Adding upstream version 6.6.15.upstream/6.6.15

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

1 files changed, 3161 insertions, 0 deletions

diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S
new file mode 100644
index 0000000000..3ac7487eca
--- /dev/null
+++ b/arch/x86/crypto/aesni-intel_asm.S
@@ -0,0 +1,3161 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Implement AES algorithm in Intel AES-NI instructions.
+ *
+ * The white paper of AES-NI instructions can be downloaded from:
+ *   http://softwarecommunity.intel.com/isn/downloads/intelavx/AES-Instructions-Set_WP.pdf
+ *
+ * Copyright (C) 2008, Intel Corp.
+ *    Author: Huang Ying <ying.huang@intel.com>
+ *            Vinodh Gopal <vinodh.gopal@intel.com>
+ *            Kahraman Akdemir
+ *
+ * Added RFC4106 AES-GCM support for 128-bit keys under the AEAD
+ * interface for 64-bit kernels.
+ *    Authors: Erdinc Ozturk (erdinc.ozturk@intel.com)
+ *             Aidan O'Mahony (aidan.o.mahony@intel.com)
+ *             Adrian Hoban <adrian.hoban@intel.com>
+ *             James Guilford (james.guilford@intel.com)
+ *             Gabriele Paoloni <gabriele.paoloni@intel.com>
+ *             Tadeusz Struk (tadeusz.struk@intel.com)
+ *             Wajdi Feghali (wajdi.k.feghali@intel.com)
+ *    Copyright (c) 2010, Intel Corporation.
+ *
+ * Ported x86_64 version to x86:
+ *    Author: Mathias Krause <minipli@googlemail.com>
+ */
+
+#include <linux/linkage.h>
+#include <asm/frame.h>
+#include <asm/nospec-branch.h>
+
+/*
+ * The following macros are used to move an (un)aligned 16 byte value to/from
+ * an XMM register.  This can done for either FP or integer values, for FP use
+ * movaps (move aligned packed single) or integer use movdqa (move double quad
+ * aligned).  It doesn't make a performance difference which instruction is used
+ * since Nehalem (original Core i7) was released.  However, the movaps is a byte
+ * shorter, so that is the one we'll use for now. (same for unaligned).
+ */
+#define MOVADQ	movaps
+#define MOVUDQ	movups
+
+#ifdef __x86_64__
+
+# constants in mergeable sections, linker can reorder and merge
+.section	.rodata.cst16.POLY, "aM", @progbits, 16
+.align 16
+POLY:   .octa 0xC2000000000000000000000000000001
+.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
+.section	.rodata.cst16.MASK1, "aM", @progbits, 16
+.align 16
+MASK1:      .octa 0x0000000000000000ffffffffffffffff
+.section	.rodata.cst16.MASK2, "aM", @progbits, 16
+.align 16
+MASK2:      .octa 0xffffffffffffffff0000000000000000
+.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
+
+
+#define	STACK_OFFSET    8*3
+
+#define AadHash 16*0
+#define AadLen 16*1
+#define InLen (16*1)+8
+#define PBlockEncKey 16*2
+#define OrigIV 16*3
+#define CurCount 16*4
+#define PBlockLen 16*5
+#define	HashKey		16*6	// store HashKey <<1 mod poly here
+#define	HashKey_2	16*7	// store HashKey^2 <<1 mod poly here
+#define	HashKey_3	16*8	// store HashKey^3 <<1 mod poly here
+#define	HashKey_4	16*9	// store HashKey^4 <<1 mod poly here
+#define	HashKey_k	16*10	// store XOR of High 64 bits and Low 64
+				// bits of  HashKey <<1 mod poly here
+				//(for Karatsuba purposes)
+#define	HashKey_2_k	16*11	// store XOR of High 64 bits and Low 64
+				// bits of  HashKey^2 <<1 mod poly here
+				// (for Karatsuba purposes)
+#define	HashKey_3_k	16*12	// store XOR of High 64 bits and Low 64
+				// bits of  HashKey^3 <<1 mod poly here
+				// (for Karatsuba purposes)
+#define	HashKey_4_k	16*13	// store XOR of High 64 bits and Low 64
+				// bits of  HashKey^4 <<1 mod poly here
+				// (for Karatsuba purposes)
+
+#define arg1 rdi
+#define arg2 rsi
+#define arg3 rdx
+#define arg4 rcx
+#define arg5 r8
+#define arg6 r9
+#define arg7 STACK_OFFSET+8(%rsp)
+#define arg8 STACK_OFFSET+16(%rsp)
+#define arg9 STACK_OFFSET+24(%rsp)
+#define arg10 STACK_OFFSET+32(%rsp)
+#define arg11 STACK_OFFSET+40(%rsp)
+#define keysize 2*15*16(%arg1)
+#endif
+
+
+#define STATE1	%xmm0
+#define STATE2	%xmm4
+#define STATE3	%xmm5
+#define STATE4	%xmm6
+#define STATE	STATE1
+#define IN1	%xmm1
+#define IN2	%xmm7
+#define IN3	%xmm8
+#define IN4	%xmm9
+#define IN	IN1
+#define KEY	%xmm2
+#define IV	%xmm3
+
+#define BSWAP_MASK %xmm10
+#define CTR	%xmm11
+#define INC	%xmm12
+
+#define GF128MUL_MASK %xmm7
+
+#ifdef __x86_64__
+#define AREG	%rax
+#define KEYP	%rdi
+#define OUTP	%rsi
+#define UKEYP	OUTP
+#define INP	%rdx
+#define LEN	%rcx
+#define IVP	%r8
+#define KLEN	%r9d
+#define T1	%r10
+#define TKEYP	T1
+#define T2	%r11
+#define TCTR_LOW T2
+#else
+#define AREG	%eax
+#define KEYP	%edi
+#define OUTP	AREG
+#define UKEYP	OUTP
+#define INP	%edx
+#define LEN	%esi
+#define IVP	%ebp
+#define KLEN	%ebx
+#define T1	%ecx
+#define TKEYP	T1
+#endif
+
+.macro FUNC_SAVE
+	push	%r12
+	push	%r13
+	push	%r14
+#
+# states of %xmm registers %xmm6:%xmm15 not saved
+# all %xmm registers are clobbered
+#
+.endm
+
+
+.macro FUNC_RESTORE
+	pop	%r14
+	pop	%r13
+	pop	%r12
+.endm
+
+# Precompute hashkeys.
+# Input: Hash subkey.
+# Output: HashKeys stored in gcm_context_data.  Only needs to be called
+# once per key.
+# clobbers r12, and tmp xmm registers.
+.macro PRECOMPUTE SUBKEY TMP1 TMP2 TMP3 TMP4 TMP5 TMP6 TMP7
+	mov	\SUBKEY, %r12
+	movdqu	(%r12), \TMP3
+	movdqa	SHUF_MASK(%rip), \TMP2
+	pshufb	\TMP2, \TMP3
+
+	# precompute HashKey<<1 mod poly from the HashKey (required for GHASH)
+
+	movdqa	\TMP3, \TMP2
+	psllq	$1, \TMP3
+	psrlq	$63, \TMP2
+	movdqa	\TMP2, \TMP1
+	pslldq	$8, \TMP2
+	psrldq	$8, \TMP1
+	por	\TMP2, \TMP3
+
+	# reduce HashKey<<1
+
+	pshufd	$0x24, \TMP1, \TMP2
+	pcmpeqd TWOONE(%rip), \TMP2
+	pand	POLY(%rip), \TMP2
+	pxor	\TMP2, \TMP3
+	movdqu	\TMP3, HashKey(%arg2)
+
+	movdqa	   \TMP3, \TMP5
+	pshufd	   $78, \TMP3, \TMP1
+	pxor	   \TMP3, \TMP1
+	movdqu	   \TMP1, HashKey_k(%arg2)
+
+	GHASH_MUL  \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
+# TMP5 = HashKey^2<<1 (mod poly)
+	movdqu	   \TMP5, HashKey_2(%arg2)
+# HashKey_2 = HashKey^2<<1 (mod poly)
+	pshufd	   $78, \TMP5, \TMP1
+	pxor	   \TMP5, \TMP1
+	movdqu	   \TMP1, HashKey_2_k(%arg2)
+
+	GHASH_MUL  \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
+# TMP5 = HashKey^3<<1 (mod poly)
+	movdqu	   \TMP5, HashKey_3(%arg2)
+	pshufd	   $78, \TMP5, \TMP1
+	pxor	   \TMP5, \TMP1
+	movdqu	   \TMP1, HashKey_3_k(%arg2)
+
+	GHASH_MUL  \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
+# TMP5 = HashKey^3<<1 (mod poly)
+	movdqu	   \TMP5, HashKey_4(%arg2)
+	pshufd	   $78, \TMP5, \TMP1
+	pxor	   \TMP5, \TMP1
+	movdqu	   \TMP1, HashKey_4_k(%arg2)
+.endm
+
+# GCM_INIT initializes a gcm_context struct to prepare for encoding/decoding.
+# Clobbers rax, r10-r13 and xmm0-xmm6, %xmm13
+.macro GCM_INIT Iv SUBKEY AAD AADLEN
+	mov \AADLEN, %r11
+	mov %r11, AadLen(%arg2) # ctx_data.aad_length = aad_length
+	xor %r11d, %r11d
+	mov %r11, InLen(%arg2) # ctx_data.in_length = 0
+	mov %r11, PBlockLen(%arg2) # ctx_data.partial_block_length = 0
+	mov %r11, PBlockEncKey(%arg2) # ctx_data.partial_block_enc_key = 0
+	mov \Iv, %rax
+	movdqu (%rax), %xmm0
+	movdqu %xmm0, OrigIV(%arg2) # ctx_data.orig_IV = iv
+
+	movdqa  SHUF_MASK(%rip), %xmm2
+	pshufb %xmm2, %xmm0
+	movdqu %xmm0, CurCount(%arg2) # ctx_data.current_counter = iv
+
+	PRECOMPUTE \SUBKEY, %xmm1, %xmm2, %xmm3, %xmm4, %xmm5, %xmm6, %xmm7
+	movdqu HashKey(%arg2), %xmm13
+
+	CALC_AAD_HASH %xmm13, \AAD, \AADLEN, %xmm0, %xmm1, %xmm2, %xmm3, \
+	%xmm4, %xmm5, %xmm6
+.endm
+
+# GCM_ENC_DEC Encodes/Decodes given data. Assumes that the passed gcm_context
+# struct has been initialized by GCM_INIT.
+# Requires the input data be at least 1 byte long because of READ_PARTIAL_BLOCK
+# Clobbers rax, r10-r13, and xmm0-xmm15
+.macro GCM_ENC_DEC operation
+	movdqu AadHash(%arg2), %xmm8
+	movdqu HashKey(%arg2), %xmm13
+	add %arg5, InLen(%arg2)
+
+	xor %r11d, %r11d # initialise the data pointer offset as zero
+	PARTIAL_BLOCK %arg3 %arg4 %arg5 %r11 %xmm8 \operation
+
+	sub %r11, %arg5		# sub partial block data used
+	mov %arg5, %r13		# save the number of bytes
+
+	and $-16, %r13		# %r13 = %r13 - (%r13 mod 16)
+	mov %r13, %r12
+	# Encrypt/Decrypt first few blocks
+
+	and	$(3<<4), %r12
+	jz	.L_initial_num_blocks_is_0_\@
+	cmp	$(2<<4), %r12
+	jb	.L_initial_num_blocks_is_1_\@
+	je	.L_initial_num_blocks_is_2_\@
+.L_initial_num_blocks_is_3_\@:
+	INITIAL_BLOCKS_ENC_DEC	%xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+%xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 5, 678, \operation
+	sub	$48, %r13
+	jmp	.L_initial_blocks_\@
+.L_initial_num_blocks_is_2_\@:
+	INITIAL_BLOCKS_ENC_DEC	%xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+%xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 6, 78, \operation
+	sub	$32, %r13
+	jmp	.L_initial_blocks_\@
+.L_initial_num_blocks_is_1_\@:
+	INITIAL_BLOCKS_ENC_DEC	%xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+%xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 7, 8, \operation
+	sub	$16, %r13
+	jmp	.L_initial_blocks_\@
+.L_initial_num_blocks_is_0_\@:
