1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (c) 1991, 92, 93, 94, 95, 96, 1997 Free Software Foundation, Inc.
3 @c UPDATE!! On future updates--
4 @c (1) check for new machine-dep cmdline options in
5 @c md_parse_option definitions in config/tc-*.c
6 @c (2) for platform-specific directives, examine md_pseudo_op
8 @c (3) for object-format specific directives, examine obj_pseudo_op
10 @c (4) portable directives in potable[] in read.c
14 @c defaults, config file may override:
17 @include asconfig.texi
19 @c common OR combinations of conditions
39 @set abnormal-separator
43 @settitle Using @value{AS}
46 @settitle Using @value{AS} (@value{TARGET})
48 @setchapternewpage odd
53 @c WARE! Some of the machine-dependent sections contain tables of machine
54 @c instructions. Except in multi-column format, these tables look silly.
55 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
56 @c the multi-col format is faked within @example sections.
58 @c Again unfortunately, the natural size that fits on a page, for these tables,
59 @c is different depending on whether or not smallbook is turned on.
60 @c This matters, because of order: text flow switches columns at each page
63 @c The format faked in this source works reasonably well for smallbook,
64 @c not well for the default large-page format. This manual expects that if you
65 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
66 @c tables in question. You can turn on one without the other at your
67 @c discretion, of course.
70 @c the insn tables look just as silly in info files regardless of smallbook,
71 @c might as well show 'em anyways.
77 * As: (as). The GNU assembler.
86 This file documents the GNU Assembler "@value{AS}".
88 Copyright (C) 1991, 92, 93, 94, 95, 96, 1997 Free Software Foundation, Inc.
90 Permission is granted to make and distribute verbatim copies of
91 this manual provided the copyright notice and this permission notice
92 are preserved on all copies.
95 Permission is granted to process this file through Tex and print the
96 results, provided the printed document carries copying permission
97 notice identical to this one except for the removal of this paragraph
98 (this paragraph not being relevant to the printed manual).
101 Permission is granted to copy and distribute modified versions of this manual
102 under the conditions for verbatim copying, provided that the entire resulting
103 derived work is distributed under the terms of a permission notice identical to
106 Permission is granted to copy and distribute translations of this manual
107 into another language, under the above conditions for modified versions.
111 @title Using @value{AS}
112 @subtitle The @sc{gnu} Assembler
114 @subtitle for the @value{TARGET} family
117 @subtitle January 1994
120 The Free Software Foundation Inc. thanks The Nice Computer
121 Company of Australia for loaning Dean Elsner to write the
122 first (Vax) version of @code{as} for Project @sc{gnu}.
123 The proprietors, management and staff of TNCCA thank FSF for
124 distracting the boss while they got some work
127 @author Dean Elsner, Jay Fenlason & friends
131 \hfill {\it Using {\tt @value{AS}}}\par
132 \hfill Edited by Cygnus Support\par
134 %"boxit" macro for figures:
135 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
136 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
137 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
138 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
139 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
142 @vskip 0pt plus 1filll
143 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 1997 Free Software Foundation, Inc.
145 Permission is granted to make and distribute verbatim copies of
146 this manual provided the copyright notice and this permission notice
147 are preserved on all copies.
149 Permission is granted to copy and distribute modified versions of this manual
150 under the conditions for verbatim copying, provided that the entire resulting
151 derived work is distributed under the terms of a permission notice identical to
154 Permission is granted to copy and distribute translations of this manual
155 into another language, under the above conditions for modified versions.
160 @top Using @value{AS}
162 This file is a user guide to the @sc{gnu} assembler @code{@value{AS}}.
164 This version of the file describes @code{@value{AS}} configured to generate
165 code for @value{TARGET} architectures.
168 * Overview:: Overview
169 * Invoking:: Command-Line Options
171 * Sections:: Sections and Relocation
173 * Expressions:: Expressions
174 * Pseudo Ops:: Assembler Directives
175 * Machine Dependencies:: Machine Dependent Features
176 * Reporting Bugs:: Reporting Bugs
177 * Acknowledgements:: Who Did What
185 This manual is a user guide to the @sc{gnu} assembler @code{@value{AS}}.
187 This version of the manual describes @code{@value{AS}} configured to generate
188 code for @value{TARGET} architectures.
192 @cindex invocation summary
193 @cindex option summary
194 @cindex summary of options
195 Here is a brief summary of how to invoke @code{@value{AS}}. For details,
196 @pxref{Invoking,,Comand-Line Options}.
198 @c We don't use deffn and friends for the following because they seem
199 @c to be limited to one line for the header.
201 @value{AS} [ -a[cdhlns][=file] ] [ -D ] [ --defsym @var{sym}=@var{val} ]
202 [ -f ] [ --gstabs ] [ --help ] [ -I @var{dir} ] [ -J ] [ -K ] [ -L ]
203 [ -o @var{objfile} ] [ -R ] [ --statistics ] [ -v ] [ -version ]
204 [ --version ] [ -W ] [ -w ] [ -x ] [ -Z ]
206 @c am29k has no machine-dependent assembler options
209 [ -mbig-endian | -mlittle-endian ]
212 [ -m[arm]1 | -m[arm]2 | -m[arm]250 | -m[arm]3 | -m[arm]6 | -m[arm]7[t][[d]m[i]] ]
213 [ -m[arm]v2 | -m[arm]v2a | -m[arm]v3 | -m[arm]v3m | -m[arm]v4 | -m[arm]v4t ]
215 [ -mfpa10 | -mfpa11 | -mfpe-old | -mno-fpu ]
217 [ -mapcs-32 | -mapcs-26 ]
224 @c Hitachi family chips have no machine-dependent assembler options
227 @c HPPA has no machine-dependent assembler options (yet).
230 @c The order here is important. See c-sparc.texi.
231 [ -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
232 -Av8plus | -Av8plusa | -Av9 | -Av9a ]
233 [ -xarch=v8plus | -xarch=v8plusa ] [ -bump ] [ -32 | -64 ]
236 @c Z8000 has no machine-dependent assembler options
239 @c see md_parse_option in tc-i960.c
240 [ -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC ]
247 [ -l ] [ -m68000 | -m68010 | -m68020 | ... ]
250 [ -nocpp ] [ -EL ] [ -EB ] [ -G @var{num} ] [ -mcpu=@var{CPU} ]
251 [ -mips1 ] [ -mips2 ] [ -mips3 ] [ -m4650 ] [ -no-m4650 ]
252 [ --trap ] [ --break ]
253 [ --emulation=@var{name} ]
255 [ -- | @var{files} @dots{} ]
260 Turn on listings, in any of a variety of ways:
264 omit false conditionals
267 omit debugging directives
270 include high-level source
276 include macro expansions
279 omit forms processing
285 set the name of the listing file
288 You may combine these options; for example, use @samp{-aln} for assembly
289 listing without forms processing. The @samp{=file} option, if used, must be
290 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
293 Ignored. This option is accepted for script compatibility with calls to
296 @item --defsym @var{sym}=@var{value}
297 Define the symbol @var{sym} to be @var{value} before assembling the input file.
298 @var{value} must be an integer constant. As in C, a leading @samp{0x}
299 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
302 ``fast''---skip whitespace and comment preprocessing (assume source is
306 Generate stabs debugging information for each assembler line. This
307 may help debugging assembler code, if the debugger can handle it.
310 Print a summary of the command line options and exit.
313 Add directory @var{dir} to the search list for @code{.include} directives.
316 Don't warn about signed overflow.
319 @ifclear DIFF-TBL-KLUGE
320 This option is accepted but has no effect on the @value{TARGET} family.
322 @ifset DIFF-TBL-KLUGE
323 Issue warnings when difference tables altered for long displacements.
327 Keep (in the symbol table) local symbols, starting with @samp{L}.
329 @item -o @var{objfile}
330 Name the object-file output from @code{@value{AS}} @var{objfile}.
333 Fold the data section into the text section.
336 Print the maximum space (in bytes) and total time (in seconds) used by
341 Print the @code{as} version.
344 Print the @code{as} version and exit.
347 Suppress warning messages.
356 Generate an object file even after errors.
358 @item -- | @var{files} @dots{}
359 Standard input, or source files to assemble.
364 The following options are available when @value{AS} is configured for
369 @cindex ARC endianness
370 @cindex endianness, ARC
371 @cindex big endian output, ARC
373 Generate ``big endian'' format output.
375 @cindex little endian output, ARC
376 @item -mlittle-endian
377 Generate ``little endian'' format output.
383 The following options are available when @value{AS} is configured for the ARM
387 @item -m[arm]1 | -m[arm]2 | -m[arm]250 | -m[arm]3 | -m[arm]6 | -m[arm]7[t][[d]m] | -m[arm]v2 | -m[arm]v2a | -m[arm]v3 | -m[arm]v3m | -m[arm]v4 | -m[arm]v4t
388 Specify which variant of the ARM architecture is the target.
389 @item -mthumb | -mall
390 Enable or disable Thumb only instruction decoding.
391 @item -mfpa10 | -mfpa11 | -mfpe-old | -mno-fpu
392 Select which Floating Point architcture is the target.
393 @item -mapcs-32 | -mapcs-26
394 Select which procedure calling convention is in use.
396 Select either big-endian (-EB) or little-endian (-EL) output.
401 The following options are available when @value{AS} is configured for
404 @cindex D10V optimization
405 @cindex optimization, D10V
407 Optimize output by parallelizing instructions.
412 The following options are available when @value{AS} is configured for the
413 Intel 80960 processor.
416 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
417 Specify which variant of the 960 architecture is the target.
420 Add code to collect statistics about branches taken.
423 Do not alter compare-and-branch instructions for long displacements;
430 The following options are available when @value{AS} is configured for the
431 Mitsubishi M32R series.
436 Specify which processor in the M32R family is the target. The default
437 is normally the M32R, but this option changes it to the M32RX.
443 The following options are available when @value{AS} is configured for the
444 Motorola 68000 series.
449 Shorten references to undefined symbols, to one word instead of two.
451 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030 | -m68040 | -m68060
452 @itemx | -m68302 | -m68331 | -m68332 | -m68333 | -m68340 | -mcpu32 | -m5200
453 Specify what processor in the 68000 family is the target. The default
454 is normally the 68020, but this can be changed at configuration time.
456 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
457 The target machine does (or does not) have a floating-point coprocessor.
458 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
459 the basic 68000 is not compatible with the 68881, a combination of the
460 two can be specified, since it's possible to do emulation of the
461 coprocessor instructions with the main processor.
463 @item -m68851 | -mno-68851
464 The target machine does (or does not) have a memory-management
465 unit coprocessor. The default is to assume an MMU for 68020 and up.
471 The following options are available when @code{@value{AS}} is configured
472 for the SPARC architecture:
475 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
476 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
477 Explicitly select a variant of the SPARC architecture.
479 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
480 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
482 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
483 UltraSPARC extensions.
485 @item -xarch=v8plus | -xarch=v8plusa
486 For compatibility with the Solaris v9 assembler. These options are
487 equivalent to -Av8plus and -Av8plusa, respectively.
490 Warn when the assembler switches to another architecture.
495 The following options are available when @value{AS} is configured for
500 This option sets the largest size of an object that can be referenced
501 implicitly with the @code{gp} register. It is only accepted for targets that
502 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
504 @cindex MIPS endianness
505 @cindex endianness, MIPS
506 @cindex big endian output, MIPS
508 Generate ``big endian'' format output.
510 @cindex little endian output, MIPS
512 Generate ``little endian'' format output.
518 Generate code for a particular MIPS Instruction Set Architecture level.
519 @samp{-mips1} corresponds to the @sc{r2000} and @sc{r3000} processors,
520 @samp{-mips2} to the @sc{r6000} processor, and @samp{-mips3} to the @sc{r4000}
525 Generate code for the MIPS @sc{r4650} chip. This tells the assembler to accept
526 the @samp{mad} and @samp{madu} instruction, and to not schedule @samp{nop}
527 instructions around accesses to the @samp{HI} and @samp{LO} registers.
528 @samp{-no-m4650} turns off this option.
530 @item -mcpu=@var{CPU}
531 Generate code for a particular MIPS cpu. This has little effect on the
532 assembler, but it is passed by @code{@value{GCC}}.
535 @item --emulation=@var{name}
536 This option causes @code{@value{AS}} to emulate @code{@value{AS}} configured
537 for some other target, in all respects, including output format (choosing
538 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
539 debugging information or store symbol table information, and default
540 endianness. The available configuration names are: @samp{mipsecoff},
541 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
542 @samp{mipsbelf}. The first two do not alter the default endianness from that
543 of the primary target for which the assembler was configured; the others change
544 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
545 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
546 selection in any case.
548 This option is currently supported only when the primary target
549 @code{@value{AS}} is configured for is a MIPS ELF or ECOFF target.
550 Furthermore, the primary target or others specified with
551 @samp{--enable-targets=@dots{}} at configuration time must include support for
552 the other format, if both are to be available. For example, the Irix 5
553 configuration includes support for both.
555 Eventually, this option will support more configurations, with more
556 fine-grained control over the assembler's behavior, and will be supported for
560 @code{@value{AS}} ignores this option. It is accepted for compatibility with
568 Control how to deal with multiplication overflow and division by zero.
569 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
570 (and only work for Instruction Set Architecture level 2 and higher);
571 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
577 * Manual:: Structure of this Manual
578 * GNU Assembler:: The GNU Assembler
579 * Object Formats:: Object File Formats
580 * Command Line:: Command Line
581 * Input Files:: Input Files
582 * Object:: Output (Object) File
583 * Errors:: Error and Warning Messages
587 @section Structure of this Manual
589 @cindex manual, structure and purpose
590 This manual is intended to describe what you need to know to use
591 @sc{gnu} @code{@value{AS}}. We cover the syntax expected in source files, including
592 notation for symbols, constants, and expressions; the directives that
593 @code{@value{AS}} understands; and of course how to invoke @code{@value{AS}}.
596 We also cover special features in the @value{TARGET}
597 configuration of @code{@value{AS}}, including assembler directives.
600 This manual also describes some of the machine-dependent features of
601 various flavors of the assembler.
604 @cindex machine instructions (not covered)
605 On the other hand, this manual is @emph{not} intended as an introduction
606 to programming in assembly language---let alone programming in general!
607 In a similar vein, we make no attempt to introduce the machine
608 architecture; we do @emph{not} describe the instruction set, standard
609 mnemonics, registers or addressing modes that are standard to a
610 particular architecture.
612 You may want to consult the manufacturer's
613 machine architecture manual for this information.
617 For information on the H8/300 machine instruction set, see @cite{H8/300
618 Series Programming Manual} (Hitachi ADE--602--025). For the H8/300H,
619 see @cite{H8/300H Series Programming Manual} (Hitachi).
622 For information on the H8/500 machine instruction set, see @cite{H8/500
623 Series Programming Manual} (Hitachi M21T001).
626 For information on the Hitachi SH machine instruction set, see
627 @cite{SH-Microcomputer User's Manual} (Hitachi Micro Systems, Inc.).
630 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
634 @c I think this is premature---doc@cygnus.com, 17jan1991
636 Throughout this manual, we assume that you are running @dfn{GNU},
637 the portable operating system from the @dfn{Free Software
638 Foundation, Inc.}. This restricts our attention to certain kinds of
639 computer (in particular, the kinds of computers that @sc{gnu} can run on);
640 once this assumption is granted examples and definitions need less
643 @code{@value{AS}} is part of a team of programs that turn a high-level
644 human-readable series of instructions into a low-level
645 computer-readable series of instructions. Different versions of
646 @code{@value{AS}} are used for different kinds of computer.
649 @c There used to be a section "Terminology" here, which defined
650 @c "contents", "byte", "word", and "long". Defining "word" to any
651 @c particular size is confusing when the .word directive may generate 16
652 @c bits on one machine and 32 bits on another; in general, for the user
653 @c version of this manual, none of these terms seem essential to define.
654 @c They were used very little even in the former draft of the manual;
655 @c this draft makes an effort to avoid them (except in names of
659 @section The GNU Assembler
661 @sc{gnu} @code{as} is really a family of assemblers.
663 This manual describes @code{@value{AS}}, a member of that family which is
664 configured for the @value{TARGET} architectures.
666 If you use (or have used) the @sc{gnu} assembler on one architecture, you
667 should find a fairly similar environment when you use it on another
668 architecture. Each version has much in common with the others,
669 including object file formats, most assembler directives (often called
670 @dfn{pseudo-ops}) and assembler syntax.@refill
672 @cindex purpose of @sc{gnu} assembler
673 @code{@value{AS}} is primarily intended to assemble the output of the
674 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
675 @code{@value{LD}}. Nevertheless, we've tried to make @code{@value{AS}}
676 assemble correctly everything that other assemblers for the same
677 machine would assemble.
679 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
682 @c This remark should appear in generic version of manual; assumption
683 @c here is that generic version sets M680x0.
684 This doesn't mean @code{@value{AS}} always uses the same syntax as another
685 assembler for the same architecture; for example, we know of several
686 incompatible versions of 680x0 assembly language syntax.
689 Unlike older assemblers, @code{@value{AS}} is designed to assemble a source
690 program in one pass of the source file. This has a subtle impact on the
691 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
694 @section Object File Formats
696 @cindex object file format
697 The @sc{gnu} assembler can be configured to produce several alternative
698 object file formats. For the most part, this does not affect how you
699 write assembly language programs; but directives for debugging symbols
700 are typically different in different file formats. @xref{Symbol
701 Attributes,,Symbol Attributes}.
704 On the @value{TARGET}, @code{@value{AS}} is configured to produce
705 @value{OBJ-NAME} format object files.
