From: Nikos Mavrogiannopoulos <nmav@gnutls.org>
Date: Sat, 16 Apr 2011 16:39:29 +0000 (+0200)
Subject: Added Andy Polyakov's version of AES-NI optimizations.
X-Git-Tag: gnutls_2_99_1~16
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=065927be43c1aef5771c0b18b28b8ef9ccdc6eae;p=thirdparty%2Fgnutls.git

Added Andy Polyakov's version of AES-NI optimizations.
---

diff --git a/NEWS b/NEWS
index db71fa72d3..2976d3c365 100644
--- a/NEWS
+++ b/NEWS
@@ -6,7 +6,7 @@ See the end for copying conditions.
 * Version 2.99.1 (unreleased)
 
 ** libgnutls: Added support for AES-NI if detected. Uses
-Intel AES-NI code.
+Andy Polyakov's AES-NI code.
 
 ** libgnutls-extra: Dropped support of LZO compression via liblzo.
 
diff --git a/configure.ac b/configure.ac
index 5dc81a18b6..85fa87a4d9 100644
--- a/configure.ac
+++ b/configure.ac
@@ -36,6 +36,7 @@ AC_MSG_RESULT([***
 
 dnl Checks for programs.
 AC_PROG_CC
+AM_PROG_AS
 AC_PROG_CXX
 gl_EARLY
 
@@ -79,18 +80,13 @@ if test "$use_accel" != "no"; then
 case $host_cpu in
   i?86 | x86_64 | amd64)
     GCC_FLAG_ADD([-maes -mpclmul],[X86])
-    AC_CHECK_PROGS(YASM, yasm)
-
-    if test "x$YASM" != "x";then
-      if test "x$X86" = "xyes";then
-        if test "$host_cpu" = "x86_64" -o "$host_cpu" = "amd64";then
-          hw_accel="x86-64"
-        else
-          hw_accel="x86"
-        fi
+
+    if test "x$X86" = "xyes";then
+      if test "$host_cpu" = "x86_64" -o "$host_cpu" = "amd64";then
+        hw_accel="x86-64"
+      else
+        hw_accel="x86"
       fi
-    else
-      AC_MSG_WARN([[yasm assembler not found. Disabling AES-NI compilation.]])
     fi
   ;;
   *)
diff --git a/lib/accelerated/intel/Makefile.am b/lib/accelerated/intel/Makefile.am
index f57b09862f..5efe894d55 100644
--- a/lib/accelerated/intel/Makefile.am
+++ b/lib/accelerated/intel/Makefile.am
@@ -32,25 +32,16 @@ if ENABLE_MINITASN1
 AM_CPPFLAGS += -I$(srcdir)/../../minitasn1
 endif
 
-EXTRA_DIST = aes-x86.h README license.txt iaes_asm_interface.h iaesni.h \
-	asm/x64_iaesx64.s asm/x86_iaesx86.s
+EXTRA_DIST = aes-x86.h README license.txt
 
 noinst_LTLIBRARIES = libintel.la
 
 libintel_la_SOURCES = aes-x86.c
 libintel_la_LIBADD =
 
-YASM_OPTS = -D__linux__
-
-x64_iaesx64.o: asm/x64_iaesx64.s
-	$(YASM) $(YASM_OPTS) -f elf64 $^ -o $@
-
-x86_iaesx86.o: asm/x86_iaesx86.s
-	$(YASM) $(YASM_OPTS) -f elf32 $^ -o $@
-
 if ASM_X86_64
-libintel_la_LIBADD += x64_iaesx64.o
+libintel_la_SOURCES += asm/appro-aes-x86-64.s
 else
-libintel_la_LIBADD += x86_iaesx86.o
+libintel_la_SOURCES += asm/appro-aes-x86.s
 endif
 
diff --git a/lib/accelerated/intel/README b/lib/accelerated/intel/README
index 187242bebd..dc1f43fa42 100644
--- a/lib/accelerated/intel/README
+++ b/lib/accelerated/intel/README
@@ -1,3 +1,3 @@
-The Intel AES library is not part of the GnuTLS library, but is used with
-GnuTLS. The license of Intel AES library is included in license.txt.
+The AES-NI implementation by Andy Polyakov is not part of the GnuTLS library, but is 
+used with GnuTLS. Its license is included in license.txt.
 
diff --git a/lib/accelerated/intel/aes-x86.c b/lib/accelerated/intel/aes-x86.c
index df4a49ba97..f1a24aa1b6 100644
--- a/lib/accelerated/intel/aes-x86.c
+++ b/lib/accelerated/intel/aes-x86.c
@@ -31,33 +31,44 @@
 #include <gnutls_errors.h>
 #include <aes-x86.h>
 #include <x86.h>
-#include "iaes_asm_interface.h"
 
 #ifdef __GNUC__
-# define ALIGN16 __attribute__ ((aligned (16))) 
+# define ALIGN16 __attribute__ ((aligned (16)))
 #else
 # define ALIGN16
 #endif
 
-typedef void (*enc_func)(sAesData*);
-
-struct aes_ctx {
-	uint8_t ALIGN16 expanded_key[16*16];
-	uint8_t ALIGN16 expanded_key_dec[16*16];
-	uint8_t iv[16];
-	enc_func enc;
-	enc_func dec;
-	size_t keysize;
+#define AES_MAXNR 14
+typedef struct
+{
+  uint32_t ALIGN16 rd_key[4 * (AES_MAXNR + 1)];
+  int rounds;
+} AES_KEY;
+
+void aesni_cbc_encrypt (const unsigned char *in, unsigned char *out,
+                        size_t len, const AES_KEY * key,
+                        unsigned char *ivec, const int enc);
+int aesni_set_decrypt_key (const unsigned char *userKey, const int bits,
+                           AES_KEY * key);
+int aesni_set_encrypt_key (const unsigned char *userKey, const int bits,
+                           AES_KEY * key);
+
+struct aes_ctx
+{
+  AES_KEY expanded_key;
+  AES_KEY expanded_key_dec;
+  uint8_t iv[16];
 };
 
 static int
 aes_cipher_init (gnutls_cipher_algorithm_t algorithm, void **_ctx)
 {
   struct aes_ctx *ctx;
-  
+
   /* we use key size to distinguish */
-  if (algorithm != GNUTLS_CIPHER_AES_128_CBC && algorithm != GNUTLS_CIPHER_AES_192_CBC 
-    && algorithm != GNUTLS_CIPHER_AES_256_CBC)
+  if (algorithm != GNUTLS_CIPHER_AES_128_CBC
+      && algorithm != GNUTLS_CIPHER_AES_192_CBC
+      && algorithm != GNUTLS_CIPHER_AES_256_CBC)
     return GNUTLS_E_INVALID_REQUEST;
 
   *_ctx = gnutls_calloc (1, sizeof (struct aes_ctx));
@@ -75,33 +86,16 @@ aes_cipher_init (gnutls_cipher_algorithm_t algorithm, void **_ctx)
 static int
 aes_cipher_setkey (void *_ctx, const void *userkey, size_t keysize)
 {
-struct aes_ctx *ctx = _ctx;
+  struct aes_ctx *ctx = _ctx;
+  int ret;
 
-  if (keysize == 128/8)
-    {
-      iEncExpandKey128((void*)userkey, ctx->expanded_key);
-      iDecExpandKey128((void*)userkey, ctx->expanded_key_dec);
-      ctx->enc = iEnc128_CBC;
-      ctx->dec = iDec128_CBC;
-    }
-  else if (keysize == 192/8)
-    {
-      iEncExpandKey192((void*)userkey, ctx->expanded_key);
-      iDecExpandKey192((void*)userkey, ctx->expanded_key_dec);
-      ctx->enc = iEnc192_CBC;
-      ctx->dec = iDec192_CBC;
-    }
-  else if (keysize == 256/8)
-    {
-      iEncExpandKey256((void*)userkey, ctx->expanded_key);
-      iDecExpandKey256((void*)userkey, ctx->expanded_key_dec);
-      ctx->enc = iEnc256_CBC;
-      ctx->dec = iDec256_CBC;
-    }
-  else
-    return gnutls_assert_val(GNUTLS_E_INVALID_REQUEST);
+  ret = aesni_set_encrypt_key (userkey, keysize * 8, &ctx->expanded_key);
+  if (ret != 0)
+    return gnutls_assert_val (GNUTLS_E_ENCRYPTION_FAILED);
 
-  ctx->keysize = keysize;
+  ret = aesni_set_decrypt_key (userkey, keysize * 8, &ctx->expanded_key_dec);
+  if (ret != 0)
+    return gnutls_assert_val (GNUTLS_E_ENCRYPTION_FAILED);
 
   return 0;
 }
@@ -117,36 +111,22 @@ aes_setiv (void *_ctx, const void *iv, size_t iv_size)
 
 static int
 aes_encrypt (void *_ctx, const void *plain, size_t plainsize,
-                   void *encr, size_t length)
+             void *encr, size_t length)
 {
-struct aes_ctx *ctx = _ctx;
-sAesData aesData;
-  
-  aesData.iv = ctx->iv;
-  aesData.in_block = (void*)plain;
-  aesData.out_block = encr;
-  aesData.expanded_key = ctx->expanded_key;
-  aesData.num_blocks = (plainsize + 1) / 16;
-
-  ctx->enc(&aesData);
+  struct aes_ctx *ctx = _ctx;
 
+  aesni_cbc_encrypt (plain, encr, plainsize, &ctx->expanded_key, ctx->iv, 1);
   return 0;
 }
 
 static int
 aes_decrypt (void *_ctx, const void *encr, size_t encrsize,
-                   void *plain, size_t length)
+             void *plain, size_t length)
 {
-struct aes_ctx *ctx = _ctx;
-sAesData aesData;
-
-  aesData.iv = ctx->iv;
-  aesData.in_block = (void*)encr;
-  aesData.out_block = plain;
-  aesData.expanded_key = ctx->expanded_key_dec;
-  aesData.num_blocks = (encrsize + 1) / 16;
+  struct aes_ctx *ctx = _ctx;
 
-  ctx->dec(&aesData);
+  aesni_cbc_encrypt (encr, plain, encrsize,
+                     &ctx->expanded_key_dec, ctx->iv, 0);
 
   return 0;
 }
@@ -166,38 +146,46 @@ static const gnutls_crypto_cipher_st cipher_struct = {
   .deinit = aes_deinit,
 };
 
-static unsigned check_optimized_aes(void)
+static unsigned
+check_optimized_aes (void)
 {
-unsigned int a,b,c,d;
-  cpuid(1, a,b,c,d);
-  
+  unsigned int a, b, c, d;
+  cpuid (1, a, b, c, d);
+
   return (c & 0x2000000);
 }
 
