Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 10 May 2016 20:31:09 +0000 (UTC)
From:      Jung-uk Kim <jkim@FreeBSD.org>
To:        src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org
Subject:   svn commit: r299389 - in head/secure/lib/libcrypto: . i386
Message-ID:  <201605102031.u4AKV9Ys060380@repo.freebsd.org>

next in thread | raw e-mail | index | archive | help
Author: jkim
Date: Tue May 10 20:31:09 2016
New Revision: 299389
URL: https://svnweb.freebsd.org/changeset/base/299389

Log:
  Make libcrypto.so position independent on i386.

Added:
  head/secure/lib/libcrypto/i386/aes-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/aes-586.s
  head/secure/lib/libcrypto/i386/aesni-x86.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/aesni-x86.s
  head/secure/lib/libcrypto/i386/bf-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/bf-586.s
  head/secure/lib/libcrypto/i386/bf-686.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/bf-686.s
  head/secure/lib/libcrypto/i386/bn-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/bn-586.s
  head/secure/lib/libcrypto/i386/cmll-x86.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/cmll-x86.s
  head/secure/lib/libcrypto/i386/co-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/co-586.s
  head/secure/lib/libcrypto/i386/crypt586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/crypt586.s
  head/secure/lib/libcrypto/i386/des-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/des-586.s
  head/secure/lib/libcrypto/i386/ghash-x86.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/ghash-x86.s
  head/secure/lib/libcrypto/i386/md5-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/md5-586.s
  head/secure/lib/libcrypto/i386/rc4-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/rc4-586.s
  head/secure/lib/libcrypto/i386/rc5-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/rc5-586.s
  head/secure/lib/libcrypto/i386/rmd-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/rmd-586.s
  head/secure/lib/libcrypto/i386/sha1-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/sha1-586.s
  head/secure/lib/libcrypto/i386/sha256-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/sha256-586.s
  head/secure/lib/libcrypto/i386/sha512-586.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/sha512-586.s
  head/secure/lib/libcrypto/i386/vpaes-x86.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/vpaes-x86.s
  head/secure/lib/libcrypto/i386/wp-mmx.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/wp-mmx.s
  head/secure/lib/libcrypto/i386/x86-gf2m.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/x86-gf2m.s
  head/secure/lib/libcrypto/i386/x86-mont.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/x86-mont.s
  head/secure/lib/libcrypto/i386/x86cpuid.S
     - copied, changed from r299388, head/secure/lib/libcrypto/i386/x86cpuid.s
Deleted:
  head/secure/lib/libcrypto/i386/aes-586.s
  head/secure/lib/libcrypto/i386/aesni-x86.s
  head/secure/lib/libcrypto/i386/bf-586.s
  head/secure/lib/libcrypto/i386/bf-686.s
  head/secure/lib/libcrypto/i386/bn-586.s
  head/secure/lib/libcrypto/i386/cmll-x86.s
  head/secure/lib/libcrypto/i386/co-586.s
  head/secure/lib/libcrypto/i386/crypt586.s
  head/secure/lib/libcrypto/i386/des-586.s
  head/secure/lib/libcrypto/i386/ghash-x86.s
  head/secure/lib/libcrypto/i386/md5-586.s
  head/secure/lib/libcrypto/i386/rc4-586.s
  head/secure/lib/libcrypto/i386/rc5-586.s
  head/secure/lib/libcrypto/i386/rmd-586.s
  head/secure/lib/libcrypto/i386/sha1-586.s
  head/secure/lib/libcrypto/i386/sha256-586.s
  head/secure/lib/libcrypto/i386/sha512-586.s
  head/secure/lib/libcrypto/i386/vpaes-x86.s
  head/secure/lib/libcrypto/i386/wp-mmx.s
  head/secure/lib/libcrypto/i386/x86-gf2m.s
  head/secure/lib/libcrypto/i386/x86-mont.s
  head/secure/lib/libcrypto/i386/x86cpuid.s
Modified:
  head/secure/lib/libcrypto/Makefile
  head/secure/lib/libcrypto/Makefile.asm

