1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
3 @c 2001, 2002, 2003, 2004
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}] [@b{--gstabs+}]
231 [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}] [@b{-K}] [@b{-L}]
232 [@b{--listing-lhs-width}=@var{NUM}] [@b{--listing-lhs-width2}=@var{NUM}]
233 [@b{--listing-rhs-width}=@var{NUM}] [@b{--listing-cont-lines}=@var{NUM}]
234 [@b{--keep-locals}] [@b{-o} @var{objfile}] [@b{-R}] [@b{--statistics}] [@b{-v}]
235 [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}] [@b{--fatal-warnings}]
236 [@b{-w}] [@b{-x}] [@b{-Z}] [@b{--target-help}] [@var{target-options}]
237 [@b{--}|@var{files} @dots{}]
239 @c Target dependent options are listed below. Keep the list sorted.
240 @c Add an empty line for separation.
242 @c am29k has no machine-dependent assembler options
246 @emph{Target Alpha options:}
248 [@b{-mdebug} | @b{-no-mdebug}]
249 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
250 [@b{-F}] [@b{-32addr}]
254 @emph{Target ARC options:}
260 @emph{Target ARM options:}
261 @c Don't document the deprecated options
262 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
263 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
264 [@b{-mfpu}=@var{floating-point-format}]
265 [@b{-mfloat-abi}=@var{abi}]
266 [@b{-meabi}=@var{ver}]
269 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
270 @b{-mapcs-reentrant}]
271 [@b{-mthumb-interwork}] [@b{-moabi}] [@b{-k}]
275 @emph{Target CRIS options:}
276 [@b{--underscore} | @b{--no-underscore}]
278 [@b{--emulation=criself} | @b{--emulation=crisaout}]
279 @c Deprecated -- deliberately not documented.
284 @emph{Target D10V options:}
289 @emph{Target D30V options:}
290 [@b{-O}|@b{-n}|@b{-N}]
293 @c Renesas family chips have no machine-dependent assembler options
296 @c HPPA has no machine-dependent assembler options (yet).
300 @emph{Target i386 options:}
301 [@b{--32}|@b{--64}] [@b{-n}]
305 @emph{Target i960 options:}
306 @c see md_parse_option in tc-i960.c
307 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
309 [@b{-b}] [@b{-no-relax}]
313 @emph{Target IA-64 options:}
314 [@b{-mconstant-gp}|@b{-mauto-pic}]
315 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
317 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
321 @emph{Target IP2K options:}
322 [@b{-mip2022}|@b{-mip2022ext}]
326 @emph{Target M32R options:}
327 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
332 @emph{Target M680X0 options:}
333 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
337 @emph{Target M68HC11 options:}
338 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
339 [@b{-mshort}|@b{-mlong}]
340 [@b{-mshort-double}|@b{-mlong-double}]
341 [@b{--force-long-branchs}] [@b{--short-branchs}]
342 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
343 [@b{--print-opcodes}] [@b{--generate-example}]
347 @emph{Target MCORE options:}
348 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
349 [@b{-mcpu=[210|340]}]
353 @emph{Target MIPS options:}
354 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
355 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
356 [@b{-non_shared}] [@b{-xgot}]
357 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
358 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
359 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
360 [@b{-mips64}] [@b{-mips64r2}]
361 [@b{-construct-floats}] [@b{-no-construct-floats}]
362 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
363 [@b{-mfix7000}] [@b{-mno-fix7000}]
364 [@b{-mips16}] [@b{-no-mips16}]
365 [@b{-mips3d}] [@b{-no-mips3d}]
366 [@b{-mdmx}] [@b{-no-mdmx}]
367 [@b{-mdebug}] [@b{-no-mdebug}]
368 [@b{-mpdr}] [@b{-mno-pdr}]
372 @emph{Target MMIX options:}
373 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
374 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
375 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
376 [@b{--linker-allocated-gregs}]
380 @emph{Target PDP11 options:}
381 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
382 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
383 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
387 @emph{Target picoJava options:}
392 @emph{Target PowerPC options:}
393 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
394 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}|
395 @b{-mbooke32}|@b{-mbooke64}]
396 [@b{-mcom}|@b{-many}|@b{-maltivec}] [@b{-memb}]
397 [@b{-mregnames}|@b{-mno-regnames}]
398 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
399 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
400 [@b{-msolaris}|@b{-mno-solaris}]
404 @emph{Target SPARC options:}
405 @c The order here is important. See c-sparc.texi.
406 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
407 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
408 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
413 @emph{Target TIC54X options:}
414 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
415 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
418 @c Z8000 has no machine-dependent assembler options
422 @emph{Target Xtensa options:}
423 [@b{--[no-]density}] [@b{--[no-]relax}] [@b{--[no-]generics}]
424 [@b{--[no-]text-section-literals}]
425 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
434 Turn on listings, in any of a variety of ways:
438 omit false conditionals
441 omit debugging directives
444 include high-level source
450 include macro expansions
453 omit forms processing
459 set the name of the listing file
462 You may combine these options; for example, use @samp{-aln} for assembly
463 listing without forms processing. The @samp{=file} option, if used, must be
464 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
467 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
470 Ignored. This option is accepted for script compatibility with calls to
473 @item --defsym @var{sym}=@var{value}
474 Define the symbol @var{sym} to be @var{value} before assembling the input file.
475 @var{value} must be an integer constant. As in C, a leading @samp{0x}
476 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
479 ``fast''---skip whitespace and comment preprocessing (assume source is
484 Generate debugging information for each assembler source line using whichever
485 debug format is preferred by the target. This currently means either STABS,
489 Generate stabs debugging information for each assembler line. This
490 may help debugging assembler code, if the debugger can handle it.
493 Generate stabs debugging information for each assembler line, with GNU
494 extensions that probably only gdb can handle, and that could make other
495 debuggers crash or refuse to read your program. This
496 may help debugging assembler code. Currently the only GNU extension is
497 the location of the current working directory at assembling time.
500 Generate DWARF2 debugging information for each assembler line. This
501 may help debugging assembler code, if the debugger can handle it. Note---this
502 option is only supported by some targets, not all of them.
505 Print a summary of the command line options and exit.
508 Print a summary of all target specific options and exit.
511 Add directory @var{dir} to the search list for @code{.include} directives.
514 Don't warn about signed overflow.
517 @ifclear DIFF-TBL-KLUGE
518 This option is accepted but has no effect on the @value{TARGET} family.
520 @ifset DIFF-TBL-KLUGE
521 Issue warnings when difference tables altered for long displacements.
526 Keep (in the symbol table) local symbols. On traditional a.out systems
527 these start with @samp{L}, but different systems have different local
530 @item --listing-lhs-width=@var{number}
531 Set the maximum width, in words, of the output data column for an assembler
532 listing to @var{number}.
534 @item --listing-lhs-width2=@var{number}
535 Set the maximum width, in words, of the output data column for continuation
536 lines in an assembler listing to @var{number}.
538 @item --listing-rhs-width=@var{number}
539 Set the maximum width of an input source line, as displayed in a listing, to
542 @item --listing-cont-lines=@var{number}
543 Set the maximum number of lines printed in a listing for a single line of input
546 @item -o @var{objfile}
547 Name the object-file output from @command{@value{AS}} @var{objfile}.
550 Fold the data section into the text section.
553 Print the maximum space (in bytes) and total time (in seconds) used by
556 @item --strip-local-absolute
557 Remove local absolute symbols from the outgoing symbol table.
561 Print the @command{as} version.
564 Print the @command{as} version and exit.
568 Suppress warning messages.
570 @item --fatal-warnings
571 Treat warnings as errors.
574 Don't suppress warning messages or treat them as errors.
583 Generate an object file even after errors.
585 @item -- | @var{files} @dots{}
586 Standard input, or source files to assemble.
591 The following options are available when @value{AS} is configured for
596 This option selects the core processor variant.
598 Select either big-endian (-EB) or little-endian (-EL) output.
603 The following options are available when @value{AS} is configured for the ARM
607 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
608 Specify which ARM processor variant is the target.
609 @item -march=@var{architecture}[+@var{extension}@dots{}]
610 Specify which ARM architecture variant is used by the target.
611 @item -mfpu=@var{floating-point-format}
612 Select which Floating Point architecture is the target.
613 @item -mfloat-abi=@var{abi}
614 Select which floating point ABI is in use.
616 Enable Thumb only instruction decoding.
617 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant | -moabi
618 Select which procedure calling convention is in use.
620 Select either big-endian (-EB) or little-endian (-EL) output.
621 @item -mthumb-interwork
622 Specify that the code has been generated with interworking between Thumb and
625 Specify that PIC code has been generated.
630 See the info pages for documentation of the CRIS-specific options.
634 The following options are available when @value{AS} is configured for
637 @cindex D10V optimization
638 @cindex optimization, D10V
640 Optimize output by parallelizing instructions.
645 The following options are available when @value{AS} is configured for a D30V
648 @cindex D30V optimization
649 @cindex optimization, D30V
651 Optimize output by parallelizing instructions.
655 Warn when nops are generated.
657 @cindex D30V nops after 32-bit multiply
659 Warn when a nop after a 32-bit multiply instruction is generated.
664 The following options are available when @value{AS} is configured for the
665 Intel 80960 processor.
668 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
669 Specify which variant of the 960 architecture is the target.
672 Add code to collect statistics about branches taken.
675 Do not alter compare-and-branch instructions for long displacements;
682 The following options are available when @value{AS} is configured for the
688 Specifies that the extended IP2022 instructions are allowed.
691 Restores the default behaviour, which restricts the permitted instructions to
692 just the basic IP2022 ones.
698 The following options are available when @value{AS} is configured for the
699 Renesas M32R (formerly Mitsubishi M32R) series.
704 Specify which processor in the M32R family is the target. The default
705 is normally the M32R, but this option changes it to the M32RX.
707 @item --warn-explicit-parallel-conflicts or --Wp
708 Produce warning messages when questionable parallel constructs are
711 @item --no-warn-explicit-parallel-conflicts or --Wnp
712 Do not produce warning messages when questionable parallel constructs are
719 The following options are available when @value{AS} is configured for the
720 Motorola 68000 series.
725 Shorten references to undefined symbols, to one word instead of two.
727 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
728 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
729 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
730 Specify what processor in the 68000 family is the target. The default
731 is normally the 68020, but this can be changed at configuration time.
733 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
734 The target machine does (or does not) have a floating-point coprocessor.
735 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
736 the basic 68000 is not compatible with the 68881, a combination of the
737 two can be specified, since it's possible to do emulation of the
738 coprocessor instructions with the main processor.
740 @item -m68851 | -mno-68851
741 The target machine does (or does not) have a memory-management
742 unit coprocessor. The default is to assume an MMU for 68020 and up.
749 For details about the PDP-11 machine dependent features options,
750 see @ref{PDP-11-Options}.
753 @item -mpic | -mno-pic
754 Generate position-independent (or position-dependent) code. The
755 default is @option{-mpic}.
758 @itemx -mall-extensions
759 Enable all instruction set extensions. This is the default.
761 @item -mno-extensions
762 Disable all instruction set extensions.
764 @item -m@var{extension} | -mno-@var{extension}
765 Enable (or disable) a particular instruction set extension.
768 Enable the instruction set extensions supported by a particular CPU, and
769 disable all other extensions.
771 @item -m@var{machine}
772 Enable the instruction set extensions supported by a particular machine
773 model, and disable all other extensions.
779 The following options are available when @value{AS} is configured for
780 a picoJava processor.
784 @cindex PJ endianness
785 @cindex endianness, PJ
786 @cindex big endian output, PJ
788 Generate ``big endian'' format output.
790 @cindex little endian output, PJ
792 Generate ``little endian'' format output.
798 The following options are available when @value{AS} is configured for the
799 Motorola 68HC11 or 68HC12 series.
803 @item -m68hc11 | -m68hc12 | -m68hcs12
804 Specify what processor is the target. The default is
805 defined by the configuration option when building the assembler.
808 Specify to use the 16-bit integer ABI.
811 Specify to use the 32-bit integer ABI.
814 Specify to use the 32-bit double ABI.
817 Specify to use the 64-bit double ABI.
819 @item --force-long-branchs
820 Relative branches are turned into absolute ones. This concerns
821 conditional branches, unconditional branches and branches to a
824 @item -S | --short-branchs
825 Do not turn relative branchs into absolute ones
826 when the offset is out of range.
828 @item --strict-direct-mode
829 Do not turn the direct addressing mode into extended addressing mode
830 when the instruction does not support direct addressing mode.
832 @item --print-insn-syntax
833 Print the syntax of instruction in case of error.
835 @item --print-opcodes
836 print the list of instructions with syntax and then exit.
838 @item --generate-example
839 print an example of instruction for each possible instruction and then exit.
840 This option is only useful for testing @command{@value{AS}}.
846 The following options are available when @command{@value{AS}} is configured
847 for the SPARC architecture:
850 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
851 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
852 Explicitly select a variant of the SPARC architecture.
854 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
855 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
857 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
858 UltraSPARC extensions.
860 @item -xarch=v8plus | -xarch=v8plusa
861 For compatibility with the Solaris v9 assembler. These options are
862 equivalent to -Av8plus and -Av8plusa, respectively.
865 Warn when the assembler switches to another architecture.
870 The following options are available when @value{AS} is configured for the 'c54x
875 Enable extended addressing mode. All addresses and relocations will assume
876 extended addressing (usually 23 bits).
877 @item -mcpu=@var{CPU_VERSION}
878 Sets the CPU version being compiled for.
879 @item -merrors-to-file @var{FILENAME}
880 Redirect error output to a file, for broken systems which don't support such
881 behaviour in the shell.
886 The following options are available when @value{AS} is configured for
887 a @sc{mips} processor.
891 This option sets the largest size of an object that can be referenced
892 implicitly with the @code{gp} register. It is only accepted for targets that
893 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
895 @cindex MIPS endianness
896 @cindex endianness, MIPS
897 @cindex big endian output, MIPS
899 Generate ``big endian'' format output.
901 @cindex little endian output, MIPS
903 Generate ``little endian'' format output.
915 Generate code for a particular @sc{mips} Instruction Set Architecture level.
916 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
917 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
918 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
919 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
921 correspond to generic
922 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
923 and @samp{MIPS64 Release 2}
924 ISA processors, respectively.
926 @item -march=@var{CPU}
927 Generate code for a particular @sc{mips} cpu.
929 @item -mtune=@var{cpu}
930 Schedule and tune for a particular @sc{mips} cpu.
934 Cause nops to be inserted if the read of the destination register
935 of an mfhi or mflo instruction occurs in the following two instructions.
939 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
940 section instead of the standard ELF .stabs sections.
944 Control generation of @code{.pdr} sections.
948 The register sizes are normally inferred from the ISA and ABI, but these
949 flags force a certain group of registers to be treated as 32 bits wide at
950 all times. @samp{-mgp32} controls the size of general-purpose registers
951 and @samp{-mfp32} controls the size of floating-point registers.
955 Generate code for the MIPS 16 processor. This is equivalent to putting
956 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
957 turns off this option.
961 Generate code for the MIPS-3D Application Specific Extension.
962 This tells the assembler to accept MIPS-3D instructions.
963 @samp{-no-mips3d} turns off this option.
967 Generate code for the MDMX Application Specific Extension.
968 This tells the assembler to accept MDMX instructions.
969 @samp{-no-mdmx} turns off this option.
971 @item --construct-floats
972 @itemx --no-construct-floats
973 The @samp{--no-construct-floats} option disables the construction of
974 double width floating point constants by loading the two halves of the
975 value into the two single width floating point registers that make up
976 the double width register. By default @samp{--construct-floats} is
977 selected, allowing construction of these floating point constants.
980 @item --emulation=@var{name}
981 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
982 for some other target, in all respects, including output format (choosing
983 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
984 debugging information or store symbol table information, and default
985 endianness. The available configuration names are: @samp{mipsecoff},
986 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
987 @samp{mipsbelf}. The first two do not alter the default endianness from that
988 of the primary target for which the assembler was configured; the others change
989 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
990 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
991 selection in any case.
993 This option is currently supported only when the primary target
994 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
995 Furthermore, the primary target or others specified with
996 @samp{--enable-targets=@dots{}} at configuration time must include support for
997 the other format, if both are to be available. For example, the Irix 5
998 configuration includes support for both.
1000 Eventually, this option will support more configurations, with more
1001 fine-grained control over the assembler's behavior, and will be supported for
1005 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1012 Control how to deal with multiplication overflow and division by zero.
1013 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1014 (and only work for Instruction Set Architecture level 2 and higher);
1015 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1019 When this option is used, @command{@value{AS}} will issue a warning every
1020 time it generates a nop instruction from a macro.
1025 The following options are available when @value{AS} is configured for
1031 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1032 The command line option @samp{-nojsri2bsr} can be used to disable it.
1036 Enable or disable the silicon filter behaviour. By default this is disabled.
1037 The default can be overridden by the @samp{-sifilter} command line option.
1040 Alter jump instructions for long displacements.
1042 @item -mcpu=[210|340]
1043 Select the cpu type on the target hardware. This controls which instructions
1047 Assemble for a big endian target.
1050 Assemble for a little endian target.
1056 See the info pages for documentation of the MMIX-specific options.
1060 The following options are available when @value{AS} is configured for
1061 an Xtensa processor.
1064 @item --density | --no-density
1065 Enable or disable use of instructions from the Xtensa code density
1066 option. This is enabled by default when the Xtensa processor supports
1067 the code density option.
1069 @item --relax | --no-relax
1070 Enable or disable instruction relaxation. This is enabled by default.
1071 Note: In the current implementation, these options also control whether
1072 assembler optimizations are performed, making these options equivalent
1073 to @option{--generics} and @option{--no-generics}.
1075 @item --generics | --no-generics
1076 Enable or disable all assembler transformations of Xtensa instructions.
1077 The default is @option{--generics};
1078 @option{--no-generics} should be used only in the rare cases when the
1079 instructions must be exactly as specified in the assembly source.
1081 @item --text-section-literals | --no-text-section-literals
1082 With @option{--text-@-section-@-literals}, literal pools are interspersed
1083 in the text section. The default is
1084 @option{--no-@-text-@-section-@-literals}, which places literals in a
1085 separate section in the output file.
1087 @item --target-align | --no-target-align
1088 Enable or disable automatic alignment to reduce branch penalties at the
1089 expense of some code density. The default is @option{--target-@-align}.
1091 @item --longcalls | --no-longcalls
1092 Enable or disable transformation of call instructions to allow calls
1093 across a greater range of addresses. The default is
1094 @option{--no-@-longcalls}.
1101 * Manual:: Structure of this Manual
1102 * GNU Assembler:: The GNU Assembler
1103 * Object Formats:: Object File Formats
1104 * Command Line:: Command Line
1105 * Input Files:: Input Files
1106 * Object:: Output (Object) File
1107 * Errors:: Error and Warning Messages
1111 @section Structure of this Manual
1113 @cindex manual, structure and purpose
1114 This manual is intended to describe what you need to know to use
1115 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1116 notation for symbols, constants, and expressions; the directives that
1117 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1120 We also cover special features in the @value{TARGET}
1121 configuration of @command{@value{AS}}, including assembler directives.
1124 This manual also describes some of the machine-dependent features of
1125 various flavors of the assembler.
1128 @cindex machine instructions (not covered)
1129 On the other hand, this manual is @emph{not} intended as an introduction
1130 to programming in assembly language---let alone programming in general!
1131 In a similar vein, we make no attempt to introduce the machine
1132 architecture; we do @emph{not} describe the instruction set, standard
1133 mnemonics, registers or addressing modes that are standard to a
1134 particular architecture.
1136 You may want to consult the manufacturer's
1137 machine architecture manual for this information.
1141 For information on the H8/300 machine instruction set, see @cite{H8/300
1142 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1143 Programming Manual} (Renesas).
