1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (c) 1991, 92, 93, 94, 95, 96, 97, 98, 2000
3 @c Free Software Foundation, Inc.
4 @c UPDATE!! On future updates--
5 @c (1) check for new machine-dep cmdline options in
6 @c md_parse_option definitions in config/tc-*.c
7 @c (2) for platform-specific directives, examine md_pseudo_op
9 @c (3) for object-format specific directives, examine obj_pseudo_op
11 @c (4) portable directives in potable[] in read.c
15 @c defaults, config file may override:
18 @include asconfig.texi
21 @c common OR combinations of conditions
41 @set abnormal-separator
45 @settitle Using @value{AS}
48 @settitle Using @value{AS} (@value{TARGET})
50 @setchapternewpage odd
55 @c WARE! Some of the machine-dependent sections contain tables of machine
56 @c instructions. Except in multi-column format, these tables look silly.
57 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
58 @c the multi-col format is faked within @example sections.
60 @c Again unfortunately, the natural size that fits on a page, for these tables,
61 @c is different depending on whether or not smallbook is turned on.
62 @c This matters, because of order: text flow switches columns at each page
65 @c The format faked in this source works reasonably well for smallbook,
66 @c not well for the default large-page format. This manual expects that if you
67 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
68 @c tables in question. You can turn on one without the other at your
69 @c discretion, of course.
72 @c the insn tables look just as silly in info files regardless of smallbook,
73 @c might as well show 'em anyways.
79 * As: (as). The GNU assembler.
88 This file documents the GNU Assembler "@value{AS}".
90 Copyright (C) 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000 Free Software Foundation, Inc.
92 Permission is granted to copy, distribute and/or modify this document
93 under the terms of the GNU Free Documentation License, Version 1.1
94 or any later version published by the Free Software Foundation;
95 with no Invariant Sections, with no Front-Cover Texts, and with no
96 Back-Cover Texts. A copy of the license is included in the
97 section entitled "GNU Free Documentation License".
100 Permission is granted to process this file through Tex and print the
101 results, provided the printed document carries copying permission
102 notice identical to this one except for the removal of this paragraph
103 (this paragraph not being relevant to the printed manual).
109 @title Using @value{AS}
110 @subtitle The @sc{gnu} Assembler
112 @subtitle for the @value{TARGET} family
115 @subtitle Version @value{VERSION}
118 The Free Software Foundation Inc. thanks The Nice Computer
119 Company of Australia for loaning Dean Elsner to write the
120 first (Vax) version of @code{as} for Project @sc{gnu}.
121 The proprietors, management and staff of TNCCA thank FSF for
122 distracting the boss while they got some work
125 @author Dean Elsner, Jay Fenlason & friends
129 \hfill {\it Using {\tt @value{AS}}}\par
130 \hfill Edited by Cygnus Support\par
132 %"boxit" macro for figures:
133 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
134 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
135 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
136 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
137 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
140 @vskip 0pt plus 1filll
141 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000 Free Software Foundation, Inc.
143 Permission is granted to copy, distribute and/or modify this document
144 under the terms of the GNU Free Documentation License, Version 1.1
145 or any later version published by the Free Software Foundation;
146 with no Invariant Sections, with no Front-Cover Texts, and with no
147 Back-Cover Texts. A copy of the license is included in the
148 section entitled "GNU Free Documentation License".
154 @top Using @value{AS}
156 This file is a user guide to the @sc{gnu} assembler @code{@value{AS}} version
159 This version of the file describes @code{@value{AS}} configured to generate
160 code for @value{TARGET} architectures.
163 This document is distributed under the terms of the GNU Free
164 Documentation License. A copy of the license is included in the
165 section entitled "GNU Free Documentation License".
168 * Overview:: Overview
169 * Invoking:: Command-Line Options
171 * Sections:: Sections and Relocation
173 * Expressions:: Expressions
174 * Pseudo Ops:: Assembler Directives
175 * Machine Dependencies:: Machine Dependent Features
176 * Reporting Bugs:: Reporting Bugs
177 * Acknowledgements:: Who Did What
178 * GNU Free Documentation License:: GNU Free Documentation License
186 This manual is a user guide to the @sc{gnu} assembler @code{@value{AS}}.
188 This version of the manual describes @code{@value{AS}} configured to generate
189 code for @value{TARGET} architectures.
193 @cindex invocation summary
194 @cindex option summary
195 @cindex summary of options
196 Here is a brief summary of how to invoke @code{@value{AS}}. For details,
197 @pxref{Invoking,,Comand-Line Options}.
199 @c We don't use deffn and friends for the following because they seem
200 @c to be limited to one line for the header.
202 @value{AS} [ -a[cdhlns][=file] ] [ -D ] [ --defsym @var{sym}=@var{val} ]
203 [ -f ] [ --gstabs ] [ --gdwarf2 ] [ --help ] [ -I @var{dir} ] [ -J ] [ -K ] [ -L ]
204 [ --keep-locals ] [ -o @var{objfile} ] [ -R ] [ --statistics ] [ -v ]
205 [ -version ] [ --version ] [ -W ] [ --warn ] [ --fatal-warnings ]
206 [ -w ] [ -x ] [ -Z ] [ --target-help ]
208 @c am29k has no machine-dependent assembler options
211 [ -mbig-endian | -mlittle-endian ]
214 [ -m[arm]1 | -m[arm]2 | -m[arm]250 | -m[arm]3 | -m[arm]6 | -m[arm]60 |
215 -m[arm]600 | -m[arm]610 | -m[arm]620 | -m[arm]7[t][[d]m[i]][fe] | -m[arm]70 |
216 -m[arm]700 | -m[arm]710[c] | -m[arm]7100 | -m[arm]7500 | -m[arm]8 |
217 -m[arm]810 | -m[arm]9 | -m[arm]920 | -m[arm]920t | -m[arm]9tdmi |
218 -mstrongarm | -mstrongarm110 | -mstrongarm1100 ]
219 [ -m[arm]v2 | -m[arm]v2a | -m[arm]v3 | -m[arm]v3m | -m[arm]v4 | -m[arm]v4t |
220 -m[arm]v5 | -[arm]v5t ]
222 [ -mfpa10 | -mfpa11 | -mfpe-old | -mno-fpu ]
224 [ -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant ]
225 [ -mthumb-interwork ]
236 @c Hitachi family chips have no machine-dependent assembler options
239 @c HPPA has no machine-dependent assembler options (yet).
245 @c The order here is important. See c-sparc.texi.
246 [ -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
247 -Av8plus | -Av8plusa | -Av9 | -Av9a ]
248 [ -xarch=v8plus | -xarch=v8plusa ] [ -bump ] [ -32 | -64 ]
251 [ -mcpu=54[123589] | -mcpu=54[56]lp ] [ -mfar-mode | -mf ]
252 [ -merrors-to-file <filename> | -me <filename> ]
255 @c Z8000 has no machine-dependent assembler options
258 @c see md_parse_option in tc-i960.c
259 [ -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC ]
263 [ --m32rx | --[no-]warn-explicit-parallel-conflicts | --W[n]p ]
266 [ -l ] [ -m68000 | -m68010 | -m68020 | ... ]
269 [ -jsri2bsr ] [ -sifilter ] [ -relax ]
273 [ -m68hc11 | -m68hc12 ]
274 [ --force-long-branchs ] [ --short-branchs ] [ --strict-direct-mode ]
275 [ --print-insn-syntax ] [ --print-opcodes ] [ --generate-example ]
278 [ -nocpp ] [ -EL ] [ -EB ] [ -G @var{num} ] [ -mcpu=@var{CPU} ]
279 [ -mips1 ] [ -mips2 ] [ -mips3 ] [ -m4650 ] [ -no-m4650 ] [ -mips32 ] [ -no-mips32 ]
280 [ --trap ] [ --break ]
281 [ --emulation=@var{name} ]
283 [ -- | @var{files} @dots{} ]
288 Turn on listings, in any of a variety of ways:
292 omit false conditionals
295 omit debugging directives
298 include high-level source
304 include macro expansions
307 omit forms processing
313 set the name of the listing file
316 You may combine these options; for example, use @samp{-aln} for assembly
317 listing without forms processing. The @samp{=file} option, if used, must be
318 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
321 Ignored. This option is accepted for script compatibility with calls to
324 @item --defsym @var{sym}=@var{value}
325 Define the symbol @var{sym} to be @var{value} before assembling the input file.
326 @var{value} must be an integer constant. As in C, a leading @samp{0x}
327 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
330 ``fast''---skip whitespace and comment preprocessing (assume source is
334 Generate stabs debugging information for each assembler line. This
335 may help debugging assembler code, if the debugger can handle it.
338 Generate DWARF2 debugging information for each assembler line. This
339 may help debugging assembler code, if the debugger can handle it. Note - this
340 option is only supported by some targets, not all of them.
343 Print a summary of the command line options and exit.
346 Print a summary of all target specific options and exit.
349 Add directory @var{dir} to the search list for @code{.include} directives.
352 Don't warn about signed overflow.
355 @ifclear DIFF-TBL-KLUGE
356 This option is accepted but has no effect on the @value{TARGET} family.
358 @ifset DIFF-TBL-KLUGE
359 Issue warnings when difference tables altered for long displacements.
364 Keep (in the symbol table) local symbols. On traditional a.out systems
365 these start with @samp{L}, but different systems have different local
368 @item -o @var{objfile}
369 Name the object-file output from @code{@value{AS}} @var{objfile}.
372 Fold the data section into the text section.
375 Print the maximum space (in bytes) and total time (in seconds) used by
378 @item --strip-local-absolute
379 Remove local absolute symbols from the outgoing symbol table.
383 Print the @code{as} version.
386 Print the @code{as} version and exit.
390 Suppress warning messages.
392 @item --fatal-warnings
393 Treat warnings as errors.
396 Don't suppress warning messages or treat them as errors.
405 Generate an object file even after errors.
407 @item -- | @var{files} @dots{}
408 Standard input, or source files to assemble.
413 The following options are available when @value{AS} is configured for
418 @cindex ARC endianness
419 @cindex endianness, ARC
420 @cindex big endian output, ARC
422 Generate ``big endian'' format output.
424 @cindex little endian output, ARC
425 @item -mlittle-endian
426 Generate ``little endian'' format output.
432 The following options are available when @value{AS} is configured for the ARM
436 @item -m[arm][1|2|3|6|7|8|9][...]
437 Specify which ARM processor variant is the target.
438 @item -m[arm]v[2|2a|3|3m|4|4t|5|5t]
439 Specify which ARM architecture variant is used by the target.
440 @item -mthumb | -mall
441 Enable or disable Thumb only instruction decoding.
442 @item -mfpa10 | -mfpa11 | -mfpe-old | -mno-fpu
443 Select which Floating Point architcture is the target.
444 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant | -moabi
445 Select which procedure calling convention is in use.
447 Select either big-endian (-EB) or little-endian (-EL) output.
448 @item -mthumb-interwork
449 Specify that the code has been generated with interworking between Thumb and
452 Specify that PIC code has been generated.
457 The following options are available when @value{AS} is configured for
460 @cindex D10V optimization
461 @cindex optimization, D10V
463 Optimize output by parallelizing instructions.
468 The following options are available when @value{AS} is configured for a D30V
471 @cindex D30V optimization
472 @cindex optimization, D30V
474 Optimize output by parallelizing instructions.
478 Warn when nops are generated.
480 @cindex D30V nops after 32-bit multiply
482 Warn when a nop after a 32-bit multiply instruction is generated.
487 The following options are available when @value{AS} is configured for the
488 Intel 80960 processor.
491 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
492 Specify which variant of the 960 architecture is the target.
495 Add code to collect statistics about branches taken.
498 Do not alter compare-and-branch instructions for long displacements;
505 The following options are available when @value{AS} is configured for the
506 Mitsubishi M32R series.
511 Specify which processor in the M32R family is the target. The default
512 is normally the M32R, but this option changes it to the M32RX.
514 @item --warn-explicit-parallel-conflicts or --Wp
515 Produce warning messages when questionable parallel constructs are
518 @item --no-warn-explicit-parallel-conflicts or --Wnp
519 Do not produce warning messages when questionable parallel constructs are
526 The following options are available when @value{AS} is configured for the
527 Motorola 68000 series.
532 Shorten references to undefined symbols, to one word instead of two.
534 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030 | -m68040 | -m68060
535 @itemx | -m68302 | -m68331 | -m68332 | -m68333 | -m68340 | -mcpu32 | -m5200
536 Specify what processor in the 68000 family is the target. The default
537 is normally the 68020, but this can be changed at configuration time.
539 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
540 The target machine does (or does not) have a floating-point coprocessor.
541 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
542 the basic 68000 is not compatible with the 68881, a combination of the
543 two can be specified, since it's possible to do emulation of the
544 coprocessor instructions with the main processor.
546 @item -m68851 | -mno-68851
547 The target machine does (or does not) have a memory-management
548 unit coprocessor. The default is to assume an MMU for 68020 and up.
554 The following options are available when @value{AS} is configured for
555 a picoJava processor.
559 @cindex PJ endianness
560 @cindex endianness, PJ
561 @cindex big endian output, PJ
563 Generate ``big endian'' format output.
565 @cindex little endian output, PJ
567 Generate ``little endian'' format output.
573 The following options are available when @value{AS} is configured for the
574 Motorola 68HC11 or 68HC12 series.
578 @item -m68hc11 | -m68hc12
579 Specify what processor is the target. The default is
580 defined by the configuration option when building the assembler.
582 @item --force-long-branchs
583 Relative branches are turned into absolute ones. This concerns
584 conditional branches, unconditional branches and branches to a
587 @item -S | --short-branchs
588 Do not turn relative branchs into absolute ones
589 when the offset is out of range.
591 @item --strict-direct-mode
592 Do not turn the direct addressing mode into extended addressing mode
593 when the instruction does not support direct addressing mode.
595 @item --print-insn-syntax
596 Print the syntax of instruction in case of error.
598 @item --print-opcodes
599 print the list of instructions with syntax and then exit.
601 @item --generate-example
602 print an example of instruction for each possible instruction and then exit.
603 This option is only useful for testing @code{@value{AS}}.
609 The following options are available when @code{@value{AS}} is configured
610 for the SPARC architecture:
613 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
614 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
615 Explicitly select a variant of the SPARC architecture.
617 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
618 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
620 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
621 UltraSPARC extensions.
623 @item -xarch=v8plus | -xarch=v8plusa
624 For compatibility with the Solaris v9 assembler. These options are
625 equivalent to -Av8plus and -Av8plusa, respectively.
628 Warn when the assembler switches to another architecture.
633 The following options are available when @value{AS} is configured for the 'c54x
638 Enable extended addressing mode. All addresses and relocations will assume
639 extended addressing (usually 23 bits).
640 @item -mcpu=@var{CPU_VERSION}
641 Sets the CPU version being compiled for.
642 @item -merrors-to-file @var{FILENAME}
643 Redirect error output to a file, for broken systems which don't support such
644 behaviour in the shell.
649 The following options are available when @value{AS} is configured for
654 This option sets the largest size of an object that can be referenced
655 implicitly with the @code{gp} register. It is only accepted for targets that
656 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
658 @cindex MIPS endianness
659 @cindex endianness, MIPS
660 @cindex big endian output, MIPS
662 Generate ``big endian'' format output.
664 @cindex little endian output, MIPS
666 Generate ``little endian'' format output.
672 Generate code for a particular MIPS Instruction Set Architecture level.
673 @samp{-mips1} corresponds to the @sc{r2000} and @sc{r3000} processors,
674 @samp{-mips2} to the @sc{r6000} processor, and @samp{-mips3} to the @sc{r4000}
679 Generate code for the MIPS @sc{r4650} chip. This tells the assembler to accept
680 the @samp{mad} and @samp{madu} instruction, and to not schedule @samp{nop}
681 instructions around accesses to the @samp{HI} and @samp{LO} registers.
682 @samp{-no-m4650} turns off this option.
686 Generate code for the @sc{MIPS32} architecture. This tells the assembler to
687 accept ISA level 2 instructions and MIPS32 extensions including some @sc{r4000}
690 @item -mcpu=@var{CPU}
691 Generate code for a particular MIPS cpu. This has little effect on the
692 assembler, but it is passed by @code{@value{GCC}}.
695 @item --emulation=@var{name}
696 This option causes @code{@value{AS}} to emulate @code{@value{AS}} configured
697 for some other target, in all respects, including output format (choosing
698 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
699 debugging information or store symbol table information, and default
700 endianness. The available configuration names are: @samp{mipsecoff},
701 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
702 @samp{mipsbelf}. The first two do not alter the default endianness from that
703 of the primary target for which the assembler was configured; the others change
704 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
705 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
706 selection in any case.
708 This option is currently supported only when the primary target
709 @code{@value{AS}} is configured for is a MIPS ELF or ECOFF target.
710 Furthermore, the primary target or others specified with
711 @samp{--enable-targets=@dots{}} at configuration time must include support for
712 the other format, if both are to be available. For example, the Irix 5
713 configuration includes support for both.
715 Eventually, this option will support more configurations, with more
716 fine-grained control over the assembler's behavior, and will be supported for
720 @code{@value{AS}} ignores this option. It is accepted for compatibility with
728 Control how to deal with multiplication overflow and division by zero.
729 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
730 (and only work for Instruction Set Architecture level 2 and higher);
731 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
737 The following options are available when @value{AS} is configured for
743 Enable or disable the JSRI to BSR transformation. By default this is enabled.
744 The command line option @samp{-nojsri2bsr} can be used to disable it.
748 Enable or disable the silicon filter behaviour. By default this is disabled.
749 The default can be overidden by the @samp{-sifilter} command line option.
752 Alter jump instructions for long displacements.
754 @item -mcpu=[210|340]
755 Select the cpu type on the target hardware. This controls which instructions
759 Assemble for a big endian target.
762 Assemble for a little endian target.
768 * Manual:: Structure of this Manual
769 * GNU Assembler:: The GNU Assembler
770 * Object Formats:: Object File Formats
771 * Command Line:: Command Line
772 * Input Files:: Input Files
773 * Object:: Output (Object) File
774 * Errors:: Error and Warning Messages
778 @section Structure of this Manual
780 @cindex manual, structure and purpose
781 This manual is intended to describe what you need to know to use
782 @sc{gnu} @code{@value{AS}}. We cover the syntax expected in source files, including
783 notation for symbols, constants, and expressions; the directives that
784 @code{@value{AS}} understands; and of course how to invoke @code{@value{AS}}.
787 We also cover special features in the @value{TARGET}
788 configuration of @code{@value{AS}}, including assembler directives.
791 This manual also describes some of the machine-dependent features of
792 various flavors of the assembler.
795 @cindex machine instructions (not covered)
796 On the other hand, this manual is @emph{not} intended as an introduction
797 to programming in assembly language---let alone programming in general!
798 In a similar vein, we make no attempt to introduce the machine
799 architecture; we do @emph{not} describe the instruction set, standard
800 mnemonics, registers or addressing modes that are standard to a
801 particular architecture.
803 You may want to consult the manufacturer's
804 machine architecture manual for this information.
808 For information on the H8/300 machine instruction set, see @cite{H8/300
809 Series Programming Manual} (Hitachi ADE--602--025). For the H8/300H,
810 see @cite{H8/300H Series Programming Manual} (Hitachi).
813 For information on the H8/500 machine instruction set, see @cite{H8/500
814 Series Programming Manual} (Hitachi M21T001).
817 For information on the Hitachi SH machine instruction set, see
818 @cite{SH-Microcomputer User's Manual} (Hitachi Micro Systems, Inc.).
821 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
825 @c I think this is premature---doc@cygnus.com, 17jan1991
827 Throughout this manual, we assume that you are running @dfn{GNU},
828 the portable operating system from the @dfn{Free Software
829 Foundation, Inc.}. This restricts our attention to certain kinds of
830 computer (in particular, the kinds of computers that @sc{gnu} can run on);
831 once this assumption is granted examples and definitions need less
834 @code{@value{AS}} is part of a team of programs that turn a high-level
835 human-readable series of instructions into a low-level
836 computer-readable series of instructions. Different versions of
837 @code{@value{AS}} are used for different kinds of computer.
840 @c There used to be a section "Terminology" here, which defined
841 @c "contents", "byte", "word", and "long". Defining "word" to any
842 @c particular size is confusing when the .word directive may generate 16
843 @c bits on one machine and 32 bits on another; in general, for the user
844 @c version of this manual, none of these terms seem essential to define.
845 @c They were used very little even in the former draft of the manual;
846 @c this draft makes an effort to avoid them (except in names of
850 @section The GNU Assembler
852 @sc{gnu} @code{as} is really a family of assemblers.
854 This manual describes @code{@value{AS}}, a member of that family which is
855 configured for the @value{TARGET} architectures.
857 If you use (or have used) the @sc{gnu} assembler on one architecture, you
858 should find a fairly similar environment when you use it on another
859 architecture. Each version has much in common with the others,
860 including object file formats, most assembler directives (often called
861 @dfn{pseudo-ops}) and assembler syntax.@refill
863 @cindex purpose of @sc{gnu} assembler
864 @code{@value{AS}} is primarily intended to assemble the output of the
865 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
866 @code{@value{LD}}. Nevertheless, we've tried to make @code{@value{AS}}
867 assemble correctly everything that other assemblers for the same
868 machine would assemble.
