diff options
-rw-r--r-- | CHANGES | 4 | ||||
-rwxr-xr-x | Configure | 14 | ||||
-rw-r--r-- | TABLE | 132 | ||||
-rwxr-xr-x | config | 10 | ||||
-rw-r--r-- | crypto/aes/asm/aes-mips.pl | 811 | ||||
-rw-r--r-- | crypto/sha/asm/sha1-mips.pl | 96 | ||||
-rw-r--r-- | crypto/sha/asm/sha512-mips.pl | 79 |
7 files changed, 984 insertions, 162 deletions
@@ -4,6 +4,10 @@ Changes between 1.0.1 and 1.0.2 [xx XXX xxxx] + *) MIPS assembly pack updates: support for MIPS32r2 and SmartMIPS ASE, + platform support for Linux and Android. + [Andy Polyakov] + *) Call OCSP Stapling callback after ciphersuite has been chosen, so the right response is stapled. Also change current certificate to the certificate actually sent. @@ -348,6 +348,13 @@ my %table=( # It's believed that majority of ARM toolchains predefine appropriate -march. # If you compiler does not, do complement config command line with one! "linux-armv4", "gcc:-DTERMIO -O3 -Wall::-D_REENTRANT::-ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR:${armv4_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)", +# Configure script adds minimally required -march for assembly support, +# if no -march was specified at command line. mips32 and mips64 below +# refer to contemporary MIPS Architecture specifications, MIPS32 and +# MIPS64, rather than to kernel bitness. +"linux-mips32", "gcc:-mabi=32 -DTERMIO -O3 -Wall::-D_REENTRANT::-ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR:${mips32_asm}:o32:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)", +"linux-mips64", "gcc:-mabi=n32 -DTERMIO -O3 -Wall::-D_REENTRANT::-ldl:SIXTY_FOUR_BIT RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR:${mips64_asm}:n32:dlfcn:linux-shared:-fPIC:-mabi=n32:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR):::32", +"linux64-mips64", "gcc:-mabi=64 -DTERMIO -O3 -Wall::-D_REENTRANT::-ldl:SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR:${mips64_asm}:64:dlfcn:linux-shared:-fPIC:-mabi=64:.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR):::64", #### IA-32 targets... "linux-ia32-icc", "icc:-DL_ENDIAN -DTERMIO -O2 -no_cpprt::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-KPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)", "linux-elf", "gcc:-DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -Wall::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)", @@ -406,6 +413,7 @@ my %table=( "android","gcc:-mandroid -I\$(ANDROID_DEV)/include -B\$(ANDROID_DEV)/lib -O3 -fomit-frame-pointer -Wall::-D_REENTRANT::-ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR:${no_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)", "android-x86","gcc:-mandroid -I\$(ANDROID_DEV)/include -B\$(ANDROID_DEV)/lib -O3 -fomit-frame-pointer -Wall::-D_REENTRANT::-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:".eval{my $asm=${x86_elf_asm};$asm=~s/:elf/:android/;$asm}.":dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)", "android-armv7","gcc:-march=armv7-a -mandroid -I\$(ANDROID_DEV)/include -B\$(ANDROID_DEV)/lib -O3 -fomit-frame-pointer -Wall::-D_REENTRANT::-ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR:${armv4_asm}:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)", +"android-mips","gcc:-mandroid -I\$(ANDROID_DEV)/include -B\$(ANDROID_DEV)/lib -O3 -Wall::-D_REENTRANT::-ldl:BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR:${mips32_asm}:o32:dlfcn:linux-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)", #### *BSD [do see comment about ${BSDthreads} above!] "BSD-generic32","gcc:-DTERMIOS -O3 -fomit-frame-pointer -Wall::${BSDthreads}:::BN_LLONG RC2_CHAR RC4_INDEX DES_INT DES_UNROLL:${no_asm}:dlfcn:bsd-gcc-shared:-fPIC::.so.