1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
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:
22 @include asconfig.texi
27 @c Configure for the generation of man pages
65 @c common OR combinations of conditions
91 @set abnormal-separator
95 @settitle Using @value{AS}
98 @settitle Using @value{AS} (@value{TARGET})
100 @setchapternewpage odd
105 @c WARE! Some of the machine-dependent sections contain tables of machine
106 @c instructions. Except in multi-column format, these tables look silly.
107 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
108 @c the multi-col format is faked within @example sections.
110 @c Again unfortunately, the natural size that fits on a page, for these tables,
111 @c is different depending on whether or not smallbook is turned on.
112 @c This matters, because of order: text flow switches columns at each page
115 @c The format faked in this source works reasonably well for smallbook,
116 @c not well for the default large-page format. This manual expects that if you
117 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
118 @c tables in question. You can turn on one without the other at your
119 @c discretion, of course.
122 @c the insn tables look just as silly in info files regardless of smallbook,
123 @c might as well show 'em anyways.
129 * As: (as). The GNU assembler.
130 * Gas: (as). The GNU assembler.
139 This file documents the GNU Assembler "@value{AS}".
141 @c man begin COPYRIGHT
142 Copyright (C) 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002 Free Software Foundation, Inc.
144 Permission is granted to copy, distribute and/or modify this document
145 under the terms of the GNU Free Documentation License, Version 1.1
146 or any later version published by the Free Software Foundation;
147 with no Invariant Sections, with no Front-Cover Texts, and with no
148 Back-Cover Texts. A copy of the license is included in the
149 section entitled ``GNU Free Documentation License''.
154 Permission is granted to process this file through Tex and print the
155 results, provided the printed document carries copying permission
156 notice identical to this one except for the removal of this paragraph
157 (this paragraph not being relevant to the printed manual).
163 @title Using @value{AS}
164 @subtitle The @sc{gnu} Assembler
166 @subtitle for the @value{TARGET} family
169 @subtitle Version @value{VERSION}
172 The Free Software Foundation Inc. thanks The Nice Computer
173 Company of Australia for loaning Dean Elsner to write the
174 first (Vax) version of @command{as} for Project @sc{gnu}.
175 The proprietors, management and staff of TNCCA thank FSF for
176 distracting the boss while they got some work
179 @author Dean Elsner, Jay Fenlason & friends
183 \hfill {\it Using {\tt @value{AS}}}\par
184 \hfill Edited by Cygnus Support\par
186 %"boxit" macro for figures:
187 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
188 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
189 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
190 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
191 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
194 @vskip 0pt plus 1filll
195 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002 Free Software Foundation, Inc.
197 Permission is granted to copy, distribute and/or modify this document
198 under the terms of the GNU Free Documentation License, Version 1.1
199 or any later version published by the Free Software Foundation;
200 with no Invariant Sections, with no Front-Cover Texts, and with no
201 Back-Cover Texts. A copy of the license is included in the
202 section entitled ``GNU Free Documentation License''.
208 @top Using @value{AS}
210 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}} version
213 This version of the file describes @command{@value{AS}} configured to generate
214 code for @value{TARGET} architectures.
217 This document is distributed under the terms of the GNU Free
218 Documentation License. A copy of the license is included in the
219 section entitled ``GNU Free Documentation License''.
222 * Overview:: Overview
223 * Invoking:: Command-Line Options
225 * Sections:: Sections and Relocation
227 * Expressions:: Expressions
228 * Pseudo Ops:: Assembler Directives
229 * Machine Dependencies:: Machine Dependent Features
230 * Reporting Bugs:: Reporting Bugs
231 * Acknowledgements:: Who Did What
232 * GNU Free Documentation License:: GNU Free Documentation License
240 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
242 This version of the manual describes @command{@value{AS}} configured to generate
243 code for @value{TARGET} architectures.
247 @cindex invocation summary
248 @cindex option summary
249 @cindex summary of options
250 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
251 @pxref{Invoking,,Command-Line Options}.
253 @c man title AS the portable GNU assembler.
257 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
261 @c We don't use deffn and friends for the following because they seem
262 @c to be limited to one line for the header.
264 @c man begin SYNOPSIS
265 @value{AS} [@b{-a}[@b{cdhlns}][=@var{file}]] [@b{-D}] [@b{--defsym} @var{sym}=@var{val}]
266 [@b{-f}] [@b{--gstabs}] [@b{--gstabs+}] [@b{--gdwarf2}] [@b{--help}]
267 [@b{-I} @var{dir}] [@b{-J}] [@b{-K}] [@b{-L}]
268 [@b{--listing-lhs-width}=@var{NUM}] [@b{--listing-lhs-width2}=@var{NUM}]
269 [@b{--listing-rhs-width}=@var{NUM}] [@b{--listing-cont-lines}=@var{NUM}]
270 [@b{--keep-locals}] [@b{-o} @var{objfile}] [@b{-R}] [@b{--statistics}] [@b{-v}]
271 [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}] [@b{--fatal-warnings}]
272 [@b{-w}] [@b{-x}] [@b{-Z}] [@b{--target-help}] [@var{target-options}]
273 [@b{--}|@var{files} @dots{}]
275 @c Target dependent options are listed below. Keep the list sorted.
276 @c Add an empty line for separation.
278 @c am29k has no machine-dependent assembler options
282 @emph{Target Alpha options:}
284 [@b{-mdebug} | @b{-no-mdebug}]
285 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
286 [@b{-F}] [@b{-32addr}]
290 @emph{Target ARC options:}
296 @emph{Target ARM options:}
297 @c Don't document the deprecated options
298 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
299 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
300 [@b{-mfpu}=@var{floating-point-format}]
301 [@b{-mfloat-abi}=@var{abi}]
304 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
305 @b{-mapcs-reentrant}]
306 [@b{-mthumb-interwork}] [@b{-moabi}] [@b{-k}]
310 @emph{Target CRIS options:}
311 [@b{--underscore} | @b{--no-underscore}]
313 [@b{--emulation=criself} | @b{--emulation=crisaout}]
314 @c Deprecated -- deliberately not documented.
319 @emph{Target D10V options:}
324 @emph{Target D30V options:}
325 [@b{-O}|@b{-n}|@b{-N}]
328 @c Renesas family chips have no machine-dependent assembler options
331 @c HPPA has no machine-dependent assembler options (yet).
335 @emph{Target i386 options:}
336 [@b{--32}|@b{--64}] [@b{-n}]
340 @emph{Target i960 options:}
341 @c see md_parse_option in tc-i960.c
342 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
344 [@b{-b}] [@b{-no-relax}]
348 @emph{Target IA-64 options:}
349 [@b{-mconstant-gp}|@b{-mauto-pic}]
350 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
352 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
356 @emph{Target IP2K options:}
357 [@b{-mip2022}|@b{-mip2022ext}]
361 @emph{Target M32R options:}
362 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
367 @emph{Target M680X0 options:}
368 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
372 @emph{Target M68HC11 options:}
373 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
374 [@b{-mshort}|@b{-mlong}]
375 [@b{-mshort-double}|@b{-mlong-double}]
376 [@b{--force-long-branchs}] [@b{--short-branchs}]
377 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
378 [@b{--print-opcodes}] [@b{--generate-example}]
382 @emph{Target MCORE options:}
383 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
384 [@b{-mcpu=[210|340]}]
388 @emph{Target MIPS options:}
389 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-n}] [@b{-O}[@var{optimization level}]]
390 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
391 [@b{-non_shared}] [@b{-xgot}] [@b{--membedded-pic}]
392 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
393 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
394 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
395 [@b{-mips64}] [@b{-mips64r2}]
396 [@b{-construct-floats}] [@b{-no-construct-floats}]
397 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
398 [@b{-mfix7000}] [@b{-mno-fix7000}]
399 [@b{-mips16}] [@b{-no-mips16}]
400 [@b{-mips3d}] [@b{-no-mips3d}]
401 [@b{-mdmx}] [@b{-no-mdmx}]
402 [@b{-mdebug}] [@b{-no-mdebug}]
403 [@b{-mpdr}] [@b{-mno-pdr}]
407 @emph{Target MMIX options:}
408 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
409 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
410 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
411 [@b{--linker-allocated-gregs}]
415 @emph{Target PDP11 options:}
416 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
417 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
418 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
422 @emph{Target picoJava options:}
427 @emph{Target PowerPC options:}
428 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
429 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}|
430 @b{-mbooke32}|@b{-mbooke64}]
431 [@b{-mcom}|@b{-many}|@b{-maltivec}] [@b{-memb}]
432 [@b{-mregnames}|@b{-mno-regnames}]
433 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
434 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
435 [@b{-msolaris}|@b{-mno-solaris}]
439 @emph{Target SPARC options:}
440 @c The order here is important. See c-sparc.texi.
441 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
442 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
443 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
448 @emph{Target TIC54X options:}
449 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
450 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
453 @c Z8000 has no machine-dependent assembler options
457 @emph{Target Xtensa options:}
458 [@b{--[no-]density}] [@b{--[no-]relax}] [@b{--[no-]generics}]
459 [@b{--[no-]text-section-literals}]
460 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
469 Turn on listings, in any of a variety of ways:
473 omit false conditionals
476 omit debugging directives
479 include high-level source
485 include macro expansions
488 omit forms processing
494 set the name of the listing file
497 You may combine these options; for example, use @samp{-aln} for assembly
498 listing without forms processing. The @samp{=file} option, if used, must be
499 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
502 Ignored. This option is accepted for script compatibility with calls to
505 @item --defsym @var{sym}=@var{value}
506 Define the symbol @var{sym} to be @var{value} before assembling the input file.
507 @var{value} must be an integer constant. As in C, a leading @samp{0x}
508 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
511 ``fast''---skip whitespace and comment preprocessing (assume source is
515 Generate stabs debugging information for each assembler line. This
516 may help debugging assembler code, if the debugger can handle it.
519 Generate stabs debugging information for each assembler line, with GNU
520 extensions that probably only gdb can handle, and that could make other
521 debuggers crash or refuse to read your program. This
522 may help debugging assembler code. Currently the only GNU extension is
523 the location of the current working directory at assembling time.
526 Generate DWARF2 debugging information for each assembler line. This
527 may help debugging assembler code, if the debugger can handle it. Note---this
528 option is only supported by some targets, not all of them.
531 Print a summary of the command line options and exit.
534 Print a summary of all target specific options and exit.
537 Add directory @var{dir} to the search list for @code{.include} directives.
540 Don't warn about signed overflow.
543 @ifclear DIFF-TBL-KLUGE
544 This option is accepted but has no effect on the @value{TARGET} family.
546 @ifset DIFF-TBL-KLUGE
547 Issue warnings when difference tables altered for long displacements.
552 Keep (in the symbol table) local symbols. On traditional a.out systems
553 these start with @samp{L}, but different systems have different local
556 @item --listing-lhs-width=@var{number}
557 Set the maximum width, in words, of the output data column for an assembler
558 listing to @var{number}.
560 @item --listing-lhs-width2=@var{number}
561 Set the maximum width, in words, of the output data column for continuation
562 lines in an assembler listing to @var{number}.
564 @item --listing-rhs-width=@var{number}
565 Set the maximum width of an input source line, as displayed in a listing, to
568 @item --listing-cont-lines=@var{number}
569 Set the maximum number of lines printed in a listing for a single line of input
572 @item -o @var{objfile}
573 Name the object-file output from @command{@value{AS}} @var{objfile}.
576 Fold the data section into the text section.
579 Print the maximum space (in bytes) and total time (in seconds) used by
582 @item --strip-local-absolute
583 Remove local absolute symbols from the outgoing symbol table.
587 Print the @command{as} version.
590 Print the @command{as} version and exit.
594 Suppress warning messages.
596 @item --fatal-warnings
597 Treat warnings as errors.
600 Don't suppress warning messages or treat them as errors.
609 Generate an object file even after errors.
611 @item -- | @var{files} @dots{}
612 Standard input, or source files to assemble.
617 The following options are available when @value{AS} is configured for
622 This option selects the core processor variant.
624 Select either big-endian (-EB) or little-endian (-EL) output.
629 The following options are available when @value{AS} is configured for the ARM
633 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
634 Specify which ARM processor variant is the target.
635 @item -march=@var{architecture}[+@var{extension}@dots{}]
636 Specify which ARM architecture variant is used by the target.
637 @item -mfpu=@var{floating-point-format}
638 Select which Floating Point architecture is the target.
639 @item -mfloat-abi=@var{abi}
640 Select which floating point ABI is in use.
642 Enable Thumb only instruction decoding.
643 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant | -moabi
644 Select which procedure calling convention is in use.
646 Select either big-endian (-EB) or little-endian (-EL) output.
647 @item -mthumb-interwork
648 Specify that the code has been generated with interworking between Thumb and
651 Specify that PIC code has been generated.
656 See the info pages for documentation of the CRIS-specific options.
660 The following options are available when @value{AS} is configured for
663 @cindex D10V optimization
664 @cindex optimization, D10V
666 Optimize output by parallelizing instructions.
671 The following options are available when @value{AS} is configured for a D30V
674 @cindex D30V optimization
675 @cindex optimization, D30V
677 Optimize output by parallelizing instructions.
681 Warn when nops are generated.
683 @cindex D30V nops after 32-bit multiply
685 Warn when a nop after a 32-bit multiply instruction is generated.
690 The following options are available when @value{AS} is configured for the
691 Intel 80960 processor.
694 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
695 Specify which variant of the 960 architecture is the target.
698 Add code to collect statistics about branches taken.
701 Do not alter compare-and-branch instructions for long displacements;
708 The following options are available when @value{AS} is configured for the
714 Specifies that the extended IP2022 instructions are allowed.
717 Restores the default behaviour, which restricts the permitted instructions to
718 just the basic IP2022 ones.
724 The following options are available when @value{AS} is configured for the
725 Renesas M32R (formerly Mitsubishi M32R) series.
730 Specify which processor in the M32R family is the target. The default
731 is normally the M32R, but this option changes it to the M32RX.
733 @item --warn-explicit-parallel-conflicts or --Wp
734 Produce warning messages when questionable parallel constructs are
737 @item --no-warn-explicit-parallel-conflicts or --Wnp
738 Do not produce warning messages when questionable parallel constructs are
745 The following options are available when @value{AS} is configured for the
746 Motorola 68000 series.
751 Shorten references to undefined symbols, to one word instead of two.
753 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
754 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
755 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
756 Specify what processor in the 68000 family is the target. The default
757 is normally the 68020, but this can be changed at configuration time.
759 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
760 The target machine does (or does not) have a floating-point coprocessor.
761 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
762 the basic 68000 is not compatible with the 68881, a combination of the
763 two can be specified, since it's possible to do emulation of the
764 coprocessor instructions with the main processor.
766 @item -m68851 | -mno-68851
767 The target machine does (or does not) have a memory-management
768 unit coprocessor. The default is to assume an MMU for 68020 and up.
775 For details about the PDP-11 machine dependent features options,
776 see @ref{PDP-11-Options}.
779 @item -mpic | -mno-pic
780 Generate position-independent (or position-dependent) code. The
781 default is @option{-mpic}.
784 @itemx -mall-extensions
785 Enable all instruction set extensions. This is the default.
787 @item -mno-extensions
788 Disable all instruction set extensions.
790 @item -m@var{extension} | -mno-@var{extension}
791 Enable (or disable) a particular instruction set extension.
794 Enable the instruction set extensions supported by a particular CPU, and
795 disable all other extensions.
797 @item -m@var{machine}
798 Enable the instruction set extensions supported by a particular machine
799 model, and disable all other extensions.
805 The following options are available when @value{AS} is configured for
806 a picoJava processor.
810 @cindex PJ endianness
811 @cindex endianness, PJ
812 @cindex big endian output, PJ
814 Generate ``big endian'' format output.
816 @cindex little endian output, PJ
818 Generate ``little endian'' format output.
824 The following options are available when @value{AS} is configured for the
825 Motorola 68HC11 or 68HC12 series.
829 @item -m68hc11 | -m68hc12 | -m68hcs12
830 Specify what processor is the target. The default is
831 defined by the configuration option when building the assembler.
834 Specify to use the 16-bit integer ABI.
837 Specify to use the 32-bit integer ABI.
840 Specify to use the 32-bit double ABI.
843 Specify to use the 64-bit double ABI.
845 @item --force-long-branchs
846 Relative branches are turned into absolute ones. This concerns
847 conditional branches, unconditional branches and branches to a
850 @item -S | --short-branchs
851 Do not turn relative branchs into absolute ones
852 when the offset is out of range.
854 @item --strict-direct-mode
855 Do not turn the direct addressing mode into extended addressing mode
856 when the instruction does not support direct addressing mode.
858 @item --print-insn-syntax
859 Print the syntax of instruction in case of error.
861 @item --print-opcodes
862 print the list of instructions with syntax and then exit.
864 @item --generate-example
865 print an example of instruction for each possible instruction and then exit.
866 This option is only useful for testing @command{@value{AS}}.
872 The following options are available when @command{@value{AS}} is configured
873 for the SPARC architecture:
876 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
877 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
878 Explicitly select a variant of the SPARC architecture.
880 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
881 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
883 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
884 UltraSPARC extensions.
886 @item -xarch=v8plus | -xarch=v8plusa
887 For compatibility with the Solaris v9 assembler. These options are
888 equivalent to -Av8plus and -Av8plusa, respectively.
891 Warn when the assembler switches to another architecture.
896 The following options are available when @value{AS} is configured for the 'c54x
901 Enable extended addressing mode. All addresses and relocations will assume
902 extended addressing (usually 23 bits).
903 @item -mcpu=@var{CPU_VERSION}
904 Sets the CPU version being compiled for.
905 @item -merrors-to-file @var{FILENAME}
906 Redirect error output to a file, for broken systems which don't support such
907 behaviour in the shell.
912 The following options are available when @value{AS} is configured for
913 a @sc{mips} processor.
917 This option sets the largest size of an object that can be referenced
918 implicitly with the @code{gp} register. It is only accepted for targets that
919 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
921 @cindex MIPS endianness
922 @cindex endianness, MIPS
923 @cindex big endian output, MIPS
925 Generate ``big endian'' format output.
927 @cindex little endian output, MIPS
929 Generate ``little endian'' format output.
941 Generate code for a particular @sc{mips} Instruction Set Architecture level.
942 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
943 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
944 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
945 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
947 correspond to generic
948 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
949 and @samp{MIPS64 Release 2}
950 ISA processors, respectively.
952 @item -march=@var{CPU}
953 Generate code for a particular @sc{mips} cpu.
955 @item -mtune=@var{cpu}
956 Schedule and tune for a particular @sc{mips} cpu.
960 Cause nops to be inserted if the read of the destination register
961 of an mfhi or mflo instruction occurs in the following two instructions.
965 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
966 section instead of the standard ELF .stabs sections.
970 Control generation of @code{.pdr} sections.
974 The register sizes are normally inferred from the ISA and ABI, but these
975 flags force a certain group of registers to be treated as 32 bits wide at
976 all times. @samp{-mgp32} controls the size of general-purpose registers
977 and @samp{-mfp32} controls the size of floating-point registers.
981 Generate code for the MIPS 16 processor. This is equivalent to putting
982 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
983 turns off this option.
987 Generate code for the MIPS-3D Application Specific Extension.
988 This tells the assembler to accept MIPS-3D instructions.
989 @samp{-no-mips3d} turns off this option.
993 Generate code for the MDMX Application Specific Extension.
994 This tells the assembler to accept MDMX instructions.
995 @samp{-no-mdmx} turns off this option.
997 @item --construct-floats
998 @itemx --no-construct-floats
999 The @samp{--no-construct-floats} option disables the construction of
1000 double width floating point constants by loading the two halves of the
1001 value into the two single width floating point registers that make up
1002 the double width register. By default @samp{--construct-floats} is
1003 selected, allowing construction of these floating point constants.
1006 @item --emulation=@var{name}
1007 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1008 for some other target, in all respects, including output format (choosing
1009 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1010 debugging information or store symbol table information, and default
1011 endianness. The available configuration names are: @samp{mipsecoff},
1012 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1013 @samp{mipsbelf}. The first two do not alter the default endianness from that
1014 of the primary target for which the assembler was configured; the others change
1015 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1016 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1017 selection in any case.
