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 make and distribute verbatim copies of
93 this manual provided the copyright notice and this permission notice
94 are preserved on all copies.
97 Permission is granted to process this file through Tex and print the
98 results, provided the printed document carries copying permission
99 notice identical to this one except for the removal of this paragraph
100 (this paragraph not being relevant to the printed manual).
103 Permission is granted to copy and distribute modified versions of this manual
104 under the conditions for verbatim copying, provided that the entire resulting
105 derived work is distributed under the terms of a permission notice identical to
108 Permission is granted to copy and distribute translations of this manual
109 into another language, under the above conditions for modified versions.
113 @title Using @value{AS}
114 @subtitle The @sc{gnu} Assembler
116 @subtitle for the @value{TARGET} family
119 @subtitle Version @value{VERSION}
122 The Free Software Foundation Inc. thanks The Nice Computer
123 Company of Australia for loaning Dean Elsner to write the
124 first (Vax) version of @code{as} for Project @sc{gnu}.
125 The proprietors, management and staff of TNCCA thank FSF for
126 distracting the boss while they got some work
129 @author Dean Elsner, Jay Fenlason & friends
133 \hfill {\it Using {\tt @value{AS}}}\par
134 \hfill Edited by Cygnus Support\par
136 %"boxit" macro for figures:
137 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
138 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
139 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
140 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
141 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
144 @vskip 0pt plus 1filll
145 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000 Free Software Foundation, Inc.
147 Permission is granted to make and distribute verbatim copies of
148 this manual provided the copyright notice and this permission notice
149 are preserved on all copies.
151 Permission is granted to copy and distribute modified versions of this manual
152 under the conditions for verbatim copying, provided that the entire resulting
153 derived work is distributed under the terms of a permission notice identical to
156 Permission is granted to copy and distribute translations of this manual
157 into another language, under the above conditions for modified versions.
162 @top Using @value{AS}
164 This file is a user guide to the @sc{gnu} assembler @code{@value{AS}} version
167 This version of the file describes @code{@value{AS}} configured to generate
168 code for @value{TARGET} architectures.
171 * Overview:: Overview
172 * Invoking:: Command-Line Options
174 * Sections:: Sections and Relocation
176 * Expressions:: Expressions
177 * Pseudo Ops:: Assembler Directives
178 * Machine Dependencies:: Machine Dependent Features
179 * Reporting Bugs:: Reporting Bugs
180 * Acknowledgements:: Who Did What
188 This manual is a user guide to the @sc{gnu} assembler @code{@value{AS}}.
190 This version of the manual describes @code{@value{AS}} configured to generate
191 code for @value{TARGET} architectures.
195 @cindex invocation summary
196 @cindex option summary
197 @cindex summary of options
198 Here is a brief summary of how to invoke @code{@value{AS}}. For details,
199 @pxref{Invoking,,Comand-Line Options}.
201 @c We don't use deffn and friends for the following because they seem
202 @c to be limited to one line for the header.
204 @value{AS} [ -a[cdhlns][=file] ] [ -D ] [ --defsym @var{sym}=@var{val} ]
205 [ -f ] [ --gstabs ] [ --gdwarf2 ] [ --help ] [ -I @var{dir} ] [ -J ] [ -K ] [ -L ]
206 [ --keep-locals ] [ -o @var{objfile} ] [ -R ] [ --statistics ] [ -v ]
207 [ -version ] [ --version ] [ -W ] [ --warn ] [ --fatal-warnings ]
210 @c am29k has no machine-dependent assembler options
213 [ -mbig-endian | -mlittle-endian ]
216 [ -m[arm]1 | -m[arm]2 | -m[arm]250 | -m[arm]3 | -m[arm]6 | -m[arm]60 |
217 -m[arm]600 | -m[arm]610 | -m[arm]620 | -m[arm]7[t][[d]m[i]][fe] | -m[arm]70 |
218 -m[arm]700 | -m[arm]710[c] | -m[arm]7100 | -m[arm]7500 | -m[arm]8 |
219 -m[arm]810 | -m[arm]9 | -m[arm]920 | -m[arm]920t | -m[arm]9tdmi |
220 -mstrongarm | -mstrongarm110 | -mstrongarm1100 ]
221 [ -m[arm]v2 | -m[arm]v2a | -m[arm]v3 | -m[arm]v3m | -m[arm]v4 | -m[arm]v4t |
222 -m[arm]v5 | -[arm]v5t ]
224 [ -mfpa10 | -mfpa11 | -mfpe-old | -mno-fpu ]
226 [ -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant ]
227 [ -mthumb-interwork ]
238 @c Hitachi family chips have no machine-dependent assembler options
241 @c HPPA has no machine-dependent assembler options (yet).
247 @c The order here is important. See c-sparc.texi.
248 [ -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
249 -Av8plus | -Av8plusa | -Av9 | -Av9a ]
250 [ -xarch=v8plus | -xarch=v8plusa ] [ -bump ] [ -32 | -64 ]
253 [ -mcpu=54[123589] | -mcpu=54[56]lp ] [ -mfar-mode | -mf ]
254 [ -merrors-to-file <filename> | -me <filename> ]
257 @c Z8000 has no machine-dependent assembler options
260 @c see md_parse_option in tc-i960.c
261 [ -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC ]
265 [ --m32rx | --[no-]warn-explicit-parallel-conflicts | --W[n]p ]
268 [ -l ] [ -m68000 | -m68010 | -m68020 | ... ]
271 [ -jsri2bsr ] [ -sifilter ] [ -relax ]
275 [ -m68hc11 | -m68hc12 ]
276 [ --force-long-branchs ] [ --short-branchs ] [ --strict-direct-mode ]
277 [ --print-insn-syntax ] [ --print-opcodes ] [ --generate-example ]
280 [ -nocpp ] [ -EL ] [ -EB ] [ -G @var{num} ] [ -mcpu=@var{CPU} ]
281 [ -mips1 ] [ -mips2 ] [ -mips3 ] [ -m4650 ] [ -no-m4650 ]
282 [ --trap ] [ --break ]
283 [ --emulation=@var{name} ]
285 [ -- | @var{files} @dots{} ]
290 Turn on listings, in any of a variety of ways:
294 omit false conditionals
297 omit debugging directives
300 include high-level source
306 include macro expansions
309 omit forms processing
315 set the name of the listing file
318 You may combine these options; for example, use @samp{-aln} for assembly
319 listing without forms processing. The @samp{=file} option, if used, must be
320 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
323 Ignored. This option is accepted for script compatibility with calls to
326 @item --defsym @var{sym}=@var{value}
327 Define the symbol @var{sym} to be @var{value} before assembling the input file.
328 @var{value} must be an integer constant. As in C, a leading @samp{0x}
329 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
332 ``fast''---skip whitespace and comment preprocessing (assume source is
336 Generate stabs debugging information for each assembler line. This
337 may help debugging assembler code, if the debugger can handle it.
340 Generate DWARF2 debugging information for each assembler line. This
341 may help debugging assembler code, if the debugger can handle it.
344 Print a summary of the command line options and exit.
347 Add directory @var{dir} to the search list for @code{.include} directives.
350 Don't warn about signed overflow.
353 @ifclear DIFF-TBL-KLUGE
354 This option is accepted but has no effect on the @value{TARGET} family.
356 @ifset DIFF-TBL-KLUGE
357 Issue warnings when difference tables altered for long displacements.
362 Keep (in the symbol table) local symbols. On traditional a.out systems
363 these start with @samp{L}, but different systems have different local
366 @item -o @var{objfile}
367 Name the object-file output from @code{@value{AS}} @var{objfile}.
370 Fold the data section into the text section.
373 Print the maximum space (in bytes) and total time (in seconds) used by
376 @item --strip-local-absolute
377 Remove local absolute symbols from the outgoing symbol table.
381 Print the @code{as} version.
384 Print the @code{as} version and exit.
388 Suppress warning messages.
390 @item --fatal-warnings
391 Treat warnings as errors.
394 Don't suppress warning messages or treat them as errors.
403 Generate an object file even after errors.
405 @item -- | @var{files} @dots{}
406 Standard input, or source files to assemble.
411 The following options are available when @value{AS} is configured for
416 @cindex ARC endianness
417 @cindex endianness, ARC
418 @cindex big endian output, ARC
420 Generate ``big endian'' format output.
422 @cindex little endian output, ARC
423 @item -mlittle-endian
424 Generate ``little endian'' format output.
430 The following options are available when @value{AS} is configured for the ARM
434 @item -m[arm][1|2|3|6|7|8|9][...]
435 Specify which ARM processor variant is the target.
436 @item -m[arm]v[2|2a|3|3m|4|4t|5|5t]
437 Specify which ARM architecture variant is used by the target.
438 @item -mthumb | -mall
439 Enable or disable Thumb only instruction decoding.
440 @item -mfpa10 | -mfpa11 | -mfpe-old | -mno-fpu
441 Select which Floating Point architcture is the target.
442 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant | -moabi
443 Select which procedure calling convention is in use.
445 Select either big-endian (-EB) or little-endian (-EL) output.
446 @item -mthumb-interwork
447 Specify that the code has been generated with interworking between Thumb and
450 Specify that PIC code has been generated.
455 The following options are available when @value{AS} is configured for
458 @cindex D10V optimization
459 @cindex optimization, D10V
461 Optimize output by parallelizing instructions.
466 The following options are available when @value{AS} is configured for a D30V
469 @cindex D30V optimization
470 @cindex optimization, D30V
472 Optimize output by parallelizing instructions.
476 Warn when nops are generated.
478 @cindex D30V nops after 32-bit multiply
480 Warn when a nop after a 32-bit multiply instruction is generated.
485 The following options are available when @value{AS} is configured for the
486 Intel 80960 processor.
489 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
490 Specify which variant of the 960 architecture is the target.
493 Add code to collect statistics about branches taken.
496 Do not alter compare-and-branch instructions for long displacements;
503 The following options are available when @value{AS} is configured for the
504 Mitsubishi M32R series.
509 Specify which processor in the M32R family is the target. The default
510 is normally the M32R, but this option changes it to the M32RX.
512 @item --warn-explicit-parallel-conflicts or --Wp
513 Produce warning messages when questionable parallel constructs are
516 @item --no-warn-explicit-parallel-conflicts or --Wnp
517 Do not produce warning messages when questionable parallel constructs are
524 The following options are available when @value{AS} is configured for the
525 Motorola 68000 series.
530 Shorten references to undefined symbols, to one word instead of two.
532 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030 | -m68040 | -m68060
533 @itemx | -m68302 | -m68331 | -m68332 | -m68333 | -m68340 | -mcpu32 | -m5200
534 Specify what processor in the 68000 family is the target. The default
535 is normally the 68020, but this can be changed at configuration time.
537 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
538 The target machine does (or does not) have a floating-point coprocessor.
539 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
540 the basic 68000 is not compatible with the 68881, a combination of the
541 two can be specified, since it's possible to do emulation of the
542 coprocessor instructions with the main processor.
544 @item -m68851 | -mno-68851
545 The target machine does (or does not) have a memory-management
546 unit coprocessor. The default is to assume an MMU for 68020 and up.
552 The following options are available when @value{AS} is configured for
553 a picoJava processor.
557 @cindex PJ endianness
558 @cindex endianness, PJ
559 @cindex big endian output, PJ
561 Generate ``big endian'' format output.
563 @cindex little endian output, PJ
565 Generate ``little endian'' format output.
571 The following options are available when @value{AS} is configured for the
572 Motorola 68HC11 or 68HC12 series.
576 @item -m68hc11 | -m68hc12
577 Specify what processor is the target. The default is
578 defined by the configuration option when building the assembler.
580 @item --force-long-branchs
581 Relative branches are turned into absolute ones. This concerns
582 conditional branches, unconditional branches and branches to a
585 @item -S | --short-branchs
586 Do not turn relative branchs into absolute ones
587 when the offset is out of range.
589 @item --strict-direct-mode
590 Do not turn the direct addressing mode into extended addressing mode
591 when the instruction does not support direct addressing mode.
593 @item --print-insn-syntax
594 Print the syntax of instruction in case of error.
596 @item --print-opcodes
597 print the list of instructions with syntax and then exit.
599 @item --generate-example
600 print an example of instruction for each possible instruction and then exit.
601 This option is only useful for testing @code{@value{AS}}.
607 The following options are available when @code{@value{AS}} is configured
608 for the SPARC architecture:
611 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
612 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
613 Explicitly select a variant of the SPARC architecture.
615 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
616 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
618 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
619 UltraSPARC extensions.
621 @item -xarch=v8plus | -xarch=v8plusa
622 For compatibility with the Solaris v9 assembler. These options are
623 equivalent to -Av8plus and -Av8plusa, respectively.
626 Warn when the assembler switches to another architecture.
631 The following options are available when @value{AS} is configured for the 'c54x
636 Enable extended addressing mode. All addresses and relocations will assume
637 extended addressing (usually 23 bits).
638 @item -mcpu=@var{CPU_VERSION}
639 Sets the CPU version being compiled for.
640 @item -merrors-to-file @var{FILENAME}
641 Redirect error output to a file, for broken systems which don't support such
642 behaviour in the shell.
647 The following options are available when @value{AS} is configured for
652 This option sets the largest size of an object that can be referenced
653 implicitly with the @code{gp} register. It is only accepted for targets that
654 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
656 @cindex MIPS endianness
657 @cindex endianness, MIPS
658 @cindex big endian output, MIPS
660 Generate ``big endian'' format output.
662 @cindex little endian output, MIPS
664 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}
677 Generate code for the MIPS @sc{r4650} chip. This tells the assembler to accept
678 the @samp{mad} and @samp{madu} instruction, and to not schedule @samp{nop}
679 instructions around accesses to the @samp{HI} and @samp{LO} registers.
680 @samp{-no-m4650} turns off this option.
682 @item -mcpu=@var{CPU}
683 Generate code for a particular MIPS cpu. This has little effect on the
684 assembler, but it is passed by @code{@value{GCC}}.
687 @item --emulation=@var{name}
688 This option causes @code{@value{AS}} to emulate @code{@value{AS}} configured
689 for some other target, in all respects, including output format (choosing
690 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
691 debugging information or store symbol table information, and default
692 endianness. The available configuration names are: @samp{mipsecoff},
693 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
694 @samp{mipsbelf}. The first two do not alter the default endianness from that
695 of the primary target for which the assembler was configured; the others change
696 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
697 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
698 selection in any case.
700 This option is currently supported only when the primary target
701 @code{@value{AS}} is configured for is a MIPS ELF or ECOFF target.
702 Furthermore, the primary target or others specified with
703 @samp{--enable-targets=@dots{}} at configuration time must include support for
704 the other format, if both are to be available. For example, the Irix 5
705 configuration includes support for both.
707 Eventually, this option will support more configurations, with more
708 fine-grained control over the assembler's behavior, and will be supported for
712 @code{@value{AS}} ignores this option. It is accepted for compatibility with
720 Control how to deal with multiplication overflow and division by zero.
721 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
722 (and only work for Instruction Set Architecture level 2 and higher);
723 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
729 The following options are available when @value{AS} is configured for
735 Enable or disable the JSRI to BSR transformation. By default this is enabled.
736 The command line option @samp{-nojsri2bsr} can be used to disable it.
740 Enable or disable the silicon filter behaviour. By default this is disabled.
741 The default can be overidden by the @samp{-sifilter} command line option.
744 Alter jump instructions for long displacements.
746 @item -mcpu=[210|340]
747 Select the cpu type on the target hardware. This controls which instructions
751 Assemble for a big endian target.
754 Assemble for a little endian target.
760 * Manual:: Structure of this Manual
761 * GNU Assembler:: The GNU Assembler
762 * Object Formats:: Object File Formats
763 * Command Line:: Command Line
764 * Input Files:: Input Files
765 * Object:: Output (Object) File
766 * Errors:: Error and Warning Messages
770 @section Structure of this Manual
772 @cindex manual, structure and purpose
773 This manual is intended to describe what you need to know to use
774 @sc{gnu} @code{@value{AS}}. We cover the syntax expected in source files, including
775 notation for symbols, constants, and expressions; the directives that
776 @code{@value{AS}} understands; and of course how to invoke @code{@value{AS}}.
779 We also cover special features in the @value{TARGET}
780 configuration of @code{@value{AS}}, including assembler directives.
783 This manual also describes some of the machine-dependent features of
784 various flavors of the assembler.
787 @cindex machine instructions (not covered)
788 On the other hand, this manual is @emph{not} intended as an introduction
789 to programming in assembly language---let alone programming in general!
790 In a similar vein, we make no attempt to introduce the machine
791 architecture; we do @emph{not} describe the instruction set, standard
792 mnemonics, registers or addressing modes that are standard to a
793 particular architecture.
795 You may want to consult the manufacturer's
796 machine architecture manual for this information.
800 For information on the H8/300 machine instruction set, see @cite{H8/300
801 Series Programming Manual} (Hitachi ADE--602--025). For the H8/300H,
802 see @cite{H8/300H Series Programming Manual} (Hitachi).
805 For information on the H8/500 machine instruction set, see @cite{H8/500
806 Series Programming Manual} (Hitachi M21T001).
809 For information on the Hitachi SH machine instruction set, see
810 @cite{SH-Microcomputer User's Manual} (Hitachi Micro Systems, Inc.).
813 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
817 @c I think this is premature---doc@cygnus.com, 17jan1991
819 Throughout this manual, we assume that you are running @dfn{GNU},
820 the portable operating system from the @dfn{Free Software
821 Foundation, Inc.}. This restricts our attention to certain kinds of
822 computer (in particular, the kinds of computers that @sc{gnu} can run on);
823 once this assumption is granted examples and definitions need less
826 @code{@value{AS}} is part of a team of programs that turn a high-level
827 human-readable series of instructions into a low-level
828 computer-readable series of instructions. Different versions of
829 @code{@value{AS}} are used for different kinds of computer.
832 @c There used to be a section "Terminology" here, which defined
833 @c "contents", "byte", "word", and "long". Defining "word" to any
834 @c particular size is confusing when the .word directive may generate 16
835 @c bits on one machine and 32 bits on another; in general, for the user
836 @c version of this manual, none of these terms seem essential to define.
837 @c They were used very little even in the former draft of the manual;
838 @c this draft makes an effort to avoid them (except in names of
842 @section The GNU Assembler
844 @sc{gnu} @code{as} is really a family of assemblers.
846 This manual describes @code{@value{AS}}, a member of that family which is
847 configured for the @value{TARGET} architectures.
