checkpatch.pl 60.3 KB
Newer Older
1
2
#!/usr/bin/perl -w
# (c) 2001, Dave Jones. <davej@codemonkey.org.uk> (the file handling bit)
3
# (c) 2005, Joel Schopp <jschopp@austin.ibm.com> (the ugly bit)
4
5
6
7
8
9
# (c) 2007, Andy Whitcroft <apw@uk.ibm.com> (new conditions, test suite, etc)
# Licensed under the terms of the GNU GPL License version 2

use strict;

my $P = $0;
10
$P =~ s@.*/@@g;
11

Andy Whitcroft's avatar
Andy Whitcroft committed
12
my $V = '0.20';
13
14
15
16
17
18
19

use Getopt::Long qw(:config no_auto_abbrev);

my $quiet = 0;
my $tree = 1;
my $chk_signoff = 1;
my $chk_patch = 1;
20
my $tst_type = 0;
21
my $tst_only;
22
my $emacs = 0;
23
my $terse = 0;
24
25
my $file = 0;
my $check = 0;
26
27
my $summary = 1;
my $mailback = 0;
28
my $summary_file = 0;
29
my $root;
30
my %debug;
31
GetOptions(
32
	'q|quiet+'	=> \$quiet,
33
34
35
	'tree!'		=> \$tree,
	'signoff!'	=> \$chk_signoff,
	'patch!'	=> \$chk_patch,
36
	'emacs!'	=> \$emacs,
37
	'terse!'	=> \$terse,
38
39
40
41
	'file!'		=> \$file,
	'subjective!'	=> \$check,
	'strict!'	=> \$check,
	'root=s'	=> \$root,
42
43
	'summary!'	=> \$summary,
	'mailback!'	=> \$mailback,
44
45
	'summary-file!'	=> \$summary_file,

46
	'debug=s'	=> \%debug,
47
	'test-type!'	=> \$tst_type,
48
	'test-only=s'	=> \$tst_only,
49
50
51
52
53
) or exit;

my $exit = 0;

if ($#ARGV < 0) {
54
	print "usage: $P [options] patchfile\n";
55
	print "version: $V\n";
56
57
58
59
60
61
62
63
64
	print "options: -q               => quiet\n";
	print "         --no-tree        => run without a kernel tree\n";
	print "         --terse          => one line per report\n";
	print "         --emacs          => emacs compile window format\n";
	print "         --file           => check a source file\n";
	print "         --strict         => enable more subjective tests\n";
	print "         --root           => path to the kernel tree root\n";
	print "         --no-summary     => suppress the per-file summary\n";
	print "         --summary-file   => include the filename in summary\n";
65
66
67
	exit(1);
}

68
69
70
71
72
73
my $dbg_values = 0;
my $dbg_possible = 0;
for my $key (keys %debug) {
	eval "\${dbg_$key} = '$debug{$key}';"
}

74
75
76
77
78
if ($terse) {
	$emacs = 1;
	$quiet++;
}

79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
if ($tree) {
	if (defined $root) {
		if (!top_of_kernel_tree($root)) {
			die "$P: $root: --root does not point at a valid tree\n";
		}
	} else {
		if (top_of_kernel_tree('.')) {
			$root = '.';
		} elsif ($0 =~ m@(.*)/scripts/[^/]*$@ &&
						top_of_kernel_tree($1)) {
			$root = $1;
		}
	}

	if (!defined $root) {
		print "Must be run from the top-level dir. of a kernel tree\n";
		exit(2);
	}
97
98
}

99
100
101
102
103
104
105
106
107
108
109
my $emitted_corrupt = 0;

our $Ident       = qr{[A-Za-z_][A-Za-z\d_]*};
our $Storage	= qr{extern|static|asmlinkage};
our $Sparse	= qr{
			__user|
			__kernel|
			__force|
			__iomem|
			__must_check|
			__init_refok|
110
			__kprobes
111
112
113
114
115
116
117
		}x;
our $Attribute	= qr{
			const|
			__read_mostly|
			__kprobes|
			__(?:mem|cpu|dev|)(?:initdata|init)
		  }x;
118
our $Modifier;
119
120
121
122
123
124
125
126
127
our $Inline	= qr{inline|__always_inline|noinline};
our $Member	= qr{->$Ident|\.$Ident|\[[^]]*\]};
our $Lval	= qr{$Ident(?:$Member)*};

our $Constant	= qr{(?:[0-9]+|0x[0-9a-fA-F]+)[UL]*};
our $Assignment	= qr{(?:\*\=|/=|%=|\+=|-=|<<=|>>=|&=|\^=|\|=|=)};
our $Operators	= qr{
			<=|>=|==|!=|
			=>|->|<<|>>|<|>|!|~|
128
			&&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/|%
129
130
		  }x;

131
132
133
134
our $NonptrType;
our $Type;
our $Declare;

