2005-04-17 21:05:57 +00:00
|
|
|
|
#!/usr/bin/env perl
|
|
|
|
|
|
2008-10-28 08:40:07 +00:00
|
|
|
|
# Ascetic x86_64 AT&T to MASM/NASM assembler translator by <appro>.
|
2005-04-17 21:05:57 +00:00
|
|
|
|
#
|
|
|
|
|
# Why AT&T to MASM and not vice versa? Several reasons. Because AT&T
|
|
|
|
|
# format is way easier to parse. Because it's simpler to "gear" from
|
|
|
|
|
# Unix ABI to Windows one [see cross-reference "card" at the end of
|
|
|
|
|
# file]. Because Linux targets were available first...
|
|
|
|
|
#
|
|
|
|
|
# In addition the script also "distills" code suitable for GNU
|
|
|
|
|
# assembler, so that it can be compiled with more rigid assemblers,
|
|
|
|
|
# such as Solaris /usr/ccs/bin/as.
|
|
|
|
|
#
|
|
|
|
|
# This translator is not designed to convert *arbitrary* assembler
|
|
|
|
|
# code from AT&T format to MASM one. It's designed to convert just
|
|
|
|
|
# enough to provide for dual-ABI OpenSSL modules development...
|
|
|
|
|
# There *are* limitations and you might have to modify your assembler
|
|
|
|
|
# code or this script to achieve the desired result...
|
|
|
|
|
#
|
|
|
|
|
# Currently recognized limitations:
|
|
|
|
|
#
|
|
|
|
|
# - can't use multiple ops per line;
|
|
|
|
|
#
|
|
|
|
|
# Dual-ABI styling rules.
|
|
|
|
|
#
|
2008-10-28 08:40:07 +00:00
|
|
|
|
# 1. Adhere to Unix register and stack layout [see cross-reference
|
|
|
|
|
# ABI "card" at the end for explanation].
|
2005-04-17 21:05:57 +00:00
|
|
|
|
# 2. Forget about "red zone," stick to more traditional blended
|
|
|
|
|
# stack frame allocation. If volatile storage is actually required
|
|
|
|
|
# that is. If not, just leave the stack as is.
|
|
|
|
|
# 3. Functions tagged with ".type name,@function" get crafted with
|
2005-05-03 15:35:14 +00:00
|
|
|
|
# unified Win64 prologue and epilogue automatically. If you want
|
2005-04-17 21:05:57 +00:00
|
|
|
|
# to take care of ABI differences yourself, tag functions as
|
2005-05-03 15:35:14 +00:00
|
|
|
|
# ".type name,@abi-omnipotent" instead.
|
|
|
|
|
# 4. To optimize the Win64 prologue you can specify number of input
|
|
|
|
|
# arguments as ".type name,@function,N." Keep in mind that if N is
|
|
|
|
|
# larger than 6, then you *have to* write "abi-omnipotent" code,
|
|
|
|
|
# because >6 cases can't be addressed with unified prologue.
|
|
|
|
|
# 5. Name local labels as .L*, do *not* use dynamic labels such as 1:
|
|
|
|
|
# (sorry about latter).
|
|
|
|
|
# 6. Don't use [or hand-code with .byte] "rep ret." "ret" mnemonic is
|
|
|
|
|
# required to identify the spots, where to inject Win64 epilogue!
|
|
|
|
|
# But on the pros, it's then prefixed with rep automatically:-)
|
2008-11-12 08:05:58 +00:00
|
|
|
|
# 7. Stick to explicit ip-relative addressing. If you have to use
|
|
|
|
|
# GOTPCREL addressing, stick to mov symbol@GOTPCREL(%rip),%r??.
|
|
|
|
|
# Both are recognized and translated to proper Win64 addressing
|
|
|
|
|
# modes. To support legacy code a synthetic directive, .picmeup,
|
|
|
|
|
# is implemented. It puts address of the *next* instruction into
|
|
|
|
|
# target register, e.g.:
|
2005-05-07 08:13:51 +00:00
|
|
|
|
#
|
|
|
|
|
# .picmeup %rax
|
|
|
|
|
# lea .Label-.(%rax),%rax
|
2008-10-28 08:40:07 +00:00
|
|
|
|
#
|
|
|
|
|
# 8. In order to provide for structured exception handling unified
|
|
|
|
|
# Win64 prologue copies %rsp value to %rax. For further details
|
|
|
|
|
# see SEH paragraph at the end.
|
2008-11-12 08:05:58 +00:00
|
|
|
|
# 9. .init segment is allowed to contain calls to functions only.
|
2010-06-01 05:56:24 +00:00
|
|
|
|
# a. If function accepts more than 4 arguments *and* >4th argument
|
|
|
|
|
# is declared as non 64-bit value, do clear its upper part.
|
2008-10-28 08:40:07 +00:00
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
my $flavour = shift;
|
|
|
|
|
my $output = shift;
|
|
|
|
|
if ($flavour =~ /\./) { $output = $flavour; undef $flavour; }
|
2007-05-14 15:57:19 +00:00
|
|
|
|
|
2012-03-13 19:19:08 +00:00
|
|
|
|
open STDOUT,">$output" || die "can't open $output: $!"
|
|
|
|
|
if (defined($output));
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
my $gas=1; $gas=0 if ($output =~ /\.asm$/);
|
|
|
|
|
my $elf=1; $elf=0 if (!$gas);
|
|
|
|
|
my $win64=0;
|
|
|
|
|
my $prefix="";
|
2008-11-14 16:40:37 +00:00
|
|
|
|
my $decor=".L";
|
2008-02-13 13:07:52 +00:00
|
|
|
|
|
2008-02-11 13:04:39 +00:00
|
|
|
|
my $masmref=8 + 50727*2**-32; # 8.00.50727 shipped with VS2005
|
2008-07-22 08:42:06 +00:00
|
|
|
|
my $masm=0;
|
2008-02-13 13:07:52 +00:00
|
|
|
|
my $PTR=" PTR";
|
|
|
|
|
|
2008-07-22 08:42:06 +00:00
|
|
|
|
my $nasmref=2.03;
|
2008-02-13 13:07:52 +00:00
|
|
|
|
my $nasm=0;
|
|
|
|
|
|
2010-06-01 05:56:24 +00:00
|
|
|
|
if ($flavour eq "mingw64") { $gas=1; $elf=0; $win64=1;
|
|
|
|
|
$prefix=`echo __USER_LABEL_PREFIX__ | $ENV{CC} -E -P -`;
|
Perl's chop / chomp considered bad, use a regexp instead
Once upon a time, there was chop, which somply chopped off the last
character of $_ or a given variable, and it was used to take off the
EOL character (\n) of strings.
... but then, you had to check for the presence of such character.
So came chomp, the better chop which checks for \n before chopping it
off. And this worked well, as long as Perl made internally sure that
all EOLs were converted to \n.
These days, though, there seems to be a mixture of perls, so lines
from files in the "wrong" environment might have \r\n as EOL, or just
\r (Mac OS, unless I'm misinformed).
So it's time we went for the more generic variant and use s|\R$||, the
better chomp which recognises all kinds of known EOLs and chops them
off.
A few chops were left alone, as they are use as surgical tools to
remove one last slash or one last comma.
NOTE: \R came with perl 5.10.0. It means that from now on, our
scripts will fail with any older version.
Reviewed-by: Rich Salz <rsalz@openssl.org>
2016-02-11 20:47:30 +00:00
|
|
|
|
$prefix =~ s|\R$||; # Better chomp
|
2010-06-01 05:56:24 +00:00
|
|
|
|
}
|
2008-11-14 16:40:37 +00:00
|
|
|
|
elsif ($flavour eq "macosx") { $gas=1; $elf=0; $prefix="_"; $decor="L\$"; }
|
|
|
|
|
elsif ($flavour eq "masm") { $gas=0; $elf=0; $masm=$masmref; $win64=1; $decor="\$L\$"; }
|
|
|
|
|
elsif ($flavour eq "nasm") { $gas=0; $elf=0; $nasm=$nasmref; $win64=1; $decor="\$L\$"; $PTR=""; }
|
2008-11-12 08:05:58 +00:00
|
|
|
|
elsif (!$gas)
|
2008-07-22 08:42:06 +00:00
|
|
|
|
{ if ($ENV{ASM} =~ m/nasm/ && `nasm -v` =~ m/version ([0-9]+)\.([0-9]+)/i)
|
|
|
|
|
{ $nasm = $1 + $2*0.01; $PTR=""; }
|
2008-02-13 13:07:52 +00:00
|
|
|
|
elsif (`ml64 2>&1` =~ m/Version ([0-9]+)\.([0-9]+)(\.([0-9]+))?/)
|
|
|
|
|
{ $masm = $1 + $2*2**-16 + $4*2**-32; }
|
2008-07-22 08:42:06 +00:00
|
|
|
|
die "no assembler found on %PATH" if (!($nasm || $masm));
|
2008-11-12 08:05:58 +00:00
|
|
|
|
$win64=1;
|
|
|
|
|
$elf=0;
|
2008-11-14 16:40:37 +00:00
|
|
|
|
$decor="\$L\$";
|
2008-02-13 13:07:52 +00:00
|
|
|
|
}
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
|
|
|
|
my $current_segment;
|
|
|
|
|
my $current_function;
|
2008-07-22 08:42:06 +00:00
|
|
|
|
my %globals;
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
|
|
|
|
{ package opcode; # pick up opcodes
|
|
|
|
|
sub re {
|
|
|
|
|
my $self = shift; # single instance in enough...
|
|
|
|
|
local *line = shift;
|
|
|
|
|
undef $ret;
|
|
|
|
|
|
2007-05-14 15:57:19 +00:00
|
|
|
|
if ($line =~ /^([a-z][a-z0-9]*)/i) {
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$self->{op} = $1;
|
|
|
|
|
$ret = $self;
|
|
|
|
|
$line = substr($line,@+[0]); $line =~ s/^\s+//;
|
|
|
|
|
|
|
|
|
|
undef $self->{sz};
|
2011-08-12 21:24:19 +00:00
|
|
|
|
if ($self->{op} =~ /^(movz)x?([bw]).*/) { # movz is pain...
