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
215 [ -m[arm]1 | -m[arm]2 | -m[arm]250 | -m[arm]3 | -m[arm]6 | -m[arm]60 |
216 -m[arm]600 | -m[arm]610 | -m[arm]620 | -m[arm]7[t][[d]m[i]][fe] | -m[arm]70 |
217 -m[arm]700 | -m[arm]710[c] | -m[arm]7100 | -m[arm]7500 | -m[arm]8 |
218 -m[arm]810 | -m[arm]9 | -m[arm]920 | -m[arm]920t | -m[arm]9tdmi |
219 -mstrongarm | -mstrongarm110 | -mstrongarm1100 ]
220 [ -m[arm]v2 | -m[arm]v2a | -m[arm]v3 | -m[arm]v3m | -m[arm]v4 | -m[arm]v4t |
221 -m[arm]v5 | -[arm]v5t ]
223 [ -mfpa10 | -mfpa11 | -mfpe-old | -mno-fpu ]
225 [ -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant ]
226 [ -mthumb-interwork ]
237 @c Hitachi family chips have no machine-dependent assembler options
240 @c HPPA has no machine-dependent assembler options (yet).
246 @c The order here is important. See c-sparc.texi.
247 [ -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
248 -Av8plus | -Av8plusa | -Av9 | -Av9a ]
249 [ -xarch=v8plus | -xarch=v8plusa ] [ -bump ] [ -32 | -64 ]
252 [ -mcpu=54[123589] | -mcpu=54[56]lp ] [ -mfar-mode | -mf ]
253 [ -merrors-to-file <filename> | -me <filename> ]
256 @c Z8000 has no machine-dependent assembler options
259 @c see md_parse_option in tc-i960.c
260 [ -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC ]
264 [ --m32rx | --[no-]warn-explicit-parallel-conflicts | --W[n]p ]
267 [ -l ] [ -m68000 | -m68010 | -m68020 | ... ]
270 [ -jsri2bsr ] [ -sifilter ] [ -relax ]
274 [ -m68hc11 | -m68hc12 ]
275 [ --force-long-branchs ] [ --short-branchs ] [ --strict-direct-mode ]
276 [ --print-insn-syntax ] [ --print-opcodes ] [ --generate-example ]
279 [ -nocpp ] [ -EL ] [ -EB ] [ -G @var{num} ] [ -mcpu=@var{CPU} ]
280 [ -mips1 ] [ -mips2 ] [ -mips3 ] [ -mips4 ] [ -mips5 ]
281 [ -mips32 ] [ -mips64 ]
282 [ -m4650 ] [ -no-m4650 ]
283 [ --trap ] [ --break ]
284 [ --emulation=@var{name} ]
286 [ -- | @var{files} @dots{} ]
291 Turn on listings, in any of a variety of ways:
295 omit false conditionals
298 omit debugging directives
301 include high-level source
307 include macro expansions
310 omit forms processing
316 set the name of the listing file
319 You may combine these options; for example, use @samp{-aln} for assembly
320 listing without forms processing. The @samp{=file} option, if used, must be
321 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
324 Ignored. This option is accepted for script compatibility with calls to
327 @item --defsym @var{sym}=@var{value}
328 Define the symbol @var{sym} to be @var{value} before assembling the input file.
329 @var{value} must be an integer constant. As in C, a leading @samp{0x}
330 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
333 ``fast''---skip whitespace and comment preprocessing (assume source is
337 Generate stabs debugging information for each assembler line. This
338 may help debugging assembler code, if the debugger can handle it.
341 Generate DWARF2 debugging information for each assembler line. This
342 may help debugging assembler code, if the debugger can handle it. Note - this
343 option is only supported by some targets, not all of them.
346 Print a summary of the command line options and exit.
349 Print a summary of all target specific options and exit.
352 Add directory @var{dir} to the search list for @code{.include} directives.
355 Don't warn about signed overflow.
358 @ifclear DIFF-TBL-KLUGE
359 This option is accepted but has no effect on the @value{TARGET} family.
361 @ifset DIFF-TBL-KLUGE
362 Issue warnings when difference tables altered for long displacements.
367 Keep (in the symbol table) local symbols. On traditional a.out systems
368 these start with @samp{L}, but different systems have different local
371 @item -o @var{objfile}
372 Name the object-file output from @code{@value{AS}} @var{objfile}.
375 Fold the data section into the text section.
378 Print the maximum space (in bytes) and total time (in seconds) used by
381 @item --strip-local-absolute
382 Remove local absolute symbols from the outgoing symbol table.
386 Print the @code{as} version.
389 Print the @code{as} version and exit.
393 Suppress warning messages.
395 @item --fatal-warnings
396 Treat warnings as errors.
399 Don't suppress warning messages or treat them as errors.
408 Generate an object file even after errors.
410 @item -- | @var{files} @dots{}
411 Standard input, or source files to assemble.
416 The following options are available when @value{AS} is configured for
421 This option selects the core processor variant.
423 Select either big-endian (-EB) or little-endian (-EL) output.
428 The following options are available when @value{AS} is configured for the ARM
432 @item -m[arm][1|2|3|6|7|8|9][...]
433 Specify which ARM processor variant is the target.
434 @item -m[arm]v[2|2a|3|3m|4|4t|5|5t]
435 Specify which ARM architecture variant is used by the target.
436 @item -mthumb | -mall
437 Enable or disable Thumb only instruction decoding.
438 @item -mfpa10 | -mfpa11 | -mfpe-old | -mno-fpu
439 Select which Floating Point architcture is the target.
440 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant | -moabi
441 Select which procedure calling convention is in use.
443 Select either big-endian (-EB) or little-endian (-EL) output.
444 @item -mthumb-interwork
445 Specify that the code has been generated with interworking between Thumb and
448 Specify that PIC code has been generated.
453 The following options are available when @value{AS} is configured for
456 @cindex D10V optimization
457 @cindex optimization, D10V
459 Optimize output by parallelizing instructions.
464 The following options are available when @value{AS} is configured for a D30V
467 @cindex D30V optimization
468 @cindex optimization, D30V
470 Optimize output by parallelizing instructions.
474 Warn when nops are generated.
476 @cindex D30V nops after 32-bit multiply
478 Warn when a nop after a 32-bit multiply instruction is generated.
483 The following options are available when @value{AS} is configured for the
484 Intel 80960 processor.
487 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
488 Specify which variant of the 960 architecture is the target.
491 Add code to collect statistics about branches taken.
494 Do not alter compare-and-branch instructions for long displacements;
501 The following options are available when @value{AS} is configured for the
502 Mitsubishi M32R series.
507 Specify which processor in the M32R family is the target. The default
508 is normally the M32R, but this option changes it to the M32RX.
510 @item --warn-explicit-parallel-conflicts or --Wp
511 Produce warning messages when questionable parallel constructs are
514 @item --no-warn-explicit-parallel-conflicts or --Wnp
515 Do not produce warning messages when questionable parallel constructs are
522 The following options are available when @value{AS} is configured for the
523 Motorola 68000 series.
528 Shorten references to undefined symbols, to one word instead of two.
530 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030 | -m68040 | -m68060
531 @itemx | -m68302 | -m68331 | -m68332 | -m68333 | -m68340 | -mcpu32 | -m5200
532 Specify what processor in the 68000 family is the target. The default
533 is normally the 68020, but this can be changed at configuration time.
535 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
536 The target machine does (or does not) have a floating-point coprocessor.
537 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
538 the basic 68000 is not compatible with the 68881, a combination of the
539 two can be specified, since it's possible to do emulation of the
540 coprocessor instructions with the main processor.
542 @item -m68851 | -mno-68851
543 The target machine does (or does not) have a memory-management
544 unit coprocessor. The default is to assume an MMU for 68020 and up.
550 The following options are available when @value{AS} is configured for
551 a picoJava processor.
555 @cindex PJ endianness
556 @cindex endianness, PJ
557 @cindex big endian output, PJ
559 Generate ``big endian'' format output.
561 @cindex little endian output, PJ
563 Generate ``little endian'' format output.
569 The following options are available when @value{AS} is configured for the
570 Motorola 68HC11 or 68HC12 series.
574 @item -m68hc11 | -m68hc12
575 Specify what processor is the target. The default is
576 defined by the configuration option when building the assembler.
578 @item --force-long-branchs
579 Relative branches are turned into absolute ones. This concerns
580 conditional branches, unconditional branches and branches to a
583 @item -S | --short-branchs
584 Do not turn relative branchs into absolute ones
585 when the offset is out of range.
587 @item --strict-direct-mode
588 Do not turn the direct addressing mode into extended addressing mode
589 when the instruction does not support direct addressing mode.
591 @item --print-insn-syntax
592 Print the syntax of instruction in case of error.
594 @item --print-opcodes
595 print the list of instructions with syntax and then exit.
597 @item --generate-example
598 print an example of instruction for each possible instruction and then exit.
599 This option is only useful for testing @code{@value{AS}}.
605 The following options are available when @code{@value{AS}} is configured
606 for the SPARC architecture:
609 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
610 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
611 Explicitly select a variant of the SPARC architecture.
613 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
614 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
616 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
617 UltraSPARC extensions.
619 @item -xarch=v8plus | -xarch=v8plusa
620 For compatibility with the Solaris v9 assembler. These options are
621 equivalent to -Av8plus and -Av8plusa, respectively.
624 Warn when the assembler switches to another architecture.
629 The following options are available when @value{AS} is configured for the 'c54x
634 Enable extended addressing mode. All addresses and relocations will assume
635 extended addressing (usually 23 bits).
636 @item -mcpu=@var{CPU_VERSION}
637 Sets the CPU version being compiled for.
638 @item -merrors-to-file @var{FILENAME}
639 Redirect error output to a file, for broken systems which don't support such
640 behaviour in the shell.
645 The following options are available when @value{AS} is configured for
650 This option sets the largest size of an object that can be referenced
651 implicitly with the @code{gp} register. It is only accepted for targets that
652 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
654 @cindex MIPS endianness
655 @cindex endianness, MIPS
656 @cindex big endian output, MIPS
658 Generate ``big endian'' format output.
660 @cindex little endian output, MIPS
662 Generate ``little endian'' format output.
670 Generate code for a particular MIPS Instruction Set Architecture level.
671 @samp{-mips1} corresponds to the @sc{r2000} and @sc{r3000} processors,
672 @samp{-mips2} to the @sc{r6000} processor, and @samp{-mips3} to the @sc{r4000}
674 @samp{-mips5}, @samp{-mips32}, and @samp{-mips64} correspond
675 to generic @sc{MIPS V}, @sc{MIPS32}, and @sc{MIPS64} ISA
676 processors, respectively.
680 Generate code for the MIPS @sc{r4650} chip. This tells the assembler to accept
681 the @samp{mad} and @samp{madu} instruction, and to not schedule @samp{nop}
682 instructions around accesses to the @samp{HI} and @samp{LO} registers.
683 @samp{-no-m4650} turns off this option.
685 @item -mcpu=@var{CPU}
686 Generate code for a particular MIPS cpu. It is exactly equivalent to
687 @samp{-m@var{cpu}}, except that there are more value of @var{cpu}
691 @item --emulation=@var{name}
692 This option causes @code{@value{AS}} to emulate @code{@value{AS}} configured
693 for some other target, in all respects, including output format (choosing
694 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
695 debugging information or store symbol table information, and default
696 endianness. The available configuration names are: @samp{mipsecoff},
697 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
698 @samp{mipsbelf}. The first two do not alter the default endianness from that
699 of the primary target for which the assembler was configured; the others change
700 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
701 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
702 selection in any case.
704 This option is currently supported only when the primary target
705 @code{@value{AS}} is configured for is a MIPS ELF or ECOFF target.
706 Furthermore, the primary target or others specified with
707 @samp{--enable-targets=@dots{}} at configuration time must include support for
708 the other format, if both are to be available. For example, the Irix 5
709 configuration includes support for both.
711 Eventually, this option will support more configurations, with more
712 fine-grained control over the assembler's behavior, and will be supported for
716 @code{@value{AS}} ignores this option. It is accepted for compatibility with
724 Control how to deal with multiplication overflow and division by zero.
725 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
726 (and only work for Instruction Set Architecture level 2 and higher);
727 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
733 The following options are available when @value{AS} is configured for
739 Enable or disable the JSRI to BSR transformation. By default this is enabled.
740 The command line option @samp{-nojsri2bsr} can be used to disable it.
744 Enable or disable the silicon filter behaviour. By default this is disabled.
745 The default can be overidden by the @samp{-sifilter} command line option.
748 Alter jump instructions for long displacements.
750 @item -mcpu=[210|340]
751 Select the cpu type on the target hardware. This controls which instructions
755 Assemble for a big endian target.
758 Assemble for a little endian target.
764 * Manual:: Structure of this Manual
765 * GNU Assembler:: The GNU Assembler
766 * Object Formats:: Object File Formats
767 * Command Line:: Command Line
768 * Input Files:: Input Files
769 * Object:: Output (Object) File
770 * Errors:: Error and Warning Messages
774 @section Structure of this Manual
776 @cindex manual, structure and purpose
777 This manual is intended to describe what you need to know to use
778 @sc{gnu} @code{@value{AS}}. We cover the syntax expected in source files, including
779 notation for symbols, constants, and expressions; the directives that
780 @code{@value{AS}} understands; and of course how to invoke @code{@value{AS}}.
783 We also cover special features in the @value{TARGET}
784 configuration of @code{@value{AS}}, including assembler directives.
787 This manual also describes some of the machine-dependent features of
788 various flavors of the assembler.
791 @cindex machine instructions (not covered)
792 On the other hand, this manual is @emph{not} intended as an introduction
793 to programming in assembly language---let alone programming in general!
794 In a similar vein, we make no attempt to introduce the machine
795 architecture; we do @emph{not} describe the instruction set, standard
796 mnemonics, registers or addressing modes that are standard to a
797 particular architecture.
799 You may want to consult the manufacturer's
800 machine architecture manual for this information.
804 For information on the H8/300 machine instruction set, see @cite{H8/300
805 Series Programming Manual} (Hitachi ADE--602--025). For the H8/300H,
806 see @cite{H8/300H Series Programming Manual} (Hitachi).
809 For information on the H8/500 machine instruction set, see @cite{H8/500
810 Series Programming Manual} (Hitachi M21T001).
813 For information on the Hitachi SH machine instruction set, see
814 @cite{SH-Microcomputer User's Manual} (Hitachi Micro Systems, Inc.).
817 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
821 @c I think this is premature---doc@cygnus.com, 17jan1991
823 Throughout this manual, we assume that you are running @dfn{GNU},
824 the portable operating system from the @dfn{Free Software
825 Foundation, Inc.}. This restricts our attention to certain kinds of
826 computer (in particular, the kinds of computers that @sc{gnu} can run on);
827 once this assumption is granted examples and definitions need less
830 @code{@value{AS}} is part of a team of programs that turn a high-level
831 human-readable series of instructions into a low-level
832 computer-readable series of instructions. Different versions of
833 @code{@value{AS}} are used for different kinds of computer.
836 @c There used to be a section "Terminology" here, which defined
837 @c "contents", "byte", "word", and "long". Defining "word" to any
838 @c particular size is confusing when the .word directive may generate 16
839 @c bits on one machine and 32 bits on another; in general, for the user
840 @c version of this manual, none of these terms seem essential to define.
841 @c They were used very little even in the former draft of the manual;
842 @c this draft makes an effort to avoid them (except in names of
846 @section The GNU Assembler
848 @sc{gnu} @code{as} is really a family of assemblers.
850 This manual describes @code{@value{AS}}, a member of that family which is
851 configured for the @value{TARGET} architectures.
853 If you use (or have used) the @sc{gnu} assembler on one architecture, you
854 should find a fairly similar environment when you use it on another
855 architecture. Each version has much in common with the others,
856 including object file formats, most assembler directives (often called
857 @dfn{pseudo-ops}) and assembler syntax.@refill
859 @cindex purpose of @sc{gnu} assembler
860 @code{@value{AS}} is primarily intended to assemble the output of the
861 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
862 @code{@value{LD}}. Nevertheless, we've tried to make @code{@value{AS}}
863 assemble correctly everything that other assemblers for the same
864 machine would assemble.
866 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
869 @c This remark should appear in generic version of manual; assumption
870 @c here is that generic version sets M680x0.
871 This doesn't mean @code{@value{AS}} always uses the same syntax as another
872 assembler for the same architecture; for example, we know of several
873 incompatible versions of 680x0 assembly language syntax.
876 Unlike older assemblers, @code{@value{AS}} is designed to assemble a source
877 program in one pass of the source file. This has a subtle impact on the
878 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
881 @section Object File Formats
883 @cindex object file format
884 The @sc{gnu} assembler can be configured to produce several alternative
885 object file formats. For the most part, this does not affect how you
886 write assembly language programs; but directives for debugging symbols
887 are typically different in different file formats. @xref{Symbol
888 Attributes,,Symbol Attributes}.
891 On the @value{TARGET}, @code{@value{AS}} is configured to produce
892 @value{OBJ-NAME} format object files.
894 @c The following should exhaust all configs that set MULTI-OBJ, ideally
896 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
897 @code{a.out} or COFF format object files.
900 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
901 @code{b.out} or COFF format object files.
904 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
905 SOM or ELF format object files.
910 @section Command Line
912 @cindex command line conventions
913 After the program name @code{@value{AS}}, the command line may contain
914 options and file names. Options may appear in any order, and may be
915 before, after, or between file names. The order of file names is
918 @cindex standard input, as input file
920 @file{--} (two hyphens) by itself names the standard input file
921 explicitly, as one of the files for @code{@value{AS}} to assemble.
923 @cindex options, command line
924 Except for @samp{--} any command line argument that begins with a
925 hyphen (@samp{-}) is an option. Each option changes the behavior of
926 @code{@value{AS}}. No option changes the way another option works. An
927 option is a @samp{-} followed by one or more letters; the case of
928 the letter is important. All options are optional.
930 Some options expect exactly one file name to follow them. The file
931 name may either immediately follow the option's letter (compatible
932 with older assemblers) or it may be the next command argument (@sc{gnu}
933 standard). These two command lines are equivalent:
936 @value{AS} -o my-object-file.o mumble.s
937 @value{AS} -omy-object-file.o mumble.s
944 @cindex source program
946 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
947 describe the program input to one run of @code{@value{AS}}. The program may
948 be in one or more files; how the source is partitioned into files
949 doesn't change the meaning of the source.
951 @c I added "con" prefix to "catenation" just to prove I can overcome my
952 @c APL training... doc@cygnus.com
953 The source program is a concatenation of the text in all the files, in the
956 Each time you run @code{@value{AS}} it assembles exactly one source
957 program. The source program is made up of one or more files.
958 (The standard input is also a file.)
960 You give @code{@value{AS}} a command line that has zero or more input file
961 names. The input files are read (from left file name to right). A
962 command line argument (in any position) that has no special meaning
963 is taken to be an input file name.
965 If you give @code{@value{AS}} no file names it attempts to read one input file
966 from the @code{@value{AS}} standard input, which is normally your terminal. You
967 may have to type @key{ctl-D} to tell @code{@value{AS}} there is no more program
970 Use @samp{--} if you need to explicitly name the standard input file
971 in your command line.
973 If the source is empty, @code{@value{AS}} produces a small, empty object
976 @subheading Filenames and Line-numbers
978 @cindex input file linenumbers
979 @cindex line numbers, in input files
980 There are two ways of locating a line in the input file (or files) and
981 either may be used in reporting error messages. One way refers to a line
982 number in a physical file; the other refers to a line number in a
983 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
985 @dfn{Physical files} are those files named in the command line given
986 to @code{@value{AS}}.
988 @dfn{Logical files} are simply names declared explicitly by assembler
989 directives; they bear no relation to physical files. Logical file names help
990 error messages reflect the original source file, when @code{@value{AS}} source
991 is itself synthesized from other files. @code{@value{AS}} understands the
992 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
993 @ref{File,,@code{.file}}.
996 @section Output (Object) File
1002 Every time you run @code{@value{AS}} it produces an output file, which is
1003 your assembly language program translated into numbers. This file
1004 is the object file. Its default name is
1012 @code{b.out} when @code{@value{AS}} is configured for the Intel 80960.
1014 You can give it another name by using the @code{-o} option. Conventionally,
1015 object file names end with @file{.o}. The default name is used for historical
1016 reasons: older assemblers were capable of assembling self-contained programs
1017 directly into a runnable program. (For some formats, this isn't currently
1018 possible, but it can be done for the @code{a.out} format.)