+	INITIAL_BLOCKS_ENC_DEC	%xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+%xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 8, 0, \operation
+.L_initial_blocks_\@:
+
+	# Main loop - Encrypt/Decrypt remaining blocks
+
+	test	%r13, %r13
+	je	.L_zero_cipher_left_\@
+	sub	$64, %r13
+	je	.L_four_cipher_left_\@
+.L_crypt_by_4_\@:
+	GHASH_4_ENCRYPT_4_PARALLEL_\operation	%xmm9, %xmm10, %xmm11, %xmm12, \
+	%xmm13, %xmm14, %xmm0, %xmm1, %xmm2, %xmm3, %xmm4, %xmm5, %xmm6, \
+	%xmm7, %xmm8, enc
+	add	$64, %r11
+	sub	$64, %r13
+	jne	.L_crypt_by_4_\@
+.L_four_cipher_left_\@:
+	GHASH_LAST_4	%xmm9, %xmm10, %xmm11, %xmm12, %xmm13, %xmm14, \
+%xmm15, %xmm1, %xmm2, %xmm3, %xmm4, %xmm8
+.L_zero_cipher_left_\@:
+	movdqu %xmm8, AadHash(%arg2)
+	movdqu %xmm0, CurCount(%arg2)
+
+	mov	%arg5, %r13
+	and	$15, %r13			# %r13 = arg5 (mod 16)
+	je	.L_multiple_of_16_bytes_\@
+
+	mov %r13, PBlockLen(%arg2)
+
+	# Handle the last <16 Byte block separately
+	paddd ONE(%rip), %xmm0                # INCR CNT to get Yn
+	movdqu %xmm0, CurCount(%arg2)
+	movdqa SHUF_MASK(%rip), %xmm10
+	pshufb %xmm10, %xmm0
+
+	ENCRYPT_SINGLE_BLOCK	%xmm0, %xmm1        # Encrypt(K, Yn)
+	movdqu %xmm0, PBlockEncKey(%arg2)
+
+	cmp	$16, %arg5
+	jge	.L_large_enough_update_\@
+
+	lea (%arg4,%r11,1), %r10
+	mov %r13, %r12
+	READ_PARTIAL_BLOCK %r10 %r12 %xmm2 %xmm1
+	jmp	.L_data_read_\@
+
+.L_large_enough_update_\@:
+	sub	$16, %r11
+	add	%r13, %r11
+
+	# receive the last <16 Byte block
+	movdqu	(%arg4, %r11, 1), %xmm1
+
+	sub	%r13, %r11
+	add	$16, %r11
+
+	lea	SHIFT_MASK+16(%rip), %r12
+	# adjust the shuffle mask pointer to be able to shift 16-r13 bytes
+	# (r13 is the number of bytes in plaintext mod 16)
+	sub	%r13, %r12
+	# get the appropriate shuffle mask
+	movdqu	(%r12), %xmm2
+	# shift right 16-r13 bytes
+	pshufb  %xmm2, %xmm1
+
+.L_data_read_\@:
+	lea ALL_F+16(%rip), %r12
+	sub %r13, %r12
+
+.ifc \operation, dec
+	movdqa  %xmm1, %xmm2
+.endif
+	pxor	%xmm1, %xmm0            # XOR Encrypt(K, Yn)
+	movdqu	(%r12), %xmm1
+	# get the appropriate mask to mask out top 16-r13 bytes of xmm0
+	pand	%xmm1, %xmm0            # mask out top 16-r13 bytes of xmm0
+.ifc \operation, dec
+	pand    %xmm1, %xmm2
+	movdqa SHUF_MASK(%rip), %xmm10
+	pshufb %xmm10 ,%xmm2
+
+	pxor %xmm2, %xmm8
+.else
+	movdqa SHUF_MASK(%rip), %xmm10
+	pshufb %xmm10,%xmm0
+
+	pxor	%xmm0, %xmm8
+.endif
+
+	movdqu %xmm8, AadHash(%arg2)
+.ifc \operation, enc
+	# GHASH computation for the last <16 byte block
+	movdqa SHUF_MASK(%rip), %xmm10
+	# shuffle xmm0 back to output as ciphertext
+	pshufb %xmm10, %xmm0
+.endif
+
+	# Output %r13 bytes
+	movq %xmm0, %rax
+	cmp $8, %r13
+	jle .L_less_than_8_bytes_left_\@
+	mov %rax, (%arg3 , %r11, 1)
+	add $8, %r11
+	psrldq $8, %xmm0
+	movq %xmm0, %rax
+	sub $8, %r13
+.L_less_than_8_bytes_left_\@:
+	mov %al,  (%arg3, %r11, 1)
+	add $1, %r11
+	shr $8, %rax
+	sub $1, %r13
+	jne .L_less_than_8_bytes_left_\@
+.L_multiple_of_16_bytes_\@:
+.endm
+
+# GCM_COMPLETE Finishes update of tag of last partial block
+# Output: Authorization Tag (AUTH_TAG)
+# Clobbers rax, r10-r12, and xmm0, xmm1, xmm5-xmm15
+.macro GCM_COMPLETE AUTHTAG AUTHTAGLEN
+	movdqu AadHash(%arg2), %xmm8
+	movdqu HashKey(%arg2), %xmm13
+
+	mov PBlockLen(%arg2), %r12
+
+	test %r12, %r12
+	je .L_partial_done\@
+
+	GHASH_MUL %xmm8, %xmm13, %xmm9, %xmm10, %xmm11, %xmm5, %xmm6
+
+.L_partial_done\@:
+	mov AadLen(%arg2), %r12  # %r13 = aadLen (number of bytes)
+	shl	$3, %r12		  # convert into number of bits
+	movd	%r12d, %xmm15		  # len(A) in %xmm15
+	mov InLen(%arg2), %r12
+	shl     $3, %r12                  # len(C) in bits (*128)
+	movq    %r12, %xmm1
+
+	pslldq	$8, %xmm15		  # %xmm15 = len(A)||0x0000000000000000
+	pxor	%xmm1, %xmm15		  # %xmm15 = len(A)||len(C)
+	pxor	%xmm15, %xmm8
+	GHASH_MUL	%xmm8, %xmm13, %xmm9, %xmm10, %xmm11, %xmm5, %xmm6
+	# final GHASH computation
+	movdqa SHUF_MASK(%rip), %xmm10
+	pshufb %xmm10, %xmm8
+
+	movdqu OrigIV(%arg2), %xmm0       # %xmm0 = Y0
+	ENCRYPT_SINGLE_BLOCK	%xmm0,  %xmm1	  # E(K, Y0)
+	pxor	%xmm8, %xmm0
+.L_return_T_\@:
+	mov	\AUTHTAG, %r10                     # %r10 = authTag
+	mov	\AUTHTAGLEN, %r11                    # %r11 = auth_tag_len
+	cmp	$16, %r11
+	je	.L_T_16_\@
+	cmp	$8, %r11
+	jl	.L_T_4_\@
+.L_T_8_\@:
+	movq	%xmm0, %rax
+	mov	%rax, (%r10)
+	add	$8, %r10
+	sub	$8, %r11
+	psrldq	$8, %xmm0
+	test	%r11, %r11
+	je	.L_return_T_done_\@
+.L_T_4_\@:
+	movd	%xmm0, %eax
+	mov	%eax, (%r10)
+	add	$4, %r10
+	sub	$4, %r11
+	psrldq	$4, %xmm0
+	test	%r11, %r11
+	je	.L_return_T_done_\@
+.L_T_123_\@:
+	movd	%xmm0, %eax
+	cmp	$2, %r11
+	jl	.L_T_1_\@
+	mov	%ax, (%r10)
+	cmp	$2, %r11
+	je	.L_return_T_done_\@
+	add	$2, %r10
+	sar	$16, %eax
+.L_T_1_\@:
+	mov	%al, (%r10)
+	jmp	.L_return_T_done_\@
+.L_T_16_\@:
+	movdqu	%xmm0, (%r10)
+.L_return_T_done_\@:
+.endm
+
+#ifdef __x86_64__
+/* GHASH_MUL MACRO to implement: Data*HashKey mod (128,127,126,121,0)
+*
+*
+* Input: A and B (128-bits each, bit-reflected)
+* Output: C = A*B*x mod poly, (i.e. >>1 )
+* To compute GH = GH*HashKey mod poly, give HK = HashKey<<1 mod poly as input
+* GH = GH * HK * x mod poly which is equivalent to GH*HashKey mod poly.
+*
+*/
+.macro GHASH_MUL GH HK TMP1 TMP2 TMP3 TMP4 TMP5
+	movdqa	  \GH, \TMP1
+	pshufd	  $78, \GH, \TMP2
+	pshufd	  $78, \HK, \TMP3
+	pxor	  \GH, \TMP2            # TMP2 = a1+a0
+	pxor	  \HK, \TMP3            # TMP3 = b1+b0
+	pclmulqdq $0x11, \HK, \TMP1     # TMP1 = a1*b1
+	pclmulqdq $0x00, \HK, \GH       # GH = a0*b0
+	pclmulqdq $0x00, \TMP3, \TMP2   # TMP2 = (a0+a1)*(b1+b0)
+	pxor	  \GH, \TMP2
+	pxor	  \TMP1, \TMP2          # TMP2 = (a0*b0)+(a1*b0)
+	movdqa	  \TMP2, \TMP3
+	pslldq	  $8, \TMP3             # left shift TMP3 2 DWs
+	psrldq	  $8, \TMP2             # right shift TMP2 2 DWs
+	pxor	  \TMP3, \GH
+	pxor	  \TMP2, \TMP1          # TMP2:GH holds the result of GH*HK
+
+        # first phase of the reduction
+
+	movdqa    \GH, \TMP2
+	movdqa    \GH, \TMP3
+	movdqa    \GH, \TMP4            # copy GH into TMP2,TMP3 and TMP4
+					# in in order to perform
+					# independent shifts
+	pslld     $31, \TMP2            # packed right shift <<31
+	pslld     $30, \TMP3            # packed right shift <<30
+	pslld     $25, \TMP4            # packed right shift <<25
+	pxor      \TMP3, \TMP2          # xor the shifted versions
+	pxor      \TMP4, \TMP2
+	movdqa    \TMP2, \TMP5
+	psrldq    $4, \TMP5             # right shift TMP5 1 DW
+	pslldq    $12, \TMP2            # left shift TMP2 3 DWs
+	pxor      \TMP2, \GH
+
+        # second phase of the reduction
+
+	movdqa    \GH,\TMP2             # copy GH into TMP2,TMP3 and TMP4
+					# in in order to perform
+					# independent shifts
+	movdqa    \GH,\TMP3
+	movdqa    \GH,\TMP4
+	psrld     $1,\TMP2              # packed left shift >>1
+	psrld     $2,\TMP3              # packed left shift >>2
+	psrld     $7,\TMP4              # packed left shift >>7
+	pxor      \TMP3,\TMP2		# xor the shifted versions
+	pxor      \TMP4,\TMP2
+	pxor      \TMP5, \TMP2
+	pxor      \TMP2, \GH
+	pxor      \TMP1, \GH            # result is in TMP1
+.endm
+
+# Reads DLEN bytes starting at DPTR and stores in XMMDst
+# where 0 < DLEN < 16
+# Clobbers %rax, DLEN and XMM1
+.macro READ_PARTIAL_BLOCK DPTR DLEN XMM1 XMMDst
+        cmp $8, \DLEN
+        jl .L_read_lt8_\@
+        mov (\DPTR), %rax
+        movq %rax, \XMMDst
+        sub $8, \DLEN
+        jz .L_done_read_partial_block_\@
+	xor %eax, %eax
+.L_read_next_byte_\@:
+        shl $8, %rax
+        mov 7(\DPTR, \DLEN, 1), %al
+        dec \DLEN
+        jnz .L_read_next_byte_\@
+        movq %rax, \XMM1
+	pslldq $8, \XMM1
+        por \XMM1, \XMMDst
+	jmp .L_done_read_partial_block_\@
+.L_read_lt8_\@:
+	xor %eax, %eax
+.L_read_next_byte_lt8_\@:
+        shl $8, %rax
+        mov -1(\DPTR, \DLEN, 1), %al
+        dec \DLEN
+        jnz .L_read_next_byte_lt8_\@
+        movq %rax, \XMMDst
+.L_done_read_partial_block_\@:
+.endm
+
+# CALC_AAD_HASH: Calculates the hash of the data which will not be encrypted.
+# clobbers r10-11, xmm14
+.macro CALC_AAD_HASH HASHKEY AAD AADLEN TMP1 TMP2 TMP3 TMP4 TMP5 \
+	TMP6 TMP7
+	MOVADQ	   SHUF_MASK(%rip), %xmm14
+	mov	   \AAD, %r10		# %r10 = AAD
+	mov	   \AADLEN, %r11		# %r11 = aadLen
+	pxor	   \TMP7, \TMP7
+	pxor	   \TMP6, \TMP6
+
+	cmp	   $16, %r11
+	jl	   .L_get_AAD_rest\@
+.L_get_AAD_blocks\@:
+	movdqu	   (%r10), \TMP7
+	pshufb	   %xmm14, \TMP7 # byte-reflect the AAD data
+	pxor	   \TMP7, \TMP6
+	GHASH_MUL  \TMP6, \HASHKEY, \TMP1, \TMP2, \TMP3, \TMP4, \TMP5
+	add	   $16, %r10
+	sub	   $16, %r11
+	cmp	   $16, %r11
+	jge	   .L_get_AAD_blocks\@
+
+	movdqu	   \TMP6, \TMP7
+
+	/* read the last <16B of AAD */
+.L_get_AAD_rest\@:
+	test	   %r11, %r11
+	je	   .L_get_AAD_done\@
+
+	READ_PARTIAL_BLOCK %r10, %r11, \TMP1, \TMP7
+	pshufb	   %xmm14, \TMP7 # byte-reflect the AAD data
+	pxor	   \TMP6, \TMP7
+	GHASH_MUL  \TMP7, \HASHKEY, \TMP1, \TMP2, \TMP3, \TMP4, \TMP5
+	movdqu \TMP7, \TMP6
+
+.L_get_AAD_done\@:
+	movdqu \TMP6, AadHash(%arg2)
+.endm
+
+# PARTIAL_BLOCK: Handles encryption/decryption and the tag partial blocks
+# between update calls.
