1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (c) 1991, 92, 93, 94, 95, 96, 97, 98, 2000
3 @c Free Software Foundation, Inc.
4 @c UPDATE!! On future updates--
5 @c (1) check for new machine-dep cmdline options in
6 @c md_parse_option definitions in config/tc-*.c
7 @c (2) for platform-specific directives, examine md_pseudo_op
9 @c (3) for object-format specific directives, examine obj_pseudo_op
11 @c (4) portable directives in potable[] in read.c
15 @c defaults, config file may override:
18 @include asconfig.texi
21 @c common OR combinations of conditions
41 @set abnormal-separator
45 @settitle Using @value{AS}
48 @settitle Using @value{AS} (@value{TARGET})
50 @setchapternewpage odd
55 @c WARE! Some of the machine-dependent sections contain tables of machine
56 @c instructions. Except in multi-column format, these tables look silly.
57 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
58 @c the multi-col format is faked within @example sections.
60 @c Again unfortunately, the natural size that fits on a page, for these tables,
61 @c is different depending on whether or not smallbook is turned on.
62 @c This matters, because of order: text flow switches columns at each page
65 @c The format faked in this source works reasonably well for smallbook,
66 @c not well for the default large-page format. This manual expects that if you
67 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
68 @c tables in question. You can turn on one without the other at your
69 @c discretion, of course.
72 @c the insn tables look just as silly in info files regardless of smallbook,
73 @c might as well show 'em anyways.
79 * As: (as). The GNU assembler.
88 This file documents the GNU Assembler "@value{AS}".
90 Copyright (C) 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000 Free Software Foundation, Inc.
92 Permission is granted to make and distribute verbatim copies of
93 this manual provided the copyright notice and this permission notice
94 are preserved on all copies.
97 Permission is granted to process this file through Tex and print the
98 results, provided the printed document carries copying permission
99 notice identical to this one except for the removal of this paragraph
100 (this paragraph not being relevant to the printed manual).
103 Permission is granted to copy and distribute modified versions of this manual
104 under the conditions for verbatim copying, provided that the entire resulting
105 derived work is distributed under the terms of a permission notice identical to
108 Permission is granted to copy and distribute translations of this manual
109 into another language, under the above conditions for modified versions.
113 @title Using @value{AS}
114 @subtitle The @sc{gnu} Assembler
116 @subtitle for the @value{TARGET} family
119 @subtitle Version @value{VERSION}
122 The Free Software Foundation Inc. thanks The Nice Computer
123 Company of Australia for loaning Dean Elsner to write the
124 first (Vax) version of @code{as} for Project @sc{gnu}.
125 The proprietors, management and staff of TNCCA thank FSF for
126 distracting the boss while they got some work
129 @author Dean Elsner, Jay Fenlason & friends
133 \hfill {\it Using {\tt @value{AS}}}\par
134 \hfill Edited by Cygnus Support\par
136 %"boxit" macro for figures:
137 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
138 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
139 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
140 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
141 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
144 @vskip 0pt plus 1filll
145 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000 Free Software Foundation, Inc.
147 Permission is granted to make and distribute verbatim copies of
148 this manual provided the copyright notice and this permission notice
149 are preserved on all copies.
151 Permission is granted to copy and distribute modified versions of this manual
152 under the conditions for verbatim copying, provided that the entire resulting
153 derived work is distributed under the terms of a permission notice identical to
156 Permission is granted to copy and distribute translations of this manual
157 into another language, under the above conditions for modified versions.
162 @top Using @value{AS}
164 This file is a user guide to the @sc{gnu} assembler @code{@value{AS}} version
167 This version of the file describes @code{@value{AS}} configured to generate
168 code for @value{TARGET} architectures.
171 * Overview:: Overview
172 * Invoking:: Command-Line Options
174 * Sections:: Sections and Relocation
176 * Expressions:: Expressions
177 * Pseudo Ops:: Assembler Directives
178 * Machine Dependencies:: Machine Dependent Features
179 * Reporting Bugs:: Reporting Bugs
180 * Acknowledgements:: Who Did What
188 This manual is a user guide to the @sc{gnu} assembler @code{@value{AS}}.
190 This version of the manual describes @code{@value{AS}} configured to generate
191 code for @value{TARGET} architectures.
195 @cindex invocation summary
196 @cindex option summary
197 @cindex summary of options
198 Here is a brief summary of how to invoke @code{@value{AS}}. For details,
199 @pxref{Invoking,,Comand-Line Options}.
201 @c We don't use deffn and friends for the following because they seem
202 @c to be limited to one line for the header.
204 @value{AS} [ -a[cdhlns][=file] ] [ -D ] [ --defsym @var{sym}=@var{val} ]
205 [ -f ] [ --gstabs ] [ --gdwarf2 ] [ --help ] [ -I @var{dir} ] [ -J ] [ -K ] [ -L ]
206 [ --keep-locals ] [ -o @var{objfile} ] [ -R ] [ --statistics ] [ -v ]
207 [ -version ] [ --version ] [ -W ] [ --warn ] [ --fatal-warnings ]
208 [ -w ] [ -x ] [ -Z ] [ --target-help ]
210 @c am29k has no machine-dependent assembler options
213 [ -mbig-endian | -mlittle-endian ]
216 [ -m[arm]1 | -m[arm]2 | -m[arm]250 | -m[arm]3 | -m[arm]6 | -m[arm]60 |
217 -m[arm]600 | -m[arm]610 | -m[arm]620 | -m[arm]7[t][[d]m[i]][fe] | -m[arm]70 |
218 -m[arm]700 | -m[arm]710[c] | -m[arm]7100 | -m[arm]7500 | -m[arm]8 |
219 -m[arm]810 | -m[arm]9 | -m[arm]920 | -m[arm]920t | -m[arm]9tdmi |
220 -mstrongarm | -mstrongarm110 | -mstrongarm1100 ]
221 [ -m[arm]v2 | -m[arm]v2a | -m[arm]v3 | -m[arm]v3m | -m[arm]v4 | -m[arm]v4t |
222 -m[arm]v5 | -[arm]v5t ]
224 [ -mfpa10 | -mfpa11 | -mfpe-old | -mno-fpu ]
226 [ -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant ]
227 [ -mthumb-interwork ]
238 @c Hitachi family chips have no machine-dependent assembler options
241 @c HPPA has no machine-dependent assembler options (yet).
247 @c The order here is important. See c-sparc.texi.
248 [ -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
249 -Av8plus | -Av8plusa | -Av9 | -Av9a ]
250 [ -xarch=v8plus | -xarch=v8plusa ] [ -bump ] [ -32 | -64 ]
253 [ -mcpu=54[123589] | -mcpu=54[56]lp ] [ -mfar-mode | -mf ]
254 [ -merrors-to-file <filename> | -me <filename> ]
257 @c Z8000 has no machine-dependent assembler options
260 @c see md_parse_option in tc-i960.c
261 [ -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC ]
265 [ --m32rx | --[no-]warn-explicit-parallel-conflicts | --W[n]p ]
268 [ -l ] [ -m68000 | -m68010 | -m68020 | ... ]
271 [ -jsri2bsr ] [ -sifilter ] [ -relax ]
275 [ -m68hc11 | -m68hc12 ]
276 [ --force-long-branchs ] [ --short-branchs ] [ --strict-direct-mode ]
277 [ --print-insn-syntax ] [ --print-opcodes ] [ --generate-example ]
280 [ -nocpp ] [ -EL ] [ -EB ] [ -G @var{num} ] [ -mcpu=@var{CPU} ]
281 [ -mips1 ] [ -mips2 ] [ -mips3 ] [ -m4650 ] [ -no-m4650 ] [ -mips32 ] [ -no-mips32 ]
282 [ --trap ] [ --break ]
283 [ --emulation=@var{name} ]
285 [ -- | @var{files} @dots{} ]
290 Turn on listings, in any of a variety of ways:
294 omit false conditionals
297 omit debugging directives
300 include high-level source
306 include macro expansions
309 omit forms processing
315 set the name of the listing file
318 You may combine these options; for example, use @samp{-aln} for assembly
319 listing without forms processing. The @samp{=file} option, if used, must be
320 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
323 Ignored. This option is accepted for script compatibility with calls to
326 @item --defsym @var{sym}=@var{value}
327 Define the symbol @var{sym} to be @var{value} before assembling the input file.
328 @var{value} must be an integer constant. As in C, a leading @samp{0x}
329 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
332 ``fast''---skip whitespace and comment preprocessing (assume source is
336 Generate stabs debugging information for each assembler line. This
337 may help debugging assembler code, if the debugger can handle it.
340 Generate DWARF2 debugging information for each assembler line. This
341 may help debugging assembler code, if the debugger can handle it. Note - this
342 option is only supported by some targets, not all of them.
345 Print a summary of the command line options and exit.
348 Print a summary of all target specific options and exit.
351 Add directory @var{dir} to the search list for @code{.include} directives.
354 Don't warn about signed overflow.
357 @ifclear DIFF-TBL-KLUGE
358 This option is accepted but has no effect on the @value{TARGET} family.
360 @ifset DIFF-TBL-KLUGE
361 Issue warnings when difference tables altered for long displacements.
366 Keep (in the symbol table) local symbols. On traditional a.out systems
367 these start with @samp{L}, but different systems have different local
370 @item -o @var{objfile}
371 Name the object-file output from @code{@value{AS}} @var{objfile}.
374 Fold the data section into the text section.
377 Print the maximum space (in bytes) and total time (in seconds) used by
380 @item --strip-local-absolute
381 Remove local absolute symbols from the outgoing symbol table.
385 Print the @code{as} version.
388 Print the @code{as} version and exit.
392 Suppress warning messages.
394 @item --fatal-warnings
395 Treat warnings as errors.
398 Don't suppress warning messages or treat them as errors.
407 Generate an object file even after errors.
409 @item -- | @var{files} @dots{}
410 Standard input, or source files to assemble.
415 The following options are available when @value{AS} is configured for
420 @cindex ARC endianness
421 @cindex endianness, ARC
422 @cindex big endian output, ARC
424 Generate ``big endian'' format output.
426 @cindex little endian output, ARC
427 @item -mlittle-endian
428 Generate ``little endian'' format output.
434 The following options are available when @value{AS} is configured for the ARM
438 @item -m[arm][1|2|3|6|7|8|9][...]
439 Specify which ARM processor variant is the target.
440 @item -m[arm]v[2|2a|3|3m|4|4t|5|5t]
441 Specify which ARM architecture variant is used by the target.
442 @item -mthumb | -mall
443 Enable or disable Thumb only instruction decoding.
444 @item -mfpa10 | -mfpa11 | -mfpe-old | -mno-fpu
445 Select which Floating Point architcture is the target.
446 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant | -moabi
447 Select which procedure calling convention is in use.
449 Select either big-endian (-EB) or little-endian (-EL) output.
450 @item -mthumb-interwork
451 Specify that the code has been generated with interworking between Thumb and
454 Specify that PIC code has been generated.
459 The following options are available when @value{AS} is configured for
462 @cindex D10V optimization
463 @cindex optimization, D10V
465 Optimize output by parallelizing instructions.
470 The following options are available when @value{AS} is configured for a D30V
473 @cindex D30V optimization
474 @cindex optimization, D30V
476 Optimize output by parallelizing instructions.
480 Warn when nops are generated.
482 @cindex D30V nops after 32-bit multiply
484 Warn when a nop after a 32-bit multiply instruction is generated.
489 The following options are available when @value{AS} is configured for the
490 Intel 80960 processor.
493 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
494 Specify which variant of the 960 architecture is the target.
497 Add code to collect statistics about branches taken.
500 Do not alter compare-and-branch instructions for long displacements;
507 The following options are available when @value{AS} is configured for the
508 Mitsubishi M32R series.
513 Specify which processor in the M32R family is the target. The default
514 is normally the M32R, but this option changes it to the M32RX.
516 @item --warn-explicit-parallel-conflicts or --Wp
517 Produce warning messages when questionable parallel constructs are
520 @item --no-warn-explicit-parallel-conflicts or --Wnp
521 Do not produce warning messages when questionable parallel constructs are
528 The following options are available when @value{AS} is configured for the
529 Motorola 68000 series.
534 Shorten references to undefined symbols, to one word instead of two.
536 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030 | -m68040 | -m68060
537 @itemx | -m68302 | -m68331 | -m68332 | -m68333 | -m68340 | -mcpu32 | -m5200
538 Specify what processor in the 68000 family is the target. The default
539 is normally the 68020, but this can be changed at configuration time.
541 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
542 The target machine does (or does not) have a floating-point coprocessor.
543 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
544 the basic 68000 is not compatible with the 68881, a combination of the
545 two can be specified, since it's possible to do emulation of the
546 coprocessor instructions with the main processor.
548 @item -m68851 | -mno-68851
549 The target machine does (or does not) have a memory-management
550 unit coprocessor. The default is to assume an MMU for 68020 and up.
556 The following options are available when @value{AS} is configured for
557 a picoJava processor.
561 @cindex PJ endianness
562 @cindex endianness, PJ
563 @cindex big endian output, PJ
565 Generate ``big endian'' format output.
567 @cindex little endian output, PJ
569 Generate ``little endian'' format output.
575 The following options are available when @value{AS} is configured for the
576 Motorola 68HC11 or 68HC12 series.
580 @item -m68hc11 | -m68hc12
581 Specify what processor is the target. The default is
582 defined by the configuration option when building the assembler.
584 @item --force-long-branchs
585 Relative branches are turned into absolute ones. This concerns
586 conditional branches, unconditional branches and branches to a
589 @item -S | --short-branchs
590 Do not turn relative branchs into absolute ones
591 when the offset is out of range.
593 @item --strict-direct-mode
594 Do not turn the direct addressing mode into extended addressing mode
595 when the instruction does not support direct addressing mode.
597 @item --print-insn-syntax
598 Print the syntax of instruction in case of error.
600 @item --print-opcodes
601 print the list of instructions with syntax and then exit.
603 @item --generate-example
604 print an example of instruction for each possible instruction and then exit.
605 This option is only useful for testing @code{@value{AS}}.
611 The following options are available when @code{@value{AS}} is configured
612 for the SPARC architecture:
615 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
616 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
617 Explicitly select a variant of the SPARC architecture.
619 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
620 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
622 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
623 UltraSPARC extensions.
625 @item -xarch=v8plus | -xarch=v8plusa
626 For compatibility with the Solaris v9 assembler. These options are
627 equivalent to -Av8plus and -Av8plusa, respectively.
630 Warn when the assembler switches to another architecture.
635 The following options are available when @value{AS} is configured for the 'c54x
640 Enable extended addressing mode. All addresses and relocations will assume
641 extended addressing (usually 23 bits).
642 @item -mcpu=@var{CPU_VERSION}
643 Sets the CPU version being compiled for.
644 @item -merrors-to-file @var{FILENAME}
645 Redirect error output to a file, for broken systems which don't support such
646 behaviour in the shell.
651 The following options are available when @value{AS} is configured for
656 This option sets the largest size of an object that can be referenced
657 implicitly with the @code{gp} register. It is only accepted for targets that
658 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
660 @cindex MIPS endianness
661 @cindex endianness, MIPS
662 @cindex big endian output, MIPS
664 Generate ``big endian'' format output.
666 @cindex little endian output, MIPS
668 Generate ``little endian'' format output.
674 Generate code for a particular MIPS Instruction Set Architecture level.
675 @samp{-mips1} corresponds to the @sc{r2000} and @sc{r3000} processors,
676 @samp{-mips2} to the @sc{r6000} processor, and @samp{-mips3} to the @sc{r4000}
681 Generate code for the MIPS @sc{r4650} chip. This tells the assembler to accept
682 the @samp{mad} and @samp{madu} instruction, and to not schedule @samp{nop}
683 instructions around accesses to the @samp{HI} and @samp{LO} registers.
684 @samp{-no-m4650} turns off this option.
688 Generate code for the @sc{MIPS32} architecture. This tells the assembler to
689 accept ISA level 2 instructions and MIPS32 extensions including some @sc{r4000}
692 @item -mcpu=@var{CPU}
693 Generate code for a particular MIPS cpu. This has little effect on the
694 assembler, but it is passed by @code{@value{GCC}}.
697 @item --emulation=@var{name}
698 This option causes @code{@value{AS}} to emulate @code{@value{AS}} configured
699 for some other target, in all respects, including output format (choosing
700 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
701 debugging information or store symbol table information, and default
702 endianness. The available configuration names are: @samp{mipsecoff},
703 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
704 @samp{mipsbelf}. The first two do not alter the default endianness from that
705 of the primary target for which the assembler was configured; the others change
706 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
707 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
708 selection in any case.
710 This option is currently supported only when the primary target
711 @code{@value{AS}} is configured for is a MIPS ELF or ECOFF target.
712 Furthermore, the primary target or others specified with
713 @samp{--enable-targets=@dots{}} at configuration time must include support for
714 the other format, if both are to be available. For example, the Irix 5
715 configuration includes support for both.
717 Eventually, this option will support more configurations, with more
718 fine-grained control over the assembler's behavior, and will be supported for
722 @code{@value{AS}} ignores this option. It is accepted for compatibility with
730 Control how to deal with multiplication overflow and division by zero.
731 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
732 (and only work for Instruction Set Architecture level 2 and higher);
733 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
739 The following options are available when @value{AS} is configured for
745 Enable or disable the JSRI to BSR transformation. By default this is enabled.
746 The command line option @samp{-nojsri2bsr} can be used to disable it.
750 Enable or disable the silicon filter behaviour. By default this is disabled.
751 The default can be overidden by the @samp{-sifilter} command line option.
754 Alter jump instructions for long displacements.
756 @item -mcpu=[210|340]
757 Select the cpu type on the target hardware. This controls which instructions
761 Assemble for a big endian target.
764 Assemble for a little endian target.
770 * Manual:: Structure of this Manual
771 * GNU Assembler:: The GNU Assembler
772 * Object Formats:: Object File Formats
773 * Command Line:: Command Line
774 * Input Files:: Input Files
775 * Object:: Output (Object) File
776 * Errors:: Error and Warning Messages
780 @section Structure of this Manual
782 @cindex manual, structure and purpose
783 This manual is intended to describe what you need to know to use
784 @sc{gnu} @code{@value{AS}}. We cover the syntax expected in source files, including
785 notation for symbols, constants, and expressions; the directives that
786 @code{@value{AS}} understands; and of course how to invoke @code{@value{AS}}.
789 We also cover special features in the @value{TARGET}
790 configuration of @code{@value{AS}}, including assembler directives.
793 This manual also describes some of the machine-dependent features of
794 various flavors of the assembler.
797 @cindex machine instructions (not covered)
798 On the other hand, this manual is @emph{not} intended as an introduction
799 to programming in assembly language---let alone programming in general!
800 In a similar vein, we make no attempt to introduce the machine
801 architecture; we do @emph{not} describe the instruction set, standard
802 mnemonics, registers or addressing modes that are standard to a
803 particular architecture.
805 You may want to consult the manufacturer's
806 machine architecture manual for this information.
810 For information on the H8/300 machine instruction set, see @cite{H8/300
811 Series Programming Manual} (Hitachi ADE--602--025). For the H8/300H,
812 see @cite{H8/300H Series Programming Manual} (Hitachi).
815 For information on the H8/500 machine instruction set, see @cite{H8/500
816 Series Programming Manual} (Hitachi M21T001).
819 For information on the Hitachi SH machine instruction set, see
820 @cite{SH-Microcomputer User's Manual} (Hitachi Micro Systems, Inc.).
