1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
3 @c 2001, 2002, 2003, 2004
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{-D}] [@b{--defsym} @var{sym}=@var{val}]
230 [@b{-f}] [@b{--gstabs}] [@b{--gstabs+}] [@b{--gdwarf2}] [@b{--help}]
231 [@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}]
268 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
269 @b{-mapcs-reentrant}]
270 [@b{-mthumb-interwork}] [@b{-moabi}] [@b{-k}]
274 @emph{Target CRIS options:}
275 [@b{--underscore} | @b{--no-underscore}]
277 [@b{--emulation=criself} | @b{--emulation=crisaout}]
278 @c Deprecated -- deliberately not documented.
283 @emph{Target D10V options:}
288 @emph{Target D30V options:}
289 [@b{-O}|@b{-n}|@b{-N}]
292 @c Renesas family chips have no machine-dependent assembler options
295 @c HPPA has no machine-dependent assembler options (yet).
299 @emph{Target i386 options:}
300 [@b{--32}|@b{--64}] [@b{-n}]
304 @emph{Target i960 options:}
305 @c see md_parse_option in tc-i960.c
306 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
308 [@b{-b}] [@b{-no-relax}]
312 @emph{Target IA-64 options:}
313 [@b{-mconstant-gp}|@b{-mauto-pic}]
314 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
316 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
320 @emph{Target IP2K options:}
321 [@b{-mip2022}|@b{-mip2022ext}]
325 @emph{Target M32R options:}
326 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
331 @emph{Target M680X0 options:}
332 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
336 @emph{Target M68HC11 options:}
337 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
338 [@b{-mshort}|@b{-mlong}]
339 [@b{-mshort-double}|@b{-mlong-double}]
340 [@b{--force-long-branchs}] [@b{--short-branchs}]
341 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
342 [@b{--print-opcodes}] [@b{--generate-example}]
346 @emph{Target MCORE options:}
347 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
348 [@b{-mcpu=[210|340]}]
352 @emph{Target MIPS options:}
353 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
354 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
355 [@b{-non_shared}] [@b{-xgot}] [@b{--membedded-pic}]
356 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
357 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
358 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
359 [@b{-mips64}] [@b{-mips64r2}]
360 [@b{-construct-floats}] [@b{-no-construct-floats}]
361 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
362 [@b{-mfix7000}] [@b{-mno-fix7000}]
363 [@b{-mips16}] [@b{-no-mips16}]
364 [@b{-mips3d}] [@b{-no-mips3d}]
365 [@b{-mdmx}] [@b{-no-mdmx}]
366 [@b{-mdebug}] [@b{-no-mdebug}]
367 [@b{-mpdr}] [@b{-mno-pdr}]
371 @emph{Target MMIX options:}
372 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
373 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
374 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
375 [@b{--linker-allocated-gregs}]
379 @emph{Target PDP11 options:}
380 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
381 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
382 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
386 @emph{Target picoJava options:}
391 @emph{Target PowerPC options:}
392 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
393 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}|
394 @b{-mbooke32}|@b{-mbooke64}]
395 [@b{-mcom}|@b{-many}|@b{-maltivec}] [@b{-memb}]
396 [@b{-mregnames}|@b{-mno-regnames}]
397 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
398 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
399 [@b{-msolaris}|@b{-mno-solaris}]
403 @emph{Target SPARC options:}
404 @c The order here is important. See c-sparc.texi.
405 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
406 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
407 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
412 @emph{Target TIC54X options:}
413 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
414 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
417 @c Z8000 has no machine-dependent assembler options
421 @emph{Target Xtensa options:}
422 [@b{--[no-]density}] [@b{--[no-]relax}] [@b{--[no-]generics}]
423 [@b{--[no-]text-section-literals}]
424 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
433 Turn on listings, in any of a variety of ways:
437 omit false conditionals
440 omit debugging directives
443 include high-level source
449 include macro expansions
452 omit forms processing
458 set the name of the listing file
461 You may combine these options; for example, use @samp{-aln} for assembly
462 listing without forms processing. The @samp{=file} option, if used, must be
463 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
466 Ignored. This option is accepted for script compatibility with calls to
469 @item --defsym @var{sym}=@var{value}
470 Define the symbol @var{sym} to be @var{value} before assembling the input file.
471 @var{value} must be an integer constant. As in C, a leading @samp{0x}
472 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
475 ``fast''---skip whitespace and comment preprocessing (assume source is
479 Generate stabs debugging information for each assembler line. This
480 may help debugging assembler code, if the debugger can handle it.
483 Generate stabs debugging information for each assembler line, with GNU
484 extensions that probably only gdb can handle, and that could make other
485 debuggers crash or refuse to read your program. This
486 may help debugging assembler code. Currently the only GNU extension is
487 the location of the current working directory at assembling time.
490 Generate DWARF2 debugging information for each assembler line. This
491 may help debugging assembler code, if the debugger can handle it. Note---this
492 option is only supported by some targets, not all of them.
495 Print a summary of the command line options and exit.
498 Print a summary of all target specific options and exit.
501 Add directory @var{dir} to the search list for @code{.include} directives.
504 Don't warn about signed overflow.
507 @ifclear DIFF-TBL-KLUGE
508 This option is accepted but has no effect on the @value{TARGET} family.
510 @ifset DIFF-TBL-KLUGE
511 Issue warnings when difference tables altered for long displacements.
516 Keep (in the symbol table) local symbols. On traditional a.out systems
517 these start with @samp{L}, but different systems have different local
520 @item --listing-lhs-width=@var{number}
521 Set the maximum width, in words, of the output data column for an assembler
522 listing to @var{number}.
524 @item --listing-lhs-width2=@var{number}
525 Set the maximum width, in words, of the output data column for continuation
526 lines in an assembler listing to @var{number}.
528 @item --listing-rhs-width=@var{number}
529 Set the maximum width of an input source line, as displayed in a listing, to
532 @item --listing-cont-lines=@var{number}
533 Set the maximum number of lines printed in a listing for a single line of input
536 @item -o @var{objfile}
537 Name the object-file output from @command{@value{AS}} @var{objfile}.
540 Fold the data section into the text section.
543 Print the maximum space (in bytes) and total time (in seconds) used by
546 @item --strip-local-absolute
547 Remove local absolute symbols from the outgoing symbol table.
551 Print the @command{as} version.
554 Print the @command{as} version and exit.
558 Suppress warning messages.
560 @item --fatal-warnings
561 Treat warnings as errors.
564 Don't suppress warning messages or treat them as errors.
573 Generate an object file even after errors.
575 @item -- | @var{files} @dots{}
576 Standard input, or source files to assemble.
581 The following options are available when @value{AS} is configured for
586 This option selects the core processor variant.
588 Select either big-endian (-EB) or little-endian (-EL) output.
593 The following options are available when @value{AS} is configured for the ARM
597 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
598 Specify which ARM processor variant is the target.
599 @item -march=@var{architecture}[+@var{extension}@dots{}]
600 Specify which ARM architecture variant is used by the target.
601 @item -mfpu=@var{floating-point-format}
602 Select which Floating Point architecture is the target.
603 @item -mfloat-abi=@var{abi}
604 Select which floating point ABI is in use.
606 Enable Thumb only instruction decoding.
607 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant | -moabi
608 Select which procedure calling convention is in use.
610 Select either big-endian (-EB) or little-endian (-EL) output.
611 @item -mthumb-interwork
612 Specify that the code has been generated with interworking between Thumb and
615 Specify that PIC code has been generated.
620 See the info pages for documentation of the CRIS-specific options.
624 The following options are available when @value{AS} is configured for
627 @cindex D10V optimization
628 @cindex optimization, D10V
630 Optimize output by parallelizing instructions.
635 The following options are available when @value{AS} is configured for a D30V
638 @cindex D30V optimization
639 @cindex optimization, D30V
641 Optimize output by parallelizing instructions.
645 Warn when nops are generated.
647 @cindex D30V nops after 32-bit multiply
649 Warn when a nop after a 32-bit multiply instruction is generated.
654 The following options are available when @value{AS} is configured for the
655 Intel 80960 processor.
658 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
659 Specify which variant of the 960 architecture is the target.
662 Add code to collect statistics about branches taken.
665 Do not alter compare-and-branch instructions for long displacements;
672 The following options are available when @value{AS} is configured for the
678 Specifies that the extended IP2022 instructions are allowed.
681 Restores the default behaviour, which restricts the permitted instructions to
682 just the basic IP2022 ones.
688 The following options are available when @value{AS} is configured for the
689 Renesas M32R (formerly Mitsubishi M32R) series.
694 Specify which processor in the M32R family is the target. The default
695 is normally the M32R, but this option changes it to the M32RX.
697 @item --warn-explicit-parallel-conflicts or --Wp
698 Produce warning messages when questionable parallel constructs are
701 @item --no-warn-explicit-parallel-conflicts or --Wnp
702 Do not produce warning messages when questionable parallel constructs are
709 The following options are available when @value{AS} is configured for the
710 Motorola 68000 series.
715 Shorten references to undefined symbols, to one word instead of two.
717 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
718 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
719 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
720 Specify what processor in the 68000 family is the target. The default
721 is normally the 68020, but this can be changed at configuration time.
723 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
724 The target machine does (or does not) have a floating-point coprocessor.
725 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
726 the basic 68000 is not compatible with the 68881, a combination of the
727 two can be specified, since it's possible to do emulation of the
728 coprocessor instructions with the main processor.
730 @item -m68851 | -mno-68851
731 The target machine does (or does not) have a memory-management
732 unit coprocessor. The default is to assume an MMU for 68020 and up.
739 For details about the PDP-11 machine dependent features options,
740 see @ref{PDP-11-Options}.
743 @item -mpic | -mno-pic
744 Generate position-independent (or position-dependent) code. The
745 default is @option{-mpic}.
748 @itemx -mall-extensions
749 Enable all instruction set extensions. This is the default.
751 @item -mno-extensions
752 Disable all instruction set extensions.
754 @item -m@var{extension} | -mno-@var{extension}
755 Enable (or disable) a particular instruction set extension.
758 Enable the instruction set extensions supported by a particular CPU, and
759 disable all other extensions.
761 @item -m@var{machine}
762 Enable the instruction set extensions supported by a particular machine
763 model, and disable all other extensions.
769 The following options are available when @value{AS} is configured for
770 a picoJava processor.
774 @cindex PJ endianness
775 @cindex endianness, PJ
776 @cindex big endian output, PJ
778 Generate ``big endian'' format output.
780 @cindex little endian output, PJ
782 Generate ``little endian'' format output.
788 The following options are available when @value{AS} is configured for the
789 Motorola 68HC11 or 68HC12 series.
793 @item -m68hc11 | -m68hc12 | -m68hcs12
794 Specify what processor is the target. The default is
795 defined by the configuration option when building the assembler.
798 Specify to use the 16-bit integer ABI.
801 Specify to use the 32-bit integer ABI.
804 Specify to use the 32-bit double ABI.
807 Specify to use the 64-bit double ABI.
809 @item --force-long-branchs
810 Relative branches are turned into absolute ones. This concerns
811 conditional branches, unconditional branches and branches to a
814 @item -S | --short-branchs
815 Do not turn relative branchs into absolute ones
816 when the offset is out of range.
818 @item --strict-direct-mode
819 Do not turn the direct addressing mode into extended addressing mode
820 when the instruction does not support direct addressing mode.
822 @item --print-insn-syntax
823 Print the syntax of instruction in case of error.
825 @item --print-opcodes
826 print the list of instructions with syntax and then exit.
828 @item --generate-example
829 print an example of instruction for each possible instruction and then exit.
830 This option is only useful for testing @command{@value{AS}}.
836 The following options are available when @command{@value{AS}} is configured
837 for the SPARC architecture:
840 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
841 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
842 Explicitly select a variant of the SPARC architecture.
844 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
845 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
847 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
848 UltraSPARC extensions.
850 @item -xarch=v8plus | -xarch=v8plusa
851 For compatibility with the Solaris v9 assembler. These options are
852 equivalent to -Av8plus and -Av8plusa, respectively.
855 Warn when the assembler switches to another architecture.
860 The following options are available when @value{AS} is configured for the 'c54x
865 Enable extended addressing mode. All addresses and relocations will assume
866 extended addressing (usually 23 bits).
867 @item -mcpu=@var{CPU_VERSION}
868 Sets the CPU version being compiled for.
869 @item -merrors-to-file @var{FILENAME}
870 Redirect error output to a file, for broken systems which don't support such
871 behaviour in the shell.
876 The following options are available when @value{AS} is configured for
877 a @sc{mips} processor.
881 This option sets the largest size of an object that can be referenced
882 implicitly with the @code{gp} register. It is only accepted for targets that
883 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
885 @cindex MIPS endianness
886 @cindex endianness, MIPS
887 @cindex big endian output, MIPS
889 Generate ``big endian'' format output.
891 @cindex little endian output, MIPS
893 Generate ``little endian'' format output.
905 Generate code for a particular @sc{mips} Instruction Set Architecture level.
906 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
907 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
908 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
909 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
911 correspond to generic
912 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
913 and @samp{MIPS64 Release 2}
914 ISA processors, respectively.
916 @item -march=@var{CPU}
917 Generate code for a particular @sc{mips} cpu.
919 @item -mtune=@var{cpu}
920 Schedule and tune for a particular @sc{mips} cpu.
924 Cause nops to be inserted if the read of the destination register
925 of an mfhi or mflo instruction occurs in the following two instructions.
929 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
930 section instead of the standard ELF .stabs sections.
934 Control generation of @code{.pdr} sections.
938 The register sizes are normally inferred from the ISA and ABI, but these
939 flags force a certain group of registers to be treated as 32 bits wide at
940 all times. @samp{-mgp32} controls the size of general-purpose registers
941 and @samp{-mfp32} controls the size of floating-point registers.
945 Generate code for the MIPS 16 processor. This is equivalent to putting
946 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
947 turns off this option.
951 Generate code for the MIPS-3D Application Specific Extension.
952 This tells the assembler to accept MIPS-3D instructions.
953 @samp{-no-mips3d} turns off this option.
957 Generate code for the MDMX Application Specific Extension.
958 This tells the assembler to accept MDMX instructions.
959 @samp{-no-mdmx} turns off this option.
961 @item --construct-floats
962 @itemx --no-construct-floats
963 The @samp{--no-construct-floats} option disables the construction of
964 double width floating point constants by loading the two halves of the
965 value into the two single width floating point registers that make up
966 the double width register. By default @samp{--construct-floats} is
967 selected, allowing construction of these floating point constants.
970 @item --emulation=@var{name}
971 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
972 for some other target, in all respects, including output format (choosing
973 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
974 debugging information or store symbol table information, and default
975 endianness. The available configuration names are: @samp{mipsecoff},
976 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
977 @samp{mipsbelf}. The first two do not alter the default endianness from that
978 of the primary target for which the assembler was configured; the others change
979 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
980 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
981 selection in any case.
983 This option is currently supported only when the primary target
984 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
985 Furthermore, the primary target or others specified with
986 @samp{--enable-targets=@dots{}} at configuration time must include support for
987 the other format, if both are to be available. For example, the Irix 5
988 configuration includes support for both.
990 Eventually, this option will support more configurations, with more
991 fine-grained control over the assembler's behavior, and will be supported for
995 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1002 Control how to deal with multiplication overflow and division by zero.
1003 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1004 (and only work for Instruction Set Architecture level 2 and higher);
1005 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1009 When this option is used, @command{@value{AS}} will issue a warning every
1010 time it generates a nop instruction from a macro.
1015 The following options are available when @value{AS} is configured for
1021 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1022 The command line option @samp{-nojsri2bsr} can be used to disable it.
1026 Enable or disable the silicon filter behaviour. By default this is disabled.
1027 The default can be overridden by the @samp{-sifilter} command line option.
1030 Alter jump instructions for long displacements.
1032 @item -mcpu=[210|340]
1033 Select the cpu type on the target hardware. This controls which instructions
1037 Assemble for a big endian target.
1040 Assemble for a little endian target.
1046 See the info pages for documentation of the MMIX-specific options.
1050 The following options are available when @value{AS} is configured for
1051 an Xtensa processor.
1054 @item --density | --no-density
1055 Enable or disable use of instructions from the Xtensa code density
1056 option. This is enabled by default when the Xtensa processor supports
1057 the code density option.
1059 @item --relax | --no-relax
1060 Enable or disable instruction relaxation. This is enabled by default.
1061 Note: In the current implementation, these options also control whether
1062 assembler optimizations are performed, making these options equivalent
1063 to @option{--generics} and @option{--no-generics}.
1065 @item --generics | --no-generics
1066 Enable or disable all assembler transformations of Xtensa instructions.
1067 The default is @option{--generics};
1068 @option{--no-generics} should be used only in the rare cases when the
1069 instructions must be exactly as specified in the assembly source.
1071 @item --text-section-literals | --no-text-section-literals
1072 With @option{--text-@-section-@-literals}, literal pools are interspersed
1073 in the text section. The default is
1074 @option{--no-@-text-@-section-@-literals}, which places literals in a
1075 separate section in the output file.
1077 @item --target-align | --no-target-align
1078 Enable or disable automatic alignment to reduce branch penalties at the
1079 expense of some code density. The default is @option{--target-@-align}.
1081 @item --longcalls | --no-longcalls
1082 Enable or disable transformation of call instructions to allow calls
1083 across a greater range of addresses. The default is
1084 @option{--no-@-longcalls}.
1091 * Manual:: Structure of this Manual
1092 * GNU Assembler:: The GNU Assembler
1093 * Object Formats:: Object File Formats
1094 * Command Line:: Command Line
1095 * Input Files:: Input Files
1096 * Object:: Output (Object) File
1097 * Errors:: Error and Warning Messages
1101 @section Structure of this Manual
1103 @cindex manual, structure and purpose
1104 This manual is intended to describe what you need to know to use
1105 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1106 notation for symbols, constants, and expressions; the directives that
1107 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1110 We also cover special features in the @value{TARGET}
1111 configuration of @command{@value{AS}}, including assembler directives.
1114 This manual also describes some of the machine-dependent features of
1115 various flavors of the assembler.
1118 @cindex machine instructions (not covered)
1119 On the other hand, this manual is @emph{not} intended as an introduction
1120 to programming in assembly language---let alone programming in general!
1121 In a similar vein, we make no attempt to introduce the machine
1122 architecture; we do @emph{not} describe the instruction set, standard
1123 mnemonics, registers or addressing modes that are standard to a
1124 particular architecture.
1126 You may want to consult the manufacturer's
1127 machine architecture manual for this information.
1131 For information on the H8/300 machine instruction set, see @cite{H8/300
1132 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1133 Programming Manual} (Renesas).
1136 For information on the H8/500 machine instruction set, see @cite{H8/500
1137 Series Programming Manual} (Renesas M21T001).
1140 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1141 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1142 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1143 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1146 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1150 @c I think this is premature---doc@cygnus.com, 17jan1991
1152 Throughout this manual, we assume that you are running @dfn{GNU},
1153 the portable operating system from the @dfn{Free Software
1154 Foundation, Inc.}. This restricts our attention to certain kinds of
1155 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1156 once this assumption is granted examples and definitions need less
1159 @command{@value{AS}} is part of a team of programs that turn a high-level
1160 human-readable series of instructions into a low-level
1161 computer-readable series of instructions. Different versions of
1162 @command{@value{AS}} are used for different kinds of computer.
1165 @c There used to be a section "Terminology" here, which defined
1166 @c "contents", "byte", "word", and "long". Defining "word" to any
1167 @c particular size is confusing when the .word directive may generate 16
1168 @c bits on one machine and 32 bits on another; in general, for the user
1169 @c version of this manual, none of these terms seem essential to define.
