1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
3 @c 2001, 2002, 2003, 2004, 2005, 2006, 2007
4 @c Free Software Foundation, Inc.
5 @c UPDATE!! On future updates--
6 @c (1) check for new machine-dep cmdline options in
7 @c md_parse_option definitions in config/tc-*.c
8 @c (2) for platform-specific directives, examine md_pseudo_op
10 @c (3) for object-format specific directives, examine obj_pseudo_op
12 @c (4) portable directives in potable[] in read.c
16 @macro gcctabopt{body}
19 @c defaults, config file may override:
24 @include asconfig.texi
29 @c common OR combinations of conditions
52 @set abnormal-separator
56 @settitle Using @value{AS}
59 @settitle Using @value{AS} (@value{TARGET})
61 @setchapternewpage odd
66 @c WARE! Some of the machine-dependent sections contain tables of machine
67 @c instructions. Except in multi-column format, these tables look silly.
68 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
69 @c the multi-col format is faked within @example sections.
71 @c Again unfortunately, the natural size that fits on a page, for these tables,
72 @c is different depending on whether or not smallbook is turned on.
73 @c This matters, because of order: text flow switches columns at each page
76 @c The format faked in this source works reasonably well for smallbook,
77 @c not well for the default large-page format. This manual expects that if you
78 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
79 @c tables in question. You can turn on one without the other at your
80 @c discretion, of course.
83 @c the insn tables look just as silly in info files regardless of smallbook,
84 @c might as well show 'em anyways.
90 * As: (as). The GNU assembler.
91 * Gas: (as). The GNU assembler.
100 This file documents the GNU Assembler "@value{AS}".
102 @c man begin COPYRIGHT
103 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002,
104 2006, 2007 Free Software Foundation, Inc.
106 Permission is granted to copy, distribute and/or modify this document
107 under the terms of the GNU Free Documentation License, Version 1.1
108 or any later version published by the Free Software Foundation;
109 with no Invariant Sections, with no Front-Cover Texts, and with no
110 Back-Cover Texts. A copy of the license is included in the
111 section entitled ``GNU Free Documentation License''.
117 @title Using @value{AS}
118 @subtitle The @sc{gnu} Assembler
120 @subtitle for the @value{TARGET} family
122 @ifset VERSION_PACKAGE
124 @subtitle @value{VERSION_PACKAGE}
127 @subtitle Version @value{VERSION}
130 The Free Software Foundation Inc.@: thanks The Nice Computer
131 Company of Australia for loaning Dean Elsner to write the
132 first (Vax) version of @command{as} for Project @sc{gnu}.
133 The proprietors, management and staff of TNCCA thank FSF for
134 distracting the boss while they got some work
137 @author Dean Elsner, Jay Fenlason & friends
141 \hfill {\it Using {\tt @value{AS}}}\par
142 \hfill Edited by Cygnus Support\par
144 %"boxit" macro for figures:
145 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
146 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
147 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
148 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
149 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
152 @vskip 0pt plus 1filll
153 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002,
154 2006, 2007 Free Software Foundation, Inc.
156 Permission is granted to copy, distribute and/or modify this document
157 under the terms of the GNU Free Documentation License, Version 1.1
158 or any later version published by the Free Software Foundation;
159 with no Invariant Sections, with no Front-Cover Texts, and with no
160 Back-Cover Texts. A copy of the license is included in the
161 section entitled ``GNU Free Documentation License''.
168 @top Using @value{AS}
170 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
171 @ifset VERSION_PACKAGE
172 @value{VERSION_PACKAGE}
174 version @value{VERSION}.
176 This version of the file describes @command{@value{AS}} configured to generate
177 code for @value{TARGET} architectures.
180 This document is distributed under the terms of the GNU Free
181 Documentation License. A copy of the license is included in the
182 section entitled ``GNU Free Documentation License''.
185 * Overview:: Overview
186 * Invoking:: Command-Line Options
188 * Sections:: Sections and Relocation
190 * Expressions:: Expressions
191 * Pseudo Ops:: Assembler Directives
192 * Machine Dependencies:: Machine Dependent Features
193 * Reporting Bugs:: Reporting Bugs
194 * Acknowledgements:: Who Did What
195 * GNU Free Documentation License:: GNU Free Documentation License
196 * AS Index:: AS Index
203 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
205 This version of the manual describes @command{@value{AS}} configured to generate
206 code for @value{TARGET} architectures.
210 @cindex invocation summary
211 @cindex option summary
212 @cindex summary of options
213 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
214 see @ref{Invoking,,Command-Line Options}.
216 @c man title AS the portable GNU assembler.
220 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
224 @c We don't use deffn and friends for the following because they seem
225 @c to be limited to one line for the header.
227 @c man begin SYNOPSIS
228 @value{AS} [@b{-a}[@b{cdhlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
229 [@b{--debug-prefix-map} @var{old}=@var{new}]
230 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
231 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
232 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
233 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
234 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}] [@b{-o}
235 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
236 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
237 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{@@@var{FILE}}]
238 [@b{--target-help}] [@var{target-options}]
239 [@b{--}|@var{files} @dots{}]
241 @c Target dependent options are listed below. Keep the list sorted.
242 @c Add an empty line for separation.
245 @emph{Target Alpha options:}
247 [@b{-mdebug} | @b{-no-mdebug}]
248 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
249 [@b{-F}] [@b{-32addr}]
253 @emph{Target ARC options:}
259 @emph{Target ARM options:}
260 @c Don't document the deprecated options
261 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
262 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
263 [@b{-mfpu}=@var{floating-point-format}]
264 [@b{-mfloat-abi}=@var{abi}]
265 [@b{-meabi}=@var{ver}]
268 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
269 @b{-mapcs-reentrant}]
270 [@b{-mthumb-interwork}] [@b{-k}]
274 @emph{Target CRIS options:}
275 [@b{--underscore} | @b{--no-underscore}]
277 [@b{--emulation=criself} | @b{--emulation=crisaout}]
278 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
279 @c Deprecated -- deliberately not documented.
284 @emph{Target D10V options:}
289 @emph{Target D30V options:}
290 [@b{-O}|@b{-n}|@b{-N}]
293 @c Renesas family chips have no machine-dependent assembler options
296 @c HPPA has no machine-dependent assembler options (yet).
300 @emph{Target i386 options:}
301 [@b{--32}|@b{--64}] [@b{-n}]
302 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}]
306 @emph{Target i960 options:}
307 @c see md_parse_option in tc-i960.c
308 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
310 [@b{-b}] [@b{-no-relax}]
314 @emph{Target IA-64 options:}
315 [@b{-mconstant-gp}|@b{-mauto-pic}]
316 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
318 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
319 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
320 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
321 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
325 @emph{Target IP2K options:}
326 [@b{-mip2022}|@b{-mip2022ext}]
330 @emph{Target M32C options:}
331 [@b{-m32c}|@b{-m16c}]
335 @emph{Target M32R options:}
336 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
341 @emph{Target M680X0 options:}
342 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
346 @emph{Target M68HC11 options:}
347 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
348 [@b{-mshort}|@b{-mlong}]
349 [@b{-mshort-double}|@b{-mlong-double}]
350 [@b{--force-long-branches}] [@b{--short-branches}]
351 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
352 [@b{--print-opcodes}] [@b{--generate-example}]
356 @emph{Target MCORE options:}
357 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
358 [@b{-mcpu=[210|340]}]
362 @emph{Target MIPS options:}
363 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
364 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
365 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
366 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
367 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
368 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
369 [@b{-mips64}] [@b{-mips64r2}]
370 [@b{-construct-floats}] [@b{-no-construct-floats}]
371 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
372 [@b{-mfix7000}] [@b{-mno-fix7000}]
373 [@b{-mips16}] [@b{-no-mips16}]
374 [@b{-msmartmips}] [@b{-mno-smartmips}]
375 [@b{-mips3d}] [@b{-no-mips3d}]
376 [@b{-mdmx}] [@b{-no-mdmx}]
377 [@b{-mdsp}] [@b{-mno-dsp}]
378 [@b{-mdspr2}] [@b{-mno-dspr2}]
379 [@b{-mmt}] [@b{-mno-mt}]
380 [@b{-mdebug}] [@b{-no-mdebug}]
381 [@b{-mpdr}] [@b{-mno-pdr}]
385 @emph{Target MMIX options:}
386 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
387 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
388 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
389 [@b{--linker-allocated-gregs}]
393 @emph{Target PDP11 options:}
394 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
395 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
396 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
400 @emph{Target picoJava options:}
405 @emph{Target PowerPC options:}
406 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
407 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}|
408 @b{-mbooke32}|@b{-mbooke64}]
409 [@b{-mcom}|@b{-many}|@b{-maltivec}] [@b{-memb}]
410 [@b{-mregnames}|@b{-mno-regnames}]
411 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
412 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
413 [@b{-msolaris}|@b{-mno-solaris}]
417 @emph{Target SPARC options:}
418 @c The order here is important. See c-sparc.texi.
419 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
420 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
421 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
426 @emph{Target TIC54X options:}
427 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
428 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
433 @emph{Target Z80 options:}
434 [@b{-z80}] [@b{-r800}]
435 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
436 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
437 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
438 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
439 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
440 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
444 @c Z8000 has no machine-dependent assembler options
448 @emph{Target Xtensa options:}
449 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
450 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
451 [@b{--[no-]transform}]
452 [@b{--rename-section} @var{oldname}=@var{newname}]
460 @include at-file.texi
463 Turn on listings, in any of a variety of ways:
467 omit false conditionals
470 omit debugging directives
473 include high-level source
479 include macro expansions
482 omit forms processing
488 set the name of the listing file
491 You may combine these options; for example, use @samp{-aln} for assembly
492 listing without forms processing. The @samp{=file} option, if used, must be
493 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
496 Begin in alternate macro mode.
498 @xref{Altmacro,,@code{.altmacro}}.
502 Ignored. This option is accepted for script compatibility with calls to
505 @item --debug-prefix-map @var{old}=@var{new}
506 When assembling files in directory @file{@var{old}}, record debugging
507 information describing them as in @file{@var{new}} instead.
509 @item --defsym @var{sym}=@var{value}
510 Define the symbol @var{sym} to be @var{value} before assembling the input file.
511 @var{value} must be an integer constant. As in C, a leading @samp{0x}
512 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
513 value. The value of the symbol can be overridden inside a source file via the
514 use of a @code{.set} pseudo-op.
517 ``fast''---skip whitespace and comment preprocessing (assume source is
522 Generate debugging information for each assembler source line using whichever
523 debug format is preferred by the target. This currently means either STABS,
527 Generate stabs debugging information for each assembler line. This
528 may help debugging assembler code, if the debugger can handle it.
531 Generate stabs debugging information for each assembler line, with GNU
532 extensions that probably only gdb can handle, and that could make other
533 debuggers crash or refuse to read your program. This
534 may help debugging assembler code. Currently the only GNU extension is
535 the location of the current working directory at assembling time.
538 Generate DWARF2 debugging information for each assembler line. This
539 may help debugging assembler code, if the debugger can handle it. Note---this
540 option is only supported by some targets, not all of them.
543 Print a summary of the command line options and exit.
546 Print a summary of all target specific options and exit.
549 Add directory @var{dir} to the search list for @code{.include} directives.
552 Don't warn about signed overflow.
555 @ifclear DIFF-TBL-KLUGE
556 This option is accepted but has no effect on the @value{TARGET} family.
558 @ifset DIFF-TBL-KLUGE
559 Issue warnings when difference tables altered for long displacements.
564 Keep (in the symbol table) local symbols. These symbols start with
565 system-specific local label prefixes, typically @samp{.L} for ELF systems
566 or @samp{L} for traditional a.out systems.
571 @item --listing-lhs-width=@var{number}
572 Set the maximum width, in words, of the output data column for an assembler
573 listing to @var{number}.
575 @item --listing-lhs-width2=@var{number}
576 Set the maximum width, in words, of the output data column for continuation
577 lines in an assembler listing to @var{number}.
579 @item --listing-rhs-width=@var{number}
580 Set the maximum width of an input source line, as displayed in a listing, to
583 @item --listing-cont-lines=@var{number}
584 Set the maximum number of lines printed in a listing for a single line of input
587 @item -o @var{objfile}
588 Name the object-file output from @command{@value{AS}} @var{objfile}.
591 Fold the data section into the text section.
593 @kindex --hash-size=@var{number}
594 Set the default size of GAS's hash tables to a prime number close to
595 @var{number}. Increasing this value can reduce the length of time it takes the
596 assembler to perform its tasks, at the expense of increasing the assembler's
597 memory requirements. Similarly reducing this value can reduce the memory
598 requirements at the expense of speed.
600 @item --reduce-memory-overheads
601 This option reduces GAS's memory requirements, at the expense of making the
602 assembly processes slower. Currently this switch is a synonym for
603 @samp{--hash-size=4051}, but in the future it may have other effects as well.
606 Print the maximum space (in bytes) and total time (in seconds) used by
609 @item --strip-local-absolute
610 Remove local absolute symbols from the outgoing symbol table.
614 Print the @command{as} version.
617 Print the @command{as} version and exit.
621 Suppress warning messages.
623 @item --fatal-warnings
624 Treat warnings as errors.
627 Don't suppress warning messages or treat them as errors.
636 Generate an object file even after errors.
638 @item -- | @var{files} @dots{}
639 Standard input, or source files to assemble.
644 The following options are available when @value{AS} is configured for
649 This option selects the core processor variant.
651 Select either big-endian (-EB) or little-endian (-EL) output.
656 The following options are available when @value{AS} is configured for the ARM
660 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
661 Specify which ARM processor variant is the target.
662 @item -march=@var{architecture}[+@var{extension}@dots{}]
663 Specify which ARM architecture variant is used by the target.
664 @item -mfpu=@var{floating-point-format}
665 Select which Floating Point architecture is the target.
666 @item -mfloat-abi=@var{abi}
667 Select which floating point ABI is in use.
669 Enable Thumb only instruction decoding.
670 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
671 Select which procedure calling convention is in use.
673 Select either big-endian (-EB) or little-endian (-EL) output.
674 @item -mthumb-interwork
675 Specify that the code has been generated with interworking between Thumb and
678 Specify that PIC code has been generated.
683 See the info pages for documentation of the CRIS-specific options.
687 The following options are available when @value{AS} is configured for
690 @cindex D10V optimization
691 @cindex optimization, D10V
693 Optimize output by parallelizing instructions.
698 The following options are available when @value{AS} is configured for a D30V
701 @cindex D30V optimization
702 @cindex optimization, D30V
704 Optimize output by parallelizing instructions.
708 Warn when nops are generated.
710 @cindex D30V nops after 32-bit multiply
712 Warn when a nop after a 32-bit multiply instruction is generated.
717 The following options are available when @value{AS} is configured for the
718 Intel 80960 processor.
721 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
722 Specify which variant of the 960 architecture is the target.
725 Add code to collect statistics about branches taken.
728 Do not alter compare-and-branch instructions for long displacements;
735 The following options are available when @value{AS} is configured for the
741 Specifies that the extended IP2022 instructions are allowed.
744 Restores the default behaviour, which restricts the permitted instructions to
745 just the basic IP2022 ones.
751 The following options are available when @value{AS} is configured for the
752 Renesas M32C and M16C processors.
757 Assemble M32C instructions.
760 Assemble M16C instructions (the default).
766 The following options are available when @value{AS} is configured for the
767 Renesas M32R (formerly Mitsubishi M32R) series.
772 Specify which processor in the M32R family is the target. The default
773 is normally the M32R, but this option changes it to the M32RX.
775 @item --warn-explicit-parallel-conflicts or --Wp
776 Produce warning messages when questionable parallel constructs are
779 @item --no-warn-explicit-parallel-conflicts or --Wnp
780 Do not produce warning messages when questionable parallel constructs are
787 The following options are available when @value{AS} is configured for the
788 Motorola 68000 series.
793 Shorten references to undefined symbols, to one word instead of two.
795 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
796 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
797 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
798 Specify what processor in the 68000 family is the target. The default
799 is normally the 68020, but this can be changed at configuration time.
801 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
802 The target machine does (or does not) have a floating-point coprocessor.
803 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
804 the basic 68000 is not compatible with the 68881, a combination of the
805 two can be specified, since it's possible to do emulation of the
806 coprocessor instructions with the main processor.
808 @item -m68851 | -mno-68851
809 The target machine does (or does not) have a memory-management
810 unit coprocessor. The default is to assume an MMU for 68020 and up.
817 For details about the PDP-11 machine dependent features options,
818 see @ref{PDP-11-Options}.
821 @item -mpic | -mno-pic
822 Generate position-independent (or position-dependent) code. The
823 default is @option{-mpic}.
826 @itemx -mall-extensions
827 Enable all instruction set extensions. This is the default.
829 @item -mno-extensions
830 Disable all instruction set extensions.
832 @item -m@var{extension} | -mno-@var{extension}
833 Enable (or disable) a particular instruction set extension.
836 Enable the instruction set extensions supported by a particular CPU, and
837 disable all other extensions.
839 @item -m@var{machine}
840 Enable the instruction set extensions supported by a particular machine
841 model, and disable all other extensions.
847 The following options are available when @value{AS} is configured for
848 a picoJava processor.
852 @cindex PJ endianness
853 @cindex endianness, PJ
854 @cindex big endian output, PJ
856 Generate ``big endian'' format output.
858 @cindex little endian output, PJ
860 Generate ``little endian'' format output.
866 The following options are available when @value{AS} is configured for the
867 Motorola 68HC11 or 68HC12 series.
871 @item -m68hc11 | -m68hc12 | -m68hcs12
872 Specify what processor is the target. The default is
873 defined by the configuration option when building the assembler.
876 Specify to use the 16-bit integer ABI.
879 Specify to use the 32-bit integer ABI.
882 Specify to use the 32-bit double ABI.
885 Specify to use the 64-bit double ABI.
887 @item --force-long-branches
888 Relative branches are turned into absolute ones. This concerns
889 conditional branches, unconditional branches and branches to a
892 @item -S | --short-branches
893 Do not turn relative branches into absolute ones
894 when the offset is out of range.
896 @item --strict-direct-mode
897 Do not turn the direct addressing mode into extended addressing mode
898 when the instruction does not support direct addressing mode.
900 @item --print-insn-syntax
901 Print the syntax of instruction in case of error.
903 @item --print-opcodes
904 print the list of instructions with syntax and then exit.
906 @item --generate-example
907 print an example of instruction for each possible instruction and then exit.
908 This option is only useful for testing @command{@value{AS}}.
914 The following options are available when @command{@value{AS}} is configured
915 for the SPARC architecture:
918 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
919 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
920 Explicitly select a variant of the SPARC architecture.
922 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
923 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
925 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
926 UltraSPARC extensions.
928 @item -xarch=v8plus | -xarch=v8plusa
929 For compatibility with the Solaris v9 assembler. These options are
930 equivalent to -Av8plus and -Av8plusa, respectively.
933 Warn when the assembler switches to another architecture.
938 The following options are available when @value{AS} is configured for the 'c54x
943 Enable extended addressing mode. All addresses and relocations will assume
944 extended addressing (usually 23 bits).
945 @item -mcpu=@var{CPU_VERSION}
946 Sets the CPU version being compiled for.
947 @item -merrors-to-file @var{FILENAME}
948 Redirect error output to a file, for broken systems which don't support such
949 behaviour in the shell.
954 The following options are available when @value{AS} is configured for
955 a @sc{mips} processor.
959 This option sets the largest size of an object that can be referenced
960 implicitly with the @code{gp} register. It is only accepted for targets that
961 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
963 @cindex MIPS endianness
964 @cindex endianness, MIPS
965 @cindex big endian output, MIPS
967 Generate ``big endian'' format output.
969 @cindex little endian output, MIPS
971 Generate ``little endian'' format output.
983 Generate code for a particular @sc{mips} Instruction Set Architecture level.
984 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
985 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
986 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
987 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
989 correspond to generic
990 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
991 and @samp{MIPS64 Release 2}
992 ISA processors, respectively.
994 @item -march=@var{CPU}
995 Generate code for a particular @sc{mips} cpu.
997 @item -mtune=@var{cpu}
998 Schedule and tune for a particular @sc{mips} cpu.
1002 Cause nops to be inserted if the read of the destination register
1003 of an mfhi or mflo instruction occurs in the following two instructions.
1007 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1008 section instead of the standard ELF .stabs sections.
1012 Control generation of @code{.pdr} sections.
1016 The register sizes are normally inferred from the ISA and ABI, but these
1017 flags force a certain group of registers to be treated as 32 bits wide at
1018 all times. @samp{-mgp32} controls the size of general-purpose registers
1019 and @samp{-mfp32} controls the size of floating-point registers.
1023 Generate code for the MIPS 16 processor. This is equivalent to putting
1024 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1025 turns off this option.
1028 @itemx -mno-smartmips
1029 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1030 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1031 @samp{-mno-smartmips} turns off this option.
1035 Generate code for the MIPS-3D Application Specific Extension.
1036 This tells the assembler to accept MIPS-3D instructions.
1037 @samp{-no-mips3d} turns off this option.
1041 Generate code for the MDMX Application Specific Extension.
1042 This tells the assembler to accept MDMX instructions.
1043 @samp{-no-mdmx} turns off this option.
1047 Generate code for the DSP Release 1 Application Specific Extension.
1048 This tells the assembler to accept DSP Release 1 instructions.
1049 @samp{-mno-dsp} turns off this option.
1053 Generate code for the DSP Release 2 Application Specific Extension.
1054 This option implies -mdsp.
1055 This tells the assembler to accept DSP Release 2 instructions.
1056 @samp{-mno-dspr2} turns off this option.
1060 Generate code for the MT Application Specific Extension.
1061 This tells the assembler to accept MT instructions.
1062 @samp{-mno-mt} turns off this option.
