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.
817 The following options are available when @value{AS} is configured for the
818 Intel 80960 processor.
821 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
822 Specify which variant of the 960 architecture is the target.
825 Add code to collect statistics about branches taken.
828 Do not alter compare-and-branch instructions for long displacements;
835 The following options are available when @value{AS} is configured for the
841 Specifies that the extended IP2022 instructions are allowed.
844 Restores the default behaviour, which restricts the permitted instructions to
845 just the basic IP2022 ones.
851 The following options are available when @value{AS} is configured for the
852 Renesas M32C and M16C processors.
857 Assemble M32C instructions.
860 Assemble M16C instructions (the default).
863 Enable support for link-time relaxations.
866 Support H'00 style hex constants in addition to 0x00 style.
872 The following options are available when @value{AS} is configured for the
873 Renesas M32R (formerly Mitsubishi M32R) series.
878 Specify which processor in the M32R family is the target. The default
879 is normally the M32R, but this option changes it to the M32RX.
881 @item --warn-explicit-parallel-conflicts or --Wp
882 Produce warning messages when questionable parallel constructs are
885 @item --no-warn-explicit-parallel-conflicts or --Wnp
886 Do not produce warning messages when questionable parallel constructs are
893 The following options are available when @value{AS} is configured for the
894 Motorola 68000 series.
899 Shorten references to undefined symbols, to one word instead of two.
901 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
902 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
903 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
904 Specify what processor in the 68000 family is the target. The default
905 is normally the 68020, but this can be changed at configuration time.
907 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
908 The target machine does (or does not) have a floating-point coprocessor.
909 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
910 the basic 68000 is not compatible with the 68881, a combination of the
911 two can be specified, since it's possible to do emulation of the
912 coprocessor instructions with the main processor.
914 @item -m68851 | -mno-68851
915 The target machine does (or does not) have a memory-management
916 unit coprocessor. The default is to assume an MMU for 68020 and up.
923 For details about the PDP-11 machine dependent features options,
924 see @ref{PDP-11-Options}.
927 @item -mpic | -mno-pic
928 Generate position-independent (or position-dependent) code. The
929 default is @option{-mpic}.
932 @itemx -mall-extensions
933 Enable all instruction set extensions. This is the default.
935 @item -mno-extensions
936 Disable all instruction set extensions.
938 @item -m@var{extension} | -mno-@var{extension}
939 Enable (or disable) a particular instruction set extension.
942 Enable the instruction set extensions supported by a particular CPU, and
943 disable all other extensions.
945 @item -m@var{machine}
946 Enable the instruction set extensions supported by a particular machine
947 model, and disable all other extensions.
953 The following options are available when @value{AS} is configured for
954 a picoJava processor.
958 @cindex PJ endianness
959 @cindex endianness, PJ
960 @cindex big endian output, PJ
962 Generate ``big endian'' format output.
964 @cindex little endian output, PJ
966 Generate ``little endian'' format output.
972 The following options are available when @value{AS} is configured for the
973 Motorola 68HC11 or 68HC12 series.
977 @item -m68hc11 | -m68hc12 | -m68hcs12
978 Specify what processor is the target. The default is
979 defined by the configuration option when building the assembler.
982 Specify to use the 16-bit integer ABI.
985 Specify to use the 32-bit integer ABI.
988 Specify to use the 32-bit double ABI.
991 Specify to use the 64-bit double ABI.
993 @item --force-long-branches
994 Relative branches are turned into absolute ones. This concerns
995 conditional branches, unconditional branches and branches to a
998 @item -S | --short-branches
999 Do not turn relative branches into absolute ones
1000 when the offset is out of range.
1002 @item --strict-direct-mode
1003 Do not turn the direct addressing mode into extended addressing mode
1004 when the instruction does not support direct addressing mode.
1006 @item --print-insn-syntax
1007 Print the syntax of instruction in case of error.
1009 @item --print-opcodes
1010 print the list of instructions with syntax and then exit.
1012 @item --generate-example
1013 print an example of instruction for each possible instruction and then exit.
1014 This option is only useful for testing @command{@value{AS}}.
1020 The following options are available when @command{@value{AS}} is configured
1021 for the SPARC architecture:
1024 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1025 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1026 Explicitly select a variant of the SPARC architecture.
1028 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1029 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1031 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1032 UltraSPARC extensions.
1034 @item -xarch=v8plus | -xarch=v8plusa
1035 For compatibility with the Solaris v9 assembler. These options are
1036 equivalent to -Av8plus and -Av8plusa, respectively.
1039 Warn when the assembler switches to another architecture.
1044 The following options are available when @value{AS} is configured for the 'c54x
1049 Enable extended addressing mode. All addresses and relocations will assume
1050 extended addressing (usually 23 bits).
1051 @item -mcpu=@var{CPU_VERSION}
1052 Sets the CPU version being compiled for.
1053 @item -merrors-to-file @var{FILENAME}
1054 Redirect error output to a file, for broken systems which don't support such
1055 behaviour in the shell.
1060 The following options are available when @value{AS} is configured for
1061 a @sc{mips} processor.
1065 This option sets the largest size of an object that can be referenced
1066 implicitly with the @code{gp} register. It is only accepted for targets that
1067 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1069 @cindex MIPS endianness
1070 @cindex endianness, MIPS
1071 @cindex big endian output, MIPS
1073 Generate ``big endian'' format output.
1075 @cindex little endian output, MIPS
1077 Generate ``little endian'' format output.
1089 Generate code for a particular @sc{mips} Instruction Set Architecture level.
1090 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1091 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1092 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1093 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1095 correspond to generic
1096 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
1097 and @samp{MIPS64 Release 2}
1098 ISA processors, respectively.
1100 @item -march=@var{CPU}
1101 Generate code for a particular @sc{mips} cpu.
1103 @item -mtune=@var{cpu}
1104 Schedule and tune for a particular @sc{mips} cpu.
1108 Cause nops to be inserted if the read of the destination register
1109 of an mfhi or mflo instruction occurs in the following two instructions.
1113 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1114 section instead of the standard ELF .stabs sections.
1118 Control generation of @code{.pdr} sections.
1122 The register sizes are normally inferred from the ISA and ABI, but these
1123 flags force a certain group of registers to be treated as 32 bits wide at
1124 all times. @samp{-mgp32} controls the size of general-purpose registers
1125 and @samp{-mfp32} controls the size of floating-point registers.
1129 Generate code for the MIPS 16 processor. This is equivalent to putting
1130 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1131 turns off this option.
1134 @itemx -mno-smartmips
1135 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1136 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1137 @samp{-mno-smartmips} turns off this option.
1141 Generate code for the MIPS-3D Application Specific Extension.
1142 This tells the assembler to accept MIPS-3D instructions.
1143 @samp{-no-mips3d} turns off this option.
1147 Generate code for the MDMX Application Specific Extension.
1148 This tells the assembler to accept MDMX instructions.
1149 @samp{-no-mdmx} turns off this option.
1153 Generate code for the DSP Release 1 Application Specific Extension.
1154 This tells the assembler to accept DSP Release 1 instructions.
1155 @samp{-mno-dsp} turns off this option.
1159 Generate code for the DSP Release 2 Application Specific Extension.
1160 This option implies -mdsp.
1161 This tells the assembler to accept DSP Release 2 instructions.
1162 @samp{-mno-dspr2} turns off this option.
1166 Generate code for the MT Application Specific Extension.
1167 This tells the assembler to accept MT instructions.
1168 @samp{-mno-mt} turns off this option.
1170 @item --construct-floats
1171 @itemx --no-construct-floats
1172 The @samp{--no-construct-floats} option disables the construction of
1173 double width floating point constants by loading the two halves of the
1174 value into the two single width floating point registers that make up
1175 the double width register. By default @samp{--construct-floats} is
1176 selected, allowing construction of these floating point constants.
1179 @item --emulation=@var{name}
1180 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1181 for some other target, in all respects, including output format (choosing
1182 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1183 debugging information or store symbol table information, and default
1184 endianness. The available configuration names are: @samp{mipsecoff},
1185 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1186 @samp{mipsbelf}. The first two do not alter the default endianness from that
1187 of the primary target for which the assembler was configured; the others change
1188 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1189 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1190 selection in any case.
1192 This option is currently supported only when the primary target
1193 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1194 Furthermore, the primary target or others specified with
1195 @samp{--enable-targets=@dots{}} at configuration time must include support for
1196 the other format, if both are to be available. For example, the Irix 5
1197 configuration includes support for both.
1199 Eventually, this option will support more configurations, with more
1200 fine-grained control over the assembler's behavior, and will be supported for
1204 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1211 Control how to deal with multiplication overflow and division by zero.
1212 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1213 (and only work for Instruction Set Architecture level 2 and higher);
1214 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1218 When this option is used, @command{@value{AS}} will issue a warning every
1219 time it generates a nop instruction from a macro.
1224 The following options are available when @value{AS} is configured for
1230 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1231 The command line option @samp{-nojsri2bsr} can be used to disable it.
1235 Enable or disable the silicon filter behaviour. By default this is disabled.
1236 The default can be overridden by the @samp{-sifilter} command line option.
1239 Alter jump instructions for long displacements.
1241 @item -mcpu=[210|340]
1242 Select the cpu type on the target hardware. This controls which instructions
1246 Assemble for a big endian target.
1249 Assemble for a little endian target.
1255 See the info pages for documentation of the MMIX-specific options.
1259 See the info pages for documentation of the RX-specific options.
1263 The following options are available when @value{AS} is configured for the s390
1269 Select the word size, either 31/32 bits or 64 bits.
1272 Select the architecture mode, either the Enterprise System
1273 Architecture (esa) or the z/Architecture mode (zarch).
1274 @item -march=@var{processor}
1275 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1276 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, or @samp{z10}.
1278 @itemx -mno-regnames
1279 Allow or disallow symbolic names for registers.
1280 @item -mwarn-areg-zero
1281 Warn whenever the operand for a base or index register has been specified
1282 but evaluates to zero.
1290 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1291 for a TMS320C6000 processor.
1295 @c man begin OPTIONS
1296 The following options are available when @value{AS} is configured for a
1297 TMS320C6000 processor.
1299 @c man begin INCLUDE
1300 @include c-tic6x.texi
1301 @c ended inside the included file
1306 @c man begin OPTIONS
1308 The following options are available when @value{AS} is configured for
1309 an Xtensa processor.
1312 @item --text-section-literals | --no-text-section-literals
1313 With @option{--text-@-section-@-literals}, literal pools are interspersed
1314 in the text section. The default is
1315 @option{--no-@-text-@-section-@-literals}, which places literals in a
1316 separate section in the output file. These options only affect literals
1317 referenced via PC-relative @code{L32R} instructions; literals for
1318 absolute mode @code{L32R} instructions are handled separately.
1320 @item --absolute-literals | --no-absolute-literals
1321 Indicate to the assembler whether @code{L32R} instructions use absolute
1322 or PC-relative addressing. The default is to assume absolute addressing
1323 if the Xtensa processor includes the absolute @code{L32R} addressing
1324 option. Otherwise, only the PC-relative @code{L32R} mode can be used.
1326 @item --target-align | --no-target-align
1327 Enable or disable automatic alignment to reduce branch penalties at the
1328 expense of some code density. The default is @option{--target-@-align}.
1330 @item --longcalls | --no-longcalls
1331 Enable or disable transformation of call instructions to allow calls
1332 across a greater range of addresses. The default is
1333 @option{--no-@-longcalls}.
1335 @item --transform | --no-transform
1336 Enable or disable all assembler transformations of Xtensa instructions.
1337 The default is @option{--transform};
1338 @option{--no-transform} should be used only in the rare cases when the
1339 instructions must be exactly as specified in the assembly source.
1341 @item --rename-section @var{oldname}=@var{newname}
1342 When generating output sections, rename the @var{oldname} section to
1348 The following options are available when @value{AS} is configured for
1349 a Z80 family processor.
1352 Assemble for Z80 processor.
1354 Assemble for R800 processor.
1355 @item -ignore-undocumented-instructions
1357 Assemble undocumented Z80 instructions that also work on R800 without warning.
1358 @item -ignore-unportable-instructions
1360 Assemble all undocumented Z80 instructions without warning.
1361 @item -warn-undocumented-instructions
1363 Issue a warning for undocumented Z80 instructions that also work on R800.
1364 @item -warn-unportable-instructions
1366 Issue a warning for undocumented Z80 instructions that do not work on R800.
1367 @item -forbid-undocumented-instructions
1369 Treat all undocumented instructions as errors.
1370 @item -forbid-unportable-instructions
1372 Treat undocumented Z80 instructions that do not work on R800 as errors.
1379 * Manual:: Structure of this Manual
1380 * GNU Assembler:: The GNU Assembler
1381 * Object Formats:: Object File Formats
1382 * Command Line:: Command Line
1383 * Input Files:: Input Files
1384 * Object:: Output (Object) File
1385 * Errors:: Error and Warning Messages
1389 @section Structure of this Manual
1391 @cindex manual, structure and purpose
1392 This manual is intended to describe what you need to know to use
1393 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1394 notation for symbols, constants, and expressions; the directives that
1395 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1398 We also cover special features in the @value{TARGET}
1399 configuration of @command{@value{AS}}, including assembler directives.
1402 This manual also describes some of the machine-dependent features of
1403 various flavors of the assembler.
1406 @cindex machine instructions (not covered)
1407 On the other hand, this manual is @emph{not} intended as an introduction
1408 to programming in assembly language---let alone programming in general!
1409 In a similar vein, we make no attempt to introduce the machine
1410 architecture; we do @emph{not} describe the instruction set, standard
1411 mnemonics, registers or addressing modes that are standard to a
1412 particular architecture.
1414 You may want to consult the manufacturer's
1415 machine architecture manual for this information.
1419 For information on the H8/300 machine instruction set, see @cite{H8/300
1420 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1421 Programming Manual} (Renesas).
1424 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1425 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1426 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1427 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1430 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1436 Throughout this manual, we assume that you are running @dfn{GNU},
1437 the portable operating system from the @dfn{Free Software
1438 Foundation, Inc.}. This restricts our attention to certain kinds of
1439 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1440 once this assumption is granted examples and definitions need less
1443 @command{@value{AS}} is part of a team of programs that turn a high-level
1444 human-readable series of instructions into a low-level
1445 computer-readable series of instructions. Different versions of
1446 @command{@value{AS}} are used for different kinds of computer.
