,0x0c0d0e0f08090a0b0405060700010203
 
 # shuffle xBxA -> 00BA
+.section	.rodata.cst32._SHUF_00BA, "aM", @progbits, 32
+.align 32
 _SHUF_00BA:
 	.octa 0xFFFFFFFFFFFFFFFF0b0a090803020100,0xFFFFFFFFFFFFFFFF0b0a090803020100
 
 # shuffle xDxC -> DC00
+.section	.rodata.cst32._SHUF_DC00, "aM", @progbits, 32
+.align 32
 _SHUF_DC00:
 	.octa 0x0b0a090803020100FFFFFFFFFFFFFFFF,0x0b0a090803020100FFFFFFFFFFFFFFFF
+
 #endif
diff --git a/arch/x86/crypto/sha256-mb/sha256_mb_mgr_flush_avx2.S b/arch/x86/crypto/sha256-mb/sha256_mb_mgr_flush_avx2.S
index a78a0694ddef..8fe6338bcc84 100644
--- a/arch/x86/crypto/sha256-mb/sha256_mb_mgr_flush_avx2.S
+++ b/arch/x86/crypto/sha256-mb/sha256_mb_mgr_flush_avx2.S
@@ -284,11 +284,13 @@ ENTRY(sha256_mb_mgr_get_comp_job_avx2)
 	ret
 ENDPROC(sha256_mb_mgr_get_comp_job_avx2)
 
-.data
-
+.section	.rodata.cst16.clear_low_nibble, "aM", @progbits, 16
 .align 16
 clear_low_nibble:
 .octa	0x000000000000000000000000FFFFFFF0
+
+.section	.rodata.cst8, "aM", @progbits, 8
+.align 8
 one:
 .quad	1
 two:
diff --git a/arch/x86/crypto/sha256-mb/sha256_mb_mgr_submit_avx2.S b/arch/x86/crypto/sha256-mb/sha256_mb_mgr_submit_avx2.S
index 7ea670e25acc..b36ae7454084 100644
--- a/arch/x86/crypto/sha256-mb/sha256_mb_mgr_submit_avx2.S
+++ b/arch/x86/crypto/sha256-mb/sha256_mb_mgr_submit_avx2.S
@@ -208,8 +208,7 @@ return_null:
 
 ENDPROC(sha256_mb_mgr_submit_avx2)
 
-.data
-
+.section	.rodata.cst16.clear_low_nibble, "aM", @progbits, 16
 .align 16
 clear_low_nibble:
 	.octa	0x000000000000000000000000FFFFFFF0
diff --git a/arch/x86/crypto/sha256-mb/sha256_x8_avx2.S b/arch/x86/crypto/sha256-mb/sha256_x8_avx2.S
index aa21aea4c722..1687c80c5995 100644
--- a/arch/x86/crypto/sha256-mb/sha256_x8_avx2.S
+++ b/arch/x86/crypto/sha256-mb/sha256_x8_avx2.S
@@ -437,7 +437,8 @@ Lrounds_16_xx:
 
 	ret
 ENDPROC(sha256_x8_avx2)
-.data
+
+.section	.rodata.K256_8, "a", @progbits
 .align 64
 K256_8:
 	.octa	0x428a2f98428a2f98428a2f98428a2f98
@@ -568,10 +569,14 @@ K256_8:
 	.octa	0xbef9a3f7bef9a3f7bef9a3f7bef9a3f7
 	.octa	0xc67178f2c67178f2c67178f2c67178f2
 	.octa	0xc67178f2c67178f2c67178f2c67178f2
+
+.section	.rodata.cst32.PSHUFFLE_BYTE_FLIP_MASK, "aM", @progbits, 32
+.align 32
 PSHUFFLE_BYTE_FLIP_MASK:
 .octa 0x0c0d0e0f08090a0b0405060700010203
 .octa 0x0c0d0e0f08090a0b0405060700010203
 
+.section	.rodata.cst256.K256, "aM", @progbits, 256
 .align 64
 .global K256
 K256:
diff --git a/arch/x86/crypto/sha256-ssse3-asm.S b/arch/x86/crypto/sha256-ssse3-asm.S
index 2cedc44e8121..39b83c93e7fd 100644
--- a/arch/x86/crypto/sha256-ssse3-asm.S
+++ b/arch/x86/crypto/sha256-ssse3-asm.S
@@ -474,7 +474,7 @@ done_hash:
 	ret
 ENDPROC(sha256_transform_ssse3)
 
