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
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
55 @set abnormal-separator
59 @settitle Using @value{AS}
62 @settitle Using @value{AS} (@value{TARGET})
64 @setchapternewpage odd
69 @c WARE! Some of the machine-dependent sections contain tables of machine
70 @c instructions. Except in multi-column format, these tables look silly.
71 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
72 @c the multi-col format is faked within @example sections.
74 @c Again unfortunately, the natural size that fits on a page, for these tables,
75 @c is different depending on whether or not smallbook is turned on.
76 @c This matters, because of order: text flow switches columns at each page
79 @c The format faked in this source works reasonably well for smallbook,
80 @c not well for the default large-page format. This manual expects that if you
81 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
82 @c tables in question. You can turn on one without the other at your
83 @c discretion, of course.
86 @c the insn tables look just as silly in info files regardless of smallbook,
87 @c might as well show 'em anyways.
93 * As: (as). The GNU assembler.
94 * Gas: (as). The GNU assembler.
103 This file documents the GNU Assembler "@value{AS}".
105 @c man begin COPYRIGHT
106 Copyright (C) 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002 Free Software Foundation, Inc.
108 Permission is granted to copy, distribute and/or modify this document
109 under the terms of the GNU Free Documentation License, Version 1.1
110 or any later version published by the Free Software Foundation;
111 with no Invariant Sections, with no Front-Cover Texts, and with no
112 Back-Cover Texts. A copy of the license is included in the
113 section entitled ``GNU Free Documentation License''.
118 Permission is granted to process this file through Tex and print the
119 results, provided the printed document carries copying permission
120 notice identical to this one except for the removal of this paragraph
121 (this paragraph not being relevant to the printed manual).
127 @title Using @value{AS}
128 @subtitle The @sc{gnu} Assembler
130 @subtitle for the @value{TARGET} family
133 @subtitle Version @value{VERSION}
136 The Free Software Foundation Inc. thanks The Nice Computer
137 Company of Australia for loaning Dean Elsner to write the
138 first (Vax) version of @command{as} for Project @sc{gnu}.
139 The proprietors, management and staff of TNCCA thank FSF for
140 distracting the boss while they got some work
143 @author Dean Elsner, Jay Fenlason & friends
147 \hfill {\it Using {\tt @value{AS}}}\par
148 \hfill Edited by Cygnus Support\par
150 %"boxit" macro for figures:
151 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
152 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
153 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
154 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
155 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
158 @vskip 0pt plus 1filll
159 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002 Free Software Foundation, Inc.
161 Permission is granted to copy, distribute and/or modify this document
162 under the terms of the GNU Free Documentation License, Version 1.1
163 or any later version published by the Free Software Foundation;
164 with no Invariant Sections, with no Front-Cover Texts, and with no
165 Back-Cover Texts. A copy of the license is included in the
166 section entitled ``GNU Free Documentation License''.
172 @top Using @value{AS}
174 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}} version
177 This version of the file describes @command{@value{AS}} configured to generate
178 code for @value{TARGET} architectures.
181 This document is distributed under the terms of the GNU Free
182 Documentation License. A copy of the license is included in the
183 section entitled ``GNU Free Documentation License''.
186 * Overview:: Overview
187 * Invoking:: Command-Line Options
189 * Sections:: Sections and Relocation
191 * Expressions:: Expressions
192 * Pseudo Ops:: Assembler Directives
193 * Machine Dependencies:: Machine Dependent Features
194 * Reporting Bugs:: Reporting Bugs
195 * Acknowledgements:: Who Did What
196 * GNU Free Documentation License:: GNU Free Documentation License
204 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
206 This version of the manual describes @command{@value{AS}} configured to generate
207 code for @value{TARGET} architectures.
211 @cindex invocation summary
212 @cindex option summary
213 @cindex summary of options
214 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
215 @pxref{Invoking,,Command-Line Options}.
217 @c man title AS the portable GNU assembler.
221 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
225 @c We don't use deffn and friends for the following because they seem
226 @c to be limited to one line for the header.
228 @c man begin SYNOPSIS
229 @value{AS} [@b{-a}[@b{cdhlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
230 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
231 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
232 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
233 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
234 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}] [@b{-o}
235 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
236 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
237 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{--target-help}]
238 [@var{target-options}] [@b{--}|@var{files} @dots{}]
240 @c Target dependent options are listed below. Keep the list sorted.
241 @c Add an empty line for separation.
243 @c am29k has no machine-dependent assembler options
247 @emph{Target Alpha options:}
249 [@b{-mdebug} | @b{-no-mdebug}]
250 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
251 [@b{-F}] [@b{-32addr}]
255 @emph{Target ARC options:}
261 @emph{Target ARM options:}
262 @c Don't document the deprecated options
263 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
264 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
265 [@b{-mfpu}=@var{floating-point-format}]
266 [@b{-mfloat-abi}=@var{abi}]
267 [@b{-meabi}=@var{ver}]
270 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
271 @b{-mapcs-reentrant}]
272 [@b{-mthumb-interwork}] [@b{-k}]
276 @emph{Target CRIS options:}
277 [@b{--underscore} | @b{--no-underscore}]
279 [@b{--emulation=criself} | @b{--emulation=crisaout}]
280 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
281 @c Deprecated -- deliberately not documented.
286 @emph{Target D10V options:}
291 @emph{Target D30V options:}
292 [@b{-O}|@b{-n}|@b{-N}]
295 @c Renesas family chips have no machine-dependent assembler options
298 @c HPPA has no machine-dependent assembler options (yet).
302 @emph{Target i386 options:}
303 [@b{--32}|@b{--64}] [@b{-n}]
307 @emph{Target i960 options:}
308 @c see md_parse_option in tc-i960.c
309 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
311 [@b{-b}] [@b{-no-relax}]
315 @emph{Target IA-64 options:}
316 [@b{-mconstant-gp}|@b{-mauto-pic}]
317 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
319 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
320 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
321 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
322 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
326 @emph{Target IP2K options:}
327 [@b{-mip2022}|@b{-mip2022ext}]
331 @emph{Target M32R options:}
332 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
337 @emph{Target M680X0 options:}
338 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
342 @emph{Target M68HC11 options:}
343 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
344 [@b{-mshort}|@b{-mlong}]
345 [@b{-mshort-double}|@b{-mlong-double}]
346 [@b{--force-long-branchs}] [@b{--short-branchs}]
347 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
348 [@b{--print-opcodes}] [@b{--generate-example}]
352 @emph{Target MCORE options:}
353 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
354 [@b{-mcpu=[210|340]}]
358 @emph{Target MIPS options:}
359 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
360 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
361 [@b{-non_shared}] [@b{-xgot}]
362 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
363 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
364 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
365 [@b{-mips64}] [@b{-mips64r2}]
366 [@b{-construct-floats}] [@b{-no-construct-floats}]
367 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
368 [@b{-mfix7000}] [@b{-mno-fix7000}]
369 [@b{-mips16}] [@b{-no-mips16}]
370 [@b{-mips3d}] [@b{-no-mips3d}]
371 [@b{-mdmx}] [@b{-no-mdmx}]
372 [@b{-mdebug}] [@b{-no-mdebug}]
373 [@b{-mpdr}] [@b{-mno-pdr}]
377 @emph{Target MMIX options:}
378 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
379 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
380 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
381 [@b{--linker-allocated-gregs}]
385 @emph{Target PDP11 options:}
386 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
387 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
388 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
392 @emph{Target picoJava options:}
397 @emph{Target PowerPC options:}
398 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
399 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}|
400 @b{-mbooke32}|@b{-mbooke64}]
401 [@b{-mcom}|@b{-many}|@b{-maltivec}] [@b{-memb}]
402 [@b{-mregnames}|@b{-mno-regnames}]
403 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
404 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
405 [@b{-msolaris}|@b{-mno-solaris}]
409 @emph{Target SPARC options:}
410 @c The order here is important. See c-sparc.texi.
411 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
412 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
413 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
418 @emph{Target TIC54X options:}
419 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
420 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
423 @c Z8000 has no machine-dependent assembler options
427 @emph{Target Xtensa options:}
428 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
429 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
430 [@b{--[no-]transform}]
431 [@b{--rename-section} @var{oldname}=@var{newname}]
440 Turn on listings, in any of a variety of ways:
444 omit false conditionals
447 omit debugging directives
450 include high-level source
456 include macro expansions
459 omit forms processing
465 set the name of the listing file
468 You may combine these options; for example, use @samp{-aln} for assembly
469 listing without forms processing. The @samp{=file} option, if used, must be
470 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
473 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
476 Ignored. This option is accepted for script compatibility with calls to
479 @item --defsym @var{sym}=@var{value}
480 Define the symbol @var{sym} to be @var{value} before assembling the input file.
481 @var{value} must be an integer constant. As in C, a leading @samp{0x}
482 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
485 ``fast''---skip whitespace and comment preprocessing (assume source is
490 Generate debugging information for each assembler source line using whichever
491 debug format is preferred by the target. This currently means either STABS,
495 Generate stabs debugging information for each assembler line. This
496 may help debugging assembler code, if the debugger can handle it.
499 Generate stabs debugging information for each assembler line, with GNU
500 extensions that probably only gdb can handle, and that could make other
501 debuggers crash or refuse to read your program. This
502 may help debugging assembler code. Currently the only GNU extension is
503 the location of the current working directory at assembling time.
506 Generate DWARF2 debugging information for each assembler line. This
507 may help debugging assembler code, if the debugger can handle it. Note---this
508 option is only supported by some targets, not all of them.
511 Print a summary of the command line options and exit.
514 Print a summary of all target specific options and exit.
517 Add directory @var{dir} to the search list for @code{.include} directives.
520 Don't warn about signed overflow.
523 @ifclear DIFF-TBL-KLUGE
524 This option is accepted but has no effect on the @value{TARGET} family.
526 @ifset DIFF-TBL-KLUGE
527 Issue warnings when difference tables altered for long displacements.
532 Keep (in the symbol table) local symbols. On traditional a.out systems
533 these start with @samp{L}, but different systems have different local
536 @item --listing-lhs-width=@var{number}
537 Set the maximum width, in words, of the output data column for an assembler
538 listing to @var{number}.
540 @item --listing-lhs-width2=@var{number}
541 Set the maximum width, in words, of the output data column for continuation
542 lines in an assembler listing to @var{number}.
544 @item --listing-rhs-width=@var{number}
545 Set the maximum width of an input source line, as displayed in a listing, to
548 @item --listing-cont-lines=@var{number}
549 Set the maximum number of lines printed in a listing for a single line of input
552 @item -o @var{objfile}
553 Name the object-file output from @command{@value{AS}} @var{objfile}.
556 Fold the data section into the text section.
558 @kindex --hash-size=@var{number}
559 Set the default size of GAS's hash tables to a prime number close to
560 @var{number}. Increasing this value can reduce the length of time it takes the
561 assembler to perform its tasks, at the expense of increasing the assembler's
562 memory requirements. Similarly reducing this value can reduce the memory
563 requirements at the expense of speed.
565 @item --reduce-memory-overheads
566 This option reduces GAS's memory requirements, at the expense of making the
567 assembly processes slower. Currently this switch is a synonym for
568 @samp{--hash-size=4051}, but in the future it may have other effects as well.
571 Print the maximum space (in bytes) and total time (in seconds) used by
574 @item --strip-local-absolute
575 Remove local absolute symbols from the outgoing symbol table.
579 Print the @command{as} version.
582 Print the @command{as} version and exit.
586 Suppress warning messages.
588 @item --fatal-warnings
589 Treat warnings as errors.
592 Don't suppress warning messages or treat them as errors.
601 Generate an object file even after errors.
603 @item -- | @var{files} @dots{}
604 Standard input, or source files to assemble.
609 The following options are available when @value{AS} is configured for
614 This option selects the core processor variant.
616 Select either big-endian (-EB) or little-endian (-EL) output.
621 The following options are available when @value{AS} is configured for the ARM
625 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
626 Specify which ARM processor variant is the target.
627 @item -march=@var{architecture}[+@var{extension}@dots{}]
628 Specify which ARM architecture variant is used by the target.
629 @item -mfpu=@var{floating-point-format}
630 Select which Floating Point architecture is the target.
631 @item -mfloat-abi=@var{abi}
632 Select which floating point ABI is in use.
634 Enable Thumb only instruction decoding.
635 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
636 Select which procedure calling convention is in use.
638 Select either big-endian (-EB) or little-endian (-EL) output.
639 @item -mthumb-interwork
640 Specify that the code has been generated with interworking between Thumb and
643 Specify that PIC code has been generated.
648 See the info pages for documentation of the CRIS-specific options.
652 The following options are available when @value{AS} is configured for
655 @cindex D10V optimization
656 @cindex optimization, D10V
658 Optimize output by parallelizing instructions.
663 The following options are available when @value{AS} is configured for a D30V
666 @cindex D30V optimization
667 @cindex optimization, D30V
669 Optimize output by parallelizing instructions.
673 Warn when nops are generated.
675 @cindex D30V nops after 32-bit multiply
677 Warn when a nop after a 32-bit multiply instruction is generated.
682 The following options are available when @value{AS} is configured for the
683 Intel 80960 processor.
686 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
687 Specify which variant of the 960 architecture is the target.
690 Add code to collect statistics about branches taken.
693 Do not alter compare-and-branch instructions for long displacements;
700 The following options are available when @value{AS} is configured for the
706 Specifies that the extended IP2022 instructions are allowed.
709 Restores the default behaviour, which restricts the permitted instructions to
710 just the basic IP2022 ones.
716 The following options are available when @value{AS} is configured for the
717 Renesas M32R (formerly Mitsubishi M32R) series.
722 Specify which processor in the M32R family is the target. The default
723 is normally the M32R, but this option changes it to the M32RX.
725 @item --warn-explicit-parallel-conflicts or --Wp
726 Produce warning messages when questionable parallel constructs are
729 @item --no-warn-explicit-parallel-conflicts or --Wnp
730 Do not produce warning messages when questionable parallel constructs are
737 The following options are available when @value{AS} is configured for the
738 Motorola 68000 series.
743 Shorten references to undefined symbols, to one word instead of two.
745 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
746 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
747 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
748 Specify what processor in the 68000 family is the target. The default
749 is normally the 68020, but this can be changed at configuration time.
751 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
752 The target machine does (or does not) have a floating-point coprocessor.
753 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
754 the basic 68000 is not compatible with the 68881, a combination of the
755 two can be specified, since it's possible to do emulation of the
756 coprocessor instructions with the main processor.
758 @item -m68851 | -mno-68851
759 The target machine does (or does not) have a memory-management
760 unit coprocessor. The default is to assume an MMU for 68020 and up.
767 For details about the PDP-11 machine dependent features options,
768 see @ref{PDP-11-Options}.
771 @item -mpic | -mno-pic
772 Generate position-independent (or position-dependent) code. The
773 default is @option{-mpic}.
776 @itemx -mall-extensions
777 Enable all instruction set extensions. This is the default.
779 @item -mno-extensions
780 Disable all instruction set extensions.
782 @item -m@var{extension} | -mno-@var{extension}
783 Enable (or disable) a particular instruction set extension.
786 Enable the instruction set extensions supported by a particular CPU, and
787 disable all other extensions.
789 @item -m@var{machine}
790 Enable the instruction set extensions supported by a particular machine
791 model, and disable all other extensions.
797 The following options are available when @value{AS} is configured for
798 a picoJava processor.
802 @cindex PJ endianness
803 @cindex endianness, PJ
804 @cindex big endian output, PJ
806 Generate ``big endian'' format output.
808 @cindex little endian output, PJ
810 Generate ``little endian'' format output.
816 The following options are available when @value{AS} is configured for the
817 Motorola 68HC11 or 68HC12 series.
821 @item -m68hc11 | -m68hc12 | -m68hcs12
822 Specify what processor is the target. The default is
823 defined by the configuration option when building the assembler.
826 Specify to use the 16-bit integer ABI.
829 Specify to use the 32-bit integer ABI.
832 Specify to use the 32-bit double ABI.
835 Specify to use the 64-bit double ABI.
837 @item --force-long-branchs
838 Relative branches are turned into absolute ones. This concerns
839 conditional branches, unconditional branches and branches to a
842 @item -S | --short-branchs
843 Do not turn relative branchs into absolute ones
844 when the offset is out of range.
846 @item --strict-direct-mode
847 Do not turn the direct addressing mode into extended addressing mode
848 when the instruction does not support direct addressing mode.
850 @item --print-insn-syntax
851 Print the syntax of instruction in case of error.
853 @item --print-opcodes
854 print the list of instructions with syntax and then exit.
856 @item --generate-example
857 print an example of instruction for each possible instruction and then exit.
858 This option is only useful for testing @command{@value{AS}}.
864 The following options are available when @command{@value{AS}} is configured
865 for the SPARC architecture:
868 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
869 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
870 Explicitly select a variant of the SPARC architecture.
872 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
873 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
875 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
876 UltraSPARC extensions.
878 @item -xarch=v8plus | -xarch=v8plusa
879 For compatibility with the Solaris v9 assembler. These options are
880 equivalent to -Av8plus and -Av8plusa, respectively.
883 Warn when the assembler switches to another architecture.
888 The following options are available when @value{AS} is configured for the 'c54x
893 Enable extended addressing mode. All addresses and relocations will assume
894 extended addressing (usually 23 bits).
895 @item -mcpu=@var{CPU_VERSION}
896 Sets the CPU version being compiled for.
897 @item -merrors-to-file @var{FILENAME}
898 Redirect error output to a file, for broken systems which don't support such
899 behaviour in the shell.
904 The following options are available when @value{AS} is configured for
905 a @sc{mips} processor.
909 This option sets the largest size of an object that can be referenced
910 implicitly with the @code{gp} register. It is only accepted for targets that
911 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
913 @cindex MIPS endianness
914 @cindex endianness, MIPS
915 @cindex big endian output, MIPS
917 Generate ``big endian'' format output.
919 @cindex little endian output, MIPS
921 Generate ``little endian'' format output.
933 Generate code for a particular @sc{mips} Instruction Set Architecture level.
934 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
935 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
936 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
937 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
939 correspond to generic
940 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
941 and @samp{MIPS64 Release 2}
942 ISA processors, respectively.
944 @item -march=@var{CPU}
945 Generate code for a particular @sc{mips} cpu.
947 @item -mtune=@var{cpu}
948 Schedule and tune for a particular @sc{mips} cpu.
952 Cause nops to be inserted if the read of the destination register
953 of an mfhi or mflo instruction occurs in the following two instructions.
957 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
958 section instead of the standard ELF .stabs sections.
962 Control generation of @code{.pdr} sections.
966 The register sizes are normally inferred from the ISA and ABI, but these
967 flags force a certain group of registers to be treated as 32 bits wide at
968 all times. @samp{-mgp32} controls the size of general-purpose registers
969 and @samp{-mfp32} controls the size of floating-point registers.
973 Generate code for the MIPS 16 processor. This is equivalent to putting
974 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
975 turns off this option.
979 Generate code for the MIPS-3D Application Specific Extension.
980 This tells the assembler to accept MIPS-3D instructions.
981 @samp{-no-mips3d} turns off this option.
985 Generate code for the MDMX Application Specific Extension.
986 This tells the assembler to accept MDMX instructions.
987 @samp{-no-mdmx} turns off this option.
989 @item --construct-floats
990 @itemx --no-construct-floats
991 The @samp{--no-construct-floats} option disables the construction of
992 double width floating point constants by loading the two halves of the
993 value into the two single width floating point registers that make up
994 the double width register. By default @samp{--construct-floats} is
995 selected, allowing construction of these floating point constants.
998 @item --emulation=@var{name}
999 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1000 for some other target, in all respects, including output format (choosing
1001 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1002 debugging information or store symbol table information, and default
1003 endianness. The available configuration names are: @samp{mipsecoff},
1004 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1005 @samp{mipsbelf}. The first two do not alter the default endianness from that
1006 of the primary target for which the assembler was configured; the others change
1007 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1008 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1009 selection in any case.
1011 This option is currently supported only when the primary target
1012 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1013 Furthermore, the primary target or others specified with
1014 @samp{--enable-targets=@dots{}} at configuration time must include support for
1015 the other format, if both are to be available. For example, the Irix 5
1016 configuration includes support for both.
1018 Eventually, this option will support more configurations, with more
1019 fine-grained control over the assembler's behavior, and will be supported for
1023 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1030 Control how to deal with multiplication overflow and division by zero.
1031 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1032 (and only work for Instruction Set Architecture level 2 and higher);
1033 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1037 When this option is used, @command{@value{AS}} will issue a warning every
1038 time it generates a nop instruction from a macro.
1043 The following options are available when @value{AS} is configured for
1049 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1050 The command line option @samp{-nojsri2bsr} can be used to disable it.