1449 @c There used to be a section "Terminology" here, which defined
1450 @c "contents", "byte", "word", and "long". Defining "word" to any
1451 @c particular size is confusing when the .word directive may generate 16
1452 @c bits on one machine and 32 bits on another; in general, for the user
1453 @c version of this manual, none of these terms seem essential to define.
1454 @c They were used very little even in the former draft of the manual;
1455 @c this draft makes an effort to avoid them (except in names of
1459 @section The GNU Assembler
1461 @c man begin DESCRIPTION
1463 @sc{gnu} @command{as} is really a family of assemblers.
1465 This manual describes @command{@value{AS}}, a member of that family which is
1466 configured for the @value{TARGET} architectures.
1468 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1469 should find a fairly similar environment when you use it on another
1470 architecture. Each version has much in common with the others,
1471 including object file formats, most assembler directives (often called
1472 @dfn{pseudo-ops}) and assembler syntax.@refill
1474 @cindex purpose of @sc{gnu} assembler
1475 @command{@value{AS}} is primarily intended to assemble the output of the
1476 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1477 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1478 assemble correctly everything that other assemblers for the same
1479 machine would assemble.
1481 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1484 @c This remark should appear in generic version of manual; assumption
1485 @c here is that generic version sets M680x0.
1486 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1487 assembler for the same architecture; for example, we know of several
1488 incompatible versions of 680x0 assembly language syntax.
1493 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1494 program in one pass of the source file. This has a subtle impact on the
1495 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1497 @node Object Formats
1498 @section Object File Formats
1500 @cindex object file format
1501 The @sc{gnu} assembler can be configured to produce several alternative
1502 object file formats. For the most part, this does not affect how you
1503 write assembly language programs; but directives for debugging symbols
1504 are typically different in different file formats. @xref{Symbol
1505 Attributes,,Symbol Attributes}.
1508 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1509 @value{OBJ-NAME} format object files.
1511 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1513 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1514 @code{b.out} or COFF format object files.
1517 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1518 SOM or ELF format object files.
1523 @section Command Line
1525 @cindex command line conventions
1527 After the program name @command{@value{AS}}, the command line may contain
1528 options and file names. Options may appear in any order, and may be
1529 before, after, or between file names. The order of file names is
1532 @cindex standard input, as input file
1534 @file{--} (two hyphens) by itself names the standard input file
1535 explicitly, as one of the files for @command{@value{AS}} to assemble.
1537 @cindex options, command line
1538 Except for @samp{--} any command line argument that begins with a
1539 hyphen (@samp{-}) is an option. Each option changes the behavior of
1540 @command{@value{AS}}. No option changes the way another option works. An
1541 option is a @samp{-} followed by one or more letters; the case of
1542 the letter is important. All options are optional.
1544 Some options expect exactly one file name to follow them. The file
1545 name may either immediately follow the option's letter (compatible
1546 with older assemblers) or it may be the next command argument (@sc{gnu}
1547 standard). These two command lines are equivalent:
1550 @value{AS} -o my-object-file.o mumble.s
1551 @value{AS} -omy-object-file.o mumble.s
1555 @section Input Files
1558 @cindex source program
1559 @cindex files, input
1560 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1561 describe the program input to one run of @command{@value{AS}}. The program may
1562 be in one or more files; how the source is partitioned into files
1563 doesn't change the meaning of the source.
1565 @c I added "con" prefix to "catenation" just to prove I can overcome my
1567 The source program is a concatenation of the text in all the files, in the
1570 @c man begin DESCRIPTION
1571 Each time you run @command{@value{AS}} it assembles exactly one source
1572 program. The source program is made up of one or more files.
1573 (The standard input is also a file.)
1575 You give @command{@value{AS}} a command line that has zero or more input file
1576 names. The input files are read (from left file name to right). A
1577 command line argument (in any position) that has no special meaning
1578 is taken to be an input file name.
1580 If you give @command{@value{AS}} no file names it attempts to read one input file
1581 from the @command{@value{AS}} standard input, which is normally your terminal. You
1582 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1585 Use @samp{--} if you need to explicitly name the standard input file
1586 in your command line.
1588 If the source is empty, @command{@value{AS}} produces a small, empty object
1593 @subheading Filenames and Line-numbers
1595 @cindex input file linenumbers
1596 @cindex line numbers, in input files
1597 There are two ways of locating a line in the input file (or files) and
1598 either may be used in reporting error messages. One way refers to a line
1599 number in a physical file; the other refers to a line number in a
1600 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1602 @dfn{Physical files} are those files named in the command line given
1603 to @command{@value{AS}}.
1605 @dfn{Logical files} are simply names declared explicitly by assembler
1606 directives; they bear no relation to physical files. Logical file names help
1607 error messages reflect the original source file, when @command{@value{AS}} source
1608 is itself synthesized from other files. @command{@value{AS}} understands the
1609 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1610 @ref{File,,@code{.file}}.
1613 @section Output (Object) File
1619 Every time you run @command{@value{AS}} it produces an output file, which is
1620 your assembly language program translated into numbers. This file
1621 is the object file. Its default name is
1629 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1631 You can give it another name by using the @option{-o} option. Conventionally,
1632 object file names end with @file{.o}. The default name is used for historical
1633 reasons: older assemblers were capable of assembling self-contained programs
1634 directly into a runnable program. (For some formats, this isn't currently
1635 possible, but it can be done for the @code{a.out} format.)
1639 The object file is meant for input to the linker @code{@value{LD}}. It contains
1640 assembled program code, information to help @code{@value{LD}} integrate
1641 the assembled program into a runnable file, and (optionally) symbolic
1642 information for the debugger.
1644 @c link above to some info file(s) like the description of a.out.
1645 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1648 @section Error and Warning Messages
1650 @c man begin DESCRIPTION
1652 @cindex error messages
1653 @cindex warning messages
1654 @cindex messages from assembler
1655 @command{@value{AS}} may write warnings and error messages to the standard error
1656 file (usually your terminal). This should not happen when a compiler
1657 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1658 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1659 grave problem that stops the assembly.
1663 @cindex format of warning messages
1664 Warning messages have the format
1667 file_name:@b{NNN}:Warning Message Text
1671 @cindex line numbers, in warnings/errors
1672 (where @b{NNN} is a line number). If a logical file name has been given
1673 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1674 the current input file is used. If a logical line number was given
1676 (@pxref{Line,,@code{.line}})
1678 then it is used to calculate the number printed,
1679 otherwise the actual line in the current source file is printed. The
1680 message text is intended to be self explanatory (in the grand Unix
1683 @cindex format of error messages
1684 Error messages have the format
1686 file_name:@b{NNN}:FATAL:Error Message Text
1688 The file name and line number are derived as for warning
1689 messages. The actual message text may be rather less explanatory
1690 because many of them aren't supposed to happen.
1693 @chapter Command-Line Options
1695 @cindex options, all versions of assembler
1696 This chapter describes command-line options available in @emph{all}
1697 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1698 for options specific
1700 to the @value{TARGET} target.
1703 to particular machine architectures.
1706 @c man begin DESCRIPTION
1708 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1709 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1710 The assembler arguments must be separated from each other (and the @samp{-Wa})
1711 by commas. For example:
1714 gcc -c -g -O -Wa,-alh,-L file.c
1718 This passes two options to the assembler: @samp{-alh} (emit a listing to
1719 standard output with high-level and assembly source) and @samp{-L} (retain
1720 local symbols in the symbol table).
1722 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1723 command-line options are automatically passed to the assembler by the compiler.
1724 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1725 precisely what options it passes to each compilation pass, including the
1731 * a:: -a[cdghlns] enable listings
1732 * alternate:: --alternate enable alternate macro syntax
1733 * D:: -D for compatibility
1734 * f:: -f to work faster
1735 * I:: -I for .include search path
1736 @ifclear DIFF-TBL-KLUGE
1737 * K:: -K for compatibility
1739 @ifset DIFF-TBL-KLUGE
1740 * K:: -K for difference tables
1743 * L:: -L to retain local symbols
1744 * listing:: --listing-XXX to configure listing output
1745 * M:: -M or --mri to assemble in MRI compatibility mode
1746 * MD:: --MD for dependency tracking
1747 * o:: -o to name the object file
1748 * R:: -R to join data and text sections
1749 * statistics:: --statistics to see statistics about assembly
1750 * traditional-format:: --traditional-format for compatible output
1751 * v:: -v to announce version
1752 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1753 * Z:: -Z to make object file even after errors
1757 @section Enable Listings: @option{-a[cdghlns]}
1767 @cindex listings, enabling
1768 @cindex assembly listings, enabling
1770 These options enable listing output from the assembler. By itself,
1771 @samp{-a} requests high-level, assembly, and symbols listing.
1772 You can use other letters to select specific options for the list:
1773 @samp{-ah} requests a high-level language listing,
1774 @samp{-al} requests an output-program assembly listing, and
1775 @samp{-as} requests a symbol table listing.
1776 High-level listings require that a compiler debugging option like
1777 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1780 Use the @samp{-ag} option to print a first section with general assembly
1781 information, like @value{AS} version, switches passed, or time stamp.
1783 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1784 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1785 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1786 omitted from the listing.
1788 Use the @samp{-ad} option to omit debugging directives from the
1791 Once you have specified one of these options, you can further control
1792 listing output and its appearance using the directives @code{.list},
1793 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1795 The @samp{-an} option turns off all forms processing.
1796 If you do not request listing output with one of the @samp{-a} options, the
1797 listing-control directives have no effect.
1799 The letters after @samp{-a} may be combined into one option,
1800 @emph{e.g.}, @samp{-aln}.
1802 Note if the assembler source is coming from the standard input (e.g.,
1804 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1805 is being used) then the listing will not contain any comments or preprocessor
1806 directives. This is because the listing code buffers input source lines from
1807 stdin only after they have been preprocessed by the assembler. This reduces
1808 memory usage and makes the code more efficient.
1811 @section @option{--alternate}
1814 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
1817 @section @option{-D}
1820 This option has no effect whatsoever, but it is accepted to make it more
1821 likely that scripts written for other assemblers also work with
1822 @command{@value{AS}}.
1825 @section Work Faster: @option{-f}
1828 @cindex trusted compiler
1829 @cindex faster processing (@option{-f})
1830 @samp{-f} should only be used when assembling programs written by a
1831 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
1832 and comment preprocessing on
1833 the input file(s) before assembling them. @xref{Preprocessing,
1837 @emph{Warning:} if you use @samp{-f} when the files actually need to be
1838 preprocessed (if they contain comments, for example), @command{@value{AS}} does
1843 @section @code{.include} Search Path: @option{-I} @var{path}
1845 @kindex -I @var{path}
1846 @cindex paths for @code{.include}
1847 @cindex search path for @code{.include}
1848 @cindex @code{include} directive search path
1849 Use this option to add a @var{path} to the list of directories
1850 @command{@value{AS}} searches for files specified in @code{.include}
1851 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
1852 many times as necessary to include a variety of paths. The current
1853 working directory is always searched first; after that, @command{@value{AS}}
1854 searches any @samp{-I} directories in the same order as they were
1855 specified (left to right) on the command line.
1858 @section Difference Tables: @option{-K}
1861 @ifclear DIFF-TBL-KLUGE
1862 On the @value{TARGET} family, this option is allowed, but has no effect. It is
1863 permitted for compatibility with the @sc{gnu} assembler on other platforms,
1864 where it can be used to warn when the assembler alters the machine code
1865 generated for @samp{.word} directives in difference tables. The @value{TARGET}
1866 family does not have the addressing limitations that sometimes lead to this
1867 alteration on other platforms.
1870 @ifset DIFF-TBL-KLUGE
1871 @cindex difference tables, warning
1872 @cindex warning for altered difference tables
1873 @command{@value{AS}} sometimes alters the code emitted for directives of the
1874 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
1875 You can use the @samp{-K} option if you want a warning issued when this
1880 @section Include Local Symbols: @option{-L}
1883 @cindex local symbols, retaining in output
1884 Symbols beginning with system-specific local label prefixes, typically
1885 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
1886 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
1887 such symbols when debugging, because they are intended for the use of
1888 programs (like compilers) that compose assembler programs, not for your
1889 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
1890 such symbols, so you do not normally debug with them.
1892 This option tells @command{@value{AS}} to retain those local symbols
1893 in the object file. Usually if you do this you also tell the linker
1894 @code{@value{LD}} to preserve those symbols.
1897 @section Configuring listing output: @option{--listing}
1899 The listing feature of the assembler can be enabled via the command line switch
1900 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
1901 hex dump of the corresponding locations in the output object file, and displays
1902 them as a listing file. The format of this listing can be controlled by
1903 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
1904 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
1905 @code{.psize} (@pxref{Psize}), and
1906 @code{.eject} (@pxref{Eject}) and also by the following switches:
1909 @item --listing-lhs-width=@samp{number}
1910 @kindex --listing-lhs-width
1911 @cindex Width of first line disassembly output
1912 Sets the maximum width, in words, of the first line of the hex byte dump. This
1913 dump appears on the left hand side of the listing output.
1915 @item --listing-lhs-width2=@samp{number}
1916 @kindex --listing-lhs-width2
1917 @cindex Width of continuation lines of disassembly output
1918 Sets the maximum width, in words, of any further lines of the hex byte dump for
1919 a given input source line. If this value is not specified, it defaults to being
1920 the same as the value specified for @samp{--listing-lhs-width}. If neither
1921 switch is used the default is to one.
1923 @item --listing-rhs-width=@samp{number}
1924 @kindex --listing-rhs-width
1925 @cindex Width of source line output
1926 Sets the maximum width, in characters, of the source line that is displayed
1927 alongside the hex dump. The default value for this parameter is 100. The
1928 source line is displayed on the right hand side of the listing output.
1930 @item --listing-cont-lines=@samp{number}
1931 @kindex --listing-cont-lines
1932 @cindex Maximum number of continuation lines
1933 Sets the maximum number of continuation lines of hex dump that will be
1934 displayed for a given single line of source input. The default value is 4.
1938 @section Assemble in MRI Compatibility Mode: @option{-M}
1941 @cindex MRI compatibility mode
1942 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
1943 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
1944 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
1945 configured target) assembler from Microtec Research. The exact nature of the
1946 MRI syntax will not be documented here; see the MRI manuals for more
1947 information. Note in particular that the handling of macros and macro
1948 arguments is somewhat different. The purpose of this option is to permit
1949 assembling existing MRI assembler code using @command{@value{AS}}.
