OpenCores
URL https://opencores.org/ocsvn/open8_urisc/open8_urisc/trunk

Subversion Repositories open8_urisc

[/] [open8_urisc/] [trunk/] [gnu/] [binutils/] [ld/] [ld.texinfo] - Blame information for rev 285

Go to most recent revision | Details | Compare with Previous | View Log

Line No. Rev Author Line
1 145 khays
\input texinfo
2
@setfilename ld.info
3
@c Copyright 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
4 166 khays
@c 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012
5 145 khays
@c Free Software Foundation, Inc.
6
@syncodeindex ky cp
7
@c man begin INCLUDE
8
@include configdoc.texi
9
@c (configdoc.texi is generated by the Makefile)
10
@include bfdver.texi
11
@c man end
12
 
13
@c @smallbook
14
 
15
@macro gcctabopt{body}
16
@code{\body\}
17
@end macro
18
 
19
@c man begin NAME
20
@ifset man
21
@c Configure for the generation of man pages
22
@set UsesEnvVars
23
@set GENERIC
24
@set ARM
25
@set C6X
26
@set H8300
27
@set HPPA
28
@set I960
29
@set M68HC11
30
@set M68K
31
@set MMIX
32
@set MSP430
33
@set POWERPC
34
@set POWERPC64
35
@set Renesas
36
@set SPU
37
@set TICOFF
38
@set WIN32
39
@set XTENSA
40
@end ifset
41
@c man end
42
 
43
@ifnottex
44
@dircategory Software development
45
@direntry
46
* Ld: (ld).                       The GNU linker.
47
@end direntry
48
@end ifnottex
49
 
50
@copying
51
This file documents the @sc{gnu} linker LD
52
@ifset VERSION_PACKAGE
53
@value{VERSION_PACKAGE}
54
@end ifset
55
version @value{VERSION}.
56
 
57
Copyright @copyright{} 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
58
2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 Free Software Foundation, Inc.
59
 
60
Permission is granted to copy, distribute and/or modify this document
61
under the terms of the GNU Free Documentation License, Version 1.3
62
or any later version published by the Free Software Foundation;
63
with no Invariant Sections, with no Front-Cover Texts, and with no
64
Back-Cover Texts.  A copy of the license is included in the
65
section entitled ``GNU Free Documentation License''.
66
@end copying
67
@iftex
68
@finalout
69
@setchapternewpage odd
70
@settitle The GNU linker
71
@titlepage
72
@title The GNU linker
73
@sp 1
74
@subtitle @code{ld}
75
@ifset VERSION_PACKAGE
76
@subtitle @value{VERSION_PACKAGE}
77
@end ifset
78
@subtitle Version @value{VERSION}
79
@author Steve Chamberlain
80
@author Ian Lance Taylor
81
@page
82
 
83
@tex
84
{\parskip=0pt
85
\hfill Red Hat Inc\par
86
\hfill nickc\@credhat.com, doc\@redhat.com\par
87
\hfill {\it The GNU linker}\par
88
\hfill Edited by Jeffrey Osier (jeffrey\@cygnus.com)\par
89
}
90
\global\parindent=0pt % Steve likes it this way.
91
@end tex
92
 
93
@vskip 0pt plus 1filll
94
@c man begin COPYRIGHT
95
Copyright @copyright{} 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998,
96
1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 Free
97
Software Foundation, Inc.
98
 
99
Permission is granted to copy, distribute and/or modify this document
100
under the terms of the GNU Free Documentation License, Version 1.3
101
or any later version published by the Free Software Foundation;
102
with no Invariant Sections, with no Front-Cover Texts, and with no
103
Back-Cover Texts.  A copy of the license is included in the
104
section entitled ``GNU Free Documentation License''.
105
@c man end
106
 
107
@end titlepage
108
@end iftex
109
@contents
110
@c FIXME: Talk about importance of *order* of args, cmds to linker!
111
 
112
@ifnottex
113
@node Top
114
@top LD
115
This file documents the @sc{gnu} linker ld
116
@ifset VERSION_PACKAGE
117
@value{VERSION_PACKAGE}
118
@end ifset
119
version @value{VERSION}.
120
 
121
This document is distributed under the terms of the GNU Free
122
Documentation License version 1.3.  A copy of the license is included
123
in the section entitled ``GNU Free Documentation License''.
124
 
125
@menu
126
* Overview::                    Overview
127
* Invocation::                  Invocation
128
* Scripts::                     Linker Scripts
129
@ifset GENERIC
130
* Machine Dependent::           Machine Dependent Features
131
@end ifset
132
@ifclear GENERIC
133
@ifset H8300
134
* H8/300::                      ld and the H8/300
135
@end ifset
136
@ifset Renesas
137
* Renesas::                     ld and other Renesas micros
138
@end ifset
139
@ifset I960
140
* i960::                        ld and the Intel 960 family
141
@end ifset
142
@ifset ARM
143
* ARM::                         ld and the ARM family
144
@end ifset
145
@ifset HPPA
146
* HPPA ELF32::                  ld and HPPA 32-bit ELF
147
@end ifset
148
@ifset M68HC11
149
* M68HC11/68HC12::              ld and the Motorola 68HC11 and 68HC12 families
150
@end ifset
151
@ifset M68K
152
* M68K::                        ld and Motorola 68K family
153
@end ifset
154
@ifset POWERPC
155
* PowerPC ELF32::               ld and PowerPC 32-bit ELF Support
156
@end ifset
157
@ifset POWERPC64
158
* PowerPC64 ELF64::             ld and PowerPC64 64-bit ELF Support
159
@end ifset
160
@ifset SPU
161
* SPU ELF::                     ld and SPU ELF Support
162
@end ifset
163
@ifset TICOFF
164
* TI COFF::                     ld and the TI COFF
165
@end ifset
166
@ifset WIN32
167
* Win32::                       ld and WIN32 (cygwin/mingw)
168
@end ifset
169
@ifset XTENSA
170
* Xtensa::                      ld and Xtensa Processors
171
@end ifset
172
@end ifclear
173
@ifclear SingleFormat
174
* BFD::                         BFD
175
@end ifclear
176
@c Following blank line required for remaining bug in makeinfo conds/menus
177
 
178
* Reporting Bugs::              Reporting Bugs
179
* MRI::                         MRI Compatible Script Files
180
* GNU Free Documentation License::  GNU Free Documentation License
181
* LD Index::                       LD Index
182
@end menu
183
@end ifnottex
184
 
185
@node Overview
186
@chapter Overview
187
 
188
@cindex @sc{gnu} linker
189
@cindex what is this?
190
 
191
@ifset man
192
@c man begin SYNOPSIS
193
ld [@b{options}] @var{objfile} @dots{}
194
@c man end
195
 
196
@c man begin SEEALSO
197
ar(1), nm(1), objcopy(1), objdump(1), readelf(1) and
198
the Info entries for @file{binutils} and
199
@file{ld}.
200
@c man end
201
@end ifset
202
 
203
@c man begin DESCRIPTION
204
 
205
@command{ld} combines a number of object and archive files, relocates
206
their data and ties up symbol references. Usually the last step in
207
compiling a program is to run @command{ld}.
208
 
209
@command{ld} accepts Linker Command Language files written in
210
a superset of AT&T's Link Editor Command Language syntax,
211
to provide explicit and total control over the linking process.
212
 
213
@ifset man
214
@c For the man only
215
This man page does not describe the command language; see the
216
@command{ld} entry in @code{info} for full details on the command
217
language and on other aspects of the GNU linker.
218
@end ifset
219
 
220
@ifclear SingleFormat
221
This version of @command{ld} uses the general purpose BFD libraries
222
to operate on object files. This allows @command{ld} to read, combine, and
223
write object files in many different formats---for example, COFF or
224
@code{a.out}.  Different formats may be linked together to produce any
225
available kind of object file.  @xref{BFD}, for more information.
226
@end ifclear
227
 
228
Aside from its flexibility, the @sc{gnu} linker is more helpful than other
229
linkers in providing diagnostic information.  Many linkers abandon
230
execution immediately upon encountering an error; whenever possible,
231
@command{ld} continues executing, allowing you to identify other errors
232
(or, in some cases, to get an output file in spite of the error).
233
 
234
@c man end
235
 
236
@node Invocation
237
@chapter Invocation
238
 
239
@c man begin DESCRIPTION
240
 
241
The @sc{gnu} linker @command{ld} is meant to cover a broad range of situations,
242
and to be as compatible as possible with other linkers.  As a result,
243
you have many choices to control its behavior.
244
 
245
@c man end
246
 
247
@ifset UsesEnvVars
248
@menu
249
* Options::                     Command Line Options
250
* Environment::                 Environment Variables
251
@end menu
252
 
253
@node Options
254
@section Command Line Options
255
@end ifset
256
 
257
@cindex command line
258
@cindex options
259
 
260
@c man begin OPTIONS
261
 
262
The linker supports a plethora of command-line options, but in actual
263
practice few of them are used in any particular context.
264
@cindex standard Unix system
265
For instance, a frequent use of @command{ld} is to link standard Unix
266
object files on a standard, supported Unix system.  On such a system, to
267
link a file @code{hello.o}:
268
 
269
@smallexample
270
ld -o @var{output} /lib/crt0.o hello.o -lc
271
@end smallexample
272
 
273
This tells @command{ld} to produce a file called @var{output} as the
274
result of linking the file @code{/lib/crt0.o} with @code{hello.o} and
275
the library @code{libc.a}, which will come from the standard search
276
directories.  (See the discussion of the @samp{-l} option below.)
277
 
278
Some of the command-line options to @command{ld} may be specified at any
279
point in the command line.  However, options which refer to files, such
280
as @samp{-l} or @samp{-T}, cause the file to be read at the point at
281
which the option appears in the command line, relative to the object
282
files and other file options.  Repeating non-file options with a
283
different argument will either have no further effect, or override prior
284
occurrences (those further to the left on the command line) of that
285
option.  Options which may be meaningfully specified more than once are
286
noted in the descriptions below.
287
 
288
@cindex object files
289
Non-option arguments are object files or archives which are to be linked
290
together.  They may follow, precede, or be mixed in with command-line
291
options, except that an object file argument may not be placed between
292
an option and its argument.
293
 
294
Usually the linker is invoked with at least one object file, but you can
295
specify other forms of binary input files using @samp{-l}, @samp{-R},
296
and the script command language.  If @emph{no} binary input files at all
297
are specified, the linker does not produce any output, and issues the
298
message @samp{No input files}.
299
 
300
If the linker cannot recognize the format of an object file, it will
301
assume that it is a linker script.  A script specified in this way
302
augments the main linker script used for the link (either the default
303
linker script or the one specified by using @samp{-T}).  This feature
304
permits the linker to link against a file which appears to be an object
305
or an archive, but actually merely defines some symbol values, or uses
306
@code{INPUT} or @code{GROUP} to load other objects.  Specifying a
307
script in this way merely augments the main linker script, with the
308
extra commands placed after the main script; use the @samp{-T} option
309
to replace the default linker script entirely, but note the effect of
310
the @code{INSERT} command.  @xref{Scripts}.
311
 
312
For options whose names are a single letter,
313
option arguments must either follow the option letter without intervening
314
whitespace, or be given as separate arguments immediately following the
315
option that requires them.
316
 
317
For options whose names are multiple letters, either one dash or two can
318
precede the option name; for example, @samp{-trace-symbol} and
319
@samp{--trace-symbol} are equivalent.  Note---there is one exception to
320
this rule.  Multiple letter options that start with a lower case 'o' can
321
only be preceded by two dashes.  This is to reduce confusion with the
322
@samp{-o} option.  So for example @samp{-omagic} sets the output file
323
name to @samp{magic} whereas @samp{--omagic} sets the NMAGIC flag on the
324
output.
325
 
326
Arguments to multiple-letter options must either be separated from the
327
option name by an equals sign, or be given as separate arguments
328
immediately following the option that requires them.  For example,
329
@samp{--trace-symbol foo} and @samp{--trace-symbol=foo} are equivalent.
330
Unique abbreviations of the names of multiple-letter options are
331
accepted.
332
 
333
Note---if the linker is being invoked indirectly, via a compiler driver
334
(e.g. @samp{gcc}) then all the linker command line options should be
335
prefixed by @samp{-Wl,} (or whatever is appropriate for the particular
336
compiler driver) like this:
337
 
338
@smallexample
339
  gcc -Wl,--start-group foo.o bar.o -Wl,--end-group
340
@end smallexample
341
 
342
This is important, because otherwise the compiler driver program may
343
silently drop the linker options, resulting in a bad link.  Confusion
344
may also arise when passing options that require values through a
345
driver, as the use of a space between option and argument acts as
346
a separator, and causes the driver to pass only the option to the linker
347
and the argument to the compiler.  In this case, it is simplest to use
348
the joined forms of both single- and multiple-letter options, such as:
349
 
350
@smallexample
351
  gcc foo.o bar.o -Wl,-eENTRY -Wl,-Map=a.map
352
@end smallexample
353
 
354
Here is a table of the generic command line switches accepted by the GNU
355
linker:
356
 
357
@table @gcctabopt
358
@include at-file.texi
359
 
360
@kindex -a @var{keyword}
361
@item -a @var{keyword}
362
This option is supported for HP/UX compatibility.  The @var{keyword}
363
argument must be one of the strings @samp{archive}, @samp{shared}, or
364
@samp{default}.  @samp{-aarchive} is functionally equivalent to
365
@samp{-Bstatic}, and the other two keywords are functionally equivalent
366
to @samp{-Bdynamic}.  This option may be used any number of times.
367
 
368
@kindex --audit @var{AUDITLIB}
369
@item --audit @var{AUDITLIB}
370
Adds @var{AUDITLIB} to the @code{DT_AUDIT} entry of the dynamic section.
371
@var{AUDITLIB} is not checked for existence, nor will it use the DT_SONAME
372
specified in the library.  If specified multiple times @code{DT_AUDIT}
373
will contain a colon separated list of audit interfaces to use. If the linker
374
finds an object with an audit entry while searching for shared libraries,
375
it will add a corresponding @code{DT_DEPAUDIT} entry in the output file.
376
This option is only meaningful on ELF platforms supporting the rtld-audit
377
interface.
378
 
379
@ifset I960
380
@cindex architectures
381
@kindex -A @var{arch}
382
@item -A @var{architecture}
383
@kindex --architecture=@var{arch}
384
@itemx --architecture=@var{architecture}
385
In the current release of @command{ld}, this option is useful only for the
386
Intel 960 family of architectures.  In that @command{ld} configuration, the
387
@var{architecture} argument identifies the particular architecture in
388
the 960 family, enabling some safeguards and modifying the
389
archive-library search path.  @xref{i960,,@command{ld} and the Intel 960
390
family}, for details.
391
 
392
Future releases of @command{ld} may support similar functionality for
393
other architecture families.
394
@end ifset
395
 
396
@ifclear SingleFormat
397
@cindex binary input format
398
@kindex -b @var{format}
399
@kindex --format=@var{format}
400
@cindex input format
401
@cindex input format
402
@item -b @var{input-format}
403
@itemx --format=@var{input-format}
404
@command{ld} may be configured to support more than one kind of object
405
file.  If your @command{ld} is configured this way, you can use the
406
@samp{-b} option to specify the binary format for input object files
407
that follow this option on the command line.  Even when @command{ld} is
408
configured to support alternative object formats, you don't usually need
409
to specify this, as @command{ld} should be configured to expect as a
410
default input format the most usual format on each machine.
411
@var{input-format} is a text string, the name of a particular format
412
supported by the BFD libraries.  (You can list the available binary
413
formats with @samp{objdump -i}.)
414
@xref{BFD}.
415
 
416
You may want to use this option if you are linking files with an unusual
417
binary format.  You can also use @samp{-b} to switch formats explicitly (when
418
linking object files of different formats), by including
419
@samp{-b @var{input-format}} before each group of object files in a
420
particular format.
421
 
422
The default format is taken from the environment variable
423
@code{GNUTARGET}.
424
@ifset UsesEnvVars
425
@xref{Environment}.
426
@end ifset
427
You can also define the input format from a script, using the command
428
@code{TARGET};
429
@ifclear man
430
see @ref{Format Commands}.
431
@end ifclear
432
@end ifclear
433
 
434
@kindex -c @var{MRI-cmdfile}
435
@kindex --mri-script=@var{MRI-cmdfile}
436
@cindex compatibility, MRI
437
@item -c @var{MRI-commandfile}
438
@itemx --mri-script=@var{MRI-commandfile}
439
For compatibility with linkers produced by MRI, @command{ld} accepts script
440
files written in an alternate, restricted command language, described in
441
@ifclear man
442
@ref{MRI,,MRI Compatible Script Files}.
443
@end ifclear
444
@ifset man
445
the MRI Compatible Script Files section of GNU ld documentation.
446
@end ifset
447
Introduce MRI script files with
448
the option @samp{-c}; use the @samp{-T} option to run linker
449
scripts written in the general-purpose @command{ld} scripting language.
450
If @var{MRI-cmdfile} does not exist, @command{ld} looks for it in the directories
451
specified by any @samp{-L} options.
452
 
453
@cindex common allocation
454
@kindex -d
455
@kindex -dc
456
@kindex -dp
457
@item -d
458
@itemx -dc
459
@itemx -dp
460
These three options are equivalent; multiple forms are supported for
461
compatibility with other linkers.  They assign space to common symbols
462
even if a relocatable output file is specified (with @samp{-r}).  The
463
script command @code{FORCE_COMMON_ALLOCATION} has the same effect.
464
@xref{Miscellaneous Commands}.
465
 
466
@kindex --depaudit @var{AUDITLIB}
467
@kindex -P @var{AUDITLIB}
468
@item --depaudit @var{AUDITLIB}
469
@itemx -P @var{AUDITLIB}
470
Adds @var{AUDITLIB} to the @code{DT_DEPAUDIT} entry of the dynamic section.
471
@var{AUDITLIB} is not checked for existence, nor will it use the DT_SONAME
472
specified in the library.  If specified multiple times @code{DT_DEPAUDIT}
473
will contain a colon separated list of audit interfaces to use.  This
474
option is only meaningful on ELF platforms supporting the rtld-audit interface.
475
The -P option is provided for Solaris compatibility.
476
 
477
@cindex entry point, from command line
478
@kindex -e @var{entry}
479
@kindex --entry=@var{entry}
480
@item -e @var{entry}
481
@itemx --entry=@var{entry}
482
Use @var{entry} as the explicit symbol for beginning execution of your
483
program, rather than the default entry point.  If there is no symbol
484
named @var{entry}, the linker will try to parse @var{entry} as a number,
485
and use that as the entry address (the number will be interpreted in
486
base 10; you may use a leading @samp{0x} for base 16, or a leading
487
@samp{0} for base 8).  @xref{Entry Point}, for a discussion of defaults
488
and other ways of specifying the entry point.
489
 
490
@kindex --exclude-libs
491
@item --exclude-libs @var{lib},@var{lib},...
492
Specifies a list of archive libraries from which symbols should not be automatically
493
exported.  The library names may be delimited by commas or colons.  Specifying
494
@code{--exclude-libs ALL} excludes symbols in all archive libraries from
495
automatic export.  This option is available only for the i386 PE targeted
496
port of the linker and for ELF targeted ports.  For i386 PE, symbols
497
explicitly listed in a .def file are still exported, regardless of this
498
option.  For ELF targeted ports, symbols affected by this option will
499
be treated as hidden.
500
 
501
@kindex --exclude-modules-for-implib
502
@item --exclude-modules-for-implib @var{module},@var{module},...
503
Specifies a list of object files or archive members, from which symbols
504
should not be automatically exported, but which should be copied wholesale
505
into the import library being generated during the link.  The module names
506
may be delimited by commas or colons, and must match exactly the filenames
507
used by @command{ld} to open the files; for archive members, this is simply
508
the member name, but for object files the name listed must include and
509
match precisely any path used to specify the input file on the linker's
510
command-line.  This option is available only for the i386 PE targeted port
511
of the linker.  Symbols explicitly listed in a .def file are still exported,
512
regardless of this option.
513
 
514
@cindex dynamic symbol table
515
@kindex -E
516
@kindex --export-dynamic
517
@kindex --no-export-dynamic
518
@item -E
519
@itemx --export-dynamic
520
@itemx --no-export-dynamic
521
When creating a dynamically linked executable, using the @option{-E}
522
option or the @option{--export-dynamic} option causes the linker to add
523
all symbols to the dynamic symbol table.  The dynamic symbol table is the
524
set of symbols which are visible from dynamic objects at run time.
525
 
526
If you do not use either of these options (or use the
527
@option{--no-export-dynamic} option to restore the default behavior), the
528
dynamic symbol table will normally contain only those symbols which are
529
referenced by some dynamic object mentioned in the link.
530
 
531
If you use @code{dlopen} to load a dynamic object which needs to refer
532
back to the symbols defined by the program, rather than some other
533
dynamic object, then you will probably need to use this option when
534
linking the program itself.
535
 
536
You can also use the dynamic list to control what symbols should
537
be added to the dynamic symbol table if the output format supports it.
538
See the description of @samp{--dynamic-list}.
539
 
540
Note that this option is specific to ELF targeted ports.  PE targets
541
support a similar function to export all symbols from a DLL or EXE; see
542
the description of @samp{--export-all-symbols} below.
543
 
544
@ifclear SingleFormat
545
@cindex big-endian objects
546
@cindex endianness
547
@kindex -EB
548
@item -EB
549
Link big-endian objects.  This affects the default output format.
550
 
551
@cindex little-endian objects
552
@kindex -EL
553
@item -EL
554
Link little-endian objects.  This affects the default output format.
555
@end ifclear
556
 
557
@kindex -f @var{name}
558
@kindex --auxiliary=@var{name}
559
@item -f @var{name}
560
@itemx --auxiliary=@var{name}
561
When creating an ELF shared object, set the internal DT_AUXILIARY field
562
to the specified name.  This tells the dynamic linker that the symbol
563
table of the shared object should be used as an auxiliary filter on the
564
symbol table of the shared object @var{name}.
565
 
566
If you later link a program against this filter object, then, when you
567
run the program, the dynamic linker will see the DT_AUXILIARY field.  If
568
the dynamic linker resolves any symbols from the filter object, it will
569
first check whether there is a definition in the shared object
570
@var{name}.  If there is one, it will be used instead of the definition
571
in the filter object.  The shared object @var{name} need not exist.
572
Thus the shared object @var{name} may be used to provide an alternative
573
implementation of certain functions, perhaps for debugging or for
574
machine specific performance.
575
 
576
This option may be specified more than once.  The DT_AUXILIARY entries
577
will be created in the order in which they appear on the command line.
578
 
579
@kindex -F @var{name}
580
@kindex --filter=@var{name}
581
@item -F @var{name}
582
@itemx --filter=@var{name}
583
When creating an ELF shared object, set the internal DT_FILTER field to
584
the specified name.  This tells the dynamic linker that the symbol table
585
of the shared object which is being created should be used as a filter
586
on the symbol table of the shared object @var{name}.
587
 
588
If you later link a program against this filter object, then, when you
589
run the program, the dynamic linker will see the DT_FILTER field.  The
590
dynamic linker will resolve symbols according to the symbol table of the
591
filter object as usual, but it will actually link to the definitions
592
found in the shared object @var{name}.  Thus the filter object can be
593
used to select a subset of the symbols provided by the object
594
@var{name}.
595
 
596
Some older linkers used the @option{-F} option throughout a compilation
597
toolchain for specifying object-file format for both input and output
598
object files.
599
@ifclear SingleFormat
600
The @sc{gnu} linker uses other mechanisms for this purpose: the
601
@option{-b}, @option{--format}, @option{--oformat} options, the
602
@code{TARGET} command in linker scripts, and the @code{GNUTARGET}
603
environment variable.
604
@end ifclear
605
The @sc{gnu} linker will ignore the @option{-F} option when not
606
creating an ELF shared object.
607
 
608
@cindex finalization function
609
@kindex -fini=@var{name}
610
@item -fini=@var{name}
611
When creating an ELF executable or shared object, call NAME when the
612
executable or shared object is unloaded, by setting DT_FINI to the
613
address of the function.  By default, the linker uses @code{_fini} as
614
the function to call.
615
 
616
@kindex -g
617
@item -g
618
Ignored.  Provided for compatibility with other tools.
619
 
620
@kindex -G @var{value}
621
@kindex --gpsize=@var{value}
622
@cindex object size
623
@item -G @var{value}
624
@itemx --gpsize=@var{value}
625
Set the maximum size of objects to be optimized using the GP register to
626
@var{size}.  This is only meaningful for object file formats such as
627
MIPS ECOFF which supports putting large and small objects into different
628
sections.  This is ignored for other object file formats.
629
 
630
@cindex runtime library name
631
@kindex -h @var{name}
632
@kindex -soname=@var{name}
633
@item -h @var{name}
634
@itemx -soname=@var{name}
635
When creating an ELF shared object, set the internal DT_SONAME field to
636
the specified name.  When an executable is linked with a shared object
637
which has a DT_SONAME field, then when the executable is run the dynamic
638
linker will attempt to load the shared object specified by the DT_SONAME
639
field rather than the using the file name given to the linker.
640
 
641
@kindex -i
642
@cindex incremental link
643
@item -i
644
Perform an incremental link (same as option @samp{-r}).
645
 
646
@cindex initialization function
647
@kindex -init=@var{name}
648
@item -init=@var{name}
649
When creating an ELF executable or shared object, call NAME when the
650
executable or shared object is loaded, by setting DT_INIT to the address
651
of the function.  By default, the linker uses @code{_init} as the
652
function to call.
653
 
654
@cindex archive files, from cmd line
655
@kindex -l @var{namespec}
656
@kindex --library=@var{namespec}
657
@item -l @var{namespec}
658
@itemx --library=@var{namespec}
659
Add the archive or object file specified by @var{namespec} to the
660
list of files to link.  This option may be used any number of times.
661
If @var{namespec} is of the form @file{:@var{filename}}, @command{ld}
662
will search the library path for a file called @var{filename}, otherwise it
663
will search the library path for a file called @file{lib@var{namespec}.a}.
664
 
665
On systems which support shared libraries, @command{ld} may also search for
666
files other than @file{lib@var{namespec}.a}.  Specifically, on ELF
667
and SunOS systems, @command{ld} will search a directory for a library
668
called @file{lib@var{namespec}.so} before searching for one called
669
@file{lib@var{namespec}.a}.  (By convention, a @code{.so} extension
670
indicates a shared library.)  Note that this behavior does not apply
671
to @file{:@var{filename}}, which always specifies a file called
672
@var{filename}.
673
 
674
The linker will search an archive only once, at the location where it is
675
specified on the command line.  If the archive defines a symbol which
676
was undefined in some object which appeared before the archive on the
677
command line, the linker will include the appropriate file(s) from the
678
archive.  However, an undefined symbol in an object appearing later on
679
the command line will not cause the linker to search the archive again.
680
 
681
See the @option{-(} option for a way to force the linker to search
682
archives multiple times.
683
 
684
You may list the same archive multiple times on the command line.
685
 
686
@ifset GENERIC
687
This type of archive searching is standard for Unix linkers.  However,
688
if you are using @command{ld} on AIX, note that it is different from the
689
behaviour of the AIX linker.
690
@end ifset
691
 
692
@cindex search directory, from cmd line
693
@kindex -L @var{dir}
694
@kindex --library-path=@var{dir}
695
@item -L @var{searchdir}
696
@itemx --library-path=@var{searchdir}
697
Add path @var{searchdir} to the list of paths that @command{ld} will search
698
for archive libraries and @command{ld} control scripts.  You may use this
699
option any number of times.  The directories are searched in the order
700
in which they are specified on the command line.  Directories specified
701
on the command line are searched before the default directories.  All
702
@option{-L} options apply to all @option{-l} options, regardless of the
703
order in which the options appear.  @option{-L} options do not affect
704
how @command{ld} searches for a linker script unless @option{-T}
705
option is specified.
706
 
707
If @var{searchdir} begins with @code{=}, then the @code{=} will be replaced
708
by the @dfn{sysroot prefix}, a path specified when the linker is configured.
709
 
710
@ifset UsesEnvVars
711
The default set of paths searched (without being specified with
712
@samp{-L}) depends on which emulation mode @command{ld} is using, and in
713
some cases also on how it was configured.  @xref{Environment}.
714
@end ifset
715
 
716
The paths can also be specified in a link script with the
717
@code{SEARCH_DIR} command.  Directories specified this way are searched
718
at the point in which the linker script appears in the command line.
719
 
720
@cindex emulation
721
@kindex -m @var{emulation}
722
@item -m @var{emulation}
723
Emulate the @var{emulation} linker.  You can list the available
724
emulations with the @samp{--verbose} or @samp{-V} options.
725
 
726
If the @samp{-m} option is not used, the emulation is taken from the
727
@code{LDEMULATION} environment variable, if that is defined.
728
 
729
Otherwise, the default emulation depends upon how the linker was
730
configured.
731
 
732
@cindex link map
733
@kindex -M
734
@kindex --print-map
735
@item -M
736
@itemx --print-map
737
Print a link map to the standard output.  A link map provides
738
information about the link, including the following:
739
 
740
@itemize @bullet
741
@item
742
Where object files are mapped into memory.
743
@item
744
How common symbols are allocated.
745
@item
746
All archive members included in the link, with a mention of the symbol
747
which caused the archive member to be brought in.
748
@item
749
The values assigned to symbols.
750
 
751
Note - symbols whose values are computed by an expression which
752
involves a reference to a previous value of the same symbol may not
753
have correct result displayed in the link map.  This is because the
754
linker discards intermediate results and only retains the final value
755
of an expression.  Under such circumstances the linker will display
756
the final value enclosed by square brackets.  Thus for example a
757
linker script containing:
758
 
759
@smallexample
760
   foo = 1
761
   foo = foo * 4
762
   foo = foo + 8
763
@end smallexample
764
 
765
will produce the following output in the link map if the @option{-M}
766
option is used:
767
 
768
@smallexample
769
   0x00000001                foo = 0x1
770
   [0x0000000c]                foo = (foo * 0x4)
771
   [0x0000000c]                foo = (foo + 0x8)
772
@end smallexample
773
 
774
See @ref{Expressions} for more information about expressions in linker
775
scripts.
776
@end itemize
777
 
778
@kindex -n
779
@cindex read-only text
780
@cindex NMAGIC
781
@kindex --nmagic
782
@item -n
783
@itemx --nmagic
784
Turn off page alignment of sections, and disable linking against shared
785
libraries.  If the output format supports Unix style magic numbers,
786
mark the output as @code{NMAGIC}.
787
 
788
@kindex -N
789
@kindex --omagic
790
@cindex read/write from cmd line
791
@cindex OMAGIC
792
@item -N
793
@itemx --omagic
794
Set the text and data sections to be readable and writable.  Also, do
795
not page-align the data segment, and disable linking against shared
796
libraries.  If the output format supports Unix style magic numbers,
797
mark the output as @code{OMAGIC}. Note: Although a writable text section
798
is allowed for PE-COFF targets, it does not conform to the format
799
specification published by Microsoft.
800
 
801
@kindex --no-omagic
802
@cindex OMAGIC
803
@item --no-omagic
804
This option negates most of the effects of the @option{-N} option.  It
805
sets the text section to be read-only, and forces the data segment to
806
be page-aligned.  Note - this option does not enable linking against
807
shared libraries.  Use @option{-Bdynamic} for this.
808
 
809
@kindex -o @var{output}
810
@kindex --output=@var{output}
811
@cindex naming the output file
812
@item -o @var{output}
813
@itemx --output=@var{output}
814
Use @var{output} as the name for the program produced by @command{ld}; if this
815
option is not specified, the name @file{a.out} is used by default.  The
816
script command @code{OUTPUT} can also specify the output file name.
817
 
818
@kindex -O @var{level}
819
@cindex generating optimized output
820
@item -O @var{level}
821
If @var{level} is a numeric values greater than zero @command{ld} optimizes
822
the output.  This might take significantly longer and therefore probably
823
should only be enabled for the final binary.  At the moment this
824
option only affects ELF shared library generation.  Future releases of
825
the linker may make more use of this option.  Also currently there is
826
no difference in the linker's behaviour for different non-zero values
827
of this option.  Again this may change with future releases.
828
 
829
@kindex -q
830
@kindex --emit-relocs
831
@cindex retain relocations in final executable
832
@item -q
833
@itemx --emit-relocs
834
Leave relocation sections and contents in fully linked executables.
835
Post link analysis and optimization tools may need this information in
836
order to perform correct modifications of executables.  This results
837
in larger executables.
838
 
839
This option is currently only supported on ELF platforms.
840
 
841
@kindex --force-dynamic
842
@cindex forcing the creation of dynamic sections
843
@item --force-dynamic
844
Force the output file to have dynamic sections.  This option is specific
845
to VxWorks targets.
846
 
847
@cindex partial link
848
@cindex relocatable output
849
@kindex -r
850
@kindex --relocatable
851
@item -r
852
@itemx --relocatable
853
Generate relocatable output---i.e., generate an output file that can in
854
turn serve as input to @command{ld}.  This is often called @dfn{partial
855
linking}.  As a side effect, in environments that support standard Unix
856
magic numbers, this option also sets the output file's magic number to
857
@code{OMAGIC}.
858
@c ; see @option{-N}.
859
If this option is not specified, an absolute file is produced.  When
860
linking C++ programs, this option @emph{will not} resolve references to
861
constructors; to do that, use @samp{-Ur}.
862
 
863
When an input file does not have the same format as the output file,
864
partial linking is only supported if that input file does not contain any
865
relocations.  Different output formats can have further restrictions; for
866
example some @code{a.out}-based formats do not support partial linking
867
with input files in other formats at all.
868
 
869
This option does the same thing as @samp{-i}.
870
 
871
@kindex -R @var{file}
872
@kindex --just-symbols=@var{file}
873
@cindex symbol-only input
874
@item -R @var{filename}
875
@itemx --just-symbols=@var{filename}
876
Read symbol names and their addresses from @var{filename}, but do not
877
relocate it or include it in the output.  This allows your output file
878
to refer symbolically to absolute locations of memory defined in other
879
programs.  You may use this option more than once.
880
 
881
For compatibility with other ELF linkers, if the @option{-R} option is
882
followed by a directory name, rather than a file name, it is treated as
883
the @option{-rpath} option.
884
 
885
@kindex -s
886
@kindex --strip-all
887
@cindex strip all symbols
888
@item -s
889
@itemx --strip-all
890
Omit all symbol information from the output file.
891
 
892
@kindex -S
893
@kindex --strip-debug
894
@cindex strip debugger symbols
895
@item -S
896
@itemx --strip-debug
897
Omit debugger symbol information (but not all symbols) from the output file.
898
 
899
@kindex -t
900
@kindex --trace
901
@cindex input files, displaying
902
@item -t
903
@itemx --trace
904
Print the names of the input files as @command{ld} processes them.
905
 
906
@kindex -T @var{script}
907
@kindex --script=@var{script}
908
@cindex script files
909
@item -T @var{scriptfile}
910
@itemx --script=@var{scriptfile}
911
Use @var{scriptfile} as the linker script.  This script replaces
912
@command{ld}'s default linker script (rather than adding to it), so
913
@var{commandfile} must specify everything necessary to describe the
914
output file.  @xref{Scripts}.  If @var{scriptfile} does not exist in
915
the current directory, @code{ld} looks for it in the directories
916
specified by any preceding @samp{-L} options.  Multiple @samp{-T}
917
options accumulate.
918
 
919
@kindex -dT @var{script}
920
@kindex --default-script=@var{script}
921
@cindex script files
922
@item -dT @var{scriptfile}
923
@itemx --default-script=@var{scriptfile}
924
Use @var{scriptfile} as the default linker script.  @xref{Scripts}.
925
 
926
This option is similar to the @option{--script} option except that
927
processing of the script is delayed until after the rest of the
928
command line has been processed.  This allows options placed after the
929
@option{--default-script} option on the command line to affect the
930
behaviour of the linker script, which can be important when the linker
931
command line cannot be directly controlled by the user.  (eg because
932
the command line is being constructed by another tool, such as
933
@samp{gcc}).
934
 
935
@kindex -u @var{symbol}
936
@kindex --undefined=@var{symbol}
937
@cindex undefined symbol
938
@item -u @var{symbol}
939
@itemx --undefined=@var{symbol}
940
Force @var{symbol} to be entered in the output file as an undefined
941
symbol.  Doing this may, for example, trigger linking of additional
942
modules from standard libraries.  @samp{-u} may be repeated with
943
different option arguments to enter additional undefined symbols.  This
944
option is equivalent to the @code{EXTERN} linker script command.
945
 
946
@kindex -Ur
947
@cindex constructors
948
@item -Ur
949
For anything other than C++ programs, this option is equivalent to
950
@samp{-r}: it generates relocatable output---i.e., an output file that can in
951
turn serve as input to @command{ld}.  When linking C++ programs, @samp{-Ur}
952
@emph{does} resolve references to constructors, unlike @samp{-r}.
953
It does not work to use @samp{-Ur} on files that were themselves linked
954
with @samp{-Ur}; once the constructor table has been built, it cannot
955
be added to.  Use @samp{-Ur} only for the last partial link, and
956
@samp{-r} for the others.
957
 
958
@kindex --unique[=@var{SECTION}]
959
@item --unique[=@var{SECTION}]
960
Creates a separate output section for every input section matching
961
@var{SECTION}, or if the optional wildcard @var{SECTION} argument is
962
missing, for every orphan input section.  An orphan section is one not
963
specifically mentioned in a linker script.  You may use this option
964
multiple times on the command line;  It prevents the normal merging of
965
input sections with the same name, overriding output section assignments
966
in a linker script.
967
 
968
@kindex -v
969
@kindex -V
970
@kindex --version
971
@cindex version
972
@item -v
973
@itemx --version
974
@itemx -V
975
Display the version number for @command{ld}.  The @option{-V} option also
976
lists the supported emulations.
977
 
978
@kindex -x
979
@kindex --discard-all
980
@cindex deleting local symbols
981
@item -x
982
@itemx --discard-all
983
Delete all local symbols.
984
 
985
@kindex -X
986
@kindex --discard-locals
987
@cindex local symbols, deleting
988
@item -X
989
@itemx --discard-locals
990
Delete all temporary local symbols.  (These symbols start with
991
system-specific local label prefixes, typically @samp{.L} for ELF systems
992
or @samp{L} for traditional a.out systems.)
993
 
994
@kindex -y @var{symbol}
995
@kindex --trace-symbol=@var{symbol}
996
@cindex symbol tracing
997
@item -y @var{symbol}
998
@itemx --trace-symbol=@var{symbol}
999
Print the name of each linked file in which @var{symbol} appears.  This
1000
option may be given any number of times.  On many systems it is necessary
1001
to prepend an underscore.
1002
 
1003
This option is useful when you have an undefined symbol in your link but
1004
don't know where the reference is coming from.
1005
 
1006
@kindex -Y @var{path}
1007
@item -Y @var{path}
1008
Add @var{path} to the default library search path.  This option exists
1009
for Solaris compatibility.
1010
 
1011
@kindex -z @var{keyword}
1012
@item -z @var{keyword}
1013
The recognized keywords are:
1014
@table @samp
1015
 
1016
@item combreloc
1017
Combines multiple reloc sections and sorts them to make dynamic symbol
1018
lookup caching possible.
1019
 
1020
@item defs
1021
Disallows undefined symbols in object files.  Undefined symbols in
1022
shared libraries are still allowed.
1023
 
1024
@item execstack
1025
Marks the object as requiring executable stack.
1026
 
1027
@item initfirst
1028
This option is only meaningful when building a shared object.
1029
It marks the object so that its runtime initialization will occur
1030
before the runtime initialization of any other objects brought into
1031
the process at the same time.  Similarly the runtime finalization of
1032
the object will occur after the runtime finalization of any other
1033
objects.
1034
 
1035
@item interpose
1036
Marks the object that its symbol table interposes before all symbols
1037
but the primary executable.
1038
 
1039
@item lazy
1040
When generating an executable or shared library, mark it to tell the
1041
dynamic linker to defer function call resolution to the point when
1042
the function is called (lazy binding), rather than at load time.
1043
Lazy binding is the default.
1044
 
1045
@item loadfltr
1046
Marks  the object that its filters be processed immediately at
1047
runtime.
1048
 
1049
@item muldefs
1050
Allows multiple definitions.
1051
 
1052
@item nocombreloc
1053
Disables multiple reloc sections combining.
1054
 
1055
@item nocopyreloc
1056
Disables production of copy relocs.
1057
 
1058
@item nodefaultlib
1059
Marks the object that the search for dependencies of this object will
1060
ignore any default library search paths.
1061
 
1062
@item nodelete
1063
Marks the object shouldn't be unloaded at runtime.
1064
 
1065
@item nodlopen
1066
Marks the object not available to @code{dlopen}.
1067
 
1068
@item nodump
1069
Marks the object can not be dumped by @code{dldump}.
1070
 
1071
@item noexecstack
1072
Marks the object as not requiring executable stack.
1073
 
