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
60 @c common OR combinations of conditions
80 @set abnormal-separator
84 @settitle Using @value{AS}
87 @settitle Using @value{AS} (@value{TARGET})
89 @setchapternewpage odd
94 @c WARE! Some of the machine-dependent sections contain tables of machine
95 @c instructions. Except in multi-column format, these tables look silly.
96 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
97 @c the multi-col format is faked within @example sections.
99 @c Again unfortunately, the natural size that fits on a page, for these tables,
100 @c is different depending on whether or not smallbook is turned on.
101 @c This matters, because of order: text flow switches columns at each page
104 @c The format faked in this source works reasonably well for smallbook,
105 @c not well for the default large-page format. This manual expects that if you
106 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
107 @c tables in question. You can turn on one without the other at your
108 @c discretion, of course.
111 @c the insn tables look just as silly in info files regardless of smallbook,
112 @c might as well show 'em anyways.
118 * As: (as). The GNU assembler.
127 This file documents the GNU Assembler "@value{AS}".
129 @c man begin COPYRIGHT
130 Copyright (C) 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002 Free Software Foundation, Inc.
132 Permission is granted to copy, distribute and/or modify this document
133 under the terms of the GNU Free Documentation License, Version 1.1
134 or any later version published by the Free Software Foundation;
135 with no Invariant Sections, with no Front-Cover Texts, and with no
136 Back-Cover Texts. A copy of the license is included in the
137 section entitled "GNU Free Documentation License".
142 Permission is granted to process this file through Tex and print the
143 results, provided the printed document carries copying permission
144 notice identical to this one except for the removal of this paragraph
145 (this paragraph not being relevant to the printed manual).
151 @title Using @value{AS}
152 @subtitle The @sc{gnu} Assembler
154 @subtitle for the @value{TARGET} family
157 @subtitle Version @value{VERSION}
160 The Free Software Foundation Inc. thanks The Nice Computer
161 Company of Australia for loaning Dean Elsner to write the
162 first (Vax) version of @command{as} for Project @sc{gnu}.
163 The proprietors, management and staff of TNCCA thank FSF for
164 distracting the boss while they got some work
167 @author Dean Elsner, Jay Fenlason & friends
171 \hfill {\it Using {\tt @value{AS}}}\par
172 \hfill Edited by Cygnus Support\par
174 %"boxit" macro for figures:
175 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
176 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
177 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
178 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
179 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
182 @vskip 0pt plus 1filll
183 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002 Free Software Foundation, Inc.
185 Permission is granted to copy, distribute and/or modify this document
186 under the terms of the GNU Free Documentation License, Version 1.1
187 or any later version published by the Free Software Foundation;
188 with no Invariant Sections, with no Front-Cover Texts, and with no
189 Back-Cover Texts. A copy of the license is included in the
190 section entitled "GNU Free Documentation License".
196 @top Using @value{AS}
198 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}} version
201 This version of the file describes @command{@value{AS}} configured to generate
202 code for @value{TARGET} architectures.
205 This document is distributed under the terms of the GNU Free
206 Documentation License. A copy of the license is included in the
207 section entitled "GNU Free Documentation License".
210 * Overview:: Overview
211 * Invoking:: Command-Line Options
213 * Sections:: Sections and Relocation
215 * Expressions:: Expressions
216 * Pseudo Ops:: Assembler Directives
217 * Machine Dependencies:: Machine Dependent Features
218 * Reporting Bugs:: Reporting Bugs
219 * Acknowledgements:: Who Did What
220 * GNU Free Documentation License:: GNU Free Documentation License
228 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
230 This version of the manual describes @command{@value{AS}} configured to generate
231 code for @value{TARGET} architectures.
235 @cindex invocation summary
236 @cindex option summary
237 @cindex summary of options
238 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
239 @pxref{Invoking,,Comand-Line Options}.
241 @c man title AS the portable GNU assembler.
245 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
249 @c We don't use deffn and friends for the following because they seem
250 @c to be limited to one line for the header.
252 @c man begin SYNOPSIS
253 @value{AS} [@b{-a}[@b{cdhlns}][=@var{file}]] [@b{-D}] [@b{--defsym} @var{sym}=@var{val}]
254 [@b{-f}] [@b{--gstabs}] [@b{--gdwarf2}] [@b{--help}] [@b{-I} @var{dir}]
255 [@b{-J}] [@b{-K}] [@b{-L}]
256 [@b{--listing-lhs-width}=@var{NUM}] [@b{--listing-lhs-width2}=@var{NUM}]
257 [@b{--listing-rhs-width}=@var{NUM}] [@b{--listing-cont-lines}=@var{NUM}]
258 [@b{--keep-locals}] [@b{-o} @var{objfile}] [@b{-R}] [@b{--statistics}] [@b{-v}]
259 [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}] [@b{--fatal-warnings}]
260 [@b{-w}] [@b{-x}] [@b{-Z}] [@b{--target-help}] [@var{target-options}]
261 [@b{--}|@var{files} @dots{}]
263 @c Target dependent options are listed below. Keep the list sorted.
264 @c Add an empty line for separation.
266 @c am29k has no machine-dependent assembler options
270 @emph{Target ARC options:}
276 @emph{Target ARM options:}
277 @c Don't document the deprecated options
278 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
279 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
280 [@b{-mfpu}=@var{floating-point-fromat}]
283 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
284 @b{-mapcs-reentrant}]
285 [@b{-mthumb-interwork}] [@b{-moabi}] [@b{-k}]
289 @emph{Target D10V options:}
294 @emph{Target D30V options:}
295 [@b{-O}|@b{-n}|@b{-N}]
298 @c Hitachi family chips have no machine-dependent assembler options
301 @c HPPA has no machine-dependent assembler options (yet).
305 @emph{Target i386 options:}
310 @emph{Target i960 options:}
311 @c see md_parse_option in tc-i960.c
312 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
314 [@b{-b}] [@b{-no-relax}]
318 @emph{Target M32R options:}
319 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
324 @emph{Target M680X0 options:}
325 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
329 @emph{Target M68HC11 options:}
330 [@b{-m68hc11}|@b{-m68hc12}]
331 [@b{--force-long-branchs}] [@b{--short-branchs}]
332 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
333 [@b{--print-opcodes}] [@b{--generate-example}]
337 @emph{Target MCORE options:}
338 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
339 [@b{-mcpu=[210|340]}]
343 @emph{Target MIPS options:}
344 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-G} @var{num}] [@b{-mcpu}=@var{CPU} ]
345 [@b{-mips1}] [@b{-mips2}] [@b{-mips3}] [@b{-mips4}] [@b{-mips5}]
346 [@b{-mips32}] [@b{-mips64}]
347 [@b{-m4650}] [@b{-no-m4650}]
348 [@b{--trap}] [@b{--break}] [@b{-n}]
349 [@b{--emulation}=@var{name} ]
353 @emph{Target MMIX options:}
354 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
355 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
356 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
357 [@b{--linker-allocated-gregs}]
361 @emph{Target PDP11 options:}
362 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
363 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
364 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
368 @emph{Target picoJava options:}
373 @emph{Target PowerPC options:}
374 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
375 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}|
376 @b{-mbooke32}|@b{-mbooke64}]
377 [@b{-mcom}|@b{-many}|@b{-maltivec}] [@b{-memb}]
378 [@b{-mregnames}|@b{-mno-regnames}]
379 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
380 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
381 [@b{-msolaris}|@b{-mno-solaris}]
385 @emph{Target SPARC options:}
386 @c The order here is important. See c-sparc.texi.
387 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
388 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
389 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
394 @emph{Target TIC54X options:}
395 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
396 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
399 @c Z8000 has no machine-dependent assembler options
408 Turn on listings, in any of a variety of ways:
412 omit false conditionals
415 omit debugging directives
418 include high-level source
424 include macro expansions
427 omit forms processing
433 set the name of the listing file
436 You may combine these options; for example, use @samp{-aln} for assembly
437 listing without forms processing. The @samp{=file} option, if used, must be
438 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
441 Ignored. This option is accepted for script compatibility with calls to
444 @item --defsym @var{sym}=@var{value}
445 Define the symbol @var{sym} to be @var{value} before assembling the input file.
446 @var{value} must be an integer constant. As in C, a leading @samp{0x}
447 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
450 ``fast''---skip whitespace and comment preprocessing (assume source is
454 Generate stabs debugging information for each assembler line. This
455 may help debugging assembler code, if the debugger can handle it.
458 Generate DWARF2 debugging information for each assembler line. This
459 may help debugging assembler code, if the debugger can handle it. Note - this
460 option is only supported by some targets, not all of them.
463 Print a summary of the command line options and exit.
466 Print a summary of all target specific options and exit.
469 Add directory @var{dir} to the search list for @code{.include} directives.
472 Don't warn about signed overflow.
475 @ifclear DIFF-TBL-KLUGE
476 This option is accepted but has no effect on the @value{TARGET} family.
478 @ifset DIFF-TBL-KLUGE
479 Issue warnings when difference tables altered for long displacements.
484 Keep (in the symbol table) local symbols. On traditional a.out systems
485 these start with @samp{L}, but different systems have different local
488 @item --listing-lhs-width=@var{number}
489 Set the maximum width, in words, of the output data column for an assembler
490 listing to @var{number}.
492 @item --listing-lhs-width2=@var{number}
493 Set the maximum width, in words, of the output data column for continuation
494 lines in an assembler listing to @var{number}.
496 @item --listing-rhs-width=@var{number}
497 Set the maximum width of an input source line, as displayed in a listing, to
500 @item --listing-cont-lines=@var{number}
501 Set the maximum number of lines printed in a listing for a single line of input
504 @item -o @var{objfile}
505 Name the object-file output from @command{@value{AS}} @var{objfile}.
508 Fold the data section into the text section.
511 Print the maximum space (in bytes) and total time (in seconds) used by
514 @item --strip-local-absolute
515 Remove local absolute symbols from the outgoing symbol table.
519 Print the @command{as} version.
522 Print the @command{as} version and exit.
526 Suppress warning messages.
528 @item --fatal-warnings
529 Treat warnings as errors.
532 Don't suppress warning messages or treat them as errors.
541 Generate an object file even after errors.
543 @item -- | @var{files} @dots{}
544 Standard input, or source files to assemble.
549 The following options are available when @value{AS} is configured for
554 This option selects the core processor variant.
556 Select either big-endian (-EB) or little-endian (-EL) output.
561 The following options are available when @value{AS} is configured for the ARM
565 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
566 Specify which ARM processor variant is the target.
567 @item -march=@var{architecture}[+@var{extension}@dots{}]
568 Specify which ARM architecture variant is used by the target.
569 @item -mfpu=@var{floating-point-format}
570 Select which Floating Point architecture is the target.
572 Enable Thumb only instruction decoding.
573 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant | -moabi
574 Select which procedure calling convention is in use.
576 Select either big-endian (-EB) or little-endian (-EL) output.
577 @item -mthumb-interwork
578 Specify that the code has been generated with interworking between Thumb and
581 Specify that PIC code has been generated.
586 The following options are available when @value{AS} is configured for
589 @cindex D10V optimization
590 @cindex optimization, D10V
592 Optimize output by parallelizing instructions.
597 The following options are available when @value{AS} is configured for a D30V
600 @cindex D30V optimization
601 @cindex optimization, D30V
603 Optimize output by parallelizing instructions.
607 Warn when nops are generated.
609 @cindex D30V nops after 32-bit multiply
611 Warn when a nop after a 32-bit multiply instruction is generated.
616 The following options are available when @value{AS} is configured for the
617 Intel 80960 processor.
620 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
621 Specify which variant of the 960 architecture is the target.
624 Add code to collect statistics about branches taken.
627 Do not alter compare-and-branch instructions for long displacements;
634 The following options are available when @value{AS} is configured for the
635 Mitsubishi M32R series.
640 Specify which processor in the M32R family is the target. The default
641 is normally the M32R, but this option changes it to the M32RX.
643 @item --warn-explicit-parallel-conflicts or --Wp
644 Produce warning messages when questionable parallel constructs are
647 @item --no-warn-explicit-parallel-conflicts or --Wnp
648 Do not produce warning messages when questionable parallel constructs are
655 The following options are available when @value{AS} is configured for the
656 Motorola 68000 series.
661 Shorten references to undefined symbols, to one word instead of two.
663 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
664 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
665 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
666 Specify what processor in the 68000 family is the target. The default
667 is normally the 68020, but this can be changed at configuration time.
669 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
670 The target machine does (or does not) have a floating-point coprocessor.
671 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
672 the basic 68000 is not compatible with the 68881, a combination of the
673 two can be specified, since it's possible to do emulation of the
674 coprocessor instructions with the main processor.
676 @item -m68851 | -mno-68851
677 The target machine does (or does not) have a memory-management
678 unit coprocessor. The default is to assume an MMU for 68020 and up.
685 For details about the PDP-11 machine dependent features options,
686 see @ref{PDP-11-Options}.
689 @item -mpic | -mno-pic
690 Generate position-independent (or position-dependent) code. The
691 default is @option{-mpic}.
694 @itemx -mall-extensions
695 Enable all instruction set extensions. This is the default.
697 @item -mno-extensions
698 Disable all instruction set extensions.
700 @item -m@var{extension} | -mno-@var{extension}
701 Enable (or disable) a particular instruction set extension.
704 Enable the instruction set extensions supported by a particular CPU, and
705 disable all other extensions.
707 @item -m@var{machine}
708 Enable the instruction set extensions supported by a particular machine
709 model, and disable all other extensions.
715 The following options are available when @value{AS} is configured for
716 a picoJava processor.
720 @cindex PJ endianness
721 @cindex endianness, PJ
722 @cindex big endian output, PJ
724 Generate ``big endian'' format output.
726 @cindex little endian output, PJ
728 Generate ``little endian'' format output.
734 The following options are available when @value{AS} is configured for the
735 Motorola 68HC11 or 68HC12 series.
739 @item -m68hc11 | -m68hc12
740 Specify what processor is the target. The default is
741 defined by the configuration option when building the assembler.
743 @item --force-long-branchs
744 Relative branches are turned into absolute ones. This concerns
745 conditional branches, unconditional branches and branches to a
748 @item -S | --short-branchs
749 Do not turn relative branchs into absolute ones
750 when the offset is out of range.
752 @item --strict-direct-mode
753 Do not turn the direct addressing mode into extended addressing mode
754 when the instruction does not support direct addressing mode.
756 @item --print-insn-syntax
757 Print the syntax of instruction in case of error.
759 @item --print-opcodes
760 print the list of instructions with syntax and then exit.
762 @item --generate-example
763 print an example of instruction for each possible instruction and then exit.
764 This option is only useful for testing @command{@value{AS}}.
770 The following options are available when @command{@value{AS}} is configured
771 for the SPARC architecture:
774 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
775 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
776 Explicitly select a variant of the SPARC architecture.
778 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
779 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
781 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
782 UltraSPARC extensions.
784 @item -xarch=v8plus | -xarch=v8plusa
785 For compatibility with the Solaris v9 assembler. These options are
786 equivalent to -Av8plus and -Av8plusa, respectively.
789 Warn when the assembler switches to another architecture.
794 The following options are available when @value{AS} is configured for the 'c54x
799 Enable extended addressing mode. All addresses and relocations will assume
800 extended addressing (usually 23 bits).
801 @item -mcpu=@var{CPU_VERSION}
802 Sets the CPU version being compiled for.
803 @item -merrors-to-file @var{FILENAME}
804 Redirect error output to a file, for broken systems which don't support such
805 behaviour in the shell.
810 The following options are available when @value{AS} is configured for
815 This option sets the largest size of an object that can be referenced
816 implicitly with the @code{gp} register. It is only accepted for targets that
817 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
819 @cindex MIPS endianness
820 @cindex endianness, MIPS
821 @cindex big endian output, MIPS
823 Generate ``big endian'' format output.
825 @cindex little endian output, MIPS
827 Generate ``little endian'' format output.
836 Generate code for a particular MIPS Instruction Set Architecture level.
837 @samp{-mips1} corresponds to the @sc{r2000} and @sc{r3000} processors,
838 @samp{-mips2} to the @sc{r6000} processor, and @samp{-mips3} to the @sc{r4000}
840 @samp{-mips5}, @samp{-mips32}, and @samp{-mips64} correspond
841 to generic @sc{MIPS V}, @sc{MIPS32}, and @sc{MIPS64} ISA
842 processors, respectively.
846 Generate code for the MIPS @sc{r4650} chip. This tells the assembler to accept
847 the @samp{mad} and @samp{madu} instruction, and to not schedule @samp{nop}
848 instructions around accesses to the @samp{HI} and @samp{LO} registers.
849 @samp{-no-m4650} turns off this option.
851 @item -mcpu=@var{CPU}
852 Generate code for a particular MIPS cpu. It is exactly equivalent to
853 @samp{-m@var{cpu}}, except that there are more value of @var{cpu}
857 @item --emulation=@var{name}
858 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
859 for some other target, in all respects, including output format (choosing
860 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
861 debugging information or store symbol table information, and default
862 endianness. The available configuration names are: @samp{mipsecoff},
863 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
864 @samp{mipsbelf}. The first two do not alter the default endianness from that
865 of the primary target for which the assembler was configured; the others change
866 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
867 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
868 selection in any case.
870 This option is currently supported only when the primary target
871 @command{@value{AS}} is configured for is a MIPS ELF or ECOFF target.
872 Furthermore, the primary target or others specified with
873 @samp{--enable-targets=@dots{}} at configuration time must include support for
874 the other format, if both are to be available. For example, the Irix 5
875 configuration includes support for both.
877 Eventually, this option will support more configurations, with more
878 fine-grained control over the assembler's behavior, and will be supported for
882 @command{@value{AS}} ignores this option. It is accepted for compatibility with
890 Control how to deal with multiplication overflow and division by zero.
891 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
892 (and only work for Instruction Set Architecture level 2 and higher);
893 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
897 When this option is used, @command{@value{AS}} will issue a warning every
898 time it generates a nop instruction from a macro.
903 The following options are available when @value{AS} is configured for
909 Enable or disable the JSRI to BSR transformation. By default this is enabled.
910 The command line option @samp{-nojsri2bsr} can be used to disable it.
914 Enable or disable the silicon filter behaviour. By default this is disabled.
915 The default can be overridden by the @samp{-sifilter} command line option.
918 Alter jump instructions for long displacements.
920 @item -mcpu=[210|340]
921 Select the cpu type on the target hardware. This controls which instructions
925 Assemble for a big endian target.
928 Assemble for a little endian target.
934 See the info pages for documentation of the MMIX-specific options.
940 * Manual:: Structure of this Manual
941 * GNU Assembler:: The GNU Assembler
942 * Object Formats:: Object File Formats
943 * Command Line:: Command Line
944 * Input Files:: Input Files
945 * Object:: Output (Object) File
946 * Errors:: Error and Warning Messages
950 @section Structure of this Manual
952 @cindex manual, structure and purpose
953 This manual is intended to describe what you need to know to use
954 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
955 notation for symbols, constants, and expressions; the directives that
956 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
959 We also cover special features in the @value{TARGET}
960 configuration of @command{@value{AS}}, including assembler directives.
963 This manual also describes some of the machine-dependent features of
964 various flavors of the assembler.
967 @cindex machine instructions (not covered)
968 On the other hand, this manual is @emph{not} intended as an introduction
969 to programming in assembly language---let alone programming in general!
970 In a similar vein, we make no attempt to introduce the machine
971 architecture; we do @emph{not} describe the instruction set, standard
972 mnemonics, registers or addressing modes that are standard to a
973 particular architecture.
975 You may want to consult the manufacturer's
976 machine architecture manual for this information.
980 For information on the H8/300 machine instruction set, see @cite{H8/300
981 Series Programming Manual} (Hitachi ADE--602--025). For the H8/300H,
982 see @cite{H8/300H Series Programming Manual} (Hitachi).
985 For information on the H8/500 machine instruction set, see @cite{H8/500
986 Series Programming Manual} (Hitachi M21T001).
989 For information on the Hitachi SH machine instruction set, see
990 @cite{SH-Microcomputer User's Manual} (Hitachi Micro Systems, Inc.).
993 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
997 @c I think this is premature---doc@cygnus.com, 17jan1991
999 Throughout this manual, we assume that you are running @dfn{GNU},
1000 the portable operating system from the @dfn{Free Software
1001 Foundation, Inc.}. This restricts our attention to certain kinds of
1002 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1003 once this assumption is granted examples and definitions need less
1006 @command{@value{AS}} is part of a team of programs that turn a high-level
1007 human-readable series of instructions into a low-level
1008 computer-readable series of instructions. Different versions of
1009 @command{@value{AS}} are used for different kinds of computer.
1012 @c There used to be a section "Terminology" here, which defined
1013 @c "contents", "byte", "word", and "long". Defining "word" to any
1014 @c particular size is confusing when the .word directive may generate 16
1015 @c bits on one machine and 32 bits on another; in general, for the user
1016 @c version of this manual, none of these terms seem essential to define.
1017 @c They were used very little even in the former draft of the manual;
1018 @c this draft makes an effort to avoid them (except in names of
1022 @section The GNU Assembler
1024 @c man begin DESCRIPTION
1026 @sc{gnu} @command{as} is really a family of assemblers.