707 @c The following should exhaust all configs that set MULTI-OBJ, ideally
709 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
710 @code{a.out} or COFF format object files.
713 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
714 @code{b.out} or COFF format object files.
717 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
718 SOM or ELF format object files.
723 @section Command Line
725 @cindex command line conventions
726 After the program name @code{@value{AS}}, the command line may contain
727 options and file names. Options may appear in any order, and may be
728 before, after, or between file names. The order of file names is
731 @cindex standard input, as input file
733 @file{--} (two hyphens) by itself names the standard input file
734 explicitly, as one of the files for @code{@value{AS}} to assemble.
736 @cindex options, command line
737 Except for @samp{--} any command line argument that begins with a
738 hyphen (@samp{-}) is an option. Each option changes the behavior of
739 @code{@value{AS}}. No option changes the way another option works. An
740 option is a @samp{-} followed by one or more letters; the case of
741 the letter is important. All options are optional.
743 Some options expect exactly one file name to follow them. The file
744 name may either immediately follow the option's letter (compatible
745 with older assemblers) or it may be the next command argument (@sc{gnu}
746 standard). These two command lines are equivalent:
749 @value{AS} -o my-object-file.o mumble.s
750 @value{AS} -omy-object-file.o mumble.s
757 @cindex source program
759 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
760 describe the program input to one run of @code{@value{AS}}. The program may
761 be in one or more files; how the source is partitioned into files
762 doesn't change the meaning of the source.
764 @c I added "con" prefix to "catenation" just to prove I can overcome my
765 @c APL training... doc@cygnus.com
766 The source program is a concatenation of the text in all the files, in the
769 Each time you run @code{@value{AS}} it assembles exactly one source
770 program. The source program is made up of one or more files.
771 (The standard input is also a file.)
773 You give @code{@value{AS}} a command line that has zero or more input file
774 names. The input files are read (from left file name to right). A
775 command line argument (in any position) that has no special meaning
776 is taken to be an input file name.
778 If you give @code{@value{AS}} no file names it attempts to read one input file
779 from the @code{@value{AS}} standard input, which is normally your terminal. You
780 may have to type @key{ctl-D} to tell @code{@value{AS}} there is no more program
783 Use @samp{--} if you need to explicitly name the standard input file
784 in your command line.
786 If the source is empty, @code{@value{AS}} produces a small, empty object
789 @subheading Filenames and Line-numbers
791 @cindex input file linenumbers
792 @cindex line numbers, in input files
793 There are two ways of locating a line in the input file (or files) and
794 either may be used in reporting error messages. One way refers to a line
795 number in a physical file; the other refers to a line number in a
796 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
798 @dfn{Physical files} are those files named in the command line given
799 to @code{@value{AS}}.
801 @dfn{Logical files} are simply names declared explicitly by assembler
802 directives; they bear no relation to physical files. Logical file names
803 help error messages reflect the original source file, when @code{@value{AS}}
804 source is itself synthesized from other files.
805 @xref{App-File,,@code{.app-file}}.
808 @section Output (Object) File
814 Every time you run @code{@value{AS}} it produces an output file, which is
815 your assembly language program translated into numbers. This file
816 is the object file. Its default name is
824 @code{b.out} when @code{@value{AS}} is configured for the Intel 80960.
826 You can give it another name by using the @code{-o} option. Conventionally,
827 object file names end with @file{.o}. The default name is used for historical
828 reasons: older assemblers were capable of assembling self-contained programs
829 directly into a runnable program. (For some formats, this isn't currently
830 possible, but it can be done for the @code{a.out} format.)
834 The object file is meant for input to the linker @code{@value{LD}}. It contains
835 assembled program code, information to help @code{@value{LD}} integrate
836 the assembled program into a runnable file, and (optionally) symbolic
837 information for the debugger.
839 @c link above to some info file(s) like the description of a.out.
840 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
843 @section Error and Warning Messages
845 @cindex error messsages
846 @cindex warning messages
847 @cindex messages from assembler
848 @code{@value{AS}} may write warnings and error messages to the standard error
849 file (usually your terminal). This should not happen when a compiler
850 runs @code{@value{AS}} automatically. Warnings report an assumption made so
851 that @code{@value{AS}} could keep assembling a flawed program; errors report a
852 grave problem that stops the assembly.
854 @cindex format of warning messages
855 Warning messages have the format
858 file_name:@b{NNN}:Warning Message Text
862 @cindex line numbers, in warnings/errors
863 (where @b{NNN} is a line number). If a logical file name has been given
864 (@pxref{App-File,,@code{.app-file}}) it is used for the filename,
865 otherwise the name of the current input file is used. If a logical line
868 (@pxref{Line,,@code{.line}})
872 (@pxref{Line,,@code{.line}})
875 (@pxref{Ln,,@code{.ln}})
878 then it is used to calculate the number printed,
879 otherwise the actual line in the current source file is printed. The
880 message text is intended to be self explanatory (in the grand Unix
883 @cindex format of error messages
884 Error messages have the format
886 file_name:@b{NNN}:FATAL:Error Message Text
888 The file name and line number are derived as for warning
889 messages. The actual message text may be rather less explanatory
890 because many of them aren't supposed to happen.
893 @chapter Command-Line Options
895 @cindex options, all versions of assembler
896 This chapter describes command-line options available in @emph{all}
897 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
899 to the @value{TARGET}.
902 to particular machine architectures.
905 If you are invoking @code{@value{AS}} via the @sc{gnu} C compiler (version 2), you
906 can use the @samp{-Wa} option to pass arguments through to the
907 assembler. The assembler arguments must be separated from each other
908 (and the @samp{-Wa}) by commas. For example:
911 gcc -c -g -O -Wa,-alh,-L file.c
915 emits a listing to standard output with high-level
918 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
919 command-line options are automatically passed to the assembler by the compiler.
920 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
921 precisely what options it passes to each compilation pass, including the
925 * a:: -a[cdhlns] enable listings
926 * D:: -D for compatibility
927 * f:: -f to work faster
928 * I:: -I for .include search path
929 @ifclear DIFF-TBL-KLUGE
930 * K:: -K for compatibility
932 @ifset DIFF-TBL-KLUGE
933 * K:: -K for difference tables
936 * L:: -L to retain local labels
937 * M:: -M or --mri to assemble in MRI compatibility mode
938 * MD:: --MD for dependency tracking
939 * o:: -o to name the object file
940 * R:: -R to join data and text sections
941 * statistics:: --statistics to see statistics about assembly
942 * v:: -v to announce version
943 * W:: -W to suppress warnings
944 * Z:: -Z to make object file even after errors
948 @section Enable Listings: @code{-a[cdhlns]}
957 @cindex listings, enabling
958 @cindex assembly listings, enabling
960 These options enable listing output from the assembler. By itself,
961 @samp{-a} requests high-level, assembly, and symbols listing.
962 You can use other letters to select specific options for the list:
963 @samp{-ah} requests a high-level language listing,
964 @samp{-al} requests an output-program assembly listing, and
965 @samp{-as} requests a symbol table listing.
966 High-level listings require that a compiler debugging option like
967 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
970 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
971 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
972 other conditional), or a true @code{.if} followed by an @code{.else}, will be
973 omitted from the listing.
975 Use the @samp{-ad} option to omit debugging directives from the
978 Once you have specified one of these options, you can further control
979 listing output and its appearance using the directives @code{.list},
980 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
982 The @samp{-an} option turns off all forms processing.
983 If you do not request listing output with one of the @samp{-a} options, the
984 listing-control directives have no effect.
986 The letters after @samp{-a} may be combined into one option,
987 @emph{e.g.}, @samp{-aln}.
993 This option has no effect whatsoever, but it is accepted to make it more
994 likely that scripts written for other assemblers also work with
998 @section Work Faster: @code{-f}
1001 @cindex trusted compiler
1002 @cindex faster processing (@code{-f})
1003 @samp{-f} should only be used when assembling programs written by a
1004 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1005 and comment preprocessing on
1006 the input file(s) before assembling them. @xref{Preprocessing,
1010 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1011 preprocessed (if they contain comments, for example), @code{@value{AS}} does
1016 @section @code{.include} search path: @code{-I} @var{path}
1018 @kindex -I @var{path}
1019 @cindex paths for @code{.include}
1020 @cindex search path for @code{.include}
1021 @cindex @code{include} directive search path
1022 Use this option to add a @var{path} to the list of directories
1023 @code{@value{AS}} searches for files specified in @code{.include}
1024 directives (@pxref{Include,,@code{.include}}). You may use @code{-I} as
1025 many times as necessary to include a variety of paths. The current
1026 working directory is always searched first; after that, @code{@value{AS}}
1027 searches any @samp{-I} directories in the same order as they were
1028 specified (left to right) on the command line.
1031 @section Difference Tables: @code{-K}
1034 @ifclear DIFF-TBL-KLUGE
1035 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1036 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1037 where it can be used to warn when the assembler alters the machine code
1038 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1039 family does not have the addressing limitations that sometimes lead to this
1040 alteration on other platforms.
1043 @ifset DIFF-TBL-KLUGE
1044 @cindex difference tables, warning
1045 @cindex warning for altered difference tables
1046 @code{@value{AS}} sometimes alters the code emitted for directives of the form
1047 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1048 You can use the @samp{-K} option if you want a warning issued when this
1053 @section Include Local Labels: @code{-L}
1056 @cindex local labels, retaining in output
1057 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1058 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1059 debugging, because they are intended for the use of programs (like
1060 compilers) that compose assembler programs, not for your notice.
1061 Normally both @code{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1062 normally debug with them.
1064 This option tells @code{@value{AS}} to retain those @samp{L@dots{}} symbols
1065 in the object file. Usually if you do this you also tell the linker
1066 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1068 By default, a local label is any label beginning with @samp{L}, but each
1069 target is allowed to redefine the local label prefix.
1071 On the HPPA local labels begin with @samp{L$}.
1074 @samp{;} for the ARM family;
1078 @section Assemble in MRI Compatibility Mode: @code{-M}
1081 @cindex MRI compatibility mode
1082 The @code{-M} or @code{--mri} option selects MRI compatibility mode. This
1083 changes the syntax and pseudo-op handling of @code{@value{AS}} to make it
1084 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1085 configured target) assembler from Microtec Research. The exact nature of the
1086 MRI syntax will not be documented here; see the MRI manuals for more
1087 information. Note in particular that the handling of macros and macro
1088 arguments is somewhat different. The purpose of this option is to permit
1089 assembling existing MRI assembler code using @code{@value{AS}}.
1091 The MRI compatibility is not complete. Certain operations of the MRI assembler
1092 depend upon its object file format, and can not be supported using other object
1093 file formats. Supporting these would require enhancing each object file format
1094 individually. These are:
1097 @item global symbols in common section
1099 The m68k MRI assembler supports common sections which are merged by the linker.
1100 Other object file formats do not support this. @code{@value{AS}} handles
1101 common sections by treating them as a single common symbol. It permits local
1102 symbols to be defined within a common section, but it can not support global
1103 symbols, since it has no way to describe them.
1105 @item complex relocations
1107 The MRI assemblers support relocations against a negated section address, and
1108 relocations which combine the start addresses of two or more sections. These
1109 are not support by other object file formats.
1111 @item @code{END} pseudo-op specifying start address
1113 The MRI @code{END} pseudo-op permits the specification of a start address.
1114 This is not supported by other object file formats. The start address may
1115 instead be specified using the @code{-e} option to the linker, or in a linker
1118 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1120 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1121 name to the output file. This is not supported by other object file formats.
1123 @item @code{ORG} pseudo-op
1125 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1126 address. This differs from the usual @code{@value{AS}} @code{.org} pseudo-op,
1127 which changes the location within the current section. Absolute sections are
1128 not supported by other object file formats. The address of a section may be
1129 assigned within a linker script.
1132 There are some other features of the MRI assembler which are not supported by
1133 @code{@value{AS}}, typically either because they are difficult or because they
1134 seem of little consequence. Some of these may be supported in future releases.
1138 @item EBCDIC strings
1140 EBCDIC strings are not supported.
1142 @item packed binary coded decimal
1144 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1145 and @code{DCB.P} pseudo-ops are not supported.
1147 @item @code{FEQU} pseudo-op
1149 The m68k @code{FEQU} pseudo-op is not supported.
1151 @item @code{NOOBJ} pseudo-op
1153 The m68k @code{NOOBJ} pseudo-op is not supported.
1155 @item @code{OPT} branch control options
1157 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1158 @code{BRL}, and @code{BRW}---are ignored. @code{@value{AS}} automatically
1159 relaxes all branches, whether forward or backward, to an appropriate size, so
1160 these options serve no purpose.
1162 @item @code{OPT} list control options
1164 The following m68k @code{OPT} list control options are ignored: @code{C},
1165 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1166 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1168 @item other @code{OPT} options
1170 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1171 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1173 @item @code{OPT} @code{D} option is default
1175 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1176 @code{OPT NOD} may be used to turn it off.
1178 @item @code{XREF} pseudo-op.
1180 The m68k @code{XREF} pseudo-op is ignored.
1182 @item @code{.debug} pseudo-op
1184 The i960 @code{.debug} pseudo-op is not supported.
1186 @item @code{.extended} pseudo-op
1188 The i960 @code{.extended} pseudo-op is not supported.
1190 @item @code{.list} pseudo-op.
1192 The various options of the i960 @code{.list} pseudo-op are not supported.
1194 @item @code{.optimize} pseudo-op
1196 The i960 @code{.optimize} pseudo-op is not supported.
1198 @item @code{.output} pseudo-op
1200 The i960 @code{.output} pseudo-op is not supported.
1202 @item @code{.setreal} pseudo-op
1204 The i960 @code{.setreal} pseudo-op is not supported.
1209 @section Dependency tracking: @code{--MD}
1212 @cindex dependency tracking
1215 @code{@value{AS}} can generate a dependency file for the file it creates. This
1216 file consists of a single rule suitable for @code{make} describing the
1217 dependencies of the main source file.
1219 The rule is written to the file named in its argument.
1221 This feature is used in the automatic updating of makefiles.
1224 @section Name the Object File: @code{-o}
1227 @cindex naming object file
1228 @cindex object file name
1229 There is always one object file output when you run @code{@value{AS}}. By
1230 default it has the name
1233 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1247 You use this option (which takes exactly one filename) to give the
1248 object file a different name.
1250 Whatever the object file is called, @code{@value{AS}} overwrites any
1251 existing file of the same name.
1254 @section Join Data and Text Sections: @code{-R}
1257 @cindex data and text sections, joining
1258 @cindex text and data sections, joining
1259 @cindex joining text and data sections
1260 @cindex merging text and data sections
1261 @code{-R} tells @code{@value{AS}} to write the object file as if all
1262 data-section data lives in the text section. This is only done at
1263 the very last moment: your binary data are the same, but data
1264 section parts are relocated differently. The data section part of
1265 your object file is zero bytes long because all its bytes are
1266 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1268 When you specify @code{-R} it would be possible to generate shorter
1269 address displacements (because we do not have to cross between text and
1270 data section). We refrain from doing this simply for compatibility with
1271 older versions of @code{@value{AS}}. In future, @code{-R} may work this way.
1274 When @code{@value{AS}} is configured for COFF output,
1275 this option is only useful if you use sections named @samp{.text} and
1280 @code{-R} is not supported for any of the HPPA targets. Using
1281 @code{-R} generates a warning from @code{@value{AS}}.
1285 @section Display Assembly Statistics: @code{--statistics}
1287 @kindex --statistics
1288 @cindex statistics, about assembly
1289 @cindex time, total for assembly
1290 @cindex space used, maximum for assembly
1291 Use @samp{--statistics} to display two statistics about the resources used by
1292 @code{@value{AS}}: the maximum amount of space allocated during the assembly
1293 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1297 @section Announce Version: @code{-v}
1301 @cindex assembler version
1302 @cindex version of assembler
1303 You can find out what version of as is running by including the
1304 option @samp{-v} (which you can also spell as @samp{-version}) on the
1308 @section Suppress Warnings: @code{-W}
1311 @cindex suppressing warnings
1312 @cindex warnings, suppressing
1313 @code{@value{AS}} should never give a warning or error message when
1314 assembling compiler output. But programs written by people often
1315 cause @code{@value{AS}} to give a warning that a particular assumption was
1316 made. All such warnings are directed to the standard error file.
1317 If you use this option, no warnings are issued. This option only
1318 affects the warning messages: it does not change any particular of how
1319 @code{@value{AS}} assembles your file. Errors, which stop the assembly, are
1323 @section Generate Object File in Spite of Errors: @code{-Z}
1324 @cindex object file, after errors
1325 @cindex errors, continuing after
1326 After an error message, @code{@value{AS}} normally produces no output. If for
1327 some reason you are interested in object file output even after
1328 @code{@value{AS}} gives an error message on your program, use the @samp{-Z}
1329 option. If there are any errors, @code{@value{AS}} continues anyways, and
1330 writes an object file after a final warning message of the form @samp{@var{n}
1331 errors, @var{m} warnings, generating bad object file.}
1336 @cindex machine-independent syntax
1337 @cindex syntax, machine-independent
1338 This chapter describes the machine-independent syntax allowed in a
1339 source file. @code{@value{AS}} syntax is similar to what many other
1340 assemblers use; it is inspired by the BSD 4.2
1345 assembler, except that @code{@value{AS}} does not assemble Vax bit-fields.
