1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
3 @c 2001, 2002, 2003, 2004, 2005, 2006, 2007
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
52 @set abnormal-separator
56 @settitle Using @value{AS}
59 @settitle Using @value{AS} (@value{TARGET})
61 @setchapternewpage odd
66 @c WARE! Some of the machine-dependent sections contain tables of machine
67 @c instructions. Except in multi-column format, these tables look silly.
68 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
69 @c the multi-col format is faked within @example sections.
71 @c Again unfortunately, the natural size that fits on a page, for these tables,
72 @c is different depending on whether or not smallbook is turned on.
73 @c This matters, because of order: text flow switches columns at each page
76 @c The format faked in this source works reasonably well for smallbook,
77 @c not well for the default large-page format. This manual expects that if you
78 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
79 @c tables in question. You can turn on one without the other at your
80 @c discretion, of course.
83 @c the insn tables look just as silly in info files regardless of smallbook,
84 @c might as well show 'em anyways.
90 * As: (as). The GNU assembler.
91 * Gas: (as). The GNU assembler.
100 This file documents the GNU Assembler "@value{AS}".
102 @c man begin COPYRIGHT
103 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002,
104 2006, 2007 Free Software Foundation, Inc.
106 Permission is granted to copy, distribute and/or modify this document
107 under the terms of the GNU Free Documentation License, Version 1.1
108 or any later version published by the Free Software Foundation;
109 with no Invariant Sections, with no Front-Cover Texts, and with no
110 Back-Cover Texts. A copy of the license is included in the
111 section entitled ``GNU Free Documentation License''.
117 @title Using @value{AS}
118 @subtitle The @sc{gnu} Assembler
120 @subtitle for the @value{TARGET} family
122 @ifset VERSION_PACKAGE
124 @subtitle @value{VERSION_PACKAGE}
127 @subtitle Version @value{VERSION}
130 The Free Software Foundation Inc.@: thanks The Nice Computer
131 Company of Australia for loaning Dean Elsner to write the
132 first (Vax) version of @command{as} for Project @sc{gnu}.
133 The proprietors, management and staff of TNCCA thank FSF for
134 distracting the boss while they got some work
137 @author Dean Elsner, Jay Fenlason & friends
141 \hfill {\it Using {\tt @value{AS}}}\par
142 \hfill Edited by Cygnus Support\par
144 %"boxit" macro for figures:
145 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
146 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
147 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
148 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
149 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
152 @vskip 0pt plus 1filll
153 Copyright @copyright{} 1991, 92, 93, 94, 95, 96, 97, 98, 99, 2000, 2001, 2002,
154 2006, 2007 Free Software Foundation, Inc.
156 Permission is granted to copy, distribute and/or modify this document
157 under the terms of the GNU Free Documentation License, Version 1.1
158 or any later version published by the Free Software Foundation;
159 with no Invariant Sections, with no Front-Cover Texts, and with no
160 Back-Cover Texts. A copy of the license is included in the
161 section entitled ``GNU Free Documentation License''.
168 @top Using @value{AS}
170 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
171 @ifset VERSION_PACKAGE
172 @value{VERSION_PACKAGE}
174 version @value{VERSION}.
176 This version of the file describes @command{@value{AS}} configured to generate
177 code for @value{TARGET} architectures.
180 This document is distributed under the terms of the GNU Free
181 Documentation License. A copy of the license is included in the
182 section entitled ``GNU Free Documentation License''.
185 * Overview:: Overview
186 * Invoking:: Command-Line Options
188 * Sections:: Sections and Relocation
190 * Expressions:: Expressions
191 * Pseudo Ops:: Assembler Directives
193 * Object Attributes:: Object Attributes
195 * Machine Dependencies:: Machine Dependent Features
196 * Reporting Bugs:: Reporting Bugs
197 * Acknowledgements:: Who Did What
198 * GNU Free Documentation License:: GNU Free Documentation License
199 * AS Index:: AS Index
206 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
208 This version of the manual describes @command{@value{AS}} configured to generate
209 code for @value{TARGET} architectures.
213 @cindex invocation summary
214 @cindex option summary
215 @cindex summary of options
216 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
217 see @ref{Invoking,,Command-Line Options}.
219 @c man title AS the portable GNU assembler.
223 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
227 @c We don't use deffn and friends for the following because they seem
228 @c to be limited to one line for the header.
230 @c man begin SYNOPSIS
231 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
232 [@b{--debug-prefix-map} @var{old}=@var{new}]
233 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
234 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
235 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
236 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
237 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}] [@b{-o}
238 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
239 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
240 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{@@@var{FILE}}]
241 [@b{--target-help}] [@var{target-options}]
242 [@b{--}|@var{files} @dots{}]
244 @c Target dependent options are listed below. Keep the list sorted.
245 @c Add an empty line for separation.
248 @emph{Target Alpha options:}
250 [@b{-mdebug} | @b{-no-mdebug}]
251 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
252 [@b{-F}] [@b{-32addr}]
256 @emph{Target ARC options:}
262 @emph{Target ARM options:}
263 @c Don't document the deprecated options
264 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
265 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
266 [@b{-mfpu}=@var{floating-point-format}]
267 [@b{-mfloat-abi}=@var{abi}]
268 [@b{-meabi}=@var{ver}]
271 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
272 @b{-mapcs-reentrant}]
273 [@b{-mthumb-interwork}] [@b{-k}]
277 @emph{Target CRIS options:}
278 [@b{--underscore} | @b{--no-underscore}]
280 [@b{--emulation=criself} | @b{--emulation=crisaout}]
281 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
282 @c Deprecated -- deliberately not documented.
287 @emph{Target D10V options:}
292 @emph{Target D30V options:}
293 [@b{-O}|@b{-n}|@b{-N}]
296 @c Renesas family chips have no machine-dependent assembler options
299 @c HPPA has no machine-dependent assembler options (yet).
303 @emph{Target i386 options:}
304 [@b{--32}|@b{--64}] [@b{-n}]
305 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
309 @emph{Target i960 options:}
310 @c see md_parse_option in tc-i960.c
311 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
313 [@b{-b}] [@b{-no-relax}]
317 @emph{Target IA-64 options:}
318 [@b{-mconstant-gp}|@b{-mauto-pic}]
319 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
321 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
322 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
323 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
324 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
328 @emph{Target IP2K options:}
329 [@b{-mip2022}|@b{-mip2022ext}]
333 @emph{Target M32C options:}
334 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
338 @emph{Target M32R options:}
339 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
344 @emph{Target M680X0 options:}
345 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
349 @emph{Target M68HC11 options:}
350 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
351 [@b{-mshort}|@b{-mlong}]
352 [@b{-mshort-double}|@b{-mlong-double}]
353 [@b{--force-long-branches}] [@b{--short-branches}]
354 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
355 [@b{--print-opcodes}] [@b{--generate-example}]
359 @emph{Target MCORE options:}
360 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
361 [@b{-mcpu=[210|340]}]
365 @emph{Target MIPS options:}
366 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
367 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
368 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
369 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
370 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
371 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
372 [@b{-mips64}] [@b{-mips64r2}]
373 [@b{-construct-floats}] [@b{-no-construct-floats}]
374 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
375 [@b{-mfix7000}] [@b{-mno-fix7000}]
376 [@b{-mips16}] [@b{-no-mips16}]
377 [@b{-msmartmips}] [@b{-mno-smartmips}]
378 [@b{-mips3d}] [@b{-no-mips3d}]
379 [@b{-mdmx}] [@b{-no-mdmx}]
380 [@b{-mdsp}] [@b{-mno-dsp}]
381 [@b{-mdspr2}] [@b{-mno-dspr2}]
382 [@b{-mmt}] [@b{-mno-mt}]
383 [@b{-mdebug}] [@b{-no-mdebug}]
384 [@b{-mpdr}] [@b{-mno-pdr}]
388 @emph{Target MMIX options:}
389 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
390 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
391 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
392 [@b{--linker-allocated-gregs}]
396 @emph{Target PDP11 options:}
397 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
398 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
399 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
403 @emph{Target picoJava options:}
408 @emph{Target PowerPC options:}
409 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
410 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}|
411 @b{-mbooke32}|@b{-mbooke64}]
412 [@b{-mcom}|@b{-many}|@b{-maltivec}|@b{-mvsx}] [@b{-memb}]
413 [@b{-mregnames}|@b{-mno-regnames}]
414 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
415 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
416 [@b{-msolaris}|@b{-mno-solaris}]
420 @emph{Target SPARC options:}
421 @c The order here is important. See c-sparc.texi.
422 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
423 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
424 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
429 @emph{Target TIC54X options:}
430 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
431 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
436 @emph{Target Z80 options:}
437 [@b{-z80}] [@b{-r800}]
438 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
439 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
440 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
441 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
442 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
443 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
447 @c Z8000 has no machine-dependent assembler options
451 @emph{Target Xtensa options:}
452 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
453 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
454 [@b{--[no-]transform}]
455 [@b{--rename-section} @var{oldname}=@var{newname}]
463 @include at-file.texi
466 Turn on listings, in any of a variety of ways:
470 omit false conditionals
473 omit debugging directives
476 include general information, like @value{AS} version and options passed
479 include high-level source
485 include macro expansions
488 omit forms processing
494 set the name of the listing file
497 You may combine these options; for example, use @samp{-aln} for assembly
498 listing without forms processing. The @samp{=file} option, if used, must be
499 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
502 Begin in alternate macro mode.
504 @xref{Altmacro,,@code{.altmacro}}.
508 Ignored. This option is accepted for script compatibility with calls to
511 @item --debug-prefix-map @var{old}=@var{new}
512 When assembling files in directory @file{@var{old}}, record debugging
513 information describing them as in @file{@var{new}} instead.
515 @item --defsym @var{sym}=@var{value}
516 Define the symbol @var{sym} to be @var{value} before assembling the input file.
517 @var{value} must be an integer constant. As in C, a leading @samp{0x}
518 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
519 value. The value of the symbol can be overridden inside a source file via the
520 use of a @code{.set} pseudo-op.
523 ``fast''---skip whitespace and comment preprocessing (assume source is
528 Generate debugging information for each assembler source line using whichever
529 debug format is preferred by the target. This currently means either STABS,
533 Generate stabs debugging information for each assembler line. This
534 may help debugging assembler code, if the debugger can handle it.
537 Generate stabs debugging information for each assembler line, with GNU
538 extensions that probably only gdb can handle, and that could make other
539 debuggers crash or refuse to read your program. This
540 may help debugging assembler code. Currently the only GNU extension is
541 the location of the current working directory at assembling time.
544 Generate DWARF2 debugging information for each assembler line. This
545 may help debugging assembler code, if the debugger can handle it. Note---this
546 option is only supported by some targets, not all of them.
549 Print a summary of the command line options and exit.
552 Print a summary of all target specific options and exit.
555 Add directory @var{dir} to the search list for @code{.include} directives.
558 Don't warn about signed overflow.
561 @ifclear DIFF-TBL-KLUGE
562 This option is accepted but has no effect on the @value{TARGET} family.
564 @ifset DIFF-TBL-KLUGE
565 Issue warnings when difference tables altered for long displacements.
570 Keep (in the symbol table) local symbols. These symbols start with
571 system-specific local label prefixes, typically @samp{.L} for ELF systems
572 or @samp{L} for traditional a.out systems.
577 @item --listing-lhs-width=@var{number}
578 Set the maximum width, in words, of the output data column for an assembler
579 listing to @var{number}.
581 @item --listing-lhs-width2=@var{number}
582 Set the maximum width, in words, of the output data column for continuation
583 lines in an assembler listing to @var{number}.
585 @item --listing-rhs-width=@var{number}
586 Set the maximum width of an input source line, as displayed in a listing, to
589 @item --listing-cont-lines=@var{number}
590 Set the maximum number of lines printed in a listing for a single line of input
593 @item -o @var{objfile}
594 Name the object-file output from @command{@value{AS}} @var{objfile}.
597 Fold the data section into the text section.
599 @kindex --hash-size=@var{number}
600 Set the default size of GAS's hash tables to a prime number close to
601 @var{number}. Increasing this value can reduce the length of time it takes the
602 assembler to perform its tasks, at the expense of increasing the assembler's
603 memory requirements. Similarly reducing this value can reduce the memory
604 requirements at the expense of speed.
606 @item --reduce-memory-overheads
607 This option reduces GAS's memory requirements, at the expense of making the
608 assembly processes slower. Currently this switch is a synonym for
609 @samp{--hash-size=4051}, but in the future it may have other effects as well.
612 Print the maximum space (in bytes) and total time (in seconds) used by
615 @item --strip-local-absolute
616 Remove local absolute symbols from the outgoing symbol table.
620 Print the @command{as} version.
623 Print the @command{as} version and exit.
627 Suppress warning messages.
629 @item --fatal-warnings
630 Treat warnings as errors.
633 Don't suppress warning messages or treat them as errors.
642 Generate an object file even after errors.
644 @item -- | @var{files} @dots{}
645 Standard input, or source files to assemble.
650 The following options are available when @value{AS} is configured for
655 This option selects the core processor variant.
657 Select either big-endian (-EB) or little-endian (-EL) output.
662 The following options are available when @value{AS} is configured for the ARM
666 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
667 Specify which ARM processor variant is the target.
668 @item -march=@var{architecture}[+@var{extension}@dots{}]
669 Specify which ARM architecture variant is used by the target.
670 @item -mfpu=@var{floating-point-format}
671 Select which Floating Point architecture is the target.
672 @item -mfloat-abi=@var{abi}
673 Select which floating point ABI is in use.
675 Enable Thumb only instruction decoding.
676 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
677 Select which procedure calling convention is in use.
679 Select either big-endian (-EB) or little-endian (-EL) output.
680 @item -mthumb-interwork
681 Specify that the code has been generated with interworking between Thumb and
684 Specify that PIC code has been generated.
689 See the info pages for documentation of the CRIS-specific options.
693 The following options are available when @value{AS} is configured for
696 @cindex D10V optimization
697 @cindex optimization, D10V
699 Optimize output by parallelizing instructions.
704 The following options are available when @value{AS} is configured for a D30V
707 @cindex D30V optimization
708 @cindex optimization, D30V
710 Optimize output by parallelizing instructions.
714 Warn when nops are generated.
716 @cindex D30V nops after 32-bit multiply
718 Warn when a nop after a 32-bit multiply instruction is generated.
723 The following options are available when @value{AS} is configured for the
724 Intel 80960 processor.
727 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
728 Specify which variant of the 960 architecture is the target.
731 Add code to collect statistics about branches taken.
734 Do not alter compare-and-branch instructions for long displacements;
741 The following options are available when @value{AS} is configured for the
747 Specifies that the extended IP2022 instructions are allowed.
750 Restores the default behaviour, which restricts the permitted instructions to
751 just the basic IP2022 ones.
757 The following options are available when @value{AS} is configured for the
758 Renesas M32C and M16C processors.
763 Assemble M32C instructions.
766 Assemble M16C instructions (the default).
769 Enable support for link-time relaxations.
772 Support H'00 style hex constants in addition to 0x00 style.
778 The following options are available when @value{AS} is configured for the
779 Renesas M32R (formerly Mitsubishi M32R) series.
784 Specify which processor in the M32R family is the target. The default
785 is normally the M32R, but this option changes it to the M32RX.
787 @item --warn-explicit-parallel-conflicts or --Wp
788 Produce warning messages when questionable parallel constructs are
791 @item --no-warn-explicit-parallel-conflicts or --Wnp
792 Do not produce warning messages when questionable parallel constructs are
799 The following options are available when @value{AS} is configured for the
800 Motorola 68000 series.
805 Shorten references to undefined symbols, to one word instead of two.
807 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
808 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
809 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
810 Specify what processor in the 68000 family is the target. The default
811 is normally the 68020, but this can be changed at configuration time.
813 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
814 The target machine does (or does not) have a floating-point coprocessor.
815 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
816 the basic 68000 is not compatible with the 68881, a combination of the
817 two can be specified, since it's possible to do emulation of the
818 coprocessor instructions with the main processor.
820 @item -m68851 | -mno-68851
821 The target machine does (or does not) have a memory-management
822 unit coprocessor. The default is to assume an MMU for 68020 and up.
829 For details about the PDP-11 machine dependent features options,
830 see @ref{PDP-11-Options}.
833 @item -mpic | -mno-pic
834 Generate position-independent (or position-dependent) code. The
835 default is @option{-mpic}.
838 @itemx -mall-extensions
839 Enable all instruction set extensions. This is the default.
841 @item -mno-extensions
842 Disable all instruction set extensions.
844 @item -m@var{extension} | -mno-@var{extension}
845 Enable (or disable) a particular instruction set extension.
848 Enable the instruction set extensions supported by a particular CPU, and
849 disable all other extensions.
851 @item -m@var{machine}
852 Enable the instruction set extensions supported by a particular machine
853 model, and disable all other extensions.
859 The following options are available when @value{AS} is configured for
860 a picoJava processor.
864 @cindex PJ endianness
865 @cindex endianness, PJ
866 @cindex big endian output, PJ
868 Generate ``big endian'' format output.
870 @cindex little endian output, PJ
872 Generate ``little endian'' format output.
878 The following options are available when @value{AS} is configured for the
879 Motorola 68HC11 or 68HC12 series.
883 @item -m68hc11 | -m68hc12 | -m68hcs12
884 Specify what processor is the target. The default is
885 defined by the configuration option when building the assembler.
888 Specify to use the 16-bit integer ABI.
891 Specify to use the 32-bit integer ABI.
894 Specify to use the 32-bit double ABI.
897 Specify to use the 64-bit double ABI.
899 @item --force-long-branches
900 Relative branches are turned into absolute ones. This concerns
901 conditional branches, unconditional branches and branches to a
904 @item -S | --short-branches
905 Do not turn relative branches into absolute ones
906 when the offset is out of range.
908 @item --strict-direct-mode
909 Do not turn the direct addressing mode into extended addressing mode
910 when the instruction does not support direct addressing mode.
912 @item --print-insn-syntax
913 Print the syntax of instruction in case of error.
915 @item --print-opcodes
916 print the list of instructions with syntax and then exit.
918 @item --generate-example
919 print an example of instruction for each possible instruction and then exit.
920 This option is only useful for testing @command{@value{AS}}.
926 The following options are available when @command{@value{AS}} is configured
927 for the SPARC architecture:
930 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
931 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
932 Explicitly select a variant of the SPARC architecture.
934 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
935 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
937 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
938 UltraSPARC extensions.
940 @item -xarch=v8plus | -xarch=v8plusa
941 For compatibility with the Solaris v9 assembler. These options are
942 equivalent to -Av8plus and -Av8plusa, respectively.
945 Warn when the assembler switches to another architecture.
950 The following options are available when @value{AS} is configured for the 'c54x
955 Enable extended addressing mode. All addresses and relocations will assume
956 extended addressing (usually 23 bits).
957 @item -mcpu=@var{CPU_VERSION}
958 Sets the CPU version being compiled for.
959 @item -merrors-to-file @var{FILENAME}
960 Redirect error output to a file, for broken systems which don't support such
961 behaviour in the shell.
966 The following options are available when @value{AS} is configured for
967 a @sc{mips} processor.
971 This option sets the largest size of an object that can be referenced
972 implicitly with the @code{gp} register. It is only accepted for targets that
973 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
975 @cindex MIPS endianness
976 @cindex endianness, MIPS
977 @cindex big endian output, MIPS
979 Generate ``big endian'' format output.
981 @cindex little endian output, MIPS
983 Generate ``little endian'' format output.
995 Generate code for a particular @sc{mips} Instruction Set Architecture level.
996 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
997 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
998 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
999 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1001 correspond to generic
1002 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
1003 and @samp{MIPS64 Release 2}
1004 ISA processors, respectively.
1006 @item -march=@var{CPU}
1007 Generate code for a particular @sc{mips} cpu.
1009 @item -mtune=@var{cpu}
1010 Schedule and tune for a particular @sc{mips} cpu.
1014 Cause nops to be inserted if the read of the destination register
1015 of an mfhi or mflo instruction occurs in the following two instructions.
1019 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1020 section instead of the standard ELF .stabs sections.
1024 Control generation of @code{.pdr} sections.
1028 The register sizes are normally inferred from the ISA and ABI, but these
1029 flags force a certain group of registers to be treated as 32 bits wide at
1030 all times. @samp{-mgp32} controls the size of general-purpose registers
1031 and @samp{-mfp32} controls the size of floating-point registers.
1035 Generate code for the MIPS 16 processor. This is equivalent to putting
1036 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1037 turns off this option.
1040 @itemx -mno-smartmips
1041 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1042 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1043 @samp{-mno-smartmips} turns off this option.
1047 Generate code for the MIPS-3D Application Specific Extension.
1048 This tells the assembler to accept MIPS-3D instructions.
1049 @samp{-no-mips3d} turns off this option.
1053 Generate code for the MDMX Application Specific Extension.
1054 This tells the assembler to accept MDMX instructions.
1055 @samp{-no-mdmx} turns off this option.
1059 Generate code for the DSP Release 1 Application Specific Extension.
1060 This tells the assembler to accept DSP Release 1 instructions.
1061 @samp{-mno-dsp} turns off this option.
1065 Generate code for the DSP Release 2 Application Specific Extension.
1066 This option implies -mdsp.
1067 This tells the assembler to accept DSP Release 2 instructions.
1068 @samp{-mno-dspr2} turns off this option.
1072 Generate code for the MT Application Specific Extension.
1073 This tells the assembler to accept MT instructions.
1074 @samp{-mno-mt} turns off this option.
1076 @item --construct-floats
1077 @itemx --no-construct-floats
1078 The @samp{--no-construct-floats} option disables the construction of
1079 double width floating point constants by loading the two halves of the
1080 value into the two single width floating point registers that make up
1081 the double width register. By default @samp{--construct-floats} is
1082 selected, allowing construction of these floating point constants.
1085 @item --emulation=@var{name}
1086 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1087 for some other target, in all respects, including output format (choosing
1088 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1089 debugging information or store symbol table information, and default
1090 endianness. The available configuration names are: @samp{mipsecoff},
1091 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1092 @samp{mipsbelf}. The first two do not alter the default endianness from that
1093 of the primary target for which the assembler was configured; the others change
1094 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1095 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1096 selection in any case.
1098 This option is currently supported only when the primary target
1099 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1100 Furthermore, the primary target or others specified with
1101 @samp{--enable-targets=@dots{}} at configuration time must include support for
1102 the other format, if both are to be available. For example, the Irix 5
1103 configuration includes support for both.