+# Requires the input data be at least 1 byte long due to READ_PARTIAL_BLOCK
+# Outputs encrypted bytes, and updates hash and partial info in gcm_data_context
+# Clobbers rax, r10, r12, r13, xmm0-6, xmm9-13
+.macro PARTIAL_BLOCK CYPH_PLAIN_OUT PLAIN_CYPH_IN PLAIN_CYPH_LEN DATA_OFFSET \
+	AAD_HASH operation
+	mov 	PBlockLen(%arg2), %r13
+	test	%r13, %r13
+	je	.L_partial_block_done_\@	# Leave Macro if no partial blocks
+	# Read in input data without over reading
+	cmp	$16, \PLAIN_CYPH_LEN
+	jl	.L_fewer_than_16_bytes_\@
+	movups	(\PLAIN_CYPH_IN), %xmm1	# If more than 16 bytes, just fill xmm
+	jmp	.L_data_read_\@
+
+.L_fewer_than_16_bytes_\@:
+	lea	(\PLAIN_CYPH_IN, \DATA_OFFSET, 1), %r10
+	mov	\PLAIN_CYPH_LEN, %r12
+	READ_PARTIAL_BLOCK %r10 %r12 %xmm0 %xmm1
+
+	mov PBlockLen(%arg2), %r13
+
+.L_data_read_\@:				# Finished reading in data
+
+	movdqu	PBlockEncKey(%arg2), %xmm9
+	movdqu	HashKey(%arg2), %xmm13
+
+	lea	SHIFT_MASK(%rip), %r12
+
+	# adjust the shuffle mask pointer to be able to shift r13 bytes
+	# r16-r13 is the number of bytes in plaintext mod 16)
+	add	%r13, %r12
+	movdqu	(%r12), %xmm2		# get the appropriate shuffle mask
+	pshufb	%xmm2, %xmm9		# shift right r13 bytes
+
+.ifc \operation, dec
+	movdqa	%xmm1, %xmm3
+	pxor	%xmm1, %xmm9		# Cyphertext XOR E(K, Yn)
+
+	mov	\PLAIN_CYPH_LEN, %r10
+	add	%r13, %r10
+	# Set r10 to be the amount of data left in CYPH_PLAIN_IN after filling
+	sub	$16, %r10
+	# Determine if if partial block is not being filled and
+	# shift mask accordingly
+	jge	.L_no_extra_mask_1_\@
+	sub	%r10, %r12
+.L_no_extra_mask_1_\@:
+
+	movdqu	ALL_F-SHIFT_MASK(%r12), %xmm1
+	# get the appropriate mask to mask out bottom r13 bytes of xmm9
+	pand	%xmm1, %xmm9		# mask out bottom r13 bytes of xmm9
+
+	pand	%xmm1, %xmm3
+	movdqa	SHUF_MASK(%rip), %xmm10
+	pshufb	%xmm10, %xmm3
+	pshufb	%xmm2, %xmm3
+	pxor	%xmm3, \AAD_HASH
+
+	test	%r10, %r10
+	jl	.L_partial_incomplete_1_\@
+
+	# GHASH computation for the last <16 Byte block
+	GHASH_MUL \AAD_HASH, %xmm13, %xmm0, %xmm10, %xmm11, %xmm5, %xmm6
+	xor	%eax, %eax
+
+	mov	%rax, PBlockLen(%arg2)
+	jmp	.L_dec_done_\@
+.L_partial_incomplete_1_\@:
+	add	\PLAIN_CYPH_LEN, PBlockLen(%arg2)
+.L_dec_done_\@:
+	movdqu	\AAD_HASH, AadHash(%arg2)
+.else
+	pxor	%xmm1, %xmm9			# Plaintext XOR E(K, Yn)
+
+	mov	\PLAIN_CYPH_LEN, %r10
+	add	%r13, %r10
+	# Set r10 to be the amount of data left in CYPH_PLAIN_IN after filling
+	sub	$16, %r10
+	# Determine if if partial block is not being filled and
+	# shift mask accordingly
+	jge	.L_no_extra_mask_2_\@
+	sub	%r10, %r12
+.L_no_extra_mask_2_\@:
+
+	movdqu	ALL_F-SHIFT_MASK(%r12), %xmm1
+	# get the appropriate mask to mask out bottom r13 bytes of xmm9
+	pand	%xmm1, %xmm9
+
+	movdqa	SHUF_MASK(%rip), %xmm1
+	pshufb	%xmm1, %xmm9
+	pshufb	%xmm2, %xmm9
+	pxor	%xmm9, \AAD_HASH
+
+	test	%r10, %r10
+	jl	.L_partial_incomplete_2_\@
+
+	# GHASH computation for the last <16 Byte block
+	GHASH_MUL \AAD_HASH, %xmm13, %xmm0, %xmm10, %xmm11, %xmm5, %xmm6
+	xor	%eax, %eax
+
+	mov	%rax, PBlockLen(%arg2)
+	jmp	.L_encode_done_\@
+.L_partial_incomplete_2_\@:
+	add	\PLAIN_CYPH_LEN, PBlockLen(%arg2)
+.L_encode_done_\@:
+	movdqu	\AAD_HASH, AadHash(%arg2)
+
+	movdqa	SHUF_MASK(%rip), %xmm10
+	# shuffle xmm9 back to output as ciphertext
+	pshufb	%xmm10, %xmm9
+	pshufb	%xmm2, %xmm9
+.endif
+	# output encrypted Bytes
+	test	%r10, %r10
+	jl	.L_partial_fill_\@
+	mov	%r13, %r12
+	mov	$16, %r13
+	# Set r13 to be the number of bytes to write out
+	sub	%r12, %r13
+	jmp	.L_count_set_\@
+.L_partial_fill_\@:
+	mov	\PLAIN_CYPH_LEN, %r13
+.L_count_set_\@:
+	movdqa	%xmm9, %xmm0
+	movq	%xmm0, %rax
+	cmp	$8, %r13
+	jle	.L_less_than_8_bytes_left_\@
+
+	mov	%rax, (\CYPH_PLAIN_OUT, \DATA_OFFSET, 1)
+	add	$8, \DATA_OFFSET
+	psrldq	$8, %xmm0
+	movq	%xmm0, %rax
+	sub	$8, %r13
+.L_less_than_8_bytes_left_\@:
+	movb	%al, (\CYPH_PLAIN_OUT, \DATA_OFFSET, 1)
+	add	$1, \DATA_OFFSET
+	shr	$8, %rax
+	sub	$1, %r13
+	jne	.L_less_than_8_bytes_left_\@
+.L_partial_block_done_\@:
+.endm # PARTIAL_BLOCK
+
+/*
+* if a = number of total plaintext bytes
+* b = floor(a/16)
+* num_initial_blocks = b mod 4
+* encrypt the initial num_initial_blocks blocks and apply ghash on
+* the ciphertext
+* %r10, %r11, %r12, %rax, %xmm5, %xmm6, %xmm7, %xmm8, %xmm9 registers
+* are clobbered
+* arg1, %arg2, %arg3 are used as a pointer only, not modified
+*/
+
+
+.macro INITIAL_BLOCKS_ENC_DEC TMP1 TMP2 TMP3 TMP4 TMP5 XMM0 XMM1 \
+	XMM2 XMM3 XMM4 XMMDst TMP6 TMP7 i i_seq operation
+	MOVADQ		SHUF_MASK(%rip), %xmm14
+
+	movdqu AadHash(%arg2), %xmm\i		    # XMM0 = Y0
+
+	# start AES for num_initial_blocks blocks
+
+	movdqu CurCount(%arg2), \XMM0                # XMM0 = Y0
+
+.if (\i == 5) || (\i == 6) || (\i == 7)
+
+	MOVADQ		ONE(%RIP),\TMP1
+	MOVADQ		0(%arg1),\TMP2
+.irpc index, \i_seq
+	paddd		\TMP1, \XMM0                 # INCR Y0
+.ifc \operation, dec
+        movdqa     \XMM0, %xmm\index
+.else
+	MOVADQ		\XMM0, %xmm\index
+.endif
+	pshufb	%xmm14, %xmm\index      # perform a 16 byte swap
+	pxor		\TMP2, %xmm\index
+.endr
+	lea	0x10(%arg1),%r10
+	mov	keysize,%eax
+	shr	$2,%eax				# 128->4, 192->6, 256->8
+	add	$5,%eax			      # 128->9, 192->11, 256->13
+
+.Laes_loop_initial_\@:
+	MOVADQ	(%r10),\TMP1
+.irpc	index, \i_seq
+	aesenc	\TMP1, %xmm\index
+.endr
+	add	$16,%r10
+	sub	$1,%eax
+	jnz	.Laes_loop_initial_\@
+
+	MOVADQ	(%r10), \TMP1
+.irpc index, \i_seq
+	aesenclast \TMP1, %xmm\index         # Last Round
+.endr
+.irpc index, \i_seq
+	movdqu	   (%arg4 , %r11, 1), \TMP1
+	pxor	   \TMP1, %xmm\index
+	movdqu	   %xmm\index, (%arg3 , %r11, 1)
+	# write back plaintext/ciphertext for num_initial_blocks
+	add	   $16, %r11
+
+.ifc \operation, dec
+	movdqa     \TMP1, %xmm\index
+.endif
+	pshufb	   %xmm14, %xmm\index
+
+		# prepare plaintext/ciphertext for GHASH computation
+.endr
+.endif
+
+        # apply GHASH on num_initial_blocks blocks
+
+.if \i == 5
+        pxor       %xmm5, %xmm6
+	GHASH_MUL  %xmm6, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+        pxor       %xmm6, %xmm7
+	GHASH_MUL  %xmm7, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+        pxor       %xmm7, %xmm8
+	GHASH_MUL  %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+.elseif \i == 6
+        pxor       %xmm6, %xmm7
+	GHASH_MUL  %xmm7, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+        pxor       %xmm7, %xmm8
+	GHASH_MUL  %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+.elseif \i == 7
+        pxor       %xmm7, %xmm8
+	GHASH_MUL  %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+.endif
+	cmp	   $64, %r13
+	jl	.L_initial_blocks_done\@
+	# no need for precomputed values
+/*
+*
+* Precomputations for HashKey parallel with encryption of first 4 blocks.
+* Haskey_i_k holds XORed values of the low and high parts of the Haskey_i
+*/
+	MOVADQ	   ONE(%RIP),\TMP1
+	paddd	   \TMP1, \XMM0              # INCR Y0
+	MOVADQ	   \XMM0, \XMM1
+	pshufb  %xmm14, \XMM1        # perform a 16 byte swap
+
+	paddd	   \TMP1, \XMM0              # INCR Y0
+	MOVADQ	   \XMM0, \XMM2
+	pshufb  %xmm14, \XMM2        # perform a 16 byte swap
+
+	paddd	   \TMP1, \XMM0              # INCR Y0
+	MOVADQ	   \XMM0, \XMM3
+	pshufb %xmm14, \XMM3        # perform a 16 byte swap
+
+	paddd	   \TMP1, \XMM0              # INCR Y0
+	MOVADQ	   \XMM0, \XMM4
+	pshufb %xmm14, \XMM4        # perform a 16 byte swap
+
+	MOVADQ	   0(%arg1),\TMP1
+	pxor	   \TMP1, \XMM1
+	pxor	   \TMP1, \XMM2
+	pxor	   \TMP1, \XMM3
+	pxor	   \TMP1, \XMM4
+.irpc index, 1234 # do 4 rounds
+	movaps 0x10*\index(%arg1), \TMP1
+	aesenc	   \TMP1, \XMM1
+	aesenc	   \TMP1, \XMM2
+	aesenc	   \TMP1, \XMM3
+	aesenc	   \TMP1, \XMM4
+.endr
+.irpc index, 56789 # do next 5 rounds
+	movaps 0x10*\index(%arg1), \TMP1
+	aesenc	   \TMP1, \XMM1
+	aesenc	   \TMP1, \XMM2
+	aesenc	   \TMP1, \XMM3
+	aesenc	   \TMP1, \XMM4
+.endr
+	lea	   0xa0(%arg1),%r10
+	mov	   keysize,%eax
+	shr	   $2,%eax			# 128->4, 192->6, 256->8
+	sub	   $4,%eax			# 128->0, 192->2, 256->4
+	jz	   .Laes_loop_pre_done\@
+
+.Laes_loop_pre_\@:
+	MOVADQ	   (%r10),\TMP2
+.irpc	index, 1234
+	aesenc	   \TMP2, %xmm\index
+.endr
+	add	   $16,%r10
+	sub	   $1,%eax
+	jnz	   .Laes_loop_pre_\@
+
+.Laes_loop_pre_done\@:
+	MOVADQ	   (%r10), \TMP2
+	aesenclast \TMP2, \XMM1
+	aesenclast \TMP2, \XMM2
+	aesenclast \TMP2, \XMM3
+	aesenclast \TMP2, \XMM4
+	movdqu	   16*0(%arg4 , %r11 , 1), \TMP1
+	pxor	   \TMP1, \XMM1
+.ifc \operation, dec
+	movdqu     \XMM1, 16*0(%arg3 , %r11 , 1)
+	movdqa     \TMP1, \XMM1
+.endif
+	movdqu	   16*1(%arg4 , %r11 , 1), \TMP1
+	pxor	   \TMP1, \XMM2
+.ifc \operation, dec
+	movdqu     \XMM2, 16*1(%arg3 , %r11 , 1)
+	movdqa     \TMP1, \XMM2
+.endif
+	movdqu	   16*2(%arg4 , %r11 , 1), \TMP1
+	pxor	   \TMP1, \XMM3
+.ifc \operation, dec
+	movdqu     \XMM3, 16*2(%arg3 , %r11 , 1)
+	movdqa     \TMP1, \XMM3
+.endif
+	movdqu	   16*3(%arg4 , %r11 , 1), \TMP1
+	pxor	   \TMP1, \XMM4
+.ifc \operation, dec
+	movdqu     \XMM4, 16*3(%arg3 , %r11 , 1)
+	movdqa     \TMP1, \XMM4
+.else
+	movdqu     \XMM1, 16*0(%arg3 , %r11 , 1)
+	movdqu     \XMM2, 16*1(%arg3 , %r11 , 1)
+	movdqu     \XMM3, 16*2(%arg3 , %r11 , 1)
+	movdqu     \XMM4, 16*3(%arg3 , %r11 , 1)
+.endif
+
+	add	   $64, %r11
+	pshufb %xmm14, \XMM1 # perform a 16 byte swap
+	pxor	   \XMMDst, \XMM1
+# combine GHASHed value with the corresponding ciphertext
+	pshufb %xmm14, \XMM2 # perform a 16 byte swap
+	pshufb %xmm14, \XMM3 # perform a 16 byte swap
+	pshufb %xmm14, \XMM4 # perform a 16 byte swap
+
+.L_initial_blocks_done\@:
+
+.endm
+
+/*
+* encrypt 4 blocks at a time
+* ghash the 4 previously encrypted ciphertext blocks
+* arg1, %arg3, %arg4 are used as pointers only, not modified
+* %r11 is the data offset value
+*/
+.macro GHASH_4_ENCRYPT_4_PARALLEL_enc TMP1 TMP2 TMP3 TMP4 TMP5 \
+TMP6 XMM0 XMM1 XMM2 XMM3 XMM4 XMM5 XMM6 XMM7 XMM8 operation
+
+	movdqa	  \XMM1, \XMM5
+	movdqa	  \XMM2, \XMM6
+	movdqa	  \XMM3, \XMM7
+	movdqa	  \XMM4, \XMM8
+
+        movdqa    SHUF_MASK(%rip), %xmm15
+        # multiply TMP5 * HashKey using karatsuba
+
+	movdqa	  \XMM5, \TMP4
+	pshufd	  $78, \XMM5, \TMP6
+	pxor	  \XMM5, \TMP6
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqu	  HashKey_4(%arg2), \TMP5
+	pclmulqdq $0x11, \TMP5, \TMP4           # TMP4 = a1*b1
+	movdqa    \XMM0, \XMM1
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqa    \XMM0, \XMM2
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqa    \XMM0, \XMM3
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqa    \XMM0, \XMM4
+	pshufb %xmm15, \XMM1	# perform a 16 byte swap
+	pclmulqdq $0x00, \TMP5, \XMM5           # XMM5 = a0*b0
+	pshufb %xmm15, \XMM2	# perform a 16 byte swap
+	pshufb %xmm15, \XMM3	# perform a 16 byte swap
+	pshufb %xmm15, \XMM4	# perform a 16 byte swap
+
+	pxor	  (%arg1), \XMM1
+	pxor	  (%arg1), \XMM2
+	pxor	  (%arg1), \XMM3
+	pxor	  (%arg1), \XMM4
+	movdqu	  HashKey_4_k(%arg2), \TMP5
+	pclmulqdq $0x00, \TMP5, \TMP6       # TMP6 = (a1+a0)*(b1+b0)
+	movaps 0x10(%arg1), \TMP1
+	aesenc	  \TMP1, \XMM1              # Round 1
+	aesenc	  \TMP1, \XMM2
+	aesenc	  \TMP1, \XMM3
+	aesenc	  \TMP1, \XMM4
+	movaps 0x20(%arg1), \TMP1
+	aesenc	  \TMP1, \XMM1              # Round 2
+	aesenc	  \TMP1, \XMM2
+	aesenc	  \TMP1, \XMM3
+	aesenc	  \TMP1, \XMM4
+	movdqa	  \XMM6, \TMP1
+	pshufd	  $78, \XMM6, \TMP2
+	pxor	  \XMM6, \TMP2
+	movdqu	  HashKey_3(%arg2), \TMP5
+	pclmulqdq $0x11, \TMP5, \TMP1       # TMP1 = a1 * b1
+	movaps 0x30(%arg1), \TMP3
+	aesenc    \TMP3, \XMM1              # Round 3
+	aesenc    \TMP3, \XMM2
+	aesenc    \TMP3, \XMM3
+	aesenc    \TMP3, \XMM4
+	pclmulqdq $0x00, \TMP5, \XMM6       # XMM6 = a0*b0
+	movaps 0x40(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1              # Round 4
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	movdqu	  HashKey_3_k(%arg2), \TMP5
+	pclmulqdq $0x00, \TMP5, \TMP2       # TMP2 = (a1+a0)*(b1+b0)
+	movaps 0x50(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1              # Round 5
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	pxor	  \TMP1, \TMP4
+# accumulate the results in TMP4:XMM5, TMP6 holds the middle part
+	pxor	  \XMM6, \XMM5
+	pxor	  \TMP2, \TMP6
+	movdqa	  \XMM7, \TMP1
+	pshufd	  $78, \XMM7, \TMP2
+	pxor	  \XMM7, \TMP2
+	movdqu	  HashKey_2(%arg2), \TMP5
+
+        # Multiply TMP5 * HashKey using karatsuba
+
+	pclmulqdq $0x11, \TMP5, \TMP1       # TMP1 = a1*b1
+	movaps 0x60(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1              # Round 6
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	pclmulqdq $0x00, \TMP5, \XMM7       # XMM7 = a0*b0
+	movaps 0x70(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1              # Round 7
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	movdqu	  HashKey_2_k(%arg2), \TMP5
+	pclmulqdq $0x00, \TMP5, \TMP2       # TMP2 = (a1+a0)*(b1+b0)
+	movaps 0x80(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1              # Round 8
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	pxor	  \TMP1, \TMP4