1028 This manual describes @command{@value{AS}}, a member of that family which is
1029 configured for the @value{TARGET} architectures.
1031 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1032 should find a fairly similar environment when you use it on another
1033 architecture. Each version has much in common with the others,
1034 including object file formats, most assembler directives (often called
1035 @dfn{pseudo-ops}) and assembler syntax.@refill
1037 @cindex purpose of @sc{gnu} assembler
1038 @command{@value{AS}} is primarily intended to assemble the output of the
1039 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1040 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1041 assemble correctly everything that other assemblers for the same
1042 machine would assemble.
1044 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1047 @c This remark should appear in generic version of manual; assumption
1048 @c here is that generic version sets M680x0.
1049 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1050 assembler for the same architecture; for example, we know of several
1051 incompatible versions of 680x0 assembly language syntax.
1056 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1057 program in one pass of the source file. This has a subtle impact on the
1058 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1060 @node Object Formats
1061 @section Object File Formats
1063 @cindex object file format
1064 The @sc{gnu} assembler can be configured to produce several alternative
1065 object file formats. For the most part, this does not affect how you
1066 write assembly language programs; but directives for debugging symbols
1067 are typically different in different file formats. @xref{Symbol
1068 Attributes,,Symbol Attributes}.
1071 On the @value{TARGET}, @command{@value{AS}} is configured to produce
1072 @value{OBJ-NAME} format object files.
1074 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1076 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1077 @code{a.out} or COFF format object files.
1080 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1081 @code{b.out} or COFF format object files.
1084 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1085 SOM or ELF format object files.
1090 @section Command Line
1092 @cindex command line conventions
1094 After the program name @command{@value{AS}}, the command line may contain
1095 options and file names. Options may appear in any order, and may be
1096 before, after, or between file names. The order of file names is
1099 @cindex standard input, as input file
1101 @file{--} (two hyphens) by itself names the standard input file
1102 explicitly, as one of the files for @command{@value{AS}} to assemble.
1104 @cindex options, command line
1105 Except for @samp{--} any command line argument that begins with a
1106 hyphen (@samp{-}) is an option. Each option changes the behavior of
1107 @command{@value{AS}}. No option changes the way another option works. An
1108 option is a @samp{-} followed by one or more letters; the case of
1109 the letter is important. All options are optional.
1111 Some options expect exactly one file name to follow them. The file
1112 name may either immediately follow the option's letter (compatible
1113 with older assemblers) or it may be the next command argument (@sc{gnu}
1114 standard). These two command lines are equivalent:
1117 @value{AS} -o my-object-file.o mumble.s
1118 @value{AS} -omy-object-file.o mumble.s
1122 @section Input Files
1125 @cindex source program
1126 @cindex files, input
1127 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1128 describe the program input to one run of @command{@value{AS}}. The program may
1129 be in one or more files; how the source is partitioned into files
1130 doesn't change the meaning of the source.
1132 @c I added "con" prefix to "catenation" just to prove I can overcome my
1133 @c APL training... doc@cygnus.com
1134 The source program is a concatenation of the text in all the files, in the
1137 @c man begin DESCRIPTION
1138 Each time you run @command{@value{AS}} it assembles exactly one source
1139 program. The source program is made up of one or more files.
1140 (The standard input is also a file.)
1142 You give @command{@value{AS}} a command line that has zero or more input file
1143 names. The input files are read (from left file name to right). A
1144 command line argument (in any position) that has no special meaning
1145 is taken to be an input file name.
1147 If you give @command{@value{AS}} no file names it attempts to read one input file
1148 from the @command{@value{AS}} standard input, which is normally your terminal. You
1149 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1152 Use @samp{--} if you need to explicitly name the standard input file
1153 in your command line.
1155 If the source is empty, @command{@value{AS}} produces a small, empty object
1160 @subheading Filenames and Line-numbers
1162 @cindex input file linenumbers
1163 @cindex line numbers, in input files
1164 There are two ways of locating a line in the input file (or files) and
1165 either may be used in reporting error messages. One way refers to a line
1166 number in a physical file; the other refers to a line number in a
1167 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1169 @dfn{Physical files} are those files named in the command line given
1170 to @command{@value{AS}}.
1172 @dfn{Logical files} are simply names declared explicitly by assembler
1173 directives; they bear no relation to physical files. Logical file names help
1174 error messages reflect the original source file, when @command{@value{AS}} source
1175 is itself synthesized from other files. @command{@value{AS}} understands the
1176 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1177 @ref{File,,@code{.file}}.
1180 @section Output (Object) File
1186 Every time you run @command{@value{AS}} it produces an output file, which is
1187 your assembly language program translated into numbers. This file
1188 is the object file. Its default name is
1196 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1198 You can give it another name by using the @option{-o} option. Conventionally,
1199 object file names end with @file{.o}. The default name is used for historical
1200 reasons: older assemblers were capable of assembling self-contained programs
1201 directly into a runnable program. (For some formats, this isn't currently
1202 possible, but it can be done for the @code{a.out} format.)
1206 The object file is meant for input to the linker @code{@value{LD}}. It contains
1207 assembled program code, information to help @code{@value{LD}} integrate
1208 the assembled program into a runnable file, and (optionally) symbolic
1209 information for the debugger.
1211 @c link above to some info file(s) like the description of a.out.
1212 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1215 @section Error and Warning Messages
1217 @c man begin DESCRIPTION
1219 @cindex error messages
1220 @cindex warning messages
1221 @cindex messages from assembler
1222 @command{@value{AS}} may write warnings and error messages to the standard error
1223 file (usually your terminal). This should not happen when a compiler
1224 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1225 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1226 grave problem that stops the assembly.
1230 @cindex format of warning messages
1231 Warning messages have the format
1234 file_name:@b{NNN}:Warning Message Text
1238 @cindex line numbers, in warnings/errors
1239 (where @b{NNN} is a line number). If a logical file name has been given
1240 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1241 the current input file is used. If a logical line number was given
1243 (@pxref{Line,,@code{.line}})
1247 (@pxref{Line,,@code{.line}})
1250 (@pxref{Ln,,@code{.ln}})
1253 then it is used to calculate the number printed,
1254 otherwise the actual line in the current source file is printed. The
1255 message text is intended to be self explanatory (in the grand Unix
1258 @cindex format of error messages
1259 Error messages have the format
1261 file_name:@b{NNN}:FATAL:Error Message Text
1263 The file name and line number are derived as for warning
1264 messages. The actual message text may be rather less explanatory
1265 because many of them aren't supposed to happen.
1268 @chapter Command-Line Options
1270 @cindex options, all versions of assembler
1271 This chapter describes command-line options available in @emph{all}
1272 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
1274 to the @value{TARGET}.
1277 to particular machine architectures.
1280 @c man begin DESCRIPTION
1282 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler (version 2),
1283 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1284 The assembler arguments must be separated from each other (and the @samp{-Wa})
1285 by commas. For example:
1288 gcc -c -g -O -Wa,-alh,-L file.c
1292 This passes two options to the assembler: @samp{-alh} (emit a listing to
1293 standard output with with high-level and assembly source) and @samp{-L} (retain
1294 local symbols in the symbol table).
1296 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1297 command-line options are automatically passed to the assembler by the compiler.
1298 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1299 precisely what options it passes to each compilation pass, including the
1305 * a:: -a[cdhlns] enable listings
1306 * D:: -D for compatibility
1307 * f:: -f to work faster
1308 * I:: -I for .include search path
1309 @ifclear DIFF-TBL-KLUGE
1310 * K:: -K for compatibility
1312 @ifset DIFF-TBL-KLUGE
1313 * K:: -K for difference tables
1316 * L:: -L to retain local labels
1317 * listing:: --listing-XXX to configure listing output
1318 * M:: -M or --mri to assemble in MRI compatibility mode
1319 * MD:: --MD for dependency tracking
1320 * o:: -o to name the object file
1321 * R:: -R to join data and text sections
1322 * statistics:: --statistics to see statistics about assembly
1323 * traditional-format:: --traditional-format for compatible output
1324 * v:: -v to announce version
1325 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1326 * Z:: -Z to make object file even after errors
1330 @section Enable Listings: @option{-a[cdhlns]}
1339 @cindex listings, enabling
1340 @cindex assembly listings, enabling
1342 These options enable listing output from the assembler. By itself,
1343 @samp{-a} requests high-level, assembly, and symbols listing.
1344 You can use other letters to select specific options for the list:
1345 @samp{-ah} requests a high-level language listing,
1346 @samp{-al} requests an output-program assembly listing, and
1347 @samp{-as} requests a symbol table listing.
1348 High-level listings require that a compiler debugging option like
1349 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1352 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1353 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1354 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1355 omitted from the listing.
1357 Use the @samp{-ad} option to omit debugging directives from the
1360 Once you have specified one of these options, you can further control
1361 listing output and its appearance using the directives @code{.list},
1362 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1364 The @samp{-an} option turns off all forms processing.
1365 If you do not request listing output with one of the @samp{-a} options, the
1366 listing-control directives have no effect.
1368 The letters after @samp{-a} may be combined into one option,
1369 @emph{e.g.}, @samp{-aln}.
1371 Note if the assembler source is coming from the standard input (eg because it
1372 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1373 is being used) then the listing will not contain any comments or preprocessor
1374 directives. This is because the listing code buffers input source lines from
1375 stdin only after they have been preprocessed by the assembler. This reduces
1376 memory usage and makes the code more efficient.
1379 @section @option{-D}
1382 This option has no effect whatsoever, but it is accepted to make it more
1383 likely that scripts written for other assemblers also work with
1384 @command{@value{AS}}.
1387 @section Work Faster: @option{-f}
1390 @cindex trusted compiler
1391 @cindex faster processing (@option{-f})
1392 @samp{-f} should only be used when assembling programs written by a
1393 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1394 and comment preprocessing on
1395 the input file(s) before assembling them. @xref{Preprocessing,
1399 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1400 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1405 @section @code{.include} search path: @option{-I} @var{path}
1407 @kindex -I @var{path}
1408 @cindex paths for @code{.include}
1409 @cindex search path for @code{.include}
1410 @cindex @code{include} directive search path
1411 Use this option to add a @var{path} to the list of directories
1412 @command{@value{AS}} searches for files specified in @code{.include}
1413 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1414 many times as necessary to include a variety of paths. The current
1415 working directory is always searched first; after that, @command{@value{AS}}
1416 searches any @samp{-I} directories in the same order as they were
1417 specified (left to right) on the command line.
1420 @section Difference Tables: @option{-K}
1423 @ifclear DIFF-TBL-KLUGE
1424 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1425 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1426 where it can be used to warn when the assembler alters the machine code
1427 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1428 family does not have the addressing limitations that sometimes lead to this
1429 alteration on other platforms.
1432 @ifset DIFF-TBL-KLUGE
1433 @cindex difference tables, warning
1434 @cindex warning for altered difference tables
1435 @command{@value{AS}} sometimes alters the code emitted for directives of the form
1436 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1437 You can use the @samp{-K} option if you want a warning issued when this
1442 @section Include Local Labels: @option{-L}
1445 @cindex local labels, retaining in output
1446 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1447 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1448 debugging, because they are intended for the use of programs (like
1449 compilers) that compose assembler programs, not for your notice.
1450 Normally both @command{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1451 normally debug with them.
1453 This option tells @command{@value{AS}} to retain those @samp{L@dots{}} symbols
1454 in the object file. Usually if you do this you also tell the linker
1455 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1457 By default, a local label is any label beginning with @samp{L}, but each
1458 target is allowed to redefine the local label prefix.
1460 On the HPPA local labels begin with @samp{L$}.
1464 @section Configuring listing output: @option{--listing}
1466 The listing feature of the assembler can be enabled via the command line switch
1467 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1468 hex dump of the corresponding locations in the output object file, and displays
1469 them as a listing file. The format of this listing can be controlled by pseudo
1470 ops inside the assembler source (@pxref{List} @pxref{Title} @pxref{Sbttl}
1471 @pxref{Psize} @pxref{Eject}) and also by the following switches:
1474 @item --listing-lhs-width=@samp{number}
1475 @kindex --listing-lhs-width
1476 @cindex Width of first line disassembly output
1477 Sets the maximum width, in words, of the first line of the hex byte dump. This
1478 dump appears on the left hand side of the listing output.
1480 @item --listing-lhs-width2=@samp{number}
1481 @kindex --listing-lhs-width2
1482 @cindex Width of continuation lines of disassembly output
1483 Sets the maximum width, in words, of any further lines of the hex byte dump for
1484 a given inut source line. If this value is not specified, it defaults to being
1485 the same as the value specified for @samp{--listing-lhs-width}. If neither
1486 switch is used the default is to one.
1488 @item --listing-rhs-width=@samp{number}
1489 @kindex --listing-rhs-width
1490 @cindex Width of source line output
1491 Sets the maximum width, in characters, of the source line that is displayed
1492 alongside the hex dump. The default value for this parameter is 100. The
1493 source line is displayed on the right hand side of the listing output.
1495 @item --listing-cont-lines=@samp{number}
1496 @kindex --listing-cont-lines
1497 @cindex Maximum number of continuation lines
1498 Sets the maximum number of continuation lines of hex dump that will be
1499 displayed for a given single line of source input. The default value is 4.
1503 @section Assemble in MRI Compatibility Mode: @option{-M}
1506 @cindex MRI compatibility mode
1507 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1508 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1509 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1510 configured target) assembler from Microtec Research. The exact nature of the
1511 MRI syntax will not be documented here; see the MRI manuals for more
1512 information. Note in particular that the handling of macros and macro
1513 arguments is somewhat different. The purpose of this option is to permit
1514 assembling existing MRI assembler code using @command{@value{AS}}.
1516 The MRI compatibility is not complete. Certain operations of the MRI assembler
1517 depend upon its object file format, and can not be supported using other object
1518 file formats. Supporting these would require enhancing each object file format
1519 individually. These are:
1522 @item global symbols in common section
1524 The m68k MRI assembler supports common sections which are merged by the linker.
1525 Other object file formats do not support this. @command{@value{AS}} handles
1526 common sections by treating them as a single common symbol. It permits local
1527 symbols to be defined within a common section, but it can not support global
1528 symbols, since it has no way to describe them.
1530 @item complex relocations
1532 The MRI assemblers support relocations against a negated section address, and
1533 relocations which combine the start addresses of two or more sections. These
1534 are not support by other object file formats.
1536 @item @code{END} pseudo-op specifying start address
1538 The MRI @code{END} pseudo-op permits the specification of a start address.
1539 This is not supported by other object file formats. The start address may
1540 instead be specified using the @option{-e} option to the linker, or in a linker
1543 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1545 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1546 name to the output file. This is not supported by other object file formats.
1548 @item @code{ORG} pseudo-op
1550 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1551 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1552 which changes the location within the current section. Absolute sections are
1553 not supported by other object file formats. The address of a section may be
1554 assigned within a linker script.
1557 There are some other features of the MRI assembler which are not supported by
1558 @command{@value{AS}}, typically either because they are difficult or because they
1559 seem of little consequence. Some of these may be supported in future releases.
1563 @item EBCDIC strings
1565 EBCDIC strings are not supported.
1567 @item packed binary coded decimal
1569 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1570 and @code{DCB.P} pseudo-ops are not supported.
1572 @item @code{FEQU} pseudo-op
1574 The m68k @code{FEQU} pseudo-op is not supported.
1576 @item @code{NOOBJ} pseudo-op
1578 The m68k @code{NOOBJ} pseudo-op is not supported.
1580 @item @code{OPT} branch control options
1582 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1583 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1584 relaxes all branches, whether forward or backward, to an appropriate size, so
1585 these options serve no purpose.
1587 @item @code{OPT} list control options
1589 The following m68k @code{OPT} list control options are ignored: @code{C},
1590 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1591 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1593 @item other @code{OPT} options
1595 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1596 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1598 @item @code{OPT} @code{D} option is default
1600 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1601 @code{OPT NOD} may be used to turn it off.
1603 @item @code{XREF} pseudo-op.
1605 The m68k @code{XREF} pseudo-op is ignored.
1607 @item @code{.debug} pseudo-op
1609 The i960 @code{.debug} pseudo-op is not supported.
1611 @item @code{.extended} pseudo-op
1613 The i960 @code{.extended} pseudo-op is not supported.
1615 @item @code{.list} pseudo-op.
1617 The various options of the i960 @code{.list} pseudo-op are not supported.
1619 @item @code{.optimize} pseudo-op
1621 The i960 @code{.optimize} pseudo-op is not supported.
1623 @item @code{.output} pseudo-op
1625 The i960 @code{.output} pseudo-op is not supported.
1627 @item @code{.setreal} pseudo-op
1629 The i960 @code{.setreal} pseudo-op is not supported.
1634 @section Dependency tracking: @option{--MD}
1637 @cindex dependency tracking
1640 @command{@value{AS}} can generate a dependency file for the file it creates. This
1641 file consists of a single rule suitable for @code{make} describing the
1642 dependencies of the main source file.
1644 The rule is written to the file named in its argument.
1646 This feature is used in the automatic updating of makefiles.
1649 @section Name the Object File: @option{-o}
1652 @cindex naming object file
1653 @cindex object file name
1654 There is always one object file output when you run @command{@value{AS}}. By
1655 default it has the name
1658 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1672 You use this option (which takes exactly one filename) to give the
1673 object file a different name.
1675 Whatever the object file is called, @command{@value{AS}} overwrites any
1676 existing file of the same name.
1679 @section Join Data and Text Sections: @option{-R}
1682 @cindex data and text sections, joining
1683 @cindex text and data sections, joining
1684 @cindex joining text and data sections
1685 @cindex merging text and data sections
1686 @option{-R} tells @command{@value{AS}} to write the object file as if all
1687 data-section data lives in the text section. This is only done at
1688 the very last moment: your binary data are the same, but data
1689 section parts are relocated differently. The data section part of
1690 your object file is zero bytes long because all its bytes are
1691 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1693 When you specify @option{-R} it would be possible to generate shorter
1694 address displacements (because we do not have to cross between text and
1695 data section). We refrain from doing this simply for compatibility with
1696 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
1699 When @command{@value{AS}} is configured for COFF output,
1700 this option is only useful if you use sections named @samp{.text} and
1705 @option{-R} is not supported for any of the HPPA targets. Using
1706 @option{-R} generates a warning from @command{@value{AS}}.
1710 @section Display Assembly Statistics: @option{--statistics}
1712 @kindex --statistics
1713 @cindex statistics, about assembly
1714 @cindex time, total for assembly
1715 @cindex space used, maximum for assembly
1716 Use @samp{--statistics} to display two statistics about the resources used by
1717 @command{@value{AS}}: the maximum amount of space allocated during the assembly
1718 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1721 @node traditional-format
1722 @section Compatible output: @option{--traditional-format}
1724 @kindex --traditional-format
1725 For some targets, the output of @command{@value{AS}} is different in some ways
1726 from the output of some existing assembler. This switch requests
1727 @command{@value{AS}} to use the traditional format instead.
1729 For example, it disables the exception frame optimizations which
1730 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
1733 @section Announce Version: @option{-v}
1737 @cindex assembler version
1738 @cindex version of assembler
1739 You can find out what version of as is running by including the
1740 option @samp{-v} (which you can also spell as @samp{-version}) on the
1744 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
1746 @command{@value{AS}} should never give a warning or error message when
1747 assembling compiler output. But programs written by people often
1748 cause @command{@value{AS}} to give a warning that a particular assumption was
1749 made. All such warnings are directed to the standard error file.
1752 @kindex @samp{--no-warn}
1753 @cindex suppressing warnings
1754 @cindex warnings, suppressing
1755 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
1756 This only affects the warning messages: it does not change any particular of
1757 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
1760 @kindex @samp{--fatal-warnings}
1761 @cindex errors, caused by warnings
1762 @cindex warnings, causing error
1763 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
1764 files that generate warnings to be in error.
1766 @kindex @samp{--warn}
1767 @cindex warnings, switching on
1768 You can switch these options off again by specifying @option{--warn}, which
1769 causes warnings to be output as usual.
1772 @section Generate Object File in Spite of Errors: @option{-Z}
1773 @cindex object file, after errors
1774 @cindex errors, continuing after
1775 After an error message, @command{@value{AS}} normally produces no output. If for
1776 some reason you are interested in object file output even after
1777 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
1778 option. If there are any errors, @command{@value{AS}} continues anyways, and
1779 writes an object file after a final warning message of the form @samp{@var{n}
1780 errors, @var{m} warnings, generating bad object file.}
1785 @cindex machine-independent syntax
1786 @cindex syntax, machine-independent
1787 This chapter describes the machine-independent syntax allowed in a
1788 source file. @command{@value{AS}} syntax is similar to what many other
1789 assemblers use; it is inspired by the BSD 4.2
1794 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
1798 * Preprocessing:: Preprocessing
1799 * Whitespace:: Whitespace
1800 * Comments:: Comments
1801 * Symbol Intro:: Symbols
1802 * Statements:: Statements
1803 * Constants:: Constants
1807 @section Preprocessing
1809 @cindex preprocessing
1810 The @command{@value{AS}} internal preprocessor:
1812 @cindex whitespace, removed by preprocessor
1814 adjusts and removes extra whitespace. It leaves one space or tab before
1815 the keywords on a line, and turns any other whitespace on the line into
1818 @cindex comments, removed by preprocessor
1820 removes all comments, replacing them with a single space, or an
1821 appropriate number of newlines.
