kris/boringssl
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
00019f2193
boringssl/crypto/fipsmodule/modes/asm/ghash-armv4.pl

555 lines
14 KiB
Perl
Raw Normal View History

Sync ARM assembly up to 609b0852e4d50251857dbbac3141ba042e35a9ae. This change was made by copying over the files as of that commit and then discarding the parts of the diff which corresponding to our own changes. Change-Id: I28c5d711f7a8cec30749b8174687434129af5209 Reviewed-on: https://boringssl-review.googlesource.com/17111 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:53:40 +01:00
#! /usr/bin/env perl
# Copyright 2010-2016 The OpenSSL Project Authors. All Rights Reserved.
#
# Licensed under the OpenSSL license (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
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
#
# ====================================================================
# 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/.
# ====================================================================
#
# April 2010
#
# The module implements "4-bit" GCM GHASH function and underlying
# single multiplication operation in GF(2^128). "4-bit" means that it
# uses 256 bytes per-key table [+32 bytes shared table]. There is no
# experimental performance data available yet. The only approximation
# that can be made at this point is based on code size. Inner loop is
# 32 instructions long and on single-issue core should execute in <40
# cycles. Having verified that gcc 3.4 didn't unroll corresponding
# loop, this assembler loop body was found to be ~3x smaller than
# compiler-generated one...
#
# July 2010
#
# Rescheduling for dual-issue pipeline resulted in 8.5% improvement on
# Cortex A8 core and ~25 cycles per processed byte (which was observed
# to be ~3 times faster than gcc-generated code:-)
#
# February 2011
#
# Profiler-assisted and platform-specific optimization resulted in 7%
# improvement on Cortex A8 core and ~23.5 cycles per byte.
#
# March 2011
#
# Add NEON implementation featuring polynomial multiplication, i.e. no
# lookup tables involved. On Cortex A8 it was measured to process one
# byte in 15 cycles or 55% faster than integer-only code.
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
#
# April 2014
#
# Switch to multiplication algorithm suggested in paper referred
# below and combine it with reduction algorithm from x86 module.
# Performance improvement over previous version varies from 65% on
# Snapdragon S4 to 110% on Cortex A9. In absolute terms Cortex A8
ARMv4 assembly pack: add Cortex-A15 performance data. (Imported from upstream's e390ae50e0bc41676994c6fa23f7b65a8afc4d7f) Change-Id: Ifee85b0936c06c42cc7c09f8327d15fec51da48a Reviewed-on: https://boringssl-review.googlesource.com/3832 Reviewed-by: Adam Langley <agl@google.com>
2015-03-09 18:51:51 +00:00
# processes one byte in 8.45 cycles, A9 - in 10.2, A15 - in 7.63,
# Snapdragon S4 - in 9.33.
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
#
Convert remaining Latin-1 files to UTF-8. See upstream's 9f0b86c68bb96d49301bbd6473c8235ca05ca06b. Generated by using upstream's script in 5a3ce86e21715a683ff0d32421ed5c6d5e84234d and then manually throwing out the false positives. (We converted a bunch of stuff already in 91157550061d5d794898fe47b95384a7ba5f7b9d.) This may require some wrestling with depot_tools to land in Chromium due to Rietveld's encoding bugs, but hopefully that will avoid future problems; Rietveld breaks if either old or new file is Latin-1. Change-Id: I26dcb20c7377f92a0c843ef5d74d440a82ea8ceb Reviewed-on: https://boringssl-review.googlesource.com/5483 Reviewed-by: Adam Langley <agl@google.com>
2015-07-22 03:29:26 +01:00
# Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
# Polynomial Multiplication on ARM Processors using the NEON Engine.
Remove trailing whitespace from Perl files. Upstream did this in 609b0852e4d50251857dbbac3141ba042e35a9ae and it's easier to apply patches if we do also. Change-Id: I5142693ed1e26640987ff16f5ea510e81bba200e Reviewed-on: https://boringssl-review.googlesource.com/13771 Commit-Queue: Adam Langley <agl@google.com> Reviewed-by: David Benjamin <davidben@google.com>
