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}]
266 [@b{-meabi}=@var{ver}]
269 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
270 @b{-mapcs-reentrant}]
271 [@b{-mthumb-interwork}] [@b{-moabi}] [@b{-k}]
275 @emph{Target CRIS options:}
276 [@b{--underscore} | @b{--no-underscore}]
278 [@b{--emulation=criself} | @b{--emulation=crisaout}]
279 @c Deprecated -- deliberately not documented.
284 @emph{Target D10V options:}
289 @emph{Target D30V options:}
290 [@b{-O}|@b{-n}|@b{-N}]
293 @c Renesas family chips have no machine-dependent assembler options
296 @c HPPA has no machine-dependent assembler options (yet).
300 @emph{Target i386 options:}
301 [@b{--32}|@b{--64}] [@b{-n}]
305 @emph{Target i960 options:}
306 @c see md_parse_option in tc-i960.c
307 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
309 [@b{-b}] [@b{-no-relax}]
313 @emph{Target IA-64 options:}
314 [@b{-mconstant-gp}|@b{-mauto-pic}]
315 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
317 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
321 @emph{Target IP2K options:}
322 [@b{-mip2022}|@b{-mip2022ext}]
326 @emph{Target M32R options:}
327 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
332 @emph{Target M680X0 options:}
333 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
337 @emph{Target M68HC11 options:}
338 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
339 [@b{-mshort}|@b{-mlong}]
340 [@b{-mshort-double}|@b{-mlong-double}]
341 [@b{--force-long-branchs}] [@b{--short-branchs}]
342 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
343 [@b{--print-opcodes}] [@b{--generate-example}]
347 @emph{Target MCORE options:}
348 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
349 [@b{-mcpu=[210|340]}]
353 @emph{Target MIPS options:}
354 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
355 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
356 [@b{-non_shared}] [@b{-xgot}]
357 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
358 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
359 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
360 [@b{-mips64}] [@b{-mips64r2}]
361 [@b{-construct-floats}] [@b{-no-construct-floats}]
362 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
363 [@b{-mfix7000}] [@b{-mno-fix7000}]
364 [@b{-mips16}] [@b{-no-mips16}]
365 [@b{-mips3d}] [@b{-no-mips3d}]
366 [@b{-mdmx}] [@b{-no-mdmx}]
367 [@b{-mdebug}] [@b{-no-mdebug}]
368 [@b{-mpdr}] [@b{-mno-pdr}]
372 @emph{Target MMIX options:}
373 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
374 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
375 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
376 [@b{--linker-allocated-gregs}]
380 @emph{Target PDP11 options:}
381 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
382 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
383 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
387 @emph{Target picoJava options:}
392 @emph{Target PowerPC options:}
393 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
394 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}|
395 @b{-mbooke32}|@b{-mbooke64}]
396 [@b{-mcom}|@b{-many}|@b{-maltivec}] [@b{-memb}]
397 [@b{-mregnames}|@b{-mno-regnames}]
398 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
399 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
400 [@b{-msolaris}|@b{-mno-solaris}]
404 @emph{Target SPARC options:}
405 @c The order here is important. See c-sparc.texi.
406 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
407 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
408 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
413 @emph{Target TIC54X options:}
414 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
415 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
418 @c Z8000 has no machine-dependent assembler options
422 @emph{Target Xtensa options:}
423 [@b{--[no-]density}] [@b{--[no-]relax}] [@b{--[no-]generics}]
424 [@b{--[no-]text-section-literals}]
425 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
434 Turn on listings, in any of a variety of ways:
438 omit false conditionals
441 omit debugging directives
444 include high-level source
450 include macro expansions
453 omit forms processing
459 set the name of the listing file
462 You may combine these options; for example, use @samp{-aln} for assembly
463 listing without forms processing. The @samp{=file} option, if used, must be
464 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
467 Ignored. This option is accepted for script compatibility with calls to
470 @item --defsym @var{sym}=@var{value}
471 Define the symbol @var{sym} to be @var{value} before assembling the input file.
472 @var{value} must be an integer constant. As in C, a leading @samp{0x}
473 indicates a hexadecimal value, and a leading @samp{0} indicates an octal value.
476 ``fast''---skip whitespace and comment preprocessing (assume source is
480 Generate stabs debugging information for each assembler line. This
481 may help debugging assembler code, if the debugger can handle it.
484 Generate stabs debugging information for each assembler line, with GNU
485 extensions that probably only gdb can handle, and that could make other
486 debuggers crash or refuse to read your program. This
487 may help debugging assembler code. Currently the only GNU extension is
488 the location of the current working directory at assembling time.
491 Generate DWARF2 debugging information for each assembler line. This
492 may help debugging assembler code, if the debugger can handle it. Note---this
493 option is only supported by some targets, not all of them.
496 Print a summary of the command line options and exit.
499 Print a summary of all target specific options and exit.
502 Add directory @var{dir} to the search list for @code{.include} directives.
505 Don't warn about signed overflow.
508 @ifclear DIFF-TBL-KLUGE
509 This option is accepted but has no effect on the @value{TARGET} family.
511 @ifset DIFF-TBL-KLUGE
512 Issue warnings when difference tables altered for long displacements.
517 Keep (in the symbol table) local symbols. On traditional a.out systems
518 these start with @samp{L}, but different systems have different local
521 @item --listing-lhs-width=@var{number}
522 Set the maximum width, in words, of the output data column for an assembler
523 listing to @var{number}.
525 @item --listing-lhs-width2=@var{number}
526 Set the maximum width, in words, of the output data column for continuation
527 lines in an assembler listing to @var{number}.
529 @item --listing-rhs-width=@var{number}
530 Set the maximum width of an input source line, as displayed in a listing, to
533 @item --listing-cont-lines=@var{number}
534 Set the maximum number of lines printed in a listing for a single line of input
537 @item -o @var{objfile}
538 Name the object-file output from @command{@value{AS}} @var{objfile}.
541 Fold the data section into the text section.
544 Print the maximum space (in bytes) and total time (in seconds) used by
547 @item --strip-local-absolute
548 Remove local absolute symbols from the outgoing symbol table.
552 Print the @command{as} version.
555 Print the @command{as} version and exit.
559 Suppress warning messages.
561 @item --fatal-warnings
562 Treat warnings as errors.
565 Don't suppress warning messages or treat them as errors.
574 Generate an object file even after errors.
576 @item -- | @var{files} @dots{}
577 Standard input, or source files to assemble.
582 The following options are available when @value{AS} is configured for
587 This option selects the core processor variant.
589 Select either big-endian (-EB) or little-endian (-EL) output.
594 The following options are available when @value{AS} is configured for the ARM
598 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
599 Specify which ARM processor variant is the target.
600 @item -march=@var{architecture}[+@var{extension}@dots{}]
601 Specify which ARM architecture variant is used by the target.
602 @item -mfpu=@var{floating-point-format}
603 Select which Floating Point architecture is the target.
604 @item -mfloat-abi=@var{abi}
605 Select which floating point ABI is in use.
607 Enable Thumb only instruction decoding.
608 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant | -moabi
609 Select which procedure calling convention is in use.
611 Select either big-endian (-EB) or little-endian (-EL) output.
612 @item -mthumb-interwork
613 Specify that the code has been generated with interworking between Thumb and
616 Specify that PIC code has been generated.
621 See the info pages for documentation of the CRIS-specific options.
625 The following options are available when @value{AS} is configured for
628 @cindex D10V optimization
629 @cindex optimization, D10V
631 Optimize output by parallelizing instructions.
636 The following options are available when @value{AS} is configured for a D30V
639 @cindex D30V optimization
640 @cindex optimization, D30V
642 Optimize output by parallelizing instructions.
646 Warn when nops are generated.
648 @cindex D30V nops after 32-bit multiply
650 Warn when a nop after a 32-bit multiply instruction is generated.
655 The following options are available when @value{AS} is configured for the
656 Intel 80960 processor.
659 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
660 Specify which variant of the 960 architecture is the target.
663 Add code to collect statistics about branches taken.
666 Do not alter compare-and-branch instructions for long displacements;
673 The following options are available when @value{AS} is configured for the
679 Specifies that the extended IP2022 instructions are allowed.
682 Restores the default behaviour, which restricts the permitted instructions to
683 just the basic IP2022 ones.
689 The following options are available when @value{AS} is configured for the
690 Renesas M32R (formerly Mitsubishi M32R) series.
695 Specify which processor in the M32R family is the target. The default
696 is normally the M32R, but this option changes it to the M32RX.
698 @item --warn-explicit-parallel-conflicts or --Wp
699 Produce warning messages when questionable parallel constructs are
702 @item --no-warn-explicit-parallel-conflicts or --Wnp
703 Do not produce warning messages when questionable parallel constructs are
710 The following options are available when @value{AS} is configured for the
711 Motorola 68000 series.
716 Shorten references to undefined symbols, to one word instead of two.
718 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
719 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
720 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
721 Specify what processor in the 68000 family is the target. The default
722 is normally the 68020, but this can be changed at configuration time.
724 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
725 The target machine does (or does not) have a floating-point coprocessor.
726 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
727 the basic 68000 is not compatible with the 68881, a combination of the
728 two can be specified, since it's possible to do emulation of the
729 coprocessor instructions with the main processor.
731 @item -m68851 | -mno-68851
732 The target machine does (or does not) have a memory-management
733 unit coprocessor. The default is to assume an MMU for 68020 and up.
740 For details about the PDP-11 machine dependent features options,
741 see @ref{PDP-11-Options}.
744 @item -mpic | -mno-pic
745 Generate position-independent (or position-dependent) code. The
746 default is @option{-mpic}.
749 @itemx -mall-extensions
750 Enable all instruction set extensions. This is the default.
752 @item -mno-extensions
753 Disable all instruction set extensions.
755 @item -m@var{extension} | -mno-@var{extension}
756 Enable (or disable) a particular instruction set extension.
759 Enable the instruction set extensions supported by a particular CPU, and
760 disable all other extensions.
762 @item -m@var{machine}
763 Enable the instruction set extensions supported by a particular machine
764 model, and disable all other extensions.
770 The following options are available when @value{AS} is configured for
771 a picoJava processor.
775 @cindex PJ endianness
776 @cindex endianness, PJ
777 @cindex big endian output, PJ
779 Generate ``big endian'' format output.
781 @cindex little endian output, PJ
783 Generate ``little endian'' format output.
789 The following options are available when @value{AS} is configured for the
790 Motorola 68HC11 or 68HC12 series.
794 @item -m68hc11 | -m68hc12 | -m68hcs12
795 Specify what processor is the target. The default is
796 defined by the configuration option when building the assembler.
799 Specify to use the 16-bit integer ABI.
802 Specify to use the 32-bit integer ABI.
805 Specify to use the 32-bit double ABI.
808 Specify to use the 64-bit double ABI.
810 @item --force-long-branchs
811 Relative branches are turned into absolute ones. This concerns
812 conditional branches, unconditional branches and branches to a
815 @item -S | --short-branchs
816 Do not turn relative branchs into absolute ones
817 when the offset is out of range.
819 @item --strict-direct-mode
820 Do not turn the direct addressing mode into extended addressing mode
821 when the instruction does not support direct addressing mode.
823 @item --print-insn-syntax
824 Print the syntax of instruction in case of error.
826 @item --print-opcodes
827 print the list of instructions with syntax and then exit.
829 @item --generate-example
830 print an example of instruction for each possible instruction and then exit.
831 This option is only useful for testing @command{@value{AS}}.
837 The following options are available when @command{@value{AS}} is configured
838 for the SPARC architecture:
841 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
842 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
843 Explicitly select a variant of the SPARC architecture.
845 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
846 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
848 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
849 UltraSPARC extensions.
851 @item -xarch=v8plus | -xarch=v8plusa
852 For compatibility with the Solaris v9 assembler. These options are
853 equivalent to -Av8plus and -Av8plusa, respectively.
856 Warn when the assembler switches to another architecture.
861 The following options are available when @value{AS} is configured for the 'c54x
866 Enable extended addressing mode. All addresses and relocations will assume
867 extended addressing (usually 23 bits).
868 @item -mcpu=@var{CPU_VERSION}
869 Sets the CPU version being compiled for.
870 @item -merrors-to-file @var{FILENAME}
871 Redirect error output to a file, for broken systems which don't support such
872 behaviour in the shell.
877 The following options are available when @value{AS} is configured for
878 a @sc{mips} processor.
882 This option sets the largest size of an object that can be referenced
883 implicitly with the @code{gp} register. It is only accepted for targets that
884 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
886 @cindex MIPS endianness
887 @cindex endianness, MIPS
888 @cindex big endian output, MIPS
890 Generate ``big endian'' format output.
892 @cindex little endian output, MIPS
894 Generate ``little endian'' format output.
906 Generate code for a particular @sc{mips} Instruction Set Architecture level.
907 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
908 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
909 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
910 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
912 correspond to generic
913 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
914 and @samp{MIPS64 Release 2}
915 ISA processors, respectively.
917 @item -march=@var{CPU}
918 Generate code for a particular @sc{mips} cpu.
920 @item -mtune=@var{cpu}
921 Schedule and tune for a particular @sc{mips} cpu.
925 Cause nops to be inserted if the read of the destination register
926 of an mfhi or mflo instruction occurs in the following two instructions.
930 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
931 section instead of the standard ELF .stabs sections.
935 Control generation of @code{.pdr} sections.
939 The register sizes are normally inferred from the ISA and ABI, but these
940 flags force a certain group of registers to be treated as 32 bits wide at
941 all times. @samp{-mgp32} controls the size of general-purpose registers
942 and @samp{-mfp32} controls the size of floating-point registers.
946 Generate code for the MIPS 16 processor. This is equivalent to putting
947 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
948 turns off this option.
952 Generate code for the MIPS-3D Application Specific Extension.
953 This tells the assembler to accept MIPS-3D instructions.
954 @samp{-no-mips3d} turns off this option.
958 Generate code for the MDMX Application Specific Extension.
959 This tells the assembler to accept MDMX instructions.
960 @samp{-no-mdmx} turns off this option.
962 @item --construct-floats
963 @itemx --no-construct-floats
964 The @samp{--no-construct-floats} option disables the construction of
965 double width floating point constants by loading the two halves of the
966 value into the two single width floating point registers that make up
967 the double width register. By default @samp{--construct-floats} is
968 selected, allowing construction of these floating point constants.
971 @item --emulation=@var{name}
972 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
973 for some other target, in all respects, including output format (choosing
974 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
975 debugging information or store symbol table information, and default
976 endianness. The available configuration names are: @samp{mipsecoff},
977 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
978 @samp{mipsbelf}. The first two do not alter the default endianness from that
979 of the primary target for which the assembler was configured; the others change
980 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
981 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
982 selection in any case.
984 This option is currently supported only when the primary target
985 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
986 Furthermore, the primary target or others specified with
987 @samp{--enable-targets=@dots{}} at configuration time must include support for
988 the other format, if both are to be available. For example, the Irix 5
989 configuration includes support for both.
991 Eventually, this option will support more configurations, with more
992 fine-grained control over the assembler's behavior, and will be supported for
996 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1003 Control how to deal with multiplication overflow and division by zero.
1004 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1005 (and only work for Instruction Set Architecture level 2 and higher);
1006 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1010 When this option is used, @command{@value{AS}} will issue a warning every
1011 time it generates a nop instruction from a macro.
1016 The following options are available when @value{AS} is configured for
1022 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1023 The command line option @samp{-nojsri2bsr} can be used to disable it.
1027 Enable or disable the silicon filter behaviour. By default this is disabled.
1028 The default can be overridden by the @samp{-sifilter} command line option.
1031 Alter jump instructions for long displacements.
1033 @item -mcpu=[210|340]
1034 Select the cpu type on the target hardware. This controls which instructions
1038 Assemble for a big endian target.
1041 Assemble for a little endian target.
1047 See the info pages for documentation of the MMIX-specific options.
1051 The following options are available when @value{AS} is configured for
1052 an Xtensa processor.
1055 @item --density | --no-density
1056 Enable or disable use of instructions from the Xtensa code density
1057 option. This is enabled by default when the Xtensa processor supports
1058 the code density option.
1060 @item --relax | --no-relax
1061 Enable or disable instruction relaxation. This is enabled by default.
1062 Note: In the current implementation, these options also control whether
1063 assembler optimizations are performed, making these options equivalent
1064 to @option{--generics} and @option{--no-generics}.
1066 @item --generics | --no-generics
1067 Enable or disable all assembler transformations of Xtensa instructions.
1068 The default is @option{--generics};
1069 @option{--no-generics} should be used only in the rare cases when the
1070 instructions must be exactly as specified in the assembly source.
1072 @item --text-section-literals | --no-text-section-literals
1073 With @option{--text-@-section-@-literals}, literal pools are interspersed
1074 in the text section. The default is
1075 @option{--no-@-text-@-section-@-literals}, which places literals in a
1076 separate section in the output file.
1078 @item --target-align | --no-target-align
1079 Enable or disable automatic alignment to reduce branch penalties at the
1080 expense of some code density. The default is @option{--target-@-align}.
1082 @item --longcalls | --no-longcalls
1083 Enable or disable transformation of call instructions to allow calls
1084 across a greater range of addresses. The default is
1085 @option{--no-@-longcalls}.
1092 * Manual:: Structure of this Manual
1093 * GNU Assembler:: The GNU Assembler
1094 * Object Formats:: Object File Formats
1095 * Command Line:: Command Line
1096 * Input Files:: Input Files
1097 * Object:: Output (Object) File
1098 * Errors:: Error and Warning Messages
1102 @section Structure of this Manual
1104 @cindex manual, structure and purpose
1105 This manual is intended to describe what you need to know to use
1106 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1107 notation for symbols, constants, and expressions; the directives that
1108 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1111 We also cover special features in the @value{TARGET}
1112 configuration of @command{@value{AS}}, including assembler directives.
1115 This manual also describes some of the machine-dependent features of
1116 various flavors of the assembler.
1119 @cindex machine instructions (not covered)
1120 On the other hand, this manual is @emph{not} intended as an introduction
1121 to programming in assembly language---let alone programming in general!
1122 In a similar vein, we make no attempt to introduce the machine
1123 architecture; we do @emph{not} describe the instruction set, standard
1124 mnemonics, registers or addressing modes that are standard to a
1125 particular architecture.
1127 You may want to consult the manufacturer's
1128 machine architecture manual for this information.
1132 For information on the H8/300 machine instruction set, see @cite{H8/300
1133 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1134 Programming Manual} (Renesas).
1137 For information on the H8/500 machine instruction set, see @cite{H8/500
1138 Series Programming Manual} (Renesas M21T001).
1141 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1142 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1143 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1144 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1147 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1151 @c I think this is premature---doc@cygnus.com, 17jan1991
1153 Throughout this manual, we assume that you are running @dfn{GNU},
1154 the portable operating system from the @dfn{Free Software
1155 Foundation, Inc.}. This restricts our attention to certain kinds of
1156 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1157 once this assumption is granted examples and definitions need less
1160 @command{@value{AS}} is part of a team of programs that turn a high-level
1161 human-readable series of instructions into a low-level
1162 computer-readable series of instructions. Different versions of
1163 @command{@value{AS}} are used for different kinds of computer.
1166 @c There used to be a section "Terminology" here, which defined
1167 @c "contents", "byte", "word", and "long". Defining "word" to any
1168 @c particular size is confusing when the .word directive may generate 16
1169 @c bits on one machine and 32 bits on another; in general, for the user
1170 @c version of this manual, none of these terms seem essential to define.
1171 @c They were used very little even in the former draft of the manual;
1172 @c this draft makes an effort to avoid them (except in names of
1176 @section The GNU Assembler
1178 @c man begin DESCRIPTION
1180 @sc{gnu} @command{as} is really a family of assemblers.
1182 This manual describes @command{@value{AS}}, a member of that family which is
1183 configured for the @value{TARGET} architectures.