870 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
873 @c This remark should appear in generic version of manual; assumption
874 @c here is that generic version sets M680x0.
875 This doesn't mean @code{@value{AS}} always uses the same syntax as another
876 assembler for the same architecture; for example, we know of several
877 incompatible versions of 680x0 assembly language syntax.
880 Unlike older assemblers, @code{@value{AS}} is designed to assemble a source
881 program in one pass of the source file. This has a subtle impact on the
882 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
885 @section Object File Formats
887 @cindex object file format
888 The @sc{gnu} assembler can be configured to produce several alternative
889 object file formats. For the most part, this does not affect how you
890 write assembly language programs; but directives for debugging symbols
891 are typically different in different file formats. @xref{Symbol
892 Attributes,,Symbol Attributes}.
895 On the @value{TARGET}, @code{@value{AS}} is configured to produce
896 @value{OBJ-NAME} format object files.
898 @c The following should exhaust all configs that set MULTI-OBJ, ideally
900 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
901 @code{a.out} or COFF format object files.
904 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
905 @code{b.out} or COFF format object files.
908 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
909 SOM or ELF format object files.
914 @section Command Line
916 @cindex command line conventions
917 After the program name @code{@value{AS}}, the command line may contain
918 options and file names. Options may appear in any order, and may be
919 before, after, or between file names. The order of file names is
922 @cindex standard input, as input file
924 @file{--} (two hyphens) by itself names the standard input file
925 explicitly, as one of the files for @code{@value{AS}} to assemble.
927 @cindex options, command line
928 Except for @samp{--} any command line argument that begins with a
929 hyphen (@samp{-}) is an option. Each option changes the behavior of
930 @code{@value{AS}}. No option changes the way another option works. An
931 option is a @samp{-} followed by one or more letters; the case of
932 the letter is important. All options are optional.
934 Some options expect exactly one file name to follow them. The file
935 name may either immediately follow the option's letter (compatible
936 with older assemblers) or it may be the next command argument (@sc{gnu}
937 standard). These two command lines are equivalent:
940 @value{AS} -o my-object-file.o mumble.s
941 @value{AS} -omy-object-file.o mumble.s
948 @cindex source program
950 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
951 describe the program input to one run of @code{@value{AS}}. The program may
952 be in one or more files; how the source is partitioned into files
953 doesn't change the meaning of the source.
955 @c I added "con" prefix to "catenation" just to prove I can overcome my
956 @c APL training... doc@cygnus.com
957 The source program is a concatenation of the text in all the files, in the
960 Each time you run @code{@value{AS}} it assembles exactly one source
961 program. The source program is made up of one or more files.
962 (The standard input is also a file.)
964 You give @code{@value{AS}} a command line that has zero or more input file
965 names. The input files are read (from left file name to right). A
966 command line argument (in any position) that has no special meaning
967 is taken to be an input file name.
969 If you give @code{@value{AS}} no file names it attempts to read one input file
970 from the @code{@value{AS}} standard input, which is normally your terminal. You
971 may have to type @key{ctl-D} to tell @code{@value{AS}} there is no more program
974 Use @samp{--} if you need to explicitly name the standard input file
975 in your command line.
977 If the source is empty, @code{@value{AS}} produces a small, empty object
980 @subheading Filenames and Line-numbers
982 @cindex input file linenumbers
983 @cindex line numbers, in input files
984 There are two ways of locating a line in the input file (or files) and
985 either may be used in reporting error messages. One way refers to a line
986 number in a physical file; the other refers to a line number in a
987 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
989 @dfn{Physical files} are those files named in the command line given
990 to @code{@value{AS}}.
992 @dfn{Logical files} are simply names declared explicitly by assembler
993 directives; they bear no relation to physical files. Logical file names help
994 error messages reflect the original source file, when @code{@value{AS}} source
995 is itself synthesized from other files. @code{@value{AS}} understands the
996 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
997 @ref{File,,@code{.file}}.
1000 @section Output (Object) File
1006 Every time you run @code{@value{AS}} it produces an output file, which is
1007 your assembly language program translated into numbers. This file
1008 is the object file. Its default name is
1016 @code{b.out} when @code{@value{AS}} is configured for the Intel 80960.
1018 You can give it another name by using the @code{-o} option. Conventionally,
1019 object file names end with @file{.o}. The default name is used for historical
1020 reasons: older assemblers were capable of assembling self-contained programs
1021 directly into a runnable program. (For some formats, this isn't currently
1022 possible, but it can be done for the @code{a.out} format.)
1026 The object file is meant for input to the linker @code{@value{LD}}. It contains
1027 assembled program code, information to help @code{@value{LD}} integrate
1028 the assembled program into a runnable file, and (optionally) symbolic
1029 information for the debugger.
1031 @c link above to some info file(s) like the description of a.out.
1032 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1035 @section Error and Warning Messages
1037 @cindex error messsages
1038 @cindex warning messages
1039 @cindex messages from assembler
1040 @code{@value{AS}} may write warnings and error messages to the standard error
1041 file (usually your terminal). This should not happen when a compiler
1042 runs @code{@value{AS}} automatically. Warnings report an assumption made so
1043 that @code{@value{AS}} could keep assembling a flawed program; errors report a
1044 grave problem that stops the assembly.
1046 @cindex format of warning messages
1047 Warning messages have the format
1050 file_name:@b{NNN}:Warning Message Text
1054 @cindex line numbers, in warnings/errors
1055 (where @b{NNN} is a line number). If a logical file name has been given
1056 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1057 the current input file is used. If a logical line number was given
1059 (@pxref{Line,,@code{.line}})
1063 (@pxref{Line,,@code{.line}})
1066 (@pxref{Ln,,@code{.ln}})
1069 then it is used to calculate the number printed,
1070 otherwise the actual line in the current source file is printed. The
1071 message text is intended to be self explanatory (in the grand Unix
1074 @cindex format of error messages
1075 Error messages have the format
1077 file_name:@b{NNN}:FATAL:Error Message Text
1079 The file name and line number are derived as for warning
1080 messages. The actual message text may be rather less explanatory
1081 because many of them aren't supposed to happen.
1084 @chapter Command-Line Options
1086 @cindex options, all versions of assembler
1087 This chapter describes command-line options available in @emph{all}
1088 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
1090 to the @value{TARGET}.
1093 to particular machine architectures.
1096 If you are invoking @code{@value{AS}} via the @sc{gnu} C compiler (version 2),
1097 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1098 The assembler arguments must be separated from each other (and the @samp{-Wa})
1099 by commas. For example:
1102 gcc -c -g -O -Wa,-alh,-L file.c
1106 This passes two options to the assembler: @samp{-alh} (emit a listing to
1107 standard output with with high-level and assembly source) and @samp{-L} (retain
1108 local symbols in the symbol table).
1110 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1111 command-line options are automatically passed to the assembler by the compiler.
1112 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1113 precisely what options it passes to each compilation pass, including the
1117 * a:: -a[cdhlns] enable listings
1118 * D:: -D for compatibility
1119 * f:: -f to work faster
1120 * I:: -I for .include search path
1121 @ifclear DIFF-TBL-KLUGE
1122 * K:: -K for compatibility
1124 @ifset DIFF-TBL-KLUGE
1125 * K:: -K for difference tables
1128 * L:: -L to retain local labels
1129 * M:: -M or --mri to assemble in MRI compatibility mode
1130 * MD:: --MD for dependency tracking
1131 * o:: -o to name the object file
1132 * R:: -R to join data and text sections
1133 * statistics:: --statistics to see statistics about assembly
1134 * traditional-format:: --traditional-format for compatible output
1135 * v:: -v to announce version
1136 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1137 * Z:: -Z to make object file even after errors
1141 @section Enable Listings: @code{-a[cdhlns]}
1150 @cindex listings, enabling
1151 @cindex assembly listings, enabling
1153 These options enable listing output from the assembler. By itself,
1154 @samp{-a} requests high-level, assembly, and symbols listing.
1155 You can use other letters to select specific options for the list:
1156 @samp{-ah} requests a high-level language listing,
1157 @samp{-al} requests an output-program assembly listing, and
1158 @samp{-as} requests a symbol table listing.
1159 High-level listings require that a compiler debugging option like
1160 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1163 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1164 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1165 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1166 omitted from the listing.
1168 Use the @samp{-ad} option to omit debugging directives from the
1171 Once you have specified one of these options, you can further control
1172 listing output and its appearance using the directives @code{.list},
1173 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1175 The @samp{-an} option turns off all forms processing.
1176 If you do not request listing output with one of the @samp{-a} options, the
1177 listing-control directives have no effect.
1179 The letters after @samp{-a} may be combined into one option,
1180 @emph{e.g.}, @samp{-aln}.
1186 This option has no effect whatsoever, but it is accepted to make it more
1187 likely that scripts written for other assemblers also work with
1191 @section Work Faster: @code{-f}
1194 @cindex trusted compiler
1195 @cindex faster processing (@code{-f})
1196 @samp{-f} should only be used when assembling programs written by a
1197 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1198 and comment preprocessing on
1199 the input file(s) before assembling them. @xref{Preprocessing,
1203 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1204 preprocessed (if they contain comments, for example), @code{@value{AS}} does
1209 @section @code{.include} search path: @code{-I} @var{path}
1211 @kindex -I @var{path}
1212 @cindex paths for @code{.include}
1213 @cindex search path for @code{.include}
1214 @cindex @code{include} directive search path
1215 Use this option to add a @var{path} to the list of directories
1216 @code{@value{AS}} searches for files specified in @code{.include}
1217 directives (@pxref{Include,,@code{.include}}). You may use @code{-I} as
1218 many times as necessary to include a variety of paths. The current
1219 working directory is always searched first; after that, @code{@value{AS}}
1220 searches any @samp{-I} directories in the same order as they were
1221 specified (left to right) on the command line.
1224 @section Difference Tables: @code{-K}
1227 @ifclear DIFF-TBL-KLUGE
1228 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1229 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1230 where it can be used to warn when the assembler alters the machine code
1231 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1232 family does not have the addressing limitations that sometimes lead to this
1233 alteration on other platforms.
1236 @ifset DIFF-TBL-KLUGE
1237 @cindex difference tables, warning
1238 @cindex warning for altered difference tables
1239 @code{@value{AS}} sometimes alters the code emitted for directives of the form
1240 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1241 You can use the @samp{-K} option if you want a warning issued when this
1246 @section Include Local Labels: @code{-L}
1249 @cindex local labels, retaining in output
1250 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1251 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1252 debugging, because they are intended for the use of programs (like
1253 compilers) that compose assembler programs, not for your notice.
1254 Normally both @code{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1255 normally debug with them.
1257 This option tells @code{@value{AS}} to retain those @samp{L@dots{}} symbols
1258 in the object file. Usually if you do this you also tell the linker
1259 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1261 By default, a local label is any label beginning with @samp{L}, but each
1262 target is allowed to redefine the local label prefix.
1264 On the HPPA local labels begin with @samp{L$}.
1268 @section Assemble in MRI Compatibility Mode: @code{-M}
1271 @cindex MRI compatibility mode
1272 The @code{-M} or @code{--mri} option selects MRI compatibility mode. This
1273 changes the syntax and pseudo-op handling of @code{@value{AS}} to make it
1274 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1275 configured target) assembler from Microtec Research. The exact nature of the
1276 MRI syntax will not be documented here; see the MRI manuals for more
1277 information. Note in particular that the handling of macros and macro
1278 arguments is somewhat different. The purpose of this option is to permit
1279 assembling existing MRI assembler code using @code{@value{AS}}.
1281 The MRI compatibility is not complete. Certain operations of the MRI assembler
1282 depend upon its object file format, and can not be supported using other object
1283 file formats. Supporting these would require enhancing each object file format
1284 individually. These are:
1287 @item global symbols in common section
1289 The m68k MRI assembler supports common sections which are merged by the linker.
1290 Other object file formats do not support this. @code{@value{AS}} handles
1291 common sections by treating them as a single common symbol. It permits local
1292 symbols to be defined within a common section, but it can not support global
1293 symbols, since it has no way to describe them.
1295 @item complex relocations
1297 The MRI assemblers support relocations against a negated section address, and
1298 relocations which combine the start addresses of two or more sections. These
1299 are not support by other object file formats.
1301 @item @code{END} pseudo-op specifying start address
1303 The MRI @code{END} pseudo-op permits the specification of a start address.
1304 This is not supported by other object file formats. The start address may
1305 instead be specified using the @code{-e} option to the linker, or in a linker
1308 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1310 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1311 name to the output file. This is not supported by other object file formats.
1313 @item @code{ORG} pseudo-op
1315 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1316 address. This differs from the usual @code{@value{AS}} @code{.org} pseudo-op,
1317 which changes the location within the current section. Absolute sections are
1318 not supported by other object file formats. The address of a section may be
1319 assigned within a linker script.
1322 There are some other features of the MRI assembler which are not supported by
1323 @code{@value{AS}}, typically either because they are difficult or because they
1324 seem of little consequence. Some of these may be supported in future releases.
1328 @item EBCDIC strings
1330 EBCDIC strings are not supported.
1332 @item packed binary coded decimal
1334 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1335 and @code{DCB.P} pseudo-ops are not supported.
1337 @item @code{FEQU} pseudo-op
1339 The m68k @code{FEQU} pseudo-op is not supported.
1341 @item @code{NOOBJ} pseudo-op
1343 The m68k @code{NOOBJ} pseudo-op is not supported.
1345 @item @code{OPT} branch control options
1347 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1348 @code{BRL}, and @code{BRW}---are ignored. @code{@value{AS}} automatically
1349 relaxes all branches, whether forward or backward, to an appropriate size, so
1350 these options serve no purpose.
1352 @item @code{OPT} list control options
1354 The following m68k @code{OPT} list control options are ignored: @code{C},
1355 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1356 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1358 @item other @code{OPT} options
1360 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1361 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1363 @item @code{OPT} @code{D} option is default
1365 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1366 @code{OPT NOD} may be used to turn it off.
1368 @item @code{XREF} pseudo-op.
1370 The m68k @code{XREF} pseudo-op is ignored.
1372 @item @code{.debug} pseudo-op
1374 The i960 @code{.debug} pseudo-op is not supported.
1376 @item @code{.extended} pseudo-op
1378 The i960 @code{.extended} pseudo-op is not supported.
1380 @item @code{.list} pseudo-op.
1382 The various options of the i960 @code{.list} pseudo-op are not supported.
1384 @item @code{.optimize} pseudo-op
1386 The i960 @code{.optimize} pseudo-op is not supported.
1388 @item @code{.output} pseudo-op
1390 The i960 @code{.output} pseudo-op is not supported.
1392 @item @code{.setreal} pseudo-op
1394 The i960 @code{.setreal} pseudo-op is not supported.
1399 @section Dependency tracking: @code{--MD}
1402 @cindex dependency tracking
1405 @code{@value{AS}} can generate a dependency file for the file it creates. This
1406 file consists of a single rule suitable for @code{make} describing the
1407 dependencies of the main source file.
1409 The rule is written to the file named in its argument.
1411 This feature is used in the automatic updating of makefiles.
1414 @section Name the Object File: @code{-o}
1417 @cindex naming object file
1418 @cindex object file name
1419 There is always one object file output when you run @code{@value{AS}}. By
1420 default it has the name
1423 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1437 You use this option (which takes exactly one filename) to give the
1438 object file a different name.
1440 Whatever the object file is called, @code{@value{AS}} overwrites any
1441 existing file of the same name.
1444 @section Join Data and Text Sections: @code{-R}
1447 @cindex data and text sections, joining
1448 @cindex text and data sections, joining
1449 @cindex joining text and data sections
1450 @cindex merging text and data sections
1451 @code{-R} tells @code{@value{AS}} to write the object file as if all
1452 data-section data lives in the text section. This is only done at
1453 the very last moment: your binary data are the same, but data
1454 section parts are relocated differently. The data section part of
1455 your object file is zero bytes long because all its bytes are
1456 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1458 When you specify @code{-R} it would be possible to generate shorter
1459 address displacements (because we do not have to cross between text and
1460 data section). We refrain from doing this simply for compatibility with
1461 older versions of @code{@value{AS}}. In future, @code{-R} may work this way.
1464 When @code{@value{AS}} is configured for COFF output,
1465 this option is only useful if you use sections named @samp{.text} and
1470 @code{-R} is not supported for any of the HPPA targets. Using
1471 @code{-R} generates a warning from @code{@value{AS}}.
1475 @section Display Assembly Statistics: @code{--statistics}
1477 @kindex --statistics
1478 @cindex statistics, about assembly
1479 @cindex time, total for assembly
1480 @cindex space used, maximum for assembly
1481 Use @samp{--statistics} to display two statistics about the resources used by
1482 @code{@value{AS}}: the maximum amount of space allocated during the assembly
1483 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1486 @node traditional-format
1487 @section Compatible output: @code{--traditional-format}
1489 @kindex --traditional-format
1490 For some targets, the output of @code{@value{AS}} is different in some ways
1491 from the output of some existing assembler. This switch requests
1492 @code{@value{AS}} to use the traditional format instead.
1494 For example, it disables the exception frame optimizations which
1495 @code{@value{AS}} normally does by default on @code{@value{GCC}} output.
1498 @section Announce Version: @code{-v}
1502 @cindex assembler version
1503 @cindex version of assembler
1504 You can find out what version of as is running by including the
1505 option @samp{-v} (which you can also spell as @samp{-version}) on the
1509 @section Control Warnings: @code{-W}, @code{--warn}, @code{--no-warn}, @code{--fatal-warnings}
1511 @code{@value{AS}} should never give a warning or error message when
1512 assembling compiler output. But programs written by people often
1513 cause @code{@value{AS}} to give a warning that a particular assumption was
1514 made. All such warnings are directed to the standard error file.
1517 @kindex @samp{--no-warn}
1518 @cindex suppressing warnings
1519 @cindex warnings, suppressing
1520 If you use the @code{-W} and @code{--no-warn} options, no warnings are issued.
1521 This only affects the warning messages: it does not change any particular of
1522 how @code{@value{AS}} assembles your file. Errors, which stop the assembly,
1525 @kindex @samp{--fatal-warnings}
1526 @cindex errors, caused by warnings
1527 @cindex warnings, causing error
1528 If you use the @code{--fatal-warnings} option, @code{@value{AS}} considers
1529 files that generate warnings to be in error.
1531 @kindex @samp{--warn}
1532 @cindex warnings, switching on
1533 You can switch these options off again by specifying @code{--warn}, which
1534 causes warnings to be output as usual.
1537 @section Generate Object File in Spite of Errors: @code{-Z}
1538 @cindex object file, after errors
1539 @cindex errors, continuing after
1540 After an error message, @code{@value{AS}} normally produces no output. If for
1541 some reason you are interested in object file output even after
1542 @code{@value{AS}} gives an error message on your program, use the @samp{-Z}
1543 option. If there are any errors, @code{@value{AS}} continues anyways, and
1544 writes an object file after a final warning message of the form @samp{@var{n}
1545 errors, @var{m} warnings, generating bad object file.}
1550 @cindex machine-independent syntax
1551 @cindex syntax, machine-independent
1552 This chapter describes the machine-independent syntax allowed in a
1553 source file. @code{@value{AS}} syntax is similar to what many other
1554 assemblers use; it is inspired by the BSD 4.2
1559 assembler, except that @code{@value{AS}} does not assemble Vax bit-fields.
1563 * Preprocessing:: Preprocessing
1564 * Whitespace:: Whitespace
1565 * Comments:: Comments
1566 * Symbol Intro:: Symbols
1567 * Statements:: Statements
1568 * Constants:: Constants
1572 @section Preprocessing
1574 @cindex preprocessing
1575 The @code{@value{AS}} internal preprocessor:
1577 @cindex whitespace, removed by preprocessor
1579 adjusts and removes extra whitespace. It leaves one space or tab before
1580 the keywords on a line, and turns any other whitespace on the line into
1583 @cindex comments, removed by preprocessor
1585 removes all comments, replacing them with a single space, or an
1586 appropriate number of newlines.
1588 @cindex constants, converted by preprocessor
1590 converts character constants into the appropriate numeric values.
1593 It does not do macro processing, include file handling, or
1594 anything else you may get from your C compiler's preprocessor. You can
1595 do include file processing with the @code{.include} directive
1596 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
1597 to get other ``CPP'' style preprocessing, by giving the input file a
1598 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
1599 Output, gcc.info, Using GNU CC}.
1601 Excess whitespace, comments, and character constants
1602 cannot be used in the portions of the input text that are not
1605 @cindex turning preprocessing on and off
1606 @cindex preprocessing, turning on and off
1609 If the first line of an input file is @code{#NO_APP} or if you use the
1610 @samp{-f} option, whitespace and comments are not removed from the input file.
1611 Within an input file, you can ask for whitespace and comment removal in
1612 specific portions of the by putting a line that says @code{#APP} before the
1613 text that may contain whitespace or comments, and putting a line that says
1614 @code{#NO_APP} after this text. This feature is mainly intend to support
1615 @code{asm} statements in compilers whose output is otherwise free of comments
1622 @dfn{Whitespace} is one or more blanks or tabs, in any order.
1623 Whitespace is used to separate symbols, and to make programs neater for
1624 people to read. Unless within character constants
1625 (@pxref{Characters,,Character Constants}), any whitespace means the same
1626 as exactly one space.