1064 @item --construct-floats
1065 @itemx --no-construct-floats
1066 The @samp{--no-construct-floats} option disables the construction of
1067 double width floating point constants by loading the two halves of the
1068 value into the two single width floating point registers that make up
1069 the double width register. By default @samp{--construct-floats} is
1070 selected, allowing construction of these floating point constants.
1073 @item --emulation=@var{name}
1074 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1075 for some other target, in all respects, including output format (choosing
1076 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1077 debugging information or store symbol table information, and default
1078 endianness. The available configuration names are: @samp{mipsecoff},
1079 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1080 @samp{mipsbelf}. The first two do not alter the default endianness from that
1081 of the primary target for which the assembler was configured; the others change
1082 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1083 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1084 selection in any case.
1086 This option is currently supported only when the primary target
1087 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1088 Furthermore, the primary target or others specified with
1089 @samp{--enable-targets=@dots{}} at configuration time must include support for
1090 the other format, if both are to be available. For example, the Irix 5
1091 configuration includes support for both.
1093 Eventually, this option will support more configurations, with more
1094 fine-grained control over the assembler's behavior, and will be supported for
1098 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1105 Control how to deal with multiplication overflow and division by zero.
1106 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1107 (and only work for Instruction Set Architecture level 2 and higher);
1108 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1112 When this option is used, @command{@value{AS}} will issue a warning every
1113 time it generates a nop instruction from a macro.
1118 The following options are available when @value{AS} is configured for
1124 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1125 The command line option @samp{-nojsri2bsr} can be used to disable it.
1129 Enable or disable the silicon filter behaviour. By default this is disabled.
1130 The default can be overridden by the @samp{-sifilter} command line option.
1133 Alter jump instructions for long displacements.
1135 @item -mcpu=[210|340]
1136 Select the cpu type on the target hardware. This controls which instructions
1140 Assemble for a big endian target.
1143 Assemble for a little endian target.
1149 See the info pages for documentation of the MMIX-specific options.
1153 The following options are available when @value{AS} is configured for
1154 an Xtensa processor.
1157 @item --text-section-literals | --no-text-section-literals
1158 With @option{--text-@-section-@-literals}, literal pools are interspersed
1159 in the text section. The default is
1160 @option{--no-@-text-@-section-@-literals}, which places literals in a
1161 separate section in the output file. These options only affect literals
1162 referenced via PC-relative @code{L32R} instructions; literals for
1163 absolute mode @code{L32R} instructions are handled separately.
1165 @item --absolute-literals | --no-absolute-literals
1166 Indicate to the assembler whether @code{L32R} instructions use absolute
1167 or PC-relative addressing. The default is to assume absolute addressing
1168 if the Xtensa processor includes the absolute @code{L32R} addressing
1169 option. Otherwise, only the PC-relative @code{L32R} mode can be used.
1171 @item --target-align | --no-target-align
1172 Enable or disable automatic alignment to reduce branch penalties at the
1173 expense of some code density. The default is @option{--target-@-align}.
1175 @item --longcalls | --no-longcalls
1176 Enable or disable transformation of call instructions to allow calls
1177 across a greater range of addresses. The default is
1178 @option{--no-@-longcalls}.
1180 @item --transform | --no-transform
1181 Enable or disable all assembler transformations of Xtensa instructions.
1182 The default is @option{--transform};
1183 @option{--no-transform} should be used only in the rare cases when the
1184 instructions must be exactly as specified in the assembly source.
1189 The following options are available when @value{AS} is configured for
1190 a Z80 family processor.
1193 Assemble for Z80 processor.
1195 Assemble for R800 processor.
1196 @item -ignore-undocumented-instructions
1198 Assemble undocumented Z80 instructions that also work on R800 without warning.
1199 @item -ignore-unportable-instructions
1201 Assemble all undocumented Z80 instructions without warning.
1202 @item -warn-undocumented-instructions
1204 Issue a warning for undocumented Z80 instructions that also work on R800.
1205 @item -warn-unportable-instructions
1207 Issue a warning for undocumented Z80 instructions that do not work on R800.
1208 @item -forbid-undocumented-instructions
1210 Treat all undocumented instructions as errors.
1211 @item -forbid-unportable-instructions
1213 Treat undocumented Z80 instructions that do not work on R800 as errors.
1220 * Manual:: Structure of this Manual
1221 * GNU Assembler:: The GNU Assembler
1222 * Object Formats:: Object File Formats
1223 * Command Line:: Command Line
1224 * Input Files:: Input Files
1225 * Object:: Output (Object) File
1226 * Errors:: Error and Warning Messages
1230 @section Structure of this Manual
1232 @cindex manual, structure and purpose
1233 This manual is intended to describe what you need to know to use
1234 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1235 notation for symbols, constants, and expressions; the directives that
1236 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1239 We also cover special features in the @value{TARGET}
1240 configuration of @command{@value{AS}}, including assembler directives.
1243 This manual also describes some of the machine-dependent features of
1244 various flavors of the assembler.
1247 @cindex machine instructions (not covered)
1248 On the other hand, this manual is @emph{not} intended as an introduction
1249 to programming in assembly language---let alone programming in general!
1250 In a similar vein, we make no attempt to introduce the machine
1251 architecture; we do @emph{not} describe the instruction set, standard
1252 mnemonics, registers or addressing modes that are standard to a
1253 particular architecture.
1255 You may want to consult the manufacturer's
1256 machine architecture manual for this information.
1260 For information on the H8/300 machine instruction set, see @cite{H8/300
1261 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1262 Programming Manual} (Renesas).
1265 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1266 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1267 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1268 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1271 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1275 @c I think this is premature---doc@cygnus.com, 17jan1991
1277 Throughout this manual, we assume that you are running @dfn{GNU},
1278 the portable operating system from the @dfn{Free Software
1279 Foundation, Inc.}. This restricts our attention to certain kinds of
1280 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1281 once this assumption is granted examples and definitions need less
1284 @command{@value{AS}} is part of a team of programs that turn a high-level
1285 human-readable series of instructions into a low-level
1286 computer-readable series of instructions. Different versions of
1287 @command{@value{AS}} are used for different kinds of computer.
1290 @c There used to be a section "Terminology" here, which defined
1291 @c "contents", "byte", "word", and "long". Defining "word" to any
1292 @c particular size is confusing when the .word directive may generate 16
1293 @c bits on one machine and 32 bits on another; in general, for the user
1294 @c version of this manual, none of these terms seem essential to define.
1295 @c They were used very little even in the former draft of the manual;
1296 @c this draft makes an effort to avoid them (except in names of
1300 @section The GNU Assembler
1302 @c man begin DESCRIPTION
1304 @sc{gnu} @command{as} is really a family of assemblers.
1306 This manual describes @command{@value{AS}}, a member of that family which is
1307 configured for the @value{TARGET} architectures.
1309 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1310 should find a fairly similar environment when you use it on another
1311 architecture. Each version has much in common with the others,
1312 including object file formats, most assembler directives (often called
1313 @dfn{pseudo-ops}) and assembler syntax.@refill
1315 @cindex purpose of @sc{gnu} assembler
1316 @command{@value{AS}} is primarily intended to assemble the output of the
1317 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1318 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1319 assemble correctly everything that other assemblers for the same
1320 machine would assemble.
1322 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1325 @c This remark should appear in generic version of manual; assumption
1326 @c here is that generic version sets M680x0.
1327 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1328 assembler for the same architecture; for example, we know of several
1329 incompatible versions of 680x0 assembly language syntax.
1334 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1335 program in one pass of the source file. This has a subtle impact on the
1336 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1338 @node Object Formats
1339 @section Object File Formats
1341 @cindex object file format
1342 The @sc{gnu} assembler can be configured to produce several alternative
1343 object file formats. For the most part, this does not affect how you
1344 write assembly language programs; but directives for debugging symbols
1345 are typically different in different file formats. @xref{Symbol
1346 Attributes,,Symbol Attributes}.
1349 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1350 @value{OBJ-NAME} format object files.
1352 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1354 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1355 @code{b.out} or COFF format object files.
1358 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1359 SOM or ELF format object files.
1364 @section Command Line
1366 @cindex command line conventions
1368 After the program name @command{@value{AS}}, the command line may contain
1369 options and file names. Options may appear in any order, and may be
1370 before, after, or between file names. The order of file names is
1373 @cindex standard input, as input file
1375 @file{--} (two hyphens) by itself names the standard input file
1376 explicitly, as one of the files for @command{@value{AS}} to assemble.
1378 @cindex options, command line
1379 Except for @samp{--} any command line argument that begins with a
1380 hyphen (@samp{-}) is an option. Each option changes the behavior of
1381 @command{@value{AS}}. No option changes the way another option works. An
1382 option is a @samp{-} followed by one or more letters; the case of
1383 the letter is important. All options are optional.
1385 Some options expect exactly one file name to follow them. The file
1386 name may either immediately follow the option's letter (compatible
1387 with older assemblers) or it may be the next command argument (@sc{gnu}
1388 standard). These two command lines are equivalent:
1391 @value{AS} -o my-object-file.o mumble.s
1392 @value{AS} -omy-object-file.o mumble.s
1396 @section Input Files
1399 @cindex source program
1400 @cindex files, input
1401 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1402 describe the program input to one run of @command{@value{AS}}. The program may
1403 be in one or more files; how the source is partitioned into files
1404 doesn't change the meaning of the source.
1406 @c I added "con" prefix to "catenation" just to prove I can overcome my
1407 @c APL training... doc@cygnus.com
1408 The source program is a concatenation of the text in all the files, in the
1411 @c man begin DESCRIPTION
1412 Each time you run @command{@value{AS}} it assembles exactly one source
1413 program. The source program is made up of one or more files.
1414 (The standard input is also a file.)
1416 You give @command{@value{AS}} a command line that has zero or more input file
1417 names. The input files are read (from left file name to right). A
1418 command line argument (in any position) that has no special meaning
1419 is taken to be an input file name.
1421 If you give @command{@value{AS}} no file names it attempts to read one input file
1422 from the @command{@value{AS}} standard input, which is normally your terminal. You
1423 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1426 Use @samp{--} if you need to explicitly name the standard input file
1427 in your command line.
1429 If the source is empty, @command{@value{AS}} produces a small, empty object
1434 @subheading Filenames and Line-numbers
1436 @cindex input file linenumbers
1437 @cindex line numbers, in input files
1438 There are two ways of locating a line in the input file (or files) and
1439 either may be used in reporting error messages. One way refers to a line
1440 number in a physical file; the other refers to a line number in a
1441 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1443 @dfn{Physical files} are those files named in the command line given
1444 to @command{@value{AS}}.
1446 @dfn{Logical files} are simply names declared explicitly by assembler
1447 directives; they bear no relation to physical files. Logical file names help
1448 error messages reflect the original source file, when @command{@value{AS}} source
1449 is itself synthesized from other files. @command{@value{AS}} understands the
1450 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1451 @ref{File,,@code{.file}}.
1454 @section Output (Object) File
1460 Every time you run @command{@value{AS}} it produces an output file, which is
1461 your assembly language program translated into numbers. This file
1462 is the object file. Its default name is
1470 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1472 You can give it another name by using the @option{-o} option. Conventionally,
1473 object file names end with @file{.o}. The default name is used for historical
1474 reasons: older assemblers were capable of assembling self-contained programs
1475 directly into a runnable program. (For some formats, this isn't currently
1476 possible, but it can be done for the @code{a.out} format.)
1480 The object file is meant for input to the linker @code{@value{LD}}. It contains
1481 assembled program code, information to help @code{@value{LD}} integrate
1482 the assembled program into a runnable file, and (optionally) symbolic
1483 information for the debugger.
1485 @c link above to some info file(s) like the description of a.out.
1486 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1489 @section Error and Warning Messages
1491 @c man begin DESCRIPTION
1493 @cindex error messages
1494 @cindex warning messages
1495 @cindex messages from assembler
1496 @command{@value{AS}} may write warnings and error messages to the standard error
1497 file (usually your terminal). This should not happen when a compiler
1498 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1499 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1500 grave problem that stops the assembly.
1504 @cindex format of warning messages
1505 Warning messages have the format
1508 file_name:@b{NNN}:Warning Message Text
1512 @cindex line numbers, in warnings/errors
1513 (where @b{NNN} is a line number). If a logical file name has been given
1514 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1515 the current input file is used. If a logical line number was given
1517 (@pxref{Line,,@code{.line}})
1519 then it is used to calculate the number printed,
1520 otherwise the actual line in the current source file is printed. The
1521 message text is intended to be self explanatory (in the grand Unix
1524 @cindex format of error messages
1525 Error messages have the format
1527 file_name:@b{NNN}:FATAL:Error Message Text
1529 The file name and line number are derived as for warning
1530 messages. The actual message text may be rather less explanatory
1531 because many of them aren't supposed to happen.
1534 @chapter Command-Line Options
1536 @cindex options, all versions of assembler
1537 This chapter describes command-line options available in @emph{all}
1538 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1539 for options specific
1541 to the @value{TARGET} target.
1544 to particular machine architectures.
1547 @c man begin DESCRIPTION
1549 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1550 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1551 The assembler arguments must be separated from each other (and the @samp{-Wa})
1552 by commas. For example:
1555 gcc -c -g -O -Wa,-alh,-L file.c
1559 This passes two options to the assembler: @samp{-alh} (emit a listing to
1560 standard output with high-level and assembly source) and @samp{-L} (retain
1561 local symbols in the symbol table).
1563 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1564 command-line options are automatically passed to the assembler by the compiler.
1565 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1566 precisely what options it passes to each compilation pass, including the
1572 * a:: -a[cdhlns] enable listings
1573 * alternate:: --alternate enable alternate macro syntax
1574 * D:: -D for compatibility
1575 * f:: -f to work faster
1576 * I:: -I for .include search path
1577 @ifclear DIFF-TBL-KLUGE
1578 * K:: -K for compatibility
1580 @ifset DIFF-TBL-KLUGE
1581 * K:: -K for difference tables
1584 * L:: -L to retain local symbols
1585 * listing:: --listing-XXX to configure listing output
1586 * M:: -M or --mri to assemble in MRI compatibility mode
1587 * MD:: --MD for dependency tracking
1588 * o:: -o to name the object file
1589 * R:: -R to join data and text sections
1590 * statistics:: --statistics to see statistics about assembly
1591 * traditional-format:: --traditional-format for compatible output
1592 * v:: -v to announce version
1593 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1594 * Z:: -Z to make object file even after errors
1598 @section Enable Listings: @option{-a[cdhlns]}
1607 @cindex listings, enabling
1608 @cindex assembly listings, enabling
1610 These options enable listing output from the assembler. By itself,
1611 @samp{-a} requests high-level, assembly, and symbols listing.
1612 You can use other letters to select specific options for the list:
1613 @samp{-ah} requests a high-level language listing,
1614 @samp{-al} requests an output-program assembly listing, and
1615 @samp{-as} requests a symbol table listing.
1616 High-level listings require that a compiler debugging option like
1617 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1620 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1621 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1622 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1623 omitted from the listing.
1625 Use the @samp{-ad} option to omit debugging directives from the
1628 Once you have specified one of these options, you can further control
1629 listing output and its appearance using the directives @code{.list},
1630 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1632 The @samp{-an} option turns off all forms processing.
1633 If you do not request listing output with one of the @samp{-a} options, the
1634 listing-control directives have no effect.
1636 The letters after @samp{-a} may be combined into one option,
1637 @emph{e.g.}, @samp{-aln}.
1639 Note if the assembler source is coming from the standard input (e.g.,
1641 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1642 is being used) then the listing will not contain any comments or preprocessor
1643 directives. This is because the listing code buffers input source lines from
1644 stdin only after they have been preprocessed by the assembler. This reduces
1645 memory usage and makes the code more efficient.
1648 @section @option{--alternate}
1651 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1654 @section @option{-D}
1657 This option has no effect whatsoever, but it is accepted to make it more
1658 likely that scripts written for other assemblers also work with
1659 @command{@value{AS}}.
1662 @section Work Faster: @option{-f}
1665 @cindex trusted compiler
1666 @cindex faster processing (@option{-f})
1667 @samp{-f} should only be used when assembling programs written by a
1668 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1669 and comment preprocessing on
1670 the input file(s) before assembling them. @xref{Preprocessing,
1674 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1675 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1680 @section @code{.include} Search Path: @option{-I} @var{path}
1682 @kindex -I @var{path}
1683 @cindex paths for @code{.include}
1684 @cindex search path for @code{.include}
1685 @cindex @code{include} directive search path
1686 Use this option to add a @var{path} to the list of directories
1687 @command{@value{AS}} searches for files specified in @code{.include}
1688 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1689 many times as necessary to include a variety of paths. The current
1690 working directory is always searched first; after that, @command{@value{AS}}
1691 searches any @samp{-I} directories in the same order as they were
1692 specified (left to right) on the command line.
1695 @section Difference Tables: @option{-K}
1698 @ifclear DIFF-TBL-KLUGE
1699 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1700 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1701 where it can be used to warn when the assembler alters the machine code
1702 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1703 family does not have the addressing limitations that sometimes lead to this
1704 alteration on other platforms.
1707 @ifset DIFF-TBL-KLUGE
1708 @cindex difference tables, warning
1709 @cindex warning for altered difference tables
1710 @command{@value{AS}} sometimes alters the code emitted for directives of the
1711 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
1712 You can use the @samp{-K} option if you want a warning issued when this
1717 @section Include Local Symbols: @option{-L}
1720 @cindex local symbols, retaining in output
1721 Symbols beginning with system-specific local label prefixes, typically
1722 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
1723 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
1724 such symbols when debugging, because they are intended for the use of
1725 programs (like compilers) that compose assembler programs, not for your
1726 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
1727 such symbols, so you do not normally debug with them.
1729 This option tells @command{@value{AS}} to retain those local symbols
1730 in the object file. Usually if you do this you also tell the linker
1731 @code{@value{LD}} to preserve those symbols.
1734 @section Configuring listing output: @option{--listing}
1736 The listing feature of the assembler can be enabled via the command line switch
1737 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1738 hex dump of the corresponding locations in the output object file, and displays
1739 them as a listing file. The format of this listing can be controlled by
1740 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
1741 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
1742 @code{.psize} (@pxref{Psize}), and
1743 @code{.eject} (@pxref{Eject}) and also by the following switches:
1746 @item --listing-lhs-width=@samp{number}
1747 @kindex --listing-lhs-width
1748 @cindex Width of first line disassembly output
1749 Sets the maximum width, in words, of the first line of the hex byte dump. This
1750 dump appears on the left hand side of the listing output.
1752 @item --listing-lhs-width2=@samp{number}
1753 @kindex --listing-lhs-width2
1754 @cindex Width of continuation lines of disassembly output
1755 Sets the maximum width, in words, of any further lines of the hex byte dump for
1756 a given input source line. If this value is not specified, it defaults to being
1757 the same as the value specified for @samp{--listing-lhs-width}. If neither
1758 switch is used the default is to one.
1760 @item --listing-rhs-width=@samp{number}
1761 @kindex --listing-rhs-width
1762 @cindex Width of source line output
1763 Sets the maximum width, in characters, of the source line that is displayed
1764 alongside the hex dump. The default value for this parameter is 100. The
1765 source line is displayed on the right hand side of the listing output.
1767 @item --listing-cont-lines=@samp{number}
1768 @kindex --listing-cont-lines
1769 @cindex Maximum number of continuation lines
1770 Sets the maximum number of continuation lines of hex dump that will be
1771 displayed for a given single line of source input. The default value is 4.
1775 @section Assemble in MRI Compatibility Mode: @option{-M}
1778 @cindex MRI compatibility mode
1779 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1780 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1781 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1782 configured target) assembler from Microtec Research. The exact nature of the
1783 MRI syntax will not be documented here; see the MRI manuals for more
1784 information. Note in particular that the handling of macros and macro
1785 arguments is somewhat different. The purpose of this option is to permit
1786 assembling existing MRI assembler code using @command{@value{AS}}.
1788 The MRI compatibility is not complete. Certain operations of the MRI assembler
1789 depend upon its object file format, and can not be supported using other object
1790 file formats. Supporting these would require enhancing each object file format
1791 individually. These are:
1794 @item global symbols in common section
1796 The m68k MRI assembler supports common sections which are merged by the linker.
1797 Other object file formats do not support this. @command{@value{AS}} handles
1798 common sections by treating them as a single common symbol. It permits local
1799 symbols to be defined within a common section, but it can not support global
1800 symbols, since it has no way to describe them.
1802 @item complex relocations
1804 The MRI assemblers support relocations against a negated section address, and
1805 relocations which combine the start addresses of two or more sections. These
1806 are not support by other object file formats.
1808 @item @code{END} pseudo-op specifying start address
1810 The MRI @code{END} pseudo-op permits the specification of a start address.
1811 This is not supported by other object file formats. The start address may
1812 instead be specified using the @option{-e} option to the linker, or in a linker
1815 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1817 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1818 name to the output file. This is not supported by other object file formats.
1820 @item @code{ORG} pseudo-op
1822 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1823 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1824 which changes the location within the current section. Absolute sections are
1825 not supported by other object file formats. The address of a section may be
1826 assigned within a linker script.
1829 There are some other features of the MRI assembler which are not supported by
1830 @command{@value{AS}}, typically either because they are difficult or because they
1831 seem of little consequence. Some of these may be supported in future releases.
1835 @item EBCDIC strings
1837 EBCDIC strings are not supported.
1839 @item packed binary coded decimal
1841 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1842 and @code{DCB.P} pseudo-ops are not supported.
1844 @item @code{FEQU} pseudo-op
1846 The m68k @code{FEQU} pseudo-op is not supported.
1848 @item @code{NOOBJ} pseudo-op
1850 The m68k @code{NOOBJ} pseudo-op is not supported.
1852 @item @code{OPT} branch control options
1854 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1855 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1856 relaxes all branches, whether forward or backward, to an appropriate size, so
1857 these options serve no purpose.
