tokhash.pl

   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:
  11 ##
  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.
  32 ##
  33 ## --------------------------------------------------------------------------
  34
  35 #
  36 # Generate a perfect hash for token parsing
  37 #
  38 # Usage: tokenhash.pl insns.dat regs.dat tokens.dat
  39 #
  40
  41 require 'phash.ph';
  42
  43 my($output, $insns_dat, $regs_dat, $tokens_dat) = @ARGV;
  44
  45 %tokens = ();
  46 @tokendata = ();
  47
  48 #
  49 # List of condition codes
  50 #
  51 @conditions = ('a', 'ae', 'b', 'be', 'c', 'e', 'g', 'ge', 'l', 'le',
  52                'na', 'nae', 'nb', 'nbe', 'nc', 'ne', 'ng', 'nge', 'nl',
  53                'nle', 'no', 'np', 'ns', 'nz', 'o', 'p', 'pe', 'po', 's', 'z');
  54
  55 #
  56 # Read insns.dat
  57 #
  58 open(ID, "< ${insns_dat}") or die "$0: cannot open $insns_dat: $!\n";
  59 while (defined($line = <ID>)) {
  60     if ($line =~ /^([A-Z0-9_]+)(|cc)\s/) {
  61         $insn = $1.$2;
  62         ($token = $1) =~ tr/A-Z/a-z/;
  63
  64         if ($2 eq '') {
  65             # Single instruction token
  66             if (!defined($tokens{$token})) {
  67                 $tokens{$token} = scalar @tokendata;
  68                 push(@tokendata, "\"${token}\", TOKEN_INSN, C_none, 0, I_${insn}");
  69             }
  70         } else {
  71             # Conditional instruction
  72             foreach $cc (@conditions) {
  73                 if (!defined($tokens{$token.$cc})) {
  74                     $tokens{$token.$cc} = scalar @tokendata;
  75                     push(@tokendata, "\"${token}${cc}\", TOKEN_INSN, C_\U$cc\E, 0, I_${insn}");
  76                 }
  77             }
  78         }
  79     }
  80 }
  81 close(ID);
  82
  83 #
  84 # Read regs.dat
  85 #
  86 open(RD, "< ${regs_dat}") or die "$0: cannot open $regs_dat: $!\n";
  87 while (defined($line = <RD>)) {
  88     if ($line =~ /^([a-z0-9_-]+)\s*\S+\s*\S+\s*[0-9]+\s*(\S*)/) {
  89         $reg = $1;
  90         $reg_flag = $2;
  91
  92         if ($reg =~ /^(.*[^0-9])([0-9]+)\-([0-9]+)(|[^0-9].*)$/) {
  93             $nregs = $3-$2+1;
  94             $reg = $1.$2.$4;
  95             $reg_nr = $2;
  96             $reg_prefix = $1;
  97             $reg_suffix = $4;
  98         } else {
  99             $nregs = 1;
 100             undef $reg_prefix, $reg_suffix;
 101         }
 102
 103         while ($nregs--) {
 104             if (defined($tokens{$reg})) {
 105                 die "Duplicate definition: $reg\n";
 106             }
 107             $tokens{$reg} = scalar @tokendata;
 108             if ($reg_flag eq '') {
 109                 push(@tokendata, "\"${reg}\", TOKEN_REG, 0, 0, R_\U${reg}\E");
 110             } else {
 111                 push(@tokendata, "\"${reg}\", TOKEN_REG, 0, ${reg_flag}, R_\U${reg}\E");
 112             }
 113
 114             if (defined($reg_prefix)) {
 115                 $reg_nr++;
 116                 $reg = sprintf("%s%u%s", $reg_prefix, $reg_nr, $reg_suffix);
 117             } else {
 118                 # Not a dashed sequence
 119                 die if ($nregs);
 120             }
 121         }
 122     }
 123 }
 124 close(RD);
 125
 126 #
 127 # Read tokens.dat
 128 #
 129 open(TD, "< ${tokens_dat}") or die "$0: cannot open $tokens_dat: $!\n";
 130 while (defined($line = <TD>)) {
 131     if ($line =~ /^\%\s+(.*)$/) {
 132         $pattern = $1;
 133     } elsif ($line =~ /^([a-z0-9_-]+)/) {
 134         $token = $1;
 135
 136         if (defined($tokens{$token})) {
 137             die "Duplicate definition: $token\n";
 138         }
 139         $tokens{$token} = scalar @tokendata;
 140
 141         $data = $pattern;
 142         if ($data =~ /^(.*)\{(.*)\}(.*)$/) {
 143             my $head = $1, $tail = $3;
 144             my $px = $2;
 145
 146             $px =~ s/\*/(.*)/g;
 147             if ($token =~ /$px/i) {
 148                 $data = $head."\U$1".$tail;
 149             } else {
 150                 die "$0: token $token doesn't match $px\n";
 151             }
 152         }
 153
 154         $data =~ s/\*/\U$token/g;
 155
 156         push(@tokendata, "\"$token\", $data");
 157     }
 158 }
 159 close(TD);
 160
 161 if ($output eq 'h') {
 162     #
 163     # tokens.h
 164     #
 165
 166     $max_len = 0;
 167     foreach $token (keys(%tokens)) {
 168         if (length($token) > $max_len) {
 169             $max_len = length($token);
 170         }
 171     }
 172
 173     print "/*\n";
 174     print " * This file is generated from insns.dat, regs.dat and token.dat\n";
 175     print " * by tokhash.pl; do not edit.\n";
 176     print " */\n";
 177     print "\n";
 178
 179     print "#ifndef NASM_TOKENS_H\n";
 180     print "#define NASM_TOKENS_H\n";
 181     print "\n";
 182     print "#define MAX_KEYWORD $max_len /* length of longest keyword */\n";
 183     print "\n";
 184     print "#endif /* NASM_TOKENS_H */\n";
 185 } elsif ($output eq 'c') {
 186     #
 187     # tokhash.c
 188     #
 189
 190     @hashinfo = gen_perfect_hash(\%tokens);
 191     if (!@hashinfo) {
 192         die "$0: no hash found\n";
 193     }
 194
 195     # Paranoia...