135
136
137
138
139
140
141
142
143
144
145
our $UTF8	= qr {
	[\x09\x0A\x0D\x20-\x7E]              # ASCII
	| [\xC2-\xDF][\x80-\xBF]             # non-overlong 2-byte
	|  \xE0[\xA0-\xBF][\x80-\xBF]        # excluding overlongs
	| [\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}  # straight 3-byte
	|  \xED[\x80-\x9F][\x80-\xBF]        # excluding surrogates
	|  \xF0[\x90-\xBF][\x80-\xBF]{2}     # planes 1-3
	| [\xF1-\xF3][\x80-\xBF]{3}          # planes 4-15
	|  \xF4[\x80-\x8F][\x80-\xBF]{2}     # plane 16
}x;

146
147
our @typeList = (
	qr{void},
148
149
150
151
152
153
154
	qr{(?:unsigned\s+)?char},
	qr{(?:unsigned\s+)?short},
	qr{(?:unsigned\s+)?int},
	qr{(?:unsigned\s+)?long},
	qr{(?:unsigned\s+)?long\s+int},
	qr{(?:unsigned\s+)?long\s+long},
	qr{(?:unsigned\s+)?long\s+long\s+int},
155
156
157
158
159
160
161
162
163
164
165
166
	qr{unsigned},
	qr{float},
	qr{double},
	qr{bool},
	qr{(?:__)?(?:u|s|be|le)(?:8|16|32|64)},
	qr{struct\s+$Ident},
	qr{union\s+$Ident},
	qr{enum\s+$Ident},
	qr{${Ident}_t},
	qr{${Ident}_handler},
	qr{${Ident}_handler_fn},
);
167
168
169
our @modifierList = (
	qr{fastcall},
);
170
171

sub build_types {
172
	my $mods = "(?:  \n" . join("|\n  ", @modifierList) . "\n)";
173
	my $all = "(?:  \n" . join("|\n  ", @typeList) . "\n)";
174
	$Modifier	= qr{(?:$Attribute|$Sparse|$mods)};
175
176
	$NonptrType	= qr{
			(?:const\s+)?
177
			(?:$mods\s+)?
178
			(?:
179
180
				(?:typeof|__typeof__)\s*\(\s*\**\s*$Ident\s*\)|
				(?:${all}\b)
181
			)
182
			(?:\s+$Modifier|\s+const)*
183
184
		  }x;
	$Type	= qr{
185
			$NonptrType
186
			(?:\s*\*+\s*const|\s*\*+|(?:\s*\[\s*\])+)?
187
			(?:\s+$Inline|\s+$Modifier)*
188
189
190
191
		  }x;
	$Declare	= qr{(?:$Storage\s+)?$Type};
}
build_types();
192
193
194

$chk_signoff = 0 if ($file);

195
196
my @dep_includes = ();
my @dep_functions = ();
197
198
199
200
my $removal = "Documentation/feature-removal-schedule.txt";
if ($tree && -f "$root/$removal") {
	open(REMOVE, "<$root/$removal") ||
				die "$P: $removal: open failed - $!\n";
201
	while (<REMOVE>) {
202
203
204
		if (/^Check:\s+(.*\S)/) {
			for my $entry (split(/[, ]+/, $1)) {
				if ($entry =~ m@include/(.*)@) {
205
206
					push(@dep_includes, $1);

207
208
209
				} elsif ($entry !~ m@/@) {
					push(@dep_functions, $entry);
				}
210
			}
211
212
213
214
		}
	}
}

215
my @rawlines = ();
216
217
my @lines = ();
my $vname;
218
219
220
221
222
223
224
for my $filename (@ARGV) {
	if ($file) {
		open(FILE, "diff -u /dev/null $filename|") ||
			die "$P: $filename: diff failed - $!\n";
	} else {
		open(FILE, "<$filename") ||
			die "$P: $filename: open failed - $!\n";
225
	}
226
227
228
229
230
	if ($filename eq '-') {
		$vname = 'Your patch';
	} else {
		$vname = $filename;
	}
231
232
233
234
235
	while (<FILE>) {
		chomp;
		push(@rawlines, $_);
	}
	close(FILE);
236
	if (!process($filename)) {
237
238
239
		$exit = 1;
	}
	@rawlines = ();
240
	@lines = ();
241
242
243
244
245
}

exit($exit);

sub top_of_kernel_tree {
246
247
248
249
250
251
252
253
254
255
256
257
	my ($root) = @_;

	my @tree_check = (
		"COPYING", "CREDITS", "Kbuild", "MAINTAINERS", "Makefile",
		"README", "Documentation", "arch", "include", "drivers",
		"fs", "init", "ipc", "kernel", "lib", "scripts",
	);

	foreach my $check (@tree_check) {
		if (! -e $root . '/' . $check) {
			return 0;
		}
258
	}
259
	return 1;
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
}

sub expand_tabs {
	my ($str) = @_;

	my $res = '';
	my $n = 0;
	for my $c (split(//, $str)) {
		if ($c eq "\t") {
			$res .= ' ';
			$n++;
			for (; ($n % 8) != 0; $n++) {
				$res .= ' ';
			}
			next;
		}
		$res .= $c;
		$n++;
	}

	return $res;
}
282
sub copy_spacing {
283
	(my $res = shift) =~ tr/\t/ /c;
284
285
	return $res;
}
286

287
288
289
290
291
292
293
294
295
296
297
298
299
sub line_stats {
	my ($line) = @_;