-.data
+.section	.rodata.cst256.K256, "aM", @progbits, 256
 .align 64
 K256:
         .long 0x428a2f98,0x71374491,0xb5c0fbcf,0xe9b5dba5
@@ -494,13 +494,19 @@ K256:
         .long 0x748f82ee,0x78a5636f,0x84c87814,0x8cc70208
         .long 0x90befffa,0xa4506ceb,0xbef9a3f7,0xc67178f2
 
+.section	.rodata.cst16.PSHUFFLE_BYTE_FLIP_MASK, "aM", @progbits, 16
+.align 16
 PSHUFFLE_BYTE_FLIP_MASK:
 	.octa 0x0c0d0e0f08090a0b0405060700010203
 
+.section	.rodata.cst16._SHUF_00BA, "aM", @progbits, 16
+.align 16
 # shuffle xBxA -> 00BA
 _SHUF_00BA:
 	.octa 0xFFFFFFFFFFFFFFFF0b0a090803020100
 
+.section	.rodata.cst16._SHUF_DC00, "aM", @progbits, 16
+.align 16
 # shuffle xDxC -> DC00
 _SHUF_DC00:
 	.octa 0x0b0a090803020100FFFFFFFFFFFFFFFF
diff --git a/arch/x86/crypto/sha256_ni_asm.S b/arch/x86/crypto/sha256_ni_asm.S
index 748cdf21a938..fb58f58ecfbc 100644
--- a/arch/x86/crypto/sha256_ni_asm.S
+++ b/arch/x86/crypto/sha256_ni_asm.S
@@ -329,7 +329,7 @@ ENTRY(sha256_ni_transform)
 	ret
 ENDPROC(sha256_ni_transform)
 
-.data
+.section	.rodata.cst256.K256, "aM", @progbits, 256
 .align 64
 K256:
 	.long	0x428a2f98,0x71374491,0xb5c0fbcf,0xe9b5dba5
@@ -349,5 +349,7 @@ K256:
 	.long	0x748f82ee,0x78a5636f,0x84c87814,0x8cc70208
 	.long	0x90befffa,0xa4506ceb,0xbef9a3f7,0xc67178f2
 
+.section	.rodata.cst16.PSHUFFLE_BYTE_FLIP_MASK, "aM", @progbits, 16
+.align 16
 PSHUFFLE_BYTE_FLIP_MASK:
 	.octa 0x0c0d0e0f08090a0b0405060700010203
diff --git a/arch/x86/crypto/sha512-avx-asm.S b/arch/x86/crypto/sha512-avx-asm.S
index 565274d6a641..39235fefe6f7 100644
--- a/arch/x86/crypto/sha512-avx-asm.S
+++ b/arch/x86/crypto/sha512-avx-asm.S
@@ -370,14 +370,17 @@ ENDPROC(sha512_transform_avx)
 ########################################################################
 ### Binary Data
 
-.data
-
+.section	.rodata.cst16.XMM_QWORD_BSWAP, "aM", @progbits, 16
 .align 16
-
 # Mask for byte-swapping a couple of qwords in an XMM register using (v)pshufb.
 XMM_QWORD_BSWAP:
 	.octa 0x08090a0b0c0d0e0f0001020304050607
 
+# Mergeable 640-byte rodata section. This allows linker to merge the table
+# with other, exactly the same 640-byte fragment of another rodata section
+# (if such section exists).
+.section	.rodata.cst640.K512, "aM", @progbits, 640
+.align 64
 # K[t] used in SHA512 hashing
 K512:
 	.quad 0x428a2f98d728ae22,0x7137449123ef65cd
diff --git a/arch/x86/crypto/sha512-avx2-asm.S b/arch/x86/crypto/sha512-avx2-asm.S
index 1f20b35d8573..7f5f6c6ec72e 100644
--- a/arch/x86/crypto/sha512-avx2-asm.S
+++ b/arch/x86/crypto/sha512-avx2-asm.S
@@ -684,8 +684,11 @@ ENDPROC(sha512_transform_rorx)
 ########################################################################
 ### Binary Data
 