1074
@item norelro
1075
Don't create an ELF @code{PT_GNU_RELRO} segment header in the object.
1076
 
1077
@item now
1078
When generating an executable or shared library, mark it to tell the
1079
dynamic linker to resolve all symbols when the program is started, or
1080
when the shared library is linked to using dlopen, instead of
1081
deferring function call resolution to the point when the function is
1082
first called.
1083
 
1084
@item origin
1085
Marks the object may contain $ORIGIN.
1086
 
1087
@item relro
1088
Create an ELF @code{PT_GNU_RELRO} segment header in the object.
1089
 
1090
@item max-page-size=@var{value}
1091
Set the emulation maximum page size to @var{value}.
1092
 
1093
@item common-page-size=@var{value}
1094
Set the emulation common page size to @var{value}.
1095
 
1096
@end table
1097
 
1098
Other keywords are ignored for Solaris compatibility.
1099
 
1100
@kindex -(
1101
@cindex groups of archives
1102
@item -( @var{archives} -)
1103
@itemx --start-group @var{archives} --end-group
1104
The @var{archives} should be a list of archive files.  They may be
1105
either explicit file names, or @samp{-l} options.
1106
 
1107
The specified archives are searched repeatedly until no new undefined
1108
references are created.  Normally, an archive is searched only once in
1109
the order that it is specified on the command line.  If a symbol in that
1110
archive is needed to resolve an undefined symbol referred to by an
1111
object in an archive that appears later on the command line, the linker
1112
would not be able to resolve that reference.  By grouping the archives,
1113
they all be searched repeatedly until all possible references are
1114
resolved.
1115
 
1116
Using this option has a significant performance cost.  It is best to use
1117
it only when there are unavoidable circular references between two or
1118
more archives.
1119
 
1120
@kindex --accept-unknown-input-arch
1121
@kindex --no-accept-unknown-input-arch
1122
@item --accept-unknown-input-arch
1123
@itemx --no-accept-unknown-input-arch
1124
Tells the linker to accept input files whose architecture cannot be
1125
recognised.  The assumption is that the user knows what they are doing
1126
and deliberately wants to link in these unknown input files.  This was
1127
the default behaviour of the linker, before release 2.14.  The default
1128
behaviour from release 2.14 onwards is to reject such input files, and
1129
so the @samp{--accept-unknown-input-arch} option has been added to
1130
restore the old behaviour.
1131
 
1132
@kindex --as-needed
1133
@kindex --no-as-needed
1134
@item --as-needed
1135
@itemx --no-as-needed
1136
This option affects ELF DT_NEEDED tags for dynamic libraries mentioned
1137
on the command line after the @option{--as-needed} option.  Normally
1138
the linker will add a DT_NEEDED tag for each dynamic library mentioned
1139
on the command line, regardless of whether the library is actually
1140
needed or not.  @option{--as-needed} causes a DT_NEEDED tag to only be
1141
emitted for a library that satisfies an undefined symbol reference
1142
from a regular object file or, if the library is not found in the
1143
DT_NEEDED lists of other libraries linked up to that point, an
1144
undefined symbol reference from another dynamic library.
1145
@option{--no-as-needed} restores the default behaviour.
1146
 
1147
@kindex --add-needed
1148
@kindex --no-add-needed
1149
@item --add-needed
1150
@itemx --no-add-needed
1151
These two options have been deprecated because of the similarity of
1152
their names to the @option{--as-needed} and @option{--no-as-needed}
1153
options.  They have been replaced by @option{--copy-dt-needed-entries}
1154
and @option{--no-copy-dt-needed-entries}.
1155
 
1156
@kindex -assert @var{keyword}
1157
@item -assert @var{keyword}
1158
This option is ignored for SunOS compatibility.
1159
 
1160
@kindex -Bdynamic
1161
@kindex -dy
1162
@kindex -call_shared
1163
@item -Bdynamic
1164
@itemx -dy
1165
@itemx -call_shared
1166
Link against dynamic libraries.  This is only meaningful on platforms
1167
for which shared libraries are supported.  This option is normally the
1168
default on such platforms.  The different variants of this option are
1169
for compatibility with various systems.  You may use this option
1170
multiple times on the command line: it affects library searching for
1171
@option{-l} options which follow it.
1172
 
1173
@kindex -Bgroup
1174
@item -Bgroup
1175
Set the @code{DF_1_GROUP} flag in the @code{DT_FLAGS_1} entry in the dynamic
1176
section.  This causes the runtime linker to handle lookups in this
1177
object and its dependencies to be performed only inside the group.
1178
@option{--unresolved-symbols=report-all} is implied.  This option is
1179
only meaningful on ELF platforms which support shared libraries.
1180
 
1181
@kindex -Bstatic
1182
@kindex -dn
1183
@kindex -non_shared
1184
@kindex -static
1185
@item -Bstatic
1186
@itemx -dn
1187
@itemx -non_shared
1188
@itemx -static
1189
Do not link against shared libraries.  This is only meaningful on
1190
platforms for which shared libraries are supported.  The different
1191
variants of this option are for compatibility with various systems.  You
1192
may use this option multiple times on the command line: it affects
1193
library searching for @option{-l} options which follow it.  This
1194
option also implies @option{--unresolved-symbols=report-all}.  This
1195
option can be used with @option{-shared}.  Doing so means that a
1196
shared library is being created but that all of the library's external
1197
references must be resolved by pulling in entries from static
1198
libraries.
1199
 
1200
@kindex -Bsymbolic
1201
@item -Bsymbolic
1202
When creating a shared library, bind references to global symbols to the
1203
definition within the shared library, if any.  Normally, it is possible
1204
for a program linked against a shared library to override the definition
1205
within the shared library.  This option is only meaningful on ELF
1206
platforms which support shared libraries.
1207
 
1208
@kindex -Bsymbolic-functions
1209
@item -Bsymbolic-functions
1210
When creating a shared library, bind references to global function
1211
symbols to the definition within the shared library, if any.
1212
This option is only meaningful on ELF platforms which support shared
1213
libraries.
1214
 
1215
@kindex --dynamic-list=@var{dynamic-list-file}
1216
@item --dynamic-list=@var{dynamic-list-file}
1217
Specify the name of a dynamic list file to the linker.  This is
1218
typically used when creating shared libraries to specify a list of
1219
global symbols whose references shouldn't be bound to the definition
1220
within the shared library, or creating dynamically linked executables
1221
to specify a list of symbols which should be added to the symbol table
1222
in the executable.  This option is only meaningful on ELF platforms
1223
which support shared libraries.
1224
 
1225
The format of the dynamic list is the same as the version node without
1226
scope and node name.  See @ref{VERSION} for more information.
1227
 
1228
@kindex --dynamic-list-data
1229
@item --dynamic-list-data
1230
Include all global data symbols to the dynamic list.
1231
 
1232
@kindex --dynamic-list-cpp-new
1233
@item --dynamic-list-cpp-new
1234
Provide the builtin dynamic list for C++ operator new and delete.  It
1235
is mainly useful for building shared libstdc++.
1236
 
1237
@kindex --dynamic-list-cpp-typeinfo
1238
@item --dynamic-list-cpp-typeinfo
1239
Provide the builtin dynamic list for C++ runtime type identification.
1240
 
1241
@kindex --check-sections
1242
@kindex --no-check-sections
1243
@item --check-sections
1244
@itemx --no-check-sections
1245
Asks the linker @emph{not} to check section addresses after they have
1246
been assigned to see if there are any overlaps.  Normally the linker will
1247
perform this check, and if it finds any overlaps it will produce
1248
suitable error messages.  The linker does know about, and does make
1249
allowances for sections in overlays.  The default behaviour can be
1250
restored by using the command line switch @option{--check-sections}.
1251
Section overlap is not usually checked for relocatable links.  You can
1252
force checking in that case by using the @option{--check-sections}
1253
option.
1254
 
1255
@kindex --copy-dt-needed-entries
1256
@kindex --no-copy-dt-needed-entries
1257
@item --copy-dt-needed-entries
1258
@itemx --no-copy-dt-needed-entries
1259
This option affects the treatment of dynamic libraries referred to
1260
by DT_NEEDED tags @emph{inside} ELF dynamic libraries mentioned on the
1261 157 khays
command line.  Normally the linker won't add a DT_NEEDED tag to the
1262 145 khays
output binary for each library mentioned in a DT_NEEDED tag in an
1263 157 khays
input dynamic library.  With @option{--copy-dt-needed-entries}
1264 145 khays
specified on the command line however any dynamic libraries that
1265 157 khays
follow it will have their DT_NEEDED entries added.  The default
1266
behaviour can be restored with @option{--no-copy-dt-needed-entries}.
1267 145 khays
 
1268
This option also has an effect on the resolution of symbols in dynamic
1269 157 khays
libraries.  With @option{--copy-dt-needed-entries} dynamic libraries
1270
mentioned on the command line will be recursively searched, following
1271
their DT_NEEDED tags to other libraries, in order to resolve symbols
1272
required by the output binary.  With the default setting however
1273
the searching of dynamic libraries that follow it will stop with the
1274
dynamic library itself.  No DT_NEEDED links will be traversed to resolve
1275 145 khays
symbols.
1276
 
1277
@cindex cross reference table
1278
@kindex --cref
1279
@item --cref
1280
Output a cross reference table.  If a linker map file is being
1281
generated, the cross reference table is printed to the map file.
1282
Otherwise, it is printed on the standard output.
1283
 
1284
The format of the table is intentionally simple, so that it may be
1285
easily processed by a script if necessary.  The symbols are printed out,
1286
sorted by name.  For each symbol, a list of file names is given.  If the
1287
symbol is defined, the first file listed is the location of the
1288
definition.  The remaining files contain references to the symbol.
1289
 
1290
@cindex common allocation
1291
@kindex --no-define-common
1292
@item --no-define-common
1293
This option inhibits the assignment of addresses to common symbols.
1294
The script command @code{INHIBIT_COMMON_ALLOCATION} has the same effect.
1295
@xref{Miscellaneous Commands}.
1296
 
1297
The @samp{--no-define-common} option allows decoupling
1298
the decision to assign addresses to Common symbols from the choice
1299
of the output file type; otherwise a non-Relocatable output type
1300
forces assigning addresses to Common symbols.
1301
Using @samp{--no-define-common} allows Common symbols that are referenced
1302
from a shared library to be assigned addresses only in the main program.
1303
This eliminates the unused duplicate space in the shared library,
1304
and also prevents any possible confusion over resolving to the wrong
1305
duplicate when there are many dynamic modules with specialized search
1306
paths for runtime symbol resolution.
1307
 
1308
@cindex symbols, from command line
1309
@kindex --defsym=@var{symbol}=@var{exp}
1310
@item --defsym=@var{symbol}=@var{expression}
1311
Create a global symbol in the output file, containing the absolute
1312
address given by @var{expression}.  You may use this option as many
1313
times as necessary to define multiple symbols in the command line.  A
1314
limited form of arithmetic is supported for the @var{expression} in this
1315
context: you may give a hexadecimal constant or the name of an existing
1316
symbol, or use @code{+} and @code{-} to add or subtract hexadecimal
1317
constants or symbols.  If you need more elaborate expressions, consider
1318
using the linker command language from a script (@pxref{Assignments,,
1319
Assignment: Symbol Definitions}).  @emph{Note:} there should be no white
1320
space between @var{symbol}, the equals sign (``@key{=}''), and
1321
@var{expression}.
1322
 
1323
@cindex demangling, from command line
1324
@kindex --demangle[=@var{style}]
1325
@kindex --no-demangle
1326
@item --demangle[=@var{style}]
1327
@itemx --no-demangle
1328
These options control whether to demangle symbol names in error messages
1329
and other output.  When the linker is told to demangle, it tries to
1330
present symbol names in a readable fashion: it strips leading
1331
underscores if they are used by the object file format, and converts C++
1332
mangled symbol names into user readable names.  Different compilers have
1333
different mangling styles.  The optional demangling style argument can be used
1334
to choose an appropriate demangling style for your compiler.  The linker will
1335
demangle by default unless the environment variable @samp{COLLECT_NO_DEMANGLE}
1336
is set.  These options may be used to override the default.
1337
 
1338
@cindex dynamic linker, from command line
1339
@kindex -I@var{file}
1340
@kindex --dynamic-linker=@var{file}
1341
@item -I@var{file}
1342
@itemx --dynamic-linker=@var{file}
1343
Set the name of the dynamic linker.  This is only meaningful when
1344
generating dynamically linked ELF executables.  The default dynamic
1345
linker is normally correct; don't use this unless you know what you are
1346
doing.
1347
 
1348
@kindex --fatal-warnings
1349
@kindex --no-fatal-warnings
1350
@item --fatal-warnings
1351
@itemx --no-fatal-warnings
1352
Treat all warnings as errors.  The default behaviour can be restored
1353
with the option @option{--no-fatal-warnings}.
1354
 
1355
@kindex --force-exe-suffix
1356
@item  --force-exe-suffix
1357
Make sure that an output file has a .exe suffix.
1358
 
1359
If a successfully built fully linked output file does not have a
1360
@code{.exe} or @code{.dll} suffix, this option forces the linker to copy
1361
the output file to one of the same name with a @code{.exe} suffix. This
1362
option is useful when using unmodified Unix makefiles on a Microsoft
1363
Windows host, since some versions of Windows won't run an image unless
1364
it ends in a @code{.exe} suffix.
1365
 
1366
@kindex --gc-sections
1367
@kindex --no-gc-sections
1368
@cindex garbage collection
1369
@item --gc-sections
1370
@itemx --no-gc-sections
1371
Enable garbage collection of unused input sections.  It is ignored on
1372
targets that do not support this option.  The default behaviour (of not
1373
performing this garbage collection) can be restored by specifying
1374
@samp{--no-gc-sections} on the command line.
1375
 
1376
@samp{--gc-sections} decides which input sections are used by
1377
examining symbols and relocations.  The section containing the entry
1378
symbol and all sections containing symbols undefined on the
1379
command-line will be kept, as will sections containing symbols
1380
referenced by dynamic objects.  Note that when building shared
1381
libraries, the linker must assume that any visible symbol is
1382
referenced.  Once this initial set of sections has been determined,
1383
the linker recursively marks as used any section referenced by their
1384
relocations.  See @samp{--entry} and @samp{--undefined}.
1385
 
1386
This option can be set when doing a partial link (enabled with option
1387
@samp{-r}).  In this case the root of symbols kept must be explicitly
1388
specified either by an @samp{--entry} or @samp{--undefined} option or by
1389
a @code{ENTRY} command in the linker script.
1390
 
1391
@kindex --print-gc-sections
1392
@kindex --no-print-gc-sections
1393
@cindex garbage collection
1394
@item --print-gc-sections
1395
@itemx --no-print-gc-sections
1396
List all sections removed by garbage collection.  The listing is
1397
printed on stderr.  This option is only effective if garbage
1398
collection has been enabled via the @samp{--gc-sections}) option.  The
1399
default behaviour (of not listing the sections that are removed) can
1400
be restored by specifying @samp{--no-print-gc-sections} on the command
1401
line.
1402
 
1403 157 khays
@kindex --print-output-format
1404
@cindex output format
1405
@item --print-output-format
1406
Print the name of the default output format (perhaps influenced by
1407
other command-line options).  This is the string that would appear
1408
in an @code{OUTPUT_FORMAT} linker script command (@pxref{File Commands}).
1409
 
1410 145 khays
@cindex help
1411
@cindex usage
1412
@kindex --help
1413
@item --help
1414
Print a summary of the command-line options on the standard output and exit.
1415
 
1416
@kindex --target-help
1417
@item --target-help
1418
Print a summary of all target specific options on the standard output and exit.
1419
 
1420
@kindex -Map=@var{mapfile}
1421
@item -Map=@var{mapfile}
1422
Print a link map to the file @var{mapfile}.  See the description of the
1423
@option{-M} option, above.
1424
 
1425
@cindex memory usage
1426
@kindex --no-keep-memory
1427
@item --no-keep-memory
1428
@command{ld} normally optimizes for speed over memory usage by caching the
1429
symbol tables of input files in memory.  This option tells @command{ld} to
1430
instead optimize for memory usage, by rereading the symbol tables as
1431
necessary.  This may be required if @command{ld} runs out of memory space
1432
while linking a large executable.
1433
 
1434
@kindex --no-undefined
1435
@kindex -z defs
1436
@item --no-undefined
1437
@itemx -z defs
1438
Report unresolved symbol references from regular object files.  This
1439
is done even if the linker is creating a non-symbolic shared library.
1440
The switch @option{--[no-]allow-shlib-undefined} controls the
1441
behaviour for reporting unresolved references found in shared
1442
libraries being linked in.
1443
 
1444
@kindex --allow-multiple-definition
1445
@kindex -z muldefs
1446
@item --allow-multiple-definition
1447
@itemx -z muldefs
1448
Normally when a symbol is defined multiple times, the linker will
1449
report a fatal error. These options allow multiple definitions and the
1450
first definition will be used.
1451
 
1452
@kindex --allow-shlib-undefined
1453
@kindex --no-allow-shlib-undefined
1454
@item --allow-shlib-undefined
1455
@itemx --no-allow-shlib-undefined
1456
Allows or disallows undefined symbols in shared libraries.
1457
This switch is similar to @option{--no-undefined} except that it
1458
determines the behaviour when the undefined symbols are in a
1459
shared library rather than a regular object file.  It does not affect
1460
how undefined symbols in regular object files are handled.
1461
 
1462
The default behaviour is to report errors for any undefined symbols
1463
referenced in shared libraries if the linker is being used to create
1464
an executable, but to allow them if the linker is being used to create
1465
a shared library.
1466
 
1467
The reasons for allowing undefined symbol references in shared
1468
libraries specified at link time are that:
1469
 
1470
@itemize @bullet
1471
@item
1472
A shared library specified at link time may not be the same as the one
1473
that is available at load time, so the symbol might actually be
1474
resolvable at load time.
1475
@item
1476
There are some operating systems, eg BeOS and HPPA, where undefined
1477
symbols in shared libraries are normal.
1478
 
1479
The BeOS kernel for example patches shared libraries at load time to
1480
select whichever function is most appropriate for the current
1481
architecture.  This is used, for example, to dynamically select an
1482
appropriate memset function.
1483
@end itemize
1484
 
1485
@kindex --no-undefined-version
1486
@item --no-undefined-version
1487
Normally when a symbol has an undefined version, the linker will ignore
1488
it. This option disallows symbols with undefined version and a fatal error
1489
will be issued instead.
1490
 
1491
@kindex --default-symver
1492
@item --default-symver
1493
Create and use a default symbol version (the soname) for unversioned
1494
exported symbols.
1495
 
1496
@kindex --default-imported-symver
1497
@item --default-imported-symver
1498
Create and use a default symbol version (the soname) for unversioned
1499
imported symbols.
1500
 
1501
@kindex --no-warn-mismatch
1502
@item --no-warn-mismatch
1503
Normally @command{ld} will give an error if you try to link together input
1504
files that are mismatched for some reason, perhaps because they have
1505
been compiled for different processors or for different endiannesses.
1506
This option tells @command{ld} that it should silently permit such possible
1507
errors.  This option should only be used with care, in cases when you
1508
have taken some special action that ensures that the linker errors are
1509
inappropriate.
1510
 
1511
@kindex --no-warn-search-mismatch
1512
@item --no-warn-search-mismatch
1513
Normally @command{ld} will give a warning if it finds an incompatible
1514
library during a library search.  This option silences the warning.
1515
 
1516
@kindex --no-whole-archive
1517
@item --no-whole-archive
1518
Turn off the effect of the @option{--whole-archive} option for subsequent
1519
archive files.
1520
 
1521
@cindex output file after errors
1522
@kindex --noinhibit-exec
1523
@item --noinhibit-exec
1524
Retain the executable output file whenever it is still usable.
1525
Normally, the linker will not produce an output file if it encounters
1526
errors during the link process; it exits without writing an output file
1527
when it issues any error whatsoever.
1528
 
1529
@kindex -nostdlib
1530
@item -nostdlib
1531
Only search library directories explicitly specified on the
1532
command line.  Library directories specified in linker scripts
1533
(including linker scripts specified on the command line) are ignored.
1534
 
1535
@ifclear SingleFormat
1536
@kindex --oformat=@var{output-format}
1537
@item --oformat=@var{output-format}
1538
@command{ld} may be configured to support more than one kind of object
1539
file.  If your @command{ld} is configured this way, you can use the
1540
@samp{--oformat} option to specify the binary format for the output
1541
object file.  Even when @command{ld} is configured to support alternative
1542
object formats, you don't usually need to specify this, as @command{ld}
1543
should be configured to produce as a default output format the most
1544
usual format on each machine.  @var{output-format} is a text string, the
1545
name of a particular format supported by the BFD libraries.  (You can
1546
list the available binary formats with @samp{objdump -i}.)  The script
1547
command @code{OUTPUT_FORMAT} can also specify the output format, but
1548
this option overrides it.  @xref{BFD}.
1549
@end ifclear
1550
 
1551
@kindex -pie
1552
@kindex --pic-executable
1553
@item -pie
1554
@itemx --pic-executable
1555
@cindex position independent executables
1556
Create a position independent executable.  This is currently only supported on
1557
ELF platforms.  Position independent executables are similar to shared
1558
libraries in that they are relocated by the dynamic linker to the virtual
1559
address the OS chooses for them (which can vary between invocations).  Like
1560
normal dynamically linked executables they can be executed and symbols
1561
defined in the executable cannot be overridden by shared libraries.
1562
 
1563
@kindex -qmagic
1564
@item -qmagic
1565
This option is ignored for Linux compatibility.
1566
 
1567
@kindex -Qy
1568
@item -Qy
1569
This option is ignored for SVR4 compatibility.
1570
 
1571
@kindex --relax
1572
@cindex synthesizing linker
1573
@cindex relaxing addressing modes
1574
@cindex --no-relax
1575
@item --relax
1576
@itemx --no-relax
1577
An option with machine dependent effects.
1578
@ifset GENERIC
1579
This option is only supported on a few targets.
1580
@end ifset
1581
@ifset H8300
1582
@xref{H8/300,,@command{ld} and the H8/300}.
1583
@end ifset
1584
@ifset I960
1585
@xref{i960,, @command{ld} and the Intel 960 family}.
1586
@end ifset
1587
@ifset XTENSA
1588
@xref{Xtensa,, @command{ld} and Xtensa Processors}.
1589
@end ifset
1590
@ifset M68HC11
1591
@xref{M68HC11/68HC12,,@command{ld} and the 68HC11 and 68HC12}.
1592
@end ifset
1593
@ifset POWERPC
1594
@xref{PowerPC ELF32,,@command{ld} and PowerPC 32-bit ELF Support}.
1595
@end ifset
1596
 
1597
On some platforms the @samp{--relax} option performs target specific,
1598
global optimizations that become possible when the linker resolves
1599
addressing in the program, such as relaxing address modes,
1600
synthesizing new instructions, selecting shorter version of current
1601
instructions, and combinig constant values.
1602
 
1603
On some platforms these link time global optimizations may make symbolic
1604
debugging of the resulting executable impossible.
1605
@ifset GENERIC
1606
This is known to be the case for the Matsushita MN10200 and MN10300
1607
family of processors.
1608
@end ifset
1609
 
1610
@ifset GENERIC
1611
On platforms where this is not supported, @samp{--relax} is accepted,
1612
but ignored.
1613
@end ifset
1614
 
1615
On platforms where @samp{--relax} is accepted the option
1616
@samp{--no-relax} can be used to disable the feature.
1617
 
1618
@cindex retaining specified symbols
1619
@cindex stripping all but some symbols
1620
@cindex symbols, retaining selectively
1621
@kindex --retain-symbols-file=@var{filename}
1622
@item --retain-symbols-file=@var{filename}
1623
Retain @emph{only} the symbols listed in the file @var{filename},
1624
discarding all others.  @var{filename} is simply a flat file, with one
1625
symbol name per line.  This option is especially useful in environments
1626
@ifset GENERIC
1627
(such as VxWorks)
1628
@end ifset
1629
where a large global symbol table is accumulated gradually, to conserve
1630
run-time memory.
1631
 
1632
@samp{--retain-symbols-file} does @emph{not} discard undefined symbols,
1633
or symbols needed for relocations.
1634
 
1635
You may only specify @samp{--retain-symbols-file} once in the command
1636
line.  It overrides @samp{-s} and @samp{-S}.
1637
 
1638
@ifset GENERIC
1639
@item -rpath=@var{dir}
1640
@cindex runtime library search path
1641
@kindex -rpath=@var{dir}
1642
Add a directory to the runtime library search path.  This is used when
1643
linking an ELF executable with shared objects.  All @option{-rpath}
1644
arguments are concatenated and passed to the runtime linker, which uses
1645
them to locate shared objects at runtime.  The @option{-rpath} option is
1646
also used when locating shared objects which are needed by shared
1647
objects explicitly included in the link; see the description of the
1648
@option{-rpath-link} option.  If @option{-rpath} is not used when linking an
1649
ELF executable, the contents of the environment variable
1650
@code{LD_RUN_PATH} will be used if it is defined.
1651
 
1652
The @option{-rpath} option may also be used on SunOS.  By default, on
1653
SunOS, the linker will form a runtime search patch out of all the
1654
@option{-L} options it is given.  If a @option{-rpath} option is used, the
1655
runtime search path will be formed exclusively using the @option{-rpath}
1656
options, ignoring the @option{-L} options.  This can be useful when using
1657
gcc, which adds many @option{-L} options which may be on NFS mounted
1658
file systems.
1659
 
1660
For compatibility with other ELF linkers, if the @option{-R} option is
1661
followed by a directory name, rather than a file name, it is treated as
1662
the @option{-rpath} option.
1663
@end ifset
1664
 
1665
@ifset GENERIC
1666
@cindex link-time runtime library search path
1667
@kindex -rpath-link=@var{dir}
1668
@item -rpath-link=@var{dir}
1669
When using ELF or SunOS, one shared library may require another.  This
1670
happens when an @code{ld -shared} link includes a shared library as one
1671
of the input files.
1672
 
1673
When the linker encounters such a dependency when doing a non-shared,
1674
non-relocatable link, it will automatically try to locate the required
1675
shared library and include it in the link, if it is not included
1676
explicitly.  In such a case, the @option{-rpath-link} option
1677
specifies the first set of directories to search.  The
1678
@option{-rpath-link} option may specify a sequence of directory names
1679
either by specifying a list of names separated by colons, or by
1680
appearing multiple times.
1681
 
1682
This option should be used with caution as it overrides the search path
1683
that may have been hard compiled into a shared library. In such a case it
1684
is possible to use unintentionally a different search path than the
1685
runtime linker would do.
1686
 
1687
The linker uses the following search paths to locate required shared
1688
libraries:
1689
@enumerate
1690
@item
1691
Any directories specified by @option{-rpath-link} options.
1692
@item
1693
Any directories specified by @option{-rpath} options.  The difference
1694
between @option{-rpath} and @option{-rpath-link} is that directories
1695
specified by @option{-rpath} options are included in the executable and
1696
used at runtime, whereas the @option{-rpath-link} option is only effective
1697
at link time. Searching @option{-rpath} in this way is only supported
1698
by native linkers and cross linkers which have been configured with
1699
the @option{--with-sysroot} option.
1700
@item
1701
On an ELF system, for native linkers, if the @option{-rpath} and
1702
@option{-rpath-link} options were not used, search the contents of the
1703
environment variable @code{LD_RUN_PATH}.
1704
@item
1705
On SunOS, if the @option{-rpath} option was not used, search any
1706
directories specified using @option{-L} options.
1707
@item
1708
For a native linker, the search the contents of the environment
1709
variable @code{LD_LIBRARY_PATH}.
1710
@item
1711
For a native ELF linker, the directories in @code{DT_RUNPATH} or
1712
@code{DT_RPATH} of a shared library are searched for shared
1713
libraries needed by it. The @code{DT_RPATH} entries are ignored if
1714
@code{DT_RUNPATH} entries exist.
1715
@item
1716
The default directories, normally @file{/lib} and @file{/usr/lib}.
1717
@item
1718
For a native linker on an ELF system, if the file @file{/etc/ld.so.conf}
1719
exists, the list of directories found in that file.
1720
@end enumerate
1721
 
1722
If the required shared library is not found, the linker will issue a
1723
warning and continue with the link.
1724
@end ifset
1725
 
1726
@kindex -shared
1727
@kindex -Bshareable
1728
@item -shared
1729
@itemx -Bshareable
1730
@cindex shared libraries
1731
Create a shared library.  This is currently only supported on ELF, XCOFF
1732
and SunOS platforms.  On SunOS, the linker will automatically create a
1733
shared library if the @option{-e} option is not used and there are
1734
undefined symbols in the link.
1735
 
1736
@kindex --sort-common
1737
@item --sort-common
1738
@itemx --sort-common=ascending
1739
@itemx --sort-common=descending
1740
This option tells @command{ld} to sort the common symbols by alignment in
1741
ascending or descending order when it places them in the appropriate output
1742
sections.  The symbol alignments considered are sixteen-byte or larger,
1743
eight-byte, four-byte, two-byte, and one-byte. This is to prevent gaps
1744
between symbols due to alignment constraints.  If no sorting order is
1745
specified, then descending order is assumed.
1746
 
1747
@kindex --sort-section=name
1748
@item --sort-section=name
1749
This option will apply @code{SORT_BY_NAME} to all wildcard section
1750
patterns in the linker script.
1751
 
1752
@kindex --sort-section=alignment
1753
@item --sort-section=alignment
1754
This option will apply @code{SORT_BY_ALIGNMENT} to all wildcard section
1755
patterns in the linker script.
1756
 
1757
@kindex --split-by-file
1758
@item --split-by-file[=@var{size}]
1759
Similar to @option{--split-by-reloc} but creates a new output section for
1760
each input file when @var{size} is reached.  @var{size} defaults to a
1761
size of 1 if not given.
1762
 
1763
@kindex --split-by-reloc
1764
@item --split-by-reloc[=@var{count}]
1765
Tries to creates extra sections in the output file so that no single
1766
output section in the file contains more than @var{count} relocations.
1767
This is useful when generating huge relocatable files for downloading into
1768
certain real time kernels with the COFF object file format; since COFF
1769
cannot represent more than 65535 relocations in a single section.  Note
1770
that this will fail to work with object file formats which do not
1771
support arbitrary sections.  The linker will not split up individual
1772
input sections for redistribution, so if a single input section contains
1773
more than @var{count} relocations one output section will contain that
1774
many relocations.  @var{count} defaults to a value of 32768.
1775
 
1776
@kindex --stats
1777
@item --stats
1778
Compute and display statistics about the operation of the linker, such
1779
as execution time and memory usage.
1780
 
1781
@kindex --sysroot=@var{directory}
1782
@item --sysroot=@var{directory}
1783
Use @var{directory} as the location of the sysroot, overriding the
1784
configure-time default.  This option is only supported by linkers
1785
that were configured using @option{--with-sysroot}.
1786
 
1787
@kindex --traditional-format
1788
@cindex traditional format
1789
@item --traditional-format
1790
For some targets, the output of @command{ld} is different in some ways from
1791
the output of some existing linker.  This switch requests @command{ld} to
1792
use the traditional format instead.
1793
 
1794
@cindex dbx
1795
For example, on SunOS, @command{ld} combines duplicate entries in the
1796
symbol string table.  This can reduce the size of an output file with
1797
full debugging information by over 30 percent.  Unfortunately, the SunOS
1798
@code{dbx} program can not read the resulting program (@code{gdb} has no
1799
trouble).  The @samp{--traditional-format} switch tells @command{ld} to not
1800
combine duplicate entries.
1801
 
1802
@kindex --section-start=@var{sectionname}=@var{org}
1803
@item --section-start=@var{sectionname}=@var{org}
1804
Locate a section in the output file at the absolute
1805
address given by @var{org}.  You may use this option as many
1806
times as necessary to locate multiple sections in the command
1807
line.
1808
@var{org} must be a single hexadecimal integer;
1809
for compatibility with other linkers, you may omit the leading
1810
@samp{0x} usually associated with hexadecimal values.  @emph{Note:} there
1811
should be no white space between @var{sectionname}, the equals
1812
sign (``@key{=}''), and @var{org}.
1813
 
1814
@kindex -Tbss=@var{org}
1815
@kindex -Tdata=@var{org}
1816
@kindex -Ttext=@var{org}
1817
@cindex segment origins, cmd line
1818
@item -Tbss=@var{org}
1819
@itemx -Tdata=@var{org}
1820
@itemx -Ttext=@var{org}
1821
Same as @option{--section-start}, with @code{.bss}, @code{.data} or
1822
@code{.text} as the @var{sectionname}.
1823
 
1824
@kindex -Ttext-segment=@var{org}
1825
@item -Ttext-segment=@var{org}
1826
@cindex text segment origin, cmd line
1827
When creating an ELF executable or shared object, it will set the address
1828
of the first byte of the text segment.
1829
 
1830
@kindex --unresolved-symbols
1831
@item --unresolved-symbols=@var{method}
1832
Determine how to handle unresolved symbols.  There are four possible
1833
values for @samp{method}:
1834
 
1835
@table @samp
1836
@item ignore-all
1837
Do not report any unresolved symbols.
1838
 
1839
@item report-all
1840
Report all unresolved symbols.  This is the default.
1841
 
1842
@item ignore-in-object-files
1843
Report unresolved symbols that are contained in shared libraries, but
1844
ignore them if they come from regular object files.
1845
 
1846
@item ignore-in-shared-libs
1847
Report unresolved symbols that come from regular object files, but
1848
ignore them if they come from shared libraries.  This can be useful
1849
when creating a dynamic binary and it is known that all the shared
1850
libraries that it should be referencing are included on the linker's
1851
command line.
1852
@end table
1853
 
1854
The behaviour for shared libraries on their own can also be controlled
1855
by the @option{--[no-]allow-shlib-undefined} option.
1856
 
1857
Normally the linker will generate an error message for each reported
1858
unresolved symbol but the option @option{--warn-unresolved-symbols}
1859
can change this to a warning.
1860
 
1861
@kindex --verbose[=@var{NUMBER}]
1862
@cindex verbose[=@var{NUMBER}]
1863
@item --dll-verbose
1864
@itemx --verbose[=@var{NUMBER}]
1865
Display the version number for @command{ld} and list the linker emulations
1866
supported.  Display which input files can and cannot be opened.  Display
1867
the linker script being used by the linker. If the optional @var{NUMBER}
1868
argument > 1, plugin symbol status will also be displayed.
1869
 
1870
@kindex --version-script=@var{version-scriptfile}
1871
@cindex version script, symbol versions
1872
@item --version-script=@var{version-scriptfile}
1873
Specify the name of a version script to the linker.  This is typically
1874
used when creating shared libraries to specify additional information
1875
about the version hierarchy for the library being created.  This option
1876
is only fully supported on ELF platforms which support shared libraries;
1877
see @ref{VERSION}.  It is partially supported on PE platforms, which can
1878
use version scripts to filter symbol visibility in auto-export mode: any
1879
symbols marked @samp{local} in the version script will not be exported.
1880
@xref{WIN32}.
1881
 
1882
@kindex --warn-common
1883
@cindex warnings, on combining symbols
1884
@cindex combining symbols, warnings on
1885
@item --warn-common
1886
Warn when a common symbol is combined with another common symbol or with
1887
a symbol definition.  Unix linkers allow this somewhat sloppy practise,
1888
but linkers on some other operating systems do not.  This option allows
1889
you to find potential problems from combining global symbols.
1890
Unfortunately, some C libraries use this practise, so you may get some
1891
warnings about symbols in the libraries as well as in your programs.
1892
 
1893
There are three kinds of global symbols, illustrated here by C examples:
1894
 
1895
@table @samp
1896
@item int i = 1;
1897
A definition, which goes in the initialized data section of the output
1898
file.
1899
 
1900
@item extern int i;
1901
An undefined reference, which does not allocate space.
1902
There must be either a definition or a common symbol for the
1903
variable somewhere.
1904
 
1905
@item int i;
1906
A common symbol.  If there are only (one or more) common symbols for a
1907
variable, it goes in the uninitialized data area of the output file.
1908
The linker merges multiple common symbols for the same variable into a
1909
single symbol.  If they are of different sizes, it picks the largest
1910
size.  The linker turns a common symbol into a declaration, if there is
1911
a definition of the same variable.
1912
@end table
1913
 
1914
The @samp{--warn-common} option can produce five kinds of warnings.
1915
Each warning consists of a pair of lines: the first describes the symbol
1916
just encountered, and the second describes the previous symbol
1917
encountered with the same name.  One or both of the two symbols will be
1918
a common symbol.
1919
 
1920
@enumerate
1921
@item
1922
Turning a common symbol into a reference, because there is already a
1923
definition for the symbol.
1924
@smallexample
1925
@var{file}(@var{section}): warning: common of `@var{symbol}'
1926
   overridden by definition
1927
@var{file}(@var{section}): warning: defined here
1928
@end smallexample
1929
 
1930
@item
1931
Turning a common symbol into a reference, because a later definition for
1932
the symbol is encountered.  This is the same as the previous case,
1933
except that the symbols are encountered in a different order.
1934
@smallexample
1935
@var{file}(@var{section}): warning: definition of `@var{symbol}'
1936
   overriding common
1937
@var{file}(@var{section}): warning: common is here
1938
@end smallexample
1939
 
1940
@item
1941
Merging a common symbol with a previous same-sized common symbol.
1942
@smallexample
1943
@var{file}(@var{section}): warning: multiple common
1944
   of `@var{symbol}'
1945
@var{file}(@var{section}): warning: previous common is here
1946
@end smallexample
1947
 
1948
@item
1949
Merging a common symbol with a previous larger common symbol.
1950
@smallexample
1951
@var{file}(@var{section}): warning: common of `@var{symbol}'
1952
   overridden by larger common
1953
@var{file}(@var{section}): warning: larger common is here
1954
@end smallexample
1955
 
1956
@item
1957
Merging a common symbol with a previous smaller common symbol.  This is
1958
the same as the previous case, except that the symbols are
1959
encountered in a different order.
1960
@smallexample
1961
@var{file}(@var{section}): warning: common of `@var{symbol}'
1962
   overriding smaller common
1963
@var{file}(@var{section}): warning: smaller common is here
1964
@end smallexample
1965
@end enumerate
1966
 
1967
@kindex --warn-constructors
1968
@item --warn-constructors
1969
Warn if any global constructors are used.  This is only useful for a few
1970
object file formats.  For formats like COFF or ELF, the linker can not
1971
detect the use of global constructors.
1972
 
1973
@kindex --warn-multiple-gp
1974
@item --warn-multiple-gp
1975
Warn if multiple global pointer values are required in the output file.
1976
This is only meaningful for certain processors, such as the Alpha.
1977
Specifically, some processors put large-valued constants in a special
1978
section.  A special register (the global pointer) points into the middle
1979
of this section, so that constants can be loaded efficiently via a
1980
base-register relative addressing mode.  Since the offset in
1981
base-register relative mode is fixed and relatively small (e.g., 16
1982
bits), this limits the maximum size of the constant pool.  Thus, in
1983
large programs, it is often necessary to use multiple global pointer
1984
values in order to be able to address all possible constants.  This
1985
option causes a warning to be issued whenever this case occurs.
1986
 
1987
@kindex --warn-once
1988
@cindex warnings, on undefined symbols
1989
@cindex undefined symbols, warnings on
1990
@item --warn-once
1991
Only warn once for each undefined symbol, rather than once per module
1992
which refers to it.
1993
 
1994
@kindex --warn-section-align
1995
@cindex warnings, on section alignment
1996
@cindex section alignment, warnings on
1997
@item --warn-section-align
1998
Warn if the address of an output section is changed because of
1999
alignment.  Typically, the alignment will be set by an input section.
2000
The address will only be changed if it not explicitly specified; that
2001
is, if the @code{SECTIONS} command does not specify a start address for
2002
the section (@pxref{SECTIONS}).
2003
 
2004
@kindex --warn-shared-textrel
2005
@item --warn-shared-textrel
2006
Warn if the linker adds a DT_TEXTREL to a shared object.
2007
 
2008
@kindex --warn-alternate-em
2009
@item --warn-alternate-em
2010
Warn if an object has alternate ELF machine code.
2011
 
2012
@kindex --warn-unresolved-symbols
2013
@item --warn-unresolved-symbols
2014
If the linker is going to report an unresolved symbol (see the option
2015
@option{--unresolved-symbols}) it will normally generate an error.
2016
This option makes it generate a warning instead.
2017
 
2018
@kindex --error-unresolved-symbols
2019
@item --error-unresolved-symbols
2020
This restores the linker's default behaviour of generating errors when
2021
it is reporting unresolved symbols.
2022
 