1170 @c They were used very little even in the former draft of the manual;
1171 @c this draft makes an effort to avoid them (except in names of
1175 @section The GNU Assembler
1177 @c man begin DESCRIPTION
1179 @sc{gnu} @command{as} is really a family of assemblers.
1181 This manual describes @command{@value{AS}}, a member of that family which is
1182 configured for the @value{TARGET} architectures.
1184 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1185 should find a fairly similar environment when you use it on another
1186 architecture. Each version has much in common with the others,
1187 including object file formats, most assembler directives (often called
1188 @dfn{pseudo-ops}) and assembler syntax.@refill
1190 @cindex purpose of @sc{gnu} assembler
1191 @command{@value{AS}} is primarily intended to assemble the output of the
1192 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1193 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1194 assemble correctly everything that other assemblers for the same
1195 machine would assemble.
1197 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1200 @c This remark should appear in generic version of manual; assumption
1201 @c here is that generic version sets M680x0.
1202 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1203 assembler for the same architecture; for example, we know of several
1204 incompatible versions of 680x0 assembly language syntax.
1209 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1210 program in one pass of the source file. This has a subtle impact on the
1211 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1213 @node Object Formats
1214 @section Object File Formats
1216 @cindex object file format
1217 The @sc{gnu} assembler can be configured to produce several alternative
1218 object file formats. For the most part, this does not affect how you
1219 write assembly language programs; but directives for debugging symbols
1220 are typically different in different file formats. @xref{Symbol
1221 Attributes,,Symbol Attributes}.
1224 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1225 @value{OBJ-NAME} format object files.
1227 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1229 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1230 @code{a.out} or COFF format object files.
1233 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1234 @code{b.out} or COFF format object files.
1237 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1238 SOM or ELF format object files.
1243 @section Command Line
1245 @cindex command line conventions
1247 After the program name @command{@value{AS}}, the command line may contain
1248 options and file names. Options may appear in any order, and may be
1249 before, after, or between file names. The order of file names is
1252 @cindex standard input, as input file
1254 @file{--} (two hyphens) by itself names the standard input file
1255 explicitly, as one of the files for @command{@value{AS}} to assemble.
1257 @cindex options, command line
1258 Except for @samp{--} any command line argument that begins with a
1259 hyphen (@samp{-}) is an option. Each option changes the behavior of
1260 @command{@value{AS}}. No option changes the way another option works. An
1261 option is a @samp{-} followed by one or more letters; the case of
1262 the letter is important. All options are optional.
1264 Some options expect exactly one file name to follow them. The file
1265 name may either immediately follow the option's letter (compatible
1266 with older assemblers) or it may be the next command argument (@sc{gnu}
1267 standard). These two command lines are equivalent:
1270 @value{AS} -o my-object-file.o mumble.s
1271 @value{AS} -omy-object-file.o mumble.s
1275 @section Input Files
1278 @cindex source program
1279 @cindex files, input
1280 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1281 describe the program input to one run of @command{@value{AS}}. The program may
1282 be in one or more files; how the source is partitioned into files
1283 doesn't change the meaning of the source.
1285 @c I added "con" prefix to "catenation" just to prove I can overcome my
1286 @c APL training... doc@cygnus.com
1287 The source program is a concatenation of the text in all the files, in the
1290 @c man begin DESCRIPTION
1291 Each time you run @command{@value{AS}} it assembles exactly one source
1292 program. The source program is made up of one or more files.
1293 (The standard input is also a file.)
1295 You give @command{@value{AS}} a command line that has zero or more input file
1296 names. The input files are read (from left file name to right). A
1297 command line argument (in any position) that has no special meaning
1298 is taken to be an input file name.
1300 If you give @command{@value{AS}} no file names it attempts to read one input file
1301 from the @command{@value{AS}} standard input, which is normally your terminal. You
1302 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1305 Use @samp{--} if you need to explicitly name the standard input file
1306 in your command line.
1308 If the source is empty, @command{@value{AS}} produces a small, empty object
1313 @subheading Filenames and Line-numbers
1315 @cindex input file linenumbers
1316 @cindex line numbers, in input files
1317 There are two ways of locating a line in the input file (or files) and
1318 either may be used in reporting error messages. One way refers to a line
1319 number in a physical file; the other refers to a line number in a
1320 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1322 @dfn{Physical files} are those files named in the command line given
1323 to @command{@value{AS}}.
1325 @dfn{Logical files} are simply names declared explicitly by assembler
1326 directives; they bear no relation to physical files. Logical file names help
1327 error messages reflect the original source file, when @command{@value{AS}} source
1328 is itself synthesized from other files. @command{@value{AS}} understands the
1329 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1330 @ref{File,,@code{.file}}.
1333 @section Output (Object) File
1339 Every time you run @command{@value{AS}} it produces an output file, which is
1340 your assembly language program translated into numbers. This file
1341 is the object file. Its default name is
1349 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1351 You can give it another name by using the @option{-o} option. Conventionally,
1352 object file names end with @file{.o}. The default name is used for historical
1353 reasons: older assemblers were capable of assembling self-contained programs
1354 directly into a runnable program. (For some formats, this isn't currently
1355 possible, but it can be done for the @code{a.out} format.)
1359 The object file is meant for input to the linker @code{@value{LD}}. It contains
1360 assembled program code, information to help @code{@value{LD}} integrate
1361 the assembled program into a runnable file, and (optionally) symbolic
1362 information for the debugger.
1364 @c link above to some info file(s) like the description of a.out.
1365 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1368 @section Error and Warning Messages
1370 @c man begin DESCRIPTION
1372 @cindex error messages
1373 @cindex warning messages
1374 @cindex messages from assembler
1375 @command{@value{AS}} may write warnings and error messages to the standard error
1376 file (usually your terminal). This should not happen when a compiler
1377 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1378 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1379 grave problem that stops the assembly.
1383 @cindex format of warning messages
1384 Warning messages have the format
1387 file_name:@b{NNN}:Warning Message Text
1391 @cindex line numbers, in warnings/errors
1392 (where @b{NNN} is a line number). If a logical file name has been given
1393 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1394 the current input file is used. If a logical line number was given
1396 (@pxref{Line,,@code{.line}})
1400 (@pxref{Line,,@code{.line}})
1403 (@pxref{Ln,,@code{.ln}})
1406 then it is used to calculate the number printed,
1407 otherwise the actual line in the current source file is printed. The
1408 message text is intended to be self explanatory (in the grand Unix
1411 @cindex format of error messages
1412 Error messages have the format
1414 file_name:@b{NNN}:FATAL:Error Message Text
1416 The file name and line number are derived as for warning
1417 messages. The actual message text may be rather less explanatory
1418 because many of them aren't supposed to happen.
1421 @chapter Command-Line Options
1423 @cindex options, all versions of assembler
1424 This chapter describes command-line options available in @emph{all}
1425 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
1427 to the @value{TARGET} target.
1430 to particular machine architectures.
1433 @c man begin DESCRIPTION
1435 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1436 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1437 The assembler arguments must be separated from each other (and the @samp{-Wa})
1438 by commas. For example:
1441 gcc -c -g -O -Wa,-alh,-L file.c
1445 This passes two options to the assembler: @samp{-alh} (emit a listing to
1446 standard output with high-level and assembly source) and @samp{-L} (retain
1447 local symbols in the symbol table).
1449 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1450 command-line options are automatically passed to the assembler by the compiler.
1451 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1452 precisely what options it passes to each compilation pass, including the
1458 * a:: -a[cdhlns] enable listings
1459 * D:: -D for compatibility
1460 * f:: -f to work faster
1461 * I:: -I for .include search path
1462 @ifclear DIFF-TBL-KLUGE
1463 * K:: -K for compatibility
1465 @ifset DIFF-TBL-KLUGE
1466 * K:: -K for difference tables
1469 * L:: -L to retain local labels
1470 * listing:: --listing-XXX to configure listing output
1471 * M:: -M or --mri to assemble in MRI compatibility mode
1472 * MD:: --MD for dependency tracking
1473 * o:: -o to name the object file
1474 * R:: -R to join data and text sections
1475 * statistics:: --statistics to see statistics about assembly
1476 * traditional-format:: --traditional-format for compatible output
1477 * v:: -v to announce version
1478 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1479 * Z:: -Z to make object file even after errors
1483 @section Enable Listings: @option{-a[cdhlns]}
1492 @cindex listings, enabling
1493 @cindex assembly listings, enabling
1495 These options enable listing output from the assembler. By itself,
1496 @samp{-a} requests high-level, assembly, and symbols listing.
1497 You can use other letters to select specific options for the list:
1498 @samp{-ah} requests a high-level language listing,
1499 @samp{-al} requests an output-program assembly listing, and
1500 @samp{-as} requests a symbol table listing.
1501 High-level listings require that a compiler debugging option like
1502 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1505 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1506 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1507 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1508 omitted from the listing.
1510 Use the @samp{-ad} option to omit debugging directives from the
1513 Once you have specified one of these options, you can further control
1514 listing output and its appearance using the directives @code{.list},
1515 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1517 The @samp{-an} option turns off all forms processing.
1518 If you do not request listing output with one of the @samp{-a} options, the
1519 listing-control directives have no effect.
1521 The letters after @samp{-a} may be combined into one option,
1522 @emph{e.g.}, @samp{-aln}.
1524 Note if the assembler source is coming from the standard input (eg because it
1525 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1526 is being used) then the listing will not contain any comments or preprocessor
1527 directives. This is because the listing code buffers input source lines from
1528 stdin only after they have been preprocessed by the assembler. This reduces
1529 memory usage and makes the code more efficient.
1532 @section @option{-D}
1535 This option has no effect whatsoever, but it is accepted to make it more
1536 likely that scripts written for other assemblers also work with
1537 @command{@value{AS}}.
1540 @section Work Faster: @option{-f}
1543 @cindex trusted compiler
1544 @cindex faster processing (@option{-f})
1545 @samp{-f} should only be used when assembling programs written by a
1546 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1547 and comment preprocessing on
1548 the input file(s) before assembling them. @xref{Preprocessing,
1552 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1553 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1558 @section @code{.include} Search Path: @option{-I} @var{path}
1560 @kindex -I @var{path}
1561 @cindex paths for @code{.include}
1562 @cindex search path for @code{.include}
1563 @cindex @code{include} directive search path
1564 Use this option to add a @var{path} to the list of directories
1565 @command{@value{AS}} searches for files specified in @code{.include}
1566 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1567 many times as necessary to include a variety of paths. The current
1568 working directory is always searched first; after that, @command{@value{AS}}
1569 searches any @samp{-I} directories in the same order as they were
1570 specified (left to right) on the command line.
1573 @section Difference Tables: @option{-K}
1576 @ifclear DIFF-TBL-KLUGE
1577 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1578 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1579 where it can be used to warn when the assembler alters the machine code
1580 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1581 family does not have the addressing limitations that sometimes lead to this
1582 alteration on other platforms.
1585 @ifset DIFF-TBL-KLUGE
1586 @cindex difference tables, warning
1587 @cindex warning for altered difference tables
1588 @command{@value{AS}} sometimes alters the code emitted for directives of the form
1589 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1590 You can use the @samp{-K} option if you want a warning issued when this
1595 @section Include Local Labels: @option{-L}
1598 @cindex local labels, retaining in output
1599 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1600 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1601 debugging, because they are intended for the use of programs (like
1602 compilers) that compose assembler programs, not for your notice.
1603 Normally both @command{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1604 normally debug with them.
1606 This option tells @command{@value{AS}} to retain those @samp{L@dots{}} symbols
1607 in the object file. Usually if you do this you also tell the linker
1608 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1610 By default, a local label is any label beginning with @samp{L}, but each
1611 target is allowed to redefine the local label prefix.
1613 On the HPPA local labels begin with @samp{L$}.
1617 @section Configuring listing output: @option{--listing}
1619 The listing feature of the assembler can be enabled via the command line switch
1620 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1621 hex dump of the corresponding locations in the output object file, and displays
1622 them as a listing file. The format of this listing can be controlled by pseudo
1623 ops inside the assembler source (@pxref{List} @pxref{Title} @pxref{Sbttl}
1624 @pxref{Psize} @pxref{Eject}) and also by the following switches:
1627 @item --listing-lhs-width=@samp{number}
1628 @kindex --listing-lhs-width
1629 @cindex Width of first line disassembly output
1630 Sets the maximum width, in words, of the first line of the hex byte dump. This
1631 dump appears on the left hand side of the listing output.
1633 @item --listing-lhs-width2=@samp{number}
1634 @kindex --listing-lhs-width2
1635 @cindex Width of continuation lines of disassembly output
1636 Sets the maximum width, in words, of any further lines of the hex byte dump for
1637 a given input source line. If this value is not specified, it defaults to being
1638 the same as the value specified for @samp{--listing-lhs-width}. If neither
1639 switch is used the default is to one.
1641 @item --listing-rhs-width=@samp{number}
1642 @kindex --listing-rhs-width
1643 @cindex Width of source line output
1644 Sets the maximum width, in characters, of the source line that is displayed
1645 alongside the hex dump. The default value for this parameter is 100. The
1646 source line is displayed on the right hand side of the listing output.
1648 @item --listing-cont-lines=@samp{number}
1649 @kindex --listing-cont-lines
1650 @cindex Maximum number of continuation lines
1651 Sets the maximum number of continuation lines of hex dump that will be
1652 displayed for a given single line of source input. The default value is 4.
1656 @section Assemble in MRI Compatibility Mode: @option{-M}
1659 @cindex MRI compatibility mode
1660 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1661 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1662 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1663 configured target) assembler from Microtec Research. The exact nature of the
1664 MRI syntax will not be documented here; see the MRI manuals for more
1665 information. Note in particular that the handling of macros and macro
1666 arguments is somewhat different. The purpose of this option is to permit
1667 assembling existing MRI assembler code using @command{@value{AS}}.
1669 The MRI compatibility is not complete. Certain operations of the MRI assembler
1670 depend upon its object file format, and can not be supported using other object
1671 file formats. Supporting these would require enhancing each object file format
1672 individually. These are:
1675 @item global symbols in common section
1677 The m68k MRI assembler supports common sections which are merged by the linker.
1678 Other object file formats do not support this. @command{@value{AS}} handles
1679 common sections by treating them as a single common symbol. It permits local
1680 symbols to be defined within a common section, but it can not support global
1681 symbols, since it has no way to describe them.
1683 @item complex relocations
1685 The MRI assemblers support relocations against a negated section address, and
1686 relocations which combine the start addresses of two or more sections. These
1687 are not support by other object file formats.
1689 @item @code{END} pseudo-op specifying start address
1691 The MRI @code{END} pseudo-op permits the specification of a start address.
1692 This is not supported by other object file formats. The start address may
1693 instead be specified using the @option{-e} option to the linker, or in a linker
1696 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1698 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1699 name to the output file. This is not supported by other object file formats.
1701 @item @code{ORG} pseudo-op
1703 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1704 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1705 which changes the location within the current section. Absolute sections are
1706 not supported by other object file formats. The address of a section may be
1707 assigned within a linker script.
1710 There are some other features of the MRI assembler which are not supported by
1711 @command{@value{AS}}, typically either because they are difficult or because they
1712 seem of little consequence. Some of these may be supported in future releases.
1716 @item EBCDIC strings
1718 EBCDIC strings are not supported.
1720 @item packed binary coded decimal
1722 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1723 and @code{DCB.P} pseudo-ops are not supported.
1725 @item @code{FEQU} pseudo-op
1727 The m68k @code{FEQU} pseudo-op is not supported.
1729 @item @code{NOOBJ} pseudo-op
1731 The m68k @code{NOOBJ} pseudo-op is not supported.
1733 @item @code{OPT} branch control options
1735 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1736 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1737 relaxes all branches, whether forward or backward, to an appropriate size, so
1738 these options serve no purpose.
1740 @item @code{OPT} list control options
1742 The following m68k @code{OPT} list control options are ignored: @code{C},
1743 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1744 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1746 @item other @code{OPT} options
1748 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1749 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1751 @item @code{OPT} @code{D} option is default
1753 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1754 @code{OPT NOD} may be used to turn it off.
1756 @item @code{XREF} pseudo-op.
1758 The m68k @code{XREF} pseudo-op is ignored.
1760 @item @code{.debug} pseudo-op
1762 The i960 @code{.debug} pseudo-op is not supported.
1764 @item @code{.extended} pseudo-op
1766 The i960 @code{.extended} pseudo-op is not supported.
1768 @item @code{.list} pseudo-op.
1770 The various options of the i960 @code{.list} pseudo-op are not supported.
1772 @item @code{.optimize} pseudo-op
1774 The i960 @code{.optimize} pseudo-op is not supported.
1776 @item @code{.output} pseudo-op
1778 The i960 @code{.output} pseudo-op is not supported.
1780 @item @code{.setreal} pseudo-op
1782 The i960 @code{.setreal} pseudo-op is not supported.
1787 @section Dependency Tracking: @option{--MD}
1790 @cindex dependency tracking
1793 @command{@value{AS}} can generate a dependency file for the file it creates. This
1794 file consists of a single rule suitable for @code{make} describing the
1795 dependencies of the main source file.
1797 The rule is written to the file named in its argument.
1799 This feature is used in the automatic updating of makefiles.
1802 @section Name the Object File: @option{-o}
1805 @cindex naming object file
1806 @cindex object file name
1807 There is always one object file output when you run @command{@value{AS}}. By
1808 default it has the name
1811 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1825 You use this option (which takes exactly one filename) to give the
1826 object file a different name.
1828 Whatever the object file is called, @command{@value{AS}} overwrites any
1829 existing file of the same name.
1832 @section Join Data and Text Sections: @option{-R}
1835 @cindex data and text sections, joining
1836 @cindex text and data sections, joining
1837 @cindex joining text and data sections
1838 @cindex merging text and data sections
1839 @option{-R} tells @command{@value{AS}} to write the object file as if all
1840 data-section data lives in the text section. This is only done at
1841 the very last moment: your binary data are the same, but data
1842 section parts are relocated differently. The data section part of
1843 your object file is zero bytes long because all its bytes are
1844 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1846 When you specify @option{-R} it would be possible to generate shorter
1847 address displacements (because we do not have to cross between text and
1848 data section). We refrain from doing this simply for compatibility with
1849 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
1852 When @command{@value{AS}} is configured for COFF or ELF output,
1853 this option is only useful if you use sections named @samp{.text} and
1858 @option{-R} is not supported for any of the HPPA targets. Using
1859 @option{-R} generates a warning from @command{@value{AS}}.
1863 @section Display Assembly Statistics: @option{--statistics}
1865 @kindex --statistics
1866 @cindex statistics, about assembly
1867 @cindex time, total for assembly
1868 @cindex space used, maximum for assembly
1869 Use @samp{--statistics} to display two statistics about the resources used by
1870 @command{@value{AS}}: the maximum amount of space allocated during the assembly
1871 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1874 @node traditional-format
1875 @section Compatible Output: @option{--traditional-format}
1877 @kindex --traditional-format
1878 For some targets, the output of @command{@value{AS}} is different in some ways
1879 from the output of some existing assembler. This switch requests
1880 @command{@value{AS}} to use the traditional format instead.
1882 For example, it disables the exception frame optimizations which
1883 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
1886 @section Announce Version: @option{-v}
1890 @cindex assembler version
1891 @cindex version of assembler
1892 You can find out what version of as is running by including the
1893 option @samp{-v} (which you can also spell as @samp{-version}) on the
1897 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
1899 @command{@value{AS}} should never give a warning or error message when
1900 assembling compiler output. But programs written by people often
1901 cause @command{@value{AS}} to give a warning that a particular assumption was
1902 made. All such warnings are directed to the standard error file.