1349 * Preprocessing:: Preprocessing
1350 * Whitespace:: Whitespace
1351 * Comments:: Comments
1352 * Symbol Intro:: Symbols
1353 * Statements:: Statements
1354 * Constants:: Constants
1358 @section Preprocessing
1360 @cindex preprocessing
1361 The @code{@value{AS}} internal preprocessor:
1363 @cindex whitespace, removed by preprocessor
1365 adjusts and removes extra whitespace. It leaves one space or tab before
1366 the keywords on a line, and turns any other whitespace on the line into
1369 @cindex comments, removed by preprocessor
1371 removes all comments, replacing them with a single space, or an
1372 appropriate number of newlines.
1374 @cindex constants, converted by preprocessor
1376 converts character constants into the appropriate numeric values.
1379 It does not do macro processing, include file handling, or
1380 anything else you may get from your C compiler's preprocessor. You can
1381 do include file processing with the @code{.include} directive
1382 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
1383 to get other ``CPP'' style preprocessing, by giving the input file a
1384 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
1385 Output, gcc.info, Using GNU CC}.
1387 Excess whitespace, comments, and character constants
1388 cannot be used in the portions of the input text that are not
1391 @cindex turning preprocessing on and off
1392 @cindex preprocessing, turning on and off
1395 If the first line of an input file is @code{#NO_APP} or if you use the
1396 @samp{-f} option, whitespace and comments are not removed from the input file.
1397 Within an input file, you can ask for whitespace and comment removal in
1398 specific portions of the by putting a line that says @code{#APP} before the
1399 text that may contain whitespace or comments, and putting a line that says
1400 @code{#NO_APP} after this text. This feature is mainly intend to support
1401 @code{asm} statements in compilers whose output is otherwise free of comments
1408 @dfn{Whitespace} is one or more blanks or tabs, in any order.
1409 Whitespace is used to separate symbols, and to make programs neater for
1410 people to read. Unless within character constants
1411 (@pxref{Characters,,Character Constants}), any whitespace means the same
1412 as exactly one space.
1418 There are two ways of rendering comments to @code{@value{AS}}. In both
1419 cases the comment is equivalent to one space.
1421 Anything from @samp{/*} through the next @samp{*/} is a comment.
1422 This means you may not nest these comments.
1426 The only way to include a newline ('\n') in a comment
1427 is to use this sort of comment.
1430 /* This sort of comment does not nest. */
1433 @cindex line comment character
1434 Anything from the @dfn{line comment} character to the next newline
1435 is considered a comment and is ignored. The line comment character is
1437 @samp{;} for the AMD 29K family;
1440 @samp{;} on the ARC;
1443 @samp{;} for the H8/300 family;
1446 @samp{!} for the H8/500 family;
1449 @samp{;} for the HPPA;
1452 @samp{#} on the i960;
1455 @samp{!} for the Hitachi SH;
1458 @samp{!} on the SPARC;
1461 @samp{#} on the m32r;
1464 @samp{|} on the 680x0;
1467 @samp{#} on the Vax;
1470 @samp{!} for the Z8000;
1473 @samp{#} on the V850;
1475 see @ref{Machine Dependencies}. @refill
1476 @c FIXME What about i386, m88k, i860?
1479 On some machines there are two different line comment characters. One
1480 character only begins a comment if it is the first non-whitespace character on
1481 a line, while the other always begins a comment.
1485 The V850 assembler also supports a double dash as starting a comment that
1486 extends to the end of the line.
1492 @cindex lines starting with @code{#}
1493 @cindex logical line numbers
1494 To be compatible with past assemblers, lines that begin with @samp{#} have a
1495 special interpretation. Following the @samp{#} should be an absolute
1496 expression (@pxref{Expressions}): the logical line number of the @emph{next}
1497 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
1498 new logical file name. The rest of the line, if any, should be whitespace.
1500 If the first non-whitespace characters on the line are not numeric,
1501 the line is ignored. (Just like a comment.)
1504 # This is an ordinary comment.
1505 # 42-6 "new_file_name" # New logical file name
1506 # This is logical line # 36.
1508 This feature is deprecated, and may disappear from future versions
1509 of @code{@value{AS}}.
1514 @cindex characters used in symbols
1515 @ifclear SPECIAL-SYMS
1516 A @dfn{symbol} is one or more characters chosen from the set of all
1517 letters (both upper and lower case), digits and the three characters
1523 A @dfn{symbol} is one or more characters chosen from the set of all
1524 letters (both upper and lower case), digits and the three characters
1525 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
1531 On most machines, you can also use @code{$} in symbol names; exceptions
1532 are noted in @ref{Machine Dependencies}.
1534 No symbol may begin with a digit. Case is significant.
1535 There is no length limit: all characters are significant. Symbols are
1536 delimited by characters not in that set, or by the beginning of a file
1537 (since the source program must end with a newline, the end of a file is
1538 not a possible symbol delimiter). @xref{Symbols}.
1539 @cindex length of symbols
1544 @cindex statements, structure of
1545 @cindex line separator character
1546 @cindex statement separator character
1548 @ifclear abnormal-separator
1549 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
1550 semicolon (@samp{;}). The newline or semicolon is considered part of
1551 the preceding statement. Newlines and semicolons within character
1552 constants are an exception: they do not end statements.
1554 @ifset abnormal-separator
1556 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
1557 sign (@samp{@@}). The newline or at sign is considered part of the
1558 preceding statement. Newlines and at signs within character constants
1559 are an exception: they do not end statements.
1562 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
1563 point (@samp{!}). The newline or exclamation point is considered part of the
1564 preceding statement. Newlines and exclamation points within character
1565 constants are an exception: they do not end statements.
1568 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
1569 H8/300) a dollar sign (@samp{$}); or (for the
1572 (@samp{;}). The newline or separator character is considered part of
1573 the preceding statement. Newlines and separators within character
1574 constants are an exception: they do not end statements.
1579 A @dfn{statement} ends at a newline character (@samp{\n}) or line
1580 separator character. (The line separator is usually @samp{;}, unless
1581 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
1582 newline or separator character is considered part of the preceding
1583 statement. Newlines and separators within character constants are an
1584 exception: they do not end statements.
1587 @cindex newline, required at file end
1588 @cindex EOF, newline must precede
1589 It is an error to end any statement with end-of-file: the last
1590 character of any input file should be a newline.@refill
1592 @cindex continuing statements
1593 @cindex multi-line statements
1594 @cindex statement on multiple lines
1595 You may write a statement on more than one line if you put a
1596 backslash (@kbd{\}) immediately in front of any newlines within the
1597 statement. When @code{@value{AS}} reads a backslashed newline both
1598 characters are ignored. You can even put backslashed newlines in
1599 the middle of symbol names without changing the meaning of your
1602 An empty statement is allowed, and may include whitespace. It is ignored.
1604 @cindex instructions and directives
1605 @cindex directives and instructions
1606 @c "key symbol" is not used elsewhere in the document; seems pedantic to
1607 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
1609 A statement begins with zero or more labels, optionally followed by a
1610 key symbol which determines what kind of statement it is. The key
1611 symbol determines the syntax of the rest of the statement. If the
1612 symbol begins with a dot @samp{.} then the statement is an assembler
1613 directive: typically valid for any computer. If the symbol begins with
1614 a letter the statement is an assembly language @dfn{instruction}: it
1615 assembles into a machine language instruction.
1617 Different versions of @code{@value{AS}} for different computers
1618 recognize different instructions. In fact, the same symbol may
1619 represent a different instruction in a different computer's assembly
1623 @cindex @code{:} (label)
1624 @cindex label (@code{:})
1625 A label is a symbol immediately followed by a colon (@code{:}).
1626 Whitespace before a label or after a colon is permitted, but you may not
1627 have whitespace between a label's symbol and its colon. @xref{Labels}.
1630 For HPPA targets, labels need not be immediately followed by a colon, but
1631 the definition of a label must begin in column zero. This also implies that
1632 only one label may be defined on each line.
1636 label: .directive followed by something
1637 another_label: # This is an empty statement.
1638 instruction operand_1, operand_2, @dots{}
1645 A constant is a number, written so that its value is known by
1646 inspection, without knowing any context. Like this:
1649 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
1650 .ascii "Ring the bell\7" # A string constant.
1651 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
1652 .float 0f-314159265358979323846264338327\
1653 95028841971.693993751E-40 # - pi, a flonum.
1658 * Characters:: Character Constants
1659 * Numbers:: Number Constants
1663 @subsection Character Constants
1665 @cindex character constants
1666 @cindex constants, character
1667 There are two kinds of character constants. A @dfn{character} stands
1668 for one character in one byte and its value may be used in
1669 numeric expressions. String constants (properly called string
1670 @emph{literals}) are potentially many bytes and their values may not be
1671 used in arithmetic expressions.
1675 * Chars:: Characters
1679 @subsubsection Strings
1681 @cindex string constants
1682 @cindex constants, string
1683 A @dfn{string} is written between double-quotes. It may contain
1684 double-quotes or null characters. The way to get special characters
1685 into a string is to @dfn{escape} these characters: precede them with
1686 a backslash @samp{\} character. For example @samp{\\} represents
1687 one backslash: the first @code{\} is an escape which tells
1688 @code{@value{AS}} to interpret the second character literally as a backslash
1689 (which prevents @code{@value{AS}} from recognizing the second @code{\} as an
1690 escape character). The complete list of escapes follows.
1692 @cindex escape codes, character
1693 @cindex character escape codes
1696 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
1698 @cindex @code{\b} (backspace character)
1699 @cindex backspace (@code{\b})
1701 Mnemonic for backspace; for ASCII this is octal code 010.
1704 @c Mnemonic for EOText; for ASCII this is octal code 004.
1706 @cindex @code{\f} (formfeed character)
1707 @cindex formfeed (@code{\f})
1709 Mnemonic for FormFeed; for ASCII this is octal code 014.
1711 @cindex @code{\n} (newline character)
1712 @cindex newline (@code{\n})
1714 Mnemonic for newline; for ASCII this is octal code 012.
1717 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
1719 @cindex @code{\r} (carriage return character)
1720 @cindex carriage return (@code{\r})
1722 Mnemonic for carriage-Return; for ASCII this is octal code 015.
1725 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
1726 @c other assemblers.
1728 @cindex @code{\t} (tab)
1729 @cindex tab (@code{\t})
1731 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
1734 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
1735 @c @item \x @var{digit} @var{digit} @var{digit}
1736 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
1738 @cindex @code{\@var{ddd}} (octal character code)
1739 @cindex octal character code (@code{\@var{ddd}})
1740 @item \ @var{digit} @var{digit} @var{digit}
1741 An octal character code. The numeric code is 3 octal digits.
1742 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
1743 for example, @code{\008} has the value 010, and @code{\009} the value 011.
1745 @cindex @code{\@var{xd...}} (hex character code)
1746 @cindex hex character code (@code{\@var{xd...}})
1747 @item \@code{x} @var{hex-digits...}
1748 A hex character code. All trailing hex digits are combined. Either upper or
1749 lower case @code{x} works.
1751 @cindex @code{\\} (@samp{\} character)
1752 @cindex backslash (@code{\\})
1754 Represents one @samp{\} character.
1757 @c Represents one @samp{'} (accent acute) character.
1758 @c This is needed in single character literals
1759 @c (@xref{Characters,,Character Constants}.) to represent
1762 @cindex @code{\"} (doublequote character)
1763 @cindex doublequote (@code{\"})
1765 Represents one @samp{"} character. Needed in strings to represent
1766 this character, because an unescaped @samp{"} would end the string.
1768 @item \ @var{anything-else}
1769 Any other character when escaped by @kbd{\} gives a warning, but
1770 assembles as if the @samp{\} was not present. The idea is that if
1771 you used an escape sequence you clearly didn't want the literal
1772 interpretation of the following character. However @code{@value{AS}} has no
1773 other interpretation, so @code{@value{AS}} knows it is giving you the wrong
1774 code and warns you of the fact.
1777 Which characters are escapable, and what those escapes represent,
1778 varies widely among assemblers. The current set is what we think
1779 the BSD 4.2 assembler recognizes, and is a subset of what most C
1780 compilers recognize. If you are in doubt, do not use an escape
1784 @subsubsection Characters
1786 @cindex single character constant
1787 @cindex character, single
1788 @cindex constant, single character
1789 A single character may be written as a single quote immediately
1790 followed by that character. The same escapes apply to characters as
1791 to strings. So if you want to write the character backslash, you
1792 must write @kbd{'\\} where the first @code{\} escapes the second
1793 @code{\}. As you can see, the quote is an acute accent, not a
1794 grave accent. A newline
1796 @ifclear abnormal-separator
1797 (or semicolon @samp{;})
1799 @ifset abnormal-separator
1801 (or at sign @samp{@@})
1804 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
1810 immediately following an acute accent is taken as a literal character
1811 and does not count as the end of a statement. The value of a character
1812 constant in a numeric expression is the machine's byte-wide code for
1813 that character. @code{@value{AS}} assumes your character code is ASCII:
1814 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
1817 @subsection Number Constants
1819 @cindex constants, number
1820 @cindex number constants
1821 @code{@value{AS}} distinguishes three kinds of numbers according to how they
1822 are stored in the target machine. @emph{Integers} are numbers that
1823 would fit into an @code{int} in the C language. @emph{Bignums} are
1824 integers, but they are stored in more than 32 bits. @emph{Flonums}
1825 are floating point numbers, described below.
1828 * Integers:: Integers
1833 * Bit Fields:: Bit Fields
1839 @subsubsection Integers
1841 @cindex constants, integer
1843 @cindex binary integers
1844 @cindex integers, binary
1845 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
1846 the binary digits @samp{01}.
1848 @cindex octal integers
1849 @cindex integers, octal
1850 An octal integer is @samp{0} followed by zero or more of the octal
1851 digits (@samp{01234567}).
1853 @cindex decimal integers
1854 @cindex integers, decimal
1855 A decimal integer starts with a non-zero digit followed by zero or
1856 more digits (@samp{0123456789}).
1858 @cindex hexadecimal integers
1859 @cindex integers, hexadecimal
1860 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
1861 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
1863 Integers have the usual values. To denote a negative integer, use
1864 the prefix operator @samp{-} discussed under expressions
1865 (@pxref{Prefix Ops,,Prefix Operators}).
1868 @subsubsection Bignums
1871 @cindex constants, bignum
1872 A @dfn{bignum} has the same syntax and semantics as an integer
1873 except that the number (or its negative) takes more than 32 bits to
1874 represent in binary. The distinction is made because in some places
1875 integers are permitted while bignums are not.
1878 @subsubsection Flonums
1880 @cindex floating point numbers
1881 @cindex constants, floating point
1883 @cindex precision, floating point
1884 A @dfn{flonum} represents a floating point number. The translation is
1885 indirect: a decimal floating point number from the text is converted by
1886 @code{@value{AS}} to a generic binary floating point number of more than
1887 sufficient precision. This generic floating point number is converted
1888 to a particular computer's floating point format (or formats) by a
1889 portion of @code{@value{AS}} specialized to that computer.
1891 A flonum is written by writing (in order)
1896 (@samp{0} is optional on the HPPA.)
1900 A letter, to tell @code{@value{AS}} the rest of the number is a flonum.
1902 @kbd{e} is recommended. Case is not important.
1904 @c FIXME: verify if flonum syntax really this vague for most cases
1905 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
1906 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
1909 On the H8/300, H8/500,
1911 and AMD 29K architectures, the letter must be
1912 one of the letters @samp{DFPRSX} (in upper or lower case).
1914 On the ARC, the letter must be one of the letters @samp{DFRS}
1915 (in upper or lower case).
1917 On the Intel 960 architecture, the letter must be
1918 one of the letters @samp{DFT} (in upper or lower case).
1920 On the HPPA architecture, the letter must be @samp{E} (upper case only).
1924 One of the letters @samp{DFPRSX} (in upper or lower case).
1927 One of the letters @samp{DFRS} (in upper or lower case).
1930 One of the letters @samp{DFPRSX} (in upper or lower case).
1933 The letter @samp{E} (upper case only).
1936 One of the letters @samp{DFT} (in upper or lower case).
1941 An optional sign: either @samp{+} or @samp{-}.
1944 An optional @dfn{integer part}: zero or more decimal digits.
1947 An optional @dfn{fractional part}: @samp{.} followed by zero
1948 or more decimal digits.
1951 An optional exponent, consisting of:
1955 An @samp{E} or @samp{e}.
1956 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
1957 @c principle this can perfectly well be different on different targets.
1959 Optional sign: either @samp{+} or @samp{-}.
1961 One or more decimal digits.
1966 At least one of the integer part or the fractional part must be
1967 present. The floating point number has the usual base-10 value.
1969 @code{@value{AS}} does all processing using integers. Flonums are computed
1970 independently of any floating point hardware in the computer running
1975 @c Bit fields are written as a general facility but are also controlled
1976 @c by a conditional-compilation flag---which is as of now (21mar91)
1977 @c turned on only by the i960 config of GAS.
1979 @subsubsection Bit Fields
1982 @cindex constants, bit field
1983 You can also define numeric constants as @dfn{bit fields}.
1984 specify two numbers separated by a colon---
1986 @var{mask}:@var{value}
1989 @code{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
1992 The resulting number is then packed
1994 @c this conditional paren in case bit fields turned on elsewhere than 960
1995 (in host-dependent byte order)
1997 into a field whose width depends on which assembler directive has the
1998 bit-field as its argument. Overflow (a result from the bitwise and
1999 requiring more binary digits to represent) is not an error; instead,
2000 more constants are generated, of the specified width, beginning with the
2001 least significant digits.@refill
2003 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2004 @code{.short}, and @code{.word} accept bit-field arguments.
