1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
3 @c 2001, 2002, 2003, 2004, 2005
4 @c Free Software Foundation, Inc.
5 @c UPDATE!! On future updates--
6 @c (1) check for new machine-dep cmdline options in
7 @c md_parse_option definitions in config/tc-*.c
8 @c (2) for platform-specific directives, examine md_pseudo_op
10 @c (3) for object-format specific directives, examine obj_pseudo_op
12 @c (4) portable directives in potable[] in read.c
16 @macro gcctabopt{body}
19 @c defaults, config file may override:
24 @include asconfig.texi
29 @c common OR combinations of conditions
55 @set abnormal-separator
59 @settitle Using @value{AS}
62 @settitle Using @value{AS} (@value{TARGET})
64 @setchapternewpage odd
69 @c WARE! Some of the machine-dependent sections contain tables of machine
70 @c instructions. Except in multi-column format, these tables look silly.
71 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
72 @c the multi-col format is faked within @example sections.
74 @c Again unfortunately, the natural size that fits on a page, for these tables,
75 @c is different depending on whether or not smallbook is turned on.
76 @c This matters, because of order: text flow switches columns at each page
79 @c The format faked in this source works reasonably well for smallbook,
80 @c not well for the default large-page format. This manual expects that if you
81 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
82 @c tables in question. You can turn on one without the other at your
83 @c discretion, of course.
86 @c the insn tables look just as silly in info files regardless of smallbook,
87 @c might as well show 'em anyways.
93 * As: (as). The GNU assembler.
94 * Gas: (as). The GNU assembler.
103 This file documents the GNU Assembler "@value{AS}".
105 @c man begin COPYRIGHT
106 Copyright (C) 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002 Free Software Foundation, Inc.
108 Permission is granted to copy, distribute and/or modify this document
109 under the terms of the GNU Free Documentation License, Version 1.1
110 or any later version published by the Free Software Foundation;
111 with no Invariant Sections, with no Front-Cover Texts, and with no
112 Back-Cover Texts. A copy of the license is included in the
113 section entitled ``GNU Free Documentation License''.
118 Permission is granted to process this file through Tex and print the
119 results, provided the printed document carries copying permission
120 notice identical to this one except for the removal of this paragraph
121 (this paragraph not being relevant to the printed manual).
127 @title Using @value{AS}
128 @subtitle The @sc{gnu} Assembler
130 @subtitle for the @value{TARGET} family
133 @subtitle Version @value{VERSION}
136 The Free Software Foundation Inc. thanks The Nice Computer
137 Company of Australia for loaning Dean Elsner to write the
138 first (Vax) version of @command{as} for Project @sc{gnu}.
139 The proprietors, management and staff of TNCCA thank FSF for
140 distracting the boss while they got some work
143 @author Dean Elsner, Jay Fenlason & friends
147 \hfill {\it Using {\tt @value{AS}}}\par
148 \hfill Edited by Cygnus Support\par
150 %"boxit" macro for figures:
151 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
152 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
153 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
154 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
155 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
158 @vskip 0pt plus 1filll
159 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002 Free Software Foundation, Inc.
161 Permission is granted to copy, distribute and/or modify this document
162 under the terms of the GNU Free Documentation License, Version 1.1
163 or any later version published by the Free Software Foundation;
164 with no Invariant Sections, with no Front-Cover Texts, and with no
165 Back-Cover Texts. A copy of the license is included in the
166 section entitled ``GNU Free Documentation License''.
172 @top Using @value{AS}
174 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}} version
177 This version of the file describes @command{@value{AS}} configured to generate
178 code for @value{TARGET} architectures.
181 This document is distributed under the terms of the GNU Free
182 Documentation License. A copy of the license is included in the
183 section entitled ``GNU Free Documentation License''.
186 * Overview:: Overview
187 * Invoking:: Command-Line Options
189 * Sections:: Sections and Relocation
191 * Expressions:: Expressions
192 * Pseudo Ops:: Assembler Directives
193 * Machine Dependencies:: Machine Dependent Features
194 * Reporting Bugs:: Reporting Bugs
195 * Acknowledgements:: Who Did What
196 * GNU Free Documentation License:: GNU Free Documentation License
204 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
206 This version of the manual describes @command{@value{AS}} configured to generate
207 code for @value{TARGET} architectures.
211 @cindex invocation summary
212 @cindex option summary
213 @cindex summary of options
214 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
215 @pxref{Invoking,,Command-Line Options}.
217 @c man title AS the portable GNU assembler.
221 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
225 @c We don't use deffn and friends for the following because they seem
226 @c to be limited to one line for the header.
228 @c man begin SYNOPSIS
229 @value{AS} [@b{-a}[@b{cdhlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
230 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}] [@b{--gstabs+}]
231 [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}] [@b{-K}] [@b{-L}]
232 [@b{--listing-lhs-width}=@var{NUM}] [@b{--listing-lhs-width2}=@var{NUM}]
233 [@b{--listing-rhs-width}=@var{NUM}] [@b{--listing-cont-lines}=@var{NUM}]
234 [@b{--keep-locals}] [@b{-o} @var{objfile}] [@b{-R}] [@b{--statistics}] [@b{-v}]
235 [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}] [@b{--fatal-warnings}]
236 [@b{-w}] [@b{-x}] [@b{-Z}] [@b{--target-help}] [@var{target-options}]
237 [@b{--}|@var{files} @dots{}]
239 @c Target dependent options are listed below. Keep the list sorted.
240 @c Add an empty line for separation.
242 @c am29k has no machine-dependent assembler options
246 @emph{Target Alpha options:}
248 [@b{-mdebug} | @b{-no-mdebug}]
249 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
250 [@b{-F}] [@b{-32addr}]
254 @emph{Target ARC options:}
260 @emph{Target ARM options:}
261 @c Don't document the deprecated options
262 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
263 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
264 [@b{-mfpu}=@var{floating-point-format}]
265 [@b{-mfloat-abi}=@var{abi}]
266 [@b{-meabi}=@var{ver}]
269 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
270 @b{-mapcs-reentrant}]
271 [@b{-mthumb-interwork}] [@b{-k}]
275 @emph{Target CRIS options:}
276 [@b{--underscore} | @b{--no-underscore}]
278 [@b{--emulation=criself} | @b{--emulation=crisaout}]
279 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
280 @c Deprecated -- deliberately not documented.
285 @emph{Target D10V options:}
290 @emph{Target D30V options:}
291 [@b{-O}|@b{-n}|@b{-N}]
294 @c Renesas family chips have no machine-dependent assembler options
297 @c HPPA has no machine-dependent assembler options (yet).
301 @emph{Target i386 options:}
302 [@b{--32}|@b{--64}] [@b{-n}]
306 @emph{Target i960 options:}
307 @c see md_parse_option in tc-i960.c
308 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
310 [@b{-b}] [@b{-no-relax}]
314 @emph{Target IA-64 options:}
315 [@b{-mconstant-gp}|@b{-mauto-pic}]
316 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
318 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
319 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
320 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
324 @emph{Target IP2K options:}
325 [@b{-mip2022}|@b{-mip2022ext}]
329 @emph{Target M32R options:}
330 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
335 @emph{Target M680X0 options:}
336 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
340 @emph{Target M68HC11 options:}
341 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
342 [@b{-mshort}|@b{-mlong}]
343 [@b{-mshort-double}|@b{-mlong-double}]
344 [@b{--force-long-branchs}] [@b{--short-branchs}]
345 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
346 [@b{--print-opcodes}] [@b{--generate-example}]
350 @emph{Target MCORE options:}
351 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
352 [@b{-mcpu=[210|340]}]
356 @emph{Target MIPS options:}
357 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
358 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
359 [@b{-non_shared}] [@b{-xgot}]
360 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
361 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
362 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
363 [@b{-mips64}] [@b{-mips64r2}]
364 [@b{-construct-floats}] [@b{-no-construct-floats}]
365 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
366 [@b{-mfix7000}] [@b{-mno-fix7000}]
367 [@b{-mips16}] [@b{-no-mips16}]
368 [@b{-mips3d}] [@b{-no-mips3d}]
369 [@b{-mdmx}] [@b{-no-mdmx}]
370 [@b{-mdebug}] [@b{-no-mdebug}]
371 [@b{-mpdr}] [@b{-mno-pdr}]
375 @emph{Target MMIX options:}
376 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
377 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
378 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
379 [@b{--linker-allocated-gregs}]
383 @emph{Target PDP11 options:}
384 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
385 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
386 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
390 @emph{Target picoJava options:}
395 @emph{Target PowerPC options:}
396 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
397 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}|
398 @b{-mbooke32}|@b{-mbooke64}]
399 [@b{-mcom}|@b{-many}|@b{-maltivec}] [@b{-memb}]
400 [@b{-mregnames}|@b{-mno-regnames}]
401 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
402 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
403 [@b{-msolaris}|@b{-mno-solaris}]
407 @emph{Target SPARC options:}
408 @c The order here is important. See c-sparc.texi.
409 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
410 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
411 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
416 @emph{Target TIC54X options:}
417 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
418 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
421 @c Z8000 has no machine-dependent assembler options
425 @emph{Target Xtensa options:}
426 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
427 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
428 [@b{--[no-]transform}]
429 [@b{--rename-section} @var{oldname}=@var{newname}]
438 Turn on listings, in any of a variety of ways:
442 omit false conditionals
445 omit debugging directives
448 include high-level source
454 include macro expansions
457 omit forms processing
463 set the name of the listing file
466 You may combine these options; for example, use @samp{-aln} for assembly
467 listing without forms processing. The @samp{=file} option, if used, must be
468 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
471 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
474 Ignored. This option is accepted for script compatibility with calls to
477 @item --defsym @var{sym}=@var{value}
478 Define the symbol @var{sym} to be @var{value} before assembling the input file.
479 @var{value} must be an integer constant. As in C, a leading @samp{0x}
480 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
483 ``fast''---skip whitespace and comment preprocessing (assume source is
488 Generate debugging information for each assembler source line using whichever
489 debug format is preferred by the target. This currently means either STABS,
493 Generate stabs debugging information for each assembler line. This
494 may help debugging assembler code, if the debugger can handle it.
497 Generate stabs debugging information for each assembler line, with GNU
498 extensions that probably only gdb can handle, and that could make other
499 debuggers crash or refuse to read your program. This
500 may help debugging assembler code. Currently the only GNU extension is
501 the location of the current working directory at assembling time.
504 Generate DWARF2 debugging information for each assembler line. This
505 may help debugging assembler code, if the debugger can handle it. Note---this
506 option is only supported by some targets, not all of them.
509 Print a summary of the command line options and exit.
512 Print a summary of all target specific options and exit.
515 Add directory @var{dir} to the search list for @code{.include} directives.
518 Don't warn about signed overflow.
521 @ifclear DIFF-TBL-KLUGE
522 This option is accepted but has no effect on the @value{TARGET} family.
524 @ifset DIFF-TBL-KLUGE
525 Issue warnings when difference tables altered for long displacements.
530 Keep (in the symbol table) local symbols. On traditional a.out systems
531 these start with @samp{L}, but different systems have different local
534 @item --listing-lhs-width=@var{number}
535 Set the maximum width, in words, of the output data column for an assembler
536 listing to @var{number}.
538 @item --listing-lhs-width2=@var{number}
539 Set the maximum width, in words, of the output data column for continuation
540 lines in an assembler listing to @var{number}.
542 @item --listing-rhs-width=@var{number}
543 Set the maximum width of an input source line, as displayed in a listing, to
546 @item --listing-cont-lines=@var{number}
547 Set the maximum number of lines printed in a listing for a single line of input
550 @item -o @var{objfile}
551 Name the object-file output from @command{@value{AS}} @var{objfile}.
554 Fold the data section into the text section.
557 Print the maximum space (in bytes) and total time (in seconds) used by
560 @item --strip-local-absolute
561 Remove local absolute symbols from the outgoing symbol table.
565 Print the @command{as} version.
568 Print the @command{as} version and exit.
572 Suppress warning messages.
574 @item --fatal-warnings
575 Treat warnings as errors.
578 Don't suppress warning messages or treat them as errors.
587 Generate an object file even after errors.
589 @item -- | @var{files} @dots{}
590 Standard input, or source files to assemble.
595 The following options are available when @value{AS} is configured for
600 This option selects the core processor variant.
602 Select either big-endian (-EB) or little-endian (-EL) output.
607 The following options are available when @value{AS} is configured for the ARM
611 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
612 Specify which ARM processor variant is the target.
613 @item -march=@var{architecture}[+@var{extension}@dots{}]
614 Specify which ARM architecture variant is used by the target.
615 @item -mfpu=@var{floating-point-format}
616 Select which Floating Point architecture is the target.
617 @item -mfloat-abi=@var{abi}
618 Select which floating point ABI is in use.
620 Enable Thumb only instruction decoding.
621 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
622 Select which procedure calling convention is in use.
624 Select either big-endian (-EB) or little-endian (-EL) output.
625 @item -mthumb-interwork
626 Specify that the code has been generated with interworking between Thumb and
629 Specify that PIC code has been generated.
634 See the info pages for documentation of the CRIS-specific options.
638 The following options are available when @value{AS} is configured for
641 @cindex D10V optimization
642 @cindex optimization, D10V
644 Optimize output by parallelizing instructions.
649 The following options are available when @value{AS} is configured for a D30V
652 @cindex D30V optimization
653 @cindex optimization, D30V
655 Optimize output by parallelizing instructions.
659 Warn when nops are generated.
661 @cindex D30V nops after 32-bit multiply
663 Warn when a nop after a 32-bit multiply instruction is generated.
668 The following options are available when @value{AS} is configured for the
669 Intel 80960 processor.
672 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
673 Specify which variant of the 960 architecture is the target.
676 Add code to collect statistics about branches taken.
679 Do not alter compare-and-branch instructions for long displacements;
686 The following options are available when @value{AS} is configured for the
692 Specifies that the extended IP2022 instructions are allowed.
695 Restores the default behaviour, which restricts the permitted instructions to
696 just the basic IP2022 ones.
702 The following options are available when @value{AS} is configured for the
703 Renesas M32R (formerly Mitsubishi M32R) series.
708 Specify which processor in the M32R family is the target. The default
709 is normally the M32R, but this option changes it to the M32RX.
711 @item --warn-explicit-parallel-conflicts or --Wp
712 Produce warning messages when questionable parallel constructs are
715 @item --no-warn-explicit-parallel-conflicts or --Wnp
716 Do not produce warning messages when questionable parallel constructs are
723 The following options are available when @value{AS} is configured for the
724 Motorola 68000 series.
729 Shorten references to undefined symbols, to one word instead of two.
731 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
732 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
733 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
734 Specify what processor in the 68000 family is the target. The default
735 is normally the 68020, but this can be changed at configuration time.
737 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
738 The target machine does (or does not) have a floating-point coprocessor.
739 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
740 the basic 68000 is not compatible with the 68881, a combination of the
741 two can be specified, since it's possible to do emulation of the
742 coprocessor instructions with the main processor.
744 @item -m68851 | -mno-68851
745 The target machine does (or does not) have a memory-management
746 unit coprocessor. The default is to assume an MMU for 68020 and up.
753 For details about the PDP-11 machine dependent features options,
754 see @ref{PDP-11-Options}.
757 @item -mpic | -mno-pic
758 Generate position-independent (or position-dependent) code. The
759 default is @option{-mpic}.
762 @itemx -mall-extensions
763 Enable all instruction set extensions. This is the default.
765 @item -mno-extensions
766 Disable all instruction set extensions.
768 @item -m@var{extension} | -mno-@var{extension}
769 Enable (or disable) a particular instruction set extension.
772 Enable the instruction set extensions supported by a particular CPU, and
773 disable all other extensions.
775 @item -m@var{machine}
776 Enable the instruction set extensions supported by a particular machine
777 model, and disable all other extensions.
783 The following options are available when @value{AS} is configured for
784 a picoJava processor.
788 @cindex PJ endianness
789 @cindex endianness, PJ
790 @cindex big endian output, PJ
792 Generate ``big endian'' format output.
794 @cindex little endian output, PJ
796 Generate ``little endian'' format output.
802 The following options are available when @value{AS} is configured for the
803 Motorola 68HC11 or 68HC12 series.
807 @item -m68hc11 | -m68hc12 | -m68hcs12
808 Specify what processor is the target. The default is
809 defined by the configuration option when building the assembler.
812 Specify to use the 16-bit integer ABI.
815 Specify to use the 32-bit integer ABI.
818 Specify to use the 32-bit double ABI.
821 Specify to use the 64-bit double ABI.
823 @item --force-long-branchs
824 Relative branches are turned into absolute ones. This concerns
825 conditional branches, unconditional branches and branches to a
828 @item -S | --short-branchs
829 Do not turn relative branchs into absolute ones
830 when the offset is out of range.
832 @item --strict-direct-mode
833 Do not turn the direct addressing mode into extended addressing mode
834 when the instruction does not support direct addressing mode.
836 @item --print-insn-syntax
837 Print the syntax of instruction in case of error.
839 @item --print-opcodes
840 print the list of instructions with syntax and then exit.
842 @item --generate-example
843 print an example of instruction for each possible instruction and then exit.
844 This option is only useful for testing @command{@value{AS}}.
850 The following options are available when @command{@value{AS}} is configured
851 for the SPARC architecture:
854 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
855 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
856 Explicitly select a variant of the SPARC architecture.
858 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
859 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
861 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
862 UltraSPARC extensions.
864 @item -xarch=v8plus | -xarch=v8plusa
865 For compatibility with the Solaris v9 assembler. These options are
866 equivalent to -Av8plus and -Av8plusa, respectively.
869 Warn when the assembler switches to another architecture.
874 The following options are available when @value{AS} is configured for the 'c54x
879 Enable extended addressing mode. All addresses and relocations will assume
880 extended addressing (usually 23 bits).
881 @item -mcpu=@var{CPU_VERSION}
882 Sets the CPU version being compiled for.
883 @item -merrors-to-file @var{FILENAME}
884 Redirect error output to a file, for broken systems which don't support such
885 behaviour in the shell.
890 The following options are available when @value{AS} is configured for
891 a @sc{mips} processor.
895 This option sets the largest size of an object that can be referenced
896 implicitly with the @code{gp} register. It is only accepted for targets that
897 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
899 @cindex MIPS endianness
900 @cindex endianness, MIPS
901 @cindex big endian output, MIPS
903 Generate ``big endian'' format output.
905 @cindex little endian output, MIPS
907 Generate ``little endian'' format output.
919 Generate code for a particular @sc{mips} Instruction Set Architecture level.
920 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
921 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
922 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
923 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
925 correspond to generic
926 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
927 and @samp{MIPS64 Release 2}
928 ISA processors, respectively.
930 @item -march=@var{CPU}
931 Generate code for a particular @sc{mips} cpu.
933 @item -mtune=@var{cpu}
934 Schedule and tune for a particular @sc{mips} cpu.
938 Cause nops to be inserted if the read of the destination register
939 of an mfhi or mflo instruction occurs in the following two instructions.
943 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
944 section instead of the standard ELF .stabs sections.
948 Control generation of @code{.pdr} sections.
952 The register sizes are normally inferred from the ISA and ABI, but these
953 flags force a certain group of registers to be treated as 32 bits wide at
954 all times. @samp{-mgp32} controls the size of general-purpose registers
955 and @samp{-mfp32} controls the size of floating-point registers.
959 Generate code for the MIPS 16 processor. This is equivalent to putting
960 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
961 turns off this option.
965 Generate code for the MIPS-3D Application Specific Extension.
966 This tells the assembler to accept MIPS-3D instructions.
967 @samp{-no-mips3d} turns off this option.
971 Generate code for the MDMX Application Specific Extension.
972 This tells the assembler to accept MDMX instructions.
973 @samp{-no-mdmx} turns off this option.
975 @item --construct-floats
976 @itemx --no-construct-floats
977 The @samp{--no-construct-floats} option disables the construction of
978 double width floating point constants by loading the two halves of the
979 value into the two single width floating point registers that make up
980 the double width register. By default @samp{--construct-floats} is
981 selected, allowing construction of these floating point constants.
984 @item --emulation=@var{name}
985 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
986 for some other target, in all respects, including output format (choosing
987 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
988 debugging information or store symbol table information, and default
989 endianness. The available configuration names are: @samp{mipsecoff},
990 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
991 @samp{mipsbelf}. The first two do not alter the default endianness from that
992 of the primary target for which the assembler was configured; the others change
993 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
994 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
995 selection in any case.
997 This option is currently supported only when the primary target
998 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
999 Furthermore, the primary target or others specified with
1000 @samp{--enable-targets=@dots{}} at configuration time must include support for
1001 the other format, if both are to be available. For example, the Irix 5
1002 configuration includes support for both.