1823 @cindex constants, converted by preprocessor
1825 converts character constants into the appropriate numeric values.
1828 It does not do macro processing, include file handling, or
1829 anything else you may get from your C compiler's preprocessor. You can
1830 do include file processing with the @code{.include} directive
1831 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
1832 to get other ``CPP'' style preprocessing, by giving the input file a
1833 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
1834 Output, gcc.info, Using GNU CC}.
1836 Excess whitespace, comments, and character constants
1837 cannot be used in the portions of the input text that are not
1840 @cindex turning preprocessing on and off
1841 @cindex preprocessing, turning on and off
1844 If the first line of an input file is @code{#NO_APP} or if you use the
1845 @samp{-f} option, whitespace and comments are not removed from the input file.
1846 Within an input file, you can ask for whitespace and comment removal in
1847 specific portions of the by putting a line that says @code{#APP} before the
1848 text that may contain whitespace or comments, and putting a line that says
1849 @code{#NO_APP} after this text. This feature is mainly intend to support
1850 @code{asm} statements in compilers whose output is otherwise free of comments
1857 @dfn{Whitespace} is one or more blanks or tabs, in any order.
1858 Whitespace is used to separate symbols, and to make programs neater for
1859 people to read. Unless within character constants
1860 (@pxref{Characters,,Character Constants}), any whitespace means the same
1861 as exactly one space.
1867 There are two ways of rendering comments to @command{@value{AS}}. In both
1868 cases the comment is equivalent to one space.
1870 Anything from @samp{/*} through the next @samp{*/} is a comment.
1871 This means you may not nest these comments.
1875 The only way to include a newline ('\n') in a comment
1876 is to use this sort of comment.
1879 /* This sort of comment does not nest. */
1882 @cindex line comment character
1883 Anything from the @dfn{line comment} character to the next newline
1884 is considered a comment and is ignored. The line comment character is
1886 @samp{;} for the AMD 29K family;
1889 @samp{;} on the ARC;
1892 @samp{@@} on the ARM;
1895 @samp{;} for the H8/300 family;
1898 @samp{!} for the H8/500 family;
1901 @samp{;} for the HPPA;
1904 @samp{#} on the i386 and x86-64;
1907 @samp{#} on the i960;
1910 @samp{;} for the PDP-11;
1913 @samp{;} for picoJava;
1916 @samp{;} for Motorola PowerPC;
1919 @samp{!} for the Hitachi SH;
1922 @samp{!} on the SPARC;
1925 @samp{#} on the m32r;
1928 @samp{|} on the 680x0;
1931 @samp{#} on the 68HC11 and 68HC12;
1934 @samp{;} on the M880x0;
1937 @samp{#} on the Vax;
1940 @samp{!} for the Z8000;
1943 @samp{#} on the V850;
1945 see @ref{Machine Dependencies}. @refill
1946 @c FIXME What about i860?
1949 On some machines there are two different line comment characters. One
1950 character only begins a comment if it is the first non-whitespace character on
1951 a line, while the other always begins a comment.
1955 The V850 assembler also supports a double dash as starting a comment that
1956 extends to the end of the line.
1962 @cindex lines starting with @code{#}
1963 @cindex logical line numbers
1964 To be compatible with past assemblers, lines that begin with @samp{#} have a
1965 special interpretation. Following the @samp{#} should be an absolute
1966 expression (@pxref{Expressions}): the logical line number of the @emph{next}
1967 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
1968 new logical file name. The rest of the line, if any, should be whitespace.
1970 If the first non-whitespace characters on the line are not numeric,
1971 the line is ignored. (Just like a comment.)
1974 # This is an ordinary comment.
1975 # 42-6 "new_file_name" # New logical file name
1976 # This is logical line # 36.
1978 This feature is deprecated, and may disappear from future versions
1979 of @command{@value{AS}}.
1984 @cindex characters used in symbols
1985 @ifclear SPECIAL-SYMS
1986 A @dfn{symbol} is one or more characters chosen from the set of all
1987 letters (both upper and lower case), digits and the three characters
1993 A @dfn{symbol} is one or more characters chosen from the set of all
1994 letters (both upper and lower case), digits and the three characters
1995 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2001 On most machines, you can also use @code{$} in symbol names; exceptions
2002 are noted in @ref{Machine Dependencies}.
2004 No symbol may begin with a digit. Case is significant.
2005 There is no length limit: all characters are significant. Symbols are
2006 delimited by characters not in that set, or by the beginning of a file
2007 (since the source program must end with a newline, the end of a file is
2008 not a possible symbol delimiter). @xref{Symbols}.
2009 @cindex length of symbols
2014 @cindex statements, structure of
2015 @cindex line separator character
2016 @cindex statement separator character
2018 @ifclear abnormal-separator
2019 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2020 semicolon (@samp{;}). The newline or semicolon is considered part of
2021 the preceding statement. Newlines and semicolons within character
2022 constants are an exception: they do not end statements.
2024 @ifset abnormal-separator
2026 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
2027 sign (@samp{@@}). The newline or at sign is considered part of the
2028 preceding statement. Newlines and at signs within character constants
2029 are an exception: they do not end statements.
2032 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2033 point (@samp{!}). The newline or exclamation point is considered part of the
2034 preceding statement. Newlines and exclamation points within character
2035 constants are an exception: they do not end statements.
2038 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2039 H8/300) a dollar sign (@samp{$}); or (for the
2042 (@samp{;}). The newline or separator character is considered part of
2043 the preceding statement. Newlines and separators within character
2044 constants are an exception: they do not end statements.
2049 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2050 separator character. (The line separator is usually @samp{;}, unless
2051 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
2052 newline or separator character is considered part of the preceding
2053 statement. Newlines and separators within character constants are an
2054 exception: they do not end statements.
2057 @cindex newline, required at file end
2058 @cindex EOF, newline must precede
2059 It is an error to end any statement with end-of-file: the last
2060 character of any input file should be a newline.@refill
2062 An empty statement is allowed, and may include whitespace. It is ignored.
2064 @cindex instructions and directives
2065 @cindex directives and instructions
2066 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2067 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2069 A statement begins with zero or more labels, optionally followed by a
2070 key symbol which determines what kind of statement it is. The key
2071 symbol determines the syntax of the rest of the statement. If the
2072 symbol begins with a dot @samp{.} then the statement is an assembler
2073 directive: typically valid for any computer. If the symbol begins with
2074 a letter the statement is an assembly language @dfn{instruction}: it
2075 assembles into a machine language instruction.
2077 Different versions of @command{@value{AS}} for different computers
2078 recognize different instructions. In fact, the same symbol may
2079 represent a different instruction in a different computer's assembly
2083 @cindex @code{:} (label)
2084 @cindex label (@code{:})
2085 A label is a symbol immediately followed by a colon (@code{:}).
2086 Whitespace before a label or after a colon is permitted, but you may not
2087 have whitespace between a label's symbol and its colon. @xref{Labels}.
2090 For HPPA targets, labels need not be immediately followed by a colon, but
2091 the definition of a label must begin in column zero. This also implies that
2092 only one label may be defined on each line.
2096 label: .directive followed by something
2097 another_label: # This is an empty statement.
2098 instruction operand_1, operand_2, @dots{}
2105 A constant is a number, written so that its value is known by
2106 inspection, without knowing any context. Like this:
2109 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2110 .ascii "Ring the bell\7" # A string constant.
2111 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2112 .float 0f-314159265358979323846264338327\
2113 95028841971.693993751E-40 # - pi, a flonum.
2118 * Characters:: Character Constants
2119 * Numbers:: Number Constants
2123 @subsection Character Constants
2125 @cindex character constants
2126 @cindex constants, character
2127 There are two kinds of character constants. A @dfn{character} stands
2128 for one character in one byte and its value may be used in
2129 numeric expressions. String constants (properly called string
2130 @emph{literals}) are potentially many bytes and their values may not be
2131 used in arithmetic expressions.
2135 * Chars:: Characters
2139 @subsubsection Strings
2141 @cindex string constants
2142 @cindex constants, string
2143 A @dfn{string} is written between double-quotes. It may contain
2144 double-quotes or null characters. The way to get special characters
2145 into a string is to @dfn{escape} these characters: precede them with
2146 a backslash @samp{\} character. For example @samp{\\} represents
2147 one backslash: the first @code{\} is an escape which tells
2148 @command{@value{AS}} to interpret the second character literally as a backslash
2149 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2150 escape character). The complete list of escapes follows.
2152 @cindex escape codes, character
2153 @cindex character escape codes
2156 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2158 @cindex @code{\b} (backspace character)
2159 @cindex backspace (@code{\b})
2161 Mnemonic for backspace; for ASCII this is octal code 010.
2164 @c Mnemonic for EOText; for ASCII this is octal code 004.
2166 @cindex @code{\f} (formfeed character)
2167 @cindex formfeed (@code{\f})
2169 Mnemonic for FormFeed; for ASCII this is octal code 014.
2171 @cindex @code{\n} (newline character)
2172 @cindex newline (@code{\n})
2174 Mnemonic for newline; for ASCII this is octal code 012.
2177 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2179 @cindex @code{\r} (carriage return character)
2180 @cindex carriage return (@code{\r})
2182 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2185 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2186 @c other assemblers.
2188 @cindex @code{\t} (tab)
2189 @cindex tab (@code{\t})
2191 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2194 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2195 @c @item \x @var{digit} @var{digit} @var{digit}
2196 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2198 @cindex @code{\@var{ddd}} (octal character code)
2199 @cindex octal character code (@code{\@var{ddd}})
2200 @item \ @var{digit} @var{digit} @var{digit}
2201 An octal character code. The numeric code is 3 octal digits.
2202 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2203 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2205 @cindex @code{\@var{xd...}} (hex character code)
2206 @cindex hex character code (@code{\@var{xd...}})
2207 @item \@code{x} @var{hex-digits...}
2208 A hex character code. All trailing hex digits are combined. Either upper or
2209 lower case @code{x} works.
2211 @cindex @code{\\} (@samp{\} character)
2212 @cindex backslash (@code{\\})
2214 Represents one @samp{\} character.
2217 @c Represents one @samp{'} (accent acute) character.
2218 @c This is needed in single character literals
2219 @c (@xref{Characters,,Character Constants}.) to represent
2222 @cindex @code{\"} (doublequote character)
2223 @cindex doublequote (@code{\"})
2225 Represents one @samp{"} character. Needed in strings to represent
2226 this character, because an unescaped @samp{"} would end the string.
2228 @item \ @var{anything-else}
2229 Any other character when escaped by @kbd{\} gives a warning, but
2230 assembles as if the @samp{\} was not present. The idea is that if
2231 you used an escape sequence you clearly didn't want the literal
2232 interpretation of the following character. However @command{@value{AS}} has no
2233 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2234 code and warns you of the fact.
2237 Which characters are escapable, and what those escapes represent,
2238 varies widely among assemblers. The current set is what we think
2239 the BSD 4.2 assembler recognizes, and is a subset of what most C
2240 compilers recognize. If you are in doubt, do not use an escape
2244 @subsubsection Characters
2246 @cindex single character constant
2247 @cindex character, single
2248 @cindex constant, single character
2249 A single character may be written as a single quote immediately
2250 followed by that character. The same escapes apply to characters as
2251 to strings. So if you want to write the character backslash, you
2252 must write @kbd{'\\} where the first @code{\} escapes the second
2253 @code{\}. As you can see, the quote is an acute accent, not a
2254 grave accent. A newline
2256 @ifclear abnormal-separator
2257 (or semicolon @samp{;})
2259 @ifset abnormal-separator
2261 (or at sign @samp{@@})
2264 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2270 immediately following an acute accent is taken as a literal character
2271 and does not count as the end of a statement. The value of a character
2272 constant in a numeric expression is the machine's byte-wide code for
2273 that character. @command{@value{AS}} assumes your character code is ASCII:
2274 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2277 @subsection Number Constants
2279 @cindex constants, number
2280 @cindex number constants
2281 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2282 are stored in the target machine. @emph{Integers} are numbers that
2283 would fit into an @code{int} in the C language. @emph{Bignums} are
2284 integers, but they are stored in more than 32 bits. @emph{Flonums}
2285 are floating point numbers, described below.
2288 * Integers:: Integers
2293 * Bit Fields:: Bit Fields
2299 @subsubsection Integers
2301 @cindex constants, integer
2303 @cindex binary integers
2304 @cindex integers, binary
2305 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2306 the binary digits @samp{01}.
2308 @cindex octal integers
2309 @cindex integers, octal
2310 An octal integer is @samp{0} followed by zero or more of the octal
2311 digits (@samp{01234567}).
2313 @cindex decimal integers
2314 @cindex integers, decimal
2315 A decimal integer starts with a non-zero digit followed by zero or
2316 more digits (@samp{0123456789}).
2318 @cindex hexadecimal integers
2319 @cindex integers, hexadecimal
2320 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2321 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2323 Integers have the usual values. To denote a negative integer, use
2324 the prefix operator @samp{-} discussed under expressions
2325 (@pxref{Prefix Ops,,Prefix Operators}).
2328 @subsubsection Bignums
2331 @cindex constants, bignum
2332 A @dfn{bignum} has the same syntax and semantics as an integer
2333 except that the number (or its negative) takes more than 32 bits to
2334 represent in binary. The distinction is made because in some places
2335 integers are permitted while bignums are not.
2338 @subsubsection Flonums
2340 @cindex floating point numbers
2341 @cindex constants, floating point
2343 @cindex precision, floating point
2344 A @dfn{flonum} represents a floating point number. The translation is
2345 indirect: a decimal floating point number from the text is converted by
2346 @command{@value{AS}} to a generic binary floating point number of more than
2347 sufficient precision. This generic floating point number is converted
2348 to a particular computer's floating point format (or formats) by a
2349 portion of @command{@value{AS}} specialized to that computer.
2351 A flonum is written by writing (in order)
2356 (@samp{0} is optional on the HPPA.)
2360 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2362 @kbd{e} is recommended. Case is not important.
2364 @c FIXME: verify if flonum syntax really this vague for most cases
2365 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2366 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2369 On the H8/300, H8/500,
2371 and AMD 29K architectures, the letter must be
2372 one of the letters @samp{DFPRSX} (in upper or lower case).
2374 On the ARC, the letter must be one of the letters @samp{DFRS}
2375 (in upper or lower case).
2377 On the Intel 960 architecture, the letter must be
2378 one of the letters @samp{DFT} (in upper or lower case).
2380 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2384 One of the letters @samp{DFPRSX} (in upper or lower case).
2387 One of the letters @samp{DFRS} (in upper or lower case).
2390 One of the letters @samp{DFPRSX} (in upper or lower case).
2393 The letter @samp{E} (upper case only).
2396 One of the letters @samp{DFT} (in upper or lower case).
2401 An optional sign: either @samp{+} or @samp{-}.
2404 An optional @dfn{integer part}: zero or more decimal digits.
2407 An optional @dfn{fractional part}: @samp{.} followed by zero
2408 or more decimal digits.
2411 An optional exponent, consisting of:
2415 An @samp{E} or @samp{e}.
2416 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2417 @c principle this can perfectly well be different on different targets.
2419 Optional sign: either @samp{+} or @samp{-}.
2421 One or more decimal digits.
2426 At least one of the integer part or the fractional part must be
2427 present. The floating point number has the usual base-10 value.
2429 @command{@value{AS}} does all processing using integers. Flonums are computed
2430 independently of any floating point hardware in the computer running
2431 @command{@value{AS}}.
2435 @c Bit fields are written as a general facility but are also controlled
2436 @c by a conditional-compilation flag---which is as of now (21mar91)
2437 @c turned on only by the i960 config of GAS.
2439 @subsubsection Bit Fields
2442 @cindex constants, bit field
2443 You can also define numeric constants as @dfn{bit fields}.
2444 specify two numbers separated by a colon---
2446 @var{mask}:@var{value}
2449 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2452 The resulting number is then packed
2454 @c this conditional paren in case bit fields turned on elsewhere than 960
2455 (in host-dependent byte order)
2457 into a field whose width depends on which assembler directive has the
2458 bit-field as its argument. Overflow (a result from the bitwise and
2459 requiring more binary digits to represent) is not an error; instead,
2460 more constants are generated, of the specified width, beginning with the
2461 least significant digits.@refill
2463 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2464 @code{.short}, and @code{.word} accept bit-field arguments.
2469 @chapter Sections and Relocation
2474 * Secs Background:: Background
2475 * Ld Sections:: Linker Sections
2476 * As Sections:: Assembler Internal Sections
2477 * Sub-Sections:: Sub-Sections
2481 @node Secs Background
2484 Roughly, a section is a range of addresses, with no gaps; all data
2485 ``in'' those addresses is treated the same for some particular purpose.
2486 For example there may be a ``read only'' section.
2488 @cindex linker, and assembler
2489 @cindex assembler, and linker
2490 The linker @code{@value{LD}} reads many object files (partial programs) and
2491 combines their contents to form a runnable program. When @command{@value{AS}}
2492 emits an object file, the partial program is assumed to start at address 0.
2493 @code{@value{LD}} assigns the final addresses for the partial program, so that
2494 different partial programs do not overlap. This is actually an
2495 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2498 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2499 addresses. These blocks slide to their run-time addresses as rigid
2500 units; their length does not change and neither does the order of bytes
2501 within them. Such a rigid unit is called a @emph{section}. Assigning
2502 run-time addresses to sections is called @dfn{relocation}. It includes
2503 the task of adjusting mentions of object-file addresses so they refer to
2504 the proper run-time addresses.
2506 For the H8/300 and H8/500,
2507 and for the Hitachi SH,
2508 @command{@value{AS}} pads sections if needed to
2509 ensure they end on a word (sixteen bit) boundary.
2512 @cindex standard assembler sections
2513 An object file written by @command{@value{AS}} has at least three sections, any
2514 of which may be empty. These are named @dfn{text}, @dfn{data} and
2519 When it generates COFF output,
2521 @command{@value{AS}} can also generate whatever other named sections you specify
2522 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2523 If you do not use any directives that place output in the @samp{.text}
2524 or @samp{.data} sections, these sections still exist, but are empty.
2529 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2531 @command{@value{AS}} can also generate whatever other named sections you
2532 specify using the @samp{.space} and @samp{.subspace} directives. See
2533 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2534 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2535 assembler directives.
2538 Additionally, @command{@value{AS}} uses different names for the standard
2539 text, data, and bss sections when generating SOM output. Program text
2540 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2541 BSS into @samp{$BSS$}.
2545 Within the object file, the text section starts at address @code{0}, the
2546 data section follows, and the bss section follows the data section.
2549 When generating either SOM or ELF output files on the HPPA, the text
2550 section starts at address @code{0}, the data section at address
2551 @code{0x4000000}, and the bss section follows the data section.
2554 To let @code{@value{LD}} know which data changes when the sections are
2555 relocated, and how to change that data, @command{@value{AS}} also writes to the
2556 object file details of the relocation needed. To perform relocation
2557 @code{@value{LD}} must know, each time an address in the object
2561 Where in the object file is the beginning of this reference to
2564 How long (in bytes) is this reference?
2566 Which section does the address refer to? What is the numeric value of
2568 (@var{address}) @minus{} (@var{start-address of section})?
2571 Is the reference to an address ``Program-Counter relative''?
2574 @cindex addresses, format of
2575 @cindex section-relative addressing
2576 In fact, every address @command{@value{AS}} ever uses is expressed as
2578 (@var{section}) + (@var{offset into section})
2581 Further, most expressions @command{@value{AS}} computes have this section-relative
2584 (For some object formats, such as SOM for the HPPA, some expressions are
2585 symbol-relative instead.)
2588 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2589 @var{N} into section @var{secname}.''
2591 Apart from text, data and bss sections you need to know about the
2592 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2593 addresses in the absolute section remain unchanged. For example, address
2594 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2595 @code{@value{LD}}. Although the linker never arranges two partial programs'
2596 data sections with overlapping addresses after linking, @emph{by definition}
2597 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2598 part of a program is always the same address when the program is running as
2599 address @code{@{absolute@ 239@}} in any other part of the program.
2601 The idea of sections is extended to the @dfn{undefined} section. Any
2602 address whose section is unknown at assembly time is by definition
2603 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2604 Since numbers are always defined, the only way to generate an undefined
2605 address is to mention an undefined symbol. A reference to a named
2606 common block would be such a symbol: its value is unknown at assembly
2607 time so it has section @emph{undefined}.
2609 By analogy the word @emph{section} is used to describe groups of sections in
2610 the linked program. @code{@value{LD}} puts all partial programs' text
2611 sections in contiguous addresses in the linked program. It is
2612 customary to refer to the @emph{text section} of a program, meaning all
2613 the addresses of all partial programs' text sections. Likewise for
2614 data and bss sections.
2616 Some sections are manipulated by @code{@value{LD}}; others are invented for
2617 use of @command{@value{AS}} and have no meaning except during assembly.
2620 @section Linker Sections
2621 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2626 @cindex named sections
2627 @cindex sections, named
2628 @item named sections
2631 @cindex text section
2632 @cindex data section
2636 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2637 separate but equal sections. Anything you can say of one section is
2640 When the program is running, however, it is
2641 customary for the text section to be unalterable. The
2642 text section is often shared among processes: it contains
2643 instructions, constants and the like. The data section of a running
2644 program is usually alterable: for example, C variables would be stored
2645 in the data section.