1906 @cindex suppressing warnings
1907 @cindex warnings, suppressing
1908 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
1909 This only affects the warning messages: it does not change any particular of
1910 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
1913 @kindex --fatal-warnings
1914 @cindex errors, caused by warnings
1915 @cindex warnings, causing error
1916 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
1917 files that generate warnings to be in error.
1920 @cindex warnings, switching on
1921 You can switch these options off again by specifying @option{--warn}, which
1922 causes warnings to be output as usual.
1925 @section Generate Object File in Spite of Errors: @option{-Z}
1926 @cindex object file, after errors
1927 @cindex errors, continuing after
1928 After an error message, @command{@value{AS}} normally produces no output. If for
1929 some reason you are interested in object file output even after
1930 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
1931 option. If there are any errors, @command{@value{AS}} continues anyways, and
1932 writes an object file after a final warning message of the form @samp{@var{n}
1933 errors, @var{m} warnings, generating bad object file.}
1938 @cindex machine-independent syntax
1939 @cindex syntax, machine-independent
1940 This chapter describes the machine-independent syntax allowed in a
1941 source file. @command{@value{AS}} syntax is similar to what many other
1942 assemblers use; it is inspired by the BSD 4.2
1947 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
1951 * Preprocessing:: Preprocessing
1952 * Whitespace:: Whitespace
1953 * Comments:: Comments
1954 * Symbol Intro:: Symbols
1955 * Statements:: Statements
1956 * Constants:: Constants
1960 @section Preprocessing
1962 @cindex preprocessing
1963 The @command{@value{AS}} internal preprocessor:
1965 @cindex whitespace, removed by preprocessor
1967 adjusts and removes extra whitespace. It leaves one space or tab before
1968 the keywords on a line, and turns any other whitespace on the line into
1971 @cindex comments, removed by preprocessor
1973 removes all comments, replacing them with a single space, or an
1974 appropriate number of newlines.
1976 @cindex constants, converted by preprocessor
1978 converts character constants into the appropriate numeric values.
1981 It does not do macro processing, include file handling, or
1982 anything else you may get from your C compiler's preprocessor. You can
1983 do include file processing with the @code{.include} directive
1984 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
1985 to get other ``CPP'' style preprocessing by giving the input file a
1986 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
1987 Output, gcc.info, Using GNU CC}.
1989 Excess whitespace, comments, and character constants
1990 cannot be used in the portions of the input text that are not
1993 @cindex turning preprocessing on and off
1994 @cindex preprocessing, turning on and off
1997 If the first line of an input file is @code{#NO_APP} or if you use the
1998 @samp{-f} option, whitespace and comments are not removed from the input file.
1999 Within an input file, you can ask for whitespace and comment removal in
2000 specific portions of the by putting a line that says @code{#APP} before the
2001 text that may contain whitespace or comments, and putting a line that says
2002 @code{#NO_APP} after this text. This feature is mainly intend to support
2003 @code{asm} statements in compilers whose output is otherwise free of comments
2010 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2011 Whitespace is used to separate symbols, and to make programs neater for
2012 people to read. Unless within character constants
2013 (@pxref{Characters,,Character Constants}), any whitespace means the same
2014 as exactly one space.
2020 There are two ways of rendering comments to @command{@value{AS}}. In both
2021 cases the comment is equivalent to one space.
2023 Anything from @samp{/*} through the next @samp{*/} is a comment.
2024 This means you may not nest these comments.
2028 The only way to include a newline ('\n') in a comment
2029 is to use this sort of comment.
2032 /* This sort of comment does not nest. */
2035 @cindex line comment character
2036 Anything from the @dfn{line comment} character to the next newline
2037 is considered a comment and is ignored. The line comment character is
2039 @samp{;} for the AMD 29K family;
2042 @samp{;} on the ARC;
2045 @samp{@@} on the ARM;
2048 @samp{;} for the H8/300 family;
2051 @samp{!} for the H8/500 family;
2054 @samp{;} for the HPPA;
2057 @samp{#} on the i386 and x86-64;
2060 @samp{#} on the i960;
2063 @samp{;} for the PDP-11;
2066 @samp{;} for picoJava;
2069 @samp{#} for Motorola PowerPC;
2072 @samp{!} for the Renesas / SuperH SH;
2075 @samp{!} on the SPARC;
2078 @samp{#} on the ip2k;
2081 @samp{#} on the m32r;
2084 @samp{|} on the 680x0;
2087 @samp{#} on the 68HC11 and 68HC12;
2090 @samp{;} on the M880x0;
2093 @samp{#} on the Vax;
2096 @samp{!} for the Z8000;
2099 @samp{#} on the V850;
2102 @samp{#} for Xtensa systems;
2104 see @ref{Machine Dependencies}. @refill
2105 @c FIXME What about i860?
2108 On some machines there are two different line comment characters. One
2109 character only begins a comment if it is the first non-whitespace character on
2110 a line, while the other always begins a comment.
2114 The V850 assembler also supports a double dash as starting a comment that
2115 extends to the end of the line.
2121 @cindex lines starting with @code{#}
2122 @cindex logical line numbers
2123 To be compatible with past assemblers, lines that begin with @samp{#} have a
2124 special interpretation. Following the @samp{#} should be an absolute
2125 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2126 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
2127 new logical file name. The rest of the line, if any, should be whitespace.
2129 If the first non-whitespace characters on the line are not numeric,
2130 the line is ignored. (Just like a comment.)
2133 # This is an ordinary comment.
2134 # 42-6 "new_file_name" # New logical file name
2135 # This is logical line # 36.
2137 This feature is deprecated, and may disappear from future versions
2138 of @command{@value{AS}}.
2143 @cindex characters used in symbols
2144 @ifclear SPECIAL-SYMS
2145 A @dfn{symbol} is one or more characters chosen from the set of all
2146 letters (both upper and lower case), digits and the three characters
2152 A @dfn{symbol} is one or more characters chosen from the set of all
2153 letters (both upper and lower case), digits and the three characters
2154 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2160 On most machines, you can also use @code{$} in symbol names; exceptions
2161 are noted in @ref{Machine Dependencies}.
2163 No symbol may begin with a digit. Case is significant.
2164 There is no length limit: all characters are significant. Symbols are
2165 delimited by characters not in that set, or by the beginning of a file
2166 (since the source program must end with a newline, the end of a file is
2167 not a possible symbol delimiter). @xref{Symbols}.
2168 @cindex length of symbols
2173 @cindex statements, structure of
2174 @cindex line separator character
2175 @cindex statement separator character
2177 @ifclear abnormal-separator
2178 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2179 semicolon (@samp{;}). The newline or semicolon is considered part of
2180 the preceding statement. Newlines and semicolons within character
2181 constants are an exception: they do not end statements.
2183 @ifset abnormal-separator
2185 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
2186 sign (@samp{@@}). The newline or at sign is considered part of the
2187 preceding statement. Newlines and at signs within character constants
2188 are an exception: they do not end statements.
2191 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2192 point (@samp{!}). The newline or exclamation point is considered part of the
2193 preceding statement. Newlines and exclamation points within character
2194 constants are an exception: they do not end statements.
2197 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2198 H8/300) a dollar sign (@samp{$}); or (for the
2201 (@samp{;}). The newline or separator character is considered part of
2202 the preceding statement. Newlines and separators within character
2203 constants are an exception: they do not end statements.
2208 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2209 separator character. (The line separator is usually @samp{;}, unless
2210 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
2211 newline or separator character is considered part of the preceding
2212 statement. Newlines and separators within character constants are an
2213 exception: they do not end statements.
2216 @cindex newline, required at file end
2217 @cindex EOF, newline must precede
2218 It is an error to end any statement with end-of-file: the last
2219 character of any input file should be a newline.@refill
2221 An empty statement is allowed, and may include whitespace. It is ignored.
2223 @cindex instructions and directives
2224 @cindex directives and instructions
2225 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2226 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2228 A statement begins with zero or more labels, optionally followed by a
2229 key symbol which determines what kind of statement it is. The key
2230 symbol determines the syntax of the rest of the statement. If the
2231 symbol begins with a dot @samp{.} then the statement is an assembler
2232 directive: typically valid for any computer. If the symbol begins with
2233 a letter the statement is an assembly language @dfn{instruction}: it
2234 assembles into a machine language instruction.
2236 Different versions of @command{@value{AS}} for different computers
2237 recognize different instructions. In fact, the same symbol may
2238 represent a different instruction in a different computer's assembly
2242 @cindex @code{:} (label)
2243 @cindex label (@code{:})
2244 A label is a symbol immediately followed by a colon (@code{:}).
2245 Whitespace before a label or after a colon is permitted, but you may not
2246 have whitespace between a label's symbol and its colon. @xref{Labels}.
2249 For HPPA targets, labels need not be immediately followed by a colon, but
2250 the definition of a label must begin in column zero. This also implies that
2251 only one label may be defined on each line.
2255 label: .directive followed by something
2256 another_label: # This is an empty statement.
2257 instruction operand_1, operand_2, @dots{}
2264 A constant is a number, written so that its value is known by
2265 inspection, without knowing any context. Like this:
2268 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2269 .ascii "Ring the bell\7" # A string constant.
2270 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2271 .float 0f-314159265358979323846264338327\
2272 95028841971.693993751E-40 # - pi, a flonum.
2277 * Characters:: Character Constants
2278 * Numbers:: Number Constants
2282 @subsection Character Constants
2284 @cindex character constants
2285 @cindex constants, character
2286 There are two kinds of character constants. A @dfn{character} stands
2287 for one character in one byte and its value may be used in
2288 numeric expressions. String constants (properly called string
2289 @emph{literals}) are potentially many bytes and their values may not be
2290 used in arithmetic expressions.
2294 * Chars:: Characters
2298 @subsubsection Strings
2300 @cindex string constants
2301 @cindex constants, string
2302 A @dfn{string} is written between double-quotes. It may contain
2303 double-quotes or null characters. The way to get special characters
2304 into a string is to @dfn{escape} these characters: precede them with
2305 a backslash @samp{\} character. For example @samp{\\} represents
2306 one backslash: the first @code{\} is an escape which tells
2307 @command{@value{AS}} to interpret the second character literally as a backslash
2308 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2309 escape character). The complete list of escapes follows.
2311 @cindex escape codes, character
2312 @cindex character escape codes
2315 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2317 @cindex @code{\b} (backspace character)
2318 @cindex backspace (@code{\b})
2320 Mnemonic for backspace; for ASCII this is octal code 010.
2323 @c Mnemonic for EOText; for ASCII this is octal code 004.
2325 @cindex @code{\f} (formfeed character)
2326 @cindex formfeed (@code{\f})
2328 Mnemonic for FormFeed; for ASCII this is octal code 014.
2330 @cindex @code{\n} (newline character)
2331 @cindex newline (@code{\n})
2333 Mnemonic for newline; for ASCII this is octal code 012.
2336 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2338 @cindex @code{\r} (carriage return character)
2339 @cindex carriage return (@code{\r})
2341 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2344 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2345 @c other assemblers.
2347 @cindex @code{\t} (tab)
2348 @cindex tab (@code{\t})
2350 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2353 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2354 @c @item \x @var{digit} @var{digit} @var{digit}
2355 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2357 @cindex @code{\@var{ddd}} (octal character code)
2358 @cindex octal character code (@code{\@var{ddd}})
2359 @item \ @var{digit} @var{digit} @var{digit}
2360 An octal character code. The numeric code is 3 octal digits.
2361 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2362 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2364 @cindex @code{\@var{xd...}} (hex character code)
2365 @cindex hex character code (@code{\@var{xd...}})
2366 @item \@code{x} @var{hex-digits...}
2367 A hex character code. All trailing hex digits are combined. Either upper or
2368 lower case @code{x} works.
2370 @cindex @code{\\} (@samp{\} character)
2371 @cindex backslash (@code{\\})
2373 Represents one @samp{\} character.
2376 @c Represents one @samp{'} (accent acute) character.
2377 @c This is needed in single character literals
2378 @c (@xref{Characters,,Character Constants}.) to represent
2381 @cindex @code{\"} (doublequote character)
2382 @cindex doublequote (@code{\"})
2384 Represents one @samp{"} character. Needed in strings to represent
2385 this character, because an unescaped @samp{"} would end the string.
2387 @item \ @var{anything-else}
2388 Any other character when escaped by @kbd{\} gives a warning, but
2389 assembles as if the @samp{\} was not present. The idea is that if
2390 you used an escape sequence you clearly didn't want the literal
2391 interpretation of the following character. However @command{@value{AS}} has no
2392 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2393 code and warns you of the fact.
2396 Which characters are escapable, and what those escapes represent,
2397 varies widely among assemblers. The current set is what we think
2398 the BSD 4.2 assembler recognizes, and is a subset of what most C
2399 compilers recognize. If you are in doubt, do not use an escape
2403 @subsubsection Characters
2405 @cindex single character constant
2406 @cindex character, single
2407 @cindex constant, single character
2408 A single character may be written as a single quote immediately
2409 followed by that character. The same escapes apply to characters as
2410 to strings. So if you want to write the character backslash, you
2411 must write @kbd{'\\} where the first @code{\} escapes the second
2412 @code{\}. As you can see, the quote is an acute accent, not a
2413 grave accent. A newline
2415 @ifclear abnormal-separator
2416 (or semicolon @samp{;})
2418 @ifset abnormal-separator
2420 (or at sign @samp{@@})
2423 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2424 Renesas SH or H8/500)
2428 immediately following an acute accent is taken as a literal character
2429 and does not count as the end of a statement. The value of a character
2430 constant in a numeric expression is the machine's byte-wide code for
2431 that character. @command{@value{AS}} assumes your character code is ASCII:
2432 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2435 @subsection Number Constants
2437 @cindex constants, number
2438 @cindex number constants
2439 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2440 are stored in the target machine. @emph{Integers} are numbers that
2441 would fit into an @code{int} in the C language. @emph{Bignums} are
2442 integers, but they are stored in more than 32 bits. @emph{Flonums}
2443 are floating point numbers, described below.
2446 * Integers:: Integers
2451 * Bit Fields:: Bit Fields
2457 @subsubsection Integers
2459 @cindex constants, integer
2461 @cindex binary integers
2462 @cindex integers, binary
2463 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2464 the binary digits @samp{01}.
2466 @cindex octal integers
2467 @cindex integers, octal
2468 An octal integer is @samp{0} followed by zero or more of the octal
2469 digits (@samp{01234567}).
2471 @cindex decimal integers
2472 @cindex integers, decimal
2473 A decimal integer starts with a non-zero digit followed by zero or
2474 more digits (@samp{0123456789}).
2476 @cindex hexadecimal integers
2477 @cindex integers, hexadecimal
2478 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2479 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2481 Integers have the usual values. To denote a negative integer, use
2482 the prefix operator @samp{-} discussed under expressions
2483 (@pxref{Prefix Ops,,Prefix Operators}).
2486 @subsubsection Bignums
2489 @cindex constants, bignum
2490 A @dfn{bignum} has the same syntax and semantics as an integer
2491 except that the number (or its negative) takes more than 32 bits to
2492 represent in binary. The distinction is made because in some places
2493 integers are permitted while bignums are not.
2496 @subsubsection Flonums
2498 @cindex floating point numbers
2499 @cindex constants, floating point
2501 @cindex precision, floating point
2502 A @dfn{flonum} represents a floating point number. The translation is
2503 indirect: a decimal floating point number from the text is converted by
2504 @command{@value{AS}} to a generic binary floating point number of more than
2505 sufficient precision. This generic floating point number is converted
2506 to a particular computer's floating point format (or formats) by a
2507 portion of @command{@value{AS}} specialized to that computer.
2509 A flonum is written by writing (in order)
2514 (@samp{0} is optional on the HPPA.)
2518 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2520 @kbd{e} is recommended. Case is not important.
2522 @c FIXME: verify if flonum syntax really this vague for most cases
2523 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2524 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2527 On the H8/300, H8/500,
2528 Renesas / SuperH SH,
2529 and AMD 29K architectures, the letter must be
2530 one of the letters @samp{DFPRSX} (in upper or lower case).
2532 On the ARC, the letter must be one of the letters @samp{DFRS}
2533 (in upper or lower case).
2535 On the Intel 960 architecture, the letter must be
2536 one of the letters @samp{DFT} (in upper or lower case).
2538 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2542 One of the letters @samp{DFPRSX} (in upper or lower case).
2545 One of the letters @samp{DFRS} (in upper or lower case).
2548 One of the letters @samp{DFPRSX} (in upper or lower case).
2551 The letter @samp{E} (upper case only).
2554 One of the letters @samp{DFT} (in upper or lower case).
2559 An optional sign: either @samp{+} or @samp{-}.
2562 An optional @dfn{integer part}: zero or more decimal digits.
2565 An optional @dfn{fractional part}: @samp{.} followed by zero
2566 or more decimal digits.
2569 An optional exponent, consisting of:
2573 An @samp{E} or @samp{e}.
2574 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2575 @c principle this can perfectly well be different on different targets.
2577 Optional sign: either @samp{+} or @samp{-}.
2579 One or more decimal digits.
2584 At least one of the integer part or the fractional part must be
2585 present. The floating point number has the usual base-10 value.
2587 @command{@value{AS}} does all processing using integers. Flonums are computed
2588 independently of any floating point hardware in the computer running
2589 @command{@value{AS}}.
2593 @c Bit fields are written as a general facility but are also controlled
2594 @c by a conditional-compilation flag---which is as of now (21mar91)
2595 @c turned on only by the i960 config of GAS.
2597 @subsubsection Bit Fields
2600 @cindex constants, bit field
2601 You can also define numeric constants as @dfn{bit fields}.
2602 specify two numbers separated by a colon---
2604 @var{mask}:@var{value}
2607 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2610 The resulting number is then packed
2612 @c this conditional paren in case bit fields turned on elsewhere than 960
2613 (in host-dependent byte order)
2615 into a field whose width depends on which assembler directive has the
2616 bit-field as its argument. Overflow (a result from the bitwise and
2617 requiring more binary digits to represent) is not an error; instead,
2618 more constants are generated, of the specified width, beginning with the
2619 least significant digits.@refill
2621 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2622 @code{.short}, and @code{.word} accept bit-field arguments.
2627 @chapter Sections and Relocation
2632 * Secs Background:: Background
2633 * Ld Sections:: Linker Sections
2634 * As Sections:: Assembler Internal Sections
2635 * Sub-Sections:: Sub-Sections
2639 @node Secs Background
2642 Roughly, a section is a range of addresses, with no gaps; all data
2643 ``in'' those addresses is treated the same for some particular purpose.
2644 For example there may be a ``read only'' section.
2646 @cindex linker, and assembler
2647 @cindex assembler, and linker
2648 The linker @code{@value{LD}} reads many object files (partial programs) and
2649 combines their contents to form a runnable program. When @command{@value{AS}}
2650 emits an object file, the partial program is assumed to start at address 0.
2651 @code{@value{LD}} assigns the final addresses for the partial program, so that
2652 different partial programs do not overlap. This is actually an
2653 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2656 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2657 addresses. These blocks slide to their run-time addresses as rigid
2658 units; their length does not change and neither does the order of bytes
2659 within them. Such a rigid unit is called a @emph{section}. Assigning
2660 run-time addresses to sections is called @dfn{relocation}. It includes
2661 the task of adjusting mentions of object-file addresses so they refer to
2662 the proper run-time addresses.
2664 For the H8/300 and H8/500,
2665 and for the Renesas / SuperH SH,
2666 @command{@value{AS}} pads sections if needed to
2667 ensure they end on a word (sixteen bit) boundary.