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$self->{op} = $1;
|
2011-08-12 21:24:19 +00:00
|
|
|
|
$self->{sz} = $2;
|
2008-07-22 08:42:06 +00:00
|
|
|
|
} elsif ($self->{op} =~ /call|jmp/) {
|
2010-04-10 13:55:05 +00:00
|
|
|
|
$self->{sz} = "";
|
2010-07-26 21:45:38 +00:00
|
|
|
|
} elsif ($self->{op} =~ /^p/ && $' !~ /^(ush|op|insrw)/) { # SSEn
|
2010-04-10 13:55:05 +00:00
|
|
|
|
$self->{sz} = "";
|
2011-06-27 09:45:10 +00:00
|
|
|
|
} elsif ($self->{op} =~ /^v/) { # VEX
|
|
|
|
|
$self->{sz} = "";
|
2014-09-11 22:06:00 +00:00
|
|
|
|
} elsif ($self->{op} =~ /mov[dq]/ && $line =~ /%xmm/) {
|
2011-05-18 16:26:03 +00:00
|
|
|
|
$self->{sz} = "";
|
2008-01-12 11:29:45 +00:00
|
|
|
|
} elsif ($self->{op} =~ /([a-z]{3,})([qlwb])$/) {
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$self->{op} = $1;
|
|
|
|
|
$self->{sz} = $2;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
$ret;
|
|
|
|
|
}
|
|
|
|
|
sub size {
|
|
|
|
|
my $self = shift;
|
|
|
|
|
my $sz = shift;
|
|
|
|
|
$self->{sz} = $sz if (defined($sz) && !defined($self->{sz}));
|
|
|
|
|
$self->{sz};
|
|
|
|
|
}
|
|
|
|
|
sub out {
|
|
|
|
|
my $self = shift;
|
2008-11-12 08:05:58 +00:00
|
|
|
|
if ($gas) {
|
2007-05-14 15:57:19 +00:00
|
|
|
|
if ($self->{op} eq "movz") { # movz is pain...
|
2005-04-17 21:05:57 +00:00
|
|
|
|
sprintf "%s%s%s",$self->{op},$self->{sz},shift;
|
2007-05-14 15:57:19 +00:00
|
|
|
|
} elsif ($self->{op} =~ /^set/) {
|
|
|
|
|
"$self->{op}";
|
2005-04-17 21:05:57 +00:00
|
|
|
|
} elsif ($self->{op} eq "ret") {
|
2008-11-12 08:05:58 +00:00
|
|
|
|
my $epilogue = "";
|
|
|
|
|
if ($win64 && $current_function->{abi} eq "svr4") {
|
|
|
|
|
$epilogue = "movq 8(%rsp),%rdi\n\t" .
|
|
|
|
|
"movq 16(%rsp),%rsi\n\t";
|
|
|
|
|
}
|
|
|
|
|
$epilogue . ".byte 0xf3,0xc3";
|
|
|
|
|
} elsif ($self->{op} eq "call" && !$elf && $current_segment eq ".init") {
|
|
|
|
|
".p2align\t3\n\t.quad";
|
2005-04-17 21:05:57 +00:00
|
|
|
|
} else {
|
|
|
|
|
"$self->{op}$self->{sz}";
|
|
|
|
|
}
|
|
|
|
|
} else {
|
2008-01-12 11:29:45 +00:00
|
|
|
|
$self->{op} =~ s/^movz/movzx/;
|
2005-04-17 21:05:57 +00:00
|
|
|
|
if ($self->{op} eq "ret") {
|
|
|
|
|
$self->{op} = "";
|
2008-11-12 08:05:58 +00:00
|
|
|
|
if ($win64 && $current_function->{abi} eq "svr4") {
|
2008-02-13 13:07:52 +00:00
|
|
|
|
$self->{op} = "mov rdi,QWORD${PTR}[8+rsp]\t;WIN64 epilogue\n\t".
|
|
|
|
|
"mov rsi,QWORD${PTR}[16+rsp]\n\t";
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
$self->{op} .= "DB\t0F3h,0C3h\t\t;repret";
|
2008-11-12 08:05:58 +00:00
|
|
|
|
} elsif ($self->{op} =~ /^(pop|push)f/) {
|
2008-10-28 08:40:07 +00:00
|
|
|
|
$self->{op} .= $self->{sz};
|
2008-11-12 08:05:58 +00:00
|
|
|
|
} elsif ($self->{op} eq "call" && $current_segment eq ".CRT\$XCU") {
|
2010-10-10 21:07:55 +00:00
|
|
|
|
$self->{op} = "\tDQ";
|
2008-11-12 08:05:58 +00:00
|
|
|
|
}
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$self->{op};
|
|
|
|
|
}
|
|
|
|
|
}
|
2008-11-12 08:05:58 +00:00
|
|
|
|
sub mnemonic {
|
|
|
|
|
my $self=shift;
|
|
|
|
|
my $op=shift;
|
|
|
|
|
$self->{op}=$op if (defined($op));
|
|
|
|
|
$self->{op};
|
|
|
|
|
}
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
{ package const; # pick up constants, which start with $
|
|
|
|
|
sub re {
|
|
|
|
|
my $self = shift; # single instance in enough...
|
|
|
|
|
local *line = shift;
|
|
|
|
|
undef $ret;
|
|
|
|
|
|
|
|
|
|
if ($line =~ /^\$([^,]+)/) {
|
|
|
|
|
$self->{value} = $1;
|
|
|
|
|
$ret = $self;
|
|
|
|
|
$line = substr($line,@+[0]); $line =~ s/^\s+//;
|
|
|
|
|
}
|
|
|
|
|
$ret;
|
|
|
|
|
}
|
|
|
|
|
sub out {
|
|
|
|
|
my $self = shift;
|
2005-05-07 08:13:51 +00:00
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
if ($gas) {
|
2005-07-12 22:11:22 +00:00
|
|
|
|
# Solaris /usr/ccs/bin/as can't handle multiplications
|
|
|
|
|
# in $self->{value}
|
2016-02-10 14:11:40 +00:00
|
|
|
|
my $value = $self->{value};
|
|
|
|
|
$value =~ s/(?<![\w\$\.])(0x?[0-9a-f]+)/oct($1)/egi;
|
|
|
|
|
if ($value =~ s/([0-9]+\s*[\*\/\%]\s*[0-9]+)/eval($1)/eg) {
|
|
|
|
|
$self->{value} = $value;
|
|
|
|
|
}
|
2005-05-07 08:13:51 +00:00
|
|
|
|
sprintf "\$%s",$self->{value};
|
|
|
|
|
} else {
|
2008-12-27 14:00:37 +00:00
|
|
|
|
$self->{value} =~ s/(0b[0-1]+)/oct($1)/eig;
|
|
|
|
|
$self->{value} =~ s/0x([0-9a-f]+)/0$1h/ig if ($masm);
|
2005-05-07 08:13:51 +00:00
|
|
|
|
sprintf "%s",$self->{value};
|
|
|
|
|
}
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
{ package ea; # pick up effective addresses: expr(%reg,%reg,scale)
|
|
|
|
|
sub re {
|
|
|
|
|
my $self = shift; # single instance in enough...
|
|
|
|
|
local *line = shift;
|
|
|
|
|
undef $ret;
|
|
|
|
|
|
2008-07-22 08:42:06 +00:00
|
|
|
|
# optional * ---vvv--- appears in indirect jmp/call
|
|
|
|
|
if ($line =~ /^(\*?)([^\(,]*)\(([%\w,]+)\)/) {
|
|
|
|
|
$self->{asterisk} = $1;
|
|
|
|
|
$self->{label} = $2;
|
|
|
|
|
($self->{base},$self->{index},$self->{scale})=split(/,/,$3);
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$self->{scale} = 1 if (!defined($self->{scale}));
|
|
|
|
|
$ret = $self;
|
|
|
|
|
$line = substr($line,@+[0]); $line =~ s/^\s+//;
|
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
if ($win64 && $self->{label} =~ s/\@GOTPCREL//) {
|
|
|
|
|
die if (opcode->mnemonic() ne "mov");
|
|
|
|
|
opcode->mnemonic("lea");
|
|
|
|
|
}
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$self->{base} =~ s/^%//;
|
|
|
|
|
$self->{index} =~ s/^%// if (defined($self->{index}));
|
|
|
|
|
}
|
|
|
|
|
$ret;
|
|
|
|
|
}
|
|
|
|
|
sub size {}
|
|
|
|
|
sub out {
|
|
|
|
|
my $self = shift;
|
|
|
|
|
my $sz = shift;
|
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
$self->{label} =~ s/([_a-z][_a-z0-9]*)/$globals{$1} or $1/gei;
|
2008-11-14 16:40:37 +00:00
|
|
|
|
$self->{label} =~ s/\.L/$decor/g;
|
2008-11-12 08:05:58 +00:00
|
|
|
|
|
2005-07-18 09:15:04 +00:00
|
|
|
|
# Silently convert all EAs to 64-bit. This is required for
|
|
|
|
|
# elder GNU assembler and results in more compact code,
|
|
|
|
|
# *but* most importantly AES module depends on this feature!
|
|
|
|
|
$self->{index} =~ s/^[er](.?[0-9xpi])[d]?$/r\1/;
|
|
|
|
|
$self->{base} =~ s/^[er](.?[0-9xpi])[d]?$/r\1/;