-.data
 
+# Mergeable 640-byte rodata section. This allows linker to merge the table
+# with other, exactly the same 640-byte fragment of another rodata section
+# (if such section exists).
+.section	.rodata.cst640.K512, "aM", @progbits, 640
 .align 64
 # K[t] used in SHA512 hashing
 K512:
@@ -730,14 +733,17 @@ K512:
 	.quad	0x4cc5d4becb3e42b6,0x597f299cfc657e2a
 	.quad	0x5fcb6fab3ad6faec,0x6c44198c4a475817
 
+.section	.rodata.cst32.PSHUFFLE_BYTE_FLIP_MASK, "aM", @progbits, 32
 .align 32
-
 # Mask for byte-swapping a couple of qwords in an XMM register using (v)pshufb.
 PSHUFFLE_BYTE_FLIP_MASK:
 	.octa 0x08090a0b0c0d0e0f0001020304050607
 	.octa 0x18191a1b1c1d1e1f1011121314151617
 
+.section	.rodata.cst32.MASK_YMM_LO, "aM", @progbits, 32
+.align 32
 MASK_YMM_LO:
 	.octa 0x00000000000000000000000000000000
 	.octa 0xFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
+
 #endif
diff --git a/arch/x86/crypto/sha512-mb/sha512_mb.c b/arch/x86/crypto/sha512-mb/sha512_mb.c
index 9c1bb6d58141..2dd3674b5a1e 100644
--- a/arch/x86/crypto/sha512-mb/sha512_mb.c
+++ b/arch/x86/crypto/sha512-mb/sha512_mb.c
@@ -221,7 +221,7 @@ static struct sha512_hash_ctx *sha512_ctx_mgr_resubmit
 }
 
 static struct sha512_hash_ctx
-		*sha512_ctx_mgr_get_comp_ctx(struct sha512_ctx_mgr *mgr)
+		*sha512_ctx_mgr_get_comp_ctx(struct mcryptd_alg_cstate *cstate)
 {
 	/*
 	 * If get_comp_job returns NULL, there are no jobs complete.
@@ -233,11 +233,17 @@ static struct sha512_hash_ctx
 	 * Otherwise, all jobs currently being managed by the hash_ctx_mgr
 	 * still need processing.
 	 */
+	struct sha512_ctx_mgr *mgr;
 	struct sha512_hash_ctx *ctx;
+	unsigned long flags;
 
+	mgr = cstate->mgr;
+	spin_lock_irqsave(&cstate->work_lock, flags);
 	ctx = (struct sha512_hash_ctx *)
 				sha512_job_mgr_get_comp_job(&mgr->mgr);
-	return sha512_ctx_mgr_resubmit(mgr, ctx);
+	ctx = sha512_ctx_mgr_resubmit(mgr, ctx);
+	spin_unlock_irqrestore(&cstate->work_lock, flags);
+	return ctx;
 }
 
 static void sha512_ctx_mgr_init(struct sha512_ctx_mgr *mgr)
@@ -246,12 +252,17 @@ static void sha512_ctx_mgr_init(struct sha512_ctx_mgr *mgr)
 }
 
 static struct sha512_hash_ctx
-			*sha512_ctx_mgr_submit(struct sha512_ctx_mgr *mgr,
+			*sha512_ctx_mgr_submit(struct mcryptd_alg_cstate *cstate,
 					  struct sha512_hash_ctx *ctx,
 					  const void *buffer,
 					  uint32_t len,
 					  int flags)
 {
+	struct sha512_ctx_mgr *mgr;
+	unsigned long irqflags;
+
+	mgr = cstate->mgr;
+	spin_lock_irqsave(&cstate->work_lock, irqflags);
 	if (flags & (~HASH_ENTIRE)) {
 		/*
 		 * User should not pass anything other than FIRST, UPDATE, or
@@ -351,20 +362,26 @@ static struct sha512_hash_ctx
 		}
 	}
 
-	return sha512_ctx_mgr_resubmit(mgr, ctx);
+	ctx = sha512_ctx_mgr_resubmit(mgr, ctx);
+	spin_unlock_irqrestore(&cstate->work_lock, irqflags);
+	return ctx;
 }
 
-static struct sha512_hash_ctx *sha512_ctx_mgr_flush(struct sha512_ctx_mgr *mgr)
+static struct sha512_hash_ctx *sha512_ctx_mgr_flush(struct mcryptd_alg_cstate *cstate)
 {
+	struct sha512_ctx_mgr *mgr;
 	struct sha512_hash_ctx *ctx;
+	unsigned long flags;
 