2023
@kindex --whole-archive
2024
@cindex including an entire archive
2025
@item --whole-archive
2026
For each archive mentioned on the command line after the
2027
@option{--whole-archive} option, include every object file in the archive
2028
in the link, rather than searching the archive for the required object
2029
files.  This is normally used to turn an archive file into a shared
2030
library, forcing every object to be included in the resulting shared
2031
library.  This option may be used more than once.
2032
 
2033
Two notes when using this option from gcc: First, gcc doesn't know
2034
about this option, so you have to use @option{-Wl,-whole-archive}.
2035
Second, don't forget to use @option{-Wl,-no-whole-archive} after your
2036
list of archives, because gcc will add its own list of archives to
2037
your link and you may not want this flag to affect those as well.
2038
 
2039
@kindex --wrap=@var{symbol}
2040
@item --wrap=@var{symbol}
2041
Use a wrapper function for @var{symbol}.  Any undefined reference to
2042
@var{symbol} will be resolved to @code{__wrap_@var{symbol}}.  Any
2043
undefined reference to @code{__real_@var{symbol}} will be resolved to
2044
@var{symbol}.
2045
 
2046
This can be used to provide a wrapper for a system function.  The
2047
wrapper function should be called @code{__wrap_@var{symbol}}.  If it
2048
wishes to call the system function, it should call
2049
@code{__real_@var{symbol}}.
2050
 
2051
Here is a trivial example:
2052
 
2053
@smallexample
2054
void *
2055
__wrap_malloc (size_t c)
2056
@{
2057
  printf ("malloc called with %zu\n", c);
2058
  return __real_malloc (c);
2059
@}
2060
@end smallexample
2061
 
2062
If you link other code with this file using @option{--wrap malloc}, then
2063
all calls to @code{malloc} will call the function @code{__wrap_malloc}
2064
instead.  The call to @code{__real_malloc} in @code{__wrap_malloc} will
2065
call the real @code{malloc} function.
2066
 
2067
You may wish to provide a @code{__real_malloc} function as well, so that
2068
links without the @option{--wrap} option will succeed.  If you do this,
2069
you should not put the definition of @code{__real_malloc} in the same
2070
file as @code{__wrap_malloc}; if you do, the assembler may resolve the
2071
call before the linker has a chance to wrap it to @code{malloc}.
2072
 
2073
@kindex --eh-frame-hdr
2074
@item --eh-frame-hdr
2075
Request creation of @code{.eh_frame_hdr} section and ELF
2076
@code{PT_GNU_EH_FRAME} segment header.
2077
 
2078 157 khays
@kindex --ld-generated-unwind-info
2079
@item --no-ld-generated-unwind-info
2080
Request creation of @code{.eh_frame} unwind info for linker
2081
generated code sections like PLT.  This option is on by default
2082
if linker generated unwind info is supported.
2083
 
2084 145 khays
@kindex --enable-new-dtags
2085
@kindex --disable-new-dtags
2086
@item --enable-new-dtags
2087
@itemx --disable-new-dtags
2088
This linker can create the new dynamic tags in ELF. But the older ELF
2089
systems may not understand them. If you specify
2090
@option{--enable-new-dtags}, the dynamic tags will be created as needed.
2091
If you specify @option{--disable-new-dtags}, no new dynamic tags will be
2092
created. By default, the new dynamic tags are not created. Note that
2093
those options are only available for ELF systems.
2094
 
2095
@kindex --hash-size=@var{number}
2096
@item --hash-size=@var{number}
2097
Set the default size of the linker's hash tables to a prime number
2098
close to @var{number}.  Increasing this value can reduce the length of
2099
time it takes the linker to perform its tasks, at the expense of
2100
increasing the linker's memory requirements.  Similarly reducing this
2101
value can reduce the memory requirements at the expense of speed.
2102
 
2103
@kindex --hash-style=@var{style}
2104
@item --hash-style=@var{style}
2105
Set the type of linker's hash table(s).  @var{style} can be either
2106
@code{sysv} for classic ELF @code{.hash} section, @code{gnu} for
2107
new style GNU @code{.gnu.hash} section or @code{both} for both
2108
the classic ELF @code{.hash} and new style GNU @code{.gnu.hash}
2109
hash tables.  The default is @code{sysv}.
2110
 
2111
@kindex --reduce-memory-overheads
2112
@item --reduce-memory-overheads
2113
This option reduces memory requirements at ld runtime, at the expense of
2114
linking speed.  This was introduced to select the old O(n^2) algorithm
2115
for link map file generation, rather than the new O(n) algorithm which uses
2116
about 40% more memory for symbol storage.
2117
 
2118
Another effect of the switch is to set the default hash table size to
2119
1021, which again saves memory at the cost of lengthening the linker's
2120
run time.  This is not done however if the @option{--hash-size} switch
2121
has been used.
2122
 
2123
The @option{--reduce-memory-overheads} switch may be also be used to
2124
enable other tradeoffs in future versions of the linker.
2125
 
2126
@kindex --build-id
2127
@kindex --build-id=@var{style}
2128
@item --build-id
2129
@itemx --build-id=@var{style}
2130
Request creation of @code{.note.gnu.build-id} ELF note section.
2131
The contents of the note are unique bits identifying this linked
2132
file.  @var{style} can be @code{uuid} to use 128 random bits,
2133
@code{sha1} to use a 160-bit @sc{SHA1} hash on the normative
2134
parts of the output contents, @code{md5} to use a 128-bit
2135
@sc{MD5} hash on the normative parts of the output contents, or
2136
@code{0x@var{hexstring}} to use a chosen bit string specified as
2137
an even number of hexadecimal digits (@code{-} and @code{:}
2138
characters between digit pairs are ignored).  If @var{style} is
2139
omitted, @code{sha1} is used.
2140
 
2141
The @code{md5} and @code{sha1} styles produces an identifier
2142
that is always the same in an identical output file, but will be
2143
unique among all nonidentical output files.  It is not intended
2144
to be compared as a checksum for the file's contents.  A linked
2145
file may be changed later by other tools, but the build ID bit
2146
string identifying the original linked file does not change.
2147
 
2148
Passing @code{none} for @var{style} disables the setting from any
2149
@code{--build-id} options earlier on the command line.
2150
@end table
2151
 
2152
@c man end
2153
 
2154
@subsection Options Specific to i386 PE Targets
2155
 
2156
@c man begin OPTIONS
2157
 
2158
The i386 PE linker supports the @option{-shared} option, which causes
2159
the output to be a dynamically linked library (DLL) instead of a
2160
normal executable.  You should name the output @code{*.dll} when you
2161
use this option.  In addition, the linker fully supports the standard
2162
@code{*.def} files, which may be specified on the linker command line
2163
like an object file (in fact, it should precede archives it exports
2164
symbols from, to ensure that they get linked in, just like a normal
2165
object file).
2166
 
2167
In addition to the options common to all targets, the i386 PE linker
2168
support additional command line options that are specific to the i386
2169
PE target.  Options that take values may be separated from their
2170
values by either a space or an equals sign.
2171
 
2172
@table @gcctabopt
2173
 
2174
@kindex --add-stdcall-alias
2175
@item --add-stdcall-alias
2176
If given, symbols with a stdcall suffix (@@@var{nn}) will be exported
2177
as-is and also with the suffix stripped.
2178
[This option is specific to the i386 PE targeted port of the linker]
2179
 
2180
@kindex --base-file
2181
@item --base-file @var{file}
2182
Use @var{file} as the name of a file in which to save the base
2183
addresses of all the relocations needed for generating DLLs with
2184
@file{dlltool}.
2185
[This is an i386 PE specific option]
2186
 
2187
@kindex --dll
2188
@item --dll
2189
Create a DLL instead of a regular executable.  You may also use
2190
@option{-shared} or specify a @code{LIBRARY} in a given @code{.def}
2191
file.
2192
[This option is specific to the i386 PE targeted port of the linker]
2193
 
2194
@kindex --enable-long-section-names
2195
@kindex --disable-long-section-names
2196
@item --enable-long-section-names
2197
@itemx --disable-long-section-names
2198
The PE variants of the Coff object format add an extension that permits
2199
the use of section names longer than eight characters, the normal limit
2200
for Coff.  By default, these names are only allowed in object files, as
2201
fully-linked executable images do not carry the Coff string table required
2202
to support the longer names.  As a GNU extension, it is possible to
2203
allow their use in executable images as well, or to (probably pointlessly!)
2204
disallow it in object files, by using these two options.  Executable images
2205
generated with these long section names are slightly non-standard, carrying
2206
as they do a string table, and may generate confusing output when examined
2207
with non-GNU PE-aware tools, such as file viewers and dumpers.  However,
2208
GDB relies on the use of PE long section names to find Dwarf-2 debug
2209
information sections in an executable image at runtime, and so if neither
2210
option is specified on the command-line, @command{ld} will enable long
2211
section names, overriding the default and technically correct behaviour,
2212
when it finds the presence of debug information while linking an executable
2213
image and not stripping symbols.
2214
[This option is valid for all PE targeted ports of the linker]
2215
 
2216
@kindex --enable-stdcall-fixup
2217
@kindex --disable-stdcall-fixup
2218
@item --enable-stdcall-fixup
2219
@itemx --disable-stdcall-fixup
2220
If the link finds a symbol that it cannot resolve, it will attempt to
2221
do ``fuzzy linking'' by looking for another defined symbol that differs
2222
only in the format of the symbol name (cdecl vs stdcall) and will
2223
resolve that symbol by linking to the match.  For example, the
2224
undefined symbol @code{_foo} might be linked to the function
2225
@code{_foo@@12}, or the undefined symbol @code{_bar@@16} might be linked
2226
to the function @code{_bar}.  When the linker does this, it prints a
2227
warning, since it normally should have failed to link, but sometimes
2228
import libraries generated from third-party dlls may need this feature
2229
to be usable.  If you specify @option{--enable-stdcall-fixup}, this
2230
feature is fully enabled and warnings are not printed.  If you specify
2231
@option{--disable-stdcall-fixup}, this feature is disabled and such
2232
mismatches are considered to be errors.
2233
[This option is specific to the i386 PE targeted port of the linker]
2234
 
2235
@kindex --leading-underscore
2236
@kindex --no-leading-underscore
2237
@item --leading-underscore
2238
@itemx --no-leading-underscore
2239
For most targets default symbol-prefix is an underscore and is defined
2240
in target's description. By this option it is possible to
2241
disable/enable the default underscore symbol-prefix.
2242
 
2243
@cindex DLLs, creating
2244
@kindex --export-all-symbols
2245
@item --export-all-symbols
2246
If given, all global symbols in the objects used to build a DLL will
2247
be exported by the DLL.  Note that this is the default if there
2248
otherwise wouldn't be any exported symbols.  When symbols are
2249
explicitly exported via DEF files or implicitly exported via function
2250
attributes, the default is to not export anything else unless this
2251
option is given.  Note that the symbols @code{DllMain@@12},
2252
@code{DllEntryPoint@@0}, @code{DllMainCRTStartup@@12}, and
2253
@code{impure_ptr} will not be automatically
2254
exported.  Also, symbols imported from other DLLs will not be
2255
re-exported, nor will symbols specifying the DLL's internal layout
2256
such as those beginning with @code{_head_} or ending with
2257
@code{_iname}.  In addition, no symbols from @code{libgcc},
2258
@code{libstd++}, @code{libmingw32}, or @code{crtX.o} will be exported.
2259
Symbols whose names begin with @code{__rtti_} or @code{__builtin_} will
2260
not be exported, to help with C++ DLLs.  Finally, there is an
2261
extensive list of cygwin-private symbols that are not exported
2262
(obviously, this applies on when building DLLs for cygwin targets).
2263
These cygwin-excludes are: @code{_cygwin_dll_entry@@12},
2264
@code{_cygwin_crt0_common@@8}, @code{_cygwin_noncygwin_dll_entry@@12},
2265
@code{_fmode}, @code{_impure_ptr}, @code{cygwin_attach_dll},
2266
@code{cygwin_premain0}, @code{cygwin_premain1}, @code{cygwin_premain2},
2267
@code{cygwin_premain3}, and @code{environ}.
2268
[This option is specific to the i386 PE targeted port of the linker]
2269
 
2270
@kindex --exclude-symbols
2271
@item --exclude-symbols @var{symbol},@var{symbol},...
2272
Specifies a list of symbols which should not be automatically
2273
exported.  The symbol names may be delimited by commas or colons.
2274
[This option is specific to the i386 PE targeted port of the linker]
2275
 
2276
@kindex --exclude-all-symbols
2277
@item --exclude-all-symbols
2278
Specifies no symbols should be automatically exported.
2279
[This option is specific to the i386 PE targeted port of the linker]
2280
 
2281
@kindex --file-alignment
2282
@item --file-alignment
2283
Specify the file alignment.  Sections in the file will always begin at
2284
file offsets which are multiples of this number.  This defaults to
2285
512.
2286
[This option is specific to the i386 PE targeted port of the linker]
2287
 
2288
@cindex heap size
2289
@kindex --heap
2290
@item --heap @var{reserve}
2291
@itemx --heap @var{reserve},@var{commit}
2292
Specify the number of bytes of memory to reserve (and optionally commit)
2293
to be used as heap for this program.  The default is 1Mb reserved, 4K
2294
committed.
2295
[This option is specific to the i386 PE targeted port of the linker]
2296
 
2297
@cindex image base
2298
@kindex --image-base
2299
@item --image-base @var{value}
2300
Use @var{value} as the base address of your program or dll.  This is
2301
the lowest memory location that will be used when your program or dll
2302
is loaded.  To reduce the need to relocate and improve performance of
2303
your dlls, each should have a unique base address and not overlap any
2304
other dlls.  The default is 0x400000 for executables, and 0x10000000
2305
for dlls.
2306
[This option is specific to the i386 PE targeted port of the linker]
2307
 
2308
@kindex --kill-at
2309
@item --kill-at
2310
If given, the stdcall suffixes (@@@var{nn}) will be stripped from
2311
symbols before they are exported.
2312
[This option is specific to the i386 PE targeted port of the linker]
2313
 
2314
@kindex --large-address-aware
2315
@item --large-address-aware
2316
If given, the appropriate bit in the ``Characteristics'' field of the COFF
2317
header is set to indicate that this executable supports virtual addresses
2318
greater than 2 gigabytes.  This should be used in conjunction with the /3GB
2319
or /USERVA=@var{value} megabytes switch in the ``[operating systems]''
2320
section of the BOOT.INI.  Otherwise, this bit has no effect.
2321
[This option is specific to PE targeted ports of the linker]
2322
 
2323
@kindex --major-image-version
2324
@item --major-image-version @var{value}
2325
Sets the major number of the ``image version''.  Defaults to 1.
2326
[This option is specific to the i386 PE targeted port of the linker]
2327
 
2328
@kindex --major-os-version
2329
@item --major-os-version @var{value}
2330
Sets the major number of the ``os version''.  Defaults to 4.
2331
[This option is specific to the i386 PE targeted port of the linker]
2332
 
2333
@kindex --major-subsystem-version
2334
@item --major-subsystem-version @var{value}
2335
Sets the major number of the ``subsystem version''.  Defaults to 4.
2336
[This option is specific to the i386 PE targeted port of the linker]
2337
 
2338
@kindex --minor-image-version
2339
@item --minor-image-version @var{value}
2340
Sets the minor number of the ``image version''.  Defaults to 0.
2341
[This option is specific to the i386 PE targeted port of the linker]
2342
 
2343
@kindex --minor-os-version
2344
@item --minor-os-version @var{value}
2345
Sets the minor number of the ``os version''.  Defaults to 0.
2346
[This option is specific to the i386 PE targeted port of the linker]
2347
 
2348
@kindex --minor-subsystem-version
2349
@item --minor-subsystem-version @var{value}
2350
Sets the minor number of the ``subsystem version''.  Defaults to 0.
2351
[This option is specific to the i386 PE targeted port of the linker]
2352
 
2353
@cindex DEF files, creating
2354
@cindex DLLs, creating
2355
@kindex --output-def
2356
@item --output-def @var{file}
2357
The linker will create the file @var{file} which will contain a DEF
2358
file corresponding to the DLL the linker is generating.  This DEF file
2359
(which should be called @code{*.def}) may be used to create an import
2360
library with @code{dlltool} or may be used as a reference to
2361
automatically or implicitly exported symbols.
2362
[This option is specific to the i386 PE targeted port of the linker]
2363
 
2364
@cindex DLLs, creating
2365
@kindex --out-implib
2366
@item --out-implib @var{file}
2367
The linker will create the file @var{file} which will contain an
2368
import lib corresponding to the DLL the linker is generating. This
2369
import lib (which should be called @code{*.dll.a} or @code{*.a}
2370
may be used to link clients against the generated DLL; this behaviour
2371
makes it possible to skip a separate @code{dlltool} import library
2372
creation step.
2373
[This option is specific to the i386 PE targeted port of the linker]
2374
 
2375
@kindex --enable-auto-image-base
2376
@item --enable-auto-image-base
2377
Automatically choose the image base for DLLs, unless one is specified
2378
using the @code{--image-base} argument.  By using a hash generated
2379
from the dllname to create unique image bases for each DLL, in-memory
2380
collisions and relocations which can delay program execution are
2381
avoided.
2382
[This option is specific to the i386 PE targeted port of the linker]
2383
 
2384
@kindex --disable-auto-image-base
2385
@item --disable-auto-image-base
2386
Do not automatically generate a unique image base.  If there is no
2387
user-specified image base (@code{--image-base}) then use the platform
2388
default.
2389
[This option is specific to the i386 PE targeted port of the linker]
2390
 
2391
@cindex DLLs, linking to
2392
@kindex --dll-search-prefix
2393
@item --dll-search-prefix @var{string}
2394
When linking dynamically to a dll without an import library,
2395
search for @code{<string><basename>.dll} in preference to
2396
@code{lib<basename>.dll}. This behaviour allows easy distinction
2397
between DLLs built for the various "subplatforms": native, cygwin,
2398
uwin, pw, etc.  For instance, cygwin DLLs typically use
2399
@code{--dll-search-prefix=cyg}.
2400
[This option is specific to the i386 PE targeted port of the linker]
2401
 
2402
@kindex --enable-auto-import
2403
@item --enable-auto-import
2404
Do sophisticated linking of @code{_symbol} to @code{__imp__symbol} for
2405
DATA imports from DLLs, and create the necessary thunking symbols when
2406
building the import libraries with those DATA exports. Note: Use of the
2407
'auto-import' extension will cause the text section of the image file
2408
to be made writable. This does not conform to the PE-COFF format
2409
specification published by Microsoft.
2410
 
2411
Note - use of the 'auto-import' extension will also cause read only
2412
data which would normally be placed into the .rdata section to be
2413
placed into the .data section instead.  This is in order to work
2414
around a problem with consts that is described here:
2415
http://www.cygwin.com/ml/cygwin/2004-09/msg01101.html
2416
 
2417
Using 'auto-import' generally will 'just work' -- but sometimes you may
2418
see this message:
2419
 
2420
"variable '<var>' can't be auto-imported. Please read the
2421
documentation for ld's @code{--enable-auto-import} for details."
2422
 
2423
This message occurs when some (sub)expression accesses an address
2424
ultimately given by the sum of two constants (Win32 import tables only
2425
allow one).  Instances where this may occur include accesses to member
2426
fields of struct variables imported from a DLL, as well as using a
2427
constant index into an array variable imported from a DLL.  Any
2428
multiword variable (arrays, structs, long long, etc) may trigger
2429
this error condition.  However, regardless of the exact data type
2430
of the offending exported variable, ld will always detect it, issue
2431
the warning, and exit.
2432
 
2433
There are several ways to address this difficulty, regardless of the
2434
data type of the exported variable:
2435
 
2436
One way is to use --enable-runtime-pseudo-reloc switch. This leaves the task
2437
of adjusting references in your client code for runtime environment, so
2438
this method works only when runtime environment supports this feature.
2439
 
2440
A second solution is to force one of the 'constants' to be a variable --
2441
that is, unknown and un-optimizable at compile time.  For arrays,
2442
there are two possibilities: a) make the indexee (the array's address)
2443
a variable, or b) make the 'constant' index a variable.  Thus:
2444
 
2445
@example
2446
extern type extern_array[];
2447
extern_array[1] -->
2448
   @{ volatile type *t=extern_array; t[1] @}
2449
@end example
2450
 
2451
or
2452
 
2453
@example
2454
extern type extern_array[];
2455
extern_array[1] -->
2456
   @{ volatile int t=1; extern_array[t] @}
2457
@end example
2458
 
2459
For structs (and most other multiword data types) the only option
2460
is to make the struct itself (or the long long, or the ...) variable:
2461
 
2462
@example
2463
extern struct s extern_struct;
2464
extern_struct.field -->
2465
   @{ volatile struct s *t=&extern_struct; t->field @}
2466
@end example
2467
 
2468
or
2469
 
2470
@example
2471
extern long long extern_ll;
2472
extern_ll -->
2473
  @{ volatile long long * local_ll=&extern_ll; *local_ll @}
2474
@end example
2475
 
2476
A third method of dealing with this difficulty is to abandon
2477
'auto-import' for the offending symbol and mark it with
2478
@code{__declspec(dllimport)}.  However, in practise that
2479
requires using compile-time #defines to indicate whether you are
2480
building a DLL, building client code that will link to the DLL, or
2481
merely building/linking to a static library.   In making the choice
2482
between the various methods of resolving the 'direct address with
2483
constant offset' problem, you should consider typical real-world usage:
2484
 
2485
Original:
2486
@example
2487
--foo.h
2488
extern int arr[];
2489
--foo.c
2490
#include "foo.h"
2491
void main(int argc, char **argv)@{
2492
  printf("%d\n",arr[1]);
2493
@}
2494
@end example
2495
 
2496
Solution 1:
2497
@example
2498
--foo.h
2499
extern int arr[];
2500
--foo.c
2501
#include "foo.h"
2502
void main(int argc, char **argv)@{
2503
  /* This workaround is for win32 and cygwin; do not "optimize" */
2504
  volatile int *parr = arr;
2505
  printf("%d\n",parr[1]);
2506
@}
2507
@end example
2508
 
2509
Solution 2:
2510
@example
2511
--foo.h
2512
/* Note: auto-export is assumed (no __declspec(dllexport)) */
2513
#if (defined(_WIN32) || defined(__CYGWIN__)) && \
2514
  !(defined(FOO_BUILD_DLL) || defined(FOO_STATIC))
2515
#define FOO_IMPORT __declspec(dllimport)
2516
#else
2517
#define FOO_IMPORT
2518
#endif
2519
extern FOO_IMPORT int arr[];
2520
--foo.c
2521
#include "foo.h"
2522
void main(int argc, char **argv)@{
2523
  printf("%d\n",arr[1]);
2524
@}
2525
@end example
2526
 
2527
A fourth way to avoid this problem is to re-code your
2528
library to use a functional interface rather than a data interface
2529
for the offending variables (e.g. set_foo() and get_foo() accessor
2530
functions).
2531
[This option is specific to the i386 PE targeted port of the linker]
2532
 
2533
@kindex --disable-auto-import
2534
@item --disable-auto-import
2535
Do not attempt to do sophisticated linking of @code{_symbol} to
2536
@code{__imp__symbol} for DATA imports from DLLs.
2537
[This option is specific to the i386 PE targeted port of the linker]
2538
 
2539
@kindex --enable-runtime-pseudo-reloc
2540
@item --enable-runtime-pseudo-reloc
2541
If your code contains expressions described in --enable-auto-import section,
2542
that is, DATA imports from DLL with non-zero offset, this switch will create
2543
a vector of 'runtime pseudo relocations' which can be used by runtime
2544
environment to adjust references to such data in your client code.
2545
[This option is specific to the i386 PE targeted port of the linker]
2546
 
2547
@kindex --disable-runtime-pseudo-reloc
2548
@item --disable-runtime-pseudo-reloc
2549
Do not create pseudo relocations for non-zero offset DATA imports from
2550
DLLs.  This is the default.
2551
[This option is specific to the i386 PE targeted port of the linker]
2552
 
2553
@kindex --enable-extra-pe-debug
2554
@item --enable-extra-pe-debug
2555
Show additional debug info related to auto-import symbol thunking.
2556
[This option is specific to the i386 PE targeted port of the linker]
2557
 
2558
@kindex --section-alignment
2559
@item --section-alignment
2560
Sets the section alignment.  Sections in memory will always begin at
2561
addresses which are a multiple of this number.  Defaults to 0x1000.
2562
[This option is specific to the i386 PE targeted port of the linker]
2563
 
2564
@cindex stack size
2565
@kindex --stack
2566
@item --stack @var{reserve}
2567
@itemx --stack @var{reserve},@var{commit}
2568
Specify the number of bytes of memory to reserve (and optionally commit)
2569
to be used as stack for this program.  The default is 2Mb reserved, 4K
2570
committed.
2571
[This option is specific to the i386 PE targeted port of the linker]
2572
 
2573
@kindex --subsystem
2574
@item --subsystem @var{which}
2575
@itemx --subsystem @var{which}:@var{major}
2576
@itemx --subsystem @var{which}:@var{major}.@var{minor}
2577
Specifies the subsystem under which your program will execute.  The
2578
legal values for @var{which} are @code{native}, @code{windows},
2579
@code{console}, @code{posix}, and @code{xbox}.  You may optionally set
2580
the subsystem version also.  Numeric values are also accepted for
2581
@var{which}.
2582
[This option is specific to the i386 PE targeted port of the linker]
2583
 
2584
The following options set flags in the @code{DllCharacteristics} field
2585
of the PE file header:
2586
[These options are specific to PE targeted ports of the linker]
2587
 
2588
@kindex --dynamicbase
2589
@item --dynamicbase
2590
The image base address may be relocated using address space layout
2591
randomization (ASLR).  This feature was introduced with MS Windows
2592
Vista for i386 PE targets.
2593
 
2594
@kindex --forceinteg
2595
@item --forceinteg
2596
Code integrity checks are enforced.
2597
 
2598
@kindex --nxcompat
2599
@item --nxcompat
2600
The image is compatible with the Data Execution Prevention.
2601
This feature was introduced with MS Windows XP SP2 for i386 PE targets.
2602
 
2603
@kindex --no-isolation
2604
@item --no-isolation
2605
Although the image understands isolation, do not isolate the image.
2606
 
2607
@kindex --no-seh
2608
@item --no-seh
2609
The image does not use SEH. No SE handler may be called from
2610
this image.
2611
 
2612
@kindex --no-bind
2613
@item --no-bind
2614
Do not bind this image.
2615
 
2616
@kindex --wdmdriver
2617
@item --wdmdriver
2618
The driver uses the MS Windows Driver Model.
2619
 
2620
@kindex --tsaware
2621
@item --tsaware
2622
The image is Terminal Server aware.
2623
 
2624
@end table
2625
 
2626
@c man end
2627
 
2628
@ifset C6X
2629
@subsection Options specific to C6X uClinux targets
2630
 
2631
@c man begin OPTIONS
2632
 
2633
The C6X uClinux target uses a binary format called DSBT to support shared
2634
libraries.  Each shared library in the system needs to have a unique index;
2635
all executables use an index of 0.
2636
 
2637
@table @gcctabopt
2638
 
2639
@kindex --dsbt-size
2640
@item --dsbt-size @var{size}
2641
This option sets the number of entires in the DSBT of the current executable
2642
or shared library to @var{size}.  The default is to create a table with 64
2643
entries.
2644
 
2645
@kindex --dsbt-index
2646
@item --dsbt-index @var{index}
2647
This option sets the DSBT index of the current executable or shared library
2648
to @var{index}.  The default is 0, which is appropriate for generating
2649
executables.  If a shared library is generated with a DSBT index of 0, the
2650
@code{R_C6000_DSBT_INDEX} relocs are copied into the output file.
2651
 
2652
@kindex --no-merge-exidx-entries
2653
The @samp{--no-merge-exidx-entries} switch disables the merging of adjacent
2654
exidx entries in frame unwind info.
2655
 
2656
@end table
2657
 
2658
@c man end
2659
@end ifset
2660
 
2661
@ifset M68HC11
2662
@subsection Options specific to Motorola 68HC11 and 68HC12 targets
2663
 
2664
@c man begin OPTIONS
2665
 
2666
The 68HC11 and 68HC12 linkers support specific options to control the
2667
memory bank switching mapping and trampoline code generation.
2668
 
2669
@table @gcctabopt
2670
 
2671
@kindex --no-trampoline
2672
@item --no-trampoline
2673
This option disables the generation of trampoline. By default a trampoline
2674
is generated for each far function which is called using a @code{jsr}
2675
instruction (this happens when a pointer to a far function is taken).
2676
 
2677
@kindex --bank-window
2678
@item --bank-window @var{name}
2679
This option indicates to the linker the name of the memory region in
2680
the @samp{MEMORY} specification that describes the memory bank window.
2681
The definition of such region is then used by the linker to compute
2682
paging and addresses within the memory window.
2683
 
2684
@end table
2685
 
2686
@c man end
2687
@end ifset
2688
 
2689
@ifset M68K
2690
@subsection Options specific to Motorola 68K target
2691
 
2692
@c man begin OPTIONS
2693
 
2694
The following options are supported to control handling of GOT generation
2695
when linking for 68K targets.
2696
 
2697
@table @gcctabopt
2698
 
2699
@kindex --got
2700
@item --got=@var{type}
2701
This option tells the linker which GOT generation scheme to use.
2702
@var{type} should be one of @samp{single}, @samp{negative},
2703
@samp{multigot} or @samp{target}.  For more information refer to the
2704
Info entry for @file{ld}.
2705
 
2706
@end table
2707
 
2708
@c man end
2709
@end ifset
2710
 
2711
@ifset UsesEnvVars
2712
@node Environment
2713
@section Environment Variables
2714
 
2715
@c man begin ENVIRONMENT
2716
 
2717
You can change the behaviour of @command{ld} with the environment variables
2718
@ifclear SingleFormat
2719
@code{GNUTARGET},
2720
@end ifclear
2721
@code{LDEMULATION} and @code{COLLECT_NO_DEMANGLE}.
2722
 
2723
@ifclear SingleFormat
2724
@kindex GNUTARGET
2725
@cindex default input format
2726
@code{GNUTARGET} determines the input-file object format if you don't
2727
use @samp{-b} (or its synonym @samp{--format}).  Its value should be one
2728
of the BFD names for an input format (@pxref{BFD}).  If there is no
2729
@code{GNUTARGET} in the environment, @command{ld} uses the natural format
2730
of the target. If @code{GNUTARGET} is set to @code{default} then BFD
2731
attempts to discover the input format by examining binary input files;
2732
this method often succeeds, but there are potential ambiguities, since
2733
there is no method of ensuring that the magic number used to specify
2734
object-file formats is unique.  However, the configuration procedure for
2735
BFD on each system places the conventional format for that system first
2736
in the search-list, so ambiguities are resolved in favor of convention.
2737
@end ifclear
2738
 
2739
@kindex LDEMULATION
2740
@cindex default emulation
2741
@cindex emulation, default
2742
@code{LDEMULATION} determines the default emulation if you don't use the
2743
@samp{-m} option.  The emulation can affect various aspects of linker
2744
behaviour, particularly the default linker script.  You can list the
2745
available emulations with the @samp{--verbose} or @samp{-V} options.  If
2746
the @samp{-m} option is not used, and the @code{LDEMULATION} environment
2747
variable is not defined, the default emulation depends upon how the
2748
linker was configured.
2749
 
2750
@kindex COLLECT_NO_DEMANGLE
2751
@cindex demangling, default
2752
Normally, the linker will default to demangling symbols.  However, if
2753
@code{COLLECT_NO_DEMANGLE} is set in the environment, then it will
2754
default to not demangling symbols.  This environment variable is used in
2755
a similar fashion by the @code{gcc} linker wrapper program.  The default
2756
may be overridden by the @samp{--demangle} and @samp{--no-demangle}
2757
options.
2758
 
2759
@c man end
2760
@end ifset
2761
 
2762
@node Scripts
2763
@chapter Linker Scripts
2764
 
2765
@cindex scripts
2766
@cindex linker scripts
2767
@cindex command files
2768
Every link is controlled by a @dfn{linker script}.  This script is
2769
written in the linker command language.
2770
 
2771
The main purpose of the linker script is to describe how the sections in
2772
the input files should be mapped into the output file, and to control
2773
the memory layout of the output file.  Most linker scripts do nothing
2774
more than this.  However, when necessary, the linker script can also
2775
direct the linker to perform many other operations, using the commands
2776
described below.
2777
 
2778
The linker always uses a linker script.  If you do not supply one
2779
yourself, the linker will use a default script that is compiled into the
2780
linker executable.  You can use the @samp{--verbose} command line option
2781
to display the default linker script.  Certain command line options,
2782
such as @samp{-r} or @samp{-N}, will affect the default linker script.
2783
 
2784
You may supply your own linker script by using the @samp{-T} command
2785
line option.  When you do this, your linker script will replace the
2786
default linker script.
2787
 
2788
You may also use linker scripts implicitly by naming them as input files
2789
to the linker, as though they were files to be linked.  @xref{Implicit
2790
Linker Scripts}.
2791
 
2792
@menu
2793
* Basic Script Concepts::       Basic Linker Script Concepts
2794
* Script Format::               Linker Script Format
2795
* Simple Example::              Simple Linker Script Example
2796
* Simple Commands::             Simple Linker Script Commands
2797
* Assignments::                 Assigning Values to Symbols
2798
* SECTIONS::                    SECTIONS Command
2799
* MEMORY::                      MEMORY Command
2800
* PHDRS::                       PHDRS Command
2801
* VERSION::                     VERSION Command
2802
* Expressions::                 Expressions in Linker Scripts
2803
* Implicit Linker Scripts::     Implicit Linker Scripts
2804
@end menu
2805
 
2806
@node Basic Script Concepts
2807
@section Basic Linker Script Concepts
2808
@cindex linker script concepts
2809
We need to define some basic concepts and vocabulary in order to
2810
describe the linker script language.
2811
 
2812
The linker combines input files into a single output file.  The output
2813
file and each input file are in a special data format known as an
2814
@dfn{object file format}.  Each file is called an @dfn{object file}.
2815
The output file is often called an @dfn{executable}, but for our
2816
purposes we will also call it an object file.  Each object file has,
2817
among other things, a list of @dfn{sections}.  We sometimes refer to a
2818
section in an input file as an @dfn{input section}; similarly, a section
2819
in the output file is an @dfn{output section}.
2820
 
2821
Each section in an object file has a name and a size.  Most sections
2822
also have an associated block of data, known as the @dfn{section
2823
contents}.  A section may be marked as @dfn{loadable}, which mean that
2824
the contents should be loaded into memory when the output file is run.
2825
A section with no contents may be @dfn{allocatable}, which means that an
2826
area in memory should be set aside, but nothing in particular should be
2827
loaded there (in some cases this memory must be zeroed out).  A section
2828
which is neither loadable nor allocatable typically contains some sort
2829
of debugging information.
2830
 
2831
Every loadable or allocatable output section has two addresses.  The
2832
first is the @dfn{VMA}, or virtual memory address.  This is the address
2833
the section will have when the output file is run.  The second is the
2834
@dfn{LMA}, or load memory address.  This is the address at which the
2835
section will be loaded.  In most cases the two addresses will be the
2836
same.  An example of when they might be different is when a data section
2837
is loaded into ROM, and then copied into RAM when the program starts up
2838
(this technique is often used to initialize global variables in a ROM
2839
based system).  In this case the ROM address would be the LMA, and the
2840
RAM address would be the VMA.
2841
 
2842
You can see the sections in an object file by using the @code{objdump}
2843
program with the @samp{-h} option.
2844
 
2845
Every object file also has a list of @dfn{symbols}, known as the
2846
@dfn{symbol table}.  A symbol may be defined or undefined.  Each symbol
2847
has a name, and each defined symbol has an address, among other
2848
information.  If you compile a C or C++ program into an object file, you
2849
will get a defined symbol for every defined function and global or
2850
static variable.  Every undefined function or global variable which is
2851
referenced in the input file will become an undefined symbol.
2852
 
2853
You can see the symbols in an object file by using the @code{nm}
2854
program, or by using the @code{objdump} program with the @samp{-t}
2855
option.
2856
 
2857
@node Script Format
2858
@section Linker Script Format
2859
@cindex linker script format
2860
Linker scripts are text files.
2861
 
2862
You write a linker script as a series of commands.  Each command is
2863
either a keyword, possibly followed by arguments, or an assignment to a
2864
symbol.  You may separate commands using semicolons.  Whitespace is
2865
generally ignored.
2866
 
2867
Strings such as file or format names can normally be entered directly.
2868
If the file name contains a character such as a comma which would
2869
otherwise serve to separate file names, you may put the file name in
2870
double quotes.  There is no way to use a double quote character in a
2871
file name.
2872
 
2873
You may include comments in linker scripts just as in C, delimited by
2874
@samp{/*} and @samp{*/}.  As in C, comments are syntactically equivalent
2875
to whitespace.
2876
 
2877
@node Simple Example
2878
@section Simple Linker Script Example
2879
@cindex linker script example
2880
@cindex example of linker script
2881
Many linker scripts are fairly simple.
2882
 
2883
The simplest possible linker script has just one command:
2884
@samp{SECTIONS}.  You use the @samp{SECTIONS} command to describe the
2885
memory layout of the output file.
2886
 
2887
The @samp{SECTIONS} command is a powerful command.  Here we will
2888
describe a simple use of it.  Let's assume your program consists only of
2889
code, initialized data, and uninitialized data.  These will be in the
2890
@samp{.text}, @samp{.data}, and @samp{.bss} sections, respectively.
2891
Let's assume further that these are the only sections which appear in
2892
your input files.
2893
 
2894
For this example, let's say that the code should be loaded at address
2895
0x10000, and that the data should start at address 0x8000000.  Here is a
2896
linker script which will do that:
2897
@smallexample
2898
SECTIONS
2899
@{
2900
  . = 0x10000;
2901
  .text : @{ *(.text) @}
2902
  . = 0x8000000;
2903
  .data : @{ *(.data) @}
2904
  .bss : @{ *(.bss) @}
2905
@}
2906
@end smallexample
2907
 
2908
You write the @samp{SECTIONS} command as the keyword @samp{SECTIONS},
2909
followed by a series of symbol assignments and output section
2910
descriptions enclosed in curly braces.
2911
 
2912
The first line inside the @samp{SECTIONS} command of the above example
2913
sets the value of the special symbol @samp{.}, which is the location
2914
counter.  If you do not specify the address of an output section in some
2915
other way (other ways are described later), the address is set from the
2916
current value of the location counter.  The location counter is then
2917
incremented by the size of the output section.  At the start of the
2918
@samp{SECTIONS} command, the location counter has the value @samp{0}.
2919
 
2920
The second line defines an output section, @samp{.text}.  The colon is
2921
required syntax which may be ignored for now.  Within the curly braces
2922
after the output section name, you list the names of the input sections
2923
which should be placed into this output section.  The @samp{*} is a
2924
wildcard which matches any file name.  The expression @samp{*(.text)}
2925
means all @samp{.text} input sections in all input files.
2926
 
2927
Since the location counter is @samp{0x10000} when the output section
2928
@samp{.text} is defined, the linker will set the address of the
2929
@samp{.text} section in the output file to be @samp{0x10000}.
2930
 
2931
The remaining lines define the @samp{.data} and @samp{.bss} sections in
2932
the output file.  The linker will place the @samp{.data} output section
2933
at address @samp{0x8000000}.  After the linker places the @samp{.data}
2934
output section, the value of the location counter will be
2935
@samp{0x8000000} plus the size of the @samp{.data} output section.  The
2936
effect is that the linker will place the @samp{.bss} output section
2937
immediately after the @samp{.data} output section in memory.
2938
 
2939
The linker will ensure that each output section has the required
2940
alignment, by increasing the location counter if necessary.  In this
2941
example, the specified addresses for the @samp{.text} and @samp{.data}
2942
sections will probably satisfy any alignment constraints, but the linker
2943
may have to create a small gap between the @samp{.data} and @samp{.bss}
2944
sections.
2945
 
2946
That's it!  That's a simple and complete linker script.
2947
 
2948
@node Simple Commands
2949
@section Simple Linker Script Commands
2950
@cindex linker script simple commands
2951
In this section we describe the simple linker script commands.
2952
 
2953
@menu
2954
* Entry Point::                 Setting the entry point
2955
* File Commands::               Commands dealing with files
2956
@ifclear SingleFormat
2957
* Format Commands::             Commands dealing with object file formats
2958
@end ifclear
2959
 
2960
* REGION_ALIAS::                Assign alias names to memory regions
2961
* Miscellaneous Commands::      Other linker script commands
2962
@end menu
2963
 