|
|
|
|
|
|
2011-07-04 13:10:50 +00:00
|
|
|
|
# Solaris /usr/ccs/bin/as can't handle multiplications
|
|
|
|
|
# in $self->{label}, new gas requires sign extension...
|
|
|
|
|
use integer;
|
|
|
|
|
$self->{label} =~ s/(?<![\w\$\.])(0x?[0-9a-f]+)/oct($1)/egi;
|
2013-10-02 22:26:09 +00:00
|
|
|
|
$self->{label} =~ s/\b([0-9]+\s*[\*\/\%]\s*[0-9]+)\b/eval($1)/eg;
|
|
|
|
|
$self->{label} =~ s/\b([0-9]+)\b/$1<<32>>32/eg;
|
2011-07-04 13:10:50 +00:00
|
|
|
|
|
2013-05-13 14:06:25 +00:00
|
|
|
|
if (!$self->{label} && $self->{index} && $self->{scale}==1 &&
|
|
|
|
|
$self->{base} =~ /(rbp|r13)/) {
|
|
|
|
|
$self->{base} = $self->{index}; $self->{index} = $1;
|
|
|
|
|
}
|
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
if ($gas) {
|
2008-11-14 16:40:37 +00:00
|
|
|
|
$self->{label} =~ s/^___imp_/__imp__/ if ($flavour eq "mingw64");
|
2005-05-03 15:35:14 +00:00
|
|
|
|
|
2005-04-17 21:05:57 +00:00
|
|
|
|
if (defined($self->{index})) {
|
2011-05-16 17:44:38 +00:00
|
|
|
|
sprintf "%s%s(%s,%%%s,%d)",$self->{asterisk},
|
|
|
|
|
$self->{label},
|
|
|
|
|
$self->{base}?"%$self->{base}":"",
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$self->{index},$self->{scale};
|
2005-05-07 08:13:51 +00:00
|
|
|
|
} else {
|
2008-07-22 08:42:06 +00:00
|
|
|
|
sprintf "%s%s(%%%s)", $self->{asterisk},$self->{label},$self->{base};
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
} else {
|
2014-07-09 18:08:01 +00:00
|
|
|
|
%szmap = ( b=>"BYTE$PTR", w=>"WORD$PTR",
|
|
|
|
|
l=>"DWORD$PTR", d=>"DWORD$PTR",
|
|
|
|
|
q=>"QWORD$PTR", o=>"OWORD$PTR",
|
|
|
|
|
x=>"XMMWORD$PTR", y=>"YMMWORD$PTR", z=>"ZMMWORD$PTR" );
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
2005-05-04 08:42:47 +00:00
|
|
|
|
$self->{label} =~ s/\./\$/g;
|
2010-06-01 05:56:24 +00:00
|
|
|
|
$self->{label} =~ s/(?<![\w\$\.])0x([0-9a-f]+)/0$1h/ig;
|
2005-05-04 08:42:47 +00:00
|
|
|
|
$self->{label} = "($self->{label})" if ($self->{label} =~ /[\*\+\-\/]/);
|
2014-07-09 18:08:01 +00:00
|
|
|
|
|
|
|
|
|
($self->{asterisk}) && ($sz="q") ||
|
|
|
|
|
(opcode->mnemonic() =~ /^v?mov([qd])$/) && ($sz=$1) ||
|
|
|
|
|
(opcode->mnemonic() =~ /^v?pinsr([qdwb])$/) && ($sz=$1) ||
|
|
|
|
|
(opcode->mnemonic() =~ /^vpbroadcast([qdwb])$/) && ($sz=$1) ||
|
|
|
|
|
(opcode->mnemonic() =~ /^vinsert[fi]128$/) && ($sz="x");
|
2005-05-04 08:42:47 +00:00
|
|
|
|
|
2005-04-17 21:05:57 +00:00
|
|
|
|
if (defined($self->{index})) {
|
2011-05-16 17:44:38 +00:00
|
|
|
|
sprintf "%s[%s%s*%d%s]",$szmap{$sz},
|
2008-02-13 13:07:52 +00:00
|
|
|
|
$self->{label}?"$self->{label}+":"",
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$self->{index},$self->{scale},
|
2011-05-16 17:44:38 +00:00
|
|
|
|
$self->{base}?"+$self->{base}":"";
|
2007-05-14 15:57:19 +00:00
|
|
|
|
} elsif ($self->{base} eq "rip") {
|
2008-02-13 13:07:52 +00:00
|
|
|
|
sprintf "%s[%s]",$szmap{$sz},$self->{label};
|
2005-05-07 08:13:51 +00:00
|
|
|
|
} else {
|
2008-02-13 13:07:52 +00:00
|
|
|
|
sprintf "%s[%s%s]",$szmap{$sz},
|
|
|
|
|
$self->{label}?"$self->{label}+":"",
|
|
|
|
|
$self->{base};
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
{ package register; # pick up registers, which start with %.
|
|
|
|
|
sub re {
|
2016-02-05 20:23:54 +00:00
|
|
|
|
my $class = shift; # multiple instances...
|
2005-04-17 21:05:57 +00:00
|
|
|
|
my $self = {};
|
|
|
|
|
local *line = shift;
|
|
|
|
|
undef $ret;
|
|
|
|
|
|
2008-07-22 08:42:06 +00:00
|
|
|
|
# optional * ---vvv--- appears in indirect jmp/call
|
|
|
|
|
if ($line =~ /^(\*?)%(\w+)/) {
|
2005-04-17 21:05:57 +00:00
|
|
|
|
bless $self,$class;
|
2008-07-22 08:42:06 +00:00
|
|
|
|
$self->{asterisk} = $1;
|
|
|
|
|
$self->{value} = $2;
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$ret = $self;
|
|
|
|
|
$line = substr($line,@+[0]); $line =~ s/^\s+//;
|
|
|
|
|
}
|
|
|
|
|
$ret;
|
|
|
|
|
}
|
|
|
|
|
sub size {
|
|
|
|
|
my $self = shift;
|
|
|
|
|
undef $ret;
|
|
|
|
|
|
|
|
|
|
if ($self->{value} =~ /^r[\d]+b$/i) { $ret="b"; }
|
|
|
|
|
elsif ($self->{value} =~ /^r[\d]+w$/i) { $ret="w"; }
|
|
|
|
|
elsif ($self->{value} =~ /^r[\d]+d$/i) { $ret="l"; }
|
|
|
|
|
elsif ($self->{value} =~ /^r[\w]+$/i) { $ret="q"; }
|
|
|
|
|
elsif ($self->{value} =~ /^[a-d][hl]$/i){ $ret="b"; }
|
|
|
|
|
elsif ($self->{value} =~ /^[\w]{2}l$/i) { $ret="b"; }
|
|
|
|
|
elsif ($self->{value} =~ /^[\w]{2}$/i) { $ret="w"; }
|
|
|
|
|
elsif ($self->{value} =~ /^e[a-z]{2}$/i){ $ret="l"; }
|
|
|
|
|
|
|
|
|
|
$ret;
|
|
|
|
|
}
|
|
|
|
|
sub out {
|
|
|
|
|
my $self = shift;
|
2008-11-12 08:05:58 +00:00
|
|
|
|
if ($gas) { sprintf "%s%%%s",$self->{asterisk},$self->{value}; }
|
2008-07-22 08:42:06 +00:00
|
|
|
|
else { $self->{value}; }
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
{ package label; # pick up labels, which end with :
|
|
|
|
|
sub re {
|
|
|
|
|
my $self = shift; # single instance is enough...
|
|
|
|
|
local *line = shift;
|
|
|
|
|
undef $ret;
|
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
if ($line =~ /(^[\.\w]+)\:/) {
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$self->{value} = $1;
|
|
|
|
|
$ret = $self;
|
|
|
|
|
$line = substr($line,@+[0]); $line =~ s/^\s+//;
|
|
|
|
|
|
2008-11-14 16:40:37 +00:00
|
|
|
|
$self->{value} =~ s/^\.L/$decor/;
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
$ret;
|
|
|
|
|
}
|
|
|
|
|
sub out {
|
|
|
|
|
my $self = shift;
|
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
if ($gas) {
|
|
|
|
|
my $func = ($globals{$self->{value}} or $self->{value}) . ":";
|
|
|
|
|
if ($win64 &&
|
|
|
|
|
$current_function->{name} eq $self->{value} &&
|
|
|
|
|
$current_function->{abi} eq "svr4") {
|
|
|
|
|
$func .= "\n";
|
|
|
|
|
$func .= " movq %rdi,8(%rsp)\n";
|
|
|
|
|
$func .= " movq %rsi,16(%rsp)\n";
|
|
|
|
|
$func .= " movq %rsp,%rax\n";
|
2008-11-14 16:40:37 +00:00
|
|
|
|
$func .= "${decor}SEH_begin_$current_function->{name}:\n";
|
2008-11-12 08:05:58 +00:00
|
|
|
|
my $narg = $current_function->{narg};
|
|
|
|
|
$narg=6 if (!defined($narg));
|
|
|
|
|
$func .= " movq %rcx,%rdi\n" if ($narg>0);
|
|
|
|
|
$func .= " movq %rdx,%rsi\n" if ($narg>1);
|
|
|
|
|
$func .= " movq %r8,%rdx\n" if ($narg>2);
|
|
|
|
|
$func .= " movq %r9,%rcx\n" if ($narg>3);
|
|
|
|
|
$func .= " movq 40(%rsp),%r8\n" if ($narg>4);
|
|
|
|
|
$func .= " movq 48(%rsp),%r9\n" if ($narg>5);
|
|
|
|
|
}
|
|
|
|
|
$func;
|
|
|
|
|
} elsif ($self->{value} ne "$current_function->{name}") {
|
2008-07-22 08:42:06 +00:00
|
|
|
|
$self->{value} .= ":" if ($masm && $ret!~m/^\$/);
|
2008-11-12 08:05:58 +00:00
|
|
|
|
$self->{value} . ":";
|
|
|
|
|
} elsif ($win64 && $current_function->{abi} eq "svr4") {
|
2008-07-22 08:42:06 +00:00
|
|
|
|
my $func = "$current_function->{name}" .
|
|
|
|
|
($nasm ? ":" : "\tPROC $current_function->{scope}") .
|
|
|
|
|
"\n";
|
|
|
|
|
$func .= " mov QWORD${PTR}[8+rsp],rdi\t;WIN64 prologue\n";
|
|
|
|
|
$func .= " mov QWORD${PTR}[16+rsp],rsi\n";
|
2008-10-28 08:40:07 +00:00
|
|
|
|
$func .= " mov rax,rsp\n";
|
2008-11-14 16:40:37 +00:00
|
|
|
|
$func .= "${decor}SEH_begin_$current_function->{name}:";
|
2008-10-28 08:40:07 +00:00
|
|
|
|
$func .= ":" if ($masm);
|
|
|
|
|
$func .= "\n";
|
2005-04-17 21:05:57 +00:00
|
|
|
|
my $narg = $current_function->{narg};
|
|
|
|
|
$narg=6 if (!defined($narg));
|
|
|
|
|
$func .= " mov rdi,rcx\n" if ($narg>0);
|
|
|
|
|
$func .= " mov rsi,rdx\n" if ($narg>1);
|
|
|
|
|
$func .= " mov rdx,r8\n" if ($narg>2);
|
|
|
|
|
$func .= " mov rcx,r9\n" if ($narg>3);
|
2008-02-13 13:07:52 +00:00
|
|
|
|
$func .= " mov r8,QWORD${PTR}[40+rsp]\n" if ($narg>4);
|
|
|
|
|
$func .= " mov r9,QWORD${PTR}[48+rsp]\n" if ($narg>5);
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$func .= "\n";
|
|
|
|
|
} else {
|
2008-07-22 08:42:06 +00:00
|
|
|
|
"$current_function->{name}".
|
|
|
|
|
($nasm ? ":" : "\tPROC $current_function->{scope}");
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
{ package expr; # pick up expressioins
|
|
|
|
|
sub re {
|
|
|
|
|
my $self = shift; # single instance is enough...