+# accumulate the results in TMP4:XMM5, TMP6 holds the middle part
+	pxor	  \XMM7, \XMM5
+	pxor	  \TMP2, \TMP6
+
+        # Multiply XMM8 * HashKey
+        # XMM8 and TMP5 hold the values for the two operands
+
+	movdqa	  \XMM8, \TMP1
+	pshufd	  $78, \XMM8, \TMP2
+	pxor	  \XMM8, \TMP2
+	movdqu	  HashKey(%arg2), \TMP5
+	pclmulqdq $0x11, \TMP5, \TMP1      # TMP1 = a1*b1
+	movaps 0x90(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1             # Round 9
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	pclmulqdq $0x00, \TMP5, \XMM8      # XMM8 = a0*b0
+	lea	  0xa0(%arg1),%r10
+	mov	  keysize,%eax
+	shr	  $2,%eax			# 128->4, 192->6, 256->8
+	sub	  $4,%eax			# 128->0, 192->2, 256->4
+	jz	  .Laes_loop_par_enc_done\@
+
+.Laes_loop_par_enc\@:
+	MOVADQ	  (%r10),\TMP3
+.irpc	index, 1234
+	aesenc	  \TMP3, %xmm\index
+.endr
+	add	  $16,%r10
+	sub	  $1,%eax
+	jnz	  .Laes_loop_par_enc\@
+
+.Laes_loop_par_enc_done\@:
+	MOVADQ	  (%r10), \TMP3
+	aesenclast \TMP3, \XMM1           # Round 10
+	aesenclast \TMP3, \XMM2
+	aesenclast \TMP3, \XMM3
+	aesenclast \TMP3, \XMM4
+	movdqu    HashKey_k(%arg2), \TMP5
+	pclmulqdq $0x00, \TMP5, \TMP2          # TMP2 = (a1+a0)*(b1+b0)
+	movdqu	  (%arg4,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM1                 # Ciphertext/Plaintext XOR EK
+	movdqu	  16(%arg4,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM2                 # Ciphertext/Plaintext XOR EK
+	movdqu	  32(%arg4,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM3                 # Ciphertext/Plaintext XOR EK
+	movdqu	  48(%arg4,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM4                 # Ciphertext/Plaintext XOR EK
+        movdqu    \XMM1, (%arg3,%r11,1)        # Write to the ciphertext buffer
+        movdqu    \XMM2, 16(%arg3,%r11,1)      # Write to the ciphertext buffer
+        movdqu    \XMM3, 32(%arg3,%r11,1)      # Write to the ciphertext buffer
+        movdqu    \XMM4, 48(%arg3,%r11,1)      # Write to the ciphertext buffer
+	pshufb %xmm15, \XMM1        # perform a 16 byte swap
+	pshufb %xmm15, \XMM2	# perform a 16 byte swap
+	pshufb %xmm15, \XMM3	# perform a 16 byte swap
+	pshufb %xmm15, \XMM4	# perform a 16 byte swap
+
+	pxor	  \TMP4, \TMP1
+	pxor	  \XMM8, \XMM5
+	pxor	  \TMP6, \TMP2
+	pxor	  \TMP1, \TMP2
+	pxor	  \XMM5, \TMP2
+	movdqa	  \TMP2, \TMP3
+	pslldq	  $8, \TMP3                    # left shift TMP3 2 DWs
+	psrldq	  $8, \TMP2                    # right shift TMP2 2 DWs
+	pxor	  \TMP3, \XMM5
+	pxor	  \TMP2, \TMP1	  # accumulate the results in TMP1:XMM5
+
+        # first phase of reduction
+
+	movdqa    \XMM5, \TMP2
+	movdqa    \XMM5, \TMP3
+	movdqa    \XMM5, \TMP4
+# move XMM5 into TMP2, TMP3, TMP4 in order to perform shifts independently
+	pslld     $31, \TMP2                   # packed right shift << 31
+	pslld     $30, \TMP3                   # packed right shift << 30
+	pslld     $25, \TMP4                   # packed right shift << 25
+	pxor      \TMP3, \TMP2	               # xor the shifted versions
+	pxor      \TMP4, \TMP2
+	movdqa    \TMP2, \TMP5
+	psrldq    $4, \TMP5                    # right shift T5 1 DW
+	pslldq    $12, \TMP2                   # left shift T2 3 DWs
+	pxor      \TMP2, \XMM5
+
+        # second phase of reduction
+
+	movdqa    \XMM5,\TMP2 # make 3 copies of XMM5 into TMP2, TMP3, TMP4
+	movdqa    \XMM5,\TMP3
+	movdqa    \XMM5,\TMP4
+	psrld     $1, \TMP2                    # packed left shift >>1
+	psrld     $2, \TMP3                    # packed left shift >>2
+	psrld     $7, \TMP4                    # packed left shift >>7
+	pxor      \TMP3,\TMP2		       # xor the shifted versions
+	pxor      \TMP4,\TMP2
+	pxor      \TMP5, \TMP2
+	pxor      \TMP2, \XMM5
+	pxor      \TMP1, \XMM5                 # result is in TMP1
+
+	pxor	  \XMM5, \XMM1
+.endm
+
+/*
+* decrypt 4 blocks at a time
+* ghash the 4 previously decrypted ciphertext blocks
+* arg1, %arg3, %arg4 are used as pointers only, not modified
+* %r11 is the data offset value
+*/
+.macro GHASH_4_ENCRYPT_4_PARALLEL_dec TMP1 TMP2 TMP3 TMP4 TMP5 \
+TMP6 XMM0 XMM1 XMM2 XMM3 XMM4 XMM5 XMM6 XMM7 XMM8 operation
+
+	movdqa	  \XMM1, \XMM5
+	movdqa	  \XMM2, \XMM6
+	movdqa	  \XMM3, \XMM7
+	movdqa	  \XMM4, \XMM8
+
+        movdqa    SHUF_MASK(%rip), %xmm15
+        # multiply TMP5 * HashKey using karatsuba
+
+	movdqa	  \XMM5, \TMP4
+	pshufd	  $78, \XMM5, \TMP6
+	pxor	  \XMM5, \TMP6
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqu	  HashKey_4(%arg2), \TMP5
+	pclmulqdq $0x11, \TMP5, \TMP4           # TMP4 = a1*b1
+	movdqa    \XMM0, \XMM1
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqa    \XMM0, \XMM2
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqa    \XMM0, \XMM3
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqa    \XMM0, \XMM4
+	pshufb %xmm15, \XMM1	# perform a 16 byte swap
+	pclmulqdq $0x00, \TMP5, \XMM5           # XMM5 = a0*b0
+	pshufb %xmm15, \XMM2	# perform a 16 byte swap
+	pshufb %xmm15, \XMM3	# perform a 16 byte swap
+	pshufb %xmm15, \XMM4	# perform a 16 byte swap
+
+	pxor	  (%arg1), \XMM1
+	pxor	  (%arg1), \XMM2
+	pxor	  (%arg1), \XMM3
+	pxor	  (%arg1), \XMM4
+	movdqu	  HashKey_4_k(%arg2), \TMP5
+	pclmulqdq $0x00, \TMP5, \TMP6       # TMP6 = (a1+a0)*(b1+b0)
+	movaps 0x10(%arg1), \TMP1
+	aesenc	  \TMP1, \XMM1              # Round 1
+	aesenc	  \TMP1, \XMM2
+	aesenc	  \TMP1, \XMM3
+	aesenc	  \TMP1, \XMM4
+	movaps 0x20(%arg1), \TMP1
+	aesenc	  \TMP1, \XMM1              # Round 2
+	aesenc	  \TMP1, \XMM2
+	aesenc	  \TMP1, \XMM3
+	aesenc	  \TMP1, \XMM4
+	movdqa	  \XMM6, \TMP1
+	pshufd	  $78, \XMM6, \TMP2
+	pxor	  \XMM6, \TMP2
+	movdqu	  HashKey_3(%arg2), \TMP5
+	pclmulqdq $0x11, \TMP5, \TMP1       # TMP1 = a1 * b1
+	movaps 0x30(%arg1), \TMP3
+	aesenc    \TMP3, \XMM1              # Round 3
+	aesenc    \TMP3, \XMM2
+	aesenc    \TMP3, \XMM3
+	aesenc    \TMP3, \XMM4
+	pclmulqdq $0x00, \TMP5, \XMM6       # XMM6 = a0*b0
+	movaps 0x40(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1              # Round 4
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	movdqu	  HashKey_3_k(%arg2), \TMP5
+	pclmulqdq $0x00, \TMP5, \TMP2       # TMP2 = (a1+a0)*(b1+b0)
+	movaps 0x50(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1              # Round 5
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	pxor	  \TMP1, \TMP4
+# accumulate the results in TMP4:XMM5, TMP6 holds the middle part
+	pxor	  \XMM6, \XMM5
+	pxor	  \TMP2, \TMP6
+	movdqa	  \XMM7, \TMP1
+	pshufd	  $78, \XMM7, \TMP2
+	pxor	  \XMM7, \TMP2
+	movdqu	  HashKey_2(%arg2), \TMP5
+
+        # Multiply TMP5 * HashKey using karatsuba
+
+	pclmulqdq $0x11, \TMP5, \TMP1       # TMP1 = a1*b1
+	movaps 0x60(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1              # Round 6
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	pclmulqdq $0x00, \TMP5, \XMM7       # XMM7 = a0*b0
+	movaps 0x70(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1              # Round 7
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	movdqu	  HashKey_2_k(%arg2), \TMP5
+	pclmulqdq $0x00, \TMP5, \TMP2       # TMP2 = (a1+a0)*(b1+b0)
+	movaps 0x80(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1              # Round 8
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	pxor	  \TMP1, \TMP4
+# accumulate the results in TMP4:XMM5, TMP6 holds the middle part
+	pxor	  \XMM7, \XMM5
+	pxor	  \TMP2, \TMP6
+
+        # Multiply XMM8 * HashKey
+        # XMM8 and TMP5 hold the values for the two operands
+
+	movdqa	  \XMM8, \TMP1
+	pshufd	  $78, \XMM8, \TMP2
+	pxor	  \XMM8, \TMP2
+	movdqu	  HashKey(%arg2), \TMP5
+	pclmulqdq $0x11, \TMP5, \TMP1      # TMP1 = a1*b1
+	movaps 0x90(%arg1), \TMP3
+	aesenc	  \TMP3, \XMM1             # Round 9
+	aesenc	  \TMP3, \XMM2
+	aesenc	  \TMP3, \XMM3
+	aesenc	  \TMP3, \XMM4
+	pclmulqdq $0x00, \TMP5, \XMM8      # XMM8 = a0*b0
+	lea	  0xa0(%arg1),%r10
+	mov	  keysize,%eax
+	shr	  $2,%eax		        # 128->4, 192->6, 256->8
+	sub	  $4,%eax			# 128->0, 192->2, 256->4
+	jz	  .Laes_loop_par_dec_done\@
+
+.Laes_loop_par_dec\@:
+	MOVADQ	  (%r10),\TMP3
+.irpc	index, 1234
+	aesenc	  \TMP3, %xmm\index
+.endr
+	add	  $16,%r10
+	sub	  $1,%eax
+	jnz	  .Laes_loop_par_dec\@
+
+.Laes_loop_par_dec_done\@:
+	MOVADQ	  (%r10), \TMP3
+	aesenclast \TMP3, \XMM1           # last round
+	aesenclast \TMP3, \XMM2
+	aesenclast \TMP3, \XMM3
+	aesenclast \TMP3, \XMM4
+	movdqu    HashKey_k(%arg2), \TMP5
+	pclmulqdq $0x00, \TMP5, \TMP2          # TMP2 = (a1+a0)*(b1+b0)
+	movdqu	  (%arg4,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM1                 # Ciphertext/Plaintext XOR EK
+	movdqu	  \XMM1, (%arg3,%r11,1)        # Write to plaintext buffer
+	movdqa    \TMP3, \XMM1
+	movdqu	  16(%arg4,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM2                 # Ciphertext/Plaintext XOR EK
+	movdqu	  \XMM2, 16(%arg3,%r11,1)      # Write to plaintext buffer
+	movdqa    \TMP3, \XMM2
+	movdqu	  32(%arg4,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM3                 # Ciphertext/Plaintext XOR EK
+	movdqu	  \XMM3, 32(%arg3,%r11,1)      # Write to plaintext buffer
+	movdqa    \TMP3, \XMM3
+	movdqu	  48(%arg4,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM4                 # Ciphertext/Plaintext XOR EK
+	movdqu	  \XMM4, 48(%arg3,%r11,1)      # Write to plaintext buffer
+	movdqa    \TMP3, \XMM4
+	pshufb %xmm15, \XMM1        # perform a 16 byte swap
+	pshufb %xmm15, \XMM2	# perform a 16 byte swap
+	pshufb %xmm15, \XMM3	# perform a 16 byte swap
+	pshufb %xmm15, \XMM4	# perform a 16 byte swap
+
+	pxor	  \TMP4, \TMP1
+	pxor	  \XMM8, \XMM5
+	pxor	  \TMP6, \TMP2
+	pxor	  \TMP1, \TMP2
+	pxor	  \XMM5, \TMP2
+	movdqa	  \TMP2, \TMP3
+	pslldq	  $8, \TMP3                    # left shift TMP3 2 DWs
+	psrldq	  $8, \TMP2                    # right shift TMP2 2 DWs
+	pxor	  \TMP3, \XMM5
+	pxor	  \TMP2, \TMP1	  # accumulate the results in TMP1:XMM5
+
+        # first phase of reduction
+
+	movdqa    \XMM5, \TMP2
+	movdqa    \XMM5, \TMP3
+	movdqa    \XMM5, \TMP4
+# move XMM5 into TMP2, TMP3, TMP4 in order to perform shifts independently
+	pslld     $31, \TMP2                   # packed right shift << 31
+	pslld     $30, \TMP3                   # packed right shift << 30
+	pslld     $25, \TMP4                   # packed right shift << 25
+	pxor      \TMP3, \TMP2	               # xor the shifted versions
+	pxor      \TMP4, \TMP2
+	movdqa    \TMP2, \TMP5
+	psrldq    $4, \TMP5                    # right shift T5 1 DW
+	pslldq    $12, \TMP2                   # left shift T2 3 DWs
+	pxor      \TMP2, \XMM5
+
+        # second phase of reduction
+
+	movdqa    \XMM5,\TMP2 # make 3 copies of XMM5 into TMP2, TMP3, TMP4
+	movdqa    \XMM5,\TMP3
+	movdqa    \XMM5,\TMP4
+	psrld     $1, \TMP2                    # packed left shift >>1
+	psrld     $2, \TMP3                    # packed left shift >>2
+	psrld     $7, \TMP4                    # packed left shift >>7
+	pxor      \TMP3,\TMP2		       # xor the shifted versions
+	pxor      \TMP4,\TMP2
+	pxor      \TMP5, \TMP2
+	pxor      \TMP2, \XMM5
+	pxor      \TMP1, \XMM5                 # result is in TMP1
+
+	pxor	  \XMM5, \XMM1
+.endm
+
+/* GHASH the last 4 ciphertext blocks. */
+.macro	GHASH_LAST_4 TMP1 TMP2 TMP3 TMP4 TMP5 TMP6 \
+TMP7 XMM1 XMM2 XMM3 XMM4 XMMDst
+
+        # Multiply TMP6 * HashKey (using Karatsuba)
+
+	movdqa	  \XMM1, \TMP6
+	pshufd	  $78, \XMM1, \TMP2
+	pxor	  \XMM1, \TMP2
+	movdqu	  HashKey_4(%arg2), \TMP5
+	pclmulqdq $0x11, \TMP5, \TMP6       # TMP6 = a1*b1
+	pclmulqdq $0x00, \TMP5, \XMM1       # XMM1 = a0*b0
+	movdqu	  HashKey_4_k(%arg2), \TMP4
+	pclmulqdq $0x00, \TMP4, \TMP2       # TMP2 = (a1+a0)*(b1+b0)
+	movdqa	  \XMM1, \XMMDst
+	movdqa	  \TMP2, \XMM1              # result in TMP6, XMMDst, XMM1
+
+        # Multiply TMP1 * HashKey (using Karatsuba)
+
+	movdqa	  \XMM2, \TMP1
+	pshufd	  $78, \XMM2, \TMP2
+	pxor	  \XMM2, \TMP2
+	movdqu	  HashKey_3(%arg2), \TMP5
+	pclmulqdq $0x11, \TMP5, \TMP1       # TMP1 = a1*b1
+	pclmulqdq $0x00, \TMP5, \XMM2       # XMM2 = a0*b0
+	movdqu	  HashKey_3_k(%arg2), \TMP4
+	pclmulqdq $0x00, \TMP4, \TMP2       # TMP2 = (a1+a0)*(b1+b0)
+	pxor	  \TMP1, \TMP6
+	pxor	  \XMM2, \XMMDst
+	pxor	  \TMP2, \XMM1
+# results accumulated in TMP6, XMMDst, XMM1
+
+        # Multiply TMP1 * HashKey (using Karatsuba)
+
+	movdqa	  \XMM3, \TMP1
+	pshufd	  $78, \XMM3, \TMP2
+	pxor	  \XMM3, \TMP2
+	movdqu	  HashKey_2(%arg2), \TMP5
+	pclmulqdq $0x11, \TMP5, \TMP1       # TMP1 = a1*b1
+	pclmulqdq $0x00, \TMP5, \XMM3       # XMM3 = a0*b0
+	movdqu	  HashKey_2_k(%arg2), \TMP4
+	pclmulqdq $0x00, \TMP4, \TMP2       # TMP2 = (a1+a0)*(b1+b0)
+	pxor	  \TMP1, \TMP6
+	pxor	  \XMM3, \XMMDst
+	pxor	  \TMP2, \XMM1   # results accumulated in TMP6, XMMDst, XMM1
+
+        # Multiply TMP1 * HashKey (using Karatsuba)
+	movdqa	  \XMM4, \TMP1
+	pshufd	  $78, \XMM4, \TMP2
+	pxor	  \XMM4, \TMP2
+	movdqu	  HashKey(%arg2), \TMP5
+	pclmulqdq $0x11, \TMP5, \TMP1	    # TMP1 = a1*b1
+	pclmulqdq $0x00, \TMP5, \XMM4       # XMM4 = a0*b0