1951 The MRI compatibility is not complete. Certain operations of the MRI assembler
1952 depend upon its object file format, and can not be supported using other object
1953 file formats. Supporting these would require enhancing each object file format
1954 individually. These are:
1957 @item global symbols in common section
1959 The m68k MRI assembler supports common sections which are merged by the linker.
1960 Other object file formats do not support this. @command{@value{AS}} handles
1961 common sections by treating them as a single common symbol. It permits local
1962 symbols to be defined within a common section, but it can not support global
1963 symbols, since it has no way to describe them.
1965 @item complex relocations
1967 The MRI assemblers support relocations against a negated section address, and
1968 relocations which combine the start addresses of two or more sections. These
1969 are not support by other object file formats.
1971 @item @code{END} pseudo-op specifying start address
1973 The MRI @code{END} pseudo-op permits the specification of a start address.
1974 This is not supported by other object file formats. The start address may
1975 instead be specified using the @option{-e} option to the linker, or in a linker
1978 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
1980 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
1981 name to the output file. This is not supported by other object file formats.
1983 @item @code{ORG} pseudo-op
1985 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
1986 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
1987 which changes the location within the current section. Absolute sections are
1988 not supported by other object file formats. The address of a section may be
1989 assigned within a linker script.
1992 There are some other features of the MRI assembler which are not supported by
1993 @command{@value{AS}}, typically either because they are difficult or because they
1994 seem of little consequence. Some of these may be supported in future releases.
1998 @item EBCDIC strings
2000 EBCDIC strings are not supported.
2002 @item packed binary coded decimal
2004 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2005 and @code{DCB.P} pseudo-ops are not supported.
2007 @item @code{FEQU} pseudo-op
2009 The m68k @code{FEQU} pseudo-op is not supported.
2011 @item @code{NOOBJ} pseudo-op
2013 The m68k @code{NOOBJ} pseudo-op is not supported.
2015 @item @code{OPT} branch control options
2017 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2018 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2019 relaxes all branches, whether forward or backward, to an appropriate size, so
2020 these options serve no purpose.
2022 @item @code{OPT} list control options
2024 The following m68k @code{OPT} list control options are ignored: @code{C},
2025 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2026 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2028 @item other @code{OPT} options
2030 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2031 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2033 @item @code{OPT} @code{D} option is default
2035 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2036 @code{OPT NOD} may be used to turn it off.
2038 @item @code{XREF} pseudo-op.
2040 The m68k @code{XREF} pseudo-op is ignored.
2042 @item @code{.debug} pseudo-op
2044 The i960 @code{.debug} pseudo-op is not supported.
2046 @item @code{.extended} pseudo-op
2048 The i960 @code{.extended} pseudo-op is not supported.
2050 @item @code{.list} pseudo-op.
2052 The various options of the i960 @code{.list} pseudo-op are not supported.
2054 @item @code{.optimize} pseudo-op
2056 The i960 @code{.optimize} pseudo-op is not supported.
2058 @item @code{.output} pseudo-op
2060 The i960 @code{.output} pseudo-op is not supported.
2062 @item @code{.setreal} pseudo-op
2064 The i960 @code{.setreal} pseudo-op is not supported.
2069 @section Dependency Tracking: @option{--MD}
2072 @cindex dependency tracking
2075 @command{@value{AS}} can generate a dependency file for the file it creates. This
2076 file consists of a single rule suitable for @code{make} describing the
2077 dependencies of the main source file.
2079 The rule is written to the file named in its argument.
2081 This feature is used in the automatic updating of makefiles.
2084 @section Name the Object File: @option{-o}
2087 @cindex naming object file
2088 @cindex object file name
2089 There is always one object file output when you run @command{@value{AS}}. By
2090 default it has the name
2093 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2107 You use this option (which takes exactly one filename) to give the
2108 object file a different name.
2110 Whatever the object file is called, @command{@value{AS}} overwrites any
2111 existing file of the same name.
2114 @section Join Data and Text Sections: @option{-R}
2117 @cindex data and text sections, joining
2118 @cindex text and data sections, joining
2119 @cindex joining text and data sections
2120 @cindex merging text and data sections
2121 @option{-R} tells @command{@value{AS}} to write the object file as if all
2122 data-section data lives in the text section. This is only done at
2123 the very last moment: your binary data are the same, but data
2124 section parts are relocated differently. The data section part of
2125 your object file is zero bytes long because all its bytes are
2126 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2128 When you specify @option{-R} it would be possible to generate shorter
2129 address displacements (because we do not have to cross between text and
2130 data section). We refrain from doing this simply for compatibility with
2131 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2134 When @command{@value{AS}} is configured for COFF or ELF output,
2135 this option is only useful if you use sections named @samp{.text} and
2140 @option{-R} is not supported for any of the HPPA targets. Using
2141 @option{-R} generates a warning from @command{@value{AS}}.
2145 @section Display Assembly Statistics: @option{--statistics}
2147 @kindex --statistics
2148 @cindex statistics, about assembly
2149 @cindex time, total for assembly
2150 @cindex space used, maximum for assembly
2151 Use @samp{--statistics} to display two statistics about the resources used by
2152 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2153 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2156 @node traditional-format
2157 @section Compatible Output: @option{--traditional-format}
2159 @kindex --traditional-format
2160 For some targets, the output of @command{@value{AS}} is different in some ways
2161 from the output of some existing assembler. This switch requests
2162 @command{@value{AS}} to use the traditional format instead.
2164 For example, it disables the exception frame optimizations which
2165 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2168 @section Announce Version: @option{-v}
2172 @cindex assembler version
2173 @cindex version of assembler
2174 You can find out what version of as is running by including the
2175 option @samp{-v} (which you can also spell as @samp{-version}) on the
2179 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2181 @command{@value{AS}} should never give a warning or error message when
2182 assembling compiler output. But programs written by people often
2183 cause @command{@value{AS}} to give a warning that a particular assumption was
2184 made. All such warnings are directed to the standard error file.
2188 @cindex suppressing warnings
2189 @cindex warnings, suppressing
2190 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2191 This only affects the warning messages: it does not change any particular of
2192 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2195 @kindex --fatal-warnings
2196 @cindex errors, caused by warnings
2197 @cindex warnings, causing error
2198 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2199 files that generate warnings to be in error.
2202 @cindex warnings, switching on
2203 You can switch these options off again by specifying @option{--warn}, which
2204 causes warnings to be output as usual.
2207 @section Generate Object File in Spite of Errors: @option{-Z}
2208 @cindex object file, after errors
2209 @cindex errors, continuing after
2210 After an error message, @command{@value{AS}} normally produces no output. If for
2211 some reason you are interested in object file output even after
2212 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2213 option. If there are any errors, @command{@value{AS}} continues anyways, and
2214 writes an object file after a final warning message of the form @samp{@var{n}
2215 errors, @var{m} warnings, generating bad object file.}
2220 @cindex machine-independent syntax
2221 @cindex syntax, machine-independent
2222 This chapter describes the machine-independent syntax allowed in a
2223 source file. @command{@value{AS}} syntax is similar to what many other
2224 assemblers use; it is inspired by the BSD 4.2
2229 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2233 * Preprocessing:: Preprocessing
2234 * Whitespace:: Whitespace
2235 * Comments:: Comments
2236 * Symbol Intro:: Symbols
2237 * Statements:: Statements
2238 * Constants:: Constants
2242 @section Preprocessing
2244 @cindex preprocessing
2245 The @command{@value{AS}} internal preprocessor:
2247 @cindex whitespace, removed by preprocessor
2249 adjusts and removes extra whitespace. It leaves one space or tab before
2250 the keywords on a line, and turns any other whitespace on the line into
2253 @cindex comments, removed by preprocessor
2255 removes all comments, replacing them with a single space, or an
2256 appropriate number of newlines.
2258 @cindex constants, converted by preprocessor
2260 converts character constants into the appropriate numeric values.
2263 It does not do macro processing, include file handling, or
2264 anything else you may get from your C compiler's preprocessor. You can
2265 do include file processing with the @code{.include} directive
2266 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2267 to get other ``CPP'' style preprocessing by giving the input file a
2268 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2269 Output, gcc.info, Using GNU CC}.
2271 Excess whitespace, comments, and character constants
2272 cannot be used in the portions of the input text that are not
2275 @cindex turning preprocessing on and off
2276 @cindex preprocessing, turning on and off
2279 If the first line of an input file is @code{#NO_APP} or if you use the
2280 @samp{-f} option, whitespace and comments are not removed from the input file.
2281 Within an input file, you can ask for whitespace and comment removal in
2282 specific portions of the by putting a line that says @code{#APP} before the
2283 text that may contain whitespace or comments, and putting a line that says
2284 @code{#NO_APP} after this text. This feature is mainly intend to support
2285 @code{asm} statements in compilers whose output is otherwise free of comments
2292 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2293 Whitespace is used to separate symbols, and to make programs neater for
2294 people to read. Unless within character constants
2295 (@pxref{Characters,,Character Constants}), any whitespace means the same
2296 as exactly one space.
2302 There are two ways of rendering comments to @command{@value{AS}}. In both
2303 cases the comment is equivalent to one space.
2305 Anything from @samp{/*} through the next @samp{*/} is a comment.
2306 This means you may not nest these comments.
2310 The only way to include a newline ('\n') in a comment
2311 is to use this sort of comment.
2314 /* This sort of comment does not nest. */
2317 @cindex line comment character
2318 Anything from the @dfn{line comment} character to the next newline
2319 is considered a comment and is ignored. The line comment character is
2321 @samp{;} on the ARC;
2324 @samp{@@} on the ARM;
2327 @samp{;} for the H8/300 family;
2330 @samp{;} for the HPPA;
2333 @samp{#} on the i386 and x86-64;
2336 @samp{#} on the i960;
2339 @samp{;} for the PDP-11;
2342 @samp{;} for picoJava;
2345 @samp{#} for Motorola PowerPC;
2348 @samp{#} for IBM S/390;
2351 @samp{#} for the Sunplus SCORE;
2354 @samp{!} for the Renesas / SuperH SH;
2357 @samp{!} on the SPARC;
2360 @samp{#} on the ip2k;
2363 @samp{#} on the m32c;
2366 @samp{#} on the m32r;
2369 @samp{|} on the 680x0;
2372 @samp{#} on the 68HC11 and 68HC12;
2378 @samp{;} on the TMS320C6X;
2381 @samp{#} on the Vax;
2384 @samp{;} for the Z80;
2387 @samp{!} for the Z8000;
2390 @samp{#} on the V850;
2393 @samp{#} for Xtensa systems;
2395 see @ref{Machine Dependencies}. @refill
2396 @c FIXME What about i860?
2399 On some machines there are two different line comment characters. One
2400 character only begins a comment if it is the first non-whitespace character on
2401 a line, while the other always begins a comment.
2405 The V850 assembler also supports a double dash as starting a comment that
2406 extends to the end of the line.
2412 @cindex lines starting with @code{#}
2413 @cindex logical line numbers
2414 To be compatible with past assemblers, lines that begin with @samp{#} have a
2415 special interpretation. Following the @samp{#} should be an absolute
2416 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2417 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2418 new logical file name. The rest of the line, if any, should be whitespace.
2420 If the first non-whitespace characters on the line are not numeric,
2421 the line is ignored. (Just like a comment.)
2424 # This is an ordinary comment.
2425 # 42-6 "new_file_name" # New logical file name
2426 # This is logical line # 36.
2428 This feature is deprecated, and may disappear from future versions
2429 of @command{@value{AS}}.
2434 @cindex characters used in symbols
2435 @ifclear SPECIAL-SYMS
2436 A @dfn{symbol} is one or more characters chosen from the set of all
2437 letters (both upper and lower case), digits and the three characters
2443 A @dfn{symbol} is one or more characters chosen from the set of all
2444 letters (both upper and lower case), digits and the three characters
2445 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2451 On most machines, you can also use @code{$} in symbol names; exceptions
2452 are noted in @ref{Machine Dependencies}.
2454 No symbol may begin with a digit. Case is significant.
2455 There is no length limit: all characters are significant. Symbols are
2456 delimited by characters not in that set, or by the beginning of a file
2457 (since the source program must end with a newline, the end of a file is
2458 not a possible symbol delimiter). @xref{Symbols}.
2459 @cindex length of symbols
2464 @cindex statements, structure of
2465 @cindex line separator character
2466 @cindex statement separator character
2468 @ifclear abnormal-separator
2469 A @dfn{statement} ends at a newline character (@samp{\n}) or at a
2470 semicolon (@samp{;}). The newline or semicolon is considered part of
2471 the preceding statement. Newlines and semicolons within character
2472 constants are an exception: they do not end statements.
2474 @ifset abnormal-separator
2476 A @dfn{statement} ends at a newline character (@samp{\n}) or an exclamation
2477 point (@samp{!}). The newline or exclamation point is considered part of the
2478 preceding statement. Newlines and exclamation points within character
2479 constants are an exception: they do not end statements.
2482 A @dfn{statement} ends at a newline character (@samp{\n}); or (for the
2483 H8/300) a dollar sign (@samp{$}); or (for the Renesas-SH) a semicolon
2484 (@samp{;}). The newline or separator character is considered part of
2485 the preceding statement. Newlines and separators within character
2486 constants are an exception: they do not end statements.
2491 A @dfn{statement} ends at a newline character (@samp{\n}) or line
2492 separator character. (The line separator is usually @samp{;}, unless this
2493 conflicts with the comment character; see @ref{Machine Dependencies}.) The
2494 newline or separator character is considered part of the preceding
2495 statement. Newlines and separators within character constants are an
2496 exception: they do not end statements.
2499 @cindex newline, required at file end
2500 @cindex EOF, newline must precede
2501 It is an error to end any statement with end-of-file: the last
2502 character of any input file should be a newline.@refill
2504 An empty statement is allowed, and may include whitespace. It is ignored.
2506 @cindex instructions and directives
2507 @cindex directives and instructions
2508 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2511 A statement begins with zero or more labels, optionally followed by a
2512 key symbol which determines what kind of statement it is. The key
2513 symbol determines the syntax of the rest of the statement. If the
2514 symbol begins with a dot @samp{.} then the statement is an assembler
2515 directive: typically valid for any computer. If the symbol begins with
2516 a letter the statement is an assembly language @dfn{instruction}: it
2517 assembles into a machine language instruction.
