From: Nikos Mavrogiannopoulos <nmav@gnutls.org>
Date: Tue, 25 Oct 2011 17:39:09 +0000 (+0200)
Subject: Added COFF versions of assembly files.
X-Git-Tag: gnutls_3_0_5~21
X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=e3e90b119bf04d8db44ee5aada6129654eab8440;p=thirdparty%2Fgnutls.git

Added COFF versions of assembly files.
---

diff --git a/configure.ac b/configure.ac
index d558994285..c881664b75 100644
--- a/configure.ac
+++ b/configure.ac
@@ -68,6 +68,16 @@ fi
 AM_CONDITIONAL(ENABLE_CXX, test "$use_cxx" != "no")
 AM_CONDITIONAL(WANT_TEST_SUITE, [test -f tests/suite/mini-eagain2.c])
 
+dnl Detect windows build
+case "$host" in
+  *-*-mingw32*)
+    have_win32=yes
+  ;;
+  *)
+  ;;
+esac
+
+AM_CONDITIONAL(WIN32, test "$have_win32" = yes)
 
 dnl Hardware Acceleration
 AC_ARG_ENABLE(hardware-acceleration,
@@ -75,6 +85,7 @@ AC_ARG_ENABLE(hardware-acceleration,
     use_accel=$enableval, use_accel=yes)
 hw_accel=none
 
+
 if test "$use_accel" != "no"; then
 case $host_cpu in
   i?86 | x86_64 | amd64)
diff --git a/lib/accelerated/x86/Makefile.am b/lib/accelerated/x86/Makefile.am
index 3530998ad0..689f977362 100644
--- a/lib/accelerated/x86/Makefile.am
+++ b/lib/accelerated/x86/Makefile.am
@@ -39,9 +39,17 @@ noinst_LTLIBRARIES = libx86.la
 libx86_la_SOURCES = sha-padlock.c hmac-padlock.c aes-x86.c aes-padlock.c aes-gcm-padlock.c aes-padlock.h aes-x86.h x86.h sha-padlock.h
 
 if ASM_X86_64
+
 AM_CFLAGS += -DASM_X86_64 -DASM_X86
 libx86_la_SOURCES += asm/appro-aes-x86-64.s asm/appro-aes-gcm-x86-64.s aes-gcm-x86.c asm/padlock-x86-64.s asm/cpuid-x86-64.s
+
 else
+
 AM_CFLAGS += -DASM_X86_32 -DASM_X86
+if WIN32
+libx86_la_SOURCES += coff/appro-aes-x86-coff.s coff/padlock-x86-coff.s asm/cpuid-x86.s
+else
 libx86_la_SOURCES += asm/appro-aes-x86.s asm/padlock-x86.s asm/cpuid-x86.s
 endif
+
+endif
diff --git a/lib/accelerated/x86/coff/appro-aes-x86-coff.s b/lib/accelerated/x86/coff/appro-aes-x86-coff.s
new file mode 100644
index 0000000000..2f9a6bc07c
--- /dev/null
+++ b/lib/accelerated/x86/coff/appro-aes-x86-coff.s
@@ -0,0 +1,2162 @@
+# Copyright (c) 2011, 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
+.def	_aesni_encrypt;	.scl	2;	.type	32;	.endef
+.align	16
+_aesni_encrypt:
+.L_aesni_encrypt_begin:
+	movl	4(%esp),%eax
+	movl	12(%esp),%edx
+	movups	(%eax),%xmm2
+	movl	240(%edx),%ecx
+	movl	8(%esp),%eax
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L000enc1_loop_1:
+.byte	102,15,56,220,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L000enc1_loop_1
+.byte	102,15,56,221,209
+	movups	%xmm2,(%eax)
+	ret
+.globl	_aesni_decrypt
+.def	_aesni_decrypt;	.scl	2;	.type	32;	.endef
+.align	16
+_aesni_decrypt:
+.L_aesni_decrypt_begin:
+	movl	4(%esp),%eax
+	movl	12(%esp),%edx
+	movups	(%eax),%xmm2
+	movl	240(%edx),%ecx
+	movl	8(%esp),%eax
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L001dec1_loop_2:
+.byte	102,15,56,222,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L001dec1_loop_2
+.byte	102,15,56,223,209
+	movups	%xmm2,(%eax)
+	ret
+.def	__aesni_encrypt3;	.scl	3;	.type	32;	.endef
+.align	16
+__aesni_encrypt3:
+	movups	(%edx),%xmm0
+	shrl	$1,%ecx
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+	pxor	%xmm0,%xmm3
+	pxor	%xmm0,%xmm4
+	movups	(%edx),%xmm0
+.L002enc3_loop:
+.byte	102,15,56,220,209
+.byte	102,15,56,220,217
+	decl	%ecx
+.byte	102,15,56,220,225
+	movups	16(%edx),%xmm1
+.byte	102,15,56,220,208
+.byte	102,15,56,220,216
+	leal	32(%edx),%edx
+.byte	102,15,56,220,224
+	movups	(%edx),%xmm0
+	jnz	.L002enc3_loop
+.byte	102,15,56,220,209
+.byte	102,15,56,220,217
+.byte	102,15,56,220,225
+.byte	102,15,56,221,208
+.byte	102,15,56,221,216
+.byte	102,15,56,221,224
+	ret
+.def	__aesni_decrypt3;	.scl	3;	.type	32;	.endef
+.align	16
+__aesni_decrypt3:
+	movups	(%edx),%xmm0
+	shrl	$1,%ecx
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+	pxor	%xmm0,%xmm3
+	pxor	%xmm0,%xmm4
+	movups	(%edx),%xmm0
+.L003dec3_loop:
+.byte	102,15,56,222,209
+.byte	102,15,56,222,217
+	decl	%ecx
+.byte	102,15,56,222,225
+	movups	16(%edx),%xmm1
+.byte	102,15,56,222,208
+.byte	102,15,56,222,216
+	leal	32(%edx),%edx
+.byte	102,15,56,222,224
+	movups	(%edx),%xmm0
+	jnz	.L003dec3_loop
+.byte	102,15,56,222,209
+.byte	102,15,56,222,217
+.byte	102,15,56,222,225
+.byte	102,15,56,223,208
+.byte	102,15,56,223,216
+.byte	102,15,56,223,224
+	ret
+.def	__aesni_encrypt4;	.scl	3;	.type	32;	.endef
+.align	16
+__aesni_encrypt4:
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	shrl	$1,%ecx
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+	pxor	%xmm0,%xmm3
+	pxor	%xmm0,%xmm4
+	pxor	%xmm0,%xmm5
+	movups	(%edx),%xmm0
+.L004enc4_loop:
+.byte	102,15,56,220,209
+.byte	102,15,56,220,217
+	decl	%ecx
+.byte	102,15,56,220,225
+.byte	102,15,56,220,233
+	movups	16(%edx),%xmm1
+.byte	102,15,56,220,208
+.byte	102,15,56,220,216
+	leal	32(%edx),%edx
+.byte	102,15,56,220,224
+.byte	102,15,56,220,232
+	movups	(%edx),%xmm0
+	jnz	.L004enc4_loop
+.byte	102,15,56,220,209
+.byte	102,15,56,220,217
+.byte	102,15,56,220,225
+.byte	102,15,56,220,233
+.byte	102,15,56,221,208
+.byte	102,15,56,221,216
+.byte	102,15,56,221,224
+.byte	102,15,56,221,232
+	ret
+.def	__aesni_decrypt4;	.scl	3;	.type	32;	.endef
+.align	16
+__aesni_decrypt4:
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	shrl	$1,%ecx
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+	pxor	%xmm0,%xmm3
+	pxor	%xmm0,%xmm4
+	pxor	%xmm0,%xmm5
+	movups	(%edx),%xmm0
+.L005dec4_loop:
+.byte	102,15,56,222,209
+.byte	102,15,56,222,217
+	decl	%ecx
+.byte	102,15,56,222,225
+.byte	102,15,56,222,233
+	movups	16(%edx),%xmm1
+.byte	102,15,56,222,208
+.byte	102,15,56,222,216
+	leal	32(%edx),%edx
+.byte	102,15,56,222,224
+.byte	102,15,56,222,232
+	movups	(%edx),%xmm0
+	jnz	.L005dec4_loop
+.byte	102,15,56,222,209
+.byte	102,15,56,222,217
+.byte	102,15,56,222,225
+.byte	102,15,56,222,233
+.byte	102,15,56,223,208
+.byte	102,15,56,223,216
+.byte	102,15,56,223,224
+.byte	102,15,56,223,232
+	ret
+.def	__aesni_encrypt6;	.scl	3;	.type	32;	.endef
+.align	16
+__aesni_encrypt6:
+	movups	(%edx),%xmm0
+	shrl	$1,%ecx
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+	pxor	%xmm0,%xmm3
+.byte	102,15,56,220,209
+	pxor	%xmm0,%xmm4
+.byte	102,15,56,220,217
+	pxor	%xmm0,%xmm5
+	decl	%ecx
+.byte	102,15,56,220,225
+	pxor	%xmm0,%xmm6
+.byte	102,15,56,220,233
+	pxor	%xmm0,%xmm7
+.byte	102,15,56,220,241
+	movups	(%edx),%xmm0
+.byte	102,15,56,220,249
+	jmp	.L_aesni_encrypt6_enter
+.align	16
+.L006enc6_loop:
+.byte	102,15,56,220,209
+.byte	102,15,56,220,217
+	decl	%ecx
+.byte	102,15,56,220,225
+.byte	102,15,56,220,233
+.byte	102,15,56,220,241
+.byte	102,15,56,220,249
+.align	16
+.L_aesni_encrypt6_enter:
+	movups	16(%edx),%xmm1
+.byte	102,15,56,220,208
+.byte	102,15,56,220,216
+	leal	32(%edx),%edx
+.byte	102,15,56,220,224
+.byte	102,15,56,220,232
+.byte	102,15,56,220,240
+.byte	102,15,56,220,248
+	movups	(%edx),%xmm0
+	jnz	.L006enc6_loop
+.byte	102,15,56,220,209
+.byte	102,15,56,220,217
+.byte	102,15,56,220,225
+.byte	102,15,56,220,233
+.byte	102,15,56,220,241
+.byte	102,15,56,220,249
+.byte	102,15,56,221,208
