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, 2008, 2009, 2010
4 @c Free Software Foundation, Inc.
5 @c UPDATE!! On future updates--
6 @c (1) check for new machine-dep cmdline options in
7 @c md_parse_option definitions in config/tc-*.c
8 @c (2) for platform-specific directives, examine md_pseudo_op
10 @c (3) for object-format specific directives, examine obj_pseudo_op
12 @c (4) portable directives in potable[] in read.c
16 @macro gcctabopt{body}
19 @c defaults, config file may override:
24 @include asconfig.texi
29 @c common OR combinations of conditions
55 @set abnormal-separator
59 @settitle Using @value{AS}
62 @settitle Using @value{AS} (@value{TARGET})
64 @setchapternewpage odd
69 @c WARE! Some of the machine-dependent sections contain tables of machine
70 @c instructions. Except in multi-column format, these tables look silly.
71 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
72 @c the multi-col format is faked within @example sections.
74 @c Again unfortunately, the natural size that fits on a page, for these tables,
75 @c is different depending on whether or not smallbook is turned on.
76 @c This matters, because of order: text flow switches columns at each page
79 @c The format faked in this source works reasonably well for smallbook,
80 @c not well for the default large-page format. This manual expects that if you
81 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
82 @c tables in question. You can turn on one without the other at your
83 @c discretion, of course.
86 @c the insn tables look just as silly in info files regardless of smallbook,
87 @c might as well show 'em anyways.
91 @dircategory Software development
93 * As: (as). The GNU assembler.
94 * Gas: (as). The GNU assembler.
102 This file documents the GNU Assembler "@value{AS}".
104 @c man begin COPYRIGHT
105 Copyright @copyright{} 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
106 2000, 2001, 2002, 2006, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
108 Permission is granted to copy, distribute and/or modify this document
109 under the terms of the GNU Free Documentation License, Version 1.3
110 or any later version published by the Free Software Foundation;
111 with no Invariant Sections, with no Front-Cover Texts, and with no
112 Back-Cover Texts. A copy of the license is included in the
113 section entitled ``GNU Free Documentation License''.
119 @title Using @value{AS}
120 @subtitle The @sc{gnu} Assembler
122 @subtitle for the @value{TARGET} family
124 @ifset VERSION_PACKAGE
126 @subtitle @value{VERSION_PACKAGE}
129 @subtitle Version @value{VERSION}
132 The Free Software Foundation Inc.@: thanks The Nice Computer
133 Company of Australia for loaning Dean Elsner to write the
134 first (Vax) version of @command{as} for Project @sc{gnu}.
135 The proprietors, management and staff of TNCCA thank FSF for
136 distracting the boss while they got some work
139 @author Dean Elsner, Jay Fenlason & friends
143 \hfill {\it Using {\tt @value{AS}}}\par
144 \hfill Edited by Cygnus Support\par
146 %"boxit" macro for figures:
147 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
148 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
149 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
150 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
151 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
154 @vskip 0pt plus 1filll
155 Copyright @copyright{} 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
156 2000, 2001, 2002, 2006, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
158 Permission is granted to copy, distribute and/or modify this document
159 under the terms of the GNU Free Documentation License, Version 1.3
160 or any later version published by the Free Software Foundation;
161 with no Invariant Sections, with no Front-Cover Texts, and with no
162 Back-Cover Texts. A copy of the license is included in the
163 section entitled ``GNU Free Documentation License''.
170 @top Using @value{AS}
172 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
173 @ifset VERSION_PACKAGE
174 @value{VERSION_PACKAGE}
176 version @value{VERSION}.
178 This version of the file describes @command{@value{AS}} configured to generate
179 code for @value{TARGET} architectures.
182 This document is distributed under the terms of the GNU Free
183 Documentation License. A copy of the license is included in the
184 section entitled ``GNU Free Documentation License''.
187 * Overview:: Overview
188 * Invoking:: Command-Line Options
190 * Sections:: Sections and Relocation
192 * Expressions:: Expressions
193 * Pseudo Ops:: Assembler Directives
195 * Object Attributes:: Object Attributes
197 * Machine Dependencies:: Machine Dependent Features
198 * Reporting Bugs:: Reporting Bugs
199 * Acknowledgements:: Who Did What
200 * GNU Free Documentation License:: GNU Free Documentation License
201 * AS Index:: AS Index
208 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
210 This version of the manual describes @command{@value{AS}} configured to generate
211 code for @value{TARGET} architectures.
215 @cindex invocation summary
216 @cindex option summary
217 @cindex summary of options
218 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
219 see @ref{Invoking,,Command-Line Options}.
221 @c man title AS the portable GNU assembler.
225 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
229 @c We don't use deffn and friends for the following because they seem
230 @c to be limited to one line for the header.
232 @c man begin SYNOPSIS
233 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
234 [@b{--compress-debug-sections}] [@b{--nocompress-debug-sections}]
235 [@b{--debug-prefix-map} @var{old}=@var{new}]
236 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
237 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
238 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
239 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
240 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}] [@b{-o}
241 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
242 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
243 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{@@@var{FILE}}]
244 [@b{--target-help}] [@var{target-options}]
245 [@b{--}|@var{files} @dots{}]
247 @c Target dependent options are listed below. Keep the list sorted.
248 @c Add an empty line for separation.
251 @emph{Target Alpha options:}
253 [@b{-mdebug} | @b{-no-mdebug}]
254 [@b{-replace} | @b{-noreplace}]
255 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
256 [@b{-F}] [@b{-32addr}]
260 @emph{Target ARC options:}
266 @emph{Target ARM options:}
267 @c Don't document the deprecated options
268 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
269 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
270 [@b{-mfpu}=@var{floating-point-format}]
271 [@b{-mfloat-abi}=@var{abi}]
272 [@b{-meabi}=@var{ver}]
275 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
276 @b{-mapcs-reentrant}]
277 [@b{-mthumb-interwork}] [@b{-k}]
281 @emph{Target Blackfin options:}
282 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
289 @emph{Target CRIS options:}
290 [@b{--underscore} | @b{--no-underscore}]
292 [@b{--emulation=criself} | @b{--emulation=crisaout}]
293 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
294 @c Deprecated -- deliberately not documented.
299 @emph{Target D10V options:}
304 @emph{Target D30V options:}
305 [@b{-O}|@b{-n}|@b{-N}]
309 @emph{Target H8/300 options:}
313 @c HPPA has no machine-dependent assembler options (yet).
317 @emph{Target i386 options:}
318 [@b{--32}|@b{--64}] [@b{-n}]
319 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
323 @emph{Target i960 options:}
324 @c see md_parse_option in tc-i960.c
325 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
327 [@b{-b}] [@b{-no-relax}]
331 @emph{Target IA-64 options:}
332 [@b{-mconstant-gp}|@b{-mauto-pic}]
333 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
335 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
336 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
337 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
338 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
342 @emph{Target IP2K options:}
343 [@b{-mip2022}|@b{-mip2022ext}]
347 @emph{Target M32C options:}
348 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
352 @emph{Target M32R options:}
353 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
358 @emph{Target M680X0 options:}
359 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
363 @emph{Target M68HC11 options:}
364 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}]
365 [@b{-mshort}|@b{-mlong}]
366 [@b{-mshort-double}|@b{-mlong-double}]
367 [@b{--force-long-branches}] [@b{--short-branches}]
368 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
369 [@b{--print-opcodes}] [@b{--generate-example}]
373 @emph{Target MCORE options:}
374 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
375 [@b{-mcpu=[210|340]}]
378 @emph{Target MICROBLAZE options:}
379 @c MicroBlaze has no machine-dependent assembler options.
383 @emph{Target MIPS options:}
384 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
385 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
386 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
387 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
388 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
389 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
390 [@b{-mips64}] [@b{-mips64r2}]
391 [@b{-construct-floats}] [@b{-no-construct-floats}]
392 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
393 [@b{-mips16}] [@b{-no-mips16}]
394 [@b{-msmartmips}] [@b{-mno-smartmips}]
395 [@b{-mips3d}] [@b{-no-mips3d}]
396 [@b{-mdmx}] [@b{-no-mdmx}]
397 [@b{-mdsp}] [@b{-mno-dsp}]
398 [@b{-mdspr2}] [@b{-mno-dspr2}]
399 [@b{-mmt}] [@b{-mno-mt}]
400 [@b{-mfix7000}] [@b{-mno-fix7000}]
401 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
402 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
403 [@b{-mdebug}] [@b{-no-mdebug}]
404 [@b{-mpdr}] [@b{-mno-pdr}]
408 @emph{Target MMIX options:}
409 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
410 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
411 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
412 [@b{--linker-allocated-gregs}]
416 @emph{Target PDP11 options:}
417 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
418 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
419 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
423 @emph{Target picoJava options:}
428 @emph{Target PowerPC options:}
429 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|
430 @b{-m403}|@b{-m405}|@b{-mppc64}|@b{-m620}|@b{-mppc64bridge}|@b{-mbooke}]
431 [@b{-mcom}|@b{-many}|@b{-maltivec}|@b{-mvsx}] [@b{-memb}]
432 [@b{-mregnames}|@b{-mno-regnames}]
433 [@b{-mrelocatable}|@b{-mrelocatable-lib}]
434 [@b{-mlittle}|@b{-mlittle-endian}|@b{-mbig}|@b{-mbig-endian}]
435 [@b{-msolaris}|@b{-mno-solaris}]
439 @emph{Target RX options:}
440 [@b{-mlittle-endian}|@b{-mbig-endian}]
441 [@b{-m32bit-ints}|@b{-m16bit-ints}]
442 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
446 @emph{Target s390 options:}
447 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
448 [@b{-mregnames}|@b{-mno-regnames}]
449 [@b{-mwarn-areg-zero}]
453 @emph{Target SCORE options:}
454 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
455 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
456 [@b{-march=score7}][@b{-march=score3}]
457 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
461 @emph{Target SPARC options:}
462 @c The order here is important. See c-sparc.texi.
463 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
464 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
465 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
470 @emph{Target TIC54X options:}
471 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
472 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
477 @emph{Target TIC6X options:}
478 [@b{-march=@var{arch}}] [@b{-matomic}|@b{-mno-atomic}]
479 [@b{-mbig-endian}|@b{-mlittle-endian}] [@b{-mdsbt}|@b{-mno-dsbt}]
480 [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}] [@b{-mpic}|@b{-mno-pic}]
485 @emph{Target Z80 options:}
486 [@b{-z80}] [@b{-r800}]
487 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
488 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
489 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
490 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
491 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
492 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
496 @c Z8000 has no machine-dependent assembler options
500 @emph{Target Xtensa options:}
501 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
502 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
503 [@b{--[no-]transform}]
504 [@b{--rename-section} @var{oldname}=@var{newname}]
512 @include at-file.texi
515 Turn on listings, in any of a variety of ways:
519 omit false conditionals
522 omit debugging directives
525 include general information, like @value{AS} version and options passed
528 include high-level source
534 include macro expansions
537 omit forms processing
543 set the name of the listing file
546 You may combine these options; for example, use @samp{-aln} for assembly
547 listing without forms processing. The @samp{=file} option, if used, must be
548 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
551 Begin in alternate macro mode.
553 @xref{Altmacro,,@code{.altmacro}}.
556 @item --compress-debug-sections
557 Compress DWARF debug sections using zlib. The debug sections are renamed
558 to begin with @samp{.zdebug}, and the resulting object file may not be
559 compatible with older linkers and object file utilities.
561 @item --nocompress-debug-sections
562 Do not compress DWARF debug sections. This is the default.
565 Ignored. This option is accepted for script compatibility with calls to
568 @item --debug-prefix-map @var{old}=@var{new}
569 When assembling files in directory @file{@var{old}}, record debugging
570 information describing them as in @file{@var{new}} instead.
572 @item --defsym @var{sym}=@var{value}
573 Define the symbol @var{sym} to be @var{value} before assembling the input file.
574 @var{value} must be an integer constant. As in C, a leading @samp{0x}
575 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
576 value. The value of the symbol can be overridden inside a source file via the
577 use of a @code{.set} pseudo-op.
580 ``fast''---skip whitespace and comment preprocessing (assume source is
585 Generate debugging information for each assembler source line using whichever
586 debug format is preferred by the target. This currently means either STABS,
590 Generate stabs debugging information for each assembler line. This
591 may help debugging assembler code, if the debugger can handle it.
594 Generate stabs debugging information for each assembler line, with GNU
595 extensions that probably only gdb can handle, and that could make other
596 debuggers crash or refuse to read your program. This
597 may help debugging assembler code. Currently the only GNU extension is
598 the location of the current working directory at assembling time.
601 Generate DWARF2 debugging information for each assembler line. This
602 may help debugging assembler code, if the debugger can handle it. Note---this
603 option is only supported by some targets, not all of them.
606 Print a summary of the command line options and exit.
609 Print a summary of all target specific options and exit.
612 Add directory @var{dir} to the search list for @code{.include} directives.
615 Don't warn about signed overflow.
618 @ifclear DIFF-TBL-KLUGE
619 This option is accepted but has no effect on the @value{TARGET} family.
621 @ifset DIFF-TBL-KLUGE
622 Issue warnings when difference tables altered for long displacements.
627 Keep (in the symbol table) local symbols. These symbols start with
628 system-specific local label prefixes, typically @samp{.L} for ELF systems
629 or @samp{L} for traditional a.out systems.
634 @item --listing-lhs-width=@var{number}
635 Set the maximum width, in words, of the output data column for an assembler
636 listing to @var{number}.
638 @item --listing-lhs-width2=@var{number}
639 Set the maximum width, in words, of the output data column for continuation
640 lines in an assembler listing to @var{number}.
642 @item --listing-rhs-width=@var{number}
643 Set the maximum width of an input source line, as displayed in a listing, to
646 @item --listing-cont-lines=@var{number}
647 Set the maximum number of lines printed in a listing for a single line of input
650 @item -o @var{objfile}
651 Name the object-file output from @command{@value{AS}} @var{objfile}.
654 Fold the data section into the text section.
656 @kindex --hash-size=@var{number}
657 Set the default size of GAS's hash tables to a prime number close to
658 @var{number}. Increasing this value can reduce the length of time it takes the
659 assembler to perform its tasks, at the expense of increasing the assembler's
660 memory requirements. Similarly reducing this value can reduce the memory
661 requirements at the expense of speed.
663 @item --reduce-memory-overheads
664 This option reduces GAS's memory requirements, at the expense of making the
665 assembly processes slower. Currently this switch is a synonym for
666 @samp{--hash-size=4051}, but in the future it may have other effects as well.
669 Print the maximum space (in bytes) and total time (in seconds) used by
672 @item --strip-local-absolute
673 Remove local absolute symbols from the outgoing symbol table.
677 Print the @command{as} version.
680 Print the @command{as} version and exit.
684 Suppress warning messages.
686 @item --fatal-warnings
687 Treat warnings as errors.
690 Don't suppress warning messages or treat them as errors.
699 Generate an object file even after errors.
701 @item -- | @var{files} @dots{}
702 Standard input, or source files to assemble.
710 @xref{Alpha Options}, for the options available when @value{AS} is configured
711 for an Alpha processor.
716 The following options are available when @value{AS} is configured for an Alpha
720 @include c-alpha.texi
721 @c ended inside the included file
728 The following options are available when @value{AS} is configured for
733 This option selects the core processor variant.
735 Select either big-endian (-EB) or little-endian (-EL) output.
740 The following options are available when @value{AS} is configured for the ARM
744 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
745 Specify which ARM processor variant is the target.
746 @item -march=@var{architecture}[+@var{extension}@dots{}]
747 Specify which ARM architecture variant is used by the target.
748 @item -mfpu=@var{floating-point-format}
749 Select which Floating Point architecture is the target.
750 @item -mfloat-abi=@var{abi}
751 Select which floating point ABI is in use.
753 Enable Thumb only instruction decoding.
754 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
755 Select which procedure calling convention is in use.
757 Select either big-endian (-EB) or little-endian (-EL) output.
758 @item -mthumb-interwork
759 Specify that the code has been generated with interworking between Thumb and
762 Specify that PIC code has been generated.
767 The following options are available when @value{AS} is configured for
768 the Blackfin processor family.
771 @item -mcpu=@var{processor}@r{[}-@var{sirevision}@r{]}
772 This option specifies the target processor. The optional @var{sirevision}
773 is not used in assembler.
775 Assemble for the FDPIC ABI.
783 See the info pages for documentation of the CRIS-specific options.
787 The following options are available when @value{AS} is configured for
790 @cindex D10V optimization
791 @cindex optimization, D10V
793 Optimize output by parallelizing instructions.
798 The following options are available when @value{AS} is configured for a D30V
801 @cindex D30V optimization
802 @cindex optimization, D30V
804 Optimize output by parallelizing instructions.
808 Warn when nops are generated.
810 @cindex D30V nops after 32-bit multiply
812 Warn when a nop after a 32-bit multiply instruction is generated.
820 @xref{i386-Options}, for the options available when @value{AS} is
821 configured for an i386 processor.
826 The following options are available when @value{AS} is configured for
831 @c ended inside the included file
838 The following options are available when @value{AS} is configured for the
839 Intel 80960 processor.
842 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
843 Specify which variant of the 960 architecture is the target.
846 Add code to collect statistics about branches taken.
849 Do not alter compare-and-branch instructions for long displacements;
856 The following options are available when @value{AS} is configured for the
862 Specifies that the extended IP2022 instructions are allowed.
865 Restores the default behaviour, which restricts the permitted instructions to
866 just the basic IP2022 ones.
872 The following options are available when @value{AS} is configured for the
873 Renesas M32C and M16C processors.
878 Assemble M32C instructions.
881 Assemble M16C instructions (the default).
884 Enable support for link-time relaxations.
887 Support H'00 style hex constants in addition to 0x00 style.
893 The following options are available when @value{AS} is configured for the
894 Renesas M32R (formerly Mitsubishi M32R) series.
899 Specify which processor in the M32R family is the target. The default
900 is normally the M32R, but this option changes it to the M32RX.
902 @item --warn-explicit-parallel-conflicts or --Wp
903 Produce warning messages when questionable parallel constructs are
906 @item --no-warn-explicit-parallel-conflicts or --Wnp
907 Do not produce warning messages when questionable parallel constructs are
914 The following options are available when @value{AS} is configured for the
915 Motorola 68000 series.
920 Shorten references to undefined symbols, to one word instead of two.
922 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
923 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
924 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
925 Specify what processor in the 68000 family is the target. The default
926 is normally the 68020, but this can be changed at configuration time.
928 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
929 The target machine does (or does not) have a floating-point coprocessor.
930 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
931 the basic 68000 is not compatible with the 68881, a combination of the
932 two can be specified, since it's possible to do emulation of the
933 coprocessor instructions with the main processor.
935 @item -m68851 | -mno-68851
936 The target machine does (or does not) have a memory-management
937 unit coprocessor. The default is to assume an MMU for 68020 and up.
944 For details about the PDP-11 machine dependent features options,
945 see @ref{PDP-11-Options}.
948 @item -mpic | -mno-pic
949 Generate position-independent (or position-dependent) code. The
950 default is @option{-mpic}.
953 @itemx -mall-extensions
954 Enable all instruction set extensions. This is the default.
956 @item -mno-extensions
957 Disable all instruction set extensions.
959 @item -m@var{extension} | -mno-@var{extension}
960 Enable (or disable) a particular instruction set extension.
963 Enable the instruction set extensions supported by a particular CPU, and
964 disable all other extensions.
966 @item -m@var{machine}
967 Enable the instruction set extensions supported by a particular machine
968 model, and disable all other extensions.
974 The following options are available when @value{AS} is configured for
975 a picoJava processor.
979 @cindex PJ endianness
980 @cindex endianness, PJ
981 @cindex big endian output, PJ
983 Generate ``big endian'' format output.
985 @cindex little endian output, PJ
987 Generate ``little endian'' format output.
993 The following options are available when @value{AS} is configured for the
994 Motorola 68HC11 or 68HC12 series.
998 @item -m68hc11 | -m68hc12 | -m68hcs12
999 Specify what processor is the target. The default is
1000 defined by the configuration option when building the assembler.
1003 Specify to use the 16-bit integer ABI.
1006 Specify to use the 32-bit integer ABI.
1008 @item -mshort-double
1009 Specify to use the 32-bit double ABI.
1012 Specify to use the 64-bit double ABI.
1014 @item --force-long-branches
1015 Relative branches are turned into absolute ones. This concerns
1016 conditional branches, unconditional branches and branches to a
1019 @item -S | --short-branches
1020 Do not turn relative branches into absolute ones
1021 when the offset is out of range.
1023 @item --strict-direct-mode
1024 Do not turn the direct addressing mode into extended addressing mode
1025 when the instruction does not support direct addressing mode.
1027 @item --print-insn-syntax
1028 Print the syntax of instruction in case of error.
1030 @item --print-opcodes
1031 print the list of instructions with syntax and then exit.
1033 @item --generate-example
1034 print an example of instruction for each possible instruction and then exit.
1035 This option is only useful for testing @command{@value{AS}}.
1041 The following options are available when @command{@value{AS}} is configured
1042 for the SPARC architecture:
1045 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1046 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1047 Explicitly select a variant of the SPARC architecture.
1049 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1050 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1052 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1053 UltraSPARC extensions.
1055 @item -xarch=v8plus | -xarch=v8plusa
1056 For compatibility with the Solaris v9 assembler. These options are
1057 equivalent to -Av8plus and -Av8plusa, respectively.
1060 Warn when the assembler switches to another architecture.
1065 The following options are available when @value{AS} is configured for the 'c54x
1070 Enable extended addressing mode. All addresses and relocations will assume
1071 extended addressing (usually 23 bits).
1072 @item -mcpu=@var{CPU_VERSION}
1073 Sets the CPU version being compiled for.
1074 @item -merrors-to-file @var{FILENAME}
1075 Redirect error output to a file, for broken systems which don't support such
1076 behaviour in the shell.
1081 The following options are available when @value{AS} is configured for
1082 a @sc{mips} processor.
1086 This option sets the largest size of an object that can be referenced
1087 implicitly with the @code{gp} register. It is only accepted for targets that
1088 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1090 @cindex MIPS endianness
1091 @cindex endianness, MIPS
1092 @cindex big endian output, MIPS
1094 Generate ``big endian'' format output.
1096 @cindex little endian output, MIPS
1098 Generate ``little endian'' format output.
1110 Generate code for a particular @sc{mips} Instruction Set Architecture level.
1111 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1112 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1113 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1114 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1116 correspond to generic
1117 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
1118 and @samp{MIPS64 Release 2}
1119 ISA processors, respectively.
1121 @item -march=@var{CPU}
1122 Generate code for a particular @sc{mips} cpu.
1124 @item -mtune=@var{cpu}
1125 Schedule and tune for a particular @sc{mips} cpu.
1129 Cause nops to be inserted if the read of the destination register
1130 of an mfhi or mflo instruction occurs in the following two instructions.
1134 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1135 section instead of the standard ELF .stabs sections.
1139 Control generation of @code{.pdr} sections.
1143 The register sizes are normally inferred from the ISA and ABI, but these
1144 flags force a certain group of registers to be treated as 32 bits wide at
1145 all times. @samp{-mgp32} controls the size of general-purpose registers
1146 and @samp{-mfp32} controls the size of floating-point registers.