2519 Different versions of @command{@value{AS}} for different computers
2520 recognize different instructions. In fact, the same symbol may
2521 represent a different instruction in a different computer's assembly
2525 @cindex @code{:} (label)
2526 @cindex label (@code{:})
2527 A label is a symbol immediately followed by a colon (@code{:}).
2528 Whitespace before a label or after a colon is permitted, but you may not
2529 have whitespace between a label's symbol and its colon. @xref{Labels}.
2532 For HPPA targets, labels need not be immediately followed by a colon, but
2533 the definition of a label must begin in column zero. This also implies that
2534 only one label may be defined on each line.
2538 label: .directive followed by something
2539 another_label: # This is an empty statement.
2540 instruction operand_1, operand_2, @dots{}
2547 A constant is a number, written so that its value is known by
2548 inspection, without knowing any context. Like this:
2551 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2552 .ascii "Ring the bell\7" # A string constant.
2553 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2554 .float 0f-314159265358979323846264338327\
2555 95028841971.693993751E-40 # - pi, a flonum.
2560 * Characters:: Character Constants
2561 * Numbers:: Number Constants
2565 @subsection Character Constants
2567 @cindex character constants
2568 @cindex constants, character
2569 There are two kinds of character constants. A @dfn{character} stands
2570 for one character in one byte and its value may be used in
2571 numeric expressions. String constants (properly called string
2572 @emph{literals}) are potentially many bytes and their values may not be
2573 used in arithmetic expressions.
2577 * Chars:: Characters
2581 @subsubsection Strings
2583 @cindex string constants
2584 @cindex constants, string
2585 A @dfn{string} is written between double-quotes. It may contain
2586 double-quotes or null characters. The way to get special characters
2587 into a string is to @dfn{escape} these characters: precede them with
2588 a backslash @samp{\} character. For example @samp{\\} represents
2589 one backslash: the first @code{\} is an escape which tells
2590 @command{@value{AS}} to interpret the second character literally as a backslash
2591 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2592 escape character). The complete list of escapes follows.
2594 @cindex escape codes, character
2595 @cindex character escape codes
2598 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2600 @cindex @code{\b} (backspace character)
2601 @cindex backspace (@code{\b})
2603 Mnemonic for backspace; for ASCII this is octal code 010.
2606 @c Mnemonic for EOText; for ASCII this is octal code 004.
2608 @cindex @code{\f} (formfeed character)
2609 @cindex formfeed (@code{\f})
2611 Mnemonic for FormFeed; for ASCII this is octal code 014.
2613 @cindex @code{\n} (newline character)
2614 @cindex newline (@code{\n})
2616 Mnemonic for newline; for ASCII this is octal code 012.
2619 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2621 @cindex @code{\r} (carriage return character)
2622 @cindex carriage return (@code{\r})
2624 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2627 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2628 @c other assemblers.
2630 @cindex @code{\t} (tab)
2631 @cindex tab (@code{\t})
2633 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2636 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2637 @c @item \x @var{digit} @var{digit} @var{digit}
2638 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2640 @cindex @code{\@var{ddd}} (octal character code)
2641 @cindex octal character code (@code{\@var{ddd}})
2642 @item \ @var{digit} @var{digit} @var{digit}
2643 An octal character code. The numeric code is 3 octal digits.
2644 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2645 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2647 @cindex @code{\@var{xd...}} (hex character code)
2648 @cindex hex character code (@code{\@var{xd...}})
2649 @item \@code{x} @var{hex-digits...}
2650 A hex character code. All trailing hex digits are combined. Either upper or
2651 lower case @code{x} works.
2653 @cindex @code{\\} (@samp{\} character)
2654 @cindex backslash (@code{\\})
2656 Represents one @samp{\} character.
2659 @c Represents one @samp{'} (accent acute) character.
2660 @c This is needed in single character literals
2661 @c (@xref{Characters,,Character Constants}.) to represent
2664 @cindex @code{\"} (doublequote character)
2665 @cindex doublequote (@code{\"})
2667 Represents one @samp{"} character. Needed in strings to represent
2668 this character, because an unescaped @samp{"} would end the string.
2670 @item \ @var{anything-else}
2671 Any other character when escaped by @kbd{\} gives a warning, but
2672 assembles as if the @samp{\} was not present. The idea is that if
2673 you used an escape sequence you clearly didn't want the literal
2674 interpretation of the following character. However @command{@value{AS}} has no
2675 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2676 code and warns you of the fact.
2679 Which characters are escapable, and what those escapes represent,
2680 varies widely among assemblers. The current set is what we think
2681 the BSD 4.2 assembler recognizes, and is a subset of what most C
2682 compilers recognize. If you are in doubt, do not use an escape
2686 @subsubsection Characters
2688 @cindex single character constant
2689 @cindex character, single
2690 @cindex constant, single character
2691 A single character may be written as a single quote immediately
2692 followed by that character. The same escapes apply to characters as
2693 to strings. So if you want to write the character backslash, you
2694 must write @kbd{'\\} where the first @code{\} escapes the second
2695 @code{\}. As you can see, the quote is an acute accent, not a
2696 grave accent. A newline
2698 @ifclear abnormal-separator
2699 (or semicolon @samp{;})
2701 @ifset abnormal-separator
2703 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2708 immediately following an acute accent is taken as a literal character
2709 and does not count as the end of a statement. The value of a character
2710 constant in a numeric expression is the machine's byte-wide code for
2711 that character. @command{@value{AS}} assumes your character code is ASCII:
2712 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2715 @subsection Number Constants
2717 @cindex constants, number
2718 @cindex number constants
2719 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2720 are stored in the target machine. @emph{Integers} are numbers that
2721 would fit into an @code{int} in the C language. @emph{Bignums} are
2722 integers, but they are stored in more than 32 bits. @emph{Flonums}
2723 are floating point numbers, described below.
2726 * Integers:: Integers
2731 * Bit Fields:: Bit Fields
2737 @subsubsection Integers
2739 @cindex constants, integer
2741 @cindex binary integers
2742 @cindex integers, binary
2743 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2744 the binary digits @samp{01}.
2746 @cindex octal integers
2747 @cindex integers, octal
2748 An octal integer is @samp{0} followed by zero or more of the octal
2749 digits (@samp{01234567}).
2751 @cindex decimal integers
2752 @cindex integers, decimal
2753 A decimal integer starts with a non-zero digit followed by zero or
2754 more digits (@samp{0123456789}).
2756 @cindex hexadecimal integers
2757 @cindex integers, hexadecimal
2758 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2759 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2761 Integers have the usual values. To denote a negative integer, use
2762 the prefix operator @samp{-} discussed under expressions
2763 (@pxref{Prefix Ops,,Prefix Operators}).
2766 @subsubsection Bignums
2769 @cindex constants, bignum
2770 A @dfn{bignum} has the same syntax and semantics as an integer
2771 except that the number (or its negative) takes more than 32 bits to
2772 represent in binary. The distinction is made because in some places
2773 integers are permitted while bignums are not.
2776 @subsubsection Flonums
2778 @cindex floating point numbers
2779 @cindex constants, floating point
2781 @cindex precision, floating point
2782 A @dfn{flonum} represents a floating point number. The translation is
2783 indirect: a decimal floating point number from the text is converted by
2784 @command{@value{AS}} to a generic binary floating point number of more than
2785 sufficient precision. This generic floating point number is converted
2786 to a particular computer's floating point format (or formats) by a
2787 portion of @command{@value{AS}} specialized to that computer.
2789 A flonum is written by writing (in order)
2794 (@samp{0} is optional on the HPPA.)
2798 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2800 @kbd{e} is recommended. Case is not important.
2802 @c FIXME: verify if flonum syntax really this vague for most cases
2803 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2804 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2807 On the H8/300, Renesas / SuperH SH,
2808 and AMD 29K architectures, the letter must be
2809 one of the letters @samp{DFPRSX} (in upper or lower case).
2811 On the ARC, the letter must be one of the letters @samp{DFRS}
2812 (in upper or lower case).
2814 On the Intel 960 architecture, the letter must be
2815 one of the letters @samp{DFT} (in upper or lower case).
2817 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2821 One of the letters @samp{DFRS} (in upper or lower case).
2824 One of the letters @samp{DFPRSX} (in upper or lower case).
2827 The letter @samp{E} (upper case only).
2830 One of the letters @samp{DFT} (in upper or lower case).
2835 An optional sign: either @samp{+} or @samp{-}.
2838 An optional @dfn{integer part}: zero or more decimal digits.
2841 An optional @dfn{fractional part}: @samp{.} followed by zero
2842 or more decimal digits.
2845 An optional exponent, consisting of:
2849 An @samp{E} or @samp{e}.
2850 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2851 @c principle this can perfectly well be different on different targets.
2853 Optional sign: either @samp{+} or @samp{-}.
2855 One or more decimal digits.
2860 At least one of the integer part or the fractional part must be
2861 present. The floating point number has the usual base-10 value.
2863 @command{@value{AS}} does all processing using integers. Flonums are computed
2864 independently of any floating point hardware in the computer running
2865 @command{@value{AS}}.
2869 @c Bit fields are written as a general facility but are also controlled
2870 @c by a conditional-compilation flag---which is as of now (21mar91)
2871 @c turned on only by the i960 config of GAS.
2873 @subsubsection Bit Fields
2876 @cindex constants, bit field
2877 You can also define numeric constants as @dfn{bit fields}.
2878 Specify two numbers separated by a colon---
2880 @var{mask}:@var{value}
2883 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2886 The resulting number is then packed
2888 @c this conditional paren in case bit fields turned on elsewhere than 960
2889 (in host-dependent byte order)
2891 into a field whose width depends on which assembler directive has the
2892 bit-field as its argument. Overflow (a result from the bitwise and
2893 requiring more binary digits to represent) is not an error; instead,
2894 more constants are generated, of the specified width, beginning with the
2895 least significant digits.@refill
2897 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2898 @code{.short}, and @code{.word} accept bit-field arguments.
2903 @chapter Sections and Relocation
2908 * Secs Background:: Background
2909 * Ld Sections:: Linker Sections
2910 * As Sections:: Assembler Internal Sections
2911 * Sub-Sections:: Sub-Sections
2915 @node Secs Background
2918 Roughly, a section is a range of addresses, with no gaps; all data
2919 ``in'' those addresses is treated the same for some particular purpose.
2920 For example there may be a ``read only'' section.
2922 @cindex linker, and assembler
2923 @cindex assembler, and linker
2924 The linker @code{@value{LD}} reads many object files (partial programs) and
2925 combines their contents to form a runnable program. When @command{@value{AS}}
2926 emits an object file, the partial program is assumed to start at address 0.
2927 @code{@value{LD}} assigns the final addresses for the partial program, so that
2928 different partial programs do not overlap. This is actually an
2929 oversimplification, but it suffices to explain how @command{@value{AS}} uses
2932 @code{@value{LD}} moves blocks of bytes of your program to their run-time
2933 addresses. These blocks slide to their run-time addresses as rigid
2934 units; their length does not change and neither does the order of bytes
2935 within them. Such a rigid unit is called a @emph{section}. Assigning
2936 run-time addresses to sections is called @dfn{relocation}. It includes
2937 the task of adjusting mentions of object-file addresses so they refer to
2938 the proper run-time addresses.
2940 For the H8/300, and for the Renesas / SuperH SH,
2941 @command{@value{AS}} pads sections if needed to
2942 ensure they end on a word (sixteen bit) boundary.
2945 @cindex standard assembler sections
2946 An object file written by @command{@value{AS}} has at least three sections, any
2947 of which may be empty. These are named @dfn{text}, @dfn{data} and
2952 When it generates COFF or ELF output,
2954 @command{@value{AS}} can also generate whatever other named sections you specify
2955 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
2956 If you do not use any directives that place output in the @samp{.text}
2957 or @samp{.data} sections, these sections still exist, but are empty.
2962 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
2964 @command{@value{AS}} can also generate whatever other named sections you
2965 specify using the @samp{.space} and @samp{.subspace} directives. See
2966 @cite{HP9000 Series 800 Assembly Language Reference Manual}
2967 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
2968 assembler directives.
2971 Additionally, @command{@value{AS}} uses different names for the standard
2972 text, data, and bss sections when generating SOM output. Program text
2973 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
2974 BSS into @samp{$BSS$}.
2978 Within the object file, the text section starts at address @code{0}, the
2979 data section follows, and the bss section follows the data section.
2982 When generating either SOM or ELF output files on the HPPA, the text
2983 section starts at address @code{0}, the data section at address
2984 @code{0x4000000}, and the bss section follows the data section.
2987 To let @code{@value{LD}} know which data changes when the sections are
2988 relocated, and how to change that data, @command{@value{AS}} also writes to the
2989 object file details of the relocation needed. To perform relocation
2990 @code{@value{LD}} must know, each time an address in the object
2994 Where in the object file is the beginning of this reference to
2997 How long (in bytes) is this reference?
2999 Which section does the address refer to? What is the numeric value of
3001 (@var{address}) @minus{} (@var{start-address of section})?
3004 Is the reference to an address ``Program-Counter relative''?
3007 @cindex addresses, format of
3008 @cindex section-relative addressing
3009 In fact, every address @command{@value{AS}} ever uses is expressed as
3011 (@var{section}) + (@var{offset into section})
3014 Further, most expressions @command{@value{AS}} computes have this section-relative
3017 (For some object formats, such as SOM for the HPPA, some expressions are
3018 symbol-relative instead.)
3021 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3022 @var{N} into section @var{secname}.''
3024 Apart from text, data and bss sections you need to know about the
3025 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3026 addresses in the absolute section remain unchanged. For example, address
3027 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3028 @code{@value{LD}}. Although the linker never arranges two partial programs'
3029 data sections with overlapping addresses after linking, @emph{by definition}
3030 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3031 part of a program is always the same address when the program is running as
3032 address @code{@{absolute@ 239@}} in any other part of the program.
3034 The idea of sections is extended to the @dfn{undefined} section. Any
3035 address whose section is unknown at assembly time is by definition
3036 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3037 Since numbers are always defined, the only way to generate an undefined
3038 address is to mention an undefined symbol. A reference to a named
3039 common block would be such a symbol: its value is unknown at assembly
3040 time so it has section @emph{undefined}.
