Skip to content
Snippets Groups Projects
checkpatch.pl 48.6 KiB
Newer Older
#!/usr/bin/perl -w
# (c) 2001, Dave Jones. <davej@codemonkey.org.uk> (the file handling bit)
# (c) 2005, Joel Schopp <jschopp@austin.ibm.com> (the ugly bit)
# (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;
$P =~ s@.*/@@g;
my $V = '0.14';

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

my $quiet = 0;
my $tree = 1;
my $chk_signoff = 1;
my $chk_patch = 1;
my $tst_type = 0;
my $emacs = 0;
my $terse = 0;
my $file = 0;
my $check = 0;
my $summary = 1;
my $mailback = 0;
my $summary_file = 0;
GetOptions(
	'q|quiet+'	=> \$quiet,
	'tree!'		=> \$tree,
	'signoff!'	=> \$chk_signoff,
	'patch!'	=> \$chk_patch,
	'emacs!'	=> \$emacs,
	'terse!'	=> \$terse,
	'file!'		=> \$file,
	'subjective!'	=> \$check,
	'strict!'	=> \$check,
	'root=s'	=> \$root,
	'summary!'	=> \$summary,
	'mailback!'	=> \$mailback,
	'summary-file!'	=> \$summary_file,

	'debug=s'	=> \%debug,
	'test-type!'	=> \$tst_type,
) or exit;

my $exit = 0;

if ($#ARGV < 0) {
	print "usage: $P [options] patchfile\n";
	print "version: $V\n";
	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";
my $dbg_values = 0;
my $dbg_possible = 0;
for my $key (keys %debug) {
	eval "\${dbg_$key} = '$debug{$key}';"
}

if ($terse) {
	$emacs = 1;
	$quiet++;
}

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);
	}
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|
			__kprobes|
			fastcall
		}x;
our $Attribute	= qr{
			const|
			__read_mostly|
			__kprobes|
			__(?:mem|cpu|dev|)(?:initdata|init)
		  }x;
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{
			<=|>=|==|!=|
			=>|->|<<|>>|<|>|!|~|
			&&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/|%
our $NonptrType;
our $Type;
our $Declare;

our @typeList = (
	qr{void},
	qr{char},
	qr{short},
	qr{int},
	qr{long},
	qr{unsigned},
	qr{float},
	qr{double},
	qr{bool},
	qr{long\s+int},
	qr{long\s+long},
	qr{long\s+long\s+int},
	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},
);

sub build_types {
	my $all = "(?:  \n" . join("|\n  ", @typeList) . "\n)";
	$NonptrType	= qr{
			\b
			(?:const\s+)?
			(?:unsigned\s+)?
			$all
			(?:\s+$Sparse|\s+const)*
			\b
		  }x;
	$Type	= qr{
			\b$NonptrType\b
			(?:\s*\*+\s*const|\s*\*+|(?:\s*\[\s*\])+)?
			(?:\s+$Inline|\s+$Sparse|\s+$Attribute)*
		  }x;
	$Declare	= qr{(?:$Storage\s+)?$Type};
}
build_types();

$chk_signoff = 0 if ($file);

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

				} elsif ($entry !~ m@/@) {
					push(@dep_functions, $entry);
				}
my @rawlines = ();
my @lines = ();
my $vname;
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 - $!\n";
	if ($filename eq '-') {
		$vname = 'Your patch';
	} else {
		$vname = $filename;
	}
	while (<FILE>) {
		chomp;
		push(@rawlines, $_);
	}
	close(FILE);
	if (!process($filename)) {
		$exit = 1;
	}
	@rawlines = ();
	@lines = ();
}

exit($exit);

sub top_of_kernel_tree {
	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;
		}
}

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 $res;
}
sub copy_spacing {
	my ($str) = @_;

	my $res = '';
	for my $c (split(//, $str)) {
		if ($c eq "\t") {
			$res .= $c;
		} else {
			$res .= ' ';
		}
	}

	return $res;
}
sub line_stats {
	my ($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));
}

sub sanitise_line {
	my ($line) = @_;

