X-Git-Url: https://git.librecmc.org/?a=blobdiff_plain;f=crypto%2Fbn%2Fasm%2Fvis3-mont.pl;h=827679b396b40a6ca778dd093db55f9653fe537a;hb=32be631ca1f2b73c92e4f7f5d23f1c3aee80ec69;hp=64dba4480fbe99da5c9b77cf6e795ea480751307;hpb=6aa36e8e5a062e31543e7796f0351ff9628832ce;p=oweals%2Fopenssl.git

diff --git a/crypto/bn/asm/vis3-mont.pl b/crypto/bn/asm/vis3-mont.pl
index 64dba4480f..827679b396 100644
--- a/crypto/bn/asm/vis3-mont.pl
+++ b/crypto/bn/asm/vis3-mont.pl
@@ -1,7 +1,7 @@
 #! /usr/bin/env perl
-# Copyright 2012-2016 The OpenSSL Project Authors. All Rights Reserved.
+# Copyright 2012-2018 The OpenSSL Project Authors. All Rights Reserved.
 #
-# Licensed under the OpenSSL license (the "License").  You may not use
+# 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
@@ -16,7 +16,7 @@
 
 # October 2012.
 #
-# SPARCv9 VIS3 Montgomery multiplicaion procedure suitable for T3 and
+# SPARCv9 VIS3 Montgomery multiplication procedure suitable for T3 and
 # onward. There are three new instructions used here: umulxhi,
 # addxc[cc] and initializing store. On T3 RSA private key operations
 # are 1.54/1.87/2.11/2.26 times faster for 512/1024/2048/4096-bit key
@@ -25,8 +25,7 @@
 # for reference purposes, because T4 has dedicated Montgomery
 # multiplication and squaring *instructions* that deliver even more.
 
-$output = pop;
-open STDOUT,">$output";
+$output = pop and open STDOUT,">$output";
 
 $frame = "STACK_FRAME";
 $bias = "STACK_BIAS";
@@ -310,23 +309,23 @@ $code.=<<___;
 	sub	$anp,	$num,	$anp
 	sub	$rp,	$num,	$rp
 
-	subc	$ovf,	%g0,	$ovf	! handle upmost overflow bit
-	and	$tp,	$ovf,	$ap
-	andn	$rp,	$ovf,	$np
-	or	$np,	$ap,	$ap	! ap=borrow?tp:rp
+	subccc	$ovf,	%g0,	$ovf	! handle upmost overflow bit
 	ba	.Lcopy
 	sub	$num,	8,	$cnt
 
 .align	16
-.Lcopy:					! copy or in-place refresh
-	ld	[$ap+0],	$t2
-	ld	[$ap+4],	$t3
-	add	$ap,	8,	$ap
+.Lcopy:					! conditional copy
+	ld	[$tp+0],	$t0
+	ld	[$tp+4],	$t1
+	ld	[$rp+0],	$t2
+	ld	[$rp+4],	$t3
 	stx	%g0,	[$tp]		! zap
 	add	$tp,	8,	$tp
 	stx	%g0,	[$anp]		! zap
 	stx	%g0,	[$anp+8]
 	add	$anp,	16,	$anp
+	movcs	%icc,	$t0,	$t2
+	movcs	%icc,	$t1,	$t3
 	st	$t3,	[$rp+0]		! flip order
 	st	$t2,	[$rp+4]
 	add	$rp,	8,	$rp
@@ -381,4 +380,4 @@ foreach (split("\n",$code)) {
 	print $_,"\n";
 }
 
-close STDOUT;
+close STDOUT or die "error closing STDOUT";