|
|
|
|
|
local *line = shift;
|
|
|
|
|
undef $ret;
|
|
|
|
|
|
|
|
|
|
if ($line =~ /(^[^,]+)/) {
|
|
|
|
|
$self->{value} = $1;
|
|
|
|
|
$ret = $self;
|
|
|
|
|
$line = substr($line,@+[0]); $line =~ s/^\s+//;
|
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
$self->{value} =~ s/\@PLT// if (!$elf);
|
|
|
|
|
$self->{value} =~ s/([_a-z][_a-z0-9]*)/$globals{$1} or $1/gei;
|
2008-11-14 16:40:37 +00:00
|
|
|
|
$self->{value} =~ s/\.L/$decor/g;
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
$ret;
|
|
|
|
|
}
|
|
|
|
|
sub out {
|
|
|
|
|
my $self = shift;
|
2013-10-02 22:26:09 +00:00
|
|
|
|
if ($nasm && opcode->mnemonic()=~m/^j(?![re]cxz)/) {
|
2008-07-22 08:42:06 +00:00
|
|
|
|
"NEAR ".$self->{value};
|
|
|
|
|
} else {
|
|
|
|
|
$self->{value};
|
|
|
|
|
}
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
{ package directive; # pick up directives, which start with .
|
|
|
|
|
sub re {
|
|
|
|
|
my $self = shift; # single instance is enough...
|
|
|
|
|
local *line = shift;
|
|
|
|
|
undef $ret;
|
|
|
|
|
my $dir;
|
2005-05-07 08:13:51 +00:00
|
|
|
|
my %opcode = # lea 2f-1f(%rip),%dst; 1: nop; 2:
|
|
|
|
|
( "%rax"=>0x01058d48, "%rcx"=>0x010d8d48,
|
|
|
|
|
"%rdx"=>0x01158d48, "%rbx"=>0x011d8d48,
|
|
|
|
|
"%rsp"=>0x01258d48, "%rbp"=>0x012d8d48,
|
|
|
|
|
"%rsi"=>0x01358d48, "%rdi"=>0x013d8d48,
|
|
|
|
|
"%r8" =>0x01058d4c, "%r9" =>0x010d8d4c,
|
|
|
|
|
"%r10"=>0x01158d4c, "%r11"=>0x011d8d4c,
|
|
|
|
|
"%r12"=>0x01258d4c, "%r13"=>0x012d8d4c,
|
|
|
|
|
"%r14"=>0x01358d4c, "%r15"=>0x013d8d4c );
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
|
|
|
|
if ($line =~ /^\s*(\.\w+)/) {
|
2008-11-12 08:05:58 +00:00
|
|
|
|
$dir = $1;
|
|
|
|
|
$ret = $self;
|
|
|
|
|
undef $self->{value};
|
|
|
|
|
$line = substr($line,@+[0]); $line =~ s/^\s+//;
|
|
|
|
|
|
|
|
|
|
SWITCH: for ($dir) {
|
|
|
|
|
/\.picmeup/ && do { if ($line =~ /(%r[\w]+)/i) {
|
|
|
|
|
$dir="\t.long";
|
|
|
|
|
$line=sprintf "0x%x,0x90000000",$opcode{$1};
|
|
|
|
|
}
|
|
|
|
|
last;
|
|
|
|
|
};
|
|
|
|
|
/\.global|\.globl|\.extern/
|
|
|
|
|
&& do { $globals{$line} = $prefix . $line;
|
|
|
|
|
$line = $globals{$line} if ($prefix);
|
|
|
|
|
last;
|
|
|
|
|
};
|
|
|
|
|
/\.type/ && do { ($sym,$type,$narg) = split(',',$line);
|
|
|
|
|
if ($type eq "\@function") {
|
|
|
|
|
undef $current_function;
|
|
|
|
|
$current_function->{name} = $sym;
|
|
|
|
|
$current_function->{abi} = "svr4";
|
|
|
|
|
$current_function->{narg} = $narg;
|
|
|
|
|
$current_function->{scope} = defined($globals{$sym})?"PUBLIC":"PRIVATE";
|
|
|
|
|
} elsif ($type eq "\@abi-omnipotent") {
|
|
|
|
|
undef $current_function;
|
|
|
|
|
$current_function->{name} = $sym;
|
|
|
|
|
$current_function->{scope} = defined($globals{$sym})?"PUBLIC":"PRIVATE";
|
|
|
|
|
}
|
|
|
|
|
$line =~ s/\@abi\-omnipotent/\@function/;
|
|
|
|
|
$line =~ s/\@function.*/\@function/;
|
|
|
|
|
last;
|
|
|
|
|
};
|
|
|
|
|
/\.asciz/ && do { if ($line =~ /^"(.*)"$/) {
|
|
|
|
|
$dir = ".byte";
|
|
|
|
|
$line = join(",",unpack("C*",$1),0);
|
|
|
|
|
}
|
|
|
|
|
last;
|
|
|
|
|
};
|
2008-11-14 16:40:37 +00:00
|
|
|
|
/\.rva|\.long|\.quad/
|
|
|
|
|
&& do { $line =~ s/([_a-z][_a-z0-9]*)/$globals{$1} or $1/gei;
|
|
|
|
|
$line =~ s/\.L/$decor/g;
|
|
|
|
|
last;
|
|
|
|
|
};
|
2008-11-12 08:05:58 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if ($gas) {
|
|
|
|
|
$self->{value} = $dir . "\t" . $line;
|
|
|
|
|
|
|
|
|
|
if ($dir =~ /\.extern/) {
|
2007-05-14 15:57:19 +00:00
|
|
|
|
$self->{value} = ""; # swallow extern
|
2008-11-12 08:05:58 +00:00
|
|
|
|
} elsif (!$elf && $dir =~ /\.type/) {
|
|
|
|
|
$self->{value} = "";
|
|
|
|
|
$self->{value} = ".def\t" . ($globals{$1} or $1) . ";\t" .
|
|
|
|
|
(defined($globals{$1})?".scl 2;":".scl 3;") .
|
|
|
|
|
"\t.type 32;\t.endef"
|
|
|
|
|
if ($win64 && $line =~ /([^,]+),\@function/);
|
|
|
|
|
} elsif (!$elf && $dir =~ /\.size/) {
|
|
|
|
|
$self->{value} = "";
|
|
|
|
|
if (defined($current_function)) {
|
2008-11-14 16:40:37 +00:00
|
|
|
|
$self->{value} .= "${decor}SEH_end_$current_function->{name}:"
|
2008-11-12 08:05:58 +00:00
|
|
|
|
if ($win64 && $current_function->{abi} eq "svr4");
|
|
|
|
|
undef $current_function;
|
|
|
|
|
}
|
|
|
|
|
} elsif (!$elf && $dir =~ /\.align/) {
|
|
|
|
|
$self->{value} = ".p2align\t" . (log($line)/log(2));
|
|
|
|
|
} elsif ($dir eq ".section") {
|
|
|
|
|
$current_segment=$line;
|
|
|
|
|
if (!$elf && $current_segment eq ".init") {
|
|
|
|
|
if ($flavour eq "macosx") { $self->{value} = ".mod_init_func"; }
|
|
|
|
|
elsif ($flavour eq "mingw64") { $self->{value} = ".section\t.ctors"; }
|
|
|
|
|
}
|
|
|
|
|
} elsif ($dir =~ /\.(text|data)/) {
|
|
|
|
|
$current_segment=".$1";
|
2011-05-18 16:24:19 +00:00
|
|
|
|
} elsif ($dir =~ /\.hidden/) {
|
|
|
|
|
if ($flavour eq "macosx") { $self->{value} = ".private_extern\t$prefix$line"; }
|
|
|
|
|
elsif ($flavour eq "mingw64") { $self->{value} = ""; }
|
|
|
|
|
} elsif ($dir =~ /\.comm/) {
|
|
|
|
|
$self->{value} = "$dir\t$prefix$line";
|
2011-11-12 13:10:00 +00:00
|
|
|
|
$self->{value} =~ s|,([0-9]+),([0-9]+)$|",$1,".log($2)/log(2)|e if ($flavour eq "macosx");
|
2005-05-07 08:13:51 +00:00
|
|
|
|
}
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$line = "";
|
|
|
|
|
return $self;
|
|
|
|
|
}
|
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
# non-gas case or nasm/masm
|
2005-04-17 21:05:57 +00:00
|
|
|
|
SWITCH: for ($dir) {
|
2008-07-22 08:42:06 +00:00
|
|
|
|
/\.text/ && do { my $v=undef;
|
2008-02-13 13:07:52 +00:00
|
|
|
|
if ($nasm) {
|
2008-07-22 08:42:06 +00:00
|
|
|
|
$v="section .text code align=64\n";
|
2008-02-13 13:07:52 +00:00
|
|
|
|
} else {
|
|
|
|
|
$v="$current_segment\tENDS\n" if ($current_segment);
|
2008-07-22 08:42:06 +00:00
|
|
|
|
$current_segment = ".text\$";
|
2008-02-13 13:07:52 +00:00
|
|
|
|
$v.="$current_segment\tSEGMENT ";
|
2014-07-09 18:08:01 +00:00
|
|
|
|
$v.=$masm>=$masmref ? "ALIGN(256)" : "PAGE";
|
2008-02-13 13:07:52 +00:00
|
|
|
|
$v.=" 'CODE'";
|
|
|
|
|
}
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$self->{value} = $v;
|
|
|
|
|
last;
|
|
|
|
|
};
|
2008-07-22 08:42:06 +00:00
|
|
|
|
/\.data/ && do { my $v=undef;
|
|
|
|
|
if ($nasm) {
|
|
|
|
|
$v="section .data data align=8\n";
|
|
|
|
|
} else {
|
|
|
|
|
$v="$current_segment\tENDS\n" if ($current_segment);
|
|
|
|
|
$current_segment = "_DATA";
|
|
|
|
|
$v.="$current_segment\tSEGMENT";
|
|
|
|
|
}
|
|
|
|
|
$self->{value} = $v;
|
|
