From 8ca28da0a798d2289b71b750dad191066600c166 Mon Sep 17 00:00:00 2001 From: Andy Polyakov Date: Thu, 15 Sep 2011 20:22:59 +0000 Subject: [PATCH] Integrate Vector Permutation AES into build system. --- Configure | 11 +- TABLE | 88 ++++++++-------- crypto/aes/Makefile | 4 + crypto/aes/asm/vpaes-x86.pl | 7 +- crypto/aes/asm/vpaes-x86_64.pl | 7 +- crypto/evp/e_aes.c | 185 +++++++++++++++++++++++++-------- 6 files changed, 206 insertions(+), 96 deletions(-) diff --git a/Configure b/Configure index 1a80afaf03..d8c93e93ed 100755 --- a/Configure +++ b/Configure @@ -123,11 +123,11 @@ my $tlib="-lnsl -lsocket"; my $bits1="THIRTY_TWO_BIT "; my $bits2="SIXTY_FOUR_BIT "; -my $x86_asm="x86cpuid.o:bn-586.o co-586.o x86-mont.o x86-gf2m.o:des-586.o crypt586.o:aes-586.o aesni-x86.o:bf-586.o:md5-586.o:sha1-586.o sha256-586.o sha512-586.o:cast-586.o:rc4-586.o:rmd-586.o:rc5-586.o:wp_block.o wp-mmx.o:cmll-x86.o:ghash-x86.o:e_padlock-x86.o"; +my $x86_asm="x86cpuid.o:bn-586.o co-586.o x86-mont.o x86-gf2m.o:des-586.o crypt586.o:aes-586.o vpaes-x86.o aesni-x86.o:bf-586.o:md5-586.o:sha1-586.o sha256-586.o sha512-586.o:cast-586.o:rc4-586.o:rmd-586.o:rc5-586.o:wp_block.o wp-mmx.o:cmll-x86.o:ghash-x86.o:e_padlock-x86.o"; my $x86_elf_asm="$x86_asm:elf"; -my $x86_64_asm="x86_64cpuid.o:x86_64-gcc.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o::aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o::md5-x86_64.o:sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o::rc4-x86_64.o rc4-md5-x86_64.o:::wp-x86_64.o:cmll-x86_64.o cmll_misc.o:ghash-x86_64.o:e_padlock-x86_64.o"; +my $x86_64_asm="x86_64cpuid.o:x86_64-gcc.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o::aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o::md5-x86_64.o:sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o::rc4-x86_64.o rc4-md5-x86_64.o:::wp-x86_64.o:cmll-x86_64.o cmll_misc.o:ghash-x86_64.o:e_padlock-x86_64.o"; my $ia64_asm="ia64cpuid.o:bn-ia64.o ia64-mont.o::aes_core.o aes_cbc.o aes-ia64.o::md5-ia64.o:sha1-ia64.o sha256-ia64.o sha512-ia64.o::rc4-ia64.o rc4_skey.o:::::ghash-ia64.o::void"; my $sparcv9_asm="sparcv9cap.o sparccpuid.o:bn-sparcv9.o sparcv9-mont.o sparcv9a-mont.o:des_enc-sparc.o fcrypt_b.o:aes_core.o aes_cbc.o aes-sparcv9.o:::sha1-sparcv9.o sha256-sparcv9.o sha512-sparcv9.o:::::::ghash-sparcv9.o::void"; my $sparcv8_asm=":sparcv8.o:des_enc-sparc.o fcrypt_b.o:::::::::::::void"; @@ -369,8 +369,8 @@ my %table=( # alongside generic one, e.g. as /lib/highgprs/libcrypto.so.x.y, for # ldconfig and run-time linker to autodiscover. Unfortunately it # doesn't work just yet, because of couple of bugs in glibc -# sysdep/s390/dl-procinfo.c affecting ldconfig and ld.so.1... -"linux32-s390x", "gcc:-m31 -Wa,-mzarch -DB_ENDIAN -DTERMIO -O3 -Wall::-D_REENTRANT::-ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL:".($s390x_asm=~s/bn\-s390x\.o/bn_asm.o/,$s390x_asm).":31:dlfcn:linux-shared:-fPIC:-m31:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR):::/highgprs", +# sysdeps/s390/dl-procinfo.c affecting ldconfig and ld.so.1... +"linux32-s390x", "gcc:-m31 -Wa,-mzarch -DB_ENDIAN -DTERMIO -O3 -Wall::-D_REENTRANT::-ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL:".eval{my $asm=$s390x_asm;$asm=~s/bn\-s390x\.o/bn_asm.o/;$asm}.":31:dlfcn:linux-shared:-fPIC:-m31:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR):::/highgprs", #### SPARC Linux setups # Ray Miller has patiently # assisted with debugging of following two configs. @@ -1551,7 +1551,8 @@ if ($aes_obj =~ /\.o$/) # aes_ctr.o is not a real file, only indication that assembler # module implements AES_ctr32_encrypt... $cflags.