1150 Generate code for the MIPS 16 processor. This is equivalent to putting
1151 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1152 turns off this option.
1155 @itemx -mno-smartmips
1156 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1157 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1158 @samp{-mno-smartmips} turns off this option.
1162 Generate code for the MIPS-3D Application Specific Extension.
1163 This tells the assembler to accept MIPS-3D instructions.
1164 @samp{-no-mips3d} turns off this option.
1168 Generate code for the MDMX Application Specific Extension.
1169 This tells the assembler to accept MDMX instructions.
1170 @samp{-no-mdmx} turns off this option.
1174 Generate code for the DSP Release 1 Application Specific Extension.
1175 This tells the assembler to accept DSP Release 1 instructions.
1176 @samp{-mno-dsp} turns off this option.
1180 Generate code for the DSP Release 2 Application Specific Extension.
1181 This option implies -mdsp.
1182 This tells the assembler to accept DSP Release 2 instructions.
1183 @samp{-mno-dspr2} turns off this option.
1187 Generate code for the MT Application Specific Extension.
1188 This tells the assembler to accept MT instructions.
1189 @samp{-mno-mt} turns off this option.
1191 @item --construct-floats
1192 @itemx --no-construct-floats
1193 The @samp{--no-construct-floats} option disables the construction of
1194 double width floating point constants by loading the two halves of the
1195 value into the two single width floating point registers that make up
1196 the double width register. By default @samp{--construct-floats} is
1197 selected, allowing construction of these floating point constants.
1200 @item --emulation=@var{name}
1201 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1202 for some other target, in all respects, including output format (choosing
1203 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1204 debugging information or store symbol table information, and default
1205 endianness. The available configuration names are: @samp{mipsecoff},
1206 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1207 @samp{mipsbelf}. The first two do not alter the default endianness from that
1208 of the primary target for which the assembler was configured; the others change
1209 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1210 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1211 selection in any case.
1213 This option is currently supported only when the primary target
1214 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1215 Furthermore, the primary target or others specified with
1216 @samp{--enable-targets=@dots{}} at configuration time must include support for
1217 the other format, if both are to be available. For example, the Irix 5
1218 configuration includes support for both.
1220 Eventually, this option will support more configurations, with more
1221 fine-grained control over the assembler's behavior, and will be supported for
1225 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1232 Control how to deal with multiplication overflow and division by zero.
1233 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1234 (and only work for Instruction Set Architecture level 2 and higher);
1235 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1239 When this option is used, @command{@value{AS}} will issue a warning every
1240 time it generates a nop instruction from a macro.
1245 The following options are available when @value{AS} is configured for
1251 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1252 The command line option @samp{-nojsri2bsr} can be used to disable it.
1256 Enable or disable the silicon filter behaviour. By default this is disabled.
1257 The default can be overridden by the @samp{-sifilter} command line option.
1260 Alter jump instructions for long displacements.
1262 @item -mcpu=[210|340]
1263 Select the cpu type on the target hardware. This controls which instructions
1267 Assemble for a big endian target.
1270 Assemble for a little endian target.
1276 See the info pages for documentation of the MMIX-specific options.
1280 See the info pages for documentation of the RX-specific options.
1284 The following options are available when @value{AS} is configured for the s390
1290 Select the word size, either 31/32 bits or 64 bits.
1293 Select the architecture mode, either the Enterprise System
1294 Architecture (esa) or the z/Architecture mode (zarch).
1295 @item -march=@var{processor}
1296 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1297 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, or @samp{z10}.
1299 @itemx -mno-regnames
1300 Allow or disallow symbolic names for registers.
1301 @item -mwarn-areg-zero
1302 Warn whenever the operand for a base or index register has been specified
1303 but evaluates to zero.
1311 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1312 for a TMS320C6000 processor.
1316 @c man begin OPTIONS
1317 The following options are available when @value{AS} is configured for a
1318 TMS320C6000 processor.
1320 @c man begin INCLUDE
1321 @include c-tic6x.texi
1322 @c ended inside the included file
1327 @c man begin OPTIONS
1329 The following options are available when @value{AS} is configured for
1330 an Xtensa processor.
1333 @item --text-section-literals | --no-text-section-literals
1334 With @option{--text-@-section-@-literals}, literal pools are interspersed
1335 in the text section. The default is
1336 @option{--no-@-text-@-section-@-literals}, which places literals in a
1337 separate section in the output file. These options only affect literals
1338 referenced via PC-relative @code{L32R} instructions; literals for
1339 absolute mode @code{L32R} instructions are handled separately.
1341 @item --absolute-literals | --no-absolute-literals
1342 Indicate to the assembler whether @code{L32R} instructions use absolute
1343 or PC-relative addressing. The default is to assume absolute addressing
1344 if the Xtensa processor includes the absolute @code{L32R} addressing
1345 option. Otherwise, only the PC-relative @code{L32R} mode can be used.
1347 @item --target-align | --no-target-align
1348 Enable or disable automatic alignment to reduce branch penalties at the
1349 expense of some code density. The default is @option{--target-@-align}.
1351 @item --longcalls | --no-longcalls
1352 Enable or disable transformation of call instructions to allow calls
1353 across a greater range of addresses. The default is
1354 @option{--no-@-longcalls}.
1356 @item --transform | --no-transform
1357 Enable or disable all assembler transformations of Xtensa instructions.
1358 The default is @option{--transform};
1359 @option{--no-transform} should be used only in the rare cases when the
1360 instructions must be exactly as specified in the assembly source.
1362 @item --rename-section @var{oldname}=@var{newname}
1363 When generating output sections, rename the @var{oldname} section to
1369 The following options are available when @value{AS} is configured for
1370 a Z80 family processor.
1373 Assemble for Z80 processor.
1375 Assemble for R800 processor.
1376 @item -ignore-undocumented-instructions
1378 Assemble undocumented Z80 instructions that also work on R800 without warning.
1379 @item -ignore-unportable-instructions
1381 Assemble all undocumented Z80 instructions without warning.
1382 @item -warn-undocumented-instructions
1384 Issue a warning for undocumented Z80 instructions that also work on R800.
1385 @item -warn-unportable-instructions
1387 Issue a warning for undocumented Z80 instructions that do not work on R800.
1388 @item -forbid-undocumented-instructions
1390 Treat all undocumented instructions as errors.
1391 @item -forbid-unportable-instructions
1393 Treat undocumented Z80 instructions that do not work on R800 as errors.
1400 * Manual:: Structure of this Manual
1401 * GNU Assembler:: The GNU Assembler
1402 * Object Formats:: Object File Formats
1403 * Command Line:: Command Line
1404 * Input Files:: Input Files
1405 * Object:: Output (Object) File
1406 * Errors:: Error and Warning Messages
1410 @section Structure of this Manual
1412 @cindex manual, structure and purpose
1413 This manual is intended to describe what you need to know to use
1414 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1415 notation for symbols, constants, and expressions; the directives that
1416 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1419 We also cover special features in the @value{TARGET}
1420 configuration of @command{@value{AS}}, including assembler directives.
1423 This manual also describes some of the machine-dependent features of
1424 various flavors of the assembler.
1427 @cindex machine instructions (not covered)
1428 On the other hand, this manual is @emph{not} intended as an introduction
1429 to programming in assembly language---let alone programming in general!
1430 In a similar vein, we make no attempt to introduce the machine
1431 architecture; we do @emph{not} describe the instruction set, standard
1432 mnemonics, registers or addressing modes that are standard to a
1433 particular architecture.
1435 You may want to consult the manufacturer's
1436 machine architecture manual for this information.
1440 For information on the H8/300 machine instruction set, see @cite{H8/300
1441 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1442 Programming Manual} (Renesas).
1445 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1446 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1447 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1448 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1451 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1455 @c I think this is premature---doc@cygnus.com, 17jan1991
1457 Throughout this manual, we assume that you are running @dfn{GNU},
1458 the portable operating system from the @dfn{Free Software
1459 Foundation, Inc.}. This restricts our attention to certain kinds of
1460 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1461 once this assumption is granted examples and definitions need less
1464 @command{@value{AS}} is part of a team of programs that turn a high-level
1465 human-readable series of instructions into a low-level
1466 computer-readable series of instructions. Different versions of
1467 @command{@value{AS}} are used for different kinds of computer.
1470 @c There used to be a section "Terminology" here, which defined
1471 @c "contents", "byte", "word", and "long". Defining "word" to any
1472 @c particular size is confusing when the .word directive may generate 16
1473 @c bits on one machine and 32 bits on another; in general, for the user
1474 @c version of this manual, none of these terms seem essential to define.
1475 @c They were used very little even in the former draft of the manual;
1476 @c this draft makes an effort to avoid them (except in names of
1480 @section The GNU Assembler
1482 @c man begin DESCRIPTION
1484 @sc{gnu} @command{as} is really a family of assemblers.
1486 This manual describes @command{@value{AS}}, a member of that family which is
1487 configured for the @value{TARGET} architectures.
1489 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1490 should find a fairly similar environment when you use it on another
1491 architecture. Each version has much in common with the others,
1492 including object file formats, most assembler directives (often called
1493 @dfn{pseudo-ops}) and assembler syntax.@refill
1495 @cindex purpose of @sc{gnu} assembler
1496 @command{@value{AS}} is primarily intended to assemble the output of the
1497 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1498 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1499 assemble correctly everything that other assemblers for the same
1500 machine would assemble.
1502 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1505 @c This remark should appear in generic version of manual; assumption
1506 @c here is that generic version sets M680x0.
1507 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1508 assembler for the same architecture; for example, we know of several
1509 incompatible versions of 680x0 assembly language syntax.
1514 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1515 program in one pass of the source file. This has a subtle impact on the
1516 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1518 @node Object Formats
1519 @section Object File Formats
1521 @cindex object file format
1522 The @sc{gnu} assembler can be configured to produce several alternative
1523 object file formats. For the most part, this does not affect how you
1524 write assembly language programs; but directives for debugging symbols
1525 are typically different in different file formats. @xref{Symbol
1526 Attributes,,Symbol Attributes}.
1529 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1530 @value{OBJ-NAME} format object files.
1532 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1534 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1535 @code{b.out} or COFF format object files.
1538 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1539 SOM or ELF format object files.
1544 @section Command Line
1546 @cindex command line conventions
1548 After the program name @command{@value{AS}}, the command line may contain
1549 options and file names. Options may appear in any order, and may be
1550 before, after, or between file names. The order of file names is
1553 @cindex standard input, as input file
1555 @file{--} (two hyphens) by itself names the standard input file
1556 explicitly, as one of the files for @command{@value{AS}} to assemble.
1558 @cindex options, command line
1559 Except for @samp{--} any command line argument that begins with a
1560 hyphen (@samp{-}) is an option. Each option changes the behavior of
1561 @command{@value{AS}}. No option changes the way another option works. An
1562 option is a @samp{-} followed by one or more letters; the case of
1563 the letter is important. All options are optional.
1565 Some options expect exactly one file name to follow them. The file
1566 name may either immediately follow the option's letter (compatible
1567 with older assemblers) or it may be the next command argument (@sc{gnu}
1568 standard). These two command lines are equivalent:
1571 @value{AS} -o my-object-file.o mumble.s
1572 @value{AS} -omy-object-file.o mumble.s
1576 @section Input Files
1579 @cindex source program
1580 @cindex files, input
1581 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1582 describe the program input to one run of @command{@value{AS}}. The program may
1583 be in one or more files; how the source is partitioned into files
1584 doesn't change the meaning of the source.
1586 @c I added "con" prefix to "catenation" just to prove I can overcome my
1587 @c APL training... doc@cygnus.com
1588 The source program is a concatenation of the text in all the files, in the
1591 @c man begin DESCRIPTION
1592 Each time you run @command{@value{AS}} it assembles exactly one source
1593 program. The source program is made up of one or more files.
1594 (The standard input is also a file.)
1596 You give @command{@value{AS}} a command line that has zero or more input file
1597 names. The input files are read (from left file name to right). A
1598 command line argument (in any position) that has no special meaning
1599 is taken to be an input file name.
1601 If you give @command{@value{AS}} no file names it attempts to read one input file
1602 from the @command{@value{AS}} standard input, which is normally your terminal. You
1603 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1606 Use @samp{--} if you need to explicitly name the standard input file
1607 in your command line.
1609 If the source is empty, @command{@value{AS}} produces a small, empty object
1614 @subheading Filenames and Line-numbers
1616 @cindex input file linenumbers
1617 @cindex line numbers, in input files
1618 There are two ways of locating a line in the input file (or files) and
1619 either may be used in reporting error messages. One way refers to a line
1620 number in a physical file; the other refers to a line number in a
1621 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1623 @dfn{Physical files} are those files named in the command line given
1624 to @command{@value{AS}}.
1626 @dfn{Logical files} are simply names declared explicitly by assembler
1627 directives; they bear no relation to physical files. Logical file names help
1628 error messages reflect the original source file, when @command{@value{AS}} source
1629 is itself synthesized from other files. @command{@value{AS}} understands the
1630 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1631 @ref{File,,@code{.file}}.
1634 @section Output (Object) File
1640 Every time you run @command{@value{AS}} it produces an output file, which is
1641 your assembly language program translated into numbers. This file
1642 is the object file. Its default name is
1650 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1652 You can give it another name by using the @option{-o} option. Conventionally,
1653 object file names end with @file{.o}. The default name is used for historical
1654 reasons: older assemblers were capable of assembling self-contained programs
1655 directly into a runnable program. (For some formats, this isn't currently
1656 possible, but it can be done for the @code{a.out} format.)
1660 The object file is meant for input to the linker @code{@value{LD}}. It contains
1661 assembled program code, information to help @code{@value{LD}} integrate
1662 the assembled program into a runnable file, and (optionally) symbolic
1663 information for the debugger.
1665 @c link above to some info file(s) like the description of a.out.
1666 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1669 @section Error and Warning Messages
1671 @c man begin DESCRIPTION
1673 @cindex error messages
1674 @cindex warning messages
1675 @cindex messages from assembler
1676 @command{@value{AS}} may write warnings and error messages to the standard error
1677 file (usually your terminal). This should not happen when a compiler
1678 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1679 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1680 grave problem that stops the assembly.
1684 @cindex format of warning messages
1685 Warning messages have the format
1688 file_name:@b{NNN}:Warning Message Text
1692 @cindex line numbers, in warnings/errors
1693 (where @b{NNN} is a line number). If a logical file name has been given
1694 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1695 the current input file is used. If a logical line number was given
1697 (@pxref{Line,,@code{.line}})
1699 then it is used to calculate the number printed,
1700 otherwise the actual line in the current source file is printed. The
1701 message text is intended to be self explanatory (in the grand Unix
1704 @cindex format of error messages
1705 Error messages have the format
1707 file_name:@b{NNN}:FATAL:Error Message Text
1709 The file name and line number are derived as for warning
1710 messages. The actual message text may be rather less explanatory
1711 because many of them aren't supposed to happen.
1714 @chapter Command-Line Options
1716 @cindex options, all versions of assembler
1717 This chapter describes command-line options available in @emph{all}
1718 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1719 for options specific
1721 to the @value{TARGET} target.
1724 to particular machine architectures.
1727 @c man begin DESCRIPTION
1729 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1730 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1731 The assembler arguments must be separated from each other (and the @samp{-Wa})
1732 by commas. For example:
1735 gcc -c -g -O -Wa,-alh,-L file.c
1739 This passes two options to the assembler: @samp{-alh} (emit a listing to
1740 standard output with high-level and assembly source) and @samp{-L} (retain
1741 local symbols in the symbol table).
1743 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1744 command-line options are automatically passed to the assembler by the compiler.
1745 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1746 precisely what options it passes to each compilation pass, including the
1752 * a:: -a[cdghlns] enable listings
1753 * alternate:: --alternate enable alternate macro syntax
1754 * D:: -D for compatibility
1755 * f:: -f to work faster
1756 * I:: -I for .include search path
1757 @ifclear DIFF-TBL-KLUGE
1758 * K:: -K for compatibility
1760 @ifset DIFF-TBL-KLUGE
1761 * K:: -K for difference tables
1764 * L:: -L to retain local symbols
1765 * listing:: --listing-XXX to configure listing output
1766 * M:: -M or --mri to assemble in MRI compatibility mode
1767 * MD:: --MD for dependency tracking
1768 * o:: -o to name the object file
1769 * R:: -R to join data and text sections
1770 * statistics:: --statistics to see statistics about assembly
1771 * traditional-format:: --traditional-format for compatible output
1772 * v:: -v to announce version
1773 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1774 * Z:: -Z to make object file even after errors
1778 @section Enable Listings: @option{-a[cdghlns]}
1788 @cindex listings, enabling
1789 @cindex assembly listings, enabling
1791 These options enable listing output from the assembler. By itself,
1792 @samp{-a} requests high-level, assembly, and symbols listing.
1793 You can use other letters to select specific options for the list:
1794 @samp{-ah} requests a high-level language listing,
1795 @samp{-al} requests an output-program assembly listing, and
1796 @samp{-as} requests a symbol table listing.
1797 High-level listings require that a compiler debugging option like
1798 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1801 Use the @samp{-ag} option to print a first section with general assembly
1802 information, like @value{AS} version, switches passed, or time stamp.
1804 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1805 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1806 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1807 omitted from the listing.
1809 Use the @samp{-ad} option to omit debugging directives from the
1812 Once you have specified one of these options, you can further control
1813 listing output and its appearance using the directives @code{.list},
1814 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1816 The @samp{-an} option turns off all forms processing.
1817 If you do not request listing output with one of the @samp{-a} options, the
1818 listing-control directives have no effect.
1820 The letters after @samp{-a} may be combined into one option,
1821 @emph{e.g.}, @samp{-aln}.
1823 Note if the assembler source is coming from the standard input (e.g.,
1825 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1826 is being used) then the listing will not contain any comments or preprocessor
1827 directives. This is because the listing code buffers input source lines from
1828 stdin only after they have been preprocessed by the assembler. This reduces
1829 memory usage and makes the code more efficient.
1832 @section @option{--alternate}
1835 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1838 @section @option{-D}
1841 This option has no effect whatsoever, but it is accepted to make it more
1842 likely that scripts written for other assemblers also work with
1843 @command{@value{AS}}.
1846 @section Work Faster: @option{-f}
1849 @cindex trusted compiler
1850 @cindex faster processing (@option{-f})
1851 @samp{-f} should only be used when assembling programs written by a
1852 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1853 and comment preprocessing on
1854 the input file(s) before assembling them. @xref{Preprocessing,
1858 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1859 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1864 @section @code{.include} Search Path: @option{-I} @var{path}
1866 @kindex -I @var{path}
1867 @cindex paths for @code{.include}
1868 @cindex search path for @code{.include}
1869 @cindex @code{include} directive search path
1870 Use this option to add a @var{path} to the list of directories
1871 @command{@value{AS}} searches for files specified in @code{.include}
1872 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1873 many times as necessary to include a variety of paths. The current
1874 working directory is always searched first; after that, @command{@value{AS}}
1875 searches any @samp{-I} directories in the same order as they were
1876 specified (left to right) on the command line.
1879 @section Difference Tables: @option{-K}
1882 @ifclear DIFF-TBL-KLUGE
1883 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1884 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1885 where it can be used to warn when the assembler alters the machine code
1886 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1887 family does not have the addressing limitations that sometimes lead to this
1888 alteration on other platforms.
1891 @ifset DIFF-TBL-KLUGE
1892 @cindex difference tables, warning
1893 @cindex warning for altered difference tables
1894 @command{@value{AS}} sometimes alters the code emitted for directives of the
1895 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
1896 You can use the @samp{-K} option if you want a warning issued when this
1901 @section Include Local Symbols: @option{-L}
1904 @cindex local symbols, retaining in output
1905 Symbols beginning with system-specific local label prefixes, typically
1906 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
1907 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
1908 such symbols when debugging, because they are intended for the use of
1909 programs (like compilers) that compose assembler programs, not for your
1910 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
1911 such symbols, so you do not normally debug with them.
1913 This option tells @command{@value{AS}} to retain those local symbols
1914 in the object file. Usually if you do this you also tell the linker
1915 @code{@value{LD}} to preserve those symbols.
1918 @section Configuring listing output: @option{--listing}
1920 The listing feature of the assembler can be enabled via the command line switch
1921 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1922 hex dump of the corresponding locations in the output object file, and displays
1923 them as a listing file. The format of this listing can be controlled by
1924 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
1925 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
1926 @code{.psize} (@pxref{Psize}), and
1927 @code{.eject} (@pxref{Eject}) and also by the following switches:
1930 @item --listing-lhs-width=@samp{number}
1931 @kindex --listing-lhs-width
1932 @cindex Width of first line disassembly output
1933 Sets the maximum width, in words, of the first line of the hex byte dump. This
1934 dump appears on the left hand side of the listing output.
1936 @item --listing-lhs-width2=@samp{number}
1937 @kindex --listing-lhs-width2
1938 @cindex Width of continuation lines of disassembly output
1939 Sets the maximum width, in words, of any further lines of the hex byte dump for
1940 a given input source line. If this value is not specified, it defaults to being
1941 the same as the value specified for @samp{--listing-lhs-width}. If neither
1942 switch is used the default is to one.
1944 @item --listing-rhs-width=@samp{number}
1945 @kindex --listing-rhs-width
1946 @cindex Width of source line output
1947 Sets the maximum width, in characters, of the source line that is displayed
1948 alongside the hex dump. The default value for this parameter is 100. The
1949 source line is displayed on the right hand side of the listing output.
1951 @item --listing-cont-lines=@samp{number}
1952 @kindex --listing-cont-lines
1953 @cindex Maximum number of continuation lines
1954 Sets the maximum number of continuation lines of hex dump that will be
1955 displayed for a given single line of source input. The default value is 4.
1959 @section Assemble in MRI Compatibility Mode: @option{-M}
1962 @cindex MRI compatibility mode
1963 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1964 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1965 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1966 configured target) assembler from Microtec Research. The exact nature of the
1967 MRI syntax will not be documented here; see the MRI manuals for more
1968 information. Note in particular that the handling of macros and macro
1969 arguments is somewhat different. The purpose of this option is to permit
1970 assembling existing MRI assembler code using @command{@value{AS}}.
1972 The MRI compatibility is not complete. Certain operations of the MRI assembler
1973 depend upon its object file format, and can not be supported using other object
1974 file formats. Supporting these would require enhancing each object file format
1975 individually. These are:
1978 @item global symbols in common section
1980 The m68k MRI assembler supports common sections which are merged by the linker.
1981 Other object file formats do not support this. @command{@value{AS}} handles
1982 common sections by treating them as a single common symbol. It permits local
1983 symbols to be defined within a common section, but it can not support global
1984 symbols, since it has no way to describe them.
1986 @item complex relocations
1988 The MRI assemblers support relocations against a negated section address, and
1989 relocations which combine the start addresses of two or more sections. These
1990 are not support by other object file formats.
1992 @item @code{END} pseudo-op specifying start address
1994 The MRI @code{END} pseudo-op permits the specification of a start address.
1995 This is not supported by other object file formats. The start address may
1996 instead be specified using the @option{-e} option to the linker, or in a linker
1999 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2001 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2002 name to the output file. This is not supported by other object file formats.