1105 Eventually, this option will support more configurations, with more
1106 fine-grained control over the assembler's behavior, and will be supported for
1110 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1117 Control how to deal with multiplication overflow and division by zero.
1118 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1119 (and only work for Instruction Set Architecture level 2 and higher);
1120 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1124 When this option is used, @command{@value{AS}} will issue a warning every
1125 time it generates a nop instruction from a macro.
1130 The following options are available when @value{AS} is configured for
1136 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1137 The command line option @samp{-nojsri2bsr} can be used to disable it.
1141 Enable or disable the silicon filter behaviour. By default this is disabled.
1142 The default can be overridden by the @samp{-sifilter} command line option.
1145 Alter jump instructions for long displacements.
1147 @item -mcpu=[210|340]
1148 Select the cpu type on the target hardware. This controls which instructions
1152 Assemble for a big endian target.
1155 Assemble for a little endian target.
1161 See the info pages for documentation of the MMIX-specific options.
1165 The following options are available when @value{AS} is configured for
1166 an Xtensa processor.
1169 @item --text-section-literals | --no-text-section-literals
1170 With @option{--text-@-section-@-literals}, literal pools are interspersed
1171 in the text section. The default is
1172 @option{--no-@-text-@-section-@-literals}, which places literals in a
1173 separate section in the output file. These options only affect literals
1174 referenced via PC-relative @code{L32R} instructions; literals for
1175 absolute mode @code{L32R} instructions are handled separately.
1177 @item --absolute-literals | --no-absolute-literals
1178 Indicate to the assembler whether @code{L32R} instructions use absolute
1179 or PC-relative addressing. The default is to assume absolute addressing
1180 if the Xtensa processor includes the absolute @code{L32R} addressing
1181 option. Otherwise, only the PC-relative @code{L32R} mode can be used.
1183 @item --target-align | --no-target-align
1184 Enable or disable automatic alignment to reduce branch penalties at the
1185 expense of some code density. The default is @option{--target-@-align}.
1187 @item --longcalls | --no-longcalls
1188 Enable or disable transformation of call instructions to allow calls
1189 across a greater range of addresses. The default is
1190 @option{--no-@-longcalls}.
1192 @item --transform | --no-transform
1193 Enable or disable all assembler transformations of Xtensa instructions.
1194 The default is @option{--transform};
1195 @option{--no-transform} should be used only in the rare cases when the
1196 instructions must be exactly as specified in the assembly source.
1198 @item --rename-section @var{oldname}=@var{newname}
1199 When generating output sections, rename the @var{oldname} section to
1205 The following options are available when @value{AS} is configured for
1206 a Z80 family processor.
1209 Assemble for Z80 processor.
1211 Assemble for R800 processor.
1212 @item -ignore-undocumented-instructions
1214 Assemble undocumented Z80 instructions that also work on R800 without warning.
1215 @item -ignore-unportable-instructions
1217 Assemble all undocumented Z80 instructions without warning.
1218 @item -warn-undocumented-instructions
1220 Issue a warning for undocumented Z80 instructions that also work on R800.
1221 @item -warn-unportable-instructions
1223 Issue a warning for undocumented Z80 instructions that do not work on R800.
1224 @item -forbid-undocumented-instructions
1226 Treat all undocumented instructions as errors.
1227 @item -forbid-unportable-instructions
1229 Treat undocumented Z80 instructions that do not work on R800 as errors.
1236 * Manual:: Structure of this Manual
1237 * GNU Assembler:: The GNU Assembler
1238 * Object Formats:: Object File Formats
1239 * Command Line:: Command Line
1240 * Input Files:: Input Files
1241 * Object:: Output (Object) File
1242 * Errors:: Error and Warning Messages
1246 @section Structure of this Manual
1248 @cindex manual, structure and purpose
1249 This manual is intended to describe what you need to know to use
1250 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1251 notation for symbols, constants, and expressions; the directives that
1252 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1255 We also cover special features in the @value{TARGET}
1256 configuration of @command{@value{AS}}, including assembler directives.
1259 This manual also describes some of the machine-dependent features of
1260 various flavors of the assembler.
1263 @cindex machine instructions (not covered)
1264 On the other hand, this manual is @emph{not} intended as an introduction
1265 to programming in assembly language---let alone programming in general!
1266 In a similar vein, we make no attempt to introduce the machine
1267 architecture; we do @emph{not} describe the instruction set, standard
1268 mnemonics, registers or addressing modes that are standard to a
1269 particular architecture.
1271 You may want to consult the manufacturer's
1272 machine architecture manual for this information.
1276 For information on the H8/300 machine instruction set, see @cite{H8/300
1277 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1278 Programming Manual} (Renesas).
1281 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1282 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1283 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1284 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1287 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1291 @c I think this is premature---doc@cygnus.com, 17jan1991
1293 Throughout this manual, we assume that you are running @dfn{GNU},
1294 the portable operating system from the @dfn{Free Software
1295 Foundation, Inc.}. This restricts our attention to certain kinds of
1296 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1297 once this assumption is granted examples and definitions need less
1300 @command{@value{AS}} is part of a team of programs that turn a high-level
1301 human-readable series of instructions into a low-level
1302 computer-readable series of instructions. Different versions of
1303 @command{@value{AS}} are used for different kinds of computer.
1306 @c There used to be a section "Terminology" here, which defined
1307 @c "contents", "byte", "word", and "long". Defining "word" to any
1308 @c particular size is confusing when the .word directive may generate 16
1309 @c bits on one machine and 32 bits on another; in general, for the user
1310 @c version of this manual, none of these terms seem essential to define.
1311 @c They were used very little even in the former draft of the manual;
1312 @c this draft makes an effort to avoid them (except in names of
1316 @section The GNU Assembler
1318 @c man begin DESCRIPTION
1320 @sc{gnu} @command{as} is really a family of assemblers.
1322 This manual describes @command{@value{AS}}, a member of that family which is
1323 configured for the @value{TARGET} architectures.
1325 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1326 should find a fairly similar environment when you use it on another
1327 architecture. Each version has much in common with the others,
1328 including object file formats, most assembler directives (often called
1329 @dfn{pseudo-ops}) and assembler syntax.@refill
1331 @cindex purpose of @sc{gnu} assembler
1332 @command{@value{AS}} is primarily intended to assemble the output of the
1333 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1334 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1335 assemble correctly everything that other assemblers for the same
1336 machine would assemble.
1338 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1341 @c This remark should appear in generic version of manual; assumption
1342 @c here is that generic version sets M680x0.
1343 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1344 assembler for the same architecture; for example, we know of several
1345 incompatible versions of 680x0 assembly language syntax.
1350 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1351 program in one pass of the source file. This has a subtle impact on the
1352 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1354 @node Object Formats
1355 @section Object File Formats
1357 @cindex object file format
1358 The @sc{gnu} assembler can be configured to produce several alternative
1359 object file formats. For the most part, this does not affect how you
1360 write assembly language programs; but directives for debugging symbols
1361 are typically different in different file formats. @xref{Symbol
1362 Attributes,,Symbol Attributes}.
1365 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1366 @value{OBJ-NAME} format object files.
1368 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1370 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1371 @code{b.out} or COFF format object files.
1374 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1375 SOM or ELF format object files.
1380 @section Command Line
1382 @cindex command line conventions
1384 After the program name @command{@value{AS}}, the command line may contain
1385 options and file names. Options may appear in any order, and may be
1386 before, after, or between file names. The order of file names is
1389 @cindex standard input, as input file
1391 @file{--} (two hyphens) by itself names the standard input file
1392 explicitly, as one of the files for @command{@value{AS}} to assemble.
1394 @cindex options, command line
1395 Except for @samp{--} any command line argument that begins with a
1396 hyphen (@samp{-}) is an option. Each option changes the behavior of
1397 @command{@value{AS}}. No option changes the way another option works. An
1398 option is a @samp{-} followed by one or more letters; the case of
1399 the letter is important. All options are optional.
1401 Some options expect exactly one file name to follow them. The file
1402 name may either immediately follow the option's letter (compatible
1403 with older assemblers) or it may be the next command argument (@sc{gnu}
1404 standard). These two command lines are equivalent:
1407 @value{AS} -o my-object-file.o mumble.s
1408 @value{AS} -omy-object-file.o mumble.s
1412 @section Input Files
1415 @cindex source program
1416 @cindex files, input
1417 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1418 describe the program input to one run of @command{@value{AS}}. The program may
1419 be in one or more files; how the source is partitioned into files
1420 doesn't change the meaning of the source.
1422 @c I added "con" prefix to "catenation" just to prove I can overcome my
1423 @c APL training... doc@cygnus.com
1424 The source program is a concatenation of the text in all the files, in the
1427 @c man begin DESCRIPTION
1428 Each time you run @command{@value{AS}} it assembles exactly one source
1429 program. The source program is made up of one or more files.
1430 (The standard input is also a file.)
1432 You give @command{@value{AS}} a command line that has zero or more input file
1433 names. The input files are read (from left file name to right). A
1434 command line argument (in any position) that has no special meaning
1435 is taken to be an input file name.
1437 If you give @command{@value{AS}} no file names it attempts to read one input file
1438 from the @command{@value{AS}} standard input, which is normally your terminal. You
1439 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1442 Use @samp{--} if you need to explicitly name the standard input file
1443 in your command line.
1445 If the source is empty, @command{@value{AS}} produces a small, empty object
1450 @subheading Filenames and Line-numbers
1452 @cindex input file linenumbers
1453 @cindex line numbers, in input files
1454 There are two ways of locating a line in the input file (or files) and
1455 either may be used in reporting error messages. One way refers to a line
1456 number in a physical file; the other refers to a line number in a
1457 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1459 @dfn{Physical files} are those files named in the command line given
1460 to @command{@value{AS}}.
1462 @dfn{Logical files} are simply names declared explicitly by assembler
1463 directives; they bear no relation to physical files. Logical file names help
1464 error messages reflect the original source file, when @command{@value{AS}} source
1465 is itself synthesized from other files. @command{@value{AS}} understands the
1466 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1467 @ref{File,,@code{.file}}.
1470 @section Output (Object) File
1476 Every time you run @command{@value{AS}} it produces an output file, which is
1477 your assembly language program translated into numbers. This file
1478 is the object file. Its default name is
1486 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1488 You can give it another name by using the @option{-o} option. Conventionally,
1489 object file names end with @file{.o}. The default name is used for historical
1490 reasons: older assemblers were capable of assembling self-contained programs
1491 directly into a runnable program. (For some formats, this isn't currently
1492 possible, but it can be done for the @code{a.out} format.)
1496 The object file is meant for input to the linker @code{@value{LD}}. It contains
1497 assembled program code, information to help @code{@value{LD}} integrate
1498 the assembled program into a runnable file, and (optionally) symbolic
1499 information for the debugger.
1501 @c link above to some info file(s) like the description of a.out.
1502 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1505 @section Error and Warning Messages
1507 @c man begin DESCRIPTION
1509 @cindex error messages
1510 @cindex warning messages
1511 @cindex messages from assembler
1512 @command{@value{AS}} may write warnings and error messages to the standard error
1513 file (usually your terminal). This should not happen when a compiler
1514 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1515 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1516 grave problem that stops the assembly.
1520 @cindex format of warning messages
1521 Warning messages have the format
1524 file_name:@b{NNN}:Warning Message Text
1528 @cindex line numbers, in warnings/errors
1529 (where @b{NNN} is a line number). If a logical file name has been given
1530 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1531 the current input file is used. If a logical line number was given
1533 (@pxref{Line,,@code{.line}})
1535 then it is used to calculate the number printed,
1536 otherwise the actual line in the current source file is printed. The
1537 message text is intended to be self explanatory (in the grand Unix
1540 @cindex format of error messages
1541 Error messages have the format
1543 file_name:@b{NNN}:FATAL:Error Message Text
1545 The file name and line number are derived as for warning
1546 messages. The actual message text may be rather less explanatory
1547 because many of them aren't supposed to happen.
1550 @chapter Command-Line Options
1552 @cindex options, all versions of assembler
1553 This chapter describes command-line options available in @emph{all}
1554 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1555 for options specific
1557 to the @value{TARGET} target.
1560 to particular machine architectures.
1563 @c man begin DESCRIPTION
1565 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1566 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1567 The assembler arguments must be separated from each other (and the @samp{-Wa})
1568 by commas. For example:
1571 gcc -c -g -O -Wa,-alh,-L file.c
1575 This passes two options to the assembler: @samp{-alh} (emit a listing to
1576 standard output with high-level and assembly source) and @samp{-L} (retain
1577 local symbols in the symbol table).
1579 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1580 command-line options are automatically passed to the assembler by the compiler.
1581 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1582 precisely what options it passes to each compilation pass, including the
1588 * a:: -a[cdghlns] enable listings
1589 * alternate:: --alternate enable alternate macro syntax
1590 * D:: -D for compatibility
1591 * f:: -f to work faster
1592 * I:: -I for .include search path
1593 @ifclear DIFF-TBL-KLUGE
1594 * K:: -K for compatibility
1596 @ifset DIFF-TBL-KLUGE
1597 * K:: -K for difference tables
1600 * L:: -L to retain local symbols
1601 * listing:: --listing-XXX to configure listing output
1602 * M:: -M or --mri to assemble in MRI compatibility mode
1603 * MD:: --MD for dependency tracking
1604 * o:: -o to name the object file
1605 * R:: -R to join data and text sections
1606 * statistics:: --statistics to see statistics about assembly
1607 * traditional-format:: --traditional-format for compatible output
1608 * v:: -v to announce version
1609 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1610 * Z:: -Z to make object file even after errors
1614 @section Enable Listings: @option{-a[cdghlns]}
1624 @cindex listings, enabling
1625 @cindex assembly listings, enabling
1627 These options enable listing output from the assembler. By itself,
1628 @samp{-a} requests high-level, assembly, and symbols listing.
1629 You can use other letters to select specific options for the list:
1630 @samp{-ah} requests a high-level language listing,
1631 @samp{-al} requests an output-program assembly listing, and
1632 @samp{-as} requests a symbol table listing.
1633 High-level listings require that a compiler debugging option like
1634 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1637 Use the @samp{-ag} option to print a first section with general assembly
1638 information, like @value{AS} version, switches passed, or time stamp.
1640 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1641 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1642 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1643 omitted from the listing.
1645 Use the @samp{-ad} option to omit debugging directives from the
1648 Once you have specified one of these options, you can further control
1649 listing output and its appearance using the directives @code{.list},
1650 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1652 The @samp{-an} option turns off all forms processing.
1653 If you do not request listing output with one of the @samp{-a} options, the
1654 listing-control directives have no effect.
1656 The letters after @samp{-a} may be combined into one option,
1657 @emph{e.g.}, @samp{-aln}.
1659 Note if the assembler source is coming from the standard input (e.g.,
1661 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1662 is being used) then the listing will not contain any comments or preprocessor
1663 directives. This is because the listing code buffers input source lines from
1664 stdin only after they have been preprocessed by the assembler. This reduces
1665 memory usage and makes the code more efficient.
1668 @section @option{--alternate}
1671 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1674 @section @option{-D}
1677 This option has no effect whatsoever, but it is accepted to make it more
1678 likely that scripts written for other assemblers also work with
1679 @command{@value{AS}}.
1682 @section Work Faster: @option{-f}
1685 @cindex trusted compiler
1686 @cindex faster processing (@option{-f})
1687 @samp{-f} should only be used when assembling programs written by a
1688 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1689 and comment preprocessing on
1690 the input file(s) before assembling them. @xref{Preprocessing,
1694 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1695 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1700 @section @code{.include} Search Path: @option{-I} @var{path}
1702 @kindex -I @var{path}
1703 @cindex paths for @code{.include}
1704 @cindex search path for @code{.include}
1705 @cindex @code{include} directive search path
1706 Use this option to add a @var{path} to the list of directories
1707 @command{@value{AS}} searches for files specified in @code{.include}
1708 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1709 many times as necessary to include a variety of paths. The current
1710 working directory is always searched first; after that, @command{@value{AS}}
1711 searches any @samp{-I} directories in the same order as they were
1712 specified (left to right) on the command line.
1715 @section Difference Tables: @option{-K}
1718 @ifclear DIFF-TBL-KLUGE
1719 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1720 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1721 where it can be used to warn when the assembler alters the machine code
1722 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1723 family does not have the addressing limitations that sometimes lead to this
1724 alteration on other platforms.
1727 @ifset DIFF-TBL-KLUGE
1728 @cindex difference tables, warning
1729 @cindex warning for altered difference tables
1730 @command{@value{AS}} sometimes alters the code emitted for directives of the
1731 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
1732 You can use the @samp{-K} option if you want a warning issued when this
1737 @section Include Local Symbols: @option{-L}
1740 @cindex local symbols, retaining in output
1741 Symbols beginning with system-specific local label prefixes, typically
1742 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
1743 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
1744 such symbols when debugging, because they are intended for the use of
1745 programs (like compilers) that compose assembler programs, not for your
1746 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
1747 such symbols, so you do not normally debug with them.
1749 This option tells @command{@value{AS}} to retain those local symbols
1750 in the object file. Usually if you do this you also tell the linker
1751 @code{@value{LD}} to preserve those symbols.
1754 @section Configuring listing output: @option{--listing}
1756 The listing feature of the assembler can be enabled via the command line switch
1757 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1758 hex dump of the corresponding locations in the output object file, and displays
1759 them as a listing file. The format of this listing can be controlled by
1760 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
1761 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
1762 @code{.psize} (@pxref{Psize}), and
1763 @code{.eject} (@pxref{Eject}) and also by the following switches:
1766 @item --listing-lhs-width=@samp{number}
1767 @kindex --listing-lhs-width
1768 @cindex Width of first line disassembly output
1769 Sets the maximum width, in words, of the first line of the hex byte dump. This
1770 dump appears on the left hand side of the listing output.
1772 @item --listing-lhs-width2=@samp{number}
1773 @kindex --listing-lhs-width2
1774 @cindex Width of continuation lines of disassembly output
1775 Sets the maximum width, in words, of any further lines of the hex byte dump for
1776 a given input source line. If this value is not specified, it defaults to being
1777 the same as the value specified for @samp{--listing-lhs-width}. If neither
1778 switch is used the default is to one.
1780 @item --listing-rhs-width=@samp{number}
1781 @kindex --listing-rhs-width
1782 @cindex Width of source line output
1783 Sets the maximum width, in characters, of the source line that is displayed
1784 alongside the hex dump. The default value for this parameter is 100. The
1785 source line is displayed on the right hand side of the listing output.
1787 @item --listing-cont-lines=@samp{number}
1788 @kindex --listing-cont-lines
1789 @cindex Maximum number of continuation lines
1790 Sets the maximum number of continuation lines of hex dump that will be
1791 displayed for a given single line of source input. The default value is 4.
1795 @section Assemble in MRI Compatibility Mode: @option{-M}
1798 @cindex MRI compatibility mode
1799 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1800 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1801 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1802 configured target) assembler from Microtec Research. The exact nature of the
1803 MRI syntax will not be documented here; see the MRI manuals for more
1804 information. Note in particular that the handling of macros and macro
1805 arguments is somewhat different. The purpose of this option is to permit
1806 assembling existing MRI assembler code using @command{@value{AS}}.
1808 The MRI compatibility is not complete. Certain operations of the MRI assembler
1809 depend upon its object file format, and can not be supported using other object
1810 file formats. Supporting these would require enhancing each object file format
1811 individually. These are:
1814 @item global symbols in common section
1816 The m68k MRI assembler supports common sections which are merged by the linker.
1817 Other object file formats do not support this. @command{@value{AS}} handles
1818 common sections by treating them as a single common symbol. It permits local
1819 symbols to be defined within a common section, but it can not support global
1820 symbols, since it has no way to describe them.
1822 @item complex relocations
1824 The MRI assemblers support relocations against a negated section address, and
1825 relocations which combine the start addresses of two or more sections. These
1826 are not support by other object file formats.
1828 @item @code{END} pseudo-op specifying start address
1830 The MRI @code{END} pseudo-op permits the specification of a start address.
1831 This is not supported by other object file formats. The start address may
1832 instead be specified using the @option{-e} option to the linker, or in a linker
1835 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1837 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1838 name to the output file. This is not supported by other object file formats.
1840 @item @code{ORG} pseudo-op
1842 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1843 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1844 which changes the location within the current section. Absolute sections are
1845 not supported by other object file formats. The address of a section may be
1846 assigned within a linker script.
1849 There are some other features of the MRI assembler which are not supported by
1850 @command{@value{AS}}, typically either because they are difficult or because they
1851 seem of little consequence. Some of these may be supported in future releases.
1855 @item EBCDIC strings
1857 EBCDIC strings are not supported.
1859 @item packed binary coded decimal
1861 Packed binary coded decimal is not supported. This means that the @code{DC.P}
1862 and @code{DCB.P} pseudo-ops are not supported.
1864 @item @code{FEQU} pseudo-op
1866 The m68k @code{FEQU} pseudo-op is not supported.
1868 @item @code{NOOBJ} pseudo-op
1870 The m68k @code{NOOBJ} pseudo-op is not supported.
1872 @item @code{OPT} branch control options
1874 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
1875 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
1876 relaxes all branches, whether forward or backward, to an appropriate size, so
1877 these options serve no purpose.
1879 @item @code{OPT} list control options
1881 The following m68k @code{OPT} list control options are ignored: @code{C},
1882 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
1883 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
1885 @item other @code{OPT} options
1887 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
1888 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
1890 @item @code{OPT} @code{D} option is default
1892 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
1893 @code{OPT NOD} may be used to turn it off.
1895 @item @code{XREF} pseudo-op.
1897 The m68k @code{XREF} pseudo-op is ignored.
1899 @item @code{.debug} pseudo-op
1901 The i960 @code{.debug} pseudo-op is not supported.
1903 @item @code{.extended} pseudo-op
1905 The i960 @code{.extended} pseudo-op is not supported.
1907 @item @code{.list} pseudo-op.
1909 The various options of the i960 @code{.list} pseudo-op are not supported.
1911 @item @code{.optimize} pseudo-op
1913 The i960 @code{.optimize} pseudo-op is not supported.
1915 @item @code{.output} pseudo-op
1917 The i960 @code{.output} pseudo-op is not supported.
1919 @item @code{.setreal} pseudo-op
1921 The i960 @code{.setreal} pseudo-op is not supported.
1926 @section Dependency Tracking: @option{--MD}
1929 @cindex dependency tracking
1932 @command{@value{AS}} can generate a dependency file for the file it creates. This
1933 file consists of a single rule suitable for @code{make} describing the
1934 dependencies of the main source file.
