version 1 showing authors affecting page license.
.
Rev |
Author |
# |
Line |
1 |
perry |
1 |
OBJDUMP |
|
|
2 |
!!!OBJDUMP |
|
|
3 |
NAME |
|
|
4 |
SYNOPSIS |
|
|
5 |
DESCRIPTION |
|
|
6 |
OPTIONS |
|
|
7 |
SEE ALSO |
|
|
8 |
COPYRIGHT |
|
|
9 |
---- |
|
|
10 |
!!NAME |
|
|
11 |
|
|
|
12 |
|
|
|
13 |
objdump - display information from object files. |
|
|
14 |
!!SYNOPSIS |
|
|
15 |
|
|
|
16 |
|
|
|
17 |
objdump [[__-a--archive-headers__] [[__-b__ |
|
|
18 |
''bfdname''__--target=__''bfdname''] |
|
|
19 |
[[__-C--demangle__[[=''style''] ] |
|
|
20 |
[[__-d--disassemble__] [[__-D--disassemble-all__] |
|
|
21 |
[[__-z--disassemble-zeroes__] [[__-EB-EL--endian=__{big |
|
|
22 |
little }] [[__-f--file-headers__] |
|
|
23 |
[[__--file-start-context__] [[__-g--debugging__] |
|
|
24 |
[[__-h--section-headers--headers__] [[__-i--info__] |
|
|
25 |
[[__-j__ ''section''__--section=__''section''] |
|
|
26 |
[[__-l--line-numbers__] [[__-S--source__] [[__-m__ |
|
|
27 |
''machine''__--architecture=__''machine''] |
|
|
28 |
[[__-M__ |
|
|
29 |
''options''__--disassembler-options=__''options''] |
|
|
30 |
[[__-p--private-headers__] [[__-r--reloc__] |
|
|
31 |
[[__-R--dynamic-reloc__] [[__-s--full-contents__] |
|
|
32 |
[[__-G--stabs__] [[__-t--syms__] |
|
|
33 |
[[__-T--dynamic-syms__] [[__-x--all-headers__] |
|
|
34 |
[[__-w--wide__] [[__--start-address=__''address''] |
|
|
35 |
[[__--stop-address=__''address''] |
|
|
36 |
[[__--prefix-addresses__] [[__--[[no-]show-raw-insn__] |
|
|
37 |
[[__--adjust-vma=__''offset''] [[__-V--version__] |
|
|
38 |
[[__-H--help__] ''objfile''... |
|
|
39 |
!!DESCRIPTION |
|
|
40 |
|
|
|
41 |
|
|
|
42 |
__objdump__ displays information about one or more object |
|
|
43 |
files. The options control what particular information to |
|
|
44 |
display. This information is mostly useful to programmers |
|
|
45 |
who are working on the compilation tools, as opposed to |
|
|
46 |
programmers who just want their program to compile and |
|
|
47 |
work. |
|
|
48 |
|
|
|
49 |
|
|
|
50 |
''objfile''... are the object files to be examined. When |
|
|
51 |
you specify archives, __objdump__ shows information on |
|
|
52 |
each of the member object files. |
|
|
53 |
!!OPTIONS |
|
|
54 |
|
|
|
55 |
|
|
|
56 |
The long and short forms of options, shown here as |
|
|
57 |
alternatives, are equivalent. At least one option from the |
|
|
58 |
list __-a,-d,-D,-f,-g,-G,-h,-H,-p,-r,-R,-S,-t,-T,-V,-x__ |
|
|
59 |
must be given. |
|
|
60 |
|
|
|
61 |
|
|
|
62 |
__-a__ |
|
|
63 |
|
|
|
64 |
|
|
|
65 |
__--archive-header__ |
|
|
66 |
|
|
|
67 |
|
|
|
68 |
If any of the ''objfile'' files are archives, display the |
|
|
69 |
archive header information (in a format similar to __ls |
|
|
70 |
-l__). Besides the information you could list with __ar |
|
|
71 |
tv__, __objdump -a__ shows the object file format of |
|
|
72 |
each archive member. |
|
|
73 |
|
|
|
74 |
|
|
|
75 |
__--adjust-vma=__''offset'' |
|
|
76 |
|
|
|
77 |
|
|
|
78 |
When dumping information, first add ''offset'' to all the |
|
|
79 |
section addresses. This is useful if the section addresses |
|
|
80 |
do not correspond to the symbol table, which can happen when |
|
|
81 |
putting sections at particular addresses when using a format |
|
|
82 |
which can not represent section addresses, such as |
|
|
83 |
a.out. |
|
|
84 |
|
|
|
85 |
|
|
|
86 |
__-b__ ''bfdname'' |
|
|
87 |
|
|
|
88 |
|
|
|
89 |
__--target=__''bfdname'' |
|
|
90 |
|
|
|
91 |
|
|
|
92 |