2004 @item @code{ORG} pseudo-op
2006 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2007 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2008 which changes the location within the current section. Absolute sections are
2009 not supported by other object file formats. The address of a section may be
2010 assigned within a linker script.
2013 There are some other features of the MRI assembler which are not supported by
2014 @command{@value{AS}}, typically either because they are difficult or because they
2015 seem of little consequence. Some of these may be supported in future releases.
2019 @item EBCDIC strings
2021 EBCDIC strings are not supported.
2023 @item packed binary coded decimal
2025 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2026 and @code{DCB.P} pseudo-ops are not supported.
2028 @item @code{FEQU} pseudo-op
2030 The m68k @code{FEQU} pseudo-op is not supported.
2032 @item @code{NOOBJ} pseudo-op
2034 The m68k @code{NOOBJ} pseudo-op is not supported.
2036 @item @code{OPT} branch control options
2038 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2039 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2040 relaxes all branches, whether forward or backward, to an appropriate size, so
2041 these options serve no purpose.
2043 @item @code{OPT} list control options
2045 The following m68k @code{OPT} list control options are ignored: @code{C},
2046 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2047 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2049 @item other @code{OPT} options
2051 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2052 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2054 @item @code{OPT} @code{D} option is default
2056 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2057 @code{OPT NOD} may be used to turn it off.
2059 @item @code{XREF} pseudo-op.
2061 The m68k @code{XREF} pseudo-op is ignored.
2063 @item @code{.debug} pseudo-op
2065 The i960 @code{.debug} pseudo-op is not supported.
2067 @item @code{.extended} pseudo-op
2069 The i960 @code{.extended} pseudo-op is not supported.
2071 @item @code{.list} pseudo-op.
2073 The various options of the i960 @code{.list} pseudo-op are not supported.
2075 @item @code{.optimize} pseudo-op
2077 The i960 @code{.optimize} pseudo-op is not supported.
2079 @item @code{.output} pseudo-op
2081 The i960 @code{.output} pseudo-op is not supported.
2083 @item @code{.setreal} pseudo-op
2085 The i960 @code{.setreal} pseudo-op is not supported.
2090 @section Dependency Tracking: @option{--MD}
2093 @cindex dependency tracking
2096 @command{@value{AS}} can generate a dependency file for the file it creates. This
2097 file consists of a single rule suitable for @code{make} describing the
2098 dependencies of the main source file.
2100 The rule is written to the file named in its argument.
2102 This feature is used in the automatic updating of makefiles.
2105 @section Name the Object File: @option{-o}
2108 @cindex naming object file
2109 @cindex object file name
2110 There is always one object file output when you run @command{@value{AS}}. By
2111 default it has the name
2114 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2128 You use this option (which takes exactly one filename) to give the
2129 object file a different name.
2131 Whatever the object file is called, @command{@value{AS}} overwrites any
2132 existing file of the same name.
2135 @section Join Data and Text Sections: @option{-R}
2138 @cindex data and text sections, joining
2139 @cindex text and data sections, joining
2140 @cindex joining text and data sections
2141 @cindex merging text and data sections
2142 @option{-R} tells @command{@value{AS}} to write the object file as if all
2143 data-section data lives in the text section. This is only done at
2144 the very last moment: your binary data are the same, but data
2145 section parts are relocated differently. The data section part of
2146 your object file is zero bytes long because all its bytes are
2147 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2149 When you specify @option{-R} it would be possible to generate shorter
2150 address displacements (because we do not have to cross between text and
2151 data section). We refrain from doing this simply for compatibility with
2152 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2155 When @command{@value{AS}} is configured for COFF or ELF output,
2156 this option is only useful if you use sections named @samp{.text} and
2161 @option{-R} is not supported for any of the HPPA targets. Using
2162 @option{-R} generates a warning from @command{@value{AS}}.
2166 @section Display Assembly Statistics: @option{--statistics}
2168 @kindex --statistics
2169 @cindex statistics, about assembly
2170 @cindex time, total for assembly
2171 @cindex space used, maximum for assembly
2172 Use @samp{--statistics} to display two statistics about the resources used by
2173 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2174 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2177 @node traditional-format
2178 @section Compatible Output: @option{--traditional-format}
2180 @kindex --traditional-format
2181 For some targets, the output of @command{@value{AS}} is different in some ways
2182 from the output of some existing assembler. This switch requests
2183 @command{@value{AS}} to use the traditional format instead.
2185 For example, it disables the exception frame optimizations which
2186 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2189 @section Announce Version: @option{-v}
2193 @cindex assembler version
2194 @cindex version of assembler
2195 You can find out what version of as is running by including the
2196 option @samp{-v} (which you can also spell as @samp{-version}) on the
2200 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2202 @command{@value{AS}} should never give a warning or error message when
2203 assembling compiler output. But programs written by people often
2204 cause @command{@value{AS}} to give a warning that a particular assumption was
2205 made. All such warnings are directed to the standard error file.
2209 @cindex suppressing warnings
2210 @cindex warnings, suppressing
2211 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2212 This only affects the warning messages: it does not change any particular of
2213 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2216 @kindex --fatal-warnings
2217 @cindex errors, caused by warnings
2218 @cindex warnings, causing error
2219 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2220 files that generate warnings to be in error.
2223 @cindex warnings, switching on
2224 You can switch these options off again by specifying @option{--warn}, which
2225 causes warnings to be output as usual.
2228 @section Generate Object File in Spite of Errors: @option{-Z}
2229 @cindex object file, after errors
2230 @cindex errors, continuing after
2231 After an error message, @command{@value{AS}} normally produces no output. If for
2232 some reason you are interested in object file output even after
2233 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2234 option. If there are any errors, @command{@value{AS}} continues anyways, and
2235 writes an object file after a final warning message of the form @samp{@var{n}
2236 errors, @var{m} warnings, generating bad object file.}
2241 @cindex machine-independent syntax
2242 @cindex syntax, machine-independent
2243 This chapter describes the machine-independent syntax allowed in a
2244 source file. @command{@value{AS}} syntax is similar to what many other
2245 assemblers use; it is inspired by the BSD 4.2
2250 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2254 * Preprocessing:: Preprocessing
2255 * Whitespace:: Whitespace
2256 * Comments:: Comments
2257 * Symbol Intro:: Symbols
2258 * Statements:: Statements
2259 * Constants:: Constants
2263 @section Preprocessing
2265 @cindex preprocessing
2266 The @command{@value{AS}} internal preprocessor:
2268 @cindex whitespace, removed by preprocessor
2270 adjusts and removes extra whitespace. It leaves one space or tab before
2271 the keywords on a line, and turns any other whitespace on the line into
2274 @cindex comments, removed by preprocessor
2276 removes all comments, replacing them with a single space, or an
2277 appropriate number of newlines.
2279 @cindex constants, converted by preprocessor
2281 converts character constants into the appropriate numeric values.
2284 It does not do macro processing, include file handling, or
2285 anything else you may get from your C compiler's preprocessor. You can
2286 do include file processing with the @code{.include} directive
2287 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2288 to get other ``CPP'' style preprocessing by giving the input file a
2289 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2290 Output, gcc.info, Using GNU CC}.
2292 Excess whitespace, comments, and character constants
2293 cannot be used in the portions of the input text that are not
2296 @cindex turning preprocessing on and off
2297 @cindex preprocessing, turning on and off
2300 If the first line of an input file is @code{#NO_APP} or if you use the
2301 @samp{-f} option, whitespace and comments are not removed from the input file.
2302 Within an input file, you can ask for whitespace and comment removal in
2303 specific portions of the by putting a line that says @code{#APP} before the
2304 text that may contain whitespace or comments, and putting a line that says
2305 @code{#NO_APP} after this text. This feature is mainly intend to support
2306 @code{asm} statements in compilers whose output is otherwise free of comments
2313 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2314 Whitespace is used to separate symbols, and to make programs neater for
2315 people to read. Unless within character constants
2316 (@pxref{Characters,,Character Constants}), any whitespace means the same
2317 as exactly one space.
2323 There are two ways of rendering comments to @command{@value{AS}}. In both
2324 cases the comment is equivalent to one space.
2326 Anything from @samp{/*} through the next @samp{*/} is a comment.
2327 This means you may not nest these comments.
2331 The only way to include a newline ('\n') in a comment
2332 is to use this sort of comment.
2335 /* This sort of comment does not nest. */
2338 @cindex line comment character
2339 Anything from the @dfn{line comment} character to the next newline
2340 is considered a comment and is ignored. The line comment character is
2342 @samp{;} on the ARC;
2345 @samp{@@} on the ARM;
2348 @samp{;} for the H8/300 family;
2351 @samp{;} for the HPPA;
2354 @samp{#} on the i386 and x86-64;
2357 @samp{#} on the i960;
2360 @samp{;} for the PDP-11;
2363 @samp{;} for picoJava;
2366 @samp{#} for Motorola PowerPC;
2369 @samp{#} for IBM S/390;
2372 @samp{#} for the Sunplus SCORE;
2375 @samp{!} for the Renesas / SuperH SH;
2378 @samp{!} on the SPARC;
2381 @samp{#} on the ip2k;
2384 @samp{#} on the m32c;
2387 @samp{#} on the m32r;
2390 @samp{|} on the 680x0;
2393 @samp{#} on the 68HC11 and 68HC12;
2399 @samp{;} on the TMS320C6X;
2402 @samp{#} on the Vax;
2405 @samp{;} for the Z80;
2408 @samp{!} for the Z8000;
2411 @samp{#} on the V850;
2414 @samp{#} for Xtensa systems;
2416 see @ref{Machine Dependencies}. @refill
2417 @c FIXME What about i860?
2420 On some machines there are two different line comment characters. One
2421 character only begins a comment if it is the first non-whitespace character on
2422 a line, while the other always begins a comment.
2426 The V850 assembler also supports a double dash as starting a comment that
2427 extends to the end of the line.
2433 @cindex lines starting with @code{#}
2434 @cindex logical line numbers
2435 To be compatible with past assemblers, lines that begin with @samp{#} have a
2436 special interpretation. Following the @samp{#} should be an absolute
2437 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2438 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2439 new logical file name. The rest of the line, if any, should be whitespace.
2441 If the first non-whitespace characters on the line are not numeric,
2442 the line is ignored. (Just like a comment.)
2445 # This is an ordinary comment.
2446 # 42-6 "new_file_name" # New logical file name
2447 # This is logical line # 36.
2449 This feature is deprecated, and may disappear from future versions
2450 of @command{@value{AS}}.
2455 @cindex characters used in symbols
2456 @ifclear SPECIAL-SYMS
2457 A @dfn{symbol} is one or more characters chosen from the set of all
2458 letters (both upper and lower case), digits and the three characters
2464 A @dfn{symbol} is one or more characters chosen from the set of all
2465 letters (both upper and lower case), digits and the three characters
2466 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2472 On most machines, you can also use @code{$} in symbol names; exceptions
2473 are noted in @ref{Machine Dependencies}.
2475 No symbol may begin with a digit. Case is significant.
2476 There is no length limit: all characters are significant. Symbols are
2477 delimited by characters not in that set, or by the beginning of a file
2478 (since the source program must end with a newline, the end of a file is
2479 not a possible symbol delimiter). @xref{Symbols}.
2480 @cindex length of symbols
2485 @cindex statements, structure of
2486 @cindex line separator character
2487 @cindex statement separator character
2489 @ifclear abnormal-separator
2490 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2491 semicolon (@samp{;}). The newline or semicolon is considered part of
2492 the preceding statement. Newlines and semicolons within character
2493 constants are an exception: they do not end statements.
2495 @ifset abnormal-separator
2497 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2498 point (@samp{!}). The newline or exclamation point is considered part of the
2499 preceding statement. Newlines and exclamation points within character
2500 constants are an exception: they do not end statements.
2503 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2504 H8/300) a dollar sign (@samp{$}); or (for the Renesas-SH) a semicolon
2505 (@samp{;}). The newline or separator character is considered part of
2506 the preceding statement. Newlines and separators within character
2507 constants are an exception: they do not end statements.
2512 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2513 separator character. (The line separator is usually @samp{;}, unless this
2514 conflicts with the comment character; see @ref{Machine Dependencies}.) The
2515 newline or separator character is considered part of the preceding
2516 statement. Newlines and separators within character constants are an
2517 exception: they do not end statements.
2520 @cindex newline, required at file end
2521 @cindex EOF, newline must precede
2522 It is an error to end any statement with end-of-file: the last
2523 character of any input file should be a newline.@refill
2525 An empty statement is allowed, and may include whitespace. It is ignored.
2527 @cindex instructions and directives
2528 @cindex directives and instructions
2529 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2530 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2532 A statement begins with zero or more labels, optionally followed by a
2533 key symbol which determines what kind of statement it is. The key
2534 symbol determines the syntax of the rest of the statement. If the
2535 symbol begins with a dot @samp{.} then the statement is an assembler
2536 directive: typically valid for any computer. If the symbol begins with
2537 a letter the statement is an assembly language @dfn{instruction}: it
2538 assembles into a machine language instruction.
2540 Different versions of @command{@value{AS}} for different computers
2541 recognize different instructions. In fact, the same symbol may
2542 represent a different instruction in a different computer's assembly
2546 @cindex @code{:} (label)
2547 @cindex label (@code{:})
2548 A label is a symbol immediately followed by a colon (@code{:}).
2549 Whitespace before a label or after a colon is permitted, but you may not
2550 have whitespace between a label's symbol and its colon. @xref{Labels}.
2553 For HPPA targets, labels need not be immediately followed by a colon, but
2554 the definition of a label must begin in column zero. This also implies that
2555 only one label may be defined on each line.
2559 label: .directive followed by something
2560 another_label: # This is an empty statement.
2561 instruction operand_1, operand_2, @dots{}
2568 A constant is a number, written so that its value is known by
2569 inspection, without knowing any context. Like this:
2572 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2573 .ascii "Ring the bell\7" # A string constant.
2574 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2575 .float 0f-314159265358979323846264338327\
2576 95028841971.693993751E-40 # - pi, a flonum.
2581 * Characters:: Character Constants
2582 * Numbers:: Number Constants
2586 @subsection Character Constants
2588 @cindex character constants
2589 @cindex constants, character
2590 There are two kinds of character constants. A @dfn{character} stands
2591 for one character in one byte and its value may be used in
2592 numeric expressions. String constants (properly called string
2593 @emph{literals}) are potentially many bytes and their values may not be
2594 used in arithmetic expressions.
2598 * Chars:: Characters
2602 @subsubsection Strings
2604 @cindex string constants
2605 @cindex constants, string
2606 A @dfn{string} is written between double-quotes. It may contain
2607 double-quotes or null characters. The way to get special characters
2608 into a string is to @dfn{escape} these characters: precede them with
2609 a backslash @samp{\} character. For example @samp{\\} represents
2610 one backslash: the first @code{\} is an escape which tells
2611 @command{@value{AS}} to interpret the second character literally as a backslash
2612 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2613 escape character). The complete list of escapes follows.
2615 @cindex escape codes, character
2616 @cindex character escape codes
2619 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2621 @cindex @code{\b} (backspace character)
2622 @cindex backspace (@code{\b})
2624 Mnemonic for backspace; for ASCII this is octal code 010.
2627 @c Mnemonic for EOText; for ASCII this is octal code 004.
2629 @cindex @code{\f} (formfeed character)
2630 @cindex formfeed (@code{\f})
2632 Mnemonic for FormFeed; for ASCII this is octal code 014.
2634 @cindex @code{\n} (newline character)
2635 @cindex newline (@code{\n})
2637 Mnemonic for newline; for ASCII this is octal code 012.
2640 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2642 @cindex @code{\r} (carriage return character)
2643 @cindex carriage return (@code{\r})
2645 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2648 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2649 @c other assemblers.
2651 @cindex @code{\t} (tab)
2652 @cindex tab (@code{\t})
2654 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2657 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2658 @c @item \x @var{digit} @var{digit} @var{digit}
2659 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2661 @cindex @code{\@var{ddd}} (octal character code)
2662 @cindex octal character code (@code{\@var{ddd}})
2663 @item \ @var{digit} @var{digit} @var{digit}
2664 An octal character code. The numeric code is 3 octal digits.
2665 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2666 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2668 @cindex @code{\@var{xd...}} (hex character code)
2669 @cindex hex character code (@code{\@var{xd...}})
2670 @item \@code{x} @var{hex-digits...}
2671 A hex character code. All trailing hex digits are combined. Either upper or
2672 lower case @code{x} works.
2674 @cindex @code{\\} (@samp{\} character)
2675 @cindex backslash (@code{\\})
2677 Represents one @samp{\} character.
2680 @c Represents one @samp{'} (accent acute) character.
2681 @c This is needed in single character literals
2682 @c (@xref{Characters,,Character Constants}.) to represent
2685 @cindex @code{\"} (doublequote character)
2686 @cindex doublequote (@code{\"})
2688 Represents one @samp{"} character. Needed in strings to represent
2689 this character, because an unescaped @samp{"} would end the string.
2691 @item \ @var{anything-else}
2692 Any other character when escaped by @kbd{\} gives a warning, but
2693 assembles as if the @samp{\} was not present. The idea is that if
2694 you used an escape sequence you clearly didn't want the literal
2695 interpretation of the following character. However @command{@value{AS}} has no
2696 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2697 code and warns you of the fact.
2700 Which characters are escapable, and what those escapes represent,
2701 varies widely among assemblers. The current set is what we think
2702 the BSD 4.2 assembler recognizes, and is a subset of what most C
2703 compilers recognize. If you are in doubt, do not use an escape
2707 @subsubsection Characters
2709 @cindex single character constant
2710 @cindex character, single
2711 @cindex constant, single character
2712 A single character may be written as a single quote immediately
2713 followed by that character. The same escapes apply to characters as
2714 to strings. So if you want to write the character backslash, you
2715 must write @kbd{'\\} where the first @code{\} escapes the second
2716 @code{\}. As you can see, the quote is an acute accent, not a
2717 grave accent. A newline
2719 @ifclear abnormal-separator
2720 (or semicolon @samp{;})
2722 @ifset abnormal-separator
2724 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2729 immediately following an acute accent is taken as a literal character
2730 and does not count as the end of a statement. The value of a character
2731 constant in a numeric expression is the machine's byte-wide code for
2732 that character. @command{@value{AS}} assumes your character code is ASCII:
2733 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2736 @subsection Number Constants
2738 @cindex constants, number
2739 @cindex number constants
2740 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2741 are stored in the target machine. @emph{Integers} are numbers that
2742 would fit into an @code{int} in the C language. @emph{Bignums} are
2743 integers, but they are stored in more than 32 bits. @emph{Flonums}
2744 are floating point numbers, described below.
2747 * Integers:: Integers
2752 * Bit Fields:: Bit Fields
2758 @subsubsection Integers
2760 @cindex constants, integer
2762 @cindex binary integers
2763 @cindex integers, binary
2764 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2765 the binary digits @samp{01}.
2767 @cindex octal integers
2768 @cindex integers, octal
2769 An octal integer is @samp{0} followed by zero or more of the octal
2770 digits (@samp{01234567}).
2772 @cindex decimal integers
2773 @cindex integers, decimal
2774 A decimal integer starts with a non-zero digit followed by zero or
2775 more digits (@samp{0123456789}).
2777 @cindex hexadecimal integers
2778 @cindex integers, hexadecimal
2779 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2780 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2782 Integers have the usual values. To denote a negative integer, use
2783 the prefix operator @samp{-} discussed under expressions
2784 (@pxref{Prefix Ops,,Prefix Operators}).
2787 @subsubsection Bignums
2790 @cindex constants, bignum
2791 A @dfn{bignum} has the same syntax and semantics as an integer
2792 except that the number (or its negative) takes more than 32 bits to
2793 represent in binary. The distinction is made because in some places
2794 integers are permitted while bignums are not.
2797 @subsubsection Flonums
2799 @cindex floating point numbers
2800 @cindex constants, floating point
2802 @cindex precision, floating point
2803 A @dfn{flonum} represents a floating point number. The translation is
2804 indirect: a decimal floating point number from the text is converted by
2805 @command{@value{AS}} to a generic binary floating point number of more than
2806 sufficient precision. This generic floating point number is converted
2807 to a particular computer's floating point format (or formats) by a
2808 portion of @command{@value{AS}} specialized to that computer.
2810 A flonum is written by writing (in order)
2815 (@samp{0} is optional on the HPPA.)
2819 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2821 @kbd{e} is recommended. Case is not important.
2823 @c FIXME: verify if flonum syntax really this vague for most cases
2824 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2825 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2828 On the H8/300, Renesas / SuperH SH,
2829 and AMD 29K architectures, the letter must be
2830 one of the letters @samp{DFPRSX} (in upper or lower case).
2832 On the ARC, the letter must be one of the letters @samp{DFRS}
2833 (in upper or lower case).
2835 On the Intel 960 architecture, the letter must be
2836 one of the letters @samp{DFT} (in upper or lower case).
2838 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2842 One of the letters @samp{DFRS} (in upper or lower case).
2845 One of the letters @samp{DFPRSX} (in upper or lower case).
2848 The letter @samp{E} (upper case only).
2851 One of the letters @samp{DFT} (in upper or lower case).
2856 An optional sign: either @samp{+} or @samp{-}.
2859 An optional @dfn{integer part}: zero or more decimal digits.
2862 An optional @dfn{fractional part}: @samp{.} followed by zero
2863 or more decimal digits.
2866 An optional exponent, consisting of:
2870 An @samp{E} or @samp{e}.
2871 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2872 @c principle this can perfectly well be different on different targets.
2874 Optional sign: either @samp{+} or @samp{-}.
2876 One or more decimal digits.
2881 At least one of the integer part or the fractional part must be
2882 present. The floating point number has the usual base-10 value.
2884 @command{@value{AS}} does all processing using integers. Flonums are computed
2885 independently of any floating point hardware in the computer running
2886 @command{@value{AS}}.
2890 @c Bit fields are written as a general facility but are also controlled
2891 @c by a conditional-compilation flag---which is as of now (21mar91)
2892 @c turned on only by the i960 config of GAS.
2894 @subsubsection Bit Fields
2897 @cindex constants, bit field
2898 You can also define numeric constants as @dfn{bit fields}.
2899 Specify two numbers separated by a colon---
2901 @var{mask}:@var{value}
2904 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2907 The resulting number is then packed
2909 @c this conditional paren in case bit fields turned on elsewhere than 960
2910 (in host-dependent byte order)
2912 into a field whose width depends on which assembler directive has the
2913 bit-field as its argument. Overflow (a result from the bitwise and
2914 requiring more binary digits to represent) is not an error; instead,
2915 more constants are generated, of the specified width, beginning with the
2916 least significant digits.@refill
2918 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2919 @code{.short}, and @code{.word} accept bit-field arguments.
2924 @chapter Sections and Relocation
2929 * Secs Background:: Background
2930 * Ld Sections:: Linker Sections
2931 * As Sections:: Assembler Internal Sections
2932 * Sub-Sections:: Sub-Sections
2936 @node Secs Background
2939 Roughly, a section is a range of addresses, with no gaps; all data
2940 ``in'' those addresses is treated the same for some particular purpose.
