HLE: Split the LOCK and REP prefix slots
[nasm/nasm.git] / pptok.pl
blobbe85b9427ca2d4886a1d8d8ba4a229c3adf19934
1 #!/usr/bin/perl
2 ## --------------------------------------------------------------------------
3 ##
4 ## Copyright 1996-2009 The NASM Authors - All Rights Reserved
5 ## See the file AUTHORS included with the NASM distribution for
6 ## the specific copyright holders.
7 ##
8 ## Redistribution and use in source and binary forms, with or without
9 ## modification, are permitted provided that the following
10 ## conditions are met:
12 ## * Redistributions of source code must retain the above copyright
13 ## notice, this list of conditions and the following disclaimer.
14 ## * Redistributions in binary form must reproduce the above
15 ## copyright notice, this list of conditions and the following
16 ## disclaimer in the documentation and/or other materials provided
17 ## with the distribution.
18 ##
19 ## THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
20 ## CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
21 ## INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
22 ## MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
23 ## DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
24 ## CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
25 ## SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
26 ## NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
27 ## LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
28 ## HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
29 ## CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
30 ## OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE,
31 ## EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
33 ## --------------------------------------------------------------------------
36 # Produce pptok.c, pptok.h and pptok.ph from pptok.dat
39 require 'phash.ph';
41 my($what, $in, $out) = @ARGV;
44 # Read pptok.dat
46 open(IN, "< $in") or die "$0: cannot open: $in\n";
47 while (defined($line = <IN>)) {
48 chomp $line;
49 $line =~ s/^\s+//; # Remove leading whitespace
50 $line =~ s/\s*\#.*$//; # Remove comments and trailing whitespace
51 next if ($line eq '');
53 if ($line =~ /^\%(.*)\*$/) {
54 push(@cctok, $1);
55 } elsif ($line =~ /^\%(.*)$/) {
56 push(@pptok, $1);
57 } elsif ($line =~ /^\*(.*)$/) {
58 push(@cond, $1);
61 close(IN);
63 @cctok = sort @cctok;
64 @cond = sort @cond;
65 @pptok = sort @pptok;
67 # Generate the expanded list including conditionals. The conditionals
68 # are at the beginning, padded to a power of 2, with the inverses
69 # interspersed; this allows a simple mask to pick out the condition.
71 while ((scalar @cond) & (scalar @cond)-1) {
72 push(@cond, undef);
75 @cptok = ();
76 foreach $ct (@cctok) {
77 foreach $cc (@cond) {
78 if (defined($cc)) {
79 push(@cptok, $ct.$cc);
80 push(@cptok, $ct.'n'.$cc);
81 } else {
82 push(@cptok, undef, undef);
86 $first_uncond = $pptok[0];
87 @pptok = (@cptok, @pptok);
89 open(OUT, "> $out") or die "$0: cannot open: $out\n";
92 # Output pptok.h
94 if ($what eq 'h') {
95 print OUT "/* Automatically generated from $in by $0 */\n";
96 print OUT "/* Do not edit */\n";
97 print OUT "\n";
99 print OUT "enum preproc_token {\n";
100 $n = 0;
101 foreach $pt (@pptok) {
102 if (defined($pt)) {
103 printf OUT " %-16s = %3d,\n", "PP_\U$pt\E", $n;
105 $n++;
107 printf OUT " %-16s = %3d\n", 'PP_INVALID', -1;
108 print OUT "};\n";
109 print OUT "\n";
111 print OUT "enum pp_conditional {\n";
112 $n = 0;
113 foreach $cc (@cond) {
114 if (defined($cc)) {
115 printf OUT " %-16s = %3d,\n", "PPC_IF\U$cc\E", $n;
117 $n += 2;
119 print OUT "};\n\n";
121 printf OUT "#define PP_COND(x) ((enum pp_conditional)((x) & 0x%x))\n",
122 (scalar(@cond)-1) << 1;
123 print OUT "#define PP_IS_COND(x) ((unsigned int)(x) < PP_\U$first_uncond\E)\n";
124 print OUT "#define PP_NEGATIVE(x) ((x) & 1)\n";
125 print OUT "\n";
127 foreach $ct (@cctok) {
128 print OUT "#define CASE_PP_\U$ct\E";
129 $pref = " \\\n";
130 foreach $cc (@cond) {
131 if (defined($cc)) {
132 print OUT "$pref\tcase PP_\U${ct}${cc}\E: \\\n";
133 print OUT "\tcase PP_\U${ct}N${cc}\E";
134 $pref = ":\\\n";
137 print OUT "\n"; # No colon or newline on the last one
142 # Output pptok.c
144 if ($what eq 'c') {
145 print OUT "/* Automatically generated from $in by $0 */\n";
146 print OUT "/* Do not edit */\n";
147 print OUT "\n";
149 my %tokens = ();
150 my @tokendata = ();
152 my $n = 0;
153 foreach $pt (@pptok) {
154 if (defined($pt)) {
155 $tokens{'%'.$pt} = $n;
156 if ($pt =~ /[\@\[\]\\_]/) {
157 # Fail on characters which look like upper-case letters
158 # to the quick-and-dirty downcasing in the prehash
159 # (see below)
160 die "$in: invalid character in token: $pt";
163 $n++;
166 my @hashinfo = gen_perfect_hash(\%tokens);
167 if (!@hashinfo) {
168 die "$0: no hash found\n";
171 # Paranoia...