1936 The rule is written to the file named in its argument.
1938 This feature is used in the automatic updating of makefiles.
1941 @section Name the Object File: @option{-o}
1944 @cindex naming object file
1945 @cindex object file name
1946 There is always one object file output when you run @command{@value{AS}}. By
1947 default it has the name
1950 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
1964 You use this option (which takes exactly one filename) to give the
1965 object file a different name.
1967 Whatever the object file is called, @command{@value{AS}} overwrites any
1968 existing file of the same name.
1971 @section Join Data and Text Sections: @option{-R}
1974 @cindex data and text sections, joining
1975 @cindex text and data sections, joining
1976 @cindex joining text and data sections
1977 @cindex merging text and data sections
1978 @option{-R} tells @command{@value{AS}} to write the object file as if all
1979 data-section data lives in the text section. This is only done at
1980 the very last moment: your binary data are the same, but data
1981 section parts are relocated differently. The data section part of
1982 your object file is zero bytes long because all its bytes are
1983 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
1985 When you specify @option{-R} it would be possible to generate shorter
1986 address displacements (because we do not have to cross between text and
1987 data section). We refrain from doing this simply for compatibility with
1988 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
1991 When @command{@value{AS}} is configured for COFF or ELF output,
1992 this option is only useful if you use sections named @samp{.text} and
1997 @option{-R} is not supported for any of the HPPA targets. Using
1998 @option{-R} generates a warning from @command{@value{AS}}.
2002 @section Display Assembly Statistics: @option{--statistics}
2004 @kindex --statistics
2005 @cindex statistics, about assembly
2006 @cindex time, total for assembly
2007 @cindex space used, maximum for assembly
2008 Use @samp{--statistics} to display two statistics about the resources used by
2009 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2010 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2013 @node traditional-format
2014 @section Compatible Output: @option{--traditional-format}
2016 @kindex --traditional-format
2017 For some targets, the output of @command{@value{AS}} is different in some ways
2018 from the output of some existing assembler. This switch requests
2019 @command{@value{AS}} to use the traditional format instead.
2021 For example, it disables the exception frame optimizations which
2022 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2025 @section Announce Version: @option{-v}
2029 @cindex assembler version
2030 @cindex version of assembler
2031 You can find out what version of as is running by including the
2032 option @samp{-v} (which you can also spell as @samp{-version}) on the
2036 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2038 @command{@value{AS}} should never give a warning or error message when
2039 assembling compiler output. But programs written by people often
2040 cause @command{@value{AS}} to give a warning that a particular assumption was
2041 made. All such warnings are directed to the standard error file.
2045 @cindex suppressing warnings
2046 @cindex warnings, suppressing
2047 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2048 This only affects the warning messages: it does not change any particular of
2049 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2052 @kindex --fatal-warnings
2053 @cindex errors, caused by warnings
2054 @cindex warnings, causing error
2055 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2056 files that generate warnings to be in error.
2059 @cindex warnings, switching on
2060 You can switch these options off again by specifying @option{--warn}, which
2061 causes warnings to be output as usual.
2064 @section Generate Object File in Spite of Errors: @option{-Z}
2065 @cindex object file, after errors
2066 @cindex errors, continuing after
2067 After an error message, @command{@value{AS}} normally produces no output. If for
2068 some reason you are interested in object file output even after
2069 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2070 option. If there are any errors, @command{@value{AS}} continues anyways, and
2071 writes an object file after a final warning message of the form @samp{@var{n}
2072 errors, @var{m} warnings, generating bad object file.}
2077 @cindex machine-independent syntax
2078 @cindex syntax, machine-independent
2079 This chapter describes the machine-independent syntax allowed in a
2080 source file. @command{@value{AS}} syntax is similar to what many other
2081 assemblers use; it is inspired by the BSD 4.2
2086 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2090 * Preprocessing:: Preprocessing
2091 * Whitespace:: Whitespace
2092 * Comments:: Comments
2093 * Symbol Intro:: Symbols
2094 * Statements:: Statements
2095 * Constants:: Constants
2099 @section Preprocessing
2101 @cindex preprocessing
2102 The @command{@value{AS}} internal preprocessor:
2104 @cindex whitespace, removed by preprocessor
2106 adjusts and removes extra whitespace. It leaves one space or tab before
2107 the keywords on a line, and turns any other whitespace on the line into
2110 @cindex comments, removed by preprocessor
2112 removes all comments, replacing them with a single space, or an
2113 appropriate number of newlines.
2115 @cindex constants, converted by preprocessor
2117 converts character constants into the appropriate numeric values.
2120 It does not do macro processing, include file handling, or
2121 anything else you may get from your C compiler's preprocessor. You can
2122 do include file processing with the @code{.include} directive
2123 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2124 to get other ``CPP'' style preprocessing by giving the input file a
2125 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2126 Output, gcc.info, Using GNU CC}.
2128 Excess whitespace, comments, and character constants
2129 cannot be used in the portions of the input text that are not
2132 @cindex turning preprocessing on and off
2133 @cindex preprocessing, turning on and off
2136 If the first line of an input file is @code{#NO_APP} or if you use the
2137 @samp{-f} option, whitespace and comments are not removed from the input file.
2138 Within an input file, you can ask for whitespace and comment removal in
2139 specific portions of the by putting a line that says @code{#APP} before the
2140 text that may contain whitespace or comments, and putting a line that says
2141 @code{#NO_APP} after this text. This feature is mainly intend to support
2142 @code{asm} statements in compilers whose output is otherwise free of comments
2149 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2150 Whitespace is used to separate symbols, and to make programs neater for
2151 people to read. Unless within character constants
2152 (@pxref{Characters,,Character Constants}), any whitespace means the same
2153 as exactly one space.
2159 There are two ways of rendering comments to @command{@value{AS}}. In both
2160 cases the comment is equivalent to one space.
2162 Anything from @samp{/*} through the next @samp{*/} is a comment.
2163 This means you may not nest these comments.
2167 The only way to include a newline ('\n') in a comment
2168 is to use this sort of comment.
2171 /* This sort of comment does not nest. */
2174 @cindex line comment character
2175 Anything from the @dfn{line comment} character to the next newline
2176 is considered a comment and is ignored. The line comment character is
2178 @samp{;} on the ARC;
2181 @samp{@@} on the ARM;
2184 @samp{;} for the H8/300 family;
2187 @samp{;} for the HPPA;
2190 @samp{#} on the i386 and x86-64;
2193 @samp{#} on the i960;
2196 @samp{;} for the PDP-11;
2199 @samp{;} for picoJava;
2202 @samp{#} for Motorola PowerPC;
2205 @samp{!} for the Renesas / SuperH SH;
2208 @samp{!} on the SPARC;
2211 @samp{#} on the ip2k;
2214 @samp{#} on the m32c;
2217 @samp{#} on the m32r;
2220 @samp{|} on the 680x0;
2223 @samp{#} on the 68HC11 and 68HC12;
2226 @samp{#} on the Vax;
2229 @samp{;} for the Z80;
2232 @samp{!} for the Z8000;
2235 @samp{#} on the V850;
2238 @samp{#} for Xtensa systems;
2240 see @ref{Machine Dependencies}. @refill
2241 @c FIXME What about i860?
2244 On some machines there are two different line comment characters. One
2245 character only begins a comment if it is the first non-whitespace character on
2246 a line, while the other always begins a comment.
2250 The V850 assembler also supports a double dash as starting a comment that
2251 extends to the end of the line.
2257 @cindex lines starting with @code{#}
2258 @cindex logical line numbers
2259 To be compatible with past assemblers, lines that begin with @samp{#} have a
2260 special interpretation. Following the @samp{#} should be an absolute
2261 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2262 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2263 new logical file name. The rest of the line, if any, should be whitespace.
2265 If the first non-whitespace characters on the line are not numeric,
2266 the line is ignored. (Just like a comment.)
2269 # This is an ordinary comment.
2270 # 42-6 "new_file_name" # New logical file name
2271 # This is logical line # 36.
2273 This feature is deprecated, and may disappear from future versions
2274 of @command{@value{AS}}.
2279 @cindex characters used in symbols
2280 @ifclear SPECIAL-SYMS
2281 A @dfn{symbol} is one or more characters chosen from the set of all
2282 letters (both upper and lower case), digits and the three characters
2288 A @dfn{symbol} is one or more characters chosen from the set of all
2289 letters (both upper and lower case), digits and the three characters
2290 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2296 On most machines, you can also use @code{$} in symbol names; exceptions
2297 are noted in @ref{Machine Dependencies}.
2299 No symbol may begin with a digit. Case is significant.
2300 There is no length limit: all characters are significant. Symbols are
2301 delimited by characters not in that set, or by the beginning of a file
2302 (since the source program must end with a newline, the end of a file is
2303 not a possible symbol delimiter). @xref{Symbols}.
2304 @cindex length of symbols
2309 @cindex statements, structure of
2310 @cindex line separator character
2311 @cindex statement separator character
2313 @ifclear abnormal-separator
2314 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2315 semicolon (@samp{;}). The newline or semicolon is considered part of
2316 the preceding statement. Newlines and semicolons within character
2317 constants are an exception: they do not end statements.
2319 @ifset abnormal-separator
2321 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2322 point (@samp{!}). The newline or exclamation point is considered part of the
2323 preceding statement. Newlines and exclamation points within character
2324 constants are an exception: they do not end statements.
2327 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2328 H8/300) a dollar sign (@samp{$}); or (for the Renesas-SH) a semicolon
2329 (@samp{;}). The newline or separator character is considered part of
2330 the preceding statement. Newlines and separators within character
2331 constants are an exception: they do not end statements.
2336 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2337 separator character. (The line separator is usually @samp{;}, unless this
2338 conflicts with the comment character; see @ref{Machine Dependencies}.) The
2339 newline or separator character is considered part of the preceding
2340 statement. Newlines and separators within character constants are an
2341 exception: they do not end statements.
2344 @cindex newline, required at file end
2345 @cindex EOF, newline must precede
2346 It is an error to end any statement with end-of-file: the last
2347 character of any input file should be a newline.@refill
2349 An empty statement is allowed, and may include whitespace. It is ignored.
2351 @cindex instructions and directives
2352 @cindex directives and instructions
2353 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2354 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2356 A statement begins with zero or more labels, optionally followed by a
2357 key symbol which determines what kind of statement it is. The key
2358 symbol determines the syntax of the rest of the statement. If the
2359 symbol begins with a dot @samp{.} then the statement is an assembler
2360 directive: typically valid for any computer. If the symbol begins with
2361 a letter the statement is an assembly language @dfn{instruction}: it
2362 assembles into a machine language instruction.
2364 Different versions of @command{@value{AS}} for different computers
2365 recognize different instructions. In fact, the same symbol may
2366 represent a different instruction in a different computer's assembly
2370 @cindex @code{:} (label)
2371 @cindex label (@code{:})
2372 A label is a symbol immediately followed by a colon (@code{:}).
2373 Whitespace before a label or after a colon is permitted, but you may not
2374 have whitespace between a label's symbol and its colon. @xref{Labels}.
2377 For HPPA targets, labels need not be immediately followed by a colon, but
2378 the definition of a label must begin in column zero. This also implies that
2379 only one label may be defined on each line.
2383 label: .directive followed by something
2384 another_label: # This is an empty statement.
2385 instruction operand_1, operand_2, @dots{}
2392 A constant is a number, written so that its value is known by
2393 inspection, without knowing any context. Like this:
2396 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2397 .ascii "Ring the bell\7" # A string constant.
2398 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2399 .float 0f-314159265358979323846264338327\
2400 95028841971.693993751E-40 # - pi, a flonum.
2405 * Characters:: Character Constants
2406 * Numbers:: Number Constants
2410 @subsection Character Constants
2412 @cindex character constants
2413 @cindex constants, character
2414 There are two kinds of character constants. A @dfn{character} stands
2415 for one character in one byte and its value may be used in
2416 numeric expressions. String constants (properly called string
2417 @emph{literals}) are potentially many bytes and their values may not be
2418 used in arithmetic expressions.
2422 * Chars:: Characters
2426 @subsubsection Strings
2428 @cindex string constants
2429 @cindex constants, string
2430 A @dfn{string} is written between double-quotes. It may contain
2431 double-quotes or null characters. The way to get special characters
2432 into a string is to @dfn{escape} these characters: precede them with
2433 a backslash @samp{\} character. For example @samp{\\} represents
2434 one backslash: the first @code{\} is an escape which tells
2435 @command{@value{AS}} to interpret the second character literally as a backslash
2436 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2437 escape character). The complete list of escapes follows.
2439 @cindex escape codes, character
2440 @cindex character escape codes
2443 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2445 @cindex @code{\b} (backspace character)
2446 @cindex backspace (@code{\b})
2448 Mnemonic for backspace; for ASCII this is octal code 010.
2451 @c Mnemonic for EOText; for ASCII this is octal code 004.
2453 @cindex @code{\f} (formfeed character)
2454 @cindex formfeed (@code{\f})
2456 Mnemonic for FormFeed; for ASCII this is octal code 014.
2458 @cindex @code{\n} (newline character)
2459 @cindex newline (@code{\n})
2461 Mnemonic for newline; for ASCII this is octal code 012.
2464 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2466 @cindex @code{\r} (carriage return character)
2467 @cindex carriage return (@code{\r})
2469 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2472 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2473 @c other assemblers.
2475 @cindex @code{\t} (tab)
2476 @cindex tab (@code{\t})
2478 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2481 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2482 @c @item \x @var{digit} @var{digit} @var{digit}
2483 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2485 @cindex @code{\@var{ddd}} (octal character code)
2486 @cindex octal character code (@code{\@var{ddd}})
2487 @item \ @var{digit} @var{digit} @var{digit}
2488 An octal character code. The numeric code is 3 octal digits.
2489 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2490 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2492 @cindex @code{\@var{xd...}} (hex character code)
2493 @cindex hex character code (@code{\@var{xd...}})
2494 @item \@code{x} @var{hex-digits...}
2495 A hex character code. All trailing hex digits are combined. Either upper or
2496 lower case @code{x} works.
2498 @cindex @code{\\} (@samp{\} character)
2499 @cindex backslash (@code{\\})
2501 Represents one @samp{\} character.
2504 @c Represents one @samp{'} (accent acute) character.
2505 @c This is needed in single character literals
2506 @c (@xref{Characters,,Character Constants}.) to represent
2509 @cindex @code{\"} (doublequote character)
2510 @cindex doublequote (@code{\"})
2512 Represents one @samp{"} character. Needed in strings to represent
2513 this character, because an unescaped @samp{"} would end the string.
2515 @item \ @var{anything-else}
2516 Any other character when escaped by @kbd{\} gives a warning, but
2517 assembles as if the @samp{\} was not present. The idea is that if
2518 you used an escape sequence you clearly didn't want the literal
2519 interpretation of the following character. However @command{@value{AS}} has no
2520 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2521 code and warns you of the fact.
2524 Which characters are escapable, and what those escapes represent,
2525 varies widely among assemblers. The current set is what we think
2526 the BSD 4.2 assembler recognizes, and is a subset of what most C
2527 compilers recognize. If you are in doubt, do not use an escape
2531 @subsubsection Characters
2533 @cindex single character constant
2534 @cindex character, single
2535 @cindex constant, single character
2536 A single character may be written as a single quote immediately
2537 followed by that character. The same escapes apply to characters as
2538 to strings. So if you want to write the character backslash, you
2539 must write @kbd{'\\} where the first @code{\} escapes the second
2540 @code{\}. As you can see, the quote is an acute accent, not a
2541 grave accent. A newline
2543 @ifclear abnormal-separator
2544 (or semicolon @samp{;})
2546 @ifset abnormal-separator
2548 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2553 immediately following an acute accent is taken as a literal character
2554 and does not count as the end of a statement. The value of a character
2555 constant in a numeric expression is the machine's byte-wide code for
2556 that character. @command{@value{AS}} assumes your character code is ASCII:
2557 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2560 @subsection Number Constants
2562 @cindex constants, number
2563 @cindex number constants
2564 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2565 are stored in the target machine. @emph{Integers} are numbers that
2566 would fit into an @code{int} in the C language. @emph{Bignums} are
2567 integers, but they are stored in more than 32 bits. @emph{Flonums}
2568 are floating point numbers, described below.
2571 * Integers:: Integers
2576 * Bit Fields:: Bit Fields
2582 @subsubsection Integers
2584 @cindex constants, integer
2586 @cindex binary integers
2587 @cindex integers, binary
2588 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2589 the binary digits @samp{01}.
2591 @cindex octal integers
2592 @cindex integers, octal
2593 An octal integer is @samp{0} followed by zero or more of the octal
2594 digits (@samp{01234567}).
2596 @cindex decimal integers
2597 @cindex integers, decimal
2598 A decimal integer starts with a non-zero digit followed by zero or
2599 more digits (@samp{0123456789}).
2601 @cindex hexadecimal integers
2602 @cindex integers, hexadecimal
2603 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2604 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2606 Integers have the usual values. To denote a negative integer, use
2607 the prefix operator @samp{-} discussed under expressions
2608 (@pxref{Prefix Ops,,Prefix Operators}).
2611 @subsubsection Bignums
2614 @cindex constants, bignum
2615 A @dfn{bignum} has the same syntax and semantics as an integer
2616 except that the number (or its negative) takes more than 32 bits to
2617 represent in binary. The distinction is made because in some places
2618 integers are permitted while bignums are not.
2621 @subsubsection Flonums
2623 @cindex floating point numbers
2624 @cindex constants, floating point
2626 @cindex precision, floating point
2627 A @dfn{flonum} represents a floating point number. The translation is
2628 indirect: a decimal floating point number from the text is converted by
2629 @command{@value{AS}} to a generic binary floating point number of more than
2630 sufficient precision. This generic floating point number is converted
2631 to a particular computer's floating point format (or formats) by a
2632 portion of @command{@value{AS}} specialized to that computer.
2634 A flonum is written by writing (in order)
2639 (@samp{0} is optional on the HPPA.)
2643 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2645 @kbd{e} is recommended. Case is not important.
2647 @c FIXME: verify if flonum syntax really this vague for most cases
2648 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2649 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2652 On the H8/300, Renesas / SuperH SH,
2653 and AMD 29K architectures, the letter must be
2654 one of the letters @samp{DFPRSX} (in upper or lower case).
2656 On the ARC, the letter must be one of the letters @samp{DFRS}
2657 (in upper or lower case).
2659 On the Intel 960 architecture, the letter must be
2660 one of the letters @samp{DFT} (in upper or lower case).
2662 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2666 One of the letters @samp{DFRS} (in upper or lower case).
2669 One of the letters @samp{DFPRSX} (in upper or lower case).
2672 The letter @samp{E} (upper case only).
2675 One of the letters @samp{DFT} (in upper or lower case).
2680 An optional sign: either @samp{+} or @samp{-}.
2683 An optional @dfn{integer part}: zero or more decimal digits.
2686 An optional @dfn{fractional part}: @samp{.} followed by zero
2687 or more decimal digits.
2690 An optional exponent, consisting of:
2694 An @samp{E} or @samp{e}.
2695 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2696 @c principle this can perfectly well be different on different targets.
2698 Optional sign: either @samp{+} or @samp{-}.
2700 One or more decimal digits.
2705 At least one of the integer part or the fractional part must be
2706 present. The floating point number has the usual base-10 value.
2708 @command{@value{AS}} does all processing using integers. Flonums are computed
2709 independently of any floating point hardware in the computer running
2710 @command{@value{AS}}.
2714 @c Bit fields are written as a general facility but are also controlled
2715 @c by a conditional-compilation flag---which is as of now (21mar91)
2716 @c turned on only by the i960 config of GAS.
2718 @subsubsection Bit Fields
2721 @cindex constants, bit field
2722 You can also define numeric constants as @dfn{bit fields}.
2723 Specify two numbers separated by a colon---
2725 @var{mask}:@var{value}
2728 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2731 The resulting number is then packed
2733 @c this conditional paren in case bit fields turned on elsewhere than 960
2734 (in host-dependent byte order)
2736 into a field whose width depends on which assembler directive has the
2737 bit-field as its argument. Overflow (a result from the bitwise and
2738 requiring more binary digits to represent) is not an error; instead,
2739 more constants are generated, of the specified width, beginning with the
2740 least significant digits.@refill
2742 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2743 @code{.short}, and @code{.word} accept bit-field arguments.
2748 @chapter Sections and Relocation
2753 * Secs Background:: Background
2754 * Ld Sections:: Linker Sections
2755 * As Sections:: Assembler Internal Sections
2756 * Sub-Sections:: Sub-Sections
2760 @node Secs Background
2763 Roughly, a section is a range of addresses, with no gaps; all data
2764 ``in'' those addresses is treated the same for some particular purpose.
2765 For example there may be a ``read only'' section.
2767 @cindex linker, and assembler
2768 @cindex assembler, and linker
2769 The linker @code{@value{LD}} reads many object files (partial programs) and
2770 combines their contents to form a runnable program. When @command{@value{AS}}
2771 emits an object file, the partial program is assumed to start at address 0.
2772 @code{@value{LD}} assigns the final addresses for the partial program, so that
2773 different partial programs do not overlap. This is actually an
2774 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2777 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2778 addresses. These blocks slide to their run-time addresses as rigid
2779 units; their length does not change and neither does the order of bytes
2780 within them. Such a rigid unit is called a @emph{section}. Assigning
2781 run-time addresses to sections is called @dfn{relocation}. It includes
2782 the task of adjusting mentions of object-file addresses so they refer to
2783 the proper run-time addresses.
2785 For the H8/300, and for the Renesas / SuperH SH,
2786 @command{@value{AS}} pads sections if needed to
2787 ensure they end on a word (sixteen bit) boundary.
2790 @cindex standard assembler sections
2791 An object file written by @command{@value{AS}} has at least three sections, any
2792 of which may be empty. These are named @dfn{text}, @dfn{data} and
2797 When it generates COFF or ELF output,
2799 @command{@value{AS}} can also generate whatever other named sections you specify
2800 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2801 If you do not use any directives that place output in the @samp{.text}
2802 or @samp{.data} sections, these sections still exist, but are empty.