3042 By analogy the word @emph{section} is used to describe groups of sections in
3043 the linked program. @code{@value{LD}} puts all partial programs' text
3044 sections in contiguous addresses in the linked program. It is
3045 customary to refer to the @emph{text section} of a program, meaning all
3046 the addresses of all partial programs' text sections. Likewise for
3047 data and bss sections.
3049 Some sections are manipulated by @code{@value{LD}}; others are invented for
3050 use of @command{@value{AS}} and have no meaning except during assembly.
3053 @section Linker Sections
3054 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3059 @cindex named sections
3060 @cindex sections, named
3061 @item named sections
3064 @cindex text section
3065 @cindex data section
3069 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3070 separate but equal sections. Anything you can say of one section is
3073 When the program is running, however, it is
3074 customary for the text section to be unalterable. The
3075 text section is often shared among processes: it contains
3076 instructions, constants and the like. The data section of a running
3077 program is usually alterable: for example, C variables would be stored
3078 in the data section.
3083 This section contains zeroed bytes when your program begins running. It
3084 is used to hold uninitialized variables or common storage. The length of
3085 each partial program's bss section is important, but because it starts
3086 out containing zeroed bytes there is no need to store explicit zero
3087 bytes in the object file. The bss section was invented to eliminate
3088 those explicit zeros from object files.
3090 @cindex absolute section
3091 @item absolute section
3092 Address 0 of this section is always ``relocated'' to runtime address 0.
3093 This is useful if you want to refer to an address that @code{@value{LD}} must
3094 not change when relocating. In this sense we speak of absolute
3095 addresses being ``unrelocatable'': they do not change during relocation.
3097 @cindex undefined section
3098 @item undefined section
3099 This ``section'' is a catch-all for address references to objects not in
3100 the preceding sections.
3101 @c FIXME: ref to some other doc on obj-file formats could go here.
3104 @cindex relocation example
3105 An idealized example of three relocatable sections follows.
3107 The example uses the traditional section names @samp{.text} and @samp{.data}.
3109 Memory addresses are on the horizontal axis.
3113 @c END TEXI2ROFF-KILL
3116 partial program # 1: |ttttt|dddd|00|
3123 partial program # 2: |TTT|DDD|000|
3126 +--+---+-----+--+----+---+-----+~~
3127 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3128 +--+---+-----+--+----+---+-----+~~
3130 addresses: 0 @dots{}
3137 \line{\it Partial program \#1: \hfil}
3138 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3139 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3141 \line{\it Partial program \#2: \hfil}
3142 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3143 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3145 \line{\it linked program: \hfil}
3146 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3147 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3148 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3149 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3151 \line{\it addresses: \hfil}
3155 @c END TEXI2ROFF-KILL
3158 @section Assembler Internal Sections
3160 @cindex internal assembler sections
3161 @cindex sections in messages, internal
3162 These sections are meant only for the internal use of @command{@value{AS}}. They
3163 have no meaning at run-time. You do not really need to know about these
3164 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3165 warning messages, so it might be helpful to have an idea of their
3166 meanings to @command{@value{AS}}. These sections are used to permit the
3167 value of every expression in your assembly language program to be a
3168 section-relative address.
3171 @cindex assembler internal logic error
3172 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3173 An internal assembler logic error has been found. This means there is a
3174 bug in the assembler.
3176 @cindex expr (internal section)
3178 The assembler stores complex expression internally as combinations of
3179 symbols. When it needs to represent an expression as a symbol, it puts
3180 it in the expr section.
3182 @c FIXME item transfer[t] vector preload
3183 @c FIXME item transfer[t] vector postload
3184 @c FIXME item register
3188 @section Sub-Sections
3190 @cindex numbered subsections
3191 @cindex grouping data
3197 fall into two sections: text and data.
3199 You may have separate groups of
3201 data in named sections
3205 data in named sections
3211 that you want to end up near to each other in the object file, even though they
3212 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3213 use @dfn{subsections} for this purpose. Within each section, there can be
3214 numbered subsections with values from 0 to 8192. Objects assembled into the
3215 same subsection go into the object file together with other objects in the same
3216 subsection. For example, a compiler might want to store constants in the text
3217 section, but might not want to have them interspersed with the program being
3218 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3219 section of code being output, and a @samp{.text 1} before each group of
3220 constants being output.
3222 Subsections are optional. If you do not use subsections, everything
3223 goes in subsection number zero.
3226 Each subsection is zero-padded up to a multiple of four bytes.
3227 (Subsections may be padded a different amount on different flavors
3228 of @command{@value{AS}}.)
3232 On the H8/300 platform, each subsection is zero-padded to a word
3233 boundary (two bytes).
3234 The same is true on the Renesas SH.
3237 @c FIXME section padding (alignment)?
3238 @c Rich Pixley says padding here depends on target obj code format; that
3239 @c doesn't seem particularly useful to say without further elaboration,
3240 @c so for now I say nothing about it. If this is a generic BFD issue,
3241 @c these paragraphs might need to vanish from this manual, and be
3242 @c discussed in BFD chapter of binutils (or some such).
3246 Subsections appear in your object file in numeric order, lowest numbered
3247 to highest. (All this to be compatible with other people's assemblers.)
3248 The object file contains no representation of subsections; @code{@value{LD}} and
3249 other programs that manipulate object files see no trace of them.
3250 They just see all your text subsections as a text section, and all your
3251 data subsections as a data section.
3253 To specify which subsection you want subsequent statements assembled
3254 into, use a numeric argument to specify it, in a @samp{.text
3255 @var{expression}} or a @samp{.data @var{expression}} statement.
3258 When generating COFF output, you
3263 can also use an extra subsection
3264 argument with arbitrary named sections: @samp{.section @var{name},
3269 When generating ELF output, you
3274 can also use the @code{.subsection} directive (@pxref{SubSection})
3275 to specify a subsection: @samp{.subsection @var{expression}}.
3277 @var{Expression} should be an absolute expression
3278 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3279 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3280 begins in @code{text 0}. For instance:
3282 .text 0 # The default subsection is text 0 anyway.
3283 .ascii "This lives in the first text subsection. *"
3285 .ascii "But this lives in the second text subsection."
3287 .ascii "This lives in the data section,"
3288 .ascii "in the first data subsection."
3290 .ascii "This lives in the first text section,"
3291 .ascii "immediately following the asterisk (*)."
3294 Each section has a @dfn{location counter} incremented by one for every byte
3295 assembled into that section. Because subsections are merely a convenience
3296 restricted to @command{@value{AS}} there is no concept of a subsection location
3297 counter. There is no way to directly manipulate a location counter---but the
3298 @code{.align} directive changes it, and any label definition captures its
3299 current value. The location counter of the section where statements are being
3300 assembled is said to be the @dfn{active} location counter.
3303 @section bss Section
3306 @cindex common variable storage
3307 The bss section is used for local common variable storage.
3308 You may allocate address space in the bss section, but you may
3309 not dictate data to load into it before your program executes. When
3310 your program starts running, all the contents of the bss
3311 section are zeroed bytes.
3313 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3314 @ref{Lcomm,,@code{.lcomm}}.
3316 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3317 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3320 When assembling for a target which supports multiple sections, such as ELF or
3321 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3322 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3323 section. Typically the section will only contain symbol definitions and
3324 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3331 Symbols are a central concept: the programmer uses symbols to name
3332 things, the linker uses symbols to link, and the debugger uses symbols
3336 @cindex debuggers, and symbol order
3337 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3338 the same order they were declared. This may break some debuggers.
3343 * Setting Symbols:: Giving Symbols Other Values
3344 * Symbol Names:: Symbol Names
3345 * Dot:: The Special Dot Symbol
3346 * Symbol Attributes:: Symbol Attributes
3353 A @dfn{label} is written as a symbol immediately followed by a colon
3354 @samp{:}. The symbol then represents the current value of the
3355 active location counter, and is, for example, a suitable instruction
3356 operand. You are warned if you use the same symbol to represent two
3357 different locations: the first definition overrides any other
3361 On the HPPA, the usual form for a label need not be immediately followed by a
3362 colon, but instead must start in column zero. Only one label may be defined on
3363 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3364 provides a special directive @code{.label} for defining labels more flexibly.
3367 @node Setting Symbols
3368 @section Giving Symbols Other Values
3370 @cindex assigning values to symbols
3371 @cindex symbol values, assigning
3372 A symbol can be given an arbitrary value by writing a symbol, followed
3373 by an equals sign @samp{=}, followed by an expression
3374 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3375 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3376 equals sign @samp{=}@samp{=} here represents an equivalent of the
3377 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3380 Blackfin does not support symbol assignment with @samp{=}.
3384 @section Symbol Names
3386 @cindex symbol names
3387 @cindex names, symbol
3388 @ifclear SPECIAL-SYMS
3389 Symbol names begin with a letter or with one of @samp{._}. On most
3390 machines, you can also use @code{$} in symbol names; exceptions are
3391 noted in @ref{Machine Dependencies}. That character may be followed by any
3392 string of digits, letters, dollar signs (unless otherwise noted for a
3393 particular target machine), and underscores.
3397 Symbol names begin with a letter or with one of @samp{._}. On the
3398 Renesas SH you can also use @code{$} in symbol names. That
3399 character may be followed by any string of digits, letters, dollar signs (save
3400 on the H8/300), and underscores.
3404 Case of letters is significant: @code{foo} is a different symbol name
3407 Each symbol has exactly one name. Each name in an assembly language program
3408 refers to exactly one symbol. You may use that symbol name any number of times
3411 @subheading Local Symbol Names
3413 @cindex local symbol names
3414 @cindex symbol names, local
3415 A local symbol is any symbol beginning with certain local label prefixes.
3416 By default, the local label prefix is @samp{.L} for ELF systems or
3417 @samp{L} for traditional a.out systems, but each target may have its own
3418 set of local label prefixes.
3420 On the HPPA local symbols begin with @samp{L$}.
3423 Local symbols are defined and used within the assembler, but they are
3424 normally not saved in object files. Thus, they are not visible when debugging.
3425 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3426 @option{-L}}) to retain the local symbols in the object files.
3428 @subheading Local Labels
3430 @cindex local labels
3431 @cindex temporary symbol names
3432 @cindex symbol names, temporary
3433 Local labels help compilers and programmers use names temporarily.
3434 They create symbols which are guaranteed to be unique over the entire scope of
3435 the input source code and which can be referred to by a simple notation.
3436 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3437 represents any positive integer). To refer to the most recent previous
3438 definition of that label write @samp{@b{N}b}, using the same number as when
3439 you defined the label. To refer to the next definition of a local label, write
3440 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3443 There is no restriction on how you can use these labels, and you can reuse them
3444 too. So that it is possible to repeatedly define the same local label (using
3445 the same number @samp{@b{N}}), although you can only refer to the most recently
3446 defined local label of that number (for a backwards reference) or the next
3447 definition of a specific local label for a forward reference. It is also worth
3448 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3449 implemented in a slightly more efficient manner than the others.
3460 Which is the equivalent of:
3463 label_1: branch label_3
3464 label_2: branch label_1
3465 label_3: branch label_4
3466 label_4: branch label_3
3469 Local label names are only a notational device. They are immediately
3470 transformed into more conventional symbol names before the assembler uses them.
3471 The symbol names are stored in the symbol table, appear in error messages, and
3472 are optionally emitted to the object file. The names are constructed using
3476 @item @emph{local label prefix}
3477 All local symbols begin with the system-specific local label prefix.
3478 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3479 that start with the local label prefix. These labels are
3480 used for symbols you are never intended to see. If you use the
3481 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3482 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3483 you may use them in debugging.
3486 This is the number that was used in the local label definition. So if the
3487 label is written @samp{55:} then the number is @samp{55}.
3490 This unusual character is included so you do not accidentally invent a symbol
3491 of the same name. The character has ASCII value of @samp{\002} (control-B).
3493 @item @emph{ordinal number}
3494 This is a serial number to keep the labels distinct. The first definition of
3495 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3496 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3497 the number @samp{1} and its 15th definition gets @samp{15} as well.
3500 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3501 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3503 @subheading Dollar Local Labels
3504 @cindex dollar local symbols
3506 @code{@value{AS}} also supports an even more local form of local labels called
3507 dollar labels. These labels go out of scope (i.e., they become undefined) as
3508 soon as a non-local label is defined. Thus they remain valid for only a small
3509 region of the input source code. Normal local labels, by contrast, remain in
3510 scope for the entire file, or until they are redefined by another occurrence of
3511 the same local label.
3513 Dollar labels are defined in exactly the same way as ordinary local labels,
3514 except that they have a dollar sign suffix to their numeric value, e.g.,
3517 They can also be distinguished from ordinary local labels by their transformed
3518 names which use ASCII character @samp{\001} (control-A) as the magic character
3519 to distinguish them from ordinary labels. For example, the fifth definition of
3520 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3523 @section The Special Dot Symbol
3525 @cindex dot (symbol)
3526 @cindex @code{.} (symbol)
3527 @cindex current address
3528 @cindex location counter
3529 The special symbol @samp{.} refers to the current address that
3530 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3531 .long .} defines @code{melvin} to contain its own address.
3532 Assigning a value to @code{.} is treated the same as a @code{.org}
3534 @ifclear no-space-dir
3535 Thus, the expression @samp{.=.+4} is the same as saying
3539 @node Symbol Attributes
3540 @section Symbol Attributes
3542 @cindex symbol attributes
3543 @cindex attributes, symbol
3544 Every symbol has, as well as its name, the attributes ``Value'' and
3545 ``Type''. Depending on output format, symbols can also have auxiliary
3548 The detailed definitions are in @file{a.out.h}.
3551 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3552 all these attributes, and probably won't warn you. This makes the
3553 symbol an externally defined symbol, which is generally what you
3557 * Symbol Value:: Value
3558 * Symbol Type:: Type
3561 * a.out Symbols:: Symbol Attributes: @code{a.out}
3565 * a.out Symbols:: Symbol Attributes: @code{a.out}
3568 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3573 * COFF Symbols:: Symbol Attributes for COFF
3576 * SOM Symbols:: Symbol Attributes for SOM
3583 @cindex value of a symbol
3584 @cindex symbol value
3585 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3586 location in the text, data, bss or absolute sections the value is the
3587 number of addresses from the start of that section to the label.