2670 @cindex standard assembler sections
2671 An object file written by @command{@value{AS}} has at least three sections, any
2672 of which may be empty. These are named @dfn{text}, @dfn{data} and
2677 When it generates COFF or ELF output,
2679 @command{@value{AS}} can also generate whatever other named sections you specify
2680 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2681 If you do not use any directives that place output in the @samp{.text}
2682 or @samp{.data} sections, these sections still exist, but are empty.
2687 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2689 @command{@value{AS}} can also generate whatever other named sections you
2690 specify using the @samp{.space} and @samp{.subspace} directives. See
2691 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2692 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2693 assembler directives.
2696 Additionally, @command{@value{AS}} uses different names for the standard
2697 text, data, and bss sections when generating SOM output. Program text
2698 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2699 BSS into @samp{$BSS$}.
2703 Within the object file, the text section starts at address @code{0}, the
2704 data section follows, and the bss section follows the data section.
2707 When generating either SOM or ELF output files on the HPPA, the text
2708 section starts at address @code{0}, the data section at address
2709 @code{0x4000000}, and the bss section follows the data section.
2712 To let @code{@value{LD}} know which data changes when the sections are
2713 relocated, and how to change that data, @command{@value{AS}} also writes to the
2714 object file details of the relocation needed. To perform relocation
2715 @code{@value{LD}} must know, each time an address in the object
2719 Where in the object file is the beginning of this reference to
2722 How long (in bytes) is this reference?
2724 Which section does the address refer to? What is the numeric value of
2726 (@var{address}) @minus{} (@var{start-address of section})?
2729 Is the reference to an address ``Program-Counter relative''?
2732 @cindex addresses, format of
2733 @cindex section-relative addressing
2734 In fact, every address @command{@value{AS}} ever uses is expressed as
2736 (@var{section}) + (@var{offset into section})
2739 Further, most expressions @command{@value{AS}} computes have this section-relative
2742 (For some object formats, such as SOM for the HPPA, some expressions are
2743 symbol-relative instead.)
2746 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2747 @var{N} into section @var{secname}.''
2749 Apart from text, data and bss sections you need to know about the
2750 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2751 addresses in the absolute section remain unchanged. For example, address
2752 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2753 @code{@value{LD}}. Although the linker never arranges two partial programs'
2754 data sections with overlapping addresses after linking, @emph{by definition}
2755 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2756 part of a program is always the same address when the program is running as
2757 address @code{@{absolute@ 239@}} in any other part of the program.
2759 The idea of sections is extended to the @dfn{undefined} section. Any
2760 address whose section is unknown at assembly time is by definition
2761 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2762 Since numbers are always defined, the only way to generate an undefined
2763 address is to mention an undefined symbol. A reference to a named
2764 common block would be such a symbol: its value is unknown at assembly
2765 time so it has section @emph{undefined}.
2767 By analogy the word @emph{section} is used to describe groups of sections in
2768 the linked program. @code{@value{LD}} puts all partial programs' text
2769 sections in contiguous addresses in the linked program. It is
2770 customary to refer to the @emph{text section} of a program, meaning all
2771 the addresses of all partial programs' text sections. Likewise for
2772 data and bss sections.
2774 Some sections are manipulated by @code{@value{LD}}; others are invented for
2775 use of @command{@value{AS}} and have no meaning except during assembly.
2778 @section Linker Sections
2779 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2784 @cindex named sections
2785 @cindex sections, named
2786 @item named sections
2789 @cindex text section
2790 @cindex data section
2794 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2795 separate but equal sections. Anything you can say of one section is
2798 When the program is running, however, it is
2799 customary for the text section to be unalterable. The
2800 text section is often shared among processes: it contains
2801 instructions, constants and the like. The data section of a running
2802 program is usually alterable: for example, C variables would be stored
2803 in the data section.
2808 This section contains zeroed bytes when your program begins running. It
2809 is used to hold uninitialized variables or common storage. The length of
2810 each partial program's bss section is important, but because it starts
2811 out containing zeroed bytes there is no need to store explicit zero
2812 bytes in the object file. The bss section was invented to eliminate
2813 those explicit zeros from object files.
2815 @cindex absolute section
2816 @item absolute section
2817 Address 0 of this section is always ``relocated'' to runtime address 0.
2818 This is useful if you want to refer to an address that @code{@value{LD}} must
2819 not change when relocating. In this sense we speak of absolute
2820 addresses being ``unrelocatable'': they do not change during relocation.
2822 @cindex undefined section
2823 @item undefined section
2824 This ``section'' is a catch-all for address references to objects not in
2825 the preceding sections.
2826 @c FIXME: ref to some other doc on obj-file formats could go here.
2829 @cindex relocation example
2830 An idealized example of three relocatable sections follows.
2832 The example uses the traditional section names @samp{.text} and @samp{.data}.
2834 Memory addresses are on the horizontal axis.
2838 @c END TEXI2ROFF-KILL
2841 partial program # 1: |ttttt|dddd|00|
2848 partial program # 2: |TTT|DDD|000|
2851 +--+---+-----+--+----+---+-----+~~
2852 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2853 +--+---+-----+--+----+---+-----+~~
2855 addresses: 0 @dots{}
2862 \line{\it Partial program \#1: \hfil}
2863 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2864 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2866 \line{\it Partial program \#2: \hfil}
2867 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2868 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2870 \line{\it linked program: \hfil}
2871 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2872 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2873 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2874 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2876 \line{\it addresses: \hfil}
2880 @c END TEXI2ROFF-KILL
2883 @section Assembler Internal Sections
2885 @cindex internal assembler sections
2886 @cindex sections in messages, internal
2887 These sections are meant only for the internal use of @command{@value{AS}}. They
2888 have no meaning at run-time. You do not really need to know about these
2889 sections for most purposes; but they can be mentioned in @command{@value{AS}}
2890 warning messages, so it might be helpful to have an idea of their
2891 meanings to @command{@value{AS}}. These sections are used to permit the
2892 value of every expression in your assembly language program to be a
2893 section-relative address.
2896 @cindex assembler internal logic error
2897 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2898 An internal assembler logic error has been found. This means there is a
2899 bug in the assembler.
2901 @cindex expr (internal section)
2903 The assembler stores complex expression internally as combinations of
2904 symbols. When it needs to represent an expression as a symbol, it puts
2905 it in the expr section.
2907 @c FIXME item transfer[t] vector preload
2908 @c FIXME item transfer[t] vector postload
2909 @c FIXME item register
2913 @section Sub-Sections
2915 @cindex numbered subsections
2916 @cindex grouping data
2922 fall into two sections: text and data.
2924 You may have separate groups of
2926 data in named sections
2930 data in named sections
2936 that you want to end up near to each other in the object file, even though they
2937 are not contiguous in the assembler source. @command{@value{AS}} allows you to
2938 use @dfn{subsections} for this purpose. Within each section, there can be
2939 numbered subsections with values from 0 to 8192. Objects assembled into the
2940 same subsection go into the object file together with other objects in the same
2941 subsection. For example, a compiler might want to store constants in the text
2942 section, but might not want to have them interspersed with the program being
2943 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2944 section of code being output, and a @samp{.text 1} before each group of
2945 constants being output.
2947 Subsections are optional. If you do not use subsections, everything
2948 goes in subsection number zero.
2951 Each subsection is zero-padded up to a multiple of four bytes.
2952 (Subsections may be padded a different amount on different flavors
2953 of @command{@value{AS}}.)
2957 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2958 boundary (two bytes).
2959 The same is true on the Renesas SH.
2962 @c FIXME section padding (alignment)?
2963 @c Rich Pixley says padding here depends on target obj code format; that
2964 @c doesn't seem particularly useful to say without further elaboration,
2965 @c so for now I say nothing about it. If this is a generic BFD issue,
2966 @c these paragraphs might need to vanish from this manual, and be
2967 @c discussed in BFD chapter of binutils (or some such).
2970 On the AMD 29K family, no particular padding is added to section or
2971 subsection sizes; @value{AS} forces no alignment on this platform.
2975 Subsections appear in your object file in numeric order, lowest numbered
2976 to highest. (All this to be compatible with other people's assemblers.)
2977 The object file contains no representation of subsections; @code{@value{LD}} and
2978 other programs that manipulate object files see no trace of them.
2979 They just see all your text subsections as a text section, and all your
2980 data subsections as a data section.
2982 To specify which subsection you want subsequent statements assembled
2983 into, use a numeric argument to specify it, in a @samp{.text
2984 @var{expression}} or a @samp{.data @var{expression}} statement.
2987 When generating COFF or ELF output, you
2992 can also use an extra subsection
2993 argument with arbitrary named sections: @samp{.section @var{name},
2996 @var{Expression} should be an absolute expression.
2997 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
2998 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
2999 begins in @code{text 0}. For instance:
3001 .text 0 # The default subsection is text 0 anyway.
3002 .ascii "This lives in the first text subsection. *"
3004 .ascii "But this lives in the second text subsection."
3006 .ascii "This lives in the data section,"
3007 .ascii "in the first data subsection."
3009 .ascii "This lives in the first text section,"
3010 .ascii "immediately following the asterisk (*)."
3013 Each section has a @dfn{location counter} incremented by one for every byte
3014 assembled into that section. Because subsections are merely a convenience
3015 restricted to @command{@value{AS}} there is no concept of a subsection location
3016 counter. There is no way to directly manipulate a location counter---but the
3017 @code{.align} directive changes it, and any label definition captures its
3018 current value. The location counter of the section where statements are being
3019 assembled is said to be the @dfn{active} location counter.
3022 @section bss Section
3025 @cindex common variable storage
3026 The bss section is used for local common variable storage.
3027 You may allocate address space in the bss section, but you may
3028 not dictate data to load into it before your program executes. When
3029 your program starts running, all the contents of the bss
3030 section are zeroed bytes.
3032 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3033 @ref{Lcomm,,@code{.lcomm}}.
3035 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3036 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
3039 When assembling for a target which supports multiple sections, such as ELF or
3040 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3041 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3042 section. Typically the section will only contain symbol definitions and
3043 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3050 Symbols are a central concept: the programmer uses symbols to name
3051 things, the linker uses symbols to link, and the debugger uses symbols
3055 @cindex debuggers, and symbol order
3056 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3057 the same order they were declared. This may break some debuggers.
3062 * Setting Symbols:: Giving Symbols Other Values
3063 * Symbol Names:: Symbol Names
3064 * Dot:: The Special Dot Symbol
3065 * Symbol Attributes:: Symbol Attributes
3072 A @dfn{label} is written as a symbol immediately followed by a colon
3073 @samp{:}. The symbol then represents the current value of the
3074 active location counter, and is, for example, a suitable instruction
3075 operand. You are warned if you use the same symbol to represent two
3076 different locations: the first definition overrides any other
3080 On the HPPA, the usual form for a label need not be immediately followed by a
3081 colon, but instead must start in column zero. Only one label may be defined on
3082 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3083 provides a special directive @code{.label} for defining labels more flexibly.
3086 @node Setting Symbols
3087 @section Giving Symbols Other Values
3089 @cindex assigning values to symbols
3090 @cindex symbol values, assigning
3091 A symbol can be given an arbitrary value by writing a symbol, followed
3092 by an equals sign @samp{=}, followed by an expression
3093 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3094 directive. @xref{Set,,@code{.set}}.
3097 @section Symbol Names
3099 @cindex symbol names
3100 @cindex names, symbol
3101 @ifclear SPECIAL-SYMS
3102 Symbol names begin with a letter or with one of @samp{._}. On most
3103 machines, you can also use @code{$} in symbol names; exceptions are
3104 noted in @ref{Machine Dependencies}. That character may be followed by any
3105 string of digits, letters, dollar signs (unless otherwise noted in
3106 @ref{Machine Dependencies}), and underscores.
3109 For the AMD 29K family, @samp{?} is also allowed in the
3110 body of a symbol name, though not at its beginning.
3115 Symbol names begin with a letter or with one of @samp{._}. On the
3116 Renesas SH or the H8/500, you can also use @code{$} in symbol names. That
3117 character may be followed by any string of digits, letters, dollar signs (save
3118 on the H8/300), and underscores.
3122 Case of letters is significant: @code{foo} is a different symbol name
3125 Each symbol has exactly one name. Each name in an assembly language program
3126 refers to exactly one symbol. You may use that symbol name any number of times
3129 @subheading Local Symbol Names
3131 @cindex local symbol names
3132 @cindex symbol names, local
3133 @cindex temporary symbol names
3134 @cindex symbol names, temporary
3135 Local symbols help compilers and programmers use names temporarily.
3136 They create symbols which are guaranteed to be unique over the entire scope of
3137 the input source code and which can be referred to by a simple notation.
3138 To define a local symbol, write a label of the form @samp{@b{N}:} (where @b{N}
3139 represents any positive integer). To refer to the most recent previous
3140 definition of that symbol write @samp{@b{N}b}, using the same number as when
3141 you defined the label. To refer to the next definition of a local label, write
3142 @samp{@b{N}f}--- The @samp{b} stands for``backwards'' and the @samp{f} stands
3145 There is no restriction on how you can use these labels, and you can reuse them
3146 too. So that it is possible to repeatedly define the same local label (using
3147 the same number @samp{@b{N}}), although you can only refer to the most recently
3148 defined local label of that number (for a backwards reference) or the next
3149 definition of a specific local label for a forward reference. It is also worth
3150 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3151 implemented in a slightly more efficient manner than the others.
3162 Which is the equivalent of:
3165 label_1: branch label_3
3166 label_2: branch label_1
3167 label_3: branch label_4
3168 label_4: branch label_3
3171 Local symbol names are only a notational device. They are immediately
3172 transformed into more conventional symbol names before the assembler uses them.
3173 The symbol names stored in the symbol table, appearing in error messages and
3174 optionally emitted to the object file. The names are constructed using these
3179 All local labels begin with @samp{L}. Normally both @command{@value{AS}} and
3180 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
3181 used for symbols you are never intended to see. If you use the
3182 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3183 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3184 you may use them in debugging.
3187 This is the number that was used in the local label definition. So if the
3188 label is written @samp{55:} then the number is @samp{55}.
3191 This unusual character is included so you do not accidentally invent a symbol
3192 of the same name. The character has ASCII value of @samp{\002} (control-B).
3194 @item @emph{ordinal number}
3195 This is a serial number to keep the labels distinct. The first definition of
3196 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3197 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3198 the number @samp{1} and its 15th defintion gets @samp{15} as well.
3201 So for example, the first @code{1:} is named @code{L1@kbd{C-B}1}, the 44th
3202 @code{3:} is named @code{L3@kbd{C-B}44}.
3204 @subheading Dollar Local Labels
3205 @cindex dollar local symbols
3207 @code{@value{AS}} also supports an even more local form of local labels called
3208 dollar labels. These labels go out of scope (ie they become undefined) as soon
3209 as a non-local label is defined. Thus they remain valid for only a small
3210 region of the input source code. Normal local labels, by contrast, remain in
3211 scope for the entire file, or until they are redefined by another occurrence of
3212 the same local label.
3214 Dollar labels are defined in exactly the same way as ordinary local labels,
3215 except that instead of being terminated by a colon, they are terminated by a
3216 dollar sign. eg @samp{@b{55$}}.
3218 They can also be distinguished from ordinary local labels by their transformed
3219 name which uses ASCII character @samp{\001} (control-A) as the magic character
3220 to distinguish them from ordinary labels. Thus the 5th defintion of @samp{6$}
3221 is named @samp{L6@kbd{C-A}5}.
3224 @section The Special Dot Symbol
3226 @cindex dot (symbol)
3227 @cindex @code{.} (symbol)
3228 @cindex current address
3229 @cindex location counter
3230 The special symbol @samp{.} refers to the current address that
3231 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3232 .long .} defines @code{melvin} to contain its own address.
3233 Assigning a value to @code{.} is treated the same as a @code{.org}
3234 directive. Thus, the expression @samp{.=.+4} is the same as saying
3235 @ifclear no-space-dir
3244 @node Symbol Attributes
3245 @section Symbol Attributes
3247 @cindex symbol attributes
3248 @cindex attributes, symbol
3249 Every symbol has, as well as its name, the attributes ``Value'' and
3250 ``Type''. Depending on output format, symbols can also have auxiliary
3253 The detailed definitions are in @file{a.out.h}.
3256 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3257 all these attributes, and probably won't warn you. This makes the
3258 symbol an externally defined symbol, which is generally what you
3262 * Symbol Value:: Value
3263 * Symbol Type:: Type
3266 * a.out Symbols:: Symbol Attributes: @code{a.out}
3270 * a.out Symbols:: Symbol Attributes: @code{a.out}
3273 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3278 * COFF Symbols:: Symbol Attributes for COFF
3281 * SOM Symbols:: Symbol Attributes for SOM
3288 @cindex value of a symbol
3289 @cindex symbol value
3290 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3291 location in the text, data, bss or absolute sections the value is the
3292 number of addresses from the start of that section to the label.
3293 Naturally for text, data and bss sections the value of a symbol changes
3294 as @code{@value{LD}} changes section base addresses during linking. Absolute
3295 symbols' values do not change during linking: that is why they are
3298 The value of an undefined symbol is treated in a special way. If it is
3299 0 then the symbol is not defined in this assembler source file, and
3300 @code{@value{LD}} tries to determine its value from other files linked into the
3301 same program. You make this kind of symbol simply by mentioning a symbol
3302 name without defining it. A non-zero value represents a @code{.comm}
3303 common declaration. The value is how much common storage to reserve, in
3304 bytes (addresses). The symbol refers to the first address of the
3310 @cindex type of a symbol
3312 The type attribute of a symbol contains relocation (section)
3313 information, any flag settings indicating that a symbol is external, and
3314 (optionally), other information for linkers and debuggers. The exact
3315 format depends on the object-code output format in use.
3320 @c The following avoids a "widow" subsection title. @group would be
3321 @c better if it were available outside examples.
3324 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3326 @cindex @code{b.out} symbol attributes
3327 @cindex symbol attributes, @code{b.out}
3328 These symbol attributes appear only when @command{@value{AS}} is configured for
3329 one of the Berkeley-descended object output formats---@code{a.out} or
3335 @subsection Symbol Attributes: @code{a.out}
3337 @cindex @code{a.out} symbol attributes
3338 @cindex symbol attributes, @code{a.out}
3344 @subsection Symbol Attributes: @code{a.out}
3346 @cindex @code{a.out} symbol attributes
3347 @cindex symbol attributes, @code{a.out}
3351 * Symbol Desc:: Descriptor
3352 * Symbol Other:: Other
3356 @subsubsection Descriptor
3358 @cindex descriptor, of @code{a.out} symbol
3359 This is an arbitrary 16-bit value. You may establish a symbol's
3360 descriptor value by using a @code{.desc} statement
3361 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3362 @command{@value{AS}}.
3365 @subsubsection Other
3367 @cindex other attribute, of @code{a.out} symbol
3368 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3373 @subsection Symbol Attributes for COFF
3375 @cindex COFF symbol attributes
3376 @cindex symbol attributes, COFF
3378 The COFF format supports a multitude of auxiliary symbol attributes;
3379 like the primary symbol attributes, they are set between @code{.def} and
3380 @code{.endef} directives.
3382 @subsubsection Primary Attributes
3384 @cindex primary attributes, COFF symbols
3385 The symbol name is set with @code{.def}; the value and type,
3386 respectively, with @code{.val} and @code{.type}.
3388 @subsubsection Auxiliary Attributes
3390 @cindex auxiliary attributes, COFF symbols
3391 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3392 @code{.size}, and @code{.tag} can generate auxiliary symbol table
3393 information for COFF.