1859 @item @code{OPT} list control options
1861 The following m68k @code{OPT} list control options are ignored: @code{C},
1862 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1863 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1865 @item other @code{OPT} options
1867 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1868 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1870 @item @code{OPT} @code{D} option is default
1872 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1873 @code{OPT NOD} may be used to turn it off.
1875 @item @code{XREF} pseudo-op.
1877 The m68k @code{XREF} pseudo-op is ignored.
1879 @item @code{.debug} pseudo-op
1881 The i960 @code{.debug} pseudo-op is not supported.
1883 @item @code{.extended} pseudo-op
1885 The i960 @code{.extended} pseudo-op is not supported.
1887 @item @code{.list} pseudo-op.
1889 The various options of the i960 @code{.list} pseudo-op are not supported.
1891 @item @code{.optimize} pseudo-op
1893 The i960 @code{.optimize} pseudo-op is not supported.
1895 @item @code{.output} pseudo-op
1897 The i960 @code{.output} pseudo-op is not supported.
1899 @item @code{.setreal} pseudo-op
1901 The i960 @code{.setreal} pseudo-op is not supported.
1906 @section Dependency Tracking: @option{--MD}
1909 @cindex dependency tracking
1912 @command{@value{AS}} can generate a dependency file for the file it creates. This
1913 file consists of a single rule suitable for @code{make} describing the
1914 dependencies of the main source file.
1916 The rule is written to the file named in its argument.
1918 This feature is used in the automatic updating of makefiles.
1921 @section Name the Object File: @option{-o}
1924 @cindex naming object file
1925 @cindex object file name
1926 There is always one object file output when you run @command{@value{AS}}. By
1927 default it has the name
1930 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1944 You use this option (which takes exactly one filename) to give the
1945 object file a different name.
1947 Whatever the object file is called, @command{@value{AS}} overwrites any
1948 existing file of the same name.
1951 @section Join Data and Text Sections: @option{-R}
1954 @cindex data and text sections, joining
1955 @cindex text and data sections, joining
1956 @cindex joining text and data sections
1957 @cindex merging text and data sections
1958 @option{-R} tells @command{@value{AS}} to write the object file as if all
1959 data-section data lives in the text section. This is only done at
1960 the very last moment: your binary data are the same, but data
1961 section parts are relocated differently. The data section part of
1962 your object file is zero bytes long because all its bytes are
1963 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1965 When you specify @option{-R} it would be possible to generate shorter
1966 address displacements (because we do not have to cross between text and
1967 data section). We refrain from doing this simply for compatibility with
1968 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
1971 When @command{@value{AS}} is configured for COFF or ELF output,
1972 this option is only useful if you use sections named @samp{.text} and
1977 @option{-R} is not supported for any of the HPPA targets. Using
1978 @option{-R} generates a warning from @command{@value{AS}}.
1982 @section Display Assembly Statistics: @option{--statistics}
1984 @kindex --statistics
1985 @cindex statistics, about assembly
1986 @cindex time, total for assembly
1987 @cindex space used, maximum for assembly
1988 Use @samp{--statistics} to display two statistics about the resources used by
1989 @command{@value{AS}}: the maximum amount of space allocated during the assembly
1990 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1993 @node traditional-format
1994 @section Compatible Output: @option{--traditional-format}
1996 @kindex --traditional-format
1997 For some targets, the output of @command{@value{AS}} is different in some ways
1998 from the output of some existing assembler. This switch requests
1999 @command{@value{AS}} to use the traditional format instead.
2001 For example, it disables the exception frame optimizations which
2002 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2005 @section Announce Version: @option{-v}
2009 @cindex assembler version
2010 @cindex version of assembler
2011 You can find out what version of as is running by including the
2012 option @samp{-v} (which you can also spell as @samp{-version}) on the
2016 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2018 @command{@value{AS}} should never give a warning or error message when
2019 assembling compiler output. But programs written by people often
2020 cause @command{@value{AS}} to give a warning that a particular assumption was
2021 made. All such warnings are directed to the standard error file.
2025 @cindex suppressing warnings
2026 @cindex warnings, suppressing
2027 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2028 This only affects the warning messages: it does not change any particular of
2029 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2032 @kindex --fatal-warnings
2033 @cindex errors, caused by warnings
2034 @cindex warnings, causing error
2035 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2036 files that generate warnings to be in error.
2039 @cindex warnings, switching on
2040 You can switch these options off again by specifying @option{--warn}, which
2041 causes warnings to be output as usual.
2044 @section Generate Object File in Spite of Errors: @option{-Z}
2045 @cindex object file, after errors
2046 @cindex errors, continuing after
2047 After an error message, @command{@value{AS}} normally produces no output. If for
2048 some reason you are interested in object file output even after
2049 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2050 option. If there are any errors, @command{@value{AS}} continues anyways, and
2051 writes an object file after a final warning message of the form @samp{@var{n}
2052 errors, @var{m} warnings, generating bad object file.}
2057 @cindex machine-independent syntax
2058 @cindex syntax, machine-independent
2059 This chapter describes the machine-independent syntax allowed in a
2060 source file. @command{@value{AS}} syntax is similar to what many other
2061 assemblers use; it is inspired by the BSD 4.2
2066 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2070 * Preprocessing:: Preprocessing
2071 * Whitespace:: Whitespace
2072 * Comments:: Comments
2073 * Symbol Intro:: Symbols
2074 * Statements:: Statements
2075 * Constants:: Constants
2079 @section Preprocessing
2081 @cindex preprocessing
2082 The @command{@value{AS}} internal preprocessor:
2084 @cindex whitespace, removed by preprocessor
2086 adjusts and removes extra whitespace. It leaves one space or tab before
2087 the keywords on a line, and turns any other whitespace on the line into
2090 @cindex comments, removed by preprocessor
2092 removes all comments, replacing them with a single space, or an
2093 appropriate number of newlines.
2095 @cindex constants, converted by preprocessor
2097 converts character constants into the appropriate numeric values.
2100 It does not do macro processing, include file handling, or
2101 anything else you may get from your C compiler's preprocessor. You can
2102 do include file processing with the @code{.include} directive
2103 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2104 to get other ``CPP'' style preprocessing by giving the input file a
2105 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2106 Output, gcc.info, Using GNU CC}.
2108 Excess whitespace, comments, and character constants
2109 cannot be used in the portions of the input text that are not
2112 @cindex turning preprocessing on and off
2113 @cindex preprocessing, turning on and off
2116 If the first line of an input file is @code{#NO_APP} or if you use the
2117 @samp{-f} option, whitespace and comments are not removed from the input file.
2118 Within an input file, you can ask for whitespace and comment removal in
2119 specific portions of the by putting a line that says @code{#APP} before the
2120 text that may contain whitespace or comments, and putting a line that says
2121 @code{#NO_APP} after this text. This feature is mainly intend to support
2122 @code{asm} statements in compilers whose output is otherwise free of comments
2129 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2130 Whitespace is used to separate symbols, and to make programs neater for
2131 people to read. Unless within character constants
2132 (@pxref{Characters,,Character Constants}), any whitespace means the same
2133 as exactly one space.
2139 There are two ways of rendering comments to @command{@value{AS}}. In both
2140 cases the comment is equivalent to one space.
2142 Anything from @samp{/*} through the next @samp{*/} is a comment.
2143 This means you may not nest these comments.
2147 The only way to include a newline ('\n') in a comment
2148 is to use this sort of comment.
2151 /* This sort of comment does not nest. */
2154 @cindex line comment character
2155 Anything from the @dfn{line comment} character to the next newline
2156 is considered a comment and is ignored. The line comment character is
2158 @samp{;} on the ARC;
2161 @samp{@@} on the ARM;
2164 @samp{;} for the H8/300 family;
2167 @samp{;} for the HPPA;
2170 @samp{#} on the i386 and x86-64;
2173 @samp{#} on the i960;
2176 @samp{;} for the PDP-11;
2179 @samp{;} for picoJava;
2182 @samp{#} for Motorola PowerPC;
2185 @samp{!} for the Renesas / SuperH SH;
2188 @samp{!} on the SPARC;
2191 @samp{#} on the ip2k;
2194 @samp{#} on the m32c;
2197 @samp{#} on the m32r;
2200 @samp{|} on the 680x0;
2203 @samp{#} on the 68HC11 and 68HC12;
2206 @samp{#} on the Vax;
2209 @samp{;} for the Z80;
2212 @samp{!} for the Z8000;
2215 @samp{#} on the V850;
2218 @samp{#} for Xtensa systems;
2220 see @ref{Machine Dependencies}. @refill
2221 @c FIXME What about i860?
2224 On some machines there are two different line comment characters. One
2225 character only begins a comment if it is the first non-whitespace character on
2226 a line, while the other always begins a comment.
2230 The V850 assembler also supports a double dash as starting a comment that
2231 extends to the end of the line.
2237 @cindex lines starting with @code{#}
2238 @cindex logical line numbers
2239 To be compatible with past assemblers, lines that begin with @samp{#} have a
2240 special interpretation. Following the @samp{#} should be an absolute
2241 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2242 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2243 new logical file name. The rest of the line, if any, should be whitespace.
2245 If the first non-whitespace characters on the line are not numeric,
2246 the line is ignored. (Just like a comment.)
2249 # This is an ordinary comment.
2250 # 42-6 "new_file_name" # New logical file name
2251 # This is logical line # 36.
2253 This feature is deprecated, and may disappear from future versions
2254 of @command{@value{AS}}.
2259 @cindex characters used in symbols
2260 @ifclear SPECIAL-SYMS
2261 A @dfn{symbol} is one or more characters chosen from the set of all
2262 letters (both upper and lower case), digits and the three characters
2268 A @dfn{symbol} is one or more characters chosen from the set of all
2269 letters (both upper and lower case), digits and the three characters
2270 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2276 On most machines, you can also use @code{$} in symbol names; exceptions
2277 are noted in @ref{Machine Dependencies}.
2279 No symbol may begin with a digit. Case is significant.
2280 There is no length limit: all characters are significant. Symbols are
2281 delimited by characters not in that set, or by the beginning of a file
2282 (since the source program must end with a newline, the end of a file is
2283 not a possible symbol delimiter). @xref{Symbols}.
2284 @cindex length of symbols
2289 @cindex statements, structure of
2290 @cindex line separator character
2291 @cindex statement separator character
2293 @ifclear abnormal-separator
2294 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2295 semicolon (@samp{;}). The newline or semicolon is considered part of
2296 the preceding statement. Newlines and semicolons within character
2297 constants are an exception: they do not end statements.
2299 @ifset abnormal-separator
2301 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2302 point (@samp{!}). The newline or exclamation point is considered part of the
2303 preceding statement. Newlines and exclamation points within character
2304 constants are an exception: they do not end statements.
2307 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2308 H8/300) a dollar sign (@samp{$}); or (for the Renesas-SH) a semicolon
2309 (@samp{;}). The newline or separator character is considered part of
2310 the preceding statement. Newlines and separators within character
2311 constants are an exception: they do not end statements.
2316 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2317 separator character. (The line separator is usually @samp{;}, unless this
2318 conflicts with the comment character; see @ref{Machine Dependencies}.) The
2319 newline or separator character is considered part of the preceding
2320 statement. Newlines and separators within character constants are an
2321 exception: they do not end statements.
2324 @cindex newline, required at file end
2325 @cindex EOF, newline must precede
2326 It is an error to end any statement with end-of-file: the last
2327 character of any input file should be a newline.@refill
2329 An empty statement is allowed, and may include whitespace. It is ignored.
2331 @cindex instructions and directives
2332 @cindex directives and instructions
2333 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2334 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2336 A statement begins with zero or more labels, optionally followed by a
2337 key symbol which determines what kind of statement it is. The key
2338 symbol determines the syntax of the rest of the statement. If the
2339 symbol begins with a dot @samp{.} then the statement is an assembler
2340 directive: typically valid for any computer. If the symbol begins with
2341 a letter the statement is an assembly language @dfn{instruction}: it
2342 assembles into a machine language instruction.
2344 Different versions of @command{@value{AS}} for different computers
2345 recognize different instructions. In fact, the same symbol may
2346 represent a different instruction in a different computer's assembly
2350 @cindex @code{:} (label)
2351 @cindex label (@code{:})
2352 A label is a symbol immediately followed by a colon (@code{:}).
2353 Whitespace before a label or after a colon is permitted, but you may not
2354 have whitespace between a label's symbol and its colon. @xref{Labels}.
2357 For HPPA targets, labels need not be immediately followed by a colon, but
2358 the definition of a label must begin in column zero. This also implies that
2359 only one label may be defined on each line.
2363 label: .directive followed by something
2364 another_label: # This is an empty statement.
2365 instruction operand_1, operand_2, @dots{}
2372 A constant is a number, written so that its value is known by
2373 inspection, without knowing any context. Like this:
2376 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2377 .ascii "Ring the bell\7" # A string constant.
2378 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2379 .float 0f-314159265358979323846264338327\
2380 95028841971.693993751E-40 # - pi, a flonum.
2385 * Characters:: Character Constants
2386 * Numbers:: Number Constants
2390 @subsection Character Constants
2392 @cindex character constants
2393 @cindex constants, character
2394 There are two kinds of character constants. A @dfn{character} stands
2395 for one character in one byte and its value may be used in
2396 numeric expressions. String constants (properly called string
2397 @emph{literals}) are potentially many bytes and their values may not be
2398 used in arithmetic expressions.
2402 * Chars:: Characters
2406 @subsubsection Strings
2408 @cindex string constants
2409 @cindex constants, string
2410 A @dfn{string} is written between double-quotes. It may contain
2411 double-quotes or null characters. The way to get special characters
2412 into a string is to @dfn{escape} these characters: precede them with
2413 a backslash @samp{\} character. For example @samp{\\} represents
2414 one backslash: the first @code{\} is an escape which tells
2415 @command{@value{AS}} to interpret the second character literally as a backslash
2416 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2417 escape character). The complete list of escapes follows.
2419 @cindex escape codes, character
2420 @cindex character escape codes
2423 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2425 @cindex @code{\b} (backspace character)
2426 @cindex backspace (@code{\b})
2428 Mnemonic for backspace; for ASCII this is octal code 010.
2431 @c Mnemonic for EOText; for ASCII this is octal code 004.
2433 @cindex @code{\f} (formfeed character)
2434 @cindex formfeed (@code{\f})
2436 Mnemonic for FormFeed; for ASCII this is octal code 014.
2438 @cindex @code{\n} (newline character)
2439 @cindex newline (@code{\n})
2441 Mnemonic for newline; for ASCII this is octal code 012.
2444 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2446 @cindex @code{\r} (carriage return character)
2447 @cindex carriage return (@code{\r})
2449 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2452 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2453 @c other assemblers.
2455 @cindex @code{\t} (tab)
2456 @cindex tab (@code{\t})
2458 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2461 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2462 @c @item \x @var{digit} @var{digit} @var{digit}
2463 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2465 @cindex @code{\@var{ddd}} (octal character code)
2466 @cindex octal character code (@code{\@var{ddd}})
2467 @item \ @var{digit} @var{digit} @var{digit}
2468 An octal character code. The numeric code is 3 octal digits.
2469 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2470 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2472 @cindex @code{\@var{xd...}} (hex character code)
2473 @cindex hex character code (@code{\@var{xd...}})
2474 @item \@code{x} @var{hex-digits...}
2475 A hex character code. All trailing hex digits are combined. Either upper or
2476 lower case @code{x} works.
2478 @cindex @code{\\} (@samp{\} character)
2479 @cindex backslash (@code{\\})
2481 Represents one @samp{\} character.
2484 @c Represents one @samp{'} (accent acute) character.
2485 @c This is needed in single character literals
2486 @c (@xref{Characters,,Character Constants}.) to represent
2489 @cindex @code{\"} (doublequote character)
2490 @cindex doublequote (@code{\"})
2492 Represents one @samp{"} character. Needed in strings to represent
2493 this character, because an unescaped @samp{"} would end the string.
2495 @item \ @var{anything-else}
2496 Any other character when escaped by @kbd{\} gives a warning, but
2497 assembles as if the @samp{\} was not present. The idea is that if
2498 you used an escape sequence you clearly didn't want the literal
2499 interpretation of the following character. However @command{@value{AS}} has no
2500 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2501 code and warns you of the fact.
2504 Which characters are escapable, and what those escapes represent,
2505 varies widely among assemblers. The current set is what we think
2506 the BSD 4.2 assembler recognizes, and is a subset of what most C
2507 compilers recognize. If you are in doubt, do not use an escape
2511 @subsubsection Characters
2513 @cindex single character constant
2514 @cindex character, single
2515 @cindex constant, single character
2516 A single character may be written as a single quote immediately
2517 followed by that character. The same escapes apply to characters as
2518 to strings. So if you want to write the character backslash, you
2519 must write @kbd{'\\} where the first @code{\} escapes the second
2520 @code{\}. As you can see, the quote is an acute accent, not a
2521 grave accent. A newline
2523 @ifclear abnormal-separator
2524 (or semicolon @samp{;})
2526 @ifset abnormal-separator
2528 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2533 immediately following an acute accent is taken as a literal character
2534 and does not count as the end of a statement. The value of a character
2535 constant in a numeric expression is the machine's byte-wide code for
2536 that character. @command{@value{AS}} assumes your character code is ASCII:
2537 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2540 @subsection Number Constants
2542 @cindex constants, number
2543 @cindex number constants
2544 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2545 are stored in the target machine. @emph{Integers} are numbers that
2546 would fit into an @code{int} in the C language. @emph{Bignums} are
2547 integers, but they are stored in more than 32 bits. @emph{Flonums}
2548 are floating point numbers, described below.
2551 * Integers:: Integers
2556 * Bit Fields:: Bit Fields
2562 @subsubsection Integers
2564 @cindex constants, integer
2566 @cindex binary integers
2567 @cindex integers, binary
2568 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2569 the binary digits @samp{01}.
2571 @cindex octal integers
2572 @cindex integers, octal
2573 An octal integer is @samp{0} followed by zero or more of the octal
2574 digits (@samp{01234567}).
2576 @cindex decimal integers
2577 @cindex integers, decimal
2578 A decimal integer starts with a non-zero digit followed by zero or
2579 more digits (@samp{0123456789}).
2581 @cindex hexadecimal integers
2582 @cindex integers, hexadecimal
2583 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2584 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2586 Integers have the usual values. To denote a negative integer, use
2587 the prefix operator @samp{-} discussed under expressions
2588 (@pxref{Prefix Ops,,Prefix Operators}).
2591 @subsubsection Bignums
2594 @cindex constants, bignum
2595 A @dfn{bignum} has the same syntax and semantics as an integer
2596 except that the number (or its negative) takes more than 32 bits to
2597 represent in binary. The distinction is made because in some places
2598 integers are permitted while bignums are not.
2601 @subsubsection Flonums
2603 @cindex floating point numbers
2604 @cindex constants, floating point
2606 @cindex precision, floating point
2607 A @dfn{flonum} represents a floating point number. The translation is
2608 indirect: a decimal floating point number from the text is converted by
2609 @command{@value{AS}} to a generic binary floating point number of more than
2610 sufficient precision. This generic floating point number is converted
2611 to a particular computer's floating point format (or formats) by a
2612 portion of @command{@value{AS}} specialized to that computer.
2614 A flonum is written by writing (in order)
2619 (@samp{0} is optional on the HPPA.)
2623 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2625 @kbd{e} is recommended. Case is not important.
2627 @c FIXME: verify if flonum syntax really this vague for most cases
2628 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2629 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2632 On the H8/300, Renesas / SuperH SH,
2633 and AMD 29K architectures, the letter must be
2634 one of the letters @samp{DFPRSX} (in upper or lower case).
2636 On the ARC, the letter must be one of the letters @samp{DFRS}
2637 (in upper or lower case).
2639 On the Intel 960 architecture, the letter must be
2640 one of the letters @samp{DFT} (in upper or lower case).
2642 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2646 One of the letters @samp{DFRS} (in upper or lower case).
2649 One of the letters @samp{DFPRSX} (in upper or lower case).
2652 The letter @samp{E} (upper case only).
2655 One of the letters @samp{DFT} (in upper or lower case).
2660 An optional sign: either @samp{+} or @samp{-}.
2663 An optional @dfn{integer part}: zero or more decimal digits.
2666 An optional @dfn{fractional part}: @samp{.} followed by zero
2667 or more decimal digits.
2670 An optional exponent, consisting of:
2674 An @samp{E} or @samp{e}.
2675 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2676 @c principle this can perfectly well be different on different targets.
2678 Optional sign: either @samp{+} or @samp{-}.
2680 One or more decimal digits.
2685 At least one of the integer part or the fractional part must be
2686 present. The floating point number has the usual base-10 value.
2688 @command{@value{AS}} does all processing using integers. Flonums are computed
2689 independently of any floating point hardware in the computer running
2690 @command{@value{AS}}.
2694 @c Bit fields are written as a general facility but are also controlled
2695 @c by a conditional-compilation flag---which is as of now (21mar91)
2696 @c turned on only by the i960 config of GAS.
2698 @subsubsection Bit Fields
2701 @cindex constants, bit field
2702 You can also define numeric constants as @dfn{bit fields}.
2703 Specify two numbers separated by a colon---
2705 @var{mask}:@var{value}
2708 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2711 The resulting number is then packed
2713 @c this conditional paren in case bit fields turned on elsewhere than 960
2714 (in host-dependent byte order)
2716 into a field whose width depends on which assembler directive has the
2717 bit-field as its argument. Overflow (a result from the bitwise and
2718 requiring more binary digits to represent) is not an error; instead,
2719 more constants are generated, of the specified width, beginning with the
2720 least significant digits.@refill
2722 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2723 @code{.short}, and @code{.word} accept bit-field arguments.
2728 @chapter Sections and Relocation
2733 * Secs Background:: Background
2734 * Ld Sections:: Linker Sections
2735 * As Sections:: Assembler Internal Sections
2736 * Sub-Sections:: Sub-Sections
2740 @node Secs Background
2743 Roughly, a section is a range of addresses, with no gaps; all data
2744 ``in'' those addresses is treated the same for some particular purpose.