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

	my $quote = '';
	my $qlen = 0;

	foreach my $c (split(//, $line)) {
		# The second backslash of a pair is not a "quote".
		if ($l eq "\\" && $c eq "\\") {
			$c = 'X';
		}
		if ($l ne "\\" && ($c eq "'" || $c eq '"')) {
			if ($quote eq '') {
				$quote = $c;
				$res .= $c;
				$l = $c;
				next;
			} elsif ($quote eq $c) {
				$quote = '';
			}
		}
		if ($quote eq "'" && $qlen > 1) {
			$quote = '';
		}
		if ($quote && $c ne "\t") {
			$res .= "X";
	# Clear out the comments.
	while ($res =~ m@(/\*.*?\*/)@g) {
		substr($res, $-[1], $+[1] - $-[1]) = $; x ($+[1] - $-[1]);
	}
	if ($res =~ m@(/\*.*)@) {
		substr($res, $-[1], $+[1] - $-[1]) = $; x ($+[1] - $-[1]);
	}
	if ($res =~ m@^.(.*\*/)@) {
		substr($res, $-[1], $+[1] - $-[1]) = $; x ($+[1] - $-[1]);
	}

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

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

sub ctx_statement_block {
	my ($linenr, $remain, $off) = @_;
	my $line = $linenr - 1;
	my $blk = '';
	my $soff = $off;
	my $coff = $off - 1;

	my $type = '';
	my $level = 0;
	my $c;
	my $len = 0;
	while (1) {
		#warn "CSB: blk<$blk>\n";
		# If we are about to drop off the end, pull in more
		# context.
		if ($off >= $len) {
			for (; $remain > 0; $line++) {
				next if ($lines[$line] =~ /^-/);
				$loff = $len;
				$blk .= $lines[$line] . "\n";
				$len = length($blk);
				$line++;
				last;
			}
			# Bail if there is no further context.
			#warn "CSB: blk<$blk> off<$off> len<$len>\n";
			if ($off >= $len) {
				last;
			}
		}
		$c = substr($blk, $off, 1);
		$remainder = substr($blk, $off);

		#warn "CSB: c<$c> type<$type> level<$level>\n";
		# Statement ends at the ';' or a close '}' at the
		# outermost level.
		if ($level == 0 && $c eq ';') {
			last;
		}

		# An else is really a conditional as long as its not else if
		if ($level == 0 && $remainder =~ /(\s+else)(?:\s|{)/ &&
					$remainder !~ /\s+else\s+if\b/) {
			$coff = $off + length($1);
		}

		if (($type eq '' || $type eq '(') && $c eq '(') {
			$level++;
			$type = '(';
		}
		if ($type eq '(' && $c eq ')') {
			$level--;
			$type = ($level != 0)? '(' : '';

			if ($level == 0 && $coff < $soff) {
				$coff = $off;
			}
		}
		if (($type eq '' || $type eq '{') && $c eq '{') {
			$level++;
			$type = '{';
		}
		if ($type eq '{' && $c eq '}') {
			$level--;
			$type = ($level != 0)? '{' : '';

			if ($level == 0) {
				last;
			}
		}
		$off++;
	}
	if ($off == $len) {
		$line++;
		$remain--;
	}

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

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

	#print "off<$off> loff<$loff>\n";

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

sub ctx_statement_full {
	my ($linenr, $remain, $off) = @_;
	my ($statement, $condition, $level);

	my (@chunks);

	($statement, $condition, $linenr, $remain, $off, $level) =
				ctx_statement_block($linenr, $remain, $off);
	#print "F: c<$condition> s<$statement>\n";
	for (;;) {
		push(@chunks, [ $condition, $statement ]);
		last if (!($remain > 0 && $condition =~ /^.\s*(?:if|else|do)/));
		($statement, $condition, $linenr, $remain, $off, $level) =
				ctx_statement_block($linenr, $remain, $off);
		#print "C: c<$condition> s<$statement>\n";
	}

	return ($level, $linenr, @chunks);
sub ctx_block_get {
	my ($linenr, $remain, $outer, $open, $close, $off) = @_;
	my $line;
	my $start = $linenr - 1;
	my $blk = '';
	my @o;
	my @c;
	my @res = ();