823 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
827 @c I think this is premature---doc@cygnus.com, 17jan1991
829 Throughout this manual, we assume that you are running @dfn{GNU},
830 the portable operating system from the @dfn{Free Software
831 Foundation, Inc.}. This restricts our attention to certain kinds of
832 computer (in particular, the kinds of computers that @sc{gnu} can run on);
833 once this assumption is granted examples and definitions need less
836 @code{@value{AS}} is part of a team of programs that turn a high-level
837 human-readable series of instructions into a low-level
838 computer-readable series of instructions. Different versions of
839 @code{@value{AS}} are used for different kinds of computer.
842 @c There used to be a section "Terminology" here, which defined
843 @c "contents", "byte", "word", and "long". Defining "word" to any
844 @c particular size is confusing when the .word directive may generate 16
845 @c bits on one machine and 32 bits on another; in general, for the user
846 @c version of this manual, none of these terms seem essential to define.
847 @c They were used very little even in the former draft of the manual;
848 @c this draft makes an effort to avoid them (except in names of
852 @section The GNU Assembler
854 @sc{gnu} @code{as} is really a family of assemblers.
856 This manual describes @code{@value{AS}}, a member of that family which is
857 configured for the @value{TARGET} architectures.
859 If you use (or have used) the @sc{gnu} assembler on one architecture, you
860 should find a fairly similar environment when you use it on another
861 architecture. Each version has much in common with the others,
862 including object file formats, most assembler directives (often called
863 @dfn{pseudo-ops}) and assembler syntax.@refill
865 @cindex purpose of @sc{gnu} assembler
866 @code{@value{AS}} is primarily intended to assemble the output of the
867 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
868 @code{@value{LD}}. Nevertheless, we've tried to make @code{@value{AS}}
869 assemble correctly everything that other assemblers for the same
870 machine would assemble.
872 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
875 @c This remark should appear in generic version of manual; assumption
876 @c here is that generic version sets M680x0.
877 This doesn't mean @code{@value{AS}} always uses the same syntax as another
878 assembler for the same architecture; for example, we know of several
879 incompatible versions of 680x0 assembly language syntax.
882 Unlike older assemblers, @code{@value{AS}} is designed to assemble a source
883 program in one pass of the source file. This has a subtle impact on the
884 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
887 @section Object File Formats
889 @cindex object file format
890 The @sc{gnu} assembler can be configured to produce several alternative
891 object file formats. For the most part, this does not affect how you
892 write assembly language programs; but directives for debugging symbols
893 are typically different in different file formats. @xref{Symbol
894 Attributes,,Symbol Attributes}.
897 On the @value{TARGET}, @code{@value{AS}} is configured to produce
898 @value{OBJ-NAME} format object files.
900 @c The following should exhaust all configs that set MULTI-OBJ, ideally
902 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
903 @code{a.out} or COFF format object files.
906 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
907 @code{b.out} or COFF format object files.
910 On the @value{TARGET}, @code{@value{AS}} can be configured to produce either
911 SOM or ELF format object files.
916 @section Command Line
918 @cindex command line conventions
919 After the program name @code{@value{AS}}, the command line may contain
920 options and file names. Options may appear in any order, and may be
921 before, after, or between file names. The order of file names is
924 @cindex standard input, as input file
926 @file{--} (two hyphens) by itself names the standard input file
927 explicitly, as one of the files for @code{@value{AS}} to assemble.
929 @cindex options, command line
930 Except for @samp{--} any command line argument that begins with a
931 hyphen (@samp{-}) is an option. Each option changes the behavior of
932 @code{@value{AS}}. No option changes the way another option works. An
933 option is a @samp{-} followed by one or more letters; the case of
934 the letter is important. All options are optional.
936 Some options expect exactly one file name to follow them. The file
937 name may either immediately follow the option's letter (compatible
938 with older assemblers) or it may be the next command argument (@sc{gnu}
939 standard). These two command lines are equivalent:
942 @value{AS} -o my-object-file.o mumble.s
943 @value{AS} -omy-object-file.o mumble.s
950 @cindex source program
952 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
953 describe the program input to one run of @code{@value{AS}}. The program may
954 be in one or more files; how the source is partitioned into files
955 doesn't change the meaning of the source.
957 @c I added "con" prefix to "catenation" just to prove I can overcome my
958 @c APL training... doc@cygnus.com
959 The source program is a concatenation of the text in all the files, in the
962 Each time you run @code{@value{AS}} it assembles exactly one source
963 program. The source program is made up of one or more files.
964 (The standard input is also a file.)
966 You give @code{@value{AS}} a command line that has zero or more input file
967 names. The input files are read (from left file name to right). A
968 command line argument (in any position) that has no special meaning
969 is taken to be an input file name.
971 If you give @code{@value{AS}} no file names it attempts to read one input file
972 from the @code{@value{AS}} standard input, which is normally your terminal. You
973 may have to type @key{ctl-D} to tell @code{@value{AS}} there is no more program
976 Use @samp{--} if you need to explicitly name the standard input file
977 in your command line.
979 If the source is empty, @code{@value{AS}} produces a small, empty object
982 @subheading Filenames and Line-numbers
984 @cindex input file linenumbers
985 @cindex line numbers, in input files
986 There are two ways of locating a line in the input file (or files) and
987 either may be used in reporting error messages. One way refers to a line
988 number in a physical file; the other refers to a line number in a
989 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
991 @dfn{Physical files} are those files named in the command line given
992 to @code{@value{AS}}.
994 @dfn{Logical files} are simply names declared explicitly by assembler
995 directives; they bear no relation to physical files. Logical file names help
996 error messages reflect the original source file, when @code{@value{AS}} source
997 is itself synthesized from other files. @code{@value{AS}} understands the
998 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
999 @ref{File,,@code{.file}}.
1002 @section Output (Object) File
1008 Every time you run @code{@value{AS}} it produces an output file, which is
1009 your assembly language program translated into numbers. This file
1010 is the object file. Its default name is
1018 @code{b.out} when @code{@value{AS}} is configured for the Intel 80960.
1020 You can give it another name by using the @code{-o} option. Conventionally,
1021 object file names end with @file{.o}. The default name is used for historical
1022 reasons: older assemblers were capable of assembling self-contained programs
1023 directly into a runnable program. (For some formats, this isn't currently
1024 possible, but it can be done for the @code{a.out} format.)
1028 The object file is meant for input to the linker @code{@value{LD}}. It contains
1029 assembled program code, information to help @code{@value{LD}} integrate
1030 the assembled program into a runnable file, and (optionally) symbolic
1031 information for the debugger.
1033 @c link above to some info file(s) like the description of a.out.
1034 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1037 @section Error and Warning Messages
1039 @cindex error messsages
1040 @cindex warning messages
1041 @cindex messages from assembler
1042 @code{@value{AS}} may write warnings and error messages to the standard error
1043 file (usually your terminal). This should not happen when a compiler
1044 runs @code{@value{AS}} automatically. Warnings report an assumption made so
1045 that @code{@value{AS}} could keep assembling a flawed program; errors report a
1046 grave problem that stops the assembly.
1048 @cindex format of warning messages
1049 Warning messages have the format
1052 file_name:@b{NNN}:Warning Message Text
1056 @cindex line numbers, in warnings/errors
1057 (where @b{NNN} is a line number). If a logical file name has been given
1058 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1059 the current input file is used. If a logical line number was given
1061 (@pxref{Line,,@code{.line}})
1065 (@pxref{Line,,@code{.line}})
1068 (@pxref{Ln,,@code{.ln}})
1071 then it is used to calculate the number printed,
1072 otherwise the actual line in the current source file is printed. The
1073 message text is intended to be self explanatory (in the grand Unix
1076 @cindex format of error messages
1077 Error messages have the format
1079 file_name:@b{NNN}:FATAL:Error Message Text
1081 The file name and line number are derived as for warning
1082 messages. The actual message text may be rather less explanatory
1083 because many of them aren't supposed to happen.
1086 @chapter Command-Line Options
1088 @cindex options, all versions of assembler
1089 This chapter describes command-line options available in @emph{all}
1090 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
1092 to the @value{TARGET}.
1095 to particular machine architectures.
1098 If you are invoking @code{@value{AS}} via the @sc{gnu} C compiler (version 2),
1099 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1100 The assembler arguments must be separated from each other (and the @samp{-Wa})
1101 by commas. For example:
1104 gcc -c -g -O -Wa,-alh,-L file.c
1108 This passes two options to the assembler: @samp{-alh} (emit a listing to
1109 standard output with with high-level and assembly source) and @samp{-L} (retain
1110 local symbols in the symbol table).
1112 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1113 command-line options are automatically passed to the assembler by the compiler.
1114 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1115 precisely what options it passes to each compilation pass, including the
1119 * a:: -a[cdhlns] enable listings
1120 * D:: -D for compatibility
1121 * f:: -f to work faster
1122 * I:: -I for .include search path
1123 @ifclear DIFF-TBL-KLUGE
1124 * K:: -K for compatibility
1126 @ifset DIFF-TBL-KLUGE
1127 * K:: -K for difference tables
1130 * L:: -L to retain local labels
1131 * M:: -M or --mri to assemble in MRI compatibility mode
1132 * MD:: --MD for dependency tracking
1133 * o:: -o to name the object file
1134 * R:: -R to join data and text sections
1135 * statistics:: --statistics to see statistics about assembly
1136 * traditional-format:: --traditional-format for compatible output
1137 * v:: -v to announce version
1138 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1139 * Z:: -Z to make object file even after errors
1143 @section Enable Listings: @code{-a[cdhlns]}
1152 @cindex listings, enabling
1153 @cindex assembly listings, enabling
1155 These options enable listing output from the assembler. By itself,
1156 @samp{-a} requests high-level, assembly, and symbols listing.
1157 You can use other letters to select specific options for the list:
1158 @samp{-ah} requests a high-level language listing,
1159 @samp{-al} requests an output-program assembly listing, and
1160 @samp{-as} requests a symbol table listing.
1161 High-level listings require that a compiler debugging option like
1162 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1165 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1166 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1167 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1168 omitted from the listing.
1170 Use the @samp{-ad} option to omit debugging directives from the
1173 Once you have specified one of these options, you can further control
1174 listing output and its appearance using the directives @code{.list},
1175 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1177 The @samp{-an} option turns off all forms processing.
1178 If you do not request listing output with one of the @samp{-a} options, the
1179 listing-control directives have no effect.
1181 The letters after @samp{-a} may be combined into one option,
1182 @emph{e.g.}, @samp{-aln}.
1188 This option has no effect whatsoever, but it is accepted to make it more
1189 likely that scripts written for other assemblers also work with
1193 @section Work Faster: @code{-f}
1196 @cindex trusted compiler
1197 @cindex faster processing (@code{-f})
1198 @samp{-f} should only be used when assembling programs written by a
1199 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1200 and comment preprocessing on
1201 the input file(s) before assembling them. @xref{Preprocessing,
1205 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1206 preprocessed (if they contain comments, for example), @code{@value{AS}} does
1211 @section @code{.include} search path: @code{-I} @var{path}
1213 @kindex -I @var{path}
1214 @cindex paths for @code{.include}
1215 @cindex search path for @code{.include}
1216 @cindex @code{include} directive search path
1217 Use this option to add a @var{path} to the list of directories
1218 @code{@value{AS}} searches for files specified in @code{.include}
1219 directives (@pxref{Include,,@code{.include}}). You may use @code{-I} as
1220 many times as necessary to include a variety of paths. The current
1221 working directory is always searched first; after that, @code{@value{AS}}
1222 searches any @samp{-I} directories in the same order as they were
1223 specified (left to right) on the command line.
1226 @section Difference Tables: @code{-K}
1229 @ifclear DIFF-TBL-KLUGE
1230 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1231 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1232 where it can be used to warn when the assembler alters the machine code
1233 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1234 family does not have the addressing limitations that sometimes lead to this
1235 alteration on other platforms.
1238 @ifset DIFF-TBL-KLUGE
1239 @cindex difference tables, warning
1240 @cindex warning for altered difference tables
1241 @code{@value{AS}} sometimes alters the code emitted for directives of the form
1242 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1243 You can use the @samp{-K} option if you want a warning issued when this
1248 @section Include Local Labels: @code{-L}
1251 @cindex local labels, retaining in output
1252 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1253 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1254 debugging, because they are intended for the use of programs (like
1255 compilers) that compose assembler programs, not for your notice.
1256 Normally both @code{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1257 normally debug with them.
1259 This option tells @code{@value{AS}} to retain those @samp{L@dots{}} symbols
1260 in the object file. Usually if you do this you also tell the linker
1261 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1263 By default, a local label is any label beginning with @samp{L}, but each
1264 target is allowed to redefine the local label prefix.
1266 On the HPPA local labels begin with @samp{L$}.
1270 @section Assemble in MRI Compatibility Mode: @code{-M}
1273 @cindex MRI compatibility mode
1274 The @code{-M} or @code{--mri} option selects MRI compatibility mode. This
1275 changes the syntax and pseudo-op handling of @code{@value{AS}} to make it
1276 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1277 configured target) assembler from Microtec Research. The exact nature of the
1278 MRI syntax will not be documented here; see the MRI manuals for more
1279 information. Note in particular that the handling of macros and macro
1280 arguments is somewhat different. The purpose of this option is to permit
1281 assembling existing MRI assembler code using @code{@value{AS}}.
1283 The MRI compatibility is not complete. Certain operations of the MRI assembler
1284 depend upon its object file format, and can not be supported using other object
1285 file formats. Supporting these would require enhancing each object file format
1286 individually. These are:
1289 @item global symbols in common section
1291 The m68k MRI assembler supports common sections which are merged by the linker.
1292 Other object file formats do not support this. @code{@value{AS}} handles
1293 common sections by treating them as a single common symbol. It permits local
1294 symbols to be defined within a common section, but it can not support global
1295 symbols, since it has no way to describe them.
1297 @item complex relocations
1299 The MRI assemblers support relocations against a negated section address, and
1300 relocations which combine the start addresses of two or more sections. These
1301 are not support by other object file formats.
1303 @item @code{END} pseudo-op specifying start address
1305 The MRI @code{END} pseudo-op permits the specification of a start address.
1306 This is not supported by other object file formats. The start address may
1307 instead be specified using the @code{-e} option to the linker, or in a linker
1310 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1312 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1313 name to the output file. This is not supported by other object file formats.
1315 @item @code{ORG} pseudo-op
1317 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1318 address. This differs from the usual @code{@value{AS}} @code{.org} pseudo-op,
1319 which changes the location within the current section. Absolute sections are
1320 not supported by other object file formats. The address of a section may be
1321 assigned within a linker script.
1324 There are some other features of the MRI assembler which are not supported by
1325 @code{@value{AS}}, typically either because they are difficult or because they
1326 seem of little consequence. Some of these may be supported in future releases.
1330 @item EBCDIC strings
1332 EBCDIC strings are not supported.
1334 @item packed binary coded decimal
1336 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1337 and @code{DCB.P} pseudo-ops are not supported.
1339 @item @code{FEQU} pseudo-op
1341 The m68k @code{FEQU} pseudo-op is not supported.
1343 @item @code{NOOBJ} pseudo-op
1345 The m68k @code{NOOBJ} pseudo-op is not supported.
1347 @item @code{OPT} branch control options
1349 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1350 @code{BRL}, and @code{BRW}---are ignored. @code{@value{AS}} automatically
1351 relaxes all branches, whether forward or backward, to an appropriate size, so
1352 these options serve no purpose.
1354 @item @code{OPT} list control options
1356 The following m68k @code{OPT} list control options are ignored: @code{C},
1357 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1358 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1360 @item other @code{OPT} options
1362 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1363 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1365 @item @code{OPT} @code{D} option is default
1367 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1368 @code{OPT NOD} may be used to turn it off.
1370 @item @code{XREF} pseudo-op.
1372 The m68k @code{XREF} pseudo-op is ignored.
1374 @item @code{.debug} pseudo-op
1376 The i960 @code{.debug} pseudo-op is not supported.
1378 @item @code{.extended} pseudo-op
1380 The i960 @code{.extended} pseudo-op is not supported.
1382 @item @code{.list} pseudo-op.
1384 The various options of the i960 @code{.list} pseudo-op are not supported.
1386 @item @code{.optimize} pseudo-op
1388 The i960 @code{.optimize} pseudo-op is not supported.
1390 @item @code{.output} pseudo-op
1392 The i960 @code{.output} pseudo-op is not supported.
1394 @item @code{.setreal} pseudo-op
1396 The i960 @code{.setreal} pseudo-op is not supported.
1401 @section Dependency tracking: @code{--MD}
1404 @cindex dependency tracking
1407 @code{@value{AS}} can generate a dependency file for the file it creates. This
1408 file consists of a single rule suitable for @code{make} describing the
1409 dependencies of the main source file.
1411 The rule is written to the file named in its argument.
1413 This feature is used in the automatic updating of makefiles.
1416 @section Name the Object File: @code{-o}
1419 @cindex naming object file
1420 @cindex object file name
1421 There is always one object file output when you run @code{@value{AS}}. By
1422 default it has the name
1425 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1439 You use this option (which takes exactly one filename) to give the
1440 object file a different name.
1442 Whatever the object file is called, @code{@value{AS}} overwrites any
1443 existing file of the same name.
1446 @section Join Data and Text Sections: @code{-R}
1449 @cindex data and text sections, joining
1450 @cindex text and data sections, joining
1451 @cindex joining text and data sections
1452 @cindex merging text and data sections
1453 @code{-R} tells @code{@value{AS}} to write the object file as if all
1454 data-section data lives in the text section. This is only done at
1455 the very last moment: your binary data are the same, but data
1456 section parts are relocated differently. The data section part of
1457 your object file is zero bytes long because all its bytes are
1458 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1460 When you specify @code{-R} it would be possible to generate shorter
1461 address displacements (because we do not have to cross between text and
1462 data section). We refrain from doing this simply for compatibility with
1463 older versions of @code{@value{AS}}. In future, @code{-R} may work this way.
1466 When @code{@value{AS}} is configured for COFF output,
1467 this option is only useful if you use sections named @samp{.text} and
1472 @code{-R} is not supported for any of the HPPA targets. Using
1473 @code{-R} generates a warning from @code{@value{AS}}.
1477 @section Display Assembly Statistics: @code{--statistics}
1479 @kindex --statistics
1480 @cindex statistics, about assembly
1481 @cindex time, total for assembly
1482 @cindex space used, maximum for assembly
1483 Use @samp{--statistics} to display two statistics about the resources used by
1484 @code{@value{AS}}: the maximum amount of space allocated during the assembly
1485 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1488 @node traditional-format
1489 @section Compatible output: @code{--traditional-format}
1491 @kindex --traditional-format
1492 For some targets, the output of @code{@value{AS}} is different in some ways
1493 from the output of some existing assembler. This switch requests
1494 @code{@value{AS}} to use the traditional format instead.
1496 For example, it disables the exception frame optimizations which
1497 @code{@value{AS}} normally does by default on @code{@value{GCC}} output.
1500 @section Announce Version: @code{-v}
1504 @cindex assembler version
1505 @cindex version of assembler
1506 You can find out what version of as is running by including the
1507 option @samp{-v} (which you can also spell as @samp{-version}) on the
1511 @section Control Warnings: @code{-W}, @code{--warn}, @code{--no-warn}, @code{--fatal-warnings}
1513 @code{@value{AS}} should never give a warning or error message when
1514 assembling compiler output. But programs written by people often
1515 cause @code{@value{AS}} to give a warning that a particular assumption was
1516 made. All such warnings are directed to the standard error file.