1004 Eventually, this option will support more configurations, with more
1005 fine-grained control over the assembler's behavior, and will be supported for
1009 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1016 Control how to deal with multiplication overflow and division by zero.
1017 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1018 (and only work for Instruction Set Architecture level 2 and higher);
1019 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1023 When this option is used, @command{@value{AS}} will issue a warning every
1024 time it generates a nop instruction from a macro.
1029 The following options are available when @value{AS} is configured for
1035 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1036 The command line option @samp{-nojsri2bsr} can be used to disable it.
1040 Enable or disable the silicon filter behaviour. By default this is disabled.
1041 The default can be overridden by the @samp{-sifilter} command line option.
1044 Alter jump instructions for long displacements.
1046 @item -mcpu=[210|340]
1047 Select the cpu type on the target hardware. This controls which instructions
1051 Assemble for a big endian target.
1054 Assemble for a little endian target.
1060 See the info pages for documentation of the MMIX-specific options.
1064 The following options are available when @value{AS} is configured for
1065 an Xtensa processor.
1068 @item --text-section-literals | --no-text-section-literals
1069 With @option{--text-@-section-@-literals}, literal pools are interspersed
1070 in the text section. The default is
1071 @option{--no-@-text-@-section-@-literals}, which places literals in a
1072 separate section in the output file. These options only affect literals
1073 referenced via PC-relative @code{L32R} instructions; literals for
1074 absolute mode @code{L32R} instructions are handled separately.
1076 @item --absolute-literals | --no-absolute-literals
1077 Indicate to the assembler whether @code{L32R} instructions use absolute
1078 or PC-relative addressing. The default is to assume absolute addressing
1079 if the Xtensa processor includes the absolute @code{L32R} addressing
1080 option. Otherwise, only the PC-relative @code{L32R} mode can be used.
1082 @item --target-align | --no-target-align
1083 Enable or disable automatic alignment to reduce branch penalties at the
1084 expense of some code density. The default is @option{--target-@-align}.
1086 @item --longcalls | --no-longcalls
1087 Enable or disable transformation of call instructions to allow calls
1088 across a greater range of addresses. The default is
1089 @option{--no-@-longcalls}.
1091 @item --transform | --no-transform
1092 Enable or disable all assembler transformations of Xtensa instructions.
1093 The default is @option{--transform};
1094 @option{--no-transform} should be used only in the rare cases when the
1095 instructions must be exactly as specified in the assembly source.
1102 * Manual:: Structure of this Manual
1103 * GNU Assembler:: The GNU Assembler
1104 * Object Formats:: Object File Formats
1105 * Command Line:: Command Line
1106 * Input Files:: Input Files
1107 * Object:: Output (Object) File
1108 * Errors:: Error and Warning Messages
1112 @section Structure of this Manual
1114 @cindex manual, structure and purpose
1115 This manual is intended to describe what you need to know to use
1116 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1117 notation for symbols, constants, and expressions; the directives that
1118 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1121 We also cover special features in the @value{TARGET}
1122 configuration of @command{@value{AS}}, including assembler directives.
1125 This manual also describes some of the machine-dependent features of
1126 various flavors of the assembler.
1129 @cindex machine instructions (not covered)
1130 On the other hand, this manual is @emph{not} intended as an introduction
1131 to programming in assembly language---let alone programming in general!
1132 In a similar vein, we make no attempt to introduce the machine
1133 architecture; we do @emph{not} describe the instruction set, standard
1134 mnemonics, registers or addressing modes that are standard to a
1135 particular architecture.
1137 You may want to consult the manufacturer's
1138 machine architecture manual for this information.
1142 For information on the H8/300 machine instruction set, see @cite{H8/300
1143 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1144 Programming Manual} (Renesas).
1147 For information on the H8/500 machine instruction set, see @cite{H8/500
1148 Series Programming Manual} (Renesas M21T001).
1151 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1152 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1153 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1154 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1157 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1161 @c I think this is premature---doc@cygnus.com, 17jan1991
1163 Throughout this manual, we assume that you are running @dfn{GNU},
1164 the portable operating system from the @dfn{Free Software
1165 Foundation, Inc.}. This restricts our attention to certain kinds of
1166 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1167 once this assumption is granted examples and definitions need less
1170 @command{@value{AS}} is part of a team of programs that turn a high-level
1171 human-readable series of instructions into a low-level
1172 computer-readable series of instructions. Different versions of
1173 @command{@value{AS}} are used for different kinds of computer.
1176 @c There used to be a section "Terminology" here, which defined
1177 @c "contents", "byte", "word", and "long". Defining "word" to any
1178 @c particular size is confusing when the .word directive may generate 16
1179 @c bits on one machine and 32 bits on another; in general, for the user
1180 @c version of this manual, none of these terms seem essential to define.
1181 @c They were used very little even in the former draft of the manual;
1182 @c this draft makes an effort to avoid them (except in names of
1186 @section The GNU Assembler
1188 @c man begin DESCRIPTION
1190 @sc{gnu} @command{as} is really a family of assemblers.
1192 This manual describes @command{@value{AS}}, a member of that family which is
1193 configured for the @value{TARGET} architectures.
1195 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1196 should find a fairly similar environment when you use it on another
1197 architecture. Each version has much in common with the others,
1198 including object file formats, most assembler directives (often called
1199 @dfn{pseudo-ops}) and assembler syntax.@refill
1201 @cindex purpose of @sc{gnu} assembler
1202 @command{@value{AS}} is primarily intended to assemble the output of the
1203 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1204 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1205 assemble correctly everything that other assemblers for the same
1206 machine would assemble.
1208 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1211 @c This remark should appear in generic version of manual; assumption
1212 @c here is that generic version sets M680x0.
1213 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1214 assembler for the same architecture; for example, we know of several
1215 incompatible versions of 680x0 assembly language syntax.
1220 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1221 program in one pass of the source file. This has a subtle impact on the
1222 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1224 @node Object Formats
1225 @section Object File Formats
1227 @cindex object file format
1228 The @sc{gnu} assembler can be configured to produce several alternative
1229 object file formats. For the most part, this does not affect how you
1230 write assembly language programs; but directives for debugging symbols
1231 are typically different in different file formats. @xref{Symbol
1232 Attributes,,Symbol Attributes}.
1235 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1236 @value{OBJ-NAME} format object files.
1238 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1240 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1241 @code{a.out} or COFF format object files.
1244 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1245 @code{b.out} or COFF format object files.
1248 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1249 SOM or ELF format object files.
1254 @section Command Line
1256 @cindex command line conventions
1258 After the program name @command{@value{AS}}, the command line may contain
1259 options and file names. Options may appear in any order, and may be
1260 before, after, or between file names. The order of file names is
1263 @cindex standard input, as input file
1265 @file{--} (two hyphens) by itself names the standard input file
1266 explicitly, as one of the files for @command{@value{AS}} to assemble.
1268 @cindex options, command line
1269 Except for @samp{--} any command line argument that begins with a
1270 hyphen (@samp{-}) is an option. Each option changes the behavior of
1271 @command{@value{AS}}. No option changes the way another option works. An
1272 option is a @samp{-} followed by one or more letters; the case of
1273 the letter is important. All options are optional.
1275 Some options expect exactly one file name to follow them. The file
1276 name may either immediately follow the option's letter (compatible
1277 with older assemblers) or it may be the next command argument (@sc{gnu}
1278 standard). These two command lines are equivalent:
1281 @value{AS} -o my-object-file.o mumble.s
1282 @value{AS} -omy-object-file.o mumble.s
1286 @section Input Files
1289 @cindex source program
1290 @cindex files, input
1291 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1292 describe the program input to one run of @command{@value{AS}}. The program may
1293 be in one or more files; how the source is partitioned into files
1294 doesn't change the meaning of the source.
1296 @c I added "con" prefix to "catenation" just to prove I can overcome my
1297 @c APL training... doc@cygnus.com
1298 The source program is a concatenation of the text in all the files, in the
1301 @c man begin DESCRIPTION
1302 Each time you run @command{@value{AS}} it assembles exactly one source
1303 program. The source program is made up of one or more files.
1304 (The standard input is also a file.)
1306 You give @command{@value{AS}} a command line that has zero or more input file
1307 names. The input files are read (from left file name to right). A
1308 command line argument (in any position) that has no special meaning
1309 is taken to be an input file name.
1311 If you give @command{@value{AS}} no file names it attempts to read one input file
1312 from the @command{@value{AS}} standard input, which is normally your terminal. You
1313 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1316 Use @samp{--} if you need to explicitly name the standard input file
1317 in your command line.
1319 If the source is empty, @command{@value{AS}} produces a small, empty object
1324 @subheading Filenames and Line-numbers
1326 @cindex input file linenumbers
1327 @cindex line numbers, in input files
1328 There are two ways of locating a line in the input file (or files) and
1329 either may be used in reporting error messages. One way refers to a line
1330 number in a physical file; the other refers to a line number in a
1331 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1333 @dfn{Physical files} are those files named in the command line given
1334 to @command{@value{AS}}.
1336 @dfn{Logical files} are simply names declared explicitly by assembler
1337 directives; they bear no relation to physical files. Logical file names help
1338 error messages reflect the original source file, when @command{@value{AS}} source
1339 is itself synthesized from other files. @command{@value{AS}} understands the
1340 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1341 @ref{File,,@code{.file}}.
1344 @section Output (Object) File
1350 Every time you run @command{@value{AS}} it produces an output file, which is
1351 your assembly language program translated into numbers. This file
1352 is the object file. Its default name is
1360 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1362 You can give it another name by using the @option{-o} option. Conventionally,
1363 object file names end with @file{.o}. The default name is used for historical
1364 reasons: older assemblers were capable of assembling self-contained programs
1365 directly into a runnable program. (For some formats, this isn't currently
1366 possible, but it can be done for the @code{a.out} format.)
1370 The object file is meant for input to the linker @code{@value{LD}}. It contains
1371 assembled program code, information to help @code{@value{LD}} integrate
1372 the assembled program into a runnable file, and (optionally) symbolic
1373 information for the debugger.
1375 @c link above to some info file(s) like the description of a.out.
1376 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1379 @section Error and Warning Messages
1381 @c man begin DESCRIPTION
1383 @cindex error messages
1384 @cindex warning messages
1385 @cindex messages from assembler
1386 @command{@value{AS}} may write warnings and error messages to the standard error
1387 file (usually your terminal). This should not happen when a compiler
1388 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1389 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1390 grave problem that stops the assembly.
1394 @cindex format of warning messages
1395 Warning messages have the format
1398 file_name:@b{NNN}:Warning Message Text
1402 @cindex line numbers, in warnings/errors
1403 (where @b{NNN} is a line number). If a logical file name has been given
1404 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1405 the current input file is used. If a logical line number was given
1407 (@pxref{Line,,@code{.line}})
1411 (@pxref{Line,,@code{.line}})
1414 (@pxref{Ln,,@code{.ln}})
1417 then it is used to calculate the number printed,
1418 otherwise the actual line in the current source file is printed. The
1419 message text is intended to be self explanatory (in the grand Unix
1422 @cindex format of error messages
1423 Error messages have the format
1425 file_name:@b{NNN}:FATAL:Error Message Text
1427 The file name and line number are derived as for warning
1428 messages. The actual message text may be rather less explanatory
1429 because many of them aren't supposed to happen.
1432 @chapter Command-Line Options
1434 @cindex options, all versions of assembler
1435 This chapter describes command-line options available in @emph{all}
1436 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
1438 to the @value{TARGET} target.
1441 to particular machine architectures.
1444 @c man begin DESCRIPTION
1446 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1447 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1448 The assembler arguments must be separated from each other (and the @samp{-Wa})
1449 by commas. For example:
1452 gcc -c -g -O -Wa,-alh,-L file.c
1456 This passes two options to the assembler: @samp{-alh} (emit a listing to
1457 standard output with high-level and assembly source) and @samp{-L} (retain
1458 local symbols in the symbol table).
1460 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1461 command-line options are automatically passed to the assembler by the compiler.
1462 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1463 precisely what options it passes to each compilation pass, including the
1469 * a:: -a[cdhlns] enable listings
1470 * alternate:: --alternate enable alternate macro syntax
1471 * D:: -D for compatibility
1472 * f:: -f to work faster
1473 * I:: -I for .include search path
1474 @ifclear DIFF-TBL-KLUGE
1475 * K:: -K for compatibility
1477 @ifset DIFF-TBL-KLUGE
1478 * K:: -K for difference tables
1481 * L:: -L to retain local labels
1482 * listing:: --listing-XXX to configure listing output
1483 * M:: -M or --mri to assemble in MRI compatibility mode
1484 * MD:: --MD for dependency tracking
1485 * o:: -o to name the object file
1486 * R:: -R to join data and text sections
1487 * statistics:: --statistics to see statistics about assembly
1488 * traditional-format:: --traditional-format for compatible output
1489 * v:: -v to announce version
1490 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1491 * Z:: -Z to make object file even after errors
1495 @section Enable Listings: @option{-a[cdhlns]}
1504 @cindex listings, enabling
1505 @cindex assembly listings, enabling
1507 These options enable listing output from the assembler. By itself,
1508 @samp{-a} requests high-level, assembly, and symbols listing.
1509 You can use other letters to select specific options for the list:
1510 @samp{-ah} requests a high-level language listing,
1511 @samp{-al} requests an output-program assembly listing, and
1512 @samp{-as} requests a symbol table listing.
1513 High-level listings require that a compiler debugging option like
1514 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1517 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1518 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1519 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1520 omitted from the listing.
1522 Use the @samp{-ad} option to omit debugging directives from the
1525 Once you have specified one of these options, you can further control
1526 listing output and its appearance using the directives @code{.list},
1527 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1529 The @samp{-an} option turns off all forms processing.
1530 If you do not request listing output with one of the @samp{-a} options, the
1531 listing-control directives have no effect.
1533 The letters after @samp{-a} may be combined into one option,
1534 @emph{e.g.}, @samp{-aln}.
1536 Note if the assembler source is coming from the standard input (eg because it
1537 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1538 is being used) then the listing will not contain any comments or preprocessor
1539 directives. This is because the listing code buffers input source lines from
1540 stdin only after they have been preprocessed by the assembler. This reduces
1541 memory usage and makes the code more efficient.
1544 @section @option{--alternate}
1547 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1550 @section @option{-D}
1553 This option has no effect whatsoever, but it is accepted to make it more
1554 likely that scripts written for other assemblers also work with
1555 @command{@value{AS}}.
1558 @section Work Faster: @option{-f}
1561 @cindex trusted compiler
1562 @cindex faster processing (@option{-f})
1563 @samp{-f} should only be used when assembling programs written by a
1564 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1565 and comment preprocessing on
1566 the input file(s) before assembling them. @xref{Preprocessing,
1570 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1571 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1576 @section @code{.include} Search Path: @option{-I} @var{path}
1578 @kindex -I @var{path}
1579 @cindex paths for @code{.include}
1580 @cindex search path for @code{.include}
1581 @cindex @code{include} directive search path
1582 Use this option to add a @var{path} to the list of directories
1583 @command{@value{AS}} searches for files specified in @code{.include}
1584 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1585 many times as necessary to include a variety of paths. The current
1586 working directory is always searched first; after that, @command{@value{AS}}
1587 searches any @samp{-I} directories in the same order as they were
1588 specified (left to right) on the command line.
1591 @section Difference Tables: @option{-K}
1594 @ifclear DIFF-TBL-KLUGE
1595 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1596 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1597 where it can be used to warn when the assembler alters the machine code
1598 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1599 family does not have the addressing limitations that sometimes lead to this
1600 alteration on other platforms.
1603 @ifset DIFF-TBL-KLUGE
1604 @cindex difference tables, warning
1605 @cindex warning for altered difference tables
1606 @command{@value{AS}} sometimes alters the code emitted for directives of the form
1607 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1608 You can use the @samp{-K} option if you want a warning issued when this
1613 @section Include Local Labels: @option{-L}
1616 @cindex local labels, retaining in output
1617 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1618 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1619 debugging, because they are intended for the use of programs (like
1620 compilers) that compose assembler programs, not for your notice.
1621 Normally both @command{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1622 normally debug with them.
1624 This option tells @command{@value{AS}} to retain those @samp{L@dots{}} symbols
1625 in the object file. Usually if you do this you also tell the linker
1626 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1628 By default, a local label is any label beginning with @samp{L}, but each
1629 target is allowed to redefine the local label prefix.
1631 On the HPPA local labels begin with @samp{L$}.
1635 @section Configuring listing output: @option{--listing}
1637 The listing feature of the assembler can be enabled via the command line switch
1638 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1639 hex dump of the corresponding locations in the output object file, and displays
1640 them as a listing file. The format of this listing can be controlled by pseudo
1641 ops inside the assembler source (@pxref{List} @pxref{Title} @pxref{Sbttl}
1642 @pxref{Psize} @pxref{Eject}) and also by the following switches:
1645 @item --listing-lhs-width=@samp{number}
1646 @kindex --listing-lhs-width
1647 @cindex Width of first line disassembly output
1648 Sets the maximum width, in words, of the first line of the hex byte dump. This
1649 dump appears on the left hand side of the listing output.
1651 @item --listing-lhs-width2=@samp{number}
1652 @kindex --listing-lhs-width2
1653 @cindex Width of continuation lines of disassembly output
1654 Sets the maximum width, in words, of any further lines of the hex byte dump for
1655 a given input source line. If this value is not specified, it defaults to being
1656 the same as the value specified for @samp{--listing-lhs-width}. If neither
1657 switch is used the default is to one.
1659 @item --listing-rhs-width=@samp{number}
1660 @kindex --listing-rhs-width
1661 @cindex Width of source line output
1662 Sets the maximum width, in characters, of the source line that is displayed
1663 alongside the hex dump. The default value for this parameter is 100. The
1664 source line is displayed on the right hand side of the listing output.
1666 @item --listing-cont-lines=@samp{number}
1667 @kindex --listing-cont-lines
1668 @cindex Maximum number of continuation lines
1669 Sets the maximum number of continuation lines of hex dump that will be
1670 displayed for a given single line of source input. The default value is 4.
1674 @section Assemble in MRI Compatibility Mode: @option{-M}
1677 @cindex MRI compatibility mode
1678 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1679 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1680 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1681 configured target) assembler from Microtec Research. The exact nature of the
1682 MRI syntax will not be documented here; see the MRI manuals for more
1683 information. Note in particular that the handling of macros and macro
1684 arguments is somewhat different. The purpose of this option is to permit
1685 assembling existing MRI assembler code using @command{@value{AS}}.
1687 The MRI compatibility is not complete. Certain operations of the MRI assembler
1688 depend upon its object file format, and can not be supported using other object
1689 file formats. Supporting these would require enhancing each object file format
1690 individually. These are:
1693 @item global symbols in common section
1695 The m68k MRI assembler supports common sections which are merged by the linker.
1696 Other object file formats do not support this. @command{@value{AS}} handles
1697 common sections by treating them as a single common symbol. It permits local
1698 symbols to be defined within a common section, but it can not support global
1699 symbols, since it has no way to describe them.
1701 @item complex relocations
1703 The MRI assemblers support relocations against a negated section address, and
1704 relocations which combine the start addresses of two or more sections. These
1705 are not support by other object file formats.
1707 @item @code{END} pseudo-op specifying start address
1709 The MRI @code{END} pseudo-op permits the specification of a start address.
1710 This is not supported by other object file formats. The start address may
1711 instead be specified using the @option{-e} option to the linker, or in a linker
1714 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1716 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1717 name to the output file. This is not supported by other object file formats.
1719 @item @code{ORG} pseudo-op
1721 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1722 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1723 which changes the location within the current section. Absolute sections are
1724 not supported by other object file formats. The address of a section may be
1725 assigned within a linker script.
1728 There are some other features of the MRI assembler which are not supported by
1729 @command{@value{AS}}, typically either because they are difficult or because they
1730 seem of little consequence. Some of these may be supported in future releases.
1734 @item EBCDIC strings
1736 EBCDIC strings are not supported.
1738 @item packed binary coded decimal
1740 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1741 and @code{DCB.P} pseudo-ops are not supported.
1743 @item @code{FEQU} pseudo-op
1745 The m68k @code{FEQU} pseudo-op is not supported.
1747 @item @code{NOOBJ} pseudo-op
1749 The m68k @code{NOOBJ} pseudo-op is not supported.
1751 @item @code{OPT} branch control options
1753 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1754 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1755 relaxes all branches, whether forward or backward, to an appropriate size, so
1756 these options serve no purpose.