2009 @chapter Sections and Relocation
2014 * Secs Background:: Background
2015 * Ld Sections:: Linker Sections
2016 * As Sections:: Assembler Internal Sections
2017 * Sub-Sections:: Sub-Sections
2021 @node Secs Background
2024 Roughly, a section is a range of addresses, with no gaps; all data
2025 ``in'' those addresses is treated the same for some particular purpose.
2026 For example there may be a ``read only'' section.
2028 @cindex linker, and assembler
2029 @cindex assembler, and linker
2030 The linker @code{@value{LD}} reads many object files (partial programs) and
2031 combines their contents to form a runnable program. When @code{@value{AS}}
2032 emits an object file, the partial program is assumed to start at address 0.
2033 @code{@value{LD}} assigns the final addresses for the partial program, so that
2034 different partial programs do not overlap. This is actually an
2035 oversimplification, but it suffices to explain how @code{@value{AS}} uses
2038 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2039 addresses. These blocks slide to their run-time addresses as rigid
2040 units; their length does not change and neither does the order of bytes
2041 within them. Such a rigid unit is called a @emph{section}. Assigning
2042 run-time addresses to sections is called @dfn{relocation}. It includes
2043 the task of adjusting mentions of object-file addresses so they refer to
2044 the proper run-time addresses.
2046 For the H8/300 and H8/500,
2047 and for the Hitachi SH,
2048 @code{@value{AS}} pads sections if needed to
2049 ensure they end on a word (sixteen bit) boundary.
2052 @cindex standard assembler sections
2053 An object file written by @code{@value{AS}} has at least three sections, any
2054 of which may be empty. These are named @dfn{text}, @dfn{data} and
2059 When it generates COFF output,
2061 @code{@value{AS}} can also generate whatever other named sections you specify
2062 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2063 If you do not use any directives that place output in the @samp{.text}
2064 or @samp{.data} sections, these sections still exist, but are empty.
2069 When @code{@value{AS}} generates SOM or ELF output for the HPPA,
2071 @code{@value{AS}} can also generate whatever other named sections you
2072 specify using the @samp{.space} and @samp{.subspace} directives. See
2073 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2074 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2075 assembler directives.
2078 Additionally, @code{@value{AS}} uses different names for the standard
2079 text, data, and bss sections when generating SOM output. Program text
2080 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2081 BSS into @samp{$BSS$}.
2085 Within the object file, the text section starts at address @code{0}, the
2086 data section follows, and the bss section follows the data section.
2089 When generating either SOM or ELF output files on the HPPA, the text
2090 section starts at address @code{0}, the data section at address
2091 @code{0x4000000}, and the bss section follows the data section.
2094 To let @code{@value{LD}} know which data changes when the sections are
2095 relocated, and how to change that data, @code{@value{AS}} also writes to the
2096 object file details of the relocation needed. To perform relocation
2097 @code{@value{LD}} must know, each time an address in the object
2101 Where in the object file is the beginning of this reference to
2104 How long (in bytes) is this reference?
2106 Which section does the address refer to? What is the numeric value of
2108 (@var{address}) @minus{} (@var{start-address of section})?
2111 Is the reference to an address ``Program-Counter relative''?
2114 @cindex addresses, format of
2115 @cindex section-relative addressing
2116 In fact, every address @code{@value{AS}} ever uses is expressed as
2118 (@var{section}) + (@var{offset into section})
2121 Further, most expressions @code{@value{AS}} computes have this section-relative
2124 (For some object formats, such as SOM for the HPPA, some expressions are
2125 symbol-relative instead.)
2128 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2129 @var{N} into section @var{secname}.''
2131 Apart from text, data and bss sections you need to know about the
2132 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2133 addresses in the absolute section remain unchanged. For example, address
2134 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2135 @code{@value{LD}}. Although the linker never arranges two partial programs'
2136 data sections with overlapping addresses after linking, @emph{by definition}
2137 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2138 part of a program is always the same address when the program is running as
2139 address @code{@{absolute@ 239@}} in any other part of the program.
2141 The idea of sections is extended to the @dfn{undefined} section. Any
2142 address whose section is unknown at assembly time is by definition
2143 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2144 Since numbers are always defined, the only way to generate an undefined
2145 address is to mention an undefined symbol. A reference to a named
2146 common block would be such a symbol: its value is unknown at assembly
2147 time so it has section @emph{undefined}.
2149 By analogy the word @emph{section} is used to describe groups of sections in
2150 the linked program. @code{@value{LD}} puts all partial programs' text
2151 sections in contiguous addresses in the linked program. It is
2152 customary to refer to the @emph{text section} of a program, meaning all
2153 the addresses of all partial programs' text sections. Likewise for
2154 data and bss sections.
2156 Some sections are manipulated by @code{@value{LD}}; others are invented for
2157 use of @code{@value{AS}} and have no meaning except during assembly.
2160 @section Linker Sections
2161 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2166 @cindex named sections
2167 @cindex sections, named
2168 @item named sections
2171 @cindex text section
2172 @cindex data section
2176 These sections hold your program. @code{@value{AS}} and @code{@value{LD}} treat them as
2177 separate but equal sections. Anything you can say of one section is
2180 When the program is running, however, it is
2181 customary for the text section to be unalterable. The
2182 text section is often shared among processes: it contains
2183 instructions, constants and the like. The data section of a running
2184 program is usually alterable: for example, C variables would be stored
2185 in the data section.
2190 This section contains zeroed bytes when your program begins running. It
2191 is used to hold unitialized variables or common storage. The length of
2192 each partial program's bss section is important, but because it starts
2193 out containing zeroed bytes there is no need to store explicit zero
2194 bytes in the object file. The bss section was invented to eliminate
2195 those explicit zeros from object files.
2197 @cindex absolute section
2198 @item absolute section
2199 Address 0 of this section is always ``relocated'' to runtime address 0.
2200 This is useful if you want to refer to an address that @code{@value{LD}} must
2201 not change when relocating. In this sense we speak of absolute
2202 addresses being ``unrelocatable'': they do not change during relocation.
2204 @cindex undefined section
2205 @item undefined section
2206 This ``section'' is a catch-all for address references to objects not in
2207 the preceding sections.
2208 @c FIXME: ref to some other doc on obj-file formats could go here.
2211 @cindex relocation example
2212 An idealized example of three relocatable sections follows.
2214 The example uses the traditional section names @samp{.text} and @samp{.data}.
2216 Memory addresses are on the horizontal axis.
2220 @c END TEXI2ROFF-KILL
2223 partial program # 1: |ttttt|dddd|00|
2230 partial program # 2: |TTT|DDD|000|
2233 +--+---+-----+--+----+---+-----+~~
2234 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2235 +--+---+-----+--+----+---+-----+~~
2237 addresses: 0 @dots{}
2244 \line{\it Partial program \#1: \hfil}
2245 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2246 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2248 \line{\it Partial program \#2: \hfil}
2249 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2250 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2252 \line{\it linked program: \hfil}
2253 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2254 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2255 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2256 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2258 \line{\it addresses: \hfil}
2262 @c END TEXI2ROFF-KILL
2265 @section Assembler Internal Sections
2267 @cindex internal assembler sections
2268 @cindex sections in messages, internal
2269 These sections are meant only for the internal use of @code{@value{AS}}. They
2270 have no meaning at run-time. You do not really need to know about these
2271 sections for most purposes; but they can be mentioned in @code{@value{AS}}
2272 warning messages, so it might be helpful to have an idea of their
2273 meanings to @code{@value{AS}}. These sections are used to permit the
2274 value of every expression in your assembly language program to be a
2275 section-relative address.
2278 @cindex assembler internal logic error
2279 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2280 An internal assembler logic error has been found. This means there is a
2281 bug in the assembler.
2283 @cindex expr (internal section)
2285 The assembler stores complex expression internally as combinations of
2286 symbols. When it needs to represent an expression as a symbol, it puts
2287 it in the expr section.
2289 @c FIXME item transfer[t] vector preload
2290 @c FIXME item transfer[t] vector postload
2291 @c FIXME item register
2295 @section Sub-Sections
2297 @cindex numbered subsections
2298 @cindex grouping data
2304 fall into two sections: text and data.
2306 You may have separate groups of
2308 data in named sections
2312 data in named sections
2318 that you want to end up near to each other in the object file, even though they
2319 are not contiguous in the assembler source. @code{@value{AS}} allows you to
2320 use @dfn{subsections} for this purpose. Within each section, there can be
2321 numbered subsections with values from 0 to 8192. Objects assembled into the
2322 same subsection go into the object file together with other objects in the same
2323 subsection. For example, a compiler might want to store constants in the text
2324 section, but might not want to have them interspersed with the program being
2325 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2326 section of code being output, and a @samp{.text 1} before each group of
2327 constants being output.
2329 Subsections are optional. If you do not use subsections, everything
2330 goes in subsection number zero.
2333 Each subsection is zero-padded up to a multiple of four bytes.
2334 (Subsections may be padded a different amount on different flavors
2335 of @code{@value{AS}}.)
2339 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2340 boundary (two bytes).
2341 The same is true on the Hitachi SH.
2344 @c FIXME section padding (alignment)?
2345 @c Rich Pixley says padding here depends on target obj code format; that
2346 @c doesn't seem particularly useful to say without further elaboration,
2347 @c so for now I say nothing about it. If this is a generic BFD issue,
2348 @c these paragraphs might need to vanish from this manual, and be
2349 @c discussed in BFD chapter of binutils (or some such).
2352 On the AMD 29K family, no particular padding is added to section or
2353 subsection sizes; @value{AS} forces no alignment on this platform.
2357 Subsections appear in your object file in numeric order, lowest numbered
2358 to highest. (All this to be compatible with other people's assemblers.)
2359 The object file contains no representation of subsections; @code{@value{LD}} and
2360 other programs that manipulate object files see no trace of them.
2361 They just see all your text subsections as a text section, and all your
2362 data subsections as a data section.
2364 To specify which subsection you want subsequent statements assembled
2365 into, use a numeric argument to specify it, in a @samp{.text
2366 @var{expression}} or a @samp{.data @var{expression}} statement.
2369 When generating COFF output, you
2374 can also use an extra subsection
2375 argument with arbitrary named sections: @samp{.section @var{name},
2378 @var{Expression} should be an absolute expression.
2379 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
2380 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
2381 begins in @code{text 0}. For instance:
2383 .text 0 # The default subsection is text 0 anyway.
2384 .ascii "This lives in the first text subsection. *"
2386 .ascii "But this lives in the second text subsection."
2388 .ascii "This lives in the data section,"
2389 .ascii "in the first data subsection."
2391 .ascii "This lives in the first text section,"
2392 .ascii "immediately following the asterisk (*)."
2395 Each section has a @dfn{location counter} incremented by one for every byte
2396 assembled into that section. Because subsections are merely a convenience
2397 restricted to @code{@value{AS}} there is no concept of a subsection location
2398 counter. There is no way to directly manipulate a location counter---but the
2399 @code{.align} directive changes it, and any label definition captures its
2400 current value. The location counter of the section where statements are being
2401 assembled is said to be the @dfn{active} location counter.
2404 @section bss Section
2407 @cindex common variable storage
2408 The bss section is used for local common variable storage.
2409 You may allocate address space in the bss section, but you may
2410 not dictate data to load into it before your program executes. When
2411 your program starts running, all the contents of the bss
2412 section are zeroed bytes.
2414 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
2415 @ref{Lcomm,,@code{.lcomm}}.
2417 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
2418 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
2421 When assembling for a target which supports multiple sections, such as ELF or
2422 COFF, you may switch into the @code{.bss} section and define symbols as usual;
2423 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
2424 section. Typically the section will only contain symbol definitions and
2425 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
2432 Symbols are a central concept: the programmer uses symbols to name
2433 things, the linker uses symbols to link, and the debugger uses symbols
2437 @cindex debuggers, and symbol order
2438 @emph{Warning:} @code{@value{AS}} does not place symbols in the object file in
2439 the same order they were declared. This may break some debuggers.
2444 * Setting Symbols:: Giving Symbols Other Values
2445 * Symbol Names:: Symbol Names
2446 * Dot:: The Special Dot Symbol
2447 * Symbol Attributes:: Symbol Attributes
2454 A @dfn{label} is written as a symbol immediately followed by a colon
2455 @samp{:}. The symbol then represents the current value of the
2456 active location counter, and is, for example, a suitable instruction
2457 operand. You are warned if you use the same symbol to represent two
2458 different locations: the first definition overrides any other
2462 On the HPPA, the usual form for a label need not be immediately followed by a
2463 colon, but instead must start in column zero. Only one label may be defined on
2464 a single line. To work around this, the HPPA version of @code{@value{AS}} also
2465 provides a special directive @code{.label} for defining labels more flexibly.
2468 @node Setting Symbols
2469 @section Giving Symbols Other Values
2471 @cindex assigning values to symbols
2472 @cindex symbol values, assigning
2473 A symbol can be given an arbitrary value by writing a symbol, followed
2474 by an equals sign @samp{=}, followed by an expression
2475 (@pxref{Expressions}). This is equivalent to using the @code{.set}
2476 directive. @xref{Set,,@code{.set}}.
2479 @section Symbol Names
2481 @cindex symbol names
2482 @cindex names, symbol
2483 @ifclear SPECIAL-SYMS
2484 Symbol names begin with a letter or with one of @samp{._}. On most
2485 machines, you can also use @code{$} in symbol names; exceptions are
2486 noted in @ref{Machine Dependencies}. That character may be followed by any
2487 string of digits, letters, dollar signs (unless otherwise noted in
2488 @ref{Machine Dependencies}), and underscores.
2491 For the AMD 29K family, @samp{?} is also allowed in the
2492 body of a symbol name, though not at its beginning.
2497 Symbol names begin with a letter or with one of @samp{._}. On the
2499 H8/500, you can also use @code{$} in symbol names. That character may
2500 be followed by any string of digits, letters, dollar signs (save on the
2501 H8/300), and underscores.
2505 Case of letters is significant: @code{foo} is a different symbol name
2508 Each symbol has exactly one name. Each name in an assembly language program
2509 refers to exactly one symbol. You may use that symbol name any number of times
2512 @subheading Local Symbol Names
2514 @cindex local symbol names
2515 @cindex symbol names, local
2516 @cindex temporary symbol names
2517 @cindex symbol names, temporary
2518 Local symbols help compilers and programmers use names temporarily.
2519 There are ten local symbol names, which are re-used throughout the
2520 program. You may refer to them using the names @samp{0} @samp{1}
2521 @dots{} @samp{9}. To define a local symbol, write a label of the form
2522 @samp{@b{N}:} (where @b{N} represents any digit). To refer to the most
2523 recent previous definition of that symbol write @samp{@b{N}b}, using the
2524 same digit as when you defined the label. To refer to the next
2525 definition of a local label, write @samp{@b{N}f}---where @b{N} gives you
2526 a choice of 10 forward references. The @samp{b} stands for
2527 ``backwards'' and the @samp{f} stands for ``forwards''.
2529 Local symbols are not emitted by the current @sc{gnu} C compiler.
2531 There is no restriction on how you can use these labels, but
2532 remember that at any point in the assembly you can refer to at most
2533 10 prior local labels and to at most 10 forward local labels.
2535 Local symbol names are only a notation device. They are immediately
2536 transformed into more conventional symbol names before the assembler
2537 uses them. The symbol names stored in the symbol table, appearing in
2538 error messages and optionally emitted to the object file have these
2543 All local labels begin with @samp{L}. Normally both @code{@value{AS}} and
2544 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
2545 used for symbols you are never intended to see. If you use the
2546 @samp{-L} option then @code{@value{AS}} retains these symbols in the
2547 object file. If you also instruct @code{@value{LD}} to retain these symbols,
2548 you may use them in debugging.
2551 If the label is written @samp{0:} then the digit is @samp{0}.
2552 If the label is written @samp{1:} then the digit is @samp{1}.
2553 And so on up through @samp{9:}.
2556 This unusual character is included so you do not accidentally invent
2557 a symbol of the same name. The character has ASCII value
2560 @item @emph{ordinal number}
2561 This is a serial number to keep the labels distinct. The first
2562 @samp{0:} gets the number @samp{1}; The 15th @samp{0:} gets the
2563 number @samp{15}; @emph{etc.}. Likewise for the other labels @samp{1:}
2567 For instance, the first @code{1:} is named @code{L1@kbd{C-A}1}, the 44th
2568 @code{3:} is named @code{L3@kbd{C-A}44}.
2571 @section The Special Dot Symbol
2573 @cindex dot (symbol)
2574 @cindex @code{.} (symbol)
2575 @cindex current address
2576 @cindex location counter
2577 The special symbol @samp{.} refers to the current address that
2578 @code{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
2579 .long .} defines @code{melvin} to contain its own address.
2580 Assigning a value to @code{.} is treated the same as a @code{.org}
2581 directive. Thus, the expression @samp{.=.+4} is the same as saying
2582 @ifclear no-space-dir
2591 @node Symbol Attributes
2592 @section Symbol Attributes
2594 @cindex symbol attributes
2595 @cindex attributes, symbol
2596 Every symbol has, as well as its name, the attributes ``Value'' and
2597 ``Type''. Depending on output format, symbols can also have auxiliary
2600 The detailed definitions are in @file{a.out.h}.
2603 If you use a symbol without defining it, @code{@value{AS}} assumes zero for
2604 all these attributes, and probably won't warn you. This makes the
2605 symbol an externally defined symbol, which is generally what you
2609 * Symbol Value:: Value
2610 * Symbol Type:: Type
2613 * a.out Symbols:: Symbol Attributes: @code{a.out}
2617 * a.out Symbols:: Symbol Attributes: @code{a.out}
2620 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
2625 * COFF Symbols:: Symbol Attributes for COFF
2628 * SOM Symbols:: Symbol Attributes for SOM
2635 @cindex value of a symbol
2636 @cindex symbol value
2637 The value of a symbol is (usually) 32 bits. For a symbol which labels a
2638 location in the text, data, bss or absolute sections the value is the
2639 number of addresses from the start of that section to the label.
