Back to home page

OSCL-LXR
 
 

     

0001 #!/usr/bin/env perl
0002 # SPDX-License-Identifier: GPL-2.0
0003 
0004 BEGIN { $Pod::Usage::Formatter = 'Pod::Text::Termcap'; }
0005 
0006 use strict;
0007 use warnings;
0008 use utf8;
0009 use Pod::Usage qw(pod2usage);
0010 use Getopt::Long;
0011 use File::Find;
0012 use IO::Handle;
0013 use Fcntl ':mode';
0014 use Cwd 'abs_path';
0015 use Data::Dumper;
0016 
0017 my $help = 0;
0018 my $hint = 0;
0019 my $man = 0;
0020 my $debug = 0;
0021 my $enable_lineno = 0;
0022 my $show_warnings = 1;
0023 my $prefix="Documentation/ABI";
0024 my $sysfs_prefix="/sys";
0025 my $search_string;
0026 
0027 # Debug options
0028 my $dbg_what_parsing = 1;
0029 my $dbg_what_open = 2;
0030 my $dbg_dump_abi_structs = 4;
0031 my $dbg_undefined = 8;
0032 
0033 $Data::Dumper::Indent = 1;
0034 $Data::Dumper::Terse = 1;
0035 
0036 #
0037 # If true, assumes that the description is formatted with ReST
0038 #
0039 my $description_is_rst = 1;
0040 
0041 GetOptions(
0042     "debug=i" => \$debug,
0043     "enable-lineno" => \$enable_lineno,
0044     "rst-source!" => \$description_is_rst,
0045     "dir=s" => \$prefix,
0046     'help|?' => \$help,
0047     "show-hints" => \$hint,
0048     "search-string=s" => \$search_string,
0049     man => \$man
0050 ) or pod2usage(2);
0051 
0052 pod2usage(1) if $help;
0053 pod2usage(-exitstatus => 0, -noperldoc, -verbose => 2) if $man;
0054 
0055 pod2usage(2) if (scalar @ARGV < 1 || @ARGV > 2);
0056 
0057 my ($cmd, $arg) = @ARGV;
0058 
0059 pod2usage(2) if ($cmd ne "search" && $cmd ne "rest" && $cmd ne "validate" && $cmd ne "undefined");
0060 pod2usage(2) if ($cmd eq "search" && !$arg);
0061 
0062 require Data::Dumper if ($debug & $dbg_dump_abi_structs);
0063 
0064 my %data;
0065 my %symbols;
0066 
0067 #
0068 # Displays an error message, printing file name and line
0069 #
0070 sub parse_error($$$$) {
0071     my ($file, $ln, $msg, $data) = @_;
0072 
0073     return if (!$show_warnings);
0074 
0075     $data =~ s/\s+$/\n/;
0076 
0077     print STDERR "Warning: file $file#$ln:\n\t$msg";
0078 
0079     if ($data ne "") {
0080         print STDERR ". Line\n\t\t$data";
0081     } else {
0082         print STDERR "\n";
0083     }
0084 }
0085 
0086 #
0087 # Parse an ABI file, storing its contents at %data
0088 #
0089 sub parse_abi {
0090     my $file = $File::Find::name;
0091 
0092     my $mode = (stat($file))[2];
0093     return if ($mode & S_IFDIR);
0094     return if ($file =~ m,/README,);
0095     return if ($file =~ m,/\.,);
0096 
0097     my $name = $file;
0098     $name =~ s,.*/,,;
0099 
0100     my $fn = $file;
0101     $fn =~ s,Documentation/ABI/,,;
0102 
0103     my $nametag = "File $fn";
0104     $data{$nametag}->{what} = "File $name";
0105     $data{$nametag}->{type} = "File";
0106     $data{$nametag}->{file} = $name;
0107     $data{$nametag}->{filepath} = $file;
0108     $data{$nametag}->{is_file} = 1;
0109     $data{$nametag}->{line_no} = 1;
0110 
0111     my $type = $file;
0112     $type =~ s,.*/(.*)/.*,$1,;
0113 
0114     my $what;
0115     my $new_what;
0116     my $tag = "";
0117     my $ln;
0118     my $xrefs;
0119     my $space;
0120     my @labels;
0121     my $label = "";
0122 
0123     print STDERR "Opening $file\n" if ($debug & $dbg_what_open);
0124     open IN, $file;
0125     while(<IN>) {
0126         $ln++;
0127         if (m/^(\S+)(:\s*)(.*)/i) {
0128             my $new_tag = lc($1);
0129             my $sep = $2;
0130             my $content = $3;
0131 
0132             if (!($new_tag =~ m/(what|where|date|kernelversion|contact|description|users)/)) {
0133                 if ($tag eq "description") {
0134                     # New "tag" is actually part of
0135                     # description. Don't consider it a tag
0136                     $new_tag = "";
0137                 } elsif ($tag ne "") {
0138                     parse_error($file, $ln, "tag '$tag' is invalid", $_);
0139                 }
0140             }
0141 
0142             # Invalid, but it is a common mistake
0143             if ($new_tag eq "where") {
0144                 parse_error($file, $ln, "tag 'Where' is invalid. Should be 'What:' instead", "");
0145                 $new_tag = "what";
0146             }
0147 
0148             if ($new_tag =~ m/what/) {
0149                 $space = "";
0150                 $content =~ s/[,.;]$//;
0151 
0152                 push @{$symbols{$content}->{file}}, " $file:" . ($ln - 1);
0153 
0154                 if ($tag =~ m/what/) {
0155                     $what .= "\xac" . $content;
0156                 } else {
0157                     if ($what) {