1758 @item @code{OPT} list control options
1760 The following m68k @code{OPT} list control options are ignored: @code{C},
1761 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1762 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1764 @item other @code{OPT} options
1766 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1767 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1769 @item @code{OPT} @code{D} option is default
1771 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1772 @code{OPT NOD} may be used to turn it off.
1774 @item @code{XREF} pseudo-op.
1776 The m68k @code{XREF} pseudo-op is ignored.
1778 @item @code{.debug} pseudo-op
1780 The i960 @code{.debug} pseudo-op is not supported.
1782 @item @code{.extended} pseudo-op
1784 The i960 @code{.extended} pseudo-op is not supported.
1786 @item @code{.list} pseudo-op.
1788 The various options of the i960 @code{.list} pseudo-op are not supported.
1790 @item @code{.optimize} pseudo-op
1792 The i960 @code{.optimize} pseudo-op is not supported.
1794 @item @code{.output} pseudo-op
1796 The i960 @code{.output} pseudo-op is not supported.
1798 @item @code{.setreal} pseudo-op
1800 The i960 @code{.setreal} pseudo-op is not supported.
1805 @section Dependency Tracking: @option{--MD}
1808 @cindex dependency tracking
1811 @command{@value{AS}} can generate a dependency file for the file it creates. This
1812 file consists of a single rule suitable for @code{make} describing the
1813 dependencies of the main source file.
1815 The rule is written to the file named in its argument.
1817 This feature is used in the automatic updating of makefiles.
1820 @section Name the Object File: @option{-o}
1823 @cindex naming object file
1824 @cindex object file name
1825 There is always one object file output when you run @command{@value{AS}}. By
1826 default it has the name
1829 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1843 You use this option (which takes exactly one filename) to give the
1844 object file a different name.
1846 Whatever the object file is called, @command{@value{AS}} overwrites any
1847 existing file of the same name.
1850 @section Join Data and Text Sections: @option{-R}
1853 @cindex data and text sections, joining
1854 @cindex text and data sections, joining
1855 @cindex joining text and data sections
1856 @cindex merging text and data sections
1857 @option{-R} tells @command{@value{AS}} to write the object file as if all
1858 data-section data lives in the text section. This is only done at
1859 the very last moment: your binary data are the same, but data
1860 section parts are relocated differently. The data section part of
1861 your object file is zero bytes long because all its bytes are
1862 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1864 When you specify @option{-R} it would be possible to generate shorter
1865 address displacements (because we do not have to cross between text and
1866 data section). We refrain from doing this simply for compatibility with
1867 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
1870 When @command{@value{AS}} is configured for COFF or ELF output,
1871 this option is only useful if you use sections named @samp{.text} and
1876 @option{-R} is not supported for any of the HPPA targets. Using
1877 @option{-R} generates a warning from @command{@value{AS}}.
1881 @section Display Assembly Statistics: @option{--statistics}
1883 @kindex --statistics
1884 @cindex statistics, about assembly
1885 @cindex time, total for assembly
1886 @cindex space used, maximum for assembly
1887 Use @samp{--statistics} to display two statistics about the resources used by
1888 @command{@value{AS}}: the maximum amount of space allocated during the assembly
1889 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1892 @node traditional-format
1893 @section Compatible Output: @option{--traditional-format}
1895 @kindex --traditional-format
1896 For some targets, the output of @command{@value{AS}} is different in some ways
1897 from the output of some existing assembler. This switch requests
1898 @command{@value{AS}} to use the traditional format instead.
1900 For example, it disables the exception frame optimizations which
1901 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
1904 @section Announce Version: @option{-v}
1908 @cindex assembler version
1909 @cindex version of assembler
1910 You can find out what version of as is running by including the
1911 option @samp{-v} (which you can also spell as @samp{-version}) on the
1915 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
1917 @command{@value{AS}} should never give a warning or error message when
1918 assembling compiler output. But programs written by people often
1919 cause @command{@value{AS}} to give a warning that a particular assumption was
1920 made. All such warnings are directed to the standard error file.
1924 @cindex suppressing warnings
1925 @cindex warnings, suppressing
1926 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
1927 This only affects the warning messages: it does not change any particular of
1928 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
1931 @kindex --fatal-warnings
1932 @cindex errors, caused by warnings
1933 @cindex warnings, causing error
1934 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
1935 files that generate warnings to be in error.
1938 @cindex warnings, switching on
1939 You can switch these options off again by specifying @option{--warn}, which
1940 causes warnings to be output as usual.
1943 @section Generate Object File in Spite of Errors: @option{-Z}
1944 @cindex object file, after errors
1945 @cindex errors, continuing after
1946 After an error message, @command{@value{AS}} normally produces no output. If for
1947 some reason you are interested in object file output even after
1948 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
1949 option. If there are any errors, @command{@value{AS}} continues anyways, and
1950 writes an object file after a final warning message of the form @samp{@var{n}
1951 errors, @var{m} warnings, generating bad object file.}
1956 @cindex machine-independent syntax
1957 @cindex syntax, machine-independent
1958 This chapter describes the machine-independent syntax allowed in a
1959 source file. @command{@value{AS}} syntax is similar to what many other
1960 assemblers use; it is inspired by the BSD 4.2
1965 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
1969 * Preprocessing:: Preprocessing
1970 * Whitespace:: Whitespace
1971 * Comments:: Comments
1972 * Symbol Intro:: Symbols
1973 * Statements:: Statements
1974 * Constants:: Constants
1978 @section Preprocessing
1980 @cindex preprocessing
1981 The @command{@value{AS}} internal preprocessor:
1983 @cindex whitespace, removed by preprocessor
1985 adjusts and removes extra whitespace. It leaves one space or tab before
1986 the keywords on a line, and turns any other whitespace on the line into
1989 @cindex comments, removed by preprocessor
1991 removes all comments, replacing them with a single space, or an
1992 appropriate number of newlines.
1994 @cindex constants, converted by preprocessor
1996 converts character constants into the appropriate numeric values.
1999 It does not do macro processing, include file handling, or
2000 anything else you may get from your C compiler's preprocessor. You can
2001 do include file processing with the @code{.include} directive
2002 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2003 to get other ``CPP'' style preprocessing by giving the input file a
2004 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
2005 Output, gcc.info, Using GNU CC}.
2007 Excess whitespace, comments, and character constants
2008 cannot be used in the portions of the input text that are not
2011 @cindex turning preprocessing on and off
2012 @cindex preprocessing, turning on and off
2015 If the first line of an input file is @code{#NO_APP} or if you use the
2016 @samp{-f} option, whitespace and comments are not removed from the input file.
2017 Within an input file, you can ask for whitespace and comment removal in
2018 specific portions of the by putting a line that says @code{#APP} before the
2019 text that may contain whitespace or comments, and putting a line that says
2020 @code{#NO_APP} after this text. This feature is mainly intend to support
2021 @code{asm} statements in compilers whose output is otherwise free of comments
2028 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2029 Whitespace is used to separate symbols, and to make programs neater for
2030 people to read. Unless within character constants
2031 (@pxref{Characters,,Character Constants}), any whitespace means the same
2032 as exactly one space.
2038 There are two ways of rendering comments to @command{@value{AS}}. In both
2039 cases the comment is equivalent to one space.
2041 Anything from @samp{/*} through the next @samp{*/} is a comment.
2042 This means you may not nest these comments.
2046 The only way to include a newline ('\n') in a comment
2047 is to use this sort of comment.
2050 /* This sort of comment does not nest. */
2053 @cindex line comment character
2054 Anything from the @dfn{line comment} character to the next newline
2055 is considered a comment and is ignored. The line comment character is
2057 @samp{;} for the AMD 29K family;
2060 @samp{;} on the ARC;
2063 @samp{@@} on the ARM;
2066 @samp{;} for the H8/300 family;
2069 @samp{!} for the H8/500 family;
2072 @samp{;} for the HPPA;
2075 @samp{#} on the i386 and x86-64;
2078 @samp{#} on the i960;
2081 @samp{;} for the PDP-11;
2084 @samp{;} for picoJava;
2087 @samp{#} for Motorola PowerPC;
2090 @samp{!} for the Renesas / SuperH SH;
2093 @samp{!} on the SPARC;
2096 @samp{#} on the ip2k;
2099 @samp{#} on the m32r;
2102 @samp{|} on the 680x0;
2105 @samp{#} on the 68HC11 and 68HC12;
2108 @samp{;} on the M880x0;
2111 @samp{#} on the Vax;
2114 @samp{!} for the Z8000;
2117 @samp{#} on the V850;
2120 @samp{#} for Xtensa systems;
2122 see @ref{Machine Dependencies}. @refill
2123 @c FIXME What about i860?
2126 On some machines there are two different line comment characters. One
2127 character only begins a comment if it is the first non-whitespace character on
2128 a line, while the other always begins a comment.
2132 The V850 assembler also supports a double dash as starting a comment that
2133 extends to the end of the line.
2139 @cindex lines starting with @code{#}
2140 @cindex logical line numbers
2141 To be compatible with past assemblers, lines that begin with @samp{#} have a
2142 special interpretation. Following the @samp{#} should be an absolute
2143 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2144 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
2145 new logical file name. The rest of the line, if any, should be whitespace.
2147 If the first non-whitespace characters on the line are not numeric,
2148 the line is ignored. (Just like a comment.)
2151 # This is an ordinary comment.
2152 # 42-6 "new_file_name" # New logical file name
2153 # This is logical line # 36.
2155 This feature is deprecated, and may disappear from future versions
2156 of @command{@value{AS}}.
2161 @cindex characters used in symbols
2162 @ifclear SPECIAL-SYMS
2163 A @dfn{symbol} is one or more characters chosen from the set of all
2164 letters (both upper and lower case), digits and the three characters
2170 A @dfn{symbol} is one or more characters chosen from the set of all
2171 letters (both upper and lower case), digits and the three characters
2172 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2178 On most machines, you can also use @code{$} in symbol names; exceptions
2179 are noted in @ref{Machine Dependencies}.
2181 No symbol may begin with a digit. Case is significant.
2182 There is no length limit: all characters are significant. Symbols are
2183 delimited by characters not in that set, or by the beginning of a file
2184 (since the source program must end with a newline, the end of a file is
2185 not a possible symbol delimiter). @xref{Symbols}.
2186 @cindex length of symbols
2191 @cindex statements, structure of
2192 @cindex line separator character
2193 @cindex statement separator character
2195 @ifclear abnormal-separator
2196 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2197 semicolon (@samp{;}). The newline or semicolon is considered part of
2198 the preceding statement. Newlines and semicolons within character
2199 constants are an exception: they do not end statements.
2201 @ifset abnormal-separator
2203 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
2204 sign (@samp{@@}). The newline or at sign is considered part of the
2205 preceding statement. Newlines and at signs within character constants
2206 are an exception: they do not end statements.
2209 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2210 point (@samp{!}). The newline or exclamation point is considered part of the
2211 preceding statement. Newlines and exclamation points within character
2212 constants are an exception: they do not end statements.
2215 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2216 H8/300) a dollar sign (@samp{$}); or (for the
2219 (@samp{;}). The newline or separator character is considered part of
2220 the preceding statement. Newlines and separators within character
2221 constants are an exception: they do not end statements.
2226 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2227 separator character. (The line separator is usually @samp{;}, unless
2228 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
2229 newline or separator character is considered part of the preceding
2230 statement. Newlines and separators within character constants are an
2231 exception: they do not end statements.
2234 @cindex newline, required at file end
2235 @cindex EOF, newline must precede
2236 It is an error to end any statement with end-of-file: the last
2237 character of any input file should be a newline.@refill
2239 An empty statement is allowed, and may include whitespace. It is ignored.
2241 @cindex instructions and directives
2242 @cindex directives and instructions
2243 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2244 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2246 A statement begins with zero or more labels, optionally followed by a
2247 key symbol which determines what kind of statement it is. The key
2248 symbol determines the syntax of the rest of the statement. If the
2249 symbol begins with a dot @samp{.} then the statement is an assembler
2250 directive: typically valid for any computer. If the symbol begins with
2251 a letter the statement is an assembly language @dfn{instruction}: it
2252 assembles into a machine language instruction.
2254 Different versions of @command{@value{AS}} for different computers
2255 recognize different instructions. In fact, the same symbol may
2256 represent a different instruction in a different computer's assembly
2260 @cindex @code{:} (label)
2261 @cindex label (@code{:})
2262 A label is a symbol immediately followed by a colon (@code{:}).
2263 Whitespace before a label or after a colon is permitted, but you may not
2264 have whitespace between a label's symbol and its colon. @xref{Labels}.
2267 For HPPA targets, labels need not be immediately followed by a colon, but
2268 the definition of a label must begin in column zero. This also implies that
2269 only one label may be defined on each line.
2273 label: .directive followed by something
2274 another_label: # This is an empty statement.
2275 instruction operand_1, operand_2, @dots{}
2282 A constant is a number, written so that its value is known by
2283 inspection, without knowing any context. Like this:
2286 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2287 .ascii "Ring the bell\7" # A string constant.
2288 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2289 .float 0f-314159265358979323846264338327\
2290 95028841971.693993751E-40 # - pi, a flonum.
2295 * Characters:: Character Constants
2296 * Numbers:: Number Constants
2300 @subsection Character Constants
2302 @cindex character constants
2303 @cindex constants, character
2304 There are two kinds of character constants. A @dfn{character} stands
2305 for one character in one byte and its value may be used in
2306 numeric expressions. String constants (properly called string
2307 @emph{literals}) are potentially many bytes and their values may not be
2308 used in arithmetic expressions.
2312 * Chars:: Characters
2316 @subsubsection Strings
2318 @cindex string constants
2319 @cindex constants, string
2320 A @dfn{string} is written between double-quotes. It may contain
2321 double-quotes or null characters. The way to get special characters
2322 into a string is to @dfn{escape} these characters: precede them with
2323 a backslash @samp{\} character. For example @samp{\\} represents
2324 one backslash: the first @code{\} is an escape which tells
2325 @command{@value{AS}} to interpret the second character literally as a backslash
2326 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2327 escape character). The complete list of escapes follows.
2329 @cindex escape codes, character
2330 @cindex character escape codes
2333 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2335 @cindex @code{\b} (backspace character)
2336 @cindex backspace (@code{\b})
2338 Mnemonic for backspace; for ASCII this is octal code 010.
2341 @c Mnemonic for EOText; for ASCII this is octal code 004.
2343 @cindex @code{\f} (formfeed character)
2344 @cindex formfeed (@code{\f})
2346 Mnemonic for FormFeed; for ASCII this is octal code 014.
2348 @cindex @code{\n} (newline character)
2349 @cindex newline (@code{\n})
2351 Mnemonic for newline; for ASCII this is octal code 012.
2354 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2356 @cindex @code{\r} (carriage return character)
2357 @cindex carriage return (@code{\r})
2359 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2362 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2363 @c other assemblers.
2365 @cindex @code{\t} (tab)
2366 @cindex tab (@code{\t})
2368 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2371 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2372 @c @item \x @var{digit} @var{digit} @var{digit}
2373 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2375 @cindex @code{\@var{ddd}} (octal character code)
2376 @cindex octal character code (@code{\@var{ddd}})
2377 @item \ @var{digit} @var{digit} @var{digit}
2378 An octal character code. The numeric code is 3 octal digits.
2379 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2380 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2382 @cindex @code{\@var{xd...}} (hex character code)
2383 @cindex hex character code (@code{\@var{xd...}})
2384 @item \@code{x} @var{hex-digits...}
2385 A hex character code. All trailing hex digits are combined. Either upper or
2386 lower case @code{x} works.
2388 @cindex @code{\\} (@samp{\} character)
2389 @cindex backslash (@code{\\})
2391 Represents one @samp{\} character.
2394 @c Represents one @samp{'} (accent acute) character.
2395 @c This is needed in single character literals
2396 @c (@xref{Characters,,Character Constants}.) to represent
2399 @cindex @code{\"} (doublequote character)
2400 @cindex doublequote (@code{\"})
2402 Represents one @samp{"} character. Needed in strings to represent
2403 this character, because an unescaped @samp{"} would end the string.
2405 @item \ @var{anything-else}
2406 Any other character when escaped by @kbd{\} gives a warning, but
2407 assembles as if the @samp{\} was not present. The idea is that if
2408 you used an escape sequence you clearly didn't want the literal
2409 interpretation of the following character. However @command{@value{AS}} has no
2410 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2411 code and warns you of the fact.
2414 Which characters are escapable, and what those escapes represent,
2415 varies widely among assemblers. The current set is what we think
2416 the BSD 4.2 assembler recognizes, and is a subset of what most C
2417 compilers recognize. If you are in doubt, do not use an escape
2421 @subsubsection Characters
2423 @cindex single character constant
2424 @cindex character, single
2425 @cindex constant, single character
2426 A single character may be written as a single quote immediately
2427 followed by that character. The same escapes apply to characters as
2428 to strings. So if you want to write the character backslash, you
2429 must write @kbd{'\\} where the first @code{\} escapes the second
2430 @code{\}. As you can see, the quote is an acute accent, not a
2431 grave accent. A newline
2433 @ifclear abnormal-separator
2434 (or semicolon @samp{;})
2436 @ifset abnormal-separator
2438 (or at sign @samp{@@})
2441 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2442 Renesas SH or H8/500)
2446 immediately following an acute accent is taken as a literal character
2447 and does not count as the end of a statement. The value of a character
2448 constant in a numeric expression is the machine's byte-wide code for
2449 that character. @command{@value{AS}} assumes your character code is ASCII:
2450 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2453 @subsection Number Constants
2455 @cindex constants, number
2456 @cindex number constants
2457 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2458 are stored in the target machine. @emph{Integers} are numbers that
2459 would fit into an @code{int} in the C language. @emph{Bignums} are
2460 integers, but they are stored in more than 32 bits. @emph{Flonums}
2461 are floating point numbers, described below.
2464 * Integers:: Integers
2469 * Bit Fields:: Bit Fields
2475 @subsubsection Integers
2477 @cindex constants, integer
2479 @cindex binary integers
2480 @cindex integers, binary
2481 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2482 the binary digits @samp{01}.
2484 @cindex octal integers
2485 @cindex integers, octal
2486 An octal integer is @samp{0} followed by zero or more of the octal
2487 digits (@samp{01234567}).
2489 @cindex decimal integers
2490 @cindex integers, decimal
2491 A decimal integer starts with a non-zero digit followed by zero or
2492 more digits (@samp{0123456789}).
2494 @cindex hexadecimal integers
2495 @cindex integers, hexadecimal
2496 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2497 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2499 Integers have the usual values. To denote a negative integer, use
2500 the prefix operator @samp{-} discussed under expressions
2501 (@pxref{Prefix Ops,,Prefix Operators}).
2504 @subsubsection Bignums
2507 @cindex constants, bignum
2508 A @dfn{bignum} has the same syntax and semantics as an integer
2509 except that the number (or its negative) takes more than 32 bits to
2510 represent in binary. The distinction is made because in some places
2511 integers are permitted while bignums are not.
2514 @subsubsection Flonums
2516 @cindex floating point numbers
2517 @cindex constants, floating point
2519 @cindex precision, floating point
2520 A @dfn{flonum} represents a floating point number. The translation is
2521 indirect: a decimal floating point number from the text is converted by
2522 @command{@value{AS}} to a generic binary floating point number of more than
2523 sufficient precision. This generic floating point number is converted
2524 to a particular computer's floating point format (or formats) by a
2525 portion of @command{@value{AS}} specialized to that computer.
2527 A flonum is written by writing (in order)
2532 (@samp{0} is optional on the HPPA.)
2536 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2538 @kbd{e} is recommended. Case is not important.
2540 @c FIXME: verify if flonum syntax really this vague for most cases
2541 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2542 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2545 On the H8/300, H8/500,
2546 Renesas / SuperH SH,
2547 and AMD 29K architectures, the letter must be
2548 one of the letters @samp{DFPRSX} (in upper or lower case).
2550 On the ARC, the letter must be one of the letters @samp{DFRS}
2551 (in upper or lower case).
2553 On the Intel 960 architecture, the letter must be
2554 one of the letters @samp{DFT} (in upper or lower case).
2556 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2560 One of the letters @samp{DFPRSX} (in upper or lower case).
2563 One of the letters @samp{DFRS} (in upper or lower case).
2566 One of the letters @samp{DFPRSX} (in upper or lower case).
2569 The letter @samp{E} (upper case only).
2572 One of the letters @samp{DFT} (in upper or lower case).
2577 An optional sign: either @samp{+} or @samp{-}.
2580 An optional @dfn{integer part}: zero or more decimal digits.
2583 An optional @dfn{fractional part}: @samp{.} followed by zero
2584 or more decimal digits.