3588 Naturally for text, data and bss sections the value of a symbol changes
3589 as @code{@value{LD}} changes section base addresses during linking. Absolute
3590 symbols' values do not change during linking: that is why they are
3593 The value of an undefined symbol is treated in a special way. If it is
3594 0 then the symbol is not defined in this assembler source file, and
3595 @code{@value{LD}} tries to determine its value from other files linked into the
3596 same program. You make this kind of symbol simply by mentioning a symbol
3597 name without defining it. A non-zero value represents a @code{.comm}
3598 common declaration. The value is how much common storage to reserve, in
3599 bytes (addresses). The symbol refers to the first address of the
3605 @cindex type of a symbol
3607 The type attribute of a symbol contains relocation (section)
3608 information, any flag settings indicating that a symbol is external, and
3609 (optionally), other information for linkers and debuggers. The exact
3610 format depends on the object-code output format in use.
3615 @c The following avoids a "widow" subsection title. @group would be
3616 @c better if it were available outside examples.
3619 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3621 @cindex @code{b.out} symbol attributes
3622 @cindex symbol attributes, @code{b.out}
3623 These symbol attributes appear only when @command{@value{AS}} is configured for
3624 one of the Berkeley-descended object output formats---@code{a.out} or
3630 @subsection Symbol Attributes: @code{a.out}
3632 @cindex @code{a.out} symbol attributes
3633 @cindex symbol attributes, @code{a.out}
3639 @subsection Symbol Attributes: @code{a.out}
3641 @cindex @code{a.out} symbol attributes
3642 @cindex symbol attributes, @code{a.out}
3646 * Symbol Desc:: Descriptor
3647 * Symbol Other:: Other
3651 @subsubsection Descriptor
3653 @cindex descriptor, of @code{a.out} symbol
3654 This is an arbitrary 16-bit value. You may establish a symbol's
3655 descriptor value by using a @code{.desc} statement
3656 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3657 @command{@value{AS}}.
3660 @subsubsection Other
3662 @cindex other attribute, of @code{a.out} symbol
3663 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3668 @subsection Symbol Attributes for COFF
3670 @cindex COFF symbol attributes
3671 @cindex symbol attributes, COFF
3673 The COFF format supports a multitude of auxiliary symbol attributes;
3674 like the primary symbol attributes, they are set between @code{.def} and
3675 @code{.endef} directives.
3677 @subsubsection Primary Attributes
3679 @cindex primary attributes, COFF symbols
3680 The symbol name is set with @code{.def}; the value and type,
3681 respectively, with @code{.val} and @code{.type}.
3683 @subsubsection Auxiliary Attributes
3685 @cindex auxiliary attributes, COFF symbols
3686 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3687 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3688 table information for COFF.
3693 @subsection Symbol Attributes for SOM
3695 @cindex SOM symbol attributes
3696 @cindex symbol attributes, SOM
3698 The SOM format for the HPPA supports a multitude of symbol attributes set with
3699 the @code{.EXPORT} and @code{.IMPORT} directives.
3701 The attributes are described in @cite{HP9000 Series 800 Assembly
3702 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3703 @code{EXPORT} assembler directive documentation.
3707 @chapter Expressions
3711 @cindex numeric values
3712 An @dfn{expression} specifies an address or numeric value.
3713 Whitespace may precede and/or follow an expression.
3715 The result of an expression must be an absolute number, or else an offset into
3716 a particular section. If an expression is not absolute, and there is not
3717 enough information when @command{@value{AS}} sees the expression to know its
3718 section, a second pass over the source program might be necessary to interpret
3719 the expression---but the second pass is currently not implemented.
3720 @command{@value{AS}} aborts with an error message in this situation.
3723 * Empty Exprs:: Empty Expressions
3724 * Integer Exprs:: Integer Expressions
3728 @section Empty Expressions
3730 @cindex empty expressions
3731 @cindex expressions, empty
3732 An empty expression has no value: it is just whitespace or null.
3733 Wherever an absolute expression is required, you may omit the
3734 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3735 is compatible with other assemblers.
3738 @section Integer Expressions
3740 @cindex integer expressions
3741 @cindex expressions, integer
3742 An @dfn{integer expression} is one or more @emph{arguments} delimited
3743 by @emph{operators}.
3746 * Arguments:: Arguments
3747 * Operators:: Operators
3748 * Prefix Ops:: Prefix Operators
3749 * Infix Ops:: Infix Operators
3753 @subsection Arguments
3755 @cindex expression arguments
3756 @cindex arguments in expressions
3757 @cindex operands in expressions
3758 @cindex arithmetic operands
3759 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3760 contexts arguments are sometimes called ``arithmetic operands''. In
3761 this manual, to avoid confusing them with the ``instruction operands'' of
3762 the machine language, we use the term ``argument'' to refer to parts of
3763 expressions only, reserving the word ``operand'' to refer only to machine
3764 instruction operands.
3766 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3767 @var{section} is one of text, data, bss, absolute,
3768 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3771 Numbers are usually integers.
3773 A number can be a flonum or bignum. In this case, you are warned
3774 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3775 these 32 bits are an integer. You may write integer-manipulating
3776 instructions that act on exotic constants, compatible with other
3779 @cindex subexpressions
3780 Subexpressions are a left parenthesis @samp{(} followed by an integer
3781 expression, followed by a right parenthesis @samp{)}; or a prefix
3782 operator followed by an argument.
3785 @subsection Operators
3787 @cindex operators, in expressions
3788 @cindex arithmetic functions
3789 @cindex functions, in expressions
3790 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3791 operators are followed by an argument. Infix operators appear
3792 between their arguments. Operators may be preceded and/or followed by
3796 @subsection Prefix Operator
3798 @cindex prefix operators
3799 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3800 one argument, which must be absolute.
3802 @c the tex/end tex stuff surrounding this small table is meant to make
3803 @c it align, on the printed page, with the similar table in the next
3804 @c section (which is inside an enumerate).
3806 \global\advance\leftskip by \itemindent
3811 @dfn{Negation}. Two's complement negation.
3813 @dfn{Complementation}. Bitwise not.
3817 \global\advance\leftskip by -\itemindent
3821 @subsection Infix Operators
3823 @cindex infix operators
3824 @cindex operators, permitted arguments
3825 @dfn{Infix operators} take two arguments, one on either side. Operators
3826 have precedence, but operations with equal precedence are performed left
3827 to right. Apart from @code{+} or @option{-}, both arguments must be
3828 absolute, and the result is absolute.
3831 @cindex operator precedence
3832 @cindex precedence of operators
3839 @dfn{Multiplication}.
3842 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3848 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3851 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3855 Intermediate precedence
3860 @dfn{Bitwise Inclusive Or}.
3866 @dfn{Bitwise Exclusive Or}.
3869 @dfn{Bitwise Or Not}.
3876 @cindex addition, permitted arguments
3877 @cindex plus, permitted arguments
3878 @cindex arguments for addition
3880 @dfn{Addition}. If either argument is absolute, the result has the section of
3881 the other argument. You may not add together arguments from different
3884 @cindex subtraction, permitted arguments
3885 @cindex minus, permitted arguments
3886 @cindex arguments for subtraction
3888 @dfn{Subtraction}. If the right argument is absolute, the
3889 result has the section of the left argument.
3890 If both arguments are in the same section, the result is absolute.
3891 You may not subtract arguments from different sections.
3892 @c FIXME is there still something useful to say about undefined - undefined ?
3894 @cindex comparison expressions
3895 @cindex expressions, comparison
3900 @dfn{Is Not Equal To}
3904 @dfn{Is Greater Than}
3906 @dfn{Is Greater Than Or Equal To}
3908 @dfn{Is Less Than Or Equal To}
3910 The comparison operators can be used as infix operators. A true results has a
3911 value of -1 whereas a false result has a value of 0. Note, these operators
3912 perform signed comparisons.
3915 @item Lowest Precedence
3924 These two logical operations can be used to combine the results of sub
3925 expressions. Note, unlike the comparison operators a true result returns a
3926 value of 1 but a false results does still return 0. Also note that the logical
3927 or operator has a slightly lower precedence than logical and.
3932 In short, it's only meaningful to add or subtract the @emph{offsets} in an
3933 address; you can only have a defined section in one of the two arguments.
3936 @chapter Assembler Directives
3938 @cindex directives, machine independent
3939 @cindex pseudo-ops, machine independent
3940 @cindex machine independent directives
3941 All assembler directives have names that begin with a period (@samp{.}).
3942 The rest of the name is letters, usually in lower case.
3944 This chapter discusses directives that are available regardless of the
3945 target machine configuration for the @sc{gnu} assembler.
3947 Some machine configurations provide additional directives.
3948 @xref{Machine Dependencies}.
3951 @ifset machine-directives
3952 @xref{Machine Dependencies}, for additional directives.
3957 * Abort:: @code{.abort}
3959 * ABORT (COFF):: @code{.ABORT}
3962 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
3963 * Altmacro:: @code{.altmacro}
3964 * Ascii:: @code{.ascii "@var{string}"}@dots{}
3965 * Asciz:: @code{.asciz "@var{string}"}@dots{}
3966 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
3967 * Byte:: @code{.byte @var{expressions}}
3968 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
3969 * Comm:: @code{.comm @var{symbol} , @var{length} }
3970 * Data:: @code{.data @var{subsection}}
3972 * Def:: @code{.def @var{name}}
3975 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
3981 * Double:: @code{.double @var{flonums}}
3982 * Eject:: @code{.eject}
3983 * Else:: @code{.else}
3984 * Elseif:: @code{.elseif}
3987 * Endef:: @code{.endef}
3990 * Endfunc:: @code{.endfunc}
3991 * Endif:: @code{.endif}
3992 * Equ:: @code{.equ @var{symbol}, @var{expression}}
3993 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
3994 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
3996 * Error:: @code{.error @var{string}}
3997 * Exitm:: @code{.exitm}
3998 * Extern:: @code{.extern}
3999 * Fail:: @code{.fail}
4000 * File:: @code{.file}
4001 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4002 * Float:: @code{.float @var{flonums}}
4003 * Func:: @code{.func}
4004 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4006 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4007 * Hidden:: @code{.hidden @var{names}}
4010 * hword:: @code{.hword @var{expressions}}
4011 * Ident:: @code{.ident}
4012 * If:: @code{.if @var{absolute expression}}
4013 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4014 * Include:: @code{.include "@var{file}"}
4015 * Int:: @code{.int @var{expressions}}
4017 * Internal:: @code{.internal @var{names}}
4020 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4021 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4022 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4023 * Lflags:: @code{.lflags}
4024 @ifclear no-line-dir
4025 * Line:: @code{.line @var{line-number}}
4028 * Linkonce:: @code{.linkonce [@var{type}]}
4029 * List:: @code{.list}
4030 * Ln:: @code{.ln @var{line-number}}
4031 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4032 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4034 * Local:: @code{.local @var{names}}
4037 * Long:: @code{.long @var{expressions}}
4039 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4042 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4043 * MRI:: @code{.mri @var{val}}
4044 * Noaltmacro:: @code{.noaltmacro}
4045 * Nolist:: @code{.nolist}
4046 * Octa:: @code{.octa @var{bignums}}
4047 * Org:: @code{.org @var{new-lc}, @var{fill}}
4048 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4050 * PopSection:: @code{.popsection}
4051 * Previous:: @code{.previous}
4054 * Print:: @code{.print @var{string}}
4056 * Protected:: @code{.protected @var{names}}
4059 * Psize:: @code{.psize @var{lines}, @var{columns}}
4060 * Purgem:: @code{.purgem @var{name}}
4062 * PushSection:: @code{.pushsection @var{name}}
4065 * Quad:: @code{.quad @var{bignums}}
4066 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4067 * Rept:: @code{.rept @var{count}}
4068 * Sbttl:: @code{.sbttl "@var{subheading}"}
4070 * Scl:: @code{.scl @var{class}}
4073 * Section:: @code{.section @var{name}[, @var{flags}]}
4076 * Set:: @code{.set @var{symbol}, @var{expression}}
4077 * Short:: @code{.short @var{expressions}}
4078 * Single:: @code{.single @var{flonums}}
4080 * Size:: @code{.size [@var{name} , @var{expression}]}
4082 @ifclear no-space-dir
4083 * Skip:: @code{.skip @var{size} , @var{fill}}
4086 * Sleb128:: @code{.sleb128 @var{expressions}}
4087 @ifclear no-space-dir
4088 * Space:: @code{.space @var{size} , @var{fill}}
4091 * Stab:: @code{.stabd, .stabn, .stabs}
4094 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4095 * Struct:: @code{.struct @var{expression}}
4097 * SubSection:: @code{.subsection}
4098 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4102 * Tag:: @code{.tag @var{structname}}
4105 * Text:: @code{.text @var{subsection}}
4106 * Title:: @code{.title "@var{heading}"}
4108 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4111 * Uleb128:: @code{.uleb128 @var{expressions}}
4113 * Val:: @code{.val @var{addr}}
4117 * Version:: @code{.version "@var{string}"}
4118 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4119 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4122 * Warning:: @code{.warning @var{string}}
4123 * Weak:: @code{.weak @var{names}}
4124 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4125 * Word:: @code{.word @var{expressions}}
4126 * Deprecated:: Deprecated Directives
4130 @section @code{.abort}
4132 @cindex @code{abort} directive
4133 @cindex stopping the assembly
4134 This directive stops the assembly immediately. It is for
4135 compatibility with other assemblers. The original idea was that the
4136 assembly language source would be piped into the assembler. If the sender
4137 of the source quit, it could use this directive tells @command{@value{AS}} to
4138 quit also. One day @code{.abort} will not be supported.
4142 @section @code{.ABORT} (COFF)
4144 @cindex @code{ABORT} directive
4145 When producing COFF output, @command{@value{AS}} accepts this directive as a
4146 synonym for @samp{.abort}.
4149 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4155 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4157 @cindex padding the location counter
4158 @cindex @code{align} directive
4159 Pad the location counter (in the current subsection) to a particular storage
4160 boundary. The first expression (which must be absolute) is the alignment
4161 required, as described below.
4163 The second expression (also absolute) gives the fill value to be stored in the
4164 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4165 padding bytes are normally zero. However, on some systems, if the section is
4166 marked as containing code and the fill value is omitted, the space is filled
4167 with no-op instructions.