 void
 register_x86_crypto (void)
 {
-int ret;
-	if (check_optimized_aes()) {
-	        fprintf(stderr, "Intel AES accelerator was detected\n");
-		ret = gnutls_crypto_single_cipher_register (GNUTLS_CIPHER_AES_128_CBC, 80, &cipher_struct);
-		if (ret < 0)
-		{
-		  gnutls_assert ();
-		}
-
-		ret = gnutls_crypto_single_cipher_register (GNUTLS_CIPHER_AES_192_CBC, 80, &cipher_struct);
-		if (ret < 0)
-		{
-		  gnutls_assert ();
-		}
-
-		ret = gnutls_crypto_single_cipher_register (GNUTLS_CIPHER_AES_256_CBC, 80, &cipher_struct);
-		if (ret < 0)
-		{
-		  gnutls_assert ();
-		}
-	}
-
-	return;
+  int ret;
+  if (check_optimized_aes ())
+    {
+      fprintf (stderr, "Intel AES accelerator was detected\n");
+      ret =
+        gnutls_crypto_single_cipher_register (GNUTLS_CIPHER_AES_128_CBC, 80,
+                                              &cipher_struct);
+      if (ret < 0)
+        {
+          gnutls_assert ();
+        }
+
+      ret =
+        gnutls_crypto_single_cipher_register (GNUTLS_CIPHER_AES_192_CBC, 80,
+                                              &cipher_struct);
+      if (ret < 0)
+        {
+          gnutls_assert ();
+        }
+
+      ret =
+        gnutls_crypto_single_cipher_register (GNUTLS_CIPHER_AES_256_CBC, 80,
+                                              &cipher_struct);
+      if (ret < 0)
+        {
+          gnutls_assert ();
+        }
+    }
+
+  return;
 }
diff --git a/lib/accelerated/intel/asm/appro-aes-x86-64.s b/lib/accelerated/intel/asm/appro-aes-x86-64.s
new file mode 100644
index 0000000000..96b7b6eadc
--- /dev/null
+++ b/lib/accelerated/intel/asm/appro-aes-x86-64.s
@@ -0,0 +1,782 @@
+# Copyright (c) 2006, Andy Polyakov by <appro@openssl.org>
+# All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions
+# are met:
+# 
+#     *	Redistributions of source code must retain copyright notices,
+#	this list of conditions and the following disclaimer.
+#
+#     *	Redistributions in binary form must reproduce the above
+# 	copyright notice, this list of conditions and the following
+#	disclaimer in the documentation and/or other materials
+#	provided with the distribution.
+#
+#     *	Neither the name of the Andy Polyakov nor the names of its
+#	copyright holder and contributors may be used to endorse or
+#	promote products derived from this software without specific
+#	prior written permission.
+#
+# ALTERNATIVELY, provided that this notice is retained in full, this
+# product may be distributed under the terms of the GNU General Public
+# License (GPL), in which case the provisions of the GPL apply INSTEAD OF
+# those given above.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER AND CONTRIBUTORS
+# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+.text	
+.globl	aesni_encrypt
+.type	aesni_encrypt,@function
+.align	16
+aesni_encrypt:
+	movups	(%rdi),%xmm0
+	movl	240(%rdx),%eax
+	movaps	(%rdx),%xmm4
+	movaps	16(%rdx),%xmm5
+	leaq	32(%rdx),%rdx
+	pxor	%xmm4,%xmm0
+.Loop_enc1_1:
+.byte	102,15,56,220,197
+	decl	%eax
+	movaps	(%rdx),%xmm5
+	leaq	16(%rdx),%rdx
+	jnz	.Loop_enc1_1	
+.byte	102,15,56,221,197
+	movups	%xmm0,(%rsi)
+	.byte	0xf3,0xc3
+.size	aesni_encrypt,.-aesni_encrypt
+
+.globl	aesni_decrypt
+.type	aesni_decrypt,@function
+.align	16
+aesni_decrypt:
+	movups	(%rdi),%xmm0
+	movl	240(%rdx),%eax
+	movaps	(%rdx),%xmm4
+	movaps	16(%rdx),%xmm5
+	leaq	32(%rdx),%rdx
+	pxor	%xmm4,%xmm0
+.Loop_dec1_2:
+.byte	102,15,56,222,197
+	decl	%eax
+	movaps	(%rdx),%xmm5
+	leaq	16(%rdx),%rdx
+	jnz	.Loop_dec1_2	
+.byte	102,15,56,223,197
+	movups	%xmm0,(%rsi)
+	.byte	0xf3,0xc3
+.size	aesni_decrypt, .-aesni_decrypt
+.type	_aesni_encrypt3,@function
+.align	16
+_aesni_encrypt3:
+	movaps	(%rcx),%xmm4
+	shrl	$1,%eax
+	movaps	16(%rcx),%xmm5
+	leaq	32(%rcx),%rcx
+	pxor	%xmm4,%xmm0
+	pxor	%xmm4,%xmm1
+	pxor	%xmm4,%xmm2
+
+.Lenc_loop3:
+.byte	102,15,56,220,197
+	movaps	(%rcx),%xmm4
+.byte	102,15,56,220,205
+	decl	%eax
+.byte	102,15,56,220,213
+.byte	102,15,56,220,196
+	movaps	16(%rcx),%xmm5
+.byte	102,15,56,220,204
+	leaq	32(%rcx),%rcx
+.byte	102,15,56,220,212
+	jnz	.Lenc_loop3
+
+.byte	102,15,56,220,197
+	movaps	(%rcx),%xmm4
+.byte	102,15,56,220,205
+.byte	102,15,56,220,213
+.byte	102,15,56,221,196
+.byte	102,15,56,221,204
+.byte	102,15,56,221,212
+	.byte	0xf3,0xc3
+.size	_aesni_encrypt3,.-_aesni_encrypt3
+.type	_aesni_decrypt3,@function
+.align	16
+_aesni_decrypt3:
+	movaps	(%rcx),%xmm4
+	shrl	$1,%eax
+	movaps	16(%rcx),%xmm5
+	leaq	32(%rcx),%rcx
+	pxor	%xmm4,%xmm0
+	pxor	%xmm4,%xmm1
+	pxor	%xmm4,%xmm2
+
+.Ldec_loop3:
+.byte	102,15,56,222,197
+	movaps	(%rcx),%xmm4
+.byte	102,15,56,222,205
+	decl	%eax
+.byte	102,15,56,222,213
+.byte	102,15,56,222,196
+	movaps	16(%rcx),%xmm5
+.byte	102,15,56,222,204
+	leaq	32(%rcx),%rcx
+.byte	102,15,56,222,212
+	jnz	.Ldec_loop3
+
+.byte	102,15,56,222,197
+	movaps	(%rcx),%xmm4
+.byte	102,15,56,222,205
+.byte	102,15,56,222,213
+.byte	102,15,56,223,196
+.byte	102,15,56,223,204
+.byte	102,15,56,223,212
+	.byte	0xf3,0xc3
+.size	_aesni_decrypt3,.-_aesni_decrypt3
+.type	_aesni_encrypt4,@function
+.align	16
+_aesni_encrypt4:
+	movaps	(%rcx),%xmm4
+	shrl	$1,%eax
+	movaps	16(%rcx),%xmm5
+	leaq	32(%rcx),%rcx
+	pxor	%xmm4,%xmm0
+	pxor	%xmm4,%xmm1
+	pxor	%xmm4,%xmm2
+	pxor	%xmm4,%xmm3
+
+.Lenc_loop4:
+.byte	102,15,56,220,197
+	movaps	(%rcx),%xmm4
+.byte	102,15,56,220,205
+	decl	%eax
+.byte	102,15,56,220,213
+.byte	102,15,56,220,221
+.byte	102,15,56,220,196
+	movaps	16(%rcx),%xmm5
+.byte	102,15,56,220,204
+	leaq	32(%rcx),%rcx
+.byte	102,15,56,220,212
+.byte	102,15,56,220,220
+	jnz	.Lenc_loop4
+
+.byte	102,15,56,220,197
+	movaps	(%rcx),%xmm4
+.byte	102,15,56,220,205
+.byte	102,15,56,220,213
+.byte	102,15,56,220,221
+.byte	102,15,56,221,196
+.byte	102,15,56,221,204
+.byte	102,15,56,221,212
+.byte	102,15,56,221,220
+	.byte	0xf3,0xc3
+.size	_aesni_encrypt4,.-_aesni_encrypt4
+.type	_aesni_decrypt4,@function
+.align	16
+_aesni_decrypt4:
+	movaps	(%rcx),%xmm4
+	shrl	$1,%eax
+	movaps	16(%rcx),%xmm5
+	leaq	32(%rcx),%rcx
+	pxor	%xmm4,%xmm0
+	pxor	%xmm4,%xmm1
+	pxor	%xmm4,%xmm2
+	pxor	%xmm4,%xmm3
+
+.Ldec_loop4:
+.byte	102,15,56,222,197
+	movaps	(%rcx),%xmm4
+.byte	102,15,56,222,205
+	decl	%eax
+.byte	102,15,56,222,213
+.byte	102,15,56,222,221
+.byte	102,15,56,222,196
+	movaps	16(%rcx),%xmm5
+.byte	102,15,56,222,204
+	leaq	32(%rcx),%rcx
+.byte	102,15,56,222,212
+.byte	102,15,56,222,220
+	jnz	.Ldec_loop4
+
+.byte	102,15,56,222,197
+	movaps	(%rcx),%xmm4
+.byte	102,15,56,222,205
+.byte	102,15,56,222,213
+.byte	102,15,56,222,221
+.byte	102,15,56,223,196
+.byte	102,15,56,223,204
+.byte	102,15,56,223,212
+.byte	102,15,56,223,220
+	.byte	0xf3,0xc3
+.size	_aesni_decrypt4,.-_aesni_decrypt4
+.globl	aesni_ecb_encrypt
+.type	aesni_ecb_encrypt,@function
+.align	16
+aesni_ecb_encrypt:
+	cmpq	$16,%rdx
+	jb	.Lecb_ret
+
+	movl	240(%rcx),%eax
+	andq	$-16,%rdx
+	movq	%rcx,%r11
+	testl	%r8d,%r8d
+	movl	%eax,%r10d
+	jz	.Lecb_decrypt
+
+	subq	$64,%rdx
+	jbe	.Lecb_enc_tail
+	jmp	.Lecb_enc_loop3
+.align	16
+.Lecb_enc_loop3:
+	movups	(%rdi),%xmm0
+	movups	16(%rdi),%xmm1
+	movups	32(%rdi),%xmm2
+	call	_aesni_encrypt3
+	subq	$48,%rdx
+	leaq	48(%rdi),%rdi
+	leaq	48(%rsi),%rsi
+	movups	%xmm0,-48(%rsi)
+	movl	%r10d,%eax
+	movups	%xmm1,-32(%rsi)
+	movq	%r11,%rcx
+	movups	%xmm2,-16(%rsi)
+	ja	.Lecb_enc_loop3
+
+.Lecb_enc_tail:
+	addq	$64,%rdx
+	jz	.Lecb_ret
+
+	cmpq	$16,%rdx
+	movups	(%rdi),%xmm0
+	je	.Lecb_enc_one
+	cmpq	$32,%rdx
+	movups	16(%rdi),%xmm1
+	je	.Lecb_enc_two
+	cmpq	$48,%rdx
+	movups	32(%rdi),%xmm2
+	je	.Lecb_enc_three
+	movups	48(%rdi),%xmm3
+	call	_aesni_encrypt4
+	movups	%xmm0,(%rsi)
+	movups	%xmm1,16(%rsi)
+	movups	%xmm2,32(%rsi)
+	movups	%xmm3,48(%rsi)
+	jmp	.Lecb_ret
+.align	16
+.Lecb_enc_one:
+	movaps	(%rcx),%xmm4
+	movaps	16(%rcx),%xmm5
+	leaq	32(%rcx),%rcx
+	pxor	%xmm4,%xmm0
+.Loop_enc1_3:
+.byte	102,15,56,220,197
+	decl	%eax
+	movaps	(%rcx),%xmm5
+	leaq	16(%rcx),%rcx
+	jnz	.Loop_enc1_3	
+.byte	102,15,56,221,197
+	movups	%xmm0,(%rsi)
+	jmp	.Lecb_ret
+.align	16
+.Lecb_enc_two:
+	call	_aesni_encrypt3
+	movups	%xmm0,(%rsi)
+	movups	%xmm1,16(%rsi)
+	jmp	.Lecb_ret
+.align	16
+.Lecb_enc_three:
+	call	_aesni_encrypt3
+	movups	%xmm0,(%rsi)
+	movups	%xmm1,16(%rsi)
+	movups	%xmm2,32(%rsi)
+	jmp	.Lecb_ret
+
+.align	16
+.Lecb_decrypt:
+	subq	$64,%rdx
+	jbe	.Lecb_dec_tail
+	jmp	.Lecb_dec_loop3
+.align	16
+.Lecb_dec_loop3:
+	movups	(%rdi),%xmm0
+	movups	16(%rdi),%xmm1
+	movups	32(%rdi),%xmm2
+	call	_aesni_decrypt3
+	subq	$48,%rdx
+	leaq	48(%rdi),%rdi
+	leaq	48(%rsi),%rsi
+	movups	%xmm0,-48(%rsi)
+	movl	%r10d,%eax
+	movups	%xmm1,-32(%rsi)
+	movq	%r11,%rcx
+	movups	%xmm2,-16(%rsi)
+	ja	.Lecb_dec_loop3
+
+.Lecb_dec_tail:
+	addq	$64,%rdx
+	jz	.Lecb_ret
+
+	cmpq	$16,%rdx
+	movups	(%rdi),%xmm0
+	je	.Lecb_dec_one
+	cmpq	$32,%rdx
+	movups	16(%rdi),%xmm1
+	je	.Lecb_dec_two
+	cmpq	$48,%rdx
+	movups	32(%rdi),%xmm2
+	je	.Lecb_dec_three
+	movups	48(%rdi),%xmm3
+	call	_aesni_decrypt4
+	movups	%xmm0,(%rsi)
+	movups	%xmm1,16(%rsi)
+	movups	%xmm2,32(%rsi)
+	movups	%xmm3,48(%rsi)
+	jmp	.Lecb_ret
+.align	16
+.Lecb_dec_one:
+	movaps	(%rcx),%xmm4
+	movaps	16(%rcx),%xmm5
+	leaq	32(%rcx),%rcx
+	pxor	%xmm4,%xmm0
+.Loop_dec1_4:
+.byte	102,15,56,222,197
+	decl	%eax
+	movaps	(%rcx),%xmm5
+	leaq	16(%rcx),%rcx
+	jnz	.Loop_dec1_4	
+.byte	102,15,56,223,197
+	movups	%xmm0,(%rsi)
+	jmp	.Lecb_ret
+.align	16
+.Lecb_dec_two:
+	call	_aesni_decrypt3
+	movups	%xmm0,(%rsi)
+	movups	%xmm1,16(%rsi)
+	jmp	.Lecb_ret
+.align	16
+.Lecb_dec_three:
+	call	_aesni_decrypt3
+	movups	%xmm0,(%rsi)
+	movups	%xmm1,16(%rsi)
+	movups	%xmm2,32(%rsi)
+
+.Lecb_ret:
+	.byte	0xf3,0xc3
+.size	aesni_ecb_encrypt,.-aesni_ecb_encrypt
+.globl	aesni_cbc_encrypt
+.type	aesni_cbc_encrypt,@function
+.align	16
+aesni_cbc_encrypt:
+	testq	%rdx,%rdx
+	jz	.Lcbc_ret
+
+	movl	240(%rcx),%r10d
+	movq	%rcx,%r11
+	testl	%r9d,%r9d
+	jz	.Lcbc_decrypt
+
+	movups	(%r8),%xmm0
+	cmpq	$16,%rdx
+	movl	%r10d,%eax
+	jb	.Lcbc_enc_tail
+	subq	$16,%rdx
+	jmp	.Lcbc_enc_loop
+.align	16
+.Lcbc_enc_loop:
+	movups	(%rdi),%xmm1
+	leaq	16(%rdi),%rdi
+	pxor	%xmm1,%xmm0
+	movaps	(%rcx),%xmm4
+	movaps	16(%rcx),%xmm5
+	leaq	32(%rcx),%rcx
+	pxor	%xmm4,%xmm0
+.Loop_enc1_5:
+.byte	102,15,56,220,197
+	decl	%eax
+	movaps	(%rcx),%xmm5
+	leaq	16(%rcx),%rcx
+	jnz	.Loop_enc1_5	
+.byte	102,15,56,221,197
+	subq	$16,%rdx
+	leaq	16(%rsi),%rsi
+	movl	%r10d,%eax
+	movq	%r11,%rcx
+	movups	%xmm0,-16(%rsi)
+	jnc	.Lcbc_enc_loop
+	addq	$16,%rdx
+	jnz	.Lcbc_enc_tail
+	movups	%xmm0,(%r8)
+	jmp	.Lcbc_ret
+
+.Lcbc_enc_tail:
+	movq	%rdx,%rcx
+	xchgq	%rdi,%rsi
+.long	0x9066A4F3	
+	movl	$16,%ecx
+	subq	%rdx,%rcx
+	xorl	%eax,%eax
+.long	0x9066AAF3	
+	leaq	-16(%rdi),%rdi
+	movl	%r10d,%eax
+	movq	%rdi,%rsi
+	movq	%r11,%rcx
+	xorq	%rdx,%rdx
+	jmp	.Lcbc_enc_loop	
+
+.align	16
+.Lcbc_decrypt:
+	movups	(%r8),%xmm6
+	subq	$64,%rdx
+	movl	%r10d,%eax
+	jbe	.Lcbc_dec_tail
+	jmp	.Lcbc_dec_loop3
+.align	16
+.Lcbc_dec_loop3:
+	movups	(%rdi),%xmm0
+	movups	16(%rdi),%xmm1
+	movups	32(%rdi),%xmm2
+	movaps	%xmm0,%xmm7
+	movaps	%xmm1,%xmm8
+	movaps	%xmm2,%xmm9
+	call	_aesni_decrypt3
+	subq	$48,%rdx
+	leaq	48(%rdi),%rdi
+	leaq	48(%rsi),%rsi
+	pxor	%xmm6,%xmm0
+	pxor	%xmm7,%xmm1
+	movaps	%xmm9,%xmm6
+	pxor	%xmm8,%xmm2
+	movups	%xmm0,-48(%rsi)
+	movl	%r10d,%eax
+	movups	%xmm1,-32(%rsi)
+	movq	%r11,%rcx
+	movups	%xmm2,-16(%rsi)
+	ja	.Lcbc_dec_loop3
+
+.Lcbc_dec_tail:
+	addq	$64,%rdx
+	movups	%xmm6,(%r8)
+	jz	.Lcbc_dec_ret
+
+	movups	(%rdi),%xmm0
+	cmpq	$16,%rdx
+	movaps	%xmm0,%xmm7
+	jbe	.Lcbc_dec_one
+	movups	16(%rdi),%xmm1
+	cmpq	$32,%rdx
+	movaps	%xmm1,%xmm8
+	jbe	.Lcbc_dec_two
+	movups	32(%rdi),%xmm2
+	cmpq	$48,%rdx
+	movaps	%xmm2,%xmm9
+	jbe	.Lcbc_dec_three
+	movups	48(%rdi),%xmm3
+	call	_aesni_decrypt4
+	pxor	%xmm6,%xmm0
+	movups	48(%rdi),%xmm6
+	pxor	%xmm7,%xmm1
+	movups	%xmm0,(%rsi)
+	pxor	%xmm8,%xmm2
+	movups	%xmm1,16(%rsi)
+	pxor	%xmm9,%xmm3
+	movups	%xmm2,32(%rsi)
+	movaps	%xmm3,%xmm0
+	leaq	48(%rsi),%rsi
+	jmp	.Lcbc_dec_tail_collected
+.align	16
+.Lcbc_dec_one:
+	movaps	(%rcx),%xmm4
+	movaps	16(%rcx),%xmm5
+	leaq	32(%rcx),%rcx
+	pxor	%xmm4,%xmm0
+.Loop_dec1_6:
+.byte	102,15,56,222,197
+	decl	%eax
+	movaps	(%rcx),%xmm5
+	leaq	16(%rcx),%rcx
+	jnz	.Loop_dec1_6	
+.byte	102,15,56,223,197
+	pxor	%xmm6,%xmm0
+	movaps	%xmm7,%xmm6
+	jmp	.Lcbc_dec_tail_collected
+.align	16
+.Lcbc_dec_two:
+	call	_aesni_decrypt3
+	pxor	%xmm6,%xmm0
+	pxor	%xmm7,%xmm1
+	movups	%xmm0,(%rsi)
+	movaps	%xmm8,%xmm6
+	movaps	%xmm1,%xmm0
+	leaq	16(%rsi),%rsi
+	jmp	.Lcbc_dec_tail_collected
+.align	16
+.Lcbc_dec_three:
+	call	_aesni_decrypt3
+	pxor	%xmm6,%xmm0
+	pxor	%xmm7,%xmm1
+	movups	%xmm0,(%rsi)
+	pxor	%xmm8,%xmm2
+	movups	%xmm1,16(%rsi)
+	movaps	%xmm9,%xmm6
+	movaps	%xmm2,%xmm0
+	leaq	32(%rsi),%rsi
+	jmp	.Lcbc_dec_tail_collected
+.align	16
+.Lcbc_dec_tail_collected:
+	andq	$15,%rdx
+	movups	%xmm6,(%r8)
+	jnz	.Lcbc_dec_tail_partial
+	movups	%xmm0,(%rsi)
+	jmp	.Lcbc_dec_ret
+.Lcbc_dec_tail_partial:
+	movaps	%xmm0,-24(%rsp)
+	movq	%rsi,%rdi
+	movq	%rdx,%rcx
+	leaq	-24(%rsp),%rsi
+.long	0x9066A4F3	
+
+.Lcbc_dec_ret:
+.Lcbc_ret:
+	.byte	0xf3,0xc3
+.size	aesni_cbc_encrypt,.-aesni_cbc_encrypt
+.globl	aesni_set_decrypt_key
+.type	aesni_set_decrypt_key,@function
+.align	16
+aesni_set_decrypt_key:
+.byte	0x48,0x83,0xEC,0x08	
+	call	_aesni_set_encrypt_key
+	shll	$4,%esi
+	testl	%eax,%eax
+	jnz	.Ldec_key_ret
+	leaq	16(%rdx,%rsi,1),%rdi
+
+	movaps	(%rdx),%xmm0
+	movaps	(%rdi),%xmm1
+	movaps	%xmm0,(%rdi)
+	movaps	%xmm1,(%rdx)
+	leaq	16(%rdx),%rdx
+	leaq	-16(%rdi),%rdi
+
+.Ldec_key_inverse:
+	movaps	(%rdx),%xmm0
+	movaps	(%rdi),%xmm1
+.byte	102,15,56,219,192
+.byte	102,15,56,219,201
+	leaq	16(%rdx),%rdx
+	leaq	-16(%rdi),%rdi
+	cmpq	%rdx,%rdi
+	movaps	%xmm0,16(%rdi)
+	movaps	%xmm1,-16(%rdx)
+	ja	.Ldec_key_inverse
+
+	movaps	(%rdx),%xmm0
+.byte	102,15,56,219,192
+	movaps	%xmm0,(%rdi)
+.Ldec_key_ret:
+	addq	$8,%rsp
+	.byte	0xf3,0xc3
+.LSEH_end_set_decrypt_key:
+.size	aesni_set_decrypt_key,.-aesni_set_decrypt_key
+.globl	aesni_set_encrypt_key
+.type	aesni_set_encrypt_key,@function
+.align	16
+aesni_set_encrypt_key:
+_aesni_set_encrypt_key:
+.byte	0x48,0x83,0xEC,0x08	
+	testq	%rdi,%rdi
+	movq	$-1,%rax
+	jz	.Lenc_key_ret
+	testq	%rdx,%rdx
+	jz	.Lenc_key_ret
+
+	movups	(%rdi),%xmm0
+	pxor	%xmm4,%xmm4
+	leaq	16(%rdx),%rax
+	cmpl	$256,%esi
+	je	.L14rounds
+	cmpl	$192,%esi
+	je	.L12rounds
+	cmpl	$128,%esi
+	jne	.Lbad_keybits
+
+.L10rounds:
+	movl	$9,%esi
+	movaps	%xmm0,(%rdx)
+.byte	102,15,58,223,200,1
+	call	.Lkey_expansion_128_cold
+.byte	102,15,58,223,200,2
+	call	.Lkey_expansion_128
+.byte	102,15,58,223,200,4
+	call	.Lkey_expansion_128
+.byte	102,15,58,223,200,8
+	call	.Lkey_expansion_128
+.byte	102,15,58,223,200,16
+	call	.Lkey_expansion_128
+.byte	102,15,58,223,200,32
+	call	.Lkey_expansion_128
+.byte	102,15,58,223,200,64
+	call	.Lkey_expansion_128
+.byte	102,15,58,223,200,128
+	call	.Lkey_expansion_128
+.byte	102,15,58,223,200,27
+	call	.Lkey_expansion_128
+.byte	102,15,58,223,200,54
+	call	.Lkey_expansion_128
+	movaps	%xmm0,(%rax)
+	movl	%esi,80(%rax)
+	xorl	%eax,%eax
+	jmp	.Lenc_key_ret
+
+.align	16
+.L12rounds:
+	movq	16(%rdi),%xmm2
+	movl	$11,%esi
+	movaps	%xmm0,(%rdx)
+.byte	102,15,58,223,202,1
+	call	.Lkey_expansion_192a_cold
+.byte	102,15,58,223,202,2
+	call	.Lkey_expansion_192b
+.byte	102,15,58,223,202,4
+	call	.Lkey_expansion_192a
+.byte	102,15,58,223,202,8
+	call	.Lkey_expansion_192b
+.byte	102,15,58,223,202,16
+	call	.Lkey_expansion_192a
+.byte	102,15,58,223,202,32
+	call	.Lkey_expansion_192b
+.byte	102,15,58,223,202,64
+	call	.Lkey_expansion_192a
+.byte	102,15,58,223,202,128
+	call	.Lkey_expansion_192b
+	movaps	%xmm0,(%rax)
+	movl	%esi,48(%rax)
+	xorq	%rax,%rax
+	jmp	.Lenc_key_ret
+
+.align	16
+.L14rounds:
+	movups	16(%rdi),%xmm2
+	movl	$13,%esi
+	leaq	16(%rax),%rax
+	movaps	%xmm0,(%rdx)
+	movaps	%xmm2,16(%rdx)
+.byte	102,15,58,223,202,1
+	call	.Lkey_expansion_256a_cold
+.byte	102,15,58,223,200,1
+	call	.Lkey_expansion_256b
+.byte	102,15,58,223,202,2
+	call	.Lkey_expansion_256a
+.byte	102,15,58,223,200,2
+	call	.Lkey_expansion_256b
+.byte	102,15,58,223,202,4
+	call	.Lkey_expansion_256a
+.byte	102,15,58,223,200,4
+	call	.Lkey_expansion_256b
+.byte	102,15,58,223,202,8
+	call	.Lkey_expansion_256a
+.byte	102,15,58,223,200,8
+	call	.Lkey_expansion_256b
+.byte	102,15,58,223,202,16
+	call	.Lkey_expansion_256a
+.byte	102,15,58,223,200,16
+	call	.Lkey_expansion_256b
+.byte	102,15,58,223,202,32
+	call	.Lkey_expansion_256a
+.byte	102,15,58,223,200,32
+	call	.Lkey_expansion_256b
+.byte	102,15,58,223,202,64
+	call	.Lkey_expansion_256a
+	movaps	%xmm0,(%rax)
+	movl	%esi,16(%rax)
+	xorq	%rax,%rax
+	jmp	.Lenc_key_ret
+
+.align	16
+.Lbad_keybits:
+	movq	$-2,%rax
+.Lenc_key_ret:
+	addq	$8,%rsp
