Add some missing cfi frame info in x86_64-mont5.pl

[openssl.git] / crypto / bn / asm / x86_64-mont5.pl

diff --git a/crypto/bn/asm/x86_64-mont5.pl b/crypto/bn/asm/x86_64-mont5.pl
index 1666fbd7a2d4a92b30754d30286a9e005e091f5a..a52ab5eb778eae9e77f1bcef230a99e47c59a6cd 100755 (executable)
--- a/crypto/bn/asm/x86_64-mont5.pl
+++ b/crypto/bn/asm/x86_64-mont5.pl
@@ -1,7 +1,7 @@
 #! /usr/bin/env perl
-# Copyright 2011-2016 The OpenSSL Project Authors. All Rights Reserved.
+# Copyright 2011-2018 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
@@ -31,9 +31,10 @@
 # the np argument is not just modulus value, but one interleaved
 # with 0. This is to optimize post-condition...
 
-$flavour = shift;
-$output  = shift;
-if ($flavour =~ /\./) { $output = $flavour; undef $flavour; }
+# $output is the last argument if it looks like a file (it has an extension)
+# $flavour is the first argument if it doesn't look like a file
+$output = $#ARGV >= 0 && $ARGV[$#ARGV] =~ m|\.\w+$| ? pop : undef;
+$flavour = $#ARGV >= 0 && $ARGV[0] !~ m|\.| ? shift : undef;
 
 $win64=0; $win64=1 if ($flavour =~ /[nm]asm|mingw64/ || $output =~ /\.asm$/);
 
@@ -42,7 +43,8 @@ $0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
 ( $xlate="${dir}../../perlasm/x86_64-xlate.pl" and -f $xlate) or
 die "can't locate x86_64-xlate.pl";
 
-open OUT,"| \"$^X\" \"$xlate\" $flavour \"$output\"";
+open OUT,"| \"$^X\" \"$xlate\" $flavour \"$output\""
+    or die "can't call $xlate: $!";
 *STDOUT=*OUT;
 
 if (`$ENV{CC} -Wa,-v -c -o /dev/null -x assembler /dev/null 2>&1`
@@ -419,22 +421,23 @@ $code.=<<___;
        mov     %rax,($rp,$i,8)         # rp[i]=tp[i]-np[i]
        mov     8($ap,$i,8),%rax        # tp[i+1]
        lea     1($i),$i                # i++
-       dec     $j                      # doesnn't affect CF!
+       dec     $j                      # doesn't affect CF!
        jnz     .Lsub
 
        sbb     \$0,%rax                # handle upmost overflow bit
+       mov     \$-1,%rbx
+       xor     %rax,%rbx
        xor     $i,$i
-       and     %rax,$ap
-       not     %rax
-       mov     $rp,$np
-       and     %rax,$np
        mov     $num,$j                 # j=num
-       or      $np,$ap                 # ap=borrow?tp:rp
-.align 16
-.Lcopy:                                        # copy or in-place refresh
-       mov     ($ap,$i,8),%rax
+
+.Lcopy:                                        # conditional copy
+       mov     ($rp,$i,8),%rcx
+       mov     (%rsp,$i,8),%rdx
+       and     %rbx,%rcx
+       and     %rax,%rdx
        mov     $i,(%rsp,$i,8)          # zap temporary vector
-       mov     %rax,($rp,$i,8)         # rp[i]=tp[i]
+       or      %rcx,%rdx
+       mov     %rdx,($rp,$i,8)         # rp[i]=tp[i]
        lea     1($i),$i
        sub     \$1,$j
        jnz     .Lcopy
@@ -579,6 +582,7 @@ $code.=<<___;
 .type  mul4x_internal,\@abi-omnipotent
 .align 32
 mul4x_internal:
+.cfi_startproc
        shl     \$5,$num                # $num was in bytes
        movd    `($win64?56:8)`(%rax),%xmm5     # load 7th argument, index
        lea     .Linc(%rip),%rax
@@ -1073,6 +1077,7 @@ $code.=<<___
 ___
 }
 $code.=<<___;
+.cfi_endproc
 .size  mul4x_internal,.-mul4x_internal
 ___
 }}}
@@ -1238,6 +1243,7 @@ $code.=<<___;
 .align 32
 bn_sqr8x_internal:
 __bn_sqr8x_internal:
+.cfi_startproc
        ##############################################################
        # Squaring part:
        #
@@ -2029,6 +2035,7 @@ __bn_sqr8x_reduction:
        cmp     %rdx,$tptr              # end of t[]?
        jb      .L8x_reduction_loop
        ret
+.cfi_endproc
 .size  bn_sqr8x_internal,.-bn_sqr8x_internal
 ___
 }\f
@@ -2041,6 +2048,7 @@ $code.=<<___;
 .type  __bn_post4x_internal,\@abi-omnipotent
 .align 32
 __bn_post4x_internal:
+.cfi_startproc
        mov     8*0($nptr),%r12
        lea     (%rdi,$num),$tptr       # %rdi was $tptr above
        mov     $num,%rcx
@@ -2091,6 +2099,7 @@ __bn_post4x_internal:
        mov     $num,%r10               # prepare for back-to-back call
        neg     $num                    # restore $num
        ret
+.cfi_endproc
 .size  __bn_post4x_internal,.-__bn_post4x_internal
 ___
 }
@@ -2100,10 +2109,12 @@ $code.=<<___;
 .type  bn_from_montgomery,\@abi-omnipotent
 .align 32
 bn_from_montgomery:
+.cfi_startproc
        testl   \$7,`($win64?"48(%rsp)":"%r9d")`
        jz      bn_from_mont8x
        xor     %eax,%eax
        ret
+.cfi_endproc
 .size  bn_from_montgomery,.-bn_from_montgomery
 
