[PATCH v2] kconfig: add kconfig-sym-check static checker

[Andrew Jones <andrew.jones@linux.dev>]

Add 'make kconfig-sym-check', a static checker that finds Kconfig
symbols referenced in expressions (select, depends on, default, etc.)
but never defined via config/menuconfig anywhere in the tree. New
dangling symbols are reported as errors (exit 1) unless they are
listed in an exclusion file, e.g.

 KCONFIG_SYM_CHECK_EXCLUDES=sym-check-excludes make kconfig-sym-check

The checker also warns about uppercase N/Y/M used as tristate literal
values following the same logic as checkpatch.

This new static checker is the script used for [1] with a few
improvements to avoid some false positives.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=216748 [1]
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>
Acked-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Tested-by: Randy Dunlap <rdunlap@infradead.org>
---

v2:
  - Added Andy's and Randy's tags
  - Accept srctree as first argument so the Makefile can drop 'cd $(srctree)' [Nathan]
  - Replace git ls-files with git+find fallback [Nathan and Andy]
  - Changes thanks to sashiko's review
    - strip quoted strings before inline comments to avoid '#' inside a string
    - use [^)]* instead of .* in macro strip regex to avoid greedy match
      eating tokens between adjacent $(macro) expansions

 Makefile                             |  7 +-
 scripts/kconfig/kconfig-sym-check.pl | 97 ++++++++++++++++++++++++++++
 2 files changed, 103 insertions(+), 1 deletion(-)
 create mode 100755 scripts/kconfig/kconfig-sym-check.pl

diff --git a/Makefile b/Makefile
index fbc67fcb6cdb..b86ccc1e08cd 100644
--- a/Makefile
+++ b/Makefile
@@ -293,6 +293,7 @@ version_h := include/generated/uapi/linux/version.h
 clean-targets := %clean mrproper cleandocs
 no-dot-config-targets := $(clean-targets) \
 			 cscope gtags TAGS tags help% %docs check% coccicheck \
+			 kconfig-sym-check \
 			 $(version_h) headers headers_% archheaders archscripts \
 			 %asm-generic kernelversion %src-pkg dt_binding_check \
 			 outputmakefile rustavailable rustfmt rustfmtcheck \
@@ -1811,6 +1812,7 @@ help:
 	@echo  '  includecheck    - Check for duplicate included header files'
 	@echo  '  headerdep       - Detect inclusion cycles in headers'
 	@echo  '  coccicheck      - Check with Coccinelle'
+	@echo  '  kconfig-sym-check - Check for dangling Kconfig symbol references'
 	@echo  '  clang-analyzer  - Check with clang static analyzer'
 	@echo  '  clang-tidy      - Check with clang-tidy'
 	@echo  ''
@@ -2232,7 +2234,7 @@ endif
 # Scripts to check various things for consistency
 # ---------------------------------------------------------------------------
 
-PHONY += includecheck versioncheck coccicheck
+PHONY += includecheck versioncheck coccicheck kconfig-sym-check
 
 includecheck:
 	find $(srctree)/* $(RCS_FIND_IGNORE) \
@@ -2247,6 +2249,9 @@ versioncheck:
 coccicheck:
 	$(Q)$(BASH) $(srctree)/scripts/$@
 
+kconfig-sym-check:
+	$(Q)$(PERL) $(srctree)/scripts/kconfig/kconfig-sym-check.pl $(srctree) $(KCONFIG_SYM_CHECK_EXCLUDES)
+
 PHONY += checkstack kernelrelease kernelversion image_name
 
 # UML needs a little special treatment here.  It wants to use the host
diff --git a/scripts/kconfig/kconfig-sym-check.pl b/scripts/kconfig/kconfig-sym-check.pl
new file mode 100755
index 000000000000..6bf32fb37929
--- /dev/null
+++ b/scripts/kconfig/kconfig-sym-check.pl
@@ -0,0 +1,97 @@
+#!/usr/bin/env perl
+# SPDX-License-Identifier: GPL-2.0
+
+use warnings;
+use strict;
+
+my $srctree = shift @ARGV;
+my $kconfig_sym_check_excludes = defined $ARGV[0] ? $ARGV[0] : undef;
+
+my @files = `git -C \Q$srctree\E ls-files '*Kconfig*' 2>/dev/null`;
+if (@files) {
+	chomp @files;
+	@files = map { "$srctree/$_" } @files;
+} else {
+	@files = `find \Q$srctree\E -name '*Kconfig*'`;
+	chomp @files;
+}
+
+my %configs = ();
+my %refs = ();
+
+foreach my $file (@files) {
+	open F, $file or die "Cannot open $file: $!";
+
+	my $help = 0;
+	my $help_level;
+	my $level;
+
+	while (<F>) {
+		chomp;
+
+		next if /^\s*$/;
+		next if /^\s*#/;
+
+		/^(\s*)/;
+		$level = length $1;
+
+		if ($help && $level < $help_level) {
+			$help = 0;
+		}
+
+		next if ($help);
+
+		if (/^\s*(help|\-\-\-help\-\-\-)$/) {
+			$help = 1;
+			$_ = <F>;
+			/^(\s*)/;
+			$help_level = length $1;
+			next;
+		}
+
+		if (/^\s*(config|menuconfig)\s+([a-zA-Z0-9_]+)\s*(#.*)?$/) {
+			$configs{$2}++;
+			next;
+		}
+
+		if (/^\s*(default|def_bool|def_tristate|select|depends\s+on|imply|visible\s+if|range|if)\s+(.+)\s*$/) {
+			my $s = $2;
+			$s =~ s/"[^"]*"//g;
+			$s =~ s/'[^']*'//g;
+			$s =~ s/#.*//;
+			$s =~ s/\$\([^)]*\)//g;
+			$s =~ s/%%[^%]*%%//g;
+			my @syms = split /[^a-zA-Z0-9_]+/, $s;
+			map {
+				$refs{$_}++ if (/[a-zA-Z]/ && $_ ne "if" && $_ ne "y" && $_ ne "n" && $_ ne "m" && !(/^0[xX]/ && !/[g-wy-zG-WY-Z]/));
+			} @syms
+		}
+	}
+
+	close F;
+}
+
+my %known_syms = ();
+if (defined $kconfig_sym_check_excludes) {
+	my $file = $kconfig_sym_check_excludes;
+	open F, $file or die "Cannot open $file: $!";
+	while (<F>) {
+		chomp;
+		next if /^\s*$/;
+		next if /^\s*#/;
+		$known_syms{$1}++ if (/^\s*([a-zA-Z0-9_]+)\s*(#.*)?$/);
+	}
+}
+
+my $ret = 0;
+foreach my $k (sort keys %refs) {
+	next if (exists $configs{$k} || exists $known_syms{$k});
+
+	print "$k";
+	print " - warning: '$k' is probably not what you want; Kconfig tristate literals are always lowercase ('n', 'y', 'm')" if ($k eq "N" || $k eq "Y" || $k eq "M");
+	print "\n";
+
+	$ret = 1;
+}
+
+exit $ret;
-- 
2.43.0