=" -DAES_CTR_ASM" if ($aes_obj =~ s/\s*aes_ctr\.o//); - $aes_obj =~ s/\s*aesni\-x86\.o// if ($no_sse2); + $aes_obj =~ s/\s*(vpaes|aesni)\-x86\.o// if ($no_sse2); + $cflags.=" -DVPAES_ASM" if ($aes_obj =~ m/vpaes/); } else { $aes_obj=$aes_enc; diff --git a/TABLE b/TABLE index dfc7739105..4292c1092c 100644 --- a/TABLE +++ b/TABLE @@ -242,7 +242,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -275,7 +275,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -308,7 +308,7 @@ $bn_ops = SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_INT DES_UNROLL $cpuid_obj = x86_64cpuid.o $bn_obj = x86_64-gcc.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o $des_obj = -$aes_obj = aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o +$aes_obj = aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o $bf_obj = $md5_obj = md5-x86_64.o $sha1_obj = sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o @@ -341,7 +341,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -407,7 +407,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -572,7 +572,7 @@ $bn_ops = DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -770,7 +770,7 @@ $bn_ops = BN_LLONG RC4_INDEX EXPORT_VAR_AS_FN RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -803,7 +803,7 @@ $bn_ops = SIXTY_FOUR_BIT RC4_CHUNK_LL DES_INT EXPORT_VAR_AS_FN $cpuid_obj = x86_64cpuid.o $bn_obj = bn_asm.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o $des_obj = -$aes_obj = aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o +$aes_obj = aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o $bf_obj = $md5_obj = md5-x86_64.o $sha1_obj = sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o @@ -1133,7 +1133,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -1166,7 +1166,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -1199,7 +1199,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -1331,7 +1331,7 @@ $bn_ops = BN_LLONG RC4_INT RC4_CHUNK DES_UNROLL BF_PTR $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -1430,7 +1430,7 @@ $bn_ops = SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL $cpuid_obj = x86_64cpuid.o $bn_obj = x86_64-gcc.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o $des_obj = -$aes_obj = aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o +$aes_obj = aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o $bf_obj = $md5_obj = md5-x86_64.o $sha1_obj = sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o @@ -1496,7 +1496,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -1562,7 +1562,7 @@ $bn_ops = BN_LLONG RC4_INDEX EXPORT_VAR_AS_FN RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -1595,7 +1595,7 @@ $bn_ops = SIXTY_FOUR_BIT RC4_CHUNK_LL DES_INT EXPORT_VAR_AS_FN $cpuid_obj = x86_64cpuid.o $bn_obj = bn_asm.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o $des_obj = -$aes_obj = aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o +$aes_obj = aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o $bf_obj = $md5_obj = md5-x86_64.o $sha1_obj = sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o @@ -1859,7 +1859,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -1892,7 +1892,7 @@ $bn_ops = BN_LLONG RC4_INT RC4_CHUNK DES_UNROLL BF_PTR $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -2024,7 +2024,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -2057,7 +2057,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -2156,7 +2156,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -2189,7 +2189,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -2321,7 +2321,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -2354,7 +2354,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -2387,7 +2387,7 @@ $bn_ops = SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_INT DES_UNROLL $cpuid_obj = x86_64cpuid.o $bn_obj = x86_64-gcc.