Modified: head/secure/lib/libcrypto/Makefile
==============================================================================
--- head/secure/lib/libcrypto/Makefile	Tue May 10 20:25:49 2016	(r299388)
+++ head/secure/lib/libcrypto/Makefile	Tue May 10 20:31:09 2016	(r299389)
@@ -26,7 +26,7 @@ SRCS=	cpt_err.c cryptlib.c cversion.c ex
 .if ${MACHINE_CPUARCH} == "amd64"
 SRCS+=	x86_64cpuid.S
 .elif ${MACHINE_CPUARCH} == "i386"
-SRCS+=	x86cpuid.s
+SRCS+=	x86cpuid.S
 .else
 SRCS+=	mem_clr.c
 .endif
@@ -38,7 +38,7 @@ SRCS+=	aes_cfb.c aes_ctr.c aes_ecb.c aes
 SRCS+=	aes-x86_64.S aesni-mb-x86_64.S aesni-sha1-x86_64.S \
 	aesni-sha256-x86_64.S aesni-x86_64.S bsaes-x86_64.S vpaes-x86_64.S
 .elif ${MACHINE_CPUARCH} == "i386"
-SRCS+=	aes-586.s aesni-x86.s vpaes-x86.s
+SRCS+=	aes-586.S aesni-x86.S vpaes-x86.S
 .else
 SRCS+=	aes_cbc.c aes_core.c
 .endif
@@ -63,9 +63,9 @@ INCS+=	asn1.h asn1_mac.h asn1t.h
 SRCS+=	bf_cfb64.c bf_ecb.c bf_ofb64.c bf_skey.c
 .if ${MACHINE_CPUARCH} == "i386"
 .if ${MACHINE_CPU:Mi686}
-SRCS+=	bf-686.s
+SRCS+=	bf-686.S
 .else
-SRCS+=	bf-586.s
+SRCS+=	bf-586.S
 .endif
 .else
 SRCS+=	bf_enc.c
@@ -87,7 +87,7 @@ SRCS+=	bn_add.c bn_blind.c bn_const.c bn
 SRCS+=	rsaz-avx2.S rsaz-x86_64.S rsaz_exp.c x86_64-gcc.c x86_64-gf2m.S \
 	x86_64-mont.S x86_64-mont5.S
 .elif ${MACHINE_CPUARCH} == "i386"
-SRCS+=	bn-586.s co-586.s x86-gf2m.s x86-mont.s
+SRCS+=	bn-586.S co-586.S x86-gf2m.S x86-mont.S
 .else
 SRCS+=	bn_asm.c
 .endif
@@ -102,7 +102,7 @@ SRCS+=	cmll_cfb.c cmll_ctr.c cmll_ecb.c 
 .if ${MACHINE_CPUARCH} == "amd64"
 SRCS+=	cmll_misc.c cmll-x86_64.S
 .elif ${MACHINE_CPUARCH} == "i386"
-SRCS+=	cmll-x86.s
+SRCS+=	cmll-x86.S
 .else
 SRCS+=	camellia.c cmll_cbc.c cmll_misc.c
 .endif
@@ -137,7 +137,7 @@ SRCS+=	cbc_cksm.c cbc_enc.c cfb64ede.c c
 	fcrypt.c ofb64ede.c ofb64enc.c ofb_enc.c pcbc_enc.c qud_cksm.c \
 	rand_key.c read2pwd.c rpc_enc.c set_key.c str2key.c xcbc_enc.c
 .if ${MACHINE_CPUARCH} == "i386"
-SRCS+=	crypt586.s des-586.s
+SRCS+=	crypt586.S des-586.S
 .else
 SRCS+=	des_enc.c fcrypt_b.c
 .endif
@@ -222,7 +222,7 @@ SRCS+=	md5_dgst.c md5_one.c
 .if ${MACHINE_CPUARCH} == "amd64"
 SRCS+=	md5-x86_64.S
 .elif ${MACHINE_CPUARCH} == "i386"
-SRCS+=	md5-586.s
+SRCS+=	md5-586.S
 .endif
 INCS+=	md5.h
 
@@ -236,7 +236,7 @@ SRCS+=	cbc128.c ccm128.c cfb128.c ctr128
 .if ${MACHINE_CPUARCH} == "amd64" 
 SRCS+=	aesni-gcm-x86_64.S ghash-x86_64.S
 .elif ${MACHINE_CPUARCH} == "i386"
-SRCS+=	ghash-x86.s
+SRCS+=	ghash-x86.S
 .endif
 INCS+=	modes.h
 
@@ -282,7 +282,7 @@ SRCS+=	rc4_utl.c
 .if ${MACHINE_CPUARCH} == "amd64" 
 SRCS+=	rc4-md5-x86_64.S rc4-x86_64.S
 .elif ${MACHINE_CPUARCH} == "i386"
-SRCS+=	rc4-586.s
+SRCS+=	rc4-586.S
 .else
 SRCS+=	rc4_enc.c rc4_skey.c
 .endif
@@ -291,7 +291,7 @@ INCS+=	rc4.h
 # rc5
 SRCS+=	rc5_ecb.c rc5_skey.c rc5cfb64.c rc5ofb64.c
 .if ${MACHINE_CPUARCH} == "i386"
-SRCS+=	rc5-586.s
+SRCS+=	rc5-586.S
 .else
 SRCS+=	rc5_enc.c
 .endif
@@ -300,7 +300,7 @@ INCS+=	rc5.h
 # ripemd
 SRCS+=	rmd_dgst.c rmd_one.c
 .if ${MACHINE_CPUARCH} == "i386"
-SRCS+=	rmd-586.s
+SRCS+=	rmd-586.S
 .endif
 INCS+=	ripemd.h
 
@@ -321,7 +321,7 @@ SRCS+=	sha1_one.c sha1dgst.c sha256.c sh
 SRCS+=	sha1-mb-x86_64.S sha1-x86_64.S sha256-mb-x86_64.S sha256-x86_64.S \
 	sha512-x86_64.S
 .elif ${MACHINE_CPUARCH} == "i386"
-SRCS+=	sha1-586.s sha256-586.s sha512-586.s
+SRCS+=	sha1-586.S sha256-586.S sha512-586.S
 .endif
 INCS+=	sha.h
 
@@ -352,7 +352,7 @@ SRCS+=	wp_dgst.c
 .if ${MACHINE_CPUARCH} == "amd64" 
 SRCS+=	wp-x86_64.S
 .elif ${MACHINE_CPUARCH} == "i386"
-SRCS+=	wp-mmx.s wp_block.c
+SRCS+=	wp-mmx.S wp_block.c
 .else
 SRCS+=	wp_block.c
 .endif
@@ -386,9 +386,6 @@ CFLAGS+=	-I${LCRYPTO_SRC}/crypto/asn1
 CFLAGS+=	-I${LCRYPTO_SRC}/crypto/evp
 CFLAGS+=	-I${LCRYPTO_SRC}/crypto/modes
 