1054 Enable or disable the silicon filter behaviour. By default this is disabled.
1055 The default can be overridden by the @samp{-sifilter} command line option.
1058 Alter jump instructions for long displacements.
1060 @item -mcpu=[210|340]
1061 Select the cpu type on the target hardware. This controls which instructions
1065 Assemble for a big endian target.
1068 Assemble for a little endian target.
1074 See the info pages for documentation of the MMIX-specific options.
1078 The following options are available when @value{AS} is configured for
1079 an Xtensa processor.
1082 @item --text-section-literals | --no-text-section-literals
1083 With @option{--text-@-section-@-literals}, literal pools are interspersed
1084 in the text section. The default is
1085 @option{--no-@-text-@-section-@-literals}, which places literals in a
1086 separate section in the output file. These options only affect literals
1087 referenced via PC-relative @code{L32R} instructions; literals for
1088 absolute mode @code{L32R} instructions are handled separately.
1090 @item --absolute-literals | --no-absolute-literals
1091 Indicate to the assembler whether @code{L32R} instructions use absolute
1092 or PC-relative addressing. The default is to assume absolute addressing
1093 if the Xtensa processor includes the absolute @code{L32R} addressing
1094 option. Otherwise, only the PC-relative @code{L32R} mode can be used.
1096 @item --target-align | --no-target-align
1097 Enable or disable automatic alignment to reduce branch penalties at the
1098 expense of some code density. The default is @option{--target-@-align}.
1100 @item --longcalls | --no-longcalls
1101 Enable or disable transformation of call instructions to allow calls
1102 across a greater range of addresses. The default is
1103 @option{--no-@-longcalls}.
1105 @item --transform | --no-transform
1106 Enable or disable all assembler transformations of Xtensa instructions.
1107 The default is @option{--transform};
1108 @option{--no-transform} should be used only in the rare cases when the
1109 instructions must be exactly as specified in the assembly source.
1116 * Manual:: Structure of this Manual
1117 * GNU Assembler:: The GNU Assembler
1118 * Object Formats:: Object File Formats
1119 * Command Line:: Command Line
1120 * Input Files:: Input Files
1121 * Object:: Output (Object) File
1122 * Errors:: Error and Warning Messages
1126 @section Structure of this Manual
1128 @cindex manual, structure and purpose
1129 This manual is intended to describe what you need to know to use
1130 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1131 notation for symbols, constants, and expressions; the directives that
1132 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1135 We also cover special features in the @value{TARGET}
1136 configuration of @command{@value{AS}}, including assembler directives.
1139 This manual also describes some of the machine-dependent features of
1140 various flavors of the assembler.
1143 @cindex machine instructions (not covered)
1144 On the other hand, this manual is @emph{not} intended as an introduction
1145 to programming in assembly language---let alone programming in general!
1146 In a similar vein, we make no attempt to introduce the machine
1147 architecture; we do @emph{not} describe the instruction set, standard
1148 mnemonics, registers or addressing modes that are standard to a
1149 particular architecture.
1151 You may want to consult the manufacturer's
1152 machine architecture manual for this information.
1156 For information on the H8/300 machine instruction set, see @cite{H8/300
1157 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1158 Programming Manual} (Renesas).
1161 For information on the H8/500 machine instruction set, see @cite{H8/500
1162 Series Programming Manual} (Renesas M21T001).
1165 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1166 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1167 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1168 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1171 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1175 @c I think this is premature---doc@cygnus.com, 17jan1991
1177 Throughout this manual, we assume that you are running @dfn{GNU},
1178 the portable operating system from the @dfn{Free Software
1179 Foundation, Inc.}. This restricts our attention to certain kinds of
1180 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1181 once this assumption is granted examples and definitions need less
1184 @command{@value{AS}} is part of a team of programs that turn a high-level
1185 human-readable series of instructions into a low-level
1186 computer-readable series of instructions. Different versions of
1187 @command{@value{AS}} are used for different kinds of computer.
1190 @c There used to be a section "Terminology" here, which defined
1191 @c "contents", "byte", "word", and "long". Defining "word" to any
1192 @c particular size is confusing when the .word directive may generate 16
1193 @c bits on one machine and 32 bits on another; in general, for the user
1194 @c version of this manual, none of these terms seem essential to define.
1195 @c They were used very little even in the former draft of the manual;
1196 @c this draft makes an effort to avoid them (except in names of
1200 @section The GNU Assembler
1202 @c man begin DESCRIPTION
1204 @sc{gnu} @command{as} is really a family of assemblers.
1206 This manual describes @command{@value{AS}}, a member of that family which is
1207 configured for the @value{TARGET} architectures.
1209 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1210 should find a fairly similar environment when you use it on another
1211 architecture. Each version has much in common with the others,
1212 including object file formats, most assembler directives (often called
1213 @dfn{pseudo-ops}) and assembler syntax.@refill
1215 @cindex purpose of @sc{gnu} assembler
1216 @command{@value{AS}} is primarily intended to assemble the output of the
1217 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1218 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1219 assemble correctly everything that other assemblers for the same
1220 machine would assemble.
1222 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1225 @c This remark should appear in generic version of manual; assumption
1226 @c here is that generic version sets M680x0.
1227 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1228 assembler for the same architecture; for example, we know of several
1229 incompatible versions of 680x0 assembly language syntax.
1234 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1235 program in one pass of the source file. This has a subtle impact on the
1236 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1238 @node Object Formats
1239 @section Object File Formats
1241 @cindex object file format
1242 The @sc{gnu} assembler can be configured to produce several alternative
1243 object file formats. For the most part, this does not affect how you
1244 write assembly language programs; but directives for debugging symbols
1245 are typically different in different file formats. @xref{Symbol
1246 Attributes,,Symbol Attributes}.
1249 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1250 @value{OBJ-NAME} format object files.
1252 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1254 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1255 @code{a.out} or COFF format object files.
1258 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1259 @code{b.out} or COFF format object files.
1262 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1263 SOM or ELF format object files.
1268 @section Command Line
1270 @cindex command line conventions
1272 After the program name @command{@value{AS}}, the command line may contain
1273 options and file names. Options may appear in any order, and may be
1274 before, after, or between file names. The order of file names is
1277 @cindex standard input, as input file
1279 @file{--} (two hyphens) by itself names the standard input file
1280 explicitly, as one of the files for @command{@value{AS}} to assemble.
1282 @cindex options, command line
1283 Except for @samp{--} any command line argument that begins with a
1284 hyphen (@samp{-}) is an option. Each option changes the behavior of
1285 @command{@value{AS}}. No option changes the way another option works. An
1286 option is a @samp{-} followed by one or more letters; the case of
1287 the letter is important. All options are optional.
1289 Some options expect exactly one file name to follow them. The file
1290 name may either immediately follow the option's letter (compatible
1291 with older assemblers) or it may be the next command argument (@sc{gnu}
1292 standard). These two command lines are equivalent:
1295 @value{AS} -o my-object-file.o mumble.s
1296 @value{AS} -omy-object-file.o mumble.s
1300 @section Input Files
1303 @cindex source program
1304 @cindex files, input
1305 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1306 describe the program input to one run of @command{@value{AS}}. The program may
1307 be in one or more files; how the source is partitioned into files
1308 doesn't change the meaning of the source.
1310 @c I added "con" prefix to "catenation" just to prove I can overcome my
1311 @c APL training... doc@cygnus.com
1312 The source program is a concatenation of the text in all the files, in the
1315 @c man begin DESCRIPTION
1316 Each time you run @command{@value{AS}} it assembles exactly one source
1317 program. The source program is made up of one or more files.
1318 (The standard input is also a file.)
1320 You give @command{@value{AS}} a command line that has zero or more input file
1321 names. The input files are read (from left file name to right). A
1322 command line argument (in any position) that has no special meaning
1323 is taken to be an input file name.
1325 If you give @command{@value{AS}} no file names it attempts to read one input file
1326 from the @command{@value{AS}} standard input, which is normally your terminal. You
1327 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1330 Use @samp{--} if you need to explicitly name the standard input file
1331 in your command line.
1333 If the source is empty, @command{@value{AS}} produces a small, empty object
1338 @subheading Filenames and Line-numbers
1340 @cindex input file linenumbers
1341 @cindex line numbers, in input files
1342 There are two ways of locating a line in the input file (or files) and
1343 either may be used in reporting error messages. One way refers to a line
1344 number in a physical file; the other refers to a line number in a
1345 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1347 @dfn{Physical files} are those files named in the command line given
1348 to @command{@value{AS}}.
1350 @dfn{Logical files} are simply names declared explicitly by assembler
1351 directives; they bear no relation to physical files. Logical file names help
1352 error messages reflect the original source file, when @command{@value{AS}} source
1353 is itself synthesized from other files. @command{@value{AS}} understands the
1354 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1355 @ref{File,,@code{.file}}.
1358 @section Output (Object) File
1364 Every time you run @command{@value{AS}} it produces an output file, which is
1365 your assembly language program translated into numbers. This file
1366 is the object file. Its default name is
1374 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1376 You can give it another name by using the @option{-o} option. Conventionally,
1377 object file names end with @file{.o}. The default name is used for historical
1378 reasons: older assemblers were capable of assembling self-contained programs
1379 directly into a runnable program. (For some formats, this isn't currently
1380 possible, but it can be done for the @code{a.out} format.)
1384 The object file is meant for input to the linker @code{@value{LD}}. It contains
1385 assembled program code, information to help @code{@value{LD}} integrate
1386 the assembled program into a runnable file, and (optionally) symbolic
1387 information for the debugger.
1389 @c link above to some info file(s) like the description of a.out.
1390 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1393 @section Error and Warning Messages
1395 @c man begin DESCRIPTION
1397 @cindex error messages
1398 @cindex warning messages
1399 @cindex messages from assembler
1400 @command{@value{AS}} may write warnings and error messages to the standard error
1401 file (usually your terminal). This should not happen when a compiler
1402 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1403 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1404 grave problem that stops the assembly.
1408 @cindex format of warning messages
1409 Warning messages have the format
1412 file_name:@b{NNN}:Warning Message Text
1416 @cindex line numbers, in warnings/errors
1417 (where @b{NNN} is a line number). If a logical file name has been given
1418 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1419 the current input file is used. If a logical line number was given
1421 (@pxref{Line,,@code{.line}})
1425 (@pxref{Line,,@code{.line}})
1428 (@pxref{Ln,,@code{.ln}})
1431 then it is used to calculate the number printed,
1432 otherwise the actual line in the current source file is printed. The
1433 message text is intended to be self explanatory (in the grand Unix
1436 @cindex format of error messages
1437 Error messages have the format
1439 file_name:@b{NNN}:FATAL:Error Message Text
1441 The file name and line number are derived as for warning
1442 messages. The actual message text may be rather less explanatory
1443 because many of them aren't supposed to happen.
1446 @chapter Command-Line Options
1448 @cindex options, all versions of assembler
1449 This chapter describes command-line options available in @emph{all}
1450 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
1452 to the @value{TARGET} target.
1455 to particular machine architectures.
1458 @c man begin DESCRIPTION
1460 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1461 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1462 The assembler arguments must be separated from each other (and the @samp{-Wa})
1463 by commas. For example:
1466 gcc -c -g -O -Wa,-alh,-L file.c
1470 This passes two options to the assembler: @samp{-alh} (emit a listing to
1471 standard output with high-level and assembly source) and @samp{-L} (retain
1472 local symbols in the symbol table).
1474 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1475 command-line options are automatically passed to the assembler by the compiler.
1476 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1477 precisely what options it passes to each compilation pass, including the
1483 * a:: -a[cdhlns] enable listings
1484 * alternate:: --alternate enable alternate macro syntax
1485 * D:: -D for compatibility
1486 * f:: -f to work faster
1487 * I:: -I for .include search path
1488 @ifclear DIFF-TBL-KLUGE
1489 * K:: -K for compatibility
1491 @ifset DIFF-TBL-KLUGE
1492 * K:: -K for difference tables
1495 * L:: -L to retain local labels
1496 * listing:: --listing-XXX to configure listing output
1497 * M:: -M or --mri to assemble in MRI compatibility mode
1498 * MD:: --MD for dependency tracking
1499 * o:: -o to name the object file
1500 * R:: -R to join data and text sections
1501 * statistics:: --statistics to see statistics about assembly
1502 * traditional-format:: --traditional-format for compatible output
1503 * v:: -v to announce version
1504 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1505 * Z:: -Z to make object file even after errors
1509 @section Enable Listings: @option{-a[cdhlns]}
1518 @cindex listings, enabling
1519 @cindex assembly listings, enabling
1521 These options enable listing output from the assembler. By itself,
1522 @samp{-a} requests high-level, assembly, and symbols listing.
1523 You can use other letters to select specific options for the list:
1524 @samp{-ah} requests a high-level language listing,
1525 @samp{-al} requests an output-program assembly listing, and
1526 @samp{-as} requests a symbol table listing.
1527 High-level listings require that a compiler debugging option like
1528 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1531 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1532 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1533 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1534 omitted from the listing.
1536 Use the @samp{-ad} option to omit debugging directives from the
1539 Once you have specified one of these options, you can further control
1540 listing output and its appearance using the directives @code{.list},
1541 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1543 The @samp{-an} option turns off all forms processing.
1544 If you do not request listing output with one of the @samp{-a} options, the
1545 listing-control directives have no effect.
1547 The letters after @samp{-a} may be combined into one option,
1548 @emph{e.g.}, @samp{-aln}.
1550 Note if the assembler source is coming from the standard input (eg because it
1551 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1552 is being used) then the listing will not contain any comments or preprocessor
1553 directives. This is because the listing code buffers input source lines from
1554 stdin only after they have been preprocessed by the assembler. This reduces
1555 memory usage and makes the code more efficient.
1558 @section @option{--alternate}
1561 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1564 @section @option{-D}
1567 This option has no effect whatsoever, but it is accepted to make it more
1568 likely that scripts written for other assemblers also work with
1569 @command{@value{AS}}.
1572 @section Work Faster: @option{-f}
1575 @cindex trusted compiler
1576 @cindex faster processing (@option{-f})
1577 @samp{-f} should only be used when assembling programs written by a
1578 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1579 and comment preprocessing on
1580 the input file(s) before assembling them. @xref{Preprocessing,
1584 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1585 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1590 @section @code{.include} Search Path: @option{-I} @var{path}
1592 @kindex -I @var{path}
1593 @cindex paths for @code{.include}
1594 @cindex search path for @code{.include}
1595 @cindex @code{include} directive search path
1596 Use this option to add a @var{path} to the list of directories
1597 @command{@value{AS}} searches for files specified in @code{.include}
1598 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1599 many times as necessary to include a variety of paths. The current
1600 working directory is always searched first; after that, @command{@value{AS}}
1601 searches any @samp{-I} directories in the same order as they were
1602 specified (left to right) on the command line.
1605 @section Difference Tables: @option{-K}
1608 @ifclear DIFF-TBL-KLUGE
1609 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1610 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1611 where it can be used to warn when the assembler alters the machine code
1612 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1613 family does not have the addressing limitations that sometimes lead to this
1614 alteration on other platforms.
1617 @ifset DIFF-TBL-KLUGE
1618 @cindex difference tables, warning
1619 @cindex warning for altered difference tables
1620 @command{@value{AS}} sometimes alters the code emitted for directives of the form
1621 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1622 You can use the @samp{-K} option if you want a warning issued when this
1627 @section Include Local Labels: @option{-L}
1630 @cindex local labels, retaining in output
1631 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1632 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1633 debugging, because they are intended for the use of programs (like
1634 compilers) that compose assembler programs, not for your notice.
1635 Normally both @command{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1636 normally debug with them.
1638 This option tells @command{@value{AS}} to retain those @samp{L@dots{}} symbols
1639 in the object file. Usually if you do this you also tell the linker
1640 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1642 By default, a local label is any label beginning with @samp{L}, but each
1643 target is allowed to redefine the local label prefix.
1645 On the HPPA local labels begin with @samp{L$}.
1649 @section Configuring listing output: @option{--listing}
1651 The listing feature of the assembler can be enabled via the command line switch
1652 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1653 hex dump of the corresponding locations in the output object file, and displays
1654 them as a listing file. The format of this listing can be controlled by pseudo
1655 ops inside the assembler source (@pxref{List} @pxref{Title} @pxref{Sbttl}
1656 @pxref{Psize} @pxref{Eject}) and also by the following switches:
1659 @item --listing-lhs-width=@samp{number}
1660 @kindex --listing-lhs-width
1661 @cindex Width of first line disassembly output
1662 Sets the maximum width, in words, of the first line of the hex byte dump. This
1663 dump appears on the left hand side of the listing output.
1665 @item --listing-lhs-width2=@samp{number}
1666 @kindex --listing-lhs-width2
1667 @cindex Width of continuation lines of disassembly output
1668 Sets the maximum width, in words, of any further lines of the hex byte dump for
1669 a given input source line. If this value is not specified, it defaults to being
1670 the same as the value specified for @samp{--listing-lhs-width}. If neither
1671 switch is used the default is to one.
1673 @item --listing-rhs-width=@samp{number}
1674 @kindex --listing-rhs-width
1675 @cindex Width of source line output
1676 Sets the maximum width, in characters, of the source line that is displayed
1677 alongside the hex dump. The default value for this parameter is 100. The
1678 source line is displayed on the right hand side of the listing output.
1680 @item --listing-cont-lines=@samp{number}
1681 @kindex --listing-cont-lines
1682 @cindex Maximum number of continuation lines
1683 Sets the maximum number of continuation lines of hex dump that will be
1684 displayed for a given single line of source input. The default value is 4.
1688 @section Assemble in MRI Compatibility Mode: @option{-M}
1691 @cindex MRI compatibility mode
1692 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1693 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1694 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1695 configured target) assembler from Microtec Research. The exact nature of the
1696 MRI syntax will not be documented here; see the MRI manuals for more
1697 information. Note in particular that the handling of macros and macro
1698 arguments is somewhat different. The purpose of this option is to permit
1699 assembling existing MRI assembler code using @command{@value{AS}}.
1701 The MRI compatibility is not complete. Certain operations of the MRI assembler
1702 depend upon its object file format, and can not be supported using other object
1703 file formats. Supporting these would require enhancing each object file format
1704 individually. These are:
1707 @item global symbols in common section
1709 The m68k MRI assembler supports common sections which are merged by the linker.
1710 Other object file formats do not support this. @command{@value{AS}} handles
1711 common sections by treating them as a single common symbol. It permits local
1712 symbols to be defined within a common section, but it can not support global
1713 symbols, since it has no way to describe them.
1715 @item complex relocations
1717 The MRI assemblers support relocations against a negated section address, and
1718 relocations which combine the start addresses of two or more sections. These
1719 are not support by other object file formats.
1721 @item @code{END} pseudo-op specifying start address
1723 The MRI @code{END} pseudo-op permits the specification of a start address.
1724 This is not supported by other object file formats. The start address may
1725 instead be specified using the @option{-e} option to the linker, or in a linker
1728 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1730 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1731 name to the output file. This is not supported by other object file formats.
1733 @item @code{ORG} pseudo-op
1735 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1736 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1737 which changes the location within the current section. Absolute sections are
1738 not supported by other object file formats. The address of a section may be
1739 assigned within a linker script.
1742 There are some other features of the MRI assembler which are not supported by
1743 @command{@value{AS}}, typically either because they are difficult or because they
1744 seem of little consequence. Some of these may be supported in future releases.
1748 @item EBCDIC strings
1750 EBCDIC strings are not supported.
1752 @item packed binary coded decimal
1754 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1755 and @code{DCB.P} pseudo-ops are not supported.
1757 @item @code{FEQU} pseudo-op
1759 The m68k @code{FEQU} pseudo-op is not supported.
1761 @item @code{NOOBJ} pseudo-op
1763 The m68k @code{NOOBJ} pseudo-op is not supported.
1765 @item @code{OPT} branch control options
1767 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1768 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1769 relaxes all branches, whether forward or backward, to an appropriate size, so
1770 these options serve no purpose.
1772 @item @code{OPT} list control options
1774 The following m68k @code{OPT} list control options are ignored: @code{C},
1775 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1776 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1778 @item other @code{OPT} options
1780 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1781 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1783 @item @code{OPT} @code{D} option is default
1785 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1786 @code{OPT NOD} may be used to turn it off.
1788 @item @code{XREF} pseudo-op.
1790 The m68k @code{XREF} pseudo-op is ignored.
1792 @item @code{.debug} pseudo-op
1794 The i960 @code{.debug} pseudo-op is not supported.
1796 @item @code{.extended} pseudo-op
1798 The i960 @code{.extended} pseudo-op is not supported.
1800 @item @code{.list} pseudo-op.