1185 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1186 should find a fairly similar environment when you use it on another
1187 architecture. Each version has much in common with the others,
1188 including object file formats, most assembler directives (often called
1189 @dfn{pseudo-ops}) and assembler syntax.@refill
1191 @cindex purpose of @sc{gnu} assembler
1192 @command{@value{AS}} is primarily intended to assemble the output of the
1193 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1194 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1195 assemble correctly everything that other assemblers for the same
1196 machine would assemble.
1198 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1201 @c This remark should appear in generic version of manual; assumption
1202 @c here is that generic version sets M680x0.
1203 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1204 assembler for the same architecture; for example, we know of several
1205 incompatible versions of 680x0 assembly language syntax.
1210 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1211 program in one pass of the source file. This has a subtle impact on the
1212 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1214 @node Object Formats
1215 @section Object File Formats
1217 @cindex object file format
1218 The @sc{gnu} assembler can be configured to produce several alternative
1219 object file formats. For the most part, this does not affect how you
1220 write assembly language programs; but directives for debugging symbols
1221 are typically different in different file formats. @xref{Symbol
1222 Attributes,,Symbol Attributes}.
1225 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1226 @value{OBJ-NAME} format object files.
1228 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1230 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1231 @code{a.out} or COFF format object files.
1234 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1235 @code{b.out} or COFF format object files.
1238 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1239 SOM or ELF format object files.
1244 @section Command Line
1246 @cindex command line conventions
1248 After the program name @command{@value{AS}}, the command line may contain
1249 options and file names. Options may appear in any order, and may be
1250 before, after, or between file names. The order of file names is
1253 @cindex standard input, as input file
1255 @file{--} (two hyphens) by itself names the standard input file
1256 explicitly, as one of the files for @command{@value{AS}} to assemble.
1258 @cindex options, command line
1259 Except for @samp{--} any command line argument that begins with a
1260 hyphen (@samp{-}) is an option. Each option changes the behavior of
1261 @command{@value{AS}}. No option changes the way another option works. An
1262 option is a @samp{-} followed by one or more letters; the case of
1263 the letter is important. All options are optional.
1265 Some options expect exactly one file name to follow them. The file
1266 name may either immediately follow the option's letter (compatible
1267 with older assemblers) or it may be the next command argument (@sc{gnu}
1268 standard). These two command lines are equivalent:
1271 @value{AS} -o my-object-file.o mumble.s
1272 @value{AS} -omy-object-file.o mumble.s
1276 @section Input Files
1279 @cindex source program
1280 @cindex files, input
1281 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1282 describe the program input to one run of @command{@value{AS}}. The program may
1283 be in one or more files; how the source is partitioned into files
1284 doesn't change the meaning of the source.
1286 @c I added "con" prefix to "catenation" just to prove I can overcome my
1287 @c APL training... doc@cygnus.com
1288 The source program is a concatenation of the text in all the files, in the
1291 @c man begin DESCRIPTION
1292 Each time you run @command{@value{AS}} it assembles exactly one source
1293 program. The source program is made up of one or more files.
1294 (The standard input is also a file.)
1296 You give @command{@value{AS}} a command line that has zero or more input file
1297 names. The input files are read (from left file name to right). A
1298 command line argument (in any position) that has no special meaning
1299 is taken to be an input file name.
1301 If you give @command{@value{AS}} no file names it attempts to read one input file
1302 from the @command{@value{AS}} standard input, which is normally your terminal. You
1303 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1306 Use @samp{--} if you need to explicitly name the standard input file
1307 in your command line.
1309 If the source is empty, @command{@value{AS}} produces a small, empty object
1314 @subheading Filenames and Line-numbers
1316 @cindex input file linenumbers
1317 @cindex line numbers, in input files
1318 There are two ways of locating a line in the input file (or files) and
1319 either may be used in reporting error messages. One way refers to a line
1320 number in a physical file; the other refers to a line number in a
1321 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1323 @dfn{Physical files} are those files named in the command line given
1324 to @command{@value{AS}}.
1326 @dfn{Logical files} are simply names declared explicitly by assembler
1327 directives; they bear no relation to physical files. Logical file names help
1328 error messages reflect the original source file, when @command{@value{AS}} source
1329 is itself synthesized from other files. @command{@value{AS}} understands the
1330 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1331 @ref{File,,@code{.file}}.
1334 @section Output (Object) File
1340 Every time you run @command{@value{AS}} it produces an output file, which is
1341 your assembly language program translated into numbers. This file
1342 is the object file. Its default name is
1350 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1352 You can give it another name by using the @option{-o} option. Conventionally,
1353 object file names end with @file{.o}. The default name is used for historical
1354 reasons: older assemblers were capable of assembling self-contained programs
1355 directly into a runnable program. (For some formats, this isn't currently
1356 possible, but it can be done for the @code{a.out} format.)
1360 The object file is meant for input to the linker @code{@value{LD}}. It contains
1361 assembled program code, information to help @code{@value{LD}} integrate
1362 the assembled program into a runnable file, and (optionally) symbolic
1363 information for the debugger.
1365 @c link above to some info file(s) like the description of a.out.
1366 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1369 @section Error and Warning Messages
1371 @c man begin DESCRIPTION
1373 @cindex error messages
1374 @cindex warning messages
1375 @cindex messages from assembler
1376 @command{@value{AS}} may write warnings and error messages to the standard error
1377 file (usually your terminal). This should not happen when a compiler
1378 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1379 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1380 grave problem that stops the assembly.
1384 @cindex format of warning messages
1385 Warning messages have the format
1388 file_name:@b{NNN}:Warning Message Text
1392 @cindex line numbers, in warnings/errors
1393 (where @b{NNN} is a line number). If a logical file name has been given
1394 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1395 the current input file is used. If a logical line number was given
1397 (@pxref{Line,,@code{.line}})
1401 (@pxref{Line,,@code{.line}})
1404 (@pxref{Ln,,@code{.ln}})
1407 then it is used to calculate the number printed,
1408 otherwise the actual line in the current source file is printed. The
1409 message text is intended to be self explanatory (in the grand Unix
1412 @cindex format of error messages
1413 Error messages have the format
1415 file_name:@b{NNN}:FATAL:Error Message Text
1417 The file name and line number are derived as for warning
1418 messages. The actual message text may be rather less explanatory
1419 because many of them aren't supposed to happen.
1422 @chapter Command-Line Options
1424 @cindex options, all versions of assembler
1425 This chapter describes command-line options available in @emph{all}
1426 versions of the @sc{gnu} assembler; @pxref{Machine Dependencies}, for options specific
1428 to the @value{TARGET} target.
1431 to particular machine architectures.
1434 @c man begin DESCRIPTION
1436 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1437 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1438 The assembler arguments must be separated from each other (and the @samp{-Wa})
1439 by commas. For example:
1442 gcc -c -g -O -Wa,-alh,-L file.c
1446 This passes two options to the assembler: @samp{-alh} (emit a listing to
1447 standard output with high-level and assembly source) and @samp{-L} (retain
1448 local symbols in the symbol table).
1450 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1451 command-line options are automatically passed to the assembler by the compiler.
1452 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1453 precisely what options it passes to each compilation pass, including the
1459 * a:: -a[cdhlns] enable listings
1460 * D:: -D for compatibility
1461 * f:: -f to work faster
1462 * I:: -I for .include search path
1463 @ifclear DIFF-TBL-KLUGE
1464 * K:: -K for compatibility
1466 @ifset DIFF-TBL-KLUGE
1467 * K:: -K for difference tables
1470 * L:: -L to retain local labels
1471 * listing:: --listing-XXX to configure listing output
1472 * M:: -M or --mri to assemble in MRI compatibility mode
1473 * MD:: --MD for dependency tracking
1474 * o:: -o to name the object file
1475 * R:: -R to join data and text sections
1476 * statistics:: --statistics to see statistics about assembly
1477 * traditional-format:: --traditional-format for compatible output
1478 * v:: -v to announce version
1479 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1480 * Z:: -Z to make object file even after errors
1484 @section Enable Listings: @option{-a[cdhlns]}
1493 @cindex listings, enabling
1494 @cindex assembly listings, enabling
1496 These options enable listing output from the assembler. By itself,
1497 @samp{-a} requests high-level, assembly, and symbols listing.
1498 You can use other letters to select specific options for the list:
1499 @samp{-ah} requests a high-level language listing,
1500 @samp{-al} requests an output-program assembly listing, and
1501 @samp{-as} requests a symbol table listing.
1502 High-level listings require that a compiler debugging option like
1503 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1506 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1507 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1508 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1509 omitted from the listing.
1511 Use the @samp{-ad} option to omit debugging directives from the
1514 Once you have specified one of these options, you can further control
1515 listing output and its appearance using the directives @code{.list},
1516 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1518 The @samp{-an} option turns off all forms processing.
1519 If you do not request listing output with one of the @samp{-a} options, the
1520 listing-control directives have no effect.
1522 The letters after @samp{-a} may be combined into one option,
1523 @emph{e.g.}, @samp{-aln}.
1525 Note if the assembler source is coming from the standard input (eg because it
1526 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1527 is being used) then the listing will not contain any comments or preprocessor
1528 directives. This is because the listing code buffers input source lines from
1529 stdin only after they have been preprocessed by the assembler. This reduces
1530 memory usage and makes the code more efficient.
1533 @section @option{-D}
1536 This option has no effect whatsoever, but it is accepted to make it more
1537 likely that scripts written for other assemblers also work with
1538 @command{@value{AS}}.
1541 @section Work Faster: @option{-f}
1544 @cindex trusted compiler
1545 @cindex faster processing (@option{-f})
1546 @samp{-f} should only be used when assembling programs written by a
1547 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1548 and comment preprocessing on
1549 the input file(s) before assembling them. @xref{Preprocessing,
1553 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1554 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1559 @section @code{.include} Search Path: @option{-I} @var{path}
1561 @kindex -I @var{path}
1562 @cindex paths for @code{.include}
1563 @cindex search path for @code{.include}
1564 @cindex @code{include} directive search path
1565 Use this option to add a @var{path} to the list of directories
1566 @command{@value{AS}} searches for files specified in @code{.include}
1567 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1568 many times as necessary to include a variety of paths. The current
1569 working directory is always searched first; after that, @command{@value{AS}}
1570 searches any @samp{-I} directories in the same order as they were
1571 specified (left to right) on the command line.
1574 @section Difference Tables: @option{-K}
1577 @ifclear DIFF-TBL-KLUGE
1578 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1579 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1580 where it can be used to warn when the assembler alters the machine code
1581 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1582 family does not have the addressing limitations that sometimes lead to this
1583 alteration on other platforms.
1586 @ifset DIFF-TBL-KLUGE
1587 @cindex difference tables, warning
1588 @cindex warning for altered difference tables
1589 @command{@value{AS}} sometimes alters the code emitted for directives of the form
1590 @samp{.word @var{sym1}-@var{sym2}}; @pxref{Word,,@code{.word}}.
1591 You can use the @samp{-K} option if you want a warning issued when this
1596 @section Include Local Labels: @option{-L}
1599 @cindex local labels, retaining in output
1600 Labels beginning with @samp{L} (upper case only) are called @dfn{local
1601 labels}. @xref{Symbol Names}. Normally you do not see such labels when
1602 debugging, because they are intended for the use of programs (like
1603 compilers) that compose assembler programs, not for your notice.
1604 Normally both @command{@value{AS}} and @code{@value{LD}} discard such labels, so you do not
1605 normally debug with them.
1607 This option tells @command{@value{AS}} to retain those @samp{L@dots{}} symbols
1608 in the object file. Usually if you do this you also tell the linker
1609 @code{@value{LD}} to preserve symbols whose names begin with @samp{L}.
1611 By default, a local label is any label beginning with @samp{L}, but each
1612 target is allowed to redefine the local label prefix.
1614 On the HPPA local labels begin with @samp{L$}.
1618 @section Configuring listing output: @option{--listing}
1620 The listing feature of the assembler can be enabled via the command line switch
1621 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1622 hex dump of the corresponding locations in the output object file, and displays
1623 them as a listing file. The format of this listing can be controlled by pseudo
1624 ops inside the assembler source (@pxref{List} @pxref{Title} @pxref{Sbttl}
1625 @pxref{Psize} @pxref{Eject}) and also by the following switches:
1628 @item --listing-lhs-width=@samp{number}
1629 @kindex --listing-lhs-width
1630 @cindex Width of first line disassembly output
1631 Sets the maximum width, in words, of the first line of the hex byte dump. This
1632 dump appears on the left hand side of the listing output.
1634 @item --listing-lhs-width2=@samp{number}
1635 @kindex --listing-lhs-width2
1636 @cindex Width of continuation lines of disassembly output
1637 Sets the maximum width, in words, of any further lines of the hex byte dump for
1638 a given input source line. If this value is not specified, it defaults to being
1639 the same as the value specified for @samp{--listing-lhs-width}. If neither
1640 switch is used the default is to one.
1642 @item --listing-rhs-width=@samp{number}
1643 @kindex --listing-rhs-width
1644 @cindex Width of source line output
1645 Sets the maximum width, in characters, of the source line that is displayed
1646 alongside the hex dump. The default value for this parameter is 100. The
1647 source line is displayed on the right hand side of the listing output.
1649 @item --listing-cont-lines=@samp{number}
1650 @kindex --listing-cont-lines
1651 @cindex Maximum number of continuation lines
1652 Sets the maximum number of continuation lines of hex dump that will be
1653 displayed for a given single line of source input. The default value is 4.
1657 @section Assemble in MRI Compatibility Mode: @option{-M}
1660 @cindex MRI compatibility mode
1661 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1662 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1663 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1664 configured target) assembler from Microtec Research. The exact nature of the
1665 MRI syntax will not be documented here; see the MRI manuals for more
1666 information. Note in particular that the handling of macros and macro
1667 arguments is somewhat different. The purpose of this option is to permit
1668 assembling existing MRI assembler code using @command{@value{AS}}.
1670 The MRI compatibility is not complete. Certain operations of the MRI assembler
1671 depend upon its object file format, and can not be supported using other object
1672 file formats. Supporting these would require enhancing each object file format
1673 individually. These are:
1676 @item global symbols in common section
1678 The m68k MRI assembler supports common sections which are merged by the linker.
1679 Other object file formats do not support this. @command{@value{AS}} handles
1680 common sections by treating them as a single common symbol. It permits local
1681 symbols to be defined within a common section, but it can not support global
1682 symbols, since it has no way to describe them.
1684 @item complex relocations
1686 The MRI assemblers support relocations against a negated section address, and
1687 relocations which combine the start addresses of two or more sections. These
1688 are not support by other object file formats.
1690 @item @code{END} pseudo-op specifying start address
1692 The MRI @code{END} pseudo-op permits the specification of a start address.
1693 This is not supported by other object file formats. The start address may
1694 instead be specified using the @option{-e} option to the linker, or in a linker
1697 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1699 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1700 name to the output file. This is not supported by other object file formats.
1702 @item @code{ORG} pseudo-op
1704 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1705 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1706 which changes the location within the current section. Absolute sections are
1707 not supported by other object file formats. The address of a section may be
1708 assigned within a linker script.
1711 There are some other features of the MRI assembler which are not supported by
1712 @command{@value{AS}}, typically either because they are difficult or because they
1713 seem of little consequence. Some of these may be supported in future releases.
1717 @item EBCDIC strings
1719 EBCDIC strings are not supported.
1721 @item packed binary coded decimal
1723 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1724 and @code{DCB.P} pseudo-ops are not supported.
1726 @item @code{FEQU} pseudo-op
1728 The m68k @code{FEQU} pseudo-op is not supported.
1730 @item @code{NOOBJ} pseudo-op
1732 The m68k @code{NOOBJ} pseudo-op is not supported.
1734 @item @code{OPT} branch control options
1736 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1737 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1738 relaxes all branches, whether forward or backward, to an appropriate size, so
1739 these options serve no purpose.
1741 @item @code{OPT} list control options
1743 The following m68k @code{OPT} list control options are ignored: @code{C},
1744 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1745 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1747 @item other @code{OPT} options
1749 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1750 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1752 @item @code{OPT} @code{D} option is default
1754 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1755 @code{OPT NOD} may be used to turn it off.
1757 @item @code{XREF} pseudo-op.
1759 The m68k @code{XREF} pseudo-op is ignored.
1761 @item @code{.debug} pseudo-op
1763 The i960 @code{.debug} pseudo-op is not supported.
1765 @item @code{.extended} pseudo-op
1767 The i960 @code{.extended} pseudo-op is not supported.
1769 @item @code{.list} pseudo-op.
1771 The various options of the i960 @code{.list} pseudo-op are not supported.
1773 @item @code{.optimize} pseudo-op
1775 The i960 @code{.optimize} pseudo-op is not supported.
1777 @item @code{.output} pseudo-op
1779 The i960 @code{.output} pseudo-op is not supported.
1781 @item @code{.setreal} pseudo-op
1783 The i960 @code{.setreal} pseudo-op is not supported.
1788 @section Dependency Tracking: @option{--MD}
1791 @cindex dependency tracking
1794 @command{@value{AS}} can generate a dependency file for the file it creates. This
1795 file consists of a single rule suitable for @code{make} describing the
1796 dependencies of the main source file.
1798 The rule is written to the file named in its argument.
1800 This feature is used in the automatic updating of makefiles.
1803 @section Name the Object File: @option{-o}
1806 @cindex naming object file
1807 @cindex object file name
1808 There is always one object file output when you run @command{@value{AS}}. By
1809 default it has the name
1812 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1826 You use this option (which takes exactly one filename) to give the
1827 object file a different name.
1829 Whatever the object file is called, @command{@value{AS}} overwrites any
1830 existing file of the same name.
1833 @section Join Data and Text Sections: @option{-R}
1836 @cindex data and text sections, joining
1837 @cindex text and data sections, joining
1838 @cindex joining text and data sections
1839 @cindex merging text and data sections
1840 @option{-R} tells @command{@value{AS}} to write the object file as if all
1841 data-section data lives in the text section. This is only done at
1842 the very last moment: your binary data are the same, but data
1843 section parts are relocated differently. The data section part of
1844 your object file is zero bytes long because all its bytes are
1845 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1847 When you specify @option{-R} it would be possible to generate shorter
1848 address displacements (because we do not have to cross between text and
1849 data section). We refrain from doing this simply for compatibility with
1850 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
1853 When @command{@value{AS}} is configured for COFF or ELF output,
1854 this option is only useful if you use sections named @samp{.text} and
1859 @option{-R} is not supported for any of the HPPA targets. Using
1860 @option{-R} generates a warning from @command{@value{AS}}.
1864 @section Display Assembly Statistics: @option{--statistics}
1866 @kindex --statistics
1867 @cindex statistics, about assembly
1868 @cindex time, total for assembly
1869 @cindex space used, maximum for assembly
1870 Use @samp{--statistics} to display two statistics about the resources used by
1871 @command{@value{AS}}: the maximum amount of space allocated during the assembly
1872 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
1875 @node traditional-format
1876 @section Compatible Output: @option{--traditional-format}
1878 @kindex --traditional-format
1879 For some targets, the output of @command{@value{AS}} is different in some ways
1880 from the output of some existing assembler. This switch requests
1881 @command{@value{AS}} to use the traditional format instead.
1883 For example, it disables the exception frame optimizations which
1884 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
1887 @section Announce Version: @option{-v}
1891 @cindex assembler version
1892 @cindex version of assembler
1893 You can find out what version of as is running by including the
1894 option @samp{-v} (which you can also spell as @samp{-version}) on the
1898 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
1900 @command{@value{AS}} should never give a warning or error message when
1901 assembling compiler output. But programs written by people often
1902 cause @command{@value{AS}} to give a warning that a particular assumption was
1903 made. All such warnings are directed to the standard error file.
1907 @cindex suppressing warnings
1908 @cindex warnings, suppressing
1909 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
1910 This only affects the warning messages: it does not change any particular of
1911 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
1914 @kindex --fatal-warnings
1915 @cindex errors, caused by warnings
1916 @cindex warnings, causing error
1917 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
1918 files that generate warnings to be in error.