|
|
|
last;
|
|
|
|
|
};
|
|
|
|
|
/\.section/ && do { my $v=undef;
|
2008-11-12 08:05:58 +00:00
|
|
|
|
$line =~ s/([^,]*).*/$1/;
|
|
|
|
|
$line = ".CRT\$XCU" if ($line eq ".init");
|
2008-07-22 08:42:06 +00:00
|
|
|
|
if ($nasm) {
|
|
|
|
|
$v="section $line";
|
|
|
|
|
if ($line=~/\.([px])data/) {
|
|
|
|
|
$v.=" rdata align=";
|
|
|
|
|
$v.=$1 eq "p"? 4 : 8;
|
2010-10-10 21:07:55 +00:00
|
|
|
|
} elsif ($line=~/\.CRT\$/i) {
|
|
|
|
|
$v.=" rdata align=8";
|
2008-07-22 08:42:06 +00:00
|
|
|
|
}
|
|
|
|
|
} else {
|
|
|
|
|
$v="$current_segment\tENDS\n" if ($current_segment);
|
|
|
|
|
$v.="$line\tSEGMENT";
|
|
|
|
|
if ($line=~/\.([px])data/) {
|
|
|
|
|
$v.=" READONLY";
|
|
|
|
|
$v.=" ALIGN(".($1 eq "p" ? 4 : 8).")" if ($masm>=$masmref);
|
2010-10-10 21:07:55 +00:00
|
|
|
|
} elsif ($line=~/\.CRT\$/i) {
|
2012-01-21 11:34:53 +00:00
|
|
|
|
$v.=" READONLY ";
|
|
|
|
|
$v.=$masm>=$masmref ? "ALIGN(8)" : "DWORD";
|
2008-07-22 08:42:06 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
$current_segment = $line;
|
|
|
|
|
$self->{value} = $v;
|
|
|
|
|
last;
|
|
|
|
|
};
|
2008-02-13 13:07:52 +00:00
|
|
|
|
/\.extern/ && do { $self->{value} = "EXTERN\t".$line;
|
2008-07-22 08:42:06 +00:00
|
|
|
|
$self->{value} .= ":NEAR" if ($masm);
|
|
|
|
|
last;
|
|
|
|
|
};
|
2008-10-28 08:40:07 +00:00
|
|
|
|
/\.globl|.global/
|
|
|
|
|
&& do { $self->{value} = $masm?"PUBLIC":"global";
|
|
|
|
|
$self->{value} .= "\t".$line;
|
2005-04-17 21:05:57 +00:00
|
|
|
|
last;
|
|
|
|
|
};
|
2005-05-07 08:13:51 +00:00
|
|
|
|
/\.size/ && do { if (defined($current_function)) {
|
2008-11-03 08:46:07 +00:00
|
|
|
|
undef $self->{value};
|
|
|
|
|
if ($current_function->{abi} eq "svr4") {
|
2008-11-14 16:40:37 +00:00
|
|
|
|
$self->{value}="${decor}SEH_end_$current_function->{name}:";
|
2008-11-03 08:46:07 +00:00
|
|
|
|
$self->{value}.=":\n" if($masm);
|
|
|
|
|
}
|
2011-10-18 09:50:23 +00:00
|
|
|
|
$self->{value}.="$current_function->{name}\tENDP" if($masm && $current_function->{name});
|
2005-04-17 21:05:57 +00:00
|
|
|
|
undef $current_function;
|
|
|
|
|
}
|
|
|
|
|
last;
|
|
|
|
|
};
|
|
|
|
|
/\.align/ && do { $self->{value} = "ALIGN\t".$line; last; };
|
2008-11-12 08:05:58 +00:00
|
|
|
|
/\.(value|long|rva|quad)/
|
2008-10-28 08:40:07 +00:00
|
|
|
|
&& do { my $sz = substr($1,0,1);
|
2010-04-10 13:55:05 +00:00
|
|
|
|
my @arr = split(/,\s*/,$line);
|
2005-04-17 21:05:57 +00:00
|
|
|
|
my $last = pop(@arr);
|
2008-02-13 13:07:52 +00:00
|
|
|
|
my $conv = sub { my $var=shift;
|
2009-05-02 21:18:52 +00:00
|
|
|
|
$var=~s/^(0b[0-1]+)/oct($1)/eig;
|
2010-04-10 13:55:05 +00:00
|
|
|
|
$var=~s/^0x([0-9a-f]+)/0$1h/ig if ($masm);
|
2008-11-12 08:05:58 +00:00
|
|
|
|
if ($sz eq "D" && ($current_segment=~/.[px]data/ || $dir eq ".rva"))
|
2008-10-28 08:40:07 +00:00
|
|
|
|
{ $var=~s/([_a-z\$\@][_a-z0-9\$\@]*)/$nasm?"$1 wrt ..imagebase":"imagerel $1"/egi; }
|
2008-07-22 08:42:06 +00:00
|
|
|
|
$var;
|
2008-02-13 13:07:52 +00:00
|
|
|
|
};
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
$sz =~ tr/bvlrq/BWDDQ/;
|
2005-04-17 21:05:57 +00:00
|
|
|
|
$self->{value} = "\tD$sz\t";
|
2008-02-13 13:07:52 +00:00
|
|
|
|
for (@arr) { $self->{value} .= &$conv($_).","; }
|
|
|
|
|
$self->{value} .= &$conv($last);
|
2005-04-17 21:05:57 +00:00
|
|
|
|
last;
|
|
|
|
|
};
|
2010-04-10 13:55:05 +00:00
|
|
|
|
/\.byte/ && do { my @str=split(/,\s*/,$line);
|
2008-12-27 14:00:37 +00:00
|
|
|
|
map(s/(0b[0-1]+)/oct($1)/eig,@str);
|
2008-12-19 11:14:38 +00:00
|
|
|
|
map(s/0x([0-9a-f]+)/0$1h/ig,@str) if ($masm);
|
2008-11-12 08:05:58 +00:00
|
|
|
|
while ($#str>15) {
|
2007-08-23 12:01:58 +00:00
|
|
|
|
$self->{value}.="DB\t"
|
2008-11-12 08:05:58 +00:00
|
|
|
|
.join(",",@str[0..15])."\n";
|
|
|
|
|
foreach (0..15) { shift @str; }
|
2006-10-17 06:43:11 +00:00
|
|
|
|
}
|
2008-11-12 08:05:58 +00:00
|
|
|
|
$self->{value}.="DB\t"
|
|
|
|
|
.join(",",@str) if (@str);
|
2006-10-17 06:43:11 +00:00
|
|
|
|
last;
|
|
|
|
|
};
|
2011-05-18 16:24:19 +00:00
|
|
|
|
/\.comm/ && do { my @str=split(/,\s*/,$line);
|
|
|
|
|
my $v=undef;
|
|
|
|
|
if ($nasm) {
|
2011-06-04 12:20:45 +00:00
|
|
|
|
$v.="common $prefix@str[0] @str[1]";
|
2011-05-18 16:24:19 +00:00
|
|
|
|
} else {
|
|
|
|
|
$v="$current_segment\tENDS\n" if ($current_segment);
|
2011-07-01 21:21:14 +00:00
|
|
|
|
$current_segment = "_DATA";
|
2011-05-18 16:24:19 +00:00
|
|
|
|
$v.="$current_segment\tSEGMENT\n";
|
|
|
|
|
$v.="COMM @str[0]:DWORD:".@str[1]/4;
|
|
|
|
|
}
|
|
|
|
|
$self->{value} = $v;
|
|
|
|
|
last;
|
|
|
|
|
};
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
$line = "";
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
$ret;
|
|
|
|
|
}
|
|
|
|
|
sub out {
|
|
|
|
|
my $self = shift;
|
|
|
|
|
$self->{value};
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2010-04-10 13:55:05 +00:00
|
|
|
|
sub rex {
|
|
|
|
|
local *opcode=shift;
|
2011-05-18 16:24:19 +00:00
|
|
|
|
my ($dst,$src,$rex)=@_;
|
|
|
|
|
|
|
|
|
|
$rex|=0x04 if($dst>=8);
|
|
|
|
|
$rex|=0x01 if($src>=8);
|
|
|
|
|
push @opcode,($rex|0x40) if ($rex);
|
2010-04-10 13:55:05 +00:00
|
|
|
|
}
|
|
|
|
|
|
2010-07-26 21:45:38 +00:00
|
|
|
|
# older gas and ml64 don't handle SSE>2 instructions
|
2010-04-10 13:55:05 +00:00
|
|
|
|
my %regrm = ( "%eax"=>0, "%ecx"=>1, "%edx"=>2, "%ebx"=>3,
|
|
|
|
|
"%esp"=>4, "%ebp"=>5, "%esi"=>6, "%edi"=>7 );
|
|
|
|
|
|
2011-05-18 16:26:03 +00:00
|
|
|
|
my $movq = sub { # elderly gas can't handle inter-register movq
|
|
|
|
|
my $arg = shift;
|
|
|
|
|
my @opcode=(0x66);
|
2011-09-12 08:25:14 +00:00
|
|
|
|
if ($arg =~ /%xmm([0-9]+),\s*%r(\w+)/) {
|
2011-05-18 16:26:03 +00:00
|
|
|
|
my ($src,$dst)=($1,$2);
|
|
|
|
|
if ($dst !~ /[0-9]+/) { $dst = $regrm{"%e$dst"}; }
|
|
|
|
|
rex(\@opcode,$src,$dst,0x8);
|
|
|
|
|
push @opcode,0x0f,0x7e;
|
|
|
|
|
push @opcode,0xc0|(($src&7)<<3)|($dst&7); # ModR/M
|
|
|
|
|
@opcode;
|
2011-09-12 08:25:14 +00:00
|
|
|
|
} elsif ($arg =~ /%r(\w+),\s*%xmm([0-9]+)/) {
|
2011-05-18 16:26:03 +00:00
|
|
|
|
my ($src,$dst)=($2,$1);
|
|
|
|
|
if ($dst !~ /[0-9]+/) { $dst = $regrm{"%e$dst"}; }
|
|
|
|
|
rex(\@opcode,$src,$dst,0x8);
|
|
|
|
|
push @opcode,0x0f,0x6e;
|
|
|
|
|
push @opcode,0xc0|(($src&7)<<3)|($dst&7); # ModR/M
|
|
|
|
|
@opcode;
|
|
|
|
|
} else {
|
|
|
|
|
();
|
|
|
|
|
}
|
|
|
|
|
};
|
|
|
|
|
|
2010-04-10 13:55:05 +00:00
|
|
|
|
my $pextrd = sub {
|
2011-09-12 08:25:14 +00:00
|
|
|
|
if (shift =~ /\$([0-9]+),\s*%xmm([0-9]+),\s*(%\w+)/) {