1802 The various options of the i960 @code{.list} pseudo-op are not supported.
1804 @item @code{.optimize} pseudo-op
1806 The i960 @code{.optimize} pseudo-op is not supported.
1808 @item @code{.output} pseudo-op
1810 The i960 @code{.output} pseudo-op is not supported.
1812 @item @code{.setreal} pseudo-op
1814 The i960 @code{.setreal} pseudo-op is not supported.
1819 @section Dependency Tracking: @option{--MD}
1822 @cindex dependency tracking
1825 @command{@value{AS}} can generate a dependency file for the file it creates. This
1826 file consists of a single rule suitable for @code{make} describing the
1827 dependencies of the main source file.
1829 The rule is written to the file named in its argument.
1831 This feature is used in the automatic updating of makefiles.
1834 @section Name the Object File: @option{-o}
1837 @cindex naming object file
1838 @cindex object file name
1839 There is always one object file output when you run @command{@value{AS}}. By
1840 default it has the name
1843 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1857 You use this option (which takes exactly one filename) to give the
1858 object file a different name.
1860 Whatever the object file is called, @command{@value{AS}} overwrites any
1861 existing file of the same name.
1864 @section Join Data and Text Sections: @option{-R}
1867 @cindex data and text sections, joining
1868 @cindex text and data sections, joining
1869 @cindex joining text and data sections
1870 @cindex merging text and data sections
1871 @option{-R} tells @command{@value{AS}} to write the object file as if all
1872 data-section data lives in the text section. This is only done at
1873 the very last moment: your binary data are the same, but data
1874 section parts are relocated differently. The data section part of
1875 your object file is zero bytes long because all its bytes are
1876 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1878 When you specify @option{-R} it would be possible to generate shorter
1879 address displacements (because we do not have to cross between text and
1880 data section). We refrain from doing this simply for compatibility with
1881 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
1884 When @command{@value{AS}} is configured for COFF or ELF output,
1885 this option is only useful if you use sections named @samp{.text} and
1890 @option{-R} is not supported for any of the HPPA targets. Using
1891 @option{-R} generates a warning from @command{@value{AS}}.
1895 @section Display Assembly Statistics: @option{--statistics}
1897 @kindex --statistics
1898 @cindex statistics, about assembly
1899 @cindex time, total for assembly
1900 @cindex space used, maximum for assembly
1901 Use @samp{--statistics} to display two statistics about the resources used by
1902 @command{@value{AS}}: the maximum amount of space allocated during the assembly
1903 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1906 @node traditional-format
1907 @section Compatible Output: @option{--traditional-format}
1909 @kindex --traditional-format
1910 For some targets, the output of @command{@value{AS}} is different in some ways
1911 from the output of some existing assembler. This switch requests
1912 @command{@value{AS}} to use the traditional format instead.
1914 For example, it disables the exception frame optimizations which
1915 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
1918 @section Announce Version: @option{-v}
1922 @cindex assembler version
1923 @cindex version of assembler
1924 You can find out what version of as is running by including the
1925 option @samp{-v} (which you can also spell as @samp{-version}) on the
1929 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
1931 @command{@value{AS}} should never give a warning or error message when
1932 assembling compiler output. But programs written by people often
1933 cause @command{@value{AS}} to give a warning that a particular assumption was
1934 made. All such warnings are directed to the standard error file.
1938 @cindex suppressing warnings
1939 @cindex warnings, suppressing
1940 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
1941 This only affects the warning messages: it does not change any particular of
1942 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
1945 @kindex --fatal-warnings
1946 @cindex errors, caused by warnings
1947 @cindex warnings, causing error
1948 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
1949 files that generate warnings to be in error.
1952 @cindex warnings, switching on
1953 You can switch these options off again by specifying @option{--warn}, which
1954 causes warnings to be output as usual.
1957 @section Generate Object File in Spite of Errors: @option{-Z}
1958 @cindex object file, after errors
1959 @cindex errors, continuing after
1960 After an error message, @command{@value{AS}} normally produces no output. If for
1961 some reason you are interested in object file output even after
1962 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
1963 option. If there are any errors, @command{@value{AS}} continues anyways, and
1964 writes an object file after a final warning message of the form @samp{@var{n}
1965 errors, @var{m} warnings, generating bad object file.}
1970 @cindex machine-independent syntax
1971 @cindex syntax, machine-independent
1972 This chapter describes the machine-independent syntax allowed in a
1973 source file. @command{@value{AS}} syntax is similar to what many other
1974 assemblers use; it is inspired by the BSD 4.2
1979 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
1983 * Preprocessing:: Preprocessing
1984 * Whitespace:: Whitespace
1985 * Comments:: Comments
1986 * Symbol Intro:: Symbols
1987 * Statements:: Statements
1988 * Constants:: Constants
1992 @section Preprocessing
1994 @cindex preprocessing
1995 The @command{@value{AS}} internal preprocessor:
1997 @cindex whitespace, removed by preprocessor
1999 adjusts and removes extra whitespace. It leaves one space or tab before
2000 the keywords on a line, and turns any other whitespace on the line into
2003 @cindex comments, removed by preprocessor
2005 removes all comments, replacing them with a single space, or an
2006 appropriate number of newlines.
2008 @cindex constants, converted by preprocessor
2010 converts character constants into the appropriate numeric values.
2013 It does not do macro processing, include file handling, or
2014 anything else you may get from your C compiler's preprocessor. You can
2015 do include file processing with the @code{.include} directive
2016 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2017 to get other ``CPP'' style preprocessing by giving the input file a
2018 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
2019 Output, gcc.info, Using GNU CC}.
2021 Excess whitespace, comments, and character constants
2022 cannot be used in the portions of the input text that are not
2025 @cindex turning preprocessing on and off
2026 @cindex preprocessing, turning on and off
2029 If the first line of an input file is @code{#NO_APP} or if you use the
2030 @samp{-f} option, whitespace and comments are not removed from the input file.
2031 Within an input file, you can ask for whitespace and comment removal in
2032 specific portions of the by putting a line that says @code{#APP} before the
2033 text that may contain whitespace or comments, and putting a line that says
2034 @code{#NO_APP} after this text. This feature is mainly intend to support
2035 @code{asm} statements in compilers whose output is otherwise free of comments
2042 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2043 Whitespace is used to separate symbols, and to make programs neater for
2044 people to read. Unless within character constants
2045 (@pxref{Characters,,Character Constants}), any whitespace means the same
2046 as exactly one space.
2052 There are two ways of rendering comments to @command{@value{AS}}. In both
2053 cases the comment is equivalent to one space.
2055 Anything from @samp{/*} through the next @samp{*/} is a comment.
2056 This means you may not nest these comments.
2060 The only way to include a newline ('\n') in a comment
2061 is to use this sort of comment.
2064 /* This sort of comment does not nest. */
2067 @cindex line comment character
2068 Anything from the @dfn{line comment} character to the next newline
2069 is considered a comment and is ignored. The line comment character is
2071 @samp{;} for the AMD 29K family;
2074 @samp{;} on the ARC;
2077 @samp{@@} on the ARM;
2080 @samp{;} for the H8/300 family;
2083 @samp{!} for the H8/500 family;
2086 @samp{;} for the HPPA;
2089 @samp{#} on the i386 and x86-64;
2092 @samp{#} on the i960;
2095 @samp{;} for the PDP-11;
2098 @samp{;} for picoJava;
2101 @samp{#} for Motorola PowerPC;
2104 @samp{!} for the Renesas / SuperH SH;
2107 @samp{!} on the SPARC;
2110 @samp{#} on the ip2k;
2113 @samp{#} on the m32r;
2116 @samp{|} on the 680x0;
2119 @samp{#} on the 68HC11 and 68HC12;
2122 @samp{;} on the M880x0;
2125 @samp{#} on the Vax;
2128 @samp{!} for the Z8000;
2131 @samp{#} on the V850;
2134 @samp{#} for Xtensa systems;
2136 see @ref{Machine Dependencies}. @refill
2137 @c FIXME What about i860?
2140 On some machines there are two different line comment characters. One
2141 character only begins a comment if it is the first non-whitespace character on
2142 a line, while the other always begins a comment.
2146 The V850 assembler also supports a double dash as starting a comment that
2147 extends to the end of the line.
2153 @cindex lines starting with @code{#}
2154 @cindex logical line numbers
2155 To be compatible with past assemblers, lines that begin with @samp{#} have a
2156 special interpretation. Following the @samp{#} should be an absolute
2157 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2158 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
2159 new logical file name. The rest of the line, if any, should be whitespace.
2161 If the first non-whitespace characters on the line are not numeric,
2162 the line is ignored. (Just like a comment.)
2165 # This is an ordinary comment.
2166 # 42-6 "new_file_name" # New logical file name
2167 # This is logical line # 36.
2169 This feature is deprecated, and may disappear from future versions
2170 of @command{@value{AS}}.
2175 @cindex characters used in symbols
2176 @ifclear SPECIAL-SYMS
2177 A @dfn{symbol} is one or more characters chosen from the set of all
2178 letters (both upper and lower case), digits and the three characters
2184 A @dfn{symbol} is one or more characters chosen from the set of all
2185 letters (both upper and lower case), digits and the three characters
2186 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2192 On most machines, you can also use @code{$} in symbol names; exceptions
2193 are noted in @ref{Machine Dependencies}.
2195 No symbol may begin with a digit. Case is significant.
2196 There is no length limit: all characters are significant. Symbols are
2197 delimited by characters not in that set, or by the beginning of a file
2198 (since the source program must end with a newline, the end of a file is
2199 not a possible symbol delimiter). @xref{Symbols}.
2200 @cindex length of symbols
2205 @cindex statements, structure of
2206 @cindex line separator character
2207 @cindex statement separator character
2209 @ifclear abnormal-separator
2210 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2211 semicolon (@samp{;}). The newline or semicolon is considered part of
2212 the preceding statement. Newlines and semicolons within character
2213 constants are an exception: they do not end statements.
2215 @ifset abnormal-separator
2217 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
2218 sign (@samp{@@}). The newline or at sign is considered part of the
2219 preceding statement. Newlines and at signs within character constants
2220 are an exception: they do not end statements.
2223 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2224 point (@samp{!}). The newline or exclamation point is considered part of the
2225 preceding statement. Newlines and exclamation points within character
2226 constants are an exception: they do not end statements.
2229 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2230 H8/300) a dollar sign (@samp{$}); or (for the
2233 (@samp{;}). The newline or separator character is considered part of
2234 the preceding statement. Newlines and separators within character
2235 constants are an exception: they do not end statements.
2240 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2241 separator character. (The line separator is usually @samp{;}, unless
2242 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
2243 newline or separator character is considered part of the preceding
2244 statement. Newlines and separators within character constants are an
2245 exception: they do not end statements.
2248 @cindex newline, required at file end
2249 @cindex EOF, newline must precede
2250 It is an error to end any statement with end-of-file: the last
2251 character of any input file should be a newline.@refill
2253 An empty statement is allowed, and may include whitespace. It is ignored.
2255 @cindex instructions and directives
2256 @cindex directives and instructions
2257 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2258 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2260 A statement begins with zero or more labels, optionally followed by a
2261 key symbol which determines what kind of statement it is. The key
2262 symbol determines the syntax of the rest of the statement. If the
2263 symbol begins with a dot @samp{.} then the statement is an assembler
2264 directive: typically valid for any computer. If the symbol begins with
2265 a letter the statement is an assembly language @dfn{instruction}: it
2266 assembles into a machine language instruction.
2268 Different versions of @command{@value{AS}} for different computers
2269 recognize different instructions. In fact, the same symbol may
2270 represent a different instruction in a different computer's assembly
2274 @cindex @code{:} (label)
2275 @cindex label (@code{:})
2276 A label is a symbol immediately followed by a colon (@code{:}).
2277 Whitespace before a label or after a colon is permitted, but you may not
2278 have whitespace between a label's symbol and its colon. @xref{Labels}.
2281 For HPPA targets, labels need not be immediately followed by a colon, but
2282 the definition of a label must begin in column zero. This also implies that
2283 only one label may be defined on each line.
2287 label: .directive followed by something
2288 another_label: # This is an empty statement.
2289 instruction operand_1, operand_2, @dots{}
2296 A constant is a number, written so that its value is known by
2297 inspection, without knowing any context. Like this:
2300 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2301 .ascii "Ring the bell\7" # A string constant.
2302 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2303 .float 0f-314159265358979323846264338327\
2304 95028841971.693993751E-40 # - pi, a flonum.
2309 * Characters:: Character Constants
2310 * Numbers:: Number Constants
2314 @subsection Character Constants
2316 @cindex character constants
2317 @cindex constants, character
2318 There are two kinds of character constants. A @dfn{character} stands
2319 for one character in one byte and its value may be used in
2320 numeric expressions. String constants (properly called string
2321 @emph{literals}) are potentially many bytes and their values may not be
2322 used in arithmetic expressions.
2326 * Chars:: Characters
2330 @subsubsection Strings
2332 @cindex string constants
2333 @cindex constants, string
2334 A @dfn{string} is written between double-quotes. It may contain
2335 double-quotes or null characters. The way to get special characters
2336 into a string is to @dfn{escape} these characters: precede them with
2337 a backslash @samp{\} character. For example @samp{\\} represents
2338 one backslash: the first @code{\} is an escape which tells
2339 @command{@value{AS}} to interpret the second character literally as a backslash
2340 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2341 escape character). The complete list of escapes follows.
2343 @cindex escape codes, character
2344 @cindex character escape codes
2347 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2349 @cindex @code{\b} (backspace character)
2350 @cindex backspace (@code{\b})
2352 Mnemonic for backspace; for ASCII this is octal code 010.
2355 @c Mnemonic for EOText; for ASCII this is octal code 004.
2357 @cindex @code{\f} (formfeed character)
2358 @cindex formfeed (@code{\f})
2360 Mnemonic for FormFeed; for ASCII this is octal code 014.
2362 @cindex @code{\n} (newline character)
2363 @cindex newline (@code{\n})
2365 Mnemonic for newline; for ASCII this is octal code 012.
2368 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2370 @cindex @code{\r} (carriage return character)
2371 @cindex carriage return (@code{\r})
2373 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2376 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2377 @c other assemblers.
2379 @cindex @code{\t} (tab)
2380 @cindex tab (@code{\t})
2382 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2385 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2386 @c @item \x @var{digit} @var{digit} @var{digit}
2387 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2389 @cindex @code{\@var{ddd}} (octal character code)
2390 @cindex octal character code (@code{\@var{ddd}})
2391 @item \ @var{digit} @var{digit} @var{digit}
2392 An octal character code. The numeric code is 3 octal digits.
2393 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2394 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2396 @cindex @code{\@var{xd...}} (hex character code)
2397 @cindex hex character code (@code{\@var{xd...}})
2398 @item \@code{x} @var{hex-digits...}
2399 A hex character code. All trailing hex digits are combined. Either upper or
2400 lower case @code{x} works.
2402 @cindex @code{\\} (@samp{\} character)
2403 @cindex backslash (@code{\\})
2405 Represents one @samp{\} character.
2408 @c Represents one @samp{'} (accent acute) character.
2409 @c This is needed in single character literals
2410 @c (@xref{Characters,,Character Constants}.) to represent
2413 @cindex @code{\"} (doublequote character)
2414 @cindex doublequote (@code{\"})
2416 Represents one @samp{"} character. Needed in strings to represent
2417 this character, because an unescaped @samp{"} would end the string.
2419 @item \ @var{anything-else}
2420 Any other character when escaped by @kbd{\} gives a warning, but
2421 assembles as if the @samp{\} was not present. The idea is that if
2422 you used an escape sequence you clearly didn't want the literal
2423 interpretation of the following character. However @command{@value{AS}} has no
2424 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2425 code and warns you of the fact.
2428 Which characters are escapable, and what those escapes represent,
2429 varies widely among assemblers. The current set is what we think
2430 the BSD 4.2 assembler recognizes, and is a subset of what most C
2431 compilers recognize. If you are in doubt, do not use an escape
2435 @subsubsection Characters
2437 @cindex single character constant
2438 @cindex character, single
2439 @cindex constant, single character
2440 A single character may be written as a single quote immediately
2441 followed by that character. The same escapes apply to characters as
2442 to strings. So if you want to write the character backslash, you
2443 must write @kbd{'\\} where the first @code{\} escapes the second
2444 @code{\}. As you can see, the quote is an acute accent, not a
2445 grave accent. A newline
2447 @ifclear abnormal-separator
2448 (or semicolon @samp{;})
2450 @ifset abnormal-separator
2452 (or at sign @samp{@@})
2455 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2456 Renesas SH or H8/500)
2460 immediately following an acute accent is taken as a literal character
2461 and does not count as the end of a statement. The value of a character
2462 constant in a numeric expression is the machine's byte-wide code for
2463 that character. @command{@value{AS}} assumes your character code is ASCII:
2464 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2467 @subsection Number Constants
2469 @cindex constants, number
2470 @cindex number constants
2471 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2472 are stored in the target machine. @emph{Integers} are numbers that
2473 would fit into an @code{int} in the C language. @emph{Bignums} are
2474 integers, but they are stored in more than 32 bits. @emph{Flonums}
2475 are floating point numbers, described below.
2478 * Integers:: Integers
2483 * Bit Fields:: Bit Fields
2489 @subsubsection Integers
2491 @cindex constants, integer
2493 @cindex binary integers
2494 @cindex integers, binary
2495 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2496 the binary digits @samp{01}.
2498 @cindex octal integers
2499 @cindex integers, octal
2500 An octal integer is @samp{0} followed by zero or more of the octal
2501 digits (@samp{01234567}).
2503 @cindex decimal integers
2504 @cindex integers, decimal
2505 A decimal integer starts with a non-zero digit followed by zero or
2506 more digits (@samp{0123456789}).
2508 @cindex hexadecimal integers
2509 @cindex integers, hexadecimal
2510 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2511 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2513 Integers have the usual values. To denote a negative integer, use
2514 the prefix operator @samp{-} discussed under expressions
2515 (@pxref{Prefix Ops,,Prefix Operators}).
2518 @subsubsection Bignums
2521 @cindex constants, bignum
2522 A @dfn{bignum} has the same syntax and semantics as an integer
2523 except that the number (or its negative) takes more than 32 bits to
2524 represent in binary. The distinction is made because in some places
2525 integers are permitted while bignums are not.
2528 @subsubsection Flonums
2530 @cindex floating point numbers
2531 @cindex constants, floating point
2533 @cindex precision, floating point
2534 A @dfn{flonum} represents a floating point number. The translation is
2535 indirect: a decimal floating point number from the text is converted by
2536 @command{@value{AS}} to a generic binary floating point number of more than
2537 sufficient precision. This generic floating point number is converted
2538 to a particular computer's floating point format (or formats) by a
2539 portion of @command{@value{AS}} specialized to that computer.
2541 A flonum is written by writing (in order)
2546 (@samp{0} is optional on the HPPA.)
2550 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2552 @kbd{e} is recommended. Case is not important.
2554 @c FIXME: verify if flonum syntax really this vague for most cases
2555 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2556 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2559 On the H8/300, H8/500,
2560 Renesas / SuperH SH,
2561 and AMD 29K architectures, the letter must be
2562 one of the letters @samp{DFPRSX} (in upper or lower case).
2564 On the ARC, the letter must be one of the letters @samp{DFRS}
2565 (in upper or lower case).
2567 On the Intel 960 architecture, the letter must be
2568 one of the letters @samp{DFT} (in upper or lower case).
2570 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2574 One of the letters @samp{DFPRSX} (in upper or lower case).
2577 One of the letters @samp{DFRS} (in upper or lower case).
2580 One of the letters @samp{DFPRSX} (in upper or lower case).
2583 The letter @samp{E} (upper case only).
2586 One of the letters @samp{DFT} (in upper or lower case).
2591 An optional sign: either @samp{+} or @samp{-}.
2594 An optional @dfn{integer part}: zero or more decimal digits.
2597 An optional @dfn{fractional part}: @samp{.} followed by zero
2598 or more decimal digits.
2601 An optional exponent, consisting of:
2605 An @samp{E} or @samp{e}.
2606 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2607 @c principle this can perfectly well be different on different targets.
2609 Optional sign: either @samp{+} or @samp{-}.
2611 One or more decimal digits.
2616 At least one of the integer part or the fractional part must be
2617 present. The floating point number has the usual base-10 value.
2619 @command{@value{AS}} does all processing using integers. Flonums are computed
2620 independently of any floating point hardware in the computer running
2621 @command{@value{AS}}.
2625 @c Bit fields are written as a general facility but are also controlled
2626 @c by a conditional-compilation flag---which is as of now (21mar91)
2627 @c turned on only by the i960 config of GAS.