+	.byte	0xf3,0xc3
+.LSEH_end_set_encrypt_key:
+
+.align	16
+.Lkey_expansion_128:
+	movaps	%xmm0,(%rax)
+	leaq	16(%rax),%rax
+.Lkey_expansion_128_cold:
+	shufps	$16,%xmm0,%xmm4
+	pxor	%xmm4,%xmm0
+	shufps	$140,%xmm0,%xmm4
+	pxor	%xmm4,%xmm0
+	pshufd	$255,%xmm1,%xmm1
+	pxor	%xmm1,%xmm0
+	.byte	0xf3,0xc3
+
+.align	16
+.Lkey_expansion_192a:
+	movaps	%xmm0,(%rax)
+	leaq	16(%rax),%rax
+.Lkey_expansion_192a_cold:
+	movaps	%xmm2,%xmm5
+.Lkey_expansion_192b_warm:
+	shufps	$16,%xmm0,%xmm4
+	movaps	%xmm2,%xmm3
+	pxor	%xmm4,%xmm0
+	shufps	$140,%xmm0,%xmm4
+	pslldq	$4,%xmm3
+	pxor	%xmm4,%xmm0
+	pshufd	$85,%xmm1,%xmm1
+	pxor	%xmm3,%xmm2
+	pxor	%xmm1,%xmm0
+	pshufd	$255,%xmm0,%xmm3
+	pxor	%xmm3,%xmm2
+	.byte	0xf3,0xc3
+
+.align	16
+.Lkey_expansion_192b:
+	movaps	%xmm0,%xmm3
+	shufps	$68,%xmm0,%xmm5
+	movaps	%xmm5,(%rax)
+	shufps	$78,%xmm2,%xmm3
+	movaps	%xmm3,16(%rax)
+	leaq	32(%rax),%rax
+	jmp	.Lkey_expansion_192b_warm
+
+.align	16
+.Lkey_expansion_256a:
+	movaps	%xmm2,(%rax)
+	leaq	16(%rax),%rax
+.Lkey_expansion_256a_cold:
+	shufps	$16,%xmm0,%xmm4
+	pxor	%xmm4,%xmm0
+	shufps	$140,%xmm0,%xmm4
+	pxor	%xmm4,%xmm0
+	pshufd	$255,%xmm1,%xmm1
+	pxor	%xmm1,%xmm0
+	.byte	0xf3,0xc3
+
+.align	16
+.Lkey_expansion_256b:
+	movaps	%xmm0,(%rax)
+	leaq	16(%rax),%rax
+
+	shufps	$16,%xmm2,%xmm4
+	pxor	%xmm4,%xmm2
+	shufps	$140,%xmm2,%xmm4
+	pxor	%xmm4,%xmm2
+	pshufd	$170,%xmm1,%xmm1
+	pxor	%xmm1,%xmm2
+	.byte	0xf3,0xc3
+.size	aesni_set_encrypt_key,.-aesni_set_encrypt_key
+.byte	65,69,83,32,102,111,114,32,73,110,116,101,108,32,65,69,83,45,78,73,44,32,67,82,89,80,84,79,71,65,77,83,32,98,121,32,60,97,112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103,62,0
+.align	64
diff --git a/lib/accelerated/intel/asm/appro-aes-x86.s b/lib/accelerated/intel/asm/appro-aes-x86.s
new file mode 100644
index 0000000000..981e356e9a
--- /dev/null
+++ b/lib/accelerated/intel/asm/appro-aes-x86.s
@@ -0,0 +1,798 @@
+# Copyright (c) 2006, Andy Polyakov by <appro@openssl.org>
+# All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions
+# are met:
+# 
+#     *	Redistributions of source code must retain copyright notices,
+#	this list of conditions and the following disclaimer.
+#
+#     *	Redistributions in binary form must reproduce the above
+# 	copyright notice, this list of conditions and the following
+#	disclaimer in the documentation and/or other materials
+#	provided with the distribution.
+#
+#     *	Neither the name of the Andy Polyakov nor the names of its
+#	copyright holder and contributors may be used to endorse or
+#	promote products derived from this software without specific
+#	prior written permission.
+#
+# ALTERNATIVELY, provided that this notice is retained in full, this
+# product may be distributed under the terms of the GNU General Public
+# License (GPL), in which case the provisions of the GPL apply INSTEAD OF
+# those given above.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER AND CONTRIBUTORS
+# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+.file	"aesni-x86.s"
+.text
+.globl	aesni_encrypt
+.type	aesni_encrypt,@function
+.align	16
+aesni_encrypt:
+.L_aesni_encrypt_begin:
+	movl	4(%esp),%eax
+	movl	12(%esp),%edx
+	movups	(%eax),%xmm0
+	movl	240(%edx),%ecx
+	movl	8(%esp),%eax
+	movups	(%edx),%xmm3
+	movups	16(%edx),%xmm4
+	leal	32(%edx),%edx
+	pxor	%xmm3,%xmm0
+.L000enc1_loop:
+	aesenc	%xmm4,%xmm0
+	decl	%ecx
+	movups	(%edx),%xmm4
+	leal	16(%edx),%edx
+	jnz	.L000enc1_loop
+	aesenclast	%xmm4,%xmm0
+	movups	%xmm0,(%eax)
+	ret
+.size	aesni_encrypt,.-.L_aesni_encrypt_begin
+.globl	aesni_decrypt
+.type	aesni_decrypt,@function
+.align	16
+aesni_decrypt:
+.L_aesni_decrypt_begin:
+	movl	4(%esp),%eax
+	movl	12(%esp),%edx
+	movups	(%eax),%xmm0
+	movl	240(%edx),%ecx
+	movl	8(%esp),%eax
+	movups	(%edx),%xmm3
+	movups	16(%edx),%xmm4
+	leal	32(%edx),%edx
+	pxor	%xmm3,%xmm0
+.L001dec1_loop:
+	aesdec	%xmm4,%xmm0
+	decl	%ecx
+	movups	(%edx),%xmm4
+	leal	16(%edx),%edx
+	jnz	.L001dec1_loop
+	aesdeclast	%xmm4,%xmm0
+	movups	%xmm0,(%eax)
+	ret
+.size	aesni_decrypt,.-.L_aesni_decrypt_begin
+.type	_aesni_encrypt3,@function
+.align	16
+_aesni_encrypt3:
+	movups	(%edx),%xmm3
+	shrl	$1,%ecx
+	movups	16(%edx),%xmm4
+	leal	32(%edx),%edx
+	pxor	%xmm3,%xmm0
+	pxor	%xmm3,%xmm1
+	pxor	%xmm3,%xmm2
+	jmp	.L002enc3_loop
+.align	16
+.L002enc3_loop:
+	aesenc	%xmm4,%xmm0
+	movups	(%edx),%xmm3
+	aesenc	%xmm4,%xmm1
+	decl	%ecx
+	aesenc	%xmm4,%xmm2
+	movups	16(%edx),%xmm4
+	aesenc	%xmm3,%xmm0
+	leal	32(%edx),%edx
+	aesenc	%xmm3,%xmm1
+	aesenc	%xmm3,%xmm2
+	jnz	.L002enc3_loop
+	aesenc	%xmm4,%xmm0
+	movups	(%edx),%xmm3
+	aesenc	%xmm4,%xmm1
+	aesenc	%xmm4,%xmm2
+	aesenclast	%xmm3,%xmm0
+	aesenclast	%xmm3,%xmm1
+	aesenclast	%xmm3,%xmm2
+	ret
+.size	_aesni_encrypt3,.-_aesni_encrypt3
+.type	_aesni_decrypt3,@function
+.align	16
+_aesni_decrypt3:
+	movups	(%edx),%xmm3
+	shrl	$1,%ecx
+	movups	16(%edx),%xmm4
+	leal	32(%edx),%edx
+	pxor	%xmm3,%xmm0
+	pxor	%xmm3,%xmm1
+	pxor	%xmm3,%xmm2
+	jmp	.L003dec3_loop
+.align	16
+.L003dec3_loop:
+	aesdec	%xmm4,%xmm0
+	movups	(%edx),%xmm3
+	aesdec	%xmm4,%xmm1
+	decl	%ecx
+	aesdec	%xmm4,%xmm2
+	movups	16(%edx),%xmm4
+	aesdec	%xmm3,%xmm0
+	leal	32(%edx),%edx
+	aesdec	%xmm3,%xmm1
+	aesdec	%xmm3,%xmm2
+	jnz	.L003dec3_loop
+	aesdec	%xmm4,%xmm0
+	movups	(%edx),%xmm3
+	aesdec	%xmm4,%xmm1
+	aesdec	%xmm4,%xmm2
+	aesdeclast	%xmm3,%xmm0
+	aesdeclast	%xmm3,%xmm1
+	aesdeclast	%xmm3,%xmm2
+	ret
+.size	_aesni_decrypt3,.-_aesni_decrypt3
+.type	_aesni_encrypt4,@function
+.align	16
+_aesni_encrypt4:
+	movups	(%edx),%xmm3
+	movups	16(%edx),%xmm4
+	shrl	$1,%ecx
+	leal	32(%edx),%edx
+	pxor	%xmm3,%xmm0
+	pxor	%xmm3,%xmm1
+	pxor	%xmm3,%xmm2
+	pxor	%xmm3,%xmm7
+	jmp	.L004enc3_loop
+.align	16
+.L004enc3_loop:
+	aesenc	%xmm4,%xmm0
+	movups	(%edx),%xmm3
+	aesenc	%xmm4,%xmm1
+	decl	%ecx
+	aesenc	%xmm4,%xmm2
+	aesenc	%xmm4,%xmm7
+	movups	16(%edx),%xmm4
+	aesenc	%xmm3,%xmm0
+	leal	32(%edx),%edx
+	aesenc	%xmm3,%xmm1
+	aesenc	%xmm3,%xmm2
+	aesenc	%xmm3,%xmm7
+	jnz	.L004enc3_loop
+	aesenc	%xmm4,%xmm0
+	movups	(%edx),%xmm3
+	aesenc	%xmm4,%xmm1
+	aesenc	%xmm4,%xmm2
+	aesenc	%xmm4,%xmm7
+	aesenclast	%xmm3,%xmm0
+	aesenclast	%xmm3,%xmm1
+	aesenclast	%xmm3,%xmm2
+	aesenclast	%xmm3,%xmm7
+	ret
+.size	_aesni_encrypt4,.-_aesni_encrypt4
+.type	_aesni_decrypt4,@function
+.align	16
+_aesni_decrypt4:
+	movups	(%edx),%xmm3
+	movups	16(%edx),%xmm4
+	shrl	$1,%ecx
+	leal	32(%edx),%edx
+	pxor	%xmm3,%xmm0
+	pxor	%xmm3,%xmm1
+	pxor	%xmm3,%xmm2
+	pxor	%xmm3,%xmm7
+	jmp	.L005dec3_loop
+.align	16
+.L005dec3_loop:
+	aesdec	%xmm4,%xmm0
+	movups	(%edx),%xmm3
+	aesdec	%xmm4,%xmm1
+	decl	%ecx
+	aesdec	%xmm4,%xmm2
+	aesdec	%xmm4,%xmm7
+	movups	16(%edx),%xmm4
+	aesdec	%xmm3,%xmm0
+	leal	32(%edx),%edx
+	aesdec	%xmm3,%xmm1
+	aesdec	%xmm3,%xmm2
+	aesdec	%xmm3,%xmm7
+	jnz	.L005dec3_loop
+	aesdec	%xmm4,%xmm0
+	movups	(%edx),%xmm3
+	aesdec	%xmm4,%xmm1
+	aesdec	%xmm4,%xmm2
+	aesdec	%xmm4,%xmm7
+	aesdeclast	%xmm3,%xmm0
+	aesdeclast	%xmm3,%xmm1
+	aesdeclast	%xmm3,%xmm2
+	aesdeclast	%xmm3,%xmm7
+	ret
+.size	_aesni_decrypt4,.-_aesni_decrypt4
+.globl	aesni_ecb_encrypt
+.type	aesni_ecb_encrypt,@function
+.align	16
+aesni_ecb_encrypt:
+.L_aesni_ecb_encrypt_begin:
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	movl	20(%esp),%esi
+	movl	24(%esp),%edi
+	movl	28(%esp),%eax
+	movl	32(%esp),%edx
+	movl	36(%esp),%ecx
+	cmpl	$16,%eax
+	jb	.L006ecb_ret
+	andl	$-16,%eax
+	testl	%ecx,%ecx
+	movl	240(%edx),%ecx
+	movl	%edx,%ebp
+	movl	%ecx,%ebx
+	jz	.L007ecb_decrypt
+	subl	$64,%eax
+	jbe	.L008ecb_enc_tail
+	jmp	.L009ecb_enc_loop3
+.align	16
+.L009ecb_enc_loop3:
+	movups	(%esi),%xmm0
+	movups	16(%esi),%xmm1
+	movups	32(%esi),%xmm2
+	call	_aesni_encrypt3
+	subl	$48,%eax
+	leal	48(%esi),%esi
+	leal	48(%edi),%edi
+	movups	%xmm0,-48(%edi)
+	movl	%ebp,%edx
+	movups	%xmm1,-32(%edi)
+	movl	%ebx,%ecx
+	movups	%xmm2,-16(%edi)
+	ja	.L009ecb_enc_loop3
+.L008ecb_enc_tail:
+	addl	$64,%eax
+	jz	.L006ecb_ret
+	cmpl	$16,%eax
+	movups	(%esi),%xmm0
+	je	.L010ecb_enc_one
+	cmpl	$32,%eax
+	movups	16(%esi),%xmm1
+	je	.L011ecb_enc_two
+	cmpl	$48,%eax
+	movups	32(%esi),%xmm2
+	je	.L012ecb_enc_three
+	movups	48(%esi),%xmm7
+	call	_aesni_encrypt4
+	movups	%xmm0,(%edi)
+	movups	%xmm1,16(%edi)
+	movups	%xmm2,32(%edi)
+	movups	%xmm7,48(%edi)
+	jmp	.L006ecb_ret
+.align	16
+.L010ecb_enc_one:
+	movups	(%edx),%xmm3
+	movups	16(%edx),%xmm4
+	leal	32(%edx),%edx
+	pxor	%xmm3,%xmm0
+.L013enc1_loop:
+	aesenc	%xmm4,%xmm0
+	decl	%ecx
+	movups	(%edx),%xmm4
+	leal	16(%edx),%edx
+	jnz	.L013enc1_loop
+	aesenclast	%xmm4,%xmm0
+	movups	%xmm0,(%edi)
+	jmp	.L006ecb_ret
+.align	16
+.L011ecb_enc_two:
+	call	_aesni_encrypt3
+	movups	%xmm0,(%edi)
+	movups	%xmm1,16(%edi)
+	jmp	.L006ecb_ret
+.align	16
+.L012ecb_enc_three:
+	call	_aesni_encrypt3
+	movups	%xmm0,(%edi)
+	movups	%xmm1,16(%edi)
+	movups	%xmm2,32(%edi)
+	jmp	.L006ecb_ret
+.align	16
+.L007ecb_decrypt:
+	subl	$64,%eax
+	jbe	.L014ecb_dec_tail
+	jmp	.L015ecb_dec_loop3
+.align	16
+.L015ecb_dec_loop3:
+	movups	(%esi),%xmm0
+	movups	16(%esi),%xmm1
+	movups	32(%esi),%xmm2
+	call	_aesni_decrypt3
+	subl	$48,%eax
+	leal	48(%esi),%esi
+	leal	48(%edi),%edi
+	movups	%xmm0,-48(%edi)
+	movl	%ebp,%edx
+	movups	%xmm1,-32(%edi)
+	movl	%ebx,%ecx
+	movups	%xmm2,-16(%edi)
+	ja	.L015ecb_dec_loop3
+.L014ecb_dec_tail:
+	addl	$64,%eax
+	jz	.L006ecb_ret
+	cmpl	$16,%eax
+	movups	(%esi),%xmm0
+	je	.L016ecb_dec_one
+	cmpl	$32,%eax
+	movups	16(%esi),%xmm1
+	je	.L017ecb_dec_two
+	cmpl	$48,%eax
+	movups	32(%esi),%xmm2
+	je	.L018ecb_dec_three
+	movups	48(%esi),%xmm7
+	call	_aesni_decrypt4
+	movups	%xmm0,(%edi)
+	movups	%xmm1,16(%edi)
+	movups	%xmm2,32(%edi)
+	movups	%xmm7,48(%edi)
+	jmp	.L006ecb_ret
+.align	16
+.L016ecb_dec_one:
+	movups	(%edx),%xmm3
+	movups	16(%edx),%xmm4
+	leal	32(%edx),%edx
+	pxor	%xmm3,%xmm0
+.L019dec1_loop:
+	aesdec	%xmm4,%xmm0
+	decl	%ecx
+	movups	(%edx),%xmm4
+	leal	16(%edx),%edx
+	jnz	.L019dec1_loop
+	aesdeclast	%xmm4,%xmm0
+	movups	%xmm0,(%edi)
+	jmp	.L006ecb_ret
+.align	16
+.L017ecb_dec_two:
+	call	_aesni_decrypt3
+	movups	%xmm0,(%edi)
+	movups	%xmm1,16(%edi)
+	jmp	.L006ecb_ret
+.align	16
+.L018ecb_dec_three:
+	call	_aesni_decrypt3
+	movups	%xmm0,(%edi)
+	movups	%xmm1,16(%edi)
+	movups	%xmm2,32(%edi)
+.L006ecb_ret:
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.size	aesni_ecb_encrypt,.-.L_aesni_ecb_encrypt_begin
+.globl	aesni_cbc_encrypt
+.type	aesni_cbc_encrypt,@function
+.align	16
+aesni_cbc_encrypt:
+.L_aesni_cbc_encrypt_begin:
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	movl	20(%esp),%esi
+	movl	24(%esp),%edi
+	movl	28(%esp),%eax
+	movl	32(%esp),%edx
+	testl	%eax,%eax
+	movl	36(%esp),%ebp
+	jz	.L020cbc_ret
+	cmpl	$0,40(%esp)
+	movups	(%ebp),%xmm5
+	movl	240(%edx),%ecx
+	movl	%edx,%ebp
+	movl	%ecx,%ebx
+	je	.L021cbc_decrypt
+	movaps	%xmm5,%xmm0
+	cmpl	$16,%eax
+	jb	.L022cbc_enc_tail
+	subl	$16,%eax
+	jmp	.L023cbc_enc_loop
+.align	16
+.L023cbc_enc_loop:
+	movups	(%esi),%xmm5
+	leal	16(%esi),%esi
+	pxor	%xmm5,%xmm0
+	movups	(%edx),%xmm3
+	movups	16(%edx),%xmm4
+	leal	32(%edx),%edx
+	pxor	%xmm3,%xmm0
+.L024enc1_loop:
+	aesenc	%xmm4,%xmm0
+	decl	%ecx
+	movups	(%edx),%xmm4
+	leal	16(%edx),%edx
+	jnz	.L024enc1_loop
+	aesenclast	%xmm4,%xmm0
+	subl	$16,%eax
+	leal	16(%edi),%edi
+	movl	%ebx,%ecx
+	movl	%ebp,%edx
+	movups	%xmm0,-16(%edi)
+	jnc	.L023cbc_enc_loop
+	addl	$16,%eax
+	jnz	.L022cbc_enc_tail
+	movaps	%xmm0,%xmm5
+	jmp	.L020cbc_ret
+.L022cbc_enc_tail:
+	movl	%eax,%ecx
+.long	2767451785
+	movl	$16,%ecx
+	subl	%eax,%ecx
+	xorl	%eax,%eax
+.long	2868115081
+	leal	-16(%edi),%edi
+	movl	%ebx,%ecx
+	movl	%edi,%esi
+	movl	%ebp,%edx
+	jmp	.L023cbc_enc_loop
+.align	16
+.L021cbc_decrypt:
+	subl	$64,%eax
+	jbe	.L025cbc_dec_tail
+	jmp	.L026cbc_dec_loop3
+.align	16
+.L026cbc_dec_loop3:
+	movups	(%esi),%xmm0
+	movups	16(%esi),%xmm1
+	movups	32(%esi),%xmm2
+	movaps	%xmm0,%xmm6
+	movaps	%xmm1,%xmm7
+	call	_aesni_decrypt3
+	subl	$48,%eax
+	leal	48(%esi),%esi
+	leal	48(%edi),%edi
+	pxor	%xmm5,%xmm0
+	pxor	%xmm6,%xmm1
+	movups	-16(%esi),%xmm5
+	pxor	%xmm7,%xmm2
+	movups	%xmm0,-48(%edi)
+	movl	%ebx,%ecx
+	movups	%xmm1,-32(%edi)
+	movl	%ebp,%edx
+	movups	%xmm2,-16(%edi)
+	ja	.L026cbc_dec_loop3
+.L025cbc_dec_tail:
+	addl	$64,%eax
+	jz	.L020cbc_ret
+	movups	(%esi),%xmm0
+	cmpl	$16,%eax
+	movaps	%xmm0,%xmm6
+	jbe	.L027cbc_dec_one
+	movups	16(%esi),%xmm1
+	cmpl	$32,%eax
+	movaps	%xmm1,%xmm7
+	jbe	.L028cbc_dec_two
+	movups	32(%esi),%xmm2
+	cmpl	$48,%eax
+	jbe	.L029cbc_dec_three
+	movups	48(%esi),%xmm7
+	call	_aesni_decrypt4
+	movups	16(%esi),%xmm3
+	movups	32(%esi),%xmm4
+	pxor	%xmm5,%xmm0
+	pxor	%xmm6,%xmm1
+	movups	48(%esi),%xmm5
+	movups	%xmm0,(%edi)
+	pxor	%xmm3,%xmm2
+	pxor	%xmm4,%xmm7
+	movups	%xmm1,16(%edi)
+	movups	%xmm2,32(%edi)
+	movaps	%xmm7,%xmm0
+	leal	48(%edi),%edi
+	jmp	.L030cbc_dec_tail_collected
+.L027cbc_dec_one:
+	movups	(%edx),%xmm3
+	movups	16(%edx),%xmm4
+	leal	32(%edx),%edx
+	pxor	%xmm3,%xmm0
+.L031dec1_loop:
+	aesdec	%xmm4,%xmm0
+	decl	%ecx
+	movups	(%edx),%xmm4
+	leal	16(%edx),%edx
+	jnz	.L031dec1_loop
+	aesdeclast	%xmm4,%xmm0
+	pxor	%xmm5,%xmm0
+	movaps	%xmm6,%xmm5
+	jmp	.L030cbc_dec_tail_collected
+.L028cbc_dec_two:
+	call	_aesni_decrypt3
+	pxor	%xmm5,%xmm0
+	pxor	%xmm6,%xmm1
+	movups	%xmm0,(%edi)
+	movaps	%xmm1,%xmm0
+	movaps	%xmm7,%xmm5
+	leal	16(%edi),%edi
+	jmp	.L030cbc_dec_tail_collected
+.L029cbc_dec_three:
+	call	_aesni_decrypt3
+	pxor	%xmm5,%xmm0
+	pxor	%xmm6,%xmm1
+	pxor	%xmm7,%xmm2
+	movups	%xmm0,(%edi)
+	movups	%xmm1,16(%edi)
+	movaps	%xmm2,%xmm0
+	movups	32(%esi),%xmm5
+	leal	32(%edi),%edi
+.L030cbc_dec_tail_collected:
+	andl	$15,%eax
+	jnz	.L032cbc_dec_tail_partial
+	movups	%xmm0,(%edi)
+	jmp	.L020cbc_ret
+.L032cbc_dec_tail_partial:
+	movl	%esp,%ebp
+	subl	$16,%esp
+	andl	$-16,%esp
+	movaps	%xmm0,(%esp)
+	movl	%esp,%esi
+	movl	%eax,%ecx
+.long	2767451785
+	movl	%ebp,%esp
+.L020cbc_ret:
+	movl	36(%esp),%ebp
+	movups	%xmm5,(%ebp)
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.size	aesni_cbc_encrypt,.-.L_aesni_cbc_encrypt_begin
+.type	_aesni_set_encrypt_key,@function
+.align	16
+_aesni_set_encrypt_key:
+	testl	%eax,%eax
+	jz	.L033bad_pointer
+	testl	%edx,%edx
+	jz	.L033bad_pointer
+	movups	(%eax),%xmm0
+	pxor	%xmm4,%xmm4
+	leal	16(%edx),%edx
+	cmpl	$256,%ecx
+	je	.L03414rounds
+	cmpl	$192,%ecx
+	je	.L03512rounds
+	cmpl	$128,%ecx
+	jne	.L036bad_keybits
+.align	16
+.L03710rounds:
+	movl	$9,%ecx
+	movups	%xmm0,-16(%edx)
+	aeskeygenassist	$1,%xmm0,%xmm1
+	call	.L038key_128_cold
+	aeskeygenassist	$2,%xmm0,%xmm1
+	call	.L039key_128
+	aeskeygenassist	$4,%xmm0,%xmm1
+	call	.L039key_128
+	aeskeygenassist	$8,%xmm0,%xmm1
+	call	.L039key_128
+	aeskeygenassist	$16,%xmm0,%xmm1
+	call	.L039key_128
+	aeskeygenassist	$32,%xmm0,%xmm1
+	call	.L039key_128
+	aeskeygenassist	$64,%xmm0,%xmm1
+	call	.L039key_128
+	aeskeygenassist	$128,%xmm0,%xmm1
+	call	.L039key_128
+	aeskeygenassist	$27,%xmm0,%xmm1
+	call	.L039key_128
+	aeskeygenassist	$54,%xmm0,%xmm1
+	call	.L039key_128
+	movups	%xmm0,(%edx)
+	movl	%ecx,80(%edx)
+	xorl	%eax,%eax
+	ret
+.align	16
+.L039key_128:
+	movups	%xmm0,(%edx)
+	leal	16(%edx),%edx
+.L038key_128_cold:
+	shufps	$16,%xmm0,%xmm4
+	pxor	%xmm4,%xmm0
+	shufps	$140,%xmm0,%xmm4
+	pxor	%xmm4,%xmm0
+	pshufd	$255,%xmm1,%xmm1
+	pxor	%xmm1,%xmm0
+	ret
+.align	16
+.L03512rounds:
+	movq	16(%eax),%xmm2
+	movl	$11,%ecx
+	movups	%xmm0,-16(%edx)
+	aeskeygenassist	$1,%xmm2,%xmm1
+	call	.L040key_192a_cold
+	aeskeygenassist	$2,%xmm2,%xmm1
+	call	.L041key_192b
+	aeskeygenassist	$4,%xmm2,%xmm1
+	call	.L042key_192a
+	aeskeygenassist	$8,%xmm2,%xmm1
+	call	.L041key_192b
+	aeskeygenassist	$16,%xmm2,%xmm1
+	call	.L042key_192a
+	aeskeygenassist	$32,%xmm2,%xmm1
+	call	.L041key_192b
+	aeskeygenassist	$64,%xmm2,%xmm1
+	call	.L042key_192a
+	aeskeygenassist	$128,%xmm2,%xmm1
+	call	.L041key_192b
+	movups	%xmm0,(%edx)
+	movl	%ecx,48(%edx)
+	xorl	%eax,%eax
+	ret
+.align	16
+.L042key_192a:
+	movups	%xmm0,(%edx)
+	leal	16(%edx),%edx
+.align	16
+.L040key_192a_cold:
+	movaps	%xmm2,%xmm5
+.L043key_192b_warm:
+	shufps	$16,%xmm0,%xmm4
+	movaps	%xmm2,%xmm3
+	pxor	%xmm4,%xmm0
+	shufps	$140,%xmm0,%xmm4
+	pslldq	$4,%xmm3
+	pxor	%xmm4,%xmm0
+	pshufd	$85,%xmm1,%xmm1
+	pxor	%xmm3,%xmm2
+	pxor	%xmm1,%xmm0
+	pshufd	$255,%xmm0,%xmm3
+	pxor	%xmm3,%xmm2
+	ret
+.align	16
+.L041key_192b:
+	movaps	%xmm0,%xmm3
+	shufps	$68,%xmm0,%xmm5
+	movups	%xmm5,(%edx)
+	shufps	$78,%xmm2,%xmm3
+	movups	%xmm3,16(%edx)
+	leal	32(%edx),%edx
+	jmp	.L043key_192b_warm
+.align	16
+.L03414rounds:
+	movups	16(%eax),%xmm2
+	movl	$13,%ecx
+	leal	16(%edx),%edx
+	movups	%xmm0,-32(%edx)
+	movups	%xmm2,-16(%edx)
+	aeskeygenassist	$1,%xmm2,%xmm1
+	call	.L044key_256a_cold
+	aeskeygenassist	$1,%xmm0,%xmm1
+	call	.L045key_256b
+	aeskeygenassist	$2,%xmm2,%xmm1
+	call	.L046key_256a