2640 Naturally for text, data and bss sections the value of a symbol changes
2641 as @code{@value{LD}} changes section base addresses during linking. Absolute
2642 symbols' values do not change during linking: that is why they are
2645 The value of an undefined symbol is treated in a special way. If it is
2646 0 then the symbol is not defined in this assembler source file, and
2647 @code{@value{LD}} tries to determine its value from other files linked into the
2648 same program. You make this kind of symbol simply by mentioning a symbol
2649 name without defining it. A non-zero value represents a @code{.comm}
2650 common declaration. The value is how much common storage to reserve, in
2651 bytes (addresses). The symbol refers to the first address of the
2657 @cindex type of a symbol
2659 The type attribute of a symbol contains relocation (section)
2660 information, any flag settings indicating that a symbol is external, and
2661 (optionally), other information for linkers and debuggers. The exact
2662 format depends on the object-code output format in use.
2667 @c The following avoids a "widow" subsection title. @group would be
2668 @c better if it were available outside examples.
2671 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
2673 @cindex @code{b.out} symbol attributes
2674 @cindex symbol attributes, @code{b.out}
2675 These symbol attributes appear only when @code{@value{AS}} is configured for
2676 one of the Berkeley-descended object output formats---@code{a.out} or
2682 @subsection Symbol Attributes: @code{a.out}
2684 @cindex @code{a.out} symbol attributes
2685 @cindex symbol attributes, @code{a.out}
2691 @subsection Symbol Attributes: @code{a.out}
2693 @cindex @code{a.out} symbol attributes
2694 @cindex symbol attributes, @code{a.out}
2698 * Symbol Desc:: Descriptor
2699 * Symbol Other:: Other
2703 @subsubsection Descriptor
2705 @cindex descriptor, of @code{a.out} symbol
2706 This is an arbitrary 16-bit value. You may establish a symbol's
2707 descriptor value by using a @code{.desc} statement
2708 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
2712 @subsubsection Other
2714 @cindex other attribute, of @code{a.out} symbol
2715 This is an arbitrary 8-bit value. It means nothing to @code{@value{AS}}.
2720 @subsection Symbol Attributes for COFF
2722 @cindex COFF symbol attributes
2723 @cindex symbol attributes, COFF
2725 The COFF format supports a multitude of auxiliary symbol attributes;
2726 like the primary symbol attributes, they are set between @code{.def} and
2727 @code{.endef} directives.
2729 @subsubsection Primary Attributes
2731 @cindex primary attributes, COFF symbols
2732 The symbol name is set with @code{.def}; the value and type,
2733 respectively, with @code{.val} and @code{.type}.
2735 @subsubsection Auxiliary Attributes
2737 @cindex auxiliary attributes, COFF symbols
2738 The @code{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
2739 @code{.size}, and @code{.tag} can generate auxiliary symbol table
2740 information for COFF.
2745 @subsection Symbol Attributes for SOM
2747 @cindex SOM symbol attributes
2748 @cindex symbol attributes, SOM
2750 The SOM format for the HPPA supports a multitude of symbol attributes set with
2751 the @code{.EXPORT} and @code{.IMPORT} directives.
2753 The attributes are described in @cite{HP9000 Series 800 Assembly
2754 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
2755 @code{EXPORT} assembler directive documentation.
2759 @chapter Expressions
2763 @cindex numeric values
2764 An @dfn{expression} specifies an address or numeric value.
2765 Whitespace may precede and/or follow an expression.
2767 The result of an expression must be an absolute number, or else an offset into
2768 a particular section. If an expression is not absolute, and there is not
2769 enough information when @code{@value{AS}} sees the expression to know its
2770 section, a second pass over the source program might be necessary to interpret
2771 the expression---but the second pass is currently not implemented.
2772 @code{@value{AS}} aborts with an error message in this situation.
2775 * Empty Exprs:: Empty Expressions
2776 * Integer Exprs:: Integer Expressions
2780 @section Empty Expressions
2782 @cindex empty expressions
2783 @cindex expressions, empty
2784 An empty expression has no value: it is just whitespace or null.
2785 Wherever an absolute expression is required, you may omit the
2786 expression, and @code{@value{AS}} assumes a value of (absolute) 0. This
2787 is compatible with other assemblers.
2790 @section Integer Expressions
2792 @cindex integer expressions
2793 @cindex expressions, integer
2794 An @dfn{integer expression} is one or more @emph{arguments} delimited
2795 by @emph{operators}.
2798 * Arguments:: Arguments
2799 * Operators:: Operators
2800 * Prefix Ops:: Prefix Operators
2801 * Infix Ops:: Infix Operators
2805 @subsection Arguments
2807 @cindex expression arguments
2808 @cindex arguments in expressions
2809 @cindex operands in expressions
2810 @cindex arithmetic operands
2811 @dfn{Arguments} are symbols, numbers or subexpressions. In other
2812 contexts arguments are sometimes called ``arithmetic operands''. In
2813 this manual, to avoid confusing them with the ``instruction operands'' of
2814 the machine language, we use the term ``argument'' to refer to parts of
2815 expressions only, reserving the word ``operand'' to refer only to machine
2816 instruction operands.
2818 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
2819 @var{section} is one of text, data, bss, absolute,
2820 or undefined. @var{NNN} is a signed, 2's complement 32 bit
2823 Numbers are usually integers.
2825 A number can be a flonum or bignum. In this case, you are warned
2826 that only the low order 32 bits are used, and @code{@value{AS}} pretends
2827 these 32 bits are an integer. You may write integer-manipulating
2828 instructions that act on exotic constants, compatible with other
2831 @cindex subexpressions
2832 Subexpressions are a left parenthesis @samp{(} followed by an integer
2833 expression, followed by a right parenthesis @samp{)}; or a prefix
2834 operator followed by an argument.
2837 @subsection Operators
2839 @cindex operators, in expressions
2840 @cindex arithmetic functions
2841 @cindex functions, in expressions
2842 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
2843 operators are followed by an argument. Infix operators appear
2844 between their arguments. Operators may be preceded and/or followed by
2848 @subsection Prefix Operator
2850 @cindex prefix operators
2851 @code{@value{AS}} has the following @dfn{prefix operators}. They each take
2852 one argument, which must be absolute.
2854 @c the tex/end tex stuff surrounding this small table is meant to make
2855 @c it align, on the printed page, with the similar table in the next
2856 @c section (which is inside an enumerate).
2858 \global\advance\leftskip by \itemindent
2863 @dfn{Negation}. Two's complement negation.
2865 @dfn{Complementation}. Bitwise not.
2869 \global\advance\leftskip by -\itemindent
2873 @subsection Infix Operators
2875 @cindex infix operators
2876 @cindex operators, permitted arguments
2877 @dfn{Infix operators} take two arguments, one on either side. Operators
2878 have precedence, but operations with equal precedence are performed left
2879 to right. Apart from @code{+} or @code{-}, both arguments must be
2880 absolute, and the result is absolute.
2883 @cindex operator precedence
2884 @cindex precedence of operators
2891 @dfn{Multiplication}.
2894 @dfn{Division}. Truncation is the same as the C operator @samp{/}
2901 @dfn{Shift Left}. Same as the C operator @samp{<<}.
2905 @dfn{Shift Right}. Same as the C operator @samp{>>}.
2909 Intermediate precedence
2914 @dfn{Bitwise Inclusive Or}.
2920 @dfn{Bitwise Exclusive Or}.
2923 @dfn{Bitwise Or Not}.
2930 @cindex addition, permitted arguments
2931 @cindex plus, permitted arguments
2932 @cindex arguments for addition
2934 @dfn{Addition}. If either argument is absolute, the result has the section of
2935 the other argument. You may not add together arguments from different
2938 @cindex subtraction, permitted arguments
2939 @cindex minus, permitted arguments
2940 @cindex arguments for subtraction
2942 @dfn{Subtraction}. If the right argument is absolute, the
2943 result has the section of the left argument.
2944 If both arguments are in the same section, the result is absolute.
2945 You may not subtract arguments from different sections.
2946 @c FIXME is there still something useful to say about undefined - undefined ?
2950 In short, it's only meaningful to add or subtract the @emph{offsets} in an
2951 address; you can only have a defined section in one of the two arguments.
2954 @chapter Assembler Directives
2956 @cindex directives, machine independent
2957 @cindex pseudo-ops, machine independent
2958 @cindex machine independent directives
2959 All assembler directives have names that begin with a period (@samp{.}).
2960 The rest of the name is letters, usually in lower case.
2962 This chapter discusses directives that are available regardless of the
2963 target machine configuration for the @sc{gnu} assembler.
2965 Some machine configurations provide additional directives.
2966 @xref{Machine Dependencies}.
2969 @ifset machine-directives
2970 @xref{Machine Dependencies} for additional directives.
2975 * Abort:: @code{.abort}
2977 * ABORT:: @code{.ABORT}
2980 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
2981 * App-File:: @code{.app-file @var{string}}
2982 * Ascii:: @code{.ascii "@var{string}"}@dots{}
2983 * Asciz:: @code{.asciz "@var{string}"}@dots{}
2984 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
2985 * Byte:: @code{.byte @var{expressions}}
2986 * Comm:: @code{.comm @var{symbol} , @var{length} }
2987 * Data:: @code{.data @var{subsection}}
2989 * Def:: @code{.def @var{name}}
2992 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
2998 * Double:: @code{.double @var{flonums}}
2999 * Eject:: @code{.eject}
3000 * Else:: @code{.else}
3002 * Endef:: @code{.endef}
3005 * Endif:: @code{.endif}
3006 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3007 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3009 * Extern:: @code{.extern}
3010 @ifclear no-file-dir
3011 * File:: @code{.file @var{string}}
3014 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3015 * Float:: @code{.float @var{flonums}}
3016 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3017 * hword:: @code{.hword @var{expressions}}
3018 * Ident:: @code{.ident}
3019 * If:: @code{.if @var{absolute expression}}
3020 * Include:: @code{.include "@var{file}"}
3021 * Int:: @code{.int @var{expressions}}
3022 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3023 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3024 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3025 * Lflags:: @code{.lflags}
3026 @ifclear no-line-dir
3027 * Line:: @code{.line @var{line-number}}
3030 * Ln:: @code{.ln @var{line-number}}
3031 * Linkonce:: @code{.linkonce [@var{type}]}
3032 * List:: @code{.list}
3033 * Long:: @code{.long @var{expressions}}
3035 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3038 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3039 * MRI:: @code{.mri @var{val}}
3041 * Nolist:: @code{.nolist}
3042 * Octa:: @code{.octa @var{bignums}}
3043 * Org:: @code{.org @var{new-lc} , @var{fill}}
3044 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3045 * Psize:: @code{.psize @var{lines}, @var{columns}}
3046 * Quad:: @code{.quad @var{bignums}}
3047 * Rept:: @code{.rept @var{count}}
3048 * Sbttl:: @code{.sbttl "@var{subheading}"}
3050 * Scl:: @code{.scl @var{class}}
3051 * Section:: @code{.section @var{name}, @var{subsection}}
3054 * Set:: @code{.set @var{symbol}, @var{expression}}
3055 * Short:: @code{.short @var{expressions}}
3056 * Single:: @code{.single @var{flonums}}
3058 * Size:: @code{.size}
3061 * Skip:: @code{.skip @var{size} , @var{fill}}
3062 * Sleb128:: @code{.sleb128 @var{expressions}}
3063 * Space:: @code{.space @var{size} , @var{fill}}
3065 * Stab:: @code{.stabd, .stabn, .stabs}
3068 * String:: @code{.string "@var{str}"}
3070 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3073 * Tag:: @code{.tag @var{structname}}
3076 * Text:: @code{.text @var{subsection}}
3077 * Title:: @code{.title "@var{heading}"}
3079 * Type:: @code{.type @var{int}}
3080 * Val:: @code{.val @var{addr}}
3083 * Uleb128:: @code{.uleb128 @var{expressions}}
3084 * Word:: @code{.word @var{expressions}}
3085 * Deprecated:: Deprecated Directives
3089 @section @code{.abort}
3091 @cindex @code{abort} directive
3092 @cindex stopping the assembly
3093 This directive stops the assembly immediately. It is for
3094 compatibility with other assemblers. The original idea was that the
3095 assembly language source would be piped into the assembler. If the sender
3096 of the source quit, it could use this directive tells @code{@value{AS}} to
3097 quit also. One day @code{.abort} will not be supported.
3101 @section @code{.ABORT}
3103 @cindex @code{ABORT} directive
3104 When producing COFF output, @code{@value{AS}} accepts this directive as a
3105 synonym for @samp{.abort}.
3108 When producing @code{b.out} output, @code{@value{AS}} accepts this directive,
3114 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3116 @cindex padding the location counter
3117 @cindex @code{align} directive
3118 Pad the location counter (in the current subsection) to a particular storage
3119 boundary. The first expression (which must be absolute) is the alignment
3120 required, as described below.
3122 The second expression (also absolute) gives the fill value to be stored in the
3123 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3124 padding bytes are normally zero. However, on some systems, if the section is
3125 marked as containing code and the fill value is omitted, the space is filled
3126 with no-op instructions.
3128 The third expression is also absolute, and is also optional. If it is present,
3129 it is the maximum number of bytes that should be skipped by this alignment
3130 directive. If doing the alignment would require skipping more bytes than the
3131 specified maximum, then the alignment is not done at all. You can omit the
3132 fill value (the second argument) entirely by simply using two commas after the
3133 required alignment; this can be useful if you want the alignment to be filled
3134 with no-op instructions when appropriate.
3136 The way the required alignment is specified varies from system to system.
3137 For the a29k, hppa, m68k, m88k, w65, sparc, and Hitachi SH, and i386 using ELF
3139 the first expression is the
3140 alignment request in bytes. For example @samp{.align 8} advances
3141 the location counter until it is a multiple of 8. If the location counter
3142 is already a multiple of 8, no change is needed.
3144 For other systems, including the i386 using a.out format, it is the
3145 number of low-order zero bits the location counter must have after
3146 advancement. For example @samp{.align 3} advances the location
3147 counter until it a multiple of 8. If the location counter is already a
3148 multiple of 8, no change is needed.
3150 This inconsistency is due to the different behaviors of the various
3151 native assemblers for these systems which GAS must emulate.
3152 GAS also provides @code{.balign} and @code{.p2align} directives,
3153 described later, which have a consistent behavior across all
3154 architectures (but are specific to GAS).
3157 @section @code{.app-file @var{string}}
3159 @cindex logical file name
3160 @cindex file name, logical
3161 @cindex @code{app-file} directive
3163 @ifclear no-file-dir
3164 (which may also be spelled @samp{.file})
3166 tells @code{@value{AS}} that we are about to start a new
3167 logical file. @var{string} is the new file name. In general, the
3168 filename is recognized whether or not it is surrounded by quotes @samp{"};
3169 but if you wish to specify an empty file name is permitted,
3170 you must give the quotes--@code{""}. This statement may go away in
3171 future: it is only recognized to be compatible with old @code{@value{AS}}
3175 @section @code{.ascii "@var{string}"}@dots{}
3177 @cindex @code{ascii} directive
3178 @cindex string literals
3179 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3180 separated by commas. It assembles each string (with no automatic
3181 trailing zero byte) into consecutive addresses.
3184 @section @code{.asciz "@var{string}"}@dots{}
3186 @cindex @code{asciz} directive
3187 @cindex zero-terminated strings
3188 @cindex null-terminated strings
3189 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3190 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3193 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3195 @cindex padding the location counter given number of bytes
3196 @cindex @code{balign} directive
3197 Pad the location counter (in the current subsection) to a particular
3198 storage boundary. The first expression (which must be absolute) is the
3199 alignment request in bytes. For example @samp{.balign 8} advances
3200 the location counter until it is a multiple of 8. If the location counter
3201 is already a multiple of 8, no change is needed.
3203 The second expression (also absolute) gives the fill value to be stored in the
3204 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3205 padding bytes are normally zero. However, on some systems, if the section is
3206 marked as containing code and the fill value is omitted, the space is filled
3207 with no-op instructions.
3209 The third expression is also absolute, and is also optional. If it is present,
3210 it is the maximum number of bytes that should be skipped by this alignment
3211 directive. If doing the alignment would require skipping more bytes than the
3212 specified maximum, then the alignment is not done at all. You can omit the
3213 fill value (the second argument) entirely by simply using two commas after the
3214 required alignment; this can be useful if you want the alignment to be filled
3215 with no-op instructions when appropriate.
3217 @cindex @code{balignw} directive
3218 @cindex @code{balignl} directive
3219 The @code{.balignw} and @code{.balignl} directives are variants of the
3220 @code{.balign} directive. The @code{.balignw} directive treats the fill
3221 pattern as a two byte word value. The @code{.balignl} directives treats the
3222 fill pattern as a four byte longword value. For example, @code{.balignw
3223 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3224 filled in with the value 0x368d (the exact placement of the bytes depends upon
3225 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3229 @section @code{.byte @var{expressions}}
3231 @cindex @code{byte} directive
3232 @cindex integers, one byte
3233 @code{.byte} expects zero or more expressions, separated by commas.