2629 @subsubsection Bit Fields
2632 @cindex constants, bit field
2633 You can also define numeric constants as @dfn{bit fields}.
2634 specify two numbers separated by a colon---
2636 @var{mask}:@var{value}
2639 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2642 The resulting number is then packed
2644 @c this conditional paren in case bit fields turned on elsewhere than 960
2645 (in host-dependent byte order)
2647 into a field whose width depends on which assembler directive has the
2648 bit-field as its argument. Overflow (a result from the bitwise and
2649 requiring more binary digits to represent) is not an error; instead,
2650 more constants are generated, of the specified width, beginning with the
2651 least significant digits.@refill
2653 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2654 @code{.short}, and @code{.word} accept bit-field arguments.
2659 @chapter Sections and Relocation
2664 * Secs Background:: Background
2665 * Ld Sections:: Linker Sections
2666 * As Sections:: Assembler Internal Sections
2667 * Sub-Sections:: Sub-Sections
2671 @node Secs Background
2674 Roughly, a section is a range of addresses, with no gaps; all data
2675 ``in'' those addresses is treated the same for some particular purpose.
2676 For example there may be a ``read only'' section.
2678 @cindex linker, and assembler
2679 @cindex assembler, and linker
2680 The linker @code{@value{LD}} reads many object files (partial programs) and
2681 combines their contents to form a runnable program. When @command{@value{AS}}
2682 emits an object file, the partial program is assumed to start at address 0.
2683 @code{@value{LD}} assigns the final addresses for the partial program, so that
2684 different partial programs do not overlap. This is actually an
2685 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2688 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2689 addresses. These blocks slide to their run-time addresses as rigid
2690 units; their length does not change and neither does the order of bytes
2691 within them. Such a rigid unit is called a @emph{section}. Assigning
2692 run-time addresses to sections is called @dfn{relocation}. It includes
2693 the task of adjusting mentions of object-file addresses so they refer to
2694 the proper run-time addresses.
2696 For the H8/300 and H8/500,
2697 and for the Renesas / SuperH SH,
2698 @command{@value{AS}} pads sections if needed to
2699 ensure they end on a word (sixteen bit) boundary.
2702 @cindex standard assembler sections
2703 An object file written by @command{@value{AS}} has at least three sections, any
2704 of which may be empty. These are named @dfn{text}, @dfn{data} and
2709 When it generates COFF or ELF output,
2711 @command{@value{AS}} can also generate whatever other named sections you specify
2712 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2713 If you do not use any directives that place output in the @samp{.text}
2714 or @samp{.data} sections, these sections still exist, but are empty.
2719 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2721 @command{@value{AS}} can also generate whatever other named sections you
2722 specify using the @samp{.space} and @samp{.subspace} directives. See
2723 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2724 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2725 assembler directives.
2728 Additionally, @command{@value{AS}} uses different names for the standard
2729 text, data, and bss sections when generating SOM output. Program text
2730 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2731 BSS into @samp{$BSS$}.
2735 Within the object file, the text section starts at address @code{0}, the
2736 data section follows, and the bss section follows the data section.
2739 When generating either SOM or ELF output files on the HPPA, the text
2740 section starts at address @code{0}, the data section at address
2741 @code{0x4000000}, and the bss section follows the data section.
2744 To let @code{@value{LD}} know which data changes when the sections are
2745 relocated, and how to change that data, @command{@value{AS}} also writes to the
2746 object file details of the relocation needed. To perform relocation
2747 @code{@value{LD}} must know, each time an address in the object
2751 Where in the object file is the beginning of this reference to
2754 How long (in bytes) is this reference?
2756 Which section does the address refer to? What is the numeric value of
2758 (@var{address}) @minus{} (@var{start-address of section})?
2761 Is the reference to an address ``Program-Counter relative''?
2764 @cindex addresses, format of
2765 @cindex section-relative addressing
2766 In fact, every address @command{@value{AS}} ever uses is expressed as
2768 (@var{section}) + (@var{offset into section})
2771 Further, most expressions @command{@value{AS}} computes have this section-relative
2774 (For some object formats, such as SOM for the HPPA, some expressions are
2775 symbol-relative instead.)
2778 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2779 @var{N} into section @var{secname}.''
2781 Apart from text, data and bss sections you need to know about the
2782 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2783 addresses in the absolute section remain unchanged. For example, address
2784 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2785 @code{@value{LD}}. Although the linker never arranges two partial programs'
2786 data sections with overlapping addresses after linking, @emph{by definition}
2787 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2788 part of a program is always the same address when the program is running as
2789 address @code{@{absolute@ 239@}} in any other part of the program.
2791 The idea of sections is extended to the @dfn{undefined} section. Any
2792 address whose section is unknown at assembly time is by definition
2793 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2794 Since numbers are always defined, the only way to generate an undefined
2795 address is to mention an undefined symbol. A reference to a named
2796 common block would be such a symbol: its value is unknown at assembly
2797 time so it has section @emph{undefined}.
2799 By analogy the word @emph{section} is used to describe groups of sections in
2800 the linked program. @code{@value{LD}} puts all partial programs' text
2801 sections in contiguous addresses in the linked program. It is
2802 customary to refer to the @emph{text section} of a program, meaning all
2803 the addresses of all partial programs' text sections. Likewise for
2804 data and bss sections.
2806 Some sections are manipulated by @code{@value{LD}}; others are invented for
2807 use of @command{@value{AS}} and have no meaning except during assembly.
2810 @section Linker Sections
2811 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2816 @cindex named sections
2817 @cindex sections, named
2818 @item named sections
2821 @cindex text section
2822 @cindex data section
2826 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2827 separate but equal sections. Anything you can say of one section is
2830 When the program is running, however, it is
2831 customary for the text section to be unalterable. The
2832 text section is often shared among processes: it contains
2833 instructions, constants and the like. The data section of a running
2834 program is usually alterable: for example, C variables would be stored
2835 in the data section.
2840 This section contains zeroed bytes when your program begins running. It
2841 is used to hold uninitialized variables or common storage. The length of
2842 each partial program's bss section is important, but because it starts
2843 out containing zeroed bytes there is no need to store explicit zero
2844 bytes in the object file. The bss section was invented to eliminate
2845 those explicit zeros from object files.
2847 @cindex absolute section
2848 @item absolute section
2849 Address 0 of this section is always ``relocated'' to runtime address 0.
2850 This is useful if you want to refer to an address that @code{@value{LD}} must
2851 not change when relocating. In this sense we speak of absolute
2852 addresses being ``unrelocatable'': they do not change during relocation.
2854 @cindex undefined section
2855 @item undefined section
2856 This ``section'' is a catch-all for address references to objects not in
2857 the preceding sections.
2858 @c FIXME: ref to some other doc on obj-file formats could go here.
2861 @cindex relocation example
2862 An idealized example of three relocatable sections follows.
2864 The example uses the traditional section names @samp{.text} and @samp{.data}.
2866 Memory addresses are on the horizontal axis.
2870 @c END TEXI2ROFF-KILL
2873 partial program # 1: |ttttt|dddd|00|
2880 partial program # 2: |TTT|DDD|000|
2883 +--+---+-----+--+----+---+-----+~~
2884 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2885 +--+---+-----+--+----+---+-----+~~
2887 addresses: 0 @dots{}
2894 \line{\it Partial program \#1: \hfil}
2895 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2896 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2898 \line{\it Partial program \#2: \hfil}
2899 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2900 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2902 \line{\it linked program: \hfil}
2903 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2904 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2905 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2906 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2908 \line{\it addresses: \hfil}
2912 @c END TEXI2ROFF-KILL
2915 @section Assembler Internal Sections
2917 @cindex internal assembler sections
2918 @cindex sections in messages, internal
2919 These sections are meant only for the internal use of @command{@value{AS}}. They
2920 have no meaning at run-time. You do not really need to know about these
2921 sections for most purposes; but they can be mentioned in @command{@value{AS}}
2922 warning messages, so it might be helpful to have an idea of their
2923 meanings to @command{@value{AS}}. These sections are used to permit the
2924 value of every expression in your assembly language program to be a
2925 section-relative address.
2928 @cindex assembler internal logic error
2929 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2930 An internal assembler logic error has been found. This means there is a
2931 bug in the assembler.
2933 @cindex expr (internal section)
2935 The assembler stores complex expression internally as combinations of
2936 symbols. When it needs to represent an expression as a symbol, it puts
2937 it in the expr section.
2939 @c FIXME item transfer[t] vector preload
2940 @c FIXME item transfer[t] vector postload
2941 @c FIXME item register
2945 @section Sub-Sections
2947 @cindex numbered subsections
2948 @cindex grouping data
2954 fall into two sections: text and data.
2956 You may have separate groups of
2958 data in named sections
2962 data in named sections
2968 that you want to end up near to each other in the object file, even though they
2969 are not contiguous in the assembler source. @command{@value{AS}} allows you to
2970 use @dfn{subsections} for this purpose. Within each section, there can be
2971 numbered subsections with values from 0 to 8192. Objects assembled into the
2972 same subsection go into the object file together with other objects in the same
2973 subsection. For example, a compiler might want to store constants in the text
2974 section, but might not want to have them interspersed with the program being
2975 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2976 section of code being output, and a @samp{.text 1} before each group of
2977 constants being output.
2979 Subsections are optional. If you do not use subsections, everything
2980 goes in subsection number zero.
2983 Each subsection is zero-padded up to a multiple of four bytes.
2984 (Subsections may be padded a different amount on different flavors
2985 of @command{@value{AS}}.)
2989 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2990 boundary (two bytes).
2991 The same is true on the Renesas SH.
2994 @c FIXME section padding (alignment)?
2995 @c Rich Pixley says padding here depends on target obj code format; that
2996 @c doesn't seem particularly useful to say without further elaboration,
2997 @c so for now I say nothing about it. If this is a generic BFD issue,
2998 @c these paragraphs might need to vanish from this manual, and be
2999 @c discussed in BFD chapter of binutils (or some such).
3002 On the AMD 29K family, no particular padding is added to section or
3003 subsection sizes; @value{AS} forces no alignment on this platform.
3007 Subsections appear in your object file in numeric order, lowest numbered
3008 to highest. (All this to be compatible with other people's assemblers.)
3009 The object file contains no representation of subsections; @code{@value{LD}} and
3010 other programs that manipulate object files see no trace of them.
3011 They just see all your text subsections as a text section, and all your
3012 data subsections as a data section.
3014 To specify which subsection you want subsequent statements assembled
3015 into, use a numeric argument to specify it, in a @samp{.text
3016 @var{expression}} or a @samp{.data @var{expression}} statement.
3019 When generating COFF output, you
3024 can also use an extra subsection
3025 argument with arbitrary named sections: @samp{.section @var{name},
3030 When generating ELF output, you
3035 can also use the @code{.subsection} directive (@pxref{SubSection})
3036 to specify a subsection: @samp{.subsection @var{expression}}.
3038 @var{Expression} should be an absolute expression.
3039 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
3040 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3041 begins in @code{text 0}. For instance:
3043 .text 0 # The default subsection is text 0 anyway.
3044 .ascii "This lives in the first text subsection. *"
3046 .ascii "But this lives in the second text subsection."
3048 .ascii "This lives in the data section,"
3049 .ascii "in the first data subsection."
3051 .ascii "This lives in the first text section,"
3052 .ascii "immediately following the asterisk (*)."
3055 Each section has a @dfn{location counter} incremented by one for every byte
3056 assembled into that section. Because subsections are merely a convenience
3057 restricted to @command{@value{AS}} there is no concept of a subsection location
3058 counter. There is no way to directly manipulate a location counter---but the
3059 @code{.align} directive changes it, and any label definition captures its
3060 current value. The location counter of the section where statements are being
3061 assembled is said to be the @dfn{active} location counter.
3064 @section bss Section
3067 @cindex common variable storage
3068 The bss section is used for local common variable storage.
3069 You may allocate address space in the bss section, but you may
3070 not dictate data to load into it before your program executes. When
3071 your program starts running, all the contents of the bss
3072 section are zeroed bytes.
3074 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3075 @ref{Lcomm,,@code{.lcomm}}.
3077 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3078 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
3081 When assembling for a target which supports multiple sections, such as ELF or
3082 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3083 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3084 section. Typically the section will only contain symbol definitions and
3085 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3092 Symbols are a central concept: the programmer uses symbols to name
3093 things, the linker uses symbols to link, and the debugger uses symbols
3097 @cindex debuggers, and symbol order
3098 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3099 the same order they were declared. This may break some debuggers.
3104 * Setting Symbols:: Giving Symbols Other Values
3105 * Symbol Names:: Symbol Names
3106 * Dot:: The Special Dot Symbol
3107 * Symbol Attributes:: Symbol Attributes
3114 A @dfn{label} is written as a symbol immediately followed by a colon
3115 @samp{:}. The symbol then represents the current value of the
3116 active location counter, and is, for example, a suitable instruction
3117 operand. You are warned if you use the same symbol to represent two
3118 different locations: the first definition overrides any other
3122 On the HPPA, the usual form for a label need not be immediately followed by a
3123 colon, but instead must start in column zero. Only one label may be defined on
3124 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3125 provides a special directive @code{.label} for defining labels more flexibly.
3128 @node Setting Symbols
3129 @section Giving Symbols Other Values
3131 @cindex assigning values to symbols
3132 @cindex symbol values, assigning
3133 A symbol can be given an arbitrary value by writing a symbol, followed
3134 by an equals sign @samp{=}, followed by an expression
3135 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3136 directive. @xref{Set,,@code{.set}}.
3139 @section Symbol Names
3141 @cindex symbol names
3142 @cindex names, symbol
3143 @ifclear SPECIAL-SYMS
3144 Symbol names begin with a letter or with one of @samp{._}. On most
3145 machines, you can also use @code{$} in symbol names; exceptions are
3146 noted in @ref{Machine Dependencies}. That character may be followed by any
3147 string of digits, letters, dollar signs (unless otherwise noted in
3148 @ref{Machine Dependencies}), and underscores.
3151 For the AMD 29K family, @samp{?} is also allowed in the
3152 body of a symbol name, though not at its beginning.
3157 Symbol names begin with a letter or with one of @samp{._}. On the
3158 Renesas SH or the H8/500, you can also use @code{$} in symbol names. That
3159 character may be followed by any string of digits, letters, dollar signs (save
3160 on the H8/300), and underscores.
3164 Case of letters is significant: @code{foo} is a different symbol name
3167 Each symbol has exactly one name. Each name in an assembly language program
3168 refers to exactly one symbol. You may use that symbol name any number of times
3171 @subheading Local Symbol Names
3173 @cindex local symbol names
3174 @cindex symbol names, local
3175 @cindex temporary symbol names
3176 @cindex symbol names, temporary
3177 Local symbols help compilers and programmers use names temporarily.
3178 They create symbols which are guaranteed to be unique over the entire scope of
3179 the input source code and which can be referred to by a simple notation.
3180 To define a local symbol, write a label of the form @samp{@b{N}:} (where @b{N}
3181 represents any positive integer). To refer to the most recent previous
3182 definition of that symbol write @samp{@b{N}b}, using the same number as when
3183 you defined the label. To refer to the next definition of a local label, write
3184 @samp{@b{N}f}--- The @samp{b} stands for``backwards'' and the @samp{f} stands
3187 There is no restriction on how you can use these labels, and you can reuse them
3188 too. So that it is possible to repeatedly define the same local label (using
3189 the same number @samp{@b{N}}), although you can only refer to the most recently
3190 defined local label of that number (for a backwards reference) or the next
3191 definition of a specific local label for a forward reference. It is also worth
3192 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3193 implemented in a slightly more efficient manner than the others.
3204 Which is the equivalent of:
3207 label_1: branch label_3
3208 label_2: branch label_1
3209 label_3: branch label_4
3210 label_4: branch label_3
3213 Local symbol names are only a notational device. They are immediately
3214 transformed into more conventional symbol names before the assembler uses them.
3215 The symbol names stored in the symbol table, appearing in error messages and
3216 optionally emitted to the object file. The names are constructed using these
3221 All local labels begin with @samp{L}. Normally both @command{@value{AS}} and
3222 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
3223 used for symbols you are never intended to see. If you use the
3224 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3225 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3226 you may use them in debugging.
3229 This is the number that was used in the local label definition. So if the
3230 label is written @samp{55:} then the number is @samp{55}.
3233 This unusual character is included so you do not accidentally invent a symbol
3234 of the same name. The character has ASCII value of @samp{\002} (control-B).
3236 @item @emph{ordinal number}
3237 This is a serial number to keep the labels distinct. The first definition of
3238 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3239 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3240 the number @samp{1} and its 15th defintion gets @samp{15} as well.
3243 So for example, the first @code{1:} is named @code{L1@kbd{C-B}1}, the 44th
3244 @code{3:} is named @code{L3@kbd{C-B}44}.
3246 @subheading Dollar Local Labels
3247 @cindex dollar local symbols
3249 @code{@value{AS}} also supports an even more local form of local labels called
3250 dollar labels. These labels go out of scope (ie they become undefined) as soon
3251 as a non-local label is defined. Thus they remain valid for only a small
3252 region of the input source code. Normal local labels, by contrast, remain in
3253 scope for the entire file, or until they are redefined by another occurrence of
3254 the same local label.
3256 Dollar labels are defined in exactly the same way as ordinary local labels,
3257 except that instead of being terminated by a colon, they are terminated by a
3258 dollar sign. eg @samp{@b{55$}}.
3260 They can also be distinguished from ordinary local labels by their transformed
3261 name which uses ASCII character @samp{\001} (control-A) as the magic character
3262 to distinguish them from ordinary labels. Thus the 5th defintion of @samp{6$}
3263 is named @samp{L6@kbd{C-A}5}.
3266 @section The Special Dot Symbol
3268 @cindex dot (symbol)
3269 @cindex @code{.} (symbol)
3270 @cindex current address
3271 @cindex location counter
3272 The special symbol @samp{.} refers to the current address that
3273 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3274 .long .} defines @code{melvin} to contain its own address.
3275 Assigning a value to @code{.} is treated the same as a @code{.org}
3276 directive. Thus, the expression @samp{.=.+4} is the same as saying
3277 @ifclear no-space-dir
3286 @node Symbol Attributes
3287 @section Symbol Attributes
3289 @cindex symbol attributes
3290 @cindex attributes, symbol
3291 Every symbol has, as well as its name, the attributes ``Value'' and
3292 ``Type''. Depending on output format, symbols can also have auxiliary
3295 The detailed definitions are in @file{a.out.h}.
3298 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3299 all these attributes, and probably won't warn you. This makes the
3300 symbol an externally defined symbol, which is generally what you
3304 * Symbol Value:: Value
3305 * Symbol Type:: Type
3308 * a.out Symbols:: Symbol Attributes: @code{a.out}
3312 * a.out Symbols:: Symbol Attributes: @code{a.out}
3315 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3320 * COFF Symbols:: Symbol Attributes for COFF
3323 * SOM Symbols:: Symbol Attributes for SOM
3330 @cindex value of a symbol
3331 @cindex symbol value
3332 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3333 location in the text, data, bss or absolute sections the value is the
3334 number of addresses from the start of that section to the label.
3335 Naturally for text, data and bss sections the value of a symbol changes
3336 as @code{@value{LD}} changes section base addresses during linking. Absolute
3337 symbols' values do not change during linking: that is why they are
3340 The value of an undefined symbol is treated in a special way. If it is
3341 0 then the symbol is not defined in this assembler source file, and
3342 @code{@value{LD}} tries to determine its value from other files linked into the
3343 same program. You make this kind of symbol simply by mentioning a symbol
3344 name without defining it. A non-zero value represents a @code{.comm}
3345 common declaration. The value is how much common storage to reserve, in
3346 bytes (addresses). The symbol refers to the first address of the
3352 @cindex type of a symbol
3354 The type attribute of a symbol contains relocation (section)
3355 information, any flag settings indicating that a symbol is external, and
3356 (optionally), other information for linkers and debuggers. The exact
3357 format depends on the object-code output format in use.
3362 @c The following avoids a "widow" subsection title. @group would be
3363 @c better if it were available outside examples.
3366 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3368 @cindex @code{b.out} symbol attributes
3369 @cindex symbol attributes, @code{b.out}
3370 These symbol attributes appear only when @command{@value{AS}} is configured for
3371 one of the Berkeley-descended object output formats---@code{a.out} or
3377 @subsection Symbol Attributes: @code{a.out}
3379 @cindex @code{a.out} symbol attributes
3380 @cindex symbol attributes, @code{a.out}
3386 @subsection Symbol Attributes: @code{a.out}
3388 @cindex @code{a.out} symbol attributes
3389 @cindex symbol attributes, @code{a.out}
3393 * Symbol Desc:: Descriptor
3394 * Symbol Other:: Other
3398 @subsubsection Descriptor
3400 @cindex descriptor, of @code{a.out} symbol
3401 This is an arbitrary 16-bit value. You may establish a symbol's
3402 descriptor value by using a @code{.desc} statement
3403 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3404 @command{@value{AS}}.