2745 For example there may be a ``read only'' section.
2747 @cindex linker, and assembler
2748 @cindex assembler, and linker
2749 The linker @code{@value{LD}} reads many object files (partial programs) and
2750 combines their contents to form a runnable program. When @command{@value{AS}}
2751 emits an object file, the partial program is assumed to start at address 0.
2752 @code{@value{LD}} assigns the final addresses for the partial program, so that
2753 different partial programs do not overlap. This is actually an
2754 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2757 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2758 addresses. These blocks slide to their run-time addresses as rigid
2759 units; their length does not change and neither does the order of bytes
2760 within them. Such a rigid unit is called a @emph{section}. Assigning
2761 run-time addresses to sections is called @dfn{relocation}. It includes
2762 the task of adjusting mentions of object-file addresses so they refer to
2763 the proper run-time addresses.
2765 For the H8/300, and for the Renesas / SuperH SH,
2766 @command{@value{AS}} pads sections if needed to
2767 ensure they end on a word (sixteen bit) boundary.
2770 @cindex standard assembler sections
2771 An object file written by @command{@value{AS}} has at least three sections, any
2772 of which may be empty. These are named @dfn{text}, @dfn{data} and
2777 When it generates COFF or ELF output,
2779 @command{@value{AS}} can also generate whatever other named sections you specify
2780 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2781 If you do not use any directives that place output in the @samp{.text}
2782 or @samp{.data} sections, these sections still exist, but are empty.
2787 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2789 @command{@value{AS}} can also generate whatever other named sections you
2790 specify using the @samp{.space} and @samp{.subspace} directives. See
2791 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2792 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2793 assembler directives.
2796 Additionally, @command{@value{AS}} uses different names for the standard
2797 text, data, and bss sections when generating SOM output. Program text
2798 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2799 BSS into @samp{$BSS$}.
2803 Within the object file, the text section starts at address @code{0}, the
2804 data section follows, and the bss section follows the data section.
2807 When generating either SOM or ELF output files on the HPPA, the text
2808 section starts at address @code{0}, the data section at address
2809 @code{0x4000000}, and the bss section follows the data section.
2812 To let @code{@value{LD}} know which data changes when the sections are
2813 relocated, and how to change that data, @command{@value{AS}} also writes to the
2814 object file details of the relocation needed. To perform relocation
2815 @code{@value{LD}} must know, each time an address in the object
2819 Where in the object file is the beginning of this reference to
2822 How long (in bytes) is this reference?
2824 Which section does the address refer to? What is the numeric value of
2826 (@var{address}) @minus{} (@var{start-address of section})?
2829 Is the reference to an address ``Program-Counter relative''?
2832 @cindex addresses, format of
2833 @cindex section-relative addressing
2834 In fact, every address @command{@value{AS}} ever uses is expressed as
2836 (@var{section}) + (@var{offset into section})
2839 Further, most expressions @command{@value{AS}} computes have this section-relative
2842 (For some object formats, such as SOM for the HPPA, some expressions are
2843 symbol-relative instead.)
2846 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2847 @var{N} into section @var{secname}.''
2849 Apart from text, data and bss sections you need to know about the
2850 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2851 addresses in the absolute section remain unchanged. For example, address
2852 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2853 @code{@value{LD}}. Although the linker never arranges two partial programs'
2854 data sections with overlapping addresses after linking, @emph{by definition}
2855 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2856 part of a program is always the same address when the program is running as
2857 address @code{@{absolute@ 239@}} in any other part of the program.
2859 The idea of sections is extended to the @dfn{undefined} section. Any
2860 address whose section is unknown at assembly time is by definition
2861 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2862 Since numbers are always defined, the only way to generate an undefined
2863 address is to mention an undefined symbol. A reference to a named
2864 common block would be such a symbol: its value is unknown at assembly
2865 time so it has section @emph{undefined}.
2867 By analogy the word @emph{section} is used to describe groups of sections in
2868 the linked program. @code{@value{LD}} puts all partial programs' text
2869 sections in contiguous addresses in the linked program. It is
2870 customary to refer to the @emph{text section} of a program, meaning all
2871 the addresses of all partial programs' text sections. Likewise for
2872 data and bss sections.
2874 Some sections are manipulated by @code{@value{LD}}; others are invented for
2875 use of @command{@value{AS}} and have no meaning except during assembly.
2878 @section Linker Sections
2879 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2884 @cindex named sections
2885 @cindex sections, named
2886 @item named sections
2889 @cindex text section
2890 @cindex data section
2894 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2895 separate but equal sections. Anything you can say of one section is
2898 When the program is running, however, it is
2899 customary for the text section to be unalterable. The
2900 text section is often shared among processes: it contains
2901 instructions, constants and the like. The data section of a running
2902 program is usually alterable: for example, C variables would be stored
2903 in the data section.
2908 This section contains zeroed bytes when your program begins running. It
2909 is used to hold uninitialized variables or common storage. The length of
2910 each partial program's bss section is important, but because it starts
2911 out containing zeroed bytes there is no need to store explicit zero
2912 bytes in the object file. The bss section was invented to eliminate
2913 those explicit zeros from object files.
2915 @cindex absolute section
2916 @item absolute section
2917 Address 0 of this section is always ``relocated'' to runtime address 0.
2918 This is useful if you want to refer to an address that @code{@value{LD}} must
2919 not change when relocating. In this sense we speak of absolute
2920 addresses being ``unrelocatable'': they do not change during relocation.
2922 @cindex undefined section
2923 @item undefined section
2924 This ``section'' is a catch-all for address references to objects not in
2925 the preceding sections.
2926 @c FIXME: ref to some other doc on obj-file formats could go here.
2929 @cindex relocation example
2930 An idealized example of three relocatable sections follows.
2932 The example uses the traditional section names @samp{.text} and @samp{.data}.
2934 Memory addresses are on the horizontal axis.
2938 @c END TEXI2ROFF-KILL
2941 partial program # 1: |ttttt|dddd|00|
2948 partial program # 2: |TTT|DDD|000|
2951 +--+---+-----+--+----+---+-----+~~
2952 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2953 +--+---+-----+--+----+---+-----+~~
2955 addresses: 0 @dots{}
2962 \line{\it Partial program \#1: \hfil}
2963 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2964 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2966 \line{\it Partial program \#2: \hfil}
2967 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2968 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2970 \line{\it linked program: \hfil}
2971 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2972 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2973 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2974 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2976 \line{\it addresses: \hfil}
2980 @c END TEXI2ROFF-KILL
2983 @section Assembler Internal Sections
2985 @cindex internal assembler sections
2986 @cindex sections in messages, internal
2987 These sections are meant only for the internal use of @command{@value{AS}}. They
2988 have no meaning at run-time. You do not really need to know about these
2989 sections for most purposes; but they can be mentioned in @command{@value{AS}}
2990 warning messages, so it might be helpful to have an idea of their
2991 meanings to @command{@value{AS}}. These sections are used to permit the
2992 value of every expression in your assembly language program to be a
2993 section-relative address.
2996 @cindex assembler internal logic error
2997 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2998 An internal assembler logic error has been found. This means there is a
2999 bug in the assembler.
3001 @cindex expr (internal section)
3003 The assembler stores complex expression internally as combinations of
3004 symbols. When it needs to represent an expression as a symbol, it puts
3005 it in the expr section.
3007 @c FIXME item transfer[t] vector preload
3008 @c FIXME item transfer[t] vector postload
3009 @c FIXME item register
3013 @section Sub-Sections
3015 @cindex numbered subsections
3016 @cindex grouping data
3022 fall into two sections: text and data.
3024 You may have separate groups of
3026 data in named sections
3030 data in named sections
3036 that you want to end up near to each other in the object file, even though they
3037 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3038 use @dfn{subsections} for this purpose. Within each section, there can be
3039 numbered subsections with values from 0 to 8192. Objects assembled into the
3040 same subsection go into the object file together with other objects in the same
3041 subsection. For example, a compiler might want to store constants in the text
3042 section, but might not want to have them interspersed with the program being
3043 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3044 section of code being output, and a @samp{.text 1} before each group of
3045 constants being output.
3047 Subsections are optional. If you do not use subsections, everything
3048 goes in subsection number zero.
3051 Each subsection is zero-padded up to a multiple of four bytes.
3052 (Subsections may be padded a different amount on different flavors
3053 of @command{@value{AS}}.)
3057 On the H8/300 platform, each subsection is zero-padded to a word
3058 boundary (two bytes).
3059 The same is true on the Renesas SH.
3062 @c FIXME section padding (alignment)?
3063 @c Rich Pixley says padding here depends on target obj code format; that
3064 @c doesn't seem particularly useful to say without further elaboration,
3065 @c so for now I say nothing about it. If this is a generic BFD issue,
3066 @c these paragraphs might need to vanish from this manual, and be
3067 @c discussed in BFD chapter of binutils (or some such).
3071 Subsections appear in your object file in numeric order, lowest numbered
3072 to highest. (All this to be compatible with other people's assemblers.)
3073 The object file contains no representation of subsections; @code{@value{LD}} and
3074 other programs that manipulate object files see no trace of them.
3075 They just see all your text subsections as a text section, and all your
3076 data subsections as a data section.
3078 To specify which subsection you want subsequent statements assembled
3079 into, use a numeric argument to specify it, in a @samp{.text
3080 @var{expression}} or a @samp{.data @var{expression}} statement.
3083 When generating COFF output, you
3088 can also use an extra subsection
3089 argument with arbitrary named sections: @samp{.section @var{name},
3094 When generating ELF output, you
3099 can also use the @code{.subsection} directive (@pxref{SubSection})
3100 to specify a subsection: @samp{.subsection @var{expression}}.
3102 @var{Expression} should be an absolute expression
3103 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3104 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3105 begins in @code{text 0}. For instance:
3107 .text 0 # The default subsection is text 0 anyway.
3108 .ascii "This lives in the first text subsection. *"
3110 .ascii "But this lives in the second text subsection."
3112 .ascii "This lives in the data section,"
3113 .ascii "in the first data subsection."
3115 .ascii "This lives in the first text section,"
3116 .ascii "immediately following the asterisk (*)."
3119 Each section has a @dfn{location counter} incremented by one for every byte
3120 assembled into that section. Because subsections are merely a convenience
3121 restricted to @command{@value{AS}} there is no concept of a subsection location
3122 counter. There is no way to directly manipulate a location counter---but the
3123 @code{.align} directive changes it, and any label definition captures its
3124 current value. The location counter of the section where statements are being
3125 assembled is said to be the @dfn{active} location counter.
3128 @section bss Section
3131 @cindex common variable storage
3132 The bss section is used for local common variable storage.
3133 You may allocate address space in the bss section, but you may
3134 not dictate data to load into it before your program executes. When
3135 your program starts running, all the contents of the bss
3136 section are zeroed bytes.
3138 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3139 @ref{Lcomm,,@code{.lcomm}}.
3141 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3142 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3145 When assembling for a target which supports multiple sections, such as ELF or
3146 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3147 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3148 section. Typically the section will only contain symbol definitions and
3149 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3156 Symbols are a central concept: the programmer uses symbols to name
3157 things, the linker uses symbols to link, and the debugger uses symbols
3161 @cindex debuggers, and symbol order
3162 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3163 the same order they were declared. This may break some debuggers.
3168 * Setting Symbols:: Giving Symbols Other Values
3169 * Symbol Names:: Symbol Names
3170 * Dot:: The Special Dot Symbol
3171 * Symbol Attributes:: Symbol Attributes
3178 A @dfn{label} is written as a symbol immediately followed by a colon
3179 @samp{:}. The symbol then represents the current value of the
3180 active location counter, and is, for example, a suitable instruction
3181 operand. You are warned if you use the same symbol to represent two
3182 different locations: the first definition overrides any other
3186 On the HPPA, the usual form for a label need not be immediately followed by a
3187 colon, but instead must start in column zero. Only one label may be defined on
3188 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3189 provides a special directive @code{.label} for defining labels more flexibly.
3192 @node Setting Symbols
3193 @section Giving Symbols Other Values
3195 @cindex assigning values to symbols
3196 @cindex symbol values, assigning
3197 A symbol can be given an arbitrary value by writing a symbol, followed
3198 by an equals sign @samp{=}, followed by an expression
3199 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3200 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3201 equals sign @samp{=}@samp{=} here represents an equivalent of the
3202 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3205 @section Symbol Names
3207 @cindex symbol names
3208 @cindex names, symbol
3209 @ifclear SPECIAL-SYMS
3210 Symbol names begin with a letter or with one of @samp{._}. On most
3211 machines, you can also use @code{$} in symbol names; exceptions are
3212 noted in @ref{Machine Dependencies}. That character may be followed by any
3213 string of digits, letters, dollar signs (unless otherwise noted for a
3214 particular target machine), and underscores.
3218 Symbol names begin with a letter or with one of @samp{._}. On the
3219 Renesas SH you can also use @code{$} in symbol names. That
3220 character may be followed by any string of digits, letters, dollar signs (save
3221 on the H8/300), and underscores.
3225 Case of letters is significant: @code{foo} is a different symbol name
3228 Each symbol has exactly one name. Each name in an assembly language program
3229 refers to exactly one symbol. You may use that symbol name any number of times
3232 @subheading Local Symbol Names
3234 @cindex local symbol names
3235 @cindex symbol names, local
3236 A local symbol is any symbol beginning with certain local label prefixes.
3237 By default, the local label prefix is @samp{.L} for ELF systems or
3238 @samp{L} for traditional a.out systems, but each target may have its own
3239 set of local label prefixes.
3241 On the HPPA local symbols begin with @samp{L$}.
3244 Local symbols are defined and used within the assembler, but they are
3245 normally not saved in object files. Thus, they are not visible when debugging.
3246 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3247 @option{-L}}) to retain the local symbols in the object files.
3249 @subheading Local Labels
3251 @cindex local labels
3252 @cindex temporary symbol names
3253 @cindex symbol names, temporary
3254 Local labels help compilers and programmers use names temporarily.
3255 They create symbols which are guaranteed to be unique over the entire scope of
3256 the input source code and which can be referred to by a simple notation.
3257 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3258 represents any positive integer). To refer to the most recent previous
3259 definition of that label write @samp{@b{N}b}, using the same number as when
3260 you defined the label. To refer to the next definition of a local label, write
3261 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3264 There is no restriction on how you can use these labels, and you can reuse them
3265 too. So that it is possible to repeatedly define the same local label (using
3266 the same number @samp{@b{N}}), although you can only refer to the most recently
3267 defined local label of that number (for a backwards reference) or the next
3268 definition of a specific local label for a forward reference. It is also worth
3269 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3270 implemented in a slightly more efficient manner than the others.
3281 Which is the equivalent of:
3284 label_1: branch label_3
3285 label_2: branch label_1
3286 label_3: branch label_4
3287 label_4: branch label_3
3290 Local label names are only a notational device. They are immediately
3291 transformed into more conventional symbol names before the assembler uses them.
3292 The symbol names are stored in the symbol table, appear in error messages, and
3293 are optionally emitted to the object file. The names are constructed using
3297 @item @emph{local label prefix}
3298 All local symbols begin with the system-specific local label prefix.
3299 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3300 that start with the local label prefix. These labels are
3301 used for symbols you are never intended to see. If you use the
3302 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3303 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3304 you may use them in debugging.
3307 This is the number that was used in the local label definition. So if the
3308 label is written @samp{55:} then the number is @samp{55}.
3311 This unusual character is included so you do not accidentally invent a symbol
3312 of the same name. The character has ASCII value of @samp{\002} (control-B).
3314 @item @emph{ordinal number}
3315 This is a serial number to keep the labels distinct. The first definition of
3316 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3317 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3318 the number @samp{1} and its 15th definition gets @samp{15} as well.
3321 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3322 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3324 @subheading Dollar Local Labels
3325 @cindex dollar local symbols
3327 @code{@value{AS}} also supports an even more local form of local labels called
3328 dollar labels. These labels go out of scope (i.e., they become undefined) as
3329 soon as a non-local label is defined. Thus they remain valid for only a small
3330 region of the input source code. Normal local labels, by contrast, remain in
3331 scope for the entire file, or until they are redefined by another occurrence of
3332 the same local label.
3334 Dollar labels are defined in exactly the same way as ordinary local labels,
3335 except that instead of being terminated by a colon, they are terminated by a
3336 dollar sign, e.g., @samp{@b{55$}}.
3338 They can also be distinguished from ordinary local labels by their transformed
3339 names which use ASCII character @samp{\001} (control-A) as the magic character
3340 to distinguish them from ordinary labels. For example, the fifth definition of
3341 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3344 @section The Special Dot Symbol
3346 @cindex dot (symbol)
3347 @cindex @code{.} (symbol)
3348 @cindex current address
3349 @cindex location counter
3350 The special symbol @samp{.} refers to the current address that
3351 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3352 .long .} defines @code{melvin} to contain its own address.
3353 Assigning a value to @code{.} is treated the same as a @code{.org}
3354 directive. Thus, the expression @samp{.=.+4} is the same as saying
3355 @ifclear no-space-dir
3359 @node Symbol Attributes
3360 @section Symbol Attributes
3362 @cindex symbol attributes
3363 @cindex attributes, symbol
3364 Every symbol has, as well as its name, the attributes ``Value'' and
3365 ``Type''. Depending on output format, symbols can also have auxiliary
3368 The detailed definitions are in @file{a.out.h}.
3371 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3372 all these attributes, and probably won't warn you. This makes the
3373 symbol an externally defined symbol, which is generally what you
3377 * Symbol Value:: Value
3378 * Symbol Type:: Type
3381 * a.out Symbols:: Symbol Attributes: @code{a.out}
3385 * a.out Symbols:: Symbol Attributes: @code{a.out}
3388 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3393 * COFF Symbols:: Symbol Attributes for COFF
3396 * SOM Symbols:: Symbol Attributes for SOM
3403 @cindex value of a symbol
3404 @cindex symbol value
3405 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3406 location in the text, data, bss or absolute sections the value is the
3407 number of addresses from the start of that section to the label.
3408 Naturally for text, data and bss sections the value of a symbol changes
3409 as @code{@value{LD}} changes section base addresses during linking. Absolute
3410 symbols' values do not change during linking: that is why they are
3413 The value of an undefined symbol is treated in a special way. If it is
3414 0 then the symbol is not defined in this assembler source file, and
3415 @code{@value{LD}} tries to determine its value from other files linked into the
3416 same program. You make this kind of symbol simply by mentioning a symbol
3417 name without defining it. A non-zero value represents a @code{.comm}
3418 common declaration. The value is how much common storage to reserve, in
3419 bytes (addresses). The symbol refers to the first address of the
3425 @cindex type of a symbol
3427 The type attribute of a symbol contains relocation (section)
3428 information, any flag settings indicating that a symbol is external, and
3429 (optionally), other information for linkers and debuggers. The exact
3430 format depends on the object-code output format in use.
3435 @c The following avoids a "widow" subsection title. @group would be
3436 @c better if it were available outside examples.
3439 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3441 @cindex @code{b.out} symbol attributes
3442 @cindex symbol attributes, @code{b.out}
3443 These symbol attributes appear only when @command{@value{AS}} is configured for
3444 one of the Berkeley-descended object output formats---@code{a.out} or
3450 @subsection Symbol Attributes: @code{a.out}
3452 @cindex @code{a.out} symbol attributes
3453 @cindex symbol attributes, @code{a.out}
3459 @subsection Symbol Attributes: @code{a.out}
3461 @cindex @code{a.out} symbol attributes
3462 @cindex symbol attributes, @code{a.out}
3466 * Symbol Desc:: Descriptor
3467 * Symbol Other:: Other
3471 @subsubsection Descriptor
3473 @cindex descriptor, of @code{a.out} symbol
3474 This is an arbitrary 16-bit value. You may establish a symbol's
3475 descriptor value by using a @code{.desc} statement
3476 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3477 @command{@value{AS}}.
3480 @subsubsection Other
3482 @cindex other attribute, of @code{a.out} symbol
3483 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3488 @subsection Symbol Attributes for COFF
3490 @cindex COFF symbol attributes
3491 @cindex symbol attributes, COFF
3493 The COFF format supports a multitude of auxiliary symbol attributes;
3494 like the primary symbol attributes, they are set between @code{.def} and
3495 @code{.endef} directives.
3497 @subsubsection Primary Attributes
3499 @cindex primary attributes, COFF symbols
3500 The symbol name is set with @code{.def}; the value and type,
3501 respectively, with @code{.val} and @code{.type}.
3503 @subsubsection Auxiliary Attributes
3505 @cindex auxiliary attributes, COFF symbols
3506 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3507 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3508 table information for COFF.
3513 @subsection Symbol Attributes for SOM
3515 @cindex SOM symbol attributes
3516 @cindex symbol attributes, SOM
3518 The SOM format for the HPPA supports a multitude of symbol attributes set with
3519 the @code{.EXPORT} and @code{.IMPORT} directives.
3521 The attributes are described in @cite{HP9000 Series 800 Assembly
3522 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3523 @code{EXPORT} assembler directive documentation.
3527 @chapter Expressions
3531 @cindex numeric values
3532 An @dfn{expression} specifies an address or numeric value.
3533 Whitespace may precede and/or follow an expression.
3535 The result of an expression must be an absolute number, or else an offset into
3536 a particular section. If an expression is not absolute, and there is not
3537 enough information when @command{@value{AS}} sees the expression to know its
3538 section, a second pass over the source program might be necessary to interpret
3539 the expression---but the second pass is currently not implemented.
3540 @command{@value{AS}} aborts with an error message in this situation.
3543 * Empty Exprs:: Empty Expressions
3544 * Integer Exprs:: Integer Expressions
3548 @section Empty Expressions
3550 @cindex empty expressions
3551 @cindex expressions, empty
3552 An empty expression has no value: it is just whitespace or null.
3553 Wherever an absolute expression is required, you may omit the
3554 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3555 is compatible with other assemblers.