2941 For example there may be a ``read only'' section.
2943 @cindex linker, and assembler
2944 @cindex assembler, and linker
2945 The linker @code{@value{LD}} reads many object files (partial programs) and
2946 combines their contents to form a runnable program. When @command{@value{AS}}
2947 emits an object file, the partial program is assumed to start at address 0.
2948 @code{@value{LD}} assigns the final addresses for the partial program, so that
2949 different partial programs do not overlap. This is actually an
2950 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2953 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2954 addresses. These blocks slide to their run-time addresses as rigid
2955 units; their length does not change and neither does the order of bytes
2956 within them. Such a rigid unit is called a @emph{section}. Assigning
2957 run-time addresses to sections is called @dfn{relocation}. It includes
2958 the task of adjusting mentions of object-file addresses so they refer to
2959 the proper run-time addresses.
2961 For the H8/300, and for the Renesas / SuperH SH,
2962 @command{@value{AS}} pads sections if needed to
2963 ensure they end on a word (sixteen bit) boundary.
2966 @cindex standard assembler sections
2967 An object file written by @command{@value{AS}} has at least three sections, any
2968 of which may be empty. These are named @dfn{text}, @dfn{data} and
2973 When it generates COFF or ELF output,
2975 @command{@value{AS}} can also generate whatever other named sections you specify
2976 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2977 If you do not use any directives that place output in the @samp{.text}
2978 or @samp{.data} sections, these sections still exist, but are empty.
2983 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2985 @command{@value{AS}} can also generate whatever other named sections you
2986 specify using the @samp{.space} and @samp{.subspace} directives. See
2987 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2988 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2989 assembler directives.
2992 Additionally, @command{@value{AS}} uses different names for the standard
2993 text, data, and bss sections when generating SOM output. Program text
2994 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2995 BSS into @samp{$BSS$}.
2999 Within the object file, the text section starts at address @code{0}, the
3000 data section follows, and the bss section follows the data section.
3003 When generating either SOM or ELF output files on the HPPA, the text
3004 section starts at address @code{0}, the data section at address
3005 @code{0x4000000}, and the bss section follows the data section.
3008 To let @code{@value{LD}} know which data changes when the sections are
3009 relocated, and how to change that data, @command{@value{AS}} also writes to the
3010 object file details of the relocation needed. To perform relocation
3011 @code{@value{LD}} must know, each time an address in the object
3015 Where in the object file is the beginning of this reference to
3018 How long (in bytes) is this reference?
3020 Which section does the address refer to? What is the numeric value of
3022 (@var{address}) @minus{} (@var{start-address of section})?
3025 Is the reference to an address ``Program-Counter relative''?
3028 @cindex addresses, format of
3029 @cindex section-relative addressing
3030 In fact, every address @command{@value{AS}} ever uses is expressed as
3032 (@var{section}) + (@var{offset into section})
3035 Further, most expressions @command{@value{AS}} computes have this section-relative
3038 (For some object formats, such as SOM for the HPPA, some expressions are
3039 symbol-relative instead.)
3042 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3043 @var{N} into section @var{secname}.''
3045 Apart from text, data and bss sections you need to know about the
3046 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3047 addresses in the absolute section remain unchanged. For example, address
3048 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3049 @code{@value{LD}}. Although the linker never arranges two partial programs'
3050 data sections with overlapping addresses after linking, @emph{by definition}
3051 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3052 part of a program is always the same address when the program is running as
3053 address @code{@{absolute@ 239@}} in any other part of the program.
3055 The idea of sections is extended to the @dfn{undefined} section. Any
3056 address whose section is unknown at assembly time is by definition
3057 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3058 Since numbers are always defined, the only way to generate an undefined
3059 address is to mention an undefined symbol. A reference to a named
3060 common block would be such a symbol: its value is unknown at assembly
3061 time so it has section @emph{undefined}.
3063 By analogy the word @emph{section} is used to describe groups of sections in
3064 the linked program. @code{@value{LD}} puts all partial programs' text
3065 sections in contiguous addresses in the linked program. It is
3066 customary to refer to the @emph{text section} of a program, meaning all
3067 the addresses of all partial programs' text sections. Likewise for
3068 data and bss sections.
3070 Some sections are manipulated by @code{@value{LD}}; others are invented for
3071 use of @command{@value{AS}} and have no meaning except during assembly.
3074 @section Linker Sections
3075 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3080 @cindex named sections
3081 @cindex sections, named
3082 @item named sections
3085 @cindex text section
3086 @cindex data section
3090 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3091 separate but equal sections. Anything you can say of one section is
3094 When the program is running, however, it is
3095 customary for the text section to be unalterable. The
3096 text section is often shared among processes: it contains
3097 instructions, constants and the like. The data section of a running
3098 program is usually alterable: for example, C variables would be stored
3099 in the data section.
3104 This section contains zeroed bytes when your program begins running. It
3105 is used to hold uninitialized variables or common storage. The length of
3106 each partial program's bss section is important, but because it starts
3107 out containing zeroed bytes there is no need to store explicit zero
3108 bytes in the object file. The bss section was invented to eliminate
3109 those explicit zeros from object files.
3111 @cindex absolute section
3112 @item absolute section
3113 Address 0 of this section is always ``relocated'' to runtime address 0.
3114 This is useful if you want to refer to an address that @code{@value{LD}} must
3115 not change when relocating. In this sense we speak of absolute
3116 addresses being ``unrelocatable'': they do not change during relocation.
3118 @cindex undefined section
3119 @item undefined section
3120 This ``section'' is a catch-all for address references to objects not in
3121 the preceding sections.
3122 @c FIXME: ref to some other doc on obj-file formats could go here.
3125 @cindex relocation example
3126 An idealized example of three relocatable sections follows.
3128 The example uses the traditional section names @samp{.text} and @samp{.data}.
3130 Memory addresses are on the horizontal axis.
3134 @c END TEXI2ROFF-KILL
3137 partial program # 1: |ttttt|dddd|00|
3144 partial program # 2: |TTT|DDD|000|
3147 +--+---+-----+--+----+---+-----+~~
3148 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3149 +--+---+-----+--+----+---+-----+~~
3151 addresses: 0 @dots{}
3158 \line{\it Partial program \#1: \hfil}
3159 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3160 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3162 \line{\it Partial program \#2: \hfil}
3163 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3164 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3166 \line{\it linked program: \hfil}
3167 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3168 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3169 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3170 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3172 \line{\it addresses: \hfil}
3176 @c END TEXI2ROFF-KILL
3179 @section Assembler Internal Sections
3181 @cindex internal assembler sections
3182 @cindex sections in messages, internal
3183 These sections are meant only for the internal use of @command{@value{AS}}. They
3184 have no meaning at run-time. You do not really need to know about these
3185 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3186 warning messages, so it might be helpful to have an idea of their
3187 meanings to @command{@value{AS}}. These sections are used to permit the
3188 value of every expression in your assembly language program to be a
3189 section-relative address.
3192 @cindex assembler internal logic error
3193 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3194 An internal assembler logic error has been found. This means there is a
3195 bug in the assembler.
3197 @cindex expr (internal section)
3199 The assembler stores complex expression internally as combinations of
3200 symbols. When it needs to represent an expression as a symbol, it puts
3201 it in the expr section.
3203 @c FIXME item transfer[t] vector preload
3204 @c FIXME item transfer[t] vector postload
3205 @c FIXME item register
3209 @section Sub-Sections
3211 @cindex numbered subsections
3212 @cindex grouping data
3218 fall into two sections: text and data.
3220 You may have separate groups of
3222 data in named sections
3226 data in named sections
3232 that you want to end up near to each other in the object file, even though they
3233 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3234 use @dfn{subsections} for this purpose. Within each section, there can be
3235 numbered subsections with values from 0 to 8192. Objects assembled into the
3236 same subsection go into the object file together with other objects in the same
3237 subsection. For example, a compiler might want to store constants in the text
3238 section, but might not want to have them interspersed with the program being
3239 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3240 section of code being output, and a @samp{.text 1} before each group of
3241 constants being output.
3243 Subsections are optional. If you do not use subsections, everything
3244 goes in subsection number zero.
3247 Each subsection is zero-padded up to a multiple of four bytes.
3248 (Subsections may be padded a different amount on different flavors
3249 of @command{@value{AS}}.)
3253 On the H8/300 platform, each subsection is zero-padded to a word
3254 boundary (two bytes).
3255 The same is true on the Renesas SH.
3258 @c FIXME section padding (alignment)?
3259 @c Rich Pixley says padding here depends on target obj code format; that
3260 @c doesn't seem particularly useful to say without further elaboration,
3261 @c so for now I say nothing about it. If this is a generic BFD issue,
3262 @c these paragraphs might need to vanish from this manual, and be
3263 @c discussed in BFD chapter of binutils (or some such).
3267 Subsections appear in your object file in numeric order, lowest numbered
3268 to highest. (All this to be compatible with other people's assemblers.)
3269 The object file contains no representation of subsections; @code{@value{LD}} and
3270 other programs that manipulate object files see no trace of them.
3271 They just see all your text subsections as a text section, and all your
3272 data subsections as a data section.
3274 To specify which subsection you want subsequent statements assembled
3275 into, use a numeric argument to specify it, in a @samp{.text
3276 @var{expression}} or a @samp{.data @var{expression}} statement.
3279 When generating COFF output, you
3284 can also use an extra subsection
3285 argument with arbitrary named sections: @samp{.section @var{name},
3290 When generating ELF output, you
3295 can also use the @code{.subsection} directive (@pxref{SubSection})
3296 to specify a subsection: @samp{.subsection @var{expression}}.
3298 @var{Expression} should be an absolute expression
3299 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3300 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3301 begins in @code{text 0}. For instance:
3303 .text 0 # The default subsection is text 0 anyway.
3304 .ascii "This lives in the first text subsection. *"
3306 .ascii "But this lives in the second text subsection."
3308 .ascii "This lives in the data section,"
3309 .ascii "in the first data subsection."
3311 .ascii "This lives in the first text section,"
3312 .ascii "immediately following the asterisk (*)."
3315 Each section has a @dfn{location counter} incremented by one for every byte
3316 assembled into that section. Because subsections are merely a convenience
3317 restricted to @command{@value{AS}} there is no concept of a subsection location
3318 counter. There is no way to directly manipulate a location counter---but the
3319 @code{.align} directive changes it, and any label definition captures its
3320 current value. The location counter of the section where statements are being
3321 assembled is said to be the @dfn{active} location counter.
3324 @section bss Section
3327 @cindex common variable storage
3328 The bss section is used for local common variable storage.
3329 You may allocate address space in the bss section, but you may
3330 not dictate data to load into it before your program executes. When
3331 your program starts running, all the contents of the bss
3332 section are zeroed bytes.
3334 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3335 @ref{Lcomm,,@code{.lcomm}}.
3337 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3338 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3341 When assembling for a target which supports multiple sections, such as ELF or
3342 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3343 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3344 section. Typically the section will only contain symbol definitions and
3345 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3352 Symbols are a central concept: the programmer uses symbols to name
3353 things, the linker uses symbols to link, and the debugger uses symbols
3357 @cindex debuggers, and symbol order
3358 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3359 the same order they were declared. This may break some debuggers.
3364 * Setting Symbols:: Giving Symbols Other Values
3365 * Symbol Names:: Symbol Names
3366 * Dot:: The Special Dot Symbol
3367 * Symbol Attributes:: Symbol Attributes
3374 A @dfn{label} is written as a symbol immediately followed by a colon
3375 @samp{:}. The symbol then represents the current value of the
3376 active location counter, and is, for example, a suitable instruction
3377 operand. You are warned if you use the same symbol to represent two
3378 different locations: the first definition overrides any other
3382 On the HPPA, the usual form for a label need not be immediately followed by a
3383 colon, but instead must start in column zero. Only one label may be defined on
3384 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3385 provides a special directive @code{.label} for defining labels more flexibly.
3388 @node Setting Symbols
3389 @section Giving Symbols Other Values
3391 @cindex assigning values to symbols
3392 @cindex symbol values, assigning
3393 A symbol can be given an arbitrary value by writing a symbol, followed
3394 by an equals sign @samp{=}, followed by an expression
3395 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3396 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3397 equals sign @samp{=}@samp{=} here represents an equivalent of the
3398 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3401 Blackfin does not support symbol assignment with @samp{=}.
3405 @section Symbol Names
3407 @cindex symbol names
3408 @cindex names, symbol
3409 @ifclear SPECIAL-SYMS
3410 Symbol names begin with a letter or with one of @samp{._}. On most
3411 machines, you can also use @code{$} in symbol names; exceptions are
3412 noted in @ref{Machine Dependencies}. That character may be followed by any
3413 string of digits, letters, dollar signs (unless otherwise noted for a
3414 particular target machine), and underscores.
3418 Symbol names begin with a letter or with one of @samp{._}. On the
3419 Renesas SH you can also use @code{$} in symbol names. That
3420 character may be followed by any string of digits, letters, dollar signs (save
3421 on the H8/300), and underscores.
3425 Case of letters is significant: @code{foo} is a different symbol name
3428 Each symbol has exactly one name. Each name in an assembly language program
3429 refers to exactly one symbol. You may use that symbol name any number of times
3432 @subheading Local Symbol Names
3434 @cindex local symbol names
3435 @cindex symbol names, local
3436 A local symbol is any symbol beginning with certain local label prefixes.
3437 By default, the local label prefix is @samp{.L} for ELF systems or
3438 @samp{L} for traditional a.out systems, but each target may have its own
3439 set of local label prefixes.
3441 On the HPPA local symbols begin with @samp{L$}.
3444 Local symbols are defined and used within the assembler, but they are
3445 normally not saved in object files. Thus, they are not visible when debugging.
3446 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3447 @option{-L}}) to retain the local symbols in the object files.
3449 @subheading Local Labels
3451 @cindex local labels
3452 @cindex temporary symbol names
3453 @cindex symbol names, temporary
3454 Local labels help compilers and programmers use names temporarily.
3455 They create symbols which are guaranteed to be unique over the entire scope of
3456 the input source code and which can be referred to by a simple notation.
3457 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3458 represents any positive integer). To refer to the most recent previous
3459 definition of that label write @samp{@b{N}b}, using the same number as when
3460 you defined the label. To refer to the next definition of a local label, write
3461 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3464 There is no restriction on how you can use these labels, and you can reuse them
3465 too. So that it is possible to repeatedly define the same local label (using
3466 the same number @samp{@b{N}}), although you can only refer to the most recently
3467 defined local label of that number (for a backwards reference) or the next
3468 definition of a specific local label for a forward reference. It is also worth
3469 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3470 implemented in a slightly more efficient manner than the others.
3481 Which is the equivalent of:
3484 label_1: branch label_3
3485 label_2: branch label_1
3486 label_3: branch label_4
3487 label_4: branch label_3
3490 Local label names are only a notational device. They are immediately
3491 transformed into more conventional symbol names before the assembler uses them.
3492 The symbol names are stored in the symbol table, appear in error messages, and
3493 are optionally emitted to the object file. The names are constructed using
3497 @item @emph{local label prefix}
3498 All local symbols begin with the system-specific local label prefix.
3499 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3500 that start with the local label prefix. These labels are
3501 used for symbols you are never intended to see. If you use the
3502 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3503 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3504 you may use them in debugging.
3507 This is the number that was used in the local label definition. So if the
3508 label is written @samp{55:} then the number is @samp{55}.
3511 This unusual character is included so you do not accidentally invent a symbol
3512 of the same name. The character has ASCII value of @samp{\002} (control-B).
3514 @item @emph{ordinal number}
3515 This is a serial number to keep the labels distinct. The first definition of
3516 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3517 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3518 the number @samp{1} and its 15th definition gets @samp{15} as well.
3521 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3522 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3524 @subheading Dollar Local Labels
3525 @cindex dollar local symbols
3527 @code{@value{AS}} also supports an even more local form of local labels called
3528 dollar labels. These labels go out of scope (i.e., they become undefined) as
3529 soon as a non-local label is defined. Thus they remain valid for only a small
3530 region of the input source code. Normal local labels, by contrast, remain in
3531 scope for the entire file, or until they are redefined by another occurrence of
3532 the same local label.
3534 Dollar labels are defined in exactly the same way as ordinary local labels,
3535 except that they have a dollar sign suffix to their numeric value, e.g.,
3538 They can also be distinguished from ordinary local labels by their transformed
3539 names which use ASCII character @samp{\001} (control-A) as the magic character
3540 to distinguish them from ordinary labels. For example, the fifth definition of
3541 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3544 @section The Special Dot Symbol
3546 @cindex dot (symbol)
3547 @cindex @code{.} (symbol)
3548 @cindex current address
3549 @cindex location counter
3550 The special symbol @samp{.} refers to the current address that
3551 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3552 .long .} defines @code{melvin} to contain its own address.
3553 Assigning a value to @code{.} is treated the same as a @code{.org}
3555 @ifclear no-space-dir
3556 Thus, the expression @samp{.=.+4} is the same as saying
3560 @node Symbol Attributes
3561 @section Symbol Attributes
3563 @cindex symbol attributes
3564 @cindex attributes, symbol
3565 Every symbol has, as well as its name, the attributes ``Value'' and
3566 ``Type''. Depending on output format, symbols can also have auxiliary
3569 The detailed definitions are in @file{a.out.h}.
3572 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3573 all these attributes, and probably won't warn you. This makes the
3574 symbol an externally defined symbol, which is generally what you
3578 * Symbol Value:: Value
3579 * Symbol Type:: Type
3582 * a.out Symbols:: Symbol Attributes: @code{a.out}
3586 * a.out Symbols:: Symbol Attributes: @code{a.out}
3589 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3594 * COFF Symbols:: Symbol Attributes for COFF
3597 * SOM Symbols:: Symbol Attributes for SOM
3604 @cindex value of a symbol
3605 @cindex symbol value
3606 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3607 location in the text, data, bss or absolute sections the value is the
3608 number of addresses from the start of that section to the label.
3609 Naturally for text, data and bss sections the value of a symbol changes
3610 as @code{@value{LD}} changes section base addresses during linking. Absolute
3611 symbols' values do not change during linking: that is why they are
3614 The value of an undefined symbol is treated in a special way. If it is
3615 0 then the symbol is not defined in this assembler source file, and
3616 @code{@value{LD}} tries to determine its value from other files linked into the
3617 same program. You make this kind of symbol simply by mentioning a symbol
3618 name without defining it. A non-zero value represents a @code{.comm}
3619 common declaration. The value is how much common storage to reserve, in
3620 bytes (addresses). The symbol refers to the first address of the
3626 @cindex type of a symbol
3628 The type attribute of a symbol contains relocation (section)
3629 information, any flag settings indicating that a symbol is external, and
3630 (optionally), other information for linkers and debuggers. The exact
3631 format depends on the object-code output format in use.
3636 @c The following avoids a "widow" subsection title. @group would be
3637 @c better if it were available outside examples.
3640 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3642 @cindex @code{b.out} symbol attributes
3643 @cindex symbol attributes, @code{b.out}
3644 These symbol attributes appear only when @command{@value{AS}} is configured for
3645 one of the Berkeley-descended object output formats---@code{a.out} or
3651 @subsection Symbol Attributes: @code{a.out}
3653 @cindex @code{a.out} symbol attributes
3654 @cindex symbol attributes, @code{a.out}
3660 @subsection Symbol Attributes: @code{a.out}
3662 @cindex @code{a.out} symbol attributes
3663 @cindex symbol attributes, @code{a.out}
3667 * Symbol Desc:: Descriptor
3668 * Symbol Other:: Other
3672 @subsubsection Descriptor
3674 @cindex descriptor, of @code{a.out} symbol
3675 This is an arbitrary 16-bit value. You may establish a symbol's
3676 descriptor value by using a @code{.desc} statement
3677 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3678 @command{@value{AS}}.
3681 @subsubsection Other
3683 @cindex other attribute, of @code{a.out} symbol
3684 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3689 @subsection Symbol Attributes for COFF
3691 @cindex COFF symbol attributes
3692 @cindex symbol attributes, COFF
3694 The COFF format supports a multitude of auxiliary symbol attributes;
3695 like the primary symbol attributes, they are set between @code{.def} and
3696 @code{.endef} directives.
3698 @subsubsection Primary Attributes
3700 @cindex primary attributes, COFF symbols
3701 The symbol name is set with @code{.def}; the value and type,
3702 respectively, with @code{.val} and @code{.type}.
3704 @subsubsection Auxiliary Attributes
3706 @cindex auxiliary attributes, COFF symbols
3707 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3708 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3709 table information for COFF.
3714 @subsection Symbol Attributes for SOM
3716 @cindex SOM symbol attributes
3717 @cindex symbol attributes, SOM
3719 The SOM format for the HPPA supports a multitude of symbol attributes set with
3720 the @code{.EXPORT} and @code{.IMPORT} directives.
3722 The attributes are described in @cite{HP9000 Series 800 Assembly
3723 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3724 @code{EXPORT} assembler directive documentation.
3728 @chapter Expressions
3732 @cindex numeric values
3733 An @dfn{expression} specifies an address or numeric value.
3734 Whitespace may precede and/or follow an expression.
3736 The result of an expression must be an absolute number, or else an offset into
3737 a particular section. If an expression is not absolute, and there is not
3738 enough information when @command{@value{AS}} sees the expression to know its
3739 section, a second pass over the source program might be necessary to interpret
3740 the expression---but the second pass is currently not implemented.
3741 @command{@value{AS}} aborts with an error message in this situation.
3744 * Empty Exprs:: Empty Expressions
3745 * Integer Exprs:: Integer Expressions
3749 @section Empty Expressions
3751 @cindex empty expressions
3752 @cindex expressions, empty
3753 An empty expression has no value: it is just whitespace or null.
3754 Wherever an absolute expression is required, you may omit the
3755 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3756 is compatible with other assemblers.
3759 @section Integer Expressions
3761 @cindex integer expressions
3762 @cindex expressions, integer
3763 An @dfn{integer expression} is one or more @emph{arguments} delimited
3764 by @emph{operators}.
3767 * Arguments:: Arguments
3768 * Operators:: Operators
3769 * Prefix Ops:: Prefix Operators
3770 * Infix Ops:: Infix Operators
3774 @subsection Arguments
3776 @cindex expression arguments
3777 @cindex arguments in expressions
3778 @cindex operands in expressions
3779 @cindex arithmetic operands
3780 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3781 contexts arguments are sometimes called ``arithmetic operands''. In
3782 this manual, to avoid confusing them with the ``instruction operands'' of
3783 the machine language, we use the term ``argument'' to refer to parts of
3784 expressions only, reserving the word ``operand'' to refer only to machine
3785 instruction operands.
3787 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3788 @var{section} is one of text, data, bss, absolute,
3789 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3792 Numbers are usually integers.
3794 A number can be a flonum or bignum. In this case, you are warned
3795 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3796 these 32 bits are an integer. You may write integer-manipulating
3797 instructions that act on exotic constants, compatible with other
3800 @cindex subexpressions
3801 Subexpressions are a left parenthesis @samp{(} followed by an integer
3802 expression, followed by a right parenthesis @samp{)}; or a prefix
3803 operator followed by an argument.