1146 For information on the H8/500 machine instruction set, see @cite{H8/500
1147 Series Programming Manual} (Renesas M21T001).
1150 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1151 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1152 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1153 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1156 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1160 @c I think this is premature---doc@cygnus.com, 17jan1991
1162 Throughout this manual, we assume that you are running @dfn{GNU},
1163 the portable operating system from the @dfn{Free Software
1164 Foundation, Inc.}. This restricts our attention to certain kinds of
1165 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1166 once this assumption is granted examples and definitions need less
1169 @command{@value{AS}} is part of a team of programs that turn a high-level
1170 human-readable series of instructions into a low-level
1171 computer-readable series of instructions. Different versions of
1172 @command{@value{AS}} are used for different kinds of computer.
1175 @c There used to be a section "Terminology" here, which defined
1176 @c "contents", "byte", "word", and "long". Defining "word" to any
1177 @c particular size is confusing when the .word directive may generate 16
1178 @c bits on one machine and 32 bits on another; in general, for the user
1179 @c version of this manual, none of these terms seem essential to define.
1180 @c They were used very little even in the former draft of the manual;
1181 @c this draft makes an effort to avoid them (except in names of
1185 @section The GNU Assembler
1187 @c man begin DESCRIPTION
1189 @sc{gnu} @command{as} is really a family of assemblers.
1191 This manual describes @command{@value{AS}}, a member of that family which is
1192 configured for the @value{TARGET} architectures.
1194 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1195 should find a fairly similar environment when you use it on another
1196 architecture. Each version has much in common with the others,
1197 including object file formats, most assembler directives (often called
1198 @dfn{pseudo-ops}) and assembler syntax.@refill
1200 @cindex purpose of @sc{gnu} assembler
1201 @command{@value{AS}} is primarily intended to assemble the output of the
1202 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1203 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1204 assemble correctly everything that other assemblers for the same
1205 machine would assemble.
1207 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1210 @c This remark should appear in generic version of manual; assumption
1211 @c here is that generic version sets M680x0.
1212 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1213 assembler for the same architecture; for example, we know of several
1214 incompatible versions of 680x0 assembly language syntax.
1219 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1220 program in one pass of the source file. This has a subtle impact on the
1221 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1223 @node Object Formats
1224 @section Object File Formats
1226 @cindex object file format
1227 The @sc{gnu} assembler can be configured to produce several alternative
1228 object file formats. For the most part, this does not affect how you
1229 write assembly language programs; but directives for debugging symbols
1230 are typically different in different file formats. @xref{Symbol
1231 Attributes,,Symbol Attributes}.
1234 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1235 @value{OBJ-NAME} format object files.
1237 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1239 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1240 @code{a.out} or COFF format object files.
1243 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1244 @code{b.out} or COFF format object files.
1247 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1248 SOM or ELF format object files.
1253 @section Command Line
1255 @cindex command line conventions
1257 After the program name @command{@value{AS}}, the command line may contain
1258 options and file names. Options may appear in any order, and may be
1259 before, after, or between file names. The order of file names is
1262 @cindex standard input, as input file
1264 @file{--} (two hyphens) by itself names the standard input file
1265 explicitly, as one of the files for @command{@value{AS}} to assemble.
1267 @cindex options, command line
1268 Except for @samp{--} any command line argument that begins with a
1269 hyphen (@samp{-}) is an option. Each option changes the behavior of
1270 @command{@value{AS}}. No option changes the way another option works. An
1271 option is a @samp{-} followed by one or more letters; the case of
1272 the letter is important. All options are optional.
1274 Some options expect exactly one file name to follow them. The file
1275 name may either immediately follow the option's letter (compatible
1276 with older assemblers) or it may be the next command argument (@sc{gnu}
1277 standard). These two command lines are equivalent:
1280 @value{AS} -o my-object-file.o mumble.s
1281 @value{AS} -omy-object-file.o mumble.s
1285 @section Input Files
1288 @cindex source program
1289 @cindex files, input
1290 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1291 describe the program input to one run of @command{@value{AS}}. The program may
1292 be in one or more files; how the source is partitioned into files
1293 doesn't change the meaning of the source.
1295 @c I added "con" prefix to "catenation" just to prove I can overcome my
1296 @c APL training... doc@cygnus.com
1297 The source program is a concatenation of the text in all the files, in the
1300 @c man begin DESCRIPTION
1301 Each time you run @command{@value{AS}} it assembles exactly one source
1302 program. The source program is made up of one or more files.
1303 (The standard input is also a file.)
1305 You give @command{@value{AS}} a command line that has zero or more input file
1306 names. The input files are read (from left file name to right). A
1307 command line argument (in any position) that has no special meaning
1308 is taken to be an input file name.
1310 If you give @command{@value{AS}} no file names it attempts to read one input file
1311 from the @command{@value{AS}} standard input, which is normally your terminal. You
1312 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1315 Use @samp{--} if you need to explicitly name the standard input file
1316 in your command line.
1318 If the source is empty, @command{@value{AS}} produces a small, empty object
1323 @subheading Filenames and Line-numbers
1325 @cindex input file linenumbers
1326 @cindex line numbers, in input files
1327 There are two ways of locating a line in the input file (or files) and
1328 either may be used in reporting error messages. One way refers to a line
1329 number in a physical file; the other refers to a line number in a
1330 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1332 @dfn{Physical files} are those files named in the command line given
1333 to @command{@value{AS}}.
1335 @dfn{Logical files} are simply names declared explicitly by assembler
1336 directives; they bear no relation to physical files. Logical file names help
1337 error messages reflect the original source file, when @command{@value{AS}} source
1338 is itself synthesized from other files. @command{@value{AS}} understands the
1339 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1340 @ref{File,,@code{.file}}.
1343 @section Output (Object) File
1349 Every time you run @command{@value{AS}} it produces an output file, which is
1350 your assembly language program translated into numbers. This file
1351 is the object file. Its default name is
1359 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1361 You can give it another name by using the @option{-o} option. Conventionally,
1362 object file names end with @file{.o}. The default name is used for historical
1363 reasons: older assemblers were capable of assembling self-contained programs
1364 directly into a runnable program. (For some formats, this isn't currently
1365 possible, but it can be done for the @code{a.out} format.)
1369 The object file is meant for input to the linker @code{@value{LD}}. It contains
1370 assembled program code, information to help @code{@value{LD}} integrate
1371 the assembled program into a runnable file, and (optionally) symbolic
1372 information for the debugger.
1374 @c link above to some info file(s) like the description of a.out.
1375 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1378 @section Error and Warning Messages
1380 @c man begin DESCRIPTION
1382 @cindex error messages
1383 @cindex warning messages
1384 @cindex messages from assembler
1385 @command{@value{AS}} may write warnings and error messages to the standard error
1386 file (usually your terminal). This should not happen when a compiler
1387 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1388 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1389 grave problem that stops the assembly.
1393 @cindex format of warning messages
1394 Warning messages have the format
1397 file_name:@b{NNN}:Warning Message Text
1401 @cindex line numbers, in warnings/errors
1402 (where @b{NNN} is a line number). If a logical file name has been given
1403 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1404 the current input file is used. If a logical line number was given
1406 (@pxref{Line,,@code{.line}})
1410 (@pxref{Line,,@code{.line}})
1413 (@pxref{Ln,,@code{.ln}})
1416 then it is used to calculate the number printed,
1417 otherwise the actual line in the current source file is printed. The
1418 message text is intended to be self explanatory (in the grand Unix
1421 @cindex format of error messages
1422 Error messages have the format
1424 file_name:@b{NNN}:FATAL:Error Message Text
1426 The file name and line number are derived as for warning
1427 messages. The actual message text may be rather less explanatory
1428 because many of them aren't supposed to happen.
1431 @chapter Command-Line Options
1433 @cindex options, all versions of assembler
1434 This chapter describes command-line options available in @emph{all}
1435 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
1437 to the @value{TARGET} target.
1440 to particular machine architectures.
1443 @c man begin DESCRIPTION
1445 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1446 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1447 The assembler arguments must be separated from each other (and the @samp{-Wa})
1448 by commas. For example:
1451 gcc -c -g -O -Wa,-alh,-L file.c
1455 This passes two options to the assembler: @samp{-alh} (emit a listing to
1456 standard output with high-level and assembly source) and @samp{-L} (retain
1457 local symbols in the symbol table).
1459 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1460 command-line options are automatically passed to the assembler by the compiler.
1461 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1462 precisely what options it passes to each compilation pass, including the
1468 * a:: -a[cdhlns] enable listings
1469 * alternate:: --alternate enable alternate macro syntax
1470 * D:: -D for compatibility
1471 * f:: -f to work faster
1472 * I:: -I for .include search path
1473 @ifclear DIFF-TBL-KLUGE
1474 * K:: -K for compatibility
1476 @ifset DIFF-TBL-KLUGE
1477 * K:: -K for difference tables
1480 * L:: -L to retain local labels
1481 * listing:: --listing-XXX to configure listing output
1482 * M:: -M or --mri to assemble in MRI compatibility mode
1483 * MD:: --MD for dependency tracking
1484 * o:: -o to name the object file
1485 * R:: -R to join data and text sections
1486 * statistics:: --statistics to see statistics about assembly
1487 * traditional-format:: --traditional-format for compatible output
1488 * v:: -v to announce version
1489 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1490 * Z:: -Z to make object file even after errors
1494 @section Enable Listings: @option{-a[cdhlns]}
1503 @cindex listings, enabling
1504 @cindex assembly listings, enabling
1506 These options enable listing output from the assembler. By itself,
1507 @samp{-a} requests high-level, assembly, and symbols listing.
1508 You can use other letters to select specific options for the list:
1509 @samp{-ah} requests a high-level language listing,
1510 @samp{-al} requests an output-program assembly listing, and
1511 @samp{-as} requests a symbol table listing.
1512 High-level listings require that a compiler debugging option like
1513 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1516 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1517 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1518 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1519 omitted from the listing.
1521 Use the @samp{-ad} option to omit debugging directives from the
1524 Once you have specified one of these options, you can further control
1525 listing output and its appearance using the directives @code{.list},
1526 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1528 The @samp{-an} option turns off all forms processing.
1529 If you do not request listing output with one of the @samp{-a} options, the
1530 listing-control directives have no effect.
1532 The letters after @samp{-a} may be combined into one option,
1533 @emph{e.g.}, @samp{-aln}.
1535 Note if the assembler source is coming from the standard input (eg because it
1536 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1537 is being used) then the listing will not contain any comments or preprocessor
1538 directives. This is because the listing code buffers input source lines from
1539 stdin only after they have been preprocessed by the assembler. This reduces
1540 memory usage and makes the code more efficient.
1543 @section @option{--alternate}
1546 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1549 @section @option{-D}
1552 This option has no effect whatsoever, but it is accepted to make it more
1553 likely that scripts written for other assemblers also work with
1554 @command{@value{AS}}.
1557 @section Work Faster: @option{-f}
1560 @cindex trusted compiler
1561 @cindex faster processing (@option{-f})
1562 @samp{-f} should only be used when assembling programs written by a
1563 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1564 and comment preprocessing on
1565 the input file(s) before assembling them. @xref{Preprocessing,
1569 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1570 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1575 @section @code{.include} Search Path: @option{-I} @var{path}
1577 @kindex -I @var{path}
1578 @cindex paths for @code{.include}
1579 @cindex search path for @code{.include}
1580 @cindex @code{include} directive search path
1581 Use this option to add a @var{path} to the list of directories
1582 @command{@value{AS}} searches for files specified in @code{.include}
1583 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1584 many times as necessary to include a variety of paths. The current
1585 working directory is always searched first; after that, @command{@value{AS}}
1586 searches any @samp{-I} directories in the same order as they were
1587 specified (left to right) on the command line.
1590 @section Difference Tables: @option{-K}
1593 @ifclear DIFF-TBL-KLUGE
1594 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1595 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1596 where it can be used to warn when the assembler alters the machine code
1597 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1598 family does not have the addressing limitations that sometimes lead to this
1599 alteration on other platforms.
1602 @ifset DIFF-TBL-KLUGE
1603 @cindex difference tables, warning
1604 @cindex warning for altered difference tables
1605 @command{@value{AS}} sometimes alters the code emitted for directives of the form
1606 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1607 You can use the @samp{-K} option if you want a warning issued when this
1612 @section Include Local Labels: @option{-L}
1615 @cindex local labels, retaining in output
1616 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1617 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1618 debugging, because they are intended for the use of programs (like
1619 compilers) that compose assembler programs, not for your notice.
1620 Normally both @command{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1621 normally debug with them.
1623 This option tells @command{@value{AS}} to retain those @samp{L@dots{}} symbols
1624 in the object file. Usually if you do this you also tell the linker
1625 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1627 By default, a local label is any label beginning with @samp{L}, but each
1628 target is allowed to redefine the local label prefix.
1630 On the HPPA local labels begin with @samp{L$}.
1634 @section Configuring listing output: @option{--listing}
1636 The listing feature of the assembler can be enabled via the command line switch
1637 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1638 hex dump of the corresponding locations in the output object file, and displays
1639 them as a listing file. The format of this listing can be controlled by pseudo
1640 ops inside the assembler source (@pxref{List} @pxref{Title} @pxref{Sbttl}
1641 @pxref{Psize} @pxref{Eject}) and also by the following switches:
1644 @item --listing-lhs-width=@samp{number}
1645 @kindex --listing-lhs-width
1646 @cindex Width of first line disassembly output
1647 Sets the maximum width, in words, of the first line of the hex byte dump. This
1648 dump appears on the left hand side of the listing output.
1650 @item --listing-lhs-width2=@samp{number}
1651 @kindex --listing-lhs-width2
1652 @cindex Width of continuation lines of disassembly output
1653 Sets the maximum width, in words, of any further lines of the hex byte dump for
1654 a given input source line. If this value is not specified, it defaults to being
1655 the same as the value specified for @samp{--listing-lhs-width}. If neither
1656 switch is used the default is to one.
1658 @item --listing-rhs-width=@samp{number}
1659 @kindex --listing-rhs-width
1660 @cindex Width of source line output
1661 Sets the maximum width, in characters, of the source line that is displayed
1662 alongside the hex dump. The default value for this parameter is 100. The
1663 source line is displayed on the right hand side of the listing output.
1665 @item --listing-cont-lines=@samp{number}
1666 @kindex --listing-cont-lines
1667 @cindex Maximum number of continuation lines
1668 Sets the maximum number of continuation lines of hex dump that will be
1669 displayed for a given single line of source input. The default value is 4.
1673 @section Assemble in MRI Compatibility Mode: @option{-M}
1676 @cindex MRI compatibility mode
1677 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1678 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1679 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1680 configured target) assembler from Microtec Research. The exact nature of the
1681 MRI syntax will not be documented here; see the MRI manuals for more
1682 information. Note in particular that the handling of macros and macro
1683 arguments is somewhat different. The purpose of this option is to permit
1684 assembling existing MRI assembler code using @command{@value{AS}}.
1686 The MRI compatibility is not complete. Certain operations of the MRI assembler
1687 depend upon its object file format, and can not be supported using other object
1688 file formats. Supporting these would require enhancing each object file format
1689 individually. These are:
1692 @item global symbols in common section
1694 The m68k MRI assembler supports common sections which are merged by the linker.
1695 Other object file formats do not support this. @command{@value{AS}} handles
1696 common sections by treating them as a single common symbol. It permits local
1697 symbols to be defined within a common section, but it can not support global
1698 symbols, since it has no way to describe them.
1700 @item complex relocations
1702 The MRI assemblers support relocations against a negated section address, and
1703 relocations which combine the start addresses of two or more sections. These
1704 are not support by other object file formats.
1706 @item @code{END} pseudo-op specifying start address
1708 The MRI @code{END} pseudo-op permits the specification of a start address.
1709 This is not supported by other object file formats. The start address may
1710 instead be specified using the @option{-e} option to the linker, or in a linker
1713 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1715 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1716 name to the output file. This is not supported by other object file formats.
1718 @item @code{ORG} pseudo-op
1720 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1721 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1722 which changes the location within the current section. Absolute sections are
1723 not supported by other object file formats. The address of a section may be
1724 assigned within a linker script.
1727 There are some other features of the MRI assembler which are not supported by
1728 @command{@value{AS}}, typically either because they are difficult or because they
1729 seem of little consequence. Some of these may be supported in future releases.
1733 @item EBCDIC strings
1735 EBCDIC strings are not supported.
1737 @item packed binary coded decimal
1739 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1740 and @code{DCB.P} pseudo-ops are not supported.
1742 @item @code{FEQU} pseudo-op
1744 The m68k @code{FEQU} pseudo-op is not supported.
1746 @item @code{NOOBJ} pseudo-op
1748 The m68k @code{NOOBJ} pseudo-op is not supported.
1750 @item @code{OPT} branch control options
1752 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1753 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1754 relaxes all branches, whether forward or backward, to an appropriate size, so
1755 these options serve no purpose.
1757 @item @code{OPT} list control options
1759 The following m68k @code{OPT} list control options are ignored: @code{C},
1760 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1761 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1763 @item other @code{OPT} options
1765 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1766 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1768 @item @code{OPT} @code{D} option is default
1770 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1771 @code{OPT NOD} may be used to turn it off.
1773 @item @code{XREF} pseudo-op.
1775 The m68k @code{XREF} pseudo-op is ignored.
1777 @item @code{.debug} pseudo-op
1779 The i960 @code{.debug} pseudo-op is not supported.
1781 @item @code{.extended} pseudo-op
1783 The i960 @code{.extended} pseudo-op is not supported.
1785 @item @code{.list} pseudo-op.
1787 The various options of the i960 @code{.list} pseudo-op are not supported.
1789 @item @code{.optimize} pseudo-op
1791 The i960 @code{.optimize} pseudo-op is not supported.
1793 @item @code{.output} pseudo-op
1795 The i960 @code{.output} pseudo-op is not supported.
1797 @item @code{.setreal} pseudo-op
1799 The i960 @code{.setreal} pseudo-op is not supported.
1804 @section Dependency Tracking: @option{--MD}
1807 @cindex dependency tracking
1810 @command{@value{AS}} can generate a dependency file for the file it creates. This
1811 file consists of a single rule suitable for @code{make} describing the
1812 dependencies of the main source file.
1814 The rule is written to the file named in its argument.
1816 This feature is used in the automatic updating of makefiles.
1819 @section Name the Object File: @option{-o}
1822 @cindex naming object file
1823 @cindex object file name
1824 There is always one object file output when you run @command{@value{AS}}. By
1825 default it has the name
1828 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1842 You use this option (which takes exactly one filename) to give the
1843 object file a different name.
1845 Whatever the object file is called, @command{@value{AS}} overwrites any
1846 existing file of the same name.
1849 @section Join Data and Text Sections: @option{-R}
1852 @cindex data and text sections, joining
1853 @cindex text and data sections, joining
1854 @cindex joining text and data sections
1855 @cindex merging text and data sections
1856 @option{-R} tells @command{@value{AS}} to write the object file as if all
1857 data-section data lives in the text section. This is only done at
1858 the very last moment: your binary data are the same, but data
1859 section parts are relocated differently. The data section part of
1860 your object file is zero bytes long because all its bytes are
1861 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1863 When you specify @option{-R} it would be possible to generate shorter
1864 address displacements (because we do not have to cross between text and
1865 data section). We refrain from doing this simply for compatibility with
1866 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
1869 When @command{@value{AS}} is configured for COFF or ELF output,
1870 this option is only useful if you use sections named @samp{.text} and
1875 @option{-R} is not supported for any of the HPPA targets. Using
1876 @option{-R} generates a warning from @command{@value{AS}}.
1880 @section Display Assembly Statistics: @option{--statistics}
1882 @kindex --statistics
1883 @cindex statistics, about assembly
1884 @cindex time, total for assembly
1885 @cindex space used, maximum for assembly
1886 Use @samp{--statistics} to display two statistics about the resources used by
1887 @command{@value{AS}}: the maximum amount of space allocated during the assembly
1888 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1891 @node traditional-format
1892 @section Compatible Output: @option{--traditional-format}
1894 @kindex --traditional-format
1895 For some targets, the output of @command{@value{AS}} is different in some ways
1896 from the output of some existing assembler. This switch requests
1897 @command{@value{AS}} to use the traditional format instead.