1022 The object file is meant for input to the linker @code{@value{LD}}. It contains
1023 assembled program code, information to help @code{@value{LD}} integrate
1024 the assembled program into a runnable file, and (optionally) symbolic
1025 information for the debugger.
1027 @c link above to some info file(s) like the description of a.out.
1028 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1031 @section Error and Warning Messages
1033 @cindex error messsages
1034 @cindex warning messages
1035 @cindex messages from assembler
1036 @code{@value{AS}} may write warnings and error messages to the standard error
1037 file (usually your terminal). This should not happen when a compiler
1038 runs @code{@value{AS}} automatically. Warnings report an assumption made so
1039 that @code{@value{AS}} could keep assembling a flawed program; errors report a
1040 grave problem that stops the assembly.
1042 @cindex format of warning messages
1043 Warning messages have the format
1046 file_name:@b{NNN}:Warning Message Text
1050 @cindex line numbers, in warnings/errors
1051 (where @b{NNN} is a line number). If a logical file name has been given
1052 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1053 the current input file is used. If a logical line number was given
1055 (@pxref{Line,,@code{.line}})
1059 (@pxref{Line,,@code{.line}})
1062 (@pxref{Ln,,@code{.ln}})
1065 then it is used to calculate the number printed,
1066 otherwise the actual line in the current source file is printed. The
1067 message text is intended to be self explanatory (in the grand Unix
1070 @cindex format of error messages
1071 Error messages have the format
1073 file_name:@b{NNN}:FATAL:Error Message Text
1075 The file name and line number are derived as for warning
1076 messages. The actual message text may be rather less explanatory
1077 because many of them aren't supposed to happen.
1080 @chapter Command-Line Options
1082 @cindex options, all versions of assembler
1083 This chapter describes command-line options available in @emph{all}
1084 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
1086 to the @value{TARGET}.
1089 to particular machine architectures.
1092 If you are invoking @code{@value{AS}} via the @sc{gnu} C compiler (version 2),
1093 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1094 The assembler arguments must be separated from each other (and the @samp{-Wa})
1095 by commas. For example:
1098 gcc -c -g -O -Wa,-alh,-L file.c
1102 This passes two options to the assembler: @samp{-alh} (emit a listing to
1103 standard output with with high-level and assembly source) and @samp{-L} (retain
1104 local symbols in the symbol table).
1106 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1107 command-line options are automatically passed to the assembler by the compiler.
1108 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1109 precisely what options it passes to each compilation pass, including the
1113 * a:: -a[cdhlns] enable listings
1114 * D:: -D for compatibility
1115 * f:: -f to work faster
1116 * I:: -I for .include search path
1117 @ifclear DIFF-TBL-KLUGE
1118 * K:: -K for compatibility
1120 @ifset DIFF-TBL-KLUGE
1121 * K:: -K for difference tables
1124 * L:: -L to retain local labels
1125 * M:: -M or --mri to assemble in MRI compatibility mode
1126 * MD:: --MD for dependency tracking
1127 * o:: -o to name the object file
1128 * R:: -R to join data and text sections
1129 * statistics:: --statistics to see statistics about assembly
1130 * traditional-format:: --traditional-format for compatible output
1131 * v:: -v to announce version
1132 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1133 * Z:: -Z to make object file even after errors
1137 @section Enable Listings: @code{-a[cdhlns]}
1146 @cindex listings, enabling
1147 @cindex assembly listings, enabling
1149 These options enable listing output from the assembler. By itself,
1150 @samp{-a} requests high-level, assembly, and symbols listing.
1151 You can use other letters to select specific options for the list:
1152 @samp{-ah} requests a high-level language listing,
1153 @samp{-al} requests an output-program assembly listing, and
1154 @samp{-as} requests a symbol table listing.
1155 High-level listings require that a compiler debugging option like
1156 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1159 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1160 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1161 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1162 omitted from the listing.
1164 Use the @samp{-ad} option to omit debugging directives from the
1167 Once you have specified one of these options, you can further control
1168 listing output and its appearance using the directives @code{.list},
1169 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1171 The @samp{-an} option turns off all forms processing.
1172 If you do not request listing output with one of the @samp{-a} options, the
1173 listing-control directives have no effect.
1175 The letters after @samp{-a} may be combined into one option,
1176 @emph{e.g.}, @samp{-aln}.
1182 This option has no effect whatsoever, but it is accepted to make it more
1183 likely that scripts written for other assemblers also work with
1187 @section Work Faster: @code{-f}
1190 @cindex trusted compiler
1191 @cindex faster processing (@code{-f})
1192 @samp{-f} should only be used when assembling programs written by a
1193 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1194 and comment preprocessing on
1195 the input file(s) before assembling them. @xref{Preprocessing,
1199 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1200 preprocessed (if they contain comments, for example), @code{@value{AS}} does
1205 @section @code{.include} search path: @code{-I} @var{path}
1207 @kindex -I @var{path}
1208 @cindex paths for @code{.include}
1209 @cindex search path for @code{.include}
1210 @cindex @code{include} directive search path
1211 Use this option to add a @var{path} to the list of directories
1212 @code{@value{AS}} searches for files specified in @code{.include}
1213 directives (@pxref{Include,,@code{.include}}). You may use @code{-I} as
1214 many times as necessary to include a variety of paths. The current
1215 working directory is always searched first; after that, @code{@value{AS}}
1216 searches any @samp{-I} directories in the same order as they were
1217 specified (left to right) on the command line.
1220 @section Difference Tables: @code{-K}
1223 @ifclear DIFF-TBL-KLUGE
1224 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1225 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1226 where it can be used to warn when the assembler alters the machine code
1227 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1228 family does not have the addressing limitations that sometimes lead to this
1229 alteration on other platforms.
1232 @ifset DIFF-TBL-KLUGE
1233 @cindex difference tables, warning
1234 @cindex warning for altered difference tables
1235 @code{@value{AS}} sometimes alters the code emitted for directives of the form
1236 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1237 You can use the @samp{-K} option if you want a warning issued when this
1242 @section Include Local Labels: @code{-L}
1245 @cindex local labels, retaining in output
1246 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1247 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1248 debugging, because they are intended for the use of programs (like
1249 compilers) that compose assembler programs, not for your notice.
1250 Normally both @code{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1251 normally debug with them.
1253 This option tells @code{@value{AS}} to retain those @samp{L@dots{}} symbols
1254 in the object file. Usually if you do this you also tell the linker
1255 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1257 By default, a local label is any label beginning with @samp{L}, but each
1258 target is allowed to redefine the local label prefix.
1260 On the HPPA local labels begin with @samp{L$}.
1264 @section Assemble in MRI Compatibility Mode: @code{-M}
1267 @cindex MRI compatibility mode
1268 The @code{-M} or @code{--mri} option selects MRI compatibility mode. This
1269 changes the syntax and pseudo-op handling of @code{@value{AS}} to make it
1270 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1271 configured target) assembler from Microtec Research. The exact nature of the
1272 MRI syntax will not be documented here; see the MRI manuals for more
1273 information. Note in particular that the handling of macros and macro
1274 arguments is somewhat different. The purpose of this option is to permit
1275 assembling existing MRI assembler code using @code{@value{AS}}.
1277 The MRI compatibility is not complete. Certain operations of the MRI assembler
1278 depend upon its object file format, and can not be supported using other object
1279 file formats. Supporting these would require enhancing each object file format
1280 individually. These are:
1283 @item global symbols in common section
1285 The m68k MRI assembler supports common sections which are merged by the linker.
1286 Other object file formats do not support this. @code{@value{AS}} handles
1287 common sections by treating them as a single common symbol. It permits local
1288 symbols to be defined within a common section, but it can not support global
1289 symbols, since it has no way to describe them.
1291 @item complex relocations
1293 The MRI assemblers support relocations against a negated section address, and
1294 relocations which combine the start addresses of two or more sections. These
1295 are not support by other object file formats.
1297 @item @code{END} pseudo-op specifying start address
1299 The MRI @code{END} pseudo-op permits the specification of a start address.
1300 This is not supported by other object file formats. The start address may
1301 instead be specified using the @code{-e} option to the linker, or in a linker
1304 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1306 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1307 name to the output file. This is not supported by other object file formats.
1309 @item @code{ORG} pseudo-op
1311 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1312 address. This differs from the usual @code{@value{AS}} @code{.org} pseudo-op,
1313 which changes the location within the current section. Absolute sections are
1314 not supported by other object file formats. The address of a section may be
1315 assigned within a linker script.
1318 There are some other features of the MRI assembler which are not supported by
1319 @code{@value{AS}}, typically either because they are difficult or because they
1320 seem of little consequence. Some of these may be supported in future releases.
1324 @item EBCDIC strings
1326 EBCDIC strings are not supported.
1328 @item packed binary coded decimal
1330 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1331 and @code{DCB.P} pseudo-ops are not supported.
1333 @item @code{FEQU} pseudo-op
1335 The m68k @code{FEQU} pseudo-op is not supported.
1337 @item @code{NOOBJ} pseudo-op
1339 The m68k @code{NOOBJ} pseudo-op is not supported.
1341 @item @code{OPT} branch control options
1343 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1344 @code{BRL}, and @code{BRW}---are ignored. @code{@value{AS}} automatically
1345 relaxes all branches, whether forward or backward, to an appropriate size, so
1346 these options serve no purpose.
1348 @item @code{OPT} list control options
1350 The following m68k @code{OPT} list control options are ignored: @code{C},
1351 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1352 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1354 @item other @code{OPT} options
1356 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1357 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1359 @item @code{OPT} @code{D} option is default
1361 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1362 @code{OPT NOD} may be used to turn it off.
1364 @item @code{XREF} pseudo-op.
1366 The m68k @code{XREF} pseudo-op is ignored.
1368 @item @code{.debug} pseudo-op
1370 The i960 @code{.debug} pseudo-op is not supported.
1372 @item @code{.extended} pseudo-op
1374 The i960 @code{.extended} pseudo-op is not supported.
1376 @item @code{.list} pseudo-op.
1378 The various options of the i960 @code{.list} pseudo-op are not supported.
1380 @item @code{.optimize} pseudo-op
1382 The i960 @code{.optimize} pseudo-op is not supported.
1384 @item @code{.output} pseudo-op
1386 The i960 @code{.output} pseudo-op is not supported.
1388 @item @code{.setreal} pseudo-op
1390 The i960 @code{.setreal} pseudo-op is not supported.
1395 @section Dependency tracking: @code{--MD}
1398 @cindex dependency tracking
1401 @code{@value{AS}} can generate a dependency file for the file it creates. This
1402 file consists of a single rule suitable for @code{make} describing the
1403 dependencies of the main source file.
1405 The rule is written to the file named in its argument.
1407 This feature is used in the automatic updating of makefiles.
1410 @section Name the Object File: @code{-o}
1413 @cindex naming object file
1414 @cindex object file name
1415 There is always one object file output when you run @code{@value{AS}}. By
1416 default it has the name
1419 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1433 You use this option (which takes exactly one filename) to give the
1434 object file a different name.
1436 Whatever the object file is called, @code{@value{AS}} overwrites any
1437 existing file of the same name.
1440 @section Join Data and Text Sections: @code{-R}
1443 @cindex data and text sections, joining
1444 @cindex text and data sections, joining
1445 @cindex joining text and data sections
1446 @cindex merging text and data sections
1447 @code{-R} tells @code{@value{AS}} to write the object file as if all
1448 data-section data lives in the text section. This is only done at
1449 the very last moment: your binary data are the same, but data
1450 section parts are relocated differently. The data section part of
1451 your object file is zero bytes long because all its bytes are
1452 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1454 When you specify @code{-R} it would be possible to generate shorter
1455 address displacements (because we do not have to cross between text and
1456 data section). We refrain from doing this simply for compatibility with
1457 older versions of @code{@value{AS}}. In future, @code{-R} may work this way.
1460 When @code{@value{AS}} is configured for COFF output,
1461 this option is only useful if you use sections named @samp{.text} and
1466 @code{-R} is not supported for any of the HPPA targets. Using
1467 @code{-R} generates a warning from @code{@value{AS}}.
1471 @section Display Assembly Statistics: @code{--statistics}
1473 @kindex --statistics
1474 @cindex statistics, about assembly
1475 @cindex time, total for assembly
1476 @cindex space used, maximum for assembly
1477 Use @samp{--statistics} to display two statistics about the resources used by
1478 @code{@value{AS}}: the maximum amount of space allocated during the assembly
1479 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1482 @node traditional-format
1483 @section Compatible output: @code{--traditional-format}
1485 @kindex --traditional-format
1486 For some targets, the output of @code{@value{AS}} is different in some ways
1487 from the output of some existing assembler. This switch requests
1488 @code{@value{AS}} to use the traditional format instead.
1490 For example, it disables the exception frame optimizations which
1491 @code{@value{AS}} normally does by default on @code{@value{GCC}} output.
1494 @section Announce Version: @code{-v}
1498 @cindex assembler version
1499 @cindex version of assembler
1500 You can find out what version of as is running by including the
1501 option @samp{-v} (which you can also spell as @samp{-version}) on the
1505 @section Control Warnings: @code{-W}, @code{--warn}, @code{--no-warn}, @code{--fatal-warnings}
1507 @code{@value{AS}} should never give a warning or error message when
1508 assembling compiler output. But programs written by people often
1509 cause @code{@value{AS}} to give a warning that a particular assumption was
1510 made. All such warnings are directed to the standard error file.
1513 @kindex @samp{--no-warn}
1514 @cindex suppressing warnings
1515 @cindex warnings, suppressing
1516 If you use the @code{-W} and @code{--no-warn} options, no warnings are issued.
1517 This only affects the warning messages: it does not change any particular of
1518 how @code{@value{AS}} assembles your file. Errors, which stop the assembly,
1521 @kindex @samp{--fatal-warnings}
1522 @cindex errors, caused by warnings
1523 @cindex warnings, causing error
1524 If you use the @code{--fatal-warnings} option, @code{@value{AS}} considers
1525 files that generate warnings to be in error.
1527 @kindex @samp{--warn}
1528 @cindex warnings, switching on
1529 You can switch these options off again by specifying @code{--warn}, which
1530 causes warnings to be output as usual.
1533 @section Generate Object File in Spite of Errors: @code{-Z}
1534 @cindex object file, after errors
1535 @cindex errors, continuing after
1536 After an error message, @code{@value{AS}} normally produces no output. If for
1537 some reason you are interested in object file output even after
1538 @code{@value{AS}} gives an error message on your program, use the @samp{-Z}
1539 option. If there are any errors, @code{@value{AS}} continues anyways, and
1540 writes an object file after a final warning message of the form @samp{@var{n}
1541 errors, @var{m} warnings, generating bad object file.}
1546 @cindex machine-independent syntax
1547 @cindex syntax, machine-independent
1548 This chapter describes the machine-independent syntax allowed in a
1549 source file. @code{@value{AS}} syntax is similar to what many other
1550 assemblers use; it is inspired by the BSD 4.2
1555 assembler, except that @code{@value{AS}} does not assemble Vax bit-fields.
1559 * Preprocessing:: Preprocessing
1560 * Whitespace:: Whitespace
1561 * Comments:: Comments
1562 * Symbol Intro:: Symbols
1563 * Statements:: Statements
1564 * Constants:: Constants
1568 @section Preprocessing
1570 @cindex preprocessing
1571 The @code{@value{AS}} internal preprocessor:
1573 @cindex whitespace, removed by preprocessor
1575 adjusts and removes extra whitespace. It leaves one space or tab before
1576 the keywords on a line, and turns any other whitespace on the line into
1579 @cindex comments, removed by preprocessor
1581 removes all comments, replacing them with a single space, or an
1582 appropriate number of newlines.
1584 @cindex constants, converted by preprocessor
1586 converts character constants into the appropriate numeric values.
1589 It does not do macro processing, include file handling, or
1590 anything else you may get from your C compiler's preprocessor. You can
1591 do include file processing with the @code{.include} directive
1592 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
1593 to get other ``CPP'' style preprocessing, by giving the input file a
1594 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
1595 Output, gcc.info, Using GNU CC}.
1597 Excess whitespace, comments, and character constants
1598 cannot be used in the portions of the input text that are not
1601 @cindex turning preprocessing on and off
1602 @cindex preprocessing, turning on and off
1605 If the first line of an input file is @code{#NO_APP} or if you use the
1606 @samp{-f} option, whitespace and comments are not removed from the input file.
1607 Within an input file, you can ask for whitespace and comment removal in
1608 specific portions of the by putting a line that says @code{#APP} before the
1609 text that may contain whitespace or comments, and putting a line that says
1610 @code{#NO_APP} after this text. This feature is mainly intend to support
1611 @code{asm} statements in compilers whose output is otherwise free of comments
1618 @dfn{Whitespace} is one or more blanks or tabs, in any order.
1619 Whitespace is used to separate symbols, and to make programs neater for
1620 people to read. Unless within character constants
1621 (@pxref{Characters,,Character Constants}), any whitespace means the same
1622 as exactly one space.
1628 There are two ways of rendering comments to @code{@value{AS}}. In both
1629 cases the comment is equivalent to one space.
1631 Anything from @samp{/*} through the next @samp{*/} is a comment.
1632 This means you may not nest these comments.
1636 The only way to include a newline ('\n') in a comment
1637 is to use this sort of comment.
1640 /* This sort of comment does not nest. */
1643 @cindex line comment character
1644 Anything from the @dfn{line comment} character to the next newline
1645 is considered a comment and is ignored. The line comment character is
1647 @samp{;} for the AMD 29K family;
1650 @samp{;} on the ARC;
1653 @samp{@@} on the ARM;
1656 @samp{;} for the H8/300 family;
1659 @samp{!} for the H8/500 family;
1662 @samp{;} for the HPPA;
1665 @samp{#} on the i386 and x86-64;
1668 @samp{#} on the i960;
1671 @samp{;} for picoJava;
1674 @samp{!} for the Hitachi SH;
1677 @samp{!} on the SPARC;
1680 @samp{#} on the m32r;
1683 @samp{|} on the 680x0;
1686 @samp{#} on the 68HC11 and 68HC12;
1689 @samp{#} on the Vax;
1692 @samp{!} for the Z8000;
1695 @samp{#} on the V850;
1697 see @ref{Machine Dependencies}. @refill
1698 @c FIXME What about m88k, i860?
1701 On some machines there are two different line comment characters. One
1702 character only begins a comment if it is the first non-whitespace character on
1703 a line, while the other always begins a comment.
1707 The V850 assembler also supports a double dash as starting a comment that
1708 extends to the end of the line.
1714 @cindex lines starting with @code{#}
1715 @cindex logical line numbers
1716 To be compatible with past assemblers, lines that begin with @samp{#} have a
1717 special interpretation. Following the @samp{#} should be an absolute
1718 expression (@pxref{Expressions}): the logical line number of the @emph{next}
1719 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
1720 new logical file name. The rest of the line, if any, should be whitespace.
1722 If the first non-whitespace characters on the line are not numeric,
1723 the line is ignored. (Just like a comment.)
1726 # This is an ordinary comment.
1727 # 42-6 "new_file_name" # New logical file name
1728 # This is logical line # 36.
1730 This feature is deprecated, and may disappear from future versions
1731 of @code{@value{AS}}.
1736 @cindex characters used in symbols
1737 @ifclear SPECIAL-SYMS
1738 A @dfn{symbol} is one or more characters chosen from the set of all
1739 letters (both upper and lower case), digits and the three characters
1745 A @dfn{symbol} is one or more characters chosen from the set of all
1746 letters (both upper and lower case), digits and the three characters
1747 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
1753 On most machines, you can also use @code{$} in symbol names; exceptions
1754 are noted in @ref{Machine Dependencies}.
1756 No symbol may begin with a digit. Case is significant.
1757 There is no length limit: all characters are significant. Symbols are
1758 delimited by characters not in that set, or by the beginning of a file
1759 (since the source program must end with a newline, the end of a file is
1760 not a possible symbol delimiter). @xref{Symbols}.
1761 @cindex length of symbols
1766 @cindex statements, structure of
1767 @cindex line separator character
1768 @cindex statement separator character
1770 @ifclear abnormal-separator
1771 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
1772 semicolon (@samp{;}). The newline or semicolon is considered part of
1773 the preceding statement. Newlines and semicolons within character
1774 constants are an exception: they do not end statements.
1776 @ifset abnormal-separator
1778 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
1779 sign (@samp{@@}). The newline or at sign is considered part of the
1780 preceding statement. Newlines and at signs within character constants
1781 are an exception: they do not end statements.