2017-02-09 20:21:08 +00:00
#
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
# http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
# ====================================================================
# Note about "528B" variant. In ARM case it makes lesser sense to
# implement it for following reasons:
#
# - performance improvement won't be anywhere near 50%, because 128-
# bit shift operation is neatly fused with 128-bit xor here, and
# "538B" variant would eliminate only 4-5 instructions out of 32
# in the inner loop (meaning that estimated improvement is ~15%);
# - ARM-based systems are often embedded ones and extra memory
# consumption might be unappreciated (for so little improvement);
#
# Byte order [in]dependence. =========================================
#
# Caller is expected to maintain specific *dword* order in Htable,
# namely with *least* significant dword of 128-bit value at *lower*
# address. This differs completely from C code and has everything to
# do with ldm instruction and order in which dwords are "consumed" by
# algorithm. *Byte* order within these dwords in turn is whatever
# *native* byte order on current platform. See gcm128.c for working
# example...
Add assembly support for 32-bit iOS. (Imported from upstream's 313e6ec11fb8a7bda1676ce5804bee8755664141) BUG=338886 Change-Id: Id635e78b9afaad5ca311e3aeed888c9aedeb9637 Reviewed-on: https://boringssl-review.googlesource.com/4490 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:07:38 +01:00
$flavour = shift;
Switch perlasm calling convention. Depending on architecture, perlasm differed on which one or both of: perl foo.pl flavor output.S perl foo.pl flavor > output.S Upstream has now unified on the first form after making a number of changes to their files (the second does not even work for their x86 files anymore). Sync those portions of our perlasm scripts with upstream and update CMakeLists.txt and generate_build_files.py per the new convention. This imports various commits like this one: 184bc45f683c76531d7e065b6553ca9086564576 (this was done by taking a diff, so I don't have the full list) Confirmed that generate_build_files.py sees no change. BUG=14 Change-Id: Id2fb5b8bc2a7369d077221b5df9a6947d41f50d2 Reviewed-on: https://boringssl-review.googlesource.com/8518 Reviewed-by: Adam Langley <agl@google.com>
2016-06-26 18:18:50 +01:00
if ($flavour=~/\w[\w\-]*\.\w+$/) { $output=$flavour; undef $flavour; }
else { while (($output=shift) && ($output!~/\w[\w\-]*\.\w+$/)) {} }
Add assembly support for 32-bit iOS. (Imported from upstream's 313e6ec11fb8a7bda1676ce5804bee8755664141) BUG=338886 Change-Id: Id635e78b9afaad5ca311e3aeed888c9aedeb9637 Reviewed-on: https://boringssl-review.googlesource.com/4490 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:07:38 +01:00
if ($flavour && $flavour ne "void") {
$0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
( $xlate="${dir}arm-xlate.pl" and -f $xlate ) or
Move modes/ into the FIPS module The changes to delocate.go are needed because modes/ does things like return the address of a module function. Both of these need to be changed from referencing the GOT to using local symbols. Rather than testing whether |ghash| is |gcm_ghash_avx|, we can just keep that information in a flag. The test for |aesni_ctr32_encrypt_blocks| is more problematic, but I believe that it's superfluous and can be dropped: if you passed in a stream function that was semantically different from |aesni_ctr32_encrypt_blocks| you would already have a bug because |CRYPTO_gcm128_[en|de]crypt_ctr32| will handle a block at the end themselves, and assume a big-endian, 32-bit counter anyway. Change-Id: I68a84ebdab6c6006e11e9467e3362d7585461385 Reviewed-on: https://boringssl-review.googlesource.com/15064 Reviewed-by: Adam Langley <agl@google.com>
2017-04-13 19:38:40 +01:00
( $xlate="${dir}../../../perlasm/arm-xlate.pl" and -f $xlate) or
Add assembly support for 32-bit iOS. (Imported from upstream's 313e6ec11fb8a7bda1676ce5804bee8755664141) BUG=338886 Change-Id: Id635e78b9afaad5ca311e3aeed888c9aedeb9637 Reviewed-on: https://boringssl-review.googlesource.com/4490 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:07:38 +01:00
die "can't locate arm-xlate.pl";
open STDOUT,"| \"$^X\" $xlate $flavour $output";
} else {
open STDOUT,">$output";
}
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
$Xi="r0"; # argument block
$Htbl="r1";
$inp="r2";
$len="r3";
$Zll="r4"; # variables
$Zlh="r5";
$Zhl="r6";
$Zhh="r7";
$Tll="r8";
$Tlh="r9";
$Thl="r10";