0158                         parse_error($file, $ln, "What '$what' doesn't have a description", "") if (!$data{$what}->{description});
0159 
0160                         foreach my $w(split /\xac/, $what) {
0161                             $symbols{$w}->{xref} = $what;
0162                         };
0163                     }
0164 
0165                     $what = $content;
0166                     $label = $content;
0167                     $new_what = 1;
0168                 }
0169                 push @labels, [($content, $label)];
0170                 $tag = $new_tag;
0171 
0172                 push @{$data{$nametag}->{symbols}}, $content if ($data{$nametag}->{what});
0173                 next;
0174             }
0175 
0176             if ($tag ne "" && $new_tag) {
0177                 $tag = $new_tag;
0178 
0179                 if ($new_what) {
0180                     @{$data{$what}->{label_list}} = @labels if ($data{$nametag}->{what});
0181                     @labels = ();
0182                     $label = "";
0183                     $new_what = 0;
0184 
0185                     $data{$what}->{type} = $type;
0186                     if (!defined($data{$what}->{file})) {
0187                         $data{$what}->{file} = $name;
0188                         $data{$what}->{filepath} = $file;
0189                     } else {
0190                         $data{$what}->{description} .= "\n\n" if (defined($data{$what}->{description}));
0191                         if ($name ne $data{$what}->{file}) {
0192                             $data{$what}->{file} .= " " . $name;
0193                             $data{$what}->{filepath} .= " " . $file;
0194                         }
0195                     }
0196                     print STDERR "\twhat: $what\n" if ($debug & $dbg_what_parsing);
0197                     $data{$what}->{line_no} = $ln;
0198                 } else {
0199                     $data{$what}->{line_no} = $ln if (!defined($data{$what}->{line_no}));
0200                 }
0201 
0202                 if (!$what) {
0203                     parse_error($file, $ln, "'What:' should come first:", $_);
0204                     next;
0205                 }
0206                 if ($new_tag eq "description") {
0207                     $sep =~ s,:, ,;
0208                     $content = ' ' x length($new_tag) . $sep . $content;
0209                     while ($content =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {}
0210                     if ($content =~ m/^(\s*)(\S.*)$/) {
0211                         # Preserve initial spaces for the first line
0212                         $space = $1;
0213                         $content = "$2\n";
0214                         $data{$what}->{$tag} .= $content;
0215                     } else {
0216                         undef($space);
0217                     }
0218 
0219                 } else {
0220                     $data{$what}->{$tag} = $content;
0221                 }
0222                 next;
0223             }
0224         }
0225 
0226         # Store any contents before tags at the database
0227         if (!$tag && $data{$nametag}->{what}) {
0228             $data{$nametag}->{description} .= $_;
0229             next;
0230         }
0231 
0232         if ($tag eq "description") {
0233             my $content = $_;
0234             while ($content =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {}
0235             if (m/^\s*\n/) {
0236                 $data{$what}->{$tag} .= "\n";
0237                 next;
0238             }
0239 
0240             if (!defined($space)) {
0241                 # Preserve initial spaces for the first line
0242                 if ($content =~ m/^(\s*)(\S.*)$/) {
0243                     $space = $1;
0244                     $content = "$2\n";
0245                 }
0246             } else {
0247                 $space = "" if (!($content =~ s/^($space)//));
0248             }
0249             $data{$what}->{$tag} .= $content;
0250 
0251             next;
0252         }
0253         if (m/^\s*(.*)/) {
0254             $data{$what}->{$tag} .= "\n$1";
0255             $data{$what}->{$tag} =~ s/\n+$//;
0256             next;
0257         }
0258 
0259         # Everything else is error
0260         parse_error($file, $ln, "Unexpected content", $_);
0261     }
0262     $data{$nametag}->{description} =~ s/^\n+// if ($data{$nametag}->{description});
0263     if ($what) {
0264         parse_error($file, $ln, "What '$what' doesn't have a description", "") if (!$data{$what}->{description});
0265 
0266         foreach my $w(split /\xac/,$what) {
0267             $symbols{$w}->{xref} = $what;