1784 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
1785 point (@samp{!}). The newline or exclamation point is considered part of the
1786 preceding statement. Newlines and exclamation points within character
1787 constants are an exception: they do not end statements.
1790 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
1791 H8/300) a dollar sign (@samp{$}); or (for the
1794 (@samp{;}). The newline or separator character is considered part of
1795 the preceding statement. Newlines and separators within character
1796 constants are an exception: they do not end statements.
1801 A @dfn{statement} ends at a newline character (@samp{\n}) or line
1802 separator character. (The line separator is usually @samp{;}, unless
1803 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
1804 newline or separator character is considered part of the preceding
1805 statement. Newlines and separators within character constants are an
1806 exception: they do not end statements.
1809 @cindex newline, required at file end
1810 @cindex EOF, newline must precede
1811 It is an error to end any statement with end-of-file: the last
1812 character of any input file should be a newline.@refill
1814 An empty statement is allowed, and may include whitespace. It is ignored.
1816 @cindex instructions and directives
1817 @cindex directives and instructions
1818 @c "key symbol" is not used elsewhere in the document; seems pedantic to
1819 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
1821 A statement begins with zero or more labels, optionally followed by a
1822 key symbol which determines what kind of statement it is. The key
1823 symbol determines the syntax of the rest of the statement. If the
1824 symbol begins with a dot @samp{.} then the statement is an assembler
1825 directive: typically valid for any computer. If the symbol begins with
1826 a letter the statement is an assembly language @dfn{instruction}: it
1827 assembles into a machine language instruction.
1829 Different versions of @code{@value{AS}} for different computers
1830 recognize different instructions. In fact, the same symbol may
1831 represent a different instruction in a different computer's assembly
1835 @cindex @code{:} (label)
1836 @cindex label (@code{:})
1837 A label is a symbol immediately followed by a colon (@code{:}).
1838 Whitespace before a label or after a colon is permitted, but you may not
1839 have whitespace between a label's symbol and its colon. @xref{Labels}.
1842 For HPPA targets, labels need not be immediately followed by a colon, but
1843 the definition of a label must begin in column zero. This also implies that
1844 only one label may be defined on each line.
1848 label: .directive followed by something
1849 another_label: # This is an empty statement.
1850 instruction operand_1, operand_2, @dots{}
1857 A constant is a number, written so that its value is known by
1858 inspection, without knowing any context. Like this:
1861 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
1862 .ascii "Ring the bell\7" # A string constant.
1863 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
1864 .float 0f-314159265358979323846264338327\
1865 95028841971.693993751E-40 # - pi, a flonum.
1870 * Characters:: Character Constants
1871 * Numbers:: Number Constants
1875 @subsection Character Constants
1877 @cindex character constants
1878 @cindex constants, character
1879 There are two kinds of character constants. A @dfn{character} stands
1880 for one character in one byte and its value may be used in
1881 numeric expressions. String constants (properly called string
1882 @emph{literals}) are potentially many bytes and their values may not be
1883 used in arithmetic expressions.
1887 * Chars:: Characters
1891 @subsubsection Strings
1893 @cindex string constants
1894 @cindex constants, string
1895 A @dfn{string} is written between double-quotes. It may contain
1896 double-quotes or null characters. The way to get special characters
1897 into a string is to @dfn{escape} these characters: precede them with
1898 a backslash @samp{\} character. For example @samp{\\} represents
1899 one backslash: the first @code{\} is an escape which tells
1900 @code{@value{AS}} to interpret the second character literally as a backslash
1901 (which prevents @code{@value{AS}} from recognizing the second @code{\} as an
1902 escape character). The complete list of escapes follows.
1904 @cindex escape codes, character
1905 @cindex character escape codes
1908 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
1910 @cindex @code{\b} (backspace character)
1911 @cindex backspace (@code{\b})
1913 Mnemonic for backspace; for ASCII this is octal code 010.
1916 @c Mnemonic for EOText; for ASCII this is octal code 004.
1918 @cindex @code{\f} (formfeed character)
1919 @cindex formfeed (@code{\f})
1921 Mnemonic for FormFeed; for ASCII this is octal code 014.
1923 @cindex @code{\n} (newline character)
1924 @cindex newline (@code{\n})
1926 Mnemonic for newline; for ASCII this is octal code 012.
1929 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
1931 @cindex @code{\r} (carriage return character)
1932 @cindex carriage return (@code{\r})
1934 Mnemonic for carriage-Return; for ASCII this is octal code 015.
1937 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
1938 @c other assemblers.
1940 @cindex @code{\t} (tab)
1941 @cindex tab (@code{\t})
1943 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
1946 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
1947 @c @item \x @var{digit} @var{digit} @var{digit}
1948 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
1950 @cindex @code{\@var{ddd}} (octal character code)
1951 @cindex octal character code (@code{\@var{ddd}})
1952 @item \ @var{digit} @var{digit} @var{digit}
1953 An octal character code. The numeric code is 3 octal digits.
1954 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
1955 for example, @code{\008} has the value 010, and @code{\009} the value 011.
1957 @cindex @code{\@var{xd...}} (hex character code)
1958 @cindex hex character code (@code{\@var{xd...}})
1959 @item \@code{x} @var{hex-digits...}
1960 A hex character code. All trailing hex digits are combined. Either upper or
1961 lower case @code{x} works.
1963 @cindex @code{\\} (@samp{\} character)
1964 @cindex backslash (@code{\\})
1966 Represents one @samp{\} character.
1969 @c Represents one @samp{'} (accent acute) character.
1970 @c This is needed in single character literals
1971 @c (@xref{Characters,,Character Constants}.) to represent
1974 @cindex @code{\"} (doublequote character)
1975 @cindex doublequote (@code{\"})
1977 Represents one @samp{"} character. Needed in strings to represent
1978 this character, because an unescaped @samp{"} would end the string.
1980 @item \ @var{anything-else}
1981 Any other character when escaped by @kbd{\} gives a warning, but
1982 assembles as if the @samp{\} was not present. The idea is that if
1983 you used an escape sequence you clearly didn't want the literal
1984 interpretation of the following character. However @code{@value{AS}} has no
1985 other interpretation, so @code{@value{AS}} knows it is giving you the wrong
1986 code and warns you of the fact.
1989 Which characters are escapable, and what those escapes represent,
1990 varies widely among assemblers. The current set is what we think
1991 the BSD 4.2 assembler recognizes, and is a subset of what most C
1992 compilers recognize. If you are in doubt, do not use an escape
1996 @subsubsection Characters
1998 @cindex single character constant
1999 @cindex character, single
2000 @cindex constant, single character
2001 A single character may be written as a single quote immediately
2002 followed by that character. The same escapes apply to characters as
2003 to strings. So if you want to write the character backslash, you
2004 must write @kbd{'\\} where the first @code{\} escapes the second
2005 @code{\}. As you can see, the quote is an acute accent, not a
2006 grave accent. A newline
2008 @ifclear abnormal-separator
2009 (or semicolon @samp{;})
2011 @ifset abnormal-separator
2013 (or at sign @samp{@@})
2016 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2022 immediately following an acute accent is taken as a literal character
2023 and does not count as the end of a statement. The value of a character
2024 constant in a numeric expression is the machine's byte-wide code for
2025 that character. @code{@value{AS}} assumes your character code is ASCII:
2026 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2029 @subsection Number Constants
2031 @cindex constants, number
2032 @cindex number constants
2033 @code{@value{AS}} distinguishes three kinds of numbers according to how they
2034 are stored in the target machine. @emph{Integers} are numbers that
2035 would fit into an @code{int} in the C language. @emph{Bignums} are
2036 integers, but they are stored in more than 32 bits. @emph{Flonums}
2037 are floating point numbers, described below.
2040 * Integers:: Integers
2045 * Bit Fields:: Bit Fields
2051 @subsubsection Integers
2053 @cindex constants, integer
2055 @cindex binary integers
2056 @cindex integers, binary
2057 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2058 the binary digits @samp{01}.
2060 @cindex octal integers
2061 @cindex integers, octal
2062 An octal integer is @samp{0} followed by zero or more of the octal
2063 digits (@samp{01234567}).
2065 @cindex decimal integers
2066 @cindex integers, decimal
2067 A decimal integer starts with a non-zero digit followed by zero or
2068 more digits (@samp{0123456789}).
2070 @cindex hexadecimal integers
2071 @cindex integers, hexadecimal
2072 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2073 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2075 Integers have the usual values. To denote a negative integer, use
2076 the prefix operator @samp{-} discussed under expressions
2077 (@pxref{Prefix Ops,,Prefix Operators}).
2080 @subsubsection Bignums
2083 @cindex constants, bignum
2084 A @dfn{bignum} has the same syntax and semantics as an integer
2085 except that the number (or its negative) takes more than 32 bits to
2086 represent in binary. The distinction is made because in some places
2087 integers are permitted while bignums are not.
2090 @subsubsection Flonums
2092 @cindex floating point numbers
2093 @cindex constants, floating point
2095 @cindex precision, floating point
2096 A @dfn{flonum} represents a floating point number. The translation is
2097 indirect: a decimal floating point number from the text is converted by
2098 @code{@value{AS}} to a generic binary floating point number of more than
2099 sufficient precision. This generic floating point number is converted
2100 to a particular computer's floating point format (or formats) by a
2101 portion of @code{@value{AS}} specialized to that computer.
2103 A flonum is written by writing (in order)
2108 (@samp{0} is optional on the HPPA.)
2112 A letter, to tell @code{@value{AS}} the rest of the number is a flonum.
2114 @kbd{e} is recommended. Case is not important.
2116 @c FIXME: verify if flonum syntax really this vague for most cases
2117 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2118 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2121 On the H8/300, H8/500,
2123 and AMD 29K architectures, the letter must be
2124 one of the letters @samp{DFPRSX} (in upper or lower case).
2126 On the ARC, the letter must be one of the letters @samp{DFRS}
2127 (in upper or lower case).
2129 On the Intel 960 architecture, the letter must be
2130 one of the letters @samp{DFT} (in upper or lower case).
2132 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2136 One of the letters @samp{DFPRSX} (in upper or lower case).
2139 One of the letters @samp{DFRS} (in upper or lower case).
2142 One of the letters @samp{DFPRSX} (in upper or lower case).
2145 The letter @samp{E} (upper case only).
2148 One of the letters @samp{DFT} (in upper or lower case).
2153 An optional sign: either @samp{+} or @samp{-}.
2156 An optional @dfn{integer part}: zero or more decimal digits.
2159 An optional @dfn{fractional part}: @samp{.} followed by zero
2160 or more decimal digits.
2163 An optional exponent, consisting of:
2167 An @samp{E} or @samp{e}.
2168 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2169 @c principle this can perfectly well be different on different targets.
2171 Optional sign: either @samp{+} or @samp{-}.
2173 One or more decimal digits.
2178 At least one of the integer part or the fractional part must be
2179 present. The floating point number has the usual base-10 value.
2181 @code{@value{AS}} does all processing using integers. Flonums are computed
2182 independently of any floating point hardware in the computer running
2187 @c Bit fields are written as a general facility but are also controlled
2188 @c by a conditional-compilation flag---which is as of now (21mar91)
2189 @c turned on only by the i960 config of GAS.
2191 @subsubsection Bit Fields
2194 @cindex constants, bit field
2195 You can also define numeric constants as @dfn{bit fields}.
2196 specify two numbers separated by a colon---
2198 @var{mask}:@var{value}
2201 @code{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2204 The resulting number is then packed
2206 @c this conditional paren in case bit fields turned on elsewhere than 960
2207 (in host-dependent byte order)
2209 into a field whose width depends on which assembler directive has the
2210 bit-field as its argument. Overflow (a result from the bitwise and
2211 requiring more binary digits to represent) is not an error; instead,
2212 more constants are generated, of the specified width, beginning with the
2213 least significant digits.@refill
2215 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2216 @code{.short}, and @code{.word} accept bit-field arguments.
2221 @chapter Sections and Relocation
2226 * Secs Background:: Background
2227 * Ld Sections:: Linker Sections
2228 * As Sections:: Assembler Internal Sections
2229 * Sub-Sections:: Sub-Sections
2233 @node Secs Background
2236 Roughly, a section is a range of addresses, with no gaps; all data
2237 ``in'' those addresses is treated the same for some particular purpose.
2238 For example there may be a ``read only'' section.
2240 @cindex linker, and assembler
2241 @cindex assembler, and linker
2242 The linker @code{@value{LD}} reads many object files (partial programs) and
2243 combines their contents to form a runnable program. When @code{@value{AS}}
2244 emits an object file, the partial program is assumed to start at address 0.
2245 @code{@value{LD}} assigns the final addresses for the partial program, so that
2246 different partial programs do not overlap. This is actually an
2247 oversimplification, but it suffices to explain how @code{@value{AS}} uses
2250 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2251 addresses. These blocks slide to their run-time addresses as rigid
2252 units; their length does not change and neither does the order of bytes
2253 within them. Such a rigid unit is called a @emph{section}. Assigning
2254 run-time addresses to sections is called @dfn{relocation}. It includes
2255 the task of adjusting mentions of object-file addresses so they refer to
2256 the proper run-time addresses.
2258 For the H8/300 and H8/500,
2259 and for the Hitachi SH,
2260 @code{@value{AS}} pads sections if needed to
2261 ensure they end on a word (sixteen bit) boundary.
2264 @cindex standard assembler sections
2265 An object file written by @code{@value{AS}} has at least three sections, any
2266 of which may be empty. These are named @dfn{text}, @dfn{data} and
2271 When it generates COFF output,
2273 @code{@value{AS}} can also generate whatever other named sections you specify
2274 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2275 If you do not use any directives that place output in the @samp{.text}
2276 or @samp{.data} sections, these sections still exist, but are empty.
2281 When @code{@value{AS}} generates SOM or ELF output for the HPPA,
2283 @code{@value{AS}} can also generate whatever other named sections you
2284 specify using the @samp{.space} and @samp{.subspace} directives. See
2285 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2286 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2287 assembler directives.
2290 Additionally, @code{@value{AS}} uses different names for the standard
2291 text, data, and bss sections when generating SOM output. Program text
2292 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2293 BSS into @samp{$BSS$}.
2297 Within the object file, the text section starts at address @code{0}, the
2298 data section follows, and the bss section follows the data section.
2301 When generating either SOM or ELF output files on the HPPA, the text
2302 section starts at address @code{0}, the data section at address
2303 @code{0x4000000}, and the bss section follows the data section.
2306 To let @code{@value{LD}} know which data changes when the sections are
2307 relocated, and how to change that data, @code{@value{AS}} also writes to the
2308 object file details of the relocation needed. To perform relocation
2309 @code{@value{LD}} must know, each time an address in the object
2313 Where in the object file is the beginning of this reference to
2316 How long (in bytes) is this reference?
2318 Which section does the address refer to? What is the numeric value of
2320 (@var{address}) @minus{} (@var{start-address of section})?
2323 Is the reference to an address ``Program-Counter relative''?
2326 @cindex addresses, format of
2327 @cindex section-relative addressing
2328 In fact, every address @code{@value{AS}} ever uses is expressed as
2330 (@var{section}) + (@var{offset into section})
2333 Further, most expressions @code{@value{AS}} computes have this section-relative
2336 (For some object formats, such as SOM for the HPPA, some expressions are
2337 symbol-relative instead.)
2340 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2341 @var{N} into section @var{secname}.''
2343 Apart from text, data and bss sections you need to know about the
2344 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2345 addresses in the absolute section remain unchanged. For example, address
2346 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2347 @code{@value{LD}}. Although the linker never arranges two partial programs'
2348 data sections with overlapping addresses after linking, @emph{by definition}
2349 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2350 part of a program is always the same address when the program is running as
2351 address @code{@{absolute@ 239@}} in any other part of the program.
2353 The idea of sections is extended to the @dfn{undefined} section. Any
2354 address whose section is unknown at assembly time is by definition
2355 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2356 Since numbers are always defined, the only way to generate an undefined
2357 address is to mention an undefined symbol. A reference to a named
2358 common block would be such a symbol: its value is unknown at assembly
2359 time so it has section @emph{undefined}.
2361 By analogy the word @emph{section} is used to describe groups of sections in
2362 the linked program. @code{@value{LD}} puts all partial programs' text
2363 sections in contiguous addresses in the linked program. It is
2364 customary to refer to the @emph{text section} of a program, meaning all
2365 the addresses of all partial programs' text sections. Likewise for
2366 data and bss sections.
2368 Some sections are manipulated by @code{@value{LD}}; others are invented for
2369 use of @code{@value{AS}} and have no meaning except during assembly.
2372 @section Linker Sections
2373 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2378 @cindex named sections
2379 @cindex sections, named
2380 @item named sections
2383 @cindex text section
2384 @cindex data section
2388 These sections hold your program. @code{@value{AS}} and @code{@value{LD}} treat them as
2389 separate but equal sections. Anything you can say of one section is
2392 When the program is running, however, it is
2393 customary for the text section to be unalterable. The
2394 text section is often shared among processes: it contains
2395 instructions, constants and the like. The data section of a running
2396 program is usually alterable: for example, C variables would be stored
2397 in the data section.
2402 This section contains zeroed bytes when your program begins running. It
2403 is used to hold unitialized variables or common storage. The length of
2404 each partial program's bss section is important, but because it starts
2405 out containing zeroed bytes there is no need to store explicit zero
2406 bytes in the object file. The bss section was invented to eliminate
2407 those explicit zeros from object files.
2409 @cindex absolute section
2410 @item absolute section
2411 Address 0 of this section is always ``relocated'' to runtime address 0.
2412 This is useful if you want to refer to an address that @code{@value{LD}} must
2413 not change when relocating. In this sense we speak of absolute
2414 addresses being ``unrelocatable'': they do not change during relocation.
2416 @cindex undefined section
2417 @item undefined section
2418 This ``section'' is a catch-all for address references to objects not in
2419 the preceding sections.
2420 @c FIXME: ref to some other doc on obj-file formats could go here.
2423 @cindex relocation example
2424 An idealized example of three relocatable sections follows.
2426 The example uses the traditional section names @samp{.text} and @samp{.data}.
2428 Memory addresses are on the horizontal axis.
2432 @c END TEXI2ROFF-KILL
2435 partial program # 1: |ttttt|dddd|00|
2442 partial program # 2: |TTT|DDD|000|
2445 +--+---+-----+--+----+---+-----+~~
2446 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2447 +--+---+-----+--+----+---+-----+~~
2449 addresses: 0 @dots{}
2456 \line{\it Partial program \#1: \hfil}
2457 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2458 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2460 \line{\it Partial program \#2: \hfil}
2461 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2462 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2464 \line{\it linked program: \hfil}
2465 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2466 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2467 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2468 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2470 \line{\it addresses: \hfil}
2474 @c END TEXI2ROFF-KILL
2477 @section Assembler Internal Sections
2479 @cindex internal assembler sections
2480 @cindex sections in messages, internal
2481 These sections are meant only for the internal use of @code{@value{AS}}. They
2482 have no meaning at run-time. You do not really need to know about these
2483 sections for most purposes; but they can be mentioned in @code{@value{AS}}
2484 warning messages, so it might be helpful to have an idea of their
2485 meanings to @code{@value{AS}}. These sections are used to permit the
2486 value of every expression in your assembly language program to be a
2487 section-relative address.
2490 @cindex assembler internal logic error
2491 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2492 An internal assembler logic error has been found. This means there is a
2493 bug in the assembler.
2495 @cindex expr (internal section)
2497 The assembler stores complex expression internally as combinations of
2498 symbols. When it needs to represent an expression as a symbol, it puts
2499 it in the expr section.
2501 @c FIXME item transfer[t] vector preload
2502 @c FIXME item transfer[t] vector postload
2503 @c FIXME item register
2507 @section Sub-Sections
2509 @cindex numbered subsections
2510 @cindex grouping data
2516 fall into two sections: text and data.
2518 You may have separate groups of
2520 data in named sections
2524 data in named sections
2530 that you want to end up near to each other in the object file, even though they
2531 are not contiguous in the assembler source. @code{@value{AS}} allows you to
2532 use @dfn{subsections} for this purpose. Within each section, there can be
2533 numbered subsections with values from 0 to 8192. Objects assembled into the
2534 same subsection go into the object file together with other objects in the same
2535 subsection. For example, a compiler might want to store constants in the text
2536 section, but might not want to have them interspersed with the program being
2537 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2538 section of code being output, and a @samp{.text 1} before each group of
2539 constants being output.
2541 Subsections are optional. If you do not use subsections, everything
2542 goes in subsection number zero.