3806 @subsection Operators
3808 @cindex operators, in expressions
3809 @cindex arithmetic functions
3810 @cindex functions, in expressions
3811 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3812 operators are followed by an argument. Infix operators appear
3813 between their arguments. Operators may be preceded and/or followed by
3817 @subsection Prefix Operator
3819 @cindex prefix operators
3820 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3821 one argument, which must be absolute.
3823 @c the tex/end tex stuff surrounding this small table is meant to make
3824 @c it align, on the printed page, with the similar table in the next
3825 @c section (which is inside an enumerate).
3827 \global\advance\leftskip by \itemindent
3832 @dfn{Negation}. Two's complement negation.
3834 @dfn{Complementation}. Bitwise not.
3838 \global\advance\leftskip by -\itemindent
3842 @subsection Infix Operators
3844 @cindex infix operators
3845 @cindex operators, permitted arguments
3846 @dfn{Infix operators} take two arguments, one on either side. Operators
3847 have precedence, but operations with equal precedence are performed left
3848 to right. Apart from @code{+} or @option{-}, both arguments must be
3849 absolute, and the result is absolute.
3852 @cindex operator precedence
3853 @cindex precedence of operators
3860 @dfn{Multiplication}.
3863 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3869 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3872 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3876 Intermediate precedence
3881 @dfn{Bitwise Inclusive Or}.
3887 @dfn{Bitwise Exclusive Or}.
3890 @dfn{Bitwise Or Not}.
3897 @cindex addition, permitted arguments
3898 @cindex plus, permitted arguments
3899 @cindex arguments for addition
3901 @dfn{Addition}. If either argument is absolute, the result has the section of
3902 the other argument. You may not add together arguments from different
3905 @cindex subtraction, permitted arguments
3906 @cindex minus, permitted arguments
3907 @cindex arguments for subtraction
3909 @dfn{Subtraction}. If the right argument is absolute, the
3910 result has the section of the left argument.
3911 If both arguments are in the same section, the result is absolute.
3912 You may not subtract arguments from different sections.
3913 @c FIXME is there still something useful to say about undefined - undefined ?
3915 @cindex comparison expressions
3916 @cindex expressions, comparison
3921 @dfn{Is Not Equal To}
3925 @dfn{Is Greater Than}
3927 @dfn{Is Greater Than Or Equal To}
3929 @dfn{Is Less Than Or Equal To}
3931 The comparison operators can be used as infix operators. A true results has a
3932 value of -1 whereas a false result has a value of 0. Note, these operators
3933 perform signed comparisons.
3936 @item Lowest Precedence
3945 These two logical operations can be used to combine the results of sub
3946 expressions. Note, unlike the comparison operators a true result returns a
3947 value of 1 but a false results does still return 0. Also note that the logical
3948 or operator has a slightly lower precedence than logical and.
3953 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3954 address; you can only have a defined section in one of the two arguments.
3957 @chapter Assembler Directives
3959 @cindex directives, machine independent
3960 @cindex pseudo-ops, machine independent
3961 @cindex machine independent directives
3962 All assembler directives have names that begin with a period (@samp{.}).
3963 The rest of the name is letters, usually in lower case.
3965 This chapter discusses directives that are available regardless of the
3966 target machine configuration for the @sc{gnu} assembler.
3968 Some machine configurations provide additional directives.
3969 @xref{Machine Dependencies}.
3972 @ifset machine-directives
3973 @xref{Machine Dependencies}, for additional directives.
3978 * Abort:: @code{.abort}
3980 * ABORT (COFF):: @code{.ABORT}
3983 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3984 * Altmacro:: @code{.altmacro}
3985 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3986 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3987 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3988 * Byte:: @code{.byte @var{expressions}}
3989 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
3990 * Comm:: @code{.comm @var{symbol} , @var{length} }
3991 * Data:: @code{.data @var{subsection}}
3993 * Def:: @code{.def @var{name}}
3996 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4002 * Double:: @code{.double @var{flonums}}
4003 * Eject:: @code{.eject}
4004 * Else:: @code{.else}
4005 * Elseif:: @code{.elseif}
4008 * Endef:: @code{.endef}
4011 * Endfunc:: @code{.endfunc}
4012 * Endif:: @code{.endif}
4013 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4014 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4015 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4017 * Error:: @code{.error @var{string}}
4018 * Exitm:: @code{.exitm}
4019 * Extern:: @code{.extern}
4020 * Fail:: @code{.fail}
4021 * File:: @code{.file}
4022 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4023 * Float:: @code{.float @var{flonums}}
4024 * Func:: @code{.func}
4025 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4027 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4028 * Hidden:: @code{.hidden @var{names}}
4031 * hword:: @code{.hword @var{expressions}}
4032 * Ident:: @code{.ident}
4033 * If:: @code{.if @var{absolute expression}}
4034 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4035 * Include:: @code{.include "@var{file}"}
4036 * Int:: @code{.int @var{expressions}}
4038 * Internal:: @code{.internal @var{names}}
4041 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4042 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4043 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4044 * Lflags:: @code{.lflags}
4045 @ifclear no-line-dir
4046 * Line:: @code{.line @var{line-number}}
4049 * Linkonce:: @code{.linkonce [@var{type}]}
4050 * List:: @code{.list}
4051 * Ln:: @code{.ln @var{line-number}}
4052 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4053 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4055 * Local:: @code{.local @var{names}}
4058 * Long:: @code{.long @var{expressions}}
4060 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4063 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4064 * MRI:: @code{.mri @var{val}}
4065 * Noaltmacro:: @code{.noaltmacro}
4066 * Nolist:: @code{.nolist}
4067 * Octa:: @code{.octa @var{bignums}}
4068 * Org:: @code{.org @var{new-lc}, @var{fill}}
4069 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4071 * PopSection:: @code{.popsection}
4072 * Previous:: @code{.previous}
4075 * Print:: @code{.print @var{string}}
4077 * Protected:: @code{.protected @var{names}}
4080 * Psize:: @code{.psize @var{lines}, @var{columns}}
4081 * Purgem:: @code{.purgem @var{name}}
4083 * PushSection:: @code{.pushsection @var{name}}
4086 * Quad:: @code{.quad @var{bignums}}
4087 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4088 * Rept:: @code{.rept @var{count}}
4089 * Sbttl:: @code{.sbttl "@var{subheading}"}
4091 * Scl:: @code{.scl @var{class}}
4094 * Section:: @code{.section @var{name}[, @var{flags}]}
4097 * Set:: @code{.set @var{symbol}, @var{expression}}
4098 * Short:: @code{.short @var{expressions}}
4099 * Single:: @code{.single @var{flonums}}
4101 * Size:: @code{.size [@var{name} , @var{expression}]}
4103 @ifclear no-space-dir
4104 * Skip:: @code{.skip @var{size} , @var{fill}}
4107 * Sleb128:: @code{.sleb128 @var{expressions}}
4108 @ifclear no-space-dir
4109 * Space:: @code{.space @var{size} , @var{fill}}
4112 * Stab:: @code{.stabd, .stabn, .stabs}
4115 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4116 * Struct:: @code{.struct @var{expression}}
4118 * SubSection:: @code{.subsection}
4119 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4123 * Tag:: @code{.tag @var{structname}}
4126 * Text:: @code{.text @var{subsection}}
4127 * Title:: @code{.title "@var{heading}"}
4129 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4132 * Uleb128:: @code{.uleb128 @var{expressions}}
4134 * Val:: @code{.val @var{addr}}
4138 * Version:: @code{.version "@var{string}"}
4139 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4140 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4143 * Warning:: @code{.warning @var{string}}
4144 * Weak:: @code{.weak @var{names}}
4145 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4146 * Word:: @code{.word @var{expressions}}
4147 * Deprecated:: Deprecated Directives
4151 @section @code{.abort}
4153 @cindex @code{abort} directive
4154 @cindex stopping the assembly
4155 This directive stops the assembly immediately. It is for
4156 compatibility with other assemblers. The original idea was that the
4157 assembly language source would be piped into the assembler. If the sender
4158 of the source quit, it could use this directive tells @command{@value{AS}} to
4159 quit also. One day @code{.abort} will not be supported.
4163 @section @code{.ABORT} (COFF)
4165 @cindex @code{ABORT} directive
4166 When producing COFF output, @command{@value{AS}} accepts this directive as a
4167 synonym for @samp{.abort}.
4170 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4176 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4178 @cindex padding the location counter
4179 @cindex @code{align} directive
4180 Pad the location counter (in the current subsection) to a particular storage
4181 boundary. The first expression (which must be absolute) is the alignment
4182 required, as described below.
4184 The second expression (also absolute) gives the fill value to be stored in the
4185 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4186 padding bytes are normally zero. However, on some systems, if the section is
4187 marked as containing code and the fill value is omitted, the space is filled
4188 with no-op instructions.
4190 The third expression is also absolute, and is also optional. If it is present,
4191 it is the maximum number of bytes that should be skipped by this alignment
4192 directive. If doing the alignment would require skipping more bytes than the
4193 specified maximum, then the alignment is not done at all. You can omit the
4194 fill value (the second argument) entirely by simply using two commas after the
4195 required alignment; this can be useful if you want the alignment to be filled
4196 with no-op instructions when appropriate.
4198 The way the required alignment is specified varies from system to system.
4199 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4200 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4201 alignment request in bytes. For example @samp{.align 8} advances
4202 the location counter until it is a multiple of 8. If the location counter
4203 is already a multiple of 8, no change is needed. For the tic54x, the
4204 first expression is the alignment request in words.
4206 For other systems, including ppc, i386 using a.out format, arm and
4207 strongarm, it is the
4208 number of low-order zero bits the location counter must have after
4209 advancement. For example @samp{.align 3} advances the location
4210 counter until it a multiple of 8. If the location counter is already a
4211 multiple of 8, no change is needed.
4213 This inconsistency is due to the different behaviors of the various
4214 native assemblers for these systems which GAS must emulate.
4215 GAS also provides @code{.balign} and @code{.p2align} directives,
4216 described later, which have a consistent behavior across all
4217 architectures (but are specific to GAS).
4220 @section @code{.altmacro}
4221 Enable alternate macro mode, enabling:
4224 @item LOCAL @var{name} [ , @dots{} ]
4225 One additional directive, @code{LOCAL}, is available. It is used to
4226 generate a string replacement for each of the @var{name} arguments, and
4227 replace any instances of @var{name} in each macro expansion. The
4228 replacement string is unique in the assembly, and different for each
4229 separate macro expansion. @code{LOCAL} allows you to write macros that
4230 define symbols, without fear of conflict between separate macro expansions.
4232 @item String delimiters
4233 You can write strings delimited in these other ways besides
4234 @code{"@var{string}"}:
4237 @item '@var{string}'
4238 You can delimit strings with single-quote characters.
4240 @item <@var{string}>
4241 You can delimit strings with matching angle brackets.
4244 @item single-character string escape
4245 To include any single character literally in a string (even if the
4246 character would otherwise have some special meaning), you can prefix the
4247 character with @samp{!} (an exclamation mark). For example, you can
4248 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4250 @item Expression results as strings
4251 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4252 and use the result as a string.
4256 @section @code{.ascii "@var{string}"}@dots{}
4258 @cindex @code{ascii} directive
4259 @cindex string literals
4260 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4261 separated by commas. It assembles each string (with no automatic
4262 trailing zero byte) into consecutive addresses.
4265 @section @code{.asciz "@var{string}"}@dots{}
4267 @cindex @code{asciz} directive
4268 @cindex zero-terminated strings
4269 @cindex null-terminated strings
4270 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4271 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4274 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4276 @cindex padding the location counter given number of bytes
4277 @cindex @code{balign} directive
4278 Pad the location counter (in the current subsection) to a particular
4279 storage boundary. The first expression (which must be absolute) is the
4280 alignment request in bytes. For example @samp{.balign 8} advances
4281 the location counter until it is a multiple of 8. If the location counter
4282 is already a multiple of 8, no change is needed.
4284 The second expression (also absolute) gives the fill value to be stored in the
4285 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4286 padding bytes are normally zero. However, on some systems, if the section is
4287 marked as containing code and the fill value is omitted, the space is filled
4288 with no-op instructions.
4290 The third expression is also absolute, and is also optional. If it is present,
4291 it is the maximum number of bytes that should be skipped by this alignment
4292 directive. If doing the alignment would require skipping more bytes than the
4293 specified maximum, then the alignment is not done at all. You can omit the
4294 fill value (the second argument) entirely by simply using two commas after the
4295 required alignment; this can be useful if you want the alignment to be filled
4296 with no-op instructions when appropriate.
4298 @cindex @code{balignw} directive
4299 @cindex @code{balignl} directive
4300 The @code{.balignw} and @code{.balignl} directives are variants of the
4301 @code{.balign} directive. The @code{.balignw} directive treats the fill
4302 pattern as a two byte word value. The @code{.balignl} directives treats the
4303 fill pattern as a four byte longword value. For example, @code{.balignw
4304 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4305 filled in with the value 0x368d (the exact placement of the bytes depends upon
4306 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4310 @section @code{.byte @var{expressions}}
4312 @cindex @code{byte} directive
4313 @cindex integers, one byte
4314 @code{.byte} expects zero or more expressions, separated by commas.
4315 Each expression is assembled into the next byte.
4317 @node CFI directives
4318 @section @code{.cfi_sections @var{section_list}}
4319 @cindex @code{cfi_sections} directive
4320 @code{.cfi_sections} may be used to specify whether CFI directives
4321 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4322 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4323 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4324 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4325 directive is not used is @code{.cfi_sections .eh_frame}.
4327 @section @code{.cfi_startproc [simple]}
4328 @cindex @code{cfi_startproc} directive
4329 @code{.cfi_startproc} is used at the beginning of each function that
4330 should have an entry in @code{.eh_frame}. It initializes some internal
4331 data structures. Don't forget to close the function by
4332 @code{.cfi_endproc}.
4334 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4335 it also emits some architecture dependent initial CFI instructions.
4337 @section @code{.cfi_endproc}
4338 @cindex @code{cfi_endproc} directive
4339 @code{.cfi_endproc} is used at the end of a function where it closes its
4340 unwind entry previously opened by
4341 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4343 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4344 @code{.cfi_personality} defines personality routine and its encoding.
4345 @var{encoding} must be a constant determining how the personality
4346 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4347 argument is not present, otherwise second argument should be
4348 a constant or a symbol name. When using indirect encodings,
4349 the symbol provided should be the location where personality
4350 can be loaded from, not the personality routine itself.
4351 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4352 no personality routine.
4354 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4355 @code{.cfi_lsda} defines LSDA and its encoding.
4356 @var{encoding} must be a constant determining how the LSDA
4357 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4358 argument is not present, otherwise second argument should be a constant
4359 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4362 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4363 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4364 address from @var{register} and add @var{offset} to it}.
4366 @section @code{.cfi_def_cfa_register @var{register}}
4367 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4368 now on @var{register} will be used instead of the old one. Offset
4371 @section @code{.cfi_def_cfa_offset @var{offset}}
4372 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4373 remains the same, but @var{offset} is new. Note that it is the
4374 absolute offset that will be added to a defined register to compute
4377 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4378 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4379 value that is added/substracted from the previous offset.
4381 @section @code{.cfi_offset @var{register}, @var{offset}}
4382 Previous value of @var{register} is saved at offset @var{offset} from
4385 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4386 Previous value of @var{register} is saved at offset @var{offset} from
4387 the current CFA register. This is transformed to @code{.cfi_offset}
4388 using the known displacement of the CFA register from the CFA.
4389 This is often easier to use, because the number will match the
4390 code it's annotating.
4392 @section @code{.cfi_register @var{register1}, @var{register2}}
4393 Previous value of @var{register1} is saved in register @var{register2}.
4395 @section @code{.cfi_restore @var{register}}
4396 @code{.cfi_restore} says that the rule for @var{register} is now the
4397 same as it was at the beginning of the function, after all initial
4398 instruction added by @code{.cfi_startproc} were executed.
4400 @section @code{.cfi_undefined @var{register}}
4401 From now on the previous value of @var{register} can't be restored anymore.
4403 @section @code{.cfi_same_value @var{register}}
4404 Current value of @var{register} is the same like in the previous frame,
4405 i.e. no restoration needed.
4407 @section @code{.cfi_remember_state},
4408 First save all current rules for all registers by @code{.cfi_remember_state},
4409 then totally screw them up by subsequent @code{.cfi_*} directives and when
4410 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4411 the previous saved state.
4413 @section @code{.cfi_return_column @var{register}}
4414 Change return column @var{register}, i.e. the return address is either
4415 directly in @var{register} or can be accessed by rules for @var{register}.
4417 @section @code{.cfi_signal_frame}
4418 Mark current function as signal trampoline.
4420 @section @code{.cfi_window_save}
4421 SPARC register window has been saved.
4423 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4424 Allows the user to add arbitrary bytes to the unwind info. One
4425 might use this to add OS-specific CFI opcodes, or generic CFI
4426 opcodes that GAS does not yet support.
4428 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4429 The current value of @var{register} is @var{label}. The value of @var{label}
4430 will be encoded in the output file according to @var{encoding}; see the
4431 description of @code{.cfi_personality} for details on this encoding.
4433 The usefulness of equating a register to a fixed label is probably
4434 limited to the return address register. Here, it can be useful to
4435 mark a code segment that has only one return address which is reached
4436 by a direct branch and no copy of the return address exists in memory
4437 or another register.
4440 @section @code{.comm @var{symbol} , @var{length} }
4442 @cindex @code{comm} directive
4443 @cindex symbol, common
4444 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4445 common symbol in one object file may be merged with a defined or common symbol
4446 of the same name in another object file. If @code{@value{LD}} does not see a
4447 definition for the symbol--just one or more common symbols--then it will
4448 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4449 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4450 the same name, and they do not all have the same size, it will allocate space
4451 using the largest size.
4454 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4455 an optional third argument. This is the desired alignment of the symbol,
4456 specified for ELF as a byte boundary (for example, an alignment of 16 means
4457 that the least significant 4 bits of the address should be zero), and for PE
4458 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4459 boundary). The alignment must be an absolute expression, and it must be a
4460 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4461 common symbol, it will use the alignment when placing the symbol. If no
4462 alignment is specified, @command{@value{AS}} will set the alignment to the
4463 largest power of two less than or equal to the size of the symbol, up to a
4464 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4465 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4466 @samp{--section-alignment} option; image file sections in PE are aligned to
4467 multiples of 4096, which is far too large an alignment for ordinary variables.
4468 It is rather the default alignment for (non-debug) sections within object
4469 (@samp{*.o}) files, which are less strictly aligned.}.
4473 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4474 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4478 @section @code{.data @var{subsection}}
4480 @cindex @code{data} directive
4481 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4482 end of the data subsection numbered @var{subsection} (which is an
4483 absolute expression). If @var{subsection} is omitted, it defaults
4488 @section @code{.def @var{name}}
4490 @cindex @code{def} directive
4491 @cindex COFF symbols, debugging
4492 @cindex debugging COFF symbols
4493 Begin defining debugging information for a symbol @var{name}; the
4494 definition extends until the @code{.endef} directive is encountered.
4497 This directive is only observed when @command{@value{AS}} is configured for COFF
4498 format output; when producing @code{b.out}, @samp{.def} is recognized,
4505 @section @code{.desc @var{symbol}, @var{abs-expression}}
4507 @cindex @code{desc} directive
4508 @cindex COFF symbol descriptor
4509 @cindex symbol descriptor, COFF
4510 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4511 to the low 16 bits of an absolute expression.
4514 The @samp{.desc} directive is not available when @command{@value{AS}} is
4515 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4516 object format. For the sake of compatibility, @command{@value{AS}} accepts
4517 it, but produces no output, when configured for COFF.
4523 @section @code{.dim}
4525 @cindex @code{dim} directive
4526 @cindex COFF auxiliary symbol information
4527 @cindex auxiliary symbol information, COFF
4528 This directive is generated by compilers to include auxiliary debugging
4529 information in the symbol table. It is only permitted inside
4530 @code{.def}/@code{.endef} pairs.
4533 @samp{.dim} is only meaningful when generating COFF format output; when
4534 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4540 @section @code{.double @var{flonums}}
4542 @cindex @code{double} directive
4543 @cindex floating point numbers (double)
4544 @code{.double} expects zero or more flonums, separated by commas. It
4545 assembles floating point numbers.
4547 The exact kind of floating point numbers emitted depends on how
4548 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4552 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4553 in @sc{ieee} format.
4558 @section @code{.eject}
4560 @cindex @code{eject} directive
4561 @cindex new page, in listings
4562 @cindex page, in listings
4563 @cindex listing control: new page
4564 Force a page break at this point, when generating assembly listings.
4567 @section @code{.else}
4569 @cindex @code{else} directive
4570 @code{.else} is part of the @command{@value{AS}} support for conditional
4571 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4572 of code to be assembled if the condition for the preceding @code{.if}
4576 @section @code{.elseif}
4578 @cindex @code{elseif} directive
4579 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4580 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4581 @code{.if} block that would otherwise fill the entire @code{.else} section.
4584 @section @code{.end}
4586 @cindex @code{end} directive
4587 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4588 process anything in the file past the @code{.end} directive.
4592 @section @code{.endef}
4594 @cindex @code{endef} directive
4595 This directive flags the end of a symbol definition begun with
4599 @samp{.endef} is only meaningful when generating COFF format output; if
4600 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4601 directive but ignores it.
4606 @section @code{.endfunc}
4607 @cindex @code{endfunc} directive
4608 @code{.endfunc} marks the end of a function specified with @code{.func}.
4611 @section @code{.endif}
4613 @cindex @code{endif} directive
4614 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4615 it marks the end of a block of code that is only assembled
4616 conditionally. @xref{If,,@code{.if}}.
4619 @section @code{.equ @var{symbol}, @var{expression}}
4621 @cindex @code{equ} directive
4622 @cindex assigning values to symbols
4623 @cindex symbols, assigning values to
4624 This directive sets the value of @var{symbol} to @var{expression}.
4625 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4628 The syntax for @code{equ} on the HPPA is
4629 @samp{@var{symbol} .equ @var{expression}}.
4633 The syntax for @code{equ} on the Z80 is
4634 @samp{@var{symbol} equ @var{expression}}.
4635 On the Z80 it is an eror if @var{symbol} is already defined,
4636 but the symbol is not protected from later redefinition.
4637 Compare @ref{Equiv}.
4641 @section @code{.equiv @var{symbol}, @var{expression}}
4642 @cindex @code{equiv} directive
4643 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4644 the assembler will signal an error if @var{symbol} is already defined. Note a
4645 symbol which has been referenced but not actually defined is considered to be
4648 Except for the contents of the error message, this is roughly equivalent to
4655 plus it protects the symbol from later redefinition.
4658 @section @code{.eqv @var{symbol}, @var{expression}}
4659 @cindex @code{eqv} directive
4660 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4661 evaluate the expression or any part of it immediately. Instead each time
4662 the resulting symbol is used in an expression, a snapshot of its current
4666 @section @code{.err}
4667 @cindex @code{err} directive
4668 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4669 message and, unless the @option{-Z} option was used, it will not generate an
4670 object file. This can be used to signal an error in conditionally compiled code.