1632 There are two ways of rendering comments to @code{@value{AS}}. In both
1633 cases the comment is equivalent to one space.
1635 Anything from @samp{/*} through the next @samp{*/} is a comment.
1636 This means you may not nest these comments.
1640 The only way to include a newline ('\n') in a comment
1641 is to use this sort of comment.
1644 /* This sort of comment does not nest. */
1647 @cindex line comment character
1648 Anything from the @dfn{line comment} character to the next newline
1649 is considered a comment and is ignored. The line comment character is
1651 @samp{;} for the AMD 29K family;
1654 @samp{;} on the ARC;
1657 @samp{@@} on the ARM;
1660 @samp{;} for the H8/300 family;
1663 @samp{!} for the H8/500 family;
1666 @samp{;} for the HPPA;
1669 @samp{#} on the i960;
1672 @samp{;} for picoJava;
1675 @samp{!} for the Hitachi SH;
1678 @samp{!} on the SPARC;
1681 @samp{#} on the m32r;
1684 @samp{|} on the 680x0;
1687 @samp{#} on the 68HC11 and 68HC12;
1690 @samp{#} on the Vax;
1693 @samp{!} for the Z8000;
1696 @samp{#} on the V850;
1698 see @ref{Machine Dependencies}. @refill
1699 @c FIXME What about i386, m88k, i860?
1702 On some machines there are two different line comment characters. One
1703 character only begins a comment if it is the first non-whitespace character on
1704 a line, while the other always begins a comment.
1708 The V850 assembler also supports a double dash as starting a comment that
1709 extends to the end of the line.
1715 @cindex lines starting with @code{#}
1716 @cindex logical line numbers
1717 To be compatible with past assemblers, lines that begin with @samp{#} have a
1718 special interpretation. Following the @samp{#} should be an absolute
1719 expression (@pxref{Expressions}): the logical line number of the @emph{next}
1720 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
1721 new logical file name. The rest of the line, if any, should be whitespace.
1723 If the first non-whitespace characters on the line are not numeric,
1724 the line is ignored. (Just like a comment.)
1727 # This is an ordinary comment.
1728 # 42-6 "new_file_name" # New logical file name
1729 # This is logical line # 36.
1731 This feature is deprecated, and may disappear from future versions
1732 of @code{@value{AS}}.
1737 @cindex characters used in symbols
1738 @ifclear SPECIAL-SYMS
1739 A @dfn{symbol} is one or more characters chosen from the set of all
1740 letters (both upper and lower case), digits and the three characters
1746 A @dfn{symbol} is one or more characters chosen from the set of all
1747 letters (both upper and lower case), digits and the three characters
1748 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
1754 On most machines, you can also use @code{$} in symbol names; exceptions
1755 are noted in @ref{Machine Dependencies}.
1757 No symbol may begin with a digit. Case is significant.
1758 There is no length limit: all characters are significant. Symbols are
1759 delimited by characters not in that set, or by the beginning of a file
1760 (since the source program must end with a newline, the end of a file is
1761 not a possible symbol delimiter). @xref{Symbols}.
1762 @cindex length of symbols
1767 @cindex statements, structure of
1768 @cindex line separator character
1769 @cindex statement separator character
1771 @ifclear abnormal-separator
1772 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
1773 semicolon (@samp{;}). The newline or semicolon is considered part of
1774 the preceding statement. Newlines and semicolons within character
1775 constants are an exception: they do not end statements.
1777 @ifset abnormal-separator
1779 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
1780 sign (@samp{@@}). The newline or at sign is considered part of the
1781 preceding statement. Newlines and at signs within character constants
1782 are an exception: they do not end statements.
1785 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
1786 point (@samp{!}). The newline or exclamation point is considered part of the
1787 preceding statement. Newlines and exclamation points within character
1788 constants are an exception: they do not end statements.
1791 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
1792 H8/300) a dollar sign (@samp{$}); or (for the
1795 (@samp{;}). The newline or separator character is considered part of
1796 the preceding statement. Newlines and separators within character
1797 constants are an exception: they do not end statements.
1802 A @dfn{statement} ends at a newline character (@samp{\n}) or line
1803 separator character. (The line separator is usually @samp{;}, unless
1804 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
1805 newline or separator character is considered part of the preceding
1806 statement. Newlines and separators within character constants are an
1807 exception: they do not end statements.
1810 @cindex newline, required at file end
1811 @cindex EOF, newline must precede
1812 It is an error to end any statement with end-of-file: the last
1813 character of any input file should be a newline.@refill
1815 An empty statement is allowed, and may include whitespace. It is ignored.
1817 @cindex instructions and directives
1818 @cindex directives and instructions
1819 @c "key symbol" is not used elsewhere in the document; seems pedantic to
1820 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
1822 A statement begins with zero or more labels, optionally followed by a
1823 key symbol which determines what kind of statement it is. The key
1824 symbol determines the syntax of the rest of the statement. If the
1825 symbol begins with a dot @samp{.} then the statement is an assembler
1826 directive: typically valid for any computer. If the symbol begins with
1827 a letter the statement is an assembly language @dfn{instruction}: it
1828 assembles into a machine language instruction.
1830 Different versions of @code{@value{AS}} for different computers
1831 recognize different instructions. In fact, the same symbol may
1832 represent a different instruction in a different computer's assembly
1836 @cindex @code{:} (label)
1837 @cindex label (@code{:})
1838 A label is a symbol immediately followed by a colon (@code{:}).
1839 Whitespace before a label or after a colon is permitted, but you may not
1840 have whitespace between a label's symbol and its colon. @xref{Labels}.
1843 For HPPA targets, labels need not be immediately followed by a colon, but
1844 the definition of a label must begin in column zero. This also implies that
1845 only one label may be defined on each line.
1849 label: .directive followed by something
1850 another_label: # This is an empty statement.
1851 instruction operand_1, operand_2, @dots{}
1858 A constant is a number, written so that its value is known by
1859 inspection, without knowing any context. Like this:
1862 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
1863 .ascii "Ring the bell\7" # A string constant.
1864 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
1865 .float 0f-314159265358979323846264338327\
1866 95028841971.693993751E-40 # - pi, a flonum.
1871 * Characters:: Character Constants
1872 * Numbers:: Number Constants
1876 @subsection Character Constants
1878 @cindex character constants
1879 @cindex constants, character
1880 There are two kinds of character constants. A @dfn{character} stands
1881 for one character in one byte and its value may be used in
1882 numeric expressions. String constants (properly called string
1883 @emph{literals}) are potentially many bytes and their values may not be
1884 used in arithmetic expressions.
1888 * Chars:: Characters
1892 @subsubsection Strings
1894 @cindex string constants
1895 @cindex constants, string
1896 A @dfn{string} is written between double-quotes. It may contain
1897 double-quotes or null characters. The way to get special characters
1898 into a string is to @dfn{escape} these characters: precede them with
1899 a backslash @samp{\} character. For example @samp{\\} represents
1900 one backslash: the first @code{\} is an escape which tells
1901 @code{@value{AS}} to interpret the second character literally as a backslash
1902 (which prevents @code{@value{AS}} from recognizing the second @code{\} as an
1903 escape character). The complete list of escapes follows.
1905 @cindex escape codes, character
1906 @cindex character escape codes
1909 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
1911 @cindex @code{\b} (backspace character)
1912 @cindex backspace (@code{\b})
1914 Mnemonic for backspace; for ASCII this is octal code 010.
1917 @c Mnemonic for EOText; for ASCII this is octal code 004.
1919 @cindex @code{\f} (formfeed character)
1920 @cindex formfeed (@code{\f})
1922 Mnemonic for FormFeed; for ASCII this is octal code 014.
1924 @cindex @code{\n} (newline character)
1925 @cindex newline (@code{\n})
1927 Mnemonic for newline; for ASCII this is octal code 012.
1930 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
1932 @cindex @code{\r} (carriage return character)
1933 @cindex carriage return (@code{\r})
1935 Mnemonic for carriage-Return; for ASCII this is octal code 015.
1938 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
1939 @c other assemblers.
1941 @cindex @code{\t} (tab)
1942 @cindex tab (@code{\t})
1944 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
1947 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
1948 @c @item \x @var{digit} @var{digit} @var{digit}
1949 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
1951 @cindex @code{\@var{ddd}} (octal character code)
1952 @cindex octal character code (@code{\@var{ddd}})
1953 @item \ @var{digit} @var{digit} @var{digit}
1954 An octal character code. The numeric code is 3 octal digits.
1955 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
1956 for example, @code{\008} has the value 010, and @code{\009} the value 011.
1958 @cindex @code{\@var{xd...}} (hex character code)
1959 @cindex hex character code (@code{\@var{xd...}})
1960 @item \@code{x} @var{hex-digits...}
1961 A hex character code. All trailing hex digits are combined. Either upper or
1962 lower case @code{x} works.
1964 @cindex @code{\\} (@samp{\} character)
1965 @cindex backslash (@code{\\})
1967 Represents one @samp{\} character.
1970 @c Represents one @samp{'} (accent acute) character.
1971 @c This is needed in single character literals
1972 @c (@xref{Characters,,Character Constants}.) to represent
1975 @cindex @code{\"} (doublequote character)
1976 @cindex doublequote (@code{\"})
1978 Represents one @samp{"} character. Needed in strings to represent
1979 this character, because an unescaped @samp{"} would end the string.
1981 @item \ @var{anything-else}
1982 Any other character when escaped by @kbd{\} gives a warning, but
1983 assembles as if the @samp{\} was not present. The idea is that if
1984 you used an escape sequence you clearly didn't want the literal
1985 interpretation of the following character. However @code{@value{AS}} has no
1986 other interpretation, so @code{@value{AS}} knows it is giving you the wrong
1987 code and warns you of the fact.
1990 Which characters are escapable, and what those escapes represent,
1991 varies widely among assemblers. The current set is what we think
1992 the BSD 4.2 assembler recognizes, and is a subset of what most C
1993 compilers recognize. If you are in doubt, do not use an escape
1997 @subsubsection Characters
1999 @cindex single character constant
2000 @cindex character, single
2001 @cindex constant, single character
2002 A single character may be written as a single quote immediately
2003 followed by that character. The same escapes apply to characters as
2004 to strings. So if you want to write the character backslash, you
2005 must write @kbd{'\\} where the first @code{\} escapes the second
2006 @code{\}. As you can see, the quote is an acute accent, not a
2007 grave accent. A newline
2009 @ifclear abnormal-separator
2010 (or semicolon @samp{;})
2012 @ifset abnormal-separator
2014 (or at sign @samp{@@})
2017 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2023 immediately following an acute accent is taken as a literal character
2024 and does not count as the end of a statement. The value of a character
2025 constant in a numeric expression is the machine's byte-wide code for
2026 that character. @code{@value{AS}} assumes your character code is ASCII:
2027 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2030 @subsection Number Constants
2032 @cindex constants, number
2033 @cindex number constants
2034 @code{@value{AS}} distinguishes three kinds of numbers according to how they
2035 are stored in the target machine. @emph{Integers} are numbers that
2036 would fit into an @code{int} in the C language. @emph{Bignums} are
2037 integers, but they are stored in more than 32 bits. @emph{Flonums}
2038 are floating point numbers, described below.
2041 * Integers:: Integers
2046 * Bit Fields:: Bit Fields
2052 @subsubsection Integers
2054 @cindex constants, integer
2056 @cindex binary integers
2057 @cindex integers, binary
2058 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2059 the binary digits @samp{01}.
2061 @cindex octal integers
2062 @cindex integers, octal
2063 An octal integer is @samp{0} followed by zero or more of the octal
2064 digits (@samp{01234567}).
2066 @cindex decimal integers
2067 @cindex integers, decimal
2068 A decimal integer starts with a non-zero digit followed by zero or
2069 more digits (@samp{0123456789}).
2071 @cindex hexadecimal integers
2072 @cindex integers, hexadecimal
2073 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2074 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2076 Integers have the usual values. To denote a negative integer, use
2077 the prefix operator @samp{-} discussed under expressions
2078 (@pxref{Prefix Ops,,Prefix Operators}).
2081 @subsubsection Bignums
2084 @cindex constants, bignum
2085 A @dfn{bignum} has the same syntax and semantics as an integer
2086 except that the number (or its negative) takes more than 32 bits to
2087 represent in binary. The distinction is made because in some places
2088 integers are permitted while bignums are not.
2091 @subsubsection Flonums
2093 @cindex floating point numbers
2094 @cindex constants, floating point
2096 @cindex precision, floating point
2097 A @dfn{flonum} represents a floating point number. The translation is
2098 indirect: a decimal floating point number from the text is converted by
2099 @code{@value{AS}} to a generic binary floating point number of more than
2100 sufficient precision. This generic floating point number is converted
2101 to a particular computer's floating point format (or formats) by a
2102 portion of @code{@value{AS}} specialized to that computer.
2104 A flonum is written by writing (in order)
2109 (@samp{0} is optional on the HPPA.)
2113 A letter, to tell @code{@value{AS}} the rest of the number is a flonum.
2115 @kbd{e} is recommended. Case is not important.
2117 @c FIXME: verify if flonum syntax really this vague for most cases
2118 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2119 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2122 On the H8/300, H8/500,
2124 and AMD 29K architectures, the letter must be
2125 one of the letters @samp{DFPRSX} (in upper or lower case).
2127 On the ARC, the letter must be one of the letters @samp{DFRS}
2128 (in upper or lower case).
2130 On the Intel 960 architecture, the letter must be
2131 one of the letters @samp{DFT} (in upper or lower case).
2133 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2137 One of the letters @samp{DFPRSX} (in upper or lower case).
2140 One of the letters @samp{DFRS} (in upper or lower case).
2143 One of the letters @samp{DFPRSX} (in upper or lower case).
2146 The letter @samp{E} (upper case only).
2149 One of the letters @samp{DFT} (in upper or lower case).
2154 An optional sign: either @samp{+} or @samp{-}.
2157 An optional @dfn{integer part}: zero or more decimal digits.
2160 An optional @dfn{fractional part}: @samp{.} followed by zero
2161 or more decimal digits.
2164 An optional exponent, consisting of:
2168 An @samp{E} or @samp{e}.
2169 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2170 @c principle this can perfectly well be different on different targets.
2172 Optional sign: either @samp{+} or @samp{-}.
2174 One or more decimal digits.
2179 At least one of the integer part or the fractional part must be
2180 present. The floating point number has the usual base-10 value.
2182 @code{@value{AS}} does all processing using integers. Flonums are computed
2183 independently of any floating point hardware in the computer running
2188 @c Bit fields are written as a general facility but are also controlled
2189 @c by a conditional-compilation flag---which is as of now (21mar91)
2190 @c turned on only by the i960 config of GAS.
2192 @subsubsection Bit Fields
2195 @cindex constants, bit field
2196 You can also define numeric constants as @dfn{bit fields}.
2197 specify two numbers separated by a colon---
2199 @var{mask}:@var{value}
2202 @code{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2205 The resulting number is then packed
2207 @c this conditional paren in case bit fields turned on elsewhere than 960
2208 (in host-dependent byte order)
2210 into a field whose width depends on which assembler directive has the
2211 bit-field as its argument. Overflow (a result from the bitwise and
2212 requiring more binary digits to represent) is not an error; instead,
2213 more constants are generated, of the specified width, beginning with the
2214 least significant digits.@refill
2216 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2217 @code{.short}, and @code{.word} accept bit-field arguments.
2222 @chapter Sections and Relocation
2227 * Secs Background:: Background
2228 * Ld Sections:: Linker Sections
2229 * As Sections:: Assembler Internal Sections
2230 * Sub-Sections:: Sub-Sections
2234 @node Secs Background
2237 Roughly, a section is a range of addresses, with no gaps; all data
2238 ``in'' those addresses is treated the same for some particular purpose.
2239 For example there may be a ``read only'' section.
2241 @cindex linker, and assembler
2242 @cindex assembler, and linker
2243 The linker @code{@value{LD}} reads many object files (partial programs) and
2244 combines their contents to form a runnable program. When @code{@value{AS}}
2245 emits an object file, the partial program is assumed to start at address 0.
2246 @code{@value{LD}} assigns the final addresses for the partial program, so that
2247 different partial programs do not overlap. This is actually an
2248 oversimplification, but it suffices to explain how @code{@value{AS}} uses
2251 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2252 addresses. These blocks slide to their run-time addresses as rigid
2253 units; their length does not change and neither does the order of bytes
2254 within them. Such a rigid unit is called a @emph{section}. Assigning
2255 run-time addresses to sections is called @dfn{relocation}. It includes
2256 the task of adjusting mentions of object-file addresses so they refer to
2257 the proper run-time addresses.
2259 For the H8/300 and H8/500,
2260 and for the Hitachi SH,
2261 @code{@value{AS}} pads sections if needed to
2262 ensure they end on a word (sixteen bit) boundary.
2265 @cindex standard assembler sections
2266 An object file written by @code{@value{AS}} has at least three sections, any
2267 of which may be empty. These are named @dfn{text}, @dfn{data} and
2272 When it generates COFF output,
2274 @code{@value{AS}} can also generate whatever other named sections you specify
2275 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2276 If you do not use any directives that place output in the @samp{.text}
2277 or @samp{.data} sections, these sections still exist, but are empty.
2282 When @code{@value{AS}} generates SOM or ELF output for the HPPA,
2284 @code{@value{AS}} can also generate whatever other named sections you
2285 specify using the @samp{.space} and @samp{.subspace} directives. See
2286 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2287 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2288 assembler directives.
2291 Additionally, @code{@value{AS}} uses different names for the standard
2292 text, data, and bss sections when generating SOM output. Program text
2293 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2294 BSS into @samp{$BSS$}.
2298 Within the object file, the text section starts at address @code{0}, the
2299 data section follows, and the bss section follows the data section.
2302 When generating either SOM or ELF output files on the HPPA, the text
2303 section starts at address @code{0}, the data section at address
2304 @code{0x4000000}, and the bss section follows the data section.
2307 To let @code{@value{LD}} know which data changes when the sections are
2308 relocated, and how to change that data, @code{@value{AS}} also writes to the
2309 object file details of the relocation needed. To perform relocation
2310 @code{@value{LD}} must know, each time an address in the object
2314 Where in the object file is the beginning of this reference to
2317 How long (in bytes) is this reference?
2319 Which section does the address refer to? What is the numeric value of
2321 (@var{address}) @minus{} (@var{start-address of section})?
2324 Is the reference to an address ``Program-Counter relative''?
2327 @cindex addresses, format of
2328 @cindex section-relative addressing
2329 In fact, every address @code{@value{AS}} ever uses is expressed as
2331 (@var{section}) + (@var{offset into section})
2334 Further, most expressions @code{@value{AS}} computes have this section-relative
2337 (For some object formats, such as SOM for the HPPA, some expressions are
2338 symbol-relative instead.)
2341 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2342 @var{N} into section @var{secname}.''
2344 Apart from text, data and bss sections you need to know about the
2345 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2346 addresses in the absolute section remain unchanged. For example, address
2347 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2348 @code{@value{LD}}. Although the linker never arranges two partial programs'
2349 data sections with overlapping addresses after linking, @emph{by definition}
2350 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2351 part of a program is always the same address when the program is running as
2352 address @code{@{absolute@ 239@}} in any other part of the program.
2354 The idea of sections is extended to the @dfn{undefined} section. Any
2355 address whose section is unknown at assembly time is by definition
2356 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2357 Since numbers are always defined, the only way to generate an undefined
2358 address is to mention an undefined symbol. A reference to a named
2359 common block would be such a symbol: its value is unknown at assembly
2360 time so it has section @emph{undefined}.
2362 By analogy the word @emph{section} is used to describe groups of sections in
2363 the linked program. @code{@value{LD}} puts all partial programs' text
2364 sections in contiguous addresses in the linked program. It is
2365 customary to refer to the @emph{text section} of a program, meaning all
2366 the addresses of all partial programs' text sections. Likewise for
2367 data and bss sections.
2369 Some sections are manipulated by @code{@value{LD}}; others are invented for
2370 use of @code{@value{AS}} and have no meaning except during assembly.
2373 @section Linker Sections
2374 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2379 @cindex named sections
2380 @cindex sections, named
2381 @item named sections
2384 @cindex text section
2385 @cindex data section
2389 These sections hold your program. @code{@value{AS}} and @code{@value{LD}} treat them as
2390 separate but equal sections. Anything you can say of one section is
2393 When the program is running, however, it is
2394 customary for the text section to be unalterable. The
2395 text section is often shared among processes: it contains
2396 instructions, constants and the like. The data section of a running
2397 program is usually alterable: for example, C variables would be stored
2398 in the data section.
2403 This section contains zeroed bytes when your program begins running. It
2404 is used to hold unitialized variables or common storage. The length of
2405 each partial program's bss section is important, but because it starts
2406 out containing zeroed bytes there is no need to store explicit zero
2407 bytes in the object file. The bss section was invented to eliminate
2408 those explicit zeros from object files.
2410 @cindex absolute section
2411 @item absolute section
2412 Address 0 of this section is always ``relocated'' to runtime address 0.
2413 This is useful if you want to refer to an address that @code{@value{LD}} must
2414 not change when relocating. In this sense we speak of absolute
2415 addresses being ``unrelocatable'': they do not change during relocation.
2417 @cindex undefined section
2418 @item undefined section
2419 This ``section'' is a catch-all for address references to objects not in
2420 the preceding sections.
