]> git.proxmox.com Git - mirror_ubuntu-bionic-kernel.git/blame - scripts/recordmcount.pl
tracing: Add regex for weak functions in recordmcount.pl
[mirror_ubuntu-bionic-kernel.git] / scripts / recordmcount.pl
CommitLineData
8da3821b
SR
1#!/usr/bin/perl -w
2# (c) 2008, Steven Rostedt <srostedt@redhat.com>
3# Licensed under the terms of the GNU GPL License version 2
4#
5# recordmcount.pl - makes a section called __mcount_loc that holds
6# all the offsets to the calls to mcount.
7#
8#
d49f6aa7
LH
9# What we want to end up with this is that each object file will have a
10# section called __mcount_loc that will hold the list of pointers to mcount
11# callers. After final linking, the vmlinux will have within .init.data the
12# list of all callers to mcount between __start_mcount_loc and __stop_mcount_loc.
13# Later on boot up, the kernel will read this list, save the locations and turn
14# them into nops. When tracing or profiling is later enabled, these locations
15# will then be converted back to pointers to some function.
8da3821b
SR
16#
17# This is no easy feat. This script is called just after the original
18# object is compiled and before it is linked.
19#
d49f6aa7
LH
20# When parse this object file using 'objdump', the references to the call
21# sites are offsets from the section that the call site is in. Hence, all
22# functions in a section that has a call site to mcount, will have the
23# offset from the beginning of the section and not the beginning of the
24# function.
25#
26# But where this section will reside finally in vmlinx is undetermined at
27# this point. So we can't use this kind of offsets to record the final
28# address of this call site.
29#
30# The trick is to change the call offset referring the start of a section to
31# referring a function symbol in this section. During the link step, 'ld' will
32# compute the final address according to the information we record.
8da3821b 33#
8da3821b
SR
34# e.g.
35#
31b6e76e 36# .section ".sched.text", "ax"
8da3821b 37# [...]
d49f6aa7
LH
38# func1:
39# [...]
40# call mcount (offset: 0x10)
41# [...]
42# ret
43# .globl fun2
44# func2: (offset: 0x20)
45# [...]
8da3821b
SR
46# [...]
47# ret
d49f6aa7 48# func3:
8da3821b 49# [...]
d49f6aa7 50# call mcount (offset: 0x30)
8da3821b
SR
51# [...]
52#
53# Both relocation offsets for the mcounts in the above example will be
d49f6aa7
LH
54# offset from .sched.text. If we choose global symbol func2 as a reference and
55# make another file called tmp.s with the new offsets:
8da3821b
SR
56#
57# .section __mcount_loc
d49f6aa7
LH
58# .quad func2 - 0x10
59# .quad func2 + 0x10
8da3821b 60#
d49f6aa7 61# We can then compile this tmp.s into tmp.o, and link it back to the original
8da3821b
SR
62# object.
63#
d49f6aa7
LH
64# In our algorithm, we will choose the first global function we meet in this
65# section as the reference. But this gets hard if there is no global functions
66# in this section. In such a case we have to select a local one. E.g. func1:
8da3821b 67#
31b6e76e 68# .section ".sched.text", "ax"
d49f6aa7 69# func1:
8da3821b 70# [...]
d49f6aa7 71# call mcount (offset: 0x10)
8da3821b
SR
72# [...]
73# ret
d49f6aa7 74# func2:
8da3821b 75# [...]
d49f6aa7 76# call mcount (offset: 0x20)
8da3821b 77# [...]
d49f6aa7 78# .section "other.section"
8da3821b
SR
79#
80# If we make the tmp.s the same as above, when we link together with
d49f6aa7 81# the original object, we will end up with two symbols for func1:
8da3821b 82# one local, one global. After final compile, we will end up with
d49f6aa7
LH
83# an undefined reference to func1 or a wrong reference to another global
84# func1 in other files.
8da3821b
SR
85#
86# Since local objects can reference local variables, we need to find
87# a way to make tmp.o reference the local objects of the original object
d49f6aa7 88# file after it is linked together. To do this, we convert func1
8da3821b 89# into a global symbol before linking tmp.o. Then after we link tmp.o
d49f6aa7
LH
90# we will only have a single symbol for func1 that is global.