1519 @kindex @samp{--no-warn}
1520 @cindex suppressing warnings
1521 @cindex warnings, suppressing
1522 If you use the @code{-W} and @code{--no-warn} options, no warnings are issued.
1523 This only affects the warning messages: it does not change any particular of
1524 how @code{@value{AS}} assembles your file. Errors, which stop the assembly,
1527 @kindex @samp{--fatal-warnings}
1528 @cindex errors, caused by warnings
1529 @cindex warnings, causing error
1530 If you use the @code{--fatal-warnings} option, @code{@value{AS}} considers
1531 files that generate warnings to be in error.
1533 @kindex @samp{--warn}
1534 @cindex warnings, switching on
1535 You can switch these options off again by specifying @code{--warn}, which
1536 causes warnings to be output as usual.
1539 @section Generate Object File in Spite of Errors: @code{-Z}
1540 @cindex object file, after errors
1541 @cindex errors, continuing after
1542 After an error message, @code{@value{AS}} normally produces no output. If for
1543 some reason you are interested in object file output even after
1544 @code{@value{AS}} gives an error message on your program, use the @samp{-Z}
1545 option. If there are any errors, @code{@value{AS}} continues anyways, and
1546 writes an object file after a final warning message of the form @samp{@var{n}
1547 errors, @var{m} warnings, generating bad object file.}
1552 @cindex machine-independent syntax
1553 @cindex syntax, machine-independent
1554 This chapter describes the machine-independent syntax allowed in a
1555 source file. @code{@value{AS}} syntax is similar to what many other
1556 assemblers use; it is inspired by the BSD 4.2
1561 assembler, except that @code{@value{AS}} does not assemble Vax bit-fields.
1565 * Preprocessing:: Preprocessing
1566 * Whitespace:: Whitespace
1567 * Comments:: Comments
1568 * Symbol Intro:: Symbols
1569 * Statements:: Statements
1570 * Constants:: Constants
1574 @section Preprocessing
1576 @cindex preprocessing
1577 The @code{@value{AS}} internal preprocessor:
1579 @cindex whitespace, removed by preprocessor
1581 adjusts and removes extra whitespace. It leaves one space or tab before
1582 the keywords on a line, and turns any other whitespace on the line into
1585 @cindex comments, removed by preprocessor
1587 removes all comments, replacing them with a single space, or an
1588 appropriate number of newlines.
1590 @cindex constants, converted by preprocessor
1592 converts character constants into the appropriate numeric values.
1595 It does not do macro processing, include file handling, or
1596 anything else you may get from your C compiler's preprocessor. You can
1597 do include file processing with the @code{.include} directive
1598 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
1599 to get other ``CPP'' style preprocessing, by giving the input file a
1600 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
1601 Output, gcc.info, Using GNU CC}.
1603 Excess whitespace, comments, and character constants
1604 cannot be used in the portions of the input text that are not
1607 @cindex turning preprocessing on and off
1608 @cindex preprocessing, turning on and off
1611 If the first line of an input file is @code{#NO_APP} or if you use the
1612 @samp{-f} option, whitespace and comments are not removed from the input file.
1613 Within an input file, you can ask for whitespace and comment removal in
1614 specific portions of the by putting a line that says @code{#APP} before the
1615 text that may contain whitespace or comments, and putting a line that says
1616 @code{#NO_APP} after this text. This feature is mainly intend to support
1617 @code{asm} statements in compilers whose output is otherwise free of comments
1624 @dfn{Whitespace} is one or more blanks or tabs, in any order.
1625 Whitespace is used to separate symbols, and to make programs neater for
1626 people to read. Unless within character constants
1627 (@pxref{Characters,,Character Constants}), any whitespace means the same
1628 as exactly one space.
1634 There are two ways of rendering comments to @code{@value{AS}}. In both
1635 cases the comment is equivalent to one space.
1637 Anything from @samp{/*} through the next @samp{*/} is a comment.
1638 This means you may not nest these comments.
1642 The only way to include a newline ('\n') in a comment
1643 is to use this sort of comment.
1646 /* This sort of comment does not nest. */
1649 @cindex line comment character
1650 Anything from the @dfn{line comment} character to the next newline
1651 is considered a comment and is ignored. The line comment character is
1653 @samp{;} for the AMD 29K family;
1656 @samp{;} on the ARC;
1659 @samp{@@} on the ARM;
1662 @samp{;} for the H8/300 family;
1665 @samp{!} for the H8/500 family;
1668 @samp{;} for the HPPA;
1671 @samp{#} on the i960;
1674 @samp{;} for picoJava;
1677 @samp{!} for the Hitachi SH;
1680 @samp{!} on the SPARC;
1683 @samp{#} on the m32r;
1686 @samp{|} on the 680x0;
1689 @samp{#} on the 68HC11 and 68HC12;
1692 @samp{#} on the Vax;
1695 @samp{!} for the Z8000;
1698 @samp{#} on the V850;
1700 see @ref{Machine Dependencies}. @refill
1701 @c FIXME What about i386, m88k, i860?
1704 On some machines there are two different line comment characters. One
1705 character only begins a comment if it is the first non-whitespace character on
1706 a line, while the other always begins a comment.
1710 The V850 assembler also supports a double dash as starting a comment that
1711 extends to the end of the line.
1717 @cindex lines starting with @code{#}
1718 @cindex logical line numbers
1719 To be compatible with past assemblers, lines that begin with @samp{#} have a
1720 special interpretation. Following the @samp{#} should be an absolute
1721 expression (@pxref{Expressions}): the logical line number of the @emph{next}
1722 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
1723 new logical file name. The rest of the line, if any, should be whitespace.
1725 If the first non-whitespace characters on the line are not numeric,
1726 the line is ignored. (Just like a comment.)
1729 # This is an ordinary comment.
1730 # 42-6 "new_file_name" # New logical file name
1731 # This is logical line # 36.
1733 This feature is deprecated, and may disappear from future versions
1734 of @code{@value{AS}}.
1739 @cindex characters used in symbols
1740 @ifclear SPECIAL-SYMS
1741 A @dfn{symbol} is one or more characters chosen from the set of all
1742 letters (both upper and lower case), digits and the three characters
1748 A @dfn{symbol} is one or more characters chosen from the set of all
1749 letters (both upper and lower case), digits and the three characters
1750 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
1756 On most machines, you can also use @code{$} in symbol names; exceptions
1757 are noted in @ref{Machine Dependencies}.
1759 No symbol may begin with a digit. Case is significant.
1760 There is no length limit: all characters are significant. Symbols are
1761 delimited by characters not in that set, or by the beginning of a file
1762 (since the source program must end with a newline, the end of a file is
1763 not a possible symbol delimiter). @xref{Symbols}.
1764 @cindex length of symbols
1769 @cindex statements, structure of
1770 @cindex line separator character
1771 @cindex statement separator character
1773 @ifclear abnormal-separator
1774 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
1775 semicolon (@samp{;}). The newline or semicolon is considered part of
1776 the preceding statement. Newlines and semicolons within character
1777 constants are an exception: they do not end statements.
1779 @ifset abnormal-separator
1781 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
1782 sign (@samp{@@}). The newline or at sign is considered part of the
1783 preceding statement. Newlines and at signs within character constants
1784 are an exception: they do not end statements.
1787 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
1788 point (@samp{!}). The newline or exclamation point is considered part of the
1789 preceding statement. Newlines and exclamation points within character
1790 constants are an exception: they do not end statements.
1793 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
1794 H8/300) a dollar sign (@samp{$}); or (for the
1797 (@samp{;}). The newline or separator character is considered part of
1798 the preceding statement. Newlines and separators within character
1799 constants are an exception: they do not end statements.
1804 A @dfn{statement} ends at a newline character (@samp{\n}) or line
1805 separator character. (The line separator is usually @samp{;}, unless
1806 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
1807 newline or separator character is considered part of the preceding
1808 statement. Newlines and separators within character constants are an
1809 exception: they do not end statements.
1812 @cindex newline, required at file end
1813 @cindex EOF, newline must precede
1814 It is an error to end any statement with end-of-file: the last
1815 character of any input file should be a newline.@refill
1817 An empty statement is allowed, and may include whitespace. It is ignored.
1819 @cindex instructions and directives
1820 @cindex directives and instructions
1821 @c "key symbol" is not used elsewhere in the document; seems pedantic to
1822 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
1824 A statement begins with zero or more labels, optionally followed by a
1825 key symbol which determines what kind of statement it is. The key
1826 symbol determines the syntax of the rest of the statement. If the
1827 symbol begins with a dot @samp{.} then the statement is an assembler
1828 directive: typically valid for any computer. If the symbol begins with
1829 a letter the statement is an assembly language @dfn{instruction}: it
1830 assembles into a machine language instruction.
1832 Different versions of @code{@value{AS}} for different computers
1833 recognize different instructions. In fact, the same symbol may
1834 represent a different instruction in a different computer's assembly
1838 @cindex @code{:} (label)
1839 @cindex label (@code{:})
1840 A label is a symbol immediately followed by a colon (@code{:}).
1841 Whitespace before a label or after a colon is permitted, but you may not
1842 have whitespace between a label's symbol and its colon. @xref{Labels}.
1845 For HPPA targets, labels need not be immediately followed by a colon, but
1846 the definition of a label must begin in column zero. This also implies that
1847 only one label may be defined on each line.
1851 label: .directive followed by something
1852 another_label: # This is an empty statement.
1853 instruction operand_1, operand_2, @dots{}
1860 A constant is a number, written so that its value is known by
1861 inspection, without knowing any context. Like this:
1864 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
1865 .ascii "Ring the bell\7" # A string constant.
1866 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
1867 .float 0f-314159265358979323846264338327\
1868 95028841971.693993751E-40 # - pi, a flonum.
1873 * Characters:: Character Constants
1874 * Numbers:: Number Constants
1878 @subsection Character Constants
1880 @cindex character constants
1881 @cindex constants, character
1882 There are two kinds of character constants. A @dfn{character} stands
1883 for one character in one byte and its value may be used in
1884 numeric expressions. String constants (properly called string
1885 @emph{literals}) are potentially many bytes and their values may not be
1886 used in arithmetic expressions.
1890 * Chars:: Characters
1894 @subsubsection Strings
1896 @cindex string constants
1897 @cindex constants, string
1898 A @dfn{string} is written between double-quotes. It may contain
1899 double-quotes or null characters. The way to get special characters
1900 into a string is to @dfn{escape} these characters: precede them with
1901 a backslash @samp{\} character. For example @samp{\\} represents
1902 one backslash: the first @code{\} is an escape which tells
1903 @code{@value{AS}} to interpret the second character literally as a backslash
1904 (which prevents @code{@value{AS}} from recognizing the second @code{\} as an
1905 escape character). The complete list of escapes follows.
1907 @cindex escape codes, character
1908 @cindex character escape codes
1911 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
1913 @cindex @code{\b} (backspace character)
1914 @cindex backspace (@code{\b})
1916 Mnemonic for backspace; for ASCII this is octal code 010.
1919 @c Mnemonic for EOText; for ASCII this is octal code 004.
1921 @cindex @code{\f} (formfeed character)
1922 @cindex formfeed (@code{\f})
1924 Mnemonic for FormFeed; for ASCII this is octal code 014.
1926 @cindex @code{\n} (newline character)
1927 @cindex newline (@code{\n})
1929 Mnemonic for newline; for ASCII this is octal code 012.
1932 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
1934 @cindex @code{\r} (carriage return character)
1935 @cindex carriage return (@code{\r})
1937 Mnemonic for carriage-Return; for ASCII this is octal code 015.
1940 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
1941 @c other assemblers.
1943 @cindex @code{\t} (tab)
1944 @cindex tab (@code{\t})
1946 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
1949 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
1950 @c @item \x @var{digit} @var{digit} @var{digit}
1951 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
1953 @cindex @code{\@var{ddd}} (octal character code)
1954 @cindex octal character code (@code{\@var{ddd}})
1955 @item \ @var{digit} @var{digit} @var{digit}
1956 An octal character code. The numeric code is 3 octal digits.
1957 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
1958 for example, @code{\008} has the value 010, and @code{\009} the value 011.
1960 @cindex @code{\@var{xd...}} (hex character code)
1961 @cindex hex character code (@code{\@var{xd...}})
1962 @item \@code{x} @var{hex-digits...}
1963 A hex character code. All trailing hex digits are combined. Either upper or
1964 lower case @code{x} works.
1966 @cindex @code{\\} (@samp{\} character)
1967 @cindex backslash (@code{\\})
1969 Represents one @samp{\} character.
1972 @c Represents one @samp{'} (accent acute) character.
1973 @c This is needed in single character literals
1974 @c (@xref{Characters,,Character Constants}.) to represent
1977 @cindex @code{\"} (doublequote character)
1978 @cindex doublequote (@code{\"})
1980 Represents one @samp{"} character. Needed in strings to represent
1981 this character, because an unescaped @samp{"} would end the string.
1983 @item \ @var{anything-else}
1984 Any other character when escaped by @kbd{\} gives a warning, but
1985 assembles as if the @samp{\} was not present. The idea is that if
1986 you used an escape sequence you clearly didn't want the literal
1987 interpretation of the following character. However @code{@value{AS}} has no
1988 other interpretation, so @code{@value{AS}} knows it is giving you the wrong
1989 code and warns you of the fact.
1992 Which characters are escapable, and what those escapes represent,
1993 varies widely among assemblers. The current set is what we think
1994 the BSD 4.2 assembler recognizes, and is a subset of what most C
1995 compilers recognize. If you are in doubt, do not use an escape
1999 @subsubsection Characters
2001 @cindex single character constant
2002 @cindex character, single
2003 @cindex constant, single character
2004 A single character may be written as a single quote immediately
2005 followed by that character. The same escapes apply to characters as
2006 to strings. So if you want to write the character backslash, you
2007 must write @kbd{'\\} where the first @code{\} escapes the second
2008 @code{\}. As you can see, the quote is an acute accent, not a
2009 grave accent. A newline
2011 @ifclear abnormal-separator
2012 (or semicolon @samp{;})
2014 @ifset abnormal-separator
2016 (or at sign @samp{@@})
2019 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2025 immediately following an acute accent is taken as a literal character
2026 and does not count as the end of a statement. The value of a character
2027 constant in a numeric expression is the machine's byte-wide code for
2028 that character. @code{@value{AS}} assumes your character code is ASCII:
2029 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2032 @subsection Number Constants
2034 @cindex constants, number
2035 @cindex number constants
2036 @code{@value{AS}} distinguishes three kinds of numbers according to how they
2037 are stored in the target machine. @emph{Integers} are numbers that
2038 would fit into an @code{int} in the C language. @emph{Bignums} are
2039 integers, but they are stored in more than 32 bits. @emph{Flonums}
2040 are floating point numbers, described below.
2043 * Integers:: Integers
2048 * Bit Fields:: Bit Fields
2054 @subsubsection Integers
2056 @cindex constants, integer
2058 @cindex binary integers
2059 @cindex integers, binary
2060 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2061 the binary digits @samp{01}.
2063 @cindex octal integers
2064 @cindex integers, octal
2065 An octal integer is @samp{0} followed by zero or more of the octal
2066 digits (@samp{01234567}).
2068 @cindex decimal integers
2069 @cindex integers, decimal
2070 A decimal integer starts with a non-zero digit followed by zero or
2071 more digits (@samp{0123456789}).
2073 @cindex hexadecimal integers
2074 @cindex integers, hexadecimal
2075 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2076 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2078 Integers have the usual values. To denote a negative integer, use
2079 the prefix operator @samp{-} discussed under expressions
2080 (@pxref{Prefix Ops,,Prefix Operators}).
2083 @subsubsection Bignums
2086 @cindex constants, bignum
2087 A @dfn{bignum} has the same syntax and semantics as an integer
2088 except that the number (or its negative) takes more than 32 bits to
2089 represent in binary. The distinction is made because in some places
2090 integers are permitted while bignums are not.
2093 @subsubsection Flonums
2095 @cindex floating point numbers
2096 @cindex constants, floating point
2098 @cindex precision, floating point
2099 A @dfn{flonum} represents a floating point number. The translation is
2100 indirect: a decimal floating point number from the text is converted by
2101 @code{@value{AS}} to a generic binary floating point number of more than
2102 sufficient precision. This generic floating point number is converted
2103 to a particular computer's floating point format (or formats) by a
2104 portion of @code{@value{AS}} specialized to that computer.
2106 A flonum is written by writing (in order)
2111 (@samp{0} is optional on the HPPA.)
2115 A letter, to tell @code{@value{AS}} the rest of the number is a flonum.
2117 @kbd{e} is recommended. Case is not important.
2119 @c FIXME: verify if flonum syntax really this vague for most cases
2120 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2121 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2124 On the H8/300, H8/500,
2126 and AMD 29K architectures, the letter must be
2127 one of the letters @samp{DFPRSX} (in upper or lower case).
2129 On the ARC, the letter must be one of the letters @samp{DFRS}
2130 (in upper or lower case).
2132 On the Intel 960 architecture, the letter must be
2133 one of the letters @samp{DFT} (in upper or lower case).
2135 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2139 One of the letters @samp{DFPRSX} (in upper or lower case).
2142 One of the letters @samp{DFRS} (in upper or lower case).
2145 One of the letters @samp{DFPRSX} (in upper or lower case).
2148 The letter @samp{E} (upper case only).
2151 One of the letters @samp{DFT} (in upper or lower case).
2156 An optional sign: either @samp{+} or @samp{-}.
2159 An optional @dfn{integer part}: zero or more decimal digits.
2162 An optional @dfn{fractional part}: @samp{.} followed by zero
2163 or more decimal digits.
2166 An optional exponent, consisting of:
2170 An @samp{E} or @samp{e}.
2171 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2172 @c principle this can perfectly well be different on different targets.
2174 Optional sign: either @samp{+} or @samp{-}.
2176 One or more decimal digits.
2181 At least one of the integer part or the fractional part must be
2182 present. The floating point number has the usual base-10 value.
2184 @code{@value{AS}} does all processing using integers. Flonums are computed
2185 independently of any floating point hardware in the computer running
2190 @c Bit fields are written as a general facility but are also controlled
2191 @c by a conditional-compilation flag---which is as of now (21mar91)
2192 @c turned on only by the i960 config of GAS.
2194 @subsubsection Bit Fields
2197 @cindex constants, bit field
2198 You can also define numeric constants as @dfn{bit fields}.
2199 specify two numbers separated by a colon---
2201 @var{mask}:@var{value}
2204 @code{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2207 The resulting number is then packed
2209 @c this conditional paren in case bit fields turned on elsewhere than 960
2210 (in host-dependent byte order)
2212 into a field whose width depends on which assembler directive has the
2213 bit-field as its argument. Overflow (a result from the bitwise and
2214 requiring more binary digits to represent) is not an error; instead,
2215 more constants are generated, of the specified width, beginning with the
2216 least significant digits.@refill
2218 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2219 @code{.short}, and @code{.word} accept bit-field arguments.
2224 @chapter Sections and Relocation
2229 * Secs Background:: Background
2230 * Ld Sections:: Linker Sections
2231 * As Sections:: Assembler Internal Sections
2232 * Sub-Sections:: Sub-Sections
2236 @node Secs Background
2239 Roughly, a section is a range of addresses, with no gaps; all data
2240 ``in'' those addresses is treated the same for some particular purpose.