1019 This option is currently supported only when the primary target
1020 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1021 Furthermore, the primary target or others specified with
1022 @samp{--enable-targets=@dots{}} at configuration time must include support for
1023 the other format, if both are to be available. For example, the Irix 5
1024 configuration includes support for both.
1026 Eventually, this option will support more configurations, with more
1027 fine-grained control over the assembler's behavior, and will be supported for
1031 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1038 Control how to deal with multiplication overflow and division by zero.
1039 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1040 (and only work for Instruction Set Architecture level 2 and higher);
1041 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1045 When this option is used, @command{@value{AS}} will issue a warning every
1046 time it generates a nop instruction from a macro.
1051 The following options are available when @value{AS} is configured for
1057 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1058 The command line option @samp{-nojsri2bsr} can be used to disable it.
1062 Enable or disable the silicon filter behaviour. By default this is disabled.
1063 The default can be overridden by the @samp{-sifilter} command line option.
1066 Alter jump instructions for long displacements.
1068 @item -mcpu=[210|340]
1069 Select the cpu type on the target hardware. This controls which instructions
1073 Assemble for a big endian target.
1076 Assemble for a little endian target.
1082 See the info pages for documentation of the MMIX-specific options.
1086 The following options are available when @value{AS} is configured for
1087 an Xtensa processor.
1090 @item --density | --no-density
1091 Enable or disable use of instructions from the Xtensa code density
1092 option. This is enabled by default when the Xtensa processor supports
1093 the code density option.
1095 @item --relax | --no-relax
1096 Enable or disable instruction relaxation. This is enabled by default.
1097 Note: In the current implementation, these options also control whether
1098 assembler optimizations are performed, making these options equivalent
1099 to @option{--generics} and @option{--no-generics}.
1101 @item --generics | --no-generics
1102 Enable or disable all assembler transformations of Xtensa instructions.
1103 The default is @option{--generics};
1104 @option{--no-generics} should be used only in the rare cases when the
1105 instructions must be exactly as specified in the assembly source.
1107 @item --text-section-literals | --no-text-section-literals
1108 With @option{--text-@-section-@-literals}, literal pools are interspersed
1109 in the text section. The default is
1110 @option{--no-@-text-@-section-@-literals}, which places literals in a
1111 separate section in the output file.
1113 @item --target-align | --no-target-align
1114 Enable or disable automatic alignment to reduce branch penalties at the
1115 expense of some code density. The default is @option{--target-@-align}.
1117 @item --longcalls | --no-longcalls
1118 Enable or disable transformation of call instructions to allow calls
1119 across a greater range of addresses. The default is
1120 @option{--no-@-longcalls}.
1127 * Manual:: Structure of this Manual
1128 * GNU Assembler:: The GNU Assembler
1129 * Object Formats:: Object File Formats
1130 * Command Line:: Command Line
1131 * Input Files:: Input Files
1132 * Object:: Output (Object) File
1133 * Errors:: Error and Warning Messages
1137 @section Structure of this Manual
1139 @cindex manual, structure and purpose
1140 This manual is intended to describe what you need to know to use
1141 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1142 notation for symbols, constants, and expressions; the directives that
1143 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1146 We also cover special features in the @value{TARGET}
1147 configuration of @command{@value{AS}}, including assembler directives.
1150 This manual also describes some of the machine-dependent features of
1151 various flavors of the assembler.
1154 @cindex machine instructions (not covered)
1155 On the other hand, this manual is @emph{not} intended as an introduction
1156 to programming in assembly language---let alone programming in general!
1157 In a similar vein, we make no attempt to introduce the machine
1158 architecture; we do @emph{not} describe the instruction set, standard
1159 mnemonics, registers or addressing modes that are standard to a
1160 particular architecture.
1162 You may want to consult the manufacturer's
1163 machine architecture manual for this information.
1167 For information on the H8/300 machine instruction set, see @cite{H8/300
1168 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1169 Programming Manual} (Renesas).
1172 For information on the H8/500 machine instruction set, see @cite{H8/500
1173 Series Programming Manual} (Renesas M21T001).
1176 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1177 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1178 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1179 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1182 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1186 @c I think this is premature---doc@cygnus.com, 17jan1991
1188 Throughout this manual, we assume that you are running @dfn{GNU},
1189 the portable operating system from the @dfn{Free Software
1190 Foundation, Inc.}. This restricts our attention to certain kinds of
1191 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1192 once this assumption is granted examples and definitions need less
1195 @command{@value{AS}} is part of a team of programs that turn a high-level
1196 human-readable series of instructions into a low-level
1197 computer-readable series of instructions. Different versions of
1198 @command{@value{AS}} are used for different kinds of computer.
1201 @c There used to be a section "Terminology" here, which defined
1202 @c "contents", "byte", "word", and "long". Defining "word" to any
1203 @c particular size is confusing when the .word directive may generate 16
1204 @c bits on one machine and 32 bits on another; in general, for the user
1205 @c version of this manual, none of these terms seem essential to define.
1206 @c They were used very little even in the former draft of the manual;
1207 @c this draft makes an effort to avoid them (except in names of
1211 @section The GNU Assembler
1213 @c man begin DESCRIPTION
1215 @sc{gnu} @command{as} is really a family of assemblers.
1217 This manual describes @command{@value{AS}}, a member of that family which is
1218 configured for the @value{TARGET} architectures.
1220 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1221 should find a fairly similar environment when you use it on another
1222 architecture. Each version has much in common with the others,
1223 including object file formats, most assembler directives (often called
1224 @dfn{pseudo-ops}) and assembler syntax.@refill
1226 @cindex purpose of @sc{gnu} assembler
1227 @command{@value{AS}} is primarily intended to assemble the output of the
1228 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1229 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1230 assemble correctly everything that other assemblers for the same
1231 machine would assemble.
1233 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1236 @c This remark should appear in generic version of manual; assumption
1237 @c here is that generic version sets M680x0.
1238 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1239 assembler for the same architecture; for example, we know of several
1240 incompatible versions of 680x0 assembly language syntax.
1245 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1246 program in one pass of the source file. This has a subtle impact on the
1247 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1249 @node Object Formats
1250 @section Object File Formats
1252 @cindex object file format
1253 The @sc{gnu} assembler can be configured to produce several alternative
1254 object file formats. For the most part, this does not affect how you
1255 write assembly language programs; but directives for debugging symbols
1256 are typically different in different file formats. @xref{Symbol
1257 Attributes,,Symbol Attributes}.
1260 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1261 @value{OBJ-NAME} format object files.
1263 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1265 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1266 @code{a.out} or COFF format object files.
1269 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1270 @code{b.out} or COFF format object files.
1273 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1274 SOM or ELF format object files.
1279 @section Command Line
1281 @cindex command line conventions
1283 After the program name @command{@value{AS}}, the command line may contain
1284 options and file names. Options may appear in any order, and may be
1285 before, after, or between file names. The order of file names is
1288 @cindex standard input, as input file
1290 @file{--} (two hyphens) by itself names the standard input file
1291 explicitly, as one of the files for @command{@value{AS}} to assemble.
1293 @cindex options, command line
1294 Except for @samp{--} any command line argument that begins with a
1295 hyphen (@samp{-}) is an option. Each option changes the behavior of
1296 @command{@value{AS}}. No option changes the way another option works. An
1297 option is a @samp{-} followed by one or more letters; the case of
1298 the letter is important. All options are optional.
1300 Some options expect exactly one file name to follow them. The file
1301 name may either immediately follow the option's letter (compatible
1302 with older assemblers) or it may be the next command argument (@sc{gnu}
1303 standard). These two command lines are equivalent:
1306 @value{AS} -o my-object-file.o mumble.s
1307 @value{AS} -omy-object-file.o mumble.s
1311 @section Input Files
1314 @cindex source program
1315 @cindex files, input
1316 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1317 describe the program input to one run of @command{@value{AS}}. The program may
1318 be in one or more files; how the source is partitioned into files
1319 doesn't change the meaning of the source.
1321 @c I added "con" prefix to "catenation" just to prove I can overcome my
1322 @c APL training... doc@cygnus.com
1323 The source program is a concatenation of the text in all the files, in the
1326 @c man begin DESCRIPTION
1327 Each time you run @command{@value{AS}} it assembles exactly one source
1328 program. The source program is made up of one or more files.
1329 (The standard input is also a file.)
1331 You give @command{@value{AS}} a command line that has zero or more input file
1332 names. The input files are read (from left file name to right). A
1333 command line argument (in any position) that has no special meaning
1334 is taken to be an input file name.
1336 If you give @command{@value{AS}} no file names it attempts to read one input file
1337 from the @command{@value{AS}} standard input, which is normally your terminal. You
1338 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1341 Use @samp{--} if you need to explicitly name the standard input file
1342 in your command line.
1344 If the source is empty, @command{@value{AS}} produces a small, empty object
1349 @subheading Filenames and Line-numbers
1351 @cindex input file linenumbers
1352 @cindex line numbers, in input files
1353 There are two ways of locating a line in the input file (or files) and
1354 either may be used in reporting error messages. One way refers to a line
1355 number in a physical file; the other refers to a line number in a
1356 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1358 @dfn{Physical files} are those files named in the command line given
1359 to @command{@value{AS}}.
1361 @dfn{Logical files} are simply names declared explicitly by assembler
1362 directives; they bear no relation to physical files. Logical file names help
1363 error messages reflect the original source file, when @command{@value{AS}} source
1364 is itself synthesized from other files. @command{@value{AS}} understands the
1365 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1366 @ref{File,,@code{.file}}.
1369 @section Output (Object) File
1375 Every time you run @command{@value{AS}} it produces an output file, which is
1376 your assembly language program translated into numbers. This file
1377 is the object file. Its default name is
1385 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1387 You can give it another name by using the @option{-o} option. Conventionally,
1388 object file names end with @file{.o}. The default name is used for historical
1389 reasons: older assemblers were capable of assembling self-contained programs
1390 directly into a runnable program. (For some formats, this isn't currently
1391 possible, but it can be done for the @code{a.out} format.)
1395 The object file is meant for input to the linker @code{@value{LD}}. It contains
1396 assembled program code, information to help @code{@value{LD}} integrate
1397 the assembled program into a runnable file, and (optionally) symbolic
1398 information for the debugger.
1400 @c link above to some info file(s) like the description of a.out.
1401 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1404 @section Error and Warning Messages
1406 @c man begin DESCRIPTION
1408 @cindex error messages
1409 @cindex warning messages
1410 @cindex messages from assembler
1411 @command{@value{AS}} may write warnings and error messages to the standard error
1412 file (usually your terminal). This should not happen when a compiler
1413 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1414 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1415 grave problem that stops the assembly.
1419 @cindex format of warning messages
1420 Warning messages have the format
1423 file_name:@b{NNN}:Warning Message Text
1427 @cindex line numbers, in warnings/errors
1428 (where @b{NNN} is a line number). If a logical file name has been given
1429 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1430 the current input file is used. If a logical line number was given
1432 (@pxref{Line,,@code{.line}})
1436 (@pxref{Line,,@code{.line}})
1439 (@pxref{Ln,,@code{.ln}})
1442 then it is used to calculate the number printed,
1443 otherwise the actual line in the current source file is printed. The
1444 message text is intended to be self explanatory (in the grand Unix
1447 @cindex format of error messages
1448 Error messages have the format
1450 file_name:@b{NNN}:FATAL:Error Message Text
1452 The file name and line number are derived as for warning
1453 messages. The actual message text may be rather less explanatory
1454 because many of them aren't supposed to happen.
1457 @chapter Command-Line Options
1459 @cindex options, all versions of assembler
1460 This chapter describes command-line options available in @emph{all}
1461 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
1463 to the @value{TARGET} target.
1466 to particular machine architectures.
1469 @c man begin DESCRIPTION
1471 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1472 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1473 The assembler arguments must be separated from each other (and the @samp{-Wa})
1474 by commas. For example:
1477 gcc -c -g -O -Wa,-alh,-L file.c
1481 This passes two options to the assembler: @samp{-alh} (emit a listing to
1482 standard output with high-level and assembly source) and @samp{-L} (retain
1483 local symbols in the symbol table).
1485 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1486 command-line options are automatically passed to the assembler by the compiler.
1487 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1488 precisely what options it passes to each compilation pass, including the
1494 * a:: -a[cdhlns] enable listings
1495 * D:: -D for compatibility
1496 * f:: -f to work faster
1497 * I:: -I for .include search path
1498 @ifclear DIFF-TBL-KLUGE
1499 * K:: -K for compatibility
1501 @ifset DIFF-TBL-KLUGE
1502 * K:: -K for difference tables
1505 * L:: -L to retain local labels
1506 * listing:: --listing-XXX to configure listing output
1507 * M:: -M or --mri to assemble in MRI compatibility mode
1508 * MD:: --MD for dependency tracking
1509 * o:: -o to name the object file
1510 * R:: -R to join data and text sections
1511 * statistics:: --statistics to see statistics about assembly
1512 * traditional-format:: --traditional-format for compatible output
1513 * v:: -v to announce version
1514 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1515 * Z:: -Z to make object file even after errors
1519 @section Enable Listings: @option{-a[cdhlns]}
1528 @cindex listings, enabling
1529 @cindex assembly listings, enabling
1531 These options enable listing output from the assembler. By itself,
1532 @samp{-a} requests high-level, assembly, and symbols listing.
1533 You can use other letters to select specific options for the list:
1534 @samp{-ah} requests a high-level language listing,
1535 @samp{-al} requests an output-program assembly listing, and
1536 @samp{-as} requests a symbol table listing.
1537 High-level listings require that a compiler debugging option like
1538 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1541 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1542 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1543 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1544 omitted from the listing.
1546 Use the @samp{-ad} option to omit debugging directives from the
1549 Once you have specified one of these options, you can further control
1550 listing output and its appearance using the directives @code{.list},
1551 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1553 The @samp{-an} option turns off all forms processing.
1554 If you do not request listing output with one of the @samp{-a} options, the
1555 listing-control directives have no effect.
1557 The letters after @samp{-a} may be combined into one option,
1558 @emph{e.g.}, @samp{-aln}.
1560 Note if the assembler source is coming from the standard input (eg because it
1561 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1562 is being used) then the listing will not contain any comments or preprocessor
1563 directives. This is because the listing code buffers input source lines from
1564 stdin only after they have been preprocessed by the assembler. This reduces
1565 memory usage and makes the code more efficient.
1568 @section @option{-D}
1571 This option has no effect whatsoever, but it is accepted to make it more
1572 likely that scripts written for other assemblers also work with
1573 @command{@value{AS}}.
1576 @section Work Faster: @option{-f}
1579 @cindex trusted compiler
1580 @cindex faster processing (@option{-f})
1581 @samp{-f} should only be used when assembling programs written by a
1582 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1583 and comment preprocessing on
1584 the input file(s) before assembling them. @xref{Preprocessing,
1588 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1589 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1594 @section @code{.include} Search Path: @option{-I} @var{path}
1596 @kindex -I @var{path}
1597 @cindex paths for @code{.include}
1598 @cindex search path for @code{.include}
1599 @cindex @code{include} directive search path
1600 Use this option to add a @var{path} to the list of directories
1601 @command{@value{AS}} searches for files specified in @code{.include}
1602 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1603 many times as necessary to include a variety of paths. The current
1604 working directory is always searched first; after that, @command{@value{AS}}
1605 searches any @samp{-I} directories in the same order as they were
1606 specified (left to right) on the command line.
1609 @section Difference Tables: @option{-K}
1612 @ifclear DIFF-TBL-KLUGE
1613 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1614 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1615 where it can be used to warn when the assembler alters the machine code
1616 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1617 family does not have the addressing limitations that sometimes lead to this
1618 alteration on other platforms.
1621 @ifset DIFF-TBL-KLUGE
1622 @cindex difference tables, warning
1623 @cindex warning for altered difference tables
1624 @command{@value{AS}} sometimes alters the code emitted for directives of the form
1625 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1626 You can use the @samp{-K} option if you want a warning issued when this
1631 @section Include Local Labels: @option{-L}
1634 @cindex local labels, retaining in output
1635 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1636 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1637 debugging, because they are intended for the use of programs (like
1638 compilers) that compose assembler programs, not for your notice.
1639 Normally both @command{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1640 normally debug with them.
1642 This option tells @command{@value{AS}} to retain those @samp{L@dots{}} symbols
1643 in the object file. Usually if you do this you also tell the linker
1644 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1646 By default, a local label is any label beginning with @samp{L}, but each
1647 target is allowed to redefine the local label prefix.
1649 On the HPPA local labels begin with @samp{L$}.
1653 @section Configuring listing output: @option{--listing}
1655 The listing feature of the assembler can be enabled via the command line switch
1656 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1657 hex dump of the corresponding locations in the output object file, and displays
1658 them as a listing file. The format of this listing can be controlled by pseudo
1659 ops inside the assembler source (@pxref{List} @pxref{Title} @pxref{Sbttl}
1660 @pxref{Psize} @pxref{Eject}) and also by the following switches:
1663 @item --listing-lhs-width=@samp{number}
1664 @kindex --listing-lhs-width
1665 @cindex Width of first line disassembly output
1666 Sets the maximum width, in words, of the first line of the hex byte dump. This
1667 dump appears on the left hand side of the listing output.
1669 @item --listing-lhs-width2=@samp{number}
1670 @kindex --listing-lhs-width2
1671 @cindex Width of continuation lines of disassembly output
1672 Sets the maximum width, in words, of any further lines of the hex byte dump for
1673 a given input source line. If this value is not specified, it defaults to being
1674 the same as the value specified for @samp{--listing-lhs-width}. If neither
1675 switch is used the default is to one.
1677 @item --listing-rhs-width=@samp{number}
1678 @kindex --listing-rhs-width
1679 @cindex Width of source line output
1680 Sets the maximum width, in characters, of the source line that is displayed
1681 alongside the hex dump. The default value for this parameter is 100. The
1682 source line is displayed on the right hand side of the listing output.
1684 @item --listing-cont-lines=@samp{number}
1685 @kindex --listing-cont-lines
1686 @cindex Maximum number of continuation lines
1687 Sets the maximum number of continuation lines of hex dump that will be
1688 displayed for a given single line of source input. The default value is 4.
1692 @section Assemble in MRI Compatibility Mode: @option{-M}
1695 @cindex MRI compatibility mode
1696 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1697 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1698 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1699 configured target) assembler from Microtec Research. The exact nature of the
1700 MRI syntax will not be documented here; see the MRI manuals for more
1701 information. Note in particular that the handling of macros and macro
1702 arguments is somewhat different. The purpose of this option is to permit
1703 assembling existing MRI assembler code using @command{@value{AS}}.
1705 The MRI compatibility is not complete. Certain operations of the MRI assembler
1706 depend upon its object file format, and can not be supported using other object
1707 file formats. Supporting these would require enhancing each object file format
1708 individually. These are:
1711 @item global symbols in common section
1713 The m68k MRI assembler supports common sections which are merged by the linker.
1714 Other object file formats do not support this. @command{@value{AS}} handles
1715 common sections by treating them as a single common symbol. It permits local
1716 symbols to be defined within a common section, but it can not support global
1717 symbols, since it has no way to describe them.
1719 @item complex relocations
1721 The MRI assemblers support relocations against a negated section address, and
1722 relocations which combine the start addresses of two or more sections. These
1723 are not support by other object file formats.
1725 @item @code{END} pseudo-op specifying start address
1727 The MRI @code{END} pseudo-op permits the specification of a start address.
1728 This is not supported by other object file formats. The start address may
1729 instead be specified using the @option{-e} option to the linker, or in a linker
1732 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1734 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1735 name to the output file. This is not supported by other object file formats.
1737 @item @code{ORG} pseudo-op
1739 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1740 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1741 which changes the location within the current section. Absolute sections are
1742 not supported by other object file formats. The address of a section may be
1743 assigned within a linker script.
1746 There are some other features of the MRI assembler which are not supported by
1747 @command{@value{AS}}, typically either because they are difficult or because they
1748 seem of little consequence. Some of these may be supported in future releases.
1752 @item EBCDIC strings
1754 EBCDIC strings are not supported.
1756 @item packed binary coded decimal
1758 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1759 and @code{DCB.P} pseudo-ops are not supported.
