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
193 * Object Attributes:: Object Attributes
195 * Machine Dependencies:: Machine Dependent Features
196 * Reporting Bugs:: Reporting Bugs
197 * Acknowledgements:: Who Did What
198 * GNU Free Documentation License:: GNU Free Documentation License
199 * AS Index:: AS Index
206 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
208 This version of the manual describes @command{@value{AS}} configured to generate
209 code for @value{TARGET} architectures.
213 @cindex invocation summary
214 @cindex option summary
215 @cindex summary of options
216 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
217 see @ref{Invoking,,Command-Line Options}.
219 @c man title AS the portable GNU assembler.
223 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
227 @c We don't use deffn and friends for the following because they seem
228 @c to be limited to one line for the header.
230 @c man begin SYNOPSIS
231 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
232 [@b{--debug-prefix-map} @var{old}=@var{new}]
233 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
234 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
235 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
236 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
237 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}] [@b{-o}
238 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
239 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
240 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{@@@var{FILE}}]
241 [@b{--target-help}] [@var{target-options}]
242 [@b{--}|@var{files} @dots{}]
244 @c Target dependent options are listed below. Keep the list sorted.
245 @c Add an empty line for separation.
248 @emph{Target Alpha options:}
250 [@b{-mdebug} | @b{-no-mdebug}]
251 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
252 [@b{-F}] [@b{-32addr}]
256 @emph{Target ARC options:}
262 @emph{Target ARM options:}
263 @c Don't document the deprecated options
264 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
265 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
266 [@b{-mfpu}=@var{floating-point-format}]
267 [@b{-mfloat-abi}=@var{abi}]
268 [@b{-meabi}=@var{ver}]
271 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
272 @b{-mapcs-reentrant}]
273 [@b{-mthumb-interwork}] [@b{-k}]
277 @emph{Target CRIS options:}
278 [@b{--underscore} | @b{--no-underscore}]
280 [@b{--emulation=criself} | @b{--emulation=crisaout}]
281 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
282 @c Deprecated -- deliberately not documented.
287 @emph{Target D10V options:}
292 @emph{Target D30V options:}
293 [@b{-O}|@b{-n}|@b{-N}]
297 @emph{Target H8/300 options:}
301 @c HPPA has no machine-dependent assembler options (yet).
305 @emph{Target i386 options:}
306 [@b{--32}|@b{--64}] [@b{-n}]
307 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
311 @emph{Target i960 options:}
312 @c see md_parse_option in tc-i960.c
313 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
315 [@b{-b}] [@b{-no-relax}]
319 @emph{Target IA-64 options:}
320 [@b{-mconstant-gp}|@b{-mauto-pic}]
321 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
323 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
324 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
325 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
326 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
330 @emph{Target IP2K options:}
331 [@b{-mip2022}|@b{-mip2022ext}]
335 @emph{Target M32C options:}
336 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
340 @emph{Target M32R options:}
341 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
346 @emph{Target M680X0 options:}
347 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
351 @emph{Target M68HC11 options:}
352 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
353 [@b{-mshort}|@b{-mlong}]
354 [@b{-mshort-double}|@b{-mlong-double}]
355 [@b{--force-long-branches}] [@b{--short-branches}]
356 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
357 [@b{--print-opcodes}] [@b{--generate-example}]
361 @emph{Target MCORE options:}
362 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
363 [@b{-mcpu=[210|340]}]
367 @emph{Target MIPS options:}
368 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
369 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
370 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
371 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
372 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
373 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
374 [@b{-mips64}] [@b{-mips64r2}]
375 [@b{-construct-floats}] [@b{-no-construct-floats}]
376 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
377 [@b{-mfix7000}] [@b{-mno-fix7000}]
378 [@b{-mips16}] [@b{-no-mips16}]
379 [@b{-msmartmips}] [@b{-mno-smartmips}]
380 [@b{-mips3d}] [@b{-no-mips3d}]
381 [@b{-mdmx}] [@b{-no-mdmx}]
382 [@b{-mdsp}] [@b{-mno-dsp}]
383 [@b{-mdspr2}] [@b{-mno-dspr2}]
384 [@b{-mmt}] [@b{-mno-mt}]
385 [@b{-mdebug}] [@b{-no-mdebug}]
386 [@b{-mpdr}] [@b{-mno-pdr}]
390 @emph{Target MMIX options:}
391 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
392 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
393 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
394 [@b{--linker-allocated-gregs}]
398 @emph{Target PDP11 options:}
399 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
400 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
401 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
405 @emph{Target picoJava options:}
410 @emph{Target PowerPC options:}
411 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
412 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}|
413 @b{-mbooke32}|@b{-mbooke64}]
414 [@b{-mcom}|@b{-many}|@b{-maltivec}|@b{-mvsx}] [@b{-memb}]
415 [@b{-mregnames}|@b{-mno-regnames}]
416 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
417 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
418 [@b{-msolaris}|@b{-mno-solaris}]
422 @emph{Target SPARC options:}
423 @c The order here is important. See c-sparc.texi.
424 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
425 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
426 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
431 @emph{Target TIC54X options:}
432 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
433 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
438 @emph{Target Z80 options:}
439 [@b{-z80}] [@b{-r800}]
440 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
441 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
442 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
443 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
444 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
445 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
449 @c Z8000 has no machine-dependent assembler options
453 @emph{Target Xtensa options:}
454 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
455 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
456 [@b{--[no-]transform}]
457 [@b{--rename-section} @var{oldname}=@var{newname}]
465 @include at-file.texi
468 Turn on listings, in any of a variety of ways:
472 omit false conditionals
475 omit debugging directives
478 include general information, like @value{AS} version and options passed
481 include high-level source
487 include macro expansions
490 omit forms processing
496 set the name of the listing file
499 You may combine these options; for example, use @samp{-aln} for assembly
500 listing without forms processing. The @samp{=file} option, if used, must be
501 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
504 Begin in alternate macro mode.
506 @xref{Altmacro,,@code{.altmacro}}.
510 Ignored. This option is accepted for script compatibility with calls to
513 @item --debug-prefix-map @var{old}=@var{new}
514 When assembling files in directory @file{@var{old}}, record debugging
515 information describing them as in @file{@var{new}} instead.
517 @item --defsym @var{sym}=@var{value}
518 Define the symbol @var{sym} to be @var{value} before assembling the input file.
519 @var{value} must be an integer constant. As in C, a leading @samp{0x}
520 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
521 value. The value of the symbol can be overridden inside a source file via the
522 use of a @code{.set} pseudo-op.
525 ``fast''---skip whitespace and comment preprocessing (assume source is
530 Generate debugging information for each assembler source line using whichever
531 debug format is preferred by the target. This currently means either STABS,
535 Generate stabs debugging information for each assembler line. This
536 may help debugging assembler code, if the debugger can handle it.
539 Generate stabs debugging information for each assembler line, with GNU
540 extensions that probably only gdb can handle, and that could make other
541 debuggers crash or refuse to read your program. This
542 may help debugging assembler code. Currently the only GNU extension is
543 the location of the current working directory at assembling time.
546 Generate DWARF2 debugging information for each assembler line. This
547 may help debugging assembler code, if the debugger can handle it. Note---this
548 option is only supported by some targets, not all of them.
551 Print a summary of the command line options and exit.
554 Print a summary of all target specific options and exit.
557 Add directory @var{dir} to the search list for @code{.include} directives.
560 Don't warn about signed overflow.
563 @ifclear DIFF-TBL-KLUGE
564 This option is accepted but has no effect on the @value{TARGET} family.
566 @ifset DIFF-TBL-KLUGE
567 Issue warnings when difference tables altered for long displacements.
572 Keep (in the symbol table) local symbols. These symbols start with
573 system-specific local label prefixes, typically @samp{.L} for ELF systems
574 or @samp{L} for traditional a.out systems.
579 @item --listing-lhs-width=@var{number}
580 Set the maximum width, in words, of the output data column for an assembler
581 listing to @var{number}.
583 @item --listing-lhs-width2=@var{number}
584 Set the maximum width, in words, of the output data column for continuation
585 lines in an assembler listing to @var{number}.
587 @item --listing-rhs-width=@var{number}
588 Set the maximum width of an input source line, as displayed in a listing, to
591 @item --listing-cont-lines=@var{number}
592 Set the maximum number of lines printed in a listing for a single line of input
595 @item -o @var{objfile}
596 Name the object-file output from @command{@value{AS}} @var{objfile}.
599 Fold the data section into the text section.
601 @kindex --hash-size=@var{number}
602 Set the default size of GAS's hash tables to a prime number close to
603 @var{number}. Increasing this value can reduce the length of time it takes the
604 assembler to perform its tasks, at the expense of increasing the assembler's
605 memory requirements. Similarly reducing this value can reduce the memory
606 requirements at the expense of speed.
608 @item --reduce-memory-overheads
609 This option reduces GAS's memory requirements, at the expense of making the
610 assembly processes slower. Currently this switch is a synonym for
611 @samp{--hash-size=4051}, but in the future it may have other effects as well.
614 Print the maximum space (in bytes) and total time (in seconds) used by
617 @item --strip-local-absolute
618 Remove local absolute symbols from the outgoing symbol table.
622 Print the @command{as} version.
625 Print the @command{as} version and exit.
629 Suppress warning messages.
631 @item --fatal-warnings
632 Treat warnings as errors.
635 Don't suppress warning messages or treat them as errors.
644 Generate an object file even after errors.
646 @item -- | @var{files} @dots{}
647 Standard input, or source files to assemble.
652 The following options are available when @value{AS} is configured for
657 This option selects the core processor variant.
659 Select either big-endian (-EB) or little-endian (-EL) output.
664 The following options are available when @value{AS} is configured for the ARM
668 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
669 Specify which ARM processor variant is the target.
670 @item -march=@var{architecture}[+@var{extension}@dots{}]
671 Specify which ARM architecture variant is used by the target.
672 @item -mfpu=@var{floating-point-format}
673 Select which Floating Point architecture is the target.
674 @item -mfloat-abi=@var{abi}
675 Select which floating point ABI is in use.
677 Enable Thumb only instruction decoding.
678 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
679 Select which procedure calling convention is in use.
681 Select either big-endian (-EB) or little-endian (-EL) output.
682 @item -mthumb-interwork
683 Specify that the code has been generated with interworking between Thumb and
686 Specify that PIC code has been generated.
691 See the info pages for documentation of the CRIS-specific options.
695 The following options are available when @value{AS} is configured for
698 @cindex D10V optimization
699 @cindex optimization, D10V
701 Optimize output by parallelizing instructions.
706 The following options are available when @value{AS} is configured for a D30V
709 @cindex D30V optimization
710 @cindex optimization, D30V
712 Optimize output by parallelizing instructions.
716 Warn when nops are generated.
718 @cindex D30V nops after 32-bit multiply
720 Warn when a nop after a 32-bit multiply instruction is generated.
725 The following options are available when @value{AS} is configured for the
726 Intel 80960 processor.
729 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
730 Specify which variant of the 960 architecture is the target.
733 Add code to collect statistics about branches taken.
736 Do not alter compare-and-branch instructions for long displacements;
743 The following options are available when @value{AS} is configured for the
749 Specifies that the extended IP2022 instructions are allowed.
752 Restores the default behaviour, which restricts the permitted instructions to
753 just the basic IP2022 ones.
759 The following options are available when @value{AS} is configured for the
760 Renesas M32C and M16C processors.
765 Assemble M32C instructions.
768 Assemble M16C instructions (the default).
771 Enable support for link-time relaxations.
774 Support H'00 style hex constants in addition to 0x00 style.
780 The following options are available when @value{AS} is configured for the
781 Renesas M32R (formerly Mitsubishi M32R) series.
786 Specify which processor in the M32R family is the target. The default
787 is normally the M32R, but this option changes it to the M32RX.
789 @item --warn-explicit-parallel-conflicts or --Wp
790 Produce warning messages when questionable parallel constructs are
793 @item --no-warn-explicit-parallel-conflicts or --Wnp
794 Do not produce warning messages when questionable parallel constructs are
801 The following options are available when @value{AS} is configured for the
802 Motorola 68000 series.
807 Shorten references to undefined symbols, to one word instead of two.
809 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
810 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
811 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
812 Specify what processor in the 68000 family is the target. The default
813 is normally the 68020, but this can be changed at configuration time.
815 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
816 The target machine does (or does not) have a floating-point coprocessor.
817 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
818 the basic 68000 is not compatible with the 68881, a combination of the
819 two can be specified, since it's possible to do emulation of the
820 coprocessor instructions with the main processor.
822 @item -m68851 | -mno-68851
823 The target machine does (or does not) have a memory-management
824 unit coprocessor. The default is to assume an MMU for 68020 and up.
831 For details about the PDP-11 machine dependent features options,
832 see @ref{PDP-11-Options}.
835 @item -mpic | -mno-pic
836 Generate position-independent (or position-dependent) code. The
837 default is @option{-mpic}.
840 @itemx -mall-extensions
841 Enable all instruction set extensions. This is the default.
843 @item -mno-extensions
844 Disable all instruction set extensions.
846 @item -m@var{extension} | -mno-@var{extension}
847 Enable (or disable) a particular instruction set extension.
850 Enable the instruction set extensions supported by a particular CPU, and
851 disable all other extensions.
853 @item -m@var{machine}
854 Enable the instruction set extensions supported by a particular machine
855 model, and disable all other extensions.
861 The following options are available when @value{AS} is configured for
862 a picoJava processor.
866 @cindex PJ endianness
867 @cindex endianness, PJ
868 @cindex big endian output, PJ
870 Generate ``big endian'' format output.
872 @cindex little endian output, PJ
874 Generate ``little endian'' format output.
880 The following options are available when @value{AS} is configured for the
881 Motorola 68HC11 or 68HC12 series.
885 @item -m68hc11 | -m68hc12 | -m68hcs12
886 Specify what processor is the target. The default is
887 defined by the configuration option when building the assembler.
890 Specify to use the 16-bit integer ABI.
893 Specify to use the 32-bit integer ABI.
896 Specify to use the 32-bit double ABI.
899 Specify to use the 64-bit double ABI.
901 @item --force-long-branches
902 Relative branches are turned into absolute ones. This concerns
903 conditional branches, unconditional branches and branches to a
906 @item -S | --short-branches
907 Do not turn relative branches into absolute ones
908 when the offset is out of range.
910 @item --strict-direct-mode
911 Do not turn the direct addressing mode into extended addressing mode
912 when the instruction does not support direct addressing mode.
914 @item --print-insn-syntax
915 Print the syntax of instruction in case of error.
917 @item --print-opcodes
918 print the list of instructions with syntax and then exit.
920 @item --generate-example
921 print an example of instruction for each possible instruction and then exit.
922 This option is only useful for testing @command{@value{AS}}.
928 The following options are available when @command{@value{AS}} is configured
929 for the SPARC architecture:
932 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
933 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
934 Explicitly select a variant of the SPARC architecture.
936 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
937 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
939 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
940 UltraSPARC extensions.
942 @item -xarch=v8plus | -xarch=v8plusa
943 For compatibility with the Solaris v9 assembler. These options are
944 equivalent to -Av8plus and -Av8plusa, respectively.
947 Warn when the assembler switches to another architecture.
952 The following options are available when @value{AS} is configured for the 'c54x
957 Enable extended addressing mode. All addresses and relocations will assume
958 extended addressing (usually 23 bits).
959 @item -mcpu=@var{CPU_VERSION}
960 Sets the CPU version being compiled for.
961 @item -merrors-to-file @var{FILENAME}
962 Redirect error output to a file, for broken systems which don't support such
963 behaviour in the shell.
968 The following options are available when @value{AS} is configured for
969 a @sc{mips} processor.
973 This option sets the largest size of an object that can be referenced
974 implicitly with the @code{gp} register. It is only accepted for targets that
975 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
977 @cindex MIPS endianness
978 @cindex endianness, MIPS
979 @cindex big endian output, MIPS
981 Generate ``big endian'' format output.
983 @cindex little endian output, MIPS
985 Generate ``little endian'' format output.
997 Generate code for a particular @sc{mips} Instruction Set Architecture level.
998 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
999 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1000 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1001 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1003 correspond to generic
1004 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
1005 and @samp{MIPS64 Release 2}
1006 ISA processors, respectively.
1008 @item -march=@var{CPU}
1009 Generate code for a particular @sc{mips} cpu.
1011 @item -mtune=@var{cpu}
1012 Schedule and tune for a particular @sc{mips} cpu.
1016 Cause nops to be inserted if the read of the destination register
1017 of an mfhi or mflo instruction occurs in the following two instructions.
1021 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1022 section instead of the standard ELF .stabs sections.
1026 Control generation of @code{.pdr} sections.
1030 The register sizes are normally inferred from the ISA and ABI, but these
1031 flags force a certain group of registers to be treated as 32 bits wide at
1032 all times. @samp{-mgp32} controls the size of general-purpose registers
1033 and @samp{-mfp32} controls the size of floating-point registers.
1037 Generate code for the MIPS 16 processor. This is equivalent to putting
1038 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1039 turns off this option.
1042 @itemx -mno-smartmips
1043 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1044 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1045 @samp{-mno-smartmips} turns off this option.
1049 Generate code for the MIPS-3D Application Specific Extension.
1050 This tells the assembler to accept MIPS-3D instructions.
1051 @samp{-no-mips3d} turns off this option.
1055 Generate code for the MDMX Application Specific Extension.
1056 This tells the assembler to accept MDMX instructions.
1057 @samp{-no-mdmx} turns off this option.
1061 Generate code for the DSP Release 1 Application Specific Extension.
1062 This tells the assembler to accept DSP Release 1 instructions.
1063 @samp{-mno-dsp} turns off this option.
1067 Generate code for the DSP Release 2 Application Specific Extension.
1068 This option implies -mdsp.
1069 This tells the assembler to accept DSP Release 2 instructions.
1070 @samp{-mno-dspr2} turns off this option.
1074 Generate code for the MT Application Specific Extension.
1075 This tells the assembler to accept MT instructions.
1076 @samp{-mno-mt} turns off this option.
1078 @item --construct-floats
1079 @itemx --no-construct-floats
1080 The @samp{--no-construct-floats} option disables the construction of
1081 double width floating point constants by loading the two halves of the
1082 value into the two single width floating point registers that make up
1083 the double width register. By default @samp{--construct-floats} is
1084 selected, allowing construction of these floating point constants.
1087 @item --emulation=@var{name}
1088 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1089 for some other target, in all respects, including output format (choosing
1090 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1091 debugging information or store symbol table information, and default
1092 endianness. The available configuration names are: @samp{mipsecoff},
1093 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1094 @samp{mipsbelf}. The first two do not alter the default endianness from that
1095 of the primary target for which the assembler was configured; the others change
1096 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1097 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1098 selection in any case.
1100 This option is currently supported only when the primary target
1101 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1102 Furthermore, the primary target or others specified with
1103 @samp{--enable-targets=@dots{}} at configuration time must include support for
1104 the other format, if both are to be available. For example, the Irix 5
1105 configuration includes support for both.
1107 Eventually, this option will support more configurations, with more
1108 fine-grained control over the assembler's behavior, and will be supported for
1112 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1119 Control how to deal with multiplication overflow and division by zero.
1120 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1121 (and only work for Instruction Set Architecture level 2 and higher);
1122 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1126 When this option is used, @command{@value{AS}} will issue a warning every
1127 time it generates a nop instruction from a macro.
1132 The following options are available when @value{AS} is configured for
1138 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1139 The command line option @samp{-nojsri2bsr} can be used to disable it.
1143 Enable or disable the silicon filter behaviour. By default this is disabled.
1144 The default can be overridden by the @samp{-sifilter} command line option.
1147 Alter jump instructions for long displacements.
1149 @item -mcpu=[210|340]
1150 Select the cpu type on the target hardware. This controls which instructions
1154 Assemble for a big endian target.
1157 Assemble for a little endian target.
1163 See the info pages for documentation of the MMIX-specific options.
1167 The following options are available when @value{AS} is configured for
1168 an Xtensa processor.
1171 @item --text-section-literals | --no-text-section-literals
1172 With @option{--text-@-section-@-literals}, literal pools are interspersed
1173 in the text section. The default is
1174 @option{--no-@-text-@-section-@-literals}, which places literals in a
1175 separate section in the output file. These options only affect literals
1176 referenced via PC-relative @code{L32R} instructions; literals for
1177 absolute mode @code{L32R} instructions are handled separately.
1179 @item --absolute-literals | --no-absolute-literals
1180 Indicate to the assembler whether @code{L32R} instructions use absolute
1181 or PC-relative addressing. The default is to assume absolute addressing
1182 if the Xtensa processor includes the absolute @code{L32R} addressing
1183 option. Otherwise, only the PC-relative @code{L32R} mode can be used.
1185 @item --target-align | --no-target-align
1186 Enable or disable automatic alignment to reduce branch penalties at the
1187 expense of some code density. The default is @option{--target-@-align}.
1189 @item --longcalls | --no-longcalls
1190 Enable or disable transformation of call instructions to allow calls
1191 across a greater range of addresses. The default is
1192 @option{--no-@-longcalls}.
1194 @item --transform | --no-transform
1195 Enable or disable all assembler transformations of Xtensa instructions.
1196 The default is @option{--transform};
1197 @option{--no-transform} should be used only in the rare cases when the
1198 instructions must be exactly as specified in the assembly source.
1200 @item --rename-section @var{oldname}=@var{newname}
1201 When generating output sections, rename the @var{oldname} section to
1207 The following options are available when @value{AS} is configured for
1208 a Z80 family processor.
