1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (c) 1991, 92, 93, 94, 95, 1996 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
36 @set abnormal-separator
40 @settitle Using @value{AS}
43 @settitle Using @value{AS} (@value{TARGET})
45 @setchapternewpage odd
50 @c WARE! Some of the machine-dependent sections contain tables of machine
51 @c instructions. Except in multi-column format, these tables look silly.
52 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
53 @c the multi-col format is faked within @example sections.
55 @c Again unfortunately, the natural size that fits on a page, for these tables,
56 @c is different depending on whether or not smallbook is turned on.
57 @c This matters, because of order: text flow switches columns at each page
60 @c The format faked in this source works reasonably well for smallbook,
61 @c not well for the default large-page format. This manual expects that if you
62 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
63 @c tables in question. You can turn on one without the other at your
64 @c discretion, of course.
67 @c the insn tables look just as silly in info files regardless of smallbook,
68 @c might as well show 'em anyways.
74 * As: (as). The GNU assembler.
83 This file documents the GNU Assembler "@value{AS}".
85 Copyright (C) 1991, 92, 93, 94, 95, 1996 Free Software Foundation, Inc.
87 Permission is granted to make and distribute verbatim copies of
88 this manual provided the copyright notice and this permission notice
89 are preserved on all copies.
92 Permission is granted to process this file through Tex and print the
93 results, provided the printed document carries copying permission
94 notice identical to this one except for the removal of this paragraph
95 (this paragraph not being relevant to the printed manual).
98 Permission is granted to copy and distribute modified versions of this manual
99 under the conditions for verbatim copying, provided that the entire resulting
100 derived work is distributed under the terms of a permission notice identical to
103 Permission is granted to copy and distribute translations of this manual
104 into another language, under the above conditions for modified versions.
108 @title Using @value{AS}
109 @subtitle The @sc{gnu} Assembler
111 @subtitle for the @value{TARGET} family
114 @subtitle January 1994
117 The Free Software Foundation Inc. thanks The Nice Computer
118 Company of Australia for loaning Dean Elsner to write the
119 first (Vax) version of @code{as} for Project @sc{gnu}.
120 The proprietors, management and staff of TNCCA thank FSF for
121 distracting the boss while they got some work
124 @author Dean Elsner, Jay Fenlason & friends
128 \hfill {\it Using {\tt @value{AS}}}\par
129 \hfill Edited by Cygnus Support\par
131 %"boxit" macro for figures:
132 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
133 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
134 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
135 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
136 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
139 @vskip 0pt plus 1filll
140 Copyright @copyright{} 1991, 92, 93, 94, 95, 1996 Free Software Foundation, Inc.
142 Permission is granted to make and distribute verbatim copies of
143 this manual provided the copyright notice and this permission notice
144 are preserved on all copies.
146 Permission is granted to copy and distribute modified versions of this manual
147 under the conditions for verbatim copying, provided that the entire resulting
148 derived work is distributed under the terms of a permission notice identical to
151 Permission is granted to copy and distribute translations of this manual
152 into another language, under the above conditions for modified versions.
157 @top Using @value{AS}
159 This file is a user guide to the @sc{gnu} assembler @code{@value{AS}}.
161 This version of the file describes @code{@value{AS}} configured to generate
162 code for @value{TARGET} architectures.
165 * Overview:: Overview
166 * Invoking:: Command-Line Options
168 * Sections:: Sections and Relocation
170 * Expressions:: Expressions
171 * Pseudo Ops:: Assembler Directives
172 * Machine Dependencies:: Machine Dependent Features
173 * Reporting Bugs:: Reporting Bugs
174 * Acknowledgements:: Who Did What
182 This manual is a user guide to the @sc{gnu} assembler @code{@value{AS}}.
184 This version of the manual describes @code{@value{AS}} configured to generate
185 code for @value{TARGET} architectures.
189 @cindex invocation summary
190 @cindex option summary
191 @cindex summary of options
192 Here is a brief summary of how to invoke @code{@value{AS}}. For details,
193 @pxref{Invoking,,Comand-Line Options}.
195 @c We don't use deffn and friends for the following because they seem
196 @c to be limited to one line for the header.
198 @value{AS} [ -a[dhlns][=file] ] [ -D ] [ --defsym @var{sym}=@var{val} ]
199 [ -f ] [ --help ] [ -I @var{dir} ] [ -J ] [ -K ] [ -L ]
200 [ -o @var{objfile} ] [ -R ] [ --statistics ] [ -v ] [ -version ]
201 [ --version ] [ -W ] [ -w ] [ -x ] [ -Z ]
203 @c am29k has no machine-dependent assembler options
205 @c start-sanitize-arc
207 [ -mbig-endian | -mlittle-endian ]
210 @c start-sanitize-d10v
217 @c Hitachi family chips have no machine-dependent assembler options
220 @c HPPA has no machine-dependent assembler options (yet).
223 @c The order here is important. See c-sparc.texi.
224 [ -Av6 | -Av7 | -Av8 | -Asparclite | -Av9 | -Av9a ]
225 [ -xarch=v8plus | -xarch=v8plusa ] [ -bump ]
228 @c Z8000 has no machine-dependent assembler options
231 @c see md_parse_option in tc-i960.c
232 [ -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC ]
236 [ -l ] [ -m68000 | -m68010 | -m68020 | ... ]
239 [ -nocpp ] [ -EL ] [ -EB ] [ -G @var{num} ] [ -mcpu=@var{CPU} ]
240 [ -mips1 ] [ -mips2 ] [ -mips3 ] [ -m4650 ] [ -no-m4650 ]
241 [ --trap ] [ --break ]
242 [ --emulation=@var{name} ]
244 [ -- | @var{files} @dots{} ]
249 Turn on listings, in any of a variety of ways:
253 omit debugging directives
256 include high-level source
262 omit forms processing
268 set the name of the listing file
271 You may combine these options; for example, use @samp{-aln} for assembly
272 listing without forms processing. The @samp{=file} option, if used, must be
273 the last one. By itself, @samp{-a} defaults to @samp{-ahls}---that is, all
277 Ignored. This option is accepted for script compatibility with calls to
280 @item --defsym @var{sym}=@var{value}
281 Define the symbol @var{sym} to be @var{value} before assembling the input file.
282 @var{value} must be an integer constant. As in C, a leading @samp{0x}
283 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
286 ``fast''---skip whitespace and comment preprocessing (assume source is
290 Print a summary of the command line options and exit.
293 Add directory @var{dir} to the search list for @code{.include} directives.
296 Don't warn about signed overflow.
299 @ifclear DIFF-TBL-KLUGE
300 This option is accepted but has no effect on the @value{TARGET} family.
302 @ifset DIFF-TBL-KLUGE
303 Issue warnings when difference tables altered for long displacements.
307 Keep (in the symbol table) local symbols, starting with @samp{L}.
309 @item -o @var{objfile}
310 Name the object-file output from @code{@value{AS}} @var{objfile}.
313 Fold the data section into the text section.
316 Print the maximum space (in bytes) and total time (in seconds) used by
321 Print the @code{as} version.
324 Print the @code{as} version and exit.
327 Suppress warning messages.
336 Generate an object file even after errors.
338 @item -- | @var{files} @dots{}
339 Standard input, or source files to assemble.
344 The following options are available when @value{AS} is configured for
349 @cindex ARC endianness
350 @cindex endianness, ARC
351 @cindex big endian output, ARC
353 Generate ``big endian'' format output.
355 @cindex little endian output, ARC
356 @item -mlittle-endian
357 Generate ``little endian'' format output.
362 @c start-sanitize-d10v
364 The following options are available when @value{AS} is configured for
367 @cindex D10V optimization
368 @cindex optimization, D10V
370 Optimize output by parallelizing instructions.
376 The following options are available when @value{AS} is configured for the
377 Intel 80960 processor.
380 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
381 Specify which variant of the 960 architecture is the target.
384 Add code to collect statistics about branches taken.
387 Do not alter compare-and-branch instructions for long displacements;
394 The following options are available when @value{AS} is configured for the
395 Motorola 68000 series.
400 Shorten references to undefined symbols, to one word instead of two.
402 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030 | -m68040 | -m68060
403 @itemx | -m68302 | -m68331 | -m68332 | -m68333 | -m68340 | -mcpu32 | -m5200
404 Specify what processor in the 68000 family is the target. The default
405 is normally the 68020, but this can be changed at configuration time.
407 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
408 The target machine does (or does not) have a floating-point coprocessor.
409 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
410 the basic 68000 is not compatible with the 68881, a combination of the
411 two can be specified, since it's possible to do emulation of the
412 coprocessor instructions with the main processor.
414 @item -m68851 | -mno-68851
415 The target machine does (or does not) have a memory-management
416 unit coprocessor. The default is to assume an MMU for 68020 and up.
422 The following options are available when @code{@value{AS}} is configured
423 for the SPARC architecture:
426 @item -Av6 | -Av7 | -Av8 | -Asparclite | -Av9 | -Av9a
427 Explicitly select a variant of the SPARC architecture.
429 @item -xarch=v8plus | -xarch=v8plusa
430 For compatibility with the Solaris v9 assembler. These options are
431 equivalent to -Av9 and -Av9a, respectively.
434 Warn when the assembler switches to another architecture.
439 The following options are available when @value{AS} is configured for
444 This option sets the largest size of an object that can be referenced
445 implicitly with the @code{gp} register. It is only accepted for targets that
446 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
448 @cindex MIPS endianness
449 @cindex endianness, MIPS
450 @cindex big endian output, MIPS
452 Generate ``big endian'' format output.
454 @cindex little endian output, MIPS
456 Generate ``little endian'' format output.
462 Generate code for a particular MIPS Instruction Set Architecture level.
463 @samp{-mips1} corresponds to the @sc{r2000} and @sc{r3000} processors,
464 @samp{-mips2} to the @sc{r6000} processor, and @samp{-mips3} to the @sc{r4000}
469 Generate code for the MIPS @sc{r4650} chip. This tells the assembler to accept
470 the @samp{mad} and @samp{madu} instruction, and to not schedule @samp{nop}
471 instructions around accesses to the @samp{HI} and @samp{LO} registers.
472 @samp{-no-m4650} turns off this option.
474 @item -mcpu=@var{CPU}
475 Generate code for a particular MIPS cpu. This has little effect on the
476 assembler, but it is passed by @code{@value{GCC}}.
479 @item --emulation=@var{name}
480 This option causes @code{@value{AS}} to emulated @code{@value{AS}} configured
481 for some other target, in all respects, including output format (choosing
482 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
483 debugging information or store symbol table information, and default
484 endianness. The available configuration names are: @samp{mipsecoff},
485 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
486 @samp{mipsbelf}. The first two do not alter the default endianness from that
487 of the primary target for which the assembler was configured; the others change
488 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
489 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
490 selection in any case.
492 This option is currently supported only when the primary target
493 @code{@value{AS}} is configured for is a MIPS ELF or ECOFF target.
494 Furthermore, the primary target or others specified with
495 @samp{--enable-targets=@dots{}} at configuration time must include support for
496 the other format, if both are to be available. For example, the Irix 5
497 configuration includes support for both.
499 Eventually, this option will support more configurations, with more
500 fine-grained control over the assembler's behavior, and will be supported for
504 @code{@value{AS}} ignores this option. It is accepted for compatibility with
512 Control how to deal with multiplication overflow and division by zero.
513 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
514 (and only work for Instruction Set Architecture level 2 and higher);
515 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
521 * Manual:: Structure of this Manual
522 * GNU Assembler:: @value{AS}, the GNU Assembler
523 * Object Formats:: Object File Formats
524 * Command Line:: Command Line
525 * Input Files:: Input Files
526 * Object:: Output (Object) File
527 * Errors:: Error and Warning Messages
531 @section Structure of this Manual
533 @cindex manual, structure and purpose
534 This manual is intended to describe what you need to know to use
535 @sc{gnu} @code{@value{AS}}. We cover the syntax expected in source files, including
536 notation for symbols, constants, and expressions; the directives that
537 @code{@value{AS}} understands; and of course how to invoke @code{@value{AS}}.
540 We also cover special features in the @value{TARGET}
541 configuration of @code{@value{AS}}, including assembler directives.
544 This manual also describes some of the machine-dependent features of
545 various flavors of the assembler.
548 @cindex machine instructions (not covered)
549 On the other hand, this manual is @emph{not} intended as an introduction
550 to programming in assembly language---let alone programming in general!
551 In a similar vein, we make no attempt to introduce the machine
552 architecture; we do @emph{not} describe the instruction set, standard
553 mnemonics, registers or addressing modes that are standard to a
554 particular architecture.
556 You may want to consult the manufacturer's
557 machine architecture manual for this information.
561 For information on the H8/300 machine instruction set, see @cite{H8/300
562 Series Programming Manual} (Hitachi ADE--602--025). For the H8/300H,
563 see @cite{H8/300H Series Programming Manual} (Hitachi).
566 For information on the H8/500 machine instruction set, see @cite{H8/500
567 Series Programming Manual} (Hitachi M21T001).
570 For information on the Hitachi SH machine instruction set, see
571 @cite{SH-Microcomputer User's Manual} (Hitachi Micro Systems, Inc.).
574 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
578 @c I think this is premature---doc@cygnus.com, 17jan1991
580 Throughout this manual, we assume that you are running @dfn{GNU},
581 the portable operating system from the @dfn{Free Software
582 Foundation, Inc.}. This restricts our attention to certain kinds of
583 computer (in particular, the kinds of computers that @sc{gnu} can run on);
584 once this assumption is granted examples and definitions need less
587 @code{@value{AS}} is part of a team of programs that turn a high-level
588 human-readable series of instructions into a low-level
589 computer-readable series of instructions. Different versions of
590 @code{@value{AS}} are used for different kinds of computer.
593 @c There used to be a section "Terminology" here, which defined
594 @c "contents", "byte", "word", and "long". Defining "word" to any
595 @c particular size is confusing when the .word directive may generate 16
596 @c bits on one machine and 32 bits on another; in general, for the user
597 @c version of this manual, none of these terms seem essential to define.
598 @c They were used very little even in the former draft of the manual;
599 @c this draft makes an effort to avoid them (except in names of
603 @section @value{AS}, the GNU Assembler
605 @sc{gnu} @code{as} is really a family of assemblers.
607 This manual describes @code{@value{AS}}, a member of that family which is
608 configured for the @value{TARGET} architectures.
610 If you use (or have used) the @sc{gnu} assembler on one architecture, you
611 should find a fairly similar environment when you use it on another
612 architecture. Each version has much in common with the others,
613 including object file formats, most assembler directives (often called
614 @dfn{pseudo-ops}) and assembler syntax.@refill
616 @cindex purpose of @sc{gnu} @code{@value{AS}}
617 @code{@value{AS}} is primarily intended to assemble the output of the
618 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
619 @code{@value{LD}}. Nevertheless, we've tried to make @code{@value{AS}}
620 assemble correctly everything that other assemblers for the same
621 machine would assemble.
623 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
626 @c This remark should appear in generic version of manual; assumption
627 @c here is that generic version sets M680x0.
628 This doesn't mean @code{@value{AS}} always uses the same syntax as another
629 assembler for the same architecture; for example, we know of several
630 incompatible versions of 680x0 assembly language syntax.
633 Unlike older assemblers, @code{@value{AS}} is designed to assemble a source
634 program in one pass of the source file. This has a subtle impact on the
635 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
638 @section Object File Formats
640 @cindex object file format
641 The @sc{gnu} assembler can be configured to produce several alternative
642 object file formats. For the most part, this does not affect how you
643 write assembly language programs; but directives for debugging symbols
644 are typically different in different file formats. @xref{Symbol
645 Attributes,,Symbol Attributes}.
648 On the @value{TARGET}, @code{@value{AS}} is configured to produce
649 @value{OBJ-NAME} format object files.
651 @c The following should exhaust all configs that set MULTI-OBJ, ideally
653 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
654 @code{a.out} or COFF format object files.
657 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
658 @code{b.out} or COFF format object files.
661 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
662 SOM or ELF format object files.
667 @section Command Line
669 @cindex command line conventions
670 After the program name @code{@value{AS}}, the command line may contain
671 options and file names. Options may appear in any order, and may be
672 before, after, or between file names. The order of file names is
675 @cindex standard input, as input file
677 @file{--} (two hyphens) by itself names the standard input file
678 explicitly, as one of the files for @code{@value{AS}} to assemble.
680 @cindex options, command line
681 Except for @samp{--} any command line argument that begins with a
682 hyphen (@samp{-}) is an option. Each option changes the behavior of
683 @code{@value{AS}}. No option changes the way another option works. An
684 option is a @samp{-} followed by one or more letters; the case of
685 the letter is important. All options are optional.