849 If you use (or have used) the @sc{gnu} assembler on one architecture, you
850 should find a fairly similar environment when you use it on another
851 architecture. Each version has much in common with the others,
852 including object file formats, most assembler directives (often called
853 @dfn{pseudo-ops}) and assembler syntax.@refill
855 @cindex purpose of @sc{gnu} assembler
856 @code{@value{AS}} is primarily intended to assemble the output of the
857 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
858 @code{@value{LD}}. Nevertheless, we've tried to make @code{@value{AS}}
859 assemble correctly everything that other assemblers for the same
860 machine would assemble.
862 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
865 @c This remark should appear in generic version of manual; assumption
866 @c here is that generic version sets M680x0.
867 This doesn't mean @code{@value{AS}} always uses the same syntax as another
868 assembler for the same architecture; for example, we know of several
869 incompatible versions of 680x0 assembly language syntax.
872 Unlike older assemblers, @code{@value{AS}} is designed to assemble a source
873 program in one pass of the source file. This has a subtle impact on the
874 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
877 @section Object File Formats
879 @cindex object file format
880 The @sc{gnu} assembler can be configured to produce several alternative
881 object file formats. For the most part, this does not affect how you
882 write assembly language programs; but directives for debugging symbols
883 are typically different in different file formats. @xref{Symbol
884 Attributes,,Symbol Attributes}.
887 On the @value{TARGET}, @code{@value{AS}} is configured to produce
888 @value{OBJ-NAME} format object files.
890 @c The following should exhaust all configs that set MULTI-OBJ, ideally
892 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
893 @code{a.out} or COFF format object files.
896 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
897 @code{b.out} or COFF format object files.
900 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
901 SOM or ELF format object files.
906 @section Command Line
908 @cindex command line conventions
909 After the program name @code{@value{AS}}, the command line may contain
910 options and file names. Options may appear in any order, and may be
911 before, after, or between file names. The order of file names is
914 @cindex standard input, as input file
916 @file{--} (two hyphens) by itself names the standard input file
917 explicitly, as one of the files for @code{@value{AS}} to assemble.
919 @cindex options, command line
920 Except for @samp{--} any command line argument that begins with a
921 hyphen (@samp{-}) is an option. Each option changes the behavior of
922 @code{@value{AS}}. No option changes the way another option works. An
923 option is a @samp{-} followed by one or more letters; the case of
924 the letter is important. All options are optional.
926 Some options expect exactly one file name to follow them. The file
927 name may either immediately follow the option's letter (compatible
928 with older assemblers) or it may be the next command argument (@sc{gnu}
929 standard). These two command lines are equivalent:
932 @value{AS} -o my-object-file.o mumble.s
933 @value{AS} -omy-object-file.o mumble.s
940 @cindex source program
942 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
943 describe the program input to one run of @code{@value{AS}}. The program may
944 be in one or more files; how the source is partitioned into files
945 doesn't change the meaning of the source.
947 @c I added "con" prefix to "catenation" just to prove I can overcome my
948 @c APL training... doc@cygnus.com
949 The source program is a concatenation of the text in all the files, in the
952 Each time you run @code{@value{AS}} it assembles exactly one source
953 program. The source program is made up of one or more files.
954 (The standard input is also a file.)
956 You give @code{@value{AS}} a command line that has zero or more input file
957 names. The input files are read (from left file name to right). A
958 command line argument (in any position) that has no special meaning
959 is taken to be an input file name.
961 If you give @code{@value{AS}} no file names it attempts to read one input file
962 from the @code{@value{AS}} standard input, which is normally your terminal. You
963 may have to type @key{ctl-D} to tell @code{@value{AS}} there is no more program
966 Use @samp{--} if you need to explicitly name the standard input file
967 in your command line.
969 If the source is empty, @code{@value{AS}} produces a small, empty object
972 @subheading Filenames and Line-numbers
974 @cindex input file linenumbers
975 @cindex line numbers, in input files
976 There are two ways of locating a line in the input file (or files) and
977 either may be used in reporting error messages. One way refers to a line
978 number in a physical file; the other refers to a line number in a
979 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
981 @dfn{Physical files} are those files named in the command line given
982 to @code{@value{AS}}.
984 @dfn{Logical files} are simply names declared explicitly by assembler
985 directives; they bear no relation to physical files. Logical file names help
986 error messages reflect the original source file, when @code{@value{AS}} source
987 is itself synthesized from other files. @code{@value{AS}} understands the
988 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
989 @ref{File,,@code{.file}}.
992 @section Output (Object) File
998 Every time you run @code{@value{AS}} it produces an output file, which is
999 your assembly language program translated into numbers. This file
1000 is the object file. Its default name is
1008 @code{b.out} when @code{@value{AS}} is configured for the Intel 80960.
1010 You can give it another name by using the @code{-o} option. Conventionally,
1011 object file names end with @file{.o}. The default name is used for historical
1012 reasons: older assemblers were capable of assembling self-contained programs
1013 directly into a runnable program. (For some formats, this isn't currently
1014 possible, but it can be done for the @code{a.out} format.)
1018 The object file is meant for input to the linker @code{@value{LD}}. It contains
1019 assembled program code, information to help @code{@value{LD}} integrate
1020 the assembled program into a runnable file, and (optionally) symbolic
1021 information for the debugger.
1023 @c link above to some info file(s) like the description of a.out.
1024 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1027 @section Error and Warning Messages
1029 @cindex error messsages
1030 @cindex warning messages
1031 @cindex messages from assembler
1032 @code{@value{AS}} may write warnings and error messages to the standard error
1033 file (usually your terminal). This should not happen when a compiler
1034 runs @code{@value{AS}} automatically. Warnings report an assumption made so
1035 that @code{@value{AS}} could keep assembling a flawed program; errors report a
1036 grave problem that stops the assembly.
1038 @cindex format of warning messages
1039 Warning messages have the format
1042 file_name:@b{NNN}:Warning Message Text
1046 @cindex line numbers, in warnings/errors
1047 (where @b{NNN} is a line number). If a logical file name has been given
1048 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1049 the current input file is used. If a logical line number was given
1051 (@pxref{Line,,@code{.line}})
1055 (@pxref{Line,,@code{.line}})
1058 (@pxref{Ln,,@code{.ln}})
1061 then it is used to calculate the number printed,
1062 otherwise the actual line in the current source file is printed. The
1063 message text is intended to be self explanatory (in the grand Unix
1066 @cindex format of error messages
1067 Error messages have the format
1069 file_name:@b{NNN}:FATAL:Error Message Text
1071 The file name and line number are derived as for warning
1072 messages. The actual message text may be rather less explanatory
1073 because many of them aren't supposed to happen.
1076 @chapter Command-Line Options
1078 @cindex options, all versions of assembler
1079 This chapter describes command-line options available in @emph{all}
1080 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
1082 to the @value{TARGET}.
1085 to particular machine architectures.
1088 If you are invoking @code{@value{AS}} via the @sc{gnu} C compiler (version 2),
1089 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1090 The assembler arguments must be separated from each other (and the @samp{-Wa})
1091 by commas. For example:
1094 gcc -c -g -O -Wa,-alh,-L file.c
1098 This passes two options to the assembler: @samp{-alh} (emit a listing to
1099 standard output with with high-level and assembly source) and @samp{-L} (retain
1100 local symbols in the symbol table).
1102 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1103 command-line options are automatically passed to the assembler by the compiler.
1104 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1105 precisely what options it passes to each compilation pass, including the
1109 * a:: -a[cdhlns] enable listings
1110 * D:: -D for compatibility
1111 * f:: -f to work faster
1112 * I:: -I for .include search path
1113 @ifclear DIFF-TBL-KLUGE
1114 * K:: -K for compatibility
1116 @ifset DIFF-TBL-KLUGE
1117 * K:: -K for difference tables
1120 * L:: -L to retain local labels
1121 * M:: -M or --mri to assemble in MRI compatibility mode
1122 * MD:: --MD for dependency tracking
1123 * o:: -o to name the object file
1124 * R:: -R to join data and text sections
1125 * statistics:: --statistics to see statistics about assembly
1126 * traditional-format:: --traditional-format for compatible output
1127 * v:: -v to announce version
1128 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1129 * Z:: -Z to make object file even after errors
1133 @section Enable Listings: @code{-a[cdhlns]}
1142 @cindex listings, enabling
1143 @cindex assembly listings, enabling
1145 These options enable listing output from the assembler. By itself,
1146 @samp{-a} requests high-level, assembly, and symbols listing.
1147 You can use other letters to select specific options for the list:
1148 @samp{-ah} requests a high-level language listing,
1149 @samp{-al} requests an output-program assembly listing, and
1150 @samp{-as} requests a symbol table listing.
1151 High-level listings require that a compiler debugging option like
1152 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1155 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1156 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1157 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1158 omitted from the listing.
1160 Use the @samp{-ad} option to omit debugging directives from the
1163 Once you have specified one of these options, you can further control
1164 listing output and its appearance using the directives @code{.list},
1165 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1167 The @samp{-an} option turns off all forms processing.
1168 If you do not request listing output with one of the @samp{-a} options, the
1169 listing-control directives have no effect.
1171 The letters after @samp{-a} may be combined into one option,
1172 @emph{e.g.}, @samp{-aln}.
1178 This option has no effect whatsoever, but it is accepted to make it more
1179 likely that scripts written for other assemblers also work with
1183 @section Work Faster: @code{-f}
1186 @cindex trusted compiler
1187 @cindex faster processing (@code{-f})
1188 @samp{-f} should only be used when assembling programs written by a
1189 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1190 and comment preprocessing on
1191 the input file(s) before assembling them. @xref{Preprocessing,
1195 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1196 preprocessed (if they contain comments, for example), @code{@value{AS}} does
1201 @section @code{.include} search path: @code{-I} @var{path}
1203 @kindex -I @var{path}
1204 @cindex paths for @code{.include}
1205 @cindex search path for @code{.include}
1206 @cindex @code{include} directive search path
1207 Use this option to add a @var{path} to the list of directories
1208 @code{@value{AS}} searches for files specified in @code{.include}
1209 directives (@pxref{Include,,@code{.include}}). You may use @code{-I} as
1210 many times as necessary to include a variety of paths. The current
1211 working directory is always searched first; after that, @code{@value{AS}}
1212 searches any @samp{-I} directories in the same order as they were
1213 specified (left to right) on the command line.
1216 @section Difference Tables: @code{-K}
1219 @ifclear DIFF-TBL-KLUGE
1220 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1221 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1222 where it can be used to warn when the assembler alters the machine code
1223 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1224 family does not have the addressing limitations that sometimes lead to this
1225 alteration on other platforms.
1228 @ifset DIFF-TBL-KLUGE
1229 @cindex difference tables, warning
1230 @cindex warning for altered difference tables
1231 @code{@value{AS}} sometimes alters the code emitted for directives of the form
1232 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1233 You can use the @samp{-K} option if you want a warning issued when this
1238 @section Include Local Labels: @code{-L}
1241 @cindex local labels, retaining in output
1242 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1243 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1244 debugging, because they are intended for the use of programs (like
1245 compilers) that compose assembler programs, not for your notice.
1246 Normally both @code{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1247 normally debug with them.
1249 This option tells @code{@value{AS}} to retain those @samp{L@dots{}} symbols
1250 in the object file. Usually if you do this you also tell the linker
1251 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1253 By default, a local label is any label beginning with @samp{L}, but each
1254 target is allowed to redefine the local label prefix.
1256 On the HPPA local labels begin with @samp{L$}.
1260 @section Assemble in MRI Compatibility Mode: @code{-M}
1263 @cindex MRI compatibility mode
1264 The @code{-M} or @code{--mri} option selects MRI compatibility mode. This
1265 changes the syntax and pseudo-op handling of @code{@value{AS}} to make it
1266 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1267 configured target) assembler from Microtec Research. The exact nature of the
1268 MRI syntax will not be documented here; see the MRI manuals for more
1269 information. Note in particular that the handling of macros and macro
1270 arguments is somewhat different. The purpose of this option is to permit
1271 assembling existing MRI assembler code using @code{@value{AS}}.
1273 The MRI compatibility is not complete. Certain operations of the MRI assembler
1274 depend upon its object file format, and can not be supported using other object
1275 file formats. Supporting these would require enhancing each object file format
1276 individually. These are:
1279 @item global symbols in common section
1281 The m68k MRI assembler supports common sections which are merged by the linker.
1282 Other object file formats do not support this. @code{@value{AS}} handles
1283 common sections by treating them as a single common symbol. It permits local
1284 symbols to be defined within a common section, but it can not support global
1285 symbols, since it has no way to describe them.
1287 @item complex relocations
1289 The MRI assemblers support relocations against a negated section address, and
1290 relocations which combine the start addresses of two or more sections. These
1291 are not support by other object file formats.
1293 @item @code{END} pseudo-op specifying start address
1295 The MRI @code{END} pseudo-op permits the specification of a start address.
1296 This is not supported by other object file formats. The start address may
1297 instead be specified using the @code{-e} option to the linker, or in a linker
1300 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1302 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1303 name to the output file. This is not supported by other object file formats.
1305 @item @code{ORG} pseudo-op
1307 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1308 address. This differs from the usual @code{@value{AS}} @code{.org} pseudo-op,
1309 which changes the location within the current section. Absolute sections are
1310 not supported by other object file formats. The address of a section may be
1311 assigned within a linker script.
1314 There are some other features of the MRI assembler which are not supported by
1315 @code{@value{AS}}, typically either because they are difficult or because they
1316 seem of little consequence. Some of these may be supported in future releases.
1320 @item EBCDIC strings
1322 EBCDIC strings are not supported.
1324 @item packed binary coded decimal
1326 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1327 and @code{DCB.P} pseudo-ops are not supported.
1329 @item @code{FEQU} pseudo-op
1331 The m68k @code{FEQU} pseudo-op is not supported.
1333 @item @code{NOOBJ} pseudo-op
1335 The m68k @code{NOOBJ} pseudo-op is not supported.
1337 @item @code{OPT} branch control options
1339 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1340 @code{BRL}, and @code{BRW}---are ignored. @code{@value{AS}} automatically
1341 relaxes all branches, whether forward or backward, to an appropriate size, so
1342 these options serve no purpose.
1344 @item @code{OPT} list control options
1346 The following m68k @code{OPT} list control options are ignored: @code{C},
1347 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1348 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1350 @item other @code{OPT} options
1352 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1353 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1355 @item @code{OPT} @code{D} option is default
1357 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1358 @code{OPT NOD} may be used to turn it off.
1360 @item @code{XREF} pseudo-op.
1362 The m68k @code{XREF} pseudo-op is ignored.
1364 @item @code{.debug} pseudo-op
1366 The i960 @code{.debug} pseudo-op is not supported.
1368 @item @code{.extended} pseudo-op
1370 The i960 @code{.extended} pseudo-op is not supported.
1372 @item @code{.list} pseudo-op.
1374 The various options of the i960 @code{.list} pseudo-op are not supported.
1376 @item @code{.optimize} pseudo-op
1378 The i960 @code{.optimize} pseudo-op is not supported.
1380 @item @code{.output} pseudo-op
1382 The i960 @code{.output} pseudo-op is not supported.
1384 @item @code{.setreal} pseudo-op
1386 The i960 @code{.setreal} pseudo-op is not supported.
1391 @section Dependency tracking: @code{--MD}
1394 @cindex dependency tracking
1397 @code{@value{AS}} can generate a dependency file for the file it creates. This
1398 file consists of a single rule suitable for @code{make} describing the
1399 dependencies of the main source file.
1401 The rule is written to the file named in its argument.
1403 This feature is used in the automatic updating of makefiles.
1406 @section Name the Object File: @code{-o}
1409 @cindex naming object file
1410 @cindex object file name
1411 There is always one object file output when you run @code{@value{AS}}. By
1412 default it has the name
1415 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1429 You use this option (which takes exactly one filename) to give the
1430 object file a different name.
1432 Whatever the object file is called, @code{@value{AS}} overwrites any
1433 existing file of the same name.
1436 @section Join Data and Text Sections: @code{-R}
1439 @cindex data and text sections, joining
1440 @cindex text and data sections, joining
1441 @cindex joining text and data sections
1442 @cindex merging text and data sections
1443 @code{-R} tells @code{@value{AS}} to write the object file as if all
1444 data-section data lives in the text section. This is only done at
1445 the very last moment: your binary data are the same, but data
1446 section parts are relocated differently. The data section part of
1447 your object file is zero bytes long because all its bytes are
1448 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1450 When you specify @code{-R} it would be possible to generate shorter
1451 address displacements (because we do not have to cross between text and
1452 data section). We refrain from doing this simply for compatibility with
1453 older versions of @code{@value{AS}}. In future, @code{-R} may work this way.
1456 When @code{@value{AS}} is configured for COFF output,
1457 this option is only useful if you use sections named @samp{.text} and
1462 @code{-R} is not supported for any of the HPPA targets. Using
1463 @code{-R} generates a warning from @code{@value{AS}}.
1467 @section Display Assembly Statistics: @code{--statistics}
1469 @kindex --statistics
1470 @cindex statistics, about assembly
1471 @cindex time, total for assembly
1472 @cindex space used, maximum for assembly
1473 Use @samp{--statistics} to display two statistics about the resources used by
1474 @code{@value{AS}}: the maximum amount of space allocated during the assembly
1475 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1478 @node traditional-format
1479 @section Compatible output: @code{--traditional-format}
1481 @kindex --traditional-format
1482 For some targets, the output of @code{@value{AS}} is different in some ways
1483 from the output of some existing assembler. This switch requests
1484 @code{@value{AS}} to use the traditional format instead.
1486 For example, it disables the exception frame optimizations which
1487 @code{@value{AS}} normally does by default on @code{@value{GCC}} output.
1490 @section Announce Version: @code{-v}
1494 @cindex assembler version
1495 @cindex version of assembler
1496 You can find out what version of as is running by including the
1497 option @samp{-v} (which you can also spell as @samp{-version}) on the
1501 @section Control Warnings: @code{-W}, @code{--warn}, @code{--no-warn}, @code{--fatal-warnings}
1503 @code{@value{AS}} should never give a warning or error message when
1504 assembling compiler output. But programs written by people often
1505 cause @code{@value{AS}} to give a warning that a particular assumption was
1506 made. All such warnings are directed to the standard error file.