1761 @item @code{FEQU} pseudo-op
1763 The m68k @code{FEQU} pseudo-op is not supported.
1765 @item @code{NOOBJ} pseudo-op
1767 The m68k @code{NOOBJ} pseudo-op is not supported.
1769 @item @code{OPT} branch control options
1771 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1772 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1773 relaxes all branches, whether forward or backward, to an appropriate size, so
1774 these options serve no purpose.
1776 @item @code{OPT} list control options
1778 The following m68k @code{OPT} list control options are ignored: @code{C},
1779 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1780 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1782 @item other @code{OPT} options
1784 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1785 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1787 @item @code{OPT} @code{D} option is default
1789 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1790 @code{OPT NOD} may be used to turn it off.
1792 @item @code{XREF} pseudo-op.
1794 The m68k @code{XREF} pseudo-op is ignored.
1796 @item @code{.debug} pseudo-op
1798 The i960 @code{.debug} pseudo-op is not supported.
1800 @item @code{.extended} pseudo-op
1802 The i960 @code{.extended} pseudo-op is not supported.
1804 @item @code{.list} pseudo-op.
1806 The various options of the i960 @code{.list} pseudo-op are not supported.
1808 @item @code{.optimize} pseudo-op
1810 The i960 @code{.optimize} pseudo-op is not supported.
1812 @item @code{.output} pseudo-op
1814 The i960 @code{.output} pseudo-op is not supported.
1816 @item @code{.setreal} pseudo-op
1818 The i960 @code{.setreal} pseudo-op is not supported.
1823 @section Dependency Tracking: @option{--MD}
1826 @cindex dependency tracking
1829 @command{@value{AS}} can generate a dependency file for the file it creates. This
1830 file consists of a single rule suitable for @code{make} describing the
1831 dependencies of the main source file.
1833 The rule is written to the file named in its argument.
1835 This feature is used in the automatic updating of makefiles.
1838 @section Name the Object File: @option{-o}
1841 @cindex naming object file
1842 @cindex object file name
1843 There is always one object file output when you run @command{@value{AS}}. By
1844 default it has the name
1847 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1861 You use this option (which takes exactly one filename) to give the
1862 object file a different name.
1864 Whatever the object file is called, @command{@value{AS}} overwrites any
1865 existing file of the same name.
1868 @section Join Data and Text Sections: @option{-R}
1871 @cindex data and text sections, joining
1872 @cindex text and data sections, joining
1873 @cindex joining text and data sections
1874 @cindex merging text and data sections
1875 @option{-R} tells @command{@value{AS}} to write the object file as if all
1876 data-section data lives in the text section. This is only done at
1877 the very last moment: your binary data are the same, but data
1878 section parts are relocated differently. The data section part of
1879 your object file is zero bytes long because all its bytes are
1880 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1882 When you specify @option{-R} it would be possible to generate shorter
1883 address displacements (because we do not have to cross between text and
1884 data section). We refrain from doing this simply for compatibility with
1885 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
1888 When @command{@value{AS}} is configured for COFF or ELF output,
1889 this option is only useful if you use sections named @samp{.text} and
1894 @option{-R} is not supported for any of the HPPA targets. Using
1895 @option{-R} generates a warning from @command{@value{AS}}.
1899 @section Display Assembly Statistics: @option{--statistics}
1901 @kindex --statistics
1902 @cindex statistics, about assembly
1903 @cindex time, total for assembly
1904 @cindex space used, maximum for assembly
1905 Use @samp{--statistics} to display two statistics about the resources used by
1906 @command{@value{AS}}: the maximum amount of space allocated during the assembly
1907 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1910 @node traditional-format
1911 @section Compatible Output: @option{--traditional-format}
1913 @kindex --traditional-format
1914 For some targets, the output of @command{@value{AS}} is different in some ways
1915 from the output of some existing assembler. This switch requests
1916 @command{@value{AS}} to use the traditional format instead.
1918 For example, it disables the exception frame optimizations which
1919 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
1922 @section Announce Version: @option{-v}
1926 @cindex assembler version
1927 @cindex version of assembler
1928 You can find out what version of as is running by including the
1929 option @samp{-v} (which you can also spell as @samp{-version}) on the
1933 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
1935 @command{@value{AS}} should never give a warning or error message when
1936 assembling compiler output. But programs written by people often
1937 cause @command{@value{AS}} to give a warning that a particular assumption was
1938 made. All such warnings are directed to the standard error file.
1942 @cindex suppressing warnings
1943 @cindex warnings, suppressing
1944 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
1945 This only affects the warning messages: it does not change any particular of
1946 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
1949 @kindex --fatal-warnings
1950 @cindex errors, caused by warnings
1951 @cindex warnings, causing error
1952 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
1953 files that generate warnings to be in error.
1956 @cindex warnings, switching on
1957 You can switch these options off again by specifying @option{--warn}, which
1958 causes warnings to be output as usual.
1961 @section Generate Object File in Spite of Errors: @option{-Z}
1962 @cindex object file, after errors
1963 @cindex errors, continuing after
1964 After an error message, @command{@value{AS}} normally produces no output. If for
1965 some reason you are interested in object file output even after
1966 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
1967 option. If there are any errors, @command{@value{AS}} continues anyways, and
1968 writes an object file after a final warning message of the form @samp{@var{n}
1969 errors, @var{m} warnings, generating bad object file.}
1974 @cindex machine-independent syntax
1975 @cindex syntax, machine-independent
1976 This chapter describes the machine-independent syntax allowed in a
1977 source file. @command{@value{AS}} syntax is similar to what many other
1978 assemblers use; it is inspired by the BSD 4.2
1983 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
1987 * Preprocessing:: Preprocessing
1988 * Whitespace:: Whitespace
1989 * Comments:: Comments
1990 * Symbol Intro:: Symbols
1991 * Statements:: Statements
1992 * Constants:: Constants
1996 @section Preprocessing
1998 @cindex preprocessing
1999 The @command{@value{AS}} internal preprocessor:
2001 @cindex whitespace, removed by preprocessor
2003 adjusts and removes extra whitespace. It leaves one space or tab before
2004 the keywords on a line, and turns any other whitespace on the line into
2007 @cindex comments, removed by preprocessor
2009 removes all comments, replacing them with a single space, or an
2010 appropriate number of newlines.
2012 @cindex constants, converted by preprocessor
2014 converts character constants into the appropriate numeric values.
2017 It does not do macro processing, include file handling, or
2018 anything else you may get from your C compiler's preprocessor. You can
2019 do include file processing with the @code{.include} directive
2020 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2021 to get other ``CPP'' style preprocessing by giving the input file a
2022 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
2023 Output, gcc.info, Using GNU CC}.
2025 Excess whitespace, comments, and character constants
2026 cannot be used in the portions of the input text that are not
2029 @cindex turning preprocessing on and off
2030 @cindex preprocessing, turning on and off
2033 If the first line of an input file is @code{#NO_APP} or if you use the
2034 @samp{-f} option, whitespace and comments are not removed from the input file.
2035 Within an input file, you can ask for whitespace and comment removal in
2036 specific portions of the by putting a line that says @code{#APP} before the
2037 text that may contain whitespace or comments, and putting a line that says
2038 @code{#NO_APP} after this text. This feature is mainly intend to support
2039 @code{asm} statements in compilers whose output is otherwise free of comments
2046 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2047 Whitespace is used to separate symbols, and to make programs neater for
2048 people to read. Unless within character constants
2049 (@pxref{Characters,,Character Constants}), any whitespace means the same
2050 as exactly one space.
2056 There are two ways of rendering comments to @command{@value{AS}}. In both
2057 cases the comment is equivalent to one space.
2059 Anything from @samp{/*} through the next @samp{*/} is a comment.
2060 This means you may not nest these comments.
2064 The only way to include a newline ('\n') in a comment
2065 is to use this sort of comment.
2068 /* This sort of comment does not nest. */
2071 @cindex line comment character
2072 Anything from the @dfn{line comment} character to the next newline
2073 is considered a comment and is ignored. The line comment character is
2075 @samp{;} for the AMD 29K family;
2078 @samp{;} on the ARC;
2081 @samp{@@} on the ARM;
2084 @samp{;} for the H8/300 family;
2087 @samp{!} for the H8/500 family;
2090 @samp{;} for the HPPA;
2093 @samp{#} on the i386 and x86-64;
2096 @samp{#} on the i960;
2099 @samp{;} for the PDP-11;
2102 @samp{;} for picoJava;
2105 @samp{#} for Motorola PowerPC;
2108 @samp{!} for the Renesas / SuperH SH;
2111 @samp{!} on the SPARC;
2114 @samp{#} on the ip2k;
2117 @samp{#} on the m32r;
2120 @samp{|} on the 680x0;
2123 @samp{#} on the 68HC11 and 68HC12;
2126 @samp{;} on the M880x0;
2129 @samp{#} on the Vax;
2132 @samp{!} for the Z8000;
2135 @samp{#} on the V850;
2138 @samp{#} for Xtensa systems;
2140 see @ref{Machine Dependencies}. @refill
2141 @c FIXME What about i860?
2144 On some machines there are two different line comment characters. One
2145 character only begins a comment if it is the first non-whitespace character on
2146 a line, while the other always begins a comment.
2150 The V850 assembler also supports a double dash as starting a comment that
2151 extends to the end of the line.
2157 @cindex lines starting with @code{#}
2158 @cindex logical line numbers
2159 To be compatible with past assemblers, lines that begin with @samp{#} have a
2160 special interpretation. Following the @samp{#} should be an absolute
2161 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2162 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
2163 new logical file name. The rest of the line, if any, should be whitespace.
2165 If the first non-whitespace characters on the line are not numeric,
2166 the line is ignored. (Just like a comment.)
2169 # This is an ordinary comment.
2170 # 42-6 "new_file_name" # New logical file name
2171 # This is logical line # 36.
2173 This feature is deprecated, and may disappear from future versions
2174 of @command{@value{AS}}.
2179 @cindex characters used in symbols
2180 @ifclear SPECIAL-SYMS
2181 A @dfn{symbol} is one or more characters chosen from the set of all
2182 letters (both upper and lower case), digits and the three characters
2188 A @dfn{symbol} is one or more characters chosen from the set of all
2189 letters (both upper and lower case), digits and the three characters
2190 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2196 On most machines, you can also use @code{$} in symbol names; exceptions
2197 are noted in @ref{Machine Dependencies}.
2199 No symbol may begin with a digit. Case is significant.
2200 There is no length limit: all characters are significant. Symbols are
2201 delimited by characters not in that set, or by the beginning of a file
2202 (since the source program must end with a newline, the end of a file is
2203 not a possible symbol delimiter). @xref{Symbols}.
2204 @cindex length of symbols
2209 @cindex statements, structure of
2210 @cindex line separator character
2211 @cindex statement separator character
2213 @ifclear abnormal-separator
2214 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2215 semicolon (@samp{;}). The newline or semicolon is considered part of
2216 the preceding statement. Newlines and semicolons within character
2217 constants are an exception: they do not end statements.
2219 @ifset abnormal-separator
2221 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
2222 sign (@samp{@@}). The newline or at sign is considered part of the
2223 preceding statement. Newlines and at signs within character constants
2224 are an exception: they do not end statements.
2227 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2228 point (@samp{!}). The newline or exclamation point is considered part of the
2229 preceding statement. Newlines and exclamation points within character
2230 constants are an exception: they do not end statements.
2233 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2234 H8/300) a dollar sign (@samp{$}); or (for the
2237 (@samp{;}). The newline or separator character is considered part of
2238 the preceding statement. Newlines and separators within character
2239 constants are an exception: they do not end statements.
2244 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2245 separator character. (The line separator is usually @samp{;}, unless
2246 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
2247 newline or separator character is considered part of the preceding
2248 statement. Newlines and separators within character constants are an
2249 exception: they do not end statements.
2252 @cindex newline, required at file end
2253 @cindex EOF, newline must precede
2254 It is an error to end any statement with end-of-file: the last
2255 character of any input file should be a newline.@refill
2257 An empty statement is allowed, and may include whitespace. It is ignored.
2259 @cindex instructions and directives
2260 @cindex directives and instructions
2261 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2262 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2264 A statement begins with zero or more labels, optionally followed by a
2265 key symbol which determines what kind of statement it is. The key
2266 symbol determines the syntax of the rest of the statement. If the
2267 symbol begins with a dot @samp{.} then the statement is an assembler
2268 directive: typically valid for any computer. If the symbol begins with
2269 a letter the statement is an assembly language @dfn{instruction}: it
2270 assembles into a machine language instruction.
2272 Different versions of @command{@value{AS}} for different computers
2273 recognize different instructions. In fact, the same symbol may
2274 represent a different instruction in a different computer's assembly
2278 @cindex @code{:} (label)
2279 @cindex label (@code{:})
2280 A label is a symbol immediately followed by a colon (@code{:}).
2281 Whitespace before a label or after a colon is permitted, but you may not
2282 have whitespace between a label's symbol and its colon. @xref{Labels}.
2285 For HPPA targets, labels need not be immediately followed by a colon, but
2286 the definition of a label must begin in column zero. This also implies that
2287 only one label may be defined on each line.
2291 label: .directive followed by something
2292 another_label: # This is an empty statement.
2293 instruction operand_1, operand_2, @dots{}
2300 A constant is a number, written so that its value is known by
2301 inspection, without knowing any context. Like this:
2304 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2305 .ascii "Ring the bell\7" # A string constant.
2306 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2307 .float 0f-314159265358979323846264338327\
2308 95028841971.693993751E-40 # - pi, a flonum.
2313 * Characters:: Character Constants
2314 * Numbers:: Number Constants
2318 @subsection Character Constants
2320 @cindex character constants
2321 @cindex constants, character
2322 There are two kinds of character constants. A @dfn{character} stands
2323 for one character in one byte and its value may be used in
2324 numeric expressions. String constants (properly called string
2325 @emph{literals}) are potentially many bytes and their values may not be
2326 used in arithmetic expressions.
2330 * Chars:: Characters
2334 @subsubsection Strings
2336 @cindex string constants
2337 @cindex constants, string
2338 A @dfn{string} is written between double-quotes. It may contain
2339 double-quotes or null characters. The way to get special characters
2340 into a string is to @dfn{escape} these characters: precede them with
2341 a backslash @samp{\} character. For example @samp{\\} represents
2342 one backslash: the first @code{\} is an escape which tells
2343 @command{@value{AS}} to interpret the second character literally as a backslash
2344 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2345 escape character). The complete list of escapes follows.
2347 @cindex escape codes, character
2348 @cindex character escape codes
2351 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2353 @cindex @code{\b} (backspace character)
2354 @cindex backspace (@code{\b})
2356 Mnemonic for backspace; for ASCII this is octal code 010.
2359 @c Mnemonic for EOText; for ASCII this is octal code 004.
2361 @cindex @code{\f} (formfeed character)
2362 @cindex formfeed (@code{\f})
2364 Mnemonic for FormFeed; for ASCII this is octal code 014.
2366 @cindex @code{\n} (newline character)
2367 @cindex newline (@code{\n})
2369 Mnemonic for newline; for ASCII this is octal code 012.
2372 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2374 @cindex @code{\r} (carriage return character)
2375 @cindex carriage return (@code{\r})
2377 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2380 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2381 @c other assemblers.
2383 @cindex @code{\t} (tab)
2384 @cindex tab (@code{\t})
2386 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2389 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2390 @c @item \x @var{digit} @var{digit} @var{digit}
2391 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2393 @cindex @code{\@var{ddd}} (octal character code)
2394 @cindex octal character code (@code{\@var{ddd}})
2395 @item \ @var{digit} @var{digit} @var{digit}
2396 An octal character code. The numeric code is 3 octal digits.
2397 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2398 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2400 @cindex @code{\@var{xd...}} (hex character code)
2401 @cindex hex character code (@code{\@var{xd...}})
2402 @item \@code{x} @var{hex-digits...}
2403 A hex character code. All trailing hex digits are combined. Either upper or
2404 lower case @code{x} works.
2406 @cindex @code{\\} (@samp{\} character)
2407 @cindex backslash (@code{\\})
2409 Represents one @samp{\} character.
2412 @c Represents one @samp{'} (accent acute) character.
2413 @c This is needed in single character literals
2414 @c (@xref{Characters,,Character Constants}.) to represent
2417 @cindex @code{\"} (doublequote character)
2418 @cindex doublequote (@code{\"})
2420 Represents one @samp{"} character. Needed in strings to represent
2421 this character, because an unescaped @samp{"} would end the string.
2423 @item \ @var{anything-else}
2424 Any other character when escaped by @kbd{\} gives a warning, but
2425 assembles as if the @samp{\} was not present. The idea is that if
2426 you used an escape sequence you clearly didn't want the literal
2427 interpretation of the following character. However @command{@value{AS}} has no
2428 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2429 code and warns you of the fact.
2432 Which characters are escapable, and what those escapes represent,
2433 varies widely among assemblers. The current set is what we think
2434 the BSD 4.2 assembler recognizes, and is a subset of what most C
2435 compilers recognize. If you are in doubt, do not use an escape
2439 @subsubsection Characters
2441 @cindex single character constant
2442 @cindex character, single
2443 @cindex constant, single character
2444 A single character may be written as a single quote immediately
2445 followed by that character. The same escapes apply to characters as
2446 to strings. So if you want to write the character backslash, you
2447 must write @kbd{'\\} where the first @code{\} escapes the second
2448 @code{\}. As you can see, the quote is an acute accent, not a
2449 grave accent. A newline
2451 @ifclear abnormal-separator
2452 (or semicolon @samp{;})
2454 @ifset abnormal-separator
2456 (or at sign @samp{@@})
2459 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2460 Renesas SH or H8/500)
2464 immediately following an acute accent is taken as a literal character
2465 and does not count as the end of a statement. The value of a character
2466 constant in a numeric expression is the machine's byte-wide code for
2467 that character. @command{@value{AS}} assumes your character code is ASCII:
2468 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2471 @subsection Number Constants
2473 @cindex constants, number
2474 @cindex number constants
2475 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2476 are stored in the target machine. @emph{Integers} are numbers that
2477 would fit into an @code{int} in the C language. @emph{Bignums} are
2478 integers, but they are stored in more than 32 bits. @emph{Flonums}
2479 are floating point numbers, described below.
2482 * Integers:: Integers
2487 * Bit Fields:: Bit Fields
2493 @subsubsection Integers
2495 @cindex constants, integer
2497 @cindex binary integers
2498 @cindex integers, binary
2499 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2500 the binary digits @samp{01}.
2502 @cindex octal integers
2503 @cindex integers, octal
2504 An octal integer is @samp{0} followed by zero or more of the octal
2505 digits (@samp{01234567}).
2507 @cindex decimal integers
2508 @cindex integers, decimal
2509 A decimal integer starts with a non-zero digit followed by zero or
2510 more digits (@samp{0123456789}).
2512 @cindex hexadecimal integers
2513 @cindex integers, hexadecimal
2514 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2515 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2517 Integers have the usual values. To denote a negative integer, use
2518 the prefix operator @samp{-} discussed under expressions
2519 (@pxref{Prefix Ops,,Prefix Operators}).
2522 @subsubsection Bignums
2525 @cindex constants, bignum
2526 A @dfn{bignum} has the same syntax and semantics as an integer
2527 except that the number (or its negative) takes more than 32 bits to
2528 represent in binary. The distinction is made because in some places
2529 integers are permitted while bignums are not.
2532 @subsubsection Flonums
2534 @cindex floating point numbers
2535 @cindex constants, floating point
2537 @cindex precision, floating point
2538 A @dfn{flonum} represents a floating point number. The translation is
2539 indirect: a decimal floating point number from the text is converted by
2540 @command{@value{AS}} to a generic binary floating point number of more than
2541 sufficient precision. This generic floating point number is converted
2542 to a particular computer's floating point format (or formats) by a
2543 portion of @command{@value{AS}} specialized to that computer.
2545 A flonum is written by writing (in order)
2550 (@samp{0} is optional on the HPPA.)
2554 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2556 @kbd{e} is recommended. Case is not important.
2558 @c FIXME: verify if flonum syntax really this vague for most cases
2559 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2560 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2563 On the H8/300, H8/500,
2564 Renesas / SuperH SH,
2565 and AMD 29K architectures, the letter must be
2566 one of the letters @samp{DFPRSX} (in upper or lower case).
2568 On the ARC, the letter must be one of the letters @samp{DFRS}
2569 (in upper or lower case).