$Thh="r11";
$nlo="r12";
################# r13 is stack pointer
$nhi="r14";
################# r15 is program counter
$rem_4bit=$inp; # used in gcm_gmult_4bit
$cnt=$len;
sub Zsmash() {
my $i=12;
my @args=@_;
for ($Zll,$Zlh,$Zhl,$Zhh) {
$code.=<<___;
#if __ARM_ARCH__>=7 && defined(__ARMEL__)
rev $_,$_
str $_,[$Xi,#$i]
#elif defined(__ARMEB__)
str $_,[$Xi,#$i]
#else
mov $Tlh,$_,lsr#8
strb $_,[$Xi,#$i+3]
mov $Thl,$_,lsr#16
strb $Tlh,[$Xi,#$i+2]
mov $Thh,$_,lsr#24
strb $Thl,[$Xi,#$i+1]
strb $Thh,[$Xi,#$i]
#endif
___
$code.="\t".shift(@args)."\n";
$i-=4;
}
}
$code=<<___;
Move arm_arch.h and fix up lots of include paths. arm_arch.h is included from ARM asm files, but lives in crypto/, not openssl/include/. Since the asm files are often built from a different location than their position in the source tree, relative include paths are unlikely to work so, rather than having crypto/ be a de-facto, second global include path, this change moves arm_arch.h to include/openssl/. It also removes entries from many include paths because they should be needed as relative includes are always based on the locations of the source file. Change-Id: I638ff43d641ca043a4fc06c0d901b11c6ff73542 Reviewed-on: https://boringssl-review.googlesource.com/5746 Reviewed-by: Adam Langley <agl@google.com>
2015-08-25 02:03:17 +01:00
#include <openssl/arm_arch.h>
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
.text
Sync ARM assembly up to 609b0852e4d50251857dbbac3141ba042e35a9ae. This change was made by copying over the files as of that commit and then discarding the parts of the diff which corresponding to our own changes. Change-Id: I28c5d711f7a8cec30749b8174687434129af5209 Reviewed-on: https://boringssl-review.googlesource.com/17111 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:53:40 +01:00
#if defined(__thumb2__) || defined(__clang__)
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
.syntax unified
Sync ARM assembly up to 609b0852e4d50251857dbbac3141ba042e35a9ae. This change was made by copying over the files as of that commit and then discarding the parts of the diff which corresponding to our own changes. Change-Id: I28c5d711f7a8cec30749b8174687434129af5209 Reviewed-on: https://boringssl-review.googlesource.com/17111 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:53:40 +01:00
#endif
#if defined(__thumb2__)
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
.thumb
#else
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
.code 32
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
#endif
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
modes/asm/ghash-armv4.pl: extend Apple fix to all clang cases. Triggered by RT#3989. (Imported from upstream's fbab8baddef8d3346ae40ff068871e2ddaf10270. This doesn't seem to affect us, but avoid getting out of sync.) Change-Id: I164e2a72e4b75e286ceaa03745ed9bcbf6c3e32e Reviewed-on: https://boringssl-review.googlesource.com/6512 Reviewed-by: Adam Langley <agl@google.com>
2015-11-16 17:49:54 +00:00
#ifdef __clang__
Add assembly support for 32-bit iOS. (Imported from upstream's 313e6ec11fb8a7bda1676ce5804bee8755664141) BUG=338886 Change-Id: Id635e78b9afaad5ca311e3aeed888c9aedeb9637 Reviewed-on: https://boringssl-review.googlesource.com/4490 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:07:38 +01:00
#define ldrplb ldrbpl
#define ldrneb ldrbne
#endif
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
.type rem_4bit,%object
.align 5
rem_4bit:
.short 0x0000,0x1C20,0x3840,0x2460
.short 0x7080,0x6CA0,0x48C0,0x54E0
.short 0xE100,0xFD20,0xD940,0xC560
.short 0x9180,0x8DA0,0xA9C0,0xB5E0
.size rem_4bit,.-rem_4bit
.type rem_4bit_get,%function
rem_4bit_get:
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
#if defined(__thumb2__)
adr $rem_4bit,rem_4bit
#else
sub $rem_4bit,pc,#8+32 @ &rem_4bit
#endif
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
b .Lrem_4bit_got
nop
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
nop
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
.size rem_4bit_get,.-rem_4bit_get
.global gcm_ghash_4bit
.type gcm_ghash_4bit,%function
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
.align 4
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
gcm_ghash_4bit:
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
#if defined(__thumb2__)
adr r12,rem_4bit
#else
sub r12,pc,#8+48 @ &rem_4bit
#endif
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
add $len,$inp,$len @ $len to point at the end
stmdb sp!,{r3-r11,lr} @ save $len/end too
ldmia r12,{r4-r11} @ copy rem_4bit ...