	# Drop the diff line leader and expand tabs
	$line =~ s/^.//;
	$line = expand_tabs($line);

	# Pick the indent from the front of the line.
	my ($white) = ($line =~ /^(\s*)/);

	return (length($line), length($white));
}

300
301
302
303
304
305
306
307
308
309
310
my $sanitise_quote = '';

sub sanitise_line_reset {
	my ($in_comment) = @_;

	if ($in_comment) {
		$sanitise_quote = '*/';
	} else {
		$sanitise_quote = '';
	}
}
311
312
313
314
315
316
sub sanitise_line {
	my ($line) = @_;

	my $res = '';
	my $l = '';

317
	my $qlen = 0;
318
319
	my $off = 0;
	my $c;
320

321
322
323
324
325
326
327
328
329
330
331
332
333
334
	# Always copy over the diff marker.
	$res = substr($line, 0, 1);

	for ($off = 1; $off < length($line); $off++) {
		$c = substr($line, $off, 1);

		# Comments we are wacking completly including the begin
		# and end, all to $;.
		if ($sanitise_quote eq '' && substr($line, $off, 2) eq '/*') {
			$sanitise_quote = '*/';

			substr($res, $off, 2, "$;$;");
			$off++;
			next;
335
		}
336
		if (substr($line, $off, 2) eq '*/') {
337
338
339
340
			$sanitise_quote = '';
			substr($res, $off, 2, "$;$;");
			$off++;
			next;
341
		}
342
343
344
345
346
347
348

		# A \ in a string means ignore the next character.
		if (($sanitise_quote eq "'" || $sanitise_quote eq '"') &&
		    $c eq "\\") {
			substr($res, $off, 2, 'XX');
			$off++;
			next;
349
		}
350
351
352
353
		# Regular quotes.
		if ($c eq "'" || $c eq '"') {
			if ($sanitise_quote eq '') {
				$sanitise_quote = $c;
354

355
356
357
358
359
360
				substr($res, $off, 1, $c);
				next;
			} elsif ($sanitise_quote eq $c) {
				$sanitise_quote = '';
			}
		}
361

362
363
364
365
366
367
368
369
		#print "SQ:$sanitise_quote\n";
		if ($off != 0 && $sanitise_quote eq '*/' && $c ne "\t") {
			substr($res, $off, 1, $;);
		} elsif ($off != 0 && $sanitise_quote && $c ne "\t") {
			substr($res, $off, 1, 'X');
		} else {
			substr($res, $off, 1, $c);
		}
370
371
372
	}

	# The pathname on a #include may be surrounded by '<' and '>'.
373
	if ($res =~ /^.\s*\#\s*include\s+\<(.*)\>/) {
374
375
376
377
		my $clean = 'X' x length($1);
		$res =~ s@\<.*\>@<$clean>@;

	# The whole of a #error is a string.
378
	} elsif ($res =~ /^.\s*\#\s*(?:error|warning)\s+(.*)\b/) {
379
		my $clean = 'X' x length($1);
380
		$res =~ s@(\#\s*(?:error|warning)\s+).*@$1$clean@;
381
382
	}

383
384
385
	return $res;
}

386
387
388
389
390
391
sub ctx_statement_block {
	my ($linenr, $remain, $off) = @_;
	my $line = $linenr - 1;
	my $blk = '';
	my $soff = $off;
	my $coff = $off - 1;
392
	my $coff_set = 0;
393

394
395
	my $loff = 0;

396
397
	my $type = '';
	my $level = 0;
398
	my $p;
399
400
	my $c;
	my $len = 0;
401
402

	my $remainder;
403
	while (1) {
404
		#warn "CSB: blk<$blk> remain<$remain>\n";
405
406
407
408
		# If we are about to drop off the end, pull in more
		# context.
		if ($off >= $len) {
			for (; $remain > 0; $line++) {
409
				next if ($lines[$line] =~ /^-/);
410
				$remain--;
411
				$loff = $len;
412
				$blk .= $lines[$line] . "\n";
413
414
415
416
417
418
				$len = length($blk);
				$line++;
				last;
			}
			# Bail if there is no further context.
			#warn "CSB: blk<$blk> off<$off> len<$len>\n";
419
			if ($off >= $len) {
420
421
422
				last;
			}
		}
423
		$p = $c;
424
		$c = substr($blk, $off, 1);
425
		$remainder = substr($blk, $off);
426

427
		#warn "CSB: c<$c> type<$type> level<$level> remainder<$remainder> coff_set<$coff_set>\n";
428
429
430
431
432
433
		# Statement ends at the ';' or a close '}' at the
		# outermost level.
		if ($level == 0 && $c eq ';') {
			last;
		}

434
		# An else is really a conditional as long as its not else if
435
436
437
438
439
440
441
442
		if ($level == 0 && $coff_set == 0 &&
				(!defined($p) || $p =~ /(?:\s|\}|\+)/) &&
				$remainder =~ /^(else)(?:\s|{)/ &&
				$remainder !~ /^else\s+if\b/) {
			$coff = $off + length($1) - 1;
			$coff_set = 1;
			#warn "CSB: mark coff<$coff> soff<$soff> 1<$1>\n";
			#warn "[" . substr($blk, $soff, $coff - $soff + 1) . "]\n";
443
444
		}