2241 For example there may be a ``read only'' section.
2243 @cindex linker, and assembler
2244 @cindex assembler, and linker
2245 The linker @code{@value{LD}} reads many object files (partial programs) and
2246 combines their contents to form a runnable program. When @code{@value{AS}}
2247 emits an object file, the partial program is assumed to start at address 0.
2248 @code{@value{LD}} assigns the final addresses for the partial program, so that
2249 different partial programs do not overlap. This is actually an
2250 oversimplification, but it suffices to explain how @code{@value{AS}} uses
2253 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2254 addresses. These blocks slide to their run-time addresses as rigid
2255 units; their length does not change and neither does the order of bytes
2256 within them. Such a rigid unit is called a @emph{section}. Assigning
2257 run-time addresses to sections is called @dfn{relocation}. It includes
2258 the task of adjusting mentions of object-file addresses so they refer to
2259 the proper run-time addresses.
2261 For the H8/300 and H8/500,
2262 and for the Hitachi SH,
2263 @code{@value{AS}} pads sections if needed to
2264 ensure they end on a word (sixteen bit) boundary.
2267 @cindex standard assembler sections
2268 An object file written by @code{@value{AS}} has at least three sections, any
2269 of which may be empty. These are named @dfn{text}, @dfn{data} and
2274 When it generates COFF output,
2276 @code{@value{AS}} can also generate whatever other named sections you specify
2277 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2278 If you do not use any directives that place output in the @samp{.text}
2279 or @samp{.data} sections, these sections still exist, but are empty.
2284 When @code{@value{AS}} generates SOM or ELF output for the HPPA,
2286 @code{@value{AS}} can also generate whatever other named sections you
2287 specify using the @samp{.space} and @samp{.subspace} directives. See
2288 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2289 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2290 assembler directives.
2293 Additionally, @code{@value{AS}} uses different names for the standard
2294 text, data, and bss sections when generating SOM output. Program text
2295 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2296 BSS into @samp{$BSS$}.
2300 Within the object file, the text section starts at address @code{0}, the
2301 data section follows, and the bss section follows the data section.
2304 When generating either SOM or ELF output files on the HPPA, the text
2305 section starts at address @code{0}, the data section at address
2306 @code{0x4000000}, and the bss section follows the data section.
2309 To let @code{@value{LD}} know which data changes when the sections are
2310 relocated, and how to change that data, @code{@value{AS}} also writes to the
2311 object file details of the relocation needed. To perform relocation
2312 @code{@value{LD}} must know, each time an address in the object
2316 Where in the object file is the beginning of this reference to
2319 How long (in bytes) is this reference?
2321 Which section does the address refer to? What is the numeric value of
2323 (@var{address}) @minus{} (@var{start-address of section})?
2326 Is the reference to an address ``Program-Counter relative''?
2329 @cindex addresses, format of
2330 @cindex section-relative addressing
2331 In fact, every address @code{@value{AS}} ever uses is expressed as
2333 (@var{section}) + (@var{offset into section})
2336 Further, most expressions @code{@value{AS}} computes have this section-relative
2339 (For some object formats, such as SOM for the HPPA, some expressions are
2340 symbol-relative instead.)
2343 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2344 @var{N} into section @var{secname}.''
2346 Apart from text, data and bss sections you need to know about the
2347 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2348 addresses in the absolute section remain unchanged. For example, address
2349 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2350 @code{@value{LD}}. Although the linker never arranges two partial programs'
2351 data sections with overlapping addresses after linking, @emph{by definition}
2352 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2353 part of a program is always the same address when the program is running as
2354 address @code{@{absolute@ 239@}} in any other part of the program.
2356 The idea of sections is extended to the @dfn{undefined} section. Any
2357 address whose section is unknown at assembly time is by definition
2358 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2359 Since numbers are always defined, the only way to generate an undefined
2360 address is to mention an undefined symbol. A reference to a named
2361 common block would be such a symbol: its value is unknown at assembly
2362 time so it has section @emph{undefined}.
2364 By analogy the word @emph{section} is used to describe groups of sections in
2365 the linked program. @code{@value{LD}} puts all partial programs' text
2366 sections in contiguous addresses in the linked program. It is
2367 customary to refer to the @emph{text section} of a program, meaning all
2368 the addresses of all partial programs' text sections. Likewise for
2369 data and bss sections.
2371 Some sections are manipulated by @code{@value{LD}}; others are invented for
2372 use of @code{@value{AS}} and have no meaning except during assembly.
2375 @section Linker Sections
2376 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2381 @cindex named sections
2382 @cindex sections, named
2383 @item named sections
2386 @cindex text section
2387 @cindex data section
2391 These sections hold your program. @code{@value{AS}} and @code{@value{LD}} treat them as
2392 separate but equal sections. Anything you can say of one section is
2395 When the program is running, however, it is
2396 customary for the text section to be unalterable. The
2397 text section is often shared among processes: it contains
2398 instructions, constants and the like. The data section of a running
2399 program is usually alterable: for example, C variables would be stored
2400 in the data section.
2405 This section contains zeroed bytes when your program begins running. It
2406 is used to hold unitialized variables or common storage. The length of
2407 each partial program's bss section is important, but because it starts
2408 out containing zeroed bytes there is no need to store explicit zero
2409 bytes in the object file. The bss section was invented to eliminate
2410 those explicit zeros from object files.
2412 @cindex absolute section
2413 @item absolute section
2414 Address 0 of this section is always ``relocated'' to runtime address 0.
2415 This is useful if you want to refer to an address that @code{@value{LD}} must
2416 not change when relocating. In this sense we speak of absolute
2417 addresses being ``unrelocatable'': they do not change during relocation.
2419 @cindex undefined section
2420 @item undefined section
2421 This ``section'' is a catch-all for address references to objects not in
2422 the preceding sections.
2423 @c FIXME: ref to some other doc on obj-file formats could go here.
2426 @cindex relocation example
2427 An idealized example of three relocatable sections follows.
2429 The example uses the traditional section names @samp{.text} and @samp{.data}.
2431 Memory addresses are on the horizontal axis.
2435 @c END TEXI2ROFF-KILL
2438 partial program # 1: |ttttt|dddd|00|
2445 partial program # 2: |TTT|DDD|000|
2448 +--+---+-----+--+----+---+-----+~~
2449 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2450 +--+---+-----+--+----+---+-----+~~
2452 addresses: 0 @dots{}
2459 \line{\it Partial program \#1: \hfil}
2460 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2461 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2463 \line{\it Partial program \#2: \hfil}
2464 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2465 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2467 \line{\it linked program: \hfil}
2468 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2469 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2470 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2471 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2473 \line{\it addresses: \hfil}
2477 @c END TEXI2ROFF-KILL
2480 @section Assembler Internal Sections
2482 @cindex internal assembler sections
2483 @cindex sections in messages, internal
2484 These sections are meant only for the internal use of @code{@value{AS}}. They
2485 have no meaning at run-time. You do not really need to know about these
2486 sections for most purposes; but they can be mentioned in @code{@value{AS}}
2487 warning messages, so it might be helpful to have an idea of their
2488 meanings to @code{@value{AS}}. These sections are used to permit the
2489 value of every expression in your assembly language program to be a
2490 section-relative address.
2493 @cindex assembler internal logic error
2494 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2495 An internal assembler logic error has been found. This means there is a
2496 bug in the assembler.
2498 @cindex expr (internal section)
2500 The assembler stores complex expression internally as combinations of
2501 symbols. When it needs to represent an expression as a symbol, it puts
2502 it in the expr section.
2504 @c FIXME item transfer[t] vector preload
2505 @c FIXME item transfer[t] vector postload
2506 @c FIXME item register
2510 @section Sub-Sections
2512 @cindex numbered subsections
2513 @cindex grouping data
2519 fall into two sections: text and data.
2521 You may have separate groups of
2523 data in named sections
2527 data in named sections
2533 that you want to end up near to each other in the object file, even though they
2534 are not contiguous in the assembler source. @code{@value{AS}} allows you to
2535 use @dfn{subsections} for this purpose. Within each section, there can be
2536 numbered subsections with values from 0 to 8192. Objects assembled into the
2537 same subsection go into the object file together with other objects in the same
2538 subsection. For example, a compiler might want to store constants in the text
2539 section, but might not want to have them interspersed with the program being
2540 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2541 section of code being output, and a @samp{.text 1} before each group of
2542 constants being output.
2544 Subsections are optional. If you do not use subsections, everything
2545 goes in subsection number zero.
2548 Each subsection is zero-padded up to a multiple of four bytes.
2549 (Subsections may be padded a different amount on different flavors
2550 of @code{@value{AS}}.)
2554 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2555 boundary (two bytes).
2556 The same is true on the Hitachi SH.
2559 @c FIXME section padding (alignment)?
2560 @c Rich Pixley says padding here depends on target obj code format; that
2561 @c doesn't seem particularly useful to say without further elaboration,
2562 @c so for now I say nothing about it. If this is a generic BFD issue,
2563 @c these paragraphs might need to vanish from this manual, and be
2564 @c discussed in BFD chapter of binutils (or some such).
2567 On the AMD 29K family, no particular padding is added to section or
2568 subsection sizes; @value{AS} forces no alignment on this platform.
2572 Subsections appear in your object file in numeric order, lowest numbered
2573 to highest. (All this to be compatible with other people's assemblers.)
2574 The object file contains no representation of subsections; @code{@value{LD}} and
2575 other programs that manipulate object files see no trace of them.
2576 They just see all your text subsections as a text section, and all your
2577 data subsections as a data section.
2579 To specify which subsection you want subsequent statements assembled
2580 into, use a numeric argument to specify it, in a @samp{.text
2581 @var{expression}} or a @samp{.data @var{expression}} statement.
2584 When generating COFF output, you
2589 can also use an extra subsection
2590 argument with arbitrary named sections: @samp{.section @var{name},
2593 @var{Expression} should be an absolute expression.
2594 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
2595 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
2596 begins in @code{text 0}. For instance:
2598 .text 0 # The default subsection is text 0 anyway.
2599 .ascii "This lives in the first text subsection. *"
2601 .ascii "But this lives in the second text subsection."
2603 .ascii "This lives in the data section,"
2604 .ascii "in the first data subsection."
2606 .ascii "This lives in the first text section,"
2607 .ascii "immediately following the asterisk (*)."
2610 Each section has a @dfn{location counter} incremented by one for every byte
2611 assembled into that section. Because subsections are merely a convenience
2612 restricted to @code{@value{AS}} there is no concept of a subsection location
2613 counter. There is no way to directly manipulate a location counter---but the
2614 @code{.align} directive changes it, and any label definition captures its
2615 current value. The location counter of the section where statements are being
2616 assembled is said to be the @dfn{active} location counter.
2619 @section bss Section
2622 @cindex common variable storage
2623 The bss section is used for local common variable storage.
2624 You may allocate address space in the bss section, but you may
2625 not dictate data to load into it before your program executes. When
2626 your program starts running, all the contents of the bss
2627 section are zeroed bytes.
2629 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
2630 @ref{Lcomm,,@code{.lcomm}}.
2632 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
2633 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
2636 When assembling for a target which supports multiple sections, such as ELF or
2637 COFF, you may switch into the @code{.bss} section and define symbols as usual;
2638 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
2639 section. Typically the section will only contain symbol definitions and
2640 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
2647 Symbols are a central concept: the programmer uses symbols to name
2648 things, the linker uses symbols to link, and the debugger uses symbols
2652 @cindex debuggers, and symbol order
2653 @emph{Warning:} @code{@value{AS}} does not place symbols in the object file in
2654 the same order they were declared. This may break some debuggers.
2659 * Setting Symbols:: Giving Symbols Other Values
2660 * Symbol Names:: Symbol Names
2661 * Dot:: The Special Dot Symbol
2662 * Symbol Attributes:: Symbol Attributes
2669 A @dfn{label} is written as a symbol immediately followed by a colon
2670 @samp{:}. The symbol then represents the current value of the
2671 active location counter, and is, for example, a suitable instruction
2672 operand. You are warned if you use the same symbol to represent two
2673 different locations: the first definition overrides any other
2677 On the HPPA, the usual form for a label need not be immediately followed by a
2678 colon, but instead must start in column zero. Only one label may be defined on
2679 a single line. To work around this, the HPPA version of @code{@value{AS}} also
2680 provides a special directive @code{.label} for defining labels more flexibly.
2683 @node Setting Symbols
2684 @section Giving Symbols Other Values
2686 @cindex assigning values to symbols
2687 @cindex symbol values, assigning
2688 A symbol can be given an arbitrary value by writing a symbol, followed
2689 by an equals sign @samp{=}, followed by an expression
2690 (@pxref{Expressions}). This is equivalent to using the @code{.set}
2691 directive. @xref{Set,,@code{.set}}.
2694 @section Symbol Names
2696 @cindex symbol names
2697 @cindex names, symbol
2698 @ifclear SPECIAL-SYMS
2699 Symbol names begin with a letter or with one of @samp{._}. On most
2700 machines, you can also use @code{$} in symbol names; exceptions are
2701 noted in @ref{Machine Dependencies}. That character may be followed by any
2702 string of digits, letters, dollar signs (unless otherwise noted in
2703 @ref{Machine Dependencies}), and underscores.
2706 For the AMD 29K family, @samp{?} is also allowed in the
2707 body of a symbol name, though not at its beginning.
2712 Symbol names begin with a letter or with one of @samp{._}. On the
2714 H8/500, you can also use @code{$} in symbol names. That character may
2715 be followed by any string of digits, letters, dollar signs (save on the
2716 H8/300), and underscores.
2720 Case of letters is significant: @code{foo} is a different symbol name
2723 Each symbol has exactly one name. Each name in an assembly language program
2724 refers to exactly one symbol. You may use that symbol name any number of times
2727 @subheading Local Symbol Names
2729 @cindex local symbol names
2730 @cindex symbol names, local
2731 @cindex temporary symbol names
2732 @cindex symbol names, temporary
2733 Local symbols help compilers and programmers use names temporarily.
2734 There are ten local symbol names, which are re-used throughout the
2735 program. You may refer to them using the names @samp{0} @samp{1}
2736 @dots{} @samp{9}. To define a local symbol, write a label of the form
2737 @samp{@b{N}:} (where @b{N} represents any digit). To refer to the most
2738 recent previous definition of that symbol write @samp{@b{N}b}, using the
2739 same digit as when you defined the label. To refer to the next
2740 definition of a local label, write @samp{@b{N}f}---where @b{N} gives you
2741 a choice of 10 forward references. The @samp{b} stands for
2742 ``backwards'' and the @samp{f} stands for ``forwards''.
2744 Local symbols are not emitted by the current @sc{gnu} C compiler.
2746 There is no restriction on how you can use these labels, but
2747 remember that at any point in the assembly you can refer to at most
2748 10 prior local labels and to at most 10 forward local labels.
2750 Local symbol names are only a notation device. They are immediately
2751 transformed into more conventional symbol names before the assembler
2752 uses them. The symbol names stored in the symbol table, appearing in
2753 error messages and optionally emitted to the object file have these
2758 All local labels begin with @samp{L}. Normally both @code{@value{AS}} and
2759 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
2760 used for symbols you are never intended to see. If you use the
2761 @samp{-L} option then @code{@value{AS}} retains these symbols in the
2762 object file. If you also instruct @code{@value{LD}} to retain these symbols,
2763 you may use them in debugging.
2766 If the label is written @samp{0:} then the digit is @samp{0}.
2767 If the label is written @samp{1:} then the digit is @samp{1}.
2768 And so on up through @samp{9:}.
2771 This unusual character is included so you do not accidentally invent
2772 a symbol of the same name. The character has ASCII value
2775 @item @emph{ordinal number}
2776 This is a serial number to keep the labels distinct. The first
2777 @samp{0:} gets the number @samp{1}; The 15th @samp{0:} gets the
2778 number @samp{15}; @emph{etc.}. Likewise for the other labels @samp{1:}
2782 For instance, the first @code{1:} is named @code{L1@kbd{C-A}1}, the 44th
2783 @code{3:} is named @code{L3@kbd{C-A}44}.
2786 @section The Special Dot Symbol
2788 @cindex dot (symbol)
2789 @cindex @code{.} (symbol)
2790 @cindex current address
2791 @cindex location counter
2792 The special symbol @samp{.} refers to the current address that
2793 @code{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
2794 .long .} defines @code{melvin} to contain its own address.
2795 Assigning a value to @code{.} is treated the same as a @code{.org}
2796 directive. Thus, the expression @samp{.=.+4} is the same as saying
2797 @ifclear no-space-dir
2806 @node Symbol Attributes
2807 @section Symbol Attributes
2809 @cindex symbol attributes
2810 @cindex attributes, symbol
2811 Every symbol has, as well as its name, the attributes ``Value'' and
2812 ``Type''. Depending on output format, symbols can also have auxiliary
2815 The detailed definitions are in @file{a.out.h}.
2818 If you use a symbol without defining it, @code{@value{AS}} assumes zero for
2819 all these attributes, and probably won't warn you. This makes the
2820 symbol an externally defined symbol, which is generally what you
2824 * Symbol Value:: Value
2825 * Symbol Type:: Type
2828 * a.out Symbols:: Symbol Attributes: @code{a.out}
2832 * a.out Symbols:: Symbol Attributes: @code{a.out}
2835 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
2840 * COFF Symbols:: Symbol Attributes for COFF
2843 * SOM Symbols:: Symbol Attributes for SOM
2850 @cindex value of a symbol
2851 @cindex symbol value
2852 The value of a symbol is (usually) 32 bits. For a symbol which labels a
2853 location in the text, data, bss or absolute sections the value is the
2854 number of addresses from the start of that section to the label.
2855 Naturally for text, data and bss sections the value of a symbol changes
2856 as @code{@value{LD}} changes section base addresses during linking. Absolute
2857 symbols' values do not change during linking: that is why they are
2860 The value of an undefined symbol is treated in a special way. If it is
2861 0 then the symbol is not defined in this assembler source file, and
2862 @code{@value{LD}} tries to determine its value from other files linked into the
2863 same program. You make this kind of symbol simply by mentioning a symbol
2864 name without defining it. A non-zero value represents a @code{.comm}
2865 common declaration. The value is how much common storage to reserve, in
2866 bytes (addresses). The symbol refers to the first address of the
2872 @cindex type of a symbol
2874 The type attribute of a symbol contains relocation (section)
2875 information, any flag settings indicating that a symbol is external, and
2876 (optionally), other information for linkers and debuggers. The exact
2877 format depends on the object-code output format in use.
2882 @c The following avoids a "widow" subsection title. @group would be
2883 @c better if it were available outside examples.
2886 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
2888 @cindex @code{b.out} symbol attributes
2889 @cindex symbol attributes, @code{b.out}
2890 These symbol attributes appear only when @code{@value{AS}} is configured for
2891 one of the Berkeley-descended object output formats---@code{a.out} or
2897 @subsection Symbol Attributes: @code{a.out}
2899 @cindex @code{a.out} symbol attributes
2900 @cindex symbol attributes, @code{a.out}
2906 @subsection Symbol Attributes: @code{a.out}
2908 @cindex @code{a.out} symbol attributes
2909 @cindex symbol attributes, @code{a.out}
2913 * Symbol Desc:: Descriptor
2914 * Symbol Other:: Other
2918 @subsubsection Descriptor
2920 @cindex descriptor, of @code{a.out} symbol
2921 This is an arbitrary 16-bit value. You may establish a symbol's
2922 descriptor value by using a @code{.desc} statement
2923 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
2927 @subsubsection Other
2929 @cindex other attribute, of @code{a.out} symbol
2930 This is an arbitrary 8-bit value. It means nothing to @code{@value{AS}}.
2935 @subsection Symbol Attributes for COFF
2937 @cindex COFF symbol attributes
2938 @cindex symbol attributes, COFF
2940 The COFF format supports a multitude of auxiliary symbol attributes;
2941 like the primary symbol attributes, they are set between @code{.def} and
2942 @code{.endef} directives.