1899 For example, it disables the exception frame optimizations which
1900 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
1903 @section Announce Version: @option{-v}
1907 @cindex assembler version
1908 @cindex version of assembler
1909 You can find out what version of as is running by including the
1910 option @samp{-v} (which you can also spell as @samp{-version}) on the
1914 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
1916 @command{@value{AS}} should never give a warning or error message when
1917 assembling compiler output. But programs written by people often
1918 cause @command{@value{AS}} to give a warning that a particular assumption was
1919 made. All such warnings are directed to the standard error file.
1923 @cindex suppressing warnings
1924 @cindex warnings, suppressing
1925 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
1926 This only affects the warning messages: it does not change any particular of
1927 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
1930 @kindex --fatal-warnings
1931 @cindex errors, caused by warnings
1932 @cindex warnings, causing error
1933 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
1934 files that generate warnings to be in error.
1937 @cindex warnings, switching on
1938 You can switch these options off again by specifying @option{--warn}, which
1939 causes warnings to be output as usual.
1942 @section Generate Object File in Spite of Errors: @option{-Z}
1943 @cindex object file, after errors
1944 @cindex errors, continuing after
1945 After an error message, @command{@value{AS}} normally produces no output. If for
1946 some reason you are interested in object file output even after
1947 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
1948 option. If there are any errors, @command{@value{AS}} continues anyways, and
1949 writes an object file after a final warning message of the form @samp{@var{n}
1950 errors, @var{m} warnings, generating bad object file.}
1955 @cindex machine-independent syntax
1956 @cindex syntax, machine-independent
1957 This chapter describes the machine-independent syntax allowed in a
1958 source file. @command{@value{AS}} syntax is similar to what many other
1959 assemblers use; it is inspired by the BSD 4.2
1964 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
1968 * Preprocessing:: Preprocessing
1969 * Whitespace:: Whitespace
1970 * Comments:: Comments
1971 * Symbol Intro:: Symbols
1972 * Statements:: Statements
1973 * Constants:: Constants
1977 @section Preprocessing
1979 @cindex preprocessing
1980 The @command{@value{AS}} internal preprocessor:
1982 @cindex whitespace, removed by preprocessor
1984 adjusts and removes extra whitespace. It leaves one space or tab before
1985 the keywords on a line, and turns any other whitespace on the line into
1988 @cindex comments, removed by preprocessor
1990 removes all comments, replacing them with a single space, or an
1991 appropriate number of newlines.
1993 @cindex constants, converted by preprocessor
1995 converts character constants into the appropriate numeric values.
1998 It does not do macro processing, include file handling, or
1999 anything else you may get from your C compiler's preprocessor. You can
2000 do include file processing with the @code{.include} directive
2001 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2002 to get other ``CPP'' style preprocessing by giving the input file a
2003 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
2004 Output, gcc.info, Using GNU CC}.
2006 Excess whitespace, comments, and character constants
2007 cannot be used in the portions of the input text that are not
2010 @cindex turning preprocessing on and off
2011 @cindex preprocessing, turning on and off
2014 If the first line of an input file is @code{#NO_APP} or if you use the
2015 @samp{-f} option, whitespace and comments are not removed from the input file.
2016 Within an input file, you can ask for whitespace and comment removal in
2017 specific portions of the by putting a line that says @code{#APP} before the
2018 text that may contain whitespace or comments, and putting a line that says
2019 @code{#NO_APP} after this text. This feature is mainly intend to support
2020 @code{asm} statements in compilers whose output is otherwise free of comments
2027 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2028 Whitespace is used to separate symbols, and to make programs neater for
2029 people to read. Unless within character constants
2030 (@pxref{Characters,,Character Constants}), any whitespace means the same
2031 as exactly one space.
2037 There are two ways of rendering comments to @command{@value{AS}}. In both
2038 cases the comment is equivalent to one space.
2040 Anything from @samp{/*} through the next @samp{*/} is a comment.
2041 This means you may not nest these comments.
2045 The only way to include a newline ('\n') in a comment
2046 is to use this sort of comment.
2049 /* This sort of comment does not nest. */
2052 @cindex line comment character
2053 Anything from the @dfn{line comment} character to the next newline
2054 is considered a comment and is ignored. The line comment character is
2056 @samp{;} for the AMD 29K family;
2059 @samp{;} on the ARC;
2062 @samp{@@} on the ARM;
2065 @samp{;} for the H8/300 family;
2068 @samp{!} for the H8/500 family;
2071 @samp{;} for the HPPA;
2074 @samp{#} on the i386 and x86-64;
2077 @samp{#} on the i960;
2080 @samp{;} for the PDP-11;
2083 @samp{;} for picoJava;
2086 @samp{#} for Motorola PowerPC;
2089 @samp{!} for the Renesas / SuperH SH;
2092 @samp{!} on the SPARC;
2095 @samp{#} on the ip2k;
2098 @samp{#} on the m32r;
2101 @samp{|} on the 680x0;
2104 @samp{#} on the 68HC11 and 68HC12;
2107 @samp{;} on the M880x0;
2110 @samp{#} on the Vax;
2113 @samp{!} for the Z8000;
2116 @samp{#} on the V850;
2119 @samp{#} for Xtensa systems;
2121 see @ref{Machine Dependencies}. @refill
2122 @c FIXME What about i860?
2125 On some machines there are two different line comment characters. One
2126 character only begins a comment if it is the first non-whitespace character on
2127 a line, while the other always begins a comment.
2131 The V850 assembler also supports a double dash as starting a comment that
2132 extends to the end of the line.
2138 @cindex lines starting with @code{#}
2139 @cindex logical line numbers
2140 To be compatible with past assemblers, lines that begin with @samp{#} have a
2141 special interpretation. Following the @samp{#} should be an absolute
2142 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2143 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
2144 new logical file name. The rest of the line, if any, should be whitespace.
2146 If the first non-whitespace characters on the line are not numeric,
2147 the line is ignored. (Just like a comment.)
2150 # This is an ordinary comment.
2151 # 42-6 "new_file_name" # New logical file name
2152 # This is logical line # 36.
2154 This feature is deprecated, and may disappear from future versions
2155 of @command{@value{AS}}.
2160 @cindex characters used in symbols
2161 @ifclear SPECIAL-SYMS
2162 A @dfn{symbol} is one or more characters chosen from the set of all
2163 letters (both upper and lower case), digits and the three characters
2169 A @dfn{symbol} is one or more characters chosen from the set of all
2170 letters (both upper and lower case), digits and the three characters
2171 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2177 On most machines, you can also use @code{$} in symbol names; exceptions
2178 are noted in @ref{Machine Dependencies}.
2180 No symbol may begin with a digit. Case is significant.
2181 There is no length limit: all characters are significant. Symbols are
2182 delimited by characters not in that set, or by the beginning of a file
2183 (since the source program must end with a newline, the end of a file is
2184 not a possible symbol delimiter). @xref{Symbols}.
2185 @cindex length of symbols
2190 @cindex statements, structure of
2191 @cindex line separator character
2192 @cindex statement separator character
2194 @ifclear abnormal-separator
2195 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2196 semicolon (@samp{;}). The newline or semicolon is considered part of
2197 the preceding statement. Newlines and semicolons within character
2198 constants are an exception: they do not end statements.
2200 @ifset abnormal-separator
2202 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
2203 sign (@samp{@@}). The newline or at sign is considered part of the
2204 preceding statement. Newlines and at signs within character constants
2205 are an exception: they do not end statements.
2208 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2209 point (@samp{!}). The newline or exclamation point is considered part of the
2210 preceding statement. Newlines and exclamation points within character
2211 constants are an exception: they do not end statements.
2214 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2215 H8/300) a dollar sign (@samp{$}); or (for the
2218 (@samp{;}). The newline or separator character is considered part of
2219 the preceding statement. Newlines and separators within character
2220 constants are an exception: they do not end statements.
2225 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2226 separator character. (The line separator is usually @samp{;}, unless
2227 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
2228 newline or separator character is considered part of the preceding
2229 statement. Newlines and separators within character constants are an
2230 exception: they do not end statements.
2233 @cindex newline, required at file end
2234 @cindex EOF, newline must precede
2235 It is an error to end any statement with end-of-file: the last
2236 character of any input file should be a newline.@refill
2238 An empty statement is allowed, and may include whitespace. It is ignored.
2240 @cindex instructions and directives
2241 @cindex directives and instructions
2242 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2243 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2245 A statement begins with zero or more labels, optionally followed by a
2246 key symbol which determines what kind of statement it is. The key
2247 symbol determines the syntax of the rest of the statement. If the
2248 symbol begins with a dot @samp{.} then the statement is an assembler
2249 directive: typically valid for any computer. If the symbol begins with
2250 a letter the statement is an assembly language @dfn{instruction}: it
2251 assembles into a machine language instruction.
2253 Different versions of @command{@value{AS}} for different computers
2254 recognize different instructions. In fact, the same symbol may
2255 represent a different instruction in a different computer's assembly
2259 @cindex @code{:} (label)
2260 @cindex label (@code{:})
2261 A label is a symbol immediately followed by a colon (@code{:}).
2262 Whitespace before a label or after a colon is permitted, but you may not
2263 have whitespace between a label's symbol and its colon. @xref{Labels}.
2266 For HPPA targets, labels need not be immediately followed by a colon, but
2267 the definition of a label must begin in column zero. This also implies that
2268 only one label may be defined on each line.
2272 label: .directive followed by something
2273 another_label: # This is an empty statement.
2274 instruction operand_1, operand_2, @dots{}
2281 A constant is a number, written so that its value is known by
2282 inspection, without knowing any context. Like this:
2285 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2286 .ascii "Ring the bell\7" # A string constant.
2287 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2288 .float 0f-314159265358979323846264338327\
2289 95028841971.693993751E-40 # - pi, a flonum.
2294 * Characters:: Character Constants
2295 * Numbers:: Number Constants
2299 @subsection Character Constants
2301 @cindex character constants
2302 @cindex constants, character
2303 There are two kinds of character constants. A @dfn{character} stands
2304 for one character in one byte and its value may be used in
2305 numeric expressions. String constants (properly called string
2306 @emph{literals}) are potentially many bytes and their values may not be
2307 used in arithmetic expressions.
2311 * Chars:: Characters
2315 @subsubsection Strings
2317 @cindex string constants
2318 @cindex constants, string
2319 A @dfn{string} is written between double-quotes. It may contain
2320 double-quotes or null characters. The way to get special characters
2321 into a string is to @dfn{escape} these characters: precede them with
2322 a backslash @samp{\} character. For example @samp{\\} represents
2323 one backslash: the first @code{\} is an escape which tells
2324 @command{@value{AS}} to interpret the second character literally as a backslash
2325 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2326 escape character). The complete list of escapes follows.
2328 @cindex escape codes, character
2329 @cindex character escape codes
2332 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2334 @cindex @code{\b} (backspace character)
2335 @cindex backspace (@code{\b})
2337 Mnemonic for backspace; for ASCII this is octal code 010.
2340 @c Mnemonic for EOText; for ASCII this is octal code 004.
2342 @cindex @code{\f} (formfeed character)
2343 @cindex formfeed (@code{\f})
2345 Mnemonic for FormFeed; for ASCII this is octal code 014.
2347 @cindex @code{\n} (newline character)
2348 @cindex newline (@code{\n})
2350 Mnemonic for newline; for ASCII this is octal code 012.
2353 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2355 @cindex @code{\r} (carriage return character)
2356 @cindex carriage return (@code{\r})
2358 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2361 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2362 @c other assemblers.
2364 @cindex @code{\t} (tab)
2365 @cindex tab (@code{\t})
2367 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2370 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2371 @c @item \x @var{digit} @var{digit} @var{digit}
2372 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2374 @cindex @code{\@var{ddd}} (octal character code)
2375 @cindex octal character code (@code{\@var{ddd}})
2376 @item \ @var{digit} @var{digit} @var{digit}
2377 An octal character code. The numeric code is 3 octal digits.
2378 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2379 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2381 @cindex @code{\@var{xd...}} (hex character code)
2382 @cindex hex character code (@code{\@var{xd...}})
2383 @item \@code{x} @var{hex-digits...}
2384 A hex character code. All trailing hex digits are combined. Either upper or
2385 lower case @code{x} works.
2387 @cindex @code{\\} (@samp{\} character)
2388 @cindex backslash (@code{\\})
2390 Represents one @samp{\} character.
2393 @c Represents one @samp{'} (accent acute) character.
2394 @c This is needed in single character literals
2395 @c (@xref{Characters,,Character Constants}.) to represent
2398 @cindex @code{\"} (doublequote character)
2399 @cindex doublequote (@code{\"})
2401 Represents one @samp{"} character. Needed in strings to represent
2402 this character, because an unescaped @samp{"} would end the string.
2404 @item \ @var{anything-else}
2405 Any other character when escaped by @kbd{\} gives a warning, but
2406 assembles as if the @samp{\} was not present. The idea is that if
2407 you used an escape sequence you clearly didn't want the literal
2408 interpretation of the following character. However @command{@value{AS}} has no
2409 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2410 code and warns you of the fact.
2413 Which characters are escapable, and what those escapes represent,
2414 varies widely among assemblers. The current set is what we think
2415 the BSD 4.2 assembler recognizes, and is a subset of what most C
2416 compilers recognize. If you are in doubt, do not use an escape
2420 @subsubsection Characters
2422 @cindex single character constant
2423 @cindex character, single
2424 @cindex constant, single character
2425 A single character may be written as a single quote immediately
2426 followed by that character. The same escapes apply to characters as
2427 to strings. So if you want to write the character backslash, you
2428 must write @kbd{'\\} where the first @code{\} escapes the second
2429 @code{\}. As you can see, the quote is an acute accent, not a
2430 grave accent. A newline
2432 @ifclear abnormal-separator
2433 (or semicolon @samp{;})
2435 @ifset abnormal-separator
2437 (or at sign @samp{@@})
2440 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2441 Renesas SH or H8/500)
2445 immediately following an acute accent is taken as a literal character
2446 and does not count as the end of a statement. The value of a character
2447 constant in a numeric expression is the machine's byte-wide code for
2448 that character. @command{@value{AS}} assumes your character code is ASCII:
2449 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2452 @subsection Number Constants
2454 @cindex constants, number
2455 @cindex number constants
2456 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2457 are stored in the target machine. @emph{Integers} are numbers that
2458 would fit into an @code{int} in the C language. @emph{Bignums} are
2459 integers, but they are stored in more than 32 bits. @emph{Flonums}
2460 are floating point numbers, described below.
2463 * Integers:: Integers
2468 * Bit Fields:: Bit Fields
2474 @subsubsection Integers
2476 @cindex constants, integer
2478 @cindex binary integers
2479 @cindex integers, binary
2480 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2481 the binary digits @samp{01}.
2483 @cindex octal integers
2484 @cindex integers, octal
2485 An octal integer is @samp{0} followed by zero or more of the octal
2486 digits (@samp{01234567}).
2488 @cindex decimal integers
2489 @cindex integers, decimal
2490 A decimal integer starts with a non-zero digit followed by zero or
2491 more digits (@samp{0123456789}).
2493 @cindex hexadecimal integers
2494 @cindex integers, hexadecimal
2495 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2496 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2498 Integers have the usual values. To denote a negative integer, use
2499 the prefix operator @samp{-} discussed under expressions
2500 (@pxref{Prefix Ops,,Prefix Operators}).
2503 @subsubsection Bignums
2506 @cindex constants, bignum
2507 A @dfn{bignum} has the same syntax and semantics as an integer
2508 except that the number (or its negative) takes more than 32 bits to
2509 represent in binary. The distinction is made because in some places
2510 integers are permitted while bignums are not.
2513 @subsubsection Flonums
2515 @cindex floating point numbers
2516 @cindex constants, floating point
2518 @cindex precision, floating point
2519 A @dfn{flonum} represents a floating point number. The translation is
2520 indirect: a decimal floating point number from the text is converted by
2521 @command{@value{AS}} to a generic binary floating point number of more than
2522 sufficient precision. This generic floating point number is converted
2523 to a particular computer's floating point format (or formats) by a
2524 portion of @command{@value{AS}} specialized to that computer.
2526 A flonum is written by writing (in order)
2531 (@samp{0} is optional on the HPPA.)
2535 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2537 @kbd{e} is recommended. Case is not important.
2539 @c FIXME: verify if flonum syntax really this vague for most cases
2540 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2541 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2544 On the H8/300, H8/500,
2545 Renesas / SuperH SH,
2546 and AMD 29K architectures, the letter must be
2547 one of the letters @samp{DFPRSX} (in upper or lower case).
2549 On the ARC, the letter must be one of the letters @samp{DFRS}
2550 (in upper or lower case).
2552 On the Intel 960 architecture, the letter must be
2553 one of the letters @samp{DFT} (in upper or lower case).
2555 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2559 One of the letters @samp{DFPRSX} (in upper or lower case).
2562 One of the letters @samp{DFRS} (in upper or lower case).
2565 One of the letters @samp{DFPRSX} (in upper or lower case).
2568 The letter @samp{E} (upper case only).
2571 One of the letters @samp{DFT} (in upper or lower case).
2576 An optional sign: either @samp{+} or @samp{-}.
2579 An optional @dfn{integer part}: zero or more decimal digits.
2582 An optional @dfn{fractional part}: @samp{.} followed by zero
2583 or more decimal digits.
2586 An optional exponent, consisting of:
2590 An @samp{E} or @samp{e}.
2591 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2592 @c principle this can perfectly well be different on different targets.
2594 Optional sign: either @samp{+} or @samp{-}.
2596 One or more decimal digits.
2601 At least one of the integer part or the fractional part must be
2602 present. The floating point number has the usual base-10 value.
2604 @command{@value{AS}} does all processing using integers. Flonums are computed
2605 independently of any floating point hardware in the computer running
2606 @command{@value{AS}}.
2610 @c Bit fields are written as a general facility but are also controlled
2611 @c by a conditional-compilation flag---which is as of now (21mar91)
2612 @c turned on only by the i960 config of GAS.
2614 @subsubsection Bit Fields
2617 @cindex constants, bit field
2618 You can also define numeric constants as @dfn{bit fields}.
2619 specify two numbers separated by a colon---
2621 @var{mask}:@var{value}
2624 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2627 The resulting number is then packed
2629 @c this conditional paren in case bit fields turned on elsewhere than 960
2630 (in host-dependent byte order)
2632 into a field whose width depends on which assembler directive has the
2633 bit-field as its argument. Overflow (a result from the bitwise and
2634 requiring more binary digits to represent) is not an error; instead,
2635 more constants are generated, of the specified width, beginning with the
2636 least significant digits.@refill
2638 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2639 @code{.short}, and @code{.word} accept bit-field arguments.
2644 @chapter Sections and Relocation
2649 * Secs Background:: Background
2650 * Ld Sections:: Linker Sections
2651 * As Sections:: Assembler Internal Sections
2652 * Sub-Sections:: Sub-Sections
2656 @node Secs Background
2659 Roughly, a section is a range of addresses, with no gaps; all data
2660 ``in'' those addresses is treated the same for some particular purpose.
2661 For example there may be a ``read only'' section.
2663 @cindex linker, and assembler
2664 @cindex assembler, and linker
2665 The linker @code{@value{LD}} reads many object files (partial programs) and
2666 combines their contents to form a runnable program. When @command{@value{AS}}
2667 emits an object file, the partial program is assumed to start at address 0.
2668 @code{@value{LD}} assigns the final addresses for the partial program, so that
2669 different partial programs do not overlap. This is actually an
2670 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2673 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2674 addresses. These blocks slide to their run-time addresses as rigid
2675 units; their length does not change and neither does the order of bytes
2676 within them. Such a rigid unit is called a @emph{section}. Assigning
2677 run-time addresses to sections is called @dfn{relocation}. It includes
2678 the task of adjusting mentions of object-file addresses so they refer to
2679 the proper run-time addresses.
2681 For the H8/300 and H8/500,
2682 and for the Renesas / SuperH SH,
2683 @command{@value{AS}} pads sections if needed to
2684 ensure they end on a word (sixteen bit) boundary.