1509 @kindex @samp{--no-warn}
1510 @cindex suppressing warnings
1511 @cindex warnings, suppressing
1512 If you use the @code{-W} and @code{--no-warn} options, no warnings are issued.
1513 This only affects the warning messages: it does not change any particular of
1514 how @code{@value{AS}} assembles your file. Errors, which stop the assembly,
1517 @kindex @samp{--fatal-warnings}
1518 @cindex errors, caused by warnings
1519 @cindex warnings, causing error
1520 If you use the @code{--fatal-warnings} option, @code{@value{AS}} considers
1521 files that generate warnings to be in error.
1523 @kindex @samp{--warn}
1524 @cindex warnings, switching on
1525 You can switch these options off again by specifying @code{--warn}, which
1526 causes warnings to be output as usual.
1529 @section Generate Object File in Spite of Errors: @code{-Z}
1530 @cindex object file, after errors
1531 @cindex errors, continuing after
1532 After an error message, @code{@value{AS}} normally produces no output. If for
1533 some reason you are interested in object file output even after
1534 @code{@value{AS}} gives an error message on your program, use the @samp{-Z}
1535 option. If there are any errors, @code{@value{AS}} continues anyways, and
1536 writes an object file after a final warning message of the form @samp{@var{n}
1537 errors, @var{m} warnings, generating bad object file.}
1542 @cindex machine-independent syntax
1543 @cindex syntax, machine-independent
1544 This chapter describes the machine-independent syntax allowed in a
1545 source file. @code{@value{AS}} syntax is similar to what many other
1546 assemblers use; it is inspired by the BSD 4.2
1551 assembler, except that @code{@value{AS}} does not assemble Vax bit-fields.
1555 * Preprocessing:: Preprocessing
1556 * Whitespace:: Whitespace
1557 * Comments:: Comments
1558 * Symbol Intro:: Symbols
1559 * Statements:: Statements
1560 * Constants:: Constants
1564 @section Preprocessing
1566 @cindex preprocessing
1567 The @code{@value{AS}} internal preprocessor:
1569 @cindex whitespace, removed by preprocessor
1571 adjusts and removes extra whitespace. It leaves one space or tab before
1572 the keywords on a line, and turns any other whitespace on the line into
1575 @cindex comments, removed by preprocessor
1577 removes all comments, replacing them with a single space, or an
1578 appropriate number of newlines.
1580 @cindex constants, converted by preprocessor
1582 converts character constants into the appropriate numeric values.
1585 It does not do macro processing, include file handling, or
1586 anything else you may get from your C compiler's preprocessor. You can
1587 do include file processing with the @code{.include} directive
1588 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
1589 to get other ``CPP'' style preprocessing, by giving the input file a
1590 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
1591 Output, gcc.info, Using GNU CC}.
1593 Excess whitespace, comments, and character constants
1594 cannot be used in the portions of the input text that are not
1597 @cindex turning preprocessing on and off
1598 @cindex preprocessing, turning on and off
1601 If the first line of an input file is @code{#NO_APP} or if you use the
1602 @samp{-f} option, whitespace and comments are not removed from the input file.
1603 Within an input file, you can ask for whitespace and comment removal in
1604 specific portions of the by putting a line that says @code{#APP} before the
1605 text that may contain whitespace or comments, and putting a line that says
1606 @code{#NO_APP} after this text. This feature is mainly intend to support
1607 @code{asm} statements in compilers whose output is otherwise free of comments
1614 @dfn{Whitespace} is one or more blanks or tabs, in any order.
1615 Whitespace is used to separate symbols, and to make programs neater for
1616 people to read. Unless within character constants
1617 (@pxref{Characters,,Character Constants}), any whitespace means the same
1618 as exactly one space.
1624 There are two ways of rendering comments to @code{@value{AS}}. In both
1625 cases the comment is equivalent to one space.
1627 Anything from @samp{/*} through the next @samp{*/} is a comment.
1628 This means you may not nest these comments.
1632 The only way to include a newline ('\n') in a comment
1633 is to use this sort of comment.
1636 /* This sort of comment does not nest. */
1639 @cindex line comment character
1640 Anything from the @dfn{line comment} character to the next newline
1641 is considered a comment and is ignored. The line comment character is
1643 @samp{;} for the AMD 29K family;
1646 @samp{;} on the ARC;
1649 @samp{@@} on the ARM;
1652 @samp{;} for the H8/300 family;
1655 @samp{!} for the H8/500 family;
1658 @samp{;} for the HPPA;
1661 @samp{#} on the i960;
1664 @samp{;} for picoJava;
1667 @samp{!} for the Hitachi SH;
1670 @samp{!} on the SPARC;
1673 @samp{#} on the m32r;
1676 @samp{|} on the 680x0;
1679 @samp{#} on the 68HC11 and 68HC12;
1682 @samp{#} on the Vax;
1685 @samp{!} for the Z8000;
1688 @samp{#} on the V850;
1690 see @ref{Machine Dependencies}. @refill
1691 @c FIXME What about i386, m88k, i860?
1694 On some machines there are two different line comment characters. One
1695 character only begins a comment if it is the first non-whitespace character on
1696 a line, while the other always begins a comment.
1700 The V850 assembler also supports a double dash as starting a comment that
1701 extends to the end of the line.
1707 @cindex lines starting with @code{#}
1708 @cindex logical line numbers
1709 To be compatible with past assemblers, lines that begin with @samp{#} have a
1710 special interpretation. Following the @samp{#} should be an absolute
1711 expression (@pxref{Expressions}): the logical line number of the @emph{next}
1712 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
1713 new logical file name. The rest of the line, if any, should be whitespace.
1715 If the first non-whitespace characters on the line are not numeric,
1716 the line is ignored. (Just like a comment.)
1719 # This is an ordinary comment.
1720 # 42-6 "new_file_name" # New logical file name
1721 # This is logical line # 36.
1723 This feature is deprecated, and may disappear from future versions
1724 of @code{@value{AS}}.
1729 @cindex characters used in symbols
1730 @ifclear SPECIAL-SYMS
1731 A @dfn{symbol} is one or more characters chosen from the set of all
1732 letters (both upper and lower case), digits and the three characters
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
1740 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
1746 On most machines, you can also use @code{$} in symbol names; exceptions
1747 are noted in @ref{Machine Dependencies}.
1749 No symbol may begin with a digit. Case is significant.
1750 There is no length limit: all characters are significant. Symbols are
1751 delimited by characters not in that set, or by the beginning of a file
1752 (since the source program must end with a newline, the end of a file is
1753 not a possible symbol delimiter). @xref{Symbols}.
1754 @cindex length of symbols
1759 @cindex statements, structure of
1760 @cindex line separator character
1761 @cindex statement separator character
1763 @ifclear abnormal-separator
1764 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
1765 semicolon (@samp{;}). The newline or semicolon is considered part of
1766 the preceding statement. Newlines and semicolons within character
1767 constants are an exception: they do not end statements.
1769 @ifset abnormal-separator
1771 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
1772 sign (@samp{@@}). The newline or at sign is considered part of the
1773 preceding statement. Newlines and at signs within character constants
1774 are an exception: they do not end statements.
1777 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
1778 point (@samp{!}). The newline or exclamation point is considered part of the
1779 preceding statement. Newlines and exclamation points within character
1780 constants are an exception: they do not end statements.
1783 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
1784 H8/300) a dollar sign (@samp{$}); or (for the
1787 (@samp{;}). The newline or separator character is considered part of
1788 the preceding statement. Newlines and separators within character
1789 constants are an exception: they do not end statements.
1794 A @dfn{statement} ends at a newline character (@samp{\n}) or line
1795 separator character. (The line separator is usually @samp{;}, unless
1796 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
1797 newline or separator character is considered part of the preceding
1798 statement. Newlines and separators within character constants are an
1799 exception: they do not end statements.
1802 @cindex newline, required at file end
1803 @cindex EOF, newline must precede
1804 It is an error to end any statement with end-of-file: the last
1805 character of any input file should be a newline.@refill
1807 An empty statement is allowed, and may include whitespace. It is ignored.
1809 @cindex instructions and directives
1810 @cindex directives and instructions
1811 @c "key symbol" is not used elsewhere in the document; seems pedantic to
1812 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
1814 A statement begins with zero or more labels, optionally followed by a
1815 key symbol which determines what kind of statement it is. The key
1816 symbol determines the syntax of the rest of the statement. If the
1817 symbol begins with a dot @samp{.} then the statement is an assembler
1818 directive: typically valid for any computer. If the symbol begins with
1819 a letter the statement is an assembly language @dfn{instruction}: it
1820 assembles into a machine language instruction.
1822 Different versions of @code{@value{AS}} for different computers
1823 recognize different instructions. In fact, the same symbol may
1824 represent a different instruction in a different computer's assembly
1828 @cindex @code{:} (label)
1829 @cindex label (@code{:})
1830 A label is a symbol immediately followed by a colon (@code{:}).
1831 Whitespace before a label or after a colon is permitted, but you may not
1832 have whitespace between a label's symbol and its colon. @xref{Labels}.
1835 For HPPA targets, labels need not be immediately followed by a colon, but
1836 the definition of a label must begin in column zero. This also implies that
1837 only one label may be defined on each line.
1841 label: .directive followed by something
1842 another_label: # This is an empty statement.
1843 instruction operand_1, operand_2, @dots{}
1850 A constant is a number, written so that its value is known by
1851 inspection, without knowing any context. Like this:
1854 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
1855 .ascii "Ring the bell\7" # A string constant.
1856 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
1857 .float 0f-314159265358979323846264338327\
1858 95028841971.693993751E-40 # - pi, a flonum.
1863 * Characters:: Character Constants
1864 * Numbers:: Number Constants
1868 @subsection Character Constants
1870 @cindex character constants
1871 @cindex constants, character
1872 There are two kinds of character constants. A @dfn{character} stands
1873 for one character in one byte and its value may be used in
1874 numeric expressions. String constants (properly called string
1875 @emph{literals}) are potentially many bytes and their values may not be
1876 used in arithmetic expressions.
1880 * Chars:: Characters
1884 @subsubsection Strings
1886 @cindex string constants
1887 @cindex constants, string
1888 A @dfn{string} is written between double-quotes. It may contain
1889 double-quotes or null characters. The way to get special characters
1890 into a string is to @dfn{escape} these characters: precede them with
1891 a backslash @samp{\} character. For example @samp{\\} represents
1892 one backslash: the first @code{\} is an escape which tells
1893 @code{@value{AS}} to interpret the second character literally as a backslash
1894 (which prevents @code{@value{AS}} from recognizing the second @code{\} as an
1895 escape character). The complete list of escapes follows.
1897 @cindex escape codes, character
1898 @cindex character escape codes
1901 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
1903 @cindex @code{\b} (backspace character)
1904 @cindex backspace (@code{\b})
1906 Mnemonic for backspace; for ASCII this is octal code 010.
1909 @c Mnemonic for EOText; for ASCII this is octal code 004.
1911 @cindex @code{\f} (formfeed character)
1912 @cindex formfeed (@code{\f})
1914 Mnemonic for FormFeed; for ASCII this is octal code 014.
1916 @cindex @code{\n} (newline character)
1917 @cindex newline (@code{\n})
1919 Mnemonic for newline; for ASCII this is octal code 012.
1922 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
1924 @cindex @code{\r} (carriage return character)
1925 @cindex carriage return (@code{\r})
1927 Mnemonic for carriage-Return; for ASCII this is octal code 015.
1930 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
1931 @c other assemblers.
1933 @cindex @code{\t} (tab)
1934 @cindex tab (@code{\t})
1936 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
1939 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
1940 @c @item \x @var{digit} @var{digit} @var{digit}
1941 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
1943 @cindex @code{\@var{ddd}} (octal character code)
1944 @cindex octal character code (@code{\@var{ddd}})
1945 @item \ @var{digit} @var{digit} @var{digit}
1946 An octal character code. The numeric code is 3 octal digits.
1947 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
1948 for example, @code{\008} has the value 010, and @code{\009} the value 011.
1950 @cindex @code{\@var{xd...}} (hex character code)
1951 @cindex hex character code (@code{\@var{xd...}})
1952 @item \@code{x} @var{hex-digits...}
1953 A hex character code. All trailing hex digits are combined. Either upper or
1954 lower case @code{x} works.
1956 @cindex @code{\\} (@samp{\} character)
1957 @cindex backslash (@code{\\})
1959 Represents one @samp{\} character.
1962 @c Represents one @samp{'} (accent acute) character.
1963 @c This is needed in single character literals
1964 @c (@xref{Characters,,Character Constants}.) to represent
1967 @cindex @code{\"} (doublequote character)
1968 @cindex doublequote (@code{\"})
1970 Represents one @samp{"} character. Needed in strings to represent
1971 this character, because an unescaped @samp{"} would end the string.
1973 @item \ @var{anything-else}
1974 Any other character when escaped by @kbd{\} gives a warning, but
1975 assembles as if the @samp{\} was not present. The idea is that if
1976 you used an escape sequence you clearly didn't want the literal
1977 interpretation of the following character. However @code{@value{AS}} has no
1978 other interpretation, so @code{@value{AS}} knows it is giving you the wrong
1979 code and warns you of the fact.
1982 Which characters are escapable, and what those escapes represent,
1983 varies widely among assemblers. The current set is what we think
1984 the BSD 4.2 assembler recognizes, and is a subset of what most C
1985 compilers recognize. If you are in doubt, do not use an escape
1989 @subsubsection Characters
1991 @cindex single character constant
1992 @cindex character, single
1993 @cindex constant, single character
1994 A single character may be written as a single quote immediately
1995 followed by that character. The same escapes apply to characters as
1996 to strings. So if you want to write the character backslash, you
1997 must write @kbd{'\\} where the first @code{\} escapes the second
1998 @code{\}. As you can see, the quote is an acute accent, not a
1999 grave accent. A newline
2001 @ifclear abnormal-separator
2002 (or semicolon @samp{;})
2004 @ifset abnormal-separator
2006 (or at sign @samp{@@})
2009 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2015 immediately following an acute accent is taken as a literal character
2016 and does not count as the end of a statement. The value of a character
2017 constant in a numeric expression is the machine's byte-wide code for
2018 that character. @code{@value{AS}} assumes your character code is ASCII:
2019 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2022 @subsection Number Constants
2024 @cindex constants, number
2025 @cindex number constants
2026 @code{@value{AS}} distinguishes three kinds of numbers according to how they
2027 are stored in the target machine. @emph{Integers} are numbers that
2028 would fit into an @code{int} in the C language. @emph{Bignums} are
2029 integers, but they are stored in more than 32 bits. @emph{Flonums}
2030 are floating point numbers, described below.
2033 * Integers:: Integers
2038 * Bit Fields:: Bit Fields
2044 @subsubsection Integers
2046 @cindex constants, integer
2048 @cindex binary integers
2049 @cindex integers, binary
2050 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2051 the binary digits @samp{01}.
2053 @cindex octal integers
2054 @cindex integers, octal
2055 An octal integer is @samp{0} followed by zero or more of the octal
2056 digits (@samp{01234567}).
2058 @cindex decimal integers
2059 @cindex integers, decimal
2060 A decimal integer starts with a non-zero digit followed by zero or
2061 more digits (@samp{0123456789}).
2063 @cindex hexadecimal integers
2064 @cindex integers, hexadecimal
2065 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2066 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2068 Integers have the usual values. To denote a negative integer, use
2069 the prefix operator @samp{-} discussed under expressions
2070 (@pxref{Prefix Ops,,Prefix Operators}).
2073 @subsubsection Bignums
2076 @cindex constants, bignum
2077 A @dfn{bignum} has the same syntax and semantics as an integer
2078 except that the number (or its negative) takes more than 32 bits to
2079 represent in binary. The distinction is made because in some places
2080 integers are permitted while bignums are not.
2083 @subsubsection Flonums
2085 @cindex floating point numbers
2086 @cindex constants, floating point
2088 @cindex precision, floating point
2089 A @dfn{flonum} represents a floating point number. The translation is
2090 indirect: a decimal floating point number from the text is converted by
2091 @code{@value{AS}} to a generic binary floating point number of more than
2092 sufficient precision. This generic floating point number is converted
2093 to a particular computer's floating point format (or formats) by a
2094 portion of @code{@value{AS}} specialized to that computer.
2096 A flonum is written by writing (in order)
2101 (@samp{0} is optional on the HPPA.)
2105 A letter, to tell @code{@value{AS}} the rest of the number is a flonum.
2107 @kbd{e} is recommended. Case is not important.
2109 @c FIXME: verify if flonum syntax really this vague for most cases
2110 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2111 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2114 On the H8/300, H8/500,
2116 and AMD 29K architectures, the letter must be
2117 one of the letters @samp{DFPRSX} (in upper or lower case).
2119 On the ARC, the letter must be one of the letters @samp{DFRS}
2120 (in upper or lower case).
2122 On the Intel 960 architecture, the letter must be
2123 one of the letters @samp{DFT} (in upper or lower case).
2125 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2129 One of the letters @samp{DFPRSX} (in upper or lower case).
2132 One of the letters @samp{DFRS} (in upper or lower case).
2135 One of the letters @samp{DFPRSX} (in upper or lower case).
2138 The letter @samp{E} (upper case only).
2141 One of the letters @samp{DFT} (in upper or lower case).
2146 An optional sign: either @samp{+} or @samp{-}.
2149 An optional @dfn{integer part}: zero or more decimal digits.
2152 An optional @dfn{fractional part}: @samp{.} followed by zero
2153 or more decimal digits.
2156 An optional exponent, consisting of:
2160 An @samp{E} or @samp{e}.
2161 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2162 @c principle this can perfectly well be different on different targets.
2164 Optional sign: either @samp{+} or @samp{-}.
2166 One or more decimal digits.
2171 At least one of the integer part or the fractional part must be
2172 present. The floating point number has the usual base-10 value.
2174 @code{@value{AS}} does all processing using integers. Flonums are computed
2175 independently of any floating point hardware in the computer running
2180 @c Bit fields are written as a general facility but are also controlled
2181 @c by a conditional-compilation flag---which is as of now (21mar91)
2182 @c turned on only by the i960 config of GAS.
2184 @subsubsection Bit Fields
2187 @cindex constants, bit field
2188 You can also define numeric constants as @dfn{bit fields}.
2189 specify two numbers separated by a colon---
2191 @var{mask}:@var{value}
2194 @code{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2197 The resulting number is then packed
2199 @c this conditional paren in case bit fields turned on elsewhere than 960
2200 (in host-dependent byte order)
2202 into a field whose width depends on which assembler directive has the
2203 bit-field as its argument. Overflow (a result from the bitwise and
2204 requiring more binary digits to represent) is not an error; instead,
2205 more constants are generated, of the specified width, beginning with the
2206 least significant digits.@refill
2208 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2209 @code{.short}, and @code{.word} accept bit-field arguments.