Specify that the object-code format for the object files is |
|
|
93 |
''bfdname''. This option may not be necessary; |
|
|
94 |
''objdump'' can automatically recognize many |
|
|
95 |
formats. |
|
|
96 |
|
|
|
97 |
|
|
|
98 |
For example, |
|
|
99 |
|
|
|
100 |
|
|
|
101 |
objdump -b oasys -m vax -h fu.o |
|
|
102 |
displays summary information from the section headers (__-h__) of ''fu.o'', which is explicitly identified (__-m__) as a VAX object file in the format produced by Oasys compilers. You can list the formats available with the __-i__ option. |
|
|
103 |
|
|
|
104 |
|
|
|
105 |
__-C__ |
|
|
106 |
|
|
|
107 |
|
|
|
108 |
__--demangle[[=__''style''__]__ |
|
|
109 |
|
|
|
110 |
|
|
|
111 |
Decode (''demangle'') low-level symbol names into |
|
|
112 |
user-level names. Besides removing any initial underscore |
|
|
113 |
prepended by the system, this makes C ++ |
|
|
114 |
function names readable. Different compilers have different |
|
|
115 |
mangling styles. The optional demangling style argument can |
|
|
116 |
be used to choose an appropriate demangling style for your |
|
|
117 |
compiler. |
|
|
118 |
|
|
|
119 |
|
|
|
120 |
__-G__ |
|
|
121 |
|
|
|
122 |
|
|
|
123 |
__--debugging__ |
|
|
124 |
|
|
|
125 |
|
|
|
126 |
Display debugging information. This attempts to parse |
|
|
127 |
debugging information stored in the file and print it out |
|
|
128 |
using a C like syntax. Only certain types of debugging |
|
|
129 |
information have been implemented. |
|
|
130 |
|
|
|
131 |
|
|
|
132 |
__-d__ |
|
|
133 |
|
|
|
134 |
|
|
|
135 |
__--disassemble__ |
|
|
136 |
|
|
|
137 |
|
|
|
138 |
Display the assembler mnemonics for the machine instructions |
|
|
139 |
from ''objfile''. This option only disassembles those |
|
|
140 |
sections which are expected to contain |
|
|
141 |
instructions. |
|
|
142 |
|
|
|
143 |
|
|
|
144 |
__-D__ |
|
|
145 |
|
|
|
146 |
|
|
|
147 |
__--disassemble-all__ |
|
|
148 |
|
|
|
149 |
|
|
|
150 |
Like __-d__, but disassemble the contents of all |
|
|
151 |
sections, not just those expected to contain |
|
|
152 |
instructions. |
|
|
153 |
|
|
|
154 |
|
|
|
155 |
__--prefix-addresses__ |
|
|
156 |
|
|
|
157 |
|
|
|
158 |
When disassembling, print the complete address on each line. |
|
|
159 |
This is the older disassembly format. |
|
|
160 |
|
|
|
161 |
|
|
|
162 |
__--disassemble-zeroes__ |
|
|
163 |
|
|
|
164 |
|
|
|
165 |
Normally the disassembly output will skip blocks of zeroes. |
|
|
166 |
This option directs the disassembler to disassemble those |
|
|
167 |
blocks, just like any other data. |
|
|
168 |
|
|
|
169 |
|
|
|
170 |
__-EB__ |
|
|
171 |
|
|
|
172 |
|
|
|
173 |
__-EL__ |
|
|
174 |
|
|
|
175 |
|
|
|
176 |
__--endian={biglittle}__ |
|
|
177 |
|
|
|
178 |
|
|
|
179 |
Specify the endianness of the object files. This only |
|
|
180 |
affects disassembly. This can be useful when disassembling a |
|
|
181 |
file format which does not describe endianness information, |
|
|
182 |
such as S-records. |
|
|
183 |
|
|
|
184 |
|
|
|
185 |
__-f__ |
|
|
186 |
|
|
|
187 |
|
|
|
188 |
__--file-header__ |
|
|
189 |
|
|
|
190 |
|
|
|
191 |
Display summary information from the overall header of each |
|
|
192 |
of the ''objfile'' files. |
|
|
193 |
|
|
|
194 |
|
|
|
195 |
__--file-start-context__ |
|
|
196 |
|
|
|
197 |
|
|
|
198 |
Specify that when displaying interlisted source |
|
|
199 |
code/disassembly (assumes __-S__) from a file that has |
|
|
200 |
not yet been displayed, extend the context to the start of |
|
|