2587 An optional exponent, consisting of:
2591 An @samp{E} or @samp{e}.
2592 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2593 @c principle this can perfectly well be different on different targets.
2595 Optional sign: either @samp{+} or @samp{-}.
2597 One or more decimal digits.
2602 At least one of the integer part or the fractional part must be
2603 present. The floating point number has the usual base-10 value.
2605 @command{@value{AS}} does all processing using integers. Flonums are computed
2606 independently of any floating point hardware in the computer running
2607 @command{@value{AS}}.
2611 @c Bit fields are written as a general facility but are also controlled
2612 @c by a conditional-compilation flag---which is as of now (21mar91)
2613 @c turned on only by the i960 config of GAS.
2615 @subsubsection Bit Fields
2618 @cindex constants, bit field
2619 You can also define numeric constants as @dfn{bit fields}.
2620 specify two numbers separated by a colon---
2622 @var{mask}:@var{value}
2625 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2628 The resulting number is then packed
2630 @c this conditional paren in case bit fields turned on elsewhere than 960
2631 (in host-dependent byte order)
2633 into a field whose width depends on which assembler directive has the
2634 bit-field as its argument. Overflow (a result from the bitwise and
2635 requiring more binary digits to represent) is not an error; instead,
2636 more constants are generated, of the specified width, beginning with the
2637 least significant digits.@refill
2639 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2640 @code{.short}, and @code{.word} accept bit-field arguments.
2645 @chapter Sections and Relocation
2650 * Secs Background:: Background
2651 * Ld Sections:: Linker Sections
2652 * As Sections:: Assembler Internal Sections
2653 * Sub-Sections:: Sub-Sections
2657 @node Secs Background
2660 Roughly, a section is a range of addresses, with no gaps; all data
2661 ``in'' those addresses is treated the same for some particular purpose.
2662 For example there may be a ``read only'' section.
2664 @cindex linker, and assembler
2665 @cindex assembler, and linker
2666 The linker @code{@value{LD}} reads many object files (partial programs) and
2667 combines their contents to form a runnable program. When @command{@value{AS}}
2668 emits an object file, the partial program is assumed to start at address 0.
2669 @code{@value{LD}} assigns the final addresses for the partial program, so that
2670 different partial programs do not overlap. This is actually an
2671 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2674 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2675 addresses. These blocks slide to their run-time addresses as rigid
2676 units; their length does not change and neither does the order of bytes
2677 within them. Such a rigid unit is called a @emph{section}. Assigning
2678 run-time addresses to sections is called @dfn{relocation}. It includes
2679 the task of adjusting mentions of object-file addresses so they refer to
2680 the proper run-time addresses.
2682 For the H8/300 and H8/500,
2683 and for the Renesas / SuperH SH,
2684 @command{@value{AS}} pads sections if needed to
2685 ensure they end on a word (sixteen bit) boundary.
2688 @cindex standard assembler sections
2689 An object file written by @command{@value{AS}} has at least three sections, any
2690 of which may be empty. These are named @dfn{text}, @dfn{data} and
2695 When it generates COFF or ELF output,
2697 @command{@value{AS}} can also generate whatever other named sections you specify
2698 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2699 If you do not use any directives that place output in the @samp{.text}
2700 or @samp{.data} sections, these sections still exist, but are empty.
2705 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2707 @command{@value{AS}} can also generate whatever other named sections you
2708 specify using the @samp{.space} and @samp{.subspace} directives. See
2709 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2710 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2711 assembler directives.
2714 Additionally, @command{@value{AS}} uses different names for the standard
2715 text, data, and bss sections when generating SOM output. Program text
2716 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2717 BSS into @samp{$BSS$}.
2721 Within the object file, the text section starts at address @code{0}, the
2722 data section follows, and the bss section follows the data section.
2725 When generating either SOM or ELF output files on the HPPA, the text
2726 section starts at address @code{0}, the data section at address
2727 @code{0x4000000}, and the bss section follows the data section.
2730 To let @code{@value{LD}} know which data changes when the sections are
2731 relocated, and how to change that data, @command{@value{AS}} also writes to the
2732 object file details of the relocation needed. To perform relocation
2733 @code{@value{LD}} must know, each time an address in the object
2737 Where in the object file is the beginning of this reference to
2740 How long (in bytes) is this reference?
2742 Which section does the address refer to? What is the numeric value of
2744 (@var{address}) @minus{} (@var{start-address of section})?
2747 Is the reference to an address ``Program-Counter relative''?
2750 @cindex addresses, format of
2751 @cindex section-relative addressing
2752 In fact, every address @command{@value{AS}} ever uses is expressed as
2754 (@var{section}) + (@var{offset into section})
2757 Further, most expressions @command{@value{AS}} computes have this section-relative
2760 (For some object formats, such as SOM for the HPPA, some expressions are
2761 symbol-relative instead.)
2764 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2765 @var{N} into section @var{secname}.''
2767 Apart from text, data and bss sections you need to know about the
2768 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2769 addresses in the absolute section remain unchanged. For example, address
2770 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2771 @code{@value{LD}}. Although the linker never arranges two partial programs'
2772 data sections with overlapping addresses after linking, @emph{by definition}
2773 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2774 part of a program is always the same address when the program is running as
2775 address @code{@{absolute@ 239@}} in any other part of the program.
2777 The idea of sections is extended to the @dfn{undefined} section. Any
2778 address whose section is unknown at assembly time is by definition
2779 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2780 Since numbers are always defined, the only way to generate an undefined
2781 address is to mention an undefined symbol. A reference to a named
2782 common block would be such a symbol: its value is unknown at assembly
2783 time so it has section @emph{undefined}.
2785 By analogy the word @emph{section} is used to describe groups of sections in
2786 the linked program. @code{@value{LD}} puts all partial programs' text
2787 sections in contiguous addresses in the linked program. It is
2788 customary to refer to the @emph{text section} of a program, meaning all
2789 the addresses of all partial programs' text sections. Likewise for
2790 data and bss sections.
2792 Some sections are manipulated by @code{@value{LD}}; others are invented for
2793 use of @command{@value{AS}} and have no meaning except during assembly.
2796 @section Linker Sections
2797 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2802 @cindex named sections
2803 @cindex sections, named
2804 @item named sections
2807 @cindex text section
2808 @cindex data section
2812 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2813 separate but equal sections. Anything you can say of one section is
2816 When the program is running, however, it is
2817 customary for the text section to be unalterable. The
2818 text section is often shared among processes: it contains
2819 instructions, constants and the like. The data section of a running
2820 program is usually alterable: for example, C variables would be stored
2821 in the data section.
2826 This section contains zeroed bytes when your program begins running. It
2827 is used to hold uninitialized variables or common storage. The length of
2828 each partial program's bss section is important, but because it starts
2829 out containing zeroed bytes there is no need to store explicit zero
2830 bytes in the object file. The bss section was invented to eliminate
2831 those explicit zeros from object files.
2833 @cindex absolute section
2834 @item absolute section
2835 Address 0 of this section is always ``relocated'' to runtime address 0.
2836 This is useful if you want to refer to an address that @code{@value{LD}} must
2837 not change when relocating. In this sense we speak of absolute
2838 addresses being ``unrelocatable'': they do not change during relocation.
2840 @cindex undefined section
2841 @item undefined section
2842 This ``section'' is a catch-all for address references to objects not in
2843 the preceding sections.
2844 @c FIXME: ref to some other doc on obj-file formats could go here.
2847 @cindex relocation example
2848 An idealized example of three relocatable sections follows.
2850 The example uses the traditional section names @samp{.text} and @samp{.data}.
2852 Memory addresses are on the horizontal axis.
2856 @c END TEXI2ROFF-KILL
2859 partial program # 1: |ttttt|dddd|00|
2866 partial program # 2: |TTT|DDD|000|
2869 +--+---+-----+--+----+---+-----+~~
2870 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2871 +--+---+-----+--+----+---+-----+~~
2873 addresses: 0 @dots{}
2880 \line{\it Partial program \#1: \hfil}
2881 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2882 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2884 \line{\it Partial program \#2: \hfil}
2885 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2886 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2888 \line{\it linked program: \hfil}
2889 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2890 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2891 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2892 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2894 \line{\it addresses: \hfil}
2898 @c END TEXI2ROFF-KILL
2901 @section Assembler Internal Sections
2903 @cindex internal assembler sections
2904 @cindex sections in messages, internal
2905 These sections are meant only for the internal use of @command{@value{AS}}. They
2906 have no meaning at run-time. You do not really need to know about these
2907 sections for most purposes; but they can be mentioned in @command{@value{AS}}
2908 warning messages, so it might be helpful to have an idea of their
2909 meanings to @command{@value{AS}}. These sections are used to permit the
2910 value of every expression in your assembly language program to be a
2911 section-relative address.
2914 @cindex assembler internal logic error
2915 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2916 An internal assembler logic error has been found. This means there is a
2917 bug in the assembler.
2919 @cindex expr (internal section)
2921 The assembler stores complex expression internally as combinations of
2922 symbols. When it needs to represent an expression as a symbol, it puts
2923 it in the expr section.
2925 @c FIXME item transfer[t] vector preload
2926 @c FIXME item transfer[t] vector postload
2927 @c FIXME item register
2931 @section Sub-Sections
2933 @cindex numbered subsections
2934 @cindex grouping data
2940 fall into two sections: text and data.
2942 You may have separate groups of
2944 data in named sections
2948 data in named sections
2954 that you want to end up near to each other in the object file, even though they
2955 are not contiguous in the assembler source. @command{@value{AS}} allows you to
2956 use @dfn{subsections} for this purpose. Within each section, there can be
2957 numbered subsections with values from 0 to 8192. Objects assembled into the
2958 same subsection go into the object file together with other objects in the same
2959 subsection. For example, a compiler might want to store constants in the text
2960 section, but might not want to have them interspersed with the program being
2961 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2962 section of code being output, and a @samp{.text 1} before each group of
2963 constants being output.
2965 Subsections are optional. If you do not use subsections, everything
2966 goes in subsection number zero.
2969 Each subsection is zero-padded up to a multiple of four bytes.
2970 (Subsections may be padded a different amount on different flavors
2971 of @command{@value{AS}}.)
2975 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2976 boundary (two bytes).
2977 The same is true on the Renesas SH.
2980 @c FIXME section padding (alignment)?
2981 @c Rich Pixley says padding here depends on target obj code format; that
2982 @c doesn't seem particularly useful to say without further elaboration,
2983 @c so for now I say nothing about it. If this is a generic BFD issue,
2984 @c these paragraphs might need to vanish from this manual, and be
2985 @c discussed in BFD chapter of binutils (or some such).
2988 On the AMD 29K family, no particular padding is added to section or
2989 subsection sizes; @value{AS} forces no alignment on this platform.
2993 Subsections appear in your object file in numeric order, lowest numbered
2994 to highest. (All this to be compatible with other people's assemblers.)
2995 The object file contains no representation of subsections; @code{@value{LD}} and
2996 other programs that manipulate object files see no trace of them.
2997 They just see all your text subsections as a text section, and all your
2998 data subsections as a data section.
3000 To specify which subsection you want subsequent statements assembled
3001 into, use a numeric argument to specify it, in a @samp{.text
3002 @var{expression}} or a @samp{.data @var{expression}} statement.
3005 When generating COFF output, you
3010 can also use an extra subsection
3011 argument with arbitrary named sections: @samp{.section @var{name},
3016 When generating ELF output, you
3021 can also use the @code{.subsection} directive (@pxref{SubSection})
3022 to specify a subsection: @samp{.subsection @var{expression}}.
3024 @var{Expression} should be an absolute expression.
3025 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
3026 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3027 begins in @code{text 0}. For instance:
3029 .text 0 # The default subsection is text 0 anyway.
3030 .ascii "This lives in the first text subsection. *"
3032 .ascii "But this lives in the second text subsection."
3034 .ascii "This lives in the data section,"
3035 .ascii "in the first data subsection."
3037 .ascii "This lives in the first text section,"
3038 .ascii "immediately following the asterisk (*)."
3041 Each section has a @dfn{location counter} incremented by one for every byte
3042 assembled into that section. Because subsections are merely a convenience
3043 restricted to @command{@value{AS}} there is no concept of a subsection location
3044 counter. There is no way to directly manipulate a location counter---but the
3045 @code{.align} directive changes it, and any label definition captures its
3046 current value. The location counter of the section where statements are being
3047 assembled is said to be the @dfn{active} location counter.
3050 @section bss Section
3053 @cindex common variable storage
3054 The bss section is used for local common variable storage.
3055 You may allocate address space in the bss section, but you may
3056 not dictate data to load into it before your program executes. When
3057 your program starts running, all the contents of the bss
3058 section are zeroed bytes.
3060 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3061 @ref{Lcomm,,@code{.lcomm}}.
3063 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3064 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
3067 When assembling for a target which supports multiple sections, such as ELF or
3068 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3069 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3070 section. Typically the section will only contain symbol definitions and
3071 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3078 Symbols are a central concept: the programmer uses symbols to name
3079 things, the linker uses symbols to link, and the debugger uses symbols
3083 @cindex debuggers, and symbol order
3084 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3085 the same order they were declared. This may break some debuggers.
3090 * Setting Symbols:: Giving Symbols Other Values
3091 * Symbol Names:: Symbol Names
3092 * Dot:: The Special Dot Symbol
3093 * Symbol Attributes:: Symbol Attributes
3100 A @dfn{label} is written as a symbol immediately followed by a colon
3101 @samp{:}. The symbol then represents the current value of the
3102 active location counter, and is, for example, a suitable instruction
3103 operand. You are warned if you use the same symbol to represent two
3104 different locations: the first definition overrides any other
3108 On the HPPA, the usual form for a label need not be immediately followed by a
3109 colon, but instead must start in column zero. Only one label may be defined on
3110 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3111 provides a special directive @code{.label} for defining labels more flexibly.
3114 @node Setting Symbols
3115 @section Giving Symbols Other Values
3117 @cindex assigning values to symbols
3118 @cindex symbol values, assigning
3119 A symbol can be given an arbitrary value by writing a symbol, followed
3120 by an equals sign @samp{=}, followed by an expression
3121 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3122 directive. @xref{Set,,@code{.set}}.
3125 @section Symbol Names
3127 @cindex symbol names
3128 @cindex names, symbol
3129 @ifclear SPECIAL-SYMS
3130 Symbol names begin with a letter or with one of @samp{._}. On most
3131 machines, you can also use @code{$} in symbol names; exceptions are
3132 noted in @ref{Machine Dependencies}. That character may be followed by any
3133 string of digits, letters, dollar signs (unless otherwise noted in
3134 @ref{Machine Dependencies}), and underscores.
3137 For the AMD 29K family, @samp{?} is also allowed in the
3138 body of a symbol name, though not at its beginning.
3143 Symbol names begin with a letter or with one of @samp{._}. On the
3144 Renesas SH or the H8/500, you can also use @code{$} in symbol names. That
3145 character may be followed by any string of digits, letters, dollar signs (save
3146 on the H8/300), and underscores.
3150 Case of letters is significant: @code{foo} is a different symbol name
3153 Each symbol has exactly one name. Each name in an assembly language program
3154 refers to exactly one symbol. You may use that symbol name any number of times
3157 @subheading Local Symbol Names
3159 @cindex local symbol names
3160 @cindex symbol names, local
3161 @cindex temporary symbol names
3162 @cindex symbol names, temporary
3163 Local symbols help compilers and programmers use names temporarily.
3164 They create symbols which are guaranteed to be unique over the entire scope of
3165 the input source code and which can be referred to by a simple notation.
3166 To define a local symbol, write a label of the form @samp{@b{N}:} (where @b{N}
3167 represents any positive integer). To refer to the most recent previous
3168 definition of that symbol write @samp{@b{N}b}, using the same number as when
3169 you defined the label. To refer to the next definition of a local label, write
3170 @samp{@b{N}f}--- The @samp{b} stands for``backwards'' and the @samp{f} stands
3173 There is no restriction on how you can use these labels, and you can reuse them
3174 too. So that it is possible to repeatedly define the same local label (using
3175 the same number @samp{@b{N}}), although you can only refer to the most recently
3176 defined local label of that number (for a backwards reference) or the next
3177 definition of a specific local label for a forward reference. It is also worth
3178 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3179 implemented in a slightly more efficient manner than the others.
3190 Which is the equivalent of:
3193 label_1: branch label_3
3194 label_2: branch label_1
3195 label_3: branch label_4
3196 label_4: branch label_3
3199 Local symbol names are only a notational device. They are immediately
3200 transformed into more conventional symbol names before the assembler uses them.
3201 The symbol names stored in the symbol table, appearing in error messages and
3202 optionally emitted to the object file. The names are constructed using these
3207 All local labels begin with @samp{L}. Normally both @command{@value{AS}} and
3208 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
3209 used for symbols you are never intended to see. If you use the
3210 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3211 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3212 you may use them in debugging.
3215 This is the number that was used in the local label definition. So if the
3216 label is written @samp{55:} then the number is @samp{55}.
3219 This unusual character is included so you do not accidentally invent a symbol
3220 of the same name. The character has ASCII value of @samp{\002} (control-B).
3222 @item @emph{ordinal number}
3223 This is a serial number to keep the labels distinct. The first definition of
3224 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3225 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3226 the number @samp{1} and its 15th defintion gets @samp{15} as well.
3229 So for example, the first @code{1:} is named @code{L1@kbd{C-B}1}, the 44th
3230 @code{3:} is named @code{L3@kbd{C-B}44}.
3232 @subheading Dollar Local Labels
3233 @cindex dollar local symbols
3235 @code{@value{AS}} also supports an even more local form of local labels called
3236 dollar labels. These labels go out of scope (ie they become undefined) as soon
3237 as a non-local label is defined. Thus they remain valid for only a small
3238 region of the input source code. Normal local labels, by contrast, remain in
3239 scope for the entire file, or until they are redefined by another occurrence of
3240 the same local label.
3242 Dollar labels are defined in exactly the same way as ordinary local labels,
3243 except that instead of being terminated by a colon, they are terminated by a
3244 dollar sign. eg @samp{@b{55$}}.
3246 They can also be distinguished from ordinary local labels by their transformed
3247 name which uses ASCII character @samp{\001} (control-A) as the magic character
3248 to distinguish them from ordinary labels. Thus the 5th defintion of @samp{6$}
3249 is named @samp{L6@kbd{C-A}5}.
3252 @section The Special Dot Symbol
3254 @cindex dot (symbol)
3255 @cindex @code{.} (symbol)
3256 @cindex current address
3257 @cindex location counter
3258 The special symbol @samp{.} refers to the current address that
3259 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3260 .long .} defines @code{melvin} to contain its own address.
3261 Assigning a value to @code{.} is treated the same as a @code{.org}
3262 directive. Thus, the expression @samp{.=.+4} is the same as saying
3263 @ifclear no-space-dir
3272 @node Symbol Attributes
3273 @section Symbol Attributes
3275 @cindex symbol attributes
3276 @cindex attributes, symbol
3277 Every symbol has, as well as its name, the attributes ``Value'' and
3278 ``Type''. Depending on output format, symbols can also have auxiliary
3281 The detailed definitions are in @file{a.out.h}.
3284 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3285 all these attributes, and probably won't warn you. This makes the
3286 symbol an externally defined symbol, which is generally what you
3290 * Symbol Value:: Value
3291 * Symbol Type:: Type
3294 * a.out Symbols:: Symbol Attributes: @code{a.out}
3298 * a.out Symbols:: Symbol Attributes: @code{a.out}
3301 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3306 * COFF Symbols:: Symbol Attributes for COFF
3309 * SOM Symbols:: Symbol Attributes for SOM
3316 @cindex value of a symbol
3317 @cindex symbol value
3318 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3319 location in the text, data, bss or absolute sections the value is the
3320 number of addresses from the start of that section to the label.
3321 Naturally for text, data and bss sections the value of a symbol changes
3322 as @code{@value{LD}} changes section base addresses during linking. Absolute
3323 symbols' values do not change during linking: that is why they are
3326 The value of an undefined symbol is treated in a special way. If it is
3327 0 then the symbol is not defined in this assembler source file, and
3328 @code{@value{LD}} tries to determine its value from other files linked into the
3329 same program. You make this kind of symbol simply by mentioning a symbol
3330 name without defining it. A non-zero value represents a @code{.comm}
3331 common declaration. The value is how much common storage to reserve, in
3332 bytes (addresses). The symbol refers to the first address of the
3338 @cindex type of a symbol
3340 The type attribute of a symbol contains relocation (section)
3341 information, any flag settings indicating that a symbol is external, and
3342 (optionally), other information for linkers and debuggers. The exact
3343 format depends on the object-code output format in use.