+	aeskeygenassist	$2,%xmm0,%xmm1
+	call	.L045key_256b
+	aeskeygenassist	$4,%xmm2,%xmm1
+	call	.L046key_256a
+	aeskeygenassist	$4,%xmm0,%xmm1
+	call	.L045key_256b
+	aeskeygenassist	$8,%xmm2,%xmm1
+	call	.L046key_256a
+	aeskeygenassist	$8,%xmm0,%xmm1
+	call	.L045key_256b
+	aeskeygenassist	$16,%xmm2,%xmm1
+	call	.L046key_256a
+	aeskeygenassist	$16,%xmm0,%xmm1
+	call	.L045key_256b
+	aeskeygenassist	$32,%xmm2,%xmm1
+	call	.L046key_256a
+	aeskeygenassist	$32,%xmm0,%xmm1
+	call	.L045key_256b
+	aeskeygenassist	$64,%xmm2,%xmm1
+	call	.L046key_256a
+	movups	%xmm0,(%edx)
+	movl	%ecx,16(%edx)
+	xorl	%eax,%eax
+	ret
+.align	16
+.L046key_256a:
+	movups	%xmm2,(%edx)
+	leal	16(%edx),%edx
+.L044key_256a_cold:
+	shufps	$16,%xmm0,%xmm4
+	pxor	%xmm4,%xmm0
+	shufps	$140,%xmm0,%xmm4
+	pxor	%xmm4,%xmm0
+	pshufd	$255,%xmm1,%xmm1
+	pxor	%xmm1,%xmm0
+	ret
+.align	16
+.L045key_256b:
+	movups	%xmm0,(%edx)
+	leal	16(%edx),%edx
+	shufps	$16,%xmm2,%xmm4
+	pxor	%xmm4,%xmm2
+	shufps	$140,%xmm2,%xmm4
+	pxor	%xmm4,%xmm2
+	pshufd	$170,%xmm1,%xmm1
+	pxor	%xmm1,%xmm2
+	ret
+.align	4
+.L033bad_pointer:
+	movl	$-1,%eax
+	ret
+.align	4
+.L036bad_keybits:
+	movl	$-2,%eax
+	ret
+.size	_aesni_set_encrypt_key,.-_aesni_set_encrypt_key
+.globl	aesni_set_encrypt_key
+.type	aesni_set_encrypt_key,@function
+.align	16
+aesni_set_encrypt_key:
+.L_aesni_set_encrypt_key_begin:
+	movl	4(%esp),%eax
+	movl	8(%esp),%ecx
+	movl	12(%esp),%edx
+	call	_aesni_set_encrypt_key
+	ret
+.size	aesni_set_encrypt_key,.-.L_aesni_set_encrypt_key_begin
+.globl	aesni_set_decrypt_key
+.type	aesni_set_decrypt_key,@function
+.align	16
+aesni_set_decrypt_key:
+.L_aesni_set_decrypt_key_begin:
+	movl	4(%esp),%eax
+	movl	8(%esp),%ecx
+	movl	12(%esp),%edx
+	call	_aesni_set_encrypt_key
+	movl	12(%esp),%edx
+	shll	$4,%ecx
+	testl	%eax,%eax
+	jnz	.L047dec_key_ret
+	leal	16(%edx,%ecx,1),%eax
+	movups	(%edx),%xmm0
+	movups	(%eax),%xmm1
+	movups	%xmm0,(%eax)
+	movups	%xmm1,(%edx)
+	leal	16(%edx),%edx
+	leal	-16(%eax),%eax
+.L048dec_key_inverse:
+	movups	(%edx),%xmm0
+	movups	(%eax),%xmm1
+	aesimc	%xmm0,%xmm0
+	aesimc	%xmm1,%xmm1
+	leal	16(%edx),%edx
+	leal	-16(%eax),%eax
+	cmpl	%edx,%eax
+	movups	%xmm0,16(%eax)
+	movups	%xmm1,-16(%edx)
+	ja	.L048dec_key_inverse
+	movups	(%edx),%xmm0
+	aesimc	%xmm0,%xmm0
+	movups	%xmm0,(%edx)
+	xorl	%eax,%eax
+.L047dec_key_ret:
+	ret
+.size	aesni_set_decrypt_key,.-.L_aesni_set_decrypt_key_begin
+.byte	65,69,83,32,102,111,114,32,73,110,116,101,108,32,65,69
+.byte	83,45,78,73,44,32,67,82,89,80,84,79,71,65,77,83
+.byte	32,98,121,32,60,97,112,112,114,111,64,111,112,101,110,115
+.byte	115,108,46,111,114,103,62,0
diff --git a/lib/accelerated/intel/asm/x64_iaesx64.s b/lib/accelerated/intel/asm/x64_iaesx64.s
deleted file mode 100755
index 433d1775a8..0000000000
--- a/lib/accelerated/intel/asm/x64_iaesx64.s
+++ /dev/null
@@ -1,2054 +0,0 @@
-[bits 64]
-[CPU intelnop]
-
-; Copyright (c) 2010, Intel Corporation
-; All rights reserved.
-; 
-; Redistribution and use in source and binary forms, with or without 
-; modification, are permitted provided that the following conditions are met:
-; 
-;     * Redistributions of source code must retain the above copyright notice, 
-;       this list of conditions and the following disclaimer.
-;     * Redistributions in binary form must reproduce the above copyright notice, 
-;       this list of conditions and the following disclaimer in the documentation 
-;       and/or other materials provided with the distribution.
-;     * Neither the name of Intel Corporation nor the names of its contributors 
-;       may be used to endorse or promote products derived from this software 
-;       without specific prior written permission.
-; 
-; THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND 
-; ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 
-; WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 
-; IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, 
-; INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, 
-; BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 
-; DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 
-; LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE 
-; OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF 
-; ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-%macro linux_setup 0
-%ifdef __linux__
-	mov rcx, rdi
-	mov rdx, rsi
-%endif
-%endmacro
-
-%macro inversekey 1
-	movdqu  xmm1,%1
-	aesimc	xmm0,xmm1
-	movdqu	%1,xmm0
-%endmacro
-
-%macro aesdeclast1 1
-	aesdeclast	xmm0,%1
-%endmacro
-
-%macro aesenclast1 1
-	aesenclast	xmm0,%1
-%endmacro
-
-%macro aesdec1 1
-	aesdec	xmm0,%1
-%endmacro
-
-%macro aesenc1 1
-	aesenc	xmm0,%1
-%endmacro
-
-
-%macro aesdeclast1_u 1
-	movdqu xmm4,%1
-	aesdeclast	xmm0,xmm4
-%endmacro
-
-%macro aesenclast1_u 1
-	movdqu xmm4,%1
-	aesenclast	xmm0,xmm4
-%endmacro
-
-%macro aesdec1_u 1
-	movdqu xmm4,%1
-	aesdec	xmm0,xmm4
-%endmacro
-
-%macro aesenc1_u 1
-	movdqu xmm4,%1
-	aesenc	xmm0,xmm4
-%endmacro
- 
-%macro aesdec4 1
-	movdqa	xmm4,%1
-
-	aesdec	xmm0,xmm4
-	aesdec	xmm1,xmm4
-	aesdec	xmm2,xmm4
-	aesdec	xmm3,xmm4
-
-%endmacro
-
-%macro aesdeclast4 1
-	movdqa	xmm4,%1
-
-	aesdeclast	xmm0,xmm4
-	aesdeclast	xmm1,xmm4
-	aesdeclast	xmm2,xmm4
-	aesdeclast	xmm3,xmm4
-
-%endmacro
-
-
-%macro aesenc4 1
-	movdqa	xmm4,%1
-
-	aesenc	xmm0,xmm4
-	aesenc	xmm1,xmm4
-	aesenc	xmm2,xmm4
-	aesenc	xmm3,xmm4
-
-%endmacro
-
-%macro aesenclast4 1
-	movdqa	xmm4,%1
-
-	aesenclast	xmm0,xmm4
-	aesenclast	xmm1,xmm4
-	aesenclast	xmm2,xmm4
-	aesenclast	xmm3,xmm4
-
-%endmacro
-
-
-%macro load_and_inc4 1
-	movdqa	xmm4,%1
-	movdqa	xmm0,xmm5
-	movdqa  xmm1,xmm5
-	paddq	xmm1,[counter_add_one wrt rip]
-	movdqa  xmm2,xmm5
-	paddq	xmm2,[counter_add_two wrt rip]
-	movdqa  xmm3,xmm5
-	paddq	xmm3,[counter_add_three wrt rip]
-	pxor	xmm0,xmm4
-	paddq	xmm5,[counter_add_four wrt rip]
-	pxor	xmm1,xmm4
-	pxor	xmm2,xmm4
-	pxor	xmm3,xmm4
-%endmacro
-
-%macro xor_with_input4 1
-	movdqu xmm4,[%1]
-	pxor xmm0,xmm4
-	movdqu xmm4,[%1+16]
-	pxor xmm1,xmm4
-	movdqu xmm4,[%1+32]
-	pxor xmm2,xmm4
-	movdqu xmm4,[%1+48]
-	pxor xmm3,xmm4
-%endmacro
-
-
-
-%macro load_and_xor4 2
-	movdqa	xmm4,%2
-	movdqu	xmm0,[%1 + 0*16]
-	pxor	xmm0,xmm4
-	movdqu	xmm1,[%1 + 1*16]
-	pxor	xmm1,xmm4
-	movdqu	xmm2,[%1 + 2*16]
-	pxor	xmm2,xmm4
-	movdqu	xmm3,[%1 + 3*16]
-	pxor	xmm3,xmm4
-%endmacro
-
-%macro store4 1
-	movdqu [%1 + 0*16],xmm0
-	movdqu [%1 + 1*16],xmm1
-	movdqu [%1 + 2*16],xmm2
-	movdqu [%1 + 3*16],xmm3
-%endmacro
-
-%macro copy_round_keys 3
-	movdqu xmm4,[%2 + ((%3)*16)]
-	movdqa [%1 + ((%3)*16)],xmm4
-%endmacro
-
-
-%macro key_expansion_1_192 1
-		;; Assumes the xmm3 includes all zeros at this point. 
-        pshufd xmm2, xmm2, 11111111b        
-        shufps xmm3, xmm1, 00010000b        
-        pxor xmm1, xmm3        
-        shufps xmm3, xmm1, 10001100b
-        pxor xmm1, xmm3        
-		pxor xmm1, xmm2		
-		movdqu [rdx+%1], xmm1			
-%endmacro
-
-; Calculate w10 and w11 using calculated w9 and known w4-w5
-%macro key_expansion_2_192 1				
-		movdqa xmm5, xmm4
-		pslldq xmm5, 4
-		shufps xmm6, xmm1, 11110000b
-		pxor xmm6, xmm5
-		pxor xmm4, xmm6
-		pshufd xmm7, xmm4, 00001110b 
-		movdqu [rdx+%1], xmm7
-%endmacro
-
-
-section .data
-align 16
-shuffle_mask:
-DD 0FFFFFFFFh
-DD 03020100h
-DD 07060504h
-DD 0B0A0908h
-
-
-align 16
-counter_add_one:
-DD 1
-DD 0
-DD 0
-DD 0
-
-counter_add_two:
-DD 2
-DD 0
-DD 0
-DD 0
-
-counter_add_three:
-DD 3
-DD 0
-DD 0
-DD 0
-
-counter_add_four:
-DD 4
-DD 0
-DD 0
-DD 0
-
-
-
-section .text
-
-align 16
-key_expansion256:
-
-    pshufd xmm2, xmm2, 011111111b
-
-    movdqa xmm4, xmm1
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pxor xmm1, xmm2
-
-    movdqu [rdx], xmm1
-    add rdx, 0x10
-    
-    aeskeygenassist xmm4, xmm1, 0
-    pshufd xmm2, xmm4, 010101010b
-
-    movdqa xmm4, xmm3
-    pshufb xmm4, xmm5
-    pxor xmm3, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm3, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm3, xmm4
-    pxor xmm3, xmm2
-
-    movdqu [rdx], xmm3
-    add rdx, 0x10
-
-    ret
-
-
-
-align 16
-key_expansion128: 
-    pshufd xmm2, xmm2, 0xFF;
-    movdqa xmm3, xmm1
-    pshufb xmm3, xmm5
-    pxor xmm1, xmm3
-    pshufb xmm3, xmm5
-    pxor xmm1, xmm3
-    pshufb xmm3, xmm5
-    pxor xmm1, xmm3
-    pxor xmm1, xmm2
-
-    ; storing the result in the key schedule array
-    movdqu [rdx], xmm1
-    add rdx, 0x10                    
-    ret
-
-
-
-
-
-
-align 16
-global iEncExpandKey128
-iEncExpandKey128:
-
-		linux_setup
-
-        movdqu xmm1, [rcx]    ; loading the key
-
-        movdqu [rdx], xmm1
-
-        movdqa xmm5, [shuffle_mask wrt rip]
-
-        add rdx,16
-
-        aeskeygenassist xmm2, xmm1, 0x1     ; Generating round key 1
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x2     ; Generating round key 2
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x4     ; Generating round key 3
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x8     ; Generating round key 4
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x10    ; Generating round key 5
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x20    ; Generating round key 6
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x40    ; Generating round key 7
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x80    ; Generating round key 8
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x1b    ; Generating round key 9
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x36    ; Generating round key 10
-        call key_expansion128
-
-		ret 
-
-
-
-align 16
-global iEncExpandKey192
-iEncExpandKey192:
-
-		linux_setup
-		sub rsp,64+8
-		movdqa	[rsp],xmm6
-		movdqa	[rsp+16],xmm7
-
-
-        movq xmm7, [rcx+16]	; loading the AES key
-        movq [rdx+16], xmm7  ; Storing key in memory where all key expansion 
-        pshufd xmm4, xmm7, 01001111b
-        movdqu xmm1, [rcx]	; loading the AES key
-        movdqu [rdx], xmm1  ; Storing key in memory where all key expansion 
-			
-        pxor xmm3, xmm3		; Set xmm3 to be all zeros. Required for the key_expansion. 
-        pxor xmm6, xmm6		; Set xmm3 to be all zeros. Required for the key_expansion. 
-
-        aeskeygenassist xmm2, xmm4, 0x1     ; Complete round key 1 and generate round key 2 
-        key_expansion_1_192 24
-		key_expansion_2_192 40				
-
-        aeskeygenassist xmm2, xmm4, 0x2     ; Generate round key 3 and part of round key 4
-        key_expansion_1_192 48
-		key_expansion_2_192 64				
-
-        aeskeygenassist xmm2, xmm4, 0x4     ; Complete round key 4 and generate round key 5
-        key_expansion_1_192 72
-		key_expansion_2_192 88
-		
-        aeskeygenassist xmm2, xmm4, 0x8     ; Generate round key 6 and part of round key 7
-        key_expansion_1_192 96
-		key_expansion_2_192 112
-		
-        aeskeygenassist xmm2, xmm4, 0x10     ; Complete round key 7 and generate round key 8 
-        key_expansion_1_192 120
-		key_expansion_2_192 136				
-
-        aeskeygenassist xmm2, xmm4, 0x20     ; Generate round key 9 and part of round key 10
-        key_expansion_1_192 144
-		key_expansion_2_192 160				
-
-        aeskeygenassist xmm2, xmm4, 0x40     ; Complete round key 10 and generate round key 11
-        key_expansion_1_192 168
-		key_expansion_2_192 184				
-
-        aeskeygenassist xmm2, xmm4, 0x80     ; Generate round key 12
-        key_expansion_1_192 192
-
-
-		movdqa	xmm6,[rsp]
-		movdqa	xmm7,[rsp+16]
-		add rsp,64+8
-
-		ret 
-
-
-
-
-align 16
-global iDecExpandKey128
-iDecExpandKey128:
-
-	linux_setup
-	push rcx
-	push rdx
-	sub rsp,16+8
-
-	call iEncExpandKey128
-
-	add rsp,16+8
-	pop rdx
-	pop rcx
-
-	inversekey [rdx + 1*16]
-	inversekey [rdx + 2*16]
-	inversekey [rdx + 3*16]
-	inversekey [rdx + 4*16]
-	inversekey [rdx + 5*16]
-	inversekey [rdx + 6*16]
-	inversekey [rdx + 7*16]
-	inversekey [rdx + 8*16]
-	inversekey [rdx + 9*16]
-
-	ret
-
-
-align 16
-global iDecExpandKey192
-iDecExpandKey192:
-
-	linux_setup
-	push rcx
-	push rdx
-	sub rsp,16+8
-
-	call iEncExpandKey192
-
-	add rsp,16+8
-	pop rdx
-	pop rcx
-
-	
-	inversekey [rdx + 1*16]
-	inversekey [rdx + 2*16]
-	inversekey [rdx + 3*16]
-	inversekey [rdx + 4*16]
-	inversekey [rdx + 5*16]
-	inversekey [rdx + 6*16]
-	inversekey [rdx + 7*16]
-	inversekey [rdx + 8*16]
-	inversekey [rdx + 9*16]
-	inversekey [rdx + 10*16]
-	inversekey [rdx + 11*16]
-
-	ret
-
-
-
-align 16
-global iDecExpandKey256
-iDecExpandKey256:
-
-	linux_setup
-	push rcx
-	push rdx
-	sub rsp,16+8
-
-	call iEncExpandKey256
-
-	add rsp,16+8
-	pop rdx
-	pop rcx
-
-	inversekey [rdx + 1*16]
-	inversekey [rdx + 2*16]
-	inversekey [rdx + 3*16]
-	inversekey [rdx + 4*16]
-	inversekey [rdx + 5*16]
-	inversekey [rdx + 6*16]
-	inversekey [rdx + 7*16]
-	inversekey [rdx + 8*16]
-	inversekey [rdx + 9*16]
-	inversekey [rdx + 10*16]
-	inversekey [rdx + 11*16]
-	inversekey [rdx + 12*16]
-	inversekey [rdx + 13*16]
-
-	ret
-	
-
-	
-	
-align 16
-global iEncExpandKey256
-iEncExpandKey256:
-
-	linux_setup
-
-    movdqu xmm1, [rcx]    ; loading the key
-    movdqu xmm3, [rcx+16]
-    movdqu [rdx], xmm1  ; Storing key in memory where all key schedule will be stored
-    movdqu [rdx+16], xmm3 
-    
-    add rdx,32
-
-    movdqa xmm5, [shuffle_mask wrt rip]  ; this mask is used by key_expansion
-
-    aeskeygenassist xmm2, xmm3, 0x1     ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x2     ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x4     ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x8     ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x10    ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x20    ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x40    ; 
-;    call key_expansion256 
-
-    pshufd xmm2, xmm2, 011111111b
-
-    movdqa xmm4, xmm1
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pxor xmm1, xmm2
-
-    movdqu [rdx], xmm1
-
-
-	ret 
-	
-	
-	
-
-
-
-align 16
-global iDec128
-iDec128:
-
-	linux_setup
-	sub rsp,16*16+8
-
-
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	sub r8,rdx
-
-	test eax,eax
-	jz end_dec128
-
-	cmp eax,4
-	jl	lp128decsingle
-
-	test	rcx,0xf
-	jz		lp128decfour
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	mov rcx,rsp	
-	
-	
-
-align 16
-lp128decfour:
-	
-	test eax,eax
-	jz end_dec128
-
-	cmp eax,4
-	jl	lp128decsingle
-
-	load_and_xor4 rdx, [rcx+10*16]
-	add rdx,16*4
-	aesdec4 [rcx+9*16]
-	aesdec4 [rcx+8*16]
-	aesdec4 [rcx+7*16]
-	aesdec4 [rcx+6*16]
-	aesdec4 [rcx+5*16]
-	aesdec4 [rcx+4*16]
-	aesdec4 [rcx+3*16]
-	aesdec4 [rcx+2*16]
-	aesdec4 [rcx+1*16]
-	aesdeclast4 [rcx+0*16]
-	
-	sub eax,4
-	store4 r8+rdx-(16*4)
-	jmp lp128decfour
-
-
-	align 16
-lp128decsingle:
-
-	movdqu xmm0, [rdx]
-	movdqu xmm4,[rcx+10*16]
-	pxor xmm0, xmm4
-	aesdec1_u [rcx+9*16]
-	aesdec1_u [rcx+8*16]
-	aesdec1_u [rcx+7*16]
-	aesdec1_u [rcx+6*16]
-	aesdec1_u [rcx+5*16]
-	aesdec1_u [rcx+4*16]
-	aesdec1_u [rcx+3*16]
-	aesdec1_u [rcx+2*16]
-	aesdec1_u [rcx+1*16]
-	aesdeclast1_u [rcx+0*16]
-
-	add rdx, 16
-	movdqu  [r8 + rdx - 16], xmm0
-	dec eax
-	jnz lp128decsingle
-
-end_dec128:
-
-	add rsp,16*16+8
-	ret
-
-
-align 16
-global iDec128_CBC
-iDec128_CBC:
-	
-	linux_setup
-	sub rsp,16*16+8
-
-	mov r9,rcx
-	mov rax,[rcx+24]
-	movdqu	xmm5,[rax]
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	
-	sub r8,rdx
-
-
-	test eax,eax
-	jz end_dec128_CBC
-
-	cmp eax,4
-	jl	lp128decsingle_CBC
-
-	test	rcx,0xf
-	jz		lp128decfour_CBC
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	mov rcx,rsp	
-
-
-align 16
-lp128decfour_CBC:
-	
-	test eax,eax
-	jz end_dec128_CBC
-
-	cmp eax,4
-	jl	lp128decsingle_CBC
-
-	load_and_xor4 rdx, [rcx+10*16]
-	add rdx,16*4
-	aesdec4 [rcx+9*16]
-	aesdec4 [rcx+8*16]
-	aesdec4 [rcx+7*16]
-	aesdec4 [rcx+6*16]
-	aesdec4 [rcx+5*16]
-	aesdec4 [rcx+4*16]
-	aesdec4 [rcx+3*16]
-	aesdec4 [rcx+2*16]
-	aesdec4 [rcx+1*16]
-	aesdeclast4 [rcx+0*16]
-
-	pxor	xmm0,xmm5
-	movdqu	xmm4,[rdx - 16*4 + 0*16]
-	pxor	xmm1,xmm4
-	movdqu	xmm4,[rdx - 16*4 + 1*16]
-	pxor	xmm2,xmm4
-	movdqu	xmm4,[rdx - 16*4 + 2*16]
-	pxor	xmm3,xmm4
-	movdqu	xmm5,[rdx - 16*4 + 3*16]
-	
-	sub eax,4
-	store4 r8+rdx-(16*4)
-	jmp lp128decfour_CBC
-
-
-	align 16
-lp128decsingle_CBC:
-
-	movdqu xmm0, [rdx]
-	movdqa	xmm1,xmm0
-	movdqu xmm4,[rcx+10*16]
-	pxor xmm0, xmm4
-	aesdec1_u [rcx+9*16]
-	aesdec1_u [rcx+8*16]
-	aesdec1_u [rcx+7*16]
-	aesdec1_u [rcx+6*16]
-	aesdec1_u [rcx+5*16]
-	aesdec1_u [rcx+4*16]
-	aesdec1_u [rcx+3*16]
-	aesdec1_u [rcx+2*16]
-	aesdec1_u [rcx+1*16]
-	aesdeclast1_u [rcx+0*16]
-
-	pxor	xmm0,xmm5
-	movdqa	xmm5,xmm1
-	add rdx, 16
-	movdqu  [r8 + rdx - 16], xmm0
-	dec eax
-	jnz lp128decsingle_CBC
-
-end_dec128_CBC:
-
-	mov	   r9,[r9+24]