4169 The third expression is also absolute, and is also optional. If it is present,
4170 it is the maximum number of bytes that should be skipped by this alignment
4171 directive. If doing the alignment would require skipping more bytes than the
4172 specified maximum, then the alignment is not done at all. You can omit the
4173 fill value (the second argument) entirely by simply using two commas after the
4174 required alignment; this can be useful if you want the alignment to be filled
4175 with no-op instructions when appropriate.
4177 The way the required alignment is specified varies from system to system.
4178 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4179 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4180 alignment request in bytes. For example @samp{.align 8} advances
4181 the location counter until it is a multiple of 8. If the location counter
4182 is already a multiple of 8, no change is needed. For the tic54x, the
4183 first expression is the alignment request in words.
4185 For other systems, including ppc, i386 using a.out format, arm and
4186 strongarm, it is the
4187 number of low-order zero bits the location counter must have after
4188 advancement. For example @samp{.align 3} advances the location
4189 counter until it a multiple of 8. If the location counter is already a
4190 multiple of 8, no change is needed.
4192 This inconsistency is due to the different behaviors of the various
4193 native assemblers for these systems which GAS must emulate.
4194 GAS also provides @code{.balign} and @code{.p2align} directives,
4195 described later, which have a consistent behavior across all
4196 architectures (but are specific to GAS).
4199 @section @code{.altmacro}
4200 Enable alternate macro mode, enabling:
4203 @item LOCAL @var{name} [ , @dots{} ]
4204 One additional directive, @code{LOCAL}, is available. It is used to
4205 generate a string replacement for each of the @var{name} arguments, and
4206 replace any instances of @var{name} in each macro expansion. The
4207 replacement string is unique in the assembly, and different for each
4208 separate macro expansion. @code{LOCAL} allows you to write macros that
4209 define symbols, without fear of conflict between separate macro expansions.
4211 @item String delimiters
4212 You can write strings delimited in these other ways besides
4213 @code{"@var{string}"}:
4216 @item '@var{string}'
4217 You can delimit strings with single-quote characters.
4219 @item <@var{string}>
4220 You can delimit strings with matching angle brackets.
4223 @item single-character string escape
4224 To include any single character literally in a string (even if the
4225 character would otherwise have some special meaning), you can prefix the
4226 character with @samp{!} (an exclamation mark). For example, you can
4227 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4229 @item Expression results as strings
4230 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4231 and use the result as a string.
4235 @section @code{.ascii "@var{string}"}@dots{}
4237 @cindex @code{ascii} directive
4238 @cindex string literals
4239 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4240 separated by commas. It assembles each string (with no automatic
4241 trailing zero byte) into consecutive addresses.
4244 @section @code{.asciz "@var{string}"}@dots{}
4246 @cindex @code{asciz} directive
4247 @cindex zero-terminated strings
4248 @cindex null-terminated strings
4249 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4250 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4253 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4255 @cindex padding the location counter given number of bytes
4256 @cindex @code{balign} directive
4257 Pad the location counter (in the current subsection) to a particular
4258 storage boundary. The first expression (which must be absolute) is the
4259 alignment request in bytes. For example @samp{.balign 8} advances
4260 the location counter until it is a multiple of 8. If the location counter
4261 is already a multiple of 8, no change is needed.
4263 The second expression (also absolute) gives the fill value to be stored in the
4264 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4265 padding bytes are normally zero. However, on some systems, if the section is
4266 marked as containing code and the fill value is omitted, the space is filled
4267 with no-op instructions.
4269 The third expression is also absolute, and is also optional. If it is present,
4270 it is the maximum number of bytes that should be skipped by this alignment
4271 directive. If doing the alignment would require skipping more bytes than the
4272 specified maximum, then the alignment is not done at all. You can omit the
4273 fill value (the second argument) entirely by simply using two commas after the
4274 required alignment; this can be useful if you want the alignment to be filled
4275 with no-op instructions when appropriate.
4277 @cindex @code{balignw} directive
4278 @cindex @code{balignl} directive
4279 The @code{.balignw} and @code{.balignl} directives are variants of the
4280 @code{.balign} directive. The @code{.balignw} directive treats the fill
4281 pattern as a two byte word value. The @code{.balignl} directives treats the
4282 fill pattern as a four byte longword value. For example, @code{.balignw
4283 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4284 filled in with the value 0x368d (the exact placement of the bytes depends upon
4285 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4289 @section @code{.byte @var{expressions}}
4291 @cindex @code{byte} directive
4292 @cindex integers, one byte
4293 @code{.byte} expects zero or more expressions, separated by commas.
4294 Each expression is assembled into the next byte.
4296 @node CFI directives
4297 @section @code{.cfi_sections @var{section_list}}
4298 @cindex @code{cfi_sections} directive
4299 @code{.cfi_sections} may be used to specify whether CFI directives
4300 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4301 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4302 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4303 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4304 directive is not used is @code{.cfi_sections .eh_frame}.
4306 @section @code{.cfi_startproc [simple]}
4307 @cindex @code{cfi_startproc} directive
4308 @code{.cfi_startproc} is used at the beginning of each function that
4309 should have an entry in @code{.eh_frame}. It initializes some internal
4310 data structures. Don't forget to close the function by
4311 @code{.cfi_endproc}.
4313 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4314 it also emits some architecture dependent initial CFI instructions.
4316 @section @code{.cfi_endproc}
4317 @cindex @code{cfi_endproc} directive
4318 @code{.cfi_endproc} is used at the end of a function where it closes its
4319 unwind entry previously opened by
4320 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4322 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4323 @code{.cfi_personality} defines personality routine and its encoding.
4324 @var{encoding} must be a constant determining how the personality
4325 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4326 argument is not present, otherwise second argument should be
4327 a constant or a symbol name. When using indirect encodings,
4328 the symbol provided should be the location where personality
4329 can be loaded from, not the personality routine itself.
4330 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4331 no personality routine.
4333 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4334 @code{.cfi_lsda} defines LSDA and its encoding.
4335 @var{encoding} must be a constant determining how the LSDA
4336 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4337 argument is not present, otherwise second argument should be a constant
4338 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4341 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4342 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4343 address from @var{register} and add @var{offset} to it}.
4345 @section @code{.cfi_def_cfa_register @var{register}}
4346 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4347 now on @var{register} will be used instead of the old one. Offset
4350 @section @code{.cfi_def_cfa_offset @var{offset}}
4351 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4352 remains the same, but @var{offset} is new. Note that it is the
4353 absolute offset that will be added to a defined register to compute
4356 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4357 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4358 value that is added/substracted from the previous offset.
4360 @section @code{.cfi_offset @var{register}, @var{offset}}
4361 Previous value of @var{register} is saved at offset @var{offset} from
4364 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4365 Previous value of @var{register} is saved at offset @var{offset} from
4366 the current CFA register. This is transformed to @code{.cfi_offset}
4367 using the known displacement of the CFA register from the CFA.
4368 This is often easier to use, because the number will match the
4369 code it's annotating.
4371 @section @code{.cfi_register @var{register1}, @var{register2}}
4372 Previous value of @var{register1} is saved in register @var{register2}.
4374 @section @code{.cfi_restore @var{register}}
4375 @code{.cfi_restore} says that the rule for @var{register} is now the
4376 same as it was at the beginning of the function, after all initial
4377 instruction added by @code{.cfi_startproc} were executed.
4379 @section @code{.cfi_undefined @var{register}}
4380 From now on the previous value of @var{register} can't be restored anymore.
4382 @section @code{.cfi_same_value @var{register}}
4383 Current value of @var{register} is the same like in the previous frame,
4384 i.e. no restoration needed.
4386 @section @code{.cfi_remember_state},
4387 First save all current rules for all registers by @code{.cfi_remember_state},
4388 then totally screw them up by subsequent @code{.cfi_*} directives and when
4389 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4390 the previous saved state.
4392 @section @code{.cfi_return_column @var{register}}
4393 Change return column @var{register}, i.e. the return address is either
4394 directly in @var{register} or can be accessed by rules for @var{register}.
4396 @section @code{.cfi_signal_frame}
4397 Mark current function as signal trampoline.
4399 @section @code{.cfi_window_save}
4400 SPARC register window has been saved.
4402 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4403 Allows the user to add arbitrary bytes to the unwind info. One
4404 might use this to add OS-specific CFI opcodes, or generic CFI
4405 opcodes that GAS does not yet support.
4407 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4408 The current value of @var{register} is @var{label}. The value of @var{label}
4409 will be encoded in the output file according to @var{encoding}; see the
4410 description of @code{.cfi_personality} for details on this encoding.
4412 The usefulness of equating a register to a fixed label is probably
4413 limited to the return address register. Here, it can be useful to
4414 mark a code segment that has only one return address which is reached
4415 by a direct branch and no copy of the return address exists in memory
4416 or another register.
4419 @section @code{.comm @var{symbol} , @var{length} }
4421 @cindex @code{comm} directive
4422 @cindex symbol, common
4423 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4424 common symbol in one object file may be merged with a defined or common symbol
4425 of the same name in another object file. If @code{@value{LD}} does not see a
4426 definition for the symbol--just one or more common symbols--then it will
4427 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4428 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4429 the same name, and they do not all have the same size, it will allocate space
4430 using the largest size.
4433 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4434 an optional third argument. This is the desired alignment of the symbol,
4435 specified for ELF as a byte boundary (for example, an alignment of 16 means
4436 that the least significant 4 bits of the address should be zero), and for PE
4437 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4438 boundary). The alignment must be an absolute expression, and it must be a
4439 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4440 common symbol, it will use the alignment when placing the symbol. If no
4441 alignment is specified, @command{@value{AS}} will set the alignment to the
4442 largest power of two less than or equal to the size of the symbol, up to a
4443 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4444 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4445 @samp{--section-alignment} option; image file sections in PE are aligned to
4446 multiples of 4096, which is far too large an alignment for ordinary variables.
4447 It is rather the default alignment for (non-debug) sections within object
4448 (@samp{*.o}) files, which are less strictly aligned.}.
4452 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4453 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4457 @section @code{.data @var{subsection}}
4459 @cindex @code{data} directive
4460 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4461 end of the data subsection numbered @var{subsection} (which is an
4462 absolute expression). If @var{subsection} is omitted, it defaults
4467 @section @code{.def @var{name}}
4469 @cindex @code{def} directive
4470 @cindex COFF symbols, debugging
4471 @cindex debugging COFF symbols
4472 Begin defining debugging information for a symbol @var{name}; the
4473 definition extends until the @code{.endef} directive is encountered.
4476 This directive is only observed when @command{@value{AS}} is configured for COFF
4477 format output; when producing @code{b.out}, @samp{.def} is recognized,
4484 @section @code{.desc @var{symbol}, @var{abs-expression}}
4486 @cindex @code{desc} directive
4487 @cindex COFF symbol descriptor
4488 @cindex symbol descriptor, COFF
4489 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4490 to the low 16 bits of an absolute expression.
4493 The @samp{.desc} directive is not available when @command{@value{AS}} is
4494 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4495 object format. For the sake of compatibility, @command{@value{AS}} accepts
4496 it, but produces no output, when configured for COFF.
4502 @section @code{.dim}
4504 @cindex @code{dim} directive
4505 @cindex COFF auxiliary symbol information
4506 @cindex auxiliary symbol information, COFF
4507 This directive is generated by compilers to include auxiliary debugging
4508 information in the symbol table. It is only permitted inside
4509 @code{.def}/@code{.endef} pairs.
4512 @samp{.dim} is only meaningful when generating COFF format output; when
4513 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4519 @section @code{.double @var{flonums}}
4521 @cindex @code{double} directive
4522 @cindex floating point numbers (double)
4523 @code{.double} expects zero or more flonums, separated by commas. It
4524 assembles floating point numbers.
4526 The exact kind of floating point numbers emitted depends on how
4527 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4531 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4532 in @sc{ieee} format.
4537 @section @code{.eject}
4539 @cindex @code{eject} directive
4540 @cindex new page, in listings
4541 @cindex page, in listings
4542 @cindex listing control: new page
4543 Force a page break at this point, when generating assembly listings.
4546 @section @code{.else}
4548 @cindex @code{else} directive
4549 @code{.else} is part of the @command{@value{AS}} support for conditional
4550 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4551 of code to be assembled if the condition for the preceding @code{.if}
4555 @section @code{.elseif}
4557 @cindex @code{elseif} directive
4558 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4559 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4560 @code{.if} block that would otherwise fill the entire @code{.else} section.
4563 @section @code{.end}
4565 @cindex @code{end} directive
4566 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4567 process anything in the file past the @code{.end} directive.
4571 @section @code{.endef}
4573 @cindex @code{endef} directive
4574 This directive flags the end of a symbol definition begun with
4578 @samp{.endef} is only meaningful when generating COFF format output; if
4579 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4580 directive but ignores it.
4585 @section @code{.endfunc}
4586 @cindex @code{endfunc} directive
4587 @code{.endfunc} marks the end of a function specified with @code{.func}.
4590 @section @code{.endif}
4592 @cindex @code{endif} directive
4593 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4594 it marks the end of a block of code that is only assembled
4595 conditionally. @xref{If,,@code{.if}}.
4598 @section @code{.equ @var{symbol}, @var{expression}}
4600 @cindex @code{equ} directive
4601 @cindex assigning values to symbols
4602 @cindex symbols, assigning values to
4603 This directive sets the value of @var{symbol} to @var{expression}.
4604 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4607 The syntax for @code{equ} on the HPPA is
4608 @samp{@var{symbol} .equ @var{expression}}.
4612 The syntax for @code{equ} on the Z80 is
4613 @samp{@var{symbol} equ @var{expression}}.
4614 On the Z80 it is an eror if @var{symbol} is already defined,
4615 but the symbol is not protected from later redefinition.
4616 Compare @ref{Equiv}.
4620 @section @code{.equiv @var{symbol}, @var{expression}}
4621 @cindex @code{equiv} directive
4622 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4623 the assembler will signal an error if @var{symbol} is already defined. Note a
4624 symbol which has been referenced but not actually defined is considered to be
4627 Except for the contents of the error message, this is roughly equivalent to
4634 plus it protects the symbol from later redefinition.