687 Some options expect exactly one file name to follow them. The file
688 name may either immediately follow the option's letter (compatible
689 with older assemblers) or it may be the next command argument (@sc{gnu}
690 standard). These two command lines are equivalent:
693 @value{AS} -o my-object-file.o mumble.s
694 @value{AS} -omy-object-file.o mumble.s
701 @cindex source program
703 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
704 describe the program input to one run of @code{@value{AS}}. The program may
705 be in one or more files; how the source is partitioned into files
706 doesn't change the meaning of the source.
708 @c I added "con" prefix to "catenation" just to prove I can overcome my
709 @c APL training... doc@cygnus.com
710 The source program is a concatenation of the text in all the files, in the
713 Each time you run @code{@value{AS}} it assembles exactly one source
714 program. The source program is made up of one or more files.
715 (The standard input is also a file.)
717 You give @code{@value{AS}} a command line that has zero or more input file
718 names. The input files are read (from left file name to right). A
719 command line argument (in any position) that has no special meaning
720 is taken to be an input file name.
722 If you give @code{@value{AS}} no file names it attempts to read one input file
723 from the @code{@value{AS}} standard input, which is normally your terminal. You
724 may have to type @key{ctl-D} to tell @code{@value{AS}} there is no more program
727 Use @samp{--} if you need to explicitly name the standard input file
728 in your command line.
730 If the source is empty, @code{@value{AS}} produces a small, empty object
733 @subheading Filenames and Line-numbers
735 @cindex input file linenumbers
736 @cindex line numbers, in input files
737 There are two ways of locating a line in the input file (or files) and
738 either may be used in reporting error messages. One way refers to a line
739 number in a physical file; the other refers to a line number in a
740 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
742 @dfn{Physical files} are those files named in the command line given
743 to @code{@value{AS}}.
745 @dfn{Logical files} are simply names declared explicitly by assembler
746 directives; they bear no relation to physical files. Logical file names
747 help error messages reflect the original source file, when @code{@value{AS}}
748 source is itself synthesized from other files.
749 @xref{App-File,,@code{.app-file}}.
752 @section Output (Object) File
758 Every time you run @code{@value{AS}} it produces an output file, which is
759 your assembly language program translated into numbers. This file
760 is the object file. Its default name is
768 @code{b.out} when @code{@value{AS}} is configured for the Intel 80960.
770 You can give it another name by using the @code{-o} option. Conventionally,
771 object file names end with @file{.o}. The default name is used for historical
772 reasons: older assemblers were capable of assembling self-contained programs
773 directly into a runnable program. (For some formats, this isn't currently
774 possible, but it can be done for the @code{a.out} format.)
778 The object file is meant for input to the linker @code{@value{LD}}. It contains
779 assembled program code, information to help @code{@value{LD}} integrate
780 the assembled program into a runnable file, and (optionally) symbolic
781 information for the debugger.
783 @c link above to some info file(s) like the description of a.out.
784 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
787 @section Error and Warning Messages
789 @cindex error messsages
790 @cindex warning messages
791 @cindex messages from @code{@value{AS}}
792 @code{@value{AS}} may write warnings and error messages to the standard error
793 file (usually your terminal). This should not happen when a compiler
794 runs @code{@value{AS}} automatically. Warnings report an assumption made so
795 that @code{@value{AS}} could keep assembling a flawed program; errors report a
796 grave problem that stops the assembly.
798 @cindex format of warning messages
799 Warning messages have the format
802 file_name:@b{NNN}:Warning Message Text
806 @cindex line numbers, in warnings/errors
807 (where @b{NNN} is a line number). If a logical file name has been given
808 (@pxref{App-File,,@code{.app-file}}) it is used for the filename,
809 otherwise the name of the current input file is used. If a logical line
812 (@pxref{Line,,@code{.line}})
816 (@pxref{Line,,@code{.line}})
819 (@pxref{Ln,,@code{.ln}})
822 then it is used to calculate the number printed,
823 otherwise the actual line in the current source file is printed. The
824 message text is intended to be self explanatory (in the grand Unix
827 @cindex format of error messages
828 Error messages have the format
830 file_name:@b{NNN}:FATAL:Error Message Text
832 The file name and line number are derived as for warning
833 messages. The actual message text may be rather less explanatory
834 because many of them aren't supposed to happen.
837 @chapter Command-Line Options
839 @cindex options, all versions of @code{@value{AS}}
840 This chapter describes command-line options available in @emph{all}
841 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
843 to the @value{TARGET}.
846 to particular machine architectures.
849 If you are invoking @code{@value{AS}} via the @sc{gnu} C compiler (version 2), you
850 can use the @samp{-Wa} option to pass arguments through to the
851 assembler. The assembler arguments must be separated from each other
852 (and the @samp{-Wa}) by commas. For example:
855 gcc -c -g -O -Wa,-alh,-L file.c
859 emits a listing to standard output with high-level
862 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
863 command-line options are automatically passed to the assembler by the compiler.
864 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
865 precisely what options it passes to each compilation pass, including the
869 * a:: -a[dhlns] enable listings
870 * D:: -D for compatibility
871 * f:: -f to work faster
872 * I:: -I for .include search path
873 @ifclear DIFF-TBL-KLUGE
874 * K:: -K for compatibility
876 @ifset DIFF-TBL-KLUGE
877 * K:: -K for difference tables
880 * L:: -L to retain local labels
881 * M:: -M or --mri to assemble in MRI compatibility mode
882 * o:: -o to name the object file
883 * R:: -R to join data and text sections
884 * statistics:: --statistics to see statistics about assembly
885 * v:: -v to announce version
886 * W:: -W to suppress warnings
887 * Z:: -Z to make object file even after errors
891 @section Enable Listings: @code{-a[dhlns]}
899 @cindex listings, enabling
900 @cindex assembly listings, enabling
902 These options enable listing output from the assembler. By itself,
903 @samp{-a} requests high-level, assembly, and symbols listing.
904 You can use other letters to select specific options for the list:
905 @samp{-ah} requests a high-level language listing,
906 @samp{-al} requests an output-program assembly listing, and
907 @samp{-as} requests a symbol table listing.
908 High-level listings require that a compiler debugging option like
909 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
912 Use the @samp{-ad} option to omit debugging directives from the
915 Once you have specified one of these options, you can further control
916 listing output and its appearance using the directives @code{.list},
917 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
919 The @samp{-an} option turns off all forms processing.
920 If you do not request listing output with one of the @samp{-a} options, the
921 listing-control directives have no effect.
923 The letters after @samp{-a} may be combined into one option,
924 @emph{e.g.}, @samp{-aln}.
930 This option has no effect whatsoever, but it is accepted to make it more
931 likely that scripts written for other assemblers also work with
935 @section Work Faster: @code{-f}
938 @cindex trusted compiler
939 @cindex faster processing (@code{-f})
940 @samp{-f} should only be used when assembling programs written by a
941 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
942 and comment preprocessing on
943 the input file(s) before assembling them. @xref{Preprocessing,
947 @emph{Warning:} if you use @samp{-f} when the files actually need to be
948 preprocessed (if they contain comments, for example), @code{@value{AS}} does
953 @section @code{.include} search path: @code{-I} @var{path}
955 @kindex -I @var{path}
956 @cindex paths for @code{.include}
957 @cindex search path for @code{.include}
958 @cindex @code{include} directive search path
959 Use this option to add a @var{path} to the list of directories
960 @code{@value{AS}} searches for files specified in @code{.include}
961 directives (@pxref{Include,,@code{.include}}). You may use @code{-I} as
962 many times as necessary to include a variety of paths. The current
963 working directory is always searched first; after that, @code{@value{AS}}
964 searches any @samp{-I} directories in the same order as they were
965 specified (left to right) on the command line.
968 @section Difference Tables: @code{-K}
971 @ifclear DIFF-TBL-KLUGE
972 On the @value{TARGET} family, this option is allowed, but has no effect. It is
973 permitted for compatibility with the @sc{gnu} assembler on other platforms,
974 where it can be used to warn when the assembler alters the machine code
975 generated for @samp{.word} directives in difference tables. The @value{TARGET}
976 family does not have the addressing limitations that sometimes lead to this
977 alteration on other platforms.
980 @ifset DIFF-TBL-KLUGE
981 @cindex difference tables, warning
982 @cindex warning for altered difference tables
983 @code{@value{AS}} sometimes alters the code emitted for directives of the form
984 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
985 You can use the @samp{-K} option if you want a warning issued when this
990 @section Include Local Labels: @code{-L}
993 @cindex local labels, retaining in output
994 Labels beginning with @samp{L} (upper case only) are called @dfn{local
995 labels}. @xref{Symbol Names}. Normally you do not see such labels when
996 debugging, because they are intended for the use of programs (like
997 compilers) that compose assembler programs, not for your notice.
998 Normally both @code{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
999 normally debug with them.
1001 This option tells @code{@value{AS}} to retain those @samp{L@dots{}} symbols
1002 in the object file. Usually if you do this you also tell the linker
1003 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1005 By default, a local label is any label beginning with @samp{L}, but each
1006 target is allowed to redefine the local label prefix.
1008 On the HPPA local labels begin with @samp{L$}.
1010 @c start-sanitize-arc
1012 On the ARC local labels begin with @samp{.L}.
1017 @section Assemble in MRI Compatibility Mode: @code{-M}
1020 @cindex MRI compatibility mode
1021 The @code{-M} or @code{--mri} option selects MRI compatibility mode. This
1022 changes the syntax and pseudo-op handling of @code{@value{AS}} to make it
1023 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1024 configured target) assembler from Microtec Research. The exact nature of the
1025 MRI syntax will not be documented here; see the MRI manuals for more
1026 information. The purpose of this option is to permit assembling existing MRI
1027 assembler code using @code{@value{AS}}.
1029 The MRI compatibility is not complete. Certain operations of the MRI assembler
1030 depend upon its object file format, and can not be supported using other object
1031 file formats. Supporting these would require enhancing each object file format
1032 individually. These are:
1035 @item global symbols in common section
1037 The m68k MRI assembler supports common sections which are merged by the linker.
1038 Other object file formats do not support this. @code{@value{AS}} handles
1039 common sections by treating them as a single common symbol. It permits local
1040 symbols to be defined within a common section, but it can not support global
1041 symbols, since it has no way to describe them.
1043 @item complex relocations
1045 The MRI assemblers support relocations against a negated section address, and
1046 relocations which combine the start addresses of two or more sections. These
1047 are not support by other object file formats.
1049 @item @code{END} pseudo-op specifying start address
1051 The MRI @code{END} pseudo-op permits the specification of a start address.
1052 This is not supported by other object file formats. The start address may
1053 instead be specified using the @code{-e} option to the linker, or in a linker
1056 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1058 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1059 name to the output file. This is not supported by other object file formats.
1061 @item @code{ORG} pseudo-op
1063 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1064 address. This differs from the usual @code{@value{AS}} @code{.org} pseudo-op,
1065 which changes the location within the current section. Absolute sections are
1066 not supported by other object file formats. The address of a section may be
1067 assigned within a linker script.
1070 There are some other features of the MRI assembler which are not supported by
1071 @code{@value{AS}}, typically either because they are difficult or because they
1072 seem of little consequence. Some of these may be supported in future releases.
1076 @item EBCDIC strings
1078 EBCDIC strings are not supported.
1080 @item packed binary coded decimal
1082 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1083 and @code{DCB.P} pseudo-ops are not supported.
1085 @item @code{FEQU} pseudo-op
1087 The m68k @code{FEQU} pseudo-op is not supported.
1089 @item @code{NOOBJ} pseudo-op
1091 The m68k @code{NOOBJ} pseudo-op is not supported.
1093 @item @code{OPT} branch control options
1095 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1096 @code{BRL}, and @code{BRW}---are ignored. @code{@value{AS}} automatically
1097 relaxes all branches, whether forward or backward, to an appropriate size, so
1098 these options serve no purpose.
1100 @item @code{OPT} list control options
1102 The following m68k @code{OPT} list control options are ignored: @code{C},
1103 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1104 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1106 @item other @code{OPT} options
1108 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1109 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1111 @item @code{OPT} @code{D} option is default
1113 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1114 @code{OPT NOD} may be used to turn it off.
1116 @item @code{XREF} pseudo-op.
1118 The m68k @code{XREF} pseudo-op is ignored.
1120 @item @code{.debug} pseudo-op
1122 The i960 @code{.debug} pseudo-op is not supported.
1124 @item @code{.extended} pseudo-op
1126 The i960 @code{.extended} pseudo-op is not supported.
1128 @item @code{.list} pseudo-op.
1130 The various options of the i960 @code{.list} pseudo-op are not supported.
1132 @item @code{.optimize} pseudo-op
1134 The i960 @code{.optimize} pseudo-op is not supported.
1136 @item @code{.output} pseudo-op
1138 The i960 @code{.output} pseudo-op is not supported.
1140 @item @code{.setreal} pseudo-op
1142 The i960 @code{.setreal} pseudo-op is not supported.
1147 @section Name the Object File: @code{-o}
1150 @cindex naming object file
1151 @cindex object file name
1152 There is always one object file output when you run @code{@value{AS}}. By
1153 default it has the name
1156 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1170 You use this option (which takes exactly one filename) to give the
1171 object file a different name.
1173 Whatever the object file is called, @code{@value{AS}} overwrites any
1174 existing file of the same name.
1177 @section Join Data and Text Sections: @code{-R}
1180 @cindex data and text sections, joining
1181 @cindex text and data sections, joining
1182 @cindex joining text and data sections
1183 @cindex merging text and data sections
1184 @code{-R} tells @code{@value{AS}} to write the object file as if all
1185 data-section data lives in the text section. This is only done at
1186 the very last moment: your binary data are the same, but data
1187 section parts are relocated differently. The data section part of
1188 your object file is zero bytes long because all its bytes are
1189 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1191 When you specify @code{-R} it would be possible to generate shorter
1192 address displacements (because we do not have to cross between text and
1193 data section). We refrain from doing this simply for compatibility with
1194 older versions of @code{@value{AS}}. In future, @code{-R} may work this way.
1197 When @code{@value{AS}} is configured for COFF output,
1198 this option is only useful if you use sections named @samp{.text} and
1203 @code{-R} is not supported for any of the HPPA targets. Using
1204 @code{-R} generates a warning from @code{@value{AS}}.
1208 @section Display Assembly Statistics: @code{--statistics}
1210 @kindex --statistics
1211 @cindex statistics, about assembly
1212 @cindex time, total for assembly
1213 @cindex space used, maximum for assembly
1214 Use @samp{--statistics} to display two statistics about the resources used by
1215 @code{@value{AS}}: the maximum amount of space allocated during the assembly
1216 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1220 @section Announce Version: @code{-v}
1224 @cindex @code{@value{AS}} version
1225 @cindex version of @code{@value{AS}}
1226 You can find out what version of as is running by including the
1227 option @samp{-v} (which you can also spell as @samp{-version}) on the
1231 @section Suppress Warnings: @code{-W}
1234 @cindex suppressing warnings
1235 @cindex warnings, suppressing
1236 @code{@value{AS}} should never give a warning or error message when
1237 assembling compiler output. But programs written by people often
1238 cause @code{@value{AS}} to give a warning that a particular assumption was
1239 made. All such warnings are directed to the standard error file.
1240 If you use this option, no warnings are issued. This option only
1241 affects the warning messages: it does not change any particular of how
1242 @code{@value{AS}} assembles your file. Errors, which stop the assembly, are
1246 @section Generate Object File in Spite of Errors: @code{-Z}
1247 @cindex object file, after errors
1248 @cindex errors, continuing after
1249 After an error message, @code{@value{AS}} normally produces no output. If for
1250 some reason you are interested in object file output even after
1251 @code{@value{AS}} gives an error message on your program, use the @samp{-Z}
1252 option. If there are any errors, @code{@value{AS}} continues anyways, and
1253 writes an object file after a final warning message of the form @samp{@var{n}
1254 errors, @var{m} warnings, generating bad object file.}
1259 @cindex machine-independent syntax
1260 @cindex syntax, machine-independent
1261 This chapter describes the machine-independent syntax allowed in a
1262 source file. @code{@value{AS}} syntax is similar to what many other
1263 assemblers use; it is inspired by the BSD 4.2
1268 assembler, except that @code{@value{AS}} does not assemble Vax bit-fields.
1272 * Preprocessing:: Preprocessing
1273 * Whitespace:: Whitespace
1274 * Comments:: Comments
1275 * Symbol Intro:: Symbols
1276 * Statements:: Statements
1277 * Constants:: Constants
1281 @section Preprocessing
1283 @cindex preprocessing
1284 The @code{@value{AS}} internal preprocessor:
1286 @cindex whitespace, removed by preprocessor
1288 adjusts and removes extra whitespace. It leaves one space or tab before
1289 the keywords on a line, and turns any other whitespace on the line into
1292 @cindex comments, removed by preprocessor
1294 removes all comments, replacing them with a single space, or an
1295 appropriate number of newlines.