+.byte	102,15,56,221,216
+.byte	102,15,56,221,224
+.byte	102,15,56,221,232
+.byte	102,15,56,221,240
+.byte	102,15,56,221,248
+	ret
+.def	__aesni_decrypt6;	.scl	3;	.type	32;	.endef
+.align	16
+__aesni_decrypt6:
+	movups	(%edx),%xmm0
+	shrl	$1,%ecx
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+	pxor	%xmm0,%xmm3
+.byte	102,15,56,222,209
+	pxor	%xmm0,%xmm4
+.byte	102,15,56,222,217
+	pxor	%xmm0,%xmm5
+	decl	%ecx
+.byte	102,15,56,222,225
+	pxor	%xmm0,%xmm6
+.byte	102,15,56,222,233
+	pxor	%xmm0,%xmm7
+.byte	102,15,56,222,241
+	movups	(%edx),%xmm0
+.byte	102,15,56,222,249
+	jmp	.L_aesni_decrypt6_enter
+.align	16
+.L007dec6_loop:
+.byte	102,15,56,222,209
+.byte	102,15,56,222,217
+	decl	%ecx
+.byte	102,15,56,222,225
+.byte	102,15,56,222,233
+.byte	102,15,56,222,241
+.byte	102,15,56,222,249
+.align	16
+.L_aesni_decrypt6_enter:
+	movups	16(%edx),%xmm1
+.byte	102,15,56,222,208
+.byte	102,15,56,222,216
+	leal	32(%edx),%edx
+.byte	102,15,56,222,224
+.byte	102,15,56,222,232
+.byte	102,15,56,222,240
+.byte	102,15,56,222,248
+	movups	(%edx),%xmm0
+	jnz	.L007dec6_loop
+.byte	102,15,56,222,209
+.byte	102,15,56,222,217
+.byte	102,15,56,222,225
+.byte	102,15,56,222,233
+.byte	102,15,56,222,241
+.byte	102,15,56,222,249
+.byte	102,15,56,223,208
+.byte	102,15,56,223,216
+.byte	102,15,56,223,224
+.byte	102,15,56,223,232
+.byte	102,15,56,223,240
+.byte	102,15,56,223,248
+	ret
+.globl	_aesni_ecb_encrypt
+.def	_aesni_ecb_encrypt;	.scl	2;	.type	32;	.endef
+.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),%ebx
+	andl	$-16,%eax
+	jz	.L008ecb_ret
+	movl	240(%edx),%ecx
+	testl	%ebx,%ebx
+	jz	.L009ecb_decrypt
+	movl	%edx,%ebp
+	movl	%ecx,%ebx
+	cmpl	$96,%eax
+	jb	.L010ecb_enc_tail
+	movdqu	(%esi),%xmm2
+	movdqu	16(%esi),%xmm3
+	movdqu	32(%esi),%xmm4
+	movdqu	48(%esi),%xmm5
+	movdqu	64(%esi),%xmm6
+	movdqu	80(%esi),%xmm7
+	leal	96(%esi),%esi
+	subl	$96,%eax
+	jmp	.L011ecb_enc_loop6_enter
+.align	16
+.L012ecb_enc_loop6:
+	movups	%xmm2,(%edi)
+	movdqu	(%esi),%xmm2
+	movups	%xmm3,16(%edi)
+	movdqu	16(%esi),%xmm3
+	movups	%xmm4,32(%edi)
+	movdqu	32(%esi),%xmm4
+	movups	%xmm5,48(%edi)
+	movdqu	48(%esi),%xmm5
+	movups	%xmm6,64(%edi)
+	movdqu	64(%esi),%xmm6
+	movups	%xmm7,80(%edi)
+	leal	96(%edi),%edi
+	movdqu	80(%esi),%xmm7
+	leal	96(%esi),%esi
+.L011ecb_enc_loop6_enter:
+	call	__aesni_encrypt6
+	movl	%ebp,%edx
+	movl	%ebx,%ecx
+	subl	$96,%eax
+	jnc	.L012ecb_enc_loop6
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+	movups	%xmm6,64(%edi)
+	movups	%xmm7,80(%edi)
+	leal	96(%edi),%edi
+	addl	$96,%eax
+	jz	.L008ecb_ret
+.L010ecb_enc_tail:
+	movups	(%esi),%xmm2
+	cmpl	$32,%eax
+	jb	.L013ecb_enc_one
+	movups	16(%esi),%xmm3
+	je	.L014ecb_enc_two
+	movups	32(%esi),%xmm4
+	cmpl	$64,%eax
+	jb	.L015ecb_enc_three
+	movups	48(%esi),%xmm5
+	je	.L016ecb_enc_four
+	movups	64(%esi),%xmm6
+	xorps	%xmm7,%xmm7
+	call	__aesni_encrypt6
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+	movups	%xmm6,64(%edi)
+	jmp	.L008ecb_ret
+.align	16
+.L013ecb_enc_one:
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L017enc1_loop_3:
+.byte	102,15,56,220,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L017enc1_loop_3
+.byte	102,15,56,221,209
+	movups	%xmm2,(%edi)
+	jmp	.L008ecb_ret
+.align	16
+.L014ecb_enc_two:
+	xorps	%xmm4,%xmm4
+	call	__aesni_encrypt3
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	jmp	.L008ecb_ret
+.align	16
+.L015ecb_enc_three:
+	call	__aesni_encrypt3
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	jmp	.L008ecb_ret
+.align	16
+.L016ecb_enc_four:
+	call	__aesni_encrypt4
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+	jmp	.L008ecb_ret
+.align	16
+.L009ecb_decrypt:
+	movl	%edx,%ebp
+	movl	%ecx,%ebx
+	cmpl	$96,%eax
+	jb	.L018ecb_dec_tail
+	movdqu	(%esi),%xmm2
+	movdqu	16(%esi),%xmm3
+	movdqu	32(%esi),%xmm4
+	movdqu	48(%esi),%xmm5
+	movdqu	64(%esi),%xmm6
+	movdqu	80(%esi),%xmm7
+	leal	96(%esi),%esi
+	subl	$96,%eax
+	jmp	.L019ecb_dec_loop6_enter
+.align	16
+.L020ecb_dec_loop6:
+	movups	%xmm2,(%edi)
+	movdqu	(%esi),%xmm2
+	movups	%xmm3,16(%edi)
+	movdqu	16(%esi),%xmm3
+	movups	%xmm4,32(%edi)
+	movdqu	32(%esi),%xmm4
+	movups	%xmm5,48(%edi)
+	movdqu	48(%esi),%xmm5
+	movups	%xmm6,64(%edi)
+	movdqu	64(%esi),%xmm6
+	movups	%xmm7,80(%edi)
+	leal	96(%edi),%edi
+	movdqu	80(%esi),%xmm7
+	leal	96(%esi),%esi
+.L019ecb_dec_loop6_enter:
+	call	__aesni_decrypt6
+	movl	%ebp,%edx
+	movl	%ebx,%ecx
+	subl	$96,%eax
+	jnc	.L020ecb_dec_loop6
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+	movups	%xmm6,64(%edi)
+	movups	%xmm7,80(%edi)
+	leal	96(%edi),%edi
+	addl	$96,%eax
+	jz	.L008ecb_ret
+.L018ecb_dec_tail:
+	movups	(%esi),%xmm2
+	cmpl	$32,%eax
+	jb	.L021ecb_dec_one
+	movups	16(%esi),%xmm3
+	je	.L022ecb_dec_two
+	movups	32(%esi),%xmm4
+	cmpl	$64,%eax
+	jb	.L023ecb_dec_three
+	movups	48(%esi),%xmm5
+	je	.L024ecb_dec_four
+	movups	64(%esi),%xmm6
+	xorps	%xmm7,%xmm7
+	call	__aesni_decrypt6
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+	movups	%xmm6,64(%edi)
+	jmp	.L008ecb_ret
+.align	16
+.L021ecb_dec_one:
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L025dec1_loop_4:
+.byte	102,15,56,222,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L025dec1_loop_4
+.byte	102,15,56,223,209
+	movups	%xmm2,(%edi)
+	jmp	.L008ecb_ret
+.align	16
+.L022ecb_dec_two:
+	xorps	%xmm4,%xmm4
+	call	__aesni_decrypt3
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	jmp	.L008ecb_ret
+.align	16
+.L023ecb_dec_three:
+	call	__aesni_decrypt3
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	jmp	.L008ecb_ret
+.align	16
+.L024ecb_dec_four:
+	call	__aesni_decrypt4
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+.L008ecb_ret:
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.globl	_aesni_ccm64_encrypt_blocks
+.def	_aesni_ccm64_encrypt_blocks;	.scl	2;	.type	32;	.endef
+.align	16
+_aesni_ccm64_encrypt_blocks:
+.L_aesni_ccm64_encrypt_blocks_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),%ebx
+	movl	40(%esp),%ecx
+	movl	%esp,%ebp
+	subl	$60,%esp
+	andl	$-16,%esp
+	movl	%ebp,48(%esp)
+	movdqu	(%ebx),%xmm7
+	movdqu	(%ecx),%xmm3
+	movl	240(%edx),%ecx
+	movl	$202182159,(%esp)
+	movl	$134810123,4(%esp)
+	movl	$67438087,8(%esp)
+	movl	$66051,12(%esp)
+	movl	$1,%ebx
+	xorl	%ebp,%ebp
+	movl	%ebx,16(%esp)
+	movl	%ebp,20(%esp)
+	movl	%ebp,24(%esp)
+	movl	%ebp,28(%esp)
+	shrl	$1,%ecx
+	leal	(%edx),%ebp
+	movdqa	(%esp),%xmm5
+	movdqa	%xmm7,%xmm2
+	movl	%ecx,%ebx
+.byte	102,15,56,0,253
+.L026ccm64_enc_outer:
+	movups	(%ebp),%xmm0
+	movl	%ebx,%ecx
+	movups	(%esi),%xmm6
+	xorps	%xmm0,%xmm2
+	movups	16(%ebp),%xmm1
+	xorps	%xmm6,%xmm0
+	leal	32(%ebp),%edx
+	xorps	%xmm0,%xmm3
+	movups	(%edx),%xmm0
+.L027ccm64_enc2_loop:
+.byte	102,15,56,220,209
+	decl	%ecx
+.byte	102,15,56,220,217
+	movups	16(%edx),%xmm1
+.byte	102,15,56,220,208
+	leal	32(%edx),%edx
+.byte	102,15,56,220,216
+	movups	(%edx),%xmm0
+	jnz	.L027ccm64_enc2_loop
+.byte	102,15,56,220,209
+.byte	102,15,56,220,217
+	paddq	16(%esp),%xmm7
+.byte	102,15,56,221,208
+.byte	102,15,56,221,216
+	decl	%eax
+	leal	16(%esi),%esi
+	xorps	%xmm2,%xmm6
+	movdqa	%xmm7,%xmm2
