directiv.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 directive parsing
  37 #
  38 # Usage:
  39 #      directiv.pl h directiv.dat directiv.h (to generate C header)
  40 #      directiv.pl c directiv.dat directiv.c (to generate C source)
  41 #
  42
  43 require 'phash.ph';
  44
  45 my($output, $directives_dat, $outfile) = @ARGV;
  46
  47 @directives = ();
  48 @specials   = ('none', 'unknown');
  49
  50 open(DD, "< ${directives_dat}\0")
  51     or die "$0: cannot open: ${directives_dat}: $!\n";
  52 while (defined($line = <DD>)) {
  53     chomp $line;
  54     if ($line =~ /^\s*([[:alnum:]]+)\s*(|[\;\#].*)$/) {
  55         push(@directives, $1);
  56     }
  57 }
  58 close(DD);
  59
  60 if ($output eq 'h') {
  61     open(H, "> ${outfile}\0")
  62         or die "$0: cannot create: ${outfile}: $!\n";
  63
  64     print H "/*\n";
  65     print H " * This     file is generated from directiv.dat\n";
  66     print H " * by directiv.pl; do not edit.\n";
  67     print H " */\n";
  68     print H "\n";
  69
  70     print H "#ifndef NASM_DIRECTIVES_H\n";
  71     print H "#define NASM_DIRECTIVES_H\n";
  72     print H "\n";
  73
  74     $c = '{';
  75     print H "enum directives ";
  76     foreach $d (@specials) {
  77         print H "$c\n    D_$d";
  78         $c = ',';
  79     }
  80     foreach $d (@directives) {
  81         print H "$c\n    D_\U$d";
  82         $c = ',';
  83     }
  84     print H "\n};\n\n";
  85     printf H "extern const char * const directives[%d];\n",
  86         scalar(@directives)+scalar(@specials);
  87     print H "enum directives find_directive(const char *token);\n\n";
  88     print H "#endif /* NASM_DIRECTIVES_H */\n";
  89 } elsif ($output eq 'c') {
  90     %directive = ();
  91     $n = 0;
  92     foreach $d (@directives) {
  93         if (exists($directive{$d})) {
  94             die "$0: $directives_dat: duplicate directive: $d\n";
  95         }
  96         $directive{$d} = $n++;  # This is zero-based, unlike the enum!
  97     }
  98
  99     @hashinfo = gen_perfect_hash(\%directive);
 100     if (!@hashinfo) {
 101         die "$0: no hash found\n";
 102     }
 103
 104     # Paranoia...
 105     verify_hash_table(\%directive, \@hashinfo);
 106
 107     ($n, $sv, $g) = @hashinfo;
 108
 109     die if ($n & ($n-1));
 110
 111     open(C, "> ${outfile}\0")
 112         or die "$0: cannot create: ${directives_c}: $!\n";
 113
 114     print C "/*\n";
 115     print C " * This file is generated from directiv.dat\n";
 116     print C " * by directiv.pl; do not edit.\n";
 117     print C " */\n";
 118     print C "\n";
 119
 120     print C "#include \"compiler.h\"\n";
 121     print C "#include <string.h>\n";
 122     print C "#include \"nasm.h\"\n";
 123     print C "#include \"hashtbl.h\"\n";
 124     print C "#include \"directiv.h\"\n";
 125     print C "\n";
 126
 127     printf C "const char * const directives[%d] =\n",
 128         scalar(@directives)+scalar(@specials);
 129     $c = '{';
 130     foreach $d (@specials) {
 131         print C "$c\n    NULL";
 132         $c = ',';
 133     }
 134     foreach $d (@directives) {
 135         print C "$c\n    \"$d\"";
 136         $c = ',';
 137     }
 138     print C "\n};\n\n";
 139
 140     print C "enum directives find_directive(const char *token)\n";
 141     print C "{\n";
 142
 143     # Put a large value in unused slots.  This makes it extremely unlikely
 144     # that any combination that involves unused slot will pass the range test.
 145     # This speeds up rejection of unrecognized tokens, i.e. identifiers.
 146     print C "#define UNUSED (65535/3)\n";
 147
 148     print C "    static const int16_t hash1[$n] = {\n";
 149     for ($i = 0; $i < $n; $i++) {
 150         my $h = ${$g}[$i*2+0];
 151         print C "        ", defined($h) ? $h : 'UNUSED', ",\n";
 152     }
 153     print C "    };\n";
 154
 155     print C "    static const int16_t hash2[$n] = {\n";
 156     for ($i = 0; $i < $n; $i++) {
 157         my $h = ${$g}[$i*2+1];
 158         print C "        ", defined($h) ? $h : 'UNUSED', ",\n";
 159     }
 160     print C "    };\n";
 161
 162     print C  "    uint32_t k1, k2;\n";
 163     print C  "    uint64_t crc;\n";
 164     # For correct overflow behavior, "ix" should be unsigned of the same
 165     # width as the hash arrays.
 166     print C  "    uint16_t ix;\n";
 167     print C  "\n";
 168     printf C "    crc = crc64i(UINT64_C(0x%08x%08x), token);\n",
 169         $$sv[0], $$sv[1];
 170     print C  "    k1 = (uint32_t)crc;\n";
 171     print C  "    k2 = (uint32_t)(crc >> 32);\n";
 172     print C  "\n";
 173     printf C "    ix = hash1[k1 & 0x%x] + hash2[k2 & 0x%x];\n", $n-1, $n-1;
 174     printf C "    if (ix >= %d)\n", scalar(@directives);
 175     print C  "        return D_unknown;\n";
 176     print C  "\n";
 177     printf C "    ix += %d;\n", scalar(@specials);
 178     print C  "    if (nasm_stricmp(token, directives[ix]))\n";
 179     print C  "        return D_unknown;\n";
 180     print C  "\n";
 181     print C  "    return ix;\n";
 182     print C  "}\n";
 183 }