91# We can convert func1 back into a local symbol and we are done.
8da3821b
SR
92#
93# Here are the steps we take:
94#
306dcf47 95# 1) Record all the local and weak symbols by using 'nm'
8da3821b
SR
96# 2) Use objdump to find all the call site offsets and sections for
97# mcount.
98# 3) Compile the list into its own object.
99# 4) Do we have to deal with local functions? If not, go to step 8.
100# 5) Make an object that converts these local functions to global symbols
101# with objcopy.
102# 6) Link together this new object with the list object.
103# 7) Convert the local functions back to local symbols and rename
104# the result as the original object.
8da3821b
SR
105# 8) Link the object with the list object.
106# 9) Move the result back to the original object.
8da3821b
SR
107#
108
109use strict;
110
111my $P = $0;
112$P =~ s@.*/@@g;
113
114my $V = '0.1';
115
e2d753fa 116if ($#ARGV != 10) {
b43f7093 117 print "usage: $P arch bits objdump objcopy cc ld nm rm mv is_module inputfile\n";
8da3821b
SR
118 print "version: $V\n";
119 exit(1);
120}
121
dce9d18a 122my ($arch, $bits, $objdump, $objcopy, $cc,
18c167fd 123 $ld, $nm, $rm, $mv, $is_module, $inputfile) = @ARGV;
8da3821b 124
25aac9dc 125# This file refers to mcount and shouldn't be ftraced, so lets' ignore it
bdd3b052 126if ($inputfile =~ m,kernel/trace/ftrace\.o$,) {
25aac9dc
SL
127 exit(0);
128}
129
34698bcb
SR
130# Acceptable sections to record.
131my %text_sections = (
132 ".text" => 1,
d144d5ee
LW
133 ".sched.text" => 1,
134 ".spinlock.text" => 1,
a0343e82 135 ".irqentry.text" => 1,
4d828949 136 ".text.unlikely" => 1,
34698bcb
SR
137);
138
8da3821b
SR
139$objdump = "objdump" if ((length $objdump) == 0);
140$objcopy = "objcopy" if ((length $objcopy) == 0);
141$cc = "gcc" if ((length $cc) == 0);
142$ld = "ld" if ((length $ld) == 0);
143$nm = "nm" if ((length $nm) == 0);
144$rm = "rm" if ((length $rm) == 0);
145$mv = "mv" if ((length $mv) == 0);
146
147#print STDERR "running: $P '$arch' '$objdump' '$objcopy' '$cc' '$ld' " .
148# "'$nm' '$rm' '$mv' '$inputfile'\n";
149
8feff1ca
SR
150my %locals; # List of local (static) functions
151my %weak; # List of weak functions
152my %convert; # List of local functions used that needs conversion
8da3821b
SR
153
154my $type;
306dcf47
LH
155my $local_regex; # Match a local function (return function)
156my $weak_regex; # Match a weak function (return function)
8da3821b 157my $section_regex; # Find the start of a section
8feff1ca
SR
158my $function_regex; # Find the name of a function
159 # (return offset and func name)
8da3821b 160my $mcount_regex; # Find the call site to mcount (return offset)
3a3d04ae 161my $alignment; # The .align value to use for $mcount_section
e58918ab 162my $section_type; # Section header plus possible alignment command
7b7edc27
LH
163my $can_use_local = 0; # If we can use local function references
164
165##
166# check_objcopy - whether objcopy supports --globalize-symbols
167#
168# --globalize-symbols came out in 2.17, we must test the version
169# of objcopy, and if it is less than 2.17, then we can not
170# record local functions.
171sub check_objcopy
172{
173 open (IN, "$objcopy --version |") or die "error running $objcopy";
174 while (<IN>) {
175 if (/objcopy.*\s(\d+)\.(\d+)/) {
176 $can_use_local = 1 if ($1 > 2 || ($1 == 2 && $2 >= 17));
177 last;
178 }
179 }
180 close (IN);
181
182 if (!$can_use_local) {
183 print STDERR "WARNING: could not find objcopy version or version " .
184 "is less than 2.17.\n" .