2545 Each subsection is zero-padded up to a multiple of four bytes.
2546 (Subsections may be padded a different amount on different flavors
2547 of @code{@value{AS}}.)
2551 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2552 boundary (two bytes).
2553 The same is true on the Hitachi SH.
2556 @c FIXME section padding (alignment)?
2557 @c Rich Pixley says padding here depends on target obj code format; that
2558 @c doesn't seem particularly useful to say without further elaboration,
2559 @c so for now I say nothing about it. If this is a generic BFD issue,
2560 @c these paragraphs might need to vanish from this manual, and be
2561 @c discussed in BFD chapter of binutils (or some such).
2564 On the AMD 29K family, no particular padding is added to section or
2565 subsection sizes; @value{AS} forces no alignment on this platform.
2569 Subsections appear in your object file in numeric order, lowest numbered
2570 to highest. (All this to be compatible with other people's assemblers.)
2571 The object file contains no representation of subsections; @code{@value{LD}} and
2572 other programs that manipulate object files see no trace of them.
2573 They just see all your text subsections as a text section, and all your
2574 data subsections as a data section.
2576 To specify which subsection you want subsequent statements assembled
2577 into, use a numeric argument to specify it, in a @samp{.text
2578 @var{expression}} or a @samp{.data @var{expression}} statement.
2581 When generating COFF output, you
2586 can also use an extra subsection
2587 argument with arbitrary named sections: @samp{.section @var{name},
2590 @var{Expression} should be an absolute expression.
2591 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
2592 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
2593 begins in @code{text 0}. For instance:
2595 .text 0 # The default subsection is text 0 anyway.
2596 .ascii "This lives in the first text subsection. *"
2598 .ascii "But this lives in the second text subsection."
2600 .ascii "This lives in the data section,"
2601 .ascii "in the first data subsection."
2603 .ascii "This lives in the first text section,"
2604 .ascii "immediately following the asterisk (*)."
2607 Each section has a @dfn{location counter} incremented by one for every byte
2608 assembled into that section. Because subsections are merely a convenience
2609 restricted to @code{@value{AS}} there is no concept of a subsection location
2610 counter. There is no way to directly manipulate a location counter---but the
2611 @code{.align} directive changes it, and any label definition captures its
2612 current value. The location counter of the section where statements are being
2613 assembled is said to be the @dfn{active} location counter.
2616 @section bss Section
2619 @cindex common variable storage
2620 The bss section is used for local common variable storage.
2621 You may allocate address space in the bss section, but you may
2622 not dictate data to load into it before your program executes. When
2623 your program starts running, all the contents of the bss
2624 section are zeroed bytes.
2626 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
2627 @ref{Lcomm,,@code{.lcomm}}.
2629 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
2630 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
2633 When assembling for a target which supports multiple sections, such as ELF or
2634 COFF, you may switch into the @code{.bss} section and define symbols as usual;
2635 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
2636 section. Typically the section will only contain symbol definitions and
2637 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
2644 Symbols are a central concept: the programmer uses symbols to name
2645 things, the linker uses symbols to link, and the debugger uses symbols
2649 @cindex debuggers, and symbol order
2650 @emph{Warning:} @code{@value{AS}} does not place symbols in the object file in
2651 the same order they were declared. This may break some debuggers.
2656 * Setting Symbols:: Giving Symbols Other Values
2657 * Symbol Names:: Symbol Names
2658 * Dot:: The Special Dot Symbol
2659 * Symbol Attributes:: Symbol Attributes
2666 A @dfn{label} is written as a symbol immediately followed by a colon
2667 @samp{:}. The symbol then represents the current value of the
2668 active location counter, and is, for example, a suitable instruction
2669 operand. You are warned if you use the same symbol to represent two
2670 different locations: the first definition overrides any other
2674 On the HPPA, the usual form for a label need not be immediately followed by a
2675 colon, but instead must start in column zero. Only one label may be defined on
2676 a single line. To work around this, the HPPA version of @code{@value{AS}} also
2677 provides a special directive @code{.label} for defining labels more flexibly.
2680 @node Setting Symbols
2681 @section Giving Symbols Other Values
2683 @cindex assigning values to symbols
2684 @cindex symbol values, assigning
2685 A symbol can be given an arbitrary value by writing a symbol, followed
2686 by an equals sign @samp{=}, followed by an expression
2687 (@pxref{Expressions}). This is equivalent to using the @code{.set}
2688 directive. @xref{Set,,@code{.set}}.
2691 @section Symbol Names
2693 @cindex symbol names
2694 @cindex names, symbol
2695 @ifclear SPECIAL-SYMS
2696 Symbol names begin with a letter or with one of @samp{._}. On most
2697 machines, you can also use @code{$} in symbol names; exceptions are
2698 noted in @ref{Machine Dependencies}. That character may be followed by any
2699 string of digits, letters, dollar signs (unless otherwise noted in
2700 @ref{Machine Dependencies}), and underscores.
2703 For the AMD 29K family, @samp{?} is also allowed in the
2704 body of a symbol name, though not at its beginning.
2709 Symbol names begin with a letter or with one of @samp{._}. On the
2711 H8/500, you can also use @code{$} in symbol names. That character may
2712 be followed by any string of digits, letters, dollar signs (save on the
2713 H8/300), and underscores.
2717 Case of letters is significant: @code{foo} is a different symbol name
2720 Each symbol has exactly one name. Each name in an assembly language program
2721 refers to exactly one symbol. You may use that symbol name any number of times
2724 @subheading Local Symbol Names
2726 @cindex local symbol names
2727 @cindex symbol names, local
2728 @cindex temporary symbol names
2729 @cindex symbol names, temporary
2730 Local symbols help compilers and programmers use names temporarily.
2731 There are ten local symbol names, which are re-used throughout the
2732 program. You may refer to them using the names @samp{0} @samp{1}
2733 @dots{} @samp{9}. To define a local symbol, write a label of the form
2734 @samp{@b{N}:} (where @b{N} represents any digit). To refer to the most
2735 recent previous definition of that symbol write @samp{@b{N}b}, using the
2736 same digit as when you defined the label. To refer to the next
2737 definition of a local label, write @samp{@b{N}f}---where @b{N} gives you
2738 a choice of 10 forward references. The @samp{b} stands for
2739 ``backwards'' and the @samp{f} stands for ``forwards''.
2741 Local symbols are not emitted by the current @sc{gnu} C compiler.
2743 There is no restriction on how you can use these labels, but
2744 remember that at any point in the assembly you can refer to at most
2745 10 prior local labels and to at most 10 forward local labels.
2747 Local symbol names are only a notation device. They are immediately
2748 transformed into more conventional symbol names before the assembler
2749 uses them. The symbol names stored in the symbol table, appearing in
2750 error messages and optionally emitted to the object file have these
2755 All local labels begin with @samp{L}. Normally both @code{@value{AS}} and
2756 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
2757 used for symbols you are never intended to see. If you use the
2758 @samp{-L} option then @code{@value{AS}} retains these symbols in the
2759 object file. If you also instruct @code{@value{LD}} to retain these symbols,
2760 you may use them in debugging.
2763 If the label is written @samp{0:} then the digit is @samp{0}.
2764 If the label is written @samp{1:} then the digit is @samp{1}.
2765 And so on up through @samp{9:}.
2768 This unusual character is included so you do not accidentally invent
2769 a symbol of the same name. The character has ASCII value
2772 @item @emph{ordinal number}
2773 This is a serial number to keep the labels distinct. The first
2774 @samp{0:} gets the number @samp{1}; The 15th @samp{0:} gets the
2775 number @samp{15}; @emph{etc.}. Likewise for the other labels @samp{1:}
2779 For instance, the first @code{1:} is named @code{L1@kbd{C-A}1}, the 44th
2780 @code{3:} is named @code{L3@kbd{C-A}44}.
2783 @section The Special Dot Symbol
2785 @cindex dot (symbol)
2786 @cindex @code{.} (symbol)
2787 @cindex current address
2788 @cindex location counter
2789 The special symbol @samp{.} refers to the current address that
2790 @code{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
2791 .long .} defines @code{melvin} to contain its own address.
2792 Assigning a value to @code{.} is treated the same as a @code{.org}
2793 directive. Thus, the expression @samp{.=.+4} is the same as saying
2794 @ifclear no-space-dir
2803 @node Symbol Attributes
2804 @section Symbol Attributes
2806 @cindex symbol attributes
2807 @cindex attributes, symbol
2808 Every symbol has, as well as its name, the attributes ``Value'' and
2809 ``Type''. Depending on output format, symbols can also have auxiliary
2812 The detailed definitions are in @file{a.out.h}.
2815 If you use a symbol without defining it, @code{@value{AS}} assumes zero for
2816 all these attributes, and probably won't warn you. This makes the
2817 symbol an externally defined symbol, which is generally what you
2821 * Symbol Value:: Value
2822 * Symbol Type:: Type
2825 * a.out Symbols:: Symbol Attributes: @code{a.out}
2829 * a.out Symbols:: Symbol Attributes: @code{a.out}
2832 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
2837 * COFF Symbols:: Symbol Attributes for COFF
2840 * SOM Symbols:: Symbol Attributes for SOM
2847 @cindex value of a symbol
2848 @cindex symbol value
2849 The value of a symbol is (usually) 32 bits. For a symbol which labels a
2850 location in the text, data, bss or absolute sections the value is the
2851 number of addresses from the start of that section to the label.
2852 Naturally for text, data and bss sections the value of a symbol changes
2853 as @code{@value{LD}} changes section base addresses during linking. Absolute
2854 symbols' values do not change during linking: that is why they are
2857 The value of an undefined symbol is treated in a special way. If it is
2858 0 then the symbol is not defined in this assembler source file, and
2859 @code{@value{LD}} tries to determine its value from other files linked into the
2860 same program. You make this kind of symbol simply by mentioning a symbol
2861 name without defining it. A non-zero value represents a @code{.comm}
2862 common declaration. The value is how much common storage to reserve, in
2863 bytes (addresses). The symbol refers to the first address of the
2869 @cindex type of a symbol
2871 The type attribute of a symbol contains relocation (section)
2872 information, any flag settings indicating that a symbol is external, and
2873 (optionally), other information for linkers and debuggers. The exact
2874 format depends on the object-code output format in use.
2879 @c The following avoids a "widow" subsection title. @group would be
2880 @c better if it were available outside examples.
2883 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
2885 @cindex @code{b.out} symbol attributes
2886 @cindex symbol attributes, @code{b.out}
2887 These symbol attributes appear only when @code{@value{AS}} is configured for
2888 one of the Berkeley-descended object output formats---@code{a.out} or
2894 @subsection Symbol Attributes: @code{a.out}
2896 @cindex @code{a.out} symbol attributes
2897 @cindex symbol attributes, @code{a.out}
2903 @subsection Symbol Attributes: @code{a.out}
2905 @cindex @code{a.out} symbol attributes
2906 @cindex symbol attributes, @code{a.out}
2910 * Symbol Desc:: Descriptor
2911 * Symbol Other:: Other
2915 @subsubsection Descriptor
2917 @cindex descriptor, of @code{a.out} symbol
2918 This is an arbitrary 16-bit value. You may establish a symbol's
2919 descriptor value by using a @code{.desc} statement
2920 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
2924 @subsubsection Other
2926 @cindex other attribute, of @code{a.out} symbol
2927 This is an arbitrary 8-bit value. It means nothing to @code{@value{AS}}.
2932 @subsection Symbol Attributes for COFF
2934 @cindex COFF symbol attributes
2935 @cindex symbol attributes, COFF
2937 The COFF format supports a multitude of auxiliary symbol attributes;
2938 like the primary symbol attributes, they are set between @code{.def} and
2939 @code{.endef} directives.
2941 @subsubsection Primary Attributes
2943 @cindex primary attributes, COFF symbols
2944 The symbol name is set with @code{.def}; the value and type,
2945 respectively, with @code{.val} and @code{.type}.
2947 @subsubsection Auxiliary Attributes
2949 @cindex auxiliary attributes, COFF symbols
2950 The @code{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
2951 @code{.size}, and @code{.tag} can generate auxiliary symbol table
2952 information for COFF.
2957 @subsection Symbol Attributes for SOM
2959 @cindex SOM symbol attributes
2960 @cindex symbol attributes, SOM
2962 The SOM format for the HPPA supports a multitude of symbol attributes set with
2963 the @code{.EXPORT} and @code{.IMPORT} directives.
2965 The attributes are described in @cite{HP9000 Series 800 Assembly
2966 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
2967 @code{EXPORT} assembler directive documentation.
2971 @chapter Expressions
2975 @cindex numeric values
2976 An @dfn{expression} specifies an address or numeric value.
2977 Whitespace may precede and/or follow an expression.
2979 The result of an expression must be an absolute number, or else an offset into
2980 a particular section. If an expression is not absolute, and there is not
2981 enough information when @code{@value{AS}} sees the expression to know its
2982 section, a second pass over the source program might be necessary to interpret
2983 the expression---but the second pass is currently not implemented.
2984 @code{@value{AS}} aborts with an error message in this situation.
2987 * Empty Exprs:: Empty Expressions
2988 * Integer Exprs:: Integer Expressions
2992 @section Empty Expressions
2994 @cindex empty expressions
2995 @cindex expressions, empty
2996 An empty expression has no value: it is just whitespace or null.
2997 Wherever an absolute expression is required, you may omit the
2998 expression, and @code{@value{AS}} assumes a value of (absolute) 0. This
2999 is compatible with other assemblers.
3002 @section Integer Expressions
3004 @cindex integer expressions
3005 @cindex expressions, integer
3006 An @dfn{integer expression} is one or more @emph{arguments} delimited
3007 by @emph{operators}.
3010 * Arguments:: Arguments
3011 * Operators:: Operators
3012 * Prefix Ops:: Prefix Operators
3013 * Infix Ops:: Infix Operators
3017 @subsection Arguments
3019 @cindex expression arguments
3020 @cindex arguments in expressions
3021 @cindex operands in expressions
3022 @cindex arithmetic operands
3023 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3024 contexts arguments are sometimes called ``arithmetic operands''. In
3025 this manual, to avoid confusing them with the ``instruction operands'' of
3026 the machine language, we use the term ``argument'' to refer to parts of
3027 expressions only, reserving the word ``operand'' to refer only to machine
3028 instruction operands.
3030 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3031 @var{section} is one of text, data, bss, absolute,
3032 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3035 Numbers are usually integers.
3037 A number can be a flonum or bignum. In this case, you are warned
3038 that only the low order 32 bits are used, and @code{@value{AS}} pretends
3039 these 32 bits are an integer. You may write integer-manipulating
3040 instructions that act on exotic constants, compatible with other
3043 @cindex subexpressions
3044 Subexpressions are a left parenthesis @samp{(} followed by an integer
3045 expression, followed by a right parenthesis @samp{)}; or a prefix
3046 operator followed by an argument.
3049 @subsection Operators
3051 @cindex operators, in expressions
3052 @cindex arithmetic functions
3053 @cindex functions, in expressions
3054 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3055 operators are followed by an argument. Infix operators appear
3056 between their arguments. Operators may be preceded and/or followed by
3060 @subsection Prefix Operator
3062 @cindex prefix operators
3063 @code{@value{AS}} has the following @dfn{prefix operators}. They each take
3064 one argument, which must be absolute.
3066 @c the tex/end tex stuff surrounding this small table is meant to make
3067 @c it align, on the printed page, with the similar table in the next
3068 @c section (which is inside an enumerate).
3070 \global\advance\leftskip by \itemindent
3075 @dfn{Negation}. Two's complement negation.
3077 @dfn{Complementation}. Bitwise not.
3081 \global\advance\leftskip by -\itemindent
3085 @subsection Infix Operators
3087 @cindex infix operators
3088 @cindex operators, permitted arguments
3089 @dfn{Infix operators} take two arguments, one on either side. Operators
3090 have precedence, but operations with equal precedence are performed left
3091 to right. Apart from @code{+} or @code{-}, both arguments must be
3092 absolute, and the result is absolute.
3095 @cindex operator precedence
3096 @cindex precedence of operators
3103 @dfn{Multiplication}.
3106 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3113 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3117 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3121 Intermediate precedence
3126 @dfn{Bitwise Inclusive Or}.
3132 @dfn{Bitwise Exclusive Or}.
3135 @dfn{Bitwise Or Not}.
3142 @cindex addition, permitted arguments
3143 @cindex plus, permitted arguments
3144 @cindex arguments for addition
3146 @dfn{Addition}. If either argument is absolute, the result has the section of
3147 the other argument. You may not add together arguments from different
3150 @cindex subtraction, permitted arguments
3151 @cindex minus, permitted arguments
3152 @cindex arguments for subtraction
3154 @dfn{Subtraction}. If the right argument is absolute, the
3155 result has the section of the left argument.
3156 If both arguments are in the same section, the result is absolute.
3157 You may not subtract arguments from different sections.
3158 @c FIXME is there still something useful to say about undefined - undefined ?
3162 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3163 address; you can only have a defined section in one of the two arguments.
3166 @chapter Assembler Directives
3168 @cindex directives, machine independent
3169 @cindex pseudo-ops, machine independent
3170 @cindex machine independent directives
3171 All assembler directives have names that begin with a period (@samp{.}).
3172 The rest of the name is letters, usually in lower case.
3174 This chapter discusses directives that are available regardless of the
3175 target machine configuration for the @sc{gnu} assembler.
3177 Some machine configurations provide additional directives.
3178 @xref{Machine Dependencies}.
3181 @ifset machine-directives
3182 @xref{Machine Dependencies} for additional directives.
3187 * Abort:: @code{.abort}
3189 * ABORT:: @code{.ABORT}
3192 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3193 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3194 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3195 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3196 * Byte:: @code{.byte @var{expressions}}
3197 * Comm:: @code{.comm @var{symbol} , @var{length} }
3198 * Data:: @code{.data @var{subsection}}
3200 * Def:: @code{.def @var{name}}
3203 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3209 * Double:: @code{.double @var{flonums}}
3210 * Eject:: @code{.eject}
3211 * Else:: @code{.else}
3212 * Elseif:: @code{.elseif}
3215 * Endef:: @code{.endef}
3218 * Endfunc:: @code{.endfunc}
3219 * Endif:: @code{.endif}
3220 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3221 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3223 * Exitm:: @code{.exitm}
3224 * Extern:: @code{.extern}
3225 * Fail:: @code{.fail}
3226 @ifclear no-file-dir
3227 * File:: @code{.file @var{string}}
3230 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3231 * Float:: @code{.float @var{flonums}}
3232 * Func:: @code{.func}
3233 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3235 * Hidden:: @code{.hidden @var{names}}
3238 * hword:: @code{.hword @var{expressions}}
3239 * Ident:: @code{.ident}
3240 * If:: @code{.if @var{absolute expression}}
3241 * Include:: @code{.include "@var{file}"}
3242 * Int:: @code{.int @var{expressions}}
3244 * Internal:: @code{.internal @var{names}}
3247 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3248 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3249 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3250 * Lflags:: @code{.lflags}
3251 @ifclear no-line-dir
3252 * Line:: @code{.line @var{line-number}}
3255 * Ln:: @code{.ln @var{line-number}}
3256 * Linkonce:: @code{.linkonce [@var{type}]}
3257 * List:: @code{.list}
3258 * Long:: @code{.long @var{expressions}}
3260 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3263 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3264 * MRI:: @code{.mri @var{val}}
3265 * Nolist:: @code{.nolist}
3266 * Octa:: @code{.octa @var{bignums}}
3267 * Org:: @code{.org @var{new-lc} , @var{fill}}
3268 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3270 * PopSection:: @code{.popsection}
3271 * Previous:: @code{.previous}
3274 * Print:: @code{.print @var{string}}
3276 * Protected:: @code{.protected @var{names}}
3279 * Psize:: @code{.psize @var{lines}, @var{columns}}
3280 * Purgem:: @code{.purgem @var{name}}
3282 * PushSection:: @code{.pushsection @var{name}}
3285 * Quad:: @code{.quad @var{bignums}}
3286 * Rept:: @code{.rept @var{count}}
3287 * Sbttl:: @code{.sbttl "@var{subheading}"}
3289 * Scl:: @code{.scl @var{class}}
3290 * Section:: @code{.section @var{name}, @var{subsection}}
3293 * Set:: @code{.set @var{symbol}, @var{expression}}
3294 * Short:: @code{.short @var{expressions}}
3295 * Single:: @code{.single @var{flonums}}
3296 * Size:: @code{.size [@var{name} , @var{expression}]}
3297 * Skip:: @code{.skip @var{size} , @var{fill}}
3298 * Sleb128:: @code{.sleb128 @var{expressions}}
3299 * Space:: @code{.space @var{size} , @var{fill}}
3301 * Stab:: @code{.stabd, .stabn, .stabs}
3304 * String:: @code{.string "@var{str}"}
3305 * Struct:: @code{.struct @var{expression}}
3307 * SubSection:: @code{.subsection}
3308 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3312 * Tag:: @code{.tag @var{structname}}
3315 * Text:: @code{.text @var{subsection}}
3316 * Title:: @code{.title "@var{heading}"}
3317 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3318 * Uleb128:: @code{.uleb128 @var{expressions}}
3320 * Val:: @code{.val @var{addr}}
3324 * Version:: @code{.version "@var{string}"}
3325 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3326 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3327 * Weak:: @code{.weak @var{names}}
3330 * Word:: @code{.word @var{expressions}}
3331 * Deprecated:: Deprecated Directives
3335 @section @code{.abort}
3337 @cindex @code{abort} directive
3338 @cindex stopping the assembly
3339 This directive stops the assembly immediately. It is for
3340 compatibility with other assemblers. The original idea was that the
3341 assembly language source would be piped into the assembler. If the sender
3342 of the source quit, it could use this directive tells @code{@value{AS}} to
3343 quit also. One day @code{.abort} will not be supported.