stmdb sp!,{r4-r11} @ ... to stack
ldrb $nlo,[$inp,#15]
ldrb $nhi,[$Xi,#15]
.Louter:
eor $nlo,$nlo,$nhi
and $nhi,$nlo,#0xf0
and $nlo,$nlo,#0x0f
mov $cnt,#14
add $Zhh,$Htbl,$nlo,lsl#4
ldmia $Zhh,{$Zll-$Zhh} @ load Htbl[nlo]
add $Thh,$Htbl,$nhi
ldrb $nlo,[$inp,#14]
and $nhi,$Zll,#0xf @ rem
ldmia $Thh,{$Tll-$Thh} @ load Htbl[nhi]
add $nhi,$nhi,$nhi
eor $Zll,$Tll,$Zll,lsr#4
ldrh $Tll,[sp,$nhi] @ rem_4bit[rem]
eor $Zll,$Zll,$Zlh,lsl#28
ldrb $nhi,[$Xi,#14]
eor $Zlh,$Tlh,$Zlh,lsr#4
eor $Zlh,$Zlh,$Zhl,lsl#28
eor $Zhl,$Thl,$Zhl,lsr#4
eor $Zhl,$Zhl,$Zhh,lsl#28
eor $Zhh,$Thh,$Zhh,lsr#4
eor $nlo,$nlo,$nhi
and $nhi,$nlo,#0xf0
and $nlo,$nlo,#0x0f
eor $Zhh,$Zhh,$Tll,lsl#16
.Linner:
add $Thh,$Htbl,$nlo,lsl#4
and $nlo,$Zll,#0xf @ rem
subs $cnt,$cnt,#1
add $nlo,$nlo,$nlo
ldmia $Thh,{$Tll-$Thh} @ load Htbl[nlo]
eor $Zll,$Tll,$Zll,lsr#4
eor $Zll,$Zll,$Zlh,lsl#28
eor $Zlh,$Tlh,$Zlh,lsr#4
eor $Zlh,$Zlh,$Zhl,lsl#28
ldrh $Tll,[sp,$nlo] @ rem_4bit[rem]
eor $Zhl,$Thl,$Zhl,lsr#4
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
#ifdef __thumb2__
it pl
#endif
Revert "Use unified ARM assembly." This reverts commit 2cd63877b5d663d1f19dca08b36de9746c044f9d. We've since imported a change to upstream which adds some #defines that should do the same thing on clang. (Though if gas accepts unified assembly too, that does seem a better approach. Ah well. Diverging on these files is expensive.) This is to reduce the diff and make applying some subsequent changes easier. Change-Id: I3f5eae2a71919b291a8de9415b894d8f0c67e3cf Reviewed-on: https://boringssl-review.googlesource.com/17107 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:10:33 +01:00
ldrplb $nlo,[$inp,$cnt]
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
eor $Zhl,$Zhl,$Zhh,lsl#28
eor $Zhh,$Thh,$Zhh,lsr#4
add $Thh,$Htbl,$nhi
and $nhi,$Zll,#0xf @ rem
eor $Zhh,$Zhh,$Tll,lsl#16 @ ^= rem_4bit[rem]
add $nhi,$nhi,$nhi
ldmia $Thh,{$Tll-$Thh} @ load Htbl[nhi]
eor $Zll,$Tll,$Zll,lsr#4
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
#ifdef __thumb2__
it pl
#endif
Revert "Use unified ARM assembly." This reverts commit 2cd63877b5d663d1f19dca08b36de9746c044f9d. We've since imported a change to upstream which adds some #defines that should do the same thing on clang. (Though if gas accepts unified assembly too, that does seem a better approach. Ah well. Diverging on these files is expensive.) This is to reduce the diff and make applying some subsequent changes easier. Change-Id: I3f5eae2a71919b291a8de9415b894d8f0c67e3cf Reviewed-on: https://boringssl-review.googlesource.com/17107 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:10:33 +01:00
ldrplb $Tll,[$Xi,$cnt]
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
eor $Zll,$Zll,$Zlh,lsl#28
eor $Zlh,$Tlh,$Zlh,lsr#4
ldrh $Tlh,[sp,$nhi]
eor $Zlh,$Zlh,$Zhl,lsl#28
eor $Zhl,$Thl,$Zhl,lsr#4
eor $Zhl,$Zhl,$Zhh,lsl#28
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
#ifdef __thumb2__
it pl
#endif
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
eorpl $nlo,$nlo,$Tll
eor $Zhh,$Thh,$Zhh,lsr#4
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
#ifdef __thumb2__
itt pl
#endif
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
andpl $nhi,$nlo,#0xf0
andpl $nlo,$nlo,#0x0f
eor $Zhh,$Zhh,$Tlh,lsl#16 @ ^= rem_4bit[rem]
bpl .Linner
ldr $len,[sp,#32] @ re-load $len/end
add $inp,$inp,#16
mov $nhi,$Zll
___
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
&Zsmash("cmp\t$inp,$len","\n".
"#ifdef __thumb2__\n".
" it ne\n".
"#endif\n".