-.if !empty(SRCS:M*.s)
-AFLAGS+=	--noexecstack
-.endif
 .if !empty(SRCS:M*.S)
 ACFLAGS+=	-Wa,--noexecstack
 .endif

Modified: head/secure/lib/libcrypto/Makefile.asm
==============================================================================
--- head/secure/lib/libcrypto/Makefile.asm	Tue May 10 20:25:49 2016	(r299388)
+++ head/secure/lib/libcrypto/Makefile.asm	Tue May 10 20:31:09 2016	(r299389)
@@ -1,8 +1,8 @@
 # $FreeBSD$
-# Use this to help generate the asm *.[Ss] files after an import.  It is not
+# Use this to help generate the asm *.S files after an import.  It is not
 # perfect by any means, but does what is needed.
-# Do a 'make -f Makefile.asm all' and it will generate *.s.  Move them
-# to the i386 subdir, and correct any exposed paths and $ FreeBSD $ tags.
+# Do a 'make -f Makefile.asm all' and it will generate *.S.  Move them
+# to the arch subdir, and correct any exposed paths and $ FreeBSD $ tags.
 
 .include "Makefile.inc"
 
@@ -127,16 +127,21 @@ SRCS+=	wp-mmx.pl
 # cpuid
 SRCS+=	x86cpuid.pl
 
-ASM=	${SRCS:S/.pl/.s/}
+ASM=	${SRCS:S/.pl/.S/}
 
 all:	${ASM}
 
-CLEANFILES+=	${SRCS:M*.pl:S/.pl$/.s/}
+CLEANFILES+=	${SRCS:M*.pl:S/.pl$/.S/}
 .SUFFIXES:	.pl
 
-.pl.s:
+.pl.S:
 	( echo '	# $$'FreeBSD'$$' ;\
-	perl ${PERLPATH} ${.IMPSRC} elf ${CFLAGS} ) > ${.TARGET}
+	echo '#ifdef PIC' ;\
+	perl ${PERLPATH} ${.IMPSRC} elf ${CFLAGS} -fpic -DPIC ;\
+	echo '#else' ;\
+	perl ${PERLPATH} ${.IMPSRC} elf ${CFLAGS} ;\
+	echo '#endif') |\
+	sed -E 's|(\.file[[:blank:]]+)".*"|\1"${.TARGET}"|' > ${.TARGET}
 .endif
 