3398 @subsection Symbol Attributes for SOM
3400 @cindex SOM symbol attributes
3401 @cindex symbol attributes, SOM
3403 The SOM format for the HPPA supports a multitude of symbol attributes set with
3404 the @code{.EXPORT} and @code{.IMPORT} directives.
3406 The attributes are described in @cite{HP9000 Series 800 Assembly
3407 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3408 @code{EXPORT} assembler directive documentation.
3412 @chapter Expressions
3416 @cindex numeric values
3417 An @dfn{expression} specifies an address or numeric value.
3418 Whitespace may precede and/or follow an expression.
3420 The result of an expression must be an absolute number, or else an offset into
3421 a particular section. If an expression is not absolute, and there is not
3422 enough information when @command{@value{AS}} sees the expression to know its
3423 section, a second pass over the source program might be necessary to interpret
3424 the expression---but the second pass is currently not implemented.
3425 @command{@value{AS}} aborts with an error message in this situation.
3428 * Empty Exprs:: Empty Expressions
3429 * Integer Exprs:: Integer Expressions
3433 @section Empty Expressions
3435 @cindex empty expressions
3436 @cindex expressions, empty
3437 An empty expression has no value: it is just whitespace or null.
3438 Wherever an absolute expression is required, you may omit the
3439 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3440 is compatible with other assemblers.
3443 @section Integer Expressions
3445 @cindex integer expressions
3446 @cindex expressions, integer
3447 An @dfn{integer expression} is one or more @emph{arguments} delimited
3448 by @emph{operators}.
3451 * Arguments:: Arguments
3452 * Operators:: Operators
3453 * Prefix Ops:: Prefix Operators
3454 * Infix Ops:: Infix Operators
3458 @subsection Arguments
3460 @cindex expression arguments
3461 @cindex arguments in expressions
3462 @cindex operands in expressions
3463 @cindex arithmetic operands
3464 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3465 contexts arguments are sometimes called ``arithmetic operands''. In
3466 this manual, to avoid confusing them with the ``instruction operands'' of
3467 the machine language, we use the term ``argument'' to refer to parts of
3468 expressions only, reserving the word ``operand'' to refer only to machine
3469 instruction operands.
3471 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3472 @var{section} is one of text, data, bss, absolute,
3473 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3476 Numbers are usually integers.
3478 A number can be a flonum or bignum. In this case, you are warned
3479 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3480 these 32 bits are an integer. You may write integer-manipulating
3481 instructions that act on exotic constants, compatible with other
3484 @cindex subexpressions
3485 Subexpressions are a left parenthesis @samp{(} followed by an integer
3486 expression, followed by a right parenthesis @samp{)}; or a prefix
3487 operator followed by an argument.
3490 @subsection Operators
3492 @cindex operators, in expressions
3493 @cindex arithmetic functions
3494 @cindex functions, in expressions
3495 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3496 operators are followed by an argument. Infix operators appear
3497 between their arguments. Operators may be preceded and/or followed by
3501 @subsection Prefix Operator
3503 @cindex prefix operators
3504 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3505 one argument, which must be absolute.
3507 @c the tex/end tex stuff surrounding this small table is meant to make
3508 @c it align, on the printed page, with the similar table in the next
3509 @c section (which is inside an enumerate).
3511 \global\advance\leftskip by \itemindent
3516 @dfn{Negation}. Two's complement negation.
3518 @dfn{Complementation}. Bitwise not.
3522 \global\advance\leftskip by -\itemindent
3526 @subsection Infix Operators
3528 @cindex infix operators
3529 @cindex operators, permitted arguments
3530 @dfn{Infix operators} take two arguments, one on either side. Operators
3531 have precedence, but operations with equal precedence are performed left
3532 to right. Apart from @code{+} or @option{-}, both arguments must be
3533 absolute, and the result is absolute.
3536 @cindex operator precedence
3537 @cindex precedence of operators
3544 @dfn{Multiplication}.
3547 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3554 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3558 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3562 Intermediate precedence
3567 @dfn{Bitwise Inclusive Or}.
3573 @dfn{Bitwise Exclusive Or}.
3576 @dfn{Bitwise Or Not}.
3583 @cindex addition, permitted arguments
3584 @cindex plus, permitted arguments
3585 @cindex arguments for addition
3587 @dfn{Addition}. If either argument is absolute, the result has the section of
3588 the other argument. You may not add together arguments from different
3591 @cindex subtraction, permitted arguments
3592 @cindex minus, permitted arguments
3593 @cindex arguments for subtraction
3595 @dfn{Subtraction}. If the right argument is absolute, the
3596 result has the section of the left argument.
3597 If both arguments are in the same section, the result is absolute.
3598 You may not subtract arguments from different sections.
3599 @c FIXME is there still something useful to say about undefined - undefined ?
3601 @cindex comparison expressions
3602 @cindex expressions, comparison
3606 @dfn{Is Not Equal To}
3610 @dfn{Is Greater Than}
3612 @dfn{Is Greater Than Or Equal To}
3614 @dfn{Is Less Than Or Equal To}
3616 The comparison operators can be used as infix operators. A true results has a
3617 value of -1 whereas a false result has a value of 0. Note, these operators
3618 perform signed comparisons.
3621 @item Lowest Precedence
3630 These two logical operations can be used to combine the results of sub
3631 expressions. Note, unlike the comparison operators a true result returns a
3632 value of 1 but a false results does still return 0. Also note that the logical
3633 or operator has a slightly lower precedence than logical and.
3638 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3639 address; you can only have a defined section in one of the two arguments.
3642 @chapter Assembler Directives
3644 @cindex directives, machine independent
3645 @cindex pseudo-ops, machine independent
3646 @cindex machine independent directives
3647 All assembler directives have names that begin with a period (@samp{.}).
3648 The rest of the name is letters, usually in lower case.
3650 This chapter discusses directives that are available regardless of the
3651 target machine configuration for the @sc{gnu} assembler.
3653 Some machine configurations provide additional directives.
3654 @xref{Machine Dependencies}.
3657 @ifset machine-directives
3658 @xref{Machine Dependencies} for additional directives.
3663 * Abort:: @code{.abort}
3665 * ABORT:: @code{.ABORT}
3668 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3669 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3670 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3671 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3672 * Byte:: @code{.byte @var{expressions}}
3673 * Comm:: @code{.comm @var{symbol} , @var{length} }
3675 * CFI directives:: @code{.cfi_startproc}, @code{.cfi_endproc}, etc.
3677 * Data:: @code{.data @var{subsection}}
3679 * Def:: @code{.def @var{name}}
3682 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3688 * Double:: @code{.double @var{flonums}}
3689 * Eject:: @code{.eject}
3690 * Else:: @code{.else}
3691 * Elseif:: @code{.elseif}
3694 * Endef:: @code{.endef}
3697 * Endfunc:: @code{.endfunc}
3698 * Endif:: @code{.endif}
3699 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3700 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3702 * Exitm:: @code{.exitm}
3703 * Extern:: @code{.extern}
3704 * Fail:: @code{.fail}
3705 @ifclear no-file-dir
3706 * File:: @code{.file @var{string}}
3709 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3710 * Float:: @code{.float @var{flonums}}
3711 * Func:: @code{.func}
3712 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3714 * Hidden:: @code{.hidden @var{names}}
3717 * hword:: @code{.hword @var{expressions}}
3718 * Ident:: @code{.ident}
3719 * If:: @code{.if @var{absolute expression}}
3720 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3721 * Include:: @code{.include "@var{file}"}
3722 * Int:: @code{.int @var{expressions}}
3724 * Internal:: @code{.internal @var{names}}
3727 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3728 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3729 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3730 * Lflags:: @code{.lflags}
3731 @ifclear no-line-dir
3732 * Line:: @code{.line @var{line-number}}
3735 * Ln:: @code{.ln @var{line-number}}
3736 * Linkonce:: @code{.linkonce [@var{type}]}
3737 * List:: @code{.list}
3738 * Long:: @code{.long @var{expressions}}
3740 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3743 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3744 * MRI:: @code{.mri @var{val}}
3745 * Nolist:: @code{.nolist}
3746 * Octa:: @code{.octa @var{bignums}}
3747 * Org:: @code{.org @var{new-lc} , @var{fill}}
3748 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3750 * PopSection:: @code{.popsection}
3751 * Previous:: @code{.previous}
3754 * Print:: @code{.print @var{string}}
3756 * Protected:: @code{.protected @var{names}}
3759 * Psize:: @code{.psize @var{lines}, @var{columns}}
3760 * Purgem:: @code{.purgem @var{name}}
3762 * PushSection:: @code{.pushsection @var{name}}
3765 * Quad:: @code{.quad @var{bignums}}
3766 * Rept:: @code{.rept @var{count}}
3767 * Sbttl:: @code{.sbttl "@var{subheading}"}
3769 * Scl:: @code{.scl @var{class}}
3772 * Section:: @code{.section @var{name}}
3775 * Set:: @code{.set @var{symbol}, @var{expression}}
3776 * Short:: @code{.short @var{expressions}}
3777 * Single:: @code{.single @var{flonums}}
3779 * Size:: @code{.size [@var{name} , @var{expression}]}
3782 * Skip:: @code{.skip @var{size} , @var{fill}}
3783 * Sleb128:: @code{.sleb128 @var{expressions}}
3784 * Space:: @code{.space @var{size} , @var{fill}}
3786 * Stab:: @code{.stabd, .stabn, .stabs}
3789 * String:: @code{.string "@var{str}"}
3790 * Struct:: @code{.struct @var{expression}}
3792 * SubSection:: @code{.subsection}
3793 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3797 * Tag:: @code{.tag @var{structname}}
3800 * Text:: @code{.text @var{subsection}}
3801 * Title:: @code{.title "@var{heading}"}
3803 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3806 * Uleb128:: @code{.uleb128 @var{expressions}}
3808 * Val:: @code{.val @var{addr}}
3812 * Version:: @code{.version "@var{string}"}
3813 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3814 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3815 * Weak:: @code{.weak @var{names}}
3818 * Word:: @code{.word @var{expressions}}
3819 * Deprecated:: Deprecated Directives
3823 @section @code{.abort}
3825 @cindex @code{abort} directive
3826 @cindex stopping the assembly
3827 This directive stops the assembly immediately. It is for
3828 compatibility with other assemblers. The original idea was that the
3829 assembly language source would be piped into the assembler. If the sender
3830 of the source quit, it could use this directive tells @command{@value{AS}} to
3831 quit also. One day @code{.abort} will not be supported.
3835 @section @code{.ABORT}
3837 @cindex @code{ABORT} directive
3838 When producing COFF output, @command{@value{AS}} accepts this directive as a
3839 synonym for @samp{.abort}.
3842 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
3848 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3850 @cindex padding the location counter
3851 @cindex @code{align} directive
3852 Pad the location counter (in the current subsection) to a particular storage
3853 boundary. The first expression (which must be absolute) is the alignment
3854 required, as described below.
3856 The second expression (also absolute) gives the fill value to be stored in the
3857 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3858 padding bytes are normally zero. However, on some systems, if the section is
3859 marked as containing code and the fill value is omitted, the space is filled
3860 with no-op instructions.
3862 The third expression is also absolute, and is also optional. If it is present,
3863 it is the maximum number of bytes that should be skipped by this alignment
3864 directive. If doing the alignment would require skipping more bytes than the
3865 specified maximum, then the alignment is not done at all. You can omit the
3866 fill value (the second argument) entirely by simply using two commas after the
3867 required alignment; this can be useful if you want the alignment to be filled
3868 with no-op instructions when appropriate.
3870 The way the required alignment is specified varies from system to system.
3871 For the a29k, arc, hppa, i386 using ELF, i860, iq2000, m68k, m88k, or32,
3872 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
3873 alignment request in bytes. For example @samp{.align 8} advances
3874 the location counter until it is a multiple of 8. If the location counter
3875 is already a multiple of 8, no change is needed. For the tic54x, the
3876 first expression is the alignment request in words.
3878 For other systems, including the i386 using a.out format, and the arm and
3879 strongarm, it is the
3880 number of low-order zero bits the location counter must have after
3881 advancement. For example @samp{.align 3} advances the location
3882 counter until it a multiple of 8. If the location counter is already a
3883 multiple of 8, no change is needed.
3885 This inconsistency is due to the different behaviors of the various
3886 native assemblers for these systems which GAS must emulate.
3887 GAS also provides @code{.balign} and @code{.p2align} directives,
3888 described later, which have a consistent behavior across all
3889 architectures (but are specific to GAS).
3892 @section @code{.ascii "@var{string}"}@dots{}
3894 @cindex @code{ascii} directive
3895 @cindex string literals
3896 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3897 separated by commas. It assembles each string (with no automatic
3898 trailing zero byte) into consecutive addresses.
3901 @section @code{.asciz "@var{string}"}@dots{}
3903 @cindex @code{asciz} directive
3904 @cindex zero-terminated strings
3905 @cindex null-terminated strings
3906 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3907 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3910 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3912 @cindex padding the location counter given number of bytes
3913 @cindex @code{balign} directive
3914 Pad the location counter (in the current subsection) to a particular
3915 storage boundary. The first expression (which must be absolute) is the
3916 alignment request in bytes. For example @samp{.balign 8} advances
3917 the location counter until it is a multiple of 8. If the location counter
3918 is already a multiple of 8, no change is needed.
3920 The second expression (also absolute) gives the fill value to be stored in the
3921 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3922 padding bytes are normally zero. However, on some systems, if the section is
3923 marked as containing code and the fill value is omitted, the space is filled
3924 with no-op instructions.
3926 The third expression is also absolute, and is also optional. If it is present,
3927 it is the maximum number of bytes that should be skipped by this alignment
3928 directive. If doing the alignment would require skipping more bytes than the
3929 specified maximum, then the alignment is not done at all. You can omit the
3930 fill value (the second argument) entirely by simply using two commas after the
3931 required alignment; this can be useful if you want the alignment to be filled
3932 with no-op instructions when appropriate.
3934 @cindex @code{balignw} directive
3935 @cindex @code{balignl} directive
3936 The @code{.balignw} and @code{.balignl} directives are variants of the
3937 @code{.balign} directive. The @code{.balignw} directive treats the fill
3938 pattern as a two byte word value. The @code{.balignl} directives treats the
3939 fill pattern as a four byte longword value. For example, @code{.balignw
3940 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3941 filled in with the value 0x368d (the exact placement of the bytes depends upon
3942 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3946 @section @code{.byte @var{expressions}}
3948 @cindex @code{byte} directive
3949 @cindex integers, one byte
3950 @code{.byte} expects zero or more expressions, separated by commas.
3951 Each expression is assembled into the next byte.
3954 @section @code{.comm @var{symbol} , @var{length} }
3956 @cindex @code{comm} directive
3957 @cindex symbol, common
3958 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
3959 common symbol in one object file may be merged with a defined or common symbol
3960 of the same name in another object file. If @code{@value{LD}} does not see a
3961 definition for the symbol--just one or more common symbols--then it will
3962 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
3963 absolute expression. If @code{@value{LD}} sees multiple common symbols with
3964 the same name, and they do not all have the same size, it will allocate space
3965 using the largest size.
3968 When using ELF, the @code{.comm} directive takes an optional third argument.
3969 This is the desired alignment of the symbol, specified as a byte boundary (for
3970 example, an alignment of 16 means that the least significant 4 bits of the
3971 address should be zero). The alignment must be an absolute expression, and it
3972 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
3973 for the common symbol, it will use the alignment when placing the symbol. If
3974 no alignment is specified, @command{@value{AS}} will set the alignment to the
3975 largest power of two less than or equal to the size of the symbol, up to a
3980 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
3981 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
3984 @node CFI directives
3985 @section @code{.cfi_startproc}
3986 @cindex @code{cfi_startproc} directive
3987 @code{.cfi_startproc} is used at the beginning of each function that
3988 should have an entry in @code{.eh_frame}. It initializes some internal
3989 data structures and emits architecture dependent initial CFI instructions.
3990 Don't forget to close the function by
3991 @code{.cfi_endproc}.
3993 @section @code{.cfi_endproc}
3994 @cindex @code{cfi_endproc} directive
3995 @code{.cfi_endproc} is used at the end of a function where it closes its
3996 unwind entry previously opened by
3997 @code{.cfi_startproc}. and emits it to @code{.eh_frame}.
3999 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4000 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4001 address from @var{register} and add @var{offset} to it}.
4003 @section @code{.cfi_def_cfa_register @var{register}}
4004 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4005 now on @var{register} will be used instead of the old one. Offset
4008 @section @code{.cfi_def_cfa_offset @var{offset}}
4009 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4010 remains the same, but @var{offset} is new. Note that it is the
4011 absolute offset that will be added to a defined register to compute
4014 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4015 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4016 value that is added/substracted from the previous offset.
4018 @section @code{.cfi_offset @var{register}, @var{offset}}
4019 Previous value of @var{register} is saved at offset @var{offset} from
4022 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4023 Previous value of @var{register} is saved at offset @var{offset} from
4024 the current CFA register. This is transformed to @code{.cfi_offset}
4025 using the known displacement of the CFA register from the CFA.
4026 This is often easier to use, because the number will match the
4027 code it's annotating.
4029 @section @code{.cfi_window_save}
4030 SPARC register window has been saved.
4032 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4033 Allows the user to add arbitrary bytes to the unwind info. One
4034 might use this to add OS-specific CFI opcodes, or generic CFI
4035 opcodes that GAS does not yet support.
4038 @section @code{.data @var{subsection}}
4040 @cindex @code{data} directive
4041 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4042 end of the data subsection numbered @var{subsection} (which is an
4043 absolute expression). If @var{subsection} is omitted, it defaults
4048 @section @code{.def @var{name}}
4050 @cindex @code{def} directive
4051 @cindex COFF symbols, debugging
4052 @cindex debugging COFF symbols
4053 Begin defining debugging information for a symbol @var{name}; the
4054 definition extends until the @code{.endef} directive is encountered.
4057 This directive is only observed when @command{@value{AS}} is configured for COFF
4058 format output; when producing @code{b.out}, @samp{.def} is recognized,
4065 @section @code{.desc @var{symbol}, @var{abs-expression}}
4067 @cindex @code{desc} directive
4068 @cindex COFF symbol descriptor
4069 @cindex symbol descriptor, COFF
4070 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4071 to the low 16 bits of an absolute expression.
4074 The @samp{.desc} directive is not available when @command{@value{AS}} is
4075 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4076 object format. For the sake of compatibility, @command{@value{AS}} accepts
4077 it, but produces no output, when configured for COFF.
4083 @section @code{.dim}
4085 @cindex @code{dim} directive
4086 @cindex COFF auxiliary symbol information
4087 @cindex auxiliary symbol information, COFF
4088 This directive is generated by compilers to include auxiliary debugging
4089 information in the symbol table. It is only permitted inside
4090 @code{.def}/@code{.endef} pairs.
4093 @samp{.dim} is only meaningful when generating COFF format output; when
4094 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4100 @section @code{.double @var{flonums}}
4102 @cindex @code{double} directive
4103 @cindex floating point numbers (double)
4104 @code{.double} expects zero or more flonums, separated by commas. It
4105 assembles floating point numbers.
4107 The exact kind of floating point numbers emitted depends on how
4108 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4112 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4113 in @sc{ieee} format.
4118 @section @code{.eject}
4120 @cindex @code{eject} directive
4121 @cindex new page, in listings
4122 @cindex page, in listings
4123 @cindex listing control: new page
4124 Force a page break at this point, when generating assembly listings.
4127 @section @code{.else}
4129 @cindex @code{else} directive
4130 @code{.else} is part of the @command{@value{AS}} support for conditional
4131 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
4132 of code to be assembled if the condition for the preceding @code{.if}
4136 @section @code{.elseif}
4138 @cindex @code{elseif} directive
4139 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4140 assembly; @pxref{If,,@code{.if}}. It is shorthand for beginning a new
4141 @code{.if} block that would otherwise fill the entire @code{.else} section.