2421 @c FIXME: ref to some other doc on obj-file formats could go here.
2424 @cindex relocation example
2425 An idealized example of three relocatable sections follows.
2427 The example uses the traditional section names @samp{.text} and @samp{.data}.
2429 Memory addresses are on the horizontal axis.
2433 @c END TEXI2ROFF-KILL
2436 partial program # 1: |ttttt|dddd|00|
2443 partial program # 2: |TTT|DDD|000|
2446 +--+---+-----+--+----+---+-----+~~
2447 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2448 +--+---+-----+--+----+---+-----+~~
2450 addresses: 0 @dots{}
2457 \line{\it Partial program \#1: \hfil}
2458 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2459 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2461 \line{\it Partial program \#2: \hfil}
2462 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2463 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2465 \line{\it linked program: \hfil}
2466 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2467 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2468 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2469 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2471 \line{\it addresses: \hfil}
2475 @c END TEXI2ROFF-KILL
2478 @section Assembler Internal Sections
2480 @cindex internal assembler sections
2481 @cindex sections in messages, internal
2482 These sections are meant only for the internal use of @code{@value{AS}}. They
2483 have no meaning at run-time. You do not really need to know about these
2484 sections for most purposes; but they can be mentioned in @code{@value{AS}}
2485 warning messages, so it might be helpful to have an idea of their
2486 meanings to @code{@value{AS}}. These sections are used to permit the
2487 value of every expression in your assembly language program to be a
2488 section-relative address.
2491 @cindex assembler internal logic error
2492 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2493 An internal assembler logic error has been found. This means there is a
2494 bug in the assembler.
2496 @cindex expr (internal section)
2498 The assembler stores complex expression internally as combinations of
2499 symbols. When it needs to represent an expression as a symbol, it puts
2500 it in the expr section.
2502 @c FIXME item transfer[t] vector preload
2503 @c FIXME item transfer[t] vector postload
2504 @c FIXME item register
2508 @section Sub-Sections
2510 @cindex numbered subsections
2511 @cindex grouping data
2517 fall into two sections: text and data.
2519 You may have separate groups of
2521 data in named sections
2525 data in named sections
2531 that you want to end up near to each other in the object file, even though they
2532 are not contiguous in the assembler source. @code{@value{AS}} allows you to
2533 use @dfn{subsections} for this purpose. Within each section, there can be
2534 numbered subsections with values from 0 to 8192. Objects assembled into the
2535 same subsection go into the object file together with other objects in the same
2536 subsection. For example, a compiler might want to store constants in the text
2537 section, but might not want to have them interspersed with the program being
2538 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2539 section of code being output, and a @samp{.text 1} before each group of
2540 constants being output.
2542 Subsections are optional. If you do not use subsections, everything
2543 goes in subsection number zero.
2546 Each subsection is zero-padded up to a multiple of four bytes.
2547 (Subsections may be padded a different amount on different flavors
2548 of @code{@value{AS}}.)
2552 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2553 boundary (two bytes).
2554 The same is true on the Hitachi SH.
2557 @c FIXME section padding (alignment)?
2558 @c Rich Pixley says padding here depends on target obj code format; that
2559 @c doesn't seem particularly useful to say without further elaboration,
2560 @c so for now I say nothing about it. If this is a generic BFD issue,
2561 @c these paragraphs might need to vanish from this manual, and be
2562 @c discussed in BFD chapter of binutils (or some such).
2565 On the AMD 29K family, no particular padding is added to section or
2566 subsection sizes; @value{AS} forces no alignment on this platform.
2570 Subsections appear in your object file in numeric order, lowest numbered
2571 to highest. (All this to be compatible with other people's assemblers.)
2572 The object file contains no representation of subsections; @code{@value{LD}} and
2573 other programs that manipulate object files see no trace of them.
2574 They just see all your text subsections as a text section, and all your
2575 data subsections as a data section.
2577 To specify which subsection you want subsequent statements assembled
2578 into, use a numeric argument to specify it, in a @samp{.text
2579 @var{expression}} or a @samp{.data @var{expression}} statement.
2582 When generating COFF output, you
2587 can also use an extra subsection
2588 argument with arbitrary named sections: @samp{.section @var{name},
2591 @var{Expression} should be an absolute expression.
2592 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
2593 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
2594 begins in @code{text 0}. For instance:
2596 .text 0 # The default subsection is text 0 anyway.
2597 .ascii "This lives in the first text subsection. *"
2599 .ascii "But this lives in the second text subsection."
2601 .ascii "This lives in the data section,"
2602 .ascii "in the first data subsection."
2604 .ascii "This lives in the first text section,"
2605 .ascii "immediately following the asterisk (*)."
2608 Each section has a @dfn{location counter} incremented by one for every byte
2609 assembled into that section. Because subsections are merely a convenience
2610 restricted to @code{@value{AS}} there is no concept of a subsection location
2611 counter. There is no way to directly manipulate a location counter---but the
2612 @code{.align} directive changes it, and any label definition captures its
2613 current value. The location counter of the section where statements are being
2614 assembled is said to be the @dfn{active} location counter.
2617 @section bss Section
2620 @cindex common variable storage
2621 The bss section is used for local common variable storage.
2622 You may allocate address space in the bss section, but you may
2623 not dictate data to load into it before your program executes. When
2624 your program starts running, all the contents of the bss
2625 section are zeroed bytes.
2627 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
2628 @ref{Lcomm,,@code{.lcomm}}.
2630 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
2631 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
2634 When assembling for a target which supports multiple sections, such as ELF or
2635 COFF, you may switch into the @code{.bss} section and define symbols as usual;
2636 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
2637 section. Typically the section will only contain symbol definitions and
2638 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
2645 Symbols are a central concept: the programmer uses symbols to name
2646 things, the linker uses symbols to link, and the debugger uses symbols
2650 @cindex debuggers, and symbol order
2651 @emph{Warning:} @code{@value{AS}} does not place symbols in the object file in
2652 the same order they were declared. This may break some debuggers.
2657 * Setting Symbols:: Giving Symbols Other Values
2658 * Symbol Names:: Symbol Names
2659 * Dot:: The Special Dot Symbol
2660 * Symbol Attributes:: Symbol Attributes
2667 A @dfn{label} is written as a symbol immediately followed by a colon
2668 @samp{:}. The symbol then represents the current value of the
2669 active location counter, and is, for example, a suitable instruction
2670 operand. You are warned if you use the same symbol to represent two
2671 different locations: the first definition overrides any other
2675 On the HPPA, the usual form for a label need not be immediately followed by a
2676 colon, but instead must start in column zero. Only one label may be defined on
2677 a single line. To work around this, the HPPA version of @code{@value{AS}} also
2678 provides a special directive @code{.label} for defining labels more flexibly.
2681 @node Setting Symbols
2682 @section Giving Symbols Other Values
2684 @cindex assigning values to symbols
2685 @cindex symbol values, assigning
2686 A symbol can be given an arbitrary value by writing a symbol, followed
2687 by an equals sign @samp{=}, followed by an expression
2688 (@pxref{Expressions}). This is equivalent to using the @code{.set}
2689 directive. @xref{Set,,@code{.set}}.
2692 @section Symbol Names
2694 @cindex symbol names
2695 @cindex names, symbol
2696 @ifclear SPECIAL-SYMS
2697 Symbol names begin with a letter or with one of @samp{._}. On most
2698 machines, you can also use @code{$} in symbol names; exceptions are
2699 noted in @ref{Machine Dependencies}. That character may be followed by any
2700 string of digits, letters, dollar signs (unless otherwise noted in
2701 @ref{Machine Dependencies}), and underscores.
2704 For the AMD 29K family, @samp{?} is also allowed in the
2705 body of a symbol name, though not at its beginning.
2710 Symbol names begin with a letter or with one of @samp{._}. On the
2712 H8/500, you can also use @code{$} in symbol names. That character may
2713 be followed by any string of digits, letters, dollar signs (save on the
2714 H8/300), and underscores.
2718 Case of letters is significant: @code{foo} is a different symbol name
2721 Each symbol has exactly one name. Each name in an assembly language program
2722 refers to exactly one symbol. You may use that symbol name any number of times
2725 @subheading Local Symbol Names
2727 @cindex local symbol names
2728 @cindex symbol names, local
2729 @cindex temporary symbol names
2730 @cindex symbol names, temporary
2731 Local symbols help compilers and programmers use names temporarily.
2732 There are ten local symbol names, which are re-used throughout the
2733 program. You may refer to them using the names @samp{0} @samp{1}
2734 @dots{} @samp{9}. To define a local symbol, write a label of the form
2735 @samp{@b{N}:} (where @b{N} represents any digit). To refer to the most
2736 recent previous definition of that symbol write @samp{@b{N}b}, using the
2737 same digit as when you defined the label. To refer to the next
2738 definition of a local label, write @samp{@b{N}f}---where @b{N} gives you
2739 a choice of 10 forward references. The @samp{b} stands for
2740 ``backwards'' and the @samp{f} stands for ``forwards''.
2742 Local symbols are not emitted by the current @sc{gnu} C compiler.
2744 There is no restriction on how you can use these labels, but
2745 remember that at any point in the assembly you can refer to at most
2746 10 prior local labels and to at most 10 forward local labels.
2748 Local symbol names are only a notation device. They are immediately
2749 transformed into more conventional symbol names before the assembler
2750 uses them. The symbol names stored in the symbol table, appearing in
2751 error messages and optionally emitted to the object file have these
2756 All local labels begin with @samp{L}. Normally both @code{@value{AS}} and
2757 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
2758 used for symbols you are never intended to see. If you use the
2759 @samp{-L} option then @code{@value{AS}} retains these symbols in the
2760 object file. If you also instruct @code{@value{LD}} to retain these symbols,
2761 you may use them in debugging.
2764 If the label is written @samp{0:} then the digit is @samp{0}.
2765 If the label is written @samp{1:} then the digit is @samp{1}.
2766 And so on up through @samp{9:}.
2769 This unusual character is included so you do not accidentally invent
2770 a symbol of the same name. The character has ASCII value
2773 @item @emph{ordinal number}
2774 This is a serial number to keep the labels distinct. The first
2775 @samp{0:} gets the number @samp{1}; The 15th @samp{0:} gets the
2776 number @samp{15}; @emph{etc.}. Likewise for the other labels @samp{1:}
2780 For instance, the first @code{1:} is named @code{L1@kbd{C-A}1}, the 44th
2781 @code{3:} is named @code{L3@kbd{C-A}44}.
2784 @section The Special Dot Symbol
2786 @cindex dot (symbol)
2787 @cindex @code{.} (symbol)
2788 @cindex current address
2789 @cindex location counter
2790 The special symbol @samp{.} refers to the current address that
2791 @code{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
2792 .long .} defines @code{melvin} to contain its own address.
2793 Assigning a value to @code{.} is treated the same as a @code{.org}
2794 directive. Thus, the expression @samp{.=.+4} is the same as saying
2795 @ifclear no-space-dir
2804 @node Symbol Attributes
2805 @section Symbol Attributes
2807 @cindex symbol attributes
2808 @cindex attributes, symbol
2809 Every symbol has, as well as its name, the attributes ``Value'' and
2810 ``Type''. Depending on output format, symbols can also have auxiliary
2813 The detailed definitions are in @file{a.out.h}.
2816 If you use a symbol without defining it, @code{@value{AS}} assumes zero for
2817 all these attributes, and probably won't warn you. This makes the
2818 symbol an externally defined symbol, which is generally what you
2822 * Symbol Value:: Value
2823 * Symbol Type:: Type
2826 * a.out Symbols:: Symbol Attributes: @code{a.out}
2830 * a.out Symbols:: Symbol Attributes: @code{a.out}
2833 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
2838 * COFF Symbols:: Symbol Attributes for COFF
2841 * SOM Symbols:: Symbol Attributes for SOM
2848 @cindex value of a symbol
2849 @cindex symbol value
2850 The value of a symbol is (usually) 32 bits. For a symbol which labels a
2851 location in the text, data, bss or absolute sections the value is the
2852 number of addresses from the start of that section to the label.
2853 Naturally for text, data and bss sections the value of a symbol changes
2854 as @code{@value{LD}} changes section base addresses during linking. Absolute
2855 symbols' values do not change during linking: that is why they are
2858 The value of an undefined symbol is treated in a special way. If it is
2859 0 then the symbol is not defined in this assembler source file, and
2860 @code{@value{LD}} tries to determine its value from other files linked into the
2861 same program. You make this kind of symbol simply by mentioning a symbol
2862 name without defining it. A non-zero value represents a @code{.comm}
2863 common declaration. The value is how much common storage to reserve, in
2864 bytes (addresses). The symbol refers to the first address of the
2870 @cindex type of a symbol
2872 The type attribute of a symbol contains relocation (section)
2873 information, any flag settings indicating that a symbol is external, and
2874 (optionally), other information for linkers and debuggers. The exact
2875 format depends on the object-code output format in use.
2880 @c The following avoids a "widow" subsection title. @group would be
2881 @c better if it were available outside examples.
2884 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
2886 @cindex @code{b.out} symbol attributes
2887 @cindex symbol attributes, @code{b.out}
2888 These symbol attributes appear only when @code{@value{AS}} is configured for
2889 one of the Berkeley-descended object output formats---@code{a.out} or
2895 @subsection Symbol Attributes: @code{a.out}
2897 @cindex @code{a.out} symbol attributes
2898 @cindex symbol attributes, @code{a.out}
2904 @subsection Symbol Attributes: @code{a.out}
2906 @cindex @code{a.out} symbol attributes
2907 @cindex symbol attributes, @code{a.out}
2911 * Symbol Desc:: Descriptor
2912 * Symbol Other:: Other
2916 @subsubsection Descriptor
2918 @cindex descriptor, of @code{a.out} symbol
2919 This is an arbitrary 16-bit value. You may establish a symbol's
2920 descriptor value by using a @code{.desc} statement
2921 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
2925 @subsubsection Other
2927 @cindex other attribute, of @code{a.out} symbol
2928 This is an arbitrary 8-bit value. It means nothing to @code{@value{AS}}.
2933 @subsection Symbol Attributes for COFF
2935 @cindex COFF symbol attributes
2936 @cindex symbol attributes, COFF
2938 The COFF format supports a multitude of auxiliary symbol attributes;
2939 like the primary symbol attributes, they are set between @code{.def} and
2940 @code{.endef} directives.
2942 @subsubsection Primary Attributes
2944 @cindex primary attributes, COFF symbols
2945 The symbol name is set with @code{.def}; the value and type,
2946 respectively, with @code{.val} and @code{.type}.
2948 @subsubsection Auxiliary Attributes
2950 @cindex auxiliary attributes, COFF symbols
2951 The @code{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
2952 @code{.size}, and @code{.tag} can generate auxiliary symbol table
2953 information for COFF.
2958 @subsection Symbol Attributes for SOM
2960 @cindex SOM symbol attributes
2961 @cindex symbol attributes, SOM
2963 The SOM format for the HPPA supports a multitude of symbol attributes set with
2964 the @code{.EXPORT} and @code{.IMPORT} directives.
2966 The attributes are described in @cite{HP9000 Series 800 Assembly
2967 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
2968 @code{EXPORT} assembler directive documentation.
2972 @chapter Expressions
2976 @cindex numeric values
2977 An @dfn{expression} specifies an address or numeric value.
2978 Whitespace may precede and/or follow an expression.
2980 The result of an expression must be an absolute number, or else an offset into
2981 a particular section. If an expression is not absolute, and there is not
2982 enough information when @code{@value{AS}} sees the expression to know its
2983 section, a second pass over the source program might be necessary to interpret
2984 the expression---but the second pass is currently not implemented.
2985 @code{@value{AS}} aborts with an error message in this situation.
2988 * Empty Exprs:: Empty Expressions
2989 * Integer Exprs:: Integer Expressions
2993 @section Empty Expressions
2995 @cindex empty expressions
2996 @cindex expressions, empty
2997 An empty expression has no value: it is just whitespace or null.
2998 Wherever an absolute expression is required, you may omit the
2999 expression, and @code{@value{AS}} assumes a value of (absolute) 0. This
3000 is compatible with other assemblers.
3003 @section Integer Expressions
3005 @cindex integer expressions
3006 @cindex expressions, integer
3007 An @dfn{integer expression} is one or more @emph{arguments} delimited
3008 by @emph{operators}.
3011 * Arguments:: Arguments
3012 * Operators:: Operators
3013 * Prefix Ops:: Prefix Operators
3014 * Infix Ops:: Infix Operators
3018 @subsection Arguments
3020 @cindex expression arguments
3021 @cindex arguments in expressions
3022 @cindex operands in expressions
3023 @cindex arithmetic operands
3024 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3025 contexts arguments are sometimes called ``arithmetic operands''. In
3026 this manual, to avoid confusing them with the ``instruction operands'' of
3027 the machine language, we use the term ``argument'' to refer to parts of
3028 expressions only, reserving the word ``operand'' to refer only to machine
3029 instruction operands.
3031 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3032 @var{section} is one of text, data, bss, absolute,
3033 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3036 Numbers are usually integers.
3038 A number can be a flonum or bignum. In this case, you are warned
3039 that only the low order 32 bits are used, and @code{@value{AS}} pretends
3040 these 32 bits are an integer. You may write integer-manipulating
3041 instructions that act on exotic constants, compatible with other
3044 @cindex subexpressions
3045 Subexpressions are a left parenthesis @samp{(} followed by an integer
3046 expression, followed by a right parenthesis @samp{)}; or a prefix
3047 operator followed by an argument.
3050 @subsection Operators
3052 @cindex operators, in expressions
3053 @cindex arithmetic functions
3054 @cindex functions, in expressions
3055 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3056 operators are followed by an argument. Infix operators appear
3057 between their arguments. Operators may be preceded and/or followed by
3061 @subsection Prefix Operator
3063 @cindex prefix operators
3064 @code{@value{AS}} has the following @dfn{prefix operators}. They each take
3065 one argument, which must be absolute.
3067 @c the tex/end tex stuff surrounding this small table is meant to make
3068 @c it align, on the printed page, with the similar table in the next
3069 @c section (which is inside an enumerate).
3071 \global\advance\leftskip by \itemindent
3076 @dfn{Negation}. Two's complement negation.
3078 @dfn{Complementation}. Bitwise not.
3082 \global\advance\leftskip by -\itemindent
3086 @subsection Infix Operators
3088 @cindex infix operators
3089 @cindex operators, permitted arguments
3090 @dfn{Infix operators} take two arguments, one on either side. Operators
3091 have precedence, but operations with equal precedence are performed left
3092 to right. Apart from @code{+} or @code{-}, both arguments must be
3093 absolute, and the result is absolute.
3096 @cindex operator precedence
3097 @cindex precedence of operators
3104 @dfn{Multiplication}.
3107 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3114 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3118 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3122 Intermediate precedence
3127 @dfn{Bitwise Inclusive Or}.
3133 @dfn{Bitwise Exclusive Or}.
3136 @dfn{Bitwise Or Not}.
3143 @cindex addition, permitted arguments
3144 @cindex plus, permitted arguments
3145 @cindex arguments for addition
3147 @dfn{Addition}. If either argument is absolute, the result has the section of
3148 the other argument. You may not add together arguments from different
3151 @cindex subtraction, permitted arguments
3152 @cindex minus, permitted arguments
3153 @cindex arguments for subtraction
3155 @dfn{Subtraction}. If the right argument is absolute, the
3156 result has the section of the left argument.
3157 If both arguments are in the same section, the result is absolute.
3158 You may not subtract arguments from different sections.
3159 @c FIXME is there still something useful to say about undefined - undefined ?
3163 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3164 address; you can only have a defined section in one of the two arguments.
3167 @chapter Assembler Directives
3169 @cindex directives, machine independent
3170 @cindex pseudo-ops, machine independent
3171 @cindex machine independent directives
3172 All assembler directives have names that begin with a period (@samp{.}).
3173 The rest of the name is letters, usually in lower case.
3175 This chapter discusses directives that are available regardless of the
3176 target machine configuration for the @sc{gnu} assembler.
3178 Some machine configurations provide additional directives.
3179 @xref{Machine Dependencies}.
3182 @ifset machine-directives
3183 @xref{Machine Dependencies} for additional directives.