 .include <bsd.prog.mk>

Copied and modified: head/secure/lib/libcrypto/i386/aes-586.S (from r299388, head/secure/lib/libcrypto/i386/aes-586.s)
==============================================================================
--- head/secure/lib/libcrypto/i386/aes-586.s	Tue May 10 20:25:49 2016	(r299388, copy source)
+++ head/secure/lib/libcrypto/i386/aes-586.S	Tue May 10 20:31:09 2016	(r299389)
@@ -1,5 +1,3251 @@
 	# $FreeBSD$
-.file	"aes-586.s"
+#ifdef PIC
+.file	"aes-586.S"
+.text
+.type	_x86_AES_encrypt_compact,@function
+.align	16
+_x86_AES_encrypt_compact:
+	movl	%edi,20(%esp)
+	xorl	(%edi),%eax
+	xorl	4(%edi),%ebx
+	xorl	8(%edi),%ecx
+	xorl	12(%edi),%edx
+	movl	240(%edi),%esi
+	leal	-2(%esi,%esi,1),%esi
+	leal	(%edi,%esi,8),%esi
+	movl	%esi,24(%esp)
+	movl	-128(%ebp),%edi
+	movl	-96(%ebp),%esi
+	movl	-64(%ebp),%edi
+	movl	-32(%ebp),%esi
+	movl	(%ebp),%edi
+	movl	32(%ebp),%esi
+	movl	64(%ebp),%edi
+	movl	96(%ebp),%esi
+.align	16
+.L000loop:
+	movl	%eax,%esi
+	andl	$255,%esi
+	movzbl	-128(%ebp,%esi,1),%esi
+	movzbl	%bh,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$8,%edi
+	xorl	%edi,%esi
+	movl	%ecx,%edi
+	shrl	$16,%edi
+	andl	$255,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$16,%edi
+	xorl	%edi,%esi
+	movl	%edx,%edi
+	shrl	$24,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$24,%edi
+	xorl	%edi,%esi
+	movl	%esi,4(%esp)
+
+	movl	%ebx,%esi
+	andl	$255,%esi
+	shrl	$16,%ebx
+	movzbl	-128(%ebp,%esi,1),%esi
+	movzbl	%ch,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$8,%edi
+	xorl	%edi,%esi
+	movl	%edx,%edi
+	shrl	$16,%edi
+	andl	$255,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$16,%edi
+	xorl	%edi,%esi
+	movl	%eax,%edi
+	shrl	$24,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$24,%edi
+	xorl	%edi,%esi
+	movl	%esi,8(%esp)
+
+	movl	%ecx,%esi
+	andl	$255,%esi
+	shrl	$24,%ecx
+	movzbl	-128(%ebp,%esi,1),%esi
+	movzbl	%dh,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$8,%edi
+	xorl	%edi,%esi
+	movl	%eax,%edi
+	shrl	$16,%edi
+	andl	$255,%edx
+	andl	$255,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$16,%edi
+	xorl	%edi,%esi
+	movzbl	%bh,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$24,%edi
+	xorl	%edi,%esi
+
+	andl	$255,%edx
+	movzbl	-128(%ebp,%edx,1),%edx
+	movzbl	%ah,%eax
+	movzbl	-128(%ebp,%eax,1),%eax
+	shll	$8,%eax
+	xorl	%eax,%edx
+	movl	4(%esp),%eax
+	andl	$255,%ebx
+	movzbl	-128(%ebp,%ebx,1),%ebx
+	shll	$16,%ebx
+	xorl	%ebx,%edx
+	movl	8(%esp),%ebx
+	movzbl	-128(%ebp,%ecx,1),%ecx
+	shll	$24,%ecx
+	xorl	%ecx,%edx
+	movl	%esi,%ecx
+
+	movl	$2155905152,%ebp
+	andl	%ecx,%ebp
+	leal	(%ecx,%ecx,1),%edi
+	movl	%ebp,%esi
+	shrl	$7,%ebp
+	andl	$4278124286,%edi
+	subl	%ebp,%esi
+	movl	%ecx,%ebp
+	andl	$454761243,%esi
+	rorl	$16,%ebp
+	xorl	%edi,%esi
+	movl	%ecx,%edi
+	xorl	%esi,%ecx
+	rorl	$24,%edi
+	xorl	%ebp,%esi
+	roll	$24,%ecx
+	xorl	%edi,%esi
+	movl	$2155905152,%ebp
+	xorl	%esi,%ecx
+	andl	%edx,%ebp
+	leal	(%edx,%edx,1),%edi
+	movl	%ebp,%esi
+	shrl	$7,%ebp
+	andl	$4278124286,%edi
+	subl	%ebp,%esi
+	movl	%edx,%ebp
+	andl	$454761243,%esi
+	rorl	$16,%ebp
+	xorl	%edi,%esi
+	movl	%edx,%edi
+	xorl	%esi,%edx
+	rorl	$24,%edi
+	xorl	%ebp,%esi
+	roll	$24,%edx
+	xorl	%edi,%esi
+	movl	$2155905152,%ebp
+	xorl	%esi,%edx
+	andl	%eax,%ebp
+	leal	(%eax,%eax,1),%edi
+	movl	%ebp,%esi
+	shrl	$7,%ebp
+	andl	$4278124286,%edi
+	subl	%ebp,%esi
+	movl	%eax,%ebp
+	andl	$454761243,%esi
+	rorl	$16,%ebp
+	xorl	%edi,%esi
+	movl	%eax,%edi
+	xorl	%esi,%eax
+	rorl	$24,%edi
+	xorl	%ebp,%esi
+	roll	$24,%eax
+	xorl	%edi,%esi
+	movl	$2155905152,%ebp
+	xorl	%esi,%eax
+	andl	%ebx,%ebp
+	leal	(%ebx,%ebx,1),%edi
+	movl	%ebp,%esi
+	shrl	$7,%ebp
+	andl	$4278124286,%edi
+	subl	%ebp,%esi
+	movl	%ebx,%ebp
+	andl	$454761243,%esi
+	rorl	$16,%ebp
+	xorl	%edi,%esi
+	movl	%ebx,%edi