445
446
447
448
449
450
451
452
453
454
		if (($type eq '' || $type eq '(') && $c eq '(') {
			$level++;
			$type = '(';
		}
		if ($type eq '(' && $c eq ')') {
			$level--;
			$type = ($level != 0)? '(' : '';

			if ($level == 0 && $coff < $soff) {
				$coff = $off;
455
456
				$coff_set = 1;
				#warn "CSB: mark coff<$coff>\n";
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
			}
		}
		if (($type eq '' || $type eq '{') && $c eq '{') {
			$level++;
			$type = '{';
		}
		if ($type eq '{' && $c eq '}') {
			$level--;
			$type = ($level != 0)? '{' : '';

			if ($level == 0) {
				last;
			}
		}
		$off++;
	}
473
	# We are truly at the end, so shuffle to the next line.
474
	if ($off == $len) {
475
		$loff = $len + 1;
476
477
478
		$line++;
		$remain--;
	}
479
480
481
482
483
484
485

	my $statement = substr($blk, $soff, $off - $soff + 1);
	my $condition = substr($blk, $soff, $coff - $soff + 1);

	#warn "STATEMENT<$statement>\n";
	#warn "CONDITION<$condition>\n";

486
	#print "coff<$coff> soff<$off> loff<$loff>\n";
487
488
489
490
491

	return ($statement, $condition,
			$line, $remain + 1, $off - $loff + 1, $level);
}

492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
sub statement_lines {
	my ($stmt) = @_;

	# Strip the diff line prefixes and rip blank lines at start and end.
	$stmt =~ s/(^|\n)./$1/g;
	$stmt =~ s/^\s*//;
	$stmt =~ s/\s*$//;

	my @stmt_lines = ($stmt =~ /\n/g);

	return $#stmt_lines + 2;
}

sub statement_rawlines {
	my ($stmt) = @_;

	my @stmt_lines = ($stmt =~ /\n/g);

	return $#stmt_lines + 2;
}

sub statement_block_size {
	my ($stmt) = @_;

	$stmt =~ s/(^|\n)./$1/g;
	$stmt =~ s/^\s*{//;
	$stmt =~ s/}\s*$//;
	$stmt =~ s/^\s*//;
	$stmt =~ s/\s*$//;

	my @stmt_lines = ($stmt =~ /\n/g);
	my @stmt_statements = ($stmt =~ /;/g);

	my $stmt_lines = $#stmt_lines + 2;
	my $stmt_statements = $#stmt_statements + 1;

	if ($stmt_lines > $stmt_statements) {
		return $stmt_lines;
	} else {
		return $stmt_statements;
	}
}

535
536
537
538
539
540
sub ctx_statement_full {
	my ($linenr, $remain, $off) = @_;
	my ($statement, $condition, $level);

	my (@chunks);

541
	# Grab the first conditional/block pair.
542
543
	($statement, $condition, $linenr, $remain, $off, $level) =
				ctx_statement_block($linenr, $remain, $off);
544
	#print "F: c<$condition> s<$statement> remain<$remain>\n";
545
546
547
548
549
550
551
	push(@chunks, [ $condition, $statement ]);
	if (!($remain > 0 && $condition =~ /^\s*(?:\n[+-])?\s*(?:if|else|do)\b/s)) {
		return ($level, $linenr, @chunks);
	}

	# Pull in the following conditional/block pairs and see if they
	# could continue the statement.
552
553
554
	for (;;) {
		($statement, $condition, $linenr, $remain, $off, $level) =
				ctx_statement_block($linenr, $remain, $off);
555
		#print "C: c<$condition> s<$statement> remain<$remain>\n";
556
		last if (!($remain > 0 && $condition =~ /^(?:\s*\n[+-])*\s*(?:else|do)\b/s));
557
558
		#print "C: push\n";
		push(@chunks, [ $condition, $statement ]);
559
560
561
	}

	return ($level, $linenr, @chunks);
562
563
}

564
sub ctx_block_get {
565
	my ($linenr, $remain, $outer, $open, $close, $off) = @_;
566
567
568
569
570
571
572
	my $line;
	my $start = $linenr - 1;
	my $blk = '';
	my @o;
	my @c;
	my @res = ();

573
	my $level = 0;
574
575
576
577
578
	for ($line = $start; $remain > 0; $line++) {
		next if ($rawlines[$line] =~ /^-/);
		$remain--;

		$blk .= $rawlines[$line];
579
580
581
582
583
584
		foreach my $c (split(//, $rawlines[$line])) {
			##print "C<$c>L<$level><$open$close>O<$off>\n";
			if ($off > 0) {
				$off--;
				next;
			}
585

586
587
588
589
590
591
592
			if ($c eq $close && $level > 0) {
				$level--;
				last if ($level == 0);
			} elsif ($c eq $open) {
				$level++;
			}
		}
593

594
		if (!$outer || $level <= 1) {
595
			push(@res, $rawlines[$line]);
596
597
		}

598
		last if ($level == 0);
599
600
	}

601
	return ($level, @res);
602
603
604
605
}
sub ctx_block_outer {
	my ($linenr, $remain) = @_;