2807 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2809 @command{@value{AS}} can also generate whatever other named sections you
2810 specify using the @samp{.space} and @samp{.subspace} directives. See
2811 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2812 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2813 assembler directives.
2816 Additionally, @command{@value{AS}} uses different names for the standard
2817 text, data, and bss sections when generating SOM output. Program text
2818 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2819 BSS into @samp{$BSS$}.
2823 Within the object file, the text section starts at address @code{0}, the
2824 data section follows, and the bss section follows the data section.
2827 When generating either SOM or ELF output files on the HPPA, the text
2828 section starts at address @code{0}, the data section at address
2829 @code{0x4000000}, and the bss section follows the data section.
2832 To let @code{@value{LD}} know which data changes when the sections are
2833 relocated, and how to change that data, @command{@value{AS}} also writes to the
2834 object file details of the relocation needed. To perform relocation
2835 @code{@value{LD}} must know, each time an address in the object
2839 Where in the object file is the beginning of this reference to
2842 How long (in bytes) is this reference?
2844 Which section does the address refer to? What is the numeric value of
2846 (@var{address}) @minus{} (@var{start-address of section})?
2849 Is the reference to an address ``Program-Counter relative''?
2852 @cindex addresses, format of
2853 @cindex section-relative addressing
2854 In fact, every address @command{@value{AS}} ever uses is expressed as
2856 (@var{section}) + (@var{offset into section})
2859 Further, most expressions @command{@value{AS}} computes have this section-relative
2862 (For some object formats, such as SOM for the HPPA, some expressions are
2863 symbol-relative instead.)
2866 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
2867 @var{N} into section @var{secname}.''
2869 Apart from text, data and bss sections you need to know about the
2870 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
2871 addresses in the absolute section remain unchanged. For example, address
2872 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
2873 @code{@value{LD}}. Although the linker never arranges two partial programs'
2874 data sections with overlapping addresses after linking, @emph{by definition}
2875 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
2876 part of a program is always the same address when the program is running as
2877 address @code{@{absolute@ 239@}} in any other part of the program.
2879 The idea of sections is extended to the @dfn{undefined} section. Any
2880 address whose section is unknown at assembly time is by definition
2881 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
2882 Since numbers are always defined, the only way to generate an undefined
2883 address is to mention an undefined symbol. A reference to a named
2884 common block would be such a symbol: its value is unknown at assembly
2885 time so it has section @emph{undefined}.
2887 By analogy the word @emph{section} is used to describe groups of sections in
2888 the linked program. @code{@value{LD}} puts all partial programs' text
2889 sections in contiguous addresses in the linked program. It is
2890 customary to refer to the @emph{text section} of a program, meaning all
2891 the addresses of all partial programs' text sections. Likewise for
2892 data and bss sections.
2894 Some sections are manipulated by @code{@value{LD}}; others are invented for
2895 use of @command{@value{AS}} and have no meaning except during assembly.
2898 @section Linker Sections
2899 @code{@value{LD}} deals with just four kinds of sections, summarized below.
2904 @cindex named sections
2905 @cindex sections, named
2906 @item named sections
2909 @cindex text section
2910 @cindex data section
2914 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
2915 separate but equal sections. Anything you can say of one section is
2918 When the program is running, however, it is
2919 customary for the text section to be unalterable. The
2920 text section is often shared among processes: it contains
2921 instructions, constants and the like. The data section of a running
2922 program is usually alterable: for example, C variables would be stored
2923 in the data section.
2928 This section contains zeroed bytes when your program begins running. It
2929 is used to hold uninitialized variables or common storage. The length of
2930 each partial program's bss section is important, but because it starts
2931 out containing zeroed bytes there is no need to store explicit zero
2932 bytes in the object file. The bss section was invented to eliminate
2933 those explicit zeros from object files.
2935 @cindex absolute section
2936 @item absolute section
2937 Address 0 of this section is always ``relocated'' to runtime address 0.
2938 This is useful if you want to refer to an address that @code{@value{LD}} must
2939 not change when relocating. In this sense we speak of absolute
2940 addresses being ``unrelocatable'': they do not change during relocation.
2942 @cindex undefined section
2943 @item undefined section
2944 This ``section'' is a catch-all for address references to objects not in
2945 the preceding sections.
2946 @c FIXME: ref to some other doc on obj-file formats could go here.
2949 @cindex relocation example
2950 An idealized example of three relocatable sections follows.
2952 The example uses the traditional section names @samp{.text} and @samp{.data}.
2954 Memory addresses are on the horizontal axis.
2958 @c END TEXI2ROFF-KILL
2961 partial program # 1: |ttttt|dddd|00|
2968 partial program # 2: |TTT|DDD|000|
2971 +--+---+-----+--+----+---+-----+~~
2972 linked program: | |TTT|ttttt| |dddd|DDD|00000|
2973 +--+---+-----+--+----+---+-----+~~
2975 addresses: 0 @dots{}
2982 \line{\it Partial program \#1: \hfil}
2983 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2984 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
2986 \line{\it Partial program \#2: \hfil}
2987 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
2988 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
2990 \line{\it linked program: \hfil}
2991 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
2992 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
2993 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
2994 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
2996 \line{\it addresses: \hfil}
3000 @c END TEXI2ROFF-KILL
3003 @section Assembler Internal Sections
3005 @cindex internal assembler sections
3006 @cindex sections in messages, internal
3007 These sections are meant only for the internal use of @command{@value{AS}}. They
3008 have no meaning at run-time. You do not really need to know about these
3009 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3010 warning messages, so it might be helpful to have an idea of their
3011 meanings to @command{@value{AS}}. These sections are used to permit the
3012 value of every expression in your assembly language program to be a
3013 section-relative address.
3016 @cindex assembler internal logic error
3017 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3018 An internal assembler logic error has been found. This means there is a
3019 bug in the assembler.
3021 @cindex expr (internal section)
3023 The assembler stores complex expression internally as combinations of
3024 symbols. When it needs to represent an expression as a symbol, it puts
3025 it in the expr section.
3027 @c FIXME item transfer[t] vector preload
3028 @c FIXME item transfer[t] vector postload
3029 @c FIXME item register
3033 @section Sub-Sections
3035 @cindex numbered subsections
3036 @cindex grouping data
3042 fall into two sections: text and data.
3044 You may have separate groups of
3046 data in named sections
3050 data in named sections
3056 that you want to end up near to each other in the object file, even though they
3057 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3058 use @dfn{subsections} for this purpose. Within each section, there can be
3059 numbered subsections with values from 0 to 8192. Objects assembled into the
3060 same subsection go into the object file together with other objects in the same
3061 subsection. For example, a compiler might want to store constants in the text
3062 section, but might not want to have them interspersed with the program being
3063 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3064 section of code being output, and a @samp{.text 1} before each group of
3065 constants being output.
3067 Subsections are optional. If you do not use subsections, everything
3068 goes in subsection number zero.
3071 Each subsection is zero-padded up to a multiple of four bytes.
3072 (Subsections may be padded a different amount on different flavors
3073 of @command{@value{AS}}.)
3077 On the H8/300 platform, each subsection is zero-padded to a word
3078 boundary (two bytes).
3079 The same is true on the Renesas SH.
3082 @c FIXME section padding (alignment)?
3083 @c Rich Pixley says padding here depends on target obj code format; that
3084 @c doesn't seem particularly useful to say without further elaboration,
3085 @c so for now I say nothing about it. If this is a generic BFD issue,
3086 @c these paragraphs might need to vanish from this manual, and be
3087 @c discussed in BFD chapter of binutils (or some such).
3091 Subsections appear in your object file in numeric order, lowest numbered
3092 to highest. (All this to be compatible with other people's assemblers.)
3093 The object file contains no representation of subsections; @code{@value{LD}} and
3094 other programs that manipulate object files see no trace of them.
3095 They just see all your text subsections as a text section, and all your
3096 data subsections as a data section.
3098 To specify which subsection you want subsequent statements assembled
3099 into, use a numeric argument to specify it, in a @samp{.text
3100 @var{expression}} or a @samp{.data @var{expression}} statement.
3103 When generating COFF output, you
3108 can also use an extra subsection
3109 argument with arbitrary named sections: @samp{.section @var{name},
3114 When generating ELF output, you
3119 can also use the @code{.subsection} directive (@pxref{SubSection})
3120 to specify a subsection: @samp{.subsection @var{expression}}.
3122 @var{Expression} should be an absolute expression
3123 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3124 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3125 begins in @code{text 0}. For instance:
3127 .text 0 # The default subsection is text 0 anyway.
3128 .ascii "This lives in the first text subsection. *"
3130 .ascii "But this lives in the second text subsection."
3132 .ascii "This lives in the data section,"
3133 .ascii "in the first data subsection."
3135 .ascii "This lives in the first text section,"
3136 .ascii "immediately following the asterisk (*)."
3139 Each section has a @dfn{location counter} incremented by one for every byte
3140 assembled into that section. Because subsections are merely a convenience
3141 restricted to @command{@value{AS}} there is no concept of a subsection location
3142 counter. There is no way to directly manipulate a location counter---but the
3143 @code{.align} directive changes it, and any label definition captures its
3144 current value. The location counter of the section where statements are being
3145 assembled is said to be the @dfn{active} location counter.
3148 @section bss Section
3151 @cindex common variable storage
3152 The bss section is used for local common variable storage.
3153 You may allocate address space in the bss section, but you may
3154 not dictate data to load into it before your program executes. When
3155 your program starts running, all the contents of the bss
3156 section are zeroed bytes.
3158 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3159 @ref{Lcomm,,@code{.lcomm}}.
3161 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3162 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3165 When assembling for a target which supports multiple sections, such as ELF or
3166 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3167 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3168 section. Typically the section will only contain symbol definitions and
3169 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3176 Symbols are a central concept: the programmer uses symbols to name
3177 things, the linker uses symbols to link, and the debugger uses symbols
3181 @cindex debuggers, and symbol order
3182 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3183 the same order they were declared. This may break some debuggers.
3188 * Setting Symbols:: Giving Symbols Other Values
3189 * Symbol Names:: Symbol Names
3190 * Dot:: The Special Dot Symbol
3191 * Symbol Attributes:: Symbol Attributes
3198 A @dfn{label} is written as a symbol immediately followed by a colon
3199 @samp{:}. The symbol then represents the current value of the
3200 active location counter, and is, for example, a suitable instruction
3201 operand. You are warned if you use the same symbol to represent two
3202 different locations: the first definition overrides any other
3206 On the HPPA, the usual form for a label need not be immediately followed by a
3207 colon, but instead must start in column zero. Only one label may be defined on
3208 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3209 provides a special directive @code{.label} for defining labels more flexibly.
3212 @node Setting Symbols
3213 @section Giving Symbols Other Values
3215 @cindex assigning values to symbols
3216 @cindex symbol values, assigning
3217 A symbol can be given an arbitrary value by writing a symbol, followed
3218 by an equals sign @samp{=}, followed by an expression
3219 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3220 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3221 equals sign @samp{=}@samp{=} here represents an equivalent of the
3222 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3225 @section Symbol Names
3227 @cindex symbol names
3228 @cindex names, symbol
3229 @ifclear SPECIAL-SYMS
3230 Symbol names begin with a letter or with one of @samp{._}. On most
3231 machines, you can also use @code{$} in symbol names; exceptions are
3232 noted in @ref{Machine Dependencies}. That character may be followed by any
3233 string of digits, letters, dollar signs (unless otherwise noted for a
3234 particular target machine), and underscores.
3238 Symbol names begin with a letter or with one of @samp{._}. On the
3239 Renesas SH you can also use @code{$} in symbol names. That
3240 character may be followed by any string of digits, letters, dollar signs (save
3241 on the H8/300), and underscores.
3245 Case of letters is significant: @code{foo} is a different symbol name
3248 Each symbol has exactly one name. Each name in an assembly language program
3249 refers to exactly one symbol. You may use that symbol name any number of times
3252 @subheading Local Symbol Names
3254 @cindex local symbol names
3255 @cindex symbol names, local
3256 A local symbol is any symbol beginning with certain local label prefixes.
3257 By default, the local label prefix is @samp{.L} for ELF systems or
3258 @samp{L} for traditional a.out systems, but each target may have its own
3259 set of local label prefixes.
3261 On the HPPA local symbols begin with @samp{L$}.
3264 Local symbols are defined and used within the assembler, but they are
3265 normally not saved in object files. Thus, they are not visible when debugging.
3266 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3267 @option{-L}}) to retain the local symbols in the object files.
3269 @subheading Local Labels
3271 @cindex local labels
3272 @cindex temporary symbol names
3273 @cindex symbol names, temporary
3274 Local labels help compilers and programmers use names temporarily.
3275 They create symbols which are guaranteed to be unique over the entire scope of
3276 the input source code and which can be referred to by a simple notation.
3277 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3278 represents any positive integer). To refer to the most recent previous
3279 definition of that label write @samp{@b{N}b}, using the same number as when
3280 you defined the label. To refer to the next definition of a local label, write
3281 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3284 There is no restriction on how you can use these labels, and you can reuse them
3285 too. So that it is possible to repeatedly define the same local label (using
3286 the same number @samp{@b{N}}), although you can only refer to the most recently
3287 defined local label of that number (for a backwards reference) or the next
3288 definition of a specific local label for a forward reference. It is also worth
3289 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3290 implemented in a slightly more efficient manner than the others.
3301 Which is the equivalent of:
3304 label_1: branch label_3
3305 label_2: branch label_1
3306 label_3: branch label_4
3307 label_4: branch label_3
3310 Local label names are only a notational device. They are immediately
3311 transformed into more conventional symbol names before the assembler uses them.
3312 The symbol names are stored in the symbol table, appear in error messages, and
3313 are optionally emitted to the object file. The names are constructed using
3317 @item @emph{local label prefix}
3318 All local symbols begin with the system-specific local label prefix.
3319 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3320 that start with the local label prefix. These labels are
3321 used for symbols you are never intended to see. If you use the
3322 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3323 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3324 you may use them in debugging.
3327 This is the number that was used in the local label definition. So if the
3328 label is written @samp{55:} then the number is @samp{55}.
3331 This unusual character is included so you do not accidentally invent a symbol
3332 of the same name. The character has ASCII value of @samp{\002} (control-B).
3334 @item @emph{ordinal number}
3335 This is a serial number to keep the labels distinct. The first definition of
3336 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3337 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3338 the number @samp{1} and its 15th definition gets @samp{15} as well.
3341 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3342 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3344 @subheading Dollar Local Labels
3345 @cindex dollar local symbols
3347 @code{@value{AS}} also supports an even more local form of local labels called
3348 dollar labels. These labels go out of scope (i.e., they become undefined) as
3349 soon as a non-local label is defined. Thus they remain valid for only a small
3350 region of the input source code. Normal local labels, by contrast, remain in
3351 scope for the entire file, or until they are redefined by another occurrence of
3352 the same local label.
3354 Dollar labels are defined in exactly the same way as ordinary local labels,
3355 except that instead of being terminated by a colon, they are terminated by a
3356 dollar sign, e.g., @samp{@b{55$}}.
3358 They can also be distinguished from ordinary local labels by their transformed
3359 names which use ASCII character @samp{\001} (control-A) as the magic character
3360 to distinguish them from ordinary labels. For example, the fifth definition of
3361 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3364 @section The Special Dot Symbol
3366 @cindex dot (symbol)
3367 @cindex @code{.} (symbol)
3368 @cindex current address
3369 @cindex location counter
3370 The special symbol @samp{.} refers to the current address that
3371 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3372 .long .} defines @code{melvin} to contain its own address.
3373 Assigning a value to @code{.} is treated the same as a @code{.org}
3374 directive. Thus, the expression @samp{.=.+4} is the same as saying
3375 @ifclear no-space-dir
3379 @node Symbol Attributes
3380 @section Symbol Attributes
3382 @cindex symbol attributes
3383 @cindex attributes, symbol
3384 Every symbol has, as well as its name, the attributes ``Value'' and
3385 ``Type''. Depending on output format, symbols can also have auxiliary
3388 The detailed definitions are in @file{a.out.h}.
3391 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3392 all these attributes, and probably won't warn you. This makes the
3393 symbol an externally defined symbol, which is generally what you
3397 * Symbol Value:: Value
3398 * Symbol Type:: Type
3401 * a.out Symbols:: Symbol Attributes: @code{a.out}
3405 * a.out Symbols:: Symbol Attributes: @code{a.out}
3408 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3413 * COFF Symbols:: Symbol Attributes for COFF
3416 * SOM Symbols:: Symbol Attributes for SOM
3423 @cindex value of a symbol
3424 @cindex symbol value
3425 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3426 location in the text, data, bss or absolute sections the value is the
3427 number of addresses from the start of that section to the label.
3428 Naturally for text, data and bss sections the value of a symbol changes
3429 as @code{@value{LD}} changes section base addresses during linking. Absolute
3430 symbols' values do not change during linking: that is why they are
3433 The value of an undefined symbol is treated in a special way. If it is
3434 0 then the symbol is not defined in this assembler source file, and
3435 @code{@value{LD}} tries to determine its value from other files linked into the
3436 same program. You make this kind of symbol simply by mentioning a symbol
3437 name without defining it. A non-zero value represents a @code{.comm}
3438 common declaration. The value is how much common storage to reserve, in
3439 bytes (addresses). The symbol refers to the first address of the
3445 @cindex type of a symbol
3447 The type attribute of a symbol contains relocation (section)
3448 information, any flag settings indicating that a symbol is external, and
3449 (optionally), other information for linkers and debuggers. The exact
3450 format depends on the object-code output format in use.
3455 @c The following avoids a "widow" subsection title. @group would be
3456 @c better if it were available outside examples.
3459 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3461 @cindex @code{b.out} symbol attributes
3462 @cindex symbol attributes, @code{b.out}
3463 These symbol attributes appear only when @command{@value{AS}} is configured for
3464 one of the Berkeley-descended object output formats---@code{a.out} or
3470 @subsection Symbol Attributes: @code{a.out}
3472 @cindex @code{a.out} symbol attributes
3473 @cindex symbol attributes, @code{a.out}
3479 @subsection Symbol Attributes: @code{a.out}
3481 @cindex @code{a.out} symbol attributes
3482 @cindex symbol attributes, @code{a.out}
3486 * Symbol Desc:: Descriptor
3487 * Symbol Other:: Other
3491 @subsubsection Descriptor
3493 @cindex descriptor, of @code{a.out} symbol
3494 This is an arbitrary 16-bit value. You may establish a symbol's
3495 descriptor value by using a @code{.desc} statement
3496 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3497 @command{@value{AS}}.
3500 @subsubsection Other
3502 @cindex other attribute, of @code{a.out} symbol
3503 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3508 @subsection Symbol Attributes for COFF
3510 @cindex COFF symbol attributes
3511 @cindex symbol attributes, COFF
3513 The COFF format supports a multitude of auxiliary symbol attributes;
3514 like the primary symbol attributes, they are set between @code{.def} and
3515 @code{.endef} directives.
3517 @subsubsection Primary Attributes
3519 @cindex primary attributes, COFF symbols
3520 The symbol name is set with @code{.def}; the value and type,
3521 respectively, with @code{.val} and @code{.type}.
3523 @subsubsection Auxiliary Attributes
3525 @cindex auxiliary attributes, COFF symbols
3526 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3527 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3528 table information for COFF.
3533 @subsection Symbol Attributes for SOM
3535 @cindex SOM symbol attributes
3536 @cindex symbol attributes, SOM
3538 The SOM format for the HPPA supports a multitude of symbol attributes set with
3539 the @code{.EXPORT} and @code{.IMPORT} directives.
3541 The attributes are described in @cite{HP9000 Series 800 Assembly
3542 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3543 @code{EXPORT} assembler directive documentation.
3547 @chapter Expressions
3551 @cindex numeric values
3552 An @dfn{expression} specifies an address or numeric value.
3553 Whitespace may precede and/or follow an expression.
3555 The result of an expression must be an absolute number, or else an offset into
3556 a particular section. If an expression is not absolute, and there is not
3557 enough information when @command{@value{AS}} sees the expression to know its
3558 section, a second pass over the source program might be necessary to interpret
3559 the expression---but the second pass is currently not implemented.
3560 @command{@value{AS}} aborts with an error message in this situation.
3563 * Empty Exprs:: Empty Expressions
3564 * Integer Exprs:: Integer Expressions
3568 @section Empty Expressions
3570 @cindex empty expressions
3571 @cindex expressions, empty
3572 An empty expression has no value: it is just whitespace or null.
3573 Wherever an absolute expression is required, you may omit the
3574 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3575 is compatible with other assemblers.
3578 @section Integer Expressions
3580 @cindex integer expressions
3581 @cindex expressions, integer
3582 An @dfn{integer expression} is one or more @emph{arguments} delimited
3583 by @emph{operators}.
3586 * Arguments:: Arguments
3587 * Operators:: Operators
3588 * Prefix Ops:: Prefix Operators
3589 * Infix Ops:: Infix Operators
3593 @subsection Arguments
3595 @cindex expression arguments
3596 @cindex arguments in expressions
3597 @cindex operands in expressions
3598 @cindex arithmetic operands
3599 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3600 contexts arguments are sometimes called ``arithmetic operands''. In
3601 this manual, to avoid confusing them with the ``instruction operands'' of
3602 the machine language, we use the term ``argument'' to refer to parts of
3603 expressions only, reserving the word ``operand'' to refer only to machine
3604 instruction operands.
3606 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3607 @var{section} is one of text, data, bss, absolute,
3608 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3611 Numbers are usually integers.
3613 A number can be a flonum or bignum. In this case, you are warned
3614 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3615 these 32 bits are an integer. You may write integer-manipulating
3616 instructions that act on exotic constants, compatible with other
3619 @cindex subexpressions
3620 Subexpressions are a left parenthesis @samp{(} followed by an integer
3621 expression, followed by a right parenthesis @samp{)}; or a prefix
3622 operator followed by an argument.
3625 @subsection Operators
3627 @cindex operators, in expressions
3628 @cindex arithmetic functions
3629 @cindex functions, in expressions
3630 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3631 operators are followed by an argument. Infix operators appear
3632 between their arguments. Operators may be preceded and/or followed by
3636 @subsection Prefix Operator
3638 @cindex prefix operators
3639 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3640 one argument, which must be absolute.
3642 @c the tex/end tex stuff surrounding this small table is meant to make
3643 @c it align, on the printed page, with the similar table in the next
3644 @c section (which is inside an enumerate).