2650 This section contains zeroed bytes when your program begins running. It
2651 is used to hold uninitialized variables or common storage. The length of
2652 each partial program's bss section is important, but because it starts
2653 out containing zeroed bytes there is no need to store explicit zero
2654 bytes in the object file. The bss section was invented to eliminate
2655 those explicit zeros from object files.
2657 @cindex absolute section
2658 @item absolute section
2659 Address 0 of this section is always ``relocated'' to runtime address 0.
2660 This is useful if you want to refer to an address that @code{@value{LD}} must
2661 not change when relocating. In this sense we speak of absolute
2662 addresses being ``unrelocatable'': they do not change during relocation.
2664 @cindex undefined section
2665 @item undefined section
2666 This ``section'' is a catch-all for address references to objects not in
2667 the preceding sections.
2668 @c FIXME: ref to some other doc on obj-file formats could go here.
2671 @cindex relocation example
2672 An idealized example of three relocatable sections follows.
2674 The example uses the traditional section names @samp{.text} and @samp{.data}.
2676 Memory addresses are on the horizontal axis.
2680 @c END TEXI2ROFF-KILL
2683 partial program # 1: |ttttt|dddd|00|
2690 partial program # 2: |TTT|DDD|000|
2693 +--+---+-----+--+----+---+-----+~~
2694 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2695 +--+---+-----+--+----+---+-----+~~
2697 addresses: 0 @dots{}
2704 \line{\it Partial program \#1: \hfil}
2705 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2706 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2708 \line{\it Partial program \#2: \hfil}
2709 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2710 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2712 \line{\it linked program: \hfil}
2713 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2714 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2715 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2716 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2718 \line{\it addresses: \hfil}
2722 @c END TEXI2ROFF-KILL
2725 @section Assembler Internal Sections
2727 @cindex internal assembler sections
2728 @cindex sections in messages, internal
2729 These sections are meant only for the internal use of @command{@value{AS}}. They
2730 have no meaning at run-time. You do not really need to know about these
2731 sections for most purposes; but they can be mentioned in @command{@value{AS}}
2732 warning messages, so it might be helpful to have an idea of their
2733 meanings to @command{@value{AS}}. These sections are used to permit the
2734 value of every expression in your assembly language program to be a
2735 section-relative address.
2738 @cindex assembler internal logic error
2739 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2740 An internal assembler logic error has been found. This means there is a
2741 bug in the assembler.
2743 @cindex expr (internal section)
2745 The assembler stores complex expression internally as combinations of
2746 symbols. When it needs to represent an expression as a symbol, it puts
2747 it in the expr section.
2749 @c FIXME item transfer[t] vector preload
2750 @c FIXME item transfer[t] vector postload
2751 @c FIXME item register
2755 @section Sub-Sections
2757 @cindex numbered subsections
2758 @cindex grouping data
2764 fall into two sections: text and data.
2766 You may have separate groups of
2768 data in named sections
2772 data in named sections
2778 that you want to end up near to each other in the object file, even though they
2779 are not contiguous in the assembler source. @command{@value{AS}} allows you to
2780 use @dfn{subsections} for this purpose. Within each section, there can be
2781 numbered subsections with values from 0 to 8192. Objects assembled into the
2782 same subsection go into the object file together with other objects in the same
2783 subsection. For example, a compiler might want to store constants in the text
2784 section, but might not want to have them interspersed with the program being
2785 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2786 section of code being output, and a @samp{.text 1} before each group of
2787 constants being output.
2789 Subsections are optional. If you do not use subsections, everything
2790 goes in subsection number zero.
2793 Each subsection is zero-padded up to a multiple of four bytes.
2794 (Subsections may be padded a different amount on different flavors
2795 of @command{@value{AS}}.)
2799 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2800 boundary (two bytes).
2801 The same is true on the Hitachi SH.
2804 @c FIXME section padding (alignment)?
2805 @c Rich Pixley says padding here depends on target obj code format; that
2806 @c doesn't seem particularly useful to say without further elaboration,
2807 @c so for now I say nothing about it. If this is a generic BFD issue,
2808 @c these paragraphs might need to vanish from this manual, and be
2809 @c discussed in BFD chapter of binutils (or some such).
2812 On the AMD 29K family, no particular padding is added to section or
2813 subsection sizes; @value{AS} forces no alignment on this platform.
2817 Subsections appear in your object file in numeric order, lowest numbered
2818 to highest. (All this to be compatible with other people's assemblers.)
2819 The object file contains no representation of subsections; @code{@value{LD}} and
2820 other programs that manipulate object files see no trace of them.
2821 They just see all your text subsections as a text section, and all your
2822 data subsections as a data section.
2824 To specify which subsection you want subsequent statements assembled
2825 into, use a numeric argument to specify it, in a @samp{.text
2826 @var{expression}} or a @samp{.data @var{expression}} statement.
2829 When generating COFF output, you
2834 can also use an extra subsection
2835 argument with arbitrary named sections: @samp{.section @var{name},
2838 @var{Expression} should be an absolute expression.
2839 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
2840 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
2841 begins in @code{text 0}. For instance:
2843 .text 0 # The default subsection is text 0 anyway.
2844 .ascii "This lives in the first text subsection. *"
2846 .ascii "But this lives in the second text subsection."
2848 .ascii "This lives in the data section,"
2849 .ascii "in the first data subsection."
2851 .ascii "This lives in the first text section,"
2852 .ascii "immediately following the asterisk (*)."
2855 Each section has a @dfn{location counter} incremented by one for every byte
2856 assembled into that section. Because subsections are merely a convenience
2857 restricted to @command{@value{AS}} there is no concept of a subsection location
2858 counter. There is no way to directly manipulate a location counter---but the
2859 @code{.align} directive changes it, and any label definition captures its
2860 current value. The location counter of the section where statements are being
2861 assembled is said to be the @dfn{active} location counter.
2864 @section bss Section
2867 @cindex common variable storage
2868 The bss section is used for local common variable storage.
2869 You may allocate address space in the bss section, but you may
2870 not dictate data to load into it before your program executes. When
2871 your program starts running, all the contents of the bss
2872 section are zeroed bytes.
2874 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
2875 @ref{Lcomm,,@code{.lcomm}}.
2877 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
2878 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
2881 When assembling for a target which supports multiple sections, such as ELF or
2882 COFF, you may switch into the @code{.bss} section and define symbols as usual;
2883 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
2884 section. Typically the section will only contain symbol definitions and
2885 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
2892 Symbols are a central concept: the programmer uses symbols to name
2893 things, the linker uses symbols to link, and the debugger uses symbols
2897 @cindex debuggers, and symbol order
2898 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
2899 the same order they were declared. This may break some debuggers.
2904 * Setting Symbols:: Giving Symbols Other Values
2905 * Symbol Names:: Symbol Names
2906 * Dot:: The Special Dot Symbol
2907 * Symbol Attributes:: Symbol Attributes
2914 A @dfn{label} is written as a symbol immediately followed by a colon
2915 @samp{:}. The symbol then represents the current value of the
2916 active location counter, and is, for example, a suitable instruction
2917 operand. You are warned if you use the same symbol to represent two
2918 different locations: the first definition overrides any other
2922 On the HPPA, the usual form for a label need not be immediately followed by a
2923 colon, but instead must start in column zero. Only one label may be defined on
2924 a single line. To work around this, the HPPA version of @command{@value{AS}} also
2925 provides a special directive @code{.label} for defining labels more flexibly.
2928 @node Setting Symbols
2929 @section Giving Symbols Other Values
2931 @cindex assigning values to symbols
2932 @cindex symbol values, assigning
2933 A symbol can be given an arbitrary value by writing a symbol, followed
2934 by an equals sign @samp{=}, followed by an expression
2935 (@pxref{Expressions}). This is equivalent to using the @code{.set}
2936 directive. @xref{Set,,@code{.set}}.
2939 @section Symbol Names
2941 @cindex symbol names
2942 @cindex names, symbol
2943 @ifclear SPECIAL-SYMS
2944 Symbol names begin with a letter or with one of @samp{._}. On most
2945 machines, you can also use @code{$} in symbol names; exceptions are
2946 noted in @ref{Machine Dependencies}. That character may be followed by any
2947 string of digits, letters, dollar signs (unless otherwise noted in
2948 @ref{Machine Dependencies}), and underscores.
2951 For the AMD 29K family, @samp{?} is also allowed in the
2952 body of a symbol name, though not at its beginning.
2957 Symbol names begin with a letter or with one of @samp{._}. On the
2959 H8/500, you can also use @code{$} in symbol names. That character may
2960 be followed by any string of digits, letters, dollar signs (save on the
2961 H8/300), and underscores.
2965 Case of letters is significant: @code{foo} is a different symbol name
2968 Each symbol has exactly one name. Each name in an assembly language program
2969 refers to exactly one symbol. You may use that symbol name any number of times
2972 @subheading Local Symbol Names
2974 @cindex local symbol names
2975 @cindex symbol names, local
2976 @cindex temporary symbol names
2977 @cindex symbol names, temporary
2978 Local symbols help compilers and programmers use names temporarily.
2979 They create symbols which are guaranteed to be unique over the entire scope of
2980 the input source code and which can be referred to by a simple notation.
2981 To define a local symbol, write a label of the form @samp{@b{N}:} (where @b{N}
2982 represents any positive integer). To refer to the most recent previous
2983 definition of that symbol write @samp{@b{N}b}, using the same number as when
2984 you defined the label. To refer to the next definition of a local label, write
2985 @samp{@b{N}f}--- The @samp{b} stands for``backwards'' and the @samp{f} stands
2988 There is no restriction on how you can use these labels, and you can reuse them
2989 too. So that it is possible to repeatedly define the same local label (using
2990 the same number @samp{@b{N}}), although you can only refer to the most recently
2991 defined local label of that number (for a backwards reference) or the next
2992 definition of a specific local label for a forward reference. It is also worth
2993 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
2994 implemented in a slightly more efficient manner than the others.
3005 Which is the equivalent of:
3008 label_1: branch label_3
3009 label_2: branch label_1
3010 label_3: branch label_4
3011 label_4: branch label_3
3014 Local symbol names are only a notational device. They are immediately
3015 transformed into more conventional symbol names before the assembler uses them.
3016 The symbol names stored in the symbol table, appearing in error messages and
3017 optionally emitted to the object file. The names are constructed using these
3022 All local labels begin with @samp{L}. Normally both @command{@value{AS}} and
3023 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
3024 used for symbols you are never intended to see. If you use the
3025 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3026 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3027 you may use them in debugging.
3030 This is the number that was used in the local label definition. So if the
3031 label is written @samp{55:} then the number is @samp{55}.
3034 This unusual character is included so you do not accidentally invent a symbol
3035 of the same name. The character has ASCII value of @samp{\002} (control-B).
3037 @item @emph{ordinal number}
3038 This is a serial number to keep the labels distinct. The first definition of
3039 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3040 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3041 the number @samp{1} and its 15th defintion gets @samp{15} as well.
3044 So for example, the first @code{1:} is named @code{L1@kbd{C-B}1}, the 44th
3045 @code{3:} is named @code{L3@kbd{C-B}44}.
3047 @subheading Dollar Local Labels
3048 @cindex dollar local symbols
3050 @code{@value{AS}} also supports an even more local form of local labels called
3051 dollar labels. These labels go out of scope (ie they become undefined) as soon
3052 as a non-local label is defined. Thus they remain valid for only a small
3053 region of the input source code. Normal local labels, by contrast, remain in
3054 scope for the entire file, or until they are redefined by another occurrence of
3055 the same local label.
3057 Dollar labels are defined in exactly the same way as ordinary local labels,
3058 except that instead of being terminated by a colon, they are terminated by a
3059 dollar sign. eg @samp{@b{55$}}.
3061 They can also be distinguished from ordinary local labels by their transformed
3062 name which uses ASCII character @samp{\001} (control-A) as the magic character
3063 to distinguish them from ordinary labels. Thus the 5th defintion of @samp{6$}
3064 is named @samp{L6@kbd{C-A}5}.
3067 @section The Special Dot Symbol
3069 @cindex dot (symbol)
3070 @cindex @code{.} (symbol)
3071 @cindex current address
3072 @cindex location counter
3073 The special symbol @samp{.} refers to the current address that
3074 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3075 .long .} defines @code{melvin} to contain its own address.
3076 Assigning a value to @code{.} is treated the same as a @code{.org}
3077 directive. Thus, the expression @samp{.=.+4} is the same as saying
3078 @ifclear no-space-dir
3087 @node Symbol Attributes
3088 @section Symbol Attributes
3090 @cindex symbol attributes
3091 @cindex attributes, symbol
3092 Every symbol has, as well as its name, the attributes ``Value'' and
3093 ``Type''. Depending on output format, symbols can also have auxiliary
3096 The detailed definitions are in @file{a.out.h}.
3099 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3100 all these attributes, and probably won't warn you. This makes the
3101 symbol an externally defined symbol, which is generally what you
3105 * Symbol Value:: Value
3106 * Symbol Type:: Type
3109 * a.out Symbols:: Symbol Attributes: @code{a.out}
3113 * a.out Symbols:: Symbol Attributes: @code{a.out}
3116 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3121 * COFF Symbols:: Symbol Attributes for COFF
3124 * SOM Symbols:: Symbol Attributes for SOM
3131 @cindex value of a symbol
3132 @cindex symbol value
3133 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3134 location in the text, data, bss or absolute sections the value is the
3135 number of addresses from the start of that section to the label.
3136 Naturally for text, data and bss sections the value of a symbol changes
3137 as @code{@value{LD}} changes section base addresses during linking. Absolute
3138 symbols' values do not change during linking: that is why they are
3141 The value of an undefined symbol is treated in a special way. If it is
3142 0 then the symbol is not defined in this assembler source file, and
3143 @code{@value{LD}} tries to determine its value from other files linked into the
3144 same program. You make this kind of symbol simply by mentioning a symbol
3145 name without defining it. A non-zero value represents a @code{.comm}
3146 common declaration. The value is how much common storage to reserve, in
3147 bytes (addresses). The symbol refers to the first address of the
3153 @cindex type of a symbol
3155 The type attribute of a symbol contains relocation (section)
3156 information, any flag settings indicating that a symbol is external, and
3157 (optionally), other information for linkers and debuggers. The exact
3158 format depends on the object-code output format in use.
3163 @c The following avoids a "widow" subsection title. @group would be
3164 @c better if it were available outside examples.
3167 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3169 @cindex @code{b.out} symbol attributes
3170 @cindex symbol attributes, @code{b.out}
3171 These symbol attributes appear only when @command{@value{AS}} is configured for
3172 one of the Berkeley-descended object output formats---@code{a.out} or
3178 @subsection Symbol Attributes: @code{a.out}
3180 @cindex @code{a.out} symbol attributes
3181 @cindex symbol attributes, @code{a.out}
3187 @subsection Symbol Attributes: @code{a.out}
3189 @cindex @code{a.out} symbol attributes
3190 @cindex symbol attributes, @code{a.out}
3194 * Symbol Desc:: Descriptor
3195 * Symbol Other:: Other
3199 @subsubsection Descriptor
3201 @cindex descriptor, of @code{a.out} symbol
3202 This is an arbitrary 16-bit value. You may establish a symbol's
3203 descriptor value by using a @code{.desc} statement
3204 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3205 @command{@value{AS}}.
3208 @subsubsection Other
3210 @cindex other attribute, of @code{a.out} symbol
3211 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3216 @subsection Symbol Attributes for COFF
3218 @cindex COFF symbol attributes
3219 @cindex symbol attributes, COFF
3221 The COFF format supports a multitude of auxiliary symbol attributes;
3222 like the primary symbol attributes, they are set between @code{.def} and
3223 @code{.endef} directives.
3225 @subsubsection Primary Attributes
3227 @cindex primary attributes, COFF symbols
3228 The symbol name is set with @code{.def}; the value and type,
3229 respectively, with @code{.val} and @code{.type}.
3231 @subsubsection Auxiliary Attributes
3233 @cindex auxiliary attributes, COFF symbols
3234 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3235 @code{.size}, and @code{.tag} can generate auxiliary symbol table
3236 information for COFF.
3241 @subsection Symbol Attributes for SOM
3243 @cindex SOM symbol attributes
3244 @cindex symbol attributes, SOM
3246 The SOM format for the HPPA supports a multitude of symbol attributes set with
3247 the @code{.EXPORT} and @code{.IMPORT} directives.
3249 The attributes are described in @cite{HP9000 Series 800 Assembly
3250 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3251 @code{EXPORT} assembler directive documentation.
3255 @chapter Expressions
3259 @cindex numeric values
3260 An @dfn{expression} specifies an address or numeric value.
3261 Whitespace may precede and/or follow an expression.
3263 The result of an expression must be an absolute number, or else an offset into
3264 a particular section. If an expression is not absolute, and there is not
3265 enough information when @command{@value{AS}} sees the expression to know its
3266 section, a second pass over the source program might be necessary to interpret
3267 the expression---but the second pass is currently not implemented.
3268 @command{@value{AS}} aborts with an error message in this situation.
3271 * Empty Exprs:: Empty Expressions
3272 * Integer Exprs:: Integer Expressions
3276 @section Empty Expressions
3278 @cindex empty expressions
3279 @cindex expressions, empty
3280 An empty expression has no value: it is just whitespace or null.
3281 Wherever an absolute expression is required, you may omit the
3282 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3283 is compatible with other assemblers.
3286 @section Integer Expressions
3288 @cindex integer expressions
3289 @cindex expressions, integer
3290 An @dfn{integer expression} is one or more @emph{arguments} delimited
3291 by @emph{operators}.
3294 * Arguments:: Arguments
3295 * Operators:: Operators
3296 * Prefix Ops:: Prefix Operators
3297 * Infix Ops:: Infix Operators
3301 @subsection Arguments
3303 @cindex expression arguments
3304 @cindex arguments in expressions
3305 @cindex operands in expressions
3306 @cindex arithmetic operands
3307 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3308 contexts arguments are sometimes called ``arithmetic operands''. In
3309 this manual, to avoid confusing them with the ``instruction operands'' of
3310 the machine language, we use the term ``argument'' to refer to parts of
3311 expressions only, reserving the word ``operand'' to refer only to machine
3312 instruction operands.
3314 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3315 @var{section} is one of text, data, bss, absolute,
3316 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3319 Numbers are usually integers.
3321 A number can be a flonum or bignum. In this case, you are warned
3322 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3323 these 32 bits are an integer. You may write integer-manipulating
3324 instructions that act on exotic constants, compatible with other
3327 @cindex subexpressions
3328 Subexpressions are a left parenthesis @samp{(} followed by an integer
3329 expression, followed by a right parenthesis @samp{)}; or a prefix
3330 operator followed by an argument.
3333 @subsection Operators
3335 @cindex operators, in expressions
3336 @cindex arithmetic functions
3337 @cindex functions, in expressions
3338 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3339 operators are followed by an argument. Infix operators appear
3340 between their arguments. Operators may be preceded and/or followed by
3344 @subsection Prefix Operator
3346 @cindex prefix operators
3347 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3348 one argument, which must be absolute.
3350 @c the tex/end tex stuff surrounding this small table is meant to make
3351 @c it align, on the printed page, with the similar table in the next
3352 @c section (which is inside an enumerate).
3354 \global\advance\leftskip by \itemindent
3359 @dfn{Negation}. Two's complement negation.
3361 @dfn{Complementation}. Bitwise not.
3365 \global\advance\leftskip by -\itemindent
3369 @subsection Infix Operators
3371 @cindex infix operators
3372 @cindex operators, permitted arguments
3373 @dfn{Infix operators} take two arguments, one on either side. Operators
3374 have precedence, but operations with equal precedence are performed left
3375 to right. Apart from @code{+} or @option{-}, both arguments must be
3376 absolute, and the result is absolute.
3379 @cindex operator precedence
3380 @cindex precedence of operators
3387 @dfn{Multiplication}.
3390 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3397 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3401 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3405 Intermediate precedence
3410 @dfn{Bitwise Inclusive Or}.
3416 @dfn{Bitwise Exclusive Or}.
3419 @dfn{Bitwise Or Not}.
3426 @cindex addition, permitted arguments
3427 @cindex plus, permitted arguments
3428 @cindex arguments for addition
3430 @dfn{Addition}. If either argument is absolute, the result has the section of
3431 the other argument. You may not add together arguments from different
3434 @cindex subtraction, permitted arguments
3435 @cindex minus, permitted arguments
3436 @cindex arguments for subtraction
3438 @dfn{Subtraction}. If the right argument is absolute, the
3439 result has the section of the left argument.
3440 If both arguments are in the same section, the result is absolute.
3441 You may not subtract arguments from different sections.
3442 @c FIXME is there still something useful to say about undefined - undefined ?
3444 @cindex comparison expressions
3445 @cindex expressions, comparison
3449 @dfn{Is Not Equal To}
3453 @dfn{Is Greater Than}
3455 @dfn{Is Greater Than Or Equal To}
3457 @dfn{Is Less Than Or Equal To}
3459 The comparison operators can be used as infix operators. A true results has a
3460 value of -1 whereas a false result has a value of 0. Note, these operators
3461 perform signed comparisons.