4673 @section @code{.error "@var{string}"}
4674 @cindex error directive
4676 Similarly to @code{.err}, this directive emits an error, but you can specify a
4677 string that will be emitted as the error message. If you don't specify the
4678 message, it defaults to @code{".error directive invoked in source file"}.
4679 @xref{Errors, ,Error and Warning Messages}.
4682 .error "This code has not been assembled and tested."
4686 @section @code{.exitm}
4687 Exit early from the current macro definition. @xref{Macro}.
4690 @section @code{.extern}
4692 @cindex @code{extern} directive
4693 @code{.extern} is accepted in the source program---for compatibility
4694 with other assemblers---but it is ignored. @command{@value{AS}} treats
4695 all undefined symbols as external.
4698 @section @code{.fail @var{expression}}
4700 @cindex @code{fail} directive
4701 Generates an error or a warning. If the value of the @var{expression} is 500
4702 or more, @command{@value{AS}} will print a warning message. If the value is less
4703 than 500, @command{@value{AS}} will print an error message. The message will
4704 include the value of @var{expression}. This can occasionally be useful inside
4705 complex nested macros or conditional assembly.
4708 @section @code{.file}
4709 @cindex @code{file} directive
4711 @ifclear no-file-dir
4712 There are two different versions of the @code{.file} directive. Targets
4713 that support DWARF2 line number information use the DWARF2 version of
4714 @code{.file}. Other targets use the default version.
4716 @subheading Default Version
4718 @cindex logical file name
4719 @cindex file name, logical
4720 This version of the @code{.file} directive tells @command{@value{AS}} that we
4721 are about to start a new logical file. The syntax is:
4727 @var{string} is the new file name. In general, the filename is
4728 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4729 to specify an empty file name, you must give the quotes--@code{""}. This
4730 statement may go away in future: it is only recognized to be compatible with
4731 old @command{@value{AS}} programs.
4733 @subheading DWARF2 Version
4736 When emitting DWARF2 line number information, @code{.file} assigns filenames
4737 to the @code{.debug_line} file name table. The syntax is:
4740 .file @var{fileno} @var{filename}
4743 The @var{fileno} operand should be a unique positive integer to use as the
4744 index of the entry in the table. The @var{filename} operand is a C string
4747 The detail of filename indices is exposed to the user because the filename
4748 table is shared with the @code{.debug_info} section of the DWARF2 debugging
4749 information, and thus the user must know the exact indices that table
4753 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4755 @cindex @code{fill} directive
4756 @cindex writing patterns in memory
4757 @cindex patterns, writing in memory
4758 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4759 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4760 may be zero or more. @var{Size} may be zero or more, but if it is
4761 more than 8, then it is deemed to have the value 8, compatible with
4762 other people's assemblers. The contents of each @var{repeat} bytes
4763 is taken from an 8-byte number. The highest order 4 bytes are
4764 zero. The lowest order 4 bytes are @var{value} rendered in the
4765 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4766 Each @var{size} bytes in a repetition is taken from the lowest order
4767 @var{size} bytes of this number. Again, this bizarre behavior is
4768 compatible with other people's assemblers.
4770 @var{size} and @var{value} are optional.
4771 If the second comma and @var{value} are absent, @var{value} is
4772 assumed zero. If the first comma and following tokens are absent,
4773 @var{size} is assumed to be 1.
4776 @section @code{.float @var{flonums}}
4778 @cindex floating point numbers (single)
4779 @cindex @code{float} directive
4780 This directive assembles zero or more flonums, separated by commas. It
4781 has the same effect as @code{.single}.
4783 The exact kind of floating point numbers emitted depends on how
4784 @command{@value{AS}} is configured.
4785 @xref{Machine Dependencies}.
4789 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4790 in @sc{ieee} format.
4795 @section @code{.func @var{name}[,@var{label}]}
4796 @cindex @code{func} directive
4797 @code{.func} emits debugging information to denote function @var{name}, and
4798 is ignored unless the file is assembled with debugging enabled.
4799 Only @samp{--gstabs[+]} is currently supported.
4800 @var{label} is the entry point of the function and if omitted @var{name}
4801 prepended with the @samp{leading char} is used.
4802 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4803 All functions are currently defined to have @code{void} return type.
4804 The function must be terminated with @code{.endfunc}.
4807 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4809 @cindex @code{global} directive
4810 @cindex symbol, making visible to linker
4811 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4812 @var{symbol} in your partial program, its value is made available to
4813 other partial programs that are linked with it. Otherwise,
4814 @var{symbol} takes its attributes from a symbol of the same name
4815 from another file linked into the same program.
4817 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4818 compatibility with other assemblers.
4821 On the HPPA, @code{.global} is not always enough to make it accessible to other
4822 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4823 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4828 @section @code{.gnu_attribute @var{tag},@var{value}}
4829 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
4832 @section @code{.hidden @var{names}}
4834 @cindex @code{hidden} directive
4836 This is one of the ELF visibility directives. The other two are
4837 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4838 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4840 This directive overrides the named symbols default visibility (which is set by
4841 their binding: local, global or weak). The directive sets the visibility to
4842 @code{hidden} which means that the symbols are not visible to other components.
4843 Such symbols are always considered to be @code{protected} as well.
4847 @section @code{.hword @var{expressions}}
4849 @cindex @code{hword} directive
4850 @cindex integers, 16-bit
4851 @cindex numbers, 16-bit
4852 @cindex sixteen bit integers
4853 This expects zero or more @var{expressions}, and emits
4854 a 16 bit number for each.
4857 This directive is a synonym for @samp{.short}; depending on the target
4858 architecture, it may also be a synonym for @samp{.word}.
4862 This directive is a synonym for @samp{.short}.
4865 This directive is a synonym for both @samp{.short} and @samp{.word}.
4870 @section @code{.ident}
4872 @cindex @code{ident} directive
4874 This directive is used by some assemblers to place tags in object files. The
4875 behavior of this directive varies depending on the target. When using the
4876 a.out object file format, @command{@value{AS}} simply accepts the directive for
4877 source-file compatibility with existing assemblers, but does not emit anything
4878 for it. When using COFF, comments are emitted to the @code{.comment} or
4879 @code{.rdata} section, depending on the target. When using ELF, comments are
4880 emitted to the @code{.comment} section.
4883 @section @code{.if @var{absolute expression}}
4885 @cindex conditional assembly
4886 @cindex @code{if} directive
4887 @code{.if} marks the beginning of a section of code which is only
4888 considered part of the source program being assembled if the argument
4889 (which must be an @var{absolute expression}) is non-zero. The end of
4890 the conditional section of code must be marked by @code{.endif}
4891 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4892 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4893 If you have several conditions to check, @code{.elseif} may be used to avoid
4894 nesting blocks if/else within each subsequent @code{.else} block.
4896 The following variants of @code{.if} are also supported:
4898 @cindex @code{ifdef} directive
4899 @item .ifdef @var{symbol}
4900 Assembles the following section of code if the specified @var{symbol}
4901 has been defined. Note a symbol which has been referenced but not yet defined
4902 is considered to be undefined.
4904 @cindex @code{ifb} directive
4905 @item .ifb @var{text}
4906 Assembles the following section of code if the operand is blank (empty).
4908 @cindex @code{ifc} directive
4909 @item .ifc @var{string1},@var{string2}
4910 Assembles the following section of code if the two strings are the same. The
4911 strings may be optionally quoted with single quotes. If they are not quoted,
4912 the first string stops at the first comma, and the second string stops at the
4913 end of the line. Strings which contain whitespace should be quoted. The
4914 string comparison is case sensitive.
4916 @cindex @code{ifeq} directive
4917 @item .ifeq @var{absolute expression}
4918 Assembles the following section of code if the argument is zero.
4920 @cindex @code{ifeqs} directive
4921 @item .ifeqs @var{string1},@var{string2}
4922 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4924 @cindex @code{ifge} directive
4925 @item .ifge @var{absolute expression}
4926 Assembles the following section of code if the argument is greater than or
4929 @cindex @code{ifgt} directive
4930 @item .ifgt @var{absolute expression}
4931 Assembles the following section of code if the argument is greater than zero.
4933 @cindex @code{ifle} directive
4934 @item .ifle @var{absolute expression}
4935 Assembles the following section of code if the argument is less than or equal
4938 @cindex @code{iflt} directive
4939 @item .iflt @var{absolute expression}
4940 Assembles the following section of code if the argument is less than zero.
4942 @cindex @code{ifnb} directive
4943 @item .ifnb @var{text}
4944 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
4945 following section of code if the operand is non-blank (non-empty).
4947 @cindex @code{ifnc} directive
4948 @item .ifnc @var{string1},@var{string2}.
4949 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4950 following section of code if the two strings are not the same.
4952 @cindex @code{ifndef} directive
4953 @cindex @code{ifnotdef} directive
4954 @item .ifndef @var{symbol}
4955 @itemx .ifnotdef @var{symbol}
4956 Assembles the following section of code if the specified @var{symbol}
4957 has not been defined. Both spelling variants are equivalent. Note a symbol
4958 which has been referenced but not yet defined is considered to be undefined.
4960 @cindex @code{ifne} directive
4961 @item .ifne @var{absolute expression}
4962 Assembles the following section of code if the argument is not equal to zero
4963 (in other words, this is equivalent to @code{.if}).
4965 @cindex @code{ifnes} directive
4966 @item .ifnes @var{string1},@var{string2}
4967 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4968 following section of code if the two strings are not the same.
4972 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4974 @cindex @code{incbin} directive
4975 @cindex binary files, including
4976 The @code{incbin} directive includes @var{file} verbatim at the current
4977 location. You can control the search paths used with the @samp{-I} command-line
4978 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4981 The @var{skip} argument skips a number of bytes from the start of the
4982 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4983 read. Note that the data is not aligned in any way, so it is the user's
4984 responsibility to make sure that proper alignment is provided both before and
4985 after the @code{incbin} directive.
4988 @section @code{.include "@var{file}"}
4990 @cindex @code{include} directive
4991 @cindex supporting files, including
4992 @cindex files, including
4993 This directive provides a way to include supporting files at specified
4994 points in your source program. The code from @var{file} is assembled as
4995 if it followed the point of the @code{.include}; when the end of the
4996 included file is reached, assembly of the original file continues. You
4997 can control the search paths used with the @samp{-I} command-line option
4998 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5002 @section @code{.int @var{expressions}}
5004 @cindex @code{int} directive
5005 @cindex integers, 32-bit
5006 Expect zero or more @var{expressions}, of any section, separated by commas.
5007 For each expression, emit a number that, at run time, is the value of that
5008 expression. The byte order and bit size of the number depends on what kind
5009 of target the assembly is for.
5013 On most forms of the H8/300, @code{.int} emits 16-bit
5014 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5021 @section @code{.internal @var{names}}
5023 @cindex @code{internal} directive
5025 This is one of the ELF visibility directives. The other two are
5026 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5027 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5029 This directive overrides the named symbols default visibility (which is set by
5030 their binding: local, global or weak). The directive sets the visibility to
5031 @code{internal} which means that the symbols are considered to be @code{hidden}
5032 (i.e., not visible to other components), and that some extra, processor specific
5033 processing must also be performed upon the symbols as well.
5037 @section @code{.irp @var{symbol},@var{values}}@dots{}
5039 @cindex @code{irp} directive
5040 Evaluate a sequence of statements assigning different values to @var{symbol}.
5041 The sequence of statements starts at the @code{.irp} directive, and is
5042 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5043 set to @var{value}, and the sequence of statements is assembled. If no
5044 @var{value} is listed, the sequence of statements is assembled once, with
5045 @var{symbol} set to the null string. To refer to @var{symbol} within the
5046 sequence of statements, use @var{\symbol}.
5048 For example, assembling
5056 is equivalent to assembling
5064 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5067 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5069 @cindex @code{irpc} directive
5070 Evaluate a sequence of statements assigning different values to @var{symbol}.
5071 The sequence of statements starts at the @code{.irpc} directive, and is
5072 terminated by an @code{.endr} directive. For each character in @var{value},
5073 @var{symbol} is set to the character, and the sequence of statements is
5074 assembled. If no @var{value} is listed, the sequence of statements is
5075 assembled once, with @var{symbol} set to the null string. To refer to
5076 @var{symbol} within the sequence of statements, use @var{\symbol}.
5078 For example, assembling
5086 is equivalent to assembling
5094 For some caveats with the spelling of @var{symbol}, see also the discussion
5098 @section @code{.lcomm @var{symbol} , @var{length}}
5100 @cindex @code{lcomm} directive
5101 @cindex local common symbols
5102 @cindex symbols, local common
5103 Reserve @var{length} (an absolute expression) bytes for a local common
5104 denoted by @var{symbol}. The section and value of @var{symbol} are
5105 those of the new local common. The addresses are allocated in the bss
5106 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5107 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5108 not visible to @code{@value{LD}}.
5111 Some targets permit a third argument to be used with @code{.lcomm}. This
5112 argument specifies the desired alignment of the symbol in the bss section.
5116 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5117 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5121 @section @code{.lflags}
5123 @cindex @code{lflags} directive (ignored)
5124 @command{@value{AS}} accepts this directive, for compatibility with other
5125 assemblers, but ignores it.
5127 @ifclear no-line-dir
5129 @section @code{.line @var{line-number}}
5131 @cindex @code{line} directive
5132 @cindex logical line number
5134 Change the logical line number. @var{line-number} must be an absolute
5135 expression. The next line has that logical line number. Therefore any other
5136 statements on the current line (after a statement separator character) are
5137 reported as on logical line number @var{line-number} @minus{} 1. One day
5138 @command{@value{AS}} will no longer support this directive: it is recognized only
5139 for compatibility with existing assembler programs.
5142 Even though this is a directive associated with the @code{a.out} or
5143 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5144 when producing COFF output, and treats @samp{.line} as though it
5145 were the COFF @samp{.ln} @emph{if} it is found outside a
5146 @code{.def}/@code{.endef} pair.
5148 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5149 used by compilers to generate auxiliary symbol information for
5154 @section @code{.linkonce [@var{type}]}
5156 @cindex @code{linkonce} directive
5157 @cindex common sections
5158 Mark the current section so that the linker only includes a single copy of it.
5159 This may be used to include the same section in several different object files,
5160 but ensure that the linker will only include it once in the final output file.
5161 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5162 Duplicate sections are detected based on the section name, so it should be
5165 This directive is only supported by a few object file formats; as of this
5166 writing, the only object file format which supports it is the Portable
5167 Executable format used on Windows NT.
5169 The @var{type} argument is optional. If specified, it must be one of the
5170 following strings. For example:
5174 Not all types may be supported on all object file formats.
5178 Silently discard duplicate sections. This is the default.
5181 Warn if there are duplicate sections, but still keep only one copy.
5184 Warn if any of the duplicates have different sizes.
5187 Warn if any of the duplicates do not have exactly the same contents.
5191 @section @code{.list}
5193 @cindex @code{list} directive
5194 @cindex listing control, turning on
5195 Control (in conjunction with the @code{.nolist} directive) whether or
5196 not assembly listings are generated. These two directives maintain an
5197 internal counter (which is zero initially). @code{.list} increments the
5198 counter, and @code{.nolist} decrements it. Assembly listings are
5199 generated whenever the counter is greater than zero.
5201 By default, listings are disabled. When you enable them (with the
5202 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5203 the initial value of the listing counter is one.
5206 @section @code{.ln @var{line-number}}
5208 @cindex @code{ln} directive
5209 @ifclear no-line-dir
5210 @samp{.ln} is a synonym for @samp{.line}.
5213 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5214 must be an absolute expression. The next line has that logical
5215 line number, so any other statements on the current line (after a
5216 statement separator character @code{;}) are reported as on logical
5217 line number @var{line-number} @minus{} 1.
5220 This directive is accepted, but ignored, when @command{@value{AS}} is
5221 configured for @code{b.out}; its effect is only associated with COFF
5227 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5228 @cindex @code{loc} directive
5229 When emitting DWARF2 line number information,
5230 the @code{.loc} directive will add a row to the @code{.debug_line} line
5231 number matrix corresponding to the immediately following assembly
5232 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5233 arguments will be applied to the @code{.debug_line} state machine before
5236 The @var{options} are a sequence of the following tokens in any order:
5240 This option will set the @code{basic_block} register in the
5241 @code{.debug_line} state machine to @code{true}.
5244 This option will set the @code{prologue_end} register in the
5245 @code{.debug_line} state machine to @code{true}.
5247 @item epilogue_begin
5248 This option will set the @code{epilogue_begin} register in the
5249 @code{.debug_line} state machine to @code{true}.
5251 @item is_stmt @var{value}
5252 This option will set the @code{is_stmt} register in the
5253 @code{.debug_line} state machine to @code{value}, which must be
5256 @item isa @var{value}
5257 This directive will set the @code{isa} register in the @code{.debug_line}
5258 state machine to @var{value}, which must be an unsigned integer.
5260 @item discriminator @var{value}
5261 This directive will set the @code{discriminator} register in the @code{.debug_line}
5262 state machine to @var{value}, which must be an unsigned integer.
5266 @node Loc_mark_labels
5267 @section @code{.loc_mark_labels @var{enable}}
5268 @cindex @code{loc_mark_labels} directive
5269 When emitting DWARF2 line number information,
5270 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5271 to the @code{.debug_line} line number matrix with the @code{basic_block}
5272 register in the state machine set whenever a code label is seen.
5273 The @var{enable} argument should be either 1 or 0, to enable or disable
5274 this function respectively.
5278 @section @code{.local @var{names}}
5280 @cindex @code{local} directive
5281 This directive, which is available for ELF targets, marks each symbol in
5282 the comma-separated list of @code{names} as a local symbol so that it
5283 will not be externally visible. If the symbols do not already exist,
5284 they will be created.
5286 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5287 accept an alignment argument, which is the case for most ELF targets,
5288 the @code{.local} directive can be used in combination with @code{.comm}
5289 (@pxref{Comm}) to define aligned local common data.
5293 @section @code{.long @var{expressions}}
5295 @cindex @code{long} directive
5296 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5299 @c no one seems to know what this is for or whether this description is
5300 @c what it really ought to do
5302 @section @code{.lsym @var{symbol}, @var{expression}}
5304 @cindex @code{lsym} directive
5305 @cindex symbol, not referenced in assembly
5306 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5307 the hash table, ensuring it cannot be referenced by name during the
5308 rest of the assembly. This sets the attributes of the symbol to be
5309 the same as the expression value:
5311 @var{other} = @var{descriptor} = 0
5312 @var{type} = @r{(section of @var{expression})}
5313 @var{value} = @var{expression}
5316 The new symbol is not flagged as external.
5320 @section @code{.macro}
5323 The commands @code{.macro} and @code{.endm} allow you to define macros that
5324 generate assembly output. For example, this definition specifies a macro
5325 @code{sum} that puts a sequence of numbers into memory:
5328 .macro sum from=0, to=5
5337 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5349 @item .macro @var{macname}
5350 @itemx .macro @var{macname} @var{macargs} @dots{}
5351 @cindex @code{macro} directive
5352 Begin the definition of a macro called @var{macname}. If your macro
5353 definition requires arguments, specify their names after the macro name,
5354 separated by commas or spaces. You can qualify the macro argument to
5355 indicate whether all invocations must specify a non-blank value (through
5356 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5357 (through @samp{:@code{vararg}}). You can supply a default value for any
5358 macro argument by following the name with @samp{=@var{deflt}}. You
5359 cannot define two macros with the same @var{macname} unless it has been
5360 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5361 definitions. For example, these are all valid @code{.macro} statements:
5365 Begin the definition of a macro called @code{comm}, which takes no
5368 @item .macro plus1 p, p1
5369 @itemx .macro plus1 p p1
5370 Either statement begins the definition of a macro called @code{plus1},
5371 which takes two arguments; within the macro definition, write
5372 @samp{\p} or @samp{\p1} to evaluate the arguments.
5374 @item .macro reserve_str p1=0 p2
5375 Begin the definition of a macro called @code{reserve_str}, with two
5376 arguments. The first argument has a default value, but not the second.
5377 After the definition is complete, you can call the macro either as
5378 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5379 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5380 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5381 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5383 @item .macro m p1:req, p2=0, p3:vararg
5384 Begin the definition of a macro called @code{m}, with at least three
5385 arguments. The first argument must always have a value specified, but
5386 not the second, which instead has a default value. The third formal
5387 will get assigned all remaining arguments specified at invocation time.
5389 When you call a macro, you can specify the argument values either by
5390 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5391 @samp{sum to=17, from=9}.
5395 Note that since each of the @var{macargs} can be an identifier exactly
5396 as any other one permitted by the target architecture, there may be
5397 occasional problems if the target hand-crafts special meanings to certain
5398 characters when they occur in a special position. For example, if the colon
5399 (@code{:}) is generally permitted to be part of a symbol name, but the
5400 architecture specific code special-cases it when occurring as the final
5401 character of a symbol (to denote a label), then the macro parameter
5402 replacement code will have no way of knowing that and consider the whole
5403 construct (including the colon) an identifier, and check only this
5404 identifier for being the subject to parameter substitution. So for example
5405 this macro definition:
5413 might not work as expected. Invoking @samp{label foo} might not create a label
5414 called @samp{foo} but instead just insert the text @samp{\l:} into the
5415 assembler source, probably generating an error about an unrecognised
5418 Similarly problems might occur with the period character (@samp{.})
5419 which is often allowed inside opcode names (and hence identifier names). So
5420 for example constructing a macro to build an opcode from a base name and a
5421 length specifier like this:
5424 .macro opcode base length
5429 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5430 instruction but instead generate some kind of error as the assembler tries to
5431 interpret the text @samp{\base.\length}.
5433 There are several possible ways around this problem:
5436 @item Insert white space
5437 If it is possible to use white space characters then this is the simplest
5446 @item Use @samp{\()}
5447 The string @samp{\()} can be used to separate the end of a macro argument from
5448 the following text. eg:
5451 .macro opcode base length
5456 @item Use the alternate macro syntax mode
5457 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5458 used as a separator. eg:
5468 Note: this problem of correctly identifying string parameters to pseudo ops
5469 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5470 and @code{.irpc} (@pxref{Irpc}) as well.
5473 @cindex @code{endm} directive
5474 Mark the end of a macro definition.
5477 @cindex @code{exitm} directive
5478 Exit early from the current macro definition.
5480 @cindex number of macros executed
5481 @cindex macros, count executed
5483 @command{@value{AS}} maintains a counter of how many macros it has
5484 executed in this pseudo-variable; you can copy that number to your
5485 output with @samp{\@@}, but @emph{only within a macro definition}.
5487 @item LOCAL @var{name} [ , @dots{} ]
5488 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5489 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5490 @xref{Altmacro,,@code{.altmacro}}.
5494 @section @code{.mri @var{val}}
5496 @cindex @code{mri} directive
5497 @cindex MRI mode, temporarily
5498 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5499 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5500 affects code assembled until the next @code{.mri} directive, or until the end
5501 of the file. @xref{M, MRI mode, MRI mode}.