" ldrneb $nlo,[$inp,#15]");
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
$code.=<<___;
bne .Louter
add sp,sp,#36
#if __ARM_ARCH__>=5
ldmia sp!,{r4-r11,pc}
#else
ldmia sp!,{r4-r11,lr}
tst lr,#1
moveq pc,lr @ be binary compatible with V4, yet
bx lr @ interoperable with Thumb ISA:-)
#endif
.size gcm_ghash_4bit,.-gcm_ghash_4bit
.global gcm_gmult_4bit
.type gcm_gmult_4bit,%function
gcm_gmult_4bit:
stmdb sp!,{r4-r11,lr}
ldrb $nlo,[$Xi,#15]
b rem_4bit_get
.Lrem_4bit_got:
and $nhi,$nlo,#0xf0
and $nlo,$nlo,#0x0f
mov $cnt,#14
add $Zhh,$Htbl,$nlo,lsl#4
ldmia $Zhh,{$Zll-$Zhh} @ load Htbl[nlo]
ldrb $nlo,[$Xi,#14]
add $Thh,$Htbl,$nhi
and $nhi,$Zll,#0xf @ rem
ldmia $Thh,{$Tll-$Thh} @ load Htbl[nhi]
add $nhi,$nhi,$nhi
eor $Zll,$Tll,$Zll,lsr#4
ldrh $Tll,[$rem_4bit,$nhi] @ rem_4bit[rem]
eor $Zll,$Zll,$Zlh,lsl#28
eor $Zlh,$Tlh,$Zlh,lsr#4
eor $Zlh,$Zlh,$Zhl,lsl#28
eor $Zhl,$Thl,$Zhl,lsr#4
eor $Zhl,$Zhl,$Zhh,lsl#28
eor $Zhh,$Thh,$Zhh,lsr#4
and $nhi,$nlo,#0xf0
eor $Zhh,$Zhh,$Tll,lsl#16
and $nlo,$nlo,#0x0f
.Loop:
add $Thh,$Htbl,$nlo,lsl#4
and $nlo,$Zll,#0xf @ rem
subs $cnt,$cnt,#1
add $nlo,$nlo,$nlo
ldmia $Thh,{$Tll-$Thh} @ load Htbl[nlo]
eor $Zll,$Tll,$Zll,lsr#4
eor $Zll,$Zll,$Zlh,lsl#28
eor $Zlh,$Tlh,$Zlh,lsr#4
eor $Zlh,$Zlh,$Zhl,lsl#28
ldrh $Tll,[$rem_4bit,$nlo] @ rem_4bit[rem]
eor $Zhl,$Thl,$Zhl,lsr#4
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
#ifdef __thumb2__
it pl
#endif
Revert "Use unified ARM assembly." This reverts commit 2cd63877b5d663d1f19dca08b36de9746c044f9d. We've since imported a change to upstream which adds some #defines that should do the same thing on clang. (Though if gas accepts unified assembly too, that does seem a better approach. Ah well. Diverging on these files is expensive.) This is to reduce the diff and make applying some subsequent changes easier. Change-Id: I3f5eae2a71919b291a8de9415b894d8f0c67e3cf Reviewed-on: https://boringssl-review.googlesource.com/17107 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:10:33 +01:00
ldrplb $nlo,[$Xi,$cnt]
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
eor $Zhl,$Zhl,$Zhh,lsl#28
eor $Zhh,$Thh,$Zhh,lsr#4
add $Thh,$Htbl,$nhi
and $nhi,$Zll,#0xf @ rem
eor $Zhh,$Zhh,$Tll,lsl#16 @ ^= rem_4bit[rem]
add $nhi,$nhi,$nhi
ldmia $Thh,{$Tll-$Thh} @ load Htbl[nhi]
eor $Zll,$Tll,$Zll,lsr#4
eor $Zll,$Zll,$Zlh,lsl#28
eor $Zlh,$Tlh,$Zlh,lsr#4
ldrh $Tll,[$rem_4bit,$nhi] @ rem_4bit[rem]
eor $Zlh,$Zlh,$Zhl,lsl#28
eor $Zhl,$Thl,$Zhl,lsr#4
eor $Zhl,$Zhl,$Zhh,lsl#28
eor $Zhh,$Thh,$Zhh,lsr#4
ARMv4 assembly pack: implement support for Thumb2. As some of ARM processors, more specifically Cortex-Mx series, are Thumb2-only, we need to support Thumb2-only builds even in assembly. (Imported from upstream's 11208dcfb9105e8afa37233185decefd45e89e17.) Change-Id: I7cb48ce6a842cf3cfdf553f6e6e6227d52d525c0 Reviewed-on: https://boringssl-review.googlesource.com/17108 Reviewed-by: Adam Langley <agl@google.com>
2017-06-12 23:31:15 +01:00
#ifdef __thumb2__
itt pl
#endif
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
andpl $nhi,$nlo,#0xf0
andpl $nlo,$nlo,#0x0f
eor $Zhh,$Zhh,$Tll,lsl#16 @ ^= rem_4bit[rem]
bpl .Loop
___
&Zsmash();
$code.=<<___;
#if __ARM_ARCH__>=5
ldmia sp!,{r4-r11,pc}
#else
ldmia sp!,{r4-r11,lr}
tst lr,#1
moveq pc,lr @ be binary compatible with V4, yet
bx lr @ interoperable with Thumb ISA:-)
#endif
.size gcm_gmult_4bit,.-gcm_gmult_4bit
___
{
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
my ($Xl,$Xm,$Xh,$IN)=map("q$_",(0..3));
my ($t0,$t1,$t2,$t3)=map("q$_",(8..12));
my ($Hlo,$Hhi,$Hhl,$k48,$k32,$k16)=map("d$_",(26..31));
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
sub clmul64x64 {
my ($r,$a,$b)=@_;
$code.=<<___;
vext.8 $t0#lo, $a, $a, #1 @ A1
vmull.p8 $t0, $t0#lo, $b @ F = A1*B
vext.8 $r#lo, $b, $b, #1 @ B1
vmull.p8 $r, $a, $r#lo @ E = A*B1
vext.8 $t1#lo, $a, $a, #2 @ A2
vmull.p8 $t1, $t1#lo, $b @ H = A2*B
vext.8 $t3#lo, $b, $b, #2 @ B2
vmull.p8 $t3, $a, $t3#lo @ G = A*B2
vext.8 $t2#lo, $a, $a, #3 @ A3
veor $t0, $t0, $r @ L = E + F
vmull.p8 $t2, $t2#lo, $b @ J = A3*B
vext.8 $r#lo, $b, $b, #3 @ B3
veor $t1, $t1, $t3 @ M = G + H
vmull.p8 $r, $a, $r#lo @ I = A*B3
veor $t0#lo, $t0#lo, $t0#hi @ t0 = (L) (P0 + P1) << 8
vand $t0#hi, $t0#hi, $k48
vext.8 $t3#lo, $b, $b, #4 @ B4
veor $t1#lo, $t1#lo, $t1#hi @ t1 = (M) (P2 + P3) << 16
vand $t1#hi, $t1#hi, $k32
vmull.p8 $t3, $a, $t3#lo @ K = A*B4
veor $t2, $t2, $r @ N = I + J
veor $t0#lo, $t0#lo, $t0#hi
veor $t1#lo, $t1#lo, $t1#hi
veor $t2#lo, $t2#lo, $t2#hi @ t2 = (N) (P4 + P5) << 24
vand $t2#hi, $t2#hi, $k16