0268         };
0269     }
0270     close IN;
0271 }
0272 
0273 sub create_labels {
0274     my %labels;
0275 
0276     foreach my $what (keys %data) {
0277         next if ($data{$what}->{file} eq "File");
0278 
0279         foreach my $p (@{$data{$what}->{label_list}}) {
0280             my ($content, $label) = @{$p};
0281             $label = "abi_" . $label . " ";
0282             $label =~ tr/A-Z/a-z/;
0283 
0284             # Convert special chars to "_"
0285             $label =~s/([\x00-\x2f\x3a-\x40\x5b-\x60\x7b-\xff])/_/g;
0286             $label =~ s,_+,_,g;
0287             $label =~ s,_$,,;
0288 
0289             # Avoid duplicated labels
0290             while (defined($labels{$label})) {
0291                 my @chars = ("A".."Z", "a".."z");
0292                 $label .= $chars[rand @chars];
0293             }
0294             $labels{$label} = 1;
0295 
0296             $data{$what}->{label} = $label;
0297 
0298             # only one label is enough
0299             last;
0300         }
0301     }
0302 }
0303 
0304 #
0305 # Outputs the book on ReST format
0306 #
0307 
0308 # \b doesn't work well with paths. So, we need to define something else:
0309 # Boundaries are punct characters, spaces and end-of-line
0310 my $start = qr {(^|\s|\() }x;
0311 my $bondary = qr { ([,.:;\)\s]|\z) }x;
0312 my $xref_match = qr { $start(\/(sys|config|proc|dev|kvd)\/[^,.:;\)\s]+)$bondary }x;
0313 my $symbols = qr { ([\x01-\x08\x0e-\x1f\x21-\x2f\x3a-\x40\x7b-\xff]) }x;
0314 
0315 sub output_rest {
0316     create_labels();
0317 
0318     my $part = "";
0319 
0320     foreach my $what (sort {
0321                 ($data{$a}->{type} eq "File") cmp ($data{$b}->{type} eq "File") ||
0322                 $a cmp $b
0323                    } keys %data) {
0324         my $type = $data{$what}->{type};
0325 
0326         my @file = split / /, $data{$what}->{file};
0327         my @filepath = split / /, $data{$what}->{filepath};
0328 
0329         if ($enable_lineno) {
0330             printf ".. LINENO %s%s#%s\n\n",
0331                    $prefix, $file[0],
0332                    $data{$what}->{line_no};
0333         }
0334 
0335         my $w = $what;
0336 
0337         if ($type ne "File") {
0338             my $cur_part = $what;
0339             if ($what =~ '/') {
0340                 if ($what =~ m#^(\/?(?:[\w\-]+\/?){1,2})#) {
0341                     $cur_part = "Symbols under $1";
0342                     $cur_part =~ s,/$,,;
0343                 }
0344             }
0345 
0346             if ($cur_part ne "" && $part ne $cur_part) {
0347                 $part = $cur_part;
0348                 my $bar = $part;
0349                 $bar =~ s/./-/g;
0350                 print "$part\n$bar\n\n";
0351             }
0352 
0353             printf ".. _%s:\n\n", $data{$what}->{label};
0354 
0355             my @names = split /\xac/,$w;
0356             my $len = 0;
0357 
0358             foreach my $name (@names) {
0359                 $name =~ s/$symbols/\\$1/g;
0360                 $name = "**$name**";
0361                 $len = length($name) if (length($name) > $len);
0362             }
0363 
0364             print "+-" . "-" x $len . "-+\n";
0365             foreach my $name (@names) {
0366                 printf "| %s", $name . " " x ($len - length($name)) . " |\n";
0367                 print "+-" . "-" x $len . "-+\n";
0368             }
0369 
0370             print "\n";
0371         }
0372 
0373         for (my $i = 0; $i < scalar(@filepath); $i++) {
0374             my $path = $filepath[$i];
0375             my $f = $file[$i];
0376 
0377             $path =~ s,.*/(.*/.*),$1,;;
0378             $path =~ s,[/\-],_,g;;
0379             my $fileref = "abi_file_".$path;
0380 
0381             if ($type eq "File") {
0382                 print ".. _$fileref:\n\n";
0383             } else {
0384                 print "Defined on file :ref:`$f <$fileref>`\n\n";
0385             }
0386         }
0387 
0388         if ($type eq "File") {
0389             my $bar = $w;
0390             $bar =~ s/./-/g;
0391             print "$w\n$bar\n\n";
0392         }
0393 
0394         my $desc = "";
0395         $desc = $data{$what}->{description} if (defined($data{$what}->{description}));
0396         $desc =~ s/\s+$/\n/;
0397 
0398         if (!($desc =~ /^\s*$/)) {
0399             if ($description_is_rst) {
0400                 # Remove title markups from the description
0401                 # Having titles inside ABI files will only work if extra
0402                 # care would be taken in order to strictly follow the same
0403                 # level order for each markup.