201 |
the file. |
|
|
202 |
|
|
|
203 |
|
|
|
204 |
__-h__ |
|
|
205 |
|
|
|
206 |
|
|
|
207 |
__--section-header__ |
|
|
208 |
|
|
|
209 |
|
|
|
210 |
__--header__ |
|
|
211 |
|
|
|
212 |
|
|
|
213 |
Display summary information from the section headers of the |
|
|
214 |
object file. |
|
|
215 |
|
|
|
216 |
|
|
|
217 |
File segments may be relocated to nonstandard addresses, for |
|
|
218 |
example by using the __-Ttext__, __-Tdata__, or |
|
|
219 |
__-Tbss__ options to __ld__. However, some object file |
|
|
220 |
formats, such as a.out, do not store the starting address of |
|
|
221 |
the file segments. In those situations, although __ld__ |
|
|
222 |
relocates the sections correctly, using __objdump -h__ to |
|
|
223 |
list the file section headers cannot show the correct |
|
|
224 |
addresses. Instead, it shows the usual addresses, which are |
|
|
225 |
implicit for the target. |
|
|
226 |
|
|
|
227 |
|
|
|
228 |
__--help__ |
|
|
229 |
|
|
|
230 |
|
|
|
231 |
Print a summary of the options to __objdump__ and |
|
|
232 |
exit. |
|
|
233 |
|
|
|
234 |
|
|
|
235 |
__-i__ |
|
|
236 |
|
|
|
237 |
|
|
|
238 |
__--info__ |
|
|
239 |
|
|
|
240 |
|
|
|
241 |
Display a list showing all architectures and object formats |
|
|
242 |
available for specification with __-b__ or |
|
|
243 |
__-m__. |
|
|
244 |
|
|
|
245 |
|
|
|
246 |
__-j__ ''name'' |
|
|
247 |
|
|
|
248 |
|
|
|
249 |
__--section=__''name'' |
|
|
250 |
|
|
|
251 |
|
|
|
252 |
Display information only for section |
|
|
253 |
''name''. |
|
|
254 |
|
|
|
255 |
|
|
|
256 |
__-l__ |
|
|
257 |
|
|
|
258 |
|
|
|
259 |
__--line-numbers__ |
|
|
260 |
|
|
|
261 |
|
|
|
262 |
Label the display (using debugging information) with the |
|
|
263 |
filename and source line numbers corresponding to the object |
|
|
264 |
code or relocs shown. Only useful with __-d__, __-D__, |
|
|
265 |
or __-r__. |
|
|
266 |
|
|
|
267 |
|
|
|
268 |
__-m__ ''machine'' |
|
|
269 |
|
|
|
270 |
|
|
|
271 |
__--architecture=__''machine'' |
|
|
272 |
|
|
|
273 |
|
|
|
274 |
Specify the architecture to use when disassembling object |
|
|
275 |
files. This can be useful when disassembling object files |
|
|
276 |
which do not describe architecture information, such as |
|
|
277 |
S-records. You can list the available architectures with the |
|
|
278 |
__-i__ option. |
|
|
279 |
|
|
|
280 |
|
|
|
281 |
__-M__ ''options'' |
|
|
282 |
|
|
|
283 |
|
|
|
284 |
__--disassembler-options=__''options'' |
|
|
285 |
|
|
|
286 |
|
|
|
287 |
Pass target specific information to the disassembler. Only |
|
|
288 |
supported on some targets. |
|
|
289 |
|
|
|
290 |
|
|
|
291 |
If the target is an ARM architecture then |
|
|
292 |
this switch can be used to select which register name set is |
|
|
293 |
used during disassembler. Specifying __-M reg-name-std__ |
|
|
294 |
(the default) will select the register names as used in |
|
|
295 |
ARM 's instruction set documentation, but |
|
|
296 |
with register 13 called 'sp', register 14 called 'lr' and |
|
|
297 |
register 15 called 'pc'. Specifying __-M reg-names-apcs__ |
|
|
298 |
will select the name set used by the ARM |
|
|
299 |
Procedure Call Standard, whilst specifying __-M |
|
|
300 |
reg-names-raw__ will just use __r__ followed by the |
|
|
301 |
register number. |
|
|
302 |
|
|
|
303 |
|
|
|
304 |
There are also two variants on the APCS |
|
|
305 |
register naming scheme enabled by __-M reg-names-atpcs__ |
|
|
306 |
and __-M reg-names-special-atpcs__ which use the |
|