1921 @cindex warnings, switching on
1922 You can switch these options off again by specifying @option{--warn}, which
1923 causes warnings to be output as usual.
1926 @section Generate Object File in Spite of Errors: @option{-Z}
1927 @cindex object file, after errors
1928 @cindex errors, continuing after
1929 After an error message, @command{@value{AS}} normally produces no output. If for
1930 some reason you are interested in object file output even after
1931 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
1932 option. If there are any errors, @command{@value{AS}} continues anyways, and
1933 writes an object file after a final warning message of the form @samp{@var{n}
1934 errors, @var{m} warnings, generating bad object file.}
1939 @cindex machine-independent syntax
1940 @cindex syntax, machine-independent
1941 This chapter describes the machine-independent syntax allowed in a
1942 source file. @command{@value{AS}} syntax is similar to what many other
1943 assemblers use; it is inspired by the BSD 4.2
1948 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
1952 * Preprocessing:: Preprocessing
1953 * Whitespace:: Whitespace
1954 * Comments:: Comments
1955 * Symbol Intro:: Symbols
1956 * Statements:: Statements
1957 * Constants:: Constants
1961 @section Preprocessing
1963 @cindex preprocessing
1964 The @command{@value{AS}} internal preprocessor:
1966 @cindex whitespace, removed by preprocessor
1968 adjusts and removes extra whitespace. It leaves one space or tab before
1969 the keywords on a line, and turns any other whitespace on the line into
1972 @cindex comments, removed by preprocessor
1974 removes all comments, replacing them with a single space, or an
1975 appropriate number of newlines.
1977 @cindex constants, converted by preprocessor
1979 converts character constants into the appropriate numeric values.
1982 It does not do macro processing, include file handling, or
1983 anything else you may get from your C compiler's preprocessor. You can
1984 do include file processing with the @code{.include} directive
1985 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
1986 to get other ``CPP'' style preprocessing by giving the input file a
1987 @samp{.S} suffix. @xref{Overall Options,, Options Controlling the Kind of
1988 Output, gcc.info, Using GNU CC}.
1990 Excess whitespace, comments, and character constants
1991 cannot be used in the portions of the input text that are not
1994 @cindex turning preprocessing on and off
1995 @cindex preprocessing, turning on and off
1998 If the first line of an input file is @code{#NO_APP} or if you use the
1999 @samp{-f} option, whitespace and comments are not removed from the input file.
2000 Within an input file, you can ask for whitespace and comment removal in
2001 specific portions of the by putting a line that says @code{#APP} before the
2002 text that may contain whitespace or comments, and putting a line that says
2003 @code{#NO_APP} after this text. This feature is mainly intend to support
2004 @code{asm} statements in compilers whose output is otherwise free of comments
2011 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2012 Whitespace is used to separate symbols, and to make programs neater for
2013 people to read. Unless within character constants
2014 (@pxref{Characters,,Character Constants}), any whitespace means the same
2015 as exactly one space.
2021 There are two ways of rendering comments to @command{@value{AS}}. In both
2022 cases the comment is equivalent to one space.
2024 Anything from @samp{/*} through the next @samp{*/} is a comment.
2025 This means you may not nest these comments.
2029 The only way to include a newline ('\n') in a comment
2030 is to use this sort of comment.
2033 /* This sort of comment does not nest. */
2036 @cindex line comment character
2037 Anything from the @dfn{line comment} character to the next newline
2038 is considered a comment and is ignored. The line comment character is
2040 @samp{;} for the AMD 29K family;
2043 @samp{;} on the ARC;
2046 @samp{@@} on the ARM;
2049 @samp{;} for the H8/300 family;
2052 @samp{!} for the H8/500 family;
2055 @samp{;} for the HPPA;
2058 @samp{#} on the i386 and x86-64;
2061 @samp{#} on the i960;
2064 @samp{;} for the PDP-11;
2067 @samp{;} for picoJava;
2070 @samp{#} for Motorola PowerPC;
2073 @samp{!} for the Renesas / SuperH SH;
2076 @samp{!} on the SPARC;
2079 @samp{#} on the ip2k;
2082 @samp{#} on the m32r;
2085 @samp{|} on the 680x0;
2088 @samp{#} on the 68HC11 and 68HC12;
2091 @samp{;} on the M880x0;
2094 @samp{#} on the Vax;
2097 @samp{!} for the Z8000;
2100 @samp{#} on the V850;
2103 @samp{#} for Xtensa systems;
2105 see @ref{Machine Dependencies}. @refill
2106 @c FIXME What about i860?
2109 On some machines there are two different line comment characters. One
2110 character only begins a comment if it is the first non-whitespace character on
2111 a line, while the other always begins a comment.
2115 The V850 assembler also supports a double dash as starting a comment that
2116 extends to the end of the line.
2122 @cindex lines starting with @code{#}
2123 @cindex logical line numbers
2124 To be compatible with past assemblers, lines that begin with @samp{#} have a
2125 special interpretation. Following the @samp{#} should be an absolute
2126 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2127 line. Then a string (@pxref{Strings,, Strings}) is allowed: if present it is a
2128 new logical file name. The rest of the line, if any, should be whitespace.
2130 If the first non-whitespace characters on the line are not numeric,
2131 the line is ignored. (Just like a comment.)
2134 # This is an ordinary comment.
2135 # 42-6 "new_file_name" # New logical file name
2136 # This is logical line # 36.
2138 This feature is deprecated, and may disappear from future versions
2139 of @command{@value{AS}}.
2144 @cindex characters used in symbols
2145 @ifclear SPECIAL-SYMS
2146 A @dfn{symbol} is one or more characters chosen from the set of all
2147 letters (both upper and lower case), digits and the three characters
2153 A @dfn{symbol} is one or more characters chosen from the set of all
2154 letters (both upper and lower case), digits and the three characters
2155 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2161 On most machines, you can also use @code{$} in symbol names; exceptions
2162 are noted in @ref{Machine Dependencies}.
2164 No symbol may begin with a digit. Case is significant.
2165 There is no length limit: all characters are significant. Symbols are
2166 delimited by characters not in that set, or by the beginning of a file
2167 (since the source program must end with a newline, the end of a file is
2168 not a possible symbol delimiter). @xref{Symbols}.
2169 @cindex length of symbols
2174 @cindex statements, structure of
2175 @cindex line separator character
2176 @cindex statement separator character
2178 @ifclear abnormal-separator
2179 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2180 semicolon (@samp{;}). The newline or semicolon is considered part of
2181 the preceding statement. Newlines and semicolons within character
2182 constants are an exception: they do not end statements.
2184 @ifset abnormal-separator
2186 A @dfn{statement} ends at a newline character (@samp{\n}) or an ``at''
2187 sign (@samp{@@}). The newline or at sign is considered part of the
2188 preceding statement. Newlines and at signs within character constants
2189 are an exception: they do not end statements.
2192 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2193 point (@samp{!}). The newline or exclamation point is considered part of the
2194 preceding statement. Newlines and exclamation points within character
2195 constants are an exception: they do not end statements.
2198 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2199 H8/300) a dollar sign (@samp{$}); or (for the
2202 (@samp{;}). The newline or separator character is considered part of
2203 the preceding statement. Newlines and separators within character
2204 constants are an exception: they do not end statements.
2209 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2210 separator character. (The line separator is usually @samp{;}, unless
2211 this conflicts with the comment character; @pxref{Machine Dependencies}.) The
2212 newline or separator character is considered part of the preceding
2213 statement. Newlines and separators within character constants are an
2214 exception: they do not end statements.
2217 @cindex newline, required at file end
2218 @cindex EOF, newline must precede
2219 It is an error to end any statement with end-of-file: the last
2220 character of any input file should be a newline.@refill
2222 An empty statement is allowed, and may include whitespace. It is ignored.
2224 @cindex instructions and directives
2225 @cindex directives and instructions
2226 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2227 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2229 A statement begins with zero or more labels, optionally followed by a
2230 key symbol which determines what kind of statement it is. The key
2231 symbol determines the syntax of the rest of the statement. If the
2232 symbol begins with a dot @samp{.} then the statement is an assembler
2233 directive: typically valid for any computer. If the symbol begins with
2234 a letter the statement is an assembly language @dfn{instruction}: it
2235 assembles into a machine language instruction.
2237 Different versions of @command{@value{AS}} for different computers
2238 recognize different instructions. In fact, the same symbol may
2239 represent a different instruction in a different computer's assembly
2243 @cindex @code{:} (label)
2244 @cindex label (@code{:})
2245 A label is a symbol immediately followed by a colon (@code{:}).
2246 Whitespace before a label or after a colon is permitted, but you may not
2247 have whitespace between a label's symbol and its colon. @xref{Labels}.
2250 For HPPA targets, labels need not be immediately followed by a colon, but
2251 the definition of a label must begin in column zero. This also implies that
2252 only one label may be defined on each line.
2256 label: .directive followed by something
2257 another_label: # This is an empty statement.
2258 instruction operand_1, operand_2, @dots{}
2265 A constant is a number, written so that its value is known by
2266 inspection, without knowing any context. Like this:
2269 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2270 .ascii "Ring the bell\7" # A string constant.
2271 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2272 .float 0f-314159265358979323846264338327\
2273 95028841971.693993751E-40 # - pi, a flonum.
2278 * Characters:: Character Constants
2279 * Numbers:: Number Constants
2283 @subsection Character Constants
2285 @cindex character constants
2286 @cindex constants, character
2287 There are two kinds of character constants. A @dfn{character} stands
2288 for one character in one byte and its value may be used in
2289 numeric expressions. String constants (properly called string
2290 @emph{literals}) are potentially many bytes and their values may not be
2291 used in arithmetic expressions.
2295 * Chars:: Characters
2299 @subsubsection Strings
2301 @cindex string constants
2302 @cindex constants, string
2303 A @dfn{string} is written between double-quotes. It may contain
2304 double-quotes or null characters. The way to get special characters
2305 into a string is to @dfn{escape} these characters: precede them with
2306 a backslash @samp{\} character. For example @samp{\\} represents
2307 one backslash: the first @code{\} is an escape which tells
2308 @command{@value{AS}} to interpret the second character literally as a backslash
2309 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2310 escape character). The complete list of escapes follows.
2312 @cindex escape codes, character
2313 @cindex character escape codes
2316 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2318 @cindex @code{\b} (backspace character)
2319 @cindex backspace (@code{\b})
2321 Mnemonic for backspace; for ASCII this is octal code 010.
2324 @c Mnemonic for EOText; for ASCII this is octal code 004.
2326 @cindex @code{\f} (formfeed character)
2327 @cindex formfeed (@code{\f})
2329 Mnemonic for FormFeed; for ASCII this is octal code 014.
2331 @cindex @code{\n} (newline character)
2332 @cindex newline (@code{\n})
2334 Mnemonic for newline; for ASCII this is octal code 012.
2337 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2339 @cindex @code{\r} (carriage return character)
2340 @cindex carriage return (@code{\r})
2342 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2345 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2346 @c other assemblers.
2348 @cindex @code{\t} (tab)
2349 @cindex tab (@code{\t})
2351 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2354 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2355 @c @item \x @var{digit} @var{digit} @var{digit}
2356 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2358 @cindex @code{\@var{ddd}} (octal character code)
2359 @cindex octal character code (@code{\@var{ddd}})
2360 @item \ @var{digit} @var{digit} @var{digit}
2361 An octal character code. The numeric code is 3 octal digits.
2362 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2363 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2365 @cindex @code{\@var{xd...}} (hex character code)
2366 @cindex hex character code (@code{\@var{xd...}})
2367 @item \@code{x} @var{hex-digits...}
2368 A hex character code. All trailing hex digits are combined. Either upper or
2369 lower case @code{x} works.
2371 @cindex @code{\\} (@samp{\} character)
2372 @cindex backslash (@code{\\})
2374 Represents one @samp{\} character.
2377 @c Represents one @samp{'} (accent acute) character.
2378 @c This is needed in single character literals
2379 @c (@xref{Characters,,Character Constants}.) to represent
2382 @cindex @code{\"} (doublequote character)
2383 @cindex doublequote (@code{\"})
2385 Represents one @samp{"} character. Needed in strings to represent
2386 this character, because an unescaped @samp{"} would end the string.
2388 @item \ @var{anything-else}
2389 Any other character when escaped by @kbd{\} gives a warning, but
2390 assembles as if the @samp{\} was not present. The idea is that if
2391 you used an escape sequence you clearly didn't want the literal
2392 interpretation of the following character. However @command{@value{AS}} has no
2393 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2394 code and warns you of the fact.
2397 Which characters are escapable, and what those escapes represent,
2398 varies widely among assemblers. The current set is what we think
2399 the BSD 4.2 assembler recognizes, and is a subset of what most C
2400 compilers recognize. If you are in doubt, do not use an escape
2404 @subsubsection Characters
2406 @cindex single character constant
2407 @cindex character, single
2408 @cindex constant, single character
2409 A single character may be written as a single quote immediately
2410 followed by that character. The same escapes apply to characters as
2411 to strings. So if you want to write the character backslash, you
2412 must write @kbd{'\\} where the first @code{\} escapes the second
2413 @code{\}. As you can see, the quote is an acute accent, not a
2414 grave accent. A newline
2416 @ifclear abnormal-separator
2417 (or semicolon @samp{;})
2419 @ifset abnormal-separator
2421 (or at sign @samp{@@})
2424 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2425 Renesas SH or H8/500)
2429 immediately following an acute accent is taken as a literal character
2430 and does not count as the end of a statement. The value of a character
2431 constant in a numeric expression is the machine's byte-wide code for
2432 that character. @command{@value{AS}} assumes your character code is ASCII:
2433 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2436 @subsection Number Constants
2438 @cindex constants, number
2439 @cindex number constants
2440 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2441 are stored in the target machine. @emph{Integers} are numbers that
2442 would fit into an @code{int} in the C language. @emph{Bignums} are
2443 integers, but they are stored in more than 32 bits. @emph{Flonums}
2444 are floating point numbers, described below.
2447 * Integers:: Integers
2452 * Bit Fields:: Bit Fields
2458 @subsubsection Integers
2460 @cindex constants, integer
2462 @cindex binary integers
2463 @cindex integers, binary
2464 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2465 the binary digits @samp{01}.
2467 @cindex octal integers
2468 @cindex integers, octal
2469 An octal integer is @samp{0} followed by zero or more of the octal
2470 digits (@samp{01234567}).
2472 @cindex decimal integers
2473 @cindex integers, decimal
2474 A decimal integer starts with a non-zero digit followed by zero or
2475 more digits (@samp{0123456789}).
2477 @cindex hexadecimal integers
2478 @cindex integers, hexadecimal
2479 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2480 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2482 Integers have the usual values. To denote a negative integer, use
2483 the prefix operator @samp{-} discussed under expressions
2484 (@pxref{Prefix Ops,,Prefix Operators}).
2487 @subsubsection Bignums
2490 @cindex constants, bignum
2491 A @dfn{bignum} has the same syntax and semantics as an integer
2492 except that the number (or its negative) takes more than 32 bits to
2493 represent in binary. The distinction is made because in some places
2494 integers are permitted while bignums are not.
2497 @subsubsection Flonums
2499 @cindex floating point numbers
2500 @cindex constants, floating point
2502 @cindex precision, floating point
2503 A @dfn{flonum} represents a floating point number. The translation is
2504 indirect: a decimal floating point number from the text is converted by
2505 @command{@value{AS}} to a generic binary floating point number of more than
2506 sufficient precision. This generic floating point number is converted
2507 to a particular computer's floating point format (or formats) by a
2508 portion of @command{@value{AS}} specialized to that computer.
2510 A flonum is written by writing (in order)
2515 (@samp{0} is optional on the HPPA.)
2519 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2521 @kbd{e} is recommended. Case is not important.
2523 @c FIXME: verify if flonum syntax really this vague for most cases
2524 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2525 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2528 On the H8/300, H8/500,
2529 Renesas / SuperH SH,
2530 and AMD 29K architectures, the letter must be
2531 one of the letters @samp{DFPRSX} (in upper or lower case).
2533 On the ARC, the letter must be one of the letters @samp{DFRS}
2534 (in upper or lower case).
2536 On the Intel 960 architecture, the letter must be
2537 one of the letters @samp{DFT} (in upper or lower case).
2539 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2543 One of the letters @samp{DFPRSX} (in upper or lower case).
2546 One of the letters @samp{DFRS} (in upper or lower case).
2549 One of the letters @samp{DFPRSX} (in upper or lower case).
2552 The letter @samp{E} (upper case only).
2555 One of the letters @samp{DFT} (in upper or lower case).
2560 An optional sign: either @samp{+} or @samp{-}.
2563 An optional @dfn{integer part}: zero or more decimal digits.
2566 An optional @dfn{fractional part}: @samp{.} followed by zero
2567 or more decimal digits.
2570 An optional exponent, consisting of:
2574 An @samp{E} or @samp{e}.
2575 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2576 @c principle this can perfectly well be different on different targets.
2578 Optional sign: either @samp{+} or @samp{-}.
2580 One or more decimal digits.
2585 At least one of the integer part or the fractional part must be
2586 present. The floating point number has the usual base-10 value.
2588 @command{@value{AS}} does all processing using integers. Flonums are computed
2589 independently of any floating point hardware in the computer running
2590 @command{@value{AS}}.
2594 @c Bit fields are written as a general facility but are also controlled
2595 @c by a conditional-compilation flag---which is as of now (21mar91)
2596 @c turned on only by the i960 config of GAS.
2598 @subsubsection Bit Fields
2601 @cindex constants, bit field
2602 You can also define numeric constants as @dfn{bit fields}.
2603 specify two numbers separated by a colon---
2605 @var{mask}:@var{value}
2608 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2611 The resulting number is then packed
2613 @c this conditional paren in case bit fields turned on elsewhere than 960
2614 (in host-dependent byte order)
2616 into a field whose width depends on which assembler directive has the
2617 bit-field as its argument. Overflow (a result from the bitwise and
2618 requiring more binary digits to represent) is not an error; instead,
2619 more constants are generated, of the specified width, beginning with the
2620 least significant digits.@refill
2622 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2623 @code{.short}, and @code{.word} accept bit-field arguments.
2628 @chapter Sections and Relocation
2633 * Secs Background:: Background
2634 * Ld Sections:: Linker Sections
2635 * As Sections:: Assembler Internal Sections
2636 * Sub-Sections:: Sub-Sections
2640 @node Secs Background
2643 Roughly, a section is a range of addresses, with no gaps; all data
2644 ``in'' those addresses is treated the same for some particular purpose.
2645 For example there may be a ``read only'' section.
2647 @cindex linker, and assembler
2648 @cindex assembler, and linker
2649 The linker @code{@value{LD}} reads many object files (partial programs) and
2650 combines their contents to form a runnable program. When @command{@value{AS}}
2651 emits an object file, the partial program is assumed to start at address 0.
2652 @code{@value{LD}} assigns the final addresses for the partial program, so that
2653 different partial programs do not overlap. This is actually an
2654 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2657 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2658 addresses. These blocks slide to their run-time addresses as rigid
2659 units; their length does not change and neither does the order of bytes
2660 within them. Such a rigid unit is called a @emph{section}. Assigning
2661 run-time addresses to sections is called @dfn{relocation}. It includes
2662 the task of adjusting mentions of object-file addresses so they refer to
2663 the proper run-time addresses.
2665 For the H8/300 and H8/500,
2666 and for the Renesas / SuperH SH,
2667 @command{@value{AS}} pads sections if needed to
2668 ensure they end on a word (sixteen bit) boundary.
2671 @cindex standard assembler sections
2672 An object file written by @command{@value{AS}} has at least three sections, any
2673 of which may be empty. These are named @dfn{text}, @dfn{data} and
2678 When it generates COFF or ELF output,
2680 @command{@value{AS}} can also generate whatever other named sections you specify
2681 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2682 If you do not use any directives that place output in the @samp{.text}
2683 or @samp{.data} sections, these sections still exist, but are empty.