0404                 $desc =~ s/\n[\-\*\=\^\~]+\n/\n\n/g;
0405 
0406                 # Enrich text by creating cross-references
0407 
0408                 my $new_desc = "";
0409                 my $init_indent = -1;
0410                 my $literal_indent = -1;
0411 
0412                 open(my $fh, "+<", \$desc);
0413                 while (my $d = <$fh>) {
0414                     my $indent = $d =~ m/^(\s+)/;
0415                     my $spaces = length($indent);
0416                     $init_indent = $indent if ($init_indent < 0);
0417                     if ($literal_indent >= 0) {
0418                         if ($spaces > $literal_indent) {
0419                             $new_desc .= $d;
0420                             next;
0421                         } else {
0422                             $literal_indent = -1;
0423                         }
0424                     } else {
0425                         if ($d =~ /()::$/ && !($d =~ /^\s*\.\./)) {
0426                             $literal_indent = $spaces;
0427                         }
0428                     }
0429 
0430                     $d =~ s,Documentation/(?!devicetree)(\S+)\.rst,:doc:`/$1`,g;
0431 
0432                     my @matches = $d =~ m,Documentation/ABI/([\w\/\-]+),g;
0433                     foreach my $f (@matches) {
0434                         my $xref = $f;
0435                         my $path = $f;
0436                         $path =~ s,.*/(.*/.*),$1,;;
0437                         $path =~ s,[/\-],_,g;;
0438                         $xref .= " <abi_file_" . $path . ">";
0439                         $d =~ s,\bDocumentation/ABI/$f\b,:ref:`$xref`,g;
0440                     }
0441 
0442                     # Seek for cross reference symbols like /sys/...
0443                     @matches = $d =~ m/$xref_match/g;
0444 
0445                     foreach my $s (@matches) {
0446                         next if (!($s =~ m,/,));
0447                         if (defined($data{$s}) && defined($data{$s}->{label})) {
0448                             my $xref = $s;
0449 
0450                             $xref =~ s/$symbols/\\$1/g;
0451                             $xref = ":ref:`$xref <" . $data{$s}->{label} . ">`";
0452 
0453                             $d =~ s,$start$s$bondary,$1$xref$2,g;
0454                         }
0455                     }
0456                     $new_desc .= $d;
0457                 }
0458                 close $fh;
0459 
0460 
0461                 print "$new_desc\n\n";
0462             } else {
0463                 $desc =~ s/^\s+//;
0464 
0465                 # Remove title markups from the description, as they won't work
0466                 $desc =~ s/\n[\-\*\=\^\~]+\n/\n\n/g;
0467 
0468                 if ($desc =~ m/\:\n/ || $desc =~ m/\n[\t ]+/  || $desc =~ m/[\x00-\x08\x0b-\x1f\x7b-\xff]/) {
0469                     # put everything inside a code block
0470                     $desc =~ s/\n/\n /g;
0471 
0472                     print "::\n\n";
0473                     print " $desc\n\n";
0474                 } else {
0475                     # Escape any special chars from description
0476                     $desc =~s/([\x00-\x08\x0b-\x1f\x21-\x2a\x2d\x2f\x3c-\x40\x5c\x5e-\x60\x7b-\xff])/\\$1/g;
0477                     print "$desc\n\n";
0478                 }
0479             }
0480         } else {
0481             print "DESCRIPTION MISSING for $what\n\n" if (!$data{$what}->{is_file});
0482         }
0483 
0484         if ($data{$what}->{symbols}) {
0485             printf "Has the following ABI:\n\n";
0486 
0487             foreach my $content(@{$data{$what}->{symbols}}) {
0488                 my $label = $data{$symbols{$content}->{xref}}->{label};
0489 
0490                 # Escape special chars from content
0491                 $content =~s/([\x00-\x1f\x21-\x2f\x3a-\x40\x7b-\xff])/\\$1/g;
0492 
0493                 print "- :ref:`$content <$label>`\n\n";
0494             }
0495         }
0496 
0497         if (defined($data{$what}->{users})) {
0498             my $users = $data{$what}->{users};
0499 
0500             $users =~ s/\n/\n\t/g;
0501             printf "Users:\n\t%s\n\n", $users if ($users ne "");
0502         }
0503 
0504     }
0505 }
0506 
0507 #
0508 # Searches for ABI symbols
0509 #
0510 sub search_symbols {
0511     foreach my $what (sort keys %data) {
0512         next if (!($what =~ m/($arg)/));
0513 
0514         my $type = $data{$what}->{type};
0515         next if ($type eq "File");
0516 
0517         my $file = $data{$what}->{filepath};
0518 
0519         $what =~ s/\xac/, /g;
0520         my $bar = $what;
0521         $bar =~ s/./-/g;
0522 
0523         print "\n$what\n$bar\n\n";
0524 
0525         my $kernelversion = $data{$what}->{kernelversion} if (defined($data{$what}->{kernelversion}));
0526         my $contact = $data{$what}->{contact} if (defined($data{$what}->{contact}));
0527         my $users = $data{$what}->{users} if (defined($data{$what}->{users}));
0528         my $date = $data{$what}->{date} if (defined($data{$what}->{date}));
0529         my $desc = $data{$what}->{description} if (defined($data{$what}->{description}));
