#!/usr/bin/env perl
#
# ====================================================================
# Written by Andy Polyakov <appro@openssl.org> for the OpenSSL
# project. The module is, however, dual licensed under OpenSSL and
# CRYPTOGAMS licenses depending on where you obtain it. For further
# details see http://www.openssl.org/~appro/cryptogams/.
# ====================================================================
#
# This module implements support for AES instructions as per PowerISA
# specification version 2.07, first implemented by POWER8 processor.
# The module is endian-agnostic in sense that it supports both big-
# and little-endian cases. Data alignment in parallelizable modes is
# handled with VSX loads and stores, which implies MSR.VSX flag being
# set. It should also be noted that ISA specification doesn't prohibit
# alignment exceptions for these instructions on page boundaries.
# Initially alignment was handled in pure AltiVec/VMX way [when data
# is aligned programmatically, which in turn guarantees exception-
# free execution], but it turned to hamper performance when vcipher
# instructions are interleaved. It's reckoned that eventual
# misalignment penalties at page boundaries are in average lower
# than additional overhead in pure AltiVec approach.
$flavour = shift;
if ($flavour =~ /64/) {
$SIZE_T =8;
$LRSAVE =2*$SIZE_T;
$STU ="stdu";
$POP ="ld";
$PUSH ="std";
$UCMP ="cmpld";
} elsif ($flavour =~ /32/) {
$SIZE_T =4;
$LRSAVE =$SIZE_T;
$STU ="stwu";
$POP ="lwz";
$PUSH ="stw";
$UCMP ="cmplw";
} else { die "nonsense $flavour"; }
$LITTLE_ENDIAN = ($flavour=~/le$/) ? $SIZE_T : 0;
$0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
( $xlate="${dir}ppc-xlate.pl" and -f $xlate ) or
( $xlate="${dir}../../perlasm/ppc-xlate.pl" and -f $xlate) or
die "can't locate ppc-xlate.pl";
open STDOUT,"| $^X $xlate $flavour ".shift || die "can't call $xlate: $!";
$FRAME=8*$SIZE_T;
$prefix="aes_p8";
$sp="r1";
$vrsave="r12";
#########################################################################
{{{ # Key setup procedures #
my ($inp,$bits,$out,$ptr,$cnt,$rounds)=map("r$_",(3..8));
my ($zero,$in0,$in1,$key,$rcon,$mask,$tmp)=map("v$_",(0..6));
my ($stage,$outperm,$outmask,$outhead,$outtail)=map("v$_",(7..11));
$code.=<<___;
.machine "any"
.text
.align 7
rcon:
.long 0x01000000, 0x01000000, 0x01000000, 0x01000000 ?rev
.long 0x1b000000, 0x1b000000, 0x1b000000, 0x1b000000 ?rev
.long 0x0d0e0f0c, 0x0d0e0f0c, 0x0d0e0f0c, 0x0d0e0f0c ?rev
.long 0,0,0,0 ?asis
Lconsts:
mflr r0
bcl 20,31,\$+4
mflr $ptr #vvvvv "distance between . and rcon
addi $ptr,$ptr,-0x48
mtlr r0
blr
.long 0
.byte 0,12,0x14,0,0,0,0,0
.asciz "AES for PowerISA 2.07, CRYPTOGAMS by <appro\@openssl.org>"
.globl .${prefix}_set_encrypt_key
.align 5
.${prefix}_set_encrypt_key:
Lset_encrypt_key:
mflr r11
lis r0,0xfff0
$PUSH r11,$LRSAVE($sp)
mfspr $vrsave,256
mtspr 256,r0