606
607
	my ($level, @r) = ctx_block_get($linenr, $remain, 1, '{', '}', 0);
	return @r;
608
609
610
611
}
sub ctx_block {
	my ($linenr, $remain) = @_;

612
613
	my ($level, @r) = ctx_block_get($linenr, $remain, 0, '{', '}', 0);
	return @r;
614
615
}
sub ctx_statement {
616
617
618
619
620
621
	my ($linenr, $remain, $off) = @_;

	my ($level, @r) = ctx_block_get($linenr, $remain, 0, '(', ')', $off);
	return @r;
}
sub ctx_block_level {
622
623
	my ($linenr, $remain) = @_;

624
	return ctx_block_get($linenr, $remain, 0, '{', '}', 0);
625
}
626
627
628
629
630
sub ctx_statement_level {
	my ($linenr, $remain, $off) = @_;

	return ctx_block_get($linenr, $remain, 0, '(', ')', $off);
}
631
632
633
634
635

sub ctx_locate_comment {
	my ($first_line, $end_line) = @_;

	# Catch a comment on the end of the line itself.
636
	my ($current_comment) = ($rawlines[$end_line - 1] =~ m@.*(/\*.*\*/)\s*(?:\\\s*)?$@);
637
638
639
640
641
642
643
	return $current_comment if (defined $current_comment);

	# Look through the context and try and figure out if there is a
	# comment.
	my $in_comment = 0;
	$current_comment = '';
	for (my $linenr = $first_line; $linenr < $end_line; $linenr++) {
644
645
		my $line = $rawlines[$linenr - 1];
		#warn "           $line\n";
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
		if ($linenr == $first_line and $line =~ m@^.\s*\*@) {
			$in_comment = 1;
		}
		if ($line =~ m@/\*@) {
			$in_comment = 1;
		}
		if (!$in_comment && $current_comment ne '') {
			$current_comment = '';
		}
		$current_comment .= $line . "\n" if ($in_comment);
		if ($line =~ m@\*/@) {
			$in_comment = 0;
		}
	}

	chomp($current_comment);
	return($current_comment);
}
sub ctx_has_comment {
	my ($first_line, $end_line) = @_;
	my $cmt = ctx_locate_comment($first_line, $end_line);

668
	##print "LINE: $rawlines[$end_line - 1 ]\n";
669
670
671
672
673
	##print "CMMT: $cmt\n";

	return ($cmt ne '');
}

674
675
676
sub cat_vet {
	my ($vet) = @_;
	my ($res, $coded);
677

678
679
680
681
682
683
	$res = '';
	while ($vet =~ /([^[:cntrl:]]*)([[:cntrl:]]|$)/g) {
		$res .= $1;
		if ($2 ne '') {
			$coded = sprintf("^%c", unpack('C', $2) + 64);
			$res .= $coded;
684
685
		}
	}
686
	$res =~ s/$/\$/;
687

688
	return $res;
689
690
}

691
my $av_preprocessor = 0;
692
my $av_pending;
693
694
695
696
my @av_paren_type;

sub annotate_reset {
	$av_preprocessor = 0;
697
698
	$av_pending = '_';
	@av_paren_type = ('E');
699
700
}

701
702
sub annotate_values {
	my ($stream, $type) = @_;
703

704
705
706
	my $res;
	my $cur = $stream;

707
	print "$stream\n" if ($dbg_values > 1);
708
709

	while (length($cur)) {
710
		@av_paren_type = ('E') if ($#av_paren_type < 0);
711
		print " <" . join('', @av_paren_type) .
712
				"> <$type> <$av_pending>" if ($dbg_values > 1);
713
		if ($cur =~ /^(\s+)/o) {
714
715
			print "WS($1)\n" if ($dbg_values > 1);
			if ($1 =~ /\n/ && $av_preprocessor) {
716
				$type = pop(@av_paren_type);
717
				$av_preprocessor = 0;
718
719
			}

720
		} elsif ($cur =~ /^($Type)\s*(?:$Ident|,|\))/) {
721
			print "DECLARE($1)\n" if ($dbg_values > 1);
722
723
			$type = 'T';

724
725
726
727
		} elsif ($cur =~ /^($Modifier)\s*/) {
			print "MODIFIER($1)\n" if ($dbg_values > 1);
			$type = 'T';

728
		} elsif ($cur =~ /^(\#\s*define\s*$Ident)(\(?)/o) {
729
			print "DEFINE($1,$2)\n" if ($dbg_values > 1);
730
			$av_preprocessor = 1;
731
732
733
734
735
736
			push(@av_paren_type, $type);
			if ($2 ne '') {
				$av_pending = 'N';
			}
			$type = 'E';

737
		} elsif ($cur =~ /^(\#\s*(?:undef\s*$Ident|include\b))/o) {
738
739
740
			print "UNDEF($1)\n" if ($dbg_values > 1);
			$av_preprocessor = 1;
			push(@av_paren_type, $type);
741