3464 @item Lowest Precedence
3473 These two logical operations can be used to combine the results of sub
3474 expressions. Note, unlike the comparison operators a true result returns a
3475 value of 1 but a false results does still return 0. Also note that the logical
3476 or operator has a slightly lower precedence than logical and.
3481 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3482 address; you can only have a defined section in one of the two arguments.
3485 @chapter Assembler Directives
3487 @cindex directives, machine independent
3488 @cindex pseudo-ops, machine independent
3489 @cindex machine independent directives
3490 All assembler directives have names that begin with a period (@samp{.}).
3491 The rest of the name is letters, usually in lower case.
3493 This chapter discusses directives that are available regardless of the
3494 target machine configuration for the @sc{gnu} assembler.
3496 Some machine configurations provide additional directives.
3497 @xref{Machine Dependencies}.
3500 @ifset machine-directives
3501 @xref{Machine Dependencies} for additional directives.
3506 * Abort:: @code{.abort}
3508 * ABORT:: @code{.ABORT}
3511 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3512 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3513 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3514 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3515 * Byte:: @code{.byte @var{expressions}}
3516 * Comm:: @code{.comm @var{symbol} , @var{length} }
3517 * Data:: @code{.data @var{subsection}}
3519 * Def:: @code{.def @var{name}}
3522 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3528 * Double:: @code{.double @var{flonums}}
3529 * Eject:: @code{.eject}
3530 * Else:: @code{.else}
3531 * Elseif:: @code{.elseif}
3534 * Endef:: @code{.endef}
3537 * Endfunc:: @code{.endfunc}
3538 * Endif:: @code{.endif}
3539 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3540 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3542 * Exitm:: @code{.exitm}
3543 * Extern:: @code{.extern}
3544 * Fail:: @code{.fail}
3545 @ifclear no-file-dir
3546 * File:: @code{.file @var{string}}
3549 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3550 * Float:: @code{.float @var{flonums}}
3551 * Func:: @code{.func}
3552 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3554 * Hidden:: @code{.hidden @var{names}}
3557 * hword:: @code{.hword @var{expressions}}
3558 * Ident:: @code{.ident}
3559 * If:: @code{.if @var{absolute expression}}
3560 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3561 * Include:: @code{.include "@var{file}"}
3562 * Int:: @code{.int @var{expressions}}
3564 * Internal:: @code{.internal @var{names}}
3567 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3568 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3569 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3570 * Lflags:: @code{.lflags}
3571 @ifclear no-line-dir
3572 * Line:: @code{.line @var{line-number}}
3575 * Ln:: @code{.ln @var{line-number}}
3576 * Linkonce:: @code{.linkonce [@var{type}]}
3577 * List:: @code{.list}
3578 * Long:: @code{.long @var{expressions}}
3580 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3583 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3584 * MRI:: @code{.mri @var{val}}
3585 * Nolist:: @code{.nolist}
3586 * Octa:: @code{.octa @var{bignums}}
3587 * Org:: @code{.org @var{new-lc} , @var{fill}}
3588 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3590 * PopSection:: @code{.popsection}
3591 * Previous:: @code{.previous}
3594 * Print:: @code{.print @var{string}}
3596 * Protected:: @code{.protected @var{names}}
3599 * Psize:: @code{.psize @var{lines}, @var{columns}}
3600 * Purgem:: @code{.purgem @var{name}}
3602 * PushSection:: @code{.pushsection @var{name}}
3605 * Quad:: @code{.quad @var{bignums}}
3606 * Rept:: @code{.rept @var{count}}
3607 * Sbttl:: @code{.sbttl "@var{subheading}"}
3609 * Scl:: @code{.scl @var{class}}
3610 * Section:: @code{.section @var{name}, @var{subsection}}
3613 * Set:: @code{.set @var{symbol}, @var{expression}}
3614 * Short:: @code{.short @var{expressions}}
3615 * Single:: @code{.single @var{flonums}}
3616 * Size:: @code{.size [@var{name} , @var{expression}]}
3617 * Skip:: @code{.skip @var{size} , @var{fill}}
3618 * Sleb128:: @code{.sleb128 @var{expressions}}
3619 * Space:: @code{.space @var{size} , @var{fill}}
3621 * Stab:: @code{.stabd, .stabn, .stabs}
3624 * String:: @code{.string "@var{str}"}
3625 * Struct:: @code{.struct @var{expression}}
3627 * SubSection:: @code{.subsection}
3628 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3632 * Tag:: @code{.tag @var{structname}}
3635 * Text:: @code{.text @var{subsection}}
3636 * Title:: @code{.title "@var{heading}"}
3637 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3638 * Uleb128:: @code{.uleb128 @var{expressions}}
3640 * Val:: @code{.val @var{addr}}
3644 * Version:: @code{.version "@var{string}"}
3645 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3646 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3647 * Weak:: @code{.weak @var{names}}
3650 * Word:: @code{.word @var{expressions}}
3651 * Deprecated:: Deprecated Directives
3655 @section @code{.abort}
3657 @cindex @code{abort} directive
3658 @cindex stopping the assembly
3659 This directive stops the assembly immediately. It is for
3660 compatibility with other assemblers. The original idea was that the
3661 assembly language source would be piped into the assembler. If the sender
3662 of the source quit, it could use this directive tells @command{@value{AS}} to
3663 quit also. One day @code{.abort} will not be supported.
3667 @section @code{.ABORT}
3669 @cindex @code{ABORT} directive
3670 When producing COFF output, @command{@value{AS}} accepts this directive as a
3671 synonym for @samp{.abort}.
3674 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
3680 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3682 @cindex padding the location counter
3683 @cindex @code{align} directive
3684 Pad the location counter (in the current subsection) to a particular storage
3685 boundary. The first expression (which must be absolute) is the alignment
3686 required, as described below.
3688 The second expression (also absolute) gives the fill value to be stored in the
3689 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3690 padding bytes are normally zero. However, on some systems, if the section is
3691 marked as containing code and the fill value is omitted, the space is filled
3692 with no-op instructions.
3694 The third expression is also absolute, and is also optional. If it is present,
3695 it is the maximum number of bytes that should be skipped by this alignment
3696 directive. If doing the alignment would require skipping more bytes than the
3697 specified maximum, then the alignment is not done at all. You can omit the
3698 fill value (the second argument) entirely by simply using two commas after the
3699 required alignment; this can be useful if you want the alignment to be filled
3700 with no-op instructions when appropriate.
3702 The way the required alignment is specified varies from system to system.
3703 For the a29k, hppa, m68k, m88k, w65, sparc, and Hitachi SH, and i386 using ELF
3705 the first expression is the
3706 alignment request in bytes. For example @samp{.align 8} advances
3707 the location counter until it is a multiple of 8. If the location counter
3708 is already a multiple of 8, no change is needed.
3710 For other systems, including the i386 using a.out format, and the arm and
3711 strongarm, it is the
3712 number of low-order zero bits the location counter must have after
3713 advancement. For example @samp{.align 3} advances the location
3714 counter until it a multiple of 8. If the location counter is already a
3715 multiple of 8, no change is needed.
3717 This inconsistency is due to the different behaviors of the various
3718 native assemblers for these systems which GAS must emulate.
3719 GAS also provides @code{.balign} and @code{.p2align} directives,
3720 described later, which have a consistent behavior across all
3721 architectures (but are specific to GAS).
3724 @section @code{.ascii "@var{string}"}@dots{}
3726 @cindex @code{ascii} directive
3727 @cindex string literals
3728 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3729 separated by commas. It assembles each string (with no automatic
3730 trailing zero byte) into consecutive addresses.
3733 @section @code{.asciz "@var{string}"}@dots{}
3735 @cindex @code{asciz} directive
3736 @cindex zero-terminated strings
3737 @cindex null-terminated strings
3738 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3739 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3742 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3744 @cindex padding the location counter given number of bytes
3745 @cindex @code{balign} directive
3746 Pad the location counter (in the current subsection) to a particular
3747 storage boundary. The first expression (which must be absolute) is the
3748 alignment request in bytes. For example @samp{.balign 8} advances
3749 the location counter until it is a multiple of 8. If the location counter
3750 is already a multiple of 8, no change is needed.
3752 The second expression (also absolute) gives the fill value to be stored in the
3753 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3754 padding bytes are normally zero. However, on some systems, if the section is
3755 marked as containing code and the fill value is omitted, the space is filled
3756 with no-op instructions.
3758 The third expression is also absolute, and is also optional. If it is present,
3759 it is the maximum number of bytes that should be skipped by this alignment
3760 directive. If doing the alignment would require skipping more bytes than the
3761 specified maximum, then the alignment is not done at all. You can omit the
3762 fill value (the second argument) entirely by simply using two commas after the
3763 required alignment; this can be useful if you want the alignment to be filled
3764 with no-op instructions when appropriate.
3766 @cindex @code{balignw} directive
3767 @cindex @code{balignl} directive
3768 The @code{.balignw} and @code{.balignl} directives are variants of the
3769 @code{.balign} directive. The @code{.balignw} directive treats the fill
3770 pattern as a two byte word value. The @code{.balignl} directives treats the
3771 fill pattern as a four byte longword value. For example, @code{.balignw
3772 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3773 filled in with the value 0x368d (the exact placement of the bytes depends upon
3774 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3778 @section @code{.byte @var{expressions}}
3780 @cindex @code{byte} directive
3781 @cindex integers, one byte
3782 @code{.byte} expects zero or more expressions, separated by commas.
3783 Each expression is assembled into the next byte.
3786 @section @code{.comm @var{symbol} , @var{length} }
3788 @cindex @code{comm} directive
3789 @cindex symbol, common
3790 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
3791 common symbol in one object file may be merged with a defined or common symbol
3792 of the same name in another object file. If @code{@value{LD}} does not see a
3793 definition for the symbol--just one or more common symbols--then it will
3794 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
3795 absolute expression. If @code{@value{LD}} sees multiple common symbols with
3796 the same name, and they do not all have the same size, it will allocate space
3797 using the largest size.
3800 When using ELF, the @code{.comm} directive takes an optional third argument.
3801 This is the desired alignment of the symbol, specified as a byte boundary (for
3802 example, an alignment of 16 means that the least significant 4 bits of the
3803 address should be zero). The alignment must be an absolute expression, and it
3804 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
3805 for the common symbol, it will use the alignment when placing the symbol. If
3806 no alignment is specified, @command{@value{AS}} will set the alignment to the
3807 largest power of two less than or equal to the size of the symbol, up to a
3812 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
3813 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
3817 @section @code{.data @var{subsection}}
3819 @cindex @code{data} directive
3820 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
3821 end of the data subsection numbered @var{subsection} (which is an
3822 absolute expression). If @var{subsection} is omitted, it defaults
3827 @section @code{.def @var{name}}
3829 @cindex @code{def} directive
3830 @cindex COFF symbols, debugging
3831 @cindex debugging COFF symbols
3832 Begin defining debugging information for a symbol @var{name}; the
3833 definition extends until the @code{.endef} directive is encountered.
3836 This directive is only observed when @command{@value{AS}} is configured for COFF
3837 format output; when producing @code{b.out}, @samp{.def} is recognized,
3844 @section @code{.desc @var{symbol}, @var{abs-expression}}
3846 @cindex @code{desc} directive
3847 @cindex COFF symbol descriptor
3848 @cindex symbol descriptor, COFF
3849 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
3850 to the low 16 bits of an absolute expression.
3853 The @samp{.desc} directive is not available when @command{@value{AS}} is
3854 configured for COFF output; it is only for @code{a.out} or @code{b.out}
3855 object format. For the sake of compatibility, @command{@value{AS}} accepts
3856 it, but produces no output, when configured for COFF.
3862 @section @code{.dim}
3864 @cindex @code{dim} directive
3865 @cindex COFF auxiliary symbol information
3866 @cindex auxiliary symbol information, COFF
3867 This directive is generated by compilers to include auxiliary debugging
3868 information in the symbol table. It is only permitted inside
3869 @code{.def}/@code{.endef} pairs.
3872 @samp{.dim} is only meaningful when generating COFF format output; when
3873 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
3879 @section @code{.double @var{flonums}}
3881 @cindex @code{double} directive
3882 @cindex floating point numbers (double)
3883 @code{.double} expects zero or more flonums, separated by commas. It
3884 assembles floating point numbers.
3886 The exact kind of floating point numbers emitted depends on how
3887 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
3891 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
3892 in @sc{ieee} format.
3897 @section @code{.eject}
3899 @cindex @code{eject} directive
3900 @cindex new page, in listings
3901 @cindex page, in listings
3902 @cindex listing control: new page
3903 Force a page break at this point, when generating assembly listings.
3906 @section @code{.else}
3908 @cindex @code{else} directive
3909 @code{.else} is part of the @command{@value{AS}} support for conditional
3910 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
3911 of code to be assembled if the condition for the preceding @code{.if}
3915 @section @code{.elseif}
3917 @cindex @code{elseif} directive
3918 @code{.elseif} is part of the @command{@value{AS}} support for conditional
3919 assembly; @pxref{If,,@code{.if}}. It is shorthand for beginning a new
3920 @code{.if} block that would otherwise fill the entire @code{.else} section.
3923 @section @code{.end}
3925 @cindex @code{end} directive
3926 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
3927 process anything in the file past the @code{.end} directive.
3931 @section @code{.endef}
3933 @cindex @code{endef} directive
3934 This directive flags the end of a symbol definition begun with
3938 @samp{.endef} is only meaningful when generating COFF format output; if
3939 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
3940 directive but ignores it.
3945 @section @code{.endfunc}
3946 @cindex @code{endfunc} directive
3947 @code{.endfunc} marks the end of a function specified with @code{.func}.
3950 @section @code{.endif}
3952 @cindex @code{endif} directive
3953 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
3954 it marks the end of a block of code that is only assembled
3955 conditionally. @xref{If,,@code{.if}}.
3958 @section @code{.equ @var{symbol}, @var{expression}}
3960 @cindex @code{equ} directive
3961 @cindex assigning values to symbols
3962 @cindex symbols, assigning values to
3963 This directive sets the value of @var{symbol} to @var{expression}.
3964 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
3967 The syntax for @code{equ} on the HPPA is
3968 @samp{@var{symbol} .equ @var{expression}}.
3972 @section @code{.equiv @var{symbol}, @var{expression}}
3973 @cindex @code{equiv} directive
3974 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
3975 the assembler will signal an error if @var{symbol} is already defined.
3977 Except for the contents of the error message, this is roughly equivalent to
3986 @section @code{.err}
3987 @cindex @code{err} directive
3988 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
3989 message and, unless the @option{-Z} option was used, it will not generate an
3990 object file. This can be used to signal error an conditionally compiled code.
3993 @section @code{.exitm}
3994 Exit early from the current macro definition. @xref{Macro}.
3997 @section @code{.extern}
3999 @cindex @code{extern} directive
4000 @code{.extern} is accepted in the source program---for compatibility
4001 with other assemblers---but it is ignored. @command{@value{AS}} treats
4002 all undefined symbols as external.
4005 @section @code{.fail @var{expression}}
4007 @cindex @code{fail} directive
4008 Generates an error or a warning. If the value of the @var{expression} is 500
4009 or more, @command{@value{AS}} will print a warning message. If the value is less
4010 than 500, @command{@value{AS}} will print an error message. The message will
4011 include the value of @var{expression}. This can occasionally be useful inside
4012 complex nested macros or conditional assembly.
4014 @ifclear no-file-dir
4016 @section @code{.file @var{string}}
4018 @cindex @code{file} directive
4019 @cindex logical file name
4020 @cindex file name, logical
4021 @code{.file} tells @command{@value{AS}} that we are about to start a new logical
4022 file. @var{string} is the new file name. In general, the filename is
4023 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4024 to specify an empty file name, you must give the quotes--@code{""}. This
4025 statement may go away in future: it is only recognized to be compatible with
4026 old @command{@value{AS}} programs.
4028 In some configurations of @command{@value{AS}}, @code{.file} has already been
4029 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
4034 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4036 @cindex @code{fill} directive
4037 @cindex writing patterns in memory
4038 @cindex patterns, writing in memory
4039 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4040 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4041 may be zero or more. @var{Size} may be zero or more, but if it is
4042 more than 8, then it is deemed to have the value 8, compatible with
4043 other people's assemblers. The contents of each @var{repeat} bytes
4044 is taken from an 8-byte number. The highest order 4 bytes are
4045 zero. The lowest order 4 bytes are @var{value} rendered in the
4046 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4047 Each @var{size} bytes in a repetition is taken from the lowest order
4048 @var{size} bytes of this number. Again, this bizarre behavior is
4049 compatible with other people's assemblers.
4051 @var{size} and @var{value} are optional.
4052 If the second comma and @var{value} are absent, @var{value} is
4053 assumed zero. If the first comma and following tokens are absent,
4054 @var{size} is assumed to be 1.
4057 @section @code{.float @var{flonums}}
4059 @cindex floating point numbers (single)
4060 @cindex @code{float} directive
4061 This directive assembles zero or more flonums, separated by commas. It
4062 has the same effect as @code{.single}.
4064 The exact kind of floating point numbers emitted depends on how
4065 @command{@value{AS}} is configured.
4066 @xref{Machine Dependencies}.
4070 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4071 in @sc{ieee} format.
4076 @section @code{.func @var{name}[,@var{label}]}
4077 @cindex @code{func} directive
4078 @code{.func} emits debugging information to denote function @var{name}, and
4079 is ignored unless the file is assembled with debugging enabled.
4080 Only @samp{--gstabs} is currently supported.
4081 @var{label} is the entry point of the function and if omitted @var{name}
4082 prepended with the @samp{leading char} is used.
4083 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4084 All functions are currently defined to have @code{void} return type.
4085 The function must be terminated with @code{.endfunc}.
4088 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4090 @cindex @code{global} directive
4091 @cindex symbol, making visible to linker
4092 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4093 @var{symbol} in your partial program, its value is made available to
4094 other partial programs that are linked with it. Otherwise,
4095 @var{symbol} takes its attributes from a symbol of the same name
4096 from another file linked into the same program.
4098 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4099 compatibility with other assemblers.
4102 On the HPPA, @code{.global} is not always enough to make it accessible to other
4103 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4104 @xref{HPPA Directives,, HPPA Assembler Directives}.
4109 @section @code{.hidden @var{names}}
4111 @cindex @code{.hidden} directive
4113 This one of the ELF visibility directives. The other two are
4114 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4115 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4117 This directive overrides the named symbols default visibility (which is set by
4118 their binding: local, global or weak). The directive sets the visibility to
4119 @code{hidden} which means that the symbols are not visible to other components.
4120 Such symbols are always considered to be @code{protected} as well.
4124 @section @code{.hword @var{expressions}}
4126 @cindex @code{hword} directive
4127 @cindex integers, 16-bit
4128 @cindex numbers, 16-bit
4129 @cindex sixteen bit integers
4130 This expects zero or more @var{expressions}, and emits
4131 a 16 bit number for each.
4134 This directive is a synonym for @samp{.short}; depending on the target
4135 architecture, it may also be a synonym for @samp{.word}.
4139 This directive is a synonym for @samp{.short}.
4142 This directive is a synonym for both @samp{.short} and @samp{.word}.
4147 @section @code{.ident}
4149 @cindex @code{ident} directive
4150 This directive is used by some assemblers to place tags in object files.
4151 @command{@value{AS}} simply accepts the directive for source-file
4152 compatibility with such assemblers, but does not actually emit anything
4156 @section @code{.if @var{absolute expression}}
4158 @cindex conditional assembly
4159 @cindex @code{if} directive
4160 @code{.if} marks the beginning of a section of code which is only
4161 considered part of the source program being assembled if the argument
4162 (which must be an @var{absolute expression}) is non-zero. The end of
4163 the conditional section of code must be marked by @code{.endif}
4164 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4165 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4166 If you have several conditions to check, @code{.elseif} may be used to avoid
4167 nesting blocks if/else within each subsequent @code{.else} block.
4169 The following variants of @code{.if} are also supported:
4171 @cindex @code{ifdef} directive
4172 @item .ifdef @var{symbol}
4173 Assembles the following section of code if the specified @var{symbol}
4176 @cindex @code{ifc} directive
4177 @item .ifc @var{string1},@var{string2}
4178 Assembles the following section of code if the two strings are the same. The
4179 strings may be optionally quoted with single quotes. If they are not quoted,
4180 the first string stops at the first comma, and the second string stops at the
4181 end of the line. Strings which contain whitespace should be quoted. The
4182 string comparison is case sensitive.
4184 @cindex @code{ifeq} directive
4185 @item .ifeq @var{absolute expression}
4186 Assembles the following section of code if the argument is zero.
4188 @cindex @code{ifeqs} directive
4189 @item .ifeqs @var{string1},@var{string2}
4190 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4192 @cindex @code{ifge} directive
4193 @item .ifge @var{absolute expression}
4194 Assembles the following section of code if the argument is greater than or
4197 @cindex @code{ifgt} directive
4198 @item .ifgt @var{absolute expression}
4199 Assembles the following section of code if the argument is greater than zero.
4201 @cindex @code{ifle} directive
4202 @item .ifle @var{absolute expression}
4203 Assembles the following section of code if the argument is less than or equal
4206 @cindex @code{iflt} directive
4207 @item .iflt @var{absolute expression}
4208 Assembles the following section of code if the argument is less than zero.
4210 @cindex @code{ifnc} directive
4211 @item .ifnc @var{string1},@var{string2}.