-	movdqu [r9],xmm5
-	add rsp,16*16+8
-	ret
-
-
-align 16
-global iDec192_CBC
-iDec192_CBC:
-
-	linux_setup
-	sub rsp,16*16+8
-
-	mov r9,rcx
-	mov rax,[rcx+24]
-	movdqu	xmm5,[rax]
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	
-	sub r8,rdx
-
-	test eax,eax
-	jz end_dec192_CBC
-
-	cmp eax,4
-	jl	lp192decsingle_CBC
-
-	test	rcx,0xf
-	jz		lp192decfour_CBC
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	copy_round_keys rsp,rcx,11
-	copy_round_keys rsp,rcx,12
-	mov rcx,rsp	
-
-
-align 16
-lp192decfour_CBC:
-	
-	test eax,eax
-	jz end_dec192_CBC
-
-	cmp eax,4
-	jl	lp192decsingle_CBC
-
-	load_and_xor4 rdx, [rcx+12*16]
-	add rdx,16*4
-	aesdec4 [rcx+11*16]
-	aesdec4 [rcx+10*16]
-	aesdec4 [rcx+9*16]
-	aesdec4 [rcx+8*16]
-	aesdec4 [rcx+7*16]
-	aesdec4 [rcx+6*16]
-	aesdec4 [rcx+5*16]
-	aesdec4 [rcx+4*16]
-	aesdec4 [rcx+3*16]
-	aesdec4 [rcx+2*16]
-	aesdec4 [rcx+1*16]
-	aesdeclast4 [rcx+0*16]
-
-	pxor	xmm0,xmm5
-	movdqu	xmm4,[rdx - 16*4 + 0*16]
-	pxor	xmm1,xmm4
-	movdqu	xmm4,[rdx - 16*4 + 1*16]
-	pxor	xmm2,xmm4
-	movdqu	xmm4,[rdx - 16*4 + 2*16]
-	pxor	xmm3,xmm4
-	movdqu	xmm5,[rdx - 16*4 + 3*16]
-	
-	sub eax,4
-	store4 r8+rdx-(16*4)
-	jmp lp192decfour_CBC
-
-
-	align 16
-lp192decsingle_CBC:
-
-	movdqu xmm0, [rdx]
-	movdqu xmm4,[rcx+12*16]
-	movdqa	xmm1,xmm0
-	pxor xmm0, xmm4
-	aesdec1_u [rcx+11*16]
-	aesdec1_u [rcx+10*16]
-	aesdec1_u [rcx+9*16]
-	aesdec1_u [rcx+8*16]
-	aesdec1_u [rcx+7*16]
-	aesdec1_u [rcx+6*16]
-	aesdec1_u [rcx+5*16]
-	aesdec1_u [rcx+4*16]
-	aesdec1_u [rcx+3*16]
-	aesdec1_u [rcx+2*16]
-	aesdec1_u [rcx+1*16]
-	aesdeclast1_u [rcx+0*16]
-
-	pxor	xmm0,xmm5
-	movdqa	xmm5,xmm1
-	add rdx, 16
-	movdqu  [r8 + rdx - 16], xmm0
-	dec eax
-	jnz lp192decsingle_CBC
-
-end_dec192_CBC:
-
-	mov	   r9,[r9+24]
-	movdqu [r9],xmm5
-	add rsp,16*16+8
-	ret
-
-
-
-
-align 16
-global iDec256_CBC
-iDec256_CBC:
-
-	linux_setup
-	sub rsp,16*16+8
-
-	mov r9,rcx
-	mov rax,[rcx+24]
-	movdqu	xmm5,[rax]
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	
-	sub r8,rdx
-
-	test eax,eax
-	jz end_dec256_CBC
-
-	cmp eax,4
-	jl	lp256decsingle_CBC
-
-	test	rcx,0xf
-	jz		lp256decfour_CBC
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	copy_round_keys rsp,rcx,11
-	copy_round_keys rsp,rcx,12
-	copy_round_keys rsp,rcx,13
-	copy_round_keys rsp,rcx,14
-	mov rcx,rsp	
-
-align 16
-lp256decfour_CBC:
-	
-	test eax,eax
-	jz end_dec256_CBC
-
-	cmp eax,4
-	jl	lp256decsingle_CBC
-
-	load_and_xor4 rdx, [rcx+14*16]
-	add rdx,16*4
-	aesdec4 [rcx+13*16]
-	aesdec4 [rcx+12*16]
-	aesdec4 [rcx+11*16]
-	aesdec4 [rcx+10*16]
-	aesdec4 [rcx+9*16]
-	aesdec4 [rcx+8*16]
-	aesdec4 [rcx+7*16]
-	aesdec4 [rcx+6*16]
-	aesdec4 [rcx+5*16]
-	aesdec4 [rcx+4*16]
-	aesdec4 [rcx+3*16]
-	aesdec4 [rcx+2*16]
-	aesdec4 [rcx+1*16]
-	aesdeclast4 [rcx+0*16]
-
-	pxor	xmm0,xmm5
-	movdqu	xmm4,[rdx - 16*4 + 0*16]
-	pxor	xmm1,xmm4
-	movdqu	xmm4,[rdx - 16*4 + 1*16]
-	pxor	xmm2,xmm4
-	movdqu	xmm4,[rdx - 16*4 + 2*16]
-	pxor	xmm3,xmm4
-	movdqu	xmm5,[rdx - 16*4 + 3*16]
-	
-	sub eax,4
-	store4 r8+rdx-(16*4)
-	jmp lp256decfour_CBC
-
-
-	align 16
-lp256decsingle_CBC:
-
-	movdqu xmm0, [rdx]
-	movdqu xmm4,[rcx+14*16]
-	movdqa	xmm1,xmm0
-	pxor xmm0, xmm4
-	aesdec1_u [rcx+13*16]
-	aesdec1_u [rcx+12*16]
-	aesdec1_u [rcx+11*16]
-	aesdec1_u [rcx+10*16]
-	aesdec1_u [rcx+9*16]
-	aesdec1_u [rcx+8*16]
-	aesdec1_u [rcx+7*16]
-	aesdec1_u [rcx+6*16]
-	aesdec1_u [rcx+5*16]
-	aesdec1_u [rcx+4*16]
-	aesdec1_u [rcx+3*16]
-	aesdec1_u [rcx+2*16]
-	aesdec1_u [rcx+1*16]
-	aesdeclast1_u [rcx+0*16]
-
-	pxor	xmm0,xmm5
-	movdqa	xmm5,xmm1
-	add rdx, 16
-	movdqu  [r8 + rdx - 16], xmm0
-	dec eax
-	jnz lp256decsingle_CBC
-
-end_dec256_CBC:
-
-	mov	   r9,[r9+24]
-	movdqu [r9],xmm5
-	add rsp,16*16+8
-	ret
-
-
-
-
-
-align 16
-global iDec192
-iDec192:
-
-	linux_setup
-	sub rsp,16*16+8
-
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	sub r8,rdx
-	
-	test eax,eax
-	jz end_dec192
-
-	cmp eax,4
-	jl	lp192decsingle
-	
-	test	rcx,0xf
-	jz		lp192decfour
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	copy_round_keys rsp,rcx,11
-	copy_round_keys rsp,rcx,12
-	mov rcx,rsp	
-
-align 16
-lp192decfour:
-	
-	test eax,eax
-	jz end_dec192
-
-	cmp eax,4
-	jl	lp192decsingle
-
-	load_and_xor4 rdx, [rcx+12*16]
-	add rdx,16*4
-	aesdec4 [rcx+11*16]
-	aesdec4 [rcx+10*16]
-	aesdec4 [rcx+9*16]
-	aesdec4 [rcx+8*16]
-	aesdec4 [rcx+7*16]
-	aesdec4 [rcx+6*16]
-	aesdec4 [rcx+5*16]
-	aesdec4 [rcx+4*16]
-	aesdec4 [rcx+3*16]
-	aesdec4 [rcx+2*16]
-	aesdec4 [rcx+1*16]
-	aesdeclast4 [rcx+0*16]
-	
-	sub eax,4
-	store4 r8+rdx-(16*4)
-	jmp lp192decfour
-
-
-	align 16
-lp192decsingle:
-
-	movdqu xmm0, [rdx]
-	movdqu xmm4,[rcx+12*16]
-	pxor xmm0, xmm4
-	aesdec1_u [rcx+11*16]
-	aesdec1_u [rcx+10*16]
-	aesdec1_u [rcx+9*16]
-	aesdec1_u [rcx+8*16]
-	aesdec1_u [rcx+7*16]
-	aesdec1_u [rcx+6*16]
-	aesdec1_u [rcx+5*16]
-	aesdec1_u [rcx+4*16]
-	aesdec1_u [rcx+3*16]
-	aesdec1_u [rcx+2*16]
-	aesdec1_u [rcx+1*16]
-	aesdeclast1_u [rcx+0*16]
-
-	add rdx, 16
-	movdqu  [r8 + rdx - 16], xmm0
-	dec eax
-	jnz lp192decsingle
-
-end_dec192:
-
-	add rsp,16*16+8
-	ret
-
-
-
-
-align 16
-global iDec256
-iDec256:
-
-	linux_setup
-	sub rsp,16*16+8
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	sub r8,rdx
-
-
-	test eax,eax
-	jz end_dec256
-	
-	cmp eax,4
-	jl lp256dec
-
-	test	rcx,0xf
-	jz		lp256dec4
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	copy_round_keys rsp,rcx,11
-	copy_round_keys rsp,rcx,12
-	copy_round_keys rsp,rcx,13
-	copy_round_keys rsp,rcx,14
-	mov rcx,rsp	
-
-	
-	align 16
-lp256dec4:	
-	test eax,eax
-	jz end_dec256
-	
-	cmp eax,4
-	jl lp256dec
-	
-	load_and_xor4 rdx,[rcx+14*16]
-	add rdx, 4*16
-	aesdec4 [rcx+13*16]
-	aesdec4 [rcx+12*16]
-	aesdec4 [rcx+11*16]
-	aesdec4 [rcx+10*16]
-	aesdec4 [rcx+9*16]
-	aesdec4 [rcx+8*16]
-	aesdec4 [rcx+7*16]
-	aesdec4 [rcx+6*16]
-	aesdec4 [rcx+5*16]
-	aesdec4 [rcx+4*16]
-	aesdec4 [rcx+3*16]
-	aesdec4 [rcx+2*16]
-	aesdec4 [rcx+1*16]
-	aesdeclast4 [rcx+0*16]
-
-	store4 r8+rdx-16*4
-	sub eax,4
-	jmp lp256dec4	
-	
-	align 16
-lp256dec:
-
-	movdqu xmm0, [rdx]
-	movdqu xmm4,[rcx+14*16]
-	add rdx, 16
-	pxor xmm0, xmm4                    ; Round 0 (only xor)
-	aesdec1_u [rcx+13*16]
-	aesdec1_u [rcx+12*16]
-	aesdec1_u [rcx+11*16]
-	aesdec1_u [rcx+10*16]
-	aesdec1_u [rcx+9*16]
-	aesdec1_u [rcx+8*16]
-	aesdec1_u [rcx+7*16]
-	aesdec1_u [rcx+6*16]
-	aesdec1_u [rcx+5*16]
-	aesdec1_u [rcx+4*16]
-	aesdec1_u [rcx+3*16]
-	aesdec1_u [rcx+2*16]
-	aesdec1_u [rcx+1*16]
-	aesdeclast1_u [rcx+0*16]
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [r8+rdx-16], xmm0
-	dec eax
-	jnz lp256dec
-
-end_dec256:
-	
-	add rsp,16*16+8
-	ret
-
-
-
-
-
-
-align 16
-global iEnc128
-iEnc128:
-
-	linux_setup
-	sub rsp,16*16+8
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	sub r8,rdx
-
-
-	test eax,eax
-	jz end_enc128
-	
-	cmp eax,4
-	jl lp128encsingle
-
-	test	rcx,0xf
-	jz		lpenc128four
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	mov rcx,rsp	
-
-
-	align 16	
-	
-lpenc128four:
-	
-	test eax,eax
-	jz end_enc128
-	
-	cmp eax,4
-	jl lp128encsingle
-
-	load_and_xor4 rdx,[rcx+0*16]
-	add rdx,4*16
-	aesenc4	[rcx+1*16]
-	aesenc4	[rcx+2*16]
-	aesenc4	[rcx+3*16]
-	aesenc4	[rcx+4*16]
-	aesenc4	[rcx+5*16]
-	aesenc4	[rcx+6*16]
-	aesenc4	[rcx+7*16]
-	aesenc4	[rcx+8*16]
-	aesenc4	[rcx+9*16]
-	aesenclast4	[rcx+10*16]
-	
-	store4 r8+rdx-16*4
-	sub eax,4
-	jmp lpenc128four
-	
-	align 16
-lp128encsingle:
-
-	movdqu xmm0, [rdx]
-	movdqu xmm4,[rcx+0*16]
-	add rdx, 16
-	pxor xmm0, xmm4
-	aesenc1_u [rcx+1*16]
-	aesenc1_u [rcx+2*16]
-	aesenc1_u [rcx+3*16]
-	aesenc1_u [rcx+4*16]     
-	aesenc1_u [rcx+5*16]
-	aesenc1_u [rcx+6*16]
-	aesenc1_u [rcx+7*16]
-	aesenc1_u [rcx+8*16]
-	aesenc1_u [rcx+9*16]
-	aesenclast1_u [rcx+10*16]
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [r8+rdx-16], xmm0
-	dec eax
-	jnz lp128encsingle
-
-end_enc128:
-
-	add rsp,16*16+8
-	ret
-
-
-align 16
-global iEnc128_CTR
-iEnc128_CTR:
-
-	linux_setup
-
-	mov r9,rcx
-	mov rax,[rcx+24]
-	movdqu xmm5,[rax]
-
-
-	sub rsp,16*16+8
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	sub r8,rdx
-
-
-	test eax,eax
-	jz end_encctr128
-	
-	cmp eax,4
-	jl lp128encctrsingle
-
-	test	rcx,0xf
-	jz		lpencctr128four
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	mov rcx,rsp	
-
-
-	align 16	
-	
-lpencctr128four:
-	
-	test eax,eax
-	jz end_encctr128
-	
-	cmp eax,4
-	jl lp128encctrsingle
-
-	load_and_inc4 [rcx+0*16]
-	add rdx,4*16
-	aesenc4	[rcx+1*16]
-	aesenc4	[rcx+2*16]
-	aesenc4	[rcx+3*16]
-	aesenc4	[rcx+4*16]
-	aesenc4	[rcx+5*16]
-	aesenc4	[rcx+6*16]
-	aesenc4	[rcx+7*16]
-	aesenc4	[rcx+8*16]
-	aesenc4	[rcx+9*16]
-	aesenclast4	[rcx+10*16]
-	xor_with_input4 rdx-(4*16)
-	
-	store4 r8+rdx-16*4
-	sub eax,4
-	jmp lpencctr128four
-	
-	align 16
-lp128encctrsingle:
-
-	movdqa xmm0,xmm5
-	paddq	xmm5,[counter_add_one wrt rip]
-	add rdx, 16
-	movdqu xmm4,[rcx+0*16]
-	pxor xmm0, xmm4
-	aesenc1_u [rcx+1*16]
-	aesenc1_u [rcx+2*16]
-	aesenc1_u [rcx+3*16]
-	aesenc1_u [rcx+4*16]     
-	aesenc1_u [rcx+5*16]
-	aesenc1_u [rcx+6*16]
-	aesenc1_u [rcx+7*16]
-	aesenc1_u [rcx+8*16]
-	aesenc1_u [rcx+9*16]
-	aesenclast1_u [rcx+10*16]
-	movdqu xmm4, [rdx-16]
-	pxor  xmm0,xmm4
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [r8+rdx-16], xmm0
-	dec eax
-	jnz lp128encctrsingle
-
-end_encctr128:
-
-	mov	   r9,[r9+24]
-	movdqu [r9],xmm5
-	add rsp,16*16+8
-	ret
-
-
-
-align 16
-global iEnc192_CTR
-iEnc192_CTR:
-
-	linux_setup
-
-	mov r9,rcx
-	mov rax,[rcx+24]
-	movdqu xmm5,[rax]
-
-
-	sub rsp,16*16+8
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	sub r8,rdx
-
-
-	test eax,eax
-	jz end_encctr192
-	
-	cmp eax,4
-	jl lp192encctrsingle
-
-	test	rcx,0xf
-	jz		lpencctr192four
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	copy_round_keys rsp,rcx,11
-	copy_round_keys rsp,rcx,12
-	mov rcx,rsp	
-
-
-	align 16	
-	
-lpencctr192four:
-	
-	test eax,eax
-	jz end_encctr192
-	
-	cmp eax,4
-	jl lp192encctrsingle
-
-	load_and_inc4 [rcx+0*16]
-	add rdx,4*16
-	aesenc4	[rcx+1*16]
-	aesenc4	[rcx+2*16]
-	aesenc4	[rcx+3*16]
-	aesenc4	[rcx+4*16]
-	aesenc4	[rcx+5*16]
-	aesenc4	[rcx+6*16]
-	aesenc4	[rcx+7*16]
-	aesenc4	[rcx+8*16]
-	aesenc4	[rcx+9*16]
-	aesenc4	[rcx+10*16]
-	aesenc4	[rcx+11*16]
-	aesenclast4	[rcx+12*16]
-	xor_with_input4 rdx-(4*16)
-	
-	store4 r8+rdx-16*4
-	sub eax,4
-	jmp lpencctr192four
-	
-	align 16
-lp192encctrsingle:
-
-	movdqa xmm5,xmm0
-	movdqu xmm4,[rcx+0*16]
-	paddq	xmm5,[counter_add_one wrt rip]
-	add rdx, 16
-	pxor xmm0, xmm4
-	aesenc1_u [rcx+1*16]
-	aesenc1_u [rcx+2*16]
-	aesenc1_u [rcx+3*16]
-	aesenc1_u [rcx+4*16]     
-	aesenc1_u [rcx+5*16]
-	aesenc1_u [rcx+6*16]
-	aesenc1_u [rcx+7*16]
-	aesenc1_u [rcx+8*16]
-	aesenc1_u [rcx+9*16]
-	aesenc1_u [rcx+10*16]
-	aesenc1_u [rcx+11*16]
-	aesenclast1_u [rcx+12*16]
-	movdqu xmm4, [rdx]
-	pxor  xmm0,xmm4
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [r8+rdx-16], xmm0
-	dec eax
-	jnz lp192encctrsingle
-
-end_encctr192:
-
-	mov	   r9,[r9+24]
-	movdqu [r9],xmm5
-	add rsp,16*16+8
-	ret
-
-
-align 16
-global iEnc256_CTR
-iEnc256_CTR:
-
-	linux_setup
-
-	mov r9,rcx
-	mov rax,[rcx+24]
-	movdqu xmm5,[rax]
-
-
-	sub rsp,16*16+8
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	sub r8,rdx
-
-
-	test eax,eax
-	jz end_encctr256
-	
-	cmp eax,4
-	jl lp256encctrsingle
-
-	test	rcx,0xf
-	jz		lpencctr256four
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	copy_round_keys rsp,rcx,11
-	copy_round_keys rsp,rcx,12
-	copy_round_keys rsp,rcx,13
-	copy_round_keys rsp,rcx,14
-	mov rcx,rsp	
-
-
-	align 16	
-	
-lpencctr256four:
-	
-	test eax,eax
-	jz end_encctr256
-	
-	cmp eax,4
-	jl lp256encctrsingle
-
-	load_and_inc4 [rcx+0*16]
-	add rdx,4*16
-	aesenc4	[rcx+1*16]
-	aesenc4	[rcx+2*16]
-	aesenc4	[rcx+3*16]
-	aesenc4	[rcx+4*16]
-	aesenc4	[rcx+5*16]
-	aesenc4	[rcx+6*16]
-	aesenc4	[rcx+7*16]
-	aesenc4	[rcx+8*16]
-	aesenc4	[rcx+9*16]
-	aesenc4	[rcx+10*16]
-	aesenc4	[rcx+11*16]
-	aesenc4	[rcx+12*16]
-	aesenc4	[rcx+13*16]
-	aesenclast4	[rcx+14*16]
-	xor_with_input4 rdx-(4*16)
-	
-	store4 r8+rdx-16*4
-	sub eax,4
-	jmp lpencctr256four
-	
-	align 16
-lp256encctrsingle:
-
-	movdqa xmm5,xmm0
-	movdqu xmm4,[rcx+0*16]
-	paddq	xmm5,[counter_add_one wrt rip]
-	add rdx, 16
-	pxor xmm0, xmm4
-	aesenc1_u [rcx+1*16]
-	aesenc1_u [rcx+2*16]
-	aesenc1_u [rcx+3*16]
-	aesenc1_u [rcx+4*16]     
-	aesenc1_u [rcx+5*16]
-	aesenc1_u [rcx+6*16]
-	aesenc1_u [rcx+7*16]
-	aesenc1_u [rcx+8*16]
-	aesenc1_u [rcx+9*16]
-	aesenc1_u [rcx+10*16]
-	aesenc1_u [rcx+11*16]
-	aesenc1_u [rcx+12*16]
-	aesenc1_u [rcx+13*16]
-	aesenclast1_u [rcx+14*16]
-	movdqu xmm4, [rdx]
-	pxor  xmm0,xmm4
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [r8+rdx-16], xmm0
-	dec eax
-	jnz lp256encctrsingle
-
-end_encctr256:
-
-	mov	   r9,[r9+24]
-	movdqu [r9],xmm5
-	add rsp,16*16+8
-	ret
-
-
-
-
-
-
-
-align 16
-global iEnc128_CBC
-iEnc128_CBC:
-
-	linux_setup
-	sub rsp,16*16+8
-	
-	mov r9,rcx
-	mov rax,[rcx+24]
-	movdqu xmm1,[rax]
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	sub r8,rdx
-
-
-	test	rcx,0xf
-	jz		lp128encsingle_CBC
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	mov rcx,rsp	
-
-
-	align 16	
-	
-lp128encsingle_CBC:
-
-	movdqu xmm0, [rdx]
-	movdqu xmm4,[rcx+0*16]
-	add rdx, 16
-	pxor xmm0, xmm1
-	pxor xmm0, xmm4
-	aesenc1 [rcx+1*16]
-	aesenc1 [rcx+2*16]
-	aesenc1 [rcx+3*16]
-	aesenc1 [rcx+4*16]     
-	aesenc1 [rcx+5*16]
-	aesenc1 [rcx+6*16]
-	aesenc1 [rcx+7*16]
-	aesenc1 [rcx+8*16]
-	aesenc1 [rcx+9*16]
-	aesenclast1 [rcx+10*16]
-	movdqa xmm1,xmm0
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [r8+rdx-16], xmm0
-	dec eax
-	jnz lp128encsingle_CBC
-
-	mov	   r9,[r9+24]
-	movdqu [r9],xmm1
-	add rsp,16*16+8
-	ret
-
-
-align 16
-global iEnc192_CBC
-iEnc192_CBC:
-
-	linux_setup
-	sub rsp,16*16+8
-	mov r9,rcx
-	mov rax,[rcx+24]
-	movdqu xmm1,[rax]
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	sub r8,rdx
-
-	test	rcx,0xf
-	jz		lp192encsingle_CBC
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	copy_round_keys rsp,rcx,11
-	copy_round_keys rsp,rcx,12
-	mov rcx,rsp	
-
-
-
-	align 16	
-	
-lp192encsingle_CBC:
-
-	movdqu xmm0, [rdx]
-	movdqu xmm4, [rcx+0*16]
-	add rdx, 16
-	pxor xmm0, xmm1
-	pxor xmm0, xmm4
-	aesenc1 [rcx+1*16]
-	aesenc1 [rcx+2*16]
-	aesenc1 [rcx+3*16]
-	aesenc1 [rcx+4*16]     
-	aesenc1 [rcx+5*16]
-	aesenc1 [rcx+6*16]
-	aesenc1 [rcx+7*16]
-	aesenc1 [rcx+8*16]
-	aesenc1 [rcx+9*16]
-	aesenc1 [rcx+10*16]
-	aesenc1 [rcx+11*16]
-	aesenclast1 [rcx+12*16]
-	movdqa xmm1,xmm0
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [r8+rdx-16], xmm0
-	dec eax
-	jnz lp192encsingle_CBC
-
-	mov	   r9,[r9+24]
-	movdqu [r9],xmm1
-
-	add rsp,16*16+8
-	ret
-
-
-align 16
-global iEnc256_CBC
-iEnc256_CBC:
-
-	linux_setup
-	sub rsp,16*16+8
-	
-	mov r9,rcx
-	mov rax,[rcx+24]
-	movdqu xmm1,[rax]
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	sub r8,rdx
-
-	test	rcx,0xf
-	jz		lp256encsingle_CBC
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	copy_round_keys rsp,rcx,11
-	copy_round_keys rsp,rcx,12
-	copy_round_keys rsp,rcx,13
-	copy_round_keys rsp,rcx,14
-	mov rcx,rsp	
-
-	align 16	
-	
-lp256encsingle_CBC:
-
-	movdqu xmm0, [rdx]
-	movdqu xmm4, [rcx+0*16]
-	add rdx, 16
-	pxor xmm0, xmm1
-	pxor xmm0, xmm4
-	aesenc1 [rcx+1*16]
-	aesenc1 [rcx+2*16]
-	aesenc1 [rcx+3*16]
-	aesenc1 [rcx+4*16]     
-	aesenc1 [rcx+5*16]
-	aesenc1 [rcx+6*16]
-	aesenc1 [rcx+7*16]
-	aesenc1 [rcx+8*16]
-	aesenc1 [rcx+9*16]
-	aesenc1 [rcx+10*16]
-	aesenc1 [rcx+11*16]
-	aesenc1 [rcx+12*16]
-	aesenc1 [rcx+13*16]
-	aesenclast1 [rcx+14*16]
-	movdqa xmm1,xmm0
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [r8+rdx-16], xmm0
-	dec eax
-	jnz lp256encsingle_CBC
-
-	mov	   r9,[r9+24]
-	movdqu [r9],xmm1
-	add rsp,16*16+8
-	ret
-
-
-
-
-align 16
-global iEnc192
-iEnc192:
-
-	linux_setup
-	sub rsp,16*16+8
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-	
-	sub r8,rdx
-
-	test eax,eax
-	jz end_enc192
-	
-	cmp eax,4
-	jl lp192encsingle
-
-	test	rcx,0xf
-	jz		lpenc192four
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	copy_round_keys rsp,rcx,11
-	copy_round_keys rsp,rcx,12
-	mov rcx,rsp	
-
-
-	align 16	
-	
-lpenc192four:
-	
-	test eax,eax
-	jz end_enc192
-	
-	cmp eax,4
-	jl lp192encsingle
-
-	load_and_xor4 rdx,[rcx+0*16]
-	add rdx,4*16
-	aesenc4	[rcx+1*16]
-	aesenc4	[rcx+2*16]
-	aesenc4	[rcx+3*16]
-	aesenc4	[rcx+4*16]
-	aesenc4	[rcx+5*16]
-	aesenc4	[rcx+6*16]
-	aesenc4	[rcx+7*16]
-	aesenc4	[rcx+8*16]
-	aesenc4	[rcx+9*16]
-	aesenc4	[rcx+10*16]
-	aesenc4	[rcx+11*16]
-	aesenclast4	[rcx+12*16]
-	
-	store4 r8+rdx-16*4
-	sub eax,4
-	jmp lpenc192four
-	
-	align 16
-lp192encsingle:
-
-	movdqu xmm0, [rdx]
-	movdqu xmm4, [rcx+0*16]
-	add rdx, 16
-	pxor xmm0, xmm4
-	aesenc1_u [rcx+1*16]
-	aesenc1_u [rcx+2*16]
-	aesenc1_u [rcx+3*16]