2214 @chapter Sections and Relocation
2219 * Secs Background:: Background
2220 * Ld Sections:: Linker Sections
2221 * As Sections:: Assembler Internal Sections
2222 * Sub-Sections:: Sub-Sections
2226 @node Secs Background
2229 Roughly, a section is a range of addresses, with no gaps; all data
2230 ``in'' those addresses is treated the same for some particular purpose.
2231 For example there may be a ``read only'' section.
2233 @cindex linker, and assembler
2234 @cindex assembler, and linker
2235 The linker @code{@value{LD}} reads many object files (partial programs) and
2236 combines their contents to form a runnable program. When @code{@value{AS}}
2237 emits an object file, the partial program is assumed to start at address 0.
2238 @code{@value{LD}} assigns the final addresses for the partial program, so that
2239 different partial programs do not overlap. This is actually an
2240 oversimplification, but it suffices to explain how @code{@value{AS}} uses
2243 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2244 addresses. These blocks slide to their run-time addresses as rigid
2245 units; their length does not change and neither does the order of bytes
2246 within them. Such a rigid unit is called a @emph{section}. Assigning
2247 run-time addresses to sections is called @dfn{relocation}. It includes
2248 the task of adjusting mentions of object-file addresses so they refer to
2249 the proper run-time addresses.
2251 For the H8/300 and H8/500,
2252 and for the Hitachi SH,
2253 @code{@value{AS}} pads sections if needed to
2254 ensure they end on a word (sixteen bit) boundary.
2257 @cindex standard assembler sections
2258 An object file written by @code{@value{AS}} has at least three sections, any
2259 of which may be empty. These are named @dfn{text}, @dfn{data} and
2264 When it generates COFF output,
2266 @code{@value{AS}} can also generate whatever other named sections you specify
2267 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2268 If you do not use any directives that place output in the @samp{.text}
2269 or @samp{.data} sections, these sections still exist, but are empty.
2274 When @code{@value{AS}} generates SOM or ELF output for the HPPA,
2276 @code{@value{AS}} can also generate whatever other named sections you
2277 specify using the @samp{.space} and @samp{.subspace} directives. See
2278 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2279 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2280 assembler directives.
2283 Additionally, @code{@value{AS}} uses different names for the standard
2284 text, data, and bss sections when generating SOM output. Program text
2285 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2286 BSS into @samp{$BSS$}.
2290 Within the object file, the text section starts at address @code{0}, the
2291 data section follows, and the bss section follows the data section.
2294 When generating either SOM or ELF output files on the HPPA, the text
2295 section starts at address @code{0}, the data section at address
2296 @code{0x4000000}, and the bss section follows the data section.
2299 To let @code{@value{LD}} know which data changes when the sections are
2300 relocated, and how to change that data, @code{@value{AS}} also writes to the
2301 object file details of the relocation needed. To perform relocation
2302 @code{@value{LD}} must know, each time an address in the object
2306 Where in the object file is the beginning of this reference to
2309 How long (in bytes) is this reference?
2311 Which section does the address refer to? What is the numeric value of
2313 (@var{address}) @minus{} (@var{start-address of section})?
2316 Is the reference to an address ``Program-Counter relative''?
2319 @cindex addresses, format of
2320 @cindex section-relative addressing
2321 In fact, every address @code{@value{AS}} ever uses is expressed as
2323 (@var{section}) + (@var{offset into section})
2326 Further, most expressions @code{@value{AS}} computes have this section-relative
2329 (For some object formats, such as SOM for the HPPA, some expressions are
2330 symbol-relative instead.)
2333 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2334 @var{N} into section @var{secname}.''
2336 Apart from text, data and bss sections you need to know about the
2337 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2338 addresses in the absolute section remain unchanged. For example, address
2339 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2340 @code{@value{LD}}. Although the linker never arranges two partial programs'
2341 data sections with overlapping addresses after linking, @emph{by definition}
2342 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2343 part of a program is always the same address when the program is running as
2344 address @code{@{absolute@ 239@}} in any other part of the program.
2346 The idea of sections is extended to the @dfn{undefined} section. Any
2347 address whose section is unknown at assembly time is by definition
2348 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2349 Since numbers are always defined, the only way to generate an undefined
2350 address is to mention an undefined symbol. A reference to a named
2351 common block would be such a symbol: its value is unknown at assembly
2352 time so it has section @emph{undefined}.
2354 By analogy the word @emph{section} is used to describe groups of sections in
2355 the linked program. @code{@value{LD}} puts all partial programs' text
2356 sections in contiguous addresses in the linked program. It is
2357 customary to refer to the @emph{text section} of a program, meaning all
2358 the addresses of all partial programs' text sections. Likewise for
2359 data and bss sections.
2361 Some sections are manipulated by @code{@value{LD}}; others are invented for
2362 use of @code{@value{AS}} and have no meaning except during assembly.
2365 @section Linker Sections
2366 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2371 @cindex named sections
2372 @cindex sections, named
2373 @item named sections
2376 @cindex text section
2377 @cindex data section
2381 These sections hold your program. @code{@value{AS}} and @code{@value{LD}} treat them as
2382 separate but equal sections. Anything you can say of one section is
2385 When the program is running, however, it is
2386 customary for the text section to be unalterable. The
2387 text section is often shared among processes: it contains
2388 instructions, constants and the like. The data section of a running
2389 program is usually alterable: for example, C variables would be stored
2390 in the data section.
2395 This section contains zeroed bytes when your program begins running. It
2396 is used to hold unitialized variables or common storage. The length of
2397 each partial program's bss section is important, but because it starts
2398 out containing zeroed bytes there is no need to store explicit zero
2399 bytes in the object file. The bss section was invented to eliminate
2400 those explicit zeros from object files.
2402 @cindex absolute section
2403 @item absolute section
2404 Address 0 of this section is always ``relocated'' to runtime address 0.
2405 This is useful if you want to refer to an address that @code{@value{LD}} must
2406 not change when relocating. In this sense we speak of absolute
2407 addresses being ``unrelocatable'': they do not change during relocation.
2409 @cindex undefined section
2410 @item undefined section
2411 This ``section'' is a catch-all for address references to objects not in
2412 the preceding sections.
2413 @c FIXME: ref to some other doc on obj-file formats could go here.
2416 @cindex relocation example
2417 An idealized example of three relocatable sections follows.
2419 The example uses the traditional section names @samp{.text} and @samp{.data}.
2421 Memory addresses are on the horizontal axis.
2425 @c END TEXI2ROFF-KILL
2428 partial program # 1: |ttttt|dddd|00|
2435 partial program # 2: |TTT|DDD|000|
2438 +--+---+-----+--+----+---+-----+~~
2439 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2440 +--+---+-----+--+----+---+-----+~~
2442 addresses: 0 @dots{}
2449 \line{\it Partial program \#1: \hfil}
2450 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2451 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2453 \line{\it Partial program \#2: \hfil}
2454 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2455 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2457 \line{\it linked program: \hfil}
2458 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2459 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2460 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2461 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2463 \line{\it addresses: \hfil}
2467 @c END TEXI2ROFF-KILL
2470 @section Assembler Internal Sections
2472 @cindex internal assembler sections
2473 @cindex sections in messages, internal
2474 These sections are meant only for the internal use of @code{@value{AS}}. They
2475 have no meaning at run-time. You do not really need to know about these
2476 sections for most purposes; but they can be mentioned in @code{@value{AS}}
2477 warning messages, so it might be helpful to have an idea of their
2478 meanings to @code{@value{AS}}. These sections are used to permit the
2479 value of every expression in your assembly language program to be a
2480 section-relative address.
2483 @cindex assembler internal logic error
2484 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2485 An internal assembler logic error has been found. This means there is a
2486 bug in the assembler.
2488 @cindex expr (internal section)
2490 The assembler stores complex expression internally as combinations of
2491 symbols. When it needs to represent an expression as a symbol, it puts
2492 it in the expr section.
2494 @c FIXME item transfer[t] vector preload
2495 @c FIXME item transfer[t] vector postload
2496 @c FIXME item register
2500 @section Sub-Sections
2502 @cindex numbered subsections
2503 @cindex grouping data
2509 fall into two sections: text and data.
2511 You may have separate groups of
2513 data in named sections
2517 data in named sections
2523 that you want to end up near to each other in the object file, even though they
2524 are not contiguous in the assembler source. @code{@value{AS}} allows you to
2525 use @dfn{subsections} for this purpose. Within each section, there can be
2526 numbered subsections with values from 0 to 8192. Objects assembled into the
2527 same subsection go into the object file together with other objects in the same
2528 subsection. For example, a compiler might want to store constants in the text
2529 section, but might not want to have them interspersed with the program being
2530 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2531 section of code being output, and a @samp{.text 1} before each group of
2532 constants being output.
2534 Subsections are optional. If you do not use subsections, everything
2535 goes in subsection number zero.
2538 Each subsection is zero-padded up to a multiple of four bytes.
2539 (Subsections may be padded a different amount on different flavors
2540 of @code{@value{AS}}.)
2544 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2545 boundary (two bytes).
2546 The same is true on the Hitachi SH.
2549 @c FIXME section padding (alignment)?
2550 @c Rich Pixley says padding here depends on target obj code format; that
2551 @c doesn't seem particularly useful to say without further elaboration,
2552 @c so for now I say nothing about it. If this is a generic BFD issue,
2553 @c these paragraphs might need to vanish from this manual, and be
2554 @c discussed in BFD chapter of binutils (or some such).
2557 On the AMD 29K family, no particular padding is added to section or
2558 subsection sizes; @value{AS} forces no alignment on this platform.
2562 Subsections appear in your object file in numeric order, lowest numbered
2563 to highest. (All this to be compatible with other people's assemblers.)
2564 The object file contains no representation of subsections; @code{@value{LD}} and
2565 other programs that manipulate object files see no trace of them.
2566 They just see all your text subsections as a text section, and all your
2567 data subsections as a data section.
2569 To specify which subsection you want subsequent statements assembled
2570 into, use a numeric argument to specify it, in a @samp{.text
2571 @var{expression}} or a @samp{.data @var{expression}} statement.
2574 When generating COFF output, you
2579 can also use an extra subsection
2580 argument with arbitrary named sections: @samp{.section @var{name},
2583 @var{Expression} should be an absolute expression.
2584 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
2585 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
2586 begins in @code{text 0}. For instance:
2588 .text 0 # The default subsection is text 0 anyway.
2589 .ascii "This lives in the first text subsection. *"
2591 .ascii "But this lives in the second text subsection."
2593 .ascii "This lives in the data section,"
2594 .ascii "in the first data subsection."
2596 .ascii "This lives in the first text section,"
2597 .ascii "immediately following the asterisk (*)."
2600 Each section has a @dfn{location counter} incremented by one for every byte
2601 assembled into that section. Because subsections are merely a convenience
2602 restricted to @code{@value{AS}} there is no concept of a subsection location
2603 counter. There is no way to directly manipulate a location counter---but the
2604 @code{.align} directive changes it, and any label definition captures its
2605 current value. The location counter of the section where statements are being
2606 assembled is said to be the @dfn{active} location counter.
2609 @section bss Section
2612 @cindex common variable storage
2613 The bss section is used for local common variable storage.
2614 You may allocate address space in the bss section, but you may
2615 not dictate data to load into it before your program executes. When
2616 your program starts running, all the contents of the bss
2617 section are zeroed bytes.
2619 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
2620 @ref{Lcomm,,@code{.lcomm}}.
2622 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
2623 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
2626 When assembling for a target which supports multiple sections, such as ELF or
2627 COFF, you may switch into the @code{.bss} section and define symbols as usual;
2628 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
2629 section. Typically the section will only contain symbol definitions and
2630 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
2637 Symbols are a central concept: the programmer uses symbols to name
2638 things, the linker uses symbols to link, and the debugger uses symbols
2642 @cindex debuggers, and symbol order
2643 @emph{Warning:} @code{@value{AS}} does not place symbols in the object file in
2644 the same order they were declared. This may break some debuggers.
2649 * Setting Symbols:: Giving Symbols Other Values
2650 * Symbol Names:: Symbol Names
2651 * Dot:: The Special Dot Symbol
2652 * Symbol Attributes:: Symbol Attributes
2659 A @dfn{label} is written as a symbol immediately followed by a colon
2660 @samp{:}. The symbol then represents the current value of the
2661 active location counter, and is, for example, a suitable instruction
2662 operand. You are warned if you use the same symbol to represent two
2663 different locations: the first definition overrides any other
2667 On the HPPA, the usual form for a label need not be immediately followed by a
2668 colon, but instead must start in column zero. Only one label may be defined on
2669 a single line. To work around this, the HPPA version of @code{@value{AS}} also
2670 provides a special directive @code{.label} for defining labels more flexibly.
2673 @node Setting Symbols
2674 @section Giving Symbols Other Values
2676 @cindex assigning values to symbols
2677 @cindex symbol values, assigning
2678 A symbol can be given an arbitrary value by writing a symbol, followed
2679 by an equals sign @samp{=}, followed by an expression
2680 (@pxref{Expressions}). This is equivalent to using the @code{.set}
2681 directive. @xref{Set,,@code{.set}}.
2684 @section Symbol Names
2686 @cindex symbol names
2687 @cindex names, symbol
2688 @ifclear SPECIAL-SYMS
2689 Symbol names begin with a letter or with one of @samp{._}. On most
2690 machines, you can also use @code{$} in symbol names; exceptions are
2691 noted in @ref{Machine Dependencies}. That character may be followed by any
2692 string of digits, letters, dollar signs (unless otherwise noted in
2693 @ref{Machine Dependencies}), and underscores.
2696 For the AMD 29K family, @samp{?} is also allowed in the
2697 body of a symbol name, though not at its beginning.
2702 Symbol names begin with a letter or with one of @samp{._}. On the
2704 H8/500, you can also use @code{$} in symbol names. That character may
2705 be followed by any string of digits, letters, dollar signs (save on the
2706 H8/300), and underscores.
2710 Case of letters is significant: @code{foo} is a different symbol name
2713 Each symbol has exactly one name. Each name in an assembly language program
2714 refers to exactly one symbol. You may use that symbol name any number of times
2717 @subheading Local Symbol Names
2719 @cindex local symbol names
2720 @cindex symbol names, local
2721 @cindex temporary symbol names
2722 @cindex symbol names, temporary
2723 Local symbols help compilers and programmers use names temporarily.
2724 There are ten local symbol names, which are re-used throughout the
2725 program. You may refer to them using the names @samp{0} @samp{1}
2726 @dots{} @samp{9}. To define a local symbol, write a label of the form
2727 @samp{@b{N}:} (where @b{N} represents any digit). To refer to the most
2728 recent previous definition of that symbol write @samp{@b{N}b}, using the
2729 same digit as when you defined the label. To refer to the next
2730 definition of a local label, write @samp{@b{N}f}---where @b{N} gives you
2731 a choice of 10 forward references. The @samp{b} stands for
2732 ``backwards'' and the @samp{f} stands for ``forwards''.
2734 Local symbols are not emitted by the current @sc{gnu} C compiler.
2736 There is no restriction on how you can use these labels, but
2737 remember that at any point in the assembly you can refer to at most
2738 10 prior local labels and to at most 10 forward local labels.
2740 Local symbol names are only a notation device. They are immediately
2741 transformed into more conventional symbol names before the assembler
2742 uses them. The symbol names stored in the symbol table, appearing in
2743 error messages and optionally emitted to the object file have these
2748 All local labels begin with @samp{L}. Normally both @code{@value{AS}} and
2749 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
2750 used for symbols you are never intended to see. If you use the
2751 @samp{-L} option then @code{@value{AS}} retains these symbols in the
2752 object file. If you also instruct @code{@value{LD}} to retain these symbols,
2753 you may use them in debugging.
2756 If the label is written @samp{0:} then the digit is @samp{0}.
2757 If the label is written @samp{1:} then the digit is @samp{1}.
2758 And so on up through @samp{9:}.
2761 This unusual character is included so you do not accidentally invent
2762 a symbol of the same name. The character has ASCII value
2765 @item @emph{ordinal number}
2766 This is a serial number to keep the labels distinct. The first
2767 @samp{0:} gets the number @samp{1}; The 15th @samp{0:} gets the
2768 number @samp{15}; @emph{etc.}. Likewise for the other labels @samp{1:}
2772 For instance, the first @code{1:} is named @code{L1@kbd{C-A}1}, the 44th
2773 @code{3:} is named @code{L3@kbd{C-A}44}.
2776 @section The Special Dot Symbol
2778 @cindex dot (symbol)
2779 @cindex @code{.} (symbol)
2780 @cindex current address
2781 @cindex location counter
2782 The special symbol @samp{.} refers to the current address that
2783 @code{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
2784 .long .} defines @code{melvin} to contain its own address.
2785 Assigning a value to @code{.} is treated the same as a @code{.org}
2786 directive. Thus, the expression @samp{.=.+4} is the same as saying
2787 @ifclear no-space-dir
2796 @node Symbol Attributes
2797 @section Symbol Attributes
2799 @cindex symbol attributes
2800 @cindex attributes, symbol
2801 Every symbol has, as well as its name, the attributes ``Value'' and
2802 ``Type''. Depending on output format, symbols can also have auxiliary
2805 The detailed definitions are in @file{a.out.h}.
2808 If you use a symbol without defining it, @code{@value{AS}} assumes zero for
2809 all these attributes, and probably won't warn you. This makes the
2810 symbol an externally defined symbol, which is generally what you
2814 * Symbol Value:: Value
2815 * Symbol Type:: Type
2818 * a.out Symbols:: Symbol Attributes: @code{a.out}
2822 * a.out Symbols:: Symbol Attributes: @code{a.out}
2825 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
2830 * COFF Symbols:: Symbol Attributes for COFF
2833 * SOM Symbols:: Symbol Attributes for SOM
2840 @cindex value of a symbol
2841 @cindex symbol value
2842 The value of a symbol is (usually) 32 bits. For a symbol which labels a
2843 location in the text, data, bss or absolute sections the value is the
2844 number of addresses from the start of that section to the label.