1297 @cindex constants, converted by preprocessor
1299 converts character constants into the appropriate numeric values.
1302 It does not do macro processing, include file handling, or
1303 anything else you may get from your C compiler's preprocessor. You can
1304 do include file processing with the @code{.include} directive
1305 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
1306 to get other ``CPP'' style preprocessing, by giving the input file a
1307 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
1308 Output, gcc.info, Using GNU CC}.
1310 Excess whitespace, comments, and character constants
1311 cannot be used in the portions of the input text that are not
1314 @cindex turning preprocessing on and off
1315 @cindex preprocessing, turning on and off
1318 If the first line of an input file is @code{#NO_APP} or if you use the
1319 @samp{-f} option, whitespace and comments are not removed from the input file.
1320 Within an input file, you can ask for whitespace and comment removal in
1321 specific portions of the by putting a line that says @code{#APP} before the
1322 text that may contain whitespace or comments, and putting a line that says
1323 @code{#NO_APP} after this text. This feature is mainly intend to support
1324 @code{asm} statements in compilers whose output is otherwise free of comments
1331 @dfn{Whitespace} is one or more blanks or tabs, in any order.
1332 Whitespace is used to separate symbols, and to make programs neater for
1333 people to read. Unless within character constants
1334 (@pxref{Characters,,Character Constants}), any whitespace means the same
1335 as exactly one space.
1341 There are two ways of rendering comments to @code{@value{AS}}. In both
1342 cases the comment is equivalent to one space.
1344 Anything from @samp{/*} through the next @samp{*/} is a comment.
1345 This means you may not nest these comments.
1349 The only way to include a newline ('\n') in a comment
1350 is to use this sort of comment.
1353 /* This sort of comment does not nest. */
1356 @cindex line comment character
1357 Anything from the @dfn{line comment} character to the next newline
1358 is considered a comment and is ignored. The line comment character is
1360 @samp{;} for the AMD 29K family;
1362 @c start-sanitize-arc
1364 @samp{;} on the ARC;
1368 @samp{;} for the H8/300 family;
1371 @samp{!} for the H8/500 family;
1374 @samp{;} for the HPPA;
1377 @samp{#} on the i960;
1380 @samp{!} for the Hitachi SH;
1383 @samp{!} on the SPARC;
1386 @samp{|} on the 680x0;
1389 @samp{#} on the Vax;
1392 @samp{!} for the Z8000;
1394 see @ref{Machine Dependencies}. @refill
1395 @c FIXME What about i386, m88k, i860?
1398 On some machines there are two different line comment characters. One
1399 character only begins a comment if it is the first non-whitespace character on
1400 a line, while the other always begins a comment.
1404 @cindex lines starting with @code{#}
1405 @cindex logical line numbers
1406 To be compatible with past assemblers, lines that begin with @samp{#} have a
1407 special interpretation. Following the @samp{#} should be an absolute
1408 expression (@pxref{Expressions}): the logical line number of the @emph{next}
1409 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
1410 new logical file name. The rest of the line, if any, should be whitespace.
1412 If the first non-whitespace characters on the line are not numeric,
1413 the line is ignored. (Just like a comment.)
1416 # This is an ordinary comment.
1417 # 42-6 "new_file_name" # New logical file name
1418 # This is logical line # 36.
1420 This feature is deprecated, and may disappear from future versions
1421 of @code{@value{AS}}.
1426 @cindex characters used in symbols
1427 @ifclear SPECIAL-SYMS
1428 A @dfn{symbol} is one or more characters chosen from the set of all
1429 letters (both upper and lower case), digits and the three characters
1435 A @dfn{symbol} is one or more characters chosen from the set of all
1436 letters (both upper and lower case), digits and the three characters
1437 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
1443 On most machines, you can also use @code{$} in symbol names; exceptions
1444 are noted in @ref{Machine Dependencies}.
1446 No symbol may begin with a digit. Case is significant.
1447 There is no length limit: all characters are significant. Symbols are
1448 delimited by characters not in that set, or by the beginning of a file
1449 (since the source program must end with a newline, the end of a file is
1450 not a possible symbol delimiter). @xref{Symbols}.
1451 @cindex length of symbols
1456 @cindex statements, structure of
1457 @cindex line separator character
1458 @cindex statement separator character
1460 @ifclear abnormal-separator
1461 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
1462 semicolon (@samp{;}). The newline or semicolon is considered part of
1463 the preceding statement. Newlines and semicolons within character
1464 constants are an exception: they do not end statements.
1466 @ifset abnormal-separator
1468 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
1469 sign (@samp{@@}). The newline or at sign is considered part of the
1470 preceding statement. Newlines and at signs within character constants
1471 are an exception: they do not end statements.
1474 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
1475 point (@samp{!}). The newline or exclamation point is considered part of the
1476 preceding statement. Newlines and exclamation points within character
1477 constants are an exception: they do not end statements.
1480 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
1481 H8/300) a dollar sign (@samp{$}); or (for the
1484 (@samp{;}). The newline or separator character is considered part of
1485 the preceding statement. Newlines and separators within character
1486 constants are an exception: they do not end statements.
1491 A @dfn{statement} ends at a newline character (@samp{\n}) or line
1492 separator character. (The line separator is usually @samp{;}, unless
1493 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
1494 newline or separator character is considered part of the preceding
1495 statement. Newlines and separators within character constants are an
1496 exception: they do not end statements.
1499 @cindex newline, required at file end
1500 @cindex EOF, newline must precede
1501 It is an error to end any statement with end-of-file: the last
1502 character of any input file should be a newline.@refill
1504 @cindex continuing statements
1505 @cindex multi-line statements
1506 @cindex statement on multiple lines
1507 You may write a statement on more than one line if you put a
1508 backslash (@kbd{\}) immediately in front of any newlines within the
1509 statement. When @code{@value{AS}} reads a backslashed newline both
1510 characters are ignored. You can even put backslashed newlines in
1511 the middle of symbol names without changing the meaning of your
1514 An empty statement is allowed, and may include whitespace. It is ignored.
1516 @cindex instructions and directives
1517 @cindex directives and instructions
1518 @c "key symbol" is not used elsewhere in the document; seems pedantic to
1519 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
1521 A statement begins with zero or more labels, optionally followed by a
1522 key symbol which determines what kind of statement it is. The key
1523 symbol determines the syntax of the rest of the statement. If the
1524 symbol begins with a dot @samp{.} then the statement is an assembler
1525 directive: typically valid for any computer. If the symbol begins with
1526 a letter the statement is an assembly language @dfn{instruction}: it
1527 assembles into a machine language instruction.
1529 Different versions of @code{@value{AS}} for different computers
1530 recognize different instructions. In fact, the same symbol may
1531 represent a different instruction in a different computer's assembly
1535 @cindex @code{:} (label)
1536 @cindex label (@code{:})
1537 A label is a symbol immediately followed by a colon (@code{:}).
1538 Whitespace before a label or after a colon is permitted, but you may not
1539 have whitespace between a label's symbol and its colon. @xref{Labels}.
1542 For HPPA targets, labels need not be immediately followed by a colon, but
1543 the definition of a label must begin in column zero. This also implies that
1544 only one label may be defined on each line.
1548 label: .directive followed by something
1549 another_label: # This is an empty statement.
1550 instruction operand_1, operand_2, @dots{}
1557 A constant is a number, written so that its value is known by
1558 inspection, without knowing any context. Like this:
1561 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
1562 .ascii "Ring the bell\7" # A string constant.
1563 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
1564 .float 0f-314159265358979323846264338327\
1565 95028841971.693993751E-40 # - pi, a flonum.
1570 * Characters:: Character Constants
1571 * Numbers:: Number Constants
1575 @subsection Character Constants
1577 @cindex character constants
1578 @cindex constants, character
1579 There are two kinds of character constants. A @dfn{character} stands
1580 for one character in one byte and its value may be used in
1581 numeric expressions. String constants (properly called string
1582 @emph{literals}) are potentially many bytes and their values may not be
1583 used in arithmetic expressions.
1587 * Chars:: Characters
1591 @subsubsection Strings
1593 @cindex string constants
1594 @cindex constants, string
1595 A @dfn{string} is written between double-quotes. It may contain
1596 double-quotes or null characters. The way to get special characters
1597 into a string is to @dfn{escape} these characters: precede them with
1598 a backslash @samp{\} character. For example @samp{\\} represents
1599 one backslash: the first @code{\} is an escape which tells
1600 @code{@value{AS}} to interpret the second character literally as a backslash
1601 (which prevents @code{@value{AS}} from recognizing the second @code{\} as an
1602 escape character). The complete list of escapes follows.
1604 @cindex escape codes, character
1605 @cindex character escape codes
1608 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
1610 @cindex @code{\b} (backspace character)
1611 @cindex backspace (@code{\b})
1613 Mnemonic for backspace; for ASCII this is octal code 010.
1616 @c Mnemonic for EOText; for ASCII this is octal code 004.
1618 @cindex @code{\f} (formfeed character)
1619 @cindex formfeed (@code{\f})
1621 Mnemonic for FormFeed; for ASCII this is octal code 014.
1623 @cindex @code{\n} (newline character)
1624 @cindex newline (@code{\n})
1626 Mnemonic for newline; for ASCII this is octal code 012.
1629 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
1631 @cindex @code{\r} (carriage return character)
1632 @cindex carriage return (@code{\r})
1634 Mnemonic for carriage-Return; for ASCII this is octal code 015.
1637 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
1638 @c other assemblers.
1640 @cindex @code{\t} (tab)
1641 @cindex tab (@code{\t})
1643 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
1646 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
1647 @c @item \x @var{digit} @var{digit} @var{digit}
1648 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
1650 @cindex @code{\@var{ddd}} (octal character code)
1651 @cindex octal character code (@code{\@var{ddd}})
1652 @item \ @var{digit} @var{digit} @var{digit}
1653 An octal character code. The numeric code is 3 octal digits.
1654 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
1655 for example, @code{\008} has the value 010, and @code{\009} the value 011.
1657 @cindex @code{\@var{xd...}} (hex character code)
1658 @cindex hex character code (@code{\@var{xd...}})
1659 @item \@code{x} @var{hex-digits...}
1660 A hex character code. All trailing hex digits are combined. Either upper or
1661 lower case @code{x} works.
1663 @cindex @code{\\} (@samp{\} character)
1664 @cindex backslash (@code{\\})
1666 Represents one @samp{\} character.
1669 @c Represents one @samp{'} (accent acute) character.
1670 @c This is needed in single character literals
1671 @c (@xref{Characters,,Character Constants}.) to represent
1674 @cindex @code{\"} (doublequote character)
1675 @cindex doublequote (@code{\"})
1677 Represents one @samp{"} character. Needed in strings to represent
1678 this character, because an unescaped @samp{"} would end the string.
1680 @item \ @var{anything-else}
1681 Any other character when escaped by @kbd{\} gives a warning, but
1682 assembles as if the @samp{\} was not present. The idea is that if
1683 you used an escape sequence you clearly didn't want the literal
1684 interpretation of the following character. However @code{@value{AS}} has no
1685 other interpretation, so @code{@value{AS}} knows it is giving you the wrong
1686 code and warns you of the fact.
1689 Which characters are escapable, and what those escapes represent,
1690 varies widely among assemblers. The current set is what we think
1691 the BSD 4.2 assembler recognizes, and is a subset of what most C
1692 compilers recognize. If you are in doubt, do not use an escape
1696 @subsubsection Characters
1698 @cindex single character constant
1699 @cindex character, single
1700 @cindex constant, single character
1701 A single character may be written as a single quote immediately
1702 followed by that character. The same escapes apply to characters as
1703 to strings. So if you want to write the character backslash, you
1704 must write @kbd{'\\} where the first @code{\} escapes the second
1705 @code{\}. As you can see, the quote is an acute accent, not a
1706 grave accent. A newline
1708 @ifclear abnormal-separator
1709 (or semicolon @samp{;})
1711 @ifset abnormal-separator
1713 (or at sign @samp{@@})
1716 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
1722 immediately following an acute accent is taken as a literal character
1723 and does not count as the end of a statement. The value of a character
1724 constant in a numeric expression is the machine's byte-wide code for
1725 that character. @code{@value{AS}} assumes your character code is ASCII:
1726 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
1729 @subsection Number Constants
1731 @cindex constants, number
1732 @cindex number constants
1733 @code{@value{AS}} distinguishes three kinds of numbers according to how they
1734 are stored in the target machine. @emph{Integers} are numbers that
1735 would fit into an @code{int} in the C language. @emph{Bignums} are
1736 integers, but they are stored in more than 32 bits. @emph{Flonums}
1737 are floating point numbers, described below.
1740 * Integers:: Integers
1745 * Bit Fields:: Bit Fields
1751 @subsubsection Integers
1753 @cindex constants, integer
1755 @cindex binary integers
1756 @cindex integers, binary
1757 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
1758 the binary digits @samp{01}.
1760 @cindex octal integers
1761 @cindex integers, octal
1762 An octal integer is @samp{0} followed by zero or more of the octal
1763 digits (@samp{01234567}).
1765 @cindex decimal integers
1766 @cindex integers, decimal
1767 A decimal integer starts with a non-zero digit followed by zero or
1768 more digits (@samp{0123456789}).
1770 @cindex hexadecimal integers
1771 @cindex integers, hexadecimal
1772 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
1773 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
1775 Integers have the usual values. To denote a negative integer, use
1776 the prefix operator @samp{-} discussed under expressions
1777 (@pxref{Prefix Ops,,Prefix Operators}).
1780 @subsubsection Bignums
1783 @cindex constants, bignum
1784 A @dfn{bignum} has the same syntax and semantics as an integer
1785 except that the number (or its negative) takes more than 32 bits to
1786 represent in binary. The distinction is made because in some places
1787 integers are permitted while bignums are not.
1790 @subsubsection Flonums
1792 @cindex floating point numbers
1793 @cindex constants, floating point
1795 @cindex precision, floating point
1796 A @dfn{flonum} represents a floating point number. The translation is
1797 indirect: a decimal floating point number from the text is converted by
1798 @code{@value{AS}} to a generic binary floating point number of more than
1799 sufficient precision. This generic floating point number is converted
1800 to a particular computer's floating point format (or formats) by a
1801 portion of @code{@value{AS}} specialized to that computer.
1803 A flonum is written by writing (in order)
1808 (@samp{0} is optional on the HPPA.)
1812 A letter, to tell @code{@value{AS}} the rest of the number is a flonum.
1814 @kbd{e} is recommended. Case is not important.
1816 @c FIXME: verify if flonum syntax really this vague for most cases
1817 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
1818 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
1821 On the H8/300, H8/500,
1823 and AMD 29K architectures, the letter must be
1824 one of the letters @samp{DFPRSX} (in upper or lower case).
1826 @c start-sanitize-arc
1827 On the ARC, the letter one of the letters @samp{DFRS}
1828 (in upper or lower case).
1831 On the Intel 960 architecture, the letter must be
1832 one of the letters @samp{DFT} (in upper or lower case).
1834 On the HPPA architecture, the letter must be @samp{E} (upper case only).
1838 One of the letters @samp{DFPRSX} (in upper or lower case).
1840 @c start-sanitize-arc
1842 One of the letters @samp{DFRS} (in upper or lower case).
1846 One of the letters @samp{DFPRSX} (in upper or lower case).
1849 The letter @samp{E} (upper case only).
1852 One of the letters @samp{DFT} (in upper or lower case).
1857 An optional sign: either @samp{+} or @samp{-}.
1860 An optional @dfn{integer part}: zero or more decimal digits.
1863 An optional @dfn{fractional part}: @samp{.} followed by zero
1864 or more decimal digits.
1867 An optional exponent, consisting of:
1871 An @samp{E} or @samp{e}.
1872 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
1873 @c principle this can perfectly well be different on different targets.
1875 Optional sign: either @samp{+} or @samp{-}.
1877 One or more decimal digits.
1882 At least one of the integer part or the fractional part must be
1883 present. The floating point number has the usual base-10 value.
1885 @code{@value{AS}} does all processing using integers. Flonums are computed
1886 independently of any floating point hardware in the computer running
1891 @c Bit fields are written as a general facility but are also controlled
1892 @c by a conditional-compilation flag---which is as of now (21mar91)
1893 @c turned on only by the i960 config of GAS.
1895 @subsubsection Bit Fields
1898 @cindex constants, bit field
1899 You can also define numeric constants as @dfn{bit fields}.
1900 specify two numbers separated by a colon---
1902 @var{mask}:@var{value}
1905 @code{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
1908 The resulting number is then packed
1910 @c this conditional paren in case bit fields turned on elsewhere than 960
1911 (in host-dependent byte order)
1913 into a field whose width depends on which assembler directive has the
1914 bit-field as its argument. Overflow (a result from the bitwise and
1915 requiring more binary digits to represent) is not an error; instead,
1916 more constants are generated, of the specified width, beginning with the
1917 least significant digits.@refill
1919 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
1920 @code{.short}, and @code{.word} accept bit-field arguments.