+	movdqu	  HashKey_k(%arg2), \TMP4
+	pclmulqdq $0x00, \TMP4, \TMP2       # TMP2 = (a1+a0)*(b1+b0)
+	pxor	  \TMP1, \TMP6
+	pxor	  \XMM4, \XMMDst
+	pxor	  \XMM1, \TMP2
+	pxor	  \TMP6, \TMP2
+	pxor	  \XMMDst, \TMP2
+	# middle section of the temp results combined as in karatsuba algorithm
+	movdqa	  \TMP2, \TMP4
+	pslldq	  $8, \TMP4                 # left shift TMP4 2 DWs
+	psrldq	  $8, \TMP2                 # right shift TMP2 2 DWs
+	pxor	  \TMP4, \XMMDst
+	pxor	  \TMP2, \TMP6
+# TMP6:XMMDst holds the result of the accumulated carry-less multiplications
+	# first phase of the reduction
+	movdqa    \XMMDst, \TMP2
+	movdqa    \XMMDst, \TMP3
+	movdqa    \XMMDst, \TMP4
+# move XMMDst into TMP2, TMP3, TMP4 in order to perform 3 shifts independently
+	pslld     $31, \TMP2                # packed right shifting << 31
+	pslld     $30, \TMP3                # packed right shifting << 30
+	pslld     $25, \TMP4                # packed right shifting << 25
+	pxor      \TMP3, \TMP2              # xor the shifted versions
+	pxor      \TMP4, \TMP2
+	movdqa    \TMP2, \TMP7
+	psrldq    $4, \TMP7                 # right shift TMP7 1 DW
+	pslldq    $12, \TMP2                # left shift TMP2 3 DWs
+	pxor      \TMP2, \XMMDst
+
+        # second phase of the reduction
+	movdqa    \XMMDst, \TMP2
+	# make 3 copies of XMMDst for doing 3 shift operations
+	movdqa    \XMMDst, \TMP3
+	movdqa    \XMMDst, \TMP4
+	psrld     $1, \TMP2                 # packed left shift >> 1
+	psrld     $2, \TMP3                 # packed left shift >> 2
+	psrld     $7, \TMP4                 # packed left shift >> 7
+	pxor      \TMP3, \TMP2              # xor the shifted versions
+	pxor      \TMP4, \TMP2
+	pxor      \TMP7, \TMP2
+	pxor      \TMP2, \XMMDst
+	pxor      \TMP6, \XMMDst            # reduced result is in XMMDst
+.endm
+
+
+/* Encryption of a single block
+* uses eax & r10
+*/
+
+.macro ENCRYPT_SINGLE_BLOCK XMM0 TMP1
+
+	pxor		(%arg1), \XMM0
+	mov		keysize,%eax
+	shr		$2,%eax			# 128->4, 192->6, 256->8
+	add		$5,%eax			# 128->9, 192->11, 256->13
+	lea		16(%arg1), %r10	  # get first expanded key address
+
+_esb_loop_\@:
+	MOVADQ		(%r10),\TMP1
+	aesenc		\TMP1,\XMM0
+	add		$16,%r10
+	sub		$1,%eax
+	jnz		_esb_loop_\@
+
+	MOVADQ		(%r10),\TMP1
+	aesenclast	\TMP1,\XMM0
+.endm
+/*****************************************************************************
+* void aesni_gcm_dec(void *aes_ctx,    // AES Key schedule. Starts on a 16 byte boundary.
+*                   struct gcm_context_data *data
+*                                      // Context data
+*                   u8 *out,           // Plaintext output. Encrypt in-place is allowed.
+*                   const u8 *in,      // Ciphertext input
+*                   u64 plaintext_len, // Length of data in bytes for decryption.
+*                   u8 *iv,            // Pre-counter block j0: 4 byte salt (from Security Association)
+*                                      // concatenated with 8 byte Initialisation Vector (from IPSec ESP Payload)
+*                                      // concatenated with 0x00000001. 16-byte aligned pointer.
+*                   u8 *hash_subkey,   // H, the Hash sub key input. Data starts on a 16-byte boundary.
+*                   const u8 *aad,     // Additional Authentication Data (AAD)
+*                   u64 aad_len,       // Length of AAD in bytes. With RFC4106 this is going to be 8 or 12 bytes
+*                   u8  *auth_tag,     // Authenticated Tag output. The driver will compare this to the
+*                                      // given authentication tag and only return the plaintext if they match.
+*                   u64 auth_tag_len); // Authenticated Tag Length in bytes. Valid values are 16
+*                                      // (most likely), 12 or 8.
+*
+* Assumptions:
+*
+* keys:
+*       keys are pre-expanded and aligned to 16 bytes. we are using the first
+*       set of 11 keys in the data structure void *aes_ctx
+*
+* iv:
+*       0                   1                   2                   3
+*       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                             Salt  (From the SA)               |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                     Initialization Vector                     |
+*       |         (This is the sequence number from IPSec header)       |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                              0x1                              |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*
+*
+*
+* AAD:
+*       AAD padded to 128 bits with 0
+*       for example, assume AAD is a u32 vector
+*
+*       if AAD is 8 bytes:
+*       AAD[3] = {A0, A1};
+*       padded AAD in xmm register = {A1 A0 0 0}
+*
+*       0                   1                   2                   3
+*       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                               SPI (A1)                        |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                     32-bit Sequence Number (A0)               |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                              0x0                              |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*
+*                                       AAD Format with 32-bit Sequence Number
+*
+*       if AAD is 12 bytes:
+*       AAD[3] = {A0, A1, A2};
+*       padded AAD in xmm register = {A2 A1 A0 0}
+*
+*       0                   1                   2                   3
+*       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                               SPI (A2)                        |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                 64-bit Extended Sequence Number {A1,A0}       |
+*       |                                                               |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                              0x0                              |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*
+*                        AAD Format with 64-bit Extended Sequence Number
+*
+* poly = x^128 + x^127 + x^126 + x^121 + 1
+*
+*****************************************************************************/
+SYM_FUNC_START(aesni_gcm_dec)
+	FUNC_SAVE
+
+	GCM_INIT %arg6, arg7, arg8, arg9
+	GCM_ENC_DEC dec
+	GCM_COMPLETE arg10, arg11
+	FUNC_RESTORE
+	RET
+SYM_FUNC_END(aesni_gcm_dec)
+
+
+/*****************************************************************************
+* void aesni_gcm_enc(void *aes_ctx,      // AES Key schedule. Starts on a 16 byte boundary.
+*                    struct gcm_context_data *data
+*                                        // Context data
+*                    u8 *out,            // Ciphertext output. Encrypt in-place is allowed.
+*                    const u8 *in,       // Plaintext input
+*                    u64 plaintext_len,  // Length of data in bytes for encryption.
+*                    u8 *iv,             // Pre-counter block j0: 4 byte salt (from Security Association)
+*                                        // concatenated with 8 byte Initialisation Vector (from IPSec ESP Payload)
+*                                        // concatenated with 0x00000001. 16-byte aligned pointer.
+*                    u8 *hash_subkey,    // H, the Hash sub key input. Data starts on a 16-byte boundary.
+*                    const u8 *aad,      // Additional Authentication Data (AAD)
+*                    u64 aad_len,        // Length of AAD in bytes. With RFC4106 this is going to be 8 or 12 bytes
+*                    u8 *auth_tag,       // Authenticated Tag output.
+*                    u64 auth_tag_len);  // Authenticated Tag Length in bytes. Valid values are 16 (most likely),
+*                                        // 12 or 8.
+*
+* Assumptions:
+*
+* keys:
+*       keys are pre-expanded and aligned to 16 bytes. we are using the
+*       first set of 11 keys in the data structure void *aes_ctx
+*
+*
+* iv:
+*       0                   1                   2                   3
+*       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                             Salt  (From the SA)               |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                     Initialization Vector                     |
+*       |         (This is the sequence number from IPSec header)       |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                              0x1                              |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*
+*
+*
+* AAD:
+*       AAD padded to 128 bits with 0
+*       for example, assume AAD is a u32 vector
+*
+*       if AAD is 8 bytes:
+*       AAD[3] = {A0, A1};
+*       padded AAD in xmm register = {A1 A0 0 0}
+*
+*       0                   1                   2                   3
+*       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                               SPI (A1)                        |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                     32-bit Sequence Number (A0)               |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                              0x0                              |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*
+*                                 AAD Format with 32-bit Sequence Number
+*
+*       if AAD is 12 bytes:
+*       AAD[3] = {A0, A1, A2};
+*       padded AAD in xmm register = {A2 A1 A0 0}
+*
+*       0                   1                   2                   3
+*       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                               SPI (A2)                        |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                 64-bit Extended Sequence Number {A1,A0}       |
+*       |                                                               |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*       |                              0x0                              |
+*       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+*
+*                         AAD Format with 64-bit Extended Sequence Number
+*
+* poly = x^128 + x^127 + x^126 + x^121 + 1
+***************************************************************************/
+SYM_FUNC_START(aesni_gcm_enc)
+	FUNC_SAVE
+
+	GCM_INIT %arg6, arg7, arg8, arg9
+	GCM_ENC_DEC enc
+
+	GCM_COMPLETE arg10, arg11
+	FUNC_RESTORE
+	RET
+SYM_FUNC_END(aesni_gcm_enc)
+
+/*****************************************************************************
+* void aesni_gcm_init(void *aes_ctx,      // AES Key schedule. Starts on a 16 byte boundary.
+*                     struct gcm_context_data *data,
+*                                         // context data
+*                     u8 *iv,             // Pre-counter block j0: 4 byte salt (from Security Association)
+*                                         // concatenated with 8 byte Initialisation Vector (from IPSec ESP Payload)
+*                                         // concatenated with 0x00000001. 16-byte aligned pointer.
+*                     u8 *hash_subkey,    // H, the Hash sub key input. Data starts on a 16-byte boundary.
+*                     const u8 *aad,      // Additional Authentication Data (AAD)
+*                     u64 aad_len)        // Length of AAD in bytes.
+*/
+SYM_FUNC_START(aesni_gcm_init)
+	FUNC_SAVE
+	GCM_INIT %arg3, %arg4,%arg5, %arg6
+	FUNC_RESTORE
+	RET
+SYM_FUNC_END(aesni_gcm_init)
+
+/*****************************************************************************
+* void aesni_gcm_enc_update(void *aes_ctx,      // AES Key schedule. Starts on a 16 byte boundary.
+*                    struct gcm_context_data *data,
+*                                        // context data
+*                    u8 *out,            // Ciphertext output. Encrypt in-place is allowed.
+*                    const u8 *in,       // Plaintext input
+*                    u64 plaintext_len,  // Length of data in bytes for encryption.
+*/
+SYM_FUNC_START(aesni_gcm_enc_update)
+	FUNC_SAVE
+	GCM_ENC_DEC enc
+	FUNC_RESTORE
+	RET
+SYM_FUNC_END(aesni_gcm_enc_update)
+
+/*****************************************************************************
+* void aesni_gcm_dec_update(void *aes_ctx,      // AES Key schedule. Starts on a 16 byte boundary.
+*                    struct gcm_context_data *data,
+*                                        // context data
+*                    u8 *out,            // Ciphertext output. Encrypt in-place is allowed.
+*                    const u8 *in,       // Plaintext input
+*                    u64 plaintext_len,  // Length of data in bytes for encryption.
+*/
+SYM_FUNC_START(aesni_gcm_dec_update)
+	FUNC_SAVE
+	GCM_ENC_DEC dec
+	FUNC_RESTORE
+	RET
+SYM_FUNC_END(aesni_gcm_dec_update)
+
+/*****************************************************************************
+* void aesni_gcm_finalize(void *aes_ctx,      // AES Key schedule. Starts on a 16 byte boundary.