2571 On the Intel 960 architecture, the letter must be
2572 one of the letters @samp{DFT} (in upper or lower case).
2574 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2578 One of the letters @samp{DFPRSX} (in upper or lower case).
2581 One of the letters @samp{DFRS} (in upper or lower case).
2584 One of the letters @samp{DFPRSX} (in upper or lower case).
2587 The letter @samp{E} (upper case only).
2590 One of the letters @samp{DFT} (in upper or lower case).
2595 An optional sign: either @samp{+} or @samp{-}.
2598 An optional @dfn{integer part}: zero or more decimal digits.
2601 An optional @dfn{fractional part}: @samp{.} followed by zero
2602 or more decimal digits.
2605 An optional exponent, consisting of:
2609 An @samp{E} or @samp{e}.
2610 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2611 @c principle this can perfectly well be different on different targets.
2613 Optional sign: either @samp{+} or @samp{-}.
2615 One or more decimal digits.
2620 At least one of the integer part or the fractional part must be
2621 present. The floating point number has the usual base-10 value.
2623 @command{@value{AS}} does all processing using integers. Flonums are computed
2624 independently of any floating point hardware in the computer running
2625 @command{@value{AS}}.
2629 @c Bit fields are written as a general facility but are also controlled
2630 @c by a conditional-compilation flag---which is as of now (21mar91)
2631 @c turned on only by the i960 config of GAS.
2633 @subsubsection Bit Fields
2636 @cindex constants, bit field
2637 You can also define numeric constants as @dfn{bit fields}.
2638 specify two numbers separated by a colon---
2640 @var{mask}:@var{value}
2643 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2646 The resulting number is then packed
2648 @c this conditional paren in case bit fields turned on elsewhere than 960
2649 (in host-dependent byte order)
2651 into a field whose width depends on which assembler directive has the
2652 bit-field as its argument. Overflow (a result from the bitwise and
2653 requiring more binary digits to represent) is not an error; instead,
2654 more constants are generated, of the specified width, beginning with the
2655 least significant digits.@refill
2657 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2658 @code{.short}, and @code{.word} accept bit-field arguments.
2663 @chapter Sections and Relocation
2668 * Secs Background:: Background
2669 * Ld Sections:: Linker Sections
2670 * As Sections:: Assembler Internal Sections
2671 * Sub-Sections:: Sub-Sections
2675 @node Secs Background
2678 Roughly, a section is a range of addresses, with no gaps; all data
2679 ``in'' those addresses is treated the same for some particular purpose.
2680 For example there may be a ``read only'' section.
2682 @cindex linker, and assembler
2683 @cindex assembler, and linker
2684 The linker @code{@value{LD}} reads many object files (partial programs) and
2685 combines their contents to form a runnable program. When @command{@value{AS}}
2686 emits an object file, the partial program is assumed to start at address 0.
2687 @code{@value{LD}} assigns the final addresses for the partial program, so that
2688 different partial programs do not overlap. This is actually an
2689 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2692 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2693 addresses. These blocks slide to their run-time addresses as rigid
2694 units; their length does not change and neither does the order of bytes
2695 within them. Such a rigid unit is called a @emph{section}. Assigning
2696 run-time addresses to sections is called @dfn{relocation}. It includes
2697 the task of adjusting mentions of object-file addresses so they refer to
2698 the proper run-time addresses.
2700 For the H8/300 and H8/500,
2701 and for the Renesas / SuperH SH,
2702 @command{@value{AS}} pads sections if needed to
2703 ensure they end on a word (sixteen bit) boundary.
2706 @cindex standard assembler sections
2707 An object file written by @command{@value{AS}} has at least three sections, any
2708 of which may be empty. These are named @dfn{text}, @dfn{data} and
2713 When it generates COFF or ELF output,
2715 @command{@value{AS}} can also generate whatever other named sections you specify
2716 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2717 If you do not use any directives that place output in the @samp{.text}
2718 or @samp{.data} sections, these sections still exist, but are empty.
2723 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2725 @command{@value{AS}} can also generate whatever other named sections you
2726 specify using the @samp{.space} and @samp{.subspace} directives. See
2727 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2728 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2729 assembler directives.
2732 Additionally, @command{@value{AS}} uses different names for the standard
2733 text, data, and bss sections when generating SOM output. Program text
2734 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2735 BSS into @samp{$BSS$}.
2739 Within the object file, the text section starts at address @code{0}, the
2740 data section follows, and the bss section follows the data section.
2743 When generating either SOM or ELF output files on the HPPA, the text
2744 section starts at address @code{0}, the data section at address
2745 @code{0x4000000}, and the bss section follows the data section.
2748 To let @code{@value{LD}} know which data changes when the sections are
2749 relocated, and how to change that data, @command{@value{AS}} also writes to the
2750 object file details of the relocation needed. To perform relocation
2751 @code{@value{LD}} must know, each time an address in the object
2755 Where in the object file is the beginning of this reference to
2758 How long (in bytes) is this reference?
2760 Which section does the address refer to? What is the numeric value of
2762 (@var{address}) @minus{} (@var{start-address of section})?
2765 Is the reference to an address ``Program-Counter relative''?
2768 @cindex addresses, format of
2769 @cindex section-relative addressing
2770 In fact, every address @command{@value{AS}} ever uses is expressed as
2772 (@var{section}) + (@var{offset into section})
2775 Further, most expressions @command{@value{AS}} computes have this section-relative
2778 (For some object formats, such as SOM for the HPPA, some expressions are
2779 symbol-relative instead.)
2782 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2783 @var{N} into section @var{secname}.''
2785 Apart from text, data and bss sections you need to know about the
2786 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2787 addresses in the absolute section remain unchanged. For example, address
2788 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2789 @code{@value{LD}}. Although the linker never arranges two partial programs'
2790 data sections with overlapping addresses after linking, @emph{by definition}
2791 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2792 part of a program is always the same address when the program is running as
2793 address @code{@{absolute@ 239@}} in any other part of the program.
2795 The idea of sections is extended to the @dfn{undefined} section. Any
2796 address whose section is unknown at assembly time is by definition
2797 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2798 Since numbers are always defined, the only way to generate an undefined
2799 address is to mention an undefined symbol. A reference to a named
2800 common block would be such a symbol: its value is unknown at assembly
2801 time so it has section @emph{undefined}.
2803 By analogy the word @emph{section} is used to describe groups of sections in
2804 the linked program. @code{@value{LD}} puts all partial programs' text
2805 sections in contiguous addresses in the linked program. It is
2806 customary to refer to the @emph{text section} of a program, meaning all
2807 the addresses of all partial programs' text sections. Likewise for
2808 data and bss sections.
2810 Some sections are manipulated by @code{@value{LD}}; others are invented for
2811 use of @command{@value{AS}} and have no meaning except during assembly.
2814 @section Linker Sections
2815 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2820 @cindex named sections
2821 @cindex sections, named
2822 @item named sections
2825 @cindex text section
2826 @cindex data section
2830 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2831 separate but equal sections. Anything you can say of one section is
2834 When the program is running, however, it is
2835 customary for the text section to be unalterable. The
2836 text section is often shared among processes: it contains
2837 instructions, constants and the like. The data section of a running
2838 program is usually alterable: for example, C variables would be stored
2839 in the data section.
2844 This section contains zeroed bytes when your program begins running. It
2845 is used to hold uninitialized variables or common storage. The length of
2846 each partial program's bss section is important, but because it starts
2847 out containing zeroed bytes there is no need to store explicit zero
2848 bytes in the object file. The bss section was invented to eliminate
2849 those explicit zeros from object files.
2851 @cindex absolute section
2852 @item absolute section
2853 Address 0 of this section is always ``relocated'' to runtime address 0.
2854 This is useful if you want to refer to an address that @code{@value{LD}} must
2855 not change when relocating. In this sense we speak of absolute
2856 addresses being ``unrelocatable'': they do not change during relocation.
2858 @cindex undefined section
2859 @item undefined section
2860 This ``section'' is a catch-all for address references to objects not in
2861 the preceding sections.
2862 @c FIXME: ref to some other doc on obj-file formats could go here.
2865 @cindex relocation example
2866 An idealized example of three relocatable sections follows.
2868 The example uses the traditional section names @samp{.text} and @samp{.data}.
2870 Memory addresses are on the horizontal axis.
2874 @c END TEXI2ROFF-KILL
2877 partial program # 1: |ttttt|dddd|00|
2884 partial program # 2: |TTT|DDD|000|
2887 +--+---+-----+--+----+---+-----+~~
2888 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2889 +--+---+-----+--+----+---+-----+~~
2891 addresses: 0 @dots{}
2898 \line{\it Partial program \#1: \hfil}
2899 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2900 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2902 \line{\it Partial program \#2: \hfil}
2903 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2904 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2906 \line{\it linked program: \hfil}
2907 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2908 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2909 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2910 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2912 \line{\it addresses: \hfil}
2916 @c END TEXI2ROFF-KILL
2919 @section Assembler Internal Sections
2921 @cindex internal assembler sections
2922 @cindex sections in messages, internal
2923 These sections are meant only for the internal use of @command{@value{AS}}. They
2924 have no meaning at run-time. You do not really need to know about these
2925 sections for most purposes; but they can be mentioned in @command{@value{AS}}
2926 warning messages, so it might be helpful to have an idea of their
2927 meanings to @command{@value{AS}}. These sections are used to permit the
2928 value of every expression in your assembly language program to be a
2929 section-relative address.
2932 @cindex assembler internal logic error
2933 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2934 An internal assembler logic error has been found. This means there is a
2935 bug in the assembler.
2937 @cindex expr (internal section)
2939 The assembler stores complex expression internally as combinations of
2940 symbols. When it needs to represent an expression as a symbol, it puts
2941 it in the expr section.
2943 @c FIXME item transfer[t] vector preload
2944 @c FIXME item transfer[t] vector postload
2945 @c FIXME item register
2949 @section Sub-Sections
2951 @cindex numbered subsections
2952 @cindex grouping data
2958 fall into two sections: text and data.
2960 You may have separate groups of
2962 data in named sections
2966 data in named sections
2972 that you want to end up near to each other in the object file, even though they
2973 are not contiguous in the assembler source. @command{@value{AS}} allows you to
2974 use @dfn{subsections} for this purpose. Within each section, there can be
2975 numbered subsections with values from 0 to 8192. Objects assembled into the
2976 same subsection go into the object file together with other objects in the same
2977 subsection. For example, a compiler might want to store constants in the text
2978 section, but might not want to have them interspersed with the program being
2979 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2980 section of code being output, and a @samp{.text 1} before each group of
2981 constants being output.
2983 Subsections are optional. If you do not use subsections, everything
2984 goes in subsection number zero.
2987 Each subsection is zero-padded up to a multiple of four bytes.
2988 (Subsections may be padded a different amount on different flavors
2989 of @command{@value{AS}}.)
2993 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2994 boundary (two bytes).
2995 The same is true on the Renesas SH.
2998 @c FIXME section padding (alignment)?
2999 @c Rich Pixley says padding here depends on target obj code format; that
3000 @c doesn't seem particularly useful to say without further elaboration,
3001 @c so for now I say nothing about it. If this is a generic BFD issue,
3002 @c these paragraphs might need to vanish from this manual, and be
3003 @c discussed in BFD chapter of binutils (or some such).
3006 On the AMD 29K family, no particular padding is added to section or
3007 subsection sizes; @value{AS} forces no alignment on this platform.
3011 Subsections appear in your object file in numeric order, lowest numbered
3012 to highest. (All this to be compatible with other people's assemblers.)
3013 The object file contains no representation of subsections; @code{@value{LD}} and
3014 other programs that manipulate object files see no trace of them.
3015 They just see all your text subsections as a text section, and all your
3016 data subsections as a data section.
3018 To specify which subsection you want subsequent statements assembled
3019 into, use a numeric argument to specify it, in a @samp{.text
3020 @var{expression}} or a @samp{.data @var{expression}} statement.
3023 When generating COFF or ELF output, you
3028 can also use an extra subsection
3029 argument with arbitrary named sections: @samp{.section @var{name},
3032 @var{Expression} should be an absolute expression.
3033 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
3034 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3035 begins in @code{text 0}. For instance:
3037 .text 0 # The default subsection is text 0 anyway.
3038 .ascii "This lives in the first text subsection. *"
3040 .ascii "But this lives in the second text subsection."
3042 .ascii "This lives in the data section,"
3043 .ascii "in the first data subsection."
3045 .ascii "This lives in the first text section,"
3046 .ascii "immediately following the asterisk (*)."
3049 Each section has a @dfn{location counter} incremented by one for every byte
3050 assembled into that section. Because subsections are merely a convenience
3051 restricted to @command{@value{AS}} there is no concept of a subsection location
3052 counter. There is no way to directly manipulate a location counter---but the
3053 @code{.align} directive changes it, and any label definition captures its
3054 current value. The location counter of the section where statements are being
3055 assembled is said to be the @dfn{active} location counter.
3058 @section bss Section
3061 @cindex common variable storage
3062 The bss section is used for local common variable storage.
3063 You may allocate address space in the bss section, but you may
3064 not dictate data to load into it before your program executes. When
3065 your program starts running, all the contents of the bss
3066 section are zeroed bytes.
3068 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3069 @ref{Lcomm,,@code{.lcomm}}.
3071 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3072 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
3075 When assembling for a target which supports multiple sections, such as ELF or
3076 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3077 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3078 section. Typically the section will only contain symbol definitions and
3079 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3086 Symbols are a central concept: the programmer uses symbols to name
3087 things, the linker uses symbols to link, and the debugger uses symbols
3091 @cindex debuggers, and symbol order
3092 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3093 the same order they were declared. This may break some debuggers.
3098 * Setting Symbols:: Giving Symbols Other Values
3099 * Symbol Names:: Symbol Names
3100 * Dot:: The Special Dot Symbol
3101 * Symbol Attributes:: Symbol Attributes
3108 A @dfn{label} is written as a symbol immediately followed by a colon
3109 @samp{:}. The symbol then represents the current value of the
3110 active location counter, and is, for example, a suitable instruction
3111 operand. You are warned if you use the same symbol to represent two
3112 different locations: the first definition overrides any other
3116 On the HPPA, the usual form for a label need not be immediately followed by a
3117 colon, but instead must start in column zero. Only one label may be defined on
3118 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3119 provides a special directive @code{.label} for defining labels more flexibly.
3122 @node Setting Symbols
3123 @section Giving Symbols Other Values
3125 @cindex assigning values to symbols
3126 @cindex symbol values, assigning
3127 A symbol can be given an arbitrary value by writing a symbol, followed
3128 by an equals sign @samp{=}, followed by an expression
3129 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3130 directive. @xref{Set,,@code{.set}}.
3133 @section Symbol Names
3135 @cindex symbol names
3136 @cindex names, symbol
3137 @ifclear SPECIAL-SYMS
3138 Symbol names begin with a letter or with one of @samp{._}. On most
3139 machines, you can also use @code{$} in symbol names; exceptions are
3140 noted in @ref{Machine Dependencies}. That character may be followed by any
3141 string of digits, letters, dollar signs (unless otherwise noted in
3142 @ref{Machine Dependencies}), and underscores.
3145 For the AMD 29K family, @samp{?} is also allowed in the
3146 body of a symbol name, though not at its beginning.
3151 Symbol names begin with a letter or with one of @samp{._}. On the
3152 Renesas SH or the H8/500, you can also use @code{$} in symbol names. That
3153 character may be followed by any string of digits, letters, dollar signs (save
3154 on the H8/300), and underscores.
3158 Case of letters is significant: @code{foo} is a different symbol name
3161 Each symbol has exactly one name. Each name in an assembly language program
3162 refers to exactly one symbol. You may use that symbol name any number of times
3165 @subheading Local Symbol Names
3167 @cindex local symbol names
3168 @cindex symbol names, local
3169 @cindex temporary symbol names
3170 @cindex symbol names, temporary
3171 Local symbols help compilers and programmers use names temporarily.
3172 They create symbols which are guaranteed to be unique over the entire scope of
3173 the input source code and which can be referred to by a simple notation.
3174 To define a local symbol, write a label of the form @samp{@b{N}:} (where @b{N}
3175 represents any positive integer). To refer to the most recent previous
3176 definition of that symbol write @samp{@b{N}b}, using the same number as when
3177 you defined the label. To refer to the next definition of a local label, write
3178 @samp{@b{N}f}--- The @samp{b} stands for``backwards'' and the @samp{f} stands
3181 There is no restriction on how you can use these labels, and you can reuse them
3182 too. So that it is possible to repeatedly define the same local label (using
3183 the same number @samp{@b{N}}), although you can only refer to the most recently
3184 defined local label of that number (for a backwards reference) or the next
3185 definition of a specific local label for a forward reference. It is also worth
3186 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3187 implemented in a slightly more efficient manner than the others.
3198 Which is the equivalent of:
3201 label_1: branch label_3
3202 label_2: branch label_1
3203 label_3: branch label_4
3204 label_4: branch label_3
3207 Local symbol names are only a notational device. They are immediately
3208 transformed into more conventional symbol names before the assembler uses them.
3209 The symbol names stored in the symbol table, appearing in error messages and
3210 optionally emitted to the object file. The names are constructed using these
3215 All local labels begin with @samp{L}. Normally both @command{@value{AS}} and
3216 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
3217 used for symbols you are never intended to see. If you use the
3218 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3219 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3220 you may use them in debugging.
3223 This is the number that was used in the local label definition. So if the
3224 label is written @samp{55:} then the number is @samp{55}.
3227 This unusual character is included so you do not accidentally invent a symbol
3228 of the same name. The character has ASCII value of @samp{\002} (control-B).
3230 @item @emph{ordinal number}
3231 This is a serial number to keep the labels distinct. The first definition of
3232 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3233 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3234 the number @samp{1} and its 15th defintion gets @samp{15} as well.
3237 So for example, the first @code{1:} is named @code{L1@kbd{C-B}1}, the 44th
3238 @code{3:} is named @code{L3@kbd{C-B}44}.
3240 @subheading Dollar Local Labels
3241 @cindex dollar local symbols
3243 @code{@value{AS}} also supports an even more local form of local labels called
3244 dollar labels. These labels go out of scope (ie they become undefined) as soon
3245 as a non-local label is defined. Thus they remain valid for only a small
3246 region of the input source code. Normal local labels, by contrast, remain in
3247 scope for the entire file, or until they are redefined by another occurrence of
3248 the same local label.
3250 Dollar labels are defined in exactly the same way as ordinary local labels,
3251 except that instead of being terminated by a colon, they are terminated by a
3252 dollar sign. eg @samp{@b{55$}}.
3254 They can also be distinguished from ordinary local labels by their transformed
3255 name which uses ASCII character @samp{\001} (control-A) as the magic character
3256 to distinguish them from ordinary labels. Thus the 5th defintion of @samp{6$}
3257 is named @samp{L6@kbd{C-A}5}.
3260 @section The Special Dot Symbol
3262 @cindex dot (symbol)
3263 @cindex @code{.} (symbol)
3264 @cindex current address
3265 @cindex location counter
3266 The special symbol @samp{.} refers to the current address that
3267 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3268 .long .} defines @code{melvin} to contain its own address.
3269 Assigning a value to @code{.} is treated the same as a @code{.org}
3270 directive. Thus, the expression @samp{.=.+4} is the same as saying
3271 @ifclear no-space-dir
3280 @node Symbol Attributes
3281 @section Symbol Attributes
3283 @cindex symbol attributes
3284 @cindex attributes, symbol
3285 Every symbol has, as well as its name, the attributes ``Value'' and
3286 ``Type''. Depending on output format, symbols can also have auxiliary
3289 The detailed definitions are in @file{a.out.h}.
3292 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3293 all these attributes, and probably won't warn you. This makes the
3294 symbol an externally defined symbol, which is generally what you
3298 * Symbol Value:: Value
3299 * Symbol Type:: Type
3302 * a.out Symbols:: Symbol Attributes: @code{a.out}
3306 * a.out Symbols:: Symbol Attributes: @code{a.out}
3309 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3314 * COFF Symbols:: Symbol Attributes for COFF
3317 * SOM Symbols:: Symbol Attributes for SOM
3324 @cindex value of a symbol
3325 @cindex symbol value
3326 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3327 location in the text, data, bss or absolute sections the value is the
3328 number of addresses from the start of that section to the label.