3558 @section Integer Expressions
3560 @cindex integer expressions
3561 @cindex expressions, integer
3562 An @dfn{integer expression} is one or more @emph{arguments} delimited
3563 by @emph{operators}.
3566 * Arguments:: Arguments
3567 * Operators:: Operators
3568 * Prefix Ops:: Prefix Operators
3569 * Infix Ops:: Infix Operators
3573 @subsection Arguments
3575 @cindex expression arguments
3576 @cindex arguments in expressions
3577 @cindex operands in expressions
3578 @cindex arithmetic operands
3579 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3580 contexts arguments are sometimes called ``arithmetic operands''. In
3581 this manual, to avoid confusing them with the ``instruction operands'' of
3582 the machine language, we use the term ``argument'' to refer to parts of
3583 expressions only, reserving the word ``operand'' to refer only to machine
3584 instruction operands.
3586 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3587 @var{section} is one of text, data, bss, absolute,
3588 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3591 Numbers are usually integers.
3593 A number can be a flonum or bignum. In this case, you are warned
3594 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3595 these 32 bits are an integer. You may write integer-manipulating
3596 instructions that act on exotic constants, compatible with other
3599 @cindex subexpressions
3600 Subexpressions are a left parenthesis @samp{(} followed by an integer
3601 expression, followed by a right parenthesis @samp{)}; or a prefix
3602 operator followed by an argument.
3605 @subsection Operators
3607 @cindex operators, in expressions
3608 @cindex arithmetic functions
3609 @cindex functions, in expressions
3610 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3611 operators are followed by an argument. Infix operators appear
3612 between their arguments. Operators may be preceded and/or followed by
3616 @subsection Prefix Operator
3618 @cindex prefix operators
3619 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3620 one argument, which must be absolute.
3622 @c the tex/end tex stuff surrounding this small table is meant to make
3623 @c it align, on the printed page, with the similar table in the next
3624 @c section (which is inside an enumerate).
3626 \global\advance\leftskip by \itemindent
3631 @dfn{Negation}. Two's complement negation.
3633 @dfn{Complementation}. Bitwise not.
3637 \global\advance\leftskip by -\itemindent
3641 @subsection Infix Operators
3643 @cindex infix operators
3644 @cindex operators, permitted arguments
3645 @dfn{Infix operators} take two arguments, one on either side. Operators
3646 have precedence, but operations with equal precedence are performed left
3647 to right. Apart from @code{+} or @option{-}, both arguments must be
3648 absolute, and the result is absolute.
3651 @cindex operator precedence
3652 @cindex precedence of operators
3659 @dfn{Multiplication}.
3662 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3668 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3671 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3675 Intermediate precedence
3680 @dfn{Bitwise Inclusive Or}.
3686 @dfn{Bitwise Exclusive Or}.
3689 @dfn{Bitwise Or Not}.
3696 @cindex addition, permitted arguments
3697 @cindex plus, permitted arguments
3698 @cindex arguments for addition
3700 @dfn{Addition}. If either argument is absolute, the result has the section of
3701 the other argument. You may not add together arguments from different
3704 @cindex subtraction, permitted arguments
3705 @cindex minus, permitted arguments
3706 @cindex arguments for subtraction
3708 @dfn{Subtraction}. If the right argument is absolute, the
3709 result has the section of the left argument.
3710 If both arguments are in the same section, the result is absolute.
3711 You may not subtract arguments from different sections.
3712 @c FIXME is there still something useful to say about undefined - undefined ?
3714 @cindex comparison expressions
3715 @cindex expressions, comparison
3720 @dfn{Is Not Equal To}
3724 @dfn{Is Greater Than}
3726 @dfn{Is Greater Than Or Equal To}
3728 @dfn{Is Less Than Or Equal To}
3730 The comparison operators can be used as infix operators. A true results has a
3731 value of -1 whereas a false result has a value of 0. Note, these operators
3732 perform signed comparisons.
3735 @item Lowest Precedence
3744 These two logical operations can be used to combine the results of sub
3745 expressions. Note, unlike the comparison operators a true result returns a
3746 value of 1 but a false results does still return 0. Also note that the logical
3747 or operator has a slightly lower precedence than logical and.
3752 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3753 address; you can only have a defined section in one of the two arguments.
3756 @chapter Assembler Directives
3758 @cindex directives, machine independent
3759 @cindex pseudo-ops, machine independent
3760 @cindex machine independent directives
3761 All assembler directives have names that begin with a period (@samp{.}).
3762 The rest of the name is letters, usually in lower case.
3764 This chapter discusses directives that are available regardless of the
3765 target machine configuration for the @sc{gnu} assembler.
3767 Some machine configurations provide additional directives.
3768 @xref{Machine Dependencies}.
3771 @ifset machine-directives
3772 @xref{Machine Dependencies}, for additional directives.
3777 * Abort:: @code{.abort}
3779 * ABORT (COFF):: @code{.ABORT}
3782 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3783 * Altmacro:: @code{.altmacro}
3784 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3785 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3786 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3787 * Byte:: @code{.byte @var{expressions}}
3788 * Comm:: @code{.comm @var{symbol} , @var{length} }
3790 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
3792 * Data:: @code{.data @var{subsection}}
3794 * Def:: @code{.def @var{name}}
3797 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3803 * Double:: @code{.double @var{flonums}}
3804 * Eject:: @code{.eject}
3805 * Else:: @code{.else}
3806 * Elseif:: @code{.elseif}
3809 * Endef:: @code{.endef}
3812 * Endfunc:: @code{.endfunc}
3813 * Endif:: @code{.endif}
3814 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3815 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3816 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
3818 * Error:: @code{.error @var{string}}
3819 * Exitm:: @code{.exitm}
3820 * Extern:: @code{.extern}
3821 * Fail:: @code{.fail}
3822 @ifclear no-file-dir
3823 * File:: @code{.file @var{string}}
3826 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3827 * Float:: @code{.float @var{flonums}}
3828 * Func:: @code{.func}
3829 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3831 * Hidden:: @code{.hidden @var{names}}
3834 * hword:: @code{.hword @var{expressions}}
3835 * Ident:: @code{.ident}
3836 * If:: @code{.if @var{absolute expression}}
3837 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3838 * Include:: @code{.include "@var{file}"}
3839 * Int:: @code{.int @var{expressions}}
3841 * Internal:: @code{.internal @var{names}}
3844 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3845 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3846 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3847 * Lflags:: @code{.lflags}
3848 @ifclear no-line-dir
3849 * Line:: @code{.line @var{line-number}}
3852 * Linkonce:: @code{.linkonce [@var{type}]}
3853 * List:: @code{.list}
3854 * Ln:: @code{.ln @var{line-number}}
3856 * LNS directives:: @code{.file}, @code{.loc}, etc.
3858 * Long:: @code{.long @var{expressions}}
3860 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3863 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3864 * MRI:: @code{.mri @var{val}}
3865 * Noaltmacro:: @code{.noaltmacro}
3866 * Nolist:: @code{.nolist}
3867 * Octa:: @code{.octa @var{bignums}}
3868 * Org:: @code{.org @var{new-lc}, @var{fill}}
3869 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3871 * PopSection:: @code{.popsection}
3872 * Previous:: @code{.previous}
3875 * Print:: @code{.print @var{string}}
3877 * Protected:: @code{.protected @var{names}}
3880 * Psize:: @code{.psize @var{lines}, @var{columns}}
3881 * Purgem:: @code{.purgem @var{name}}
3883 * PushSection:: @code{.pushsection @var{name}}
3886 * Quad:: @code{.quad @var{bignums}}
3887 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
3888 * Rept:: @code{.rept @var{count}}
3889 * Sbttl:: @code{.sbttl "@var{subheading}"}
3891 * Scl:: @code{.scl @var{class}}
3894 * Section:: @code{.section @var{name}}
3897 * Set:: @code{.set @var{symbol}, @var{expression}}
3898 * Short:: @code{.short @var{expressions}}
3899 * Single:: @code{.single @var{flonums}}
3901 * Size:: @code{.size [@var{name} , @var{expression}]}
3904 * Skip:: @code{.skip @var{size} , @var{fill}}
3905 * Sleb128:: @code{.sleb128 @var{expressions}}
3906 * Space:: @code{.space @var{size} , @var{fill}}
3908 * Stab:: @code{.stabd, .stabn, .stabs}
3911 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
3912 * Struct:: @code{.struct @var{expression}}
3914 * SubSection:: @code{.subsection}
3915 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3919 * Tag:: @code{.tag @var{structname}}
3922 * Text:: @code{.text @var{subsection}}
3923 * Title:: @code{.title "@var{heading}"}
3925 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3928 * Uleb128:: @code{.uleb128 @var{expressions}}
3930 * Val:: @code{.val @var{addr}}
3934 * Version:: @code{.version "@var{string}"}
3935 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3936 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3939 * Warning:: @code{.warning @var{string}}
3940 * Weak:: @code{.weak @var{names}}
3941 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
3942 * Word:: @code{.word @var{expressions}}
3943 * Deprecated:: Deprecated Directives
3947 @section @code{.abort}
3949 @cindex @code{abort} directive
3950 @cindex stopping the assembly
3951 This directive stops the assembly immediately. It is for
3952 compatibility with other assemblers. The original idea was that the
3953 assembly language source would be piped into the assembler. If the sender
3954 of the source quit, it could use this directive tells @command{@value{AS}} to
3955 quit also. One day @code{.abort} will not be supported.
3959 @section @code{.ABORT} (COFF)
3961 @cindex @code{ABORT} directive
3962 When producing COFF output, @command{@value{AS}} accepts this directive as a
3963 synonym for @samp{.abort}.
3966 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
3972 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3974 @cindex padding the location counter
3975 @cindex @code{align} directive
3976 Pad the location counter (in the current subsection) to a particular storage
3977 boundary. The first expression (which must be absolute) is the alignment
3978 required, as described below.
3980 The second expression (also absolute) gives the fill value to be stored in the
3981 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3982 padding bytes are normally zero. However, on some systems, if the section is
3983 marked as containing code and the fill value is omitted, the space is filled
3984 with no-op instructions.
3986 The third expression is also absolute, and is also optional. If it is present,
3987 it is the maximum number of bytes that should be skipped by this alignment
3988 directive. If doing the alignment would require skipping more bytes than the
3989 specified maximum, then the alignment is not done at all. You can omit the
3990 fill value (the second argument) entirely by simply using two commas after the
3991 required alignment; this can be useful if you want the alignment to be filled
3992 with no-op instructions when appropriate.
3994 The way the required alignment is specified varies from system to system.
3995 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
3996 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
3997 alignment request in bytes. For example @samp{.align 8} advances
3998 the location counter until it is a multiple of 8. If the location counter
3999 is already a multiple of 8, no change is needed. For the tic54x, the
4000 first expression is the alignment request in words.
4002 For other systems, including the i386 using a.out format, and the arm and
4003 strongarm, it is the
4004 number of low-order zero bits the location counter must have after
4005 advancement. For example @samp{.align 3} advances the location
4006 counter until it a multiple of 8. If the location counter is already a
4007 multiple of 8, no change is needed.
4009 This inconsistency is due to the different behaviors of the various
4010 native assemblers for these systems which GAS must emulate.
4011 GAS also provides @code{.balign} and @code{.p2align} directives,
4012 described later, which have a consistent behavior across all
4013 architectures (but are specific to GAS).
4016 @section @code{.ascii "@var{string}"}@dots{}
4018 @cindex @code{ascii} directive
4019 @cindex string literals
4020 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4021 separated by commas. It assembles each string (with no automatic
4022 trailing zero byte) into consecutive addresses.
4025 @section @code{.asciz "@var{string}"}@dots{}
4027 @cindex @code{asciz} directive
4028 @cindex zero-terminated strings
4029 @cindex null-terminated strings
4030 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4031 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4034 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4036 @cindex padding the location counter given number of bytes
4037 @cindex @code{balign} directive
4038 Pad the location counter (in the current subsection) to a particular
4039 storage boundary. The first expression (which must be absolute) is the
4040 alignment request in bytes. For example @samp{.balign 8} advances
4041 the location counter until it is a multiple of 8. If the location counter
4042 is already a multiple of 8, no change is needed.
4044 The second expression (also absolute) gives the fill value to be stored in the
4045 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4046 padding bytes are normally zero. However, on some systems, if the section is
4047 marked as containing code and the fill value is omitted, the space is filled
4048 with no-op instructions.
4050 The third expression is also absolute, and is also optional. If it is present,
4051 it is the maximum number of bytes that should be skipped by this alignment
4052 directive. If doing the alignment would require skipping more bytes than the
4053 specified maximum, then the alignment is not done at all. You can omit the
4054 fill value (the second argument) entirely by simply using two commas after the
4055 required alignment; this can be useful if you want the alignment to be filled
4056 with no-op instructions when appropriate.
4058 @cindex @code{balignw} directive
4059 @cindex @code{balignl} directive
4060 The @code{.balignw} and @code{.balignl} directives are variants of the
4061 @code{.balign} directive. The @code{.balignw} directive treats the fill
4062 pattern as a two byte word value. The @code{.balignl} directives treats the
4063 fill pattern as a four byte longword value. For example, @code{.balignw
4064 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4065 filled in with the value 0x368d (the exact placement of the bytes depends upon
4066 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4070 @section @code{.byte @var{expressions}}
4072 @cindex @code{byte} directive
4073 @cindex integers, one byte
4074 @code{.byte} expects zero or more expressions, separated by commas.
4075 Each expression is assembled into the next byte.
4078 @section @code{.comm @var{symbol} , @var{length} }
4080 @cindex @code{comm} directive
4081 @cindex symbol, common
4082 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4083 common symbol in one object file may be merged with a defined or common symbol
4084 of the same name in another object file. If @code{@value{LD}} does not see a
4085 definition for the symbol--just one or more common symbols--then it will
4086 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4087 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4088 the same name, and they do not all have the same size, it will allocate space
4089 using the largest size.
4092 When using ELF, the @code{.comm} directive takes an optional third argument.
4093 This is the desired alignment of the symbol, specified as a byte boundary (for
4094 example, an alignment of 16 means that the least significant 4 bits of the
4095 address should be zero). The alignment must be an absolute expression, and it
4096 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
4097 for the common symbol, it will use the alignment when placing the symbol. If
4098 no alignment is specified, @command{@value{AS}} will set the alignment to the
4099 largest power of two less than or equal to the size of the symbol, up to a
4104 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4105 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4108 @node CFI directives
4109 @section @code{.cfi_startproc [simple]}
4110 @cindex @code{cfi_startproc} directive
4111 @code{.cfi_startproc} is used at the beginning of each function that
4112 should have an entry in @code{.eh_frame}. It initializes some internal
4113 data structures. Don't forget to close the function by
4114 @code{.cfi_endproc}.
4116 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4117 it also emits some architecture dependent initial CFI instructions.
4119 @section @code{.cfi_endproc}
4120 @cindex @code{cfi_endproc} directive
4121 @code{.cfi_endproc} is used at the end of a function where it closes its
4122 unwind entry previously opened by
4123 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4125 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4126 @code{.cfi_personality} defines personality routine and its encoding.
4127 @var{encoding} must be a constant determining how the personality
4128 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4129 argument is not present, otherwise second argument should be
4130 a constant or a symbol name. When using indirect encodings,
4131 the symbol provided should be the location where personality
4132 can be loaded from, not the personality routine itself.
4133 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4134 no personality routine.
4136 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4137 @code{.cfi_lsda} defines LSDA and its encoding.
4138 @var{encoding} must be a constant determining how the LSDA
4139 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4140 argument is not present, otherwise second argument should be a constant
4141 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4144 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4145 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4146 address from @var{register} and add @var{offset} to it}.
4148 @section @code{.cfi_def_cfa_register @var{register}}
4149 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4150 now on @var{register} will be used instead of the old one. Offset
4153 @section @code{.cfi_def_cfa_offset @var{offset}}
4154 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4155 remains the same, but @var{offset} is new. Note that it is the
4156 absolute offset that will be added to a defined register to compute
4159 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4160 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4161 value that is added/substracted from the previous offset.
4163 @section @code{.cfi_offset @var{register}, @var{offset}}
4164 Previous value of @var{register} is saved at offset @var{offset} from
4167 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4168 Previous value of @var{register} is saved at offset @var{offset} from
4169 the current CFA register. This is transformed to @code{.cfi_offset}
4170 using the known displacement of the CFA register from the CFA.
4171 This is often easier to use, because the number will match the
4172 code it's annotating.
4174 @section @code{.cfi_register @var{register1}, @var{register2}}
4175 Previous value of @var{register1} is saved in register @var{register2}.
4177 @section @code{.cfi_restore @var{register}}
4178 @code{.cfi_restore} says that the rule for @var{register} is now the
4179 same as it was at the beginning of the function, after all initial
4180 instruction added by @code{.cfi_startproc} were executed.
4182 @section @code{.cfi_undefined @var{register}}
4183 From now on the previous value of @var{register} can't be restored anymore.
4185 @section @code{.cfi_same_value @var{register}}
4186 Current value of @var{register} is the same like in the previous frame,
4187 i.e. no restoration needed.
4189 @section @code{.cfi_remember_state},
4190 First save all current rules for all registers by @code{.cfi_remember_state},
4191 then totally screw them up by subsequent @code{.cfi_*} directives and when
4192 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4193 the previous saved state.
4195 @section @code{.cfi_return_column @var{register}}
4196 Change return column @var{register}, i.e. the return address is either
4197 directly in @var{register} or can be accessed by rules for @var{register}.
4199 @section @code{.cfi_signal_frame}
4200 Mark current function as signal trampoline.
4202 @section @code{.cfi_window_save}
4203 SPARC register window has been saved.
4205 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4206 Allows the user to add arbitrary bytes to the unwind info. One
4207 might use this to add OS-specific CFI opcodes, or generic CFI
4208 opcodes that GAS does not yet support.
4210 @node LNS directives
4211 @section @code{.file @var{fileno} @var{filename}}
4212 @cindex @code{file} directive
4213 When emitting dwarf2 line number information @code{.file} assigns filenames
4214 to the @code{.debug_line} file name table. The @var{fileno} operand should
4215 be a unique positive integer to use as the index of the entry in the table.
4216 The @var{filename} operand is a C string literal.
4218 The detail of filename indices is exposed to the user because the filename
4219 table is shared with the @code{.debug_info} section of the dwarf2 debugging
4220 information, and thus the user must know the exact indices that table
4223 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
4224 @cindex @code{loc} directive
4225 The @code{.loc} directive will add row to the @code{.debug_line} line
4226 number matrix corresponding to the immediately following assembly
4227 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
4228 arguments will be applied to the @code{.debug_line} state machine before
4231 The @var{options} are a sequence of the following tokens in any order:
4235 This option will set the @code{basic_block} register in the
4236 @code{.debug_line} state machine to @code{true}.
4239 This option will set the @code{prologue_end} register in the
4240 @code{.debug_line} state machine to @code{true}.
4242 @item epilogue_begin
4243 This option will set the @code{epilogue_begin} register in the
4244 @code{.debug_line} state machine to @code{true}.
4246 @item is_stmt @var{value}
4247 This option will set the @code{is_stmt} register in the
4248 @code{.debug_line} state machine to @code{value}, which must be
4251 @item isa @var{value}
4252 This directive will set the @code{isa} register in the @code{.debug_line}
4253 state machine to @var{value}, which must be an unsigned integer.
4257 @section @code{.loc_mark_blocks @var{enable}}
4258 @cindex @code{loc_mark_blocks} directive
4259 The @code{.loc_mark_blocks} directive makes the assembler emit an entry
4260 to the @code{.debug_line} line number matrix with the @code{basic_block}
4261 register in the state machine set whenever a code label is seen.
4262 The @var{enable} argument should be either 1 or 0, to enable or disable
4263 this function respectively.
4266 @section @code{.data @var{subsection}}
4268 @cindex @code{data} directive
4269 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4270 end of the data subsection numbered @var{subsection} (which is an
4271 absolute expression). If @var{subsection} is omitted, it defaults
4276 @section @code{.def @var{name}}
4278 @cindex @code{def} directive
4279 @cindex COFF symbols, debugging
4280 @cindex debugging COFF symbols
4281 Begin defining debugging information for a symbol @var{name}; the
4282 definition extends until the @code{.endef} directive is encountered.
4285 This directive is only observed when @command{@value{AS}} is configured for COFF
4286 format output; when producing @code{b.out}, @samp{.def} is recognized,
4293 @section @code{.desc @var{symbol}, @var{abs-expression}}
4295 @cindex @code{desc} directive
4296 @cindex COFF symbol descriptor
4297 @cindex symbol descriptor, COFF
4298 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4299 to the low 16 bits of an absolute expression.
4302 The @samp{.desc} directive is not available when @command{@value{AS}} is
4303 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4304 object format. For the sake of compatibility, @command{@value{AS}} accepts
4305 it, but produces no output, when configured for COFF.
4311 @section @code{.dim}
4313 @cindex @code{dim} directive
4314 @cindex COFF auxiliary symbol information
4315 @cindex auxiliary symbol information, COFF
4316 This directive is generated by compilers to include auxiliary debugging
4317 information in the symbol table. It is only permitted inside
4318 @code{.def}/@code{.endef} pairs.
4321 @samp{.dim} is only meaningful when generating COFF format output; when
4322 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4328 @section @code{.double @var{flonums}}
4330 @cindex @code{double} directive
4331 @cindex floating point numbers (double)
4332 @code{.double} expects zero or more flonums, separated by commas. It
4333 assembles floating point numbers.
4335 The exact kind of floating point numbers emitted depends on how
4336 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4340 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4341 in @sc{ieee} format.