|
307 |
ARM/Thumb Procedure Call Standard naming conventions. |
|
|
308 |
(Either with the normal register name or the special |
|
|
309 |
register names). |
|
|
310 |
|
|
|
311 |
|
|
|
312 |
This option can also be used for ARM |
|
|
313 |
architectures to force the disassembler to interpret all |
|
|
314 |
instructions as Thumb instructions by using the switch |
|
|
315 |
__--disassembler-options=force-thumb__. This can be |
|
|
316 |
useful when attempting to disassemble thumb code produced by |
|
|
317 |
other compilers. |
|
|
318 |
|
|
|
319 |
|
|
|
320 |
For the x86, some of the options duplicate functions of the |
|
|
321 |
__-m__ switch, but allow finer grained control. Multiple |
|
|
322 |
selections from the following may be specified as a comma |
|
|
323 |
separated string. __x86-64__, __i386__ and |
|
|
324 |
__i8086__ select disassembly for the given architecture. |
|
|
325 |
__intel__ and __att__ select between intel syntax mode |
|
|
326 |
and AT syntax mode. __addr32__, |
|
|
327 |
__addr16__, __data32__ and __data16__ specify the |
|
|
328 |
default address size and operand size. These four options |
|
|
329 |
will be overridden if __x86-64__, __i386__ or |
|
|
330 |
__i8086__ appear later in the option string. Lastly, |
|
|
331 |
__suffix__, when in AT mode, |
|
|
332 |
instructs the dissassembler to print a mnemonic suffix even |
|
|
333 |
when the suffix could be inferred by the |
|
|
334 |
operands. |
|
|
335 |
|
|
|
336 |
|
|
|
337 |
For PPC , __booke__, __booke32__ and |
|
|
338 |
__booke64__ select disassembly of BookE instructions. |
|
|
339 |
__32__ and __64__ select PowerPC and PowerPC64 |
|
|
340 |
disassembly, respectively. |
|
|
341 |
|
|
|
342 |
|
|
|
343 |
__-p__ |
|
|
344 |
|
|
|
345 |
|
|
|
346 |
__--private-headers__ |
|
|
347 |
|
|
|
348 |
|
|
|
349 |
Print information that is specific to the object file |
|
|
350 |
format. The exact information printed depends upon the |
|
|
351 |
object file format. For some object file formats, no |
|
|
352 |
additional information is printed. |
|
|
353 |
|
|
|
354 |
|
|
|
355 |
__-r__ |
|
|
356 |
|
|
|
357 |
|
|
|
358 |
__--reloc__ |
|
|
359 |
|
|
|
360 |
|
|
|
361 |
Print the relocation entries of the file. If used with |
|
|
362 |
__-d__ or __-D__, the relocations are printed |
|
|
363 |
interspersed with the disassembly. |
|
|
364 |
|
|
|
365 |
|
|
|
366 |
__-R__ |
|
|
367 |
|
|
|
368 |
|
|
|
369 |
__--dynamic-reloc__ |
|
|
370 |
|
|
|
371 |
|
|
|
372 |
Print the dynamic relocation entries of the file. This is |
|
|
373 |
only meaningful for dynamic objects, such as certain types |
|
|
374 |
of shared libraries. |
|
|
375 |
|
|
|
376 |
|
|
|
377 |
__-s__ |
|
|
378 |
|
|
|
379 |
|
|
|
380 |
__--full-contents__ |
|
|
381 |
|
|
|
382 |
|
|
|
383 |
Display the full contents of any sections |
|
|
384 |
requested. |
|
|
385 |
|
|
|
386 |
|
|
|
387 |
__-S__ |
|
|
388 |
|
|
|
389 |
|
|
|
390 |
__--source__ |
|
|
391 |
|
|
|
392 |
|
|
|
393 |
Display source code intermixed with disassembly, if |
|
|
394 |
possible. Implies __-d__. |
|
|
395 |
|
|
|
396 |
|
|
|
397 |
__--show-raw-insn__ |
|
|
398 |
|
|
|
399 |
|
|
|
400 |
When disassembling instructions, print the instruction in |
|
|
401 |
hex as well as in symbolic form. This is the default except |
|
|
402 |
when __--prefix-addresses__ is used. |
|
|
403 |
|
|
|
404 |
|
|
|
405 |
__--no-show-raw-insn__ |
|
|
406 |
|
|
|
407 |
|
|
|
408 |
When disassembling instructions, do not print the |
|
|
409 |