2688 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2690 @command{@value{AS}} can also generate whatever other named sections you
2691 specify using the @samp{.space} and @samp{.subspace} directives. See
2692 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2693 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2694 assembler directives.
2697 Additionally, @command{@value{AS}} uses different names for the standard
2698 text, data, and bss sections when generating SOM output. Program text
2699 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2700 BSS into @samp{$BSS$}.
2704 Within the object file, the text section starts at address @code{0}, the
2705 data section follows, and the bss section follows the data section.
2708 When generating either SOM or ELF output files on the HPPA, the text
2709 section starts at address @code{0}, the data section at address
2710 @code{0x4000000}, and the bss section follows the data section.
2713 To let @code{@value{LD}} know which data changes when the sections are
2714 relocated, and how to change that data, @command{@value{AS}} also writes to the
2715 object file details of the relocation needed. To perform relocation
2716 @code{@value{LD}} must know, each time an address in the object
2720 Where in the object file is the beginning of this reference to
2723 How long (in bytes) is this reference?
2725 Which section does the address refer to? What is the numeric value of
2727 (@var{address}) @minus{} (@var{start-address of section})?
2730 Is the reference to an address ``Program-Counter relative''?
2733 @cindex addresses, format of
2734 @cindex section-relative addressing
2735 In fact, every address @command{@value{AS}} ever uses is expressed as
2737 (@var{section}) + (@var{offset into section})
2740 Further, most expressions @command{@value{AS}} computes have this section-relative
2743 (For some object formats, such as SOM for the HPPA, some expressions are
2744 symbol-relative instead.)
2747 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2748 @var{N} into section @var{secname}.''
2750 Apart from text, data and bss sections you need to know about the
2751 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2752 addresses in the absolute section remain unchanged. For example, address
2753 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2754 @code{@value{LD}}. Although the linker never arranges two partial programs'
2755 data sections with overlapping addresses after linking, @emph{by definition}
2756 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2757 part of a program is always the same address when the program is running as
2758 address @code{@{absolute@ 239@}} in any other part of the program.
2760 The idea of sections is extended to the @dfn{undefined} section. Any
2761 address whose section is unknown at assembly time is by definition
2762 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2763 Since numbers are always defined, the only way to generate an undefined
2764 address is to mention an undefined symbol. A reference to a named
2765 common block would be such a symbol: its value is unknown at assembly
2766 time so it has section @emph{undefined}.
2768 By analogy the word @emph{section} is used to describe groups of sections in
2769 the linked program. @code{@value{LD}} puts all partial programs' text
2770 sections in contiguous addresses in the linked program. It is
2771 customary to refer to the @emph{text section} of a program, meaning all
2772 the addresses of all partial programs' text sections. Likewise for
2773 data and bss sections.
2775 Some sections are manipulated by @code{@value{LD}}; others are invented for
2776 use of @command{@value{AS}} and have no meaning except during assembly.
2779 @section Linker Sections
2780 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2785 @cindex named sections
2786 @cindex sections, named
2787 @item named sections
2790 @cindex text section
2791 @cindex data section
2795 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2796 separate but equal sections. Anything you can say of one section is
2799 When the program is running, however, it is
2800 customary for the text section to be unalterable. The
2801 text section is often shared among processes: it contains
2802 instructions, constants and the like. The data section of a running
2803 program is usually alterable: for example, C variables would be stored
2804 in the data section.
2809 This section contains zeroed bytes when your program begins running. It
2810 is used to hold uninitialized variables or common storage. The length of
2811 each partial program's bss section is important, but because it starts
2812 out containing zeroed bytes there is no need to store explicit zero
2813 bytes in the object file. The bss section was invented to eliminate
2814 those explicit zeros from object files.
2816 @cindex absolute section
2817 @item absolute section
2818 Address 0 of this section is always ``relocated'' to runtime address 0.
2819 This is useful if you want to refer to an address that @code{@value{LD}} must
2820 not change when relocating. In this sense we speak of absolute
2821 addresses being ``unrelocatable'': they do not change during relocation.
2823 @cindex undefined section
2824 @item undefined section
2825 This ``section'' is a catch-all for address references to objects not in
2826 the preceding sections.
2827 @c FIXME: ref to some other doc on obj-file formats could go here.
2830 @cindex relocation example
2831 An idealized example of three relocatable sections follows.
2833 The example uses the traditional section names @samp{.text} and @samp{.data}.
2835 Memory addresses are on the horizontal axis.
2839 @c END TEXI2ROFF-KILL
2842 partial program # 1: |ttttt|dddd|00|
2849 partial program # 2: |TTT|DDD|000|
2852 +--+---+-----+--+----+---+-----+~~
2853 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2854 +--+---+-----+--+----+---+-----+~~
2856 addresses: 0 @dots{}
2863 \line{\it Partial program \#1: \hfil}
2864 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2865 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2867 \line{\it Partial program \#2: \hfil}
2868 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2869 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2871 \line{\it linked program: \hfil}
2872 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2873 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2874 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2875 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2877 \line{\it addresses: \hfil}
2881 @c END TEXI2ROFF-KILL
2884 @section Assembler Internal Sections
2886 @cindex internal assembler sections
2887 @cindex sections in messages, internal
2888 These sections are meant only for the internal use of @command{@value{AS}}. They
2889 have no meaning at run-time. You do not really need to know about these
2890 sections for most purposes; but they can be mentioned in @command{@value{AS}}
2891 warning messages, so it might be helpful to have an idea of their
2892 meanings to @command{@value{AS}}. These sections are used to permit the
2893 value of every expression in your assembly language program to be a
2894 section-relative address.
2897 @cindex assembler internal logic error
2898 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
2899 An internal assembler logic error has been found. This means there is a
2900 bug in the assembler.
2902 @cindex expr (internal section)
2904 The assembler stores complex expression internally as combinations of
2905 symbols. When it needs to represent an expression as a symbol, it puts
2906 it in the expr section.
2908 @c FIXME item transfer[t] vector preload
2909 @c FIXME item transfer[t] vector postload
2910 @c FIXME item register
2914 @section Sub-Sections
2916 @cindex numbered subsections
2917 @cindex grouping data
2923 fall into two sections: text and data.
2925 You may have separate groups of
2927 data in named sections
2931 data in named sections
2937 that you want to end up near to each other in the object file, even though they
2938 are not contiguous in the assembler source. @command{@value{AS}} allows you to
2939 use @dfn{subsections} for this purpose. Within each section, there can be
2940 numbered subsections with values from 0 to 8192. Objects assembled into the
2941 same subsection go into the object file together with other objects in the same
2942 subsection. For example, a compiler might want to store constants in the text
2943 section, but might not want to have them interspersed with the program being
2944 assembled. In this case, the compiler could issue a @samp{.text 0} before each
2945 section of code being output, and a @samp{.text 1} before each group of
2946 constants being output.
2948 Subsections are optional. If you do not use subsections, everything
2949 goes in subsection number zero.
2952 Each subsection is zero-padded up to a multiple of four bytes.
2953 (Subsections may be padded a different amount on different flavors
2954 of @command{@value{AS}}.)
2958 On the H8/300 and H8/500 platforms, each subsection is zero-padded to a word
2959 boundary (two bytes).
2960 The same is true on the Renesas SH.
2963 @c FIXME section padding (alignment)?
2964 @c Rich Pixley says padding here depends on target obj code format; that
2965 @c doesn't seem particularly useful to say without further elaboration,
2966 @c so for now I say nothing about it. If this is a generic BFD issue,
2967 @c these paragraphs might need to vanish from this manual, and be
2968 @c discussed in BFD chapter of binutils (or some such).
2971 On the AMD 29K family, no particular padding is added to section or
2972 subsection sizes; @value{AS} forces no alignment on this platform.
2976 Subsections appear in your object file in numeric order, lowest numbered
2977 to highest. (All this to be compatible with other people's assemblers.)
2978 The object file contains no representation of subsections; @code{@value{LD}} and
2979 other programs that manipulate object files see no trace of them.
2980 They just see all your text subsections as a text section, and all your
2981 data subsections as a data section.
2983 To specify which subsection you want subsequent statements assembled
2984 into, use a numeric argument to specify it, in a @samp{.text
2985 @var{expression}} or a @samp{.data @var{expression}} statement.
2988 When generating COFF output, you
2993 can also use an extra subsection
2994 argument with arbitrary named sections: @samp{.section @var{name},
2999 When generating ELF output, you
3004 can also use the @code{.subsection} directive (@pxref{SubSection})
3005 to specify a subsection: @samp{.subsection @var{expression}}.
3007 @var{Expression} should be an absolute expression.
3008 (@xref{Expressions}.) If you just say @samp{.text} then @samp{.text 0}
3009 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3010 begins in @code{text 0}. For instance:
3012 .text 0 # The default subsection is text 0 anyway.
3013 .ascii "This lives in the first text subsection. *"
3015 .ascii "But this lives in the second text subsection."
3017 .ascii "This lives in the data section,"
3018 .ascii "in the first data subsection."
3020 .ascii "This lives in the first text section,"
3021 .ascii "immediately following the asterisk (*)."
3024 Each section has a @dfn{location counter} incremented by one for every byte
3025 assembled into that section. Because subsections are merely a convenience
3026 restricted to @command{@value{AS}} there is no concept of a subsection location
3027 counter. There is no way to directly manipulate a location counter---but the
3028 @code{.align} directive changes it, and any label definition captures its
3029 current value. The location counter of the section where statements are being
3030 assembled is said to be the @dfn{active} location counter.
3033 @section bss Section
3036 @cindex common variable storage
3037 The bss section is used for local common variable storage.
3038 You may allocate address space in the bss section, but you may
3039 not dictate data to load into it before your program executes. When
3040 your program starts running, all the contents of the bss
3041 section are zeroed bytes.
3043 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3044 @ref{Lcomm,,@code{.lcomm}}.
3046 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3047 another form of uninitialized symbol; see @xref{Comm,,@code{.comm}}.
3050 When assembling for a target which supports multiple sections, such as ELF or
3051 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3052 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3053 section. Typically the section will only contain symbol definitions and
3054 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3061 Symbols are a central concept: the programmer uses symbols to name
3062 things, the linker uses symbols to link, and the debugger uses symbols
3066 @cindex debuggers, and symbol order
3067 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3068 the same order they were declared. This may break some debuggers.
3073 * Setting Symbols:: Giving Symbols Other Values
3074 * Symbol Names:: Symbol Names
3075 * Dot:: The Special Dot Symbol
3076 * Symbol Attributes:: Symbol Attributes
3083 A @dfn{label} is written as a symbol immediately followed by a colon
3084 @samp{:}. The symbol then represents the current value of the
3085 active location counter, and is, for example, a suitable instruction
3086 operand. You are warned if you use the same symbol to represent two
3087 different locations: the first definition overrides any other
3091 On the HPPA, the usual form for a label need not be immediately followed by a
3092 colon, but instead must start in column zero. Only one label may be defined on
3093 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3094 provides a special directive @code{.label} for defining labels more flexibly.
3097 @node Setting Symbols
3098 @section Giving Symbols Other Values
3100 @cindex assigning values to symbols
3101 @cindex symbol values, assigning
3102 A symbol can be given an arbitrary value by writing a symbol, followed
3103 by an equals sign @samp{=}, followed by an expression
3104 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3105 directive. @xref{Set,,@code{.set}}.
3108 @section Symbol Names
3110 @cindex symbol names
3111 @cindex names, symbol
3112 @ifclear SPECIAL-SYMS
3113 Symbol names begin with a letter or with one of @samp{._}. On most
3114 machines, you can also use @code{$} in symbol names; exceptions are
3115 noted in @ref{Machine Dependencies}. That character may be followed by any
3116 string of digits, letters, dollar signs (unless otherwise noted in
3117 @ref{Machine Dependencies}), and underscores.
3120 For the AMD 29K family, @samp{?} is also allowed in the
3121 body of a symbol name, though not at its beginning.
3126 Symbol names begin with a letter or with one of @samp{._}. On the
3127 Renesas SH or the H8/500, you can also use @code{$} in symbol names. That
3128 character may be followed by any string of digits, letters, dollar signs (save
3129 on the H8/300), and underscores.
3133 Case of letters is significant: @code{foo} is a different symbol name
3136 Each symbol has exactly one name. Each name in an assembly language program
3137 refers to exactly one symbol. You may use that symbol name any number of times
3140 @subheading Local Symbol Names
3142 @cindex local symbol names
3143 @cindex symbol names, local
3144 @cindex temporary symbol names
3145 @cindex symbol names, temporary
3146 Local symbols help compilers and programmers use names temporarily.
3147 They create symbols which are guaranteed to be unique over the entire scope of
3148 the input source code and which can be referred to by a simple notation.
3149 To define a local symbol, write a label of the form @samp{@b{N}:} (where @b{N}
3150 represents any positive integer). To refer to the most recent previous
3151 definition of that symbol write @samp{@b{N}b}, using the same number as when
3152 you defined the label. To refer to the next definition of a local label, write
3153 @samp{@b{N}f}--- The @samp{b} stands for``backwards'' and the @samp{f} stands
3156 There is no restriction on how you can use these labels, and you can reuse them
3157 too. So that it is possible to repeatedly define the same local label (using
3158 the same number @samp{@b{N}}), although you can only refer to the most recently
3159 defined local label of that number (for a backwards reference) or the next
3160 definition of a specific local label for a forward reference. It is also worth
3161 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3162 implemented in a slightly more efficient manner than the others.
3173 Which is the equivalent of:
3176 label_1: branch label_3
3177 label_2: branch label_1
3178 label_3: branch label_4
3179 label_4: branch label_3
3182 Local symbol names are only a notational device. They are immediately
3183 transformed into more conventional symbol names before the assembler uses them.
3184 The symbol names stored in the symbol table, appearing in error messages and
3185 optionally emitted to the object file. The names are constructed using these
3190 All local labels begin with @samp{L}. Normally both @command{@value{AS}} and
3191 @code{@value{LD}} forget symbols that start with @samp{L}. These labels are
3192 used for symbols you are never intended to see. If you use the
3193 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3194 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3195 you may use them in debugging.
3198 This is the number that was used in the local label definition. So if the
3199 label is written @samp{55:} then the number is @samp{55}.
3202 This unusual character is included so you do not accidentally invent a symbol
3203 of the same name. The character has ASCII value of @samp{\002} (control-B).
3205 @item @emph{ordinal number}
3206 This is a serial number to keep the labels distinct. The first definition of
3207 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3208 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3209 the number @samp{1} and its 15th defintion gets @samp{15} as well.
3212 So for example, the first @code{1:} is named @code{L1@kbd{C-B}1}, the 44th
3213 @code{3:} is named @code{L3@kbd{C-B}44}.
3215 @subheading Dollar Local Labels
3216 @cindex dollar local symbols
3218 @code{@value{AS}} also supports an even more local form of local labels called
3219 dollar labels. These labels go out of scope (ie they become undefined) as soon
3220 as a non-local label is defined. Thus they remain valid for only a small
3221 region of the input source code. Normal local labels, by contrast, remain in
3222 scope for the entire file, or until they are redefined by another occurrence of
3223 the same local label.
3225 Dollar labels are defined in exactly the same way as ordinary local labels,
3226 except that instead of being terminated by a colon, they are terminated by a
3227 dollar sign. eg @samp{@b{55$}}.
3229 They can also be distinguished from ordinary local labels by their transformed
3230 name which uses ASCII character @samp{\001} (control-A) as the magic character
3231 to distinguish them from ordinary labels. Thus the 5th defintion of @samp{6$}
3232 is named @samp{L6@kbd{C-A}5}.
3235 @section The Special Dot Symbol
3237 @cindex dot (symbol)
3238 @cindex @code{.} (symbol)
3239 @cindex current address
3240 @cindex location counter
3241 The special symbol @samp{.} refers to the current address that
3242 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3243 .long .} defines @code{melvin} to contain its own address.
3244 Assigning a value to @code{.} is treated the same as a @code{.org}
3245 directive. Thus, the expression @samp{.=.+4} is the same as saying
3246 @ifclear no-space-dir
3255 @node Symbol Attributes
3256 @section Symbol Attributes
3258 @cindex symbol attributes
3259 @cindex attributes, symbol
3260 Every symbol has, as well as its name, the attributes ``Value'' and
3261 ``Type''. Depending on output format, symbols can also have auxiliary
3264 The detailed definitions are in @file{a.out.h}.
3267 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3268 all these attributes, and probably won't warn you. This makes the
3269 symbol an externally defined symbol, which is generally what you
3273 * Symbol Value:: Value
3274 * Symbol Type:: Type
3277 * a.out Symbols:: Symbol Attributes: @code{a.out}
3281 * a.out Symbols:: Symbol Attributes: @code{a.out}
3284 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3289 * COFF Symbols:: Symbol Attributes for COFF
3292 * SOM Symbols:: Symbol Attributes for SOM
3299 @cindex value of a symbol
3300 @cindex symbol value
3301 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3302 location in the text, data, bss or absolute sections the value is the
3303 number of addresses from the start of that section to the label.
3304 Naturally for text, data and bss sections the value of a symbol changes
3305 as @code{@value{LD}} changes section base addresses during linking. Absolute
3306 symbols' values do not change during linking: that is why they are
3309 The value of an undefined symbol is treated in a special way. If it is
3310 0 then the symbol is not defined in this assembler source file, and
3311 @code{@value{LD}} tries to determine its value from other files linked into the
3312 same program. You make this kind of symbol simply by mentioning a symbol
3313 name without defining it. A non-zero value represents a @code{.comm}
3314 common declaration. The value is how much common storage to reserve, in
3315 bytes (addresses). The symbol refers to the first address of the
3321 @cindex type of a symbol
3323 The type attribute of a symbol contains relocation (section)
3324 information, any flag settings indicating that a symbol is external, and
3325 (optionally), other information for linkers and debuggers. The exact
3326 format depends on the object-code output format in use.
3331 @c The following avoids a "widow" subsection title. @group would be
3332 @c better if it were available outside examples.
3335 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3337 @cindex @code{b.out} symbol attributes
3338 @cindex symbol attributes, @code{b.out}
3339 These symbol attributes appear only when @command{@value{AS}} is configured for
3340 one of the Berkeley-descended object output formats---@code{a.out} or
3346 @subsection Symbol Attributes: @code{a.out}
3348 @cindex @code{a.out} symbol attributes
3349 @cindex symbol attributes, @code{a.out}
3355 @subsection Symbol Attributes: @code{a.out}
3357 @cindex @code{a.out} symbol attributes
3358 @cindex symbol attributes, @code{a.out}
3362 * Symbol Desc:: Descriptor
3363 * Symbol Other:: Other
3367 @subsubsection Descriptor
3369 @cindex descriptor, of @code{a.out} symbol
3370 This is an arbitrary 16-bit value. You may establish a symbol's
3371 descriptor value by using a @code{.desc} statement
3372 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3373 @command{@value{AS}}.
3376 @subsubsection Other
3378 @cindex other attribute, of @code{a.out} symbol
3379 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3384 @subsection Symbol Attributes for COFF
3386 @cindex COFF symbol attributes
3387 @cindex symbol attributes, COFF
3389 The COFF format supports a multitude of auxiliary symbol attributes;
3390 like the primary symbol attributes, they are set between @code{.def} and
3391 @code{.endef} directives.
3393 @subsubsection Primary Attributes
3395 @cindex primary attributes, COFF symbols
3396 The symbol name is set with @code{.def}; the value and type,
3397 respectively, with @code{.val} and @code{.type}.
3399 @subsubsection Auxiliary Attributes
3401 @cindex auxiliary attributes, COFF symbols
3402 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3403 @code{.size}, and @code{.tag} can generate auxiliary symbol table
3404 information for COFF.
3409 @subsection Symbol Attributes for SOM
3411 @cindex SOM symbol attributes
3412 @cindex symbol attributes, SOM
3414 The SOM format for the HPPA supports a multitude of symbol attributes set with
3415 the @code{.EXPORT} and @code{.IMPORT} directives.