2845 Naturally for text, data and bss sections the value of a symbol changes
2846 as @code{@value{LD}} changes section base addresses during linking. Absolute
2847 symbols' values do not change during linking: that is why they are
2850 The value of an undefined symbol is treated in a special way. If it is
2851 0 then the symbol is not defined in this assembler source file, and
2852 @code{@value{LD}} tries to determine its value from other files linked into the
2853 same program. You make this kind of symbol simply by mentioning a symbol
2854 name without defining it. A non-zero value represents a @code{.comm}
2855 common declaration. The value is how much common storage to reserve, in
2856 bytes (addresses). The symbol refers to the first address of the
2862 @cindex type of a symbol
2864 The type attribute of a symbol contains relocation (section)
2865 information, any flag settings indicating that a symbol is external, and
2866 (optionally), other information for linkers and debuggers. The exact
2867 format depends on the object-code output format in use.
2872 @c The following avoids a "widow" subsection title. @group would be
2873 @c better if it were available outside examples.
2876 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
2878 @cindex @code{b.out} symbol attributes
2879 @cindex symbol attributes, @code{b.out}
2880 These symbol attributes appear only when @code{@value{AS}} is configured for
2881 one of the Berkeley-descended object output formats---@code{a.out} or
2887 @subsection Symbol Attributes: @code{a.out}
2889 @cindex @code{a.out} symbol attributes
2890 @cindex symbol attributes, @code{a.out}
2896 @subsection Symbol Attributes: @code{a.out}
2898 @cindex @code{a.out} symbol attributes
2899 @cindex symbol attributes, @code{a.out}
2903 * Symbol Desc:: Descriptor
2904 * Symbol Other:: Other
2908 @subsubsection Descriptor
2910 @cindex descriptor, of @code{a.out} symbol
2911 This is an arbitrary 16-bit value. You may establish a symbol's
2912 descriptor value by using a @code{.desc} statement
2913 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
2917 @subsubsection Other
2919 @cindex other attribute, of @code{a.out} symbol
2920 This is an arbitrary 8-bit value. It means nothing to @code{@value{AS}}.
2925 @subsection Symbol Attributes for COFF
2927 @cindex COFF symbol attributes
2928 @cindex symbol attributes, COFF
2930 The COFF format supports a multitude of auxiliary symbol attributes;
2931 like the primary symbol attributes, they are set between @code{.def} and
2932 @code{.endef} directives.
2934 @subsubsection Primary Attributes
2936 @cindex primary attributes, COFF symbols
2937 The symbol name is set with @code{.def}; the value and type,
2938 respectively, with @code{.val} and @code{.type}.
2940 @subsubsection Auxiliary Attributes
2942 @cindex auxiliary attributes, COFF symbols
2943 The @code{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
2944 @code{.size}, and @code{.tag} can generate auxiliary symbol table
2945 information for COFF.
2950 @subsection Symbol Attributes for SOM
2952 @cindex SOM symbol attributes
2953 @cindex symbol attributes, SOM
2955 The SOM format for the HPPA supports a multitude of symbol attributes set with
2956 the @code{.EXPORT} and @code{.IMPORT} directives.
2958 The attributes are described in @cite{HP9000 Series 800 Assembly
2959 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
2960 @code{EXPORT} assembler directive documentation.
2964 @chapter Expressions
2968 @cindex numeric values
2969 An @dfn{expression} specifies an address or numeric value.
2970 Whitespace may precede and/or follow an expression.
2972 The result of an expression must be an absolute number, or else an offset into
2973 a particular section. If an expression is not absolute, and there is not
2974 enough information when @code{@value{AS}} sees the expression to know its
2975 section, a second pass over the source program might be necessary to interpret
2976 the expression---but the second pass is currently not implemented.
2977 @code{@value{AS}} aborts with an error message in this situation.
2980 * Empty Exprs:: Empty Expressions
2981 * Integer Exprs:: Integer Expressions
2985 @section Empty Expressions
2987 @cindex empty expressions
2988 @cindex expressions, empty
2989 An empty expression has no value: it is just whitespace or null.
2990 Wherever an absolute expression is required, you may omit the
2991 expression, and @code{@value{AS}} assumes a value of (absolute) 0. This
2992 is compatible with other assemblers.
2995 @section Integer Expressions
2997 @cindex integer expressions
2998 @cindex expressions, integer
2999 An @dfn{integer expression} is one or more @emph{arguments} delimited
3000 by @emph{operators}.
3003 * Arguments:: Arguments
3004 * Operators:: Operators
3005 * Prefix Ops:: Prefix Operators
3006 * Infix Ops:: Infix Operators
3010 @subsection Arguments
3012 @cindex expression arguments
3013 @cindex arguments in expressions
3014 @cindex operands in expressions
3015 @cindex arithmetic operands
3016 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3017 contexts arguments are sometimes called ``arithmetic operands''. In
3018 this manual, to avoid confusing them with the ``instruction operands'' of
3019 the machine language, we use the term ``argument'' to refer to parts of
3020 expressions only, reserving the word ``operand'' to refer only to machine
3021 instruction operands.
3023 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3024 @var{section} is one of text, data, bss, absolute,
3025 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3028 Numbers are usually integers.
3030 A number can be a flonum or bignum. In this case, you are warned
3031 that only the low order 32 bits are used, and @code{@value{AS}} pretends
3032 these 32 bits are an integer. You may write integer-manipulating
3033 instructions that act on exotic constants, compatible with other
3036 @cindex subexpressions
3037 Subexpressions are a left parenthesis @samp{(} followed by an integer
3038 expression, followed by a right parenthesis @samp{)}; or a prefix
3039 operator followed by an argument.
3042 @subsection Operators
3044 @cindex operators, in expressions
3045 @cindex arithmetic functions
3046 @cindex functions, in expressions
3047 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3048 operators are followed by an argument. Infix operators appear
3049 between their arguments. Operators may be preceded and/or followed by
3053 @subsection Prefix Operator
3055 @cindex prefix operators
3056 @code{@value{AS}} has the following @dfn{prefix operators}. They each take
3057 one argument, which must be absolute.
3059 @c the tex/end tex stuff surrounding this small table is meant to make
3060 @c it align, on the printed page, with the similar table in the next
3061 @c section (which is inside an enumerate).
3063 \global\advance\leftskip by \itemindent
3068 @dfn{Negation}. Two's complement negation.
3070 @dfn{Complementation}. Bitwise not.
3074 \global\advance\leftskip by -\itemindent
3078 @subsection Infix Operators
3080 @cindex infix operators
3081 @cindex operators, permitted arguments
3082 @dfn{Infix operators} take two arguments, one on either side. Operators
3083 have precedence, but operations with equal precedence are performed left
3084 to right. Apart from @code{+} or @code{-}, both arguments must be
3085 absolute, and the result is absolute.
3088 @cindex operator precedence
3089 @cindex precedence of operators
3096 @dfn{Multiplication}.
3099 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3106 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3110 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3114 Intermediate precedence
3119 @dfn{Bitwise Inclusive Or}.
3125 @dfn{Bitwise Exclusive Or}.
3128 @dfn{Bitwise Or Not}.
3135 @cindex addition, permitted arguments
3136 @cindex plus, permitted arguments
3137 @cindex arguments for addition
3139 @dfn{Addition}. If either argument is absolute, the result has the section of
3140 the other argument. You may not add together arguments from different
3143 @cindex subtraction, permitted arguments
3144 @cindex minus, permitted arguments
3145 @cindex arguments for subtraction
3147 @dfn{Subtraction}. If the right argument is absolute, the
3148 result has the section of the left argument.
3149 If both arguments are in the same section, the result is absolute.
3150 You may not subtract arguments from different sections.
3151 @c FIXME is there still something useful to say about undefined - undefined ?
3155 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3156 address; you can only have a defined section in one of the two arguments.
3159 @chapter Assembler Directives
3161 @cindex directives, machine independent
3162 @cindex pseudo-ops, machine independent
3163 @cindex machine independent directives
3164 All assembler directives have names that begin with a period (@samp{.}).
3165 The rest of the name is letters, usually in lower case.
3167 This chapter discusses directives that are available regardless of the
3168 target machine configuration for the @sc{gnu} assembler.
3170 Some machine configurations provide additional directives.
3171 @xref{Machine Dependencies}.
3174 @ifset machine-directives
3175 @xref{Machine Dependencies} for additional directives.
3180 * Abort:: @code{.abort}
3182 * ABORT:: @code{.ABORT}
3185 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3186 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3187 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3188 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3189 * Byte:: @code{.byte @var{expressions}}
3190 * Comm:: @code{.comm @var{symbol} , @var{length} }
3191 * Data:: @code{.data @var{subsection}}
3193 * Def:: @code{.def @var{name}}
3196 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3202 * Double:: @code{.double @var{flonums}}
3203 * Eject:: @code{.eject}
3204 * Else:: @code{.else}
3205 * Elseif:: @code{.elseif}
3208 * Endef:: @code{.endef}
3211 * Endfunc:: @code{.endfunc}
3212 * Endif:: @code{.endif}
3213 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3214 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3216 * Exitm:: @code{.exitm}
3217 * Extern:: @code{.extern}
3218 * Fail:: @code{.fail}
3219 @ifclear no-file-dir
3220 * File:: @code{.file @var{string}}
3223 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3224 * Float:: @code{.float @var{flonums}}
3225 * Func:: @code{.func}
3226 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3227 * hword:: @code{.hword @var{expressions}}
3228 * Ident:: @code{.ident}
3229 * If:: @code{.if @var{absolute expression}}
3230 * Include:: @code{.include "@var{file}"}
3231 * Int:: @code{.int @var{expressions}}
3232 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3233 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3234 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3235 * Lflags:: @code{.lflags}
3236 @ifclear no-line-dir
3237 * Line:: @code{.line @var{line-number}}
3240 * Ln:: @code{.ln @var{line-number}}
3241 * Linkonce:: @code{.linkonce [@var{type}]}
3242 * List:: @code{.list}
3243 * Long:: @code{.long @var{expressions}}
3245 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3248 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3249 * MRI:: @code{.mri @var{val}}
3251 * Nolist:: @code{.nolist}
3252 * Octa:: @code{.octa @var{bignums}}
3253 * Org:: @code{.org @var{new-lc} , @var{fill}}
3254 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3255 * Print:: @code{.print @var{string}}
3256 * Psize:: @code{.psize @var{lines}, @var{columns}}
3257 * Purgem:: @code{.purgem @var{name}}
3258 * Quad:: @code{.quad @var{bignums}}
3259 * Rept:: @code{.rept @var{count}}
3260 * Sbttl:: @code{.sbttl "@var{subheading}"}
3262 * Scl:: @code{.scl @var{class}}
3263 * Section:: @code{.section @var{name}, @var{subsection}}
3266 * Set:: @code{.set @var{symbol}, @var{expression}}
3267 * Short:: @code{.short @var{expressions}}
3268 * Single:: @code{.single @var{flonums}}
3270 * Size:: @code{.size}
3273 * Skip:: @code{.skip @var{size} , @var{fill}}
3274 * Sleb128:: @code{.sleb128 @var{expressions}}
3275 * Space:: @code{.space @var{size} , @var{fill}}
3277 * Stab:: @code{.stabd, .stabn, .stabs}
3280 * String:: @code{.string "@var{str}"}
3281 * Struct:: @code{.struct @var{expression}}
3283 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3286 * Tag:: @code{.tag @var{structname}}
3289 * Text:: @code{.text @var{subsection}}
3290 * Title:: @code{.title "@var{heading}"}
3292 * Type:: @code{.type @var{int}}
3293 * Val:: @code{.val @var{addr}}
3296 * Visibility:: @code{.internal @var{name}, .hidden @var{name}, .protected @var{name}}
3299 * Uleb128:: @code{.uleb128 @var{expressions}}
3300 * Word:: @code{.word @var{expressions}}
3301 * Deprecated:: Deprecated Directives
3305 @section @code{.abort}
3307 @cindex @code{abort} directive
3308 @cindex stopping the assembly
3309 This directive stops the assembly immediately. It is for
3310 compatibility with other assemblers. The original idea was that the
3311 assembly language source would be piped into the assembler. If the sender
3312 of the source quit, it could use this directive tells @code{@value{AS}} to
3313 quit also. One day @code{.abort} will not be supported.
3317 @section @code{.ABORT}
3319 @cindex @code{ABORT} directive
3320 When producing COFF output, @code{@value{AS}} accepts this directive as a
3321 synonym for @samp{.abort}.
3324 When producing @code{b.out} output, @code{@value{AS}} accepts this directive,
3330 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3332 @cindex padding the location counter
3333 @cindex @code{align} directive
3334 Pad the location counter (in the current subsection) to a particular storage
3335 boundary. The first expression (which must be absolute) is the alignment
3336 required, as described below.
3338 The second expression (also absolute) gives the fill value to be stored in the
3339 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3340 padding bytes are normally zero. However, on some systems, if the section is
3341 marked as containing code and the fill value is omitted, the space is filled
3342 with no-op instructions.
3344 The third expression is also absolute, and is also optional. If it is present,
3345 it is the maximum number of bytes that should be skipped by this alignment
3346 directive. If doing the alignment would require skipping more bytes than the
3347 specified maximum, then the alignment is not done at all. You can omit the
3348 fill value (the second argument) entirely by simply using two commas after the
3349 required alignment; this can be useful if you want the alignment to be filled
3350 with no-op instructions when appropriate.
3352 The way the required alignment is specified varies from system to system.
3353 For the a29k, hppa, m68k, m88k, w65, sparc, and Hitachi SH, and i386 using ELF
3355 the first expression is the
3356 alignment request in bytes. For example @samp{.align 8} advances
3357 the location counter until it is a multiple of 8. If the location counter
3358 is already a multiple of 8, no change is needed.
3360 For other systems, including the i386 using a.out format, and the arm and
3361 strongarm, it is the
3362 number of low-order zero bits the location counter must have after
3363 advancement. For example @samp{.align 3} advances the location
3364 counter until it a multiple of 8. If the location counter is already a
3365 multiple of 8, no change is needed.
3367 This inconsistency is due to the different behaviors of the various
3368 native assemblers for these systems which GAS must emulate.
3369 GAS also provides @code{.balign} and @code{.p2align} directives,
3370 described later, which have a consistent behavior across all
3371 architectures (but are specific to GAS).
3374 @section @code{.ascii "@var{string}"}@dots{}
3376 @cindex @code{ascii} directive
3377 @cindex string literals
3378 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3379 separated by commas. It assembles each string (with no automatic
3380 trailing zero byte) into consecutive addresses.
3383 @section @code{.asciz "@var{string}"}@dots{}
3385 @cindex @code{asciz} directive
3386 @cindex zero-terminated strings
3387 @cindex null-terminated strings
3388 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3389 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3392 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3394 @cindex padding the location counter given number of bytes
3395 @cindex @code{balign} directive
3396 Pad the location counter (in the current subsection) to a particular
3397 storage boundary. The first expression (which must be absolute) is the
3398 alignment request in bytes. For example @samp{.balign 8} advances
3399 the location counter until it is a multiple of 8. If the location counter
3400 is already a multiple of 8, no change is needed.
3402 The second expression (also absolute) gives the fill value to be stored in the
3403 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3404 padding bytes are normally zero. However, on some systems, if the section is
3405 marked as containing code and the fill value is omitted, the space is filled
3406 with no-op instructions.
3408 The third expression is also absolute, and is also optional. If it is present,
3409 it is the maximum number of bytes that should be skipped by this alignment
3410 directive. If doing the alignment would require skipping more bytes than the
3411 specified maximum, then the alignment is not done at all. You can omit the
3412 fill value (the second argument) entirely by simply using two commas after the
3413 required alignment; this can be useful if you want the alignment to be filled
3414 with no-op instructions when appropriate.
3416 @cindex @code{balignw} directive
3417 @cindex @code{balignl} directive
3418 The @code{.balignw} and @code{.balignl} directives are variants of the
3419 @code{.balign} directive. The @code{.balignw} directive treats the fill
3420 pattern as a two byte word value. The @code{.balignl} directives treats the
3421 fill pattern as a four byte longword value. For example, @code{.balignw
3422 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3423 filled in with the value 0x368d (the exact placement of the bytes depends upon
3424 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3428 @section @code{.byte @var{expressions}}
3430 @cindex @code{byte} directive
3431 @cindex integers, one byte
3432 @code{.byte} expects zero or more expressions, separated by commas.
3433 Each expression is assembled into the next byte.
3436 @section @code{.comm @var{symbol} , @var{length} }
3438 @cindex @code{comm} directive
3439 @cindex symbol, common
3440 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
3441 common symbol in one object file may be merged with a defined or common symbol
3442 of the same name in another object file. If @code{@value{LD}} does not see a
3443 definition for the symbol--just one or more common symbols--then it will
3444 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
3445 absolute expression. If @code{@value{LD}} sees multiple common symbols with
3446 the same name, and they do not all have the same size, it will allocate space
3447 using the largest size.
3450 When using ELF, the @code{.comm} directive takes an optional third argument.
3451 This is the desired alignment of the symbol, specified as a byte boundary (for
3452 example, an alignment of 16 means that the least significant 4 bits of the
3453 address should be zero). The alignment must be an absolute expression, and it
3454 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
3455 for the common symbol, it will use the alignment when placing the symbol. If
3456 no alignment is specified, @code{@value{AS}} will set the alignment to the
3457 largest power of two less than or equal to the size of the symbol, up to a
3462 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
3463 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
3467 @section @code{.data @var{subsection}}
3469 @cindex @code{data} directive
3470 @code{.data} tells @code{@value{AS}} to assemble the following statements onto the
3471 end of the data subsection numbered @var{subsection} (which is an
3472 absolute expression). If @var{subsection} is omitted, it defaults
3477 @section @code{.def @var{name}}
3479 @cindex @code{def} directive
3480 @cindex COFF symbols, debugging
3481 @cindex debugging COFF symbols
3482 Begin defining debugging information for a symbol @var{name}; the
3483 definition extends until the @code{.endef} directive is encountered.
3486 This directive is only observed when @code{@value{AS}} is configured for COFF
3487 format output; when producing @code{b.out}, @samp{.def} is recognized,
3494 @section @code{.desc @var{symbol}, @var{abs-expression}}
3496 @cindex @code{desc} directive
3497 @cindex COFF symbol descriptor
3498 @cindex symbol descriptor, COFF
3499 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
3500 to the low 16 bits of an absolute expression.
3503 The @samp{.desc} directive is not available when @code{@value{AS}} is
3504 configured for COFF output; it is only for @code{a.out} or @code{b.out}
3505 object format. For the sake of compatibility, @code{@value{AS}} accepts
3506 it, but produces no output, when configured for COFF.
3512 @section @code{.dim}
3514 @cindex @code{dim} directive
3515 @cindex COFF auxiliary symbol information
3516 @cindex auxiliary symbol information, COFF
3517 This directive is generated by compilers to include auxiliary debugging
3518 information in the symbol table. It is only permitted inside
3519 @code{.def}/@code{.endef} pairs.
