1#!/usr/bin/env perl
2#
3#	namespace.pl.  Mon Aug 30 2004
4#
5#	Perform a name space analysis on the linux kernel.
6#
7#	Copyright Keith Owens <kaos@ocs.com.au>.  GPL.
8#
9#	Invoke by changing directory to the top of the kernel object
10#	tree then namespace.pl, no parameters.
11#
12#	Tuned for 2.1.x kernels with the new module handling, it will
13#	work with 2.0 kernels as well.
14#
15#	Last change 2.6.9-rc1, adding support for separate source and object
16#	trees.
17#
18#	The source must be compiled/assembled first, the object files
19#	are the primary input to this script.  Incomplete or missing
20#	objects will result in a flawed analysis.  Compile both vmlinux
21#	and modules.
22#
23#	Even with complete objects, treat the result of the analysis
24#	with caution.  Some external references are only used by
25#	certain architectures, others with certain combinations of
26#	configuration parameters.  Ideally the source should include
27#	something like
28#
29#	#ifndef CONFIG_...
30#	static
31#	#endif
32#	symbol_definition;
33#
34#	so the symbols are defined as static unless a particular
35#	CONFIG_... requires it to be external.
36#
37#	A symbol that is suffixed with '(export only)' has these properties
38#
39#	* It is global.
40#	* It is marked EXPORT_SYMBOL or EXPORT_SYMBOL_GPL, either in the same
41#	  source file or a different source file.
42#	* Given the current .config, nothing uses the symbol.
43#
44#	The symbol is a candidate for conversion to static, plus removal of the
45#	export.  But be careful that a different .config might use the symbol.
46#
47#
48#	Name space analysis and cleanup is an iterative process.  You cannot
49#	expect to find all the problems in a single pass.
50#
51#	* Identify possibly unnecessary global declarations, verify that they
52#	  really are unnecessary and change them to static.
53#	* Compile and fix up gcc warnings about static, removing dead symbols
54#	  as necessary.
55#	* make clean and rebuild with different configs (especially
56#	  CONFIG_MODULES=n) to see which symbols are being defined when the
57#	  config does not require them.  These symbols bloat the kernel object
58#	  for no good reason, which is frustrating for embedded systems.
59#	* Wrap config sensitive symbols in #ifdef CONFIG_foo, as long as the
60#	  code does not get too ugly.
61#	* Repeat the name space analysis until you can live with with the
62#	  result.
63#
64
65use warnings;
66use strict;
67use File::Find;
68
69my $nm = ($ENV{'NM'} || "nm") . " -p";
70my $objdump = ($ENV{'OBJDUMP'} || "objdump") . " -s -j .comment";
71my $srctree = "";
72my $objtree = "";
73$srctree = "$ENV{'srctree'}/" if (exists($ENV{'srctree'}));
74$objtree = "$ENV{'objtree'}/" if (exists($ENV{'objtree'}));
75
76if ($#ARGV != -1) {
77	print STDERR "usage: $0 takes no parameters\n";
78	die("giving up\n");
79}
80
81my %nmdata = ();	# nm data for each object
82my %def = ();		# all definitions for each name
83my %ksymtab = ();	# names that appear in __ksymtab_
84my %ref = ();		# $ref{$name} exists if there is a true external reference to $name
85my %export = ();	# $export{$name} exists if there is an EXPORT_... of $name
86
87my %nmexception = (
88    'fs/ext3/bitmap'			=> 1,
89    'fs/ext4/bitmap'			=> 1,
90    'arch/x86/lib/thunk_32'		=> 1,
91    'arch/x86/lib/cmpxchg'		=> 1,
92    'arch/x86/vdso/vdso32/note'		=> 1,
93    'lib/irq_regs'			=> 1,
94    'usr/initramfs_data'		=> 1,
95    'drivers/scsi/aic94xx/aic94xx_dump'	=> 1,
96    'drivers/scsi/libsas/sas_dump'	=> 1,
97    'lib/dec_and_lock'			=> 1,
98    'drivers/ide/ide-probe-mini'	=> 1,
99    'usr/initramfs_data'		=> 1,
100    'drivers/acpi/acpia/exdump'		=> 1,