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o $des_obj = -$aes_obj = aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o +$aes_obj = aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o $bf_obj = $md5_obj = md5-x86_64.o $sha1_obj = sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o @@ -2420,7 +2420,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -2585,7 +2585,7 @@ $bn_ops = SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_INT DES_UNROLL $cpuid_obj = x86_64cpuid.o $bn_obj = x86_64-gcc.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o $des_obj = -$aes_obj = aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o +$aes_obj = aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o $bf_obj = $md5_obj = md5-x86_64.o $sha1_obj = sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o @@ -2618,7 +2618,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -2651,7 +2651,7 @@ $bn_ops = SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_INT DES_UNROLL $cpuid_obj = x86_64cpuid.o $bn_obj = x86_64-gcc.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o $des_obj = -$aes_obj = aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o +$aes_obj = aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o $bf_obj = $md5_obj = md5-x86_64.o $sha1_obj = sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o @@ -2816,7 +2816,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -3410,7 +3410,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -3773,7 +3773,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -3839,7 +3839,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -3938,7 +3938,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -4202,7 +4202,7 @@ $bn_ops = SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_INT DES_UNROLL $cpuid_obj = x86_64cpuid.o $bn_obj = x86_64-gcc.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o $des_obj = -$aes_obj = aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o +$aes_obj = aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o $bf_obj = $md5_obj = md5-x86_64.o $sha1_obj = sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o @@ -4334,7 +4334,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT EXPORT_V $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -4367,7 +4367,7 @@ $bn_ops = SIXTY_FOUR_BIT RC4_CHUNK_LL DES_INT EXPORT_VAR_AS_FN $cpuid_obj = x86_64cpuid.o $bn_obj = x86_64-gcc.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o $des_obj = -$aes_obj = aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o +$aes_obj = aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o $bf_obj = $md5_obj = md5-x86_64.o $sha1_obj = sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o @@ -4961,7 +4961,7 @@ $bn_ops = DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -4994,7 +4994,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -5258,7 +5258,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -5357,7 +5357,7 @@ $bn_ops = SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_INT DES_UNROLL $cpuid_obj = x86_64cpuid.o $bn_obj = x86_64-gcc.