vext.8 $t0, $t0, $t0, #15
veor $t3#lo, $t3#lo, $t3#hi @ t3 = (K) (P6 + P7) << 32
vmov.i64 $t3#hi, #0
vext.8 $t1, $t1, $t1, #14
veor $t2#lo, $t2#lo, $t2#hi
vmull.p8 $r, $a, $b @ D = A*B
vext.8 $t3, $t3, $t3, #12
vext.8 $t2, $t2, $t2, #13
veor $t0, $t0, $t1
veor $t2, $t2, $t3
veor $r, $r, $t0
veor $r, $r, $t2
___
}
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
$code.=<<___;
Remove inconsistency in ARM support. This facilitates "universal" builds, ones that target multiple architectures, e.g. ARMv5 through ARMv7. (Imported from upstream's c1669e1c205dc8e695fb0c10a655f434e758b9f7) This is a change from a while ago which was a source of divergence between our perlasm and upstream's. This change in upstream came with the following comment in Configure: Note that -march is not among compiler options in below linux-armv4 target line. Not specifying one is intentional to give you choice to: a) rely on your compiler default by not specifying one; b) specify your target platform explicitly for optimal performance, e.g. -march=armv6 or -march=armv7-a; c) build "universal" binary that targets *range* of platforms by specifying minimum and maximum supported architecture; As for c) option. It actually makes no sense to specify maximum to be less than ARMv7, because it's the least requirement for run-time switch between platform-specific code paths. And without run-time switch performance would be equivalent to one for minimum. Secondly, there are some natural limitations that you'd have to accept and respect. Most notably you can *not* build "universal" binary for big-endian platform. This is because ARMv7 processor always picks instructions in little-endian order. Another similar limitation is that -mthumb can't "cross" -march=armv6t2 boundary, because that's where it became Thumb-2. Well, this limitation is a bit artificial, because it's not really impossible, but it's deemed too tricky to support. And of course you have to be sure that your binutils are actually up to the task of handling maximum target platform. Change-Id: Ie5f674d603393f0a1354a0d0973987484a4a650c Reviewed-on: https://boringssl-review.googlesource.com/4488 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:27:38 +01:00
#if __ARM_MAX_ARCH__>=7
.arch armv7-a
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
.fpu neon
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
.global gcm_init_neon
.type gcm_init_neon,%function
.align 4
gcm_init_neon:
Add assembly support for 32-bit iOS. (Imported from upstream's 313e6ec11fb8a7bda1676ce5804bee8755664141) BUG=338886 Change-Id: Id635e78b9afaad5ca311e3aeed888c9aedeb9637 Reviewed-on: https://boringssl-review.googlesource.com/4490 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:07:38 +01:00
vld1.64 $IN#hi,[r1]! @ load H
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
vmov.i8 $t0,#0xe1
Add assembly support for 32-bit iOS. (Imported from upstream's 313e6ec11fb8a7bda1676ce5804bee8755664141) BUG=338886 Change-Id: Id635e78b9afaad5ca311e3aeed888c9aedeb9637 Reviewed-on: https://boringssl-review.googlesource.com/4490 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:07:38 +01:00
vld1.64 $IN#lo,[r1]
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
vshl.i64 $t0#hi,#57
vshr.u64 $t0#lo,#63 @ t0=0xc2....01
vdup.8 $t1,$IN#hi[7]
vshr.u64 $Hlo,$IN#lo,#63
vshr.s8 $t1,#7 @ broadcast carry bit
vshl.i64 $IN,$IN,#1
vand $t0,$t0,$t1
vorr $IN#hi,$Hlo @ H<<<=1
veor $IN,$IN,$t0 @ twisted H
vstmia r0,{$IN}
ARM assembly pack: get ARMv7 instruction endianness right. Pointer out and suggested by: Ard Biesheuvel. (Imported from upstream's 5dcf70a1c57c2019bfad640fe14fd4a73212860a) This is from a while ago, but it's one source of divergence between our copy of these files and master's. Change-Id: I6525a27f25eb86a92420c32996af47ecc42ee020 Reviewed-on: https://boringssl-review.googlesource.com/4487 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:21:51 +01:00
ret @ bx lr
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
.size gcm_init_neon,.-gcm_init_neon
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
.global gcm_gmult_neon
.type gcm_gmult_neon,%function
.align 4
gcm_gmult_neon:
Add assembly support for 32-bit iOS. (Imported from upstream's 313e6ec11fb8a7bda1676ce5804bee8755664141) BUG=338886 Change-Id: Id635e78b9afaad5ca311e3aeed888c9aedeb9637 Reviewed-on: https://boringssl-review.googlesource.com/4490 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:07:38 +01:00
vld1.64 $IN#hi,[$Xi]! @ load Xi
vld1.64 $IN#lo,[$Xi]!