4212 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4213 following section of code if the two strings are not the same.
4215 @cindex @code{ifndef} directive
4216 @cindex @code{ifnotdef} directive
4217 @item .ifndef @var{symbol}
4218 @itemx .ifnotdef @var{symbol}
4219 Assembles the following section of code if the specified @var{symbol}
4220 has not been defined. Both spelling variants are equivalent.
4222 @cindex @code{ifne} directive
4223 @item .ifne @var{absolute expression}
4224 Assembles the following section of code if the argument is not equal to zero
4225 (in other words, this is equivalent to @code{.if}).
4227 @cindex @code{ifnes} directive
4228 @item .ifnes @var{string1},@var{string2}
4229 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4230 following section of code if the two strings are not the same.
4234 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4236 @cindex @code{incbin} directive
4237 @cindex binary files, including
4238 The @code{incbin} directive includes @var{file} verbatim at the current
4239 location. You can control the search paths used with the @samp{-I} command-line
4240 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4243 The @var{skip} argument skips a number of bytes from the start of the
4244 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4245 read. Note that the data is not aligned in any way, so it is the user's
4246 responsibility to make sure that proper alignment is provided both before and
4247 after the @code{incbin} directive.
4250 @section @code{.include "@var{file}"}
4252 @cindex @code{include} directive
4253 @cindex supporting files, including
4254 @cindex files, including
4255 This directive provides a way to include supporting files at specified
4256 points in your source program. The code from @var{file} is assembled as
4257 if it followed the point of the @code{.include}; when the end of the
4258 included file is reached, assembly of the original file continues. You
4259 can control the search paths used with the @samp{-I} command-line option
4260 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4264 @section @code{.int @var{expressions}}
4266 @cindex @code{int} directive
4267 @cindex integers, 32-bit
4268 Expect zero or more @var{expressions}, of any section, separated by commas.
4269 For each expression, emit a number that, at run time, is the value of that
4270 expression. The byte order and bit size of the number depends on what kind
4271 of target the assembly is for.
4275 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
4276 integers. On the H8/300H and the Hitachi SH, however, @code{.int} emits
4283 @section @code{.internal @var{names}}
4285 @cindex @code{.internal} directive
4287 This one of the ELF visibility directives. The other two are
4288 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4289 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4291 This directive overrides the named symbols default visibility (which is set by
4292 their binding: local, global or weak). The directive sets the visibility to
4293 @code{internal} which means that the symbols are considered to be @code{hidden}
4294 (ie not visible to other components), and that some extra, processor specific
4295 processing must also be performed upon the symbols as well.
4299 @section @code{.irp @var{symbol},@var{values}}@dots{}
4301 @cindex @code{irp} directive
4302 Evaluate a sequence of statements assigning different values to @var{symbol}.
4303 The sequence of statements starts at the @code{.irp} directive, and is
4304 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4305 set to @var{value}, and the sequence of statements is assembled. If no
4306 @var{value} is listed, the sequence of statements is assembled once, with
4307 @var{symbol} set to the null string. To refer to @var{symbol} within the
4308 sequence of statements, use @var{\symbol}.
4310 For example, assembling
4318 is equivalent to assembling
4327 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4329 @cindex @code{irpc} directive
4330 Evaluate a sequence of statements assigning different values to @var{symbol}.
4331 The sequence of statements starts at the @code{.irpc} directive, and is
4332 terminated by an @code{.endr} directive. For each character in @var{value},
4333 @var{symbol} is set to the character, and the sequence of statements is
4334 assembled. If no @var{value} is listed, the sequence of statements is
4335 assembled once, with @var{symbol} set to the null string. To refer to
4336 @var{symbol} within the sequence of statements, use @var{\symbol}.
4338 For example, assembling
4346 is equivalent to assembling
4355 @section @code{.lcomm @var{symbol} , @var{length}}
4357 @cindex @code{lcomm} directive
4358 @cindex local common symbols
4359 @cindex symbols, local common
4360 Reserve @var{length} (an absolute expression) bytes for a local common
4361 denoted by @var{symbol}. The section and value of @var{symbol} are
4362 those of the new local common. The addresses are allocated in the bss
4363 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4364 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4365 not visible to @code{@value{LD}}.
4368 Some targets permit a third argument to be used with @code{.lcomm}. This
4369 argument specifies the desired alignment of the symbol in the bss section.
4373 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4374 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4378 @section @code{.lflags}
4380 @cindex @code{lflags} directive (ignored)
4381 @command{@value{AS}} accepts this directive, for compatibility with other
4382 assemblers, but ignores it.
4384 @ifclear no-line-dir
4386 @section @code{.line @var{line-number}}
4388 @cindex @code{line} directive
4392 @section @code{.ln @var{line-number}}
4394 @cindex @code{ln} directive
4396 @cindex logical line number
4398 Change the logical line number. @var{line-number} must be an absolute
4399 expression. The next line has that logical line number. Therefore any other
4400 statements on the current line (after a statement separator character) are
4401 reported as on logical line number @var{line-number} @minus{} 1. One day
4402 @command{@value{AS}} will no longer support this directive: it is recognized only
4403 for compatibility with existing assembler programs.
4407 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
4408 not available; use the synonym @code{.ln} in that context.
4413 @ifclear no-line-dir
4414 Even though this is a directive associated with the @code{a.out} or
4415 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
4416 when producing COFF output, and treats @samp{.line} as though it
4417 were the COFF @samp{.ln} @emph{if} it is found outside a
4418 @code{.def}/@code{.endef} pair.
4420 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4421 used by compilers to generate auxiliary symbol information for
4426 @section @code{.linkonce [@var{type}]}
4428 @cindex @code{linkonce} directive
4429 @cindex common sections
4430 Mark the current section so that the linker only includes a single copy of it.
4431 This may be used to include the same section in several different object files,
4432 but ensure that the linker will only include it once in the final output file.
4433 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4434 Duplicate sections are detected based on the section name, so it should be
4437 This directive is only supported by a few object file formats; as of this
4438 writing, the only object file format which supports it is the Portable
4439 Executable format used on Windows NT.
4441 The @var{type} argument is optional. If specified, it must be one of the
4442 following strings. For example:
4446 Not all types may be supported on all object file formats.
4450 Silently discard duplicate sections. This is the default.
4453 Warn if there are duplicate sections, but still keep only one copy.
4456 Warn if any of the duplicates have different sizes.
4459 Warn if any of the duplicates do not have exactly the same contents.
4463 @section @code{.ln @var{line-number}}
4465 @cindex @code{ln} directive
4466 @ifclear no-line-dir
4467 @samp{.ln} is a synonym for @samp{.line}.
4470 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
4471 must be an absolute expression. The next line has that logical
4472 line number, so any other statements on the current line (after a
4473 statement separator character @code{;}) are reported as on logical
4474 line number @var{line-number} @minus{} 1.
4477 This directive is accepted, but ignored, when @command{@value{AS}} is
4478 configured for @code{b.out}; its effect is only associated with COFF
4484 @section @code{.mri @var{val}}
4486 @cindex @code{mri} directive
4487 @cindex MRI mode, temporarily
4488 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
4489 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
4490 affects code assembled until the next @code{.mri} directive, or until the end
4491 of the file. @xref{M, MRI mode, MRI mode}.
4494 @section @code{.list}
4496 @cindex @code{list} directive
4497 @cindex listing control, turning on
4498 Control (in conjunction with the @code{.nolist} directive) whether or
4499 not assembly listings are generated. These two directives maintain an
4500 internal counter (which is zero initially). @code{.list} increments the
4501 counter, and @code{.nolist} decrements it. Assembly listings are
4502 generated whenever the counter is greater than zero.
4504 By default, listings are disabled. When you enable them (with the
4505 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4506 the initial value of the listing counter is one.
4509 @section @code{.long @var{expressions}}
4511 @cindex @code{long} directive
4512 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
4515 @c no one seems to know what this is for or whether this description is
4516 @c what it really ought to do
4518 @section @code{.lsym @var{symbol}, @var{expression}}
4520 @cindex @code{lsym} directive
4521 @cindex symbol, not referenced in assembly
4522 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
4523 the hash table, ensuring it cannot be referenced by name during the
4524 rest of the assembly. This sets the attributes of the symbol to be
4525 the same as the expression value:
4527 @var{other} = @var{descriptor} = 0
4528 @var{type} = @r{(section of @var{expression})}
4529 @var{value} = @var{expression}
4532 The new symbol is not flagged as external.
4536 @section @code{.macro}
4539 The commands @code{.macro} and @code{.endm} allow you to define macros that
4540 generate assembly output. For example, this definition specifies a macro
4541 @code{sum} that puts a sequence of numbers into memory:
4544 .macro sum from=0, to=5
4553 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
4565 @item .macro @var{macname}
4566 @itemx .macro @var{macname} @var{macargs} @dots{}
4567 @cindex @code{macro} directive
4568 Begin the definition of a macro called @var{macname}. If your macro
4569 definition requires arguments, specify their names after the macro name,
4570 separated by commas or spaces. You can supply a default value for any
4571 macro argument by following the name with @samp{=@var{deflt}}. For
4572 example, these are all valid @code{.macro} statements:
4576 Begin the definition of a macro called @code{comm}, which takes no
4579 @item .macro plus1 p, p1
4580 @itemx .macro plus1 p p1
4581 Either statement begins the definition of a macro called @code{plus1},
4582 which takes two arguments; within the macro definition, write
4583 @samp{\p} or @samp{\p1} to evaluate the arguments.
4585 @item .macro reserve_str p1=0 p2
4586 Begin the definition of a macro called @code{reserve_str}, with two
4587 arguments. The first argument has a default value, but not the second.
4588 After the definition is complete, you can call the macro either as
4589 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
4590 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
4591 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
4592 @samp{0}, and @samp{\p2} evaluating to @var{b}).
4595 When you call a macro, you can specify the argument values either by
4596 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
4597 @samp{sum to=17, from=9}.
4600 @cindex @code{endm} directive
4601 Mark the end of a macro definition.
4604 @cindex @code{exitm} directive
4605 Exit early from the current macro definition.
4607 @cindex number of macros executed
4608 @cindex macros, count executed
4610 @command{@value{AS}} maintains a counter of how many macros it has
4611 executed in this pseudo-variable; you can copy that number to your
4612 output with @samp{\@@}, but @emph{only within a macro definition}.
4615 @item LOCAL @var{name} [ , @dots{} ]
4616 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
4617 macro syntax'' with @samp{-a} or @samp{--alternate}.} @xref{Alternate,,
4618 Alternate macro syntax}.
4620 Generate a string replacement for each of the @var{name} arguments, and
4621 replace any instances of @var{name} in each macro expansion. The
4622 replacement string is unique in the assembly, and different for each
4623 separate macro expansion. @code{LOCAL} allows you to write macros that
4624 define symbols, without fear of conflict between separate macro expansions.
4629 @section @code{.nolist}
4631 @cindex @code{nolist} directive
4632 @cindex listing control, turning off
4633 Control (in conjunction with the @code{.list} directive) whether or
4634 not assembly listings are generated. These two directives maintain an
4635 internal counter (which is zero initially). @code{.list} increments the
4636 counter, and @code{.nolist} decrements it. Assembly listings are
4637 generated whenever the counter is greater than zero.
4640 @section @code{.octa @var{bignums}}
4642 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
4643 @cindex @code{octa} directive
4644 @cindex integer, 16-byte
4645 @cindex sixteen byte integer
4646 This directive expects zero or more bignums, separated by commas. For each
4647 bignum, it emits a 16-byte integer.
4649 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
4650 hence @emph{octa}-word for 16 bytes.
4653 @section @code{.org @var{new-lc} , @var{fill}}
4655 @cindex @code{org} directive
4656 @cindex location counter, advancing
4657 @cindex advancing location counter
4658 @cindex current address, advancing
4659 Advance the location counter of the current section to
4660 @var{new-lc}. @var{new-lc} is either an absolute expression or an
4661 expression with the same section as the current subsection. That is,
4662 you can't use @code{.org} to cross sections: if @var{new-lc} has the
4663 wrong section, the @code{.org} directive is ignored. To be compatible
4664 with former assemblers, if the section of @var{new-lc} is absolute,
4665 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
4666 is the same as the current subsection.
4668 @code{.org} may only increase the location counter, or leave it
4669 unchanged; you cannot use @code{.org} to move the location counter
4672 @c double negative used below "not undefined" because this is a specific
4673 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
4674 @c section. doc@cygnus.com 18feb91
4675 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
4676 may not be undefined. If you really detest this restriction we eagerly await
4677 a chance to share your improved assembler.
4679 Beware that the origin is relative to the start of the section, not
4680 to the start of the subsection. This is compatible with other
4681 people's assemblers.
4683 When the location counter (of the current subsection) is advanced, the
4684 intervening bytes are filled with @var{fill} which should be an
4685 absolute expression. If the comma and @var{fill} are omitted,
4686 @var{fill} defaults to zero.
4689 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4691 @cindex padding the location counter given a power of two
4692 @cindex @code{p2align} directive
4693 Pad the location counter (in the current subsection) to a particular
4694 storage boundary. The first expression (which must be absolute) is the
4695 number of low-order zero bits the location counter must have after
4696 advancement. For example @samp{.p2align 3} advances the location
4697 counter until it a multiple of 8. If the location counter is already a
4698 multiple of 8, no change is needed.
4700 The second expression (also absolute) gives the fill value to be stored in the
4701 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4702 padding bytes are normally zero. However, on some systems, if the section is
4703 marked as containing code and the fill value is omitted, the space is filled
4704 with no-op instructions.
4706 The third expression is also absolute, and is also optional. If it is present,
4707 it is the maximum number of bytes that should be skipped by this alignment
4708 directive. If doing the alignment would require skipping more bytes than the
4709 specified maximum, then the alignment is not done at all. You can omit the
4710 fill value (the second argument) entirely by simply using two commas after the
4711 required alignment; this can be useful if you want the alignment to be filled
4712 with no-op instructions when appropriate.
4714 @cindex @code{p2alignw} directive
4715 @cindex @code{p2alignl} directive
4716 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
4717 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
4718 pattern as a two byte word value. The @code{.p2alignl} directives treats the
4719 fill pattern as a four byte longword value. For example, @code{.p2alignw
4720 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4721 filled in with the value 0x368d (the exact placement of the bytes depends upon
4722 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4727 @section @code{.previous}
4729 @cindex @code{.previous} directive
4730 @cindex Section Stack
4731 This is one of the ELF section stack manipulation directives. The others are
4732 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
4733 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
4734 (@pxref{PopSection}).
4736 This directive swaps the current section (and subsection) with most recently
4737 referenced section (and subsection) prior to this one. Multiple
4738 @code{.previous} directives in a row will flip between two sections (and their
4741 In terms of the section stack, this directive swaps the current section with
4742 the top section on the section stack.
4747 @section @code{.popsection}
4749 @cindex @code{.popsection} directive
4750 @cindex Section Stack
4751 This is one of the ELF section stack manipulation directives. The others are
4752 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
4753 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
4756 This directive replaces the current section (and subsection) with the top
4757 section (and subsection) on the section stack. This section is popped off the
4762 @section @code{.print @var{string}}
4764 @cindex @code{print} directive
4765 @command{@value{AS}} will print @var{string} on the standard output during
4766 assembly. You must put @var{string} in double quotes.
4770 @section @code{.protected @var{names}}
4772 @cindex @code{.protected} directive
4774 This one of the ELF visibility directives. The other two are
4775 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
4777 This directive overrides the named symbols default visibility (which is set by
4778 their binding: local, global or weak). The directive sets the visibility to
4779 @code{protected} which means that any references to the symbols from within the
4780 components that defines them must be resolved to the definition in that
4781 component, even if a definition in another component would normally preempt
4786 @section @code{.psize @var{lines} , @var{columns}}
4788 @cindex @code{psize} directive
4789 @cindex listing control: paper size
4790 @cindex paper size, for listings
4791 Use this directive to declare the number of lines---and, optionally, the
4792 number of columns---to use for each page, when generating listings.
4794 If you do not use @code{.psize}, listings use a default line-count
4795 of 60. You may omit the comma and @var{columns} specification; the
4796 default width is 200 columns.
4798 @command{@value{AS}} generates formfeeds whenever the specified number of
4799 lines is exceeded (or whenever you explicitly request one, using
4802 If you specify @var{lines} as @code{0}, no formfeeds are generated save
4803 those explicitly specified with @code{.eject}.
4806 @section @code{.purgem @var{name}}
4808 @cindex @code{purgem} directive
4809 Undefine the macro @var{name}, so that later uses of the string will not be
4810 expanded. @xref{Macro}.
4814 @section @code{.pushsection @var{name} , @var{subsection}}
4816 @cindex @code{.pushsection} directive
4817 @cindex Section Stack
4818 This is one of the ELF section stack manipulation directives. The others are
4819 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
4820 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
4823 This directive is a synonym for @code{.section}. It pushes the current section
4824 (and subsection) onto the top of the section stack, and then replaces the
4825 current section and subsection with @code{name} and @code{subsection}.
4829 @section @code{.quad @var{bignums}}
4831 @cindex @code{quad} directive
4832 @code{.quad} expects zero or more bignums, separated by commas. For
4833 each bignum, it emits
4835 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
4836 warning message; and just takes the lowest order 8 bytes of the bignum.
4837 @cindex eight-byte integer
4838 @cindex integer, 8-byte
4840 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
4841 hence @emph{quad}-word for 8 bytes.
4844 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
4845 warning message; and just takes the lowest order 16 bytes of the bignum.
4846 @cindex sixteen-byte integer
4847 @cindex integer, 16-byte
4851 @section @code{.rept @var{count}}
4853 @cindex @code{rept} directive
4854 Repeat the sequence of lines between the @code{.rept} directive and the next
4855 @code{.endr} directive @var{count} times.
4857 For example, assembling
4865 is equivalent to assembling
4874 @section @code{.sbttl "@var{subheading}"}
4876 @cindex @code{sbttl} directive
4877 @cindex subtitles for listings
4878 @cindex listing control: subtitle
4879 Use @var{subheading} as the title (third line, immediately after the
4880 title line) when generating assembly listings.
4882 This directive affects subsequent pages, as well as the current page if
4883 it appears within ten lines of the top of a page.
4887 @section @code{.scl @var{class}}
4889 @cindex @code{scl} directive
4890 @cindex symbol storage class (COFF)
4891 @cindex COFF symbol storage class
4892 Set the storage-class value for a symbol. This directive may only be
4893 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
4894 whether a symbol is static or external, or it may record further
4895 symbolic debugging information.
4898 The @samp{.scl} directive is primarily associated with COFF output; when
4899 configured to generate @code{b.out} output format, @command{@value{AS}}
4900 accepts this directive but ignores it.
4905 @section @code{.section @var{name}} (COFF version)
4907 @cindex @code{section} directive
4908 @cindex named section
4909 Use the @code{.section} directive to assemble the following code into a section
4912 This directive is only supported for targets that actually support arbitrarily
4913 named sections; on @code{a.out} targets, for example, it is not accepted, even
4914 with a standard @code{a.out} section name.
4916 For COFF targets, the @code{.section} directive is used in one of the following
4920 .section @var{name}[, "@var{flags}"]
4921 .section @var{name}[, @var{subsegment}]
4924 If the optional argument is quoted, it is taken as flags to use for the
4925 section. Each flag is a single character. The following flags are recognized:
4928 bss section (uninitialized data)
4930 section is not loaded
4940 shared section (meaningful for PE targets)
4943 If no flags are specified, the default flags depend upon the section name. If
4944 the section name is not recognized, the default will be for the section to be
4945 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
4946 from the section, rather than adding them, so if they are used on their own it
4947 will be as if no flags had been specified at all.
4949 If the optional argument to the @code{.section} directive is not quoted, it is
4950 taken as a subsegment number (@pxref{Sub-Sections}).
4953 @section @code{.section @var{name}} (ELF version)
4955 @cindex @code{section} directive
4956 @cindex named section
4958 @cindex Section Stack
4959 This is one of the ELF section stack manipulation directives. The others are
4960 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
4961 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
4962 @code{.previous} (@pxref{Previous}).
4965 For ELF targets, the @code{.section} directive is used like this:
4968 .section @var{name} [, "@var{flags}"[, @@@var{type}[, @@@var{entsize}]]]
4971 The optional @var{flags} argument is a quoted string which may contain any
4972 combination of the following characters:
4975 section is allocatable
4979 section is executable
4981 section is mergeable
4983 section contains zero terminated strings
4986 The optional @var{type} argument may contain one of the following constants:
4989 section contains data
4991 section does not contain data (i.e., section only occupies space)
4994 If @var{flags} contains @code{M} flag, @var{type} argument must be specified
4995 as well as @var{entsize} argument. Sections with @code{M} flag but not
4996 @code{S} flag must contain fixed size constants, each @var{entsize} octets
4997 long. Sections with both @code{M} and @code{S} must contain zero terminated
4998 strings where each character is @var{entsize} bytes long. The linker may remove
4999 duplicates within sections with the same name, same entity size and same flags.
5001 If no flags are specified, the default flags depend upon the section name. If
5002 the section name is not recognized, the default will be for the section to have
5003 none of the above flags: it will not be allocated in memory, nor writable, nor
5004 executable. The section will contain data.