3188 * Abort:: @code{.abort}
3190 * ABORT:: @code{.ABORT}
3193 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3194 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3195 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3196 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3197 * Byte:: @code{.byte @var{expressions}}
3198 * Comm:: @code{.comm @var{symbol} , @var{length} }
3199 * Data:: @code{.data @var{subsection}}
3201 * Def:: @code{.def @var{name}}
3204 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3210 * Double:: @code{.double @var{flonums}}
3211 * Eject:: @code{.eject}
3212 * Else:: @code{.else}
3213 * Elseif:: @code{.elseif}
3216 * Endef:: @code{.endef}
3219 * Endfunc:: @code{.endfunc}
3220 * Endif:: @code{.endif}
3221 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3222 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3224 * Exitm:: @code{.exitm}
3225 * Extern:: @code{.extern}
3226 * Fail:: @code{.fail}
3227 @ifclear no-file-dir
3228 * File:: @code{.file @var{string}}
3231 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3232 * Float:: @code{.float @var{flonums}}
3233 * Func:: @code{.func}
3234 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3236 * Hidden:: @code{.hidden @var{names}}
3239 * hword:: @code{.hword @var{expressions}}
3240 * Ident:: @code{.ident}
3241 * If:: @code{.if @var{absolute expression}}
3242 * Include:: @code{.include "@var{file}"}
3243 * Int:: @code{.int @var{expressions}}
3245 * Internal:: @code{.internal @var{names}}
3248 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3249 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3250 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3251 * Lflags:: @code{.lflags}
3252 @ifclear no-line-dir
3253 * Line:: @code{.line @var{line-number}}
3256 * Ln:: @code{.ln @var{line-number}}
3257 * Linkonce:: @code{.linkonce [@var{type}]}
3258 * List:: @code{.list}
3259 * Long:: @code{.long @var{expressions}}
3261 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3264 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3265 * MRI:: @code{.mri @var{val}}
3266 * Nolist:: @code{.nolist}
3267 * Octa:: @code{.octa @var{bignums}}
3268 * Org:: @code{.org @var{new-lc} , @var{fill}}
3269 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3271 * PopSection:: @code{.popsection}
3272 * Previous:: @code{.previous}
3275 * Print:: @code{.print @var{string}}
3277 * Protected:: @code{.protected @var{names}}
3280 * Psize:: @code{.psize @var{lines}, @var{columns}}
3281 * Purgem:: @code{.purgem @var{name}}
3283 * PushSection:: @code{.pushsection @var{name}}
3286 * Quad:: @code{.quad @var{bignums}}
3287 * Rept:: @code{.rept @var{count}}
3288 * Sbttl:: @code{.sbttl "@var{subheading}"}
3290 * Scl:: @code{.scl @var{class}}
3291 * Section:: @code{.section @var{name}, @var{subsection}}
3294 * Set:: @code{.set @var{symbol}, @var{expression}}
3295 * Short:: @code{.short @var{expressions}}
3296 * Single:: @code{.single @var{flonums}}
3297 * Size:: @code{.size [@var{name} , @var{expression}]}
3298 * Skip:: @code{.skip @var{size} , @var{fill}}
3299 * Sleb128:: @code{.sleb128 @var{expressions}}
3300 * Space:: @code{.space @var{size} , @var{fill}}
3302 * Stab:: @code{.stabd, .stabn, .stabs}
3305 * String:: @code{.string "@var{str}"}
3306 * Struct:: @code{.struct @var{expression}}
3308 * SubSection:: @code{.subsection}
3309 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3313 * Tag:: @code{.tag @var{structname}}
3316 * Text:: @code{.text @var{subsection}}
3317 * Title:: @code{.title "@var{heading}"}
3318 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3319 * Uleb128:: @code{.uleb128 @var{expressions}}
3321 * Val:: @code{.val @var{addr}}
3325 * Version:: @code{.version "@var{string}"}
3326 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3327 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3328 * Weak:: @code{.weak @var{names}}
3331 * Word:: @code{.word @var{expressions}}
3332 * Deprecated:: Deprecated Directives
3336 @section @code{.abort}
3338 @cindex @code{abort} directive
3339 @cindex stopping the assembly
3340 This directive stops the assembly immediately. It is for
3341 compatibility with other assemblers. The original idea was that the
3342 assembly language source would be piped into the assembler. If the sender
3343 of the source quit, it could use this directive tells @code{@value{AS}} to
3344 quit also. One day @code{.abort} will not be supported.
3348 @section @code{.ABORT}
3350 @cindex @code{ABORT} directive
3351 When producing COFF output, @code{@value{AS}} accepts this directive as a
3352 synonym for @samp{.abort}.
3355 When producing @code{b.out} output, @code{@value{AS}} accepts this directive,
3361 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3363 @cindex padding the location counter
3364 @cindex @code{align} directive
3365 Pad the location counter (in the current subsection) to a particular storage
3366 boundary. The first expression (which must be absolute) is the alignment
3367 required, as described below.
3369 The second expression (also absolute) gives the fill value to be stored in the
3370 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3371 padding bytes are normally zero. However, on some systems, if the section is
3372 marked as containing code and the fill value is omitted, the space is filled
3373 with no-op instructions.
3375 The third expression is also absolute, and is also optional. If it is present,
3376 it is the maximum number of bytes that should be skipped by this alignment
3377 directive. If doing the alignment would require skipping more bytes than the
3378 specified maximum, then the alignment is not done at all. You can omit the
3379 fill value (the second argument) entirely by simply using two commas after the
3380 required alignment; this can be useful if you want the alignment to be filled
3381 with no-op instructions when appropriate.
3383 The way the required alignment is specified varies from system to system.
3384 For the a29k, hppa, m68k, m88k, w65, sparc, and Hitachi SH, and i386 using ELF
3386 the first expression is the
3387 alignment request in bytes. For example @samp{.align 8} advances
3388 the location counter until it is a multiple of 8. If the location counter
3389 is already a multiple of 8, no change is needed.
3391 For other systems, including the i386 using a.out format, and the arm and
3392 strongarm, it is the
3393 number of low-order zero bits the location counter must have after
3394 advancement. For example @samp{.align 3} advances the location
3395 counter until it a multiple of 8. If the location counter is already a
3396 multiple of 8, no change is needed.
3398 This inconsistency is due to the different behaviors of the various
3399 native assemblers for these systems which GAS must emulate.
3400 GAS also provides @code{.balign} and @code{.p2align} directives,
3401 described later, which have a consistent behavior across all
3402 architectures (but are specific to GAS).
3405 @section @code{.ascii "@var{string}"}@dots{}
3407 @cindex @code{ascii} directive
3408 @cindex string literals
3409 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3410 separated by commas. It assembles each string (with no automatic
3411 trailing zero byte) into consecutive addresses.
3414 @section @code{.asciz "@var{string}"}@dots{}
3416 @cindex @code{asciz} directive
3417 @cindex zero-terminated strings
3418 @cindex null-terminated strings
3419 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3420 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3423 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3425 @cindex padding the location counter given number of bytes
3426 @cindex @code{balign} directive
3427 Pad the location counter (in the current subsection) to a particular
3428 storage boundary. The first expression (which must be absolute) is the
3429 alignment request in bytes. For example @samp{.balign 8} advances
3430 the location counter until it is a multiple of 8. If the location counter
3431 is already a multiple of 8, no change is needed.
3433 The second expression (also absolute) gives the fill value to be stored in the
3434 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3435 padding bytes are normally zero. However, on some systems, if the section is
3436 marked as containing code and the fill value is omitted, the space is filled
3437 with no-op instructions.
3439 The third expression is also absolute, and is also optional. If it is present,
3440 it is the maximum number of bytes that should be skipped by this alignment
3441 directive. If doing the alignment would require skipping more bytes than the
3442 specified maximum, then the alignment is not done at all. You can omit the
3443 fill value (the second argument) entirely by simply using two commas after the
3444 required alignment; this can be useful if you want the alignment to be filled
3445 with no-op instructions when appropriate.
3447 @cindex @code{balignw} directive
3448 @cindex @code{balignl} directive
3449 The @code{.balignw} and @code{.balignl} directives are variants of the
3450 @code{.balign} directive. The @code{.balignw} directive treats the fill
3451 pattern as a two byte word value. The @code{.balignl} directives treats the
3452 fill pattern as a four byte longword value. For example, @code{.balignw
3453 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3454 filled in with the value 0x368d (the exact placement of the bytes depends upon
3455 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3459 @section @code{.byte @var{expressions}}
3461 @cindex @code{byte} directive
3462 @cindex integers, one byte
3463 @code{.byte} expects zero or more expressions, separated by commas.
3464 Each expression is assembled into the next byte.
3467 @section @code{.comm @var{symbol} , @var{length} }
3469 @cindex @code{comm} directive
3470 @cindex symbol, common
3471 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
3472 common symbol in one object file may be merged with a defined or common symbol
3473 of the same name in another object file. If @code{@value{LD}} does not see a
3474 definition for the symbol--just one or more common symbols--then it will
3475 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
3476 absolute expression. If @code{@value{LD}} sees multiple common symbols with
3477 the same name, and they do not all have the same size, it will allocate space
3478 using the largest size.
3481 When using ELF, the @code{.comm} directive takes an optional third argument.
3482 This is the desired alignment of the symbol, specified as a byte boundary (for
3483 example, an alignment of 16 means that the least significant 4 bits of the
3484 address should be zero). The alignment must be an absolute expression, and it
3485 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
3486 for the common symbol, it will use the alignment when placing the symbol. If
3487 no alignment is specified, @code{@value{AS}} will set the alignment to the
3488 largest power of two less than or equal to the size of the symbol, up to a
3493 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
3494 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
3498 @section @code{.data @var{subsection}}
3500 @cindex @code{data} directive
3501 @code{.data} tells @code{@value{AS}} to assemble the following statements onto the
3502 end of the data subsection numbered @var{subsection} (which is an
3503 absolute expression). If @var{subsection} is omitted, it defaults
3508 @section @code{.def @var{name}}
3510 @cindex @code{def} directive
3511 @cindex COFF symbols, debugging
3512 @cindex debugging COFF symbols
3513 Begin defining debugging information for a symbol @var{name}; the
3514 definition extends until the @code{.endef} directive is encountered.
3517 This directive is only observed when @code{@value{AS}} is configured for COFF
3518 format output; when producing @code{b.out}, @samp{.def} is recognized,
3525 @section @code{.desc @var{symbol}, @var{abs-expression}}
3527 @cindex @code{desc} directive
3528 @cindex COFF symbol descriptor
3529 @cindex symbol descriptor, COFF
3530 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
3531 to the low 16 bits of an absolute expression.
3534 The @samp{.desc} directive is not available when @code{@value{AS}} is
3535 configured for COFF output; it is only for @code{a.out} or @code{b.out}
3536 object format. For the sake of compatibility, @code{@value{AS}} accepts
3537 it, but produces no output, when configured for COFF.
3543 @section @code{.dim}
3545 @cindex @code{dim} directive
3546 @cindex COFF auxiliary symbol information
3547 @cindex auxiliary symbol information, COFF
3548 This directive is generated by compilers to include auxiliary debugging
3549 information in the symbol table. It is only permitted inside
3550 @code{.def}/@code{.endef} pairs.
3553 @samp{.dim} is only meaningful when generating COFF format output; when
3554 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
3560 @section @code{.double @var{flonums}}
3562 @cindex @code{double} directive
3563 @cindex floating point numbers (double)
3564 @code{.double} expects zero or more flonums, separated by commas. It
3565 assembles floating point numbers.
3567 The exact kind of floating point numbers emitted depends on how
3568 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
3572 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
3573 in @sc{ieee} format.
3578 @section @code{.eject}
3580 @cindex @code{eject} directive
3581 @cindex new page, in listings
3582 @cindex page, in listings
3583 @cindex listing control: new page
3584 Force a page break at this point, when generating assembly listings.
3587 @section @code{.else}
3589 @cindex @code{else} directive
3590 @code{.else} is part of the @code{@value{AS}} support for conditional
3591 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
3592 of code to be assembled if the condition for the preceding @code{.if}
3596 @section @code{.elseif}
3598 @cindex @code{elseif} directive
3599 @code{.elseif} is part of the @code{@value{AS}} support for conditional
3600 assembly; @pxref{If,,@code{.if}}. It is shorthand for beginning a new
3601 @code{.if} block that would otherwise fill the entire @code{.else} section.
3604 @section @code{.end}
3606 @cindex @code{end} directive
3607 @code{.end} marks the end of the assembly file. @code{@value{AS}} does not
3608 process anything in the file past the @code{.end} directive.
3612 @section @code{.endef}
3614 @cindex @code{endef} directive
3615 This directive flags the end of a symbol definition begun with
3619 @samp{.endef} is only meaningful when generating COFF format output; if
3620 @code{@value{AS}} is configured to generate @code{b.out}, it accepts this
3621 directive but ignores it.
3626 @section @code{.endfunc}
3627 @cindex @code{endfunc} directive
3628 @code{.endfunc} marks the end of a function specified with @code{.func}.
3631 @section @code{.endif}
3633 @cindex @code{endif} directive
3634 @code{.endif} is part of the @code{@value{AS}} support for conditional assembly;
3635 it marks the end of a block of code that is only assembled
3636 conditionally. @xref{If,,@code{.if}}.
3639 @section @code{.equ @var{symbol}, @var{expression}}
3641 @cindex @code{equ} directive
3642 @cindex assigning values to symbols
3643 @cindex symbols, assigning values to
3644 This directive sets the value of @var{symbol} to @var{expression}.
3645 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
3648 The syntax for @code{equ} on the HPPA is
3649 @samp{@var{symbol} .equ @var{expression}}.
3653 @section @code{.equiv @var{symbol}, @var{expression}}
3654 @cindex @code{equiv} directive
3655 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
3656 the assembler will signal an error if @var{symbol} is already defined.
3658 Except for the contents of the error message, this is roughly equivalent to
3667 @section @code{.err}
3668 @cindex @code{err} directive
3669 If @code{@value{AS}} assembles a @code{.err} directive, it will print an error
3670 message and, unless the @code{-Z} option was used, it will not generate an
3671 object file. This can be used to signal error an conditionally compiled code.
3674 @section @code{.exitm}
3675 Exit early from the current macro definition. @xref{Macro}.
3678 @section @code{.extern}
3680 @cindex @code{extern} directive
3681 @code{.extern} is accepted in the source program---for compatibility
3682 with other assemblers---but it is ignored. @code{@value{AS}} treats
3683 all undefined symbols as external.
3686 @section @code{.fail @var{expression}}
3688 @cindex @code{fail} directive
3689 Generates an error or a warning. If the value of the @var{expression} is 500
3690 or more, @code{@value{AS}} will print a warning message. If the value is less
3691 than 500, @code{@value{AS}} will print an error message. The message will
3692 include the value of @var{expression}. This can occasionally be useful inside
3693 complex nested macros or conditional assembly.
3695 @ifclear no-file-dir
3697 @section @code{.file @var{string}}
3699 @cindex @code{file} directive
3700 @cindex logical file name
3701 @cindex file name, logical
3702 @code{.file} tells @code{@value{AS}} that we are about to start a new logical
3703 file. @var{string} is the new file name. In general, the filename is
3704 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
3705 to specify an empty file name, you must give the quotes--@code{""}. This
3706 statement may go away in future: it is only recognized to be compatible with
3707 old @code{@value{AS}} programs.
3709 In some configurations of @code{@value{AS}}, @code{.file} has already been
3710 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
3715 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
3717 @cindex @code{fill} directive
3718 @cindex writing patterns in memory
3719 @cindex patterns, writing in memory
3720 @var{result}, @var{size} and @var{value} are absolute expressions.
3721 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
3722 may be zero or more. @var{Size} may be zero or more, but if it is
3723 more than 8, then it is deemed to have the value 8, compatible with
3724 other people's assemblers. The contents of each @var{repeat} bytes
3725 is taken from an 8-byte number. The highest order 4 bytes are
3726 zero. The lowest order 4 bytes are @var{value} rendered in the
3727 byte-order of an integer on the computer @code{@value{AS}} is assembling for.
3728 Each @var{size} bytes in a repetition is taken from the lowest order
3729 @var{size} bytes of this number. Again, this bizarre behavior is
3730 compatible with other people's assemblers.
3732 @var{size} and @var{value} are optional.
3733 If the second comma and @var{value} are absent, @var{value} is
3734 assumed zero. If the first comma and following tokens are absent,
3735 @var{size} is assumed to be 1.
3738 @section @code{.float @var{flonums}}
3740 @cindex floating point numbers (single)
3741 @cindex @code{float} directive
3742 This directive assembles zero or more flonums, separated by commas. It
3743 has the same effect as @code{.single}.
3745 The exact kind of floating point numbers emitted depends on how
3746 @code{@value{AS}} is configured.
3747 @xref{Machine Dependencies}.
3751 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
3752 in @sc{ieee} format.
3757 @section @code{.func @var{name}[,@var{label}]}
3758 @cindex @code{func} directive
3759 @code{.func} emits debugging information to denote function @var{name}, and
3760 is ignored unless the file is assembled with debugging enabled.
3761 Only @samp{--gstabs} is currently supported.
3762 @var{label} is the entry point of the function and if omitted @var{name}
3763 prepended with the @samp{leading char} is used.
3764 @samp{leading char} is usually @code{_} or nothing, depending on the target.
3765 All functions are currently defined to have @code{void} return type.
3766 The function must be terminated with @code{.endfunc}.
3769 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3771 @cindex @code{global} directive
3772 @cindex symbol, making visible to linker
3773 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
3774 @var{symbol} in your partial program, its value is made available to
3775 other partial programs that are linked with it. Otherwise,
3776 @var{symbol} takes its attributes from a symbol of the same name
3777 from another file linked into the same program.
3779 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
3780 compatibility with other assemblers.
3783 On the HPPA, @code{.global} is not always enough to make it accessible to other
3784 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
3785 @xref{HPPA Directives,, HPPA Assembler Directives}.
3790 @section @code{.hidden @var{names}}
3792 @cindex @code{.hidden} directive
3794 This one of the ELF visibility directives. The other two are
3795 @pxref{Internal,,@code{.internal}} and @pxref{Protected,,@code{.protected}}
3797 This directive overrides the named symbols default visibility (which is set by
3798 their binding: local, global or weak). The directive sets the visibility to
3799 @code{hidden} which means that the symbols are not visible to other components.
3800 Such symbols are always considered to be @code{protected} as well.
3804 @section @code{.hword @var{expressions}}
3806 @cindex @code{hword} directive
3807 @cindex integers, 16-bit
3808 @cindex numbers, 16-bit
3809 @cindex sixteen bit integers
3810 This expects zero or more @var{expressions}, and emits
3811 a 16 bit number for each.
3814 This directive is a synonym for @samp{.short}; depending on the target
3815 architecture, it may also be a synonym for @samp{.word}.
3819 This directive is a synonym for @samp{.short}.
3822 This directive is a synonym for both @samp{.short} and @samp{.word}.
3827 @section @code{.ident}
3829 @cindex @code{ident} directive
3830 This directive is used by some assemblers to place tags in object files.
3831 @code{@value{AS}} simply accepts the directive for source-file
3832 compatibility with such assemblers, but does not actually emit anything
3836 @section @code{.if @var{absolute expression}}
3838 @cindex conditional assembly
3839 @cindex @code{if} directive
3840 @code{.if} marks the beginning of a section of code which is only
3841 considered part of the source program being assembled if the argument
3842 (which must be an @var{absolute expression}) is non-zero. The end of
3843 the conditional section of code must be marked by @code{.endif}
3844 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
3845 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
3846 If you have several conditions to check, @code{.elseif} may be used to avoid
3847 nesting blocks if/else within each subsequent @code{.else} block.
3849 The following variants of @code{.if} are also supported:
3851 @cindex @code{ifdef} directive
3852 @item .ifdef @var{symbol}
3853 Assembles the following section of code if the specified @var{symbol}
3856 @cindex @code{ifc} directive
3857 @item .ifc @var{string1},@var{string2}
3858 Assembles the following section of code if the two strings are the same. The
3859 strings may be optionally quoted with single quotes. If they are not quoted,
3860 the first string stops at the first comma, and the second string stops at the
3861 end of the line. Strings which contain whitespace should be quoted. The
3862 string comparison is case sensitive.
3864 @cindex @code{ifeq} directive
3865 @item .ifeq @var{absolute expression}
3866 Assembles the following section of code if the argument is zero.
3868 @cindex @code{ifeqs} directive
3869 @item .ifeqs @var{string1},@var{string2}
3870 Another form of @code{.ifc}. The strings must be quoted using double quotes.
3872 @cindex @code{ifge} directive
3873 @item .ifge @var{absolute expression}
3874 Assembles the following section of code if the argument is greater than or
3877 @cindex @code{ifgt} directive
3878 @item .ifgt @var{absolute expression}
3879 Assembles the following section of code if the argument is greater than zero.
3881 @cindex @code{ifle} directive
3882 @item .ifle @var{absolute expression}
3883 Assembles the following section of code if the argument is less than or equal
3886 @cindex @code{iflt} directive
3887 @item .iflt @var{absolute expression}
3888 Assembles the following section of code if the argument is less than zero.
3890 @cindex @code{ifnc} directive
3891 @item .ifnc @var{string1},@var{string2}.
3892 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
3893 following section of code if the two strings are not the same.
3895 @cindex @code{ifndef} directive
3896 @cindex @code{ifnotdef} directive
3897 @item .ifndef @var{symbol}
3898 @itemx .ifnotdef @var{symbol}
3899 Assembles the following section of code if the specified @var{symbol}
3900 has not been defined. Both spelling variants are equivalent.
3902 @cindex @code{ifne} directive
3903 @item .ifne @var{absolute expression}
3904 Assembles the following section of code if the argument is not equal to zero
3905 (in other words, this is equivalent to @code{.if}).
3907 @cindex @code{ifnes} directive
3908 @item .ifnes @var{string1},@var{string2}
3909 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
3910 following section of code if the two strings are not the same.