+	movups	%xmm6,(%edi)
+	leal	16(%edi),%edi
+.byte	102,15,56,0,213
+	jnz	.L026ccm64_enc_outer
+	movl	48(%esp),%esp
+	movl	40(%esp),%edi
+	movups	%xmm3,(%edi)
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.globl	_aesni_ccm64_decrypt_blocks
+.def	_aesni_ccm64_decrypt_blocks;	.scl	2;	.type	32;	.endef
+.align	16
+_aesni_ccm64_decrypt_blocks:
+.L_aesni_ccm64_decrypt_blocks_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),%ebx
+	movl	40(%esp),%ecx
+	movl	%esp,%ebp
+	subl	$60,%esp
+	andl	$-16,%esp
+	movl	%ebp,48(%esp)
+	movdqu	(%ebx),%xmm7
+	movdqu	(%ecx),%xmm3
+	movl	240(%edx),%ecx
+	movl	$202182159,(%esp)
+	movl	$134810123,4(%esp)
+	movl	$67438087,8(%esp)
+	movl	$66051,12(%esp)
+	movl	$1,%ebx
+	xorl	%ebp,%ebp
+	movl	%ebx,16(%esp)
+	movl	%ebp,20(%esp)
+	movl	%ebp,24(%esp)
+	movl	%ebp,28(%esp)
+	movdqa	(%esp),%xmm5
+	movdqa	%xmm7,%xmm2
+	movl	%edx,%ebp
+	movl	%ecx,%ebx
+.byte	102,15,56,0,253
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L028enc1_loop_5:
+.byte	102,15,56,220,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L028enc1_loop_5
+.byte	102,15,56,221,209
+	movups	(%esi),%xmm6
+	paddq	16(%esp),%xmm7
+	leal	16(%esi),%esi
+	jmp	.L029ccm64_dec_outer
+.align	16
+.L029ccm64_dec_outer:
+	xorps	%xmm2,%xmm6
+	movdqa	%xmm7,%xmm2
+	movl	%ebx,%ecx
+	movups	%xmm6,(%edi)
+	leal	16(%edi),%edi
+.byte	102,15,56,0,213
+	subl	$1,%eax
+	jz	.L030ccm64_dec_break
+	movups	(%ebp),%xmm0
+	shrl	$1,%ecx
+	movups	16(%ebp),%xmm1
+	xorps	%xmm0,%xmm6
+	leal	32(%ebp),%edx
+	xorps	%xmm0,%xmm2
+	xorps	%xmm6,%xmm3
+	movups	(%edx),%xmm0
+.L031ccm64_dec2_loop:
+.byte	102,15,56,220,209
+	decl	%ecx
+.byte	102,15,56,220,217
+	movups	16(%edx),%xmm1
+.byte	102,15,56,220,208
+	leal	32(%edx),%edx
+.byte	102,15,56,220,216
+	movups	(%edx),%xmm0
+	jnz	.L031ccm64_dec2_loop
+	movups	(%esi),%xmm6
+	paddq	16(%esp),%xmm7
+.byte	102,15,56,220,209
+.byte	102,15,56,220,217
+	leal	16(%esi),%esi
+.byte	102,15,56,221,208
+.byte	102,15,56,221,216
+	jmp	.L029ccm64_dec_outer
+.align	16
+.L030ccm64_dec_break:
+	movl	%ebp,%edx
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	xorps	%xmm0,%xmm6
+	leal	32(%edx),%edx
+	xorps	%xmm6,%xmm3
+.L032enc1_loop_6:
+.byte	102,15,56,220,217
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L032enc1_loop_6
+.byte	102,15,56,221,217
+	movl	48(%esp),%esp
+	movl	40(%esp),%edi
+	movups	%xmm3,(%edi)
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.globl	_aesni_ctr32_encrypt_blocks
+.def	_aesni_ctr32_encrypt_blocks;	.scl	2;	.type	32;	.endef
+.align	16
+_aesni_ctr32_encrypt_blocks:
+.L_aesni_ctr32_encrypt_blocks_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),%ebx
+	movl	%esp,%ebp
+	subl	$88,%esp
+	andl	$-16,%esp
+	movl	%ebp,80(%esp)
+	cmpl	$1,%eax
+	je	.L033ctr32_one_shortcut
+	movdqu	(%ebx),%xmm7
+	movl	$202182159,(%esp)
+	movl	$134810123,4(%esp)
+	movl	$67438087,8(%esp)
+	movl	$66051,12(%esp)
+	movl	$6,%ecx
+	xorl	%ebp,%ebp
+	movl	%ecx,16(%esp)
+	movl	%ecx,20(%esp)
+	movl	%ecx,24(%esp)
+	movl	%ebp,28(%esp)
+.byte	102,15,58,22,251,3
+.byte	102,15,58,34,253,3
+	movl	240(%edx),%ecx
+	bswap	%ebx
+	pxor	%xmm1,%xmm1
+	pxor	%xmm0,%xmm0
+	movdqa	(%esp),%xmm2
+.byte	102,15,58,34,203,0
+	leal	3(%ebx),%ebp
+.byte	102,15,58,34,197,0
+	incl	%ebx
+.byte	102,15,58,34,203,1
+	incl	%ebp
+.byte	102,15,58,34,197,1
+	incl	%ebx
+.byte	102,15,58,34,203,2
+	incl	%ebp
+.byte	102,15,58,34,197,2
+	movdqa	%xmm1,48(%esp)
+.byte	102,15,56,0,202
+	movdqa	%xmm0,64(%esp)
+.byte	102,15,56,0,194
+	pshufd	$192,%xmm1,%xmm2
+	pshufd	$128,%xmm1,%xmm3
+	cmpl	$6,%eax
+	jb	.L034ctr32_tail
+	movdqa	%xmm7,32(%esp)
+	shrl	$1,%ecx
+	movl	%edx,%ebp
+	movl	%ecx,%ebx
+	subl	$6,%eax
+	jmp	.L035ctr32_loop6
+.align	16
+.L035ctr32_loop6:
+	pshufd	$64,%xmm1,%xmm4
+	movdqa	32(%esp),%xmm1
+	pshufd	$192,%xmm0,%xmm5
+	por	%xmm1,%xmm2
+	pshufd	$128,%xmm0,%xmm6
+	por	%xmm1,%xmm3
+	pshufd	$64,%xmm0,%xmm7
+	por	%xmm1,%xmm4
+	por	%xmm1,%xmm5
+	por	%xmm1,%xmm6
+	por	%xmm1,%xmm7
+	movups	(%ebp),%xmm0
+	movups	16(%ebp),%xmm1
+	leal	32(%ebp),%edx
+	decl	%ecx
+	pxor	%xmm0,%xmm2
+	pxor	%xmm0,%xmm3
+.byte	102,15,56,220,209
+	pxor	%xmm0,%xmm4
+.byte	102,15,56,220,217
+	pxor	%xmm0,%xmm5
+.byte	102,15,56,220,225
+	pxor	%xmm0,%xmm6
+.byte	102,15,56,220,233
+	pxor	%xmm0,%xmm7
+.byte	102,15,56,220,241
+	movups	(%edx),%xmm0
+.byte	102,15,56,220,249
+	call	.L_aesni_encrypt6_enter
+	movups	(%esi),%xmm1
+	movups	16(%esi),%xmm0
+	xorps	%xmm1,%xmm2
+	movups	32(%esi),%xmm1
+	xorps	%xmm0,%xmm3
+	movups	%xmm2,(%edi)
+	movdqa	16(%esp),%xmm0
+	xorps	%xmm1,%xmm4
+	movdqa	48(%esp),%xmm1
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	paddd	%xmm0,%xmm1
+	paddd	64(%esp),%xmm0
+	movdqa	(%esp),%xmm2
+	movups	48(%esi),%xmm3
+	movups	64(%esi),%xmm4
+	xorps	%xmm3,%xmm5
+	movups	80(%esi),%xmm3
+	leal	96(%esi),%esi
+	movdqa	%xmm1,48(%esp)
+.byte	102,15,56,0,202
+	xorps	%xmm4,%xmm6
+	movups	%xmm5,48(%edi)
+	xorps	%xmm3,%xmm7
+	movdqa	%xmm0,64(%esp)
+.byte	102,15,56,0,194
+	movups	%xmm6,64(%edi)
+	pshufd	$192,%xmm1,%xmm2
+	movups	%xmm7,80(%edi)
+	leal	96(%edi),%edi
+	movl	%ebx,%ecx
+	pshufd	$128,%xmm1,%xmm3
+	subl	$6,%eax
+	jnc	.L035ctr32_loop6
+	addl	$6,%eax
+	jz	.L036ctr32_ret
+	movl	%ebp,%edx
+	leal	1(,%ecx,2),%ecx
+	movdqa	32(%esp),%xmm7
+.L034ctr32_tail:
+	por	%xmm7,%xmm2
+	cmpl	$2,%eax
+	jb	.L037ctr32_one
+	pshufd	$64,%xmm1,%xmm4
+	por	%xmm7,%xmm3
+	je	.L038ctr32_two
+	pshufd	$192,%xmm0,%xmm5
+	por	%xmm7,%xmm4
+	cmpl	$4,%eax
+	jb	.L039ctr32_three
+	pshufd	$128,%xmm0,%xmm6
+	por	%xmm7,%xmm5
+	je	.L040ctr32_four
+	por	%xmm7,%xmm6
+	call	__aesni_encrypt6
+	movups	(%esi),%xmm1
+	movups	16(%esi),%xmm0
+	xorps	%xmm1,%xmm2
+	movups	32(%esi),%xmm1
+	xorps	%xmm0,%xmm3
+	movups	48(%esi),%xmm0
+	xorps	%xmm1,%xmm4
+	movups	64(%esi),%xmm1
+	xorps	%xmm0,%xmm5
+	movups	%xmm2,(%edi)
+	xorps	%xmm1,%xmm6
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+	movups	%xmm6,64(%edi)
+	jmp	.L036ctr32_ret
+.align	16
+.L033ctr32_one_shortcut:
+	movups	(%ebx),%xmm2
+	movl	240(%edx),%ecx
+.L037ctr32_one:
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L041enc1_loop_7:
+.byte	102,15,56,220,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L041enc1_loop_7
+.byte	102,15,56,221,209
+	movups	(%esi),%xmm6
+	xorps	%xmm2,%xmm6
+	movups	%xmm6,(%edi)
+	jmp	.L036ctr32_ret
+.align	16
+.L038ctr32_two:
+	call	__aesni_encrypt3
+	movups	(%esi),%xmm5
+	movups	16(%esi),%xmm6
+	xorps	%xmm5,%xmm2
+	xorps	%xmm6,%xmm3
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	jmp	.L036ctr32_ret
+.align	16
+.L039ctr32_three:
+	call	__aesni_encrypt3
+	movups	(%esi),%xmm5
+	movups	16(%esi),%xmm6
+	xorps	%xmm5,%xmm2
+	movups	32(%esi),%xmm7
+	xorps	%xmm6,%xmm3
+	movups	%xmm2,(%edi)
+	xorps	%xmm7,%xmm4
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	jmp	.L036ctr32_ret
+.align	16
+.L040ctr32_four:
+	call	__aesni_encrypt4
+	movups	(%esi),%xmm6
+	movups	16(%esi),%xmm7
+	movups	32(%esi),%xmm1
+	xorps	%xmm6,%xmm2
+	movups	48(%esi),%xmm0
+	xorps	%xmm7,%xmm3
+	movups	%xmm2,(%edi)
+	xorps	%xmm1,%xmm4
+	movups	%xmm3,16(%edi)
+	xorps	%xmm0,%xmm5
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+.L036ctr32_ret:
+	movl	80(%esp),%esp
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.globl	_aesni_xts_encrypt
+.def	_aesni_xts_encrypt;	.scl	2;	.type	32;	.endef
+.align	16
+_aesni_xts_encrypt:
+.L_aesni_xts_encrypt_begin:
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	movl	36(%esp),%edx
+	movl	40(%esp),%esi
+	movl	240(%edx),%ecx
+	movups	(%esi),%xmm2
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L042enc1_loop_8:
+.byte	102,15,56,220,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L042enc1_loop_8
+.byte	102,15,56,221,209
+	movl	20(%esp),%esi
+	movl	24(%esp),%edi
+	movl	28(%esp),%eax
+	movl	32(%esp),%edx
+	movl	%esp,%ebp
+	subl	$120,%esp
+	movl	240(%edx),%ecx
+	andl	$-16,%esp
+	movl	$135,96(%esp)
+	movl	$0,100(%esp)
+	movl	$1,104(%esp)
+	movl	$0,108(%esp)
+	movl	%eax,112(%esp)
+	movl	%ebp,116(%esp)
+	movdqa	%xmm2,%xmm1
+	pxor	%xmm0,%xmm0
+	movdqa	96(%esp),%xmm3
+	pcmpgtd	%xmm1,%xmm0
+	andl	$-16,%eax
+	movl	%edx,%ebp
+	movl	%ecx,%ebx
+	subl	$96,%eax
+	jc	.L043xts_enc_short
+	shrl	$1,%ecx
+	movl	%ecx,%ebx
+	jmp	.L044xts_enc_loop6
+.align	16
+.L044xts_enc_loop6:
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,(%esp)
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,16(%esp)
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,32(%esp)
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,48(%esp)
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	pshufd	$19,%xmm0,%xmm7
+	movdqa	%xmm1,64(%esp)
+	paddq	%xmm1,%xmm1
+	movups	(%ebp),%xmm0
+	pand	%xmm3,%xmm7
+	movups	(%esi),%xmm2
+	pxor	%xmm1,%xmm7
+	movdqu	16(%esi),%xmm3
+	xorps	%xmm0,%xmm2
+	movdqu	32(%esi),%xmm4
+	pxor	%xmm0,%xmm3
+	movdqu	48(%esi),%xmm5
+	pxor	%xmm0,%xmm4
+	movdqu	64(%esi),%xmm6
+	pxor	%xmm0,%xmm5
+	movdqu	80(%esi),%xmm1
+	pxor	%xmm0,%xmm6
+	leal	96(%esi),%esi
+	pxor	(%esp),%xmm2
+	movdqa	%xmm7,80(%esp)
+	pxor	%xmm1,%xmm7
+	movups	16(%ebp),%xmm1
+	leal	32(%ebp),%edx
+	pxor	16(%esp),%xmm3
+.byte	102,15,56,220,209
+	pxor	32(%esp),%xmm4
+.byte	102,15,56,220,217
+	pxor	48(%esp),%xmm5
+	decl	%ecx
+.byte	102,15,56,220,225
+	pxor	64(%esp),%xmm6
+.byte	102,15,56,220,233
+	pxor	%xmm0,%xmm7
+.byte	102,15,56,220,241
+	movups	(%edx),%xmm0
+.byte	102,15,56,220,249
+	call	.L_aesni_encrypt6_enter
+	movdqa	80(%esp),%xmm1
+	pxor	%xmm0,%xmm0
+	xorps	(%esp),%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	xorps	16(%esp),%xmm3
+	movups	%xmm2,(%edi)
+	xorps	32(%esp),%xmm4
+	movups	%xmm3,16(%edi)
+	xorps	48(%esp),%xmm5
+	movups	%xmm4,32(%edi)
+	xorps	64(%esp),%xmm6
+	movups	%xmm5,48(%edi)
+	xorps	%xmm1,%xmm7
+	movups	%xmm6,64(%edi)
+	pshufd	$19,%xmm0,%xmm2
+	movups	%xmm7,80(%edi)
+	leal	96(%edi),%edi
+	movdqa	96(%esp),%xmm3
+	pxor	%xmm0,%xmm0
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	movl	%ebx,%ecx
+	pxor	%xmm2,%xmm1
+	subl	$96,%eax
+	jnc	.L044xts_enc_loop6
+	leal	1(,%ecx,2),%ecx
+	movl	%ebp,%edx
+	movl	%ecx,%ebx
+.L043xts_enc_short:
+	addl	$96,%eax
+	jz	.L045xts_enc_done6x
+	movdqa	%xmm1,%xmm5
+	cmpl	$32,%eax
+	jb	.L046xts_enc_one
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	je	.L047xts_enc_two
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,%xmm6
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	cmpl	$64,%eax
+	jb	.L048xts_enc_three
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,%xmm7
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	movdqa	%xmm5,(%esp)
+	movdqa	%xmm6,16(%esp)
+	je	.L049xts_enc_four
+	movdqa	%xmm7,32(%esp)
+	pshufd	$19,%xmm0,%xmm7
+	movdqa	%xmm1,48(%esp)
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm7
+	pxor	%xmm1,%xmm7
+	movdqu	(%esi),%xmm2
+	movdqu	16(%esi),%xmm3
+	movdqu	32(%esi),%xmm4
+	pxor	(%esp),%xmm2
+	movdqu	48(%esi),%xmm5
+	pxor	16(%esp),%xmm3
+	movdqu	64(%esi),%xmm6
+	pxor	32(%esp),%xmm4
+	leal	80(%esi),%esi
+	pxor	48(%esp),%xmm5
+	movdqa	%xmm7,64(%esp)
+	pxor	%xmm7,%xmm6
+	call	__aesni_encrypt6
+	movaps	64(%esp),%xmm1
+	xorps	(%esp),%xmm2
+	xorps	16(%esp),%xmm3
+	xorps	32(%esp),%xmm4
+	movups	%xmm2,(%edi)
+	xorps	48(%esp),%xmm5
+	movups	%xmm3,16(%edi)
+	xorps	%xmm1,%xmm6
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+	movups	%xmm6,64(%edi)
+	leal	80(%edi),%edi
+	jmp	.L050xts_enc_done
+.align	16
+.L046xts_enc_one:
+	movups	(%esi),%xmm2
+	leal	16(%esi),%esi
+	xorps	%xmm5,%xmm2
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L051enc1_loop_9:
+.byte	102,15,56,220,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L051enc1_loop_9
+.byte	102,15,56,221,209
+	xorps	%xmm5,%xmm2
+	movups	%xmm2,(%edi)
+	leal	16(%edi),%edi
+	movdqa	%xmm5,%xmm1
+	jmp	.L050xts_enc_done
+.align	16
+.L047xts_enc_two:
+	movaps	%xmm1,%xmm6
+	movups	(%esi),%xmm2
+	movups	16(%esi),%xmm3
+	leal	32(%esi),%esi
+	xorps	%xmm5,%xmm2
+	xorps	%xmm6,%xmm3
+	xorps	%xmm4,%xmm4
+	call	__aesni_encrypt3
+	xorps	%xmm5,%xmm2
+	xorps	%xmm6,%xmm3
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	leal	32(%edi),%edi
+	movdqa	%xmm6,%xmm1
+	jmp	.L050xts_enc_done
+.align	16
+.L048xts_enc_three:
+	movaps	%xmm1,%xmm7
+	movups	(%esi),%xmm2
+	movups	16(%esi),%xmm3
+	movups	32(%esi),%xmm4
+	leal	48(%esi),%esi
+	xorps	%xmm5,%xmm2
+	xorps	%xmm6,%xmm3
+	xorps	%xmm7,%xmm4
+	call	__aesni_encrypt3
+	xorps	%xmm5,%xmm2
+	xorps	%xmm6,%xmm3
+	xorps	%xmm7,%xmm4
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	leal	48(%edi),%edi
+	movdqa	%xmm7,%xmm1
+	jmp	.L050xts_enc_done
+.align	16
+.L049xts_enc_four:
+	movaps	%xmm1,%xmm6
+	movups	(%esi),%xmm2
+	movups	16(%esi),%xmm3
+	movups	32(%esi),%xmm4
+	xorps	(%esp),%xmm2
+	movups	48(%esi),%xmm5
+	leal	64(%esi),%esi
+	xorps	16(%esp),%xmm3
+	xorps	%xmm7,%xmm4
+	xorps	%xmm6,%xmm5
+	call	__aesni_encrypt4
+	xorps	(%esp),%xmm2
+	xorps	16(%esp),%xmm3
+	xorps	%xmm7,%xmm4
+	movups	%xmm2,(%edi)
+	xorps	%xmm6,%xmm5
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+	leal	64(%edi),%edi
+	movdqa	%xmm6,%xmm1
+	jmp	.L050xts_enc_done
+.align	16
+.L045xts_enc_done6x:
+	movl	112(%esp),%eax
+	andl	$15,%eax
+	jz	.L052xts_enc_ret
+	movdqa	%xmm1,%xmm5
+	movl	%eax,112(%esp)
+	jmp	.L053xts_enc_steal
+.align	16
+.L050xts_enc_done:
+	movl	112(%esp),%eax
+	pxor	%xmm0,%xmm0
+	andl	$15,%eax
+	jz	.L052xts_enc_ret
+	pcmpgtd	%xmm1,%xmm0
+	movl	%eax,112(%esp)
+	pshufd	$19,%xmm0,%xmm5
+	paddq	%xmm1,%xmm1
+	pand	96(%esp),%xmm5
+	pxor	%xmm1,%xmm5
+.L053xts_enc_steal:
+	movzbl	(%esi),%ecx
+	movzbl	-16(%edi),%edx
+	leal	1(%esi),%esi
+	movb	%cl,-16(%edi)
+	movb	%dl,(%edi)
+	leal	1(%edi),%edi
+	subl	$1,%eax
+	jnz	.L053xts_enc_steal
+	subl	112(%esp),%edi
+	movl	%ebp,%edx
+	movl	%ebx,%ecx
+	movups	-16(%edi),%xmm2
+	xorps	%xmm5,%xmm2
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L054enc1_loop_10:
+.byte	102,15,56,220,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L054enc1_loop_10
+.byte	102,15,56,221,209
+	xorps	%xmm5,%xmm2
+	movups	%xmm2,-16(%edi)
+.L052xts_enc_ret:
+	movl	116(%esp),%esp
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.globl	_aesni_xts_decrypt
+.def	_aesni_xts_decrypt;	.scl	2;	.type	32;	.endef
+.align	16
+_aesni_xts_decrypt:
+.L_aesni_xts_decrypt_begin:
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	movl	36(%esp),%edx