3407 @subsubsection Other
3409 @cindex other attribute, of @code{a.out} symbol
3410 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3415 @subsection Symbol Attributes for COFF
3417 @cindex COFF symbol attributes
3418 @cindex symbol attributes, COFF
3420 The COFF format supports a multitude of auxiliary symbol attributes;
3421 like the primary symbol attributes, they are set between @code{.def} and
3422 @code{.endef} directives.
3424 @subsubsection Primary Attributes
3426 @cindex primary attributes, COFF symbols
3427 The symbol name is set with @code{.def}; the value and type,
3428 respectively, with @code{.val} and @code{.type}.
3430 @subsubsection Auxiliary Attributes
3432 @cindex auxiliary attributes, COFF symbols
3433 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3434 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3435 table information for COFF.
3440 @subsection Symbol Attributes for SOM
3442 @cindex SOM symbol attributes
3443 @cindex symbol attributes, SOM
3445 The SOM format for the HPPA supports a multitude of symbol attributes set with
3446 the @code{.EXPORT} and @code{.IMPORT} directives.
3448 The attributes are described in @cite{HP9000 Series 800 Assembly
3449 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3450 @code{EXPORT} assembler directive documentation.
3454 @chapter Expressions
3458 @cindex numeric values
3459 An @dfn{expression} specifies an address or numeric value.
3460 Whitespace may precede and/or follow an expression.
3462 The result of an expression must be an absolute number, or else an offset into
3463 a particular section. If an expression is not absolute, and there is not
3464 enough information when @command{@value{AS}} sees the expression to know its
3465 section, a second pass over the source program might be necessary to interpret
3466 the expression---but the second pass is currently not implemented.
3467 @command{@value{AS}} aborts with an error message in this situation.
3470 * Empty Exprs:: Empty Expressions
3471 * Integer Exprs:: Integer Expressions
3475 @section Empty Expressions
3477 @cindex empty expressions
3478 @cindex expressions, empty
3479 An empty expression has no value: it is just whitespace or null.
3480 Wherever an absolute expression is required, you may omit the
3481 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3482 is compatible with other assemblers.
3485 @section Integer Expressions
3487 @cindex integer expressions
3488 @cindex expressions, integer
3489 An @dfn{integer expression} is one or more @emph{arguments} delimited
3490 by @emph{operators}.
3493 * Arguments:: Arguments
3494 * Operators:: Operators
3495 * Prefix Ops:: Prefix Operators
3496 * Infix Ops:: Infix Operators
3500 @subsection Arguments
3502 @cindex expression arguments
3503 @cindex arguments in expressions
3504 @cindex operands in expressions
3505 @cindex arithmetic operands
3506 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3507 contexts arguments are sometimes called ``arithmetic operands''. In
3508 this manual, to avoid confusing them with the ``instruction operands'' of
3509 the machine language, we use the term ``argument'' to refer to parts of
3510 expressions only, reserving the word ``operand'' to refer only to machine
3511 instruction operands.
3513 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3514 @var{section} is one of text, data, bss, absolute,
3515 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3518 Numbers are usually integers.
3520 A number can be a flonum or bignum. In this case, you are warned
3521 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3522 these 32 bits are an integer. You may write integer-manipulating
3523 instructions that act on exotic constants, compatible with other
3526 @cindex subexpressions
3527 Subexpressions are a left parenthesis @samp{(} followed by an integer
3528 expression, followed by a right parenthesis @samp{)}; or a prefix
3529 operator followed by an argument.
3532 @subsection Operators
3534 @cindex operators, in expressions
3535 @cindex arithmetic functions
3536 @cindex functions, in expressions
3537 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3538 operators are followed by an argument. Infix operators appear
3539 between their arguments. Operators may be preceded and/or followed by
3543 @subsection Prefix Operator
3545 @cindex prefix operators
3546 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3547 one argument, which must be absolute.
3549 @c the tex/end tex stuff surrounding this small table is meant to make
3550 @c it align, on the printed page, with the similar table in the next
3551 @c section (which is inside an enumerate).
3553 \global\advance\leftskip by \itemindent
3558 @dfn{Negation}. Two's complement negation.
3560 @dfn{Complementation}. Bitwise not.
3564 \global\advance\leftskip by -\itemindent
3568 @subsection Infix Operators
3570 @cindex infix operators
3571 @cindex operators, permitted arguments
3572 @dfn{Infix operators} take two arguments, one on either side. Operators
3573 have precedence, but operations with equal precedence are performed left
3574 to right. Apart from @code{+} or @option{-}, both arguments must be
3575 absolute, and the result is absolute.
3578 @cindex operator precedence
3579 @cindex precedence of operators
3586 @dfn{Multiplication}.
3589 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3596 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3600 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3604 Intermediate precedence
3609 @dfn{Bitwise Inclusive Or}.
3615 @dfn{Bitwise Exclusive Or}.
3618 @dfn{Bitwise Or Not}.
3625 @cindex addition, permitted arguments
3626 @cindex plus, permitted arguments
3627 @cindex arguments for addition
3629 @dfn{Addition}. If either argument is absolute, the result has the section of
3630 the other argument. You may not add together arguments from different
3633 @cindex subtraction, permitted arguments
3634 @cindex minus, permitted arguments
3635 @cindex arguments for subtraction
3637 @dfn{Subtraction}. If the right argument is absolute, the
3638 result has the section of the left argument.
3639 If both arguments are in the same section, the result is absolute.
3640 You may not subtract arguments from different sections.
3641 @c FIXME is there still something useful to say about undefined - undefined ?
3643 @cindex comparison expressions
3644 @cindex expressions, comparison
3648 @dfn{Is Not Equal To}
3652 @dfn{Is Greater Than}
3654 @dfn{Is Greater Than Or Equal To}
3656 @dfn{Is Less Than Or Equal To}
3658 The comparison operators can be used as infix operators. A true results has a
3659 value of -1 whereas a false result has a value of 0. Note, these operators
3660 perform signed comparisons.
3663 @item Lowest Precedence
3672 These two logical operations can be used to combine the results of sub
3673 expressions. Note, unlike the comparison operators a true result returns a
3674 value of 1 but a false results does still return 0. Also note that the logical
3675 or operator has a slightly lower precedence than logical and.
3680 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3681 address; you can only have a defined section in one of the two arguments.
3684 @chapter Assembler Directives
3686 @cindex directives, machine independent
3687 @cindex pseudo-ops, machine independent
3688 @cindex machine independent directives
3689 All assembler directives have names that begin with a period (@samp{.}).
3690 The rest of the name is letters, usually in lower case.
3692 This chapter discusses directives that are available regardless of the
3693 target machine configuration for the @sc{gnu} assembler.
3695 Some machine configurations provide additional directives.
3696 @xref{Machine Dependencies}.
3699 @ifset machine-directives
3700 @xref{Machine Dependencies} for additional directives.
3705 * Abort:: @code{.abort}
3707 * ABORT:: @code{.ABORT}
3710 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3711 * Altmacro:: @code{.altmacro}
3712 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3713 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3714 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3715 * Byte:: @code{.byte @var{expressions}}
3716 * Comm:: @code{.comm @var{symbol} , @var{length} }
3718 * CFI directives:: @code{.cfi_startproc}, @code{.cfi_endproc}, etc.
3720 * Data:: @code{.data @var{subsection}}
3722 * Def:: @code{.def @var{name}}
3725 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3731 * Double:: @code{.double @var{flonums}}
3732 * Eject:: @code{.eject}
3733 * Else:: @code{.else}
3734 * Elseif:: @code{.elseif}
3737 * Endef:: @code{.endef}
3740 * Endfunc:: @code{.endfunc}
3741 * Endif:: @code{.endif}
3742 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3743 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3745 * Error:: @code{.error @var{string}}
3746 * Exitm:: @code{.exitm}
3747 * Extern:: @code{.extern}
3748 * Fail:: @code{.fail}
3749 @ifclear no-file-dir
3750 * File:: @code{.file @var{string}}
3753 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3754 * Float:: @code{.float @var{flonums}}
3755 * Func:: @code{.func}
3756 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3758 * Hidden:: @code{.hidden @var{names}}
3761 * hword:: @code{.hword @var{expressions}}
3762 * Ident:: @code{.ident}
3763 * If:: @code{.if @var{absolute expression}}
3764 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3765 * Include:: @code{.include "@var{file}"}
3766 * Int:: @code{.int @var{expressions}}
3768 * Internal:: @code{.internal @var{names}}
3771 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3772 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3773 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3774 * Lflags:: @code{.lflags}
3775 @ifclear no-line-dir
3776 * Line:: @code{.line @var{line-number}}
3779 * Ln:: @code{.ln @var{line-number}}
3780 * Linkonce:: @code{.linkonce [@var{type}]}
3781 * List:: @code{.list}
3782 * Long:: @code{.long @var{expressions}}
3784 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3787 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3788 * MRI:: @code{.mri @var{val}}
3789 * Noaltmacro:: @code{.noaltmacro}
3790 * Nolist:: @code{.nolist}
3791 * Octa:: @code{.octa @var{bignums}}
3792 * Org:: @code{.org @var{new-lc} , @var{fill}}
3793 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3795 * PopSection:: @code{.popsection}
3796 * Previous:: @code{.previous}
3799 * Print:: @code{.print @var{string}}
3801 * Protected:: @code{.protected @var{names}}
3804 * Psize:: @code{.psize @var{lines}, @var{columns}}
3805 * Purgem:: @code{.purgem @var{name}}
3807 * PushSection:: @code{.pushsection @var{name}}
3810 * Quad:: @code{.quad @var{bignums}}
3811 * Rept:: @code{.rept @var{count}}
3812 * Sbttl:: @code{.sbttl "@var{subheading}"}
3814 * Scl:: @code{.scl @var{class}}
3817 * Section:: @code{.section @var{name}}
3820 * Set:: @code{.set @var{symbol}, @var{expression}}
3821 * Short:: @code{.short @var{expressions}}
3822 * Single:: @code{.single @var{flonums}}
3824 * Size:: @code{.size [@var{name} , @var{expression}]}
3827 * Skip:: @code{.skip @var{size} , @var{fill}}
3828 * Sleb128:: @code{.sleb128 @var{expressions}}
3829 * Space:: @code{.space @var{size} , @var{fill}}
3831 * Stab:: @code{.stabd, .stabn, .stabs}
3834 * String:: @code{.string "@var{str}"}
3835 * Struct:: @code{.struct @var{expression}}
3837 * SubSection:: @code{.subsection}
3838 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3842 * Tag:: @code{.tag @var{structname}}
3845 * Text:: @code{.text @var{subsection}}
3846 * Title:: @code{.title "@var{heading}"}
3848 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3851 * Uleb128:: @code{.uleb128 @var{expressions}}
3853 * Val:: @code{.val @var{addr}}
3857 * Version:: @code{.version "@var{string}"}
3858 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3859 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3862 * Warning:: @code{.warning @var{string}}
3863 * Weak:: @code{.weak @var{names}}
3864 * Word:: @code{.word @var{expressions}}
3865 * Deprecated:: Deprecated Directives
3869 @section @code{.abort}
3871 @cindex @code{abort} directive
3872 @cindex stopping the assembly
3873 This directive stops the assembly immediately. It is for
3874 compatibility with other assemblers. The original idea was that the
3875 assembly language source would be piped into the assembler. If the sender
3876 of the source quit, it could use this directive tells @command{@value{AS}} to
3877 quit also. One day @code{.abort} will not be supported.
3881 @section @code{.ABORT}
3883 @cindex @code{ABORT} directive
3884 When producing COFF output, @command{@value{AS}} accepts this directive as a
3885 synonym for @samp{.abort}.
3888 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
3894 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3896 @cindex padding the location counter
3897 @cindex @code{align} directive
3898 Pad the location counter (in the current subsection) to a particular storage
3899 boundary. The first expression (which must be absolute) is the alignment
3900 required, as described below.
3902 The second expression (also absolute) gives the fill value to be stored in the
3903 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3904 padding bytes are normally zero. However, on some systems, if the section is
3905 marked as containing code and the fill value is omitted, the space is filled
3906 with no-op instructions.
3908 The third expression is also absolute, and is also optional. If it is present,
3909 it is the maximum number of bytes that should be skipped by this alignment
3910 directive. If doing the alignment would require skipping more bytes than the
3911 specified maximum, then the alignment is not done at all. You can omit the
3912 fill value (the second argument) entirely by simply using two commas after the
3913 required alignment; this can be useful if you want the alignment to be filled
3914 with no-op instructions when appropriate.
3916 The way the required alignment is specified varies from system to system.
3917 For the a29k, arc, hppa, i386 using ELF, i860, iq2000, m68k, m88k, or32,
3918 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
3919 alignment request in bytes. For example @samp{.align 8} advances
3920 the location counter until it is a multiple of 8. If the location counter
3921 is already a multiple of 8, no change is needed. For the tic54x, the
3922 first expression is the alignment request in words.
3924 For other systems, including the i386 using a.out format, and the arm and
3925 strongarm, it is the
3926 number of low-order zero bits the location counter must have after
3927 advancement. For example @samp{.align 3} advances the location
3928 counter until it a multiple of 8. If the location counter is already a
3929 multiple of 8, no change is needed.
3931 This inconsistency is due to the different behaviors of the various
3932 native assemblers for these systems which GAS must emulate.
3933 GAS also provides @code{.balign} and @code{.p2align} directives,
3934 described later, which have a consistent behavior across all
3935 architectures (but are specific to GAS).
3938 @section @code{.ascii "@var{string}"}@dots{}
3940 @cindex @code{ascii} directive
3941 @cindex string literals
3942 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3943 separated by commas. It assembles each string (with no automatic
3944 trailing zero byte) into consecutive addresses.
3947 @section @code{.asciz "@var{string}"}@dots{}
3949 @cindex @code{asciz} directive
3950 @cindex zero-terminated strings
3951 @cindex null-terminated strings
3952 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3953 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3956 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3958 @cindex padding the location counter given number of bytes
3959 @cindex @code{balign} directive
3960 Pad the location counter (in the current subsection) to a particular
3961 storage boundary. The first expression (which must be absolute) is the
3962 alignment request in bytes. For example @samp{.balign 8} advances
3963 the location counter until it is a multiple of 8. If the location counter
3964 is already a multiple of 8, no change is needed.
3966 The second expression (also absolute) gives the fill value to be stored in the
3967 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3968 padding bytes are normally zero. However, on some systems, if the section is
3969 marked as containing code and the fill value is omitted, the space is filled
3970 with no-op instructions.
3972 The third expression is also absolute, and is also optional. If it is present,
3973 it is the maximum number of bytes that should be skipped by this alignment
3974 directive. If doing the alignment would require skipping more bytes than the
3975 specified maximum, then the alignment is not done at all. You can omit the
3976 fill value (the second argument) entirely by simply using two commas after the
3977 required alignment; this can be useful if you want the alignment to be filled
3978 with no-op instructions when appropriate.
3980 @cindex @code{balignw} directive
3981 @cindex @code{balignl} directive
3982 The @code{.balignw} and @code{.balignl} directives are variants of the
3983 @code{.balign} directive. The @code{.balignw} directive treats the fill
3984 pattern as a two byte word value. The @code{.balignl} directives treats the
3985 fill pattern as a four byte longword value. For example, @code{.balignw
3986 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3987 filled in with the value 0x368d (the exact placement of the bytes depends upon
3988 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3992 @section @code{.byte @var{expressions}}
3994 @cindex @code{byte} directive
3995 @cindex integers, one byte
3996 @code{.byte} expects zero or more expressions, separated by commas.
3997 Each expression is assembled into the next byte.
4000 @section @code{.comm @var{symbol} , @var{length} }
4002 @cindex @code{comm} directive
4003 @cindex symbol, common
4004 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4005 common symbol in one object file may be merged with a defined or common symbol
4006 of the same name in another object file. If @code{@value{LD}} does not see a
4007 definition for the symbol--just one or more common symbols--then it will
4008 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4009 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4010 the same name, and they do not all have the same size, it will allocate space
4011 using the largest size.
4014 When using ELF, the @code{.comm} directive takes an optional third argument.
4015 This is the desired alignment of the symbol, specified as a byte boundary (for
4016 example, an alignment of 16 means that the least significant 4 bits of the
4017 address should be zero). The alignment must be an absolute expression, and it
4018 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
4019 for the common symbol, it will use the alignment when placing the symbol. If
4020 no alignment is specified, @command{@value{AS}} will set the alignment to the
4021 largest power of two less than or equal to the size of the symbol, up to a
4026 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4027 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4030 @node CFI directives
4031 @section @code{.cfi_startproc}
4032 @cindex @code{cfi_startproc} directive
4033 @code{.cfi_startproc} is used at the beginning of each function that
4034 should have an entry in @code{.eh_frame}. It initializes some internal
4035 data structures and emits architecture dependent initial CFI instructions.
4036 Don't forget to close the function by
4037 @code{.cfi_endproc}.
4039 @section @code{.cfi_endproc}
4040 @cindex @code{cfi_endproc} directive
4041 @code{.cfi_endproc} is used at the end of a function where it closes its
4042 unwind entry previously opened by
4043 @code{.cfi_startproc}. and emits it to @code{.eh_frame}.
4045 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4046 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4047 address from @var{register} and add @var{offset} to it}.
4049 @section @code{.cfi_def_cfa_register @var{register}}
4050 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4051 now on @var{register} will be used instead of the old one. Offset
4054 @section @code{.cfi_def_cfa_offset @var{offset}}
4055 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4056 remains the same, but @var{offset} is new. Note that it is the
4057 absolute offset that will be added to a defined register to compute
4060 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4061 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4062 value that is added/substracted from the previous offset.
4064 @section @code{.cfi_offset @var{register}, @var{offset}}
4065 Previous value of @var{register} is saved at offset @var{offset} from
4068 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4069 Previous value of @var{register} is saved at offset @var{offset} from
4070 the current CFA register. This is transformed to @code{.cfi_offset}
4071 using the known displacement of the CFA register from the CFA.
4072 This is often easier to use, because the number will match the
4073 code it's annotating.
4075 @section @code{.cfi_window_save}
4076 SPARC register window has been saved.
4078 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4079 Allows the user to add arbitrary bytes to the unwind info. One
4080 might use this to add OS-specific CFI opcodes, or generic CFI
4081 opcodes that GAS does not yet support.
4084 @section @code{.data @var{subsection}}
4086 @cindex @code{data} directive
4087 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4088 end of the data subsection numbered @var{subsection} (which is an
4089 absolute expression). If @var{subsection} is omitted, it defaults
4094 @section @code{.def @var{name}}
4096 @cindex @code{def} directive
4097 @cindex COFF symbols, debugging
4098 @cindex debugging COFF symbols
4099 Begin defining debugging information for a symbol @var{name}; the
4100 definition extends until the @code{.endef} directive is encountered.
4103 This directive is only observed when @command{@value{AS}} is configured for COFF
4104 format output; when producing @code{b.out}, @samp{.def} is recognized,
4111 @section @code{.desc @var{symbol}, @var{abs-expression}}
4113 @cindex @code{desc} directive
4114 @cindex COFF symbol descriptor
4115 @cindex symbol descriptor, COFF
4116 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4117 to the low 16 bits of an absolute expression.
4120 The @samp{.desc} directive is not available when @command{@value{AS}} is
4121 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4122 object format. For the sake of compatibility, @command{@value{AS}} accepts
4123 it, but produces no output, when configured for COFF.
4129 @section @code{.dim}
4131 @cindex @code{dim} directive
4132 @cindex COFF auxiliary symbol information
4133 @cindex auxiliary symbol information, COFF
4134 This directive is generated by compilers to include auxiliary debugging
4135 information in the symbol table. It is only permitted inside
4136 @code{.def}/@code{.endef} pairs.
4139 @samp{.dim} is only meaningful when generating COFF format output; when
4140 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4146 @section @code{.double @var{flonums}}
4148 @cindex @code{double} directive
4149 @cindex floating point numbers (double)
4150 @code{.double} expects zero or more flonums, separated by commas. It
4151 assembles floating point numbers.
4153 The exact kind of floating point numbers emitted depends on how
4154 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4158 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4159 in @sc{ieee} format.
4164 @section @code{.eject}
4166 @cindex @code{eject} directive
4167 @cindex new page, in listings
4168 @cindex page, in listings
4169 @cindex listing control: new page
4170 Force a page break at this point, when generating assembly listings.
4173 @section @code{.else}
4175 @cindex @code{else} directive
4176 @code{.else} is part of the @command{@value{AS}} support for conditional
4177 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
4178 of code to be assembled if the condition for the preceding @code{.if}
4182 @section @code{.elseif}
4184 @cindex @code{elseif} directive
4185 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4186 assembly; @pxref{If,,@code{.if}}. It is shorthand for beginning a new
4187 @code{.if} block that would otherwise fill the entire @code{.else} section.
4190 @section @code{.end}
4192 @cindex @code{end} directive
4193 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4194 process anything in the file past the @code{.end} directive.