3348 @c The following avoids a "widow" subsection title. @group would be
3349 @c better if it were available outside examples.
3352 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3354 @cindex @code{b.out} symbol attributes
3355 @cindex symbol attributes, @code{b.out}
3356 These symbol attributes appear only when @command{@value{AS}} is configured for
3357 one of the Berkeley-descended object output formats---@code{a.out} or
3363 @subsection Symbol Attributes: @code{a.out}
3365 @cindex @code{a.out} symbol attributes
3366 @cindex symbol attributes, @code{a.out}
3372 @subsection Symbol Attributes: @code{a.out}
3374 @cindex @code{a.out} symbol attributes
3375 @cindex symbol attributes, @code{a.out}
3379 * Symbol Desc:: Descriptor
3380 * Symbol Other:: Other
3384 @subsubsection Descriptor
3386 @cindex descriptor, of @code{a.out} symbol
3387 This is an arbitrary 16-bit value. You may establish a symbol's
3388 descriptor value by using a @code{.desc} statement
3389 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3390 @command{@value{AS}}.
3393 @subsubsection Other
3395 @cindex other attribute, of @code{a.out} symbol
3396 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3401 @subsection Symbol Attributes for COFF
3403 @cindex COFF symbol attributes
3404 @cindex symbol attributes, COFF
3406 The COFF format supports a multitude of auxiliary symbol attributes;
3407 like the primary symbol attributes, they are set between @code{.def} and
3408 @code{.endef} directives.
3410 @subsubsection Primary Attributes
3412 @cindex primary attributes, COFF symbols
3413 The symbol name is set with @code{.def}; the value and type,
3414 respectively, with @code{.val} and @code{.type}.
3416 @subsubsection Auxiliary Attributes
3418 @cindex auxiliary attributes, COFF symbols
3419 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3420 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3421 table information for COFF.
3426 @subsection Symbol Attributes for SOM
3428 @cindex SOM symbol attributes
3429 @cindex symbol attributes, SOM
3431 The SOM format for the HPPA supports a multitude of symbol attributes set with
3432 the @code{.EXPORT} and @code{.IMPORT} directives.
3434 The attributes are described in @cite{HP9000 Series 800 Assembly
3435 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3436 @code{EXPORT} assembler directive documentation.
3440 @chapter Expressions
3444 @cindex numeric values
3445 An @dfn{expression} specifies an address or numeric value.
3446 Whitespace may precede and/or follow an expression.
3448 The result of an expression must be an absolute number, or else an offset into
3449 a particular section. If an expression is not absolute, and there is not
3450 enough information when @command{@value{AS}} sees the expression to know its
3451 section, a second pass over the source program might be necessary to interpret
3452 the expression---but the second pass is currently not implemented.
3453 @command{@value{AS}} aborts with an error message in this situation.
3456 * Empty Exprs:: Empty Expressions
3457 * Integer Exprs:: Integer Expressions
3461 @section Empty Expressions
3463 @cindex empty expressions
3464 @cindex expressions, empty
3465 An empty expression has no value: it is just whitespace or null.
3466 Wherever an absolute expression is required, you may omit the
3467 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3468 is compatible with other assemblers.
3471 @section Integer Expressions
3473 @cindex integer expressions
3474 @cindex expressions, integer
3475 An @dfn{integer expression} is one or more @emph{arguments} delimited
3476 by @emph{operators}.
3479 * Arguments:: Arguments
3480 * Operators:: Operators
3481 * Prefix Ops:: Prefix Operators
3482 * Infix Ops:: Infix Operators
3486 @subsection Arguments
3488 @cindex expression arguments
3489 @cindex arguments in expressions
3490 @cindex operands in expressions
3491 @cindex arithmetic operands
3492 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3493 contexts arguments are sometimes called ``arithmetic operands''. In
3494 this manual, to avoid confusing them with the ``instruction operands'' of
3495 the machine language, we use the term ``argument'' to refer to parts of
3496 expressions only, reserving the word ``operand'' to refer only to machine
3497 instruction operands.
3499 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3500 @var{section} is one of text, data, bss, absolute,
3501 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3504 Numbers are usually integers.
3506 A number can be a flonum or bignum. In this case, you are warned
3507 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3508 these 32 bits are an integer. You may write integer-manipulating
3509 instructions that act on exotic constants, compatible with other
3512 @cindex subexpressions
3513 Subexpressions are a left parenthesis @samp{(} followed by an integer
3514 expression, followed by a right parenthesis @samp{)}; or a prefix
3515 operator followed by an argument.
3518 @subsection Operators
3520 @cindex operators, in expressions
3521 @cindex arithmetic functions
3522 @cindex functions, in expressions
3523 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3524 operators are followed by an argument. Infix operators appear
3525 between their arguments. Operators may be preceded and/or followed by
3529 @subsection Prefix Operator
3531 @cindex prefix operators
3532 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3533 one argument, which must be absolute.
3535 @c the tex/end tex stuff surrounding this small table is meant to make
3536 @c it align, on the printed page, with the similar table in the next
3537 @c section (which is inside an enumerate).
3539 \global\advance\leftskip by \itemindent
3544 @dfn{Negation}. Two's complement negation.
3546 @dfn{Complementation}. Bitwise not.
3550 \global\advance\leftskip by -\itemindent
3554 @subsection Infix Operators
3556 @cindex infix operators
3557 @cindex operators, permitted arguments
3558 @dfn{Infix operators} take two arguments, one on either side. Operators
3559 have precedence, but operations with equal precedence are performed left
3560 to right. Apart from @code{+} or @option{-}, both arguments must be
3561 absolute, and the result is absolute.
3564 @cindex operator precedence
3565 @cindex precedence of operators
3572 @dfn{Multiplication}.
3575 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3582 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3586 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3590 Intermediate precedence
3595 @dfn{Bitwise Inclusive Or}.
3601 @dfn{Bitwise Exclusive Or}.
3604 @dfn{Bitwise Or Not}.
3611 @cindex addition, permitted arguments
3612 @cindex plus, permitted arguments
3613 @cindex arguments for addition
3615 @dfn{Addition}. If either argument is absolute, the result has the section of
3616 the other argument. You may not add together arguments from different
3619 @cindex subtraction, permitted arguments
3620 @cindex minus, permitted arguments
3621 @cindex arguments for subtraction
3623 @dfn{Subtraction}. If the right argument is absolute, the
3624 result has the section of the left argument.
3625 If both arguments are in the same section, the result is absolute.
3626 You may not subtract arguments from different sections.
3627 @c FIXME is there still something useful to say about undefined - undefined ?
3629 @cindex comparison expressions
3630 @cindex expressions, comparison
3634 @dfn{Is Not Equal To}
3638 @dfn{Is Greater Than}
3640 @dfn{Is Greater Than Or Equal To}
3642 @dfn{Is Less Than Or Equal To}
3644 The comparison operators can be used as infix operators. A true results has a
3645 value of -1 whereas a false result has a value of 0. Note, these operators
3646 perform signed comparisons.
3649 @item Lowest Precedence
3658 These two logical operations can be used to combine the results of sub
3659 expressions. Note, unlike the comparison operators a true result returns a
3660 value of 1 but a false results does still return 0. Also note that the logical
3661 or operator has a slightly lower precedence than logical and.
3666 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3667 address; you can only have a defined section in one of the two arguments.
3670 @chapter Assembler Directives
3672 @cindex directives, machine independent
3673 @cindex pseudo-ops, machine independent
3674 @cindex machine independent directives
3675 All assembler directives have names that begin with a period (@samp{.}).
3676 The rest of the name is letters, usually in lower case.
3678 This chapter discusses directives that are available regardless of the
3679 target machine configuration for the @sc{gnu} assembler.
3681 Some machine configurations provide additional directives.
3682 @xref{Machine Dependencies}.
3685 @ifset machine-directives
3686 @xref{Machine Dependencies} for additional directives.
3691 * Abort:: @code{.abort}
3693 * ABORT:: @code{.ABORT}
3696 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3697 * Altmacro:: @code{.altmacro}
3698 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3699 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3700 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3701 * Byte:: @code{.byte @var{expressions}}
3702 * Comm:: @code{.comm @var{symbol} , @var{length} }
3704 * CFI directives:: @code{.cfi_startproc}, @code{.cfi_endproc}, etc.
3706 * Data:: @code{.data @var{subsection}}
3708 * Def:: @code{.def @var{name}}
3711 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3717 * Double:: @code{.double @var{flonums}}
3718 * Eject:: @code{.eject}
3719 * Else:: @code{.else}
3720 * Elseif:: @code{.elseif}
3723 * Endef:: @code{.endef}
3726 * Endfunc:: @code{.endfunc}
3727 * Endif:: @code{.endif}
3728 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3729 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3731 * Error:: @code{.error @var{string}}
3732 * Exitm:: @code{.exitm}
3733 * Extern:: @code{.extern}
3734 * Fail:: @code{.fail}
3735 @ifclear no-file-dir
3736 * File:: @code{.file @var{string}}
3739 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3740 * Float:: @code{.float @var{flonums}}
3741 * Func:: @code{.func}
3742 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3744 * Hidden:: @code{.hidden @var{names}}
3747 * hword:: @code{.hword @var{expressions}}
3748 * Ident:: @code{.ident}
3749 * If:: @code{.if @var{absolute expression}}
3750 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3751 * Include:: @code{.include "@var{file}"}
3752 * Int:: @code{.int @var{expressions}}
3754 * Internal:: @code{.internal @var{names}}
3757 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3758 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3759 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3760 * Lflags:: @code{.lflags}
3761 @ifclear no-line-dir
3762 * Line:: @code{.line @var{line-number}}
3765 * Ln:: @code{.ln @var{line-number}}
3766 * Linkonce:: @code{.linkonce [@var{type}]}
3767 * List:: @code{.list}
3768 * Long:: @code{.long @var{expressions}}
3770 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3773 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3774 * MRI:: @code{.mri @var{val}}
3775 * Noaltmacro:: @code{.noaltmacro}
3776 * Nolist:: @code{.nolist}
3777 * Octa:: @code{.octa @var{bignums}}
3778 * Org:: @code{.org @var{new-lc} , @var{fill}}
3779 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3781 * PopSection:: @code{.popsection}
3782 * Previous:: @code{.previous}
3785 * Print:: @code{.print @var{string}}
3787 * Protected:: @code{.protected @var{names}}
3790 * Psize:: @code{.psize @var{lines}, @var{columns}}
3791 * Purgem:: @code{.purgem @var{name}}
3793 * PushSection:: @code{.pushsection @var{name}}
3796 * Quad:: @code{.quad @var{bignums}}
3797 * Rept:: @code{.rept @var{count}}
3798 * Sbttl:: @code{.sbttl "@var{subheading}"}
3800 * Scl:: @code{.scl @var{class}}
3803 * Section:: @code{.section @var{name}}
3806 * Set:: @code{.set @var{symbol}, @var{expression}}
3807 * Short:: @code{.short @var{expressions}}
3808 * Single:: @code{.single @var{flonums}}
3810 * Size:: @code{.size [@var{name} , @var{expression}]}
3813 * Skip:: @code{.skip @var{size} , @var{fill}}
3814 * Sleb128:: @code{.sleb128 @var{expressions}}
3815 * Space:: @code{.space @var{size} , @var{fill}}
3817 * Stab:: @code{.stabd, .stabn, .stabs}
3820 * String:: @code{.string "@var{str}"}
3821 * Struct:: @code{.struct @var{expression}}
3823 * SubSection:: @code{.subsection}
3824 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3828 * Tag:: @code{.tag @var{structname}}
3831 * Text:: @code{.text @var{subsection}}
3832 * Title:: @code{.title "@var{heading}"}
3834 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3837 * Uleb128:: @code{.uleb128 @var{expressions}}
3839 * Val:: @code{.val @var{addr}}
3843 * Version:: @code{.version "@var{string}"}
3844 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3845 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3848 * Warning:: @code{.warning @var{string}}
3849 * Weak:: @code{.weak @var{names}}
3850 * Word:: @code{.word @var{expressions}}
3851 * Deprecated:: Deprecated Directives
3855 @section @code{.abort}
3857 @cindex @code{abort} directive
3858 @cindex stopping the assembly
3859 This directive stops the assembly immediately. It is for
3860 compatibility with other assemblers. The original idea was that the
3861 assembly language source would be piped into the assembler. If the sender
3862 of the source quit, it could use this directive tells @command{@value{AS}} to
3863 quit also. One day @code{.abort} will not be supported.
3867 @section @code{.ABORT}
3869 @cindex @code{ABORT} directive
3870 When producing COFF output, @command{@value{AS}} accepts this directive as a
3871 synonym for @samp{.abort}.
3874 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
3880 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3882 @cindex padding the location counter
3883 @cindex @code{align} directive
3884 Pad the location counter (in the current subsection) to a particular storage
3885 boundary. The first expression (which must be absolute) is the alignment
3886 required, as described below.
3888 The second expression (also absolute) gives the fill value to be stored in the
3889 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3890 padding bytes are normally zero. However, on some systems, if the section is
3891 marked as containing code and the fill value is omitted, the space is filled
3892 with no-op instructions.
3894 The third expression is also absolute, and is also optional. If it is present,
3895 it is the maximum number of bytes that should be skipped by this alignment
3896 directive. If doing the alignment would require skipping more bytes than the
3897 specified maximum, then the alignment is not done at all. You can omit the
3898 fill value (the second argument) entirely by simply using two commas after the
3899 required alignment; this can be useful if you want the alignment to be filled
3900 with no-op instructions when appropriate.
3902 The way the required alignment is specified varies from system to system.
3903 For the a29k, arc, hppa, i386 using ELF, i860, iq2000, m68k, m88k, or32,
3904 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
3905 alignment request in bytes. For example @samp{.align 8} advances
3906 the location counter until it is a multiple of 8. If the location counter
3907 is already a multiple of 8, no change is needed. For the tic54x, the
3908 first expression is the alignment request in words.
3910 For other systems, including the i386 using a.out format, and the arm and
3911 strongarm, it is the
3912 number of low-order zero bits the location counter must have after
3913 advancement. For example @samp{.align 3} advances the location
3914 counter until it a multiple of 8. If the location counter is already a
3915 multiple of 8, no change is needed.
3917 This inconsistency is due to the different behaviors of the various
3918 native assemblers for these systems which GAS must emulate.
3919 GAS also provides @code{.balign} and @code{.p2align} directives,
3920 described later, which have a consistent behavior across all
3921 architectures (but are specific to GAS).
3924 @section @code{.ascii "@var{string}"}@dots{}
3926 @cindex @code{ascii} directive
3927 @cindex string literals
3928 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3929 separated by commas. It assembles each string (with no automatic
3930 trailing zero byte) into consecutive addresses.
3933 @section @code{.asciz "@var{string}"}@dots{}
3935 @cindex @code{asciz} directive
3936 @cindex zero-terminated strings
3937 @cindex null-terminated strings
3938 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3939 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3942 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3944 @cindex padding the location counter given number of bytes
3945 @cindex @code{balign} directive
3946 Pad the location counter (in the current subsection) to a particular
3947 storage boundary. The first expression (which must be absolute) is the
3948 alignment request in bytes. For example @samp{.balign 8} advances
3949 the location counter until it is a multiple of 8. If the location counter
3950 is already a multiple of 8, no change is needed.
3952 The second expression (also absolute) gives the fill value to be stored in the
3953 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3954 padding bytes are normally zero. However, on some systems, if the section is
3955 marked as containing code and the fill value is omitted, the space is filled
3956 with no-op instructions.
3958 The third expression is also absolute, and is also optional. If it is present,
3959 it is the maximum number of bytes that should be skipped by this alignment
3960 directive. If doing the alignment would require skipping more bytes than the
3961 specified maximum, then the alignment is not done at all. You can omit the
3962 fill value (the second argument) entirely by simply using two commas after the
3963 required alignment; this can be useful if you want the alignment to be filled
3964 with no-op instructions when appropriate.
3966 @cindex @code{balignw} directive
3967 @cindex @code{balignl} directive
3968 The @code{.balignw} and @code{.balignl} directives are variants of the
3969 @code{.balign} directive. The @code{.balignw} directive treats the fill
3970 pattern as a two byte word value. The @code{.balignl} directives treats the
3971 fill pattern as a four byte longword value. For example, @code{.balignw
3972 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3973 filled in with the value 0x368d (the exact placement of the bytes depends upon
3974 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3978 @section @code{.byte @var{expressions}}
3980 @cindex @code{byte} directive
3981 @cindex integers, one byte
3982 @code{.byte} expects zero or more expressions, separated by commas.
3983 Each expression is assembled into the next byte.
3986 @section @code{.comm @var{symbol} , @var{length} }
3988 @cindex @code{comm} directive
3989 @cindex symbol, common
3990 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
3991 common symbol in one object file may be merged with a defined or common symbol
3992 of the same name in another object file. If @code{@value{LD}} does not see a
3993 definition for the symbol--just one or more common symbols--then it will
3994 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
3995 absolute expression. If @code{@value{LD}} sees multiple common symbols with
3996 the same name, and they do not all have the same size, it will allocate space
3997 using the largest size.
4000 When using ELF, the @code{.comm} directive takes an optional third argument.
4001 This is the desired alignment of the symbol, specified as a byte boundary (for
4002 example, an alignment of 16 means that the least significant 4 bits of the
4003 address should be zero). The alignment must be an absolute expression, and it
4004 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
4005 for the common symbol, it will use the alignment when placing the symbol. If
4006 no alignment is specified, @command{@value{AS}} will set the alignment to the
4007 largest power of two less than or equal to the size of the symbol, up to a
4012 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4013 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4016 @node CFI directives
4017 @section @code{.cfi_startproc}
4018 @cindex @code{cfi_startproc} directive
4019 @code{.cfi_startproc} is used at the beginning of each function that
4020 should have an entry in @code{.eh_frame}. It initializes some internal
4021 data structures and emits architecture dependent initial CFI instructions.
4022 Don't forget to close the function by
4023 @code{.cfi_endproc}.
4025 @section @code{.cfi_endproc}
4026 @cindex @code{cfi_endproc} directive
4027 @code{.cfi_endproc} is used at the end of a function where it closes its
4028 unwind entry previously opened by
4029 @code{.cfi_startproc}. and emits it to @code{.eh_frame}.
4031 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4032 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4033 address from @var{register} and add @var{offset} to it}.
4035 @section @code{.cfi_def_cfa_register @var{register}}
4036 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4037 now on @var{register} will be used instead of the old one. Offset
4040 @section @code{.cfi_def_cfa_offset @var{offset}}
4041 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4042 remains the same, but @var{offset} is new. Note that it is the
4043 absolute offset that will be added to a defined register to compute
4046 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4047 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4048 value that is added/substracted from the previous offset.
4050 @section @code{.cfi_offset @var{register}, @var{offset}}
4051 Previous value of @var{register} is saved at offset @var{offset} from
4054 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4055 Previous value of @var{register} is saved at offset @var{offset} from
4056 the current CFA register. This is transformed to @code{.cfi_offset}
4057 using the known displacement of the CFA register from the CFA.
4058 This is often easier to use, because the number will match the
4059 code it's annotating.
4061 @section @code{.cfi_window_save}
4062 SPARC register window has been saved.
4064 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4065 Allows the user to add arbitrary bytes to the unwind info. One
4066 might use this to add OS-specific CFI opcodes, or generic CFI
4067 opcodes that GAS does not yet support.
4070 @section @code{.data @var{subsection}}
4072 @cindex @code{data} directive
4073 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4074 end of the data subsection numbered @var{subsection} (which is an
4075 absolute expression). If @var{subsection} is omitted, it defaults
4080 @section @code{.def @var{name}}
4082 @cindex @code{def} directive
4083 @cindex COFF symbols, debugging
4084 @cindex debugging COFF symbols
4085 Begin defining debugging information for a symbol @var{name}; the
4086 definition extends until the @code{.endef} directive is encountered.
4089 This directive is only observed when @command{@value{AS}} is configured for COFF
4090 format output; when producing @code{b.out}, @samp{.def} is recognized,
4097 @section @code{.desc @var{symbol}, @var{abs-expression}}
4099 @cindex @code{desc} directive
4100 @cindex COFF symbol descriptor
4101 @cindex symbol descriptor, COFF
4102 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4103 to the low 16 bits of an absolute expression.
4106 The @samp{.desc} directive is not available when @command{@value{AS}} is
4107 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4108 object format. For the sake of compatibility, @command{@value{AS}} accepts
4109 it, but produces no output, when configured for COFF.