+	movl	40(%esp),%esi
+	movl	240(%edx),%ecx
+	movups	(%esi),%xmm2
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L055enc1_loop_11:
+.byte	102,15,56,220,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L055enc1_loop_11
+.byte	102,15,56,221,209
+	movl	20(%esp),%esi
+	movl	24(%esp),%edi
+	movl	28(%esp),%eax
+	movl	32(%esp),%edx
+	movl	%esp,%ebp
+	subl	$120,%esp
+	andl	$-16,%esp
+	xorl	%ebx,%ebx
+	testl	$15,%eax
+	setnz	%bl
+	shll	$4,%ebx
+	subl	%ebx,%eax
+	movl	$135,96(%esp)
+	movl	$0,100(%esp)
+	movl	$1,104(%esp)
+	movl	$0,108(%esp)
+	movl	%eax,112(%esp)
+	movl	%ebp,116(%esp)
+	movl	240(%edx),%ecx
+	movl	%edx,%ebp
+	movl	%ecx,%ebx
+	movdqa	%xmm2,%xmm1
+	pxor	%xmm0,%xmm0
+	movdqa	96(%esp),%xmm3
+	pcmpgtd	%xmm1,%xmm0
+	andl	$-16,%eax
+	subl	$96,%eax
+	jc	.L056xts_dec_short
+	shrl	$1,%ecx
+	movl	%ecx,%ebx
+	jmp	.L057xts_dec_loop6
+.align	16
+.L057xts_dec_loop6:
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,(%esp)
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,16(%esp)
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,32(%esp)
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,48(%esp)
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	pshufd	$19,%xmm0,%xmm7
+	movdqa	%xmm1,64(%esp)
+	paddq	%xmm1,%xmm1
+	movups	(%ebp),%xmm0
+	pand	%xmm3,%xmm7
+	movups	(%esi),%xmm2
+	pxor	%xmm1,%xmm7
+	movdqu	16(%esi),%xmm3
+	xorps	%xmm0,%xmm2
+	movdqu	32(%esi),%xmm4
+	pxor	%xmm0,%xmm3
+	movdqu	48(%esi),%xmm5
+	pxor	%xmm0,%xmm4
+	movdqu	64(%esi),%xmm6
+	pxor	%xmm0,%xmm5
+	movdqu	80(%esi),%xmm1
+	pxor	%xmm0,%xmm6
+	leal	96(%esi),%esi
+	pxor	(%esp),%xmm2
+	movdqa	%xmm7,80(%esp)
+	pxor	%xmm1,%xmm7
+	movups	16(%ebp),%xmm1
+	leal	32(%ebp),%edx
+	pxor	16(%esp),%xmm3
+.byte	102,15,56,222,209
+	pxor	32(%esp),%xmm4
+.byte	102,15,56,222,217
+	pxor	48(%esp),%xmm5
+	decl	%ecx
+.byte	102,15,56,222,225
+	pxor	64(%esp),%xmm6
+.byte	102,15,56,222,233
+	pxor	%xmm0,%xmm7
+.byte	102,15,56,222,241
+	movups	(%edx),%xmm0
+.byte	102,15,56,222,249
+	call	.L_aesni_decrypt6_enter
+	movdqa	80(%esp),%xmm1
+	pxor	%xmm0,%xmm0
+	xorps	(%esp),%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	xorps	16(%esp),%xmm3
+	movups	%xmm2,(%edi)
+	xorps	32(%esp),%xmm4
+	movups	%xmm3,16(%edi)
+	xorps	48(%esp),%xmm5
+	movups	%xmm4,32(%edi)
+	xorps	64(%esp),%xmm6
+	movups	%xmm5,48(%edi)
+	xorps	%xmm1,%xmm7
+	movups	%xmm6,64(%edi)
+	pshufd	$19,%xmm0,%xmm2
+	movups	%xmm7,80(%edi)
+	leal	96(%edi),%edi
+	movdqa	96(%esp),%xmm3
+	pxor	%xmm0,%xmm0
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	movl	%ebx,%ecx
+	pxor	%xmm2,%xmm1
+	subl	$96,%eax
+	jnc	.L057xts_dec_loop6
+	leal	1(,%ecx,2),%ecx
+	movl	%ebp,%edx
+	movl	%ecx,%ebx
+.L056xts_dec_short:
+	addl	$96,%eax
+	jz	.L058xts_dec_done6x
+	movdqa	%xmm1,%xmm5
+	cmpl	$32,%eax
+	jb	.L059xts_dec_one
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	je	.L060xts_dec_two
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,%xmm6
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	cmpl	$64,%eax
+	jb	.L061xts_dec_three
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	%xmm1,%xmm7
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+	movdqa	%xmm5,(%esp)
+	movdqa	%xmm6,16(%esp)
+	je	.L062xts_dec_four
+	movdqa	%xmm7,32(%esp)
+	pshufd	$19,%xmm0,%xmm7
+	movdqa	%xmm1,48(%esp)
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm7
+	pxor	%xmm1,%xmm7
+	movdqu	(%esi),%xmm2
+	movdqu	16(%esi),%xmm3
+	movdqu	32(%esi),%xmm4
+	pxor	(%esp),%xmm2
+	movdqu	48(%esi),%xmm5
+	pxor	16(%esp),%xmm3
+	movdqu	64(%esi),%xmm6
+	pxor	32(%esp),%xmm4
+	leal	80(%esi),%esi
+	pxor	48(%esp),%xmm5
+	movdqa	%xmm7,64(%esp)
+	pxor	%xmm7,%xmm6
+	call	__aesni_decrypt6
+	movaps	64(%esp),%xmm1
+	xorps	(%esp),%xmm2
+	xorps	16(%esp),%xmm3
+	xorps	32(%esp),%xmm4
+	movups	%xmm2,(%edi)
+	xorps	48(%esp),%xmm5
+	movups	%xmm3,16(%edi)
+	xorps	%xmm1,%xmm6
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+	movups	%xmm6,64(%edi)
+	leal	80(%edi),%edi
+	jmp	.L063xts_dec_done
+.align	16
+.L059xts_dec_one:
+	movups	(%esi),%xmm2
+	leal	16(%esi),%esi
+	xorps	%xmm5,%xmm2
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L064dec1_loop_12:
+.byte	102,15,56,222,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L064dec1_loop_12
+.byte	102,15,56,223,209
+	xorps	%xmm5,%xmm2
+	movups	%xmm2,(%edi)
+	leal	16(%edi),%edi
+	movdqa	%xmm5,%xmm1
+	jmp	.L063xts_dec_done
+.align	16
+.L060xts_dec_two:
+	movaps	%xmm1,%xmm6
+	movups	(%esi),%xmm2
+	movups	16(%esi),%xmm3
+	leal	32(%esi),%esi
+	xorps	%xmm5,%xmm2
+	xorps	%xmm6,%xmm3
+	call	__aesni_decrypt3
+	xorps	%xmm5,%xmm2
+	xorps	%xmm6,%xmm3
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	leal	32(%edi),%edi
+	movdqa	%xmm6,%xmm1
+	jmp	.L063xts_dec_done
+.align	16
+.L061xts_dec_three:
+	movaps	%xmm1,%xmm7
+	movups	(%esi),%xmm2
+	movups	16(%esi),%xmm3
+	movups	32(%esi),%xmm4
+	leal	48(%esi),%esi
+	xorps	%xmm5,%xmm2
+	xorps	%xmm6,%xmm3
+	xorps	%xmm7,%xmm4
+	call	__aesni_decrypt3
+	xorps	%xmm5,%xmm2
+	xorps	%xmm6,%xmm3
+	xorps	%xmm7,%xmm4
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	leal	48(%edi),%edi
+	movdqa	%xmm7,%xmm1
+	jmp	.L063xts_dec_done
+.align	16
+.L062xts_dec_four:
+	movaps	%xmm1,%xmm6
+	movups	(%esi),%xmm2
+	movups	16(%esi),%xmm3
+	movups	32(%esi),%xmm4
+	xorps	(%esp),%xmm2
+	movups	48(%esi),%xmm5
+	leal	64(%esi),%esi
+	xorps	16(%esp),%xmm3
+	xorps	%xmm7,%xmm4
+	xorps	%xmm6,%xmm5
+	call	__aesni_decrypt4
+	xorps	(%esp),%xmm2
+	xorps	16(%esp),%xmm3
+	xorps	%xmm7,%xmm4
+	movups	%xmm2,(%edi)
+	xorps	%xmm6,%xmm5
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+	leal	64(%edi),%edi
+	movdqa	%xmm6,%xmm1
+	jmp	.L063xts_dec_done
+.align	16
+.L058xts_dec_done6x:
+	movl	112(%esp),%eax
+	andl	$15,%eax
+	jz	.L065xts_dec_ret
+	movl	%eax,112(%esp)
+	jmp	.L066xts_dec_only_one_more
+.align	16
+.L063xts_dec_done:
+	movl	112(%esp),%eax
+	pxor	%xmm0,%xmm0
+	andl	$15,%eax
+	jz	.L065xts_dec_ret
+	pcmpgtd	%xmm1,%xmm0
+	movl	%eax,112(%esp)
+	pshufd	$19,%xmm0,%xmm2
+	pxor	%xmm0,%xmm0
+	movdqa	96(%esp),%xmm3
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm2
+	pcmpgtd	%xmm1,%xmm0
+	pxor	%xmm2,%xmm1
+.L066xts_dec_only_one_more:
+	pshufd	$19,%xmm0,%xmm5
+	movdqa	%xmm1,%xmm6
+	paddq	%xmm1,%xmm1
+	pand	%xmm3,%xmm5
+	pxor	%xmm1,%xmm5
+	movl	%ebp,%edx
+	movl	%ebx,%ecx
+	movups	(%esi),%xmm2
+	xorps	%xmm5,%xmm2
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L067dec1_loop_13:
+.byte	102,15,56,222,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L067dec1_loop_13
+.byte	102,15,56,223,209
+	xorps	%xmm5,%xmm2
+	movups	%xmm2,(%edi)
+.L068xts_dec_steal:
+	movzbl	16(%esi),%ecx
+	movzbl	(%edi),%edx
+	leal	1(%esi),%esi
+	movb	%cl,(%edi)
+	movb	%dl,16(%edi)
+	leal	1(%edi),%edi
+	subl	$1,%eax
+	jnz	.L068xts_dec_steal
+	subl	112(%esp),%edi
+	movl	%ebp,%edx
+	movl	%ebx,%ecx
+	movups	(%edi),%xmm2
+	xorps	%xmm6,%xmm2
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L069dec1_loop_14:
+.byte	102,15,56,222,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L069dec1_loop_14
+.byte	102,15,56,223,209
+	xorps	%xmm6,%xmm2
+	movups	%xmm2,(%edi)
+.L065xts_dec_ret:
+	movl	116(%esp),%esp