\$(SHLIB_MAJOR).\$(SHLIB_MINOR)", @@ -1202,6 +1210,12 @@ if ($target =~ /^mingw/ && `$cc --target-help 2>&1` !~ m/\-mno\-cygwin/m) $shared_ldflag =~ s/\-mno\-cygwin\s*//; } +if ($target =~ /linux.*\-mips/ && !$no_asm && $flags !~ /\-m(ips|arch=)/) { + # minimally required architecture flags for assembly modules + $cflags="-mips2 $cflags" if ($target =~ /mips32/); + $cflags="-mips3 $cflags" if ($target =~ /mips64/); +} + my $no_shared_warn=0; my $no_user_cflags=0; @@ -1089,6 +1089,39 @@ $ranlib = $arflags = $multilib = +*** android-mips +$cc = gcc +$cflags = -mandroid -I$(ANDROID_DEV)/include -B$(ANDROID_DEV)/lib -O3 -Wall +$unistd = +$thread_cflag = -D_REENTRANT +$sys_id = +$lflags = -ldl +$bn_ops = BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR +$cpuid_obj = +$bn_obj = bn-mips.o mips-mont.o +$des_obj = +$aes_obj = aes_cbc.o aes-mips.o +$bf_obj = +$md5_obj = +$sha1_obj = sha1-mips.o sha256-mips.o +$cast_obj = +$rc4_obj = +$rmd160_obj = +$rc5_obj = +$wp_obj = +$cmll_obj = +$modes_obj = +$engines_obj = +$perlasm_scheme = o32 +$dso_scheme = dlfcn +$shared_target= linux-shared +$shared_cflag = -fPIC +$shared_ldflag = +$shared_extension = .so.$(SHLIB_MAJOR).$(SHLIB_MINOR) +$ranlib = +$arflags = +$multilib = + *** android-x86 $cc = gcc $cflags = -mandroid -I$(ANDROID_DEV)/include -B$(ANDROID_DEV)/lib -O3 -fomit-frame-pointer -Wall @@ -4191,6 +4224,72 @@ $ranlib = $arflags = $multilib = +*** linux-mips32 +$cc = gcc +$cflags = -mabi=32 -DTERMIO -O3 -Wall +$unistd = +$thread_cflag = -D_REENTRANT +$sys_id = +$lflags = -ldl +$bn_ops = BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR +$cpuid_obj = +$bn_obj = bn-mips.o mips-mont.o +$des_obj = +$aes_obj = aes_cbc.o aes-mips.o +$bf_obj = +$md5_obj = +$sha1_obj = sha1-mips.o sha256-mips.o +$cast_obj = +$rc4_obj = +$rmd160_obj = +$rc5_obj = +$wp_obj = +$cmll_obj = +$modes_obj = +$engines_obj = +$perlasm_scheme = o32 +$dso_scheme = dlfcn +$shared_target= linux-shared +$shared_cflag = -fPIC +$shared_ldflag = +$shared_extension = .so.$(SHLIB_MAJOR).$(SHLIB_MINOR) +$ranlib = +$arflags = +$multilib = + +*** linux-mips64 +$cc = gcc +$cflags = -mabi=n32 -DTERMIO -O3 -Wall +$unistd = +$thread_cflag = -D_REENTRANT +$sys_id = +$lflags = -ldl +$bn_ops = SIXTY_FOUR_BIT RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR +$cpuid_obj = +$bn_obj = bn-mips.o mips-mont.o +$des_obj = +$aes_obj = aes_cbc.o aes-mips.o +$bf_obj = +$md5_obj = +$sha1_obj = sha1-mips.o sha256-mips.o sha512-mips.o +$cast_obj = +$rc4_obj = +$rmd160_obj = +$rc5_obj = +$wp_obj = +$cmll_obj = +$modes_obj = +$engines_obj = +$perlasm_scheme = n32 +$dso_scheme = dlfcn +$shared_target= linux-shared +$shared_cflag = -fPIC +$shared_ldflag = -mabi=n32 +$shared_extension = .so.$(SHLIB_MAJOR).$(SHLIB_MINOR) +$ranlib = +$arflags = +$multilib = 32 + *** linux-ppc $cc = gcc $cflags = -DB_ENDIAN -DTERMIO -O3 -Wall @@ -4422,6 +4521,39 @@ $ranlib = $arflags = $multilib = /highgprs +*** linux64-mips64 +$cc = gcc +$cflags = -mabi=64 -DTERMIO -O3 -Wall +$unistd = +$thread_cflag = -D_REENTRANT +$sys_id = +$lflags = -ldl +$bn_ops = SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR +$cpuid_obj = +$bn_obj = bn-mips.o mips-mont.o +$des_obj = +$aes_obj = aes_cbc.o aes-mips.o +$bf_obj = +$md5_obj = +$sha1_obj = sha1-mips.o sha256-mips.o sha512-mips.o +$cast_obj = +$rc4_obj = +$rmd160_obj = +$rc5_obj = +$wp_obj = +$cmll_obj = +$modes_obj = +$engines_obj = +$perlasm_scheme = 64 +$dso_scheme = dlfcn +$shared_target= linux-shared +$shared_cflag = -fPIC +$shared_ldflag = -mabi=64 +$shared_extension = .so.