5504 @section @code{.noaltmacro}
5505 Disable alternate macro mode. @xref{Altmacro}.
5508 @section @code{.nolist}
5510 @cindex @code{nolist} directive
5511 @cindex listing control, turning off
5512 Control (in conjunction with the @code{.list} directive) whether or
5513 not assembly listings are generated. These two directives maintain an
5514 internal counter (which is zero initially). @code{.list} increments the
5515 counter, and @code{.nolist} decrements it. Assembly listings are
5516 generated whenever the counter is greater than zero.
5519 @section @code{.octa @var{bignums}}
5521 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5522 @cindex @code{octa} directive
5523 @cindex integer, 16-byte
5524 @cindex sixteen byte integer
5525 This directive expects zero or more bignums, separated by commas. For each
5526 bignum, it emits a 16-byte integer.
5528 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5529 hence @emph{octa}-word for 16 bytes.
5532 @section @code{.org @var{new-lc} , @var{fill}}
5534 @cindex @code{org} directive
5535 @cindex location counter, advancing
5536 @cindex advancing location counter
5537 @cindex current address, advancing
5538 Advance the location counter of the current section to
5539 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5540 expression with the same section as the current subsection. That is,
5541 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5542 wrong section, the @code{.org} directive is ignored. To be compatible
5543 with former assemblers, if the section of @var{new-lc} is absolute,
5544 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5545 is the same as the current subsection.
5547 @code{.org} may only increase the location counter, or leave it
5548 unchanged; you cannot use @code{.org} to move the location counter
5551 @c double negative used below "not undefined" because this is a specific
5552 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5553 @c section. doc@cygnus.com 18feb91
5554 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5555 may not be undefined. If you really detest this restriction we eagerly await
5556 a chance to share your improved assembler.
5558 Beware that the origin is relative to the start of the section, not
5559 to the start of the subsection. This is compatible with other
5560 people's assemblers.
5562 When the location counter (of the current subsection) is advanced, the
5563 intervening bytes are filled with @var{fill} which should be an
5564 absolute expression. If the comma and @var{fill} are omitted,
5565 @var{fill} defaults to zero.
5568 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5570 @cindex padding the location counter given a power of two
5571 @cindex @code{p2align} directive
5572 Pad the location counter (in the current subsection) to a particular
5573 storage boundary. The first expression (which must be absolute) is the
5574 number of low-order zero bits the location counter must have after
5575 advancement. For example @samp{.p2align 3} advances the location
5576 counter until it a multiple of 8. If the location counter is already a
5577 multiple of 8, no change is needed.
5579 The second expression (also absolute) gives the fill value to be stored in the
5580 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5581 padding bytes are normally zero. However, on some systems, if the section is
5582 marked as containing code and the fill value is omitted, the space is filled
5583 with no-op instructions.
5585 The third expression is also absolute, and is also optional. If it is present,
5586 it is the maximum number of bytes that should be skipped by this alignment
5587 directive. If doing the alignment would require skipping more bytes than the
5588 specified maximum, then the alignment is not done at all. You can omit the
5589 fill value (the second argument) entirely by simply using two commas after the
5590 required alignment; this can be useful if you want the alignment to be filled
5591 with no-op instructions when appropriate.
5593 @cindex @code{p2alignw} directive
5594 @cindex @code{p2alignl} directive
5595 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5596 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5597 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5598 fill pattern as a four byte longword value. For example, @code{.p2alignw
5599 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5600 filled in with the value 0x368d (the exact placement of the bytes depends upon
5601 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5606 @section @code{.popsection}
5608 @cindex @code{popsection} directive
5609 @cindex Section Stack
5610 This is one of the ELF section stack manipulation directives. The others are
5611 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5612 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5615 This directive replaces the current section (and subsection) with the top
5616 section (and subsection) on the section stack. This section is popped off the
5622 @section @code{.previous}
5624 @cindex @code{previous} directive
5625 @cindex Section Stack
5626 This is one of the ELF section stack manipulation directives. The others are
5627 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5628 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5629 (@pxref{PopSection}).
5631 This directive swaps the current section (and subsection) with most recently
5632 referenced section/subsection pair prior to this one. Multiple
5633 @code{.previous} directives in a row will flip between two sections (and their
5634 subsections). For example:
5646 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5652 # Now in section A subsection 1
5656 # Now in section B subsection 0
5659 # Now in section B subsection 1
5662 # Now in section B subsection 0
5666 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5667 section B and 0x9abc into subsection 1 of section B.
5669 In terms of the section stack, this directive swaps the current section with
5670 the top section on the section stack.
5674 @section @code{.print @var{string}}
5676 @cindex @code{print} directive
5677 @command{@value{AS}} will print @var{string} on the standard output during
5678 assembly. You must put @var{string} in double quotes.
5682 @section @code{.protected @var{names}}
5684 @cindex @code{protected} directive
5686 This is one of the ELF visibility directives. The other two are
5687 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5689 This directive overrides the named symbols default visibility (which is set by
5690 their binding: local, global or weak). The directive sets the visibility to
5691 @code{protected} which means that any references to the symbols from within the
5692 components that defines them must be resolved to the definition in that
5693 component, even if a definition in another component would normally preempt
5698 @section @code{.psize @var{lines} , @var{columns}}
5700 @cindex @code{psize} directive
5701 @cindex listing control: paper size
5702 @cindex paper size, for listings
5703 Use this directive to declare the number of lines---and, optionally, the
5704 number of columns---to use for each page, when generating listings.
5706 If you do not use @code{.psize}, listings use a default line-count
5707 of 60. You may omit the comma and @var{columns} specification; the
5708 default width is 200 columns.
5710 @command{@value{AS}} generates formfeeds whenever the specified number of
5711 lines is exceeded (or whenever you explicitly request one, using
5714 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5715 those explicitly specified with @code{.eject}.
5718 @section @code{.purgem @var{name}}
5720 @cindex @code{purgem} directive
5721 Undefine the macro @var{name}, so that later uses of the string will not be
5722 expanded. @xref{Macro}.
5726 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5728 @cindex @code{pushsection} directive
5729 @cindex Section Stack
5730 This is one of the ELF section stack manipulation directives. The others are
5731 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5732 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5735 This directive pushes the current section (and subsection) onto the
5736 top of the section stack, and then replaces the current section and
5737 subsection with @code{name} and @code{subsection}. The optional
5738 @code{flags}, @code{type} and @code{arguments} are treated the same
5739 as in the @code{.section} (@pxref{Section}) directive.
5743 @section @code{.quad @var{bignums}}
5745 @cindex @code{quad} directive
5746 @code{.quad} expects zero or more bignums, separated by commas. For
5747 each bignum, it emits
5749 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5750 warning message; and just takes the lowest order 8 bytes of the bignum.
5751 @cindex eight-byte integer
5752 @cindex integer, 8-byte
5754 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5755 hence @emph{quad}-word for 8 bytes.
5758 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5759 warning message; and just takes the lowest order 16 bytes of the bignum.
5760 @cindex sixteen-byte integer
5761 @cindex integer, 16-byte
5765 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5767 @cindex @code{reloc} directive
5768 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5769 @var{expression}. If @var{offset} is a number, the relocation is generated in
5770 the current section. If @var{offset} is an expression that resolves to a
5771 symbol plus offset, the relocation is generated in the given symbol's section.
5772 @var{expression}, if present, must resolve to a symbol plus addend or to an
5773 absolute value, but note that not all targets support an addend. e.g. ELF REL
5774 targets such as i386 store an addend in the section contents rather than in the
5775 relocation. This low level interface does not support addends stored in the
5779 @section @code{.rept @var{count}}
5781 @cindex @code{rept} directive
5782 Repeat the sequence of lines between the @code{.rept} directive and the next
5783 @code{.endr} directive @var{count} times.
5785 For example, assembling
5793 is equivalent to assembling
5802 @section @code{.sbttl "@var{subheading}"}
5804 @cindex @code{sbttl} directive
5805 @cindex subtitles for listings
5806 @cindex listing control: subtitle
5807 Use @var{subheading} as the title (third line, immediately after the
5808 title line) when generating assembly listings.
5810 This directive affects subsequent pages, as well as the current page if
5811 it appears within ten lines of the top of a page.
5815 @section @code{.scl @var{class}}
5817 @cindex @code{scl} directive
5818 @cindex symbol storage class (COFF)
5819 @cindex COFF symbol storage class
5820 Set the storage-class value for a symbol. This directive may only be
5821 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5822 whether a symbol is static or external, or it may record further
5823 symbolic debugging information.
5826 The @samp{.scl} directive is primarily associated with COFF output; when
5827 configured to generate @code{b.out} output format, @command{@value{AS}}
5828 accepts this directive but ignores it.
5834 @section @code{.section @var{name}}
5836 @cindex named section
5837 Use the @code{.section} directive to assemble the following code into a section
5840 This directive is only supported for targets that actually support arbitrarily
5841 named sections; on @code{a.out} targets, for example, it is not accepted, even
5842 with a standard @code{a.out} section name.
5846 @c only print the extra heading if both COFF and ELF are set
5847 @subheading COFF Version
5850 @cindex @code{section} directive (COFF version)
5851 For COFF targets, the @code{.section} directive is used in one of the following
5855 .section @var{name}[, "@var{flags}"]
5856 .section @var{name}[, @var{subsection}]
5859 If the optional argument is quoted, it is taken as flags to use for the
5860 section. Each flag is a single character. The following flags are recognized:
5863 bss section (uninitialized data)
5865 section is not loaded
5875 shared section (meaningful for PE targets)
5877 ignored. (For compatibility with the ELF version)
5879 section is not readable (meaningful for PE targets)
5881 single-digit power-of-two section alignment (GNU extension)
5884 If no flags are specified, the default flags depend upon the section name. If
5885 the section name is not recognized, the default will be for the section to be
5886 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5887 from the section, rather than adding them, so if they are used on their own it
5888 will be as if no flags had been specified at all.
5890 If the optional argument to the @code{.section} directive is not quoted, it is
5891 taken as a subsection number (@pxref{Sub-Sections}).
5896 @c only print the extra heading if both COFF and ELF are set
5897 @subheading ELF Version
5900 @cindex Section Stack
5901 This is one of the ELF section stack manipulation directives. The others are
5902 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5903 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5904 @code{.previous} (@pxref{Previous}).
5906 @cindex @code{section} directive (ELF version)
5907 For ELF targets, the @code{.section} directive is used like this:
5910 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
5913 The optional @var{flags} argument is a quoted string which may contain any
5914 combination of the following characters:
5917 section is allocatable
5919 section is excluded from executable and shared library.
5923 section is executable
5925 section is mergeable
5927 section contains zero terminated strings
5929 section is a member of a section group
5931 section is used for thread-local-storage
5933 section is a member of the previously-current section's group, if any
5936 The optional @var{type} argument may contain one of the following constants:
5939 section contains data
5941 section does not contain data (i.e., section only occupies space)
5943 section contains data which is used by things other than the program
5945 section contains an array of pointers to init functions
5947 section contains an array of pointers to finish functions
5948 @item @@preinit_array
5949 section contains an array of pointers to pre-init functions
5952 Many targets only support the first three section types.
5954 Note on targets where the @code{@@} character is the start of a comment (eg
5955 ARM) then another character is used instead. For example the ARM port uses the
5958 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5959 be specified as well as an extra argument---@var{entsize}---like this:
5962 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5965 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5966 constants, each @var{entsize} octets long. Sections with both @code{M} and
5967 @code{S} must contain zero terminated strings where each character is
5968 @var{entsize} bytes long. The linker may remove duplicates within sections with
5969 the same name, same entity size and same flags. @var{entsize} must be an
5970 absolute expression. For sections with both @code{M} and @code{S}, a string
5971 which is a suffix of a larger string is considered a duplicate. Thus
5972 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
5973 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
5975 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5976 be present along with an additional field like this:
5979 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5982 The @var{GroupName} field specifies the name of the section group to which this
5983 particular section belongs. The optional linkage field can contain:
5986 indicates that only one copy of this section should be retained
5991 Note: if both the @var{M} and @var{G} flags are present then the fields for
5992 the Merge flag should come first, like this:
5995 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5998 If @var{flags} contains the @code{?} symbol then it may not also contain the
5999 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6000 present. Instead, @code{?} says to consider the section that's current before
6001 this directive. If that section used @code{G}, then the new section will use
6002 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6003 If not, then the @code{?} symbol has no effect.
6005 If no flags are specified, the default flags depend upon the section name. If
6006 the section name is not recognized, the default will be for the section to have
6007 none of the above flags: it will not be allocated in memory, nor writable, nor
6008 executable. The section will contain data.
6010 For ELF targets, the assembler supports another type of @code{.section}
6011 directive for compatibility with the Solaris assembler:
6014 .section "@var{name}"[, @var{flags}...]
6017 Note that the section name is quoted. There may be a sequence of comma
6021 section is allocatable
6025 section is executable
6027 section is excluded from executable and shared library.
6029 section is used for thread local storage
6032 This directive replaces the current section and subsection. See the
6033 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6034 some examples of how this directive and the other section stack directives
6040 @section @code{.set @var{symbol}, @var{expression}}
6042 @cindex @code{set} directive
6043 @cindex symbol value, setting
6044 Set the value of @var{symbol} to @var{expression}. This
6045 changes @var{symbol}'s value and type to conform to
6046 @var{expression}. If @var{symbol} was flagged as external, it remains
6047 flagged (@pxref{Symbol Attributes}).
6049 You may @code{.set} a symbol many times in the same assembly.
6051 If you @code{.set} a global symbol, the value stored in the object
6052 file is the last value stored into it.
6055 On Z80 @code{set} is a real instruction, use
6056 @samp{@var{symbol} defl @var{expression}} instead.
6060 @section @code{.short @var{expressions}}
6062 @cindex @code{short} directive
6064 @code{.short} is normally the same as @samp{.word}.
6065 @xref{Word,,@code{.word}}.
6067 In some configurations, however, @code{.short} and @code{.word} generate
6068 numbers of different lengths. @xref{Machine Dependencies}.
6072 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6075 This expects zero or more @var{expressions}, and emits
6076 a 16 bit number for each.
6081 @section @code{.single @var{flonums}}
6083 @cindex @code{single} directive
6084 @cindex floating point numbers (single)
6085 This directive assembles zero or more flonums, separated by commas. It
6086 has the same effect as @code{.float}.
6088 The exact kind of floating point numbers emitted depends on how
6089 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6093 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6094 numbers in @sc{ieee} format.
6100 @section @code{.size}
6102 This directive is used to set the size associated with a symbol.
6106 @c only print the extra heading if both COFF and ELF are set
6107 @subheading COFF Version
6110 @cindex @code{size} directive (COFF version)
6111 For COFF targets, the @code{.size} directive is only permitted inside
6112 @code{.def}/@code{.endef} pairs. It is used like this:
6115 .size @var{expression}
6119 @samp{.size} is only meaningful when generating COFF format output; when
6120 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6127 @c only print the extra heading if both COFF and ELF are set
6128 @subheading ELF Version
6131 @cindex @code{size} directive (ELF version)
6132 For ELF targets, the @code{.size} directive is used like this:
6135 .size @var{name} , @var{expression}
6138 This directive sets the size associated with a symbol @var{name}.
6139 The size in bytes is computed from @var{expression} which can make use of label
6140 arithmetic. This directive is typically used to set the size of function
6145 @ifclear no-space-dir
6147 @section @code{.skip @var{size} , @var{fill}}
6149 @cindex @code{skip} directive
6150 @cindex filling memory
6151 This directive emits @var{size} bytes, each of value @var{fill}. Both
6152 @var{size} and @var{fill} are absolute expressions. If the comma and
6153 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6158 @section @code{.sleb128 @var{expressions}}
6160 @cindex @code{sleb128} directive
6161 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6162 compact, variable length representation of numbers used by the DWARF
6163 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6165 @ifclear no-space-dir
6167 @section @code{.space @var{size} , @var{fill}}
6169 @cindex @code{space} directive
6170 @cindex filling memory
6171 This directive emits @var{size} bytes, each of value @var{fill}. Both
6172 @var{size} and @var{fill} are absolute expressions. If the comma
6173 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6178 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6179 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6180 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6181 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6189 @section @code{.stabd, .stabn, .stabs}
6191 @cindex symbolic debuggers, information for
6192 @cindex @code{stab@var{x}} directives
6193 There are three directives that begin @samp{.stab}.
6194 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6195 The symbols are not entered in the @command{@value{AS}} hash table: they
6196 cannot be referenced elsewhere in the source file.
6197 Up to five fields are required:
6201 This is the symbol's name. It may contain any character except
6202 @samp{\000}, so is more general than ordinary symbol names. Some
6203 debuggers used to code arbitrarily complex structures into symbol names
6207 An absolute expression. The symbol's type is set to the low 8 bits of
6208 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6209 and debuggers choke on silly bit patterns.
6212 An absolute expression. The symbol's ``other'' attribute is set to the
6213 low 8 bits of this expression.
6216 An absolute expression. The symbol's descriptor is set to the low 16
6217 bits of this expression.
6220 An absolute expression which becomes the symbol's value.
6223 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6224 or @code{.stabs} statement, the symbol has probably already been created;
6225 you get a half-formed symbol in your object file. This is
6226 compatible with earlier assemblers!
6229 @cindex @code{stabd} directive
6230 @item .stabd @var{type} , @var{other} , @var{desc}
6232 The ``name'' of the symbol generated is not even an empty string.
6233 It is a null pointer, for compatibility. Older assemblers used a
6234 null pointer so they didn't waste space in object files with empty
6237 The symbol's value is set to the location counter,
6238 relocatably. When your program is linked, the value of this symbol
6239 is the address of the location counter when the @code{.stabd} was
6242 @cindex @code{stabn} directive
6243 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6244 The name of the symbol is set to the empty string @code{""}.
6246 @cindex @code{stabs} directive
6247 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6248 All five fields are specified.
6254 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6255 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6257 @cindex string, copying to object file
6258 @cindex string8, copying to object file
6259 @cindex string16, copying to object file
6260 @cindex string32, copying to object file
6261 @cindex string64, copying to object file
6262 @cindex @code{string} directive
6263 @cindex @code{string8} directive
6264 @cindex @code{string16} directive
6265 @cindex @code{string32} directive
6266 @cindex @code{string64} directive
6268 Copy the characters in @var{str} to the object file. You may specify more than
6269 one string to copy, separated by commas. Unless otherwise specified for a
6270 particular machine, the assembler marks the end of each string with a 0 byte.
6271 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6273 The variants @code{string16}, @code{string32} and @code{string64} differ from
6274 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6275 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6276 are stored in target endianness byte order.
6282 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6283 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6288 @section @code{.struct @var{expression}}
6290 @cindex @code{struct} directive
6291 Switch to the absolute section, and set the section offset to @var{expression},
6292 which must be an absolute expression. You might use this as follows:
6301 This would define the symbol @code{field1} to have the value 0, the symbol
6302 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6303 value 8. Assembly would be left in the absolute section, and you would need to
6304 use a @code{.section} directive of some sort to change to some other section
6305 before further assembly.
6309 @section @code{.subsection @var{name}}
6311 @cindex @code{subsection} directive
6312 @cindex Section Stack
6313 This is one of the ELF section stack manipulation directives. The others are
6314 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6315 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6318 This directive replaces the current subsection with @code{name}. The current
6319 section is not changed. The replaced subsection is put onto the section stack
6320 in place of the then current top of stack subsection.
6325 @section @code{.symver}
6326 @cindex @code{symver} directive
6327 @cindex symbol versioning
6328 @cindex versions of symbols
6329 Use the @code{.symver} directive to bind symbols to specific version nodes
6330 within a source file. This is only supported on ELF platforms, and is
6331 typically used when assembling files to be linked into a shared library.
6332 There are cases where it may make sense to use this in objects to be bound
6333 into an application itself so as to override a versioned symbol from a
6336 For ELF targets, the @code{.symver} directive can be used like this:
6338 .symver @var{name}, @var{name2@@nodename}
6340 If the symbol @var{name} is defined within the file
6341 being assembled, the @code{.symver} directive effectively creates a symbol
6342 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6343 just don't try and create a regular alias is that the @var{@@} character isn't
6344 permitted in symbol names. The @var{name2} part of the name is the actual name
6345 of the symbol by which it will be externally referenced. The name @var{name}
6346 itself is merely a name of convenience that is used so that it is possible to
6347 have definitions for multiple versions of a function within a single source
6348 file, and so that the compiler can unambiguously know which version of a
6349 function is being mentioned. The @var{nodename} portion of the alias should be
6350 the name of a node specified in the version script supplied to the linker when
6351 building a shared library. If you are attempting to override a versioned
6352 symbol from a shared library, then @var{nodename} should correspond to the
6353 nodename of the symbol you are trying to override.
6355 If the symbol @var{name} is not defined within the file being assembled, all
6356 references to @var{name} will be changed to @var{name2@@nodename}. If no
6357 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6360 Another usage of the @code{.symver} directive is:
6362 .symver @var{name}, @var{name2@@@@nodename}
6364 In this case, the symbol @var{name} must exist and be defined within
6365 the file being assembled. It is similar to @var{name2@@nodename}. The
6366 difference is @var{name2@@@@nodename} will also be used to resolve
6367 references to @var{name2} by the linker.
6369 The third usage of the @code{.symver} directive is:
6371 .symver @var{name}, @var{name2@@@@@@nodename}
6373 When @var{name} is not defined within the
6374 file being assembled, it is treated as @var{name2@@nodename}. When
6375 @var{name} is defined within the file being assembled, the symbol
6376 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6381 @section @code{.tag @var{structname}}
6383 @cindex COFF structure debugging
6384 @cindex structure debugging, COFF
6385 @cindex @code{tag} directive
6386 This directive is generated by compilers to include auxiliary debugging
6387 information in the symbol table. It is only permitted inside
6388 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6389 definitions in the symbol table with instances of those structures.
6392 @samp{.tag} is only used when generating COFF format output; when
6393 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6399 @section @code{.text @var{subsection}}
6401 @cindex @code{text} directive
6402 Tells @command{@value{AS}} to assemble the following statements onto the end of
6403 the text subsection numbered @var{subsection}, which is an absolute
6404 expression. If @var{subsection} is omitted, subsection number zero
6408 @section @code{.title "@var{heading}"}
6410 @cindex @code{title} directive
6411 @cindex listing control: title line
6412 Use @var{heading} as the title (second line, immediately after the
6413 source file name and pagenumber) when generating assembly listings.
6415 This directive affects subsequent pages, as well as the current page if
6416 it appears within ten lines of the top of a page.
6420 @section @code{.type}
6422 This directive is used to set the type of a symbol.
6426 @c only print the extra heading if both COFF and ELF are set
6427 @subheading COFF Version
6430 @cindex COFF symbol type
6431 @cindex symbol type, COFF
6432 @cindex @code{type} directive (COFF version)
6433 For COFF targets, this directive is permitted only within
6434 @code{.def}/@code{.endef} pairs. It is used like this:
6440 This records the integer @var{int} as the type attribute of a symbol table
6444 @samp{.type} is associated only with COFF format output; when
6445 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6446 directive but ignores it.