 .type  bn_from_mont8x,\@function,6
@@ -2399,6 +2410,7 @@ bn_mulx4x_mont_gather5:
 .type  mulx4x_internal,\@abi-omnipotent
 .align 32
 mulx4x_internal:
+.cfi_startproc
        mov     $num,8(%rsp)            # save -$num (it was in bytes)
        mov     $num,%r10
        neg     $num                    # restore $num
@@ -2421,7 +2433,7 @@ my $N=$STRIDE/4;          # should match cache line size
 $code.=<<___;
        movdqa  0(%rax),%xmm0           # 00000001000000010000000000000000
        movdqa  16(%rax),%xmm1          # 00000002000000020000000200000002
-       lea     88-112(%rsp,%r10),%r10  # place the mask after tp[num+1] (+ICache optimizaton)
+       lea     88-112(%rsp,%r10),%r10  # place the mask after tp[num+1] (+ICache optimization)
        lea     128($bp),$bptr          # size optimization
 
        pshufd  \$0,%xmm5,%xmm5         # broadcast index
@@ -2749,6 +2761,7 @@ $code.=<<___;
        mov     8*2(%rbp),%r14
        mov     8*3(%rbp),%r15
        jmp     .Lsqrx4x_sub_entry      # common post-condition
+.cfi_endproc
 .size  mulx4x_internal,.-mulx4x_internal
 ___
 }\f{
@@ -2909,6 +2922,7 @@ bn_powerx5:
 .align 32
 bn_sqrx8x_internal:
 __bn_sqrx8x_internal:
+.cfi_startproc
        ##################################################################
        # Squaring part:
        #
@@ -3541,6 +3555,7 @@ __bn_sqrx8x_reduction:
        cmp     8+8(%rsp),%r8           # end of t[]?
        jb      .Lsqrx8x_reduction_loop
        ret
+.cfi_endproc
 .size  bn_sqrx8x_internal,.-bn_sqrx8x_internal
 ___
 }\f
@@ -3552,6 +3567,7 @@ my ($rptr,$nptr)=("%rdx","%rbp");
 $code.=<<___;
 .align 32
 __bn_postx4x_internal:
+.cfi_startproc
        mov     8*0($nptr),%r12
        mov     %rcx,%r10               # -$num
        mov     %rcx,%r9                # -$num
@@ -3599,6 +3615,7 @@ __bn_postx4x_internal:
        neg     %r9                     # restore $num
 
        ret
+.cfi_endproc
 .size  __bn_postx4x_internal,.-__bn_postx4x_internal
 ___
 }
@@ -3615,6 +3632,7 @@ $code.=<<___;
 .type  bn_get_bits5,\@abi-omnipotent
 .align 16
 bn_get_bits5:
+.cfi_startproc
        lea     0($inp),%r10
        lea     1($inp),%r11
        mov     $num,%ecx
@@ -3628,12 +3646,14 @@ bn_get_bits5:
        shrl    %cl,%eax
        and     \$31,%eax
        ret
+.cfi_endproc
 .size  bn_get_bits5,.-bn_get_bits5
 
 .globl bn_scatter5
 .type  bn_scatter5,\@abi-omnipotent
 .align 16
 bn_scatter5:
+.cfi_startproc
        cmp     \$0, $num
        jz      .Lscatter_epilogue
        lea     ($tbl,$idx,8),$tbl
@@ -3646,6 +3666,7 @@ bn_scatter5:
        jnz     .Lscatter
 .Lscatter_epilogue:
        ret
+.cfi_endproc
 .size  bn_scatter5,.-bn_scatter5
 
 .globl bn_gather5
@@ -3653,6 +3674,7 @@ bn_scatter5:
 .align 32
 bn_gather5:
 .LSEH_begin_bn_gather5:                        # Win64 thing, but harmless in other cases
+.cfi_startproc
        # I can't trust assembler to use specific encoding:-(
        .byte   0x4c,0x8d,0x14,0x24                     #lea    (%rsp),%r10
        .byte   0x48,0x81,0xec,0x08,0x01,0x00,0x00      #sub    $0x108,%rsp
@@ -3737,6 +3759,7 @@ $code.=<<___;
        lea     (%r10),%rsp
        ret
 .LSEH_end_bn_gather5:
+.cfi_endproc
 .size  bn_gather5,.-bn_gather5
 ___
 }