1211 Assemble for Z80 processor.
1213 Assemble for R800 processor.
1214 @item -ignore-undocumented-instructions
1216 Assemble undocumented Z80 instructions that also work on R800 without warning.
1217 @item -ignore-unportable-instructions
1219 Assemble all undocumented Z80 instructions without warning.
1220 @item -warn-undocumented-instructions
1222 Issue a warning for undocumented Z80 instructions that also work on R800.
1223 @item -warn-unportable-instructions
1225 Issue a warning for undocumented Z80 instructions that do not work on R800.
1226 @item -forbid-undocumented-instructions
1228 Treat all undocumented instructions as errors.
1229 @item -forbid-unportable-instructions
1231 Treat undocumented Z80 instructions that do not work on R800 as errors.
1238 * Manual:: Structure of this Manual
1239 * GNU Assembler:: The GNU Assembler
1240 * Object Formats:: Object File Formats
1241 * Command Line:: Command Line
1242 * Input Files:: Input Files
1243 * Object:: Output (Object) File
1244 * Errors:: Error and Warning Messages
1248 @section Structure of this Manual
1250 @cindex manual, structure and purpose
1251 This manual is intended to describe what you need to know to use
1252 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1253 notation for symbols, constants, and expressions; the directives that
1254 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1257 We also cover special features in the @value{TARGET}
1258 configuration of @command{@value{AS}}, including assembler directives.
1261 This manual also describes some of the machine-dependent features of
1262 various flavors of the assembler.
1265 @cindex machine instructions (not covered)
1266 On the other hand, this manual is @emph{not} intended as an introduction
1267 to programming in assembly language---let alone programming in general!
1268 In a similar vein, we make no attempt to introduce the machine
1269 architecture; we do @emph{not} describe the instruction set, standard
1270 mnemonics, registers or addressing modes that are standard to a
1271 particular architecture.
1273 You may want to consult the manufacturer's
1274 machine architecture manual for this information.
1278 For information on the H8/300 machine instruction set, see @cite{H8/300
1279 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1280 Programming Manual} (Renesas).
1283 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1284 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1285 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1286 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1289 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1293 @c I think this is premature---doc@cygnus.com, 17jan1991
1295 Throughout this manual, we assume that you are running @dfn{GNU},
1296 the portable operating system from the @dfn{Free Software
1297 Foundation, Inc.}. This restricts our attention to certain kinds of
1298 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1299 once this assumption is granted examples and definitions need less
1302 @command{@value{AS}} is part of a team of programs that turn a high-level
1303 human-readable series of instructions into a low-level
1304 computer-readable series of instructions. Different versions of
1305 @command{@value{AS}} are used for different kinds of computer.
1308 @c There used to be a section "Terminology" here, which defined
1309 @c "contents", "byte", "word", and "long". Defining "word" to any
1310 @c particular size is confusing when the .word directive may generate 16
1311 @c bits on one machine and 32 bits on another; in general, for the user
1312 @c version of this manual, none of these terms seem essential to define.
1313 @c They were used very little even in the former draft of the manual;
1314 @c this draft makes an effort to avoid them (except in names of
1318 @section The GNU Assembler
1320 @c man begin DESCRIPTION
1322 @sc{gnu} @command{as} is really a family of assemblers.
1324 This manual describes @command{@value{AS}}, a member of that family which is
1325 configured for the @value{TARGET} architectures.
1327 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1328 should find a fairly similar environment when you use it on another
1329 architecture. Each version has much in common with the others,
1330 including object file formats, most assembler directives (often called
1331 @dfn{pseudo-ops}) and assembler syntax.@refill
1333 @cindex purpose of @sc{gnu} assembler
1334 @command{@value{AS}} is primarily intended to assemble the output of the
1335 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1336 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1337 assemble correctly everything that other assemblers for the same
1338 machine would assemble.
1340 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1343 @c This remark should appear in generic version of manual; assumption
1344 @c here is that generic version sets M680x0.
1345 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1346 assembler for the same architecture; for example, we know of several
1347 incompatible versions of 680x0 assembly language syntax.
1352 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1353 program in one pass of the source file. This has a subtle impact on the
1354 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1356 @node Object Formats
1357 @section Object File Formats
1359 @cindex object file format
1360 The @sc{gnu} assembler can be configured to produce several alternative
1361 object file formats. For the most part, this does not affect how you
1362 write assembly language programs; but directives for debugging symbols
1363 are typically different in different file formats. @xref{Symbol
1364 Attributes,,Symbol Attributes}.
1367 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1368 @value{OBJ-NAME} format object files.
1370 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1372 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1373 @code{b.out} or COFF format object files.
1376 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1377 SOM or ELF format object files.
1382 @section Command Line
1384 @cindex command line conventions
1386 After the program name @command{@value{AS}}, the command line may contain
1387 options and file names. Options may appear in any order, and may be
1388 before, after, or between file names. The order of file names is
1391 @cindex standard input, as input file
1393 @file{--} (two hyphens) by itself names the standard input file
1394 explicitly, as one of the files for @command{@value{AS}} to assemble.
1396 @cindex options, command line
1397 Except for @samp{--} any command line argument that begins with a
1398 hyphen (@samp{-}) is an option. Each option changes the behavior of
1399 @command{@value{AS}}. No option changes the way another option works. An
1400 option is a @samp{-} followed by one or more letters; the case of
1401 the letter is important. All options are optional.
1403 Some options expect exactly one file name to follow them. The file
1404 name may either immediately follow the option's letter (compatible
1405 with older assemblers) or it may be the next command argument (@sc{gnu}
1406 standard). These two command lines are equivalent:
1409 @value{AS} -o my-object-file.o mumble.s
1410 @value{AS} -omy-object-file.o mumble.s
1414 @section Input Files
1417 @cindex source program
1418 @cindex files, input
1419 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1420 describe the program input to one run of @command{@value{AS}}. The program may
1421 be in one or more files; how the source is partitioned into files
1422 doesn't change the meaning of the source.
1424 @c I added "con" prefix to "catenation" just to prove I can overcome my
1425 @c APL training... doc@cygnus.com
1426 The source program is a concatenation of the text in all the files, in the
1429 @c man begin DESCRIPTION
1430 Each time you run @command{@value{AS}} it assembles exactly one source
1431 program. The source program is made up of one or more files.
1432 (The standard input is also a file.)
1434 You give @command{@value{AS}} a command line that has zero or more input file
1435 names. The input files are read (from left file name to right). A
1436 command line argument (in any position) that has no special meaning
1437 is taken to be an input file name.
1439 If you give @command{@value{AS}} no file names it attempts to read one input file
1440 from the @command{@value{AS}} standard input, which is normally your terminal. You
1441 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1444 Use @samp{--} if you need to explicitly name the standard input file
1445 in your command line.
1447 If the source is empty, @command{@value{AS}} produces a small, empty object
1452 @subheading Filenames and Line-numbers
1454 @cindex input file linenumbers
1455 @cindex line numbers, in input files
1456 There are two ways of locating a line in the input file (or files) and
1457 either may be used in reporting error messages. One way refers to a line
1458 number in a physical file; the other refers to a line number in a
1459 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1461 @dfn{Physical files} are those files named in the command line given
1462 to @command{@value{AS}}.
1464 @dfn{Logical files} are simply names declared explicitly by assembler
1465 directives; they bear no relation to physical files. Logical file names help
1466 error messages reflect the original source file, when @command{@value{AS}} source
1467 is itself synthesized from other files. @command{@value{AS}} understands the
1468 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1469 @ref{File,,@code{.file}}.
1472 @section Output (Object) File
1478 Every time you run @command{@value{AS}} it produces an output file, which is
1479 your assembly language program translated into numbers. This file
1480 is the object file. Its default name is
1488 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1490 You can give it another name by using the @option{-o} option. Conventionally,
1491 object file names end with @file{.o}. The default name is used for historical
1492 reasons: older assemblers were capable of assembling self-contained programs
1493 directly into a runnable program. (For some formats, this isn't currently
1494 possible, but it can be done for the @code{a.out} format.)
1498 The object file is meant for input to the linker @code{@value{LD}}. It contains
1499 assembled program code, information to help @code{@value{LD}} integrate
1500 the assembled program into a runnable file, and (optionally) symbolic
1501 information for the debugger.
1503 @c link above to some info file(s) like the description of a.out.
1504 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1507 @section Error and Warning Messages
1509 @c man begin DESCRIPTION
1511 @cindex error messages
1512 @cindex warning messages
1513 @cindex messages from assembler
1514 @command{@value{AS}} may write warnings and error messages to the standard error
1515 file (usually your terminal). This should not happen when a compiler
1516 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1517 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1518 grave problem that stops the assembly.
1522 @cindex format of warning messages
1523 Warning messages have the format
1526 file_name:@b{NNN}:Warning Message Text
1530 @cindex line numbers, in warnings/errors
1531 (where @b{NNN} is a line number). If a logical file name has been given
1532 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1533 the current input file is used. If a logical line number was given
1535 (@pxref{Line,,@code{.line}})
1537 then it is used to calculate the number printed,
1538 otherwise the actual line in the current source file is printed. The
1539 message text is intended to be self explanatory (in the grand Unix
1542 @cindex format of error messages
1543 Error messages have the format
1545 file_name:@b{NNN}:FATAL:Error Message Text
1547 The file name and line number are derived as for warning
1548 messages. The actual message text may be rather less explanatory
1549 because many of them aren't supposed to happen.
1552 @chapter Command-Line Options
1554 @cindex options, all versions of assembler
1555 This chapter describes command-line options available in @emph{all}
1556 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1557 for options specific
1559 to the @value{TARGET} target.
1562 to particular machine architectures.
1565 @c man begin DESCRIPTION
1567 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1568 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1569 The assembler arguments must be separated from each other (and the @samp{-Wa})
1570 by commas. For example:
1573 gcc -c -g -O -Wa,-alh,-L file.c
1577 This passes two options to the assembler: @samp{-alh} (emit a listing to
1578 standard output with high-level and assembly source) and @samp{-L} (retain
1579 local symbols in the symbol table).
1581 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1582 command-line options are automatically passed to the assembler by the compiler.
1583 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1584 precisely what options it passes to each compilation pass, including the
1590 * a:: -a[cdghlns] enable listings
1591 * alternate:: --alternate enable alternate macro syntax
1592 * D:: -D for compatibility
1593 * f:: -f to work faster
1594 * I:: -I for .include search path
1595 @ifclear DIFF-TBL-KLUGE
1596 * K:: -K for compatibility
1598 @ifset DIFF-TBL-KLUGE
1599 * K:: -K for difference tables
1602 * L:: -L to retain local symbols
1603 * listing:: --listing-XXX to configure listing output
1604 * M:: -M or --mri to assemble in MRI compatibility mode
1605 * MD:: --MD for dependency tracking
1606 * o:: -o to name the object file
1607 * R:: -R to join data and text sections
1608 * statistics:: --statistics to see statistics about assembly
1609 * traditional-format:: --traditional-format for compatible output
1610 * v:: -v to announce version
1611 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1612 * Z:: -Z to make object file even after errors
1616 @section Enable Listings: @option{-a[cdghlns]}
1626 @cindex listings, enabling
1627 @cindex assembly listings, enabling
1629 These options enable listing output from the assembler. By itself,
1630 @samp{-a} requests high-level, assembly, and symbols listing.
1631 You can use other letters to select specific options for the list:
1632 @samp{-ah} requests a high-level language listing,
1633 @samp{-al} requests an output-program assembly listing, and
1634 @samp{-as} requests a symbol table listing.
1635 High-level listings require that a compiler debugging option like
1636 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1639 Use the @samp{-ag} option to print a first section with general assembly
1640 information, like @value{AS} version, switches passed, or time stamp.
1642 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1643 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1644 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1645 omitted from the listing.
1647 Use the @samp{-ad} option to omit debugging directives from the
1650 Once you have specified one of these options, you can further control
1651 listing output and its appearance using the directives @code{.list},
1652 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1654 The @samp{-an} option turns off all forms processing.
1655 If you do not request listing output with one of the @samp{-a} options, the
1656 listing-control directives have no effect.
1658 The letters after @samp{-a} may be combined into one option,
1659 @emph{e.g.}, @samp{-aln}.
1661 Note if the assembler source is coming from the standard input (e.g.,
1663 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1664 is being used) then the listing will not contain any comments or preprocessor
1665 directives. This is because the listing code buffers input source lines from
1666 stdin only after they have been preprocessed by the assembler. This reduces
1667 memory usage and makes the code more efficient.
1670 @section @option{--alternate}
1673 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1676 @section @option{-D}
1679 This option has no effect whatsoever, but it is accepted to make it more
1680 likely that scripts written for other assemblers also work with
1681 @command{@value{AS}}.
1684 @section Work Faster: @option{-f}
1687 @cindex trusted compiler
1688 @cindex faster processing (@option{-f})
1689 @samp{-f} should only be used when assembling programs written by a
1690 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1691 and comment preprocessing on
1692 the input file(s) before assembling them. @xref{Preprocessing,
1696 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1697 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1702 @section @code{.include} Search Path: @option{-I} @var{path}
1704 @kindex -I @var{path}
1705 @cindex paths for @code{.include}
1706 @cindex search path for @code{.include}
1707 @cindex @code{include} directive search path
1708 Use this option to add a @var{path} to the list of directories
1709 @command{@value{AS}} searches for files specified in @code{.include}
1710 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1711 many times as necessary to include a variety of paths. The current
1712 working directory is always searched first; after that, @command{@value{AS}}
1713 searches any @samp{-I} directories in the same order as they were
1714 specified (left to right) on the command line.
1717 @section Difference Tables: @option{-K}
1720 @ifclear DIFF-TBL-KLUGE
1721 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1722 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1723 where it can be used to warn when the assembler alters the machine code
1724 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1725 family does not have the addressing limitations that sometimes lead to this
1726 alteration on other platforms.
1729 @ifset DIFF-TBL-KLUGE
1730 @cindex difference tables, warning
1731 @cindex warning for altered difference tables
1732 @command{@value{AS}} sometimes alters the code emitted for directives of the
1733 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
1734 You can use the @samp{-K} option if you want a warning issued when this
1739 @section Include Local Symbols: @option{-L}
1742 @cindex local symbols, retaining in output
1743 Symbols beginning with system-specific local label prefixes, typically
1744 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
1745 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
1746 such symbols when debugging, because they are intended for the use of
1747 programs (like compilers) that compose assembler programs, not for your
1748 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
1749 such symbols, so you do not normally debug with them.
1751 This option tells @command{@value{AS}} to retain those local symbols
1752 in the object file. Usually if you do this you also tell the linker
1753 @code{@value{LD}} to preserve those symbols.
1756 @section Configuring listing output: @option{--listing}
1758 The listing feature of the assembler can be enabled via the command line switch
1759 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1760 hex dump of the corresponding locations in the output object file, and displays
1761 them as a listing file. The format of this listing can be controlled by
1762 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
1763 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
1764 @code{.psize} (@pxref{Psize}), and
1765 @code{.eject} (@pxref{Eject}) and also by the following switches:
1768 @item --listing-lhs-width=@samp{number}
1769 @kindex --listing-lhs-width
1770 @cindex Width of first line disassembly output
1771 Sets the maximum width, in words, of the first line of the hex byte dump. This
1772 dump appears on the left hand side of the listing output.
1774 @item --listing-lhs-width2=@samp{number}
1775 @kindex --listing-lhs-width2
1776 @cindex Width of continuation lines of disassembly output
1777 Sets the maximum width, in words, of any further lines of the hex byte dump for
1778 a given input source line. If this value is not specified, it defaults to being
1779 the same as the value specified for @samp{--listing-lhs-width}. If neither
1780 switch is used the default is to one.
1782 @item --listing-rhs-width=@samp{number}
1783 @kindex --listing-rhs-width
1784 @cindex Width of source line output
1785 Sets the maximum width, in characters, of the source line that is displayed
1786 alongside the hex dump. The default value for this parameter is 100. The
1787 source line is displayed on the right hand side of the listing output.
1789 @item --listing-cont-lines=@samp{number}
1790 @kindex --listing-cont-lines
1791 @cindex Maximum number of continuation lines
1792 Sets the maximum number of continuation lines of hex dump that will be
1793 displayed for a given single line of source input. The default value is 4.
1797 @section Assemble in MRI Compatibility Mode: @option{-M}
1800 @cindex MRI compatibility mode
1801 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1802 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1803 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1804 configured target) assembler from Microtec Research. The exact nature of the
1805 MRI syntax will not be documented here; see the MRI manuals for more
1806 information. Note in particular that the handling of macros and macro
1807 arguments is somewhat different. The purpose of this option is to permit
1808 assembling existing MRI assembler code using @command{@value{AS}}.
1810 The MRI compatibility is not complete. Certain operations of the MRI assembler
1811 depend upon its object file format, and can not be supported using other object
1812 file formats. Supporting these would require enhancing each object file format
1813 individually. These are:
1816 @item global symbols in common section
1818 The m68k MRI assembler supports common sections which are merged by the linker.
1819 Other object file formats do not support this. @command{@value{AS}} handles
1820 common sections by treating them as a single common symbol. It permits local
1821 symbols to be defined within a common section, but it can not support global
1822 symbols, since it has no way to describe them.
1824 @item complex relocations
1826 The MRI assemblers support relocations against a negated section address, and
1827 relocations which combine the start addresses of two or more sections. These
1828 are not support by other object file formats.
1830 @item @code{END} pseudo-op specifying start address
1832 The MRI @code{END} pseudo-op permits the specification of a start address.
1833 This is not supported by other object file formats. The start address may
1834 instead be specified using the @option{-e} option to the linker, or in a linker
1837 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1839 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1840 name to the output file. This is not supported by other object file formats.
1842 @item @code{ORG} pseudo-op
1844 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1845 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1846 which changes the location within the current section. Absolute sections are
1847 not supported by other object file formats. The address of a section may be
1848 assigned within a linker script.
1851 There are some other features of the MRI assembler which are not supported by
1852 @command{@value{AS}}, typically either because they are difficult or because they
1853 seem of little consequence. Some of these may be supported in future releases.
1857 @item EBCDIC strings
1859 EBCDIC strings are not supported.
1861 @item packed binary coded decimal
1863 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1864 and @code{DCB.P} pseudo-ops are not supported.
1866 @item @code{FEQU} pseudo-op
1868 The m68k @code{FEQU} pseudo-op is not supported.
1870 @item @code{NOOBJ} pseudo-op
1872 The m68k @code{NOOBJ} pseudo-op is not supported.
1874 @item @code{OPT} branch control options
1876 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1877 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1878 relaxes all branches, whether forward or backward, to an appropriate size, so
1879 these options serve no purpose.
1881 @item @code{OPT} list control options
1883 The following m68k @code{OPT} list control options are ignored: @code{C},
1884 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1885 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1887 @item other @code{OPT} options
1889 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1890 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1892 @item @code{OPT} @code{D} option is default
1894 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1895 @code{OPT NOD} may be used to turn it off.
1897 @item @code{XREF} pseudo-op.
1899 The m68k @code{XREF} pseudo-op is ignored.
1901 @item @code{.debug} pseudo-op
1903 The i960 @code{.debug} pseudo-op is not supported.
1905 @item @code{.extended} pseudo-op
1907 The i960 @code{.extended} pseudo-op is not supported.
1909 @item @code{.list} pseudo-op.
1911 The various options of the i960 @code{.list} pseudo-op are not supported.
1913 @item @code{.optimize} pseudo-op
1915 The i960 @code{.optimize} pseudo-op is not supported.
1917 @item @code{.output} pseudo-op
1919 The i960 @code{.output} pseudo-op is not supported.
1921 @item @code{.setreal} pseudo-op
1923 The i960 @code{.setreal} pseudo-op is not supported.
1928 @section Dependency Tracking: @option{--MD}
1931 @cindex dependency tracking
1934 @command{@value{AS}} can generate a dependency file for the file it creates. This
1935 file consists of a single rule suitable for @code{make} describing the
1936 dependencies of the main source file.
1938 The rule is written to the file named in its argument.
1940 This feature is used in the automatic updating of makefiles.
1943 @section Name the Object File: @option{-o}
1946 @cindex naming object file
1947 @cindex object file name
1948 There is always one object file output when you run @command{@value{AS}}. By
1949 default it has the name
1952 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1966 You use this option (which takes exactly one filename) to give the
1967 object file a different name.
1969 Whatever the object file is called, @command{@value{AS}} overwrites any
1970 existing file of the same name.
1973 @section Join Data and Text Sections: @option{-R}
1976 @cindex data and text sections, joining
1977 @cindex text and data sections, joining
1978 @cindex joining text and data sections
1979 @cindex merging text and data sections
1980 @option{-R} tells @command{@value{AS}} to write the object file as if all
1981 data-section data lives in the text section. This is only done at
1982 the very last moment: your binary data are the same, but data
1983 section parts are relocated differently. The data section part of
1984 your object file is zero bytes long because all its bytes are
1985 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1987 When you specify @option{-R} it would be possible to generate shorter
1988 address displacements (because we do not have to cross between text and
1989 data section). We refrain from doing this simply for compatibility with
1990 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
1993 When @command{@value{AS}} is configured for COFF or ELF output,
1994 this option is only useful if you use sections named @samp{.text} and
1999 @option{-R} is not supported for any of the HPPA targets. Using
2000 @option{-R} generates a warning from @command{@value{AS}}.
2004 @section Display Assembly Statistics: @option{--statistics}
2006 @kindex --statistics
2007 @cindex statistics, about assembly
2008 @cindex time, total for assembly
2009 @cindex space used, maximum for assembly
2010 Use @samp{--statistics} to display two statistics about the resources used by
2011 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2012 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2015 @node traditional-format
2016 @section Compatible Output: @option{--traditional-format}
2018 @kindex --traditional-format
2019 For some targets, the output of @command{@value{AS}} is different in some ways
2020 from the output of some existing assembler. This switch requests
2021 @command{@value{AS}} to use the traditional format instead.
2023 For example, it disables the exception frame optimizations which
2024 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2027 @section Announce Version: @option{-v}
2031 @cindex assembler version
2032 @cindex version of assembler
2033 You can find out what version of as is running by including the
2034 option @samp{-v} (which you can also spell as @samp{-version}) on the
2038 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2040 @command{@value{AS}} should never give a warning or error message when
2041 assembling compiler output. But programs written by people often
2042 cause @command{@value{AS}} to give a warning that a particular assumption was
2043 made. All such warnings are directed to the standard error file.