+*                    struct gcm_context_data *data,
+*                                        // context data
+*                    u8 *auth_tag,       // Authenticated Tag output.
+*                    u64 auth_tag_len);  // Authenticated Tag Length in bytes. Valid values are 16 (most likely),
+*                                        // 12 or 8.
+*/
+SYM_FUNC_START(aesni_gcm_finalize)
+	FUNC_SAVE
+	GCM_COMPLETE %arg3 %arg4
+	FUNC_RESTORE
+	RET
+SYM_FUNC_END(aesni_gcm_finalize)
+
+#endif
+
+SYM_FUNC_START_LOCAL(_key_expansion_256a)
+	pshufd $0b11111111, %xmm1, %xmm1
+	shufps $0b00010000, %xmm0, %xmm4
+	pxor %xmm4, %xmm0
+	shufps $0b10001100, %xmm0, %xmm4
+	pxor %xmm4, %xmm0
+	pxor %xmm1, %xmm0
+	movaps %xmm0, (TKEYP)
+	add $0x10, TKEYP
+	RET
+SYM_FUNC_END(_key_expansion_256a)
+SYM_FUNC_ALIAS_LOCAL(_key_expansion_128, _key_expansion_256a)
+
+SYM_FUNC_START_LOCAL(_key_expansion_192a)
+	pshufd $0b01010101, %xmm1, %xmm1
+	shufps $0b00010000, %xmm0, %xmm4
+	pxor %xmm4, %xmm0
+	shufps $0b10001100, %xmm0, %xmm4
+	pxor %xmm4, %xmm0
+	pxor %xmm1, %xmm0
+
+	movaps %xmm2, %xmm5
+	movaps %xmm2, %xmm6
+	pslldq $4, %xmm5
+	pshufd $0b11111111, %xmm0, %xmm3
+	pxor %xmm3, %xmm2
+	pxor %xmm5, %xmm2
+
+	movaps %xmm0, %xmm1
+	shufps $0b01000100, %xmm0, %xmm6
+	movaps %xmm6, (TKEYP)
+	shufps $0b01001110, %xmm2, %xmm1
+	movaps %xmm1, 0x10(TKEYP)
+	add $0x20, TKEYP
+	RET
+SYM_FUNC_END(_key_expansion_192a)
+
+SYM_FUNC_START_LOCAL(_key_expansion_192b)
+	pshufd $0b01010101, %xmm1, %xmm1
+	shufps $0b00010000, %xmm0, %xmm4
+	pxor %xmm4, %xmm0
+	shufps $0b10001100, %xmm0, %xmm4
+	pxor %xmm4, %xmm0
+	pxor %xmm1, %xmm0
+
+	movaps %xmm2, %xmm5
+	pslldq $4, %xmm5
+	pshufd $0b11111111, %xmm0, %xmm3
+	pxor %xmm3, %xmm2
+	pxor %xmm5, %xmm2
+
+	movaps %xmm0, (TKEYP)
+	add $0x10, TKEYP
+	RET
+SYM_FUNC_END(_key_expansion_192b)
+
+SYM_FUNC_START_LOCAL(_key_expansion_256b)
+	pshufd $0b10101010, %xmm1, %xmm1
+	shufps $0b00010000, %xmm2, %xmm4
+	pxor %xmm4, %xmm2
+	shufps $0b10001100, %xmm2, %xmm4
+	pxor %xmm4, %xmm2
+	pxor %xmm1, %xmm2
+	movaps %xmm2, (TKEYP)
+	add $0x10, TKEYP
+	RET
+SYM_FUNC_END(_key_expansion_256b)
+
+/*
+ * int aesni_set_key(struct crypto_aes_ctx *ctx, const u8 *in_key,
+ *                   unsigned int key_len)
+ */
+SYM_FUNC_START(aesni_set_key)
+	FRAME_BEGIN
+#ifndef __x86_64__
+	pushl KEYP
+	movl (FRAME_OFFSET+8)(%esp), KEYP	# ctx
+	movl (FRAME_OFFSET+12)(%esp), UKEYP	# in_key
+	movl (FRAME_OFFSET+16)(%esp), %edx	# key_len
+#endif
+	movups (UKEYP), %xmm0		# user key (first 16 bytes)
+	movaps %xmm0, (KEYP)
+	lea 0x10(KEYP), TKEYP		# key addr
+	movl %edx, 480(KEYP)
+	pxor %xmm4, %xmm4		# xmm4 is assumed 0 in _key_expansion_x
+	cmp $24, %dl
+	jb .Lenc_key128
+	je .Lenc_key192
+	movups 0x10(UKEYP), %xmm2	# other user key
+	movaps %xmm2, (TKEYP)
+	add $0x10, TKEYP
+	aeskeygenassist $0x1, %xmm2, %xmm1	# round 1
+	call _key_expansion_256a
+	aeskeygenassist $0x1, %xmm0, %xmm1
+	call _key_expansion_256b
+	aeskeygenassist $0x2, %xmm2, %xmm1	# round 2
+	call _key_expansion_256a
+	aeskeygenassist $0x2, %xmm0, %xmm1
+	call _key_expansion_256b
+	aeskeygenassist $0x4, %xmm2, %xmm1	# round 3
+	call _key_expansion_256a
+	aeskeygenassist $0x4, %xmm0, %xmm1
+	call _key_expansion_256b
+	aeskeygenassist $0x8, %xmm2, %xmm1	# round 4
+	call _key_expansion_256a
+	aeskeygenassist $0x8, %xmm0, %xmm1
+	call _key_expansion_256b
+	aeskeygenassist $0x10, %xmm2, %xmm1	# round 5
+	call _key_expansion_256a
+	aeskeygenassist $0x10, %xmm0, %xmm1
+	call _key_expansion_256b
+	aeskeygenassist $0x20, %xmm2, %xmm1	# round 6
+	call _key_expansion_256a
+	aeskeygenassist $0x20, %xmm0, %xmm1
+	call _key_expansion_256b
+	aeskeygenassist $0x40, %xmm2, %xmm1	# round 7
+	call _key_expansion_256a
+	jmp .Ldec_key
+.Lenc_key192:
+	movq 0x10(UKEYP), %xmm2		# other user key
+	aeskeygenassist $0x1, %xmm2, %xmm1	# round 1
+	call _key_expansion_192a
+	aeskeygenassist $0x2, %xmm2, %xmm1	# round 2
+	call _key_expansion_192b
+	aeskeygenassist $0x4, %xmm2, %xmm1	# round 3
+	call _key_expansion_192a
+	aeskeygenassist $0x8, %xmm2, %xmm1	# round 4
+	call _key_expansion_192b
+	aeskeygenassist $0x10, %xmm2, %xmm1	# round 5
+	call _key_expansion_192a
+	aeskeygenassist $0x20, %xmm2, %xmm1	# round 6
+	call _key_expansion_192b
+	aeskeygenassist $0x40, %xmm2, %xmm1	# round 7
+	call _key_expansion_192a
+	aeskeygenassist $0x80, %xmm2, %xmm1	# round 8
+	call _key_expansion_192b
+	jmp .Ldec_key
+.Lenc_key128:
+	aeskeygenassist $0x1, %xmm0, %xmm1	# round 1
+	call _key_expansion_128
+	aeskeygenassist $0x2, %xmm0, %xmm1	# round 2
+	call _key_expansion_128
+	aeskeygenassist $0x4, %xmm0, %xmm1	# round 3
+	call _key_expansion_128
+	aeskeygenassist $0x8, %xmm0, %xmm1	# round 4
+	call _key_expansion_128
+	aeskeygenassist $0x10, %xmm0, %xmm1	# round 5
+	call _key_expansion_128
+	aeskeygenassist $0x20, %xmm0, %xmm1	# round 6
+	call _key_expansion_128
+	aeskeygenassist $0x40, %xmm0, %xmm1	# round 7
+	call _key_expansion_128
+	aeskeygenassist $0x80, %xmm0, %xmm1	# round 8
+	call _key_expansion_128
+	aeskeygenassist $0x1b, %xmm0, %xmm1	# round 9
+	call _key_expansion_128
+	aeskeygenassist $0x36, %xmm0, %xmm1	# round 10
+	call _key_expansion_128
+.Ldec_key:
+	sub $0x10, TKEYP
+	movaps (KEYP), %xmm0
+	movaps (TKEYP), %xmm1
+	movaps %xmm0, 240(TKEYP)
+	movaps %xmm1, 240(KEYP)
+	add $0x10, KEYP
+	lea 240-16(TKEYP), UKEYP
+.align 4
+.Ldec_key_loop:
+	movaps (KEYP), %xmm0
+	aesimc %xmm0, %xmm1
+	movaps %xmm1, (UKEYP)
+	add $0x10, KEYP
+	sub $0x10, UKEYP
+	cmp TKEYP, KEYP
+	jb .Ldec_key_loop
+	xor AREG, AREG
+#ifndef __x86_64__
+	popl KEYP
+#endif
+	FRAME_END
+	RET
+SYM_FUNC_END(aesni_set_key)
+
+/*
+ * void aesni_enc(const void *ctx, u8 *dst, const u8 *src)
+ */
+SYM_FUNC_START(aesni_enc)
+	FRAME_BEGIN
+#ifndef __x86_64__
+	pushl KEYP
+	pushl KLEN
+	movl (FRAME_OFFSET+12)(%esp), KEYP	# ctx
+	movl (FRAME_OFFSET+16)(%esp), OUTP	# dst
+	movl (FRAME_OFFSET+20)(%esp), INP	# src
+#endif
+	movl 480(KEYP), KLEN		# key length
+	movups (INP), STATE		# input
+	call _aesni_enc1
+	movups STATE, (OUTP)		# output
+#ifndef __x86_64__
+	popl KLEN
+	popl KEYP
+#endif
+	FRAME_END
+	RET
+SYM_FUNC_END(aesni_enc)
+
+/*
+ * _aesni_enc1:		internal ABI
+ * input:
+ *	KEYP:		key struct pointer
+ *	KLEN:		round count
+ *	STATE:		initial state (input)
+ * output:
+ *	STATE:		finial state (output)
+ * changed:
+ *	KEY
+ *	TKEYP (T1)
+ */
+SYM_FUNC_START_LOCAL(_aesni_enc1)
+	movaps (KEYP), KEY		# key
+	mov KEYP, TKEYP
+	pxor KEY, STATE		# round 0
+	add $0x30, TKEYP
+	cmp $24, KLEN
+	jb .Lenc128
+	lea 0x20(TKEYP), TKEYP
+	je .Lenc192
+	add $0x20, TKEYP
+	movaps -0x60(TKEYP), KEY
+	aesenc KEY, STATE
+	movaps -0x50(TKEYP), KEY
+	aesenc KEY, STATE
+.align 4
+.Lenc192:
+	movaps -0x40(TKEYP), KEY
+	aesenc KEY, STATE
+	movaps -0x30(TKEYP), KEY
+	aesenc KEY, STATE
+.align 4
+.Lenc128:
+	movaps -0x20(TKEYP), KEY
+	aesenc KEY, STATE
+	movaps -0x10(TKEYP), KEY
+	aesenc KEY, STATE
+	movaps (TKEYP), KEY
+	aesenc KEY, STATE
+	movaps 0x10(TKEYP), KEY
+	aesenc KEY, STATE
+	movaps 0x20(TKEYP), KEY
+	aesenc KEY, STATE
+	movaps 0x30(TKEYP), KEY
+	aesenc KEY, STATE
+	movaps 0x40(TKEYP), KEY
+	aesenc KEY, STATE
+	movaps 0x50(TKEYP), KEY
+	aesenc KEY, STATE
+	movaps 0x60(TKEYP), KEY
+	aesenc KEY, STATE
+	movaps 0x70(TKEYP), KEY
+	aesenclast KEY, STATE
+	RET
+SYM_FUNC_END(_aesni_enc1)
+
+/*
+ * _aesni_enc4:	internal ABI
+ * input:
+ *	KEYP:		key struct pointer
+ *	KLEN:		round count
+ *	STATE1:		initial state (input)
+ *	STATE2
+ *	STATE3
+ *	STATE4
+ * output:
+ *	STATE1:		finial state (output)
+ *	STATE2
+ *	STATE3
+ *	STATE4
+ * changed:
+ *	KEY
+ *	TKEYP (T1)
+ */
+SYM_FUNC_START_LOCAL(_aesni_enc4)
+	movaps (KEYP), KEY		# key
+	mov KEYP, TKEYP
+	pxor KEY, STATE1		# round 0
+	pxor KEY, STATE2
+	pxor KEY, STATE3
+	pxor KEY, STATE4
+	add $0x30, TKEYP
+	cmp $24, KLEN
+	jb .L4enc128
+	lea 0x20(TKEYP), TKEYP
+	je .L4enc192
+	add $0x20, TKEYP
+	movaps -0x60(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+	movaps -0x50(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+#.align 4
+.L4enc192:
+	movaps -0x40(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+	movaps -0x30(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+#.align 4
+.L4enc128:
+	movaps -0x20(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+	movaps -0x10(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+	movaps (TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+	movaps 0x10(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+	movaps 0x20(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+	movaps 0x30(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+	movaps 0x40(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+	movaps 0x50(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+	movaps 0x60(TKEYP), KEY
+	aesenc KEY, STATE1
+	aesenc KEY, STATE2
+	aesenc KEY, STATE3
+	aesenc KEY, STATE4
+	movaps 0x70(TKEYP), KEY
+	aesenclast KEY, STATE1		# last round
+	aesenclast KEY, STATE2
+	aesenclast KEY, STATE3
+	aesenclast KEY, STATE4
+	RET
+SYM_FUNC_END(_aesni_enc4)
+
+/*
+ * void aesni_dec (const void *ctx, u8 *dst, const u8 *src)
+ */
+SYM_FUNC_START(aesni_dec)
+	FRAME_BEGIN
+#ifndef __x86_64__
+	pushl KEYP
+	pushl KLEN
+	movl (FRAME_OFFSET+12)(%esp), KEYP	# ctx
+	movl (FRAME_OFFSET+16)(%esp), OUTP	# dst
+	movl (FRAME_OFFSET+20)(%esp), INP	# src
+#endif
+	mov 480(KEYP), KLEN		# key length
+	add $240, KEYP
+	movups (INP), STATE		# input
+	call _aesni_dec1
+	movups STATE, (OUTP)		#output
+#ifndef __x86_64__
+	popl KLEN
+	popl KEYP
+#endif
+	FRAME_END
+	RET
+SYM_FUNC_END(aesni_dec)
+
+/*
+ * _aesni_dec1:		internal ABI
+ * input:
+ *	KEYP:		key struct pointer