3329 Naturally for text, data and bss sections the value of a symbol changes
3330 as @code{@value{LD}} changes section base addresses during linking. Absolute
3331 symbols' values do not change during linking: that is why they are
3334 The value of an undefined symbol is treated in a special way. If it is
3335 0 then the symbol is not defined in this assembler source file, and
3336 @code{@value{LD}} tries to determine its value from other files linked into the
3337 same program. You make this kind of symbol simply by mentioning a symbol
3338 name without defining it. A non-zero value represents a @code{.comm}
3339 common declaration. The value is how much common storage to reserve, in
3340 bytes (addresses). The symbol refers to the first address of the
3346 @cindex type of a symbol
3348 The type attribute of a symbol contains relocation (section)
3349 information, any flag settings indicating that a symbol is external, and
3350 (optionally), other information for linkers and debuggers. The exact
3351 format depends on the object-code output format in use.
3356 @c The following avoids a "widow" subsection title. @group would be
3357 @c better if it were available outside examples.
3360 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3362 @cindex @code{b.out} symbol attributes
3363 @cindex symbol attributes, @code{b.out}
3364 These symbol attributes appear only when @command{@value{AS}} is configured for
3365 one of the Berkeley-descended object output formats---@code{a.out} or
3371 @subsection Symbol Attributes: @code{a.out}
3373 @cindex @code{a.out} symbol attributes
3374 @cindex symbol attributes, @code{a.out}
3380 @subsection Symbol Attributes: @code{a.out}
3382 @cindex @code{a.out} symbol attributes
3383 @cindex symbol attributes, @code{a.out}
3387 * Symbol Desc:: Descriptor
3388 * Symbol Other:: Other
3392 @subsubsection Descriptor
3394 @cindex descriptor, of @code{a.out} symbol
3395 This is an arbitrary 16-bit value. You may establish a symbol's
3396 descriptor value by using a @code{.desc} statement
3397 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3398 @command{@value{AS}}.
3401 @subsubsection Other
3403 @cindex other attribute, of @code{a.out} symbol
3404 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3409 @subsection Symbol Attributes for COFF
3411 @cindex COFF symbol attributes
3412 @cindex symbol attributes, COFF
3414 The COFF format supports a multitude of auxiliary symbol attributes;
3415 like the primary symbol attributes, they are set between @code{.def} and
3416 @code{.endef} directives.
3418 @subsubsection Primary Attributes
3420 @cindex primary attributes, COFF symbols
3421 The symbol name is set with @code{.def}; the value and type,
3422 respectively, with @code{.val} and @code{.type}.
3424 @subsubsection Auxiliary Attributes
3426 @cindex auxiliary attributes, COFF symbols
3427 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3428 @code{.size}, and @code{.tag} can generate auxiliary symbol table
3429 information for COFF.
3434 @subsection Symbol Attributes for SOM
3436 @cindex SOM symbol attributes
3437 @cindex symbol attributes, SOM
3439 The SOM format for the HPPA supports a multitude of symbol attributes set with
3440 the @code{.EXPORT} and @code{.IMPORT} directives.
3442 The attributes are described in @cite{HP9000 Series 800 Assembly
3443 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3444 @code{EXPORT} assembler directive documentation.
3448 @chapter Expressions
3452 @cindex numeric values
3453 An @dfn{expression} specifies an address or numeric value.
3454 Whitespace may precede and/or follow an expression.
3456 The result of an expression must be an absolute number, or else an offset into
3457 a particular section. If an expression is not absolute, and there is not
3458 enough information when @command{@value{AS}} sees the expression to know its
3459 section, a second pass over the source program might be necessary to interpret
3460 the expression---but the second pass is currently not implemented.
3461 @command{@value{AS}} aborts with an error message in this situation.
3464 * Empty Exprs:: Empty Expressions
3465 * Integer Exprs:: Integer Expressions
3469 @section Empty Expressions
3471 @cindex empty expressions
3472 @cindex expressions, empty
3473 An empty expression has no value: it is just whitespace or null.
3474 Wherever an absolute expression is required, you may omit the
3475 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3476 is compatible with other assemblers.
3479 @section Integer Expressions
3481 @cindex integer expressions
3482 @cindex expressions, integer
3483 An @dfn{integer expression} is one or more @emph{arguments} delimited
3484 by @emph{operators}.
3487 * Arguments:: Arguments
3488 * Operators:: Operators
3489 * Prefix Ops:: Prefix Operators
3490 * Infix Ops:: Infix Operators
3494 @subsection Arguments
3496 @cindex expression arguments
3497 @cindex arguments in expressions
3498 @cindex operands in expressions
3499 @cindex arithmetic operands
3500 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3501 contexts arguments are sometimes called ``arithmetic operands''. In
3502 this manual, to avoid confusing them with the ``instruction operands'' of
3503 the machine language, we use the term ``argument'' to refer to parts of
3504 expressions only, reserving the word ``operand'' to refer only to machine
3505 instruction operands.
3507 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3508 @var{section} is one of text, data, bss, absolute,
3509 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3512 Numbers are usually integers.
3514 A number can be a flonum or bignum. In this case, you are warned
3515 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3516 these 32 bits are an integer. You may write integer-manipulating
3517 instructions that act on exotic constants, compatible with other
3520 @cindex subexpressions
3521 Subexpressions are a left parenthesis @samp{(} followed by an integer
3522 expression, followed by a right parenthesis @samp{)}; or a prefix
3523 operator followed by an argument.
3526 @subsection Operators
3528 @cindex operators, in expressions
3529 @cindex arithmetic functions
3530 @cindex functions, in expressions
3531 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3532 operators are followed by an argument. Infix operators appear
3533 between their arguments. Operators may be preceded and/or followed by
3537 @subsection Prefix Operator
3539 @cindex prefix operators
3540 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3541 one argument, which must be absolute.
3543 @c the tex/end tex stuff surrounding this small table is meant to make
3544 @c it align, on the printed page, with the similar table in the next
3545 @c section (which is inside an enumerate).
3547 \global\advance\leftskip by \itemindent
3552 @dfn{Negation}. Two's complement negation.
3554 @dfn{Complementation}. Bitwise not.
3558 \global\advance\leftskip by -\itemindent
3562 @subsection Infix Operators
3564 @cindex infix operators
3565 @cindex operators, permitted arguments
3566 @dfn{Infix operators} take two arguments, one on either side. Operators
3567 have precedence, but operations with equal precedence are performed left
3568 to right. Apart from @code{+} or @option{-}, both arguments must be
3569 absolute, and the result is absolute.
3572 @cindex operator precedence
3573 @cindex precedence of operators
3580 @dfn{Multiplication}.
3583 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3590 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3594 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3598 Intermediate precedence
3603 @dfn{Bitwise Inclusive Or}.
3609 @dfn{Bitwise Exclusive Or}.
3612 @dfn{Bitwise Or Not}.
3619 @cindex addition, permitted arguments
3620 @cindex plus, permitted arguments
3621 @cindex arguments for addition
3623 @dfn{Addition}. If either argument is absolute, the result has the section of
3624 the other argument. You may not add together arguments from different
3627 @cindex subtraction, permitted arguments
3628 @cindex minus, permitted arguments
3629 @cindex arguments for subtraction
3631 @dfn{Subtraction}. If the right argument is absolute, the
3632 result has the section of the left argument.
3633 If both arguments are in the same section, the result is absolute.
3634 You may not subtract arguments from different sections.
3635 @c FIXME is there still something useful to say about undefined - undefined ?
3637 @cindex comparison expressions
3638 @cindex expressions, comparison
3642 @dfn{Is Not Equal To}
3646 @dfn{Is Greater Than}
3648 @dfn{Is Greater Than Or Equal To}
3650 @dfn{Is Less Than Or Equal To}
3652 The comparison operators can be used as infix operators. A true results has a
3653 value of -1 whereas a false result has a value of 0. Note, these operators
3654 perform signed comparisons.
3657 @item Lowest Precedence
3666 These two logical operations can be used to combine the results of sub
3667 expressions. Note, unlike the comparison operators a true result returns a
3668 value of 1 but a false results does still return 0. Also note that the logical
3669 or operator has a slightly lower precedence than logical and.
3674 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3675 address; you can only have a defined section in one of the two arguments.
3678 @chapter Assembler Directives
3680 @cindex directives, machine independent
3681 @cindex pseudo-ops, machine independent
3682 @cindex machine independent directives
3683 All assembler directives have names that begin with a period (@samp{.}).
3684 The rest of the name is letters, usually in lower case.
3686 This chapter discusses directives that are available regardless of the
3687 target machine configuration for the @sc{gnu} assembler.
3689 Some machine configurations provide additional directives.
3690 @xref{Machine Dependencies}.
3693 @ifset machine-directives
3694 @xref{Machine Dependencies} for additional directives.
3699 * Abort:: @code{.abort}
3701 * ABORT:: @code{.ABORT}
3704 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3705 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3706 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3707 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3708 * Byte:: @code{.byte @var{expressions}}
3709 * Comm:: @code{.comm @var{symbol} , @var{length} }
3711 * CFI directives:: @code{.cfi_startproc}, @code{.cfi_endproc}, etc.
3713 * Data:: @code{.data @var{subsection}}
3715 * Def:: @code{.def @var{name}}
3718 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3724 * Double:: @code{.double @var{flonums}}
3725 * Eject:: @code{.eject}
3726 * Else:: @code{.else}
3727 * Elseif:: @code{.elseif}
3730 * Endef:: @code{.endef}
3733 * Endfunc:: @code{.endfunc}
3734 * Endif:: @code{.endif}
3735 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3736 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3738 * Exitm:: @code{.exitm}
3739 * Extern:: @code{.extern}
3740 * Fail:: @code{.fail}
3741 @ifclear no-file-dir
3742 * File:: @code{.file @var{string}}
3745 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3746 * Float:: @code{.float @var{flonums}}
3747 * Func:: @code{.func}
3748 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3750 * Hidden:: @code{.hidden @var{names}}
3753 * hword:: @code{.hword @var{expressions}}
3754 * Ident:: @code{.ident}
3755 * If:: @code{.if @var{absolute expression}}
3756 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3757 * Include:: @code{.include "@var{file}"}
3758 * Int:: @code{.int @var{expressions}}
3760 * Internal:: @code{.internal @var{names}}
3763 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3764 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3765 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3766 * Lflags:: @code{.lflags}
3767 @ifclear no-line-dir
3768 * Line:: @code{.line @var{line-number}}
3771 * Ln:: @code{.ln @var{line-number}}
3772 * Linkonce:: @code{.linkonce [@var{type}]}
3773 * List:: @code{.list}
3774 * Long:: @code{.long @var{expressions}}
3776 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3779 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3780 * MRI:: @code{.mri @var{val}}
3781 * Nolist:: @code{.nolist}
3782 * Octa:: @code{.octa @var{bignums}}
3783 * Org:: @code{.org @var{new-lc} , @var{fill}}
3784 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3786 * PopSection:: @code{.popsection}
3787 * Previous:: @code{.previous}
3790 * Print:: @code{.print @var{string}}
3792 * Protected:: @code{.protected @var{names}}
3795 * Psize:: @code{.psize @var{lines}, @var{columns}}
3796 * Purgem:: @code{.purgem @var{name}}
3798 * PushSection:: @code{.pushsection @var{name}}
3801 * Quad:: @code{.quad @var{bignums}}
3802 * Rept:: @code{.rept @var{count}}
3803 * Sbttl:: @code{.sbttl "@var{subheading}"}
3805 * Scl:: @code{.scl @var{class}}
3808 * Section:: @code{.section @var{name}}
3811 * Set:: @code{.set @var{symbol}, @var{expression}}
3812 * Short:: @code{.short @var{expressions}}
3813 * Single:: @code{.single @var{flonums}}
3815 * Size:: @code{.size [@var{name} , @var{expression}]}
3818 * Skip:: @code{.skip @var{size} , @var{fill}}
3819 * Sleb128:: @code{.sleb128 @var{expressions}}
3820 * Space:: @code{.space @var{size} , @var{fill}}
3822 * Stab:: @code{.stabd, .stabn, .stabs}
3825 * String:: @code{.string "@var{str}"}
3826 * Struct:: @code{.struct @var{expression}}
3828 * SubSection:: @code{.subsection}
3829 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3833 * Tag:: @code{.tag @var{structname}}
3836 * Text:: @code{.text @var{subsection}}
3837 * Title:: @code{.title "@var{heading}"}
3839 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3842 * Uleb128:: @code{.uleb128 @var{expressions}}
3844 * Val:: @code{.val @var{addr}}
3848 * Version:: @code{.version "@var{string}"}
3849 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3850 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3851 * Weak:: @code{.weak @var{names}}
3854 * Word:: @code{.word @var{expressions}}
3855 * Deprecated:: Deprecated Directives
3859 @section @code{.abort}
3861 @cindex @code{abort} directive
3862 @cindex stopping the assembly
3863 This directive stops the assembly immediately. It is for
3864 compatibility with other assemblers. The original idea was that the
3865 assembly language source would be piped into the assembler. If the sender
3866 of the source quit, it could use this directive tells @command{@value{AS}} to
3867 quit also. One day @code{.abort} will not be supported.
3871 @section @code{.ABORT}
3873 @cindex @code{ABORT} directive
3874 When producing COFF output, @command{@value{AS}} accepts this directive as a
3875 synonym for @samp{.abort}.
3878 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
3884 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3886 @cindex padding the location counter
3887 @cindex @code{align} directive
3888 Pad the location counter (in the current subsection) to a particular storage
3889 boundary. The first expression (which must be absolute) is the alignment
3890 required, as described below.
3892 The second expression (also absolute) gives the fill value to be stored in the
3893 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3894 padding bytes are normally zero. However, on some systems, if the section is
3895 marked as containing code and the fill value is omitted, the space is filled
3896 with no-op instructions.
3898 The third expression is also absolute, and is also optional. If it is present,
3899 it is the maximum number of bytes that should be skipped by this alignment
3900 directive. If doing the alignment would require skipping more bytes than the
3901 specified maximum, then the alignment is not done at all. You can omit the
3902 fill value (the second argument) entirely by simply using two commas after the
3903 required alignment; this can be useful if you want the alignment to be filled
3904 with no-op instructions when appropriate.
3906 The way the required alignment is specified varies from system to system.
3907 For the a29k, arc, hppa, i386 using ELF, i860, iq2000, m68k, m88k, or32,
3908 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
3909 alignment request in bytes. For example @samp{.align 8} advances
3910 the location counter until it is a multiple of 8. If the location counter
3911 is already a multiple of 8, no change is needed. For the tic54x, the
3912 first expression is the alignment request in words.
3914 For other systems, including the i386 using a.out format, and the arm and
3915 strongarm, it is the
3916 number of low-order zero bits the location counter must have after
3917 advancement. For example @samp{.align 3} advances the location
3918 counter until it a multiple of 8. If the location counter is already a
3919 multiple of 8, no change is needed.
3921 This inconsistency is due to the different behaviors of the various
3922 native assemblers for these systems which GAS must emulate.
3923 GAS also provides @code{.balign} and @code{.p2align} directives,
3924 described later, which have a consistent behavior across all
3925 architectures (but are specific to GAS).
3928 @section @code{.ascii "@var{string}"}@dots{}
3930 @cindex @code{ascii} directive
3931 @cindex string literals
3932 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3933 separated by commas. It assembles each string (with no automatic
3934 trailing zero byte) into consecutive addresses.
3937 @section @code{.asciz "@var{string}"}@dots{}
3939 @cindex @code{asciz} directive
3940 @cindex zero-terminated strings
3941 @cindex null-terminated strings
3942 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3943 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3946 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3948 @cindex padding the location counter given number of bytes
3949 @cindex @code{balign} directive
3950 Pad the location counter (in the current subsection) to a particular
3951 storage boundary. The first expression (which must be absolute) is the
3952 alignment request in bytes. For example @samp{.balign 8} advances
3953 the location counter until it is a multiple of 8. If the location counter
3954 is already a multiple of 8, no change is needed.
3956 The second expression (also absolute) gives the fill value to be stored in the
3957 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3958 padding bytes are normally zero. However, on some systems, if the section is
3959 marked as containing code and the fill value is omitted, the space is filled
3960 with no-op instructions.
3962 The third expression is also absolute, and is also optional. If it is present,
3963 it is the maximum number of bytes that should be skipped by this alignment
3964 directive. If doing the alignment would require skipping more bytes than the
3965 specified maximum, then the alignment is not done at all. You can omit the
3966 fill value (the second argument) entirely by simply using two commas after the
3967 required alignment; this can be useful if you want the alignment to be filled
3968 with no-op instructions when appropriate.
3970 @cindex @code{balignw} directive
3971 @cindex @code{balignl} directive
3972 The @code{.balignw} and @code{.balignl} directives are variants of the
3973 @code{.balign} directive. The @code{.balignw} directive treats the fill
3974 pattern as a two byte word value. The @code{.balignl} directives treats the
3975 fill pattern as a four byte longword value. For example, @code{.balignw
3976 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3977 filled in with the value 0x368d (the exact placement of the bytes depends upon
3978 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3982 @section @code{.byte @var{expressions}}
3984 @cindex @code{byte} directive
3985 @cindex integers, one byte
3986 @code{.byte} expects zero or more expressions, separated by commas.
3987 Each expression is assembled into the next byte.
3990 @section @code{.comm @var{symbol} , @var{length} }
3992 @cindex @code{comm} directive
3993 @cindex symbol, common
3994 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
3995 common symbol in one object file may be merged with a defined or common symbol
3996 of the same name in another object file. If @code{@value{LD}} does not see a
3997 definition for the symbol--just one or more common symbols--then it will
3998 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
3999 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4000 the same name, and they do not all have the same size, it will allocate space
4001 using the largest size.
4004 When using ELF, the @code{.comm} directive takes an optional third argument.
4005 This is the desired alignment of the symbol, specified as a byte boundary (for
4006 example, an alignment of 16 means that the least significant 4 bits of the
4007 address should be zero). The alignment must be an absolute expression, and it
4008 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
4009 for the common symbol, it will use the alignment when placing the symbol. If
4010 no alignment is specified, @command{@value{AS}} will set the alignment to the
4011 largest power of two less than or equal to the size of the symbol, up to a
4016 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4017 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4020 @node CFI directives
4021 @section @code{.cfi_startproc}
4022 @cindex @code{cfi_startproc} directive
4023 @code{.cfi_startproc} is used at the beginning of each function that
4024 should have an entry in @code{.eh_frame}. It initializes some internal
4025 data structures and emits architecture dependent initial CFI instructions.
4026 Don't forget to close the function by
4027 @code{.cfi_endproc}.
4029 @section @code{.cfi_endproc}
4030 @cindex @code{cfi_endproc} directive
4031 @code{.cfi_endproc} is used at the end of a function where it closes its
4032 unwind entry previously opened by
4033 @code{.cfi_startproc}. and emits it to @code{.eh_frame}.
4035 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4036 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4037 address from @var{register} and add @var{offset} to it}.
4039 @section @code{.cfi_def_cfa_register @var{register}}
4040 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4041 now on @var{register} will be used instead of the old one. Offset
4044 @section @code{.cfi_def_cfa_offset @var{offset}}
4045 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4046 remains the same, but @var{offset} is new. Note that it is the
4047 absolute offset that will be added to a defined register to compute
4050 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4051 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4052 value that is added/substracted from the previous offset.
4054 @section @code{.cfi_offset @var{register}, @var{offset}}
4055 Previous value of @var{register} is saved at offset @var{offset} from
4058 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4059 Previous value of @var{register} is saved at offset @var{offset} from
4060 the current CFA register. This is transformed to @code{.cfi_offset}
4061 using the known displacement of the CFA register from the CFA.
4062 This is often easier to use, because the number will match the
4063 code it's annotating.
4065 @section @code{.cfi_window_save}
4066 SPARC register window has been saved.
4068 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4069 Allows the user to add arbitrary bytes to the unwind info. One
4070 might use this to add OS-specific CFI opcodes, or generic CFI
4071 opcodes that GAS does not yet support.