1925 @chapter Sections and Relocation
1930 * Secs Background:: Background
1931 * Ld Sections:: @value{LD} Sections
1932 * As Sections:: @value{AS} Internal Sections
1933 * Sub-Sections:: Sub-Sections
1937 @node Secs Background
1940 Roughly, a section is a range of addresses, with no gaps; all data
1941 ``in'' those addresses is treated the same for some particular purpose.
1942 For example there may be a ``read only'' section.
1944 @cindex linker, and assembler
1945 @cindex assembler, and linker
1946 The linker @code{@value{LD}} reads many object files (partial programs) and
1947 combines their contents to form a runnable program. When @code{@value{AS}}
1948 emits an object file, the partial program is assumed to start at address 0.
1949 @code{@value{LD}} assigns the final addresses for the partial program, so that
1950 different partial programs do not overlap. This is actually an
1951 oversimplification, but it suffices to explain how @code{@value{AS}} uses
1954 @code{@value{LD}} moves blocks of bytes of your program to their run-time
1955 addresses. These blocks slide to their run-time addresses as rigid
1956 units; their length does not change and neither does the order of bytes
1957 within them. Such a rigid unit is called a @emph{section}. Assigning
1958 run-time addresses to sections is called @dfn{relocation}. It includes
1959 the task of adjusting mentions of object-file addresses so they refer to
1960 the proper run-time addresses.
1962 For the H8/300 and H8/500,
1963 and for the Hitachi SH,
1964 @code{@value{AS}} pads sections if needed to
1965 ensure they end on a word (sixteen bit) boundary.
1968 @cindex standard @code{@value{AS}} sections
1969 An object file written by @code{@value{AS}} has at least three sections, any
1970 of which may be empty. These are named @dfn{text}, @dfn{data} and
1975 When it generates COFF output,
1977 @code{@value{AS}} can also generate whatever other named sections you specify
1978 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
1979 If you do not use any directives that place output in the @samp{.text}
1980 or @samp{.data} sections, these sections still exist, but are empty.
1985 When @code{@value{AS}} generates SOM or ELF output for the HPPA,
1987 @code{@value{AS}} can also generate whatever other named sections you
1988 specify using the @samp{.space} and @samp{.subspace} directives. See
1989 @cite{HP9000 Series 800 Assembly Language Reference Manual}
1990 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
1991 assembler directives.
1994 Additionally, @code{@value{AS}} uses different names for the standard
1995 text, data, and bss sections when generating SOM output. Program text
1996 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
1997 BSS into @samp{$BSS$}.
2001 Within the object file, the text section starts at address @code{0}, the
2002 data section follows, and the bss section follows the data section.
2005 When generating either SOM or ELF output files on the HPPA, the text
2006 section starts at address @code{0}, the data section at address
2007 @code{0x4000000}, and the bss section follows the data section.
2010 To let @code{@value{LD}} know which data changes when the sections are
2011 relocated, and how to change that data, @code{@value{AS}} also writes to the
2012 object file details of the relocation needed. To perform relocation
2013 @code{@value{LD}} must know, each time an address in the object
2017 Where in the object file is the beginning of this reference to
2020 How long (in bytes) is this reference?
2022 Which section does the address refer to? What is the numeric value of
2024 (@var{address}) @minus{} (@var{start-address of section})?
2027 Is the reference to an address ``Program-Counter relative''?
2030 @cindex addresses, format of
2031 @cindex section-relative addressing
2032 In fact, every address @code{@value{AS}} ever uses is expressed as
2034 (@var{section}) + (@var{offset into section})
2037 Further, most expressions @code{@value{AS}} computes have this section-relative
2040 (For some object formats, such as SOM for the HPPA, some expressions are
2041 symbol-relative instead.)
2044 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2045 @var{N} into section @var{secname}.''
2047 Apart from text, data and bss sections you need to know about the
2048 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2049 addresses in the absolute section remain unchanged. For example, address
2050 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2051 @code{@value{LD}}. Although the linker never arranges two partial programs'
2052 data sections with overlapping addresses after linking, @emph{by definition}
2053 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2054 part of a program is always the same address when the program is running as
2055 address @code{@{absolute@ 239@}} in any other part of the program.
2057 The idea of sections is extended to the @dfn{undefined} section. Any
2058 address whose section is unknown at assembly time is by definition
2059 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2060 Since numbers are always defined, the only way to generate an undefined
2061 address is to mention an undefined symbol. A reference to a named
2062 common block would be such a symbol: its value is unknown at assembly
2063 time so it has section @emph{undefined}.
2065 By analogy the word @emph{section} is used to describe groups of sections in
2066 the linked program. @code{@value{LD}} puts all partial programs' text
2067 sections in contiguous addresses in the linked program. It is
2068 customary to refer to the @emph{text section} of a program, meaning all
2069 the addresses of all partial programs' text sections. Likewise for
2070 data and bss sections.
2072 Some sections are manipulated by @code{@value{LD}}; others are invented for
2073 use of @code{@value{AS}} and have no meaning except during assembly.
2076 @section @value{LD} Sections
2077 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2082 @cindex named sections
2083 @cindex sections, named
2084 @item named sections
2087 @cindex text section
2088 @cindex data section
2092 These sections hold your program. @code{@value{AS}} and @code{@value{LD}} treat them as
2093 separate but equal sections. Anything you can say of one section is
2096 When the program is running, however, it is
2097 customary for the text section to be unalterable. The
2098 text section is often shared among processes: it contains
2099 instructions, constants and the like. The data section of a running
2100 program is usually alterable: for example, C variables would be stored
2101 in the data section.
2106 This section contains zeroed bytes when your program begins running. It
2107 is used to hold unitialized variables or common storage. The length of
2108 each partial program's bss section is important, but because it starts
2109 out containing zeroed bytes there is no need to store explicit zero
2110 bytes in the object file. The bss section was invented to eliminate
2111 those explicit zeros from object files.
2113 @cindex absolute section
2114 @item absolute section
2115 Address 0 of this section is always ``relocated'' to runtime address 0.
2116 This is useful if you want to refer to an address that @code{@value{LD}} must
2117 not change when relocating. In this sense we speak of absolute
2118 addresses being ``unrelocatable'': they do not change during relocation.
2120 @cindex undefined section
2121 @item undefined section
2122 This ``section'' is a catch-all for address references to objects not in
2123 the preceding sections.
2124 @c FIXME: ref to some other doc on obj-file formats could go here.
2127 @cindex relocation example
2128 An idealized example of three relocatable sections follows.
2130 The example uses the traditional section names @samp{.text} and @samp{.data}.
2132 Memory addresses are on the horizontal axis.
2136 @c END TEXI2ROFF-KILL
2139 partial program # 1: |ttttt|dddd|00|
2146 partial program # 2: |TTT|DDD|000|
2149 +--+---+-----+--+----+---+-----+~~
2150 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2151 +--+---+-----+--+----+---+-----+~~
2153 addresses: 0 @dots{}
2160 \line{\it Partial program \#1: \hfil}
2161 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2162 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2164 \line{\it Partial program \#2: \hfil}
2165 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2166 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2168 \line{\it linked program: \hfil}
2169 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2170 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2171 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2172 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2174 \line{\it addresses: \hfil}
2178 @c END TEXI2ROFF-KILL
2181 @section @value{AS} Internal Sections
2183 @cindex internal @code{@value{AS}} sections
2184 @cindex sections in messages, internal
2185 These sections are meant only for the internal use of @code{@value{AS}}. They
2186 have no meaning at run-time. You do not really need to know about these
2187 sections for most purposes; but they can be mentioned in @code{@value{AS}}
2188 warning messages, so it might be helpful to have an idea of their
2189 meanings to @code{@value{AS}}. These sections are used to permit the
2190 value of every expression in your assembly language program to be a
2191 section-relative address.
2194 @cindex assembler internal logic error
2195 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2196 An internal assembler logic error has been found. This means there is a
2197 bug in the assembler.
2199 @cindex expr (internal section)
2201 The assembler stores complex expression internally as combinations of
2202 symbols. When it needs to represent an expression as a symbol, it puts
2203 it in the expr section.
2205 @c FIXME item transfer[t] vector preload
2206 @c FIXME item transfer[t] vector postload
2207 @c FIXME item register
2211 @section Sub-Sections
2213 @cindex numbered subsections
2214 @cindex grouping data
2220 fall into two sections: text and data.
2222 You may have separate groups of
2224 data in named sections
2228 data in named sections
2234 that you want to end up near to each other in the object file, even though they
2235 are not contiguous in the assembler source. @code{@value{AS}} allows you to
2236 use @dfn{subsections} for this purpose. Within each section, there can be
2237 numbered subsections with values from 0 to 8192. Objects assembled into the
2238 same subsection go into the object file together with other objects in the same
2239 subsection. For example, a compiler might want to store constants in the text
2240 section, but might not want to have them interspersed with the program being
2241 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2242 section of code being output, and a @samp{.text 1} before each group of
2243 constants being output.
2245 Subsections are optional. If you do not use subsections, everything
2246 goes in subsection number zero.
2249 Each subsection is zero-padded up to a multiple of four bytes.
2250 (Subsections may be padded a different amount on different flavors
2251 of @code{@value{AS}}.)
2255 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2256 boundary (two bytes).
2257 The same is true on the Hitachi SH.
2260 @c FIXME section padding (alignment)?
2261 @c Rich Pixley says padding here depends on target obj code format; that
2262 @c doesn't seem particularly useful to say without further elaboration,
2263 @c so for now I say nothing about it. If this is a generic BFD issue,
2264 @c these paragraphs might need to vanish from this manual, and be
2265 @c discussed in BFD chapter of binutils (or some such).
2268 On the AMD 29K family, no particular padding is added to section or
2269 subsection sizes; @value{AS} forces no alignment on this platform.
2273 Subsections appear in your object file in numeric order, lowest numbered
2274 to highest. (All this to be compatible with other people's assemblers.)
2275 The object file contains no representation of subsections; @code{@value{LD}} and
2276 other programs that manipulate object files see no trace of them.
2277 They just see all your text subsections as a text section, and all your
2278 data subsections as a data section.
2280 To specify which subsection you want subsequent statements assembled
2281 into, use a numeric argument to specify it, in a @samp{.text
2282 @var{expression}} or a @samp{.data @var{expression}} statement.
2285 When generating COFF output, you
2290 can also use an extra subsection
2291 argument with arbitrary named sections: @samp{.section @var{name},
2294 @var{Expression} should be an absolute expression.
2295 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
2296 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
2297 begins in @code{text 0}. For instance:
2299 .text 0 # The default subsection is text 0 anyway.
2300 .ascii "This lives in the first text subsection. *"
2302 .ascii "But this lives in the second text subsection."
2304 .ascii "This lives in the data section,"
2305 .ascii "in the first data subsection."
2307 .ascii "This lives in the first text section,"
2308 .ascii "immediately following the asterisk (*)."
2311 Each section has a @dfn{location counter} incremented by one for every byte
2312 assembled into that section. Because subsections are merely a convenience
2313 restricted to @code{@value{AS}} there is no concept of a subsection location
2314 counter. There is no way to directly manipulate a location counter---but the
2315 @code{.align} directive changes it, and any label definition captures its
2316 current value. The location counter of the section where statements are being
2317 assembled is said to be the @dfn{active} location counter.
2320 @section bss Section
2323 @cindex common variable storage
2324 The bss section is used for local common variable storage.
2325 You may allocate address space in the bss section, but you may
2326 not dictate data to load into it before your program executes. When
2327 your program starts running, all the contents of the bss
2328 section are zeroed bytes.
2330 Addresses in the bss section are allocated with special directives; you
2331 may not assemble anything directly into the bss section. Hence there
2332 are no bss subsections. @xref{Comm,,@code{.comm}},
2333 @pxref{Lcomm,,@code{.lcomm}}.
2339 Symbols are a central concept: the programmer uses symbols to name
2340 things, the linker uses symbols to link, and the debugger uses symbols
2344 @cindex debuggers, and symbol order
2345 @emph{Warning:} @code{@value{AS}} does not place symbols in the object file in
2346 the same order they were declared. This may break some debuggers.
2351 * Setting Symbols:: Giving Symbols Other Values
2352 * Symbol Names:: Symbol Names
2353 * Dot:: The Special Dot Symbol
2354 * Symbol Attributes:: Symbol Attributes
2361 A @dfn{label} is written as a symbol immediately followed by a colon
2362 @samp{:}. The symbol then represents the current value of the
2363 active location counter, and is, for example, a suitable instruction
2364 operand. You are warned if you use the same symbol to represent two
2365 different locations: the first definition overrides any other
2369 On the HPPA, the usual form for a label need not be immediately followed by a
2370 colon, but instead must start in column zero. Only one label may be defined on
2371 a single line. To work around this, the HPPA version of @code{@value{AS}} also
2372 provides a special directive @code{.label} for defining labels more flexibly.
2375 @node Setting Symbols
2376 @section Giving Symbols Other Values
2378 @cindex assigning values to symbols
2379 @cindex symbol values, assigning
2380 A symbol can be given an arbitrary value by writing a symbol, followed
2381 by an equals sign @samp{=}, followed by an expression
2382 (@pxref{Expressions}). This is equivalent to using the @code{.set}
2383 directive. @xref{Set,,@code{.set}}.
2386 @section Symbol Names
2388 @cindex symbol names
2389 @cindex names, symbol
2390 @ifclear SPECIAL-SYMS
2391 Symbol names begin with a letter or with one of @samp{._}. On most
2392 machines, you can also use @code{$} in symbol names; exceptions are
2393 noted in @ref{Machine Dependencies}. That character may be followed by any
2394 string of digits, letters, dollar signs (unless otherwise noted in
2395 @ref{Machine Dependencies}), and underscores.
2398 For the AMD 29K family, @samp{?} is also allowed in the
2399 body of a symbol name, though not at its beginning.
2404 Symbol names begin with a letter or with one of @samp{._}. On the
2406 H8/500, you can also use @code{$} in symbol names. That character may
2407 be followed by any string of digits, letters, dollar signs (save on the
2408 H8/300), and underscores.
2412 Case of letters is significant: @code{foo} is a different symbol name
2415 Each symbol has exactly one name. Each name in an assembly language program
2416 refers to exactly one symbol. You may use that symbol name any number of times
2419 @subheading Local Symbol Names
2421 @cindex local symbol names
2422 @cindex symbol names, local
2423 @cindex temporary symbol names
2424 @cindex symbol names, temporary
2425 Local symbols help compilers and programmers use names temporarily.
2426 There are ten local symbol names, which are re-used throughout the
2427 program. You may refer to them using the names @samp{0} @samp{1}
2428 @dots{} @samp{9}. To define a local symbol, write a label of the form
2429 @samp{@b{N}:} (where @b{N} represents any digit). To refer to the most
2430 recent previous definition of that symbol write @samp{@b{N}b}, using the
2431 same digit as when you defined the label. To refer to the next
2432 definition of a local label, write @samp{@b{N}f}---where @b{N} gives you
2433 a choice of 10 forward references. The @samp{b} stands for
2434 ``backwards'' and the @samp{f} stands for ``forwards''.
2436 Local symbols are not emitted by the current @sc{gnu} C compiler.
2438 There is no restriction on how you can use these labels, but
2439 remember that at any point in the assembly you can refer to at most
2440 10 prior local labels and to at most 10 forward local labels.
2442 Local symbol names are only a notation device. They are immediately
2443 transformed into more conventional symbol names before the assembler
2444 uses them. The symbol names stored in the symbol table, appearing in
2445 error messages and optionally emitted to the object file have these
2450 All local labels begin with @samp{L}. Normally both @code{@value{AS}} and
2451 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
2452 used for symbols you are never intended to see. If you use the
2453 @samp{-L} option then @code{@value{AS}} retains these symbols in the
2454 object file. If you also instruct @code{@value{LD}} to retain these symbols,
2455 you may use them in debugging.
2458 If the label is written @samp{0:} then the digit is @samp{0}.
2459 If the label is written @samp{1:} then the digit is @samp{1}.
2460 And so on up through @samp{9:}.
2463 This unusual character is included so you do not accidentally invent
2464 a symbol of the same name. The character has ASCII value
2467 @item @emph{ordinal number}
2468 This is a serial number to keep the labels distinct. The first
2469 @samp{0:} gets the number @samp{1}; The 15th @samp{0:} gets the
2470 number @samp{15}; @emph{etc.}. Likewise for the other labels @samp{1:}
2474 For instance, the first @code{1:} is named @code{L1@ctrl{A}1}, the 44th
2475 @code{3:} is named @code{L3@ctrl{A}44}.