5006 For ELF targets, the assembler supports another type of @code{.section}
5007 directive for compatibility with the Solaris assembler:
5010 .section "@var{name}"[, @var{flags}...]
5013 Note that the section name is quoted. There may be a sequence of comma
5017 section is allocatable
5021 section is executable
5024 This directive replaces the current section and subsection. The replaced
5025 section and subsection are pushed onto the section stack. See the contents of
5026 the gas testsuite directory @code{gas/testsuite/gas/elf} for some examples of
5027 how this directive and the other section stack directives work.
5030 @section @code{.set @var{symbol}, @var{expression}}
5032 @cindex @code{set} directive
5033 @cindex symbol value, setting
5034 Set the value of @var{symbol} to @var{expression}. This
5035 changes @var{symbol}'s value and type to conform to
5036 @var{expression}. If @var{symbol} was flagged as external, it remains
5037 flagged (@pxref{Symbol Attributes}).
5039 You may @code{.set} a symbol many times in the same assembly.
5041 If you @code{.set} a global symbol, the value stored in the object
5042 file is the last value stored into it.
5045 The syntax for @code{set} on the HPPA is
5046 @samp{@var{symbol} .set @var{expression}}.
5050 @section @code{.short @var{expressions}}
5052 @cindex @code{short} directive
5054 @code{.short} is normally the same as @samp{.word}.
5055 @xref{Word,,@code{.word}}.
5057 In some configurations, however, @code{.short} and @code{.word} generate
5058 numbers of different lengths; @pxref{Machine Dependencies}.
5062 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5065 This expects zero or more @var{expressions}, and emits
5066 a 16 bit number for each.
5071 @section @code{.single @var{flonums}}
5073 @cindex @code{single} directive
5074 @cindex floating point numbers (single)
5075 This directive assembles zero or more flonums, separated by commas. It
5076 has the same effect as @code{.float}.
5078 The exact kind of floating point numbers emitted depends on how
5079 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5083 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5084 numbers in @sc{ieee} format.
5089 @section @code{.size} (COFF version)
5091 @cindex @code{size} directive
5092 This directive is generated by compilers to include auxiliary debugging
5093 information in the symbol table. It is only permitted inside
5094 @code{.def}/@code{.endef} pairs.
5097 @samp{.size} is only meaningful when generating COFF format output; when
5098 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5102 @section @code{.size @var{name} , @var{expression}} (ELF version)
5103 @cindex @code{size} directive
5105 This directive is used to set the size associated with a symbol @var{name}.
5106 The size in bytes is computed from @var{expression} which can make use of label
5107 arithmetic. This directive is typically used to set the size of function
5111 @section @code{.sleb128 @var{expressions}}
5113 @cindex @code{sleb128} directive
5114 @var{sleb128} stands for ``signed little endian base 128.'' This is a
5115 compact, variable length representation of numbers used by the DWARF
5116 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
5118 @ifclear no-space-dir
5120 @section @code{.skip @var{size} , @var{fill}}
5122 @cindex @code{skip} directive
5123 @cindex filling memory
5124 This directive emits @var{size} bytes, each of value @var{fill}. Both
5125 @var{size} and @var{fill} are absolute expressions. If the comma and
5126 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
5130 @section @code{.space @var{size} , @var{fill}}
5132 @cindex @code{space} directive
5133 @cindex filling memory
5134 This directive emits @var{size} bytes, each of value @var{fill}. Both
5135 @var{size} and @var{fill} are absolute expressions. If the comma
5136 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
5141 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
5142 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
5143 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
5144 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
5153 @section @code{.space}
5154 @cindex @code{space} directive
5156 On the AMD 29K, this directive is ignored; it is accepted for
5157 compatibility with other AMD 29K assemblers.
5160 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
5161 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
5167 @section @code{.stabd, .stabn, .stabs}
5169 @cindex symbolic debuggers, information for
5170 @cindex @code{stab@var{x}} directives
5171 There are three directives that begin @samp{.stab}.
5172 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
5173 The symbols are not entered in the @command{@value{AS}} hash table: they
5174 cannot be referenced elsewhere in the source file.
5175 Up to five fields are required:
5179 This is the symbol's name. It may contain any character except
5180 @samp{\000}, so is more general than ordinary symbol names. Some
5181 debuggers used to code arbitrarily complex structures into symbol names
5185 An absolute expression. The symbol's type is set to the low 8 bits of
5186 this expression. Any bit pattern is permitted, but @code{@value{LD}}
5187 and debuggers choke on silly bit patterns.
5190 An absolute expression. The symbol's ``other'' attribute is set to the
5191 low 8 bits of this expression.
5194 An absolute expression. The symbol's descriptor is set to the low 16
5195 bits of this expression.
5198 An absolute expression which becomes the symbol's value.
5201 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
5202 or @code{.stabs} statement, the symbol has probably already been created;
5203 you get a half-formed symbol in your object file. This is
5204 compatible with earlier assemblers!
5207 @cindex @code{stabd} directive
5208 @item .stabd @var{type} , @var{other} , @var{desc}
5210 The ``name'' of the symbol generated is not even an empty string.
5211 It is a null pointer, for compatibility. Older assemblers used a
5212 null pointer so they didn't waste space in object files with empty
5215 The symbol's value is set to the location counter,
5216 relocatably. When your program is linked, the value of this symbol
5217 is the address of the location counter when the @code{.stabd} was
5220 @cindex @code{stabn} directive
5221 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
5222 The name of the symbol is set to the empty string @code{""}.
5224 @cindex @code{stabs} directive
5225 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
5226 All five fields are specified.
5232 @section @code{.string} "@var{str}"
5234 @cindex string, copying to object file
5235 @cindex @code{string} directive
5237 Copy the characters in @var{str} to the object file. You may specify more than
5238 one string to copy, separated by commas. Unless otherwise specified for a
5239 particular machine, the assembler marks the end of each string with a 0 byte.
5240 You can use any of the escape sequences described in @ref{Strings,,Strings}.
5243 @section @code{.struct @var{expression}}
5245 @cindex @code{struct} directive
5246 Switch to the absolute section, and set the section offset to @var{expression},
5247 which must be an absolute expression. You might use this as follows:
5256 This would define the symbol @code{field1} to have the value 0, the symbol
5257 @code{field2} to have the value 4, and the symbol @code{field3} to have the
5258 value 8. Assembly would be left in the absolute section, and you would need to
5259 use a @code{.section} directive of some sort to change to some other section
5260 before further assembly.
5264 @section @code{.subsection @var{name}}
5266 @cindex @code{.subsection} directive
5267 @cindex Section Stack
5268 This is one of the ELF section stack manipulation directives. The others are
5269 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
5270 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5273 This directive replaces the current subsection with @code{name}. The current
5274 section is not changed. The replaced subsection is put onto the section stack
5275 in place of the then current top of stack subsection.
5280 @section @code{.symver}
5281 @cindex @code{symver} directive
5282 @cindex symbol versioning
5283 @cindex versions of symbols
5284 Use the @code{.symver} directive to bind symbols to specific version nodes
5285 within a source file. This is only supported on ELF platforms, and is
5286 typically used when assembling files to be linked into a shared library.
5287 There are cases where it may make sense to use this in objects to be bound
5288 into an application itself so as to override a versioned symbol from a
5291 For ELF targets, the @code{.symver} directive can be used like this:
5293 .symver @var{name}, @var{name2@@nodename}
5295 If the symbol @var{name} is defined within the file
5296 being assembled, the @code{.symver} directive effectively creates a symbol
5297 alias with the name @var{name2@@nodename}, and in fact the main reason that we
5298 just don't try and create a regular alias is that the @var{@@} character isn't
5299 permitted in symbol names. The @var{name2} part of the name is the actual name
5300 of the symbol by which it will be externally referenced. The name @var{name}
5301 itself is merely a name of convenience that is used so that it is possible to
5302 have definitions for multiple versions of a function within a single source
5303 file, and so that the compiler can unambiguously know which version of a
5304 function is being mentioned. The @var{nodename} portion of the alias should be
5305 the name of a node specified in the version script supplied to the linker when
5306 building a shared library. If you are attempting to override a versioned
5307 symbol from a shared library, then @var{nodename} should correspond to the
5308 nodename of the symbol you are trying to override.
5310 If the symbol @var{name} is not defined within the file being assembled, all
5311 references to @var{name} will be changed to @var{name2@@nodename}. If no
5312 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
5315 Another usage of the @code{.symver} directive is:
5317 .symver @var{name}, @var{name2@@@@nodename}
5319 In this case, the symbol @var{name} must exist and be defined within
5320 the file being assembled. It is similar to @var{name2@@nodename}. The
5321 difference is @var{name2@@@@nodename} will also be used to resolve
5322 references to @var{name2} by the linker.
5324 The third usage of the @code{.symver} directive is:
5326 .symver @var{name}, @var{name2@@@@@@nodename}
5328 When @var{name} is not defined within the
5329 file being assembled, it is treated as @var{name2@@nodename}. When
5330 @var{name} is defined within the file being assembled, the symbol
5331 name, @var{name}, will be changed to @var{name2@@@@nodename}.
5336 @section @code{.tag @var{structname}}
5338 @cindex COFF structure debugging
5339 @cindex structure debugging, COFF
5340 @cindex @code{tag} directive
5341 This directive is generated by compilers to include auxiliary debugging
5342 information in the symbol table. It is only permitted inside
5343 @code{.def}/@code{.endef} pairs. Tags are used to link structure
5344 definitions in the symbol table with instances of those structures.
5347 @samp{.tag} is only used when generating COFF format output; when
5348 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5354 @section @code{.text @var{subsection}}
5356 @cindex @code{text} directive
5357 Tells @command{@value{AS}} to assemble the following statements onto the end of
5358 the text subsection numbered @var{subsection}, which is an absolute
5359 expression. If @var{subsection} is omitted, subsection number zero
5363 @section @code{.title "@var{heading}"}
5365 @cindex @code{title} directive
5366 @cindex listing control: title line
5367 Use @var{heading} as the title (second line, immediately after the
5368 source file name and pagenumber) when generating assembly listings.
5370 This directive affects subsequent pages, as well as the current page if
5371 it appears within ten lines of the top of a page.
5374 @section @code{.type @var{int}} (COFF version)
5376 @cindex COFF symbol type
5377 @cindex symbol type, COFF
5378 @cindex @code{type} directive
5379 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5380 records the integer @var{int} as the type attribute of a symbol table entry.
5383 @samp{.type} is associated only with COFF format output; when
5384 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
5385 directive but ignores it.
5388 @section @code{.type @var{name} , @var{type description}} (ELF version)
5390 @cindex ELF symbol type
5391 @cindex symbol type, ELF
5392 @cindex @code{type} directive
5393 This directive is used to set the type of symbol @var{name} to be either a
5394 function symbol or an object symbol. There are five different syntaxes
5395 supported for the @var{type description} field, in order to provide
5396 compatibility with various other assemblers. The syntaxes supported are:
5399 .type <name>,#function
5400 .type <name>,#object
5402 .type <name>,@@function
5403 .type <name>,@@object
5405 .type <name>,%function
5406 .type <name>,%object
5408 .type <name>,"function"
5409 .type <name>,"object"
5411 .type <name> STT_FUNCTION
5412 .type <name> STT_OBJECT
5416 @section @code{.uleb128 @var{expressions}}
5418 @cindex @code{uleb128} directive
5419 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
5420 compact, variable length representation of numbers used by the DWARF
5421 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
5425 @section @code{.val @var{addr}}
5427 @cindex @code{val} directive
5428 @cindex COFF value attribute
5429 @cindex value attribute, COFF
5430 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5431 records the address @var{addr} as the value attribute of a symbol table
5435 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
5436 configured for @code{b.out}, it accepts this directive but ignores it.
5442 @section @code{.version "@var{string}"}
5444 @cindex @code{.version}
5445 This directive creates a @code{.note} section and places into it an ELF
5446 formatted note of type NT_VERSION. The note's name is set to @code{string}.
5451 @section @code{.vtable_entry @var{table}, @var{offset}}
5453 @cindex @code{.vtable_entry}
5454 This directive finds or creates a symbol @code{table} and creates a
5455 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
5458 @section @code{.vtable_inherit @var{child}, @var{parent}}
5460 @cindex @code{.vtable_inherit}
5461 This directive finds the symbol @code{child} and finds or creates the symbol
5462 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
5463 parent whose addend is the value of the child symbol. As a special case the
5464 parent name of @code{0} is treated as refering the @code{*ABS*} section.
5469 @section @code{.weak @var{names}}
5471 @cindex @code{.weak}
5472 This directive sets the weak attribute on the comma separated list of symbol
5473 @code{names}. If the symbols do not already exist, they will be created.
5477 @section @code{.word @var{expressions}}
5479 @cindex @code{word} directive
5480 This directive expects zero or more @var{expressions}, of any section,
5481 separated by commas.
5484 For each expression, @command{@value{AS}} emits a 32-bit number.
5487 For each expression, @command{@value{AS}} emits a 16-bit number.
5492 The size of the number emitted, and its byte order,
5493 depend on what target computer the assembly is for.
5496 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
5497 @c happen---32-bit addressability, period; no long/short jumps.
5498 @ifset DIFF-TBL-KLUGE
5499 @cindex difference tables altered
5500 @cindex altered difference tables
5502 @emph{Warning: Special Treatment to support Compilers}
5506 Machines with a 32-bit address space, but that do less than 32-bit
5507 addressing, require the following special treatment. If the machine of
5508 interest to you does 32-bit addressing (or doesn't require it;
5509 @pxref{Machine Dependencies}), you can ignore this issue.
5512 In order to assemble compiler output into something that works,
5513 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
5514 Directives of the form @samp{.word sym1-sym2} are often emitted by
5515 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
5516 directive of the form @samp{.word sym1-sym2}, and the difference between
5517 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
5518 creates a @dfn{secondary jump table}, immediately before the next label.
5519 This secondary jump table is preceded by a short-jump to the
5520 first byte after the secondary table. This short-jump prevents the flow
5521 of control from accidentally falling into the new table. Inside the
5522 table is a long-jump to @code{sym2}. The original @samp{.word}
5523 contains @code{sym1} minus the address of the long-jump to
5526 If there were several occurrences of @samp{.word sym1-sym2} before the
5527 secondary jump table, all of them are adjusted. If there was a
5528 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
5529 long-jump to @code{sym4} is included in the secondary jump table,
5530 and the @code{.word} directives are adjusted to contain @code{sym3}
5531 minus the address of the long-jump to @code{sym4}; and so on, for as many
5532 entries in the original jump table as necessary.
5535 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
5536 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
5537 assembly language programmers.
5540 @c end DIFF-TBL-KLUGE
5543 @section Deprecated Directives
5545 @cindex deprecated directives
5546 @cindex obsolescent directives
5547 One day these directives won't work.
5548 They are included for compatibility with older assemblers.
5555 @node Machine Dependencies
5556 @chapter Machine Dependent Features
5558 @cindex machine dependencies
5559 The machine instruction sets are (almost by definition) different on
5560 each machine where @command{@value{AS}} runs. Floating point representations
5561 vary as well, and @command{@value{AS}} often supports a few additional
5562 directives or command-line options for compatibility with other
5563 assemblers on a particular platform. Finally, some versions of
5564 @command{@value{AS}} support special pseudo-instructions for branch
5567 This chapter discusses most of these differences, though it does not
5568 include details on any machine's instruction set. For details on that
5569 subject, see the hardware manufacturer's manual.
5573 * AMD29K-Dependent:: AMD 29K Dependent Features
5576 * ARC-Dependent:: ARC Dependent Features
5579 * ARM-Dependent:: ARM Dependent Features
5582 * D10V-Dependent:: D10V Dependent Features
5585 * D30V-Dependent:: D30V Dependent Features
5588 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
5591 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
5594 * HPPA-Dependent:: HPPA Dependent Features
5597 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
5600 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
5603 * i860-Dependent:: Intel 80860 Dependent Features
5606 * i960-Dependent:: Intel 80960 Dependent Features
5609 * M32R-Dependent:: M32R Dependent Features
5612 * M68K-Dependent:: M680x0 Dependent Features
5615 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
5618 * M88K-Dependent:: M880x0 Dependent Features
5621 * MIPS-Dependent:: MIPS Dependent Features
5624 * MMIX-Dependent:: MMIX Dependent Features
5627 * SH-Dependent:: Hitachi SH Dependent Features
5628 * SH64-Dependent:: Hitachi SH64 Dependent Features
5631 * PDP-11-Dependent:: PDP-11 Dependent Features
5634 * PJ-Dependent:: picoJava Dependent Features
5637 * PPC-Dependent:: PowerPC Dependent Features
5640 * Sparc-Dependent:: SPARC Dependent Features
5643 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
5646 * V850-Dependent:: V850 Dependent Features
5649 * Z8000-Dependent:: Z8000 Dependent Features
5652 * Vax-Dependent:: VAX Dependent Features
5659 @c The following major nodes are *sections* in the GENERIC version, *chapters*
5660 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
5661 @c peculiarity: to preserve cross-references, there must be a node called
5662 @c "Machine Dependencies". Hence the conditional nodenames in each
5663 @c major node below. Node defaulting in makeinfo requires adjacency of
5664 @c node and sectioning commands; hence the repetition of @chapter BLAH
5665 @c in both conditional blocks.
5672 @include c-a29k.texi
5681 @node Machine Dependencies
5682 @chapter Machine Dependent Features
5684 The machine instruction sets are different on each Hitachi chip family,
5685 and there are also some syntax differences among the families. This
5686 chapter describes the specific @command{@value{AS}} features for each
5690 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
5691 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
5692 * SH-Dependent:: Hitachi SH Dependent Features
5699 @include c-d10v.texi
5703 @include c-d30v.texi
5707 @include c-h8300.texi
5711 @include c-h8500.texi
5715 @include c-hppa.texi
5719 @include c-i370.texi
5723 @include c-i386.texi
5727 @include c-i860.texi
5731 @include c-i960.texi
5735 @include c-m32r.texi
5739 @include c-m68k.texi
5743 @include c-m68hc11.texi
5747 @include c-m88k.texi
5751 @include c-mips.texi
5755 @include c-mmix.texi
5759 @include c-ns32k.texi
5763 @include c-pdp11.texi
5776 @include c-sh64.texi
5780 @include c-sparc.texi
5784 @include c-tic54x.texi
5796 @include c-v850.texi
5800 @c reverse effect of @down at top of generic Machine-Dep chapter
5804 @node Reporting Bugs
5805 @chapter Reporting Bugs
5806 @cindex bugs in assembler
5807 @cindex reporting bugs in assembler
5809 Your bug reports play an essential role in making @command{@value{AS}} reliable.
5811 Reporting a bug may help you by bringing a solution to your problem, or it may
5812 not. But in any case the principal function of a bug report is to help the
5813 entire community by making the next version of @command{@value{AS}} work better.
5814 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
5816 In order for a bug report to serve its purpose, you must include the
5817 information that enables us to fix the bug.
5820 * Bug Criteria:: Have you found a bug?
5821 * Bug Reporting:: How to report bugs
5825 @section Have you found a bug?
5826 @cindex bug criteria
5828 If you are not sure whether you have found a bug, here are some guidelines:
5831 @cindex fatal signal
5832 @cindex assembler crash
5833 @cindex crash of assembler
5835 If the assembler gets a fatal signal, for any input whatever, that is a
5836 @command{@value{AS}} bug. Reliable assemblers never crash.
5838 @cindex error on valid input
5840 If @command{@value{AS}} produces an error message for valid input, that is a bug.
5842 @cindex invalid input
5844 If @command{@value{AS}} does not produce an error message for invalid input, that
5845 is a bug. However, you should note that your idea of ``invalid input'' might
5846 be our idea of ``an extension'' or ``support for traditional practice''.
5849 If you are an experienced user of assemblers, your suggestions for improvement
5850 of @command{@value{AS}} are welcome in any case.
5854 @section How to report bugs
5856 @cindex assembler bugs, reporting
5858 A number of companies and individuals offer support for @sc{gnu} products. If
5859 you obtained @command{@value{AS}} from a support organization, we recommend you
5860 contact that organization first.
5862 You can find contact information for many support companies and
5863 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
5866 In any event, we also recommend that you send bug reports for @command{@value{AS}}
5867 to @samp{bug-binutils@@gnu.org}.
5869 The fundamental principle of reporting bugs usefully is this:
5870 @strong{report all the facts}. If you are not sure whether to state a
5871 fact or leave it out, state it!
5873 Often people omit facts because they think they know what causes the problem
5874 and assume that some details do not matter. Thus, you might assume that the
5875 name of a symbol you use in an example does not matter. Well, probably it does
5876 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
5877 happens to fetch from the location where that name is stored in memory;
5878 perhaps, if the name were different, the contents of that location would fool
5879 the assembler into doing the right thing despite the bug. Play it safe and
5880 give a specific, complete example. That is the easiest thing for you to do,
5881 and the most helpful.
5883 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
5884 it is new to us. Therefore, always write your bug reports on the assumption
5885 that the bug has not been reported previously.