2047 @cindex suppressing warnings
2048 @cindex warnings, suppressing
2049 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2050 This only affects the warning messages: it does not change any particular of
2051 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2054 @kindex --fatal-warnings
2055 @cindex errors, caused by warnings
2056 @cindex warnings, causing error
2057 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2058 files that generate warnings to be in error.
2061 @cindex warnings, switching on
2062 You can switch these options off again by specifying @option{--warn}, which
2063 causes warnings to be output as usual.
2066 @section Generate Object File in Spite of Errors: @option{-Z}
2067 @cindex object file, after errors
2068 @cindex errors, continuing after
2069 After an error message, @command{@value{AS}} normally produces no output. If for
2070 some reason you are interested in object file output even after
2071 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2072 option. If there are any errors, @command{@value{AS}} continues anyways, and
2073 writes an object file after a final warning message of the form @samp{@var{n}
2074 errors, @var{m} warnings, generating bad object file.}
2079 @cindex machine-independent syntax
2080 @cindex syntax, machine-independent
2081 This chapter describes the machine-independent syntax allowed in a
2082 source file. @command{@value{AS}} syntax is similar to what many other
2083 assemblers use; it is inspired by the BSD 4.2
2088 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2092 * Preprocessing:: Preprocessing
2093 * Whitespace:: Whitespace
2094 * Comments:: Comments
2095 * Symbol Intro:: Symbols
2096 * Statements:: Statements
2097 * Constants:: Constants
2101 @section Preprocessing
2103 @cindex preprocessing
2104 The @command{@value{AS}} internal preprocessor:
2106 @cindex whitespace, removed by preprocessor
2108 adjusts and removes extra whitespace. It leaves one space or tab before
2109 the keywords on a line, and turns any other whitespace on the line into
2112 @cindex comments, removed by preprocessor
2114 removes all comments, replacing them with a single space, or an
2115 appropriate number of newlines.
2117 @cindex constants, converted by preprocessor
2119 converts character constants into the appropriate numeric values.
2122 It does not do macro processing, include file handling, or
2123 anything else you may get from your C compiler's preprocessor. You can
2124 do include file processing with the @code{.include} directive
2125 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2126 to get other ``CPP'' style preprocessing by giving the input file a
2127 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2128 Output, gcc.info, Using GNU CC}.
2130 Excess whitespace, comments, and character constants
2131 cannot be used in the portions of the input text that are not
2134 @cindex turning preprocessing on and off
2135 @cindex preprocessing, turning on and off
2138 If the first line of an input file is @code{#NO_APP} or if you use the
2139 @samp{-f} option, whitespace and comments are not removed from the input file.
2140 Within an input file, you can ask for whitespace and comment removal in
2141 specific portions of the by putting a line that says @code{#APP} before the
2142 text that may contain whitespace or comments, and putting a line that says
2143 @code{#NO_APP} after this text. This feature is mainly intend to support
2144 @code{asm} statements in compilers whose output is otherwise free of comments
2151 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2152 Whitespace is used to separate symbols, and to make programs neater for
2153 people to read. Unless within character constants
2154 (@pxref{Characters,,Character Constants}), any whitespace means the same
2155 as exactly one space.
2161 There are two ways of rendering comments to @command{@value{AS}}. In both
2162 cases the comment is equivalent to one space.
2164 Anything from @samp{/*} through the next @samp{*/} is a comment.
2165 This means you may not nest these comments.
2169 The only way to include a newline ('\n') in a comment
2170 is to use this sort of comment.
2173 /* This sort of comment does not nest. */
2176 @cindex line comment character
2177 Anything from the @dfn{line comment} character to the next newline
2178 is considered a comment and is ignored. The line comment character is
2180 @samp{;} on the ARC;
2183 @samp{@@} on the ARM;
2186 @samp{;} for the H8/300 family;
2189 @samp{;} for the HPPA;
2192 @samp{#} on the i386 and x86-64;
2195 @samp{#} on the i960;
2198 @samp{;} for the PDP-11;
2201 @samp{;} for picoJava;
2204 @samp{#} for Motorola PowerPC;
2207 @samp{!} for the Renesas / SuperH SH;
2210 @samp{!} on the SPARC;
2213 @samp{#} on the ip2k;
2216 @samp{#} on the m32c;
2219 @samp{#} on the m32r;
2222 @samp{|} on the 680x0;
2225 @samp{#} on the 68HC11 and 68HC12;
2228 @samp{#} on the Vax;
2231 @samp{;} for the Z80;
2234 @samp{!} for the Z8000;
2237 @samp{#} on the V850;
2240 @samp{#} for Xtensa systems;
2242 see @ref{Machine Dependencies}. @refill
2243 @c FIXME What about i860?
2246 On some machines there are two different line comment characters. One
2247 character only begins a comment if it is the first non-whitespace character on
2248 a line, while the other always begins a comment.
2252 The V850 assembler also supports a double dash as starting a comment that
2253 extends to the end of the line.
2259 @cindex lines starting with @code{#}
2260 @cindex logical line numbers
2261 To be compatible with past assemblers, lines that begin with @samp{#} have a
2262 special interpretation. Following the @samp{#} should be an absolute
2263 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2264 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2265 new logical file name. The rest of the line, if any, should be whitespace.
2267 If the first non-whitespace characters on the line are not numeric,
2268 the line is ignored. (Just like a comment.)
2271 # This is an ordinary comment.
2272 # 42-6 "new_file_name" # New logical file name
2273 # This is logical line # 36.
2275 This feature is deprecated, and may disappear from future versions
2276 of @command{@value{AS}}.
2281 @cindex characters used in symbols
2282 @ifclear SPECIAL-SYMS
2283 A @dfn{symbol} is one or more characters chosen from the set of all
2284 letters (both upper and lower case), digits and the three characters
2290 A @dfn{symbol} is one or more characters chosen from the set of all
2291 letters (both upper and lower case), digits and the three characters
2292 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2298 On most machines, you can also use @code{$} in symbol names; exceptions
2299 are noted in @ref{Machine Dependencies}.
2301 No symbol may begin with a digit. Case is significant.
2302 There is no length limit: all characters are significant. Symbols are
2303 delimited by characters not in that set, or by the beginning of a file
2304 (since the source program must end with a newline, the end of a file is
2305 not a possible symbol delimiter). @xref{Symbols}.
2306 @cindex length of symbols
2311 @cindex statements, structure of
2312 @cindex line separator character
2313 @cindex statement separator character
2315 @ifclear abnormal-separator
2316 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2317 semicolon (@samp{;}). The newline or semicolon is considered part of
2318 the preceding statement. Newlines and semicolons within character
2319 constants are an exception: they do not end statements.
2321 @ifset abnormal-separator
2323 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2324 point (@samp{!}). The newline or exclamation point is considered part of the
2325 preceding statement. Newlines and exclamation points within character
2326 constants are an exception: they do not end statements.
2329 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2330 H8/300) a dollar sign (@samp{$}); or (for the Renesas-SH) a semicolon
2331 (@samp{;}). The newline or separator character is considered part of
2332 the preceding statement. Newlines and separators within character
2333 constants are an exception: they do not end statements.
2338 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2339 separator character. (The line separator is usually @samp{;}, unless this
2340 conflicts with the comment character; see @ref{Machine Dependencies}.) The
2341 newline or separator character is considered part of the preceding
2342 statement. Newlines and separators within character constants are an
2343 exception: they do not end statements.
2346 @cindex newline, required at file end
2347 @cindex EOF, newline must precede
2348 It is an error to end any statement with end-of-file: the last
2349 character of any input file should be a newline.@refill
2351 An empty statement is allowed, and may include whitespace. It is ignored.
2353 @cindex instructions and directives
2354 @cindex directives and instructions
2355 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2356 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2358 A statement begins with zero or more labels, optionally followed by a
2359 key symbol which determines what kind of statement it is. The key
2360 symbol determines the syntax of the rest of the statement. If the
2361 symbol begins with a dot @samp{.} then the statement is an assembler
2362 directive: typically valid for any computer. If the symbol begins with
2363 a letter the statement is an assembly language @dfn{instruction}: it
2364 assembles into a machine language instruction.
2366 Different versions of @command{@value{AS}} for different computers
2367 recognize different instructions. In fact, the same symbol may
2368 represent a different instruction in a different computer's assembly
2372 @cindex @code{:} (label)
2373 @cindex label (@code{:})
2374 A label is a symbol immediately followed by a colon (@code{:}).
2375 Whitespace before a label or after a colon is permitted, but you may not
2376 have whitespace between a label's symbol and its colon. @xref{Labels}.
2379 For HPPA targets, labels need not be immediately followed by a colon, but
2380 the definition of a label must begin in column zero. This also implies that
2381 only one label may be defined on each line.
2385 label: .directive followed by something
2386 another_label: # This is an empty statement.
2387 instruction operand_1, operand_2, @dots{}
2394 A constant is a number, written so that its value is known by
2395 inspection, without knowing any context. Like this:
2398 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2399 .ascii "Ring the bell\7" # A string constant.
2400 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2401 .float 0f-314159265358979323846264338327\
2402 95028841971.693993751E-40 # - pi, a flonum.
2407 * Characters:: Character Constants
2408 * Numbers:: Number Constants
2412 @subsection Character Constants
2414 @cindex character constants
2415 @cindex constants, character
2416 There are two kinds of character constants. A @dfn{character} stands
2417 for one character in one byte and its value may be used in
2418 numeric expressions. String constants (properly called string
2419 @emph{literals}) are potentially many bytes and their values may not be
2420 used in arithmetic expressions.
2424 * Chars:: Characters
2428 @subsubsection Strings
2430 @cindex string constants
2431 @cindex constants, string
2432 A @dfn{string} is written between double-quotes. It may contain
2433 double-quotes or null characters. The way to get special characters
2434 into a string is to @dfn{escape} these characters: precede them with
2435 a backslash @samp{\} character. For example @samp{\\} represents
2436 one backslash: the first @code{\} is an escape which tells
2437 @command{@value{AS}} to interpret the second character literally as a backslash
2438 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2439 escape character). The complete list of escapes follows.
2441 @cindex escape codes, character
2442 @cindex character escape codes
2445 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2447 @cindex @code{\b} (backspace character)
2448 @cindex backspace (@code{\b})
2450 Mnemonic for backspace; for ASCII this is octal code 010.
2453 @c Mnemonic for EOText; for ASCII this is octal code 004.
2455 @cindex @code{\f} (formfeed character)
2456 @cindex formfeed (@code{\f})
2458 Mnemonic for FormFeed; for ASCII this is octal code 014.
2460 @cindex @code{\n} (newline character)
2461 @cindex newline (@code{\n})
2463 Mnemonic for newline; for ASCII this is octal code 012.
2466 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2468 @cindex @code{\r} (carriage return character)
2469 @cindex carriage return (@code{\r})
2471 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2474 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2475 @c other assemblers.
2477 @cindex @code{\t} (tab)
2478 @cindex tab (@code{\t})
2480 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2483 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2484 @c @item \x @var{digit} @var{digit} @var{digit}
2485 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2487 @cindex @code{\@var{ddd}} (octal character code)
2488 @cindex octal character code (@code{\@var{ddd}})
2489 @item \ @var{digit} @var{digit} @var{digit}
2490 An octal character code. The numeric code is 3 octal digits.
2491 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2492 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2494 @cindex @code{\@var{xd...}} (hex character code)
2495 @cindex hex character code (@code{\@var{xd...}})
2496 @item \@code{x} @var{hex-digits...}
2497 A hex character code. All trailing hex digits are combined. Either upper or
2498 lower case @code{x} works.
2500 @cindex @code{\\} (@samp{\} character)
2501 @cindex backslash (@code{\\})
2503 Represents one @samp{\} character.
2506 @c Represents one @samp{'} (accent acute) character.
2507 @c This is needed in single character literals
2508 @c (@xref{Characters,,Character Constants}.) to represent
2511 @cindex @code{\"} (doublequote character)
2512 @cindex doublequote (@code{\"})
2514 Represents one @samp{"} character. Needed in strings to represent
2515 this character, because an unescaped @samp{"} would end the string.
2517 @item \ @var{anything-else}
2518 Any other character when escaped by @kbd{\} gives a warning, but
2519 assembles as if the @samp{\} was not present. The idea is that if
2520 you used an escape sequence you clearly didn't want the literal
2521 interpretation of the following character. However @command{@value{AS}} has no
2522 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2523 code and warns you of the fact.
2526 Which characters are escapable, and what those escapes represent,
2527 varies widely among assemblers. The current set is what we think
2528 the BSD 4.2 assembler recognizes, and is a subset of what most C
2529 compilers recognize. If you are in doubt, do not use an escape
2533 @subsubsection Characters
2535 @cindex single character constant
2536 @cindex character, single
2537 @cindex constant, single character
2538 A single character may be written as a single quote immediately
2539 followed by that character. The same escapes apply to characters as
2540 to strings. So if you want to write the character backslash, you
2541 must write @kbd{'\\} where the first @code{\} escapes the second
2542 @code{\}. As you can see, the quote is an acute accent, not a
2543 grave accent. A newline
2545 @ifclear abnormal-separator
2546 (or semicolon @samp{;})
2548 @ifset abnormal-separator
2550 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2555 immediately following an acute accent is taken as a literal character
2556 and does not count as the end of a statement. The value of a character
2557 constant in a numeric expression is the machine's byte-wide code for
2558 that character. @command{@value{AS}} assumes your character code is ASCII:
2559 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2562 @subsection Number Constants
2564 @cindex constants, number
2565 @cindex number constants
2566 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2567 are stored in the target machine. @emph{Integers} are numbers that
2568 would fit into an @code{int} in the C language. @emph{Bignums} are
2569 integers, but they are stored in more than 32 bits. @emph{Flonums}
2570 are floating point numbers, described below.
2573 * Integers:: Integers
2578 * Bit Fields:: Bit Fields
2584 @subsubsection Integers
2586 @cindex constants, integer
2588 @cindex binary integers
2589 @cindex integers, binary
2590 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2591 the binary digits @samp{01}.
2593 @cindex octal integers
2594 @cindex integers, octal
2595 An octal integer is @samp{0} followed by zero or more of the octal
2596 digits (@samp{01234567}).
2598 @cindex decimal integers
2599 @cindex integers, decimal
2600 A decimal integer starts with a non-zero digit followed by zero or
2601 more digits (@samp{0123456789}).
2603 @cindex hexadecimal integers
2604 @cindex integers, hexadecimal
2605 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2606 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2608 Integers have the usual values. To denote a negative integer, use
2609 the prefix operator @samp{-} discussed under expressions
2610 (@pxref{Prefix Ops,,Prefix Operators}).
2613 @subsubsection Bignums
2616 @cindex constants, bignum
2617 A @dfn{bignum} has the same syntax and semantics as an integer
2618 except that the number (or its negative) takes more than 32 bits to
2619 represent in binary. The distinction is made because in some places
2620 integers are permitted while bignums are not.
2623 @subsubsection Flonums
2625 @cindex floating point numbers
2626 @cindex constants, floating point
2628 @cindex precision, floating point
2629 A @dfn{flonum} represents a floating point number. The translation is
2630 indirect: a decimal floating point number from the text is converted by
2631 @command{@value{AS}} to a generic binary floating point number of more than
2632 sufficient precision. This generic floating point number is converted
2633 to a particular computer's floating point format (or formats) by a
2634 portion of @command{@value{AS}} specialized to that computer.
2636 A flonum is written by writing (in order)
2641 (@samp{0} is optional on the HPPA.)
2645 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2647 @kbd{e} is recommended. Case is not important.
2649 @c FIXME: verify if flonum syntax really this vague for most cases
2650 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2651 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2654 On the H8/300, Renesas / SuperH SH,
2655 and AMD 29K architectures, the letter must be
2656 one of the letters @samp{DFPRSX} (in upper or lower case).
2658 On the ARC, the letter must be one of the letters @samp{DFRS}
2659 (in upper or lower case).
2661 On the Intel 960 architecture, the letter must be
2662 one of the letters @samp{DFT} (in upper or lower case).
2664 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2668 One of the letters @samp{DFRS} (in upper or lower case).
2671 One of the letters @samp{DFPRSX} (in upper or lower case).
2674 The letter @samp{E} (upper case only).
2677 One of the letters @samp{DFT} (in upper or lower case).
2682 An optional sign: either @samp{+} or @samp{-}.
2685 An optional @dfn{integer part}: zero or more decimal digits.
2688 An optional @dfn{fractional part}: @samp{.} followed by zero
2689 or more decimal digits.
2692 An optional exponent, consisting of:
2696 An @samp{E} or @samp{e}.
2697 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2698 @c principle this can perfectly well be different on different targets.
2700 Optional sign: either @samp{+} or @samp{-}.
2702 One or more decimal digits.
2707 At least one of the integer part or the fractional part must be
2708 present. The floating point number has the usual base-10 value.
2710 @command{@value{AS}} does all processing using integers. Flonums are computed
2711 independently of any floating point hardware in the computer running
2712 @command{@value{AS}}.
2716 @c Bit fields are written as a general facility but are also controlled
2717 @c by a conditional-compilation flag---which is as of now (21mar91)
2718 @c turned on only by the i960 config of GAS.
2720 @subsubsection Bit Fields
2723 @cindex constants, bit field
2724 You can also define numeric constants as @dfn{bit fields}.
2725 Specify two numbers separated by a colon---
2727 @var{mask}:@var{value}
2730 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2733 The resulting number is then packed
2735 @c this conditional paren in case bit fields turned on elsewhere than 960
2736 (in host-dependent byte order)
2738 into a field whose width depends on which assembler directive has the
2739 bit-field as its argument. Overflow (a result from the bitwise and
2740 requiring more binary digits to represent) is not an error; instead,
2741 more constants are generated, of the specified width, beginning with the
2742 least significant digits.@refill
2744 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2745 @code{.short}, and @code{.word} accept bit-field arguments.
2750 @chapter Sections and Relocation
2755 * Secs Background:: Background
2756 * Ld Sections:: Linker Sections
2757 * As Sections:: Assembler Internal Sections
2758 * Sub-Sections:: Sub-Sections
2762 @node Secs Background
2765 Roughly, a section is a range of addresses, with no gaps; all data
2766 ``in'' those addresses is treated the same for some particular purpose.
2767 For example there may be a ``read only'' section.
2769 @cindex linker, and assembler
2770 @cindex assembler, and linker
2771 The linker @code{@value{LD}} reads many object files (partial programs) and
2772 combines their contents to form a runnable program. When @command{@value{AS}}
2773 emits an object file, the partial program is assumed to start at address 0.
2774 @code{@value{LD}} assigns the final addresses for the partial program, so that
2775 different partial programs do not overlap. This is actually an
2776 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2779 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2780 addresses. These blocks slide to their run-time addresses as rigid
2781 units; their length does not change and neither does the order of bytes
2782 within them. Such a rigid unit is called a @emph{section}. Assigning
2783 run-time addresses to sections is called @dfn{relocation}. It includes
2784 the task of adjusting mentions of object-file addresses so they refer to
2785 the proper run-time addresses.
2787 For the H8/300, and for the Renesas / SuperH SH,
2788 @command{@value{AS}} pads sections if needed to
2789 ensure they end on a word (sixteen bit) boundary.
2792 @cindex standard assembler sections
2793 An object file written by @command{@value{AS}} has at least three sections, any
2794 of which may be empty. These are named @dfn{text}, @dfn{data} and
2799 When it generates COFF or ELF output,
2801 @command{@value{AS}} can also generate whatever other named sections you specify
2802 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2803 If you do not use any directives that place output in the @samp{.text}
2804 or @samp{.data} sections, these sections still exist, but are empty.
2809 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2811 @command{@value{AS}} can also generate whatever other named sections you
2812 specify using the @samp{.space} and @samp{.subspace} directives. See
2813 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2814 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2815 assembler directives.
2818 Additionally, @command{@value{AS}} uses different names for the standard
2819 text, data, and bss sections when generating SOM output. Program text
2820 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2821 BSS into @samp{$BSS$}.
2825 Within the object file, the text section starts at address @code{0}, the
2826 data section follows, and the bss section follows the data section.
2829 When generating either SOM or ELF output files on the HPPA, the text
2830 section starts at address @code{0}, the data section at address
2831 @code{0x4000000}, and the bss section follows the data section.
2834 To let @code{@value{LD}} know which data changes when the sections are
2835 relocated, and how to change that data, @command{@value{AS}} also writes to the
2836 object file details of the relocation needed. To perform relocation
2837 @code{@value{LD}} must know, each time an address in the object
2841 Where in the object file is the beginning of this reference to
2844 How long (in bytes) is this reference?
2846 Which section does the address refer to? What is the numeric value of
2848 (@var{address}) @minus{} (@var{start-address of section})?
2851 Is the reference to an address ``Program-Counter relative''?
2854 @cindex addresses, format of
2855 @cindex section-relative addressing
2856 In fact, every address @command{@value{AS}} ever uses is expressed as
2858 (@var{section}) + (@var{offset into section})
2861 Further, most expressions @command{@value{AS}} computes have this section-relative
2864 (For some object formats, such as SOM for the HPPA, some expressions are
2865 symbol-relative instead.)
2868 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2869 @var{N} into section @var{secname}.''
2871 Apart from text, data and bss sections you need to know about the
2872 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2873 addresses in the absolute section remain unchanged. For example, address
2874 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2875 @code{@value{LD}}. Although the linker never arranges two partial programs'
2876 data sections with overlapping addresses after linking, @emph{by definition}
2877 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2878 part of a program is always the same address when the program is running as
2879 address @code{@{absolute@ 239@}} in any other part of the program.