3522 @samp{.dim} is only meaningful when generating COFF format output; when
3523 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
3529 @section @code{.double @var{flonums}}
3531 @cindex @code{double} directive
3532 @cindex floating point numbers (double)
3533 @code{.double} expects zero or more flonums, separated by commas. It
3534 assembles floating point numbers.
3536 The exact kind of floating point numbers emitted depends on how
3537 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
3541 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
3542 in @sc{ieee} format.
3547 @section @code{.eject}
3549 @cindex @code{eject} directive
3550 @cindex new page, in listings
3551 @cindex page, in listings
3552 @cindex listing control: new page
3553 Force a page break at this point, when generating assembly listings.
3556 @section @code{.else}
3558 @cindex @code{else} directive
3559 @code{.else} is part of the @code{@value{AS}} support for conditional
3560 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
3561 of code to be assembled if the condition for the preceding @code{.if}
3565 @section @code{.elseif}
3567 @cindex @code{elseif} directive
3568 @code{.elseif} is part of the @code{@value{AS}} support for conditional
3569 assembly; @pxref{If,,@code{.if}}. It is shorthand for beginning a new
3570 @code{.if} block that would otherwise fill the entire @code{.else} section.
3573 @section @code{.end}
3575 @cindex @code{end} directive
3576 @code{.end} marks the end of the assembly file. @code{@value{AS}} does not
3577 process anything in the file past the @code{.end} directive.
3581 @section @code{.endef}
3583 @cindex @code{endef} directive
3584 This directive flags the end of a symbol definition begun with
3588 @samp{.endef} is only meaningful when generating COFF format output; if
3589 @code{@value{AS}} is configured to generate @code{b.out}, it accepts this
3590 directive but ignores it.
3595 @section @code{.endfunc}
3596 @cindex @code{endfunc} directive
3597 @code{.endfunc} marks the end of a function specified with @code{.func}.
3600 @section @code{.endif}
3602 @cindex @code{endif} directive
3603 @code{.endif} is part of the @code{@value{AS}} support for conditional assembly;
3604 it marks the end of a block of code that is only assembled
3605 conditionally. @xref{If,,@code{.if}}.
3608 @section @code{.equ @var{symbol}, @var{expression}}
3610 @cindex @code{equ} directive
3611 @cindex assigning values to symbols
3612 @cindex symbols, assigning values to
3613 This directive sets the value of @var{symbol} to @var{expression}.
3614 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
3617 The syntax for @code{equ} on the HPPA is
3618 @samp{@var{symbol} .equ @var{expression}}.
3622 @section @code{.equiv @var{symbol}, @var{expression}}
3623 @cindex @code{equiv} directive
3624 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
3625 the assembler will signal an error if @var{symbol} is already defined.
3627 Except for the contents of the error message, this is roughly equivalent to
3636 @section @code{.err}
3637 @cindex @code{err} directive
3638 If @code{@value{AS}} assembles a @code{.err} directive, it will print an error
3639 message and, unless the @code{-Z} option was used, it will not generate an
3640 object file. This can be used to signal error an conditionally compiled code.
3643 @section @code{.exitm}
3644 Exit early from the current macro definition. @xref{Macro}.
3647 @section @code{.extern}
3649 @cindex @code{extern} directive
3650 @code{.extern} is accepted in the source program---for compatibility
3651 with other assemblers---but it is ignored. @code{@value{AS}} treats
3652 all undefined symbols as external.
3655 @section @code{.fail @var{expression}}
3657 @cindex @code{fail} directive
3658 Generates an error or a warning. If the value of the @var{expression} is 500
3659 or more, @code{@value{AS}} will print a warning message. If the value is less
3660 than 500, @code{@value{AS}} will print an error message. The message will
3661 include the value of @var{expression}. This can occasionally be useful inside
3662 complex nested macros or conditional assembly.
3664 @ifclear no-file-dir
3666 @section @code{.file @var{string}}
3668 @cindex @code{file} directive
3669 @cindex logical file name
3670 @cindex file name, logical
3671 @code{.file} tells @code{@value{AS}} that we are about to start a new logical
3672 file. @var{string} is the new file name. In general, the filename is
3673 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
3674 to specify an empty file name, you must give the quotes--@code{""}. This
3675 statement may go away in future: it is only recognized to be compatible with
3676 old @code{@value{AS}} programs.
3678 In some configurations of @code{@value{AS}}, @code{.file} has already been
3679 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
3684 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
3686 @cindex @code{fill} directive
3687 @cindex writing patterns in memory
3688 @cindex patterns, writing in memory
3689 @var{result}, @var{size} and @var{value} are absolute expressions.
3690 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
3691 may be zero or more. @var{Size} may be zero or more, but if it is
3692 more than 8, then it is deemed to have the value 8, compatible with
3693 other people's assemblers. The contents of each @var{repeat} bytes
3694 is taken from an 8-byte number. The highest order 4 bytes are
3695 zero. The lowest order 4 bytes are @var{value} rendered in the
3696 byte-order of an integer on the computer @code{@value{AS}} is assembling for.
3697 Each @var{size} bytes in a repetition is taken from the lowest order
3698 @var{size} bytes of this number. Again, this bizarre behavior is
3699 compatible with other people's assemblers.
3701 @var{size} and @var{value} are optional.
3702 If the second comma and @var{value} are absent, @var{value} is
3703 assumed zero. If the first comma and following tokens are absent,
3704 @var{size} is assumed to be 1.
3707 @section @code{.float @var{flonums}}
3709 @cindex floating point numbers (single)
3710 @cindex @code{float} directive
3711 This directive assembles zero or more flonums, separated by commas. It
3712 has the same effect as @code{.single}.
3714 The exact kind of floating point numbers emitted depends on how
3715 @code{@value{AS}} is configured.
3716 @xref{Machine Dependencies}.
3720 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
3721 in @sc{ieee} format.
3726 @section @code{.func @var{name}[,@var{label}]}
3727 @cindex @code{func} directive
3728 @code{.func} emits debugging information to denote function @var{name}, and
3729 is ignored unless the file is assembled with debugging enabled.
3730 Only @samp{--gstabs} is currently supported.
3731 @var{label} is the entry point of the function and if omitted @var{name}
3732 prepended with the @samp{leading char} is used.
3733 @samp{leading char} is usually @code{_} or nothing, depending on the target.
3734 All functions are currently defined to have @code{void} return type.
3735 The function must be terminated with @code{.endfunc}.
3738 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3740 @cindex @code{global} directive
3741 @cindex symbol, making visible to linker
3742 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
3743 @var{symbol} in your partial program, its value is made available to
3744 other partial programs that are linked with it. Otherwise,
3745 @var{symbol} takes its attributes from a symbol of the same name
3746 from another file linked into the same program.
3748 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
3749 compatibility with other assemblers.
3752 On the HPPA, @code{.global} is not always enough to make it accessible to other
3753 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
3754 @xref{HPPA Directives,, HPPA Assembler Directives}.
3758 @section @code{.hword @var{expressions}}
3760 @cindex @code{hword} directive
3761 @cindex integers, 16-bit
3762 @cindex numbers, 16-bit
3763 @cindex sixteen bit integers
3764 This expects zero or more @var{expressions}, and emits
3765 a 16 bit number for each.
3768 This directive is a synonym for @samp{.short}; depending on the target
3769 architecture, it may also be a synonym for @samp{.word}.
3773 This directive is a synonym for @samp{.short}.
3776 This directive is a synonym for both @samp{.short} and @samp{.word}.
3781 @section @code{.ident}
3783 @cindex @code{ident} directive
3784 This directive is used by some assemblers to place tags in object files.
3785 @code{@value{AS}} simply accepts the directive for source-file
3786 compatibility with such assemblers, but does not actually emit anything
3790 @section @code{.if @var{absolute expression}}
3792 @cindex conditional assembly
3793 @cindex @code{if} directive
3794 @code{.if} marks the beginning of a section of code which is only
3795 considered part of the source program being assembled if the argument
3796 (which must be an @var{absolute expression}) is non-zero. The end of
3797 the conditional section of code must be marked by @code{.endif}
3798 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
3799 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
3800 If you have several conditions to check, @code{.elseif} may be used to avoid
3801 nesting blocks if/else within each subsequent @code{.else} block.
3803 The following variants of @code{.if} are also supported:
3805 @cindex @code{ifdef} directive
3806 @item .ifdef @var{symbol}
3807 Assembles the following section of code if the specified @var{symbol}
3810 @cindex @code{ifc} directive
3811 @item .ifc @var{string1},@var{string2}
3812 Assembles the following section of code if the two strings are the same. The
3813 strings may be optionally quoted with single quotes. If they are not quoted,
3814 the first string stops at the first comma, and the second string stops at the
3815 end of the line. Strings which contain whitespace should be quoted. The
3816 string comparison is case sensitive.
3818 @cindex @code{ifeq} directive
3819 @item .ifeq @var{absolute expression}
3820 Assembles the following section of code if the argument is zero.
3822 @cindex @code{ifeqs} directive
3823 @item .ifeqs @var{string1},@var{string2}
3824 Another form of @code{.ifc}. The strings must be quoted using double quotes.
3826 @cindex @code{ifge} directive
3827 @item .ifge @var{absolute expression}
3828 Assembles the following section of code if the argument is greater than or
3831 @cindex @code{ifgt} directive
3832 @item .ifgt @var{absolute expression}
3833 Assembles the following section of code if the argument is greater than zero.
3835 @cindex @code{ifle} directive
3836 @item .ifle @var{absolute expression}
3837 Assembles the following section of code if the argument is less than or equal
3840 @cindex @code{iflt} directive
3841 @item .iflt @var{absolute expression}
3842 Assembles the following section of code if the argument is less than zero.
3844 @cindex @code{ifnc} directive
3845 @item .ifnc @var{string1},@var{string2}.
3846 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
3847 following section of code if the two strings are not the same.
3849 @cindex @code{ifndef} directive
3850 @cindex @code{ifnotdef} directive
3851 @item .ifndef @var{symbol}
3852 @itemx .ifnotdef @var{symbol}
3853 Assembles the following section of code if the specified @var{symbol}
3854 has not been defined. Both spelling variants are equivalent.
3856 @cindex @code{ifne} directive
3857 @item .ifne @var{absolute expression}
3858 Assembles the following section of code if the argument is not equal to zero
3859 (in other words, this is equivalent to @code{.if}).
3861 @cindex @code{ifnes} directive
3862 @item .ifnes @var{string1},@var{string2}
3863 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
3864 following section of code if the two strings are not the same.
3868 @section @code{.include "@var{file}"}
3870 @cindex @code{include} directive
3871 @cindex supporting files, including
3872 @cindex files, including
3873 This directive provides a way to include supporting files at specified
3874 points in your source program. The code from @var{file} is assembled as
3875 if it followed the point of the @code{.include}; when the end of the
3876 included file is reached, assembly of the original file continues. You
3877 can control the search paths used with the @samp{-I} command-line option
3878 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
3882 @section @code{.int @var{expressions}}
3884 @cindex @code{int} directive
3885 @cindex integers, 32-bit
3886 Expect zero or more @var{expressions}, of any section, separated by commas.
3887 For each expression, emit a number that, at run time, is the value of that
3888 expression. The byte order and bit size of the number depends on what kind
3889 of target the assembly is for.
3893 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
3894 integers. On the H8/300H and the Hitachi SH, however, @code{.int} emits
3900 @section @code{.irp @var{symbol},@var{values}}@dots{}
3902 @cindex @code{irp} directive
3903 Evaluate a sequence of statements assigning different values to @var{symbol}.
3904 The sequence of statements starts at the @code{.irp} directive, and is
3905 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
3906 set to @var{value}, and the sequence of statements is assembled. If no
3907 @var{value} is listed, the sequence of statements is assembled once, with
3908 @var{symbol} set to the null string. To refer to @var{symbol} within the
3909 sequence of statements, use @var{\symbol}.
3911 For example, assembling
3919 is equivalent to assembling
3928 @section @code{.irpc @var{symbol},@var{values}}@dots{}
3930 @cindex @code{irpc} directive
3931 Evaluate a sequence of statements assigning different values to @var{symbol}.
3932 The sequence of statements starts at the @code{.irpc} directive, and is
3933 terminated by an @code{.endr} directive. For each character in @var{value},
3934 @var{symbol} is set to the character, and the sequence of statements is
3935 assembled. If no @var{value} is listed, the sequence of statements is
3936 assembled once, with @var{symbol} set to the null string. To refer to
3937 @var{symbol} within the sequence of statements, use @var{\symbol}.
3939 For example, assembling
3947 is equivalent to assembling
3956 @section @code{.lcomm @var{symbol} , @var{length}}
3958 @cindex @code{lcomm} directive
3959 @cindex local common symbols
3960 @cindex symbols, local common
3961 Reserve @var{length} (an absolute expression) bytes for a local common
3962 denoted by @var{symbol}. The section and value of @var{symbol} are
3963 those of the new local common. The addresses are allocated in the bss
3964 section, so that at run-time the bytes start off zeroed. @var{Symbol}
3965 is not declared global (@pxref{Global,,@code{.global}}), so is normally
3966 not visible to @code{@value{LD}}.
3969 Some targets permit a third argument to be used with @code{.lcomm}. This
3970 argument specifies the desired alignment of the symbol in the bss section.
3974 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
3975 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
3979 @section @code{.lflags}
3981 @cindex @code{lflags} directive (ignored)
3982 @code{@value{AS}} accepts this directive, for compatibility with other
3983 assemblers, but ignores it.
3985 @ifclear no-line-dir
3987 @section @code{.line @var{line-number}}
3989 @cindex @code{line} directive
3993 @section @code{.ln @var{line-number}}
3995 @cindex @code{ln} directive
3997 @cindex logical line number
3999 Change the logical line number. @var{line-number} must be an absolute
4000 expression. The next line has that logical line number. Therefore any other
4001 statements on the current line (after a statement separator character) are
4002 reported as on logical line number @var{line-number} @minus{} 1. One day
4003 @code{@value{AS}} will no longer support this directive: it is recognized only
4004 for compatibility with existing assembler programs.
4008 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
4009 not available; use the synonym @code{.ln} in that context.
4014 @ifclear no-line-dir
4015 Even though this is a directive associated with the @code{a.out} or
4016 @code{b.out} object-code formats, @code{@value{AS}} still recognizes it
4017 when producing COFF output, and treats @samp{.line} as though it
4018 were the COFF @samp{.ln} @emph{if} it is found outside a
4019 @code{.def}/@code{.endef} pair.
4021 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4022 used by compilers to generate auxiliary symbol information for
4027 @section @code{.linkonce [@var{type}]}
4029 @cindex @code{linkonce} directive
4030 @cindex common sections
4031 Mark the current section so that the linker only includes a single copy of it.
4032 This may be used to include the same section in several different object files,
4033 but ensure that the linker will only include it once in the final output file.
4034 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4035 Duplicate sections are detected based on the section name, so it should be
4038 This directive is only supported by a few object file formats; as of this
4039 writing, the only object file format which supports it is the Portable
4040 Executable format used on Windows NT.
4042 The @var{type} argument is optional. If specified, it must be one of the
4043 following strings. For example:
4047 Not all types may be supported on all object file formats.
4051 Silently discard duplicate sections. This is the default.
4054 Warn if there are duplicate sections, but still keep only one copy.
4057 Warn if any of the duplicates have different sizes.
4060 Warn if any of the duplicates do not have exactly the same contents.
4064 @section @code{.ln @var{line-number}}
4066 @cindex @code{ln} directive
4067 @ifclear no-line-dir
4068 @samp{.ln} is a synonym for @samp{.line}.
4071 Tell @code{@value{AS}} to change the logical line number. @var{line-number}
4072 must be an absolute expression. The next line has that logical
4073 line number, so any other statements on the current line (after a
4074 statement separator character @code{;}) are reported as on logical
4075 line number @var{line-number} @minus{} 1.
4078 This directive is accepted, but ignored, when @code{@value{AS}} is
4079 configured for @code{b.out}; its effect is only associated with COFF
4085 @section @code{.mri @var{val}}
4087 @cindex @code{mri} directive
4088 @cindex MRI mode, temporarily
4089 If @var{val} is non-zero, this tells @code{@value{AS}} to enter MRI mode. If
4090 @var{val} is zero, this tells @code{@value{AS}} to exit MRI mode. This change
4091 affects code assembled until the next @code{.mri} directive, or until the end
4092 of the file. @xref{M, MRI mode, MRI mode}.
4095 @section @code{.list}
4097 @cindex @code{list} directive
4098 @cindex listing control, turning on
4099 Control (in conjunction with the @code{.nolist} directive) whether or
4100 not assembly listings are generated. These two directives maintain an
4101 internal counter (which is zero initially). @code{.list} increments the
4102 counter, and @code{.nolist} decrements it. Assembly listings are
4103 generated whenever the counter is greater than zero.
4105 By default, listings are disabled. When you enable them (with the
4106 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4107 the initial value of the listing counter is one.
4110 @section @code{.long @var{expressions}}
4112 @cindex @code{long} directive
4113 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
4116 @c no one seems to know what this is for or whether this description is
4117 @c what it really ought to do
4119 @section @code{.lsym @var{symbol}, @var{expression}}
4121 @cindex @code{lsym} directive
4122 @cindex symbol, not referenced in assembly
4123 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
4124 the hash table, ensuring it cannot be referenced by name during the
4125 rest of the assembly. This sets the attributes of the symbol to be
4126 the same as the expression value:
4128 @var{other} = @var{descriptor} = 0
4129 @var{type} = @r{(section of @var{expression})}
4130 @var{value} = @var{expression}
4133 The new symbol is not flagged as external.
4137 @section @code{.macro}
4140 The commands @code{.macro} and @code{.endm} allow you to define macros that
4141 generate assembly output. For example, this definition specifies a macro
4142 @code{sum} that puts a sequence of numbers into memory:
4145 .macro sum from=0, to=5
4154 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
4166 @item .macro @var{macname}
4167 @itemx .macro @var{macname} @var{macargs} @dots{}
4168 @cindex @code{macro} directive
4169 Begin the definition of a macro called @var{macname}. If your macro
4170 definition requires arguments, specify their names after the macro name,
4171 separated by commas or spaces. You can supply a default value for any
4172 macro argument by following the name with @samp{=@var{deflt}}. For
4173 example, these are all valid @code{.macro} statements:
4177 Begin the definition of a macro called @code{comm}, which takes no
4180 @item .macro plus1 p, p1
4181 @itemx .macro plus1 p p1
4182 Either statement begins the definition of a macro called @code{plus1},
4183 which takes two arguments; within the macro definition, write
4184 @samp{\p} or @samp{\p1} to evaluate the arguments.