instruction bytes. This is the default when |
|
|
410 |
__--prefix-addresses__ is used. |
|
|
411 |
|
|
|
412 |
|
|
|
413 |
__-G__ |
|
|
414 |
|
|
|
415 |
|
|
|
416 |
__--stabs__ |
|
|
417 |
|
|
|
418 |
|
|
|
419 |
Display the full contents of any sections requested. Display |
|
|
420 |
the contents of the .stab and .stab.index and .stab.excl |
|
|
421 |
sections from an ELF file. This is only |
|
|
422 |
useful on systems (such as Solaris 2.0) in which |
|
|
423 |
.stab debugging symbol-table entries are carried in |
|
|
424 |
an ELF section. In most other file formats, |
|
|
425 |
debugging symbol-table entries are interleaved with linkage |
|
|
426 |
symbols, and are visible in the __--syms__ |
|
|
427 |
output. |
|
|
428 |
|
|
|
429 |
|
|
|
430 |
__--start-address=__''address'' |
|
|
431 |
|
|
|
432 |
|
|
|
433 |
Start displaying data at the specified address. This affects |
|
|
434 |
the output of the __-d__, __-r__ and __-s__ |
|
|
435 |
options. |
|
|
436 |
|
|
|
437 |
|
|
|
438 |
__--stop-address=__''address'' |
|
|
439 |
|
|
|
440 |
|
|
|
441 |
Stop displaying data at the specified address. This affects |
|
|
442 |
the output of the __-d__, __-r__ and __-s__ |
|
|
443 |
options. |
|
|
444 |
|
|
|
445 |
|
|
|
446 |
__-t__ |
|
|
447 |
|
|
|
448 |
|
|
|
449 |
__--syms__ |
|
|
450 |
|
|
|
451 |
|
|
|
452 |
Print the symbol table entries of the file. This is similar |
|
|
453 |
to the information provided by the __nm__ |
|
|
454 |
program. |
|
|
455 |
|
|
|
456 |
|
|
|
457 |
__-T__ |
|
|
458 |
|
|
|
459 |
|
|
|
460 |
__--dynamic-syms__ |
|
|
461 |
|
|
|
462 |
|
|
|
463 |
Print the dynamic symbol table entries of the file. This is |
|
|
464 |
only meaningful for dynamic objects, such as certain types |
|
|
465 |
of shared libraries. This is similar to the information |
|
|
466 |
provided by the __nm__ program when given the __-D__ |
|
|
467 |
(__--dynamic__) option. |
|
|
468 |
|
|
|
469 |
|
|
|
470 |
__--version__ |
|
|
471 |
|
|
|
472 |
|
|
|
473 |
Print the version number of __objdump__ and |
|
|
474 |
exit. |
|
|
475 |
|
|
|
476 |
|
|
|
477 |
__-x__ |
|
|
478 |
|
|
|
479 |
|
|
|
480 |
__--all-header__ |
|
|
481 |
|
|
|
482 |
|
|
|
483 |
Display all available header information, including the |
|
|
484 |
symbol table and relocation entries. Using __-x__ is |
|
|
485 |
equivalent to specifying all of __-a -f -h -r |
|
|
486 |
-t__. |
|
|
487 |
|
|
|
488 |
|
|
|
489 |
__-w__ |
|
|
490 |
|
|
|
491 |
|
|
|
492 |
__--wide__ |
|
|
493 |
|
|
|
494 |
|
|
|
495 |
Format some lines for output devices that have more than 80 |
|
|
496 |
columns. Also do not truncate symbol names when they are |
|
|
497 |
displayed. |
|
|
498 |
!!SEE ALSO |
|
|
499 |
|
|
|
500 |
|
|
|
501 |
nm(1), readelf(1), and the Info entries for |
|
|
502 |
''binutils''. |
|
|
503 |
!!COPYRIGHT |
|
|
504 |
|
|
|
505 |
|
|
|
506 |
Copyright (c) 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, |
|
|
507 |
2001, 2002 Free Software Foundation, Inc. |
|
|
508 |
|
|
|
509 |
|
|
|
510 |
Permission is granted to copy, distribute and/or modify this |
|
|
511 |
document under the terms of the GNU Free |
|
|
512 |
Documentation License, Version 1.1 or any later version |
|
|
513 |
published by the Free Software Foundation; with no Invariant |
|
|
514 |
Sections, with no Front-Cover Texts, and with no Back-Cover |
|
|
515 |
Texts. A copy of the license is included in the section |
|
|
516 |
entitled `` GNU Free Documentation |
|
|
517 |
License''. |
|
|
518 |
---- |