742
		} elsif ($cur =~ /^(\#\s*(?:ifdef|ifndef|if))/o) {
743
			print "PRE_START($1)\n" if ($dbg_values > 1);
744
			$av_preprocessor = 1;
745
746
747

			push(@av_paren_type, $type);
			push(@av_paren_type, $type);
748
			$type = 'E';
749

750
		} elsif ($cur =~ /^(\#\s*(?:else|elif))/o) {
751
752
753
754
755
			print "PRE_RESTART($1)\n" if ($dbg_values > 1);
			$av_preprocessor = 1;

			push(@av_paren_type, $av_paren_type[$#av_paren_type]);

756
			$type = 'E';
757

758
		} elsif ($cur =~ /^(\#\s*(?:endif))/o) {
759
760
761
762
763
764
765
766
			print "PRE_END($1)\n" if ($dbg_values > 1);

			$av_preprocessor = 1;

			# Assume all arms of the conditional end as this
			# one does, and continue as if the #endif was not here.
			pop(@av_paren_type);
			push(@av_paren_type, $type);
767
			$type = 'E';
768
769

		} elsif ($cur =~ /^(\\\n)/o) {
770
			print "PRECONT($1)\n" if ($dbg_values > 1);
771

772
773
774
775
776
		} elsif ($cur =~ /^(__attribute__)\s*\(?/o) {
			print "ATTR($1)\n" if ($dbg_values > 1);
			$av_pending = $type;
			$type = 'N';

777
		} elsif ($cur =~ /^(sizeof)\s*(\()?/o) {
778
			print "SIZEOF($1)\n" if ($dbg_values > 1);
779
			if (defined $2) {
780
				$av_pending = 'V';
781
782
783
			}
			$type = 'N';

784
		} elsif ($cur =~ /^(if|while|typeof|__typeof__|for)\b/o) {
785
			print "COND($1)\n" if ($dbg_values > 1);
786
			$av_pending = 'N';
787
788
			$type = 'N';

789
		} elsif ($cur =~/^(return|case|else|goto)/o) {
790
			print "KEYWORD($1)\n" if ($dbg_values > 1);
791
792
793
			$type = 'N';

		} elsif ($cur =~ /^(\()/o) {
794
			print "PAREN('$1')\n" if ($dbg_values > 1);
795
796
			push(@av_paren_type, $av_pending);
			$av_pending = '_';
797
798
799
			$type = 'N';

		} elsif ($cur =~ /^(\))/o) {
800
801
802
			my $new_type = pop(@av_paren_type);
			if ($new_type ne '_') {
				$type = $new_type;
803
804
				print "PAREN('$1') -> $type\n"
							if ($dbg_values > 1);
805
			} else {
806
				print "PAREN('$1')\n" if ($dbg_values > 1);
807
808
			}

809
		} elsif ($cur =~ /^($Ident)\s*\(/o) {
810
			print "FUNC($1)\n" if ($dbg_values > 1);
811
			$type = 'V';
812
			$av_pending = 'V';
813
814

		} elsif ($cur =~ /^($Ident|$Constant)/o) {
815
			print "IDENT($1)\n" if ($dbg_values > 1);
816
817
818
			$type = 'V';

		} elsif ($cur =~ /^($Assignment)/o) {
819
			print "ASSIGN($1)\n" if ($dbg_values > 1);
820
821
			$type = 'N';

822
		} elsif ($cur =~/^(;|{|})/) {
823
			print "END($1)\n" if ($dbg_values > 1);
824
825
			$type = 'E';

826
		} elsif ($cur =~ /^(;|\?|:|\[)/o) {
827
			print "CLOSE($1)\n" if ($dbg_values > 1);
828
829
830
			$type = 'N';

		} elsif ($cur =~ /^($Operators)/o) {
831
			print "OP($1)\n" if ($dbg_values > 1);
832
833
834
835
836
			if ($1 ne '++' && $1 ne '--') {
				$type = 'N';
			}

		} elsif ($cur =~ /(^.)/o) {
837
			print "C($1)\n" if ($dbg_values > 1);
838
839
840
841
842
		}
		if (defined $1) {
			$cur = substr($cur, length($1));
			$res .= $type x length($1);
		}
843
	}
844

845
	return $res;
846
847
}

848
sub possible {
849
	my ($possible, $line) = @_;
850

851
	print "CHECK<$possible> ($line)\n" if ($dbg_possible > 1);
852
853
854
	if ($possible !~ /^(?:$Storage|$Type|DEFINE_\S+)$/ &&
	    $possible ne 'goto' && $possible ne 'return' &&
	    $possible ne 'case' && $possible ne 'else' &&
855
	    $possible ne 'asm' && $possible ne '__asm__' &&
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
	    $possible !~ /^(typedef|struct|enum)\b/) {
		# Check for modifiers.
		$possible =~ s/\s*$Storage\s*//g;
		$possible =~ s/\s*$Sparse\s*//g;
		if ($possible =~ /^\s*$/) {

		} elsif ($possible =~ /\s/) {
			$possible =~ s/\s*$Type\s*//g;
			warn "MODIFIER: $possible ($line)\n" if ($dbg_possible);
			push(@modifierList, $possible);

		} else {
			warn "POSSIBLE: $possible ($line)\n" if ($dbg_possible);
			push(@typeList, $possible);
		}
871
872
873
874
		build_types();
	}
}