|
2010-04-10 13:55:05 +00:00
|
|
|
|
my @opcode=(0x66);
|
|
|
|
|
$imm=$1;
|
|
|
|
|
$src=$2;
|
2010-07-26 21:45:38 +00:00
|
|
|
|
$dst=$3;
|
2010-04-10 13:55:05 +00:00
|
|
|
|
if ($dst =~ /%r([0-9]+)d/) { $dst = $1; }
|
|
|
|
|
elsif ($dst =~ /%e/) { $dst = $regrm{$dst}; }
|
|
|
|
|
rex(\@opcode,$src,$dst);
|
|
|
|
|
push @opcode,0x0f,0x3a,0x16;
|
|
|
|
|
push @opcode,0xc0|(($src&7)<<3)|($dst&7); # ModR/M
|
|
|
|
|
push @opcode,$imm;
|
2010-07-26 21:45:38 +00:00
|
|
|
|
@opcode;
|
2010-04-10 13:55:05 +00:00
|
|
|
|
} else {
|
2010-07-26 21:45:38 +00:00
|
|
|
|
();
|
2010-04-10 13:55:05 +00:00
|
|
|
|
}
|
2010-07-26 21:45:38 +00:00
|
|
|
|
};
|
2010-04-10 13:55:05 +00:00
|
|
|
|
|
|
|
|
|
my $pinsrd = sub {
|
2011-09-12 08:25:14 +00:00
|
|
|
|
if (shift =~ /\$([0-9]+),\s*(%\w+),\s*%xmm([0-9]+)/) {
|
2010-04-10 13:55:05 +00:00
|
|
|
|
my @opcode=(0x66);
|
|
|
|
|
$imm=$1;
|
2010-07-26 21:45:38 +00:00
|
|
|
|
$src=$2;
|
|
|
|
|
$dst=$3;
|
2011-05-16 17:44:38 +00:00
|
|
|
|
if ($src =~ /%r([0-9]+)/) { $src = $1; }
|
2010-04-10 13:55:05 +00:00
|
|
|
|
elsif ($src =~ /%e/) { $src = $regrm{$src}; }
|
|
|
|
|
rex(\@opcode,$dst,$src);
|
|
|
|
|
push @opcode,0x0f,0x3a,0x22;
|
|
|
|
|
push @opcode,0xc0|(($dst&7)<<3)|($src&7); # ModR/M
|
|
|
|
|
push @opcode,$imm;
|
2010-07-26 21:45:38 +00:00
|
|
|
|
@opcode;
|
2010-04-10 13:55:05 +00:00
|
|
|
|
} else {
|
2010-07-26 21:45:38 +00:00
|
|
|
|
();
|
2010-04-10 13:55:05 +00:00
|
|
|
|
}
|
2010-07-26 21:45:38 +00:00
|
|
|
|
};
|
2010-04-10 13:55:05 +00:00
|
|
|
|
|
|
|
|
|
my $pshufb = sub {
|
2011-09-12 08:25:14 +00:00
|
|
|
|
if (shift =~ /%xmm([0-9]+),\s*%xmm([0-9]+)/) {
|
2010-04-10 13:55:05 +00:00
|
|
|
|
my @opcode=(0x66);
|
2010-07-26 21:45:38 +00:00
|
|
|
|
rex(\@opcode,$2,$1);
|
2010-04-10 13:55:05 +00:00
|
|
|
|
push @opcode,0x0f,0x38,0x00;
|
2010-07-26 21:45:38 +00:00
|
|
|
|
push @opcode,0xc0|($1&7)|(($2&7)<<3); # ModR/M
|
|
|
|
|
@opcode;
|
2010-04-10 13:55:05 +00:00
|
|
|
|
} else {
|
2010-07-26 21:45:38 +00:00
|
|
|
|
();
|
2010-04-10 13:55:05 +00:00
|
|
|
|
}
|
2010-07-26 21:45:38 +00:00
|
|
|
|
};
|
2010-04-10 13:55:05 +00:00
|
|
|
|
|
2011-05-16 17:44:38 +00:00
|
|
|
|
my $palignr = sub {
|
2011-09-12 08:25:14 +00:00
|
|
|
|
if (shift =~ /\$([0-9]+),\s*%xmm([0-9]+),\s*%xmm([0-9]+)/) {
|
2011-05-16 17:44:38 +00:00
|
|
|
|
my @opcode=(0x66);
|
|
|
|
|
rex(\@opcode,$3,$2);
|
|
|
|
|
push @opcode,0x0f,0x3a,0x0f;
|
|
|
|
|
push @opcode,0xc0|($2&7)|(($3&7)<<3); # ModR/M
|
|
|
|
|
push @opcode,$1;
|
|
|
|
|
@opcode;
|
|
|
|
|
} else {
|
|
|
|
|
();
|
|
|
|
|
}
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
my $pclmulqdq = sub {
|
|
|
|
|
if (shift =~ /\$([x0-9a-f]+),\s*%xmm([0-9]+),\s*%xmm([0-9]+)/) {
|
|
|
|
|
my @opcode=(0x66);
|
|
|
|
|
rex(\@opcode,$3,$2);
|
|
|
|
|
push @opcode,0x0f,0x3a,0x44;
|
|
|
|
|
push @opcode,0xc0|($2&7)|(($3&7)<<3); # ModR/M
|
|
|
|
|
my $c=$1;
|
|
|
|
|
push @opcode,$c=~/^0/?oct($c):$c;
|
|
|
|
|
@opcode;
|
|
|
|
|
} else {
|
|
|
|
|
();
|
|
|
|
|
}
|
|
|
|
|
};
|
|
|
|
|
|
2011-06-04 12:20:45 +00:00
|
|
|
|
my $rdrand = sub {
|
|
|
|
|
if (shift =~ /%[er](\w+)/) {
|
|
|
|
|
my @opcode=();
|
|
|
|
|
my $dst=$1;
|
|
|
|
|
if ($dst !~ /[0-9]+/) { $dst = $regrm{"%e$dst"}; }
|
|
|
|
|
rex(\@opcode,0,$1,8);
|
|
|
|
|
push @opcode,0x0f,0xc7,0xf0|($dst&7);
|
|
|
|
|
@opcode;
|
|
|
|
|
} else {
|
|
|
|
|
();
|
|
|
|
|
}
|
|
|
|
|
};
|
|
|
|
|
|
2014-02-14 16:24:12 +00:00
|
|
|
|
my $rdseed = sub {
|
|
|
|
|
if (shift =~ /%[er](\w+)/) {
|
|
|
|
|
my @opcode=();
|
|
|
|
|
my $dst=$1;
|
|
|
|
|
if ($dst !~ /[0-9]+/) { $dst = $regrm{"%e$dst"}; }
|
|
|
|
|
rex(\@opcode,0,$1,8);
|
|
|
|
|
push @opcode,0x0f,0xc7,0xf8|($dst&7);
|
|
|
|
|
@opcode;
|
|
|
|
|
} else {
|
|
|
|
|
();
|
|
|
|
|
}
|
|
|
|
|
};
|
|
|
|
|
|
2012-08-13 12:34:36 +00:00
|
|
|
|
sub rxb {
|
|
|
|
|
local *opcode=shift;
|
|
|
|
|
my ($dst,$src1,$src2,$rxb)=@_;
|
|
|
|
|
|
|
|
|
|
$rxb|=0x7<<5;
|
|
|
|
|
$rxb&=~(0x04<<5) if($dst>=8);
|
|
|
|
|
$rxb&=~(0x01<<5) if($src1>=8);
|
|
|
|
|
$rxb&=~(0x02<<5) if($src2>=8);
|
|
|
|
|
push @opcode,$rxb;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
my $vprotd = sub {
|
|
|
|
|
if (shift =~ /\$([x0-9a-f]+),\s*%xmm([0-9]+),\s*%xmm([0-9]+)/) {
|
|
|
|
|
my @opcode=(0x8f);
|
|
|
|
|
rxb(\@opcode,$3,$2,-1,0x08);
|
|
|
|
|
push @opcode,0x78,0xc2;
|
|
|
|
|
push @opcode,0xc0|($2&7)|(($3&7)<<3); # ModR/M
|
|
|
|
|
my $c=$1;
|
|
|
|
|
push @opcode,$c=~/^0/?oct($c):$c;
|
|
|
|
|
@opcode;
|
|
|
|
|
} else {
|
|
|
|
|
();
|
|
|
|
|
}
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
my $vprotq = sub {
|
|
|
|
|
if (shift =~ /\$([x0-9a-f]+),\s*%xmm([0-9]+),\s*%xmm([0-9]+)/) {
|
|
|
|
|
my @opcode=(0x8f);
|
|
|
|
|
rxb(\@opcode,$3,$2,-1,0x08);
|
|
|
|
|
push @opcode,0x78,0xc3;
|
|
|
|
|
push @opcode,0xc0|($2&7)|(($3&7)<<3); # ModR/M
|
|
|
|
|
my $c=$1;
|
|
|
|
|
push @opcode,$c=~/^0/?oct($c):$c;
|
|
|
|
|
@opcode;
|
|
|
|
|
} else {
|
|
|
|
|
();
|
|
|
|
|
}
|
|
|
|
|
};
|
|
|
|
|
|
2008-07-22 08:42:06 +00:00
|
|
|
|
if ($nasm) {
|
|
|
|
|
print <<___;
|
|
|
|
|
default rel
|
2010-07-26 21:45:38 +00:00
|
|
|
|
%define XMMWORD
|
2014-07-09 18:08:01 +00:00
|
|
|
|
%define YMMWORD
|
|
|
|
|
%define ZMMWORD
|
2008-07-22 08:42:06 +00:00
|
|
|
|
___
|
|
|
|
|
} elsif ($masm) {
|
|
|
|
|
print <<___;
|
|
|
|
|
OPTION DOTNAME
|
|
|
|
|
___
|
|
|
|
|
}
|
2016-02-13 07:53:13 +00:00
|
|
|
|
while(defined($line=<>)) {
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
Perl's chop / chomp considered bad, use a regexp instead
Once upon a time, there was chop, which somply chopped off the last
character of $_ or a given variable, and it was used to take off the
EOL character (\n) of strings.
... but then, you had to check for the presence of such character.
So came chomp, the better chop which checks for \n before chopping it
off. And this worked well, as long as Perl made internally sure that
all EOLs were converted to \n.
These days, though, there seems to be a mixture of perls, so lines
from files in the "wrong" environment might have \r\n as EOL, or just
\r (Mac OS, unless I'm misinformed).
So it's time we went for the more generic variant and use s|\R$||, the
better chomp which recognises all kinds of known EOLs and chops them
off.
A few chops were left alone, as they are use as surgical tools to
remove one last slash or one last comma.
NOTE: \R came with perl 5.10.0. It means that from now on, our
scripts will fail with any older version.