0530 
0531         $kernelversion =~ s/^\s+// if ($kernelversion);
0532         $contact =~ s/^\s+// if ($contact);
0533         if ($users) {
0534             $users =~ s/^\s+//;
0535             $users =~ s/\n//g;
0536         }
0537         $date =~ s/^\s+// if ($date);
0538         $desc =~ s/^\s+// if ($desc);
0539 
0540         printf "Kernel version:\t\t%s\n", $kernelversion if ($kernelversion);
0541         printf "Date:\t\t\t%s\n", $date if ($date);
0542         printf "Contact:\t\t%s\n", $contact if ($contact);
0543         printf "Users:\t\t\t%s\n", $users if ($users);
0544         print "Defined on file(s):\t$file\n\n";
0545         print "Description:\n\n$desc";
0546     }
0547 }
0548 
0549 # Exclude /sys/kernel/debug and /sys/kernel/tracing from the search path
0550 sub dont_parse_special_attributes {
0551     if (($File::Find::dir =~ m,^/sys/kernel,)) {
0552         return grep {!/(debug|tracing)/ } @_;
0553     }
0554 
0555     if (($File::Find::dir =~ m,^/sys/fs,)) {
0556         return grep {!/(pstore|bpf|fuse)/ } @_;
0557     }
0558 
0559     return @_
0560 }
0561 
0562 my %leaf;
0563 my %aliases;
0564 my @files;
0565 my %root;
0566 
0567 sub graph_add_file {
0568     my $file = shift;
0569     my $type = shift;
0570 
0571     my $dir = $file;
0572     $dir =~ s,^(.*/).*,$1,;
0573     $file =~ s,.*/,,;
0574 
0575     my $name;
0576     my $file_ref = \%root;
0577     foreach my $edge(split "/", $dir) {
0578         $name .= "$edge/";
0579         if (!defined ${$file_ref}{$edge}) {
0580             ${$file_ref}{$edge} = { };
0581         }
0582         $file_ref = \%{$$file_ref{$edge}};
0583         ${$file_ref}{"__name"} = [ $name ];
0584     }
0585     $name .= "$file";
0586     ${$file_ref}{$file} = {
0587         "__name" => [ $name ]
0588     };
0589 
0590     return \%{$$file_ref{$file}};
0591 }
0592 
0593 sub graph_add_link {
0594     my $file = shift;
0595     my $link = shift;
0596 
0597     # Traverse graph to find the reference
0598     my $file_ref = \%root;
0599     foreach my $edge(split "/", $file) {
0600         $file_ref = \%{$$file_ref{$edge}} || die "Missing node!";
0601     }
0602 
0603     # do a BFS
0604 
0605     my @queue;
0606     my %seen;
0607     my $st;
0608 
0609     push @queue, $file_ref;
0610     $seen{$start}++;
0611 
0612     while (@queue) {
0613         my $v = shift @queue;
0614         my @child = keys(%{$v});
0615 
0616         foreach my $c(@child) {
0617             next if $seen{$$v{$c}};
0618             next if ($c eq "__name");
0619 
0620             if (!defined($$v{$c}{"__name"})) {
0621                 printf STDERR "Error: Couldn't find a non-empty name on a children of $file/.*: ";
0622                 print STDERR Dumper(%{$v});
0623                 exit;
0624             }
0625 
0626             # Add new name
0627             my $name = @{$$v{$c}{"__name"}}[0];
0628             if ($name =~ s#^$file/#$link/#) {
0629                 push @{$$v{$c}{"__name"}}, $name;
0630             }
0631             # Add child to the queue and mark as seen
0632             push @queue, $$v{$c};
0633             $seen{$c}++;
0634         }
0635     }
0636 }
0637 
0638 my $escape_symbols = qr { ([\x01-\x08\x0e-\x1f\x21-\x29\x2b-\x2d\x3a-\x40\x7b-\xfe]) }x;
0639 sub parse_existing_sysfs {
0640     my $file = $File::Find::name;
0641 
0642     my $mode = (lstat($file))[2];
0643     my $abs_file = abs_path($file);
0644 
0645     my @tmp;
0646     push @tmp, $file;
0647     push @tmp, $abs_file if ($abs_file ne $file);
0648 
0649     foreach my $f(@tmp) {
0650         # Ignore cgroup, as this is big and has zero docs under ABI
0651         return if ($f =~ m#^/sys/fs/cgroup/#);
0652 
0653         # Ignore firmware as it is documented elsewhere
0654         # Either ACPI or under Documentation/devicetree/bindings/
0655         return if ($f =~ m#^/sys/firmware/#);
0656 
0657         # Ignore some sysfs nodes that aren't actually part of ABI
0658         return if ($f =~ m#/sections|notes/#);
0659 
0660         # Would need to check at
0661         # Documentation/admin-guide/kernel-parameters.txt, but this
0662         # is not easily parseable.
0663         return if ($f =~ m#/parameters/#);
0664     }
0665 
0666     if (S_ISLNK($mode)) {
0667         $aliases{$file} = $abs_file;
0668         return;
0669     }
0670 
0671     return if (S_ISDIR($mode));
0672 
0673     # Trivial: file is defined exactly the same way at ABI What:
0674     return if (defined($data{$file}));
0675     return if (defined($data{$abs_file}));
0676 
0677     push @files, graph_add_file($abs_file, "file");
0678 }
0679 
0680 sub get_leave($)
0681 {
0682     my $what = shift;
0683     my $leave;
0684 
0685     my $l = $what;
0686     my $stop = 1;
0687 
0688     $leave = $l;
0689     $leave =~ s,/$,,;
0690     $leave =~ s,.*/,,;
0691     $leave =~ s/[\(\)]//g;
0692 
0693     # $leave is used to improve search performance at
0694     # check_undefined_symbols, as the algorithm there can seek
0695     # for a small number of "what". It also allows giving a
0696     # hint about a leave with the same name somewhere else.