101    'drivers/acpi/acpia/rsdump'		=> 1,
102    'drivers/acpi/acpia/nsdumpdv'	=> 1,
103    'drivers/acpi/acpia/nsdump'		=> 1,
104    'arch/ia64/sn/kernel/sn2/io'	=> 1,
105    'arch/ia64/kernel/gate-data'	=> 1,
106    'security/capability'		=> 1,
107    'fs/ntfs/sysctl'			=> 1,
108    'fs/jfs/jfs_debug'			=> 1,
109);
110
111my %nameexception = (
112    'mod_use_count_'	 => 1,
113    '__initramfs_end'	=> 1,
114    '__initramfs_start'	=> 1,
115    '_einittext'	=> 1,
116    '_sinittext'	=> 1,
117    'kallsyms_names'	=> 1,
118    'kallsyms_num_syms'	=> 1,
119    'kallsyms_addresses'=> 1,
120    'kallsyms_offsets'	=> 1,
121    'kallsyms_relative_base'=> 1,
122    '__this_module'	=> 1,
123    '_etext'		=> 1,
124    '_edata'		=> 1,
125    '_end'		=> 1,
126    '__bss_start'	=> 1,
127    '_text'		=> 1,
128    '_stext'		=> 1,
129    '__gp'		=> 1,
130    'ia64_unw_start'	=> 1,
131    'ia64_unw_end'	=> 1,
132    '__init_begin'	=> 1,
133    '__init_end'	=> 1,
134    '__bss_stop'	=> 1,
135    '__nosave_begin'	=> 1,
136    '__nosave_end'	=> 1,
137    'pg0'		=> 1,
138    'vdso_enabled'	=> 1,
139    '__stack_chk_fail'  => 1,
140    'VDSO32_PRELINK'	=> 1,
141    'VDSO32_vsyscall'	=> 1,
142    'VDSO32_rt_sigreturn'=>1,
143    'VDSO32_sigreturn'	=> 1,
144);
145
146
147&find(\&linux_objects, '.');	# find the objects and do_nm on them
148&list_multiply_defined();
149&resolve_external_references();
150&list_extra_externals();
151
152exit(0);
153
154sub linux_objects
155{
156	# Select objects, ignoring objects which are only created by
157	# merging other objects.  Also ignore all of modules, scripts
158	# and compressed.  Most conglomerate objects are handled by do_nm,
159	# this list only contains the special cases.  These include objects
160	# that are linked from just one other object and objects for which
161	# there is really no permanent source file.
162	my $basename = $_;
163	$_ = $File::Find::name;
164	s:^\./::;
165	if (/.*\.o$/ &&
166		! (
167		m:/built-in.a$:
168		|| m:arch/x86/vdso/:
169		|| m:arch/x86/boot/:
170		|| m:arch/ia64/ia32/ia32.o$:
171		|| m:arch/ia64/kernel/gate-syms.o$:
172		|| m:arch/ia64/lib/__divdi3.o$:
173		|| m:arch/ia64/lib/__divsi3.o$:
174		|| m:arch/ia64/lib/__moddi3.o$:
175		|| m:arch/ia64/lib/__modsi3.o$:
176		|| m:arch/ia64/lib/__udivdi3.o$:
177		|| m:arch/ia64/lib/__udivsi3.o$:
178		|| m:arch/ia64/lib/__umoddi3.o$:
179		|| m:arch/ia64/lib/__umodsi3.o$:
180		|| m:arch/ia64/scripts/check_gas_for_hint.o$:
181		|| m:arch/ia64/sn/kernel/xp.o$:
182		|| m:boot/bbootsect.o$:
183		|| m:boot/bsetup.o$:
184		|| m:/bootsect.o$:
185		|| m:/boot/setup.o$:
186		|| m:/compressed/:
187		|| m:drivers/cdrom/driver.o$:
188		|| m:drivers/char/drm/tdfx_drv.o$:
189		|| m:drivers/ide/ide-detect.o$:
190		|| m:drivers/ide/pci/idedriver-pci.o$:
191		|| m:drivers/media/media.o$:
192		|| m:drivers/scsi/sd_mod.o$:
193		|| m:drivers/video/video.o$:
194		|| m:fs/devpts/devpts.o$:
195		|| m:fs/exportfs/exportfs.o$:
196		|| m:fs/hugetlbfs/hugetlbfs.o$:
197		|| m:fs/msdos/msdos.o$:
198		|| m:fs/nls/nls.o$:
199		|| m:fs/ramfs/ramfs.o$:
200		|| m:fs/romfs/romfs.o$:
201		|| m:fs/vfat/vfat.o$:
202		|| m:init/mounts.o$:
203		|| m:^modules/:
204		|| m:net/netlink/netlink.o$:
205		|| m:net/sched/sched.o$:
206		|| m:/piggy.o$:
207		|| m:^scripts/:
208		|| m:sound/.*/snd-:
209		|| m:^.*/\.tmp_:
210		|| m:^\.tmp_:
211		|| m:/vmlinux-obj.o$:
212		|| m:^tools/:
213		)
214	) {
215		do_nm($basename, $_);
216	}
217	$_ = $basename;		# File::Find expects $_ untouched (undocumented)
218}
219
220sub do_nm
221{
222	my ($basename, $fullname) = @_;
223	my ($source, $type, $name);
224	if (! -e $basename) {
225		printf STDERR "$basename does not exist\n";