+	mgr = cstate->mgr;
+	spin_lock_irqsave(&cstate->work_lock, flags);
 	while (1) {
 		ctx = (struct sha512_hash_ctx *)
 					sha512_job_mgr_flush(&mgr->mgr);
 
 		/* If flush returned 0, there are no more jobs in flight. */
 		if (!ctx)
-			return NULL;
+			break;
 
 		/*
 		 * If flush returned a job, resubmit the job to finish
@@ -378,8 +395,10 @@ static struct sha512_hash_ctx *sha512_ctx_mgr_flush(struct sha512_ctx_mgr *mgr)
 		 * the sha512_ctx_mgr still need processing. Loop.
 		 */
 		if (ctx)
-			return ctx;
+			break;
 	}
+	spin_unlock_irqrestore(&cstate->work_lock, flags);
+	return ctx;
 }
 
 static int sha512_mb_init(struct ahash_request *areq)
@@ -439,11 +458,11 @@ static int sha_finish_walk(struct mcryptd_hash_request_ctx **ret_rctx,
 		sha_ctx = (struct sha512_hash_ctx *)
 						ahash_request_ctx(&rctx->areq);
 		kernel_fpu_begin();
-		sha_ctx = sha512_ctx_mgr_submit(cstate->mgr, sha_ctx,
+		sha_ctx = sha512_ctx_mgr_submit(cstate, sha_ctx,
 						rctx->walk.data, nbytes, flag);
 		if (!sha_ctx) {
 			if (flush)
-				sha_ctx = sha512_ctx_mgr_flush(cstate->mgr);
+				sha_ctx = sha512_ctx_mgr_flush(cstate);
 		}
 		kernel_fpu_end();
 		if (sha_ctx)
@@ -471,11 +490,12 @@ static int sha_complete_job(struct mcryptd_hash_request_ctx *rctx,
 	struct sha512_hash_ctx *sha_ctx;
 	struct mcryptd_hash_request_ctx *req_ctx;
 	int ret;
+	unsigned long flags;
 
 	/* remove from work list */
-	spin_lock(&cstate->work_lock);
+	spin_lock_irqsave(&cstate->work_lock, flags);
 	list_del(&rctx->waiter);
-	spin_unlock(&cstate->work_lock);
+	spin_unlock_irqrestore(&cstate->work_lock, flags);
 
 	if (irqs_disabled())
 		rctx->complete(&req->base, err);
@@ -486,14 +506,14 @@ static int sha_complete_job(struct mcryptd_hash_request_ctx *rctx,
 	}
 
 	/* check to see if there are other jobs that are done */
-	sha_ctx = sha512_ctx_mgr_get_comp_ctx(cstate->mgr);
+	sha_ctx = sha512_ctx_mgr_get_comp_ctx(cstate);
 	while (sha_ctx) {
 		req_ctx = cast_hash_to_mcryptd_ctx(sha_ctx);
 		ret = sha_finish_walk(&req_ctx, cstate, false);
 		if (req_ctx) {
-			spin_lock(&cstate->work_lock);
+			spin_lock_irqsave(&cstate->work_lock, flags);
 			list_del(&req_ctx->waiter);
-			spin_unlock(&cstate->work_lock);
+			spin_unlock_irqrestore(&cstate->work_lock, flags);
 
 			req = cast_mcryptd_ctx_to_req(req_ctx);
 			if (irqs_disabled())
@@ -504,7 +524,7 @@ static int sha_complete_job(struct mcryptd_hash_request_ctx *rctx,
 				local_bh_enable();
 			}
 		}
-		sha_ctx = sha512_ctx_mgr_get_comp_ctx(cstate->mgr);
+		sha_ctx = sha512_ctx_mgr_get_comp_ctx(cstate);
 	}
 
 	return 0;
@@ -515,6 +535,7 @@ static void sha512_mb_add_list(struct mcryptd_hash_request_ctx *rctx,
 {
 	unsigned long next_flush;
 	unsigned long delay = usecs_to_jiffies(FLUSH_INTERVAL);
+	unsigned long flags;
 
 	/* initialize tag */
 	rctx->tag.arrival = jiffies;    /* tag the arrival time */
@@ -522,9 +543,9 @@ static void sha512_mb_add_list(struct mcryptd_hash_request_ctx *rctx,
 	next_flush = rctx->tag.arrival + delay;
 	rctx->tag.expire = next_flush;
 