2944 @subsubsection Primary Attributes
2946 @cindex primary attributes, COFF symbols
2947 The symbol name is set with @code{.def}; the value and type,
2948 respectively, with @code{.val} and @code{.type}.
2950 @subsubsection Auxiliary Attributes
2952 @cindex auxiliary attributes, COFF symbols
2953 The @code{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
2954 @code{.size}, and @code{.tag} can generate auxiliary symbol table
2955 information for COFF.
2960 @subsection Symbol Attributes for SOM
2962 @cindex SOM symbol attributes
2963 @cindex symbol attributes, SOM
2965 The SOM format for the HPPA supports a multitude of symbol attributes set with
2966 the @code{.EXPORT} and @code{.IMPORT} directives.
2968 The attributes are described in @cite{HP9000 Series 800 Assembly
2969 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
2970 @code{EXPORT} assembler directive documentation.
2974 @chapter Expressions
2978 @cindex numeric values
2979 An @dfn{expression} specifies an address or numeric value.
2980 Whitespace may precede and/or follow an expression.
2982 The result of an expression must be an absolute number, or else an offset into
2983 a particular section. If an expression is not absolute, and there is not
2984 enough information when @code{@value{AS}} sees the expression to know its
2985 section, a second pass over the source program might be necessary to interpret
2986 the expression---but the second pass is currently not implemented.
2987 @code{@value{AS}} aborts with an error message in this situation.
2990 * Empty Exprs:: Empty Expressions
2991 * Integer Exprs:: Integer Expressions
2995 @section Empty Expressions
2997 @cindex empty expressions
2998 @cindex expressions, empty
2999 An empty expression has no value: it is just whitespace or null.
3000 Wherever an absolute expression is required, you may omit the
3001 expression, and @code{@value{AS}} assumes a value of (absolute) 0. This
3002 is compatible with other assemblers.
3005 @section Integer Expressions
3007 @cindex integer expressions
3008 @cindex expressions, integer
3009 An @dfn{integer expression} is one or more @emph{arguments} delimited
3010 by @emph{operators}.
3013 * Arguments:: Arguments
3014 * Operators:: Operators
3015 * Prefix Ops:: Prefix Operators
3016 * Infix Ops:: Infix Operators
3020 @subsection Arguments
3022 @cindex expression arguments
3023 @cindex arguments in expressions
3024 @cindex operands in expressions
3025 @cindex arithmetic operands
3026 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3027 contexts arguments are sometimes called ``arithmetic operands''. In
3028 this manual, to avoid confusing them with the ``instruction operands'' of
3029 the machine language, we use the term ``argument'' to refer to parts of
3030 expressions only, reserving the word ``operand'' to refer only to machine
3031 instruction operands.
3033 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3034 @var{section} is one of text, data, bss, absolute,
3035 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3038 Numbers are usually integers.
3040 A number can be a flonum or bignum. In this case, you are warned
3041 that only the low order 32 bits are used, and @code{@value{AS}} pretends
3042 these 32 bits are an integer. You may write integer-manipulating
3043 instructions that act on exotic constants, compatible with other
3046 @cindex subexpressions
3047 Subexpressions are a left parenthesis @samp{(} followed by an integer
3048 expression, followed by a right parenthesis @samp{)}; or a prefix
3049 operator followed by an argument.
3052 @subsection Operators
3054 @cindex operators, in expressions
3055 @cindex arithmetic functions
3056 @cindex functions, in expressions
3057 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3058 operators are followed by an argument. Infix operators appear
3059 between their arguments. Operators may be preceded and/or followed by
3063 @subsection Prefix Operator
3065 @cindex prefix operators
3066 @code{@value{AS}} has the following @dfn{prefix operators}. They each take
3067 one argument, which must be absolute.
3069 @c the tex/end tex stuff surrounding this small table is meant to make
3070 @c it align, on the printed page, with the similar table in the next
3071 @c section (which is inside an enumerate).
3073 \global\advance\leftskip by \itemindent
3078 @dfn{Negation}. Two's complement negation.
3080 @dfn{Complementation}. Bitwise not.
3084 \global\advance\leftskip by -\itemindent
3088 @subsection Infix Operators
3090 @cindex infix operators
3091 @cindex operators, permitted arguments
3092 @dfn{Infix operators} take two arguments, one on either side. Operators
3093 have precedence, but operations with equal precedence are performed left
3094 to right. Apart from @code{+} or @code{-}, both arguments must be
3095 absolute, and the result is absolute.
3098 @cindex operator precedence
3099 @cindex precedence of operators
3106 @dfn{Multiplication}.
3109 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3116 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3120 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3124 Intermediate precedence
3129 @dfn{Bitwise Inclusive Or}.
3135 @dfn{Bitwise Exclusive Or}.
3138 @dfn{Bitwise Or Not}.
3145 @cindex addition, permitted arguments
3146 @cindex plus, permitted arguments
3147 @cindex arguments for addition
3149 @dfn{Addition}. If either argument is absolute, the result has the section of
3150 the other argument. You may not add together arguments from different
3153 @cindex subtraction, permitted arguments
3154 @cindex minus, permitted arguments
3155 @cindex arguments for subtraction
3157 @dfn{Subtraction}. If the right argument is absolute, the
3158 result has the section of the left argument.
3159 If both arguments are in the same section, the result is absolute.
3160 You may not subtract arguments from different sections.
3161 @c FIXME is there still something useful to say about undefined - undefined ?
3165 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3166 address; you can only have a defined section in one of the two arguments.
3169 @chapter Assembler Directives
3171 @cindex directives, machine independent
3172 @cindex pseudo-ops, machine independent
3173 @cindex machine independent directives
3174 All assembler directives have names that begin with a period (@samp{.}).
3175 The rest of the name is letters, usually in lower case.
3177 This chapter discusses directives that are available regardless of the
3178 target machine configuration for the @sc{gnu} assembler.
3180 Some machine configurations provide additional directives.
3181 @xref{Machine Dependencies}.
3184 @ifset machine-directives
3185 @xref{Machine Dependencies} for additional directives.
3190 * Abort:: @code{.abort}
3192 * ABORT:: @code{.ABORT}
3195 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3196 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3197 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3198 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3199 * Byte:: @code{.byte @var{expressions}}
3200 * Comm:: @code{.comm @var{symbol} , @var{length} }
3201 * Data:: @code{.data @var{subsection}}
3203 * Def:: @code{.def @var{name}}
3206 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3212 * Double:: @code{.double @var{flonums}}
3213 * Eject:: @code{.eject}
3214 * Else:: @code{.else}
3215 * Elseif:: @code{.elseif}
3218 * Endef:: @code{.endef}
3221 * Endfunc:: @code{.endfunc}
3222 * Endif:: @code{.endif}
3223 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3224 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3226 * Exitm:: @code{.exitm}
3227 * Extern:: @code{.extern}
3228 * Fail:: @code{.fail}
3229 @ifclear no-file-dir
3230 * File:: @code{.file @var{string}}
3233 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3234 * Float:: @code{.float @var{flonums}}
3235 * Func:: @code{.func}
3236 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3238 * Hidden:: @code{.hidden @var{names}}
3241 * hword:: @code{.hword @var{expressions}}
3242 * Ident:: @code{.ident}
3243 * If:: @code{.if @var{absolute expression}}
3244 * Include:: @code{.include "@var{file}"}
3245 * Int:: @code{.int @var{expressions}}
3247 * Internal:: @code{.internal @var{names}}
3250 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3251 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3252 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3253 * Lflags:: @code{.lflags}
3254 @ifclear no-line-dir
3255 * Line:: @code{.line @var{line-number}}
3258 * Ln:: @code{.ln @var{line-number}}
3259 * Linkonce:: @code{.linkonce [@var{type}]}
3260 * List:: @code{.list}
3261 * Long:: @code{.long @var{expressions}}
3263 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3266 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3267 * MRI:: @code{.mri @var{val}}
3268 * Nolist:: @code{.nolist}
3269 * Octa:: @code{.octa @var{bignums}}
3270 * Org:: @code{.org @var{new-lc} , @var{fill}}
3271 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3273 * PopSection:: @code{.popsection}
3274 * Previous:: @code{.previous}
3277 * Print:: @code{.print @var{string}}
3279 * Protected:: @code{.protected @var{names}}
3282 * Psize:: @code{.psize @var{lines}, @var{columns}}
3283 * Purgem:: @code{.purgem @var{name}}
3285 * PushSection:: @code{.pushsection @var{name}}
3288 * Quad:: @code{.quad @var{bignums}}
3289 * Rept:: @code{.rept @var{count}}
3290 * Sbttl:: @code{.sbttl "@var{subheading}"}
3292 * Scl:: @code{.scl @var{class}}
3293 * Section:: @code{.section @var{name}, @var{subsection}}
3296 * Set:: @code{.set @var{symbol}, @var{expression}}
3297 * Short:: @code{.short @var{expressions}}
3298 * Single:: @code{.single @var{flonums}}
3299 * Size:: @code{.size [@var{name} , @var{expression}]}
3300 * Skip:: @code{.skip @var{size} , @var{fill}}
3301 * Sleb128:: @code{.sleb128 @var{expressions}}
3302 * Space:: @code{.space @var{size} , @var{fill}}
3304 * Stab:: @code{.stabd, .stabn, .stabs}
3307 * String:: @code{.string "@var{str}"}
3308 * Struct:: @code{.struct @var{expression}}
3310 * SubSection:: @code{.subsection}
3311 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3315 * Tag:: @code{.tag @var{structname}}
3318 * Text:: @code{.text @var{subsection}}
3319 * Title:: @code{.title "@var{heading}"}
3320 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3321 * Uleb128:: @code{.uleb128 @var{expressions}}
3323 * Val:: @code{.val @var{addr}}
3327 * Version:: @code{.version "@var{string}"}
3328 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3329 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3330 * Weak:: @code{.weak @var{names}}
3333 * Word:: @code{.word @var{expressions}}
3334 * Deprecated:: Deprecated Directives
3338 @section @code{.abort}
3340 @cindex @code{abort} directive
3341 @cindex stopping the assembly
3342 This directive stops the assembly immediately. It is for
3343 compatibility with other assemblers. The original idea was that the
3344 assembly language source would be piped into the assembler. If the sender
3345 of the source quit, it could use this directive tells @code{@value{AS}} to
3346 quit also. One day @code{.abort} will not be supported.
3350 @section @code{.ABORT}
3352 @cindex @code{ABORT} directive
3353 When producing COFF output, @code{@value{AS}} accepts this directive as a
3354 synonym for @samp{.abort}.
3357 When producing @code{b.out} output, @code{@value{AS}} accepts this directive,
3363 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3365 @cindex padding the location counter
3366 @cindex @code{align} directive
3367 Pad the location counter (in the current subsection) to a particular storage
3368 boundary. The first expression (which must be absolute) is the alignment
3369 required, as described below.
3371 The second expression (also absolute) gives the fill value to be stored in the
3372 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3373 padding bytes are normally zero. However, on some systems, if the section is
3374 marked as containing code and the fill value is omitted, the space is filled
3375 with no-op instructions.
3377 The third expression is also absolute, and is also optional. If it is present,
3378 it is the maximum number of bytes that should be skipped by this alignment
3379 directive. If doing the alignment would require skipping more bytes than the
3380 specified maximum, then the alignment is not done at all. You can omit the
3381 fill value (the second argument) entirely by simply using two commas after the
3382 required alignment; this can be useful if you want the alignment to be filled
3383 with no-op instructions when appropriate.
3385 The way the required alignment is specified varies from system to system.
3386 For the a29k, hppa, m68k, m88k, w65, sparc, and Hitachi SH, and i386 using ELF
3388 the first expression is the
3389 alignment request in bytes. For example @samp{.align 8} advances
3390 the location counter until it is a multiple of 8. If the location counter
3391 is already a multiple of 8, no change is needed.
3393 For other systems, including the i386 using a.out format, and the arm and
3394 strongarm, it is the
3395 number of low-order zero bits the location counter must have after
3396 advancement. For example @samp{.align 3} advances the location
3397 counter until it a multiple of 8. If the location counter is already a
3398 multiple of 8, no change is needed.
3400 This inconsistency is due to the different behaviors of the various
3401 native assemblers for these systems which GAS must emulate.
3402 GAS also provides @code{.balign} and @code{.p2align} directives,
3403 described later, which have a consistent behavior across all
3404 architectures (but are specific to GAS).
3407 @section @code{.ascii "@var{string}"}@dots{}
3409 @cindex @code{ascii} directive
3410 @cindex string literals
3411 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3412 separated by commas. It assembles each string (with no automatic
3413 trailing zero byte) into consecutive addresses.
3416 @section @code{.asciz "@var{string}"}@dots{}
3418 @cindex @code{asciz} directive
3419 @cindex zero-terminated strings
3420 @cindex null-terminated strings
3421 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3422 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3425 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3427 @cindex padding the location counter given number of bytes
3428 @cindex @code{balign} directive
3429 Pad the location counter (in the current subsection) to a particular
3430 storage boundary. The first expression (which must be absolute) is the
3431 alignment request in bytes. For example @samp{.balign 8} advances
3432 the location counter until it is a multiple of 8. If the location counter
3433 is already a multiple of 8, no change is needed.
3435 The second expression (also absolute) gives the fill value to be stored in the
3436 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3437 padding bytes are normally zero. However, on some systems, if the section is
3438 marked as containing code and the fill value is omitted, the space is filled
3439 with no-op instructions.
3441 The third expression is also absolute, and is also optional. If it is present,
3442 it is the maximum number of bytes that should be skipped by this alignment
3443 directive. If doing the alignment would require skipping more bytes than the
3444 specified maximum, then the alignment is not done at all. You can omit the
3445 fill value (the second argument) entirely by simply using two commas after the
3446 required alignment; this can be useful if you want the alignment to be filled
3447 with no-op instructions when appropriate.
3449 @cindex @code{balignw} directive
3450 @cindex @code{balignl} directive
3451 The @code{.balignw} and @code{.balignl} directives are variants of the
3452 @code{.balign} directive. The @code{.balignw} directive treats the fill
3453 pattern as a two byte word value. The @code{.balignl} directives treats the
3454 fill pattern as a four byte longword value. For example, @code{.balignw
3455 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3456 filled in with the value 0x368d (the exact placement of the bytes depends upon
3457 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3461 @section @code{.byte @var{expressions}}
3463 @cindex @code{byte} directive
3464 @cindex integers, one byte
3465 @code{.byte} expects zero or more expressions, separated by commas.
3466 Each expression is assembled into the next byte.
3469 @section @code{.comm @var{symbol} , @var{length} }
3471 @cindex @code{comm} directive
3472 @cindex symbol, common
3473 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
3474 common symbol in one object file may be merged with a defined or common symbol
3475 of the same name in another object file. If @code{@value{LD}} does not see a
3476 definition for the symbol--just one or more common symbols--then it will
3477 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
3478 absolute expression. If @code{@value{LD}} sees multiple common symbols with
3479 the same name, and they do not all have the same size, it will allocate space
3480 using the largest size.
3483 When using ELF, the @code{.comm} directive takes an optional third argument.
3484 This is the desired alignment of the symbol, specified as a byte boundary (for
3485 example, an alignment of 16 means that the least significant 4 bits of the
3486 address should be zero). The alignment must be an absolute expression, and it
3487 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
3488 for the common symbol, it will use the alignment when placing the symbol. If
3489 no alignment is specified, @code{@value{AS}} will set the alignment to the
3490 largest power of two less than or equal to the size of the symbol, up to a
3495 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
3496 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
3500 @section @code{.data @var{subsection}}
3502 @cindex @code{data} directive
3503 @code{.data} tells @code{@value{AS}} to assemble the following statements onto the
3504 end of the data subsection numbered @var{subsection} (which is an
3505 absolute expression). If @var{subsection} is omitted, it defaults
3510 @section @code{.def @var{name}}
3512 @cindex @code{def} directive
3513 @cindex COFF symbols, debugging
3514 @cindex debugging COFF symbols
3515 Begin defining debugging information for a symbol @var{name}; the
3516 definition extends until the @code{.endef} directive is encountered.
3519 This directive is only observed when @code{@value{AS}} is configured for COFF
3520 format output; when producing @code{b.out}, @samp{.def} is recognized,
3527 @section @code{.desc @var{symbol}, @var{abs-expression}}
3529 @cindex @code{desc} directive
3530 @cindex COFF symbol descriptor
3531 @cindex symbol descriptor, COFF
3532 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
3533 to the low 16 bits of an absolute expression.
3536 The @samp{.desc} directive is not available when @code{@value{AS}} is
3537 configured for COFF output; it is only for @code{a.out} or @code{b.out}
3538 object format. For the sake of compatibility, @code{@value{AS}} accepts
3539 it, but produces no output, when configured for COFF.
3545 @section @code{.dim}
3547 @cindex @code{dim} directive
3548 @cindex COFF auxiliary symbol information
3549 @cindex auxiliary symbol information, COFF
3550 This directive is generated by compilers to include auxiliary debugging
3551 information in the symbol table. It is only permitted inside
3552 @code{.def}/@code{.endef} pairs.
3555 @samp{.dim} is only meaningful when generating COFF format output; when
3556 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
3562 @section @code{.double @var{flonums}}
3564 @cindex @code{double} directive
3565 @cindex floating point numbers (double)
3566 @code{.double} expects zero or more flonums, separated by commas. It
3567 assembles floating point numbers.
3569 The exact kind of floating point numbers emitted depends on how
3570 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
3574 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
3575 in @sc{ieee} format.
3580 @section @code{.eject}
3582 @cindex @code{eject} directive
3583 @cindex new page, in listings
3584 @cindex page, in listings
3585 @cindex listing control: new page
3586 Force a page break at this point, when generating assembly listings.
3589 @section @code{.else}
3591 @cindex @code{else} directive
3592 @code{.else} is part of the @code{@value{AS}} support for conditional
3593 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
3594 of code to be assembled if the condition for the preceding @code{.if}
3598 @section @code{.elseif}
3600 @cindex @code{elseif} directive
3601 @code{.elseif} is part of the @code{@value{AS}} support for conditional
3602 assembly; @pxref{If,,@code{.if}}. It is shorthand for beginning a new
3603 @code{.if} block that would otherwise fill the entire @code{.else} section.
3606 @section @code{.end}
3608 @cindex @code{end} directive
3609 @code{.end} marks the end of the assembly file. @code{@value{AS}} does not
3610 process anything in the file past the @code{.end} directive.
3614 @section @code{.endef}
3616 @cindex @code{endef} directive
3617 This directive flags the end of a symbol definition begun with
3621 @samp{.endef} is only meaningful when generating COFF format output; if
3622 @code{@value{AS}} is configured to generate @code{b.out}, it accepts this
3623 directive but ignores it.
3628 @section @code{.endfunc}
3629 @cindex @code{endfunc} directive
3630 @code{.endfunc} marks the end of a function specified with @code{.func}.
