#! /usr/bin/env perl
# Copyright 2015-2016 The OpenSSL Project Authors. All Rights Reserved.
#
# Licensed under the Apache License 2.0 (the "License"). You may not use
# this file except in compliance with the License. You can obtain a copy
# in the file LICENSE in the source distribution or at
# https://www.openssl.org/source/license.html
# ====================================================================
# 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/.
# ====================================================================
# March 2015
#
# "Teaser" Montgomery multiplication module for ARMv8. Needs more
# work. While it does improve RSA sign performance by 20-30% (less for
# longer keys) on most processors, for some reason RSA2048 is not
# faster and RSA4096 goes 15-20% slower on Cortex-A57. Multiplication
# instruction issue rate is limited on processor in question, meaning
# that dedicated squaring procedure is a must. Well, actually all
# contemporary AArch64 processors seem to have limited multiplication
# issue rate, i.e. they can't issue multiplication every cycle, which
# explains moderate improvement coefficients in comparison to
# compiler-generated code. Recall that compiler is instructed to use
# umulh and therefore u