2687 @cindex standard assembler sections
2688 An object file written by @command{@value{AS}} has at least three sections, any
2689 of which may be empty. These are named @dfn{text}, @dfn{data} and
2694 When it generates COFF or ELF output,
2696 @command{@value{AS}} can also generate whatever other named sections you specify
2697 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2698 If you do not use any directives that place output in the @samp{.text}
2699 or @samp{.data} sections, these sections still exist, but are empty.
2704 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2706 @command{@value{AS}} can also generate whatever other named sections you
2707 specify using the @samp{.space} and @samp{.subspace} directives. See
2708 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2709 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2710 assembler directives.
2713 Additionally, @command{@value{AS}} uses different names for the standard
2714 text, data, and bss sections when generating SOM output. Program text
2715 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2716 BSS into @samp{$BSS$}.
2720 Within the object file, the text section starts at address @code{0}, the
2721 data section follows, and the bss section follows the data section.
2724 When generating either SOM or ELF output files on the HPPA, the text
2725 section starts at address @code{0}, the data section at address
2726 @code{0x4000000}, and the bss section follows the data section.
2729 To let @code{@value{LD}} know which data changes when the sections are
2730 relocated, and how to change that data, @command{@value{AS}} also writes to the
2731 object file details of the relocation needed. To perform relocation
2732 @code{@value{LD}} must know, each time an address in the object
2736 Where in the object file is the beginning of this reference to
2739 How long (in bytes) is this reference?
2741 Which section does the address refer to? What is the numeric value of
2743 (@var{address}) @minus{} (@var{start-address of section})?
2746 Is the reference to an address ``Program-Counter relative''?
2749 @cindex addresses, format of
2750 @cindex section-relative addressing
2751 In fact, every address @command{@value{AS}} ever uses is expressed as
2753 (@var{section}) + (@var{offset into section})
2756 Further, most expressions @command{@value{AS}} computes have this section-relative
2759 (For some object formats, such as SOM for the HPPA, some expressions are
2760 symbol-relative instead.)
2763 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2764 @var{N} into section @var{secname}.''
2766 Apart from text, data and bss sections you need to know about the
2767 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2768 addresses in the absolute section remain unchanged. For example, address
2769 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2770 @code{@value{LD}}. Although the linker never arranges two partial programs'
2771 data sections with overlapping addresses after linking, @emph{by definition}
2772 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2773 part of a program is always the same address when the program is running as
2774 address @code{@{absolute@ 239@}} in any other part of the program.
2776 The idea of sections is extended to the @dfn{undefined} section. Any
2777 address whose section is unknown at assembly time is by definition
2778 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2779 Since numbers are always defined, the only way to generate an undefined
2780 address is to mention an undefined symbol. A reference to a named
2781 common block would be such a symbol: its value is unknown at assembly
2782 time so it has section @emph{undefined}.
2784 By analogy the word @emph{section} is used to describe groups of sections in
2785 the linked program. @code{@value{LD}} puts all partial programs' text
2786 sections in contiguous addresses in the linked program. It is
2787 customary to refer to the @emph{text section} of a program, meaning all
2788 the addresses of all partial programs' text sections. Likewise for
2789 data and bss sections.
2791 Some sections are manipulated by @code{@value{LD}}; others are invented for
2792 use of @command{@value{AS}} and have no meaning except during assembly.
2795 @section Linker Sections
2796 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2801 @cindex named sections
2802 @cindex sections, named
2803 @item named sections
2806 @cindex text section
2807 @cindex data section
2811 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2812 separate but equal sections. Anything you can say of one section is
2815 When the program is running, however, it is
2816 customary for the text section to be unalterable. The
2817 text section is often shared among processes: it contains
2818 instructions, constants and the like. The data section of a running
2819 program is usually alterable: for example, C variables would be stored
2820 in the data section.
2825 This section contains zeroed bytes when your program begins running. It
2826 is used to hold uninitialized variables or common storage. The length of
2827 each partial program's bss section is important, but because it starts
2828 out containing zeroed bytes there is no need to store explicit zero
2829 bytes in the object file. The bss section was invented to eliminate
2830 those explicit zeros from object files.
2832 @cindex absolute section
2833 @item absolute section
2834 Address 0 of this section is always ``relocated'' to runtime address 0.
2835 This is useful if you want to refer to an address that @code{@value{LD}} must
2836 not change when relocating. In this sense we speak of absolute
2837 addresses being ``unrelocatable'': they do not change during relocation.
2839 @cindex undefined section
2840 @item undefined section
2841 This ``section'' is a catch-all for address references to objects not in
2842 the preceding sections.
2843 @c FIXME: ref to some other doc on obj-file formats could go here.
2846 @cindex relocation example
2847 An idealized example of three relocatable sections follows.
2849 The example uses the traditional section names @samp{.text} and @samp{.data}.
2851 Memory addresses are on the horizontal axis.
2855 @c END TEXI2ROFF-KILL
2858 partial program # 1: |ttttt|dddd|00|
2865 partial program # 2: |TTT|DDD|000|
2868 +--+---+-----+--+----+---+-----+~~
2869 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2870 +--+---+-----+--+----+---+-----+~~
2872 addresses: 0 @dots{}
2879 \line{\it Partial program \#1: \hfil}
2880 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2881 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2883 \line{\it Partial program \#2: \hfil}
2884 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2885 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2887 \line{\it linked program: \hfil}
2888 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2889 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2890 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2891 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2893 \line{\it addresses: \hfil}
2897 @c END TEXI2ROFF-KILL
2900 @section Assembler Internal Sections
2902 @cindex internal assembler sections
2903 @cindex sections in messages, internal
2904 These sections are meant only for the internal use of @command{@value{AS}}. They
2905 have no meaning at run-time. You do not really need to know about these
2906 sections for most purposes; but they can be mentioned in @command{@value{AS}}
2907 warning messages, so it might be helpful to have an idea of their
2908 meanings to @command{@value{AS}}. These sections are used to permit the
2909 value of every expression in your assembly language program to be a
2910 section-relative address.
2913 @cindex assembler internal logic error
2914 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2915 An internal assembler logic error has been found. This means there is a
2916 bug in the assembler.
2918 @cindex expr (internal section)
2920 The assembler stores complex expression internally as combinations of
2921 symbols. When it needs to represent an expression as a symbol, it puts
2922 it in the expr section.
2924 @c FIXME item transfer[t] vector preload
2925 @c FIXME item transfer[t] vector postload
2926 @c FIXME item register
2930 @section Sub-Sections
2932 @cindex numbered subsections
2933 @cindex grouping data
2939 fall into two sections: text and data.
2941 You may have separate groups of
2943 data in named sections
2947 data in named sections
2953 that you want to end up near to each other in the object file, even though they
2954 are not contiguous in the assembler source. @command{@value{AS}} allows you to
2955 use @dfn{subsections} for this purpose. Within each section, there can be
2956 numbered subsections with values from 0 to 8192. Objects assembled into the
2957 same subsection go into the object file together with other objects in the same
2958 subsection. For example, a compiler might want to store constants in the text
2959 section, but might not want to have them interspersed with the program being
2960 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2961 section of code being output, and a @samp{.text 1} before each group of
2962 constants being output.
2964 Subsections are optional. If you do not use subsections, everything
2965 goes in subsection number zero.
2968 Each subsection is zero-padded up to a multiple of four bytes.
2969 (Subsections may be padded a different amount on different flavors
2970 of @command{@value{AS}}.)
2974 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2975 boundary (two bytes).
2976 The same is true on the Renesas SH.
2979 @c FIXME section padding (alignment)?
2980 @c Rich Pixley says padding here depends on target obj code format; that
2981 @c doesn't seem particularly useful to say without further elaboration,
2982 @c so for now I say nothing about it. If this is a generic BFD issue,
2983 @c these paragraphs might need to vanish from this manual, and be
2984 @c discussed in BFD chapter of binutils (or some such).
2987 On the AMD 29K family, no particular padding is added to section or
2988 subsection sizes; @value{AS} forces no alignment on this platform.
2992 Subsections appear in your object file in numeric order, lowest numbered
2993 to highest. (All this to be compatible with other people's assemblers.)
2994 The object file contains no representation of subsections; @code{@value{LD}} and
2995 other programs that manipulate object files see no trace of them.
2996 They just see all your text subsections as a text section, and all your
2997 data subsections as a data section.
2999 To specify which subsection you want subsequent statements assembled
3000 into, use a numeric argument to specify it, in a @samp{.text
3001 @var{expression}} or a @samp{.data @var{expression}} statement.
3004 When generating COFF output, you
3009 can also use an extra subsection
3010 argument with arbitrary named sections: @samp{.section @var{name},
3015 When generating ELF output, you
3020 can also use the @code{.subsection} directive (@pxref{SubSection})
3021 to specify a subsection: @samp{.subsection @var{expression}}.
3023 @var{Expression} should be an absolute expression.
3024 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
3025 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3026 begins in @code{text 0}. For instance:
3028 .text 0 # The default subsection is text 0 anyway.
3029 .ascii "This lives in the first text subsection. *"
3031 .ascii "But this lives in the second text subsection."
3033 .ascii "This lives in the data section,"
3034 .ascii "in the first data subsection."
3036 .ascii "This lives in the first text section,"
3037 .ascii "immediately following the asterisk (*)."
3040 Each section has a @dfn{location counter} incremented by one for every byte
3041 assembled into that section. Because subsections are merely a convenience
3042 restricted to @command{@value{AS}} there is no concept of a subsection location
3043 counter. There is no way to directly manipulate a location counter---but the
3044 @code{.align} directive changes it, and any label definition captures its
3045 current value. The location counter of the section where statements are being
3046 assembled is said to be the @dfn{active} location counter.
3049 @section bss Section
3052 @cindex common variable storage
3053 The bss section is used for local common variable storage.
3054 You may allocate address space in the bss section, but you may
3055 not dictate data to load into it before your program executes. When
3056 your program starts running, all the contents of the bss
3057 section are zeroed bytes.
3059 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3060 @ref{Lcomm,,@code{.lcomm}}.
3062 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3063 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
3066 When assembling for a target which supports multiple sections, such as ELF or
3067 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3068 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3069 section. Typically the section will only contain symbol definitions and
3070 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3077 Symbols are a central concept: the programmer uses symbols to name
3078 things, the linker uses symbols to link, and the debugger uses symbols
3082 @cindex debuggers, and symbol order
3083 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3084 the same order they were declared. This may break some debuggers.
3089 * Setting Symbols:: Giving Symbols Other Values
3090 * Symbol Names:: Symbol Names
3091 * Dot:: The Special Dot Symbol
3092 * Symbol Attributes:: Symbol Attributes
3099 A @dfn{label} is written as a symbol immediately followed by a colon
3100 @samp{:}. The symbol then represents the current value of the
3101 active location counter, and is, for example, a suitable instruction
3102 operand. You are warned if you use the same symbol to represent two
3103 different locations: the first definition overrides any other
3107 On the HPPA, the usual form for a label need not be immediately followed by a
3108 colon, but instead must start in column zero. Only one label may be defined on
3109 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3110 provides a special directive @code{.label} for defining labels more flexibly.
3113 @node Setting Symbols
3114 @section Giving Symbols Other Values
3116 @cindex assigning values to symbols
3117 @cindex symbol values, assigning
3118 A symbol can be given an arbitrary value by writing a symbol, followed
3119 by an equals sign @samp{=}, followed by an expression
3120 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3121 directive. @xref{Set,,@code{.set}}.
3124 @section Symbol Names
3126 @cindex symbol names
3127 @cindex names, symbol
3128 @ifclear SPECIAL-SYMS
3129 Symbol names begin with a letter or with one of @samp{._}. On most
3130 machines, you can also use @code{$} in symbol names; exceptions are
3131 noted in @ref{Machine Dependencies}. That character may be followed by any
3132 string of digits, letters, dollar signs (unless otherwise noted in
3133 @ref{Machine Dependencies}), and underscores.
3136 For the AMD 29K family, @samp{?} is also allowed in the
3137 body of a symbol name, though not at its beginning.
3142 Symbol names begin with a letter or with one of @samp{._}. On the
3143 Renesas SH or the H8/500, you can also use @code{$} in symbol names. That
3144 character may be followed by any string of digits, letters, dollar signs (save
3145 on the H8/300), and underscores.
3149 Case of letters is significant: @code{foo} is a different symbol name
3152 Each symbol has exactly one name. Each name in an assembly language program
3153 refers to exactly one symbol. You may use that symbol name any number of times
3156 @subheading Local Symbol Names
3158 @cindex local symbol names
3159 @cindex symbol names, local
3160 @cindex temporary symbol names
3161 @cindex symbol names, temporary
3162 Local symbols help compilers and programmers use names temporarily.
3163 They create symbols which are guaranteed to be unique over the entire scope of
3164 the input source code and which can be referred to by a simple notation.
3165 To define a local symbol, write a label of the form @samp{@b{N}:} (where @b{N}
3166 represents any positive integer). To refer to the most recent previous
3167 definition of that symbol write @samp{@b{N}b}, using the same number as when
3168 you defined the label. To refer to the next definition of a local label, write
3169 @samp{@b{N}f}--- The @samp{b} stands for``backwards'' and the @samp{f} stands
3172 There is no restriction on how you can use these labels, and you can reuse them
3173 too. So that it is possible to repeatedly define the same local label (using
3174 the same number @samp{@b{N}}), although you can only refer to the most recently
3175 defined local label of that number (for a backwards reference) or the next
3176 definition of a specific local label for a forward reference. It is also worth
3177 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3178 implemented in a slightly more efficient manner than the others.
3189 Which is the equivalent of:
3192 label_1: branch label_3
3193 label_2: branch label_1
3194 label_3: branch label_4
3195 label_4: branch label_3
3198 Local symbol names are only a notational device. They are immediately
3199 transformed into more conventional symbol names before the assembler uses them.
3200 The symbol names stored in the symbol table, appearing in error messages and
3201 optionally emitted to the object file. The names are constructed using these
3206 All local labels begin with @samp{L}. Normally both @command{@value{AS}} and
3207 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
3208 used for symbols you are never intended to see. If you use the
3209 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3210 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3211 you may use them in debugging.
3214 This is the number that was used in the local label definition. So if the
3215 label is written @samp{55:} then the number is @samp{55}.
3218 This unusual character is included so you do not accidentally invent a symbol
3219 of the same name. The character has ASCII value of @samp{\002} (control-B).
3221 @item @emph{ordinal number}
3222 This is a serial number to keep the labels distinct. The first definition of
3223 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3224 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3225 the number @samp{1} and its 15th defintion gets @samp{15} as well.
3228 So for example, the first @code{1:} is named @code{L1@kbd{C-B}1}, the 44th
3229 @code{3:} is named @code{L3@kbd{C-B}44}.
3231 @subheading Dollar Local Labels
3232 @cindex dollar local symbols
3234 @code{@value{AS}} also supports an even more local form of local labels called
3235 dollar labels. These labels go out of scope (ie they become undefined) as soon
3236 as a non-local label is defined. Thus they remain valid for only a small
3237 region of the input source code. Normal local labels, by contrast, remain in
3238 scope for the entire file, or until they are redefined by another occurrence of
3239 the same local label.
3241 Dollar labels are defined in exactly the same way as ordinary local labels,
3242 except that instead of being terminated by a colon, they are terminated by a
3243 dollar sign. eg @samp{@b{55$}}.
3245 They can also be distinguished from ordinary local labels by their transformed
3246 name which uses ASCII character @samp{\001} (control-A) as the magic character
3247 to distinguish them from ordinary labels. Thus the 5th defintion of @samp{6$}
3248 is named @samp{L6@kbd{C-A}5}.
3251 @section The Special Dot Symbol
3253 @cindex dot (symbol)
3254 @cindex @code{.} (symbol)
3255 @cindex current address
3256 @cindex location counter
3257 The special symbol @samp{.} refers to the current address that
3258 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3259 .long .} defines @code{melvin} to contain its own address.
3260 Assigning a value to @code{.} is treated the same as a @code{.org}
3261 directive. Thus, the expression @samp{.=.+4} is the same as saying
3262 @ifclear no-space-dir
3271 @node Symbol Attributes
3272 @section Symbol Attributes
3274 @cindex symbol attributes
3275 @cindex attributes, symbol
3276 Every symbol has, as well as its name, the attributes ``Value'' and
3277 ``Type''. Depending on output format, symbols can also have auxiliary
3280 The detailed definitions are in @file{a.out.h}.
3283 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3284 all these attributes, and probably won't warn you. This makes the
3285 symbol an externally defined symbol, which is generally what you
3289 * Symbol Value:: Value
3290 * Symbol Type:: Type
3293 * a.out Symbols:: Symbol Attributes: @code{a.out}
3297 * a.out Symbols:: Symbol Attributes: @code{a.out}
3300 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3305 * COFF Symbols:: Symbol Attributes for COFF
3308 * SOM Symbols:: Symbol Attributes for SOM
3315 @cindex value of a symbol
3316 @cindex symbol value
3317 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3318 location in the text, data, bss or absolute sections the value is the
3319 number of addresses from the start of that section to the label.
3320 Naturally for text, data and bss sections the value of a symbol changes
3321 as @code{@value{LD}} changes section base addresses during linking. Absolute
3322 symbols' values do not change during linking: that is why they are
3325 The value of an undefined symbol is treated in a special way. If it is
3326 0 then the symbol is not defined in this assembler source file, and
3327 @code{@value{LD}} tries to determine its value from other files linked into the
3328 same program. You make this kind of symbol simply by mentioning a symbol
3329 name without defining it. A non-zero value represents a @code{.comm}
3330 common declaration. The value is how much common storage to reserve, in
3331 bytes (addresses). The symbol refers to the first address of the
3337 @cindex type of a symbol
3339 The type attribute of a symbol contains relocation (section)
3340 information, any flag settings indicating that a symbol is external, and
3341 (optionally), other information for linkers and debuggers. The exact
3342 format depends on the object-code output format in use.
3347 @c The following avoids a "widow" subsection title. @group would be
3348 @c better if it were available outside examples.
3351 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3353 @cindex @code{b.out} symbol attributes
3354 @cindex symbol attributes, @code{b.out}
3355 These symbol attributes appear only when @command{@value{AS}} is configured for
3356 one of the Berkeley-descended object output formats---@code{a.out} or
3362 @subsection Symbol Attributes: @code{a.out}
3364 @cindex @code{a.out} symbol attributes
3365 @cindex symbol attributes, @code{a.out}
3371 @subsection Symbol Attributes: @code{a.out}
3373 @cindex @code{a.out} symbol attributes
3374 @cindex symbol attributes, @code{a.out}
3378 * Symbol Desc:: Descriptor
3379 * Symbol Other:: Other
3383 @subsubsection Descriptor
3385 @cindex descriptor, of @code{a.out} symbol
3386 This is an arbitrary 16-bit value. You may establish a symbol's
3387 descriptor value by using a @code{.desc} statement
3388 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3389 @command{@value{AS}}.
3392 @subsubsection Other
3394 @cindex other attribute, of @code{a.out} symbol
3395 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3400 @subsection Symbol Attributes for COFF
3402 @cindex COFF symbol attributes
3403 @cindex symbol attributes, COFF
3405 The COFF format supports a multitude of auxiliary symbol attributes;
3406 like the primary symbol attributes, they are set between @code{.def} and
3407 @code{.endef} directives.
3409 @subsubsection Primary Attributes
3411 @cindex primary attributes, COFF symbols
3412 The symbol name is set with @code{.def}; the value and type,
3413 respectively, with @code{.val} and @code{.type}.
3415 @subsubsection Auxiliary Attributes
3417 @cindex auxiliary attributes, COFF symbols
3418 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3419 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3420 table information for COFF.
3425 @subsection Symbol Attributes for SOM
3427 @cindex SOM symbol attributes
3428 @cindex symbol attributes, SOM
3430 The SOM format for the HPPA supports a multitude of symbol attributes set with
3431 the @code{.EXPORT} and @code{.IMPORT} directives.