2881 The idea of sections is extended to the @dfn{undefined} section. Any
2882 address whose section is unknown at assembly time is by definition
2883 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2884 Since numbers are always defined, the only way to generate an undefined
2885 address is to mention an undefined symbol. A reference to a named
2886 common block would be such a symbol: its value is unknown at assembly
2887 time so it has section @emph{undefined}.
2889 By analogy the word @emph{section} is used to describe groups of sections in
2890 the linked program. @code{@value{LD}} puts all partial programs' text
2891 sections in contiguous addresses in the linked program. It is
2892 customary to refer to the @emph{text section} of a program, meaning all
2893 the addresses of all partial programs' text sections. Likewise for
2894 data and bss sections.
2896 Some sections are manipulated by @code{@value{LD}}; others are invented for
2897 use of @command{@value{AS}} and have no meaning except during assembly.
2900 @section Linker Sections
2901 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2906 @cindex named sections
2907 @cindex sections, named
2908 @item named sections
2911 @cindex text section
2912 @cindex data section
2916 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2917 separate but equal sections. Anything you can say of one section is
2920 When the program is running, however, it is
2921 customary for the text section to be unalterable. The
2922 text section is often shared among processes: it contains
2923 instructions, constants and the like. The data section of a running
2924 program is usually alterable: for example, C variables would be stored
2925 in the data section.
2930 This section contains zeroed bytes when your program begins running. It
2931 is used to hold uninitialized variables or common storage. The length of
2932 each partial program's bss section is important, but because it starts
2933 out containing zeroed bytes there is no need to store explicit zero
2934 bytes in the object file. The bss section was invented to eliminate
2935 those explicit zeros from object files.
2937 @cindex absolute section
2938 @item absolute section
2939 Address 0 of this section is always ``relocated'' to runtime address 0.
2940 This is useful if you want to refer to an address that @code{@value{LD}} must
2941 not change when relocating. In this sense we speak of absolute
2942 addresses being ``unrelocatable'': they do not change during relocation.
2944 @cindex undefined section
2945 @item undefined section
2946 This ``section'' is a catch-all for address references to objects not in
2947 the preceding sections.
2948 @c FIXME: ref to some other doc on obj-file formats could go here.
2951 @cindex relocation example
2952 An idealized example of three relocatable sections follows.
2954 The example uses the traditional section names @samp{.text} and @samp{.data}.
2956 Memory addresses are on the horizontal axis.
2960 @c END TEXI2ROFF-KILL
2963 partial program # 1: |ttttt|dddd|00|
2970 partial program # 2: |TTT|DDD|000|
2973 +--+---+-----+--+----+---+-----+~~
2974 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2975 +--+---+-----+--+----+---+-----+~~
2977 addresses: 0 @dots{}
2984 \line{\it Partial program \#1: \hfil}
2985 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2986 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2988 \line{\it Partial program \#2: \hfil}
2989 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2990 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2992 \line{\it linked program: \hfil}
2993 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2994 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2995 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2996 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2998 \line{\it addresses: \hfil}
3002 @c END TEXI2ROFF-KILL
3005 @section Assembler Internal Sections
3007 @cindex internal assembler sections
3008 @cindex sections in messages, internal
3009 These sections are meant only for the internal use of @command{@value{AS}}. They
3010 have no meaning at run-time. You do not really need to know about these
3011 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3012 warning messages, so it might be helpful to have an idea of their
3013 meanings to @command{@value{AS}}. These sections are used to permit the
3014 value of every expression in your assembly language program to be a
3015 section-relative address.
3018 @cindex assembler internal logic error
3019 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3020 An internal assembler logic error has been found. This means there is a
3021 bug in the assembler.
3023 @cindex expr (internal section)
3025 The assembler stores complex expression internally as combinations of
3026 symbols. When it needs to represent an expression as a symbol, it puts
3027 it in the expr section.
3029 @c FIXME item transfer[t] vector preload
3030 @c FIXME item transfer[t] vector postload
3031 @c FIXME item register
3035 @section Sub-Sections
3037 @cindex numbered subsections
3038 @cindex grouping data
3044 fall into two sections: text and data.
3046 You may have separate groups of
3048 data in named sections
3052 data in named sections
3058 that you want to end up near to each other in the object file, even though they
3059 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3060 use @dfn{subsections} for this purpose. Within each section, there can be
3061 numbered subsections with values from 0 to 8192. Objects assembled into the
3062 same subsection go into the object file together with other objects in the same
3063 subsection. For example, a compiler might want to store constants in the text
3064 section, but might not want to have them interspersed with the program being
3065 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3066 section of code being output, and a @samp{.text 1} before each group of
3067 constants being output.
3069 Subsections are optional. If you do not use subsections, everything
3070 goes in subsection number zero.
3073 Each subsection is zero-padded up to a multiple of four bytes.
3074 (Subsections may be padded a different amount on different flavors
3075 of @command{@value{AS}}.)
3079 On the H8/300 platform, each subsection is zero-padded to a word
3080 boundary (two bytes).
3081 The same is true on the Renesas SH.
3084 @c FIXME section padding (alignment)?
3085 @c Rich Pixley says padding here depends on target obj code format; that
3086 @c doesn't seem particularly useful to say without further elaboration,
3087 @c so for now I say nothing about it. If this is a generic BFD issue,
3088 @c these paragraphs might need to vanish from this manual, and be
3089 @c discussed in BFD chapter of binutils (or some such).
3093 Subsections appear in your object file in numeric order, lowest numbered
3094 to highest. (All this to be compatible with other people's assemblers.)
3095 The object file contains no representation of subsections; @code{@value{LD}} and
3096 other programs that manipulate object files see no trace of them.
3097 They just see all your text subsections as a text section, and all your
3098 data subsections as a data section.
3100 To specify which subsection you want subsequent statements assembled
3101 into, use a numeric argument to specify it, in a @samp{.text
3102 @var{expression}} or a @samp{.data @var{expression}} statement.
3105 When generating COFF output, you
3110 can also use an extra subsection
3111 argument with arbitrary named sections: @samp{.section @var{name},
3116 When generating ELF output, you
3121 can also use the @code{.subsection} directive (@pxref{SubSection})
3122 to specify a subsection: @samp{.subsection @var{expression}}.
3124 @var{Expression} should be an absolute expression
3125 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3126 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3127 begins in @code{text 0}. For instance:
3129 .text 0 # The default subsection is text 0 anyway.
3130 .ascii "This lives in the first text subsection. *"
3132 .ascii "But this lives in the second text subsection."
3134 .ascii "This lives in the data section,"
3135 .ascii "in the first data subsection."
3137 .ascii "This lives in the first text section,"
3138 .ascii "immediately following the asterisk (*)."
3141 Each section has a @dfn{location counter} incremented by one for every byte
3142 assembled into that section. Because subsections are merely a convenience
3143 restricted to @command{@value{AS}} there is no concept of a subsection location
3144 counter. There is no way to directly manipulate a location counter---but the
3145 @code{.align} directive changes it, and any label definition captures its
3146 current value. The location counter of the section where statements are being
3147 assembled is said to be the @dfn{active} location counter.
3150 @section bss Section
3153 @cindex common variable storage
3154 The bss section is used for local common variable storage.
3155 You may allocate address space in the bss section, but you may
3156 not dictate data to load into it before your program executes. When
3157 your program starts running, all the contents of the bss
3158 section are zeroed bytes.
3160 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3161 @ref{Lcomm,,@code{.lcomm}}.
3163 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3164 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3167 When assembling for a target which supports multiple sections, such as ELF or
3168 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3169 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3170 section. Typically the section will only contain symbol definitions and
3171 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3178 Symbols are a central concept: the programmer uses symbols to name
3179 things, the linker uses symbols to link, and the debugger uses symbols
3183 @cindex debuggers, and symbol order
3184 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3185 the same order they were declared. This may break some debuggers.
3190 * Setting Symbols:: Giving Symbols Other Values
3191 * Symbol Names:: Symbol Names
3192 * Dot:: The Special Dot Symbol
3193 * Symbol Attributes:: Symbol Attributes
3200 A @dfn{label} is written as a symbol immediately followed by a colon
3201 @samp{:}. The symbol then represents the current value of the
3202 active location counter, and is, for example, a suitable instruction
3203 operand. You are warned if you use the same symbol to represent two
3204 different locations: the first definition overrides any other
3208 On the HPPA, the usual form for a label need not be immediately followed by a
3209 colon, but instead must start in column zero. Only one label may be defined on
3210 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3211 provides a special directive @code{.label} for defining labels more flexibly.
3214 @node Setting Symbols
3215 @section Giving Symbols Other Values
3217 @cindex assigning values to symbols
3218 @cindex symbol values, assigning
3219 A symbol can be given an arbitrary value by writing a symbol, followed
3220 by an equals sign @samp{=}, followed by an expression
3221 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3222 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3223 equals sign @samp{=}@samp{=} here represents an equivalent of the
3224 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3227 @section Symbol Names
3229 @cindex symbol names
3230 @cindex names, symbol
3231 @ifclear SPECIAL-SYMS
3232 Symbol names begin with a letter or with one of @samp{._}. On most
3233 machines, you can also use @code{$} in symbol names; exceptions are
3234 noted in @ref{Machine Dependencies}. That character may be followed by any
3235 string of digits, letters, dollar signs (unless otherwise noted for a
3236 particular target machine), and underscores.
3240 Symbol names begin with a letter or with one of @samp{._}. On the
3241 Renesas SH you can also use @code{$} in symbol names. That
3242 character may be followed by any string of digits, letters, dollar signs (save
3243 on the H8/300), and underscores.
3247 Case of letters is significant: @code{foo} is a different symbol name
3250 Each symbol has exactly one name. Each name in an assembly language program
3251 refers to exactly one symbol. You may use that symbol name any number of times
3254 @subheading Local Symbol Names
3256 @cindex local symbol names
3257 @cindex symbol names, local
3258 A local symbol is any symbol beginning with certain local label prefixes.
3259 By default, the local label prefix is @samp{.L} for ELF systems or
3260 @samp{L} for traditional a.out systems, but each target may have its own
3261 set of local label prefixes.
3263 On the HPPA local symbols begin with @samp{L$}.
3266 Local symbols are defined and used within the assembler, but they are
3267 normally not saved in object files. Thus, they are not visible when debugging.
3268 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3269 @option{-L}}) to retain the local symbols in the object files.
3271 @subheading Local Labels
3273 @cindex local labels
3274 @cindex temporary symbol names
3275 @cindex symbol names, temporary
3276 Local labels help compilers and programmers use names temporarily.
3277 They create symbols which are guaranteed to be unique over the entire scope of
3278 the input source code and which can be referred to by a simple notation.
3279 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3280 represents any positive integer). To refer to the most recent previous
3281 definition of that label write @samp{@b{N}b}, using the same number as when
3282 you defined the label. To refer to the next definition of a local label, write
3283 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3286 There is no restriction on how you can use these labels, and you can reuse them
3287 too. So that it is possible to repeatedly define the same local label (using
3288 the same number @samp{@b{N}}), although you can only refer to the most recently
3289 defined local label of that number (for a backwards reference) or the next
3290 definition of a specific local label for a forward reference. It is also worth
3291 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3292 implemented in a slightly more efficient manner than the others.
3303 Which is the equivalent of:
3306 label_1: branch label_3
3307 label_2: branch label_1
3308 label_3: branch label_4
3309 label_4: branch label_3
3312 Local label names are only a notational device. They are immediately
3313 transformed into more conventional symbol names before the assembler uses them.
3314 The symbol names are stored in the symbol table, appear in error messages, and
3315 are optionally emitted to the object file. The names are constructed using
3319 @item @emph{local label prefix}
3320 All local symbols begin with the system-specific local label prefix.
3321 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3322 that start with the local label prefix. These labels are
3323 used for symbols you are never intended to see. If you use the
3324 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3325 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3326 you may use them in debugging.
3329 This is the number that was used in the local label definition. So if the
3330 label is written @samp{55:} then the number is @samp{55}.
3333 This unusual character is included so you do not accidentally invent a symbol
3334 of the same name. The character has ASCII value of @samp{\002} (control-B).
3336 @item @emph{ordinal number}
3337 This is a serial number to keep the labels distinct. The first definition of
3338 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3339 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3340 the number @samp{1} and its 15th definition gets @samp{15} as well.
3343 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3344 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3346 @subheading Dollar Local Labels
3347 @cindex dollar local symbols
3349 @code{@value{AS}} also supports an even more local form of local labels called
3350 dollar labels. These labels go out of scope (i.e., they become undefined) as
3351 soon as a non-local label is defined. Thus they remain valid for only a small
3352 region of the input source code. Normal local labels, by contrast, remain in
3353 scope for the entire file, or until they are redefined by another occurrence of
3354 the same local label.
3356 Dollar labels are defined in exactly the same way as ordinary local labels,
3357 except that instead of being terminated by a colon, they are terminated by a
3358 dollar sign, e.g., @samp{@b{55$}}.
3360 They can also be distinguished from ordinary local labels by their transformed
3361 names which use ASCII character @samp{\001} (control-A) as the magic character
3362 to distinguish them from ordinary labels. For example, the fifth definition of
3363 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3366 @section The Special Dot Symbol
3368 @cindex dot (symbol)
3369 @cindex @code{.} (symbol)
3370 @cindex current address
3371 @cindex location counter
3372 The special symbol @samp{.} refers to the current address that
3373 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3374 .long .} defines @code{melvin} to contain its own address.
3375 Assigning a value to @code{.} is treated the same as a @code{.org}
3376 directive. Thus, the expression @samp{.=.+4} is the same as saying
3377 @ifclear no-space-dir
3381 @node Symbol Attributes
3382 @section Symbol Attributes
3384 @cindex symbol attributes
3385 @cindex attributes, symbol
3386 Every symbol has, as well as its name, the attributes ``Value'' and
3387 ``Type''. Depending on output format, symbols can also have auxiliary
3390 The detailed definitions are in @file{a.out.h}.
3393 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3394 all these attributes, and probably won't warn you. This makes the
3395 symbol an externally defined symbol, which is generally what you
3399 * Symbol Value:: Value
3400 * Symbol Type:: Type
3403 * a.out Symbols:: Symbol Attributes: @code{a.out}
3407 * a.out Symbols:: Symbol Attributes: @code{a.out}
3410 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3415 * COFF Symbols:: Symbol Attributes for COFF
3418 * SOM Symbols:: Symbol Attributes for SOM
3425 @cindex value of a symbol
3426 @cindex symbol value
3427 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3428 location in the text, data, bss or absolute sections the value is the
3429 number of addresses from the start of that section to the label.
3430 Naturally for text, data and bss sections the value of a symbol changes
3431 as @code{@value{LD}} changes section base addresses during linking. Absolute
3432 symbols' values do not change during linking: that is why they are
3435 The value of an undefined symbol is treated in a special way. If it is
3436 0 then the symbol is not defined in this assembler source file, and
3437 @code{@value{LD}} tries to determine its value from other files linked into the
3438 same program. You make this kind of symbol simply by mentioning a symbol
3439 name without defining it. A non-zero value represents a @code{.comm}
3440 common declaration. The value is how much common storage to reserve, in
3441 bytes (addresses). The symbol refers to the first address of the
3447 @cindex type of a symbol
3449 The type attribute of a symbol contains relocation (section)
3450 information, any flag settings indicating that a symbol is external, and
3451 (optionally), other information for linkers and debuggers. The exact
3452 format depends on the object-code output format in use.
3457 @c The following avoids a "widow" subsection title. @group would be
3458 @c better if it were available outside examples.
3461 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3463 @cindex @code{b.out} symbol attributes
3464 @cindex symbol attributes, @code{b.out}
3465 These symbol attributes appear only when @command{@value{AS}} is configured for
3466 one of the Berkeley-descended object output formats---@code{a.out} or
3472 @subsection Symbol Attributes: @code{a.out}
3474 @cindex @code{a.out} symbol attributes
3475 @cindex symbol attributes, @code{a.out}
3481 @subsection Symbol Attributes: @code{a.out}
3483 @cindex @code{a.out} symbol attributes
3484 @cindex symbol attributes, @code{a.out}
3488 * Symbol Desc:: Descriptor
3489 * Symbol Other:: Other
3493 @subsubsection Descriptor
3495 @cindex descriptor, of @code{a.out} symbol
3496 This is an arbitrary 16-bit value. You may establish a symbol's
3497 descriptor value by using a @code{.desc} statement
3498 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3499 @command{@value{AS}}.
3502 @subsubsection Other
3504 @cindex other attribute, of @code{a.out} symbol
3505 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3510 @subsection Symbol Attributes for COFF
3512 @cindex COFF symbol attributes
3513 @cindex symbol attributes, COFF
3515 The COFF format supports a multitude of auxiliary symbol attributes;
3516 like the primary symbol attributes, they are set between @code{.def} and
3517 @code{.endef} directives.
3519 @subsubsection Primary Attributes
3521 @cindex primary attributes, COFF symbols
3522 The symbol name is set with @code{.def}; the value and type,
3523 respectively, with @code{.val} and @code{.type}.
3525 @subsubsection Auxiliary Attributes
3527 @cindex auxiliary attributes, COFF symbols
3528 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3529 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3530 table information for COFF.
3535 @subsection Symbol Attributes for SOM
3537 @cindex SOM symbol attributes
3538 @cindex symbol attributes, SOM
3540 The SOM format for the HPPA supports a multitude of symbol attributes set with
3541 the @code{.EXPORT} and @code{.IMPORT} directives.
3543 The attributes are described in @cite{HP9000 Series 800 Assembly
3544 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3545 @code{EXPORT} assembler directive documentation.
3549 @chapter Expressions
3553 @cindex numeric values
3554 An @dfn{expression} specifies an address or numeric value.
3555 Whitespace may precede and/or follow an expression.
3557 The result of an expression must be an absolute number, or else an offset into
3558 a particular section. If an expression is not absolute, and there is not
3559 enough information when @command{@value{AS}} sees the expression to know its
3560 section, a second pass over the source program might be necessary to interpret
3561 the expression---but the second pass is currently not implemented.
3562 @command{@value{AS}} aborts with an error message in this situation.
3565 * Empty Exprs:: Empty Expressions
3566 * Integer Exprs:: Integer Expressions
3570 @section Empty Expressions
3572 @cindex empty expressions
3573 @cindex expressions, empty
3574 An empty expression has no value: it is just whitespace or null.
3575 Wherever an absolute expression is required, you may omit the
3576 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3577 is compatible with other assemblers.
3580 @section Integer Expressions
3582 @cindex integer expressions
3583 @cindex expressions, integer
3584 An @dfn{integer expression} is one or more @emph{arguments} delimited
3585 by @emph{operators}.
3588 * Arguments:: Arguments
3589 * Operators:: Operators
3590 * Prefix Ops:: Prefix Operators
3591 * Infix Ops:: Infix Operators
3595 @subsection Arguments
3597 @cindex expression arguments
3598 @cindex arguments in expressions
3599 @cindex operands in expressions
3600 @cindex arithmetic operands
3601 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3602 contexts arguments are sometimes called ``arithmetic operands''. In
3603 this manual, to avoid confusing them with the ``instruction operands'' of
3604 the machine language, we use the term ``argument'' to refer to parts of
3605 expressions only, reserving the word ``operand'' to refer only to machine
3606 instruction operands.
3608 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3609 @var{section} is one of text, data, bss, absolute,
3610 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3613 Numbers are usually integers.
3615 A number can be a flonum or bignum. In this case, you are warned
3616 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3617 these 32 bits are an integer. You may write integer-manipulating
3618 instructions that act on exotic constants, compatible with other
3621 @cindex subexpressions
3622 Subexpressions are a left parenthesis @samp{(} followed by an integer
3623 expression, followed by a right parenthesis @samp{)}; or a prefix
3624 operator followed by an argument.
3627 @subsection Operators
3629 @cindex operators, in expressions
3630 @cindex arithmetic functions
3631 @cindex functions, in expressions
3632 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3633 operators are followed by an argument. Infix operators appear
3634 between their arguments. Operators may be preceded and/or followed by
3638 @subsection Prefix Operator
3640 @cindex prefix operators
3641 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3642 one argument, which must be absolute.
3644 @c the tex/end tex stuff surrounding this small table is meant to make
3645 @c it align, on the printed page, with the similar table in the next
3646 @c section (which is inside an enumerate).
3648 \global\advance\leftskip by \itemindent
3653 @dfn{Negation}. Two's complement negation.
3655 @dfn{Complementation}. Bitwise not.
3659 \global\advance\leftskip by -\itemindent
3663 @subsection Infix Operators
3665 @cindex infix operators
3666 @cindex operators, permitted arguments
3667 @dfn{Infix operators} take two arguments, one on either side. Operators
3668 have precedence, but operations with equal precedence are performed left
3669 to right. Apart from @code{+} or @option{-}, both arguments must be
3670 absolute, and the result is absolute.
3673 @cindex operator precedence
3674 @cindex precedence of operators
3681 @dfn{Multiplication}.
3684 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3690 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3693 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3697 Intermediate precedence
3702 @dfn{Bitwise Inclusive Or}.
3708 @dfn{Bitwise Exclusive Or}.
3711 @dfn{Bitwise Or Not}.
3718 @cindex addition, permitted arguments
3719 @cindex plus, permitted arguments
3720 @cindex arguments for addition
3722 @dfn{Addition}. If either argument is absolute, the result has the section of
3723 the other argument. You may not add together arguments from different
3726 @cindex subtraction, permitted arguments
3727 @cindex minus, permitted arguments
3728 @cindex arguments for subtraction
3730 @dfn{Subtraction}. If the right argument is absolute, the
3731 result has the section of the left argument.
3732 If both arguments are in the same section, the result is absolute.
3733 You may not subtract arguments from different sections.
3734 @c FIXME is there still something useful to say about undefined - undefined ?