3347 @section @code{.ABORT}
3349 @cindex @code{ABORT} directive
3350 When producing COFF output, @code{@value{AS}} accepts this directive as a
3351 synonym for @samp{.abort}.
3354 When producing @code{b.out} output, @code{@value{AS}} accepts this directive,
3360 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3362 @cindex padding the location counter
3363 @cindex @code{align} directive
3364 Pad the location counter (in the current subsection) to a particular storage
3365 boundary. The first expression (which must be absolute) is the alignment
3366 required, as described below.
3368 The second expression (also absolute) gives the fill value to be stored in the
3369 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3370 padding bytes are normally zero. However, on some systems, if the section is
3371 marked as containing code and the fill value is omitted, the space is filled
3372 with no-op instructions.
3374 The third expression is also absolute, and is also optional. If it is present,
3375 it is the maximum number of bytes that should be skipped by this alignment
3376 directive. If doing the alignment would require skipping more bytes than the
3377 specified maximum, then the alignment is not done at all. You can omit the
3378 fill value (the second argument) entirely by simply using two commas after the
3379 required alignment; this can be useful if you want the alignment to be filled
3380 with no-op instructions when appropriate.
3382 The way the required alignment is specified varies from system to system.
3383 For the a29k, hppa, m68k, m88k, w65, sparc, and Hitachi SH, and i386 using ELF
3385 the first expression is the
3386 alignment request in bytes. For example @samp{.align 8} advances
3387 the location counter until it is a multiple of 8. If the location counter
3388 is already a multiple of 8, no change is needed.
3390 For other systems, including the i386 using a.out format, and the arm and
3391 strongarm, it is the
3392 number of low-order zero bits the location counter must have after
3393 advancement. For example @samp{.align 3} advances the location
3394 counter until it a multiple of 8. If the location counter is already a
3395 multiple of 8, no change is needed.
3397 This inconsistency is due to the different behaviors of the various
3398 native assemblers for these systems which GAS must emulate.
3399 GAS also provides @code{.balign} and @code{.p2align} directives,
3400 described later, which have a consistent behavior across all
3401 architectures (but are specific to GAS).
3404 @section @code{.ascii "@var{string}"}@dots{}
3406 @cindex @code{ascii} directive
3407 @cindex string literals
3408 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3409 separated by commas. It assembles each string (with no automatic
3410 trailing zero byte) into consecutive addresses.
3413 @section @code{.asciz "@var{string}"}@dots{}
3415 @cindex @code{asciz} directive
3416 @cindex zero-terminated strings
3417 @cindex null-terminated strings
3418 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3419 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3422 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3424 @cindex padding the location counter given number of bytes
3425 @cindex @code{balign} directive
3426 Pad the location counter (in the current subsection) to a particular
3427 storage boundary. The first expression (which must be absolute) is the
3428 alignment request in bytes. For example @samp{.balign 8} advances
3429 the location counter until it is a multiple of 8. If the location counter
3430 is already a multiple of 8, no change is needed.
3432 The second expression (also absolute) gives the fill value to be stored in the
3433 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3434 padding bytes are normally zero. However, on some systems, if the section is
3435 marked as containing code and the fill value is omitted, the space is filled
3436 with no-op instructions.
3438 The third expression is also absolute, and is also optional. If it is present,
3439 it is the maximum number of bytes that should be skipped by this alignment
3440 directive. If doing the alignment would require skipping more bytes than the
3441 specified maximum, then the alignment is not done at all. You can omit the
3442 fill value (the second argument) entirely by simply using two commas after the
3443 required alignment; this can be useful if you want the alignment to be filled
3444 with no-op instructions when appropriate.
3446 @cindex @code{balignw} directive
3447 @cindex @code{balignl} directive
3448 The @code{.balignw} and @code{.balignl} directives are variants of the
3449 @code{.balign} directive. The @code{.balignw} directive treats the fill
3450 pattern as a two byte word value. The @code{.balignl} directives treats the
3451 fill pattern as a four byte longword value. For example, @code{.balignw
3452 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3453 filled in with the value 0x368d (the exact placement of the bytes depends upon
3454 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3458 @section @code{.byte @var{expressions}}
3460 @cindex @code{byte} directive
3461 @cindex integers, one byte
3462 @code{.byte} expects zero or more expressions, separated by commas.
3463 Each expression is assembled into the next byte.
3466 @section @code{.comm @var{symbol} , @var{length} }
3468 @cindex @code{comm} directive
3469 @cindex symbol, common
3470 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
3471 common symbol in one object file may be merged with a defined or common symbol
3472 of the same name in another object file. If @code{@value{LD}} does not see a
3473 definition for the symbol--just one or more common symbols--then it will
3474 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
3475 absolute expression. If @code{@value{LD}} sees multiple common symbols with
3476 the same name, and they do not all have the same size, it will allocate space
3477 using the largest size.
3480 When using ELF, the @code{.comm} directive takes an optional third argument.
3481 This is the desired alignment of the symbol, specified as a byte boundary (for
3482 example, an alignment of 16 means that the least significant 4 bits of the
3483 address should be zero). The alignment must be an absolute expression, and it
3484 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
3485 for the common symbol, it will use the alignment when placing the symbol. If
3486 no alignment is specified, @code{@value{AS}} will set the alignment to the
3487 largest power of two less than or equal to the size of the symbol, up to a
3492 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
3493 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
3497 @section @code{.data @var{subsection}}
3499 @cindex @code{data} directive
3500 @code{.data} tells @code{@value{AS}} to assemble the following statements onto the
3501 end of the data subsection numbered @var{subsection} (which is an
3502 absolute expression). If @var{subsection} is omitted, it defaults
3507 @section @code{.def @var{name}}
3509 @cindex @code{def} directive
3510 @cindex COFF symbols, debugging
3511 @cindex debugging COFF symbols
3512 Begin defining debugging information for a symbol @var{name}; the
3513 definition extends until the @code{.endef} directive is encountered.
3516 This directive is only observed when @code{@value{AS}} is configured for COFF
3517 format output; when producing @code{b.out}, @samp{.def} is recognized,
3524 @section @code{.desc @var{symbol}, @var{abs-expression}}
3526 @cindex @code{desc} directive
3527 @cindex COFF symbol descriptor
3528 @cindex symbol descriptor, COFF
3529 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
3530 to the low 16 bits of an absolute expression.
3533 The @samp{.desc} directive is not available when @code{@value{AS}} is
3534 configured for COFF output; it is only for @code{a.out} or @code{b.out}
3535 object format. For the sake of compatibility, @code{@value{AS}} accepts
3536 it, but produces no output, when configured for COFF.
3542 @section @code{.dim}
3544 @cindex @code{dim} directive
3545 @cindex COFF auxiliary symbol information
3546 @cindex auxiliary symbol information, COFF
3547 This directive is generated by compilers to include auxiliary debugging
3548 information in the symbol table. It is only permitted inside
3549 @code{.def}/@code{.endef} pairs.
3552 @samp{.dim} is only meaningful when generating COFF format output; when
3553 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
3559 @section @code{.double @var{flonums}}
3561 @cindex @code{double} directive
3562 @cindex floating point numbers (double)
3563 @code{.double} expects zero or more flonums, separated by commas. It
3564 assembles floating point numbers.
3566 The exact kind of floating point numbers emitted depends on how
3567 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
3571 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
3572 in @sc{ieee} format.
3577 @section @code{.eject}
3579 @cindex @code{eject} directive
3580 @cindex new page, in listings
3581 @cindex page, in listings
3582 @cindex listing control: new page
3583 Force a page break at this point, when generating assembly listings.
3586 @section @code{.else}
3588 @cindex @code{else} directive
3589 @code{.else} is part of the @code{@value{AS}} support for conditional
3590 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
3591 of code to be assembled if the condition for the preceding @code{.if}
3595 @section @code{.elseif}
3597 @cindex @code{elseif} directive
3598 @code{.elseif} is part of the @code{@value{AS}} support for conditional
3599 assembly; @pxref{If,,@code{.if}}. It is shorthand for beginning a new
3600 @code{.if} block that would otherwise fill the entire @code{.else} section.
3603 @section @code{.end}
3605 @cindex @code{end} directive
3606 @code{.end} marks the end of the assembly file. @code{@value{AS}} does not
3607 process anything in the file past the @code{.end} directive.
3611 @section @code{.endef}
3613 @cindex @code{endef} directive
3614 This directive flags the end of a symbol definition begun with
3618 @samp{.endef} is only meaningful when generating COFF format output; if
3619 @code{@value{AS}} is configured to generate @code{b.out}, it accepts this
3620 directive but ignores it.
3625 @section @code{.endfunc}
3626 @cindex @code{endfunc} directive
3627 @code{.endfunc} marks the end of a function specified with @code{.func}.
3630 @section @code{.endif}
3632 @cindex @code{endif} directive
3633 @code{.endif} is part of the @code{@value{AS}} support for conditional assembly;
3634 it marks the end of a block of code that is only assembled
3635 conditionally. @xref{If,,@code{.if}}.
3638 @section @code{.equ @var{symbol}, @var{expression}}
3640 @cindex @code{equ} directive
3641 @cindex assigning values to symbols
3642 @cindex symbols, assigning values to
3643 This directive sets the value of @var{symbol} to @var{expression}.
3644 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
3647 The syntax for @code{equ} on the HPPA is
3648 @samp{@var{symbol} .equ @var{expression}}.
3652 @section @code{.equiv @var{symbol}, @var{expression}}
3653 @cindex @code{equiv} directive
3654 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
3655 the assembler will signal an error if @var{symbol} is already defined.
3657 Except for the contents of the error message, this is roughly equivalent to
3666 @section @code{.err}
3667 @cindex @code{err} directive
3668 If @code{@value{AS}} assembles a @code{.err} directive, it will print an error
3669 message and, unless the @code{-Z} option was used, it will not generate an
3670 object file. This can be used to signal error an conditionally compiled code.
3673 @section @code{.exitm}
3674 Exit early from the current macro definition. @xref{Macro}.
3677 @section @code{.extern}
3679 @cindex @code{extern} directive
3680 @code{.extern} is accepted in the source program---for compatibility
3681 with other assemblers---but it is ignored. @code{@value{AS}} treats
3682 all undefined symbols as external.
3685 @section @code{.fail @var{expression}}
3687 @cindex @code{fail} directive
3688 Generates an error or a warning. If the value of the @var{expression} is 500
3689 or more, @code{@value{AS}} will print a warning message. If the value is less
3690 than 500, @code{@value{AS}} will print an error message. The message will
3691 include the value of @var{expression}. This can occasionally be useful inside
3692 complex nested macros or conditional assembly.
3694 @ifclear no-file-dir
3696 @section @code{.file @var{string}}
3698 @cindex @code{file} directive
3699 @cindex logical file name
3700 @cindex file name, logical
3701 @code{.file} tells @code{@value{AS}} that we are about to start a new logical
3702 file. @var{string} is the new file name. In general, the filename is
3703 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
3704 to specify an empty file name, you must give the quotes--@code{""}. This
3705 statement may go away in future: it is only recognized to be compatible with
3706 old @code{@value{AS}} programs.
3708 In some configurations of @code{@value{AS}}, @code{.file} has already been
3709 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
3714 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
3716 @cindex @code{fill} directive
3717 @cindex writing patterns in memory
3718 @cindex patterns, writing in memory
3719 @var{result}, @var{size} and @var{value} are absolute expressions.
3720 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
3721 may be zero or more. @var{Size} may be zero or more, but if it is
3722 more than 8, then it is deemed to have the value 8, compatible with
3723 other people's assemblers. The contents of each @var{repeat} bytes
3724 is taken from an 8-byte number. The highest order 4 bytes are
3725 zero. The lowest order 4 bytes are @var{value} rendered in the
3726 byte-order of an integer on the computer @code{@value{AS}} is assembling for.
3727 Each @var{size} bytes in a repetition is taken from the lowest order
3728 @var{size} bytes of this number. Again, this bizarre behavior is
3729 compatible with other people's assemblers.
3731 @var{size} and @var{value} are optional.
3732 If the second comma and @var{value} are absent, @var{value} is
3733 assumed zero. If the first comma and following tokens are absent,
3734 @var{size} is assumed to be 1.
3737 @section @code{.float @var{flonums}}
3739 @cindex floating point numbers (single)
3740 @cindex @code{float} directive
3741 This directive assembles zero or more flonums, separated by commas. It
3742 has the same effect as @code{.single}.
3744 The exact kind of floating point numbers emitted depends on how
3745 @code{@value{AS}} is configured.
3746 @xref{Machine Dependencies}.
3750 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
3751 in @sc{ieee} format.
3756 @section @code{.func @var{name}[,@var{label}]}
3757 @cindex @code{func} directive
3758 @code{.func} emits debugging information to denote function @var{name}, and
3759 is ignored unless the file is assembled with debugging enabled.
3760 Only @samp{--gstabs} is currently supported.
3761 @var{label} is the entry point of the function and if omitted @var{name}
3762 prepended with the @samp{leading char} is used.
3763 @samp{leading char} is usually @code{_} or nothing, depending on the target.
3764 All functions are currently defined to have @code{void} return type.
3765 The function must be terminated with @code{.endfunc}.
3768 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3770 @cindex @code{global} directive
3771 @cindex symbol, making visible to linker
3772 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
3773 @var{symbol} in your partial program, its value is made available to
3774 other partial programs that are linked with it. Otherwise,
3775 @var{symbol} takes its attributes from a symbol of the same name
3776 from another file linked into the same program.
3778 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
3779 compatibility with other assemblers.
3782 On the HPPA, @code{.global} is not always enough to make it accessible to other
3783 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
3784 @xref{HPPA Directives,, HPPA Assembler Directives}.
3789 @section @code{.hidden @var{names}}
3791 @cindex @code{.hidden} directive
3793 This one of the ELF visibility directives. The other two are
3794 @pxref{Internal,,@code{.internal}} and @pxref{Protected,,@code{.protected}}
3796 This directive overrides the named symbols default visibility (which is set by
3797 their binding: local, global or weak). The directive sets the visibility to
3798 @code{hidden} which means that the symbols are not visible to other components.
3799 Such symbols are always considered to be @code{protected} as well.
3803 @section @code{.hword @var{expressions}}
3805 @cindex @code{hword} directive
3806 @cindex integers, 16-bit
3807 @cindex numbers, 16-bit
3808 @cindex sixteen bit integers
3809 This expects zero or more @var{expressions}, and emits
3810 a 16 bit number for each.
3813 This directive is a synonym for @samp{.short}; depending on the target
3814 architecture, it may also be a synonym for @samp{.word}.
3818 This directive is a synonym for @samp{.short}.
3821 This directive is a synonym for both @samp{.short} and @samp{.word}.
3826 @section @code{.ident}
3828 @cindex @code{ident} directive
3829 This directive is used by some assemblers to place tags in object files.
3830 @code{@value{AS}} simply accepts the directive for source-file
3831 compatibility with such assemblers, but does not actually emit anything
3835 @section @code{.if @var{absolute expression}}
3837 @cindex conditional assembly
3838 @cindex @code{if} directive
3839 @code{.if} marks the beginning of a section of code which is only
3840 considered part of the source program being assembled if the argument
3841 (which must be an @var{absolute expression}) is non-zero. The end of
3842 the conditional section of code must be marked by @code{.endif}
3843 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
3844 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
3845 If you have several conditions to check, @code{.elseif} may be used to avoid
3846 nesting blocks if/else within each subsequent @code{.else} block.
3848 The following variants of @code{.if} are also supported:
3850 @cindex @code{ifdef} directive
3851 @item .ifdef @var{symbol}
3852 Assembles the following section of code if the specified @var{symbol}
3855 @cindex @code{ifc} directive
3856 @item .ifc @var{string1},@var{string2}
3857 Assembles the following section of code if the two strings are the same. The
3858 strings may be optionally quoted with single quotes. If they are not quoted,
3859 the first string stops at the first comma, and the second string stops at the
3860 end of the line. Strings which contain whitespace should be quoted. The
3861 string comparison is case sensitive.
3863 @cindex @code{ifeq} directive
3864 @item .ifeq @var{absolute expression}
3865 Assembles the following section of code if the argument is zero.
3867 @cindex @code{ifeqs} directive
3868 @item .ifeqs @var{string1},@var{string2}
3869 Another form of @code{.ifc}. The strings must be quoted using double quotes.
3871 @cindex @code{ifge} directive
3872 @item .ifge @var{absolute expression}
3873 Assembles the following section of code if the argument is greater than or
3876 @cindex @code{ifgt} directive
3877 @item .ifgt @var{absolute expression}
3878 Assembles the following section of code if the argument is greater than zero.
3880 @cindex @code{ifle} directive
3881 @item .ifle @var{absolute expression}
3882 Assembles the following section of code if the argument is less than or equal
3885 @cindex @code{iflt} directive
3886 @item .iflt @var{absolute expression}
3887 Assembles the following section of code if the argument is less than zero.
3889 @cindex @code{ifnc} directive
3890 @item .ifnc @var{string1},@var{string2}.
3891 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
3892 following section of code if the two strings are not the same.
3894 @cindex @code{ifndef} directive
3895 @cindex @code{ifnotdef} directive
3896 @item .ifndef @var{symbol}
3897 @itemx .ifnotdef @var{symbol}
3898 Assembles the following section of code if the specified @var{symbol}
3899 has not been defined. Both spelling variants are equivalent.
3901 @cindex @code{ifne} directive
3902 @item .ifne @var{absolute expression}
3903 Assembles the following section of code if the argument is not equal to zero
3904 (in other words, this is equivalent to @code{.if}).
3906 @cindex @code{ifnes} directive
3907 @item .ifnes @var{string1},@var{string2}
3908 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
3909 following section of code if the two strings are not the same.
3913 @section @code{.include "@var{file}"}
3915 @cindex @code{include} directive
3916 @cindex supporting files, including
3917 @cindex files, including
3918 This directive provides a way to include supporting files at specified
3919 points in your source program. The code from @var{file} is assembled as
3920 if it followed the point of the @code{.include}; when the end of the
3921 included file is reached, assembly of the original file continues. You
3922 can control the search paths used with the @samp{-I} command-line option
3923 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
3927 @section @code{.int @var{expressions}}
3929 @cindex @code{int} directive
3930 @cindex integers, 32-bit
3931 Expect zero or more @var{expressions}, of any section, separated by commas.
3932 For each expression, emit a number that, at run time, is the value of that
3933 expression. The byte order and bit size of the number depends on what kind
3934 of target the assembly is for.
3938 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
3939 integers. On the H8/300H and the Hitachi SH, however, @code{.int} emits
3946 @section @code{.internal @var{names}}
3948 @cindex @code{.internal} directive
3950 This one of the ELF visibility directives. The other two are
3951 @pxref{Hidden,,@code{.hidden}} and @pxref{Protected,,@code{.protected}}
3953 This directive overrides the named symbols default visibility (which is set by
3954 their binding: local, global or weak). The directive sets the visibility to
3955 @code{internal} which means that the symbols are considered to be @code{hidden}
3956 (ie not visible to other components), and that some extra, processor specific
3957 processing must also be performed upon the symbols as well.