3633 @section @code{.endif}
3635 @cindex @code{endif} directive
3636 @code{.endif} is part of the @code{@value{AS}} support for conditional assembly;
3637 it marks the end of a block of code that is only assembled
3638 conditionally. @xref{If,,@code{.if}}.
3641 @section @code{.equ @var{symbol}, @var{expression}}
3643 @cindex @code{equ} directive
3644 @cindex assigning values to symbols
3645 @cindex symbols, assigning values to
3646 This directive sets the value of @var{symbol} to @var{expression}.
3647 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
3650 The syntax for @code{equ} on the HPPA is
3651 @samp{@var{symbol} .equ @var{expression}}.
3655 @section @code{.equiv @var{symbol}, @var{expression}}
3656 @cindex @code{equiv} directive
3657 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
3658 the assembler will signal an error if @var{symbol} is already defined.
3660 Except for the contents of the error message, this is roughly equivalent to
3669 @section @code{.err}
3670 @cindex @code{err} directive
3671 If @code{@value{AS}} assembles a @code{.err} directive, it will print an error
3672 message and, unless the @code{-Z} option was used, it will not generate an
3673 object file. This can be used to signal error an conditionally compiled code.
3676 @section @code{.exitm}
3677 Exit early from the current macro definition. @xref{Macro}.
3680 @section @code{.extern}
3682 @cindex @code{extern} directive
3683 @code{.extern} is accepted in the source program---for compatibility
3684 with other assemblers---but it is ignored. @code{@value{AS}} treats
3685 all undefined symbols as external.
3688 @section @code{.fail @var{expression}}
3690 @cindex @code{fail} directive
3691 Generates an error or a warning. If the value of the @var{expression} is 500
3692 or more, @code{@value{AS}} will print a warning message. If the value is less
3693 than 500, @code{@value{AS}} will print an error message. The message will
3694 include the value of @var{expression}. This can occasionally be useful inside
3695 complex nested macros or conditional assembly.
3697 @ifclear no-file-dir
3699 @section @code{.file @var{string}}
3701 @cindex @code{file} directive
3702 @cindex logical file name
3703 @cindex file name, logical
3704 @code{.file} tells @code{@value{AS}} that we are about to start a new logical
3705 file. @var{string} is the new file name. In general, the filename is
3706 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
3707 to specify an empty file name, you must give the quotes--@code{""}. This
3708 statement may go away in future: it is only recognized to be compatible with
3709 old @code{@value{AS}} programs.
3711 In some configurations of @code{@value{AS}}, @code{.file} has already been
3712 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
3717 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
3719 @cindex @code{fill} directive
3720 @cindex writing patterns in memory
3721 @cindex patterns, writing in memory
3722 @var{result}, @var{size} and @var{value} are absolute expressions.
3723 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
3724 may be zero or more. @var{Size} may be zero or more, but if it is
3725 more than 8, then it is deemed to have the value 8, compatible with
3726 other people's assemblers. The contents of each @var{repeat} bytes
3727 is taken from an 8-byte number. The highest order 4 bytes are
3728 zero. The lowest order 4 bytes are @var{value} rendered in the
3729 byte-order of an integer on the computer @code{@value{AS}} is assembling for.
3730 Each @var{size} bytes in a repetition is taken from the lowest order
3731 @var{size} bytes of this number. Again, this bizarre behavior is
3732 compatible with other people's assemblers.
3734 @var{size} and @var{value} are optional.
3735 If the second comma and @var{value} are absent, @var{value} is
3736 assumed zero. If the first comma and following tokens are absent,
3737 @var{size} is assumed to be 1.
3740 @section @code{.float @var{flonums}}
3742 @cindex floating point numbers (single)
3743 @cindex @code{float} directive
3744 This directive assembles zero or more flonums, separated by commas. It
3745 has the same effect as @code{.single}.
3747 The exact kind of floating point numbers emitted depends on how
3748 @code{@value{AS}} is configured.
3749 @xref{Machine Dependencies}.
3753 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
3754 in @sc{ieee} format.
3759 @section @code{.func @var{name}[,@var{label}]}
3760 @cindex @code{func} directive
3761 @code{.func} emits debugging information to denote function @var{name}, and
3762 is ignored unless the file is assembled with debugging enabled.
3763 Only @samp{--gstabs} is currently supported.
3764 @var{label} is the entry point of the function and if omitted @var{name}
3765 prepended with the @samp{leading char} is used.
3766 @samp{leading char} is usually @code{_} or nothing, depending on the target.
3767 All functions are currently defined to have @code{void} return type.
3768 The function must be terminated with @code{.endfunc}.
3771 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3773 @cindex @code{global} directive
3774 @cindex symbol, making visible to linker
3775 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
3776 @var{symbol} in your partial program, its value is made available to
3777 other partial programs that are linked with it. Otherwise,
3778 @var{symbol} takes its attributes from a symbol of the same name
3779 from another file linked into the same program.
3781 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
3782 compatibility with other assemblers.
3785 On the HPPA, @code{.global} is not always enough to make it accessible to other
3786 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
3787 @xref{HPPA Directives,, HPPA Assembler Directives}.
3792 @section @code{.hidden @var{names}}
3794 @cindex @code{.hidden} directive
3796 This one of the ELF visibility directives. The other two are
3797 @pxref{Internal,,@code{.internal}} and @pxref{Protected,,@code{.protected}}
3799 This directive overrides the named symbols default visibility (which is set by
3800 their binding: local, global or weak). The directive sets the visibility to
3801 @code{hidden} which means that the symbols are not visible to other components.
3802 Such symbols are always considered to be @code{protected} as well.
3806 @section @code{.hword @var{expressions}}
3808 @cindex @code{hword} directive
3809 @cindex integers, 16-bit
3810 @cindex numbers, 16-bit
3811 @cindex sixteen bit integers
3812 This expects zero or more @var{expressions}, and emits
3813 a 16 bit number for each.
3816 This directive is a synonym for @samp{.short}; depending on the target
3817 architecture, it may also be a synonym for @samp{.word}.
3821 This directive is a synonym for @samp{.short}.
3824 This directive is a synonym for both @samp{.short} and @samp{.word}.
3829 @section @code{.ident}
3831 @cindex @code{ident} directive
3832 This directive is used by some assemblers to place tags in object files.
3833 @code{@value{AS}} simply accepts the directive for source-file
3834 compatibility with such assemblers, but does not actually emit anything
3838 @section @code{.if @var{absolute expression}}
3840 @cindex conditional assembly
3841 @cindex @code{if} directive
3842 @code{.if} marks the beginning of a section of code which is only
3843 considered part of the source program being assembled if the argument
3844 (which must be an @var{absolute expression}) is non-zero. The end of
3845 the conditional section of code must be marked by @code{.endif}
3846 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
3847 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
3848 If you have several conditions to check, @code{.elseif} may be used to avoid
3849 nesting blocks if/else within each subsequent @code{.else} block.
3851 The following variants of @code{.if} are also supported:
3853 @cindex @code{ifdef} directive
3854 @item .ifdef @var{symbol}
3855 Assembles the following section of code if the specified @var{symbol}
3858 @cindex @code{ifc} directive
3859 @item .ifc @var{string1},@var{string2}
3860 Assembles the following section of code if the two strings are the same. The
3861 strings may be optionally quoted with single quotes. If they are not quoted,
3862 the first string stops at the first comma, and the second string stops at the
3863 end of the line. Strings which contain whitespace should be quoted. The
3864 string comparison is case sensitive.
3866 @cindex @code{ifeq} directive
3867 @item .ifeq @var{absolute expression}
3868 Assembles the following section of code if the argument is zero.
3870 @cindex @code{ifeqs} directive
3871 @item .ifeqs @var{string1},@var{string2}
3872 Another form of @code{.ifc}. The strings must be quoted using double quotes.
3874 @cindex @code{ifge} directive
3875 @item .ifge @var{absolute expression}
3876 Assembles the following section of code if the argument is greater than or
3879 @cindex @code{ifgt} directive
3880 @item .ifgt @var{absolute expression}
3881 Assembles the following section of code if the argument is greater than zero.
3883 @cindex @code{ifle} directive
3884 @item .ifle @var{absolute expression}
3885 Assembles the following section of code if the argument is less than or equal
3888 @cindex @code{iflt} directive
3889 @item .iflt @var{absolute expression}
3890 Assembles the following section of code if the argument is less than zero.
3892 @cindex @code{ifnc} directive
3893 @item .ifnc @var{string1},@var{string2}.
3894 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
3895 following section of code if the two strings are not the same.
3897 @cindex @code{ifndef} directive
3898 @cindex @code{ifnotdef} directive
3899 @item .ifndef @var{symbol}
3900 @itemx .ifnotdef @var{symbol}
3901 Assembles the following section of code if the specified @var{symbol}
3902 has not been defined. Both spelling variants are equivalent.
3904 @cindex @code{ifne} directive
3905 @item .ifne @var{absolute expression}
3906 Assembles the following section of code if the argument is not equal to zero
3907 (in other words, this is equivalent to @code{.if}).
3909 @cindex @code{ifnes} directive
3910 @item .ifnes @var{string1},@var{string2}
3911 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
3912 following section of code if the two strings are not the same.
3916 @section @code{.include "@var{file}"}
3918 @cindex @code{include} directive
3919 @cindex supporting files, including
3920 @cindex files, including
3921 This directive provides a way to include supporting files at specified
3922 points in your source program. The code from @var{file} is assembled as
3923 if it followed the point of the @code{.include}; when the end of the
3924 included file is reached, assembly of the original file continues. You
3925 can control the search paths used with the @samp{-I} command-line option
3926 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
3930 @section @code{.int @var{expressions}}
3932 @cindex @code{int} directive
3933 @cindex integers, 32-bit
3934 Expect zero or more @var{expressions}, of any section, separated by commas.
3935 For each expression, emit a number that, at run time, is the value of that
3936 expression. The byte order and bit size of the number depends on what kind
3937 of target the assembly is for.
3941 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
3942 integers. On the H8/300H and the Hitachi SH, however, @code{.int} emits
3949 @section @code{.internal @var{names}}
3951 @cindex @code{.internal} directive
3953 This one of the ELF visibility directives. The other two are
3954 @pxref{Hidden,,@code{.hidden}} and @pxref{Protected,,@code{.protected}}
3956 This directive overrides the named symbols default visibility (which is set by
3957 their binding: local, global or weak). The directive sets the visibility to
3958 @code{internal} which means that the symbols are considered to be @code{hidden}
3959 (ie not visible to other components), and that some extra, processor specific
3960 processing must also be performed upon the symbols as well.
3964 @section @code{.irp @var{symbol},@var{values}}@dots{}
3966 @cindex @code{irp} directive
3967 Evaluate a sequence of statements assigning different values to @var{symbol}.
3968 The sequence of statements starts at the @code{.irp} directive, and is
3969 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
3970 set to @var{value}, and the sequence of statements is assembled. If no
3971 @var{value} is listed, the sequence of statements is assembled once, with
3972 @var{symbol} set to the null string. To refer to @var{symbol} within the
3973 sequence of statements, use @var{\symbol}.
3975 For example, assembling
3983 is equivalent to assembling
3992 @section @code{.irpc @var{symbol},@var{values}}@dots{}
3994 @cindex @code{irpc} directive
3995 Evaluate a sequence of statements assigning different values to @var{symbol}.
3996 The sequence of statements starts at the @code{.irpc} directive, and is
3997 terminated by an @code{.endr} directive. For each character in @var{value},
3998 @var{symbol} is set to the character, and the sequence of statements is
3999 assembled. If no @var{value} is listed, the sequence of statements is
4000 assembled once, with @var{symbol} set to the null string. To refer to
4001 @var{symbol} within the sequence of statements, use @var{\symbol}.
4003 For example, assembling
4011 is equivalent to assembling
4020 @section @code{.lcomm @var{symbol} , @var{length}}
4022 @cindex @code{lcomm} directive
4023 @cindex local common symbols
4024 @cindex symbols, local common
4025 Reserve @var{length} (an absolute expression) bytes for a local common
4026 denoted by @var{symbol}. The section and value of @var{symbol} are
4027 those of the new local common. The addresses are allocated in the bss
4028 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4029 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4030 not visible to @code{@value{LD}}.
4033 Some targets permit a third argument to be used with @code{.lcomm}. This
4034 argument specifies the desired alignment of the symbol in the bss section.
4038 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4039 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4043 @section @code{.lflags}
4045 @cindex @code{lflags} directive (ignored)
4046 @code{@value{AS}} accepts this directive, for compatibility with other
4047 assemblers, but ignores it.
4049 @ifclear no-line-dir
4051 @section @code{.line @var{line-number}}
4053 @cindex @code{line} directive
4057 @section @code{.ln @var{line-number}}
4059 @cindex @code{ln} directive
4061 @cindex logical line number
4063 Change the logical line number. @var{line-number} must be an absolute
4064 expression. The next line has that logical line number. Therefore any other
4065 statements on the current line (after a statement separator character) are
4066 reported as on logical line number @var{line-number} @minus{} 1. One day
4067 @code{@value{AS}} will no longer support this directive: it is recognized only
4068 for compatibility with existing assembler programs.
4072 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
4073 not available; use the synonym @code{.ln} in that context.
4078 @ifclear no-line-dir
4079 Even though this is a directive associated with the @code{a.out} or
4080 @code{b.out} object-code formats, @code{@value{AS}} still recognizes it
4081 when producing COFF output, and treats @samp{.line} as though it
4082 were the COFF @samp{.ln} @emph{if} it is found outside a
4083 @code{.def}/@code{.endef} pair.
4085 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4086 used by compilers to generate auxiliary symbol information for
4091 @section @code{.linkonce [@var{type}]}
4093 @cindex @code{linkonce} directive
4094 @cindex common sections
4095 Mark the current section so that the linker only includes a single copy of it.
4096 This may be used to include the same section in several different object files,
4097 but ensure that the linker will only include it once in the final output file.
4098 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4099 Duplicate sections are detected based on the section name, so it should be
4102 This directive is only supported by a few object file formats; as of this
4103 writing, the only object file format which supports it is the Portable
4104 Executable format used on Windows NT.
4106 The @var{type} argument is optional. If specified, it must be one of the
4107 following strings. For example:
4111 Not all types may be supported on all object file formats.
4115 Silently discard duplicate sections. This is the default.
4118 Warn if there are duplicate sections, but still keep only one copy.
4121 Warn if any of the duplicates have different sizes.
4124 Warn if any of the duplicates do not have exactly the same contents.
4128 @section @code{.ln @var{line-number}}
4130 @cindex @code{ln} directive
4131 @ifclear no-line-dir
4132 @samp{.ln} is a synonym for @samp{.line}.
4135 Tell @code{@value{AS}} to change the logical line number. @var{line-number}
4136 must be an absolute expression. The next line has that logical
4137 line number, so any other statements on the current line (after a
4138 statement separator character @code{;}) are reported as on logical
4139 line number @var{line-number} @minus{} 1.
4142 This directive is accepted, but ignored, when @code{@value{AS}} is
4143 configured for @code{b.out}; its effect is only associated with COFF
4149 @section @code{.mri @var{val}}
4151 @cindex @code{mri} directive
4152 @cindex MRI mode, temporarily
4153 If @var{val} is non-zero, this tells @code{@value{AS}} to enter MRI mode. If
4154 @var{val} is zero, this tells @code{@value{AS}} to exit MRI mode. This change
4155 affects code assembled until the next @code{.mri} directive, or until the end
4156 of the file. @xref{M, MRI mode, MRI mode}.
4159 @section @code{.list}
4161 @cindex @code{list} directive
4162 @cindex listing control, turning on
4163 Control (in conjunction with the @code{.nolist} directive) whether or
4164 not assembly listings are generated. These two directives maintain an
4165 internal counter (which is zero initially). @code{.list} increments the
4166 counter, and @code{.nolist} decrements it. Assembly listings are
4167 generated whenever the counter is greater than zero.
4169 By default, listings are disabled. When you enable them (with the
4170 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4171 the initial value of the listing counter is one.
4174 @section @code{.long @var{expressions}}
4176 @cindex @code{long} directive
4177 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
4180 @c no one seems to know what this is for or whether this description is
4181 @c what it really ought to do
4183 @section @code{.lsym @var{symbol}, @var{expression}}
4185 @cindex @code{lsym} directive
4186 @cindex symbol, not referenced in assembly
4187 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
4188 the hash table, ensuring it cannot be referenced by name during the
4189 rest of the assembly. This sets the attributes of the symbol to be
4190 the same as the expression value:
4192 @var{other} = @var{descriptor} = 0
4193 @var{type} = @r{(section of @var{expression})}
4194 @var{value} = @var{expression}
4197 The new symbol is not flagged as external.
4201 @section @code{.macro}
4204 The commands @code{.macro} and @code{.endm} allow you to define macros that
4205 generate assembly output. For example, this definition specifies a macro
4206 @code{sum} that puts a sequence of numbers into memory:
4209 .macro sum from=0, to=5
4218 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
4230 @item .macro @var{macname}
4231 @itemx .macro @var{macname} @var{macargs} @dots{}
4232 @cindex @code{macro} directive
4233 Begin the definition of a macro called @var{macname}. If your macro
4234 definition requires arguments, specify their names after the macro name,
4235 separated by commas or spaces. You can supply a default value for any
4236 macro argument by following the name with @samp{=@var{deflt}}. For
4237 example, these are all valid @code{.macro} statements:
4241 Begin the definition of a macro called @code{comm}, which takes no
4244 @item .macro plus1 p, p1
4245 @itemx .macro plus1 p p1
4246 Either statement begins the definition of a macro called @code{plus1},
4247 which takes two arguments; within the macro definition, write
4248 @samp{\p} or @samp{\p1} to evaluate the arguments.
4250 @item .macro reserve_str p1=0 p2
4251 Begin the definition of a macro called @code{reserve_str}, with two
4252 arguments. The first argument has a default value, but not the second.
4253 After the definition is complete, you can call the macro either as
4254 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
4255 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
4256 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
4257 @samp{0}, and @samp{\p2} evaluating to @var{b}).
4260 When you call a macro, you can specify the argument values either by
4261 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
4262 @samp{sum to=17, from=9}.
4265 @cindex @code{endm} directive
4266 Mark the end of a macro definition.
4269 @cindex @code{exitm} directive
4270 Exit early from the current macro definition.
4272 @cindex number of macros executed
4273 @cindex macros, count executed
4275 @code{@value{AS}} maintains a counter of how many macros it has
4276 executed in this pseudo-variable; you can copy that number to your
4277 output with @samp{\@@}, but @emph{only within a macro definition}.
4280 @item LOCAL @var{name} [ , @dots{} ]
4281 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
4282 macro syntax'' with @samp{-a} or @samp{--alternate}.} @xref{Alternate,,
4283 Alternate macro syntax}.
4285 Generate a string replacement for each of the @var{name} arguments, and
4286 replace any instances of @var{name} in each macro expansion. The
4287 replacement string is unique in the assembly, and different for each
4288 separate macro expansion. @code{LOCAL} allows you to write macros that
4289 define symbols, without fear of conflict between separate macro expansions.
4294 @section @code{.nolist}
4296 @cindex @code{nolist} directive
4297 @cindex listing control, turning off
4298 Control (in conjunction with the @code{.list} directive) whether or
4299 not assembly listings are generated. These two directives maintain an
4300 internal counter (which is zero initially). @code{.list} increments the
4301 counter, and @code{.nolist} decrements it. Assembly listings are
4302 generated whenever the counter is greater than zero.