+	xorl	%esi,%ebx
+	rorl	$24,%edi
+	xorl	%ebp,%esi
+	roll	$24,%ebx
+	xorl	%edi,%esi
+	xorl	%esi,%ebx
+	movl	20(%esp),%edi
+	movl	28(%esp),%ebp
+	addl	$16,%edi
+	xorl	(%edi),%eax
+	xorl	4(%edi),%ebx
+	xorl	8(%edi),%ecx
+	xorl	12(%edi),%edx
+	cmpl	24(%esp),%edi
+	movl	%edi,20(%esp)
+	jb	.L000loop
+	movl	%eax,%esi
+	andl	$255,%esi
+	movzbl	-128(%ebp,%esi,1),%esi
+	movzbl	%bh,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$8,%edi
+	xorl	%edi,%esi
+	movl	%ecx,%edi
+	shrl	$16,%edi
+	andl	$255,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$16,%edi
+	xorl	%edi,%esi
+	movl	%edx,%edi
+	shrl	$24,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$24,%edi
+	xorl	%edi,%esi
+	movl	%esi,4(%esp)
+
+	movl	%ebx,%esi
+	andl	$255,%esi
+	shrl	$16,%ebx
+	movzbl	-128(%ebp,%esi,1),%esi
+	movzbl	%ch,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$8,%edi
+	xorl	%edi,%esi
+	movl	%edx,%edi
+	shrl	$16,%edi
+	andl	$255,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$16,%edi
+	xorl	%edi,%esi
+	movl	%eax,%edi
+	shrl	$24,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$24,%edi
+	xorl	%edi,%esi
+	movl	%esi,8(%esp)
+
+	movl	%ecx,%esi
+	andl	$255,%esi
+	shrl	$24,%ecx
+	movzbl	-128(%ebp,%esi,1),%esi
+	movzbl	%dh,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$8,%edi
+	xorl	%edi,%esi
+	movl	%eax,%edi
+	shrl	$16,%edi
+	andl	$255,%edx
+	andl	$255,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$16,%edi
+	xorl	%edi,%esi
+	movzbl	%bh,%edi
+	movzbl	-128(%ebp,%edi,1),%edi
+	shll	$24,%edi
+	xorl	%edi,%esi
+
+	movl	20(%esp),%edi
+	andl	$255,%edx
+	movzbl	-128(%ebp,%edx,1),%edx
+	movzbl	%ah,%eax
+	movzbl	-128(%ebp,%eax,1),%eax
+	shll	$8,%eax
+	xorl	%eax,%edx
+	movl	4(%esp),%eax
+	andl	$255,%ebx
+	movzbl	-128(%ebp,%ebx,1),%ebx
+	shll	$16,%ebx
+	xorl	%ebx,%edx
+	movl	8(%esp),%ebx
+	movzbl	-128(%ebp,%ecx,1),%ecx
+	shll	$24,%ecx
+	xorl	%ecx,%edx
+	movl	%esi,%ecx
+
+	xorl	16(%edi),%eax
+	xorl	20(%edi),%ebx
+	xorl	24(%edi),%ecx
+	xorl	28(%edi),%edx
+	ret
+.size	_x86_AES_encrypt_compact,.-_x86_AES_encrypt_compact
+.type	_sse_AES_encrypt_compact,@function
+.align	16
+_sse_AES_encrypt_compact:
+	pxor	(%edi),%mm0
+	pxor	8(%edi),%mm4
+	movl	240(%edi),%esi
+	leal	-2(%esi,%esi,1),%esi
+	leal	(%edi,%esi,8),%esi
+	movl	%esi,24(%esp)
+	movl	$454761243,%eax
+	movl	%eax,8(%esp)
+	movl	%eax,12(%esp)
+	movl	-128(%ebp),%eax
+	movl	-96(%ebp),%ebx
+	movl	-64(%ebp),%ecx
+	movl	-32(%ebp),%edx
+	movl	(%ebp),%eax
+	movl	32(%ebp),%ebx
+	movl	64(%ebp),%ecx
+	movl	96(%ebp),%edx
+.align	16
+.L001loop:
+	pshufw	$8,%mm0,%mm1
+	pshufw	$13,%mm4,%mm5
+	movd	%mm1,%eax
+	movd	%mm5,%ebx
+	movl	%edi,20(%esp)
+	movzbl	%al,%esi
+	movzbl	%ah,%edx
+	pshufw	$13,%mm0,%mm2
+	movzbl	-128(%ebp,%esi,1),%ecx
+	movzbl	%bl,%edi
+	movzbl	-128(%ebp,%edx,1),%edx
+	shrl	$16,%eax
+	shll	$8,%edx
+	movzbl	-128(%ebp,%edi,1),%esi
+	movzbl	%bh,%edi
+	shll	$16,%esi
+	pshufw	$8,%mm4,%mm6
+	orl	%esi,%ecx
+	movzbl	-128(%ebp,%edi,1),%esi
+	movzbl	%ah,%edi
+	shll	$24,%esi
+	shrl	$16,%ebx
+	orl	%esi,%edx
+	movzbl	-128(%ebp,%edi,1),%esi
+	movzbl	%bh,%edi
+	shll	$8,%esi
+	orl	%esi,%ecx
+	movzbl	-128(%ebp,%edi,1),%esi
+	movzbl	%al,%edi
+	shll	$24,%esi
+	orl	%esi,%ecx
+	movzbl	-128(%ebp,%edi,1),%esi
+	movzbl	%bl,%edi
+	movd	%mm2,%eax
+	movd	%ecx,%mm0
+	movzbl	-128(%ebp,%edi,1),%ecx
+	movzbl	%ah,%edi
+	shll	$16,%ecx
+	movd	%mm6,%ebx
+	orl	%esi,%ecx
+	movzbl	-128(%ebp,%edi,1),%esi
+	movzbl	%bh,%edi
+	shll	$24,%esi
+	orl	%esi,%ecx
+	movzbl	-128(%ebp,%edi,1),%esi
+	movzbl	%bl,%edi
+	shll	$8,%esi
+	shrl	$16,%ebx
+	orl	%esi,%ecx
+	movzbl	-128(%ebp,%edi,1),%esi
+	movzbl	%al,%edi
+	shrl	$16,%eax
+	movd	%ecx,%mm1
+	movzbl	-128(%ebp,%edi,1),%ecx
+	movzbl	%ah,%edi
+	shll	$16,%ecx
+	andl	$255,%eax
+	orl	%esi,%ecx
+	punpckldq	%mm1,%mm0
+	movzbl	-128(%ebp,%edi,1),%esi
+	movzbl	%bh,%edi