3961 @section @code{.irp @var{symbol},@var{values}}@dots{}
3963 @cindex @code{irp} directive
3964 Evaluate a sequence of statements assigning different values to @var{symbol}.
3965 The sequence of statements starts at the @code{.irp} directive, and is
3966 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
3967 set to @var{value}, and the sequence of statements is assembled. If no
3968 @var{value} is listed, the sequence of statements is assembled once, with
3969 @var{symbol} set to the null string. To refer to @var{symbol} within the
3970 sequence of statements, use @var{\symbol}.
3972 For example, assembling
3980 is equivalent to assembling
3989 @section @code{.irpc @var{symbol},@var{values}}@dots{}
3991 @cindex @code{irpc} directive
3992 Evaluate a sequence of statements assigning different values to @var{symbol}.
3993 The sequence of statements starts at the @code{.irpc} directive, and is
3994 terminated by an @code{.endr} directive. For each character in @var{value},
3995 @var{symbol} is set to the character, and the sequence of statements is
3996 assembled. If no @var{value} is listed, the sequence of statements is
3997 assembled once, with @var{symbol} set to the null string. To refer to
3998 @var{symbol} within the sequence of statements, use @var{\symbol}.
4000 For example, assembling
4008 is equivalent to assembling
4017 @section @code{.lcomm @var{symbol} , @var{length}}
4019 @cindex @code{lcomm} directive
4020 @cindex local common symbols
4021 @cindex symbols, local common
4022 Reserve @var{length} (an absolute expression) bytes for a local common
4023 denoted by @var{symbol}. The section and value of @var{symbol} are
4024 those of the new local common. The addresses are allocated in the bss
4025 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4026 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4027 not visible to @code{@value{LD}}.
4030 Some targets permit a third argument to be used with @code{.lcomm}. This
4031 argument specifies the desired alignment of the symbol in the bss section.
4035 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4036 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4040 @section @code{.lflags}
4042 @cindex @code{lflags} directive (ignored)
4043 @code{@value{AS}} accepts this directive, for compatibility with other
4044 assemblers, but ignores it.
4046 @ifclear no-line-dir
4048 @section @code{.line @var{line-number}}
4050 @cindex @code{line} directive
4054 @section @code{.ln @var{line-number}}
4056 @cindex @code{ln} directive
4058 @cindex logical line number
4060 Change the logical line number. @var{line-number} must be an absolute
4061 expression. The next line has that logical line number. Therefore any other
4062 statements on the current line (after a statement separator character) are
4063 reported as on logical line number @var{line-number} @minus{} 1. One day
4064 @code{@value{AS}} will no longer support this directive: it is recognized only
4065 for compatibility with existing assembler programs.
4069 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
4070 not available; use the synonym @code{.ln} in that context.
4075 @ifclear no-line-dir
4076 Even though this is a directive associated with the @code{a.out} or
4077 @code{b.out} object-code formats, @code{@value{AS}} still recognizes it
4078 when producing COFF output, and treats @samp{.line} as though it
4079 were the COFF @samp{.ln} @emph{if} it is found outside a
4080 @code{.def}/@code{.endef} pair.
4082 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4083 used by compilers to generate auxiliary symbol information for
4088 @section @code{.linkonce [@var{type}]}
4090 @cindex @code{linkonce} directive
4091 @cindex common sections
4092 Mark the current section so that the linker only includes a single copy of it.
4093 This may be used to include the same section in several different object files,
4094 but ensure that the linker will only include it once in the final output file.
4095 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4096 Duplicate sections are detected based on the section name, so it should be
4099 This directive is only supported by a few object file formats; as of this
4100 writing, the only object file format which supports it is the Portable
4101 Executable format used on Windows NT.
4103 The @var{type} argument is optional. If specified, it must be one of the
4104 following strings. For example:
4108 Not all types may be supported on all object file formats.
4112 Silently discard duplicate sections. This is the default.
4115 Warn if there are duplicate sections, but still keep only one copy.
4118 Warn if any of the duplicates have different sizes.
4121 Warn if any of the duplicates do not have exactly the same contents.
4125 @section @code{.ln @var{line-number}}
4127 @cindex @code{ln} directive
4128 @ifclear no-line-dir
4129 @samp{.ln} is a synonym for @samp{.line}.
4132 Tell @code{@value{AS}} to change the logical line number. @var{line-number}
4133 must be an absolute expression. The next line has that logical
4134 line number, so any other statements on the current line (after a
4135 statement separator character @code{;}) are reported as on logical
4136 line number @var{line-number} @minus{} 1.
4139 This directive is accepted, but ignored, when @code{@value{AS}} is
4140 configured for @code{b.out}; its effect is only associated with COFF
4146 @section @code{.mri @var{val}}
4148 @cindex @code{mri} directive
4149 @cindex MRI mode, temporarily
4150 If @var{val} is non-zero, this tells @code{@value{AS}} to enter MRI mode. If
4151 @var{val} is zero, this tells @code{@value{AS}} to exit MRI mode. This change
4152 affects code assembled until the next @code{.mri} directive, or until the end
4153 of the file. @xref{M, MRI mode, MRI mode}.
4156 @section @code{.list}
4158 @cindex @code{list} directive
4159 @cindex listing control, turning on
4160 Control (in conjunction with the @code{.nolist} directive) whether or
4161 not assembly listings are generated. These two directives maintain an
4162 internal counter (which is zero initially). @code{.list} increments the
4163 counter, and @code{.nolist} decrements it. Assembly listings are
4164 generated whenever the counter is greater than zero.
4166 By default, listings are disabled. When you enable them (with the
4167 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4168 the initial value of the listing counter is one.
4171 @section @code{.long @var{expressions}}
4173 @cindex @code{long} directive
4174 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
4177 @c no one seems to know what this is for or whether this description is
4178 @c what it really ought to do
4180 @section @code{.lsym @var{symbol}, @var{expression}}
4182 @cindex @code{lsym} directive
4183 @cindex symbol, not referenced in assembly
4184 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
4185 the hash table, ensuring it cannot be referenced by name during the
4186 rest of the assembly. This sets the attributes of the symbol to be
4187 the same as the expression value:
4189 @var{other} = @var{descriptor} = 0
4190 @var{type} = @r{(section of @var{expression})}
4191 @var{value} = @var{expression}
4194 The new symbol is not flagged as external.
4198 @section @code{.macro}
4201 The commands @code{.macro} and @code{.endm} allow you to define macros that
4202 generate assembly output. For example, this definition specifies a macro
4203 @code{sum} that puts a sequence of numbers into memory:
4206 .macro sum from=0, to=5
4215 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
4227 @item .macro @var{macname}
4228 @itemx .macro @var{macname} @var{macargs} @dots{}
4229 @cindex @code{macro} directive
4230 Begin the definition of a macro called @var{macname}. If your macro
4231 definition requires arguments, specify their names after the macro name,
4232 separated by commas or spaces. You can supply a default value for any
4233 macro argument by following the name with @samp{=@var{deflt}}. For
4234 example, these are all valid @code{.macro} statements:
4238 Begin the definition of a macro called @code{comm}, which takes no
4241 @item .macro plus1 p, p1
4242 @itemx .macro plus1 p p1
4243 Either statement begins the definition of a macro called @code{plus1},
4244 which takes two arguments; within the macro definition, write
4245 @samp{\p} or @samp{\p1} to evaluate the arguments.
4247 @item .macro reserve_str p1=0 p2
4248 Begin the definition of a macro called @code{reserve_str}, with two
4249 arguments. The first argument has a default value, but not the second.
4250 After the definition is complete, you can call the macro either as
4251 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
4252 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
4253 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
4254 @samp{0}, and @samp{\p2} evaluating to @var{b}).
4257 When you call a macro, you can specify the argument values either by
4258 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
4259 @samp{sum to=17, from=9}.
4262 @cindex @code{endm} directive
4263 Mark the end of a macro definition.
4266 @cindex @code{exitm} directive
4267 Exit early from the current macro definition.
4269 @cindex number of macros executed
4270 @cindex macros, count executed
4272 @code{@value{AS}} maintains a counter of how many macros it has
4273 executed in this pseudo-variable; you can copy that number to your
4274 output with @samp{\@@}, but @emph{only within a macro definition}.
4277 @item LOCAL @var{name} [ , @dots{} ]
4278 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
4279 macro syntax'' with @samp{-a} or @samp{--alternate}.} @xref{Alternate,,
4280 Alternate macro syntax}.
4282 Generate a string replacement for each of the @var{name} arguments, and
4283 replace any instances of @var{name} in each macro expansion. The
4284 replacement string is unique in the assembly, and different for each
4285 separate macro expansion. @code{LOCAL} allows you to write macros that
4286 define symbols, without fear of conflict between separate macro expansions.
4291 @section @code{.nolist}
4293 @cindex @code{nolist} directive
4294 @cindex listing control, turning off
4295 Control (in conjunction with the @code{.list} directive) whether or
4296 not assembly listings are generated. These two directives maintain an
4297 internal counter (which is zero initially). @code{.list} increments the
4298 counter, and @code{.nolist} decrements it. Assembly listings are
4299 generated whenever the counter is greater than zero.
4302 @section @code{.octa @var{bignums}}
4304 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
4305 @cindex @code{octa} directive
4306 @cindex integer, 16-byte
4307 @cindex sixteen byte integer
4308 This directive expects zero or more bignums, separated by commas. For each
4309 bignum, it emits a 16-byte integer.
4311 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
4312 hence @emph{octa}-word for 16 bytes.
4315 @section @code{.org @var{new-lc} , @var{fill}}
4317 @cindex @code{org} directive
4318 @cindex location counter, advancing
4319 @cindex advancing location counter
4320 @cindex current address, advancing
4321 Advance the location counter of the current section to
4322 @var{new-lc}. @var{new-lc} is either an absolute expression or an
4323 expression with the same section as the current subsection. That is,
4324 you can't use @code{.org} to cross sections: if @var{new-lc} has the
4325 wrong section, the @code{.org} directive is ignored. To be compatible
4326 with former assemblers, if the section of @var{new-lc} is absolute,
4327 @code{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
4328 is the same as the current subsection.
4330 @code{.org} may only increase the location counter, or leave it
4331 unchanged; you cannot use @code{.org} to move the location counter
4334 @c double negative used below "not undefined" because this is a specific
4335 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
4336 @c section. doc@cygnus.com 18feb91
4337 Because @code{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
4338 may not be undefined. If you really detest this restriction we eagerly await
4339 a chance to share your improved assembler.
4341 Beware that the origin is relative to the start of the section, not
4342 to the start of the subsection. This is compatible with other
4343 people's assemblers.
4345 When the location counter (of the current subsection) is advanced, the
4346 intervening bytes are filled with @var{fill} which should be an
4347 absolute expression. If the comma and @var{fill} are omitted,
4348 @var{fill} defaults to zero.
4351 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4353 @cindex padding the location counter given a power of two
4354 @cindex @code{p2align} directive
4355 Pad the location counter (in the current subsection) to a particular
4356 storage boundary. The first expression (which must be absolute) is the
4357 number of low-order zero bits the location counter must have after
4358 advancement. For example @samp{.p2align 3} advances the location
4359 counter until it a multiple of 8. If the location counter is already a
4360 multiple of 8, no change is needed.
4362 The second expression (also absolute) gives the fill value to be stored in the
4363 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4364 padding bytes are normally zero. However, on some systems, if the section is
4365 marked as containing code and the fill value is omitted, the space is filled
4366 with no-op instructions.
4368 The third expression is also absolute, and is also optional. If it is present,
4369 it is the maximum number of bytes that should be skipped by this alignment
4370 directive. If doing the alignment would require skipping more bytes than the
4371 specified maximum, then the alignment is not done at all. You can omit the
4372 fill value (the second argument) entirely by simply using two commas after the
4373 required alignment; this can be useful if you want the alignment to be filled
4374 with no-op instructions when appropriate.
4376 @cindex @code{p2alignw} directive
4377 @cindex @code{p2alignl} directive
4378 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
4379 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
4380 pattern as a two byte word value. The @code{.p2alignl} directives treats the
4381 fill pattern as a four byte longword value. For example, @code{.p2alignw
4382 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4383 filled in with the value 0x368d (the exact placement of the bytes depends upon
4384 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4389 @section @code{.previous}
4391 @cindex @code{.previous} directive
4392 @cindex Section Stack
4393 This is one of the ELF section stack manipulation directives. The others are
4394 @pxref{Section}, @xref{SubSection}, @pxref{PushSection}, and
4397 This directive swaps the current section (and subsection) with most recently
4398 referenced section (and subsection) prior to this one. Multiple
4399 @code{.previous} directives in a row will flip between two sections (and their
4402 In terms of the section stack, this directive swaps the current section with
4403 the top section on the section stack.
4408 @section @code{.popsection}
4410 @cindex @code{.popsection} directive
4411 @cindex Section Stack
4412 This is one of the ELF section stack manipulation directives. The others are
4413 @pxref{Section}, @xref{SubSection}, @pxref{PushSection}, and
4416 This directive replaces the current section (and subsection) with the top
4417 section (and subsection) on the section stack. This section is popped off the
4422 @section @code{.print @var{string}}
4424 @cindex @code{print} directive
4425 @code{@value{AS}} will print @var{string} on the standard output during
4426 assembly. You must put @var{string} in double quotes.
4430 @section @code{.protected @var{names}}
4432 @cindex @code{.protected} directive
4434 This one of the ELF visibility directives. The other two are
4435 @pxref{Hidden} and @pxref{Internal}
4437 This directive overrides the named symbols default visibility (which is set by
4438 their binding: local, global or weak). The directive sets the visibility to
4439 @code{protected} which means that any references to the symbols from within the
4440 components that defines them must be resolved to the definition in that
4441 component, even if a definition in another component would normally preempt
4446 @section @code{.psize @var{lines} , @var{columns}}
4448 @cindex @code{psize} directive
4449 @cindex listing control: paper size
4450 @cindex paper size, for listings
4451 Use this directive to declare the number of lines---and, optionally, the
4452 number of columns---to use for each page, when generating listings.
4454 If you do not use @code{.psize}, listings use a default line-count
4455 of 60. You may omit the comma and @var{columns} specification; the
4456 default width is 200 columns.
4458 @code{@value{AS}} generates formfeeds whenever the specified number of
4459 lines is exceeded (or whenever you explicitly request one, using
4462 If you specify @var{lines} as @code{0}, no formfeeds are generated save
4463 those explicitly specified with @code{.eject}.
4466 @section @code{.purgem @var{name}}
4468 @cindex @code{purgem} directive
4469 Undefine the macro @var{name}, so that later uses of the string will not be
4470 expanded. @xref{Macro}.
4474 @section @code{.pushsection @var{name} , @var{subsection}}
4476 @cindex @code{.pushsection} directive
4477 @cindex Section Stack
4478 This is one of the ELF section stack manipulation directives. The others are
4479 @pxref{Section}, @xref{SubSection}, @pxref{PopSection}, and
4482 This directive is a synonym for @code{.section}. It psuhes the current section
4483 (and subsection) onto the top of the section stack, and then replaces the
4484 current section and subsection with @code{name} and @code{subsection}.
4488 @section @code{.quad @var{bignums}}
4490 @cindex @code{quad} directive
4491 @code{.quad} expects zero or more bignums, separated by commas. For
4492 each bignum, it emits
4494 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
4495 warning message; and just takes the lowest order 8 bytes of the bignum.
4496 @cindex eight-byte integer
4497 @cindex integer, 8-byte
4499 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
4500 hence @emph{quad}-word for 8 bytes.
4503 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
4504 warning message; and just takes the lowest order 16 bytes of the bignum.
4505 @cindex sixteen-byte integer
4506 @cindex integer, 16-byte
4510 @section @code{.rept @var{count}}
4512 @cindex @code{rept} directive
4513 Repeat the sequence of lines between the @code{.rept} directive and the next
4514 @code{.endr} directive @var{count} times.
4516 For example, assembling
4524 is equivalent to assembling
4533 @section @code{.sbttl "@var{subheading}"}
4535 @cindex @code{sbttl} directive
4536 @cindex subtitles for listings
4537 @cindex listing control: subtitle
4538 Use @var{subheading} as the title (third line, immediately after the
4539 title line) when generating assembly listings.
4541 This directive affects subsequent pages, as well as the current page if
4542 it appears within ten lines of the top of a page.
4546 @section @code{.scl @var{class}}
4548 @cindex @code{scl} directive
4549 @cindex symbol storage class (COFF)
4550 @cindex COFF symbol storage class
4551 Set the storage-class value for a symbol. This directive may only be
4552 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
4553 whether a symbol is static or external, or it may record further
4554 symbolic debugging information.
4557 The @samp{.scl} directive is primarily associated with COFF output; when
4558 configured to generate @code{b.out} output format, @code{@value{AS}}
4559 accepts this directive but ignores it.
4564 @section @code{.section @var{name}} (COFF version)
4566 @cindex @code{section} directive
4567 @cindex named section
4568 Use the @code{.section} directive to assemble the following code into a section
4571 This directive is only supported for targets that actually support arbitrarily
4572 named sections; on @code{a.out} targets, for example, it is not accepted, even
4573 with a standard @code{a.out} section name.
4575 For COFF targets, the @code{.section} directive is used in one of the following
4579 .section @var{name}[, "@var{flags}"]
4580 .section @var{name}[, @var{subsegment}]
4583 If the optional argument is quoted, it is taken as flags to use for the
4584 section. Each flag is a single character. The following flags are recognized:
4587 bss section (uninitialized data)
4589 section is not loaded
4599 shared section (meaningful for PE targets)
4602 If no flags are specified, the default flags depend upon the section name. If
4603 the section name is not recognized, the default will be for the section to be
4604 loaded and writable.
4606 If the optional argument to the @code{.section} directive is not quoted, it is
4607 taken as a subsegment number (@pxref{Sub-Sections}).
4610 @section @code{.section @var{name}} (ELF version)
4612 @cindex @code{section} directive
4613 @cindex named section
4615 @cindex Section Stack
4616 This is one of the ELF section stack manipulation directives. The others are
4617 @xref{SubSection}, @pxref{PushSection}@pxref{PopSection}, and
4621 For ELF targets, the @code{.section} directive is used like this:
4624 .section @var{name} [, "@var{flags}"[, @@@var{type}]]
4627 The optional @var{flags} argument is a quoted string which may contain any
4628 combintion of the following characters:
4631 section is allocatable
4635 section is executable
4638 The optional @var{type} argument may contain one of the following constants:
4641 section contains data
4643 section does not contain data (i.e., section only occupies space)
4646 If no flags are specified, the default flags depend upon the section name. If
4647 the section name is not recognized, the default will be for the section to have
4648 none of the above flags: it will not be allocated in memory, nor writable, nor
4649 executable. The section will contain data.
4651 For ELF targets, the assembler supports another type of @code{.section}
4652 directive for compatibility with the Solaris assembler:
4655 .section "@var{name}"[, @var{flags}...]
4658 Note that the section name is quoted. There may be a sequence of comma
4662 section is allocatable
4666 section is executable
4669 This directive replaces the current section and subsection. The replaced
4670 section and subsection are pushed onto the section stack. See the contents of
4671 the gas testsuite directory @code{gas/testsuite/gas/elf} for some examples of
4672 how this directive and the other section stack directives work.
4675 @section @code{.set @var{symbol}, @var{expression}}
4677 @cindex @code{set} directive
4678 @cindex symbol value, setting
4679 Set the value of @var{symbol} to @var{expression}. This
4680 changes @var{symbol}'s value and type to conform to
4681 @var{expression}. If @var{symbol} was flagged as external, it remains
4682 flagged (@pxref{Symbol Attributes}).
4684 You may @code{.set} a symbol many times in the same assembly.
4686 If you @code{.set} a global symbol, the value stored in the object
4687 file is the last value stored into it.
4690 The syntax for @code{set} on the HPPA is
4691 @samp{@var{symbol} .set @var{expression}}.
4695 @section @code{.short @var{expressions}}
4697 @cindex @code{short} directive
4699 @code{.short} is normally the same as @samp{.word}.
4700 @xref{Word,,@code{.word}}.
4702 In some configurations, however, @code{.short} and @code{.word} generate
4703 numbers of different lengths; @pxref{Machine Dependencies}.
4707 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
4710 This expects zero or more @var{expressions}, and emits
4711 a 16 bit number for each.
4716 @section @code{.single @var{flonums}}
4718 @cindex @code{single} directive
4719 @cindex floating point numbers (single)
4720 This directive assembles zero or more flonums, separated by commas. It
4721 has the same effect as @code{.float}.