Reviewed-by: Rich Salz <rsalz@openssl.org>
2016-02-11 20:47:30 +00:00
|
|
|
|
$line =~ s|\R$||; # Better chomp
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
2005-05-03 15:35:14 +00:00
|
|
|
|
$line =~ s|[#!].*$||; # get rid of asm-style comments...
|
|
|
|
|
$line =~ s|/\*.*\*/||; # ... and C-style comments...
|
|
|
|
|
$line =~ s|^\s+||; # ... and skip white spaces in beginning
|
2013-12-09 20:23:19 +00:00
|
|
|
|
$line =~ s|\s+$||; # ... and at the end
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
|
|
|
|
undef $label;
|
|
|
|
|
undef $opcode;
|
2008-12-17 19:56:48 +00:00
|
|
|
|
undef @args;
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
|
|
|
|
if ($label=label->re(\$line)) { print $label->out(); }
|
|
|
|
|
|
|
|
|
|
if (directive->re(\$line)) {
|
|
|
|
|
printf "%s",directive->out();
|
2010-07-26 21:45:38 +00:00
|
|
|
|
} elsif ($opcode=opcode->re(\$line)) {
|
|
|
|
|
my $asm = eval("\$".$opcode->mnemonic());
|
|
|
|
|
undef @bytes;
|
|
|
|
|
|
|
|
|
|
if ((ref($asm) eq 'CODE') && scalar(@bytes=&$asm($line))) {
|
|
|
|
|
print $gas?".byte\t":"DB\t",join(',',@bytes),"\n";
|
|
|
|
|
next;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
ARGUMENT: while (1) {
|
2008-12-17 19:56:48 +00:00
|
|
|
|
my $arg;
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
2008-12-17 19:56:48 +00:00
|
|
|
|
if ($arg=register->re(\$line)) { opcode->size($arg->size()); }
|
|
|
|
|
elsif ($arg=const->re(\$line)) { }
|
|
|
|
|
elsif ($arg=ea->re(\$line)) { }
|
|
|
|
|
elsif ($arg=expr->re(\$line)) { }
|
|
|
|
|
else { last ARGUMENT; }
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
2008-12-17 19:56:48 +00:00
|
|
|
|
push @args,$arg;
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
2008-12-17 19:56:48 +00:00
|
|
|
|
last ARGUMENT if ($line !~ /^,/);
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
2008-12-17 19:56:48 +00:00
|
|
|
|
$line =~ s/^,\s*//;
|
2005-04-17 21:05:57 +00:00
|
|
|
|
} # ARGUMENT:
|
|
|
|
|
|
2008-12-17 19:56:48 +00:00
|
|
|
|
if ($#args>=0) {
|
|
|
|
|
my $insn;
|
2010-07-26 21:45:38 +00:00
|
|
|
|
my $sz=opcode->size();
|
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
if ($gas) {
|
2008-12-17 19:56:48 +00:00
|
|
|
|
$insn = $opcode->out($#args>=1?$args[$#args]->size():$sz);
|
2010-04-10 13:55:05 +00:00
|
|
|
|
@args = map($_->out($sz),@args);
|
2010-07-26 21:45:38 +00:00
|
|
|
|
printf "\t%s\t%s",$insn,join(",",@args);
|
2005-05-07 08:13:51 +00:00
|
|
|
|
} else {
|
2008-12-17 19:56:48 +00:00
|
|
|
|
$insn = $opcode->out();
|
2010-05-13 15:26:46 +00:00
|
|
|
|
foreach (@args) {
|
|
|
|
|
my $arg = $_->out();
|
2010-07-26 21:45:38 +00:00
|
|
|
|
# $insn.=$sz compensates for movq, pinsrw, ...
|
|
|
|
|
if ($arg =~ /^xmm[0-9]+$/) { $insn.=$sz; $sz="x" if(!$sz); last; }
|
2013-06-30 21:07:33 +00:00
|
|
|
|
if ($arg =~ /^ymm[0-9]+$/) { $insn.=$sz; $sz="y" if(!$sz); last; }
|
2014-07-09 18:08:01 +00:00
|
|
|
|
if ($arg =~ /^zmm[0-9]+$/) { $insn.=$sz; $sz="z" if(!$sz); last; }
|
2010-07-26 21:45:38 +00:00
|
|
|
|
if ($arg =~ /^mm[0-9]+$/) { $insn.=$sz; $sz="q" if(!$sz); last; }
|
2010-05-13 15:26:46 +00:00
|
|
|
|
}
|
2008-12-17 19:56:48 +00:00
|
|
|
|
@args = reverse(@args);
|
2008-02-13 13:07:52 +00:00
|
|
|
|
undef $sz if ($nasm && $opcode->mnemonic() eq "lea");
|
2010-04-10 13:55:05 +00:00
|
|
|
|
printf "\t%s\t%s",$insn,join(",",map($_->out($sz),@args));
|
2005-04-17 21:05:57 +00:00
|
|
|
|
}
|
|
|
|
|
} else {
|
|
|
|
|
printf "\t%s",$opcode->out();
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
print $line,"\n";
|
|
|
|
|
}
|
|
|
|
|
|
2008-11-12 08:05:58 +00:00
|
|
|
|
print "\n$current_segment\tENDS\n" if ($current_segment && $masm);
|
|
|
|
|
print "END\n" if ($masm);
|
2005-04-17 21:05:57 +00:00
|
|
|
|
|
|
|
|
|
close STDOUT;