+	shll	$24,%esi
+	andl	$255,%ebx
+	movzbl	-128(%ebp,%eax,1),%eax
+	orl	%esi,%ecx
+	shll	$16,%eax
+	movzbl	-128(%ebp,%edi,1),%esi
+	orl	%eax,%edx
+	shll	$8,%esi
+	movzbl	-128(%ebp,%ebx,1),%ebx
+	orl	%esi,%ecx
+	orl	%ebx,%edx
+	movl	20(%esp),%edi
+	movd	%ecx,%mm4
+	movd	%edx,%mm5
+	punpckldq	%mm5,%mm4
+	addl	$16,%edi
+	cmpl	24(%esp),%edi
+	ja	.L002out
+	movq	8(%esp),%mm2
+	pxor	%mm3,%mm3
+	pxor	%mm7,%mm7
+	movq	%mm0,%mm1
+	movq	%mm4,%mm5
+	pcmpgtb	%mm0,%mm3
+	pcmpgtb	%mm4,%mm7
+	pand	%mm2,%mm3
+	pand	%mm2,%mm7
+	pshufw	$177,%mm0,%mm2
+	pshufw	$177,%mm4,%mm6
+	paddb	%mm0,%mm0
+	paddb	%mm4,%mm4
+	pxor	%mm3,%mm0
+	pxor	%mm7,%mm4
+	pshufw	$177,%mm2,%mm3
+	pshufw	$177,%mm6,%mm7
+	pxor	%mm0,%mm1
+	pxor	%mm4,%mm5
+	pxor	%mm2,%mm0
+	pxor	%mm6,%mm4
+	movq	%mm3,%mm2
+	movq	%mm7,%mm6
+	pslld	$8,%mm3
+	pslld	$8,%mm7
+	psrld	$24,%mm2
+	psrld	$24,%mm6
+	pxor	%mm3,%mm0
+	pxor	%mm7,%mm4
+	pxor	%mm2,%mm0
+	pxor	%mm6,%mm4
+	movq	%mm1,%mm3
+	movq	%mm5,%mm7
+	movq	(%edi),%mm2
+	movq	8(%edi),%mm6
+	psrld	$8,%mm1
+	psrld	$8,%mm5
+	movl	-128(%ebp),%eax
+	pslld	$24,%mm3
+	pslld	$24,%mm7
+	movl	-64(%ebp),%ebx
+	pxor	%mm1,%mm0
+	pxor	%mm5,%mm4
+	movl	(%ebp),%ecx
+	pxor	%mm3,%mm0
+	pxor	%mm7,%mm4
+	movl	64(%ebp),%edx
+	pxor	%mm2,%mm0
+	pxor	%mm6,%mm4
+	jmp	.L001loop
+.align	16
+.L002out:
+	pxor	(%edi),%mm0
+	pxor	8(%edi),%mm4
+	ret
+.size	_sse_AES_encrypt_compact,.-_sse_AES_encrypt_compact
+.type	_x86_AES_encrypt,@function
+.align	16
+_x86_AES_encrypt:
+	movl	%edi,20(%esp)
+	xorl	(%edi),%eax
+	xorl	4(%edi),%ebx
+	xorl	8(%edi),%ecx
+	xorl	12(%edi),%edx
+	movl	240(%edi),%esi
+	leal	-2(%esi,%esi,1),%esi
+	leal	(%edi,%esi,8),%esi
+	movl	%esi,24(%esp)
+.align	16
+.L003loop:
+	movl	%eax,%esi
+	andl	$255,%esi
+	movl	(%ebp,%esi,8),%esi
+	movzbl	%bh,%edi
+	xorl	3(%ebp,%edi,8),%esi
+	movl	%ecx,%edi
+	shrl	$16,%edi
+	andl	$255,%edi
+	xorl	2(%ebp,%edi,8),%esi
+	movl	%edx,%edi
+	shrl	$24,%edi
+	xorl	1(%ebp,%edi,8),%esi
+	movl	%esi,4(%esp)
+
+	movl	%ebx,%esi
+	andl	$255,%esi
+	shrl	$16,%ebx
+	movl	(%ebp,%esi,8),%esi
+	movzbl	%ch,%edi
+	xorl	3(%ebp,%edi,8),%esi
+	movl	%edx,%edi
+	shrl	$16,%edi
+	andl	$255,%edi
+	xorl	2(%ebp,%edi,8),%esi
+	movl	%eax,%edi
+	shrl	$24,%edi
+	xorl	1(%ebp,%edi,8),%esi
+	movl	%esi,8(%esp)
+
+	movl	%ecx,%esi
+	andl	$255,%esi
+	shrl	$24,%ecx
+	movl	(%ebp,%esi,8),%esi
+	movzbl	%dh,%edi
+	xorl	3(%ebp,%edi,8),%esi
+	movl	%eax,%edi
+	shrl	$16,%edi
+	andl	$255,%edx
+	andl	$255,%edi
+	xorl	2(%ebp,%edi,8),%esi
+	movzbl	%bh,%edi
+	xorl	1(%ebp,%edi,8),%esi
+
+	movl	20(%esp),%edi
+	movl	(%ebp,%edx,8),%edx
+	movzbl	%ah,%eax
+	xorl	3(%ebp,%eax,8),%edx
+	movl	4(%esp),%eax
+	andl	$255,%ebx
+	xorl	2(%ebp,%ebx,8),%edx
+	movl	8(%esp),%ebx
+	xorl	1(%ebp,%ecx,8),%edx
+	movl	%esi,%ecx
+
+	addl	$16,%edi
+	xorl	(%edi),%eax
+	xorl	4(%edi),%ebx
+	xorl	8(%edi),%ecx
+	xorl	12(%edi),%edx
+	cmpl	24(%esp),%edi
+	movl	%edi,20(%esp)
+	jb	.L003loop
+	movl	%eax,%esi
+	andl	$255,%esi
+	movl	2(%ebp,%esi,8),%esi
+	andl	$255,%esi
+	movzbl	%bh,%edi
+	movl	(%ebp,%edi,8),%edi
+	andl	$65280,%edi
+	xorl	%edi,%esi
+	movl	%ecx,%edi
+	shrl	$16,%edi
+	andl	$255,%edi
+	movl	(%ebp,%edi,8),%edi
+	andl	$16711680,%edi
+	xorl	%edi,%esi
+	movl	%edx,%edi
+	shrl	$24,%edi
+	movl	2(%ebp,%edi,8),%edi
+	andl	$4278190080,%edi
+	xorl	%edi,%esi
+	movl	%esi,4(%esp)
+	movl	%ebx,%esi
+	andl	$255,%esi
+	shrl	$16,%ebx
+	movl	2(%ebp,%esi,8),%esi
+	andl	$255,%esi
+	movzbl	%ch,%edi
+	movl	(%ebp,%edi,8),%edi
+	andl	$65280,%edi
+	xorl	%edi,%esi
+	movl	%edx,%edi
+	shrl	$16,%edi
+	andl	$255,%edi
+	movl	(%ebp,%edi,8),%edi
+	andl	$16711680,%edi
+	xorl	%edi,%esi
+	movl	%eax,%edi
+	shrl	$24,%edi
+	movl	2(%ebp,%edi,8),%edi
+	andl	$4278190080,%edi
+	xorl	%edi,%esi
+	movl	%esi,8(%esp)
+	movl	%ecx,%esi
+	andl	$255,%esi
+	shrl	$24,%ecx
+	movl	2(%ebp,%esi,8),%esi
+	andl	$255,%esi
+	movzbl	%dh,%edi