3736 @cindex comparison expressions
3737 @cindex expressions, comparison
3742 @dfn{Is Not Equal To}
3746 @dfn{Is Greater Than}
3748 @dfn{Is Greater Than Or Equal To}
3750 @dfn{Is Less Than Or Equal To}
3752 The comparison operators can be used as infix operators. A true results has a
3753 value of -1 whereas a false result has a value of 0. Note, these operators
3754 perform signed comparisons.
3757 @item Lowest Precedence
3766 These two logical operations can be used to combine the results of sub
3767 expressions. Note, unlike the comparison operators a true result returns a
3768 value of 1 but a false results does still return 0. Also note that the logical
3769 or operator has a slightly lower precedence than logical and.
3774 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3775 address; you can only have a defined section in one of the two arguments.
3778 @chapter Assembler Directives
3780 @cindex directives, machine independent
3781 @cindex pseudo-ops, machine independent
3782 @cindex machine independent directives
3783 All assembler directives have names that begin with a period (@samp{.}).
3784 The rest of the name is letters, usually in lower case.
3786 This chapter discusses directives that are available regardless of the
3787 target machine configuration for the @sc{gnu} assembler.
3789 Some machine configurations provide additional directives.
3790 @xref{Machine Dependencies}.
3793 @ifset machine-directives
3794 @xref{Machine Dependencies}, for additional directives.
3799 * Abort:: @code{.abort}
3801 * ABORT (COFF):: @code{.ABORT}
3804 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3805 * Altmacro:: @code{.altmacro}
3806 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3807 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3808 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3809 * Byte:: @code{.byte @var{expressions}}
3810 * Comm:: @code{.comm @var{symbol} , @var{length} }
3812 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
3814 * Data:: @code{.data @var{subsection}}
3816 * Def:: @code{.def @var{name}}
3819 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3825 * Double:: @code{.double @var{flonums}}
3826 * Eject:: @code{.eject}
3827 * Else:: @code{.else}
3828 * Elseif:: @code{.elseif}
3831 * Endef:: @code{.endef}
3834 * Endfunc:: @code{.endfunc}
3835 * Endif:: @code{.endif}
3836 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3837 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3838 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
3840 * Error:: @code{.error @var{string}}
3841 * Exitm:: @code{.exitm}
3842 * Extern:: @code{.extern}
3843 * Fail:: @code{.fail}
3844 @ifclear no-file-dir
3845 * File:: @code{.file @var{string}}
3848 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3849 * Float:: @code{.float @var{flonums}}
3850 * Func:: @code{.func}
3851 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3853 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
3854 * Hidden:: @code{.hidden @var{names}}
3857 * hword:: @code{.hword @var{expressions}}
3858 * Ident:: @code{.ident}
3859 * If:: @code{.if @var{absolute expression}}
3860 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3861 * Include:: @code{.include "@var{file}"}
3862 * Int:: @code{.int @var{expressions}}
3864 * Internal:: @code{.internal @var{names}}
3867 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3868 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3869 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3870 * Lflags:: @code{.lflags}
3871 @ifclear no-line-dir
3872 * Line:: @code{.line @var{line-number}}
3875 * Linkonce:: @code{.linkonce [@var{type}]}
3876 * List:: @code{.list}
3877 * Ln:: @code{.ln @var{line-number}}
3879 * LNS directives:: @code{.file}, @code{.loc}, etc.
3881 * Long:: @code{.long @var{expressions}}
3883 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3886 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3887 * MRI:: @code{.mri @var{val}}
3888 * Noaltmacro:: @code{.noaltmacro}
3889 * Nolist:: @code{.nolist}
3890 * Octa:: @code{.octa @var{bignums}}
3891 * Org:: @code{.org @var{new-lc}, @var{fill}}
3892 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3894 * PopSection:: @code{.popsection}
3895 * Previous:: @code{.previous}
3898 * Print:: @code{.print @var{string}}
3900 * Protected:: @code{.protected @var{names}}
3903 * Psize:: @code{.psize @var{lines}, @var{columns}}
3904 * Purgem:: @code{.purgem @var{name}}
3906 * PushSection:: @code{.pushsection @var{name}}
3909 * Quad:: @code{.quad @var{bignums}}
3910 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
3911 * Rept:: @code{.rept @var{count}}
3912 * Sbttl:: @code{.sbttl "@var{subheading}"}
3914 * Scl:: @code{.scl @var{class}}
3917 * Section:: @code{.section @var{name}[, @var{flags}]}
3920 * Set:: @code{.set @var{symbol}, @var{expression}}
3921 * Short:: @code{.short @var{expressions}}
3922 * Single:: @code{.single @var{flonums}}
3924 * Size:: @code{.size [@var{name} , @var{expression}]}
3927 * Skip:: @code{.skip @var{size} , @var{fill}}
3928 * Sleb128:: @code{.sleb128 @var{expressions}}
3929 * Space:: @code{.space @var{size} , @var{fill}}
3931 * Stab:: @code{.stabd, .stabn, .stabs}
3934 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
3935 * Struct:: @code{.struct @var{expression}}
3937 * SubSection:: @code{.subsection}
3938 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3942 * Tag:: @code{.tag @var{structname}}
3945 * Text:: @code{.text @var{subsection}}
3946 * Title:: @code{.title "@var{heading}"}
3948 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3951 * Uleb128:: @code{.uleb128 @var{expressions}}
3953 * Val:: @code{.val @var{addr}}
3957 * Version:: @code{.version "@var{string}"}
3958 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3959 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3962 * Warning:: @code{.warning @var{string}}
3963 * Weak:: @code{.weak @var{names}}
3964 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
3965 * Word:: @code{.word @var{expressions}}
3966 * Deprecated:: Deprecated Directives
3970 @section @code{.abort}
3972 @cindex @code{abort} directive
3973 @cindex stopping the assembly
3974 This directive stops the assembly immediately. It is for
3975 compatibility with other assemblers. The original idea was that the
3976 assembly language source would be piped into the assembler. If the sender
3977 of the source quit, it could use this directive tells @command{@value{AS}} to
3978 quit also. One day @code{.abort} will not be supported.
3982 @section @code{.ABORT} (COFF)
3984 @cindex @code{ABORT} directive
3985 When producing COFF output, @command{@value{AS}} accepts this directive as a
3986 synonym for @samp{.abort}.
3989 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
3995 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3997 @cindex padding the location counter
3998 @cindex @code{align} directive
3999 Pad the location counter (in the current subsection) to a particular storage
4000 boundary. The first expression (which must be absolute) is the alignment
4001 required, as described below.
4003 The second expression (also absolute) gives the fill value to be stored in the
4004 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4005 padding bytes are normally zero. However, on some systems, if the section is
4006 marked as containing code and the fill value is omitted, the space is filled
4007 with no-op instructions.
4009 The third expression is also absolute, and is also optional. If it is present,
4010 it is the maximum number of bytes that should be skipped by this alignment
4011 directive. If doing the alignment would require skipping more bytes than the
4012 specified maximum, then the alignment is not done at all. You can omit the
4013 fill value (the second argument) entirely by simply using two commas after the
4014 required alignment; this can be useful if you want the alignment to be filled
4015 with no-op instructions when appropriate.
4017 The way the required alignment is specified varies from system to system.
4018 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4019 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4020 alignment request in bytes. For example @samp{.align 8} advances
4021 the location counter until it is a multiple of 8. If the location counter
4022 is already a multiple of 8, no change is needed. For the tic54x, the
4023 first expression is the alignment request in words.
4025 For other systems, including ppc, i386 using a.out format, arm and
4026 strongarm, it is the
4027 number of low-order zero bits the location counter must have after
4028 advancement. For example @samp{.align 3} advances the location
4029 counter until it a multiple of 8. If the location counter is already a
4030 multiple of 8, no change is needed.
4032 This inconsistency is due to the different behaviors of the various
4033 native assemblers for these systems which GAS must emulate.
4034 GAS also provides @code{.balign} and @code{.p2align} directives,
4035 described later, which have a consistent behavior across all
4036 architectures (but are specific to GAS).
4039 @section @code{.ascii "@var{string}"}@dots{}
4041 @cindex @code{ascii} directive
4042 @cindex string literals
4043 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4044 separated by commas. It assembles each string (with no automatic
4045 trailing zero byte) into consecutive addresses.
4048 @section @code{.asciz "@var{string}"}@dots{}
4050 @cindex @code{asciz} directive
4051 @cindex zero-terminated strings
4052 @cindex null-terminated strings
4053 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4054 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4057 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4059 @cindex padding the location counter given number of bytes
4060 @cindex @code{balign} directive
4061 Pad the location counter (in the current subsection) to a particular
4062 storage boundary. The first expression (which must be absolute) is the
4063 alignment request in bytes. For example @samp{.balign 8} advances
4064 the location counter until it is a multiple of 8. If the location counter
4065 is already a multiple of 8, no change is needed.
4067 The second expression (also absolute) gives the fill value to be stored in the
4068 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4069 padding bytes are normally zero. However, on some systems, if the section is
4070 marked as containing code and the fill value is omitted, the space is filled
4071 with no-op instructions.
4073 The third expression is also absolute, and is also optional. If it is present,
4074 it is the maximum number of bytes that should be skipped by this alignment
4075 directive. If doing the alignment would require skipping more bytes than the
4076 specified maximum, then the alignment is not done at all. You can omit the
4077 fill value (the second argument) entirely by simply using two commas after the
4078 required alignment; this can be useful if you want the alignment to be filled
4079 with no-op instructions when appropriate.
4081 @cindex @code{balignw} directive
4082 @cindex @code{balignl} directive
4083 The @code{.balignw} and @code{.balignl} directives are variants of the
4084 @code{.balign} directive. The @code{.balignw} directive treats the fill
4085 pattern as a two byte word value. The @code{.balignl} directives treats the
4086 fill pattern as a four byte longword value. For example, @code{.balignw
4087 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4088 filled in with the value 0x368d (the exact placement of the bytes depends upon
4089 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4093 @section @code{.byte @var{expressions}}
4095 @cindex @code{byte} directive
4096 @cindex integers, one byte
4097 @code{.byte} expects zero or more expressions, separated by commas.
4098 Each expression is assembled into the next byte.
4101 @section @code{.comm @var{symbol} , @var{length} }
4103 @cindex @code{comm} directive
4104 @cindex symbol, common
4105 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4106 common symbol in one object file may be merged with a defined or common symbol
4107 of the same name in another object file. If @code{@value{LD}} does not see a
4108 definition for the symbol--just one or more common symbols--then it will
4109 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4110 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4111 the same name, and they do not all have the same size, it will allocate space
4112 using the largest size.
4115 When using ELF, the @code{.comm} directive takes an optional third argument.
4116 This is the desired alignment of the symbol, specified as a byte boundary (for
4117 example, an alignment of 16 means that the least significant 4 bits of the
4118 address should be zero). The alignment must be an absolute expression, and it
4119 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
4120 for the common symbol, it will use the alignment when placing the symbol. If
4121 no alignment is specified, @command{@value{AS}} will set the alignment to the
4122 largest power of two less than or equal to the size of the symbol, up to a
4127 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4128 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4131 @node CFI directives
4132 @section @code{.cfi_startproc [simple]}
4133 @cindex @code{cfi_startproc} directive
4134 @code{.cfi_startproc} is used at the beginning of each function that
4135 should have an entry in @code{.eh_frame}. It initializes some internal
4136 data structures. Don't forget to close the function by
4137 @code{.cfi_endproc}.
4139 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4140 it also emits some architecture dependent initial CFI instructions.
4142 @section @code{.cfi_endproc}
4143 @cindex @code{cfi_endproc} directive
4144 @code{.cfi_endproc} is used at the end of a function where it closes its
4145 unwind entry previously opened by
4146 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4148 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4149 @code{.cfi_personality} defines personality routine and its encoding.
4150 @var{encoding} must be a constant determining how the personality
4151 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4152 argument is not present, otherwise second argument should be
4153 a constant or a symbol name. When using indirect encodings,
4154 the symbol provided should be the location where personality
4155 can be loaded from, not the personality routine itself.
4156 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4157 no personality routine.
4159 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4160 @code{.cfi_lsda} defines LSDA and its encoding.
4161 @var{encoding} must be a constant determining how the LSDA
4162 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4163 argument is not present, otherwise second argument should be a constant
4164 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4167 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4168 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4169 address from @var{register} and add @var{offset} to it}.
4171 @section @code{.cfi_def_cfa_register @var{register}}
4172 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4173 now on @var{register} will be used instead of the old one. Offset
4176 @section @code{.cfi_def_cfa_offset @var{offset}}
4177 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4178 remains the same, but @var{offset} is new. Note that it is the
4179 absolute offset that will be added to a defined register to compute
4182 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4183 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4184 value that is added/substracted from the previous offset.
4186 @section @code{.cfi_offset @var{register}, @var{offset}}
4187 Previous value of @var{register} is saved at offset @var{offset} from
4190 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4191 Previous value of @var{register} is saved at offset @var{offset} from
4192 the current CFA register. This is transformed to @code{.cfi_offset}
4193 using the known displacement of the CFA register from the CFA.
4194 This is often easier to use, because the number will match the
4195 code it's annotating.
4197 @section @code{.cfi_register @var{register1}, @var{register2}}
4198 Previous value of @var{register1} is saved in register @var{register2}.
4200 @section @code{.cfi_restore @var{register}}
4201 @code{.cfi_restore} says that the rule for @var{register} is now the
4202 same as it was at the beginning of the function, after all initial
4203 instruction added by @code{.cfi_startproc} were executed.
4205 @section @code{.cfi_undefined @var{register}}
4206 From now on the previous value of @var{register} can't be restored anymore.
4208 @section @code{.cfi_same_value @var{register}}
4209 Current value of @var{register} is the same like in the previous frame,
4210 i.e. no restoration needed.
4212 @section @code{.cfi_remember_state},
4213 First save all current rules for all registers by @code{.cfi_remember_state},
4214 then totally screw them up by subsequent @code{.cfi_*} directives and when
4215 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4216 the previous saved state.
4218 @section @code{.cfi_return_column @var{register}}
4219 Change return column @var{register}, i.e. the return address is either
4220 directly in @var{register} or can be accessed by rules for @var{register}.
4222 @section @code{.cfi_signal_frame}
4223 Mark current function as signal trampoline.
4225 @section @code{.cfi_window_save}
4226 SPARC register window has been saved.
4228 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4229 Allows the user to add arbitrary bytes to the unwind info. One
4230 might use this to add OS-specific CFI opcodes, or generic CFI
4231 opcodes that GAS does not yet support.
4233 @node LNS directives
4234 @section @code{.file @var{fileno} @var{filename}}
4235 @cindex @code{file} directive
4236 When emitting dwarf2 line number information @code{.file} assigns filenames
4237 to the @code{.debug_line} file name table. The @var{fileno} operand should
4238 be a unique positive integer to use as the index of the entry in the table.
4239 The @var{filename} operand is a C string literal.
4241 The detail of filename indices is exposed to the user because the filename
4242 table is shared with the @code{.debug_info} section of the dwarf2 debugging
4243 information, and thus the user must know the exact indices that table
4246 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
4247 @cindex @code{loc} directive
4248 The @code{.loc} directive will add row to the @code{.debug_line} line
4249 number matrix corresponding to the immediately following assembly
4250 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
4251 arguments will be applied to the @code{.debug_line} state machine before
4254 The @var{options} are a sequence of the following tokens in any order:
4258 This option will set the @code{basic_block} register in the
4259 @code{.debug_line} state machine to @code{true}.
4262 This option will set the @code{prologue_end} register in the
4263 @code{.debug_line} state machine to @code{true}.
4265 @item epilogue_begin
4266 This option will set the @code{epilogue_begin} register in the
4267 @code{.debug_line} state machine to @code{true}.
4269 @item is_stmt @var{value}
4270 This option will set the @code{is_stmt} register in the
4271 @code{.debug_line} state machine to @code{value}, which must be
4274 @item isa @var{value}
4275 This directive will set the @code{isa} register in the @code{.debug_line}
4276 state machine to @var{value}, which must be an unsigned integer.
4280 @section @code{.loc_mark_labels @var{enable}}
4281 @cindex @code{loc_mark_labels} directive
4282 The @code{.loc_mark_labels} directive makes the assembler emit an entry
4283 to the @code{.debug_line} line number matrix with the @code{basic_block}
4284 register in the state machine set whenever a code label is seen.
4285 The @var{enable} argument should be either 1 or 0, to enable or disable
4286 this function respectively.
4289 @section @code{.data @var{subsection}}
4291 @cindex @code{data} directive
4292 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4293 end of the data subsection numbered @var{subsection} (which is an
4294 absolute expression). If @var{subsection} is omitted, it defaults
4299 @section @code{.def @var{name}}
4301 @cindex @code{def} directive
4302 @cindex COFF symbols, debugging
4303 @cindex debugging COFF symbols
4304 Begin defining debugging information for a symbol @var{name}; the
4305 definition extends until the @code{.endef} directive is encountered.
4308 This directive is only observed when @command{@value{AS}} is configured for COFF
4309 format output; when producing @code{b.out}, @samp{.def} is recognized,
4316 @section @code{.desc @var{symbol}, @var{abs-expression}}
4318 @cindex @code{desc} directive
4319 @cindex COFF symbol descriptor
4320 @cindex symbol descriptor, COFF
4321 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4322 to the low 16 bits of an absolute expression.
4325 The @samp{.desc} directive is not available when @command{@value{AS}} is
4326 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4327 object format. For the sake of compatibility, @command{@value{AS}} accepts
4328 it, but produces no output, when configured for COFF.
4334 @section @code{.dim}
4336 @cindex @code{dim} directive
4337 @cindex COFF auxiliary symbol information
4338 @cindex auxiliary symbol information, COFF
4339 This directive is generated by compilers to include auxiliary debugging
4340 information in the symbol table. It is only permitted inside
4341 @code{.def}/@code{.endef} pairs.
4344 @samp{.dim} is only meaningful when generating COFF format output; when
4345 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4351 @section @code{.double @var{flonums}}
4353 @cindex @code{double} directive
4354 @cindex floating point numbers (double)
4355 @code{.double} expects zero or more flonums, separated by commas. It
4356 assembles floating point numbers.
4358 The exact kind of floating point numbers emitted depends on how
4359 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4363 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4364 in @sc{ieee} format.
4369 @section @code{.eject}
4371 @cindex @code{eject} directive
4372 @cindex new page, in listings
4373 @cindex page, in listings
4374 @cindex listing control: new page
4375 Force a page break at this point, when generating assembly listings.
4378 @section @code{.else}
4380 @cindex @code{else} directive
4381 @code{.else} is part of the @command{@value{AS}} support for conditional
4382 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4383 of code to be assembled if the condition for the preceding @code{.if}
4387 @section @code{.elseif}
4389 @cindex @code{elseif} directive
4390 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4391 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4392 @code{.if} block that would otherwise fill the entire @code{.else} section.
4395 @section @code{.end}
4397 @cindex @code{end} directive
4398 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4399 process anything in the file past the @code{.end} directive.
4403 @section @code{.endef}
4405 @cindex @code{endef} directive
4406 This directive flags the end of a symbol definition begun with
4410 @samp{.endef} is only meaningful when generating COFF format output; if
4411 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4412 directive but ignores it.
4417 @section @code{.endfunc}
4418 @cindex @code{endfunc} directive
4419 @code{.endfunc} marks the end of a function specified with @code{.func}.
4422 @section @code{.endif}
4424 @cindex @code{endif} directive
4425 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4426 it marks the end of a block of code that is only assembled
4427 conditionally. @xref{If,,@code{.if}}.
4430 @section @code{.equ @var{symbol}, @var{expression}}
4432 @cindex @code{equ} directive
4433 @cindex assigning values to symbols
4434 @cindex symbols, assigning values to
4435 This directive sets the value of @var{symbol} to @var{expression}.
4436 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4439 The syntax for @code{equ} on the HPPA is
4440 @samp{@var{symbol} .equ @var{expression}}.
4444 The syntax for @code{equ} on the Z80 is
4445 @samp{@var{symbol} equ @var{expression}}.
4446 On the Z80 it is an eror if @var{symbol} is already defined,
4447 but the symbol is not protected from later redefinition.
4448 Compare @ref{Equiv}.
4452 @section @code{.equiv @var{symbol}, @var{expression}}
4453 @cindex @code{equiv} directive
4454 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4455 the assembler will signal an error if @var{symbol} is already defined. Note a
4456 symbol which has been referenced but not actually defined is considered to be
4459 Except for the contents of the error message, this is roughly equivalent to
4466 plus it protects the symbol from later redefinition.
4469 @section @code{.eqv @var{symbol}, @var{expression}}
4470 @cindex @code{eqv} directive
4471 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4472 evaluate the expression or any part of it immediately. Instead each time
4473 the resulting symbol is used in an expression, a snapshot of its current
4477 @section @code{.err}
4478 @cindex @code{err} directive
4479 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4480 message and, unless the @option{-Z} option was used, it will not generate an
4481 object file. This can be used to signal an error in conditionally compiled code.
4484 @section @code{.error "@var{string}"}
4485 @cindex error directive
4487 Similarly to @code{.err}, this directive emits an error, but you can specify a
4488 string that will be emitted as the error message. If you don't specify the
4489 message, it defaults to @code{".error directive invoked in source file"}.
4490 @xref{Errors, ,Error and Warning Messages}.
4493 .error "This code has not been assembled and tested."
4497 @section @code{.exitm}
4498 Exit early from the current macro definition. @xref{Macro}.
4501 @section @code{.extern}
4503 @cindex @code{extern} directive
4504 @code{.extern} is accepted in the source program---for compatibility
4505 with other assemblers---but it is ignored. @command{@value{AS}} treats
4506 all undefined symbols as external.
4509 @section @code{.fail @var{expression}}
4511 @cindex @code{fail} directive
4512 Generates an error or a warning. If the value of the @var{expression} is 500
4513 or more, @command{@value{AS}} will print a warning message. If the value is less
4514 than 500, @command{@value{AS}} will print an error message. The message will
4515 include the value of @var{expression}. This can occasionally be useful inside
4516 complex nested macros or conditional assembly.