3234 Each expression is assembled into the next byte.
3237 @section @code{.comm @var{symbol} , @var{length} }
3239 @cindex @code{comm} directive
3240 @cindex symbol, common
3241 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
3242 common symbol in one object file may be merged with a defined or common symbol
3243 of the same name in another object file. If @code{@value{LD}} does not see a
3244 definition for the symbol--just one or more common symbols--then it will
3245 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
3246 absolute expression. If @code{@value{LD}} sees multiple common symbols with
3247 the same name, and they do not all have the same size, it will allocate space
3248 using the largest size.
3251 When using ELF, the @code{.comm} directive takes an optional third argument.
3252 This is the desired alignment of the symbol, specified as a byte boundary (for
3253 example, an alignment of 16 means that the least significant 4 bits of the
3254 address should be zero). The alignment must be an absolute expression, and it
3255 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
3256 for the common symbol, it will use the alignment when placing the symbol. If
3257 no alignment is specified, @code{@value{AS}} will set the alignment to the
3258 largest power of two less than or equal to the size of the symbol, up to a
3263 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
3264 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
3268 @section @code{.data @var{subsection}}
3270 @cindex @code{data} directive
3271 @code{.data} tells @code{@value{AS}} to assemble the following statements onto the
3272 end of the data subsection numbered @var{subsection} (which is an
3273 absolute expression). If @var{subsection} is omitted, it defaults
3278 @section @code{.def @var{name}}
3280 @cindex @code{def} directive
3281 @cindex COFF symbols, debugging
3282 @cindex debugging COFF symbols
3283 Begin defining debugging information for a symbol @var{name}; the
3284 definition extends until the @code{.endef} directive is encountered.
3287 This directive is only observed when @code{@value{AS}} is configured for COFF
3288 format output; when producing @code{b.out}, @samp{.def} is recognized,
3295 @section @code{.desc @var{symbol}, @var{abs-expression}}
3297 @cindex @code{desc} directive
3298 @cindex COFF symbol descriptor
3299 @cindex symbol descriptor, COFF
3300 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
3301 to the low 16 bits of an absolute expression.
3304 The @samp{.desc} directive is not available when @code{@value{AS}} is
3305 configured for COFF output; it is only for @code{a.out} or @code{b.out}
3306 object format. For the sake of compatibility, @code{@value{AS}} accepts
3307 it, but produces no output, when configured for COFF.
3313 @section @code{.dim}
3315 @cindex @code{dim} directive
3316 @cindex COFF auxiliary symbol information
3317 @cindex auxiliary symbol information, COFF
3318 This directive is generated by compilers to include auxiliary debugging
3319 information in the symbol table. It is only permitted inside
3320 @code{.def}/@code{.endef} pairs.
3323 @samp{.dim} is only meaningful when generating COFF format output; when
3324 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
3330 @section @code{.double @var{flonums}}
3332 @cindex @code{double} directive
3333 @cindex floating point numbers (double)
3334 @code{.double} expects zero or more flonums, separated by commas. It
3335 assembles floating point numbers.
3337 The exact kind of floating point numbers emitted depends on how
3338 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
3342 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
3343 in @sc{ieee} format.
3348 @section @code{.eject}
3350 @cindex @code{eject} directive
3351 @cindex new page, in listings
3352 @cindex page, in listings
3353 @cindex listing control: new page
3354 Force a page break at this point, when generating assembly listings.
3357 @section @code{.else}
3359 @cindex @code{else} directive
3360 @code{.else} is part of the @code{@value{AS}} support for conditional
3361 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
3362 of code to be assembled if the condition for the preceding @code{.if}
3366 @node End, Endef, Else, Pseudo Ops
3367 @section @code{.end}
3369 @cindex @code{end} directive
3370 This doesn't do anything---but isn't an s_ignore, so I suspect it's
3371 meant to do something eventually (which is why it isn't documented here
3372 as "for compatibility with blah").
3377 @section @code{.endef}
3379 @cindex @code{endef} directive
3380 This directive flags the end of a symbol definition begun with
3384 @samp{.endef} is only meaningful when generating COFF format output; if
3385 @code{@value{AS}} is configured to generate @code{b.out}, it accepts this
3386 directive but ignores it.
3391 @section @code{.endif}
3393 @cindex @code{endif} directive
3394 @code{.endif} is part of the @code{@value{AS}} support for conditional assembly;
3395 it marks the end of a block of code that is only assembled
3396 conditionally. @xref{If,,@code{.if}}.
3399 @section @code{.equ @var{symbol}, @var{expression}}
3401 @cindex @code{equ} directive
3402 @cindex assigning values to symbols
3403 @cindex symbols, assigning values to
3404 This directive sets the value of @var{symbol} to @var{expression}.
3405 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
3408 The syntax for @code{equ} on the HPPA is
3409 @samp{@var{symbol} .equ @var{expression}}.
3413 @section @code{.equiv @var{symbol}, @var{expression}}
3414 @cindex @code{equiv} directive
3415 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
3416 the assembler will signal an error if @var{symbol} is already defined.
3418 Except for the contents of the error message, this is roughly equivalent to
3427 @section @code{.err}
3428 @cindex @code{err} directive
3429 If @code{@value{AS}} assembles a @code{.err} directive, it will print an error
3430 message and, unless the @code{-Z} option was used, it will not generate an
3431 object file. This can be used to signal error an conditionally compiled code.
3434 @section @code{.extern}
3436 @cindex @code{extern} directive
3437 @code{.extern} is accepted in the source program---for compatibility
3438 with other assemblers---but it is ignored. @code{@value{AS}} treats
3439 all undefined symbols as external.
3441 @ifclear no-file-dir
3443 @section @code{.file @var{string}}
3445 @cindex @code{file} directive
3446 @cindex logical file name
3447 @cindex file name, logical
3448 @code{.file} (which may also be spelled @samp{.app-file}) tells
3449 @code{@value{AS}} that we are about to start a new logical file.
3450 @var{string} is the new file name. In general, the filename is
3451 recognized whether or not it is surrounded by quotes @samp{"}; but if
3452 you wish to specify an empty file name, you must give the
3453 quotes--@code{""}. This statement may go away in future: it is only
3454 recognized to be compatible with old @code{@value{AS}} programs.
3456 In some configurations of @code{@value{AS}}, @code{.file} has already been
3457 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
3462 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
3464 @cindex @code{fill} directive
3465 @cindex writing patterns in memory
3466 @cindex patterns, writing in memory
3467 @var{result}, @var{size} and @var{value} are absolute expressions.
3468 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
3469 may be zero or more. @var{Size} may be zero or more, but if it is
3470 more than 8, then it is deemed to have the value 8, compatible with
3471 other people's assemblers. The contents of each @var{repeat} bytes
3472 is taken from an 8-byte number. The highest order 4 bytes are
3473 zero. The lowest order 4 bytes are @var{value} rendered in the
3474 byte-order of an integer on the computer @code{@value{AS}} is assembling for.
3475 Each @var{size} bytes in a repetition is taken from the lowest order
3476 @var{size} bytes of this number. Again, this bizarre behavior is
3477 compatible with other people's assemblers.
3479 @var{size} and @var{value} are optional.
3480 If the second comma and @var{value} are absent, @var{value} is
3481 assumed zero. If the first comma and following tokens are absent,
3482 @var{size} is assumed to be 1.
3485 @section @code{.float @var{flonums}}
3487 @cindex floating point numbers (single)
3488 @cindex @code{float} directive
3489 This directive assembles zero or more flonums, separated by commas. It
3490 has the same effect as @code{.single}.
3492 The exact kind of floating point numbers emitted depends on how
3493 @code{@value{AS}} is configured.
3494 @xref{Machine Dependencies}.
3498 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
3499 in @sc{ieee} format.
3504 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3506 @cindex @code{global} directive
3507 @cindex symbol, making visible to linker
3508 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
3509 @var{symbol} in your partial program, its value is made available to
3510 other partial programs that are linked with it. Otherwise,
3511 @var{symbol} takes its attributes from a symbol of the same name
3512 from another file linked into the same program.
3514 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
3515 compatibility with other assemblers.
3518 On the HPPA, @code{.global} is not always enough to make it accessible to other
3519 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
3520 @xref{HPPA Directives,, HPPA Assembler Directives}.
3524 @section @code{.hword @var{expressions}}
3526 @cindex @code{hword} directive
3527 @cindex integers, 16-bit
3528 @cindex numbers, 16-bit
3529 @cindex sixteen bit integers
3530 This expects zero or more @var{expressions}, and emits
3531 a 16 bit number for each.
3534 This directive is a synonym for @samp{.short}; depending on the target
3535 architecture, it may also be a synonym for @samp{.word}.
3539 This directive is a synonym for @samp{.short}.
3542 This directive is a synonym for both @samp{.short} and @samp{.word}.
3547 @section @code{.ident}
3549 @cindex @code{ident} directive
3550 This directive is used by some assemblers to place tags in object files.
3551 @code{@value{AS}} simply accepts the directive for source-file
3552 compatibility with such assemblers, but does not actually emit anything
3556 @section @code{.if @var{absolute expression}}
3558 @cindex conditional assembly
3559 @cindex @code{if} directive
3560 @code{.if} marks the beginning of a section of code which is only
3561 considered part of the source program being assembled if the argument
3562 (which must be an @var{absolute expression}) is non-zero. The end of
3563 the conditional section of code must be marked by @code{.endif}
3564 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
3565 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
3567 The following variants of @code{.if} are also supported:
3569 @cindex @code{ifdef} directive
3570 @item .ifdef @var{symbol}
3571 Assembles the following section of code if the specified @var{symbol}
3575 @cindex @code{ifeqs} directive
3577 Not yet implemented.
3580 @cindex @code{ifndef} directive
3581 @cindex @code{ifnotdef} directive
3582 @item .ifndef @var{symbol}
3583 @itemx .ifnotdef @var{symbol}
3584 Assembles the following section of code if the specified @var{symbol}
3585 has not been defined. Both spelling variants are equivalent.
3589 Not yet implemented.
3594 @section @code{.include "@var{file}"}
3596 @cindex @code{include} directive
3597 @cindex supporting files, including
3598 @cindex files, including
3599 This directive provides a way to include supporting files at specified
3600 points in your source program. The code from @var{file} is assembled as
3601 if it followed the point of the @code{.include}; when the end of the
3602 included file is reached, assembly of the original file continues. You
3603 can control the search paths used with the @samp{-I} command-line option
3604 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
3608 @section @code{.int @var{expressions}}
3610 @cindex @code{int} directive
3611 @cindex integers, 32-bit
3612 Expect zero or more @var{expressions}, of any section, separated by commas.
3613 For each expression, emit a number that, at run time, is the value of that
3614 expression. The byte order and bit size of the number depends on what kind
3615 of target the assembly is for.
3619 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
3620 integers. On the H8/300H and the Hitachi SH, however, @code{.int} emits
3626 @section @code{.irp @var{symbol},@var{values}}@dots{}
3628 @cindex @code{irp} directive
3629 Evaluate a sequence of statements assigning different values to @var{symbol}.
3630 The sequence of statements starts at the @code{.irp} directive, and is
3631 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
3632 set to @var{value}, and the sequence of statements is assembled. If no
3633 @var{value} is listed, the sequence of statements is assembled once, with
3634 @var{symbol} set to the null string. To refer to @var{symbol} within the
3635 sequence of statements, use @var{\symbol}.
3637 For example, assembling
3645 is equivalent to assembling
3654 @section @code{.irpc @var{symbol},@var{values}}@dots{}
3656 @cindex @code{irpc} directive
3657 Evaluate a sequence of statements assigning different values to @var{symbol}.
3658 The sequence of statements starts at the @code{.irpc} directive, and is
3659 terminated by an @code{.endr} directive. For each character in @var{value},
3660 @var{symbol} is set to the character, and the sequence of statements is
3661 assembled. If no @var{value} is listed, the sequence of statements is
3662 assembled once, with @var{symbol} set to the null string. To refer to
3663 @var{symbol} within the sequence of statements, use @var{\symbol}.
3665 For example, assembling
3673 is equivalent to assembling
3682 @section @code{.lcomm @var{symbol} , @var{length}}
3684 @cindex @code{lcomm} directive
3685 @cindex local common symbols
3686 @cindex symbols, local common
3687 Reserve @var{length} (an absolute expression) bytes for a local common
3688 denoted by @var{symbol}. The section and value of @var{symbol} are
3689 those of the new local common. The addresses are allocated in the bss
3690 section, so that at run-time the bytes start off zeroed. @var{Symbol}
3691 is not declared global (@pxref{Global,,@code{.global}}), so is normally
3692 not visible to @code{@value{LD}}.
3695 Some targets permit a third argument to be used with @code{.lcomm}. This
3696 argument specifies the desired alignment of the symbol in the bss section.
3700 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
3701 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
3705 @section @code{.lflags}
3707 @cindex @code{lflags} directive (ignored)
3708 @code{@value{AS}} accepts this directive, for compatibility with other
3709 assemblers, but ignores it.
3711 @ifclear no-line-dir
3713 @section @code{.line @var{line-number}}
3715 @cindex @code{line} directive
3719 @section @code{.ln @var{line-number}}
3721 @cindex @code{ln} directive
3723 @cindex logical line number
3725 Change the logical line number. @var{line-number} must be an absolute
3726 expression. The next line has that logical line number. Therefore any other
3727 statements on the current line (after a statement separator character) are
3728 reported as on logical line number @var{line-number} @minus{} 1. One day
3729 @code{@value{AS}} will no longer support this directive: it is recognized only
3730 for compatibility with existing assembler programs.
3734 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
3735 not available; use the synonym @code{.ln} in that context.
3740 @ifclear no-line-dir
3741 Even though this is a directive associated with the @code{a.out} or
3742 @code{b.out} object-code formats, @code{@value{AS}} still recognizes it
3743 when producing COFF output, and treats @samp{.line} as though it
3744 were the COFF @samp{.ln} @emph{if} it is found outside a
3745 @code{.def}/@code{.endef} pair.
3747 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
3748 used by compilers to generate auxiliary symbol information for
3753 @section @code{.linkonce [@var{type}]}
3755 @cindex @code{linkonce} directive
3756 @cindex common sections
3757 Mark the current section so that the linker only includes a single copy of it.
3758 This may be used to include the same section in several different object files,
3759 but ensure that the linker will only include it once in the final output file.
3760 The @code{.linkonce} pseudo-op must be used for each instance of the section.
3761 Duplicate sections are detected based on the section name, so it should be
3764 This directive is only supported by a few object file formats; as of this
3765 writing, the only object file format which supports it is the Portable
3766 Executable format used on Windows NT.
3768 The @var{type} argument is optional. If specified, it must be one of the
3769 following strings. For example:
3773 Not all types may be supported on all object file formats.
3777 Silently discard duplicate sections. This is the default.
3780 Warn if there are duplicate sections, but still keep only one copy.
3783 Warn if any of the duplicates have different sizes.
3786 Warn if any of the duplicates do not have exactly the same contents.
3790 @section @code{.ln @var{line-number}}
3792 @cindex @code{ln} directive
3793 @ifclear no-line-dir
3794 @samp{.ln} is a synonym for @samp{.line}.
3797 Tell @code{@value{AS}} to change the logical line number. @var{line-number}
3798 must be an absolute expression. The next line has that logical
3799 line number, so any other statements on the current line (after a
3800 statement separator character @code{;}) are reported as on logical
3801 line number @var{line-number} @minus{} 1.
3804 This directive is accepted, but ignored, when @code{@value{AS}} is
3805 configured for @code{b.out}; its effect is only associated with COFF
3811 @section @code{.mri @var{val}}
3813 @cindex @code{mri} directive
3814 @cindex MRI mode, temporarily
3815 If @var{val} is non-zero, this tells @code{@value{AS}} to enter MRI mode. If
3816 @var{val} is zero, this tells @code{@value{AS}} to exit MRI mode. This change
3817 affects code assembled until the next @code{.mri} directive, or until the end
3818 of the file. @xref{M, MRI mode, MRI mode}.
3821 @section @code{.list}
3823 @cindex @code{list} directive
3824 @cindex listing control, turning on
3825 Control (in conjunction with the @code{.nolist} directive) whether or
3826 not assembly listings are generated. These two directives maintain an
3827 internal counter (which is zero initially). @code{.list} increments the
3828 counter, and @code{.nolist} decrements it. Assembly listings are
3829 generated whenever the counter is greater than zero.
3831 By default, listings are disabled. When you enable them (with the
3832 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
3833 the initial value of the listing counter is one.
3836 @section @code{.long @var{expressions}}
3838 @cindex @code{long} directive
3839 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
3842 @c no one seems to know what this is for or whether this description is
3843 @c what it really ought to do
3845 @section @code{.lsym @var{symbol}, @var{expression}}
3847 @cindex @code{lsym} directive
3848 @cindex symbol, not referenced in assembly
3849 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
3850 the hash table, ensuring it cannot be referenced by name during the
3851 rest of the assembly. This sets the attributes of the symbol to be
3852 the same as the expression value:
3854 @var{other} = @var{descriptor} = 0
3855 @var{type} = @r{(section of @var{expression})}
3856 @var{value} = @var{expression}
3859 The new symbol is not flagged as external.
3863 @section @code{.macro}
3866 The commands @code{.macro} and @code{.endm} allow you to define macros that
3867 generate assembly output. For example, this definition specifies a macro
3868 @code{sum} that puts a sequence of numbers into memory:
3871 .macro sum from=0, to=5
3880 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
3892 @item .macro @var{macname}
3893 @itemx .macro @var{macname} @var{macargs} @dots{}
3894 @cindex @code{macro} directive
3895 Begin the definition of a macro called @var{macname}. If your macro
3896 definition requires arguments, specify their names after the macro name,
3897 separated by commas or spaces. You can supply a default value for any
3898 macro argument by following the name with @samp{=@var{deflt}}. For
3899 example, these are all valid @code{.macro} statements:
3903 Begin the definition of a macro called @code{comm}, which takes no
3906 @item .macro plus1 p, p1
3907 @itemx .macro plus1 p p1
3908 Either statement begins the definition of a macro called @code{plus1},
3909 which takes two arguments; within the macro definition, write
3910 @samp{\p} or @samp{\p1} to evaluate the arguments.