4144 @section @code{.end}
4146 @cindex @code{end} directive
4147 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4148 process anything in the file past the @code{.end} directive.
4152 @section @code{.endef}
4154 @cindex @code{endef} directive
4155 This directive flags the end of a symbol definition begun with
4159 @samp{.endef} is only meaningful when generating COFF format output; if
4160 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4161 directive but ignores it.
4166 @section @code{.endfunc}
4167 @cindex @code{endfunc} directive
4168 @code{.endfunc} marks the end of a function specified with @code{.func}.
4171 @section @code{.endif}
4173 @cindex @code{endif} directive
4174 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4175 it marks the end of a block of code that is only assembled
4176 conditionally. @xref{If,,@code{.if}}.
4179 @section @code{.equ @var{symbol}, @var{expression}}
4181 @cindex @code{equ} directive
4182 @cindex assigning values to symbols
4183 @cindex symbols, assigning values to
4184 This directive sets the value of @var{symbol} to @var{expression}.
4185 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
4188 The syntax for @code{equ} on the HPPA is
4189 @samp{@var{symbol} .equ @var{expression}}.
4193 @section @code{.equiv @var{symbol}, @var{expression}}
4194 @cindex @code{equiv} directive
4195 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4196 the assembler will signal an error if @var{symbol} is already defined. Note a
4197 symbol which has been referenced but not actually defined is considered to be
4200 Except for the contents of the error message, this is roughly equivalent to
4209 @section @code{.err}
4210 @cindex @code{err} directive
4211 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4212 message and, unless the @option{-Z} option was used, it will not generate an
4213 object file. This can be used to signal error an conditionally compiled code.
4216 @section @code{.exitm}
4217 Exit early from the current macro definition. @xref{Macro}.
4220 @section @code{.extern}
4222 @cindex @code{extern} directive
4223 @code{.extern} is accepted in the source program---for compatibility
4224 with other assemblers---but it is ignored. @command{@value{AS}} treats
4225 all undefined symbols as external.
4228 @section @code{.fail @var{expression}}
4230 @cindex @code{fail} directive
4231 Generates an error or a warning. If the value of the @var{expression} is 500
4232 or more, @command{@value{AS}} will print a warning message. If the value is less
4233 than 500, @command{@value{AS}} will print an error message. The message will
4234 include the value of @var{expression}. This can occasionally be useful inside
4235 complex nested macros or conditional assembly.
4237 @ifclear no-file-dir
4239 @section @code{.file @var{string}}
4241 @cindex @code{file} directive
4242 @cindex logical file name
4243 @cindex file name, logical
4244 @code{.file} tells @command{@value{AS}} that we are about to start a new logical
4245 file. @var{string} is the new file name. In general, the filename is
4246 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4247 to specify an empty file name, you must give the quotes--@code{""}. This
4248 statement may go away in future: it is only recognized to be compatible with
4249 old @command{@value{AS}} programs.
4251 In some configurations of @command{@value{AS}}, @code{.file} has already been
4252 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
4257 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4259 @cindex @code{fill} directive
4260 @cindex writing patterns in memory
4261 @cindex patterns, writing in memory
4262 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4263 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4264 may be zero or more. @var{Size} may be zero or more, but if it is
4265 more than 8, then it is deemed to have the value 8, compatible with
4266 other people's assemblers. The contents of each @var{repeat} bytes
4267 is taken from an 8-byte number. The highest order 4 bytes are
4268 zero. The lowest order 4 bytes are @var{value} rendered in the
4269 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4270 Each @var{size} bytes in a repetition is taken from the lowest order
4271 @var{size} bytes of this number. Again, this bizarre behavior is
4272 compatible with other people's assemblers.
4274 @var{size} and @var{value} are optional.
4275 If the second comma and @var{value} are absent, @var{value} is
4276 assumed zero. If the first comma and following tokens are absent,
4277 @var{size} is assumed to be 1.
4280 @section @code{.float @var{flonums}}
4282 @cindex floating point numbers (single)
4283 @cindex @code{float} directive
4284 This directive assembles zero or more flonums, separated by commas. It
4285 has the same effect as @code{.single}.
4287 The exact kind of floating point numbers emitted depends on how
4288 @command{@value{AS}} is configured.
4289 @xref{Machine Dependencies}.
4293 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4294 in @sc{ieee} format.
4299 @section @code{.func @var{name}[,@var{label}]}
4300 @cindex @code{func} directive
4301 @code{.func} emits debugging information to denote function @var{name}, and
4302 is ignored unless the file is assembled with debugging enabled.
4303 Only @samp{--gstabs[+]} is currently supported.
4304 @var{label} is the entry point of the function and if omitted @var{name}
4305 prepended with the @samp{leading char} is used.
4306 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4307 All functions are currently defined to have @code{void} return type.
4308 The function must be terminated with @code{.endfunc}.
4311 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4313 @cindex @code{global} directive
4314 @cindex symbol, making visible to linker
4315 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4316 @var{symbol} in your partial program, its value is made available to
4317 other partial programs that are linked with it. Otherwise,
4318 @var{symbol} takes its attributes from a symbol of the same name
4319 from another file linked into the same program.
4321 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4322 compatibility with other assemblers.
4325 On the HPPA, @code{.global} is not always enough to make it accessible to other
4326 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4327 @xref{HPPA Directives,, HPPA Assembler Directives}.
4332 @section @code{.hidden @var{names}}
4334 @cindex @code{hidden} directive
4336 This one of the ELF visibility directives. The other two are
4337 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4338 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4340 This directive overrides the named symbols default visibility (which is set by
4341 their binding: local, global or weak). The directive sets the visibility to
4342 @code{hidden} which means that the symbols are not visible to other components.
4343 Such symbols are always considered to be @code{protected} as well.
4347 @section @code{.hword @var{expressions}}
4349 @cindex @code{hword} directive
4350 @cindex integers, 16-bit
4351 @cindex numbers, 16-bit
4352 @cindex sixteen bit integers
4353 This expects zero or more @var{expressions}, and emits
4354 a 16 bit number for each.
4357 This directive is a synonym for @samp{.short}; depending on the target
4358 architecture, it may also be a synonym for @samp{.word}.
4362 This directive is a synonym for @samp{.short}.
4365 This directive is a synonym for both @samp{.short} and @samp{.word}.
4370 @section @code{.ident}
4372 @cindex @code{ident} directive
4373 This directive is used by some assemblers to place tags in object files.
4374 @command{@value{AS}} simply accepts the directive for source-file
4375 compatibility with such assemblers, but does not actually emit anything
4379 @section @code{.if @var{absolute expression}}
4381 @cindex conditional assembly
4382 @cindex @code{if} directive
4383 @code{.if} marks the beginning of a section of code which is only
4384 considered part of the source program being assembled if the argument
4385 (which must be an @var{absolute expression}) is non-zero. The end of
4386 the conditional section of code must be marked by @code{.endif}
4387 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4388 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4389 If you have several conditions to check, @code{.elseif} may be used to avoid
4390 nesting blocks if/else within each subsequent @code{.else} block.
4392 The following variants of @code{.if} are also supported:
4394 @cindex @code{ifdef} directive
4395 @item .ifdef @var{symbol}
4396 Assembles the following section of code if the specified @var{symbol}
4397 has been defined. Note a symbol which has been referenced but not yet defined
4398 is considered to be undefined.
4400 @cindex @code{ifc} directive
4401 @item .ifc @var{string1},@var{string2}
4402 Assembles the following section of code if the two strings are the same. The
4403 strings may be optionally quoted with single quotes. If they are not quoted,
4404 the first string stops at the first comma, and the second string stops at the
4405 end of the line. Strings which contain whitespace should be quoted. The
4406 string comparison is case sensitive.
4408 @cindex @code{ifeq} directive
4409 @item .ifeq @var{absolute expression}
4410 Assembles the following section of code if the argument is zero.
4412 @cindex @code{ifeqs} directive
4413 @item .ifeqs @var{string1},@var{string2}
4414 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4416 @cindex @code{ifge} directive
4417 @item .ifge @var{absolute expression}
4418 Assembles the following section of code if the argument is greater than or
4421 @cindex @code{ifgt} directive
4422 @item .ifgt @var{absolute expression}
4423 Assembles the following section of code if the argument is greater than zero.
4425 @cindex @code{ifle} directive
4426 @item .ifle @var{absolute expression}
4427 Assembles the following section of code if the argument is less than or equal
4430 @cindex @code{iflt} directive
4431 @item .iflt @var{absolute expression}
4432 Assembles the following section of code if the argument is less than zero.
4434 @cindex @code{ifnc} directive
4435 @item .ifnc @var{string1},@var{string2}.
4436 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4437 following section of code if the two strings are not the same.
4439 @cindex @code{ifndef} directive
4440 @cindex @code{ifnotdef} directive
4441 @item .ifndef @var{symbol}
4442 @itemx .ifnotdef @var{symbol}
4443 Assembles the following section of code if the specified @var{symbol}
4444 has not been defined. Both spelling variants are equivalent. Note a symbol
4445 which has been referenced but not yet defined is considered to be undefined.
4447 @cindex @code{ifne} directive
4448 @item .ifne @var{absolute expression}
4449 Assembles the following section of code if the argument is not equal to zero
4450 (in other words, this is equivalent to @code{.if}).
4452 @cindex @code{ifnes} directive
4453 @item .ifnes @var{string1},@var{string2}
4454 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4455 following section of code if the two strings are not the same.
4459 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4461 @cindex @code{incbin} directive
4462 @cindex binary files, including
4463 The @code{incbin} directive includes @var{file} verbatim at the current
4464 location. You can control the search paths used with the @samp{-I} command-line
4465 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4468 The @var{skip} argument skips a number of bytes from the start of the
4469 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4470 read. Note that the data is not aligned in any way, so it is the user's
4471 responsibility to make sure that proper alignment is provided both before and
4472 after the @code{incbin} directive.
4475 @section @code{.include "@var{file}"}
4477 @cindex @code{include} directive
4478 @cindex supporting files, including
4479 @cindex files, including
4480 This directive provides a way to include supporting files at specified
4481 points in your source program. The code from @var{file} is assembled as
4482 if it followed the point of the @code{.include}; when the end of the
4483 included file is reached, assembly of the original file continues. You
4484 can control the search paths used with the @samp{-I} command-line option
4485 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4489 @section @code{.int @var{expressions}}
4491 @cindex @code{int} directive
4492 @cindex integers, 32-bit
4493 Expect zero or more @var{expressions}, of any section, separated by commas.
4494 For each expression, emit a number that, at run time, is the value of that
4495 expression. The byte order and bit size of the number depends on what kind
4496 of target the assembly is for.
4500 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
4501 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4508 @section @code{.internal @var{names}}
4510 @cindex @code{internal} directive
4512 This one of the ELF visibility directives. The other two are
4513 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4514 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4516 This directive overrides the named symbols default visibility (which is set by
4517 their binding: local, global or weak). The directive sets the visibility to
4518 @code{internal} which means that the symbols are considered to be @code{hidden}
4519 (i.e., not visible to other components), and that some extra, processor specific
4520 processing must also be performed upon the symbols as well.
4524 @section @code{.irp @var{symbol},@var{values}}@dots{}
4526 @cindex @code{irp} directive
4527 Evaluate a sequence of statements assigning different values to @var{symbol}.
4528 The sequence of statements starts at the @code{.irp} directive, and is
4529 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4530 set to @var{value}, and the sequence of statements is assembled. If no
4531 @var{value} is listed, the sequence of statements is assembled once, with
4532 @var{symbol} set to the null string. To refer to @var{symbol} within the
4533 sequence of statements, use @var{\symbol}.
4535 For example, assembling
4543 is equivalent to assembling
4552 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4554 @cindex @code{irpc} directive
4555 Evaluate a sequence of statements assigning different values to @var{symbol}.
4556 The sequence of statements starts at the @code{.irpc} directive, and is
4557 terminated by an @code{.endr} directive. For each character in @var{value},
4558 @var{symbol} is set to the character, and the sequence of statements is
4559 assembled. If no @var{value} is listed, the sequence of statements is
4560 assembled once, with @var{symbol} set to the null string. To refer to
4561 @var{symbol} within the sequence of statements, use @var{\symbol}.
4563 For example, assembling
4571 is equivalent to assembling
4580 @section @code{.lcomm @var{symbol} , @var{length}}
4582 @cindex @code{lcomm} directive
4583 @cindex local common symbols
4584 @cindex symbols, local common
4585 Reserve @var{length} (an absolute expression) bytes for a local common
4586 denoted by @var{symbol}. The section and value of @var{symbol} are
4587 those of the new local common. The addresses are allocated in the bss
4588 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4589 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4590 not visible to @code{@value{LD}}.
4593 Some targets permit a third argument to be used with @code{.lcomm}. This
4594 argument specifies the desired alignment of the symbol in the bss section.
4598 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4599 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4603 @section @code{.lflags}
4605 @cindex @code{lflags} directive (ignored)
4606 @command{@value{AS}} accepts this directive, for compatibility with other
4607 assemblers, but ignores it.
4609 @ifclear no-line-dir
4611 @section @code{.line @var{line-number}}
4613 @cindex @code{line} directive
4617 @section @code{.ln @var{line-number}}
4619 @cindex @code{ln} directive
4621 @cindex logical line number
4623 Change the logical line number. @var{line-number} must be an absolute
4624 expression. The next line has that logical line number. Therefore any other
4625 statements on the current line (after a statement separator character) are
4626 reported as on logical line number @var{line-number} @minus{} 1. One day
4627 @command{@value{AS}} will no longer support this directive: it is recognized only
4628 for compatibility with existing assembler programs.
4632 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
4633 not available; use the synonym @code{.ln} in that context.
4638 @ifclear no-line-dir
4639 Even though this is a directive associated with the @code{a.out} or
4640 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
4641 when producing COFF output, and treats @samp{.line} as though it
4642 were the COFF @samp{.ln} @emph{if} it is found outside a
4643 @code{.def}/@code{.endef} pair.
4645 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4646 used by compilers to generate auxiliary symbol information for
4651 @section @code{.linkonce [@var{type}]}
4653 @cindex @code{linkonce} directive
4654 @cindex common sections
4655 Mark the current section so that the linker only includes a single copy of it.
4656 This may be used to include the same section in several different object files,
4657 but ensure that the linker will only include it once in the final output file.
4658 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4659 Duplicate sections are detected based on the section name, so it should be
4662 This directive is only supported by a few object file formats; as of this
4663 writing, the only object file format which supports it is the Portable
4664 Executable format used on Windows NT.
4666 The @var{type} argument is optional. If specified, it must be one of the
4667 following strings. For example:
4671 Not all types may be supported on all object file formats.
4675 Silently discard duplicate sections. This is the default.
4678 Warn if there are duplicate sections, but still keep only one copy.
4681 Warn if any of the duplicates have different sizes.
4684 Warn if any of the duplicates do not have exactly the same contents.
4688 @section @code{.ln @var{line-number}}
4690 @cindex @code{ln} directive
4691 @ifclear no-line-dir
4692 @samp{.ln} is a synonym for @samp{.line}.
4695 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
4696 must be an absolute expression. The next line has that logical
4697 line number, so any other statements on the current line (after a
4698 statement separator character @code{;}) are reported as on logical
4699 line number @var{line-number} @minus{} 1.
4702 This directive is accepted, but ignored, when @command{@value{AS}} is
4703 configured for @code{b.out}; its effect is only associated with COFF
4709 @section @code{.mri @var{val}}
4711 @cindex @code{mri} directive
4712 @cindex MRI mode, temporarily
4713 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
4714 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
4715 affects code assembled until the next @code{.mri} directive, or until the end
4716 of the file. @xref{M, MRI mode, MRI mode}.
4719 @section @code{.list}
4721 @cindex @code{list} directive
4722 @cindex listing control, turning on
4723 Control (in conjunction with the @code{.nolist} directive) whether or
4724 not assembly listings are generated. These two directives maintain an
4725 internal counter (which is zero initially). @code{.list} increments the
4726 counter, and @code{.nolist} decrements it. Assembly listings are
4727 generated whenever the counter is greater than zero.
4729 By default, listings are disabled. When you enable them (with the
4730 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4731 the initial value of the listing counter is one.
4734 @section @code{.long @var{expressions}}
4736 @cindex @code{long} directive
4737 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
4740 @c no one seems to know what this is for or whether this description is
4741 @c what it really ought to do
4743 @section @code{.lsym @var{symbol}, @var{expression}}
4745 @cindex @code{lsym} directive
4746 @cindex symbol, not referenced in assembly
4747 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
4748 the hash table, ensuring it cannot be referenced by name during the
4749 rest of the assembly. This sets the attributes of the symbol to be
4750 the same as the expression value:
4752 @var{other} = @var{descriptor} = 0
4753 @var{type} = @r{(section of @var{expression})}
4754 @var{value} = @var{expression}
4757 The new symbol is not flagged as external.
4761 @section @code{.macro}
4764 The commands @code{.macro} and @code{.endm} allow you to define macros that
4765 generate assembly output. For example, this definition specifies a macro
4766 @code{sum} that puts a sequence of numbers into memory:
4769 .macro sum from=0, to=5
4778 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
4790 @item .macro @var{macname}
4791 @itemx .macro @var{macname} @var{macargs} @dots{}
4792 @cindex @code{macro} directive
4793 Begin the definition of a macro called @var{macname}. If your macro
4794 definition requires arguments, specify their names after the macro name,
4795 separated by commas or spaces. You can supply a default value for any
4796 macro argument by following the name with @samp{=@var{deflt}}. For
4797 example, these are all valid @code{.macro} statements:
4801 Begin the definition of a macro called @code{comm}, which takes no
4804 @item .macro plus1 p, p1
4805 @itemx .macro plus1 p p1
4806 Either statement begins the definition of a macro called @code{plus1},
4807 which takes two arguments; within the macro definition, write
4808 @samp{\p} or @samp{\p1} to evaluate the arguments.
4810 @item .macro reserve_str p1=0 p2
4811 Begin the definition of a macro called @code{reserve_str}, with two
4812 arguments. The first argument has a default value, but not the second.
4813 After the definition is complete, you can call the macro either as
4814 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
4815 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
4816 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
4817 @samp{0}, and @samp{\p2} evaluating to @var{b}).
4820 When you call a macro, you can specify the argument values either by
4821 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
4822 @samp{sum to=17, from=9}.
4825 @cindex @code{endm} directive
4826 Mark the end of a macro definition.
4829 @cindex @code{exitm} directive
4830 Exit early from the current macro definition.
4832 @cindex number of macros executed
4833 @cindex macros, count executed
4835 @command{@value{AS}} maintains a counter of how many macros it has
4836 executed in this pseudo-variable; you can copy that number to your
4837 output with @samp{\@@}, but @emph{only within a macro definition}.
4840 @item LOCAL @var{name} [ , @dots{} ]
4841 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
4842 macro syntax'' with @samp{-a} or @samp{--alternate}.} @xref{Alternate,,
4843 Alternate macro syntax}.
4845 Generate a string replacement for each of the @var{name} arguments, and
4846 replace any instances of @var{name} in each macro expansion. The
4847 replacement string is unique in the assembly, and different for each
4848 separate macro expansion. @code{LOCAL} allows you to write macros that
4849 define symbols, without fear of conflict between separate macro expansions.
4854 @section @code{.nolist}
4856 @cindex @code{nolist} directive
4857 @cindex listing control, turning off
4858 Control (in conjunction with the @code{.list} directive) whether or
4859 not assembly listings are generated. These two directives maintain an
4860 internal counter (which is zero initially). @code{.list} increments the
4861 counter, and @code{.nolist} decrements it. Assembly listings are
4862 generated whenever the counter is greater than zero.