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.globl	_aesni_cbc_encrypt
+.def	_aesni_cbc_encrypt;	.scl	2;	.type	32;	.endef
+.align	16
+_aesni_cbc_encrypt:
+.L_aesni_cbc_encrypt_begin:
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	movl	20(%esp),%esi
+	movl	%esp,%ebx
+	movl	24(%esp),%edi
+	subl	$24,%ebx
+	movl	28(%esp),%eax
+	andl	$-16,%ebx
+	movl	32(%esp),%edx
+	movl	36(%esp),%ebp
+	testl	%eax,%eax
+	jz	.L070cbc_abort
+	cmpl	$0,40(%esp)
+	xchgl	%esp,%ebx
+	movups	(%ebp),%xmm7
+	movl	240(%edx),%ecx
+	movl	%edx,%ebp
+	movl	%ebx,16(%esp)
+	movl	%ecx,%ebx
+	je	.L071cbc_decrypt
+	movaps	%xmm7,%xmm2
+	cmpl	$16,%eax
+	jb	.L072cbc_enc_tail
+	subl	$16,%eax
+	jmp	.L073cbc_enc_loop
+.align	16
+.L073cbc_enc_loop:
+	movups	(%esi),%xmm7
+	leal	16(%esi),%esi
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	xorps	%xmm0,%xmm7
+	leal	32(%edx),%edx
+	xorps	%xmm7,%xmm2
+.L074enc1_loop_15:
+.byte	102,15,56,220,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L074enc1_loop_15
+.byte	102,15,56,221,209
+	movl	%ebx,%ecx
+	movl	%ebp,%edx
+	movups	%xmm2,(%edi)
+	leal	16(%edi),%edi
+	subl	$16,%eax
+	jnc	.L073cbc_enc_loop
+	addl	$16,%eax
+	jnz	.L072cbc_enc_tail
+	movaps	%xmm2,%xmm7
+	jmp	.L075cbc_ret
+.L072cbc_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	.L073cbc_enc_loop
+.align	16
+.L071cbc_decrypt:
+	cmpl	$80,%eax
+	jbe	.L076cbc_dec_tail
+	movaps	%xmm7,(%esp)
+	subl	$80,%eax
+	jmp	.L077cbc_dec_loop6_enter
+.align	16
+.L078cbc_dec_loop6:
+	movaps	%xmm0,(%esp)
+	movups	%xmm7,(%edi)
+	leal	16(%edi),%edi
+.L077cbc_dec_loop6_enter:
+	movdqu	(%esi),%xmm2
+	movdqu	16(%esi),%xmm3
+	movdqu	32(%esi),%xmm4
+	movdqu	48(%esi),%xmm5
+	movdqu	64(%esi),%xmm6
+	movdqu	80(%esi),%xmm7
+	call	__aesni_decrypt6
+	movups	(%esi),%xmm1
+	movups	16(%esi),%xmm0
+	xorps	(%esp),%xmm2
+	xorps	%xmm1,%xmm3
+	movups	32(%esi),%xmm1
+	xorps	%xmm0,%xmm4
+	movups	48(%esi),%xmm0
+	xorps	%xmm1,%xmm5
+	movups	64(%esi),%xmm1
+	xorps	%xmm0,%xmm6
+	movups	80(%esi),%xmm0
+	xorps	%xmm1,%xmm7
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	leal	96(%esi),%esi
+	movups	%xmm4,32(%edi)
+	movl	%ebx,%ecx
+	movups	%xmm5,48(%edi)
+	movl	%ebp,%edx
+	movups	%xmm6,64(%edi)
+	leal	80(%edi),%edi
+	subl	$96,%eax
+	ja	.L078cbc_dec_loop6
+	movaps	%xmm7,%xmm2
+	movaps	%xmm0,%xmm7
+	addl	$80,%eax
+	jle	.L079cbc_dec_tail_collected
+	movups	%xmm2,(%edi)
+	leal	16(%edi),%edi
+.L076cbc_dec_tail:
+	movups	(%esi),%xmm2
+	movaps	%xmm2,%xmm6
+	cmpl	$16,%eax
+	jbe	.L080cbc_dec_one
+	movups	16(%esi),%xmm3
+	movaps	%xmm3,%xmm5
+	cmpl	$32,%eax
+	jbe	.L081cbc_dec_two
+	movups	32(%esi),%xmm4
+	cmpl	$48,%eax
+	jbe	.L082cbc_dec_three
+	movups	48(%esi),%xmm5
+	cmpl	$64,%eax
+	jbe	.L083cbc_dec_four
+	movups	64(%esi),%xmm6
+	movaps	%xmm7,(%esp)
+	movups	(%esi),%xmm2
+	xorps	%xmm7,%xmm7
+	call	__aesni_decrypt6
+	movups	(%esi),%xmm1
+	movups	16(%esi),%xmm0
+	xorps	(%esp),%xmm2
+	xorps	%xmm1,%xmm3
+	movups	32(%esi),%xmm1
+	xorps	%xmm0,%xmm4
+	movups	48(%esi),%xmm0
+	xorps	%xmm1,%xmm5
+	movups	64(%esi),%xmm7
+	xorps	%xmm0,%xmm6
+	movups	%xmm2,(%edi)
+	movups	%xmm3,16(%edi)
+	movups	%xmm4,32(%edi)
+	movups	%xmm5,48(%edi)
+	leal	64(%edi),%edi
+	movaps	%xmm6,%xmm2
+	subl	$80,%eax
+	jmp	.L079cbc_dec_tail_collected
+.align	16
+.L080cbc_dec_one:
+	movups	(%edx),%xmm0
+	movups	16(%edx),%xmm1
+	leal	32(%edx),%edx
+	xorps	%xmm0,%xmm2
+.L084dec1_loop_16:
+.byte	102,15,56,222,209
+	decl	%ecx
+	movups	(%edx),%xmm1
+	leal	16(%edx),%edx
+	jnz	.L084dec1_loop_16
+.byte	102,15,56,223,209
+	xorps	%xmm7,%xmm2
+	movaps	%xmm6,%xmm7
+	subl	$16,%eax
+	jmp	.L079cbc_dec_tail_collected
+.align	16
+.L081cbc_dec_two:
+	xorps	%xmm4,%xmm4
+	call	__aesni_decrypt3
+	xorps	%xmm7,%xmm2
+	xorps	%xmm6,%xmm3
+	movups	%xmm2,(%edi)
+	movaps	%xmm3,%xmm2
+	leal	16(%edi),%edi
+	movaps	%xmm5,%xmm7
+	subl	$32,%eax
+	jmp	.L079cbc_dec_tail_collected
+.align	16
+.L082cbc_dec_three:
+	call	__aesni_decrypt3
+	xorps	%xmm7,%xmm2
+	xorps	%xmm6,%xmm3
+	xorps	%xmm5,%xmm4
+	movups	%xmm2,(%edi)
+	movaps	%xmm4,%xmm2
+	movups	%xmm3,16(%edi)
+	leal	32(%edi),%edi
+	movups	32(%esi),%xmm7
+	subl	$48,%eax
+	jmp	.L079cbc_dec_tail_collected
+.align	16
+.L083cbc_dec_four:
+	call	__aesni_decrypt4
+	movups	16(%esi),%xmm1
+	movups	32(%esi),%xmm0
+	xorps	%xmm7,%xmm2
+	movups	48(%esi),%xmm7
+	xorps	%xmm6,%xmm3
+	movups	%xmm2,(%edi)
+	xorps	%xmm1,%xmm4
+	movups	%xmm3,16(%edi)
+	xorps	%xmm0,%xmm5
+	movups	%xmm4,32(%edi)
+	leal	48(%edi),%edi
+	movaps	%xmm5,%xmm2
+	subl	$64,%eax
+.L079cbc_dec_tail_collected:
+	andl	$15,%eax
+	jnz	.L085cbc_dec_tail_partial
+	movups	%xmm2,(%edi)
+	jmp	.L075cbc_ret
+.align	16
+.L085cbc_dec_tail_partial:
+	movaps	%xmm2,(%esp)
+	movl	$16,%ecx
+	movl	%esp,%esi
+	subl	%eax,%ecx
+.long	2767451785
+.L075cbc_ret:
+	movl	16(%esp),%esp
+	movl	36(%esp),%ebp
+	movups	%xmm7,(%ebp)
+.L070cbc_abort:
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.def	__aesni_set_encrypt_key;	.scl	3;	.type	32;	.endef
+.align	16
+__aesni_set_encrypt_key:
+	testl	%eax,%eax
+	jz	.L086bad_pointer
+	testl	%edx,%edx
+	jz	.L086bad_pointer
+	movups	(%eax),%xmm0
+	xorps	%xmm4,%xmm4
+	leal	16(%edx),%edx
+	cmpl	$256,%ecx
+	je	.L08714rounds
+	cmpl	$192,%ecx
+	je	.L08812rounds
+	cmpl	$128,%ecx
+	jne	.L089bad_keybits
+.align	16
+.L09010rounds:
+	movl	$9,%ecx
+	movups	%xmm0,-16(%edx)
+.byte	102,15,58,223,200,1
+	call	.L091key_128_cold
+.byte	102,15,58,223,200,2
+	call	.L092key_128
+.byte	102,15,58,223,200,4
+	call	.L092key_128
+.byte	102,15,58,223,200,8
+	call	.L092key_128
+.byte	102,15,58,223,200,16
+	call	.L092key_128
+.byte	102,15,58,223,200,32
+	call	.L092key_128
+.byte	102,15,58,223,200,64
+	call	.L092key_128
+.byte	102,15,58,223,200,128
+	call	.L092key_128
+.byte	102,15,58,223,200,27
+	call	.L092key_128
+.byte	102,15,58,223,200,54
+	call	.L092key_128
+	movups	%xmm0,(%edx)
+	movl	%ecx,80(%edx)
+	xorl	%eax,%eax
+	ret
+.align	16
+.L092key_128:
+	movups	%xmm0,(%edx)
+	leal	16(%edx),%edx
+.L091key_128_cold:
+	shufps	$16,%xmm0,%xmm4
+	xorps	%xmm4,%xmm0
+	shufps	$140,%xmm0,%xmm4
+	xorps	%xmm4,%xmm0
+	shufps	$255,%xmm1,%xmm1
+	xorps	%xmm1,%xmm0
+	ret
+.align	16
+.L08812rounds:
+	movq	16(%eax),%xmm2
+	movl	$11,%ecx
+	movups	%xmm0,-16(%edx)
+.byte	102,15,58,223,202,1
+	call	.L093key_192a_cold
+.byte	102,15,58,223,202,2
+	call	.L094key_192b
+.byte	102,15,58,223,202,4
+	call	.L095key_192a
+.byte	102,15,58,223,202,8
+	call	.L094key_192b
+.byte	102,15,58,223,202,16
+	call	.L095key_192a
+.byte	102,15,58,223,202,32
+	call	.L094key_192b
+.byte	102,15,58,223,202,64
+	call	.L095key_192a
+.byte	102,15,58,223,202,128
+	call	.L094key_192b
+	movups	%xmm0,(%edx)
+	movl	%ecx,48(%edx)
+	xorl	%eax,%eax
+	ret
+.align	16
+.L095key_192a:
+	movups	%xmm0,(%edx)
+	leal	16(%edx),%edx
+.align	16
+.L093key_192a_cold:
+	movaps	%xmm2,%xmm5
+.L096key_192b_warm:
+	shufps	$16,%xmm0,%xmm4
+	movdqa	%xmm2,%xmm3
+	xorps	%xmm4,%xmm0
+	shufps	$140,%xmm0,%xmm4
+	pslldq	$4,%xmm3
+	xorps	%xmm4,%xmm0
+	pshufd	$85,%xmm1,%xmm1
+	pxor	%xmm3,%xmm2
+	pxor	%xmm1,%xmm0
+	pshufd	$255,%xmm0,%xmm3
+	pxor	%xmm3,%xmm2
+	ret
+.align	16
+.L094key_192b:
+	movaps	%xmm0,%xmm3
+	shufps	$68,%xmm0,%xmm5