3912 @item .macro reserve_str p1=0 p2
3913 Begin the definition of a macro called @code{reserve_str}, with two
3914 arguments. The first argument has a default value, but not the second.
3915 After the definition is complete, you can call the macro either as
3916 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
3917 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
3918 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
3919 @samp{0}, and @samp{\p2} evaluating to @var{b}).
3922 When you call a macro, you can specify the argument values either by
3923 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
3924 @samp{sum to=17, from=9}.
3927 @cindex @code{endm} directive
3928 Mark the end of a macro definition.
3931 @cindex @code{exitm} directive
3932 Exit early from the current macro definition.
3934 @cindex number of macros executed
3935 @cindex macros, count executed
3937 @code{@value{AS}} maintains a counter of how many macros it has
3938 executed in this pseudo-variable; you can copy that number to your
3939 output with @samp{\@@}, but @emph{only within a macro definition}.
3942 @item LOCAL @var{name} [ , @dots{} ]
3943 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
3944 macro syntax'' with @samp{-a} or @samp{--alternate}.} @xref{Alternate,,
3945 Alternate macro syntax}.
3947 Generate a string replacement for each of the @var{name} arguments, and
3948 replace any instances of @var{name} in each macro expansion. The
3949 replacement string is unique in the assembly, and different for each
3950 separate macro expansion. @code{LOCAL} allows you to write macros that
3951 define symbols, without fear of conflict between separate macro expansions.
3956 @section @code{.nolist}
3958 @cindex @code{nolist} directive
3959 @cindex listing control, turning off
3960 Control (in conjunction with the @code{.list} directive) whether or
3961 not assembly listings are generated. These two directives maintain an
3962 internal counter (which is zero initially). @code{.list} increments the
3963 counter, and @code{.nolist} decrements it. Assembly listings are
3964 generated whenever the counter is greater than zero.
3967 @section @code{.octa @var{bignums}}
3969 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
3970 @cindex @code{octa} directive
3971 @cindex integer, 16-byte
3972 @cindex sixteen byte integer
3973 This directive expects zero or more bignums, separated by commas. For each
3974 bignum, it emits a 16-byte integer.
3976 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
3977 hence @emph{octa}-word for 16 bytes.
3980 @section @code{.org @var{new-lc} , @var{fill}}
3982 @cindex @code{org} directive
3983 @cindex location counter, advancing
3984 @cindex advancing location counter
3985 @cindex current address, advancing
3986 Advance the location counter of the current section to
3987 @var{new-lc}. @var{new-lc} is either an absolute expression or an
3988 expression with the same section as the current subsection. That is,
3989 you can't use @code{.org} to cross sections: if @var{new-lc} has the
3990 wrong section, the @code{.org} directive is ignored. To be compatible
3991 with former assemblers, if the section of @var{new-lc} is absolute,
3992 @code{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
3993 is the same as the current subsection.
3995 @code{.org} may only increase the location counter, or leave it
3996 unchanged; you cannot use @code{.org} to move the location counter
3999 @c double negative used below "not undefined" because this is a specific
4000 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
4001 @c section. doc@cygnus.com 18feb91
4002 Because @code{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
4003 may not be undefined. If you really detest this restriction we eagerly await
4004 a chance to share your improved assembler.
4006 Beware that the origin is relative to the start of the section, not
4007 to the start of the subsection. This is compatible with other
4008 people's assemblers.
4010 When the location counter (of the current subsection) is advanced, the
4011 intervening bytes are filled with @var{fill} which should be an
4012 absolute expression. If the comma and @var{fill} are omitted,
4013 @var{fill} defaults to zero.
4016 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4018 @cindex padding the location counter given a power of two
4019 @cindex @code{p2align} directive
4020 Pad the location counter (in the current subsection) to a particular
4021 storage boundary. The first expression (which must be absolute) is the
4022 number of low-order zero bits the location counter must have after
4023 advancement. For example @samp{.p2align 3} advances the location
4024 counter until it a multiple of 8. If the location counter is already a
4025 multiple of 8, no change is needed.
4027 The second expression (also absolute) gives the fill value to be stored in the
4028 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4029 padding bytes are normally zero. However, on some systems, if the section is
4030 marked as containing code and the fill value is omitted, the space is filled
4031 with no-op instructions.
4033 The third expression is also absolute, and is also optional. If it is present,
4034 it is the maximum number of bytes that should be skipped by this alignment
4035 directive. If doing the alignment would require skipping more bytes than the
4036 specified maximum, then the alignment is not done at all. You can omit the
4037 fill value (the second argument) entirely by simply using two commas after the
4038 required alignment; this can be useful if you want the alignment to be filled
4039 with no-op instructions when appropriate.
4041 @cindex @code{p2alignw} directive
4042 @cindex @code{p2alignl} directive
4043 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
4044 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
4045 pattern as a two byte word value. The @code{.p2alignl} directives treats the
4046 fill pattern as a four byte longword value. For example, @code{.p2alignw
4047 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4048 filled in with the value 0x368d (the exact placement of the bytes depends upon
4049 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4053 @section @code{.psize @var{lines} , @var{columns}}
4055 @cindex @code{psize} directive
4056 @cindex listing control: paper size
4057 @cindex paper size, for listings
4058 Use this directive to declare the number of lines---and, optionally, the
4059 number of columns---to use for each page, when generating listings.
4061 If you do not use @code{.psize}, listings use a default line-count
4062 of 60. You may omit the comma and @var{columns} specification; the
4063 default width is 200 columns.
4065 @code{@value{AS}} generates formfeeds whenever the specified number of
4066 lines is exceeded (or whenever you explicitly request one, using
4069 If you specify @var{lines} as @code{0}, no formfeeds are generated save
4070 those explicitly specified with @code{.eject}.
4073 @section @code{.quad @var{bignums}}
4075 @cindex @code{quad} directive
4076 @code{.quad} expects zero or more bignums, separated by commas. For
4077 each bignum, it emits
4079 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
4080 warning message; and just takes the lowest order 8 bytes of the bignum.
4081 @cindex eight-byte integer
4082 @cindex integer, 8-byte
4084 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
4085 hence @emph{quad}-word for 8 bytes.
4088 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
4089 warning message; and just takes the lowest order 16 bytes of the bignum.
4090 @cindex sixteen-byte integer
4091 @cindex integer, 16-byte
4095 @section @code{.rept @var{count}}
4097 @cindex @code{rept} directive
4098 Repeat the sequence of lines between the @code{.rept} directive and the next
4099 @code{.endr} directive @var{count} times.
4101 For example, assembling
4109 is equivalent to assembling
4118 @section @code{.sbttl "@var{subheading}"}
4120 @cindex @code{sbttl} directive
4121 @cindex subtitles for listings
4122 @cindex listing control: subtitle
4123 Use @var{subheading} as the title (third line, immediately after the
4124 title line) when generating assembly listings.
4126 This directive affects subsequent pages, as well as the current page if
4127 it appears within ten lines of the top of a page.
4131 @section @code{.scl @var{class}}
4133 @cindex @code{scl} directive
4134 @cindex symbol storage class (COFF)
4135 @cindex COFF symbol storage class
4136 Set the storage-class value for a symbol. This directive may only be
4137 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
4138 whether a symbol is static or external, or it may record further
4139 symbolic debugging information.
4142 The @samp{.scl} directive is primarily associated with COFF output; when
4143 configured to generate @code{b.out} output format, @code{@value{AS}}
4144 accepts this directive but ignores it.
4149 @section @code{.section @var{name}}
4151 @cindex @code{section} directive
4152 @cindex named section
4153 Use the @code{.section} directive to assemble the following code into a section
4156 This directive is only supported for targets that actually support arbitrarily
4157 named sections; on @code{a.out} targets, for example, it is not accepted, even
4158 with a standard @code{a.out} section name.
4161 For COFF targets, the @code{.section} directive is used in one of the following
4164 .section @var{name}[, "@var{flags}"]
4165 .section @var{name}[, @var{subsegment}]
4168 If the optional argument is quoted, it is taken as flags to use for the
4169 section. Each flag is a single character. The following flags are recognized:
4172 bss section (uninitialized data)
4174 section is not loaded
4185 If no flags are specified, the default flags depend upon the section name. If
4186 the section name is not recognized, the default will be for the section to be
4187 loaded and writable.
4189 If the optional argument to the @code{.section} directive is not quoted, it is
4190 taken as a subsegment number (@pxref{Sub-Sections}).
4194 For ELF targets, the @code{.section} directive is used like this:
4196 .section @var{name}[, "@var{flags}"[, @@@var{type}]]
4198 The optional @var{flags} argument is a quoted string which may contain any
4199 combintion of the following characters:
4202 section is allocatable
4206 section is executable
4209 The optional @var{type} argument may contain one of the following constants:
4212 section contains data
4214 section does not contain data (i.e., section only occupies space)
4217 If no flags are specified, the default flags depend upon the section name. If
4218 the section name is not recognized, the default will be for the section to have
4219 none of the above flags: it will not be allocated in memory, nor writable, nor
4220 executable. The section will contain data.
4222 For ELF targets, the assembler supports another type of @code{.section}
4223 directive for compatibility with the Solaris assembler:
4225 .section "@var{name}"[, @var{flags}...]
4227 Note that the section name is quoted. There may be a sequence of comma
4231 section is allocatable
4235 section is executable
4240 @section @code{.set @var{symbol}, @var{expression}}
4242 @cindex @code{set} directive
4243 @cindex symbol value, setting
4244 Set the value of @var{symbol} to @var{expression}. This
4245 changes @var{symbol}'s value and type to conform to
4246 @var{expression}. If @var{symbol} was flagged as external, it remains
4247 flagged (@pxref{Symbol Attributes}).
4249 You may @code{.set} a symbol many times in the same assembly.
4251 If you @code{.set} a global symbol, the value stored in the object
4252 file is the last value stored into it.
4255 The syntax for @code{set} on the HPPA is
4256 @samp{@var{symbol} .set @var{expression}}.
4260 @section @code{.short @var{expressions}}
4262 @cindex @code{short} directive
4264 @code{.short} is normally the same as @samp{.word}.
4265 @xref{Word,,@code{.word}}.
4267 In some configurations, however, @code{.short} and @code{.word} generate
4268 numbers of different lengths; @pxref{Machine Dependencies}.
4272 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
4275 This expects zero or more @var{expressions}, and emits
4276 a 16 bit number for each.
4281 @section @code{.single @var{flonums}}
4283 @cindex @code{single} directive
4284 @cindex floating point numbers (single)
4285 This directive assembles zero or more flonums, separated by commas. It
4286 has the same effect as @code{.float}.
4288 The exact kind of floating point numbers emitted depends on how
4289 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
4293 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
4294 numbers in @sc{ieee} format.
4300 @section @code{.size}
4302 @cindex @code{size} directive
4303 This directive is generated by compilers to include auxiliary debugging
4304 information in the symbol table. It is only permitted inside
4305 @code{.def}/@code{.endef} pairs.
4308 @samp{.size} is only meaningful when generating COFF format output; when
4309 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4315 @section @code{.sleb128 @var{expressions}}
4317 @cindex @code{sleb128} directive
4318 @var{sleb128} stands for ``signed little endian base 128.'' This is a
4319 compact, variable length representation of numbers used by the DWARF
4320 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
4322 @ifclear no-space-dir
4324 @section @code{.skip @var{size} , @var{fill}}
4326 @cindex @code{skip} directive
4327 @cindex filling memory
4328 This directive emits @var{size} bytes, each of value @var{fill}. Both
4329 @var{size} and @var{fill} are absolute expressions. If the comma and
4330 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
4334 @section @code{.space @var{size} , @var{fill}}
4336 @cindex @code{space} directive
4337 @cindex filling memory
4338 This directive emits @var{size} bytes, each of value @var{fill}. Both
4339 @var{size} and @var{fill} are absolute expressions. If the comma
4340 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
4345 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
4346 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
4347 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
4348 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
4357 @section @code{.space}
4358 @cindex @code{space} directive
4360 On the AMD 29K, this directive is ignored; it is accepted for
4361 compatibility with other AMD 29K assemblers.
4364 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
4365 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
4371 @section @code{.stabd, .stabn, .stabs}
4373 @cindex symbolic debuggers, information for
4374 @cindex @code{stab@var{x}} directives
4375 There are three directives that begin @samp{.stab}.
4376 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
4377 The symbols are not entered in the @code{@value{AS}} hash table: they
4378 cannot be referenced elsewhere in the source file.
4379 Up to five fields are required:
4383 This is the symbol's name. It may contain any character except
4384 @samp{\000}, so is more general than ordinary symbol names. Some
4385 debuggers used to code arbitrarily complex structures into symbol names
4389 An absolute expression. The symbol's type is set to the low 8 bits of
4390 this expression. Any bit pattern is permitted, but @code{@value{LD}}
4391 and debuggers choke on silly bit patterns.
4394 An absolute expression. The symbol's ``other'' attribute is set to the
4395 low 8 bits of this expression.
4398 An absolute expression. The symbol's descriptor is set to the low 16
4399 bits of this expression.
4402 An absolute expression which becomes the symbol's value.
4405 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
4406 or @code{.stabs} statement, the symbol has probably already been created;
4407 you get a half-formed symbol in your object file. This is
4408 compatible with earlier assemblers!
4411 @cindex @code{stabd} directive
4412 @item .stabd @var{type} , @var{other} , @var{desc}
4414 The ``name'' of the symbol generated is not even an empty string.
4415 It is a null pointer, for compatibility. Older assemblers used a
4416 null pointer so they didn't waste space in object files with empty
4419 The symbol's value is set to the location counter,
4420 relocatably. When your program is linked, the value of this symbol
4421 is the address of the location counter when the @code{.stabd} was
4424 @cindex @code{stabn} directive
4425 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
4426 The name of the symbol is set to the empty string @code{""}.
4428 @cindex @code{stabs} directive
4429 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
4430 All five fields are specified.
4436 @section @code{.string} "@var{str}"
4438 @cindex string, copying to object file
4439 @cindex @code{string} directive
4441 Copy the characters in @var{str} to the object file. You may specify more than
4442 one string to copy, separated by commas. Unless otherwise specified for a
4443 particular machine, the assembler marks the end of each string with a 0 byte.
4444 You can use any of the escape sequences described in @ref{Strings,,Strings}.
4448 @section @code{.symver}
4449 @cindex @code{symver} directive
4450 @cindex symbol versioning
4451 @cindex versions of symbols
4452 Use the @code{.symver} directive to bind symbols to specific version nodes
4453 within a source file. This is only supported on ELF platforms, and is
4454 typically used when assembling files to be linked into a shared library.
4455 There are cases where it may make sense to use this in objects to be bound
4456 into an application itself so as to override a versioned symbol from a
4459 For ELF targets, the @code{.symver} directive is used like this:
4461 .symver @var{name}, @var{name2@@nodename}
4463 In this case, the symbol @var{name} must exist and be defined within the file
4464 being assembled. The @code{.versym} directive effectively creates a symbol
4465 alias with the name @var{name2@@nodename}, and in fact the main reason that we
4466 just don't try and create a regular alias is that the @var{@@} character isn't
4467 permitted in symbol names. The @var{name2} part of the name is the actual name
4468 of the symbol by which it will be externally referenced. The name @var{name}
4469 itself is merely a name of convenience that is used so that it is possible to
4470 have definitions for multiple versions of a function within a single source
4471 file, and so that the compiler can unambiguously know which version of a
4472 function is being mentioned. The @var{nodename} portion of the alias should be
4473 the name of a node specified in the version script supplied to the linker when
4474 building a shared library. If you are attempting to override a versioned
4475 symbol from a shared library, then @var{nodename} should correspond to the
4476 nodename of the symbol you are trying to override.
4481 @section @code{.tag @var{structname}}
4483 @cindex COFF structure debugging
4484 @cindex structure debugging, COFF
4485 @cindex @code{tag} directive
4486 This directive is generated by compilers to include auxiliary debugging
4487 information in the symbol table. It is only permitted inside
4488 @code{.def}/@code{.endef} pairs. Tags are used to link structure
4489 definitions in the symbol table with instances of those structures.
4492 @samp{.tag} is only used when generating COFF format output; when
4493 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4499 @section @code{.text @var{subsection}}
4501 @cindex @code{text} directive
4502 Tells @code{@value{AS}} to assemble the following statements onto the end of
4503 the text subsection numbered @var{subsection}, which is an absolute
4504 expression. If @var{subsection} is omitted, subsection number zero
4508 @section @code{.title "@var{heading}"}
4510 @cindex @code{title} directive
4511 @cindex listing control: title line
4512 Use @var{heading} as the title (second line, immediately after the
4513 source file name and pagenumber) when generating assembly listings.
4515 This directive affects subsequent pages, as well as the current page if
4516 it appears within ten lines of the top of a page.
4520 @section @code{.type @var{int}}
4522 @cindex COFF symbol type
4523 @cindex symbol type, COFF
4524 @cindex @code{type} directive
4525 This directive, permitted only within @code{.def}/@code{.endef} pairs,
4526 records the integer @var{int} as the type attribute of a symbol table entry.