2478 @section The Special Dot Symbol
2480 @cindex dot (symbol)
2481 @cindex @code{.} (symbol)
2482 @cindex current address
2483 @cindex location counter
2484 The special symbol @samp{.} refers to the current address that
2485 @code{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
2486 .long .} defines @code{melvin} to contain its own address.
2487 Assigning a value to @code{.} is treated the same as a @code{.org}
2488 directive. Thus, the expression @samp{.=.+4} is the same as saying
2489 @ifclear no-space-dir
2498 @node Symbol Attributes
2499 @section Symbol Attributes
2501 @cindex symbol attributes
2502 @cindex attributes, symbol
2503 Every symbol has, as well as its name, the attributes ``Value'' and
2504 ``Type''. Depending on output format, symbols can also have auxiliary
2507 The detailed definitions are in @file{a.out.h}.
2510 If you use a symbol without defining it, @code{@value{AS}} assumes zero for
2511 all these attributes, and probably won't warn you. This makes the
2512 symbol an externally defined symbol, which is generally what you
2516 * Symbol Value:: Value
2517 * Symbol Type:: Type
2520 * a.out Symbols:: Symbol Attributes: @code{a.out}
2524 * a.out Symbols:: Symbol Attributes: @code{a.out}
2527 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
2532 * COFF Symbols:: Symbol Attributes for COFF
2535 * SOM Symbols:: Symbol Attributes for SOM
2542 @cindex value of a symbol
2543 @cindex symbol value
2544 The value of a symbol is (usually) 32 bits. For a symbol which labels a
2545 location in the text, data, bss or absolute sections the value is the
2546 number of addresses from the start of that section to the label.
2547 Naturally for text, data and bss sections the value of a symbol changes
2548 as @code{@value{LD}} changes section base addresses during linking. Absolute
2549 symbols' values do not change during linking: that is why they are
2552 The value of an undefined symbol is treated in a special way. If it is
2553 0 then the symbol is not defined in this assembler source file, and
2554 @code{@value{LD}} tries to determine its value from other files linked into the
2555 same program. You make this kind of symbol simply by mentioning a symbol
2556 name without defining it. A non-zero value represents a @code{.comm}
2557 common declaration. The value is how much common storage to reserve, in
2558 bytes (addresses). The symbol refers to the first address of the
2564 @cindex type of a symbol
2566 The type attribute of a symbol contains relocation (section)
2567 information, any flag settings indicating that a symbol is external, and
2568 (optionally), other information for linkers and debuggers. The exact
2569 format depends on the object-code output format in use.
2574 @c The following avoids a "widow" subsection title. @group would be
2575 @c better if it were available outside examples.
2578 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
2580 @cindex @code{b.out} symbol attributes
2581 @cindex symbol attributes, @code{b.out}
2582 These symbol attributes appear only when @code{@value{AS}} is configured for
2583 one of the Berkeley-descended object output formats---@code{a.out} or
2589 @subsection Symbol Attributes: @code{a.out}
2591 @cindex @code{a.out} symbol attributes
2592 @cindex symbol attributes, @code{a.out}
2598 @subsection Symbol Attributes: @code{a.out}
2600 @cindex @code{a.out} symbol attributes
2601 @cindex symbol attributes, @code{a.out}
2605 * Symbol Desc:: Descriptor
2606 * Symbol Other:: Other
2610 @subsubsection Descriptor
2612 @cindex descriptor, of @code{a.out} symbol
2613 This is an arbitrary 16-bit value. You may establish a symbol's
2614 descriptor value by using a @code{.desc} statement
2615 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
2619 @subsubsection Other
2621 @cindex other attribute, of @code{a.out} symbol
2622 This is an arbitrary 8-bit value. It means nothing to @code{@value{AS}}.
2627 @subsection Symbol Attributes for COFF
2629 @cindex COFF symbol attributes
2630 @cindex symbol attributes, COFF
2632 The COFF format supports a multitude of auxiliary symbol attributes;
2633 like the primary symbol attributes, they are set between @code{.def} and
2634 @code{.endef} directives.
2636 @subsubsection Primary Attributes
2638 @cindex primary attributes, COFF symbols
2639 The symbol name is set with @code{.def}; the value and type,
2640 respectively, with @code{.val} and @code{.type}.
2642 @subsubsection Auxiliary Attributes
2644 @cindex auxiliary attributes, COFF symbols
2645 The @code{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
2646 @code{.size}, and @code{.tag} can generate auxiliary symbol table
2647 information for COFF.
2652 @subsection Symbol Attributes for SOM
2654 @cindex SOM symbol attributes
2655 @cindex symbol attributes, SOM
2657 The SOM format for the HPPA supports a multitude of symbol attributes set with
2658 the @code{.EXPORT} and @code{.IMPORT} directives.
2660 The attributes are described in @cite{HP9000 Series 800 Assembly
2661 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
2662 @code{EXPORT} assembler directive documentation.
2666 @chapter Expressions
2670 @cindex numeric values
2671 An @dfn{expression} specifies an address or numeric value.
2672 Whitespace may precede and/or follow an expression.
2674 The result of an expression must be an absolute number, or else an offset into
2675 a particular section. If an expression is not absolute, and there is not
2676 enough information when @code{@value{AS}} sees the expression to know its
2677 section, a second pass over the source program might be necessary to interpret
2678 the expression---but the second pass is currently not implemented.
2679 @code{@value{AS}} aborts with an error message in this situation.
2682 * Empty Exprs:: Empty Expressions
2683 * Integer Exprs:: Integer Expressions
2687 @section Empty Expressions
2689 @cindex empty expressions
2690 @cindex expressions, empty
2691 An empty expression has no value: it is just whitespace or null.
2692 Wherever an absolute expression is required, you may omit the
2693 expression, and @code{@value{AS}} assumes a value of (absolute) 0. This
2694 is compatible with other assemblers.
2697 @section Integer Expressions
2699 @cindex integer expressions
2700 @cindex expressions, integer
2701 An @dfn{integer expression} is one or more @emph{arguments} delimited
2702 by @emph{operators}.
2705 * Arguments:: Arguments
2706 * Operators:: Operators
2707 * Prefix Ops:: Prefix Operators
2708 * Infix Ops:: Infix Operators
2712 @subsection Arguments
2714 @cindex expression arguments
2715 @cindex arguments in expressions
2716 @cindex operands in expressions
2717 @cindex arithmetic operands
2718 @dfn{Arguments} are symbols, numbers or subexpressions. In other
2719 contexts arguments are sometimes called ``arithmetic operands''. In
2720 this manual, to avoid confusing them with the ``instruction operands'' of
2721 the machine language, we use the term ``argument'' to refer to parts of
2722 expressions only, reserving the word ``operand'' to refer only to machine
2723 instruction operands.
2725 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
2726 @var{section} is one of text, data, bss, absolute,
2727 or undefined. @var{NNN} is a signed, 2's complement 32 bit
2730 Numbers are usually integers.
2732 A number can be a flonum or bignum. In this case, you are warned
2733 that only the low order 32 bits are used, and @code{@value{AS}} pretends
2734 these 32 bits are an integer. You may write integer-manipulating
2735 instructions that act on exotic constants, compatible with other
2738 @cindex subexpressions
2739 Subexpressions are a left parenthesis @samp{(} followed by an integer
2740 expression, followed by a right parenthesis @samp{)}; or a prefix
2741 operator followed by an argument.
2744 @subsection Operators
2746 @cindex operators, in expressions
2747 @cindex arithmetic functions
2748 @cindex functions, in expressions
2749 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
2750 operators are followed by an argument. Infix operators appear
2751 between their arguments. Operators may be preceded and/or followed by
2755 @subsection Prefix Operator
2757 @cindex prefix operators
2758 @code{@value{AS}} has the following @dfn{prefix operators}. They each take
2759 one argument, which must be absolute.
2761 @c the tex/end tex stuff surrounding this small table is meant to make
2762 @c it align, on the printed page, with the similar table in the next
2763 @c section (which is inside an enumerate).
2765 \global\advance\leftskip by \itemindent
2770 @dfn{Negation}. Two's complement negation.
2772 @dfn{Complementation}. Bitwise not.
2776 \global\advance\leftskip by -\itemindent
2780 @subsection Infix Operators
2782 @cindex infix operators
2783 @cindex operators, permitted arguments
2784 @dfn{Infix operators} take two arguments, one on either side. Operators
2785 have precedence, but operations with equal precedence are performed left
2786 to right. Apart from @code{+} or @code{-}, both arguments must be
2787 absolute, and the result is absolute.
2790 @cindex operator precedence
2791 @cindex precedence of operators
2798 @dfn{Multiplication}.
2801 @dfn{Division}. Truncation is the same as the C operator @samp{/}
2808 @dfn{Shift Left}. Same as the C operator @samp{<<}.
2812 @dfn{Shift Right}. Same as the C operator @samp{>>}.
2816 Intermediate precedence
2821 @dfn{Bitwise Inclusive Or}.
2827 @dfn{Bitwise Exclusive Or}.
2830 @dfn{Bitwise Or Not}.
2837 @cindex addition, permitted arguments
2838 @cindex plus, permitted arguments
2839 @cindex arguments for addition
2841 @dfn{Addition}. If either argument is absolute, the result has the section of
2842 the other argument. You may not add together arguments from different
2845 @cindex subtraction, permitted arguments
2846 @cindex minus, permitted arguments
2847 @cindex arguments for subtraction
2849 @dfn{Subtraction}. If the right argument is absolute, the
2850 result has the section of the left argument.
2851 If both arguments are in the same section, the result is absolute.
2852 You may not subtract arguments from different sections.
2853 @c FIXME is there still something useful to say about undefined - undefined ?
2857 In short, it's only meaningful to add or subtract the @emph{offsets} in an
2858 address; you can only have a defined section in one of the two arguments.
2861 @chapter Assembler Directives
2863 @cindex directives, machine independent
2864 @cindex pseudo-ops, machine independent
2865 @cindex machine independent directives
2866 All assembler directives have names that begin with a period (@samp{.}).
2867 The rest of the name is letters, usually in lower case.
2869 This chapter discusses directives that are available regardless of the
2870 target machine configuration for the @sc{gnu} assembler.
2872 Some machine configurations provide additional directives.
2873 @xref{Machine Dependencies}.
2876 @ifset machine-directives
2877 @xref{Machine Dependencies} for additional directives.
2882 * Abort:: @code{.abort}
2884 * ABORT:: @code{.ABORT}
2887 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
2888 * App-File:: @code{.app-file @var{string}}
2889 * Ascii:: @code{.ascii "@var{string}"}@dots{}
2890 * Asciz:: @code{.asciz "@var{string}"}@dots{}
2891 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
2892 * Byte:: @code{.byte @var{expressions}}
2893 * Comm:: @code{.comm @var{symbol} , @var{length} }
2894 * Data:: @code{.data @var{subsection}}
2896 * Def:: @code{.def @var{name}}
2899 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
2905 * Double:: @code{.double @var{flonums}}
2906 * Eject:: @code{.eject}
2907 * Else:: @code{.else}
2909 * Endef:: @code{.endef}
2912 * Endif:: @code{.endif}
2913 * Equ:: @code{.equ @var{symbol}, @var{expression}}
2914 * Extern:: @code{.extern}
2915 @ifclear no-file-dir
2916 * File:: @code{.file @var{string}}
2919 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
2920 * Float:: @code{.float @var{flonums}}
2921 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
2922 * hword:: @code{.hword @var{expressions}}
2923 * Ident:: @code{.ident}
2924 * If:: @code{.if @var{absolute expression}}
2925 * Include:: @code{.include "@var{file}"}
2926 * Int:: @code{.int @var{expressions}}
2927 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
2928 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
2929 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
2930 * Lflags:: @code{.lflags}
2931 @ifclear no-line-dir
2932 * Line:: @code{.line @var{line-number}}
2935 * Ln:: @code{.ln @var{line-number}}
2936 * Linkonce:: @code{.linkonce [@var{type}]}
2937 * List:: @code{.list}
2938 * Long:: @code{.long @var{expressions}}
2940 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
2943 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
2944 * MRI:: @code{.mri @var{val}}
2946 * Nolist:: @code{.nolist}
2947 * Octa:: @code{.octa @var{bignums}}
2948 * Org:: @code{.org @var{new-lc} , @var{fill}}
2949 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
2950 * Psize:: @code{.psize @var{lines}, @var{columns}}
2951 * Quad:: @code{.quad @var{bignums}}
2952 * Rept:: @code{.rept @var{count}}
2953 * Sbttl:: @code{.sbttl "@var{subheading}"}
2955 * Scl:: @code{.scl @var{class}}
2958 * Section:: @code{.section @var{name}, @var{subsection}}
2961 * Set:: @code{.set @var{symbol}, @var{expression}}
2962 * Short:: @code{.short @var{expressions}}
2963 * Single:: @code{.single @var{flonums}}
2965 * Size:: @code{.size}
2968 * Skip:: @code{.skip @var{size} , @var{fill}}
2969 * Space:: @code{.space @var{size} , @var{fill}}
2971 * Stab:: @code{.stabd, .stabn, .stabs}
2974 * String:: @code{.string "@var{str}"}
2976 * Tag:: @code{.tag @var{structname}}
2979 * Text:: @code{.text @var{subsection}}
2980 * Title:: @code{.title "@var{heading}"}
2982 * Type:: @code{.type @var{int}}
2983 * Val:: @code{.val @var{addr}}
2986 * Word:: @code{.word @var{expressions}}
2987 * Deprecated:: Deprecated Directives
2991 @section @code{.abort}
2993 @cindex @code{abort} directive
2994 @cindex stopping the assembly
2995 This directive stops the assembly immediately. It is for
2996 compatibility with other assemblers. The original idea was that the
2997 assembly language source would be piped into the assembler. If the sender
2998 of the source quit, it could use this directive tells @code{@value{AS}} to
2999 quit also. One day @code{.abort} will not be supported.
3003 @section @code{.ABORT}
3005 @cindex @code{ABORT} directive
3006 When producing COFF output, @code{@value{AS}} accepts this directive as a
3007 synonym for @samp{.abort}.
3010 When producing @code{b.out} output, @code{@value{AS}} accepts this directive,
3016 @section @code{.align @var{abs-expr} , @var{abs-expr}}
3018 @cindex padding the location counter
3019 @cindex @code{align} directive
3020 Pad the location counter (in the current subsection) to a particular
3021 storage boundary. The first expression (which must be absolute) is the
3022 alignment required, as described below.
3023 The second expression (also absolute) gives the value to be stored in
3024 the padding bytes. It (and the comma) may be omitted. If it is
3025 omitted, the padding bytes are zero.
3026 For the alpha, if the section is marked as containing code and the
3027 padding expression is omitted, then the space is filled with no-ops.
3029 The way the required alignment is specified varies from system to system.
3030 For the a29k, hppa, m68k, m88k, w65, sparc, and Hitachi SH, and i386 using ELF
3032 the first expression is the
3033 alignment request in bytes. For example @samp{.align 8} advances
3034 the location counter until it is a multiple of 8. If the location counter
3035 is already a multiple of 8, no change is needed.
3037 For other systems, including the i386 using a.out format, it is the
3038 number of low-order zero bits the location counter must have after
3039 advancement. For example @samp{.align 3} advances the location
3040 counter until it a multiple of 8. If the location counter is already a
3041 multiple of 8, no change is needed.
3043 This inconsistency is due to the different behaviors of the various
3044 native assemblers for these systems which GAS must emulate.
3045 GAS also provides @code{.balign} and @code{.p2align} directives,
3046 described later, which have a consistent behavior across all
3047 architectures (but are specific to GAS).
3050 @section @code{.app-file @var{string}}
3052 @cindex logical file name
3053 @cindex file name, logical
3054 @cindex @code{app-file} directive
3056 @ifclear no-file-dir
3057 (which may also be spelled @samp{.file})
3059 tells @code{@value{AS}} that we are about to start a new
3060 logical file. @var{string} is the new file name. In general, the
3061 filename is recognized whether or not it is surrounded by quotes @samp{"};
3062 but if you wish to specify an empty file name is permitted,
3063 you must give the quotes--@code{""}. This statement may go away in
3064 future: it is only recognized to be compatible with old @code{@value{AS}}
3068 @section @code{.ascii "@var{string}"}@dots{}
3070 @cindex @code{ascii} directive
3071 @cindex string literals
3072 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3073 separated by commas. It assembles each string (with no automatic
3074 trailing zero byte) into consecutive addresses.
3077 @section @code{.asciz "@var{string}"}@dots{}
3079 @cindex @code{asciz} directive
3080 @cindex zero-terminated strings
3081 @cindex null-terminated strings
3082 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3083 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3086 @section @code{.balign[wl] @var{abs-expr} , @var{abs-expr}}
3088 @cindex padding the location counter given number of bytes
3089 @cindex @code{balign} directive
3090 Pad the location counter (in the current subsection) to a particular
3091 storage boundary. The first expression (which must be absolute) is the
3092 alignment request in bytes. For example @samp{.balign 8} advances
3093 the location counter until it is a multiple of 8. If the location counter
3094 is already a multiple of 8, no change is needed.