-	spin_lock(&cstate->work_lock);
+	spin_lock_irqsave(&cstate->work_lock, flags);
 	list_add_tail(&rctx->waiter, &cstate->work_list);
-	spin_unlock(&cstate->work_lock);
+	spin_unlock_irqrestore(&cstate->work_lock, flags);
 
 	mcryptd_arm_flusher(cstate, delay);
 }
@@ -565,7 +586,7 @@ static int sha512_mb_update(struct ahash_request *areq)
 	sha_ctx = (struct sha512_hash_ctx *) ahash_request_ctx(areq);
 	sha512_mb_add_list(rctx, cstate);
 	kernel_fpu_begin();
-	sha_ctx = sha512_ctx_mgr_submit(cstate->mgr, sha_ctx, rctx->walk.data,
+	sha_ctx = sha512_ctx_mgr_submit(cstate, sha_ctx, rctx->walk.data,
 							nbytes, HASH_UPDATE);
 	kernel_fpu_end();
 
@@ -628,7 +649,7 @@ static int sha512_mb_finup(struct ahash_request *areq)
 	sha512_mb_add_list(rctx, cstate);
 
 	kernel_fpu_begin();
-	sha_ctx = sha512_ctx_mgr_submit(cstate->mgr, sha_ctx, rctx->walk.data,
+	sha_ctx = sha512_ctx_mgr_submit(cstate, sha_ctx, rctx->walk.data,
 								nbytes, flag);
 	kernel_fpu_end();
 
@@ -677,8 +698,7 @@ static int sha512_mb_final(struct ahash_request *areq)
 	/* flag HASH_FINAL and 0 data size */
 	sha512_mb_add_list(rctx, cstate);
 	kernel_fpu_begin();
-	sha_ctx = sha512_ctx_mgr_submit(cstate->mgr, sha_ctx, &data, 0,
-								HASH_LAST);
+	sha_ctx = sha512_ctx_mgr_submit(cstate, sha_ctx, &data, 0, HASH_LAST);
 	kernel_fpu_end();
 
 	/* check if anything is returned */
@@ -940,7 +960,7 @@ static unsigned long sha512_mb_flusher(struct mcryptd_alg_cstate *cstate)
 			break;
 		kernel_fpu_begin();
 		sha_ctx = (struct sha512_hash_ctx *)
-					sha512_ctx_mgr_flush(cstate->mgr);
+					sha512_ctx_mgr_flush(cstate);
 		kernel_fpu_end();
 		if (!sha_ctx) {
 			pr_err("sha512_mb error: nothing got flushed for"
diff --git a/arch/x86/crypto/sha512-mb/sha512_mb_mgr_flush_avx2.S b/arch/x86/crypto/sha512-mb/sha512_mb_mgr_flush_avx2.S
index 3ddba19a0db6..7c629caebc05 100644
--- a/arch/x86/crypto/sha512-mb/sha512_mb_mgr_flush_avx2.S
+++ b/arch/x86/crypto/sha512-mb/sha512_mb_mgr_flush_avx2.S
@@ -280,12 +280,18 @@ ENTRY(sha512_mb_mgr_get_comp_job_avx2)
 	pop     %rbx
         ret
 ENDPROC(sha512_mb_mgr_get_comp_job_avx2)
-.data
 
-.align 16
+.section	.rodata.cst8.one, "aM", @progbits, 8
+.align 8
 one:
 .quad  1
+
+.section	.rodata.cst8.two, "aM", @progbits, 8
+.align 8
 two:
 .quad  2
+
+.section	.rodata.cst8.three, "aM", @progbits, 8
+.align 8
 three:
 .quad  3
diff --git a/arch/x86/crypto/sha512-mb/sha512_mb_mgr_submit_avx2.S b/arch/x86/crypto/sha512-mb/sha512_mb_mgr_submit_avx2.S
index 815f07bdd1f8..4ba709ba78e5 100644
--- a/arch/x86/crypto/sha512-mb/sha512_mb_mgr_submit_avx2.S
+++ b/arch/x86/crypto/sha512-mb/sha512_mb_mgr_submit_avx2.S
@@ -209,8 +209,9 @@ return_null:
 	xor     job_rax, job_rax
 	jmp     return
 ENDPROC(sha512_mb_mgr_submit_avx2)
-.data
 