4074 @section @code{.data @var{subsection}}
4076 @cindex @code{data} directive
4077 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4078 end of the data subsection numbered @var{subsection} (which is an
4079 absolute expression). If @var{subsection} is omitted, it defaults
4084 @section @code{.def @var{name}}
4086 @cindex @code{def} directive
4087 @cindex COFF symbols, debugging
4088 @cindex debugging COFF symbols
4089 Begin defining debugging information for a symbol @var{name}; the
4090 definition extends until the @code{.endef} directive is encountered.
4093 This directive is only observed when @command{@value{AS}} is configured for COFF
4094 format output; when producing @code{b.out}, @samp{.def} is recognized,
4101 @section @code{.desc @var{symbol}, @var{abs-expression}}
4103 @cindex @code{desc} directive
4104 @cindex COFF symbol descriptor
4105 @cindex symbol descriptor, COFF
4106 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4107 to the low 16 bits of an absolute expression.
4110 The @samp{.desc} directive is not available when @command{@value{AS}} is
4111 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4112 object format. For the sake of compatibility, @command{@value{AS}} accepts
4113 it, but produces no output, when configured for COFF.
4119 @section @code{.dim}
4121 @cindex @code{dim} directive
4122 @cindex COFF auxiliary symbol information
4123 @cindex auxiliary symbol information, COFF
4124 This directive is generated by compilers to include auxiliary debugging
4125 information in the symbol table. It is only permitted inside
4126 @code{.def}/@code{.endef} pairs.
4129 @samp{.dim} is only meaningful when generating COFF format output; when
4130 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4136 @section @code{.double @var{flonums}}
4138 @cindex @code{double} directive
4139 @cindex floating point numbers (double)
4140 @code{.double} expects zero or more flonums, separated by commas. It
4141 assembles floating point numbers.
4143 The exact kind of floating point numbers emitted depends on how
4144 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4148 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4149 in @sc{ieee} format.
4154 @section @code{.eject}
4156 @cindex @code{eject} directive
4157 @cindex new page, in listings
4158 @cindex page, in listings
4159 @cindex listing control: new page
4160 Force a page break at this point, when generating assembly listings.
4163 @section @code{.else}
4165 @cindex @code{else} directive
4166 @code{.else} is part of the @command{@value{AS}} support for conditional
4167 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
4168 of code to be assembled if the condition for the preceding @code{.if}
4172 @section @code{.elseif}
4174 @cindex @code{elseif} directive
4175 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4176 assembly; @pxref{If,,@code{.if}}. It is shorthand for beginning a new
4177 @code{.if} block that would otherwise fill the entire @code{.else} section.
4180 @section @code{.end}
4182 @cindex @code{end} directive
4183 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4184 process anything in the file past the @code{.end} directive.
4188 @section @code{.endef}
4190 @cindex @code{endef} directive
4191 This directive flags the end of a symbol definition begun with
4195 @samp{.endef} is only meaningful when generating COFF format output; if
4196 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4197 directive but ignores it.
4202 @section @code{.endfunc}
4203 @cindex @code{endfunc} directive
4204 @code{.endfunc} marks the end of a function specified with @code{.func}.
4207 @section @code{.endif}
4209 @cindex @code{endif} directive
4210 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4211 it marks the end of a block of code that is only assembled
4212 conditionally. @xref{If,,@code{.if}}.
4215 @section @code{.equ @var{symbol}, @var{expression}}
4217 @cindex @code{equ} directive
4218 @cindex assigning values to symbols
4219 @cindex symbols, assigning values to
4220 This directive sets the value of @var{symbol} to @var{expression}.
4221 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
4224 The syntax for @code{equ} on the HPPA is
4225 @samp{@var{symbol} .equ @var{expression}}.
4229 @section @code{.equiv @var{symbol}, @var{expression}}
4230 @cindex @code{equiv} directive
4231 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4232 the assembler will signal an error if @var{symbol} is already defined. Note a
4233 symbol which has been referenced but not actually defined is considered to be
4236 Except for the contents of the error message, this is roughly equivalent to
4245 @section @code{.err}
4246 @cindex @code{err} directive
4247 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4248 message and, unless the @option{-Z} option was used, it will not generate an
4249 object file. This can be used to signal error an conditionally compiled code.
4252 @section @code{.exitm}
4253 Exit early from the current macro definition. @xref{Macro}.
4256 @section @code{.extern}
4258 @cindex @code{extern} directive
4259 @code{.extern} is accepted in the source program---for compatibility
4260 with other assemblers---but it is ignored. @command{@value{AS}} treats
4261 all undefined symbols as external.
4264 @section @code{.fail @var{expression}}
4266 @cindex @code{fail} directive
4267 Generates an error or a warning. If the value of the @var{expression} is 500
4268 or more, @command{@value{AS}} will print a warning message. If the value is less
4269 than 500, @command{@value{AS}} will print an error message. The message will
4270 include the value of @var{expression}. This can occasionally be useful inside
4271 complex nested macros or conditional assembly.
4273 @ifclear no-file-dir
4275 @section @code{.file @var{string}}
4277 @cindex @code{file} directive
4278 @cindex logical file name
4279 @cindex file name, logical
4280 @code{.file} tells @command{@value{AS}} that we are about to start a new logical
4281 file. @var{string} is the new file name. In general, the filename is
4282 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4283 to specify an empty file name, you must give the quotes--@code{""}. This
4284 statement may go away in future: it is only recognized to be compatible with
4285 old @command{@value{AS}} programs.
4287 In some configurations of @command{@value{AS}}, @code{.file} has already been
4288 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
4293 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4295 @cindex @code{fill} directive
4296 @cindex writing patterns in memory
4297 @cindex patterns, writing in memory
4298 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4299 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4300 may be zero or more. @var{Size} may be zero or more, but if it is
4301 more than 8, then it is deemed to have the value 8, compatible with
4302 other people's assemblers. The contents of each @var{repeat} bytes
4303 is taken from an 8-byte number. The highest order 4 bytes are
4304 zero. The lowest order 4 bytes are @var{value} rendered in the
4305 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4306 Each @var{size} bytes in a repetition is taken from the lowest order
4307 @var{size} bytes of this number. Again, this bizarre behavior is
4308 compatible with other people's assemblers.
4310 @var{size} and @var{value} are optional.
4311 If the second comma and @var{value} are absent, @var{value} is
4312 assumed zero. If the first comma and following tokens are absent,
4313 @var{size} is assumed to be 1.
4316 @section @code{.float @var{flonums}}
4318 @cindex floating point numbers (single)
4319 @cindex @code{float} directive
4320 This directive assembles zero or more flonums, separated by commas. It
4321 has the same effect as @code{.single}.
4323 The exact kind of floating point numbers emitted depends on how
4324 @command{@value{AS}} is configured.
4325 @xref{Machine Dependencies}.
4329 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4330 in @sc{ieee} format.
4335 @section @code{.func @var{name}[,@var{label}]}
4336 @cindex @code{func} directive
4337 @code{.func} emits debugging information to denote function @var{name}, and
4338 is ignored unless the file is assembled with debugging enabled.
4339 Only @samp{--gstabs[+]} is currently supported.
4340 @var{label} is the entry point of the function and if omitted @var{name}
4341 prepended with the @samp{leading char} is used.
4342 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4343 All functions are currently defined to have @code{void} return type.
4344 The function must be terminated with @code{.endfunc}.
4347 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4349 @cindex @code{global} directive
4350 @cindex symbol, making visible to linker
4351 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4352 @var{symbol} in your partial program, its value is made available to
4353 other partial programs that are linked with it. Otherwise,
4354 @var{symbol} takes its attributes from a symbol of the same name
4355 from another file linked into the same program.
4357 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4358 compatibility with other assemblers.
4361 On the HPPA, @code{.global} is not always enough to make it accessible to other
4362 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4363 @xref{HPPA Directives,, HPPA Assembler Directives}.
4368 @section @code{.hidden @var{names}}
4370 @cindex @code{hidden} directive
4372 This one of the ELF visibility directives. The other two are
4373 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4374 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4376 This directive overrides the named symbols default visibility (which is set by
4377 their binding: local, global or weak). The directive sets the visibility to
4378 @code{hidden} which means that the symbols are not visible to other components.
4379 Such symbols are always considered to be @code{protected} as well.
4383 @section @code{.hword @var{expressions}}
4385 @cindex @code{hword} directive
4386 @cindex integers, 16-bit
4387 @cindex numbers, 16-bit
4388 @cindex sixteen bit integers
4389 This expects zero or more @var{expressions}, and emits
4390 a 16 bit number for each.
4393 This directive is a synonym for @samp{.short}; depending on the target
4394 architecture, it may also be a synonym for @samp{.word}.
4398 This directive is a synonym for @samp{.short}.
4401 This directive is a synonym for both @samp{.short} and @samp{.word}.
4406 @section @code{.ident}
4408 @cindex @code{ident} directive
4409 This directive is used by some assemblers to place tags in object files.
4410 @command{@value{AS}} simply accepts the directive for source-file
4411 compatibility with such assemblers, but does not actually emit anything
4415 @section @code{.if @var{absolute expression}}
4417 @cindex conditional assembly
4418 @cindex @code{if} directive
4419 @code{.if} marks the beginning of a section of code which is only
4420 considered part of the source program being assembled if the argument
4421 (which must be an @var{absolute expression}) is non-zero. The end of
4422 the conditional section of code must be marked by @code{.endif}
4423 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4424 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4425 If you have several conditions to check, @code{.elseif} may be used to avoid
4426 nesting blocks if/else within each subsequent @code{.else} block.
4428 The following variants of @code{.if} are also supported:
4430 @cindex @code{ifdef} directive
4431 @item .ifdef @var{symbol}
4432 Assembles the following section of code if the specified @var{symbol}
4433 has been defined. Note a symbol which has been referenced but not yet defined
4434 is considered to be undefined.
4436 @cindex @code{ifc} directive
4437 @item .ifc @var{string1},@var{string2}
4438 Assembles the following section of code if the two strings are the same. The
4439 strings may be optionally quoted with single quotes. If they are not quoted,
4440 the first string stops at the first comma, and the second string stops at the
4441 end of the line. Strings which contain whitespace should be quoted. The
4442 string comparison is case sensitive.
4444 @cindex @code{ifeq} directive
4445 @item .ifeq @var{absolute expression}
4446 Assembles the following section of code if the argument is zero.
4448 @cindex @code{ifeqs} directive
4449 @item .ifeqs @var{string1},@var{string2}
4450 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4452 @cindex @code{ifge} directive
4453 @item .ifge @var{absolute expression}
4454 Assembles the following section of code if the argument is greater than or
4457 @cindex @code{ifgt} directive
4458 @item .ifgt @var{absolute expression}
4459 Assembles the following section of code if the argument is greater than zero.
4461 @cindex @code{ifle} directive
4462 @item .ifle @var{absolute expression}
4463 Assembles the following section of code if the argument is less than or equal
4466 @cindex @code{iflt} directive
4467 @item .iflt @var{absolute expression}
4468 Assembles the following section of code if the argument is less than zero.
4470 @cindex @code{ifnc} directive
4471 @item .ifnc @var{string1},@var{string2}.
4472 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4473 following section of code if the two strings are not the same.
4475 @cindex @code{ifndef} directive
4476 @cindex @code{ifnotdef} directive
4477 @item .ifndef @var{symbol}
4478 @itemx .ifnotdef @var{symbol}
4479 Assembles the following section of code if the specified @var{symbol}
4480 has not been defined. Both spelling variants are equivalent. Note a symbol
4481 which has been referenced but not yet defined is considered to be undefined.
4483 @cindex @code{ifne} directive
4484 @item .ifne @var{absolute expression}
4485 Assembles the following section of code if the argument is not equal to zero
4486 (in other words, this is equivalent to @code{.if}).
4488 @cindex @code{ifnes} directive
4489 @item .ifnes @var{string1},@var{string2}
4490 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4491 following section of code if the two strings are not the same.
4495 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4497 @cindex @code{incbin} directive
4498 @cindex binary files, including
4499 The @code{incbin} directive includes @var{file} verbatim at the current
4500 location. You can control the search paths used with the @samp{-I} command-line
4501 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4504 The @var{skip} argument skips a number of bytes from the start of the
4505 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4506 read. Note that the data is not aligned in any way, so it is the user's
4507 responsibility to make sure that proper alignment is provided both before and
4508 after the @code{incbin} directive.
4511 @section @code{.include "@var{file}"}
4513 @cindex @code{include} directive
4514 @cindex supporting files, including
4515 @cindex files, including
4516 This directive provides a way to include supporting files at specified
4517 points in your source program. The code from @var{file} is assembled as
4518 if it followed the point of the @code{.include}; when the end of the
4519 included file is reached, assembly of the original file continues. You
4520 can control the search paths used with the @samp{-I} command-line option
4521 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4525 @section @code{.int @var{expressions}}
4527 @cindex @code{int} directive
4528 @cindex integers, 32-bit
4529 Expect zero or more @var{expressions}, of any section, separated by commas.
4530 For each expression, emit a number that, at run time, is the value of that
4531 expression. The byte order and bit size of the number depends on what kind
4532 of target the assembly is for.
4536 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
4537 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4544 @section @code{.internal @var{names}}
4546 @cindex @code{internal} directive
4548 This one of the ELF visibility directives. The other two are
4549 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4550 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4552 This directive overrides the named symbols default visibility (which is set by
4553 their binding: local, global or weak). The directive sets the visibility to
4554 @code{internal} which means that the symbols are considered to be @code{hidden}
4555 (i.e., not visible to other components), and that some extra, processor specific
4556 processing must also be performed upon the symbols as well.
4560 @section @code{.irp @var{symbol},@var{values}}@dots{}
4562 @cindex @code{irp} directive
4563 Evaluate a sequence of statements assigning different values to @var{symbol}.
4564 The sequence of statements starts at the @code{.irp} directive, and is
4565 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4566 set to @var{value}, and the sequence of statements is assembled. If no
4567 @var{value} is listed, the sequence of statements is assembled once, with
4568 @var{symbol} set to the null string. To refer to @var{symbol} within the
4569 sequence of statements, use @var{\symbol}.
4571 For example, assembling
4579 is equivalent to assembling
4588 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4590 @cindex @code{irpc} directive
4591 Evaluate a sequence of statements assigning different values to @var{symbol}.
4592 The sequence of statements starts at the @code{.irpc} directive, and is
4593 terminated by an @code{.endr} directive. For each character in @var{value},
4594 @var{symbol} is set to the character, and the sequence of statements is
4595 assembled. If no @var{value} is listed, the sequence of statements is
4596 assembled once, with @var{symbol} set to the null string. To refer to
4597 @var{symbol} within the sequence of statements, use @var{\symbol}.
4599 For example, assembling
4607 is equivalent to assembling
4616 @section @code{.lcomm @var{symbol} , @var{length}}
4618 @cindex @code{lcomm} directive
4619 @cindex local common symbols
4620 @cindex symbols, local common
4621 Reserve @var{length} (an absolute expression) bytes for a local common
4622 denoted by @var{symbol}. The section and value of @var{symbol} are
4623 those of the new local common. The addresses are allocated in the bss
4624 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4625 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4626 not visible to @code{@value{LD}}.
4629 Some targets permit a third argument to be used with @code{.lcomm}. This
4630 argument specifies the desired alignment of the symbol in the bss section.
4634 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4635 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4639 @section @code{.lflags}
4641 @cindex @code{lflags} directive (ignored)
4642 @command{@value{AS}} accepts this directive, for compatibility with other
4643 assemblers, but ignores it.
4645 @ifclear no-line-dir
4647 @section @code{.line @var{line-number}}
4649 @cindex @code{line} directive
4653 @section @code{.ln @var{line-number}}
4655 @cindex @code{ln} directive
4657 @cindex logical line number
4659 Change the logical line number. @var{line-number} must be an absolute
4660 expression. The next line has that logical line number. Therefore any other
4661 statements on the current line (after a statement separator character) are
4662 reported as on logical line number @var{line-number} @minus{} 1. One day
4663 @command{@value{AS}} will no longer support this directive: it is recognized only
4664 for compatibility with existing assembler programs.
4668 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
4669 not available; use the synonym @code{.ln} in that context.
4674 @ifclear no-line-dir
4675 Even though this is a directive associated with the @code{a.out} or
4676 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
4677 when producing COFF output, and treats @samp{.line} as though it
4678 were the COFF @samp{.ln} @emph{if} it is found outside a
4679 @code{.def}/@code{.endef} pair.
4681 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4682 used by compilers to generate auxiliary symbol information for
4687 @section @code{.linkonce [@var{type}]}
4689 @cindex @code{linkonce} directive
4690 @cindex common sections
4691 Mark the current section so that the linker only includes a single copy of it.
4692 This may be used to include the same section in several different object files,
4693 but ensure that the linker will only include it once in the final output file.
4694 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4695 Duplicate sections are detected based on the section name, so it should be
4698 This directive is only supported by a few object file formats; as of this
4699 writing, the only object file format which supports it is the Portable
4700 Executable format used on Windows NT.
4702 The @var{type} argument is optional. If specified, it must be one of the
4703 following strings. For example:
4707 Not all types may be supported on all object file formats.
4711 Silently discard duplicate sections. This is the default.
4714 Warn if there are duplicate sections, but still keep only one copy.
4717 Warn if any of the duplicates have different sizes.
4720 Warn if any of the duplicates do not have exactly the same contents.
4724 @section @code{.ln @var{line-number}}
4726 @cindex @code{ln} directive
4727 @ifclear no-line-dir
4728 @samp{.ln} is a synonym for @samp{.line}.
4731 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
4732 must be an absolute expression. The next line has that logical
4733 line number, so any other statements on the current line (after a
4734 statement separator character @code{;}) are reported as on logical
4735 line number @var{line-number} @minus{} 1.
4738 This directive is accepted, but ignored, when @command{@value{AS}} is
4739 configured for @code{b.out}; its effect is only associated with COFF
4745 @section @code{.mri @var{val}}
4747 @cindex @code{mri} directive
4748 @cindex MRI mode, temporarily
4749 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
4750 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
4751 affects code assembled until the next @code{.mri} directive, or until the end
4752 of the file. @xref{M, MRI mode, MRI mode}.
4755 @section @code{.list}
4757 @cindex @code{list} directive
4758 @cindex listing control, turning on
4759 Control (in conjunction with the @code{.nolist} directive) whether or
4760 not assembly listings are generated. These two directives maintain an
4761 internal counter (which is zero initially). @code{.list} increments the
4762 counter, and @code{.nolist} decrements it. Assembly listings are
4763 generated whenever the counter is greater than zero.
4765 By default, listings are disabled. When you enable them (with the
4766 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4767 the initial value of the listing counter is one.
4770 @section @code{.long @var{expressions}}
4772 @cindex @code{long} directive
4773 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
4776 @c no one seems to know what this is for or whether this description is
4777 @c what it really ought to do
4779 @section @code{.lsym @var{symbol}, @var{expression}}
4781 @cindex @code{lsym} directive
4782 @cindex symbol, not referenced in assembly
4783 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
4784 the hash table, ensuring it cannot be referenced by name during the
4785 rest of the assembly. This sets the attributes of the symbol to be
4786 the same as the expression value:
4788 @var{other} = @var{descriptor} = 0
4789 @var{type} = @r{(section of @var{expression})}
4790 @var{value} = @var{expression}
4793 The new symbol is not flagged as external.
4797 @section @code{.macro}
4800 The commands @code{.macro} and @code{.endm} allow you to define macros that
4801 generate assembly output. For example, this definition specifies a macro
4802 @code{sum} that puts a sequence of numbers into memory:
4805 .macro sum from=0, to=5
4814 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
4826 @item .macro @var{macname}
4827 @itemx .macro @var{macname} @var{macargs} @dots{}
4828 @cindex @code{macro} directive
4829 Begin the definition of a macro called @var{macname}. If your macro
4830 definition requires arguments, specify their names after the macro name,
4831 separated by commas or spaces. You can supply a default value for any
4832 macro argument by following the name with @samp{=@var{deflt}}. For
4833 example, these are all valid @code{.macro} statements:
4837 Begin the definition of a macro called @code{comm}, which takes no
4840 @item .macro plus1 p, p1
4841 @itemx .macro plus1 p p1
4842 Either statement begins the definition of a macro called @code{plus1},
4843 which takes two arguments; within the macro definition, write
4844 @samp{\p} or @samp{\p1} to evaluate the arguments.