0697     # However, there are a few occurences where the leave is
0698     # either a wildcard or a number. Just group such cases
0699     # altogether.
0700     if ($leave =~ m/\.\*/ || $leave eq "" || $leave =~ /\\d/) {
0701         $leave = "others";
0702     }
0703 
0704     return $leave;
0705 }
0706 
0707 my @not_found;
0708 
0709 sub check_file($$)
0710 {
0711     my $file_ref = shift;
0712     my $names_ref = shift;
0713     my @names = @{$names_ref};
0714     my $file = $names[0];
0715 
0716     my $found_string;
0717 
0718     my $leave = get_leave($file);
0719     if (!defined($leaf{$leave})) {
0720         $leave = "others";
0721     }
0722     my @expr = @{$leaf{$leave}->{expr}};
0723     die ("\rmissing rules for $leave") if (!defined($leaf{$leave}));
0724 
0725     my $path = $file;
0726     $path =~ s,(.*/).*,$1,;
0727 
0728     if ($search_string) {
0729         return if (!($file =~ m#$search_string#));
0730         $found_string = 1;
0731     }
0732 
0733     for (my $i = 0; $i < @names; $i++) {
0734         if ($found_string && $hint) {
0735             if (!$i) {
0736                 print STDERR "--> $names[$i]\n";
0737             } else {
0738                 print STDERR "    $names[$i]\n";
0739             }
0740         }
0741         foreach my $re (@expr) {
0742             print STDERR "$names[$i] =~ /^$re\$/\n" if ($debug && $dbg_undefined);
0743             if ($names[$i] =~ $re) {
0744                 return;
0745             }
0746         }
0747     }
0748 
0749     if ($leave ne "others") {
0750         my @expr = @{$leaf{"others"}->{expr}};
0751         for (my $i = 0; $i < @names; $i++) {
0752             foreach my $re (@expr) {
0753                 print STDERR "$names[$i] =~ /^$re\$/\n" if ($debug && $dbg_undefined);
0754                 if ($names[$i] =~ $re) {
0755                     return;
0756                 }
0757             }
0758         }
0759     }
0760 
0761     push @not_found, $file if (!$search_string || $found_string);
0762 
0763     if ($hint && (!$search_string || $found_string)) {
0764         my $what = $leaf{$leave}->{what};
0765         $what =~ s/\xac/\n\t/g;
0766         if ($leave ne "others") {
0767             print STDERR "\r    more likely regexes:\n\t$what\n";
0768         } else {
0769             print STDERR "\r    tested regexes:\n\t$what\n";
0770         }
0771     }
0772 }
0773 
0774 sub check_undefined_symbols {
0775     my $num_files = scalar @files;
0776     my $next_i = 0;
0777     my $start_time = times;
0778 
0779     @files = sort @files;
0780 
0781     my $last_time = $start_time;
0782 
0783     # When either debug or hint is enabled, there's no sense showing
0784     # progress, as the progress will be overriden.
0785     if ($hint || ($debug && $dbg_undefined)) {
0786         $next_i = $num_files;
0787     }
0788 
0789     my $is_console;
0790     $is_console = 1 if (-t STDERR);
0791 
0792     for (my $i = 0; $i < $num_files; $i++) {
0793         my $file_ref = $files[$i];
0794         my @names = @{$$file_ref{"__name"}};
0795 
0796         check_file($file_ref, \@names);
0797 
0798         my $cur_time = times;
0799 
0800         if ($i == $next_i || $cur_time > $last_time + 1) {
0801             my $percent = $i * 100 / $num_files;
0802 
0803             my $tm = $cur_time - $start_time;
0804             my $time = sprintf "%d:%02d", int($tm), 60 * ($tm - int($tm));
0805 
0806             printf STDERR "\33[2K\r", if ($is_console);
0807             printf STDERR "%s: processing sysfs files... %i%%: $names[0]", $time, $percent;
0808             printf STDERR "\n", if (!$is_console);
0809             STDERR->flush();
0810 
0811             $next_i = int (($percent + 1) * $num_files / 100);
0812             $last_time = $cur_time;
0813         }
0814     }
0815 
0816     my $cur_time = times;
0817     my $tm = $cur_time - $start_time;
0818     my $time = sprintf "%d:%02d", int($tm), 60 * ($tm - int($tm));
0819 
0820     printf STDERR "\33[2K\r", if ($is_console);
0821     printf STDERR "%s: processing sysfs files... done\n", $time;
0822 
0823     foreach my $file (@not_found) {
0824         print "$file not found.\n";
0825     }
0826 }
0827 
0828 sub undefined_symbols {
0829     print STDERR "Reading $sysfs_prefix directory contents...";
0830     find({
0831         wanted =>\&parse_existing_sysfs,
0832         preprocess =>\&dont_parse_special_attributes,
0833         no_chdir => 1
0834          }, $sysfs_prefix);
0835     print STDERR "done.\n";
0836 
0837     $leaf{"others"}->{what} = "";
0838 
0839     print STDERR "Converting ABI What fields into regexes...";
0840     foreach my $w (sort keys %data) {
0841         foreach my $what (split /\xac/,$w) {