+/* UNUSED?
+.section	.rodata.cst16, "aM", @progbits, 16
 .align 16
 H0:     .int  0x6a09e667
 H1:     .int  0xbb67ae85
@@ -220,3 +221,4 @@ H4:     .int  0x510e527f
 H5:     .int  0x9b05688c
 H6:     .int  0x1f83d9ab
 H7:     .int  0x5be0cd19
+*/
diff --git a/arch/x86/crypto/sha512-mb/sha512_x4_avx2.S b/arch/x86/crypto/sha512-mb/sha512_x4_avx2.S
index 31ab1eff6413..e22e907643a6 100644
--- a/arch/x86/crypto/sha512-mb/sha512_x4_avx2.S
+++ b/arch/x86/crypto/sha512-mb/sha512_x4_avx2.S
@@ -361,7 +361,7 @@ Lrounds_16_xx:
 	ret
 ENDPROC(sha512_x4_avx2)
 
-.data
+.section	.rodata.K512_4, "a", @progbits
 .align 64
 K512_4:
 	.octa 0x428a2f98d728ae22428a2f98d728ae22,\
@@ -525,5 +525,7 @@ K512_4:
 	.octa 0x6c44198c4a4758176c44198c4a475817,\
 		0x6c44198c4a4758176c44198c4a475817
 
+.section	.rodata.cst32.PSHUFFLE_BYTE_FLIP_MASK, "aM", @progbits, 32
+.align 32
 PSHUFFLE_BYTE_FLIP_MASK: .octa 0x08090a0b0c0d0e0f0001020304050607
                          .octa 0x18191a1b1c1d1e1f1011121314151617
diff --git a/arch/x86/crypto/sha512-ssse3-asm.S b/arch/x86/crypto/sha512-ssse3-asm.S
index e610e29cbc81..66bbd9058a90 100644
--- a/arch/x86/crypto/sha512-ssse3-asm.S
+++ b/arch/x86/crypto/sha512-ssse3-asm.S
@@ -369,14 +369,17 @@ ENDPROC(sha512_transform_ssse3)
 ########################################################################
 ### Binary Data
 
-.data
-
+.section	.rodata.cst16.XMM_QWORD_BSWAP, "aM", @progbits, 16
 .align 16
-
 # Mask for byte-swapping a couple of qwords in an XMM register using (v)pshufb.
 XMM_QWORD_BSWAP:
 	.octa 0x08090a0b0c0d0e0f0001020304050607
 
+# Mergeable 640-byte rodata section. This allows linker to merge the table
+# with other, exactly the same 640-byte fragment of another rodata section
+# (if such section exists).
+.section	.rodata.cst640.K512, "aM", @progbits, 640
+.align 64
 # K[t] used in SHA512 hashing
 K512:
 	.quad 0x428a2f98d728ae22,0x7137449123ef65cd
diff --git a/arch/x86/crypto/twofish-avx-x86_64-asm_64.S b/arch/x86/crypto/twofish-avx-x86_64-asm_64.S
index dc66273e610d..b3f49d286348 100644
--- a/arch/x86/crypto/twofish-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/twofish-avx-x86_64-asm_64.S
@@ -29,11 +29,13 @@
 
 .file "twofish-avx-x86_64-asm_64.S"
 
-.data
+.section	.rodata.cst16.bswap128_mask, "aM", @progbits, 16
 .align 16
-
 .Lbswap128_mask:
 	.byte 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0
+
+.section	.rodata.cst16.xts_gf128mul_and_shl1_mask, "aM", @progbits, 16
+.align 16
 .Lxts_gf128mul_and_shl1_mask:
 	.byte 0x87, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0
author	Linus Torvalds <torvalds@linux-foundation.org>	2017-02-23 09:54:19 -0800
committer	Linus Torvalds <torvalds@linux-foundation.org>	2017-02-23 09:54:19 -0800
commit	5bcbe22ca47da04cda3a858cef67f55b550c1d13 (patch)
tree	49bd61e32eb2d652085a49182436322a3e0e9840 /arch
parent	1db934a5b77a9e37c4742c704fde6af233187a98 (diff)
parent	12cb3a1c4184f891d965d1f39f8cfcc9ef617647 (diff)