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

		$blk .= $rawlines[$line];
		foreach my $c (split(//, $rawlines[$line])) {
			##print "C<$c>L<$level><$open$close>O<$off>\n";
			if ($off > 0) {
				$off--;
				next;
			}
			if ($c eq $close && $level > 0) {
				$level--;
				last if ($level == 0);
			} elsif ($c eq $open) {
				$level++;
			}
		}
		if (!$outer || $level <= 1) {
			push(@res, $rawlines[$line]);
		last if ($level == 0);
	return ($level, @res);
}
sub ctx_block_outer {
	my ($linenr, $remain) = @_;

	my ($level, @r) = ctx_block_get($linenr, $remain, 1, '{', '}', 0);
	return @r;
}
sub ctx_block {
	my ($linenr, $remain) = @_;

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

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

	return ctx_block_get($linenr, $remain, 0, '{', '}', 0);
sub ctx_statement_level {
	my ($linenr, $remain, $off) = @_;

	return ctx_block_get($linenr, $remain, 0, '(', ')', $off);
}

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

	# Catch a comment on the end of the line itself.
	my ($current_comment) = ($rawlines[$end_line - 1] =~ m@.*(/\*.*\*/)\s*$@);
	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++) {
		my $line = $rawlines[$linenr - 1];
		#warn "           $line\n";
		if ($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);

	##print "LINE: $rawlines[$end_line - 1 ]\n";
	##print "CMMT: $cmt\n";

	return ($cmt ne '');
}

sub cat_vet {
	my ($vet) = @_;
	my ($res, $coded);
	$res = '';
	while ($vet =~ /([^[:cntrl:]]*)([[:cntrl:]]|$)/g) {
		$res .= $1;
		if ($2 ne '') {
			$coded = sprintf("^%c", unpack('C', $2) + 64);
			$res .= $coded;
	$res =~ s/$/\$/;
	return $res;
my $av_preprocessor = 0;
my $av_paren = 0;
my @av_paren_type;

sub annotate_reset {
	$av_preprocessor = 0;
	$av_paren = 0;
	@av_paren_type = ();
}

sub annotate_values {
	my ($stream, $type) = @_;
	my $res;
	my $cur = $stream;

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

	while (length($cur)) {
		print " <$type> " if ($dbg_values > 1);
		if ($cur =~ /^(\s+)/o) {
			print "WS($1)\n" if ($dbg_values > 1);
			if ($1 =~ /\n/ && $av_preprocessor) {
				$av_preprocessor = 0;
		} elsif ($cur =~ /^($Type)/) {
			print "DECLARE($1)\n" if ($dbg_values > 1);
			$type = 'T';

		} elsif ($cur =~ /^(#\s*define\s*$Ident)(\(?)/o) {
			print "DEFINE($1)\n" if ($dbg_values > 1);
			$av_preprocessor = 1;
			$av_paren_type[$av_paren] = 'N';
		} elsif ($cur =~ /^(#\s*(?:ifdef|ifndef|if|else|elif|endif))/o) {
			print "PRE($1)\n" if ($dbg_values > 1);
			$av_preprocessor = 1;
			$type = 'N';

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

		} elsif ($cur =~ /^(sizeof)\s*(\()?/o) {
			print "SIZEOF($1)\n" if ($dbg_values > 1);
			if (defined $2) {
				$av_paren_type[$av_paren] = 'V';
		} elsif ($cur =~ /^(if|while|typeof|__typeof__|for)\b/o) {
			print "COND($1)\n" if ($dbg_values > 1);
			$av_paren_type[$av_paren] = 'N';
			$type = 'N';

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

		} elsif ($cur =~ /^(\()/o) {
			print "PAREN('$1')\n" if ($dbg_values > 1);
			$av_paren++;
			$type = 'N';

		} elsif ($cur =~ /^(\))/o) {
			$av_paren-- if ($av_paren > 0);
			if (defined $av_paren_type[$av_paren]) {
				$type = $av_paren_type[$av_paren];
				undef $av_paren_type[$av_paren];
				print "PAREN('$1') -> $type\n"
							if ($dbg_values > 1);
				print "PAREN('$1')\n" if ($dbg_values > 1);
			}