4846 @item .macro reserve_str p1=0 p2
4847 Begin the definition of a macro called @code{reserve_str}, with two
4848 arguments. The first argument has a default value, but not the second.
4849 After the definition is complete, you can call the macro either as
4850 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
4851 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
4852 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
4853 @samp{0}, and @samp{\p2} evaluating to @var{b}).
4856 When you call a macro, you can specify the argument values either by
4857 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
4858 @samp{sum to=17, from=9}.
4861 @cindex @code{endm} directive
4862 Mark the end of a macro definition.
4865 @cindex @code{exitm} directive
4866 Exit early from the current macro definition.
4868 @cindex number of macros executed
4869 @cindex macros, count executed
4871 @command{@value{AS}} maintains a counter of how many macros it has
4872 executed in this pseudo-variable; you can copy that number to your
4873 output with @samp{\@@}, but @emph{only within a macro definition}.
4876 @item LOCAL @var{name} [ , @dots{} ]
4877 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
4878 macro syntax'' with @samp{-a} or @samp{--alternate}.} @xref{Alternate,,
4879 Alternate macro syntax}.
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.
4890 @section @code{.nolist}
4892 @cindex @code{nolist} directive
4893 @cindex listing control, turning off
4894 Control (in conjunction with the @code{.list} directive) whether or
4895 not assembly listings are generated. These two directives maintain an
4896 internal counter (which is zero initially). @code{.list} increments the
4897 counter, and @code{.nolist} decrements it. Assembly listings are
4898 generated whenever the counter is greater than zero.
4901 @section @code{.octa @var{bignums}}
4903 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
4904 @cindex @code{octa} directive
4905 @cindex integer, 16-byte
4906 @cindex sixteen byte integer
4907 This directive expects zero or more bignums, separated by commas. For each
4908 bignum, it emits a 16-byte integer.
4910 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
4911 hence @emph{octa}-word for 16 bytes.
4914 @section @code{.org @var{new-lc} , @var{fill}}
4916 @cindex @code{org} directive
4917 @cindex location counter, advancing
4918 @cindex advancing location counter
4919 @cindex current address, advancing
4920 Advance the location counter of the current section to
4921 @var{new-lc}. @var{new-lc} is either an absolute expression or an
4922 expression with the same section as the current subsection. That is,
4923 you can't use @code{.org} to cross sections: if @var{new-lc} has the
4924 wrong section, the @code{.org} directive is ignored. To be compatible
4925 with former assemblers, if the section of @var{new-lc} is absolute,
4926 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
4927 is the same as the current subsection.
4929 @code{.org} may only increase the location counter, or leave it
4930 unchanged; you cannot use @code{.org} to move the location counter
4933 @c double negative used below "not undefined" because this is a specific
4934 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
4935 @c section. doc@cygnus.com 18feb91
4936 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
4937 may not be undefined. If you really detest this restriction we eagerly await
4938 a chance to share your improved assembler.
4940 Beware that the origin is relative to the start of the section, not
4941 to the start of the subsection. This is compatible with other
4942 people's assemblers.
4944 When the location counter (of the current subsection) is advanced, the
4945 intervening bytes are filled with @var{fill} which should be an
4946 absolute expression. If the comma and @var{fill} are omitted,
4947 @var{fill} defaults to zero.
4950 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4952 @cindex padding the location counter given a power of two
4953 @cindex @code{p2align} directive
4954 Pad the location counter (in the current subsection) to a particular
4955 storage boundary. The first expression (which must be absolute) is the
4956 number of low-order zero bits the location counter must have after
4957 advancement. For example @samp{.p2align 3} advances the location
4958 counter until it a multiple of 8. If the location counter is already a
4959 multiple of 8, no change is needed.
4961 The second expression (also absolute) gives the fill value to be stored in the
4962 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4963 padding bytes are normally zero. However, on some systems, if the section is
4964 marked as containing code and the fill value is omitted, the space is filled
4965 with no-op instructions.
4967 The third expression is also absolute, and is also optional. If it is present,
4968 it is the maximum number of bytes that should be skipped by this alignment
4969 directive. If doing the alignment would require skipping more bytes than the
4970 specified maximum, then the alignment is not done at all. You can omit the
4971 fill value (the second argument) entirely by simply using two commas after the
4972 required alignment; this can be useful if you want the alignment to be filled
4973 with no-op instructions when appropriate.
4975 @cindex @code{p2alignw} directive
4976 @cindex @code{p2alignl} directive
4977 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
4978 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
4979 pattern as a two byte word value. The @code{.p2alignl} directives treats the
4980 fill pattern as a four byte longword value. For example, @code{.p2alignw
4981 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4982 filled in with the value 0x368d (the exact placement of the bytes depends upon
4983 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4988 @section @code{.previous}
4990 @cindex @code{previous} directive
4991 @cindex Section Stack
4992 This is one of the ELF section stack manipulation directives. The others are
4993 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
4994 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
4995 (@pxref{PopSection}).
4997 This directive swaps the current section (and subsection) with most recently
4998 referenced section (and subsection) prior to this one. Multiple
4999 @code{.previous} directives in a row will flip between two sections (and their
5002 In terms of the section stack, this directive swaps the current section with
5003 the top section on the section stack.
5008 @section @code{.popsection}
5010 @cindex @code{popsection} directive
5011 @cindex Section Stack
5012 This is one of the ELF section stack manipulation directives. The others are
5013 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5014 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5017 This directive replaces the current section (and subsection) with the top
5018 section (and subsection) on the section stack. This section is popped off the
5023 @section @code{.print @var{string}}
5025 @cindex @code{print} directive
5026 @command{@value{AS}} will print @var{string} on the standard output during
5027 assembly. You must put @var{string} in double quotes.
5031 @section @code{.protected @var{names}}
5033 @cindex @code{protected} directive
5035 This one of the ELF visibility directives. The other two are
5036 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5038 This directive overrides the named symbols default visibility (which is set by
5039 their binding: local, global or weak). The directive sets the visibility to
5040 @code{protected} which means that any references to the symbols from within the
5041 components that defines them must be resolved to the definition in that
5042 component, even if a definition in another component would normally preempt
5047 @section @code{.psize @var{lines} , @var{columns}}
5049 @cindex @code{psize} directive
5050 @cindex listing control: paper size
5051 @cindex paper size, for listings
5052 Use this directive to declare the number of lines---and, optionally, the
5053 number of columns---to use for each page, when generating listings.
5055 If you do not use @code{.psize}, listings use a default line-count
5056 of 60. You may omit the comma and @var{columns} specification; the
5057 default width is 200 columns.
5059 @command{@value{AS}} generates formfeeds whenever the specified number of
5060 lines is exceeded (or whenever you explicitly request one, using
5063 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5064 those explicitly specified with @code{.eject}.
5067 @section @code{.purgem @var{name}}
5069 @cindex @code{purgem} directive
5070 Undefine the macro @var{name}, so that later uses of the string will not be
5071 expanded. @xref{Macro}.
5075 @section @code{.pushsection @var{name} , @var{subsection}}
5077 @cindex @code{pushsection} directive
5078 @cindex Section Stack
5079 This is one of the ELF section stack manipulation directives. The others are
5080 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5081 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5084 This directive is a synonym for @code{.section}. It pushes the current section
5085 (and subsection) onto the top of the section stack, and then replaces the
5086 current section and subsection with @code{name} and @code{subsection}.
5090 @section @code{.quad @var{bignums}}
5092 @cindex @code{quad} directive
5093 @code{.quad} expects zero or more bignums, separated by commas. For
5094 each bignum, it emits
5096 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5097 warning message; and just takes the lowest order 8 bytes of the bignum.
5098 @cindex eight-byte integer
5099 @cindex integer, 8-byte
5101 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5102 hence @emph{quad}-word for 8 bytes.
5105 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5106 warning message; and just takes the lowest order 16 bytes of the bignum.
5107 @cindex sixteen-byte integer
5108 @cindex integer, 16-byte
5112 @section @code{.rept @var{count}}
5114 @cindex @code{rept} directive
5115 Repeat the sequence of lines between the @code{.rept} directive and the next
5116 @code{.endr} directive @var{count} times.
5118 For example, assembling
5126 is equivalent to assembling
5135 @section @code{.sbttl "@var{subheading}"}
5137 @cindex @code{sbttl} directive
5138 @cindex subtitles for listings
5139 @cindex listing control: subtitle
5140 Use @var{subheading} as the title (third line, immediately after the
5141 title line) when generating assembly listings.
5143 This directive affects subsequent pages, as well as the current page if
5144 it appears within ten lines of the top of a page.
5148 @section @code{.scl @var{class}}
5150 @cindex @code{scl} directive
5151 @cindex symbol storage class (COFF)
5152 @cindex COFF symbol storage class
5153 Set the storage-class value for a symbol. This directive may only be
5154 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5155 whether a symbol is static or external, or it may record further
5156 symbolic debugging information.
5159 The @samp{.scl} directive is primarily associated with COFF output; when
5160 configured to generate @code{b.out} output format, @command{@value{AS}}
5161 accepts this directive but ignores it.
5167 @section @code{.section @var{name}}
5169 @cindex named section
5170 Use the @code{.section} directive to assemble the following code into a section
5173 This directive is only supported for targets that actually support arbitrarily
5174 named sections; on @code{a.out} targets, for example, it is not accepted, even
5175 with a standard @code{a.out} section name.
5179 @c only print the extra heading if both COFF and ELF are set
5180 @subheading COFF Version
5183 @cindex @code{section} directive (COFF version)
5184 For COFF targets, the @code{.section} directive is used in one of the following
5188 .section @var{name}[, "@var{flags}"]
5189 .section @var{name}[, @var{subsegment}]
5192 If the optional argument is quoted, it is taken as flags to use for the
5193 section. Each flag is a single character. The following flags are recognized:
5196 bss section (uninitialized data)
5198 section is not loaded
5208 shared section (meaningful for PE targets)
5210 ignored. (For compatibility with the ELF version)
5213 If no flags are specified, the default flags depend upon the section name. If
5214 the section name is not recognized, the default will be for the section to be
5215 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5216 from the section, rather than adding them, so if they are used on their own it
5217 will be as if no flags had been specified at all.
5219 If the optional argument to the @code{.section} directive is not quoted, it is
5220 taken as a subsegment number (@pxref{Sub-Sections}).
5225 @c only print the extra heading if both COFF and ELF are set
5226 @subheading ELF Version
5229 @cindex Section Stack
5230 This is one of the ELF section stack manipulation directives. The others are
5231 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5232 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5233 @code{.previous} (@pxref{Previous}).
5235 @cindex @code{section} directive (ELF version)
5236 For ELF targets, the @code{.section} directive is used like this:
5239 .section @var{name} [, "@var{flags}"[, @@@var{type}[, @@@var{entsize}]]]
5242 The optional @var{flags} argument is a quoted string which may contain any
5243 combination of the following characters:
5246 section is allocatable
5250 section is executable
5252 section is mergeable
5254 section contains zero terminated strings
5257 The optional @var{type} argument may contain one of the following constants:
5260 section contains data
5262 section does not contain data (i.e., section only occupies space)
5265 Note on targets where the @code{@@} character is the start of a comment (eg
5266 ARM) then another character is used instead. For example the ARM port uses the
5269 If @var{flags} contains @code{M} flag, @var{type} argument must be specified
5270 as well as @var{entsize} argument. Sections with @code{M} flag but not
5271 @code{S} flag must contain fixed size constants, each @var{entsize} octets
5272 long. Sections with both @code{M} and @code{S} must contain zero terminated
5273 strings where each character is @var{entsize} bytes long. The linker may remove
5274 duplicates within sections with the same name, same entity size and same flags.
5276 If no flags are specified, the default flags depend upon the section name. If
5277 the section name is not recognized, the default will be for the section to have
5278 none of the above flags: it will not be allocated in memory, nor writable, nor
5279 executable. The section will contain data.
5281 For ELF targets, the assembler supports another type of @code{.section}
5282 directive for compatibility with the Solaris assembler:
5285 .section "@var{name}"[, @var{flags}...]
5288 Note that the section name is quoted. There may be a sequence of comma
5292 section is allocatable
5296 section is executable
5299 This directive replaces the current section and subsection. The replaced
5300 section and subsection are pushed onto the section stack. See the contents of
5301 the gas testsuite directory @code{gas/testsuite/gas/elf} for some examples of
5302 how this directive and the other section stack directives work.
5307 @section @code{.set @var{symbol}, @var{expression}}
5309 @cindex @code{set} directive
5310 @cindex symbol value, setting
5311 Set the value of @var{symbol} to @var{expression}. This
5312 changes @var{symbol}'s value and type to conform to
5313 @var{expression}. If @var{symbol} was flagged as external, it remains
5314 flagged (@pxref{Symbol Attributes}).
5316 You may @code{.set} a symbol many times in the same assembly.
5318 If you @code{.set} a global symbol, the value stored in the object
5319 file is the last value stored into it.
5322 The syntax for @code{set} on the HPPA is
5323 @samp{@var{symbol} .set @var{expression}}.
5327 @section @code{.short @var{expressions}}
5329 @cindex @code{short} directive
5331 @code{.short} is normally the same as @samp{.word}.
5332 @xref{Word,,@code{.word}}.
5334 In some configurations, however, @code{.short} and @code{.word} generate
5335 numbers of different lengths; @pxref{Machine Dependencies}.
5339 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5342 This expects zero or more @var{expressions}, and emits
5343 a 16 bit number for each.
5348 @section @code{.single @var{flonums}}
5350 @cindex @code{single} directive
5351 @cindex floating point numbers (single)
5352 This directive assembles zero or more flonums, separated by commas. It
5353 has the same effect as @code{.float}.
5355 The exact kind of floating point numbers emitted depends on how
5356 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5360 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5361 numbers in @sc{ieee} format.
5367 @section @code{.size}
5369 This directive is used to set the size associated with a symbol.
5373 @c only print the extra heading if both COFF and ELF are set
5374 @subheading COFF Version
5377 @cindex @code{size} directive (COFF version)
5378 For COFF targets, the @code{.size} directive is only permitted inside
5379 @code{.def}/@code{.endef} pairs. It is used like this:
5382 .size @var{expression}
5386 @samp{.size} is only meaningful when generating COFF format output; when
5387 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5394 @c only print the extra heading if both COFF and ELF are set
5395 @subheading ELF Version
5398 @cindex @code{size} directive (ELF version)
5399 For ELF targets, the @code{.size} directive is used like this:
5402 .size @var{name} , @var{expression}
5405 This directive sets the size associated with a symbol @var{name}.
5406 The size in bytes is computed from @var{expression} which can make use of label
5407 arithmetic. This directive is typically used to set the size of function
5413 @section @code{.sleb128 @var{expressions}}
5415 @cindex @code{sleb128} directive
5416 @var{sleb128} stands for ``signed little endian base 128.'' This is a
5417 compact, variable length representation of numbers used by the DWARF
5418 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
5420 @ifclear no-space-dir
5422 @section @code{.skip @var{size} , @var{fill}}
5424 @cindex @code{skip} directive
5425 @cindex filling memory
5426 This directive emits @var{size} bytes, each of value @var{fill}. Both
5427 @var{size} and @var{fill} are absolute expressions. If the comma and
5428 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
5432 @section @code{.space @var{size} , @var{fill}}
5434 @cindex @code{space} directive
5435 @cindex filling memory
5436 This directive emits @var{size} bytes, each of value @var{fill}. Both
5437 @var{size} and @var{fill} are absolute expressions. If the comma
5438 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
5443 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
5444 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
5445 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
5446 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
5455 @section @code{.space}
5456 @cindex @code{space} directive
5458 On the AMD 29K, this directive is ignored; it is accepted for
5459 compatibility with other AMD 29K assemblers.
5462 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
5463 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
5469 @section @code{.stabd, .stabn, .stabs}
5471 @cindex symbolic debuggers, information for
5472 @cindex @code{stab@var{x}} directives
5473 There are three directives that begin @samp{.stab}.
5474 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
5475 The symbols are not entered in the @command{@value{AS}} hash table: they
5476 cannot be referenced elsewhere in the source file.
5477 Up to five fields are required:
5481 This is the symbol's name. It may contain any character except
5482 @samp{\000}, so is more general than ordinary symbol names. Some
5483 debuggers used to code arbitrarily complex structures into symbol names
5487 An absolute expression. The symbol's type is set to the low 8 bits of
5488 this expression. Any bit pattern is permitted, but @code{@value{LD}}
5489 and debuggers choke on silly bit patterns.
5492 An absolute expression. The symbol's ``other'' attribute is set to the
5493 low 8 bits of this expression.
5496 An absolute expression. The symbol's descriptor is set to the low 16
5497 bits of this expression.
5500 An absolute expression which becomes the symbol's value.
5503 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
5504 or @code{.stabs} statement, the symbol has probably already been created;
5505 you get a half-formed symbol in your object file. This is
5506 compatible with earlier assemblers!
5509 @cindex @code{stabd} directive
5510 @item .stabd @var{type} , @var{other} , @var{desc}
5512 The ``name'' of the symbol generated is not even an empty string.
5513 It is a null pointer, for compatibility. Older assemblers used a
5514 null pointer so they didn't waste space in object files with empty
5517 The symbol's value is set to the location counter,
5518 relocatably. When your program is linked, the value of this symbol
5519 is the address of the location counter when the @code{.stabd} was
5522 @cindex @code{stabn} directive
5523 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
5524 The name of the symbol is set to the empty string @code{""}.
5526 @cindex @code{stabs} directive
5527 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
5528 All five fields are specified.
5534 @section @code{.string} "@var{str}"
5536 @cindex string, copying to object file
5537 @cindex @code{string} directive
5539 Copy the characters in @var{str} to the object file. You may specify more than
5540 one string to copy, separated by commas. Unless otherwise specified for a
5541 particular machine, the assembler marks the end of each string with a 0 byte.
5542 You can use any of the escape sequences described in @ref{Strings,,Strings}.
5545 @section @code{.struct @var{expression}}
5547 @cindex @code{struct} directive
5548 Switch to the absolute section, and set the section offset to @var{expression},
5549 which must be an absolute expression. You might use this as follows:
5558 This would define the symbol @code{field1} to have the value 0, the symbol
5559 @code{field2} to have the value 4, and the symbol @code{field3} to have the
5560 value 8. Assembly would be left in the absolute section, and you would need to
5561 use a @code{.section} directive of some sort to change to some other section
5562 before further assembly.
5566 @section @code{.subsection @var{name}}
5568 @cindex @code{subsection} directive
5569 @cindex Section Stack
5570 This is one of the ELF section stack manipulation directives. The others are
5571 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
5572 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5575 This directive replaces the current subsection with @code{name}. The current
5576 section is not changed. The replaced subsection is put onto the section stack
5577 in place of the then current top of stack subsection.
5582 @section @code{.symver}
5583 @cindex @code{symver} directive
5584 @cindex symbol versioning
5585 @cindex versions of symbols
5586 Use the @code{.symver} directive to bind symbols to specific version nodes
5587 within a source file. This is only supported on ELF platforms, and is
5588 typically used when assembling files to be linked into a shared library.
5589 There are cases where it may make sense to use this in objects to be bound
5590 into an application itself so as to override a versioned symbol from a
5593 For ELF targets, the @code{.symver} directive can be used like this:
5595 .symver @var{name}, @var{name2@@nodename}
5597 If the symbol @var{name} is defined within the file
5598 being assembled, the @code{.symver} directive effectively creates a symbol
5599 alias with the name @var{name2@@nodename}, and in fact the main reason that we
5600 just don't try and create a regular alias is that the @var{@@} character isn't
5601 permitted in symbol names. The @var{name2} part of the name is the actual name
5602 of the symbol by which it will be externally referenced. The name @var{name}
5603 itself is merely a name of convenience that is used so that it is possible to
5604 have definitions for multiple versions of a function within a single source
5605 file, and so that the compiler can unambiguously know which version of a
5606 function is being mentioned. The @var{nodename} portion of the alias should be
5607 the name of a node specified in the version script supplied to the linker when
5608 building a shared library. If you are attempting to override a versioned
5609 symbol from a shared library, then @var{nodename} should correspond to the
5610 nodename of the symbol you are trying to override.