185 "\tLocal function references is disabled.\n";
186 }
187}
8da3821b 188
dce9d18a
SR
189if ($arch eq "x86") {
190 if ($bits == 64) {
191 $arch = "x86_64";
192 } else {
193 $arch = "i386";
194 }
195}
196
c204f726
SR
197#
198# We base the defaults off of i386, the other archs may
199# feel free to change them in the below if statements.
200#
306dcf47
LH
201$local_regex = "^[0-9a-fA-F]+\\s+t\\s+(\\S+)";
202$weak_regex = "^[0-9a-fA-F]+\\s+([wW])\\s+(\\S+)";
c204f726
SR
203$section_regex = "Disassembly of section\\s+(\\S+):";
204$function_regex = "^([0-9a-fA-F]+)\\s+<(.*?)>:";
205$mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\smcount\$";
e58918ab 206$section_type = '@progbits';
c204f726
SR
207$type = ".long";
208
8da3821b 209if ($arch eq "x86_64") {
8da3821b
SR
210 $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\smcount([+-]0x[0-9a-zA-Z]+)?\$";
211 $type = ".quad";
7d5222a6 212 $alignment = 8;
d74fcd1e
SR
213
214 # force flags for this arch
215 $ld .= " -m elf_x86_64";
216 $objdump .= " -M x86-64";
217 $objcopy .= " -O elf64-x86-64";
218 $cc .= " -m64";
219
8da3821b 220} elsif ($arch eq "i386") {
7d5222a6 221 $alignment = 4;
d74fcd1e
SR
222
223 # force flags for this arch
224 $ld .= " -m elf_i386";
225 $objdump .= " -M i386";
226 $objcopy .= " -O elf32-i386";
227 $cc .= " -m32";
228
dfd9f7ab
HC
229} elsif ($arch eq "s390" && $bits == 32) {
230 $mcount_regex = "^\\s*([0-9a-fA-F]+):\\s*R_390_32\\s+_mcount\$";
231 $alignment = 4;
232 $ld .= " -m elf_s390";
233 $cc .= " -m31";
234
235} elsif ($arch eq "s390" && $bits == 64) {
236 $mcount_regex = "^\\s*([0-9a-fA-F]+):\\s*R_390_(PC|PLT)32DBL\\s+_mcount\\+0x2\$";
237 $alignment = 8;
238 $type = ".quad";
239 $ld .= " -m elf64_s390";
240 $cc .= " -m64";
241
0da85c09 242} elsif ($arch eq "sh") {
3a3d04ae 243 $alignment = 2;
0da85c09
MF
244
245 # force flags for this arch
246 $ld .= " -m shlelf_linux";
247 $objcopy .= " -O elf32-sh-linux";
248 $cc .= " -m32";
249
42e007d0 250} elsif ($arch eq "powerpc") {
306dcf47 251 $local_regex = "^[0-9a-fA-F]+\\s+t\\s+(\\.?\\S+)";
42e007d0
SR
252 $function_regex = "^([0-9a-fA-F]+)\\s+<(\\.?.*?)>:";
253 $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\s\\.?_mcount\$";
c204f726 254
42e007d0
SR
255 if ($bits == 64) {
256 $type = ".quad";
42e007d0
SR
257 }
258
e58918ab
JR
259} elsif ($arch eq "arm") {
260 $alignment = 2;
261 $section_type = '%progbits';
262
418071eb
SL
263} elsif ($arch eq "ia64") {
264 $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\s_mcount\$";
265 $type = "data8";
266
267 if ($is_module eq "0") {
268 $cc .= " -mconstant-gp";
269 }
9be12f9b
DM
270} elsif ($arch eq "sparc64") {
271 # In the objdump output there are giblets like:
272 # 0000000000000000 <igmp_net_exit-0x18>:
273 # As there's some data blobs that get emitted into the
274 # text section before the first instructions and the first
275 # real symbols. We don't want to match that, so to combat
276 # this we use '\w' so we'll match just plain symbol names,
277 # and not those that also include hex offsets inside of the
278 # '<>' brackets. Actually the generic function_regex setting
279 # could safely use this too.
280 $function_regex = "^([0-9a-fA-F]+)\\s+<(\\w*?)>:";
281
282 # Sparc64 calls '_mcount' instead of plain 'mcount'.