4346 @section @code{.eject}
4348 @cindex @code{eject} directive
4349 @cindex new page, in listings
4350 @cindex page, in listings
4351 @cindex listing control: new page
4352 Force a page break at this point, when generating assembly listings.
4355 @section @code{.else}
4357 @cindex @code{else} directive
4358 @code{.else} is part of the @command{@value{AS}} support for conditional
4359 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4360 of code to be assembled if the condition for the preceding @code{.if}
4364 @section @code{.elseif}
4366 @cindex @code{elseif} directive
4367 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4368 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4369 @code{.if} block that would otherwise fill the entire @code{.else} section.
4372 @section @code{.end}
4374 @cindex @code{end} directive
4375 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4376 process anything in the file past the @code{.end} directive.
4380 @section @code{.endef}
4382 @cindex @code{endef} directive
4383 This directive flags the end of a symbol definition begun with
4387 @samp{.endef} is only meaningful when generating COFF format output; if
4388 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4389 directive but ignores it.
4394 @section @code{.endfunc}
4395 @cindex @code{endfunc} directive
4396 @code{.endfunc} marks the end of a function specified with @code{.func}.
4399 @section @code{.endif}
4401 @cindex @code{endif} directive
4402 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4403 it marks the end of a block of code that is only assembled
4404 conditionally. @xref{If,,@code{.if}}.
4407 @section @code{.equ @var{symbol}, @var{expression}}
4409 @cindex @code{equ} directive
4410 @cindex assigning values to symbols
4411 @cindex symbols, assigning values to
4412 This directive sets the value of @var{symbol} to @var{expression}.
4413 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4416 The syntax for @code{equ} on the HPPA is
4417 @samp{@var{symbol} .equ @var{expression}}.
4421 The syntax for @code{equ} on the Z80 is
4422 @samp{@var{symbol} equ @var{expression}}.
4423 On the Z80 it is an eror if @var{symbol} is already defined,
4424 but the symbol is not protected from later redefinition.
4425 Compare @ref{Equiv}.
4429 @section @code{.equiv @var{symbol}, @var{expression}}
4430 @cindex @code{equiv} directive
4431 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4432 the assembler will signal an error if @var{symbol} is already defined. Note a
4433 symbol which has been referenced but not actually defined is considered to be
4436 Except for the contents of the error message, this is roughly equivalent to
4443 plus it protects the symbol from later redefinition.
4446 @section @code{.eqv @var{symbol}, @var{expression}}
4447 @cindex @code{eqv} directive
4448 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4449 evaluate the expression or any part of it immediately. Instead each time
4450 the resulting symbol is used in an expression, a snapshot of its current
4454 @section @code{.err}
4455 @cindex @code{err} directive
4456 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4457 message and, unless the @option{-Z} option was used, it will not generate an
4458 object file. This can be used to signal an error in conditionally compiled code.
4461 @section @code{.error "@var{string}"}
4462 @cindex error directive
4464 Similarly to @code{.err}, this directive emits an error, but you can specify a
4465 string that will be emitted as the error message. If you don't specify the
4466 message, it defaults to @code{".error directive invoked in source file"}.
4467 @xref{Errors, ,Error and Warning Messages}.
4470 .error "This code has not been assembled and tested."
4474 @section @code{.exitm}
4475 Exit early from the current macro definition. @xref{Macro}.
4478 @section @code{.extern}
4480 @cindex @code{extern} directive
4481 @code{.extern} is accepted in the source program---for compatibility
4482 with other assemblers---but it is ignored. @command{@value{AS}} treats
4483 all undefined symbols as external.
4486 @section @code{.fail @var{expression}}
4488 @cindex @code{fail} directive
4489 Generates an error or a warning. If the value of the @var{expression} is 500
4490 or more, @command{@value{AS}} will print a warning message. If the value is less
4491 than 500, @command{@value{AS}} will print an error message. The message will
4492 include the value of @var{expression}. This can occasionally be useful inside
4493 complex nested macros or conditional assembly.
4495 @ifclear no-file-dir
4497 @section @code{.file @var{string}}
4499 @cindex @code{file} directive
4500 @cindex logical file name
4501 @cindex file name, logical
4502 @code{.file} tells @command{@value{AS}} that we are about to start a new logical
4503 file. @var{string} is the new file name. In general, the filename is
4504 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4505 to specify an empty file name, you must give the quotes--@code{""}. This
4506 statement may go away in future: it is only recognized to be compatible with
4507 old @command{@value{AS}} programs.
4511 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4513 @cindex @code{fill} directive
4514 @cindex writing patterns in memory
4515 @cindex patterns, writing in memory
4516 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4517 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4518 may be zero or more. @var{Size} may be zero or more, but if it is
4519 more than 8, then it is deemed to have the value 8, compatible with
4520 other people's assemblers. The contents of each @var{repeat} bytes
4521 is taken from an 8-byte number. The highest order 4 bytes are
4522 zero. The lowest order 4 bytes are @var{value} rendered in the
4523 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4524 Each @var{size} bytes in a repetition is taken from the lowest order
4525 @var{size} bytes of this number. Again, this bizarre behavior is
4526 compatible with other people's assemblers.
4528 @var{size} and @var{value} are optional.
4529 If the second comma and @var{value} are absent, @var{value} is
4530 assumed zero. If the first comma and following tokens are absent,
4531 @var{size} is assumed to be 1.
4534 @section @code{.float @var{flonums}}
4536 @cindex floating point numbers (single)
4537 @cindex @code{float} directive
4538 This directive assembles zero or more flonums, separated by commas. It
4539 has the same effect as @code{.single}.
4541 The exact kind of floating point numbers emitted depends on how
4542 @command{@value{AS}} is configured.
4543 @xref{Machine Dependencies}.
4547 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4548 in @sc{ieee} format.
4553 @section @code{.func @var{name}[,@var{label}]}
4554 @cindex @code{func} directive
4555 @code{.func} emits debugging information to denote function @var{name}, and
4556 is ignored unless the file is assembled with debugging enabled.
4557 Only @samp{--gstabs[+]} is currently supported.
4558 @var{label} is the entry point of the function and if omitted @var{name}
4559 prepended with the @samp{leading char} is used.
4560 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4561 All functions are currently defined to have @code{void} return type.
4562 The function must be terminated with @code{.endfunc}.
4565 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4567 @cindex @code{global} directive
4568 @cindex symbol, making visible to linker
4569 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4570 @var{symbol} in your partial program, its value is made available to
4571 other partial programs that are linked with it. Otherwise,
4572 @var{symbol} takes its attributes from a symbol of the same name
4573 from another file linked into the same program.
4575 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4576 compatibility with other assemblers.
4579 On the HPPA, @code{.global} is not always enough to make it accessible to other
4580 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4581 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4586 @section @code{.hidden @var{names}}
4588 @cindex @code{hidden} directive
4590 This is one of the ELF visibility directives. The other two are
4591 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4592 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4594 This directive overrides the named symbols default visibility (which is set by
4595 their binding: local, global or weak). The directive sets the visibility to
4596 @code{hidden} which means that the symbols are not visible to other components.
4597 Such symbols are always considered to be @code{protected} as well.
4601 @section @code{.hword @var{expressions}}
4603 @cindex @code{hword} directive
4604 @cindex integers, 16-bit
4605 @cindex numbers, 16-bit
4606 @cindex sixteen bit integers
4607 This expects zero or more @var{expressions}, and emits
4608 a 16 bit number for each.
4611 This directive is a synonym for @samp{.short}; depending on the target
4612 architecture, it may also be a synonym for @samp{.word}.
4616 This directive is a synonym for @samp{.short}.
4619 This directive is a synonym for both @samp{.short} and @samp{.word}.
4624 @section @code{.ident}
4626 @cindex @code{ident} directive
4628 This directive is used by some assemblers to place tags in object files. The
4629 behavior of this directive varies depending on the target. When using the
4630 a.out object file format, @command{@value{AS}} simply accepts the directive for
4631 source-file compatibility with existing assemblers, but does not emit anything
4632 for it. When using COFF, comments are emitted to the @code{.comment} or
4633 @code{.rdata} section, depending on the target. When using ELF, comments are
4634 emitted to the @code{.comment} section.
4637 @section @code{.if @var{absolute expression}}
4639 @cindex conditional assembly
4640 @cindex @code{if} directive
4641 @code{.if} marks the beginning of a section of code which is only
4642 considered part of the source program being assembled if the argument
4643 (which must be an @var{absolute expression}) is non-zero. The end of
4644 the conditional section of code must be marked by @code{.endif}
4645 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4646 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4647 If you have several conditions to check, @code{.elseif} may be used to avoid
4648 nesting blocks if/else within each subsequent @code{.else} block.
4650 The following variants of @code{.if} are also supported:
4652 @cindex @code{ifdef} directive
4653 @item .ifdef @var{symbol}
4654 Assembles the following section of code if the specified @var{symbol}
4655 has been defined. Note a symbol which has been referenced but not yet defined
4656 is considered to be undefined.
4658 @cindex @code{ifb} directive
4659 @item .ifb @var{text}
4660 Assembles the following section of code if the operand is blank (empty).
4662 @cindex @code{ifc} directive
4663 @item .ifc @var{string1},@var{string2}
4664 Assembles the following section of code if the two strings are the same. The
4665 strings may be optionally quoted with single quotes. If they are not quoted,
4666 the first string stops at the first comma, and the second string stops at the
4667 end of the line. Strings which contain whitespace should be quoted. The
4668 string comparison is case sensitive.
4670 @cindex @code{ifeq} directive
4671 @item .ifeq @var{absolute expression}
4672 Assembles the following section of code if the argument is zero.
4674 @cindex @code{ifeqs} directive
4675 @item .ifeqs @var{string1},@var{string2}
4676 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4678 @cindex @code{ifge} directive
4679 @item .ifge @var{absolute expression}
4680 Assembles the following section of code if the argument is greater than or
4683 @cindex @code{ifgt} directive
4684 @item .ifgt @var{absolute expression}
4685 Assembles the following section of code if the argument is greater than zero.
4687 @cindex @code{ifle} directive
4688 @item .ifle @var{absolute expression}
4689 Assembles the following section of code if the argument is less than or equal
4692 @cindex @code{iflt} directive
4693 @item .iflt @var{absolute expression}
4694 Assembles the following section of code if the argument is less than zero.
4696 @cindex @code{ifnb} directive
4697 @item .ifnb @var{text}
4698 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
4699 following section of code if the operand is non-blank (non-empty).
4701 @cindex @code{ifnc} directive
4702 @item .ifnc @var{string1},@var{string2}.
4703 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4704 following section of code if the two strings are not the same.
4706 @cindex @code{ifndef} directive
4707 @cindex @code{ifnotdef} directive
4708 @item .ifndef @var{symbol}
4709 @itemx .ifnotdef @var{symbol}
4710 Assembles the following section of code if the specified @var{symbol}
4711 has not been defined. Both spelling variants are equivalent. Note a symbol
4712 which has been referenced but not yet defined is considered to be undefined.
4714 @cindex @code{ifne} directive
4715 @item .ifne @var{absolute expression}
4716 Assembles the following section of code if the argument is not equal to zero
4717 (in other words, this is equivalent to @code{.if}).
4719 @cindex @code{ifnes} directive
4720 @item .ifnes @var{string1},@var{string2}
4721 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4722 following section of code if the two strings are not the same.
4726 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4728 @cindex @code{incbin} directive
4729 @cindex binary files, including
4730 The @code{incbin} directive includes @var{file} verbatim at the current
4731 location. You can control the search paths used with the @samp{-I} command-line
4732 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4735 The @var{skip} argument skips a number of bytes from the start of the
4736 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4737 read. Note that the data is not aligned in any way, so it is the user's
4738 responsibility to make sure that proper alignment is provided both before and
4739 after the @code{incbin} directive.
4742 @section @code{.include "@var{file}"}
4744 @cindex @code{include} directive
4745 @cindex supporting files, including
4746 @cindex files, including
4747 This directive provides a way to include supporting files at specified
4748 points in your source program. The code from @var{file} is assembled as
4749 if it followed the point of the @code{.include}; when the end of the
4750 included file is reached, assembly of the original file continues. You
4751 can control the search paths used with the @samp{-I} command-line option
4752 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4756 @section @code{.int @var{expressions}}
4758 @cindex @code{int} directive
4759 @cindex integers, 32-bit
4760 Expect zero or more @var{expressions}, of any section, separated by commas.
4761 For each expression, emit a number that, at run time, is the value of that
4762 expression. The byte order and bit size of the number depends on what kind
4763 of target the assembly is for.
4767 On most forms of the H8/300, @code{.int} emits 16-bit
4768 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4775 @section @code{.internal @var{names}}
4777 @cindex @code{internal} directive
4779 This is one of the ELF visibility directives. The other two are
4780 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4781 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4783 This directive overrides the named symbols default visibility (which is set by
4784 their binding: local, global or weak). The directive sets the visibility to
4785 @code{internal} which means that the symbols are considered to be @code{hidden}
4786 (i.e., not visible to other components), and that some extra, processor specific
4787 processing must also be performed upon the symbols as well.
4791 @section @code{.irp @var{symbol},@var{values}}@dots{}
4793 @cindex @code{irp} directive
4794 Evaluate a sequence of statements assigning different values to @var{symbol}.
4795 The sequence of statements starts at the @code{.irp} directive, and is
4796 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4797 set to @var{value}, and the sequence of statements is assembled. If no
4798 @var{value} is listed, the sequence of statements is assembled once, with
4799 @var{symbol} set to the null string. To refer to @var{symbol} within the
4800 sequence of statements, use @var{\symbol}.
4802 For example, assembling
4810 is equivalent to assembling
4818 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
4821 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4823 @cindex @code{irpc} directive
4824 Evaluate a sequence of statements assigning different values to @var{symbol}.
4825 The sequence of statements starts at the @code{.irpc} directive, and is
4826 terminated by an @code{.endr} directive. For each character in @var{value},
4827 @var{symbol} is set to the character, and the sequence of statements is
4828 assembled. If no @var{value} is listed, the sequence of statements is
4829 assembled once, with @var{symbol} set to the null string. To refer to
4830 @var{symbol} within the sequence of statements, use @var{\symbol}.
4832 For example, assembling
4840 is equivalent to assembling
4848 For some caveats with the spelling of @var{symbol}, see also the discussion
4852 @section @code{.lcomm @var{symbol} , @var{length}}
4854 @cindex @code{lcomm} directive
4855 @cindex local common symbols
4856 @cindex symbols, local common
4857 Reserve @var{length} (an absolute expression) bytes for a local common
4858 denoted by @var{symbol}. The section and value of @var{symbol} are
4859 those of the new local common. The addresses are allocated in the bss
4860 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4861 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4862 not visible to @code{@value{LD}}.
4865 Some targets permit a third argument to be used with @code{.lcomm}. This
4866 argument specifies the desired alignment of the symbol in the bss section.
4870 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4871 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4875 @section @code{.lflags}
4877 @cindex @code{lflags} directive (ignored)
4878 @command{@value{AS}} accepts this directive, for compatibility with other
4879 assemblers, but ignores it.
4881 @ifclear no-line-dir
4883 @section @code{.line @var{line-number}}
4885 @cindex @code{line} directive
4889 @section @code{.ln @var{line-number}}
4891 @cindex @code{ln} directive
4893 @cindex logical line number
4895 Change the logical line number. @var{line-number} must be an absolute
4896 expression. The next line has that logical line number. Therefore any other
4897 statements on the current line (after a statement separator character) are
4898 reported as on logical line number @var{line-number} @minus{} 1. One day
4899 @command{@value{AS}} will no longer support this directive: it is recognized only
4900 for compatibility with existing assembler programs.
4904 @ifclear no-line-dir
4905 Even though this is a directive associated with the @code{a.out} or
4906 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
4907 when producing COFF output, and treats @samp{.line} as though it
4908 were the COFF @samp{.ln} @emph{if} it is found outside a
4909 @code{.def}/@code{.endef} pair.
4911 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4912 used by compilers to generate auxiliary symbol information for
4917 @section @code{.linkonce [@var{type}]}
4919 @cindex @code{linkonce} directive
4920 @cindex common sections
4921 Mark the current section so that the linker only includes a single copy of it.
4922 This may be used to include the same section in several different object files,
4923 but ensure that the linker will only include it once in the final output file.
4924 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4925 Duplicate sections are detected based on the section name, so it should be
4928 This directive is only supported by a few object file formats; as of this
4929 writing, the only object file format which supports it is the Portable
4930 Executable format used on Windows NT.
4932 The @var{type} argument is optional. If specified, it must be one of the
4933 following strings. For example:
4937 Not all types may be supported on all object file formats.
4941 Silently discard duplicate sections. This is the default.
4944 Warn if there are duplicate sections, but still keep only one copy.
4947 Warn if any of the duplicates have different sizes.
4950 Warn if any of the duplicates do not have exactly the same contents.
4954 @section @code{.ln @var{line-number}}
4956 @cindex @code{ln} directive
4957 @ifclear no-line-dir
4958 @samp{.ln} is a synonym for @samp{.line}.
4961 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
4962 must be an absolute expression. The next line has that logical
4963 line number, so any other statements on the current line (after a
4964 statement separator character @code{;}) are reported as on logical
4965 line number @var{line-number} @minus{} 1.
4968 This directive is accepted, but ignored, when @command{@value{AS}} is
4969 configured for @code{b.out}; its effect is only associated with COFF
4975 @section @code{.mri @var{val}}
4977 @cindex @code{mri} directive
4978 @cindex MRI mode, temporarily
4979 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
4980 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
4981 affects code assembled until the next @code{.mri} directive, or until the end
4982 of the file. @xref{M, MRI mode, MRI mode}.
4985 @section @code{.list}
4987 @cindex @code{list} directive
4988 @cindex listing control, turning on
4989 Control (in conjunction with the @code{.nolist} directive) whether or
4990 not assembly listings are generated. These two directives maintain an
4991 internal counter (which is zero initially). @code{.list} increments the
4992 counter, and @code{.nolist} decrements it. Assembly listings are
4993 generated whenever the counter is greater than zero.
4995 By default, listings are disabled. When you enable them (with the
4996 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4997 the initial value of the listing counter is one.
5000 @section @code{.long @var{expressions}}
5002 @cindex @code{long} directive
5003 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5006 @c no one seems to know what this is for or whether this description is
5007 @c what it really ought to do
5009 @section @code{.lsym @var{symbol}, @var{expression}}
5011 @cindex @code{lsym} directive
5012 @cindex symbol, not referenced in assembly
5013 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5014 the hash table, ensuring it cannot be referenced by name during the
5015 rest of the assembly. This sets the attributes of the symbol to be
5016 the same as the expression value:
5018 @var{other} = @var{descriptor} = 0
5019 @var{type} = @r{(section of @var{expression})}
5020 @var{value} = @var{expression}
5023 The new symbol is not flagged as external.
5027 @section @code{.macro}
5030 The commands @code{.macro} and @code{.endm} allow you to define macros that
5031 generate assembly output. For example, this definition specifies a macro
5032 @code{sum} that puts a sequence of numbers into memory:
5035 .macro sum from=0, to=5
5044 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5056 @item .macro @var{macname}
5057 @itemx .macro @var{macname} @var{macargs} @dots{}
5058 @cindex @code{macro} directive
5059 Begin the definition of a macro called @var{macname}. If your macro
5060 definition requires arguments, specify their names after the macro name,
5061 separated by commas or spaces. You can qualify the macro argument to
5062 indicate whether all invocations must specify a non-blank value (through
5063 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5064 (through @samp{:@code{vararg}}). You can supply a default value for any
5065 macro argument by following the name with @samp{=@var{deflt}}. You
5066 cannot define two macros with the same @var{macname} unless it has been
5067 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5068 definitions. For example, these are all valid @code{.macro} statements:
5072 Begin the definition of a macro called @code{comm}, which takes no
5075 @item .macro plus1 p, p1
5076 @itemx .macro plus1 p p1
5077 Either statement begins the definition of a macro called @code{plus1},
5078 which takes two arguments; within the macro definition, write
5079 @samp{\p} or @samp{\p1} to evaluate the arguments.
5081 @item .macro reserve_str p1=0 p2
5082 Begin the definition of a macro called @code{reserve_str}, with two
5083 arguments. The first argument has a default value, but not the second.
5084 After the definition is complete, you can call the macro either as
5085 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5086 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5087 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5088 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5090 @item .macro m p1:req, p2=0, p3:vararg
5091 Begin the definition of a macro called @code{m}, with at least three
5092 arguments. The first argument must always have a value specified, but
5093 not the second, which instead has a default value. The third formal
5094 will get assigned all remaining arguments specified at invocation time.
5096 When you call a macro, you can specify the argument values either by
5097 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5098 @samp{sum to=17, from=9}.
5102 Note that since each of the @var{macargs} can be an identifier exactly
5103 as any other one permitted by the target architecture, there may be
5104 occasional problems if the target hand-crafts special meanings to certain
5105 characters when they occur in a special position. For example, if the colon
5106 (@code{:}) is generally permitted to be part of a symbol name, but the
5107 architecture specific code special-cases it when occurring as the final
5108 character of a symbol (to denote a label), then the macro parameter
5109 replacement code will have no way of knowing that and consider the whole
5110 construct (including the colon) an identifier, and check only this
5111 identifier for being the subject to parameter substitution. So for example
5112 this macro definition:
5120 might not work as expected. Invoking @samp{label foo} might not create a label
5121 called @samp{foo} but instead just insert the text @samp{\l:} into the
5122 assembler source, probably generating an error about an unrecognised
5125 Similarly problems might occur with the period character (@samp{.})
5126 which is often allowed inside opcode names (and hence identifier names). So
5127 for example constructing a macro to build an opcode from a base name and a
5128 length specifier like this:
5131 .macro opcode base length
5136 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5137 instruction but instead generate some kind of error as the assembler tries to
5138 interpret the text @samp{\base.\length}.