-	aesenc1_u [rcx+4*16]     
-	aesenc1_u [rcx+5*16]
-	aesenc1_u [rcx+6*16]
-	aesenc1_u [rcx+7*16]
-	aesenc1_u [rcx+8*16]
-	aesenc1_u [rcx+9*16]
-	aesenc1_u [rcx+10*16]
-	aesenc1_u [rcx+11*16]
-	aesenclast1_u [rcx+12*16]
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [r8+rdx-16], xmm0
-	dec eax
-	jnz lp192encsingle
-
-end_enc192:
-
-	add rsp,16*16+8
-	ret
-
-
-
-
-
-
-align 16
-global iEnc256
-iEnc256:
-
-	linux_setup
-	sub rsp,16*16+8
-	
-	mov eax,[rcx+32] ; numblocks
-	mov rdx,[rcx]
-	mov r8,[rcx+8]
-	mov rcx,[rcx+16]
-
-	sub r8,rdx	
-
-
-	test eax,eax
-	jz end_enc256
-
-	cmp eax,4
-	jl lp256enc
-
-	test	rcx,0xf
-	jz		lp256enc4
-	
-	copy_round_keys rsp,rcx,0
-	copy_round_keys rsp,rcx,1
-	copy_round_keys rsp,rcx,2
-	copy_round_keys rsp,rcx,3
-	copy_round_keys rsp,rcx,4
-	copy_round_keys rsp,rcx,5
-	copy_round_keys rsp,rcx,6
-	copy_round_keys rsp,rcx,7
-	copy_round_keys rsp,rcx,8
-	copy_round_keys rsp,rcx,9
-	copy_round_keys rsp,rcx,10
-	copy_round_keys rsp,rcx,11
-	copy_round_keys rsp,rcx,12
-	copy_round_keys rsp,rcx,13
-	copy_round_keys rsp,rcx,14
-	mov rcx,rsp	
-
-
-	align 16
-	
-lp256enc4:
-	test eax,eax
-	jz end_enc256
-
-	cmp eax,4
-	jl lp256enc
-
-
-	load_and_xor4 rdx,[rcx+0*16]
-	add rdx, 16*4
-	aesenc4 [rcx+1*16]
-	aesenc4 [rcx+2*16]
-	aesenc4 [rcx+3*16]
-	aesenc4 [rcx+4*16]
-	aesenc4 [rcx+5*16]
-	aesenc4 [rcx+6*16]
-	aesenc4 [rcx+7*16]
-	aesenc4 [rcx+8*16]
-	aesenc4 [rcx+9*16]
-	aesenc4 [rcx+10*16]
-	aesenc4 [rcx+11*16]
-	aesenc4 [rcx+12*16]
-	aesenc4 [rcx+13*16]
-	aesenclast4 [rcx+14*16]
-
-	store4  r8+rdx-16*4
-	sub eax,4
-	jmp lp256enc4
-	
-	align 16
-lp256enc:
-
-	movdqu xmm0, [rdx]
-	movdqu xmm4, [rcx+0*16]
-	add rdx, 16
-	pxor xmm0, xmm4
-	aesenc1_u [rcx+1*16]
-	aesenc1_u [rcx+2*16]
-	aesenc1_u [rcx+3*16]
-	aesenc1_u [rcx+4*16]
-	aesenc1_u [rcx+5*16]
-	aesenc1_u [rcx+6*16]
-	aesenc1_u [rcx+7*16]
-	aesenc1_u [rcx+8*16]
-	aesenc1_u [rcx+9*16]
-	aesenc1_u [rcx+10*16]
-	aesenc1_u [rcx+11*16]
-	aesenc1_u [rcx+12*16]
-	aesenc1_u [rcx+13*16]
-	aesenclast1_u [rcx+14*16]
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [r8+rdx-16], xmm0
-	dec eax
-	jnz lp256enc
-
-end_enc256:
-
-	add rsp,16*16+8
-	ret
diff --git a/lib/accelerated/intel/asm/x86_iaesx86.s b/lib/accelerated/intel/asm/x86_iaesx86.s
deleted file mode 100755
index 16fda3b19b..0000000000
--- a/lib/accelerated/intel/asm/x86_iaesx86.s
+++ /dev/null
@@ -1,2183 +0,0 @@
-[bits 32]
-[CPU intelnop]
-
-; Copyright (c) 2010, Intel Corporation
-; All rights reserved.
-; 
-; Redistribution and use in source and binary forms, with or without 
-; modification, are permitted provided that the following conditions are met:
-; 
-;     * Redistributions of source code must retain the above copyright notice, 
-;       this list of conditions and the following disclaimer.
-;     * Redistributions in binary form must reproduce the above copyright notice, 
-;       this list of conditions and the following disclaimer in the documentation 
-;       and/or other materials provided with the distribution.
-;     * Neither the name of Intel Corporation nor the names of its contributors 
-;       may be used to endorse or promote products derived from this software 
-;       without specific prior written permission.
-; 
-; THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND 
-; ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 
-; WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 
-; IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, 
-; INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, 
-; BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 
-; DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 
-; LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE 
-; OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF 
-; ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
-
-%macro inversekey 1
-	movdqu  xmm1,%1
-	aesimc	xmm0,xmm1
-	movdqu	%1,xmm0
-%endmacro
-
-
-%macro aesdec4 1
-	movdqa	xmm4,%1
-
-	aesdec	xmm0,xmm4
-	aesdec	xmm1,xmm4
-	aesdec	xmm2,xmm4
-	aesdec	xmm3,xmm4
-
-%endmacro
-
-
-%macro aesdeclast4 1
-	movdqa	xmm4,%1
-
-	aesdeclast	xmm0,xmm4
-	aesdeclast	xmm1,xmm4
-	aesdeclast	xmm2,xmm4
-	aesdeclast	xmm3,xmm4
-
-%endmacro
-
-
-%macro aesenc4 1
-	movdqa	xmm4,%1
-
-	aesenc	xmm0,xmm4
-	aesenc	xmm1,xmm4
-	aesenc	xmm2,xmm4
-	aesenc	xmm3,xmm4
-
-%endmacro
-
-%macro aesenclast4 1
-	movdqa	xmm4,%1
-
-	aesenclast	xmm0,xmm4
-	aesenclast	xmm1,xmm4
-	aesenclast	xmm2,xmm4
-	aesenclast	xmm3,xmm4
-
-%endmacro
-
-
-%macro aesdeclast1 1
-	aesdeclast	xmm0,%1
-%endmacro
-
-%macro aesenclast1 1
-	aesenclast	xmm0,%1
-%endmacro
-
-%macro aesdec1 1
-	aesdec	xmm0,%1
-%endmacro
-
-;abab
-%macro aesenc1 1
-	aesenc	xmm0,%1
-%endmacro
-
-
-%macro aesdeclast1_u 1
-	movdqu xmm4,%1
-	aesdeclast	xmm0,xmm4
-%endmacro
-
-%macro aesenclast1_u 1
-	movdqu xmm4,%1
-	aesenclast	xmm0,xmm4
-%endmacro
-
-%macro aesdec1_u 1
-	movdqu xmm4,%1
-	aesdec	xmm0,xmm4
-%endmacro
-
-%macro aesenc1_u 1
-	movdqu xmm4,%1
-	aesenc	xmm0,xmm4
-%endmacro
-
-
-%macro load_and_xor4 2
-	movdqa	xmm4,%2
-	movdqu	xmm0,[%1 + 0*16]
-	pxor	xmm0,xmm4
-	movdqu	xmm1,[%1 + 1*16]
-	pxor	xmm1,xmm4
-	movdqu	xmm2,[%1 + 2*16]
-	pxor	xmm2,xmm4
-	movdqu	xmm3,[%1 + 3*16]
-	pxor	xmm3,xmm4
-%endmacro
-
-
-%macro load_and_inc4 1
-	movdqa	xmm4,%1
-	movdqa	xmm0,xmm5
-	movdqa  xmm1,xmm5
-	paddq	xmm1,[counter_add_one]
-	movdqa  xmm2,xmm5
-	paddq	xmm2,[counter_add_two]
-	movdqa  xmm3,xmm5
-	paddq	xmm3,[counter_add_three]
-	pxor	xmm0,xmm4
-	paddq	xmm5,[counter_add_four]
-	pxor	xmm1,xmm4
-	pxor	xmm2,xmm4
-	pxor	xmm3,xmm4
-%endmacro
-
-%macro xor_with_input4 1
-	movdqu xmm4,[%1]
-	pxor xmm0,xmm4
-	movdqu xmm4,[%1+16]
-	pxor xmm1,xmm4
-	movdqu xmm4,[%1+32]
-	pxor xmm2,xmm4
-	movdqu xmm4,[%1+48]
-	pxor xmm3,xmm4
-%endmacro
-
-%macro store4 1
-	movdqu [%1 + 0*16],xmm0
-	movdqu [%1 + 1*16],xmm1
-	movdqu [%1 + 2*16],xmm2
-	movdqu [%1 + 3*16],xmm3
-%endmacro
-
-
-%macro copy_round_keys 3
-	movdqu xmm4,[%2 + ((%3)*16)]
-	movdqa [%1 + ((%3)*16)],xmm4
-%endmacro
-
-;abab
-%macro copy_round_keyx 3
-	movdqu xmm4,[%2 + ((%3)*16)]
-	movdqa %1,xmm4
-%endmacro
-
-
-
-%macro key_expansion_1_192 1
-		;; Assumes the xmm3 includes all zeros at this point. 
-        pshufd xmm2, xmm2, 11111111b        
-        shufps xmm3, xmm1, 00010000b        
-        pxor xmm1, xmm3        
-        shufps xmm3, xmm1, 10001100b
-        pxor xmm1, xmm3        
-		pxor xmm1, xmm2		
-		movdqu [edx+%1], xmm1			
-%endmacro
-
-; Calculate w10 and w11 using calculated w9 and known w4-w5
-%macro key_expansion_2_192 1				
-		movdqa xmm5, xmm4
-		pslldq xmm5, 4
-		shufps xmm6, xmm1, 11110000b
-		pxor xmm6, xmm5
-		pxor xmm4, xmm6
-		pshufd xmm7, xmm4, 00001110b 
-		movdqu [edx+%1], xmm7
-%endmacro
-
-
-
-
-
-section .data
-align 16
-shuffle_mask:
-DD 0FFFFFFFFh
-DD 03020100h
-DD 07060504h
-DD 0B0A0908h
-
-align 16
-counter_add_one:
-DD 1
-DD 0
-DD 0
-DD 0
-
-counter_add_two:
-DD 2
-DD 0
-DD 0
-DD 0
-
-counter_add_three:
-DD 3
-DD 0
-DD 0
-DD 0
-
-counter_add_four:
-DD 4
-DD 0
-DD 0
-DD 0
-
-
-section .text
-
-
-
-align 16
-key_expansion256:
-
-    pshufd xmm2, xmm2, 011111111b
-
-    movdqu xmm4, xmm1
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pxor xmm1, xmm2
-
-    movdqu [edx], xmm1
-    add edx, 0x10
-    
-    aeskeygenassist xmm4, xmm1, 0
-    pshufd xmm2, xmm4, 010101010b
-
-    movdqu xmm4, xmm3
-    pshufb xmm4, xmm5
-    pxor xmm3, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm3, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm3, xmm4
-    pxor xmm3, xmm2
-
-    movdqu [edx], xmm3
-    add edx, 0x10
-
-    ret
-
-
-
-align 16
-key_expansion128: 
-    pshufd xmm2, xmm2, 0xFF;
-    movdqu xmm3, xmm1
-    pshufb xmm3, xmm5
-    pxor xmm1, xmm3
-    pshufb xmm3, xmm5
-    pxor xmm1, xmm3
-    pshufb xmm3, xmm5
-    pxor xmm1, xmm3
-    pxor xmm1, xmm2
-
-    ; storing the result in the key schedule array
-    movdqu [edx], xmm1
-    add edx, 0x10                    
-    ret
-
-
-
-align 16
-global _iEncExpandKey128
-_iEncExpandKey128:
-
-	mov ecx,[esp-4+8]		;input
-	mov edx,[esp-4+12]		;ctx
-
-        movdqu xmm1, [ecx]    ; loading the key
-
-        movdqu [edx], xmm1
-
-        movdqa xmm5, [shuffle_mask]
-
-        add edx,16
-
-        aeskeygenassist xmm2, xmm1, 0x1     ; Generating round key 1
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x2     ; Generating round key 2
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x4     ; Generating round key 3
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x8     ; Generating round key 4
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x10    ; Generating round key 5
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x20    ; Generating round key 6
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x40    ; Generating round key 7
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x80    ; Generating round key 8
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x1b    ; Generating round key 9
-        call key_expansion128
-        aeskeygenassist xmm2, xmm1, 0x36    ; Generating round key 10
-        call key_expansion128
-
-	ret
-
-
-align 16
-global _iEncExpandKey192
-_iEncExpandKey192:
-
-	mov ecx,[esp-4+8]		;input
-	mov edx,[esp-4+12]		;ctx
-
-        movq xmm7, [ecx+16]	; loading the AES key
-        movq [edx+16], xmm7  ; Storing key in memory where all key expansion 
-        pshufd xmm4, xmm7, 01001111b
-        movdqu xmm1, [ecx]	; loading the AES key
-        movdqu [edx], xmm1  ; Storing key in memory where all key expansion 
-			
-        pxor xmm3, xmm3		; Set xmm3 to be all zeros. Required for the key_expansion. 
-        pxor xmm6, xmm6		; Set xmm3 to be all zeros. Required for the key_expansion. 
-
-        aeskeygenassist xmm2, xmm4, 0x1     ; Complete round key 1 and generate round key 2 
-        key_expansion_1_192 24
-	key_expansion_2_192 40				
-
-        aeskeygenassist xmm2, xmm4, 0x2     ; Generate round key 3 and part of round key 4
-        key_expansion_1_192 48
-	key_expansion_2_192 64				
-
-        aeskeygenassist xmm2, xmm4, 0x4     ; Complete round key 4 and generate round key 5
-        key_expansion_1_192 72
-	key_expansion_2_192 88
-		
-        aeskeygenassist xmm2, xmm4, 0x8     ; Generate round key 6 and part of round key 7
-        key_expansion_1_192 96
-	key_expansion_2_192 112
-		
-        aeskeygenassist xmm2, xmm4, 0x10     ; Complete round key 7 and generate round key 8 
-        key_expansion_1_192 120
-	key_expansion_2_192 136				
-
-        aeskeygenassist xmm2, xmm4, 0x20     ; Generate round key 9 and part of round key 10
-        key_expansion_1_192 144
-	key_expansion_2_192 160				
-
-        aeskeygenassist xmm2, xmm4, 0x40     ; Complete round key 10 and generate round key 11
-        key_expansion_1_192 168
-	key_expansion_2_192 184				
-
-        aeskeygenassist xmm2, xmm4, 0x80     ; Generate round key 12
-        key_expansion_1_192 192
-
-	ret
-
-
-
-
-
-
-align 16
-global _iDecExpandKey128
-_iDecExpandKey128:
-	push DWORD [esp+8]
-	push DWORD [esp+8]
-	
-	call _iEncExpandKey128
-	add esp,8
-
-	mov edx,[esp-4+12]		;ctx
-	
-	inversekey	[edx + 1*16]
-	inversekey	[edx + 2*16]
-	inversekey	[edx + 3*16]
-	inversekey	[edx + 4*16]
-	inversekey	[edx + 5*16]
-	inversekey	[edx + 6*16]
-	inversekey	[edx + 7*16]
-	inversekey	[edx + 8*16]
-	inversekey	[edx + 9*16]
-
-	ret
-
-
-
-
-align 16
-global _iDecExpandKey192
-_iDecExpandKey192:
-	push DWORD [esp+8]
-	push DWORD [esp+8]
-	
-	call _iEncExpandKey192
-	add esp,8
-
-	mov edx,[esp-4+12]		;ctx
-	
-	inversekey	[edx + 1*16]
-	inversekey	[edx + 2*16]
-	inversekey	[edx + 3*16]
-	inversekey	[edx + 4*16]
-	inversekey	[edx + 5*16]
-	inversekey	[edx + 6*16]
-	inversekey	[edx + 7*16]
-	inversekey	[edx + 8*16]
-	inversekey	[edx + 9*16]
-	inversekey	[edx + 10*16]
-	inversekey	[edx + 11*16]
-
-	ret
-
-
-
-
-align 16
-global _iDecExpandKey256
-_iDecExpandKey256:
-	push DWORD [esp+8]
-	push DWORD [esp+8]
-	
-	call _iEncExpandKey256
-	add esp, 8
-
-	mov edx, [esp-4+12]		;expanded key
-	
-	inversekey	[edx + 1*16]
-	inversekey	[edx + 2*16]
-	inversekey	[edx + 3*16]
-	inversekey	[edx + 4*16]
-	inversekey	[edx + 5*16]
-	inversekey	[edx + 6*16]
-	inversekey	[edx + 7*16]
-	inversekey	[edx + 8*16]
-	inversekey	[edx + 9*16]
-	inversekey	[edx + 10*16]
-	inversekey	[edx + 11*16]
-	inversekey	[edx + 12*16]
-	inversekey	[edx + 13*16]
-
-	ret
-	
-
-	
-	
-align 16
-global _iEncExpandKey256
-_iEncExpandKey256:
-	mov ecx, [esp-4+8]		;input
-	mov edx, [esp-4+12]		;expanded key
-
-
-    movdqu xmm1, [ecx]    ; loading the key
-    movdqu xmm3, [ecx+16]
-    movdqu [edx], xmm1  ; Storing key in memory where all key schedule will be stored
-    movdqu [edx+16], xmm3 
-    
-    add edx,32
-
-    movdqa xmm5, [shuffle_mask]  ; this mask is used by key_expansion
-
-    aeskeygenassist xmm2, xmm3, 0x1     ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x2     ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x4     ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x8     ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x10    ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x20    ; 
-    call key_expansion256
-    aeskeygenassist xmm2, xmm3, 0x40    ; 
-;    call key_expansion256 
-
-    pshufd xmm2, xmm2, 011111111b
-
-    movdqu xmm4, xmm1
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pshufb xmm4, xmm5
-    pxor xmm1, xmm4
-    pxor xmm1, xmm2
-
-    movdqu [edx], xmm1
-
-
-	ret
-	
-	
-	
-	
-	
-
-align 16
-global _iDec128
-_iDec128:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-	
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-
-	sub edi,esi
-	
-	test eax,eax
-	jz end_dec128
-
-	cmp eax,4
-	jl	lp128decsingle
-
-	test	ecx,0xf
-	jz		lp128decfour
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	mov ecx,esp	
-	
-
-align 16
-lp128decfour:
-	
-	test eax,eax
-	jz end_dec128
-
-	cmp eax,4
-	jl	lp128decsingle
-
-	load_and_xor4 esi, [ecx+10*16]
-	add esi,16*4
-	aesdec4 [ecx+9*16]
-	aesdec4 [ecx+8*16]
-	aesdec4 [ecx+7*16]
-	aesdec4 [ecx+6*16]
-	aesdec4 [ecx+5*16]
-	aesdec4 [ecx+4*16]
-	aesdec4 [ecx+3*16]
-	aesdec4 [ecx+2*16]
-	aesdec4 [ecx+1*16]
-	aesdeclast4 [ecx+0*16]
-	
-	sub eax,4
-	store4 esi+edi-(16*4)
-	jmp lp128decfour
-
-
-	align 16
-lp128decsingle:
-
-	movdqu xmm0, [esi]
-	movdqu xmm4,[ecx+10*16]
-	pxor xmm0, xmm4
-	aesdec1_u  [ecx+9*16]
-	aesdec1_u  [ecx+8*16]
-	aesdec1_u  [ecx+7*16]
-	aesdec1_u  [ecx+6*16]
-	aesdec1_u  [ecx+5*16]
-	aesdec1_u  [ecx+4*16]
-	aesdec1_u  [ecx+3*16]
-	aesdec1_u  [ecx+2*16]
-	aesdec1_u  [ecx+1*16]
-	aesdeclast1_u [ecx+0*16]
-
-	add esi, 16
-	movdqu  [edi+esi - 16], xmm0
-	dec eax
-	jnz lp128decsingle
-
-end_dec128:
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-	
-	ret
-
-
-
-align 16
-global _iDec128_CBC
-_iDec128_CBC:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	sub esp,16*16
-	and esp,0xfffffff0
-	
-	mov eax,[ecx+12]
-	movdqu xmm5,[eax]	;iv
-	
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	
-	sub edi,esi
-
-	test eax,eax
-	jz end_dec128_CBC
-
-	cmp eax,4
-	jl	lp128decsingle_CBC
-
-	test	ecx,0xf
-	jz		lp128decfour_CBC
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	mov ecx,esp	
-
-
-align 16
-lp128decfour_CBC:
-	
-	test eax,eax
-	jz end_dec128_CBC
-
-	cmp eax,4
-	jl	lp128decsingle_CBC
-
-	load_and_xor4 esi, [ecx+10*16]
-	add esi,16*4
-	aesdec4 [ecx+9*16]
-	aesdec4 [ecx+8*16]
-	aesdec4 [ecx+7*16]
-	aesdec4 [ecx+6*16]
-	aesdec4 [ecx+5*16]
-	aesdec4 [ecx+4*16]
-	aesdec4 [ecx+3*16]
-	aesdec4 [ecx+2*16]
-	aesdec4 [ecx+1*16]
-	aesdeclast4 [ecx+0*16]
-	
-	pxor	xmm0,xmm5
-	movdqu	xmm4,[esi- 16*4 + 0*16]
-	pxor	xmm1,xmm4
-	movdqu	xmm4,[esi- 16*4 + 1*16]
-	pxor	xmm2,xmm4
-	movdqu	xmm4,[esi- 16*4 + 2*16]
-	pxor	xmm3,xmm4
-	movdqu	xmm5,[esi- 16*4 + 3*16]
-	
-	sub eax,4
-	store4 esi+edi-(16*4)
-	jmp lp128decfour_CBC
-
-
-	align 16
-lp128decsingle_CBC:
-
-	movdqu xmm0, [esi]
-	movdqa xmm1,xmm0
-	movdqu xmm4,[ecx+10*16]
-	pxor xmm0, xmm4
-	aesdec1_u  [ecx+9*16]
-	aesdec1_u  [ecx+8*16]
-	aesdec1_u  [ecx+7*16]
-	aesdec1_u  [ecx+6*16]
-	aesdec1_u  [ecx+5*16]
-	aesdec1_u  [ecx+4*16]
-	aesdec1_u  [ecx+3*16]
-	aesdec1_u  [ecx+2*16]
-	aesdec1_u  [ecx+1*16]
-	aesdeclast1_u [ecx+0*16]
-	
-	pxor	xmm0,xmm5
-	movdqa	xmm5,xmm1
-	
-	add esi, 16
-	movdqu  [edi+esi - 16], xmm0
-	dec eax
-	jnz lp128decsingle_CBC
-
-end_dec128_CBC:
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-
-	mov ecx,[esp-4+8]   ; first arg
-	mov ecx,[ecx+12]
-	movdqu	[ecx],xmm5 ; store last iv for chaining
-	
-	ret
-
-
-
-
-
-
-align 16
-global _iDec192
-_iDec192:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-	