875
876
my $prefix = '';

877
sub report {
878
879
880
	if (defined $tst_only && $_[0] !~ /\Q$tst_only\E/) {
		return 0;
	}
881
882
883
884
	my $line = $prefix . $_[0];

	$line = (split('\n', $line))[0] . "\n" if ($terse);

885
	push(our @report, $line);
886
887

	return 1;
888
889
}
sub report_dump {
890
	our @report;
891
}
892
sub ERROR {
893
894
895
896
	if (report("ERROR: $_[0]\n")) {
		our $clean = 0;
		our $cnt_error++;
	}
897
898
}
sub WARN {
899
900
901
902
	if (report("WARNING: $_[0]\n")) {
		our $clean = 0;
		our $cnt_warn++;
	}
903
904
}
sub CHK {
905
	if ($check && report("CHECK: $_[0]\n")) {
906
907
908
		our $clean = 0;
		our $cnt_chk++;
	}
909
910
}

911
912
913
914
915
sub process {
	my $filename = shift;

	my $linenr=0;
	my $prevline="";
916
	my $prevrawline="";
917
	my $stashline="";
918
	my $stashrawline="";
919

920
	my $length;
921
922
923
924
	my $indent;
	my $previndent=0;
	my $stashindent=0;

925
	our $clean = 1;
926
927
928
	my $signoff = 0;
	my $is_patch = 0;

929
	our @report = ();
930
931
932
933
934
	our $cnt_lines = 0;
	our $cnt_error = 0;
	our $cnt_warn = 0;
	our $cnt_chk = 0;

935
936
937
938
939
940
	# Trace the real file/line as we go.
	my $realfile = '';
	my $realline = 0;
	my $realcnt = 0;
	my $here = '';
	my $in_comment = 0;
941
	my $comment_edge = 0;
942
943
	my $first_line = 0;

944
945
946
	my $prev_values = 'E';

	# suppression flags
947
	my %suppress_ifbraces;
948

949
	# Pre-scan the patch sanitizing the lines.
950
	# Pre-scan the patch looking for any __setup documentation.
951
	#
952
953
	my @setup_docs = ();
	my $setup_docs = 0;
954
955

	sanitise_line_reset();
956
957
	my $line;
	foreach my $rawline (@rawlines) {
958
959
		$linenr++;
		$line = $rawline;
960

961
		if ($rawline=~/^\+\+\+\s+(\S+)/) {
962
963
964
965
			$setup_docs = 0;
			if ($1 =~ m@Documentation/kernel-parameters.txt$@) {
				$setup_docs = 1;
			}
966
967
968
969
970
971
972
973
974
			#next;
		}
		if ($rawline=~/^\@\@ -\d+(?:,\d+)? \+(\d+)(,(\d+))? \@\@/) {
			$realline=$1-1;
			if (defined $2) {
				$realcnt=$3+1;
			} else {
				$realcnt=1+1;
			}
975
			$in_comment = 0;
976
977
978
979
980
981

			# Guestimate if this is a continuing comment.  Run
			# the context looking for a comment "edge".  If this
			# edge is a close comment then we must be in a comment
			# at context start.
			my $edge;
982
			for (my $ln = $linenr + 1; $ln < ($linenr + $realcnt); $ln++) {
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
				next if ($line =~ /^-/);
				($edge) = ($rawlines[$ln - 1] =~ m@(/\*|\*/)@);
				last if (defined $edge);
			}
			if (defined $edge && $edge eq '*/') {
				$in_comment = 1;
			}

			# Guestimate if this is a continuing comment.  If this
			# is the start of a diff block and this line starts
			# ' *' then it is very likely a comment.
			if (!defined $edge &&
			    $rawlines[$linenr] =~ m@^.\s* \*(?:\s|$)@)
			{
				$in_comment = 1;
			}

			##print "COMMENT:$in_comment edge<$edge> $rawline\n";
			sanitise_line_reset($in_comment);

1003
		} elsif ($realcnt && $rawline =~ /^(?:\+| |$)/) {
1004
			# Standardise the strings and chars within the input to
1005
			# simplify matching -- only bother with positive lines.
1006
			$line = sanitise_line($rawline);
1007
		}
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
		push(@lines, $line);

		if ($realcnt > 1) {
			$realcnt-- if ($line =~ /^(?:\+| |$)/);
		} else {
			$realcnt = 0;
		}

		#print "==>$rawline\n";
		#print "-->$line\n";
1018
1019
1020
1021
1022
1023

		if ($setup_docs && $line =~ /^\+/) {
			push(@setup_docs, $line);
		}
	}

1024
1025
	$prefix = '';

1026
1027
	$realcnt = 0;
	$linenr = 0;
1028
1029
1030
	foreach my $line (@lines) {
		$linenr++;

1031
		my $rawline = $rawlines[$linenr - 1];
1032

1033
#extract the line range in the file after the patch is applied
1034
		if ($line=~/^\@\@ -\d+(?:,\d+)? \+(\d+)(,(\d+))? \@\@/) {
1035
			$is_patch = 1;
1036
			$first_line = $linenr + 1;
1037
1038
1039
1040
1041
1042
			$realline=$1-1;
			if (defined $2) {
				$realcnt=$3+1;
			} else {
				$realcnt=1+1;
			}
1043
			annotate_reset();
1044
1045
			$prev_values = 'E';