5140 There are several possible ways around this problem:
5143 @item Insert white space
5144 If it is possible to use white space characters then this is the simplest
5153 @item Use @samp{\()}
5154 The string @samp{\()} can be used to separate the end of a macro argument from
5155 the following text. eg:
5158 .macro opcode base length
5163 @item Use the alternate macro syntax mode
5164 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5165 used as a separator. eg:
5175 Note: this problem of correctly identifying string parameters to pseudo ops
5176 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5177 and @code{.irpc} (@pxref{Irpc}) as well.
5180 @cindex @code{endm} directive
5181 Mark the end of a macro definition.
5184 @cindex @code{exitm} directive
5185 Exit early from the current macro definition.
5187 @cindex number of macros executed
5188 @cindex macros, count executed
5190 @command{@value{AS}} maintains a counter of how many macros it has
5191 executed in this pseudo-variable; you can copy that number to your
5192 output with @samp{\@@}, but @emph{only within a macro definition}.
5194 @item LOCAL @var{name} [ , @dots{} ]
5195 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5196 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5197 @xref{Altmacro,,@code{.altmacro}}.
5201 @section @code{.altmacro}
5202 Enable alternate macro mode, enabling:
5205 @item LOCAL @var{name} [ , @dots{} ]
5206 One additional directive, @code{LOCAL}, is available. It is used to
5207 generate a string replacement for each of the @var{name} arguments, and
5208 replace any instances of @var{name} in each macro expansion. The
5209 replacement string is unique in the assembly, and different for each
5210 separate macro expansion. @code{LOCAL} allows you to write macros that
5211 define symbols, without fear of conflict between separate macro expansions.
5213 @item String delimiters
5214 You can write strings delimited in these other ways besides
5215 @code{"@var{string}"}:
5218 @item '@var{string}'
5219 You can delimit strings with single-quote characters.
5221 @item <@var{string}>
5222 You can delimit strings with matching angle brackets.
5225 @item single-character string escape
5226 To include any single character literally in a string (even if the
5227 character would otherwise have some special meaning), you can prefix the
5228 character with @samp{!} (an exclamation mark). For example, you can
5229 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
5231 @item Expression results as strings
5232 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
5233 and use the result as a string.
5237 @section @code{.noaltmacro}
5238 Disable alternate macro mode. @xref{Altmacro}.
5241 @section @code{.nolist}
5243 @cindex @code{nolist} directive
5244 @cindex listing control, turning off
5245 Control (in conjunction with the @code{.list} directive) whether or
5246 not assembly listings are generated. These two directives maintain an
5247 internal counter (which is zero initially). @code{.list} increments the
5248 counter, and @code{.nolist} decrements it. Assembly listings are
5249 generated whenever the counter is greater than zero.
5252 @section @code{.octa @var{bignums}}
5254 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5255 @cindex @code{octa} directive
5256 @cindex integer, 16-byte
5257 @cindex sixteen byte integer
5258 This directive expects zero or more bignums, separated by commas. For each
5259 bignum, it emits a 16-byte integer.
5261 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5262 hence @emph{octa}-word for 16 bytes.
5265 @section @code{.org @var{new-lc} , @var{fill}}
5267 @cindex @code{org} directive
5268 @cindex location counter, advancing
5269 @cindex advancing location counter
5270 @cindex current address, advancing
5271 Advance the location counter of the current section to
5272 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5273 expression with the same section as the current subsection. That is,
5274 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5275 wrong section, the @code{.org} directive is ignored. To be compatible
5276 with former assemblers, if the section of @var{new-lc} is absolute,
5277 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5278 is the same as the current subsection.
5280 @code{.org} may only increase the location counter, or leave it
5281 unchanged; you cannot use @code{.org} to move the location counter
5284 @c double negative used below "not undefined" because this is a specific
5285 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5286 @c section. doc@cygnus.com 18feb91
5287 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5288 may not be undefined. If you really detest this restriction we eagerly await
5289 a chance to share your improved assembler.
5291 Beware that the origin is relative to the start of the section, not
5292 to the start of the subsection. This is compatible with other
5293 people's assemblers.
5295 When the location counter (of the current subsection) is advanced, the
5296 intervening bytes are filled with @var{fill} which should be an
5297 absolute expression. If the comma and @var{fill} are omitted,
5298 @var{fill} defaults to zero.
5301 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5303 @cindex padding the location counter given a power of two
5304 @cindex @code{p2align} directive
5305 Pad the location counter (in the current subsection) to a particular
5306 storage boundary. The first expression (which must be absolute) is the
5307 number of low-order zero bits the location counter must have after
5308 advancement. For example @samp{.p2align 3} advances the location
5309 counter until it a multiple of 8. If the location counter is already a
5310 multiple of 8, no change is needed.
5312 The second expression (also absolute) gives the fill value to be stored in the
5313 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5314 padding bytes are normally zero. However, on some systems, if the section is
5315 marked as containing code and the fill value is omitted, the space is filled
5316 with no-op instructions.
5318 The third expression is also absolute, and is also optional. If it is present,
5319 it is the maximum number of bytes that should be skipped by this alignment
5320 directive. If doing the alignment would require skipping more bytes than the
5321 specified maximum, then the alignment is not done at all. You can omit the
5322 fill value (the second argument) entirely by simply using two commas after the
5323 required alignment; this can be useful if you want the alignment to be filled
5324 with no-op instructions when appropriate.
5326 @cindex @code{p2alignw} directive
5327 @cindex @code{p2alignl} directive
5328 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5329 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5330 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5331 fill pattern as a four byte longword value. For example, @code{.p2alignw
5332 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5333 filled in with the value 0x368d (the exact placement of the bytes depends upon
5334 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5339 @section @code{.previous}
5341 @cindex @code{previous} directive
5342 @cindex Section Stack
5343 This is one of the ELF section stack manipulation directives. The others are
5344 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5345 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5346 (@pxref{PopSection}).
5348 This directive swaps the current section (and subsection) with most recently
5349 referenced section/subsection pair prior to this one. Multiple
5350 @code{.previous} directives in a row will flip between two sections (and their
5351 subsections). For example:
5363 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5369 # Now in section A subsection 1
5373 # Now in section B subsection 0
5376 # Now in section B subsection 1
5379 # Now in section B subsection 0
5383 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5384 section B and 0x9abc into subsection 1 of section B.
5386 In terms of the section stack, this directive swaps the current section with
5387 the top section on the section stack.
5392 @section @code{.popsection}
5394 @cindex @code{popsection} directive
5395 @cindex Section Stack
5396 This is one of the ELF section stack manipulation directives. The others are
5397 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5398 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5401 This directive replaces the current section (and subsection) with the top
5402 section (and subsection) on the section stack. This section is popped off the
5407 @section @code{.print @var{string}}
5409 @cindex @code{print} directive
5410 @command{@value{AS}} will print @var{string} on the standard output during
5411 assembly. You must put @var{string} in double quotes.
5415 @section @code{.protected @var{names}}
5417 @cindex @code{protected} directive
5419 This is one of the ELF visibility directives. The other two are
5420 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5422 This directive overrides the named symbols default visibility (which is set by
5423 their binding: local, global or weak). The directive sets the visibility to
5424 @code{protected} which means that any references to the symbols from within the
5425 components that defines them must be resolved to the definition in that
5426 component, even if a definition in another component would normally preempt
5431 @section @code{.psize @var{lines} , @var{columns}}
5433 @cindex @code{psize} directive
5434 @cindex listing control: paper size
5435 @cindex paper size, for listings
5436 Use this directive to declare the number of lines---and, optionally, the
5437 number of columns---to use for each page, when generating listings.
5439 If you do not use @code{.psize}, listings use a default line-count
5440 of 60. You may omit the comma and @var{columns} specification; the
5441 default width is 200 columns.
5443 @command{@value{AS}} generates formfeeds whenever the specified number of
5444 lines is exceeded (or whenever you explicitly request one, using
5447 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5448 those explicitly specified with @code{.eject}.
5451 @section @code{.purgem @var{name}}
5453 @cindex @code{purgem} directive
5454 Undefine the macro @var{name}, so that later uses of the string will not be
5455 expanded. @xref{Macro}.
5459 @section @code{.pushsection @var{name} , @var{subsection}}
5461 @cindex @code{pushsection} directive
5462 @cindex Section Stack
5463 This is one of the ELF section stack manipulation directives. The others are
5464 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5465 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5468 This directive pushes the current section (and subsection) onto the
5469 top of the section stack, and then replaces the current section and
5470 subsection with @code{name} and @code{subsection}.
5474 @section @code{.quad @var{bignums}}
5476 @cindex @code{quad} directive
5477 @code{.quad} expects zero or more bignums, separated by commas. For
5478 each bignum, it emits
5480 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5481 warning message; and just takes the lowest order 8 bytes of the bignum.
5482 @cindex eight-byte integer
5483 @cindex integer, 8-byte
5485 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5486 hence @emph{quad}-word for 8 bytes.
5489 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5490 warning message; and just takes the lowest order 16 bytes of the bignum.
5491 @cindex sixteen-byte integer
5492 @cindex integer, 16-byte
5496 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5498 @cindex @code{reloc} directive
5499 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5500 @var{expression}. If @var{offset} is a number, the relocation is generated in
5501 the current section. If @var{offset} is an expression that resolves to a
5502 symbol plus offset, the relocation is generated in the given symbol's section.
5503 @var{expression}, if present, must resolve to a symbol plus addend or to an
5504 absolute value, but note that not all targets support an addend. e.g. ELF REL
5505 targets such as i386 store an addend in the section contents rather than in the
5506 relocation. This low level interface does not support addends stored in the
5510 @section @code{.rept @var{count}}
5512 @cindex @code{rept} directive
5513 Repeat the sequence of lines between the @code{.rept} directive and the next
5514 @code{.endr} directive @var{count} times.
5516 For example, assembling
5524 is equivalent to assembling
5533 @section @code{.sbttl "@var{subheading}"}
5535 @cindex @code{sbttl} directive
5536 @cindex subtitles for listings
5537 @cindex listing control: subtitle
5538 Use @var{subheading} as the title (third line, immediately after the
5539 title line) when generating assembly listings.
5541 This directive affects subsequent pages, as well as the current page if
5542 it appears within ten lines of the top of a page.
5546 @section @code{.scl @var{class}}
5548 @cindex @code{scl} directive
5549 @cindex symbol storage class (COFF)
5550 @cindex COFF symbol storage class
5551 Set the storage-class value for a symbol. This directive may only be
5552 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5553 whether a symbol is static or external, or it may record further
5554 symbolic debugging information.
5557 The @samp{.scl} directive is primarily associated with COFF output; when
5558 configured to generate @code{b.out} output format, @command{@value{AS}}
5559 accepts this directive but ignores it.
5565 @section @code{.section @var{name}}
5567 @cindex named section
5568 Use the @code{.section} directive to assemble the following code into a section
5571 This directive is only supported for targets that actually support arbitrarily
5572 named sections; on @code{a.out} targets, for example, it is not accepted, even
5573 with a standard @code{a.out} section name.
5577 @c only print the extra heading if both COFF and ELF are set
5578 @subheading COFF Version
5581 @cindex @code{section} directive (COFF version)
5582 For COFF targets, the @code{.section} directive is used in one of the following
5586 .section @var{name}[, "@var{flags}"]
5587 .section @var{name}[, @var{subsegment}]
5590 If the optional argument is quoted, it is taken as flags to use for the
5591 section. Each flag is a single character. The following flags are recognized:
5594 bss section (uninitialized data)
5596 section is not loaded
5606 shared section (meaningful for PE targets)
5608 ignored. (For compatibility with the ELF version)
5611 If no flags are specified, the default flags depend upon the section name. If
5612 the section name is not recognized, the default will be for the section to be
5613 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5614 from the section, rather than adding them, so if they are used on their own it
5615 will be as if no flags had been specified at all.
5617 If the optional argument to the @code{.section} directive is not quoted, it is
5618 taken as a subsegment number (@pxref{Sub-Sections}).
5623 @c only print the extra heading if both COFF and ELF are set
5624 @subheading ELF Version
5627 @cindex Section Stack
5628 This is one of the ELF section stack manipulation directives. The others are
5629 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5630 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5631 @code{.previous} (@pxref{Previous}).
5633 @cindex @code{section} directive (ELF version)
5634 For ELF targets, the @code{.section} directive is used like this:
5637 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
5640 The optional @var{flags} argument is a quoted string which may contain any
5641 combination of the following characters:
5644 section is allocatable
5648 section is executable
5650 section is mergeable
5652 section contains zero terminated strings
5654 section is a member of a section group
5656 section is used for thread-local-storage
5659 The optional @var{type} argument may contain one of the following constants:
5662 section contains data
5664 section does not contain data (i.e., section only occupies space)
5666 section contains data which is used by things other than the program
5668 section contains an array of pointers to init functions
5670 section contains an array of pointers to finish functions
5671 @item @@preinit_array
5672 section contains an array of pointers to pre-init functions
5675 Many targets only support the first three section types.
5677 Note on targets where the @code{@@} character is the start of a comment (eg
5678 ARM) then another character is used instead. For example the ARM port uses the
5681 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5682 be specified as well as an extra argument---@var{entsize}---like this:
5685 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5688 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5689 constants, each @var{entsize} octets long. Sections with both @code{M} and
5690 @code{S} must contain zero terminated strings where each character is
5691 @var{entsize} bytes long. The linker may remove duplicates within sections with
5692 the same name, same entity size and same flags. @var{entsize} must be an
5693 absolute expression.
5695 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5696 be present along with an additional field like this:
5699 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5702 The @var{GroupName} field specifies the name of the section group to which this
5703 particular section belongs. The optional linkage field can contain:
5706 indicates that only one copy of this section should be retained
5711 Note: if both the @var{M} and @var{G} flags are present then the fields for
5712 the Merge flag should come first, like this:
5715 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5718 If no flags are specified, the default flags depend upon the section name. If
5719 the section name is not recognized, the default will be for the section to have
5720 none of the above flags: it will not be allocated in memory, nor writable, nor
5721 executable. The section will contain data.
5723 For ELF targets, the assembler supports another type of @code{.section}
5724 directive for compatibility with the Solaris assembler:
5727 .section "@var{name}"[, @var{flags}...]
5730 Note that the section name is quoted. There may be a sequence of comma
5734 section is allocatable
5738 section is executable
5740 section is used for thread local storage
5743 This directive replaces the current section and subsection. See the
5744 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
5745 some examples of how this directive and the other section stack directives
5751 @section @code{.set @var{symbol}, @var{expression}}
5753 @cindex @code{set} directive
5754 @cindex symbol value, setting
5755 Set the value of @var{symbol} to @var{expression}. This
5756 changes @var{symbol}'s value and type to conform to
5757 @var{expression}. If @var{symbol} was flagged as external, it remains
5758 flagged (@pxref{Symbol Attributes}).
5760 You may @code{.set} a symbol many times in the same assembly.
5762 If you @code{.set} a global symbol, the value stored in the object
5763 file is the last value stored into it.
5766 The syntax for @code{set} on the HPPA is
5767 @samp{@var{symbol} .set @var{expression}}.
5771 On Z80 @code{set} is a real instruction, use
5772 @samp{@var{symbol} defl @var{expression}} instead.
5776 @section @code{.short @var{expressions}}
5778 @cindex @code{short} directive
5780 @code{.short} is normally the same as @samp{.word}.
5781 @xref{Word,,@code{.word}}.
5783 In some configurations, however, @code{.short} and @code{.word} generate
5784 numbers of different lengths. @xref{Machine Dependencies}.
5788 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5791 This expects zero or more @var{expressions}, and emits
5792 a 16 bit number for each.
5797 @section @code{.single @var{flonums}}
5799 @cindex @code{single} directive
5800 @cindex floating point numbers (single)
5801 This directive assembles zero or more flonums, separated by commas. It
5802 has the same effect as @code{.float}.
5804 The exact kind of floating point numbers emitted depends on how
5805 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5809 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5810 numbers in @sc{ieee} format.
5816 @section @code{.size}
5818 This directive is used to set the size associated with a symbol.
5822 @c only print the extra heading if both COFF and ELF are set
5823 @subheading COFF Version
5826 @cindex @code{size} directive (COFF version)
5827 For COFF targets, the @code{.size} directive is only permitted inside
5828 @code{.def}/@code{.endef} pairs. It is used like this:
5831 .size @var{expression}
5835 @samp{.size} is only meaningful when generating COFF format output; when
5836 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5843 @c only print the extra heading if both COFF and ELF are set
5844 @subheading ELF Version
5847 @cindex @code{size} directive (ELF version)
5848 For ELF targets, the @code{.size} directive is used like this:
5851 .size @var{name} , @var{expression}
5854 This directive sets the size associated with a symbol @var{name}.
5855 The size in bytes is computed from @var{expression} which can make use of label
5856 arithmetic. This directive is typically used to set the size of function
5862 @section @code{.sleb128 @var{expressions}}
5864 @cindex @code{sleb128} directive
5865 @var{sleb128} stands for ``signed little endian base 128.'' This is a
5866 compact, variable length representation of numbers used by the DWARF
5867 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
5869 @ifclear no-space-dir
5871 @section @code{.skip @var{size} , @var{fill}}
5873 @cindex @code{skip} directive
5874 @cindex filling memory
5875 This directive emits @var{size} bytes, each of value @var{fill}. Both
5876 @var{size} and @var{fill} are absolute expressions. If the comma and
5877 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
5881 @section @code{.space @var{size} , @var{fill}}
5883 @cindex @code{space} directive
5884 @cindex filling memory
5885 This directive emits @var{size} bytes, each of value @var{fill}. Both
5886 @var{size} and @var{fill} are absolute expressions. If the comma
5887 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
5892 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
5893 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
5894 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
5895 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
5903 @section @code{.stabd, .stabn, .stabs}
5905 @cindex symbolic debuggers, information for
5906 @cindex @code{stab@var{x}} directives
5907 There are three directives that begin @samp{.stab}.
5908 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
5909 The symbols are not entered in the @command{@value{AS}} hash table: they
5910 cannot be referenced elsewhere in the source file.
5911 Up to five fields are required:
5915 This is the symbol's name. It may contain any character except
5916 @samp{\000}, so is more general than ordinary symbol names. Some
5917 debuggers used to code arbitrarily complex structures into symbol names
5921 An absolute expression. The symbol's type is set to the low 8 bits of
5922 this expression. Any bit pattern is permitted, but @code{@value{LD}}
5923 and debuggers choke on silly bit patterns.
5926 An absolute expression. The symbol's ``other'' attribute is set to the
5927 low 8 bits of this expression.
5930 An absolute expression. The symbol's descriptor is set to the low 16
5931 bits of this expression.
5934 An absolute expression which becomes the symbol's value.
5937 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
5938 or @code{.stabs} statement, the symbol has probably already been created;
5939 you get a half-formed symbol in your object file. This is
5940 compatible with earlier assemblers!
5943 @cindex @code{stabd} directive
5944 @item .stabd @var{type} , @var{other} , @var{desc}
5946 The ``name'' of the symbol generated is not even an empty string.
5947 It is a null pointer, for compatibility. Older assemblers used a
5948 null pointer so they didn't waste space in object files with empty
5951 The symbol's value is set to the location counter,
5952 relocatably. When your program is linked, the value of this symbol
5953 is the address of the location counter when the @code{.stabd} was
5956 @cindex @code{stabn} directive
5957 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
5958 The name of the symbol is set to the empty string @code{""}.
5960 @cindex @code{stabs} directive
5961 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
5962 All five fields are specified.
5968 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
5969 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
5971 @cindex string, copying to object file
5972 @cindex string8, copying to object file
5973 @cindex string16, copying to object file
5974 @cindex string32, copying to object file
5975 @cindex string64, copying to object file
5976 @cindex @code{string} directive
5977 @cindex @code{string8} directive
5978 @cindex @code{string16} directive
5979 @cindex @code{string32} directive
5980 @cindex @code{string64} directive
5982 Copy the characters in @var{str} to the object file. You may specify more than
5983 one string to copy, separated by commas. Unless otherwise specified for a
5984 particular machine, the assembler marks the end of each string with a 0 byte.
5985 You can use any of the escape sequences described in @ref{Strings,,Strings}.
5987 The variants @code{string16}, @code{string32} and @code{string64} differ from
5988 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
5989 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
5990 are stored in target endianness byte order.
5996 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
5997 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6002 @section @code{.struct @var{expression}}
6004 @cindex @code{struct} directive
6005 Switch to the absolute section, and set the section offset to @var{expression},
6006 which must be an absolute expression. You might use this as follows:
6015 This would define the symbol @code{field1} to have the value 0, the symbol
6016 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6017 value 8. Assembly would be left in the absolute section, and you would need to
6018 use a @code{.section} directive of some sort to change to some other section
6019 before further assembly.
6023 @section @code{.subsection @var{name}}
6025 @cindex @code{subsection} directive
6026 @cindex Section Stack
6027 This is one of the ELF section stack manipulation directives. The others are
6028 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6029 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6032 This directive replaces the current subsection with @code{name}. The current
6033 section is not changed. The replaced subsection is put onto the section stack
6034 in place of the then current top of stack subsection.
6039 @section @code{.symver}
6040 @cindex @code{symver} directive
6041 @cindex symbol versioning
6042 @cindex versions of symbols
6043 Use the @code{.symver} directive to bind symbols to specific version nodes
6044 within a source file. This is only supported on ELF platforms, and is
6045 typically used when assembling files to be linked into a shared library.
6046 There are cases where it may make sense to use this in objects to be bound
6047 into an application itself so as to override a versioned symbol from a
6050 For ELF targets, the @code{.symver} directive can be used like this:
6052 .symver @var{name}, @var{name2@@nodename}
6054 If the symbol @var{name} is defined within the file
6055 being assembled, the @code{.symver} directive effectively creates a symbol
6056 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6057 just don't try and create a regular alias is that the @var{@@} character isn't
6058 permitted in symbol names. The @var{name2} part of the name is the actual name
6059 of the symbol by which it will be externally referenced. The name @var{name}
6060 itself is merely a name of convenience that is used so that it is possible to
6061 have definitions for multiple versions of a function within a single source
6062 file, and so that the compiler can unambiguously know which version of a
6063 function is being mentioned. The @var{nodename} portion of the alias should be
6064 the name of a node specified in the version script supplied to the linker when
6065 building a shared library. If you are attempting to override a versioned
6066 symbol from a shared library, then @var{nodename} should correspond to the
6067 nodename of the symbol you are trying to override.