4115 @section @code{.dim}
4117 @cindex @code{dim} directive
4118 @cindex COFF auxiliary symbol information
4119 @cindex auxiliary symbol information, COFF
4120 This directive is generated by compilers to include auxiliary debugging
4121 information in the symbol table. It is only permitted inside
4122 @code{.def}/@code{.endef} pairs.
4125 @samp{.dim} is only meaningful when generating COFF format output; when
4126 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4132 @section @code{.double @var{flonums}}
4134 @cindex @code{double} directive
4135 @cindex floating point numbers (double)
4136 @code{.double} expects zero or more flonums, separated by commas. It
4137 assembles floating point numbers.
4139 The exact kind of floating point numbers emitted depends on how
4140 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4144 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4145 in @sc{ieee} format.
4150 @section @code{.eject}
4152 @cindex @code{eject} directive
4153 @cindex new page, in listings
4154 @cindex page, in listings
4155 @cindex listing control: new page
4156 Force a page break at this point, when generating assembly listings.
4159 @section @code{.else}
4161 @cindex @code{else} directive
4162 @code{.else} is part of the @command{@value{AS}} support for conditional
4163 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
4164 of code to be assembled if the condition for the preceding @code{.if}
4168 @section @code{.elseif}
4170 @cindex @code{elseif} directive
4171 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4172 assembly; @pxref{If,,@code{.if}}. It is shorthand for beginning a new
4173 @code{.if} block that would otherwise fill the entire @code{.else} section.
4176 @section @code{.end}
4178 @cindex @code{end} directive
4179 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4180 process anything in the file past the @code{.end} directive.
4184 @section @code{.endef}
4186 @cindex @code{endef} directive
4187 This directive flags the end of a symbol definition begun with
4191 @samp{.endef} is only meaningful when generating COFF format output; if
4192 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4193 directive but ignores it.
4198 @section @code{.endfunc}
4199 @cindex @code{endfunc} directive
4200 @code{.endfunc} marks the end of a function specified with @code{.func}.
4203 @section @code{.endif}
4205 @cindex @code{endif} directive
4206 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4207 it marks the end of a block of code that is only assembled
4208 conditionally. @xref{If,,@code{.if}}.
4211 @section @code{.equ @var{symbol}, @var{expression}}
4213 @cindex @code{equ} directive
4214 @cindex assigning values to symbols
4215 @cindex symbols, assigning values to
4216 This directive sets the value of @var{symbol} to @var{expression}.
4217 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
4220 The syntax for @code{equ} on the HPPA is
4221 @samp{@var{symbol} .equ @var{expression}}.
4225 @section @code{.equiv @var{symbol}, @var{expression}}
4226 @cindex @code{equiv} directive
4227 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4228 the assembler will signal an error if @var{symbol} is already defined. Note a
4229 symbol which has been referenced but not actually defined is considered to be
4232 Except for the contents of the error message, this is roughly equivalent to
4241 @section @code{.err}
4242 @cindex @code{err} directive
4243 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4244 message and, unless the @option{-Z} option was used, it will not generate an
4245 object file. This can be used to signal error an conditionally compiled code.
4248 @section @code{.error "@var{string}"}
4249 @cindex error directive
4251 Similarly to @code{.err}, this directive emits an error, but you can specify a
4252 string that will be emitted as the error message. If you don't specify the
4253 message, it defaults to @code{".error directive invoked in source file"}.
4254 @xref{Errors, ,Error and Warning Messages}.
4257 .error "This code has not been assembled and tested."
4261 @section @code{.exitm}
4262 Exit early from the current macro definition. @xref{Macro}.
4265 @section @code{.extern}
4267 @cindex @code{extern} directive
4268 @code{.extern} is accepted in the source program---for compatibility
4269 with other assemblers---but it is ignored. @command{@value{AS}} treats
4270 all undefined symbols as external.
4273 @section @code{.fail @var{expression}}
4275 @cindex @code{fail} directive
4276 Generates an error or a warning. If the value of the @var{expression} is 500
4277 or more, @command{@value{AS}} will print a warning message. If the value is less
4278 than 500, @command{@value{AS}} will print an error message. The message will
4279 include the value of @var{expression}. This can occasionally be useful inside
4280 complex nested macros or conditional assembly.
4282 @ifclear no-file-dir
4284 @section @code{.file @var{string}}
4286 @cindex @code{file} directive
4287 @cindex logical file name
4288 @cindex file name, logical
4289 @code{.file} tells @command{@value{AS}} that we are about to start a new logical
4290 file. @var{string} is the new file name. In general, the filename is
4291 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4292 to specify an empty file name, you must give the quotes--@code{""}. This
4293 statement may go away in future: it is only recognized to be compatible with
4294 old @command{@value{AS}} programs.
4296 In some configurations of @command{@value{AS}}, @code{.file} has already been
4297 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
4302 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4304 @cindex @code{fill} directive
4305 @cindex writing patterns in memory
4306 @cindex patterns, writing in memory
4307 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4308 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4309 may be zero or more. @var{Size} may be zero or more, but if it is
4310 more than 8, then it is deemed to have the value 8, compatible with
4311 other people's assemblers. The contents of each @var{repeat} bytes
4312 is taken from an 8-byte number. The highest order 4 bytes are
4313 zero. The lowest order 4 bytes are @var{value} rendered in the
4314 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4315 Each @var{size} bytes in a repetition is taken from the lowest order
4316 @var{size} bytes of this number. Again, this bizarre behavior is
4317 compatible with other people's assemblers.
4319 @var{size} and @var{value} are optional.
4320 If the second comma and @var{value} are absent, @var{value} is
4321 assumed zero. If the first comma and following tokens are absent,
4322 @var{size} is assumed to be 1.
4325 @section @code{.float @var{flonums}}
4327 @cindex floating point numbers (single)
4328 @cindex @code{float} directive
4329 This directive assembles zero or more flonums, separated by commas. It
4330 has the same effect as @code{.single}.
4332 The exact kind of floating point numbers emitted depends on how
4333 @command{@value{AS}} is configured.
4334 @xref{Machine Dependencies}.
4338 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4339 in @sc{ieee} format.
4344 @section @code{.func @var{name}[,@var{label}]}
4345 @cindex @code{func} directive
4346 @code{.func} emits debugging information to denote function @var{name}, and
4347 is ignored unless the file is assembled with debugging enabled.
4348 Only @samp{--gstabs[+]} is currently supported.
4349 @var{label} is the entry point of the function and if omitted @var{name}
4350 prepended with the @samp{leading char} is used.
4351 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4352 All functions are currently defined to have @code{void} return type.
4353 The function must be terminated with @code{.endfunc}.
4356 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4358 @cindex @code{global} directive
4359 @cindex symbol, making visible to linker
4360 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4361 @var{symbol} in your partial program, its value is made available to
4362 other partial programs that are linked with it. Otherwise,
4363 @var{symbol} takes its attributes from a symbol of the same name
4364 from another file linked into the same program.
4366 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4367 compatibility with other assemblers.
4370 On the HPPA, @code{.global} is not always enough to make it accessible to other
4371 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4372 @xref{HPPA Directives,, HPPA Assembler Directives}.
4377 @section @code{.hidden @var{names}}
4379 @cindex @code{hidden} directive
4381 This is one of the ELF visibility directives. The other two are
4382 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4383 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4385 This directive overrides the named symbols default visibility (which is set by
4386 their binding: local, global or weak). The directive sets the visibility to
4387 @code{hidden} which means that the symbols are not visible to other components.
4388 Such symbols are always considered to be @code{protected} as well.
4392 @section @code{.hword @var{expressions}}
4394 @cindex @code{hword} directive
4395 @cindex integers, 16-bit
4396 @cindex numbers, 16-bit
4397 @cindex sixteen bit integers
4398 This expects zero or more @var{expressions}, and emits
4399 a 16 bit number for each.
4402 This directive is a synonym for @samp{.short}; depending on the target
4403 architecture, it may also be a synonym for @samp{.word}.
4407 This directive is a synonym for @samp{.short}.
4410 This directive is a synonym for both @samp{.short} and @samp{.word}.
4415 @section @code{.ident}
4417 @cindex @code{ident} directive
4418 This directive is used by some assemblers to place tags in object files.
4419 @command{@value{AS}} simply accepts the directive for source-file
4420 compatibility with such assemblers, but does not actually emit anything
4424 @section @code{.if @var{absolute expression}}
4426 @cindex conditional assembly
4427 @cindex @code{if} directive
4428 @code{.if} marks the beginning of a section of code which is only
4429 considered part of the source program being assembled if the argument
4430 (which must be an @var{absolute expression}) is non-zero. The end of
4431 the conditional section of code must be marked by @code{.endif}
4432 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4433 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4434 If you have several conditions to check, @code{.elseif} may be used to avoid
4435 nesting blocks if/else within each subsequent @code{.else} block.
4437 The following variants of @code{.if} are also supported:
4439 @cindex @code{ifdef} directive
4440 @item .ifdef @var{symbol}
4441 Assembles the following section of code if the specified @var{symbol}
4442 has been defined. Note a symbol which has been referenced but not yet defined
4443 is considered to be undefined.
4445 @cindex @code{ifc} directive
4446 @item .ifc @var{string1},@var{string2}
4447 Assembles the following section of code if the two strings are the same. The
4448 strings may be optionally quoted with single quotes. If they are not quoted,
4449 the first string stops at the first comma, and the second string stops at the
4450 end of the line. Strings which contain whitespace should be quoted. The
4451 string comparison is case sensitive.
4453 @cindex @code{ifeq} directive
4454 @item .ifeq @var{absolute expression}
4455 Assembles the following section of code if the argument is zero.
4457 @cindex @code{ifeqs} directive
4458 @item .ifeqs @var{string1},@var{string2}
4459 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4461 @cindex @code{ifge} directive
4462 @item .ifge @var{absolute expression}
4463 Assembles the following section of code if the argument is greater than or
4466 @cindex @code{ifgt} directive
4467 @item .ifgt @var{absolute expression}
4468 Assembles the following section of code if the argument is greater than zero.
4470 @cindex @code{ifle} directive
4471 @item .ifle @var{absolute expression}
4472 Assembles the following section of code if the argument is less than or equal
4475 @cindex @code{iflt} directive
4476 @item .iflt @var{absolute expression}
4477 Assembles the following section of code if the argument is less than zero.
4479 @cindex @code{ifnc} directive
4480 @item .ifnc @var{string1},@var{string2}.
4481 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4482 following section of code if the two strings are not the same.
4484 @cindex @code{ifndef} directive
4485 @cindex @code{ifnotdef} directive
4486 @item .ifndef @var{symbol}
4487 @itemx .ifnotdef @var{symbol}
4488 Assembles the following section of code if the specified @var{symbol}
4489 has not been defined. Both spelling variants are equivalent. Note a symbol
4490 which has been referenced but not yet defined is considered to be undefined.
4492 @cindex @code{ifne} directive
4493 @item .ifne @var{absolute expression}
4494 Assembles the following section of code if the argument is not equal to zero
4495 (in other words, this is equivalent to @code{.if}).
4497 @cindex @code{ifnes} directive
4498 @item .ifnes @var{string1},@var{string2}
4499 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4500 following section of code if the two strings are not the same.
4504 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4506 @cindex @code{incbin} directive
4507 @cindex binary files, including
4508 The @code{incbin} directive includes @var{file} verbatim at the current
4509 location. You can control the search paths used with the @samp{-I} command-line
4510 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4513 The @var{skip} argument skips a number of bytes from the start of the
4514 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4515 read. Note that the data is not aligned in any way, so it is the user's
4516 responsibility to make sure that proper alignment is provided both before and
4517 after the @code{incbin} directive.
4520 @section @code{.include "@var{file}"}
4522 @cindex @code{include} directive
4523 @cindex supporting files, including
4524 @cindex files, including
4525 This directive provides a way to include supporting files at specified
4526 points in your source program. The code from @var{file} is assembled as
4527 if it followed the point of the @code{.include}; when the end of the
4528 included file is reached, assembly of the original file continues. You
4529 can control the search paths used with the @samp{-I} command-line option
4530 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4534 @section @code{.int @var{expressions}}
4536 @cindex @code{int} directive
4537 @cindex integers, 32-bit
4538 Expect zero or more @var{expressions}, of any section, separated by commas.
4539 For each expression, emit a number that, at run time, is the value of that
4540 expression. The byte order and bit size of the number depends on what kind
4541 of target the assembly is for.
4545 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
4546 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4553 @section @code{.internal @var{names}}
4555 @cindex @code{internal} directive
4557 This is one of the ELF visibility directives. The other two are
4558 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4559 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4561 This directive overrides the named symbols default visibility (which is set by
4562 their binding: local, global or weak). The directive sets the visibility to
4563 @code{internal} which means that the symbols are considered to be @code{hidden}
4564 (i.e., not visible to other components), and that some extra, processor specific
4565 processing must also be performed upon the symbols as well.
4569 @section @code{.irp @var{symbol},@var{values}}@dots{}
4571 @cindex @code{irp} directive
4572 Evaluate a sequence of statements assigning different values to @var{symbol}.
4573 The sequence of statements starts at the @code{.irp} directive, and is
4574 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4575 set to @var{value}, and the sequence of statements is assembled. If no
4576 @var{value} is listed, the sequence of statements is assembled once, with
4577 @var{symbol} set to the null string. To refer to @var{symbol} within the
4578 sequence of statements, use @var{\symbol}.
4580 For example, assembling
4588 is equivalent to assembling
4597 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4599 @cindex @code{irpc} directive
4600 Evaluate a sequence of statements assigning different values to @var{symbol}.
4601 The sequence of statements starts at the @code{.irpc} directive, and is
4602 terminated by an @code{.endr} directive. For each character in @var{value},
4603 @var{symbol} is set to the character, and the sequence of statements is
4604 assembled. If no @var{value} is listed, the sequence of statements is
4605 assembled once, with @var{symbol} set to the null string. To refer to
4606 @var{symbol} within the sequence of statements, use @var{\symbol}.
4608 For example, assembling
4616 is equivalent to assembling
4625 @section @code{.lcomm @var{symbol} , @var{length}}
4627 @cindex @code{lcomm} directive
4628 @cindex local common symbols
4629 @cindex symbols, local common
4630 Reserve @var{length} (an absolute expression) bytes for a local common
4631 denoted by @var{symbol}. The section and value of @var{symbol} are
4632 those of the new local common. The addresses are allocated in the bss
4633 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4634 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4635 not visible to @code{@value{LD}}.
4638 Some targets permit a third argument to be used with @code{.lcomm}. This
4639 argument specifies the desired alignment of the symbol in the bss section.
4643 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4644 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4648 @section @code{.lflags}
4650 @cindex @code{lflags} directive (ignored)
4651 @command{@value{AS}} accepts this directive, for compatibility with other
4652 assemblers, but ignores it.
4654 @ifclear no-line-dir
4656 @section @code{.line @var{line-number}}
4658 @cindex @code{line} directive
4662 @section @code{.ln @var{line-number}}
4664 @cindex @code{ln} directive
4666 @cindex logical line number
4668 Change the logical line number. @var{line-number} must be an absolute
4669 expression. The next line has that logical line number. Therefore any other
4670 statements on the current line (after a statement separator character) are
4671 reported as on logical line number @var{line-number} @minus{} 1. One day
4672 @command{@value{AS}} will no longer support this directive: it is recognized only
4673 for compatibility with existing assembler programs.
4677 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
4678 not available; use the synonym @code{.ln} in that context.
4683 @ifclear no-line-dir
4684 Even though this is a directive associated with the @code{a.out} or
4685 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
4686 when producing COFF output, and treats @samp{.line} as though it
4687 were the COFF @samp{.ln} @emph{if} it is found outside a
4688 @code{.def}/@code{.endef} pair.
4690 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4691 used by compilers to generate auxiliary symbol information for
4696 @section @code{.linkonce [@var{type}]}
4698 @cindex @code{linkonce} directive
4699 @cindex common sections
4700 Mark the current section so that the linker only includes a single copy of it.
4701 This may be used to include the same section in several different object files,
4702 but ensure that the linker will only include it once in the final output file.
4703 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4704 Duplicate sections are detected based on the section name, so it should be
4707 This directive is only supported by a few object file formats; as of this
4708 writing, the only object file format which supports it is the Portable
4709 Executable format used on Windows NT.
4711 The @var{type} argument is optional. If specified, it must be one of the
4712 following strings. For example:
4716 Not all types may be supported on all object file formats.
4720 Silently discard duplicate sections. This is the default.
4723 Warn if there are duplicate sections, but still keep only one copy.
4726 Warn if any of the duplicates have different sizes.
4729 Warn if any of the duplicates do not have exactly the same contents.
4733 @section @code{.ln @var{line-number}}
4735 @cindex @code{ln} directive
4736 @ifclear no-line-dir
4737 @samp{.ln} is a synonym for @samp{.line}.
4740 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
4741 must be an absolute expression. The next line has that logical
4742 line number, so any other statements on the current line (after a
4743 statement separator character @code{;}) are reported as on logical
4744 line number @var{line-number} @minus{} 1.
4747 This directive is accepted, but ignored, when @command{@value{AS}} is
4748 configured for @code{b.out}; its effect is only associated with COFF
4754 @section @code{.mri @var{val}}
4756 @cindex @code{mri} directive
4757 @cindex MRI mode, temporarily
4758 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
4759 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
4760 affects code assembled until the next @code{.mri} directive, or until the end
4761 of the file. @xref{M, MRI mode, MRI mode}.
4764 @section @code{.list}
4766 @cindex @code{list} directive
4767 @cindex listing control, turning on
4768 Control (in conjunction with the @code{.nolist} directive) whether or
4769 not assembly listings are generated. These two directives maintain an
4770 internal counter (which is zero initially). @code{.list} increments the
4771 counter, and @code{.nolist} decrements it. Assembly listings are
4772 generated whenever the counter is greater than zero.
4774 By default, listings are disabled. When you enable them (with the
4775 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4776 the initial value of the listing counter is one.
4779 @section @code{.long @var{expressions}}
4781 @cindex @code{long} directive
4782 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
4785 @c no one seems to know what this is for or whether this description is
4786 @c what it really ought to do
4788 @section @code{.lsym @var{symbol}, @var{expression}}
4790 @cindex @code{lsym} directive
4791 @cindex symbol, not referenced in assembly
4792 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
4793 the hash table, ensuring it cannot be referenced by name during the
4794 rest of the assembly. This sets the attributes of the symbol to be
4795 the same as the expression value:
4797 @var{other} = @var{descriptor} = 0
4798 @var{type} = @r{(section of @var{expression})}
4799 @var{value} = @var{expression}
4802 The new symbol is not flagged as external.
4806 @section @code{.macro}
4809 The commands @code{.macro} and @code{.endm} allow you to define macros that
4810 generate assembly output. For example, this definition specifies a macro
4811 @code{sum} that puts a sequence of numbers into memory:
4814 .macro sum from=0, to=5
4823 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
4835 @item .macro @var{macname}
4836 @itemx .macro @var{macname} @var{macargs} @dots{}
4837 @cindex @code{macro} directive
4838 Begin the definition of a macro called @var{macname}. If your macro
4839 definition requires arguments, specify their names after the macro name,
4840 separated by commas or spaces. You can supply a default value for any
4841 macro argument by following the name with @samp{=@var{deflt}}. For
4842 example, these are all valid @code{.macro} statements:
4846 Begin the definition of a macro called @code{comm}, which takes no
4849 @item .macro plus1 p, p1
4850 @itemx .macro plus1 p p1
4851 Either statement begins the definition of a macro called @code{plus1},
4852 which takes two arguments; within the macro definition, write
4853 @samp{\p} or @samp{\p1} to evaluate the arguments.
4855 @item .macro reserve_str p1=0 p2
4856 Begin the definition of a macro called @code{reserve_str}, with two
4857 arguments. The first argument has a default value, but not the second.
4858 After the definition is complete, you can call the macro either as
4859 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
4860 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
4861 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
4862 @samp{0}, and @samp{\p2} evaluating to @var{b}).
4865 When you call a macro, you can specify the argument values either by
4866 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
4867 @samp{sum to=17, from=9}.
4870 @cindex @code{endm} directive
4871 Mark the end of a macro definition.
4874 @cindex @code{exitm} directive
4875 Exit early from the current macro definition.
4877 @cindex number of macros executed
4878 @cindex macros, count executed
4880 @command{@value{AS}} maintains a counter of how many macros it has
4881 executed in this pseudo-variable; you can copy that number to your
4882 output with @samp{\@@}, but @emph{only within a macro definition}.
4884 @item LOCAL @var{name} [ , @dots{} ]
4885 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
4886 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
4887 @xref{Altmacro,,@code{.altmacro}}.
