Check for the presence of _WIN32 rather than its value.

[oweals/openssl.git] / crypto / bn / asm / sparcv9-mont.pl

diff --git a/crypto/bn/asm/sparcv9-mont.pl b/crypto/bn/asm/sparcv9-mont.pl
index acdf6928ca52c1c65b93861364aa90faafcb4e70..c36ce36806788a6552f7f61158f4cbdeb3c22099 100644 (file)
--- a/crypto/bn/asm/sparcv9-mont.pl
+++ b/crypto/bn/asm/sparcv9-mont.pl
@@ -1,9 +1,17 @@
-#!/usr/bin/env perl
+#! /usr/bin/env perl
+# Copyright 2005-2016 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
+
 
 # ====================================================================
 # Written by Andy Polyakov <appro@fy.chalmers.se> for the OpenSSL
-# project. Rights for redistribution and usage in source and binary
-# forms are granted according to the OpenSSL license.
+# 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/.
 # ====================================================================
 
 # December 2005
@@ -12,7 +20,7 @@
 # for undertaken effort are multiple. First of all, UltraSPARC is not
 # the whole SPARCv9 universe and other VIS-free implementations deserve
 # optimized code as much. Secondly, newly introduced UltraSPARC T1,
-# a.k.a. Niagara, has shared FPU and concurrent FPU-intensive pathes,
+# a.k.a. Niagara, has shared FPU and concurrent FPU-intensive paths,
 # such as sparcv9a-mont, will simply sink it. Yes, T1 is equipped with
 # several integrated RSA/DSA accelerator circuits accessible through
 # kernel driver [only(*)], but having decent user-land software
@@ -41,6 +49,9 @@
 # module still have hidden potential [see TODO list there], which is
 # estimated to be larger than 20%...
 
+$output = pop;
+open STDOUT,">$output";
+
 # int bn_mul_mont(
 $rp="%i0";     # BN_ULONG *rp,
 $ap="%i1";     # const BN_ULONG *ap,
@@ -49,10 +60,8 @@ $np="%i3";   # const BN_ULONG *np,
 $n0="%i4";     # const BN_ULONG *n0,
 $num="%i5";    # int num);
 
-$bits=32;
-for (@ARGV)    { $bits=64 if (/\-m64/ || /\-xarch\=v9/); }
-if ($bits==64) { $bias=2047; $frame=192; }
-else           { $bias=0;    $frame=128; }
+$frame="STACK_FRAME";
+$bias="STACK_BIAS";
 
 $car0="%o0";
 $car1="%o1";
@@ -75,6 +84,8 @@ $tpj="%l7";
 $fname="bn_mul_mont_int";
 
 $code=<<___;
+#include "sparc_arch.h"
+
 .section       ".text",#alloc,#execinstr
 
 .global        $fname
@@ -104,7 +115,7 @@ $fname:
        ld      [$np],$car1             ! np[0]
        sub     %o7,$bias,%sp           ! alloca
        ld      [$np+4],$npj            ! np[1]
-       be,pt   `$bits==32?"%icc":"%xcc"`,.Lbn_sqr_mont
+       be,pt   SIZE_T_CC,.Lbn_sqr_mont
        mov     12,$j
 
        mulx    $car0,$mul0,$car0       ! ap[0]*bp[0]
@@ -254,44 +265,32 @@ $fname:
 .Ltail:
        add     $np,$num,$np
        add     $rp,$num,$rp
-
-       cmp     $car2,0                 ! clears %icc.c
-       bne,pn  %icc,.Lsub
+       mov     $tp,$ap
        sub     %g0,$num,%o7            ! k=-num
-
-       cmp     $car1,$npj              ! compare top-most $tp and $np words
-       bcs,pt  %icc,.Lcopy             ! %icc.c is clean if not taken
-       nop
-
-.align 16,0x1000000
+       ba      .Lsub
+       subcc   %g0,%g0,%g0             ! clear %icc.c
+.align 16
 .Lsub:
        ld      [$tp+%o7],%o0
        ld      [$np+%o7],%o1
-       subccc  %o0,%o1,%o1
-       st      %o1,[$rp+%o7]
+       subccc  %o0,%o1,%o1             ! tp[j]-np[j]
+       add     $rp,%o7,$i
        add     %o7,4,%o7
        brnz    %o7,.Lsub
-       nop
-       subccc  $car2,0,$car2
-       bcc     %icc,.Lzap
+       st      %o1,[$i]
+       subc    $car2,0,$car2           ! handle upmost overflow bit
+       and     $tp,$car2,$ap
+       andn    $rp,$car2,$np
+       or      $ap,$np,$ap
        sub     %g0,$num,%o7
 
-.align 16,0x1000000
 .Lcopy:
-       ld      [$tp+%o7],%o0
+       ld      [$ap+%o7],%o0           ! copy or in-place refresh
+       st      %g0,[$tp+%o7]           ! zap tp
        st      %o0,[$rp+%o7]
        add     %o7,4,%o7
        brnz    %o7,.Lcopy
        nop
-       ba      .Lzap
-       sub     %g0,$num,%o7
-
-.align 32
-.Lzap:
-       st      %g0,[$tp+%o7]
-       add     %o7,4,%o7
-       brnz    %o7,.Lzap
-       nop
        mov     1,%i0
        ret
        restore
@@ -609,6 +608,8 @@ $code.=<<___;
        add     $tp,8,$tp
 .type  $fname,#function
 .size  $fname,(.-$fname)
+.asciz "Montgomery Multipltication for SPARCv9, CRYPTOGAMS by <appro\@openssl.org>"
+.align 32
 ___
 $code =~ s/\`([^\`]*)\`/eval($1)/gem;
 print $code;