+ *	KLEN:		key length
+ *	STATE:		initial state (input)
+ * output:
+ *	STATE:		finial state (output)
+ * changed:
+ *	KEY
+ *	TKEYP (T1)
+ */
+SYM_FUNC_START_LOCAL(_aesni_dec1)
+	movaps (KEYP), KEY		# key
+	mov KEYP, TKEYP
+	pxor KEY, STATE		# round 0
+	add $0x30, TKEYP
+	cmp $24, KLEN
+	jb .Ldec128
+	lea 0x20(TKEYP), TKEYP
+	je .Ldec192
+	add $0x20, TKEYP
+	movaps -0x60(TKEYP), KEY
+	aesdec KEY, STATE
+	movaps -0x50(TKEYP), KEY
+	aesdec KEY, STATE
+.align 4
+.Ldec192:
+	movaps -0x40(TKEYP), KEY
+	aesdec KEY, STATE
+	movaps -0x30(TKEYP), KEY
+	aesdec KEY, STATE
+.align 4
+.Ldec128:
+	movaps -0x20(TKEYP), KEY
+	aesdec KEY, STATE
+	movaps -0x10(TKEYP), KEY
+	aesdec KEY, STATE
+	movaps (TKEYP), KEY
+	aesdec KEY, STATE
+	movaps 0x10(TKEYP), KEY
+	aesdec KEY, STATE
+	movaps 0x20(TKEYP), KEY
+	aesdec KEY, STATE
+	movaps 0x30(TKEYP), KEY
+	aesdec KEY, STATE
+	movaps 0x40(TKEYP), KEY
+	aesdec KEY, STATE
+	movaps 0x50(TKEYP), KEY
+	aesdec KEY, STATE
+	movaps 0x60(TKEYP), KEY
+	aesdec KEY, STATE
+	movaps 0x70(TKEYP), KEY
+	aesdeclast KEY, STATE
+	RET
+SYM_FUNC_END(_aesni_dec1)
+
+/*
+ * _aesni_dec4:	internal ABI
+ * input:
+ *	KEYP:		key struct pointer
+ *	KLEN:		key length
+ *	STATE1:		initial state (input)
+ *	STATE2
+ *	STATE3
+ *	STATE4
+ * output:
+ *	STATE1:		finial state (output)
+ *	STATE2
+ *	STATE3
+ *	STATE4
+ * changed:
+ *	KEY
+ *	TKEYP (T1)
+ */
+SYM_FUNC_START_LOCAL(_aesni_dec4)
+	movaps (KEYP), KEY		# key
+	mov KEYP, TKEYP
+	pxor KEY, STATE1		# round 0
+	pxor KEY, STATE2
+	pxor KEY, STATE3
+	pxor KEY, STATE4
+	add $0x30, TKEYP
+	cmp $24, KLEN
+	jb .L4dec128
+	lea 0x20(TKEYP), TKEYP
+	je .L4dec192
+	add $0x20, TKEYP
+	movaps -0x60(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+	movaps -0x50(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+.align 4
+.L4dec192:
+	movaps -0x40(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+	movaps -0x30(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+.align 4
+.L4dec128:
+	movaps -0x20(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+	movaps -0x10(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+	movaps (TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+	movaps 0x10(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+	movaps 0x20(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+	movaps 0x30(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+	movaps 0x40(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+	movaps 0x50(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+	movaps 0x60(TKEYP), KEY
+	aesdec KEY, STATE1
+	aesdec KEY, STATE2
+	aesdec KEY, STATE3
+	aesdec KEY, STATE4
+	movaps 0x70(TKEYP), KEY
+	aesdeclast KEY, STATE1		# last round
+	aesdeclast KEY, STATE2
+	aesdeclast KEY, STATE3
+	aesdeclast KEY, STATE4
+	RET
+SYM_FUNC_END(_aesni_dec4)
+
+/*
+ * void aesni_ecb_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
+ *		      size_t len)
+ */
+SYM_FUNC_START(aesni_ecb_enc)
+	FRAME_BEGIN
+#ifndef __x86_64__
+	pushl LEN
+	pushl KEYP
+	pushl KLEN
+	movl (FRAME_OFFSET+16)(%esp), KEYP	# ctx
+	movl (FRAME_OFFSET+20)(%esp), OUTP	# dst
+	movl (FRAME_OFFSET+24)(%esp), INP	# src
+	movl (FRAME_OFFSET+28)(%esp), LEN	# len
+#endif
+	test LEN, LEN		# check length
+	jz .Lecb_enc_ret
+	mov 480(KEYP), KLEN
+	cmp $16, LEN
+	jb .Lecb_enc_ret
+	cmp $64, LEN
+	jb .Lecb_enc_loop1
+.align 4
+.Lecb_enc_loop4:
+	movups (INP), STATE1
+	movups 0x10(INP), STATE2
+	movups 0x20(INP), STATE3
+	movups 0x30(INP), STATE4
+	call _aesni_enc4
+	movups STATE1, (OUTP)
+	movups STATE2, 0x10(OUTP)
+	movups STATE3, 0x20(OUTP)
+	movups STATE4, 0x30(OUTP)
+	sub $64, LEN
+	add $64, INP
+	add $64, OUTP
+	cmp $64, LEN
+	jge .Lecb_enc_loop4
+	cmp $16, LEN
+	jb .Lecb_enc_ret
+.align 4
+.Lecb_enc_loop1:
+	movups (INP), STATE1
+	call _aesni_enc1
+	movups STATE1, (OUTP)
+	sub $16, LEN
+	add $16, INP
+	add $16, OUTP
+	cmp $16, LEN
+	jge .Lecb_enc_loop1
+.Lecb_enc_ret:
+#ifndef __x86_64__
+	popl KLEN
+	popl KEYP
+	popl LEN
+#endif
+	FRAME_END
+	RET
+SYM_FUNC_END(aesni_ecb_enc)
+
+/*
+ * void aesni_ecb_dec(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
+ *		      size_t len);
+ */
+SYM_FUNC_START(aesni_ecb_dec)
+	FRAME_BEGIN
+#ifndef __x86_64__
+	pushl LEN
+	pushl KEYP
+	pushl KLEN
+	movl (FRAME_OFFSET+16)(%esp), KEYP	# ctx
+	movl (FRAME_OFFSET+20)(%esp), OUTP	# dst
+	movl (FRAME_OFFSET+24)(%esp), INP	# src
+	movl (FRAME_OFFSET+28)(%esp), LEN	# len
+#endif
+	test LEN, LEN
+	jz .Lecb_dec_ret
+	mov 480(KEYP), KLEN
+	add $240, KEYP
+	cmp $16, LEN
+	jb .Lecb_dec_ret
+	cmp $64, LEN
+	jb .Lecb_dec_loop1
+.align 4
+.Lecb_dec_loop4:
+	movups (INP), STATE1
+	movups 0x10(INP), STATE2
+	movups 0x20(INP), STATE3
+	movups 0x30(INP), STATE4
+	call _aesni_dec4
+	movups STATE1, (OUTP)
+	movups STATE2, 0x10(OUTP)
+	movups STATE3, 0x20(OUTP)
+	movups STATE4, 0x30(OUTP)
+	sub $64, LEN
+	add $64, INP
+	add $64, OUTP
+	cmp $64, LEN
+	jge .Lecb_dec_loop4
+	cmp $16, LEN
+	jb .Lecb_dec_ret
+.align 4
+.Lecb_dec_loop1:
+	movups (INP), STATE1
+	call _aesni_dec1
+	movups STATE1, (OUTP)
+	sub $16, LEN
+	add $16, INP
+	add $16, OUTP
+	cmp $16, LEN
+	jge .Lecb_dec_loop1
+.Lecb_dec_ret:
+#ifndef __x86_64__
+	popl KLEN
+	popl KEYP
+	popl LEN
+#endif
+	FRAME_END
+	RET
+SYM_FUNC_END(aesni_ecb_dec)
+
+/*
+ * void aesni_cbc_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
+ *		      size_t len, u8 *iv)
+ */
+SYM_FUNC_START(aesni_cbc_enc)
+	FRAME_BEGIN
+#ifndef __x86_64__
+	pushl IVP
+	pushl LEN
+	pushl KEYP
+	pushl KLEN
+	movl (FRAME_OFFSET+20)(%esp), KEYP	# ctx
+	movl (FRAME_OFFSET+24)(%esp), OUTP	# dst
+	movl (FRAME_OFFSET+28)(%esp), INP	# src
+	movl (FRAME_OFFSET+32)(%esp), LEN	# len
+	movl (FRAME_OFFSET+36)(%esp), IVP	# iv
+#endif
+	cmp $16, LEN
+	jb .Lcbc_enc_ret
+	mov 480(KEYP), KLEN
+	movups (IVP), STATE	# load iv as initial state
+.align 4
+.Lcbc_enc_loop:
+	movups (INP), IN	# load input
+	pxor IN, STATE
+	call _aesni_enc1
+	movups STATE, (OUTP)	# store output
+	sub $16, LEN
+	add $16, INP
+	add $16, OUTP
+	cmp $16, LEN
+	jge .Lcbc_enc_loop
+	movups STATE, (IVP)
+.Lcbc_enc_ret:
+#ifndef __x86_64__
+	popl KLEN
+	popl KEYP
+	popl LEN
+	popl IVP
+#endif
+	FRAME_END
+	RET
+SYM_FUNC_END(aesni_cbc_enc)
+
+/*
+ * void aesni_cbc_dec(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
+ *		      size_t len, u8 *iv)
+ */
+SYM_FUNC_START(aesni_cbc_dec)
+	FRAME_BEGIN
+#ifndef __x86_64__
+	pushl IVP
+	pushl LEN
+	pushl KEYP
+	pushl KLEN
+	movl (FRAME_OFFSET+20)(%esp), KEYP	# ctx
+	movl (FRAME_OFFSET+24)(%esp), OUTP	# dst
+	movl (FRAME_OFFSET+28)(%esp), INP	# src
+	movl (FRAME_OFFSET+32)(%esp), LEN	# len
+	movl (FRAME_OFFSET+36)(%esp), IVP	# iv
+#endif
+	cmp $16, LEN
+	jb .Lcbc_dec_just_ret
+	mov 480(KEYP), KLEN
+	add $240, KEYP
+	movups (IVP), IV
+	cmp $64, LEN
+	jb .Lcbc_dec_loop1
+.align 4
+.Lcbc_dec_loop4:
+	movups (INP), IN1
+	movaps IN1, STATE1
+	movups 0x10(INP), IN2
+	movaps IN2, STATE2
+#ifdef __x86_64__
+	movups 0x20(INP), IN3
+	movaps IN3, STATE3
+	movups 0x30(INP), IN4
+	movaps IN4, STATE4
+#else
+	movups 0x20(INP), IN1
+	movaps IN1, STATE3
+	movups 0x30(INP), IN2
+	movaps IN2, STATE4
+#endif
+	call _aesni_dec4
+	pxor IV, STATE1
+#ifdef __x86_64__
+	pxor IN1, STATE2
+	pxor IN2, STATE3
+	pxor IN3, STATE4
+	movaps IN4, IV
+#else
+	pxor IN1, STATE4
+	movaps IN2, IV
+	movups (INP), IN1
+	pxor IN1, STATE2
+	movups 0x10(INP), IN2
+	pxor IN2, STATE3
+#endif
+	movups STATE1, (OUTP)
+	movups STATE2, 0x10(OUTP)
+	movups STATE3, 0x20(OUTP)
+	movups STATE4, 0x30(OUTP)
+	sub $64, LEN
+	add $64, INP
+	add $64, OUTP
+	cmp $64, LEN
+	jge .Lcbc_dec_loop4
+	cmp $16, LEN
+	jb .Lcbc_dec_ret
+.align 4
+.Lcbc_dec_loop1:
+	movups (INP), IN
+	movaps IN, STATE
+	call _aesni_dec1
+	pxor IV, STATE
+	movups STATE, (OUTP)
+	movaps IN, IV
+	sub $16, LEN
+	add $16, INP
+	add $16, OUTP
+	cmp $16, LEN
+	jge .Lcbc_dec_loop1
+.Lcbc_dec_ret:
+	movups IV, (IVP)
+.Lcbc_dec_just_ret:
+#ifndef __x86_64__
+	popl KLEN
+	popl KEYP
+	popl LEN
+	popl IVP
+#endif
+	FRAME_END
+	RET
+SYM_FUNC_END(aesni_cbc_dec)
+
+/*
+ * void aesni_cts_cbc_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
+ *			  size_t len, u8 *iv)
+ */
+SYM_FUNC_START(aesni_cts_cbc_enc)
+	FRAME_BEGIN
+#ifndef __x86_64__
+	pushl IVP
+	pushl LEN
+	pushl KEYP
+	pushl KLEN
+	movl (FRAME_OFFSET+20)(%esp), KEYP	# ctx
+	movl (FRAME_OFFSET+24)(%esp), OUTP	# dst
+	movl (FRAME_OFFSET+28)(%esp), INP	# src
+	movl (FRAME_OFFSET+32)(%esp), LEN	# len
+	movl (FRAME_OFFSET+36)(%esp), IVP	# iv
+	lea .Lcts_permute_table, T1
+#else
+	lea .Lcts_permute_table(%rip), T1
+#endif
+	mov 480(KEYP), KLEN
+	movups (IVP), STATE
+	sub $16, LEN
+	mov T1, IVP
+	add $32, IVP
+	add LEN, T1
+	sub LEN, IVP
+	movups (T1), %xmm4
+	movups (IVP), %xmm5
+
+	movups (INP), IN1
+	add LEN, INP
+	movups (INP), IN2
+
+	pxor IN1, STATE
+	call _aesni_enc1
+
+	pshufb %xmm5, IN2
+	pxor STATE, IN2
+	pshufb %xmm4, STATE
+	add OUTP, LEN
+	movups STATE, (LEN)
+
+	movaps IN2, STATE
+	call _aesni_enc1
+	movups STATE, (OUTP)
+
+#ifndef __x86_64__
+	popl KLEN
+	popl KEYP
+	popl LEN
+	popl IVP
+#endif
+	FRAME_END
+	RET
+SYM_FUNC_END(aesni_cts_cbc_enc)
+
+/*
+ * void aesni_cts_cbc_dec(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
+ *			  size_t len, u8 *iv)
+ */
+SYM_FUNC_START(aesni_cts_cbc_dec)
+	FRAME_BEGIN
+#ifndef __x86_64__
+	pushl IVP
+	pushl LEN
+	pushl KEYP
+	pushl KLEN
+	movl (FRAME_OFFSET+20)(%esp), KEYP	# ctx
+	movl (FRAME_OFFSET+24)(%esp), OUTP	# dst
+	movl (FRAME_OFFSET+28)(%esp), INP	# src
+	movl (FRAME_OFFSET+32)(%esp), LEN	# len
+	movl (FRAME_OFFSET+36)(%esp), IVP	# iv
+	lea .Lcts_permute_table, T1
+#else
+	lea .Lcts_permute_table(%rip), T1
+#endif
+	mov 480(KEYP), KLEN