4891 @section @code{.altmacro}
4892 Enable alternate macro mode, enabling:
4895 @item LOCAL @var{name} [ , @dots{} ]
4896 One additional directive, @code{LOCAL}, is available. It is used to
4897 generate a string replacement for each of the @var{name} arguments, and
4898 replace any instances of @var{name} in each macro expansion. The
4899 replacement string is unique in the assembly, and different for each
4900 separate macro expansion. @code{LOCAL} allows you to write macros that
4901 define symbols, without fear of conflict between separate macro expansions.
4903 @item String delimiters
4904 You can write strings delimited in these other ways besides
4905 @code{"@var{string}"}:
4908 @item '@var{string}'
4909 You can delimit strings with single-quote charaters.
4911 @item <@var{string}>
4912 You can delimit strings with matching angle brackets.
4915 @item single-character string escape
4916 To include any single character literally in a string (even if the
4917 character would otherwise have some special meaning), you can prefix the
4918 character with @samp{!} (an exclamation mark). For example, you can
4919 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4921 @item Expression results as strings
4922 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4923 and use the result as a string.
4927 @section @code{.noaltmacro}
4928 Disable alternate macro mode. @ref{Altmacro}
4931 @section @code{.nolist}
4933 @cindex @code{nolist} directive
4934 @cindex listing control, turning off
4935 Control (in conjunction with the @code{.list} directive) whether or
4936 not assembly listings are generated. These two directives maintain an
4937 internal counter (which is zero initially). @code{.list} increments the
4938 counter, and @code{.nolist} decrements it. Assembly listings are
4939 generated whenever the counter is greater than zero.
4942 @section @code{.octa @var{bignums}}
4944 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
4945 @cindex @code{octa} directive
4946 @cindex integer, 16-byte
4947 @cindex sixteen byte integer
4948 This directive expects zero or more bignums, separated by commas. For each
4949 bignum, it emits a 16-byte integer.
4951 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
4952 hence @emph{octa}-word for 16 bytes.
4955 @section @code{.org @var{new-lc} , @var{fill}}
4957 @cindex @code{org} directive
4958 @cindex location counter, advancing
4959 @cindex advancing location counter
4960 @cindex current address, advancing
4961 Advance the location counter of the current section to
4962 @var{new-lc}. @var{new-lc} is either an absolute expression or an
4963 expression with the same section as the current subsection. That is,
4964 you can't use @code{.org} to cross sections: if @var{new-lc} has the
4965 wrong section, the @code{.org} directive is ignored. To be compatible
4966 with former assemblers, if the section of @var{new-lc} is absolute,
4967 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
4968 is the same as the current subsection.
4970 @code{.org} may only increase the location counter, or leave it
4971 unchanged; you cannot use @code{.org} to move the location counter
4974 @c double negative used below "not undefined" because this is a specific
4975 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
4976 @c section. doc@cygnus.com 18feb91
4977 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
4978 may not be undefined. If you really detest this restriction we eagerly await
4979 a chance to share your improved assembler.
4981 Beware that the origin is relative to the start of the section, not
4982 to the start of the subsection. This is compatible with other
4983 people's assemblers.
4985 When the location counter (of the current subsection) is advanced, the
4986 intervening bytes are filled with @var{fill} which should be an
4987 absolute expression. If the comma and @var{fill} are omitted,
4988 @var{fill} defaults to zero.
4991 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4993 @cindex padding the location counter given a power of two
4994 @cindex @code{p2align} directive
4995 Pad the location counter (in the current subsection) to a particular
4996 storage boundary. The first expression (which must be absolute) is the
4997 number of low-order zero bits the location counter must have after
4998 advancement. For example @samp{.p2align 3} advances the location
4999 counter until it a multiple of 8. If the location counter is already a
5000 multiple of 8, no change is needed.
5002 The second expression (also absolute) gives the fill value to be stored in the
5003 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5004 padding bytes are normally zero. However, on some systems, if the section is
5005 marked as containing code and the fill value is omitted, the space is filled
5006 with no-op instructions.
5008 The third expression is also absolute, and is also optional. If it is present,
5009 it is the maximum number of bytes that should be skipped by this alignment
5010 directive. If doing the alignment would require skipping more bytes than the
5011 specified maximum, then the alignment is not done at all. You can omit the
5012 fill value (the second argument) entirely by simply using two commas after the
5013 required alignment; this can be useful if you want the alignment to be filled
5014 with no-op instructions when appropriate.
5016 @cindex @code{p2alignw} directive
5017 @cindex @code{p2alignl} directive
5018 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5019 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5020 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5021 fill pattern as a four byte longword value. For example, @code{.p2alignw
5022 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5023 filled in with the value 0x368d (the exact placement of the bytes depends upon
5024 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5029 @section @code{.previous}
5031 @cindex @code{previous} directive
5032 @cindex Section Stack
5033 This is one of the ELF section stack manipulation directives. The others are
5034 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5035 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5036 (@pxref{PopSection}).
5038 This directive swaps the current section (and subsection) with most recently
5039 referenced section (and subsection) prior to this one. Multiple
5040 @code{.previous} directives in a row will flip between two sections (and their
5043 In terms of the section stack, this directive swaps the current section with
5044 the top section on the section stack.
5049 @section @code{.popsection}
5051 @cindex @code{popsection} directive
5052 @cindex Section Stack
5053 This is one of the ELF section stack manipulation directives. The others are
5054 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5055 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5058 This directive replaces the current section (and subsection) with the top
5059 section (and subsection) on the section stack. This section is popped off the
5064 @section @code{.print @var{string}}
5066 @cindex @code{print} directive
5067 @command{@value{AS}} will print @var{string} on the standard output during
5068 assembly. You must put @var{string} in double quotes.
5072 @section @code{.protected @var{names}}
5074 @cindex @code{protected} directive
5076 This is one of the ELF visibility directives. The other two are
5077 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5079 This directive overrides the named symbols default visibility (which is set by
5080 their binding: local, global or weak). The directive sets the visibility to
5081 @code{protected} which means that any references to the symbols from within the
5082 components that defines them must be resolved to the definition in that
5083 component, even if a definition in another component would normally preempt
5088 @section @code{.psize @var{lines} , @var{columns}}
5090 @cindex @code{psize} directive
5091 @cindex listing control: paper size
5092 @cindex paper size, for listings
5093 Use this directive to declare the number of lines---and, optionally, the
5094 number of columns---to use for each page, when generating listings.
5096 If you do not use @code{.psize}, listings use a default line-count
5097 of 60. You may omit the comma and @var{columns} specification; the
5098 default width is 200 columns.
5100 @command{@value{AS}} generates formfeeds whenever the specified number of
5101 lines is exceeded (or whenever you explicitly request one, using
5104 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5105 those explicitly specified with @code{.eject}.
5108 @section @code{.purgem @var{name}}
5110 @cindex @code{purgem} directive
5111 Undefine the macro @var{name}, so that later uses of the string will not be
5112 expanded. @xref{Macro}.
5116 @section @code{.pushsection @var{name} , @var{subsection}}
5118 @cindex @code{pushsection} directive
5119 @cindex Section Stack
5120 This is one of the ELF section stack manipulation directives. The others are
5121 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5122 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5125 This directive pushes the current section (and subsection) onto the
5126 top of the section stack, and then replaces the current section and
5127 subsection with @code{name} and @code{subsection}.
5131 @section @code{.quad @var{bignums}}
5133 @cindex @code{quad} directive
5134 @code{.quad} expects zero or more bignums, separated by commas. For
5135 each bignum, it emits
5137 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5138 warning message; and just takes the lowest order 8 bytes of the bignum.
5139 @cindex eight-byte integer
5140 @cindex integer, 8-byte
5142 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5143 hence @emph{quad}-word for 8 bytes.
5146 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5147 warning message; and just takes the lowest order 16 bytes of the bignum.
5148 @cindex sixteen-byte integer
5149 @cindex integer, 16-byte
5153 @section @code{.rept @var{count}}
5155 @cindex @code{rept} directive
5156 Repeat the sequence of lines between the @code{.rept} directive and the next
5157 @code{.endr} directive @var{count} times.
5159 For example, assembling
5167 is equivalent to assembling
5176 @section @code{.sbttl "@var{subheading}"}
5178 @cindex @code{sbttl} directive
5179 @cindex subtitles for listings
5180 @cindex listing control: subtitle
5181 Use @var{subheading} as the title (third line, immediately after the
5182 title line) when generating assembly listings.
5184 This directive affects subsequent pages, as well as the current page if
5185 it appears within ten lines of the top of a page.
5189 @section @code{.scl @var{class}}
5191 @cindex @code{scl} directive
5192 @cindex symbol storage class (COFF)
5193 @cindex COFF symbol storage class
5194 Set the storage-class value for a symbol. This directive may only be
5195 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5196 whether a symbol is static or external, or it may record further
5197 symbolic debugging information.
5200 The @samp{.scl} directive is primarily associated with COFF output; when
5201 configured to generate @code{b.out} output format, @command{@value{AS}}
5202 accepts this directive but ignores it.
5208 @section @code{.section @var{name}}
5210 @cindex named section
5211 Use the @code{.section} directive to assemble the following code into a section
5214 This directive is only supported for targets that actually support arbitrarily
5215 named sections; on @code{a.out} targets, for example, it is not accepted, even
5216 with a standard @code{a.out} section name.
5220 @c only print the extra heading if both COFF and ELF are set
5221 @subheading COFF Version
5224 @cindex @code{section} directive (COFF version)
5225 For COFF targets, the @code{.section} directive is used in one of the following
5229 .section @var{name}[, "@var{flags}"]
5230 .section @var{name}[, @var{subsegment}]
5233 If the optional argument is quoted, it is taken as flags to use for the
5234 section. Each flag is a single character. The following flags are recognized:
5237 bss section (uninitialized data)
5239 section is not loaded
5249 shared section (meaningful for PE targets)
5251 ignored. (For compatibility with the ELF version)
5254 If no flags are specified, the default flags depend upon the section name. If
5255 the section name is not recognized, the default will be for the section to be
5256 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5257 from the section, rather than adding them, so if they are used on their own it
5258 will be as if no flags had been specified at all.
5260 If the optional argument to the @code{.section} directive is not quoted, it is
5261 taken as a subsegment number (@pxref{Sub-Sections}).
5266 @c only print the extra heading if both COFF and ELF are set
5267 @subheading ELF Version
5270 @cindex Section Stack
5271 This is one of the ELF section stack manipulation directives. The others are
5272 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5273 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5274 @code{.previous} (@pxref{Previous}).
5276 @cindex @code{section} directive (ELF version)
5277 For ELF targets, the @code{.section} directive is used like this:
5280 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]
5283 The optional @var{flags} argument is a quoted string which may contain any
5284 combination of the following characters:
5287 section is allocatable
5291 section is executable
5293 section is mergeable
5295 section contains zero terminated strings
5297 section is a member of a section group
5299 section is used for thread-local-storage
5302 The optional @var{type} argument may contain one of the following constants:
5305 section contains data
5307 section does not contain data (i.e., section only occupies space)
5309 section contains data which is used by things other than the program
5311 section contains an array of pointers to init functions
5313 section contains an array of pointers to finish functions
5314 @item @@preinit_array
5315 section contains an array of pointers to pre-init functions
5318 Many targets only support the first three section types.
5320 Note on targets where the @code{@@} character is the start of a comment (eg
5321 ARM) then another character is used instead. For example the ARM port uses the
5324 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5325 be specified as well as an extra argument - @var{entsize} - like this:
5328 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5331 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5332 constants, each @var{entsize} octets long. Sections with both @code{M} and
5333 @code{S} must contain zero terminated strings where each character is
5334 @var{entsize} bytes long. The linker may remove duplicates within sections with
5335 the same name, same entity size and same flags. @var{entsize} must be an
5336 absolute expression.
5338 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5339 be present along with an additional field like this:
5342 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5345 The @var{GroupName} field specifies the name of the section group to which this
5346 particular section belongs. The optional linkage field can contain:
5349 indicates that only one copy of this section should be retained
5354 Note - if both the @var{M} and @var{G} flags are present then the fields for
5355 the Merge flag should come first, like this:
5358 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5361 If no flags are specified, the default flags depend upon the section name. If
5362 the section name is not recognized, the default will be for the section to have
5363 none of the above flags: it will not be allocated in memory, nor writable, nor
5364 executable. The section will contain data.
5366 For ELF targets, the assembler supports another type of @code{.section}
5367 directive for compatibility with the Solaris assembler:
5370 .section "@var{name}"[, @var{flags}...]
5373 Note that the section name is quoted. There may be a sequence of comma
5377 section is allocatable
5381 section is executable
5383 section is used for thread local storage
5386 This directive replaces the current section and subsection. See the
5387 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
5388 some examples of how this directive and the other section stack directives
5394 @section @code{.set @var{symbol}, @var{expression}}
5396 @cindex @code{set} directive
5397 @cindex symbol value, setting
5398 Set the value of @var{symbol} to @var{expression}. This
5399 changes @var{symbol}'s value and type to conform to
5400 @var{expression}. If @var{symbol} was flagged as external, it remains
5401 flagged (@pxref{Symbol Attributes}).
5403 You may @code{.set} a symbol many times in the same assembly.
5405 If you @code{.set} a global symbol, the value stored in the object
5406 file is the last value stored into it.
5409 The syntax for @code{set} on the HPPA is
5410 @samp{@var{symbol} .set @var{expression}}.
5414 @section @code{.short @var{expressions}}
5416 @cindex @code{short} directive
5418 @code{.short} is normally the same as @samp{.word}.
5419 @xref{Word,,@code{.word}}.
5421 In some configurations, however, @code{.short} and @code{.word} generate
5422 numbers of different lengths; @pxref{Machine Dependencies}.
5426 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5429 This expects zero or more @var{expressions}, and emits
5430 a 16 bit number for each.
5435 @section @code{.single @var{flonums}}
5437 @cindex @code{single} directive
5438 @cindex floating point numbers (single)
5439 This directive assembles zero or more flonums, separated by commas. It
5440 has the same effect as @code{.float}.
5442 The exact kind of floating point numbers emitted depends on how
5443 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5447 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5448 numbers in @sc{ieee} format.
5454 @section @code{.size}
5456 This directive is used to set the size associated with a symbol.
5460 @c only print the extra heading if both COFF and ELF are set
5461 @subheading COFF Version
5464 @cindex @code{size} directive (COFF version)
5465 For COFF targets, the @code{.size} directive is only permitted inside
5466 @code{.def}/@code{.endef} pairs. It is used like this:
5469 .size @var{expression}
5473 @samp{.size} is only meaningful when generating COFF format output; when
5474 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5481 @c only print the extra heading if both COFF and ELF are set
5482 @subheading ELF Version
5485 @cindex @code{size} directive (ELF version)
5486 For ELF targets, the @code{.size} directive is used like this:
5489 .size @var{name} , @var{expression}
5492 This directive sets the size associated with a symbol @var{name}.
5493 The size in bytes is computed from @var{expression} which can make use of label
5494 arithmetic. This directive is typically used to set the size of function
5500 @section @code{.sleb128 @var{expressions}}
5502 @cindex @code{sleb128} directive
5503 @var{sleb128} stands for ``signed little endian base 128.'' This is a
5504 compact, variable length representation of numbers used by the DWARF
5505 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
5507 @ifclear no-space-dir
5509 @section @code{.skip @var{size} , @var{fill}}
5511 @cindex @code{skip} directive
5512 @cindex filling memory
5513 This directive emits @var{size} bytes, each of value @var{fill}. Both
5514 @var{size} and @var{fill} are absolute expressions. If the comma and
5515 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
5519 @section @code{.space @var{size} , @var{fill}}
5521 @cindex @code{space} directive
5522 @cindex filling memory
5523 This directive emits @var{size} bytes, each of value @var{fill}. Both
5524 @var{size} and @var{fill} are absolute expressions. If the comma
5525 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
5530 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
5531 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
5532 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
5533 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
5542 @section @code{.space}
5543 @cindex @code{space} directive
5545 On the AMD 29K, this directive is ignored; it is accepted for
5546 compatibility with other AMD 29K assemblers.
5549 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
5550 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
5556 @section @code{.stabd, .stabn, .stabs}
5558 @cindex symbolic debuggers, information for
5559 @cindex @code{stab@var{x}} directives
5560 There are three directives that begin @samp{.stab}.
5561 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
5562 The symbols are not entered in the @command{@value{AS}} hash table: they
5563 cannot be referenced elsewhere in the source file.
5564 Up to five fields are required:
5568 This is the symbol's name. It may contain any character except
5569 @samp{\000}, so is more general than ordinary symbol names. Some
5570 debuggers used to code arbitrarily complex structures into symbol names
5574 An absolute expression. The symbol's type is set to the low 8 bits of
5575 this expression. Any bit pattern is permitted, but @code{@value{LD}}
5576 and debuggers choke on silly bit patterns.
5579 An absolute expression. The symbol's ``other'' attribute is set to the
5580 low 8 bits of this expression.
5583 An absolute expression. The symbol's descriptor is set to the low 16
5584 bits of this expression.
5587 An absolute expression which becomes the symbol's value.
5590 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
5591 or @code{.stabs} statement, the symbol has probably already been created;
5592 you get a half-formed symbol in your object file. This is
5593 compatible with earlier assemblers!
5596 @cindex @code{stabd} directive
5597 @item .stabd @var{type} , @var{other} , @var{desc}
5599 The ``name'' of the symbol generated is not even an empty string.
5600 It is a null pointer, for compatibility. Older assemblers used a
5601 null pointer so they didn't waste space in object files with empty
5604 The symbol's value is set to the location counter,
5605 relocatably. When your program is linked, the value of this symbol
5606 is the address of the location counter when the @code{.stabd} was
5609 @cindex @code{stabn} directive
5610 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
5611 The name of the symbol is set to the empty string @code{""}.
5613 @cindex @code{stabs} directive
5614 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
5615 All five fields are specified.
5621 @section @code{.string} "@var{str}"
5623 @cindex string, copying to object file
5624 @cindex @code{string} directive
5626 Copy the characters in @var{str} to the object file. You may specify more than
5627 one string to copy, separated by commas. Unless otherwise specified for a
5628 particular machine, the assembler marks the end of each string with a 0 byte.
5629 You can use any of the escape sequences described in @ref{Strings,,Strings}.
5632 @section @code{.struct @var{expression}}
5634 @cindex @code{struct} directive
5635 Switch to the absolute section, and set the section offset to @var{expression},
5636 which must be an absolute expression. You might use this as follows:
5645 This would define the symbol @code{field1} to have the value 0, the symbol
5646 @code{field2} to have the value 4, and the symbol @code{field3} to have the
5647 value 8. Assembly would be left in the absolute section, and you would need to
5648 use a @code{.section} directive of some sort to change to some other section
5649 before further assembly.
5653 @section @code{.subsection @var{name}}
5655 @cindex @code{subsection} directive
5656 @cindex Section Stack
5657 This is one of the ELF section stack manipulation directives. The others are
5658 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
5659 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5662 This directive replaces the current subsection with @code{name}. The current
5663 section is not changed. The replaced subsection is put onto the section stack
5664 in place of the then current top of stack subsection.
5669 @section @code{.symver}
5670 @cindex @code{symver} directive
5671 @cindex symbol versioning
5672 @cindex versions of symbols
5673 Use the @code{.symver} directive to bind symbols to specific version nodes
5674 within a source file. This is only supported on ELF platforms, and is
5675 typically used when assembling files to be linked into a shared library.
5676 There are cases where it may make sense to use this in objects to be bound
5677 into an application itself so as to override a versioned symbol from a
5680 For ELF targets, the @code{.symver} directive can be used like this:
5682 .symver @var{name}, @var{name2@@nodename}
5684 If the symbol @var{name} is defined within the file
5685 being assembled, the @code{.symver} directive effectively creates a symbol
5686 alias with the name @var{name2@@nodename}, and in fact the main reason that we
5687 just don't try and create a regular alias is that the @var{@@} character isn't
5688 permitted in symbol names. The @var{name2} part of the name is the actual name
5689 of the symbol by which it will be externally referenced. The name @var{name}
5690 itself is merely a name of convenience that is used so that it is possible to
5691 have definitions for multiple versions of a function within a single source
5692 file, and so that the compiler can unambiguously know which version of a
5693 function is being mentioned. The @var{nodename} portion of the alias should be
5694 the name of a node specified in the version script supplied to the linker when
5695 building a shared library. If you are attempting to override a versioned
5696 symbol from a shared library, then @var{nodename} should correspond to the
5697 nodename of the symbol you are trying to override.