6452 @c only print the extra heading if both COFF and ELF are set
6453 @subheading ELF Version
6456 @cindex ELF symbol type
6457 @cindex symbol type, ELF
6458 @cindex @code{type} directive (ELF version)
6459 For ELF targets, the @code{.type} directive is used like this:
6462 .type @var{name} , @var{type description}
6465 This sets the type of symbol @var{name} to be either a
6466 function symbol or an object symbol. There are five different syntaxes
6467 supported for the @var{type description} field, in order to provide
6468 compatibility with various other assemblers.
6470 Because some of the characters used in these syntaxes (such as @samp{@@} and
6471 @samp{#}) are comment characters for some architectures, some of the syntaxes
6472 below do not work on all architectures. The first variant will be accepted by
6473 the GNU assembler on all architectures so that variant should be used for
6474 maximum portability, if you do not need to assemble your code with other
6477 The syntaxes supported are:
6480 .type <name> STT_<TYPE_IN_UPPER_CASE>
6481 .type <name>,#<type>
6482 .type <name>,@@<type>
6483 .type <name>,%<type>
6484 .type <name>,"<type>"
6487 The types supported are:
6492 Mark the symbol as being a function name.
6495 @itemx gnu_indirect_function
6496 Mark the symbol as an indirect function when evaluated during reloc
6497 processing. (This is only supported on Linux targeted assemblers).
6501 Mark the symbol as being a data object.
6505 Mark the symbol as being a thead-local data object.
6509 Mark the symbol as being a common data object.
6513 Does not mark the symbol in any way. It is supported just for completeness.
6515 @item gnu_unique_object
6516 Marks the symbol as being a globally unique data object. The dynamic linker
6517 will make sure that in the entire process there is just one symbol with this
6518 name and type in use. (This is only supported on Linux targeted assemblers).
6522 Note: Some targets support extra types in addition to those listed above.
6528 @section @code{.uleb128 @var{expressions}}
6530 @cindex @code{uleb128} directive
6531 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6532 compact, variable length representation of numbers used by the DWARF
6533 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6537 @section @code{.val @var{addr}}
6539 @cindex @code{val} directive
6540 @cindex COFF value attribute
6541 @cindex value attribute, COFF
6542 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6543 records the address @var{addr} as the value attribute of a symbol table
6547 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6548 configured for @code{b.out}, it accepts this directive but ignores it.
6554 @section @code{.version "@var{string}"}
6556 @cindex @code{version} directive
6557 This directive creates a @code{.note} section and places into it an ELF
6558 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6563 @section @code{.vtable_entry @var{table}, @var{offset}}
6565 @cindex @code{vtable_entry} directive
6566 This directive finds or creates a symbol @code{table} and creates a
6567 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6570 @section @code{.vtable_inherit @var{child}, @var{parent}}
6572 @cindex @code{vtable_inherit} directive
6573 This directive finds the symbol @code{child} and finds or creates the symbol
6574 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6575 parent whose addend is the value of the child symbol. As a special case the
6576 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6580 @section @code{.warning "@var{string}"}
6581 @cindex warning directive
6582 Similar to the directive @code{.error}
6583 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6586 @section @code{.weak @var{names}}
6588 @cindex @code{weak} directive
6589 This directive sets the weak attribute on the comma separated list of symbol
6590 @code{names}. If the symbols do not already exist, they will be created.
6592 On COFF targets other than PE, weak symbols are a GNU extension. This
6593 directive sets the weak attribute on the comma separated list of symbol
6594 @code{names}. If the symbols do not already exist, they will be created.
6596 On the PE target, weak symbols are supported natively as weak aliases.
6597 When a weak symbol is created that is not an alias, GAS creates an
6598 alternate symbol to hold the default value.
6601 @section @code{.weakref @var{alias}, @var{target}}
6603 @cindex @code{weakref} directive
6604 This directive creates an alias to the target symbol that enables the symbol to
6605 be referenced with weak-symbol semantics, but without actually making it weak.
6606 If direct references or definitions of the symbol are present, then the symbol
6607 will not be weak, but if all references to it are through weak references, the
6608 symbol will be marked as weak in the symbol table.
6610 The effect is equivalent to moving all references to the alias to a separate
6611 assembly source file, renaming the alias to the symbol in it, declaring the
6612 symbol as weak there, and running a reloadable link to merge the object files
6613 resulting from the assembly of the new source file and the old source file that
6614 had the references to the alias removed.
6616 The alias itself never makes to the symbol table, and is entirely handled
6617 within the assembler.
6620 @section @code{.word @var{expressions}}
6622 @cindex @code{word} directive
6623 This directive expects zero or more @var{expressions}, of any section,
6624 separated by commas.
6627 For each expression, @command{@value{AS}} emits a 32-bit number.
6630 For each expression, @command{@value{AS}} emits a 16-bit number.
6635 The size of the number emitted, and its byte order,
6636 depend on what target computer the assembly is for.
6639 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6640 @c happen---32-bit addressability, period; no long/short jumps.
6641 @ifset DIFF-TBL-KLUGE
6642 @cindex difference tables altered
6643 @cindex altered difference tables
6645 @emph{Warning: Special Treatment to support Compilers}
6649 Machines with a 32-bit address space, but that do less than 32-bit
6650 addressing, require the following special treatment. If the machine of
6651 interest to you does 32-bit addressing (or doesn't require it;
6652 @pxref{Machine Dependencies}), you can ignore this issue.
6655 In order to assemble compiler output into something that works,
6656 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6657 Directives of the form @samp{.word sym1-sym2} are often emitted by
6658 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6659 directive of the form @samp{.word sym1-sym2}, and the difference between
6660 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6661 creates a @dfn{secondary jump table}, immediately before the next label.
6662 This secondary jump table is preceded by a short-jump to the
6663 first byte after the secondary table. This short-jump prevents the flow
6664 of control from accidentally falling into the new table. Inside the
6665 table is a long-jump to @code{sym2}. The original @samp{.word}
6666 contains @code{sym1} minus the address of the long-jump to
6669 If there were several occurrences of @samp{.word sym1-sym2} before the
6670 secondary jump table, all of them are adjusted. If there was a
6671 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6672 long-jump to @code{sym4} is included in the secondary jump table,
6673 and the @code{.word} directives are adjusted to contain @code{sym3}
6674 minus the address of the long-jump to @code{sym4}; and so on, for as many
6675 entries in the original jump table as necessary.
6678 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6679 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6680 assembly language programmers.
6683 @c end DIFF-TBL-KLUGE
6686 @section Deprecated Directives
6688 @cindex deprecated directives
6689 @cindex obsolescent directives
6690 One day these directives won't work.
6691 They are included for compatibility with older assemblers.
6698 @node Object Attributes
6699 @chapter Object Attributes
6700 @cindex object attributes
6702 @command{@value{AS}} assembles source files written for a specific architecture
6703 into object files for that architecture. But not all object files are alike.
6704 Many architectures support incompatible variations. For instance, floating
6705 point arguments might be passed in floating point registers if the object file
6706 requires hardware floating point support---or floating point arguments might be
6707 passed in integer registers if the object file supports processors with no
6708 hardware floating point unit. Or, if two objects are built for different
6709 generations of the same architecture, the combination may require the
6710 newer generation at run-time.
6712 This information is useful during and after linking. At link time,
6713 @command{@value{LD}} can warn about incompatible object files. After link
6714 time, tools like @command{gdb} can use it to process the linked file
6717 Compatibility information is recorded as a series of object attributes. Each
6718 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6719 string, and indicates who sets the meaning of the tag. The tag is an integer,
6720 and indicates what property the attribute describes. The value may be a string
6721 or an integer, and indicates how the property affects this object. Missing
6722 attributes are the same as attributes with a zero value or empty string value.
6724 Object attributes were developed as part of the ABI for the ARM Architecture.
6725 The file format is documented in @cite{ELF for the ARM Architecture}.
6728 * GNU Object Attributes:: @sc{gnu} Object Attributes
6729 * Defining New Object Attributes:: Defining New Object Attributes
6732 @node GNU Object Attributes
6733 @section @sc{gnu} Object Attributes
6735 The @code{.gnu_attribute} directive records an object attribute
6736 with vendor @samp{gnu}.
6738 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6739 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6740 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6741 2} is set for architecture-independent attributes and clear for
6742 architecture-dependent ones.
6744 @subsection Common @sc{gnu} attributes
6746 These attributes are valid on all architectures.
6749 @item Tag_compatibility (32)
6750 The compatibility attribute takes an integer flag value and a vendor name. If
6751 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6752 then the file is only compatible with the named toolchain. If it is greater
6753 than 1, the file can only be processed by other toolchains under some private
6754 arrangement indicated by the flag value and the vendor name.
6757 @subsection MIPS Attributes
6760 @item Tag_GNU_MIPS_ABI_FP (4)
6761 The floating-point ABI used by this object file. The value will be:
6765 0 for files not affected by the floating-point ABI.
6767 1 for files using the hardware floating-point with a standard double-precision
6770 2 for files using the hardware floating-point ABI with a single-precision FPU.
6772 3 for files using the software floating-point ABI.
6774 4 for files using the hardware floating-point ABI with 64-bit wide
6775 double-precision floating-point registers and 32-bit wide general
6780 @subsection PowerPC Attributes
6783 @item Tag_GNU_Power_ABI_FP (4)
6784 The floating-point ABI used by this object file. The value will be:
6788 0 for files not affected by the floating-point ABI.
6790 1 for files using double-precision hardware floating-point ABI.
6792 2 for files using the software floating-point ABI.
6794 3 for files using single-precision hardware floating-point ABI.
6797 @item Tag_GNU_Power_ABI_Vector (8)
6798 The vector ABI used by this object file. The value will be:
6802 0 for files not affected by the vector ABI.
6804 1 for files using general purpose registers to pass vectors.
6806 2 for files using AltiVec registers to pass vectors.
6808 3 for files using SPE registers to pass vectors.
6812 @node Defining New Object Attributes
6813 @section Defining New Object Attributes
6815 If you want to define a new @sc{gnu} object attribute, here are the places you
6816 will need to modify. New attributes should be discussed on the @samp{binutils}
6821 This manual, which is the official register of attributes.
6823 The header for your architecture @file{include/elf}, to define the tag.
6825 The @file{bfd} support file for your architecture, to merge the attribute
6826 and issue any appropriate link warnings.
6828 Test cases in @file{ld/testsuite} for merging and link warnings.
6830 @file{binutils/readelf.c} to display your attribute.
6832 GCC, if you want the compiler to mark the attribute automatically.
6838 @node Machine Dependencies
6839 @chapter Machine Dependent Features
6841 @cindex machine dependencies
6842 The machine instruction sets are (almost by definition) different on
6843 each machine where @command{@value{AS}} runs. Floating point representations
6844 vary as well, and @command{@value{AS}} often supports a few additional
6845 directives or command-line options for compatibility with other
6846 assemblers on a particular platform. Finally, some versions of
6847 @command{@value{AS}} support special pseudo-instructions for branch
6850 This chapter discusses most of these differences, though it does not
6851 include details on any machine's instruction set. For details on that
6852 subject, see the hardware manufacturer's manual.
6856 * Alpha-Dependent:: Alpha Dependent Features
6859 * ARC-Dependent:: ARC Dependent Features
6862 * ARM-Dependent:: ARM Dependent Features
6865 * AVR-Dependent:: AVR Dependent Features
6868 * Blackfin-Dependent:: Blackfin Dependent Features
6871 * CR16-Dependent:: CR16 Dependent Features
6874 * CRIS-Dependent:: CRIS Dependent Features
6877 * D10V-Dependent:: D10V Dependent Features
6880 * D30V-Dependent:: D30V Dependent Features
6883 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6886 * HPPA-Dependent:: HPPA Dependent Features
6889 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6892 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6895 * i860-Dependent:: Intel 80860 Dependent Features
6898 * i960-Dependent:: Intel 80960 Dependent Features
6901 * IA-64-Dependent:: Intel IA-64 Dependent Features
6904 * IP2K-Dependent:: IP2K Dependent Features
6907 * LM32-Dependent:: LM32 Dependent Features
6910 * M32C-Dependent:: M32C Dependent Features
6913 * M32R-Dependent:: M32R Dependent Features
6916 * M68K-Dependent:: M680x0 Dependent Features
6919 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6922 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
6925 * MIPS-Dependent:: MIPS Dependent Features
6928 * MMIX-Dependent:: MMIX Dependent Features
6931 * MSP430-Dependent:: MSP430 Dependent Features
6934 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6935 * SH64-Dependent:: SuperH SH64 Dependent Features
6938 * PDP-11-Dependent:: PDP-11 Dependent Features
6941 * PJ-Dependent:: picoJava Dependent Features
6944 * PPC-Dependent:: PowerPC Dependent Features
6947 * RX-Dependent:: RX Dependent Features
6950 * S/390-Dependent:: IBM S/390 Dependent Features
6953 * SCORE-Dependent:: SCORE Dependent Features
6956 * Sparc-Dependent:: SPARC Dependent Features
6959 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6962 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
6965 * V850-Dependent:: V850 Dependent Features
6968 * Xtensa-Dependent:: Xtensa Dependent Features
6971 * Z80-Dependent:: Z80 Dependent Features
6974 * Z8000-Dependent:: Z8000 Dependent Features
6977 * Vax-Dependent:: VAX Dependent Features
6984 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6985 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6986 @c peculiarity: to preserve cross-references, there must be a node called
6987 @c "Machine Dependencies". Hence the conditional nodenames in each
6988 @c major node below. Node defaulting in makeinfo requires adjacency of
6989 @c node and sectioning commands; hence the repetition of @chapter BLAH
6990 @c in both conditional blocks.
6993 @include c-alpha.texi
7009 @include c-bfin.texi
7013 @include c-cr16.texi
7017 @include c-cris.texi
7022 @node Machine Dependencies
7023 @chapter Machine Dependent Features
7025 The machine instruction sets are different on each Renesas chip family,
7026 and there are also some syntax differences among the families. This
7027 chapter describes the specific @command{@value{AS}} features for each
7031 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7032 * SH-Dependent:: Renesas SH Dependent Features
7039 @include c-d10v.texi
7043 @include c-d30v.texi
7047 @include c-h8300.texi
7051 @include c-hppa.texi
7055 @include c-i370.texi
7059 @include c-i386.texi
7063 @include c-i860.texi
7067 @include c-i960.texi
7071 @include c-ia64.texi
7075 @include c-ip2k.texi
7079 @include c-lm32.texi
7083 @include c-m32c.texi
7087 @include c-m32r.texi
7091 @include c-m68k.texi
7095 @include c-m68hc11.texi
7099 @include c-microblaze.texi
7103 @include c-mips.texi
7107 @include c-mmix.texi
7111 @include c-msp430.texi
7115 @include c-ns32k.texi
7119 @include c-pdp11.texi
7135 @include c-s390.texi
7139 @include c-score.texi
7144 @include c-sh64.texi
7148 @include c-sparc.texi
7152 @include c-tic54x.texi
7156 @include c-tic6x.texi
7172 @include c-v850.texi
7176 @include c-xtensa.texi
7180 @c reverse effect of @down at top of generic Machine-Dep chapter
7184 @node Reporting Bugs
7185 @chapter Reporting Bugs
7186 @cindex bugs in assembler
7187 @cindex reporting bugs in assembler
7189 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7191 Reporting a bug may help you by bringing a solution to your problem, or it may
7192 not. But in any case the principal function of a bug report is to help the
7193 entire community by making the next version of @command{@value{AS}} work better.
7194 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7196 In order for a bug report to serve its purpose, you must include the
7197 information that enables us to fix the bug.
7200 * Bug Criteria:: Have you found a bug?
7201 * Bug Reporting:: How to report bugs
7205 @section Have You Found a Bug?
7206 @cindex bug criteria
7208 If you are not sure whether you have found a bug, here are some guidelines:
7211 @cindex fatal signal
7212 @cindex assembler crash
7213 @cindex crash of assembler
7215 If the assembler gets a fatal signal, for any input whatever, that is a
7216 @command{@value{AS}} bug. Reliable assemblers never crash.
7218 @cindex error on valid input
7220 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7222 @cindex invalid input
7224 If @command{@value{AS}} does not produce an error message for invalid input, that
7225 is a bug. However, you should note that your idea of ``invalid input'' might
7226 be our idea of ``an extension'' or ``support for traditional practice''.
7229 If you are an experienced user of assemblers, your suggestions for improvement
7230 of @command{@value{AS}} are welcome in any case.
7234 @section How to Report Bugs
7236 @cindex assembler bugs, reporting
7238 A number of companies and individuals offer support for @sc{gnu} products. If
7239 you obtained @command{@value{AS}} from a support organization, we recommend you
7240 contact that organization first.
7242 You can find contact information for many support companies and
7243 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7247 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7251 The fundamental principle of reporting bugs usefully is this:
7252 @strong{report all the facts}. If you are not sure whether to state a
7253 fact or leave it out, state it!
7255 Often people omit facts because they think they know what causes the problem
7256 and assume that some details do not matter. Thus, you might assume that the
7257 name of a symbol you use in an example does not matter. Well, probably it does
7258 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7259 happens to fetch from the location where that name is stored in memory;
7260 perhaps, if the name were different, the contents of that location would fool
7261 the assembler into doing the right thing despite the bug. Play it safe and
7262 give a specific, complete example. That is the easiest thing for you to do,
7263 and the most helpful.
7265 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7266 it is new to us. Therefore, always write your bug reports on the assumption
7267 that the bug has not been reported previously.
7269 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7270 bell?'' This cannot help us fix a bug, so it is basically useless. We
7271 respond by asking for enough details to enable us to investigate.
7272 You might as well expedite matters by sending them to begin with.
7274 To enable us to fix the bug, you should include all these things:
7278 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7279 it with the @samp{--version} argument.
7281 Without this, we will not know whether there is any point in looking for
7282 the bug in the current version of @command{@value{AS}}.
7285 Any patches you may have applied to the @command{@value{AS}} source.
7288 The type of machine you are using, and the operating system name and
7292 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7296 The command arguments you gave the assembler to assemble your example and
7297 observe the bug. To guarantee you will not omit something important, list them
7298 all. A copy of the Makefile (or the output from make) is sufficient.
7300 If we were to try to guess the arguments, we would probably guess wrong
7301 and then we might not encounter the bug.
7304 A complete input file that will reproduce the bug. If the bug is observed when
7305 the assembler is invoked via a compiler, send the assembler source, not the
7306 high level language source. Most compilers will produce the assembler source
7307 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7308 the options @samp{-v --save-temps}; this will save the assembler source in a
7309 file with an extension of @file{.s}, and also show you exactly how
7310 @command{@value{AS}} is being run.
7313 A description of what behavior you observe that you believe is
7314 incorrect. For example, ``It gets a fatal signal.''
7316 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7317 will certainly notice it. But if the bug is incorrect output, we might not
7318 notice unless it is glaringly wrong. You might as well not give us a chance to
7321 Even if the problem you experience is a fatal signal, you should still say so
7322 explicitly. Suppose something strange is going on, such as, your copy of
7323 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7324 library on your system. (This has happened!) Your copy might crash and ours
7325 would not. If you told us to expect a crash, then when ours fails to crash, we
7326 would know that the bug was not happening for us. If you had not told us to
7327 expect a crash, then we would not be able to draw any conclusion from our
7331 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7332 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7333 option. Always send diffs from the old file to the new file. If you even
7334 discuss something in the @command{@value{AS}} source, refer to it by context, not
7337 The line numbers in our development sources will not match those in your
7338 sources. Your line numbers would convey no useful information to us.
7341 Here are some things that are not necessary:
7345 A description of the envelope of the bug.
7347 Often people who encounter a bug spend a lot of time investigating
7348 which changes to the input file will make the bug go away and which
7349 changes will not affect it.
7351 This is often time consuming and not very useful, because the way we
7352 will find the bug is by running a single example under the debugger
7353 with breakpoints, not by pure deduction from a series of examples.
7354 We recommend that you save your time for something else.
7356 Of course, if you can find a simpler example to report @emph{instead}
7357 of the original one, that is a convenience for us. Errors in the
7358 output will be easier to spot, running under the debugger will take
7359 less time, and so on.
7361 However, simplification is not vital; if you do not want to do this,
7362 report the bug anyway and send us the entire test case you used.
7365 A patch for the bug.
7367 A patch for the bug does help us if it is a good one. But do not omit
7368 the necessary information, such as the test case, on the assumption that
7369 a patch is all we need. We might see problems with your patch and decide
7370 to fix the problem another way, or we might not understand it at all.
7372 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7373 construct an example that will make the program follow a certain path through
7374 the code. If you do not send us the example, we will not be able to construct
7375 one, so we will not be able to verify that the bug is fixed.
7377 And if we cannot understand what bug you are trying to fix, or why your
7378 patch should be an improvement, we will not install it. A test case will
7379 help us to understand.
7382 A guess about what the bug is or what it depends on.
7384 Such guesses are usually wrong. Even we cannot guess right about such
7385 things without first using the debugger to find the facts.
7388 @node Acknowledgements
7389 @chapter Acknowledgements
7391 If you have contributed to GAS and your name isn't listed here,
7392 it is not meant as a slight. We just don't know about it. Send mail to the
7393 maintainer, and we'll correct the situation. Currently
7395 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
7397 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7400 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7401 information and the 68k series machines, most of the preprocessing pass, and
7402 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7404 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7405 many bug fixes, including merging support for several processors, breaking GAS
7406 up to handle multiple object file format back ends (including heavy rewrite,
7407 testing, an integration of the coff and b.out back ends), adding configuration
7408 including heavy testing and verification of cross assemblers and file splits
7409 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7410 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7411 port (including considerable amounts of reverse engineering), a SPARC opcode
7412 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7413 assertions and made them work, much other reorganization, cleanup, and lint.
7415 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7416 in format-specific I/O modules.
7418 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7419 has done much work with it since.
7421 The Intel 80386 machine description was written by Eliot Dresselhaus.
7423 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7425 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7426 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7428 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7429 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7430 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7431 support a.out format.
7433 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7434 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7435 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7436 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7439 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7440 simplified the configuration of which versions accept which directives. He
7441 updated the 68k machine description so that Motorola's opcodes always produced
7442 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7443 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7444 cross-compilation support, and one bug in relaxation that took a week and
7445 required the proverbial one-bit fix.
7447 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7448 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7449 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7450 PowerPC assembler, and made a few other minor patches.
7452 Steve Chamberlain made GAS able to generate listings.
7454 Hewlett-Packard contributed support for the HP9000/300.
7456 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7457 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7458 formats). This work was supported by both the Center for Software Science at
7459 the University of Utah and Cygnus Support.
7461 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7462 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7463 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7464 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7465 and some initial 64-bit support).
7467 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7469 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7470 support for openVMS/Alpha.
7472 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7475 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7476 Inc.@: added support for Xtensa processors.
7478 Several engineers at Cygnus Support have also provided many small bug fixes and
7479 configuration enhancements.
7481 Jon Beniston added support for the Lattice Mico32 architecture.
7483 Many others have contributed large or small bugfixes and enhancements. If
7484 you have contributed significant work and are not mentioned on this list, and
7485 want to be, let us know. Some of the history has been lost; we are not
7486 intentionally leaving anyone out.
7488 @node GNU Free Documentation License
7489 @appendix GNU Free Documentation License
7493 @unnumbered AS Index