5612 If the symbol @var{name} is not defined within the file being assembled, all
5613 references to @var{name} will be changed to @var{name2@@nodename}. If no
5614 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
5617 Another usage of the @code{.symver} directive is:
5619 .symver @var{name}, @var{name2@@@@nodename}
5621 In this case, the symbol @var{name} must exist and be defined within
5622 the file being assembled. It is similar to @var{name2@@nodename}. The
5623 difference is @var{name2@@@@nodename} will also be used to resolve
5624 references to @var{name2} by the linker.
5626 The third usage of the @code{.symver} directive is:
5628 .symver @var{name}, @var{name2@@@@@@nodename}
5630 When @var{name} is not defined within the
5631 file being assembled, it is treated as @var{name2@@nodename}. When
5632 @var{name} is defined within the file being assembled, the symbol
5633 name, @var{name}, will be changed to @var{name2@@@@nodename}.
5638 @section @code{.tag @var{structname}}
5640 @cindex COFF structure debugging
5641 @cindex structure debugging, COFF
5642 @cindex @code{tag} directive
5643 This directive is generated by compilers to include auxiliary debugging
5644 information in the symbol table. It is only permitted inside
5645 @code{.def}/@code{.endef} pairs. Tags are used to link structure
5646 definitions in the symbol table with instances of those structures.
5649 @samp{.tag} is only used when generating COFF format output; when
5650 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5656 @section @code{.text @var{subsection}}
5658 @cindex @code{text} directive
5659 Tells @command{@value{AS}} to assemble the following statements onto the end of
5660 the text subsection numbered @var{subsection}, which is an absolute
5661 expression. If @var{subsection} is omitted, subsection number zero
5665 @section @code{.title "@var{heading}"}
5667 @cindex @code{title} directive
5668 @cindex listing control: title line
5669 Use @var{heading} as the title (second line, immediately after the
5670 source file name and pagenumber) when generating assembly listings.
5672 This directive affects subsequent pages, as well as the current page if
5673 it appears within ten lines of the top of a page.
5677 @section @code{.type}
5679 This directive is used to set the type of a symbol.
5683 @c only print the extra heading if both COFF and ELF are set
5684 @subheading COFF Version
5687 @cindex COFF symbol type
5688 @cindex symbol type, COFF
5689 @cindex @code{type} directive (COFF version)
5690 For COFF targets, this directive is permitted only within
5691 @code{.def}/@code{.endef} pairs. It is used like this:
5697 This records the integer @var{int} as the type attribute of a symbol table
5701 @samp{.type} is associated only with COFF format output; when
5702 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
5703 directive but ignores it.
5709 @c only print the extra heading if both COFF and ELF are set
5710 @subheading ELF Version
5713 @cindex ELF symbol type
5714 @cindex symbol type, ELF
5715 @cindex @code{type} directive (ELF version)
5716 For ELF targets, the @code{.type} directive is used like this:
5719 .type @var{name} , @var{type description}
5722 This sets the type of symbol @var{name} to be either a
5723 function symbol or an object symbol. There are five different syntaxes
5724 supported for the @var{type description} field, in order to provide
5725 compatibility with various other assemblers. The syntaxes supported are:
5728 .type <name>,#function
5729 .type <name>,#object
5731 .type <name>,@@function
5732 .type <name>,@@object
5734 .type <name>,%function
5735 .type <name>,%object
5737 .type <name>,"function"
5738 .type <name>,"object"
5740 .type <name> STT_FUNCTION
5741 .type <name> STT_OBJECT
5747 @section @code{.uleb128 @var{expressions}}
5749 @cindex @code{uleb128} directive
5750 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
5751 compact, variable length representation of numbers used by the DWARF
5752 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
5756 @section @code{.val @var{addr}}
5758 @cindex @code{val} directive
5759 @cindex COFF value attribute
5760 @cindex value attribute, COFF
5761 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5762 records the address @var{addr} as the value attribute of a symbol table
5766 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
5767 configured for @code{b.out}, it accepts this directive but ignores it.
5773 @section @code{.version "@var{string}"}
5775 @cindex @code{version} directive
5776 This directive creates a @code{.note} section and places into it an ELF
5777 formatted note of type NT_VERSION. The note's name is set to @code{string}.
5782 @section @code{.vtable_entry @var{table}, @var{offset}}
5784 @cindex @code{vtable_entry}
5785 This directive finds or creates a symbol @code{table} and creates a
5786 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
5789 @section @code{.vtable_inherit @var{child}, @var{parent}}
5791 @cindex @code{vtable_inherit}
5792 This directive finds the symbol @code{child} and finds or creates the symbol
5793 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
5794 parent whose addend is the value of the child symbol. As a special case the
5795 parent name of @code{0} is treated as refering the @code{*ABS*} section.
5800 @section @code{.weak @var{names}}
5802 @cindex @code{weak} directive
5803 This directive sets the weak attribute on the comma separated list of symbol
5804 @code{names}. If the symbols do not already exist, they will be created.
5808 @section @code{.word @var{expressions}}
5810 @cindex @code{word} directive
5811 This directive expects zero or more @var{expressions}, of any section,
5812 separated by commas.
5815 For each expression, @command{@value{AS}} emits a 32-bit number.
5818 For each expression, @command{@value{AS}} emits a 16-bit number.
5823 The size of the number emitted, and its byte order,
5824 depend on what target computer the assembly is for.
5827 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
5828 @c happen---32-bit addressability, period; no long/short jumps.
5829 @ifset DIFF-TBL-KLUGE
5830 @cindex difference tables altered
5831 @cindex altered difference tables
5833 @emph{Warning: Special Treatment to support Compilers}
5837 Machines with a 32-bit address space, but that do less than 32-bit
5838 addressing, require the following special treatment. If the machine of
5839 interest to you does 32-bit addressing (or doesn't require it;
5840 @pxref{Machine Dependencies}), you can ignore this issue.
5843 In order to assemble compiler output into something that works,
5844 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
5845 Directives of the form @samp{.word sym1-sym2} are often emitted by
5846 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
5847 directive of the form @samp{.word sym1-sym2}, and the difference between
5848 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
5849 creates a @dfn{secondary jump table}, immediately before the next label.
5850 This secondary jump table is preceded by a short-jump to the
5851 first byte after the secondary table. This short-jump prevents the flow
5852 of control from accidentally falling into the new table. Inside the
5853 table is a long-jump to @code{sym2}. The original @samp{.word}
5854 contains @code{sym1} minus the address of the long-jump to
5857 If there were several occurrences of @samp{.word sym1-sym2} before the
5858 secondary jump table, all of them are adjusted. If there was a
5859 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
5860 long-jump to @code{sym4} is included in the secondary jump table,
5861 and the @code{.word} directives are adjusted to contain @code{sym3}
5862 minus the address of the long-jump to @code{sym4}; and so on, for as many
5863 entries in the original jump table as necessary.
5866 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
5867 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
5868 assembly language programmers.
5871 @c end DIFF-TBL-KLUGE
5874 @section Deprecated Directives
5876 @cindex deprecated directives
5877 @cindex obsolescent directives
5878 One day these directives won't work.
5879 They are included for compatibility with older assemblers.
5886 @node Machine Dependencies
5887 @chapter Machine Dependent Features
5889 @cindex machine dependencies
5890 The machine instruction sets are (almost by definition) different on
5891 each machine where @command{@value{AS}} runs. Floating point representations
5892 vary as well, and @command{@value{AS}} often supports a few additional
5893 directives or command-line options for compatibility with other
5894 assemblers on a particular platform. Finally, some versions of
5895 @command{@value{AS}} support special pseudo-instructions for branch
5898 This chapter discusses most of these differences, though it does not
5899 include details on any machine's instruction set. For details on that
5900 subject, see the hardware manufacturer's manual.
5904 * AMD29K-Dependent:: AMD 29K Dependent Features
5907 * Alpha-Dependent:: Alpha Dependent Features
5910 * ARC-Dependent:: ARC Dependent Features
5913 * ARM-Dependent:: ARM Dependent Features
5916 * CRIS-Dependent:: CRIS Dependent Features
5919 * D10V-Dependent:: D10V Dependent Features
5922 * D30V-Dependent:: D30V Dependent Features
5925 * H8/300-Dependent:: Renesas H8/300 Dependent Features
5928 * H8/500-Dependent:: Renesas H8/500 Dependent Features
5931 * HPPA-Dependent:: HPPA Dependent Features
5934 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
5937 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
5940 * i860-Dependent:: Intel 80860 Dependent Features
5943 * i960-Dependent:: Intel 80960 Dependent Features
5946 * IP2K-Dependent:: IP2K Dependent Features
5949 * M32R-Dependent:: M32R Dependent Features
5952 * M68K-Dependent:: M680x0 Dependent Features
5955 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
5958 * M88K-Dependent:: M880x0 Dependent Features
5961 * MIPS-Dependent:: MIPS Dependent Features
5964 * MMIX-Dependent:: MMIX Dependent Features
5967 * MSP430-Dependent:: MSP430 Dependent Features
5970 * SH-Dependent:: Renesas / SuperH SH Dependent Features
5971 * SH64-Dependent:: SuperH SH64 Dependent Features
5974 * PDP-11-Dependent:: PDP-11 Dependent Features
5977 * PJ-Dependent:: picoJava Dependent Features
5980 * PPC-Dependent:: PowerPC Dependent Features
5983 * Sparc-Dependent:: SPARC Dependent Features
5986 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
5989 * V850-Dependent:: V850 Dependent Features
5992 * Xtensa-Dependent:: Xtensa Dependent Features
5995 * Z8000-Dependent:: Z8000 Dependent Features
5998 * Vax-Dependent:: VAX Dependent Features
6005 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6006 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6007 @c peculiarity: to preserve cross-references, there must be a node called
6008 @c "Machine Dependencies". Hence the conditional nodenames in each
6009 @c major node below. Node defaulting in makeinfo requires adjacency of
6010 @c node and sectioning commands; hence the repetition of @chapter BLAH
6011 @c in both conditional blocks.
6014 @include c-a29k.texi
6018 @include c-alpha.texi
6030 @include c-cris.texi
6035 @node Machine Dependencies
6036 @chapter Machine Dependent Features
6038 The machine instruction sets are different on each Renesas chip family,
6039 and there are also some syntax differences among the families. This
6040 chapter describes the specific @command{@value{AS}} features for each
6044 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6045 * H8/500-Dependent:: Renesas H8/500 Dependent Features
6046 * SH-Dependent:: Renesas SH Dependent Features
6053 @include c-d10v.texi
6057 @include c-d30v.texi
6061 @include c-h8300.texi
6065 @include c-h8500.texi
6069 @include c-hppa.texi
6073 @include c-i370.texi
6077 @include c-i386.texi
6081 @include c-i860.texi
6085 @include c-i960.texi
6089 @include c-ia64.texi
6093 @include c-ip2k.texi
6097 @include c-m32r.texi
6101 @include c-m68k.texi
6105 @include c-m68hc11.texi
6109 @include c-m88k.texi
6113 @include c-mips.texi
6117 @include c-mmix.texi
6121 @include c-msp430.texi
6125 @include c-ns32k.texi
6129 @include c-pdp11.texi
6142 @include c-sh64.texi
6146 @include c-sparc.texi
6150 @include c-tic54x.texi
6162 @include c-v850.texi
6166 @include c-xtensa.texi
6170 @c reverse effect of @down at top of generic Machine-Dep chapter
6174 @node Reporting Bugs
6175 @chapter Reporting Bugs
6176 @cindex bugs in assembler
6177 @cindex reporting bugs in assembler
6179 Your bug reports play an essential role in making @command{@value{AS}} reliable.
6181 Reporting a bug may help you by bringing a solution to your problem, or it may
6182 not. But in any case the principal function of a bug report is to help the
6183 entire community by making the next version of @command{@value{AS}} work better.
6184 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
6186 In order for a bug report to serve its purpose, you must include the
6187 information that enables us to fix the bug.
6190 * Bug Criteria:: Have you found a bug?
6191 * Bug Reporting:: How to report bugs
6195 @section Have You Found a Bug?
6196 @cindex bug criteria
6198 If you are not sure whether you have found a bug, here are some guidelines:
6201 @cindex fatal signal
6202 @cindex assembler crash
6203 @cindex crash of assembler
6205 If the assembler gets a fatal signal, for any input whatever, that is a
6206 @command{@value{AS}} bug. Reliable assemblers never crash.
6208 @cindex error on valid input
6210 If @command{@value{AS}} produces an error message for valid input, that is a bug.
6212 @cindex invalid input
6214 If @command{@value{AS}} does not produce an error message for invalid input, that
6215 is a bug. However, you should note that your idea of ``invalid input'' might
6216 be our idea of ``an extension'' or ``support for traditional practice''.
6219 If you are an experienced user of assemblers, your suggestions for improvement
6220 of @command{@value{AS}} are welcome in any case.
6224 @section How to Report Bugs
6226 @cindex assembler bugs, reporting
6228 A number of companies and individuals offer support for @sc{gnu} products. If
6229 you obtained @command{@value{AS}} from a support organization, we recommend you
6230 contact that organization first.
6232 You can find contact information for many support companies and
6233 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
6236 In any event, we also recommend that you send bug reports for @command{@value{AS}}
6237 to @samp{bug-binutils@@gnu.org}.
6239 The fundamental principle of reporting bugs usefully is this:
6240 @strong{report all the facts}. If you are not sure whether to state a
6241 fact or leave it out, state it!
6243 Often people omit facts because they think they know what causes the problem
6244 and assume that some details do not matter. Thus, you might assume that the
6245 name of a symbol you use in an example does not matter. Well, probably it does
6246 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
6247 happens to fetch from the location where that name is stored in memory;
6248 perhaps, if the name were different, the contents of that location would fool
6249 the assembler into doing the right thing despite the bug. Play it safe and
6250 give a specific, complete example. That is the easiest thing for you to do,
6251 and the most helpful.
6253 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
6254 it is new to us. Therefore, always write your bug reports on the assumption
6255 that the bug has not been reported previously.
6257 Sometimes people give a few sketchy facts and ask, ``Does this ring a
6258 bell?'' This cannot help us fix a bug, so it is basically useless. We
6259 respond by asking for enough details to enable us to investigate.
6260 You might as well expedite matters by sending them to begin with.
6262 To enable us to fix the bug, you should include all these things:
6266 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
6267 it with the @samp{--version} argument.
6269 Without this, we will not know whether there is any point in looking for
6270 the bug in the current version of @command{@value{AS}}.
6273 Any patches you may have applied to the @command{@value{AS}} source.
6276 The type of machine you are using, and the operating system name and
6280 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
6284 The command arguments you gave the assembler to assemble your example and
6285 observe the bug. To guarantee you will not omit something important, list them
6286 all. A copy of the Makefile (or the output from make) is sufficient.
6288 If we were to try to guess the arguments, we would probably guess wrong
6289 and then we might not encounter the bug.
6292 A complete input file that will reproduce the bug. If the bug is observed when
6293 the assembler is invoked via a compiler, send the assembler source, not the
6294 high level language source. Most compilers will produce the assembler source
6295 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
6296 the options @samp{-v --save-temps}; this will save the assembler source in a
6297 file with an extension of @file{.s}, and also show you exactly how
6298 @command{@value{AS}} is being run.
6301 A description of what behavior you observe that you believe is
6302 incorrect. For example, ``It gets a fatal signal.''
6304 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
6305 will certainly notice it. But if the bug is incorrect output, we might not
6306 notice unless it is glaringly wrong. You might as well not give us a chance to
6309 Even if the problem you experience is a fatal signal, you should still say so
6310 explicitly. Suppose something strange is going on, such as, your copy of
6311 @command{@value{AS}} is out of synch, or you have encountered a bug in the C
6312 library on your system. (This has happened!) Your copy might crash and ours
6313 would not. If you told us to expect a crash, then when ours fails to crash, we
6314 would know that the bug was not happening for us. If you had not told us to
6315 expect a crash, then we would not be able to draw any conclusion from our
6319 If you wish to suggest changes to the @command{@value{AS}} source, send us context
6320 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
6321 option. Always send diffs from the old file to the new file. If you even
6322 discuss something in the @command{@value{AS}} source, refer to it by context, not
6325 The line numbers in our development sources will not match those in your
6326 sources. Your line numbers would convey no useful information to us.
6329 Here are some things that are not necessary:
6333 A description of the envelope of the bug.
6335 Often people who encounter a bug spend a lot of time investigating
6336 which changes to the input file will make the bug go away and which
6337 changes will not affect it.
6339 This is often time consuming and not very useful, because the way we
6340 will find the bug is by running a single example under the debugger
6341 with breakpoints, not by pure deduction from a series of examples.
6342 We recommend that you save your time for something else.
6344 Of course, if you can find a simpler example to report @emph{instead}
6345 of the original one, that is a convenience for us. Errors in the
6346 output will be easier to spot, running under the debugger will take
6347 less time, and so on.
6349 However, simplification is not vital; if you do not want to do this,
6350 report the bug anyway and send us the entire test case you used.
6353 A patch for the bug.
6355 A patch for the bug does help us if it is a good one. But do not omit
6356 the necessary information, such as the test case, on the assumption that
6357 a patch is all we need. We might see problems with your patch and decide
6358 to fix the problem another way, or we might not understand it at all.
6360 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
6361 construct an example that will make the program follow a certain path through
6362 the code. If you do not send us the example, we will not be able to construct
6363 one, so we will not be able to verify that the bug is fixed.
6365 And if we cannot understand what bug you are trying to fix, or why your
6366 patch should be an improvement, we will not install it. A test case will
6367 help us to understand.
6370 A guess about what the bug is or what it depends on.
6372 Such guesses are usually wrong. Even we cannot guess right about such
6373 things without first using the debugger to find the facts.
6376 @node Acknowledgements
6377 @chapter Acknowledgements
6379 If you have contributed to @command{@value{AS}} and your name isn't listed here,
6380 it is not meant as a slight. We just don't know about it. Send mail to the
6381 maintainer, and we'll correct the situation. Currently
6383 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
6385 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
6388 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
6389 information and the 68k series machines, most of the preprocessing pass, and
6390 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
6392 K. Richard Pixley maintained GAS for a while, adding various enhancements and
6393 many bug fixes, including merging support for several processors, breaking GAS
6394 up to handle multiple object file format back ends (including heavy rewrite,
6395 testing, an integration of the coff and b.out back ends), adding configuration
6396 including heavy testing and verification of cross assemblers and file splits
6397 and renaming, converted GAS to strictly ANSI C including full prototypes, added
6398 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
6399 port (including considerable amounts of reverse engineering), a SPARC opcode
6400 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
6401 assertions and made them work, much other reorganization, cleanup, and lint.
6403 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
6404 in format-specific I/O modules.
6406 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
6407 has done much work with it since.
6409 The Intel 80386 machine description was written by Eliot Dresselhaus.
6411 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
6413 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
6414 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
6416 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
6417 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
6418 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
6419 support a.out format.
6421 Support for the Zilog Z8k and Renesas H8/300 and H8/500 processors (tc-z8k,
6422 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
6423 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
6424 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
6427 John Gilmore built the AMD 29000 support, added @code{.include} support, and
6428 simplified the configuration of which versions accept which directives. He
6429 updated the 68k machine description so that Motorola's opcodes always produced
6430 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
6431 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
6432 cross-compilation support, and one bug in relaxation that took a week and
6433 required the proverbial one-bit fix.
6435 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
6436 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
6437 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
6438 PowerPC assembler, and made a few other minor patches.
6440 Steve Chamberlain made @command{@value{AS}} able to generate listings.
6442 Hewlett-Packard contributed support for the HP9000/300.
6444 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
6445 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
6446 formats). This work was supported by both the Center for Software Science at
6447 the University of Utah and Cygnus Support.
6449 Support for ELF format files has been worked on by Mark Eichin of Cygnus
6450 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
6451 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
6452 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
6453 and some initial 64-bit support).
6455 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
6457 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
6458 support for openVMS/Alpha.
6460 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
6463 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
6464 Inc. added support for Xtensa processors.
6466 Several engineers at Cygnus Support have also provided many small bug fixes and
6467 configuration enhancements.
6469 Many others have contributed large or small bugfixes and enhancements. If
6470 you have contributed significant work and are not mentioned on this list, and
6471 want to be, let us know. Some of the history has been lost; we are not
6472 intentionally leaving anyone out.
projects / deliverable / binutils-gdb.git / blob