2964
@node Entry Point
2965
@subsection Setting the Entry Point
2966
@kindex ENTRY(@var{symbol})
2967
@cindex start of execution
2968
@cindex first instruction
2969
@cindex entry point
2970
The first instruction to execute in a program is called the @dfn{entry
2971
point}.  You can use the @code{ENTRY} linker script command to set the
2972
entry point.  The argument is a symbol name:
2973
@smallexample
2974
ENTRY(@var{symbol})
2975
@end smallexample
2976
 
2977
There are several ways to set the entry point.  The linker will set the
2978
entry point by trying each of the following methods in order, and
2979
stopping when one of them succeeds:
2980
@itemize @bullet
2981
@item
2982
the @samp{-e} @var{entry} command-line option;
2983
@item
2984
the @code{ENTRY(@var{symbol})} command in a linker script;
2985
@item
2986
the value of a target specific symbol, if it is defined;  For many
2987
targets this is @code{start}, but PE and BeOS based systems for example
2988
check a list of possible entry symbols, matching the first one found.
2989
@item
2990
the address of the first byte of the @samp{.text} section, if present;
2991
@item
2992
The address @code{0}.
2993
@end itemize
2994
 
2995
@node File Commands
2996
@subsection Commands Dealing with Files
2997
@cindex linker script file commands
2998
Several linker script commands deal with files.
2999
 
3000
@table @code
3001
@item INCLUDE @var{filename}
3002
@kindex INCLUDE @var{filename}
3003
@cindex including a linker script
3004
Include the linker script @var{filename} at this point.  The file will
3005
be searched for in the current directory, and in any directory specified
3006
with the @option{-L} option.  You can nest calls to @code{INCLUDE} up to
3007
10 levels deep.
3008
 
3009
You can place @code{INCLUDE} directives at the top level, in @code{MEMORY} or
3010
@code{SECTIONS} commands, or in output section descriptions.
3011
 
3012
@item INPUT(@var{file}, @var{file}, @dots{})
3013
@itemx INPUT(@var{file} @var{file} @dots{})
3014
@kindex INPUT(@var{files})
3015
@cindex input files in linker scripts
3016
@cindex input object files in linker scripts
3017
@cindex linker script input object files
3018
The @code{INPUT} command directs the linker to include the named files
3019
in the link, as though they were named on the command line.
3020
 
3021
For example, if you always want to include @file{subr.o} any time you do
3022
a link, but you can't be bothered to put it on every link command line,
3023
then you can put @samp{INPUT (subr.o)} in your linker script.
3024
 
3025
In fact, if you like, you can list all of your input files in the linker
3026
script, and then invoke the linker with nothing but a @samp{-T} option.
3027
 
3028
In case a @dfn{sysroot prefix} is configured, and the filename starts
3029
with the @samp{/} character, and the script being processed was
3030
located inside the @dfn{sysroot prefix}, the filename will be looked
3031
for in the @dfn{sysroot prefix}.  Otherwise, the linker will try to
3032
open the file in the current directory.  If it is not found, the
3033
linker will search through the archive library search path.  See the
3034
description of @samp{-L} in @ref{Options,,Command Line Options}.
3035
 
3036
If you use @samp{INPUT (-l@var{file})}, @command{ld} will transform the
3037
name to @code{lib@var{file}.a}, as with the command line argument
3038
@samp{-l}.
3039
 
3040
When you use the @code{INPUT} command in an implicit linker script, the
3041
files will be included in the link at the point at which the linker
3042
script file is included.  This can affect archive searching.
3043
 
3044
@item GROUP(@var{file}, @var{file}, @dots{})
3045
@itemx GROUP(@var{file} @var{file} @dots{})
3046
@kindex GROUP(@var{files})
3047
@cindex grouping input files
3048
The @code{GROUP} command is like @code{INPUT}, except that the named
3049
files should all be archives, and they are searched repeatedly until no
3050
new undefined references are created.  See the description of @samp{-(}
3051
in @ref{Options,,Command Line Options}.
3052
 
3053
@item AS_NEEDED(@var{file}, @var{file}, @dots{})
3054
@itemx AS_NEEDED(@var{file} @var{file} @dots{})
3055
@kindex AS_NEEDED(@var{files})
3056
This construct can appear only inside of the @code{INPUT} or @code{GROUP}
3057
commands, among other filenames.  The files listed will be handled
3058
as if they appear directly in the @code{INPUT} or @code{GROUP} commands,
3059
with the exception of ELF shared libraries, that will be added only
3060
when they are actually needed.  This construct essentially enables
3061
@option{--as-needed} option for all the files listed inside of it
3062
and restores previous @option{--as-needed} resp. @option{--no-as-needed}
3063
setting afterwards.
3064
 
3065
@item OUTPUT(@var{filename})
3066
@kindex OUTPUT(@var{filename})
3067
@cindex output file name in linker script
3068
The @code{OUTPUT} command names the output file.  Using
3069
@code{OUTPUT(@var{filename})} in the linker script is exactly like using
3070
@samp{-o @var{filename}} on the command line (@pxref{Options,,Command
3071
Line Options}).  If both are used, the command line option takes
3072
precedence.
3073
 
3074
You can use the @code{OUTPUT} command to define a default name for the
3075
output file other than the usual default of @file{a.out}.
3076
 
3077
@item SEARCH_DIR(@var{path})
3078
@kindex SEARCH_DIR(@var{path})
3079
@cindex library search path in linker script
3080
@cindex archive search path in linker script
3081
@cindex search path in linker script
3082
The @code{SEARCH_DIR} command adds @var{path} to the list of paths where
3083
@command{ld} looks for archive libraries.  Using
3084
@code{SEARCH_DIR(@var{path})} is exactly like using @samp{-L @var{path}}
3085
on the command line (@pxref{Options,,Command Line Options}).  If both
3086
are used, then the linker will search both paths.  Paths specified using
3087
the command line option are searched first.
3088
 
3089
@item STARTUP(@var{filename})
3090
@kindex STARTUP(@var{filename})
3091
@cindex first input file
3092
The @code{STARTUP} command is just like the @code{INPUT} command, except
3093
that @var{filename} will become the first input file to be linked, as
3094
though it were specified first on the command line.  This may be useful
3095
when using a system in which the entry point is always the start of the
3096
first file.
3097
@end table
3098
 
3099
@ifclear SingleFormat
3100
@node Format Commands
3101
@subsection Commands Dealing with Object File Formats
3102
A couple of linker script commands deal with object file formats.
3103
 
3104
@table @code
3105
@item OUTPUT_FORMAT(@var{bfdname})
3106
@itemx OUTPUT_FORMAT(@var{default}, @var{big}, @var{little})
3107
@kindex OUTPUT_FORMAT(@var{bfdname})
3108
@cindex output file format in linker script
3109
The @code{OUTPUT_FORMAT} command names the BFD format to use for the
3110
output file (@pxref{BFD}).  Using @code{OUTPUT_FORMAT(@var{bfdname})} is
3111
exactly like using @samp{--oformat @var{bfdname}} on the command line
3112
(@pxref{Options,,Command Line Options}).  If both are used, the command
3113
line option takes precedence.
3114
 
3115
You can use @code{OUTPUT_FORMAT} with three arguments to use different
3116
formats based on the @samp{-EB} and @samp{-EL} command line options.
3117
This permits the linker script to set the output format based on the
3118
desired endianness.
3119
 
3120
If neither @samp{-EB} nor @samp{-EL} are used, then the output format
3121
will be the first argument, @var{default}.  If @samp{-EB} is used, the
3122
output format will be the second argument, @var{big}.  If @samp{-EL} is
3123
used, the output format will be the third argument, @var{little}.
3124
 
3125
For example, the default linker script for the MIPS ELF target uses this
3126
command:
3127
@smallexample
3128
OUTPUT_FORMAT(elf32-bigmips, elf32-bigmips, elf32-littlemips)
3129
@end smallexample
3130
This says that the default format for the output file is
3131
@samp{elf32-bigmips}, but if the user uses the @samp{-EL} command line
3132
option, the output file will be created in the @samp{elf32-littlemips}
3133
format.
3134
 
3135
@item TARGET(@var{bfdname})
3136
@kindex TARGET(@var{bfdname})
3137
@cindex input file format in linker script
3138
The @code{TARGET} command names the BFD format to use when reading input
3139
files.  It affects subsequent @code{INPUT} and @code{GROUP} commands.
3140
This command is like using @samp{-b @var{bfdname}} on the command line
3141
(@pxref{Options,,Command Line Options}).  If the @code{TARGET} command
3142
is used but @code{OUTPUT_FORMAT} is not, then the last @code{TARGET}
3143
command is also used to set the format for the output file.  @xref{BFD}.
3144
@end table
3145
@end ifclear
3146
 
3147
@node REGION_ALIAS
3148
@subsection Assign alias names to memory regions
3149
@kindex REGION_ALIAS(@var{alias}, @var{region})
3150
@cindex region alias
3151
@cindex region names
3152
 
3153
Alias names can be added to existing memory regions created with the
3154
@ref{MEMORY} command.  Each name corresponds to at most one memory region.
3155
 
3156
@smallexample
3157
REGION_ALIAS(@var{alias}, @var{region})
3158
@end smallexample
3159
 
3160
The @code{REGION_ALIAS} function creates an alias name @var{alias} for the
3161
memory region @var{region}.  This allows a flexible mapping of output sections
3162
to memory regions.  An example follows.
3163
 
3164
Suppose we have an application for embedded systems which come with various
3165
memory storage devices.  All have a general purpose, volatile memory @code{RAM}
3166
that allows code execution or data storage.  Some may have a read-only,
3167
non-volatile memory @code{ROM} that allows code execution and read-only data
3168
access.  The last variant is a read-only, non-volatile memory @code{ROM2} with
3169
read-only data access and no code execution capability.  We have four output
3170
sections:
3171
 
3172
@itemize @bullet
3173
@item
3174
@code{.text} program code;
3175
@item
3176
@code{.rodata} read-only data;
3177
@item
3178
@code{.data} read-write initialized data;
3179
@item
3180
@code{.bss} read-write zero initialized data.
3181
@end itemize
3182
 
3183
The goal is to provide a linker command file that contains a system independent
3184
part defining the output sections and a system dependent part mapping the
3185
output sections to the memory regions available on the system.  Our embedded
3186
systems come with three different memory setups @code{A}, @code{B} and
3187
@code{C}:
3188
@multitable @columnfractions .25 .25 .25 .25
3189
@item Section @tab Variant A @tab Variant B @tab Variant C
3190
@item .text @tab RAM @tab ROM @tab ROM
3191
@item .rodata @tab RAM @tab ROM @tab ROM2
3192
@item .data @tab RAM @tab RAM/ROM @tab RAM/ROM2
3193
@item .bss @tab RAM @tab RAM @tab RAM
3194
@end multitable
3195
The notation @code{RAM/ROM} or @code{RAM/ROM2} means that this section is
3196
loaded into region @code{ROM} or @code{ROM2} respectively.  Please note that
3197
the load address of the @code{.data} section starts in all three variants at
3198
the end of the @code{.rodata} section.
3199
 
3200
The base linker script that deals with the output sections follows.  It
3201
includes the system dependent @code{linkcmds.memory} file that describes the
3202
memory layout:
3203
@smallexample
3204
INCLUDE linkcmds.memory
3205
 
3206
SECTIONS
3207
  @{
3208
    .text :
3209
      @{
3210
        *(.text)
3211
      @} > REGION_TEXT
3212
    .rodata :
3213
      @{
3214
        *(.rodata)
3215
        rodata_end = .;
3216
      @} > REGION_RODATA
3217
    .data : AT (rodata_end)
3218
      @{
3219
        data_start = .;
3220
        *(.data)
3221
      @} > REGION_DATA
3222
    data_size = SIZEOF(.data);
3223
    data_load_start = LOADADDR(.data);
3224
    .bss :
3225
      @{
3226
        *(.bss)
3227
      @} > REGION_BSS
3228
  @}
3229
@end smallexample
3230
 
3231
Now we need three different @code{linkcmds.memory} files to define memory
3232
regions and alias names.  The content of @code{linkcmds.memory} for the three
3233
variants @code{A}, @code{B} and @code{C}:
3234
@table @code
3235
@item A
3236
Here everything goes into the @code{RAM}.
3237
@smallexample
3238
MEMORY
3239
  @{
3240
    RAM : ORIGIN = 0, LENGTH = 4M
3241
  @}
3242
 
3243
REGION_ALIAS("REGION_TEXT", RAM);
3244
REGION_ALIAS("REGION_RODATA", RAM);
3245
REGION_ALIAS("REGION_DATA", RAM);
3246
REGION_ALIAS("REGION_BSS", RAM);
3247
@end smallexample
3248
@item B
3249
Program code and read-only data go into the @code{ROM}.  Read-write data goes
3250
into the @code{RAM}.  An image of the initialized data is loaded into the
3251
@code{ROM} and will be copied during system start into the @code{RAM}.
3252
@smallexample
3253
MEMORY
3254
  @{
3255
    ROM : ORIGIN = 0, LENGTH = 3M
3256
    RAM : ORIGIN = 0x10000000, LENGTH = 1M
3257
  @}
3258
 
3259
REGION_ALIAS("REGION_TEXT", ROM);
3260
REGION_ALIAS("REGION_RODATA", ROM);
3261
REGION_ALIAS("REGION_DATA", RAM);
3262
REGION_ALIAS("REGION_BSS", RAM);
3263
@end smallexample
3264
@item C
3265
Program code goes into the @code{ROM}.  Read-only data goes into the
3266
@code{ROM2}.  Read-write data goes into the @code{RAM}.  An image of the
3267
initialized data is loaded into the @code{ROM2} and will be copied during
3268
system start into the @code{RAM}.
3269
@smallexample
3270
MEMORY
3271
  @{
3272
    ROM : ORIGIN = 0, LENGTH = 2M
3273
    ROM2 : ORIGIN = 0x10000000, LENGTH = 1M
3274
    RAM : ORIGIN = 0x20000000, LENGTH = 1M
3275
  @}
3276
 
3277
REGION_ALIAS("REGION_TEXT", ROM);
3278
REGION_ALIAS("REGION_RODATA", ROM2);
3279
REGION_ALIAS("REGION_DATA", RAM);
3280
REGION_ALIAS("REGION_BSS", RAM);
3281
@end smallexample
3282
@end table
3283
 
3284
It is possible to write a common system initialization routine to copy the
3285
@code{.data} section from @code{ROM} or @code{ROM2} into the @code{RAM} if
3286
necessary:
3287
@smallexample
3288
#include <string.h>
3289
 
3290
extern char data_start [];
3291
extern char data_size [];
3292
extern char data_load_start [];
3293
 
3294
void copy_data(void)
3295
@{
3296
  if (data_start != data_load_start)
3297
    @{
3298
      memcpy(data_start, data_load_start, (size_t) data_size);
3299
    @}
3300
@}
3301
@end smallexample
3302
 
3303
@node Miscellaneous Commands
3304
@subsection Other Linker Script Commands
3305
There are a few other linker scripts commands.
3306
 
3307
@table @code
3308
@item ASSERT(@var{exp}, @var{message})
3309
@kindex ASSERT
3310
@cindex assertion in linker script
3311
Ensure that @var{exp} is non-zero.  If it is zero, then exit the linker
3312
with an error code, and print @var{message}.
3313
 
3314
@item EXTERN(@var{symbol} @var{symbol} @dots{})
3315
@kindex EXTERN
3316
@cindex undefined symbol in linker script
3317
Force @var{symbol} to be entered in the output file as an undefined
3318
symbol.  Doing this may, for example, trigger linking of additional
3319
modules from standard libraries.  You may list several @var{symbol}s for
3320
each @code{EXTERN}, and you may use @code{EXTERN} multiple times.  This
3321
command has the same effect as the @samp{-u} command-line option.
3322
 
3323
@item FORCE_COMMON_ALLOCATION
3324
@kindex FORCE_COMMON_ALLOCATION
3325
@cindex common allocation in linker script
3326
This command has the same effect as the @samp{-d} command-line option:
3327
to make @command{ld} assign space to common symbols even if a relocatable
3328
output file is specified (@samp{-r}).
3329
 
3330
@item INHIBIT_COMMON_ALLOCATION
3331
@kindex INHIBIT_COMMON_ALLOCATION
3332
@cindex common allocation in linker script
3333
This command has the same effect as the @samp{--no-define-common}
3334
command-line option: to make @code{ld} omit the assignment of addresses
3335
to common symbols even for a non-relocatable output file.
3336
 
3337
@item INSERT [ AFTER | BEFORE ] @var{output_section}
3338
@kindex INSERT
3339
@cindex insert user script into default script
3340
This command is typically used in a script specified by @samp{-T} to
3341
augment the default @code{SECTIONS} with, for example, overlays.  It
3342
inserts all prior linker script statements after (or before)
3343
@var{output_section}, and also causes @samp{-T} to not override the
3344
default linker script.  The exact insertion point is as for orphan
3345
sections.  @xref{Location Counter}.  The insertion happens after the
3346
linker has mapped input sections to output sections.  Prior to the
3347
insertion, since @samp{-T} scripts are parsed before the default
3348
linker script, statements in the @samp{-T} script occur before the
3349
default linker script statements in the internal linker representation
3350
of the script.  In particular, input section assignments will be made
3351
to @samp{-T} output sections before those in the default script.  Here
3352
is an example of how a @samp{-T} script using @code{INSERT} might look:
3353
 
3354
@smallexample
3355
SECTIONS
3356
@{
3357
  OVERLAY :
3358
  @{
3359
    .ov1 @{ ov1*(.text) @}
3360
    .ov2 @{ ov2*(.text) @}
3361
  @}
3362
@}
3363
INSERT AFTER .text;
3364
@end smallexample
3365
 
3366
@item NOCROSSREFS(@var{section} @var{section} @dots{})
3367
@kindex NOCROSSREFS(@var{sections})
3368
@cindex cross references
3369
This command may be used to tell @command{ld} to issue an error about any
3370
references among certain output sections.
3371
 
3372
In certain types of programs, particularly on embedded systems when
3373
using overlays, when one section is loaded into memory, another section
3374
will not be.  Any direct references between the two sections would be
3375
errors.  For example, it would be an error if code in one section called
3376
a function defined in the other section.
3377
 
3378
The @code{NOCROSSREFS} command takes a list of output section names.  If
3379
@command{ld} detects any cross references between the sections, it reports
3380
an error and returns a non-zero exit status.  Note that the
3381
@code{NOCROSSREFS} command uses output section names, not input section
3382
names.
3383
 
3384
@ifclear SingleFormat
3385
@item OUTPUT_ARCH(@var{bfdarch})
3386
@kindex OUTPUT_ARCH(@var{bfdarch})
3387
@cindex machine architecture
3388
@cindex architecture
3389
Specify a particular output machine architecture.  The argument is one
3390
of the names used by the BFD library (@pxref{BFD}).  You can see the
3391
architecture of an object file by using the @code{objdump} program with
3392
the @samp{-f} option.
3393
@end ifclear
3394
 
3395
@item LD_FEATURE(@var{string})
3396
@kindex LD_FEATURE(@var{string})
3397
This command may be used to modify @command{ld} behavior.  If
3398
@var{string} is @code{"SANE_EXPR"} then absolute symbols and numbers
3399
in a script are simply treated as numbers everywhere.
3400
@xref{Expression Section}.
3401
@end table
3402
 
3403
@node Assignments
3404
@section Assigning Values to Symbols
3405
@cindex assignment in scripts
3406
@cindex symbol definition, scripts
3407
@cindex variables, defining
3408
You may assign a value to a symbol in a linker script.  This will define
3409
the symbol and place it into the symbol table with a global scope.
3410
 
3411
@menu
3412
* Simple Assignments::          Simple Assignments
3413
* PROVIDE::                     PROVIDE
3414
* PROVIDE_HIDDEN::              PROVIDE_HIDDEN
3415
* Source Code Reference::       How to use a linker script defined symbol in source code
3416
@end menu
3417
 
3418
@node Simple Assignments
3419
@subsection Simple Assignments
3420
 
3421
You may assign to a symbol using any of the C assignment operators:
3422
 
3423
@table @code
3424
@item @var{symbol} = @var{expression} ;
3425
@itemx @var{symbol} += @var{expression} ;
3426
@itemx @var{symbol} -= @var{expression} ;
3427
@itemx @var{symbol} *= @var{expression} ;
3428
@itemx @var{symbol} /= @var{expression} ;
3429
@itemx @var{symbol} <<= @var{expression} ;
3430
@itemx @var{symbol} >>= @var{expression} ;
3431
@itemx @var{symbol} &= @var{expression} ;
3432
@itemx @var{symbol} |= @var{expression} ;
3433
@end table
3434
 
3435
The first case will define @var{symbol} to the value of
3436
@var{expression}.  In the other cases, @var{symbol} must already be
3437
defined, and the value will be adjusted accordingly.
3438
 
3439
The special symbol name @samp{.} indicates the location counter.  You
3440
may only use this within a @code{SECTIONS} command.  @xref{Location Counter}.
3441
 
3442
The semicolon after @var{expression} is required.
3443
 
3444
Expressions are defined below; see @ref{Expressions}.
3445
 
3446
You may write symbol assignments as commands in their own right, or as
3447
statements within a @code{SECTIONS} command, or as part of an output
3448
section description in a @code{SECTIONS} command.
3449
 
3450
The section of the symbol will be set from the section of the
3451
expression; for more information, see @ref{Expression Section}.
3452
 
3453
Here is an example showing the three different places that symbol
3454
assignments may be used:
3455
 
3456
@smallexample
3457
floating_point = 0;
3458
SECTIONS
3459
@{
3460
  .text :
3461
    @{
3462
      *(.text)
3463
      _etext = .;
3464
    @}
3465
  _bdata = (. + 3) & ~ 3;
3466
  .data : @{ *(.data) @}
3467
@}
3468
@end smallexample
3469
@noindent
3470
In this example, the symbol @samp{floating_point} will be defined as
3471
zero.  The symbol @samp{_etext} will be defined as the address following
3472
the last @samp{.text} input section.  The symbol @samp{_bdata} will be
3473
defined as the address following the @samp{.text} output section aligned
3474
upward to a 4 byte boundary.
3475
 
3476
@node PROVIDE
3477
@subsection PROVIDE
3478
@cindex PROVIDE
3479
In some cases, it is desirable for a linker script to define a symbol
3480
only if it is referenced and is not defined by any object included in
3481
the link.  For example, traditional linkers defined the symbol
3482
@samp{etext}.  However, ANSI C requires that the user be able to use
3483
@samp{etext} as a function name without encountering an error.  The
3484
@code{PROVIDE} keyword may be used to define a symbol, such as
3485
@samp{etext}, only if it is referenced but not defined.  The syntax is
3486
@code{PROVIDE(@var{symbol} = @var{expression})}.
3487
 
3488
Here is an example of using @code{PROVIDE} to define @samp{etext}:
3489
@smallexample
3490
SECTIONS
3491
@{
3492
  .text :
3493
    @{
3494
      *(.text)
3495
      _etext = .;
3496
      PROVIDE(etext = .);
3497
    @}
3498
@}
3499
@end smallexample
3500
 
3501
In this example, if the program defines @samp{_etext} (with a leading
3502
underscore), the linker will give a multiple definition error.  If, on
3503
the other hand, the program defines @samp{etext} (with no leading
3504
underscore), the linker will silently use the definition in the program.
3505
If the program references @samp{etext} but does not define it, the
3506
linker will use the definition in the linker script.
3507
 
3508
@node PROVIDE_HIDDEN
3509
@subsection PROVIDE_HIDDEN
3510
@cindex PROVIDE_HIDDEN
3511
Similar to @code{PROVIDE}.  For ELF targeted ports, the symbol will be
3512
hidden and won't be exported.
3513
 
3514
@node Source Code Reference
3515
@subsection Source Code Reference
3516
 
3517
Accessing a linker script defined variable from source code is not
3518
intuitive.  In particular a linker script symbol is not equivalent to
3519
a variable declaration in a high level language, it is instead a
3520
symbol that does not have a value.
3521
 
3522
Before going further, it is important to note that compilers often
3523
transform names in the source code into different names when they are
3524
stored in the symbol table.  For example, Fortran compilers commonly
3525
prepend or append an underscore, and C++ performs extensive @samp{name
3526
mangling}.  Therefore there might be a discrepancy between the name
3527
of a variable as it is used in source code and the name of the same
3528
variable as it is defined in a linker script.  For example in C a
3529
linker script variable might be referred to as:
3530
 
3531
@smallexample
3532
  extern int foo;
3533
@end smallexample
3534
 
3535
But in the linker script it might be defined as:
3536
 
3537
@smallexample
3538
  _foo = 1000;
3539
@end smallexample
3540
 
3541
In the remaining examples however it is assumed that no name
3542
transformation has taken place.
3543
 
3544
When a symbol is declared in a high level language such as C, two
3545
things happen.  The first is that the compiler reserves enough space
3546
in the program's memory to hold the @emph{value} of the symbol.  The
3547
second is that the compiler creates an entry in the program's symbol
3548
table which holds the symbol's @emph{address}.  ie the symbol table
3549
contains the address of the block of memory holding the symbol's
3550
value.  So for example the following C declaration, at file scope:
3551
 
3552
@smallexample
3553
  int foo = 1000;
3554
@end smallexample
3555
 
3556
creates a entry called @samp{foo} in the symbol table.  This entry
3557
holds the address of an @samp{int} sized block of memory where the
3558
number 1000 is initially stored.
3559
 
3560
When a program references a symbol the compiler generates code that
3561
first accesses the symbol table to find the address of the symbol's
3562
memory block and then code to read the value from that memory block.
3563
So:
3564
 
3565
@smallexample
3566
  foo = 1;
3567
@end smallexample
3568
 
3569
looks up the symbol @samp{foo} in the symbol table, gets the address
3570
associated with this symbol and then writes the value 1 into that
3571
address.  Whereas:
3572
 
3573
@smallexample
3574
  int * a = & foo;
3575
@end smallexample
3576
 
3577
looks up the symbol @samp{foo} in the symbol table, gets it address
3578
and then copies this address into the block of memory associated with
3579
the variable @samp{a}.
3580
 
3581
Linker scripts symbol declarations, by contrast, create an entry in
3582
the symbol table but do not assign any memory to them.  Thus they are
3583
an address without a value.  So for example the linker script definition:
3584
 
3585
@smallexample
3586
  foo = 1000;
3587
@end smallexample
3588
 
3589
creates an entry in the symbol table called @samp{foo} which holds
3590
the address of memory location 1000, but nothing special is stored at
3591
address 1000.  This means that you cannot access the @emph{value} of a
3592
linker script defined symbol - it has no value - all you can do is
3593
access the @emph{address} of a linker script defined symbol.
3594
 
3595
Hence when you are using a linker script defined symbol in source code
3596
you should always take the address of the symbol, and never attempt to
3597
use its value.  For example suppose you want to copy the contents of a
3598
section of memory called .ROM into a section called .FLASH and the
3599
linker script contains these declarations:
3600
 
3601
@smallexample
3602
@group
3603
  start_of_ROM   = .ROM;
3604
  end_of_ROM     = .ROM + sizeof (.ROM) - 1;
3605
  start_of_FLASH = .FLASH;
3606
@end group
3607
@end smallexample
3608
 
3609
Then the C source code to perform the copy would be:
3610
 
3611
@smallexample
3612
@group
3613
  extern char start_of_ROM, end_of_ROM, start_of_FLASH;
3614
 
3615
  memcpy (& start_of_FLASH, & start_of_ROM, & end_of_ROM - & start_of_ROM);
3616
@end group
3617
@end smallexample
3618
 
3619
Note the use of the @samp{&} operators.  These are correct.
3620
 
3621
@node SECTIONS
3622
@section SECTIONS Command
3623
@kindex SECTIONS
3624
The @code{SECTIONS} command tells the linker how to map input sections
3625
into output sections, and how to place the output sections in memory.
3626
 
3627
The format of the @code{SECTIONS} command is:
3628
@smallexample
3629
SECTIONS
3630
@{
3631
  @var{sections-command}
3632
  @var{sections-command}
3633
  @dots{}
3634
@}
3635
@end smallexample
3636
 
3637
Each @var{sections-command} may of be one of the following:
3638
 
3639
@itemize @bullet
3640
@item
3641
an @code{ENTRY} command (@pxref{Entry Point,,Entry command})
3642
@item
3643
a symbol assignment (@pxref{Assignments})
3644
@item
3645
an output section description
3646
@item
3647
an overlay description
3648
@end itemize
3649
 
3650
The @code{ENTRY} command and symbol assignments are permitted inside the
3651
@code{SECTIONS} command for convenience in using the location counter in
3652
those commands.  This can also make the linker script easier to
3653
understand because you can use those commands at meaningful points in
3654
the layout of the output file.
3655
 
3656
Output section descriptions and overlay descriptions are described
3657
below.
3658
 
3659
If you do not use a @code{SECTIONS} command in your linker script, the
3660
linker will place each input section into an identically named output
3661
section in the order that the sections are first encountered in the
3662
input files.  If all input sections are present in the first file, for
3663
example, the order of sections in the output file will match the order
3664
in the first input file.  The first section will be at address zero.
3665
 
3666
@menu
3667
* Output Section Description::  Output section description
3668
* Output Section Name::         Output section name
3669
* Output Section Address::      Output section address
3670
* Input Section::               Input section description
3671
* Output Section Data::         Output section data
3672
* Output Section Keywords::     Output section keywords
3673
* Output Section Discarding::   Output section discarding
3674
* Output Section Attributes::   Output section attributes
3675
* Overlay Description::         Overlay description
3676
@end menu
3677
 
3678
@node Output Section Description
3679
@subsection Output Section Description
3680
The full description of an output section looks like this:
3681
@smallexample
3682
@group
3683
@var{section} [@var{address}] [(@var{type})] :
3684
  [AT(@var{lma})]
3685
  [ALIGN(@var{section_align})]
3686
  [SUBALIGN(@var{subsection_align})]
3687
  [@var{constraint}]
3688
  @{
3689
    @var{output-section-command}
3690
    @var{output-section-command}
3691
    @dots{}
3692
  @} [>@var{region}] [AT>@var{lma_region}] [:@var{phdr} :@var{phdr} @dots{}] [=@var{fillexp}]
3693
@end group
3694
@end smallexample
3695
 
3696
Most output sections do not use most of the optional section attributes.
3697
 
3698
The whitespace around @var{section} is required, so that the section
3699
name is unambiguous.  The colon and the curly braces are also required.
3700
The line breaks and other white space are optional.
3701
 
3702
Each @var{output-section-command} may be one of the following:
3703
 
3704
@itemize @bullet
3705
@item
3706
a symbol assignment (@pxref{Assignments})
3707
@item
3708
an input section description (@pxref{Input Section})
3709
@item
3710
data values to include directly (@pxref{Output Section Data})
3711
@item
3712
a special output section keyword (@pxref{Output Section Keywords})
3713
@end itemize
3714
 
3715
@node Output Section Name
3716
@subsection Output Section Name
3717
@cindex name, section
3718
@cindex section name
3719
The name of the output section is @var{section}.  @var{section} must
3720
meet the constraints of your output format.  In formats which only
3721
support a limited number of sections, such as @code{a.out}, the name
3722
must be one of the names supported by the format (@code{a.out}, for
3723
example, allows only @samp{.text}, @samp{.data} or @samp{.bss}). If the
3724
output format supports any number of sections, but with numbers and not
3725
names (as is the case for Oasys), the name should be supplied as a
3726
quoted numeric string.  A section name may consist of any sequence of
3727
characters, but a name which contains any unusual characters such as
3728
commas must be quoted.
3729
 
3730
The output section name @samp{/DISCARD/} is special; @ref{Output Section
3731
Discarding}.
3732
 
3733
@node Output Section Address
3734
@subsection Output Section Address
3735
@cindex address, section
3736
@cindex section address
3737
The @var{address} is an expression for the VMA (the virtual memory
3738
address) of the output section.  This address is optional, but if it
3739
is provided then the output address will be set exactly as specified.
3740
 
3741
If the output address is not specified then one will be chosen for the
3742
section, based on the heuristic below.  This address will be adjusted
3743
to fit the alignment requirement of the output section.  The
3744
alignment requirement is the strictest alignment of any input section
3745
contained within the output section.
3746
 
3747
The output section address heuristic is as follows:
3748
 
3749
@itemize @bullet
3750
@item
3751
If an output memory @var{region} is set for the section then it
3752
is added to this region and its address will be the next free address
3753
in that region.
3754
 
3755
@item
3756
If the MEMORY command has been used to create a list of memory
3757
regions then the first region which has attributes compatible with the
3758
section is selected to contain it.  The section's output address will
3759
be the next free address in that region; @ref{MEMORY}.
3760
 
3761
@item
3762
If no memory regions were specified, or none match the section then
3763
the output address will be based on the current value of the location
3764
counter.
3765
@end itemize
3766
 
3767
@noindent
3768
For example:
3769
 
3770
@smallexample
3771
.text . : @{ *(.text) @}
3772
@end smallexample
3773
 
3774
@noindent
3775
and
3776
 
3777
@smallexample
3778
.text : @{ *(.text) @}
3779
@end smallexample
3780
 
3781
@noindent
3782
are subtly different.  The first will set the address of the
3783
@samp{.text} output section to the current value of the location
3784
counter.  The second will set it to the current value of the location
3785
counter aligned to the strictest alignment of any of the @samp{.text}
3786
input sections.
3787
 
3788
The @var{address} may be an arbitrary expression; @ref{Expressions}.
3789
For example, if you want to align the section on a 0x10 byte boundary,
3790
so that the lowest four bits of the section address are zero, you could
3791
do something like this:
3792
@smallexample
3793
.text ALIGN(0x10) : @{ *(.text) @}
3794
@end smallexample
3795
@noindent
3796
This works because @code{ALIGN} returns the current location counter
3797
aligned upward to the specified value.
3798
 
3799
Specifying @var{address} for a section will change the value of the
3800
location counter, provided that the section is non-empty.  (Empty
3801
sections are ignored).
3802
 
3803
@node Input Section
3804
@subsection Input Section Description
3805
@cindex input sections
3806
@cindex mapping input sections to output sections
3807
The most common output section command is an input section description.
3808
 
3809
The input section description is the most basic linker script operation.
3810
You use output sections to tell the linker how to lay out your program
3811
in memory.  You use input section descriptions to tell the linker how to
3812
map the input files into your memory layout.
3813
 
3814
@menu
3815
* Input Section Basics::        Input section basics
3816
* Input Section Wildcards::     Input section wildcard patterns
3817
* Input Section Common::        Input section for common symbols
3818
* Input Section Keep::          Input section and garbage collection
3819
* Input Section Example::       Input section example
3820
@end menu
3821
 
3822
@node Input Section Basics
3823
@subsubsection Input Section Basics
3824
@cindex input section basics
3825
An input section description consists of a file name optionally followed
3826
by a list of section names in parentheses.
3827
 
3828
The file name and the section name may be wildcard patterns, which we
3829
describe further below (@pxref{Input Section Wildcards}).
3830
 
3831
The most common input section description is to include all input
3832
sections with a particular name in the output section.  For example, to
3833
include all input @samp{.text} sections, you would write:
3834
@smallexample
3835
*(.text)
3836
@end smallexample
3837
@noindent
3838
Here the @samp{*} is a wildcard which matches any file name.  To exclude a list
3839
of files from matching the file name wildcard, EXCLUDE_FILE may be used to
3840
match all files except the ones specified in the EXCLUDE_FILE list.  For
3841
example:
3842
@smallexample
3843
*(EXCLUDE_FILE (*crtend.o *otherfile.o) .ctors)
3844
@end smallexample
3845
will cause all .ctors sections from all files except @file{crtend.o} and
3846
@file{otherfile.o} to be included.
3847
 
3848
There are two ways to include more than one section:
3849
@smallexample
3850
*(.text .rdata)
3851
*(.text) *(.rdata)
3852
@end smallexample
3853
@noindent
3854
The difference between these is the order in which the @samp{.text} and
3855
@samp{.rdata} input sections will appear in the output section.  In the
3856
first example, they will be intermingled, appearing in the same order as
3857
they are found in the linker input.  In the second example, all
3858
@samp{.text} input sections will appear first, followed by all
3859
@samp{.rdata} input sections.
3860
 
3861
You can specify a file name to include sections from a particular file.
3862
You would do this if one or more of your files contain special data that
3863
needs to be at a particular location in memory.  For example:
3864
@smallexample
3865
data.o(.data)
3866
@end smallexample
3867
 
3868 157 khays
To refine the sections that are included based on the section flags
3869
of an input section, INPUT_SECTION_FLAGS may be used.
3870
 
3871
Here is a simple example for using Section header flags for ELF sections:
3872
 
3873
@smallexample
3874
@group
3875
SECTIONS @{
3876
  .text : @{ INPUT_SECTION_FLAGS (SHF_MERGE & SHF_STRINGS) *(.text) @}
3877
  .text2 :  @{ INPUT_SECTION_FLAGS (!SHF_WRITE) *(.text) @}
3878
@}
3879
@end group
3880
@end smallexample
3881
 
3882
In this example, the output section @samp{.text} will be comprised of any
3883
input section matching the name *(.text) whose section header flags
3884
@code{SHF_MERGE} and @code{SHF_STRINGS} are set.  The output section
3885
@samp{.text2} will be comprised of any input section matching the name *(.text)
3886
whose section header flag @code{SHF_WRITE} is clear.
3887
 
3888 145 khays
You can also specify files within archives by writing a pattern
3889
matching the archive, a colon, then the pattern matching the file,
3890
with no whitespace around the colon.
3891
 
3892
@table @samp
3893
@item archive:file
3894
matches file within archive
3895
@item archive:
3896
matches the whole archive
3897
@item :file
3898
matches file but not one in an archive
3899
@end table
3900
 
3901
Either one or both of @samp{archive} and @samp{file} can contain shell
3902
wildcards.  On DOS based file systems, the linker will assume that a
3903
single letter followed by a colon is a drive specifier, so
3904
@samp{c:myfile.o} is a simple file specification, not @samp{myfile.o}
3905
within an archive called @samp{c}.  @samp{archive:file} filespecs may
3906
also be used within an @code{EXCLUDE_FILE} list, but may not appear in
3907
other linker script contexts.  For instance, you cannot extract a file
3908
from an archive by using @samp{archive:file} in an @code{INPUT}
3909
command.
3910
 
3911
If you use a file name without a list of sections, then all sections in
3912
the input file will be included in the output section.  This is not
3913
commonly done, but it may by useful on occasion.  For example:
3914
@smallexample
3915
data.o
3916
@end smallexample
3917
 
3918
When you use a file name which is not an @samp{archive:file} specifier
3919
and does not contain any wild card
3920
characters, the linker will first see if you also specified the file
3921
name on the linker command line or in an @code{INPUT} command.  If you
3922
did not, the linker will attempt to open the file as an input file, as
3923
though it appeared on the command line.  Note that this differs from an
3924
@code{INPUT} command, because the linker will not search for the file in
3925
the archive search path.
3926
 
3927
@node Input Section Wildcards
3928
@subsubsection Input Section Wildcard Patterns
3929
@cindex input section wildcards
3930
@cindex wildcard file name patterns
3931
@cindex file name wildcard patterns
3932
@cindex section name wildcard patterns
3933
In an input section description, either the file name or the section
3934
name or both may be wildcard patterns.
3935
 
3936
The file name of @samp{*} seen in many examples is a simple wildcard
3937
pattern for the file name.
3938
 
3939
The wildcard patterns are like those used by the Unix shell.
3940
 
3941
@table @samp
3942
@item *
3943
matches any number of characters
3944
@item ?
3945
matches any single character
3946
@item [@var{chars}]
3947
matches a single instance of any of the @var{chars}; the @samp{-}
3948
character may be used to specify a range of characters, as in
3949
@samp{[a-z]} to match any lower case letter
3950
@item \
3951
quotes the following character
3952
@end table
3953
 
3954
When a file name is matched with a wildcard, the wildcard characters
3955
will not match a @samp{/} character (used to separate directory names on
3956
Unix).  A pattern consisting of a single @samp{*} character is an
3957
exception; it will always match any file name, whether it contains a
3958
@samp{/} or not.  In a section name, the wildcard characters will match
3959
a @samp{/} character.
3960
 
3961
File name wildcard patterns only match files which are explicitly
3962
specified on the command line or in an @code{INPUT} command.  The linker
3963
does not search directories to expand wildcards.
3964
 
3965
If a file name matches more than one wildcard pattern, or if a file name
3966
appears explicitly and is also matched by a wildcard pattern, the linker
3967
will use the first match in the linker script.  For example, this
3968
sequence of input section descriptions is probably in error, because the
3969
@file{data.o} rule will not be used:
3970
@smallexample
3971
.data : @{ *(.data) @}
3972
.data1 : @{ data.o(.data) @}
3973
@end smallexample
3974
 