3646 \global\advance\leftskip by \itemindent
3651 @dfn{Negation}. Two's complement negation.
3653 @dfn{Complementation}. Bitwise not.
3657 \global\advance\leftskip by -\itemindent
3661 @subsection Infix Operators
3663 @cindex infix operators
3664 @cindex operators, permitted arguments
3665 @dfn{Infix operators} take two arguments, one on either side. Operators
3666 have precedence, but operations with equal precedence are performed left
3667 to right. Apart from @code{+} or @option{-}, both arguments must be
3668 absolute, and the result is absolute.
3671 @cindex operator precedence
3672 @cindex precedence of operators
3679 @dfn{Multiplication}.
3682 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3688 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3691 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3695 Intermediate precedence
3700 @dfn{Bitwise Inclusive Or}.
3706 @dfn{Bitwise Exclusive Or}.
3709 @dfn{Bitwise Or Not}.
3716 @cindex addition, permitted arguments
3717 @cindex plus, permitted arguments
3718 @cindex arguments for addition
3720 @dfn{Addition}. If either argument is absolute, the result has the section of
3721 the other argument. You may not add together arguments from different
3724 @cindex subtraction, permitted arguments
3725 @cindex minus, permitted arguments
3726 @cindex arguments for subtraction
3728 @dfn{Subtraction}. If the right argument is absolute, the
3729 result has the section of the left argument.
3730 If both arguments are in the same section, the result is absolute.
3731 You may not subtract arguments from different sections.
3732 @c FIXME is there still something useful to say about undefined - undefined ?
3734 @cindex comparison expressions
3735 @cindex expressions, comparison
3740 @dfn{Is Not Equal To}
3744 @dfn{Is Greater Than}
3746 @dfn{Is Greater Than Or Equal To}
3748 @dfn{Is Less Than Or Equal To}
3750 The comparison operators can be used as infix operators. A true results has a
3751 value of -1 whereas a false result has a value of 0. Note, these operators
3752 perform signed comparisons.
3755 @item Lowest Precedence
3764 These two logical operations can be used to combine the results of sub
3765 expressions. Note, unlike the comparison operators a true result returns a
3766 value of 1 but a false results does still return 0. Also note that the logical
3767 or operator has a slightly lower precedence than logical and.
3772 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3773 address; you can only have a defined section in one of the two arguments.
3776 @chapter Assembler Directives
3778 @cindex directives, machine independent
3779 @cindex pseudo-ops, machine independent
3780 @cindex machine independent directives
3781 All assembler directives have names that begin with a period (@samp{.}).
3782 The rest of the name is letters, usually in lower case.
3784 This chapter discusses directives that are available regardless of the
3785 target machine configuration for the @sc{gnu} assembler.
3787 Some machine configurations provide additional directives.
3788 @xref{Machine Dependencies}.
3791 @ifset machine-directives
3792 @xref{Machine Dependencies}, for additional directives.
3797 * Abort:: @code{.abort}
3799 * ABORT (COFF):: @code{.ABORT}
3802 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3803 * Altmacro:: @code{.altmacro}
3804 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3805 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3806 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3807 * Byte:: @code{.byte @var{expressions}}
3808 * Comm:: @code{.comm @var{symbol} , @var{length} }
3810 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
3812 * Data:: @code{.data @var{subsection}}
3814 * Def:: @code{.def @var{name}}
3817 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3823 * Double:: @code{.double @var{flonums}}
3824 * Eject:: @code{.eject}
3825 * Else:: @code{.else}
3826 * Elseif:: @code{.elseif}
3829 * Endef:: @code{.endef}
3832 * Endfunc:: @code{.endfunc}
3833 * Endif:: @code{.endif}
3834 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3835 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3836 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
3838 * Error:: @code{.error @var{string}}
3839 * Exitm:: @code{.exitm}
3840 * Extern:: @code{.extern}
3841 * Fail:: @code{.fail}
3842 @ifclear no-file-dir
3843 * File:: @code{.file @var{string}}
3846 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
3847 * Float:: @code{.float @var{flonums}}
3848 * Func:: @code{.func}
3849 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
3851 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
3852 * Hidden:: @code{.hidden @var{names}}
3855 * hword:: @code{.hword @var{expressions}}
3856 * Ident:: @code{.ident}
3857 * If:: @code{.if @var{absolute expression}}
3858 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
3859 * Include:: @code{.include "@var{file}"}
3860 * Int:: @code{.int @var{expressions}}
3862 * Internal:: @code{.internal @var{names}}
3865 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
3866 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
3867 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
3868 * Lflags:: @code{.lflags}
3869 @ifclear no-line-dir
3870 * Line:: @code{.line @var{line-number}}
3873 * Linkonce:: @code{.linkonce [@var{type}]}
3874 * List:: @code{.list}
3875 * Ln:: @code{.ln @var{line-number}}
3877 * LNS directives:: @code{.file}, @code{.loc}, etc.
3879 * Long:: @code{.long @var{expressions}}
3881 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
3884 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
3885 * MRI:: @code{.mri @var{val}}
3886 * Noaltmacro:: @code{.noaltmacro}
3887 * Nolist:: @code{.nolist}
3888 * Octa:: @code{.octa @var{bignums}}
3889 * Org:: @code{.org @var{new-lc}, @var{fill}}
3890 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3892 * PopSection:: @code{.popsection}
3893 * Previous:: @code{.previous}
3896 * Print:: @code{.print @var{string}}
3898 * Protected:: @code{.protected @var{names}}
3901 * Psize:: @code{.psize @var{lines}, @var{columns}}
3902 * Purgem:: @code{.purgem @var{name}}
3904 * PushSection:: @code{.pushsection @var{name}}
3907 * Quad:: @code{.quad @var{bignums}}
3908 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
3909 * Rept:: @code{.rept @var{count}}
3910 * Sbttl:: @code{.sbttl "@var{subheading}"}
3912 * Scl:: @code{.scl @var{class}}
3915 * Section:: @code{.section @var{name}[, @var{flags}]}
3918 * Set:: @code{.set @var{symbol}, @var{expression}}
3919 * Short:: @code{.short @var{expressions}}
3920 * Single:: @code{.single @var{flonums}}
3922 * Size:: @code{.size [@var{name} , @var{expression}]}
3925 * Skip:: @code{.skip @var{size} , @var{fill}}
3926 * Sleb128:: @code{.sleb128 @var{expressions}}
3927 * Space:: @code{.space @var{size} , @var{fill}}
3929 * Stab:: @code{.stabd, .stabn, .stabs}
3932 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
3933 * Struct:: @code{.struct @var{expression}}
3935 * SubSection:: @code{.subsection}
3936 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
3940 * Tag:: @code{.tag @var{structname}}
3943 * Text:: @code{.text @var{subsection}}
3944 * Title:: @code{.title "@var{heading}"}
3946 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
3949 * Uleb128:: @code{.uleb128 @var{expressions}}
3951 * Val:: @code{.val @var{addr}}
3955 * Version:: @code{.version "@var{string}"}
3956 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
3957 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
3960 * Warning:: @code{.warning @var{string}}
3961 * Weak:: @code{.weak @var{names}}
3962 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
3963 * Word:: @code{.word @var{expressions}}
3964 * Deprecated:: Deprecated Directives
3968 @section @code{.abort}
3970 @cindex @code{abort} directive
3971 @cindex stopping the assembly
3972 This directive stops the assembly immediately. It is for
3973 compatibility with other assemblers. The original idea was that the
3974 assembly language source would be piped into the assembler. If the sender
3975 of the source quit, it could use this directive tells @command{@value{AS}} to
3976 quit also. One day @code{.abort} will not be supported.
3980 @section @code{.ABORT} (COFF)
3982 @cindex @code{ABORT} directive
3983 When producing COFF output, @command{@value{AS}} accepts this directive as a
3984 synonym for @samp{.abort}.
3987 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
3993 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
3995 @cindex padding the location counter
3996 @cindex @code{align} directive
3997 Pad the location counter (in the current subsection) to a particular storage
3998 boundary. The first expression (which must be absolute) is the alignment
3999 required, as described below.
4001 The second expression (also absolute) gives the fill value to be stored in the
4002 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4003 padding bytes are normally zero. However, on some systems, if the section is
4004 marked as containing code and the fill value is omitted, the space is filled
4005 with no-op instructions.
4007 The third expression is also absolute, and is also optional. If it is present,
4008 it is the maximum number of bytes that should be skipped by this alignment
4009 directive. If doing the alignment would require skipping more bytes than the
4010 specified maximum, then the alignment is not done at all. You can omit the
4011 fill value (the second argument) entirely by simply using two commas after the
4012 required alignment; this can be useful if you want the alignment to be filled
4013 with no-op instructions when appropriate.
4015 The way the required alignment is specified varies from system to system.
4016 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4017 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4018 alignment request in bytes. For example @samp{.align 8} advances
4019 the location counter until it is a multiple of 8. If the location counter
4020 is already a multiple of 8, no change is needed. For the tic54x, the
4021 first expression is the alignment request in words.
4023 For other systems, including the i386 using a.out format, and the arm and
4024 strongarm, it is the
4025 number of low-order zero bits the location counter must have after
4026 advancement. For example @samp{.align 3} advances the location
4027 counter until it a multiple of 8. If the location counter is already a
4028 multiple of 8, no change is needed.
4030 This inconsistency is due to the different behaviors of the various
4031 native assemblers for these systems which GAS must emulate.
4032 GAS also provides @code{.balign} and @code{.p2align} directives,
4033 described later, which have a consistent behavior across all
4034 architectures (but are specific to GAS).
4037 @section @code{.ascii "@var{string}"}@dots{}
4039 @cindex @code{ascii} directive
4040 @cindex string literals
4041 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4042 separated by commas. It assembles each string (with no automatic
4043 trailing zero byte) into consecutive addresses.
4046 @section @code{.asciz "@var{string}"}@dots{}
4048 @cindex @code{asciz} directive
4049 @cindex zero-terminated strings
4050 @cindex null-terminated strings
4051 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4052 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4055 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4057 @cindex padding the location counter given number of bytes
4058 @cindex @code{balign} directive
4059 Pad the location counter (in the current subsection) to a particular
4060 storage boundary. The first expression (which must be absolute) is the
4061 alignment request in bytes. For example @samp{.balign 8} advances
4062 the location counter until it is a multiple of 8. If the location counter
4063 is already a multiple of 8, no change is needed.
4065 The second expression (also absolute) gives the fill value to be stored in the
4066 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4067 padding bytes are normally zero. However, on some systems, if the section is
4068 marked as containing code and the fill value is omitted, the space is filled
4069 with no-op instructions.
4071 The third expression is also absolute, and is also optional. If it is present,
4072 it is the maximum number of bytes that should be skipped by this alignment
4073 directive. If doing the alignment would require skipping more bytes than the
4074 specified maximum, then the alignment is not done at all. You can omit the
4075 fill value (the second argument) entirely by simply using two commas after the
4076 required alignment; this can be useful if you want the alignment to be filled
4077 with no-op instructions when appropriate.
4079 @cindex @code{balignw} directive
4080 @cindex @code{balignl} directive
4081 The @code{.balignw} and @code{.balignl} directives are variants of the
4082 @code{.balign} directive. The @code{.balignw} directive treats the fill
4083 pattern as a two byte word value. The @code{.balignl} directives treats the
4084 fill pattern as a four byte longword value. For example, @code{.balignw
4085 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4086 filled in with the value 0x368d (the exact placement of the bytes depends upon
4087 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4091 @section @code{.byte @var{expressions}}
4093 @cindex @code{byte} directive
4094 @cindex integers, one byte
4095 @code{.byte} expects zero or more expressions, separated by commas.
4096 Each expression is assembled into the next byte.
4099 @section @code{.comm @var{symbol} , @var{length} }
4101 @cindex @code{comm} directive
4102 @cindex symbol, common
4103 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4104 common symbol in one object file may be merged with a defined or common symbol
4105 of the same name in another object file. If @code{@value{LD}} does not see a
4106 definition for the symbol--just one or more common symbols--then it will
4107 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4108 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4109 the same name, and they do not all have the same size, it will allocate space
4110 using the largest size.
4113 When using ELF, the @code{.comm} directive takes an optional third argument.
4114 This is the desired alignment of the symbol, specified as a byte boundary (for
4115 example, an alignment of 16 means that the least significant 4 bits of the
4116 address should be zero). The alignment must be an absolute expression, and it
4117 must be a power of two. If @code{@value{LD}} allocates uninitialized memory
4118 for the common symbol, it will use the alignment when placing the symbol. If
4119 no alignment is specified, @command{@value{AS}} will set the alignment to the
4120 largest power of two less than or equal to the size of the symbol, up to a
4125 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4126 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4129 @node CFI directives
4130 @section @code{.cfi_startproc [simple]}
4131 @cindex @code{cfi_startproc} directive
4132 @code{.cfi_startproc} is used at the beginning of each function that
4133 should have an entry in @code{.eh_frame}. It initializes some internal
4134 data structures. Don't forget to close the function by
4135 @code{.cfi_endproc}.
4137 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4138 it also emits some architecture dependent initial CFI instructions.
4140 @section @code{.cfi_endproc}
4141 @cindex @code{cfi_endproc} directive
4142 @code{.cfi_endproc} is used at the end of a function where it closes its
4143 unwind entry previously opened by
4144 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4146 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4147 @code{.cfi_personality} defines personality routine and its encoding.
4148 @var{encoding} must be a constant determining how the personality
4149 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4150 argument is not present, otherwise second argument should be
4151 a constant or a symbol name. When using indirect encodings,
4152 the symbol provided should be the location where personality
4153 can be loaded from, not the personality routine itself.
4154 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4155 no personality routine.
4157 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4158 @code{.cfi_lsda} defines LSDA and its encoding.
4159 @var{encoding} must be a constant determining how the LSDA
4160 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4161 argument is not present, otherwise second argument should be a constant
4162 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4165 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4166 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4167 address from @var{register} and add @var{offset} to it}.
4169 @section @code{.cfi_def_cfa_register @var{register}}
4170 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4171 now on @var{register} will be used instead of the old one. Offset
4174 @section @code{.cfi_def_cfa_offset @var{offset}}
4175 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4176 remains the same, but @var{offset} is new. Note that it is the
4177 absolute offset that will be added to a defined register to compute
4180 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4181 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4182 value that is added/substracted from the previous offset.
4184 @section @code{.cfi_offset @var{register}, @var{offset}}
4185 Previous value of @var{register} is saved at offset @var{offset} from
4188 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4189 Previous value of @var{register} is saved at offset @var{offset} from
4190 the current CFA register. This is transformed to @code{.cfi_offset}
4191 using the known displacement of the CFA register from the CFA.
4192 This is often easier to use, because the number will match the
4193 code it's annotating.
4195 @section @code{.cfi_register @var{register1}, @var{register2}}
4196 Previous value of @var{register1} is saved in register @var{register2}.
4198 @section @code{.cfi_restore @var{register}}
4199 @code{.cfi_restore} says that the rule for @var{register} is now the
4200 same as it was at the beginning of the function, after all initial
4201 instruction added by @code{.cfi_startproc} were executed.
4203 @section @code{.cfi_undefined @var{register}}
4204 From now on the previous value of @var{register} can't be restored anymore.
4206 @section @code{.cfi_same_value @var{register}}
4207 Current value of @var{register} is the same like in the previous frame,
4208 i.e. no restoration needed.
4210 @section @code{.cfi_remember_state},
4211 First save all current rules for all registers by @code{.cfi_remember_state},
4212 then totally screw them up by subsequent @code{.cfi_*} directives and when
4213 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4214 the previous saved state.
4216 @section @code{.cfi_return_column @var{register}}
4217 Change return column @var{register}, i.e. the return address is either
4218 directly in @var{register} or can be accessed by rules for @var{register}.
4220 @section @code{.cfi_signal_frame}
4221 Mark current function as signal trampoline.
4223 @section @code{.cfi_window_save}
4224 SPARC register window has been saved.
4226 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4227 Allows the user to add arbitrary bytes to the unwind info. One
4228 might use this to add OS-specific CFI opcodes, or generic CFI
4229 opcodes that GAS does not yet support.
4231 @node LNS directives
4232 @section @code{.file @var{fileno} @var{filename}}
4233 @cindex @code{file} directive
4234 When emitting dwarf2 line number information @code{.file} assigns filenames
4235 to the @code{.debug_line} file name table. The @var{fileno} operand should
4236 be a unique positive integer to use as the index of the entry in the table.
4237 The @var{filename} operand is a C string literal.
4239 The detail of filename indices is exposed to the user because the filename
4240 table is shared with the @code{.debug_info} section of the dwarf2 debugging
4241 information, and thus the user must know the exact indices that table
4244 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
4245 @cindex @code{loc} directive
4246 The @code{.loc} directive will add row to the @code{.debug_line} line
4247 number matrix corresponding to the immediately following assembly
4248 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
4249 arguments will be applied to the @code{.debug_line} state machine before
4252 The @var{options} are a sequence of the following tokens in any order:
4256 This option will set the @code{basic_block} register in the
4257 @code{.debug_line} state machine to @code{true}.
4260 This option will set the @code{prologue_end} register in the
4261 @code{.debug_line} state machine to @code{true}.
4263 @item epilogue_begin
4264 This option will set the @code{epilogue_begin} register in the
4265 @code{.debug_line} state machine to @code{true}.
4267 @item is_stmt @var{value}
4268 This option will set the @code{is_stmt} register in the
4269 @code{.debug_line} state machine to @code{value}, which must be
4272 @item isa @var{value}
4273 This directive will set the @code{isa} register in the @code{.debug_line}
4274 state machine to @var{value}, which must be an unsigned integer.
4278 @section @code{.loc_mark_labels @var{enable}}
4279 @cindex @code{loc_mark_labels} directive
4280 The @code{.loc_mark_labels} directive makes the assembler emit an entry
4281 to the @code{.debug_line} line number matrix with the @code{basic_block}
4282 register in the state machine set whenever a code label is seen.
4283 The @var{enable} argument should be either 1 or 0, to enable or disable
4284 this function respectively.
4287 @section @code{.data @var{subsection}}
4289 @cindex @code{data} directive
4290 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4291 end of the data subsection numbered @var{subsection} (which is an
4292 absolute expression). If @var{subsection} is omitted, it defaults
4297 @section @code{.def @var{name}}
4299 @cindex @code{def} directive
4300 @cindex COFF symbols, debugging
4301 @cindex debugging COFF symbols
4302 Begin defining debugging information for a symbol @var{name}; the
4303 definition extends until the @code{.endef} directive is encountered.
4306 This directive is only observed when @command{@value{AS}} is configured for COFF
4307 format output; when producing @code{b.out}, @samp{.def} is recognized,
4314 @section @code{.desc @var{symbol}, @var{abs-expression}}
4316 @cindex @code{desc} directive
4317 @cindex COFF symbol descriptor
4318 @cindex symbol descriptor, COFF
4319 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4320 to the low 16 bits of an absolute expression.
4323 The @samp{.desc} directive is not available when @command{@value{AS}} is
4324 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4325 object format. For the sake of compatibility, @command{@value{AS}} accepts
4326 it, but produces no output, when configured for COFF.
4332 @section @code{.dim}
4334 @cindex @code{dim} directive
4335 @cindex COFF auxiliary symbol information
4336 @cindex auxiliary symbol information, COFF
4337 This directive is generated by compilers to include auxiliary debugging
4338 information in the symbol table. It is only permitted inside
4339 @code{.def}/@code{.endef} pairs.
4342 @samp{.dim} is only meaningful when generating COFF format output; when
4343 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4349 @section @code{.double @var{flonums}}
4351 @cindex @code{double} directive
4352 @cindex floating point numbers (double)
4353 @code{.double} expects zero or more flonums, separated by commas. It
4354 assembles floating point numbers.
4356 The exact kind of floating point numbers emitted depends on how
4357 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4361 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4362 in @sc{ieee} format.
4367 @section @code{.eject}
4369 @cindex @code{eject} directive
4370 @cindex new page, in listings
4371 @cindex page, in listings
4372 @cindex listing control: new page
4373 Force a page break at this point, when generating assembly listings.
4376 @section @code{.else}
4378 @cindex @code{else} directive
4379 @code{.else} is part of the @command{@value{AS}} support for conditional
4380 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4381 of code to be assembled if the condition for the preceding @code{.if}
4385 @section @code{.elseif}
4387 @cindex @code{elseif} directive
4388 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4389 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4390 @code{.if} block that would otherwise fill the entire @code{.else} section.
4393 @section @code{.end}
4395 @cindex @code{end} directive
4396 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4397 process anything in the file past the @code{.end} directive.
4401 @section @code{.endef}
4403 @cindex @code{endef} directive
4404 This directive flags the end of a symbol definition begun with
4408 @samp{.endef} is only meaningful when generating COFF format output; if
4409 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4410 directive but ignores it.
4415 @section @code{.endfunc}
4416 @cindex @code{endfunc} directive
4417 @code{.endfunc} marks the end of a function specified with @code{.func}.
4420 @section @code{.endif}
4422 @cindex @code{endif} directive
4423 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4424 it marks the end of a block of code that is only assembled
4425 conditionally. @xref{If,,@code{.if}}.
4428 @section @code{.equ @var{symbol}, @var{expression}}
4430 @cindex @code{equ} directive
4431 @cindex assigning values to symbols
4432 @cindex symbols, assigning values to
4433 This directive sets the value of @var{symbol} to @var{expression}.
4434 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4437 The syntax for @code{equ} on the HPPA is
4438 @samp{@var{symbol} .equ @var{expression}}.
4442 The syntax for @code{equ} on the Z80 is
4443 @samp{@var{symbol} equ @var{expression}}.
4444 On the Z80 it is an eror if @var{symbol} is already defined,
4445 but the symbol is not protected from later redefinition.
4446 Compare @ref{Equiv}.
4450 @section @code{.equiv @var{symbol}, @var{expression}}
4451 @cindex @code{equiv} directive
4452 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4453 the assembler will signal an error if @var{symbol} is already defined. Note a
4454 symbol which has been referenced but not actually defined is considered to be
4457 Except for the contents of the error message, this is roughly equivalent to
4464 plus it protects the symbol from later redefinition.