|
|
|
|
|
|
2008-10-28 08:40:07 +00:00
|
|
|
|
#################################################
|
2005-04-17 21:05:57 +00:00
|
|
|
|
# Cross-reference x86_64 ABI "card"
|
|
|
|
|
#
|
|
|
|
|
# Unix Win64
|
|
|
|
|
# %rax * *
|
|
|
|
|
# %rbx - -
|
|
|
|
|
# %rcx #4 #1
|
|
|
|
|
# %rdx #3 #2
|
|
|
|
|
# %rsi #2 -
|
|
|
|
|
# %rdi #1 -
|
|
|
|
|
# %rbp - -
|
|
|
|
|
# %rsp - -
|
|
|
|
|
# %r8 #5 #3
|
|
|
|
|
# %r9 #6 #4
|
|
|
|
|
# %r10 * *
|
|
|
|
|
# %r11 * *
|
|
|
|
|
# %r12 - -
|
|
|
|
|
# %r13 - -
|
|
|
|
|
# %r14 - -
|
|
|
|
|
# %r15 - -
|
|
|
|
|
#
|
|
|
|
|
# (*) volatile register
|
|
|
|
|
# (-) preserved by callee
|
|
|
|
|
# (#) Nth argument, volatile
|
|
|
|
|
#
|
|
|
|
|
# In Unix terms top of stack is argument transfer area for arguments
|
2016-02-05 20:23:54 +00:00
|
|
|
|
# which could not be accommodated in registers. Or in other words 7th
|
2005-04-17 21:05:57 +00:00
|
|
|
|
# [integer] argument resides at 8(%rsp) upon function entry point.
|
|
|
|
|
# 128 bytes above %rsp constitute a "red zone" which is not touched
|
|
|
|
|
# by signal handlers and can be used as temporal storage without
|
|
|
|
|
# allocating a frame.
|
|
|
|
|
#
|
|
|
|
|
# In Win64 terms N*8 bytes on top of stack is argument transfer area,
|
|
|
|
|
# which belongs to/can be overwritten by callee. N is the number of
|
|
|
|
|
# arguments passed to callee, *but* not less than 4! This means that
|
|
|
|
|
# upon function entry point 5th argument resides at 40(%rsp), as well
|
|
|
|
|
# as that 32 bytes from 8(%rsp) can always be used as temporal
|
2005-11-06 11:59:39 +00:00
|
|
|
|
# storage [without allocating a frame]. One can actually argue that
|
|
|
|
|
# one can assume a "red zone" above stack pointer under Win64 as well.
|
2005-11-09 17:20:26 +00:00
|
|
|
|
# Point is that at apparently no occasion Windows kernel would alter
|
|
|
|
|
# the area above user stack pointer in true asynchronous manner...
|
2005-04-17 21:05:57 +00:00
|
|
|
|
#
|
|
|
|
|
# All the above means that if assembler programmer adheres to Unix
|
|
|
|
|
# register and stack layout, but disregards the "red zone" existense,
|
|
|
|
|
# it's possible to use following prologue and epilogue to "gear" from
|
|
|
|
|
# Unix to Win64 ABI in leaf functions with not more than 6 arguments.
|
|
|
|
|
#
|
|
|
|
|
# omnipotent_function:
|
|
|
|
|
# ifdef WIN64
|
|
|
|
|
# movq %rdi,8(%rsp)
|
|
|
|
|
# movq %rsi,16(%rsp)
|
|
|
|
|
# movq %rcx,%rdi ; if 1st argument is actually present
|
|
|
|
|
# movq %rdx,%rsi ; if 2nd argument is actually ...
|
|
|
|
|
# movq %r8,%rdx ; if 3rd argument is ...
|
|
|
|
|
# movq %r9,%rcx ; if 4th argument ...
|
|
|
|
|
# movq 40(%rsp),%r8 ; if 5th ...
|
|
|
|
|
# movq 48(%rsp),%r9 ; if 6th ...
|
|
|
|
|
# endif
|
|
|
|
|
# ...
|
|
|
|
|
# ifdef WIN64
|
|
|
|
|
# movq 8(%rsp),%rdi
|
|
|
|
|
# movq 16(%rsp),%rsi
|
|
|
|
|
# endif
|
|
|
|
|
# ret
|
2008-05-03 18:34:59 +00:00
|
|
|
|
#
|
2008-10-28 08:40:07 +00:00
|
|
|
|
#################################################
|
|
|
|
|
# Win64 SEH, Structured Exception Handling.
|
|
|
|
|
#
|
2008-05-03 18:34:59 +00:00
|
|
|
|
# Unlike on Unix systems(*) lack of Win64 stack unwinding information
|
|
|
|
|
# has undesired side-effect at run-time: if an exception is raised in
|
|
|
|
|
# assembler subroutine such as those in question (basically we're
|
|
|
|
|
# referring to segmentation violations caused by malformed input
|
|
|
|
|
# parameters), the application is briskly terminated without invoking
|
|
|
|
|
# any exception handlers, most notably without generating memory dump
|
|
|
|
|
# or any user notification whatsoever. This poses a problem. It's
|
|
|
|
|
# possible to address it by registering custom language-specific
|
|
|
|
|
# handler that would restore processor context to the state at
|
|
|
|
|
# subroutine entry point and return "exception is not handled, keep
|
|
|
|
|
# unwinding" code. Writing such handler can be a challenge... But it's
|
|
|
|
|
# doable, though requires certain coding convention. Consider following
|
|
|
|
|
# snippet:
|
|
|
|
|
#
|
2008-10-28 08:40:07 +00:00
|
|
|
|
# .type function,@function
|
2008-05-03 18:34:59 +00:00
|
|
|
|
# function:
|
|
|
|
|
# movq %rsp,%rax # copy rsp to volatile register
|
|
|
|
|
# pushq %r15 # save non-volatile registers
|
|
|
|
|
# pushq %rbx
|
|
|
|
|
# pushq %rbp
|
|
|
|
|
# movq %rsp,%r11
|
|
|
|
|
# subq %rdi,%r11 # prepare [variable] stack frame
|
|
|
|
|
# andq $-64,%r11
|
|
|
|
|
# movq %rax,0(%r11) # check for exceptions
|
|
|
|
|
# movq %r11,%rsp # allocate [variable] stack frame
|
|
|
|
|
# movq %rax,0(%rsp) # save original rsp value
|
|
|
|
|
# magic_point:
|
|
|
|
|
# ...
|
|
|
|
|
# movq 0(%rsp),%rcx # pull original rsp value
|
|
|
|
|
# movq -24(%rcx),%rbp # restore non-volatile registers
|
|
|
|
|
# movq -16(%rcx),%rbx
|
|
|
|
|
# movq -8(%rcx),%r15
|
|
|
|
|
# movq %rcx,%rsp # restore original rsp
|
|
|
|
|
# ret
|
2008-10-28 08:40:07 +00:00
|
|
|
|
# .size function,.-function
|
2008-05-03 18:34:59 +00:00
|
|
|
|
#
|
|
|
|
|
# The key is that up to magic_point copy of original rsp value remains
|
|
|
|
|
# in chosen volatile register and no non-volatile register, except for
|
|
|
|
|
# rsp, is modified. While past magic_point rsp remains constant till
|
|
|
|
|
# the very end of the function. In this case custom language-specific
|
|
|
|
|
# exception handler would look like this:
|
|
|
|
|
#
|
|
|
|
|
# EXCEPTION_DISPOSITION handler (EXCEPTION_RECORD *rec,ULONG64 frame,
|
|
|
|
|
# CONTEXT *context,DISPATCHER_CONTEXT *disp)
|
2008-10-28 08:40:07 +00:00
|
|
|
|
# { ULONG64 *rsp = (ULONG64 *)context->Rax;
|
|
|
|
|
# if (context->Rip >= magic_point)
|
2008-05-03 18:34:59 +00:00
|
|
|
|
# { rsp = ((ULONG64 **)context->Rsp)[0];
|
|
|
|
|
# context->Rbp = rsp[-3];
|
|
|
|
|
# context->Rbx = rsp[-2];
|
|
|
|
|
# context->R15 = rsp[-1];
|
|
|
|
|
# }
|
|
|
|
|
# context->Rsp = (ULONG64)rsp;
|
|
|
|
|
# context->Rdi = rsp[1];
|
|
|
|
|
# context->Rsi = rsp[2];
|
|
|
|
|
#
|
|
|
|
|
# memcpy (disp->ContextRecord,context,sizeof(CONTEXT));
|
|
|
|
|
# RtlVirtualUnwind(UNW_FLAG_NHANDLER,disp->ImageBase,
|
|
|
|
|
# dips->ControlPc,disp->FunctionEntry,disp->ContextRecord,
|
|
|
|
|
# &disp->HandlerData,&disp->EstablisherFrame,NULL);
|
|
|
|
|
# return ExceptionContinueSearch;
|
|
|
|
|
# }
|
|
|
|
|
#
|
|
|
|
|
# It's appropriate to implement this handler in assembler, directly in
|
|
|
|
|
# function's module. In order to do that one has to know members'
|
|
|
|
|
# offsets in CONTEXT and DISPATCHER_CONTEXT structures and some constant
|
|
|
|
|
# values. Here they are:
|
|
|
|
|
#
|
|
|
|
|
# CONTEXT.Rax 120
|
|
|
|
|
# CONTEXT.Rcx 128
|
|
|
|
|
# CONTEXT.Rdx 136
|
|
|
|
|
# CONTEXT.Rbx 144
|
|
|
|
|
# CONTEXT.Rsp 152
|
|
|
|
|
# CONTEXT.Rbp 160
|
|
|
|
|
# CONTEXT.Rsi 168
|
|
|
|
|
# CONTEXT.Rdi 176
|
|
|
|
|
# CONTEXT.R8 184
|
|
|
|
|
# CONTEXT.R9 192
|
|
|
|
|
# CONTEXT.R10 200
|
|
|
|
|
# CONTEXT.R11 208
|
|
|
|
|
# CONTEXT.R12 216
|
|
|
|
|
# CONTEXT.R13 224
|
|
|
|
|
# CONTEXT.R14 232
|
|
|
|
|
# CONTEXT.R15 240
|
|
|
|
|
# CONTEXT.Rip 248
|
2009-05-12 20:29:27 +00:00
|
|
|
|
# CONTEXT.Xmm6 512
|
2008-05-03 18:34:59 +00:00
|
|
|
|
# sizeof(CONTEXT) 1232
|
|
|
|
|
# DISPATCHER_CONTEXT.ControlPc 0
|
|
|
|
|
# DISPATCHER_CONTEXT.ImageBase 8
|
|
|
|
|
# DISPATCHER_CONTEXT.FunctionEntry 16
|
|
|
|
|
# DISPATCHER_CONTEXT.EstablisherFrame 24
|
|
|
|
|
# DISPATCHER_CONTEXT.TargetIp 32
|
|
|
|
|
# DISPATCHER_CONTEXT.ContextRecord 40
|
|
|
|
|
# DISPATCHER_CONTEXT.LanguageHandler 48
|
|
|
|
|
# DISPATCHER_CONTEXT.HandlerData 56
|
|
|
|
|
# UNW_FLAG_NHANDLER 0
|
|
|
|
|
# ExceptionContinueSearch 1
|
|
|
|
|
#
|
2008-10-28 08:40:07 +00:00
|
|
|
|
# In order to tie the handler to the function one has to compose
|
|
|
|
|
# couple of structures: one for .xdata segment and one for .pdata.
|
|
|
|
|
#
|
2008-05-03 18:34:59 +00:00
|
|
|
|
# UNWIND_INFO structure for .xdata segment would be
|
2008-10-28 08:40:07 +00:00
|
|
|
|
#
|
|
|
|
|
# function_unwind_info:
|
|
|
|
|
# .byte 9,0,0,0
|
2008-11-12 08:05:58 +00:00
|
|
|
|
# .rva handler
|
2008-10-28 08:40:07 +00:00
|
|
|
|
#
|
|
|
|
|
# This structure designates exception handler for a function with
|
|
|
|
|
# zero-length prologue, no stack frame or frame register.
|
|
|
|
|
#
|
|
|
|
|
# To facilitate composing of .pdata structures, auto-generated "gear"
|
|
|
|
|
# prologue copies rsp value to rax and denotes next instruction with
|
2008-11-12 08:05:58 +00:00
|
|
|
|
# .LSEH_begin_{function_name} label. This essentially defines the SEH
|
2008-10-28 08:40:07 +00:00
|
|
|
|
# styling rule mentioned in the beginning. Position of this label is
|
|
|
|
|
# chosen in such manner that possible exceptions raised in the "gear"
|
|
|
|
|
# prologue would be accounted to caller and unwound from latter's frame.
|
2008-11-12 08:05:58 +00:00
|
|
|
|
# End of function is marked with respective .LSEH_end_{function_name}
|
2008-10-28 08:40:07 +00:00
|
|
|
|
# label. To summarize, .pdata segment would contain
|
|
|
|
|
#
|
2008-11-12 08:05:58 +00:00
|
|
|
|
# .rva .LSEH_begin_function
|
|
|
|
|
# .rva .LSEH_end_function
|
|
|
|
|
# .rva function_unwind_info
|
2008-10-28 08:40:07 +00:00
|
|
|
|
#
|
2013-06-12 23:22:32 +00:00
|
|
|
|
# Reference to function_unwind_info from .xdata segment is the anchor.
|
2008-11-12 08:05:58 +00:00
|
|
|
|
# In case you wonder why references are 32-bit .rvas and not 64-bit
|
2008-10-28 08:40:07 +00:00
|
|
|
|
# .quads. References put into these two segments are required to be
|
|
|
|
|
# *relative* to the base address of the current binary module, a.k.a.
|
|
|
|
|
# image base. No Win64 module, be it .exe or .dll, can be larger than
|
|
|
|
|
# 2GB and thus such relative references can be and are accommodated in
|
|
|
|
|
# 32 bits.
|
|
|
|
|
#
|
|
|
|
|
# Having reviewed the example function code, one can argue that "movq
|
|
|
|
|
# %rsp,%rax" above is redundant. It is not! Keep in mind that on Unix
|
|
|
|
|
# rax would contain an undefined value. If this "offends" you, use
|
|
|
|
|
# another register and refrain from modifying rax till magic_point is
|
|
|
|
|
# reached, i.e. as if it was a non-volatile register. If more registers
|
|
|
|
|
# are required prior [variable] frame setup is completed, note that
|
|
|
|
|
# nobody says that you can have only one "magic point." You can
|
|
|
|
|
# "liberate" non-volatile registers by denoting last stack off-load
|
|
|
|
|
# instruction and reflecting it in finer grade unwind logic in handler.
|
|
|
|
|
# After all, isn't it why it's called *language-specific* handler...
|
|
|
|
|
#
|
|
|
|
|
# Attentive reader can notice that exceptions would be mishandled in
|
|
|
|
|
# auto-generated "gear" epilogue. Well, exception effectively can't
|
|
|
|
|
# occur there, because if memory area used by it was subject to
|
|
|
|
|
# segmentation violation, then it would be raised upon call to the
|
|
|
|
|
# function (and as already mentioned be accounted to caller, which is
|
|
|
|
|
# not a problem). If you're still not comfortable, then define tail
|
|
|
|
|
# "magic point" just prior ret instruction and have handler treat it...
|
2008-05-03 18:34:59 +00:00
|
|
|
|
#
|
|
|
|
|
# (*) Note that we're talking about run-time, not debug-time. Lack of
|
|
|
|
|
# unwind information makes debugging hard on both Windows and
|
|
|
|
|
# Unix. "Unlike" referes to the fact that on Unix signal handler
|
|
|
|
|
# will always be invoked, core dumped and appropriate exit code
|
|
|
|
|
# returned to parent (for user notification).
|