4518 @ifclear no-file-dir
4520 @section @code{.file @var{string}}
4522 @cindex @code{file} directive
4523 @cindex logical file name
4524 @cindex file name, logical
4525 @code{.file} tells @command{@value{AS}} that we are about to start a new logical
4526 file. @var{string} is the new file name. In general, the filename is
4527 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4528 to specify an empty file name, you must give the quotes--@code{""}. This
4529 statement may go away in future: it is only recognized to be compatible with
4530 old @command{@value{AS}} programs.
4534 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4536 @cindex @code{fill} directive
4537 @cindex writing patterns in memory
4538 @cindex patterns, writing in memory
4539 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4540 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4541 may be zero or more. @var{Size} may be zero or more, but if it is
4542 more than 8, then it is deemed to have the value 8, compatible with
4543 other people's assemblers. The contents of each @var{repeat} bytes
4544 is taken from an 8-byte number. The highest order 4 bytes are
4545 zero. The lowest order 4 bytes are @var{value} rendered in the
4546 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4547 Each @var{size} bytes in a repetition is taken from the lowest order
4548 @var{size} bytes of this number. Again, this bizarre behavior is
4549 compatible with other people's assemblers.
4551 @var{size} and @var{value} are optional.
4552 If the second comma and @var{value} are absent, @var{value} is
4553 assumed zero. If the first comma and following tokens are absent,
4554 @var{size} is assumed to be 1.
4557 @section @code{.float @var{flonums}}
4559 @cindex floating point numbers (single)
4560 @cindex @code{float} directive
4561 This directive assembles zero or more flonums, separated by commas. It
4562 has the same effect as @code{.single}.
4564 The exact kind of floating point numbers emitted depends on how
4565 @command{@value{AS}} is configured.
4566 @xref{Machine Dependencies}.
4570 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4571 in @sc{ieee} format.
4576 @section @code{.func @var{name}[,@var{label}]}
4577 @cindex @code{func} directive
4578 @code{.func} emits debugging information to denote function @var{name}, and
4579 is ignored unless the file is assembled with debugging enabled.
4580 Only @samp{--gstabs[+]} is currently supported.
4581 @var{label} is the entry point of the function and if omitted @var{name}
4582 prepended with the @samp{leading char} is used.
4583 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4584 All functions are currently defined to have @code{void} return type.
4585 The function must be terminated with @code{.endfunc}.
4588 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4590 @cindex @code{global} directive
4591 @cindex symbol, making visible to linker
4592 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4593 @var{symbol} in your partial program, its value is made available to
4594 other partial programs that are linked with it. Otherwise,
4595 @var{symbol} takes its attributes from a symbol of the same name
4596 from another file linked into the same program.
4598 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4599 compatibility with other assemblers.
4602 On the HPPA, @code{.global} is not always enough to make it accessible to other
4603 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4604 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4609 @section @code{.gnu_attribute @var{tag},@var{value}}
4610 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
4613 @section @code{.hidden @var{names}}
4615 @cindex @code{hidden} directive
4617 This is one of the ELF visibility directives. The other two are
4618 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4619 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4621 This directive overrides the named symbols default visibility (which is set by
4622 their binding: local, global or weak). The directive sets the visibility to
4623 @code{hidden} which means that the symbols are not visible to other components.
4624 Such symbols are always considered to be @code{protected} as well.
4628 @section @code{.hword @var{expressions}}
4630 @cindex @code{hword} directive
4631 @cindex integers, 16-bit
4632 @cindex numbers, 16-bit
4633 @cindex sixteen bit integers
4634 This expects zero or more @var{expressions}, and emits
4635 a 16 bit number for each.
4638 This directive is a synonym for @samp{.short}; depending on the target
4639 architecture, it may also be a synonym for @samp{.word}.
4643 This directive is a synonym for @samp{.short}.
4646 This directive is a synonym for both @samp{.short} and @samp{.word}.
4651 @section @code{.ident}
4653 @cindex @code{ident} directive
4655 This directive is used by some assemblers to place tags in object files. The
4656 behavior of this directive varies depending on the target. When using the
4657 a.out object file format, @command{@value{AS}} simply accepts the directive for
4658 source-file compatibility with existing assemblers, but does not emit anything
4659 for it. When using COFF, comments are emitted to the @code{.comment} or
4660 @code{.rdata} section, depending on the target. When using ELF, comments are
4661 emitted to the @code{.comment} section.
4664 @section @code{.if @var{absolute expression}}
4666 @cindex conditional assembly
4667 @cindex @code{if} directive
4668 @code{.if} marks the beginning of a section of code which is only
4669 considered part of the source program being assembled if the argument
4670 (which must be an @var{absolute expression}) is non-zero. The end of
4671 the conditional section of code must be marked by @code{.endif}
4672 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4673 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4674 If you have several conditions to check, @code{.elseif} may be used to avoid
4675 nesting blocks if/else within each subsequent @code{.else} block.
4677 The following variants of @code{.if} are also supported:
4679 @cindex @code{ifdef} directive
4680 @item .ifdef @var{symbol}
4681 Assembles the following section of code if the specified @var{symbol}
4682 has been defined. Note a symbol which has been referenced but not yet defined
4683 is considered to be undefined.
4685 @cindex @code{ifb} directive
4686 @item .ifb @var{text}
4687 Assembles the following section of code if the operand is blank (empty).
4689 @cindex @code{ifc} directive
4690 @item .ifc @var{string1},@var{string2}
4691 Assembles the following section of code if the two strings are the same. The
4692 strings may be optionally quoted with single quotes. If they are not quoted,
4693 the first string stops at the first comma, and the second string stops at the
4694 end of the line. Strings which contain whitespace should be quoted. The
4695 string comparison is case sensitive.
4697 @cindex @code{ifeq} directive
4698 @item .ifeq @var{absolute expression}
4699 Assembles the following section of code if the argument is zero.
4701 @cindex @code{ifeqs} directive
4702 @item .ifeqs @var{string1},@var{string2}
4703 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4705 @cindex @code{ifge} directive
4706 @item .ifge @var{absolute expression}
4707 Assembles the following section of code if the argument is greater than or
4710 @cindex @code{ifgt} directive
4711 @item .ifgt @var{absolute expression}
4712 Assembles the following section of code if the argument is greater than zero.
4714 @cindex @code{ifle} directive
4715 @item .ifle @var{absolute expression}
4716 Assembles the following section of code if the argument is less than or equal
4719 @cindex @code{iflt} directive
4720 @item .iflt @var{absolute expression}
4721 Assembles the following section of code if the argument is less than zero.
4723 @cindex @code{ifnb} directive
4724 @item .ifnb @var{text}
4725 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
4726 following section of code if the operand is non-blank (non-empty).
4728 @cindex @code{ifnc} directive
4729 @item .ifnc @var{string1},@var{string2}.
4730 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4731 following section of code if the two strings are not the same.
4733 @cindex @code{ifndef} directive
4734 @cindex @code{ifnotdef} directive
4735 @item .ifndef @var{symbol}
4736 @itemx .ifnotdef @var{symbol}
4737 Assembles the following section of code if the specified @var{symbol}
4738 has not been defined. Both spelling variants are equivalent. Note a symbol
4739 which has been referenced but not yet defined is considered to be undefined.
4741 @cindex @code{ifne} directive
4742 @item .ifne @var{absolute expression}
4743 Assembles the following section of code if the argument is not equal to zero
4744 (in other words, this is equivalent to @code{.if}).
4746 @cindex @code{ifnes} directive
4747 @item .ifnes @var{string1},@var{string2}
4748 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4749 following section of code if the two strings are not the same.
4753 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4755 @cindex @code{incbin} directive
4756 @cindex binary files, including
4757 The @code{incbin} directive includes @var{file} verbatim at the current
4758 location. You can control the search paths used with the @samp{-I} command-line
4759 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4762 The @var{skip} argument skips a number of bytes from the start of the
4763 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4764 read. Note that the data is not aligned in any way, so it is the user's
4765 responsibility to make sure that proper alignment is provided both before and
4766 after the @code{incbin} directive.
4769 @section @code{.include "@var{file}"}
4771 @cindex @code{include} directive
4772 @cindex supporting files, including
4773 @cindex files, including
4774 This directive provides a way to include supporting files at specified
4775 points in your source program. The code from @var{file} is assembled as
4776 if it followed the point of the @code{.include}; when the end of the
4777 included file is reached, assembly of the original file continues. You
4778 can control the search paths used with the @samp{-I} command-line option
4779 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4783 @section @code{.int @var{expressions}}
4785 @cindex @code{int} directive
4786 @cindex integers, 32-bit
4787 Expect zero or more @var{expressions}, of any section, separated by commas.
4788 For each expression, emit a number that, at run time, is the value of that
4789 expression. The byte order and bit size of the number depends on what kind
4790 of target the assembly is for.
4794 On most forms of the H8/300, @code{.int} emits 16-bit
4795 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4802 @section @code{.internal @var{names}}
4804 @cindex @code{internal} directive
4806 This is one of the ELF visibility directives. The other two are
4807 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4808 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4810 This directive overrides the named symbols default visibility (which is set by
4811 their binding: local, global or weak). The directive sets the visibility to
4812 @code{internal} which means that the symbols are considered to be @code{hidden}
4813 (i.e., not visible to other components), and that some extra, processor specific
4814 processing must also be performed upon the symbols as well.
4818 @section @code{.irp @var{symbol},@var{values}}@dots{}
4820 @cindex @code{irp} directive
4821 Evaluate a sequence of statements assigning different values to @var{symbol}.
4822 The sequence of statements starts at the @code{.irp} directive, and is
4823 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4824 set to @var{value}, and the sequence of statements is assembled. If no
4825 @var{value} is listed, the sequence of statements is assembled once, with
4826 @var{symbol} set to the null string. To refer to @var{symbol} within the
4827 sequence of statements, use @var{\symbol}.
4829 For example, assembling
4837 is equivalent to assembling
4845 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
4848 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4850 @cindex @code{irpc} directive
4851 Evaluate a sequence of statements assigning different values to @var{symbol}.
4852 The sequence of statements starts at the @code{.irpc} directive, and is
4853 terminated by an @code{.endr} directive. For each character in @var{value},
4854 @var{symbol} is set to the character, and the sequence of statements is
4855 assembled. If no @var{value} is listed, the sequence of statements is
4856 assembled once, with @var{symbol} set to the null string. To refer to
4857 @var{symbol} within the sequence of statements, use @var{\symbol}.
4859 For example, assembling
4867 is equivalent to assembling
4875 For some caveats with the spelling of @var{symbol}, see also the discussion
4879 @section @code{.lcomm @var{symbol} , @var{length}}
4881 @cindex @code{lcomm} directive
4882 @cindex local common symbols
4883 @cindex symbols, local common
4884 Reserve @var{length} (an absolute expression) bytes for a local common
4885 denoted by @var{symbol}. The section and value of @var{symbol} are
4886 those of the new local common. The addresses are allocated in the bss
4887 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4888 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4889 not visible to @code{@value{LD}}.
4892 Some targets permit a third argument to be used with @code{.lcomm}. This
4893 argument specifies the desired alignment of the symbol in the bss section.
4897 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4898 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4902 @section @code{.lflags}
4904 @cindex @code{lflags} directive (ignored)
4905 @command{@value{AS}} accepts this directive, for compatibility with other
4906 assemblers, but ignores it.
4908 @ifclear no-line-dir
4910 @section @code{.line @var{line-number}}
4912 @cindex @code{line} directive
4916 @section @code{.ln @var{line-number}}
4918 @cindex @code{ln} directive
4920 @cindex logical line number
4922 Change the logical line number. @var{line-number} must be an absolute
4923 expression. The next line has that logical line number. Therefore any other
4924 statements on the current line (after a statement separator character) are
4925 reported as on logical line number @var{line-number} @minus{} 1. One day
4926 @command{@value{AS}} will no longer support this directive: it is recognized only
4927 for compatibility with existing assembler programs.
4931 @ifclear no-line-dir
4932 Even though this is a directive associated with the @code{a.out} or
4933 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
4934 when producing COFF output, and treats @samp{.line} as though it
4935 were the COFF @samp{.ln} @emph{if} it is found outside a
4936 @code{.def}/@code{.endef} pair.
4938 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4939 used by compilers to generate auxiliary symbol information for
4944 @section @code{.linkonce [@var{type}]}
4946 @cindex @code{linkonce} directive
4947 @cindex common sections
4948 Mark the current section so that the linker only includes a single copy of it.
4949 This may be used to include the same section in several different object files,
4950 but ensure that the linker will only include it once in the final output file.
4951 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4952 Duplicate sections are detected based on the section name, so it should be
4955 This directive is only supported by a few object file formats; as of this
4956 writing, the only object file format which supports it is the Portable
4957 Executable format used on Windows NT.
4959 The @var{type} argument is optional. If specified, it must be one of the
4960 following strings. For example:
4964 Not all types may be supported on all object file formats.
4968 Silently discard duplicate sections. This is the default.
4971 Warn if there are duplicate sections, but still keep only one copy.
4974 Warn if any of the duplicates have different sizes.
4977 Warn if any of the duplicates do not have exactly the same contents.
4981 @section @code{.ln @var{line-number}}
4983 @cindex @code{ln} directive
4984 @ifclear no-line-dir
4985 @samp{.ln} is a synonym for @samp{.line}.
4988 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
4989 must be an absolute expression. The next line has that logical
4990 line number, so any other statements on the current line (after a
4991 statement separator character @code{;}) are reported as on logical
4992 line number @var{line-number} @minus{} 1.
4995 This directive is accepted, but ignored, when @command{@value{AS}} is
4996 configured for @code{b.out}; its effect is only associated with COFF
5002 @section @code{.mri @var{val}}
5004 @cindex @code{mri} directive
5005 @cindex MRI mode, temporarily
5006 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5007 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5008 affects code assembled until the next @code{.mri} directive, or until the end
5009 of the file. @xref{M, MRI mode, MRI mode}.
5012 @section @code{.list}
5014 @cindex @code{list} directive
5015 @cindex listing control, turning on
5016 Control (in conjunction with the @code{.nolist} directive) whether or
5017 not assembly listings are generated. These two directives maintain an
5018 internal counter (which is zero initially). @code{.list} increments the
5019 counter, and @code{.nolist} decrements it. Assembly listings are
5020 generated whenever the counter is greater than zero.
5022 By default, listings are disabled. When you enable them (with the
5023 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5024 the initial value of the listing counter is one.
5027 @section @code{.long @var{expressions}}
5029 @cindex @code{long} directive
5030 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5033 @c no one seems to know what this is for or whether this description is
5034 @c what it really ought to do
5036 @section @code{.lsym @var{symbol}, @var{expression}}
5038 @cindex @code{lsym} directive
5039 @cindex symbol, not referenced in assembly
5040 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5041 the hash table, ensuring it cannot be referenced by name during the
5042 rest of the assembly. This sets the attributes of the symbol to be
5043 the same as the expression value:
5045 @var{other} = @var{descriptor} = 0
5046 @var{type} = @r{(section of @var{expression})}
5047 @var{value} = @var{expression}
5050 The new symbol is not flagged as external.
5054 @section @code{.macro}
5057 The commands @code{.macro} and @code{.endm} allow you to define macros that
5058 generate assembly output. For example, this definition specifies a macro
5059 @code{sum} that puts a sequence of numbers into memory:
5062 .macro sum from=0, to=5
5071 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5083 @item .macro @var{macname}
5084 @itemx .macro @var{macname} @var{macargs} @dots{}
5085 @cindex @code{macro} directive
5086 Begin the definition of a macro called @var{macname}. If your macro
5087 definition requires arguments, specify their names after the macro name,
5088 separated by commas or spaces. You can qualify the macro argument to
5089 indicate whether all invocations must specify a non-blank value (through
5090 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5091 (through @samp{:@code{vararg}}). You can supply a default value for any
5092 macro argument by following the name with @samp{=@var{deflt}}. You
5093 cannot define two macros with the same @var{macname} unless it has been
5094 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5095 definitions. For example, these are all valid @code{.macro} statements:
5099 Begin the definition of a macro called @code{comm}, which takes no
5102 @item .macro plus1 p, p1
5103 @itemx .macro plus1 p p1
5104 Either statement begins the definition of a macro called @code{plus1},
5105 which takes two arguments; within the macro definition, write
5106 @samp{\p} or @samp{\p1} to evaluate the arguments.
5108 @item .macro reserve_str p1=0 p2
5109 Begin the definition of a macro called @code{reserve_str}, with two
5110 arguments. The first argument has a default value, but not the second.
5111 After the definition is complete, you can call the macro either as
5112 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5113 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5114 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5115 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5117 @item .macro m p1:req, p2=0, p3:vararg
5118 Begin the definition of a macro called @code{m}, with at least three
5119 arguments. The first argument must always have a value specified, but
5120 not the second, which instead has a default value. The third formal
5121 will get assigned all remaining arguments specified at invocation time.
5123 When you call a macro, you can specify the argument values either by
5124 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5125 @samp{sum to=17, from=9}.
5129 Note that since each of the @var{macargs} can be an identifier exactly
5130 as any other one permitted by the target architecture, there may be
5131 occasional problems if the target hand-crafts special meanings to certain
5132 characters when they occur in a special position. For example, if the colon
5133 (@code{:}) is generally permitted to be part of a symbol name, but the
5134 architecture specific code special-cases it when occurring as the final
5135 character of a symbol (to denote a label), then the macro parameter
5136 replacement code will have no way of knowing that and consider the whole
5137 construct (including the colon) an identifier, and check only this
5138 identifier for being the subject to parameter substitution. So for example
5139 this macro definition:
5147 might not work as expected. Invoking @samp{label foo} might not create a label
5148 called @samp{foo} but instead just insert the text @samp{\l:} into the
5149 assembler source, probably generating an error about an unrecognised
5152 Similarly problems might occur with the period character (@samp{.})
5153 which is often allowed inside opcode names (and hence identifier names). So
5154 for example constructing a macro to build an opcode from a base name and a
5155 length specifier like this:
5158 .macro opcode base length
5163 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5164 instruction but instead generate some kind of error as the assembler tries to
5165 interpret the text @samp{\base.\length}.
5167 There are several possible ways around this problem:
5170 @item Insert white space
5171 If it is possible to use white space characters then this is the simplest
5180 @item Use @samp{\()}
5181 The string @samp{\()} can be used to separate the end of a macro argument from
5182 the following text. eg:
5185 .macro opcode base length
5190 @item Use the alternate macro syntax mode
5191 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5192 used as a separator. eg:
5202 Note: this problem of correctly identifying string parameters to pseudo ops
5203 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5204 and @code{.irpc} (@pxref{Irpc}) as well.
5207 @cindex @code{endm} directive
5208 Mark the end of a macro definition.
5211 @cindex @code{exitm} directive
5212 Exit early from the current macro definition.
5214 @cindex number of macros executed
5215 @cindex macros, count executed
5217 @command{@value{AS}} maintains a counter of how many macros it has
5218 executed in this pseudo-variable; you can copy that number to your
5219 output with @samp{\@@}, but @emph{only within a macro definition}.
5221 @item LOCAL @var{name} [ , @dots{} ]
5222 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5223 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5224 @xref{Altmacro,,@code{.altmacro}}.
5228 @section @code{.altmacro}
5229 Enable alternate macro mode, enabling:
5232 @item LOCAL @var{name} [ , @dots{} ]
5233 One additional directive, @code{LOCAL}, is available. It is used to
5234 generate a string replacement for each of the @var{name} arguments, and
5235 replace any instances of @var{name} in each macro expansion. The
5236 replacement string is unique in the assembly, and different for each
5237 separate macro expansion. @code{LOCAL} allows you to write macros that
5238 define symbols, without fear of conflict between separate macro expansions.
5240 @item String delimiters
5241 You can write strings delimited in these other ways besides
5242 @code{"@var{string}"}:
5245 @item '@var{string}'
5246 You can delimit strings with single-quote characters.
5248 @item <@var{string}>
5249 You can delimit strings with matching angle brackets.
5252 @item single-character string escape
5253 To include any single character literally in a string (even if the
5254 character would otherwise have some special meaning), you can prefix the
5255 character with @samp{!} (an exclamation mark). For example, you can
5256 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
5258 @item Expression results as strings
5259 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
5260 and use the result as a string.
5264 @section @code{.noaltmacro}
5265 Disable alternate macro mode. @xref{Altmacro}.
5268 @section @code{.nolist}
5270 @cindex @code{nolist} directive
5271 @cindex listing control, turning off
5272 Control (in conjunction with the @code{.list} directive) whether or
5273 not assembly listings are generated. These two directives maintain an
5274 internal counter (which is zero initially). @code{.list} increments the
5275 counter, and @code{.nolist} decrements it. Assembly listings are
5276 generated whenever the counter is greater than zero.
5279 @section @code{.octa @var{bignums}}
5281 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5282 @cindex @code{octa} directive
5283 @cindex integer, 16-byte
5284 @cindex sixteen byte integer
5285 This directive expects zero or more bignums, separated by commas. For each
5286 bignum, it emits a 16-byte integer.
5288 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5289 hence @emph{octa}-word for 16 bytes.
5292 @section @code{.org @var{new-lc} , @var{fill}}
5294 @cindex @code{org} directive
5295 @cindex location counter, advancing
5296 @cindex advancing location counter
5297 @cindex current address, advancing
5298 Advance the location counter of the current section to
5299 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5300 expression with the same section as the current subsection. That is,
5301 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5302 wrong section, the @code{.org} directive is ignored. To be compatible
5303 with former assemblers, if the section of @var{new-lc} is absolute,
5304 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5305 is the same as the current subsection.
5307 @code{.org} may only increase the location counter, or leave it
5308 unchanged; you cannot use @code{.org} to move the location counter
5311 @c double negative used below "not undefined" because this is a specific
5312 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5313 @c section. doc@cygnus.com 18feb91
5314 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5315 may not be undefined. If you really detest this restriction we eagerly await
5316 a chance to share your improved assembler.