4865 @section @code{.octa @var{bignums}}
4867 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
4868 @cindex @code{octa} directive
4869 @cindex integer, 16-byte
4870 @cindex sixteen byte integer
4871 This directive expects zero or more bignums, separated by commas. For each
4872 bignum, it emits a 16-byte integer.
4874 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
4875 hence @emph{octa}-word for 16 bytes.
4878 @section @code{.org @var{new-lc} , @var{fill}}
4880 @cindex @code{org} directive
4881 @cindex location counter, advancing
4882 @cindex advancing location counter
4883 @cindex current address, advancing
4884 Advance the location counter of the current section to
4885 @var{new-lc}. @var{new-lc} is either an absolute expression or an
4886 expression with the same section as the current subsection. That is,
4887 you can't use @code{.org} to cross sections: if @var{new-lc} has the
4888 wrong section, the @code{.org} directive is ignored. To be compatible
4889 with former assemblers, if the section of @var{new-lc} is absolute,
4890 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
4891 is the same as the current subsection.
4893 @code{.org} may only increase the location counter, or leave it
4894 unchanged; you cannot use @code{.org} to move the location counter
4897 @c double negative used below "not undefined" because this is a specific
4898 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
4899 @c section. doc@cygnus.com 18feb91
4900 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
4901 may not be undefined. If you really detest this restriction we eagerly await
4902 a chance to share your improved assembler.
4904 Beware that the origin is relative to the start of the section, not
4905 to the start of the subsection. This is compatible with other
4906 people's assemblers.
4908 When the location counter (of the current subsection) is advanced, the
4909 intervening bytes are filled with @var{fill} which should be an
4910 absolute expression. If the comma and @var{fill} are omitted,
4911 @var{fill} defaults to zero.
4914 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4916 @cindex padding the location counter given a power of two
4917 @cindex @code{p2align} directive
4918 Pad the location counter (in the current subsection) to a particular
4919 storage boundary. The first expression (which must be absolute) is the
4920 number of low-order zero bits the location counter must have after
4921 advancement. For example @samp{.p2align 3} advances the location
4922 counter until it a multiple of 8. If the location counter is already a
4923 multiple of 8, no change is needed.
4925 The second expression (also absolute) gives the fill value to be stored in the
4926 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4927 padding bytes are normally zero. However, on some systems, if the section is
4928 marked as containing code and the fill value is omitted, the space is filled
4929 with no-op instructions.
4931 The third expression is also absolute, and is also optional. If it is present,
4932 it is the maximum number of bytes that should be skipped by this alignment
4933 directive. If doing the alignment would require skipping more bytes than the
4934 specified maximum, then the alignment is not done at all. You can omit the
4935 fill value (the second argument) entirely by simply using two commas after the
4936 required alignment; this can be useful if you want the alignment to be filled
4937 with no-op instructions when appropriate.
4939 @cindex @code{p2alignw} directive
4940 @cindex @code{p2alignl} directive
4941 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
4942 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
4943 pattern as a two byte word value. The @code{.p2alignl} directives treats the
4944 fill pattern as a four byte longword value. For example, @code{.p2alignw
4945 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4946 filled in with the value 0x368d (the exact placement of the bytes depends upon
4947 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4952 @section @code{.previous}
4954 @cindex @code{previous} directive
4955 @cindex Section Stack
4956 This is one of the ELF section stack manipulation directives. The others are
4957 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
4958 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
4959 (@pxref{PopSection}).
4961 This directive swaps the current section (and subsection) with most recently
4962 referenced section (and subsection) prior to this one. Multiple
4963 @code{.previous} directives in a row will flip between two sections (and their
4966 In terms of the section stack, this directive swaps the current section with
4967 the top section on the section stack.
4972 @section @code{.popsection}
4974 @cindex @code{popsection} directive
4975 @cindex Section Stack
4976 This is one of the ELF section stack manipulation directives. The others are
4977 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
4978 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
4981 This directive replaces the current section (and subsection) with the top
4982 section (and subsection) on the section stack. This section is popped off the
4987 @section @code{.print @var{string}}
4989 @cindex @code{print} directive
4990 @command{@value{AS}} will print @var{string} on the standard output during
4991 assembly. You must put @var{string} in double quotes.
4995 @section @code{.protected @var{names}}
4997 @cindex @code{protected} directive
4999 This one of the ELF visibility directives. The other two are
5000 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5002 This directive overrides the named symbols default visibility (which is set by
5003 their binding: local, global or weak). The directive sets the visibility to
5004 @code{protected} which means that any references to the symbols from within the
5005 components that defines them must be resolved to the definition in that
5006 component, even if a definition in another component would normally preempt
5011 @section @code{.psize @var{lines} , @var{columns}}
5013 @cindex @code{psize} directive
5014 @cindex listing control: paper size
5015 @cindex paper size, for listings
5016 Use this directive to declare the number of lines---and, optionally, the
5017 number of columns---to use for each page, when generating listings.
5019 If you do not use @code{.psize}, listings use a default line-count
5020 of 60. You may omit the comma and @var{columns} specification; the
5021 default width is 200 columns.
5023 @command{@value{AS}} generates formfeeds whenever the specified number of
5024 lines is exceeded (or whenever you explicitly request one, using
5027 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5028 those explicitly specified with @code{.eject}.
5031 @section @code{.purgem @var{name}}
5033 @cindex @code{purgem} directive
5034 Undefine the macro @var{name}, so that later uses of the string will not be
5035 expanded. @xref{Macro}.
5039 @section @code{.pushsection @var{name} , @var{subsection}}
5041 @cindex @code{pushsection} directive
5042 @cindex Section Stack
5043 This is one of the ELF section stack manipulation directives. The others are
5044 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5045 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5048 This directive is a synonym for @code{.section}. It pushes the current section
5049 (and subsection) onto the top of the section stack, and then replaces the
5050 current section and subsection with @code{name} and @code{subsection}.
5054 @section @code{.quad @var{bignums}}
5056 @cindex @code{quad} directive
5057 @code{.quad} expects zero or more bignums, separated by commas. For
5058 each bignum, it emits
5060 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5061 warning message; and just takes the lowest order 8 bytes of the bignum.
5062 @cindex eight-byte integer
5063 @cindex integer, 8-byte
5065 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5066 hence @emph{quad}-word for 8 bytes.
5069 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5070 warning message; and just takes the lowest order 16 bytes of the bignum.
5071 @cindex sixteen-byte integer
5072 @cindex integer, 16-byte
5076 @section @code{.rept @var{count}}
5078 @cindex @code{rept} directive
5079 Repeat the sequence of lines between the @code{.rept} directive and the next
5080 @code{.endr} directive @var{count} times.
5082 For example, assembling
5090 is equivalent to assembling
5099 @section @code{.sbttl "@var{subheading}"}
5101 @cindex @code{sbttl} directive
5102 @cindex subtitles for listings
5103 @cindex listing control: subtitle
5104 Use @var{subheading} as the title (third line, immediately after the
5105 title line) when generating assembly listings.
5107 This directive affects subsequent pages, as well as the current page if
5108 it appears within ten lines of the top of a page.
5112 @section @code{.scl @var{class}}
5114 @cindex @code{scl} directive
5115 @cindex symbol storage class (COFF)
5116 @cindex COFF symbol storage class
5117 Set the storage-class value for a symbol. This directive may only be
5118 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5119 whether a symbol is static or external, or it may record further
5120 symbolic debugging information.
5123 The @samp{.scl} directive is primarily associated with COFF output; when
5124 configured to generate @code{b.out} output format, @command{@value{AS}}
5125 accepts this directive but ignores it.
5131 @section @code{.section @var{name}}
5133 @cindex named section
5134 Use the @code{.section} directive to assemble the following code into a section
5137 This directive is only supported for targets that actually support arbitrarily
5138 named sections; on @code{a.out} targets, for example, it is not accepted, even
5139 with a standard @code{a.out} section name.
5143 @c only print the extra heading if both COFF and ELF are set
5144 @subheading COFF Version
5147 @cindex @code{section} directive (COFF version)
5148 For COFF targets, the @code{.section} directive is used in one of the following
5152 .section @var{name}[, "@var{flags}"]
5153 .section @var{name}[, @var{subsegment}]
5156 If the optional argument is quoted, it is taken as flags to use for the
5157 section. Each flag is a single character. The following flags are recognized:
5160 bss section (uninitialized data)
5162 section is not loaded
5172 shared section (meaningful for PE targets)
5174 ignored. (For compatibility with the ELF version)
5177 If no flags are specified, the default flags depend upon the section name. If
5178 the section name is not recognized, the default will be for the section to be
5179 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5180 from the section, rather than adding them, so if they are used on their own it
5181 will be as if no flags had been specified at all.
5183 If the optional argument to the @code{.section} directive is not quoted, it is
5184 taken as a subsegment number (@pxref{Sub-Sections}).
5189 @c only print the extra heading if both COFF and ELF are set
5190 @subheading ELF Version
5193 @cindex Section Stack
5194 This is one of the ELF section stack manipulation directives. The others are
5195 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5196 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5197 @code{.previous} (@pxref{Previous}).
5199 @cindex @code{section} directive (ELF version)
5200 For ELF targets, the @code{.section} directive is used like this:
5203 .section @var{name} [, "@var{flags}"[, @@@var{type}[, @@@var{entsize}]]]
5206 The optional @var{flags} argument is a quoted string which may contain any
5207 combination of the following characters:
5210 section is allocatable
5214 section is executable
5216 section is mergeable
5218 section contains zero terminated strings
5221 The optional @var{type} argument may contain one of the following constants:
5224 section contains data
5226 section does not contain data (i.e., section only occupies space)
5229 Note on targets where the @code{@@} character is the start of a comment (eg
5230 ARM) then another character is used instead. For example the ARM port uses the
5233 If @var{flags} contains @code{M} flag, @var{type} argument must be specified
5234 as well as @var{entsize} argument. Sections with @code{M} flag but not
5235 @code{S} flag must contain fixed size constants, each @var{entsize} octets
5236 long. Sections with both @code{M} and @code{S} must contain zero terminated
5237 strings where each character is @var{entsize} bytes long. The linker may remove
5238 duplicates within sections with the same name, same entity size and same flags.
5240 If no flags are specified, the default flags depend upon the section name. If
5241 the section name is not recognized, the default will be for the section to have
5242 none of the above flags: it will not be allocated in memory, nor writable, nor
5243 executable. The section will contain data.
5245 For ELF targets, the assembler supports another type of @code{.section}
5246 directive for compatibility with the Solaris assembler:
5249 .section "@var{name}"[, @var{flags}...]
5252 Note that the section name is quoted. There may be a sequence of comma
5256 section is allocatable
5260 section is executable
5263 This directive replaces the current section and subsection. The replaced
5264 section and subsection are pushed onto the section stack. See the contents of
5265 the gas testsuite directory @code{gas/testsuite/gas/elf} for some examples of
5266 how this directive and the other section stack directives work.
5271 @section @code{.set @var{symbol}, @var{expression}}
5273 @cindex @code{set} directive
5274 @cindex symbol value, setting
5275 Set the value of @var{symbol} to @var{expression}. This
5276 changes @var{symbol}'s value and type to conform to
5277 @var{expression}. If @var{symbol} was flagged as external, it remains
5278 flagged (@pxref{Symbol Attributes}).
5280 You may @code{.set} a symbol many times in the same assembly.
5282 If you @code{.set} a global symbol, the value stored in the object
5283 file is the last value stored into it.
5286 The syntax for @code{set} on the HPPA is
5287 @samp{@var{symbol} .set @var{expression}}.
5291 @section @code{.short @var{expressions}}
5293 @cindex @code{short} directive
5295 @code{.short} is normally the same as @samp{.word}.
5296 @xref{Word,,@code{.word}}.
5298 In some configurations, however, @code{.short} and @code{.word} generate
5299 numbers of different lengths; @pxref{Machine Dependencies}.
5303 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5306 This expects zero or more @var{expressions}, and emits
5307 a 16 bit number for each.
5312 @section @code{.single @var{flonums}}
5314 @cindex @code{single} directive
5315 @cindex floating point numbers (single)
5316 This directive assembles zero or more flonums, separated by commas. It
5317 has the same effect as @code{.float}.
5319 The exact kind of floating point numbers emitted depends on how
5320 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5324 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5325 numbers in @sc{ieee} format.
5331 @section @code{.size}
5333 This directive is used to set the size associated with a symbol.
5337 @c only print the extra heading if both COFF and ELF are set
5338 @subheading COFF Version
5341 @cindex @code{size} directive (COFF version)
5342 For COFF targets, the @code{.size} directive is only permitted inside
5343 @code{.def}/@code{.endef} pairs. It is used like this:
5346 .size @var{expression}
5350 @samp{.size} is only meaningful when generating COFF format output; when
5351 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5358 @c only print the extra heading if both COFF and ELF are set
5359 @subheading ELF Version
5362 @cindex @code{size} directive (ELF version)
5363 For ELF targets, the @code{.size} directive is used like this:
5366 .size @var{name} , @var{expression}
5369 This directive sets the size associated with a symbol @var{name}.
5370 The size in bytes is computed from @var{expression} which can make use of label
5371 arithmetic. This directive is typically used to set the size of function
5377 @section @code{.sleb128 @var{expressions}}
5379 @cindex @code{sleb128} directive
5380 @var{sleb128} stands for ``signed little endian base 128.'' This is a
5381 compact, variable length representation of numbers used by the DWARF
5382 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
5384 @ifclear no-space-dir
5386 @section @code{.skip @var{size} , @var{fill}}
5388 @cindex @code{skip} directive
5389 @cindex filling memory
5390 This directive emits @var{size} bytes, each of value @var{fill}. Both
5391 @var{size} and @var{fill} are absolute expressions. If the comma and
5392 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
5396 @section @code{.space @var{size} , @var{fill}}
5398 @cindex @code{space} directive
5399 @cindex filling memory
5400 This directive emits @var{size} bytes, each of value @var{fill}. Both
5401 @var{size} and @var{fill} are absolute expressions. If the comma
5402 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
5407 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
5408 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
5409 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
5410 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
5419 @section @code{.space}
5420 @cindex @code{space} directive
5422 On the AMD 29K, this directive is ignored; it is accepted for
5423 compatibility with other AMD 29K assemblers.
5426 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
5427 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
5433 @section @code{.stabd, .stabn, .stabs}
5435 @cindex symbolic debuggers, information for
5436 @cindex @code{stab@var{x}} directives
5437 There are three directives that begin @samp{.stab}.
5438 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
5439 The symbols are not entered in the @command{@value{AS}} hash table: they
5440 cannot be referenced elsewhere in the source file.
5441 Up to five fields are required:
5445 This is the symbol's name. It may contain any character except
5446 @samp{\000}, so is more general than ordinary symbol names. Some
5447 debuggers used to code arbitrarily complex structures into symbol names
5451 An absolute expression. The symbol's type is set to the low 8 bits of
5452 this expression. Any bit pattern is permitted, but @code{@value{LD}}
5453 and debuggers choke on silly bit patterns.
5456 An absolute expression. The symbol's ``other'' attribute is set to the
5457 low 8 bits of this expression.
5460 An absolute expression. The symbol's descriptor is set to the low 16
5461 bits of this expression.
5464 An absolute expression which becomes the symbol's value.
5467 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
5468 or @code{.stabs} statement, the symbol has probably already been created;
5469 you get a half-formed symbol in your object file. This is
5470 compatible with earlier assemblers!
5473 @cindex @code{stabd} directive
5474 @item .stabd @var{type} , @var{other} , @var{desc}
5476 The ``name'' of the symbol generated is not even an empty string.
5477 It is a null pointer, for compatibility. Older assemblers used a
5478 null pointer so they didn't waste space in object files with empty
5481 The symbol's value is set to the location counter,
5482 relocatably. When your program is linked, the value of this symbol
5483 is the address of the location counter when the @code{.stabd} was
5486 @cindex @code{stabn} directive
5487 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
5488 The name of the symbol is set to the empty string @code{""}.
5490 @cindex @code{stabs} directive
5491 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
5492 All five fields are specified.
5498 @section @code{.string} "@var{str}"
5500 @cindex string, copying to object file
5501 @cindex @code{string} directive
5503 Copy the characters in @var{str} to the object file. You may specify more than
5504 one string to copy, separated by commas. Unless otherwise specified for a
5505 particular machine, the assembler marks the end of each string with a 0 byte.
5506 You can use any of the escape sequences described in @ref{Strings,,Strings}.
5509 @section @code{.struct @var{expression}}
5511 @cindex @code{struct} directive
5512 Switch to the absolute section, and set the section offset to @var{expression},
5513 which must be an absolute expression. You might use this as follows:
5522 This would define the symbol @code{field1} to have the value 0, the symbol
5523 @code{field2} to have the value 4, and the symbol @code{field3} to have the
5524 value 8. Assembly would be left in the absolute section, and you would need to
5525 use a @code{.section} directive of some sort to change to some other section
5526 before further assembly.
5530 @section @code{.subsection @var{name}}
5532 @cindex @code{subsection} directive
5533 @cindex Section Stack
5534 This is one of the ELF section stack manipulation directives. The others are
5535 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
5536 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5539 This directive replaces the current subsection with @code{name}. The current
5540 section is not changed. The replaced subsection is put onto the section stack
5541 in place of the then current top of stack subsection.
5546 @section @code{.symver}
5547 @cindex @code{symver} directive
5548 @cindex symbol versioning
5549 @cindex versions of symbols
5550 Use the @code{.symver} directive to bind symbols to specific version nodes
5551 within a source file. This is only supported on ELF platforms, and is
5552 typically used when assembling files to be linked into a shared library.
5553 There are cases where it may make sense to use this in objects to be bound
5554 into an application itself so as to override a versioned symbol from a
5557 For ELF targets, the @code{.symver} directive can be used like this:
5559 .symver @var{name}, @var{name2@@nodename}
5561 If the symbol @var{name} is defined within the file
5562 being assembled, the @code{.symver} directive effectively creates a symbol
5563 alias with the name @var{name2@@nodename}, and in fact the main reason that we
5564 just don't try and create a regular alias is that the @var{@@} character isn't
5565 permitted in symbol names. The @var{name2} part of the name is the actual name
5566 of the symbol by which it will be externally referenced. The name @var{name}
5567 itself is merely a name of convenience that is used so that it is possible to
5568 have definitions for multiple versions of a function within a single source
5569 file, and so that the compiler can unambiguously know which version of a
5570 function is being mentioned. The @var{nodename} portion of the alias should be
5571 the name of a node specified in the version script supplied to the linker when
5572 building a shared library. If you are attempting to override a versioned
5573 symbol from a shared library, then @var{nodename} should correspond to the
5574 nodename of the symbol you are trying to override.
5576 If the symbol @var{name} is not defined within the file being assembled, all
5577 references to @var{name} will be changed to @var{name2@@nodename}. If no
5578 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
5581 Another usage of the @code{.symver} directive is:
5583 .symver @var{name}, @var{name2@@@@nodename}
5585 In this case, the symbol @var{name} must exist and be defined within
5586 the file being assembled. It is similar to @var{name2@@nodename}. The
5587 difference is @var{name2@@@@nodename} will also be used to resolve
5588 references to @var{name2} by the linker.
5590 The third usage of the @code{.symver} directive is:
5592 .symver @var{name}, @var{name2@@@@@@nodename}
5594 When @var{name} is not defined within the
5595 file being assembled, it is treated as @var{name2@@nodename}. When
5596 @var{name} is defined within the file being assembled, the symbol
5597 name, @var{name}, will be changed to @var{name2@@@@nodename}.