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o $des_obj = -$aes_obj = aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o +$aes_obj = aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o $bf_obj = $md5_obj = md5-x86_64.o $sha1_obj = sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o @@ -5390,7 +5390,7 @@ $bn_ops = SIXTY_FOUR_BIT_LONG RC4_CHUNK DES_INT DES_UNROLL $cpuid_obj = x86_64cpuid.o $bn_obj = x86_64-gcc.o x86_64-mont.o x86_64-mont5.o x86_64-gf2m.o modexp512-x86_64.o $des_obj = -$aes_obj = aes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o +$aes_obj = aes-x86_64.o vpaes-x86_64.o aesni-x86_64.o aesni-sha1-x86_64.o $bf_obj = $md5_obj = md5-x86_64.o $sha1_obj = sha1-x86_64.o sha256-x86_64.o sha512-x86_64.o @@ -5720,7 +5720,7 @@ $bn_ops = BN_LLONG MD2_CHAR RC4_INDEX DES_PTR DES_RISC1 DES_UNROLL $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o @@ -5753,7 +5753,7 @@ $bn_ops = BN_LLONG DES_PTR DES_RISC1 DES_UNROLL RC4_INDEX MD2_INT $cpuid_obj = x86cpuid.o $bn_obj = bn-586.o co-586.o x86-mont.o x86-gf2m.o $des_obj = des-586.o crypt586.o -$aes_obj = aes-586.o aesni-x86.o +$aes_obj = aes-586.o vpaes-x86.o aesni-x86.o $bf_obj = bf-586.o $md5_obj = md5-586.o $sha1_obj = sha1-586.o sha256-586.o sha512-586.o diff --git a/crypto/aes/Makefile b/crypto/aes/Makefile index ae16e659e4..78d5984140 100644 --- a/crypto/aes/Makefile +++ b/crypto/aes/Makefile @@ -50,11 +50,15 @@ aes-ia64.s: asm/aes-ia64.S aes-586.s: asm/aes-586.pl ../perlasm/x86asm.pl $(PERL) asm/aes-586.pl $(PERLASM_SCHEME) $(CFLAGS) $(PROCESSOR) > $@ +vpaes-x86.s: asm/vpaes-x86.pl ../perlasm/x86asm.pl + $(PERL) asm/vpaes-x86.pl $(PERLASM_SCHEME) $(CFLAGS) $(PROCESSOR) > $@ aesni-x86.s: asm/aesni-x86.pl ../perlasm/x86asm.pl $(PERL) asm/aesni-x86.pl $(PERLASM_SCHEME) $(CFLAGS) $(PROCESSOR) > $@ aes-x86_64.s: asm/aes-x86_64.pl $(PERL) asm/aes-x86_64.pl $(PERLASM_SCHEME) > $@ +vpaes-x86_64.s: asm/vpaes-x86_64.pl + $(PERL) asm/vpaes-x86_64.pl $(PERLASM_SCHEME) > $@ aesni-x86_64.s: asm/aesni-x86_64.pl $(PERL) asm/aesni-x86_64.pl $(PERLASM_SCHEME) > $@ aesni-sha1-x86_64.s: asm/aesni-sha1-x86_64.pl diff --git a/crypto/aes/asm/vpaes-x86.pl b/crypto/aes/asm/vpaes-x86.pl index efe68dff41..f2414ccd23 100644 --- a/crypto/aes/asm/vpaes-x86.pl +++ b/crypto/aes/asm/vpaes-x86.pl @@ -21,7 +21,7 @@ # about its alignment... # # Performance summary. aes-586.pl column lists large-block CBC -# encrypt/decrypt/with-hypert-hreading-off(*) results in cycles per +# encrypt/decrypt/with-hyper-threading-off(*) results in cycles per # byte processed with 128-bit key, and vpaes-x86.pl column - # encrypt/decrypt. # @@ -41,7 +41,8 @@ # # (***) Less impressive improvement on Core 2 and Atom is due to slow # pshufb, yet it's respectable +32%/65% improvement on Core 2 -# and +58%/40% on Atom. +# and +58%/40% on Atom (as implied, over "hyper-threading-safe" +# code path). # # @@ -51,7 +52,7 @@ require "x86asm.pl"; &asm_init($ARGV[0],"vpaes-x86.pl",$x86only = $ARGV[$#ARGV] eq "386"); -$PREFIX="AES"; +$PREFIX="vpaes"; my ($round, $base, $magic, $key, $const, $inp, $out)= ("eax", "ebx", "ecx", "edx","ebp", "esi","edi"); diff --git a/crypto/aes/asm/vpaes-x86_64.pl b/crypto/aes/asm/vpaes-x86_64.pl index 3ea85d23e8..01011defbd 100644 --- a/crypto/aes/asm/vpaes-x86_64.pl +++ b/crypto/aes/asm/vpaes-x86_64.pl @@ -21,7 +21,7 @@ # about its alignment... # # Performance