5318 Beware that the origin is relative to the start of the section, not
5319 to the start of the subsection. This is compatible with other
5320 people's assemblers.
5322 When the location counter (of the current subsection) is advanced, the
5323 intervening bytes are filled with @var{fill} which should be an
5324 absolute expression. If the comma and @var{fill} are omitted,
5325 @var{fill} defaults to zero.
5328 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5330 @cindex padding the location counter given a power of two
5331 @cindex @code{p2align} directive
5332 Pad the location counter (in the current subsection) to a particular
5333 storage boundary. The first expression (which must be absolute) is the
5334 number of low-order zero bits the location counter must have after
5335 advancement. For example @samp{.p2align 3} advances the location
5336 counter until it a multiple of 8. If the location counter is already a
5337 multiple of 8, no change is needed.
5339 The second expression (also absolute) gives the fill value to be stored in the
5340 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5341 padding bytes are normally zero. However, on some systems, if the section is
5342 marked as containing code and the fill value is omitted, the space is filled
5343 with no-op instructions.
5345 The third expression is also absolute, and is also optional. If it is present,
5346 it is the maximum number of bytes that should be skipped by this alignment
5347 directive. If doing the alignment would require skipping more bytes than the
5348 specified maximum, then the alignment is not done at all. You can omit the
5349 fill value (the second argument) entirely by simply using two commas after the
5350 required alignment; this can be useful if you want the alignment to be filled
5351 with no-op instructions when appropriate.
5353 @cindex @code{p2alignw} directive
5354 @cindex @code{p2alignl} directive
5355 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5356 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5357 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5358 fill pattern as a four byte longword value. For example, @code{.p2alignw
5359 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5360 filled in with the value 0x368d (the exact placement of the bytes depends upon
5361 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5366 @section @code{.previous}
5368 @cindex @code{previous} directive
5369 @cindex Section Stack
5370 This is one of the ELF section stack manipulation directives. The others are
5371 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5372 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5373 (@pxref{PopSection}).
5375 This directive swaps the current section (and subsection) with most recently
5376 referenced section/subsection pair prior to this one. Multiple
5377 @code{.previous} directives in a row will flip between two sections (and their
5378 subsections). For example:
5390 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5396 # Now in section A subsection 1
5400 # Now in section B subsection 0
5403 # Now in section B subsection 1
5406 # Now in section B subsection 0
5410 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5411 section B and 0x9abc into subsection 1 of section B.
5413 In terms of the section stack, this directive swaps the current section with
5414 the top section on the section stack.
5419 @section @code{.popsection}
5421 @cindex @code{popsection} directive
5422 @cindex Section Stack
5423 This is one of the ELF section stack manipulation directives. The others are
5424 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5425 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5428 This directive replaces the current section (and subsection) with the top
5429 section (and subsection) on the section stack. This section is popped off the
5434 @section @code{.print @var{string}}
5436 @cindex @code{print} directive
5437 @command{@value{AS}} will print @var{string} on the standard output during
5438 assembly. You must put @var{string} in double quotes.
5442 @section @code{.protected @var{names}}
5444 @cindex @code{protected} directive
5446 This is one of the ELF visibility directives. The other two are
5447 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5449 This directive overrides the named symbols default visibility (which is set by
5450 their binding: local, global or weak). The directive sets the visibility to
5451 @code{protected} which means that any references to the symbols from within the
5452 components that defines them must be resolved to the definition in that
5453 component, even if a definition in another component would normally preempt
5458 @section @code{.psize @var{lines} , @var{columns}}
5460 @cindex @code{psize} directive
5461 @cindex listing control: paper size
5462 @cindex paper size, for listings
5463 Use this directive to declare the number of lines---and, optionally, the
5464 number of columns---to use for each page, when generating listings.
5466 If you do not use @code{.psize}, listings use a default line-count
5467 of 60. You may omit the comma and @var{columns} specification; the
5468 default width is 200 columns.
5470 @command{@value{AS}} generates formfeeds whenever the specified number of
5471 lines is exceeded (or whenever you explicitly request one, using
5474 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5475 those explicitly specified with @code{.eject}.
5478 @section @code{.purgem @var{name}}
5480 @cindex @code{purgem} directive
5481 Undefine the macro @var{name}, so that later uses of the string will not be
5482 expanded. @xref{Macro}.
5486 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5488 @cindex @code{pushsection} directive
5489 @cindex Section Stack
5490 This is one of the ELF section stack manipulation directives. The others are
5491 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5492 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5495 This directive pushes the current section (and subsection) onto the
5496 top of the section stack, and then replaces the current section and
5497 subsection with @code{name} and @code{subsection}. The optional
5498 @code{flags}, @code{type} and @code{arguments} are treated the same
5499 as in the @code{.section} (@pxref{Section}) directive.
5503 @section @code{.quad @var{bignums}}
5505 @cindex @code{quad} directive
5506 @code{.quad} expects zero or more bignums, separated by commas. For
5507 each bignum, it emits
5509 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5510 warning message; and just takes the lowest order 8 bytes of the bignum.
5511 @cindex eight-byte integer
5512 @cindex integer, 8-byte
5514 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5515 hence @emph{quad}-word for 8 bytes.
5518 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5519 warning message; and just takes the lowest order 16 bytes of the bignum.
5520 @cindex sixteen-byte integer
5521 @cindex integer, 16-byte
5525 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5527 @cindex @code{reloc} directive
5528 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5529 @var{expression}. If @var{offset} is a number, the relocation is generated in
5530 the current section. If @var{offset} is an expression that resolves to a
5531 symbol plus offset, the relocation is generated in the given symbol's section.
5532 @var{expression}, if present, must resolve to a symbol plus addend or to an
5533 absolute value, but note that not all targets support an addend. e.g. ELF REL
5534 targets such as i386 store an addend in the section contents rather than in the
5535 relocation. This low level interface does not support addends stored in the
5539 @section @code{.rept @var{count}}
5541 @cindex @code{rept} directive
5542 Repeat the sequence of lines between the @code{.rept} directive and the next
5543 @code{.endr} directive @var{count} times.
5545 For example, assembling
5553 is equivalent to assembling
5562 @section @code{.sbttl "@var{subheading}"}
5564 @cindex @code{sbttl} directive
5565 @cindex subtitles for listings
5566 @cindex listing control: subtitle
5567 Use @var{subheading} as the title (third line, immediately after the
5568 title line) when generating assembly listings.
5570 This directive affects subsequent pages, as well as the current page if
5571 it appears within ten lines of the top of a page.
5575 @section @code{.scl @var{class}}
5577 @cindex @code{scl} directive
5578 @cindex symbol storage class (COFF)
5579 @cindex COFF symbol storage class
5580 Set the storage-class value for a symbol. This directive may only be
5581 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5582 whether a symbol is static or external, or it may record further
5583 symbolic debugging information.
5586 The @samp{.scl} directive is primarily associated with COFF output; when
5587 configured to generate @code{b.out} output format, @command{@value{AS}}
5588 accepts this directive but ignores it.
5594 @section @code{.section @var{name}}
5596 @cindex named section
5597 Use the @code{.section} directive to assemble the following code into a section
5600 This directive is only supported for targets that actually support arbitrarily
5601 named sections; on @code{a.out} targets, for example, it is not accepted, even
5602 with a standard @code{a.out} section name.
5606 @c only print the extra heading if both COFF and ELF are set
5607 @subheading COFF Version
5610 @cindex @code{section} directive (COFF version)
5611 For COFF targets, the @code{.section} directive is used in one of the following
5615 .section @var{name}[, "@var{flags}"]
5616 .section @var{name}[, @var{subsection}]
5619 If the optional argument is quoted, it is taken as flags to use for the
5620 section. Each flag is a single character. The following flags are recognized:
5623 bss section (uninitialized data)
5625 section is not loaded
5635 shared section (meaningful for PE targets)
5637 ignored. (For compatibility with the ELF version)
5640 If no flags are specified, the default flags depend upon the section name. If
5641 the section name is not recognized, the default will be for the section to be
5642 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5643 from the section, rather than adding them, so if they are used on their own it
5644 will be as if no flags had been specified at all.
5646 If the optional argument to the @code{.section} directive is not quoted, it is
5647 taken as a subsection number (@pxref{Sub-Sections}).
5652 @c only print the extra heading if both COFF and ELF are set
5653 @subheading ELF Version
5656 @cindex Section Stack
5657 This is one of the ELF section stack manipulation directives. The others are
5658 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5659 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5660 @code{.previous} (@pxref{Previous}).
5662 @cindex @code{section} directive (ELF version)
5663 For ELF targets, the @code{.section} directive is used like this:
5666 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
5669 The optional @var{flags} argument is a quoted string which may contain any
5670 combination of the following characters:
5673 section is allocatable
5677 section is executable
5679 section is mergeable
5681 section contains zero terminated strings
5683 section is a member of a section group
5685 section is used for thread-local-storage
5688 The optional @var{type} argument may contain one of the following constants:
5691 section contains data
5693 section does not contain data (i.e., section only occupies space)
5695 section contains data which is used by things other than the program
5697 section contains an array of pointers to init functions
5699 section contains an array of pointers to finish functions
5700 @item @@preinit_array
5701 section contains an array of pointers to pre-init functions
5704 Many targets only support the first three section types.
5706 Note on targets where the @code{@@} character is the start of a comment (eg
5707 ARM) then another character is used instead. For example the ARM port uses the
5710 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5711 be specified as well as an extra argument---@var{entsize}---like this:
5714 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5717 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5718 constants, each @var{entsize} octets long. Sections with both @code{M} and
5719 @code{S} must contain zero terminated strings where each character is
5720 @var{entsize} bytes long. The linker may remove duplicates within sections with
5721 the same name, same entity size and same flags. @var{entsize} must be an
5722 absolute expression.
5724 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5725 be present along with an additional field like this:
5728 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5731 The @var{GroupName} field specifies the name of the section group to which this
5732 particular section belongs. The optional linkage field can contain:
5735 indicates that only one copy of this section should be retained
5740 Note: if both the @var{M} and @var{G} flags are present then the fields for
5741 the Merge flag should come first, like this:
5744 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5747 If no flags are specified, the default flags depend upon the section name. If
5748 the section name is not recognized, the default will be for the section to have
5749 none of the above flags: it will not be allocated in memory, nor writable, nor
5750 executable. The section will contain data.
5752 For ELF targets, the assembler supports another type of @code{.section}
5753 directive for compatibility with the Solaris assembler:
5756 .section "@var{name}"[, @var{flags}...]
5759 Note that the section name is quoted. There may be a sequence of comma
5763 section is allocatable
5767 section is executable
5769 section is used for thread local storage
5772 This directive replaces the current section and subsection. See the
5773 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
5774 some examples of how this directive and the other section stack directives
5780 @section @code{.set @var{symbol}, @var{expression}}
5782 @cindex @code{set} directive
5783 @cindex symbol value, setting
5784 Set the value of @var{symbol} to @var{expression}. This
5785 changes @var{symbol}'s value and type to conform to
5786 @var{expression}. If @var{symbol} was flagged as external, it remains
5787 flagged (@pxref{Symbol Attributes}).
5789 You may @code{.set} a symbol many times in the same assembly.
5791 If you @code{.set} a global symbol, the value stored in the object
5792 file is the last value stored into it.
5795 The syntax for @code{set} on the HPPA is
5796 @samp{@var{symbol} .set @var{expression}}.
5800 On Z80 @code{set} is a real instruction, use
5801 @samp{@var{symbol} defl @var{expression}} instead.
5805 @section @code{.short @var{expressions}}
5807 @cindex @code{short} directive
5809 @code{.short} is normally the same as @samp{.word}.
5810 @xref{Word,,@code{.word}}.
5812 In some configurations, however, @code{.short} and @code{.word} generate
5813 numbers of different lengths. @xref{Machine Dependencies}.
5817 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5820 This expects zero or more @var{expressions}, and emits
5821 a 16 bit number for each.
5826 @section @code{.single @var{flonums}}
5828 @cindex @code{single} directive
5829 @cindex floating point numbers (single)
5830 This directive assembles zero or more flonums, separated by commas. It
5831 has the same effect as @code{.float}.
5833 The exact kind of floating point numbers emitted depends on how
5834 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5838 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5839 numbers in @sc{ieee} format.
5845 @section @code{.size}
5847 This directive is used to set the size associated with a symbol.
5851 @c only print the extra heading if both COFF and ELF are set
5852 @subheading COFF Version
5855 @cindex @code{size} directive (COFF version)
5856 For COFF targets, the @code{.size} directive is only permitted inside
5857 @code{.def}/@code{.endef} pairs. It is used like this:
5860 .size @var{expression}
5864 @samp{.size} is only meaningful when generating COFF format output; when
5865 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5872 @c only print the extra heading if both COFF and ELF are set
5873 @subheading ELF Version
5876 @cindex @code{size} directive (ELF version)
5877 For ELF targets, the @code{.size} directive is used like this:
5880 .size @var{name} , @var{expression}
5883 This directive sets the size associated with a symbol @var{name}.
5884 The size in bytes is computed from @var{expression} which can make use of label
5885 arithmetic. This directive is typically used to set the size of function
5891 @section @code{.sleb128 @var{expressions}}
5893 @cindex @code{sleb128} directive
5894 @var{sleb128} stands for ``signed little endian base 128.'' This is a
5895 compact, variable length representation of numbers used by the DWARF
5896 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
5898 @ifclear no-space-dir
5900 @section @code{.skip @var{size} , @var{fill}}
5902 @cindex @code{skip} directive
5903 @cindex filling memory
5904 This directive emits @var{size} bytes, each of value @var{fill}. Both
5905 @var{size} and @var{fill} are absolute expressions. If the comma and
5906 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
5910 @section @code{.space @var{size} , @var{fill}}
5912 @cindex @code{space} directive
5913 @cindex filling memory
5914 This directive emits @var{size} bytes, each of value @var{fill}. Both
5915 @var{size} and @var{fill} are absolute expressions. If the comma
5916 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
5921 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
5922 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
5923 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
5924 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
5932 @section @code{.stabd, .stabn, .stabs}
5934 @cindex symbolic debuggers, information for
5935 @cindex @code{stab@var{x}} directives
5936 There are three directives that begin @samp{.stab}.
5937 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
5938 The symbols are not entered in the @command{@value{AS}} hash table: they
5939 cannot be referenced elsewhere in the source file.
5940 Up to five fields are required:
5944 This is the symbol's name. It may contain any character except
5945 @samp{\000}, so is more general than ordinary symbol names. Some
5946 debuggers used to code arbitrarily complex structures into symbol names
5950 An absolute expression. The symbol's type is set to the low 8 bits of
5951 this expression. Any bit pattern is permitted, but @code{@value{LD}}
5952 and debuggers choke on silly bit patterns.
5955 An absolute expression. The symbol's ``other'' attribute is set to the
5956 low 8 bits of this expression.
5959 An absolute expression. The symbol's descriptor is set to the low 16
5960 bits of this expression.
5963 An absolute expression which becomes the symbol's value.
5966 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
5967 or @code{.stabs} statement, the symbol has probably already been created;
5968 you get a half-formed symbol in your object file. This is
5969 compatible with earlier assemblers!
5972 @cindex @code{stabd} directive
5973 @item .stabd @var{type} , @var{other} , @var{desc}
5975 The ``name'' of the symbol generated is not even an empty string.
5976 It is a null pointer, for compatibility. Older assemblers used a
5977 null pointer so they didn't waste space in object files with empty
5980 The symbol's value is set to the location counter,
5981 relocatably. When your program is linked, the value of this symbol
5982 is the address of the location counter when the @code{.stabd} was
5985 @cindex @code{stabn} directive
5986 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
5987 The name of the symbol is set to the empty string @code{""}.
5989 @cindex @code{stabs} directive
5990 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
5991 All five fields are specified.
5997 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
5998 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6000 @cindex string, copying to object file
6001 @cindex string8, copying to object file
6002 @cindex string16, copying to object file
6003 @cindex string32, copying to object file
6004 @cindex string64, copying to object file
6005 @cindex @code{string} directive
6006 @cindex @code{string8} directive
6007 @cindex @code{string16} directive
6008 @cindex @code{string32} directive
6009 @cindex @code{string64} directive
6011 Copy the characters in @var{str} to the object file. You may specify more than
6012 one string to copy, separated by commas. Unless otherwise specified for a
6013 particular machine, the assembler marks the end of each string with a 0 byte.
6014 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6016 The variants @code{string16}, @code{string32} and @code{string64} differ from
6017 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6018 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6019 are stored in target endianness byte order.
6025 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6026 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6031 @section @code{.struct @var{expression}}
6033 @cindex @code{struct} directive
6034 Switch to the absolute section, and set the section offset to @var{expression},
6035 which must be an absolute expression. You might use this as follows:
6044 This would define the symbol @code{field1} to have the value 0, the symbol
6045 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6046 value 8. Assembly would be left in the absolute section, and you would need to
6047 use a @code{.section} directive of some sort to change to some other section
6048 before further assembly.
6052 @section @code{.subsection @var{name}}
6054 @cindex @code{subsection} directive
6055 @cindex Section Stack
6056 This is one of the ELF section stack manipulation directives. The others are
6057 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6058 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6061 This directive replaces the current subsection with @code{name}. The current
6062 section is not changed. The replaced subsection is put onto the section stack
6063 in place of the then current top of stack subsection.
6068 @section @code{.symver}
6069 @cindex @code{symver} directive
6070 @cindex symbol versioning
6071 @cindex versions of symbols
6072 Use the @code{.symver} directive to bind symbols to specific version nodes
6073 within a source file. This is only supported on ELF platforms, and is
6074 typically used when assembling files to be linked into a shared library.
6075 There are cases where it may make sense to use this in objects to be bound
6076 into an application itself so as to override a versioned symbol from a
6079 For ELF targets, the @code{.symver} directive can be used like this:
6081 .symver @var{name}, @var{name2@@nodename}
6083 If the symbol @var{name} is defined within the file
6084 being assembled, the @code{.symver} directive effectively creates a symbol
6085 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6086 just don't try and create a regular alias is that the @var{@@} character isn't
6087 permitted in symbol names. The @var{name2} part of the name is the actual name
6088 of the symbol by which it will be externally referenced. The name @var{name}
6089 itself is merely a name of convenience that is used so that it is possible to
6090 have definitions for multiple versions of a function within a single source
6091 file, and so that the compiler can unambiguously know which version of a
6092 function is being mentioned. The @var{nodename} portion of the alias should be
6093 the name of a node specified in the version script supplied to the linker when
6094 building a shared library. If you are attempting to override a versioned
6095 symbol from a shared library, then @var{nodename} should correspond to the
6096 nodename of the symbol you are trying to override.
6098 If the symbol @var{name} is not defined within the file being assembled, all
6099 references to @var{name} will be changed to @var{name2@@nodename}. If no
6100 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6103 Another usage of the @code{.symver} directive is:
6105 .symver @var{name}, @var{name2@@@@nodename}
6107 In this case, the symbol @var{name} must exist and be defined within
6108 the file being assembled. It is similar to @var{name2@@nodename}. The
6109 difference is @var{name2@@@@nodename} will also be used to resolve
6110 references to @var{name2} by the linker.
6112 The third usage of the @code{.symver} directive is:
6114 .symver @var{name}, @var{name2@@@@@@nodename}
6116 When @var{name} is not defined within the
6117 file being assembled, it is treated as @var{name2@@nodename}. When
6118 @var{name} is defined within the file being assembled, the symbol
6119 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6124 @section @code{.tag @var{structname}}
6126 @cindex COFF structure debugging
6127 @cindex structure debugging, COFF
6128 @cindex @code{tag} directive
6129 This directive is generated by compilers to include auxiliary debugging
6130 information in the symbol table. It is only permitted inside
6131 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6132 definitions in the symbol table with instances of those structures.
6135 @samp{.tag} is only used when generating COFF format output; when
6136 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6142 @section @code{.text @var{subsection}}
6144 @cindex @code{text} directive
6145 Tells @command{@value{AS}} to assemble the following statements onto the end of
6146 the text subsection numbered @var{subsection}, which is an absolute
6147 expression. If @var{subsection} is omitted, subsection number zero
6151 @section @code{.title "@var{heading}"}
6153 @cindex @code{title} directive
6154 @cindex listing control: title line
6155 Use @var{heading} as the title (second line, immediately after the
6156 source file name and pagenumber) when generating assembly listings.
6158 This directive affects subsequent pages, as well as the current page if
6159 it appears within ten lines of the top of a page.
6163 @section @code{.type}
6165 This directive is used to set the type of a symbol.
6169 @c only print the extra heading if both COFF and ELF are set
6170 @subheading COFF Version
6173 @cindex COFF symbol type
6174 @cindex symbol type, COFF
6175 @cindex @code{type} directive (COFF version)
6176 For COFF targets, this directive is permitted only within
6177 @code{.def}/@code{.endef} pairs. It is used like this:
6183 This records the integer @var{int} as the type attribute of a symbol table
6187 @samp{.type} is associated only with COFF format output; when
6188 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6189 directive but ignores it.
6195 @c only print the extra heading if both COFF and ELF are set
6196 @subheading ELF Version
6199 @cindex ELF symbol type
6200 @cindex symbol type, ELF
6201 @cindex @code{type} directive (ELF version)
6202 For ELF targets, the @code{.type} directive is used like this:
6205 .type @var{name} , @var{type description}
6208 This sets the type of symbol @var{name} to be either a
6209 function symbol or an object symbol. There are five different syntaxes
6210 supported for the @var{type description} field, in order to provide
6211 compatibility with various other assemblers.