3417 The attributes are described in @cite{HP9000 Series 800 Assembly
3418 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3419 @code{EXPORT} assembler directive documentation.
3423 @chapter Expressions
3427 @cindex numeric values
3428 An @dfn{expression} specifies an address or numeric value.
3429 Whitespace may precede and/or follow an expression.
3431 The result of an expression must be an absolute number, or else an offset into
3432 a particular section. If an expression is not absolute, and there is not
3433 enough information when @command{@value{AS}} sees the expression to know its
3434 section, a second pass over the source program might be necessary to interpret
3435 the expression---but the second pass is currently not implemented.
3436 @command{@value{AS}} aborts with an error message in this situation.
3439 * Empty Exprs:: Empty Expressions
3440 * Integer Exprs:: Integer Expressions
3444 @section Empty Expressions
3446 @cindex empty expressions
3447 @cindex expressions, empty
3448 An empty expression has no value: it is just whitespace or null.
3449 Wherever an absolute expression is required, you may omit the
3450 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3451 is compatible with other assemblers.
3454 @section Integer Expressions
3456 @cindex integer expressions
3457 @cindex expressions, integer
3458 An @dfn{integer expression} is one or more @emph{arguments} delimited
3459 by @emph{operators}.
3462 * Arguments:: Arguments
3463 * Operators:: Operators
3464 * Prefix Ops:: Prefix Operators
3465 * Infix Ops:: Infix Operators
3469 @subsection Arguments
3471 @cindex expression arguments
3472 @cindex arguments in expressions
3473 @cindex operands in expressions
3474 @cindex arithmetic operands
3475 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3476 contexts arguments are sometimes called ``arithmetic operands''. In
3477 this manual, to avoid confusing them with the ``instruction operands'' of
3478 the machine language, we use the term ``argument'' to refer to parts of
3479 expressions only, reserving the word ``operand'' to refer only to machine
3480 instruction operands.
3482 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3483 @var{section} is one of text, data, bss, absolute,
3484 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3487 Numbers are usually integers.
3489 A number can be a flonum or bignum. In this case, you are warned
3490 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3491 these 32 bits are an integer. You may write integer-manipulating
3492 instructions that act on exotic constants, compatible with other
3495 @cindex subexpressions
3496 Subexpressions are a left parenthesis @samp{(} followed by an integer
3497 expression, followed by a right parenthesis @samp{)}; or a prefix
3498 operator followed by an argument.
3501 @subsection Operators
3503 @cindex operators, in expressions
3504 @cindex arithmetic functions
3505 @cindex functions, in expressions
3506 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3507 operators are followed by an argument. Infix operators appear
3508 between their arguments. Operators may be preceded and/or followed by
3512 @subsection Prefix Operator
3514 @cindex prefix operators
3515 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3516 one argument, which must be absolute.
3518 @c the tex/end tex stuff surrounding this small table is meant to make
3519 @c it align, on the printed page, with the similar table in the next
3520 @c section (which is inside an enumerate).
3522 \global\advance\leftskip by \itemindent
3527 @dfn{Negation}. Two's complement negation.
3529 @dfn{Complementation}. Bitwise not.
3533 \global\advance\leftskip by -\itemindent
3537 @subsection Infix Operators
3539 @cindex infix operators
3540 @cindex operators, permitted arguments
3541 @dfn{Infix operators} take two arguments, one on either side. Operators
3542 have precedence, but operations with equal precedence are performed left
3543 to right. Apart from @code{+} or @option{-}, both arguments must be
3544 absolute, and the result is absolute.
3547 @cindex operator precedence
3548 @cindex precedence of operators
3555 @dfn{Multiplication}.
3558 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3565 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3569 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3573 Intermediate precedence
3578 @dfn{Bitwise Inclusive Or}.
3584 @dfn{Bitwise Exclusive Or}.
3587 @dfn{Bitwise Or Not}.
3594 @cindex addition, permitted arguments
3595 @cindex plus, permitted arguments
3596 @cindex arguments for addition
3598 @dfn{Addition}. If either argument is absolute, the result has the section of
3599 the other argument. You may not add together arguments from different
3602 @cindex subtraction, permitted arguments
3603 @cindex minus, permitted arguments
3604 @cindex arguments for subtraction
3606 @dfn{Subtraction}. If the right argument is absolute, the
3607 result has the section of the left argument.
3608 If both arguments are in the same section, the result is absolute.
3609 You may not subtract arguments from different sections.
3610 @c FIXME is there still something useful to say about undefined - undefined ?
3612 @cindex comparison expressions
3613 @cindex expressions, comparison
3617 @dfn{Is Not Equal To}
3621 @dfn{Is Greater Than}
3623 @dfn{Is Greater Than Or Equal To}
3625 @dfn{Is Less Than Or Equal To}
3627 The comparison operators can be used as infix operators. A true results has a
3628 value of -1 whereas a false result has a value of 0. Note, these operators
3629 perform signed comparisons.
3632 @item Lowest Precedence
3641 These two logical operations can be used to combine the results of sub
3642 expressions. Note, unlike the comparison operators a true result returns a
3643 value of 1 but a false results does still return 0. Also note that the logical
3644 or operator has a slightly lower precedence than logical and.
3649 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3650 address; you can only have a defined section in one of the two arguments.
3653 @chapter Assembler Directives
3655 @cindex directives, machine independent
3656 @cindex pseudo-ops, machine independent
3657 @cindex machine independent directives
3658 All assembler directives have names that begin with a period (@samp{.}).
3659 The rest of the name is letters, usually in lower case.
3661 This chapter discusses directives that are available regardless of the
3662 target machine configuration for the @sc{gnu} assembler.
3664 Some machine configurations provide additional directives.
3665 @xref{Machine Dependencies}.
3668 @ifset machine-directives
3669 @xref{Machine Dependencies} for additional directives.
3674 * Abort:: @code{.abort}
3676 * ABORT:: @code{.ABORT}
3679 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3680 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3681 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3682 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3683 * Byte:: @code{.byte @var{expressions}}
3684 * Comm:: @code{.comm @var{symbol} , @var{length} }
3686 * CFI directives:: @code{.cfi_startproc}, @code{.cfi_endproc}, etc.
3688 * Data:: @code{.data @var{subsection}}
3690 * Def:: @code{.def @var{name}}
3693 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3699 * Double:: @code{.double @var{flonums}}
3700 * Eject:: @code{.eject}
3701 * Else:: @code{.else}
3702 * Elseif:: @code{.elseif}
3705 * Endef:: @code{.endef}
3708 * Endfunc:: @code{.endfunc}
3709 * Endif:: @code{.endif}
3710 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3711 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3713 * Exitm:: @code{.exitm}
3714 * Extern:: @code{.extern}
3715 * Fail:: @code{.fail}
3716 @ifclear no-file-dir
3717 * File:: @code{.file @var{string}}
3720 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3721 * Float:: @code{.float @var{flonums}}
3722 * Func:: @code{.func}
3723 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3725 * Hidden:: @code{.hidden @var{names}}
3728 * hword:: @code{.hword @var{expressions}}
3729 * Ident:: @code{.ident}
3730 * If:: @code{.if @var{absolute expression}}
3731 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3732 * Include:: @code{.include "@var{file}"}
3733 * Int:: @code{.int @var{expressions}}
3735 * Internal:: @code{.internal @var{names}}
3738 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3739 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3740 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3741 * Lflags:: @code{.lflags}
3742 @ifclear no-line-dir
3743 * Line:: @code{.line @var{line-number}}
3746 * Ln:: @code{.ln @var{line-number}}
3747 * Linkonce:: @code{.linkonce [@var{type}]}
3748 * List:: @code{.list}
3749 * Long:: @code{.long @var{expressions}}
3751 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3754 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3755 * MRI:: @code{.mri @var{val}}
3756 * Nolist:: @code{.nolist}
3757 * Octa:: @code{.octa @var{bignums}}
3758 * Org:: @code{.org @var{new-lc} , @var{fill}}
3759 * P2align:: @code{.p2align @var{abs-expr} , @var{abs-expr}}
3761 * PopSection:: @code{.popsection}
3762 * Previous:: @code{.previous}
3765 * Print:: @code{.print @var{string}}
3767 * Protected:: @code{.protected @var{names}}
3770 * Psize:: @code{.psize @var{lines}, @var{columns}}
3771 * Purgem:: @code{.purgem @var{name}}
3773 * PushSection:: @code{.pushsection @var{name}}
3776 * Quad:: @code{.quad @var{bignums}}
3777 * Rept:: @code{.rept @var{count}}
3778 * Sbttl:: @code{.sbttl "@var{subheading}"}
3780 * Scl:: @code{.scl @var{class}}
3783 * Section:: @code{.section @var{name}}
3786 * Set:: @code{.set @var{symbol}, @var{expression}}
3787 * Short:: @code{.short @var{expressions}}
3788 * Single:: @code{.single @var{flonums}}
3790 * Size:: @code{.size [@var{name} , @var{expression}]}
3793 * Skip:: @code{.skip @var{size} , @var{fill}}
3794 * Sleb128:: @code{.sleb128 @var{expressions}}
3795 * Space:: @code{.space @var{size} , @var{fill}}
3797 * Stab:: @code{.stabd, .stabn, .stabs}
3800 * String:: @code{.string "@var{str}"}
3801 * Struct:: @code{.struct @var{expression}}
3803 * SubSection:: @code{.subsection}
3804 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3808 * Tag:: @code{.tag @var{structname}}
3811 * Text:: @code{.text @var{subsection}}
3812 * Title:: @code{.title "@var{heading}"}
3814 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3817 * Uleb128:: @code{.uleb128 @var{expressions}}
3819 * Val:: @code{.val @var{addr}}
3823 * Version:: @code{.version "@var{string}"}
3824 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3825 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3826 * Weak:: @code{.weak @var{names}}
3829 * Word:: @code{.word @var{expressions}}
3830 * Deprecated:: Deprecated Directives
3834 @section @code{.abort}
3836 @cindex @code{abort} directive
3837 @cindex stopping the assembly
3838 This directive stops the assembly immediately. It is for
3839 compatibility with other assemblers. The original idea was that the
3840 assembly language source would be piped into the assembler. If the sender
3841 of the source quit, it could use this directive tells @command{@value{AS}} to
3842 quit also. One day @code{.abort} will not be supported.
3846 @section @code{.ABORT}
3848 @cindex @code{ABORT} directive
3849 When producing COFF output, @command{@value{AS}} accepts this directive as a
3850 synonym for @samp{.abort}.
3853 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
3859 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3861 @cindex padding the location counter
3862 @cindex @code{align} directive
3863 Pad the location counter (in the current subsection) to a particular storage
3864 boundary. The first expression (which must be absolute) is the alignment
3865 required, as described below.
3867 The second expression (also absolute) gives the fill value to be stored in the
3868 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3869 padding bytes are normally zero. However, on some systems, if the section is
3870 marked as containing code and the fill value is omitted, the space is filled
3871 with no-op instructions.
3873 The third expression is also absolute, and is also optional. If it is present,
3874 it is the maximum number of bytes that should be skipped by this alignment
3875 directive. If doing the alignment would require skipping more bytes than the
3876 specified maximum, then the alignment is not done at all. You can omit the
3877 fill value (the second argument) entirely by simply using two commas after the
3878 required alignment; this can be useful if you want the alignment to be filled
3879 with no-op instructions when appropriate.
3881 The way the required alignment is specified varies from system to system.
3882 For the a29k, arc, hppa, i386 using ELF, i860, iq2000, m68k, m88k, or32,
3883 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
3884 alignment request in bytes. For example @samp{.align 8} advances
3885 the location counter until it is a multiple of 8. If the location counter
3886 is already a multiple of 8, no change is needed. For the tic54x, the
3887 first expression is the alignment request in words.
3889 For other systems, including the i386 using a.out format, and the arm and
3890 strongarm, it is the
3891 number of low-order zero bits the location counter must have after
3892 advancement. For example @samp{.align 3} advances the location
3893 counter until it a multiple of 8. If the location counter is already a
3894 multiple of 8, no change is needed.
3896 This inconsistency is due to the different behaviors of the various
3897 native assemblers for these systems which GAS must emulate.
3898 GAS also provides @code{.balign} and @code{.p2align} directives,
3899 described later, which have a consistent behavior across all
3900 architectures (but are specific to GAS).
3903 @section @code{.ascii "@var{string}"}@dots{}
3905 @cindex @code{ascii} directive
3906 @cindex string literals
3907 @code{.ascii} expects zero or more string literals (@pxref{Strings})
3908 separated by commas. It assembles each string (with no automatic
3909 trailing zero byte) into consecutive addresses.
3912 @section @code{.asciz "@var{string}"}@dots{}
3914 @cindex @code{asciz} directive
3915 @cindex zero-terminated strings
3916 @cindex null-terminated strings
3917 @code{.asciz} is just like @code{.ascii}, but each string is followed by
3918 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
3921 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3923 @cindex padding the location counter given number of bytes
3924 @cindex @code{balign} directive
3925 Pad the location counter (in the current subsection) to a particular
3926 storage boundary. The first expression (which must be absolute) is the
3927 alignment request in bytes. For example @samp{.balign 8} advances
3928 the location counter until it is a multiple of 8. If the location counter
3929 is already a multiple of 8, no change is needed.
3931 The second expression (also absolute) gives the fill value to be stored in the
3932 padding bytes. It (and the comma) may be omitted. If it is omitted, the
3933 padding bytes are normally zero. However, on some systems, if the section is
3934 marked as containing code and the fill value is omitted, the space is filled
3935 with no-op instructions.
3937 The third expression is also absolute, and is also optional. If it is present,
3938 it is the maximum number of bytes that should be skipped by this alignment
3939 directive. If doing the alignment would require skipping more bytes than the
3940 specified maximum, then the alignment is not done at all. You can omit the
3941 fill value (the second argument) entirely by simply using two commas after the
3942 required alignment; this can be useful if you want the alignment to be filled
3943 with no-op instructions when appropriate.
3945 @cindex @code{balignw} directive
3946 @cindex @code{balignl} directive
3947 The @code{.balignw} and @code{.balignl} directives are variants of the
3948 @code{.balign} directive. The @code{.balignw} directive treats the fill
3949 pattern as a two byte word value. The @code{.balignl} directives treats the
3950 fill pattern as a four byte longword value. For example, @code{.balignw
3951 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
3952 filled in with the value 0x368d (the exact placement of the bytes depends upon
3953 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
3957 @section @code{.byte @var{expressions}}
3959 @cindex @code{byte} directive
3960 @cindex integers, one byte
3961 @code{.byte} expects zero or more expressions, separated by commas.
3962 Each expression is assembled into the next byte.
3965 @section @code{.comm @var{symbol} , @var{length} }
3967 @cindex @code{comm} directive
3968 @cindex symbol, common
3969 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
3970 common symbol in one object file may be merged with a defined or common symbol
3971 of the same name in another object file. If @code{@value{LD}} does not see a
3972 definition for the symbol--just one or more common symbols--then it will
3973 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
3974 absolute expression. If @code{@value{LD}} sees multiple common symbols with
3975 the same name, and they do not all have the same size, it will allocate space
3976 using the largest size.
3979 When using ELF, the @code{.comm} directive takes an optional third argument.
3980 This is the desired alignment of the symbol, specified as a byte boundary (for
3981 example, an alignment of 16 means that the least significant 4 bits of the
3982 address should be zero). The alignment must be an absolute expression, and it
3983 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
3984 for the common symbol, it will use the alignment when placing the symbol. If
3985 no alignment is specified, @command{@value{AS}} will set the alignment to the
3986 largest power of two less than or equal to the size of the symbol, up to a
3991 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
3992 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
3995 @node CFI directives
3996 @section @code{.cfi_startproc}
3997 @cindex @code{cfi_startproc} directive
3998 @code{.cfi_startproc} is used at the beginning of each function that
3999 should have an entry in @code{.eh_frame}. It initializes some internal
4000 data structures and emits architecture dependent initial CFI instructions.
4001 Don't forget to close the function by
4002 @code{.cfi_endproc}.
4004 @section @code{.cfi_endproc}
4005 @cindex @code{cfi_endproc} directive
4006 @code{.cfi_endproc} is used at the end of a function where it closes its
4007 unwind entry previously opened by
4008 @code{.cfi_startproc}. and emits it to @code{.eh_frame}.
4010 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4011 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4012 address from @var{register} and add @var{offset} to it}.
4014 @section @code{.cfi_def_cfa_register @var{register}}
4015 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4016 now on @var{register} will be used instead of the old one. Offset
4019 @section @code{.cfi_def_cfa_offset @var{offset}}
4020 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4021 remains the same, but @var{offset} is new. Note that it is the
4022 absolute offset that will be added to a defined register to compute
4025 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4026 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4027 value that is added/substracted from the previous offset.
4029 @section @code{.cfi_offset @var{register}, @var{offset}}
4030 Previous value of @var{register} is saved at offset @var{offset} from
4033 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4034 Previous value of @var{register} is saved at offset @var{offset} from
4035 the current CFA register. This is transformed to @code{.cfi_offset}
4036 using the known displacement of the CFA register from the CFA.
4037 This is often easier to use, because the number will match the
4038 code it's annotating.
4040 @section @code{.cfi_window_save}
4041 SPARC register window has been saved.
4043 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4044 Allows the user to add arbitrary bytes to the unwind info. One
4045 might use this to add OS-specific CFI opcodes, or generic CFI
4046 opcodes that GAS does not yet support.
4049 @section @code{.data @var{subsection}}
4051 @cindex @code{data} directive
4052 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4053 end of the data subsection numbered @var{subsection} (which is an
4054 absolute expression). If @var{subsection} is omitted, it defaults
4059 @section @code{.def @var{name}}
4061 @cindex @code{def} directive
4062 @cindex COFF symbols, debugging
4063 @cindex debugging COFF symbols
4064 Begin defining debugging information for a symbol @var{name}; the
4065 definition extends until the @code{.endef} directive is encountered.
4068 This directive is only observed when @command{@value{AS}} is configured for COFF
4069 format output; when producing @code{b.out}, @samp{.def} is recognized,
4076 @section @code{.desc @var{symbol}, @var{abs-expression}}
4078 @cindex @code{desc} directive
4079 @cindex COFF symbol descriptor
4080 @cindex symbol descriptor, COFF
4081 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4082 to the low 16 bits of an absolute expression.
4085 The @samp{.desc} directive is not available when @command{@value{AS}} is
4086 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4087 object format. For the sake of compatibility, @command{@value{AS}} accepts
4088 it, but produces no output, when configured for COFF.
4094 @section @code{.dim}
4096 @cindex @code{dim} directive
4097 @cindex COFF auxiliary symbol information
4098 @cindex auxiliary symbol information, COFF
4099 This directive is generated by compilers to include auxiliary debugging
4100 information in the symbol table. It is only permitted inside
4101 @code{.def}/@code{.endef} pairs.
4104 @samp{.dim} is only meaningful when generating COFF format output; when
4105 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4111 @section @code{.double @var{flonums}}
4113 @cindex @code{double} directive
4114 @cindex floating point numbers (double)
4115 @code{.double} expects zero or more flonums, separated by commas. It
4116 assembles floating point numbers.
4118 The exact kind of floating point numbers emitted depends on how
4119 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4123 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4124 in @sc{ieee} format.
4129 @section @code{.eject}
4131 @cindex @code{eject} directive
4132 @cindex new page, in listings
4133 @cindex page, in listings
4134 @cindex listing control: new page
4135 Force a page break at this point, when generating assembly listings.
4138 @section @code{.else}
4140 @cindex @code{else} directive
4141 @code{.else} is part of the @command{@value{AS}} support for conditional
4142 assembly; @pxref{If,,@code{.if}}. It marks the beginning of a section
4143 of code to be assembled if the condition for the preceding @code{.if}
4147 @section @code{.elseif}
4149 @cindex @code{elseif} directive
4150 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4151 assembly; @pxref{If,,@code{.if}}. It is shorthand for beginning a new
4152 @code{.if} block that would otherwise fill the entire @code{.else} section.