+	movl	(%ebp,%edi,8),%edi
+	andl	$65280,%edi
+	xorl	%edi,%esi
+	movl	%eax,%edi
+	shrl	$16,%edi
+	andl	$255,%edx
+	andl	$255,%edi
+	movl	(%ebp,%edi,8),%edi
+	andl	$16711680,%edi
+	xorl	%edi,%esi
+	movzbl	%bh,%edi
+	movl	2(%ebp,%edi,8),%edi
+	andl	$4278190080,%edi
+	xorl	%edi,%esi
+	movl	20(%esp),%edi
+	andl	$255,%edx
+	movl	2(%ebp,%edx,8),%edx
+	andl	$255,%edx
+	movzbl	%ah,%eax
+	movl	(%ebp,%eax,8),%eax
+	andl	$65280,%eax
+	xorl	%eax,%edx
+	movl	4(%esp),%eax
+	andl	$255,%ebx
+	movl	(%ebp,%ebx,8),%ebx
+	andl	$16711680,%ebx
+	xorl	%ebx,%edx
+	movl	8(%esp),%ebx
+	movl	2(%ebp,%ecx,8),%ecx
+	andl	$4278190080,%ecx
+	xorl	%ecx,%edx
+	movl	%esi,%ecx
+	addl	$16,%edi
+	xorl	(%edi),%eax
+	xorl	4(%edi),%ebx
+	xorl	8(%edi),%ecx
+	xorl	12(%edi),%edx
+	ret
+.align	64
+.LAES_Te:
+.long	2774754246,2774754246
+.long	2222750968,2222750968
+.long	2574743534,2574743534
+.long	2373680118,2373680118
+.long	234025727,234025727
+.long	3177933782,3177933782
+.long	2976870366,2976870366
+.long	1422247313,1422247313
+.long	1345335392,1345335392
+.long	50397442,50397442
+.long	2842126286,2842126286
+.long	2099981142,2099981142
+.long	436141799,436141799
+.long	1658312629,1658312629
+.long	3870010189,3870010189
+.long	2591454956,2591454956
+.long	1170918031,1170918031
+.long	2642575903,2642575903
+.long	1086966153,1086966153
+.long	2273148410,2273148410
+.long	368769775,368769775
+.long	3948501426,3948501426
+.long	3376891790,3376891790
+.long	200339707,200339707
+.long	3970805057,3970805057
+.long	1742001331,1742001331
+.long	4255294047,4255294047
+.long	3937382213,3937382213
+.long	3214711843,3214711843
+.long	4154762323,4154762323
+.long	2524082916,2524082916
+.long	1539358875,1539358875
+.long	3266819957,3266819957
+.long	486407649,486407649
+.long	2928907069,2928907069
+.long	1780885068,1780885068
+.long	1513502316,1513502316
+.long	1094664062,1094664062
+.long	49805301,49805301
+.long	1338821763,1338821763
+.long	1546925160,1546925160
+.long	4104496465,4104496465
+.long	887481809,887481809
+.long	150073849,150073849
+.long	2473685474,2473685474
+.long	1943591083,1943591083
+.long	1395732834,1395732834
+.long	1058346282,1058346282
+.long	201589768,201589768
+.long	1388824469,1388824469
+.long	1696801606,1696801606
+.long	1589887901,1589887901
+.long	672667696,672667696
+.long	2711000631,2711000631
+.long	251987210,251987210
+.long	3046808111,3046808111
+.long	151455502,151455502
+.long	907153956,907153956
+.long	2608889883,2608889883
+.long	1038279391,1038279391
+.long	652995533,652995533
+.long	1764173646,1764173646
+.long	3451040383,3451040383
+.long	2675275242,2675275242
+.long	453576978,453576978
+.long	2659418909,2659418909
+.long	1949051992,1949051992
+.long	773462580,773462580
+.long	756751158,756751158
+.long	2993581788,2993581788
+.long	3998898868,3998898868
+.long	4221608027,4221608027
+.long	4132590244,4132590244
+.long	1295727478,1295727478
+.long	1641469623,1641469623
+.long	3467883389,3467883389
+.long	2066295122,2066295122
+.long	1055122397,1055122397
+.long	1898917726,1898917726
+.long	2542044179,2542044179
+.long	4115878822,4115878822
+.long	1758581177,1758581177
+.long	0,0
+.long	753790401,753790401
+.long	1612718144,1612718144
+.long	536673507,536673507
+.long	3367088505,3367088505
+.long	3982187446,3982187446
+.long	3194645204,3194645204
+.long	1187761037,1187761037
+.long	3653156455,3653156455
+.long	1262041458,1262041458
+.long	3729410708,3729410708
+.long	3561770136,3561770136
+.long	3898103984,3898103984
+.long	1255133061,1255133061
+.long	1808847035,1808847035
+.long	720367557,720367557
+.long	3853167183,3853167183
+.long	385612781,385612781
+.long	3309519750,3309519750
+.long	3612167578,3612167578
+.long	1429418854,1429418854
+.long	2491778321,2491778321
+.long	3477423498,3477423498