226		return;
227	}
228	if ($fullname !~ /\.o$/) {
229		printf STDERR "$fullname is not an object file\n";
230		return;
231	}
232	($source = $basename) =~ s/\.o$//;
233	if (-e "$source.c" || -e "$source.S") {
234		$source = "$objtree$File::Find::dir/$source";
235	} else {
236		$source = "$srctree$File::Find::dir/$source";
237	}
238	if (! -e "$source.c" && ! -e "$source.S") {
239		# No obvious source, exclude the object if it is conglomerate
240	        open(my $objdumpdata, "$objdump $basename|")
241		    or die "$objdump $fullname failed $!\n";
242
243		my $comment;
244		while (<$objdumpdata>) {
245			chomp();
246			if (/^In archive/) {
247				# Archives are always conglomerate
248				$comment = "GCC:GCC:";
249				last;
250			}
251			next if (! /^[ 0-9a-f]{5,} /);
252			$comment .= substr($_, 43);
253		}
254		close($objdumpdata);
255
256		if (!defined($comment) || $comment !~ /GCC\:.*GCC\:/m) {
257			printf STDERR "No source file found for $fullname\n";
258		}
259		return;
260	}
261	open (my $nmdata, "$nm $basename|")
262	    or die "$nm $fullname failed $!\n";
263
264	my @nmdata;
265	while (<$nmdata>) {
266		chop;
267		($type, $name) = (split(/ +/, $_, 3))[1..2];
268		# Expected types
269		# A absolute symbol
270		# B weak external reference to data that has been resolved
271		# C global variable, uninitialised
272		# D global variable, initialised
273		# G global variable, initialised, small data section
274		# R global array, initialised
275		# S global variable, uninitialised, small bss
276		# T global label/procedure
277		# U external reference
278		# W weak external reference to text that has been resolved
279		# V similar to W, but the value of the weak symbol becomes zero with no error.
280		# a assembler equate
281		# b static variable, uninitialised
282		# d static variable, initialised
283		# g static variable, initialised, small data section
284		# r static array, initialised
285		# s static variable, uninitialised, small bss
286		# t static label/procedures
287		# w weak external reference to text that has not been resolved
288		# v similar to w
289		# ? undefined type, used a lot by modules
290		if ($type !~ /^[ABCDGRSTUWVabdgrstwv?]$/) {
291			printf STDERR "nm output for $fullname contains unknown type '$_'\n";
292		}
293		elsif ($name =~ /\./) {
294			# name with '.' is local static
295		}
296		else {
297			$type = 'R' if ($type eq '?');	# binutils replaced ? with R at one point
298			# binutils keeps changing the type for exported symbols, force it to R
299			$type = 'R' if ($name =~ /^__ksymtab/ || $name =~ /^__kstrtab/);
300			$name =~ s/_R[a-f0-9]{8}$//;	# module versions adds this
301			if ($type =~ /[ABCDGRSTWV]/ &&
302				$name ne 'init_module' &&
303				$name ne 'cleanup_module' &&
304				$name ne 'Using_Versions' &&
305				$name !~ /^Version_[0-9]+$/ &&
306				$name !~ /^__parm_/ &&
307				$name !~ /^__kstrtab/ &&
308				$name !~ /^__ksymtab/ &&
309				$name !~ /^__kcrctab_/ &&
310				$name !~ /^__exitcall_/ &&
311				$name !~ /^__initcall_/ &&
312				$name !~ /^__kdb_initcall_/ &&
313				$name !~ /^__kdb_exitcall_/ &&
314				$name !~ /^__module_/ &&
315				$name !~ /^__mod_/ &&
316				$name !~ /^__crc_/ &&
317				$name ne '__this_module' &&
318				$name ne 'kernel_version') {
319				if (!exists($def{$name})) {
320					$def{$name} = [];
321				}
322				push(@{$def{$name}}, $fullname);
323			}
324			push(@nmdata, "$type $name");
325			if ($name =~ /^__ksymtab_/) {
326				$name = substr($name, 10);
327				if (!exists($ksymtab{$name})) {
328					$ksymtab{$name} = [];
329				}
330				push(@{$ksymtab{$name}}, $fullname);
331			}
332		}
333	}
334	close($nmdata);
335