6213 Because some of the characters used in these syntaxes (such as @samp{@@} and
6214 @samp{#}) are comment characters for some architectures, some of the syntaxes
6215 below do not work on all architectures. The first variant will be accepted by
6216 the GNU assembler on all architectures so that variant should be used for
6217 maximum portability, if you do not need to assemble your code with other
6220 The syntaxes supported are:
6223 .type <name> STT_<TYPE_IN_UPPER_CASE>
6224 .type <name>,#<type>
6225 .type <name>,@@<type>
6226 .type <name>,%>type>
6227 .type <name>,"<type>"
6230 The types supported are:
6235 Mark the symbol as being a function name.
6239 Mark the symbol as being a data object.
6243 Mark the symbol as being a thead-local data object.
6247 Mark the symbol as being a common data object.
6250 Note: Some targets support extra types in addition to those listed above.
6256 @section @code{.uleb128 @var{expressions}}
6258 @cindex @code{uleb128} directive
6259 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6260 compact, variable length representation of numbers used by the DWARF
6261 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6265 @section @code{.val @var{addr}}
6267 @cindex @code{val} directive
6268 @cindex COFF value attribute
6269 @cindex value attribute, COFF
6270 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6271 records the address @var{addr} as the value attribute of a symbol table
6275 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6276 configured for @code{b.out}, it accepts this directive but ignores it.
6282 @section @code{.version "@var{string}"}
6284 @cindex @code{version} directive
6285 This directive creates a @code{.note} section and places into it an ELF
6286 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6291 @section @code{.vtable_entry @var{table}, @var{offset}}
6293 @cindex @code{vtable_entry} directive
6294 This directive finds or creates a symbol @code{table} and creates a
6295 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6298 @section @code{.vtable_inherit @var{child}, @var{parent}}
6300 @cindex @code{vtable_inherit} directive
6301 This directive finds the symbol @code{child} and finds or creates the symbol
6302 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6303 parent whose addend is the value of the child symbol. As a special case the
6304 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6308 @section @code{.warning "@var{string}"}
6309 @cindex warning directive
6310 Similar to the directive @code{.error}
6311 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6314 @section @code{.weak @var{names}}
6316 @cindex @code{weak} directive
6317 This directive sets the weak attribute on the comma separated list of symbol
6318 @code{names}. If the symbols do not already exist, they will be created.
6320 On COFF targets other than PE, weak symbols are a GNU extension. This
6321 directive sets the weak attribute on the comma separated list of symbol
6322 @code{names}. If the symbols do not already exist, they will be created.
6324 On the PE target, weak symbols are supported natively as weak aliases.
6325 When a weak symbol is created that is not an alias, GAS creates an
6326 alternate symbol to hold the default value.
6329 @section @code{.weakref @var{alias}, @var{target}}
6331 @cindex @code{weakref} directive
6332 This directive creates an alias to the target symbol that enables the symbol to
6333 be referenced with weak-symbol semantics, but without actually making it weak.
6334 If direct references or definitions of the symbol are present, then the symbol
6335 will not be weak, but if all references to it are through weak references, the
6336 symbol will be marked as weak in the symbol table.
6338 The effect is equivalent to moving all references to the alias to a separate
6339 assembly source file, renaming the alias to the symbol in it, declaring the
6340 symbol as weak there, and running a reloadable link to merge the object files
6341 resulting from the assembly of the new source file and the old source file that
6342 had the references to the alias removed.
6344 The alias itself never makes to the symbol table, and is entirely handled
6345 within the assembler.
6348 @section @code{.word @var{expressions}}
6350 @cindex @code{word} directive
6351 This directive expects zero or more @var{expressions}, of any section,
6352 separated by commas.
6355 For each expression, @command{@value{AS}} emits a 32-bit number.
6358 For each expression, @command{@value{AS}} emits a 16-bit number.
6363 The size of the number emitted, and its byte order,
6364 depend on what target computer the assembly is for.
6367 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6368 @c happen---32-bit addressability, period; no long/short jumps.
6369 @ifset DIFF-TBL-KLUGE
6370 @cindex difference tables altered
6371 @cindex altered difference tables
6373 @emph{Warning: Special Treatment to support Compilers}
6377 Machines with a 32-bit address space, but that do less than 32-bit
6378 addressing, require the following special treatment. If the machine of
6379 interest to you does 32-bit addressing (or doesn't require it;
6380 @pxref{Machine Dependencies}), you can ignore this issue.
6383 In order to assemble compiler output into something that works,
6384 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6385 Directives of the form @samp{.word sym1-sym2} are often emitted by
6386 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6387 directive of the form @samp{.word sym1-sym2}, and the difference between
6388 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6389 creates a @dfn{secondary jump table}, immediately before the next label.
6390 This secondary jump table is preceded by a short-jump to the
6391 first byte after the secondary table. This short-jump prevents the flow
6392 of control from accidentally falling into the new table. Inside the
6393 table is a long-jump to @code{sym2}. The original @samp{.word}
6394 contains @code{sym1} minus the address of the long-jump to
6397 If there were several occurrences of @samp{.word sym1-sym2} before the
6398 secondary jump table, all of them are adjusted. If there was a
6399 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6400 long-jump to @code{sym4} is included in the secondary jump table,
6401 and the @code{.word} directives are adjusted to contain @code{sym3}
6402 minus the address of the long-jump to @code{sym4}; and so on, for as many
6403 entries in the original jump table as necessary.
6406 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6407 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6408 assembly language programmers.
6411 @c end DIFF-TBL-KLUGE
6414 @section Deprecated Directives
6416 @cindex deprecated directives
6417 @cindex obsolescent directives
6418 One day these directives won't work.
6419 They are included for compatibility with older assemblers.
6426 @node Object Attributes
6427 @chapter Object Attributes
6428 @cindex object attributes
6430 @command{@value{AS}} assembles source files written for a specific architecture
6431 into object files for that architecture. But not all object files are alike.
6432 Many architectures support incompatible variations. For instance, floating
6433 point arguments might be passed in floating point registers if the object file
6434 requires hardware floating point support---or floating point arguments might be
6435 passed in integer registers if the object file supports processors with no
6436 hardware floating point unit. Or, if two objects are built for different
6437 generations of the same architecture, the combination may require the
6438 newer generation at run-time.
6440 This information is useful during and after linking. At link time,
6441 @command{@value{LD}} can warn about incompatible object files. After link
6442 time, tools like @command{gdb} can use it to process the linked file
6445 Compatibility information is recorded as a series of object attributes. Each
6446 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6447 string, and indicates who sets the meaning of the tag. The tag is an integer,
6448 and indicates what property the attribute describes. The value may be a string
6449 or an integer, and indicates how the property affects this object. Missing
6450 attributes are the same as attributes with a zero value or empty string value.
6452 Object attributes were developed as part of the ABI for the ARM Architecture.
6453 The file format is documented in @cite{ELF for the ARM Architecture}.
6456 * GNU Object Attributes:: @sc{gnu} Object Attributes
6457 * Defining New Object Attributes:: Defining New Object Attributes
6460 @node GNU Object Attributes
6461 @section @sc{gnu} Object Attributes
6463 The @code{.gnu_attribute} directive records an object attribute
6464 with vendor @samp{gnu}.
6466 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6467 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6468 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6469 2} is set for architecture-independent attributes and clear for
6470 architecture-dependent ones.
6472 @subsection Common @sc{gnu} attributes
6474 These attributes are valid on all architectures.
6477 @item Tag_compatibility (32)
6478 The compatibility attribute takes an integer flag value and a vendor name. If
6479 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6480 then the file is only compatible with the named toolchain. If it is greater
6481 than 1, the file can only be processed by other toolchains under some private
6482 arrangement indicated by the flag value and the vendor name.
6485 @subsection MIPS Attributes
6488 @item Tag_GNU_MIPS_ABI_FP (4)
6489 The floating-point ABI used by this object file. The value will be:
6493 0 for files not affected by the floating-point ABI.
6495 1 for files using the hardware floating-point with a standard double-precision
6498 2 for files using the hardware floating-point ABI with a single-precision FPU.
6500 3 for files using the software floating-point ABI.
6502 4 for files using the hardware floating-point ABI with 64-bit wide
6503 double-precision floating-point registers and 32-bit wide general
6508 @subsection PowerPC Attributes
6511 @item Tag_GNU_Power_ABI_FP (4)
6512 The floating-point ABI used by this object file. The value will be:
6516 0 for files not affected by the floating-point ABI.
6518 1 for files using double-precision hardware floating-point ABI.
6520 2 for files using the software floating-point ABI.
6522 3 for files using single-precision hardware floating-point ABI.
6525 @item Tag_GNU_Power_ABI_Vector (8)
6526 The vector ABI used by this object file. The value will be:
6530 0 for files not affected by the vector ABI.
6532 1 for files using general purpose registers to pass vectors.
6534 2 for files using AltiVec registers to pass vectors.
6536 3 for files using SPE registers to pass vectors.
6540 @node Defining New Object Attributes
6541 @section Defining New Object Attributes
6543 If you want to define a new @sc{gnu} object attribute, here are the places you
6544 will need to modify. New attributes should be discussed on the @samp{binutils}
6549 This manual, which is the official register of attributes.
6551 The header for your architecture @file{include/elf}, to define the tag.
6553 The @file{bfd} support file for your architecture, to merge the attribute
6554 and issue any appropriate link warnings.
6556 Test cases in @file{ld/testsuite} for merging and link warnings.
6558 @file{binutils/readelf.c} to display your attribute.
6560 GCC, if you want the compiler to mark the attribute automatically.
6566 @node Machine Dependencies
6567 @chapter Machine Dependent Features
6569 @cindex machine dependencies
6570 The machine instruction sets are (almost by definition) different on
6571 each machine where @command{@value{AS}} runs. Floating point representations
6572 vary as well, and @command{@value{AS}} often supports a few additional
6573 directives or command-line options for compatibility with other
6574 assemblers on a particular platform. Finally, some versions of
6575 @command{@value{AS}} support special pseudo-instructions for branch
6578 This chapter discusses most of these differences, though it does not
6579 include details on any machine's instruction set. For details on that
6580 subject, see the hardware manufacturer's manual.
6584 * Alpha-Dependent:: Alpha Dependent Features
6587 * ARC-Dependent:: ARC Dependent Features
6590 * ARM-Dependent:: ARM Dependent Features
6593 * AVR-Dependent:: AVR Dependent Features
6596 * BFIN-Dependent:: BFIN Dependent Features
6599 * CR16-Dependent:: CR16 Dependent Features
6602 * CRIS-Dependent:: CRIS Dependent Features
6605 * D10V-Dependent:: D10V Dependent Features
6608 * D30V-Dependent:: D30V Dependent Features
6611 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6614 * HPPA-Dependent:: HPPA Dependent Features
6617 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6620 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6623 * i860-Dependent:: Intel 80860 Dependent Features
6626 * i960-Dependent:: Intel 80960 Dependent Features
6629 * IA-64-Dependent:: Intel IA-64 Dependent Features
6632 * IP2K-Dependent:: IP2K Dependent Features
6635 * M32C-Dependent:: M32C Dependent Features
6638 * M32R-Dependent:: M32R Dependent Features
6641 * M68K-Dependent:: M680x0 Dependent Features
6644 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6647 * MIPS-Dependent:: MIPS Dependent Features
6650 * MMIX-Dependent:: MMIX Dependent Features
6653 * MSP430-Dependent:: MSP430 Dependent Features
6656 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6657 * SH64-Dependent:: SuperH SH64 Dependent Features
6660 * PDP-11-Dependent:: PDP-11 Dependent Features
6663 * PJ-Dependent:: picoJava Dependent Features
6666 * PPC-Dependent:: PowerPC Dependent Features
6669 * Sparc-Dependent:: SPARC Dependent Features
6672 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6675 * V850-Dependent:: V850 Dependent Features
6678 * Xtensa-Dependent:: Xtensa Dependent Features
6681 * Z80-Dependent:: Z80 Dependent Features
6684 * Z8000-Dependent:: Z8000 Dependent Features
6687 * Vax-Dependent:: VAX Dependent Features
6694 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6695 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6696 @c peculiarity: to preserve cross-references, there must be a node called
6697 @c "Machine Dependencies". Hence the conditional nodenames in each
6698 @c major node below. Node defaulting in makeinfo requires adjacency of
6699 @c node and sectioning commands; hence the repetition of @chapter BLAH
6700 @c in both conditional blocks.
6703 @include c-alpha.texi
6719 @include c-bfin.texi
6723 @include c-cr16.texi
6727 @include c-cris.texi
6732 @node Machine Dependencies
6733 @chapter Machine Dependent Features
6735 The machine instruction sets are different on each Renesas chip family,
6736 and there are also some syntax differences among the families. This
6737 chapter describes the specific @command{@value{AS}} features for each
6741 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6742 * SH-Dependent:: Renesas SH Dependent Features
6749 @include c-d10v.texi
6753 @include c-d30v.texi
6757 @include c-h8300.texi
6761 @include c-hppa.texi
6765 @include c-i370.texi
6769 @include c-i386.texi
6773 @include c-i860.texi
6777 @include c-i960.texi
6781 @include c-ia64.texi
6785 @include c-ip2k.texi
6789 @include c-m32c.texi
6793 @include c-m32r.texi
6797 @include c-m68k.texi
6801 @include c-m68hc11.texi
6805 @include c-mips.texi
6809 @include c-mmix.texi
6813 @include c-msp430.texi
6817 @include c-ns32k.texi
6821 @include c-pdp11.texi
6834 @include c-sh64.texi
6838 @include c-sparc.texi
6842 @include c-tic54x.texi
6858 @include c-v850.texi
6862 @include c-xtensa.texi
6866 @c reverse effect of @down at top of generic Machine-Dep chapter
6870 @node Reporting Bugs
6871 @chapter Reporting Bugs
6872 @cindex bugs in assembler
6873 @cindex reporting bugs in assembler
6875 Your bug reports play an essential role in making @command{@value{AS}} reliable.
6877 Reporting a bug may help you by bringing a solution to your problem, or it may
6878 not. But in any case the principal function of a bug report is to help the
6879 entire community by making the next version of @command{@value{AS}} work better.
6880 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
6882 In order for a bug report to serve its purpose, you must include the
6883 information that enables us to fix the bug.
6886 * Bug Criteria:: Have you found a bug?
6887 * Bug Reporting:: How to report bugs
6891 @section Have You Found a Bug?
6892 @cindex bug criteria
6894 If you are not sure whether you have found a bug, here are some guidelines:
6897 @cindex fatal signal
6898 @cindex assembler crash
6899 @cindex crash of assembler
6901 If the assembler gets a fatal signal, for any input whatever, that is a
6902 @command{@value{AS}} bug. Reliable assemblers never crash.
6904 @cindex error on valid input
6906 If @command{@value{AS}} produces an error message for valid input, that is a bug.
6908 @cindex invalid input
6910 If @command{@value{AS}} does not produce an error message for invalid input, that
6911 is a bug. However, you should note that your idea of ``invalid input'' might
6912 be our idea of ``an extension'' or ``support for traditional practice''.
6915 If you are an experienced user of assemblers, your suggestions for improvement
6916 of @command{@value{AS}} are welcome in any case.
6920 @section How to Report Bugs
6922 @cindex assembler bugs, reporting
6924 A number of companies and individuals offer support for @sc{gnu} products. If
6925 you obtained @command{@value{AS}} from a support organization, we recommend you
6926 contact that organization first.
6928 You can find contact information for many support companies and
6929 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
6933 In any event, we also recommend that you send bug reports for @command{@value{AS}}
6937 The fundamental principle of reporting bugs usefully is this:
6938 @strong{report all the facts}. If you are not sure whether to state a
6939 fact or leave it out, state it!
6941 Often people omit facts because they think they know what causes the problem
6942 and assume that some details do not matter. Thus, you might assume that the
6943 name of a symbol you use in an example does not matter. Well, probably it does
6944 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
6945 happens to fetch from the location where that name is stored in memory;
6946 perhaps, if the name were different, the contents of that location would fool
6947 the assembler into doing the right thing despite the bug. Play it safe and
6948 give a specific, complete example. That is the easiest thing for you to do,
6949 and the most helpful.
6951 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
6952 it is new to us. Therefore, always write your bug reports on the assumption
6953 that the bug has not been reported previously.
6955 Sometimes people give a few sketchy facts and ask, ``Does this ring a
6956 bell?'' This cannot help us fix a bug, so it is basically useless. We
6957 respond by asking for enough details to enable us to investigate.
6958 You might as well expedite matters by sending them to begin with.
6960 To enable us to fix the bug, you should include all these things:
6964 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
6965 it with the @samp{--version} argument.
6967 Without this, we will not know whether there is any point in looking for
6968 the bug in the current version of @command{@value{AS}}.
6971 Any patches you may have applied to the @command{@value{AS}} source.
6974 The type of machine you are using, and the operating system name and
6978 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
6982 The command arguments you gave the assembler to assemble your example and
6983 observe the bug. To guarantee you will not omit something important, list them
6984 all. A copy of the Makefile (or the output from make) is sufficient.
6986 If we were to try to guess the arguments, we would probably guess wrong
6987 and then we might not encounter the bug.
6990 A complete input file that will reproduce the bug. If the bug is observed when
6991 the assembler is invoked via a compiler, send the assembler source, not the
6992 high level language source. Most compilers will produce the assembler source
6993 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
6994 the options @samp{-v --save-temps}; this will save the assembler source in a
6995 file with an extension of @file{.s}, and also show you exactly how
6996 @command{@value{AS}} is being run.
6999 A description of what behavior you observe that you believe is
7000 incorrect. For example, ``It gets a fatal signal.''
7002 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7003 will certainly notice it. But if the bug is incorrect output, we might not
7004 notice unless it is glaringly wrong. You might as well not give us a chance to
7007 Even if the problem you experience is a fatal signal, you should still say so
7008 explicitly. Suppose something strange is going on, such as, your copy of
7009 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7010 library on your system. (This has happened!) Your copy might crash and ours
7011 would not. If you told us to expect a crash, then when ours fails to crash, we
7012 would know that the bug was not happening for us. If you had not told us to
7013 expect a crash, then we would not be able to draw any conclusion from our
7017 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7018 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7019 option. Always send diffs from the old file to the new file. If you even
7020 discuss something in the @command{@value{AS}} source, refer to it by context, not
7023 The line numbers in our development sources will not match those in your
7024 sources. Your line numbers would convey no useful information to us.
7027 Here are some things that are not necessary:
7031 A description of the envelope of the bug.
7033 Often people who encounter a bug spend a lot of time investigating
7034 which changes to the input file will make the bug go away and which
7035 changes will not affect it.
7037 This is often time consuming and not very useful, because the way we
7038 will find the bug is by running a single example under the debugger
7039 with breakpoints, not by pure deduction from a series of examples.
7040 We recommend that you save your time for something else.
7042 Of course, if you can find a simpler example to report @emph{instead}
7043 of the original one, that is a convenience for us. Errors in the
7044 output will be easier to spot, running under the debugger will take
7045 less time, and so on.
7047 However, simplification is not vital; if you do not want to do this,
7048 report the bug anyway and send us the entire test case you used.
7051 A patch for the bug.
7053 A patch for the bug does help us if it is a good one. But do not omit
7054 the necessary information, such as the test case, on the assumption that
7055 a patch is all we need. We might see problems with your patch and decide
7056 to fix the problem another way, or we might not understand it at all.
7058 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7059 construct an example that will make the program follow a certain path through
7060 the code. If you do not send us the example, we will not be able to construct
7061 one, so we will not be able to verify that the bug is fixed.
7063 And if we cannot understand what bug you are trying to fix, or why your
7064 patch should be an improvement, we will not install it. A test case will
7065 help us to understand.
7068 A guess about what the bug is or what it depends on.
7070 Such guesses are usually wrong. Even we cannot guess right about such
7071 things without first using the debugger to find the facts.
7074 @node Acknowledgements
7075 @chapter Acknowledgements
7077 If you have contributed to GAS and your name isn't listed here,
7078 it is not meant as a slight. We just don't know about it. Send mail to the
7079 maintainer, and we'll correct the situation. Currently
7081 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
7083 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7086 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7087 information and the 68k series machines, most of the preprocessing pass, and
7088 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7090 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7091 many bug fixes, including merging support for several processors, breaking GAS
7092 up to handle multiple object file format back ends (including heavy rewrite,
7093 testing, an integration of the coff and b.out back ends), adding configuration
7094 including heavy testing and verification of cross assemblers and file splits
7095 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7096 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7097 port (including considerable amounts of reverse engineering), a SPARC opcode
7098 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7099 assertions and made them work, much other reorganization, cleanup, and lint.
7101 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7102 in format-specific I/O modules.
7104 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7105 has done much work with it since.
7107 The Intel 80386 machine description was written by Eliot Dresselhaus.
7109 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7111 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7112 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7114 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7115 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7116 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7117 support a.out format.
7119 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7120 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7121 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7122 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7125 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7126 simplified the configuration of which versions accept which directives. He
7127 updated the 68k machine description so that Motorola's opcodes always produced
7128 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7129 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7130 cross-compilation support, and one bug in relaxation that took a week and
7131 required the proverbial one-bit fix.
7133 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7134 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7135 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7136 PowerPC assembler, and made a few other minor patches.
7138 Steve Chamberlain made GAS able to generate listings.
7140 Hewlett-Packard contributed support for the HP9000/300.
7142 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7143 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7144 formats). This work was supported by both the Center for Software Science at
7145 the University of Utah and Cygnus Support.
7147 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7148 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7149 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7150 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7151 and some initial 64-bit support).
7153 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7155 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7156 support for openVMS/Alpha.
7158 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7161 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7162 Inc.@: added support for Xtensa processors.
7164 Several engineers at Cygnus Support have also provided many small bug fixes and
7165 configuration enhancements.
7167 Many others have contributed large or small bugfixes and enhancements. If
7168 you have contributed significant work and are not mentioned on this list, and
7169 want to be, let us know. Some of the history has been lost; we are not
7170 intentionally leaving anyone out.
7175 @unnumbered AS Index
projects / platform / upstream / binutils.git / blob