4723 The exact kind of floating point numbers emitted depends on how
4724 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
4728 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
4729 numbers in @sc{ieee} format.
4734 @section @code{.size} (COFF version)
4736 @cindex @code{size} directive
4737 This directive is generated by compilers to include auxiliary debugging
4738 information in the symbol table. It is only permitted inside
4739 @code{.def}/@code{.endef} pairs.
4742 @samp{.size} is only meaningful when generating COFF format output; when
4743 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4747 @section @code{.size @var{name} , @var{expression}} (ELF version)
4748 @cindex @code{size} directive
4750 This directive is used to set the size associated with a symbol @var{name}.
4751 The size in bytes is computed from @var{expression} which can make use of label
4752 arithmetic. This directive is typically used to set the size of function
4756 @section @code{.sleb128 @var{expressions}}
4758 @cindex @code{sleb128} directive
4759 @var{sleb128} stands for ``signed little endian base 128.'' This is a
4760 compact, variable length representation of numbers used by the DWARF
4761 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
4763 @ifclear no-space-dir
4765 @section @code{.skip @var{size} , @var{fill}}
4767 @cindex @code{skip} directive
4768 @cindex filling memory
4769 This directive emits @var{size} bytes, each of value @var{fill}. Both
4770 @var{size} and @var{fill} are absolute expressions. If the comma and
4771 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
4775 @section @code{.space @var{size} , @var{fill}}
4777 @cindex @code{space} directive
4778 @cindex filling memory
4779 This directive emits @var{size} bytes, each of value @var{fill}. Both
4780 @var{size} and @var{fill} are absolute expressions. If the comma
4781 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
4786 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
4787 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
4788 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
4789 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
4798 @section @code{.space}
4799 @cindex @code{space} directive
4801 On the AMD 29K, this directive is ignored; it is accepted for
4802 compatibility with other AMD 29K assemblers.
4805 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
4806 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
4812 @section @code{.stabd, .stabn, .stabs}
4814 @cindex symbolic debuggers, information for
4815 @cindex @code{stab@var{x}} directives
4816 There are three directives that begin @samp{.stab}.
4817 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
4818 The symbols are not entered in the @code{@value{AS}} hash table: they
4819 cannot be referenced elsewhere in the source file.
4820 Up to five fields are required:
4824 This is the symbol's name. It may contain any character except
4825 @samp{\000}, so is more general than ordinary symbol names. Some
4826 debuggers used to code arbitrarily complex structures into symbol names
4830 An absolute expression. The symbol's type is set to the low 8 bits of
4831 this expression. Any bit pattern is permitted, but @code{@value{LD}}
4832 and debuggers choke on silly bit patterns.
4835 An absolute expression. The symbol's ``other'' attribute is set to the
4836 low 8 bits of this expression.
4839 An absolute expression. The symbol's descriptor is set to the low 16
4840 bits of this expression.
4843 An absolute expression which becomes the symbol's value.
4846 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
4847 or @code{.stabs} statement, the symbol has probably already been created;
4848 you get a half-formed symbol in your object file. This is
4849 compatible with earlier assemblers!
4852 @cindex @code{stabd} directive
4853 @item .stabd @var{type} , @var{other} , @var{desc}
4855 The ``name'' of the symbol generated is not even an empty string.
4856 It is a null pointer, for compatibility. Older assemblers used a
4857 null pointer so they didn't waste space in object files with empty
4860 The symbol's value is set to the location counter,
4861 relocatably. When your program is linked, the value of this symbol
4862 is the address of the location counter when the @code{.stabd} was
4865 @cindex @code{stabn} directive
4866 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
4867 The name of the symbol is set to the empty string @code{""}.
4869 @cindex @code{stabs} directive
4870 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
4871 All five fields are specified.
4877 @section @code{.string} "@var{str}"
4879 @cindex string, copying to object file
4880 @cindex @code{string} directive
4882 Copy the characters in @var{str} to the object file. You may specify more than
4883 one string to copy, separated by commas. Unless otherwise specified for a
4884 particular machine, the assembler marks the end of each string with a 0 byte.
4885 You can use any of the escape sequences described in @ref{Strings,,Strings}.
4888 @section @code{.struct @var{expression}}
4890 @cindex @code{struct} directive
4891 Switch to the absolute section, and set the section offset to @var{expression},
4892 which must be an absolute expression. You might use this as follows:
4901 This would define the symbol @code{field1} to have the value 0, the symbol
4902 @code{field2} to have the value 4, and the symbol @code{field3} to have the
4903 value 8. Assembly would be left in the absolute section, and you would need to
4904 use a @code{.section} directive of some sort to change to some other section
4905 before further assembly.
4909 @section @code{.subsection @var{name}}
4911 @cindex @code{.subsection} directive
4912 @cindex Section Stack
4913 This is one of the ELF section stack manipulation directives. The others are
4914 @pxref{Section}, @xref{PushSection}, @pxref{PopSection}, and
4917 This directive replaces the current subsection with @code{name}. The current
4918 section is not changed. The replaced subsection is put onto the section stack
4919 in place of the then current top of stack subsection.
4924 @section @code{.symver}
4925 @cindex @code{symver} directive
4926 @cindex symbol versioning
4927 @cindex versions of symbols
4928 Use the @code{.symver} directive to bind symbols to specific version nodes
4929 within a source file. This is only supported on ELF platforms, and is
4930 typically used when assembling files to be linked into a shared library.
4931 There are cases where it may make sense to use this in objects to be bound
4932 into an application itself so as to override a versioned symbol from a
4935 For ELF targets, the @code{.symver} directive can be used like this:
4937 .symver @var{name}, @var{name2@@nodename}
4939 If the symbol @var{name} is defined within the file
4940 being assembled, the @code{.symver} directive effectively creates a symbol
4941 alias with the name @var{name2@@nodename}, and in fact the main reason that we
4942 just don't try and create a regular alias is that the @var{@@} character isn't
4943 permitted in symbol names. The @var{name2} part of the name is the actual name
4944 of the symbol by which it will be externally referenced. The name @var{name}
4945 itself is merely a name of convenience that is used so that it is possible to
4946 have definitions for multiple versions of a function within a single source
4947 file, and so that the compiler can unambiguously know which version of a
4948 function is being mentioned. The @var{nodename} portion of the alias should be
4949 the name of a node specified in the version script supplied to the linker when
4950 building a shared library. If you are attempting to override a versioned
4951 symbol from a shared library, then @var{nodename} should correspond to the
4952 nodename of the symbol you are trying to override.
4954 If the symbol @var{name} is not defined within the file being assembled, all
4955 references to @var{name} will be changed to @var{name2@@nodename}. If no
4956 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
4959 Another usage of the @code{.symver} directive is:
4961 .symver @var{name}, @var{name2@@@@nodename}
4963 In this case, the symbol @var{name} must exist and be defined within
4964 the file being assembled. It is similiar to @var{name2@@nodename}. The
4965 difference is @var{name2@@@@nodename} will also be used to resolve
4966 references to @var{name2} by the linker.
4968 The third usage of the @code{.symver} directive is:
4970 .symver @var{name}, @var{name2@@@@@@nodename}
4972 When @var{name} is not defined within the
4973 file being assembled, it is treated as @var{name2@@nodename}. When
4974 @var{name} is defined within the file being assembled, the symbol
4975 name, @var{name}, will be changed to @var{name2@@@@nodename}.
4980 @section @code{.tag @var{structname}}
4982 @cindex COFF structure debugging
4983 @cindex structure debugging, COFF
4984 @cindex @code{tag} directive
4985 This directive is generated by compilers to include auxiliary debugging
4986 information in the symbol table. It is only permitted inside
4987 @code{.def}/@code{.endef} pairs. Tags are used to link structure
4988 definitions in the symbol table with instances of those structures.
4991 @samp{.tag} is only used when generating COFF format output; when
4992 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4998 @section @code{.text @var{subsection}}
5000 @cindex @code{text} directive
5001 Tells @code{@value{AS}} to assemble the following statements onto the end of
5002 the text subsection numbered @var{subsection}, which is an absolute
5003 expression. If @var{subsection} is omitted, subsection number zero
5007 @section @code{.title "@var{heading}"}
5009 @cindex @code{title} directive
5010 @cindex listing control: title line
5011 Use @var{heading} as the title (second line, immediately after the
5012 source file name and pagenumber) when generating assembly listings.
5014 This directive affects subsequent pages, as well as the current page if
5015 it appears within ten lines of the top of a page.
5018 @section @code{.type @var{int}} (COFF version)
5020 @cindex COFF symbol type
5021 @cindex symbol type, COFF
5022 @cindex @code{type} directive
5023 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5024 records the integer @var{int} as the type attribute of a symbol table entry.
5027 @samp{.type} is associated only with COFF format output; when
5028 @code{@value{AS}} is configured for @code{b.out} output, it accepts this
5029 directive but ignores it.
5032 @section @code{.type @var{name} , @var{type description}} (ELF version)
5034 @cindex ELF symbol type
5035 @cindex symbol type, ELF
5036 @cindex @code{type} directive
5037 This directive is used to set the type of symbol @var{name} to be either a
5038 function symbol or an ojbect symbol. There are five different syntaxes
5039 supported for the @var{type description} field, in order to provide
5040 comptability with various other assemblers. The syntaxes supported are:
5043 .type <name>,#function
5044 .type <name>,#object
5046 .type <name>,@@function
5047 .type <name>,@@object
5049 .type <name>,%function
5050 .type <name>,%object
5052 .type <name>,"function"
5053 .type <name>,"object"
5055 .type <name> STT_FUNCTION
5056 .type <name> STT_OBJECT
5060 @section @code{.uleb128 @var{expressions}}
5062 @cindex @code{uleb128} directive
5063 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
5064 compact, variable length representation of numbers used by the DWARF
5065 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
5069 @section @code{.val @var{addr}}
5071 @cindex @code{val} directive
5072 @cindex COFF value attribute
5073 @cindex value attribute, COFF
5074 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5075 records the address @var{addr} as the value attribute of a symbol table
5079 @samp{.val} is used only for COFF output; when @code{@value{AS}} is
5080 configured for @code{b.out}, it accepts this directive but ignores it.
5086 @section @code{.version "@var{string}"}
5088 @cindex @code{.version}
5089 This directive creates a @code{.note} section and places into it an ELF
5090 formatted note of type NT_VERSION. The note's name is set to @code{string}.
5095 @section @code{.vtable_entry @var{table}, @var{offset}}
5097 @cindex @code{.vtable_entry}
5098 This directive finds or creates a symbol @code{table} and creates a
5099 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
5102 @section @code{.vtable_inherit @var{child}, @var{parent}}
5104 @cindex @code{.vtable_inherit}
5105 This directive finds the symbol @code{child} and finds or creates the symbol
5106 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
5107 parent whoes addend is the value of the child symbol. As a special case the
5108 parent name of @code{0} is treated as refering the @code{*ABS*} section.
5113 @section @code{.weak @var{names}}
5115 @cindex @code{.weak}
5116 This directive sets the weak attribute on the comma seperated list of symbol
5117 @code{names}. If the symbols do not already exist, they will be created.
5121 @section @code{.word @var{expressions}}
5123 @cindex @code{word} directive
5124 This directive expects zero or more @var{expressions}, of any section,
5125 separated by commas.
5128 For each expression, @code{@value{AS}} emits a 32-bit number.
5131 For each expression, @code{@value{AS}} emits a 16-bit number.
5136 The size of the number emitted, and its byte order,
5137 depend on what target computer the assembly is for.
5140 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
5141 @c happen---32-bit addressability, period; no long/short jumps.
5142 @ifset DIFF-TBL-KLUGE
5143 @cindex difference tables altered
5144 @cindex altered difference tables
5146 @emph{Warning: Special Treatment to support Compilers}
5150 Machines with a 32-bit address space, but that do less than 32-bit
5151 addressing, require the following special treatment. If the machine of
5152 interest to you does 32-bit addressing (or doesn't require it;
5153 @pxref{Machine Dependencies}), you can ignore this issue.
5156 In order to assemble compiler output into something that works,
5157 @code{@value{AS}} occasionlly does strange things to @samp{.word} directives.
5158 Directives of the form @samp{.word sym1-sym2} are often emitted by
5159 compilers as part of jump tables. Therefore, when @code{@value{AS}} assembles a
5160 directive of the form @samp{.word sym1-sym2}, and the difference between
5161 @code{sym1} and @code{sym2} does not fit in 16 bits, @code{@value{AS}}
5162 creates a @dfn{secondary jump table}, immediately before the next label.
5163 This secondary jump table is preceded by a short-jump to the
5164 first byte after the secondary table. This short-jump prevents the flow
5165 of control from accidentally falling into the new table. Inside the
5166 table is a long-jump to @code{sym2}. The original @samp{.word}
5167 contains @code{sym1} minus the address of the long-jump to
5170 If there were several occurrences of @samp{.word sym1-sym2} before the
5171 secondary jump table, all of them are adjusted. If there was a
5172 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
5173 long-jump to @code{sym4} is included in the secondary jump table,
5174 and the @code{.word} directives are adjusted to contain @code{sym3}
5175 minus the address of the long-jump to @code{sym4}; and so on, for as many
5176 entries in the original jump table as necessary.
5179 @emph{This feature may be disabled by compiling @code{@value{AS}} with the
5180 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
5181 assembly language programmers.
5184 @c end DIFF-TBL-KLUGE
5187 @section Deprecated Directives
5189 @cindex deprecated directives
5190 @cindex obsolescent directives
5191 One day these directives won't work.
5192 They are included for compatibility with older assemblers.
5199 @node Machine Dependencies
5200 @chapter Machine Dependent Features
5202 @cindex machine dependencies
5203 The machine instruction sets are (almost by definition) different on
5204 each machine where @code{@value{AS}} runs. Floating point representations
5205 vary as well, and @code{@value{AS}} often supports a few additional
5206 directives or command-line options for compatibility with other
5207 assemblers on a particular platform. Finally, some versions of
5208 @code{@value{AS}} support special pseudo-instructions for branch
5211 This chapter discusses most of these differences, though it does not
5212 include details on any machine's instruction set. For details on that
5213 subject, see the hardware manufacturer's manual.
5217 * AMD29K-Dependent:: AMD 29K Dependent Features
5220 * ARC-Dependent:: ARC Dependent Features
5223 * ARM-Dependent:: ARM Dependent Features
5226 * D10V-Dependent:: D10V Dependent Features
5229 * D30V-Dependent:: D30V Dependent Features
5232 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
5235 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
5238 * HPPA-Dependent:: HPPA Dependent Features
5241 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
5244 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
5247 * i860-Dependent:: Intel 80860 Dependent Features
5250 * i960-Dependent:: Intel 80960 Dependent Features
5253 * M32R-Dependent:: M32R Dependent Features
5256 * M68K-Dependent:: M680x0 Dependent Features
5259 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
5262 * MIPS-Dependent:: MIPS Dependent Features
5265 * SH-Dependent:: Hitachi SH Dependent Features
5268 * PJ-Dependent:: picoJava Dependent Features
5271 * Sparc-Dependent:: SPARC Dependent Features
5274 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
5277 * V850-Dependent:: V850 Dependent Features
5280 * Z8000-Dependent:: Z8000 Dependent Features
5283 * Vax-Dependent:: VAX Dependent Features
5290 @c The following major nodes are *sections* in the GENERIC version, *chapters*
5291 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
5292 @c peculiarity: to preserve cross-references, there must be a node called
5293 @c "Machine Dependencies". Hence the conditional nodenames in each
5294 @c major node below. Node defaulting in makeinfo requires adjacency of
5295 @c node and sectioning commands; hence the repetition of @chapter BLAH
5296 @c in both conditional blocks.
5303 @include c-a29k.texi
5312 @node Machine Dependencies
5313 @chapter Machine Dependent Features
5315 The machine instruction sets are different on each Hitachi chip family,
5316 and there are also some syntax differences among the families. This
5317 chapter describes the specific @code{@value{AS}} features for each
5321 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
5322 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
5323 * SH-Dependent:: Hitachi SH Dependent Features
5330 @include c-d10v.texi
5334 @include c-d30v.texi
5338 @include c-h8300.texi
5342 @include c-h8500.texi
5346 @include c-hppa.texi
5350 @include c-i370.texi
5354 @include c-i386.texi
5358 @include c-i860.texi
5362 @include c-i960.texi
5366 @include c-m32r.texi
5370 @include c-m68k.texi
5374 @include c-m68hc11.texi
5378 @include c-mips.texi
5382 @include c-ns32k.texi
5394 @include c-sparc.texi
5398 @include c-tic54x.texi
5410 @include c-v850.texi
5414 @c reverse effect of @down at top of generic Machine-Dep chapter
5418 @node Reporting Bugs
5419 @chapter Reporting Bugs
5420 @cindex bugs in assembler
5421 @cindex reporting bugs in assembler
5423 Your bug reports play an essential role in making @code{@value{AS}} reliable.
5425 Reporting a bug may help you by bringing a solution to your problem, or it may
5426 not. But in any case the principal function of a bug report is to help the
5427 entire community by making the next version of @code{@value{AS}} work better.
5428 Bug reports are your contribution to the maintenance of @code{@value{AS}}.
5430 In order for a bug report to serve its purpose, you must include the
5431 information that enables us to fix the bug.
5434 * Bug Criteria:: Have you found a bug?
5435 * Bug Reporting:: How to report bugs
5439 @section Have you found a bug?
5440 @cindex bug criteria
5442 If you are not sure whether you have found a bug, here are some guidelines:
5445 @cindex fatal signal
5446 @cindex assembler crash
5447 @cindex crash of assembler
5449 If the assembler gets a fatal signal, for any input whatever, that is a
5450 @code{@value{AS}} bug. Reliable assemblers never crash.
5452 @cindex error on valid input
5454 If @code{@value{AS}} produces an error message for valid input, that is a bug.
5456 @cindex invalid input
5458 If @code{@value{AS}} does not produce an error message for invalid input, that
5459 is a bug. However, you should note that your idea of ``invalid input'' might
5460 be our idea of ``an extension'' or ``support for traditional practice''.
5463 If you are an experienced user of assemblers, your suggestions for improvement
5464 of @code{@value{AS}} are welcome in any case.
5468 @section How to report bugs
5470 @cindex assembler bugs, reporting
5472 A number of companies and individuals offer support for @sc{gnu} products. If
5473 you obtained @code{@value{AS}} from a support organization, we recommend you
5474 contact that organization first.
5476 You can find contact information for many support companies and
5477 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
5480 In any event, we also recommend that you send bug reports for @code{@value{AS}}
5481 to @samp{bug-binutils@@gnu.org}.
5483 The fundamental principle of reporting bugs usefully is this:
5484 @strong{report all the facts}. If you are not sure whether to state a
5485 fact or leave it out, state it!
5487 Often people omit facts because they think they know what causes the problem
5488 and assume that some details do not matter. Thus, you might assume that the
5489 name of a symbol you use in an example does not matter. Well, probably it does
5490 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
5491 happens to fetch from the location where that name is stored in memory;
5492 perhaps, if the name were different, the contents of that location would fool
5493 the assembler into doing the right thing despite the bug. Play it safe and
5494 give a specific, complete example. That is the easiest thing for you to do,
5495 and the most helpful.
5497 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
5498 it is new to us. Therefore, always write your bug reports on the assumption
5499 that the bug has not been reported previously.
5501 Sometimes people give a few sketchy facts and ask, ``Does this ring a
5502 bell?'' Those bug reports are useless, and we urge everyone to
5503 @emph{refuse to respond to them} except to chide the sender to report
5506 To enable us to fix the bug, you should include all these things:
5510 The version of @code{@value{AS}}. @code{@value{AS}} announces it if you start
5511 it with the @samp{--version} argument.
5513 Without this, we will not know whether there is any point in looking for
5514 the bug in the current version of @code{@value{AS}}.
5517 Any patches you may have applied to the @code{@value{AS}} source.
5520 The type of machine you are using, and the operating system name and
5524 What compiler (and its version) was used to compile @code{@value{AS}}---e.g.
5528 The command arguments you gave the assembler to assemble your example and
5529 observe the bug. To guarantee you will not omit something important, list them
5530 all. A copy of the Makefile (or the output from make) is sufficient.
5532 If we were to try to guess the arguments, we would probably guess wrong
5533 and then we might not encounter the bug.
5536 A complete input file that will reproduce the bug. If the bug is observed when
5537 the assembler is invoked via a compiler, send the assembler source, not the
5538 high level language source. Most compilers will produce the assembler source
5539 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
5540 the options @samp{-v --save-temps}; this will save the assembler source in a
5541 file with an extension of @file{.s}, and also show you exactly how
5542 @code{@value{AS}} is being run.