5602 @section @code{.tag @var{structname}}
5604 @cindex COFF structure debugging
5605 @cindex structure debugging, COFF
5606 @cindex @code{tag} directive
5607 This directive is generated by compilers to include auxiliary debugging
5608 information in the symbol table. It is only permitted inside
5609 @code{.def}/@code{.endef} pairs. Tags are used to link structure
5610 definitions in the symbol table with instances of those structures.
5613 @samp{.tag} is only used when generating COFF format output; when
5614 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5620 @section @code{.text @var{subsection}}
5622 @cindex @code{text} directive
5623 Tells @command{@value{AS}} to assemble the following statements onto the end of
5624 the text subsection numbered @var{subsection}, which is an absolute
5625 expression. If @var{subsection} is omitted, subsection number zero
5629 @section @code{.title "@var{heading}"}
5631 @cindex @code{title} directive
5632 @cindex listing control: title line
5633 Use @var{heading} as the title (second line, immediately after the
5634 source file name and pagenumber) when generating assembly listings.
5636 This directive affects subsequent pages, as well as the current page if
5637 it appears within ten lines of the top of a page.
5641 @section @code{.type}
5643 This directive is used to set the type of a symbol.
5647 @c only print the extra heading if both COFF and ELF are set
5648 @subheading COFF Version
5651 @cindex COFF symbol type
5652 @cindex symbol type, COFF
5653 @cindex @code{type} directive (COFF version)
5654 For COFF targets, this directive is permitted only within
5655 @code{.def}/@code{.endef} pairs. It is used like this:
5661 This records the integer @var{int} as the type attribute of a symbol table
5665 @samp{.type} is associated only with COFF format output; when
5666 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
5667 directive but ignores it.
5673 @c only print the extra heading if both COFF and ELF are set
5674 @subheading ELF Version
5677 @cindex ELF symbol type
5678 @cindex symbol type, ELF
5679 @cindex @code{type} directive (ELF version)
5680 For ELF targets, the @code{.type} directive is used like this:
5683 .type @var{name} , @var{type description}
5686 This sets the type of symbol @var{name} to be either a
5687 function symbol or an object symbol. There are five different syntaxes
5688 supported for the @var{type description} field, in order to provide
5689 compatibility with various other assemblers. The syntaxes supported are:
5692 .type <name>,#function
5693 .type <name>,#object
5695 .type <name>,@@function
5696 .type <name>,@@object
5698 .type <name>,%function
5699 .type <name>,%object
5701 .type <name>,"function"
5702 .type <name>,"object"
5704 .type <name> STT_FUNCTION
5705 .type <name> STT_OBJECT
5711 @section @code{.uleb128 @var{expressions}}
5713 @cindex @code{uleb128} directive
5714 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
5715 compact, variable length representation of numbers used by the DWARF
5716 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
5720 @section @code{.val @var{addr}}
5722 @cindex @code{val} directive
5723 @cindex COFF value attribute
5724 @cindex value attribute, COFF
5725 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5726 records the address @var{addr} as the value attribute of a symbol table
5730 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
5731 configured for @code{b.out}, it accepts this directive but ignores it.
5737 @section @code{.version "@var{string}"}
5739 @cindex @code{version} directive
5740 This directive creates a @code{.note} section and places into it an ELF
5741 formatted note of type NT_VERSION. The note's name is set to @code{string}.
5746 @section @code{.vtable_entry @var{table}, @var{offset}}
5748 @cindex @code{vtable_entry}
5749 This directive finds or creates a symbol @code{table} and creates a
5750 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
5753 @section @code{.vtable_inherit @var{child}, @var{parent}}
5755 @cindex @code{vtable_inherit}
5756 This directive finds the symbol @code{child} and finds or creates the symbol
5757 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
5758 parent whose addend is the value of the child symbol. As a special case the
5759 parent name of @code{0} is treated as refering the @code{*ABS*} section.
5764 @section @code{.weak @var{names}}
5766 @cindex @code{weak} directive
5767 This directive sets the weak attribute on the comma separated list of symbol
5768 @code{names}. If the symbols do not already exist, they will be created.
5772 @section @code{.word @var{expressions}}
5774 @cindex @code{word} directive
5775 This directive expects zero or more @var{expressions}, of any section,
5776 separated by commas.
5779 For each expression, @command{@value{AS}} emits a 32-bit number.
5782 For each expression, @command{@value{AS}} emits a 16-bit number.
5787 The size of the number emitted, and its byte order,
5788 depend on what target computer the assembly is for.
5791 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
5792 @c happen---32-bit addressability, period; no long/short jumps.
5793 @ifset DIFF-TBL-KLUGE
5794 @cindex difference tables altered
5795 @cindex altered difference tables
5797 @emph{Warning: Special Treatment to support Compilers}
5801 Machines with a 32-bit address space, but that do less than 32-bit
5802 addressing, require the following special treatment. If the machine of
5803 interest to you does 32-bit addressing (or doesn't require it;
5804 @pxref{Machine Dependencies}), you can ignore this issue.
5807 In order to assemble compiler output into something that works,
5808 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
5809 Directives of the form @samp{.word sym1-sym2} are often emitted by
5810 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
5811 directive of the form @samp{.word sym1-sym2}, and the difference between
5812 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
5813 creates a @dfn{secondary jump table}, immediately before the next label.
5814 This secondary jump table is preceded by a short-jump to the
5815 first byte after the secondary table. This short-jump prevents the flow
5816 of control from accidentally falling into the new table. Inside the
5817 table is a long-jump to @code{sym2}. The original @samp{.word}
5818 contains @code{sym1} minus the address of the long-jump to
5821 If there were several occurrences of @samp{.word sym1-sym2} before the
5822 secondary jump table, all of them are adjusted. If there was a
5823 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
5824 long-jump to @code{sym4} is included in the secondary jump table,
5825 and the @code{.word} directives are adjusted to contain @code{sym3}
5826 minus the address of the long-jump to @code{sym4}; and so on, for as many
5827 entries in the original jump table as necessary.
5830 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
5831 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
5832 assembly language programmers.
5835 @c end DIFF-TBL-KLUGE
5838 @section Deprecated Directives
5840 @cindex deprecated directives
5841 @cindex obsolescent directives
5842 One day these directives won't work.
5843 They are included for compatibility with older assemblers.
5850 @node Machine Dependencies
5851 @chapter Machine Dependent Features
5853 @cindex machine dependencies
5854 The machine instruction sets are (almost by definition) different on
5855 each machine where @command{@value{AS}} runs. Floating point representations
5856 vary as well, and @command{@value{AS}} often supports a few additional
5857 directives or command-line options for compatibility with other
5858 assemblers on a particular platform. Finally, some versions of
5859 @command{@value{AS}} support special pseudo-instructions for branch
5862 This chapter discusses most of these differences, though it does not
5863 include details on any machine's instruction set. For details on that
5864 subject, see the hardware manufacturer's manual.
5868 * AMD29K-Dependent:: AMD 29K Dependent Features
5871 * Alpha-Dependent:: Alpha Dependent Features
5874 * ARC-Dependent:: ARC Dependent Features
5877 * ARM-Dependent:: ARM Dependent Features
5880 * CRIS-Dependent:: CRIS Dependent Features
5883 * D10V-Dependent:: D10V Dependent Features
5886 * D30V-Dependent:: D30V Dependent Features
5889 * H8/300-Dependent:: Renesas H8/300 Dependent Features
5892 * H8/500-Dependent:: Renesas H8/500 Dependent Features
5895 * HPPA-Dependent:: HPPA Dependent Features
5898 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
5901 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
5904 * i860-Dependent:: Intel 80860 Dependent Features
5907 * i960-Dependent:: Intel 80960 Dependent Features
5910 * IP2K-Dependent:: IP2K Dependent Features
5913 * M32R-Dependent:: M32R Dependent Features
5916 * M68K-Dependent:: M680x0 Dependent Features
5919 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
5922 * M88K-Dependent:: M880x0 Dependent Features
5925 * MIPS-Dependent:: MIPS Dependent Features
5928 * MMIX-Dependent:: MMIX Dependent Features
5931 * MSP430-Dependent:: MSP430 Dependent Features
5934 * SH-Dependent:: Renesas / SuperH SH Dependent Features
5935 * SH64-Dependent:: SuperH SH64 Dependent Features
5938 * PDP-11-Dependent:: PDP-11 Dependent Features
5941 * PJ-Dependent:: picoJava Dependent Features
5944 * PPC-Dependent:: PowerPC Dependent Features
5947 * Sparc-Dependent:: SPARC Dependent Features
5950 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
5953 * V850-Dependent:: V850 Dependent Features
5956 * Xtensa-Dependent:: Xtensa Dependent Features
5959 * Z8000-Dependent:: Z8000 Dependent Features
5962 * Vax-Dependent:: VAX Dependent Features
5969 @c The following major nodes are *sections* in the GENERIC version, *chapters*
5970 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
5971 @c peculiarity: to preserve cross-references, there must be a node called
5972 @c "Machine Dependencies". Hence the conditional nodenames in each
5973 @c major node below. Node defaulting in makeinfo requires adjacency of
5974 @c node and sectioning commands; hence the repetition of @chapter BLAH
5975 @c in both conditional blocks.
5978 @include c-a29k.texi
5982 @include c-alpha.texi
5994 @include c-cris.texi
5999 @node Machine Dependencies
6000 @chapter Machine Dependent Features
6002 The machine instruction sets are different on each Renesas chip family,
6003 and there are also some syntax differences among the families. This
6004 chapter describes the specific @command{@value{AS}} features for each
6008 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6009 * H8/500-Dependent:: Renesas H8/500 Dependent Features
6010 * SH-Dependent:: Renesas SH Dependent Features
6017 @include c-d10v.texi
6021 @include c-d30v.texi
6025 @include c-h8300.texi
6029 @include c-h8500.texi
6033 @include c-hppa.texi
6037 @include c-i370.texi
6041 @include c-i386.texi
6045 @include c-i860.texi
6049 @include c-i960.texi
6053 @include c-ia64.texi
6057 @include c-ip2k.texi
6061 @include c-m32r.texi
6065 @include c-m68k.texi
6069 @include c-m68hc11.texi
6073 @include c-m88k.texi
6077 @include c-mips.texi
6081 @include c-mmix.texi
6085 @include c-msp430.texi
6089 @include c-ns32k.texi
6093 @include c-pdp11.texi
6106 @include c-sh64.texi
6110 @include c-sparc.texi
6114 @include c-tic54x.texi
6126 @include c-v850.texi
6130 @include c-xtensa.texi
6134 @c reverse effect of @down at top of generic Machine-Dep chapter
6138 @node Reporting Bugs
6139 @chapter Reporting Bugs
6140 @cindex bugs in assembler
6141 @cindex reporting bugs in assembler
6143 Your bug reports play an essential role in making @command{@value{AS}} reliable.
6145 Reporting a bug may help you by bringing a solution to your problem, or it may
6146 not. But in any case the principal function of a bug report is to help the
6147 entire community by making the next version of @command{@value{AS}} work better.
6148 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
6150 In order for a bug report to serve its purpose, you must include the
6151 information that enables us to fix the bug.
6154 * Bug Criteria:: Have you found a bug?
6155 * Bug Reporting:: How to report bugs
6159 @section Have You Found a Bug?
6160 @cindex bug criteria
6162 If you are not sure whether you have found a bug, here are some guidelines:
6165 @cindex fatal signal
6166 @cindex assembler crash
6167 @cindex crash of assembler
6169 If the assembler gets a fatal signal, for any input whatever, that is a
6170 @command{@value{AS}} bug. Reliable assemblers never crash.
6172 @cindex error on valid input
6174 If @command{@value{AS}} produces an error message for valid input, that is a bug.
6176 @cindex invalid input
6178 If @command{@value{AS}} does not produce an error message for invalid input, that
6179 is a bug. However, you should note that your idea of ``invalid input'' might
6180 be our idea of ``an extension'' or ``support for traditional practice''.
6183 If you are an experienced user of assemblers, your suggestions for improvement
6184 of @command{@value{AS}} are welcome in any case.
6188 @section How to Report Bugs
6190 @cindex assembler bugs, reporting
6192 A number of companies and individuals offer support for @sc{gnu} products. If
6193 you obtained @command{@value{AS}} from a support organization, we recommend you
6194 contact that organization first.
6196 You can find contact information for many support companies and
6197 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
6200 In any event, we also recommend that you send bug reports for @command{@value{AS}}
6201 to @samp{bug-binutils@@gnu.org}.
6203 The fundamental principle of reporting bugs usefully is this:
6204 @strong{report all the facts}. If you are not sure whether to state a
6205 fact or leave it out, state it!
6207 Often people omit facts because they think they know what causes the problem
6208 and assume that some details do not matter. Thus, you might assume that the
6209 name of a symbol you use in an example does not matter. Well, probably it does
6210 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
6211 happens to fetch from the location where that name is stored in memory;
6212 perhaps, if the name were different, the contents of that location would fool
6213 the assembler into doing the right thing despite the bug. Play it safe and
6214 give a specific, complete example. That is the easiest thing for you to do,
6215 and the most helpful.
6217 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
6218 it is new to us. Therefore, always write your bug reports on the assumption
6219 that the bug has not been reported previously.
6221 Sometimes people give a few sketchy facts and ask, ``Does this ring a
6222 bell?'' This cannot help us fix a bug, so it is basically useless. We
6223 respond by asking for enough details to enable us to investigate.
6224 You might as well expedite matters by sending them to begin with.
6226 To enable us to fix the bug, you should include all these things:
6230 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
6231 it with the @samp{--version} argument.
6233 Without this, we will not know whether there is any point in looking for
6234 the bug in the current version of @command{@value{AS}}.
6237 Any patches you may have applied to the @command{@value{AS}} source.
6240 The type of machine you are using, and the operating system name and
6244 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
6248 The command arguments you gave the assembler to assemble your example and
6249 observe the bug. To guarantee you will not omit something important, list them
6250 all. A copy of the Makefile (or the output from make) is sufficient.
6252 If we were to try to guess the arguments, we would probably guess wrong
6253 and then we might not encounter the bug.
6256 A complete input file that will reproduce the bug. If the bug is observed when
6257 the assembler is invoked via a compiler, send the assembler source, not the
6258 high level language source. Most compilers will produce the assembler source
6259 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
6260 the options @samp{-v --save-temps}; this will save the assembler source in a
6261 file with an extension of @file{.s}, and also show you exactly how
6262 @command{@value{AS}} is being run.
6265 A description of what behavior you observe that you believe is
6266 incorrect. For example, ``It gets a fatal signal.''
6268 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
6269 will certainly notice it. But if the bug is incorrect output, we might not
6270 notice unless it is glaringly wrong. You might as well not give us a chance to
6273 Even if the problem you experience is a fatal signal, you should still say so
6274 explicitly. Suppose something strange is going on, such as, your copy of
6275 @command{@value{AS}} is out of synch, or you have encountered a bug in the C
6276 library on your system. (This has happened!) Your copy might crash and ours
6277 would not. If you told us to expect a crash, then when ours fails to crash, we
6278 would know that the bug was not happening for us. If you had not told us to
6279 expect a crash, then we would not be able to draw any conclusion from our
6283 If you wish to suggest changes to the @command{@value{AS}} source, send us context
6284 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
6285 option. Always send diffs from the old file to the new file. If you even
6286 discuss something in the @command{@value{AS}} source, refer to it by context, not
6289 The line numbers in our development sources will not match those in your
6290 sources. Your line numbers would convey no useful information to us.
6293 Here are some things that are not necessary:
6297 A description of the envelope of the bug.
6299 Often people who encounter a bug spend a lot of time investigating
6300 which changes to the input file will make the bug go away and which
6301 changes will not affect it.
6303 This is often time consuming and not very useful, because the way we
6304 will find the bug is by running a single example under the debugger
6305 with breakpoints, not by pure deduction from a series of examples.
6306 We recommend that you save your time for something else.
6308 Of course, if you can find a simpler example to report @emph{instead}
6309 of the original one, that is a convenience for us. Errors in the
6310 output will be easier to spot, running under the debugger will take
6311 less time, and so on.
6313 However, simplification is not vital; if you do not want to do this,
6314 report the bug anyway and send us the entire test case you used.
6317 A patch for the bug.
6319 A patch for the bug does help us if it is a good one. But do not omit
6320 the necessary information, such as the test case, on the assumption that
6321 a patch is all we need. We might see problems with your patch and decide
6322 to fix the problem another way, or we might not understand it at all.
6324 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
6325 construct an example that will make the program follow a certain path through
6326 the code. If you do not send us the example, we will not be able to construct
6327 one, so we will not be able to verify that the bug is fixed.
6329 And if we cannot understand what bug you are trying to fix, or why your
6330 patch should be an improvement, we will not install it. A test case will
6331 help us to understand.
6334 A guess about what the bug is or what it depends on.
6336 Such guesses are usually wrong. Even we cannot guess right about such
6337 things without first using the debugger to find the facts.
6340 @node Acknowledgements
6341 @chapter Acknowledgements
6343 If you have contributed to @command{@value{AS}} and your name isn't listed here,
6344 it is not meant as a slight. We just don't know about it. Send mail to the
6345 maintainer, and we'll correct the situation. Currently
6347 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
6349 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
6352 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
6353 information and the 68k series machines, most of the preprocessing pass, and
6354 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
6356 K. Richard Pixley maintained GAS for a while, adding various enhancements and
6357 many bug fixes, including merging support for several processors, breaking GAS
6358 up to handle multiple object file format back ends (including heavy rewrite,
6359 testing, an integration of the coff and b.out back ends), adding configuration
6360 including heavy testing and verification of cross assemblers and file splits
6361 and renaming, converted GAS to strictly ANSI C including full prototypes, added
6362 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
6363 port (including considerable amounts of reverse engineering), a SPARC opcode
6364 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
6365 assertions and made them work, much other reorganization, cleanup, and lint.
6367 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
6368 in format-specific I/O modules.
6370 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
6371 has done much work with it since.
6373 The Intel 80386 machine description was written by Eliot Dresselhaus.
6375 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
6377 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
6378 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
6380 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
6381 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
6382 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
6383 support a.out format.
6385 Support for the Zilog Z8k and Renesas H8/300 and H8/500 processors (tc-z8k,
6386 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
6387 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
6388 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
6391 John Gilmore built the AMD 29000 support, added @code{.include} support, and
6392 simplified the configuration of which versions accept which directives. He
6393 updated the 68k machine description so that Motorola's opcodes always produced
6394 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
6395 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
6396 cross-compilation support, and one bug in relaxation that took a week and
6397 required the proverbial one-bit fix.
6399 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
6400 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
6401 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
6402 PowerPC assembler, and made a few other minor patches.
6404 Steve Chamberlain made @command{@value{AS}} able to generate listings.
6406 Hewlett-Packard contributed support for the HP9000/300.
6408 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
6409 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
6410 formats). This work was supported by both the Center for Software Science at
6411 the University of Utah and Cygnus Support.
6413 Support for ELF format files has been worked on by Mark Eichin of Cygnus
6414 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
6415 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
6416 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
6417 and some initial 64-bit support).
6419 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
6421 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
6422 support for openVMS/Alpha.
6424 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
6427 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
6428 Inc. added support for Xtensa processors.
6430 Several engineers at Cygnus Support have also provided many small bug fixes and
6431 configuration enhancements.
6433 Many others have contributed large or small bugfixes and enhancements. If
6434 you have contributed significant work and are not mentioned on this list, and
6435 want to be, let us know. Some of the history has been lost; we are not
6436 intentionally leaving anyone out.
projects / dragonfly.git / blob