summary. aes-x86_64.pl column lists large-block CBC -# encrypt/decrypt/with-hypert-hreading-off(*) results in cycles per +# encrypt/decrypt/with-hyper-threading-off(*) results in cycles per # byte processed with 128-bit key, and vpaes-x86_64.pl column - # encrypt/decrypt. # @@ -40,7 +40,8 @@ # (**) "Core 2" refers to initial 65nm design, a.k.a. Conroe. # # (***) Less impressive improvement on Core 2 and Atom is due to slow -# pshufb, yet it's respectable +40%/78% improvement on Core 2. +# pshufb, yet it's respectable +40%/78% improvement on Core 2 +# (as implied, over "hyper-threading-safe" code path). # # @@ -57,7 +58,7 @@ die "can't locate x86_64-xlate.pl"; open STDOUT,"| $^X $xlate $flavour $output"; -$PREFIX="AES"; +$PREFIX="vpaes"; $code.=<<___; .text diff --git a/crypto/evp/e_aes.c b/crypto/evp/e_aes.c index 64a2f49dd3..558f1a197f 100644 --- a/crypto/evp/e_aes.c +++ b/crypto/evp/e_aes.c @@ -64,6 +64,14 @@ typedef struct { AES_KEY ks; + void (*block)(const unsigned char *in, unsigned char *out, + const AES_KEY *key); + void (*cbc)(const unsigned char *in, + unsigned char *out, + size_t length, + const AES_KEY *key, + unsigned char *ivec, int enc); + } EVP_AES_KEY; typedef struct @@ -98,17 +106,40 @@ typedef struct #define MAXBITCHUNK ((size_t)1<<(sizeof(size_t)*8-4)) +#ifdef VPAES_ASM +int vpaes_set_encrypt_key(const unsigned char *userKey, int bits, + AES_KEY *key); +int vpaes_set_decrypt_key(const unsigned char *userKey, int bits, + AES_KEY *key); + +void vpaes_encrypt(const unsigned char *in, unsigned char *out, + const AES_KEY *key); +void vpaes_decrypt(const unsigned char *in, unsigned char *out, + const AES_KEY *key); + +void vpaes_cbc_encrypt(const unsigned char *in, + unsigned char *out, + size_t length, + const AES_KEY *key, + unsigned char *ivec, int enc); +#endif + #if defined(AES_ASM) && !defined(I386_ONLY) && ( \ ((defined(__i386) || defined(__i386__) || \ defined(_M_IX86)) && defined(OPENSSL_IA32_SSE2))|| \ defined(__x86_64) || defined(__x86_64__) || \ defined(_M_AMD64) || defined(_M_X64) || \ defined(__INTEL__) ) + +extern unsigned int OPENSSL_ia32cap_P[2]; + +#ifdef VPAES_ASM +#define VPAES_CAPABLE (OPENSSL_ia32cap_P[1]&(1<<(41-32))) +#endif /* * AES-NI section */ -extern unsigned int OPENSSL_ia32cap_P[2]; -#define AESNI_CAPABLE (1<<(57-32)) +#define AESNI_CAPABLE (OPENSSL_ia32cap_P[1]&(1<<(57-32))) int aesni_set_encrypt_key(const unsigned char *userKey, int bits, AES_KEY *key); @@ -594,8 +625,7 @@ static const EVP_CIPHER aes_##keylen##_##mode = { \ sizeof(EVP_AES_KEY), \ NULL,NULL,NULL,NULL }; \ const EVP_CIPHER *EVP_aes_##keylen##_##mode(void) \ -{ return (OPENSSL_ia32cap_P[1]&AESNI_CAPABLE)? \ - &aesni_##keylen##_##mode:&aes_##keylen##_##mode; } +{ return AESNI_CAPABLE?&aesni_##keylen##_##mode:&aes_##keylen##_##mode; } #define BLOCK_CIPHER_custom(nid,keylen,blocksize,ivlen,mode,MODE,flags) \ static const EVP_CIPHER aesni_##keylen##_##mode = { \ @@ -617,8 +647,7 @@ static const EVP_CIPHER aes_##keylen##_##mode = { \ sizeof(EVP_AES_##MODE##_CTX), \ NULL,NULL,aes_##mode##_ctrl,NULL }; \ const EVP_CIPHER *EVP_aes_##keylen##_##mode(void) \ -{ return (OPENSSL_ia32cap_P[1]&AESNI_CAPABLE)? \ - &aesni_##keylen##_##mode:&aes_##keylen##_##mode; } +{ return AESNI_CAPABLE?