3096 The second expression (also absolute) gives the value to be stored in
3097 the padding bytes. It (and the comma) may be omitted. If it is
3098 omitted, the padding bytes are zero.
3100 @cindex @code{balignw} directive
3101 @cindex @code{balignl} directive
3102 The @code{.balignw} and @code{.balignl} directives are variants of the
3103 @code{.balign} directive. The @code{.balignw} directive treats the fill
3104 pattern as a two byte word value. The @code{.balignl} directives treats the
3105 fill pattern as a four byte longword value. For example, @code{.balignw
3106 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3107 filled in with the value 0x368d (the exact placement of the bytes depends upon
3108 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3112 @section @code{.byte @var{expressions}}
3114 @cindex @code{byte} directive
3115 @cindex integers, one byte
3116 @code{.byte} expects zero or more expressions, separated by commas.
3117 Each expression is assembled into the next byte.
3120 @section @code{.comm @var{symbol} , @var{length} }
3122 @cindex @code{comm} directive
3123 @cindex symbol, common
3124 @code{.comm} declares a named common area in the bss section. Normally
3125 @code{@value{LD}} reserves memory addresses for it during linking, so no partial
3126 program defines the location of the symbol. Use @code{.comm} to tell
3127 @code{@value{LD}} that it must be at least @var{length} bytes long. @code{@value{LD}}
3128 allocates space for each @code{.comm} symbol that is at least as
3129 long as the longest @code{.comm} request in any of the partial programs
3130 linked. @var{length} is an absolute expression.
3133 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
3134 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
3138 @section @code{.data @var{subsection}}
3140 @cindex @code{data} directive
3141 @code{.data} tells @code{@value{AS}} to assemble the following statements onto the
3142 end of the data subsection numbered @var{subsection} (which is an
3143 absolute expression). If @var{subsection} is omitted, it defaults
3148 @section @code{.def @var{name}}
3150 @cindex @code{def} directive
3151 @cindex COFF symbols, debugging
3152 @cindex debugging COFF symbols
3153 Begin defining debugging information for a symbol @var{name}; the
3154 definition extends until the @code{.endef} directive is encountered.
3157 This directive is only observed when @code{@value{AS}} is configured for COFF
3158 format output; when producing @code{b.out}, @samp{.def} is recognized,
3165 @section @code{.desc @var{symbol}, @var{abs-expression}}
3167 @cindex @code{desc} directive
3168 @cindex COFF symbol descriptor
3169 @cindex symbol descriptor, COFF
3170 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
3171 to the low 16 bits of an absolute expression.
3174 The @samp{.desc} directive is not available when @code{@value{AS}} is
3175 configured for COFF output; it is only for @code{a.out} or @code{b.out}
3176 object format. For the sake of compatibility, @code{@value{AS}} accepts
3177 it, but produces no output, when configured for COFF.
3183 @section @code{.dim}
3185 @cindex @code{dim} directive
3186 @cindex COFF auxiliary symbol information
3187 @cindex auxiliary symbol information, COFF
3188 This directive is generated by compilers to include auxiliary debugging
3189 information in the symbol table. It is only permitted inside
3190 @code{.def}/@code{.endef} pairs.
3193 @samp{.dim} is only meaningful when generating COFF format output; when
3194 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
3200 @section @code{.double @var{flonums}}
3202 @cindex @code{double} directive
3203 @cindex floating point numbers (double)
3204 @code{.double} expects zero or more flonums, separated by commas. It
3205 assembles floating point numbers.
3207 The exact kind of floating point numbers emitted depends on how
3208 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
3212 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
3213 in @sc{ieee} format.
3218 @section @code{.eject}
3220 @cindex @code{eject} directive
3221 @cindex new page, in listings
3222 @cindex page, in listings
3223 @cindex listing control: new page
3224 Force a page break at this point, when generating assembly listings.
3227 @section @code{.else}
3229 @cindex @code{else} directive
3230 @code{.else} is part of the @code{@value{AS}} support for conditional
3231 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
3232 of code to be assembled if the condition for the preceding @code{.if}
3236 @node End, Endef, Else, Pseudo Ops
3237 @section @code{.end}
3239 @cindex @code{end} directive
3240 This doesn't do anything---but isn't an s_ignore, so I suspect it's
3241 meant to do something eventually (which is why it isn't documented here
3242 as "for compatibility with blah").
3247 @section @code{.endef}
3249 @cindex @code{endef} directive
3250 This directive flags the end of a symbol definition begun with
3254 @samp{.endef} is only meaningful when generating COFF format output; if
3255 @code{@value{AS}} is configured to generate @code{b.out}, it accepts this
3256 directive but ignores it.
3261 @section @code{.endif}
3263 @cindex @code{endif} directive
3264 @code{.endif} is part of the @code{@value{AS}} support for conditional assembly;
3265 it marks the end of a block of code that is only assembled
3266 conditionally. @xref{If,,@code{.if}}.
3269 @section @code{.equ @var{symbol}, @var{expression}}
3271 @cindex @code{equ} directive
3272 @cindex assigning values to symbols
3273 @cindex symbols, assigning values to
3274 This directive sets the value of @var{symbol} to @var{expression}.
3275 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
3278 The syntax for @code{equ} on the HPPA is
3279 @samp{@var{symbol} .equ @var{expression}}.
3283 @section @code{.extern}
3285 @cindex @code{extern} directive
3286 @code{.extern} is accepted in the source program---for compatibility
3287 with other assemblers---but it is ignored. @code{@value{AS}} treats
3288 all undefined symbols as external.
3290 @ifclear no-file-dir
3292 @section @code{.file @var{string}}
3294 @cindex @code{file} directive
3295 @cindex logical file name
3296 @cindex file name, logical
3297 @code{.file} (which may also be spelled @samp{.app-file}) tells
3298 @code{@value{AS}} that we are about to start a new logical file.
3299 @var{string} is the new file name. In general, the filename is
3300 recognized whether or not it is surrounded by quotes @samp{"}; but if
3301 you wish to specify an empty file name, you must give the
3302 quotes--@code{""}. This statement may go away in future: it is only
3303 recognized to be compatible with old @code{@value{AS}} programs.
3305 In some configurations of @code{@value{AS}}, @code{.file} has already been
3306 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
3311 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
3313 @cindex @code{fill} directive
3314 @cindex writing patterns in memory
3315 @cindex patterns, writing in memory
3316 @var{result}, @var{size} and @var{value} are absolute expressions.
3317 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
3318 may be zero or more. @var{Size} may be zero or more, but if it is
3319 more than 8, then it is deemed to have the value 8, compatible with
3320 other people's assemblers. The contents of each @var{repeat} bytes
3321 is taken from an 8-byte number. The highest order 4 bytes are
3322 zero. The lowest order 4 bytes are @var{value} rendered in the
3323 byte-order of an integer on the computer @code{@value{AS}} is assembling for.
3324 Each @var{size} bytes in a repetition is taken from the lowest order
3325 @var{size} bytes of this number. Again, this bizarre behavior is
3326 compatible with other people's assemblers.
3328 @var{size} and @var{value} are optional.
3329 If the second comma and @var{value} are absent, @var{value} is
3330 assumed zero. If the first comma and following tokens are absent,
3331 @var{size} is assumed to be 1.
3334 @section @code{.float @var{flonums}}
3336 @cindex floating point numbers (single)
3337 @cindex @code{float} directive
3338 This directive assembles zero or more flonums, separated by commas. It
3339 has the same effect as @code{.single}.
3341 The exact kind of floating point numbers emitted depends on how
3342 @code{@value{AS}} is configured.
3343 @xref{Machine Dependencies}.
3347 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
3348 in @sc{ieee} format.
3353 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3355 @cindex @code{global} directive
3356 @cindex symbol, making visible to linker
3357 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
3358 @var{symbol} in your partial program, its value is made available to
3359 other partial programs that are linked with it. Otherwise,
3360 @var{symbol} takes its attributes from a symbol of the same name
3361 from another file linked into the same program.
3363 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
3364 compatibility with other assemblers.
3367 On the HPPA, @code{.global} is not always enough to make it accessible to other
3368 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
3369 @xref{HPPA Directives,, HPPA Assembler Directives}.
3373 @section @code{.hword @var{expressions}}
3375 @cindex @code{hword} directive
3376 @cindex integers, 16-bit
3377 @cindex numbers, 16-bit
3378 @cindex sixteen bit integers
3379 This expects zero or more @var{expressions}, and emits
3380 a 16 bit number for each.
3383 This directive is a synonym for @samp{.short}; depending on the target
3384 architecture, it may also be a synonym for @samp{.word}.
3388 This directive is a synonym for @samp{.short}.
3391 This directive is a synonym for both @samp{.short} and @samp{.word}.
3396 @section @code{.ident}
3398 @cindex @code{ident} directive
3399 This directive is used by some assemblers to place tags in object files.
3400 @code{@value{AS}} simply accepts the directive for source-file
3401 compatibility with such assemblers, but does not actually emit anything
3405 @section @code{.if @var{absolute expression}}
3407 @cindex conditional assembly
3408 @cindex @code{if} directive
3409 @code{.if} marks the beginning of a section of code which is only
3410 considered part of the source program being assembled if the argument
3411 (which must be an @var{absolute expression}) is non-zero. The end of
3412 the conditional section of code must be marked by @code{.endif}
3413 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
3414 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
3416 The following variants of @code{.if} are also supported:
3418 @cindex @code{ifdef} directive
3419 @item .ifdef @var{symbol}
3420 Assembles the following section of code if the specified @var{symbol}
3424 @cindex @code{ifeqs} directive
3426 Not yet implemented.
3429 @cindex @code{ifndef} directive
3430 @cindex @code{ifnotdef} directive
3431 @item .ifndef @var{symbol}
3432 @itemx .ifnotdef @var{symbol}
3433 Assembles the following section of code if the specified @var{symbol}
3434 has not been defined. Both spelling variants are equivalent.
3438 Not yet implemented.
3443 @section @code{.include "@var{file}"}
3445 @cindex @code{include} directive
3446 @cindex supporting files, including
3447 @cindex files, including
3448 This directive provides a way to include supporting files at specified
3449 points in your source program. The code from @var{file} is assembled as
3450 if it followed the point of the @code{.include}; when the end of the
3451 included file is reached, assembly of the original file continues. You
3452 can control the search paths used with the @samp{-I} command-line option
3453 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
3457 @section @code{.int @var{expressions}}
3459 @cindex @code{int} directive
3460 @cindex integers, 32-bit
3461 Expect zero or more @var{expressions}, of any section, separated by commas.
3462 For each expression, emit a number that, at run time, is the value of that
3463 expression. The byte order and bit size of the number depends on what kind
3464 of target the assembly is for.
3468 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
3469 integers. On the H8/300H and the Hitachi SH, however, @code{.int} emits
3475 @section @code{.irp @var{symbol},@var{values}}@dots{}
3477 @cindex @code{irp} directive
3478 Evaluate a sequence of statements assigning different values to @var{symbol}.
3479 The sequence of statements starts at the @code{.irp} directive, and is
3480 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
3481 set to @var{value}, and the sequence of statements is assembled. If no
3482 @var{value} is listed, the sequence of statements is assembled once, with
3483 @var{symbol} set to the null string. To refer to @var{symbol} within the
3484 sequence of statements, use @var{\symbol}.
3486 For example, assembling
3494 is equivalent to assembling
3503 @section @code{.irpc @var{symbol},@var{values}}@dots{}
3505 @cindex @code{irpc} directive
3506 Evaluate a sequence of statements assigning different values to @var{symbol}.
3507 The sequence of statements starts at the @code{.irpc} directive, and is
3508 terminated by an @code{.endr} directive. For each character in @var{value},
3509 @var{symbol} is set to the character, and the sequence of statements is
3510 assembled. If no @var{value} is listed, the sequence of statements is
3511 assembled once, with @var{symbol} set to the null string. To refer to
3512 @var{symbol} within the sequence of statements, use @var{\symbol}.
3514 For example, assembling
3522 is equivalent to assembling
3531 @section @code{.lcomm @var{symbol} , @var{length}}
3533 @cindex @code{lcomm} directive
3534 @cindex local common symbols
3535 @cindex symbols, local common
3536 Reserve @var{length} (an absolute expression) bytes for a local common
3537 denoted by @var{symbol}. The section and value of @var{symbol} are
3538 those of the new local common. The addresses are allocated in the bss
3539 section, so that at run-time the bytes start off zeroed. @var{Symbol}
3540 is not declared global (@pxref{Global,,@code{.global}}), so is normally
3541 not visible to @code{@value{LD}}.
3544 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
3545 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
3549 @section @code{.lflags}
3551 @cindex @code{lflags} directive (ignored)
3552 @code{@value{AS}} accepts this directive, for compatibility with other
3553 assemblers, but ignores it.
3555 @ifclear no-line-dir
3557 @section @code{.line @var{line-number}}
3559 @cindex @code{line} directive
3563 @section @code{.ln @var{line-number}}
3565 @cindex @code{ln} directive
3567 @cindex logical line number
3569 Change the logical line number. @var{line-number} must be an absolute
3570 expression. The next line has that logical line number. Therefore any other
3571 statements on the current line (after a statement separator character) are
3572 reported as on logical line number @var{line-number} @minus{} 1. One day
3573 @code{@value{AS}} will no longer support this directive: it is recognized only
3574 for compatibility with existing assembler programs.
3578 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
3579 not available; use the synonym @code{.ln} in that context.
3584 @ifclear no-line-dir
3585 Even though this is a directive associated with the @code{a.out} or
3586 @code{b.out} object-code formats, @code{@value{AS}} still recognizes it
3587 when producing COFF output, and treats @samp{.line} as though it
3588 were the COFF @samp{.ln} @emph{if} it is found outside a
3589 @code{.def}/@code{.endef} pair.
3591 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
3592 used by compilers to generate auxiliary symbol information for
3597 @section @code{.linkonce [@var{type}]}
3599 @cindex @code{linkonce} directive
3600 @cindex common sections
3601 Mark the current section so that the linker only includes a single copy of it.
3602 This may be used to include the same section in several different object files,
3603 but ensure that the linker will only include it once in the final output file.
3604 The @code{.linkonce} pseudo-op must be used for each instance of the section.
3605 Duplicate sections are detected based on the section name, so it should be
3608 This directive is only supported by a few object file formats; as of this
3609 writing, the only object file format which supports it is the Portable
3610 Executable format used on Windows NT.
3612 The @var{type} argument is optional. If specified, it must be one of the
3613 following strings. For example:
3617 Not all types may be supported on all object file formats.
3621 Silently discard duplicate sections. This is the default.
3624 Warn if there are duplicate sections, but still keep only one copy.
3627 Warn if any of the duplicates have different sizes.
3630 Warn if any of the duplicates do not have exactly the same contents.
3634 @section @code{.ln @var{line-number}}
3636 @cindex @code{ln} directive
3637 @ifclear no-line-dir
3638 @samp{.ln} is a synonym for @samp{.line}.
3641 Tell @code{@value{AS}} to change the logical line number. @var{line-number}
3642 must be an absolute expression. The next line has that logical
3643 line number, so any other statements on the current line (after a
3644 statement separator character @code{;}) are reported as on logical
3645 line number @var{line-number} @minus{} 1.
3648 This directive is accepted, but ignored, when @code{@value{AS}} is
3649 configured for @code{b.out}; its effect is only associated with COFF
3655 @section @code{.mri @var{val}}
3657 @cindex @code{mri} directive
3658 @cindex MRI mode, temporarily
3659 If @var{val} is non-zero, this tells @code{@value{AS}} to enter MRI mode. If
3660 @var{val} is zero, this tells @code{@value{AS}} to exit MRI mode. This change
3661 affects code assembled until the next @code{.mri} directive, or until the end
3662 of the file. @xref{M, MRI mode, MRI mode}.
3665 @section @code{.list}
3667 @cindex @code{list} directive
3668 @cindex listing control, turning on
3669 Control (in conjunction with the @code{.nolist} directive) whether or
3670 not assembly listings are generated. These two directives maintain an
3671 internal counter (which is zero initially). @code{.list} increments the
3672 counter, and @code{.nolist} decrements it. Assembly listings are
3673 generated whenever the counter is greater than zero.
3675 By default, listings are disabled. When you enable them (with the
3676 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
3677 the initial value of the listing counter is one.