0842             next if (!($what =~ m/^$sysfs_prefix/));
0843 
0844             # Convert what into regular expressions
0845 
0846             # Escape dot characters
0847             $what =~ s/\./\xf6/g;
0848 
0849             # Temporarily change [0-9]+ type of patterns
0850             $what =~ s/\[0\-9\]\+/\xff/g;
0851 
0852             # Temporarily change [\d+-\d+] type of patterns
0853             $what =~ s/\[0\-\d+\]/\xff/g;
0854             $what =~ s/\[(\d+)\]/\xf4$1\xf5/g;
0855 
0856             # Temporarily change [0-9] type of patterns
0857             $what =~ s/\[(\d)\-(\d)\]/\xf4$1-$2\xf5/g;
0858 
0859             # Handle multiple option patterns
0860             $what =~ s/[\{\<\[]([\w_]+)(?:[,|]+([\w_]+)){1,}[\}\>\]]/($1|$2)/g;
0861 
0862             # Handle wildcards
0863             $what =~ s,\*,.*,g;
0864             $what =~ s,/\xf6..,/.*,g;
0865             $what =~ s/\<[^\>]+\>/.*/g;
0866             $what =~ s/\{[^\}]+\}/.*/g;
0867             $what =~ s/\[[^\]]+\]/.*/g;
0868 
0869             $what =~ s/[XYZ]/.*/g;
0870 
0871             # Recover [0-9] type of patterns
0872             $what =~ s/\xf4/[/g;
0873             $what =~ s/\xf5/]/g;
0874 
0875             # Remove duplicated spaces
0876             $what =~ s/\s+/ /g;
0877 
0878             # Special case: this ABI has a parenthesis on it
0879             $what =~ s/sqrt\(x^2\+y^2\+z^2\)/sqrt\(x^2\+y^2\+z^2\)/;
0880 
0881             # Special case: drop comparition as in:
0882             #   What: foo = <something>
0883             # (this happens on a few IIO definitions)
0884             $what =~ s,\s*\=.*$,,;
0885 
0886             # Escape all other symbols
0887             $what =~ s/$escape_symbols/\\$1/g;
0888             $what =~ s/\\\\/\\/g;
0889             $what =~ s/\\([\[\]\(\)\|])/$1/g;
0890             $what =~ s/(\d+)\\(-\d+)/$1$2/g;
0891 
0892             $what =~ s/\xff/\\d+/g;
0893 
0894             # Special case: IIO ABI which a parenthesis.
0895             $what =~ s/sqrt(.*)/sqrt\(.*\)/;
0896 
0897             # Simplify regexes with multiple .*
0898             $what =~ s#(?:\.\*){2,}##g;
0899 #           $what =~ s#\.\*/\.\*#.*#g;
0900 
0901             # Recover dot characters
0902             $what =~ s/\xf6/\./g;
0903 
0904             my $leave = get_leave($what);
0905 
0906             my $added = 0;
0907             foreach my $l (split /\|/, $leave) {
0908                 if (defined($leaf{$l})) {
0909                     next if ($leaf{$l}->{what} =~ m/\b$what\b/);
0910                     $leaf{$l}->{what} .= "\xac" . $what;
0911                     $added = 1;
0912                 } else {
0913                     $leaf{$l}->{what} = $what;
0914                     $added = 1;
0915                 }
0916             }
0917             if ($search_string && $added) {
0918                 print STDERR "What: $what\n" if ($what =~ m#$search_string#);
0919             }
0920 
0921         }
0922     }
0923     # Compile regexes
0924     foreach my $l (sort keys %leaf) {
0925         my @expr;
0926         foreach my $w(sort split /\xac/, $leaf{$l}->{what}) {
0927             push @expr, qr /^$w$/;
0928         }
0929         $leaf{$l}->{expr} = \@expr;
0930     }
0931 
0932     # Take links into account
0933     foreach my $link (sort keys %aliases) {
0934         my $abs_file = $aliases{$link};
0935         graph_add_link($abs_file, $link);
0936     }
0937     print STDERR "done.\n";
0938 
0939     check_undefined_symbols;
0940 }
0941 
0942 # Ensure that the prefix will always end with a slash
0943 # While this is not needed for find, it makes the patch nicer
0944 # with --enable-lineno
0945 $prefix =~ s,/?$,/,;
0946 
0947 if ($cmd eq "undefined" || $cmd eq "search") {
0948     $show_warnings = 0;
0949 }
0950 #
0951 # Parses all ABI files located at $prefix dir
0952 #
0953 find({wanted =>\&parse_abi, no_chdir => 1}, $prefix);
0954 
0955 print STDERR Data::Dumper->Dump([\%data], [qw(*data)]) if ($debug & $dbg_dump_abi_structs);
0956 
0957 #
0958 # Handles the command
0959 #
0960 if ($cmd eq "undefined") {
0961     undefined_symbols;
0962 } elsif ($cmd eq "search") {
0963     search_symbols;
0964 } else {
0965     if ($cmd eq "rest") {
0966         output_rest;
0967     }
0968 
0969     # Warn about duplicated ABI entries
0970     foreach my $what(sort keys %symbols) {
0971         my @files = @{$symbols{$what}->{file}};
0972 
0973         next if (scalar(@files) == 1);
0974 
0975         printf STDERR "Warning: $what is defined %d times: @files\n",
0976             scalar(@files);
0977     }
0978 }
0979 
0980 __END__
0981 
0982 =head1 NAME
0983 
0984 get_abi.pl - parse the Linux ABI files and produce a ReST book.