&aesni_##keylen##_##mode:&aes_##keylen##_##mode; } #else @@ -661,13 +690,40 @@ static int aes_init_key(EVP_CIPHER_CTX *ctx, const unsigned char *key, const unsigned char *iv, int enc) { int ret; + EVP_AES_KEY *dat = (EVP_AES_KEY *)ctx->cipher_data; if (((ctx->cipher->flags & EVP_CIPH_MODE) == EVP_CIPH_ECB_MODE || (ctx->cipher->flags & EVP_CIPH_MODE) == EVP_CIPH_CBC_MODE) - && !enc) - ret = AES_set_decrypt_key(key, ctx->key_len * 8, ctx->cipher_data); + && !enc) +#ifdef VPAES_CAPABLE + if (VPAES_CAPABLE) + { + ret = vpaes_set_decrypt_key(key,ctx->key_len*8,&dat->ks); + dat->block = vpaes_decrypt; + dat->cbc = vpaes_cbc_encrypt; + } + else +#endif + { + ret = AES_set_decrypt_key(key,ctx->key_len*8,&dat->ks); + dat->block = AES_decrypt; + dat->cbc = AES_cbc_encrypt; + } else - ret = AES_set_encrypt_key(key, ctx->key_len * 8, ctx->cipher_data); +#ifdef VPAES_CAPABLE + if (VPAES_CAPABLE) + { + ret = vpaes_set_encrypt_key(key,ctx->key_len*8,&dat->ks); + dat->block = vpaes_encrypt; + dat->cbc = vpaes_cbc_encrypt; + } + else +#endif + { + ret = AES_set_encrypt_key(key,ctx->key_len*8,&dat->ks); + dat->block = AES_encrypt; + dat->cbc = AES_cbc_encrypt; + } if(ret < 0) { @@ -681,7 +737,9 @@ static int aes_init_key(EVP_CIPHER_CTX *ctx, const unsigned char *key, static int aes_cbc_cipher(EVP_CIPHER_CTX *ctx,unsigned char *out, const unsigned char *in, size_t len) { - AES_cbc_encrypt(in,out,len,ctx->cipher_data,ctx->iv,ctx->encrypt); + EVP_AES_KEY *dat = (EVP_AES_KEY *)ctx->cipher_data; + + (*dat->cbc)(in,out,len,&dat->ks,ctx->iv,ctx->encrypt); return 1; } @@ -691,16 +749,12 @@ static int aes_ecb_cipher(EVP_CIPHER_CTX *ctx,unsigned char *out, { size_t bl = ctx->cipher->block_size; size_t i; + EVP_AES_KEY *dat = (EVP_AES_KEY *)ctx->cipher_data; if (lenencrypt) { - for (i=0,len-=bl;i<=len;i+=bl) - AES_encrypt(in+i,out+i,ctx->cipher_data); - } else { - for (i=0,len-=bl;i<=len;i+=bl) - AES_decrypt(in+i,out+i,ctx->cipher_data); - } + for (i=0,len-=bl;i<=len;i+=bl) + (*dat->block)(in+i,out+i,&dat->ks); return 1; } @@ -708,50 +762,58 @@ static int aes_ecb_cipher(EVP_CIPHER_CTX *ctx,unsigned char *out, static int aes_ofb_cipher(EVP_CIPHER_CTX *ctx,unsigned char *out, const unsigned char *in,size_t len) { - CRYPTO_ofb128_encrypt(in,out,len,ctx->cipher_data, + EVP_AES_KEY *dat = (EVP_AES_KEY *)ctx->cipher_data; + + CRYPTO_ofb128_encrypt(in,out,len,&dat->ks, ctx->iv,&ctx->num, - (block128_f)AES_encrypt); + (block128_f)dat->block); return 1; } static int aes_cfb_cipher(EVP_CIPHER_CTX *ctx,unsigned char *out, const unsigned char *in,size_t len) { - CRYPTO_cfb128_encrypt(in,out,len,ctx->cipher_data, + EVP_AES_KEY *dat = (EVP_AES_KEY *)ctx->cipher_data; + + CRYPTO_cfb128_encrypt(in,out,len,&dat->ks, ctx->iv,&ctx->num,ctx->encrypt, - (block128_f)AES_encrypt); + (block128_f)dat->block); return 1; } static int aes_cfb8_cipher(EVP_CIPHER_CTX *ctx,unsigned char *out, const unsigned char *in,size_t len) { - CRYPTO_cfb128_8_encrypt(in,out,len,ctx->cipher_data, + EVP_AES_KEY *dat = (EVP_AES_KEY *)ctx->cipher_data; + + CRYPTO_cfb128_8_encrypt(in,out,len,&dat->ks, ctx->iv,&ctx->num,ctx->encrypt, - (block128_f)AES_encrypt); + (block128_f)dat->block); return 1; } static int aes_cfb1_cipher(EVP_CIPHER_CTX *ctx,unsigned char *out, const unsigned char *in,size_t len) { + EVP_AES_KEY *dat = (EVP_AES_KEY *)ctx->cipher_data; + if (ctx->flags&EVP_CIPH_FLAG_LENGTH_BITS) { - CRYPTO_cfb128_1_encrypt(in,out,len,ctx->cipher_data, + CRYPTO_cfb128_1_encrypt(in,out,len,&dat->ks, ctx->iv,&ctx->num,ctx->encrypt, - (block128_f)AES_encrypt); + (block128_f)dat->block); return 1; } while (len>=MAXBITCHUNK) { - CRYPTO_cfb128_1_encrypt(in,out,MAXBITCHUNK*8,ctx->cipher_data, + CRYPTO_cfb128_1_encrypt(in,out,MAXBITCHUNK*8,&dat->ks, ctx->iv,&ctx->num,ctx->encrypt, - (block128_f)AES_encrypt); + (block128_f)dat->block); len-=MAXBITCHUNK; } if (len) - CRYPTO_cfb128_1_encrypt(in,out,len*8,ctx->cipher_data, + CRYPTO_cfb128_1_encrypt(in,out,len*8,&dat->ks, ctx->iv,&ctx->num,ctx->encrypt, - (block128_f)AES_encrypt); + (block128_f)dat->block); return 1; } @@ -759,20 +821,18 @@ static int aes_cfb1_cipher(EVP_CIPHER_CTX *ctx,unsigned char *out, static int aes_ctr_cipher (EVP_CIPHER_CTX *ctx, unsigned char *out, const unsigned char *in, size_t len) { - unsigned int num; - num = ctx->num; + unsigned int num = ctx->num; + EVP_AES_KEY *dat = (EVP_AES_KEY *)ctx->cipher_data; #ifdef AES_CTR_ASM void AES_ctr32_encrypt(const unsigned char *in, unsigned char *out, size_t blocks, const AES_KEY *key, const unsigned char ivec[AES_BLOCK_SIZE]); - CRYPTO_ctr128_encrypt_ctr32(in,out,len, - &((EVP_AES_KEY *)ctx->cipher_data)->ks, + CRYPTO_ctr128_encrypt_ctr32(in,out,len,&dat->ks, ctx->iv,ctx->buf,&num,(ctr128_f)AES_ctr32_encrypt); #else - CRYPTO_ctr128_encrypt(in,out,len, - &((EVP_AES_KEY *)ctx->cipher_data)->ks, - ctx->iv,ctx->buf,&num,(block128_f)AES_encrypt); + CRYPTO_ctr128_encrypt(in,out,len,&dat->ks, + ctx->iv,ctx->buf,&num,(block128_f)dat->block); #endif ctx->num = (size_t)num; return 1; @@ -929,9 +989,20 @@ static int aes_gcm_init_key(EVP_CIPHER_CTX *ctx, const unsigned char *key, if (!iv && !key) return 1; if (key) - { + { do { +#ifdef VPAES_CAPABLE + if (VPAES_CAPABLE) + { + vpaes_set_encrypt_key(key,ctx->key_len*8,&gctx->ks); + CRYPTO_gcm128_init(&gctx->gcm,&gctx->ks, + (block128_f)vpaes_encrypt); + break; + } +#endif AES_set_encrypt_key(key, ctx->key_len * 8, &gctx->ks); CRYPTO_gcm128_init(&gctx->gcm, &gctx->ks, (block128_f)AES_encrypt); + } while (0); + /* If we have an iv can set it directly, otherwise use * saved IV. */ @@ -1100,9 +1171,31 @@ static int aes_xts_init_key(EVP_CIPHER_CTX *ctx, const unsigned char *key, if (!iv && !key) return 1; - if (key) + if (key) do { /* key_len is two AES keys */ +#ifdef VPAES_CAPABLE + if (VPAES_CAPABLE) + { + if (enc) + { + vpaes_set_encrypt_key(key, ctx->key_len * 4, &xctx->ks1); + xctx->xts.block1 = (block128_f)vpaes_encrypt; + } + else + { + vpaes_set_decrypt_key(key, ctx->key_len * 4, &xctx->ks1); + xctx->xts.block1 = (block128_f)vpaes_decrypt; + } + + vpaes_set_encrypt_key(key + ctx->key_len/2, + ctx->key_len * 4, &xctx->ks2); + xctx->xts.block2 = (block128_f)vpaes_encrypt; + + xctx->xts.key1 = &xctx->ks1; + break; + } +#endif if (enc) { AES_set_encrypt_key(key, ctx->key_len * 4, &xctx->ks1); @@ -1119,7 +1212,7 @@ static int aes_xts_init_key(EVP_CIPHER_CTX *ctx, const unsigned char *key, xctx->xts.block2 = (block128_f)AES_encrypt; xctx->xts.key1 = &xctx->ks1; - } + } while (0); if (iv) { @@ -1215,13 +1308,23 @@ static int aes_ccm_init_key(EVP_CIPHER_CTX *ctx, const unsigned char *key, EVP_AES_CCM_CTX *cctx = ctx->cipher_data; if (!iv && !key) return 1; - if (key) + if (key) do { +#ifdef VPAES_CAPABLE + if (VPAES_CAPABLE) + { + vpaes_set_encrypt_key(key, ctx->key_len*8, &cctx->ks); + CRYPTO_ccm128_init(&cctx->ccm, cctx->M, cctx->L, + &cctx->ks, (block128_f)vpaes_encrypt); + cctx->key_set = 1; + break; + } +#endif AES_set_encrypt_key(key, ctx->key_len * 8, &cctx->ks); CRYPTO_ccm128_init(&cctx->ccm, cctx->M, cctx->L, &cctx->ks, (block128_f)AES_encrypt); cctx->key_set = 1; - } + } while (0); if (iv) { memcpy(ctx->iv, iv, 15 - cctx->L); -- 2.40.0