4305 @section @code{.octa @var{bignums}}
4307 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
4308 @cindex @code{octa} directive
4309 @cindex integer, 16-byte
4310 @cindex sixteen byte integer
4311 This directive expects zero or more bignums, separated by commas. For each
4312 bignum, it emits a 16-byte integer.
4314 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
4315 hence @emph{octa}-word for 16 bytes.
4318 @section @code{.org @var{new-lc} , @var{fill}}
4320 @cindex @code{org} directive
4321 @cindex location counter, advancing
4322 @cindex advancing location counter
4323 @cindex current address, advancing
4324 Advance the location counter of the current section to
4325 @var{new-lc}. @var{new-lc} is either an absolute expression or an
4326 expression with the same section as the current subsection. That is,
4327 you can't use @code{.org} to cross sections: if @var{new-lc} has the
4328 wrong section, the @code{.org} directive is ignored. To be compatible
4329 with former assemblers, if the section of @var{new-lc} is absolute,
4330 @code{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
4331 is the same as the current subsection.
4333 @code{.org} may only increase the location counter, or leave it
4334 unchanged; you cannot use @code{.org} to move the location counter
4337 @c double negative used below "not undefined" because this is a specific
4338 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
4339 @c section. doc@cygnus.com 18feb91
4340 Because @code{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
4341 may not be undefined. If you really detest this restriction we eagerly await
4342 a chance to share your improved assembler.
4344 Beware that the origin is relative to the start of the section, not
4345 to the start of the subsection. This is compatible with other
4346 people's assemblers.
4348 When the location counter (of the current subsection) is advanced, the
4349 intervening bytes are filled with @var{fill} which should be an
4350 absolute expression. If the comma and @var{fill} are omitted,
4351 @var{fill} defaults to zero.
4354 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4356 @cindex padding the location counter given a power of two
4357 @cindex @code{p2align} directive
4358 Pad the location counter (in the current subsection) to a particular
4359 storage boundary. The first expression (which must be absolute) is the
4360 number of low-order zero bits the location counter must have after
4361 advancement. For example @samp{.p2align 3} advances the location
4362 counter until it a multiple of 8. If the location counter is already a
4363 multiple of 8, no change is needed.
4365 The second expression (also absolute) gives the fill value to be stored in the
4366 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4367 padding bytes are normally zero. However, on some systems, if the section is
4368 marked as containing code and the fill value is omitted, the space is filled
4369 with no-op instructions.
4371 The third expression is also absolute, and is also optional. If it is present,
4372 it is the maximum number of bytes that should be skipped by this alignment
4373 directive. If doing the alignment would require skipping more bytes than the
4374 specified maximum, then the alignment is not done at all. You can omit the
4375 fill value (the second argument) entirely by simply using two commas after the
4376 required alignment; this can be useful if you want the alignment to be filled
4377 with no-op instructions when appropriate.
4379 @cindex @code{p2alignw} directive
4380 @cindex @code{p2alignl} directive
4381 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
4382 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
4383 pattern as a two byte word value. The @code{.p2alignl} directives treats the
4384 fill pattern as a four byte longword value. For example, @code{.p2alignw
4385 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4386 filled in with the value 0x368d (the exact placement of the bytes depends upon
4387 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4392 @section @code{.previous}
4394 @cindex @code{.previous} directive
4395 @cindex Section Stack
4396 This is one of the ELF section stack manipulation directives. The others are
4397 @pxref{Section}, @xref{SubSection}, @pxref{PushSection}, and
4400 This directive swaps the current section (and subsection) with most recently
4401 referenced section (and subsection) prior to this one. Multiple
4402 @code{.previous} directives in a row will flip between two sections (and their
4405 In terms of the section stack, this directive swaps the current section with
4406 the top section on the section stack.
4411 @section @code{.popsection}
4413 @cindex @code{.popsection} directive
4414 @cindex Section Stack
4415 This is one of the ELF section stack manipulation directives. The others are
4416 @pxref{Section}, @xref{SubSection}, @pxref{PushSection}, and
4419 This directive replaces the current section (and subsection) with the top
4420 section (and subsection) on the section stack. This section is popped off the
4425 @section @code{.print @var{string}}
4427 @cindex @code{print} directive
4428 @code{@value{AS}} will print @var{string} on the standard output during
4429 assembly. You must put @var{string} in double quotes.
4433 @section @code{.protected @var{names}}
4435 @cindex @code{.protected} directive
4437 This one of the ELF visibility directives. The other two are
4438 @pxref{Hidden} and @pxref{Internal}
4440 This directive overrides the named symbols default visibility (which is set by
4441 their binding: local, global or weak). The directive sets the visibility to
4442 @code{protected} which means that any references to the symbols from within the
4443 components that defines them must be resolved to the definition in that
4444 component, even if a definition in another component would normally preempt
4449 @section @code{.psize @var{lines} , @var{columns}}
4451 @cindex @code{psize} directive
4452 @cindex listing control: paper size
4453 @cindex paper size, for listings
4454 Use this directive to declare the number of lines---and, optionally, the
4455 number of columns---to use for each page, when generating listings.
4457 If you do not use @code{.psize}, listings use a default line-count
4458 of 60. You may omit the comma and @var{columns} specification; the
4459 default width is 200 columns.
4461 @code{@value{AS}} generates formfeeds whenever the specified number of
4462 lines is exceeded (or whenever you explicitly request one, using
4465 If you specify @var{lines} as @code{0}, no formfeeds are generated save
4466 those explicitly specified with @code{.eject}.
4469 @section @code{.purgem @var{name}}
4471 @cindex @code{purgem} directive
4472 Undefine the macro @var{name}, so that later uses of the string will not be
4473 expanded. @xref{Macro}.
4477 @section @code{.pushsection @var{name} , @var{subsection}}
4479 @cindex @code{.pushsection} directive
4480 @cindex Section Stack
4481 This is one of the ELF section stack manipulation directives. The others are
4482 @pxref{Section}, @xref{SubSection}, @pxref{PopSection}, and
4485 This directive is a synonym for @code{.section}. It psuhes the current section
4486 (and subsection) onto the top of the section stack, and then replaces the
4487 current section and subsection with @code{name} and @code{subsection}.
4491 @section @code{.quad @var{bignums}}
4493 @cindex @code{quad} directive
4494 @code{.quad} expects zero or more bignums, separated by commas. For
4495 each bignum, it emits
4497 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
4498 warning message; and just takes the lowest order 8 bytes of the bignum.
4499 @cindex eight-byte integer
4500 @cindex integer, 8-byte
4502 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
4503 hence @emph{quad}-word for 8 bytes.
4506 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
4507 warning message; and just takes the lowest order 16 bytes of the bignum.
4508 @cindex sixteen-byte integer
4509 @cindex integer, 16-byte
4513 @section @code{.rept @var{count}}
4515 @cindex @code{rept} directive
4516 Repeat the sequence of lines between the @code{.rept} directive and the next
4517 @code{.endr} directive @var{count} times.
4519 For example, assembling
4527 is equivalent to assembling
4536 @section @code{.sbttl "@var{subheading}"}
4538 @cindex @code{sbttl} directive
4539 @cindex subtitles for listings
4540 @cindex listing control: subtitle
4541 Use @var{subheading} as the title (third line, immediately after the
4542 title line) when generating assembly listings.
4544 This directive affects subsequent pages, as well as the current page if
4545 it appears within ten lines of the top of a page.
4549 @section @code{.scl @var{class}}
4551 @cindex @code{scl} directive
4552 @cindex symbol storage class (COFF)
4553 @cindex COFF symbol storage class
4554 Set the storage-class value for a symbol. This directive may only be
4555 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
4556 whether a symbol is static or external, or it may record further
4557 symbolic debugging information.
4560 The @samp{.scl} directive is primarily associated with COFF output; when
4561 configured to generate @code{b.out} output format, @code{@value{AS}}
4562 accepts this directive but ignores it.
4567 @section @code{.section @var{name}} (COFF version)
4569 @cindex @code{section} directive
4570 @cindex named section
4571 Use the @code{.section} directive to assemble the following code into a section
4574 This directive is only supported for targets that actually support arbitrarily
4575 named sections; on @code{a.out} targets, for example, it is not accepted, even
4576 with a standard @code{a.out} section name.
4578 For COFF targets, the @code{.section} directive is used in one of the following
4582 .section @var{name}[, "@var{flags}"]
4583 .section @var{name}[, @var{subsegment}]
4586 If the optional argument is quoted, it is taken as flags to use for the
4587 section. Each flag is a single character. The following flags are recognized:
4590 bss section (uninitialized data)
4592 section is not loaded
4602 shared section (meaningful for PE targets)
4605 If no flags are specified, the default flags depend upon the section name. If
4606 the section name is not recognized, the default will be for the section to be
4607 loaded and writable.
4609 If the optional argument to the @code{.section} directive is not quoted, it is
4610 taken as a subsegment number (@pxref{Sub-Sections}).
4613 @section @code{.section @var{name}} (ELF version)
4615 @cindex @code{section} directive
4616 @cindex named section
4618 @cindex Section Stack
4619 This is one of the ELF section stack manipulation directives. The others are
4620 @xref{SubSection}, @pxref{PushSection}@pxref{PopSection}, and
4624 For ELF targets, the @code{.section} directive is used like this:
4627 .section @var{name} [, "@var{flags}"[, @@@var{type}]]
4630 The optional @var{flags} argument is a quoted string which may contain any
4631 combintion of the following characters:
4634 section is allocatable
4638 section is executable
4641 The optional @var{type} argument may contain one of the following constants:
4644 section contains data
4646 section does not contain data (i.e., section only occupies space)
4649 If no flags are specified, the default flags depend upon the section name. If
4650 the section name is not recognized, the default will be for the section to have
4651 none of the above flags: it will not be allocated in memory, nor writable, nor
4652 executable. The section will contain data.
4654 For ELF targets, the assembler supports another type of @code{.section}
4655 directive for compatibility with the Solaris assembler:
4658 .section "@var{name}"[, @var{flags}...]
4661 Note that the section name is quoted. There may be a sequence of comma
4665 section is allocatable
4669 section is executable
4672 This directive replaces the current section and subsection. The replaced
4673 section and subsection are pushed onto the section stack. See the contents of
4674 the gas testsuite directory @code{gas/testsuite/gas/elf} for some examples of
4675 how this directive and the other section stack directives work.
4678 @section @code{.set @var{symbol}, @var{expression}}
4680 @cindex @code{set} directive
4681 @cindex symbol value, setting
4682 Set the value of @var{symbol} to @var{expression}. This
4683 changes @var{symbol}'s value and type to conform to
4684 @var{expression}. If @var{symbol} was flagged as external, it remains
4685 flagged (@pxref{Symbol Attributes}).
4687 You may @code{.set} a symbol many times in the same assembly.
4689 If you @code{.set} a global symbol, the value stored in the object
4690 file is the last value stored into it.
4693 The syntax for @code{set} on the HPPA is
4694 @samp{@var{symbol} .set @var{expression}}.
4698 @section @code{.short @var{expressions}}
4700 @cindex @code{short} directive
4702 @code{.short} is normally the same as @samp{.word}.
4703 @xref{Word,,@code{.word}}.
4705 In some configurations, however, @code{.short} and @code{.word} generate
4706 numbers of different lengths; @pxref{Machine Dependencies}.
4710 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
4713 This expects zero or more @var{expressions}, and emits
4714 a 16 bit number for each.
4719 @section @code{.single @var{flonums}}
4721 @cindex @code{single} directive
4722 @cindex floating point numbers (single)
4723 This directive assembles zero or more flonums, separated by commas. It
4724 has the same effect as @code{.float}.
4726 The exact kind of floating point numbers emitted depends on how
4727 @code{@value{AS}} is configured. @xref{Machine Dependencies}.
4731 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
4732 numbers in @sc{ieee} format.
4737 @section @code{.size} (COFF version)
4739 @cindex @code{size} directive
4740 This directive is generated by compilers to include auxiliary debugging
4741 information in the symbol table. It is only permitted inside
4742 @code{.def}/@code{.endef} pairs.
4745 @samp{.size} is only meaningful when generating COFF format output; when
4746 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4750 @section @code{.size @var{name} , @var{expression}} (ELF version)
4751 @cindex @code{size} directive
4753 This directive is used to set the size associated with a symbol @var{name}.
4754 The size in bytes is computed from @var{expression} which can make use of label
4755 arithmetic. This directive is typically used to set the size of function
4759 @section @code{.sleb128 @var{expressions}}
4761 @cindex @code{sleb128} directive
4762 @var{sleb128} stands for ``signed little endian base 128.'' This is a
4763 compact, variable length representation of numbers used by the DWARF
4764 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
4766 @ifclear no-space-dir
4768 @section @code{.skip @var{size} , @var{fill}}
4770 @cindex @code{skip} directive
4771 @cindex filling memory
4772 This directive emits @var{size} bytes, each of value @var{fill}. Both
4773 @var{size} and @var{fill} are absolute expressions. If the comma and
4774 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
4778 @section @code{.space @var{size} , @var{fill}}
4780 @cindex @code{space} directive
4781 @cindex filling memory
4782 This directive emits @var{size} bytes, each of value @var{fill}. Both
4783 @var{size} and @var{fill} are absolute expressions. If the comma
4784 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
4789 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
4790 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
4791 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
4792 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
4801 @section @code{.space}
4802 @cindex @code{space} directive
4804 On the AMD 29K, this directive is ignored; it is accepted for
4805 compatibility with other AMD 29K assemblers.
4808 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
4809 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
4815 @section @code{.stabd, .stabn, .stabs}
4817 @cindex symbolic debuggers, information for
4818 @cindex @code{stab@var{x}} directives
4819 There are three directives that begin @samp{.stab}.
4820 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
4821 The symbols are not entered in the @code{@value{AS}} hash table: they
4822 cannot be referenced elsewhere in the source file.
4823 Up to five fields are required:
4827 This is the symbol's name. It may contain any character except
4828 @samp{\000}, so is more general than ordinary symbol names. Some
4829 debuggers used to code arbitrarily complex structures into symbol names
4833 An absolute expression. The symbol's type is set to the low 8 bits of
4834 this expression. Any bit pattern is permitted, but @code{@value{LD}}
4835 and debuggers choke on silly bit patterns.
4838 An absolute expression. The symbol's ``other'' attribute is set to the
4839 low 8 bits of this expression.
4842 An absolute expression. The symbol's descriptor is set to the low 16
4843 bits of this expression.
4846 An absolute expression which becomes the symbol's value.
4849 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
4850 or @code{.stabs} statement, the symbol has probably already been created;
4851 you get a half-formed symbol in your object file. This is
4852 compatible with earlier assemblers!
4855 @cindex @code{stabd} directive
4856 @item .stabd @var{type} , @var{other} , @var{desc}
4858 The ``name'' of the symbol generated is not even an empty string.
4859 It is a null pointer, for compatibility. Older assemblers used a
4860 null pointer so they didn't waste space in object files with empty
4863 The symbol's value is set to the location counter,
4864 relocatably. When your program is linked, the value of this symbol
4865 is the address of the location counter when the @code{.stabd} was
4868 @cindex @code{stabn} directive
4869 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
4870 The name of the symbol is set to the empty string @code{""}.
4872 @cindex @code{stabs} directive
4873 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
4874 All five fields are specified.
4880 @section @code{.string} "@var{str}"
4882 @cindex string, copying to object file
4883 @cindex @code{string} directive
4885 Copy the characters in @var{str} to the object file. You may specify more than
4886 one string to copy, separated by commas. Unless otherwise specified for a
4887 particular machine, the assembler marks the end of each string with a 0 byte.
4888 You can use any of the escape sequences described in @ref{Strings,,Strings}.
4891 @section @code{.struct @var{expression}}
4893 @cindex @code{struct} directive
4894 Switch to the absolute section, and set the section offset to @var{expression},
4895 which must be an absolute expression. You might use this as follows:
4904 This would define the symbol @code{field1} to have the value 0, the symbol
4905 @code{field2} to have the value 4, and the symbol @code{field3} to have the
4906 value 8. Assembly would be left in the absolute section, and you would need to
4907 use a @code{.section} directive of some sort to change to some other section
4908 before further assembly.
4912 @section @code{.subsection @var{name}}
4914 @cindex @code{.subsection} directive
4915 @cindex Section Stack
4916 This is one of the ELF section stack manipulation directives. The others are
4917 @pxref{Section}, @xref{PushSection}, @pxref{PopSection}, and
4920 This directive replaces the current subsection with @code{name}. The current
4921 section is not changed. The replaced subsection is put onto the section stack
4922 in place of the then current top of stack subsection.
4927 @section @code{.symver}
4928 @cindex @code{symver} directive
4929 @cindex symbol versioning
4930 @cindex versions of symbols
4931 Use the @code{.symver} directive to bind symbols to specific version nodes
4932 within a source file. This is only supported on ELF platforms, and is
4933 typically used when assembling files to be linked into a shared library.
4934 There are cases where it may make sense to use this in objects to be bound
4935 into an application itself so as to override a versioned symbol from a
4938 For ELF targets, the @code{.symver} directive is used like this:
4940 .symver @var{name}, @var{name2@@nodename}
4942 In this case, the symbol @var{name} must exist and be defined within the file
4943 being assembled. The @code{.versym} directive effectively creates a symbol
4944 alias with the name @var{name2@@nodename}, and in fact the main reason that we
4945 just don't try and create a regular alias is that the @var{@@} character isn't
4946 permitted in symbol names. The @var{name2} part of the name is the actual name
4947 of the symbol by which it will be externally referenced. The name @var{name}
4948 itself is merely a name of convenience that is used so that it is possible to
4949 have definitions for multiple versions of a function within a single source
4950 file, and so that the compiler can unambiguously know which version of a
4951 function is being mentioned. The @var{nodename} portion of the alias should be
4952 the name of a node specified in the version script supplied to the linker when
4953 building a shared library. If you are attempting to override a versioned
4954 symbol from a shared library, then @var{nodename} should correspond to the
4955 nodename of the symbol you are trying to override.
4960 @section @code{.tag @var{structname}}
4962 @cindex COFF structure debugging
4963 @cindex structure debugging, COFF
4964 @cindex @code{tag} directive
4965 This directive is generated by compilers to include auxiliary debugging
4966 information in the symbol table. It is only permitted inside
4967 @code{.def}/@code{.endef} pairs. Tags are used to link structure
4968 definitions in the symbol table with instances of those structures.
4971 @samp{.tag} is only used when generating COFF format output; when
4972 @code{@value{AS}} is generating @code{b.out}, it accepts this directive but
4978 @section @code{.text @var{subsection}}
4980 @cindex @code{text} directive
4981 Tells @code{@value{AS}} to assemble the following statements onto the end of
4982 the text subsection numbered @var{subsection}, which is an absolute
4983 expression. If @var{subsection} is omitted, subsection number zero
4987 @section @code{.title "@var{heading}"}
4989 @cindex @code{title} directive
4990 @cindex listing control: title line
4991 Use @var{heading} as the title (second line, immediately after the
4992 source file name and pagenumber) when generating assembly listings.
4994 This directive affects subsequent pages, as well as the current page if
4995 it appears within ten lines of the top of a page.
4998 @section @code{.type @var{int}} (COFF version)
5000 @cindex COFF symbol type
5001 @cindex symbol type, COFF
5002 @cindex @code{type} directive
5003 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5004 records the integer @var{int} as the type attribute of a symbol table entry.
5007 @samp{.type} is associated only with COFF format output; when
5008 @code{@value{AS}} is configured for @code{b.out} output, it accepts this
5009 directive but ignores it.