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
vmov.i64 $k48,#0x0000ffffffffffff
vldmia $Htbl,{$Hlo-$Hhi} @ load twisted H
vmov.i64 $k32,#0x00000000ffffffff
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
#ifdef __ARMEL__
vrev64.8 $IN,$IN
#endif
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
vmov.i64 $k16,#0x000000000000ffff
veor $Hhl,$Hlo,$Hhi @ Karatsuba pre-processing
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
mov $len,#16
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
b .Lgmult_neon
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
.size gcm_gmult_neon,.-gcm_gmult_neon
.global gcm_ghash_neon
.type gcm_ghash_neon,%function
.align 4
gcm_ghash_neon:
Add assembly support for 32-bit iOS. (Imported from upstream's 313e6ec11fb8a7bda1676ce5804bee8755664141) BUG=338886 Change-Id: Id635e78b9afaad5ca311e3aeed888c9aedeb9637 Reviewed-on: https://boringssl-review.googlesource.com/4490 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:07:38 +01:00
vld1.64 $Xl#hi,[$Xi]! @ load Xi
vld1.64 $Xl#lo,[$Xi]!
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
vmov.i64 $k48,#0x0000ffffffffffff
vldmia $Htbl,{$Hlo-$Hhi} @ load twisted H
vmov.i64 $k32,#0x00000000ffffffff
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
#ifdef __ARMEL__
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
vrev64.8 $Xl,$Xl
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
#endif
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
vmov.i64 $k16,#0x000000000000ffff
veor $Hhl,$Hlo,$Hhi @ Karatsuba pre-processing
.Loop_neon:
vld1.64 $IN#hi,[$inp]! @ load inp
vld1.64 $IN#lo,[$inp]!
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
#ifdef __ARMEL__
vrev64.8 $IN,$IN
#endif
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
veor $IN,$Xl @ inp^=Xi
.Lgmult_neon:
___
Convert remaining Latin-1 files to UTF-8. See upstream's 9f0b86c68bb96d49301bbd6473c8235ca05ca06b. Generated by using upstream's script in 5a3ce86e21715a683ff0d32421ed5c6d5e84234d and then manually throwing out the false positives. (We converted a bunch of stuff already in 91157550061d5d794898fe47b95384a7ba5f7b9d.) This may require some wrestling with depot_tools to land in Chromium due to Rietveld's encoding bugs, but hopefully that will avoid future problems; Rietveld breaks if either old or new file is Latin-1. Change-Id: I26dcb20c7377f92a0c843ef5d74d440a82ea8ceb Reviewed-on: https://boringssl-review.googlesource.com/5483 Reviewed-by: Adam Langley <agl@google.com>
2015-07-22 03:29:26 +01:00
&clmul64x64 ($Xl,$Hlo,"$IN#lo"); # H.lo·Xi.lo
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
$code.=<<___;
veor $IN#lo,$IN#lo,$IN#hi @ Karatsuba pre-processing
___
Convert remaining Latin-1 files to UTF-8. See upstream's 9f0b86c68bb96d49301bbd6473c8235ca05ca06b. Generated by using upstream's script in 5a3ce86e21715a683ff0d32421ed5c6d5e84234d and then manually throwing out the false positives. (We converted a bunch of stuff already in 91157550061d5d794898fe47b95384a7ba5f7b9d.) This may require some wrestling with depot_tools to land in Chromium due to Rietveld's encoding bugs, but hopefully that will avoid future problems; Rietveld breaks if either old or new file is Latin-1. Change-Id: I26dcb20c7377f92a0c843ef5d74d440a82ea8ceb Reviewed-on: https://boringssl-review.googlesource.com/5483 Reviewed-by: Adam Langley <agl@google.com>
2015-07-22 03:29:26 +01:00
&clmul64x64 ($Xm,$Hhl,"$IN#lo"); # (H.lo+H.hi)·(Xi.lo+Xi.hi)
&clmul64x64 ($Xh,$Hhi,"$IN#hi"); # H.hi·Xi.hi
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
$code.=<<___;
veor $Xm,$Xm,$Xl @ Karatsuba post-processing
veor $Xm,$Xm,$Xh
veor $Xl#hi,$Xl#hi,$Xm#lo