5887 Sometimes people give a few sketchy facts and ask, ``Does this ring a
5888 bell?'' Those bug reports are useless, and we urge everyone to
5889 @emph{refuse to respond to them} except to chide the sender to report
5892 To enable us to fix the bug, you should include all these things:
5896 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
5897 it with the @samp{--version} argument.
5899 Without this, we will not know whether there is any point in looking for
5900 the bug in the current version of @command{@value{AS}}.
5903 Any patches you may have applied to the @command{@value{AS}} source.
5906 The type of machine you are using, and the operating system name and
5910 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
5914 The command arguments you gave the assembler to assemble your example and
5915 observe the bug. To guarantee you will not omit something important, list them
5916 all. A copy of the Makefile (or the output from make) is sufficient.
5918 If we were to try to guess the arguments, we would probably guess wrong
5919 and then we might not encounter the bug.
5922 A complete input file that will reproduce the bug. If the bug is observed when
5923 the assembler is invoked via a compiler, send the assembler source, not the
5924 high level language source. Most compilers will produce the assembler source
5925 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
5926 the options @samp{-v --save-temps}; this will save the assembler source in a
5927 file with an extension of @file{.s}, and also show you exactly how
5928 @command{@value{AS}} is being run.
5931 A description of what behavior you observe that you believe is
5932 incorrect. For example, ``It gets a fatal signal.''
5934 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
5935 will certainly notice it. But if the bug is incorrect output, we might not
5936 notice unless it is glaringly wrong. You might as well not give us a chance to
5939 Even if the problem you experience is a fatal signal, you should still say so
5940 explicitly. Suppose something strange is going on, such as, your copy of
5941 @command{@value{AS}} is out of synch, or you have encountered a bug in the C
5942 library on your system. (This has happened!) Your copy might crash and ours
5943 would not. If you told us to expect a crash, then when ours fails to crash, we
5944 would know that the bug was not happening for us. If you had not told us to
5945 expect a crash, then we would not be able to draw any conclusion from our
5949 If you wish to suggest changes to the @command{@value{AS}} source, send us context
5950 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
5951 option. Always send diffs from the old file to the new file. If you even
5952 discuss something in the @command{@value{AS}} source, refer to it by context, not
5955 The line numbers in our development sources will not match those in your
5956 sources. Your line numbers would convey no useful information to us.
5959 Here are some things that are not necessary:
5963 A description of the envelope of the bug.
5965 Often people who encounter a bug spend a lot of time investigating
5966 which changes to the input file will make the bug go away and which
5967 changes will not affect it.
5969 This is often time consuming and not very useful, because the way we
5970 will find the bug is by running a single example under the debugger
5971 with breakpoints, not by pure deduction from a series of examples.
5972 We recommend that you save your time for something else.
5974 Of course, if you can find a simpler example to report @emph{instead}
5975 of the original one, that is a convenience for us. Errors in the
5976 output will be easier to spot, running under the debugger will take
5977 less time, and so on.
5979 However, simplification is not vital; if you do not want to do this,
5980 report the bug anyway and send us the entire test case you used.
5983 A patch for the bug.
5985 A patch for the bug does help us if it is a good one. But do not omit
5986 the necessary information, such as the test case, on the assumption that
5987 a patch is all we need. We might see problems with your patch and decide
5988 to fix the problem another way, or we might not understand it at all.
5990 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
5991 construct an example that will make the program follow a certain path through
5992 the code. If you do not send us the example, we will not be able to construct
5993 one, so we will not be able to verify that the bug is fixed.
5995 And if we cannot understand what bug you are trying to fix, or why your
5996 patch should be an improvement, we will not install it. A test case will
5997 help us to understand.
6000 A guess about what the bug is or what it depends on.
6002 Such guesses are usually wrong. Even we cannot guess right about such
6003 things without first using the debugger to find the facts.
6006 @node Acknowledgements
6007 @chapter Acknowledgements
6009 If you have contributed to @command{@value{AS}} and your name isn't listed here,
6010 it is not meant as a slight. We just don't know about it. Send mail to the
6011 maintainer, and we'll correct the situation. Currently
6013 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
6015 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
6018 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
6019 information and the 68k series machines, most of the preprocessing pass, and
6020 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
6022 K. Richard Pixley maintained GAS for a while, adding various enhancements and
6023 many bug fixes, including merging support for several processors, breaking GAS
6024 up to handle multiple object file format back ends (including heavy rewrite,
6025 testing, an integration of the coff and b.out back ends), adding configuration
6026 including heavy testing and verification of cross assemblers and file splits
6027 and renaming, converted GAS to strictly ANSI C including full prototypes, added
6028 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
6029 port (including considerable amounts of reverse engineering), a SPARC opcode
6030 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
6031 assertions and made them work, much other reorganization, cleanup, and lint.
6033 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
6034 in format-specific I/O modules.
6036 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
6037 has done much work with it since.
6039 The Intel 80386 machine description was written by Eliot Dresselhaus.
6041 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
6043 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
6044 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
6046 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
6047 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
6048 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
6049 support a.out format.
6051 Support for the Zilog Z8k and Hitachi H8/300 and H8/500 processors (tc-z8k,
6052 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
6053 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
6054 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
6057 John Gilmore built the AMD 29000 support, added @code{.include} support, and
6058 simplified the configuration of which versions accept which directives. He
6059 updated the 68k machine description so that Motorola's opcodes always produced
6060 fixed-size instructions (e.g. @code{jsr}), while synthetic instructions
6061 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
6062 cross-compilation support, and one bug in relaxation that took a week and
6063 required the proverbial one-bit fix.
6065 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
6066 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
6067 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
6068 PowerPC assembler, and made a few other minor patches.
6070 Steve Chamberlain made @command{@value{AS}} able to generate listings.
6072 Hewlett-Packard contributed support for the HP9000/300.
6074 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
6075 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
6076 formats). This work was supported by both the Center for Software Science at
6077 the University of Utah and Cygnus Support.
6079 Support for ELF format files has been worked on by Mark Eichin of Cygnus
6080 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
6081 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
6082 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
6083 and some initial 64-bit support).
6085 Linas Vepstas added GAS support for the ESA/390 "IBM 370" architecture.
6087 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
6088 support for openVMS/Alpha.
6090 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
6093 Several engineers at Cygnus Support have also provided many small bug fixes and
6094 configuration enhancements.
6096 Many others have contributed large or small bugfixes and enhancements. If
6097 you have contributed significant work and are not mentioned on this list, and
6098 want to be, let us know. Some of the history has been lost; we are not
6099 intentionally leaving anyone out.
6101 @node GNU Free Documentation License
6102 @chapter GNU Free Documentation License
6104 GNU Free Documentation License
6106 Version 1.1, March 2000
6108 Copyright (C) 2000 Free Software Foundation, Inc.
6109 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
6111 Everyone is permitted to copy and distribute verbatim copies
6112 of this license document, but changing it is not allowed.
6117 The purpose of this License is to make a manual, textbook, or other
6118 written document "free" in the sense of freedom: to assure everyone
6119 the effective freedom to copy and redistribute it, with or without
6120 modifying it, either commercially or noncommercially. Secondarily,
6121 this License preserves for the author and publisher a way to get
6122 credit for their work, while not being considered responsible for
6123 modifications made by others.
6125 This License is a kind of "copyleft", which means that derivative
6126 works of the document must themselves be free in the same sense. It
6127 complements the GNU General Public License, which is a copyleft
6128 license designed for free software.
6130 We have designed this License in order to use it for manuals for free
6131 software, because free software needs free documentation: a free
6132 program should come with manuals providing the same freedoms that the
6133 software does. But this License is not limited to software manuals;
6134 it can be used for any textual work, regardless of subject matter or
6135 whether it is published as a printed book. We recommend this License
6136 principally for works whose purpose is instruction or reference.
6139 1. APPLICABILITY AND DEFINITIONS
6141 This License applies to any manual or other work that contains a
6142 notice placed by the copyright holder saying it can be distributed
6143 under the terms of this License. The "Document", below, refers to any
6144 such manual or work. Any member of the public is a licensee, and is
6147 A "Modified Version" of the Document means any work containing the
6148 Document or a portion of it, either copied verbatim, or with
6149 modifications and/or translated into another language.
6151 A "Secondary Section" is a named appendix or a front-matter section of
6152 the Document that deals exclusively with the relationship of the
6153 publishers or authors of the Document to the Document's overall subject
6154 (or to related matters) and contains nothing that could fall directly
6155 within that overall subject. (For example, if the Document is in part a
6156 textbook of mathematics, a Secondary Section may not explain any
6157 mathematics.) The relationship could be a matter of historical
6158 connection with the subject or with related matters, or of legal,
6159 commercial, philosophical, ethical or political position regarding
6162 The "Invariant Sections" are certain Secondary Sections whose titles
6163 are designated, as being those of Invariant Sections, in the notice
6164 that says that the Document is released under this License.
6166 The "Cover Texts" are certain short passages of text that are listed,
6167 as Front-Cover Texts or Back-Cover Texts, in the notice that says that
6168 the Document is released under this License.
6170 A "Transparent" copy of the Document means a machine-readable copy,
6171 represented in a format whose specification is available to the
6172 general public, whose contents can be viewed and edited directly and
6173 straightforwardly with generic text editors or (for images composed of
6174 pixels) generic paint programs or (for drawings) some widely available
6175 drawing editor, and that is suitable for input to text formatters or
6176 for automatic translation to a variety of formats suitable for input
6177 to text formatters. A copy made in an otherwise Transparent file
6178 format whose markup has been designed to thwart or discourage
6179 subsequent modification by readers is not Transparent. A copy that is
6180 not "Transparent" is called "Opaque".
6182 Examples of suitable formats for Transparent copies include plain
6183 ASCII without markup, Texinfo input format, LaTeX input format, SGML
6184 or XML using a publicly available DTD, and standard-conforming simple
6185 HTML designed for human modification. Opaque formats include
6186 PostScript, PDF, proprietary formats that can be read and edited only
6187 by proprietary word processors, SGML or XML for which the DTD and/or
6188 processing tools are not generally available, and the
6189 machine-generated HTML produced by some word processors for output
6192 The "Title Page" means, for a printed book, the title page itself,
6193 plus such following pages as are needed to hold, legibly, the material
6194 this License requires to appear in the title page. For works in
6195 formats which do not have any title page as such, "Title Page" means
6196 the text near the most prominent appearance of the work's title,
6197 preceding the beginning of the body of the text.
6202 You may copy and distribute the Document in any medium, either
6203 commercially or noncommercially, provided that this License, the
6204 copyright notices, and the license notice saying this License applies
6205 to the Document are reproduced in all copies, and that you add no other
6206 conditions whatsoever to those of this License. You may not use
6207 technical measures to obstruct or control the reading or further
6208 copying of the copies you make or distribute. However, you may accept
6209 compensation in exchange for copies. If you distribute a large enough
6210 number of copies you must also follow the conditions in section 3.
6212 You may also lend copies, under the same conditions stated above, and
6213 you may publicly display copies.
6216 3. COPYING IN QUANTITY
6218 If you publish printed copies of the Document numbering more than 100,
6219 and the Document's license notice requires Cover Texts, you must enclose
6220 the copies in covers that carry, clearly and legibly, all these Cover
6221 Texts: Front-Cover Texts on the front cover, and Back-Cover Texts on
6222 the back cover. Both covers must also clearly and legibly identify
6223 you as the publisher of these copies. The front cover must present
6224 the full title with all words of the title equally prominent and
6225 visible. You may add other material on the covers in addition.
6226 Copying with changes limited to the covers, as long as they preserve
6227 the title of the Document and satisfy these conditions, can be treated
6228 as verbatim copying in other respects.
6230 If the required texts for either cover are too voluminous to fit
6231 legibly, you should put the first ones listed (as many as fit
6232 reasonably) on the actual cover, and continue the rest onto adjacent
6235 If you publish or distribute Opaque copies of the Document numbering
6236 more than 100, you must either include a machine-readable Transparent
6237 copy along with each Opaque copy, or state in or with each Opaque copy
6238 a publicly-accessible computer-network location containing a complete
6239 Transparent copy of the Document, free of added material, which the
6240 general network-using public has access to download anonymously at no
6241 charge using public-standard network protocols. If you use the latter
6242 option, you must take reasonably prudent steps, when you begin
6243 distribution of Opaque copies in quantity, to ensure that this
6244 Transparent copy will remain thus accessible at the stated location
6245 until at least one year after the last time you distribute an Opaque
6246 copy (directly or through your agents or retailers) of that edition to
6249 It is requested, but not required, that you contact the authors of the
6250 Document well before redistributing any large number of copies, to give
6251 them a chance to provide you with an updated version of the Document.
6256 You may copy and distribute a Modified Version of the Document under
6257 the conditions of sections 2 and 3 above, provided that you release
6258 the Modified Version under precisely this License, with the Modified
6259 Version filling the role of the Document, thus licensing distribution
6260 and modification of the Modified Version to whoever possesses a copy
6261 of it. In addition, you must do these things in the Modified Version:
6263 A. Use in the Title Page (and on the covers, if any) a title distinct
6264 from that of the Document, and from those of previous versions
6265 (which should, if there were any, be listed in the History section
6266 of the Document). You may use the same title as a previous version
6267 if the original publisher of that version gives permission.
6268 B. List on the Title Page, as authors, one or more persons or entities
6269 responsible for authorship of the modifications in the Modified
6270 Version, together with at least five of the principal authors of the
6271 Document (all of its principal authors, if it has less than five).
6272 C. State on the Title page the name of the publisher of the
6273 Modified Version, as the publisher.
6274 D. Preserve all the copyright notices of the Document.
6275 E. Add an appropriate copyright notice for your modifications
6276 adjacent to the other copyright notices.
6277 F. Include, immediately after the copyright notices, a license notice
6278 giving the public permission to use the Modified Version under the
6279 terms of this License, in the form shown in the Addendum below.
6280 G. Preserve in that license notice the full lists of Invariant Sections
6281 and required Cover Texts given in the Document's license notice.
6282 H. Include an unaltered copy of this License.
6283 I. Preserve the section entitled "History", and its title, and add to
6284 it an item stating at least the title, year, new authors, and
6285 publisher of the Modified Version as given on the Title Page. If
6286 there is no section entitled "History" in the Document, create one
6287 stating the title, year, authors, and publisher of the Document as
6288 given on its Title Page, then add an item describing the Modified
6289 Version as stated in the previous sentence.
6290 J. Preserve the network location, if any, given in the Document for
6291 public access to a Transparent copy of the Document, and likewise
6292 the network locations given in the Document for previous versions
6293 it was based on. These may be placed in the "History" section.
6294 You may omit a network location for a work that was published at
6295 least four years before the Document itself, or if the original
6296 publisher of the version it refers to gives permission.
6297 K. In any section entitled "Acknowledgements" or "Dedications",
6298 preserve the section's title, and preserve in the section all the
6299 substance and tone of each of the contributor acknowledgements
6300 and/or dedications given therein.
6301 L. Preserve all the Invariant Sections of the Document,
6302 unaltered in their text and in their titles. Section numbers
6303 or the equivalent are not considered part of the section titles.
6304 M. Delete any section entitled "Endorsements". Such a section
6305 may not be included in the Modified Version.
6306 N. Do not retitle any existing section as "Endorsements"
6307 or to conflict in title with any Invariant Section.
6309 If the Modified Version includes new front-matter sections or
6310 appendices that qualify as Secondary Sections and contain no material
6311 copied from the Document, you may at your option designate some or all
6312 of these sections as invariant. To do this, add their titles to the
6313 list of Invariant Sections in the Modified Version's license notice.
6314 These titles must be distinct from any other section titles.
6316 You may add a section entitled "Endorsements", provided it contains
6317 nothing but endorsements of your Modified Version by various
6318 parties--for example, statements of peer review or that the text has
6319 been approved by an organization as the authoritative definition of a
6322 You may add a passage of up to five words as a Front-Cover Text, and a
6323 passage of up to 25 words as a Back-Cover Text, to the end of the list
6324 of Cover Texts in the Modified Version. Only one passage of
6325 Front-Cover Text and one of Back-Cover Text may be added by (or
6326 through arrangements made by) any one entity. If the Document already
6327 includes a cover text for the same cover, previously added by you or
6328 by arrangement made by the same entity you are acting on behalf of,
6329 you may not add another; but you may replace the old one, on explicit
6330 permission from the previous publisher that added the old one.
6332 The author(s) and publisher(s) of the Document do not by this License
6333 give permission to use their names for publicity for or to assert or
6334 imply endorsement of any Modified Version.
6337 5. COMBINING DOCUMENTS
6339 You may combine the Document with other documents released under this
6340 License, under the terms defined in section 4 above for modified
6341 versions, provided that you include in the combination all of the
6342 Invariant Sections of all of the original documents, unmodified, and
6343 list them all as Invariant Sections of your combined work in its
6346 The combined work need only contain one copy of this License, and
6347 multiple identical Invariant Sections may be replaced with a single
6348 copy. If there are multiple Invariant Sections with the same name but
6349 different contents, make the title of each such section unique by
6350 adding at the end of it, in parentheses, the name of the original
6351 author or publisher of that section if known, or else a unique number.
6352 Make the same adjustment to the section titles in the list of
6353 Invariant Sections in the license notice of the combined work.
6355 In the combination, you must combine any sections entitled "History"
6356 in the various original documents, forming one section entitled
6357 "History"; likewise combine any sections entitled "Acknowledgements",
6358 and any sections entitled "Dedications". You must delete all sections
6359 entitled "Endorsements."
6362 6. COLLECTIONS OF DOCUMENTS
6364 You may make a collection consisting of the Document and other documents
6365 released under this License, and replace the individual copies of this
6366 License in the various documents with a single copy that is included in
6367 the collection, provided that you follow the rules of this License for
6368 verbatim copying of each of the documents in all other respects.
6370 You may extract a single document from such a collection, and distribute
6371 it individually under this License, provided you insert a copy of this
6372 License into the extracted document, and follow this License in all
6373 other respects regarding verbatim copying of that document.
6376 7. AGGREGATION WITH INDEPENDENT WORKS
6378 A compilation of the Document or its derivatives with other separate
6379 and independent documents or works, in or on a volume of a storage or
6380 distribution medium, does not as a whole count as a Modified Version
6381 of the Document, provided no compilation copyright is claimed for the
6382 compilation. Such a compilation is called an "aggregate", and this
6383 License does not apply to the other self-contained works thus compiled
6384 with the Document, on account of their being thus compiled, if they
6385 are not themselves derivative works of the Document.
6387 If the Cover Text requirement of section 3 is applicable to these
6388 copies of the Document, then if the Document is less than one quarter
6389 of the entire aggregate, the Document's Cover Texts may be placed on
6390 covers that surround only the Document within the aggregate.
6391 Otherwise they must appear on covers around the whole aggregate.
6396 Translation is considered a kind of modification, so you may
6397 distribute translations of the Document under the terms of section 4.
6398 Replacing Invariant Sections with translations requires special
6399 permission from their copyright holders, but you may include
6400 translations of some or all Invariant Sections in addition to the
6401 original versions of these Invariant Sections. You may include a
6402 translation of this License provided that you also include the
6403 original English version of this License. In case of a disagreement
6404 between the translation and the original English version of this
6405 License, the original English version will prevail.
6410 You may not copy, modify, sublicense, or distribute the Document except
6411 as expressly provided for under this License. Any other attempt to
6412 copy, modify, sublicense or distribute the Document is void, and will
6413 automatically terminate your rights under this License. However,
6414 parties who have received copies, or rights, from you under this
6415 License will not have their licenses terminated so long as such
6416 parties remain in full compliance.
6419 10. FUTURE REVISIONS OF THIS LICENSE
6421 The Free Software Foundation may publish new, revised versions
6422 of the GNU Free Documentation License from time to time. Such new
6423 versions will be similar in spirit to the present version, but may
6424 differ in detail to address new problems or concerns. See
6425 http://www.gnu.org/copyleft/.
6427 Each version of the License is given a distinguishing version number.
6428 If the Document specifies that a particular numbered version of this
6429 License "or any later version" applies to it, you have the option of
6430 following the terms and conditions either of that specified version or
6431 of any later version that has been published (not as a draft) by the
6432 Free Software Foundation. If the Document does not specify a version
6433 number of this License, you may choose any version ever published (not
6434 as a draft) by the Free Software Foundation.
6437 ADDENDUM: How to use this License for your documents
6439 To use this License in a document you have written, include a copy of
6440 the License in the document and put the following copyright and
6441 license notices just after the title page:
6444 Copyright (c) YEAR YOUR NAME.
6445 Permission is granted to copy, distribute and/or modify this document
6446 under the terms of the GNU Free Documentation License, Version 1.1
6447 or any later version published by the Free Software Foundation;
6448 with the Invariant Sections being LIST THEIR TITLES, with the
6449 Front-Cover Texts being LIST, and with the Back-Cover Texts being LIST.
6450 A copy of the license is included in the section entitled "GNU
6451 Free Documentation License".
6454 If you have no Invariant Sections, write "with no Invariant Sections"
6455 instead of saying which ones are invariant. If you have no
6456 Front-Cover Texts, write "no Front-Cover Texts" instead of
6457 "Front-Cover Texts being LIST"; likewise for Back-Cover Texts.
6459 If your document contains nontrivial examples of program code, we
6460 recommend releasing these examples in parallel under your choice of
6461 free software license, such as the GNU General Public License,
6462 to permit their use in free software.