4186 @item .macro reserve_str p1=0 p2
4187 Begin the definition of a macro called @code{reserve_str}, with two
4188 arguments. The first argument has a default value, but not the second.
4189 After the definition is complete, you can call the macro either as
4190 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
4191 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
4192 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
4193 @samp{0}, and @samp{\p2} evaluating to @var{b}).
4196 When you call a macro, you can specify the argument values either by
4197 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
4198 @samp{sum to=17, from=9}.
4201 @cindex @code{endm} directive
4202 Mark the end of a macro definition.
4205 @cindex @code{exitm} directive
4206 Exit early from the current macro definition.
4208 @cindex number of macros executed
4209 @cindex macros, count executed
4211 @code{@value{AS}} maintains a counter of how many macros it has
4212 executed in this pseudo-variable; you can copy that number to your
4213 output with @samp{\@@}, but @emph{only within a macro definition}.
4216 @item LOCAL @var{name} [ , @dots{} ]
4217 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
4218 macro syntax'' with @samp{-a} or @samp{--alternate}.} @xref{Alternate,,
4219 Alternate macro syntax}.
4221 Generate a string replacement for each of the @var{name} arguments, and
4222 replace any instances of @var{name} in each macro expansion. The
4223 replacement string is unique in the assembly, and different for each
4224 separate macro expansion. @code{LOCAL} allows you to write macros that
4225 define symbols, without fear of conflict between separate macro expansions.
4230 @section @code{.nolist}
4232 @cindex @code{nolist} directive
4233 @cindex listing control, turning off
4234 Control (in conjunction with the @code{.list} directive) whether or
4235 not assembly listings are generated. These two directives maintain an
4236 internal counter (which is zero initially). @code{.list} increments the
4237 counter, and @code{.nolist} decrements it. Assembly listings are
4238 generated whenever the counter is greater than zero.
4241 @section @code{.octa @var{bignums}}
4243 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
4244 @cindex @code{octa} directive
4245 @cindex integer, 16-byte
4246 @cindex sixteen byte integer
4247 This directive expects zero or more bignums, separated by commas. For each
4248 bignum, it emits a 16-byte integer.
4250 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
4251 hence @emph{octa}-word for 16 bytes.
4254 @section @code{.org @var{new-lc} , @var{fill}}
4256 @cindex @code{org} directive
4257 @cindex location counter, advancing
4258 @cindex advancing location counter
4259 @cindex current address, advancing
4260 Advance the location counter of the current section to
4261 @var{new-lc}. @var{new-lc} is either an absolute expression or an
4262 expression with the same section as the current subsection. That is,
4263 you can't use @code{.org} to cross sections: if @var{new-lc} has the
4264 wrong section, the @code{.org} directive is ignored. To be compatible
4265 with former assemblers, if the section of @var{new-lc} is absolute,
4266 @code{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
4267 is the same as the current subsection.
4269 @code{.org} may only increase the location counter, or leave it
4270 unchanged; you cannot use @code{.org} to move the location counter
4273 @c double negative used below "not undefined" because this is a specific
4274 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
4275 @c section. doc@cygnus.com 18feb91
4276 Because @code{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
4277 may not be undefined. If you really detest this restriction we eagerly await
4278 a chance to share your improved assembler.
4280 Beware that the origin is relative to the start of the section, not
4281 to the start of the subsection. This is compatible with other
4282 people's assemblers.
4284 When the location counter (of the current subsection) is advanced, the
4285 intervening bytes are filled with @var{fill} which should be an
4286 absolute expression. If the comma and @var{fill} are omitted,
4287 @var{fill} defaults to zero.
4290 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4292 @cindex padding the location counter given a power of two
4293 @cindex @code{p2align} directive
4294 Pad the location counter (in the current subsection) to a particular
4295 storage boundary. The first expression (which must be absolute) is the
4296 number of low-order zero bits the location counter must have after
4297 advancement. For example @samp{.p2align 3} advances the location
4298 counter until it a multiple of 8. If the location counter is already a
4299 multiple of 8, no change is needed.
4301 The second expression (also absolute) gives the fill value to be stored in the
4302 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4303 padding bytes are normally zero. However, on some systems, if the section is
4304 marked as containing code and the fill value is omitted, the space is filled
4305 with no-op instructions.
4307 The third expression is also absolute, and is also optional. If it is present,
4308 it is the maximum number of bytes that should be skipped by this alignment
4309 directive. If doing the alignment would require skipping more bytes than the
4310 specified maximum, then the alignment is not done at all. You can omit the
4311 fill value (the second argument) entirely by simply using two commas after the
4312 required alignment; this can be useful if you want the alignment to be filled
4313 with no-op instructions when appropriate.
4315 @cindex @code{p2alignw} directive
4316 @cindex @code{p2alignl} directive
4317 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
4318 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
4319 pattern as a two byte word value. The @code{.p2alignl} directives treats the
4320 fill pattern as a four byte longword value. For example, @code{.p2alignw
4321 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4322 filled in with the value 0x368d (the exact placement of the bytes depends upon
4323 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4327 @section @code{.print @var{string}}
4329 @cindex @code{print} directive
4330 @code{@value{AS}} will print @var{string} on the standard output during
4331 assembly. You must put @var{string} in double quotes.
4334 @section @code{.psize @var{lines} , @var{columns}}
4336 @cindex @code{psize} directive
4337 @cindex listing control: paper size
4338 @cindex paper size, for listings
4339 Use this directive to declare the number of lines---and, optionally, the
4340 number of columns---to use for each page, when generating listings.
4342 If you do not use @code{.psize}, listings use a default line-count
4343 of 60. You may omit the comma and @var{columns} specification; the
4344 default width is 200 columns.
4346 @code{@value{AS}} generates formfeeds whenever the specified number of
4347 lines is exceeded (or whenever you explicitly request one, using
4350 If you specify @var{lines} as @code{0}, no formfeeds are generated save
4351 those explicitly specified with @code{.eject}.
4354 @section @code{.purgem @var{name}}
4356 @cindex @code{purgem} directive
4357 Undefine the macro @var{name}, so that later uses of the string will not be
4358 expanded. @xref{Macro}.
4361 @section @code{.quad @var{bignums}}
4363 @cindex @code{quad} directive
4364 @code{.quad} expects zero or more bignums, separated by commas. For
4365 each bignum, it emits
4367 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
4368 warning message; and just takes the lowest order 8 bytes of the bignum.
4369 @cindex eight-byte integer
4370 @cindex integer, 8-byte
4372 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
4373 hence @emph{quad}-word for 8 bytes.
4376 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
4377 warning message; and just takes the lowest order 16 bytes of the bignum.
4378 @cindex sixteen-byte integer
4379 @cindex integer, 16-byte
4383 @section @code{.rept @var{count}}
4385 @cindex @code{rept} directive
4386 Repeat the sequence of lines between the @code{.rept} directive and the next
4387 @code{.endr} directive @var{count} times.
4389 For example, assembling
4397 is equivalent to assembling
4406 @section @code{.sbttl "@var{subheading}"}
4408 @cindex @code{sbttl} directive
4409 @cindex subtitles for listings
4410 @cindex listing control: subtitle
4411 Use @var{subheading} as the title (third line, immediately after the
4412 title line) when generating assembly listings.
4414 This directive affects subsequent pages, as well as the current page if
4415 it appears within ten lines of the top of a page.
4419 @section @code{.scl @var{class}}
4421 @cindex @code{scl} directive
4422 @cindex symbol storage class (COFF)
4423 @cindex COFF symbol storage class
4424 Set the storage-class value for a symbol. This directive may only be
4425 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
4426 whether a symbol is static or external, or it may record further
4427 symbolic debugging information.
4430 The @samp{.scl} directive is primarily associated with COFF output; when
4431 configured to generate @code{b.out} output format, @code{@value{AS}}
4432 accepts this directive but ignores it.
4437 @section @code{.section @var{name}}
4439 @cindex @code{section} directive
4440 @cindex named section
4441 Use the @code{.section} directive to assemble the following code into a section
4444 This directive is only supported for targets that actually support arbitrarily
4445 named sections; on @code{a.out} targets, for example, it is not accepted, even
4446 with a standard @code{a.out} section name.
4449 For COFF targets, the @code{.section} directive is used in one of the following
4452 .section @var{name}[, "@var{flags}"]
4453 .section @var{name}[, @var{subsegment}]
4456 If the optional argument is quoted, it is taken as flags to use for the
4457 section. Each flag is a single character. The following flags are recognized:
4460 bss section (uninitialized data)
4462 section is not loaded
4472 shared section (meaningful for PE targets)
4475 If no flags are specified, the default flags depend upon the section name. If
4476 the section name is not recognized, the default will be for the section to be
4477 loaded and writable.
4479 If the optional argument to the @code{.section} directive is not quoted, it is
4480 taken as a subsegment number (@pxref{Sub-Sections}).
4484 For ELF targets, the @code{.section} directive is used like this:
4486 .section @var{name}[, "@var{flags}"[, @@@var{type}]]
4488 The optional @var{flags} argument is a quoted string which may contain any
4489 combintion of the following characters:
4492 section is allocatable
4496 section is executable
4499 The optional @var{type} argument may contain one of the following constants:
4502 section contains data
4504 section does not contain data (i.e., section only occupies space)
4507 If no flags are specified, the default flags depend upon the section name. If
4508 the section name is not recognized, the default will be for the section to have
4509 none of the above flags: it will not be allocated in memory, nor writable, nor
4510 executable. The section will contain data.
4512 For ELF targets, the assembler supports another type of @code{.section}
4513 directive for compatibility with the Solaris assembler:
4515 .section "@var{name}"[, @var{flags}...]
4517 Note that the section name is quoted. There may be a sequence of comma
4521 section is allocatable
4525 section is executable
4530 @section @code{.set @var{symbol}, @var{expression}}
4532 @cindex @code{set} directive
4533 @cindex symbol value, setting
4534 Set the value of @var{symbol} to @var{expression}. This
4535 changes @var{symbol}'s value and type to conform to
4536 @var{expression}. If @var{symbol} was flagged as external, it remains
4537 flagged (@pxref{Symbol Attributes}).
4539 You may @code{.set} a symbol many times in the same assembly.
4541 If you @code{.set} a global symbol, the value stored in the object
4542 file is the last value stored into it.
4545 The syntax for @code{set} on the HPPA is
4546 @samp{@var{symbol} .set @var{expression}}.
4550 @section @code{.short @var{expressions}}
4552 @cindex @code{short} directive
4554 @code{.short} is normally the same as @samp{.word}.
4555 @xref{Word,,@code{.word}}.
4557 In some configurations, however, @code{.short} and @code{.word} generate
4558 numbers of different lengths; @pxref{Machine Dependencies}.
4562 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
4565 This expects zero or more @var{expressions}, and emits
4566 a 16 bit number for each.
4571 @section @code{.single @var{flonums}}
4573 @cindex @code{single} directive
4574 @cindex floating point numbers (single)
4575 This directive assembles zero or more flonums, separated by commas. It
4576 has the same effect as @code{.float}.
4578 The exact kind of floating point numbers emitted depends on how
4579 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
4583 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
4584 numbers in @sc{ieee} format.
4590 @section @code{.size}
4592 @cindex @code{size} directive
4593 This directive is generated by compilers to include auxiliary debugging
4594 information in the symbol table. It is only permitted inside
4595 @code{.def}/@code{.endef} pairs.
4598 @samp{.size} is only meaningful when generating COFF format output; when
4599 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4605 @section @code{.sleb128 @var{expressions}}
4607 @cindex @code{sleb128} directive
4608 @var{sleb128} stands for ``signed little endian base 128.'' This is a
4609 compact, variable length representation of numbers used by the DWARF
4610 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
4612 @ifclear no-space-dir
4614 @section @code{.skip @var{size} , @var{fill}}
4616 @cindex @code{skip} directive
4617 @cindex filling memory
4618 This directive emits @var{size} bytes, each of value @var{fill}. Both
4619 @var{size} and @var{fill} are absolute expressions. If the comma and
4620 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
4624 @section @code{.space @var{size} , @var{fill}}
4626 @cindex @code{space} directive
4627 @cindex filling memory
4628 This directive emits @var{size} bytes, each of value @var{fill}. Both
4629 @var{size} and @var{fill} are absolute expressions. If the comma
4630 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
4635 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
4636 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
4637 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
4638 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
4647 @section @code{.space}
4648 @cindex @code{space} directive
4650 On the AMD 29K, this directive is ignored; it is accepted for
4651 compatibility with other AMD 29K assemblers.
4654 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
4655 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
4661 @section @code{.stabd, .stabn, .stabs}
4663 @cindex symbolic debuggers, information for
4664 @cindex @code{stab@var{x}} directives
4665 There are three directives that begin @samp{.stab}.
4666 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
4667 The symbols are not entered in the @code{@value{AS}} hash table: they
4668 cannot be referenced elsewhere in the source file.
4669 Up to five fields are required:
4673 This is the symbol's name. It may contain any character except
4674 @samp{\000}, so is more general than ordinary symbol names. Some
4675 debuggers used to code arbitrarily complex structures into symbol names
4679 An absolute expression. The symbol's type is set to the low 8 bits of
4680 this expression. Any bit pattern is permitted, but @code{@value{LD}}
4681 and debuggers choke on silly bit patterns.
4684 An absolute expression. The symbol's ``other'' attribute is set to the
4685 low 8 bits of this expression.
4688 An absolute expression. The symbol's descriptor is set to the low 16
4689 bits of this expression.
4692 An absolute expression which becomes the symbol's value.
4695 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
4696 or @code{.stabs} statement, the symbol has probably already been created;
4697 you get a half-formed symbol in your object file. This is
4698 compatible with earlier assemblers!
4701 @cindex @code{stabd} directive
4702 @item .stabd @var{type} , @var{other} , @var{desc}
4704 The ``name'' of the symbol generated is not even an empty string.
4705 It is a null pointer, for compatibility. Older assemblers used a
4706 null pointer so they didn't waste space in object files with empty
4709 The symbol's value is set to the location counter,
4710 relocatably. When your program is linked, the value of this symbol
4711 is the address of the location counter when the @code{.stabd} was
4714 @cindex @code{stabn} directive
4715 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
4716 The name of the symbol is set to the empty string @code{""}.
4718 @cindex @code{stabs} directive
4719 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
4720 All five fields are specified.
4726 @section @code{.string} "@var{str}"
4728 @cindex string, copying to object file
4729 @cindex @code{string} directive
4731 Copy the characters in @var{str} to the object file. You may specify more than
4732 one string to copy, separated by commas. Unless otherwise specified for a
4733 particular machine, the assembler marks the end of each string with a 0 byte.
4734 You can use any of the escape sequences described in @ref{Strings,,Strings}.
4737 @section @code{.struct @var{expression}}
4739 @cindex @code{struct} directive
4740 Switch to the absolute section, and set the section offset to @var{expression},
4741 which must be an absolute expression. You might use this as follows:
4750 This would define the symbol @code{field1} to have the value 0, the symbol
4751 @code{field2} to have the value 4, and the symbol @code{field3} to have the
4752 value 8. Assembly would be left in the absolute section, and you would need to
4753 use a @code{.section} directive of some sort to change to some other section
4754 before further assembly.
4758 @section @code{.symver}
4759 @cindex @code{symver} directive
4760 @cindex symbol versioning
4761 @cindex versions of symbols
4762 Use the @code{.symver} directive to bind symbols to specific version nodes
4763 within a source file. This is only supported on ELF platforms, and is
4764 typically used when assembling files to be linked into a shared library.
4765 There are cases where it may make sense to use this in objects to be bound
4766 into an application itself so as to override a versioned symbol from a
4769 For ELF targets, the @code{.symver} directive is used like this:
4771 .symver @var{name}, @var{name2@@nodename}
4773 In this case, the symbol @var{name} must exist and be defined within the file
4774 being assembled. The @code{.versym} directive effectively creates a symbol
4775 alias with the name @var{name2@@nodename}, and in fact the main reason that we
4776 just don't try and create a regular alias is that the @var{@@} character isn't
4777 permitted in symbol names. The @var{name2} part of the name is the actual name
4778 of the symbol by which it will be externally referenced. The name @var{name}
4779 itself is merely a name of convenience that is used so that it is possible to
4780 have definitions for multiple versions of a function within a single source
4781 file, and so that the compiler can unambiguously know which version of a
4782 function is being mentioned. The @var{nodename} portion of the alias should be
4783 the name of a node specified in the version script supplied to the linker when
4784 building a shared library. If you are attempting to override a versioned
4785 symbol from a shared library, then @var{nodename} should correspond to the
4786 nodename of the symbol you are trying to override.
4791 @section @code{.tag @var{structname}}
4793 @cindex COFF structure debugging
4794 @cindex structure debugging, COFF
4795 @cindex @code{tag} directive
4796 This directive is generated by compilers to include auxiliary debugging
4797 information in the symbol table. It is only permitted inside
4798 @code{.def}/@code{.endef} pairs. Tags are used to link structure
4799 definitions in the symbol table with instances of those structures.
4802 @samp{.tag} is only used when generating COFF format output; when
4803 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4809 @section @code{.text @var{subsection}}
4811 @cindex @code{text} directive
4812 Tells @code{@value{AS}} to assemble the following statements onto the end of
4813 the text subsection numbered @var{subsection}, which is an absolute
4814 expression. If @var{subsection} is omitted, subsection number zero
4818 @section @code{.title "@var{heading}"}
4820 @cindex @code{title} directive
4821 @cindex listing control: title line
4822 Use @var{heading} as the title (second line, immediately after the
4823 source file name and pagenumber) when generating assembly listings.
4825 This directive affects subsequent pages, as well as the current page if
4826 it appears within ten lines of the top of a page.
4830 @section @code{.type @var{int}}
4832 @cindex COFF symbol type
4833 @cindex symbol type, COFF
4834 @cindex @code{type} directive
4835 This directive, permitted only within @code{.def}/@code{.endef} pairs,
4836 records the integer @var{int} as the type attribute of a symbol table entry.
4839 @samp{.type} is associated only with COFF format output; when
4840 @code{@value{AS}} is configured for @code{b.out} output, it accepts this
4841 directive but ignores it.