172 verify_hash_table(\%tokens, \@hashinfo);
174 ($n, $sv, $g) = @hashinfo;
175 $sv2 = $sv+2;
177 die if ($n & ($n-1));
179 print OUT "#include \"compiler.h\"\n";
180 print OUT "#include <inttypes.h>\n";
181 print OUT "#include <ctype.h>\n";
182 print OUT "#include \"nasmlib.h\"\n";
183 print OUT "#include \"hashtbl.h\"\n";
184 print OUT "#include \"preproc.h\"\n";
185 print OUT "\n";
187 # Note that this is global.
188 printf OUT "const char * const pp_directives[%d] = {\n", scalar(@pptok);
189 foreach $d (@pptok) {
190 if (defined($d)) {
191 print OUT " \"%$d\",\n";
192 } else {
193 print OUT " NULL,\n";
196 print OUT "};\n";
198 printf OUT "const uint8_t pp_directives_len[%d] = {\n", scalar(@pptok);
199 foreach $d (@pptok) {
200 printf OUT " %d,\n", defined($d) ? length($d)+1 : 0;
202 print OUT "};\n";
204 print OUT "enum preproc_token pp_token_hash(const char *token)\n";
205 print OUT "{\n";
207 # Put a large value in unused slots. This makes it extremely unlikely
208 # that any combination that involves unused slot will pass the range test.
209 # This speeds up rejection of unrecognized tokens, i.e. identifiers.
210 print OUT "#define UNUSED 16383\n";
212 print OUT " static const int16_t hash1[$n] = {\n";
213 for ($i = 0; $i < $n; $i++) {
214 my $h = ${$g}[$i*2+0];
215 print OUT " ", defined($h) ? $h : 'UNUSED', ",\n";
217 print OUT " };\n";
219 print OUT " static const int16_t hash2[$n] = {\n";
220 for ($i = 0; $i < $n; $i++) {
221 my $h = ${$g}[$i*2+1];
222 print OUT " ", defined($h) ? $h : 'UNUSED', ",\n";
224 print OUT " };\n";
226 print OUT " uint32_t k1, k2;\n";
227 print OUT " uint64_t crc;\n";
228 # For correct overflow behavior, "ix" should be unsigned of the same
229 # width as the hash arrays.
230 print OUT " uint16_t ix;\n";
231 print OUT "\n";
233 printf OUT " crc = crc64i(UINT64_C(0x%08x%08x), token);\n",
234 $$sv[0], $$sv[1];
235 print OUT " k1 = (uint32_t)crc;\n";
236 print OUT " k2 = (uint32_t)(crc >> 32);\n";
237 print OUT "\n";
238 printf OUT " ix = hash1[k1 & 0x%x] + hash2[k2 & 0x%x];\n", $n-1, $n-1;
239 printf OUT " if (ix >= %d)\n", scalar(@pptok);
240 print OUT " return PP_INVALID;\n";
241 print OUT "\n";
243 print OUT " if (!pp_directives[ix] || nasm_stricmp(pp_directives[ix], token))\n";
244 print OUT " return PP_INVALID;\n";
245 print OUT "\n";
246 print OUT " return ix;\n";
247 print OUT "}\n";
251 # Output pptok.ph
253 if ($what eq 'ph') {
254 print OUT "# Automatically generated from $in by $0\n";
255 print OUT "# Do not edit\n";
256 print OUT "\n";
258 print OUT "%pptok_hash = (\n";
259 $n = 0;
260 foreach $tok (@pptok) {
261 if (defined($tok)) {
262 printf OUT " '%%%s' => %d,\n", $tok, $n;
264 $n++;
266 print OUT ");\n";
267 print OUT "1;\n";