4467 @section @code{.eqv @var{symbol}, @var{expression}}
4468 @cindex @code{eqv} directive
4469 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4470 evaluate the expression or any part of it immediately. Instead each time
4471 the resulting symbol is used in an expression, a snapshot of its current
4475 @section @code{.err}
4476 @cindex @code{err} directive
4477 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4478 message and, unless the @option{-Z} option was used, it will not generate an
4479 object file. This can be used to signal an error in conditionally compiled code.
4482 @section @code{.error "@var{string}"}
4483 @cindex error directive
4485 Similarly to @code{.err}, this directive emits an error, but you can specify a
4486 string that will be emitted as the error message. If you don't specify the
4487 message, it defaults to @code{".error directive invoked in source file"}.
4488 @xref{Errors, ,Error and Warning Messages}.
4491 .error "This code has not been assembled and tested."
4495 @section @code{.exitm}
4496 Exit early from the current macro definition. @xref{Macro}.
4499 @section @code{.extern}
4501 @cindex @code{extern} directive
4502 @code{.extern} is accepted in the source program---for compatibility
4503 with other assemblers---but it is ignored. @command{@value{AS}} treats
4504 all undefined symbols as external.
4507 @section @code{.fail @var{expression}}
4509 @cindex @code{fail} directive
4510 Generates an error or a warning. If the value of the @var{expression} is 500
4511 or more, @command{@value{AS}} will print a warning message. If the value is less
4512 than 500, @command{@value{AS}} will print an error message. The message will
4513 include the value of @var{expression}. This can occasionally be useful inside
4514 complex nested macros or conditional assembly.
4516 @ifclear no-file-dir
4518 @section @code{.file @var{string}}
4520 @cindex @code{file} directive
4521 @cindex logical file name
4522 @cindex file name, logical
4523 @code{.file} tells @command{@value{AS}} that we are about to start a new logical
4524 file. @var{string} is the new file name. In general, the filename is
4525 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4526 to specify an empty file name, you must give the quotes--@code{""}. This
4527 statement may go away in future: it is only recognized to be compatible with
4528 old @command{@value{AS}} programs.
4532 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4534 @cindex @code{fill} directive
4535 @cindex writing patterns in memory
4536 @cindex patterns, writing in memory
4537 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4538 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4539 may be zero or more. @var{Size} may be zero or more, but if it is
4540 more than 8, then it is deemed to have the value 8, compatible with
4541 other people's assemblers. The contents of each @var{repeat} bytes
4542 is taken from an 8-byte number. The highest order 4 bytes are
4543 zero. The lowest order 4 bytes are @var{value} rendered in the
4544 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4545 Each @var{size} bytes in a repetition is taken from the lowest order
4546 @var{size} bytes of this number. Again, this bizarre behavior is
4547 compatible with other people's assemblers.
4549 @var{size} and @var{value} are optional.
4550 If the second comma and @var{value} are absent, @var{value} is
4551 assumed zero. If the first comma and following tokens are absent,
4552 @var{size} is assumed to be 1.
4555 @section @code{.float @var{flonums}}
4557 @cindex floating point numbers (single)
4558 @cindex @code{float} directive
4559 This directive assembles zero or more flonums, separated by commas. It
4560 has the same effect as @code{.single}.
4562 The exact kind of floating point numbers emitted depends on how
4563 @command{@value{AS}} is configured.
4564 @xref{Machine Dependencies}.
4568 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4569 in @sc{ieee} format.
4574 @section @code{.func @var{name}[,@var{label}]}
4575 @cindex @code{func} directive
4576 @code{.func} emits debugging information to denote function @var{name}, and
4577 is ignored unless the file is assembled with debugging enabled.
4578 Only @samp{--gstabs[+]} is currently supported.
4579 @var{label} is the entry point of the function and if omitted @var{name}
4580 prepended with the @samp{leading char} is used.
4581 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4582 All functions are currently defined to have @code{void} return type.
4583 The function must be terminated with @code{.endfunc}.
4586 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4588 @cindex @code{global} directive
4589 @cindex symbol, making visible to linker
4590 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4591 @var{symbol} in your partial program, its value is made available to
4592 other partial programs that are linked with it. Otherwise,
4593 @var{symbol} takes its attributes from a symbol of the same name
4594 from another file linked into the same program.
4596 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4597 compatibility with other assemblers.
4600 On the HPPA, @code{.global} is not always enough to make it accessible to other
4601 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4602 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4607 @section @code{.gnu_attribute @var{tag},@var{value}}
4608 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
4611 @section @code{.hidden @var{names}}
4613 @cindex @code{hidden} directive
4615 This is one of the ELF visibility directives. The other two are
4616 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4617 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4619 This directive overrides the named symbols default visibility (which is set by
4620 their binding: local, global or weak). The directive sets the visibility to
4621 @code{hidden} which means that the symbols are not visible to other components.
4622 Such symbols are always considered to be @code{protected} as well.
4626 @section @code{.hword @var{expressions}}
4628 @cindex @code{hword} directive
4629 @cindex integers, 16-bit
4630 @cindex numbers, 16-bit
4631 @cindex sixteen bit integers
4632 This expects zero or more @var{expressions}, and emits
4633 a 16 bit number for each.
4636 This directive is a synonym for @samp{.short}; depending on the target
4637 architecture, it may also be a synonym for @samp{.word}.
4641 This directive is a synonym for @samp{.short}.
4644 This directive is a synonym for both @samp{.short} and @samp{.word}.
4649 @section @code{.ident}
4651 @cindex @code{ident} directive
4653 This directive is used by some assemblers to place tags in object files. The
4654 behavior of this directive varies depending on the target. When using the
4655 a.out object file format, @command{@value{AS}} simply accepts the directive for
4656 source-file compatibility with existing assemblers, but does not emit anything
4657 for it. When using COFF, comments are emitted to the @code{.comment} or
4658 @code{.rdata} section, depending on the target. When using ELF, comments are
4659 emitted to the @code{.comment} section.
4662 @section @code{.if @var{absolute expression}}
4664 @cindex conditional assembly
4665 @cindex @code{if} directive
4666 @code{.if} marks the beginning of a section of code which is only
4667 considered part of the source program being assembled if the argument
4668 (which must be an @var{absolute expression}) is non-zero. The end of
4669 the conditional section of code must be marked by @code{.endif}
4670 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4671 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4672 If you have several conditions to check, @code{.elseif} may be used to avoid
4673 nesting blocks if/else within each subsequent @code{.else} block.
4675 The following variants of @code{.if} are also supported:
4677 @cindex @code{ifdef} directive
4678 @item .ifdef @var{symbol}
4679 Assembles the following section of code if the specified @var{symbol}
4680 has been defined. Note a symbol which has been referenced but not yet defined
4681 is considered to be undefined.
4683 @cindex @code{ifb} directive
4684 @item .ifb @var{text}
4685 Assembles the following section of code if the operand is blank (empty).
4687 @cindex @code{ifc} directive
4688 @item .ifc @var{string1},@var{string2}
4689 Assembles the following section of code if the two strings are the same. The
4690 strings may be optionally quoted with single quotes. If they are not quoted,
4691 the first string stops at the first comma, and the second string stops at the
4692 end of the line. Strings which contain whitespace should be quoted. The
4693 string comparison is case sensitive.
4695 @cindex @code{ifeq} directive
4696 @item .ifeq @var{absolute expression}
4697 Assembles the following section of code if the argument is zero.
4699 @cindex @code{ifeqs} directive
4700 @item .ifeqs @var{string1},@var{string2}
4701 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4703 @cindex @code{ifge} directive
4704 @item .ifge @var{absolute expression}
4705 Assembles the following section of code if the argument is greater than or
4708 @cindex @code{ifgt} directive
4709 @item .ifgt @var{absolute expression}
4710 Assembles the following section of code if the argument is greater than zero.
4712 @cindex @code{ifle} directive
4713 @item .ifle @var{absolute expression}
4714 Assembles the following section of code if the argument is less than or equal
4717 @cindex @code{iflt} directive
4718 @item .iflt @var{absolute expression}
4719 Assembles the following section of code if the argument is less than zero.
4721 @cindex @code{ifnb} directive
4722 @item .ifnb @var{text}
4723 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
4724 following section of code if the operand is non-blank (non-empty).
4726 @cindex @code{ifnc} directive
4727 @item .ifnc @var{string1},@var{string2}.
4728 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4729 following section of code if the two strings are not the same.
4731 @cindex @code{ifndef} directive
4732 @cindex @code{ifnotdef} directive
4733 @item .ifndef @var{symbol}
4734 @itemx .ifnotdef @var{symbol}
4735 Assembles the following section of code if the specified @var{symbol}
4736 has not been defined. Both spelling variants are equivalent. Note a symbol
4737 which has been referenced but not yet defined is considered to be undefined.
4739 @cindex @code{ifne} directive
4740 @item .ifne @var{absolute expression}
4741 Assembles the following section of code if the argument is not equal to zero
4742 (in other words, this is equivalent to @code{.if}).
4744 @cindex @code{ifnes} directive
4745 @item .ifnes @var{string1},@var{string2}
4746 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4747 following section of code if the two strings are not the same.
4751 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4753 @cindex @code{incbin} directive
4754 @cindex binary files, including
4755 The @code{incbin} directive includes @var{file} verbatim at the current
4756 location. You can control the search paths used with the @samp{-I} command-line
4757 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4760 The @var{skip} argument skips a number of bytes from the start of the
4761 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4762 read. Note that the data is not aligned in any way, so it is the user's
4763 responsibility to make sure that proper alignment is provided both before and
4764 after the @code{incbin} directive.
4767 @section @code{.include "@var{file}"}
4769 @cindex @code{include} directive
4770 @cindex supporting files, including
4771 @cindex files, including
4772 This directive provides a way to include supporting files at specified
4773 points in your source program. The code from @var{file} is assembled as
4774 if it followed the point of the @code{.include}; when the end of the
4775 included file is reached, assembly of the original file continues. You
4776 can control the search paths used with the @samp{-I} command-line option
4777 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4781 @section @code{.int @var{expressions}}
4783 @cindex @code{int} directive
4784 @cindex integers, 32-bit
4785 Expect zero or more @var{expressions}, of any section, separated by commas.
4786 For each expression, emit a number that, at run time, is the value of that
4787 expression. The byte order and bit size of the number depends on what kind
4788 of target the assembly is for.
4792 On most forms of the H8/300, @code{.int} emits 16-bit
4793 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
4800 @section @code{.internal @var{names}}
4802 @cindex @code{internal} directive
4804 This is one of the ELF visibility directives. The other two are
4805 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
4806 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4808 This directive overrides the named symbols default visibility (which is set by
4809 their binding: local, global or weak). The directive sets the visibility to
4810 @code{internal} which means that the symbols are considered to be @code{hidden}
4811 (i.e., not visible to other components), and that some extra, processor specific
4812 processing must also be performed upon the symbols as well.
4816 @section @code{.irp @var{symbol},@var{values}}@dots{}
4818 @cindex @code{irp} directive
4819 Evaluate a sequence of statements assigning different values to @var{symbol}.
4820 The sequence of statements starts at the @code{.irp} directive, and is
4821 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
4822 set to @var{value}, and the sequence of statements is assembled. If no
4823 @var{value} is listed, the sequence of statements is assembled once, with
4824 @var{symbol} set to the null string. To refer to @var{symbol} within the
4825 sequence of statements, use @var{\symbol}.
4827 For example, assembling
4835 is equivalent to assembling
4843 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
4846 @section @code{.irpc @var{symbol},@var{values}}@dots{}
4848 @cindex @code{irpc} directive
4849 Evaluate a sequence of statements assigning different values to @var{symbol}.
4850 The sequence of statements starts at the @code{.irpc} directive, and is
4851 terminated by an @code{.endr} directive. For each character in @var{value},
4852 @var{symbol} is set to the character, and the sequence of statements is
4853 assembled. If no @var{value} is listed, the sequence of statements is
4854 assembled once, with @var{symbol} set to the null string. To refer to
4855 @var{symbol} within the sequence of statements, use @var{\symbol}.
4857 For example, assembling
4865 is equivalent to assembling
4873 For some caveats with the spelling of @var{symbol}, see also the discussion
4877 @section @code{.lcomm @var{symbol} , @var{length}}
4879 @cindex @code{lcomm} directive
4880 @cindex local common symbols
4881 @cindex symbols, local common
4882 Reserve @var{length} (an absolute expression) bytes for a local common
4883 denoted by @var{symbol}. The section and value of @var{symbol} are
4884 those of the new local common. The addresses are allocated in the bss
4885 section, so that at run-time the bytes start off zeroed. @var{Symbol}
4886 is not declared global (@pxref{Global,,@code{.global}}), so is normally
4887 not visible to @code{@value{LD}}.
4890 Some targets permit a third argument to be used with @code{.lcomm}. This
4891 argument specifies the desired alignment of the symbol in the bss section.
4895 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
4896 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
4900 @section @code{.lflags}
4902 @cindex @code{lflags} directive (ignored)
4903 @command{@value{AS}} accepts this directive, for compatibility with other
4904 assemblers, but ignores it.
4906 @ifclear no-line-dir
4908 @section @code{.line @var{line-number}}
4910 @cindex @code{line} directive
4914 @section @code{.ln @var{line-number}}
4916 @cindex @code{ln} directive
4918 @cindex logical line number
4920 Change the logical line number. @var{line-number} must be an absolute
4921 expression. The next line has that logical line number. Therefore any other
4922 statements on the current line (after a statement separator character) are
4923 reported as on logical line number @var{line-number} @minus{} 1. One day
4924 @command{@value{AS}} will no longer support this directive: it is recognized only
4925 for compatibility with existing assembler programs.
4929 @ifclear no-line-dir
4930 Even though this is a directive associated with the @code{a.out} or
4931 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
4932 when producing COFF output, and treats @samp{.line} as though it
4933 were the COFF @samp{.ln} @emph{if} it is found outside a
4934 @code{.def}/@code{.endef} pair.
4936 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
4937 used by compilers to generate auxiliary symbol information for
4942 @section @code{.linkonce [@var{type}]}
4944 @cindex @code{linkonce} directive
4945 @cindex common sections
4946 Mark the current section so that the linker only includes a single copy of it.
4947 This may be used to include the same section in several different object files,
4948 but ensure that the linker will only include it once in the final output file.
4949 The @code{.linkonce} pseudo-op must be used for each instance of the section.
4950 Duplicate sections are detected based on the section name, so it should be
4953 This directive is only supported by a few object file formats; as of this
4954 writing, the only object file format which supports it is the Portable
4955 Executable format used on Windows NT.
4957 The @var{type} argument is optional. If specified, it must be one of the
4958 following strings. For example:
4962 Not all types may be supported on all object file formats.
4966 Silently discard duplicate sections. This is the default.
4969 Warn if there are duplicate sections, but still keep only one copy.
4972 Warn if any of the duplicates have different sizes.
4975 Warn if any of the duplicates do not have exactly the same contents.
4979 @section @code{.ln @var{line-number}}
4981 @cindex @code{ln} directive
4982 @ifclear no-line-dir
4983 @samp{.ln} is a synonym for @samp{.line}.
4986 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
4987 must be an absolute expression. The next line has that logical
4988 line number, so any other statements on the current line (after a
4989 statement separator character @code{;}) are reported as on logical
4990 line number @var{line-number} @minus{} 1.
4993 This directive is accepted, but ignored, when @command{@value{AS}} is
4994 configured for @code{b.out}; its effect is only associated with COFF
5000 @section @code{.mri @var{val}}
5002 @cindex @code{mri} directive
5003 @cindex MRI mode, temporarily
5004 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5005 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5006 affects code assembled until the next @code{.mri} directive, or until the end
5007 of the file. @xref{M, MRI mode, MRI mode}.
5010 @section @code{.list}
5012 @cindex @code{list} directive
5013 @cindex listing control, turning on
5014 Control (in conjunction with the @code{.nolist} directive) whether or
5015 not assembly listings are generated. These two directives maintain an
5016 internal counter (which is zero initially). @code{.list} increments the
5017 counter, and @code{.nolist} decrements it. Assembly listings are
5018 generated whenever the counter is greater than zero.
5020 By default, listings are disabled. When you enable them (with the
5021 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5022 the initial value of the listing counter is one.
5025 @section @code{.long @var{expressions}}
5027 @cindex @code{long} directive
5028 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5031 @c no one seems to know what this is for or whether this description is
5032 @c what it really ought to do
5034 @section @code{.lsym @var{symbol}, @var{expression}}
5036 @cindex @code{lsym} directive
5037 @cindex symbol, not referenced in assembly
5038 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5039 the hash table, ensuring it cannot be referenced by name during the
5040 rest of the assembly. This sets the attributes of the symbol to be
5041 the same as the expression value:
5043 @var{other} = @var{descriptor} = 0
5044 @var{type} = @r{(section of @var{expression})}
5045 @var{value} = @var{expression}
5048 The new symbol is not flagged as external.
5052 @section @code{.macro}
5055 The commands @code{.macro} and @code{.endm} allow you to define macros that
5056 generate assembly output. For example, this definition specifies a macro
5057 @code{sum} that puts a sequence of numbers into memory:
5060 .macro sum from=0, to=5
5069 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5081 @item .macro @var{macname}
5082 @itemx .macro @var{macname} @var{macargs} @dots{}
5083 @cindex @code{macro} directive
5084 Begin the definition of a macro called @var{macname}. If your macro
5085 definition requires arguments, specify their names after the macro name,
5086 separated by commas or spaces. You can qualify the macro argument to
5087 indicate whether all invocations must specify a non-blank value (through
5088 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5089 (through @samp{:@code{vararg}}). You can supply a default value for any
5090 macro argument by following the name with @samp{=@var{deflt}}. You
5091 cannot define two macros with the same @var{macname} unless it has been
5092 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5093 definitions. For example, these are all valid @code{.macro} statements:
5097 Begin the definition of a macro called @code{comm}, which takes no
5100 @item .macro plus1 p, p1
5101 @itemx .macro plus1 p p1
5102 Either statement begins the definition of a macro called @code{plus1},
5103 which takes two arguments; within the macro definition, write
5104 @samp{\p} or @samp{\p1} to evaluate the arguments.
5106 @item .macro reserve_str p1=0 p2
5107 Begin the definition of a macro called @code{reserve_str}, with two
5108 arguments. The first argument has a default value, but not the second.
5109 After the definition is complete, you can call the macro either as
5110 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5111 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5112 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5113 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5115 @item .macro m p1:req, p2=0, p3:vararg
5116 Begin the definition of a macro called @code{m}, with at least three
5117 arguments. The first argument must always have a value specified, but
5118 not the second, which instead has a default value. The third formal
5119 will get assigned all remaining arguments specified at invocation time.
5121 When you call a macro, you can specify the argument values either by
5122 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5123 @samp{sum to=17, from=9}.
5127 Note that since each of the @var{macargs} can be an identifier exactly
5128 as any other one permitted by the target architecture, there may be
5129 occasional problems if the target hand-crafts special meanings to certain
5130 characters when they occur in a special position. For example, if the colon
5131 (@code{:}) is generally permitted to be part of a symbol name, but the
5132 architecture specific code special-cases it when occurring as the final
5133 character of a symbol (to denote a label), then the macro parameter
5134 replacement code will have no way of knowing that and consider the whole
5135 construct (including the colon) an identifier, and check only this
5136 identifier for being the subject to parameter substitution. So for example
5137 this macro definition:
5145 might not work as expected. Invoking @samp{label foo} might not create a label
5146 called @samp{foo} but instead just insert the text @samp{\l:} into the
5147 assembler source, probably generating an error about an unrecognised
5150 Similarly problems might occur with the period character (@samp{.})
5151 which is often allowed inside opcode names (and hence identifier names). So
5152 for example constructing a macro to build an opcode from a base name and a
5153 length specifier like this:
5156 .macro opcode base length
5161 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5162 instruction but instead generate some kind of error as the assembler tries to
5163 interpret the text @samp{\base.\length}.
5165 There are several possible ways around this problem:
5168 @item Insert white space
5169 If it is possible to use white space characters then this is the simplest
5178 @item Use @samp{\()}
5179 The string @samp{\()} can be used to separate the end of a macro argument from
5180 the following text. eg:
5183 .macro opcode base length
5188 @item Use the alternate macro syntax mode
5189 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5190 used as a separator. eg:
5200 Note: this problem of correctly identifying string parameters to pseudo ops
5201 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5202 and @code{.irpc} (@pxref{Irpc}) as well.
5205 @cindex @code{endm} directive
5206 Mark the end of a macro definition.
5209 @cindex @code{exitm} directive
5210 Exit early from the current macro definition.
5212 @cindex number of macros executed
5213 @cindex macros, count executed
5215 @command{@value{AS}} maintains a counter of how many macros it has
5216 executed in this pseudo-variable; you can copy that number to your
5217 output with @samp{\@@}, but @emph{only within a macro definition}.
5219 @item LOCAL @var{name} [ , @dots{} ]
5220 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5221 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5222 @xref{Altmacro,,@code{.altmacro}}.
5226 @section @code{.altmacro}
5227 Enable alternate macro mode, enabling:
5230 @item LOCAL @var{name} [ , @dots{} ]
5231 One additional directive, @code{LOCAL}, is available. It is used to
5232 generate a string replacement for each of the @var{name} arguments, and
5233 replace any instances of @var{name} in each macro expansion. The
5234 replacement string is unique in the assembly, and different for each
5235 separate macro expansion. @code{LOCAL} allows you to write macros that
5236 define symbols, without fear of conflict between separate macro expansions.
5238 @item String delimiters
5239 You can write strings delimited in these other ways besides
5240 @code{"@var{string}"}:
5243 @item '@var{string}'
5244 You can delimit strings with single-quote characters.
5246 @item <@var{string}>
5247 You can delimit strings with matching angle brackets.
5250 @item single-character string escape
5251 To include any single character literally in a string (even if the
5252 character would otherwise have some special meaning), you can prefix the
5253 character with @samp{!} (an exclamation mark). For example, you can
5254 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
5256 @item Expression results as strings
5257 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
5258 and use the result as a string.
5262 @section @code{.noaltmacro}
5263 Disable alternate macro mode. @xref{Altmacro}.
5266 @section @code{.nolist}
5268 @cindex @code{nolist} directive
5269 @cindex listing control, turning off
5270 Control (in conjunction with the @code{.list} directive) whether or
5271 not assembly listings are generated. These two directives maintain an
5272 internal counter (which is zero initially). @code{.list} increments the
5273 counter, and @code{.nolist} decrements it. Assembly listings are
5274 generated whenever the counter is greater than zero.