5699 If the symbol @var{name} is not defined within the file being assembled, all
5700 references to @var{name} will be changed to @var{name2@@nodename}. If no
5701 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
5704 Another usage of the @code{.symver} directive is:
5706 .symver @var{name}, @var{name2@@@@nodename}
5708 In this case, the symbol @var{name} must exist and be defined within
5709 the file being assembled. It is similar to @var{name2@@nodename}. The
5710 difference is @var{name2@@@@nodename} will also be used to resolve
5711 references to @var{name2} by the linker.
5713 The third usage of the @code{.symver} directive is:
5715 .symver @var{name}, @var{name2@@@@@@nodename}
5717 When @var{name} is not defined within the
5718 file being assembled, it is treated as @var{name2@@nodename}. When
5719 @var{name} is defined within the file being assembled, the symbol
5720 name, @var{name}, will be changed to @var{name2@@@@nodename}.
5725 @section @code{.tag @var{structname}}
5727 @cindex COFF structure debugging
5728 @cindex structure debugging, COFF
5729 @cindex @code{tag} directive
5730 This directive is generated by compilers to include auxiliary debugging
5731 information in the symbol table. It is only permitted inside
5732 @code{.def}/@code{.endef} pairs. Tags are used to link structure
5733 definitions in the symbol table with instances of those structures.
5736 @samp{.tag} is only used when generating COFF format output; when
5737 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5743 @section @code{.text @var{subsection}}
5745 @cindex @code{text} directive
5746 Tells @command{@value{AS}} to assemble the following statements onto the end of
5747 the text subsection numbered @var{subsection}, which is an absolute
5748 expression. If @var{subsection} is omitted, subsection number zero
5752 @section @code{.title "@var{heading}"}
5754 @cindex @code{title} directive
5755 @cindex listing control: title line
5756 Use @var{heading} as the title (second line, immediately after the
5757 source file name and pagenumber) when generating assembly listings.
5759 This directive affects subsequent pages, as well as the current page if
5760 it appears within ten lines of the top of a page.
5764 @section @code{.type}
5766 This directive is used to set the type of a symbol.
5770 @c only print the extra heading if both COFF and ELF are set
5771 @subheading COFF Version
5774 @cindex COFF symbol type
5775 @cindex symbol type, COFF
5776 @cindex @code{type} directive (COFF version)
5777 For COFF targets, this directive is permitted only within
5778 @code{.def}/@code{.endef} pairs. It is used like this:
5784 This records the integer @var{int} as the type attribute of a symbol table
5788 @samp{.type} is associated only with COFF format output; when
5789 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
5790 directive but ignores it.
5796 @c only print the extra heading if both COFF and ELF are set
5797 @subheading ELF Version
5800 @cindex ELF symbol type
5801 @cindex symbol type, ELF
5802 @cindex @code{type} directive (ELF version)
5803 For ELF targets, the @code{.type} directive is used like this:
5806 .type @var{name} , @var{type description}
5809 This sets the type of symbol @var{name} to be either a
5810 function symbol or an object symbol. There are five different syntaxes
5811 supported for the @var{type description} field, in order to provide
5812 compatibility with various other assemblers. The syntaxes supported are:
5815 .type <name>,#function
5816 .type <name>,#object
5818 .type <name>,@@function
5819 .type <name>,@@object
5821 .type <name>,%function
5822 .type <name>,%object
5824 .type <name>,"function"
5825 .type <name>,"object"
5827 .type <name> STT_FUNCTION
5828 .type <name> STT_OBJECT
5834 @section @code{.uleb128 @var{expressions}}
5836 @cindex @code{uleb128} directive
5837 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
5838 compact, variable length representation of numbers used by the DWARF
5839 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
5843 @section @code{.val @var{addr}}
5845 @cindex @code{val} directive
5846 @cindex COFF value attribute
5847 @cindex value attribute, COFF
5848 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5849 records the address @var{addr} as the value attribute of a symbol table
5853 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
5854 configured for @code{b.out}, it accepts this directive but ignores it.
5860 @section @code{.version "@var{string}"}
5862 @cindex @code{version} directive
5863 This directive creates a @code{.note} section and places into it an ELF
5864 formatted note of type NT_VERSION. The note's name is set to @code{string}.
5869 @section @code{.vtable_entry @var{table}, @var{offset}}
5871 @cindex @code{vtable_entry} directive
5872 This directive finds or creates a symbol @code{table} and creates a
5873 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
5876 @section @code{.vtable_inherit @var{child}, @var{parent}}
5878 @cindex @code{vtable_inherit} directive
5879 This directive finds the symbol @code{child} and finds or creates the symbol
5880 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
5881 parent whose addend is the value of the child symbol. As a special case the
5882 parent name of @code{0} is treated as refering the @code{*ABS*} section.
5886 @section @code{.warning "@var{string}"}
5887 @cindex warning directive
5888 Similar to the directive @code{.error}
5889 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
5892 @section @code{.weak @var{names}}
5894 @cindex @code{weak} directive
5895 This directive sets the weak attribute on the comma separated list of symbol
5896 @code{names}. If the symbols do not already exist, they will be created.
5898 On COFF targets other than PE, weak symbols are a GNU extension. This
5899 directive sets the weak attribute on the comma separated list of symbol
5900 @code{names}. If the symbols do not already exist, they will be created.
5902 On the PE target, weak symbols are supported natively as weak aliases.
5903 When a weak symbol is created that is not an alias, GAS creates an
5904 alternate symbol to hold the default value.
5907 @section @code{.word @var{expressions}}
5909 @cindex @code{word} directive
5910 This directive expects zero or more @var{expressions}, of any section,
5911 separated by commas.
5914 For each expression, @command{@value{AS}} emits a 32-bit number.
5917 For each expression, @command{@value{AS}} emits a 16-bit number.
5922 The size of the number emitted, and its byte order,
5923 depend on what target computer the assembly is for.
5926 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
5927 @c happen---32-bit addressability, period; no long/short jumps.
5928 @ifset DIFF-TBL-KLUGE
5929 @cindex difference tables altered
5930 @cindex altered difference tables
5932 @emph{Warning: Special Treatment to support Compilers}
5936 Machines with a 32-bit address space, but that do less than 32-bit
5937 addressing, require the following special treatment. If the machine of
5938 interest to you does 32-bit addressing (or doesn't require it;
5939 @pxref{Machine Dependencies}), you can ignore this issue.
5942 In order to assemble compiler output into something that works,
5943 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
5944 Directives of the form @samp{.word sym1-sym2} are often emitted by
5945 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
5946 directive of the form @samp{.word sym1-sym2}, and the difference between
5947 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
5948 creates a @dfn{secondary jump table}, immediately before the next label.
5949 This secondary jump table is preceded by a short-jump to the
5950 first byte after the secondary table. This short-jump prevents the flow
5951 of control from accidentally falling into the new table. Inside the
5952 table is a long-jump to @code{sym2}. The original @samp{.word}
5953 contains @code{sym1} minus the address of the long-jump to
5956 If there were several occurrences of @samp{.word sym1-sym2} before the
5957 secondary jump table, all of them are adjusted. If there was a
5958 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
5959 long-jump to @code{sym4} is included in the secondary jump table,
5960 and the @code{.word} directives are adjusted to contain @code{sym3}
5961 minus the address of the long-jump to @code{sym4}; and so on, for as many
5962 entries in the original jump table as necessary.
5965 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
5966 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
5967 assembly language programmers.
5970 @c end DIFF-TBL-KLUGE
5973 @section Deprecated Directives
5975 @cindex deprecated directives
5976 @cindex obsolescent directives
5977 One day these directives won't work.
5978 They are included for compatibility with older assemblers.
5985 @node Machine Dependencies
5986 @chapter Machine Dependent Features
5988 @cindex machine dependencies
5989 The machine instruction sets are (almost by definition) different on
5990 each machine where @command{@value{AS}} runs. Floating point representations
5991 vary as well, and @command{@value{AS}} often supports a few additional
5992 directives or command-line options for compatibility with other
5993 assemblers on a particular platform. Finally, some versions of
5994 @command{@value{AS}} support special pseudo-instructions for branch
5997 This chapter discusses most of these differences, though it does not
5998 include details on any machine's instruction set. For details on that
5999 subject, see the hardware manufacturer's manual.
6003 * AMD29K-Dependent:: AMD 29K Dependent Features
6006 * Alpha-Dependent:: Alpha Dependent Features
6009 * ARC-Dependent:: ARC Dependent Features
6012 * ARM-Dependent:: ARM Dependent Features
6015 * CRIS-Dependent:: CRIS Dependent Features
6018 * D10V-Dependent:: D10V Dependent Features
6021 * D30V-Dependent:: D30V Dependent Features
6024 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6027 * H8/500-Dependent:: Renesas H8/500 Dependent Features
6030 * HPPA-Dependent:: HPPA Dependent Features
6033 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6036 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6039 * i860-Dependent:: Intel 80860 Dependent Features
6042 * i960-Dependent:: Intel 80960 Dependent Features
6045 * IA-64-Dependent:: Intel IA-64 Dependent Features
6048 * IP2K-Dependent:: IP2K Dependent Features
6051 * M32R-Dependent:: M32R Dependent Features
6054 * M68K-Dependent:: M680x0 Dependent Features
6057 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6060 * M88K-Dependent:: M880x0 Dependent Features
6063 * MIPS-Dependent:: MIPS Dependent Features
6066 * MMIX-Dependent:: MMIX Dependent Features
6069 * MSP430-Dependent:: MSP430 Dependent Features
6072 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6073 * SH64-Dependent:: SuperH SH64 Dependent Features
6076 * PDP-11-Dependent:: PDP-11 Dependent Features
6079 * PJ-Dependent:: picoJava Dependent Features
6082 * PPC-Dependent:: PowerPC Dependent Features
6085 * Sparc-Dependent:: SPARC Dependent Features
6088 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6091 * V850-Dependent:: V850 Dependent Features
6094 * Xtensa-Dependent:: Xtensa Dependent Features
6097 * Z8000-Dependent:: Z8000 Dependent Features
6100 * Vax-Dependent:: VAX Dependent Features
6107 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6108 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6109 @c peculiarity: to preserve cross-references, there must be a node called
6110 @c "Machine Dependencies". Hence the conditional nodenames in each
6111 @c major node below. Node defaulting in makeinfo requires adjacency of
6112 @c node and sectioning commands; hence the repetition of @chapter BLAH
6113 @c in both conditional blocks.
6116 @include c-a29k.texi
6120 @include c-alpha.texi
6132 @include c-cris.texi
6137 @node Machine Dependencies
6138 @chapter Machine Dependent Features
6140 The machine instruction sets are different on each Renesas chip family,
6141 and there are also some syntax differences among the families. This
6142 chapter describes the specific @command{@value{AS}} features for each
6146 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6147 * H8/500-Dependent:: Renesas H8/500 Dependent Features
6148 * SH-Dependent:: Renesas SH Dependent Features
6155 @include c-d10v.texi
6159 @include c-d30v.texi
6163 @include c-h8300.texi
6167 @include c-h8500.texi
6171 @include c-hppa.texi
6175 @include c-i370.texi
6179 @include c-i386.texi
6183 @include c-i860.texi
6187 @include c-i960.texi
6191 @include c-ia64.texi
6195 @include c-ip2k.texi
6199 @include c-m32r.texi
6203 @include c-m68k.texi
6207 @include c-m68hc11.texi
6211 @include c-m88k.texi
6215 @include c-mips.texi
6219 @include c-mmix.texi
6223 @include c-msp430.texi
6227 @include c-ns32k.texi
6231 @include c-pdp11.texi
6244 @include c-sh64.texi
6248 @include c-sparc.texi
6252 @include c-tic54x.texi
6264 @include c-v850.texi
6268 @include c-xtensa.texi
6272 @c reverse effect of @down at top of generic Machine-Dep chapter
6276 @node Reporting Bugs
6277 @chapter Reporting Bugs
6278 @cindex bugs in assembler
6279 @cindex reporting bugs in assembler
6281 Your bug reports play an essential role in making @command{@value{AS}} reliable.
6283 Reporting a bug may help you by bringing a solution to your problem, or it may
6284 not. But in any case the principal function of a bug report is to help the
6285 entire community by making the next version of @command{@value{AS}} work better.
6286 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
6288 In order for a bug report to serve its purpose, you must include the
6289 information that enables us to fix the bug.
6292 * Bug Criteria:: Have you found a bug?
6293 * Bug Reporting:: How to report bugs
6297 @section Have You Found a Bug?
6298 @cindex bug criteria
6300 If you are not sure whether you have found a bug, here are some guidelines:
6303 @cindex fatal signal
6304 @cindex assembler crash
6305 @cindex crash of assembler
6307 If the assembler gets a fatal signal, for any input whatever, that is a
6308 @command{@value{AS}} bug. Reliable assemblers never crash.
6310 @cindex error on valid input
6312 If @command{@value{AS}} produces an error message for valid input, that is a bug.
6314 @cindex invalid input
6316 If @command{@value{AS}} does not produce an error message for invalid input, that
6317 is a bug. However, you should note that your idea of ``invalid input'' might
6318 be our idea of ``an extension'' or ``support for traditional practice''.
6321 If you are an experienced user of assemblers, your suggestions for improvement
6322 of @command{@value{AS}} are welcome in any case.
6326 @section How to Report Bugs
6328 @cindex assembler bugs, reporting
6330 A number of companies and individuals offer support for @sc{gnu} products. If
6331 you obtained @command{@value{AS}} from a support organization, we recommend you
6332 contact that organization first.
6334 You can find contact information for many support companies and
6335 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
6338 In any event, we also recommend that you send bug reports for @command{@value{AS}}
6339 to @samp{bug-binutils@@gnu.org}.
6341 The fundamental principle of reporting bugs usefully is this:
6342 @strong{report all the facts}. If you are not sure whether to state a
6343 fact or leave it out, state it!
6345 Often people omit facts because they think they know what causes the problem
6346 and assume that some details do not matter. Thus, you might assume that the
6347 name of a symbol you use in an example does not matter. Well, probably it does
6348 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
6349 happens to fetch from the location where that name is stored in memory;
6350 perhaps, if the name were different, the contents of that location would fool
6351 the assembler into doing the right thing despite the bug. Play it safe and
6352 give a specific, complete example. That is the easiest thing for you to do,
6353 and the most helpful.
6355 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
6356 it is new to us. Therefore, always write your bug reports on the assumption
6357 that the bug has not been reported previously.
6359 Sometimes people give a few sketchy facts and ask, ``Does this ring a
6360 bell?'' This cannot help us fix a bug, so it is basically useless. We
6361 respond by asking for enough details to enable us to investigate.
6362 You might as well expedite matters by sending them to begin with.
6364 To enable us to fix the bug, you should include all these things:
6368 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
6369 it with the @samp{--version} argument.
6371 Without this, we will not know whether there is any point in looking for
6372 the bug in the current version of @command{@value{AS}}.
6375 Any patches you may have applied to the @command{@value{AS}} source.
6378 The type of machine you are using, and the operating system name and
6382 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
6386 The command arguments you gave the assembler to assemble your example and
6387 observe the bug. To guarantee you will not omit something important, list them
6388 all. A copy of the Makefile (or the output from make) is sufficient.
6390 If we were to try to guess the arguments, we would probably guess wrong
6391 and then we might not encounter the bug.
6394 A complete input file that will reproduce the bug. If the bug is observed when
6395 the assembler is invoked via a compiler, send the assembler source, not the
6396 high level language source. Most compilers will produce the assembler source
6397 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
6398 the options @samp{-v --save-temps}; this will save the assembler source in a
6399 file with an extension of @file{.s}, and also show you exactly how
6400 @command{@value{AS}} is being run.
6403 A description of what behavior you observe that you believe is
6404 incorrect. For example, ``It gets a fatal signal.''
6406 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
6407 will certainly notice it. But if the bug is incorrect output, we might not
6408 notice unless it is glaringly wrong. You might as well not give us a chance to
6411 Even if the problem you experience is a fatal signal, you should still say so
6412 explicitly. Suppose something strange is going on, such as, your copy of
6413 @command{@value{AS}} is out of synch, or you have encountered a bug in the C
6414 library on your system. (This has happened!) Your copy might crash and ours
6415 would not. If you told us to expect a crash, then when ours fails to crash, we
6416 would know that the bug was not happening for us. If you had not told us to
6417 expect a crash, then we would not be able to draw any conclusion from our
6421 If you wish to suggest changes to the @command{@value{AS}} source, send us context
6422 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
6423 option. Always send diffs from the old file to the new file. If you even
6424 discuss something in the @command{@value{AS}} source, refer to it by context, not
6427 The line numbers in our development sources will not match those in your
6428 sources. Your line numbers would convey no useful information to us.
6431 Here are some things that are not necessary:
6435 A description of the envelope of the bug.
6437 Often people who encounter a bug spend a lot of time investigating
6438 which changes to the input file will make the bug go away and which
6439 changes will not affect it.
6441 This is often time consuming and not very useful, because the way we
6442 will find the bug is by running a single example under the debugger
6443 with breakpoints, not by pure deduction from a series of examples.
6444 We recommend that you save your time for something else.
6446 Of course, if you can find a simpler example to report @emph{instead}
6447 of the original one, that is a convenience for us. Errors in the
6448 output will be easier to spot, running under the debugger will take
6449 less time, and so on.
6451 However, simplification is not vital; if you do not want to do this,
6452 report the bug anyway and send us the entire test case you used.
6455 A patch for the bug.
6457 A patch for the bug does help us if it is a good one. But do not omit
6458 the necessary information, such as the test case, on the assumption that
6459 a patch is all we need. We might see problems with your patch and decide
6460 to fix the problem another way, or we might not understand it at all.
6462 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
6463 construct an example that will make the program follow a certain path through
6464 the code. If you do not send us the example, we will not be able to construct
6465 one, so we will not be able to verify that the bug is fixed.
6467 And if we cannot understand what bug you are trying to fix, or why your
6468 patch should be an improvement, we will not install it. A test case will
6469 help us to understand.
6472 A guess about what the bug is or what it depends on.
6474 Such guesses are usually wrong. Even we cannot guess right about such
6475 things without first using the debugger to find the facts.
6478 @node Acknowledgements
6479 @chapter Acknowledgements
6481 If you have contributed to GAS and your name isn't listed here,
6482 it is not meant as a slight. We just don't know about it. Send mail to the
6483 maintainer, and we'll correct the situation. Currently
6485 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
6487 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
6490 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
6491 information and the 68k series machines, most of the preprocessing pass, and
6492 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
6494 K. Richard Pixley maintained GAS for a while, adding various enhancements and
6495 many bug fixes, including merging support for several processors, breaking GAS
6496 up to handle multiple object file format back ends (including heavy rewrite,
6497 testing, an integration of the coff and b.out back ends), adding configuration
6498 including heavy testing and verification of cross assemblers and file splits
6499 and renaming, converted GAS to strictly ANSI C including full prototypes, added
6500 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
6501 port (including considerable amounts of reverse engineering), a SPARC opcode
6502 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
6503 assertions and made them work, much other reorganization, cleanup, and lint.
6505 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
6506 in format-specific I/O modules.
6508 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
6509 has done much work with it since.
6511 The Intel 80386 machine description was written by Eliot Dresselhaus.
6513 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
6515 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
6516 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
6518 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
6519 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
6520 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
6521 support a.out format.
6523 Support for the Zilog Z8k and Renesas H8/300 and H8/500 processors (tc-z8k,
6524 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
6525 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
6526 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
6529 John Gilmore built the AMD 29000 support, added @code{.include} support, and
6530 simplified the configuration of which versions accept which directives. He
6531 updated the 68k machine description so that Motorola's opcodes always produced
6532 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
6533 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
6534 cross-compilation support, and one bug in relaxation that took a week and
6535 required the proverbial one-bit fix.
6537 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
6538 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
6539 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
6540 PowerPC assembler, and made a few other minor patches.
6542 Steve Chamberlain made GAS able to generate listings.
6544 Hewlett-Packard contributed support for the HP9000/300.
6546 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
6547 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
6548 formats). This work was supported by both the Center for Software Science at
6549 the University of Utah and Cygnus Support.
6551 Support for ELF format files has been worked on by Mark Eichin of Cygnus
6552 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
6553 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
6554 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
6555 and some initial 64-bit support).
6557 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
6559 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
6560 support for openVMS/Alpha.
6562 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
6565 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
6566 Inc. added support for Xtensa processors.
6568 Several engineers at Cygnus Support have also provided many small bug fixes and
6569 configuration enhancements.
6571 Many others have contributed large or small bugfixes and enhancements. If
6572 you have contributed significant work and are not mentioned on this list, and
6573 want to be, let us know. Some of the history has been lost; we are not
6574 intentionally leaving anyone out.
projects / deliverable / binutils-gdb.git / blob