3433 The attributes are described in @cite{HP9000 Series 800 Assembly
3434 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3435 @code{EXPORT} assembler directive documentation.
3439 @chapter Expressions
3443 @cindex numeric values
3444 An @dfn{expression} specifies an address or numeric value.
3445 Whitespace may precede and/or follow an expression.
3447 The result of an expression must be an absolute number, or else an offset into
3448 a particular section. If an expression is not absolute, and there is not
3449 enough information when @command{@value{AS}} sees the expression to know its
3450 section, a second pass over the source program might be necessary to interpret
3451 the expression---but the second pass is currently not implemented.
3452 @command{@value{AS}} aborts with an error message in this situation.
3455 * Empty Exprs:: Empty Expressions
3456 * Integer Exprs:: Integer Expressions
3460 @section Empty Expressions
3462 @cindex empty expressions
3463 @cindex expressions, empty
3464 An empty expression has no value: it is just whitespace or null.
3465 Wherever an absolute expression is required, you may omit the
3466 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3467 is compatible with other assemblers.
3470 @section Integer Expressions
3472 @cindex integer expressions
3473 @cindex expressions, integer
3474 An @dfn{integer expression} is one or more @emph{arguments} delimited
3475 by @emph{operators}.
3478 * Arguments:: Arguments
3479 * Operators:: Operators
3480 * Prefix Ops:: Prefix Operators
3481 * Infix Ops:: Infix Operators
3485 @subsection Arguments
3487 @cindex expression arguments
3488 @cindex arguments in expressions
3489 @cindex operands in expressions
3490 @cindex arithmetic operands
3491 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3492 contexts arguments are sometimes called ``arithmetic operands''. In
3493 this manual, to avoid confusing them with the ``instruction operands'' of
3494 the machine language, we use the term ``argument'' to refer to parts of
3495 expressions only, reserving the word ``operand'' to refer only to machine
3496 instruction operands.
3498 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3499 @var{section} is one of text, data, bss, absolute,
3500 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3503 Numbers are usually integers.
3505 A number can be a flonum or bignum. In this case, you are warned
3506 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3507 these 32 bits are an integer. You may write integer-manipulating
3508 instructions that act on exotic constants, compatible with other
3511 @cindex subexpressions
3512 Subexpressions are a left parenthesis @samp{(} followed by an integer
3513 expression, followed by a right parenthesis @samp{)}; or a prefix
3514 operator followed by an argument.
3517 @subsection Operators
3519 @cindex operators, in expressions
3520 @cindex arithmetic functions
3521 @cindex functions, in expressions
3522 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3523 operators are followed by an argument. Infix operators appear
3524 between their arguments. Operators may be preceded and/or followed by
3528 @subsection Prefix Operator
3530 @cindex prefix operators
3531 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3532 one argument, which must be absolute.
3534 @c the tex/end tex stuff surrounding this small table is meant to make
3535 @c it align, on the printed page, with the similar table in the next
3536 @c section (which is inside an enumerate).
3538 \global\advance\leftskip by \itemindent
3543 @dfn{Negation}. Two's complement negation.
3545 @dfn{Complementation}. Bitwise not.
3549 \global\advance\leftskip by -\itemindent
3553 @subsection Infix Operators
3555 @cindex infix operators
3556 @cindex operators, permitted arguments
3557 @dfn{Infix operators} take two arguments, one on either side. Operators
3558 have precedence, but operations with equal precedence are performed left
3559 to right. Apart from @code{+} or @option{-}, both arguments must be
3560 absolute, and the result is absolute.
3563 @cindex operator precedence
3564 @cindex precedence of operators
3571 @dfn{Multiplication}.
3574 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3581 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3585 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3589 Intermediate precedence
3594 @dfn{Bitwise Inclusive Or}.
3600 @dfn{Bitwise Exclusive Or}.
3603 @dfn{Bitwise Or Not}.
3610 @cindex addition, permitted arguments
3611 @cindex plus, permitted arguments
3612 @cindex arguments for addition
3614 @dfn{Addition}. If either argument is absolute, the result has the section of
3615 the other argument. You may not add together arguments from different
3618 @cindex subtraction, permitted arguments
3619 @cindex minus, permitted arguments
3620 @cindex arguments for subtraction
3622 @dfn{Subtraction}. If the right argument is absolute, the
3623 result has the section of the left argument.
3624 If both arguments are in the same section, the result is absolute.
3625 You may not subtract arguments from different sections.
3626 @c FIXME is there still something useful to say about undefined - undefined ?
3628 @cindex comparison expressions
3629 @cindex expressions, comparison
3633 @dfn{Is Not Equal To}
3637 @dfn{Is Greater Than}
3639 @dfn{Is Greater Than Or Equal To}
3641 @dfn{Is Less Than Or Equal To}
3643 The comparison operators can be used as infix operators. A true results has a
3644 value of -1 whereas a false result has a value of 0. Note, these operators
3645 perform signed comparisons.
3648 @item Lowest Precedence
3657 These two logical operations can be used to combine the results of sub
3658 expressions. Note, unlike the comparison operators a true result returns a
3659 value of 1 but a false results does still return 0. Also note that the logical
3660 or operator has a slightly lower precedence than logical and.
3665 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3666 address; you can only have a defined section in one of the two arguments.
3669 @chapter Assembler Directives
3671 @cindex directives, machine independent
3672 @cindex pseudo-ops, machine independent
3673 @cindex machine independent directives
3674 All assembler directives have names that begin with a period (@samp{.}).
3675 The rest of the name is letters, usually in lower case.
3677 This chapter discusses directives that are available regardless of the
3678 target machine configuration for the @sc{gnu} assembler.
3680 Some machine configurations provide additional directives.
3681 @xref{Machine Dependencies}.
3684 @ifset machine-directives
3685 @xref{Machine Dependencies} for additional directives.
3690 * Abort:: @code{.abort}
3692 * ABORT:: @code{.ABORT}
3695 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3696 * Altmacro:: @code{.altmacro}
3697 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3698 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3699 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3700 * Byte:: @code{.byte @var{expressions}}
3701 * Comm:: @code{.comm @var{symbol} , @var{length} }
3703 * CFI directives:: @code{.cfi_startproc}, @code{.cfi_endproc}, etc.
3705 * Data:: @code{.data @var{subsection}}
3707 * Def:: @code{.def @var{name}}
3710 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3716 * Double:: @code{.double @var{flonums}}
3717 * Eject:: @code{.eject}
3718 * Else:: @code{.else}
3719 * Elseif:: @code{.elseif}
3722 * Endef:: @code{.endef}
3725 * Endfunc:: @code{.endfunc}
3726 * Endif:: @code{.endif}
3727 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3728 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3730 * Exitm:: @code{.exitm}
3731 * Extern:: @code{.extern}
3732 * Fail:: @code{.fail}
3733 @ifclear no-file-dir
3734 * File:: @code{.file @var{string}}
3737 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3738 * Float:: @code{.float @var{flonums}}
3739 * Func:: @code{.func}
3740 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3742 * Hidden:: @code{.hidden @var{names}}
3745 * hword:: @code{.hword @var{expressions}}
3746 * Ident:: @code{.ident}
3747 * If:: @code{.if @var{absolute expression}}
3748 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3749 * Include:: @code{.include "@var{file}"}
3750 * Int:: @code{.int @var{expressions}}
3752 * Internal:: @code{.internal @var{names}}
3755 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3756 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3757 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3758 * Lflags:: @code{.lflags}
3759 @ifclear no-line-dir
3760 * Line:: @code{.line @var{line-number}}
3763 * Ln:: @code{.ln @var{line-number}}
3764 * Linkonce:: @code{.linkonce [@var{type}]}
3765 * List:: @code{.list}
3766 * Long:: @code{.long @var{expressions}}
3768 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3771 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3772 * MRI:: @code{.mri @var{val}}
3773 * Noaltmacro:: @code{.noaltmacro}
3774 * Nolist:: @code{.nolist}
3775 * Octa:: @code{.octa @var{bignums}}
3776 * Org:: @code{.org @var{new-lc} , @var{fill}}
3777 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3779 * PopSection:: @code{.popsection}
3780 * Previous:: @code{.previous}
3783 * Print:: @code{.print @var{string}}
3785 * Protected:: @code{.protected @var{names}}
3788 * Psize:: @code{.psize @var{lines}, @var{columns}}
3789 * Purgem:: @code{.purgem @var{name}}
3791 * PushSection:: @code{.pushsection @var{name}}
3794 * Quad:: @code{.quad @var{bignums}}
3795 * Rept:: @code{.rept @var{count}}
3796 * Sbttl:: @code{.sbttl "@var{subheading}"}
3798 * Scl:: @code{.scl @var{class}}
3801 * Section:: @code{.section @var{name}}
3804 * Set:: @code{.set @var{symbol}, @var{expression}}
3805 * Short:: @code{.short @var{expressions}}
3806 * Single:: @code{.single @var{flonums}}
3808 * Size:: @code{.size [@var{name} , @var{expression}]}
3811 * Skip:: @code{.skip @var{size} , @var{fill}}
3812 * Sleb128:: @code{.sleb128 @var{expressions}}
3813 * Space:: @code{.space @var{size} , @var{fill}}
3815 * Stab:: @code{.stabd, .stabn, .stabs}
3818 * String:: @code{.string "@var{str}"}
3819 * Struct:: @code{.struct @var{expression}}
3821 * SubSection:: @code{.subsection}
3822 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3826 * Tag:: @code{.tag @var{structname}}
3829 * Text:: @code{.text @var{subsection}}
3830 * Title:: @code{.title "@var{heading}"}
3832 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3835 * Uleb128:: @code{.uleb128 @var{expressions}}
3837 * Val:: @code{.val @var{addr}}
3841 * Version:: @code{.version "@var{string}"}
3842 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3843 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3846 * Weak:: @code{.weak @var{names}}
3847 * Word:: @code{.word @var{expressions}}
3848 * Deprecated:: Deprecated Directives
3852 @section @code{.abort}
3854 @cindex @code{abort} directive
3855 @cindex stopping the assembly
3856 This directive stops the assembly immediately. It is for
3857 compatibility with other assemblers. The original idea was that the
3858 assembly language source would be piped into the assembler. If the sender
3859 of the source quit, it could use this directive tells @command{@value{AS}} to
3860 quit also. One day @code{.abort} will not be supported.
3864 @section @code{.ABORT}
3866 @cindex @code{ABORT} directive
3867 When producing COFF output, @command{@value{AS}} accepts this directive as a
3868 synonym for @samp{.abort}.
3871 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
3877 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3879 @cindex padding the location counter
3880 @cindex @code{align} directive
3881 Pad the location counter (in the current subsection) to a particular storage
3882 boundary. The first expression (which must be absolute) is the alignment
3883 required, as described below.
3885 The second expression (also absolute) gives the fill value to be stored in the
3886 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3887 padding bytes are normally zero. However, on some systems, if the section is
3888 marked as containing code and the fill value is omitted, the space is filled
3889 with no-op instructions.
3891 The third expression is also absolute, and is also optional. If it is present,
3892 it is the maximum number of bytes that should be skipped by this alignment
3893 directive. If doing the alignment would require skipping more bytes than the
3894 specified maximum, then the alignment is not done at all. You can omit the
3895 fill value (the second argument) entirely by simply using two commas after the
3896 required alignment; this can be useful if you want the alignment to be filled
3897 with no-op instructions when appropriate.
3899 The way the required alignment is specified varies from system to system.
3900 For the a29k, arc, hppa, i386 using ELF, i860, iq2000, m68k, m88k, or32,
3901 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
3902 alignment request in bytes. For example @samp{.align 8} advances
3903 the location counter until it is a multiple of 8. If the location counter
3904 is already a multiple of 8, no change is needed. For the tic54x, the
3905 first expression is the alignment request in words.
3907 For other systems, including the i386 using a.out format, and the arm and
3908 strongarm, it is the
3909 number of low-order zero bits the location counter must have after
3910 advancement. For example @samp{.align 3} advances the location
3911 counter until it a multiple of 8. If the location counter is already a
3912 multiple of 8, no change is needed.
3914 This inconsistency is due to the different behaviors of the various
3915 native assemblers for these systems which GAS must emulate.
3916 GAS also provides @code{.balign} and @code{.p2align} directives,
3917 described later, which have a consistent behavior across all
3918 architectures (but are specific to GAS).
3921 @section @code{.ascii "@var{string}"}@dots{}
3923 @cindex @code{ascii} directive
3924 @cindex string literals
3925 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3926 separated by commas. It assembles each string (with no automatic
3927 trailing zero byte) into consecutive addresses.
3930 @section @code{.asciz "@var{string}"}@dots{}
3932 @cindex @code{asciz} directive
3933 @cindex zero-terminated strings
3934 @cindex null-terminated strings
3935 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3936 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3939 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3941 @cindex padding the location counter given number of bytes
3942 @cindex @code{balign} directive
3943 Pad the location counter (in the current subsection) to a particular
3944 storage boundary. The first expression (which must be absolute) is the
3945 alignment request in bytes. For example @samp{.balign 8} advances
3946 the location counter until it is a multiple of 8. If the location counter
3947 is already a multiple of 8, no change is needed.
3949 The second expression (also absolute) gives the fill value to be stored in the
3950 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3951 padding bytes are normally zero. However, on some systems, if the section is
3952 marked as containing code and the fill value is omitted, the space is filled
3953 with no-op instructions.
3955 The third expression is also absolute, and is also optional. If it is present,
3956 it is the maximum number of bytes that should be skipped by this alignment
3957 directive. If doing the alignment would require skipping more bytes than the
3958 specified maximum, then the alignment is not done at all. You can omit the
3959 fill value (the second argument) entirely by simply using two commas after the
3960 required alignment; this can be useful if you want the alignment to be filled
3961 with no-op instructions when appropriate.
3963 @cindex @code{balignw} directive
3964 @cindex @code{balignl} directive
3965 The @code{.balignw} and @code{.balignl} directives are variants of the
3966 @code{.balign} directive. The @code{.balignw} directive treats the fill
3967 pattern as a two byte word value. The @code{.balignl} directives treats the
3968 fill pattern as a four byte longword value. For example, @code{.balignw
3969 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3970 filled in with the value 0x368d (the exact placement of the bytes depends upon
3971 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3975 @section @code{.byte @var{expressions}}
3977 @cindex @code{byte} directive
3978 @cindex integers, one byte
3979 @code{.byte} expects zero or more expressions, separated by commas.
3980 Each expression is assembled into the next byte.
3983 @section @code{.comm @var{symbol} , @var{length} }
3985 @cindex @code{comm} directive
3986 @cindex symbol, common
3987 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
3988 common symbol in one object file may be merged with a defined or common symbol
3989 of the same name in another object file. If @code{@value{LD}} does not see a
3990 definition for the symbol--just one or more common symbols--then it will
3991 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
3992 absolute expression. If @code{@value{LD}} sees multiple common symbols with
3993 the same name, and they do not all have the same size, it will allocate space
3994 using the largest size.
3997 When using ELF, the @code{.comm} directive takes an optional third argument.
3998 This is the desired alignment of the symbol, specified as a byte boundary (for
3999 example, an alignment of 16 means that the least significant 4 bits of the
4000 address should be zero). The alignment must be an absolute expression, and it
4001 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
4002 for the common symbol, it will use the alignment when placing the symbol. If
4003 no alignment is specified, @command{@value{AS}} will set the alignment to the
4004 largest power of two less than or equal to the size of the symbol, up to a
4009 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4010 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4013 @node CFI directives
4014 @section @code{.cfi_startproc}
4015 @cindex @code{cfi_startproc} directive
4016 @code{.cfi_startproc} is used at the beginning of each function that
4017 should have an entry in @code{.eh_frame}. It initializes some internal
4018 data structures and emits architecture dependent initial CFI instructions.
4019 Don't forget to close the function by
4020 @code{.cfi_endproc}.
4022 @section @code{.cfi_endproc}
4023 @cindex @code{cfi_endproc} directive
4024 @code{.cfi_endproc} is used at the end of a function where it closes its
4025 unwind entry previously opened by
4026 @code{.cfi_startproc}. and emits it to @code{.eh_frame}.
4028 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4029 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4030 address from @var{register} and add @var{offset} to it}.
4032 @section @code{.cfi_def_cfa_register @var{register}}
4033 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4034 now on @var{register} will be used instead of the old one. Offset
4037 @section @code{.cfi_def_cfa_offset @var{offset}}
4038 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4039 remains the same, but @var{offset} is new. Note that it is the
4040 absolute offset that will be added to a defined register to compute
4043 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4044 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4045 value that is added/substracted from the previous offset.
4047 @section @code{.cfi_offset @var{register}, @var{offset}}
4048 Previous value of @var{register} is saved at offset @var{offset} from
4051 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4052 Previous value of @var{register} is saved at offset @var{offset} from
4053 the current CFA register. This is transformed to @code{.cfi_offset}
4054 using the known displacement of the CFA register from the CFA.
4055 This is often easier to use, because the number will match the
4056 code it's annotating.
4058 @section @code{.cfi_window_save}
4059 SPARC register window has been saved.
4061 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4062 Allows the user to add arbitrary bytes to the unwind info. One
4063 might use this to add OS-specific CFI opcodes, or generic CFI
4064 opcodes that GAS does not yet support.
4067 @section @code{.data @var{subsection}}
4069 @cindex @code{data} directive
4070 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4071 end of the data subsection numbered @var{subsection} (which is an
4072 absolute expression). If @var{subsection} is omitted, it defaults
4077 @section @code{.def @var{name}}
4079 @cindex @code{def} directive
4080 @cindex COFF symbols, debugging
4081 @cindex debugging COFF symbols
4082 Begin defining debugging information for a symbol @var{name}; the
4083 definition extends until the @code{.endef} directive is encountered.
4086 This directive is only observed when @command{@value{AS}} is configured for COFF
4087 format output; when producing @code{b.out}, @samp{.def} is recognized,
4094 @section @code{.desc @var{symbol}, @var{abs-expression}}
4096 @cindex @code{desc} directive
4097 @cindex COFF symbol descriptor
4098 @cindex symbol descriptor, COFF
4099 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4100 to the low 16 bits of an absolute expression.
4103 The @samp{.desc} directive is not available when @command{@value{AS}} is
4104 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4105 object format. For the sake of compatibility, @command{@value{AS}} accepts
4106 it, but produces no output, when configured for COFF.
4112 @section @code{.dim}
4114 @cindex @code{dim} directive
4115 @cindex COFF auxiliary symbol information
4116 @cindex auxiliary symbol information, COFF
4117 This directive is generated by compilers to include auxiliary debugging
4118 information in the symbol table. It is only permitted inside
4119 @code{.def}/@code{.endef} pairs.
4122 @samp{.dim} is only meaningful when generating COFF format output; when
4123 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4129 @section @code{.double @var{flonums}}
4131 @cindex @code{double} directive
4132 @cindex floating point numbers (double)
4133 @code{.double} expects zero or more flonums, separated by commas. It
4134 assembles floating point numbers.
4136 The exact kind of floating point numbers emitted depends on how
4137 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4141 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4142 in @sc{ieee} format.
4147 @section @code{.eject}
4149 @cindex @code{eject} directive
4150 @cindex new page, in listings
4151 @cindex page, in listings
4152 @cindex listing control: new page
4153 Force a page break at this point, when generating assembly listings.
4156 @section @code{.else}
4158 @cindex @code{else} directive
4159 @code{.else} is part of the @command{@value{AS}} support for conditional
4160 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
4161 of code to be assembled if the condition for the preceding @code{.if}
4165 @section @code{.elseif}
4167 @cindex @code{elseif} directive
4168 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4169 assembly; @pxref{If,,@code{.if}}. It is shorthand for beginning a new
4170 @code{.if} block that would otherwise fill the entire @code{.else} section.
4173 @section @code{.end}
4175 @cindex @code{end} directive
4176 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4177 process anything in the file past the @code{.end} directive.
4181 @section @code{.endef}
4183 @cindex @code{endef} directive
4184 This directive flags the end of a symbol definition begun with
4188 @samp{.endef} is only meaningful when generating COFF format output; if
4189 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4190 directive but ignores it.