4198 @section @code{.endef}
4200 @cindex @code{endef} directive
4201 This directive flags the end of a symbol definition begun with
4205 @samp{.endef} is only meaningful when generating COFF format output; if
4206 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4207 directive but ignores it.
4212 @section @code{.endfunc}
4213 @cindex @code{endfunc} directive
4214 @code{.endfunc} marks the end of a function specified with @code{.func}.
4217 @section @code{.endif}
4219 @cindex @code{endif} directive
4220 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4221 it marks the end of a block of code that is only assembled
4222 conditionally. @xref{If,,@code{.if}}.
4225 @section @code{.equ @var{symbol}, @var{expression}}
4227 @cindex @code{equ} directive
4228 @cindex assigning values to symbols
4229 @cindex symbols, assigning values to
4230 This directive sets the value of @var{symbol} to @var{expression}.
4231 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
4234 The syntax for @code{equ} on the HPPA is
4235 @samp{@var{symbol} .equ @var{expression}}.
4239 @section @code{.equiv @var{symbol}, @var{expression}}
4240 @cindex @code{equiv} directive
4241 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4242 the assembler will signal an error if @var{symbol} is already defined. Note a
4243 symbol which has been referenced but not actually defined is considered to be
4246 Except for the contents of the error message, this is roughly equivalent to
4255 @section @code{.err}
4256 @cindex @code{err} directive
4257 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4258 message and, unless the @option{-Z} option was used, it will not generate an
4259 object file. This can be used to signal error an conditionally compiled code.
4262 @section @code{.error "@var{string}"}
4263 @cindex error directive
4265 Similarly to @code{.err}, this directive emits an error, but you can specify a
4266 string that will be emitted as the error message. If you don't specify the
4267 message, it defaults to @code{".error directive invoked in source file"}.
4268 @xref{Errors, ,Error and Warning Messages}.
4271 .error "This code has not been assembled and tested."
4275 @section @code{.exitm}
4276 Exit early from the current macro definition. @xref{Macro}.
4279 @section @code{.extern}
4281 @cindex @code{extern} directive
4282 @code{.extern} is accepted in the source program---for compatibility
4283 with other assemblers---but it is ignored. @command{@value{AS}} treats
4284 all undefined symbols as external.
4287 @section @code{.fail @var{expression}}
4289 @cindex @code{fail} directive
4290 Generates an error or a warning. If the value of the @var{expression} is 500
4291 or more, @command{@value{AS}} will print a warning message. If the value is less
4292 than 500, @command{@value{AS}} will print an error message. The message will
4293 include the value of @var{expression}. This can occasionally be useful inside
4294 complex nested macros or conditional assembly.
4296 @ifclear no-file-dir
4298 @section @code{.file @var{string}}
4300 @cindex @code{file} directive
4301 @cindex logical file name
4302 @cindex file name, logical
4303 @code{.file} tells @command{@value{AS}} that we are about to start a new logical
4304 file. @var{string} is the new file name. In general, the filename is
4305 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4306 to specify an empty file name, you must give the quotes--@code{""}. This
4307 statement may go away in future: it is only recognized to be compatible with
4308 old @command{@value{AS}} programs.
4310 In some configurations of @command{@value{AS}}, @code{.file} has already been
4311 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
4316 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4318 @cindex @code{fill} directive
4319 @cindex writing patterns in memory
4320 @cindex patterns, writing in memory
4321 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4322 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4323 may be zero or more. @var{Size} may be zero or more, but if it is
4324 more than 8, then it is deemed to have the value 8, compatible with
4325 other people's assemblers. The contents of each @var{repeat} bytes
4326 is taken from an 8-byte number. The highest order 4 bytes are
4327 zero. The lowest order 4 bytes are @var{value} rendered in the
4328 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4329 Each @var{size} bytes in a repetition is taken from the lowest order
4330 @var{size} bytes of this number. Again, this bizarre behavior is
4331 compatible with other people's assemblers.
4333 @var{size} and @var{value} are optional.
4334 If the second comma and @var{value} are absent, @var{value} is
4335 assumed zero. If the first comma and following tokens are absent,
4336 @var{size} is assumed to be 1.
4339 @section @code{.float @var{flonums}}
4341 @cindex floating point numbers (single)
4342 @cindex @code{float} directive
4343 This directive assembles zero or more flonums, separated by commas. It
4344 has the same effect as @code{.single}.
4346 The exact kind of floating point numbers emitted depends on how
4347 @command{@value{AS}} is configured.
4348 @xref{Machine Dependencies}.
4352 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4353 in @sc{ieee} format.
4358 @section @code{.func @var{name}[,@var{label}]}
4359 @cindex @code{func} directive
4360 @code{.func} emits debugging information to denote function @var{name}, and
4361 is ignored unless the file is assembled with debugging enabled.
4362 Only @samp{--gstabs[+]} is currently supported.
4363 @var{label} is the entry point of the function and if omitted @var{name}
4364 prepended with the @samp{leading char} is used.
4365 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4366 All functions are currently defined to have @code{void} return type.
4367 The function must be terminated with @code{.endfunc}.
4370 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4372 @cindex @code{global} directive
4373 @cindex symbol, making visible to linker
4374 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4375 @var{symbol} in your partial program, its value is made available to
4376 other partial programs that are linked with it. Otherwise,
4377 @var{symbol} takes its attributes from a symbol of the same name
4378 from another file linked into the same program.
4380 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4381 compatibility with other assemblers.
4384 On the HPPA, @code{.global} is not always enough to make it accessible to other
4385 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4386 @xref{HPPA Directives,, HPPA Assembler Directives}.
4391 @section @code{.hidden @var{names}}
4393 @cindex @code{hidden} directive
4395 This is one of the ELF visibility directives. The other two are
4396 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4397 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4399 This directive overrides the named symbols default visibility (which is set by
4400 their binding: local, global or weak). The directive sets the visibility to
4401 @code{hidden} which means that the symbols are not visible to other components.
4402 Such symbols are always considered to be @code{protected} as well.
4406 @section @code{.hword @var{expressions}}
4408 @cindex @code{hword} directive
4409 @cindex integers, 16-bit
4410 @cindex numbers, 16-bit
4411 @cindex sixteen bit integers
4412 This expects zero or more @var{expressions}, and emits
4413 a 16 bit number for each.
4416 This directive is a synonym for @samp{.short}; depending on the target
4417 architecture, it may also be a synonym for @samp{.word}.
4421 This directive is a synonym for @samp{.short}.
4424 This directive is a synonym for both @samp{.short} and @samp{.word}.
4429 @section @code{.ident}
4431 @cindex @code{ident} directive
4432 This directive is used by some assemblers to place tags in object files.
4433 @command{@value{AS}} simply accepts the directive for source-file
4434 compatibility with such assemblers, but does not actually emit anything
4438 @section @code{.if @var{absolute expression}}
4440 @cindex conditional assembly
4441 @cindex @code{if} directive
4442 @code{.if} marks the beginning of a section of code which is only
4443 considered part of the source program being assembled if the argument
4444 (which must be an @var{absolute expression}) is non-zero. The end of
4445 the conditional section of code must be marked by @code{.endif}
4446 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4447 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4448 If you have several conditions to check, @code{.elseif} may be used to avoid
4449 nesting blocks if/else within each subsequent @code{.else} block.
4451 The following variants of @code{.if} are also supported:
4453 @cindex @code{ifdef} directive
4454 @item .ifdef @var{symbol}
4455 Assembles the following section of code if the specified @var{symbol}
4456 has been defined. Note a symbol which has been referenced but not yet defined
4457 is considered to be undefined.
4459 @cindex @code{ifb} directive
4460 @item .ifb @var{text}
4461 Assembles the following section of code if the operand is blank (empty).
4463 @cindex @code{ifc} directive
4464 @item .ifc @var{string1},@var{string2}
4465 Assembles the following section of code if the two strings are the same. The
4466 strings may be optionally quoted with single quotes. If they are not quoted,
4467 the first string stops at the first comma, and the second string stops at the
4468 end of the line. Strings which contain whitespace should be quoted. The
4469 string comparison is case sensitive.
4471 @cindex @code{ifeq} directive
4472 @item .ifeq @var{absolute expression}
4473 Assembles the following section of code if the argument is zero.
4475 @cindex @code{ifeqs} directive
4476 @item .ifeqs @var{string1},@var{string2}
4477 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4479 @cindex @code{ifge} directive
4480 @item .ifge @var{absolute expression}
4481 Assembles the following section of code if the argument is greater than or
4484 @cindex @code{ifgt} directive
4485 @item .ifgt @var{absolute expression}
4486 Assembles the following section of code if the argument is greater than zero.
4488 @cindex @code{ifle} directive
4489 @item .ifle @var{absolute expression}
4490 Assembles the following section of code if the argument is less than or equal
4493 @cindex @code{iflt} directive
4494 @item .iflt @var{absolute expression}
4495 Assembles the following section of code if the argument is less than zero.
4497 @cindex @code{ifnb} directive
4498 @item .ifnb @var{text}
4499 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
4500 following section of code if the operand is non-blank (non-empty).
4502 @cindex @code{ifnc} directive
4503 @item .ifnc @var{string1},@var{string2}.
4504 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4505 following section of code if the two strings are not the same.
4507 @cindex @code{ifndef} directive
4508 @cindex @code{ifnotdef} directive
4509 @item .ifndef @var{symbol}
4510 @itemx .ifnotdef @var{symbol}
4511 Assembles the following section of code if the specified @var{symbol}
4512 has not been defined. Both spelling variants are equivalent. Note a symbol
4513 which has been referenced but not yet defined is considered to be undefined.
4515 @cindex @code{ifne} directive
4516 @item .ifne @var{absolute expression}
4517 Assembles the following section of code if the argument is not equal to zero
4518 (in other words, this is equivalent to @code{.if}).
4520 @cindex @code{ifnes} directive
4521 @item .ifnes @var{string1},@var{string2}
4522 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4523 following section of code if the two strings are not the same.
4527 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4529 @cindex @code{incbin} directive
4530 @cindex binary files, including
4531 The @code{incbin} directive includes @var{file} verbatim at the current
4532 location. You can control the search paths used with the @samp{-I} command-line
4533 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4536 The @var{skip} argument skips a number of bytes from the start of the
4537 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4538 read. Note that the data is not aligned in any way, so it is the user's
4539 responsibility to make sure that proper alignment is provided both before and
4540 after the @code{incbin} directive.
4543 @section @code{.include "@var{file}"}
4545 @cindex @code{include} directive
4546 @cindex supporting files, including
4547 @cindex files, including
4548 This directive provides a way to include supporting files at specified
4549 points in your source program. The code from @var{file} is assembled as
4550 if it followed the point of the @code{.include}; when the end of the
4551 included file is reached, assembly of the original file continues. You
4552 can control the search paths used with the @samp{-I} command-line option
4553 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4557 @section @code{.int @var{expressions}}
4559 @cindex @code{int} directive
4560 @cindex integers, 32-bit
4561 Expect zero or more @var{expressions}, of any section, separated by commas.
4562 For each expression, emit a number that, at run time, is the value of that
4563 expression. The byte order and bit size of the number depends on what kind
4564 of target the assembly is for.
4568 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
4569 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4576 @section @code{.internal @var{names}}
4578 @cindex @code{internal} directive
4580 This is one of the ELF visibility directives. The other two are
4581 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4582 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4584 This directive overrides the named symbols default visibility (which is set by
4585 their binding: local, global or weak). The directive sets the visibility to
4586 @code{internal} which means that the symbols are considered to be @code{hidden}
4587 (i.e., not visible to other components), and that some extra, processor specific
4588 processing must also be performed upon the symbols as well.
4592 @section @code{.irp @var{symbol},@var{values}}@dots{}
4594 @cindex @code{irp} directive
4595 Evaluate a sequence of statements assigning different values to @var{symbol}.
4596 The sequence of statements starts at the @code{.irp} directive, and is
4597 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4598 set to @var{value}, and the sequence of statements is assembled. If no
4599 @var{value} is listed, the sequence of statements is assembled once, with
4600 @var{symbol} set to the null string. To refer to @var{symbol} within the
4601 sequence of statements, use @var{\symbol}.
4603 For example, assembling
4611 is equivalent to assembling
4619 For some caveats with the spelling of @var{symbol}, see also the discussion
4623 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4625 @cindex @code{irpc} directive
4626 Evaluate a sequence of statements assigning different values to @var{symbol}.
4627 The sequence of statements starts at the @code{.irpc} directive, and is
4628 terminated by an @code{.endr} directive. For each character in @var{value},
4629 @var{symbol} is set to the character, and the sequence of statements is
4630 assembled. If no @var{value} is listed, the sequence of statements is
4631 assembled once, with @var{symbol} set to the null string. To refer to
4632 @var{symbol} within the sequence of statements, use @var{\symbol}.
4634 For example, assembling
4642 is equivalent to assembling
4650 For some caveats with the spelling of @var{symbol}, see also the discussion
4654 @section @code{.lcomm @var{symbol} , @var{length}}
4656 @cindex @code{lcomm} directive
4657 @cindex local common symbols
4658 @cindex symbols, local common
4659 Reserve @var{length} (an absolute expression) bytes for a local common
4660 denoted by @var{symbol}. The section and value of @var{symbol} are
4661 those of the new local common. The addresses are allocated in the bss
4662 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4663 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4664 not visible to @code{@value{LD}}.
4667 Some targets permit a third argument to be used with @code{.lcomm}. This
4668 argument specifies the desired alignment of the symbol in the bss section.
4672 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4673 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4677 @section @code{.lflags}
4679 @cindex @code{lflags} directive (ignored)
4680 @command{@value{AS}} accepts this directive, for compatibility with other
4681 assemblers, but ignores it.
4683 @ifclear no-line-dir
4685 @section @code{.line @var{line-number}}
4687 @cindex @code{line} directive
4691 @section @code{.ln @var{line-number}}
4693 @cindex @code{ln} directive
4695 @cindex logical line number
4697 Change the logical line number. @var{line-number} must be an absolute
4698 expression. The next line has that logical line number. Therefore any other
4699 statements on the current line (after a statement separator character) are
4700 reported as on logical line number @var{line-number} @minus{} 1. One day
4701 @command{@value{AS}} will no longer support this directive: it is recognized only
4702 for compatibility with existing assembler programs.
4706 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
4707 not available; use the synonym @code{.ln} in that context.
4712 @ifclear no-line-dir
4713 Even though this is a directive associated with the @code{a.out} or
4714 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
4715 when producing COFF output, and treats @samp{.line} as though it
4716 were the COFF @samp{.ln} @emph{if} it is found outside a
4717 @code{.def}/@code{.endef} pair.
4719 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4720 used by compilers to generate auxiliary symbol information for
4725 @section @code{.linkonce [@var{type}]}
4727 @cindex @code{linkonce} directive
4728 @cindex common sections
4729 Mark the current section so that the linker only includes a single copy of it.
4730 This may be used to include the same section in several different object files,
4731 but ensure that the linker will only include it once in the final output file.
4732 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4733 Duplicate sections are detected based on the section name, so it should be
4736 This directive is only supported by a few object file formats; as of this
4737 writing, the only object file format which supports it is the Portable
4738 Executable format used on Windows NT.
4740 The @var{type} argument is optional. If specified, it must be one of the
4741 following strings. For example:
4745 Not all types may be supported on all object file formats.
4749 Silently discard duplicate sections. This is the default.
4752 Warn if there are duplicate sections, but still keep only one copy.
4755 Warn if any of the duplicates have different sizes.
4758 Warn if any of the duplicates do not have exactly the same contents.
4762 @section @code{.ln @var{line-number}}
4764 @cindex @code{ln} directive
4765 @ifclear no-line-dir
4766 @samp{.ln} is a synonym for @samp{.line}.
4769 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
4770 must be an absolute expression. The next line has that logical
4771 line number, so any other statements on the current line (after a
4772 statement separator character @code{;}) are reported as on logical
4773 line number @var{line-number} @minus{} 1.
4776 This directive is accepted, but ignored, when @command{@value{AS}} is
4777 configured for @code{b.out}; its effect is only associated with COFF
4783 @section @code{.mri @var{val}}
4785 @cindex @code{mri} directive
4786 @cindex MRI mode, temporarily
4787 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
4788 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
4789 affects code assembled until the next @code{.mri} directive, or until the end
4790 of the file. @xref{M, MRI mode, MRI mode}.
4793 @section @code{.list}
4795 @cindex @code{list} directive
4796 @cindex listing control, turning on
4797 Control (in conjunction with the @code{.nolist} directive) whether or
4798 not assembly listings are generated. These two directives maintain an
4799 internal counter (which is zero initially). @code{.list} increments the
4800 counter, and @code{.nolist} decrements it. Assembly listings are
4801 generated whenever the counter is greater than zero.
4803 By default, listings are disabled. When you enable them (with the
4804 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4805 the initial value of the listing counter is one.
4808 @section @code{.long @var{expressions}}
4810 @cindex @code{long} directive
4811 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
4814 @c no one seems to know what this is for or whether this description is
4815 @c what it really ought to do
4817 @section @code{.lsym @var{symbol}, @var{expression}}
4819 @cindex @code{lsym} directive
4820 @cindex symbol, not referenced in assembly
4821 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
4822 the hash table, ensuring it cannot be referenced by name during the
4823 rest of the assembly. This sets the attributes of the symbol to be
4824 the same as the expression value:
4826 @var{other} = @var{descriptor} = 0
4827 @var{type} = @r{(section of @var{expression})}
4828 @var{value} = @var{expression}
4831 The new symbol is not flagged as external.
4835 @section @code{.macro}
4838 The commands @code{.macro} and @code{.endm} allow you to define macros that
4839 generate assembly output. For example, this definition specifies a macro
4840 @code{sum} that puts a sequence of numbers into memory:
4843 .macro sum from=0, to=5
4852 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
4864 @item .macro @var{macname}
4865 @itemx .macro @var{macname} @var{macargs} @dots{}
4866 @cindex @code{macro} directive
4867 Begin the definition of a macro called @var{macname}. If your macro
4868 definition requires arguments, specify their names after the macro name,
4869 separated by commas or spaces. You can qualify the macro argument to
4870 indicate whether all invocations must specify a non-blank value (through
4871 @samp{:@code{req}}), or whether it takes all of the remaining arguments
4872 (through @samp{:@code{vararg}}). You can supply a default value for any
4873 macro argument by following the name with @samp{=@var{deflt}}. You
4874 cannot define two macros with the same @var{macname} unless it has been
4875 subject to the @code{.purgem} directive (@xref{Purgem}.) between the two
4876 definitions. For example, these are all valid @code{.macro} statements:
4880 Begin the definition of a macro called @code{comm}, which takes no
4883 @item .macro plus1 p, p1
4884 @itemx .macro plus1 p p1
4885 Either statement begins the definition of a macro called @code{plus1},
4886 which takes two arguments; within the macro definition, write
4887 @samp{\p} or @samp{\p1} to evaluate the arguments.
4889 @item .macro reserve_str p1=0 p2
4890 Begin the definition of a macro called @code{reserve_str}, with two
4891 arguments. The first argument has a default value, but not the second.
4892 After the definition is complete, you can call the macro either as
4893 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
4894 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
4895 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
4896 @samp{0}, and @samp{\p2} evaluating to @var{b}).
4899 @item .macro m p1:req, p2=0, p3:vararg
4900 Begin the definition of a macro called @code{m}, with at least three
4901 arguments. The first argument must always have a value specified, but
4902 not the second, which instead has a default value. The third formal
4903 will get assigned all remaining arguments specified at invocation time.
4905 When you call a macro, you can specify the argument values either by
4906 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
4907 @samp{sum to=17, from=9}.
4909 Note that since each of the @var{macargs} can be an identifier exactly
4910 as any other one permitted by the target architecture, there may be
4911 occasional problems if the target hand-crafts special meanings to certain
4912 characters when they occur in a special position. For example, if colon
4913 (@code{:}) is generally permitted to be part of a symbol name, but the
4914 architecture specific code special-cases it when occuring as the final
4915 character of a symbol (to denote a label), then the macro parameter
4916 replacement code will have no way of knowing that and consider the whole
4917 construct (including the colon) an identifier, and check only this
4918 identifier for being the subject to parameter substitution. In this
4919 example, besides the potential of just separating identifier and colon
4920 by white space, using alternate macro syntax (@xref{Altmacro}.) and
4921 ampersand (@code{&}) as the character to separate literal text from macro
4922 parameters (or macro parameters from one another) would provide a way to
4923 achieve the same effect:
4932 This applies identically to the identifiers used in @code{.irp} (@xref{Irp}.)
4933 and @code{.irpc} (@xref{Irpc}.).