6069 If the symbol @var{name} is not defined within the file being assembled, all
6070 references to @var{name} will be changed to @var{name2@@nodename}. If no
6071 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6074 Another usage of the @code{.symver} directive is:
6076 .symver @var{name}, @var{name2@@@@nodename}
6078 In this case, the symbol @var{name} must exist and be defined within
6079 the file being assembled. It is similar to @var{name2@@nodename}. The
6080 difference is @var{name2@@@@nodename} will also be used to resolve
6081 references to @var{name2} by the linker.
6083 The third usage of the @code{.symver} directive is:
6085 .symver @var{name}, @var{name2@@@@@@nodename}
6087 When @var{name} is not defined within the
6088 file being assembled, it is treated as @var{name2@@nodename}. When
6089 @var{name} is defined within the file being assembled, the symbol
6090 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6095 @section @code{.tag @var{structname}}
6097 @cindex COFF structure debugging
6098 @cindex structure debugging, COFF
6099 @cindex @code{tag} directive
6100 This directive is generated by compilers to include auxiliary debugging
6101 information in the symbol table. It is only permitted inside
6102 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6103 definitions in the symbol table with instances of those structures.
6106 @samp{.tag} is only used when generating COFF format output; when
6107 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6113 @section @code{.text @var{subsection}}
6115 @cindex @code{text} directive
6116 Tells @command{@value{AS}} to assemble the following statements onto the end of
6117 the text subsection numbered @var{subsection}, which is an absolute
6118 expression. If @var{subsection} is omitted, subsection number zero
6122 @section @code{.title "@var{heading}"}
6124 @cindex @code{title} directive
6125 @cindex listing control: title line
6126 Use @var{heading} as the title (second line, immediately after the
6127 source file name and pagenumber) when generating assembly listings.
6129 This directive affects subsequent pages, as well as the current page if
6130 it appears within ten lines of the top of a page.
6134 @section @code{.type}
6136 This directive is used to set the type of a symbol.
6140 @c only print the extra heading if both COFF and ELF are set
6141 @subheading COFF Version
6144 @cindex COFF symbol type
6145 @cindex symbol type, COFF
6146 @cindex @code{type} directive (COFF version)
6147 For COFF targets, this directive is permitted only within
6148 @code{.def}/@code{.endef} pairs. It is used like this:
6154 This records the integer @var{int} as the type attribute of a symbol table
6158 @samp{.type} is associated only with COFF format output; when
6159 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6160 directive but ignores it.
6166 @c only print the extra heading if both COFF and ELF are set
6167 @subheading ELF Version
6170 @cindex ELF symbol type
6171 @cindex symbol type, ELF
6172 @cindex @code{type} directive (ELF version)
6173 For ELF targets, the @code{.type} directive is used like this:
6176 .type @var{name} , @var{type description}
6179 This sets the type of symbol @var{name} to be either a
6180 function symbol or an object symbol. There are five different syntaxes
6181 supported for the @var{type description} field, in order to provide
6182 compatibility with various other assemblers.
6184 Because some of the characters used in these syntaxes (such as @samp{@@} and
6185 @samp{#}) are comment characters for some architectures, some of the syntaxes
6186 below do not work on all architectures. The first variant will be accepted by
6187 the GNU assembler on all architectures so that variant should be used for
6188 maximum portability, if you do not need to assemble your code with other
6191 The syntaxes supported are:
6194 .type <name> STT_FUNCTION
6195 .type <name> STT_OBJECT
6197 .type <name>,#function
6198 .type <name>,#object
6200 .type <name>,@@function
6201 .type <name>,@@object
6203 .type <name>,%function
6204 .type <name>,%object
6206 .type <name>,"function"
6207 .type <name>,"object"
6213 @section @code{.uleb128 @var{expressions}}
6215 @cindex @code{uleb128} directive
6216 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6217 compact, variable length representation of numbers used by the DWARF
6218 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6222 @section @code{.val @var{addr}}
6224 @cindex @code{val} directive
6225 @cindex COFF value attribute
6226 @cindex value attribute, COFF
6227 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6228 records the address @var{addr} as the value attribute of a symbol table
6232 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6233 configured for @code{b.out}, it accepts this directive but ignores it.
6239 @section @code{.version "@var{string}"}
6241 @cindex @code{version} directive
6242 This directive creates a @code{.note} section and places into it an ELF
6243 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6248 @section @code{.vtable_entry @var{table}, @var{offset}}
6250 @cindex @code{vtable_entry} directive
6251 This directive finds or creates a symbol @code{table} and creates a
6252 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6255 @section @code{.vtable_inherit @var{child}, @var{parent}}
6257 @cindex @code{vtable_inherit} directive
6258 This directive finds the symbol @code{child} and finds or creates the symbol
6259 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6260 parent whose addend is the value of the child symbol. As a special case the
6261 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6265 @section @code{.warning "@var{string}"}
6266 @cindex warning directive
6267 Similar to the directive @code{.error}
6268 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6271 @section @code{.weak @var{names}}
6273 @cindex @code{weak} directive
6274 This directive sets the weak attribute on the comma separated list of symbol
6275 @code{names}. If the symbols do not already exist, they will be created.
6277 On COFF targets other than PE, weak symbols are a GNU extension. This
6278 directive sets the weak attribute on the comma separated list of symbol
6279 @code{names}. If the symbols do not already exist, they will be created.
6281 On the PE target, weak symbols are supported natively as weak aliases.
6282 When a weak symbol is created that is not an alias, GAS creates an
6283 alternate symbol to hold the default value.
6286 @section @code{.weakref @var{alias}, @var{target}}
6288 @cindex @code{weakref} directive
6289 This directive creates an alias to the target symbol that enables the symbol to
6290 be referenced with weak-symbol semantics, but without actually making it weak.
6291 If direct references or definitions of the symbol are present, then the symbol
6292 will not be weak, but if all references to it are through weak references, the
6293 symbol will be marked as weak in the symbol table.
6295 The effect is equivalent to moving all references to the alias to a separate
6296 assembly source file, renaming the alias to the symbol in it, declaring the
6297 symbol as weak there, and running a reloadable link to merge the object files
6298 resulting from the assembly of the new source file and the old source file that
6299 had the references to the alias removed.
6301 The alias itself never makes to the symbol table, and is entirely handled
6302 within the assembler.
6305 @section @code{.word @var{expressions}}
6307 @cindex @code{word} directive
6308 This directive expects zero or more @var{expressions}, of any section,
6309 separated by commas.
6312 For each expression, @command{@value{AS}} emits a 32-bit number.
6315 For each expression, @command{@value{AS}} emits a 16-bit number.
6320 The size of the number emitted, and its byte order,
6321 depend on what target computer the assembly is for.
6324 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6325 @c happen---32-bit addressability, period; no long/short jumps.
6326 @ifset DIFF-TBL-KLUGE
6327 @cindex difference tables altered
6328 @cindex altered difference tables
6330 @emph{Warning: Special Treatment to support Compilers}
6334 Machines with a 32-bit address space, but that do less than 32-bit
6335 addressing, require the following special treatment. If the machine of
6336 interest to you does 32-bit addressing (or doesn't require it;
6337 @pxref{Machine Dependencies}), you can ignore this issue.
6340 In order to assemble compiler output into something that works,
6341 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6342 Directives of the form @samp{.word sym1-sym2} are often emitted by
6343 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6344 directive of the form @samp{.word sym1-sym2}, and the difference between
6345 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6346 creates a @dfn{secondary jump table}, immediately before the next label.
6347 This secondary jump table is preceded by a short-jump to the
6348 first byte after the secondary table. This short-jump prevents the flow
6349 of control from accidentally falling into the new table. Inside the
6350 table is a long-jump to @code{sym2}. The original @samp{.word}
6351 contains @code{sym1} minus the address of the long-jump to
6354 If there were several occurrences of @samp{.word sym1-sym2} before the
6355 secondary jump table, all of them are adjusted. If there was a
6356 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6357 long-jump to @code{sym4} is included in the secondary jump table,
6358 and the @code{.word} directives are adjusted to contain @code{sym3}
6359 minus the address of the long-jump to @code{sym4}; and so on, for as many
6360 entries in the original jump table as necessary.
6363 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6364 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6365 assembly language programmers.
6368 @c end DIFF-TBL-KLUGE
6371 @section Deprecated Directives
6373 @cindex deprecated directives
6374 @cindex obsolescent directives
6375 One day these directives won't work.
6376 They are included for compatibility with older assemblers.
6383 @node Machine Dependencies
6384 @chapter Machine Dependent Features
6386 @cindex machine dependencies
6387 The machine instruction sets are (almost by definition) different on
6388 each machine where @command{@value{AS}} runs. Floating point representations
6389 vary as well, and @command{@value{AS}} often supports a few additional
6390 directives or command-line options for compatibility with other
6391 assemblers on a particular platform. Finally, some versions of
6392 @command{@value{AS}} support special pseudo-instructions for branch
6395 This chapter discusses most of these differences, though it does not
6396 include details on any machine's instruction set. For details on that
6397 subject, see the hardware manufacturer's manual.
6401 * Alpha-Dependent:: Alpha Dependent Features
6404 * ARC-Dependent:: ARC Dependent Features
6407 * ARM-Dependent:: ARM Dependent Features
6410 * AVR-Dependent:: AVR Dependent Features
6413 * BFIN-Dependent:: BFIN Dependent Features
6416 * CR16-Dependent:: CR16 Dependent Features
6419 * CRIS-Dependent:: CRIS Dependent Features
6422 * D10V-Dependent:: D10V Dependent Features
6425 * D30V-Dependent:: D30V Dependent Features
6428 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6431 * HPPA-Dependent:: HPPA Dependent Features
6434 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6437 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6440 * i860-Dependent:: Intel 80860 Dependent Features
6443 * i960-Dependent:: Intel 80960 Dependent Features
6446 * IA-64-Dependent:: Intel IA-64 Dependent Features
6449 * IP2K-Dependent:: IP2K Dependent Features
6452 * M32C-Dependent:: M32C Dependent Features
6455 * M32R-Dependent:: M32R Dependent Features
6458 * M68K-Dependent:: M680x0 Dependent Features
6461 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6464 * MIPS-Dependent:: MIPS Dependent Features
6467 * MMIX-Dependent:: MMIX Dependent Features
6470 * MSP430-Dependent:: MSP430 Dependent Features
6473 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6474 * SH64-Dependent:: SuperH SH64 Dependent Features
6477 * PDP-11-Dependent:: PDP-11 Dependent Features
6480 * PJ-Dependent:: picoJava Dependent Features
6483 * PPC-Dependent:: PowerPC Dependent Features
6486 * Sparc-Dependent:: SPARC Dependent Features
6489 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6492 * V850-Dependent:: V850 Dependent Features
6495 * Xtensa-Dependent:: Xtensa Dependent Features
6498 * Z80-Dependent:: Z80 Dependent Features
6501 * Z8000-Dependent:: Z8000 Dependent Features
6504 * Vax-Dependent:: VAX Dependent Features
6511 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6512 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6513 @c peculiarity: to preserve cross-references, there must be a node called
6514 @c "Machine Dependencies". Hence the conditional nodenames in each
6515 @c major node below. Node defaulting in makeinfo requires adjacency of
6516 @c node and sectioning commands; hence the repetition of @chapter BLAH
6517 @c in both conditional blocks.
6520 @include c-alpha.texi
6536 @include c-bfin.texi
6540 @include c-cr16.texi
6544 @include c-cris.texi
6549 @node Machine Dependencies
6550 @chapter Machine Dependent Features
6552 The machine instruction sets are different on each Renesas chip family,
6553 and there are also some syntax differences among the families. This
6554 chapter describes the specific @command{@value{AS}} features for each
6558 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6559 * SH-Dependent:: Renesas SH Dependent Features
6566 @include c-d10v.texi
6570 @include c-d30v.texi
6574 @include c-h8300.texi
6578 @include c-hppa.texi
6582 @include c-i370.texi
6586 @include c-i386.texi
6590 @include c-i860.texi
6594 @include c-i960.texi
6598 @include c-ia64.texi
6602 @include c-ip2k.texi
6606 @include c-m32c.texi
6610 @include c-m32r.texi
6614 @include c-m68k.texi
6618 @include c-m68hc11.texi
6622 @include c-mips.texi
6626 @include c-mmix.texi
6630 @include c-msp430.texi
6634 @include c-ns32k.texi
6638 @include c-pdp11.texi
6651 @include c-sh64.texi
6655 @include c-sparc.texi
6659 @include c-tic54x.texi
6675 @include c-v850.texi
6679 @include c-xtensa.texi
6683 @c reverse effect of @down at top of generic Machine-Dep chapter
6687 @node Reporting Bugs
6688 @chapter Reporting Bugs
6689 @cindex bugs in assembler
6690 @cindex reporting bugs in assembler
6692 Your bug reports play an essential role in making @command{@value{AS}} reliable.
6694 Reporting a bug may help you by bringing a solution to your problem, or it may
6695 not. But in any case the principal function of a bug report is to help the
6696 entire community by making the next version of @command{@value{AS}} work better.
6697 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
6699 In order for a bug report to serve its purpose, you must include the
6700 information that enables us to fix the bug.
6703 * Bug Criteria:: Have you found a bug?
6704 * Bug Reporting:: How to report bugs
6708 @section Have You Found a Bug?
6709 @cindex bug criteria
6711 If you are not sure whether you have found a bug, here are some guidelines:
6714 @cindex fatal signal
6715 @cindex assembler crash
6716 @cindex crash of assembler
6718 If the assembler gets a fatal signal, for any input whatever, that is a
6719 @command{@value{AS}} bug. Reliable assemblers never crash.
6721 @cindex error on valid input
6723 If @command{@value{AS}} produces an error message for valid input, that is a bug.
6725 @cindex invalid input
6727 If @command{@value{AS}} does not produce an error message for invalid input, that
6728 is a bug. However, you should note that your idea of ``invalid input'' might
6729 be our idea of ``an extension'' or ``support for traditional practice''.
6732 If you are an experienced user of assemblers, your suggestions for improvement
6733 of @command{@value{AS}} are welcome in any case.
6737 @section How to Report Bugs
6739 @cindex assembler bugs, reporting
6741 A number of companies and individuals offer support for @sc{gnu} products. If
6742 you obtained @command{@value{AS}} from a support organization, we recommend you
6743 contact that organization first.
6745 You can find contact information for many support companies and
6746 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
6750 In any event, we also recommend that you send bug reports for @command{@value{AS}}
6754 The fundamental principle of reporting bugs usefully is this:
6755 @strong{report all the facts}. If you are not sure whether to state a
6756 fact or leave it out, state it!
6758 Often people omit facts because they think they know what causes the problem
6759 and assume that some details do not matter. Thus, you might assume that the
6760 name of a symbol you use in an example does not matter. Well, probably it does
6761 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
6762 happens to fetch from the location where that name is stored in memory;
6763 perhaps, if the name were different, the contents of that location would fool
6764 the assembler into doing the right thing despite the bug. Play it safe and
6765 give a specific, complete example. That is the easiest thing for you to do,
6766 and the most helpful.
6768 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
6769 it is new to us. Therefore, always write your bug reports on the assumption
6770 that the bug has not been reported previously.
6772 Sometimes people give a few sketchy facts and ask, ``Does this ring a
6773 bell?'' This cannot help us fix a bug, so it is basically useless. We
6774 respond by asking for enough details to enable us to investigate.
6775 You might as well expedite matters by sending them to begin with.
6777 To enable us to fix the bug, you should include all these things:
6781 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
6782 it with the @samp{--version} argument.
6784 Without this, we will not know whether there is any point in looking for
6785 the bug in the current version of @command{@value{AS}}.
6788 Any patches you may have applied to the @command{@value{AS}} source.
6791 The type of machine you are using, and the operating system name and
6795 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
6799 The command arguments you gave the assembler to assemble your example and
6800 observe the bug. To guarantee you will not omit something important, list them
6801 all. A copy of the Makefile (or the output from make) is sufficient.
6803 If we were to try to guess the arguments, we would probably guess wrong
6804 and then we might not encounter the bug.
6807 A complete input file that will reproduce the bug. If the bug is observed when
6808 the assembler is invoked via a compiler, send the assembler source, not the
6809 high level language source. Most compilers will produce the assembler source
6810 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
6811 the options @samp{-v --save-temps}; this will save the assembler source in a
6812 file with an extension of @file{.s}, and also show you exactly how
6813 @command{@value{AS}} is being run.
6816 A description of what behavior you observe that you believe is
6817 incorrect. For example, ``It gets a fatal signal.''
6819 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
6820 will certainly notice it. But if the bug is incorrect output, we might not
6821 notice unless it is glaringly wrong. You might as well not give us a chance to
6824 Even if the problem you experience is a fatal signal, you should still say so
6825 explicitly. Suppose something strange is going on, such as, your copy of
6826 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
6827 library on your system. (This has happened!) Your copy might crash and ours
6828 would not. If you told us to expect a crash, then when ours fails to crash, we
6829 would know that the bug was not happening for us. If you had not told us to
6830 expect a crash, then we would not be able to draw any conclusion from our
6834 If you wish to suggest changes to the @command{@value{AS}} source, send us context
6835 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
6836 option. Always send diffs from the old file to the new file. If you even
6837 discuss something in the @command{@value{AS}} source, refer to it by context, not
6840 The line numbers in our development sources will not match those in your
6841 sources. Your line numbers would convey no useful information to us.
6844 Here are some things that are not necessary:
6848 A description of the envelope of the bug.
6850 Often people who encounter a bug spend a lot of time investigating
6851 which changes to the input file will make the bug go away and which
6852 changes will not affect it.
6854 This is often time consuming and not very useful, because the way we
6855 will find the bug is by running a single example under the debugger
6856 with breakpoints, not by pure deduction from a series of examples.
6857 We recommend that you save your time for something else.
6859 Of course, if you can find a simpler example to report @emph{instead}
6860 of the original one, that is a convenience for us. Errors in the
6861 output will be easier to spot, running under the debugger will take
6862 less time, and so on.
6864 However, simplification is not vital; if you do not want to do this,
6865 report the bug anyway and send us the entire test case you used.
6868 A patch for the bug.
6870 A patch for the bug does help us if it is a good one. But do not omit
6871 the necessary information, such as the test case, on the assumption that
6872 a patch is all we need. We might see problems with your patch and decide
6873 to fix the problem another way, or we might not understand it at all.
6875 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
6876 construct an example that will make the program follow a certain path through
6877 the code. If you do not send us the example, we will not be able to construct
6878 one, so we will not be able to verify that the bug is fixed.
6880 And if we cannot understand what bug you are trying to fix, or why your
6881 patch should be an improvement, we will not install it. A test case will
6882 help us to understand.
6885 A guess about what the bug is or what it depends on.
6887 Such guesses are usually wrong. Even we cannot guess right about such
6888 things without first using the debugger to find the facts.
6891 @node Acknowledgements
6892 @chapter Acknowledgements
6894 If you have contributed to GAS and your name isn't listed here,
6895 it is not meant as a slight. We just don't know about it. Send mail to the
6896 maintainer, and we'll correct the situation. Currently
6898 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
6900 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
6903 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
6904 information and the 68k series machines, most of the preprocessing pass, and
6905 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
6907 K. Richard Pixley maintained GAS for a while, adding various enhancements and
6908 many bug fixes, including merging support for several processors, breaking GAS
6909 up to handle multiple object file format back ends (including heavy rewrite,
6910 testing, an integration of the coff and b.out back ends), adding configuration
6911 including heavy testing and verification of cross assemblers and file splits
6912 and renaming, converted GAS to strictly ANSI C including full prototypes, added
6913 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
6914 port (including considerable amounts of reverse engineering), a SPARC opcode
6915 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
6916 assertions and made them work, much other reorganization, cleanup, and lint.
6918 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
6919 in format-specific I/O modules.
6921 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
6922 has done much work with it since.
6924 The Intel 80386 machine description was written by Eliot Dresselhaus.
6926 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
6928 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
6929 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
6931 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
6932 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
6933 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
6934 support a.out format.
6936 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
6937 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
6938 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
6939 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
6942 John Gilmore built the AMD 29000 support, added @code{.include} support, and
6943 simplified the configuration of which versions accept which directives. He
6944 updated the 68k machine description so that Motorola's opcodes always produced
6945 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
6946 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
6947 cross-compilation support, and one bug in relaxation that took a week and
6948 required the proverbial one-bit fix.
6950 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
6951 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
6952 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
6953 PowerPC assembler, and made a few other minor patches.
6955 Steve Chamberlain made GAS able to generate listings.
6957 Hewlett-Packard contributed support for the HP9000/300.
6959 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
6960 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
6961 formats). This work was supported by both the Center for Software Science at
6962 the University of Utah and Cygnus Support.
6964 Support for ELF format files has been worked on by Mark Eichin of Cygnus
6965 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
6966 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
6967 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
6968 and some initial 64-bit support).
6970 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
6972 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
6973 support for openVMS/Alpha.
6975 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
6978 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
6979 Inc.@: added support for Xtensa processors.
6981 Several engineers at Cygnus Support have also provided many small bug fixes and
6982 configuration enhancements.
6984 Many others have contributed large or small bugfixes and enhancements. If
6985 you have contributed significant work and are not mentioned on this list, and
6986 want to be, let us know. Some of the history has been lost; we are not
6987 intentionally leaving anyone out.
6992 @unnumbered AS Index
projects / platform / upstream / binutils.git / blob