4195 @section @code{.endfunc}
4196 @cindex @code{endfunc} directive
4197 @code{.endfunc} marks the end of a function specified with @code{.func}.
4200 @section @code{.endif}
4202 @cindex @code{endif} directive
4203 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4204 it marks the end of a block of code that is only assembled
4205 conditionally. @xref{If,,@code{.if}}.
4208 @section @code{.equ @var{symbol}, @var{expression}}
4210 @cindex @code{equ} directive
4211 @cindex assigning values to symbols
4212 @cindex symbols, assigning values to
4213 This directive sets the value of @var{symbol} to @var{expression}.
4214 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
4217 The syntax for @code{equ} on the HPPA is
4218 @samp{@var{symbol} .equ @var{expression}}.
4222 @section @code{.equiv @var{symbol}, @var{expression}}
4223 @cindex @code{equiv} directive
4224 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4225 the assembler will signal an error if @var{symbol} is already defined. Note a
4226 symbol which has been referenced but not actually defined is considered to be
4229 Except for the contents of the error message, this is roughly equivalent to
4238 @section @code{.err}
4239 @cindex @code{err} directive
4240 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4241 message and, unless the @option{-Z} option was used, it will not generate an
4242 object file. This can be used to signal error an conditionally compiled code.
4245 @section @code{.exitm}
4246 Exit early from the current macro definition. @xref{Macro}.
4249 @section @code{.extern}
4251 @cindex @code{extern} directive
4252 @code{.extern} is accepted in the source program---for compatibility
4253 with other assemblers---but it is ignored. @command{@value{AS}} treats
4254 all undefined symbols as external.
4257 @section @code{.fail @var{expression}}
4259 @cindex @code{fail} directive
4260 Generates an error or a warning. If the value of the @var{expression} is 500
4261 or more, @command{@value{AS}} will print a warning message. If the value is less
4262 than 500, @command{@value{AS}} will print an error message. The message will
4263 include the value of @var{expression}. This can occasionally be useful inside
4264 complex nested macros or conditional assembly.
4266 @ifclear no-file-dir
4268 @section @code{.file @var{string}}
4270 @cindex @code{file} directive
4271 @cindex logical file name
4272 @cindex file name, logical
4273 @code{.file} tells @command{@value{AS}} that we are about to start a new logical
4274 file. @var{string} is the new file name. In general, the filename is
4275 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4276 to specify an empty file name, you must give the quotes--@code{""}. This
4277 statement may go away in future: it is only recognized to be compatible with
4278 old @command{@value{AS}} programs.
4280 In some configurations of @command{@value{AS}}, @code{.file} has already been
4281 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
4286 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4288 @cindex @code{fill} directive
4289 @cindex writing patterns in memory
4290 @cindex patterns, writing in memory
4291 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4292 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4293 may be zero or more. @var{Size} may be zero or more, but if it is
4294 more than 8, then it is deemed to have the value 8, compatible with
4295 other people's assemblers. The contents of each @var{repeat} bytes
4296 is taken from an 8-byte number. The highest order 4 bytes are
4297 zero. The lowest order 4 bytes are @var{value} rendered in the
4298 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4299 Each @var{size} bytes in a repetition is taken from the lowest order
4300 @var{size} bytes of this number. Again, this bizarre behavior is
4301 compatible with other people's assemblers.
4303 @var{size} and @var{value} are optional.
4304 If the second comma and @var{value} are absent, @var{value} is
4305 assumed zero. If the first comma and following tokens are absent,
4306 @var{size} is assumed to be 1.
4309 @section @code{.float @var{flonums}}
4311 @cindex floating point numbers (single)
4312 @cindex @code{float} directive
4313 This directive assembles zero or more flonums, separated by commas. It
4314 has the same effect as @code{.single}.
4316 The exact kind of floating point numbers emitted depends on how
4317 @command{@value{AS}} is configured.
4318 @xref{Machine Dependencies}.
4322 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4323 in @sc{ieee} format.
4328 @section @code{.func @var{name}[,@var{label}]}
4329 @cindex @code{func} directive
4330 @code{.func} emits debugging information to denote function @var{name}, and
4331 is ignored unless the file is assembled with debugging enabled.
4332 Only @samp{--gstabs[+]} is currently supported.
4333 @var{label} is the entry point of the function and if omitted @var{name}
4334 prepended with the @samp{leading char} is used.
4335 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4336 All functions are currently defined to have @code{void} return type.
4337 The function must be terminated with @code{.endfunc}.
4340 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4342 @cindex @code{global} directive
4343 @cindex symbol, making visible to linker
4344 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4345 @var{symbol} in your partial program, its value is made available to
4346 other partial programs that are linked with it. Otherwise,
4347 @var{symbol} takes its attributes from a symbol of the same name
4348 from another file linked into the same program.
4350 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4351 compatibility with other assemblers.
4354 On the HPPA, @code{.global} is not always enough to make it accessible to other
4355 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4356 @xref{HPPA Directives,, HPPA Assembler Directives}.
4361 @section @code{.hidden @var{names}}
4363 @cindex @code{hidden} directive
4365 This is one of the ELF visibility directives. The other two are
4366 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4367 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4369 This directive overrides the named symbols default visibility (which is set by
4370 their binding: local, global or weak). The directive sets the visibility to
4371 @code{hidden} which means that the symbols are not visible to other components.
4372 Such symbols are always considered to be @code{protected} as well.
4376 @section @code{.hword @var{expressions}}
4378 @cindex @code{hword} directive
4379 @cindex integers, 16-bit
4380 @cindex numbers, 16-bit
4381 @cindex sixteen bit integers
4382 This expects zero or more @var{expressions}, and emits
4383 a 16 bit number for each.
4386 This directive is a synonym for @samp{.short}; depending on the target
4387 architecture, it may also be a synonym for @samp{.word}.
4391 This directive is a synonym for @samp{.short}.
4394 This directive is a synonym for both @samp{.short} and @samp{.word}.
4399 @section @code{.ident}
4401 @cindex @code{ident} directive
4402 This directive is used by some assemblers to place tags in object files.
4403 @command{@value{AS}} simply accepts the directive for source-file
4404 compatibility with such assemblers, but does not actually emit anything
4408 @section @code{.if @var{absolute expression}}
4410 @cindex conditional assembly
4411 @cindex @code{if} directive
4412 @code{.if} marks the beginning of a section of code which is only
4413 considered part of the source program being assembled if the argument
4414 (which must be an @var{absolute expression}) is non-zero. The end of
4415 the conditional section of code must be marked by @code{.endif}
4416 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4417 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4418 If you have several conditions to check, @code{.elseif} may be used to avoid
4419 nesting blocks if/else within each subsequent @code{.else} block.
4421 The following variants of @code{.if} are also supported:
4423 @cindex @code{ifdef} directive
4424 @item .ifdef @var{symbol}
4425 Assembles the following section of code if the specified @var{symbol}
4426 has been defined. Note a symbol which has been referenced but not yet defined
4427 is considered to be undefined.
4429 @cindex @code{ifc} directive
4430 @item .ifc @var{string1},@var{string2}
4431 Assembles the following section of code if the two strings are the same. The
4432 strings may be optionally quoted with single quotes. If they are not quoted,
4433 the first string stops at the first comma, and the second string stops at the
4434 end of the line. Strings which contain whitespace should be quoted. The
4435 string comparison is case sensitive.
4437 @cindex @code{ifeq} directive
4438 @item .ifeq @var{absolute expression}
4439 Assembles the following section of code if the argument is zero.
4441 @cindex @code{ifeqs} directive
4442 @item .ifeqs @var{string1},@var{string2}
4443 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4445 @cindex @code{ifge} directive
4446 @item .ifge @var{absolute expression}
4447 Assembles the following section of code if the argument is greater than or
4450 @cindex @code{ifgt} directive
4451 @item .ifgt @var{absolute expression}
4452 Assembles the following section of code if the argument is greater than zero.
4454 @cindex @code{ifle} directive
4455 @item .ifle @var{absolute expression}
4456 Assembles the following section of code if the argument is less than or equal
4459 @cindex @code{iflt} directive
4460 @item .iflt @var{absolute expression}
4461 Assembles the following section of code if the argument is less than zero.
4463 @cindex @code{ifnc} directive
4464 @item .ifnc @var{string1},@var{string2}.
4465 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4466 following section of code if the two strings are not the same.
4468 @cindex @code{ifndef} directive
4469 @cindex @code{ifnotdef} directive
4470 @item .ifndef @var{symbol}
4471 @itemx .ifnotdef @var{symbol}
4472 Assembles the following section of code if the specified @var{symbol}
4473 has not been defined. Both spelling variants are equivalent. Note a symbol
4474 which has been referenced but not yet defined is considered to be undefined.
4476 @cindex @code{ifne} directive
4477 @item .ifne @var{absolute expression}
4478 Assembles the following section of code if the argument is not equal to zero
4479 (in other words, this is equivalent to @code{.if}).
4481 @cindex @code{ifnes} directive
4482 @item .ifnes @var{string1},@var{string2}
4483 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4484 following section of code if the two strings are not the same.
4488 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4490 @cindex @code{incbin} directive
4491 @cindex binary files, including
4492 The @code{incbin} directive includes @var{file} verbatim at the current
4493 location. You can control the search paths used with the @samp{-I} command-line
4494 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4497 The @var{skip} argument skips a number of bytes from the start of the
4498 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4499 read. Note that the data is not aligned in any way, so it is the user's
4500 responsibility to make sure that proper alignment is provided both before and
4501 after the @code{incbin} directive.
4504 @section @code{.include "@var{file}"}
4506 @cindex @code{include} directive
4507 @cindex supporting files, including
4508 @cindex files, including
4509 This directive provides a way to include supporting files at specified
4510 points in your source program. The code from @var{file} is assembled as
4511 if it followed the point of the @code{.include}; when the end of the
4512 included file is reached, assembly of the original file continues. You
4513 can control the search paths used with the @samp{-I} command-line option
4514 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4518 @section @code{.int @var{expressions}}
4520 @cindex @code{int} directive
4521 @cindex integers, 32-bit
4522 Expect zero or more @var{expressions}, of any section, separated by commas.
4523 For each expression, emit a number that, at run time, is the value of that
4524 expression. The byte order and bit size of the number depends on what kind
4525 of target the assembly is for.
4529 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
4530 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4537 @section @code{.internal @var{names}}
4539 @cindex @code{internal} directive
4541 This is one of the ELF visibility directives. The other two are
4542 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4543 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4545 This directive overrides the named symbols default visibility (which is set by
4546 their binding: local, global or weak). The directive sets the visibility to
4547 @code{internal} which means that the symbols are considered to be @code{hidden}
4548 (i.e., not visible to other components), and that some extra, processor specific
4549 processing must also be performed upon the symbols as well.
4553 @section @code{.irp @var{symbol},@var{values}}@dots{}
4555 @cindex @code{irp} directive
4556 Evaluate a sequence of statements assigning different values to @var{symbol}.
4557 The sequence of statements starts at the @code{.irp} directive, and is
4558 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4559 set to @var{value}, and the sequence of statements is assembled. If no
4560 @var{value} is listed, the sequence of statements is assembled once, with
4561 @var{symbol} set to the null string. To refer to @var{symbol} within the
4562 sequence of statements, use @var{\symbol}.
4564 For example, assembling
4572 is equivalent to assembling
4581 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4583 @cindex @code{irpc} directive
4584 Evaluate a sequence of statements assigning different values to @var{symbol}.
4585 The sequence of statements starts at the @code{.irpc} directive, and is
4586 terminated by an @code{.endr} directive. For each character in @var{value},
4587 @var{symbol} is set to the character, and the sequence of statements is
4588 assembled. If no @var{value} is listed, the sequence of statements is
4589 assembled once, with @var{symbol} set to the null string. To refer to
4590 @var{symbol} within the sequence of statements, use @var{\symbol}.
4592 For example, assembling
4600 is equivalent to assembling
4609 @section @code{.lcomm @var{symbol} , @var{length}}
4611 @cindex @code{lcomm} directive
4612 @cindex local common symbols
4613 @cindex symbols, local common
4614 Reserve @var{length} (an absolute expression) bytes for a local common
4615 denoted by @var{symbol}. The section and value of @var{symbol} are
4616 those of the new local common. The addresses are allocated in the bss
4617 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4618 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4619 not visible to @code{@value{LD}}.
4622 Some targets permit a third argument to be used with @code{.lcomm}. This
4623 argument specifies the desired alignment of the symbol in the bss section.
4627 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4628 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4632 @section @code{.lflags}
4634 @cindex @code{lflags} directive (ignored)
4635 @command{@value{AS}} accepts this directive, for compatibility with other
4636 assemblers, but ignores it.
4638 @ifclear no-line-dir
4640 @section @code{.line @var{line-number}}
4642 @cindex @code{line} directive
4646 @section @code{.ln @var{line-number}}
4648 @cindex @code{ln} directive
4650 @cindex logical line number
4652 Change the logical line number. @var{line-number} must be an absolute
4653 expression. The next line has that logical line number. Therefore any other
4654 statements on the current line (after a statement separator character) are
4655 reported as on logical line number @var{line-number} @minus{} 1. One day
4656 @command{@value{AS}} will no longer support this directive: it is recognized only
4657 for compatibility with existing assembler programs.
4661 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
4662 not available; use the synonym @code{.ln} in that context.
4667 @ifclear no-line-dir
4668 Even though this is a directive associated with the @code{a.out} or
4669 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
4670 when producing COFF output, and treats @samp{.line} as though it
4671 were the COFF @samp{.ln} @emph{if} it is found outside a
4672 @code{.def}/@code{.endef} pair.
4674 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4675 used by compilers to generate auxiliary symbol information for
4680 @section @code{.linkonce [@var{type}]}
4682 @cindex @code{linkonce} directive
4683 @cindex common sections
4684 Mark the current section so that the linker only includes a single copy of it.
4685 This may be used to include the same section in several different object files,
4686 but ensure that the linker will only include it once in the final output file.
4687 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4688 Duplicate sections are detected based on the section name, so it should be
4691 This directive is only supported by a few object file formats; as of this
4692 writing, the only object file format which supports it is the Portable
4693 Executable format used on Windows NT.
4695 The @var{type} argument is optional. If specified, it must be one of the
4696 following strings. For example:
4700 Not all types may be supported on all object file formats.
4704 Silently discard duplicate sections. This is the default.
4707 Warn if there are duplicate sections, but still keep only one copy.
4710 Warn if any of the duplicates have different sizes.
4713 Warn if any of the duplicates do not have exactly the same contents.
4717 @section @code{.ln @var{line-number}}
4719 @cindex @code{ln} directive
4720 @ifclear no-line-dir
4721 @samp{.ln} is a synonym for @samp{.line}.
4724 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
4725 must be an absolute expression. The next line has that logical
4726 line number, so any other statements on the current line (after a
4727 statement separator character @code{;}) are reported as on logical
4728 line number @var{line-number} @minus{} 1.
4731 This directive is accepted, but ignored, when @command{@value{AS}} is
4732 configured for @code{b.out}; its effect is only associated with COFF
4738 @section @code{.mri @var{val}}
4740 @cindex @code{mri} directive
4741 @cindex MRI mode, temporarily
4742 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
4743 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
4744 affects code assembled until the next @code{.mri} directive, or until the end
4745 of the file. @xref{M, MRI mode, MRI mode}.
4748 @section @code{.list}
4750 @cindex @code{list} directive
4751 @cindex listing control, turning on
4752 Control (in conjunction with the @code{.nolist} directive) whether or
4753 not assembly listings are generated. These two directives maintain an
4754 internal counter (which is zero initially). @code{.list} increments the
4755 counter, and @code{.nolist} decrements it. Assembly listings are
4756 generated whenever the counter is greater than zero.
4758 By default, listings are disabled. When you enable them (with the
4759 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4760 the initial value of the listing counter is one.
4763 @section @code{.long @var{expressions}}
4765 @cindex @code{long} directive
4766 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
4769 @c no one seems to know what this is for or whether this description is
4770 @c what it really ought to do
4772 @section @code{.lsym @var{symbol}, @var{expression}}
4774 @cindex @code{lsym} directive
4775 @cindex symbol, not referenced in assembly
4776 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
4777 the hash table, ensuring it cannot be referenced by name during the
4778 rest of the assembly. This sets the attributes of the symbol to be
4779 the same as the expression value:
4781 @var{other} = @var{descriptor} = 0
4782 @var{type} = @r{(section of @var{expression})}
4783 @var{value} = @var{expression}
4786 The new symbol is not flagged as external.
4790 @section @code{.macro}
4793 The commands @code{.macro} and @code{.endm} allow you to define macros that
4794 generate assembly output. For example, this definition specifies a macro
4795 @code{sum} that puts a sequence of numbers into memory:
4798 .macro sum from=0, to=5
4807 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
4819 @item .macro @var{macname}
4820 @itemx .macro @var{macname} @var{macargs} @dots{}
4821 @cindex @code{macro} directive
4822 Begin the definition of a macro called @var{macname}. If your macro
4823 definition requires arguments, specify their names after the macro name,
4824 separated by commas or spaces. You can supply a default value for any
4825 macro argument by following the name with @samp{=@var{deflt}}. For
4826 example, these are all valid @code{.macro} statements:
4830 Begin the definition of a macro called @code{comm}, which takes no
4833 @item .macro plus1 p, p1
4834 @itemx .macro plus1 p p1
4835 Either statement begins the definition of a macro called @code{plus1},
4836 which takes two arguments; within the macro definition, write
4837 @samp{\p} or @samp{\p1} to evaluate the arguments.
4839 @item .macro reserve_str p1=0 p2
4840 Begin the definition of a macro called @code{reserve_str}, with two
4841 arguments. The first argument has a default value, but not the second.
4842 After the definition is complete, you can call the macro either as
4843 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
4844 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
4845 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
4846 @samp{0}, and @samp{\p2} evaluating to @var{b}).
4849 When you call a macro, you can specify the argument values either by
4850 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
4851 @samp{sum to=17, from=9}.
4854 @cindex @code{endm} directive
4855 Mark the end of a macro definition.
4858 @cindex @code{exitm} directive
4859 Exit early from the current macro definition.
4861 @cindex number of macros executed
4862 @cindex macros, count executed
4864 @command{@value{AS}} maintains a counter of how many macros it has
4865 executed in this pseudo-variable; you can copy that number to your
4866 output with @samp{\@@}, but @emph{only within a macro definition}.
4868 @item LOCAL @var{name} [ , @dots{} ]
4869 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
4870 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
4871 @xref{Altmacro,,@code{.altmacro}}.
4875 @section @code{.altmacro}
4876 Enable alternate macro mode, enabling:
4879 @item LOCAL @var{name} [ , @dots{} ]
4880 One additional directive, @code{LOCAL}, is available. It is used to
4881 generate a string replacement for each of the @var{name} arguments, and
4882 replace any instances of @var{name} in each macro expansion. The
4883 replacement string is unique in the assembly, and different for each
4884 separate macro expansion. @code{LOCAL} allows you to write macros that
4885 define symbols, without fear of conflict between separate macro expansions.
4887 @item String delimiters
4888 You can write strings delimited in these other ways besides
4889 @code{"@var{string}"}:
4892 @item '@var{string}'
4893 You can delimit strings with single-quote charaters.
4895 @item <@var{string}>
4896 You can delimit strings with matching angle brackets.
4899 @item single-character string escape
4900 To include any single character literally in a string (even if the
4901 character would otherwise have some special meaning), you can prefix the
4902 character with @samp{!} (an exclamation mark). For example, you can
4903 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4905 @item Expression results as strings
4906 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4907 and use the result as a string.
4911 @section @code{.noaltmacro}
4912 Disable alternate macro mode. @ref{Altmacro}
4915 @section @code{.nolist}
4917 @cindex @code{nolist} directive
4918 @cindex listing control, turning off
4919 Control (in conjunction with the @code{.list} directive) whether or
4920 not assembly listings are generated. These two directives maintain an
4921 internal counter (which is zero initially). @code{.list} increments the
4922 counter, and @code{.nolist} decrements it. Assembly listings are
4923 generated whenever the counter is greater than zero.