+	movups	%xmm5,(%edx)
+	shufps	$78,%xmm2,%xmm3
+	movups	%xmm3,16(%edx)
+	leal	32(%edx),%edx
+	jmp	.L096key_192b_warm
+.align	16
+.L08714rounds:
+	movups	16(%eax),%xmm2
+	movl	$13,%ecx
+	leal	16(%edx),%edx
+	movups	%xmm0,-32(%edx)
+	movups	%xmm2,-16(%edx)
+.byte	102,15,58,223,202,1
+	call	.L097key_256a_cold
+.byte	102,15,58,223,200,1
+	call	.L098key_256b
+.byte	102,15,58,223,202,2
+	call	.L099key_256a
+.byte	102,15,58,223,200,2
+	call	.L098key_256b
+.byte	102,15,58,223,202,4
+	call	.L099key_256a
+.byte	102,15,58,223,200,4
+	call	.L098key_256b
+.byte	102,15,58,223,202,8
+	call	.L099key_256a
+.byte	102,15,58,223,200,8
+	call	.L098key_256b
+.byte	102,15,58,223,202,16
+	call	.L099key_256a
+.byte	102,15,58,223,200,16
+	call	.L098key_256b
+.byte	102,15,58,223,202,32
+	call	.L099key_256a
+.byte	102,15,58,223,200,32
+	call	.L098key_256b
+.byte	102,15,58,223,202,64
+	call	.L099key_256a
+	movups	%xmm0,(%edx)
+	movl	%ecx,16(%edx)
+	xorl	%eax,%eax
+	ret
+.align	16
+.L099key_256a:
+	movups	%xmm2,(%edx)
+	leal	16(%edx),%edx
+.L097key_256a_cold:
+	shufps	$16,%xmm0,%xmm4
+	xorps	%xmm4,%xmm0
+	shufps	$140,%xmm0,%xmm4
+	xorps	%xmm4,%xmm0
+	shufps	$255,%xmm1,%xmm1
+	xorps	%xmm1,%xmm0
+	ret
+.align	16
+.L098key_256b:
+	movups	%xmm0,(%edx)
+	leal	16(%edx),%edx
+	shufps	$16,%xmm2,%xmm4
+	xorps	%xmm4,%xmm2
+	shufps	$140,%xmm2,%xmm4
+	xorps	%xmm4,%xmm2
+	shufps	$170,%xmm1,%xmm1
+	xorps	%xmm1,%xmm2
+	ret
+.align	4
+.L086bad_pointer:
+	movl	$-1,%eax
+	ret
+.align	4
+.L089bad_keybits:
+	movl	$-2,%eax
+	ret
+.globl	_aesni_set_encrypt_key
+.def	_aesni_set_encrypt_key;	.scl	2;	.type	32;	.endef
+.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
+.globl	_aesni_set_decrypt_key
+.def	_aesni_set_decrypt_key;	.scl	2;	.type	32;	.endef
+.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	.L100dec_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
+.L101dec_key_inverse:
+	movups	(%edx),%xmm0
+	movups	(%eax),%xmm1
+.byte	102,15,56,219,192
+.byte	102,15,56,219,201
+	leal	16(%edx),%edx
+	leal	-16(%eax),%eax
+	movups	%xmm0,16(%eax)
+	movups	%xmm1,-16(%edx)
+	cmpl	%edx,%eax
+	ja	.L101dec_key_inverse
+	movups	(%edx),%xmm0
+.byte	102,15,56,219,192
+	movups	%xmm0,(%edx)
+	xorl	%eax,%eax
+.L100dec_key_ret:
+	ret
+.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/x86/coff/padlock-x86-coff.s b/lib/accelerated/x86/coff/padlock-x86-coff.s
new file mode 100644
index 0000000000..c1014bd122
--- /dev/null
+++ b/lib/accelerated/x86/coff/padlock-x86-coff.s
@@ -0,0 +1,615 @@
+# Copyright (c) 2011, 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	"./engines/asm/e_padlock-x86.s"
+.text
+.globl	_padlock_capability
+.def	_padlock_capability;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_capability:
+.L_padlock_capability_begin:
+	pushl	%ebx
+	pushfl
+	popl	%eax
+	movl	%eax,%ecx
+	xorl	$2097152,%eax
+	pushl	%eax
+	popfl
+	pushfl
+	popl	%eax
+	xorl	%eax,%ecx
+	xorl	%eax,%eax
+	btl	$21,%ecx
+	jnc	.L000noluck
+	.byte	0x0f,0xa2
+	xorl	%eax,%eax
+	cmpl	$0x746e6543,%ebx
+	jne	.L000noluck
+	cmpl	$0x48727561,%edx
+	jne	.L000noluck
+	cmpl	$0x736c7561,%ecx
+	jne	.L000noluck
+	movl	$3221225472,%eax
+	.byte	0x0f,0xa2
+	movl	%eax,%edx
+	xorl	%eax,%eax
+	cmpl	$3221225473,%edx
+	jb	.L000noluck
+	movl	$1,%eax
+	.byte	0x0f,0xa2
+	orl	$15,%eax
+	xorl	%ebx,%ebx
+	andl	$4095,%eax
+	cmpl	$1791,%eax
+	sete	%bl
+	movl	$3221225473,%eax
+	pushl	%ebx
+	.byte	0x0f,0xa2
+	popl	%ebx
+	movl	%edx,%eax
+	shll	$4,%ebx
+	andl	$4294967279,%eax
+	orl	%ebx,%eax
+.L000noluck:
+	popl	%ebx
+	ret
+.globl	_padlock_key_bswap
+.def	_padlock_key_bswap;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_key_bswap:
+.L_padlock_key_bswap_begin:
+	movl	4(%esp),%edx
+	movl	240(%edx),%ecx
+.L001bswap_loop:
+	movl	(%edx),%eax
+	bswap	%eax
+	movl	%eax,(%edx)
+	leal	4(%edx),%edx
+	subl	$1,%ecx
+	jnz	.L001bswap_loop
+	ret
+.globl	_padlock_verify_context
+.def	_padlock_verify_context;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_verify_context:
+.L_padlock_verify_context_begin:
+	movl	4(%esp),%edx
+	leal	.Lpadlock_saved_context,%eax
+	pushfl
+	call	__padlock_verify_ctx
+.L002verify_pic_point:
+	leal	4(%esp),%esp
+	ret
+.def	__padlock_verify_ctx;	.scl	3;	.type	32;	.endef
+.align	16
+__padlock_verify_ctx:
+	btl	$30,4(%esp)
+	jnc	.L003verified
+	cmpl	(%eax),%edx
+	je	.L003verified
+	pushfl
+	popfl
+.L003verified:
+	movl	%edx,(%eax)
+	ret
+.globl	_padlock_reload_key
+.def	_padlock_reload_key;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_reload_key:
+.L_padlock_reload_key_begin:
+	pushfl
+	popfl
+	ret
+.globl	_padlock_aes_block
+.def	_padlock_aes_block;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_aes_block:
+.L_padlock_aes_block_begin:
+	pushl	%edi
+	pushl	%esi
+	pushl	%ebx
+	movl	16(%esp),%edi
+	movl	20(%esp),%esi
+	movl	24(%esp),%edx
+	movl	$1,%ecx
+	leal	32(%edx),%ebx
+	leal	16(%edx),%edx
+.byte	243,15,167,200
+	popl	%ebx
+	popl	%esi
+	popl	%edi
+	ret
+.globl	_padlock_ecb_encrypt
+.def	_padlock_ecb_encrypt;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_ecb_encrypt:
+.L_padlock_ecb_encrypt_begin:
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	movl	20(%esp),%edi
+	movl	24(%esp),%esi
+	movl	28(%esp),%edx
+	movl	32(%esp),%ecx
+	testl	$15,%edx
+	jnz	.L004ecb_abort
+	testl	$15,%ecx
+	jnz	.L004ecb_abort
+	leal	.Lpadlock_saved_context,%eax
+	pushfl
+	cld
+	call	__padlock_verify_ctx
+.L005ecb_pic_point:
+	leal	16(%edx),%edx
+	xorl	%eax,%eax
+	xorl	%ebx,%ebx
+	cmpl	$128,%ecx
+	jbe	.L006ecb_short
+	testl	$32,(%edx)
+	jnz	.L007ecb_aligned
+	testl	$15,%edi
+	setz	%al
+	testl	$15,%esi
+	setz	%bl
+	testl	%ebx,%eax
+	jnz	.L007ecb_aligned
+	negl	%eax
+	movl	$512,%ebx
+	notl	%eax
+	leal	-24(%esp),%ebp
+	cmpl	%ebx,%ecx
+	cmovcl	%ecx,%ebx
+	andl	%ebx,%eax
+	movl	%ecx,%ebx
+	negl	%eax
+	andl	$511,%ebx
+	leal	(%eax,%ebp,1),%esp
+	andl	$-16,%esp
+	jmp	.L008ecb_loop
+.align	16
+.L008ecb_loop:
+	movl	%edi,(%ebp)
+	movl	%esi,4(%ebp)
+	movl	%ecx,8(%ebp)
+	movl	%ebx,%ecx
+	movl	%ebx,12(%ebp)
+	testl	$15,%edi
+	cmovnzl	%esp,%edi
+	testl	$15,%esi
+	jz	.L009ecb_inp_aligned
+	shrl	$2,%ecx
+.byte	243,165
+	subl	%ebx,%edi
+	movl	%ebx,%ecx
+	movl	%edi,%esi
+.L009ecb_inp_aligned:
+	leal	-16(%edx),%eax
+	leal	16(%edx),%ebx
+	shrl	$4,%ecx
+.byte	243,15,167,200
+	movl	(%ebp),%edi
+	movl	12(%ebp),%ebx
+	testl	$15,%edi
+	jz	.L010ecb_out_aligned
+	movl	%ebx,%ecx
+	shrl	$2,%ecx
+	leal	(%esp),%esi
+.byte	243,165
+	subl	%ebx,%edi
+.L010ecb_out_aligned:
+	movl	4(%ebp),%esi
+	movl	8(%ebp),%ecx
+	addl	%ebx,%edi
+	addl	%ebx,%esi
+	subl	%ebx,%ecx
+	movl	$512,%ebx
+	jnz	.L008ecb_loop
+	cmpl	%ebp,%esp
+	je	.L011ecb_done
+	pxor	%xmm0,%xmm0
+	leal	(%esp),%eax
+.L012ecb_bzero:
+	movaps	%xmm0,(%eax)
+	leal	16(%eax),%eax
+	cmpl	%eax,%ebp
+	ja	.L012ecb_bzero
+.L011ecb_done:
+	leal	24(%ebp),%esp
+	jmp	.L013ecb_exit
+.align	16
+.L006ecb_short:
+	xorl	%eax,%eax
+	leal	-24(%esp),%ebp
+	subl	%ecx,%eax
+	leal	(%eax,%ebp,1),%esp
+	andl	$-16,%esp
+	xorl	%ebx,%ebx
+.L014ecb_short_copy:
+	movups	(%esi,%ebx,1),%xmm0
+	leal	16(%ebx),%ebx
+	cmpl	%ebx,%ecx
+	movaps	%xmm0,-16(%esp,%ebx,1)
+	ja	.L014ecb_short_copy