		} elsif ($cur =~ /^($Ident)\(/o) {
			print "FUNC($1)\n" if ($dbg_values > 1);
			$av_paren_type[$av_paren] = 'V';

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

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

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

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

		} elsif ($cur =~ /(^.)/o) {
			print "C($1)\n" if ($dbg_values > 1);
		}
		if (defined $1) {
			$cur = substr($cur, length($1));
			$res .= $type x length($1);
		}
	return $res;
sub possible {
	my ($possible, $line) = @_;

	#print "CHECK<$possible>\n";
	if ($possible !~ /^(?:$Storage|$Type|DEFINE_\S+)$/ &&
	    $possible ne 'goto' && $possible ne 'return' &&
	    $possible ne 'struct' && $possible ne 'enum' &&
	    $possible ne 'case' && $possible ne 'else' &&
	    $possible ne 'typedef') {
		warn "POSSIBLE: $possible ($line)\n" if ($dbg_possible);
		push(@typeList, $possible);
		build_types();
	}
}

my $prefix = '';

sub report {
	my $line = $prefix . $_[0];

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

	push(our @report, $line);
}
sub report_dump {
	our @report;
sub ERROR {
	report("ERROR: $_[0]\n");
	our $clean = 0;
	our $cnt_error++;
	report("WARNING: $_[0]\n");
	our $clean = 0;
	our $cnt_warn++;
	if ($check) {
		report("CHECK: $_[0]\n");
		our $clean = 0;
		our $cnt_chk++;
	}
sub process {
	my $filename = shift;

	my $linenr=0;
	my $prevline="";
	my $prevrawline="";
	my $stashline="";
	my $stashrawline="";
	my $length;
	my $indent;
	my $previndent=0;
	my $stashindent=0;

	our $clean = 1;
	my $signoff = 0;
	my $is_patch = 0;

	our @report = ();
	our $cnt_lines = 0;
	our $cnt_error = 0;
	our $cnt_warn = 0;
	our $cnt_chk = 0;

	# Trace the real file/line as we go.
	my $realfile = '';
	my $realline = 0;
	my $realcnt = 0;
	my $here = '';
	my $in_comment = 0;
	my $comment_edge = 0;
	my $first_line = 0;

	my $prev_values = 'E';

	# suppression flags
	my $suppress_ifbraces = 0;
	# Pre-scan the patch sanitizing the lines.
	# Pre-scan the patch looking for any __setup documentation.
	my @setup_docs = ();
	my $setup_docs = 0;
	my $line;
	foreach my $rawline (@rawlines) {
		# Standardise the strings and chars within the input to
		# simplify matching.
		$line = sanitise_line($rawline);
		push(@lines, $line);

		##print "==>$rawline\n";
		##print "-->$line\n";

		if ($line=~/^\+\+\+\s+(\S+)/) {
			$setup_docs = 0;
			if ($1 =~ m@Documentation/kernel-parameters.txt$@) {
				$setup_docs = 1;
			}
			next;
		}

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

	foreach my $line (@lines) {
		$linenr++;

		my $rawline = $rawlines[$linenr - 1];
#extract the filename as it passes
		if ($line=~/^\+\+\+\s+(\S+)/) {
			$realfile=$1;
			$realfile =~ s@^[^/]*/@@;
			$in_comment = 0;
			next;
		}
#extract the line range in the file after the patch is applied
		if ($line=~/^\@\@ -\d+(?:,\d+)? \+(\d+)(,(\d+))? \@\@/) {
			$is_patch = 1;
			$first_line = $linenr + 1;
			$in_comment = 0;
			$realline=$1-1;
			if (defined $2) {
				$realcnt=$3+1;
			} else {
				$realcnt=1+1;
			}
			annotate_reset();
			$prev_values = 'E';

			$suppress_ifbraces = $linenr - 1;
# track the line number as we move through the hunk, note that
# new versions of GNU diff omit the leading space on completely
# blank context lines so we need to count that too.
		if ($line =~ /^( |\+|$)/) {
			$realline++;
			$realcnt-- if ($realcnt != 0);
			# 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.
			if ($linenr == $first_line) {
				my $edge;
				for (my $ln = $first_line; $ln < ($linenr + $realcnt); $ln++) {
					($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 ($linenr == $first_line and $rawline =~ m@^.\s* \*(?:\s|$)@) {
				$in_comment = 1;
			}