veor $Xh#lo,$Xh#lo,$Xm#hi @ Xh|Xl - 256-bit result
@ equivalent of reduction_avx from ghash-x86_64.pl
vshl.i64 $t1,$Xl,#57 @ 1st phase
vshl.i64 $t2,$Xl,#62
veor $t2,$t2,$t1 @
vshl.i64 $t1,$Xl,#63
veor $t2, $t2, $t1 @
veor $Xl#hi,$Xl#hi,$t2#lo @
veor $Xh#lo,$Xh#lo,$t2#hi
vshr.u64 $t2,$Xl,#1 @ 2nd phase
veor $Xh,$Xh,$Xl
veor $Xl,$Xl,$t2 @
vshr.u64 $t2,$t2,#6
vshr.u64 $Xl,$Xl,#1 @
veor $Xl,$Xl,$Xh @
veor $Xl,$Xl,$t2 @
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
subs $len,#16
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
bne .Loop_neon
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
#ifdef __ARMEL__
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
vrev64.8 $Xl,$Xl
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
#endif
Remove trailing whitespace from Perl files. Upstream did this in 609b0852e4d50251857dbbac3141ba042e35a9ae and it's easier to apply patches if we do also. Change-Id: I5142693ed1e26640987ff16f5ea510e81bba200e Reviewed-on: https://boringssl-review.googlesource.com/13771 Commit-Queue: Adam Langley <agl@google.com> Reviewed-by: David Benjamin <davidben@google.com>
2017-02-09 20:21:08 +00:00
sub $Xi,#16
Add assembly support for 32-bit iOS. (Imported from upstream's 313e6ec11fb8a7bda1676ce5804bee8755664141) BUG=338886 Change-Id: Id635e78b9afaad5ca311e3aeed888c9aedeb9637 Reviewed-on: https://boringssl-review.googlesource.com/4490 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:07:38 +01:00
vst1.64 $Xl#hi,[$Xi]! @ write out Xi
vst1.64 $Xl#lo,[$Xi]
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
ARM assembly pack: get ARMv7 instruction endianness right. Pointer out and suggested by: Ard Biesheuvel. (Imported from upstream's 5dcf70a1c57c2019bfad640fe14fd4a73212860a) This is from a while ago, but it's one source of divergence between our copy of these files and master's. Change-Id: I6525a27f25eb86a92420c32996af47ecc42ee020 Reviewed-on: https://boringssl-review.googlesource.com/4487 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:21:51 +01:00
ret @ bx lr
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
.size gcm_ghash_neon,.-gcm_ghash_neon
#endif
___
}
$code.=<<___;
.asciz "GHASH for ARMv4/NEON, CRYPTOGAMS by <appro\@openssl.org>"
.align 2
___
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
foreach (split("\n",$code)) {
s/\`([^\`]*)\`/eval $1/geo;
s/\bq([0-9]+)#(lo|hi)/sprintf "d%d",2*$1+($2 eq "hi")/geo or
ARM assembly pack: get ARMv7 instruction endianness right. Pointer out and suggested by: Ard Biesheuvel. (Imported from upstream's 5dcf70a1c57c2019bfad640fe14fd4a73212860a) This is from a while ago, but it's one source of divergence between our copy of these files and master's. Change-Id: I6525a27f25eb86a92420c32996af47ecc42ee020 Reviewed-on: https://boringssl-review.googlesource.com/4487 Reviewed-by: Adam Langley <agl@google.com>
2015-04-21 02:21:51 +01:00
s/\bret\b/bx lr/go or
bn/asm/armv4-gf2m.pl, modes/asm/ghash-armv4.pl: faster multiplication algorithm suggested in following paper: Câmara, D.; Gouvêa, C. P. L.; López, J. & Dahab, R.: Fast Software Polynomial Multiplication on ARM Processors using the NEON Engine. http://conradoplg.cryptoland.net/files/2010/12/mocrysen13.pdf (Imported from upstream's 0fb3d5b4fdc76b8d4a4700d03480cda135c6c117)
2014-06-20 20:00:00 +01:00
s/\bbx\s+lr\b/.word\t0xe12fff1e/go; # make it possible to compile with -march=armv4
print $_,"\n";
}
Inital import. Initial fork from f2d678e6e89b6508147086610e985d4e8416e867 (1.0.2 beta). (This change contains substantial changes from the original and effectively starts a new history.)
2014-06-20 20:00:00 +01:00
close STDOUT; # enforce flush
Reference in New Issue Copy Permalink