4926 @section @code{.octa @var{bignums}}
4928 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
4929 @cindex @code{octa} directive
4930 @cindex integer, 16-byte
4931 @cindex sixteen byte integer
4932 This directive expects zero or more bignums, separated by commas. For each
4933 bignum, it emits a 16-byte integer.
4935 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
4936 hence @emph{octa}-word for 16 bytes.
4939 @section @code{.org @var{new-lc} , @var{fill}}
4941 @cindex @code{org} directive
4942 @cindex location counter, advancing
4943 @cindex advancing location counter
4944 @cindex current address, advancing
4945 Advance the location counter of the current section to
4946 @var{new-lc}. @var{new-lc} is either an absolute expression or an
4947 expression with the same section as the current subsection. That is,
4948 you can't use @code{.org} to cross sections: if @var{new-lc} has the
4949 wrong section, the @code{.org} directive is ignored. To be compatible
4950 with former assemblers, if the section of @var{new-lc} is absolute,
4951 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
4952 is the same as the current subsection.
4954 @code{.org} may only increase the location counter, or leave it
4955 unchanged; you cannot use @code{.org} to move the location counter
4958 @c double negative used below "not undefined" because this is a specific
4959 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
4960 @c section. doc@cygnus.com 18feb91
4961 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
4962 may not be undefined. If you really detest this restriction we eagerly await
4963 a chance to share your improved assembler.
4965 Beware that the origin is relative to the start of the section, not
4966 to the start of the subsection. This is compatible with other
4967 people's assemblers.
4969 When the location counter (of the current subsection) is advanced, the
4970 intervening bytes are filled with @var{fill} which should be an
4971 absolute expression. If the comma and @var{fill} are omitted,
4972 @var{fill} defaults to zero.
4975 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4977 @cindex padding the location counter given a power of two
4978 @cindex @code{p2align} directive
4979 Pad the location counter (in the current subsection) to a particular
4980 storage boundary. The first expression (which must be absolute) is the
4981 number of low-order zero bits the location counter must have after
4982 advancement. For example @samp{.p2align 3} advances the location
4983 counter until it a multiple of 8. If the location counter is already a
4984 multiple of 8, no change is needed.
4986 The second expression (also absolute) gives the fill value to be stored in the
4987 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4988 padding bytes are normally zero. However, on some systems, if the section is
4989 marked as containing code and the fill value is omitted, the space is filled
4990 with no-op instructions.
4992 The third expression is also absolute, and is also optional. If it is present,
4993 it is the maximum number of bytes that should be skipped by this alignment
4994 directive. If doing the alignment would require skipping more bytes than the
4995 specified maximum, then the alignment is not done at all. You can omit the
4996 fill value (the second argument) entirely by simply using two commas after the
4997 required alignment; this can be useful if you want the alignment to be filled
4998 with no-op instructions when appropriate.
5000 @cindex @code{p2alignw} directive
5001 @cindex @code{p2alignl} directive
5002 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5003 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5004 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5005 fill pattern as a four byte longword value. For example, @code{.p2alignw
5006 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5007 filled in with the value 0x368d (the exact placement of the bytes depends upon
5008 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5013 @section @code{.previous}
5015 @cindex @code{previous} directive
5016 @cindex Section Stack
5017 This is one of the ELF section stack manipulation directives. The others are
5018 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5019 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5020 (@pxref{PopSection}).
5022 This directive swaps the current section (and subsection) with most recently
5023 referenced section (and subsection) prior to this one. Multiple
5024 @code{.previous} directives in a row will flip between two sections (and their
5027 In terms of the section stack, this directive swaps the current section with
5028 the top section on the section stack.
5033 @section @code{.popsection}
5035 @cindex @code{popsection} directive
5036 @cindex Section Stack
5037 This is one of the ELF section stack manipulation directives. The others are
5038 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5039 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5042 This directive replaces the current section (and subsection) with the top
5043 section (and subsection) on the section stack. This section is popped off the
5048 @section @code{.print @var{string}}
5050 @cindex @code{print} directive
5051 @command{@value{AS}} will print @var{string} on the standard output during
5052 assembly. You must put @var{string} in double quotes.
5056 @section @code{.protected @var{names}}
5058 @cindex @code{protected} directive
5060 This is one of the ELF visibility directives. The other two are
5061 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5063 This directive overrides the named symbols default visibility (which is set by
5064 their binding: local, global or weak). The directive sets the visibility to
5065 @code{protected} which means that any references to the symbols from within the
5066 components that defines them must be resolved to the definition in that
5067 component, even if a definition in another component would normally preempt
5072 @section @code{.psize @var{lines} , @var{columns}}
5074 @cindex @code{psize} directive
5075 @cindex listing control: paper size
5076 @cindex paper size, for listings
5077 Use this directive to declare the number of lines---and, optionally, the
5078 number of columns---to use for each page, when generating listings.
5080 If you do not use @code{.psize}, listings use a default line-count
5081 of 60. You may omit the comma and @var{columns} specification; the
5082 default width is 200 columns.
5084 @command{@value{AS}} generates formfeeds whenever the specified number of
5085 lines is exceeded (or whenever you explicitly request one, using
5088 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5089 those explicitly specified with @code{.eject}.
5092 @section @code{.purgem @var{name}}
5094 @cindex @code{purgem} directive
5095 Undefine the macro @var{name}, so that later uses of the string will not be
5096 expanded. @xref{Macro}.
5100 @section @code{.pushsection @var{name} , @var{subsection}}
5102 @cindex @code{pushsection} directive
5103 @cindex Section Stack
5104 This is one of the ELF section stack manipulation directives. The others are
5105 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5106 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5109 This directive pushes the current section (and subsection) onto the
5110 top of the section stack, and then replaces the current section and
5111 subsection with @code{name} and @code{subsection}.
5115 @section @code{.quad @var{bignums}}
5117 @cindex @code{quad} directive
5118 @code{.quad} expects zero or more bignums, separated by commas. For
5119 each bignum, it emits
5121 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5122 warning message; and just takes the lowest order 8 bytes of the bignum.
5123 @cindex eight-byte integer
5124 @cindex integer, 8-byte
5126 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5127 hence @emph{quad}-word for 8 bytes.
5130 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5131 warning message; and just takes the lowest order 16 bytes of the bignum.
5132 @cindex sixteen-byte integer
5133 @cindex integer, 16-byte
5137 @section @code{.rept @var{count}}
5139 @cindex @code{rept} directive
5140 Repeat the sequence of lines between the @code{.rept} directive and the next
5141 @code{.endr} directive @var{count} times.
5143 For example, assembling
5151 is equivalent to assembling
5160 @section @code{.sbttl "@var{subheading}"}
5162 @cindex @code{sbttl} directive
5163 @cindex subtitles for listings
5164 @cindex listing control: subtitle
5165 Use @var{subheading} as the title (third line, immediately after the
5166 title line) when generating assembly listings.
5168 This directive affects subsequent pages, as well as the current page if
5169 it appears within ten lines of the top of a page.
5173 @section @code{.scl @var{class}}
5175 @cindex @code{scl} directive
5176 @cindex symbol storage class (COFF)
5177 @cindex COFF symbol storage class
5178 Set the storage-class value for a symbol. This directive may only be
5179 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5180 whether a symbol is static or external, or it may record further
5181 symbolic debugging information.
5184 The @samp{.scl} directive is primarily associated with COFF output; when
5185 configured to generate @code{b.out} output format, @command{@value{AS}}
5186 accepts this directive but ignores it.
5192 @section @code{.section @var{name}}
5194 @cindex named section
5195 Use the @code{.section} directive to assemble the following code into a section
5198 This directive is only supported for targets that actually support arbitrarily
5199 named sections; on @code{a.out} targets, for example, it is not accepted, even
5200 with a standard @code{a.out} section name.
5204 @c only print the extra heading if both COFF and ELF are set
5205 @subheading COFF Version
5208 @cindex @code{section} directive (COFF version)
5209 For COFF targets, the @code{.section} directive is used in one of the following
5213 .section @var{name}[, "@var{flags}"]
5214 .section @var{name}[, @var{subsegment}]
5217 If the optional argument is quoted, it is taken as flags to use for the
5218 section. Each flag is a single character. The following flags are recognized:
5221 bss section (uninitialized data)
5223 section is not loaded
5233 shared section (meaningful for PE targets)
5235 ignored. (For compatibility with the ELF version)
5238 If no flags are specified, the default flags depend upon the section name. If
5239 the section name is not recognized, the default will be for the section to be
5240 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5241 from the section, rather than adding them, so if they are used on their own it
5242 will be as if no flags had been specified at all.
5244 If the optional argument to the @code{.section} directive is not quoted, it is
5245 taken as a subsegment number (@pxref{Sub-Sections}).
5250 @c only print the extra heading if both COFF and ELF are set
5251 @subheading ELF Version
5254 @cindex Section Stack
5255 This is one of the ELF section stack manipulation directives. The others are
5256 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5257 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5258 @code{.previous} (@pxref{Previous}).
5260 @cindex @code{section} directive (ELF version)
5261 For ELF targets, the @code{.section} directive is used like this:
5264 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]
5267 The optional @var{flags} argument is a quoted string which may contain any
5268 combination of the following characters:
5271 section is allocatable
5275 section is executable
5277 section is mergeable
5279 section contains zero terminated strings
5281 section is a member of a section group
5283 section is used for thread-local-storage
5286 The optional @var{type} argument may contain one of the following constants:
5289 section contains data
5291 section does not contain data (i.e., section only occupies space)
5293 section contains data which is used by things other than the program
5296 Note on targets where the @code{@@} character is the start of a comment (eg
5297 ARM) then another character is used instead. For example the ARM port uses the
5300 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5301 be specified as well as an extra argument - @var{entsize} - like this:
5304 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5307 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5308 constants, each @var{entsize} octets long. Sections with both @code{M} and
5309 @code{S} must contain zero terminated strings where each character is
5310 @var{entsize} bytes long. The linker may remove duplicates within sections with
5311 the same name, same entity size and same flags. @var{entsize} must be an
5312 absolute expression.
5314 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5315 be present along with an additional field like this:
5318 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5321 The @var{GroupName} field specifies the name of the section group to which this
5322 particular section belongs. The optional linkage field can contain:
5325 indicates that only one copy of this section should be retained
5330 Note - if both the @var{M} and @var{G} flags are present then the fields for
5331 the Merge flag should come first, like this:
5334 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5337 If no flags are specified, the default flags depend upon the section name. If
5338 the section name is not recognized, the default will be for the section to have
5339 none of the above flags: it will not be allocated in memory, nor writable, nor
5340 executable. The section will contain data.
5342 For ELF targets, the assembler supports another type of @code{.section}
5343 directive for compatibility with the Solaris assembler:
5346 .section "@var{name}"[, @var{flags}...]
5349 Note that the section name is quoted. There may be a sequence of comma
5353 section is allocatable
5357 section is executable
5359 section is used for thread local storage
5362 This directive replaces the current section and subsection. See the
5363 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
5364 some examples of how this directive and the other section stack directives
5370 @section @code{.set @var{symbol}, @var{expression}}
5372 @cindex @code{set} directive
5373 @cindex symbol value, setting
5374 Set the value of @var{symbol} to @var{expression}. This
5375 changes @var{symbol}'s value and type to conform to
5376 @var{expression}. If @var{symbol} was flagged as external, it remains
5377 flagged (@pxref{Symbol Attributes}).
5379 You may @code{.set} a symbol many times in the same assembly.
5381 If you @code{.set} a global symbol, the value stored in the object
5382 file is the last value stored into it.
5385 The syntax for @code{set} on the HPPA is
5386 @samp{@var{symbol} .set @var{expression}}.
5390 @section @code{.short @var{expressions}}
5392 @cindex @code{short} directive
5394 @code{.short} is normally the same as @samp{.word}.
5395 @xref{Word,,@code{.word}}.
5397 In some configurations, however, @code{.short} and @code{.word} generate
5398 numbers of different lengths; @pxref{Machine Dependencies}.
5402 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5405 This expects zero or more @var{expressions}, and emits
5406 a 16 bit number for each.
5411 @section @code{.single @var{flonums}}
5413 @cindex @code{single} directive
5414 @cindex floating point numbers (single)
5415 This directive assembles zero or more flonums, separated by commas. It
5416 has the same effect as @code{.float}.
5418 The exact kind of floating point numbers emitted depends on how
5419 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5423 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5424 numbers in @sc{ieee} format.
5430 @section @code{.size}
5432 This directive is used to set the size associated with a symbol.
5436 @c only print the extra heading if both COFF and ELF are set
5437 @subheading COFF Version
5440 @cindex @code{size} directive (COFF version)
5441 For COFF targets, the @code{.size} directive is only permitted inside
5442 @code{.def}/@code{.endef} pairs. It is used like this:
5445 .size @var{expression}
5449 @samp{.size} is only meaningful when generating COFF format output; when
5450 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5457 @c only print the extra heading if both COFF and ELF are set
5458 @subheading ELF Version
5461 @cindex @code{size} directive (ELF version)
5462 For ELF targets, the @code{.size} directive is used like this:
5465 .size @var{name} , @var{expression}
5468 This directive sets the size associated with a symbol @var{name}.
5469 The size in bytes is computed from @var{expression} which can make use of label
5470 arithmetic. This directive is typically used to set the size of function
5476 @section @code{.sleb128 @var{expressions}}
5478 @cindex @code{sleb128} directive
5479 @var{sleb128} stands for ``signed little endian base 128.'' This is a
5480 compact, variable length representation of numbers used by the DWARF
5481 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
5483 @ifclear no-space-dir
5485 @section @code{.skip @var{size} , @var{fill}}
5487 @cindex @code{skip} directive
5488 @cindex filling memory
5489 This directive emits @var{size} bytes, each of value @var{fill}. Both
5490 @var{size} and @var{fill} are absolute expressions. If the comma and
5491 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
5495 @section @code{.space @var{size} , @var{fill}}
5497 @cindex @code{space} directive
5498 @cindex filling memory
5499 This directive emits @var{size} bytes, each of value @var{fill}. Both
5500 @var{size} and @var{fill} are absolute expressions. If the comma
5501 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
5506 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
5507 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
5508 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
5509 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
5518 @section @code{.space}
5519 @cindex @code{space} directive
5521 On the AMD 29K, this directive is ignored; it is accepted for
5522 compatibility with other AMD 29K assemblers.
5525 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
5526 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
5532 @section @code{.stabd, .stabn, .stabs}
5534 @cindex symbolic debuggers, information for
5535 @cindex @code{stab@var{x}} directives
5536 There are three directives that begin @samp{.stab}.
5537 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
5538 The symbols are not entered in the @command{@value{AS}} hash table: they
5539 cannot be referenced elsewhere in the source file.
5540 Up to five fields are required:
5544 This is the symbol's name. It may contain any character except
5545 @samp{\000}, so is more general than ordinary symbol names. Some
5546 debuggers used to code arbitrarily complex structures into symbol names
5550 An absolute expression. The symbol's type is set to the low 8 bits of
5551 this expression. Any bit pattern is permitted, but @code{@value{LD}}
5552 and debuggers choke on silly bit patterns.
5555 An absolute expression. The symbol's ``other'' attribute is set to the
5556 low 8 bits of this expression.
5559 An absolute expression. The symbol's descriptor is set to the low 16
5560 bits of this expression.
5563 An absolute expression which becomes the symbol's value.
5566 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
5567 or @code{.stabs} statement, the symbol has probably already been created;
5568 you get a half-formed symbol in your object file. This is
5569 compatible with earlier assemblers!
5572 @cindex @code{stabd} directive
5573 @item .stabd @var{type} , @var{other} , @var{desc}
5575 The ``name'' of the symbol generated is not even an empty string.
5576 It is a null pointer, for compatibility. Older assemblers used a
5577 null pointer so they didn't waste space in object files with empty
5580 The symbol's value is set to the location counter,
5581 relocatably. When your program is linked, the value of this symbol
5582 is the address of the location counter when the @code{.stabd} was
5585 @cindex @code{stabn} directive
5586 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
5587 The name of the symbol is set to the empty string @code{""}.
5589 @cindex @code{stabs} directive
5590 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
5591 All five fields are specified.
5597 @section @code{.string} "@var{str}"
5599 @cindex string, copying to object file
5600 @cindex @code{string} directive
5602 Copy the characters in @var{str} to the object file. You may specify more than
5603 one string to copy, separated by commas. Unless otherwise specified for a
5604 particular machine, the assembler marks the end of each string with a 0 byte.
5605 You can use any of the escape sequences described in @ref{Strings,,Strings}.
5608 @section @code{.struct @var{expression}}
5610 @cindex @code{struct} directive
5611 Switch to the absolute section, and set the section offset to @var{expression},
5612 which must be an absolute expression. You might use this as follows:
5621 This would define the symbol @code{field1} to have the value 0, the symbol
5622 @code{field2} to have the value 4, and the symbol @code{field3} to have the
5623 value 8. Assembly would be left in the absolute section, and you would need to
5624 use a @code{.section} directive of some sort to change to some other section
5625 before further assembly.
5629 @section @code{.subsection @var{name}}
5631 @cindex @code{subsection} directive
5632 @cindex Section Stack
5633 This is one of the ELF section stack manipulation directives. The others are
5634 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
5635 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5638 This directive replaces the current subsection with @code{name}. The current
5639 section is not changed. The replaced subsection is put onto the section stack
5640 in place of the then current top of stack subsection.
5645 @section @code{.symver}
5646 @cindex @code{symver} directive
5647 @cindex symbol versioning
5648 @cindex versions of symbols
5649 Use the @code{.symver} directive to bind symbols to specific version nodes
5650 within a source file. This is only supported on ELF platforms, and is
5651 typically used when assembling files to be linked into a shared library.
5652 There are cases where it may make sense to use this in objects to be bound
5653 into an application itself so as to override a versioned symbol from a
5656 For ELF targets, the @code{.symver} directive can be used like this:
5658 .symver @var{name}, @var{name2@@nodename}
5660 If the symbol @var{name} is defined within the file
5661 being assembled, the @code{.symver} directive effectively creates a symbol
5662 alias with the name @var{name2@@nodename}, and in fact the main reason that we
5663 just don't try and create a regular alias is that the @var{@@} character isn't
5664 permitted in symbol names. The @var{name2} part of the name is the actual name
5665 of the symbol by which it will be externally referenced. The name @var{name}
5666 itself is merely a name of convenience that is used so that it is possible to
5667 have definitions for multiple versions of a function within a single source
5668 file, and so that the compiler can unambiguously know which version of a
5669 function is being mentioned. The @var{nodename} portion of the alias should be
5670 the name of a node specified in the version script supplied to the linker when
5671 building a shared library. If you are attempting to override a versioned
5672 symbol from a shared library, then @var{nodename} should correspond to the
5673 nodename of the symbol you are trying to override.
5675 If the symbol @var{name} is not defined within the file being assembled, all
5676 references to @var{name} will be changed to @var{name2@@nodename}. If no
5677 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
5680 Another usage of the @code{.symver} directive is:
5682 .symver @var{name}, @var{name2@@@@nodename}
5684 In this case, the symbol @var{name} must exist and be defined within
5685 the file being assembled. It is similar to @var{name2@@nodename}. The
5686 difference is @var{name2@@@@nodename} will also be used to resolve
5687 references to @var{name2} by the linker.
5689 The third usage of the @code{.symver} directive is:
5691 .symver @var{name}, @var{name2@@@@@@nodename}
5693 When @var{name} is not defined within the
5694 file being assembled, it is treated as @var{name2@@nodename}. When
5695 @var{name} is defined within the file being assembled, the symbol
5696 name, @var{name}, will be changed to @var{name2@@@@nodename}.