283 $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\s_mcount\$";
284
285 $alignment = 8;
286 $type = ".xword";
287 $ld .= " -m elf64_sparc";
288 $cc .= " -m64";
289 $objcopy .= " -O elf64-sparc";
8da3821b
SR
290} else {
291 die "Arch $arch is not supported with CONFIG_FTRACE_MCOUNT_RECORD";
292}
293
294my $text_found = 0;
295my $read_function = 0;
296my $opened = 0;
8da3821b
SR
297my $mcount_section = "__mcount_loc";
298
299my $dirname;
300my $filename;
301my $prefix;
302my $ext;
303
304if ($inputfile =~ m,^(.*)/([^/]*)$,) {
305 $dirname = $1;
306 $filename = $2;
307} else {
308 $dirname = ".";
309 $filename = $inputfile;
310}
311
312if ($filename =~ m,^(.*)(\.\S),) {
313 $prefix = $1;
314 $ext = $2;
315} else {
316 $prefix = $filename;
317 $ext = "";
318}
319
320my $mcount_s = $dirname . "/.tmp_mc_" . $prefix . ".s";
321my $mcount_o = $dirname . "/.tmp_mc_" . $prefix . ".o";
322
7b7edc27 323check_objcopy();
f2f8458e 324
8da3821b 325#
8feff1ca 326# Step 1: find all the local (static functions) and weak symbols.
306dcf47 327# 't' is local, 'w/W' is weak
8da3821b
SR
328#
329open (IN, "$nm $inputfile|") || die "error running $nm";
330while (<IN>) {
306dcf47 331 if (/$local_regex/) {
8da3821b 332 $locals{$1} = 1;
306dcf47 333 } elsif (/$weak_regex/) {
8feff1ca 334 $weak{$2} = $1;
8da3821b
SR
335 }
336}
337close(IN);
338
8feff1ca
SR
339my @offsets; # Array of offsets of mcount callers
340my $ref_func; # reference function to use for offsets
341my $offset = 0; # offset of ref_func to section beginning
342
343##
344# update_funcs - print out the current mcount callers
345#
346# Go through the list of offsets to callers and write them to
347# the output file in a format that can be read by an assembler.
348#
349sub update_funcs
350{
351 return if ($#offsets < 0);
352
353 defined($ref_func) || die "No function to reference";
354
355 # A section only had a weak function, to represent it.
356 # Unfortunately, a weak function may be overwritten by another
357 # function of the same name, making all these offsets incorrect.
358 # To be safe, we simply print a warning and bail.
359 if (defined $weak{$ref_func}) {
360 print STDERR
361 "$inputfile: WARNING: referencing weak function" .
362 " $ref_func for mcount\n";
363 return;
364 }
365
366 # is this function static? If so, note this fact.
367 if (defined $locals{$ref_func}) {
f2f8458e
SR
368
369 # only use locals if objcopy supports globalize-symbols
7b7edc27 370 if (!$can_use_local) {
f2f8458e
SR
371 return;
372 }
8feff1ca
SR
373 $convert{$ref_func} = 1;
374 }
375
376 # Loop through all the mcount caller offsets and print a reference
377 # to the caller based from the ref_func.
378 for (my $i=0; $i <= $#offsets; $i++) {
379 if (!$opened) {
380 open(FILE, ">$mcount_s") || die "can't create $mcount_s\n";
381 $opened = 1;
e58918ab 382 print FILE "\t.section $mcount_section,\"a\",$section_type\n";
42e007d0 383 print FILE "\t.align $alignment\n" if (defined($alignment));
8feff1ca
SR
384 }
385 printf FILE "\t%s %s + %d\n", $type, $ref_func, $offsets[$i] - $offset;
386 }
387}
388
8da3821b
SR
389#
390# Step 2: find the sections and mcount call sites
391#
b43f7093 392open(IN, "$objdump -hdr $inputfile|") || die "error running $objdump";
8da3821b 393
8feff1ca
SR
394my $text;
395
db24c7dc
LH
396
397# read headers first
b43f7093
SR
398my $read_headers = 1;
399
8da3821b 400while (<IN>) {
db24c7dc
LH
401
402 if ($read_headers && /$mcount_section/) {
403 #
404 # Somehow the make process can execute this script on an
405 # object twice. If it does, we would duplicate the mcount
406 # section and it will cause the function tracer self test
407 # to fail. Check if the mcount section exists, and if it does,
408 # warn and exit.