3914 @section @code{.include "@var{file}"}
3916 @cindex @code{include} directive
3917 @cindex supporting files, including
3918 @cindex files, including
3919 This directive provides a way to include supporting files at specified
3920 points in your source program. The code from @var{file} is assembled as
3921 if it followed the point of the @code{.include}; when the end of the
3922 included file is reached, assembly of the original file continues. You
3923 can control the search paths used with the @samp{-I} command-line option
3924 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
3928 @section @code{.int @var{expressions}}
3930 @cindex @code{int} directive
3931 @cindex integers, 32-bit
3932 Expect zero or more @var{expressions}, of any section, separated by commas.
3933 For each expression, emit a number that, at run time, is the value of that
3934 expression. The byte order and bit size of the number depends on what kind
3935 of target the assembly is for.
3939 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
3940 integers. On the H8/300H and the Hitachi SH, however, @code{.int} emits
3947 @section @code{.internal @var{names}}
3949 @cindex @code{.internal} directive
3951 This one of the ELF visibility directives. The other two are
3952 @pxref{Hidden,,@code{.hidden}} and @pxref{Protected,,@code{.protected}}
3954 This directive overrides the named symbols default visibility (which is set by
3955 their binding: local, global or weak). The directive sets the visibility to
3956 @code{internal} which means that the symbols are considered to be @code{hidden}
3957 (ie not visible to other components), and that some extra, processor specific
3958 processing must also be performed upon the symbols as well.
3962 @section @code{.irp @var{symbol},@var{values}}@dots{}
3964 @cindex @code{irp} directive
3965 Evaluate a sequence of statements assigning different values to @var{symbol}.
3966 The sequence of statements starts at the @code{.irp} directive, and is
3967 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
3968 set to @var{value}, and the sequence of statements is assembled. If no
3969 @var{value} is listed, the sequence of statements is assembled once, with
3970 @var{symbol} set to the null string. To refer to @var{symbol} within the
3971 sequence of statements, use @var{\symbol}.
3973 For example, assembling
3981 is equivalent to assembling
3990 @section @code{.irpc @var{symbol},@var{values}}@dots{}
3992 @cindex @code{irpc} directive
3993 Evaluate a sequence of statements assigning different values to @var{symbol}.
3994 The sequence of statements starts at the @code{.irpc} directive, and is
3995 terminated by an @code{.endr} directive. For each character in @var{value},
3996 @var{symbol} is set to the character, and the sequence of statements is
3997 assembled. If no @var{value} is listed, the sequence of statements is
3998 assembled once, with @var{symbol} set to the null string. To refer to
3999 @var{symbol} within the sequence of statements, use @var{\symbol}.
4001 For example, assembling
4009 is equivalent to assembling
4018 @section @code{.lcomm @var{symbol} , @var{length}}
4020 @cindex @code{lcomm} directive
4021 @cindex local common symbols
4022 @cindex symbols, local common
4023 Reserve @var{length} (an absolute expression) bytes for a local common
4024 denoted by @var{symbol}. The section and value of @var{symbol} are
4025 those of the new local common. The addresses are allocated in the bss
4026 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4027 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4028 not visible to @code{@value{LD}}.
4031 Some targets permit a third argument to be used with @code{.lcomm}. This
4032 argument specifies the desired alignment of the symbol in the bss section.
4036 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4037 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4041 @section @code{.lflags}
4043 @cindex @code{lflags} directive (ignored)
4044 @code{@value{AS}} accepts this directive, for compatibility with other
4045 assemblers, but ignores it.
4047 @ifclear no-line-dir
4049 @section @code{.line @var{line-number}}
4051 @cindex @code{line} directive
4055 @section @code{.ln @var{line-number}}
4057 @cindex @code{ln} directive
4059 @cindex logical line number
4061 Change the logical line number. @var{line-number} must be an absolute
4062 expression. The next line has that logical line number. Therefore any other
4063 statements on the current line (after a statement separator character) are
4064 reported as on logical line number @var{line-number} @minus{} 1. One day
4065 @code{@value{AS}} will no longer support this directive: it is recognized only
4066 for compatibility with existing assembler programs.
4070 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
4071 not available; use the synonym @code{.ln} in that context.
4076 @ifclear no-line-dir
4077 Even though this is a directive associated with the @code{a.out} or
4078 @code{b.out} object-code formats, @code{@value{AS}} still recognizes it
4079 when producing COFF output, and treats @samp{.line} as though it
4080 were the COFF @samp{.ln} @emph{if} it is found outside a
4081 @code{.def}/@code{.endef} pair.
4083 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4084 used by compilers to generate auxiliary symbol information for
4089 @section @code{.linkonce [@var{type}]}
4091 @cindex @code{linkonce} directive
4092 @cindex common sections
4093 Mark the current section so that the linker only includes a single copy of it.
4094 This may be used to include the same section in several different object files,
4095 but ensure that the linker will only include it once in the final output file.
4096 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4097 Duplicate sections are detected based on the section name, so it should be
4100 This directive is only supported by a few object file formats; as of this
4101 writing, the only object file format which supports it is the Portable
4102 Executable format used on Windows NT.
4104 The @var{type} argument is optional. If specified, it must be one of the
4105 following strings. For example:
4109 Not all types may be supported on all object file formats.
4113 Silently discard duplicate sections. This is the default.
4116 Warn if there are duplicate sections, but still keep only one copy.
4119 Warn if any of the duplicates have different sizes.
4122 Warn if any of the duplicates do not have exactly the same contents.
4126 @section @code{.ln @var{line-number}}
4128 @cindex @code{ln} directive
4129 @ifclear no-line-dir
4130 @samp{.ln} is a synonym for @samp{.line}.
4133 Tell @code{@value{AS}} to change the logical line number. @var{line-number}
4134 must be an absolute expression. The next line has that logical
4135 line number, so any other statements on the current line (after a
4136 statement separator character @code{;}) are reported as on logical
4137 line number @var{line-number} @minus{} 1.
4140 This directive is accepted, but ignored, when @code{@value{AS}} is
4141 configured for @code{b.out}; its effect is only associated with COFF
4147 @section @code{.mri @var{val}}
4149 @cindex @code{mri} directive
4150 @cindex MRI mode, temporarily
4151 If @var{val} is non-zero, this tells @code{@value{AS}} to enter MRI mode. If
4152 @var{val} is zero, this tells @code{@value{AS}} to exit MRI mode. This change
4153 affects code assembled until the next @code{.mri} directive, or until the end
4154 of the file. @xref{M, MRI mode, MRI mode}.
4157 @section @code{.list}
4159 @cindex @code{list} directive
4160 @cindex listing control, turning on
4161 Control (in conjunction with the @code{.nolist} directive) whether or
4162 not assembly listings are generated. These two directives maintain an
4163 internal counter (which is zero initially). @code{.list} increments the
4164 counter, and @code{.nolist} decrements it. Assembly listings are
4165 generated whenever the counter is greater than zero.
4167 By default, listings are disabled. When you enable them (with the
4168 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4169 the initial value of the listing counter is one.
4172 @section @code{.long @var{expressions}}
4174 @cindex @code{long} directive
4175 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
4178 @c no one seems to know what this is for or whether this description is
4179 @c what it really ought to do
4181 @section @code{.lsym @var{symbol}, @var{expression}}
4183 @cindex @code{lsym} directive
4184 @cindex symbol, not referenced in assembly
4185 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
4186 the hash table, ensuring it cannot be referenced by name during the
4187 rest of the assembly. This sets the attributes of the symbol to be
4188 the same as the expression value:
4190 @var{other} = @var{descriptor} = 0
4191 @var{type} = @r{(section of @var{expression})}
4192 @var{value} = @var{expression}
4195 The new symbol is not flagged as external.
4199 @section @code{.macro}
4202 The commands @code{.macro} and @code{.endm} allow you to define macros that
4203 generate assembly output. For example, this definition specifies a macro
4204 @code{sum} that puts a sequence of numbers into memory:
4207 .macro sum from=0, to=5
4216 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
4228 @item .macro @var{macname}
4229 @itemx .macro @var{macname} @var{macargs} @dots{}
4230 @cindex @code{macro} directive
4231 Begin the definition of a macro called @var{macname}. If your macro
4232 definition requires arguments, specify their names after the macro name,
4233 separated by commas or spaces. You can supply a default value for any
4234 macro argument by following the name with @samp{=@var{deflt}}. For
4235 example, these are all valid @code{.macro} statements:
4239 Begin the definition of a macro called @code{comm}, which takes no
4242 @item .macro plus1 p, p1
4243 @itemx .macro plus1 p p1
4244 Either statement begins the definition of a macro called @code{plus1},
4245 which takes two arguments; within the macro definition, write
4246 @samp{\p} or @samp{\p1} to evaluate the arguments.
4248 @item .macro reserve_str p1=0 p2
4249 Begin the definition of a macro called @code{reserve_str}, with two
4250 arguments. The first argument has a default value, but not the second.
4251 After the definition is complete, you can call the macro either as
4252 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
4253 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
4254 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
4255 @samp{0}, and @samp{\p2} evaluating to @var{b}).
4258 When you call a macro, you can specify the argument values either by
4259 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
4260 @samp{sum to=17, from=9}.
4263 @cindex @code{endm} directive
4264 Mark the end of a macro definition.
4267 @cindex @code{exitm} directive
4268 Exit early from the current macro definition.
4270 @cindex number of macros executed
4271 @cindex macros, count executed
4273 @code{@value{AS}} maintains a counter of how many macros it has
4274 executed in this pseudo-variable; you can copy that number to your
4275 output with @samp{\@@}, but @emph{only within a macro definition}.
4278 @item LOCAL @var{name} [ , @dots{} ]
4279 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
4280 macro syntax'' with @samp{-a} or @samp{--alternate}.} @xref{Alternate,,
4281 Alternate macro syntax}.
4283 Generate a string replacement for each of the @var{name} arguments, and
4284 replace any instances of @var{name} in each macro expansion. The
4285 replacement string is unique in the assembly, and different for each
4286 separate macro expansion. @code{LOCAL} allows you to write macros that
4287 define symbols, without fear of conflict between separate macro expansions.
4292 @section @code{.nolist}
4294 @cindex @code{nolist} directive
4295 @cindex listing control, turning off
4296 Control (in conjunction with the @code{.list} directive) whether or
4297 not assembly listings are generated. These two directives maintain an
4298 internal counter (which is zero initially). @code{.list} increments the
4299 counter, and @code{.nolist} decrements it. Assembly listings are
4300 generated whenever the counter is greater than zero.
4303 @section @code{.octa @var{bignums}}
4305 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
4306 @cindex @code{octa} directive
4307 @cindex integer, 16-byte
4308 @cindex sixteen byte integer
4309 This directive expects zero or more bignums, separated by commas. For each
4310 bignum, it emits a 16-byte integer.
4312 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
4313 hence @emph{octa}-word for 16 bytes.
4316 @section @code{.org @var{new-lc} , @var{fill}}
4318 @cindex @code{org} directive
4319 @cindex location counter, advancing
4320 @cindex advancing location counter
4321 @cindex current address, advancing
4322 Advance the location counter of the current section to
4323 @var{new-lc}. @var{new-lc} is either an absolute expression or an
4324 expression with the same section as the current subsection. That is,
4325 you can't use @code{.org} to cross sections: if @var{new-lc} has the
4326 wrong section, the @code{.org} directive is ignored. To be compatible
4327 with former assemblers, if the section of @var{new-lc} is absolute,
4328 @code{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
4329 is the same as the current subsection.
4331 @code{.org} may only increase the location counter, or leave it
4332 unchanged; you cannot use @code{.org} to move the location counter
4335 @c double negative used below "not undefined" because this is a specific
4336 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
4337 @c section. doc@cygnus.com 18feb91
4338 Because @code{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
4339 may not be undefined. If you really detest this restriction we eagerly await
4340 a chance to share your improved assembler.
4342 Beware that the origin is relative to the start of the section, not
4343 to the start of the subsection. This is compatible with other
4344 people's assemblers.
4346 When the location counter (of the current subsection) is advanced, the
4347 intervening bytes are filled with @var{fill} which should be an
4348 absolute expression. If the comma and @var{fill} are omitted,
4349 @var{fill} defaults to zero.
4352 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4354 @cindex padding the location counter given a power of two
4355 @cindex @code{p2align} directive
4356 Pad the location counter (in the current subsection) to a particular
4357 storage boundary. The first expression (which must be absolute) is the
4358 number of low-order zero bits the location counter must have after
4359 advancement. For example @samp{.p2align 3} advances the location
4360 counter until it a multiple of 8. If the location counter is already a
4361 multiple of 8, no change is needed.
4363 The second expression (also absolute) gives the fill value to be stored in the
4364 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4365 padding bytes are normally zero. However, on some systems, if the section is
4366 marked as containing code and the fill value is omitted, the space is filled
4367 with no-op instructions.
4369 The third expression is also absolute, and is also optional. If it is present,
4370 it is the maximum number of bytes that should be skipped by this alignment
4371 directive. If doing the alignment would require skipping more bytes than the
4372 specified maximum, then the alignment is not done at all. You can omit the
4373 fill value (the second argument) entirely by simply using two commas after the
4374 required alignment; this can be useful if you want the alignment to be filled
4375 with no-op instructions when appropriate.
4377 @cindex @code{p2alignw} directive
4378 @cindex @code{p2alignl} directive
4379 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
4380 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
4381 pattern as a two byte word value. The @code{.p2alignl} directives treats the
4382 fill pattern as a four byte longword value. For example, @code{.p2alignw
4383 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4384 filled in with the value 0x368d (the exact placement of the bytes depends upon
4385 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4390 @section @code{.previous}
4392 @cindex @code{.previous} directive
4393 @cindex Section Stack
4394 This is one of the ELF section stack manipulation directives. The others are
4395 @pxref{Section}, @xref{SubSection}, @pxref{PushSection}, and
4398 This directive swaps the current section (and subsection) with most recently
4399 referenced section (and subsection) prior to this one. Multiple
4400 @code{.previous} directives in a row will flip between two sections (and their
4403 In terms of the section stack, this directive swaps the current section with
4404 the top section on the section stack.
4409 @section @code{.popsection}
4411 @cindex @code{.popsection} directive
4412 @cindex Section Stack
4413 This is one of the ELF section stack manipulation directives. The others are
4414 @pxref{Section}, @xref{SubSection}, @pxref{PushSection}, and
4417 This directive replaces the current section (and subsection) with the top
4418 section (and subsection) on the section stack. This section is popped off the
4423 @section @code{.print @var{string}}
4425 @cindex @code{print} directive
4426 @code{@value{AS}} will print @var{string} on the standard output during
4427 assembly. You must put @var{string} in double quotes.
4431 @section @code{.protected @var{names}}
4433 @cindex @code{.protected} directive
4435 This one of the ELF visibility directives. The other two are
4436 @pxref{Hidden} and @pxref{Internal}
4438 This directive overrides the named symbols default visibility (which is set by
4439 their binding: local, global or weak). The directive sets the visibility to
4440 @code{protected} which means that any references to the symbols from within the
4441 components that defines them must be resolved to the definition in that
4442 component, even if a definition in another component would normally preempt
4447 @section @code{.psize @var{lines} , @var{columns}}
4449 @cindex @code{psize} directive
4450 @cindex listing control: paper size
4451 @cindex paper size, for listings
4452 Use this directive to declare the number of lines---and, optionally, the
4453 number of columns---to use for each page, when generating listings.
4455 If you do not use @code{.psize}, listings use a default line-count
4456 of 60. You may omit the comma and @var{columns} specification; the
4457 default width is 200 columns.
4459 @code{@value{AS}} generates formfeeds whenever the specified number of
4460 lines is exceeded (or whenever you explicitly request one, using
4463 If you specify @var{lines} as @code{0}, no formfeeds are generated save
4464 those explicitly specified with @code{.eject}.
4467 @section @code{.purgem @var{name}}
4469 @cindex @code{purgem} directive
4470 Undefine the macro @var{name}, so that later uses of the string will not be
4471 expanded. @xref{Macro}.
4475 @section @code{.pushsection @var{name} , @var{subsection}}
4477 @cindex @code{.pushsection} directive
4478 @cindex Section Stack
4479 This is one of the ELF section stack manipulation directives. The others are
4480 @pxref{Section}, @xref{SubSection}, @pxref{PopSection}, and
4483 This directive is a synonym for @code{.section}. It psuhes the current section
4484 (and subsection) onto the top of the section stack, and then replaces the
4485 current section and subsection with @code{name} and @code{subsection}.
4489 @section @code{.quad @var{bignums}}
4491 @cindex @code{quad} directive
4492 @code{.quad} expects zero or more bignums, separated by commas. For
4493 each bignum, it emits
4495 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
4496 warning message; and just takes the lowest order 8 bytes of the bignum.
4497 @cindex eight-byte integer
4498 @cindex integer, 8-byte
4500 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
4501 hence @emph{quad}-word for 8 bytes.
4504 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
4505 warning message; and just takes the lowest order 16 bytes of the bignum.
4506 @cindex sixteen-byte integer
4507 @cindex integer, 16-byte
4511 @section @code{.rept @var{count}}
4513 @cindex @code{rept} directive
4514 Repeat the sequence of lines between the @code{.rept} directive and the next
4515 @code{.endr} directive @var{count} times.
4517 For example, assembling
4525 is equivalent to assembling
4534 @section @code{.sbttl "@var{subheading}"}
4536 @cindex @code{sbttl} directive
4537 @cindex subtitles for listings
4538 @cindex listing control: subtitle
4539 Use @var{subheading} as the title (third line, immediately after the
4540 title line) when generating assembly listings.
4542 This directive affects subsequent pages, as well as the current page if
4543 it appears within ten lines of the top of a page.
4547 @section @code{.scl @var{class}}
4549 @cindex @code{scl} directive
4550 @cindex symbol storage class (COFF)
4551 @cindex COFF symbol storage class
4552 Set the storage-class value for a symbol. This directive may only be
4553 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
4554 whether a symbol is static or external, or it may record further
4555 symbolic debugging information.
4558 The @samp{.scl} directive is primarily associated with COFF output; when
4559 configured to generate @code{b.out} output format, @code{@value{AS}}
4560 accepts this directive but ignores it.
4565 @section @code{.section @var{name}} (COFF version)
4567 @cindex @code{section} directive
4568 @cindex named section
4569 Use the @code{.section} directive to assemble the following code into a section
4572 This directive is only supported for targets that actually support arbitrarily
4573 named sections; on @code{a.out} targets, for example, it is not accepted, even
4574 with a standard @code{a.out} section name.
4576 For COFF targets, the @code{.section} directive is used in one of the following
4580 .section @var{name}[, "@var{flags}"]
4581 .section @var{name}[, @var{subsegment}]
4584 If the optional argument is quoted, it is taken as flags to use for the
4585 section. Each flag is a single character. The following flags are recognized:
4588 bss section (uninitialized data)
4590 section is not loaded
4600 shared section (meaningful for PE targets)
4603 If no flags are specified, the default flags depend upon the section name. If
4604 the section name is not recognized, the default will be for the section to be
4605 loaded and writable.
4607 If the optional argument to the @code{.section} directive is not quoted, it is
4608 taken as a subsegment number (@pxref{Sub-Sections}).
4611 @section @code{.section @var{name}} (ELF version)
4613 @cindex @code{section} directive
4614 @cindex named section
4616 @cindex Section Stack
4617 This is one of the ELF section stack manipulation directives. The others are
4618 @xref{SubSection}, @pxref{PushSection}@pxref{PopSection}, and
4622 For ELF targets, the @code{.section} directive is used like this:
4625 .section @var{name} [, "@var{flags}"[, @@@var{type}]]
4628 The optional @var{flags} argument is a quoted string which may contain any
4629 combintion of the following characters:
4632 section is allocatable
4636 section is executable
4639 The optional @var{type} argument may contain one of the following constants:
4642 section contains data
4644 section does not contain data (i.e., section only occupies space)
4647 If no flags are specified, the default flags depend upon the section name. If
4648 the section name is not recognized, the default will be for the section to have
4649 none of the above flags: it will not be allocated in memory, nor writable, nor
4650 executable. The section will contain data.
4652 For ELF targets, the assembler supports another type of @code{.section}
4653 directive for compatibility with the Solaris assembler:
4656 .section "@var{name}"[, @var{flags}...]
4659 Note that the section name is quoted. There may be a sequence of comma
4663 section is allocatable
4667 section is executable
4670 This directive replaces the current section and subsection. The replaced
4671 section and subsection are pushed onto the section stack. See the contents of
4672 the gas testsuite directory @code{gas/testsuite/gas/elf} for some examples of
4673 how this directive and the other section stack directives work.
4676 @section @code{.set @var{symbol}, @var{expression}}
4678 @cindex @code{set} directive
4679 @cindex symbol value, setting
4680 Set the value of @var{symbol} to @var{expression}. This
4681 changes @var{symbol}'s value and type to conform to
4682 @var{expression}. If @var{symbol} was flagged as external, it remains
4683 flagged (@pxref{Symbol Attributes}).
4685 You may @code{.set} a symbol many times in the same assembly.
4687 If you @code{.set} a global symbol, the value stored in the object
4688 file is the last value stored into it.
4691 The syntax for @code{set} on the HPPA is
4692 @samp{@var{symbol} .set @var{expression}}.
4696 @section @code{.short @var{expressions}}
4698 @cindex @code{short} directive
4700 @code{.short} is normally the same as @samp{.word}.
4701 @xref{Word,,@code{.word}}.
4703 In some configurations, however, @code{.short} and @code{.word} generate
4704 numbers of different lengths; @pxref{Machine Dependencies}.