 196     verify_hash_table(\%tokens, \@hashinfo);
 197
 198     ($n, $sv, $g) = @hashinfo;
 199     $sv2 = $sv+2;
 200
 201     die if ($n & ($n-1));
 202
 203     print "/*\n";
 204     print " * This file is generated from insns.dat, regs.dat and token.dat\n";
 205     print " * by tokhash.pl; do not edit.\n";
 206     print " */\n";
 207     print "\n";
 208
 209     print "#include \"compiler.h\"\n";
 210     print "#include <string.h>\n";
 211     print "#include \"nasm.h\"\n";
 212     print "#include \"hashtbl.h\"\n";
 213     print "#include \"insns.h\"\n";
 214     print "\n";
 215
 216     # These somewhat odd sizes and ordering thereof are due to the
 217     # relative ranges of the types; this makes it fit in 16 bytes on
 218     # 64-bit machines and 12 bytes on 32-bit machines.
 219     print "struct tokendata {\n";
 220     print "    const char *string;\n";
 221     print "    int16_t tokentype;\n";
 222     print "    int8_t aux;\n";
 223     print "    int8_t tokflag;\n";
 224     print "    int32_t num;\n";
 225     print "};\n";
 226     print "\n";
 227
 228     print "int nasm_token_hash(const char *token, struct tokenval *tv)\n";
 229     print "{\n";
 230
 231     # Put a large value in unused slots.  This makes it extremely unlikely
 232     # that any combination that involves unused slot will pass the range test.
 233     # This speeds up rejection of unrecognized tokens, i.e. identifiers.
 234     print "#define UNUSED 16383\n";
 235
 236     print "    static const int16_t hash1[$n] = {\n";
 237     for ($i = 0; $i < $n; $i++) {
 238         my $h = ${$g}[$i*2+0];
 239         print "        ", defined($h) ? $h : 'UNUSED', ",\n";
 240     }
 241     print "    };\n";
 242
 243     print "    static const int16_t hash2[$n] = {\n";
 244     for ($i = 0; $i < $n; $i++) {
 245         my $h = ${$g}[$i*2+1];
 246         print "        ", defined($h) ? $h : 'UNUSED', ",\n";
 247     }
 248     print "    };\n";
 249
 250     printf "    static const struct tokendata tokendata[%d] = {\n", scalar(@tokendata);
 251     foreach $d (@tokendata) {
 252         print "        { ", $d, " },\n";
 253     }
 254     print  "    };\n";
 255
 256     print  "    uint32_t k1, k2;\n";
 257     print  "    uint64_t crc;\n";
 258     # For correct overflow behavior, "ix" should be unsigned of the same
 259     # width as the hash arrays.
 260     print  "    uint16_t ix;\n";
 261     print  "    const struct tokendata *data;\n";
 262     print  "\n";
 263     printf "    tv->t_flag = 0;\n";
 264     printf "    crc = crc64(UINT64_C(0x%08x%08x), token);\n",
 265         $$sv[0], $$sv[1];
 266     print  "    k1 = (uint32_t)crc;\n";
 267     print  "    k2 = (uint32_t)(crc >> 32);\n";
 268     print  "\n";
 269     printf "    ix = hash1[k1 & 0x%x] + hash2[k2 & 0x%x];\n", $n-1, $n-1;
 270     printf "    if (ix >= %d)\n", scalar(@tokendata);
 271     print  "        return tv->t_type = TOKEN_ID;\n";
 272     print  "\n";
 273     print  "    data = &tokendata[ix];\n";
 274
 275     print  "    if (strcmp(data->string, token))\n";
 276     print  "        return tv->t_type = TOKEN_ID;\n";
 277     print  "\n";
 278     print  "    tv->t_integer = data->num;\n";
 279     print  "    tv->t_inttwo  = data->aux;\n";
 280     print  "    tv->t_flag    = data->tokflag;\n";
 281     print  "    return tv->t_type = data->tokentype;\n";
 282     print  "}\n";
 283 }