409 #
410 print STDERR "ERROR: $mcount_section already in $inputfile\n" .
411 "\tThis may be an indication that your build is corrupted.\n" .
412 "\tDelete $inputfile and try again. If the same object file\n" .
413 "\tstill causes an issue, then disable CONFIG_DYNAMIC_FTRACE.\n";
414 exit(-1);
415 }
416
8da3821b
SR
417 # is it a section?
418 if (/$section_regex/) {
b43f7093 419 $read_headers = 0;
34698bcb
SR
420
421 # Only record text sections that we know are safe
422 if (defined($text_sections{$1})) {
423 $read_function = 1;
424 } else {
425 $read_function = 0;
426 }
8feff1ca 427 # print out any recorded offsets
7dbdee2e 428 update_funcs() if (defined($ref_func));
8feff1ca
SR
429
430 # reset all markers and arrays
8da3821b 431 $text_found = 0;
8feff1ca
SR
432 undef($ref_func);
433 undef(@offsets);
434
8da3821b
SR
435 # section found, now is this a start of a function?
436 } elsif ($read_function && /$function_regex/) {
8da3821b 437 $text_found = 1;
8feff1ca
SR
438 $text = $2;
439
440 # if this is either a local function or a weak function
441 # keep looking for functions that are global that
442 # we can use safely.
443 if (!defined($locals{$text}) && !defined($weak{$text})) {
444 $ref_func = $text;
445 $read_function = 0;
bd171d5f 446 $offset = hex $1;
8feff1ca
SR
447 } else {
448 # if we already have a function, and this is weak, skip it
3f6e968e
SR
449 if (!defined($ref_func) && !defined($weak{$text}) &&
450 # PPC64 can have symbols that start with .L and
451 # gcc considers these special. Don't use them!
452 $text !~ /^\.L/) {
8feff1ca 453 $ref_func = $text;
bd171d5f 454 $offset = hex $1;
8feff1ca 455 }
8da3821b 456 }
8feff1ca 457 }
8feff1ca
SR
458 # is this a call site to mcount? If so, record it to print later
459 if ($text_found && /$mcount_regex/) {
460 $offsets[$#offsets + 1] = hex $1;
8da3821b
SR
461 }
462}
463
8feff1ca 464# dump out anymore offsets that may have been found
7dbdee2e 465update_funcs() if (defined($ref_func));
8feff1ca 466
8da3821b
SR
467# If we did not find any mcount callers, we are done (do nothing).
468if (!$opened) {
469 exit(0);
470}
471
472close(FILE);
473
474#
475# Step 3: Compile the file that holds the list of call sites to mcount.
476#
477`$cc -o $mcount_o -c $mcount_s`;
478
479my @converts = keys %convert;
480
481#
482# Step 4: Do we have sections that started with local functions?
483#
484if ($#converts >= 0) {
485 my $globallist = "";
486 my $locallist = "";
487
488 foreach my $con (@converts) {
489 $globallist .= " --globalize-symbol $con";
490 $locallist .= " --localize-symbol $con";
491 }
492
493 my $globalobj = $dirname . "/.tmp_gl_" . $filename;
494 my $globalmix = $dirname . "/.tmp_mx_" . $filename;
495
496 #
497 # Step 5: set up each local function as a global
498 #
499 `$objcopy $globallist $inputfile $globalobj`;
500
501 #
502 # Step 6: Link the global version to our list.
503 #
504 `$ld -r $globalobj $mcount_o -o $globalmix`;
505
506 #
507 # Step 7: Convert the local functions back into local symbols
508 #
509 `$objcopy $locallist $globalmix $inputfile`;
510
511 # Remove the temp files
512 `$rm $globalobj $globalmix`;
513
514} else {
515
516 my $mix = $dirname . "/.tmp_mx_" . $filename;
517
518 #
519 # Step 8: Link the object with our list of call sites object.
520 #
521 `$ld -r $inputfile $mcount_o -o $mix`;
522
523 #
524 # Step 9: Move the result back to the original object.
525 #
526 `$mv $mix $inputfile`;
527}
528
529# Clean up the temp files
530`$rm $mcount_o $mcount_s`;
531
532exit(0);