4708 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
4711 This expects zero or more @var{expressions}, and emits
4712 a 16 bit number for each.
4717 @section @code{.single @var{flonums}}
4719 @cindex @code{single} directive
4720 @cindex floating point numbers (single)
4721 This directive assembles zero or more flonums, separated by commas. It
4722 has the same effect as @code{.float}.
4724 The exact kind of floating point numbers emitted depends on how
4725 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
4729 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
4730 numbers in @sc{ieee} format.
4735 @section @code{.size} (COFF version)
4737 @cindex @code{size} directive
4738 This directive is generated by compilers to include auxiliary debugging
4739 information in the symbol table. It is only permitted inside
4740 @code{.def}/@code{.endef} pairs.
4743 @samp{.size} is only meaningful when generating COFF format output; when
4744 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4748 @section @code{.size @var{name} , @var{expression}} (ELF version)
4749 @cindex @code{size} directive
4751 This directive is used to set the size associated with a symbol @var{name}.
4752 The size in bytes is computed from @var{expression} which can make use of label
4753 arithmetic. This directive is typically used to set the size of function
4757 @section @code{.sleb128 @var{expressions}}
4759 @cindex @code{sleb128} directive
4760 @var{sleb128} stands for ``signed little endian base 128.'' This is a
4761 compact, variable length representation of numbers used by the DWARF
4762 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
4764 @ifclear no-space-dir
4766 @section @code{.skip @var{size} , @var{fill}}
4768 @cindex @code{skip} directive
4769 @cindex filling memory
4770 This directive emits @var{size} bytes, each of value @var{fill}. Both
4771 @var{size} and @var{fill} are absolute expressions. If the comma and
4772 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
4776 @section @code{.space @var{size} , @var{fill}}
4778 @cindex @code{space} directive
4779 @cindex filling memory
4780 This directive emits @var{size} bytes, each of value @var{fill}. Both
4781 @var{size} and @var{fill} are absolute expressions. If the comma
4782 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
4787 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
4788 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
4789 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
4790 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
4799 @section @code{.space}
4800 @cindex @code{space} directive
4802 On the AMD 29K, this directive is ignored; it is accepted for
4803 compatibility with other AMD 29K assemblers.
4806 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
4807 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
4813 @section @code{.stabd, .stabn, .stabs}
4815 @cindex symbolic debuggers, information for
4816 @cindex @code{stab@var{x}} directives
4817 There are three directives that begin @samp{.stab}.
4818 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
4819 The symbols are not entered in the @code{@value{AS}} hash table: they
4820 cannot be referenced elsewhere in the source file.
4821 Up to five fields are required:
4825 This is the symbol's name. It may contain any character except
4826 @samp{\000}, so is more general than ordinary symbol names. Some
4827 debuggers used to code arbitrarily complex structures into symbol names
4831 An absolute expression. The symbol's type is set to the low 8 bits of
4832 this expression. Any bit pattern is permitted, but @code{@value{LD}}
4833 and debuggers choke on silly bit patterns.
4836 An absolute expression. The symbol's ``other'' attribute is set to the
4837 low 8 bits of this expression.
4840 An absolute expression. The symbol's descriptor is set to the low 16
4841 bits of this expression.
4844 An absolute expression which becomes the symbol's value.
4847 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
4848 or @code{.stabs} statement, the symbol has probably already been created;
4849 you get a half-formed symbol in your object file. This is
4850 compatible with earlier assemblers!
4853 @cindex @code{stabd} directive
4854 @item .stabd @var{type} , @var{other} , @var{desc}
4856 The ``name'' of the symbol generated is not even an empty string.
4857 It is a null pointer, for compatibility. Older assemblers used a
4858 null pointer so they didn't waste space in object files with empty
4861 The symbol's value is set to the location counter,
4862 relocatably. When your program is linked, the value of this symbol
4863 is the address of the location counter when the @code{.stabd} was
4866 @cindex @code{stabn} directive
4867 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
4868 The name of the symbol is set to the empty string @code{""}.
4870 @cindex @code{stabs} directive
4871 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
4872 All five fields are specified.
4878 @section @code{.string} "@var{str}"
4880 @cindex string, copying to object file
4881 @cindex @code{string} directive
4883 Copy the characters in @var{str} to the object file. You may specify more than
4884 one string to copy, separated by commas. Unless otherwise specified for a
4885 particular machine, the assembler marks the end of each string with a 0 byte.
4886 You can use any of the escape sequences described in @ref{Strings,,Strings}.
4889 @section @code{.struct @var{expression}}
4891 @cindex @code{struct} directive
4892 Switch to the absolute section, and set the section offset to @var{expression},
4893 which must be an absolute expression. You might use this as follows:
4902 This would define the symbol @code{field1} to have the value 0, the symbol
4903 @code{field2} to have the value 4, and the symbol @code{field3} to have the
4904 value 8. Assembly would be left in the absolute section, and you would need to
4905 use a @code{.section} directive of some sort to change to some other section
4906 before further assembly.
4910 @section @code{.subsection @var{name}}
4912 @cindex @code{.subsection} directive
4913 @cindex Section Stack
4914 This is one of the ELF section stack manipulation directives. The others are
4915 @pxref{Section}, @xref{PushSection}, @pxref{PopSection}, and
4918 This directive replaces the current subsection with @code{name}. The current
4919 section is not changed. The replaced subsection is put onto the section stack
4920 in place of the then current top of stack subsection.
4925 @section @code{.symver}
4926 @cindex @code{symver} directive
4927 @cindex symbol versioning
4928 @cindex versions of symbols
4929 Use the @code{.symver} directive to bind symbols to specific version nodes
4930 within a source file. This is only supported on ELF platforms, and is
4931 typically used when assembling files to be linked into a shared library.
4932 There are cases where it may make sense to use this in objects to be bound
4933 into an application itself so as to override a versioned symbol from a
4936 For ELF targets, the @code{.symver} directive is used like this:
4938 .symver @var{name}, @var{name2@@nodename}
4940 If the symbol @var{name} is defined within the file
4941 being assembled, the @code{.versym} directive effectively creates a symbol
4942 alias with the name @var{name2@@nodename}, and in fact the main reason that we
4943 just don't try and create a regular alias is that the @var{@@} character isn't
4944 permitted in symbol names. The @var{name2} part of the name is the actual name
4945 of the symbol by which it will be externally referenced. The name @var{name}
4946 itself is merely a name of convenience that is used so that it is possible to
4947 have definitions for multiple versions of a function within a single source
4948 file, and so that the compiler can unambiguously know which version of a
4949 function is being mentioned. The @var{nodename} portion of the alias should be
4950 the name of a node specified in the version script supplied to the linker when
4951 building a shared library. If you are attempting to override a versioned
4952 symbol from a shared library, then @var{nodename} should correspond to the
4953 nodename of the symbol you are trying to override.
4955 If the symbol @var{name} is not defined within the file being assembled, all
4956 references to @var{name} will be changed to @var{name2@@nodename}. If no
4957 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
4963 @section @code{.tag @var{structname}}
4965 @cindex COFF structure debugging
4966 @cindex structure debugging, COFF
4967 @cindex @code{tag} directive
4968 This directive is generated by compilers to include auxiliary debugging
4969 information in the symbol table. It is only permitted inside
4970 @code{.def}/@code{.endef} pairs. Tags are used to link structure
4971 definitions in the symbol table with instances of those structures.
4974 @samp{.tag} is only used when generating COFF format output; when
4975 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4981 @section @code{.text @var{subsection}}
4983 @cindex @code{text} directive
4984 Tells @code{@value{AS}} to assemble the following statements onto the end of
4985 the text subsection numbered @var{subsection}, which is an absolute
4986 expression. If @var{subsection} is omitted, subsection number zero
4990 @section @code{.title "@var{heading}"}
4992 @cindex @code{title} directive
4993 @cindex listing control: title line
4994 Use @var{heading} as the title (second line, immediately after the
4995 source file name and pagenumber) when generating assembly listings.
4997 This directive affects subsequent pages, as well as the current page if
4998 it appears within ten lines of the top of a page.
5001 @section @code{.type @var{int}} (COFF version)
5003 @cindex COFF symbol type
5004 @cindex symbol type, COFF
5005 @cindex @code{type} directive
5006 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5007 records the integer @var{int} as the type attribute of a symbol table entry.
5010 @samp{.type} is associated only with COFF format output; when
5011 @code{@value{AS}} is configured for @code{b.out} output, it accepts this
5012 directive but ignores it.
5015 @section @code{.type @var{name} , @var{type description}} (ELF version)
5017 @cindex ELF symbol type
5018 @cindex symbol type, ELF
5019 @cindex @code{type} directive
5020 This directive is used to set the type of symbol @var{name} to be either a
5021 function symbol or an ojbect symbol. There are five different syntaxes
5022 supported for the @var{type description} field, in order to provide
5023 comptability with various other assemblers. The syntaxes supported are:
5026 .type <name>,#function
5027 .type <name>,#object
5029 .type <name>,@@function
5030 .type <name>,@@object
5032 .type <name>,%function
5033 .type <name>,%object
5035 .type <name>,"function"
5036 .type <name>,"object"
5038 .type <name> STT_FUNCTION
5039 .type <name> STT_OBJECT
5043 @section @code{.uleb128 @var{expressions}}
5045 @cindex @code{uleb128} directive
5046 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
5047 compact, variable length representation of numbers used by the DWARF
5048 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
5052 @section @code{.val @var{addr}}
5054 @cindex @code{val} directive
5055 @cindex COFF value attribute
5056 @cindex value attribute, COFF
5057 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5058 records the address @var{addr} as the value attribute of a symbol table
5062 @samp{.val} is used only for COFF output; when @code{@value{AS}} is
5063 configured for @code{b.out}, it accepts this directive but ignores it.
5069 @section @code{.version "@var{string}"}
5071 @cindex @code{.version}
5072 This directive creates a @code{.note} section and places into it an ELF
5073 formatted note of type NT_VERSION. The note's name is set to @code{string}.
5078 @section @code{.vtable_entry @var{table}, @var{offset}}
5080 @cindex @code{.vtable_entry}
5081 This directive finds or creates a symbol @code{table} and creates a
5082 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
5085 @section @code{.vtable_inherit @var{child}, @var{parent}}
5087 @cindex @code{.vtable_inherit}
5088 This directive finds the symbol @code{child} and finds or creates the symbol
5089 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
5090 parent whoes addend is the value of the child symbol. As a special case the
5091 parent name of @code{0} is treated as refering the @code{*ABS*} section.
5096 @section @code{.weak @var{names}}
5098 @cindex @code{.weak}
5099 This directive sets the weak attribute on the comma seperated list of symbol
5100 @code{names}. If the symbols do not already exist, they will be created.
5104 @section @code{.word @var{expressions}}
5106 @cindex @code{word} directive
5107 This directive expects zero or more @var{expressions}, of any section,
5108 separated by commas.
5111 For each expression, @code{@value{AS}} emits a 32-bit number.
5114 For each expression, @code{@value{AS}} emits a 16-bit number.
5119 The size of the number emitted, and its byte order,
5120 depend on what target computer the assembly is for.
5123 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
5124 @c happen---32-bit addressability, period; no long/short jumps.
5125 @ifset DIFF-TBL-KLUGE
5126 @cindex difference tables altered
5127 @cindex altered difference tables
5129 @emph{Warning: Special Treatment to support Compilers}
5133 Machines with a 32-bit address space, but that do less than 32-bit
5134 addressing, require the following special treatment. If the machine of
5135 interest to you does 32-bit addressing (or doesn't require it;
5136 @pxref{Machine Dependencies}), you can ignore this issue.
5139 In order to assemble compiler output into something that works,
5140 @code{@value{AS}} occasionlly does strange things to @samp{.word} directives.
5141 Directives of the form @samp{.word sym1-sym2} are often emitted by
5142 compilers as part of jump tables. Therefore, when @code{@value{AS}} assembles a
5143 directive of the form @samp{.word sym1-sym2}, and the difference between
5144 @code{sym1} and @code{sym2} does not fit in 16 bits, @code{@value{AS}}
5145 creates a @dfn{secondary jump table}, immediately before the next label.
5146 This secondary jump table is preceded by a short-jump to the
5147 first byte after the secondary table. This short-jump prevents the flow
5148 of control from accidentally falling into the new table. Inside the
5149 table is a long-jump to @code{sym2}. The original @samp{.word}
5150 contains @code{sym1} minus the address of the long-jump to
5153 If there were several occurrences of @samp{.word sym1-sym2} before the
5154 secondary jump table, all of them are adjusted. If there was a
5155 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
5156 long-jump to @code{sym4} is included in the secondary jump table,
5157 and the @code{.word} directives are adjusted to contain @code{sym3}
5158 minus the address of the long-jump to @code{sym4}; and so on, for as many
5159 entries in the original jump table as necessary.
5162 @emph{This feature may be disabled by compiling @code{@value{AS}} with the
5163 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
5164 assembly language programmers.
5167 @c end DIFF-TBL-KLUGE
5170 @section Deprecated Directives
5172 @cindex deprecated directives
5173 @cindex obsolescent directives
5174 One day these directives won't work.
5175 They are included for compatibility with older assemblers.
5182 @node Machine Dependencies
5183 @chapter Machine Dependent Features
5185 @cindex machine dependencies
5186 The machine instruction sets are (almost by definition) different on
5187 each machine where @code{@value{AS}} runs. Floating point representations
5188 vary as well, and @code{@value{AS}} often supports a few additional
5189 directives or command-line options for compatibility with other
5190 assemblers on a particular platform. Finally, some versions of
5191 @code{@value{AS}} support special pseudo-instructions for branch
5194 This chapter discusses most of these differences, though it does not
5195 include details on any machine's instruction set. For details on that
5196 subject, see the hardware manufacturer's manual.
5200 * AMD29K-Dependent:: AMD 29K Dependent Features
5203 * ARC-Dependent:: ARC Dependent Features
5206 * ARM-Dependent:: ARM Dependent Features
5209 * D10V-Dependent:: D10V Dependent Features
5212 * D30V-Dependent:: D30V Dependent Features
5215 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
5218 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
5221 * HPPA-Dependent:: HPPA Dependent Features
5224 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
5227 * i386-Dependent:: Intel 80386 Dependent Features
5230 * i860-Dependent:: Intel 80860 Dependent Features
5233 * i960-Dependent:: Intel 80960 Dependent Features
5236 * M32R-Dependent:: M32R Dependent Features
5239 * M68K-Dependent:: M680x0 Dependent Features
5242 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
5245 * MIPS-Dependent:: MIPS Dependent Features
5248 * SH-Dependent:: Hitachi SH Dependent Features
5251 * PJ-Dependent:: picoJava Dependent Features
5254 * Sparc-Dependent:: SPARC Dependent Features
5257 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
5260 * V850-Dependent:: V850 Dependent Features
5263 * Z8000-Dependent:: Z8000 Dependent Features
5266 * Vax-Dependent:: VAX Dependent Features
5273 @c The following major nodes are *sections* in the GENERIC version, *chapters*
5274 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
5275 @c peculiarity: to preserve cross-references, there must be a node called
5276 @c "Machine Dependencies". Hence the conditional nodenames in each
5277 @c major node below. Node defaulting in makeinfo requires adjacency of
5278 @c node and sectioning commands; hence the repetition of @chapter BLAH
5279 @c in both conditional blocks.
5285 @chapter ARC Dependent Features
5288 @node Machine Dependencies
5289 @chapter ARC Dependent Features
5294 * ARC-Opts:: Options
5295 * ARC-Float:: Floating Point
5296 * ARC-Directives:: Sparc Machine Directives
5302 @cindex options for ARC
5304 @cindex architectures, ARC
5305 @cindex ARC architectures
5306 The ARC chip family includes several successive levels (or other
5307 variants) of chip, using the same core instruction set, but including
5308 a few additional instructions at each level.
5310 By default, @code{@value{AS}} assumes the core instruction set (ARC
5311 base). The @code{.cpu} pseudo-op is intended to be used to select
5315 @cindex @code{-mbig-endian} option (ARC)
5316 @cindex @code{-mlittle-endian} option (ARC)
5317 @cindex ARC big-endian output
5318 @cindex ARC little-endian output
5319 @cindex big-endian output, ARC
5320 @cindex little-endian output, ARC
5322 @itemx -mlittle-endian
5323 Any @sc{arc} configuration of @code{@value{AS}} can select big-endian or
5324 little-endian output at run time (unlike most other @sc{gnu} development
5325 tools, which must be configured for one or the other). Use
5326 @samp{-mbig-endian} to select big-endian output, and @samp{-mlittle-endian}
5331 @section Floating Point
5333 @cindex floating point, ARC (@sc{ieee})
5334 @cindex ARC floating point (@sc{ieee})
5335 The ARC cpu family currently does not have hardware floating point
5336 support. Software floating point support is provided by @code{GCC}
5337 and uses @sc{ieee} floating-point numbers.
5339 @node ARC-Directives
5340 @section ARC Machine Directives
5342 @cindex ARC machine directives
5343 @cindex machine directives, ARC
5344 The ARC version of @code{@value{AS}} supports the following additional
5349 @cindex @code{cpu} directive, SPARC
5350 This must be followed by the desired cpu.
5351 The ARC is intended to be customizable, @code{.cpu} is used to
5352 select the desired variant [though currently there are none].
5359 @include c-a29k.texi
5368 @node Machine Dependencies
5369 @chapter Machine Dependent Features
5371 The machine instruction sets are different on each Hitachi chip family,
5372 and there are also some syntax differences among the families. This
5373 chapter describes the specific @code{@value{AS}} features for each
5377 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
5378 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
5379 * SH-Dependent:: Hitachi SH Dependent Features
5386 @include c-d10v.texi
5390 @include c-d30v.texi
5394 @include c-h8300.texi
5398 @include c-h8500.texi
5402 @include c-hppa.texi
5406 @include c-i370.texi
5410 @include c-i386.texi
5414 @include c-i860.texi
5418 @include c-i960.texi
5422 @include c-m32r.texi
5426 @include c-m68k.texi
5430 @include c-m68hc11.texi
5434 @include c-mips.texi
5438 @include c-ns32k.texi
5450 @include c-sparc.texi
5454 @include c-tic54x.texi
5466 @include c-v850.texi
5470 @c reverse effect of @down at top of generic Machine-Dep chapter
5474 @node Reporting Bugs
5475 @chapter Reporting Bugs
5476 @cindex bugs in assembler
5477 @cindex reporting bugs in assembler
5479 Your bug reports play an essential role in making @code{@value{AS}} reliable.
5481 Reporting a bug may help you by bringing a solution to your problem, or it may
5482 not. But in any case the principal function of a bug report is to help the
5483 entire community by making the next version of @code{@value{AS}} work better.
5484 Bug reports are your contribution to the maintenance of @code{@value{AS}}.
5486 In order for a bug report to serve its purpose, you must include the
5487 information that enables us to fix the bug.
5490 * Bug Criteria:: Have you found a bug?
5491 * Bug Reporting:: How to report bugs
5495 @section Have you found a bug?
5496 @cindex bug criteria
5498 If you are not sure whether you have found a bug, here are some guidelines:
5501 @cindex fatal signal
5502 @cindex assembler crash
5503 @cindex crash of assembler
5505 If the assembler gets a fatal signal, for any input whatever, that is a
5506 @code{@value{AS}} bug. Reliable assemblers never crash.
5508 @cindex error on valid input
5510 If @code{@value{AS}} produces an error message for valid input, that is a bug.
5512 @cindex invalid input
5514 If @code{@value{AS}} does not produce an error message for invalid input, that
5515 is a bug. However, you should note that your idea of ``invalid input'' might
5516 be our idea of ``an extension'' or ``support for traditional practice''.
5519 If you are an experienced user of assemblers, your suggestions for improvement
5520 of @code{@value{AS}} are welcome in any case.
5524 @section How to report bugs
5526 @cindex assembler bugs, reporting
5528 A number of companies and individuals offer support for @sc{gnu} products. If
5529 you obtained @code{@value{AS}} from a support organization, we recommend you
5530 contact that organization first.
5532 You can find contact information for many support companies and
5533 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
5536 In any event, we also recommend that you send bug reports for @code{@value{AS}}
5537 to @samp{bug-gnu-utils@@gnu.org}.
5539 The fundamental principle of reporting bugs usefully is this:
5540 @strong{report all the facts}. If you are not sure whether to state a
5541 fact or leave it out, state it!
5543 Often people omit facts because they think they know what causes the problem
5544 and assume that some details do not matter. Thus, you might assume that the
5545 name of a symbol you use in an example does not matter. Well, probably it does
5546 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
5547 happens to fetch from the location where that name is stored in memory;
5548 perhaps, if the name were different, the contents of that location would fool
5549 the assembler into doing the right thing despite the bug. Play it safe and
5550 give a specific, complete example. That is the easiest thing for you to do,
5551 and the most helpful.
5553 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
5554 it is new to us. Therefore, always write your bug reports on the assumption
5555 that the bug has not been reported previously.
5557 Sometimes people give a few sketchy facts and ask, ``Does this ring a
5558 bell?'' Those bug reports are useless, and we urge everyone to
5559 @emph{refuse to respond to them} except to chide the sender to report
5562 To enable us to fix the bug, you should include all these things:
5566 The version of @code{@value{AS}}. @code{@value{AS}} announces it if you start
5567 it with the @samp{--version} argument.