5012 @section @code{.type @var{name} , @var{type description}} (ELF version)
5014 @cindex ELF symbol type
5015 @cindex symbol type, ELF
5016 @cindex @code{type} directive
5017 This directive is used to set the type of symbol @var{name} to be either a
5018 function symbol or an ojbect symbol. There are five different syntaxes
5019 supported for the @var{type description} field, in order to provide
5020 comptability with various other assemblers. The syntaxes supported are:
5023 .type <name>,#function
5024 .type <name>,#object
5026 .type <name>,@@function
5027 .type <name>,@@object
5029 .type <name>,%function
5030 .type <name>,%object
5032 .type <name>,"function"
5033 .type <name>,"object"
5035 .type <name> STT_FUNCTION
5036 .type <name> STT_OBJECT
5040 @section @code{.uleb128 @var{expressions}}
5042 @cindex @code{uleb128} directive
5043 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
5044 compact, variable length representation of numbers used by the DWARF
5045 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
5049 @section @code{.val @var{addr}}
5051 @cindex @code{val} directive
5052 @cindex COFF value attribute
5053 @cindex value attribute, COFF
5054 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5055 records the address @var{addr} as the value attribute of a symbol table
5059 @samp{.val} is used only for COFF output; when @code{@value{AS}} is
5060 configured for @code{b.out}, it accepts this directive but ignores it.
5066 @section @code{.version "@var{string}"}
5068 @cindex @code{.version}
5069 This directive creates a @code{.note} section and places into it an ELF
5070 formatted note of type NT_VERSION. The note's name is set to @code{string}.
5075 @section @code{.vtable_entry @var{table}, @var{offset}}
5077 @cindex @code{.vtable_entry}
5078 This directive finds or creates a symbol @code{table} and creates a
5079 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
5082 @section @code{.vtable_inherit @var{child}, @var{parent}}
5084 @cindex @code{.vtable_inherit}
5085 This directive finds the symbol @code{child} and finds or creates the symbol
5086 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
5087 parent whoes addend is the value of the child symbol. As a special case the
5088 parent name of @code{0} is treated as refering the @code{*ABS*} section.
5093 @section @code{.weak @var{names}}
5095 @cindex @code{.weak}
5096 This directive sets the weak attribute on the comma seperated list of symbol
5097 @code{names}. If the symbols do not already exist, they will be created.
5101 @section @code{.word @var{expressions}}
5103 @cindex @code{word} directive
5104 This directive expects zero or more @var{expressions}, of any section,
5105 separated by commas.
5108 For each expression, @code{@value{AS}} emits a 32-bit number.
5111 For each expression, @code{@value{AS}} emits a 16-bit number.
5116 The size of the number emitted, and its byte order,
5117 depend on what target computer the assembly is for.
5120 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
5121 @c happen---32-bit addressability, period; no long/short jumps.
5122 @ifset DIFF-TBL-KLUGE
5123 @cindex difference tables altered
5124 @cindex altered difference tables
5126 @emph{Warning: Special Treatment to support Compilers}
5130 Machines with a 32-bit address space, but that do less than 32-bit
5131 addressing, require the following special treatment. If the machine of
5132 interest to you does 32-bit addressing (or doesn't require it;
5133 @pxref{Machine Dependencies}), you can ignore this issue.
5136 In order to assemble compiler output into something that works,
5137 @code{@value{AS}} occasionlly does strange things to @samp{.word} directives.
5138 Directives of the form @samp{.word sym1-sym2} are often emitted by
5139 compilers as part of jump tables. Therefore, when @code{@value{AS}} assembles a
5140 directive of the form @samp{.word sym1-sym2}, and the difference between
5141 @code{sym1} and @code{sym2} does not fit in 16 bits, @code{@value{AS}}
5142 creates a @dfn{secondary jump table}, immediately before the next label.
5143 This secondary jump table is preceded by a short-jump to the
5144 first byte after the secondary table. This short-jump prevents the flow
5145 of control from accidentally falling into the new table. Inside the
5146 table is a long-jump to @code{sym2}. The original @samp{.word}
5147 contains @code{sym1} minus the address of the long-jump to
5150 If there were several occurrences of @samp{.word sym1-sym2} before the
5151 secondary jump table, all of them are adjusted. If there was a
5152 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
5153 long-jump to @code{sym4} is included in the secondary jump table,
5154 and the @code{.word} directives are adjusted to contain @code{sym3}
5155 minus the address of the long-jump to @code{sym4}; and so on, for as many
5156 entries in the original jump table as necessary.
5159 @emph{This feature may be disabled by compiling @code{@value{AS}} with the
5160 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
5161 assembly language programmers.
5164 @c end DIFF-TBL-KLUGE
5167 @section Deprecated Directives
5169 @cindex deprecated directives
5170 @cindex obsolescent directives
5171 One day these directives won't work.
5172 They are included for compatibility with older assemblers.
5179 @node Machine Dependencies
5180 @chapter Machine Dependent Features
5182 @cindex machine dependencies
5183 The machine instruction sets are (almost by definition) different on
5184 each machine where @code{@value{AS}} runs. Floating point representations
5185 vary as well, and @code{@value{AS}} often supports a few additional
5186 directives or command-line options for compatibility with other
5187 assemblers on a particular platform. Finally, some versions of
5188 @code{@value{AS}} support special pseudo-instructions for branch
5191 This chapter discusses most of these differences, though it does not
5192 include details on any machine's instruction set. For details on that
5193 subject, see the hardware manufacturer's manual.
5197 * AMD29K-Dependent:: AMD 29K Dependent Features
5200 * ARC-Dependent:: ARC Dependent Features
5203 * ARM-Dependent:: ARM Dependent Features
5206 * D10V-Dependent:: D10V Dependent Features
5209 * D30V-Dependent:: D30V Dependent Features
5212 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
5215 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
5218 * HPPA-Dependent:: HPPA Dependent Features
5221 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
5224 * i386-Dependent:: Intel 80386 Dependent Features
5227 * i860-Dependent:: Intel 80860 Dependent Features
5230 * i960-Dependent:: Intel 80960 Dependent Features
5233 * M32R-Dependent:: M32R Dependent Features
5236 * M68K-Dependent:: M680x0 Dependent Features
5239 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
5242 * MIPS-Dependent:: MIPS Dependent Features
5245 * SH-Dependent:: Hitachi SH Dependent Features
5248 * PJ-Dependent:: picoJava Dependent Features
5251 * Sparc-Dependent:: SPARC Dependent Features
5254 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
5257 * V850-Dependent:: V850 Dependent Features
5260 * Z8000-Dependent:: Z8000 Dependent Features
5263 * Vax-Dependent:: VAX Dependent Features
5270 @c The following major nodes are *sections* in the GENERIC version, *chapters*
5271 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
5272 @c peculiarity: to preserve cross-references, there must be a node called
5273 @c "Machine Dependencies". Hence the conditional nodenames in each
5274 @c major node below. Node defaulting in makeinfo requires adjacency of
5275 @c node and sectioning commands; hence the repetition of @chapter BLAH
5276 @c in both conditional blocks.
5282 @chapter ARC Dependent Features
5285 @node Machine Dependencies
5286 @chapter ARC Dependent Features
5291 * ARC-Opts:: Options
5292 * ARC-Float:: Floating Point
5293 * ARC-Directives:: Sparc Machine Directives
5299 @cindex options for ARC
5301 @cindex architectures, ARC
5302 @cindex ARC architectures
5303 The ARC chip family includes several successive levels (or other
5304 variants) of chip, using the same core instruction set, but including
5305 a few additional instructions at each level.
5307 By default, @code{@value{AS}} assumes the core instruction set (ARC
5308 base). The @code{.cpu} pseudo-op is intended to be used to select
5312 @cindex @code{-mbig-endian} option (ARC)
5313 @cindex @code{-mlittle-endian} option (ARC)
5314 @cindex ARC big-endian output
5315 @cindex ARC little-endian output
5316 @cindex big-endian output, ARC
5317 @cindex little-endian output, ARC
5319 @itemx -mlittle-endian
5320 Any @sc{arc} configuration of @code{@value{AS}} can select big-endian or
5321 little-endian output at run time (unlike most other @sc{gnu} development
5322 tools, which must be configured for one or the other). Use
5323 @samp{-mbig-endian} to select big-endian output, and @samp{-mlittle-endian}
5328 @section Floating Point
5330 @cindex floating point, ARC (@sc{ieee})
5331 @cindex ARC floating point (@sc{ieee})
5332 The ARC cpu family currently does not have hardware floating point
5333 support. Software floating point support is provided by @code{GCC}
5334 and uses @sc{ieee} floating-point numbers.
5336 @node ARC-Directives
5337 @section ARC Machine Directives
5339 @cindex ARC machine directives
5340 @cindex machine directives, ARC
5341 The ARC version of @code{@value{AS}} supports the following additional
5346 @cindex @code{cpu} directive, SPARC
5347 This must be followed by the desired cpu.
5348 The ARC is intended to be customizable, @code{.cpu} is used to
5349 select the desired variant [though currently there are none].
5356 @include c-a29k.texi
5365 @node Machine Dependencies
5366 @chapter Machine Dependent Features
5368 The machine instruction sets are different on each Hitachi chip family,
5369 and there are also some syntax differences among the families. This
5370 chapter describes the specific @code{@value{AS}} features for each
5374 * H8/300-Dependent:: Hitachi H8/300 Dependent Features
5375 * H8/500-Dependent:: Hitachi H8/500 Dependent Features
5376 * SH-Dependent:: Hitachi SH Dependent Features
5383 @include c-d10v.texi
5387 @include c-d30v.texi
5391 @include c-h8300.texi
5395 @include c-h8500.texi
5399 @include c-hppa.texi
5403 @include c-i370.texi
5407 @include c-i386.texi
5411 @include c-i860.texi
5415 @include c-i960.texi
5419 @include c-m32r.texi
5423 @include c-m68k.texi
5427 @include c-m68hc11.texi
5431 @include c-mips.texi
5435 @include c-ns32k.texi
5447 @include c-sparc.texi
5451 @include c-tic54x.texi
5463 @include c-v850.texi
5467 @c reverse effect of @down at top of generic Machine-Dep chapter
5471 @node Reporting Bugs
5472 @chapter Reporting Bugs
5473 @cindex bugs in assembler
5474 @cindex reporting bugs in assembler
5476 Your bug reports play an essential role in making @code{@value{AS}} reliable.
5478 Reporting a bug may help you by bringing a solution to your problem, or it may
5479 not. But in any case the principal function of a bug report is to help the
5480 entire community by making the next version of @code{@value{AS}} work better.
5481 Bug reports are your contribution to the maintenance of @code{@value{AS}}.
5483 In order for a bug report to serve its purpose, you must include the
5484 information that enables us to fix the bug.
5487 * Bug Criteria:: Have you found a bug?
5488 * Bug Reporting:: How to report bugs
5492 @section Have you found a bug?
5493 @cindex bug criteria
5495 If you are not sure whether you have found a bug, here are some guidelines:
5498 @cindex fatal signal
5499 @cindex assembler crash
5500 @cindex crash of assembler
5502 If the assembler gets a fatal signal, for any input whatever, that is a
5503 @code{@value{AS}} bug. Reliable assemblers never crash.
5505 @cindex error on valid input
5507 If @code{@value{AS}} produces an error message for valid input, that is a bug.
5509 @cindex invalid input
5511 If @code{@value{AS}} does not produce an error message for invalid input, that
5512 is a bug. However, you should note that your idea of ``invalid input'' might
5513 be our idea of ``an extension'' or ``support for traditional practice''.
5516 If you are an experienced user of assemblers, your suggestions for improvement
5517 of @code{@value{AS}} are welcome in any case.
5521 @section How to report bugs
5523 @cindex assembler bugs, reporting
5525 A number of companies and individuals offer support for @sc{gnu} products. If
5526 you obtained @code{@value{AS}} from a support organization, we recommend you
5527 contact that organization first.
5529 You can find contact information for many support companies and
5530 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
5533 In any event, we also recommend that you send bug reports for @code{@value{AS}}
5534 to @samp{bug-gnu-utils@@gnu.org}.
5536 The fundamental principle of reporting bugs usefully is this:
5537 @strong{report all the facts}. If you are not sure whether to state a
5538 fact or leave it out, state it!
5540 Often people omit facts because they think they know what causes the problem
5541 and assume that some details do not matter. Thus, you might assume that the
5542 name of a symbol you use in an example does not matter. Well, probably it does
5543 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
5544 happens to fetch from the location where that name is stored in memory;
5545 perhaps, if the name were different, the contents of that location would fool
5546 the assembler into doing the right thing despite the bug. Play it safe and
5547 give a specific, complete example. That is the easiest thing for you to do,
5548 and the most helpful.
5550 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
5551 it is new to us. Therefore, always write your bug reports on the assumption
5552 that the bug has not been reported previously.
5554 Sometimes people give a few sketchy facts and ask, ``Does this ring a
5555 bell?'' Those bug reports are useless, and we urge everyone to
5556 @emph{refuse to respond to them} except to chide the sender to report
5559 To enable us to fix the bug, you should include all these things:
5563 The version of @code{@value{AS}}. @code{@value{AS}} announces it if you start
5564 it with the @samp{--version} argument.
5566 Without this, we will not know whether there is any point in looking for
5567 the bug in the current version of @code{@value{AS}}.
5570 Any patches you may have applied to the @code{@value{AS}} source.
5573 The type of machine you are using, and the operating system name and
5577 What compiler (and its version) was used to compile @code{@value{AS}}---e.g.
5581 The command arguments you gave the assembler to assemble your example and
5582 observe the bug. To guarantee you will not omit something important, list them
5583 all. A copy of the Makefile (or the output from make) is sufficient.
5585 If we were to try to guess the arguments, we would probably guess wrong
5586 and then we might not encounter the bug.
5589 A complete input file that will reproduce the bug. If the bug is observed when
5590 the assembler is invoked via a compiler, send the assembler source, not the
5591 high level language source. Most compilers will produce the assembler source
5592 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
5593 the options @samp{-v --save-temps}; this will save the assembler source in a
5594 file with an extension of @file{.s}, and also show you exactly how
5595 @code{@value{AS}} is being run.
5598 A description of what behavior you observe that you believe is
5599 incorrect. For example, ``It gets a fatal signal.''
5601 Of course, if the bug is that @code{@value{AS}} gets a fatal signal, then we
5602 will certainly notice it. But if the bug is incorrect output, we might not
5603 notice unless it is glaringly wrong. You might as well not give us a chance to
5606 Even if the problem you experience is a fatal signal, you should still say so
5607 explicitly. Suppose something strange is going on, such as, your copy of
5608 @code{@value{AS}} is out of synch, or you have encountered a bug in the C
5609 library on your system. (This has happened!) Your copy might crash and ours
5610 would not. If you told us to expect a crash, then when ours fails to crash, we
5611 would know that the bug was not happening for us. If you had not told us to
5612 expect a crash, then we would not be able to draw any conclusion from our
5616 If you wish to suggest changes to the @code{@value{AS}} source, send us context
5617 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
5618 option. Always send diffs from the old file to the new file. If you even
5619 discuss something in the @code{@value{AS}} source, refer to it by context, not
5622 The line numbers in our development sources will not match those in your
5623 sources. Your line numbers would convey no useful information to us.
5626 Here are some things that are not necessary:
5630 A description of the envelope of the bug.
5632 Often people who encounter a bug spend a lot of time investigating
5633 which changes to the input file will make the bug go away and which
5634 changes will not affect it.
5636 This is often time consuming and not very useful, because the way we
5637 will find the bug is by running a single example under the debugger
5638 with breakpoints, not by pure deduction from a series of examples.
5639 We recommend that you save your time for something else.
5641 Of course, if you can find a simpler example to report @emph{instead}
5642 of the original one, that is a convenience for us. Errors in the
5643 output will be easier to spot, running under the debugger will take
5644 less time, and so on.
5646 However, simplification is not vital; if you do not want to do this,
5647 report the bug anyway and send us the entire test case you used.
5650 A patch for the bug.
5652 A patch for the bug does help us if it is a good one. But do not omit
5653 the necessary information, such as the test case, on the assumption that
5654 a patch is all we need. We might see problems with your patch and decide
5655 to fix the problem another way, or we might not understand it at all.
5657 Sometimes with a program as complicated as @code{@value{AS}} it is very hard to
5658 construct an example that will make the program follow a certain path through
5659 the code. If you do not send us the example, we will not be able to construct
5660 one, so we will not be able to verify that the bug is fixed.
5662 And if we cannot understand what bug you are trying to fix, or why your
5663 patch should be an improvement, we will not install it. A test case will
5664 help us to understand.
5667 A guess about what the bug is or what it depends on.
5669 Such guesses are usually wrong. Even we cannot guess right about such
5670 things without first using the debugger to find the facts.
5673 @node Acknowledgements
5674 @chapter Acknowledgements
5676 If you have contributed to @code{@value{AS}} and your name isn't listed here,
5677 it is not meant as a slight. We just don't know about it. Send mail to the
5678 maintainer, and we'll correct the situation. Currently
5680 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
5682 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
5685 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
5686 information and the 68k series machines, most of the preprocessing pass, and
5687 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
5689 K. Richard Pixley maintained GAS for a while, adding various enhancements and
5690 many bug fixes, including merging support for several processors, breaking GAS
5691 up to handle multiple object file format back ends (including heavy rewrite,
5692 testing, an integration of the coff and b.out back ends), adding configuration
5693 including heavy testing and verification of cross assemblers and file splits
5694 and renaming, converted GAS to strictly ANSI C including full prototypes, added
5695 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
5696 port (including considerable amounts of reverse engineering), a SPARC opcode
5697 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
5698 assertions and made them work, much other reorganization, cleanup, and lint.
5700 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
5701 in format-specific I/O modules.
5703 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
5704 has done much work with it since.
5706 The Intel 80386 machine description was written by Eliot Dresselhaus.
5708 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
5710 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
5711 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
5713 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
5714 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
5715 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
5716 support a.out format.
5718 Support for the Zilog Z8k and Hitachi H8/300 and H8/500 processors (tc-z8k,
5719 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
5720 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
5721 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
5724 John Gilmore built the AMD 29000 support, added @code{.include} support, and
5725 simplified the configuration of which versions accept which directives. He
5726 updated the 68k machine description so that Motorola's opcodes always produced
5727 fixed-size instructions (e.g. @code{jsr}), while synthetic instructions
5728 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
5729 cross-compilation support, and one bug in relaxation that took a week and
5730 required the proverbial one-bit fix.
5732 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
5733 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
5734 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
5735 PowerPC assembler, and made a few other minor patches.
5737 Steve Chamberlain made @code{@value{AS}} able to generate listings.
5739 Hewlett-Packard contributed support for the HP9000/300.
5741 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
5742 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
5743 formats). This work was supported by both the Center for Software Science at
5744 the University of Utah and Cygnus Support.
5746 Support for ELF format files has been worked on by Mark Eichin of Cygnus
5747 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
5748 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
5749 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
5750 and some initial 64-bit support).
5752 Linas Vepstas added GAS support for the ESA/390 "IBM 370" architecture.
5754 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
5755 support for openVMS/Alpha.
5757 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
5760 Several engineers at Cygnus Support have also provided many small bug fixes and
5761 configuration enhancements.
5763 Many others have contributed large or small bugfixes and enhancements. If
5764 you have contributed significant work and are not mentioned on this list, and
5765 want to be, let us know. Some of the history has been lost; we are not
5766 intentionally leaving anyone out.