336	if ($#nmdata < 0) {
337	    printf "No nm data for $fullname\n"
338		unless $nmexception{$fullname};
339	    return;
340	}
341	$nmdata{$fullname} = \@nmdata;
342}
343
344sub drop_def
345{
346	my ($object, $name) = @_;
347	my $nmdata = $nmdata{$object};
348	my ($i, $j);
349	for ($i = 0; $i <= $#{$nmdata}; ++$i) {
350		if ($name eq (split(' ', $nmdata->[$i], 2))[1]) {
351			splice(@{$nmdata{$object}}, $i, 1);
352			my $def = $def{$name};
353			for ($j = 0; $j < $#{$def{$name}}; ++$j) {
354				if ($def{$name}[$j] eq $object) {
355					splice(@{$def{$name}}, $j, 1);
356				}
357			}
358			last;
359		}
360	}
361}
362
363sub list_multiply_defined
364{
365	foreach my $name (keys(%def)) {
366		if ($#{$def{$name}} > 0) {
367			# Special case for cond_syscall
368			if ($#{$def{$name}} == 1 &&
369			   ($name =~ /^sys_/ || $name =~ /^compat_sys_/ ||
370			    $name =~ /^sys32_/)) {
371				if($def{$name}[0] eq "kernel/sys_ni.o" ||
372				   $def{$name}[1] eq "kernel/sys_ni.o") {
373					&drop_def("kernel/sys_ni.o", $name);
374					next;
375				}
376			}
377
378			printf "$name is multiply defined in :-\n";
379			foreach my $module (@{$def{$name}}) {
380				printf "\t$module\n";
381			}
382		}
383	}
384}
385
386sub resolve_external_references
387{
388	my ($kstrtab, $ksymtab, $export);
389
390	printf "\n";
391	foreach my $object (keys(%nmdata)) {
392		my $nmdata = $nmdata{$object};
393		for (my $i = 0; $i <= $#{$nmdata}; ++$i) {
394			my ($type, $name) = split(' ', $nmdata->[$i], 2);
395			if ($type eq "U" || $type eq "w") {
396				if (exists($def{$name}) || exists($ksymtab{$name})) {
397					# add the owning object to the nmdata
398					$nmdata->[$i] = "$type $name $object";
399					# only count as a reference if it is not EXPORT_...
400					$kstrtab = "R __kstrtab_$name";
401					$ksymtab = "R __ksymtab_$name";
402					$export = 0;
403					for (my $j = 0; $j <= $#{$nmdata}; ++$j) {
404						if ($nmdata->[$j] eq $kstrtab ||
405						    $nmdata->[$j] eq $ksymtab) {
406							$export = 1;
407							last;
408						}
409					}
410					if ($export) {
411						$export{$name} = "";
412					}
413					else {
414						$ref{$name} = ""
415					}
416				}
417				elsif ( ! $nameexception{$name}
418					&& $name !~ /^__sched_text_/
419					&& $name !~ /^__start_/
420					&& $name !~ /^__end_/
421					&& $name !~ /^__stop_/
422					&& $name !~ /^__scheduling_functions_.*_here/
423					&& $name !~ /^__.*initcall_/
424					&& $name !~ /^__.*per_cpu_start/
425					&& $name !~ /^__.*per_cpu_end/
426					&& $name !~ /^__alt_instructions/
427					&& $name !~ /^__setup_/
428					&& $name !~ /^__mod_timer/
429					&& $name !~ /^__mod_page_state/
430					&& $name !~ /^init_module/
431					&& $name !~ /^cleanup_module/
432				) {
433					printf "Cannot resolve ";
434					printf "weak " if ($type eq "w");
435					printf "reference to $name from $object\n";
436				}
437			}
438		}
439	}
440}
441
442sub list_extra_externals
443{
444	my %noref = ();
445
446	foreach my $name (keys(%def)) {
447		if (! exists($ref{$name})) {
448			my @module = @{$def{$name}};
449			foreach my $module (@module) {
450				if (! exists($noref{$module})) {
451					$noref{$module} = [];
452				}
453				push(@{$noref{$module}}, $name);
454			}
455		}
456	}
457	if (%noref) {
458		printf "\nExternally defined symbols with no external references\n";
459		foreach my $module (sort(keys(%noref))) {
460			printf "  $module\n";
461			foreach (sort(@{$noref{$module}})) {
462			    my $export;
463			    if (exists($export{$_})) {
464				$export = " (export only)";
465			    } else {
466				$export = "";
467			    }
468			    printf "    $_$export\n";
469			}
470		}
471	}
472}
473