3975
@cindex SORT_BY_NAME
3976
Normally, the linker will place files and sections matched by wildcards
3977
in the order in which they are seen during the link.  You can change
3978
this by using the @code{SORT_BY_NAME} keyword, which appears before a wildcard
3979
pattern in parentheses (e.g., @code{SORT_BY_NAME(.text*)}).  When the
3980
@code{SORT_BY_NAME} keyword is used, the linker will sort the files or sections
3981
into ascending order by name before placing them in the output file.
3982
 
3983
@cindex SORT_BY_ALIGNMENT
3984
@code{SORT_BY_ALIGNMENT} is very similar to @code{SORT_BY_NAME}. The
3985
difference is @code{SORT_BY_ALIGNMENT} will sort sections into
3986
ascending order by alignment before placing them in the output file.
3987
 
3988
@cindex SORT_BY_INIT_PRIORITY
3989
@code{SORT_BY_INIT_PRIORITY} is very similar to @code{SORT_BY_NAME}. The
3990
difference is @code{SORT_BY_INIT_PRIORITY} will sort sections into
3991
ascending order by numerical value of the GCC init_priority attribute
3992
encoded in the section name before placing them in the output file.
3993
 
3994
@cindex SORT
3995
@code{SORT} is an alias for @code{SORT_BY_NAME}.
3996
 
3997
When there are nested section sorting commands in linker script, there
3998
can be at most 1 level of nesting for section sorting commands.
3999
 
4000
@enumerate
4001
@item
4002
@code{SORT_BY_NAME} (@code{SORT_BY_ALIGNMENT} (wildcard section pattern)).
4003
It will sort the input sections by name first, then by alignment if 2
4004
sections have the same name.
4005
@item
4006
@code{SORT_BY_ALIGNMENT} (@code{SORT_BY_NAME} (wildcard section pattern)).
4007
It will sort the input sections by alignment first, then by name if 2
4008
sections have the same alignment.
4009
@item
4010
@code{SORT_BY_NAME} (@code{SORT_BY_NAME} (wildcard section pattern)) is
4011
treated the same as @code{SORT_BY_NAME} (wildcard section pattern).
4012
@item
4013
@code{SORT_BY_ALIGNMENT} (@code{SORT_BY_ALIGNMENT} (wildcard section pattern))
4014
is treated the same as @code{SORT_BY_ALIGNMENT} (wildcard section pattern).
4015
@item
4016
All other nested section sorting commands are invalid.
4017
@end enumerate
4018
 
4019
When both command line section sorting option and linker script
4020
section sorting command are used, section sorting command always
4021
takes precedence over the command line option.
4022
 
4023
If the section sorting command in linker script isn't nested, the
4024
command line option will make the section sorting command to be
4025
treated as nested sorting command.
4026
 
4027
@enumerate
4028
@item
4029
@code{SORT_BY_NAME} (wildcard section pattern ) with
4030
@option{--sort-sections alignment} is equivalent to
4031
@code{SORT_BY_NAME} (@code{SORT_BY_ALIGNMENT} (wildcard section pattern)).
4032
@item
4033
@code{SORT_BY_ALIGNMENT} (wildcard section pattern) with
4034
@option{--sort-section name} is equivalent to
4035
@code{SORT_BY_ALIGNMENT} (@code{SORT_BY_NAME} (wildcard section pattern)).
4036
@end enumerate
4037
 
4038
If the section sorting command in linker script is nested, the
4039
command line option will be ignored.
4040
 
4041
If you ever get confused about where input sections are going, use the
4042
@samp{-M} linker option to generate a map file.  The map file shows
4043
precisely how input sections are mapped to output sections.
4044
 
4045
This example shows how wildcard patterns might be used to partition
4046
files.  This linker script directs the linker to place all @samp{.text}
4047
sections in @samp{.text} and all @samp{.bss} sections in @samp{.bss}.
4048
The linker will place the @samp{.data} section from all files beginning
4049
with an upper case character in @samp{.DATA}; for all other files, the
4050
linker will place the @samp{.data} section in @samp{.data}.
4051
@smallexample
4052
@group
4053
SECTIONS @{
4054
  .text : @{ *(.text) @}
4055
  .DATA : @{ [A-Z]*(.data) @}
4056
  .data : @{ *(.data) @}
4057
  .bss : @{ *(.bss) @}
4058
@}
4059
@end group
4060
@end smallexample
4061
 
4062
@node Input Section Common
4063
@subsubsection Input Section for Common Symbols
4064
@cindex common symbol placement
4065
@cindex uninitialized data placement
4066
A special notation is needed for common symbols, because in many object
4067
file formats common symbols do not have a particular input section.  The
4068
linker treats common symbols as though they are in an input section
4069
named @samp{COMMON}.
4070
 
4071
You may use file names with the @samp{COMMON} section just as with any
4072
other input sections.  You can use this to place common symbols from a
4073
particular input file in one section while common symbols from other
4074
input files are placed in another section.
4075
 
4076
In most cases, common symbols in input files will be placed in the
4077
@samp{.bss} section in the output file.  For example:
4078
@smallexample
4079
.bss @{ *(.bss) *(COMMON) @}
4080
@end smallexample
4081
 
4082
@cindex scommon section
4083
@cindex small common symbols
4084
Some object file formats have more than one type of common symbol.  For
4085
example, the MIPS ELF object file format distinguishes standard common
4086
symbols and small common symbols.  In this case, the linker will use a
4087
different special section name for other types of common symbols.  In
4088
the case of MIPS ELF, the linker uses @samp{COMMON} for standard common
4089
symbols and @samp{.scommon} for small common symbols.  This permits you
4090
to map the different types of common symbols into memory at different
4091
locations.
4092
 
4093
@cindex [COMMON]
4094
You will sometimes see @samp{[COMMON]} in old linker scripts.  This
4095
notation is now considered obsolete.  It is equivalent to
4096
@samp{*(COMMON)}.
4097
 
4098
@node Input Section Keep
4099
@subsubsection Input Section and Garbage Collection
4100
@cindex KEEP
4101
@cindex garbage collection
4102
When link-time garbage collection is in use (@samp{--gc-sections}),
4103
it is often useful to mark sections that should not be eliminated.
4104
This is accomplished by surrounding an input section's wildcard entry
4105
with @code{KEEP()}, as in @code{KEEP(*(.init))} or
4106
@code{KEEP(SORT_BY_NAME(*)(.ctors))}.
4107
 
4108
@node Input Section Example
4109
@subsubsection Input Section Example
4110
The following example is a complete linker script.  It tells the linker
4111
to read all of the sections from file @file{all.o} and place them at the
4112
start of output section @samp{outputa} which starts at location
4113
@samp{0x10000}.  All of section @samp{.input1} from file @file{foo.o}
4114
follows immediately, in the same output section.  All of section
4115
@samp{.input2} from @file{foo.o} goes into output section
4116
@samp{outputb}, followed by section @samp{.input1} from @file{foo1.o}.
4117
All of the remaining @samp{.input1} and @samp{.input2} sections from any
4118
files are written to output section @samp{outputc}.
4119
 
4120
@smallexample
4121
@group
4122
SECTIONS @{
4123
  outputa 0x10000 :
4124
    @{
4125
    all.o
4126
    foo.o (.input1)
4127
    @}
4128
@end group
4129
@group
4130
  outputb :
4131
    @{
4132
    foo.o (.input2)
4133
    foo1.o (.input1)
4134
    @}
4135
@end group
4136
@group
4137
  outputc :
4138
    @{
4139
    *(.input1)
4140
    *(.input2)
4141
    @}
4142
@}
4143
@end group
4144
@end smallexample
4145
 
4146
@node Output Section Data
4147
@subsection Output Section Data
4148
@cindex data
4149
@cindex section data
4150
@cindex output section data
4151
@kindex BYTE(@var{expression})
4152
@kindex SHORT(@var{expression})
4153
@kindex LONG(@var{expression})
4154
@kindex QUAD(@var{expression})
4155
@kindex SQUAD(@var{expression})
4156
You can include explicit bytes of data in an output section by using
4157
@code{BYTE}, @code{SHORT}, @code{LONG}, @code{QUAD}, or @code{SQUAD} as
4158
an output section command.  Each keyword is followed by an expression in
4159
parentheses providing the value to store (@pxref{Expressions}).  The
4160
value of the expression is stored at the current value of the location
4161
counter.
4162
 
4163
The @code{BYTE}, @code{SHORT}, @code{LONG}, and @code{QUAD} commands
4164
store one, two, four, and eight bytes (respectively).  After storing the
4165
bytes, the location counter is incremented by the number of bytes
4166
stored.
4167
 
4168
For example, this will store the byte 1 followed by the four byte value
4169
of the symbol @samp{addr}:
4170
@smallexample
4171
BYTE(1)
4172
LONG(addr)
4173
@end smallexample
4174
 
4175
When using a 64 bit host or target, @code{QUAD} and @code{SQUAD} are the
4176
same; they both store an 8 byte, or 64 bit, value.  When both host and
4177
target are 32 bits, an expression is computed as 32 bits.  In this case
4178
@code{QUAD} stores a 32 bit value zero extended to 64 bits, and
4179
@code{SQUAD} stores a 32 bit value sign extended to 64 bits.
4180
 
4181
If the object file format of the output file has an explicit endianness,
4182
which is the normal case, the value will be stored in that endianness.
4183
When the object file format does not have an explicit endianness, as is
4184
true of, for example, S-records, the value will be stored in the
4185
endianness of the first input object file.
4186
 
4187
Note---these commands only work inside a section description and not
4188
between them, so the following will produce an error from the linker:
4189
@smallexample
4190
SECTIONS @{@ .text : @{@ *(.text) @}@ LONG(1) .data : @{@ *(.data) @}@ @}@
4191
@end smallexample
4192
whereas this will work:
4193
@smallexample
4194
SECTIONS @{@ .text : @{@ *(.text) ; LONG(1) @}@ .data : @{@ *(.data) @}@ @}@
4195
@end smallexample
4196
 
4197
@kindex FILL(@var{expression})
4198
@cindex holes, filling
4199
@cindex unspecified memory
4200
You may use the @code{FILL} command to set the fill pattern for the
4201
current section.  It is followed by an expression in parentheses.  Any
4202
otherwise unspecified regions of memory within the section (for example,
4203
gaps left due to the required alignment of input sections) are filled
4204
with the value of the expression, repeated as
4205
necessary.  A @code{FILL} statement covers memory locations after the
4206
point at which it occurs in the section definition; by including more
4207
than one @code{FILL} statement, you can have different fill patterns in
4208
different parts of an output section.
4209
 
4210
This example shows how to fill unspecified regions of memory with the
4211
value @samp{0x90}:
4212
@smallexample
4213
FILL(0x90909090)
4214
@end smallexample
4215
 
4216
The @code{FILL} command is similar to the @samp{=@var{fillexp}} output
4217
section attribute, but it only affects the
4218
part of the section following the @code{FILL} command, rather than the
4219
entire section.  If both are used, the @code{FILL} command takes
4220
precedence.  @xref{Output Section Fill}, for details on the fill
4221
expression.
4222
 
4223
@node Output Section Keywords
4224
@subsection Output Section Keywords
4225
There are a couple of keywords which can appear as output section
4226
commands.
4227
 
4228
@table @code
4229
@kindex CREATE_OBJECT_SYMBOLS
4230
@cindex input filename symbols
4231
@cindex filename symbols
4232
@item CREATE_OBJECT_SYMBOLS
4233
The command tells the linker to create a symbol for each input file.
4234
The name of each symbol will be the name of the corresponding input
4235
file.  The section of each symbol will be the output section in which
4236
the @code{CREATE_OBJECT_SYMBOLS} command appears.
4237
 
4238
This is conventional for the a.out object file format.  It is not
4239
normally used for any other object file format.
4240
 
4241
@kindex CONSTRUCTORS
4242
@cindex C++ constructors, arranging in link
4243
@cindex constructors, arranging in link
4244
@item CONSTRUCTORS
4245
When linking using the a.out object file format, the linker uses an
4246
unusual set construct to support C++ global constructors and
4247
destructors.  When linking object file formats which do not support
4248
arbitrary sections, such as ECOFF and XCOFF, the linker will
4249
automatically recognize C++ global constructors and destructors by name.
4250
For these object file formats, the @code{CONSTRUCTORS} command tells the
4251
linker to place constructor information in the output section where the
4252
@code{CONSTRUCTORS} command appears.  The @code{CONSTRUCTORS} command is
4253
ignored for other object file formats.
4254
 
4255
The symbol @w{@code{__CTOR_LIST__}} marks the start of the global
4256
constructors, and the symbol @w{@code{__CTOR_END__}} marks the end.
4257
Similarly, @w{@code{__DTOR_LIST__}} and @w{@code{__DTOR_END__}} mark
4258
the start and end of the global destructors.  The
4259
first word in the list is the number of entries, followed by the address
4260
of each constructor or destructor, followed by a zero word.  The
4261
compiler must arrange to actually run the code.  For these object file
4262
formats @sc{gnu} C++ normally calls constructors from a subroutine
4263
@code{__main}; a call to @code{__main} is automatically inserted into
4264
the startup code for @code{main}.  @sc{gnu} C++ normally runs
4265
destructors either by using @code{atexit}, or directly from the function
4266
@code{exit}.
4267
 
4268
For object file formats such as @code{COFF} or @code{ELF} which support
4269
arbitrary section names, @sc{gnu} C++ will normally arrange to put the
4270
addresses of global constructors and destructors into the @code{.ctors}
4271
and @code{.dtors} sections.  Placing the following sequence into your
4272
linker script will build the sort of table which the @sc{gnu} C++
4273
runtime code expects to see.
4274
 
4275
@smallexample
4276
      __CTOR_LIST__ = .;
4277
      LONG((__CTOR_END__ - __CTOR_LIST__) / 4 - 2)
4278
      *(.ctors)
4279
      LONG(0)
4280
      __CTOR_END__ = .;
4281
      __DTOR_LIST__ = .;
4282
      LONG((__DTOR_END__ - __DTOR_LIST__) / 4 - 2)
4283
      *(.dtors)
4284
      LONG(0)
4285
      __DTOR_END__ = .;
4286
@end smallexample
4287
 
4288
If you are using the @sc{gnu} C++ support for initialization priority,
4289
which provides some control over the order in which global constructors
4290
are run, you must sort the constructors at link time to ensure that they
4291
are executed in the correct order.  When using the @code{CONSTRUCTORS}
4292
command, use @samp{SORT_BY_NAME(CONSTRUCTORS)} instead.  When using the
4293
@code{.ctors} and @code{.dtors} sections, use @samp{*(SORT_BY_NAME(.ctors))} and
4294
@samp{*(SORT_BY_NAME(.dtors))} instead of just @samp{*(.ctors)} and
4295
@samp{*(.dtors)}.
4296
 
4297
Normally the compiler and linker will handle these issues automatically,
4298
and you will not need to concern yourself with them.  However, you may
4299
need to consider this if you are using C++ and writing your own linker
4300
scripts.
4301
 
4302
@end table
4303
 
4304
@node Output Section Discarding
4305
@subsection Output Section Discarding
4306
@cindex discarding sections
4307
@cindex sections, discarding
4308
@cindex removing sections
4309
The linker will not create output sections with no contents.  This is
4310
for convenience when referring to input sections that may or may not
4311
be present in any of the input files.  For example:
4312
@smallexample
4313
.foo : @{ *(.foo) @}
4314
@end smallexample
4315
@noindent
4316
will only create a @samp{.foo} section in the output file if there is a
4317
@samp{.foo} section in at least one input file, and if the input
4318
sections are not all empty.  Other link script directives that allocate
4319
space in an output section will also create the output section.
4320
 
4321
The linker will ignore address assignments (@pxref{Output Section Address})
4322
on discarded output sections, except when the linker script defines
4323
symbols in the output section.  In that case the linker will obey
4324
the address assignments, possibly advancing dot even though the
4325
section is discarded.
4326
 
4327
@cindex /DISCARD/
4328
The special output section name @samp{/DISCARD/} may be used to discard
4329
input sections.  Any input sections which are assigned to an output
4330
section named @samp{/DISCARD/} are not included in the output file.
4331
 
4332
@node Output Section Attributes
4333
@subsection Output Section Attributes
4334
@cindex output section attributes
4335
We showed above that the full description of an output section looked
4336
like this:
4337
 
4338
@smallexample
4339
@group
4340
@var{section} [@var{address}] [(@var{type})] :
4341
  [AT(@var{lma})]
4342
  [ALIGN(@var{section_align})]
4343
  [SUBALIGN(@var{subsection_align})]
4344
  [@var{constraint}]
4345
  @{
4346
    @var{output-section-command}
4347
    @var{output-section-command}
4348
    @dots{}
4349
  @} [>@var{region}] [AT>@var{lma_region}] [:@var{phdr} :@var{phdr} @dots{}] [=@var{fillexp}]
4350
@end group
4351
@end smallexample
4352
 
4353
We've already described @var{section}, @var{address}, and
4354
@var{output-section-command}.  In this section we will describe the
4355
remaining section attributes.
4356
 
4357
@menu
4358
* Output Section Type::         Output section type
4359
* Output Section LMA::          Output section LMA
4360
* Forced Output Alignment::     Forced Output Alignment
4361
* Forced Input Alignment::      Forced Input Alignment
4362
* Output Section Constraint::   Output section constraint
4363
* Output Section Region::       Output section region
4364
* Output Section Phdr::         Output section phdr
4365
* Output Section Fill::         Output section fill
4366
@end menu
4367
 
4368
@node Output Section Type
4369
@subsubsection Output Section Type
4370
Each output section may have a type.  The type is a keyword in
4371
parentheses.  The following types are defined:
4372
 
4373
@table @code
4374
@item NOLOAD
4375
The section should be marked as not loadable, so that it will not be
4376
loaded into memory when the program is run.
4377
@item DSECT
4378
@itemx COPY
4379
@itemx INFO
4380
@itemx OVERLAY
4381
These type names are supported for backward compatibility, and are
4382
rarely used.  They all have the same effect: the section should be
4383
marked as not allocatable, so that no memory is allocated for the
4384
section when the program is run.
4385
@end table
4386
 
4387
@kindex NOLOAD
4388
@cindex prevent unnecessary loading
4389
@cindex loading, preventing
4390
The linker normally sets the attributes of an output section based on
4391
the input sections which map into it.  You can override this by using
4392
the section type.  For example, in the script sample below, the
4393
@samp{ROM} section is addressed at memory location @samp{0} and does not
4394
need to be loaded when the program is run.
4395
@smallexample
4396
@group
4397
SECTIONS @{
4398
  ROM 0 (NOLOAD) : @{ @dots{} @}
4399
  @dots{}
4400
@}
4401
@end group
4402
@end smallexample
4403
 
4404
@node Output Section LMA
4405
@subsubsection Output Section LMA
4406
@kindex AT>@var{lma_region}
4407
@kindex AT(@var{lma})
4408
@cindex load address
4409
@cindex section load address
4410
Every section has a virtual address (VMA) and a load address (LMA); see
4411
@ref{Basic Script Concepts}.  The virtual address is specified by the
4412
@pxref{Output Section Address} described earlier.  The load address is
4413
specified by the @code{AT} or @code{AT>} keywords.  Specifying a load
4414
address is optional.
4415
 
4416
The @code{AT} keyword takes an expression as an argument.  This
4417
specifies the exact load address of the section.  The @code{AT>} keyword
4418
takes the name of a memory region as an argument.  @xref{MEMORY}.  The
4419
load address of the section is set to the next free address in the
4420
region, aligned to the section's alignment requirements.
4421
 
4422
If neither @code{AT} nor @code{AT>} is specified for an allocatable
4423
section, the linker will use the following heuristic to determine the
4424
load address:
4425
 
4426
@itemize @bullet
4427
@item
4428
If the section has a specific VMA address, then this is used as
4429
the LMA address as well.
4430
 
4431
@item
4432
If the section is not allocatable then its LMA is set to its VMA.
4433
 
4434
@item
4435
Otherwise if a memory region can be found that is compatible
4436
with the current section, and this region contains at least one
4437
section, then the LMA is set so the difference between the
4438
VMA and LMA is the same as the difference between the VMA and LMA of
4439
the last section in the located region.
4440
 
4441
@item
4442
If no memory regions have been declared then a default region
4443
that covers the entire address space is used in the previous step.
4444
 
4445
@item
4446
If no suitable region could be found, or there was no previous
4447
section then the LMA is set equal to the VMA.
4448
@end itemize
4449
 
4450
@cindex ROM initialized data
4451
@cindex initialized data in ROM
4452
This feature is designed to make it easy to build a ROM image.  For
4453
example, the following linker script creates three output sections: one
4454
called @samp{.text}, which starts at @code{0x1000}, one called
4455
@samp{.mdata}, which is loaded at the end of the @samp{.text} section
4456
even though its VMA is @code{0x2000}, and one called @samp{.bss} to hold
4457
uninitialized data at address @code{0x3000}.  The symbol @code{_data} is
4458
defined with the value @code{0x2000}, which shows that the location
4459
counter holds the VMA value, not the LMA value.
4460
 
4461
@smallexample
4462
@group
4463
SECTIONS
4464
  @{
4465
  .text 0x1000 : @{ *(.text) _etext = . ; @}
4466
  .mdata 0x2000 :
4467
    AT ( ADDR (.text) + SIZEOF (.text) )
4468
    @{ _data = . ; *(.data); _edata = . ;  @}
4469
  .bss 0x3000 :
4470
    @{ _bstart = . ;  *(.bss) *(COMMON) ; _bend = . ;@}
4471
@}
4472
@end group
4473
@end smallexample
4474
 
4475
The run-time initialization code for use with a program generated with
4476
this linker script would include something like the following, to copy
4477
the initialized data from the ROM image to its runtime address.  Notice
4478
how this code takes advantage of the symbols defined by the linker
4479
script.
4480
 
4481
@smallexample
4482
@group
4483
extern char _etext, _data, _edata, _bstart, _bend;
4484
char *src = &_etext;
4485
char *dst = &_data;
4486
 
4487
/* ROM has data at end of text; copy it.  */
4488
while (dst < &_edata)
4489
  *dst++ = *src++;
4490
 
4491
/* Zero bss.  */
4492
for (dst = &_bstart; dst< &_bend; dst++)
4493
  *dst = 0;
4494
@end group
4495
@end smallexample
4496
 
4497
@node Forced Output Alignment
4498
@subsubsection Forced Output Alignment
4499
@kindex ALIGN(@var{section_align})
4500
@cindex forcing output section alignment
4501
@cindex output section alignment
4502
You can increase an output section's alignment by using ALIGN.
4503
 
4504
@node Forced Input Alignment
4505
@subsubsection Forced Input Alignment
4506
@kindex SUBALIGN(@var{subsection_align})
4507
@cindex forcing input section alignment
4508
@cindex input section alignment
4509
You can force input section alignment within an output section by using
4510
SUBALIGN.  The value specified overrides any alignment given by input
4511
sections, whether larger or smaller.
4512
 
4513
@node Output Section Constraint
4514
@subsubsection Output Section Constraint
4515
@kindex ONLY_IF_RO
4516
@kindex ONLY_IF_RW
4517
@cindex constraints on output sections
4518
You can specify that an output section should only be created if all
4519
of its input sections are read-only or all of its input sections are
4520
read-write by using the keyword @code{ONLY_IF_RO} and
4521
@code{ONLY_IF_RW} respectively.
4522
 
4523
@node Output Section Region
4524
@subsubsection Output Section Region
4525
@kindex >@var{region}
4526
@cindex section, assigning to memory region
4527
@cindex memory regions and sections
4528
You can assign a section to a previously defined region of memory by
4529
using @samp{>@var{region}}.  @xref{MEMORY}.
4530
 
4531
Here is a simple example:
4532
@smallexample
4533
@group
4534
MEMORY @{ rom : ORIGIN = 0x1000, LENGTH = 0x1000 @}
4535
SECTIONS @{ ROM : @{ *(.text) @} >rom @}
4536
@end group
4537
@end smallexample
4538
 
4539
@node Output Section Phdr
4540
@subsubsection Output Section Phdr
4541
@kindex :@var{phdr}
4542
@cindex section, assigning to program header
4543
@cindex program headers and sections
4544
You can assign a section to a previously defined program segment by
4545
using @samp{:@var{phdr}}.  @xref{PHDRS}.  If a section is assigned to
4546
one or more segments, then all subsequent allocated sections will be
4547
assigned to those segments as well, unless they use an explicitly
4548
@code{:@var{phdr}} modifier.  You can use @code{:NONE} to tell the
4549
linker to not put the section in any segment at all.
4550
 
4551
Here is a simple example:
4552
@smallexample
4553
@group
4554
PHDRS @{ text PT_LOAD ; @}
4555
SECTIONS @{ .text : @{ *(.text) @} :text @}
4556
@end group
4557
@end smallexample
4558
 
4559
@node Output Section Fill
4560
@subsubsection Output Section Fill
4561
@kindex =@var{fillexp}
4562
@cindex section fill pattern
4563
@cindex fill pattern, entire section
4564
You can set the fill pattern for an entire section by using
4565
@samp{=@var{fillexp}}.  @var{fillexp} is an expression
4566
(@pxref{Expressions}).  Any otherwise unspecified regions of memory
4567
within the output section (for example, gaps left due to the required
4568
alignment of input sections) will be filled with the value, repeated as
4569
necessary.  If the fill expression is a simple hex number, ie. a string
4570
of hex digit starting with @samp{0x} and without a trailing @samp{k} or @samp{M}, then
4571
an arbitrarily long sequence of hex digits can be used to specify the
4572
fill pattern;  Leading zeros become part of the pattern too.  For all
4573
other cases, including extra parentheses or a unary @code{+}, the fill
4574
pattern is the four least significant bytes of the value of the
4575
expression.  In all cases, the number is big-endian.
4576
 
4577
You can also change the fill value with a @code{FILL} command in the
4578
output section commands; (@pxref{Output Section Data}).
4579
 
4580
Here is a simple example:
4581
@smallexample
4582
@group
4583
SECTIONS @{ .text : @{ *(.text) @} =0x90909090 @}
4584
@end group
4585
@end smallexample
4586
 
4587
@node Overlay Description
4588
@subsection Overlay Description
4589
@kindex OVERLAY
4590
@cindex overlays
4591
An overlay description provides an easy way to describe sections which
4592
are to be loaded as part of a single memory image but are to be run at
4593
the same memory address.  At run time, some sort of overlay manager will
4594
copy the overlaid sections in and out of the runtime memory address as
4595
required, perhaps by simply manipulating addressing bits.  This approach
4596
can be useful, for example, when a certain region of memory is faster
4597
than another.
4598
 
4599
Overlays are described using the @code{OVERLAY} command.  The
4600
@code{OVERLAY} command is used within a @code{SECTIONS} command, like an
4601
output section description.  The full syntax of the @code{OVERLAY}
4602
command is as follows:
4603
@smallexample
4604
@group
4605
OVERLAY [@var{start}] : [NOCROSSREFS] [AT ( @var{ldaddr} )]
4606
  @{
4607
    @var{secname1}
4608
      @{
4609
        @var{output-section-command}
4610
        @var{output-section-command}
4611
        @dots{}
4612
      @} [:@var{phdr}@dots{}] [=@var{fill}]
4613
    @var{secname2}
4614
      @{
4615
        @var{output-section-command}
4616
        @var{output-section-command}
4617
        @dots{}
4618
      @} [:@var{phdr}@dots{}] [=@var{fill}]
4619
    @dots{}
4620
  @} [>@var{region}] [:@var{phdr}@dots{}] [=@var{fill}]
4621
@end group
4622
@end smallexample
4623
 
4624
Everything is optional except @code{OVERLAY} (a keyword), and each
4625
section must have a name (@var{secname1} and @var{secname2} above).  The
4626
section definitions within the @code{OVERLAY} construct are identical to
4627
those within the general @code{SECTIONS} contruct (@pxref{SECTIONS}),
4628
except that no addresses and no memory regions may be defined for
4629
sections within an @code{OVERLAY}.
4630
 
4631
The sections are all defined with the same starting address.  The load
4632
addresses of the sections are arranged such that they are consecutive in
4633
memory starting at the load address used for the @code{OVERLAY} as a
4634
whole (as with normal section definitions, the load address is optional,
4635
and defaults to the start address; the start address is also optional,
4636
and defaults to the current value of the location counter).
4637
 
4638
If the @code{NOCROSSREFS} keyword is used, and there any references
4639
among the sections, the linker will report an error.  Since the sections
4640
all run at the same address, it normally does not make sense for one
4641
section to refer directly to another.  @xref{Miscellaneous Commands,
4642
NOCROSSREFS}.
4643
 
4644
For each section within the @code{OVERLAY}, the linker automatically
4645
provides two symbols.  The symbol @code{__load_start_@var{secname}} is
4646
defined as the starting load address of the section.  The symbol
4647
@code{__load_stop_@var{secname}} is defined as the final load address of
4648
the section.  Any characters within @var{secname} which are not legal
4649
within C identifiers are removed.  C (or assembler) code may use these
4650
symbols to move the overlaid sections around as necessary.
4651
 
4652
At the end of the overlay, the value of the location counter is set to
4653
the start address of the overlay plus the size of the largest section.
4654
 
4655
Here is an example.  Remember that this would appear inside a
4656
@code{SECTIONS} construct.
4657
@smallexample
4658
@group
4659
  OVERLAY 0x1000 : AT (0x4000)
4660
   @{
4661
     .text0 @{ o1/*.o(.text) @}
4662
     .text1 @{ o2/*.o(.text) @}
4663
   @}
4664
@end group
4665
@end smallexample
4666
@noindent
4667
This will define both @samp{.text0} and @samp{.text1} to start at
4668
address 0x1000.  @samp{.text0} will be loaded at address 0x4000, and
4669
@samp{.text1} will be loaded immediately after @samp{.text0}.  The
4670
following symbols will be defined if referenced: @code{__load_start_text0},
4671
@code{__load_stop_text0}, @code{__load_start_text1},
4672
@code{__load_stop_text1}.
4673
 
4674
C code to copy overlay @code{.text1} into the overlay area might look
4675
like the following.
4676
 
4677
@smallexample
4678
@group
4679
  extern char __load_start_text1, __load_stop_text1;
4680
  memcpy ((char *) 0x1000, &__load_start_text1,
4681
          &__load_stop_text1 - &__load_start_text1);
4682
@end group
4683
@end smallexample
4684
 
4685
Note that the @code{OVERLAY} command is just syntactic sugar, since
4686
everything it does can be done using the more basic commands.  The above
4687
example could have been written identically as follows.
4688
 
4689
@smallexample
4690
@group
4691
  .text0 0x1000 : AT (0x4000) @{ o1/*.o(.text) @}
4692
  PROVIDE (__load_start_text0 = LOADADDR (.text0));
4693
  PROVIDE (__load_stop_text0 = LOADADDR (.text0) + SIZEOF (.text0));
4694
  .text1 0x1000 : AT (0x4000 + SIZEOF (.text0)) @{ o2/*.o(.text) @}
4695
  PROVIDE (__load_start_text1 = LOADADDR (.text1));
4696
  PROVIDE (__load_stop_text1 = LOADADDR (.text1) + SIZEOF (.text1));
4697
  . = 0x1000 + MAX (SIZEOF (.text0), SIZEOF (.text1));
4698
@end group
4699
@end smallexample
4700
 
4701
@node MEMORY
4702
@section MEMORY Command
4703
@kindex MEMORY
4704
@cindex memory regions
4705
@cindex regions of memory
4706
@cindex allocating memory
4707
@cindex discontinuous memory
4708
The linker's default configuration permits allocation of all available
4709
memory.  You can override this by using the @code{MEMORY} command.
4710
 
4711
The @code{MEMORY} command describes the location and size of blocks of
4712
memory in the target.  You can use it to describe which memory regions
4713
may be used by the linker, and which memory regions it must avoid.  You
4714
can then assign sections to particular memory regions.  The linker will
4715
set section addresses based on the memory regions, and will warn about
4716
regions that become too full.  The linker will not shuffle sections
4717
around to fit into the available regions.
4718
 
4719
A linker script may contain at most one use of the @code{MEMORY}
4720
command.  However, you can define as many blocks of memory within it as
4721
you wish.  The syntax is:
4722
@smallexample
4723
@group
4724
MEMORY
4725
  @{
4726
    @var{name} [(@var{attr})] : ORIGIN = @var{origin}, LENGTH = @var{len}
4727
    @dots{}
4728
  @}
4729
@end group
4730
@end smallexample
4731
 
4732
The @var{name} is a name used in the linker script to refer to the
4733
region.  The region name has no meaning outside of the linker script.
4734
Region names are stored in a separate name space, and will not conflict
4735
with symbol names, file names, or section names.  Each memory region
4736
must have a distinct name within the @code{MEMORY} command.  However you can
4737
add later alias names to existing memory regions with the @ref{REGION_ALIAS}
4738
command.
4739
 
4740
@cindex memory region attributes
4741
The @var{attr} string is an optional list of attributes that specify
4742
whether to use a particular memory region for an input section which is
4743
not explicitly mapped in the linker script.  As described in
4744
@ref{SECTIONS}, if you do not specify an output section for some input
4745
section, the linker will create an output section with the same name as
4746
the input section.  If you define region attributes, the linker will use
4747
them to select the memory region for the output section that it creates.
4748
 
4749
The @var{attr} string must consist only of the following characters:
4750
@table @samp
4751
@item R
4752
Read-only section
4753
@item W
4754
Read/write section
4755
@item X
4756
Executable section
4757
@item A
4758
Allocatable section
4759
@item I
4760
Initialized section
4761
@item L
4762
Same as @samp{I}
4763
@item !
4764
Invert the sense of any of the attributes that follow
4765
@end table
4766
 
4767
If a unmapped section matches any of the listed attributes other than
4768
@samp{!}, it will be placed in the memory region.  The @samp{!}
4769
attribute reverses this test, so that an unmapped section will be placed
4770
in the memory region only if it does not match any of the listed
4771
attributes.
4772
 
4773
@kindex ORIGIN =
4774
@kindex o =
4775
@kindex org =
4776
The @var{origin} is an numerical expression for the start address of
4777
the memory region.  The expression must evaluate to a constant and it
4778
cannot involve any symbols.  The keyword @code{ORIGIN} may be
4779
abbreviated to @code{org} or @code{o} (but not, for example,
4780
@code{ORG}).
4781
 
4782
@kindex LENGTH =
4783
@kindex len =
4784
@kindex l =
4785
The @var{len} is an expression for the size in bytes of the memory
4786
region.  As with the @var{origin} expression, the expression must
4787
be numerical only and must evaluate to a constant.  The keyword
4788
@code{LENGTH} may be abbreviated to @code{len} or @code{l}.
4789
 
4790
In the following example, we specify that there are two memory regions
4791
available for allocation: one starting at @samp{0} for 256 kilobytes,
4792
and the other starting at @samp{0x40000000} for four megabytes.  The
4793
linker will place into the @samp{rom} memory region every section which
4794
is not explicitly mapped into a memory region, and is either read-only
4795
or executable.  The linker will place other sections which are not
4796
explicitly mapped into a memory region into the @samp{ram} memory
4797
region.
4798
 
4799
@smallexample
4800
@group
4801
MEMORY
4802
  @{
4803
    rom (rx)  : ORIGIN = 0, LENGTH = 256K
4804
    ram (!rx) : org = 0x40000000, l = 4M
4805
  @}
4806
@end group
4807
@end smallexample
4808
 
4809
Once you define a memory region, you can direct the linker to place
4810
specific output sections into that memory region by using the
4811
@samp{>@var{region}} output section attribute.  For example, if you have
4812
a memory region named @samp{mem}, you would use @samp{>mem} in the
4813
output section definition.  @xref{Output Section Region}.  If no address
4814
was specified for the output section, the linker will set the address to
4815
the next available address within the memory region.  If the combined
4816
output sections directed to a memory region are too large for the
4817
region, the linker will issue an error message.
4818
 
4819
It is possible to access the origin and length of a memory in an
4820
expression via the @code{ORIGIN(@var{memory})} and
4821
@code{LENGTH(@var{memory})} functions:
4822
 
4823
@smallexample
4824
@group
4825
  _fstack = ORIGIN(ram) + LENGTH(ram) - 4;
4826
@end group
4827
@end smallexample
4828
 
4829
@node PHDRS
4830
@section PHDRS Command
4831
@kindex PHDRS
4832
@cindex program headers
4833
@cindex ELF program headers
4834
@cindex program segments
4835
@cindex segments, ELF
4836
The ELF object file format uses @dfn{program headers}, also knows as
4837
@dfn{segments}.  The program headers describe how the program should be
4838
loaded into memory.  You can print them out by using the @code{objdump}
4839
program with the @samp{-p} option.
4840
 
4841
When you run an ELF program on a native ELF system, the system loader
4842
reads the program headers in order to figure out how to load the
4843
program.  This will only work if the program headers are set correctly.
4844
This manual does not describe the details of how the system loader
4845
interprets program headers; for more information, see the ELF ABI.
4846
 
4847
The linker will create reasonable program headers by default.  However,
4848
in some cases, you may need to specify the program headers more
4849
precisely.  You may use the @code{PHDRS} command for this purpose.  When
4850
the linker sees the @code{PHDRS} command in the linker script, it will
4851
not create any program headers other than the ones specified.
4852
 
4853
The linker only pays attention to the @code{PHDRS} command when
4854
generating an ELF output file.  In other cases, the linker will simply
4855
ignore @code{PHDRS}.
4856
 
4857
This is the syntax of the @code{PHDRS} command.  The words @code{PHDRS},
4858
@code{FILEHDR}, @code{AT}, and @code{FLAGS} are keywords.
4859
 
4860
@smallexample
4861
@group
4862
PHDRS
4863
@{
4864
  @var{name} @var{type} [ FILEHDR ] [ PHDRS ] [ AT ( @var{address} ) ]
4865
        [ FLAGS ( @var{flags} ) ] ;
4866
@}
4867
@end group
4868
@end smallexample
4869
 
4870
The @var{name} is used only for reference in the @code{SECTIONS} command
4871
of the linker script.  It is not put into the output file.  Program
4872
header names are stored in a separate name space, and will not conflict
4873
with symbol names, file names, or section names.  Each program header
4874
must have a distinct name.  The headers are processed in order and it
4875
is usual for them to map to sections in ascending load address order.
4876
 
4877
Certain program header types describe segments of memory which the
4878
system loader will load from the file.  In the linker script, you
4879
specify the contents of these segments by placing allocatable output
4880
sections in the segments.  You use the @samp{:@var{phdr}} output section
4881
attribute to place a section in a particular segment.  @xref{Output
4882
Section Phdr}.
4883
 
4884
It is normal to put certain sections in more than one segment.  This
4885
merely implies that one segment of memory contains another.  You may
4886
repeat @samp{:@var{phdr}}, using it once for each segment which should
4887
contain the section.
4888
 
4889
If you place a section in one or more segments using @samp{:@var{phdr}},
4890
then the linker will place all subsequent allocatable sections which do
4891
not specify @samp{:@var{phdr}} in the same segments.  This is for
4892
convenience, since generally a whole set of contiguous sections will be
4893
placed in a single segment.  You can use @code{:NONE} to override the
4894
default segment and tell the linker to not put the section in any
4895
segment at all.
4896
 
4897
@kindex FILEHDR
4898
@kindex PHDRS
4899
You may use the @code{FILEHDR} and @code{PHDRS} keywords after
4900
the program header type to further describe the contents of the segment.
4901
The @code{FILEHDR} keyword means that the segment should include the ELF
4902
file header.  The @code{PHDRS} keyword means that the segment should
4903
include the ELF program headers themselves.  If applied to a loadable
4904
segment (@code{PT_LOAD}), all prior loadable segments must have one of
4905
these keywords.
4906
 
4907
The @var{type} may be one of the following.  The numbers indicate the
4908
value of the keyword.
4909
 
4910
@table @asis
4911
@item @code{PT_NULL} (0)
4912
Indicates an unused program header.
4913
 
4914
@item @code{PT_LOAD} (1)
4915
Indicates that this program header describes a segment to be loaded from
4916
the file.
4917
 
4918
@item @code{PT_DYNAMIC} (2)
4919
Indicates a segment where dynamic linking information can be found.
4920
 
4921
@item @code{PT_INTERP} (3)
4922
Indicates a segment where the name of the program interpreter may be
4923
found.
4924
 
4925
@item @code{PT_NOTE} (4)
4926
Indicates a segment holding note information.
4927
 
4928
@item @code{PT_SHLIB} (5)
4929
A reserved program header type, defined but not specified by the ELF
4930
ABI.
4931
 
4932
@item @code{PT_PHDR} (6)
4933
Indicates a segment where the program headers may be found.
4934
 
4935
@item @var{expression}
4936
An expression giving the numeric type of the program header.  This may
4937
be used for types not defined above.
4938
@end table
4939
 