5545 A description of what behavior you observe that you believe is
5546 incorrect. For example, ``It gets a fatal signal.''
5548 Of course, if the bug is that @code{@value{AS}} gets a fatal signal, then we
5549 will certainly notice it. But if the bug is incorrect output, we might not
5550 notice unless it is glaringly wrong. You might as well not give us a chance to
5553 Even if the problem you experience is a fatal signal, you should still say so
5554 explicitly. Suppose something strange is going on, such as, your copy of
5555 @code{@value{AS}} is out of synch, or you have encountered a bug in the C
5556 library on your system. (This has happened!) Your copy might crash and ours
5557 would not. If you told us to expect a crash, then when ours fails to crash, we
5558 would know that the bug was not happening for us. If you had not told us to
5559 expect a crash, then we would not be able to draw any conclusion from our
5563 If you wish to suggest changes to the @code{@value{AS}} source, send us context
5564 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
5565 option. Always send diffs from the old file to the new file. If you even
5566 discuss something in the @code{@value{AS}} source, refer to it by context, not
5569 The line numbers in our development sources will not match those in your
5570 sources. Your line numbers would convey no useful information to us.
5573 Here are some things that are not necessary:
5577 A description of the envelope of the bug.
5579 Often people who encounter a bug spend a lot of time investigating
5580 which changes to the input file will make the bug go away and which
5581 changes will not affect it.
5583 This is often time consuming and not very useful, because the way we
5584 will find the bug is by running a single example under the debugger
5585 with breakpoints, not by pure deduction from a series of examples.
5586 We recommend that you save your time for something else.
5588 Of course, if you can find a simpler example to report @emph{instead}
5589 of the original one, that is a convenience for us. Errors in the
5590 output will be easier to spot, running under the debugger will take
5591 less time, and so on.
5593 However, simplification is not vital; if you do not want to do this,
5594 report the bug anyway and send us the entire test case you used.
5597 A patch for the bug.
5599 A patch for the bug does help us if it is a good one. But do not omit
5600 the necessary information, such as the test case, on the assumption that
5601 a patch is all we need. We might see problems with your patch and decide
5602 to fix the problem another way, or we might not understand it at all.
5604 Sometimes with a program as complicated as @code{@value{AS}} it is very hard to
5605 construct an example that will make the program follow a certain path through
5606 the code. If you do not send us the example, we will not be able to construct
5607 one, so we will not be able to verify that the bug is fixed.
5609 And if we cannot understand what bug you are trying to fix, or why your
5610 patch should be an improvement, we will not install it. A test case will
5611 help us to understand.
5614 A guess about what the bug is or what it depends on.
5616 Such guesses are usually wrong. Even we cannot guess right about such
5617 things without first using the debugger to find the facts.
5620 @node Acknowledgements
5621 @chapter Acknowledgements
5623 If you have contributed to @code{@value{AS}} and your name isn't listed here,
5624 it is not meant as a slight. We just don't know about it. Send mail to the
5625 maintainer, and we'll correct the situation. Currently
5627 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
5629 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
5632 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
5633 information and the 68k series machines, most of the preprocessing pass, and
5634 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
5636 K. Richard Pixley maintained GAS for a while, adding various enhancements and
5637 many bug fixes, including merging support for several processors, breaking GAS
5638 up to handle multiple object file format back ends (including heavy rewrite,
5639 testing, an integration of the coff and b.out back ends), adding configuration
5640 including heavy testing and verification of cross assemblers and file splits
5641 and renaming, converted GAS to strictly ANSI C including full prototypes, added
5642 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
5643 port (including considerable amounts of reverse engineering), a SPARC opcode
5644 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
5645 assertions and made them work, much other reorganization, cleanup, and lint.
5647 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
5648 in format-specific I/O modules.
5650 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
5651 has done much work with it since.
5653 The Intel 80386 machine description was written by Eliot Dresselhaus.
5655 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
5657 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
5658 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
5660 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
5661 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
5662 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
5663 support a.out format.
5665 Support for the Zilog Z8k and Hitachi H8/300 and H8/500 processors (tc-z8k,
5666 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
5667 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
5668 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
5671 John Gilmore built the AMD 29000 support, added @code{.include} support, and
5672 simplified the configuration of which versions accept which directives. He
5673 updated the 68k machine description so that Motorola's opcodes always produced
5674 fixed-size instructions (e.g. @code{jsr}), while synthetic instructions
5675 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
5676 cross-compilation support, and one bug in relaxation that took a week and
5677 required the proverbial one-bit fix.
5679 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
5680 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
5681 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
5682 PowerPC assembler, and made a few other minor patches.
5684 Steve Chamberlain made @code{@value{AS}} able to generate listings.
5686 Hewlett-Packard contributed support for the HP9000/300.
5688 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
5689 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
5690 formats). This work was supported by both the Center for Software Science at
5691 the University of Utah and Cygnus Support.
5693 Support for ELF format files has been worked on by Mark Eichin of Cygnus
5694 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
5695 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
5696 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
5697 and some initial 64-bit support).
5699 Linas Vepstas added GAS support for the ESA/390 "IBM 370" architecture.
5701 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
5702 support for openVMS/Alpha.
5704 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
5707 Several engineers at Cygnus Support have also provided many small bug fixes and
5708 configuration enhancements.
5710 Many others have contributed large or small bugfixes and enhancements. If
5711 you have contributed significant work and are not mentioned on this list, and
5712 want to be, let us know. Some of the history has been lost; we are not
5713 intentionally leaving anyone out.
5715 @node GNU Free Documentation License
5716 @chapter GNU Free Documentation License
5718 GNU Free Documentation License
5720 Version 1.1, March 2000
5722 Copyright (C) 2000 Free Software Foundation, Inc.
5723 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
5725 Everyone is permitted to copy and distribute verbatim copies
5726 of this license document, but changing it is not allowed.
5731 The purpose of this License is to make a manual, textbook, or other
5732 written document "free" in the sense of freedom: to assure everyone
5733 the effective freedom to copy and redistribute it, with or without
5734 modifying it, either commercially or noncommercially. Secondarily,
5735 this License preserves for the author and publisher a way to get
5736 credit for their work, while not being considered responsible for
5737 modifications made by others.
5739 This License is a kind of "copyleft", which means that derivative
5740 works of the document must themselves be free in the same sense. It
5741 complements the GNU General Public License, which is a copyleft
5742 license designed for free software.
5744 We have designed this License in order to use it for manuals for free
5745 software, because free software needs free documentation: a free
5746 program should come with manuals providing the same freedoms that the
5747 software does. But this License is not limited to software manuals;
5748 it can be used for any textual work, regardless of subject matter or
5749 whether it is published as a printed book. We recommend this License
5750 principally for works whose purpose is instruction or reference.
5753 1. APPLICABILITY AND DEFINITIONS
5755 This License applies to any manual or other work that contains a
5756 notice placed by the copyright holder saying it can be distributed
5757 under the terms of this License. The "Document", below, refers to any
5758 such manual or work. Any member of the public is a licensee, and is
5761 A "Modified Version" of the Document means any work containing the
5762 Document or a portion of it, either copied verbatim, or with
5763 modifications and/or translated into another language.
5765 A "Secondary Section" is a named appendix or a front-matter section of
5766 the Document that deals exclusively with the relationship of the
5767 publishers or authors of the Document to the Document's overall subject
5768 (or to related matters) and contains nothing that could fall directly
5769 within that overall subject. (For example, if the Document is in part a
5770 textbook of mathematics, a Secondary Section may not explain any
5771 mathematics.) The relationship could be a matter of historical
5772 connection with the subject or with related matters, or of legal,
5773 commercial, philosophical, ethical or political position regarding
5776 The "Invariant Sections" are certain Secondary Sections whose titles
5777 are designated, as being those of Invariant Sections, in the notice
5778 that says that the Document is released under this License.
5780 The "Cover Texts" are certain short passages of text that are listed,
5781 as Front-Cover Texts or Back-Cover Texts, in the notice that says that
5782 the Document is released under this License.
5784 A "Transparent" copy of the Document means a machine-readable copy,
5785 represented in a format whose specification is available to the
5786 general public, whose contents can be viewed and edited directly and
5787 straightforwardly with generic text editors or (for images composed of
5788 pixels) generic paint programs or (for drawings) some widely available
5789 drawing editor, and that is suitable for input to text formatters or
5790 for automatic translation to a variety of formats suitable for input
5791 to text formatters. A copy made in an otherwise Transparent file
5792 format whose markup has been designed to thwart or discourage
5793 subsequent modification by readers is not Transparent. A copy that is
5794 not "Transparent" is called "Opaque".
5796 Examples of suitable formats for Transparent copies include plain
5797 ASCII without markup, Texinfo input format, LaTeX input format, SGML
5798 or XML using a publicly available DTD, and standard-conforming simple
5799 HTML designed for human modification. Opaque formats include
5800 PostScript, PDF, proprietary formats that can be read and edited only
5801 by proprietary word processors, SGML or XML for which the DTD and/or
5802 processing tools are not generally available, and the
5803 machine-generated HTML produced by some word processors for output
5806 The "Title Page" means, for a printed book, the title page itself,
5807 plus such following pages as are needed to hold, legibly, the material
5808 this License requires to appear in the title page. For works in
5809 formats which do not have any title page as such, "Title Page" means
5810 the text near the most prominent appearance of the work's title,
5811 preceding the beginning of the body of the text.
5816 You may copy and distribute the Document in any medium, either
5817 commercially or noncommercially, provided that this License, the
5818 copyright notices, and the license notice saying this License applies
5819 to the Document are reproduced in all copies, and that you add no other
5820 conditions whatsoever to those of this License. You may not use
5821 technical measures to obstruct or control the reading or further
5822 copying of the copies you make or distribute. However, you may accept
5823 compensation in exchange for copies. If you distribute a large enough
5824 number of copies you must also follow the conditions in section 3.
5826 You may also lend copies, under the same conditions stated above, and
5827 you may publicly display copies.
5830 3. COPYING IN QUANTITY
5832 If you publish printed copies of the Document numbering more than 100,
5833 and the Document's license notice requires Cover Texts, you must enclose
5834 the copies in covers that carry, clearly and legibly, all these Cover
5835 Texts: Front-Cover Texts on the front cover, and Back-Cover Texts on
5836 the back cover. Both covers must also clearly and legibly identify
5837 you as the publisher of these copies. The front cover must present
5838 the full title with all words of the title equally prominent and
5839 visible. You may add other material on the covers in addition.
5840 Copying with changes limited to the covers, as long as they preserve
5841 the title of the Document and satisfy these conditions, can be treated
5842 as verbatim copying in other respects.
5844 If the required texts for either cover are too voluminous to fit
5845 legibly, you should put the first ones listed (as many as fit
5846 reasonably) on the actual cover, and continue the rest onto adjacent
5849 If you publish or distribute Opaque copies of the Document numbering
5850 more than 100, you must either include a machine-readable Transparent
5851 copy along with each Opaque copy, or state in or with each Opaque copy
5852 a publicly-accessible computer-network location containing a complete
5853 Transparent copy of the Document, free of added material, which the
5854 general network-using public has access to download anonymously at no
5855 charge using public-standard network protocols. If you use the latter
5856 option, you must take reasonably prudent steps, when you begin
5857 distribution of Opaque copies in quantity, to ensure that this
5858 Transparent copy will remain thus accessible at the stated location
5859 until at least one year after the last time you distribute an Opaque
5860 copy (directly or through your agents or retailers) of that edition to
5863 It is requested, but not required, that you contact the authors of the
5864 Document well before redistributing any large number of copies, to give
5865 them a chance to provide you with an updated version of the Document.
5870 You may copy and distribute a Modified Version of the Document under
5871 the conditions of sections 2 and 3 above, provided that you release
5872 the Modified Version under precisely this License, with the Modified
5873 Version filling the role of the Document, thus licensing distribution
5874 and modification of the Modified Version to whoever possesses a copy
5875 of it. In addition, you must do these things in the Modified Version:
5877 A. Use in the Title Page (and on the covers, if any) a title distinct
5878 from that of the Document, and from those of previous versions
5879 (which should, if there were any, be listed in the History section
5880 of the Document). You may use the same title as a previous version
5881 if the original publisher of that version gives permission.
5882 B. List on the Title Page, as authors, one or more persons or entities
5883 responsible for authorship of the modifications in the Modified
5884 Version, together with at least five of the principal authors of the
5885 Document (all of its principal authors, if it has less than five).
5886 C. State on the Title page the name of the publisher of the
5887 Modified Version, as the publisher.
5888 D. Preserve all the copyright notices of the Document.
5889 E. Add an appropriate copyright notice for your modifications
5890 adjacent to the other copyright notices.
5891 F. Include, immediately after the copyright notices, a license notice
5892 giving the public permission to use the Modified Version under the
5893 terms of this License, in the form shown in the Addendum below.
5894 G. Preserve in that license notice the full lists of Invariant Sections
5895 and required Cover Texts given in the Document's license notice.
5896 H. Include an unaltered copy of this License.
5897 I. Preserve the section entitled "History", and its title, and add to
5898 it an item stating at least the title, year, new authors, and
5899 publisher of the Modified Version as given on the Title Page. If
5900 there is no section entitled "History" in the Document, create one
5901 stating the title, year, authors, and publisher of the Document as
5902 given on its Title Page, then add an item describing the Modified
5903 Version as stated in the previous sentence.
5904 J. Preserve the network location, if any, given in the Document for
5905 public access to a Transparent copy of the Document, and likewise
5906 the network locations given in the Document for previous versions
5907 it was based on. These may be placed in the "History" section.
5908 You may omit a network location for a work that was published at
5909 least four years before the Document itself, or if the original
5910 publisher of the version it refers to gives permission.
5911 K. In any section entitled "Acknowledgements" or "Dedications",
5912 preserve the section's title, and preserve in the section all the
5913 substance and tone of each of the contributor acknowledgements
5914 and/or dedications given therein.
5915 L. Preserve all the Invariant Sections of the Document,
5916 unaltered in their text and in their titles. Section numbers
5917 or the equivalent are not considered part of the section titles.
5918 M. Delete any section entitled "Endorsements". Such a section
5919 may not be included in the Modified Version.
5920 N. Do not retitle any existing section as "Endorsements"
5921 or to conflict in title with any Invariant Section.
5923 If the Modified Version includes new front-matter sections or
5924 appendices that qualify as Secondary Sections and contain no material
5925 copied from the Document, you may at your option designate some or all
5926 of these sections as invariant. To do this, add their titles to the
5927 list of Invariant Sections in the Modified Version's license notice.
5928 These titles must be distinct from any other section titles.
5930 You may add a section entitled "Endorsements", provided it contains
5931 nothing but endorsements of your Modified Version by various
5932 parties--for example, statements of peer review or that the text has
5933 been approved by an organization as the authoritative definition of a
5936 You may add a passage of up to five words as a Front-Cover Text, and a
5937 passage of up to 25 words as a Back-Cover Text, to the end of the list
5938 of Cover Texts in the Modified Version. Only one passage of
5939 Front-Cover Text and one of Back-Cover Text may be added by (or
5940 through arrangements made by) any one entity. If the Document already
5941 includes a cover text for the same cover, previously added by you or
5942 by arrangement made by the same entity you are acting on behalf of,
5943 you may not add another; but you may replace the old one, on explicit
5944 permission from the previous publisher that added the old one.
5946 The author(s) and publisher(s) of the Document do not by this License
5947 give permission to use their names for publicity for or to assert or
5948 imply endorsement of any Modified Version.
5951 5. COMBINING DOCUMENTS
5953 You may combine the Document with other documents released under this
5954 License, under the terms defined in section 4 above for modified
5955 versions, provided that you include in the combination all of the
5956 Invariant Sections of all of the original documents, unmodified, and
5957 list them all as Invariant Sections of your combined work in its
5960 The combined work need only contain one copy of this License, and
5961 multiple identical Invariant Sections may be replaced with a single
5962 copy. If there are multiple Invariant Sections with the same name but
5963 different contents, make the title of each such section unique by
5964 adding at the end of it, in parentheses, the name of the original
5965 author or publisher of that section if known, or else a unique number.
5966 Make the same adjustment to the section titles in the list of
5967 Invariant Sections in the license notice of the combined work.
5969 In the combination, you must combine any sections entitled "History"
5970 in the various original documents, forming one section entitled
5971 "History"; likewise combine any sections entitled "Acknowledgements",
5972 and any sections entitled "Dedications". You must delete all sections
5973 entitled "Endorsements."
5976 6. COLLECTIONS OF DOCUMENTS
5978 You may make a collection consisting of the Document and other documents
5979 released under this License, and replace the individual copies of this
5980 License in the various documents with a single copy that is included in
5981 the collection, provided that you follow the rules of this License for
5982 verbatim copying of each of the documents in all other respects.
5984 You may extract a single document from such a collection, and distribute
5985 it individually under this License, provided you insert a copy of this
5986 License into the extracted document, and follow this License in all
5987 other respects regarding verbatim copying of that document.
5990 7. AGGREGATION WITH INDEPENDENT WORKS
5992 A compilation of the Document or its derivatives with other separate
5993 and independent documents or works, in or on a volume of a storage or
5994 distribution medium, does not as a whole count as a Modified Version
5995 of the Document, provided no compilation copyright is claimed for the
5996 compilation. Such a compilation is called an "aggregate", and this
5997 License does not apply to the other self-contained works thus compiled
5998 with the Document, on account of their being thus compiled, if they
5999 are not themselves derivative works of the Document.
6001 If the Cover Text requirement of section 3 is applicable to these
6002 copies of the Document, then if the Document is less than one quarter
6003 of the entire aggregate, the Document's Cover Texts may be placed on
6004 covers that surround only the Document within the aggregate.
6005 Otherwise they must appear on covers around the whole aggregate.
6010 Translation is considered a kind of modification, so you may
6011 distribute translations of the Document under the terms of section 4.
6012 Replacing Invariant Sections with translations requires special
6013 permission from their copyright holders, but you may include
6014 translations of some or all Invariant Sections in addition to the
6015 original versions of these Invariant Sections. You may include a
6016 translation of this License provided that you also include the
6017 original English version of this License. In case of a disagreement
6018 between the translation and the original English version of this
6019 License, the original English version will prevail.
6024 You may not copy, modify, sublicense, or distribute the Document except
6025 as expressly provided for under this License. Any other attempt to
6026 copy, modify, sublicense or distribute the Document is void, and will
6027 automatically terminate your rights under this License. However,
6028 parties who have received copies, or rights, from you under this
6029 License will not have their licenses terminated so long as such
6030 parties remain in full compliance.
6033 10. FUTURE REVISIONS OF THIS LICENSE
6035 The Free Software Foundation may publish new, revised versions
6036 of the GNU Free Documentation License from time to time. Such new
6037 versions will be similar in spirit to the present version, but may
6038 differ in detail to address new problems or concerns. See
6039 http://www.gnu.org/copyleft/.
6041 Each version of the License is given a distinguishing version number.
6042 If the Document specifies that a particular numbered version of this
6043 License "or any later version" applies to it, you have the option of
6044 following the terms and conditions either of that specified version or
6045 of any later version that has been published (not as a draft) by the
6046 Free Software Foundation. If the Document does not specify a version
6047 number of this License, you may choose any version ever published (not
6048 as a draft) by the Free Software Foundation.
6051 ADDENDUM: How to use this License for your documents
6053 To use this License in a document you have written, include a copy of
6054 the License in the document and put the following copyright and
6055 license notices just after the title page:
6058 Copyright (c) YEAR YOUR NAME.
6059 Permission is granted to copy, distribute and/or modify this document
6060 under the terms of the GNU Free Documentation License, Version 1.1
6061 or any later version published by the Free Software Foundation;
6062 with the Invariant Sections being LIST THEIR TITLES, with the
6063 Front-Cover Texts being LIST, and with the Back-Cover Texts being LIST.
6064 A copy of the license is included in the section entitled "GNU
6065 Free Documentation License".
6068 If you have no Invariant Sections, write "with no Invariant Sections"
6069 instead of saying which ones are invariant. If you have no
6070 Front-Cover Texts, write "no Front-Cover Texts" instead of
6071 "Front-Cover Texts being LIST"; likewise for Back-Cover Texts.
6073 If your document contains nontrivial examples of program code, we
6074 recommend releasing these examples in parallel under your choice of
6075 free software license, such as the GNU General Public License,
6076 to permit their use in free software.