1046
			%suppress_ifbraces = ();
1047
1048
			next;

1049
1050
1051
# track the line number as we move through the hunk, note that
# new versions of GNU diff omit the leading space on completely
# blank context lines so we need to count that too.
1052
		} elsif ($line =~ /^( |\+|$)/) {
1053
			$realline++;
1054
			$realcnt-- if ($realcnt != 0);
1055

1056
			# Measure the line length and indent.
1057
			($length, $indent) = line_stats($rawline);
1058
1059
1060
1061

			# Track the previous line.
			($prevline, $stashline) = ($stashline, $line);
			($previndent, $stashindent) = ($stashindent, $indent);
1062
1063
			($prevrawline, $stashrawline) = ($stashrawline, $rawline);

1064
			#warn "line<$line>\n";
1065

1066
1067
		} elsif ($realcnt == 1) {
			$realcnt--;
1068
1069
1070
		}

#make up the handle for any error we report on this line
1071
1072
1073
		$prefix = "$filename:$realline: " if ($emacs && $file);
		$prefix = "$filename:$linenr: " if ($emacs && !$file);

1074
1075
		$here = "#$linenr: " if (!$file);
		$here = "#$realline: " if ($file);
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087

		# extract the filename as it passes
		if ($line=~/^\+\+\+\s+(\S+)/) {
			$realfile = $1;
			$realfile =~ s@^[^/]*/@@;

			if ($realfile =~ m@include/asm/@) {
				ERROR("do not modify files in include/asm, change architecture specific files in include/asm-<architecture>\n" . "$here$rawline\n");
			}
			next;
		}

1088
		$here .= "FILE: $realfile:$realline:" if ($realcnt != 0);
1089

1090
1091
1092
		my $hereline = "$here\n$rawline\n";
		my $herecurr = "$here\n$rawline\n";
		my $hereprev = "$here\n$prevrawline\n$rawline\n";
1093

1094
1095
		$cnt_lines++ if ($realcnt != 0);

1096
#check the patch for a signoff:
1097
		if ($line =~ /^\s*signed-off-by:/i) {
1098
1099
			# This is a signoff, if ugly, so do not double report.
			$signoff++;
1100
			if (!($line =~ /^\s*Signed-off-by:/)) {
1101
1102
				WARN("Signed-off-by: is the preferred form\n" .
					$herecurr);
1103
1104
			}
			if ($line =~ /^\s*signed-off-by:\S/i) {
1105
				WARN("space required after Signed-off-by:\n" .
1106
					$herecurr);
1107
1108
1109
			}
		}

1110
# Check for wrappage within a valid hunk of the file
1111
		if ($realcnt != 0 && $line !~ m{^(?:\+|-| |\\ No newline|$)}) {
1112
			ERROR("patch seems to be corrupt (line wrapped?)\n" .
1113
				$herecurr) if (!$emitted_corrupt++);
1114
1115
1116
1117
		}

# UTF-8 regex found at http://www.w3.org/International/questions/qa-forms-utf-8.en.php
		if (($realfile =~ /^$/ || $line =~ /^\+/) &&
1118
1119
1120
1121
1122
1123
1124
1125
		    $rawline !~ m/^$UTF8*$/) {
			my ($utf8_prefix) = ($rawline =~ /^($UTF8*)/);

			my $blank = copy_spacing($rawline);
			my $ptr = substr($blank, 0, length($utf8_prefix)) . "^";
			my $hereptr = "$hereline$ptr\n";

			ERROR("Invalid UTF-8, patch and commit message should be encoded in UTF-8\n" . $hereptr);
1126
1127
1128
1129
		}

#ignore lines being removed
		if ($line=~/^-/) {next;}
1130

1131
1132
# check we are in a valid source file if not then ignore this hunk
		next if ($realfile !~ /\.(h|c|s|S|pl|sh)$/);
1133
1134

#trailing whitespace
1135
		if ($line =~ /^\+.*\015/) {
1136
			my $herevet = "$here\n" . cat_vet($rawline) . "\n";
1137
1138
			ERROR("DOS line endings\n" . $herevet);

1139
1140
		} elsif ($rawline =~ /^\+.*\S\s+$/ || $rawline =~ /^\+\s+$/) {
			my $herevet = "$here\n" . cat_vet($rawline) . "\n";
1141
			ERROR("trailing whitespace\n" . $herevet);
1142
1143
		}
#80 column limit
1144
		if ($line =~ /^\+/ && $prevrawline !~ /\/\*\*/ &&
1145
1146
1147
		    $rawline !~ /^.\s*\*\s*\@$Ident\s/ &&
		    $line !~ /^\+\s*printk\s*\(\s*(?:KERN_\S+\s*)?"[X\t]*"\s*(?:,|\)\s*;)\s*$/ &&
		    $length > 80)
1148
		{
1149
			WARN("line over 80 characters\n" . $herecurr);
1150
1151
		}