4940
You can specify that a segment should be loaded at a particular address
4941
in memory by using an @code{AT} expression.  This is identical to the
4942
@code{AT} command used as an output section attribute (@pxref{Output
4943
Section LMA}).  The @code{AT} command for a program header overrides the
4944
output section attribute.
4945
 
4946
The linker will normally set the segment flags based on the sections
4947
which comprise the segment.  You may use the @code{FLAGS} keyword to
4948
explicitly specify the segment flags.  The value of @var{flags} must be
4949
an integer.  It is used to set the @code{p_flags} field of the program
4950
header.
4951
 
4952
Here is an example of @code{PHDRS}.  This shows a typical set of program
4953
headers used on a native ELF system.
4954
 
4955
@example
4956
@group
4957
PHDRS
4958
@{
4959
  headers PT_PHDR PHDRS ;
4960
  interp PT_INTERP ;
4961
  text PT_LOAD FILEHDR PHDRS ;
4962
  data PT_LOAD ;
4963
  dynamic PT_DYNAMIC ;
4964
@}
4965
 
4966
SECTIONS
4967
@{
4968
  . = SIZEOF_HEADERS;
4969
  .interp : @{ *(.interp) @} :text :interp
4970
  .text : @{ *(.text) @} :text
4971
  .rodata : @{ *(.rodata) @} /* defaults to :text */
4972
  @dots{}
4973
  . = . + 0x1000; /* move to a new page in memory */
4974
  .data : @{ *(.data) @} :data
4975
  .dynamic : @{ *(.dynamic) @} :data :dynamic
4976
  @dots{}
4977
@}
4978
@end group
4979
@end example
4980
 
4981
@node VERSION
4982
@section VERSION Command
4983
@kindex VERSION @{script text@}
4984
@cindex symbol versions
4985
@cindex version script
4986
@cindex versions of symbols
4987
The linker supports symbol versions when using ELF.  Symbol versions are
4988
only useful when using shared libraries.  The dynamic linker can use
4989
symbol versions to select a specific version of a function when it runs
4990
a program that may have been linked against an earlier version of the
4991
shared library.
4992
 
4993
You can include a version script directly in the main linker script, or
4994
you can supply the version script as an implicit linker script.  You can
4995
also use the @samp{--version-script} linker option.
4996
 
4997
The syntax of the @code{VERSION} command is simply
4998
@smallexample
4999
VERSION @{ version-script-commands @}
5000
@end smallexample
5001
 
5002
The format of the version script commands is identical to that used by
5003
Sun's linker in Solaris 2.5.  The version script defines a tree of
5004
version nodes.  You specify the node names and interdependencies in the
5005
version script.  You can specify which symbols are bound to which
5006
version nodes, and you can reduce a specified set of symbols to local
5007
scope so that they are not globally visible outside of the shared
5008
library.
5009
 
5010
The easiest way to demonstrate the version script language is with a few
5011
examples.
5012
 
5013
@smallexample
5014
VERS_1.1 @{
5015
         global:
5016
                 foo1;
5017
         local:
5018
                 old*;
5019
                 original*;
5020
                 new*;
5021
@};
5022
 
5023
VERS_1.2 @{
5024
                 foo2;
5025
@} VERS_1.1;
5026
 
5027
VERS_2.0 @{
5028
                 bar1; bar2;
5029
         extern "C++" @{
5030
                 ns::*;
5031
                 "f(int, double)";
5032
         @};
5033
@} VERS_1.2;
5034
@end smallexample
5035
 
5036
This example version script defines three version nodes.  The first
5037
version node defined is @samp{VERS_1.1}; it has no other dependencies.
5038
The script binds the symbol @samp{foo1} to @samp{VERS_1.1}.  It reduces
5039
a number of symbols to local scope so that they are not visible outside
5040
of the shared library; this is done using wildcard patterns, so that any
5041
symbol whose name begins with @samp{old}, @samp{original}, or @samp{new}
5042
is matched.  The wildcard patterns available are the same as those used
5043
in the shell when matching filenames (also known as ``globbing'').
5044
However, if you specify the symbol name inside double quotes, then the
5045
name is treated as literal, rather than as a glob pattern.
5046
 
5047
Next, the version script defines node @samp{VERS_1.2}.  This node
5048
depends upon @samp{VERS_1.1}.  The script binds the symbol @samp{foo2}
5049
to the version node @samp{VERS_1.2}.
5050
 
5051
Finally, the version script defines node @samp{VERS_2.0}.  This node
5052
depends upon @samp{VERS_1.2}.  The scripts binds the symbols @samp{bar1}
5053
and @samp{bar2} are bound to the version node @samp{VERS_2.0}.
5054
 
5055
When the linker finds a symbol defined in a library which is not
5056
specifically bound to a version node, it will effectively bind it to an
5057
unspecified base version of the library.  You can bind all otherwise
5058
unspecified symbols to a given version node by using @samp{global: *;}
5059
somewhere in the version script.  Note that it's slightly crazy to use
5060
wildcards in a global spec except on the last version node.  Global
5061
wildcards elsewhere run the risk of accidentally adding symbols to the
5062
set exported for an old version.  That's wrong since older versions
5063
ought to have a fixed set of symbols.
5064
 
5065
The names of the version nodes have no specific meaning other than what
5066
they might suggest to the person reading them.  The @samp{2.0} version
5067
could just as well have appeared in between @samp{1.1} and @samp{1.2}.
5068
However, this would be a confusing way to write a version script.
5069
 
5070
Node name can be omitted, provided it is the only version node
5071
in the version script.  Such version script doesn't assign any versions to
5072
symbols, only selects which symbols will be globally visible out and which
5073
won't.
5074
 
5075
@smallexample
5076
@{ global: foo; bar; local: *; @};
5077
@end smallexample
5078
 
5079
When you link an application against a shared library that has versioned
5080
symbols, the application itself knows which version of each symbol it
5081
requires, and it also knows which version nodes it needs from each
5082
shared library it is linked against.  Thus at runtime, the dynamic
5083
loader can make a quick check to make sure that the libraries you have
5084
linked against do in fact supply all of the version nodes that the
5085
application will need to resolve all of the dynamic symbols.  In this
5086
way it is possible for the dynamic linker to know with certainty that
5087
all external symbols that it needs will be resolvable without having to
5088
search for each symbol reference.
5089
 
5090
The symbol versioning is in effect a much more sophisticated way of
5091
doing minor version checking that SunOS does.  The fundamental problem
5092
that is being addressed here is that typically references to external
5093
functions are bound on an as-needed basis, and are not all bound when
5094
the application starts up.  If a shared library is out of date, a
5095
required interface may be missing; when the application tries to use
5096
that interface, it may suddenly and unexpectedly fail.  With symbol
5097
versioning, the user will get a warning when they start their program if
5098
the libraries being used with the application are too old.
5099
 
5100
There are several GNU extensions to Sun's versioning approach.  The
5101
first of these is the ability to bind a symbol to a version node in the
5102
source file where the symbol is defined instead of in the versioning
5103
script.  This was done mainly to reduce the burden on the library
5104
maintainer.  You can do this by putting something like:
5105
@smallexample
5106
__asm__(".symver original_foo,foo@@VERS_1.1");
5107
@end smallexample
5108
@noindent
5109
in the C source file.  This renames the function @samp{original_foo} to
5110
be an alias for @samp{foo} bound to the version node @samp{VERS_1.1}.
5111
The @samp{local:} directive can be used to prevent the symbol
5112
@samp{original_foo} from being exported. A @samp{.symver} directive
5113
takes precedence over a version script.
5114
 
5115
The second GNU extension is to allow multiple versions of the same
5116
function to appear in a given shared library.  In this way you can make
5117
an incompatible change to an interface without increasing the major
5118
version number of the shared library, while still allowing applications
5119
linked against the old interface to continue to function.
5120
 
5121
To do this, you must use multiple @samp{.symver} directives in the
5122
source file.  Here is an example:
5123
 
5124
@smallexample
5125
__asm__(".symver original_foo,foo@@");
5126
__asm__(".symver old_foo,foo@@VERS_1.1");
5127
__asm__(".symver old_foo1,foo@@VERS_1.2");
5128
__asm__(".symver new_foo,foo@@@@VERS_2.0");
5129
@end smallexample
5130
 
5131
In this example, @samp{foo@@} represents the symbol @samp{foo} bound to the
5132
unspecified base version of the symbol.  The source file that contains this
5133
example would define 4 C functions: @samp{original_foo}, @samp{old_foo},
5134
@samp{old_foo1}, and @samp{new_foo}.
5135
 
5136
When you have multiple definitions of a given symbol, there needs to be
5137
some way to specify a default version to which external references to
5138
this symbol will be bound.  You can do this with the
5139
@samp{foo@@@@VERS_2.0} type of @samp{.symver} directive.  You can only
5140
declare one version of a symbol as the default in this manner; otherwise
5141
you would effectively have multiple definitions of the same symbol.
5142
 
5143
If you wish to bind a reference to a specific version of the symbol
5144
within the shared library, you can use the aliases of convenience
5145
(i.e., @samp{old_foo}), or you can use the @samp{.symver} directive to
5146
specifically bind to an external version of the function in question.
5147
 
5148
You can also specify the language in the version script:
5149
 
5150
@smallexample
5151
VERSION extern "lang" @{ version-script-commands @}
5152
@end smallexample
5153
 
5154
The supported @samp{lang}s are @samp{C}, @samp{C++}, and @samp{Java}.
5155
The linker will iterate over the list of symbols at the link time and
5156
demangle them according to @samp{lang} before matching them to the
5157
patterns specified in @samp{version-script-commands}.  The default
5158
@samp{lang} is @samp{C}.
5159
 
5160
Demangled names may contains spaces and other special characters.  As
5161
described above, you can use a glob pattern to match demangled names,
5162
or you can use a double-quoted string to match the string exactly.  In
5163
the latter case, be aware that minor differences (such as differing
5164
whitespace) between the version script and the demangler output will
5165
cause a mismatch.  As the exact string generated by the demangler
5166
might change in the future, even if the mangled name does not, you
5167
should check that all of your version directives are behaving as you
5168
expect when you upgrade.
5169
 
5170
@node Expressions
5171
@section Expressions in Linker Scripts
5172
@cindex expressions
5173
@cindex arithmetic
5174
The syntax for expressions in the linker script language is identical to
5175
that of C expressions.  All expressions are evaluated as integers.  All
5176
expressions are evaluated in the same size, which is 32 bits if both the
5177
host and target are 32 bits, and is otherwise 64 bits.
5178
 
5179
You can use and set symbol values in expressions.
5180
 
5181
The linker defines several special purpose builtin functions for use in
5182
expressions.
5183
 
5184
@menu
5185
* Constants::                   Constants
5186
* Symbolic Constants::          Symbolic constants
5187
* Symbols::                     Symbol Names
5188
* Orphan Sections::             Orphan Sections
5189
* Location Counter::            The Location Counter
5190
* Operators::                   Operators
5191
* Evaluation::                  Evaluation
5192
* Expression Section::          The Section of an Expression
5193
* Builtin Functions::           Builtin Functions
5194
@end menu
5195
 
5196
@node Constants
5197
@subsection Constants
5198
@cindex integer notation
5199
@cindex constants in linker scripts
5200
All constants are integers.
5201
 
5202
As in C, the linker considers an integer beginning with @samp{0} to be
5203
octal, and an integer beginning with @samp{0x} or @samp{0X} to be
5204
hexadecimal.  Alternatively the linker accepts suffixes of @samp{h} or
5205
@samp{H} for hexadeciaml, @samp{o} or @samp{O} for octal, @samp{b} or
5206
@samp{B} for binary and @samp{d} or @samp{D} for decimal.  Any integer
5207
value without a prefix or a suffix is considered to be decimal.
5208
 
5209
@cindex scaled integers
5210
@cindex K and M integer suffixes
5211
@cindex M and K integer suffixes
5212
@cindex suffixes for integers
5213
@cindex integer suffixes
5214
In addition, you can use the suffixes @code{K} and @code{M} to scale a
5215
constant by
5216
@c TEXI2ROFF-KILL
5217
@ifnottex
5218
@c END TEXI2ROFF-KILL
5219
@code{1024} or @code{1024*1024}
5220
@c TEXI2ROFF-KILL
5221
@end ifnottex
5222
@tex
5223
${\rm 1024}$ or ${\rm 1024}^2$
5224
@end tex
5225
@c END TEXI2ROFF-KILL
5226
respectively.  For example, the following
5227
all refer to the same quantity:
5228
 
5229
@smallexample
5230
_fourk_1 = 4K;
5231
_fourk_2 = 4096;
5232
_fourk_3 = 0x1000;
5233
_fourk_4 = 10000o;
5234
@end smallexample
5235
 
5236
Note - the @code{K} and @code{M} suffixes cannot be used in
5237
conjunction with the base suffixes mentioned above.
5238
 
5239
@node Symbolic Constants
5240
@subsection Symbolic Constants
5241
@cindex symbolic constants
5242
@kindex CONSTANT
5243
It is possible to refer to target specific constants via the use of
5244
the @code{CONSTANT(@var{name})} operator, where @var{name} is one of:
5245
 
5246
@table @code
5247
@item MAXPAGESIZE
5248
@kindex MAXPAGESIZE
5249
The target's maximum page size.
5250
 
5251
@item COMMONPAGESIZE
5252
@kindex COMMONPAGESIZE
5253
The target's default page size.
5254
@end table
5255
 
5256
So for example:
5257
 
5258
@smallexample
5259
  .text ALIGN (CONSTANT (MAXPAGESIZE)) : @{ *(.text) @}
5260
@end smallexample
5261
 
5262
will create a text section aligned to the largest page boundary
5263
supported by the target.
5264
 
5265
@node Symbols
5266
@subsection Symbol Names
5267
@cindex symbol names
5268
@cindex names
5269
@cindex quoted symbol names
5270
@kindex "
5271
Unless quoted, symbol names start with a letter, underscore, or period
5272
and may include letters, digits, underscores, periods, and hyphens.
5273
Unquoted symbol names must not conflict with any keywords.  You can
5274
specify a symbol which contains odd characters or has the same name as a
5275
keyword by surrounding the symbol name in double quotes:
5276
@smallexample
5277
"SECTION" = 9;
5278
"with a space" = "also with a space" + 10;
5279
@end smallexample
5280
 
5281
Since symbols can contain many non-alphabetic characters, it is safest
5282
to delimit symbols with spaces.  For example, @samp{A-B} is one symbol,
5283
whereas @samp{A - B} is an expression involving subtraction.
5284
 
5285
@node Orphan Sections
5286
@subsection Orphan Sections
5287
@cindex orphan
5288
Orphan sections are sections present in the input files which
5289
are not explicitly placed into the output file by the linker
5290
script.  The linker will still copy these sections into the
5291
output file, but it has to guess as to where they should be
5292
placed.  The linker uses a simple heuristic to do this.  It
5293
attempts to place orphan sections after non-orphan sections of the
5294
same attribute, such as code vs data, loadable vs non-loadable, etc.
5295
If there is not enough room to do this then it places
5296
at the end of the file.
5297
 
5298
For ELF targets, the attribute of the section includes section type as
5299
well as section flag.
5300
 
5301
If an orphaned section's name is representable as a C identifier then
5302
the linker will automatically @pxref{PROVIDE} two symbols:
5303
__start_SECNAME and __end_SECNAME, where SECNAME is the name of the
5304
section.  These indicate the start address and end address of the
5305
orphaned section respectively.  Note: most section names are not
5306
representable as C identifiers because they contain a @samp{.}
5307
character.
5308
 
5309
@node Location Counter
5310
@subsection The Location Counter
5311
@kindex .
5312
@cindex dot
5313
@cindex location counter
5314
@cindex current output location
5315
The special linker variable @dfn{dot} @samp{.} always contains the
5316
current output location counter.  Since the @code{.} always refers to a
5317
location in an output section, it may only appear in an expression
5318
within a @code{SECTIONS} command.  The @code{.} symbol may appear
5319
anywhere that an ordinary symbol is allowed in an expression.
5320
 
5321
@cindex holes
5322
Assigning a value to @code{.} will cause the location counter to be
5323
moved.  This may be used to create holes in the output section.  The
5324
location counter may not be moved backwards inside an output section,
5325
and may not be moved backwards outside of an output section if so
5326
doing creates areas with overlapping LMAs.
5327
 
5328
@smallexample
5329
SECTIONS
5330
@{
5331
  output :
5332
    @{
5333
      file1(.text)
5334
      . = . + 1000;
5335
      file2(.text)
5336
      . += 1000;
5337
      file3(.text)
5338
    @} = 0x12345678;
5339
@}
5340
@end smallexample
5341
@noindent
5342
In the previous example, the @samp{.text} section from @file{file1} is
5343
located at the beginning of the output section @samp{output}.  It is
5344
followed by a 1000 byte gap.  Then the @samp{.text} section from
5345
@file{file2} appears, also with a 1000 byte gap following before the
5346
@samp{.text} section from @file{file3}.  The notation @samp{= 0x12345678}
5347
specifies what data to write in the gaps (@pxref{Output Section Fill}).
5348
 
5349
@cindex dot inside sections
5350
Note: @code{.} actually refers to the byte offset from the start of the
5351
current containing object.  Normally this is the @code{SECTIONS}
5352
statement, whose start address is 0, hence @code{.} can be used as an
5353
absolute address.  If @code{.} is used inside a section description
5354
however, it refers to the byte offset from the start of that section,
5355
not an absolute address.  Thus in a script like this:
5356
 
5357
@smallexample
5358
SECTIONS
5359
@{
5360
    . = 0x100
5361
    .text: @{
5362
      *(.text)
5363
      . = 0x200
5364
    @}
5365
    . = 0x500
5366
    .data: @{
5367
      *(.data)
5368
      . += 0x600
5369
    @}
5370
@}
5371
@end smallexample
5372
 
5373
The @samp{.text} section will be assigned a starting address of 0x100
5374
and a size of exactly 0x200 bytes, even if there is not enough data in
5375
the @samp{.text} input sections to fill this area.  (If there is too
5376
much data, an error will be produced because this would be an attempt to
5377
move @code{.} backwards).  The @samp{.data} section will start at 0x500
5378
and it will have an extra 0x600 bytes worth of space after the end of
5379
the values from the @samp{.data} input sections and before the end of
5380
the @samp{.data} output section itself.
5381
 
5382
@cindex dot outside sections
5383
Setting symbols to the value of the location counter outside of an
5384
output section statement can result in unexpected values if the linker
5385
needs to place orphan sections.  For example, given the following:
5386
 
5387
@smallexample
5388
SECTIONS
5389
@{
5390
    start_of_text = . ;
5391
    .text: @{ *(.text) @}
5392
    end_of_text = . ;
5393
 
5394
    start_of_data = . ;
5395
    .data: @{ *(.data) @}
5396
    end_of_data = . ;
5397
@}
5398
@end smallexample
5399
 
5400
If the linker needs to place some input section, e.g. @code{.rodata},
5401
not mentioned in the script, it might choose to place that section
5402
between @code{.text} and @code{.data}.  You might think the linker
5403
should place @code{.rodata} on the blank line in the above script, but
5404
blank lines are of no particular significance to the linker.  As well,
5405
the linker doesn't associate the above symbol names with their
5406
sections.  Instead, it assumes that all assignments or other
5407
statements belong to the previous output section, except for the
5408
special case of an assignment to @code{.}.  I.e., the linker will
5409
place the orphan @code{.rodata} section as if the script was written
5410
as follows:
5411
 
5412
@smallexample
5413
SECTIONS
5414
@{
5415
    start_of_text = . ;
5416
    .text: @{ *(.text) @}
5417
    end_of_text = . ;
5418
 
5419
    start_of_data = . ;
5420
    .rodata: @{ *(.rodata) @}
5421
    .data: @{ *(.data) @}
5422
    end_of_data = . ;
5423
@}
5424
@end smallexample
5425
 
5426
This may or may not be the script author's intention for the value of
5427
@code{start_of_data}.  One way to influence the orphan section
5428
placement is to assign the location counter to itself, as the linker
5429
assumes that an assignment to @code{.} is setting the start address of
5430
a following output section and thus should be grouped with that
5431
section.  So you could write:
5432
 
5433
@smallexample
5434
SECTIONS
5435
@{
5436
    start_of_text = . ;
5437
    .text: @{ *(.text) @}
5438
    end_of_text = . ;
5439
 
5440
    . = . ;
5441
    start_of_data = . ;
5442
    .data: @{ *(.data) @}
5443
    end_of_data = . ;
5444
@}
5445
@end smallexample
5446
 
5447
Now, the orphan @code{.rodata} section will be placed between
5448
@code{end_of_text} and @code{start_of_data}.
5449
 
5450
@need 2000
5451
@node Operators
5452
@subsection Operators
5453
@cindex operators for arithmetic
5454
@cindex arithmetic operators
5455
@cindex precedence in expressions
5456
The linker recognizes the standard C set of arithmetic operators, with
5457
the standard bindings and precedence levels:
5458
@c TEXI2ROFF-KILL
5459
@ifnottex
5460
@c END TEXI2ROFF-KILL
5461
@smallexample
5462
precedence      associativity   Operators                Notes
5463
(highest)
5464
1               left            !  -  ~                  (1)
5465
2               left            *  /  %
5466
3               left            +  -
5467
4               left            >>  <<
5468
5               left            ==  !=  >  <  <=  >=
5469
6               left            &
5470
7               left            |
5471
8               left            &&
5472
9               left            ||
5473
10              right           ? :
5474
11              right           &=  +=  -=  *=  /=       (2)
5475
(lowest)
5476
@end smallexample
5477
Notes:
5478
(1) Prefix operators
5479
(2) @xref{Assignments}.
5480
@c TEXI2ROFF-KILL
5481
@end ifnottex
5482
@tex
5483
\vskip \baselineskip
5484
%"lispnarrowing" is the extra indent used generally for smallexample
5485
\hskip\lispnarrowing\vbox{\offinterlineskip
5486
\hrule
5487
\halign
5488
{\vrule#&\strut\hfil\ #\ \hfil&\vrule#&\strut\hfil\ #\ \hfil&\vrule#&\strut\hfil\ {\tt #}\ \hfil&\vrule#\cr
5489
height2pt&\omit&&\omit&&\omit&\cr
5490
&Precedence&&  Associativity  &&{\rm Operators}&\cr
5491
height2pt&\omit&&\omit&&\omit&\cr
5492
\noalign{\hrule}
5493
height2pt&\omit&&\omit&&\omit&\cr
5494
&highest&&&&&\cr
5495
% '176 is tilde, '~' in tt font
5496
&1&&left&&\qquad-          \char'176\      !\qquad\dag&\cr
5497
&2&&left&&*          /        \%&\cr
5498
&3&&left&&+          -&\cr
5499
&4&&left&&>>         <<&\cr
5500
&5&&left&&==         !=       >      <      <=      >=&\cr
5501
&6&&left&&\&&\cr
5502
&7&&left&&|&\cr
5503
&8&&left&&{\&\&}&\cr
5504
&9&&left&&||&\cr
5505
&10&&right&&?        :&\cr
5506
&11&&right&&\qquad\&=      +=       -=     *=     /=\qquad\ddag&\cr
5507
&lowest&&&&&\cr
5508
height2pt&\omit&&\omit&&\omit&\cr}
5509
\hrule}
5510
@end tex
5511
@iftex
5512
{
5513
@obeylines@parskip=0pt@parindent=0pt
5514
@dag@quad Prefix operators.
5515
@ddag@quad @xref{Assignments}.
5516
}
5517
@end iftex
5518
@c END TEXI2ROFF-KILL
5519
 
5520
@node Evaluation
5521
@subsection Evaluation
5522
@cindex lazy evaluation
5523
@cindex expression evaluation order
5524
The linker evaluates expressions lazily.  It only computes the value of
5525
an expression when absolutely necessary.
5526
 
5527
The linker needs some information, such as the value of the start
5528
address of the first section, and the origins and lengths of memory
5529
regions, in order to do any linking at all.  These values are computed
5530
as soon as possible when the linker reads in the linker script.
5531
 
5532
However, other values (such as symbol values) are not known or needed
5533
until after storage allocation.  Such values are evaluated later, when
5534
other information (such as the sizes of output sections) is available
5535
for use in the symbol assignment expression.
5536
 
5537
The sizes of sections cannot be known until after allocation, so
5538
assignments dependent upon these are not performed until after
5539
allocation.
5540
 
5541
Some expressions, such as those depending upon the location counter
5542
@samp{.}, must be evaluated during section allocation.
5543
 
5544
If the result of an expression is required, but the value is not
5545
available, then an error results.  For example, a script like the
5546
following
5547
@smallexample
5548
@group
5549
SECTIONS
5550
  @{
5551
    .text 9+this_isnt_constant :
5552
      @{ *(.text) @}
5553
  @}
5554
@end group
5555
@end smallexample
5556
@noindent
5557
will cause the error message @samp{non constant expression for initial
5558
address}.
5559
 
5560
@node Expression Section
5561
@subsection The Section of an Expression
5562
@cindex expression sections
5563
@cindex absolute expressions
5564
@cindex relative expressions
5565
@cindex absolute and relocatable symbols
5566
@cindex relocatable and absolute symbols
5567
@cindex symbols, relocatable and absolute
5568
Addresses and symbols may be section relative, or absolute.  A section
5569
relative symbol is relocatable.  If you request relocatable output
5570
using the @samp{-r} option, a further link operation may change the
5571
value of a section relative symbol.  On the other hand, an absolute
5572
symbol will retain the same value throughout any further link
5573
operations.
5574
 
5575
Some terms in linker expressions are addresses.  This is true of
5576
section relative symbols and for builtin functions that return an
5577
address, such as @code{ADDR}, @code{LOADADDR}, @code{ORIGIN} and
5578
@code{SEGMENT_START}.  Other terms are simply numbers, or are builtin
5579
functions that return a non-address value, such as @code{LENGTH}.
5580
One complication is that unless you set @code{LD_FEATURE ("SANE_EXPR")}
5581
(@pxref{Miscellaneous Commands}), numbers and absolute symbols are treated
5582
differently depending on their location, for compatibility with older
5583
versions of @code{ld}.  Expressions appearing outside an output
5584
section definition treat all numbers as absolute addresses.
5585
Expressions appearing inside an output section definition treat
5586
absolute symbols as numbers.  If @code{LD_FEATURE ("SANE_EXPR")} is
5587
given, then absolute symbols and numbers are simply treated as numbers
5588
everywhere.
5589
 
5590
In the following simple example,
5591
 
5592
@smallexample
5593
@group
5594
SECTIONS
5595
  @{
5596
    . = 0x100;
5597
    __executable_start = 0x100;
5598
    .data :
5599
    @{
5600
      . = 0x10;
5601
      __data_start = 0x10;
5602
      *(.data)
5603
    @}
5604
    @dots{}
5605
  @}
5606
@end group
5607
@end smallexample
5608
 
5609
both @code{.} and @code{__executable_start} are set to the absolute
5610
address 0x100 in the first two assignments, then both @code{.} and
5611
@code{__data_start} are set to 0x10 relative to the @code{.data}
5612
section in the second two assignments.
5613
 
5614
For expressions involving numbers, relative addresses and absolute
5615
addresses, ld follows these rules to evaluate terms:
5616
 
5617
@itemize @bullet
5618
@item
5619
Unary operations on a relative address, and binary operations on two
5620
relative addresses in the same section or between one relative address
5621
and a number, apply the operator to the offset part of the address(es).
5622
@item
5623
Unary operations on an absolute address, and binary operations on one
5624
or more absolute addresses or on two relative addresses not in the
5625
same section, first convert any non-absolute term to an absolute
5626
address before applying the operator.
5627
@end itemize
5628
 
5629
The result section of each sub-expression is as follows:
5630
 
5631
@itemize @bullet
5632
@item
5633
An operation involving only numbers results in a number.
5634
@item
5635
The result of comparisons, @samp{&&} and @samp{||} is also a number.
5636
@item
5637
The result of other binary arithmetic and logical operations on two
5638
relative addresses in the same section or two absolute addresess
5639
(after above conversions) is also a number.
5640
@item
5641
The result of other operations on relative addresses or one
5642
relative address and a number, is a relative address in the same
5643
section as the relative operand(s).
5644
@item
5645
The result of other operations on absolute addresses (after above
5646
conversions) is an absolute address.
5647
@end itemize
5648
 
5649
You can use the builtin function @code{ABSOLUTE} to force an expression
5650
to be absolute when it would otherwise be relative.  For example, to
5651
create an absolute symbol set to the address of the end of the output
5652
section @samp{.data}:
5653
@smallexample
5654
SECTIONS
5655
  @{
5656
    .data : @{ *(.data) _edata = ABSOLUTE(.); @}
5657
  @}
5658
@end smallexample
5659
@noindent
5660
If @samp{ABSOLUTE} were not used, @samp{_edata} would be relative to the
5661
@samp{.data} section.
5662
 
5663
Using @code{LOADADDR} also forces an expression absolute, since this
5664
particular builtin function returns an absolute address.
5665
 
5666
@node Builtin Functions
5667
@subsection Builtin Functions
5668
@cindex functions in expressions
5669
The linker script language includes a number of builtin functions for
5670
use in linker script expressions.
5671
 
5672
@table @code
5673
@item ABSOLUTE(@var{exp})
5674
@kindex ABSOLUTE(@var{exp})
5675
@cindex expression, absolute
5676
Return the absolute (non-relocatable, as opposed to non-negative) value
5677
of the expression @var{exp}.  Primarily useful to assign an absolute
5678
value to a symbol within a section definition, where symbol values are
5679
normally section relative.  @xref{Expression Section}.
5680
 
5681
@item ADDR(@var{section})
5682
@kindex ADDR(@var{section})
5683
@cindex section address in expression
5684
Return the address (VMA) of the named @var{section}.  Your
5685
script must previously have defined the location of that section.  In
5686
the following example, @code{start_of_output_1}, @code{symbol_1} and
5687
@code{symbol_2} are assigned equivalent values, except that
5688
@code{symbol_1} will be relative to the @code{.output1} section while
5689
the other two will be absolute:
5690
@smallexample
5691
@group
5692
SECTIONS @{ @dots{}
5693
  .output1 :
5694
    @{
5695
    start_of_output_1 = ABSOLUTE(.);
5696
    @dots{}
5697
    @}
5698
  .output :
5699
    @{
5700
    symbol_1 = ADDR(.output1);
5701
    symbol_2 = start_of_output_1;
5702
    @}
5703
@dots{} @}
5704
@end group
5705
@end smallexample
5706
 
5707
@item ALIGN(@var{align})
5708
@itemx ALIGN(@var{exp},@var{align})
5709
@kindex ALIGN(@var{align})
5710
@kindex ALIGN(@var{exp},@var{align})
5711
@cindex round up location counter
5712
@cindex align location counter
5713
@cindex round up expression
5714
@cindex align expression
5715
Return the location counter (@code{.}) or arbitrary expression aligned
5716
to the next @var{align} boundary.  The single operand @code{ALIGN}
5717
doesn't change the value of the location counter---it just does
5718
arithmetic on it.  The two operand @code{ALIGN} allows an arbitrary
5719
expression to be aligned upwards (@code{ALIGN(@var{align})} is
5720
equivalent to @code{ALIGN(., @var{align})}).
5721
 
5722
Here is an example which aligns the output @code{.data} section to the
5723
next @code{0x2000} byte boundary after the preceding section and sets a
5724
variable within the section to the next @code{0x8000} boundary after the
5725
input sections:
5726
@smallexample
5727
@group
5728
SECTIONS @{ @dots{}
5729
  .data ALIGN(0x2000): @{
5730
    *(.data)
5731
    variable = ALIGN(0x8000);
5732
  @}
5733
@dots{} @}
5734
@end group
5735
@end smallexample
5736
@noindent
5737
The first use of @code{ALIGN} in this example specifies the location of
5738
a section because it is used as the optional @var{address} attribute of
5739
a section definition (@pxref{Output Section Address}).  The second use
5740
of @code{ALIGN} is used to defines the value of a symbol.
5741
 
5742
The builtin function @code{NEXT} is closely related to @code{ALIGN}.
5743
 
5744
@item ALIGNOF(@var{section})
5745
@kindex ALIGNOF(@var{section})
5746
@cindex section alignment
5747
Return the alignment in bytes of the named @var{section}, if that section has
5748
been allocated.  If the section has not been allocated when this is
5749
evaluated, the linker will report an error. In the following example,
5750
the alignment of the @code{.output} section is stored as the first
5751
value in that section.
5752
@smallexample
5753
@group
5754
SECTIONS@{ @dots{}
5755
  .output @{
5756
    LONG (ALIGNOF (.output))
5757
    @dots{}
5758
    @}
5759
@dots{} @}
5760
@end group
5761
@end smallexample
5762
 
5763
@item BLOCK(@var{exp})
5764
@kindex BLOCK(@var{exp})
5765
This is a synonym for @code{ALIGN}, for compatibility with older linker
5766
scripts.  It is most often seen when setting the address of an output
5767
section.
5768
 
5769
@item DATA_SEGMENT_ALIGN(@var{maxpagesize}, @var{commonpagesize})
5770
@kindex DATA_SEGMENT_ALIGN(@var{maxpagesize}, @var{commonpagesize})
5771
This is equivalent to either
5772
@smallexample
5773
(ALIGN(@var{maxpagesize}) + (. & (@var{maxpagesize} - 1)))
5774
@end smallexample
5775
or
5776
@smallexample
5777
(ALIGN(@var{maxpagesize}) + (. & (@var{maxpagesize} - @var{commonpagesize})))
5778
@end smallexample
5779
@noindent
5780
depending on whether the latter uses fewer @var{commonpagesize} sized pages
5781
for the data segment (area between the result of this expression and
5782
@code{DATA_SEGMENT_END}) than the former or not.
5783
If the latter form is used, it means @var{commonpagesize} bytes of runtime
5784
memory will be saved at the expense of up to @var{commonpagesize} wasted
5785
bytes in the on-disk file.
5786
 
5787
This expression can only be used directly in @code{SECTIONS} commands, not in
5788
any output section descriptions and only once in the linker script.
5789
@var{commonpagesize} should be less or equal to @var{maxpagesize} and should
5790
be the system page size the object wants to be optimized for (while still
5791
working on system page sizes up to @var{maxpagesize}).
5792
 
5793
@noindent
5794
Example:
5795
@smallexample
5796
  . = DATA_SEGMENT_ALIGN(0x10000, 0x2000);
5797
@end smallexample
5798
 
5799
@item DATA_SEGMENT_END(@var{exp})
5800
@kindex DATA_SEGMENT_END(@var{exp})
5801
This defines the end of data segment for @code{DATA_SEGMENT_ALIGN}
5802
evaluation purposes.
5803
 
5804
@smallexample
5805
  . = DATA_SEGMENT_END(.);
5806
@end smallexample
5807
 
5808
@item DATA_SEGMENT_RELRO_END(@var{offset}, @var{exp})
5809
@kindex DATA_SEGMENT_RELRO_END(@var{offset}, @var{exp})
5810
This defines the end of the @code{PT_GNU_RELRO} segment when
5811
@samp{-z relro} option is used.  Second argument is returned.
5812
When @samp{-z relro} option is not present, @code{DATA_SEGMENT_RELRO_END}
5813
does nothing, otherwise @code{DATA_SEGMENT_ALIGN} is padded so that
5814
@var{exp} + @var{offset} is aligned to the most commonly used page
5815
boundary for particular target.  If present in the linker script,
5816
it must always come in between @code{DATA_SEGMENT_ALIGN} and
5817
@code{DATA_SEGMENT_END}.
5818
 
5819
@smallexample
5820
  . = DATA_SEGMENT_RELRO_END(24, .);
5821
@end smallexample
5822
 
5823
@item DEFINED(@var{symbol})
5824
@kindex DEFINED(@var{symbol})
5825
@cindex symbol defaults
5826
Return 1 if @var{symbol} is in the linker global symbol table and is
5827
defined before the statement using DEFINED in the script, otherwise
5828
return 0.  You can use this function to provide
5829
default values for symbols.  For example, the following script fragment
5830
shows how to set a global symbol @samp{begin} to the first location in
5831
the @samp{.text} section---but if a symbol called @samp{begin} already
5832
existed, its value is preserved:
5833
 
5834
@smallexample
5835
@group
5836
SECTIONS @{ @dots{}
5837
  .text : @{
5838
    begin = DEFINED(begin) ? begin : . ;
5839
    @dots{}
5840
  @}
5841
  @dots{}
5842
@}
5843
@end group
5844
@end smallexample
5845
 
5846
@item LENGTH(@var{memory})
5847
@kindex LENGTH(@var{memory})
5848
Return the length of the memory region named @var{memory}.
5849
 
5850
@item LOADADDR(@var{section})
5851
@kindex LOADADDR(@var{section})
5852
@cindex section load address in expression
5853
Return the absolute LMA of the named @var{section}.  (@pxref{Output
5854
Section LMA}).
5855
 
5856
@kindex MAX
5857
@item MAX(@var{exp1}, @var{exp2})
5858
Returns the maximum of @var{exp1} and @var{exp2}.
5859
 
5860
@kindex MIN
5861
@item MIN(@var{exp1}, @var{exp2})
5862
Returns the minimum of @var{exp1} and @var{exp2}.
5863
 
5864
@item NEXT(@var{exp})
5865
@kindex NEXT(@var{exp})
5866
@cindex unallocated address, next
5867
Return the next unallocated address that is a multiple of @var{exp}.
5868
This function is closely related to @code{ALIGN(@var{exp})}; unless you
5869
use the @code{MEMORY} command to define discontinuous memory for the
5870
output file, the two functions are equivalent.
5871
 
5872
@item ORIGIN(@var{memory})
5873
@kindex ORIGIN(@var{memory})
5874
Return the origin of the memory region named @var{memory}.
5875
 
5876
@item SEGMENT_START(@var{segment}, @var{default})
5877
@kindex SEGMENT_START(@var{segment}, @var{default})
5878
Return the base address of the named @var{segment}.  If an explicit
5879
value has been given for this segment (with a command-line @samp{-T}
5880
option) that value will be returned; otherwise the value will be
5881
@var{default}.  At present, the @samp{-T} command-line option can only
5882
be used to set the base address for the ``text'', ``data'', and
5883
``bss'' sections, but you can use @code{SEGMENT_START} with any segment
5884
name.
5885
 
5886
@item SIZEOF(@var{section})
5887
@kindex SIZEOF(@var{section})
5888
@cindex section size
5889
Return the size in bytes of the named @var{section}, if that section has
5890
been allocated.  If the section has not been allocated when this is
5891
evaluated, the linker will report an error.  In the following example,
5892
@code{symbol_1} and @code{symbol_2} are assigned identical values:
5893
@smallexample
5894
@group
5895
SECTIONS@{ @dots{}
5896
  .output @{
5897
    .start = . ;
5898
    @dots{}
5899
    .end = . ;
5900
    @}
5901
  symbol_1 = .end - .start ;
5902
  symbol_2 = SIZEOF(.output);
5903
@dots{} @}
5904
@end group
5905
@end smallexample
5906
 
5907
@item SIZEOF_HEADERS
5908
@itemx sizeof_headers
5909
@kindex SIZEOF_HEADERS
5910
@cindex header size
5911
Return the size in bytes of the output file's headers.  This is
5912
information which appears at the start of the output file.  You can use
5913
this number when setting the start address of the first section, if you
5914
choose, to facilitate paging.
5915
 
5916
@cindex not enough room for program headers
5917
@cindex program headers, not enough room
5918
When producing an ELF output file, if the linker script uses the
5919
@code{SIZEOF_HEADERS} builtin function, the linker must compute the
5920
number of program headers before it has determined all the section
5921
addresses and sizes.  If the linker later discovers that it needs
5922
additional program headers, it will report an error @samp{not enough
5923
room for program headers}.  To avoid this error, you must avoid using
5924
the @code{SIZEOF_HEADERS} function, or you must rework your linker
5925
script to avoid forcing the linker to use additional program headers, or
5926
you must define the program headers yourself using the @code{PHDRS}
5927
command (@pxref{PHDRS}).
5928
@end table
5929
 
5930
@node Implicit Linker Scripts
5931
@section Implicit Linker Scripts
5932
@cindex implicit linker scripts
5933
If you specify a linker input file which the linker can not recognize as
5934
an object file or an archive file, it will try to read the file as a
5935
linker script.  If the file can not be parsed as a linker script, the
5936
linker will report an error.
5937
 
5938
An implicit linker script will not replace the default linker script.
5939
 
5940
Typically an implicit linker script would contain only symbol
5941
assignments, or the @code{INPUT}, @code{GROUP}, or @code{VERSION}
5942
commands.
5943
 
5944
Any input files read because of an implicit linker script will be read
5945
at the position in the command line where the implicit linker script was
5946
read.  This can affect archive searching.
5947
 