+.long	284817897,284817897
+.long	100794884,100794884
+.long	2172616702,2172616702
+.long	4031795360,4031795360
+.long	1144798328,1144798328
+.long	3131023141,3131023141
+.long	3819481163,3819481163
+.long	4082192802,4082192802
+.long	4272137053,4272137053
+.long	3225436288,3225436288
+.long	2324664069,2324664069
+.long	2912064063,2912064063
+.long	3164445985,3164445985
+.long	1211644016,1211644016
+.long	83228145,83228145
+.long	3753688163,3753688163
+.long	3249976951,3249976951
+.long	1977277103,1977277103
+.long	1663115586,1663115586
+.long	806359072,806359072
+.long	452984805,452984805
+.long	250868733,250868733
+.long	1842533055,1842533055
+.long	1288555905,1288555905
+.long	336333848,336333848
+.long	890442534,890442534
+.long	804056259,804056259
+.long	3781124030,3781124030
+.long	2727843637,2727843637
+.long	3427026056,3427026056
+.long	957814574,957814574
+.long	1472513171,1472513171
+.long	4071073621,4071073621
+.long	2189328124,2189328124
+.long	1195195770,1195195770
+.long	2892260552,2892260552
+.long	3881655738,3881655738
+.long	723065138,723065138
+.long	2507371494,2507371494
+.long	2690670784,2690670784
+.long	2558624025,2558624025
+.long	3511635870,3511635870
+.long	2145180835,2145180835
+.long	1713513028,1713513028
+.long	2116692564,2116692564
+.long	2878378043,2878378043
+.long	2206763019,2206763019
+.long	3393603212,3393603212
+.long	703524551,703524551
+.long	3552098411,3552098411
+.long	1007948840,1007948840
+.long	2044649127,2044649127
+.long	3797835452,3797835452
+.long	487262998,487262998
+.long	1994120109,1994120109
+.long	1004593371,1004593371
+.long	1446130276,1446130276
+.long	1312438900,1312438900
+.long	503974420,503974420
+.long	3679013266,3679013266
+.long	168166924,168166924
+.long	1814307912,1814307912
+.long	3831258296,3831258296
+.long	1573044895,1573044895
+.long	1859376061,1859376061
+.long	4021070915,4021070915
+.long	2791465668,2791465668
+.long	2828112185,2828112185
+.long	2761266481,2761266481
+.long	937747667,937747667
+.long	2339994098,2339994098
+.long	854058965,854058965
+.long	1137232011,1137232011
+.long	1496790894,1496790894
+.long	3077402074,3077402074
+.long	2358086913,2358086913
+.long	1691735473,1691735473
+.long	3528347292,3528347292
+.long	3769215305,3769215305
+.long	3027004632,3027004632
+.long	4199962284,4199962284
+.long	133494003,133494003
+.long	636152527,636152527
+.long	2942657994,2942657994
+.long	2390391540,2390391540
+.long	3920539207,3920539207
+.long	403179536,403179536
+.long	3585784431,3585784431
+.long	2289596656,2289596656
+.long	1864705354,1864705354
+.long	1915629148,1915629148
+.long	605822008,605822008
+.long	4054230615,4054230615
+.long	3350508659,3350508659
+.long	1371981463,1371981463
+.long	602466507,602466507
+.long	2094914977,2094914977
+.long	2624877800,2624877800
+.long	555687742,555687742
+.long	3712699286,3712699286
+.long	3703422305,3703422305
+.long	2257292045,2257292045
+.long	2240449039,2240449039
+.long	2423288032,2423288032
+.long	1111375484,1111375484
+.long	3300242801,3300242801
+.long	2858837708,2858837708
+.long	3628615824,3628615824
+.long	84083462,84083462
+.long	32962295,32962295
+.long	302911004,302911004
+.long	2741068226,2741068226
+.long	1597322602,1597322602
+.long	4183250862,4183250862
+.long	3501832553,3501832553
+.long	2441512471,2441512471
+.long	1489093017,1489093017
+.long	656219450,656219450
+.long	3114180135,3114180135
+.long	954327513,954327513
+.long	335083755,335083755
+.long	3013122091,3013122091
+.long	856756514,856756514
+.long	3144247762,3144247762
+.long	1893325225,1893325225
+.long	2307821063,2307821063
+.long	2811532339,2811532339
+.long	3063651117,3063651117

*** DIFF OUTPUT TRUNCATED AT 1000 LINES ***



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201605102031.u4AKV9Ys060380>