4155 @section @code{.end}
4157 @cindex @code{end} directive
4158 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4159 process anything in the file past the @code{.end} directive.
4163 @section @code{.endef}
4165 @cindex @code{endef} directive
4166 This directive flags the end of a symbol definition begun with
4170 @samp{.endef} is only meaningful when generating COFF format output; if
4171 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4172 directive but ignores it.
4177 @section @code{.endfunc}
4178 @cindex @code{endfunc} directive
4179 @code{.endfunc} marks the end of a function specified with @code{.func}.
4182 @section @code{.endif}
4184 @cindex @code{endif} directive
4185 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4186 it marks the end of a block of code that is only assembled
4187 conditionally. @xref{If,,@code{.if}}.
4190 @section @code{.equ @var{symbol}, @var{expression}}
4192 @cindex @code{equ} directive
4193 @cindex assigning values to symbols
4194 @cindex symbols, assigning values to
4195 This directive sets the value of @var{symbol} to @var{expression}.
4196 It is synonymous with @samp{.set}; @pxref{Set,,@code{.set}}.
4199 The syntax for @code{equ} on the HPPA is
4200 @samp{@var{symbol} .equ @var{expression}}.
4204 @section @code{.equiv @var{symbol}, @var{expression}}
4205 @cindex @code{equiv} directive
4206 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4207 the assembler will signal an error if @var{symbol} is already defined. Note a
4208 symbol which has been referenced but not actually defined is considered to be
4211 Except for the contents of the error message, this is roughly equivalent to
4220 @section @code{.err}
4221 @cindex @code{err} directive
4222 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4223 message and, unless the @option{-Z} option was used, it will not generate an
4224 object file. This can be used to signal error an conditionally compiled code.
4227 @section @code{.exitm}
4228 Exit early from the current macro definition. @xref{Macro}.
4231 @section @code{.extern}
4233 @cindex @code{extern} directive
4234 @code{.extern} is accepted in the source program---for compatibility
4235 with other assemblers---but it is ignored. @command{@value{AS}} treats
4236 all undefined symbols as external.
4239 @section @code{.fail @var{expression}}
4241 @cindex @code{fail} directive
4242 Generates an error or a warning. If the value of the @var{expression} is 500
4243 or more, @command{@value{AS}} will print a warning message. If the value is less
4244 than 500, @command{@value{AS}} will print an error message. The message will
4245 include the value of @var{expression}. This can occasionally be useful inside
4246 complex nested macros or conditional assembly.
4248 @ifclear no-file-dir
4250 @section @code{.file @var{string}}
4252 @cindex @code{file} directive
4253 @cindex logical file name
4254 @cindex file name, logical
4255 @code{.file} tells @command{@value{AS}} that we are about to start a new logical
4256 file. @var{string} is the new file name. In general, the filename is
4257 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4258 to specify an empty file name, you must give the quotes--@code{""}. This
4259 statement may go away in future: it is only recognized to be compatible with
4260 old @command{@value{AS}} programs.
4262 In some configurations of @command{@value{AS}}, @code{.file} has already been
4263 removed to avoid conflicts with other assemblers. @xref{Machine Dependencies}.
4268 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4270 @cindex @code{fill} directive
4271 @cindex writing patterns in memory
4272 @cindex patterns, writing in memory
4273 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4274 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4275 may be zero or more. @var{Size} may be zero or more, but if it is
4276 more than 8, then it is deemed to have the value 8, compatible with
4277 other people's assemblers. The contents of each @var{repeat} bytes
4278 is taken from an 8-byte number. The highest order 4 bytes are
4279 zero. The lowest order 4 bytes are @var{value} rendered in the
4280 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4281 Each @var{size} bytes in a repetition is taken from the lowest order
4282 @var{size} bytes of this number. Again, this bizarre behavior is
4283 compatible with other people's assemblers.
4285 @var{size} and @var{value} are optional.
4286 If the second comma and @var{value} are absent, @var{value} is
4287 assumed zero. If the first comma and following tokens are absent,
4288 @var{size} is assumed to be 1.
4291 @section @code{.float @var{flonums}}
4293 @cindex floating point numbers (single)
4294 @cindex @code{float} directive
4295 This directive assembles zero or more flonums, separated by commas. It
4296 has the same effect as @code{.single}.
4298 The exact kind of floating point numbers emitted depends on how
4299 @command{@value{AS}} is configured.
4300 @xref{Machine Dependencies}.
4304 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4305 in @sc{ieee} format.
4310 @section @code{.func @var{name}[,@var{label}]}
4311 @cindex @code{func} directive
4312 @code{.func} emits debugging information to denote function @var{name}, and
4313 is ignored unless the file is assembled with debugging enabled.
4314 Only @samp{--gstabs[+]} is currently supported.
4315 @var{label} is the entry point of the function and if omitted @var{name}
4316 prepended with the @samp{leading char} is used.
4317 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4318 All functions are currently defined to have @code{void} return type.
4319 The function must be terminated with @code{.endfunc}.
4322 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4324 @cindex @code{global} directive
4325 @cindex symbol, making visible to linker
4326 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4327 @var{symbol} in your partial program, its value is made available to
4328 other partial programs that are linked with it. Otherwise,
4329 @var{symbol} takes its attributes from a symbol of the same name
4330 from another file linked into the same program.
4332 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4333 compatibility with other assemblers.
4336 On the HPPA, @code{.global} is not always enough to make it accessible to other
4337 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4338 @xref{HPPA Directives,, HPPA Assembler Directives}.
4343 @section @code{.hidden @var{names}}
4345 @cindex @code{hidden} directive
4347 This is one of the ELF visibility directives. The other two are
4348 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4349 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4351 This directive overrides the named symbols default visibility (which is set by
4352 their binding: local, global or weak). The directive sets the visibility to
4353 @code{hidden} which means that the symbols are not visible to other components.
4354 Such symbols are always considered to be @code{protected} as well.
4358 @section @code{.hword @var{expressions}}
4360 @cindex @code{hword} directive
4361 @cindex integers, 16-bit
4362 @cindex numbers, 16-bit
4363 @cindex sixteen bit integers
4364 This expects zero or more @var{expressions}, and emits
4365 a 16 bit number for each.
4368 This directive is a synonym for @samp{.short}; depending on the target
4369 architecture, it may also be a synonym for @samp{.word}.
4373 This directive is a synonym for @samp{.short}.
4376 This directive is a synonym for both @samp{.short} and @samp{.word}.
4381 @section @code{.ident}
4383 @cindex @code{ident} directive
4384 This directive is used by some assemblers to place tags in object files.
4385 @command{@value{AS}} simply accepts the directive for source-file
4386 compatibility with such assemblers, but does not actually emit anything
4390 @section @code{.if @var{absolute expression}}
4392 @cindex conditional assembly
4393 @cindex @code{if} directive
4394 @code{.if} marks the beginning of a section of code which is only
4395 considered part of the source program being assembled if the argument
4396 (which must be an @var{absolute expression}) is non-zero. The end of
4397 the conditional section of code must be marked by @code{.endif}
4398 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4399 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4400 If you have several conditions to check, @code{.elseif} may be used to avoid
4401 nesting blocks if/else within each subsequent @code{.else} block.
4403 The following variants of @code{.if} are also supported:
4405 @cindex @code{ifdef} directive
4406 @item .ifdef @var{symbol}
4407 Assembles the following section of code if the specified @var{symbol}
4408 has been defined. Note a symbol which has been referenced but not yet defined
4409 is considered to be undefined.
4411 @cindex @code{ifc} directive
4412 @item .ifc @var{string1},@var{string2}
4413 Assembles the following section of code if the two strings are the same. The
4414 strings may be optionally quoted with single quotes. If they are not quoted,
4415 the first string stops at the first comma, and the second string stops at the
4416 end of the line. Strings which contain whitespace should be quoted. The
4417 string comparison is case sensitive.
4419 @cindex @code{ifeq} directive
4420 @item .ifeq @var{absolute expression}
4421 Assembles the following section of code if the argument is zero.
4423 @cindex @code{ifeqs} directive
4424 @item .ifeqs @var{string1},@var{string2}
4425 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4427 @cindex @code{ifge} directive
4428 @item .ifge @var{absolute expression}
4429 Assembles the following section of code if the argument is greater than or
4432 @cindex @code{ifgt} directive
4433 @item .ifgt @var{absolute expression}
4434 Assembles the following section of code if the argument is greater than zero.
4436 @cindex @code{ifle} directive
4437 @item .ifle @var{absolute expression}
4438 Assembles the following section of code if the argument is less than or equal
4441 @cindex @code{iflt} directive
4442 @item .iflt @var{absolute expression}
4443 Assembles the following section of code if the argument is less than zero.
4445 @cindex @code{ifnc} directive
4446 @item .ifnc @var{string1},@var{string2}.
4447 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4448 following section of code if the two strings are not the same.
4450 @cindex @code{ifndef} directive
4451 @cindex @code{ifnotdef} directive
4452 @item .ifndef @var{symbol}
4453 @itemx .ifnotdef @var{symbol}
4454 Assembles the following section of code if the specified @var{symbol}
4455 has not been defined. Both spelling variants are equivalent. Note a symbol
4456 which has been referenced but not yet defined is considered to be undefined.
4458 @cindex @code{ifne} directive
4459 @item .ifne @var{absolute expression}
4460 Assembles the following section of code if the argument is not equal to zero
4461 (in other words, this is equivalent to @code{.if}).
4463 @cindex @code{ifnes} directive
4464 @item .ifnes @var{string1},@var{string2}
4465 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4466 following section of code if the two strings are not the same.
4470 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4472 @cindex @code{incbin} directive
4473 @cindex binary files, including
4474 The @code{incbin} directive includes @var{file} verbatim at the current
4475 location. You can control the search paths used with the @samp{-I} command-line
4476 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4479 The @var{skip} argument skips a number of bytes from the start of the
4480 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4481 read. Note that the data is not aligned in any way, so it is the user's
4482 responsibility to make sure that proper alignment is provided both before and
4483 after the @code{incbin} directive.
4486 @section @code{.include "@var{file}"}
4488 @cindex @code{include} directive
4489 @cindex supporting files, including
4490 @cindex files, including
4491 This directive provides a way to include supporting files at specified
4492 points in your source program. The code from @var{file} is assembled as
4493 if it followed the point of the @code{.include}; when the end of the
4494 included file is reached, assembly of the original file continues. You
4495 can control the search paths used with the @samp{-I} command-line option
4496 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4500 @section @code{.int @var{expressions}}
4502 @cindex @code{int} directive
4503 @cindex integers, 32-bit
4504 Expect zero or more @var{expressions}, of any section, separated by commas.
4505 For each expression, emit a number that, at run time, is the value of that
4506 expression. The byte order and bit size of the number depends on what kind
4507 of target the assembly is for.
4511 On the H8/500 and most forms of the H8/300, @code{.int} emits 16-bit
4512 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4519 @section @code{.internal @var{names}}
4521 @cindex @code{internal} directive
4523 This is one of the ELF visibility directives. The other two are
4524 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4525 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4527 This directive overrides the named symbols default visibility (which is set by
4528 their binding: local, global or weak). The directive sets the visibility to
4529 @code{internal} which means that the symbols are considered to be @code{hidden}
4530 (i.e., not visible to other components), and that some extra, processor specific
4531 processing must also be performed upon the symbols as well.
4535 @section @code{.irp @var{symbol},@var{values}}@dots{}
4537 @cindex @code{irp} directive
4538 Evaluate a sequence of statements assigning different values to @var{symbol}.
4539 The sequence of statements starts at the @code{.irp} directive, and is
4540 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4541 set to @var{value}, and the sequence of statements is assembled. If no
4542 @var{value} is listed, the sequence of statements is assembled once, with
4543 @var{symbol} set to the null string. To refer to @var{symbol} within the
4544 sequence of statements, use @var{\symbol}.
4546 For example, assembling
4554 is equivalent to assembling
4563 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4565 @cindex @code{irpc} directive
4566 Evaluate a sequence of statements assigning different values to @var{symbol}.
4567 The sequence of statements starts at the @code{.irpc} directive, and is
4568 terminated by an @code{.endr} directive. For each character in @var{value},
4569 @var{symbol} is set to the character, and the sequence of statements is
4570 assembled. If no @var{value} is listed, the sequence of statements is
4571 assembled once, with @var{symbol} set to the null string. To refer to
4572 @var{symbol} within the sequence of statements, use @var{\symbol}.
4574 For example, assembling
4582 is equivalent to assembling
4591 @section @code{.lcomm @var{symbol} , @var{length}}
4593 @cindex @code{lcomm} directive
4594 @cindex local common symbols
4595 @cindex symbols, local common
4596 Reserve @var{length} (an absolute expression) bytes for a local common
4597 denoted by @var{symbol}. The section and value of @var{symbol} are
4598 those of the new local common. The addresses are allocated in the bss
4599 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4600 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4601 not visible to @code{@value{LD}}.
4604 Some targets permit a third argument to be used with @code{.lcomm}. This
4605 argument specifies the desired alignment of the symbol in the bss section.
4609 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4610 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4614 @section @code{.lflags}
4616 @cindex @code{lflags} directive (ignored)
4617 @command{@value{AS}} accepts this directive, for compatibility with other
4618 assemblers, but ignores it.
4620 @ifclear no-line-dir
4622 @section @code{.line @var{line-number}}
4624 @cindex @code{line} directive
4628 @section @code{.ln @var{line-number}}
4630 @cindex @code{ln} directive
4632 @cindex logical line number
4634 Change the logical line number. @var{line-number} must be an absolute
4635 expression. The next line has that logical line number. Therefore any other
4636 statements on the current line (after a statement separator character) are
4637 reported as on logical line number @var{line-number} @minus{} 1. One day
4638 @command{@value{AS}} will no longer support this directive: it is recognized only
4639 for compatibility with existing assembler programs.
4643 @emph{Warning:} In the AMD29K configuration of @value{AS}, this command is
4644 not available; use the synonym @code{.ln} in that context.
4649 @ifclear no-line-dir
4650 Even though this is a directive associated with the @code{a.out} or
4651 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
4652 when producing COFF output, and treats @samp{.line} as though it
4653 were the COFF @samp{.ln} @emph{if} it is found outside a
4654 @code{.def}/@code{.endef} pair.
4656 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4657 used by compilers to generate auxiliary symbol information for
4662 @section @code{.linkonce [@var{type}]}
4664 @cindex @code{linkonce} directive
4665 @cindex common sections
4666 Mark the current section so that the linker only includes a single copy of it.
4667 This may be used to include the same section in several different object files,
4668 but ensure that the linker will only include it once in the final output file.
4669 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4670 Duplicate sections are detected based on the section name, so it should be
4673 This directive is only supported by a few object file formats; as of this
4674 writing, the only object file format which supports it is the Portable
4675 Executable format used on Windows NT.
4677 The @var{type} argument is optional. If specified, it must be one of the
4678 following strings. For example:
4682 Not all types may be supported on all object file formats.
4686 Silently discard duplicate sections. This is the default.
4689 Warn if there are duplicate sections, but still keep only one copy.
4692 Warn if any of the duplicates have different sizes.
4695 Warn if any of the duplicates do not have exactly the same contents.
4699 @section @code{.ln @var{line-number}}
4701 @cindex @code{ln} directive
4702 @ifclear no-line-dir
4703 @samp{.ln} is a synonym for @samp{.line}.
4706 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
4707 must be an absolute expression. The next line has that logical
4708 line number, so any other statements on the current line (after a
4709 statement separator character @code{;}) are reported as on logical
4710 line number @var{line-number} @minus{} 1.
4713 This directive is accepted, but ignored, when @command{@value{AS}} is
4714 configured for @code{b.out}; its effect is only associated with COFF
4720 @section @code{.mri @var{val}}
4722 @cindex @code{mri} directive
4723 @cindex MRI mode, temporarily
4724 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
4725 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
4726 affects code assembled until the next @code{.mri} directive, or until the end
4727 of the file. @xref{M, MRI mode, MRI mode}.
4730 @section @code{.list}
4732 @cindex @code{list} directive
4733 @cindex listing control, turning on
4734 Control (in conjunction with the @code{.nolist} directive) whether or
4735 not assembly listings are generated. These two directives maintain an
4736 internal counter (which is zero initially). @code{.list} increments the
4737 counter, and @code{.nolist} decrements it. Assembly listings are
4738 generated whenever the counter is greater than zero.
4740 By default, listings are disabled. When you enable them (with the
4741 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
4742 the initial value of the listing counter is one.
4745 @section @code{.long @var{expressions}}
4747 @cindex @code{long} directive
4748 @code{.long} is the same as @samp{.int}, @pxref{Int,,@code{.int}}.
4751 @c no one seems to know what this is for or whether this description is
4752 @c what it really ought to do
4754 @section @code{.lsym @var{symbol}, @var{expression}}
4756 @cindex @code{lsym} directive
4757 @cindex symbol, not referenced in assembly
4758 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
4759 the hash table, ensuring it cannot be referenced by name during the
4760 rest of the assembly. This sets the attributes of the symbol to be
4761 the same as the expression value:
4763 @var{other} = @var{descriptor} = 0
4764 @var{type} = @r{(section of @var{expression})}
4765 @var{value} = @var{expression}
4768 The new symbol is not flagged as external.
4772 @section @code{.macro}
4775 The commands @code{.macro} and @code{.endm} allow you to define macros that
4776 generate assembly output. For example, this definition specifies a macro
4777 @code{sum} that puts a sequence of numbers into memory:
4780 .macro sum from=0, to=5
4789 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
4801 @item .macro @var{macname}
4802 @itemx .macro @var{macname} @var{macargs} @dots{}
4803 @cindex @code{macro} directive
4804 Begin the definition of a macro called @var{macname}. If your macro
4805 definition requires arguments, specify their names after the macro name,
4806 separated by commas or spaces. You can supply a default value for any
4807 macro argument by following the name with @samp{=@var{deflt}}. For
4808 example, these are all valid @code{.macro} statements:
4812 Begin the definition of a macro called @code{comm}, which takes no
4815 @item .macro plus1 p, p1
4816 @itemx .macro plus1 p p1
4817 Either statement begins the definition of a macro called @code{plus1},
4818 which takes two arguments; within the macro definition, write
4819 @samp{\p} or @samp{\p1} to evaluate the arguments.
4821 @item .macro reserve_str p1=0 p2
4822 Begin the definition of a macro called @code{reserve_str}, with two
4823 arguments. The first argument has a default value, but not the second.
4824 After the definition is complete, you can call the macro either as
4825 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
4826 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
4827 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
4828 @samp{0}, and @samp{\p2} evaluating to @var{b}).
4831 When you call a macro, you can specify the argument values either by
4832 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
4833 @samp{sum to=17, from=9}.
4836 @cindex @code{endm} directive
4837 Mark the end of a macro definition.
4840 @cindex @code{exitm} directive
4841 Exit early from the current macro definition.
4843 @cindex number of macros executed
4844 @cindex macros, count executed
4846 @command{@value{AS}} maintains a counter of how many macros it has
4847 executed in this pseudo-variable; you can copy that number to your
4848 output with @samp{\@@}, but @emph{only within a macro definition}.
4851 @item LOCAL @var{name} [ , @dots{} ]
4852 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
4853 macro syntax'' with @samp{-a} or @samp{--alternate}.} @xref{Alternate,,
4854 Alternate macro syntax}.
4856 Generate a string replacement for each of the @var{name} arguments, and
4857 replace any instances of @var{name} in each macro expansion. The
4858 replacement string is unique in the assembly, and different for each
4859 separate macro expansion. @code{LOCAL} allows you to write macros that
4860 define symbols, without fear of conflict between separate macro expansions.
4865 @section @code{.nolist}
4867 @cindex @code{nolist} directive
4868 @cindex listing control, turning off
4869 Control (in conjunction with the @code{.list} directive) whether or
4870 not assembly listings are generated. These two directives maintain an
4871 internal counter (which is zero initially). @code{.list} increments the
4872 counter, and @code{.nolist} decrements it. Assembly listings are
4873 generated whenever the counter is greater than zero.