4847 @section @code{.val @var{addr}}
4849 @cindex @code{val} directive
4850 @cindex COFF value attribute
4851 @cindex value attribute, COFF
4852 This directive, permitted only within @code{.def}/@code{.endef} pairs,
4853 records the address @var{addr} as the value attribute of a symbol table
4857 @samp{.val} is used only for COFF output; when @code{@value{AS}} is
4858 configured for @code{b.out}, it accepts this directive but ignores it.
4863 @section @code{.uleb128 @var{expressions}}
4865 @cindex @code{uleb128} directive
4866 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
4867 compact, variable length representation of numbers used by the DWARF
4868 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
4872 @section @code{.internal}, @code{.hidden}, @code{.protected}
4873 @cindex @code{internal} directive
4874 @cindex @code{hidden} directive
4875 @cindex @code{protected} directive
4876 @cindex symbol visibility
4878 These directives can be used to set the visibility of a specified symbol. By
4879 default a symbol's visibility is set by its binding (local, global or weak),
4880 but these directives can be used to override that.
4882 A visibility of @code{protected} means that any references to the symbol from
4883 within the component that defines the symbol must be resolved to the definition
4884 in that component, even if a definition in another component would normally
4887 A visibility of @code{hidden} means that the symbol is not visible to other
4888 components. Such a symbol is always considered to be protected as well.
4890 A visibility of @code{internal} is the same as a visibility of @code{hidden},
4891 except that some extra, processor specific processing must also be performed
4894 For ELF targets, the directives are used like this:
4897 .internal @var{name}
4899 .protected @var{name}
4905 @section @code{.word @var{expressions}}
4907 @cindex @code{word} directive
4908 This directive expects zero or more @var{expressions}, of any section,
4909 separated by commas.
4912 For each expression, @code{@value{AS}} emits a 32-bit number.
4915 For each expression, @code{@value{AS}} emits a 16-bit number.
4920 The size of the number emitted, and its byte order,
4921 depend on what target computer the assembly is for.
4924 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
4925 @c happen---32-bit addressability, period; no long/short jumps.
4926 @ifset DIFF-TBL-KLUGE
4927 @cindex difference tables altered
4928 @cindex altered difference tables
4930 @emph{Warning: Special Treatment to support Compilers}
4934 Machines with a 32-bit address space, but that do less than 32-bit
4935 addressing, require the following special treatment. If the machine of
4936 interest to you does 32-bit addressing (or doesn't require it;
4937 @pxref{Machine Dependencies}), you can ignore this issue.
4940 In order to assemble compiler output into something that works,
4941 @code{@value{AS}} occasionlly does strange things to @samp{.word} directives.
4942 Directives of the form @samp{.word sym1-sym2} are often emitted by
4943 compilers as part of jump tables. Therefore, when @code{@value{AS}} assembles a
4944 directive of the form @samp{.word sym1-sym2}, and the difference between
4945 @code{sym1} and @code{sym2} does not fit in 16 bits, @code{@value{AS}}
4946 creates a @dfn{secondary jump table}, immediately before the next label.
4947 This secondary jump table is preceded by a short-jump to the
4948 first byte after the secondary table. This short-jump prevents the flow
4949 of control from accidentally falling into the new table. Inside the
4950 table is a long-jump to @code{sym2}. The original @samp{.word}
4951 contains @code{sym1} minus the address of the long-jump to
4954 If there were several occurrences of @samp{.word sym1-sym2} before the
4955 secondary jump table, all of them are adjusted. If there was a
4956 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
4957 long-jump to @code{sym4} is included in the secondary jump table,
4958 and the @code{.word} directives are adjusted to contain @code{sym3}
4959 minus the address of the long-jump to @code{sym4}; and so on, for as many
4960 entries in the original jump table as necessary.
4963 @emph{This feature may be disabled by compiling @code{@value{AS}} with the
4964 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
4965 assembly language programmers.
4968 @c end DIFF-TBL-KLUGE
4971 @section Deprecated Directives
4973 @cindex deprecated directives
4974 @cindex obsolescent directives
4975 One day these directives won't work.
4976 They are included for compatibility with older assemblers.
4983 @node Machine Dependencies
4984 @chapter Machine Dependent Features
4986 @cindex machine dependencies
4987 The machine instruction sets are (almost by definition) different on
4988 each machine where @code{@value{AS}} runs. Floating point representations
4989 vary as well, and @code{@value{AS}} often supports a few additional
4990 directives or command-line options for compatibility with other
4991 assemblers on a particular platform. Finally, some versions of
4992 @code{@value{AS}} support special pseudo-instructions for branch
4995 This chapter discusses most of these differences, though it does not
4996 include details on any machine's instruction set. For details on that
4997 subject, see the hardware manufacturer's manual.
5001 * AMD29K-Dependent:: AMD 29K Dependent Features
5004 * ARC-Dependent:: ARC Dependent Features
5007 * ARM-Dependent:: ARM Dependent Features
5010 * D10V-Dependent:: D10V Dependent Features
5013 * D30V-Dependent:: D30V Dependent Features
5016 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
5019 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
5022 * HPPA-Dependent:: HPPA Dependent Features
5025 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
5028 * i386-Dependent:: Intel 80386 Dependent Features
5031 * i860-Dependent:: Intel 80860 Dependent Features
5034 * i960-Dependent:: Intel 80960 Dependent Features
5037 * M32R-Dependent:: M32R Dependent Features
5040 * M68K-Dependent:: M680x0 Dependent Features
5043 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
5046 * MIPS-Dependent:: MIPS Dependent Features
5049 * SH-Dependent:: Hitachi SH Dependent Features
5052 * PJ-Dependent:: picoJava Dependent Features
5055 * Sparc-Dependent:: SPARC Dependent Features
5058 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
5061 * V850-Dependent:: V850 Dependent Features
5064 * Z8000-Dependent:: Z8000 Dependent Features
5067 * Vax-Dependent:: VAX Dependent Features
5074 @c The following major nodes are *sections* in the GENERIC version, *chapters*
5075 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
5076 @c peculiarity: to preserve cross-references, there must be a node called
5077 @c "Machine Dependencies". Hence the conditional nodenames in each
5078 @c major node below. Node defaulting in makeinfo requires adjacency of
5079 @c node and sectioning commands; hence the repetition of @chapter BLAH
5080 @c in both conditional blocks.
5086 @chapter ARC Dependent Features
5089 @node Machine Dependencies
5090 @chapter ARC Dependent Features
5095 * ARC-Opts:: Options
5096 * ARC-Float:: Floating Point
5097 * ARC-Directives:: Sparc Machine Directives
5103 @cindex options for ARC
5105 @cindex architectures, ARC
5106 @cindex ARC architectures
5107 The ARC chip family includes several successive levels (or other
5108 variants) of chip, using the same core instruction set, but including
5109 a few additional instructions at each level.
5111 By default, @code{@value{AS}} assumes the core instruction set (ARC
5112 base). The @code{.cpu} pseudo-op is intended to be used to select
5116 @cindex @code{-mbig-endian} option (ARC)
5117 @cindex @code{-mlittle-endian} option (ARC)
5118 @cindex ARC big-endian output
5119 @cindex ARC little-endian output
5120 @cindex big-endian output, ARC
5121 @cindex little-endian output, ARC
5123 @itemx -mlittle-endian
5124 Any @sc{arc} configuration of @code{@value{AS}} can select big-endian or
5125 little-endian output at run time (unlike most other @sc{gnu} development
5126 tools, which must be configured for one or the other). Use
5127 @samp{-mbig-endian} to select big-endian output, and @samp{-mlittle-endian}
5132 @section Floating Point
5134 @cindex floating point, ARC (@sc{ieee})
5135 @cindex ARC floating point (@sc{ieee})
5136 The ARC cpu family currently does not have hardware floating point
5137 support. Software floating point support is provided by @code{GCC}
5138 and uses @sc{ieee} floating-point numbers.
5140 @node ARC-Directives
5141 @section ARC Machine Directives
5143 @cindex ARC machine directives
5144 @cindex machine directives, ARC
5145 The ARC version of @code{@value{AS}} supports the following additional
5150 @cindex @code{cpu} directive, SPARC
5151 This must be followed by the desired cpu.
5152 The ARC is intended to be customizable, @code{.cpu} is used to
5153 select the desired variant [though currently there are none].
5160 @include c-a29k.texi
5169 @node Machine Dependencies
5170 @chapter Machine Dependent Features
5172 The machine instruction sets are different on each Hitachi chip family,
5173 and there are also some syntax differences among the families. This
5174 chapter describes the specific @code{@value{AS}} features for each
5178 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
5179 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
5180 * SH-Dependent:: Hitachi SH Dependent Features
5187 @include c-d10v.texi
5191 @include c-d30v.texi
5195 @include c-h8300.texi
5199 @include c-h8500.texi
5203 @include c-hppa.texi
5207 @include c-i370.texi
5211 @include c-i386.texi
5215 @include c-i860.texi
5219 @include c-i960.texi
5223 @include c-m32r.texi
5227 @include c-m68k.texi
5231 @include c-m68hc11.texi
5235 @include c-mips.texi
5239 @include c-ns32k.texi
5251 @include c-sparc.texi
5255 @include c-tic54x.texi
5267 @include c-v850.texi
5271 @c reverse effect of @down at top of generic Machine-Dep chapter
5275 @node Reporting Bugs
5276 @chapter Reporting Bugs
5277 @cindex bugs in assembler
5278 @cindex reporting bugs in assembler
5280 Your bug reports play an essential role in making @code{@value{AS}} reliable.
5282 Reporting a bug may help you by bringing a solution to your problem, or it may
5283 not. But in any case the principal function of a bug report is to help the
5284 entire community by making the next version of @code{@value{AS}} work better.
5285 Bug reports are your contribution to the maintenance of @code{@value{AS}}.
5287 In order for a bug report to serve its purpose, you must include the
5288 information that enables us to fix the bug.
5291 * Bug Criteria:: Have you found a bug?
5292 * Bug Reporting:: How to report bugs
5296 @section Have you found a bug?
5297 @cindex bug criteria
5299 If you are not sure whether you have found a bug, here are some guidelines:
5302 @cindex fatal signal
5303 @cindex assembler crash
5304 @cindex crash of assembler
5306 If the assembler gets a fatal signal, for any input whatever, that is a
5307 @code{@value{AS}} bug. Reliable assemblers never crash.
5309 @cindex error on valid input
5311 If @code{@value{AS}} produces an error message for valid input, that is a bug.
5313 @cindex invalid input
5315 If @code{@value{AS}} does not produce an error message for invalid input, that
5316 is a bug. However, you should note that your idea of ``invalid input'' might
5317 be our idea of ``an extension'' or ``support for traditional practice''.
5320 If you are an experienced user of assemblers, your suggestions for improvement
5321 of @code{@value{AS}} are welcome in any case.
5325 @section How to report bugs
5327 @cindex assembler bugs, reporting
5329 A number of companies and individuals offer support for @sc{gnu} products. If
5330 you obtained @code{@value{AS}} from a support organization, we recommend you
5331 contact that organization first.
5333 You can find contact information for many support companies and
5334 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
5337 In any event, we also recommend that you send bug reports for @code{@value{AS}}
5338 to @samp{bug-gnu-utils@@gnu.org}.
5340 The fundamental principle of reporting bugs usefully is this:
5341 @strong{report all the facts}. If you are not sure whether to state a
5342 fact or leave it out, state it!
5344 Often people omit facts because they think they know what causes the problem
5345 and assume that some details do not matter. Thus, you might assume that the
5346 name of a symbol you use in an example does not matter. Well, probably it does
5347 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
5348 happens to fetch from the location where that name is stored in memory;
5349 perhaps, if the name were different, the contents of that location would fool
5350 the assembler into doing the right thing despite the bug. Play it safe and
5351 give a specific, complete example. That is the easiest thing for you to do,
5352 and the most helpful.
5354 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
5355 it is new to us. Therefore, always write your bug reports on the assumption
5356 that the bug has not been reported previously.
5358 Sometimes people give a few sketchy facts and ask, ``Does this ring a
5359 bell?'' Those bug reports are useless, and we urge everyone to
5360 @emph{refuse to respond to them} except to chide the sender to report
5363 To enable us to fix the bug, you should include all these things:
5367 The version of @code{@value{AS}}. @code{@value{AS}} announces it if you start
5368 it with the @samp{--version} argument.
5370 Without this, we will not know whether there is any point in looking for
5371 the bug in the current version of @code{@value{AS}}.
5374 Any patches you may have applied to the @code{@value{AS}} source.
5377 The type of machine you are using, and the operating system name and
5381 What compiler (and its version) was used to compile @code{@value{AS}}---e.g.
5385 The command arguments you gave the assembler to assemble your example and
5386 observe the bug. To guarantee you will not omit something important, list them
5387 all. A copy of the Makefile (or the output from make) is sufficient.
5389 If we were to try to guess the arguments, we would probably guess wrong
5390 and then we might not encounter the bug.
5393 A complete input file that will reproduce the bug. If the bug is observed when
5394 the assembler is invoked via a compiler, send the assembler source, not the
5395 high level language source. Most compilers will produce the assembler source
5396 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
5397 the options @samp{-v --save-temps}; this will save the assembler source in a
5398 file with an extension of @file{.s}, and also show you exactly how
5399 @code{@value{AS}} is being run.
5402 A description of what behavior you observe that you believe is
5403 incorrect. For example, ``It gets a fatal signal.''
5405 Of course, if the bug is that @code{@value{AS}} gets a fatal signal, then we
5406 will certainly notice it. But if the bug is incorrect output, we might not
5407 notice unless it is glaringly wrong. You might as well not give us a chance to
5410 Even if the problem you experience is a fatal signal, you should still say so
5411 explicitly. Suppose something strange is going on, such as, your copy of
5412 @code{@value{AS}} is out of synch, or you have encountered a bug in the C
5413 library on your system. (This has happened!) Your copy might crash and ours
5414 would not. If you told us to expect a crash, then when ours fails to crash, we
5415 would know that the bug was not happening for us. If you had not told us to
5416 expect a crash, then we would not be able to draw any conclusion from our
5420 If you wish to suggest changes to the @code{@value{AS}} source, send us context
5421 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
5422 option. Always send diffs from the old file to the new file. If you even
5423 discuss something in the @code{@value{AS}} source, refer to it by context, not
5426 The line numbers in our development sources will not match those in your
5427 sources. Your line numbers would convey no useful information to us.
5430 Here are some things that are not necessary:
5434 A description of the envelope of the bug.
5436 Often people who encounter a bug spend a lot of time investigating
5437 which changes to the input file will make the bug go away and which
5438 changes will not affect it.
5440 This is often time consuming and not very useful, because the way we
5441 will find the bug is by running a single example under the debugger
5442 with breakpoints, not by pure deduction from a series of examples.
5443 We recommend that you save your time for something else.
5445 Of course, if you can find a simpler example to report @emph{instead}
5446 of the original one, that is a convenience for us. Errors in the
5447 output will be easier to spot, running under the debugger will take
5448 less time, and so on.
5450 However, simplification is not vital; if you do not want to do this,
5451 report the bug anyway and send us the entire test case you used.
5454 A patch for the bug.
5456 A patch for the bug does help us if it is a good one. But do not omit
5457 the necessary information, such as the test case, on the assumption that
5458 a patch is all we need. We might see problems with your patch and decide
5459 to fix the problem another way, or we might not understand it at all.
5461 Sometimes with a program as complicated as @code{@value{AS}} it is very hard to
5462 construct an example that will make the program follow a certain path through
5463 the code. If you do not send us the example, we will not be able to construct
5464 one, so we will not be able to verify that the bug is fixed.
5466 And if we cannot understand what bug you are trying to fix, or why your
5467 patch should be an improvement, we will not install it. A test case will
5468 help us to understand.
5471 A guess about what the bug is or what it depends on.
5473 Such guesses are usually wrong. Even we cannot guess right about such
5474 things without first using the debugger to find the facts.
5477 @node Acknowledgements
5478 @chapter Acknowledgements
5480 If you have contributed to @code{@value{AS}} and your name isn't listed here,
5481 it is not meant as a slight. We just don't know about it. Send mail to the
5482 maintainer, and we'll correct the situation. Currently
5484 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
5486 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
5489 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
5490 information and the 68k series machines, most of the preprocessing pass, and
5491 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
5493 K. Richard Pixley maintained GAS for a while, adding various enhancements and
5494 many bug fixes, including merging support for several processors, breaking GAS
5495 up to handle multiple object file format back ends (including heavy rewrite,
5496 testing, an integration of the coff and b.out back ends), adding configuration
5497 including heavy testing and verification of cross assemblers and file splits
5498 and renaming, converted GAS to strictly ANSI C including full prototypes, added
5499 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
5500 port (including considerable amounts of reverse engineering), a SPARC opcode
5501 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
5502 assertions and made them work, much other reorganization, cleanup, and lint.
5504 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
5505 in format-specific I/O modules.
5507 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
5508 has done much work with it since.
5510 The Intel 80386 machine description was written by Eliot Dresselhaus.
5512 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
5514 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
5515 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
5517 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
5518 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
5519 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
5520 support a.out format.
5522 Support for the Zilog Z8k and Hitachi H8/300 and H8/500 processors (tc-z8k,
5523 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
5524 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
5525 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
5528 John Gilmore built the AMD 29000 support, added @code{.include} support, and
5529 simplified the configuration of which versions accept which directives. He
5530 updated the 68k machine description so that Motorola's opcodes always produced
5531 fixed-size instructions (e.g. @code{jsr}), while synthetic instructions
5532 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
5533 cross-compilation support, and one bug in relaxation that took a week and
5534 required the proverbial one-bit fix.
5536 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
5537 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
5538 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
5539 PowerPC assembler, and made a few other minor patches.
5541 Steve Chamberlain made @code{@value{AS}} able to generate listings.
5543 Hewlett-Packard contributed support for the HP9000/300.
5545 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
5546 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
5547 formats). This work was supported by both the Center for Software Science at
5548 the University of Utah and Cygnus Support.
5550 Support for ELF format files has been worked on by Mark Eichin of Cygnus
5551 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
5552 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
5553 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
5554 and some initial 64-bit support).
5556 Linas Vepstas added GAS support for the ESA/390 "IBM 370" architecture.
5558 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
5559 support for openVMS/Alpha.
5561 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
5564 Several engineers at Cygnus Support have also provided many small bug fixes and
5565 configuration enhancements.
5567 Many others have contributed large or small bugfixes and enhancements. If
5568 you have contributed significant work and are not mentioned on this list, and
5569 want to be, let us know. Some of the history has been lost; we are not
5570 intentionally leaving anyone out.
projects / external / binutils.git / blob