5701 @section @code{.tag @var{structname}}
5703 @cindex COFF structure debugging
5704 @cindex structure debugging, COFF
5705 @cindex @code{tag} directive
5706 This directive is generated by compilers to include auxiliary debugging
5707 information in the symbol table. It is only permitted inside
5708 @code{.def}/@code{.endef} pairs. Tags are used to link structure
5709 definitions in the symbol table with instances of those structures.
5712 @samp{.tag} is only used when generating COFF format output; when
5713 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5719 @section @code{.text @var{subsection}}
5721 @cindex @code{text} directive
5722 Tells @command{@value{AS}} to assemble the following statements onto the end of
5723 the text subsection numbered @var{subsection}, which is an absolute
5724 expression. If @var{subsection} is omitted, subsection number zero
5728 @section @code{.title "@var{heading}"}
5730 @cindex @code{title} directive
5731 @cindex listing control: title line
5732 Use @var{heading} as the title (second line, immediately after the
5733 source file name and pagenumber) when generating assembly listings.
5735 This directive affects subsequent pages, as well as the current page if
5736 it appears within ten lines of the top of a page.
5740 @section @code{.type}
5742 This directive is used to set the type of a symbol.
5746 @c only print the extra heading if both COFF and ELF are set
5747 @subheading COFF Version
5750 @cindex COFF symbol type
5751 @cindex symbol type, COFF
5752 @cindex @code{type} directive (COFF version)
5753 For COFF targets, this directive is permitted only within
5754 @code{.def}/@code{.endef} pairs. It is used like this:
5760 This records the integer @var{int} as the type attribute of a symbol table
5764 @samp{.type} is associated only with COFF format output; when
5765 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
5766 directive but ignores it.
5772 @c only print the extra heading if both COFF and ELF are set
5773 @subheading ELF Version
5776 @cindex ELF symbol type
5777 @cindex symbol type, ELF
5778 @cindex @code{type} directive (ELF version)
5779 For ELF targets, the @code{.type} directive is used like this:
5782 .type @var{name} , @var{type description}
5785 This sets the type of symbol @var{name} to be either a
5786 function symbol or an object symbol. There are five different syntaxes
5787 supported for the @var{type description} field, in order to provide
5788 compatibility with various other assemblers. The syntaxes supported are:
5791 .type <name>,#function
5792 .type <name>,#object
5794 .type <name>,@@function
5795 .type <name>,@@object
5797 .type <name>,%function
5798 .type <name>,%object
5800 .type <name>,"function"
5801 .type <name>,"object"
5803 .type <name> STT_FUNCTION
5804 .type <name> STT_OBJECT
5810 @section @code{.uleb128 @var{expressions}}
5812 @cindex @code{uleb128} directive
5813 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
5814 compact, variable length representation of numbers used by the DWARF
5815 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
5819 @section @code{.val @var{addr}}
5821 @cindex @code{val} directive
5822 @cindex COFF value attribute
5823 @cindex value attribute, COFF
5824 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5825 records the address @var{addr} as the value attribute of a symbol table
5829 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
5830 configured for @code{b.out}, it accepts this directive but ignores it.
5836 @section @code{.version "@var{string}"}
5838 @cindex @code{version} directive
5839 This directive creates a @code{.note} section and places into it an ELF
5840 formatted note of type NT_VERSION. The note's name is set to @code{string}.
5845 @section @code{.vtable_entry @var{table}, @var{offset}}
5847 @cindex @code{vtable_entry}
5848 This directive finds or creates a symbol @code{table} and creates a
5849 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
5852 @section @code{.vtable_inherit @var{child}, @var{parent}}
5854 @cindex @code{vtable_inherit}
5855 This directive finds the symbol @code{child} and finds or creates the symbol
5856 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
5857 parent whose addend is the value of the child symbol. As a special case the
5858 parent name of @code{0} is treated as refering the @code{*ABS*} section.
5862 @section @code{.weak @var{names}}
5864 @cindex @code{weak} directive
5865 This directive sets the weak attribute on the comma separated list of symbol
5866 @code{names}. If the symbols do not already exist, they will be created.
5868 Weak symbols are supported in COFF as a GNU extension. This directive
5869 sets the weak attribute on the comma separated list of symbol
5870 @code{names}. If the symbols do not already exist, they will be created.
5873 @code{.weak @var{name} [ < = | == > @var{alternate}] [, ...]}
5876 On the PE target, weak aliases are supported natively. Weak aliases
5877 (usually called "weak externals" in PE) are created when an alternate
5878 name is specified. When a weak symbol is linked and the symbol is not
5879 defined, the weak symbol becomes an alias for the alternate symbol. If
5880 one equal sign is used, the linker searches for defined symbols within
5881 other objects and libraries. This is the usual mode, historically
5882 called "lazy externals." Otherwise, when two equal signs are used,
5883 the linker searches for defined symbols only within other objects.
5885 Non-alias weak symbols are supported on PE as a GNU extension.
5888 @section @code{.word @var{expressions}}
5890 @cindex @code{word} directive
5891 This directive expects zero or more @var{expressions}, of any section,
5892 separated by commas.
5895 For each expression, @command{@value{AS}} emits a 32-bit number.
5898 For each expression, @command{@value{AS}} emits a 16-bit number.
5903 The size of the number emitted, and its byte order,
5904 depend on what target computer the assembly is for.
5907 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
5908 @c happen---32-bit addressability, period; no long/short jumps.
5909 @ifset DIFF-TBL-KLUGE
5910 @cindex difference tables altered
5911 @cindex altered difference tables
5913 @emph{Warning: Special Treatment to support Compilers}
5917 Machines with a 32-bit address space, but that do less than 32-bit
5918 addressing, require the following special treatment. If the machine of
5919 interest to you does 32-bit addressing (or doesn't require it;
5920 @pxref{Machine Dependencies}), you can ignore this issue.
5923 In order to assemble compiler output into something that works,
5924 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
5925 Directives of the form @samp{.word sym1-sym2} are often emitted by
5926 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
5927 directive of the form @samp{.word sym1-sym2}, and the difference between
5928 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
5929 creates a @dfn{secondary jump table}, immediately before the next label.
5930 This secondary jump table is preceded by a short-jump to the
5931 first byte after the secondary table. This short-jump prevents the flow
5932 of control from accidentally falling into the new table. Inside the
5933 table is a long-jump to @code{sym2}. The original @samp{.word}
5934 contains @code{sym1} minus the address of the long-jump to
5937 If there were several occurrences of @samp{.word sym1-sym2} before the
5938 secondary jump table, all of them are adjusted. If there was a
5939 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
5940 long-jump to @code{sym4} is included in the secondary jump table,
5941 and the @code{.word} directives are adjusted to contain @code{sym3}
5942 minus the address of the long-jump to @code{sym4}; and so on, for as many
5943 entries in the original jump table as necessary.
5946 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
5947 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
5948 assembly language programmers.
5951 @c end DIFF-TBL-KLUGE
5954 @section Deprecated Directives
5956 @cindex deprecated directives
5957 @cindex obsolescent directives
5958 One day these directives won't work.
5959 They are included for compatibility with older assemblers.
5966 @node Machine Dependencies
5967 @chapter Machine Dependent Features
5969 @cindex machine dependencies
5970 The machine instruction sets are (almost by definition) different on
5971 each machine where @command{@value{AS}} runs. Floating point representations
5972 vary as well, and @command{@value{AS}} often supports a few additional
5973 directives or command-line options for compatibility with other
5974 assemblers on a particular platform. Finally, some versions of
5975 @command{@value{AS}} support special pseudo-instructions for branch
5978 This chapter discusses most of these differences, though it does not
5979 include details on any machine's instruction set. For details on that
5980 subject, see the hardware manufacturer's manual.
5984 * AMD29K-Dependent:: AMD 29K Dependent Features
5987 * Alpha-Dependent:: Alpha Dependent Features
5990 * ARC-Dependent:: ARC Dependent Features
5993 * ARM-Dependent:: ARM Dependent Features
5996 * CRIS-Dependent:: CRIS Dependent Features
5999 * D10V-Dependent:: D10V Dependent Features
6002 * D30V-Dependent:: D30V Dependent Features
6005 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6008 * H8/500-Dependent:: Renesas H8/500 Dependent Features
6011 * HPPA-Dependent:: HPPA Dependent Features
6014 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6017 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6020 * i860-Dependent:: Intel 80860 Dependent Features
6023 * i960-Dependent:: Intel 80960 Dependent Features
6026 * IP2K-Dependent:: IP2K Dependent Features
6029 * M32R-Dependent:: M32R Dependent Features
6032 * M68K-Dependent:: M680x0 Dependent Features
6035 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6038 * M88K-Dependent:: M880x0 Dependent Features
6041 * MIPS-Dependent:: MIPS Dependent Features
6044 * MMIX-Dependent:: MMIX Dependent Features
6047 * MSP430-Dependent:: MSP430 Dependent Features
6050 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6051 * SH64-Dependent:: SuperH SH64 Dependent Features
6054 * PDP-11-Dependent:: PDP-11 Dependent Features
6057 * PJ-Dependent:: picoJava Dependent Features
6060 * PPC-Dependent:: PowerPC Dependent Features
6063 * Sparc-Dependent:: SPARC Dependent Features
6066 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6069 * V850-Dependent:: V850 Dependent Features
6072 * Xtensa-Dependent:: Xtensa Dependent Features
6075 * Z8000-Dependent:: Z8000 Dependent Features
6078 * Vax-Dependent:: VAX Dependent Features
6085 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6086 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6087 @c peculiarity: to preserve cross-references, there must be a node called
6088 @c "Machine Dependencies". Hence the conditional nodenames in each
6089 @c major node below. Node defaulting in makeinfo requires adjacency of
6090 @c node and sectioning commands; hence the repetition of @chapter BLAH
6091 @c in both conditional blocks.
6094 @include c-a29k.texi
6098 @include c-alpha.texi
6110 @include c-cris.texi
6115 @node Machine Dependencies
6116 @chapter Machine Dependent Features
6118 The machine instruction sets are different on each Renesas chip family,
6119 and there are also some syntax differences among the families. This
6120 chapter describes the specific @command{@value{AS}} features for each
6124 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6125 * H8/500-Dependent:: Renesas H8/500 Dependent Features
6126 * SH-Dependent:: Renesas SH Dependent Features
6133 @include c-d10v.texi
6137 @include c-d30v.texi
6141 @include c-h8300.texi
6145 @include c-h8500.texi
6149 @include c-hppa.texi
6153 @include c-i370.texi
6157 @include c-i386.texi
6161 @include c-i860.texi
6165 @include c-i960.texi
6169 @include c-ia64.texi
6173 @include c-ip2k.texi
6177 @include c-m32r.texi
6181 @include c-m68k.texi
6185 @include c-m68hc11.texi
6189 @include c-m88k.texi
6193 @include c-mips.texi
6197 @include c-mmix.texi
6201 @include c-msp430.texi
6205 @include c-ns32k.texi
6209 @include c-pdp11.texi
6222 @include c-sh64.texi
6226 @include c-sparc.texi
6230 @include c-tic54x.texi
6242 @include c-v850.texi
6246 @include c-xtensa.texi
6250 @c reverse effect of @down at top of generic Machine-Dep chapter
6254 @node Reporting Bugs
6255 @chapter Reporting Bugs
6256 @cindex bugs in assembler
6257 @cindex reporting bugs in assembler
6259 Your bug reports play an essential role in making @command{@value{AS}} reliable.
6261 Reporting a bug may help you by bringing a solution to your problem, or it may
6262 not. But in any case the principal function of a bug report is to help the
6263 entire community by making the next version of @command{@value{AS}} work better.
6264 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
6266 In order for a bug report to serve its purpose, you must include the
6267 information that enables us to fix the bug.
6270 * Bug Criteria:: Have you found a bug?
6271 * Bug Reporting:: How to report bugs
6275 @section Have You Found a Bug?
6276 @cindex bug criteria
6278 If you are not sure whether you have found a bug, here are some guidelines:
6281 @cindex fatal signal
6282 @cindex assembler crash
6283 @cindex crash of assembler
6285 If the assembler gets a fatal signal, for any input whatever, that is a
6286 @command{@value{AS}} bug. Reliable assemblers never crash.
6288 @cindex error on valid input
6290 If @command{@value{AS}} produces an error message for valid input, that is a bug.
6292 @cindex invalid input
6294 If @command{@value{AS}} does not produce an error message for invalid input, that
6295 is a bug. However, you should note that your idea of ``invalid input'' might
6296 be our idea of ``an extension'' or ``support for traditional practice''.
6299 If you are an experienced user of assemblers, your suggestions for improvement
6300 of @command{@value{AS}} are welcome in any case.
6304 @section How to Report Bugs
6306 @cindex assembler bugs, reporting
6308 A number of companies and individuals offer support for @sc{gnu} products. If
6309 you obtained @command{@value{AS}} from a support organization, we recommend you
6310 contact that organization first.
6312 You can find contact information for many support companies and
6313 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
6316 In any event, we also recommend that you send bug reports for @command{@value{AS}}
6317 to @samp{bug-binutils@@gnu.org}.
6319 The fundamental principle of reporting bugs usefully is this:
6320 @strong{report all the facts}. If you are not sure whether to state a
6321 fact or leave it out, state it!
6323 Often people omit facts because they think they know what causes the problem
6324 and assume that some details do not matter. Thus, you might assume that the
6325 name of a symbol you use in an example does not matter. Well, probably it does
6326 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
6327 happens to fetch from the location where that name is stored in memory;
6328 perhaps, if the name were different, the contents of that location would fool
6329 the assembler into doing the right thing despite the bug. Play it safe and
6330 give a specific, complete example. That is the easiest thing for you to do,
6331 and the most helpful.
6333 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
6334 it is new to us. Therefore, always write your bug reports on the assumption
6335 that the bug has not been reported previously.
6337 Sometimes people give a few sketchy facts and ask, ``Does this ring a
6338 bell?'' This cannot help us fix a bug, so it is basically useless. We
6339 respond by asking for enough details to enable us to investigate.
6340 You might as well expedite matters by sending them to begin with.
6342 To enable us to fix the bug, you should include all these things:
6346 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
6347 it with the @samp{--version} argument.
6349 Without this, we will not know whether there is any point in looking for
6350 the bug in the current version of @command{@value{AS}}.
6353 Any patches you may have applied to the @command{@value{AS}} source.
6356 The type of machine you are using, and the operating system name and
6360 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
6364 The command arguments you gave the assembler to assemble your example and
6365 observe the bug. To guarantee you will not omit something important, list them
6366 all. A copy of the Makefile (or the output from make) is sufficient.
6368 If we were to try to guess the arguments, we would probably guess wrong
6369 and then we might not encounter the bug.
6372 A complete input file that will reproduce the bug. If the bug is observed when
6373 the assembler is invoked via a compiler, send the assembler source, not the
6374 high level language source. Most compilers will produce the assembler source
6375 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
6376 the options @samp{-v --save-temps}; this will save the assembler source in a
6377 file with an extension of @file{.s}, and also show you exactly how
6378 @command{@value{AS}} is being run.
6381 A description of what behavior you observe that you believe is
6382 incorrect. For example, ``It gets a fatal signal.''
6384 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
6385 will certainly notice it. But if the bug is incorrect output, we might not
6386 notice unless it is glaringly wrong. You might as well not give us a chance to
6389 Even if the problem you experience is a fatal signal, you should still say so
6390 explicitly. Suppose something strange is going on, such as, your copy of
6391 @command{@value{AS}} is out of synch, or you have encountered a bug in the C
6392 library on your system. (This has happened!) Your copy might crash and ours
6393 would not. If you told us to expect a crash, then when ours fails to crash, we
6394 would know that the bug was not happening for us. If you had not told us to
6395 expect a crash, then we would not be able to draw any conclusion from our
6399 If you wish to suggest changes to the @command{@value{AS}} source, send us context
6400 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
6401 option. Always send diffs from the old file to the new file. If you even
6402 discuss something in the @command{@value{AS}} source, refer to it by context, not
6405 The line numbers in our development sources will not match those in your
6406 sources. Your line numbers would convey no useful information to us.
6409 Here are some things that are not necessary:
6413 A description of the envelope of the bug.
6415 Often people who encounter a bug spend a lot of time investigating
6416 which changes to the input file will make the bug go away and which
6417 changes will not affect it.
6419 This is often time consuming and not very useful, because the way we
6420 will find the bug is by running a single example under the debugger
6421 with breakpoints, not by pure deduction from a series of examples.
6422 We recommend that you save your time for something else.
6424 Of course, if you can find a simpler example to report @emph{instead}
6425 of the original one, that is a convenience for us. Errors in the
6426 output will be easier to spot, running under the debugger will take
6427 less time, and so on.
6429 However, simplification is not vital; if you do not want to do this,
6430 report the bug anyway and send us the entire test case you used.
6433 A patch for the bug.
6435 A patch for the bug does help us if it is a good one. But do not omit
6436 the necessary information, such as the test case, on the assumption that
6437 a patch is all we need. We might see problems with your patch and decide
6438 to fix the problem another way, or we might not understand it at all.
6440 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
6441 construct an example that will make the program follow a certain path through
6442 the code. If you do not send us the example, we will not be able to construct
6443 one, so we will not be able to verify that the bug is fixed.
6445 And if we cannot understand what bug you are trying to fix, or why your
6446 patch should be an improvement, we will not install it. A test case will
6447 help us to understand.
6450 A guess about what the bug is or what it depends on.
6452 Such guesses are usually wrong. Even we cannot guess right about such
6453 things without first using the debugger to find the facts.
6456 @node Acknowledgements
6457 @chapter Acknowledgements
6459 If you have contributed to @command{@value{AS}} and your name isn't listed here,
6460 it is not meant as a slight. We just don't know about it. Send mail to the
6461 maintainer, and we'll correct the situation. Currently
6463 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
6465 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
6468 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
6469 information and the 68k series machines, most of the preprocessing pass, and
6470 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
6472 K. Richard Pixley maintained GAS for a while, adding various enhancements and
6473 many bug fixes, including merging support for several processors, breaking GAS
6474 up to handle multiple object file format back ends (including heavy rewrite,
6475 testing, an integration of the coff and b.out back ends), adding configuration
6476 including heavy testing and verification of cross assemblers and file splits
6477 and renaming, converted GAS to strictly ANSI C including full prototypes, added
6478 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
6479 port (including considerable amounts of reverse engineering), a SPARC opcode
6480 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
6481 assertions and made them work, much other reorganization, cleanup, and lint.
6483 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
6484 in format-specific I/O modules.
6486 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
6487 has done much work with it since.
6489 The Intel 80386 machine description was written by Eliot Dresselhaus.
6491 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
6493 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
6494 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
6496 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
6497 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
6498 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
6499 support a.out format.
6501 Support for the Zilog Z8k and Renesas H8/300 and H8/500 processors (tc-z8k,
6502 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
6503 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
6504 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
6507 John Gilmore built the AMD 29000 support, added @code{.include} support, and
6508 simplified the configuration of which versions accept which directives. He
6509 updated the 68k machine description so that Motorola's opcodes always produced
6510 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
6511 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
6512 cross-compilation support, and one bug in relaxation that took a week and
6513 required the proverbial one-bit fix.
6515 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
6516 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
6517 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
6518 PowerPC assembler, and made a few other minor patches.
6520 Steve Chamberlain made @command{@value{AS}} able to generate listings.
6522 Hewlett-Packard contributed support for the HP9000/300.
6524 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
6525 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
6526 formats). This work was supported by both the Center for Software Science at
6527 the University of Utah and Cygnus Support.
6529 Support for ELF format files has been worked on by Mark Eichin of Cygnus
6530 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
6531 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
6532 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
6533 and some initial 64-bit support).
6535 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
6537 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
6538 support for openVMS/Alpha.
6540 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
6543 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
6544 Inc. added support for Xtensa processors.
6546 Several engineers at Cygnus Support have also provided many small bug fixes and
6547 configuration enhancements.
6549 Many others have contributed large or small bugfixes and enhancements. If
6550 you have contributed significant work and are not mentioned on this list, and
6551 want to be, let us know. Some of the history has been lost; we are not
6552 intentionally leaving anyone out.
projects / deliverable / binutils-gdb.git / blob