3680 @section @code{.long @var{expressions}}
3682 @cindex @code{long} directive
3683 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
3686 @c no one seems to know what this is for or whether this description is
3687 @c what it really ought to do
3689 @section @code{.lsym @var{symbol}, @var{expression}}
3691 @cindex @code{lsym} directive
3692 @cindex symbol, not referenced in assembly
3693 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
3694 the hash table, ensuring it cannot be referenced by name during the
3695 rest of the assembly. This sets the attributes of the symbol to be
3696 the same as the expression value:
3698 @var{other} = @var{descriptor} = 0
3699 @var{type} = @r{(section of @var{expression})}
3700 @var{value} = @var{expression}
3703 The new symbol is not flagged as external.
3707 @section @code{.macro}
3710 The commands @code{.macro} and @code{.endm} allow you to define macros that
3711 generate assembly output. For example, this definition specifies a macro
3712 @code{sum} that puts a sequence of numbers into memory:
3715 .macro sum from=0, to=5
3724 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
3736 @item .macro @var{macname}
3737 @itemx .macro @var{macname} @var{macargs} @dots{}
3738 @cindex @code{macro} directive
3739 Begin the definition of a macro called @var{macname}. If your macro
3740 definition requires arguments, specify their names after the macro name,
3741 separated by commas or spaces. You can supply a default value for any
3742 macro argument by following the name with @samp{=@var{deflt}}. For
3743 example, these are all valid @code{.macro} statements:
3747 Begin the definition of a macro called @code{comm}, which takes no
3750 @item .macro plus1 p, p1
3751 @itemx .macro plus1 p p1
3752 Either statement begins the definition of a macro called @code{plus1},
3753 which takes two arguments; within the macro definition, write
3754 @samp{\p} or @samp{\p1} to evaluate the arguments.
3756 @item .macro reserve_str p1=0 p2
3757 Begin the definition of a macro called @code{reserve_str}, with two
3758 arguments. The first argument has a default value, but not the second.
3759 After the definition is complete, you can call the macro either as
3760 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
3761 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
3762 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
3763 @samp{0}, and @samp{\p2} evaluating to @var{b}).
3766 When you call a macro, you can specify the argument values either by
3767 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
3768 @samp{sum to=17, from=9}.
3771 @cindex @code{endm} directive
3772 Mark the end of a macro definition.
3775 @cindex @code{exitm} directive
3776 Exit early from the current macro definition.
3778 @cindex number of macros executed
3779 @cindex macros, count executed
3781 @code{@value{AS}} maintains a counter of how many macros it has
3782 executed in this pseudo-variable; you can copy that number to your
3783 output with @samp{\@@}, but @emph{only within a macro definition}.
3786 @item LOCAL @var{name} [ , @dots{} ]
3787 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
3788 macro syntax'' with @samp{-a} or @samp{--alternate}.} @xref{Alternate,,
3789 Alternate macro syntax}.
3791 Generate a string replacement for each of the @var{name} arguments, and
3792 replace any instances of @var{name} in each macro expansion. The
3793 replacement string is unique in the assembly, and different for each
3794 separate macro expansion. @code{LOCAL} allows you to write macros that
3795 define symbols, without fear of conflict between separate macro expansions.
3800 @section @code{.nolist}
3802 @cindex @code{nolist} directive
3803 @cindex listing control, turning off
3804 Control (in conjunction with the @code{.list} directive) whether or
3805 not assembly listings are generated. These two directives maintain an
3806 internal counter (which is zero initially). @code{.list} increments the
3807 counter, and @code{.nolist} decrements it. Assembly listings are
3808 generated whenever the counter is greater than zero.
3811 @section @code{.octa @var{bignums}}
3813 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
3814 @cindex @code{octa} directive
3815 @cindex integer, 16-byte
3816 @cindex sixteen byte integer
3817 This directive expects zero or more bignums, separated by commas. For each
3818 bignum, it emits a 16-byte integer.
3820 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
3821 hence @emph{octa}-word for 16 bytes.
3824 @section @code{.org @var{new-lc} , @var{fill}}
3826 @cindex @code{org} directive
3827 @cindex location counter, advancing
3828 @cindex advancing location counter
3829 @cindex current address, advancing
3830 Advance the location counter of the current section to
3831 @var{new-lc}. @var{new-lc} is either an absolute expression or an
3832 expression with the same section as the current subsection. That is,
3833 you can't use @code{.org} to cross sections: if @var{new-lc} has the
3834 wrong section, the @code{.org} directive is ignored. To be compatible
3835 with former assemblers, if the section of @var{new-lc} is absolute,
3836 @code{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
3837 is the same as the current subsection.
3839 @code{.org} may only increase the location counter, or leave it
3840 unchanged; you cannot use @code{.org} to move the location counter
3843 @c double negative used below "not undefined" because this is a specific
3844 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
3845 @c section. doc@cygnus.com 18feb91
3846 Because @code{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
3847 may not be undefined. If you really detest this restriction we eagerly await
3848 a chance to share your improved assembler.
3850 Beware that the origin is relative to the start of the section, not
3851 to the start of the subsection. This is compatible with other
3852 people's assemblers.
3854 When the location counter (of the current subsection) is advanced, the
3855 intervening bytes are filled with @var{fill} which should be an
3856 absolute expression. If the comma and @var{fill} are omitted,
3857 @var{fill} defaults to zero.
3860 @section @code{.p2align[wl] @var{abs-expr} , @var{abs-expr}}
3862 @cindex padding the location counter given a power of two
3863 @cindex @code{p2align} directive
3864 Pad the location counter (in the current subsection) to a particular
3865 storage boundary. The first expression (which must be absolute) is the
3866 number of low-order zero bits the location counter must have after
3867 advancement. For example @samp{.p2align 3} advances the location
3868 counter until it a multiple of 8. If the location counter is already a
3869 multiple of 8, no change is needed.
3871 The second expression (also absolute) gives the value to be stored in
3872 the padding bytes. It (and the comma) may be omitted. If it is
3873 omitted, the padding bytes are zero.
3875 @cindex @code{p2alignw} directive
3876 @cindex @code{p2alignl} directive
3877 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
3878 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
3879 pattern as a two byte word value. The @code{.p2alignl} directives treats the
3880 fill pattern as a four byte longword value. For example, @code{.p2alignw
3881 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3882 filled in with the value 0x368d (the exact placement of the bytes depends upon
3883 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3887 @section @code{.psize @var{lines} , @var{columns}}
3889 @cindex @code{psize} directive
3890 @cindex listing control: paper size
3891 @cindex paper size, for listings
3892 Use this directive to declare the number of lines---and, optionally, the
3893 number of columns---to use for each page, when generating listings.
3895 If you do not use @code{.psize}, listings use a default line-count
3896 of 60. You may omit the comma and @var{columns} specification; the
3897 default width is 200 columns.
3899 @code{@value{AS}} generates formfeeds whenever the specified number of
3900 lines is exceeded (or whenever you explicitly request one, using
3903 If you specify @var{lines} as @code{0}, no formfeeds are generated save
3904 those explicitly specified with @code{.eject}.
3907 @section @code{.quad @var{bignums}}
3909 @cindex @code{quad} directive
3910 @code{.quad} expects zero or more bignums, separated by commas. For
3911 each bignum, it emits
3913 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
3914 warning message; and just takes the lowest order 8 bytes of the bignum.
3915 @cindex eight-byte integer
3916 @cindex integer, 8-byte
3918 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
3919 hence @emph{quad}-word for 8 bytes.
3922 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
3923 warning message; and just takes the lowest order 16 bytes of the bignum.
3924 @cindex sixteen-byte integer
3925 @cindex integer, 16-byte
3929 @section @code{.rept @var{count}}
3931 @cindex @code{rept} directive
3932 Repeat the sequence of lines between the @code{.rept} directive and the next
3933 @code{.endr} directive @var{count} times.
3935 For example, assembling
3943 is equivalent to assembling
3952 @section @code{.sbttl "@var{subheading}"}
3954 @cindex @code{sbttl} directive
3955 @cindex subtitles for listings
3956 @cindex listing control: subtitle
3957 Use @var{subheading} as the title (third line, immediately after the
3958 title line) when generating assembly listings.
3960 This directive affects subsequent pages, as well as the current page if
3961 it appears within ten lines of the top of a page.
3965 @section @code{.scl @var{class}}
3967 @cindex @code{scl} directive
3968 @cindex symbol storage class (COFF)
3969 @cindex COFF symbol storage class
3970 Set the storage-class value for a symbol. This directive may only be
3971 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
3972 whether a symbol is static or external, or it may record further
3973 symbolic debugging information.
3976 The @samp{.scl} directive is primarily associated with COFF output; when
3977 configured to generate @code{b.out} output format, @code{@value{AS}}
3978 accepts this directive but ignores it.
3983 @section @code{.section @var{name}}
3985 @cindex @code{section} directive
3986 @cindex named section
3987 Use the @code{.section} directive to assemble the following code into a section
3990 This directive is only supported for targets that actually support arbitrarily
3991 named sections; on @code{a.out} targets, for example, it is not accepted, even
3992 with a standard @code{a.out} section name.
3995 For COFF targets, the @code{.section} directive is used in one of the following
3998 .section @var{name}[, "@var{flags}"]
3999 .section @var{name}[, @var{subsegment}]
4002 If the optional argument is quoted, it is taken as flags to use for the
4003 section. Each flag is a single character. The following flags are recognized:
4006 bss section (uninitialized data)
4008 section is not loaded
4017 If no flags are specified, the default flags depend upon the section name. If
4018 the section name is not recognized, the default will be for the section to be
4019 loaded and writable.
4021 If the optional argument to the @code{.section} directive is not quoted, it is
4022 taken as a subsegment number (@pxref{Sub-Sections}).
4026 For ELF targets, the @code{.section} directive is used like this:
4028 .section @var{name}[, "@var{flags}"[, @@@var{type}]]
4030 The optional @var{flags} argument is a quoted string which may contain any
4031 combintion of the following characters:
4034 section is allocatable
4038 section is executable
4041 The optional @var{type} argument may contain one of the following constants:
4044 section contains data
4046 section does not contain data (i.e., section only occupies space)
4049 If no flags are specified, the default flags depend upon the section name. If
4050 the section name is not recognized, the default will be for the section to have
4051 none of the above flags: it will not be allocated in memory, nor writable, nor
4052 executable. The section will contain data.
4054 For ELF targets, the assembler supports another type of @code{.section}
4055 directive for compatibility with the Solaris assembler:
4057 .section "@var{name}"[, @var{flags}...]
4059 Note that the section name is quoted. There may be a sequence of comma
4063 section is allocatable
4067 section is executable
4072 @section @code{.set @var{symbol}, @var{expression}}
4074 @cindex @code{set} directive
4075 @cindex symbol value, setting
4076 Set the value of @var{symbol} to @var{expression}. This
4077 changes @var{symbol}'s value and type to conform to
4078 @var{expression}. If @var{symbol} was flagged as external, it remains
4079 flagged. (@xref{Symbol Attributes}.)
4081 You may @code{.set} a symbol many times in the same assembly.
4083 If you @code{.set} a global symbol, the value stored in the object
4084 file is the last value stored into it.
4087 The syntax for @code{set} on the HPPA is
4088 @samp{@var{symbol} .set @var{expression}}.
4092 @section @code{.short @var{expressions}}
4094 @cindex @code{short} directive
4096 @code{.short} is normally the same as @samp{.word}.
4097 @xref{Word,,@code{.word}}.
4099 In some configurations, however, @code{.short} and @code{.word} generate
4100 numbers of different lengths; @pxref{Machine Dependencies}.
4104 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
4107 This expects zero or more @var{expressions}, and emits
4108 a 16 bit number for each.
4113 @section @code{.single @var{flonums}}
4115 @cindex @code{single} directive
4116 @cindex floating point numbers (single)
4117 This directive assembles zero or more flonums, separated by commas. It
4118 has the same effect as @code{.float}.
4120 The exact kind of floating point numbers emitted depends on how
4121 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
4125 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
4126 numbers in @sc{ieee} format.
4132 @section @code{.size}
4134 @cindex @code{size} directive
4135 This directive is generated by compilers to include auxiliary debugging
4136 information in the symbol table. It is only permitted inside
4137 @code{.def}/@code{.endef} pairs.
4140 @samp{.size} is only meaningful when generating COFF format output; when
4141 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4146 @ifclear no-space-dir
4148 @section @code{.skip @var{size} , @var{fill}}
4150 @cindex @code{skip} directive
4151 @cindex filling memory
4152 This directive emits @var{size} bytes, each of value @var{fill}. Both
4153 @var{size} and @var{fill} are absolute expressions. If the comma and
4154 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
4158 @section @code{.space @var{size} , @var{fill}}
4160 @cindex @code{space} directive
4161 @cindex filling memory
4162 This directive emits @var{size} bytes, each of value @var{fill}. Both
4163 @var{size} and @var{fill} are absolute expressions. If the comma
4164 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
4169 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
4170 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
4171 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
4172 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
4181 @section @code{.space}
4182 @cindex @code{space} directive
4184 On the AMD 29K, this directive is ignored; it is accepted for
4185 compatibility with other AMD 29K assemblers.
4188 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
4189 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
4195 @section @code{.stabd, .stabn, .stabs}
4197 @cindex symbolic debuggers, information for
4198 @cindex @code{stab@var{x}} directives
4199 There are three directives that begin @samp{.stab}.
4200 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
4201 The symbols are not entered in the @code{@value{AS}} hash table: they
4202 cannot be referenced elsewhere in the source file.
4203 Up to five fields are required:
4207 This is the symbol's name. It may contain any character except
4208 @samp{\000}, so is more general than ordinary symbol names. Some
4209 debuggers used to code arbitrarily complex structures into symbol names
4213 An absolute expression. The symbol's type is set to the low 8 bits of
4214 this expression. Any bit pattern is permitted, but @code{@value{LD}}
4215 and debuggers choke on silly bit patterns.
4218 An absolute expression. The symbol's ``other'' attribute is set to the
4219 low 8 bits of this expression.
4222 An absolute expression. The symbol's descriptor is set to the low 16
4223 bits of this expression.
4226 An absolute expression which becomes the symbol's value.
4229 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
4230 or @code{.stabs} statement, the symbol has probably already been created;
4231 you get a half-formed symbol in your object file. This is
4232 compatible with earlier assemblers!
4235 @cindex @code{stabd} directive
4236 @item .stabd @var{type} , @var{other} , @var{desc}
4238 The ``name'' of the symbol generated is not even an empty string.
4239 It is a null pointer, for compatibility. Older assemblers used a
4240 null pointer so they didn't waste space in object files with empty
4243 The symbol's value is set to the location counter,
4244 relocatably. When your program is linked, the value of this symbol
4245 is the address of the location counter when the @code{.stabd} was
4248 @cindex @code{stabn} directive
4249 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
4250 The name of the symbol is set to the empty string @code{""}.
4252 @cindex @code{stabs} directive
4253 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
4254 All five fields are specified.
4260 @section @code{.string} "@var{str}"
4262 @cindex string, copying to object file
4263 @cindex @code{string} directive
4265 Copy the characters in @var{str} to the object file. You may specify more than
4266 one string to copy, separated by commas. Unless otherwise specified for a
4267 particular machine, the assembler marks the end of each string with a 0 byte.
4268 You can use any of the escape sequences described in @ref{Strings,,Strings}.
4272 @section @code{.tag @var{structname}}
4274 @cindex COFF structure debugging
4275 @cindex structure debugging, COFF
4276 @cindex @code{tag} directive
4277 This directive is generated by compilers to include auxiliary debugging
4278 information in the symbol table. It is only permitted inside
4279 @code{.def}/@code{.endef} pairs. Tags are used to link structure
4280 definitions in the symbol table with instances of those structures.
4283 @samp{.tag} is only used when generating COFF format output; when
4284 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4290 @section @code{.text @var{subsection}}
4292 @cindex @code{text} directive
4293 Tells @code{@value{AS}} to assemble the following statements onto the end of
4294 the text subsection numbered @var{subsection}, which is an absolute
4295 expression. If @var{subsection} is omitted, subsection number zero
4299 @section @code{.title "@var{heading}"}
4301 @cindex @code{title} directive
4302 @cindex listing control: title line
4303 Use @var{heading} as the title (second line, immediately after the
4304 source file name and pagenumber) when generating assembly listings.
4306 This directive affects subsequent pages, as well as the current page if
4307 it appears within ten lines of the top of a page.
4311 @section @code{.type @var{int}}
4313 @cindex COFF symbol type
4314 @cindex symbol type, COFF
4315 @cindex @code{type} directive
4316 This directive, permitted only within @code{.def}/@code{.endef} pairs,
4317 records the integer @var{int} as the type attribute of a symbol table entry.
4320 @samp{.type} is associated only with COFF format output; when
4321 @code{@value{AS}} is configured for @code{b.out} output, it accepts this
4322 directive but ignores it.
4328 @section @code{.val @var{addr}}
4330 @cindex @code{val} directive
4331 @cindex COFF value attribute
4332 @cindex value attribute, COFF
4333 This directive, permitted only within @code{.def}/@code{.endef} pairs,
4334 records the address @var{addr} as the value attribute of a symbol table
4338 @samp{.val} is used only for COFF output; when @code{@value{AS}} is
4339 configured for @code{b.out}, it accepts this directive but ignores it.