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	
-	sub edi,esi
-
-	test eax,eax
-	jz end_dec192
-
-	cmp eax,4
-	jl	lp192decsingle
-
-	test	ecx,0xf
-	jz		lp192decfour
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	copy_round_keys esp,ecx,11
-	copy_round_keys esp,ecx,12
-	mov ecx,esp	
-
-
-align 16
-lp192decfour:
-	
-	test eax,eax
-	jz end_dec192
-
-	cmp eax,4
-	jl	lp192decsingle
-
-	load_and_xor4 esi, [ecx+12*16]
-	add esi,16*4
-	aesdec4 [ecx+11*16]
-	aesdec4 [ecx+10*16]
-	aesdec4 [ecx+9*16]
-	aesdec4 [ecx+8*16]
-	aesdec4 [ecx+7*16]
-	aesdec4 [ecx+6*16]
-	aesdec4 [ecx+5*16]
-	aesdec4 [ecx+4*16]
-	aesdec4 [ecx+3*16]
-	aesdec4 [ecx+2*16]
-	aesdec4 [ecx+1*16]
-	aesdeclast4 [ecx+0*16]
-	
-	sub eax,4
-	store4 esi+edi-(16*4)
-	jmp lp192decfour
-
-
-	align 16
-lp192decsingle:
-
-	movdqu xmm0, [esi]
-	movdqu xmm4,[ecx+12*16]
-	pxor xmm0, xmm4
-	aesdec1_u [ecx+11*16]
-	aesdec1_u  [ecx+10*16]
-	aesdec1_u  [ecx+9*16]
-	aesdec1_u  [ecx+8*16]
-	aesdec1_u  [ecx+7*16]
-	aesdec1_u  [ecx+6*16]
-	aesdec1_u  [ecx+5*16]
-	aesdec1_u  [ecx+4*16]
-	aesdec1_u  [ecx+3*16]
-	aesdec1_u  [ecx+2*16]
-	aesdec1_u  [ecx+1*16]
-	aesdeclast1_u  [ecx+0*16]
-
-	add esi, 16
-	movdqu  [edi+esi - 16], xmm0
-	dec eax
-	jnz lp192decsingle
-
-end_dec192:
-
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-	
-	ret
-
-
-align 16
-global _iDec192_CBC
-_iDec192_CBC:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov eax,[ecx+12]
-	movdqu xmm5,[eax]	;iv
-	
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	
-	sub edi,esi
-
-	test eax,eax
-	jz end_dec192_CBC
-
-	cmp eax,4
-	jl	lp192decsingle_CBC
-
-	test	ecx,0xf
-	jz		lp192decfour_CBC
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	copy_round_keys esp,ecx,11
-	copy_round_keys esp,ecx,12
-	mov ecx,esp	
-
-align 16
-lp192decfour_CBC:
-	
-	test eax,eax
-	jz end_dec192_CBC
-
-	cmp eax,4
-	jl	lp192decsingle_CBC
-
-	load_and_xor4 esi, [ecx+12*16]
-	add esi,16*4
-	aesdec4 [ecx+11*16]
-	aesdec4 [ecx+10*16]
-	aesdec4 [ecx+9*16]
-	aesdec4 [ecx+8*16]
-	aesdec4 [ecx+7*16]
-	aesdec4 [ecx+6*16]
-	aesdec4 [ecx+5*16]
-	aesdec4 [ecx+4*16]
-	aesdec4 [ecx+3*16]
-	aesdec4 [ecx+2*16]
-	aesdec4 [ecx+1*16]
-	aesdeclast4 [ecx+0*16]
-	
-	pxor	xmm0,xmm5
-	movdqu	xmm4,[esi- 16*4 + 0*16]
-	pxor	xmm1,xmm4
-	movdqu	xmm4,[esi- 16*4 + 1*16]
-	pxor	xmm2,xmm4
-	movdqu	xmm4,[esi- 16*4 + 2*16]
-	pxor	xmm3,xmm4
-	movdqu	xmm5,[esi- 16*4 + 3*16]
-	
-	sub eax,4
-	store4 esi+edi-(16*4)
-	jmp lp192decfour_CBC
-
-
-	align 16
-lp192decsingle_CBC:
-
-	movdqu xmm0, [esi]
-	movdqu xmm4,[ecx+12*16]
-	movdqa xmm1,xmm0
-	pxor xmm0, xmm4
-	aesdec1_u [ecx+11*16]
-	aesdec1_u [ecx+10*16]
-	aesdec1_u [ecx+9*16]
-	aesdec1_u [ecx+8*16]
-	aesdec1_u [ecx+7*16]
-	aesdec1_u [ecx+6*16]
-	aesdec1_u [ecx+5*16]
-	aesdec1_u [ecx+4*16]
-	aesdec1_u [ecx+3*16]
-	aesdec1_u [ecx+2*16]
-	aesdec1_u [ecx+1*16]
-	aesdeclast1_u [ecx+0*16]
-	
-	pxor	xmm0,xmm5
-	movdqa	xmm5,xmm1
-	
-	add esi, 16
-	movdqu  [edi+esi - 16], xmm0
-	dec eax
-	jnz lp192decsingle_CBC
-
-end_dec192_CBC:
-
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-
-	mov ecx,[esp-4+8]
-	mov ecx,[ecx+12]
-	movdqu	[ecx],xmm5 ; store last iv for chaining
-	
-	ret
-
-
-
-
-
-align 16
-global _iDec256
-_iDec256:
-	mov ecx, [esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	
-	sub edi,esi
-
-
-	test eax,eax
-	jz end_dec256
-	
-	cmp eax,4
-	jl lp256dec
-
-	test	ecx,0xf
-	jz	lp256dec4
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	copy_round_keys esp,ecx,11
-	copy_round_keys esp,ecx,12
-	copy_round_keys esp,ecx,13
-	copy_round_keys esp,ecx,14
-	mov ecx,esp	
-	
-	align 16
-lp256dec4:
-	test eax,eax
-	jz end_dec256
-	
-	cmp eax,4
-	jl lp256dec
-	
-	load_and_xor4 esi,[ecx+14*16]
-	add esi, 4*16
-	aesdec4 [ecx+13*16]
-	aesdec4 [ecx+12*16]
-	aesdec4 [ecx+11*16]
-	aesdec4 [ecx+10*16]
-	aesdec4 [ecx+9*16]
-	aesdec4 [ecx+8*16]
-	aesdec4 [ecx+7*16]
-	aesdec4 [ecx+6*16]
-	aesdec4 [ecx+5*16]
-	aesdec4 [ecx+4*16]
-	aesdec4 [ecx+3*16]
-	aesdec4 [ecx+2*16]
-	aesdec4 [ecx+1*16]
-	aesdeclast4 [ecx+0*16]
-
-	store4 esi+edi-16*4
-	sub eax,4
-	jmp lp256dec4	
-	
-	align 16
-lp256dec:
-
-	movdqu xmm0, [esi]
-	movdqu xmm4,[ecx+14*16]
-	add esi, 16
-	pxor xmm0, xmm4                     ; Round 0 (only xor)
-	aesdec1_u  [ecx+13*16]
-	aesdec1_u  [ecx+12*16]
-	aesdec1_u  [ecx+11*16]
-	aesdec1_u  [ecx+10*16]
-	aesdec1_u  [ecx+9*16]
-	aesdec1_u  [ecx+8*16]
-	aesdec1_u  [ecx+7*16]
-	aesdec1_u  [ecx+6*16]
-	aesdec1_u  [ecx+5*16]
-	aesdec1_u  [ecx+4*16]
-	aesdec1_u  [ecx+3*16]
-	aesdec1_u  [ecx+2*16]
-	aesdec1_u  [ecx+1*16]
-	aesdeclast1_u  [ecx+0*16]
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [esi+edi-16], xmm0
-	dec eax
-	jnz lp256dec
-
-end_dec256:
-
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-	
-	ret
-
-
-
-
-align 16
-global _iDec256_CBC
-_iDec256_CBC:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov eax,[ecx+12]
-	movdqu xmm5,[eax]	;iv
-	
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	
-	sub edi,esi
-
-	test eax,eax
-	jz end_dec256_CBC
-
-	cmp eax,4
-	jl	lp256decsingle_CBC
-
-	test	ecx,0xf
-	jz	lp256decfour_CBC
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	copy_round_keys esp,ecx,11
-	copy_round_keys esp,ecx,12
-	copy_round_keys esp,ecx,13
-	copy_round_keys esp,ecx,14
-	mov ecx,esp	
-
-align 16
-lp256decfour_CBC:
-	
-	test eax,eax
-	jz end_dec256_CBC
-
-	cmp eax,4
-	jl	lp256decsingle_CBC
-
-	load_and_xor4 esi, [ecx+14*16]
-	add esi,16*4
-	aesdec4 [ecx+13*16]
-	aesdec4 [ecx+12*16]
-	aesdec4 [ecx+11*16]
-	aesdec4 [ecx+10*16]
-	aesdec4 [ecx+9*16]
-	aesdec4 [ecx+8*16]
-	aesdec4 [ecx+7*16]
-	aesdec4 [ecx+6*16]
-	aesdec4 [ecx+5*16]
-	aesdec4 [ecx+4*16]
-	aesdec4 [ecx+3*16]
-	aesdec4 [ecx+2*16]
-	aesdec4 [ecx+1*16]
-	aesdeclast4 [ecx+0*16]
-	
-	pxor	xmm0,xmm5
-	movdqu	xmm4,[esi- 16*4 + 0*16]
-	pxor	xmm1,xmm4
-	movdqu	xmm4,[esi- 16*4 + 1*16]
-	pxor	xmm2,xmm4
-	movdqu	xmm4,[esi- 16*4 + 2*16]
-	pxor	xmm3,xmm4
-	movdqu	xmm5,[esi- 16*4 + 3*16]
-	
-	sub eax,4
-	store4 esi+edi-(16*4)
-	jmp lp256decfour_CBC
-
-
-	align 16
-lp256decsingle_CBC:
-
-	movdqu xmm0, [esi]
-	movdqa xmm1,xmm0
-	movdqu xmm4, [ecx+14*16]
-	pxor xmm0, xmm4
-	aesdec1_u  [ecx+13*16]
-	aesdec1_u  [ecx+12*16]
-	aesdec1_u  [ecx+11*16]
-	aesdec1_u  [ecx+10*16]
-	aesdec1_u  [ecx+9*16]
-	aesdec1_u  [ecx+8*16]
-	aesdec1_u  [ecx+7*16]
-	aesdec1_u  [ecx+6*16]
-	aesdec1_u  [ecx+5*16]
-	aesdec1_u  [ecx+4*16]
-	aesdec1_u  [ecx+3*16]
-	aesdec1_u  [ecx+2*16]
-	aesdec1_u  [ecx+1*16]
-	aesdeclast1_u  [ecx+0*16]
-	
-	pxor	xmm0,xmm5
-	movdqa	xmm5,xmm1
-	
-	add esi, 16
-	movdqu  [edi+esi - 16], xmm0
-	dec eax
-	jnz lp256decsingle_CBC
-
-end_dec256_CBC:
-
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-
-	mov ecx,[esp-4+8]  ; first arg
-	mov ecx,[ecx+12]
-	movdqu	[ecx],xmm5 ; store last iv for chaining
-	
-	ret
-
-
-
-
-
-
-
-
-
-align 16
-global _iEnc128
-_iEnc128:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	
-	sub edi,esi
-
-	test eax,eax
-	jz end_enc128
-	
-	cmp eax,4
-	jl lp128encsingle
-
-	test	ecx,0xf
-	jz		lpenc128four
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	mov ecx,esp	
-
-
-	align 16	
-	
-lpenc128four:
-	
-	test eax,eax
-	jz end_enc128
-	
-	cmp eax,4
-	jl lp128encsingle
-
-	load_and_xor4 esi,[ecx+0*16]
-	add esi,4*16
-	aesenc4	[ecx+1*16]
-	aesenc4	[ecx+2*16]
-	aesenc4	[ecx+3*16]
-	aesenc4	[ecx+4*16]
-	aesenc4	[ecx+5*16]
-	aesenc4	[ecx+6*16]
-	aesenc4	[ecx+7*16]
-	aesenc4	[ecx+8*16]
-	aesenc4	[ecx+9*16]
-	aesenclast4	[ecx+10*16]
-	
-	store4 esi+edi-16*4
-	sub eax,4
-	jmp lpenc128four
-	
-	align 16
-lp128encsingle:
-
-	movdqu xmm0, [esi]
-	add esi, 16
-	movdqu xmm4,[ecx+0*16]
-	pxor xmm0, xmm4
-	aesenc1_u  [ecx+1*16]
-	aesenc1_u  [ecx+2*16]
-	aesenc1_u  [ecx+3*16]
-	aesenc1_u  [ecx+4*16]     
-	aesenc1_u  [ecx+5*16]
-	aesenc1_u  [ecx+6*16]
-	aesenc1_u  [ecx+7*16]
-	aesenc1_u  [ecx+8*16]
-	aesenc1_u  [ecx+9*16]
-	aesenclast1_u  [ecx+10*16]
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [esi+edi-16], xmm0
-	dec eax
-	jnz lp128encsingle
-
-end_enc128:
-
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-	
-	ret
-
-
-align 16
-global _iEnc128_CTR
-_iEnc128_CTR:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov	eax,[ecx+12]
-	movdqu xmm5,[eax]	;initial counter
-
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	
-	sub edi,esi
-
-	test eax,eax
-	jz end_encctr128
-	
-	cmp eax,4
-	jl lp128encctrsingle
-
-	test	ecx,0xf
-	jz		lpencctr128four
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	mov ecx,esp	
-
-
-	align 16	
-	
-lpencctr128four:
-	
-	test eax,eax
-	jz end_encctr128
-	
-	cmp eax,4
-	jl lp128encsingle
-
-	load_and_inc4 [ecx+0*16]
-	add esi,4*16
-	aesenc4	[ecx+1*16]
-	aesenc4	[ecx+2*16]
-	aesenc4	[ecx+3*16]
-	aesenc4	[ecx+4*16]
-	aesenc4	[ecx+5*16]
-	aesenc4	[ecx+6*16]
-	aesenc4	[ecx+7*16]
-	aesenc4	[ecx+8*16]
-	aesenc4	[ecx+9*16]
-	aesenclast4	[ecx+10*16]
-	xor_with_input4 esi-(4*16)
-	
-	store4 esi+edi-16*4
-	sub eax,4
-	jmp lpencctr128four
-	
-	align 16
-lp128encctrsingle:
-
-	movdqa	xmm0,xmm5
-	paddq	xmm5,[counter_add_one]
-	add esi, 16
-	movdqu xmm4,[ecx+0*16]
-	pxor xmm0, xmm4
-	aesenc1_u [ecx+1*16]
-	aesenc1_u [ecx+2*16]
-	aesenc1_u [ecx+3*16]
-	aesenc1_u [ecx+4*16]     
-	aesenc1_u [ecx+5*16]
-	aesenc1_u [ecx+6*16]
-	aesenc1_u [ecx+7*16]
-	aesenc1_u [ecx+8*16]
-	aesenc1_u [ecx+9*16]
-	aesenclast1_u [ecx+10*16]
-	movdqu xmm4, [esi-16]
-	pxor	xmm0,xmm4
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [esi+edi-16], xmm0
-	dec eax
-	jnz lp128encctrsingle
-
-end_encctr128:
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-
-	mov ecx,[esp-4+8]  ; first arg
-	mov ecx,[ecx+12]
-	movdqu	[ecx],xmm5 ; store last counter for chaining
-	
-	ret
-
-
-align 16
-global _iEnc192_CTR
-_iEnc192_CTR:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov	eax,[ecx+12]
-	movdqu xmm5,[eax]	;initial counter
-
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	
-	sub edi,esi
-
-	test eax,eax
-	jz end_encctr192
-	
-	cmp eax,4
-	jl lp192encctrsingle
-
-	test	ecx,0xf
-	jz		lpencctr128four
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	copy_round_keys esp,ecx,11
-	copy_round_keys esp,ecx,12
-	mov ecx,esp	
-
-
-	align 16	
-	
-lpencctr192four:
-	
-	test eax,eax
-	jz end_encctr192
-	
-	cmp eax,4
-	jl lp192encsingle
-
-	load_and_inc4 [ecx+0*16]
-	add esi,4*16
-	aesenc4	[ecx+1*16]
-	aesenc4	[ecx+2*16]
-	aesenc4	[ecx+3*16]
-	aesenc4	[ecx+4*16]
-	aesenc4	[ecx+5*16]
-	aesenc4	[ecx+6*16]
-	aesenc4	[ecx+7*16]
-	aesenc4	[ecx+8*16]
-	aesenc4	[ecx+9*16]
-	aesenc4	[ecx+10*16]
-	aesenc4	[ecx+11*16]
-	aesenclast4	[ecx+12*16]
-	xor_with_input4 esi-(4*16)
-	
-	store4 esi+edi-16*4
-	sub eax,4
-	jmp lpencctr192four
-	
-	align 16
-lp192encctrsingle:
-
-	movdqa	xmm0,xmm5
-	paddq	xmm5,[counter_add_one]
-	add esi, 16
-	movdqu xmm4,[ecx+0*16]
-	pxor xmm0, xmm4
-	aesenc1_u  [ecx+1*16]
-	aesenc1_u  [ecx+2*16]
-	aesenc1_u  [ecx+3*16]
-	aesenc1_u  [ecx+4*16]     
-	aesenc1_u  [ecx+5*16]
-	aesenc1_u  [ecx+6*16]
-	aesenc1_u  [ecx+7*16]
-	aesenc1_u  [ecx+8*16]
-	aesenc1_u  [ecx+9*16]
-	aesenc1_u  [ecx+10*16]
-	aesenc1_u  [ecx+11*16]
-	aesenclast1_u  [ecx+12*16]
-	movdqu xmm4, [esi-16]
-	pxor	xmm0,xmm4
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [esi+edi-16], xmm0
-	dec eax
-	jnz lp192encctrsingle
-
-end_encctr192:
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-
-	mov ecx,[esp-4+8]  ; first arg
-	mov ecx,[ecx+12]
-	movdqu	[ecx],xmm5 ; store last counter for chaining
-	
-	ret
-
-
-align 16
-global _iEnc256_CTR
-_iEnc256_CTR:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov	eax,[ecx+12]
-	movdqu xmm5,[eax]	;initial counter
-
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	
-	sub edi,esi
-
-	test eax,eax
-	jz end_encctr256
-	
-	cmp eax,4
-	jl lp256encctrsingle
-
-	test	ecx,0xf
-	jz		lpencctr128four
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	copy_round_keys esp,ecx,11
-	copy_round_keys esp,ecx,12
-	copy_round_keys esp,ecx,13
-	copy_round_keys esp,ecx,14
-	mov ecx,esp	
-
-
-	align 16	
-	
-lpencctr256four:
-	
-	test eax,eax
-	jz end_encctr256
-	
-	cmp eax,4
-	jl lp256encctrsingle
-
-	load_and_inc4 [ecx+0*16]
-	add esi,4*16
-	aesenc4	[ecx+1*16]
-	aesenc4	[ecx+2*16]
-	aesenc4	[ecx+3*16]
-	aesenc4	[ecx+4*16]
-	aesenc4	[ecx+5*16]
-	aesenc4	[ecx+6*16]
-	aesenc4	[ecx+7*16]
-	aesenc4	[ecx+8*16]
-	aesenc4	[ecx+9*16]
-	aesenc4	[ecx+10*16]
-	aesenc4	[ecx+11*16]
-	aesenc4	[ecx+12*16]
-	aesenc4	[ecx+13*16]
-	aesenclast4	[ecx+14*16]
-	xor_with_input4 esi-(4*16)
-	
-	store4 esi+edi-16*4
-	sub eax,4
-	jmp lpencctr256four
-	
-	align 16
-	
-lp256encctrsingle:
-
-	movdqa	xmm0,xmm5
-	paddq	xmm5,[counter_add_one]
-	add esi, 16
-	movdqu xmm4,[ecx+0*16]
-	pxor xmm0, xmm4
-	aesenc1_u  [ecx+1*16]
-	aesenc1_u  [ecx+2*16]
-	aesenc1_u  [ecx+3*16]
-	aesenc1_u  [ecx+4*16]     
-	aesenc1_u  [ecx+5*16]
-	aesenc1_u  [ecx+6*16]
-	aesenc1_u  [ecx+7*16]
-	aesenc1_u  [ecx+8*16]
-	aesenc1_u  [ecx+9*16]
-	aesenc1_u  [ecx+10*16]
-	aesenc1_u  [ecx+11*16]
-	aesenc1_u  [ecx+12*16]
-	aesenc1_u  [ecx+13*16]
-	aesenclast1_u  [ecx+14*16]
-	movdqu xmm4, [esi-16]
-	pxor	xmm0,xmm4
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [esi+edi-16], xmm0
-	dec eax
-	jnz lp256encctrsingle
-
-end_encctr256:
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-
-	mov ecx,[esp-4+8]  ; first arg
-	mov ecx,[ecx+12]
-	movdqu	[ecx],xmm5 ; store last counter for chaining
-	
-	ret
-
-
-
-
-
-
-align 16
-global _iEnc128_CBC
-_iEnc128_CBC:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov	eax,[ecx+12]
-	movdqu xmm1,[eax]	;iv	
-	
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	sub edi,esi
-
-	test	ecx,0xf
-	jz		lp128encsingle_CBC
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	mov ecx,esp	
-
-	align 16	
-	
-lp128encsingle_CBC:
-
-	movdqu xmm0, [esi]
-	add esi, 16
-	pxor xmm0, xmm1
-	movdqu xmm4,[ecx+0*16]
-	pxor xmm0, xmm4
-	aesenc1  [ecx+1*16]
-	aesenc1  [ecx+2*16]
-	aesenc1  [ecx+3*16]
-	aesenc1  [ecx+4*16]     
-	aesenc1  [ecx+5*16]
-	aesenc1  [ecx+6*16]
-	aesenc1  [ecx+7*16]
-	aesenc1  [ecx+8*16]
-	aesenc1  [ecx+9*16]
-	aesenclast1  [ecx+10*16]
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [esi+edi-16], xmm0
-	movdqa xmm1,xmm0
-	dec eax
-	jnz lp128encsingle_CBC
-
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-	mov ecx,[esp-4+8]  ; first arg
-	mov ecx,[ecx+12]
-	movdqu	[ecx],xmm1 ; store last iv for chaining
-	
-	ret
-
-
-align 16
-global _iEnc192_CBC
-_iEnc192_CBC:
-	mov ecx,[esp-4+8]  ; first arg
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov	eax,[ecx+12]
-	movdqu xmm1,[eax]	;iv	
-	
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	sub edi,esi
-
-	test	ecx,0xf
-	jz		lp192encsingle_CBC
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	copy_round_keys esp,ecx,11
-	copy_round_keys esp,ecx,12
-	mov ecx,esp	
-
-	align 16	
-	
-lp192encsingle_CBC:
-
-	movdqu xmm0, [esi]
-	add esi, 16
-	pxor xmm0, xmm1
-	movdqu xmm4,[ecx+0*16]
-	pxor xmm0, xmm4
-	aesenc1  [ecx+1*16]
-	aesenc1  [ecx+2*16]
-	aesenc1  [ecx+3*16]
-	aesenc1  [ecx+4*16]     
-	aesenc1  [ecx+5*16]
-	aesenc1  [ecx+6*16]
-	aesenc1  [ecx+7*16]
-	aesenc1  [ecx+8*16]
-	aesenc1  [ecx+9*16]
-	aesenc1  [ecx+10*16]
-	aesenc1  [ecx+11*16]
-	aesenclast1  [ecx+12*16]
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [esi+edi-16], xmm0
-	movdqa xmm1,xmm0
-	dec eax
-	jnz lp192encsingle_CBC
-
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-	mov ecx,[esp-4+8]  ; first arg
-	mov ecx,[ecx+12]
-	movdqu	[ecx],xmm1 ; store last iv for chaining
-	
-	ret
-
-align 16
-global _iEnc256_CBC
-_iEnc256_CBC:
-	mov ecx,[esp-4+8]  ; first arg
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov	eax,[ecx+12]
-	movdqu xmm1,[eax]	;iv	
-	
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	sub edi,esi
-
-	test	ecx,0xf
-	jz		lp256encsingle_CBC
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	copy_round_keys esp,ecx,11
-	copy_round_keys esp,ecx,12
-	copy_round_keys esp,ecx,13
-	copy_round_keys esp,ecx,14
-	mov ecx,esp	
-
-	align 16	
-	
-lp256encsingle_CBC:
-
-;abab
-	movdqu xmm0, [esi]
-	add esi, 16
-	pxor xmm0, xmm1
-	movdqu xmm4,[ecx+0*16]
-	pxor xmm0, xmm4
-	aesenc1 [ecx+1*16]
-	aesenc1 [ecx+2*16]
-	aesenc1 [ecx+3*16]
-	aesenc1 [ecx+4*16]     
-	aesenc1 [ecx+5*16]
-	aesenc1 [ecx+6*16]
-	aesenc1 [ecx+7*16]
-	aesenc1 [ecx+8*16]
-	aesenc1 [ecx+9*16]
-	aesenc1 [ecx+10*16]
-	aesenc1 [ecx+11*16]
-	aesenc1 [ecx+12*16]
-	aesenc1 [ecx+13*16]
-	aesenclast1 [ecx+14*16]
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [esi+edi-16], xmm0
-	movdqa xmm1,xmm0
-	dec eax
-	jnz lp256encsingle_CBC
-
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-	mov ecx,[esp-4+8]
-	mov ecx,[ecx+12]
-	movdqu	[ecx],xmm1 ; store last iv for chaining
-	
-	ret
-
-
-
-
-
-align 16
-global _iEnc192