5569 Without this, we will not know whether there is any point in looking for
5570 the bug in the current version of @code{@value{AS}}.
5573 Any patches you may have applied to the @code{@value{AS}} source.
5576 The type of machine you are using, and the operating system name and
5580 What compiler (and its version) was used to compile @code{@value{AS}}---e.g.
5584 The command arguments you gave the assembler to assemble your example and
5585 observe the bug. To guarantee you will not omit something important, list them
5586 all. A copy of the Makefile (or the output from make) is sufficient.
5588 If we were to try to guess the arguments, we would probably guess wrong
5589 and then we might not encounter the bug.
5592 A complete input file that will reproduce the bug. If the bug is observed when
5593 the assembler is invoked via a compiler, send the assembler source, not the
5594 high level language source. Most compilers will produce the assembler source
5595 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
5596 the options @samp{-v --save-temps}; this will save the assembler source in a
5597 file with an extension of @file{.s}, and also show you exactly how
5598 @code{@value{AS}} is being run.
5601 A description of what behavior you observe that you believe is
5602 incorrect. For example, ``It gets a fatal signal.''
5604 Of course, if the bug is that @code{@value{AS}} gets a fatal signal, then we
5605 will certainly notice it. But if the bug is incorrect output, we might not
5606 notice unless it is glaringly wrong. You might as well not give us a chance to
5609 Even if the problem you experience is a fatal signal, you should still say so
5610 explicitly. Suppose something strange is going on, such as, your copy of
5611 @code{@value{AS}} is out of synch, or you have encountered a bug in the C
5612 library on your system. (This has happened!) Your copy might crash and ours
5613 would not. If you told us to expect a crash, then when ours fails to crash, we
5614 would know that the bug was not happening for us. If you had not told us to
5615 expect a crash, then we would not be able to draw any conclusion from our
5619 If you wish to suggest changes to the @code{@value{AS}} source, send us context
5620 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
5621 option. Always send diffs from the old file to the new file. If you even
5622 discuss something in the @code{@value{AS}} source, refer to it by context, not
5625 The line numbers in our development sources will not match those in your
5626 sources. Your line numbers would convey no useful information to us.
5629 Here are some things that are not necessary:
5633 A description of the envelope of the bug.
5635 Often people who encounter a bug spend a lot of time investigating
5636 which changes to the input file will make the bug go away and which
5637 changes will not affect it.
5639 This is often time consuming and not very useful, because the way we
5640 will find the bug is by running a single example under the debugger
5641 with breakpoints, not by pure deduction from a series of examples.
5642 We recommend that you save your time for something else.
5644 Of course, if you can find a simpler example to report @emph{instead}
5645 of the original one, that is a convenience for us. Errors in the
5646 output will be easier to spot, running under the debugger will take
5647 less time, and so on.
5649 However, simplification is not vital; if you do not want to do this,
5650 report the bug anyway and send us the entire test case you used.
5653 A patch for the bug.
5655 A patch for the bug does help us if it is a good one. But do not omit
5656 the necessary information, such as the test case, on the assumption that
5657 a patch is all we need. We might see problems with your patch and decide
5658 to fix the problem another way, or we might not understand it at all.
5660 Sometimes with a program as complicated as @code{@value{AS}} it is very hard to
5661 construct an example that will make the program follow a certain path through
5662 the code. If you do not send us the example, we will not be able to construct
5663 one, so we will not be able to verify that the bug is fixed.
5665 And if we cannot understand what bug you are trying to fix, or why your
5666 patch should be an improvement, we will not install it. A test case will
5667 help us to understand.
5670 A guess about what the bug is or what it depends on.
5672 Such guesses are usually wrong. Even we cannot guess right about such
5673 things without first using the debugger to find the facts.
5676 @node Acknowledgements
5677 @chapter Acknowledgements
5679 If you have contributed to @code{@value{AS}} and your name isn't listed here,
5680 it is not meant as a slight. We just don't know about it. Send mail to the
5681 maintainer, and we'll correct the situation. Currently
5683 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
5685 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
5688 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
5689 information and the 68k series machines, most of the preprocessing pass, and
5690 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
5692 K. Richard Pixley maintained GAS for a while, adding various enhancements and
5693 many bug fixes, including merging support for several processors, breaking GAS
5694 up to handle multiple object file format back ends (including heavy rewrite,
5695 testing, an integration of the coff and b.out back ends), adding configuration
5696 including heavy testing and verification of cross assemblers and file splits
5697 and renaming, converted GAS to strictly ANSI C including full prototypes, added
5698 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
5699 port (including considerable amounts of reverse engineering), a SPARC opcode
5700 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
5701 assertions and made them work, much other reorganization, cleanup, and lint.
5703 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
5704 in format-specific I/O modules.
5706 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
5707 has done much work with it since.
5709 The Intel 80386 machine description was written by Eliot Dresselhaus.
5711 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
5713 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
5714 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
5716 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
5717 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
5718 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
5719 support a.out format.
5721 Support for the Zilog Z8k and Hitachi H8/300 and H8/500 processors (tc-z8k,
5722 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
5723 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
5724 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
5727 John Gilmore built the AMD 29000 support, added @code{.include} support, and
5728 simplified the configuration of which versions accept which directives. He
5729 updated the 68k machine description so that Motorola's opcodes always produced
5730 fixed-size instructions (e.g. @code{jsr}), while synthetic instructions
5731 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
5732 cross-compilation support, and one bug in relaxation that took a week and
5733 required the proverbial one-bit fix.
5735 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
5736 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
5737 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
5738 PowerPC assembler, and made a few other minor patches.
5740 Steve Chamberlain made @code{@value{AS}} able to generate listings.
5742 Hewlett-Packard contributed support for the HP9000/300.
5744 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
5745 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
5746 formats). This work was supported by both the Center for Software Science at
5747 the University of Utah and Cygnus Support.
5749 Support for ELF format files has been worked on by Mark Eichin of Cygnus
5750 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
5751 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
5752 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
5753 and some initial 64-bit support).
5755 Linas Vepstas added GAS support for the ESA/390 "IBM 370" architecture.
5757 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
5758 support for openVMS/Alpha.
5760 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
5763 Several engineers at Cygnus Support have also provided many small bug fixes and
5764 configuration enhancements.
5766 Many others have contributed large or small bugfixes and enhancements. If
5767 you have contributed significant work and are not mentioned on this list, and
5768 want to be, let us know. Some of the history has been lost; we are not
5769 intentionally leaving anyone out.
5771 @node GNU Free Documentation License
5772 @chapter GNU Free Documentation License
5774 GNU Free Documentation License
5776 Version 1.1, March 2000
5778 Copyright (C) 2000 Free Software Foundation, Inc.
5779 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
5781 Everyone is permitted to copy and distribute verbatim copies
5782 of this license document, but changing it is not allowed.
5787 The purpose of this License is to make a manual, textbook, or other
5788 written document "free" in the sense of freedom: to assure everyone
5789 the effective freedom to copy and redistribute it, with or without
5790 modifying it, either commercially or noncommercially. Secondarily,
5791 this License preserves for the author and publisher a way to get
5792 credit for their work, while not being considered responsible for
5793 modifications made by others.
5795 This License is a kind of "copyleft", which means that derivative
5796 works of the document must themselves be free in the same sense. It
5797 complements the GNU General Public License, which is a copyleft
5798 license designed for free software.
5800 We have designed this License in order to use it for manuals for free
5801 software, because free software needs free documentation: a free
5802 program should come with manuals providing the same freedoms that the
5803 software does. But this License is not limited to software manuals;
5804 it can be used for any textual work, regardless of subject matter or
5805 whether it is published as a printed book. We recommend this License
5806 principally for works whose purpose is instruction or reference.
5809 1. APPLICABILITY AND DEFINITIONS
5811 This License applies to any manual or other work that contains a
5812 notice placed by the copyright holder saying it can be distributed
5813 under the terms of this License. The "Document", below, refers to any
5814 such manual or work. Any member of the public is a licensee, and is
5817 A "Modified Version" of the Document means any work containing the
5818 Document or a portion of it, either copied verbatim, or with
5819 modifications and/or translated into another language.
5821 A "Secondary Section" is a named appendix or a front-matter section of
5822 the Document that deals exclusively with the relationship of the
5823 publishers or authors of the Document to the Document's overall subject
5824 (or to related matters) and contains nothing that could fall directly
5825 within that overall subject. (For example, if the Document is in part a
5826 textbook of mathematics, a Secondary Section may not explain any
5827 mathematics.) The relationship could be a matter of historical
5828 connection with the subject or with related matters, or of legal,
5829 commercial, philosophical, ethical or political position regarding
5832 The "Invariant Sections" are certain Secondary Sections whose titles
5833 are designated, as being those of Invariant Sections, in the notice
5834 that says that the Document is released under this License.
5836 The "Cover Texts" are certain short passages of text that are listed,
5837 as Front-Cover Texts or Back-Cover Texts, in the notice that says that
5838 the Document is released under this License.
5840 A "Transparent" copy of the Document means a machine-readable copy,
5841 represented in a format whose specification is available to the
5842 general public, whose contents can be viewed and edited directly and
5843 straightforwardly with generic text editors or (for images composed of
5844 pixels) generic paint programs or (for drawings) some widely available
5845 drawing editor, and that is suitable for input to text formatters or
5846 for automatic translation to a variety of formats suitable for input
5847 to text formatters. A copy made in an otherwise Transparent file
5848 format whose markup has been designed to thwart or discourage
5849 subsequent modification by readers is not Transparent. A copy that is
5850 not "Transparent" is called "Opaque".
5852 Examples of suitable formats for Transparent copies include plain
5853 ASCII without markup, Texinfo input format, LaTeX input format, SGML
5854 or XML using a publicly available DTD, and standard-conforming simple
5855 HTML designed for human modification. Opaque formats include
5856 PostScript, PDF, proprietary formats that can be read and edited only
5857 by proprietary word processors, SGML or XML for which the DTD and/or
5858 processing tools are not generally available, and the
5859 machine-generated HTML produced by some word processors for output
5862 The "Title Page" means, for a printed book, the title page itself,
5863 plus such following pages as are needed to hold, legibly, the material
5864 this License requires to appear in the title page. For works in
5865 formats which do not have any title page as such, "Title Page" means
5866 the text near the most prominent appearance of the work's title,
5867 preceding the beginning of the body of the text.
5872 You may copy and distribute the Document in any medium, either
5873 commercially or noncommercially, provided that this License, the
5874 copyright notices, and the license notice saying this License applies
5875 to the Document are reproduced in all copies, and that you add no other
5876 conditions whatsoever to those of this License. You may not use
5877 technical measures to obstruct or control the reading or further
5878 copying of the copies you make or distribute. However, you may accept
5879 compensation in exchange for copies. If you distribute a large enough
5880 number of copies you must also follow the conditions in section 3.
5882 You may also lend copies, under the same conditions stated above, and
5883 you may publicly display copies.
5886 3. COPYING IN QUANTITY
5888 If you publish printed copies of the Document numbering more than 100,
5889 and the Document's license notice requires Cover Texts, you must enclose
5890 the copies in covers that carry, clearly and legibly, all these Cover
5891 Texts: Front-Cover Texts on the front cover, and Back-Cover Texts on
5892 the back cover. Both covers must also clearly and legibly identify
5893 you as the publisher of these copies. The front cover must present
5894 the full title with all words of the title equally prominent and
5895 visible. You may add other material on the covers in addition.
5896 Copying with changes limited to the covers, as long as they preserve
5897 the title of the Document and satisfy these conditions, can be treated
5898 as verbatim copying in other respects.
5900 If the required texts for either cover are too voluminous to fit
5901 legibly, you should put the first ones listed (as many as fit
5902 reasonably) on the actual cover, and continue the rest onto adjacent
5905 If you publish or distribute Opaque copies of the Document numbering
5906 more than 100, you must either include a machine-readable Transparent
5907 copy along with each Opaque copy, or state in or with each Opaque copy
5908 a publicly-accessible computer-network location containing a complete
5909 Transparent copy of the Document, free of added material, which the
5910 general network-using public has access to download anonymously at no
5911 charge using public-standard network protocols. If you use the latter
5912 option, you must take reasonably prudent steps, when you begin
5913 distribution of Opaque copies in quantity, to ensure that this
5914 Transparent copy will remain thus accessible at the stated location
5915 until at least one year after the last time you distribute an Opaque
5916 copy (directly or through your agents or retailers) of that edition to
5919 It is requested, but not required, that you contact the authors of the
5920 Document well before redistributing any large number of copies, to give
5921 them a chance to provide you with an updated version of the Document.
5926 You may copy and distribute a Modified Version of the Document under
5927 the conditions of sections 2 and 3 above, provided that you release
5928 the Modified Version under precisely this License, with the Modified
5929 Version filling the role of the Document, thus licensing distribution
5930 and modification of the Modified Version to whoever possesses a copy
5931 of it. In addition, you must do these things in the Modified Version:
5933 A. Use in the Title Page (and on the covers, if any) a title distinct
5934 from that of the Document, and from those of previous versions
5935 (which should, if there were any, be listed in the History section
5936 of the Document). You may use the same title as a previous version
5937 if the original publisher of that version gives permission.
5938 B. List on the Title Page, as authors, one or more persons or entities
5939 responsible for authorship of the modifications in the Modified
5940 Version, together with at least five of the principal authors of the
5941 Document (all of its principal authors, if it has less than five).
5942 C. State on the Title page the name of the publisher of the
5943 Modified Version, as the publisher.
5944 D. Preserve all the copyright notices of the Document.
5945 E. Add an appropriate copyright notice for your modifications
5946 adjacent to the other copyright notices.
5947 F. Include, immediately after the copyright notices, a license notice
5948 giving the public permission to use the Modified Version under the
5949 terms of this License, in the form shown in the Addendum below.
5950 G. Preserve in that license notice the full lists of Invariant Sections
5951 and required Cover Texts given in the Document's license notice.
5952 H. Include an unaltered copy of this License.
5953 I. Preserve the section entitled "History", and its title, and add to
5954 it an item stating at least the title, year, new authors, and
5955 publisher of the Modified Version as given on the Title Page. If
5956 there is no section entitled "History" in the Document, create one
5957 stating the title, year, authors, and publisher of the Document as
5958 given on its Title Page, then add an item describing the Modified
5959 Version as stated in the previous sentence.
5960 J. Preserve the network location, if any, given in the Document for
5961 public access to a Transparent copy of the Document, and likewise
5962 the network locations given in the Document for previous versions
5963 it was based on. These may be placed in the "History" section.
5964 You may omit a network location for a work that was published at
5965 least four years before the Document itself, or if the original
5966 publisher of the version it refers to gives permission.
5967 K. In any section entitled "Acknowledgements" or "Dedications",
5968 preserve the section's title, and preserve in the section all the
5969 substance and tone of each of the contributor acknowledgements
5970 and/or dedications given therein.
5971 L. Preserve all the Invariant Sections of the Document,
5972 unaltered in their text and in their titles. Section numbers
5973 or the equivalent are not considered part of the section titles.
5974 M. Delete any section entitled "Endorsements". Such a section
5975 may not be included in the Modified Version.
5976 N. Do not retitle any existing section as "Endorsements"
5977 or to conflict in title with any Invariant Section.
5979 If the Modified Version includes new front-matter sections or
5980 appendices that qualify as Secondary Sections and contain no material
5981 copied from the Document, you may at your option designate some or all
5982 of these sections as invariant. To do this, add their titles to the
5983 list of Invariant Sections in the Modified Version's license notice.
5984 These titles must be distinct from any other section titles.
5986 You may add a section entitled "Endorsements", provided it contains
5987 nothing but endorsements of your Modified Version by various
5988 parties--for example, statements of peer review or that the text has
5989 been approved by an organization as the authoritative definition of a
5992 You may add a passage of up to five words as a Front-Cover Text, and a
5993 passage of up to 25 words as a Back-Cover Text, to the end of the list
5994 of Cover Texts in the Modified Version. Only one passage of
5995 Front-Cover Text and one of Back-Cover Text may be added by (or
5996 through arrangements made by) any one entity. If the Document already
5997 includes a cover text for the same cover, previously added by you or
5998 by arrangement made by the same entity you are acting on behalf of,
5999 you may not add another; but you may replace the old one, on explicit
6000 permission from the previous publisher that added the old one.
6002 The author(s) and publisher(s) of the Document do not by this License
6003 give permission to use their names for publicity for or to assert or
6004 imply endorsement of any Modified Version.
6007 5. COMBINING DOCUMENTS
6009 You may combine the Document with other documents released under this
6010 License, under the terms defined in section 4 above for modified
6011 versions, provided that you include in the combination all of the
6012 Invariant Sections of all of the original documents, unmodified, and
6013 list them all as Invariant Sections of your combined work in its
6016 The combined work need only contain one copy of this License, and
6017 multiple identical Invariant Sections may be replaced with a single
6018 copy. If there are multiple Invariant Sections with the same name but
6019 different contents, make the title of each such section unique by
6020 adding at the end of it, in parentheses, the name of the original
6021 author or publisher of that section if known, or else a unique number.
6022 Make the same adjustment to the section titles in the list of
6023 Invariant Sections in the license notice of the combined work.
6025 In the combination, you must combine any sections entitled "History"
6026 in the various original documents, forming one section entitled
6027 "History"; likewise combine any sections entitled "Acknowledgements",
6028 and any sections entitled "Dedications". You must delete all sections
6029 entitled "Endorsements."
6032 6. COLLECTIONS OF DOCUMENTS
6034 You may make a collection consisting of the Document and other documents
6035 released under this License, and replace the individual copies of this
6036 License in the various documents with a single copy that is included in
6037 the collection, provided that you follow the rules of this License for
6038 verbatim copying of each of the documents in all other respects.
6040 You may extract a single document from such a collection, and distribute
6041 it individually under this License, provided you insert a copy of this
6042 License into the extracted document, and follow this License in all
6043 other respects regarding verbatim copying of that document.
6046 7. AGGREGATION WITH INDEPENDENT WORKS
6048 A compilation of the Document or its derivatives with other separate
6049 and independent documents or works, in or on a volume of a storage or
6050 distribution medium, does not as a whole count as a Modified Version
6051 of the Document, provided no compilation copyright is claimed for the
6052 compilation. Such a compilation is called an "aggregate", and this
6053 License does not apply to the other self-contained works thus compiled
6054 with the Document, on account of their being thus compiled, if they
6055 are not themselves derivative works of the Document.
6057 If the Cover Text requirement of section 3 is applicable to these
6058 copies of the Document, then if the Document is less than one quarter
6059 of the entire aggregate, the Document's Cover Texts may be placed on
6060 covers that surround only the Document within the aggregate.
6061 Otherwise they must appear on covers around the whole aggregate.
6066 Translation is considered a kind of modification, so you may
6067 distribute translations of the Document under the terms of section 4.
6068 Replacing Invariant Sections with translations requires special
6069 permission from their copyright holders, but you may include
6070 translations of some or all Invariant Sections in addition to the
6071 original versions of these Invariant Sections. You may include a
6072 translation of this License provided that you also include the
6073 original English version of this License. In case of a disagreement
6074 between the translation and the original English version of this
6075 License, the original English version will prevail.
6080 You may not copy, modify, sublicense, or distribute the Document except
6081 as expressly provided for under this License. Any other attempt to
6082 copy, modify, sublicense or distribute the Document is void, and will
6083 automatically terminate your rights under this License. However,
6084 parties who have received copies, or rights, from you under this
6085 License will not have their licenses terminated so long as such
6086 parties remain in full compliance.
6089 10. FUTURE REVISIONS OF THIS LICENSE
6091 The Free Software Foundation may publish new, revised versions
6092 of the GNU Free Documentation License from time to time. Such new
6093 versions will be similar in spirit to the present version, but may
6094 differ in detail to address new problems or concerns. See
6095 http://www.gnu.org/copyleft/.
6097 Each version of the License is given a distinguishing version number.
6098 If the Document specifies that a particular numbered version of this
6099 License "or any later version" applies to it, you have the option of
6100 following the terms and conditions either of that specified version or
6101 of any later version that has been published (not as a draft) by the
6102 Free Software Foundation. If the Document does not specify a version
6103 number of this License, you may choose any version ever published (not
6104 as a draft) by the Free Software Foundation.
6107 ADDENDUM: How to use this License for your documents
6109 To use this License in a document you have written, include a copy of
6110 the License in the document and put the following copyright and
6111 license notices just after the title page:
6114 Copyright (c) YEAR YOUR NAME.
6115 Permission is granted to copy, distribute and/or modify this document
6116 under the terms of the GNU Free Documentation License, Version 1.1
6117 or any later version published by the Free Software Foundation;
6118 with the Invariant Sections being LIST THEIR TITLES, with the
6119 Front-Cover Texts being LIST, and with the Back-Cover Texts being LIST.
6120 A copy of the license is included in the section entitled "GNU
6121 Free Documentation License".
6124 If you have no Invariant Sections, write "with no Invariant Sections"
6125 instead of saying which ones are invariant. If you have no
6126 Front-Cover Texts, write "no Front-Cover Texts" instead of
6127 "Front-Cover Texts being LIST"; likewise for Back-Cover Texts.
6129 If your document contains nontrivial examples of program code, we
6130 recommend releasing these examples in parallel under your choice of
6131 free software license, such as the GNU General Public License,
6132 to permit their use in free software.