4344 @section @code{.word @var{expressions}}
4346 @cindex @code{word} directive
4347 This directive expects zero or more @var{expressions}, of any section,
4348 separated by commas.
4351 For each expression, @code{@value{AS}} emits a 32-bit number.
4354 For each expression, @code{@value{AS}} emits a 16-bit number.
4359 The size of the number emitted, and its byte order,
4360 depend on what target computer the assembly is for.
4363 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
4364 @c happen---32-bit addressability, period; no long/short jumps.
4365 @ifset DIFF-TBL-KLUGE
4366 @cindex difference tables altered
4367 @cindex altered difference tables
4369 @emph{Warning: Special Treatment to support Compilers}
4373 Machines with a 32-bit address space, but that do less than 32-bit
4374 addressing, require the following special treatment. If the machine of
4375 interest to you does 32-bit addressing (or doesn't require it;
4376 @pxref{Machine Dependencies}), you can ignore this issue.
4379 In order to assemble compiler output into something that works,
4380 @code{@value{AS}} occasionlly does strange things to @samp{.word} directives.
4381 Directives of the form @samp{.word sym1-sym2} are often emitted by
4382 compilers as part of jump tables. Therefore, when @code{@value{AS}} assembles a
4383 directive of the form @samp{.word sym1-sym2}, and the difference between
4384 @code{sym1} and @code{sym2} does not fit in 16 bits, @code{@value{AS}}
4385 creates a @dfn{secondary jump table}, immediately before the next label.
4386 This secondary jump table is preceded by a short-jump to the
4387 first byte after the secondary table. This short-jump prevents the flow
4388 of control from accidentally falling into the new table. Inside the
4389 table is a long-jump to @code{sym2}. The original @samp{.word}
4390 contains @code{sym1} minus the address of the long-jump to
4393 If there were several occurrences of @samp{.word sym1-sym2} before the
4394 secondary jump table, all of them are adjusted. If there was a
4395 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
4396 long-jump to @code{sym4} is included in the secondary jump table,
4397 and the @code{.word} directives are adjusted to contain @code{sym3}
4398 minus the address of the long-jump to @code{sym4}; and so on, for as many
4399 entries in the original jump table as necessary.
4402 @emph{This feature may be disabled by compiling @code{@value{AS}} with the
4403 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
4404 assembly language programmers.
4407 @c end DIFF-TBL-KLUGE
4410 @section Deprecated Directives
4412 @cindex deprecated directives
4413 @cindex obsolescent directives
4414 One day these directives won't work.
4415 They are included for compatibility with older assemblers.
4423 @node Machine Dependencies
4424 @chapter Machine Dependent Features
4426 @cindex machine dependencies
4427 The machine instruction sets are (almost by definition) different on
4428 each machine where @code{@value{AS}} runs. Floating point representations
4429 vary as well, and @code{@value{AS}} often supports a few additional
4430 directives or command-line options for compatibility with other
4431 assemblers on a particular platform. Finally, some versions of
4432 @code{@value{AS}} support special pseudo-instructions for branch
4435 This chapter discusses most of these differences, though it does not
4436 include details on any machine's instruction set. For details on that
4437 subject, see the hardware manufacturer's manual.
4441 * AMD29K-Dependent:: AMD 29K Dependent Features
4443 @c start-sanitize-arc
4445 * ARC-Dependent:: ARC Dependent Features
4448 @c start-sanitize-d10v
4450 * D10V-Dependent:: D10V Dependent Features
4452 @c end-sanitize-d10v
4454 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
4457 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
4460 * HPPA-Dependent:: HPPA Dependent Features
4463 * i386-Dependent:: Intel 80386 Dependent Features
4466 * i960-Dependent:: Intel 80960 Dependent Features
4469 * M68K-Dependent:: M680x0 Dependent Features
4472 * MIPS-Dependent:: MIPS Dependent Features
4475 * SH-Dependent:: Hitachi SH Dependent Features
4478 * Sparc-Dependent:: SPARC Dependent Features
4481 * Z8000-Dependent:: Z8000 Dependent Features
4484 * Vax-Dependent:: VAX Dependent Features
4491 @c The following major nodes are *sections* in the GENERIC version, *chapters*
4492 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
4493 @c peculiarity: to preserve cross-references, there must be a node called
4494 @c "Machine Dependencies". Hence the conditional nodenames in each
4495 @c major node below. Node defaulting in makeinfo requires adjacency of
4496 @c node and sectioning commands; hence the repetition of @chapter BLAH
4497 @c in both conditional blocks.
4499 @c start-sanitize-arc
4504 @chapter ARC Dependent Features
4507 @node Machine Dependencies
4508 @chapter ARC Dependent Features
4513 * ARC-Opts:: Options
4514 * ARC-Float:: Floating Point
4515 * ARC-Directives:: Sparc Machine Directives
4521 @cindex options for ARC
4523 @cindex architectures, ARC
4524 @cindex ARC architectures
4525 The ARC chip family includes several successive levels (or other
4526 variants) of chip, using the same core instruction set, but including
4527 a few additional instructions at each level.
4529 By default, @code{@value{AS}} assumes the core instruction set (ARC
4530 base). The @code{.cpu} pseudo-op is used to select a different variant.
4533 @cindex @code{-mbig-endian} option (ARC)
4534 @cindex @code{-mlittle-endian} option (ARC)
4535 @cindex ARC big-endian output
4536 @cindex ARC little-endian output
4537 @cindex big-endian output, ARC
4538 @cindex little-endian output, ARC
4540 @itemx -mlittle-endian
4541 Any @sc{arc} configuration of @code{@value{AS}} can select big-endian or
4542 little-endian output at run time (unlike most other @sc{gnu} development
4543 tools, which must be configured for one or the other). Use
4544 @samp{-mbig-endian} to select big-endian output, and @samp{-mlittle-endian}
4549 @section Floating Point
4551 @cindex floating point, ARC (@sc{ieee})
4552 @cindex ARC floating point (@sc{ieee})
4553 The ARC cpu family currently does not have hardware floating point
4554 support. Software floating point support is provided by @code{GCC}
4555 and uses @sc{ieee} floating-point numbers.
4557 @node ARC-Directives
4558 @section ARC Machine Directives
4560 @cindex ARC machine directives
4561 @cindex machine directives, ARC
4562 The ARC version of @code{@value{AS}} supports the following additional
4567 @cindex @code{cpu} directive, SPARC
4568 This must be followed by the desired cpu. It must be one of
4569 @code{base}, @code{host}, @code{graphics}, or @code{audio}.
4577 @include c-a29k.texi
4582 @node Machine Dependencies
4583 @chapter Machine Dependent Features
4585 The machine instruction sets are different on each Hitachi chip family,
4586 and there are also some syntax differences among the families. This
4587 chapter describes the specific @code{@value{AS}} features for each
4591 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
4592 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
4593 * SH-Dependent:: Hitachi SH Dependent Features
4599 @c start-sanitize-d10v
4601 @include c-d10v.texi
4603 @c end-sanitize-d10v
4606 @include c-h8300.texi
4610 @include c-h8500.texi
4614 @include c-hppa.texi
4618 @include c-i386.texi
4622 @include c-i960.texi
4626 @include c-m68k.texi
4630 @include c-mips.texi
4634 @include c-ns32k.texi
4642 @include c-sparc.texi
4654 @c reverse effect of @down at top of generic Machine-Dep chapter
4658 @node Reporting Bugs
4659 @chapter Reporting Bugs
4660 @cindex bugs in @code{@value{AS}}
4661 @cindex reporting bugs in @code{@value{AS}}
4663 Your bug reports play an essential role in making @code{@value{AS}} reliable.
4665 Reporting a bug may help you by bringing a solution to your problem, or it may
4666 not. But in any case the principal function of a bug report is to help the
4667 entire community by making the next version of @code{@value{AS}} work better.
4668 Bug reports are your contribution to the maintenance of @code{@value{AS}}.
4670 In order for a bug report to serve its purpose, you must include the
4671 information that enables us to fix the bug.
4674 * Bug Criteria:: Have you found a bug?
4675 * Bug Reporting:: How to report bugs
4679 @section Have you found a bug?
4680 @cindex bug criteria
4682 If you are not sure whether you have found a bug, here are some guidelines:
4685 @cindex fatal signal
4686 @cindex assembler crash
4687 @cindex crash of assembler
4689 If the assembler gets a fatal signal, for any input whatever, that is a
4690 @code{@value{AS}} bug. Reliable assemblers never crash.
4692 @cindex error on valid input
4694 If @code{@value{AS}} produces an error message for valid input, that is a bug.
4696 @cindex invalid input
4698 If @code{@value{AS}} does not produce an error message for invalid input, that
4699 is a bug. However, you should note that your idea of ``invalid input'' might
4700 be our idea of ``an extension'' or ``support for traditional practice''.
4703 If you are an experienced user of assemblers, your suggestions for improvement
4704 of @code{@value{AS}} are welcome in any case.
4708 @section How to report bugs
4710 @cindex @code{@value{AS}} bugs, reporting
4712 A number of companies and individuals offer support for @sc{gnu} products. If
4713 you obtained @code{@value{AS}} from a support organization, we recommend you
4714 contact that organization first.
4716 You can find contact information for many support companies and
4717 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
4720 In any event, we also recommend that you send bug reports for @code{@value{AS}}
4721 to @samp{bug-gnu-utils@@prep.ai.mit.edu}.
4723 The fundamental principle of reporting bugs usefully is this:
4724 @strong{report all the facts}. If you are not sure whether to state a
4725 fact or leave it out, state it!
4727 Often people omit facts because they think they know what causes the problem
4728 and assume that some details do not matter. Thus, you might assume that the
4729 name of a symbol you use in an example does not matter. Well, probably it does
4730 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
4731 happens to fetch from the location where that name is stored in memory;
4732 perhaps, if the name were different, the contents of that location would fool
4733 the assembler into doing the right thing despite the bug. Play it safe and
4734 give a specific, complete example. That is the easiest thing for you to do,
4735 and the most helpful.
4737 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
4738 it is new to us. Therefore, always write your bug reports on the assumption
4739 that the bug has not been reported previously.
4741 Sometimes people give a few sketchy facts and ask, ``Does this ring a
4742 bell?'' Those bug reports are useless, and we urge everyone to
4743 @emph{refuse to respond to them} except to chide the sender to report
4746 To enable us to fix the bug, you should include all these things:
4750 The version of @code{@value{AS}}. @code{@value{AS}} announces it if you start
4751 it with the @samp{--version} argument.
4753 Without this, we will not know whether there is any point in looking for
4754 the bug in the current version of @code{@value{AS}}.
4757 Any patches you may have applied to the @code{@value{AS}} source.
4760 The type of machine you are using, and the operating system name and
4764 What compiler (and its version) was used to compile @code{@value{AS}}---e.g.
4768 The command arguments you gave the assembler to assemble your example and
4769 observe the bug. To guarantee you will not omit something important, list them
4770 all. A copy of the Makefile (or the output from make) is sufficient.
4772 If we were to try to guess the arguments, we would probably guess wrong
4773 and then we might not encounter the bug.
4776 A complete input file that will reproduce the bug. If the bug is observed when
4777 the assembler is invoked via a compiler, send the assembler source, not the
4778 high level language source. Most compilers will produce the assembler source
4779 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
4780 the options @samp{-v --save-temps}; this will save the assembler source in a
4781 file with an extension of @file{.s}, and also show you exactly how
4782 @code{@value{AS}} is being run.
4785 A description of what behavior you observe that you believe is
4786 incorrect. For example, ``It gets a fatal signal.''
4788 Of course, if the bug is that @code{@value{AS}} gets a fatal signal, then we
4789 will certainly notice it. But if the bug is incorrect output, we might not
4790 notice unless it is glaringly wrong. You might as well not give us a chance to
4793 Even if the problem you experience is a fatal signal, you should still say so
4794 explicitly. Suppose something strange is going on, such as, your copy of
4795 @code{@value{AS}} is out of synch, or you have encountered a bug in the C
4796 library on your system. (This has happened!) Your copy might crash and ours
4797 would not. If you told us to expect a crash, then when ours fails to crash, we
4798 would know that the bug was not happening for us. If you had not told us to
4799 expect a crash, then we would not be able to draw any conclusion from our
4803 If you wish to suggest changes to the @code{@value{AS}} source, send us context
4804 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
4805 option. Always send diffs from the old file to the new file. If you even
4806 discuss something in the @code{@value{AS}} source, refer to it by context, not
4809 The line numbers in our development sources will not match those in your
4810 sources. Your line numbers would convey no useful information to us.
4813 Here are some things that are not necessary:
4817 A description of the envelope of the bug.
4819 Often people who encounter a bug spend a lot of time investigating
4820 which changes to the input file will make the bug go away and which
4821 changes will not affect it.
4823 This is often time consuming and not very useful, because the way we
4824 will find the bug is by running a single example under the debugger
4825 with breakpoints, not by pure deduction from a series of examples.
4826 We recommend that you save your time for something else.
4828 Of course, if you can find a simpler example to report @emph{instead}
4829 of the original one, that is a convenience for us. Errors in the
4830 output will be easier to spot, running under the debugger will take
4831 less time, and so on.
4833 However, simplification is not vital; if you do not want to do this,
4834 report the bug anyway and send us the entire test case you used.
4837 A patch for the bug.
4839 A patch for the bug does help us if it is a good one. But do not omit
4840 the necessary information, such as the test case, on the assumption that
4841 a patch is all we need. We might see problems with your patch and decide
4842 to fix the problem another way, or we might not understand it at all.
4844 Sometimes with a program as complicated as @code{@value{AS}} it is very hard to
4845 construct an example that will make the program follow a certain path through
4846 the code. If you do not send us the example, we will not be able to construct
4847 one, so we will not be able to verify that the bug is fixed.
4849 And if we cannot understand what bug you are trying to fix, or why your
4850 patch should be an improvement, we will not install it. A test case will
4851 help us to understand.
4854 A guess about what the bug is or what it depends on.
4856 Such guesses are usually wrong. Even we cannot guess right about such
4857 things without first using the debugger to find the facts.
4860 @node Acknowledgements
4861 @chapter Acknowledgements
4863 If you have contributed to @code{@value{AS}} and your name isn't listed here,
4864 it is not meant as a slight. We just don't know about it. Send mail to the
4865 maintainer, and we'll correct the situation. Currently
4867 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
4869 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
4872 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
4873 information and the 68k series machines, most of the preprocessing pass, and
4874 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
4876 K. Richard Pixley maintained GAS for a while, adding various enhancements and
4877 many bug fixes, including merging support for several processors, breaking GAS
4878 up to handle multiple object file format back ends (including heavy rewrite,
4879 testing, an integration of the coff and b.out back ends), adding configuration
4880 including heavy testing and verification of cross assemblers and file splits
4881 and renaming, converted GAS to strictly ANSI C including full prototypes, added
4882 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
4883 port (including considerable amounts of reverse engineering), a SPARC opcode
4884 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
4885 assertions and made them work, much other reorganization, cleanup, and lint.
4887 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
4888 in format-specific I/O modules.
4890 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
4891 has done much work with it since.
4893 The Intel 80386 machine description was written by Eliot Dresselhaus.
4895 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
4897 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
4898 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
4900 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
4901 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
4902 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
4903 support a.out format.
4905 Support for the Zilog Z8k and Hitachi H8/300 and H8/500 processors (tc-z8k,
4906 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
4907 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
4908 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
4911 John Gilmore built the AMD 29000 support, added @code{.include} support, and
4912 simplified the configuration of which versions accept which directives. He
4913 updated the 68k machine description so that Motorola's opcodes always produced
4914 fixed-size instructions (e.g. @code{jsr}), while synthetic instructions
4915 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
4916 cross-compilation support, and one bug in relaxation that took a week and
4917 required the proverbial one-bit fix.
4919 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
4920 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
4921 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
4922 PowerPC assembler, and made a few other minor patches.
4924 Steve Chamberlain made @code{@value{AS}} able to generate listings.
4926 Hewlett-Packard contributed support for the HP9000/300.
4928 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
4929 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
4930 formats). This work was supported by both the Center for Software Science at
4931 the University of Utah and Cygnus Support.
4933 Support for ELF format files has been worked on by Mark Eichin of Cygnus
4934 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
4935 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
4936 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
4937 and some initial 64-bit support).
4939 Richard Henderson rewrote the Alpha assembler.
4941 Several engineers at Cygnus Support have also provided many small bug fixes and
4942 configuration enhancements.
4944 Many others have contributed large or small bugfixes and enhancements. If
4945 you have contributed significant work and are not mentioned on this list, and
4946 want to be, let us know. Some of the history has been lost; we are not
4947 intentionally leaving anyone out.