$(SHLIB_MAJOR).$(SHLIB_MINOR) +$ranlib = +$arflags = +$multilib = 64 + *** linux64-s390x $cc = gcc $cflags = -m64 -DB_ENDIAN -DTERMIO -O3 -Wall @@ -596,6 +596,16 @@ case "$GUESSOS" in OUT="linux-ppc" ;; ppc-*-linux2) OUT="linux-ppc" ;; + mips64*-*-linux2) + echo "WARNING! If you wish to build 64-bit library, then you have to" + echo " invoke './Configure linux64-mips64' *manually*." + if [ "$TEST" = "false" -a -t 1 ]; then + echo " You have about 5 seconds to press Ctrl-C to abort." + (trap "stty `stty -g`" 2 0; stty -icanon min 0 time 50; read waste) <&1 + fi + OUT="linux-mips64" + ;; + mips*-*-linux2) OUT="linux-mips32" ;; ppc60x-*-vxworks*) OUT="vxworks-ppc60x" ;; ppcgen-*-vxworks*) OUT="vxworks-ppcgen" ;; pentium-*-vxworks*) OUT="vxworks-pentium" ;; diff --git a/crypto/aes/asm/aes-mips.pl b/crypto/aes/asm/aes-mips.pl index 07ac70e77b..1fdc6bf85c 100644 --- a/crypto/aes/asm/aes-mips.pl +++ b/crypto/aes/asm/aes-mips.pl @@ -20,6 +20,13 @@ # thing about this module is its endian neutrality, which means that # it processes data without ever changing byte order... +# September 2012 +# +# Add MIPS32R2 (~10% less instructions) and SmartMIPS ASE (further +# ~25% less instructions) code. Note that there is no run-time switch, +# instead, code path is chosen upon pre-process time, pass -mips32r2 +# or/and -msmartmips. + ###################################################################### # There is a number of MIPS ABI in use, O32 and N32/64 are most # widely used. Then there is a new contender: NUBI. It appears that if @@ -52,6 +59,7 @@ $flavour = shift || "o32"; # supported flavours are o32,n32,64,nubi32,nubi64 if ($flavour =~ /64|n32/i) { $PTR_ADD="dadd"; # incidentally works even on n32 $PTR_SUB="dsub"; # incidentally works even on n32 + $PTR_INS="dins"; $REG_S="sd"; $REG_L="ld"; $PTR_SLL="dsll"; # incidentally works even on n32 @@ -59,6 +67,7 @@ if ($flavour =~ /64|n32/i) { } else { $PTR_ADD="add"; $PTR_SUB="sub"; + $PTR_INS="ins"; $REG_S="sw"; $REG_L="lw"; $PTR_SLL="sll"; @@ -89,7 +98,11 @@ $code.=<<___; # include <openssl/fipssyms.h> #endif -#if !defined(__vxworks) || defined(__pic__) +#if defined(__mips_smartmips) && !defined(_MIPS_ARCH_MIPS32R2) +#define _MIPS_ARCH_MIPS32R2 +#endif + +#if !defined(__mips_eabi) && (!defined(__vxworks) || defined(__pic__)) .option pic2 #endif .set noat @@ -125,6 +138,89 @@ _mips_AES_encrypt: xor $s3,$t3 sub $cnt,1 +#if defined(__mips_smartmips) + ext $i0,$s1,16,8 +.Loop_enc: + ext $i1,$s2,16,8 + ext $i2,$s3,16,8 + ext $i3,$s0,16,8 + lwxs $t0,$i0($Tbl) # Te1[s1>>16] + ext $i0,$s2,8,8 + lwxs $t1,$i1($Tbl) # Te1[s2>>16] + ext $i1,$s3,8,8 + lwxs $t2,$i2($Tbl) # Te1[s3>>16] + ext $i2,$s0,8,8 + lwxs $t3,$i3($Tbl) # Te1[s0>>16] + ext $i3,$s1,8,8 + + lwxs $t4,$i0($Tbl) # Te2[s2>>8] + ext $i0,$s3,0,8 + lwxs $t5,$i1($Tbl) # Te2[s3>>8] + ext $i1,$s0,0,8 + lwxs $t6,$i2($Tbl) # Te2[s0>>8] + ext $i2,$s1,0,8 + lwxs $t7,$i3($Tbl) # Te2[s1>>8] + ext $i3,$s2,0,8 + + lwxs $t8,$i0($Tbl) # Te3[s3] + ext $i0,$s0,24,8 + lwxs $t9,$i1($Tbl) # Te3[s0] + ext $i1,$s1,24,8 + lwxs $t10,$i2($Tbl) # Te3[s1] + ext $i2,$s2,24,8 + lwxs $t11,$i3($Tbl) # Te3[s2] + ext $i3,$s3,24,8 + + rotr $t0,$t0,8 + rotr $t1,$t1,8 + rotr $t2,$t2,8 + rotr $t3,$t3,8 + + rotr $t4,$t4,16 + rotr $t5,$t5,16 + rotr $t6,$t6,16 + rotr $t7,$t7,16 + + xor $t0,$t4 + lwxs $t4,$i0($Tbl) # Te0[s0>>24] + xor $t1,$t5 + lwxs $t5,$i1($Tbl) # Te0[s1>>24] + xor $t2,$t6 + lwxs $t6,$i2($Tbl) # Te0[s2>>24] + xor $t3,$t7 + lwxs $t7,$i3($Tbl) # Te0[s3>>24] + + rotr $t8,$t8,24 + lw $s0,0($key0) + rotr $t9,$t9,24 + lw $s1,4($key0) + rotr $t10,$t10,24 + lw $s2,8($key0) + rotr $t11,$t11,24 + lw $s3,12($key0) + + xor $t0,$t8 + xor $t1,$t9 + xor $t2,$t10 + xor $t3,$t11 + + xor $t0,$t4 + xor $t1,$t5 + xor $t2,$t6 + xor $t3,$t7 + + sub $cnt,1 + $PTR_ADD $key0,16 + xor $s0,$t0 + xor $s1,$t1 + xor $s2,$t2 + xor $s3,$t3 + .set noreorder + bnez $cnt,.Loop_enc + ext $i0,$s1,16,8 + + _xtr $i0,$s1,16-2 +#else _xtr $i0,$s1,16-2 .Loop_enc: _xtr $i1,$s2,16-2 @@ -138,19 +234,29 @@ _mips_AES_encrypt: $PTR_ADD $i1,$Tbl $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl +#if defined(_MIPS_ARCH_MIPS32R2) || defined(_MIPS_ARCH_MIPS64R2) + lw $t0,0($i0) # Te1[s1>>16] + _xtr $i0,$s2,8-2 + lw $t1,0($i1) # Te1[s2>>16] + _xtr $i1,$s3,8-2 + lw $t2,0($i2) # Te1[s3>>16] + _xtr $i2,$s0,8-2 + lw $t3,0($i3) # Te1[s0>>16] + _xtr $i3,$s1,8-2 +#else lwl $t0,3($i0) # Te1[s1>>16] lwl $t1,3($i1) # Te1[s2>>16] lwl $t2,3($i2) # Te1[s3>>16] lwl $t3,3($i3) # Te1[s0>>16] lwr $t0,2($i0) # Te1[s1>>16] - lwr $t1,2($i1) # Te1[s2>>16] - lwr $t2,2($i2) # Te1[s3>>16] - lwr $t3,2($i3) # Te1[s0>>16] - _xtr $i0,$s2,8-2 + lwr $t1,2($i1) # Te1[s2>>16] _xtr $i1,$s3,8-2 + lwr $t2,2($i2) # Te1[s3>>16] _xtr $i2,$s0,8-2 + lwr $t3,2($i3) # Te1[s0>>16] _xtr $i3,$s1,8-2 +#endif and $i0,0x3fc and $i1,0x3fc and $i2,0x3fc @@ -159,19 +265,88 @@ _mips_AES_encrypt: $PTR_ADD $i1,$Tbl $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl +#if defined(_MIPS_ARCH_MIPS32R2) || defined(_MIPS_ARCH_MIPS64R2) + rotr $t0,$t0,8 + rotr $t1,$t1,8 + rotr $t2,$t2,8 + rotr $t3,$t3,8 +# if defined(_MIPSEL) + lw $t4,0($i0) # Te2[s2>>8] + _xtr $i0,$s3,0-2 + lw $t5,0($i1) # Te2[s3>>8] + _xtr $i1,$s0,0-2 + lw $t6,0($i2) # Te2[s0>>8] + _xtr $i2,$s1,0-2 + lw $t7,0($i3) # Te2[s1>>8] + _xtr $i3,$s2,0-2 + + and $i0,0x3fc + and $i1,0x3fc + and $i2,0x3fc + and $i3,0x3fc + $PTR_ADD $i0,$Tbl + $PTR_ADD $i1,$Tbl + $PTR_ADD $i2,$Tbl + $PTR_ADD $i3,$Tbl + lw $t8,0($i0) # Te3[s3] + $PTR_INS $i0,$s0,2,8 + lw $t9,0($i1) # Te3[s0] + $PTR_INS $i1,$s1,2,8 + lw $t10,0($i2) # Te3[s1] + $PTR_INS $i2,$s2,2,8 + lw $t11,0($i3) # Te3[s2] + $PTR_INS $i3,$s3,2,8 +# else + lw $t4,0($i0) # Te2[s2>>8] + $PTR_INS $i0,$s3,2,8 + lw $t5,0($i1) # Te2[s3>>8] + $PTR_INS $i1,$s0,2,8 + lw $t6,0($i2) # Te2[s0>>8] + $PTR_INS $i2,$s1,2,8 + lw $t7,0($i3) # Te2[s1>>8] + $PTR_INS $i3,$s2,2,8 + + lw $t8,0($i0) # Te3[s3] + _xtr $i0,$s0,24-2 + lw $t9,0($i1) # Te3[s0] + _xtr $i1,$s1,24-2 + lw $t10,0($i2) # Te3[s1] + _xtr $i2,$s2,24-2 + lw $t11,0($i3) # Te3[s2] + _xtr $i3,$s3,24-2 + + and $i0,0x3fc + and $i1,0x3fc + and $i2,0x3fc + and $i3,0x3fc + $PTR_ADD $i0,$Tbl + $PTR_ADD $i1,$Tbl + $PTR_ADD $i2,$Tbl + $PTR_ADD $i3,$Tbl +# endif + rotr $t4,$t4,16 + rotr $t5,$t5,16 + rotr $t6,$t6,16 + rotr $t7,$t7,16 + + rotr $t8,$t8,24 + rotr $t9,$t9,24 + rotr $t10,$t10,24 + rotr $t11,$t11,24 +#else lwl $t4,2($i0) # Te2[s2>>8] lwl $t5,2($i1) # Te2[s3>>8] lwl $t6,2($i2) # Te2[s0>>8] lwl $t7,2($i3) # Te2[s1>>8] lwr $t4,1($i0) # Te2[s2>>8] - lwr $t5,1($i1) # Te2[s3>>8] - lwr $t6,1($i2) # Te2[s0>>8] - lwr $t7,1($i3) # Te2[s1>>8] - _xtr $i0,$s3,0-2 + lwr $t5,1($i1) # Te2[s3>>8] _xtr $i1,$s0,0-2 + lwr $t6,1($i2) # Te2[s0>>8] _xtr $i2,$s1,0-2 + lwr $t7,1($i3) # Te2[s1>>8] _xtr $i3,$s2,0-2 + and $i0,0x3fc and $i1,0x3fc and $i2,0x3fc @@ -185,14 +360,14 @@ _mips_AES_encrypt: lwl $t10,1($i2) # Te3[s1] lwl $t11,1($i3) # Te3[s2] lwr $t8,0($i0) # Te3[s3] - lwr $t9,0($i1) # Te3[s0] - lwr $t10,0($i2) # Te3[s1] - lwr $t11,0($i3) # Te3[s2] - _xtr $i0,$s0,24-2 + lwr $t9,0($i1) # Te3[s0] _xtr $i1,$s1,24-2 + lwr $t10,0($i2) # Te3[s1] _xtr $i2,$s2,24-2 + lwr $t11,0($i3) # Te3[s2] _xtr $i3,$s3,24-2 + and $i0,0x3fc and $i1,0x3fc and $i2,0x3fc @@ -201,24 +376,24 @@ _mips_AES_encrypt: $PTR_ADD $i1,$Tbl $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl +#endif xor $t0,$t4 - xor $t1,$t5 - xor $t2,$t6 - xor $t3,$t7 lw $t4,0($i0) # Te0[s0>>24] + xor $t1,$t5 lw $t5,0($i1) # Te0[s1>>24] + xor $t2,$t6 lw $t6,0($i2) # Te0[s2>>24] + xor $t3,$t7 lw $t7,0($i3) # Te0[s3>>24] - lw $s0,0($key0) - lw $s1,4($key0) - lw $s2,8($key0) - lw $s3,12($key0) - xor $t0,$t8 + lw $s0,0($key0) xor $t1,$t9 + lw $s1,4($key0) xor $t2,$t10 + lw $s2,8($key0) xor $t3,$t11 + lw $s3,12($key0) xor $t0,$t4 xor $t1,$t5 @@ -234,6 +409,7 @@ _mips_AES_encrypt: .set noreorder bnez $cnt,.Loop_enc _xtr $i0,$s1,16-2 +#endif .set reorder _xtr $i1,$s2,16-2 @@ -248,14 +424,14 @@ _mips_AES_encrypt: $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl lbu $t0,2($i0) # Te4[s1>>16] - lbu $t1,2($i1) # Te4[s2>>16] - lbu $t2,2($i2) # Te4[s3>>16] - lbu $t3,2($i3) # Te4[s0>>16] - _xtr $i0,$s2,8-2 + lbu $t1,2($i1) # Te4[s2>>16] _xtr $i1,$s3,8-2 + lbu $t2,2($i2) # Te4[s3>>16] _xtr $i2,$s0,8-2 + lbu $t3,2($i3) # Te4[s0>>16] _xtr $i3,$s1,8-2 + and $i0,0x3fc and $i1,0x3fc and $i2,0x3fc @@ -264,15 +440,44 @@ _mips_AES_encrypt: $PTR_ADD $i1,$Tbl $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl +#if defined(_MIPS_ARCH_MIPS32R2) || defined(_MIPS_ARCH_MIPS64R2) +# if defined(_MIPSEL) lbu $t4,2($i0) # Te4[s2>>8] + $PTR_INS $i0,$s0,2,8 lbu $t5,2($i1) # Te4[s3>>8] + $PTR_INS $i1,$s1,2,8 lbu $t6,2($i2) # Te4[s0>>8] + $PTR_INS $i2,$s2,2,8 lbu $t7,2($i3) # Te4[s1>>8] + $PTR_INS $i3,$s3,2,8 + lbu $t8,2($i0) # Te4[s0>>24] + _xtr $i0,$s3,0-2 + lbu $t9,2($i1) # Te4[s1>>24] + _xtr $i1,$s0,0-2 + lbu $t10,2($i2) # Te4[s2>>24] + _xtr $i2,$s1,0-2 + lbu $t11,2($i3) # Te4[s3>>24] + _xtr $i3,$s2,0-2 + + and $i0,0x3fc + and $i1,0x3fc + and $i2,0x3fc + and $i3,0x3fc + $PTR_ADD $i0,$Tbl + $PTR_ADD $i1,$Tbl + $PTR_ADD $i2,$Tbl + $PTR_ADD $i3,$Tbl +# else + lbu $t4,2($i0) # Te4[s2>>8] _xtr $i0,$s0,24-2 + lbu $t5,2($i1) # Te4[s3>>8] _xtr $i1,$s1,24-2 + lbu $t6,2($i2) # Te4[s0>>8] _xtr $i2,$s2,24-2 + lbu $t7,2($i3) # Te4[s1>>8] _xtr $i3,$s3,24-2 + and $i0,0x3fc and $i1,0x3fc and $i2,0x3fc @@ -282,18 +487,76 @@ _mips_AES_encrypt: $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl lbu $t8,2($i0) # Te4[s0>>24] + $PTR_INS $i0,$s3,2,8 lbu $t9,2($i1) # Te4[s1>>24] + $PTR_INS $i1,$s0,2,8 lbu $t10,2($i2) # Te4[s2>>24] + $PTR_INS $i2,$s1,2,8 lbu $t11,2($i3) # Te4[s3>>24] + $PTR_INS $i3,$s2,2,8 +# endif + _ins $t0,16 + _ins $t1,16 + _ins $t2,16 + _ins $t3,16 + _ins2 $t0,$t4,8 + lbu $t4,2($i0) # Te4[s3] + _ins2 $t1,$t5,8 + lbu $t5,2($i1) # Te4[s0] + _ins2 $t2,$t6,8 + lbu $t6,2($i2) # Te4[s1] + _ins2 $t3,$t7,8 + lbu $t7,2($i3) # Te4[s2] + + _ins2 $t0,$t8,24 + lw $s0,0($key0) + _ins2 $t1,$t9,24 + lw $s1,4($key0) + _ins2 $t2,$t10,24 + lw $s2,8($key0) + _ins2 $t3,$t11,24 + lw $s3,12($key0) + + _ins2 $t0,$t4,0 + _ins2 $t1,$t5,0 + _ins2 $t2,$t6,0 + _ins2 $t3,$t7,0 +#else + lbu $t4,2($i0) # Te4[s2>>8] + _xtr $i0,$s0,24-2 + lbu $t5,2($i1) # Te4[s3>>8] + _xtr $i1,$s1,24-2 + lbu $t6,2($i2) # Te4[s0>>8] + _xtr $i2,$s2,24-2 + lbu $t7,2($i3) # Te4[s1>>8] + _xtr $i3,$s3,24-2 + + and $i0,0x3fc + and $i1,0x3fc + and $i2,0x3fc + and $i3,0x3fc + $PTR_ADD $i0,$Tbl + $PTR_ADD $i1,$Tbl + $PTR_ADD $i2,$Tbl + $PTR_ADD $i3,$Tbl + lbu $t8,2($i0) # Te4[s0>>24] _xtr $i0,$s3,0-2 + lbu $t9,2($i1) # Te4[s1>>24] _xtr $i1,$s0,0-2 + lbu $t10,2($i2) # Te4[s2>>24] _xtr $i2,$s1,0-2 + lbu $t11,2($i3) # Te4[s3>>24] _xtr $i3,$s2,0-2 + and $i0,0x3fc and $i1,0x3fc and $i2,0x3fc and $i3,0x3fc + $PTR_ADD $i0,$Tbl + $PTR_ADD $i1,$Tbl + $PTR_ADD $i2,$Tbl + $PTR_ADD $i3,$Tbl _ins $t0,16 _ins $t1,16 @@ -306,27 +569,21 @@ _mips_AES_encrypt: _ins $t7,8 xor $t0,$t4 - xor $t1,$t5 - xor $t2,$t6 - xor $t3,$t7 - - $PTR_ADD $i0,$Tbl - $PTR_ADD $i1,$Tbl - $PTR_ADD $i2,$Tbl - $PTR_ADD $i3,$Tbl lbu $t4,2($i0) # Te4[s3] + xor $t1,$t5 lbu $t5,2($i1) # Te4[s0] + xor $t2,$t6 lbu $t6,2($i2) # Te4[s1] + xor $t3,$t7 lbu $t7,2($i3) # Te4[s2] _ins $t8,24 - _ins $t9,24 - _ins $t10,24 - _ins $t11,24 - lw $s0,0($key0) + _ins $t9,24 lw $s1,4($key0) + _ins $t10,24 lw $s2,8($key0) + _ins $t11,24 lw $s3,12($key0) xor $t0,$t8 @@ -343,7 +600,7 @@ _mips_AES_encrypt: xor $t1,$t5 xor $t2,$t6 xor $t3,$t7 - +#endif xor $s0,$t0 xor $s1,$t1 xor $s2,$t2 @@ -455,6 +712,89 @@ _mips_AES_decrypt: xor $s3,$t3 sub $cnt,1 +#if defined(__mips_smartmips) + ext $i0,$s3,16,8 +.Loop_dec: + ext $i1,$s0,16,8 + ext $i2,$s1,16,8 + ext $i3,$s2,16,8 + lwxs $t0,$i0($Tbl) # Td1[s3>>16] + ext $i0,$s2,8,8 + lwxs $t1,$i1($Tbl) # Td1[s0>>16] + ext $i1,$s3,8,8 + lwxs $t2,$i2($Tbl) # Td1[s1>>16] + ext $i2,$s0,8,8 + lwxs $t3,$i3($Tbl) # Td1[s2>>16] + ext $i3,$s1,8,8 + + lwxs $t4,$i0($Tbl) # Td2[s2>>8] + ext $i0,$s1,0,8 + lwxs $t5,$i1($Tbl) # Td2[s3>>8] + ext $i1,$s2,0,8 + lwxs $t6,$i2($Tbl) # Td2[s0>>8] + ext $i2,$s3,0,8 + lwxs $t7,$i3($Tbl) # Td2[s1>>8] + ext $i3,$s0,0,8 + + lwxs $t8,$i0($Tbl) # Td3[s1] + ext $i0,$s0,24,8 + lwxs $t9,$i1($Tbl) # Td3[s2] + ext $i1,$s1,24,8 + lwxs $t10,$i2($Tbl) # Td3[s3] + ext $i2,$s2,24,8 + lwxs $t11,$i3($Tbl) # Td3[s0] + ext $i3,$s3,24,8 + + rotr $t0,$t0,8 + rotr $t1,$t1,8 + rotr $t2,$t2,8 + rotr $t3,$t3,8 + + rotr $t4,$t4,16 + rotr $t5,$t5,16 + rotr $t6,$t6,16 + rotr $t7,$t7,16 + + xor $t0,$t4 + lwxs $t4,$i0($Tbl) # Td0[s0>>24] + xor $t1,$t5 + lwxs $t5,$i1($Tbl) # Td0[s1>>24] + xor $t2,$t6 + lwxs $t6,$i2($Tbl) # Td0[s2>>24] + xor $t3,$t7 + lwxs $t7,$i3($Tbl) # Td0[s3>>24] + + rotr $t8,$t8,24 + lw $s0,0($key0) + rotr $t9,$t9,24 + lw $s1,4($key0) + rotr $t10,$t10,24 + lw $s2,8($key0) + rotr $t11,$t11,24 + lw $s3,12($key0) + + xor $t0,$t8 + xor $t1,$t9 + xor $t2,$t10 + xor $t3,$t11 + + xor $t0,$t4 + xor $t1,$t5 + xor $t2,$t6 + xor $t3,$t7 + + sub $cnt,1 + $PTR_ADD $key0,16 + xor $s0,$t0 + xor $s1,$t1 + xor $s2,$t2 + xor $s3,$t3 + .set noreorder + bnez $cnt,.Loop_dec + ext $i0,$s3,16,8 + + _xtr $i0,$s3,16-2 +#else _xtr $i0,$s3,16-2 .Loop_dec: _xtr $i1,$s0,16-2 @@ -468,19 +808,88 @@ _mips_AES_decrypt: $PTR_ADD $i1,$Tbl $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl +#if defined(_MIPS_ARCH_MIPS32R2) || defined(_MIPS_ARCH_MIPS64R2) + lw $t0,0($i0) # Td1[s3>>16] + _xtr $i0,$s2,8-2 + lw $t1,0($i1) # Td1[s0>>16] + _xtr $i1,$s3,8-2 + lw $t2,0($i2) # Td1[s1>>16] + _xtr $i2,$s0,8-2 + lw $t3,0($i3) # Td1[s2>>16] + _xtr $i3,$s1,8-2 +#else lwl $t0,3($i0) # Td1[s3>>16] lwl $t1,3($i1) # Td1[s0>>16] lwl $t2,3($i2) # Td1[s1>>16] lwl $t3,3($i3) # Td1[s2>>16] lwr $t0,2($i0) # Td1[s3>>16] - lwr $t1,2($i1) # Td1[s0>>16] - lwr $t2,2($i2) # Td1[s1>>16] - lwr $t3,2($i3) # Td1[s2>>16] - _xtr $i0,$s2,8-2 + lwr $t1,2($i1) # Td1[s0>>16] _xtr $i1,$s3,8-2 + lwr $t2,2($i2) # Td1[s1>>16] _xtr $i2,$s0,8-2 + lwr $t3,2($i3) # Td1[s2>>16] _xtr $i3,$s1,8-2 +#endif + + and $i0,0x3fc + and $i1,0x3fc + and $i2,0x3fc + and $i3,0x3fc + $PTR_ADD $i0,$Tbl + $PTR_ADD $i1,$Tbl + $PTR_ADD $i2,$Tbl + $PTR_ADD $i3,$Tbl +#if defined(_MIPS_ARCH_MIPS32R2) || defined(_MIPS_ARCH_MIPS64R2) + rotr $t0,$t0,8 + rotr $t1,$t1,8 + rotr $t2,$t2,8 + rotr $t3,$t3,8 +# if defined(_MIPSEL) + lw $t4,0($i0) # Td2[s2>>8] + _xtr $i0,$s1,0-2 + lw $t5,0($i1) # Td2[s3>>8] + _xtr $i1,$s2,0-2 + lw $t6,0($i2) # Td2[s0>>8] + _xtr $i2,$s3,0-2 + lw $t7,0($i3) # Td2[s1>>8] + _xtr $i3,$s0,0-2 + + and $i0,0x3fc + and $i1,0x3fc + and $i2,0x3fc + and $i3,0x3fc + $PTR_ADD $i0,$Tbl + $PTR_ADD $i1,$Tbl + $PTR_ADD $i2,$Tbl + $PTR_ADD $i3,$Tbl + lw $t8,0($i0) # Td3[s1] + $PTR_INS $i0,$s0,2,8 + lw $t9,0($i1) # Td3[s2] + $PTR_INS $i1,$s1,2,8 + lw $t10,0($i2) # Td3[s3] + $PTR_INS $i2,$s2,2,8 + lw $t11,0($i3) # Td3[s0] + $PTR_INS $i3,$s3,2,8 +#else + lw $t4,0($i0) # Td2[s2>>8] + $PTR_INS $i0,$s1,2,8 + lw $t5,0($i1) # Td2[s3>>8] + $PTR_INS $i1,$s2,2,8 + lw $t6,0($i2) # Td2[s0>>8] + $PTR_INS $i2,$s3,2,8 + lw $t7,0($i3) # Td2[s1>>8] + $PTR_INS $i3,$s0,2,8 + + lw $t8,0($i0) # Td3[s1] + _xtr $i0,$s0,24-2 + lw $t9,0($i1) # Td3[s2] + _xtr $i1,$s1,24-2 + lw $t10,0($i2) # Td3[s3] + _xtr $i2,$s2,24-2 + lw $t11,0($i3) # Td3[s0] + _xtr $i3,$s3,24-2 + and $i0,0x3fc and $i1,0x3fc and $i2,0x3fc @@ -489,19 +898,30 @@ _mips_AES_decrypt: $PTR_ADD $i1,$Tbl $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl +#endif + rotr $t4,$t4,16 + rotr $t5,$t5,16 + rotr $t6,$t6,16 + rotr $t7,$t7,16 + + rotr $t8,$t8,24 + rotr $t9,$t9,24 + rotr $t10,$t10,24 + rotr $t11,$t11,24 +#else lwl $t4,2($i0) # Td2[s2>>8] lwl $t5,2($i1) # Td2[s3>>8] lwl $t6,2($i2) # Td2[s0>>8] lwl $t7,2($i3) # Td2[s1>>8] lwr $t4,1($i0) # Td2[s2>>8] - lwr $t5,1($i1) # Td2[s3>>8] - lwr $t6,1($i2) # Td2[s0>>8] - lwr $t7,1($i3) # Td2[s1>>8] - _xtr $i0,$s1,0-2 + lwr $t5,1($i1) # Td2[s3>>8] _xtr $i1,$s2,0-2 + lwr $t6,1($i2) # Td2[s0>>8] _xtr $i2,$s3,0-2 + lwr $t7,1($i3) # Td2[s1>>8] _xtr $i3,$s0,0-2 + and $i0,0x3fc and $i1,0x3fc and $i2,0x3fc @@ -515,14 +935,14 @@ _mips_AES_decrypt: lwl $t10,1($i2) # Td3[s3] lwl $t11,1($i3) # Td3[s0] lwr $t8,0($i0) # Td3[s1] - lwr $t9,0($i1) # Td3[s2] - lwr $t10,0($i2) # Td3[s3] - lwr $t11,0($i3) # Td3[s0] - _xtr $i0,$s0,24-2 + lwr $t9,0($i1) # Td3[s2] _xtr $i1,$s1,24-2 + lwr $t10,0($i2) # Td3[s3] _xtr $i2,$s2,24-2 + lwr $t11,0($i3) # Td3[s0] _xtr $i3,$s3,24-2 + and $i0,0x3fc and $i1,0x3fc and $i2,0x3fc @@ -531,27 +951,25 @@ _mips_AES_decrypt: $PTR_ADD $i1,$Tbl $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl +#endif xor $t0,$t4 - xor $t1,$t5 - xor $t2,$t6 - xor $t3,$t7 - - lw $t4,0($i0) # Td0[s0>>24] + xor $t1,$t5 lw $t5,0($i1) # Td0[s1>>24] + xor $t2,$t6 lw $t6,0($i2) # Td0[s2>>24] + xor $t3,$t7 lw $t7,0($i3) # Td0[s3>>24] - lw $s0,0($key0) - lw $s1,4($key0) - lw $s2,8($key0) - lw $s3,12($key0) - xor $t0,$t8 + lw $s0,0($key0) xor $t1,$t9 + lw $s1,4($key0) xor $t2,$t10 + lw $s2,8($key0) xor $t3,$t11 + lw $s3,12($key0) xor $t0,$t4 xor $t1,$t5 @@ -567,38 +985,39 @@ _mips_AES_decrypt: .set noreorder bnez $cnt,.Loop_dec _xtr $i0,$s3,16-2 +#endif .set reorder lw $t4,1024($Tbl) # prefetch Td4 - lw $t5,1024+32($Tbl) - lw $t6,1024+64($Tbl) - lw $t7,1024+96($Tbl) - lw $t8,1024+128($Tbl) - lw $t9,1024+160($Tbl) - lw $t10,1024+192($Tbl) - lw $t11,1024+224($Tbl) - _xtr $i0,$s3,16 + lw $t5,1024+32($Tbl) _xtr $i1,$s0,16 + lw $t6,1024+64($Tbl) _xtr $i2,$s1,16 + lw $t7,1024+96($Tbl) _xtr $i3,$s2,16 + lw $t8,1024+128($Tbl) and $i0,0xff + lw $t9,1024+160($Tbl) and $i1,0xff + lw $t10,1024+192($Tbl) and $i2,0xff + lw $t11,1024+224($Tbl) and $i3,0xff + $PTR_ADD $i0,$Tbl $PTR_ADD $i1,$Tbl $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl lbu $t0,1024($i0) # Td4[s3>>16] - lbu $t1,1024($i1) # Td4[s0>>16] - lbu $t2,1024($i2) # Td4[s1>>16] - lbu $t3,1024($i3) # Td4[s2>>16] - _xtr $i0,$s2,8 + lbu $t1,1024($i1) # Td4[s0>>16] _xtr $i1,$s3,8 + lbu $t2,1024($i2) # Td4[s1>>16] _xtr $i2,$s0,8 + lbu $t3,1024($i3) # Td4[s2>>16] _xtr $i3,$s1,8 + and $i0,0xff and $i1,0xff and $i2,0xff @@ -607,29 +1026,108 @@ _mips_AES_decrypt: $PTR_ADD $i1,$Tbl $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl +#if defined(_MIPS_ARCH_MIPS32R2) || defined(_MIPS_ARCH_MIPS64R2) +# if defined(_MIPSEL) lbu $t4,1024($i0) # Td4[s2>>8] + $PTR_INS $i0,$s0,0,8 lbu $t5,1024($i1) # Td4[s3>>8] + $PTR_INS $i1,$s1,0,8 lbu $t6,1024($i2) # Td4[s0>>8] + $PTR_INS $i2,$s2,0,8 lbu $t7,1024($i3) # Td4[s1>>8] + $PTR_INS $i3,$s3,0,8 + lbu $t8,1024($i0) # Td4[s0>>24] + _xtr $i0,$s1,0 + lbu $t9,1024($i1) # Td4[s1>>24] + _xtr $i1,$s2,0 + lbu $t10,1024($i2) # Td4[s2>>24] + _xtr $i2,$s3,0 + lbu $t11,1024($i3) # Td4[s3>>24] + _xtr $i3,$s0,0 + + $PTR_ADD $i0,$Tbl + $PTR_ADD $i1,$Tbl + $PTR_ADD $i2,$Tbl + $PTR_ADD $i3,$Tbl +# else + lbu $t4,1024($i0) # Td4[s2>>8] _xtr $i0,$s0,24 + lbu $t5,1024($i1) # Td4[s3>>8] _xtr $i1,$s1,24 + lbu $t6,1024($i2) # Td4[s0>>8] _xtr $i2,$s2,24 + lbu $t7,1024($i3) # Td4[s1>>8] _xtr $i3,$s3,24 + $PTR_ADD $i0,$Tbl $PTR_ADD $i1,$Tbl $PTR_ADD $i2,$Tbl $PTR_ADD $i3,$Tbl lbu $t8,1024($i0) # Td4[s0>>24] + $PTR_INS $i0,$s1,0,8 lbu $t9,1024($i1) # Td4[s1>>24] + $PTR_INS $i1,$s2,0,8 lbu $t10,1024($i2) # Td4[s2>>24] + $PTR_INS $i2,$s3,0,8 lbu $t11,1024($i3) # Td4[s3>>24] + $PTR_INS $i3,$s0,0,8 +# endif + _ins $t0,16 + _ins $t1,16 + _ins $t2,16 + _ins $t3,16 + + _ins2 $t0,$t4,8 + lbu $t4,1024($i0) # Td4[s1] + _ins2 $t1,$t5,8 + lbu $t5,1024($i1) # Td4[s2] + _ins2 $t2,$t6,8 + lbu $t6,1024($i2) # Td4[s3] + _ins2 $t3,$t7,8 + lbu $t7,1024($i3) # Td4[s0] + + _ins2 $t0,$t8,24 + lw $s0,0($key0) + _ins2 $t1,$t9,24 + lw $s1,4($key0) + _ins2 $t2,$t10,24 + lw $s2,8($key0) + _ins2 $t3,$t11,24 + lw $s3,12($key0) + _ins2 $t0,$t4,0 + _ins2 $t1,$t5,0 + _ins2 $t2,$t6,0 + _ins2 $t3,$t7,0 +#else + lbu $t4,1024($i0) # Td4[s2>>8] + _xtr $i0,$s0,24 + lbu $t5,1024($i1) # Td4[s3>>8] + _xtr $i1,$s1,24 + lbu $t6,1024($i2) # Td4[s0>>8] + _xtr $i2,$s2,24 + lbu $t7,1024($i3) # Td4[s1>>8] + _xtr $i3,$s3,24 + + $PTR_ADD $i0,$Tbl + $PTR_ADD $i1,$Tbl + $PTR_ADD $i2,$Tbl + $PTR_ADD $i3,$Tbl + lbu $t8,1024($i0) # Td4[s0>>24] _xtr $i0,$s1,0 + lbu $t9,1024($i1) # Td4[s1>>24] _xtr $i1,$s2,0 + lbu $t10,1024($i2) # Td4[s2>>24] _xtr $i2,$s3,0 + lbu $t11,1024($i3) # Td4[s3>>24] _xtr $i3,$s0,0 + $PTR_ADD $i0,$Tbl + $PTR_ADD $i1,$Tbl + $PTR_ADD $i2,$Tbl + $PTR_ADD $i3,$Tbl + _ins $t0,16 _ins $t1,16 _ins $t2,16 @@ -641,44 +1139,38 @@ _mips_AES_de |