4936 @cindex @code{endm} directive
4937 Mark the end of a macro definition.
4940 @cindex @code{exitm} directive
4941 Exit early from the current macro definition.
4943 @cindex number of macros executed
4944 @cindex macros, count executed
4946 @command{@value{AS}} maintains a counter of how many macros it has
4947 executed in this pseudo-variable; you can copy that number to your
4948 output with @samp{\@@}, but @emph{only within a macro definition}.
4950 @item LOCAL @var{name} [ , @dots{} ]
4951 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
4952 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
4953 @xref{Altmacro,,@code{.altmacro}}.
4957 @section @code{.altmacro}
4958 Enable alternate macro mode, enabling:
4961 @item LOCAL @var{name} [ , @dots{} ]
4962 One additional directive, @code{LOCAL}, is available. It is used to
4963 generate a string replacement for each of the @var{name} arguments, and
4964 replace any instances of @var{name} in each macro expansion. The
4965 replacement string is unique in the assembly, and different for each
4966 separate macro expansion. @code{LOCAL} allows you to write macros that
4967 define symbols, without fear of conflict between separate macro expansions.
4969 @item String delimiters
4970 You can write strings delimited in these other ways besides
4971 @code{"@var{string}"}:
4974 @item '@var{string}'
4975 You can delimit strings with single-quote charaters.
4977 @item <@var{string}>
4978 You can delimit strings with matching angle brackets.
4981 @item single-character string escape
4982 To include any single character literally in a string (even if the
4983 character would otherwise have some special meaning), you can prefix the
4984 character with @samp{!} (an exclamation mark). For example, you can
4985 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4987 @item Expression results as strings
4988 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4989 and use the result as a string.
4993 @section @code{.noaltmacro}
4994 Disable alternate macro mode. @ref{Altmacro}
4997 @section @code{.nolist}
4999 @cindex @code{nolist} directive
5000 @cindex listing control, turning off
5001 Control (in conjunction with the @code{.list} directive) whether or
5002 not assembly listings are generated. These two directives maintain an
5003 internal counter (which is zero initially). @code{.list} increments the
5004 counter, and @code{.nolist} decrements it. Assembly listings are
5005 generated whenever the counter is greater than zero.
5008 @section @code{.octa @var{bignums}}
5010 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5011 @cindex @code{octa} directive
5012 @cindex integer, 16-byte
5013 @cindex sixteen byte integer
5014 This directive expects zero or more bignums, separated by commas. For each
5015 bignum, it emits a 16-byte integer.
5017 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5018 hence @emph{octa}-word for 16 bytes.
5021 @section @code{.org @var{new-lc} , @var{fill}}
5023 @cindex @code{org} directive
5024 @cindex location counter, advancing
5025 @cindex advancing location counter
5026 @cindex current address, advancing
5027 Advance the location counter of the current section to
5028 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5029 expression with the same section as the current subsection. That is,
5030 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5031 wrong section, the @code{.org} directive is ignored. To be compatible
5032 with former assemblers, if the section of @var{new-lc} is absolute,
5033 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5034 is the same as the current subsection.
5036 @code{.org} may only increase the location counter, or leave it
5037 unchanged; you cannot use @code{.org} to move the location counter
5040 @c double negative used below "not undefined" because this is a specific
5041 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5042 @c section. doc@cygnus.com 18feb91
5043 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5044 may not be undefined. If you really detest this restriction we eagerly await
5045 a chance to share your improved assembler.
5047 Beware that the origin is relative to the start of the section, not
5048 to the start of the subsection. This is compatible with other
5049 people's assemblers.
5051 When the location counter (of the current subsection) is advanced, the
5052 intervening bytes are filled with @var{fill} which should be an
5053 absolute expression. If the comma and @var{fill} are omitted,
5054 @var{fill} defaults to zero.
5057 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5059 @cindex padding the location counter given a power of two
5060 @cindex @code{p2align} directive
5061 Pad the location counter (in the current subsection) to a particular
5062 storage boundary. The first expression (which must be absolute) is the
5063 number of low-order zero bits the location counter must have after
5064 advancement. For example @samp{.p2align 3} advances the location
5065 counter until it a multiple of 8. If the location counter is already a
5066 multiple of 8, no change is needed.
5068 The second expression (also absolute) gives the fill value to be stored in the
5069 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5070 padding bytes are normally zero. However, on some systems, if the section is
5071 marked as containing code and the fill value is omitted, the space is filled
5072 with no-op instructions.
5074 The third expression is also absolute, and is also optional. If it is present,
5075 it is the maximum number of bytes that should be skipped by this alignment
5076 directive. If doing the alignment would require skipping more bytes than the
5077 specified maximum, then the alignment is not done at all. You can omit the
5078 fill value (the second argument) entirely by simply using two commas after the
5079 required alignment; this can be useful if you want the alignment to be filled
5080 with no-op instructions when appropriate.
5082 @cindex @code{p2alignw} directive
5083 @cindex @code{p2alignl} directive
5084 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5085 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5086 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5087 fill pattern as a four byte longword value. For example, @code{.p2alignw
5088 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5089 filled in with the value 0x368d (the exact placement of the bytes depends upon
5090 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5095 @section @code{.previous}
5097 @cindex @code{previous} directive
5098 @cindex Section Stack
5099 This is one of the ELF section stack manipulation directives. The others are
5100 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5101 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5102 (@pxref{PopSection}).
5104 This directive swaps the current section (and subsection) with most recently
5105 referenced section (and subsection) prior to this one. Multiple
5106 @code{.previous} directives in a row will flip between two sections (and their
5109 In terms of the section stack, this directive swaps the current section with
5110 the top section on the section stack.
5115 @section @code{.popsection}
5117 @cindex @code{popsection} directive
5118 @cindex Section Stack
5119 This is one of the ELF section stack manipulation directives. The others are
5120 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5121 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5124 This directive replaces the current section (and subsection) with the top
5125 section (and subsection) on the section stack. This section is popped off the
5130 @section @code{.print @var{string}}
5132 @cindex @code{print} directive
5133 @command{@value{AS}} will print @var{string} on the standard output during
5134 assembly. You must put @var{string} in double quotes.
5138 @section @code{.protected @var{names}}
5140 @cindex @code{protected} directive
5142 This is one of the ELF visibility directives. The other two are
5143 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5145 This directive overrides the named symbols default visibility (which is set by
5146 their binding: local, global or weak). The directive sets the visibility to
5147 @code{protected} which means that any references to the symbols from within the
5148 components that defines them must be resolved to the definition in that
5149 component, even if a definition in another component would normally preempt
5154 @section @code{.psize @var{lines} , @var{columns}}
5156 @cindex @code{psize} directive
5157 @cindex listing control: paper size
5158 @cindex paper size, for listings
5159 Use this directive to declare the number of lines---and, optionally, the
5160 number of columns---to use for each page, when generating listings.
5162 If you do not use @code{.psize}, listings use a default line-count
5163 of 60. You may omit the comma and @var{columns} specification; the
5164 default width is 200 columns.
5166 @command{@value{AS}} generates formfeeds whenever the specified number of
5167 lines is exceeded (or whenever you explicitly request one, using
5170 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5171 those explicitly specified with @code{.eject}.
5174 @section @code{.purgem @var{name}}
5176 @cindex @code{purgem} directive
5177 Undefine the macro @var{name}, so that later uses of the string will not be
5178 expanded. @xref{Macro}.
5182 @section @code{.pushsection @var{name} , @var{subsection}}
5184 @cindex @code{pushsection} directive
5185 @cindex Section Stack
5186 This is one of the ELF section stack manipulation directives. The others are
5187 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5188 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5191 This directive pushes the current section (and subsection) onto the
5192 top of the section stack, and then replaces the current section and
5193 subsection with @code{name} and @code{subsection}.
5197 @section @code{.quad @var{bignums}}
5199 @cindex @code{quad} directive
5200 @code{.quad} expects zero or more bignums, separated by commas. For
5201 each bignum, it emits
5203 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5204 warning message; and just takes the lowest order 8 bytes of the bignum.
5205 @cindex eight-byte integer
5206 @cindex integer, 8-byte
5208 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5209 hence @emph{quad}-word for 8 bytes.
5212 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5213 warning message; and just takes the lowest order 16 bytes of the bignum.
5214 @cindex sixteen-byte integer
5215 @cindex integer, 16-byte
5219 @section @code{.rept @var{count}}
5221 @cindex @code{rept} directive
5222 Repeat the sequence of lines between the @code{.rept} directive and the next
5223 @code{.endr} directive @var{count} times.
5225 For example, assembling
5233 is equivalent to assembling
5242 @section @code{.sbttl "@var{subheading}"}
5244 @cindex @code{sbttl} directive
5245 @cindex subtitles for listings
5246 @cindex listing control: subtitle
5247 Use @var{subheading} as the title (third line, immediately after the
5248 title line) when generating assembly listings.
5250 This directive affects subsequent pages, as well as the current page if
5251 it appears within ten lines of the top of a page.
5255 @section @code{.scl @var{class}}
5257 @cindex @code{scl} directive
5258 @cindex symbol storage class (COFF)
5259 @cindex COFF symbol storage class
5260 Set the storage-class value for a symbol. This directive may only be
5261 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5262 whether a symbol is static or external, or it may record further
5263 symbolic debugging information.
5266 The @samp{.scl} directive is primarily associated with COFF output; when
5267 configured to generate @code{b.out} output format, @command{@value{AS}}
5268 accepts this directive but ignores it.
5274 @section @code{.section @var{name}}
5276 @cindex named section
5277 Use the @code{.section} directive to assemble the following code into a section
5280 This directive is only supported for targets that actually support arbitrarily
5281 named sections; on @code{a.out} targets, for example, it is not accepted, even
5282 with a standard @code{a.out} section name.
5286 @c only print the extra heading if both COFF and ELF are set
5287 @subheading COFF Version
5290 @cindex @code{section} directive (COFF version)
5291 For COFF targets, the @code{.section} directive is used in one of the following
5295 .section @var{name}[, "@var{flags}"]
5296 .section @var{name}[, @var{subsegment}]
5299 If the optional argument is quoted, it is taken as flags to use for the
5300 section. Each flag is a single character. The following flags are recognized:
5303 bss section (uninitialized data)
5305 section is not loaded
5315 shared section (meaningful for PE targets)
5317 ignored. (For compatibility with the ELF version)
5320 If no flags are specified, the default flags depend upon the section name. If
5321 the section name is not recognized, the default will be for the section to be
5322 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5323 from the section, rather than adding them, so if they are used on their own it
5324 will be as if no flags had been specified at all.
5326 If the optional argument to the @code{.section} directive is not quoted, it is
5327 taken as a subsegment number (@pxref{Sub-Sections}).
5332 @c only print the extra heading if both COFF and ELF are set
5333 @subheading ELF Version
5336 @cindex Section Stack
5337 This is one of the ELF section stack manipulation directives. The others are
5338 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5339 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5340 @code{.previous} (@pxref{Previous}).
5342 @cindex @code{section} directive (ELF version)
5343 For ELF targets, the @code{.section} directive is used like this:
5346 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]
5349 The optional @var{flags} argument is a quoted string which may contain any
5350 combination of the following characters:
5353 section is allocatable
5357 section is executable
5359 section is mergeable
5361 section contains zero terminated strings
5363 section is a member of a section group
5365 section is used for thread-local-storage
5368 The optional @var{type} argument may contain one of the following constants:
5371 section contains data
5373 section does not contain data (i.e., section only occupies space)
5375 section contains data which is used by things other than the program
5377 section contains an array of pointers to init functions
5379 section contains an array of pointers to finish functions
5380 @item @@preinit_array
5381 section contains an array of pointers to pre-init functions
5384 Many targets only support the first three section types.
5386 Note on targets where the @code{@@} character is the start of a comment (eg
5387 ARM) then another character is used instead. For example the ARM port uses the
5390 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5391 be specified as well as an extra argument - @var{entsize} - like this:
5394 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5397 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5398 constants, each @var{entsize} octets long. Sections with both @code{M} and
5399 @code{S} must contain zero terminated strings where each character is
5400 @var{entsize} bytes long. The linker may remove duplicates within sections with
5401 the same name, same entity size and same flags. @var{entsize} must be an
5402 absolute expression.
5404 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5405 be present along with an additional field like this:
5408 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5411 The @var{GroupName} field specifies the name of the section group to which this
5412 particular section belongs. The optional linkage field can contain:
5415 indicates that only one copy of this section should be retained
5420 Note - if both the @var{M} and @var{G} flags are present then the fields for
5421 the Merge flag should come first, like this:
5424 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5427 If no flags are specified, the default flags depend upon the section name. If
5428 the section name is not recognized, the default will be for the section to have
5429 none of the above flags: it will not be allocated in memory, nor writable, nor
5430 executable. The section will contain data.
5432 For ELF targets, the assembler supports another type of @code{.section}
5433 directive for compatibility with the Solaris assembler:
5436 .section "@var{name}"[, @var{flags}...]
5439 Note that the section name is quoted. There may be a sequence of comma
5443 section is allocatable
5447 section is executable
5449 section is used for thread local storage
5452 This directive replaces the current section and subsection. See the
5453 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
5454 some examples of how this directive and the other section stack directives
5460 @section @code{.set @var{symbol}, @var{expression}}
5462 @cindex @code{set} directive
5463 @cindex symbol value, setting
5464 Set the value of @var{symbol} to @var{expression}. This
5465 changes @var{symbol}'s value and type to conform to
5466 @var{expression}. If @var{symbol} was flagged as external, it remains
5467 flagged (@pxref{Symbol Attributes}).
5469 You may @code{.set} a symbol many times in the same assembly.
5471 If you @code{.set} a global symbol, the value stored in the object
5472 file is the last value stored into it.
5475 The syntax for @code{set} on the HPPA is
5476 @samp{@var{symbol} .set @var{expression}}.
5480 @section @code{.short @var{expressions}}
5482 @cindex @code{short} directive
5484 @code{.short} is normally the same as @samp{.word}.
5485 @xref{Word,,@code{.word}}.
5487 In some configurations, however, @code{.short} and @code{.word} generate
5488 numbers of different lengths; @pxref{Machine Dependencies}.
5492 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5495 This expects zero or more @var{expressions}, and emits
5496 a 16 bit number for each.
5501 @section @code{.single @var{flonums}}
5503 @cindex @code{single} directive
5504 @cindex floating point numbers (single)
5505 This directive assembles zero or more flonums, separated by commas. It
5506 has the same effect as @code{.float}.
5508 The exact kind of floating point numbers emitted depends on how
5509 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5513 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5514 numbers in @sc{ieee} format.
5520 @section @code{.size}
5522 This directive is used to set the size associated with a symbol.
5526 @c only print the extra heading if both COFF and ELF are set
5527 @subheading COFF Version
5530 @cindex @code{size} directive (COFF version)
5531 For COFF targets, the @code{.size} directive is only permitted inside
5532 @code{.def}/@code{.endef} pairs. It is used like this:
5535 .size @var{expression}
5539 @samp{.size} is only meaningful when generating COFF format output; when
5540 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5547 @c only print the extra heading if both COFF and ELF are set
5548 @subheading ELF Version
5551 @cindex @code{size} directive (ELF version)
5552 For ELF targets, the @code{.size} directive is used like this:
5555 .size @var{name} , @var{expression}
5558 This directive sets the size associated with a symbol @var{name}.
5559 The size in bytes is computed from @var{expression} which can make use of label
5560 arithmetic. This directive is typically used to set the size of function
5566 @section @code{.sleb128 @var{expressions}}
5568 @cindex @code{sleb128} directive
5569 @var{sleb128} stands for ``signed little endian base 128.'' This is a
5570 compact, variable length representation of numbers used by the DWARF
5571 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
5573 @ifclear no-space-dir
5575 @section @code{.skip @var{size} , @var{fill}}
5577 @cindex @code{skip} directive
5578 @cindex filling memory
5579 This directive emits @var{size} bytes, each of value @var{fill}. Both
5580 @var{size} and @var{fill} are absolute expressions. If the comma and
5581 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
5585 @section @code{.space @var{size} , @var{fill}}
5587 @cindex @code{space} directive
5588 @cindex filling memory
5589 This directive emits @var{size} bytes, each of value @var{fill}. Both
5590 @var{size} and @var{fill} are absolute expressions. If the comma
5591 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
5596 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
5597 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
5598 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
5599 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
5608 @section @code{.space}
5609 @cindex @code{space} directive
5611 On the AMD 29K, this directive is ignored; it is accepted for
5612 compatibility with other AMD 29K assemblers.
5615 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
5616 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
5622 @section @code{.stabd, .stabn, .stabs}
5624 @cindex symbolic debuggers, information for
5625 @cindex @code{stab@var{x}} directives
5626 There are three directives that begin @samp{.stab}.
5627 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
5628 The symbols are not entered in the @command{@value{AS}} hash table: they
5629 cannot be referenced elsewhere in the source file.
5630 Up to five fields are required:
5634 This is the symbol's name. It may contain any character except
5635 @samp{\000}, so is more general than ordinary symbol names. Some
5636 debuggers used to code arbitrarily complex structures into symbol names
5640 An absolute expression. The symbol's type is set to the low 8 bits of
5641 this expression. Any bit pattern is permitted, but @code{@value{LD}}
5642 and debuggers choke on silly bit patterns.
5645 An absolute expression. The symbol's ``other'' attribute is set to the
5646 low 8 bits of this expression.
5649 An absolute expression. The symbol's descriptor is set to the low 16
5650 bits of this expression.
5653 An absolute expression which becomes the symbol's value.
5656 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
5657 or @code{.stabs} statement, the symbol has probably already been created;
5658 you get a half-formed symbol in your object file. This is
5659 compatible with earlier assemblers!
5662 @cindex @code{stabd} directive
5663 @item .stabd @var{type} , @var{other} , @var{desc}
5665 The ``name'' of the symbol generated is not even an empty string.
5666 It is a null pointer, for compatibility. Older assemblers used a
5667 null pointer so they didn't waste space in object files with empty
5670 The symbol's value is set to the location counter,
5671 relocatably. When your program is linked, the value of this symbol
5672 is the address of the location counter when the @code{.stabd} was
5675 @cindex @code{stabn} directive
5676 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
5677 The name of the symbol is set to the empty string @code{""}.
5679 @cindex @code{stabs} directive
5680 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
5681 All five fields are specified.
5687 @section @code{.string} "@var{str}"
5689 @cindex string, copying to object file
5690 @cindex @code{string} directive
5692 Copy the characters in @var{str} to the object file. You may specify more than
5693 one string to copy, separated by commas. Unless otherwise specified for a
5694 particular machine, the assembler marks the end of each string with a 0 byte.
5695 You can use any of the escape sequences described in @ref{Strings,,Strings}.
5698 @section @code{.struct @var{expression}}
5700 @cindex @code{struct} directive
5701 Switch to the absolute section, and set the section offset to @var{expression},
5702 which must be an absolute expression. You might use this as follows:
5711 This would define the symbol @code{field1} to have the value 0, the symbol
5712 @code{field2} to have the value 4, and the symbol @code{field3} to have the
5713 value 8. Assembly would be left in the absolute section, and you would need to
5714 use a @code{.section} directive of some sort to change to some other section
5715 before further assembly.
5719 @section @code{.subsection @var{name}}
5721 @cindex @code{subsection} directive
5722 @cindex Section Stack
5723 This is one of the ELF section stack manipulation directives. The others are
5724 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
5725 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5728 This directive replaces the current subsection with @code{name}. The current
5729 section is not changed. The replaced subsection is put onto the section stack
5730 in place of the then current top of stack subsection.
5735 @section @code{.symver}
5736 @cindex @code{symver} directive
5737 @cindex symbol versioning
5738 @cindex versions of symbols
5739 Use the @code{.symver} directive to bind symbols to specific version nodes
5740 within a source file. This is only supported on ELF platforms, and is
5741 typically used when assembling files to be linked into a shared library.
5742 There are cases where it may make sense to use this in objects to be bound
5743 into an application itself so as to override a versioned symbol from a
5746 For ELF targets, the @code{.symver} directive can be used like this:
5748 .symver @var{name}, @var{name2@@nodename}
5750 If the symbol @var{name} is defined within the file
5751 being assembled, the @code{.symver} directive effectively creates a symbol
5752 alias with the name @var{name2@@nodename}, and in fact the main reason that we
5753 just don't try and create a regular alias is that the @var{@@} character isn't
5754 permitted in symbol names. The @var{name2} part of the name is the actual name
5755 of the symbol by which it will be externally referenced. The name @var{name}
5756 itself is merely a name of convenience that is used so that it is possible to
5757 have definitions for multiple versions of a function within a single source
5758 file, and so that the compiler can unambiguously know which version of a
5759 function is being mentioned. The @var{nodename} portion of the alias should be
5760 the name of a node specified in the version script supplied to the linker when
5761 building a shared library. If you are attempting to override a versioned
5762 symbol from a shared library, then @var{nodename} should correspond to the
5763 nodename of the symbol you are trying to override.