5277 @section @code{.octa @var{bignums}}
5279 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5280 @cindex @code{octa} directive
5281 @cindex integer, 16-byte
5282 @cindex sixteen byte integer
5283 This directive expects zero or more bignums, separated by commas. For each
5284 bignum, it emits a 16-byte integer.
5286 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5287 hence @emph{octa}-word for 16 bytes.
5290 @section @code{.org @var{new-lc} , @var{fill}}
5292 @cindex @code{org} directive
5293 @cindex location counter, advancing
5294 @cindex advancing location counter
5295 @cindex current address, advancing
5296 Advance the location counter of the current section to
5297 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5298 expression with the same section as the current subsection. That is,
5299 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5300 wrong section, the @code{.org} directive is ignored. To be compatible
5301 with former assemblers, if the section of @var{new-lc} is absolute,
5302 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5303 is the same as the current subsection.
5305 @code{.org} may only increase the location counter, or leave it
5306 unchanged; you cannot use @code{.org} to move the location counter
5309 @c double negative used below "not undefined" because this is a specific
5310 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5311 @c section. doc@cygnus.com 18feb91
5312 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5313 may not be undefined. If you really detest this restriction we eagerly await
5314 a chance to share your improved assembler.
5316 Beware that the origin is relative to the start of the section, not
5317 to the start of the subsection. This is compatible with other
5318 people's assemblers.
5320 When the location counter (of the current subsection) is advanced, the
5321 intervening bytes are filled with @var{fill} which should be an
5322 absolute expression. If the comma and @var{fill} are omitted,
5323 @var{fill} defaults to zero.
5326 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5328 @cindex padding the location counter given a power of two
5329 @cindex @code{p2align} directive
5330 Pad the location counter (in the current subsection) to a particular
5331 storage boundary. The first expression (which must be absolute) is the
5332 number of low-order zero bits the location counter must have after
5333 advancement. For example @samp{.p2align 3} advances the location
5334 counter until it a multiple of 8. If the location counter is already a
5335 multiple of 8, no change is needed.
5337 The second expression (also absolute) gives the fill value to be stored in the
5338 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5339 padding bytes are normally zero. However, on some systems, if the section is
5340 marked as containing code and the fill value is omitted, the space is filled
5341 with no-op instructions.
5343 The third expression is also absolute, and is also optional. If it is present,
5344 it is the maximum number of bytes that should be skipped by this alignment
5345 directive. If doing the alignment would require skipping more bytes than the
5346 specified maximum, then the alignment is not done at all. You can omit the
5347 fill value (the second argument) entirely by simply using two commas after the
5348 required alignment; this can be useful if you want the alignment to be filled
5349 with no-op instructions when appropriate.
5351 @cindex @code{p2alignw} directive
5352 @cindex @code{p2alignl} directive
5353 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5354 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5355 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5356 fill pattern as a four byte longword value. For example, @code{.p2alignw
5357 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5358 filled in with the value 0x368d (the exact placement of the bytes depends upon
5359 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5364 @section @code{.previous}
5366 @cindex @code{previous} directive
5367 @cindex Section Stack
5368 This is one of the ELF section stack manipulation directives. The others are
5369 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5370 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5371 (@pxref{PopSection}).
5373 This directive swaps the current section (and subsection) with most recently
5374 referenced section/subsection pair prior to this one. Multiple
5375 @code{.previous} directives in a row will flip between two sections (and their
5376 subsections). For example:
5388 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5394 # Now in section A subsection 1
5398 # Now in section B subsection 0
5401 # Now in section B subsection 1
5404 # Now in section B subsection 0
5408 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5409 section B and 0x9abc into subsection 1 of section B.
5411 In terms of the section stack, this directive swaps the current section with
5412 the top section on the section stack.
5417 @section @code{.popsection}
5419 @cindex @code{popsection} directive
5420 @cindex Section Stack
5421 This is one of the ELF section stack manipulation directives. The others are
5422 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5423 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5426 This directive replaces the current section (and subsection) with the top
5427 section (and subsection) on the section stack. This section is popped off the
5432 @section @code{.print @var{string}}
5434 @cindex @code{print} directive
5435 @command{@value{AS}} will print @var{string} on the standard output during
5436 assembly. You must put @var{string} in double quotes.
5440 @section @code{.protected @var{names}}
5442 @cindex @code{protected} directive
5444 This is one of the ELF visibility directives. The other two are
5445 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5447 This directive overrides the named symbols default visibility (which is set by
5448 their binding: local, global or weak). The directive sets the visibility to
5449 @code{protected} which means that any references to the symbols from within the
5450 components that defines them must be resolved to the definition in that
5451 component, even if a definition in another component would normally preempt
5456 @section @code{.psize @var{lines} , @var{columns}}
5458 @cindex @code{psize} directive
5459 @cindex listing control: paper size
5460 @cindex paper size, for listings
5461 Use this directive to declare the number of lines---and, optionally, the
5462 number of columns---to use for each page, when generating listings.
5464 If you do not use @code{.psize}, listings use a default line-count
5465 of 60. You may omit the comma and @var{columns} specification; the
5466 default width is 200 columns.
5468 @command{@value{AS}} generates formfeeds whenever the specified number of
5469 lines is exceeded (or whenever you explicitly request one, using
5472 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5473 those explicitly specified with @code{.eject}.
5476 @section @code{.purgem @var{name}}
5478 @cindex @code{purgem} directive
5479 Undefine the macro @var{name}, so that later uses of the string will not be
5480 expanded. @xref{Macro}.
5484 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5486 @cindex @code{pushsection} directive
5487 @cindex Section Stack
5488 This is one of the ELF section stack manipulation directives. The others are
5489 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5490 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5493 This directive pushes the current section (and subsection) onto the
5494 top of the section stack, and then replaces the current section and
5495 subsection with @code{name} and @code{subsection}. The optional
5496 @code{flags}, @code{type} and @code{arguments} are treated the same
5497 as in the @code{.section} (@pxref{Section}) directive.
5501 @section @code{.quad @var{bignums}}
5503 @cindex @code{quad} directive
5504 @code{.quad} expects zero or more bignums, separated by commas. For
5505 each bignum, it emits
5507 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5508 warning message; and just takes the lowest order 8 bytes of the bignum.
5509 @cindex eight-byte integer
5510 @cindex integer, 8-byte
5512 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5513 hence @emph{quad}-word for 8 bytes.
5516 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5517 warning message; and just takes the lowest order 16 bytes of the bignum.
5518 @cindex sixteen-byte integer
5519 @cindex integer, 16-byte
5523 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5525 @cindex @code{reloc} directive
5526 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5527 @var{expression}. If @var{offset} is a number, the relocation is generated in
5528 the current section. If @var{offset} is an expression that resolves to a
5529 symbol plus offset, the relocation is generated in the given symbol's section.
5530 @var{expression}, if present, must resolve to a symbol plus addend or to an
5531 absolute value, but note that not all targets support an addend. e.g. ELF REL
5532 targets such as i386 store an addend in the section contents rather than in the
5533 relocation. This low level interface does not support addends stored in the
5537 @section @code{.rept @var{count}}
5539 @cindex @code{rept} directive
5540 Repeat the sequence of lines between the @code{.rept} directive and the next
5541 @code{.endr} directive @var{count} times.
5543 For example, assembling
5551 is equivalent to assembling
5560 @section @code{.sbttl "@var{subheading}"}
5562 @cindex @code{sbttl} directive
5563 @cindex subtitles for listings
5564 @cindex listing control: subtitle
5565 Use @var{subheading} as the title (third line, immediately after the
5566 title line) when generating assembly listings.
5568 This directive affects subsequent pages, as well as the current page if
5569 it appears within ten lines of the top of a page.
5573 @section @code{.scl @var{class}}
5575 @cindex @code{scl} directive
5576 @cindex symbol storage class (COFF)
5577 @cindex COFF symbol storage class
5578 Set the storage-class value for a symbol. This directive may only be
5579 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5580 whether a symbol is static or external, or it may record further
5581 symbolic debugging information.
5584 The @samp{.scl} directive is primarily associated with COFF output; when
5585 configured to generate @code{b.out} output format, @command{@value{AS}}
5586 accepts this directive but ignores it.
5592 @section @code{.section @var{name}}
5594 @cindex named section
5595 Use the @code{.section} directive to assemble the following code into a section
5598 This directive is only supported for targets that actually support arbitrarily
5599 named sections; on @code{a.out} targets, for example, it is not accepted, even
5600 with a standard @code{a.out} section name.
5604 @c only print the extra heading if both COFF and ELF are set
5605 @subheading COFF Version
5608 @cindex @code{section} directive (COFF version)
5609 For COFF targets, the @code{.section} directive is used in one of the following
5613 .section @var{name}[, "@var{flags}"]
5614 .section @var{name}[, @var{subsection}]
5617 If the optional argument is quoted, it is taken as flags to use for the
5618 section. Each flag is a single character. The following flags are recognized:
5621 bss section (uninitialized data)
5623 section is not loaded
5633 shared section (meaningful for PE targets)
5635 ignored. (For compatibility with the ELF version)
5638 If no flags are specified, the default flags depend upon the section name. If
5639 the section name is not recognized, the default will be for the section to be
5640 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5641 from the section, rather than adding them, so if they are used on their own it
5642 will be as if no flags had been specified at all.
5644 If the optional argument to the @code{.section} directive is not quoted, it is
5645 taken as a subsection number (@pxref{Sub-Sections}).
5650 @c only print the extra heading if both COFF and ELF are set
5651 @subheading ELF Version
5654 @cindex Section Stack
5655 This is one of the ELF section stack manipulation directives. The others are
5656 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5657 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5658 @code{.previous} (@pxref{Previous}).
5660 @cindex @code{section} directive (ELF version)
5661 For ELF targets, the @code{.section} directive is used like this:
5664 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
5667 The optional @var{flags} argument is a quoted string which may contain any
5668 combination of the following characters:
5671 section is allocatable
5675 section is executable
5677 section is mergeable
5679 section contains zero terminated strings
5681 section is a member of a section group
5683 section is used for thread-local-storage
5686 The optional @var{type} argument may contain one of the following constants:
5689 section contains data
5691 section does not contain data (i.e., section only occupies space)
5693 section contains data which is used by things other than the program
5695 section contains an array of pointers to init functions
5697 section contains an array of pointers to finish functions
5698 @item @@preinit_array
5699 section contains an array of pointers to pre-init functions
5702 Many targets only support the first three section types.
5704 Note on targets where the @code{@@} character is the start of a comment (eg
5705 ARM) then another character is used instead. For example the ARM port uses the
5708 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5709 be specified as well as an extra argument---@var{entsize}---like this:
5712 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5715 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5716 constants, each @var{entsize} octets long. Sections with both @code{M} and
5717 @code{S} must contain zero terminated strings where each character is
5718 @var{entsize} bytes long. The linker may remove duplicates within sections with
5719 the same name, same entity size and same flags. @var{entsize} must be an
5720 absolute expression.
5722 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5723 be present along with an additional field like this:
5726 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5729 The @var{GroupName} field specifies the name of the section group to which this
5730 particular section belongs. The optional linkage field can contain:
5733 indicates that only one copy of this section should be retained
5738 Note: if both the @var{M} and @var{G} flags are present then the fields for
5739 the Merge flag should come first, like this:
5742 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5745 If no flags are specified, the default flags depend upon the section name. If
5746 the section name is not recognized, the default will be for the section to have
5747 none of the above flags: it will not be allocated in memory, nor writable, nor
5748 executable. The section will contain data.
5750 For ELF targets, the assembler supports another type of @code{.section}
5751 directive for compatibility with the Solaris assembler:
5754 .section "@var{name}"[, @var{flags}...]
5757 Note that the section name is quoted. There may be a sequence of comma
5761 section is allocatable
5765 section is executable
5767 section is used for thread local storage
5770 This directive replaces the current section and subsection. See the
5771 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
5772 some examples of how this directive and the other section stack directives
5778 @section @code{.set @var{symbol}, @var{expression}}
5780 @cindex @code{set} directive
5781 @cindex symbol value, setting
5782 Set the value of @var{symbol} to @var{expression}. This
5783 changes @var{symbol}'s value and type to conform to
5784 @var{expression}. If @var{symbol} was flagged as external, it remains
5785 flagged (@pxref{Symbol Attributes}).
5787 You may @code{.set} a symbol many times in the same assembly.
5789 If you @code{.set} a global symbol, the value stored in the object
5790 file is the last value stored into it.
5793 The syntax for @code{set} on the HPPA is
5794 @samp{@var{symbol} .set @var{expression}}.
5798 On Z80 @code{set} is a real instruction, use
5799 @samp{@var{symbol} defl @var{expression}} instead.
5803 @section @code{.short @var{expressions}}
5805 @cindex @code{short} directive
5807 @code{.short} is normally the same as @samp{.word}.
5808 @xref{Word,,@code{.word}}.
5810 In some configurations, however, @code{.short} and @code{.word} generate
5811 numbers of different lengths. @xref{Machine Dependencies}.
5815 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
5818 This expects zero or more @var{expressions}, and emits
5819 a 16 bit number for each.
5824 @section @code{.single @var{flonums}}
5826 @cindex @code{single} directive
5827 @cindex floating point numbers (single)
5828 This directive assembles zero or more flonums, separated by commas. It
5829 has the same effect as @code{.float}.
5831 The exact kind of floating point numbers emitted depends on how
5832 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5836 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
5837 numbers in @sc{ieee} format.
5843 @section @code{.size}
5845 This directive is used to set the size associated with a symbol.
5849 @c only print the extra heading if both COFF and ELF are set
5850 @subheading COFF Version
5853 @cindex @code{size} directive (COFF version)
5854 For COFF targets, the @code{.size} directive is only permitted inside
5855 @code{.def}/@code{.endef} pairs. It is used like this:
5858 .size @var{expression}
5862 @samp{.size} is only meaningful when generating COFF format output; when
5863 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5870 @c only print the extra heading if both COFF and ELF are set
5871 @subheading ELF Version
5874 @cindex @code{size} directive (ELF version)
5875 For ELF targets, the @code{.size} directive is used like this:
5878 .size @var{name} , @var{expression}
5881 This directive sets the size associated with a symbol @var{name}.
5882 The size in bytes is computed from @var{expression} which can make use of label
5883 arithmetic. This directive is typically used to set the size of function
5889 @section @code{.sleb128 @var{expressions}}
5891 @cindex @code{sleb128} directive
5892 @var{sleb128} stands for ``signed little endian base 128.'' This is a
5893 compact, variable length representation of numbers used by the DWARF
5894 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
5896 @ifclear no-space-dir
5898 @section @code{.skip @var{size} , @var{fill}}
5900 @cindex @code{skip} directive
5901 @cindex filling memory
5902 This directive emits @var{size} bytes, each of value @var{fill}. Both
5903 @var{size} and @var{fill} are absolute expressions. If the comma and
5904 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
5908 @section @code{.space @var{size} , @var{fill}}
5910 @cindex @code{space} directive
5911 @cindex filling memory
5912 This directive emits @var{size} bytes, each of value @var{fill}. Both
5913 @var{size} and @var{fill} are absolute expressions. If the comma
5914 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
5919 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
5920 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
5921 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
5922 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
5930 @section @code{.stabd, .stabn, .stabs}
5932 @cindex symbolic debuggers, information for
5933 @cindex @code{stab@var{x}} directives
5934 There are three directives that begin @samp{.stab}.
5935 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
5936 The symbols are not entered in the @command{@value{AS}} hash table: they
5937 cannot be referenced elsewhere in the source file.
5938 Up to five fields are required:
5942 This is the symbol's name. It may contain any character except
5943 @samp{\000}, so is more general than ordinary symbol names. Some
5944 debuggers used to code arbitrarily complex structures into symbol names
5948 An absolute expression. The symbol's type is set to the low 8 bits of
5949 this expression. Any bit pattern is permitted, but @code{@value{LD}}
5950 and debuggers choke on silly bit patterns.
5953 An absolute expression. The symbol's ``other'' attribute is set to the
5954 low 8 bits of this expression.
5957 An absolute expression. The symbol's descriptor is set to the low 16
5958 bits of this expression.
5961 An absolute expression which becomes the symbol's value.
5964 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
5965 or @code{.stabs} statement, the symbol has probably already been created;
5966 you get a half-formed symbol in your object file. This is
5967 compatible with earlier assemblers!
5970 @cindex @code{stabd} directive
5971 @item .stabd @var{type} , @var{other} , @var{desc}
5973 The ``name'' of the symbol generated is not even an empty string.
5974 It is a null pointer, for compatibility. Older assemblers used a
5975 null pointer so they didn't waste space in object files with empty
5978 The symbol's value is set to the location counter,
5979 relocatably. When your program is linked, the value of this symbol
5980 is the address of the location counter when the @code{.stabd} was
5983 @cindex @code{stabn} directive
5984 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
5985 The name of the symbol is set to the empty string @code{""}.
5987 @cindex @code{stabs} directive
5988 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
5989 All five fields are specified.
5995 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
5996 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
5998 @cindex string, copying to object file
5999 @cindex string8, copying to object file
6000 @cindex string16, copying to object file
6001 @cindex string32, copying to object file
6002 @cindex string64, copying to object file
6003 @cindex @code{string} directive
6004 @cindex @code{string8} directive
6005 @cindex @code{string16} directive
6006 @cindex @code{string32} directive
6007 @cindex @code{string64} directive
6009 Copy the characters in @var{str} to the object file. You may specify more than
6010 one string to copy, separated by commas. Unless otherwise specified for a
6011 particular machine, the assembler marks the end of each string with a 0 byte.
6012 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6014 The variants @code{string16}, @code{string32} and @code{string64} differ from
6015 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6016 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6017 are stored in target endianness byte order.
6023 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6024 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6029 @section @code{.struct @var{expression}}
6031 @cindex @code{struct} directive
6032 Switch to the absolute section, and set the section offset to @var{expression},
6033 which must be an absolute expression. You might use this as follows:
6042 This would define the symbol @code{field1} to have the value 0, the symbol
6043 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6044 value 8. Assembly would be left in the absolute section, and you would need to
6045 use a @code{.section} directive of some sort to change to some other section
6046 before further assembly.
6050 @section @code{.subsection @var{name}}
6052 @cindex @code{subsection} directive
6053 @cindex Section Stack
6054 This is one of the ELF section stack manipulation directives. The others are
6055 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6056 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6059 This directive replaces the current subsection with @code{name}. The current
6060 section is not changed. The replaced subsection is put onto the section stack
6061 in place of the then current top of stack subsection.
6066 @section @code{.symver}
6067 @cindex @code{symver} directive
6068 @cindex symbol versioning
6069 @cindex versions of symbols
6070 Use the @code{.symver} directive to bind symbols to specific version nodes
6071 within a source file. This is only supported on ELF platforms, and is
6072 typically used when assembling files to be linked into a shared library.
6073 There are cases where it may make sense to use this in objects to be bound
6074 into an application itself so as to override a versioned symbol from a
6077 For ELF targets, the @code{.symver} directive can be used like this:
6079 .symver @var{name}, @var{name2@@nodename}
6081 If the symbol @var{name} is defined within the file
6082 being assembled, the @code{.symver} directive effectively creates a symbol
6083 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6084 just don't try and create a regular alias is that the @var{@@} character isn't
6085 permitted in symbol names. The @var{name2} part of the name is the actual name
6086 of the symbol by which it will be externally referenced. The name @var{name}
6087 itself is merely a name of convenience that is used so that it is possible to
6088 have definitions for multiple versions of a function within a single source
6089 file, and so that the compiler can unambiguously know which version of a
6090 function is being mentioned. The @var{nodename} portion of the alias should be
6091 the name of a node specified in the version script supplied to the linker when
6092 building a shared library. If you are attempting to override a versioned
6093 symbol from a shared library, then @var{nodename} should correspond to the
6094 nodename of the symbol you are trying to override.
6096 If the symbol @var{name} is not defined within the file being assembled, all
6097 references to @var{name} will be changed to @var{name2@@nodename}. If no
6098 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6101 Another usage of the @code{.symver} directive is:
6103 .symver @var{name}, @var{name2@@@@nodename}
6105 In this case, the symbol @var{name} must exist and be defined within
6106 the file being assembled. It is similar to @var{name2@@nodename}. The
6107 difference is @var{name2@@@@nodename} will also be used to resolve
6108 references to @var{name2} by the linker.
6110 The third usage of the @code{.symver} directive is:
6112 .symver @var{name}, @var{name2@@@@@@nodename}
6114 When @var{name} is not defined within the
6115 file being assembled, it is treated as @var{name2@@nodename}. When
6116 @var{name} is defined within the file being assembled, the symbol
6117 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6122 @section @code{.tag @var{structname}}
6124 @cindex COFF structure debugging
6125 @cindex structure debugging, COFF
6126 @cindex @code{tag} directive
6127 This directive is generated by compilers to include auxiliary debugging
6128 information in the symbol table. It is only permitted inside
6129 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6130 definitions in the symbol table with instances of those structures.
6133 @samp{.tag} is only used when generating COFF format output; when
6134 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6140 @section @code{.text @var{subsection}}
6142 @cindex @code{text} directive
6143 Tells @command{@value{AS}} to assemble the following statements onto the end of
6144 the text subsection numbered @var{subsection}, which is an absolute
6145 expression. If @var{subsection} is omitted, subsection number zero
6149 @section @code{.title "@var{heading}"}
6151 @cindex @code{title} directive
6152 @cindex listing control: title line
6153 Use @var{heading} as the title (second line, immediately after the
6154 source file name and pagenumber) when generating assembly listings.
6156 This directive affects subsequent pages, as well as the current page if
6157 it appears within ten lines of the top of a page.
6161 @section @code{.type}
6163 This directive is used to set the type of a symbol.
6167 @c only print the extra heading if both COFF and ELF are set
6168 @subheading COFF Version
6171 @cindex COFF symbol type
6172 @cindex symbol type, COFF
6173 @cindex @code{type} directive (COFF version)
6174 For COFF targets, this directive is permitted only within
6175 @code{.def}/@code{.endef} pairs. It is used like this:
6181 This records the integer @var{int} as the type attribute of a symbol table
6185 @samp{.type} is associated only with COFF format output; when
6186 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6187 directive but ignores it.