0985 
0986 =head1 SYNOPSIS
0987 
0988 B<get_abi.pl> [--debug <level>] [--enable-lineno] [--man] [--help]
0989            [--(no-)rst-source] [--dir=<dir>] [--show-hints]
0990            [--search-string <regex>]
0991            <COMMAND> [<ARGUMENT>]
0992 
0993 Where B<COMMAND> can be:
0994 
0995 =over 8
0996 
0997 B<search> I<SEARCH_REGEX> - search for I<SEARCH_REGEX> inside ABI
0998 
0999 B<rest>                   - output the ABI in ReST markup language
1000 
1001 B<validate>               - validate the ABI contents
1002 
1003 B<undefined>              - existing symbols at the system that aren't
1004                             defined at Documentation/ABI
1005 
1006 =back
1007 
1008 =head1 OPTIONS
1009 
1010 =over 8
1011 
1012 =item B<--dir>
1013 
1014 Changes the location of the ABI search. By default, it uses
1015 the Documentation/ABI directory.
1016 
1017 =item B<--rst-source> and B<--no-rst-source>
1018 
1019 The input file may be using ReST syntax or not. Those two options allow
1020 selecting between a rst-compliant source ABI (B<--rst-source>), or a
1021 plain text that may be violating ReST spec, so it requres some escaping
1022 logic (B<--no-rst-source>).
1023 
1024 =item B<--enable-lineno>
1025 
1026 Enable output of .. LINENO lines.
1027 
1028 =item B<--debug> I<debug level>
1029 
1030 Print debug information according with the level, which is given by the
1031 following bitmask:
1032 
1033     -  1: Debug parsing What entries from ABI files;
1034     -  2: Shows what files are opened from ABI files;
1035     -  4: Dump the structs used to store the contents of the ABI files.
1036 
1037 =item B<--show-hints>
1038 
1039 Show hints about possible definitions for the missing ABI symbols.
1040 Used only when B<undefined>.
1041 
1042 =item B<--search-string> I<regex string>
1043 
1044 Show only occurences that match a search string.
1045 Used only when B<undefined>.
1046 
1047 =item B<--help>
1048 
1049 Prints a brief help message and exits.
1050 
1051 =item B<--man>
1052 
1053 Prints the manual page and exits.
1054 
1055 =back
1056 
1057 =head1 DESCRIPTION
1058 
1059 Parse the Linux ABI files from ABI DIR (usually located at Documentation/ABI),
1060 allowing to search for ABI symbols or to produce a ReST book containing
1061 the Linux ABI documentation.
1062 
1063 =head1 EXAMPLES
1064 
1065 Search for all stable symbols with the word "usb":
1066 
1067 =over 8
1068 
1069 $ scripts/get_abi.pl search usb --dir Documentation/ABI/stable
1070 
1071 =back
1072 
1073 Search for all symbols that match the regex expression "usb.*cap":
1074 
1075 =over 8
1076 
1077 $ scripts/get_abi.pl search usb.*cap
1078 
1079 =back
1080 
1081 Output all obsoleted symbols in ReST format
1082 
1083 =over 8
1084 
1085 $ scripts/get_abi.pl rest --dir Documentation/ABI/obsolete
1086 
1087 =back
1088 
1089 =head1 BUGS
1090 
1091 Report bugs to Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
1092 
1093 =head1 COPYRIGHT
1094 
1095 Copyright (c) 2016-2021 by Mauro Carvalho Chehab <mchehab+huawei@kernel.org>.
1096 
1097 License GPLv2: GNU GPL version 2 <http://gnu.org/licenses/gpl.html>.
1098 
1099 This is free software: you are free to change and redistribute it.
1100 There is NO WARRANTY, to the extent permitted by law.
1101 
1102 =cut
[ Source navigation ] [ Diff markup ] [ Identifier search ] [ general search ]

This page was automatically generated by the 2.1.0 LXR engine.

The LXR team
Valid CSS 2.1! Valid HTML 4.01!