5765 If the symbol @var{name} is not defined within the file being assembled, all
5766 references to @var{name} will be changed to @var{name2@@nodename}. If no
5767 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
5770 Another usage of the @code{.symver} directive is:
5772 .symver @var{name}, @var{name2@@@@nodename}
5774 In this case, the symbol @var{name} must exist and be defined within
5775 the file being assembled. It is similar to @var{name2@@nodename}. The
5776 difference is @var{name2@@@@nodename} will also be used to resolve
5777 references to @var{name2} by the linker.
5779 The third usage of the @code{.symver} directive is:
5781 .symver @var{name}, @var{name2@@@@@@nodename}
5783 When @var{name} is not defined within the
5784 file being assembled, it is treated as @var{name2@@nodename}. When
5785 @var{name} is defined within the file being assembled, the symbol
5786 name, @var{name}, will be changed to @var{name2@@@@nodename}.
5791 @section @code{.tag @var{structname}}
5793 @cindex COFF structure debugging
5794 @cindex structure debugging, COFF
5795 @cindex @code{tag} directive
5796 This directive is generated by compilers to include auxiliary debugging
5797 information in the symbol table. It is only permitted inside
5798 @code{.def}/@code{.endef} pairs. Tags are used to link structure
5799 definitions in the symbol table with instances of those structures.
5802 @samp{.tag} is only used when generating COFF format output; when
5803 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5809 @section @code{.text @var{subsection}}
5811 @cindex @code{text} directive
5812 Tells @command{@value{AS}} to assemble the following statements onto the end of
5813 the text subsection numbered @var{subsection}, which is an absolute
5814 expression. If @var{subsection} is omitted, subsection number zero
5818 @section @code{.title "@var{heading}"}
5820 @cindex @code{title} directive
5821 @cindex listing control: title line
5822 Use @var{heading} as the title (second line, immediately after the
5823 source file name and pagenumber) when generating assembly listings.
5825 This directive affects subsequent pages, as well as the current page if
5826 it appears within ten lines of the top of a page.
5830 @section @code{.type}
5832 This directive is used to set the type of a symbol.
5836 @c only print the extra heading if both COFF and ELF are set
5837 @subheading COFF Version
5840 @cindex COFF symbol type
5841 @cindex symbol type, COFF
5842 @cindex @code{type} directive (COFF version)
5843 For COFF targets, this directive is permitted only within
5844 @code{.def}/@code{.endef} pairs. It is used like this:
5850 This records the integer @var{int} as the type attribute of a symbol table
5854 @samp{.type} is associated only with COFF format output; when
5855 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
5856 directive but ignores it.
5862 @c only print the extra heading if both COFF and ELF are set
5863 @subheading ELF Version
5866 @cindex ELF symbol type
5867 @cindex symbol type, ELF
5868 @cindex @code{type} directive (ELF version)
5869 For ELF targets, the @code{.type} directive is used like this:
5872 .type @var{name} , @var{type description}
5875 This sets the type of symbol @var{name} to be either a
5876 function symbol or an object symbol. There are five different syntaxes
5877 supported for the @var{type description} field, in order to provide
5878 compatibility with various other assemblers. The syntaxes supported are:
5881 .type <name>,#function
5882 .type <name>,#object
5884 .type <name>,@@function
5885 .type <name>,@@object
5887 .type <name>,%function
5888 .type <name>,%object
5890 .type <name>,"function"
5891 .type <name>,"object"
5893 .type <name> STT_FUNCTION
5894 .type <name> STT_OBJECT
5900 @section @code{.uleb128 @var{expressions}}
5902 @cindex @code{uleb128} directive
5903 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
5904 compact, variable length representation of numbers used by the DWARF
5905 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
5909 @section @code{.val @var{addr}}
5911 @cindex @code{val} directive
5912 @cindex COFF value attribute
5913 @cindex value attribute, COFF
5914 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5915 records the address @var{addr} as the value attribute of a symbol table
5919 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
5920 configured for @code{b.out}, it accepts this directive but ignores it.
5926 @section @code{.version "@var{string}"}
5928 @cindex @code{version} directive
5929 This directive creates a @code{.note} section and places into it an ELF
5930 formatted note of type NT_VERSION. The note's name is set to @code{string}.
5935 @section @code{.vtable_entry @var{table}, @var{offset}}
5937 @cindex @code{vtable_entry} directive
5938 This directive finds or creates a symbol @code{table} and creates a
5939 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
5942 @section @code{.vtable_inherit @var{child}, @var{parent}}
5944 @cindex @code{vtable_inherit} directive
5945 This directive finds the symbol @code{child} and finds or creates the symbol
5946 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
5947 parent whose addend is the value of the child symbol. As a special case the
5948 parent name of @code{0} is treated as refering the @code{*ABS*} section.
5952 @section @code{.warning "@var{string}"}
5953 @cindex warning directive
5954 Similar to the directive @code{.error}
5955 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
5958 @section @code{.weak @var{names}}
5960 @cindex @code{weak} directive
5961 This directive sets the weak attribute on the comma separated list of symbol
5962 @code{names}. If the symbols do not already exist, they will be created.
5964 On COFF targets other than PE, weak symbols are a GNU extension. This
5965 directive sets the weak attribute on the comma separated list of symbol
5966 @code{names}. If the symbols do not already exist, they will be created.
5968 On the PE target, weak symbols are supported natively as weak aliases.
5969 When a weak symbol is created that is not an alias, GAS creates an
5970 alternate symbol to hold the default value.
5973 @section @code{.word @var{expressions}}
5975 @cindex @code{word} directive
5976 This directive expects zero or more @var{expressions}, of any section,
5977 separated by commas.
5980 For each expression, @command{@value{AS}} emits a 32-bit number.
5983 For each expression, @command{@value{AS}} emits a 16-bit number.
5988 The size of the number emitted, and its byte order,
5989 depend on what target computer the assembly is for.
5992 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
5993 @c happen---32-bit addressability, period; no long/short jumps.
5994 @ifset DIFF-TBL-KLUGE
5995 @cindex difference tables altered
5996 @cindex altered difference tables
5998 @emph{Warning: Special Treatment to support Compilers}
6002 Machines with a 32-bit address space, but that do less than 32-bit
6003 addressing, require the following special treatment. If the machine of
6004 interest to you does 32-bit addressing (or doesn't require it;
6005 @pxref{Machine Dependencies}), you can ignore this issue.
6008 In order to assemble compiler output into something that works,
6009 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6010 Directives of the form @samp{.word sym1-sym2} are often emitted by
6011 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6012 directive of the form @samp{.word sym1-sym2}, and the difference between
6013 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6014 creates a @dfn{secondary jump table}, immediately before the next label.
6015 This secondary jump table is preceded by a short-jump to the
6016 first byte after the secondary table. This short-jump prevents the flow
6017 of control from accidentally falling into the new table. Inside the
6018 table is a long-jump to @code{sym2}. The original @samp{.word}
6019 contains @code{sym1} minus the address of the long-jump to
6022 If there were several occurrences of @samp{.word sym1-sym2} before the
6023 secondary jump table, all of them are adjusted. If there was a
6024 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6025 long-jump to @code{sym4} is included in the secondary jump table,
6026 and the @code{.word} directives are adjusted to contain @code{sym3}
6027 minus the address of the long-jump to @code{sym4}; and so on, for as many
6028 entries in the original jump table as necessary.
6031 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6032 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6033 assembly language programmers.
6036 @c end DIFF-TBL-KLUGE
6039 @section Deprecated Directives
6041 @cindex deprecated directives
6042 @cindex obsolescent directives
6043 One day these directives won't work.
6044 They are included for compatibility with older assemblers.
6051 @node Machine Dependencies
6052 @chapter Machine Dependent Features
6054 @cindex machine dependencies
6055 The machine instruction sets are (almost by definition) different on
6056 each machine where @command{@value{AS}} runs. Floating point representations
6057 vary as well, and @command{@value{AS}} often supports a few additional
6058 directives or command-line options for compatibility with other
6059 assemblers on a particular platform. Finally, some versions of
6060 @command{@value{AS}} support special pseudo-instructions for branch
6063 This chapter discusses most of these differences, though it does not
6064 include details on any machine's instruction set. For details on that
6065 subject, see the hardware manufacturer's manual.
6069 * AMD29K-Dependent:: AMD 29K Dependent Features
6072 * Alpha-Dependent:: Alpha Dependent Features
6075 * ARC-Dependent:: ARC Dependent Features
6078 * ARM-Dependent:: ARM Dependent Features
6081 * CRIS-Dependent:: CRIS Dependent Features
6084 * D10V-Dependent:: D10V Dependent Features
6087 * D30V-Dependent:: D30V Dependent Features
6090 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6093 * H8/500-Dependent:: Renesas H8/500 Dependent Features
6096 * HPPA-Dependent:: HPPA Dependent Features
6099 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6102 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6105 * i860-Dependent:: Intel 80860 Dependent Features
6108 * i960-Dependent:: Intel 80960 Dependent Features
6111 * IA-64-Dependent:: Intel IA-64 Dependent Features
6114 * IP2K-Dependent:: IP2K Dependent Features
6117 * M32R-Dependent:: M32R Dependent Features
6120 * M68K-Dependent:: M680x0 Dependent Features
6123 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6126 * M88K-Dependent:: M880x0 Dependent Features
6129 * MIPS-Dependent:: MIPS Dependent Features
6132 * MMIX-Dependent:: MMIX Dependent Features
6135 * MSP430-Dependent:: MSP430 Dependent Features
6138 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6139 * SH64-Dependent:: SuperH SH64 Dependent Features
6142 * PDP-11-Dependent:: PDP-11 Dependent Features
6145 * PJ-Dependent:: picoJava Dependent Features
6148 * PPC-Dependent:: PowerPC Dependent Features
6151 * Sparc-Dependent:: SPARC Dependent Features
6154 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6157 * V850-Dependent:: V850 Dependent Features
6160 * Xtensa-Dependent:: Xtensa Dependent Features
6163 * Z8000-Dependent:: Z8000 Dependent Features
6166 * Vax-Dependent:: VAX Dependent Features
6173 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6174 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6175 @c peculiarity: to preserve cross-references, there must be a node called
6176 @c "Machine Dependencies". Hence the conditional nodenames in each
6177 @c major node below. Node defaulting in makeinfo requires adjacency of
6178 @c node and sectioning commands; hence the repetition of @chapter BLAH
6179 @c in both conditional blocks.
6182 @include c-a29k.texi
6186 @include c-alpha.texi
6198 @include c-cris.texi
6203 @node Machine Dependencies
6204 @chapter Machine Dependent Features
6206 The machine instruction sets are different on each Renesas chip family,
6207 and there are also some syntax differences among the families. This
6208 chapter describes the specific @command{@value{AS}} features for each
6212 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6213 * H8/500-Dependent:: Renesas H8/500 Dependent Features
6214 * SH-Dependent:: Renesas SH Dependent Features
6221 @include c-d10v.texi
6225 @include c-d30v.texi
6229 @include c-h8300.texi
6233 @include c-h8500.texi
6237 @include c-hppa.texi
6241 @include c-i370.texi
6245 @include c-i386.texi
6249 @include c-i860.texi
6253 @include c-i960.texi
6257 @include c-ia64.texi
6261 @include c-ip2k.texi
6265 @include c-m32r.texi
6269 @include c-m68k.texi
6273 @include c-m68hc11.texi
6277 @include c-m88k.texi
6281 @include c-mips.texi
6285 @include c-mmix.texi
6289 @include c-msp430.texi
6293 @include c-ns32k.texi
6297 @include c-pdp11.texi
6310 @include c-sh64.texi
6314 @include c-sparc.texi
6318 @include c-tic54x.texi
6330 @include c-v850.texi
6334 @include c-xtensa.texi
6338 @c reverse effect of @down at top of generic Machine-Dep chapter
6342 @node Reporting Bugs
6343 @chapter Reporting Bugs
6344 @cindex bugs in assembler
6345 @cindex reporting bugs in assembler
6347 Your bug reports play an essential role in making @command{@value{AS}} reliable.
6349 Reporting a bug may help you by bringing a solution to your problem, or it may
6350 not. But in any case the principal function of a bug report is to help the
6351 entire community by making the next version of @command{@value{AS}} work better.
6352 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
6354 In order for a bug report to serve its purpose, you must include the
6355 information that enables us to fix the bug.
6358 * Bug Criteria:: Have you found a bug?
6359 * Bug Reporting:: How to report bugs
6363 @section Have You Found a Bug?
6364 @cindex bug criteria
6366 If you are not sure whether you have found a bug, here are some guidelines:
6369 @cindex fatal signal
6370 @cindex assembler crash
6371 @cindex crash of assembler
6373 If the assembler gets a fatal signal, for any input whatever, that is a
6374 @command{@value{AS}} bug. Reliable assemblers never crash.
6376 @cindex error on valid input
6378 If @command{@value{AS}} produces an error message for valid input, that is a bug.
6380 @cindex invalid input
6382 If @command{@value{AS}} does not produce an error message for invalid input, that
6383 is a bug. However, you should note that your idea of ``invalid input'' might
6384 be our idea of ``an extension'' or ``support for traditional practice''.
6387 If you are an experienced user of assemblers, your suggestions for improvement
6388 of @command{@value{AS}} are welcome in any case.
6392 @section How to Report Bugs
6394 @cindex assembler bugs, reporting
6396 A number of companies and individuals offer support for @sc{gnu} products. If
6397 you obtained @command{@value{AS}} from a support organization, we recommend you
6398 contact that organization first.
6400 You can find contact information for many support companies and
6401 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
6404 In any event, we also recommend that you send bug reports for @command{@value{AS}}
6405 to @samp{bug-binutils@@gnu.org}.
6407 The fundamental principle of reporting bugs usefully is this:
6408 @strong{report all the facts}. If you are not sure whether to state a
6409 fact or leave it out, state it!
6411 Often people omit facts because they think they know what causes the problem
6412 and assume that some details do not matter. Thus, you might assume that the
6413 name of a symbol you use in an example does not matter. Well, probably it does
6414 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
6415 happens to fetch from the location where that name is stored in memory;
6416 perhaps, if the name were different, the contents of that location would fool
6417 the assembler into doing the right thing despite the bug. Play it safe and
6418 give a specific, complete example. That is the easiest thing for you to do,
6419 and the most helpful.
6421 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
6422 it is new to us. Therefore, always write your bug reports on the assumption
6423 that the bug has not been reported previously.
6425 Sometimes people give a few sketchy facts and ask, ``Does this ring a
6426 bell?'' This cannot help us fix a bug, so it is basically useless. We
6427 respond by asking for enough details to enable us to investigate.
6428 You might as well expedite matters by sending them to begin with.
6430 To enable us to fix the bug, you should include all these things:
6434 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
6435 it with the @samp{--version} argument.
6437 Without this, we will not know whether there is any point in looking for
6438 the bug in the current version of @command{@value{AS}}.
6441 Any patches you may have applied to the @command{@value{AS}} source.
6444 The type of machine you are using, and the operating system name and
6448 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
6452 The command arguments you gave the assembler to assemble your example and
6453 observe the bug. To guarantee you will not omit something important, list them
6454 all. A copy of the Makefile (or the output from make) is sufficient.
6456 If we were to try to guess the arguments, we would probably guess wrong
6457 and then we might not encounter the bug.
6460 A complete input file that will reproduce the bug. If the bug is observed when
6461 the assembler is invoked via a compiler, send the assembler source, not the
6462 high level language source. Most compilers will produce the assembler source
6463 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
6464 the options @samp{-v --save-temps}; this will save the assembler source in a
6465 file with an extension of @file{.s}, and also show you exactly how
6466 @command{@value{AS}} is being run.
6469 A description of what behavior you observe that you believe is
6470 incorrect. For example, ``It gets a fatal signal.''
6472 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
6473 will certainly notice it. But if the bug is incorrect output, we might not
6474 notice unless it is glaringly wrong. You might as well not give us a chance to
6477 Even if the problem you experience is a fatal signal, you should still say so
6478 explicitly. Suppose something strange is going on, such as, your copy of
6479 @command{@value{AS}} is out of synch, or you have encountered a bug in the C
6480 library on your system. (This has happened!) Your copy might crash and ours
6481 would not. If you told us to expect a crash, then when ours fails to crash, we
6482 would know that the bug was not happening for us. If you had not told us to
6483 expect a crash, then we would not be able to draw any conclusion from our
6487 If you wish to suggest changes to the @command{@value{AS}} source, send us context
6488 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
6489 option. Always send diffs from the old file to the new file. If you even
6490 discuss something in the @command{@value{AS}} source, refer to it by context, not
6493 The line numbers in our development sources will not match those in your
6494 sources. Your line numbers would convey no useful information to us.
6497 Here are some things that are not necessary:
6501 A description of the envelope of the bug.
6503 Often people who encounter a bug spend a lot of time investigating
6504 which changes to the input file will make the bug go away and which
6505 changes will not affect it.
6507 This is often time consuming and not very useful, because the way we
6508 will find the bug is by running a single example under the debugger
6509 with breakpoints, not by pure deduction from a series of examples.
6510 We recommend that you save your time for something else.
6512 Of course, if you can find a simpler example to report @emph{instead}
6513 of the original one, that is a convenience for us. Errors in the
6514 output will be easier to spot, running under the debugger will take
6515 less time, and so on.
6517 However, simplification is not vital; if you do not want to do this,
6518 report the bug anyway and send us the entire test case you used.
6521 A patch for the bug.
6523 A patch for the bug does help us if it is a good one. But do not omit
6524 the necessary information, such as the test case, on the assumption that
6525 a patch is all we need. We might see problems with your patch and decide
6526 to fix the problem another way, or we might not understand it at all.
6528 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
6529 construct an example that will make the program follow a certain path through
6530 the code. If you do not send us the example, we will not be able to construct
6531 one, so we will not be able to verify that the bug is fixed.
6533 And if we cannot understand what bug you are trying to fix, or why your
6534 patch should be an improvement, we will not install it. A test case will
6535 help us to understand.
6538 A guess about what the bug is or what it depends on.
6540 Such guesses are usually wrong. Even we cannot guess right about such
6541 things without first using the debugger to find the facts.
6544 @node Acknowledgements
6545 @chapter Acknowledgements
6547 If you have contributed to GAS and your name isn't listed here,
6548 it is not meant as a slight. We just don't know about it. Send mail to the
6549 maintainer, and we'll correct the situation. Currently
6551 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
6553 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
6556 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
6557 information and the 68k series machines, most of the preprocessing pass, and
6558 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
6560 K. Richard Pixley maintained GAS for a while, adding various enhancements and
6561 many bug fixes, including merging support for several processors, breaking GAS
6562 up to handle multiple object file format back ends (including heavy rewrite,
6563 testing, an integration of the coff and b.out back ends), adding configuration
6564 including heavy testing and verification of cross assemblers and file splits
6565 and renaming, converted GAS to strictly ANSI C including full prototypes, added
6566 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
6567 port (including considerable amounts of reverse engineering), a SPARC opcode
6568 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
6569 assertions and made them work, much other reorganization, cleanup, and lint.
6571 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
6572 in format-specific I/O modules.
6574 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
6575 has done much work with it since.
6577 The Intel 80386 machine description was written by Eliot Dresselhaus.
6579 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
6581 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
6582 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
6584 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
6585 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
6586 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
6587 support a.out format.
6589 Support for the Zilog Z8k and Renesas H8/300 and H8/500 processors (tc-z8k,
6590 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
6591 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
6592 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
6595 John Gilmore built the AMD 29000 support, added @code{.include} support, and
6596 simplified the configuration of which versions accept which directives. He
6597 updated the 68k machine description so that Motorola's opcodes always produced
6598 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
6599 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
6600 cross-compilation support, and one bug in relaxation that took a week and
6601 required the proverbial one-bit fix.
6603 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
6604 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
6605 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
6606 PowerPC assembler, and made a few other minor patches.
6608 Steve Chamberlain made GAS able to generate listings.
6610 Hewlett-Packard contributed support for the HP9000/300.
6612 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
6613 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
6614 formats). This work was supported by both the Center for Software Science at
6615 the University of Utah and Cygnus Support.
6617 Support for ELF format files has been worked on by Mark Eichin of Cygnus
6618 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
6619 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
6620 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
6621 and some initial 64-bit support).
6623 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
6625 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
6626 support for openVMS/Alpha.
6628 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
6631 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
6632 Inc. added support for Xtensa processors.
6634 Several engineers at Cygnus Support have also provided many small bug fixes and
6635 configuration enhancements.
6637 Many others have contributed large or small bugfixes and enhancements. If
6638 you have contributed significant work and are not mentioned on this list, and
6639 want to be, let us know. Some of the history has been lost; we are not
6640 intentionally leaving anyone out.