4529 @samp{.type} is associated only with COFF format output; when
4530 @code{@value{AS}} is configured for @code{b.out} output, it accepts this
4531 directive but ignores it.
4537 @section @code{.val @var{addr}}
4539 @cindex @code{val} directive
4540 @cindex COFF value attribute
4541 @cindex value attribute, COFF
4542 This directive, permitted only within @code{.def}/@code{.endef} pairs,
4543 records the address @var{addr} as the value attribute of a symbol table
4547 @samp{.val} is used only for COFF output; when @code{@value{AS}} is
4548 configured for @code{b.out}, it accepts this directive but ignores it.
4553 @section @code{.uleb128 @var{expressions}}
4555 @cindex @code{uleb128} directive
4556 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
4557 compact, variable length representation of numbers used by the DWARF
4558 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
4561 @section @code{.word @var{expressions}}
4563 @cindex @code{word} directive
4564 This directive expects zero or more @var{expressions}, of any section,
4565 separated by commas.
4568 For each expression, @code{@value{AS}} emits a 32-bit number.
4571 For each expression, @code{@value{AS}} emits a 16-bit number.
4576 The size of the number emitted, and its byte order,
4577 depend on what target computer the assembly is for.
4580 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
4581 @c happen---32-bit addressability, period; no long/short jumps.
4582 @ifset DIFF-TBL-KLUGE
4583 @cindex difference tables altered
4584 @cindex altered difference tables
4586 @emph{Warning: Special Treatment to support Compilers}
4590 Machines with a 32-bit address space, but that do less than 32-bit
4591 addressing, require the following special treatment. If the machine of
4592 interest to you does 32-bit addressing (or doesn't require it;
4593 @pxref{Machine Dependencies}), you can ignore this issue.
4596 In order to assemble compiler output into something that works,
4597 @code{@value{AS}} occasionlly does strange things to @samp{.word} directives.
4598 Directives of the form @samp{.word sym1-sym2} are often emitted by
4599 compilers as part of jump tables. Therefore, when @code{@value{AS}} assembles a
4600 directive of the form @samp{.word sym1-sym2}, and the difference between
4601 @code{sym1} and @code{sym2} does not fit in 16 bits, @code{@value{AS}}
4602 creates a @dfn{secondary jump table}, immediately before the next label.
4603 This secondary jump table is preceded by a short-jump to the
4604 first byte after the secondary table. This short-jump prevents the flow
4605 of control from accidentally falling into the new table. Inside the
4606 table is a long-jump to @code{sym2}. The original @samp{.word}
4607 contains @code{sym1} minus the address of the long-jump to
4610 If there were several occurrences of @samp{.word sym1-sym2} before the
4611 secondary jump table, all of them are adjusted. If there was a
4612 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
4613 long-jump to @code{sym4} is included in the secondary jump table,
4614 and the @code{.word} directives are adjusted to contain @code{sym3}
4615 minus the address of the long-jump to @code{sym4}; and so on, for as many
4616 entries in the original jump table as necessary.
4619 @emph{This feature may be disabled by compiling @code{@value{AS}} with the
4620 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
4621 assembly language programmers.
4624 @c end DIFF-TBL-KLUGE
4627 @section Deprecated Directives
4629 @cindex deprecated directives
4630 @cindex obsolescent directives
4631 One day these directives won't work.
4632 They are included for compatibility with older assemblers.
4640 @node Machine Dependencies
4641 @chapter Machine Dependent Features
4643 @cindex machine dependencies
4644 The machine instruction sets are (almost by definition) different on
4645 each machine where @code{@value{AS}} runs. Floating point representations
4646 vary as well, and @code{@value{AS}} often supports a few additional
4647 directives or command-line options for compatibility with other
4648 assemblers on a particular platform. Finally, some versions of
4649 @code{@value{AS}} support special pseudo-instructions for branch
4652 This chapter discusses most of these differences, though it does not
4653 include details on any machine's instruction set. For details on that
4654 subject, see the hardware manufacturer's manual.
4658 * AMD29K-Dependent:: AMD 29K Dependent Features
4661 * ARC-Dependent:: ARC Dependent Features
4664 * ARM-Dependent:: ARM Dependent Features
4667 * D10V-Dependent:: D10V Dependent Features
4670 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
4673 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
4676 * HPPA-Dependent:: HPPA Dependent Features
4679 * i386-Dependent:: Intel 80386 Dependent Features
4682 * i960-Dependent:: Intel 80960 Dependent Features
4685 * M32R-Dependent:: M32R Dependent Features
4688 * M68K-Dependent:: M680x0 Dependent Features
4691 * MIPS-Dependent:: MIPS Dependent Features
4694 * SH-Dependent:: Hitachi SH Dependent Features
4697 * Sparc-Dependent:: SPARC Dependent Features
4700 * V850-Dependent:: V850 Dependent Features
4703 * Z8000-Dependent:: Z8000 Dependent Features
4706 * Vax-Dependent:: VAX Dependent Features
4713 @c The following major nodes are *sections* in the GENERIC version, *chapters*
4714 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
4715 @c peculiarity: to preserve cross-references, there must be a node called
4716 @c "Machine Dependencies". Hence the conditional nodenames in each
4717 @c major node below. Node defaulting in makeinfo requires adjacency of
4718 @c node and sectioning commands; hence the repetition of @chapter BLAH
4719 @c in both conditional blocks.
4725 @chapter ARC Dependent Features
4728 @node Machine Dependencies
4729 @chapter ARC Dependent Features
4734 * ARC-Opts:: Options
4735 * ARC-Float:: Floating Point
4736 * ARC-Directives:: Sparc Machine Directives
4742 @cindex options for ARC
4744 @cindex architectures, ARC
4745 @cindex ARC architectures
4746 The ARC chip family includes several successive levels (or other
4747 variants) of chip, using the same core instruction set, but including
4748 a few additional instructions at each level.
4750 By default, @code{@value{AS}} assumes the core instruction set (ARC
4751 base). The @code{.cpu} pseudo-op is intended to be used to select
4755 @cindex @code{-mbig-endian} option (ARC)
4756 @cindex @code{-mlittle-endian} option (ARC)
4757 @cindex ARC big-endian output
4758 @cindex ARC little-endian output
4759 @cindex big-endian output, ARC
4760 @cindex little-endian output, ARC
4762 @itemx -mlittle-endian
4763 Any @sc{arc} configuration of @code{@value{AS}} can select big-endian or
4764 little-endian output at run time (unlike most other @sc{gnu} development
4765 tools, which must be configured for one or the other). Use
4766 @samp{-mbig-endian} to select big-endian output, and @samp{-mlittle-endian}
4771 @section Floating Point
4773 @cindex floating point, ARC (@sc{ieee})
4774 @cindex ARC floating point (@sc{ieee})
4775 The ARC cpu family currently does not have hardware floating point
4776 support. Software floating point support is provided by @code{GCC}
4777 and uses @sc{ieee} floating-point numbers.
4779 @node ARC-Directives
4780 @section ARC Machine Directives
4782 @cindex ARC machine directives
4783 @cindex machine directives, ARC
4784 The ARC version of @code{@value{AS}} supports the following additional
4789 @cindex @code{cpu} directive, SPARC
4790 This must be followed by the desired cpu.
4791 The ARC is intended to be customizable, @code{.cpu} is used to
4792 select the desired variant [though currently there are none].
4799 @include c-a29k.texi
4808 @node Machine Dependencies
4809 @chapter Machine Dependent Features
4811 The machine instruction sets are different on each Hitachi chip family,
4812 and there are also some syntax differences among the families. This
4813 chapter describes the specific @code{@value{AS}} features for each
4817 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
4818 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
4819 * SH-Dependent:: Hitachi SH Dependent Features
4826 @include c-d10v.texi
4830 @include c-h8300.texi
4834 @include c-h8500.texi
4838 @include c-hppa.texi
4842 @include c-i386.texi
4846 @include c-i960.texi
4850 @include c-m32r.texi
4854 @include c-m68k.texi
4858 @include c-mips.texi
4862 @include c-ns32k.texi
4870 @include c-sparc.texi
4882 @include c-v850.texi
4886 @c reverse effect of @down at top of generic Machine-Dep chapter
4890 @node Reporting Bugs
4891 @chapter Reporting Bugs
4892 @cindex bugs in assembler
4893 @cindex reporting bugs in assembler
4895 Your bug reports play an essential role in making @code{@value{AS}} reliable.
4897 Reporting a bug may help you by bringing a solution to your problem, or it may
4898 not. But in any case the principal function of a bug report is to help the
4899 entire community by making the next version of @code{@value{AS}} work better.
4900 Bug reports are your contribution to the maintenance of @code{@value{AS}}.
4902 In order for a bug report to serve its purpose, you must include the
4903 information that enables us to fix the bug.
4906 * Bug Criteria:: Have you found a bug?
4907 * Bug Reporting:: How to report bugs
4911 @section Have you found a bug?
4912 @cindex bug criteria
4914 If you are not sure whether you have found a bug, here are some guidelines:
4917 @cindex fatal signal
4918 @cindex assembler crash
4919 @cindex crash of assembler
4921 If the assembler gets a fatal signal, for any input whatever, that is a
4922 @code{@value{AS}} bug. Reliable assemblers never crash.
4924 @cindex error on valid input
4926 If @code{@value{AS}} produces an error message for valid input, that is a bug.
4928 @cindex invalid input
4930 If @code{@value{AS}} does not produce an error message for invalid input, that
4931 is a bug. However, you should note that your idea of ``invalid input'' might
4932 be our idea of ``an extension'' or ``support for traditional practice''.
4935 If you are an experienced user of assemblers, your suggestions for improvement
4936 of @code{@value{AS}} are welcome in any case.
4940 @section How to report bugs
4942 @cindex assembler bugs, reporting
4944 A number of companies and individuals offer support for @sc{gnu} products. If
4945 you obtained @code{@value{AS}} from a support organization, we recommend you
4946 contact that organization first.
4948 You can find contact information for many support companies and
4949 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
4952 In any event, we also recommend that you send bug reports for @code{@value{AS}}
4953 to @samp{bug-gnu-utils@@prep.ai.mit.edu}.
4955 The fundamental principle of reporting bugs usefully is this:
4956 @strong{report all the facts}. If you are not sure whether to state a
4957 fact or leave it out, state it!
4959 Often people omit facts because they think they know what causes the problem
4960 and assume that some details do not matter. Thus, you might assume that the
4961 name of a symbol you use in an example does not matter. Well, probably it does
4962 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
4963 happens to fetch from the location where that name is stored in memory;
4964 perhaps, if the name were different, the contents of that location would fool
4965 the assembler into doing the right thing despite the bug. Play it safe and
4966 give a specific, complete example. That is the easiest thing for you to do,
4967 and the most helpful.
4969 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
4970 it is new to us. Therefore, always write your bug reports on the assumption
4971 that the bug has not been reported previously.
4973 Sometimes people give a few sketchy facts and ask, ``Does this ring a
4974 bell?'' Those bug reports are useless, and we urge everyone to
4975 @emph{refuse to respond to them} except to chide the sender to report
4978 To enable us to fix the bug, you should include all these things:
4982 The version of @code{@value{AS}}. @code{@value{AS}} announces it if you start
4983 it with the @samp{--version} argument.
4985 Without this, we will not know whether there is any point in looking for
4986 the bug in the current version of @code{@value{AS}}.
4989 Any patches you may have applied to the @code{@value{AS}} source.
4992 The type of machine you are using, and the operating system name and
4996 What compiler (and its version) was used to compile @code{@value{AS}}---e.g.
5000 The command arguments you gave the assembler to assemble your example and
5001 observe the bug. To guarantee you will not omit something important, list them
5002 all. A copy of the Makefile (or the output from make) is sufficient.
5004 If we were to try to guess the arguments, we would probably guess wrong
5005 and then we might not encounter the bug.
5008 A complete input file that will reproduce the bug. If the bug is observed when
5009 the assembler is invoked via a compiler, send the assembler source, not the
5010 high level language source. Most compilers will produce the assembler source
5011 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
5012 the options @samp{-v --save-temps}; this will save the assembler source in a
5013 file with an extension of @file{.s}, and also show you exactly how
5014 @code{@value{AS}} is being run.
5017 A description of what behavior you observe that you believe is
5018 incorrect. For example, ``It gets a fatal signal.''
5020 Of course, if the bug is that @code{@value{AS}} gets a fatal signal, then we
5021 will certainly notice it. But if the bug is incorrect output, we might not
5022 notice unless it is glaringly wrong. You might as well not give us a chance to
5025 Even if the problem you experience is a fatal signal, you should still say so
5026 explicitly. Suppose something strange is going on, such as, your copy of
5027 @code{@value{AS}} is out of synch, or you have encountered a bug in the C
5028 library on your system. (This has happened!) Your copy might crash and ours
5029 would not. If you told us to expect a crash, then when ours fails to crash, we
5030 would know that the bug was not happening for us. If you had not told us to
5031 expect a crash, then we would not be able to draw any conclusion from our
5035 If you wish to suggest changes to the @code{@value{AS}} source, send us context
5036 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
5037 option. Always send diffs from the old file to the new file. If you even
5038 discuss something in the @code{@value{AS}} source, refer to it by context, not
5041 The line numbers in our development sources will not match those in your
5042 sources. Your line numbers would convey no useful information to us.
5045 Here are some things that are not necessary:
5049 A description of the envelope of the bug.
5051 Often people who encounter a bug spend a lot of time investigating
5052 which changes to the input file will make the bug go away and which
5053 changes will not affect it.
5055 This is often time consuming and not very useful, because the way we
5056 will find the bug is by running a single example under the debugger
5057 with breakpoints, not by pure deduction from a series of examples.
5058 We recommend that you save your time for something else.
5060 Of course, if you can find a simpler example to report @emph{instead}
5061 of the original one, that is a convenience for us. Errors in the
5062 output will be easier to spot, running under the debugger will take
5063 less time, and so on.
5065 However, simplification is not vital; if you do not want to do this,
5066 report the bug anyway and send us the entire test case you used.
5069 A patch for the bug.
5071 A patch for the bug does help us if it is a good one. But do not omit
5072 the necessary information, such as the test case, on the assumption that
5073 a patch is all we need. We might see problems with your patch and decide
5074 to fix the problem another way, or we might not understand it at all.
5076 Sometimes with a program as complicated as @code{@value{AS}} it is very hard to
5077 construct an example that will make the program follow a certain path through
5078 the code. If you do not send us the example, we will not be able to construct
5079 one, so we will not be able to verify that the bug is fixed.
5081 And if we cannot understand what bug you are trying to fix, or why your
5082 patch should be an improvement, we will not install it. A test case will
5083 help us to understand.
5086 A guess about what the bug is or what it depends on.
5088 Such guesses are usually wrong. Even we cannot guess right about such
5089 things without first using the debugger to find the facts.
5092 @node Acknowledgements
5093 @chapter Acknowledgements
5095 If you have contributed to @code{@value{AS}} and your name isn't listed here,
5096 it is not meant as a slight. We just don't know about it. Send mail to the
5097 maintainer, and we'll correct the situation. Currently
5099 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
5101 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
5104 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
5105 information and the 68k series machines, most of the preprocessing pass, and
5106 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
5108 K. Richard Pixley maintained GAS for a while, adding various enhancements and
5109 many bug fixes, including merging support for several processors, breaking GAS
5110 up to handle multiple object file format back ends (including heavy rewrite,
5111 testing, an integration of the coff and b.out back ends), adding configuration
5112 including heavy testing and verification of cross assemblers and file splits
5113 and renaming, converted GAS to strictly ANSI C including full prototypes, added
5114 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
5115 port (including considerable amounts of reverse engineering), a SPARC opcode
5116 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
5117 assertions and made them work, much other reorganization, cleanup, and lint.
5119 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
5120 in format-specific I/O modules.
5122 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
5123 has done much work with it since.
5125 The Intel 80386 machine description was written by Eliot Dresselhaus.
5127 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
5129 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
5130 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
5132 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
5133 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
5134 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
5135 support a.out format.
5137 Support for the Zilog Z8k and Hitachi H8/300 and H8/500 processors (tc-z8k,
5138 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
5139 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
5140 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
5143 John Gilmore built the AMD 29000 support, added @code{.include} support, and
5144 simplified the configuration of which versions accept which directives. He
5145 updated the 68k machine description so that Motorola's opcodes always produced
5146 fixed-size instructions (e.g. @code{jsr}), while synthetic instructions
5147 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
5148 cross-compilation support, and one bug in relaxation that took a week and
5149 required the proverbial one-bit fix.
5151 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
5152 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
5153 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
5154 PowerPC assembler, and made a few other minor patches.
5156 Steve Chamberlain made @code{@value{AS}} able to generate listings.
5158 Hewlett-Packard contributed support for the HP9000/300.
5160 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
5161 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
5162 formats). This work was supported by both the Center for Software Science at
5163 the University of Utah and Cygnus Support.
5165 Support for ELF format files has been worked on by Mark Eichin of Cygnus
5166 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
5167 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
5168 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
5169 and some initial 64-bit support).
5171 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
5172 support for openVMS/Alpha.
5174 Several engineers at Cygnus Support have also provided many small bug fixes and
5175 configuration enhancements.
5177 Many others have contributed large or small bugfixes and enhancements. If
5178 you have contributed significant work and are not mentioned on this list, and
5179 want to be, let us know. Some of the history has been lost; we are not
5180 intentionally leaving anyone out.