6193 @c only print the extra heading if both COFF and ELF are set
6194 @subheading ELF Version
6197 @cindex ELF symbol type
6198 @cindex symbol type, ELF
6199 @cindex @code{type} directive (ELF version)
6200 For ELF targets, the @code{.type} directive is used like this:
6203 .type @var{name} , @var{type description}
6206 This sets the type of symbol @var{name} to be either a
6207 function symbol or an object symbol. There are five different syntaxes
6208 supported for the @var{type description} field, in order to provide
6209 compatibility with various other assemblers.
6211 Because some of the characters used in these syntaxes (such as @samp{@@} and
6212 @samp{#}) are comment characters for some architectures, some of the syntaxes
6213 below do not work on all architectures. The first variant will be accepted by
6214 the GNU assembler on all architectures so that variant should be used for
6215 maximum portability, if you do not need to assemble your code with other
6218 The syntaxes supported are:
6221 .type <name> STT_<TYPE_IN_UPPER_CASE>
6222 .type <name>,#<type>
6223 .type <name>,@@<type>
6224 .type <name>,%>type>
6225 .type <name>,"<type>"
6228 The types supported are:
6233 Mark the symbol as being a function name.
6237 Mark the symbol as being a data object.
6241 Mark the symbol as being a thead-local data object.
6245 Mark the symbol as being a common data object.
6248 Note: Some targets support extra types in addition to those listed above.
6254 @section @code{.uleb128 @var{expressions}}
6256 @cindex @code{uleb128} directive
6257 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6258 compact, variable length representation of numbers used by the DWARF
6259 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6263 @section @code{.val @var{addr}}
6265 @cindex @code{val} directive
6266 @cindex COFF value attribute
6267 @cindex value attribute, COFF
6268 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6269 records the address @var{addr} as the value attribute of a symbol table
6273 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6274 configured for @code{b.out}, it accepts this directive but ignores it.
6280 @section @code{.version "@var{string}"}
6282 @cindex @code{version} directive
6283 This directive creates a @code{.note} section and places into it an ELF
6284 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6289 @section @code{.vtable_entry @var{table}, @var{offset}}
6291 @cindex @code{vtable_entry} directive
6292 This directive finds or creates a symbol @code{table} and creates a
6293 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6296 @section @code{.vtable_inherit @var{child}, @var{parent}}
6298 @cindex @code{vtable_inherit} directive
6299 This directive finds the symbol @code{child} and finds or creates the symbol
6300 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6301 parent whose addend is the value of the child symbol. As a special case the
6302 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6306 @section @code{.warning "@var{string}"}
6307 @cindex warning directive
6308 Similar to the directive @code{.error}
6309 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6312 @section @code{.weak @var{names}}
6314 @cindex @code{weak} directive
6315 This directive sets the weak attribute on the comma separated list of symbol
6316 @code{names}. If the symbols do not already exist, they will be created.
6318 On COFF targets other than PE, weak symbols are a GNU extension. This
6319 directive sets the weak attribute on the comma separated list of symbol
6320 @code{names}. If the symbols do not already exist, they will be created.
6322 On the PE target, weak symbols are supported natively as weak aliases.
6323 When a weak symbol is created that is not an alias, GAS creates an
6324 alternate symbol to hold the default value.
6327 @section @code{.weakref @var{alias}, @var{target}}
6329 @cindex @code{weakref} directive
6330 This directive creates an alias to the target symbol that enables the symbol to
6331 be referenced with weak-symbol semantics, but without actually making it weak.
6332 If direct references or definitions of the symbol are present, then the symbol
6333 will not be weak, but if all references to it are through weak references, the
6334 symbol will be marked as weak in the symbol table.
6336 The effect is equivalent to moving all references to the alias to a separate
6337 assembly source file, renaming the alias to the symbol in it, declaring the
6338 symbol as weak there, and running a reloadable link to merge the object files
6339 resulting from the assembly of the new source file and the old source file that
6340 had the references to the alias removed.
6342 The alias itself never makes to the symbol table, and is entirely handled
6343 within the assembler.
6346 @section @code{.word @var{expressions}}
6348 @cindex @code{word} directive
6349 This directive expects zero or more @var{expressions}, of any section,
6350 separated by commas.
6353 For each expression, @command{@value{AS}} emits a 32-bit number.
6356 For each expression, @command{@value{AS}} emits a 16-bit number.
6361 The size of the number emitted, and its byte order,
6362 depend on what target computer the assembly is for.
6365 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6366 @c happen---32-bit addressability, period; no long/short jumps.
6367 @ifset DIFF-TBL-KLUGE
6368 @cindex difference tables altered
6369 @cindex altered difference tables
6371 @emph{Warning: Special Treatment to support Compilers}
6375 Machines with a 32-bit address space, but that do less than 32-bit
6376 addressing, require the following special treatment. If the machine of
6377 interest to you does 32-bit addressing (or doesn't require it;
6378 @pxref{Machine Dependencies}), you can ignore this issue.
6381 In order to assemble compiler output into something that works,
6382 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6383 Directives of the form @samp{.word sym1-sym2} are often emitted by
6384 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6385 directive of the form @samp{.word sym1-sym2}, and the difference between
6386 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6387 creates a @dfn{secondary jump table}, immediately before the next label.
6388 This secondary jump table is preceded by a short-jump to the
6389 first byte after the secondary table. This short-jump prevents the flow
6390 of control from accidentally falling into the new table. Inside the
6391 table is a long-jump to @code{sym2}. The original @samp{.word}
6392 contains @code{sym1} minus the address of the long-jump to
6395 If there were several occurrences of @samp{.word sym1-sym2} before the
6396 secondary jump table, all of them are adjusted. If there was a
6397 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6398 long-jump to @code{sym4} is included in the secondary jump table,
6399 and the @code{.word} directives are adjusted to contain @code{sym3}
6400 minus the address of the long-jump to @code{sym4}; and so on, for as many
6401 entries in the original jump table as necessary.
6404 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6405 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6406 assembly language programmers.
6409 @c end DIFF-TBL-KLUGE
6412 @section Deprecated Directives
6414 @cindex deprecated directives
6415 @cindex obsolescent directives
6416 One day these directives won't work.
6417 They are included for compatibility with older assemblers.
6424 @node Object Attributes
6425 @chapter Object Attributes
6426 @cindex object attributes
6428 @command{@value{AS}} assembles source files written for a specific architecture
6429 into object files for that architecture. But not all object files are alike.
6430 Many architectures support incompatible variations. For instance, floating
6431 point arguments might be passed in floating point registers if the object file
6432 requires hardware floating point support---or floating point arguments might be
6433 passed in integer registers if the object file supports processors with no
6434 hardware floating point unit. Or, if two objects are built for different
6435 generations of the same architecture, the combination may require the
6436 newer generation at run-time.
6438 This information is useful during and after linking. At link time,
6439 @command{@value{LD}} can warn about incompatible object files. After link
6440 time, tools like @command{gdb} can use it to process the linked file
6443 Compatibility information is recorded as a series of object attributes. Each
6444 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6445 string, and indicates who sets the meaning of the tag. The tag is an integer,
6446 and indicates what property the attribute describes. The value may be a string
6447 or an integer, and indicates how the property affects this object. Missing
6448 attributes are the same as attributes with a zero value or empty string value.
6450 Object attributes were developed as part of the ABI for the ARM Architecture.
6451 The file format is documented in @cite{ELF for the ARM Architecture}.
6454 * GNU Object Attributes:: @sc{gnu} Object Attributes
6455 * Defining New Object Attributes:: Defining New Object Attributes
6458 @node GNU Object Attributes
6459 @section @sc{gnu} Object Attributes
6461 The @code{.gnu_attribute} directive records an object attribute
6462 with vendor @samp{gnu}.
6464 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6465 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6466 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6467 2} is set for architecture-independent attributes and clear for
6468 architecture-dependent ones.
6470 @subsection Common @sc{gnu} attributes
6472 These attributes are valid on all architectures.
6475 @item Tag_compatibility (32)
6476 The compatibility attribute takes an integer flag value and a vendor name. If
6477 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6478 then the file is only compatible with the named toolchain. If it is greater
6479 than 1, the file can only be processed by other toolchains under some private
6480 arrangement indicated by the flag value and the vendor name.
6483 @subsection MIPS Attributes
6486 @item Tag_GNU_MIPS_ABI_FP (4)
6487 The floating-point ABI used by this object file. The value will be:
6491 0 for files not affected by the floating-point ABI.
6493 1 for files using the hardware floating-point with a standard double-precision
6496 2 for files using the hardware floating-point ABI with a single-precision FPU.
6498 3 for files using the software floating-point ABI.
6500 4 for files using the hardware floating-point ABI with 64-bit wide
6501 double-precision floating-point registers and 32-bit wide general
6506 @subsection PowerPC Attributes
6509 @item Tag_GNU_Power_ABI_FP (4)
6510 The floating-point ABI used by this object file. The value will be:
6514 0 for files not affected by the floating-point ABI.
6516 1 for files using double-precision hardware floating-point ABI.
6518 2 for files using the software floating-point ABI.
6520 3 for files using single-precision hardware floating-point ABI.
6523 @item Tag_GNU_Power_ABI_Vector (8)
6524 The vector ABI used by this object file. The value will be:
6528 0 for files not affected by the vector ABI.
6530 1 for files using general purpose registers to pass vectors.
6532 2 for files using AltiVec registers to pass vectors.
6534 3 for files using SPE registers to pass vectors.
6538 @node Defining New Object Attributes
6539 @section Defining New Object Attributes
6541 If you want to define a new @sc{gnu} object attribute, here are the places you
6542 will need to modify. New attributes should be discussed on the @samp{binutils}
6547 This manual, which is the official register of attributes.
6549 The header for your architecture @file{include/elf}, to define the tag.
6551 The @file{bfd} support file for your architecture, to merge the attribute
6552 and issue any appropriate link warnings.
6554 Test cases in @file{ld/testsuite} for merging and link warnings.
6556 @file{binutils/readelf.c} to display your attribute.
6558 GCC, if you want the compiler to mark the attribute automatically.
6564 @node Machine Dependencies
6565 @chapter Machine Dependent Features
6567 @cindex machine dependencies
6568 The machine instruction sets are (almost by definition) different on
6569 each machine where @command{@value{AS}} runs. Floating point representations
6570 vary as well, and @command{@value{AS}} often supports a few additional
6571 directives or command-line options for compatibility with other
6572 assemblers on a particular platform. Finally, some versions of
6573 @command{@value{AS}} support special pseudo-instructions for branch
6576 This chapter discusses most of these differences, though it does not
6577 include details on any machine's instruction set. For details on that
6578 subject, see the hardware manufacturer's manual.
6582 * Alpha-Dependent:: Alpha Dependent Features
6585 * ARC-Dependent:: ARC Dependent Features
6588 * ARM-Dependent:: ARM Dependent Features
6591 * AVR-Dependent:: AVR Dependent Features
6594 * BFIN-Dependent:: BFIN Dependent Features
6597 * CR16-Dependent:: CR16 Dependent Features
6600 * CRIS-Dependent:: CRIS Dependent Features
6603 * D10V-Dependent:: D10V Dependent Features
6606 * D30V-Dependent:: D30V Dependent Features
6609 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6612 * HPPA-Dependent:: HPPA Dependent Features
6615 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6618 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6621 * i860-Dependent:: Intel 80860 Dependent Features
6624 * i960-Dependent:: Intel 80960 Dependent Features
6627 * IA-64-Dependent:: Intel IA-64 Dependent Features
6630 * IP2K-Dependent:: IP2K Dependent Features
6633 * M32C-Dependent:: M32C Dependent Features
6636 * M32R-Dependent:: M32R Dependent Features
6639 * M68K-Dependent:: M680x0 Dependent Features
6642 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6645 * MIPS-Dependent:: MIPS Dependent Features
6648 * MMIX-Dependent:: MMIX Dependent Features
6651 * MSP430-Dependent:: MSP430 Dependent Features
6654 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6655 * SH64-Dependent:: SuperH SH64 Dependent Features
6658 * PDP-11-Dependent:: PDP-11 Dependent Features
6661 * PJ-Dependent:: picoJava Dependent Features
6664 * PPC-Dependent:: PowerPC Dependent Features
6667 * Sparc-Dependent:: SPARC Dependent Features
6670 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6673 * V850-Dependent:: V850 Dependent Features
6676 * Xtensa-Dependent:: Xtensa Dependent Features
6679 * Z80-Dependent:: Z80 Dependent Features
6682 * Z8000-Dependent:: Z8000 Dependent Features
6685 * Vax-Dependent:: VAX Dependent Features
6692 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6693 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6694 @c peculiarity: to preserve cross-references, there must be a node called
6695 @c "Machine Dependencies". Hence the conditional nodenames in each
6696 @c major node below. Node defaulting in makeinfo requires adjacency of
6697 @c node and sectioning commands; hence the repetition of @chapter BLAH
6698 @c in both conditional blocks.
6701 @include c-alpha.texi
6717 @include c-bfin.texi
6721 @include c-cr16.texi
6725 @include c-cris.texi
6730 @node Machine Dependencies
6731 @chapter Machine Dependent Features
6733 The machine instruction sets are different on each Renesas chip family,
6734 and there are also some syntax differences among the families. This
6735 chapter describes the specific @command{@value{AS}} features for each
6739 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6740 * SH-Dependent:: Renesas SH Dependent Features
6747 @include c-d10v.texi
6751 @include c-d30v.texi
6755 @include c-h8300.texi
6759 @include c-hppa.texi
6763 @include c-i370.texi
6767 @include c-i386.texi
6771 @include c-i860.texi
6775 @include c-i960.texi
6779 @include c-ia64.texi
6783 @include c-ip2k.texi
6787 @include c-m32c.texi
6791 @include c-m32r.texi
6795 @include c-m68k.texi
6799 @include c-m68hc11.texi
6803 @include c-mips.texi
6807 @include c-mmix.texi
6811 @include c-msp430.texi
6815 @include c-ns32k.texi
6819 @include c-pdp11.texi
6832 @include c-sh64.texi
6836 @include c-sparc.texi
6840 @include c-tic54x.texi
6856 @include c-v850.texi
6860 @include c-xtensa.texi
6864 @c reverse effect of @down at top of generic Machine-Dep chapter
6868 @node Reporting Bugs
6869 @chapter Reporting Bugs
6870 @cindex bugs in assembler
6871 @cindex reporting bugs in assembler
6873 Your bug reports play an essential role in making @command{@value{AS}} reliable.
6875 Reporting a bug may help you by bringing a solution to your problem, or it may
6876 not. But in any case the principal function of a bug report is to help the
6877 entire community by making the next version of @command{@value{AS}} work better.
6878 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
6880 In order for a bug report to serve its purpose, you must include the
6881 information that enables us to fix the bug.
6884 * Bug Criteria:: Have you found a bug?
6885 * Bug Reporting:: How to report bugs
6889 @section Have You Found a Bug?
6890 @cindex bug criteria
6892 If you are not sure whether you have found a bug, here are some guidelines:
6895 @cindex fatal signal
6896 @cindex assembler crash
6897 @cindex crash of assembler
6899 If the assembler gets a fatal signal, for any input whatever, that is a
6900 @command{@value{AS}} bug. Reliable assemblers never crash.
6902 @cindex error on valid input
6904 If @command{@value{AS}} produces an error message for valid input, that is a bug.
6906 @cindex invalid input
6908 If @command{@value{AS}} does not produce an error message for invalid input, that
6909 is a bug. However, you should note that your idea of ``invalid input'' might
6910 be our idea of ``an extension'' or ``support for traditional practice''.
6913 If you are an experienced user of assemblers, your suggestions for improvement
6914 of @command{@value{AS}} are welcome in any case.
6918 @section How to Report Bugs
6920 @cindex assembler bugs, reporting
6922 A number of companies and individuals offer support for @sc{gnu} products. If
6923 you obtained @command{@value{AS}} from a support organization, we recommend you
6924 contact that organization first.
6926 You can find contact information for many support companies and
6927 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
6931 In any event, we also recommend that you send bug reports for @command{@value{AS}}
6935 The fundamental principle of reporting bugs usefully is this:
6936 @strong{report all the facts}. If you are not sure whether to state a
6937 fact or leave it out, state it!
6939 Often people omit facts because they think they know what causes the problem
6940 and assume that some details do not matter. Thus, you might assume that the
6941 name of a symbol you use in an example does not matter. Well, probably it does
6942 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
6943 happens to fetch from the location where that name is stored in memory;
6944 perhaps, if the name were different, the contents of that location would fool
6945 the assembler into doing the right thing despite the bug. Play it safe and
6946 give a specific, complete example. That is the easiest thing for you to do,
6947 and the most helpful.
6949 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
6950 it is new to us. Therefore, always write your bug reports on the assumption
6951 that the bug has not been reported previously.
6953 Sometimes people give a few sketchy facts and ask, ``Does this ring a
6954 bell?'' This cannot help us fix a bug, so it is basically useless. We
6955 respond by asking for enough details to enable us to investigate.
6956 You might as well expedite matters by sending them to begin with.
6958 To enable us to fix the bug, you should include all these things:
6962 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
6963 it with the @samp{--version} argument.
6965 Without this, we will not know whether there is any point in looking for
6966 the bug in the current version of @command{@value{AS}}.
6969 Any patches you may have applied to the @command{@value{AS}} source.
6972 The type of machine you are using, and the operating system name and
6976 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
6980 The command arguments you gave the assembler to assemble your example and
6981 observe the bug. To guarantee you will not omit something important, list them
6982 all. A copy of the Makefile (or the output from make) is sufficient.
6984 If we were to try to guess the arguments, we would probably guess wrong
6985 and then we might not encounter the bug.
6988 A complete input file that will reproduce the bug. If the bug is observed when
6989 the assembler is invoked via a compiler, send the assembler source, not the
6990 high level language source. Most compilers will produce the assembler source
6991 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
6992 the options @samp{-v --save-temps}; this will save the assembler source in a
6993 file with an extension of @file{.s}, and also show you exactly how
6994 @command{@value{AS}} is being run.
6997 A description of what behavior you observe that you believe is
6998 incorrect. For example, ``It gets a fatal signal.''
7000 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7001 will certainly notice it. But if the bug is incorrect output, we might not
7002 notice unless it is glaringly wrong. You might as well not give us a chance to
7005 Even if the problem you experience is a fatal signal, you should still say so
7006 explicitly. Suppose something strange is going on, such as, your copy of
7007 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7008 library on your system. (This has happened!) Your copy might crash and ours
7009 would not. If you told us to expect a crash, then when ours fails to crash, we
7010 would know that the bug was not happening for us. If you had not told us to
7011 expect a crash, then we would not be able to draw any conclusion from our
7015 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7016 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7017 option. Always send diffs from the old file to the new file. If you even
7018 discuss something in the @command{@value{AS}} source, refer to it by context, not
7021 The line numbers in our development sources will not match those in your
7022 sources. Your line numbers would convey no useful information to us.
7025 Here are some things that are not necessary:
7029 A description of the envelope of the bug.
7031 Often people who encounter a bug spend a lot of time investigating
7032 which changes to the input file will make the bug go away and which
7033 changes will not affect it.
7035 This is often time consuming and not very useful, because the way we
7036 will find the bug is by running a single example under the debugger
7037 with breakpoints, not by pure deduction from a series of examples.
7038 We recommend that you save your time for something else.
7040 Of course, if you can find a simpler example to report @emph{instead}
7041 of the original one, that is a convenience for us. Errors in the
7042 output will be easier to spot, running under the debugger will take
7043 less time, and so on.
7045 However, simplification is not vital; if you do not want to do this,
7046 report the bug anyway and send us the entire test case you used.
7049 A patch for the bug.
7051 A patch for the bug does help us if it is a good one. But do not omit
7052 the necessary information, such as the test case, on the assumption that
7053 a patch is all we need. We might see problems with your patch and decide
7054 to fix the problem another way, or we might not understand it at all.
7056 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7057 construct an example that will make the program follow a certain path through
7058 the code. If you do not send us the example, we will not be able to construct
7059 one, so we will not be able to verify that the bug is fixed.
7061 And if we cannot understand what bug you are trying to fix, or why your
7062 patch should be an improvement, we will not install it. A test case will
7063 help us to understand.
7066 A guess about what the bug is or what it depends on.
7068 Such guesses are usually wrong. Even we cannot guess right about such
7069 things without first using the debugger to find the facts.
7072 @node Acknowledgements
7073 @chapter Acknowledgements
7075 If you have contributed to GAS and your name isn't listed here,
7076 it is not meant as a slight. We just don't know about it. Send mail to the
7077 maintainer, and we'll correct the situation. Currently
7079 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
7081 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7084 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7085 information and the 68k series machines, most of the preprocessing pass, and
7086 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7088 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7089 many bug fixes, including merging support for several processors, breaking GAS
7090 up to handle multiple object file format back ends (including heavy rewrite,
7091 testing, an integration of the coff and b.out back ends), adding configuration
7092 including heavy testing and verification of cross assemblers and file splits
7093 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7094 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7095 port (including considerable amounts of reverse engineering), a SPARC opcode
7096 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7097 assertions and made them work, much other reorganization, cleanup, and lint.
7099 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7100 in format-specific I/O modules.
7102 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7103 has done much work with it since.
7105 The Intel 80386 machine description was written by Eliot Dresselhaus.
7107 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7109 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7110 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7112 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7113 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7114 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7115 support a.out format.
7117 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7118 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7119 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7120 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7123 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7124 simplified the configuration of which versions accept which directives. He
7125 updated the 68k machine description so that Motorola's opcodes always produced
7126 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7127 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7128 cross-compilation support, and one bug in relaxation that took a week and
7129 required the proverbial one-bit fix.
7131 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7132 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7133 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7134 PowerPC assembler, and made a few other minor patches.
7136 Steve Chamberlain made GAS able to generate listings.
7138 Hewlett-Packard contributed support for the HP9000/300.
7140 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7141 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7142 formats). This work was supported by both the Center for Software Science at
7143 the University of Utah and Cygnus Support.
7145 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7146 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7147 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7148 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7149 and some initial 64-bit support).
7151 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7153 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7154 support for openVMS/Alpha.
7156 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7159 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7160 Inc.@: added support for Xtensa processors.
7162 Several engineers at Cygnus Support have also provided many small bug fixes and
7163 configuration enhancements.
7165 Many others have contributed large or small bugfixes and enhancements. If
7166 you have contributed significant work and are not mentioned on this list, and
7167 want to be, let us know. Some of the history has been lost; we are not
7168 intentionally leaving anyone out.
7173 @unnumbered AS Index
OSDN Git Service