5948
@ifset GENERIC
5949
@node Machine Dependent
5950
@chapter Machine Dependent Features
5951
 
5952
@cindex machine dependencies
5953
@command{ld} has additional features on some platforms; the following
5954
sections describe them.  Machines where @command{ld} has no additional
5955
functionality are not listed.
5956
 
5957
@menu
5958
@ifset H8300
5959
* H8/300::                      @command{ld} and the H8/300
5960
@end ifset
5961
@ifset I960
5962
* i960::                        @command{ld} and the Intel 960 family
5963
@end ifset
5964
@ifset ARM
5965
* ARM::                         @command{ld} and the ARM family
5966
@end ifset
5967
@ifset HPPA
5968
* HPPA ELF32::                  @command{ld} and HPPA 32-bit ELF
5969
@end ifset
5970
@ifset M68K
5971
* M68K::                        @command{ld} and the Motorola 68K family
5972
@end ifset
5973
@ifset MMIX
5974
* MMIX::                        @command{ld} and MMIX
5975
@end ifset
5976
@ifset MSP430
5977
* MSP430::                      @command{ld} and MSP430
5978
@end ifset
5979
@ifset M68HC11
5980
* M68HC11/68HC12::              @code{ld} and the Motorola 68HC11 and 68HC12 families
5981
@end ifset
5982
@ifset POWERPC
5983
* PowerPC ELF32::               @command{ld} and PowerPC 32-bit ELF Support
5984
@end ifset
5985
@ifset POWERPC64
5986
* PowerPC64 ELF64::             @command{ld} and PowerPC64 64-bit ELF Support
5987
@end ifset
5988
@ifset SPU
5989
* SPU ELF::                     @command{ld} and SPU ELF Support
5990
@end ifset
5991
@ifset TICOFF
5992
* TI COFF::                     @command{ld} and TI COFF
5993
@end ifset
5994
@ifset WIN32
5995
* WIN32::                       @command{ld} and WIN32 (cygwin/mingw)
5996
@end ifset
5997
@ifset XTENSA
5998
* Xtensa::                      @command{ld} and Xtensa Processors
5999
@end ifset
6000
@end menu
6001
@end ifset
6002
 
6003
@ifset H8300
6004
@ifclear GENERIC
6005
@raisesections
6006
@end ifclear
6007
 
6008
@node H8/300
6009
@section @command{ld} and the H8/300
6010
 
6011
@cindex H8/300 support
6012
For the H8/300, @command{ld} can perform these global optimizations when
6013
you specify the @samp{--relax} command-line option.
6014
 
6015
@table @emph
6016
@cindex relaxing on H8/300
6017
@item relaxing address modes
6018
@command{ld} finds all @code{jsr} and @code{jmp} instructions whose
6019
targets are within eight bits, and turns them into eight-bit
6020
program-counter relative @code{bsr} and @code{bra} instructions,
6021
respectively.
6022
 
6023
@cindex synthesizing on H8/300
6024
@item synthesizing instructions
6025
@c FIXME: specifically mov.b, or any mov instructions really?
6026
@command{ld} finds all @code{mov.b} instructions which use the
6027
sixteen-bit absolute address form, but refer to the top
6028
page of memory, and changes them to use the eight-bit address form.
6029
(That is: the linker turns @samp{mov.b @code{@@}@var{aa}:16} into
6030
@samp{mov.b @code{@@}@var{aa}:8} whenever the address @var{aa} is in the
6031
top page of memory).
6032
 
6033
@item bit manipulation instructions
6034
@command{ld} finds all bit manipulation instructions like @code{band, bclr,
6035
biand, bild, bior, bist, bixor, bld, bnot, bor, bset, bst, btst, bxor}
6036
which use 32 bit and 16 bit absolute address form, but refer to the top
6037
page of memory, and changes them to use the 8 bit address form.
6038
(That is: the linker turns @samp{bset #xx:3,@code{@@}@var{aa}:32} into
6039
@samp{bset #xx:3,@code{@@}@var{aa}:8} whenever the address @var{aa} is in
6040
the top page of memory).
6041
 
6042
@item system control instructions
6043
@command{ld} finds all @code{ldc.w, stc.w} instructions which use the
6044
32 bit absolute address form, but refer to the top page of memory, and
6045
changes them to use 16 bit address form.
6046
(That is: the linker turns @samp{ldc.w @code{@@}@var{aa}:32,ccr} into
6047
@samp{ldc.w @code{@@}@var{aa}:16,ccr} whenever the address @var{aa} is in
6048
the top page of memory).
6049
@end table
6050
 
6051
@ifclear GENERIC
6052
@lowersections
6053
@end ifclear
6054
@end ifset
6055
 
6056
@ifclear GENERIC
6057
@ifset Renesas
6058
@c This stuff is pointless to say unless you're especially concerned
6059
@c with Renesas chips; don't enable it for generic case, please.
6060
@node Renesas
6061
@chapter @command{ld} and Other Renesas Chips
6062
 
6063
@command{ld} also supports the Renesas (formerly Hitachi) H8/300H,
6064
H8/500, and SH chips.  No special features, commands, or command-line
6065
options are required for these chips.
6066
@end ifset
6067
@end ifclear
6068
 
6069
@ifset I960
6070
@ifclear GENERIC
6071
@raisesections
6072
@end ifclear
6073
 
6074
@node i960
6075
@section @command{ld} and the Intel 960 Family
6076
 
6077
@cindex i960 support
6078
 
6079
You can use the @samp{-A@var{architecture}} command line option to
6080
specify one of the two-letter names identifying members of the 960
6081
family; the option specifies the desired output target, and warns of any
6082
incompatible instructions in the input files.  It also modifies the
6083
linker's search strategy for archive libraries, to support the use of
6084
libraries specific to each particular architecture, by including in the
6085
search loop names suffixed with the string identifying the architecture.
6086
 
6087
For example, if your @command{ld} command line included @w{@samp{-ACA}} as
6088
well as @w{@samp{-ltry}}, the linker would look (in its built-in search
6089
paths, and in any paths you specify with @samp{-L}) for a library with
6090
the names
6091
 
6092
@smallexample
6093
@group
6094
try
6095
libtry.a
6096
tryca
6097
libtryca.a
6098
@end group
6099
@end smallexample
6100
 
6101
@noindent
6102
The first two possibilities would be considered in any event; the last
6103
two are due to the use of @w{@samp{-ACA}}.
6104
 
6105
You can meaningfully use @samp{-A} more than once on a command line, since
6106
the 960 architecture family allows combination of target architectures; each
6107
use will add another pair of name variants to search for when @w{@samp{-l}}
6108
specifies a library.
6109
 
6110
@cindex @option{--relax} on i960
6111
@cindex relaxing on i960
6112
@command{ld} supports the @samp{--relax} option for the i960 family.  If
6113
you specify @samp{--relax}, @command{ld} finds all @code{balx} and
6114
@code{calx} instructions whose targets are within 24 bits, and turns
6115
them into 24-bit program-counter relative @code{bal} and @code{cal}
6116
instructions, respectively.  @command{ld} also turns @code{cal}
6117
instructions into @code{bal} instructions when it determines that the
6118
target subroutine is a leaf routine (that is, the target subroutine does
6119
not itself call any subroutines).
6120
 
6121
@cindex Cortex-A8 erratum workaround
6122
@kindex --fix-cortex-a8
6123
@kindex --no-fix-cortex-a8
6124
The @samp{--fix-cortex-a8} switch enables a link-time workaround for an erratum in certain Cortex-A8 processors.  The workaround is enabled by default if you are targeting the ARM v7-A architecture profile.  It can be enabled otherwise by specifying @samp{--fix-cortex-a8}, or disabled unconditionally by specifying @samp{--no-fix-cortex-a8}.
6125
 
6126
The erratum only affects Thumb-2 code.  Please contact ARM for further details.
6127
 
6128
@kindex --merge-exidx-entries
6129
@kindex --no-merge-exidx-entries
6130
The @samp{--no-merge-exidx-entries} switch disables the merging of adjacent exidx entries in debuginfo.
6131
 
6132
@ifclear GENERIC
6133
@lowersections
6134
@end ifclear
6135
@end ifset
6136
 
6137
@ifset ARM
6138
@ifclear GENERIC
6139
@raisesections
6140
@end ifclear
6141
 
6142
@ifset M68HC11
6143
@ifclear GENERIC
6144
@raisesections
6145
@end ifclear
6146
 
6147
@node M68HC11/68HC12
6148
@section @command{ld} and the Motorola 68HC11 and 68HC12 families
6149
 
6150
@cindex M68HC11 and 68HC12 support
6151
 
6152
@subsection Linker Relaxation
6153
 
6154
For the Motorola 68HC11, @command{ld} can perform these global
6155
optimizations when you specify the @samp{--relax} command-line option.
6156
 
6157
@table @emph
6158
@cindex relaxing on M68HC11
6159
@item relaxing address modes
6160
@command{ld} finds all @code{jsr} and @code{jmp} instructions whose
6161
targets are within eight bits, and turns them into eight-bit
6162
program-counter relative @code{bsr} and @code{bra} instructions,
6163
respectively.
6164
 
6165
@command{ld} also looks at all 16-bit extended addressing modes and
6166
transforms them in a direct addressing mode when the address is in
6167
page 0 (between 0 and 0x0ff).
6168
 
6169
@item relaxing gcc instruction group
6170
When @command{gcc} is called with @option{-mrelax}, it can emit group
6171
of instructions that the linker can optimize to use a 68HC11 direct
6172
addressing mode. These instructions consists of @code{bclr} or
6173
@code{bset} instructions.
6174
 
6175
@end table
6176
 
6177
@subsection Trampoline Generation
6178
 
6179
@cindex trampoline generation on M68HC11
6180
@cindex trampoline generation on M68HC12
6181
For 68HC11 and 68HC12, @command{ld} can generate trampoline code to
6182
call a far function using a normal @code{jsr} instruction. The linker
6183
will also change the relocation to some far function to use the
6184
trampoline address instead of the function address. This is typically the
6185
case when a pointer to a function is taken. The pointer will in fact
6186
point to the function trampoline.
6187
 
6188
@ifclear GENERIC
6189
@lowersections
6190
@end ifclear
6191
@end ifset
6192
 
6193
@node ARM
6194
@section @command{ld} and the ARM family
6195
 
6196
@cindex ARM interworking support
6197
@kindex --support-old-code
6198
For the ARM, @command{ld} will generate code stubs to allow functions calls
6199
between ARM and Thumb code.  These stubs only work with code that has
6200
been compiled and assembled with the @samp{-mthumb-interwork} command
6201
line option.  If it is necessary to link with old ARM object files or
6202
libraries, which have not been compiled with the -mthumb-interwork
6203
option then the @samp{--support-old-code} command line switch should be
6204
given to the linker.  This will make it generate larger stub functions
6205
which will work with non-interworking aware ARM code.  Note, however,
6206
the linker does not support generating stubs for function calls to
6207
non-interworking aware Thumb code.
6208
 
6209
@cindex thumb entry point
6210
@cindex entry point, thumb
6211
@kindex --thumb-entry=@var{entry}
6212
The @samp{--thumb-entry} switch is a duplicate of the generic
6213
@samp{--entry} switch, in that it sets the program's starting address.
6214
But it also sets the bottom bit of the address, so that it can be
6215
branched to using a BX instruction, and the program will start
6216
executing in Thumb mode straight away.
6217
 
6218
@cindex PE import table prefixing
6219
@kindex --use-nul-prefixed-import-tables
6220
The @samp{--use-nul-prefixed-import-tables} switch is specifying, that
6221
the import tables idata4 and idata5 have to be generated with a zero
6222
elememt prefix for import libraries. This is the old style to generate
6223
import tables. By default this option is turned off.
6224
 
6225
@cindex BE8
6226
@kindex --be8
6227
The @samp{--be8} switch instructs @command{ld} to generate BE8 format
6228
executables.  This option is only valid when linking big-endian objects.
6229
The resulting image will contain big-endian data and little-endian code.
6230
 
6231
@cindex TARGET1
6232
@kindex --target1-rel
6233
@kindex --target1-abs
6234
The @samp{R_ARM_TARGET1} relocation is typically used for entries in the
6235
@samp{.init_array} section.  It is interpreted as either @samp{R_ARM_REL32}
6236
or @samp{R_ARM_ABS32}, depending on the target.  The @samp{--target1-rel}
6237
and @samp{--target1-abs} switches override the default.
6238
 
6239
@cindex TARGET2
6240
@kindex --target2=@var{type}
6241
The @samp{--target2=type} switch overrides the default definition of the
6242
@samp{R_ARM_TARGET2} relocation.  Valid values for @samp{type}, their
6243
meanings, and target defaults are as follows:
6244
@table @samp
6245
@item rel
6246
@samp{R_ARM_REL32} (arm*-*-elf, arm*-*-eabi)
6247
@item abs
6248
@samp{R_ARM_ABS32} (arm*-*-symbianelf)
6249
@item got-rel
6250
@samp{R_ARM_GOT_PREL} (arm*-*-linux, arm*-*-*bsd)
6251
@end table
6252
 
6253
@cindex FIX_V4BX
6254
@kindex --fix-v4bx
6255
The @samp{R_ARM_V4BX} relocation (defined by the ARM AAELF
6256
specification) enables objects compiled for the ARMv4 architecture to be
6257
interworking-safe when linked with other objects compiled for ARMv4t, but
6258
also allows pure ARMv4 binaries to be built from the same ARMv4 objects.
6259
 
6260
In the latter case, the switch @option{--fix-v4bx} must be passed to the
6261
linker, which causes v4t @code{BX rM} instructions to be rewritten as
6262
@code{MOV PC,rM}, since v4 processors do not have a @code{BX} instruction.
6263
 
6264
In the former case, the switch should not be used, and @samp{R_ARM_V4BX}
6265
relocations are ignored.
6266
 
6267
@cindex FIX_V4BX_INTERWORKING
6268
@kindex --fix-v4bx-interworking
6269
Replace @code{BX rM} instructions identified by @samp{R_ARM_V4BX}
6270
relocations with a branch to the following veneer:
6271
 
6272
@smallexample
6273
TST rM, #1
6274
MOVEQ PC, rM
6275
BX Rn
6276
@end smallexample
6277
 
6278
This allows generation of libraries/applications that work on ARMv4 cores
6279
and are still interworking safe.  Note that the above veneer clobbers the
6280
condition flags, so may cause incorrect progrm behavior in rare cases.
6281
 
6282
@cindex USE_BLX
6283
@kindex --use-blx
6284
The @samp{--use-blx} switch enables the linker to use ARM/Thumb
6285
BLX instructions (available on ARMv5t and above) in various
6286
situations. Currently it is used to perform calls via the PLT from Thumb
6287
code using BLX rather than using BX and a mode-switching stub before
6288
each PLT entry. This should lead to such calls executing slightly faster.
6289
 
6290
This option is enabled implicitly for SymbianOS, so there is no need to
6291
specify it if you are using that target.
6292
 
6293
@cindex VFP11_DENORM_FIX
6294
@kindex --vfp11-denorm-fix
6295
The @samp{--vfp11-denorm-fix} switch enables a link-time workaround for a
6296
bug in certain VFP11 coprocessor hardware, which sometimes allows
6297
instructions with denorm operands (which must be handled by support code)
6298
to have those operands overwritten by subsequent instructions before
6299
the support code can read the intended values.
6300
 
6301
The bug may be avoided in scalar mode if you allow at least one
6302
intervening instruction between a VFP11 instruction which uses a register
6303
and another instruction which writes to the same register, or at least two
6304
intervening instructions if vector mode is in use. The bug only affects
6305
full-compliance floating-point mode: you do not need this workaround if
6306
you are using "runfast" mode. Please contact ARM for further details.
6307
 
6308
If you know you are using buggy VFP11 hardware, you can
6309
enable this workaround by specifying the linker option
6310
@samp{--vfp-denorm-fix=scalar} if you are using the VFP11 scalar
6311
mode only, or @samp{--vfp-denorm-fix=vector} if you are using
6312
vector mode (the latter also works for scalar code). The default is
6313
@samp{--vfp-denorm-fix=none}.
6314
 
6315
If the workaround is enabled, instructions are scanned for
6316
potentially-troublesome sequences, and a veneer is created for each
6317
such sequence which may trigger the erratum. The veneer consists of the
6318
first instruction of the sequence and a branch back to the subsequent
6319
instruction. The original instruction is then replaced with a branch to
6320
the veneer. The extra cycles required to call and return from the veneer
6321
are sufficient to avoid the erratum in both the scalar and vector cases.
6322
 
6323 157 khays
@cindex ARM1176 erratum workaround
6324
@kindex --fix-arm1176
6325
@kindex --no-fix-arm1176
6326
The @samp{--fix-arm1176} switch enables a link-time workaround for an erratum
6327
in certain ARM1176 processors.  The workaround is enabled by default if you
6328
are targetting ARM v6 (excluding ARM v6T2) or earlier.  It can be disabled
6329
unconditionally by specifying @samp{--no-fix-arm1176}.
6330
 
6331
Further information is available in the ``ARM1176JZ-S and ARM1176JZF-S
6332
Programmer Advice Notice'' available on the ARM documentaion website at:
6333
http://infocenter.arm.com/.
6334
 
6335 145 khays
@cindex NO_ENUM_SIZE_WARNING
6336
@kindex --no-enum-size-warning
6337
The @option{--no-enum-size-warning} switch prevents the linker from
6338
warning when linking object files that specify incompatible EABI
6339
enumeration size attributes.  For example, with this switch enabled,
6340
linking of an object file using 32-bit enumeration values with another
6341
using enumeration values fitted into the smallest possible space will
6342
not be diagnosed.
6343
 
6344
@cindex NO_WCHAR_SIZE_WARNING
6345
@kindex --no-wchar-size-warning
6346
The @option{--no-wchar-size-warning} switch prevents the linker from
6347
warning when linking object files that specify incompatible EABI
6348
@code{wchar_t} size attributes.  For example, with this switch enabled,
6349
linking of an object file using 32-bit @code{wchar_t} values with another
6350
using 16-bit @code{wchar_t} values will not be diagnosed.
6351
 
6352
@cindex PIC_VENEER
6353
@kindex --pic-veneer
6354
The @samp{--pic-veneer} switch makes the linker use PIC sequences for
6355
ARM/Thumb interworking veneers, even if the rest of the binary
6356
is not PIC.  This avoids problems on uClinux targets where
6357
@samp{--emit-relocs} is used to generate relocatable binaries.
6358
 
6359
@cindex STUB_GROUP_SIZE
6360
@kindex --stub-group-size=@var{N}
6361
The linker will automatically generate and insert small sequences of
6362
code into a linked ARM ELF executable whenever an attempt is made to
6363
perform a function call to a symbol that is too far away.  The
6364
placement of these sequences of instructions - called stubs - is
6365
controlled by the command line option @option{--stub-group-size=N}.
6366
The placement is important because a poor choice can create a need for
6367
duplicate stubs, increasing the code sizw.  The linker will try to
6368
group stubs together in order to reduce interruptions to the flow of
6369
code, but it needs guidance as to how big these groups should be and
6370
where they should be placed.
6371
 
6372
The value of @samp{N}, the parameter to the
6373
@option{--stub-group-size=} option controls where the stub groups are
6374
placed.  If it is negative then all stubs are placed after the first
6375
branch that needs them.  If it is positive then the stubs can be
6376
placed either before or after the branches that need them.  If the
6377
value of @samp{N} is 1 (either +1 or -1) then the linker will choose
6378
exactly where to place groups of stubs, using its built in heuristics.
6379
A value of @samp{N} greater than 1 (or smaller than -1) tells the
6380
linker that a single group of stubs can service at most @samp{N} bytes
6381
from the input sections.
6382
 
6383
The default, if @option{--stub-group-size=} is not specified, is
6384
@samp{N = +1}.
6385
 
6386
Farcalls stubs insertion is fully supported for the ARM-EABI target
6387
only, because it relies on object files properties not present
6388
otherwise.
6389
 
6390
@ifclear GENERIC
6391
@lowersections
6392
@end ifclear
6393
@end ifset
6394
 
6395
@ifset HPPA
6396
@ifclear GENERIC
6397
@raisesections
6398
@end ifclear
6399
 
6400
@node HPPA ELF32
6401
@section @command{ld} and HPPA 32-bit ELF Support
6402
@cindex HPPA multiple sub-space stubs
6403
@kindex --multi-subspace
6404
When generating a shared library, @command{ld} will by default generate
6405
import stubs suitable for use with a single sub-space application.
6406
The @samp{--multi-subspace} switch causes @command{ld} to generate export
6407
stubs, and different (larger) import stubs suitable for use with
6408
multiple sub-spaces.
6409
 
6410
@cindex HPPA stub grouping
6411
@kindex --stub-group-size=@var{N}
6412
Long branch stubs and import/export stubs are placed by @command{ld} in
6413
stub sections located between groups of input sections.
6414
@samp{--stub-group-size} specifies the maximum size of a group of input
6415
sections handled by one stub section.  Since branch offsets are signed,
6416
a stub section may serve two groups of input sections, one group before
6417
the stub section, and one group after it.  However, when using
6418
conditional branches that require stubs, it may be better (for branch
6419
prediction) that stub sections only serve one group of input sections.
6420
A negative value for @samp{N} chooses this scheme, ensuring that
6421
branches to stubs always use a negative offset.  Two special values of
6422
@samp{N} are recognized, @samp{1} and @samp{-1}.  These both instruct
6423
@command{ld} to automatically size input section groups for the branch types
6424
detected, with the same behaviour regarding stub placement as other
6425
positive or negative values of @samp{N} respectively.
6426
 
6427
Note that @samp{--stub-group-size} does not split input sections.  A
6428
single input section larger than the group size specified will of course
6429
create a larger group (of one section).  If input sections are too
6430
large, it may not be possible for a branch to reach its stub.
6431
 
6432
@ifclear GENERIC
6433
@lowersections
6434
@end ifclear
6435
@end ifset
6436
 
6437
@ifset M68K
6438
@ifclear GENERIC
6439
@raisesections
6440
@end ifclear
6441
 
6442
@node M68K
6443
@section @command{ld} and the Motorola 68K family
6444
 
6445
@cindex Motorola 68K GOT generation
6446
@kindex --got=@var{type}
6447
The @samp{--got=@var{type}} option lets you choose the GOT generation scheme.
6448
The choices are @samp{single}, @samp{negative}, @samp{multigot} and
6449
@samp{target}.  When @samp{target} is selected the linker chooses
6450
the default GOT generation scheme for the current target.
6451
@samp{single} tells the linker to generate a single GOT with
6452
entries only at non-negative offsets.
6453
@samp{negative} instructs the linker to generate a single GOT with
6454
entries at both negative and positive offsets.  Not all environments
6455
support such GOTs.
6456
@samp{multigot} allows the linker to generate several GOTs in the
6457
output file.  All GOT references from a single input object
6458
file access the same GOT, but references from different input object
6459
files might access different GOTs.  Not all environments support such GOTs.
6460
 
6461
@ifclear GENERIC
6462
@lowersections
6463
@end ifclear
6464
@end ifset
6465
 
6466
@ifset MMIX
6467
@ifclear GENERIC
6468
@raisesections
6469
@end ifclear
6470
 
6471
@node MMIX
6472
@section @code{ld} and MMIX
6473
For MMIX, there is a choice of generating @code{ELF} object files or
6474
@code{mmo} object files when linking.  The simulator @code{mmix}
6475
understands the @code{mmo} format.  The binutils @code{objcopy} utility
6476
can translate between the two formats.
6477
 
6478
There is one special section, the @samp{.MMIX.reg_contents} section.
6479
Contents in this section is assumed to correspond to that of global
6480
registers, and symbols referring to it are translated to special symbols,
6481
equal to registers.  In a final link, the start address of the
6482
@samp{.MMIX.reg_contents} section corresponds to the first allocated
6483
global register multiplied by 8.  Register @code{$255} is not included in
6484
this section; it is always set to the program entry, which is at the
6485
symbol @code{Main} for @code{mmo} files.
6486
 
6487
Global symbols with the prefix @code{__.MMIX.start.}, for example
6488
@code{__.MMIX.start..text} and @code{__.MMIX.start..data} are special.
6489
The default linker script uses these to set the default start address
6490
of a section.
6491
 
6492
Initial and trailing multiples of zero-valued 32-bit words in a section,
6493
are left out from an mmo file.
6494
 
6495
@ifclear GENERIC
6496
@lowersections
6497
@end ifclear
6498
@end ifset
6499
 
6500
@ifset MSP430
6501
@ifclear GENERIC
6502
@raisesections
6503
@end ifclear
6504
 
6505
@node  MSP430
6506
@section @code{ld} and MSP430
6507
For the MSP430 it is possible to select the MPU architecture.  The flag @samp{-m [mpu type]}
6508
will select an appropriate linker script for selected MPU type.  (To get a list of known MPUs
6509
just pass @samp{-m help} option to the linker).
6510
 
6511
@cindex MSP430 extra sections
6512
The linker will recognize some extra sections which are MSP430 specific:
6513
 
6514
@table @code
6515
@item @samp{.vectors}
6516
Defines a portion of ROM where interrupt vectors located.
6517
 
6518
@item @samp{.bootloader}
6519
Defines the bootloader portion of the ROM (if applicable).  Any code
6520
in this section will be uploaded to the MPU.
6521
 
6522
@item @samp{.infomem}
6523
Defines an information memory section (if applicable).  Any code in
6524
this section will be uploaded to the MPU.
6525
 
6526
@item @samp{.infomemnobits}
6527
This is the same as the @samp{.infomem} section except that any code
6528
in this section will not be uploaded to the MPU.
6529
 
6530
@item @samp{.noinit}
6531
Denotes a portion of RAM located above @samp{.bss} section.
6532
 
6533
The last two sections are used by gcc.
6534
@end table
6535
 
6536
@ifclear GENERIC
6537
@lowersections
6538
@end ifclear
6539
@end ifset
6540
 
6541
@ifset POWERPC
6542
@ifclear GENERIC
6543
@raisesections
6544
@end ifclear
6545
 
6546
@node PowerPC ELF32
6547
@section @command{ld} and PowerPC 32-bit ELF Support
6548
@cindex PowerPC long branches
6549
@kindex --relax on PowerPC
6550
Branches on PowerPC processors are limited to a signed 26-bit
6551
displacement, which may result in @command{ld} giving
6552
@samp{relocation truncated to fit} errors with very large programs.
6553
@samp{--relax} enables the generation of trampolines that can access
6554
the entire 32-bit address space.  These trampolines are inserted at
6555
section boundaries, so may not themselves be reachable if an input
6556
section exceeds 33M in size.  You may combine @samp{-r} and
6557
@samp{--relax} to add trampolines in a partial link.  In that case
6558
both branches to undefined symbols and inter-section branches are also
6559
considered potentially out of range, and trampolines inserted.
6560
 
6561
@cindex PowerPC ELF32 options
6562
@table @option
6563
@cindex PowerPC PLT
6564
@kindex --bss-plt
6565
@item --bss-plt
6566
Current PowerPC GCC accepts a @samp{-msecure-plt} option that
6567
generates code capable of using a newer PLT and GOT layout that has
6568
the security advantage of no executable section ever needing to be
6569
writable and no writable section ever being executable.  PowerPC
6570
@command{ld} will generate this layout, including stubs to access the
6571
PLT, if all input files (including startup and static libraries) were
6572
compiled with @samp{-msecure-plt}.  @samp{--bss-plt} forces the old
6573
BSS PLT (and GOT layout) which can give slightly better performance.
6574
 
6575
@kindex --secure-plt
6576
@item --secure-plt
6577
@command{ld} will use the new PLT and GOT layout if it is linking new
6578
@samp{-fpic} or @samp{-fPIC} code, but does not do so automatically
6579
when linking non-PIC code.  This option requests the new PLT and GOT
6580
layout.  A warning will be given if some object file requires the old
6581
style BSS PLT.
6582
 
6583
@cindex PowerPC GOT
6584
@kindex --sdata-got
6585
@item --sdata-got
6586
The new secure PLT and GOT are placed differently relative to other
6587
sections compared to older BSS PLT and GOT placement.  The location of
6588
@code{.plt} must change because the new secure PLT is an initialized
6589
section while the old PLT is uninitialized.  The reason for the
6590
@code{.got} change is more subtle:  The new placement allows
6591
@code{.got} to be read-only in applications linked with
6592
@samp{-z relro -z now}.  However, this placement means that
6593
@code{.sdata} cannot always be used in shared libraries, because the
6594
PowerPC ABI accesses @code{.sdata} in shared libraries from the GOT
6595
pointer.  @samp{--sdata-got} forces the old GOT placement.  PowerPC
6596
GCC doesn't use @code{.sdata} in shared libraries, so this option is
6597
really only useful for other compilers that may do so.
6598
 
6599
@cindex PowerPC stub symbols
6600
@kindex --emit-stub-syms
6601
@item --emit-stub-syms
6602
This option causes @command{ld} to label linker stubs with a local
6603
symbol that encodes the stub type and destination.
6604
 
6605
@cindex PowerPC TLS optimization
6606
@kindex --no-tls-optimize
6607
@item --no-tls-optimize
6608
PowerPC @command{ld} normally performs some optimization of code
6609
sequences used to access Thread-Local Storage.  Use this option to
6610
disable the optimization.
6611
@end table
6612
 
6613
@ifclear GENERIC
6614
@lowersections
6615
@end ifclear
6616
@end ifset
6617
 
6618
@ifset POWERPC64
6619
@ifclear GENERIC
6620
@raisesections
6621
@end ifclear
6622
 
6623
@node PowerPC64 ELF64
6624
@section @command{ld} and PowerPC64 64-bit ELF Support
6625
 
6626
@cindex PowerPC64 ELF64 options
6627
@table @option
6628
@cindex PowerPC64 stub grouping
6629
@kindex --stub-group-size
6630
@item --stub-group-size
6631
Long branch stubs, PLT call stubs  and TOC adjusting stubs are placed
6632
by @command{ld} in stub sections located between groups of input sections.
6633
@samp{--stub-group-size} specifies the maximum size of a group of input
6634
sections handled by one stub section.  Since branch offsets are signed,
6635
a stub section may serve two groups of input sections, one group before
6636
the stub section, and one group after it.  However, when using
6637
conditional branches that require stubs, it may be better (for branch
6638
prediction) that stub sections only serve one group of input sections.
6639
A negative value for @samp{N} chooses this scheme, ensuring that
6640
branches to stubs always use a negative offset.  Two special values of
6641
@samp{N} are recognized, @samp{1} and @samp{-1}.  These both instruct
6642
@command{ld} to automatically size input section groups for the branch types
6643
detected, with the same behaviour regarding stub placement as other
6644
positive or negative values of @samp{N} respectively.
6645
 
6646
Note that @samp{--stub-group-size} does not split input sections.  A
6647
single input section larger than the group size specified will of course
6648
create a larger group (of one section).  If input sections are too
6649
large, it may not be possible for a branch to reach its stub.
6650
 
6651
@cindex PowerPC64 stub symbols
6652
@kindex --emit-stub-syms
6653
@item --emit-stub-syms
6654
This option causes @command{ld} to label linker stubs with a local
6655
symbol that encodes the stub type and destination.
6656
 
6657
@cindex PowerPC64 dot symbols
6658
@kindex --dotsyms
6659
@kindex --no-dotsyms
6660
@item --dotsyms, --no-dotsyms
6661
These two options control how @command{ld} interprets version patterns
6662
in a version script.  Older PowerPC64 compilers emitted both a
6663
function descriptor symbol with the same name as the function, and a
6664
code entry symbol with the name prefixed by a dot (@samp{.}).  To
6665
properly version a function @samp{foo}, the version script thus needs
6666
to control both @samp{foo} and @samp{.foo}.  The option
6667
@samp{--dotsyms}, on by default, automatically adds the required
6668
dot-prefixed patterns.  Use @samp{--no-dotsyms} to disable this
6669
feature.
6670
 
6671
@cindex PowerPC64 TLS optimization
6672
@kindex --no-tls-optimize
6673
@item --no-tls-optimize
6674
PowerPC64 @command{ld} normally performs some optimization of code
6675
sequences used to access Thread-Local Storage.  Use this option to
6676
disable the optimization.
6677
 
6678
@cindex PowerPC64 OPD optimization
6679
@kindex --no-opd-optimize
6680
@item --no-opd-optimize
6681
PowerPC64 @command{ld} normally removes @code{.opd} section entries
6682
corresponding to deleted link-once functions, or functions removed by
6683
the action of @samp{--gc-sections} or linker script @code{/DISCARD/}.
6684
Use this option to disable @code{.opd} optimization.
6685
 
6686
@cindex PowerPC64 OPD spacing
6687
@kindex --non-overlapping-opd
6688
@item --non-overlapping-opd
6689
Some PowerPC64 compilers have an option to generate compressed
6690
@code{.opd} entries spaced 16 bytes apart, overlapping the third word,
6691
the static chain pointer (unused in C) with the first word of the next
6692
entry.  This option expands such entries to the full 24 bytes.
6693
 
6694
@cindex PowerPC64 TOC optimization
6695
@kindex --no-toc-optimize
6696
@item --no-toc-optimize
6697
PowerPC64 @command{ld} normally removes unused @code{.toc} section
6698
entries.  Such entries are detected by examining relocations that
6699
reference the TOC in code sections.  A reloc in a deleted code section
6700
marks a TOC word as unneeded, while a reloc in a kept code section
6701
marks a TOC word as needed.  Since the TOC may reference itself, TOC
6702
relocs are also examined.  TOC words marked as both needed and
6703
unneeded will of course be kept.  TOC words without any referencing
6704
reloc are assumed to be part of a multi-word entry, and are kept or
6705
discarded as per the nearest marked preceding word.  This works
6706
reliably for compiler generated code, but may be incorrect if assembly
6707
code is used to insert TOC entries.  Use this option to disable the
6708
optimization.
6709
 
6710
@cindex PowerPC64 multi-TOC
6711
@kindex --no-multi-toc
6712
@item --no-multi-toc
6713 166 khays
If given any toc option besides @code{-mcmodel=medium} or
6714
@code{-mcmodel=large}, PowerPC64 GCC generates code for a TOC model
6715
where TOC
6716 145 khays
entries are accessed with a 16-bit offset from r2.  This limits the
6717
total TOC size to 64K.  PowerPC64 @command{ld} extends this limit by
6718
grouping code sections such that each group uses less than 64K for its
6719
TOC entries, then inserts r2 adjusting stubs between inter-group
6720
calls.  @command{ld} does not split apart input sections, so cannot
6721
help if a single input file has a @code{.toc} section that exceeds
6722
64K, most likely from linking multiple files with @command{ld -r}.
6723
Use this option to turn off this feature.
6724 166 khays
 
6725
@cindex PowerPC64 TOC sorting
6726
@kindex --no-toc-sort
6727
@item --no-toc-sort
6728
By default, @command{ld} sorts TOC sections so that those whose file
6729
happens to have a section called @code{.init} or @code{.fini} are
6730
placed first, followed by TOC sections referenced by code generated
6731
with PowerPC64 gcc's @code{-mcmodel=small}, and lastly TOC sections
6732
referenced only by code generated with PowerPC64 gcc's
6733
@code{-mcmodel=medium} or @code{-mcmodel=large} options.  Doing this
6734
results in better TOC grouping for multi-TOC.  Use this option to turn
6735
off this feature.
6736
 
6737
@cindex PowerPC64 PLT stub alignment
6738
@kindex --plt-align
6739
@kindex --no-plt-align
6740
@item --plt-align
6741
@itemx --no-plt-align
6742
Use these options to control whether individual PLT call stubs are
6743
aligned to a 32-byte boundary, or to the specified power of two
6744
boundary when using @code{--plt-align=}.  By default PLT call stubs
6745
are packed tightly.
6746
 
6747
@cindex PowerPC64 PLT call stub static chain
6748
@kindex --plt-static-chain
6749
@kindex --no-plt-static-chain
6750
@item --plt-static-chain
6751
@itemx --no-plt-static-chain
6752
Use these options to control whether PLT call stubs load the static
6753
chain pointer (r11).  @code{ld} defaults to not loading the static
6754
chain since there is never any need to do so on a PLT call.
6755
 
6756
@cindex PowerPC64 PLT call stub thread safety
6757
@kindex --plt-thread-safe
6758
@kindex --no-plt-thread-safe
6759
@item --plt-thread-safe
6760
@itemx --no-thread-safe
6761
With power7's weakly ordered memory model, it is possible when using
6762
lazy binding for ld.so to update a plt entry in one thread and have
6763
another thread see the individual plt entry words update in the wrong
6764
order, despite ld.so carefully writing in the correct order and using
6765
memory write barriers.  To avoid this we need some sort of read
6766
barrier in the call stub, or use LD_BIND_NOW=1.  By default, @code{ld}
6767
looks for calls to commonly used functions that create threads, and if
6768
seen, adds the necessary barriers.  Use these options to change the
6769
default behaviour.
6770 145 khays
@end table
6771
 
6772
@ifclear GENERIC
6773
@lowersections
6774
@end ifclear
6775
@end ifset
6776
 
6777
@ifset SPU
6778
@ifclear GENERIC
6779
@raisesections
6780
@end ifclear
6781
 
6782
@node SPU ELF
6783
@section @command{ld} and SPU ELF Support
6784
 
6785
@cindex SPU ELF options
6786
@table @option
6787
 
6788
@cindex SPU plugins
6789
@kindex --plugin
6790
@item --plugin
6791
This option marks an executable as a PIC plugin module.
6792
 
6793
@cindex SPU overlays
6794
@kindex --no-overlays
6795
@item --no-overlays
6796
Normally, @command{ld} recognizes calls to functions within overlay
6797
regions, and redirects such calls to an overlay manager via a stub.
6798
@command{ld} also provides a built-in overlay manager.  This option
6799
turns off all this special overlay handling.
6800
 
6801
@cindex SPU overlay stub symbols
6802
@kindex --emit-stub-syms
6803
@item --emit-stub-syms
6804
This option causes @command{ld} to label overlay stubs with a local
6805
symbol that encodes the stub type and destination.
6806
 
6807
@cindex SPU extra overlay stubs
6808
@kindex --extra-overlay-stubs
6809
@item --extra-overlay-stubs
6810
This option causes @command{ld} to add overlay call stubs on all
6811
function calls out of overlay regions.  Normally stubs are not added
6812
on calls to non-overlay regions.
6813
 
6814
@cindex SPU local store size
6815
@kindex --local-store=lo:hi
6816
@item --local-store=lo:hi
6817
@command{ld} usually checks that a final executable for SPU fits in
6818
the address range 0 to 256k.  This option may be used to change the
6819
range.  Disable the check entirely with @option{--local-store=0:0}.
6820
 
6821
@cindex SPU
6822
@kindex --stack-analysis
6823
@item --stack-analysis
6824
SPU local store space is limited.  Over-allocation of stack space
6825
unnecessarily limits space available for code and data, while
6826
under-allocation results in runtime failures.  If given this option,
6827
@command{ld} will provide an estimate of maximum stack usage.
6828
@command{ld} does this by examining symbols in code sections to
6829
determine the extents of functions, and looking at function prologues
6830
for stack adjusting instructions.  A call-graph is created by looking
6831
for relocations on branch instructions.  The graph is then searched
6832
for the maximum stack usage path.  Note that this analysis does not
6833
find calls made via function pointers, and does not handle recursion
6834
and other cycles in the call graph.  Stack usage may be
6835
under-estimated if your code makes such calls.  Also, stack usage for
6836
dynamic allocation, e.g. alloca, will not be detected.  If a link map
6837
is requested, detailed information about each function's stack usage
6838
and calls will be given.
6839
 
6840
@cindex SPU
6841
@kindex --emit-stack-syms
6842
@item --emit-stack-syms
6843
This option, if given along with @option{--stack-analysis} will result
6844
in @command{ld} emitting stack sizing symbols for each function.
6845
These take the form @code{__stack_<function_name>} for global
6846
functions, and @code{__stack_<number>_<function_name>} for static
6847
functions.  @code{<number>} is the section id in hex.  The value of
6848
such symbols is the stack requirement for the corresponding function.
6849
The symbol size will be zero, type @code{STT_NOTYPE}, binding
6850
@code{STB_LOCAL}, and section @code{SHN_ABS}.
6851
@end table
6852
 
6853
@ifclear GENERIC
6854
@lowersections
6855
@end ifclear
6856
@end ifset
6857
 
6858
@ifset TICOFF
6859
@ifclear GENERIC
6860
@raisesections
6861
@end ifclear
6862
 
6863
@node TI COFF
6864
@section @command{ld}'s Support for Various TI COFF Versions
6865
@cindex TI COFF versions
6866
@kindex --format=@var{version}
6867
The @samp{--format} switch allows selection of one of the various
6868
TI COFF versions.  The latest of this writing is 2; versions 0 and 1 are
6869
also supported.  The TI COFF versions also vary in header byte-order
6870
format; @command{ld} will read any version or byte order, but the output
6871
header format depends on the default specified by the specific target.
6872
 