+	movl	%esp,%esi
+	movl	%ecx,%ebx
+	jmp	.L008ecb_loop
+.align	16
+.L007ecb_aligned:
+	leal	-16(%edx),%eax
+	leal	16(%edx),%ebx
+	shrl	$4,%ecx
+.byte	243,15,167,200
+.L013ecb_exit:
+	movl	$1,%eax
+	leal	4(%esp),%esp
+.L004ecb_abort:
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.globl	_padlock_cbc_encrypt
+.def	_padlock_cbc_encrypt;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_cbc_encrypt:
+.L_padlock_cbc_encrypt_begin:
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	movl	20(%esp),%edi
+	movl	24(%esp),%esi
+	movl	28(%esp),%edx
+	movl	32(%esp),%ecx
+	testl	$15,%edx
+	jnz	.L015cbc_abort
+	testl	$15,%ecx
+	jnz	.L015cbc_abort
+	leal	.Lpadlock_saved_context,%eax
+	pushfl
+	cld
+	call	__padlock_verify_ctx
+.L016cbc_pic_point:
+	leal	16(%edx),%edx
+	xorl	%eax,%eax
+	xorl	%ebx,%ebx
+	cmpl	$64,%ecx
+	jbe	.L017cbc_short
+	testl	$32,(%edx)
+	jnz	.L018cbc_aligned
+	testl	$15,%edi
+	setz	%al
+	testl	$15,%esi
+	setz	%bl
+	testl	%ebx,%eax
+	jnz	.L018cbc_aligned
+	negl	%eax
+	movl	$512,%ebx
+	notl	%eax
+	leal	-24(%esp),%ebp
+	cmpl	%ebx,%ecx
+	cmovcl	%ecx,%ebx
+	andl	%ebx,%eax
+	movl	%ecx,%ebx
+	negl	%eax
+	andl	$511,%ebx
+	leal	(%eax,%ebp,1),%esp
+	andl	$-16,%esp
+	jmp	.L019cbc_loop
+.align	16
+.L019cbc_loop:
+	movl	%edi,(%ebp)
+	movl	%esi,4(%ebp)
+	movl	%ecx,8(%ebp)
+	movl	%ebx,%ecx
+	movl	%ebx,12(%ebp)
+	testl	$15,%edi
+	cmovnzl	%esp,%edi
+	testl	$15,%esi
+	jz	.L020cbc_inp_aligned
+	shrl	$2,%ecx
+.byte	243,165
+	subl	%ebx,%edi
+	movl	%ebx,%ecx
+	movl	%edi,%esi
+.L020cbc_inp_aligned:
+	leal	-16(%edx),%eax
+	leal	16(%edx),%ebx
+	shrl	$4,%ecx
+.byte	243,15,167,208
+	movaps	(%eax),%xmm0
+	movaps	%xmm0,-16(%edx)
+	movl	(%ebp),%edi
+	movl	12(%ebp),%ebx
+	testl	$15,%edi
+	jz	.L021cbc_out_aligned
+	movl	%ebx,%ecx
+	shrl	$2,%ecx
+	leal	(%esp),%esi
+.byte	243,165
+	subl	%ebx,%edi
+.L021cbc_out_aligned:
+	movl	4(%ebp),%esi
+	movl	8(%ebp),%ecx
+	addl	%ebx,%edi
+	addl	%ebx,%esi
+	subl	%ebx,%ecx
+	movl	$512,%ebx
+	jnz	.L019cbc_loop
+	cmpl	%ebp,%esp
+	je	.L022cbc_done
+	pxor	%xmm0,%xmm0
+	leal	(%esp),%eax
+.L023cbc_bzero:
+	movaps	%xmm0,(%eax)
+	leal	16(%eax),%eax
+	cmpl	%eax,%ebp
+	ja	.L023cbc_bzero
+.L022cbc_done:
+	leal	24(%ebp),%esp
+	jmp	.L024cbc_exit
+.align	16
+.L017cbc_short:
+	xorl	%eax,%eax
+	leal	-24(%esp),%ebp
+	subl	%ecx,%eax
+	leal	(%eax,%ebp,1),%esp
+	andl	$-16,%esp
+	xorl	%ebx,%ebx
+.L025cbc_short_copy:
+	movups	(%esi,%ebx,1),%xmm0
+	leal	16(%ebx),%ebx
+	cmpl	%ebx,%ecx
+	movaps	%xmm0,-16(%esp,%ebx,1)
+	ja	.L025cbc_short_copy
+	movl	%esp,%esi
+	movl	%ecx,%ebx
+	jmp	.L019cbc_loop
+.align	16
+.L018cbc_aligned:
+	leal	-16(%edx),%eax
+	leal	16(%edx),%ebx
+	shrl	$4,%ecx
+.byte	243,15,167,208
+	movaps	(%eax),%xmm0
+	movaps	%xmm0,-16(%edx)
+.L024cbc_exit:
+	movl	$1,%eax
+	leal	4(%esp),%esp
+.L015cbc_abort:
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.globl	_padlock_xstore
+.def	_padlock_xstore;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_xstore:
+.L_padlock_xstore_begin:
+	pushl	%edi
+	movl	8(%esp),%edi
+	movl	12(%esp),%edx
+.byte	15,167,192
+	popl	%edi
+	ret
+.def	__win32_segv_handler;	.scl	3;	.type	32;	.endef
+.align	16
+__win32_segv_handler:
+	movl	$1,%eax
+	movl	4(%esp),%edx
+	movl	12(%esp),%ecx
+	cmpl	$3221225477,(%edx)
+	jne	.L026ret
+	addl	$4,184(%ecx)
+	movl	$0,%eax
+.L026ret:
+	ret
+.globl	_padlock_sha1_oneshot
+.def	_padlock_sha1_oneshot;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_sha1_oneshot:
+.L_padlock_sha1_oneshot_begin:
+	pushl	%edi
+	pushl	%esi
+	xorl	%eax,%eax
+	movl	12(%esp),%edi
+	movl	16(%esp),%esi
+	movl	20(%esp),%ecx
+	pushl	__win32_segv_handler
+.byte	100,255,48
+.byte	100,137,32
+	movl	%esp,%edx
+	addl	$-128,%esp
+	movups	(%edi),%xmm0
+	andl	$-16,%esp
+	movl	16(%edi),%eax
+	movaps	%xmm0,(%esp)
+	movl	%esp,%edi
+	movl	%eax,16(%esp)
+	xorl	%eax,%eax
+.byte	243,15,166,200
+	movaps	(%esp),%xmm0
+	movl	16(%esp),%eax
+	movl	%edx,%esp
+.byte	100,143,5,0,0,0,0
+	leal	4(%esp),%esp
+	movl	16(%esp),%edi
+	movups	%xmm0,(%edi)
+	movl	%eax,16(%edi)
+	popl	%esi
+	popl	%edi
+	ret
+.globl	_padlock_sha1_blocks
+.def	_padlock_sha1_blocks;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_sha1_blocks:
+.L_padlock_sha1_blocks_begin:
+	pushl	%edi
+	pushl	%esi
+	movl	12(%esp),%edi
+	movl	16(%esp),%esi
+	movl	%esp,%edx
+	movl	20(%esp),%ecx
+	addl	$-128,%esp
+	movups	(%edi),%xmm0
+	andl	$-16,%esp
+	movl	16(%edi),%eax
+	movaps	%xmm0,(%esp)
+	movl	%esp,%edi
+	movl	%eax,16(%esp)
+	movl	$-1,%eax
+.byte	243,15,166,200
+	movaps	(%esp),%xmm0
+	movl	16(%esp),%eax
+	movl	%edx,%esp
+	movl	12(%esp),%edi
+	movups	%xmm0,(%edi)
+	movl	%eax,16(%edi)
+	popl	%esi
+	popl	%edi
+	ret
+.globl	_padlock_sha256_oneshot
+.def	_padlock_sha256_oneshot;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_sha256_oneshot:
+.L_padlock_sha256_oneshot_begin:
+	pushl	%edi
+	pushl	%esi
+	xorl	%eax,%eax
+	movl	12(%esp),%edi
+	movl	16(%esp),%esi
+	movl	20(%esp),%ecx
+	pushl	__win32_segv_handler
+.byte	100,255,48
+.byte	100,137,32
+	movl	%esp,%edx
+	addl	$-128,%esp
+	movups	(%edi),%xmm0
+	andl	$-16,%esp
+	movups	16(%edi),%xmm1
+	movaps	%xmm0,(%esp)
+	movl	%esp,%edi
+	movaps	%xmm1,16(%esp)
+	xorl	%eax,%eax
+.byte	243,15,166,208
+	movaps	(%esp),%xmm0
+	movaps	16(%esp),%xmm1
+	movl	%edx,%esp
+.byte	100,143,5,0,0,0,0
+	leal	4(%esp),%esp
+	movl	16(%esp),%edi
+	movups	%xmm0,(%edi)
+	movups	%xmm1,16(%edi)
+	popl	%esi
+	popl	%edi
+	ret
+.globl	_padlock_sha256_blocks
+.def	_padlock_sha256_blocks;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_sha256_blocks:
+.L_padlock_sha256_blocks_begin:
+	pushl	%edi
+	pushl	%esi
+	movl	12(%esp),%edi
+	movl	16(%esp),%esi
+	movl	20(%esp),%ecx
+	movl	%esp,%edx
+	addl	$-128,%esp
+	movups	(%edi),%xmm0
+	andl	$-16,%esp
+	movups	16(%edi),%xmm1
+	movaps	%xmm0,(%esp)
+	movl	%esp,%edi
+	movaps	%xmm1,16(%esp)
+	movl	$-1,%eax
+.byte	243,15,166,208
+	movaps	(%esp),%xmm0
+	movaps	16(%esp),%xmm1
+	movl	%edx,%esp
+	movl	12(%esp),%edi
+	movups	%xmm0,(%edi)
+	movups	%xmm1,16(%edi)
+	popl	%esi
+	popl	%edi
+	ret
+.globl	_padlock_sha512_blocks
+.def	_padlock_sha512_blocks;	.scl	2;	.type	32;	.endef
+.align	16
+_padlock_sha512_blocks:
+.L_padlock_sha512_blocks_begin:
+	pushl	%edi
+	pushl	%esi
+	movl	12(%esp),%edi
+	movl	16(%esp),%esi
+	movl	20(%esp),%ecx
+	movl	%esp,%edx
+	addl	$-128,%esp
+	movups	(%edi),%xmm0
+	andl	$-16,%esp
+	movups	16(%edi),%xmm1
+	movups	32(%edi),%xmm2
+	movups	48(%edi),%xmm3
+	movaps	%xmm0,(%esp)
+	movl	%esp,%edi
+	movaps	%xmm1,16(%esp)
+	movaps	%xmm2,32(%esp)
+	movaps	%xmm3,48(%esp)
+.byte	243,15,166,224
+	movaps	(%esp),%xmm0
+	movaps	16(%esp),%xmm1
+	movaps	32(%esp),%xmm2
+	movaps	48(%esp),%xmm3
+	movl	%edx,%esp
+	movl	12(%esp),%edi
+	movups	%xmm0,(%edi)
+	movups	%xmm1,16(%edi)
+	movups	%xmm2,32(%edi)
+	movups	%xmm3,48(%edi)
+	popl	%esi
+	popl	%edi
+	ret
+.byte	86,73,65,32,80,97,100,108,111,99,107,32,120,56,54,32
+.byte	109,111,100,117,108,101,44,32,67,82,89,80,84,79,71,65
+.byte	77,83,32,98,121,32,60,97,112,112,114,111,64,111,112,101
+.byte	110,115,115,108,46,111,114,103,62,0
+.align	16
+.data
+.align	4
+.Lpadlock_saved_context:
+.long	0