4876 @section @code{.octa @var{bignums}}
4878 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
4879 @cindex @code{octa} directive
4880 @cindex integer, 16-byte
4881 @cindex sixteen byte integer
4882 This directive expects zero or more bignums, separated by commas. For each
4883 bignum, it emits a 16-byte integer.
4885 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
4886 hence @emph{octa}-word for 16 bytes.
4889 @section @code{.org @var{new-lc} , @var{fill}}
4891 @cindex @code{org} directive
4892 @cindex location counter, advancing
4893 @cindex advancing location counter
4894 @cindex current address, advancing
4895 Advance the location counter of the current section to
4896 @var{new-lc}. @var{new-lc} is either an absolute expression or an
4897 expression with the same section as the current subsection. That is,
4898 you can't use @code{.org} to cross sections: if @var{new-lc} has the
4899 wrong section, the @code{.org} directive is ignored. To be compatible
4900 with former assemblers, if the section of @var{new-lc} is absolute,
4901 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
4902 is the same as the current subsection.
4904 @code{.org} may only increase the location counter, or leave it
4905 unchanged; you cannot use @code{.org} to move the location counter
4908 @c double negative used below "not undefined" because this is a specific
4909 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
4910 @c section. doc@cygnus.com 18feb91
4911 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
4912 may not be undefined. If you really detest this restriction we eagerly await
4913 a chance to share your improved assembler.
4915 Beware that the origin is relative to the start of the section, not
4916 to the start of the subsection. This is compatible with other
4917 people's assemblers.
4919 When the location counter (of the current subsection) is advanced, the
4920 intervening bytes are filled with @var{fill} which should be an
4921 absolute expression. If the comma and @var{fill} are omitted,
4922 @var{fill} defaults to zero.
4925 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4927 @cindex padding the location counter given a power of two
4928 @cindex @code{p2align} directive
4929 Pad the location counter (in the current subsection) to a particular
4930 storage boundary. The first expression (which must be absolute) is the
4931 number of low-order zero bits the location counter must have after
4932 advancement. For example @samp{.p2align 3} advances the location
4933 counter until it a multiple of 8. If the location counter is already a
4934 multiple of 8, no change is needed.
4936 The second expression (also absolute) gives the fill value to be stored in the
4937 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4938 padding bytes are normally zero. However, on some systems, if the section is
4939 marked as containing code and the fill value is omitted, the space is filled
4940 with no-op instructions.
4942 The third expression is also absolute, and is also optional. If it is present,
4943 it is the maximum number of bytes that should be skipped by this alignment
4944 directive. If doing the alignment would require skipping more bytes than the
4945 specified maximum, then the alignment is not done at all. You can omit the
4946 fill value (the second argument) entirely by simply using two commas after the
4947 required alignment; this can be useful if you want the alignment to be filled
4948 with no-op instructions when appropriate.
4950 @cindex @code{p2alignw} directive
4951 @cindex @code{p2alignl} directive
4952 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
4953 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
4954 pattern as a two byte word value. The @code{.p2alignl} directives treats the
4955 fill pattern as a four byte longword value. For example, @code{.p2alignw
4956 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4957 filled in with the value 0x368d (the exact placement of the bytes depends upon
4958 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4963 @section @code{.previous}
4965 @cindex @code{previous} directive
4966 @cindex Section Stack
4967 This is one of the ELF section stack manipulation directives. The others are
4968 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
4969 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
4970 (@pxref{PopSection}).
4972 This directive swaps the current section (and subsection) with most recently
4973 referenced section (and subsection) prior to this one. Multiple
4974 @code{.previous} directives in a row will flip between two sections (and their
4977 In terms of the section stack, this directive swaps the current section with
4978 the top section on the section stack.
4983 @section @code{.popsection}
4985 @cindex @code{popsection} directive
4986 @cindex Section Stack
4987 This is one of the ELF section stack manipulation directives. The others are
4988 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
4989 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
4992 This directive replaces the current section (and subsection) with the top
4993 section (and subsection) on the section stack. This section is popped off the
4998 @section @code{.print @var{string}}
5000 @cindex @code{print} directive
5001 @command{@value{AS}} will print @var{string} on the standard output during
5002 assembly. You must put @var{string} in double quotes.
5006 @section @code{.protected @var{names}}
5008 @cindex @code{protected} directive
5010 This is one of the ELF visibility directives. The other two are
5011 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5013 This directive overrides the named symbols default visibility (which is set by
5014 their binding: local, global or weak). The directive sets the visibility to
5015 @code{protected} which means that any references to the symbols from within the
5016 components that defines them must be resolved to the definition in that
5017 component, even if a definition in another component would normally preempt
5022 @section @code{.psize @var{lines} , @var{columns}}
5024 @cindex @code{psize} directive
5025 @cindex listing control: paper size
5026 @cindex paper size, for listings
5027 Use this directive to declare the number of lines---and, optionally, the
5028 number of columns---to use for each page, when generating listings.
5030 If you do not use @code{.psize}, listings use a default line-count
5031 of 60. You may omit the comma and @var{columns} specification; the
5032 default width is 200 columns.
5034 @command{@value{AS}} generates formfeeds whenever the specified number of
5035 lines is exceeded (or whenever you explicitly request one, using
5038 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5039 those explicitly specified with @code{.eject}.
5042 @section @code{.purgem @var{name}}
5044 @cindex @code{purgem} directive
5045 Undefine the macro @var{name}, so that later uses of the string will not be
5046 expanded. @xref{Macro}.
5050 @section @code{.pushsection @var{name} , @var{subsection}}
5052 @cindex @code{pushsection} directive
5053 @cindex Section Stack
5054 This is one of the ELF section stack manipulation directives. The others are
5055 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5056 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5059 This directive is a synonym for @code{.section}. It pushes the current section
5060 (and subsection) onto the top of the section stack, and then replaces the
5061 current section and subsection with @code{name} and @code{subsection}.
5065 @section @code{.quad @var{bignums}}
5067 @cindex @code{quad} directive
5068 @code{.quad} expects zero or more bignums, separated by commas. For
5069 each bignum, it emits
5071 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5072 warning message; and just takes the lowest order 8 bytes of the bignum.
5073 @cindex eight-byte integer
5074 @cindex integer, 8-byte
5076 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5077 hence @emph{quad}-word for 8 bytes.
5080 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5081 warning message; and just takes the lowest order 16 bytes of the bignum.
5082 @cindex sixteen-byte integer
5083 @cindex integer, 16-byte
5087 @section @code{.rept @var{count}}
5089 @cindex @code{rept} directive
5090 Repeat the sequence of lines between the @code{.rept} directive and the next
5091 @code{.endr} directive @var{count} times.
5093 For example, assembling
5101 is equivalent to assembling
5110 @section @code{.sbttl "@var{subheading}"}
5112 @cindex @code{sbttl} directive
5113 @cindex subtitles for listings
5114 @cindex listing control: subtitle
5115 Use @var{subheading} as the title (third line, immediately after the
5116 title line) when generating assembly listings.
5118 This directive affects subsequent pages, as well as the current page if
5119 it appears within ten lines of the top of a page.
5123 @section @code{.scl @var{class}}
5125 @cindex @code{scl} directive
5126 @cindex symbol storage class (COFF)
5127 @cindex COFF symbol storage class
5128 Set the storage-class value for a symbol. This directive may only be
5129 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5130 whether a symbol is static or external, or it may record further
5131 symbolic debugging information.
5134 The @samp{.scl} directive is primarily associated with COFF output; when
5135 configured to generate @code{b.out} output format, @command{@value{AS}}
5136 accepts this directive but ignores it.
5142 @section @code{.section @var{name}}
5144 @cindex named section
5145 Use the @code{.section} directive to assemble the following code into a section
5148 This directive is only supported for targets that actually support arbitrarily
5149 named sections; on @code{a.out} targets, for example, it is not accepted, even
5150 with a standard @code{a.out} section name.
5154 @c only print the extra heading if both COFF and ELF are set
5155 @subheading COFF Version
5158 @cindex @code{section} directive (COFF version)
5159 For COFF targets, the @code{.section} directive is used in one of the following
5163 .section @var{name}[, "@var{flags}"]
5164 .section @var{name}[, @var{subsegment}]
5167 If the optional argument is quoted, it is taken as flags to use for the
5168 section. Each flag is a single character. The following flags are recognized:
5171 bss section (uninitialized data)
5173 section is not loaded
5183 shared section (meaningful for PE targets)
5185 ignored. (For compatibility with the ELF version)
5188 If no flags are specified, the default flags depend upon the section name. If
5189 the section name is not recognized, the default will be for the section to be
5190 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5191 from the section, rather than adding them, so if they are used on their own it
5192 will be as if no flags had been specified at all.
5194 If the optional argument to the @code{.section} directive is not quoted, it is
5195 taken as a subsegment number (@pxref{Sub-Sections}).
5200 @c only print the extra heading if both COFF and ELF are set
5201 @subheading ELF Version
5204 @cindex Section Stack
5205 This is one of the ELF section stack manipulation directives. The others are
5206 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5207 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5208 @code{.previous} (@pxref{Previous}).
5210 @cindex @code{section} directive (ELF version)
5211 For ELF targets, the @code{.section} directive is used like this:
5214 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]
5217 The optional @var{flags} argument is a quoted string which may contain any
5218 combination of the following characters:
5221 section is allocatable
5225 section is executable
5227 section is mergeable
5229 section contains zero terminated strings
5231 section is a member of a section group
5233 section is used for thread-local-storage
5236 The optional @var{type} argument may contain one of the following constants:
5239 section contains data
5241 section does not contain data (i.e., section only occupies space)
5243 section contains data which is used by things other than the program
5246 Note on targets where the @code{@@} character is the start of a comment (eg
5247 ARM) then another character is used instead. For example the ARM port uses the
5250 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5251 be specified as well as an extra argument - @var{entsize} - like this:
5254 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5257 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5258 constants, each @var{entsize} octets long. Sections with both @code{M} and
5259 @code{S} must contain zero terminated strings where each character is
5260 @var{entsize} bytes long. The linker may remove duplicates within sections with
5261 the same name, same entity size and same flags. @var{entsize} must be an
5262 absolute expression.
5264 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5265 be present along with an additional field like this:
5268 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5271 The @var{GroupName} field specifies the name of the section group to which this
5272 particular section belongs. The optional linkage field can contain:
5275 indicates that only one copy of this section should be retained
5280 Note - if both the @var{M} and @var{G} flags are present then the fields for
5281 the Merge flag should come first, like this:
5284 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5287 If no flags are specified, the default flags depend upon the section name. If
5288 the section name is not recognized, the default will be for the section to have
5289 none of the above flags: it will not be allocated in memory, nor writable, nor
5290 executable. The section will contain data.
5292 For ELF targets, the assembler supports another type of @code{.section}
5293 directive for compatibility with the Solaris assembler:
5296 .section "@var{name}"[, @var{flags}...]
5299 Note that the section name is quoted. There may be a sequence of comma
5303 section is allocatable
5307 section is executable
5309 section is used for thread local storage
5312 This directive replaces the current section and subsection. The replaced
5313 section and subsection are pushed onto the section stack. See the contents of
5314 the gas testsuite directory @code{gas/testsuite/gas/elf} for some examples of
5315 how this directive and the other section stack directives work.
5320 @section @code{.set @var{symbol}, @var{expression}}
5322 @cindex @code{set} directive
5323 @cindex symbol value, setting
5324 Set the value of @var{symbol} to @var{expression}. This
5325 changes @var{symbol}'s value and type to conform to
5326 @var{expression}. If @var{symbol} was flagged as external, it remains
5327 flagged (@pxref{Symbol Attributes}).
5329 You may @code{.set} a symbol many times in the same assembly.
5331 If you @code{.set} a global symbol, the value stored in the object
5332 file is the last value stored into it.
5335 The syntax for @code{set} on the HPPA is
5336 @samp{@var{symbol} .set @var{expression}}.
5340 @section @code{.short @var{expressions}}
5342 @cindex @code{short} directive
5344 @code{.short} is normally the same as @samp{.word}.
5345 @xref{Word,,@code{.word}}.
5347 In some configurations, however, @code{.short} and @code{.word} generate
5348 numbers of different lengths; @pxref{Machine Dependencies}.
5352 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5355 This expects zero or more @var{expressions}, and emits
5356 a 16 bit number for each.
5361 @section @code{.single @var{flonums}}
5363 @cindex @code{single} directive
5364 @cindex floating point numbers (single)
5365 This directive assembles zero or more flonums, separated by commas. It
5366 has the same effect as @code{.float}.
5368 The exact kind of floating point numbers emitted depends on how
5369 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5373 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5374 numbers in @sc{ieee} format.
5380 @section @code{.size}
5382 This directive is used to set the size associated with a symbol.
5386 @c only print the extra heading if both COFF and ELF are set
5387 @subheading COFF Version
5390 @cindex @code{size} directive (COFF version)
5391 For COFF targets, the @code{.size} directive is only permitted inside
5392 @code{.def}/@code{.endef} pairs. It is used like this:
5395 .size @var{expression}
5399 @samp{.size} is only meaningful when generating COFF format output; when
5400 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5407 @c only print the extra heading if both COFF and ELF are set
5408 @subheading ELF Version
5411 @cindex @code{size} directive (ELF version)
5412 For ELF targets, the @code{.size} directive is used like this:
5415 .size @var{name} , @var{expression}
5418 This directive sets the size associated with a symbol @var{name}.
5419 The size in bytes is computed from @var{expression} which can make use of label
5420 arithmetic. This directive is typically used to set the size of function
5426 @section @code{.sleb128 @var{expressions}}
5428 @cindex @code{sleb128} directive
5429 @var{sleb128} stands for ``signed little endian base 128.'' This is a
5430 compact, variable length representation of numbers used by the DWARF
5431 symbolic debugging format. @xref{Uleb128,@code{.uleb128}}.
5433 @ifclear no-space-dir
5435 @section @code{.skip @var{size} , @var{fill}}
5437 @cindex @code{skip} directive
5438 @cindex filling memory
5439 This directive emits @var{size} bytes, each of value @var{fill}. Both
5440 @var{size} and @var{fill} are absolute expressions. If the comma and
5441 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
5445 @section @code{.space @var{size} , @var{fill}}
5447 @cindex @code{space} directive
5448 @cindex filling memory
5449 This directive emits @var{size} bytes, each of value @var{fill}. Both
5450 @var{size} and @var{fill} are absolute expressions. If the comma
5451 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
5456 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
5457 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
5458 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
5459 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
5468 @section @code{.space}
5469 @cindex @code{space} directive
5471 On the AMD 29K, this directive is ignored; it is accepted for
5472 compatibility with other AMD 29K assemblers.
5475 @emph{Warning:} In most versions of the @sc{gnu} assembler, the directive
5476 @code{.space} has the effect of @code{.block} @xref{Machine Dependencies}.
5482 @section @code{.stabd, .stabn, .stabs}
5484 @cindex symbolic debuggers, information for
5485 @cindex @code{stab@var{x}} directives
5486 There are three directives that begin @samp{.stab}.
5487 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
5488 The symbols are not entered in the @command{@value{AS}} hash table: they
5489 cannot be referenced elsewhere in the source file.
5490 Up to five fields are required:
5494 This is the symbol's name. It may contain any character except
5495 @samp{\000}, so is more general than ordinary symbol names. Some
5496 debuggers used to code arbitrarily complex structures into symbol names
5500 An absolute expression. The symbol's type is set to the low 8 bits of
5501 this expression. Any bit pattern is permitted, but @code{@value{LD}}
5502 and debuggers choke on silly bit patterns.
5505 An absolute expression. The symbol's ``other'' attribute is set to the
5506 low 8 bits of this expression.
5509 An absolute expression. The symbol's descriptor is set to the low 16
5510 bits of this expression.
5513 An absolute expression which becomes the symbol's value.
5516 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
5517 or @code{.stabs} statement, the symbol has probably already been created;
5518 you get a half-formed symbol in your object file. This is
5519 compatible with earlier assemblers!
5522 @cindex @code{stabd} directive
5523 @item .stabd @var{type} , @var{other} , @var{desc}
5525 The ``name'' of the symbol generated is not even an empty string.
5526 It is a null pointer, for compatibility. Older assemblers used a
5527 null pointer so they didn't waste space in object files with empty
5530 The symbol's value is set to the location counter,
5531 relocatably. When your program is linked, the value of this symbol
5532 is the address of the location counter when the @code{.stabd} was
5535 @cindex @code{stabn} directive
5536 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
5537 The name of the symbol is set to the empty string @code{""}.
5539 @cindex @code{stabs} directive
5540 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
5541 All five fields are specified.
5547 @section @code{.string} "@var{str}"
5549 @cindex string, copying to object file
5550 @cindex @code{string} directive
5552 Copy the characters in @var{str} to the object file. You may specify more than
5553 one string to copy, separated by commas. Unless otherwise specified for a
5554 particular machine, the assembler marks the end of each string with a 0 byte.
5555 You can use any of the escape sequences described in @ref{Strings,,Strings}.
5558 @section @code{.struct @var{expression}}
5560 @cindex @code{struct} directive
5561 Switch to the absolute section, and set the section offset to @var{expression},
5562 which must be an absolute expression. You might use this as follows:
5571 This would define the symbol @code{field1} to have the value 0, the symbol
5572 @code{field2} to have the value 4, and the symbol @code{field3} to have the
5573 value 8. Assembly would be left in the absolute section, and you would need to
5574 use a @code{.section} directive of some sort to change to some other section
5575 before further assembly.
5579 @section @code{.subsection @var{name}}
5581 @cindex @code{subsection} directive
5582 @cindex Section Stack
5583 This is one of the ELF section stack manipulation directives. The others are
5584 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
5585 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5588 This directive replaces the current subsection with @code{name}. The current
5589 section is not changed. The replaced subsection is put onto the section stack
5590 in place of the then current top of stack subsection.
5595 @section @code{.symver}
5596 @cindex @code{symver} directive
5597 @cindex symbol versioning
5598 @cindex versions of symbols
5599 Use the @code{.symver} directive to bind symbols to specific version nodes
5600 within a source file. This is only supported on ELF platforms, and is
5601 typically used when assembling files to be linked into a shared library.
5602 There are cases where it may make sense to use this in objects to be bound
5603 into an application itself so as to override a versioned symbol from a
5606 For ELF targets, the @code{.symver} directive can be used like this:
5608 .symver @var{name}, @var{name2@@nodename}
5610 If the symbol @var{name} is defined within the file
5611 being assembled, the @code{.symver} directive effectively creates a symbol
5612 alias with the name @var{name2@@nodename}, and in fact the main reason that we
5613 just don't try and create a regular alias is that the @var{@@} character isn't
5614 permitted in symbol names. The @var{name2} part of the name is the actual name
5615 of the symbol by which it will be externally referenced. The name @var{name}
5616 itself is merely a name of convenience that is used so that it is possible to
5617 have definitions for multiple versions of a function within a single source
5618 file, and so that the compiler can unambiguously know which version of a
5619 function is being mentioned. The @var{nodename} portion of the alias should be
5620 the name of a node specified in the version script supplied to the linker when
5621 building a shared library. If you are attempting to override a versioned
5622 symbol from a shared library, then @var{nodename} should correspond to the
5623 nodename of the symbol you are trying to override.
5625 If the symbol @var{name} is not defined within the file being assembled, all
5626 references to @var{name} will be changed to @var{name2@@nodename}. If no
5627 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
5630 Another usage of the @code{.symver} directive is:
5632 .symver @var{name}, @var{name2@@@@nodename}
5634 In this case, the symbol @var{name} must exist and be defined within
5635 the file being assembled. It is similar to @var{name2@@nodename}. The
5636 difference is @var{name2@@@@nodename} will also be used to resolve
5637 references to @var{name2} by the linker.