6873
@ifclear GENERIC
6874
@lowersections
6875
@end ifclear
6876
@end ifset
6877
 
6878
@ifset WIN32
6879
@ifclear GENERIC
6880
@raisesections
6881
@end ifclear
6882
 
6883
@node WIN32
6884
@section @command{ld} and WIN32 (cygwin/mingw)
6885
 
6886
This section describes some of the win32 specific @command{ld} issues.
6887
See @ref{Options,,Command Line Options} for detailed description of the
6888
command line options mentioned here.
6889
 
6890
@table @emph
6891
@cindex import libraries
6892
@item import libraries
6893
The standard Windows linker creates and uses so-called import
6894
libraries, which contains information for linking to dll's.  They are
6895
regular static archives and are handled as any other static
6896
archive.  The cygwin and mingw ports of @command{ld} have specific
6897
support for creating such libraries provided with the
6898
@samp{--out-implib} command line option.
6899
 
6900
@item   exporting DLL symbols
6901
@cindex exporting DLL symbols
6902
The cygwin/mingw @command{ld} has several ways to export symbols for dll's.
6903
 
6904
@table @emph
6905
@item   using auto-export functionality
6906
@cindex using auto-export functionality
6907
By default @command{ld} exports symbols with the auto-export functionality,
6908
which is controlled by the following command line options:
6909
 
6910
@itemize
6911
@item --export-all-symbols   [This is the default]
6912
@item --exclude-symbols
6913
@item --exclude-libs
6914
@item --exclude-modules-for-implib
6915
@item --version-script
6916
@end itemize
6917
 
6918
When auto-export is in operation, @command{ld} will export all the non-local
6919
(global and common) symbols it finds in a DLL, with the exception of a few
6920
symbols known to belong to the system's runtime and libraries.  As it will
6921
often not be desirable to export all of a DLL's symbols, which may include
6922
private functions that are not part of any public interface, the command-line
6923
options listed above may be used to filter symbols out from the list for
6924
exporting.  The @samp{--output-def} option can be used in order to see the
6925
final list of exported symbols with all exclusions taken into effect.
6926
 
6927
If @samp{--export-all-symbols} is not given explicitly on the
6928
command line, then the default auto-export behavior will be @emph{disabled}
6929
if either of the following are true:
6930
 
6931
@itemize
6932
@item A DEF file is used.
6933
@item Any symbol in any object file was marked with the __declspec(dllexport) attribute.
6934
@end itemize
6935
 
6936
@item   using a DEF file
6937
@cindex using a DEF file
6938
Another way of exporting symbols is using a DEF file.  A DEF file is
6939
an ASCII file containing definitions of symbols which should be
6940
exported when a dll is created.  Usually it is named @samp{<dll
6941
name>.def} and is added as any other object file to the linker's
6942
command line.  The file's name must end in @samp{.def} or @samp{.DEF}.
6943
 
6944
@example
6945
gcc -o <output> <objectfiles> <dll name>.def
6946
@end example
6947
 
6948
Using a DEF file turns off the normal auto-export behavior, unless the
6949
@samp{--export-all-symbols} option is also used.
6950
 
6951
Here is an example of a DEF file for a shared library called @samp{xyz.dll}:
6952
 
6953
@example
6954
LIBRARY "xyz.dll" BASE=0x20000000
6955
 
6956
EXPORTS
6957
foo
6958
bar
6959
_bar = bar
6960
another_foo = abc.dll.afoo
6961
var1 DATA
6962
doo = foo == foo2
6963
eoo DATA == var1
6964
@end example
6965
 
6966
This example defines a DLL with a non-default base address and seven
6967
symbols in the export table. The third exported symbol @code{_bar} is an
6968
alias for the second. The fourth symbol, @code{another_foo} is resolved
6969
by "forwarding" to another module and treating it as an alias for
6970
@code{afoo} exported from the DLL @samp{abc.dll}. The final symbol
6971
@code{var1} is declared to be a data object. The @samp{doo} symbol in
6972
export library is an alias of @samp{foo}, which gets the string name
6973
in export table @samp{foo2}. The @samp{eoo} symbol is an data export
6974
symbol, which gets in export table the name @samp{var1}.
6975
 
6976
The optional @code{LIBRARY <name>} command indicates the @emph{internal}
6977
name of the output DLL. If @samp{<name>} does not include a suffix,
6978
the default library suffix, @samp{.DLL} is appended.
6979
 
6980
When the .DEF file is used to build an application, rather than a
6981
library, the @code{NAME <name>} command should be used instead of
6982
@code{LIBRARY}. If @samp{<name>} does not include a suffix, the default
6983
executable suffix, @samp{.EXE} is appended.
6984
 
6985
With either @code{LIBRARY <name>} or @code{NAME <name>} the optional
6986
specification @code{BASE = <number>} may be used to specify a
6987
non-default base address for the image.
6988
 
6989
If neither @code{LIBRARY <name>} nor  @code{NAME <name>} is specified,
6990
or they specify an empty string, the internal name is the same as the
6991
filename specified on the command line.
6992
 
6993
The complete specification of an export symbol is:
6994
 
6995
@example
6996
EXPORTS
6997
  ( (  ( <name1> [ = <name2> ] )
6998
     | ( <name1> = <module-name> . <external-name>))
6999
  [ @@ <integer> ] [NONAME] [DATA] [CONSTANT] [PRIVATE] [== <name3>] ) *
7000
@end example
7001
 
7002
Declares @samp{<name1>} as an exported symbol from the DLL, or declares
7003
@samp{<name1>} as an exported alias for @samp{<name2>}; or declares
7004
@samp{<name1>} as a "forward" alias for the symbol
7005
@samp{<external-name>} in the DLL @samp{<module-name>}.
7006
Optionally, the symbol may be exported by the specified ordinal
7007
@samp{<integer>} alias. The optional @samp{<name3>} is the to be used
7008
string in import/export table for the symbol.
7009
 
7010
The optional keywords that follow the declaration indicate:
7011
 
7012
@code{NONAME}: Do not put the symbol name in the DLL's export table.  It
7013
will still be exported by its ordinal alias (either the value specified
7014
by the .def specification or, otherwise, the value assigned by the
7015
linker). The symbol name, however, does remain visible in the import
7016
library (if any), unless @code{PRIVATE} is also specified.
7017
 
7018
@code{DATA}: The symbol is a variable or object, rather than a function.
7019
The import lib will export only an indirect reference to @code{foo} as
7020
the symbol @code{_imp__foo} (ie, @code{foo} must be resolved as
7021
@code{*_imp__foo}).
7022
 
7023
@code{CONSTANT}: Like @code{DATA}, but put the undecorated @code{foo} as
7024
well as @code{_imp__foo} into the import library. Both refer to the
7025
read-only import address table's pointer to the variable, not to the
7026
variable itself. This can be dangerous. If the user code fails to add
7027
the @code{dllimport} attribute and also fails to explicitly add the
7028
extra indirection that the use of the attribute enforces, the
7029
application will behave unexpectedly.
7030
 
7031
@code{PRIVATE}: Put the symbol in the DLL's export table, but do not put
7032
it into the static import library used to resolve imports at link time. The
7033
symbol can still be imported using the @code{LoadLibrary/GetProcAddress}
7034
API at runtime or by by using the GNU ld extension of linking directly to
7035
the DLL without an import library.
7036
 
7037
See ld/deffilep.y in the binutils sources for the full specification of
7038
other DEF file statements
7039
 
7040
@cindex creating a DEF file
7041
While linking a shared dll, @command{ld} is able to create a DEF file
7042
with the @samp{--output-def <file>} command line option.
7043
 
7044
@item   Using decorations
7045
@cindex Using decorations
7046
Another way of marking symbols for export is to modify the source code
7047
itself, so that when building the DLL each symbol to be exported is
7048
declared as:
7049
 
7050
@example
7051
__declspec(dllexport) int a_variable
7052
__declspec(dllexport) void a_function(int with_args)
7053
@end example
7054
 
7055
All such symbols will be exported from the DLL.  If, however,
7056
any of the object files in the DLL contain symbols decorated in
7057
this way, then the normal auto-export behavior is disabled, unless
7058
the @samp{--export-all-symbols} option is also used.
7059
 
7060
Note that object files that wish to access these symbols must @emph{not}
7061
decorate them with dllexport.  Instead, they should use dllimport,
7062
instead:
7063
 
7064
@example
7065
__declspec(dllimport) int a_variable
7066
__declspec(dllimport) void a_function(int with_args)
7067
@end example
7068
 
7069
This complicates the structure of library header files, because
7070
when included by the library itself the header must declare the
7071
variables and functions as dllexport, but when included by client
7072
code the header must declare them as dllimport.  There are a number
7073
of idioms that are typically used to do this; often client code can
7074
omit the __declspec() declaration completely.  See
7075
@samp{--enable-auto-import} and @samp{automatic data imports} for more
7076
information.
7077
@end table
7078
 
7079
@cindex automatic data imports
7080
@item automatic data imports
7081
The standard Windows dll format supports data imports from dlls only
7082
by adding special decorations (dllimport/dllexport), which let the
7083
compiler produce specific assembler instructions to deal with this
7084
issue.  This increases the effort necessary to port existing Un*x
7085
code to these platforms, especially for large
7086
c++ libraries and applications.  The auto-import feature, which was
7087
initially provided by Paul Sokolovsky, allows one to omit the
7088
decorations to achieve a behavior that conforms to that on POSIX/Un*x
7089
platforms. This feature is enabled with the @samp{--enable-auto-import}
7090
command-line option, although it is enabled by default on cygwin/mingw.
7091
The @samp{--enable-auto-import} option itself now serves mainly to
7092
suppress any warnings that are ordinarily emitted when linked objects
7093
trigger the feature's use.
7094
 
7095
auto-import of variables does not always work flawlessly without
7096
additional assistance.  Sometimes, you will see this message
7097
 
7098
"variable '<var>' can't be auto-imported. Please read the
7099
documentation for ld's @code{--enable-auto-import} for details."
7100
 
7101
The @samp{--enable-auto-import} documentation explains why this error
7102
occurs, and several methods that can be used to overcome this difficulty.
7103
One of these methods is the @emph{runtime pseudo-relocs} feature, described
7104
below.
7105
 
7106
@cindex runtime pseudo-relocation
7107
For complex variables imported from DLLs (such as structs or classes),
7108
object files typically contain a base address for the variable and an
7109
offset (@emph{addend}) within the variable--to specify a particular
7110
field or public member, for instance.  Unfortunately, the runtime loader used
7111
in win32 environments is incapable of fixing these references at runtime
7112
without the additional information supplied by dllimport/dllexport decorations.
7113
The standard auto-import feature described above is unable to resolve these
7114
references.
7115
 
7116
The @samp{--enable-runtime-pseudo-relocs} switch allows these references to
7117
be resolved without error, while leaving the task of adjusting the references
7118
themselves (with their non-zero addends) to specialized code provided by the
7119
runtime environment.  Recent versions of the cygwin and mingw environments and
7120
compilers provide this runtime support; older versions do not.  However, the
7121
support is only necessary on the developer's platform; the compiled result will
7122
run without error on an older system.
7123
 
7124
@samp{--enable-runtime-pseudo-relocs} is not the default; it must be explicitly
7125
enabled as needed.
7126
 
7127
@cindex direct linking to a dll
7128
@item direct linking to a dll
7129
The cygwin/mingw ports of @command{ld} support the direct linking,
7130
including data symbols, to a dll without the usage of any import
7131
libraries.  This is much faster and uses much less memory than does the
7132
traditional import library method, especially when linking large
7133
libraries or applications.  When @command{ld} creates an import lib, each
7134
function or variable exported from the dll is stored in its own bfd, even
7135
though a single bfd could contain many exports.  The overhead involved in
7136
storing, loading, and processing so many bfd's is quite large, and explains the
7137
tremendous time, memory, and storage needed to link against particularly
7138
large or complex libraries when using import libs.
7139
 
7140
Linking directly to a dll uses no extra command-line switches other than
7141
@samp{-L} and @samp{-l}, because @command{ld} already searches for a number
7142
of names to match each library.  All that is needed from the developer's
7143
perspective is an understanding of this search, in order to force ld to
7144
select the dll instead of an import library.
7145
 
7146
 
7147
For instance, when ld is called with the argument @samp{-lxxx} it will attempt
7148
to find, in the first directory of its search path,
7149
 
7150
@example
7151
libxxx.dll.a
7152
xxx.dll.a
7153
libxxx.a
7154
xxx.lib
7155
cygxxx.dll (*)
7156
libxxx.dll
7157
xxx.dll
7158
@end example
7159
 
7160
before moving on to the next directory in the search path.
7161
 
7162
(*) Actually, this is not @samp{cygxxx.dll} but in fact is @samp{<prefix>xxx.dll},
7163
where @samp{<prefix>} is set by the @command{ld} option
7164
@samp{--dll-search-prefix=<prefix>}. In the case of cygwin, the standard gcc spec
7165
file includes @samp{--dll-search-prefix=cyg}, so in effect we actually search for
7166
@samp{cygxxx.dll}.
7167
 
7168
Other win32-based unix environments, such as mingw or pw32, may use other
7169
@samp{<prefix>}es, although at present only cygwin makes use of this feature.  It
7170
was originally intended to help avoid name conflicts among dll's built for the
7171
various win32/un*x environments, so that (for example) two versions of a zlib dll
7172
could coexist on the same machine.
7173
 
7174
The generic cygwin/mingw path layout uses a @samp{bin} directory for
7175
applications and dll's and a @samp{lib} directory for the import
7176
libraries (using cygwin nomenclature):
7177
 
7178
@example
7179
bin/
7180
        cygxxx.dll
7181
lib/
7182
        libxxx.dll.a   (in case of dll's)
7183
        libxxx.a       (in case of static archive)
7184
@end example
7185
 
7186
Linking directly to a dll without using the import library can be
7187
done two ways:
7188
 
7189
1. Use the dll directly by adding the @samp{bin} path to the link line
7190
@example
7191
gcc -Wl,-verbose  -o a.exe -L../bin/ -lxxx
7192
@end example
7193
 
7194
However, as the dll's often have version numbers appended to their names
7195
(@samp{cygncurses-5.dll}) this will often fail, unless one specifies
7196
@samp{-L../bin -lncurses-5} to include the version.  Import libs are generally
7197
not versioned, and do not have this difficulty.
7198
 
7199
2. Create a symbolic link from the dll to a file in the @samp{lib}
7200
directory according to the above mentioned search pattern.  This
7201
should be used to avoid unwanted changes in the tools needed for
7202
making the app/dll.
7203
 
7204
@example
7205
ln -s bin/cygxxx.dll lib/[cyg|lib|]xxx.dll[.a]
7206
@end example
7207
 
7208
Then you can link without any make environment changes.
7209
 
7210
@example
7211
gcc -Wl,-verbose  -o a.exe -L../lib/ -lxxx
7212
@end example
7213
 
7214
This technique also avoids the version number problems, because the following is
7215
perfectly legal
7216
 
7217
@example
7218
bin/
7219
        cygxxx-5.dll
7220
lib/
7221
        libxxx.dll.a -> ../bin/cygxxx-5.dll
7222
@end example
7223
 
7224
Linking directly to a dll without using an import lib will work
7225
even when auto-import features are exercised, and even when
7226
@samp{--enable-runtime-pseudo-relocs} is used.
7227
 
7228
Given the improvements in speed and memory usage, one might justifiably
7229
wonder why import libraries are used at all.  There are three reasons:
7230
 
7231
1. Until recently, the link-directly-to-dll functionality did @emph{not}
7232
work with auto-imported data.
7233
 
7234
2. Sometimes it is necessary to include pure static objects within the
7235
import library (which otherwise contains only bfd's for indirection
7236
symbols that point to the exports of a dll).  Again, the import lib
7237
for the cygwin kernel makes use of this ability, and it is not
7238
possible to do this without an import lib.
7239
 
7240
3. Symbol aliases can only be resolved using an import lib.  This is
7241
critical when linking against OS-supplied dll's (eg, the win32 API)
7242
in which symbols are usually exported as undecorated aliases of their
7243
stdcall-decorated assembly names.
7244
 
7245
So, import libs are not going away.  But the ability to replace
7246
true import libs with a simple symbolic link to (or a copy of)
7247
a dll, in many cases, is a useful addition to the suite of tools
7248
binutils makes available to the win32 developer.  Given the
7249
massive improvements in memory requirements during linking, storage
7250
requirements, and linking speed, we expect that many developers
7251
will soon begin to use this feature whenever possible.
7252
 
7253
@item symbol aliasing
7254
@table @emph
7255
@item adding additional names
7256
Sometimes, it is useful to export symbols with additional names.
7257
A symbol @samp{foo} will be exported as @samp{foo}, but it can also be
7258
exported as @samp{_foo} by using special directives in the DEF file
7259
when creating the dll.  This will affect also the optional created
7260
import library.  Consider the following DEF file:
7261
 
7262
@example
7263
LIBRARY "xyz.dll" BASE=0x61000000
7264
 
7265
EXPORTS
7266
foo
7267
_foo = foo
7268
@end example
7269
 
7270
The line @samp{_foo = foo} maps the symbol @samp{foo} to @samp{_foo}.
7271
 
7272
Another method for creating a symbol alias is to create it in the
7273
source code using the "weak" attribute:
7274
 
7275
@example
7276
void foo () @{ /* Do something.  */; @}
7277
void _foo () __attribute__ ((weak, alias ("foo")));
7278
@end example
7279
 
7280
See the gcc manual for more information about attributes and weak
7281
symbols.
7282
 
7283
@item renaming symbols
7284
Sometimes it is useful to rename exports.  For instance, the cygwin
7285
kernel does this regularly.  A symbol @samp{_foo} can be exported as
7286
@samp{foo} but not as @samp{_foo} by using special directives in the
7287
DEF file. (This will also affect the import library, if it is
7288
created).  In the following example:
7289
 
7290
@example
7291
LIBRARY "xyz.dll" BASE=0x61000000
7292
 
7293
EXPORTS
7294
_foo = foo
7295
@end example
7296
 
7297
The line @samp{_foo = foo} maps the exported symbol @samp{foo} to
7298
@samp{_foo}.
7299
@end table
7300
 
7301
Note: using a DEF file disables the default auto-export behavior,
7302
unless the @samp{--export-all-symbols} command line option is used.
7303
If, however, you are trying to rename symbols, then you should list
7304
@emph{all} desired exports in the DEF file, including the symbols
7305
that are not being renamed, and do @emph{not} use the
7306
@samp{--export-all-symbols} option.  If you list only the
7307
renamed symbols in the DEF file, and use @samp{--export-all-symbols}
7308
to handle the other symbols, then the both the new names @emph{and}
7309
the original names for the renamed symbols will be exported.
7310
In effect, you'd be aliasing those symbols, not renaming them,
7311
which is probably not what you wanted.
7312
 
7313
@cindex weak externals
7314
@item weak externals
7315
The Windows object format, PE, specifies a form of weak symbols called
7316
weak externals.  When a weak symbol is linked and the symbol is not
7317
defined, the weak symbol becomes an alias for some other symbol.  There
7318
are three variants of weak externals:
7319
@itemize
7320
@item Definition is searched for in objects and libraries, historically
7321
called lazy externals.
7322
@item Definition is searched for only in other objects, not in libraries.
7323
This form is not presently implemented.
7324
@item No search; the symbol is an alias.  This form is not presently
7325
implemented.
7326
@end itemize
7327
As a GNU extension, weak symbols that do not specify an alternate symbol
7328
are supported.  If the symbol is undefined when linking, the symbol
7329
uses a default value.
7330
 
7331
@cindex aligned common symbols
7332
@item aligned common symbols
7333
As a GNU extension to the PE file format, it is possible to specify the
7334
desired alignment for a common symbol.  This information is conveyed from
7335
the assembler or compiler to the linker by means of GNU-specific commands
7336
carried in the object file's @samp{.drectve} section, which are recognized
7337
by @command{ld} and respected when laying out the common symbols.  Native
7338
tools will be able to process object files employing this GNU extension,
7339
but will fail to respect the alignment instructions, and may issue noisy
7340
warnings about unknown linker directives.
7341
@end table
7342
 
7343
@ifclear GENERIC
7344
@lowersections
7345
@end ifclear
7346
@end ifset
7347
 
7348
@ifset XTENSA
7349
@ifclear GENERIC
7350
@raisesections
7351
@end ifclear
7352
 
7353
@node Xtensa
7354
@section @code{ld} and Xtensa Processors
7355
 
7356
@cindex Xtensa processors
7357
The default @command{ld} behavior for Xtensa processors is to interpret
7358
@code{SECTIONS} commands so that lists of explicitly named sections in a
7359
specification with a wildcard file will be interleaved when necessary to
7360
keep literal pools within the range of PC-relative load offsets.  For
7361
example, with the command:
7362
 
7363
@smallexample
7364
SECTIONS
7365
@{
7366
  .text : @{
7367
    *(.literal .text)
7368
  @}
7369
@}
7370
@end smallexample
7371
 
7372
@noindent
7373
@command{ld} may interleave some of the @code{.literal}
7374
and @code{.text} sections from different object files to ensure that the
7375
literal pools are within the range of PC-relative load offsets.  A valid
7376
interleaving might place the @code{.literal} sections from an initial
7377
group of files followed by the @code{.text} sections of that group of
7378
files.  Then, the @code{.literal} sections from the rest of the files
7379
and the @code{.text} sections from the rest of the files would follow.
7380
 
7381
@cindex @option{--relax} on Xtensa
7382
@cindex relaxing on Xtensa
7383
Relaxation is enabled by default for the Xtensa version of @command{ld} and
7384
provides two important link-time optimizations.  The first optimization
7385
is to combine identical literal values to reduce code size.  A redundant
7386
literal will be removed and all the @code{L32R} instructions that use it
7387
will be changed to reference an identical literal, as long as the
7388
location of the replacement literal is within the offset range of all
7389
the @code{L32R} instructions.  The second optimization is to remove
7390
unnecessary overhead from assembler-generated ``longcall'' sequences of
7391
@code{L32R}/@code{CALLX@var{n}} when the target functions are within
7392
range of direct @code{CALL@var{n}} instructions.
7393
 
7394
For each of these cases where an indirect call sequence can be optimized
7395
to a direct call, the linker will change the @code{CALLX@var{n}}
7396
instruction to a @code{CALL@var{n}} instruction, remove the @code{L32R}
7397
instruction, and remove the literal referenced by the @code{L32R}
7398
instruction if it is not used for anything else.  Removing the
7399
@code{L32R} instruction always reduces code size but can potentially
7400
hurt performance by changing the alignment of subsequent branch targets.
7401
By default, the linker will always preserve alignments, either by
7402
switching some instructions between 24-bit encodings and the equivalent
7403
density instructions or by inserting a no-op in place of the @code{L32R}
7404
instruction that was removed.  If code size is more important than
7405
performance, the @option{--size-opt} option can be used to prevent the
7406
linker from widening density instructions or inserting no-ops, except in
7407
a few cases where no-ops are required for correctness.
7408
 
7409
The following Xtensa-specific command-line options can be used to
7410
control the linker:
7411
 
7412
@cindex Xtensa options
7413
@table @option
7414
@item --size-opt
7415
When optimizing indirect calls to direct calls, optimize for code size
7416
more than performance.  With this option, the linker will not insert
7417
no-ops or widen density instructions to preserve branch target
7418
alignment.  There may still be some cases where no-ops are required to
7419
preserve the correctness of the code.
7420
@end table
7421
 
7422
@ifclear GENERIC
7423
@lowersections
7424
@end ifclear
7425
@end ifset
7426
 
7427
@ifclear SingleFormat
7428
@node BFD
7429
@chapter BFD
7430
 
7431
@cindex back end
7432
@cindex object file management
7433
@cindex object formats available
7434
@kindex objdump -i
7435
The linker accesses object and archive files using the BFD libraries.
7436
These libraries allow the linker to use the same routines to operate on
7437
object files whatever the object file format.  A different object file
7438
format can be supported simply by creating a new BFD back end and adding
7439
it to the library.  To conserve runtime memory, however, the linker and
7440
associated tools are usually configured to support only a subset of the
7441
object file formats available.  You can use @code{objdump -i}
7442
(@pxref{objdump,,objdump,binutils.info,The GNU Binary Utilities}) to
7443
list all the formats available for your configuration.
7444
 
7445
@cindex BFD requirements
7446
@cindex requirements for BFD
7447
As with most implementations, BFD is a compromise between
7448
several conflicting requirements. The major factor influencing
7449
BFD design was efficiency: any time used converting between
7450
formats is time which would not have been spent had BFD not
7451
been involved. This is partly offset by abstraction payback; since
7452
BFD simplifies applications and back ends, more time and care
7453
may be spent optimizing algorithms for a greater speed.
7454
 
7455
One minor artifact of the BFD solution which you should bear in
7456
mind is the potential for information loss.  There are two places where
7457
useful information can be lost using the BFD mechanism: during
7458
conversion and during output. @xref{BFD information loss}.
7459
 
7460
@menu
7461
* BFD outline::                 How it works: an outline of BFD
7462
@end menu
7463
 
7464
@node BFD outline
7465
@section How It Works: An Outline of BFD
7466
@cindex opening object files
7467
@include bfdsumm.texi
7468
@end ifclear
7469
 
7470
@node Reporting Bugs
7471
@chapter Reporting Bugs
7472
@cindex bugs in @command{ld}
7473
@cindex reporting bugs in @command{ld}
7474
 
7475
Your bug reports play an essential role in making @command{ld} reliable.
7476
 
7477
Reporting a bug may help you by bringing a solution to your problem, or
7478
it may not.  But in any case the principal function of a bug report is
7479
to help the entire community by making the next version of @command{ld}
7480
work better.  Bug reports are your contribution to the maintenance of
7481
@command{ld}.
7482
 
7483
In order for a bug report to serve its purpose, you must include the
7484
information that enables us to fix the bug.
7485
 
7486
@menu
7487
* Bug Criteria::                Have you found a bug?
7488
* Bug Reporting::               How to report bugs
7489
@end menu
7490
 
7491
@node Bug Criteria
7492
@section Have You Found a Bug?
7493
@cindex bug criteria
7494
 
7495
If you are not sure whether you have found a bug, here are some guidelines:
7496
 
7497
@itemize @bullet
7498
@cindex fatal signal
7499
@cindex linker crash
7500
@cindex crash of linker
7501
@item
7502
If the linker gets a fatal signal, for any input whatever, that is a
7503
@command{ld} bug.  Reliable linkers never crash.
7504
 
7505
@cindex error on valid input
7506
@item
7507
If @command{ld} produces an error message for valid input, that is a bug.
7508
 
7509
@cindex invalid input
7510
@item
7511
If @command{ld} does not produce an error message for invalid input, that
7512
may be a bug.  In the general case, the linker can not verify that
7513
object files are correct.
7514
 
7515
@item
7516
If you are an experienced user of linkers, your suggestions for
7517
improvement of @command{ld} are welcome in any case.
7518
@end itemize
7519
 
7520
@node Bug Reporting
7521
@section How to Report Bugs
7522
@cindex bug reports
7523
@cindex @command{ld} bugs, reporting
7524
 
7525
A number of companies and individuals offer support for @sc{gnu}
7526
products.  If you obtained @command{ld} from a support organization, we
7527
recommend you contact that organization first.
7528
 
7529
You can find contact information for many support companies and
7530
individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7531
distribution.
7532
 
7533
@ifset BUGURL
7534
Otherwise, send bug reports for @command{ld} to
7535
@value{BUGURL}.
7536
@end ifset
7537
 
7538
The fundamental principle of reporting bugs usefully is this:
7539
@strong{report all the facts}.  If you are not sure whether to state a
7540
fact or leave it out, state it!
7541
 
7542
Often people omit facts because they think they know what causes the
7543
problem and assume that some details do not matter.  Thus, you might
7544
assume that the name of a symbol you use in an example does not
7545
matter.  Well, probably it does not, but one cannot be sure.  Perhaps
7546
the bug is a stray memory reference which happens to fetch from the
7547
location where that name is stored in memory; perhaps, if the name
7548
were different, the contents of that location would fool the linker
7549
into doing the right thing despite the bug.  Play it safe and give a
7550
specific, complete example.  That is the easiest thing for you to do,
7551
and the most helpful.
7552
 
7553
Keep in mind that the purpose of a bug report is to enable us to fix
7554
the bug if it is new to us.  Therefore, always write your bug reports
7555
on the assumption that the bug has not been reported previously.
7556
 
7557
Sometimes people give a few sketchy facts and ask, ``Does this ring a
7558
bell?''  This cannot help us fix a bug, so it is basically useless.  We
7559
respond by asking for enough details to enable us to investigate.
7560
You might as well expedite matters by sending them to begin with.
7561
 
7562
To enable us to fix the bug, you should include all these things:
7563
 
7564
@itemize @bullet
7565
@item
7566
The version of @command{ld}.  @command{ld} announces it if you start it with
7567
the @samp{--version} argument.
7568
 
7569
Without this, we will not know whether there is any point in looking for
7570
the bug in the current version of @command{ld}.
7571
 
7572
@item
7573
Any patches you may have applied to the @command{ld} source, including any
7574
patches made to the @code{BFD} library.
7575
 
7576
@item
7577
The type of machine you are using, and the operating system name and
7578
version number.
7579
 
7580
@item
7581
What compiler (and its version) was used to compile @command{ld}---e.g.
7582
``@code{gcc-2.7}''.
7583
 
7584
@item
7585
The command arguments you gave the linker to link your example and
7586
observe the bug.  To guarantee you will not omit something important,
7587
list them all.  A copy of the Makefile (or the output from make) is
7588
sufficient.
7589
 
7590
If we were to try to guess the arguments, we would probably guess wrong
7591
and then we might not encounter the bug.
7592
 
7593
@item
7594
A complete input file, or set of input files, that will reproduce the
7595
bug.  It is generally most helpful to send the actual object files
7596
provided that they are reasonably small.  Say no more than 10K.  For
7597
bigger files you can either make them available by FTP or HTTP or else
7598
state that you are willing to send the object file(s) to whomever
7599
requests them.  (Note - your email will be going to a mailing list, so
7600
we do not want to clog it up with large attachments).  But small
7601
attachments are best.
7602
 
7603
If the source files were assembled using @code{gas} or compiled using
7604
@code{gcc}, then it may be OK to send the source files rather than the
7605
object files.  In this case, be sure to say exactly what version of
7606
@code{gas} or @code{gcc} was used to produce the object files.  Also say
7607
how @code{gas} or @code{gcc} were configured.
7608
 
7609
@item
7610
A description of what behavior you observe that you believe is
7611
incorrect.  For example, ``It gets a fatal signal.''
7612
 
7613
Of course, if the bug is that @command{ld} gets a fatal signal, then we
7614
will certainly notice it.  But if the bug is incorrect output, we might
7615
not notice unless it is glaringly wrong.  You might as well not give us
7616
a chance to make a mistake.
7617
 
7618
Even if the problem you experience is a fatal signal, you should still
7619
say so explicitly.  Suppose something strange is going on, such as, your
7620
copy of @command{ld} is out of sync, or you have encountered a bug in the
7621
C library on your system.  (This has happened!)  Your copy might crash
7622
and ours would not.  If you told us to expect a crash, then when ours
7623
fails to crash, we would know that the bug was not happening for us.  If
7624
you had not told us to expect a crash, then we would not be able to draw
7625
any conclusion from our observations.
7626
 
7627
@item
7628
If you wish to suggest changes to the @command{ld} source, send us context
7629
diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or
7630
@samp{-p} option.  Always send diffs from the old file to the new file.
7631
If you even discuss something in the @command{ld} source, refer to it by
7632
context, not by line number.
7633
 
7634
The line numbers in our development sources will not match those in your
7635
sources.  Your line numbers would convey no useful information to us.
7636
@end itemize
7637
 
7638
Here are some things that are not necessary:
7639
 
7640
@itemize @bullet
7641
@item
7642
A description of the envelope of the bug.
7643
 
7644
Often people who encounter a bug spend a lot of time investigating
7645
which changes to the input file will make the bug go away and which
7646
changes will not affect it.
7647
 
7648
This is often time consuming and not very useful, because the way we
7649
will find the bug is by running a single example under the debugger
7650
with breakpoints, not by pure deduction from a series of examples.
7651
We recommend that you save your time for something else.
7652
 
7653
Of course, if you can find a simpler example to report @emph{instead}
7654
of the original one, that is a convenience for us.  Errors in the
7655
output will be easier to spot, running under the debugger will take
7656
less time, and so on.
7657
 
7658
However, simplification is not vital; if you do not want to do this,
7659
report the bug anyway and send us the entire test case you used.
7660
 
7661
@item
7662
A patch for the bug.
7663
 
7664
A patch for the bug does help us if it is a good one.  But do not omit
7665
the necessary information, such as the test case, on the assumption that
7666
a patch is all we need.  We might see problems with your patch and decide
7667
to fix the problem another way, or we might not understand it at all.
7668
 
7669
Sometimes with a program as complicated as @command{ld} it is very hard to
7670
construct an example that will make the program follow a certain path
7671
through the code.  If you do not send us the example, we will not be
7672
able to construct one, so we will not be able to verify that the bug is
7673
fixed.
7674
 
7675
And if we cannot understand what bug you are trying to fix, or why your
7676
patch should be an improvement, we will not install it.  A test case will
7677
help us to understand.
7678
 
7679
@item
7680
A guess about what the bug is or what it depends on.
7681
 
7682
Such guesses are usually wrong.  Even we cannot guess right about such
7683
things without first using the debugger to find the facts.
7684
@end itemize
7685
 
7686
@node MRI
7687
@appendix MRI Compatible Script Files
7688
@cindex MRI compatibility
7689
To aid users making the transition to @sc{gnu} @command{ld} from the MRI
7690
linker, @command{ld} can use MRI compatible linker scripts as an
7691
alternative to the more general-purpose linker scripting language
7692
described in @ref{Scripts}.  MRI compatible linker scripts have a much
7693
simpler command set than the scripting language otherwise used with
7694
@command{ld}.  @sc{gnu} @command{ld} supports the most commonly used MRI
7695
linker commands; these commands are described here.
7696
 
7697
In general, MRI scripts aren't of much use with the @code{a.out} object
7698
file format, since it only has three sections and MRI scripts lack some
7699
features to make use of them.
7700
 
7701
You can specify a file containing an MRI-compatible script using the
7702
@samp{-c} command-line option.
7703
 
7704
Each command in an MRI-compatible script occupies its own line; each
7705
command line starts with the keyword that identifies the command (though
7706
blank lines are also allowed for punctuation).  If a line of an
7707
MRI-compatible script begins with an unrecognized keyword, @command{ld}
7708
issues a warning message, but continues processing the script.
7709
 
7710
Lines beginning with @samp{*} are comments.
7711
 
7712
You can write these commands using all upper-case letters, or all
7713
lower case; for example, @samp{chip} is the same as @samp{CHIP}.
7714
The following list shows only the upper-case form of each command.
7715
 
7716
@table @code
7717
@cindex @code{ABSOLUTE} (MRI)
7718
@item ABSOLUTE @var{secname}
7719
@itemx ABSOLUTE @var{secname}, @var{secname}, @dots{} @var{secname}
7720
Normally, @command{ld} includes in the output file all sections from all
7721
the input files.  However, in an MRI-compatible script, you can use the
7722
@code{ABSOLUTE} command to restrict the sections that will be present in
7723
your output program.  If the @code{ABSOLUTE} command is used at all in a
7724
script, then only the sections named explicitly in @code{ABSOLUTE}
7725
commands will appear in the linker output.  You can still use other
7726
input sections (whatever you select on the command line, or using
7727
@code{LOAD}) to resolve addresses in the output file.
7728
 
7729
@cindex @code{ALIAS} (MRI)
7730
@item ALIAS @var{out-secname}, @var{in-secname}
7731
Use this command to place the data from input section @var{in-secname}
7732
in a section called @var{out-secname} in the linker output file.
7733
 
7734
@var{in-secname} may be an integer.
7735
 
7736
@cindex @code{ALIGN} (MRI)
7737
@item ALIGN @var{secname} = @var{expression}
7738
Align the section called @var{secname} to @var{expression}.  The
7739
@var{expression} should be a power of two.
7740
 
7741
@cindex @code{BASE} (MRI)
7742
@item BASE @var{expression}
7743
Use the value of @var{expression} as the lowest address (other than
7744
absolute addresses) in the output file.
7745
 
7746
@cindex @code{CHIP} (MRI)
7747
@item CHIP @var{expression}
7748
@itemx CHIP @var{expression}, @var{expression}
7749
This command does nothing; it is accepted only for compatibility.
7750
 
7751
@cindex @code{END} (MRI)
7752
@item END
7753
This command does nothing whatever; it's only accepted for compatibility.
7754
 
7755
@cindex @code{FORMAT} (MRI)
7756
@item FORMAT @var{output-format}
7757
Similar to the @code{OUTPUT_FORMAT} command in the more general linker
7758
language, but restricted to one of these output formats:
7759
 
7760
@enumerate
7761
@item
7762
S-records, if @var{output-format} is @samp{S}
7763
 
7764
@item
7765
IEEE, if @var{output-format} is @samp{IEEE}
7766
 
7767
@item
7768
COFF (the @samp{coff-m68k} variant in BFD), if @var{output-format} is
7769
@samp{COFF}
7770
@end enumerate
7771
 
7772
@cindex @code{LIST} (MRI)
7773
@item LIST @var{anything}@dots{}
7774
Print (to the standard output file) a link map, as produced by the
7775
@command{ld} command-line option @samp{-M}.
7776
 
7777
The keyword @code{LIST} may be followed by anything on the
7778
same line, with no change in its effect.
7779
 
7780
@cindex @code{LOAD} (MRI)
7781
@item LOAD @var{filename}
7782
@itemx LOAD @var{filename}, @var{filename}, @dots{} @var{filename}
7783
Include one or more object file @var{filename} in the link; this has the
7784
same effect as specifying @var{filename} directly on the @command{ld}
7785
command line.
7786
 
7787
@cindex @code{NAME} (MRI)
7788
@item NAME @var{output-name}
7789
@var{output-name} is the name for the program produced by @command{ld}; the
7790
MRI-compatible command @code{NAME} is equivalent to the command-line
7791
option @samp{-o} or the general script language command @code{OUTPUT}.
7792
 
7793
@cindex @code{ORDER} (MRI)
7794
@item ORDER @var{secname}, @var{secname}, @dots{} @var{secname}
7795
@itemx ORDER @var{secname} @var{secname} @var{secname}
7796
Normally, @command{ld} orders the sections in its output file in the
7797
order in which they first appear in the input files.  In an MRI-compatible
7798
script, you can override this ordering with the @code{ORDER} command.  The
7799
sections you list with @code{ORDER} will appear first in your output
7800
file, in the order specified.
7801
 
7802
@cindex @code{PUBLIC} (MRI)
7803
@item PUBLIC @var{name}=@var{expression}
7804
@itemx PUBLIC @var{name},@var{expression}
7805
@itemx PUBLIC @var{name} @var{expression}
7806
Supply a value (@var{expression}) for external symbol
7807
@var{name} used in the linker input files.
7808
 
7809
@cindex @code{SECT} (MRI)
7810
@item SECT @var{secname}, @var{expression}
7811
@itemx SECT @var{secname}=@var{expression}
7812
@itemx SECT @var{secname} @var{expression}
7813
You can use any of these three forms of the @code{SECT} command to
7814
specify the start address (@var{expression}) for section @var{secname}.
7815
If you have more than one @code{SECT} statement for the same
7816
@var{secname}, only the @emph{first} sets the start address.
7817
@end table
7818
 
7819
@node GNU Free Documentation License
7820
@appendix GNU Free Documentation License
7821
@include fdl.texi
7822
 
7823
@node LD Index
7824
@unnumbered LD Index
7825
 
7826
@printindex cp
7827
 
7828
@tex
7829
% I think something like @colophon should be in texinfo.  In the
7830
% meantime:
7831
\long\def\colophon{\hbox to0pt{}\vfill
7832
\centerline{The body of this manual is set in}
7833
\centerline{\fontname\tenrm,}
7834
\centerline{with headings in {\bf\fontname\tenbf}}
7835
\centerline{and examples in {\tt\fontname\tentt}.}
7836
\centerline{{\it\fontname\tenit\/} and}
7837
\centerline{{\sl\fontname\tensl\/}}
7838
\centerline{are used for emphasis.}\vfill}
7839
\page\colophon
7840
% Blame: doc@cygnus.com, 28mar91.
7841
@end tex
7842
 
7843
@bye

powered by: WebSVN 2.1.0

© copyright 1999-2024 OpenCores.org, equivalent to Oliscience, all rights reserved. OpenCores®, registered trademark.