-_iEnc192:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	
-	sub edi,esi
-
-	test eax,eax
-	jz end_enc192
-	
-	cmp eax,4
-	jl lp192encsingle
-
-	test	ecx,0xf
-	jz		lpenc192four
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	copy_round_keys esp,ecx,11
-	copy_round_keys esp,ecx,12
-	mov ecx,esp	
-
-	align 16	
-	
-lpenc192four:
-	
-	test eax,eax
-	jz end_enc192
-	
-	cmp eax,4
-	jl lp192encsingle
-
-	load_and_xor4 esi,[ecx+0*16]
-	add esi,4*16
-	aesenc4	[ecx+1*16]
-	aesenc4	[ecx+2*16]
-	aesenc4	[ecx+3*16]
-	aesenc4	[ecx+4*16]
-	aesenc4	[ecx+5*16]
-	aesenc4	[ecx+6*16]
-	aesenc4	[ecx+7*16]
-	aesenc4	[ecx+8*16]
-	aesenc4	[ecx+9*16]
-	aesenc4	[ecx+10*16]
-	aesenc4	[ecx+11*16]
-	aesenclast4	[ecx+12*16]
-	
-	store4 esi+edi-16*4
-	sub eax,4
-	jmp lpenc192four
-	
-	align 16
-lp192encsingle:
-
-	movdqu xmm0, [esi]
-	add esi, 16
-	movdqu xmm4,[ecx+0*16]
-	pxor xmm0, xmm4
-	aesenc1_u [ecx+1*16]
-	aesenc1_u [ecx+2*16]
-	aesenc1_u [ecx+3*16]
-	aesenc1_u [ecx+4*16]     
-	aesenc1_u [ecx+5*16]
-	aesenc1_u [ecx+6*16]
-	aesenc1_u [ecx+7*16]
-	aesenc1_u [ecx+8*16]
-	aesenc1_u [ecx+9*16]
-	aesenc1_u [ecx+10*16]
-	aesenc1_u [ecx+11*16]
-	aesenclast1_u [ecx+12*16]
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [esi+edi-16], xmm0
-	dec eax
-	jnz lp192encsingle
-
-end_enc192:
-
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-	
-	ret
-
-
-
-
-align 16
-global _iEnc256
-_iEnc256:
-	mov ecx,[esp-4+8]
-	
-	push esi
-	push edi
-	push ebp
-	mov ebp,esp
-	
-	sub esp,16*16
-	and esp,0xfffffff0
-
-	mov eax,[ecx+16] ; numblocks
-	mov esi,[ecx]
-	mov edi,[ecx+4]
-	mov ecx,[ecx+8]
-	
-	sub edi,esi	
-
-	test eax,eax
-	jz end_enc256
-
-	cmp eax,4
-	jl lp256enc
-
-	test	ecx,0xf
-	jz	lp256enc4
-	
-	copy_round_keys esp,ecx,0
-	copy_round_keys esp,ecx,1
-	copy_round_keys esp,ecx,2
-	copy_round_keys esp,ecx,3
-	copy_round_keys esp,ecx,4
-	copy_round_keys esp,ecx,5
-	copy_round_keys esp,ecx,6
-	copy_round_keys esp,ecx,7
-	copy_round_keys esp,ecx,8
-	copy_round_keys esp,ecx,9
-	copy_round_keys esp,ecx,10
-	copy_round_keys esp,ecx,11
-	copy_round_keys esp,ecx,12
-	copy_round_keys esp,ecx,13
-	copy_round_keys esp,ecx,14
-	mov ecx,esp	
-
-
-
-	align 16
-	
-lp256enc4:
-	test eax,eax
-	jz end_enc256
-
-	cmp eax,4
-	jl lp256enc
-
-
-	load_and_xor4 esi,[ecx+0*16]
-	add esi, 16*4
-	aesenc4 [ecx+1*16]
-	aesenc4 [ecx+2*16]
-	aesenc4 [ecx+3*16]
-	aesenc4 [ecx+4*16]
-	aesenc4 [ecx+5*16]
-	aesenc4 [ecx+6*16]
-	aesenc4 [ecx+7*16]
-	aesenc4 [ecx+8*16]
-	aesenc4 [ecx+9*16]
-	aesenc4 [ecx+10*16]
-	aesenc4 [ecx+11*16]
-	aesenc4 [ecx+12*16]
-	aesenc4 [ecx+13*16]
-	aesenclast4 [ecx+14*16]
-
-	store4  esi+edi-16*4
-	sub eax,4
-	jmp lp256enc4
-	
-	align 16
-lp256enc:
-
-	movdqu xmm0, [esi]
-	add esi, 16
-	movdqu xmm4,[ecx+0*16]
-	pxor xmm0, xmm4
-	aesenc1_u [ecx+1*16]
-	aesenc1_u [ecx+2*16]
-	aesenc1_u [ecx+3*16]
-	aesenc1_u [ecx+4*16]
-	aesenc1_u [ecx+5*16]
-	aesenc1_u [ecx+6*16]
-	aesenc1_u [ecx+7*16]
-	aesenc1_u [ecx+8*16]
-	aesenc1_u [ecx+9*16]
-	aesenc1_u [ecx+10*16]
-	aesenc1_u [ecx+11*16]
-	aesenc1_u [ecx+12*16]
-	aesenc1_u [ecx+13*16]
-	aesenclast1_u [ecx+14*16]
-
-		; Store output encrypted data into CIPHERTEXT array
-	movdqu  [esi+edi-16], xmm0
-	dec eax
-	jnz lp256enc
-
-end_enc256:
-
-
-	mov esp,ebp
-	pop ebp
-	pop edi
-	pop esi
-	
-	ret
diff --git a/lib/accelerated/intel/iaes_asm_interface.h b/lib/accelerated/intel/iaes_asm_interface.h
deleted file mode 100755
index 093d2a56e3..0000000000
--- a/lib/accelerated/intel/iaes_asm_interface.h
+++ /dev/null
@@ -1,126 +0,0 @@
-/* 
- * Copyright (c) 2010, Intel Corporation
- * All rights reserved.
- * 
- * Redistribution and use in source and binary forms, with or without 
- * modification, are permitted provided that the following conditions are met:
- * 
- *     * Redistributions of source code must retain the above copyright notice, 
- *       this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright notice, 
- *       this list of conditions and the following disclaimer in the documentation 
- *       and/or other materials provided with the distribution.
- *     * Neither the name of Intel Corporation nor the names of its contributors 
- *       may be used to endorse or promote products derived from this software 
- *       without specific prior written permission.
- * 
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND 
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 
- * IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, 
- * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, 
- * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 
- * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 
- * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE 
- * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF 
- * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- * 
-*/
-
-#ifndef _INTEL_AES_ASM_INTERFACE_H__
-#define _INTEL_AES_ASM_INTERFACE_H__
-
-
-#include "iaesni.h"
-
-
-
-//structure to pass aes processing data to asm level functions
-typedef struct _sAesData
-{
-	_AES_IN		UCHAR	*in_block;
-	_AES_OUT	UCHAR	*out_block;
-	_AES_IN		UCHAR	*expanded_key;		
-	_AES_INOUT	UCHAR	*iv;					// for CBC mode
-	_AES_IN		size_t	num_blocks;
-} sAesData;
-
-#if (__cplusplus)
-extern "C"
-{
-#endif
-#if 0
-#define MYSTDCALL __stdcall
-#else
-#define MYSTDCALL 
-#endif
-
-#ifdef __linux__
-#ifndef __LP64__
-#define iEncExpandKey256 _iEncExpandKey256
-#define iEncExpandKey192 _iEncExpandKey192
-#define iEncExpandKey128 _iEncExpandKey128
-#define iDecExpandKey256 _iDecExpandKey256
-#define iDecExpandKey192 _iDecExpandKey192
-#define iDecExpandKey128 _iDecExpandKey128
-#define iEnc128 _iEnc128
-#define iDec128 _iDec128
-#define iEnc256 _iEnc256
-#define iDec256 _iDec256
-#define iEnc192 _iEnc192
-#define iDec192 _iDec192
-#define iEnc128_CBC _iEnc128_CBC
-#define iDec128_CBC _iDec128_CBC
-#define iEnc256_CBC _iEnc256_CBC
-#define iDec256_CBC _iDec256_CBC
-#define iEnc192_CBC _iEnc192_CBC
-#define iDec192_CBC _iDec192_CBC
-#define iEnc128_CTR _iEnc128_CTR
-#define iEnc192_CTR _iEnc192_CTR
-#define iEnc256_CTR _iEnc256_CTR
-#define do_rdtsc    _do_rdtsc
-#endif
-#endif
-	// prepearing the different key rounds, for enc/dec in asm
-	// expnaded key should be 16-byte aligned
-	// expanded key should have enough space to hold all key rounds (16 bytes per round) - 256 bytes would cover all cases (AES256 has 14 rounds + 1 xor)
-	void MYSTDCALL iEncExpandKey256(_AES_IN UCHAR *key, _AES_OUT UCHAR *expanded_key);
-	void MYSTDCALL iEncExpandKey192(_AES_IN UCHAR *key, _AES_OUT UCHAR *expanded_key);
-	void MYSTDCALL iEncExpandKey128(_AES_IN UCHAR *key, _AES_OUT UCHAR *expanded_key);
-
-	void MYSTDCALL iDecExpandKey256(UCHAR *key, _AES_OUT UCHAR *expanded_key);
-	void MYSTDCALL iDecExpandKey192(UCHAR *key, _AES_OUT UCHAR *expanded_key);
-	void MYSTDCALL iDecExpandKey128(UCHAR *key, _AES_OUT UCHAR *expanded_key);
-
-
-	//enc/dec asm functions
-	void MYSTDCALL iEnc128(sAesData *data);
-	void MYSTDCALL iDec128(sAesData *data);
-	void MYSTDCALL iEnc256(sAesData *data);
-	void MYSTDCALL iDec256(sAesData *data);
-	void MYSTDCALL iEnc192(sAesData *data);
-	void MYSTDCALL iDec192(sAesData *data);
-
-	void MYSTDCALL iEnc128_CBC(sAesData *data);
-	void MYSTDCALL iDec128_CBC(sAesData *data);
-	void MYSTDCALL iEnc256_CBC(sAesData *data);
-	void MYSTDCALL iDec256_CBC(sAesData *data);
-	void MYSTDCALL iEnc192_CBC(sAesData *data);
-	void MYSTDCALL iDec192_CBC(sAesData *data);
-
-
-	void MYSTDCALL iEnc128_CTR(sAesData *data);
-	void MYSTDCALL iEnc256_CTR(sAesData *data);
-	void MYSTDCALL iEnc192_CTR(sAesData *data);
-
-	// rdtsc function
-	unsigned long long do_rdtsc(void);
-
-
-#if (__cplusplus)
-}
-#endif
-
-
-#endif
-
diff --git a/lib/accelerated/intel/iaesni.h b/lib/accelerated/intel/iaesni.h
deleted file mode 100755
index b358faa7ee..0000000000
--- a/lib/accelerated/intel/iaesni.h
+++ /dev/null
@@ -1,147 +0,0 @@
-/* 
- * Copyright (c) 2010, Intel Corporation
- * All rights reserved.
- * 
- * Redistribution and use in source and binary forms, with or without 
- * modification, are permitted provided that the following conditions are met:
- * 
- *     * Redistributions of source code must retain the above copyright notice, 
- *       this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright notice, 
- *       this list of conditions and the following disclaimer in the documentation 
- *       and/or other materials provided with the distribution.
- *     * Neither the name of Intel Corporation nor the names of its contributors 
- *       may be used to endorse or promote products derived from this software 
- *       without specific prior written permission.
- * 
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND 
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 
- * IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, 
- * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, 
- * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 
- * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 
- * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE 
- * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF 
- * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- * 
-*/
-
-
-#ifndef _IAESNI_H__
-#define _IAESNI_H__
-
-#include <stdlib.h>
-
-#define AES_INSTRCTIONS_CPUID_BIT (1<<25)
-
-//indicates input param
-#define _AES_IN	
-
-//indicates output param
-#define _AES_OUT
-
-//indicates input/output param - based on context
-#define _AES_INOUT
-
-typedef unsigned char UCHAR;
-
-
-#ifndef bool
-#define bool BOOL
-#endif
-//test if the processor actually supports the above functions
-//executing one the functions below without processor support will cause UD fault
-//bool check_for_aes_instructions(void);
-#if (__cplusplus)
-extern "C" {
-#endif
-int check_for_aes_instructions(void);
-
-#define ROUND_KEYS_UNALIGNED_TESTING
-
-#ifdef __linux__
-
-#ifdef ROUND_KEYS_UNALIGNED_TESTING
-
-#define DEFINE_ROUND_KEYS \
-	UCHAR __attribute__ ((aligned (16))) _expandedKey[16*16];	\
-	UCHAR *expandedKey = _expandedKey + 4;	\
-
-
-#else
-
-
-
-#define DEFINE_ROUND_KEYS \
-	UCHAR __attribute__ ((aligned (16))) _expandedKey[16*16];	\
-	UCHAR *expandedKey = _expandedKey;	\
-
-#endif
-
-#else // if not __linux__
-
-#ifdef ROUND_KEYS_UNALIGNED_TESTING
-
-#define DEFINE_ROUND_KEYS \
-	__declspec(align(16)) UCHAR _expandedKey[16*16];	\
-	UCHAR *expandedKey = _expandedKey + 4;	\
-
-
-#else
-
-
-
-#define DEFINE_ROUND_KEYS \
-	__declspec(align(16)) UCHAR _expandedKey[16*16];	\
-	UCHAR *expandedKey = _expandedKey;	\
-
-
-#endif
-
-#endif
-
-
-
-// encryption functions
-// plainText is pointer to input stream
-// cipherText is pointer to buffer to be filled with encrypted (cipher text) data
-// key is pointer to enc key (sizes are 16 bytes for AES-128, 24 bytes for AES-192, 32 for AES-256)
-// numBlocks is number of 16 bytes blocks to process - note that encryption is done of full 16 byte blocks
-void intel_AES_enc128(_AES_IN UCHAR *plainText, _AES_OUT UCHAR *cipherText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks);
-void intel_AES_enc192(_AES_IN UCHAR *plainText, _AES_OUT UCHAR *cipherText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks);
-void intel_AES_enc256(_AES_IN UCHAR *plainText, _AES_OUT UCHAR *cipherText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks);
-
-
-void intel_AES_enc128_CBC(_AES_IN UCHAR *plainText, _AES_OUT UCHAR *cipherText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks, _AES_IN UCHAR *iv);
-void intel_AES_enc192_CBC(_AES_IN UCHAR *plainText, _AES_OUT UCHAR *cipherText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks, _AES_IN UCHAR *iv);
-void intel_AES_enc256_CBC(_AES_IN UCHAR *plainText, _AES_OUT UCHAR *cipherText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks, _AES_IN UCHAR *iv);
-
-
-// encryption functions
-// cipherText is pointer to encrypted stream
-// plainText is pointer to buffer to be filled with original (plain text) data
-// key is pointer to enc key (sizes are 16 bytes for AES-128, 24 bytes for AES-192, 32 for AES-256)
-// numBlocks is number of 16 bytes blocks to process - note that decryption is done of full 16 byte blocks
-void intel_AES_dec128(_AES_IN UCHAR *cipherText, _AES_OUT UCHAR *plainText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks);
-void intel_AES_dec192(_AES_IN UCHAR *cipherText, _AES_OUT UCHAR *plainText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks);
-void intel_AES_dec256(_AES_IN UCHAR *cipherText, _AES_OUT UCHAR *plainText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks);
-
-void intel_AES_dec128_CBC(_AES_IN UCHAR *cipherText, _AES_OUT UCHAR *plainText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks, _AES_IN UCHAR *iv);
-void intel_AES_dec192_CBC(_AES_IN UCHAR *cipherText, _AES_OUT UCHAR *plainText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks, _AES_IN UCHAR *iv);
-void intel_AES_dec256_CBC(_AES_IN UCHAR *cipherText, _AES_OUT UCHAR *plainText, _AES_IN UCHAR *key, _AES_IN size_t numBlocks, _AES_IN UCHAR *iv);
-
-void intel_AES_encdec128_CTR(_AES_IN UCHAR *input, _AES_OUT UCHAR *output, _AES_IN UCHAR *key, _AES_IN size_t numBlocks, _AES_IN UCHAR *initial_counter);
-void intel_AES_encdec192_CTR(_AES_IN UCHAR *input, _AES_OUT UCHAR *output, _AES_IN UCHAR *key, _AES_IN size_t numBlocks, _AES_IN UCHAR *initial_counter);
-void intel_AES_encdec256_CTR(_AES_IN UCHAR *input, _AES_OUT UCHAR *output, _AES_IN UCHAR *key, _AES_IN size_t numBlocks, _AES_IN UCHAR *initial_counter);
-
-
-#if (__cplusplus)
-}
-#endif
-
-
-#endif
-
-
-
diff --git a/lib/accelerated/intel/license.txt b/lib/accelerated/intel/license.txt
index a93c3231d1..c87ba42b08 100755
--- a/lib/accelerated/intel/license.txt
+++ b/lib/accelerated/intel/license.txt
@@ -1,34 +1,43 @@
-/* intel_aes_lib source files come from Intel.
- * Modified by Patrick Fay
- *
-Copyright (c) 2010, Intel Corporation
-All rights reserved.
-
-Redistribution and use in source and binary forms, with or without 
-modification, are permitted provided that the following conditions are met:
-
-    * Redistributions of source code must retain the above copyright notice, 
-      this list of conditions and the following disclaimer.
-    * Redistributions in binary form must reproduce the above copyright notice, 
-      this list of conditions and the following disclaimer in the documentation 
-      and/or other materials provided with the distribution.
-    * Neither the name of Intel Corporation nor the names of its contributors 
-      may be used to endorse or promote products derived from this software 
-      without specific prior written permission.
-
-THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND 
-ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 
-WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 
-IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, 
-INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, 
-BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 
-DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 
-LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE 
-OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF 
-ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-
- ---------------------------------------------------------------------------
- Issue Date: Aug 6, 2010
- */
-
-Other source code files use the license shown in the source code file.
+====================================================================
+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/.
+====================================================================
+
+Copyright (c) 2006, CRYPTOGAMS by <appro@openssl.org>
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions
+are met:
+
+      *	Redistributions of source code must retain copyright notices,
+	this list of conditions and the following disclaimer.
+
+      *	Redistributions in binary form must reproduce the above
+	copyright notice, this list of conditions and the following
+	disclaimer in the documentation and/or other materials
+	provided with the distribution.
+
+      *	Neither the name of the CRYPTOGAMS nor the names of its
+	copyright holder and contributors may be used to endorse or
+	promote products derived from this software without specific
+	prior written permission.
+
+ALTERNATIVELY, provided that this notice is retained in full, this
+product may be distributed under the terms of the GNU General Public
+License (GPL), in which case the provisions of the GPL apply INSTEAD OF
+those given above.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.