5639 The third usage of the @code{.symver} directive is:
5641 .symver @var{name}, @var{name2@@@@@@nodename}
5643 When @var{name} is not defined within the
5644 file being assembled, it is treated as @var{name2@@nodename}. When
5645 @var{name} is defined within the file being assembled, the symbol
5646 name, @var{name}, will be changed to @var{name2@@@@nodename}.
5651 @section @code{.tag @var{structname}}
5653 @cindex COFF structure debugging
5654 @cindex structure debugging, COFF
5655 @cindex @code{tag} directive
5656 This directive is generated by compilers to include auxiliary debugging
5657 information in the symbol table. It is only permitted inside
5658 @code{.def}/@code{.endef} pairs. Tags are used to link structure
5659 definitions in the symbol table with instances of those structures.
5662 @samp{.tag} is only used when generating COFF format output; when
5663 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5669 @section @code{.text @var{subsection}}
5671 @cindex @code{text} directive
5672 Tells @command{@value{AS}} to assemble the following statements onto the end of
5673 the text subsection numbered @var{subsection}, which is an absolute
5674 expression. If @var{subsection} is omitted, subsection number zero
5678 @section @code{.title "@var{heading}"}
5680 @cindex @code{title} directive
5681 @cindex listing control: title line
5682 Use @var{heading} as the title (second line, immediately after the
5683 source file name and pagenumber) when generating assembly listings.
5685 This directive affects subsequent pages, as well as the current page if
5686 it appears within ten lines of the top of a page.
5690 @section @code{.type}
5692 This directive is used to set the type of a symbol.
5696 @c only print the extra heading if both COFF and ELF are set
5697 @subheading COFF Version
5700 @cindex COFF symbol type
5701 @cindex symbol type, COFF
5702 @cindex @code{type} directive (COFF version)
5703 For COFF targets, this directive is permitted only within
5704 @code{.def}/@code{.endef} pairs. It is used like this:
5710 This records the integer @var{int} as the type attribute of a symbol table
5714 @samp{.type} is associated only with COFF format output; when
5715 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
5716 directive but ignores it.
5722 @c only print the extra heading if both COFF and ELF are set
5723 @subheading ELF Version
5726 @cindex ELF symbol type
5727 @cindex symbol type, ELF
5728 @cindex @code{type} directive (ELF version)
5729 For ELF targets, the @code{.type} directive is used like this:
5732 .type @var{name} , @var{type description}
5735 This sets the type of symbol @var{name} to be either a
5736 function symbol or an object symbol. There are five different syntaxes
5737 supported for the @var{type description} field, in order to provide
5738 compatibility with various other assemblers. The syntaxes supported are:
5741 .type <name>,#function
5742 .type <name>,#object
5744 .type <name>,@@function
5745 .type <name>,@@object
5747 .type <name>,%function
5748 .type <name>,%object
5750 .type <name>,"function"
5751 .type <name>,"object"
5753 .type <name> STT_FUNCTION
5754 .type <name> STT_OBJECT
5760 @section @code{.uleb128 @var{expressions}}
5762 @cindex @code{uleb128} directive
5763 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
5764 compact, variable length representation of numbers used by the DWARF
5765 symbolic debugging format. @xref{Sleb128,@code{.sleb128}}.
5769 @section @code{.val @var{addr}}
5771 @cindex @code{val} directive
5772 @cindex COFF value attribute
5773 @cindex value attribute, COFF
5774 This directive, permitted only within @code{.def}/@code{.endef} pairs,
5775 records the address @var{addr} as the value attribute of a symbol table
5779 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
5780 configured for @code{b.out}, it accepts this directive but ignores it.
5786 @section @code{.version "@var{string}"}
5788 @cindex @code{version} directive
5789 This directive creates a @code{.note} section and places into it an ELF
5790 formatted note of type NT_VERSION. The note's name is set to @code{string}.
5795 @section @code{.vtable_entry @var{table}, @var{offset}}
5797 @cindex @code{vtable_entry}
5798 This directive finds or creates a symbol @code{table} and creates a
5799 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
5802 @section @code{.vtable_inherit @var{child}, @var{parent}}
5804 @cindex @code{vtable_inherit}
5805 This directive finds the symbol @code{child} and finds or creates the symbol
5806 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
5807 parent whose addend is the value of the child symbol. As a special case the
5808 parent name of @code{0} is treated as refering the @code{*ABS*} section.
5813 @section @code{.weak @var{names}}
5815 @cindex @code{weak} directive
5816 This directive sets the weak attribute on the comma separated list of symbol
5817 @code{names}. If the symbols do not already exist, they will be created.
5821 @section @code{.word @var{expressions}}
5823 @cindex @code{word} directive
5824 This directive expects zero or more @var{expressions}, of any section,
5825 separated by commas.
5828 For each expression, @command{@value{AS}} emits a 32-bit number.
5831 For each expression, @command{@value{AS}} emits a 16-bit number.
5836 The size of the number emitted, and its byte order,
5837 depend on what target computer the assembly is for.
5840 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
5841 @c happen---32-bit addressability, period; no long/short jumps.
5842 @ifset DIFF-TBL-KLUGE
5843 @cindex difference tables altered
5844 @cindex altered difference tables
5846 @emph{Warning: Special Treatment to support Compilers}
5850 Machines with a 32-bit address space, but that do less than 32-bit
5851 addressing, require the following special treatment. If the machine of
5852 interest to you does 32-bit addressing (or doesn't require it;
5853 @pxref{Machine Dependencies}), you can ignore this issue.
5856 In order to assemble compiler output into something that works,
5857 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
5858 Directives of the form @samp{.word sym1-sym2} are often emitted by
5859 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
5860 directive of the form @samp{.word sym1-sym2}, and the difference between
5861 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
5862 creates a @dfn{secondary jump table}, immediately before the next label.
5863 This secondary jump table is preceded by a short-jump to the
5864 first byte after the secondary table. This short-jump prevents the flow
5865 of control from accidentally falling into the new table. Inside the
5866 table is a long-jump to @code{sym2}. The original @samp{.word}
5867 contains @code{sym1} minus the address of the long-jump to
5870 If there were several occurrences of @samp{.word sym1-sym2} before the
5871 secondary jump table, all of them are adjusted. If there was a
5872 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
5873 long-jump to @code{sym4} is included in the secondary jump table,
5874 and the @code{.word} directives are adjusted to contain @code{sym3}
5875 minus the address of the long-jump to @code{sym4}; and so on, for as many
5876 entries in the original jump table as necessary.
5879 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
5880 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
5881 assembly language programmers.
5884 @c end DIFF-TBL-KLUGE
5887 @section Deprecated Directives
5889 @cindex deprecated directives
5890 @cindex obsolescent directives
5891 One day these directives won't work.
5892 They are included for compatibility with older assemblers.
5899 @node Machine Dependencies
5900 @chapter Machine Dependent Features
5902 @cindex machine dependencies
5903 The machine instruction sets are (almost by definition) different on
5904 each machine where @command{@value{AS}} runs. Floating point representations
5905 vary as well, and @command{@value{AS}} often supports a few additional
5906 directives or command-line options for compatibility with other
5907 assemblers on a particular platform. Finally, some versions of
5908 @command{@value{AS}} support special pseudo-instructions for branch
5911 This chapter discusses most of these differences, though it does not
5912 include details on any machine's instruction set. For details on that
5913 subject, see the hardware manufacturer's manual.
5917 * AMD29K-Dependent:: AMD 29K Dependent Features
5920 * Alpha-Dependent:: Alpha Dependent Features
5923 * ARC-Dependent:: ARC Dependent Features
5926 * ARM-Dependent:: ARM Dependent Features
5929 * CRIS-Dependent:: CRIS Dependent Features
5932 * D10V-Dependent:: D10V Dependent Features
5935 * D30V-Dependent:: D30V Dependent Features
5938 * H8/300-Dependent:: Renesas H8/300 Dependent Features
5941 * H8/500-Dependent:: Renesas H8/500 Dependent Features
5944 * HPPA-Dependent:: HPPA Dependent Features
5947 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
5950 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
5953 * i860-Dependent:: Intel 80860 Dependent Features
5956 * i960-Dependent:: Intel 80960 Dependent Features
5959 * IP2K-Dependent:: IP2K Dependent Features
5962 * M32R-Dependent:: M32R Dependent Features
5965 * M68K-Dependent:: M680x0 Dependent Features
5968 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
5971 * M88K-Dependent:: M880x0 Dependent Features
5974 * MIPS-Dependent:: MIPS Dependent Features
5977 * MMIX-Dependent:: MMIX Dependent Features
5980 * MSP430-Dependent:: MSP430 Dependent Features
5983 * SH-Dependent:: Renesas / SuperH SH Dependent Features
5984 * SH64-Dependent:: SuperH SH64 Dependent Features
5987 * PDP-11-Dependent:: PDP-11 Dependent Features
5990 * PJ-Dependent:: picoJava Dependent Features
5993 * PPC-Dependent:: PowerPC Dependent Features
5996 * Sparc-Dependent:: SPARC Dependent Features
5999 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6002 * V850-Dependent:: V850 Dependent Features
6005 * Xtensa-Dependent:: Xtensa Dependent Features
6008 * Z8000-Dependent:: Z8000 Dependent Features
6011 * Vax-Dependent:: VAX Dependent Features
6018 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6019 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6020 @c peculiarity: to preserve cross-references, there must be a node called
6021 @c "Machine Dependencies". Hence the conditional nodenames in each
6022 @c major node below. Node defaulting in makeinfo requires adjacency of
6023 @c node and sectioning commands; hence the repetition of @chapter BLAH
6024 @c in both conditional blocks.
6027 @include c-a29k.texi
6031 @include c-alpha.texi
6043 @include c-cris.texi
6048 @node Machine Dependencies
6049 @chapter Machine Dependent Features
6051 The machine instruction sets are different on each Renesas chip family,
6052 and there are also some syntax differences among the families. This
6053 chapter describes the specific @command{@value{AS}} features for each
6057 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6058 * H8/500-Dependent:: Renesas H8/500 Dependent Features
6059 * SH-Dependent:: Renesas SH Dependent Features
6066 @include c-d10v.texi
6070 @include c-d30v.texi
6074 @include c-h8300.texi
6078 @include c-h8500.texi
6082 @include c-hppa.texi
6086 @include c-i370.texi
6090 @include c-i386.texi
6094 @include c-i860.texi
6098 @include c-i960.texi
6102 @include c-ia64.texi
6106 @include c-ip2k.texi
6110 @include c-m32r.texi
6114 @include c-m68k.texi
6118 @include c-m68hc11.texi
6122 @include c-m88k.texi
6126 @include c-mips.texi
6130 @include c-mmix.texi
6134 @include c-msp430.texi
6138 @include c-ns32k.texi
6142 @include c-pdp11.texi
6155 @include c-sh64.texi
6159 @include c-sparc.texi
6163 @include c-tic54x.texi
6175 @include c-v850.texi
6179 @include c-xtensa.texi
6183 @c reverse effect of @down at top of generic Machine-Dep chapter
6187 @node Reporting Bugs
6188 @chapter Reporting Bugs
6189 @cindex bugs in assembler
6190 @cindex reporting bugs in assembler
6192 Your bug reports play an essential role in making @command{@value{AS}} reliable.
6194 Reporting a bug may help you by bringing a solution to your problem, or it may
6195 not. But in any case the principal function of a bug report is to help the
6196 entire community by making the next version of @command{@value{AS}} work better.
6197 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
6199 In order for a bug report to serve its purpose, you must include the
6200 information that enables us to fix the bug.
6203 * Bug Criteria:: Have you found a bug?
6204 * Bug Reporting:: How to report bugs
6208 @section Have You Found a Bug?
6209 @cindex bug criteria
6211 If you are not sure whether you have found a bug, here are some guidelines:
6214 @cindex fatal signal
6215 @cindex assembler crash
6216 @cindex crash of assembler
6218 If the assembler gets a fatal signal, for any input whatever, that is a
6219 @command{@value{AS}} bug. Reliable assemblers never crash.
6221 @cindex error on valid input
6223 If @command{@value{AS}} produces an error message for valid input, that is a bug.
6225 @cindex invalid input
6227 If @command{@value{AS}} does not produce an error message for invalid input, that
6228 is a bug. However, you should note that your idea of ``invalid input'' might
6229 be our idea of ``an extension'' or ``support for traditional practice''.
6232 If you are an experienced user of assemblers, your suggestions for improvement
6233 of @command{@value{AS}} are welcome in any case.
6237 @section How to Report Bugs
6239 @cindex assembler bugs, reporting
6241 A number of companies and individuals offer support for @sc{gnu} products. If
6242 you obtained @command{@value{AS}} from a support organization, we recommend you
6243 contact that organization first.
6245 You can find contact information for many support companies and
6246 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
6249 In any event, we also recommend that you send bug reports for @command{@value{AS}}
6250 to @samp{bug-binutils@@gnu.org}.
6252 The fundamental principle of reporting bugs usefully is this:
6253 @strong{report all the facts}. If you are not sure whether to state a
6254 fact or leave it out, state it!
6256 Often people omit facts because they think they know what causes the problem
6257 and assume that some details do not matter. Thus, you might assume that the
6258 name of a symbol you use in an example does not matter. Well, probably it does
6259 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
6260 happens to fetch from the location where that name is stored in memory;
6261 perhaps, if the name were different, the contents of that location would fool
6262 the assembler into doing the right thing despite the bug. Play it safe and
6263 give a specific, complete example. That is the easiest thing for you to do,
6264 and the most helpful.
6266 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
6267 it is new to us. Therefore, always write your bug reports on the assumption
6268 that the bug has not been reported previously.
6270 Sometimes people give a few sketchy facts and ask, ``Does this ring a
6271 bell?'' This cannot help us fix a bug, so it is basically useless. We
6272 respond by asking for enough details to enable us to investigate.
6273 You might as well expedite matters by sending them to begin with.
6275 To enable us to fix the bug, you should include all these things:
6279 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
6280 it with the @samp{--version} argument.
6282 Without this, we will not know whether there is any point in looking for
6283 the bug in the current version of @command{@value{AS}}.
6286 Any patches you may have applied to the @command{@value{AS}} source.
6289 The type of machine you are using, and the operating system name and
6293 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
6297 The command arguments you gave the assembler to assemble your example and
6298 observe the bug. To guarantee you will not omit something important, list them
6299 all. A copy of the Makefile (or the output from make) is sufficient.
6301 If we were to try to guess the arguments, we would probably guess wrong
6302 and then we might not encounter the bug.
6305 A complete input file that will reproduce the bug. If the bug is observed when
6306 the assembler is invoked via a compiler, send the assembler source, not the
6307 high level language source. Most compilers will produce the assembler source
6308 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
6309 the options @samp{-v --save-temps}; this will save the assembler source in a
6310 file with an extension of @file{.s}, and also show you exactly how
6311 @command{@value{AS}} is being run.
6314 A description of what behavior you observe that you believe is
6315 incorrect. For example, ``It gets a fatal signal.''
6317 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
6318 will certainly notice it. But if the bug is incorrect output, we might not
6319 notice unless it is glaringly wrong. You might as well not give us a chance to
6322 Even if the problem you experience is a fatal signal, you should still say so
6323 explicitly. Suppose something strange is going on, such as, your copy of
6324 @command{@value{AS}} is out of synch, or you have encountered a bug in the C
6325 library on your system. (This has happened!) Your copy might crash and ours
6326 would not. If you told us to expect a crash, then when ours fails to crash, we
6327 would know that the bug was not happening for us. If you had not told us to
6328 expect a crash, then we would not be able to draw any conclusion from our
6332 If you wish to suggest changes to the @command{@value{AS}} source, send us context
6333 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
6334 option. Always send diffs from the old file to the new file. If you even
6335 discuss something in the @command{@value{AS}} source, refer to it by context, not
6338 The line numbers in our development sources will not match those in your
6339 sources. Your line numbers would convey no useful information to us.
6342 Here are some things that are not necessary:
6346 A description of the envelope of the bug.
6348 Often people who encounter a bug spend a lot of time investigating
6349 which changes to the input file will make the bug go away and which
6350 changes will not affect it.
6352 This is often time consuming and not very useful, because the way we
6353 will find the bug is by running a single example under the debugger
6354 with breakpoints, not by pure deduction from a series of examples.
6355 We recommend that you save your time for something else.
6357 Of course, if you can find a simpler example to report @emph{instead}
6358 of the original one, that is a convenience for us. Errors in the
6359 output will be easier to spot, running under the debugger will take
6360 less time, and so on.
6362 However, simplification is not vital; if you do not want to do this,
6363 report the bug anyway and send us the entire test case you used.
6366 A patch for the bug.
6368 A patch for the bug does help us if it is a good one. But do not omit
6369 the necessary information, such as the test case, on the assumption that
6370 a patch is all we need. We might see problems with your patch and decide
6371 to fix the problem another way, or we might not understand it at all.
6373 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
6374 construct an example that will make the program follow a certain path through
6375 the code. If you do not send us the example, we will not be able to construct
6376 one, so we will not be able to verify that the bug is fixed.
6378 And if we cannot understand what bug you are trying to fix, or why your
6379 patch should be an improvement, we will not install it. A test case will
6380 help us to understand.
6383 A guess about what the bug is or what it depends on.
6385 Such guesses are usually wrong. Even we cannot guess right about such
6386 things without first using the debugger to find the facts.
6389 @node Acknowledgements
6390 @chapter Acknowledgements
6392 If you have contributed to @command{@value{AS}} and your name isn't listed here,
6393 it is not meant as a slight. We just don't know about it. Send mail to the
6394 maintainer, and we'll correct the situation. Currently
6396 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
6398 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
6401 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
6402 information and the 68k series machines, most of the preprocessing pass, and
6403 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
6405 K. Richard Pixley maintained GAS for a while, adding various enhancements and
6406 many bug fixes, including merging support for several processors, breaking GAS
6407 up to handle multiple object file format back ends (including heavy rewrite,
6408 testing, an integration of the coff and b.out back ends), adding configuration
6409 including heavy testing and verification of cross assemblers and file splits
6410 and renaming, converted GAS to strictly ANSI C including full prototypes, added
6411 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
6412 port (including considerable amounts of reverse engineering), a SPARC opcode
6413 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
6414 assertions and made them work, much other reorganization, cleanup, and lint.
6416 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
6417 in format-specific I/O modules.
6419 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
6420 has done much work with it since.
6422 The Intel 80386 machine description was written by Eliot Dresselhaus.
6424 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
6426 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
6427 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
6429 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
6430 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
6431 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
6432 support a.out format.
6434 Support for the Zilog Z8k and Renesas H8/300 and H8/500 processors (tc-z8k,
6435 tc-h8300, tc-h8500), and IEEE 695 object file format (obj-ieee), was written by
6436 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
6437 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
6440 John Gilmore built the AMD 29000 support, added @code{.include} support, and
6441 simplified the configuration of which versions accept which directives. He
6442 updated the 68k machine description so that Motorola's opcodes always produced
6443 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
6444 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
6445 cross-compilation support, and one bug in relaxation that took a week and
6446 required the proverbial one-bit fix.
6448 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
6449 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
6450 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
6451 PowerPC assembler, and made a few other minor patches.
6453 Steve Chamberlain made @command{@value{AS}} able to generate listings.
6455 Hewlett-Packard contributed support for the HP9000/300.
6457 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
6458 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
6459 formats). This work was supported by both the Center for Software Science at
6460 the University of Utah and Cygnus Support.
6462 Support for ELF format files has been worked on by Mark Eichin of Cygnus
6463 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
6464 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
6465 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
6466 and some initial 64-bit support).
6468 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
6470 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
6471 support for openVMS/Alpha.
6473 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
6476 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
6477 Inc. added support for Xtensa processors.
6479 Several engineers at Cygnus Support have also provided many small bug fixes and
6480 configuration enhancements.
6482 Many others have contributed large or small bugfixes and enhancements. If
6483 you have contributed significant work and are not mentioned on this list, and
6484 want to be, let us know. Some of the history has been lost; we are not
6485 intentionally leaving anyone out.
projects / binutils-gdb.git / blob