+	add $240, KEYP
+	movups (IVP), IV
+	sub $16, LEN
+	mov T1, IVP
+	add $32, IVP
+	add LEN, T1
+	sub LEN, IVP
+	movups (T1), %xmm4
+
+	movups (INP), STATE
+	add LEN, INP
+	movups (INP), IN1
+
+	call _aesni_dec1
+	movaps STATE, IN2
+	pshufb %xmm4, STATE
+	pxor IN1, STATE
+
+	add OUTP, LEN
+	movups STATE, (LEN)
+
+	movups (IVP), %xmm0
+	pshufb %xmm0, IN1
+	pblendvb IN2, IN1
+	movaps IN1, STATE
+	call _aesni_dec1
+
+	pxor IV, STATE
+	movups STATE, (OUTP)
+
+#ifndef __x86_64__
+	popl KLEN
+	popl KEYP
+	popl LEN
+	popl IVP
+#endif
+	FRAME_END
+	RET
+SYM_FUNC_END(aesni_cts_cbc_dec)
+
+.pushsection .rodata
+.align 16
+.Lcts_permute_table:
+	.byte		0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80
+	.byte		0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80
+	.byte		0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07
+	.byte		0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f
+	.byte		0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80
+	.byte		0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80
+#ifdef __x86_64__
+.Lbswap_mask:
+	.byte 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0
+#endif
+.popsection
+
+#ifdef __x86_64__
+/*
+ * _aesni_inc_init:	internal ABI
+ *	setup registers used by _aesni_inc
+ * input:
+ *	IV
+ * output:
+ *	CTR:	== IV, in little endian
+ *	TCTR_LOW: == lower qword of CTR
+ *	INC:	== 1, in little endian
+ *	BSWAP_MASK == endian swapping mask
+ */
+SYM_FUNC_START_LOCAL(_aesni_inc_init)
+	movaps .Lbswap_mask(%rip), BSWAP_MASK
+	movaps IV, CTR
+	pshufb BSWAP_MASK, CTR
+	mov $1, TCTR_LOW
+	movq TCTR_LOW, INC
+	movq CTR, TCTR_LOW
+	RET
+SYM_FUNC_END(_aesni_inc_init)
+
+/*
+ * _aesni_inc:		internal ABI
+ *	Increase IV by 1, IV is in big endian
+ * input:
+ *	IV
+ *	CTR:	== IV, in little endian
+ *	TCTR_LOW: == lower qword of CTR
+ *	INC:	== 1, in little endian
+ *	BSWAP_MASK == endian swapping mask
+ * output:
+ *	IV:	Increase by 1
+ * changed:
+ *	CTR:	== output IV, in little endian
+ *	TCTR_LOW: == lower qword of CTR
+ */
+SYM_FUNC_START_LOCAL(_aesni_inc)
+	paddq INC, CTR
+	add $1, TCTR_LOW
+	jnc .Linc_low
+	pslldq $8, INC
+	paddq INC, CTR
+	psrldq $8, INC
+.Linc_low:
+	movaps CTR, IV
+	pshufb BSWAP_MASK, IV
+	RET
+SYM_FUNC_END(_aesni_inc)
+
+/*
+ * void aesni_ctr_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
+ *		      size_t len, u8 *iv)
+ */
+SYM_FUNC_START(aesni_ctr_enc)
+	FRAME_BEGIN
+	cmp $16, LEN
+	jb .Lctr_enc_just_ret
+	mov 480(KEYP), KLEN
+	movups (IVP), IV
+	call _aesni_inc_init
+	cmp $64, LEN
+	jb .Lctr_enc_loop1
+.align 4
+.Lctr_enc_loop4:
+	movaps IV, STATE1
+	call _aesni_inc
+	movups (INP), IN1
+	movaps IV, STATE2
+	call _aesni_inc
+	movups 0x10(INP), IN2
+	movaps IV, STATE3
+	call _aesni_inc
+	movups 0x20(INP), IN3
+	movaps IV, STATE4
+	call _aesni_inc
+	movups 0x30(INP), IN4
+	call _aesni_enc4
+	pxor IN1, STATE1
+	movups STATE1, (OUTP)
+	pxor IN2, STATE2
+	movups STATE2, 0x10(OUTP)
+	pxor IN3, STATE3
+	movups STATE3, 0x20(OUTP)
+	pxor IN4, STATE4
+	movups STATE4, 0x30(OUTP)
+	sub $64, LEN
+	add $64, INP
+	add $64, OUTP
+	cmp $64, LEN
+	jge .Lctr_enc_loop4
+	cmp $16, LEN
+	jb .Lctr_enc_ret
+.align 4
+.Lctr_enc_loop1:
+	movaps IV, STATE
+	call _aesni_inc
+	movups (INP), IN
+	call _aesni_enc1
+	pxor IN, STATE
+	movups STATE, (OUTP)
+	sub $16, LEN
+	add $16, INP
+	add $16, OUTP
+	cmp $16, LEN
+	jge .Lctr_enc_loop1
+.Lctr_enc_ret:
+	movups IV, (IVP)
+.Lctr_enc_just_ret:
+	FRAME_END
+	RET
+SYM_FUNC_END(aesni_ctr_enc)
+
+#endif
+
+.section	.rodata.cst16.gf128mul_x_ble_mask, "aM", @progbits, 16
+.align 16
+.Lgf128mul_x_ble_mask:
+	.octa 0x00000000000000010000000000000087
+.previous
+
+/*
+ * _aesni_gf128mul_x_ble:		internal ABI
+ *	Multiply in GF(2^128) for XTS IVs
+ * input:
+ *	IV:	current IV
+ *	GF128MUL_MASK == mask with 0x87 and 0x01
+ * output:
+ *	IV:	next IV
+ * changed:
+ *	CTR:	== temporary value
+ */
+#define _aesni_gf128mul_x_ble() \
+	pshufd $0x13, IV, KEY; \
+	paddq IV, IV; \
+	psrad $31, KEY; \
+	pand GF128MUL_MASK, KEY; \
+	pxor KEY, IV;
+
+/*
+ * void aesni_xts_encrypt(const struct crypto_aes_ctx *ctx, u8 *dst,
+ *			  const u8 *src, unsigned int len, le128 *iv)
+ */
+SYM_FUNC_START(aesni_xts_encrypt)
+	FRAME_BEGIN
+#ifndef __x86_64__
+	pushl IVP
+	pushl LEN
+	pushl KEYP
+	pushl KLEN
+	movl (FRAME_OFFSET+20)(%esp), KEYP	# ctx
+	movl (FRAME_OFFSET+24)(%esp), OUTP	# dst
+	movl (FRAME_OFFSET+28)(%esp), INP	# src
+	movl (FRAME_OFFSET+32)(%esp), LEN	# len
+	movl (FRAME_OFFSET+36)(%esp), IVP	# iv
+	movdqa .Lgf128mul_x_ble_mask, GF128MUL_MASK
+#else
+	movdqa .Lgf128mul_x_ble_mask(%rip), GF128MUL_MASK
+#endif
+	movups (IVP), IV
+
+	mov 480(KEYP), KLEN
+
+.Lxts_enc_loop4:
+	sub $64, LEN
+	jl .Lxts_enc_1x
+
+	movdqa IV, STATE1
+	movdqu 0x00(INP), IN
+	pxor IN, STATE1
+	movdqu IV, 0x00(OUTP)
+
+	_aesni_gf128mul_x_ble()
+	movdqa IV, STATE2
+	movdqu 0x10(INP), IN
+	pxor IN, STATE2
+	movdqu IV, 0x10(OUTP)
+
+	_aesni_gf128mul_x_ble()
+	movdqa IV, STATE3
+	movdqu 0x20(INP), IN
+	pxor IN, STATE3
+	movdqu IV, 0x20(OUTP)
+
+	_aesni_gf128mul_x_ble()
+	movdqa IV, STATE4
+	movdqu 0x30(INP), IN
+	pxor IN, STATE4
+	movdqu IV, 0x30(OUTP)
+
+	call _aesni_enc4
+
+	movdqu 0x00(OUTP), IN
+	pxor IN, STATE1
+	movdqu STATE1, 0x00(OUTP)
+
+	movdqu 0x10(OUTP), IN
+	pxor IN, STATE2
+	movdqu STATE2, 0x10(OUTP)
+
+	movdqu 0x20(OUTP), IN
+	pxor IN, STATE3
+	movdqu STATE3, 0x20(OUTP)
+
+	movdqu 0x30(OUTP), IN
+	pxor IN, STATE4
+	movdqu STATE4, 0x30(OUTP)
+
+	_aesni_gf128mul_x_ble()
+
+	add $64, INP
+	add $64, OUTP
+	test LEN, LEN
+	jnz .Lxts_enc_loop4
+
+.Lxts_enc_ret_iv:
+	movups IV, (IVP)
+
+.Lxts_enc_ret:
+#ifndef __x86_64__
+	popl KLEN
+	popl KEYP
+	popl LEN
+	popl IVP
+#endif
+	FRAME_END
+	RET
+
+.Lxts_enc_1x:
+	add $64, LEN
+	jz .Lxts_enc_ret_iv
+	sub $16, LEN
+	jl .Lxts_enc_cts4
+
+.Lxts_enc_loop1:
+	movdqu (INP), STATE
+	pxor IV, STATE
+	call _aesni_enc1
+	pxor IV, STATE
+	_aesni_gf128mul_x_ble()
+
+	test LEN, LEN
+	jz .Lxts_enc_out
+
+	add $16, INP
+	sub $16, LEN
+	jl .Lxts_enc_cts1
+
+	movdqu STATE, (OUTP)
+	add $16, OUTP
+	jmp .Lxts_enc_loop1
+
+.Lxts_enc_out:
+	movdqu STATE, (OUTP)
+	jmp .Lxts_enc_ret_iv
+
+.Lxts_enc_cts4:
+	movdqa STATE4, STATE
+	sub $16, OUTP
+
+.Lxts_enc_cts1:
+#ifndef __x86_64__
+	lea .Lcts_permute_table, T1
+#else
+	lea .Lcts_permute_table(%rip), T1
+#endif
+	add LEN, INP		/* rewind input pointer */
+	add $16, LEN		/* # bytes in final block */
+	movups (INP), IN1
+
+	mov T1, IVP
+	add $32, IVP
+	add LEN, T1
+	sub LEN, IVP
+	add OUTP, LEN
+
+	movups (T1), %xmm4
+	movaps STATE, IN2
+	pshufb %xmm4, STATE
+	movups STATE, (LEN)
+
+	movups (IVP), %xmm0
+	pshufb %xmm0, IN1
+	pblendvb IN2, IN1
+	movaps IN1, STATE
+
+	pxor IV, STATE
+	call _aesni_enc1
+	pxor IV, STATE
+
+	movups STATE, (OUTP)
+	jmp .Lxts_enc_ret
+SYM_FUNC_END(aesni_xts_encrypt)
+
+/*
+ * void aesni_xts_decrypt(const struct crypto_aes_ctx *ctx, u8 *dst,
+ *			  const u8 *src, unsigned int len, le128 *iv)
+ */
+SYM_FUNC_START(aesni_xts_decrypt)
+	FRAME_BEGIN
+#ifndef __x86_64__
+	pushl IVP
+	pushl LEN
+	pushl KEYP
+	pushl KLEN
+	movl (FRAME_OFFSET+20)(%esp), KEYP	# ctx
+	movl (FRAME_OFFSET+24)(%esp), OUTP	# dst
+	movl (FRAME_OFFSET+28)(%esp), INP	# src
+	movl (FRAME_OFFSET+32)(%esp), LEN	# len
+	movl (FRAME_OFFSET+36)(%esp), IVP	# iv
+	movdqa .Lgf128mul_x_ble_mask, GF128MUL_MASK
+#else
+	movdqa .Lgf128mul_x_ble_mask(%rip), GF128MUL_MASK
+#endif
+	movups (IVP), IV
+
+	mov 480(KEYP), KLEN
+	add $240, KEYP
+
+	test $15, LEN
+	jz .Lxts_dec_loop4
+	sub $16, LEN
+
+.Lxts_dec_loop4:
+	sub $64, LEN
+	jl .Lxts_dec_1x
+
+	movdqa IV, STATE1
+	movdqu 0x00(INP), IN
+	pxor IN, STATE1
+	movdqu IV, 0x00(OUTP)
+
+	_aesni_gf128mul_x_ble()
+	movdqa IV, STATE2
+	movdqu 0x10(INP), IN
+	pxor IN, STATE2
+	movdqu IV, 0x10(OUTP)
+
+	_aesni_gf128mul_x_ble()
+	movdqa IV, STATE3
+	movdqu 0x20(INP), IN
+	pxor IN, STATE3
+	movdqu IV, 0x20(OUTP)
+
+	_aesni_gf128mul_x_ble()
+	movdqa IV, STATE4
+	movdqu 0x30(INP), IN
+	pxor IN, STATE4
+	movdqu IV, 0x30(OUTP)
+
+	call _aesni_dec4
+
+	movdqu 0x00(OUTP), IN
+	pxor IN, STATE1
+	movdqu STATE1, 0x00(OUTP)
+
+	movdqu 0x10(OUTP), IN
+	pxor IN, STATE2
+	movdqu STATE2, 0x10(OUTP)
+
+	movdqu 0x20(OUTP), IN
+	pxor IN, STATE3
+	movdqu STATE3, 0x20(OUTP)
+
+	movdqu 0x30(OUTP), IN
+	pxor IN, STATE4
+	movdqu STATE4, 0x30(OUTP)
+
+	_aesni_gf128mul_x_ble()
+
+	add $64, INP
+	add $64, OUTP
+	test LEN, LEN
+	jnz .Lxts_dec_loop4
+
+.Lxts_dec_ret_iv:
+	movups IV, (IVP)
+
+.Lxts_dec_ret:
+#ifndef __x86_64__
+	popl KLEN
+	popl KEYP
+	popl LEN
+	popl IVP
+#endif
+	FRAME_END
+	RET
+
+.Lxts_dec_1x:
+	add $64, LEN
+	jz .Lxts_dec_ret_iv
+
+.Lxts_dec_loop1:
+	movdqu (INP), STATE
+
+	add $16, INP
+	sub $16, LEN
+	jl .Lxts_dec_cts1
+
+	pxor IV, STATE
+	call _aesni_dec1
+	pxor IV, STATE
+	_aesni_gf128mul_x_ble()
+
+	test LEN, LEN
+	jz .Lxts_dec_out
+
+	movdqu STATE, (OUTP)
+	add $16, OUTP
+	jmp .Lxts_dec_loop1
+
+.Lxts_dec_out:
+	movdqu STATE, (OUTP)
+	jmp .Lxts_dec_ret_iv
+
+.Lxts_dec_cts1:
+	movdqa IV, STATE4
+	_aesni_gf128mul_x_ble()
+
+	pxor IV, STATE
+	call _aesni_dec1
+	pxor IV, STATE
+
+#ifndef __x86_64__
+	lea .Lcts_permute_table, T1
+#else
+	lea .Lcts_permute_table(%rip), T1
+#endif
+	add LEN, INP		/* rewind input pointer */
+	add $16, LEN		/* # bytes in final block */
+	movups (INP), IN1
+
+	mov T1, IVP
+	add $32, IVP
+	add LEN, T1
+	sub LEN, IVP
+	add OUTP, LEN
+
+	movups (T1), %xmm4
+	movaps STATE, IN2
+	pshufb %xmm4, STATE
+	movups STATE, (LEN)
+
+	movups (IVP), %xmm0
+	pshufb %xmm0, IN1
+	pblendvb IN2, IN1
+	movaps IN1, STATE
+
+	pxor STATE4, STATE
+	call _aesni_dec1
+	pxor STATE4, STATE
+
+	movups STATE, (OUTP)
+	jmp .Lxts_dec_ret
+SYM_FUNC_END(aesni_xts_decrypt)