			# Find the last comment edge on _this_ line.
			$comment_edge = 0;
			while (($rawline =~ m@(/\*|\*/)@g)) {
				if ($1 eq '/*') {
					$in_comment = 1;
				} else {
					$in_comment = 0;
				}
				$comment_edge = 1;
			# Measure the line length and indent.
			($length, $indent) = line_stats($rawline);

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

			#warn "ic<$in_comment> ce<$comment_edge> line<$line>\n";
		} elsif ($realcnt == 1) {
			$realcnt--;
		}

#make up the handle for any error we report on this line
		$here = "#$linenr: " if (!$file);
		$here = "#$realline: " if ($file);
		$here .= "FILE: $realfile:$realline:" if ($realcnt != 0);
		my $hereline = "$here\n$rawline\n";
		my $herecurr = "$here\n$rawline\n";
		my $hereprev = "$here\n$prevrawline\n$rawline\n";
		$prefix = "$filename:$realline: " if ($emacs && $file);
		$prefix = "$filename:$linenr: " if ($emacs && !$file);
		$cnt_lines++ if ($realcnt != 0);

#check the patch for a signoff:
		if ($line =~ /^\s*signed-off-by:/i) {
			# This is a signoff, if ugly, so do not double report.
			$signoff++;
			if (!($line =~ /^\s*Signed-off-by:/)) {
				WARN("Signed-off-by: is the preferred form\n" .
					$herecurr);
			}
			if ($line =~ /^\s*signed-off-by:\S/i) {
				WARN("need space after Signed-off-by:\n" .
					$herecurr);
# Check for wrappage within a valid hunk of the file
		if ($realcnt != 0 && $line !~ m{^(?:\+|-| |\\ No newline|$)}) {
			ERROR("patch seems to be corrupt (line wrapped?)\n" .
				$herecurr) if (!$emitted_corrupt++);
		}

# UTF-8 regex found at http://www.w3.org/International/questions/qa-forms-utf-8.en.php
		if (($realfile =~ /^$/ || $line =~ /^\+/) &&
		     !($rawline =~ m/^(
				[\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 )) {
			ERROR("Invalid UTF-8, patch and commit message should be encoded in UTF-8\n" . $herecurr);
		}

#ignore lines being removed
		if ($line=~/^-/) {next;}
# check we are in a valid source file if not then ignore this hunk
		next if ($realfile !~ /\.(h|c|s|S|pl|sh)$/);

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

		} elsif ($rawline =~ /^\+.*\S\s+$/ || $rawline =~ /^\+\s+$/) {
			my $herevet = "$here\n" . cat_vet($rawline) . "\n";
			ERROR("trailing whitespace\n" . $herevet);
		}
#80 column limit
		if ($line =~ /^\+/ && !($prevrawline=~/\/\*\*/) && $length > 80) {
			WARN("line over 80 characters\n" . $herecurr);
# check for adding lines without a newline.
		if ($line =~ /^\+/ && defined $lines[$linenr] && $lines[$linenr] =~ /^\\ No newline at end of file/) {
			WARN("adding a line without newline at end of file\n" . $herecurr);
		}

# check we are in a valid source file *.[hc] if not then ignore this hunk
		next if ($realfile !~ /\.[hc]$/);

# at the beginning of a line any tabs must come first and anything
# more than 8 must use tabs.
		if ($rawline =~ /^\+\s* \t\s*\S/ ||
		    $rawline =~ /^\+\s*        \s*/) {
			my $herevet = "$here\n" . cat_vet($rawline) . "\n";
			ERROR("use tabs not spaces\n" . $herevet);
# check for RCS/CVS revision markers
		if ($rawline =~ /\$(Revision|Log|Id)(?:\$|)/) {
			WARN("CVS style keyword markers, these will _not_ be updated\n". $herecurr);
		}

# The rest of our checks refer specifically to C style
# only apply those _outside_ comments.  Only skip
# lines in the middle of comments.
		next if (!$comment_edge && $in_comment);
# Check for potential 'bare' types