4637 @section @code{.eqv @var{symbol}, @var{expression}}
4638 @cindex @code{eqv} directive
4639 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4640 evaluate the expression or any part of it immediately. Instead each time
4641 the resulting symbol is used in an expression, a snapshot of its current
4645 @section @code{.err}
4646 @cindex @code{err} directive
4647 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4648 message and, unless the @option{-Z} option was used, it will not generate an
4649 object file. This can be used to signal an error in conditionally compiled code.
4652 @section @code{.error "@var{string}"}
4653 @cindex error directive
4655 Similarly to @code{.err}, this directive emits an error, but you can specify a
4656 string that will be emitted as the error message. If you don't specify the
4657 message, it defaults to @code{".error directive invoked in source file"}.
4658 @xref{Errors, ,Error and Warning Messages}.
4661 .error "This code has not been assembled and tested."
4665 @section @code{.exitm}
4666 Exit early from the current macro definition. @xref{Macro}.
4669 @section @code{.extern}
4671 @cindex @code{extern} directive
4672 @code{.extern} is accepted in the source program---for compatibility
4673 with other assemblers---but it is ignored. @command{@value{AS}} treats
4674 all undefined symbols as external.
4677 @section @code{.fail @var{expression}}
4679 @cindex @code{fail} directive
4680 Generates an error or a warning. If the value of the @var{expression} is 500
4681 or more, @command{@value{AS}} will print a warning message. If the value is less
4682 than 500, @command{@value{AS}} will print an error message. The message will
4683 include the value of @var{expression}. This can occasionally be useful inside
4684 complex nested macros or conditional assembly.
4687 @section @code{.file}
4688 @cindex @code{file} directive
4690 @ifclear no-file-dir
4691 There are two different versions of the @code{.file} directive. Targets
4692 that support DWARF2 line number information use the DWARF2 version of
4693 @code{.file}. Other targets use the default version.
4695 @subheading Default Version
4697 @cindex logical file name
4698 @cindex file name, logical
4699 This version of the @code{.file} directive tells @command{@value{AS}} that we
4700 are about to start a new logical file. The syntax is:
4706 @var{string} is the new file name. In general, the filename is
4707 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4708 to specify an empty file name, you must give the quotes--@code{""}. This
4709 statement may go away in future: it is only recognized to be compatible with
4710 old @command{@value{AS}} programs.
4712 @subheading DWARF2 Version
4715 When emitting DWARF2 line number information, @code{.file} assigns filenames
4716 to the @code{.debug_line} file name table. The syntax is:
4719 .file @var{fileno} @var{filename}
4722 The @var{fileno} operand should be a unique positive integer to use as the
4723 index of the entry in the table. The @var{filename} operand is a C string
4726 The detail of filename indices is exposed to the user because the filename
4727 table is shared with the @code{.debug_info} section of the DWARF2 debugging
4728 information, and thus the user must know the exact indices that table
4732 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4734 @cindex @code{fill} directive
4735 @cindex writing patterns in memory
4736 @cindex patterns, writing in memory
4737 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4738 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4739 may be zero or more. @var{Size} may be zero or more, but if it is
4740 more than 8, then it is deemed to have the value 8, compatible with
4741 other people's assemblers. The contents of each @var{repeat} bytes
4742 is taken from an 8-byte number. The highest order 4 bytes are
4743 zero. The lowest order 4 bytes are @var{value} rendered in the
4744 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4745 Each @var{size} bytes in a repetition is taken from the lowest order
4746 @var{size} bytes of this number. Again, this bizarre behavior is
4747 compatible with other people's assemblers.
4749 @var{size} and @var{value} are optional.
4750 If the second comma and @var{value} are absent, @var{value} is
4751 assumed zero. If the first comma and following tokens are absent,
4752 @var{size} is assumed to be 1.
4755 @section @code{.float @var{flonums}}
4757 @cindex floating point numbers (single)
4758 @cindex @code{float} directive
4759 This directive assembles zero or more flonums, separated by commas. It
4760 has the same effect as @code{.single}.
4762 The exact kind of floating point numbers emitted depends on how
4763 @command{@value{AS}} is configured.
4764 @xref{Machine Dependencies}.
4768 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4769 in @sc{ieee} format.
4774 @section @code{.func @var{name}[,@var{label}]}
4775 @cindex @code{func} directive
4776 @code{.func} emits debugging information to denote function @var{name}, and
4777 is ignored unless the file is assembled with debugging enabled.
4778 Only @samp{--gstabs[+]} is currently supported.
4779 @var{label} is the entry point of the function and if omitted @var{name}
4780 prepended with the @samp{leading char} is used.
4781 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4782 All functions are currently defined to have @code{void} return type.
4783 The function must be terminated with @code{.endfunc}.
4786 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4788 @cindex @code{global} directive
4789 @cindex symbol, making visible to linker
4790 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4791 @var{symbol} in your partial program, its value is made available to
4792 other partial programs that are linked with it. Otherwise,
4793 @var{symbol} takes its attributes from a symbol of the same name
4794 from another file linked into the same program.
4796 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4797 compatibility with other assemblers.
4800 On the HPPA, @code{.global} is not always enough to make it accessible to other
4801 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4802 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4807 @section @code{.gnu_attribute @var{tag},@var{value}}
4808 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
4811 @section @code{.hidden @var{names}}
4813 @cindex @code{hidden} directive
4815 This is one of the ELF visibility directives. The other two are
4816 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4817 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4819 This directive overrides the named symbols default visibility (which is set by
4820 their binding: local, global or weak). The directive sets the visibility to
4821 @code{hidden} which means that the symbols are not visible to other components.
4822 Such symbols are always considered to be @code{protected} as well.
4826 @section @code{.hword @var{expressions}}
4828 @cindex @code{hword} directive
4829 @cindex integers, 16-bit
4830 @cindex numbers, 16-bit
4831 @cindex sixteen bit integers
4832 This expects zero or more @var{expressions}, and emits
4833 a 16 bit number for each.
4836 This directive is a synonym for @samp{.short}; depending on the target
4837 architecture, it may also be a synonym for @samp{.word}.
4841 This directive is a synonym for @samp{.short}.
4844 This directive is a synonym for both @samp{.short} and @samp{.word}.
4849 @section @code{.ident}
4851 @cindex @code{ident} directive
4853 This directive is used by some assemblers to place tags in object files. The
4854 behavior of this directive varies depending on the target. When using the
4855 a.out object file format, @command{@value{AS}} simply accepts the directive for
4856 source-file compatibility with existing assemblers, but does not emit anything
4857 for it. When using COFF, comments are emitted to the @code{.comment} or
4858 @code{.rdata} section, depending on the target. When using ELF, comments are
4859 emitted to the @code{.comment} section.
4862 @section @code{.if @var{absolute expression}}
4864 @cindex conditional assembly
4865 @cindex @code{if} directive
4866 @code{.if} marks the beginning of a section of code which is only
4867 considered part of the source program being assembled if the argument
4868 (which must be an @var{absolute expression}) is non-zero. The end of
4869 the conditional section of code must be marked by @code{.endif}
4870 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
4871 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
4872 If you have several conditions to check, @code{.elseif} may be used to avoid
4873 nesting blocks if/else within each subsequent @code{.else} block.
4875 The following variants of @code{.if} are also supported:
4877 @cindex @code{ifdef} directive
4878 @item .ifdef @var{symbol}
4879 Assembles the following section of code if the specified @var{symbol}
4880 has been defined. Note a symbol which has been referenced but not yet defined
4881 is considered to be undefined.
4883 @cindex @code{ifb} directive
4884 @item .ifb @var{text}
4885 Assembles the following section of code if the operand is blank (empty).
4887 @cindex @code{ifc} directive
4888 @item .ifc @var{string1},@var{string2}
4889 Assembles the following section of code if the two strings are the same. The
4890 strings may be optionally quoted with single quotes. If they are not quoted,
4891 the first string stops at the first comma, and the second string stops at the
4892 end of the line. Strings which contain whitespace should be quoted. The
4893 string comparison is case sensitive.
4895 @cindex @code{ifeq} directive
4896 @item .ifeq @var{absolute expression}
4897 Assembles the following section of code if the argument is zero.
4899 @cindex @code{ifeqs} directive
4900 @item .ifeqs @var{string1},@var{string2}
4901 Another form of @code{.ifc}. The strings must be quoted using double quotes.
4903 @cindex @code{ifge} directive
4904 @item .ifge @var{absolute expression}
4905 Assembles the following section of code if the argument is greater than or
4908 @cindex @code{ifgt} directive
4909 @item .ifgt @var{absolute expression}
4910 Assembles the following section of code if the argument is greater than zero.
4912 @cindex @code{ifle} directive
4913 @item .ifle @var{absolute expression}
4914 Assembles the following section of code if the argument is less than or equal
4917 @cindex @code{iflt} directive
4918 @item .iflt @var{absolute expression}
4919 Assembles the following section of code if the argument is less than zero.
4921 @cindex @code{ifnb} directive
4922 @item .ifnb @var{text}
4923 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
4924 following section of code if the operand is non-blank (non-empty).
4926 @cindex @code{ifnc} directive
4927 @item .ifnc @var{string1},@var{string2}.
4928 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
4929 following section of code if the two strings are not the same.
4931 @cindex @code{ifndef} directive
4932 @cindex @code{ifnotdef} directive
4933 @item .ifndef @var{symbol}
4934 @itemx .ifnotdef @var{symbol}
4935 Assembles the following section of code if the specified @var{symbol}
4936 has not been defined. Both spelling variants are equivalent. Note a symbol
4937 which has been referenced but not yet defined is considered to be undefined.
4939 @cindex @code{ifne} directive
4940 @item .ifne @var{absolute expression}
4941 Assembles the following section of code if the argument is not equal to zero
4942 (in other words, this is equivalent to @code{.if}).
4944 @cindex @code{ifnes} directive
4945 @item .ifnes @var{string1},@var{string2}
4946 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
4947 following section of code if the two strings are not the same.
4951 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4953 @cindex @code{incbin} directive
4954 @cindex binary files, including
4955 The @code{incbin} directive includes @var{file} verbatim at the current
4956 location. You can control the search paths used with the @samp{-I} command-line
4957 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4960 The @var{skip} argument skips a number of bytes from the start of the
4961 @var{file}. The @var{count} argument indicates the maximum number of bytes to
4962 read. Note that the data is not aligned in any way, so it is the user's
4963 responsibility to make sure that proper alignment is provided both before and
4964 after the @code{incbin} directive.
4967 @section @code{.include "@var{file}"}
4969 @cindex @code{include} directive
4970 @cindex supporting files, including
4971 @cindex files, including
4972 This directive provides a way to include supporting files at specified
4973 points in your source program. The code from @var{file} is assembled as
4974 if it followed the point of the @code{.include}; when the end of the
4975 included file is reached, assembly of the original file continues. You
4976 can control the search paths used with the @samp{-I} command-line option
4977 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
4981 @section @code{.int @var{expressions}}
4983 @cindex @code{int} directive
4984 @cindex integers, 32-bit
4985 Expect zero or more @var{expressions}, of any section, separated by commas.
4986 For each expression, emit a number that, at run time, is the value of that
4987 expression. The byte order and bit size of the number depends on what kind
4988 of target the assembly is for.
4992 On most forms of the H8/300, @code{.int} emits 16-bit
4993 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5000 @section @code{.internal @var{names}}
5002 @cindex @code{internal} directive
5004 This is one of the ELF visibility directives. The other two are
5005 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5006 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5008 This directive overrides the named symbols default visibility (which is set by
5009 their binding: local, global or weak). The directive sets the visibility to
5010 @code{internal} which means that the symbols are considered to be @code{hidden}
5011 (i.e., not visible to other components), and that some extra, processor specific
5012 processing must also be performed upon the symbols as well.
5016 @section @code{.irp @var{symbol},@var{values}}@dots{}
5018 @cindex @code{irp} directive
5019 Evaluate a sequence of statements assigning different values to @var{symbol}.
5020 The sequence of statements starts at the @code{.irp} directive, and is
5021 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5022 set to @var{value}, and the sequence of statements is assembled. If no
5023 @var{value} is listed, the sequence of statements is assembled once, with
5024 @var{symbol} set to the null string. To refer to @var{symbol} within the
5025 sequence of statements, use @var{\symbol}.
5027 For example, assembling
5035 is equivalent to assembling
5043 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5046 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5048 @cindex @code{irpc} directive
5049 Evaluate a sequence of statements assigning different values to @var{symbol}.
5050 The sequence of statements starts at the @code{.irpc} directive, and is
5051 terminated by an @code{.endr} directive. For each character in @var{value},
5052 @var{symbol} is set to the character, and the sequence of statements is
5053 assembled. If no @var{value} is listed, the sequence of statements is
5054 assembled once, with @var{symbol} set to the null string. To refer to
5055 @var{symbol} within the sequence of statements, use @var{\symbol}.
5057 For example, assembling
5065 is equivalent to assembling
5073 For some caveats with the spelling of @var{symbol}, see also the discussion
5077 @section @code{.lcomm @var{symbol} , @var{length}}
5079 @cindex @code{lcomm} directive
5080 @cindex local common symbols
5081 @cindex symbols, local common
5082 Reserve @var{length} (an absolute expression) bytes for a local common
5083 denoted by @var{symbol}. The section and value of @var{symbol} are
5084 those of the new local common. The addresses are allocated in the bss
5085 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5086 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5087 not visible to @code{@value{LD}}.
5090 Some targets permit a third argument to be used with @code{.lcomm}. This
5091 argument specifies the desired alignment of the symbol in the bss section.
5095 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5096 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5100 @section @code{.lflags}
5102 @cindex @code{lflags} directive (ignored)
5103 @command{@value{AS}} accepts this directive, for compatibility with other
5104 assemblers, but ignores it.
5106 @ifclear no-line-dir
5108 @section @code{.line @var{line-number}}
5110 @cindex @code{line} directive
5111 @cindex logical line number
5113 Change the logical line number. @var{line-number} must be an absolute
5114 expression. The next line has that logical line number. Therefore any other
5115 statements on the current line (after a statement separator character) are
5116 reported as on logical line number @var{line-number} @minus{} 1. One day
5117 @command{@value{AS}} will no longer support this directive: it is recognized only
5118 for compatibility with existing assembler programs.
5121 Even though this is a directive associated with the @code{a.out} or
5122 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5123 when producing COFF output, and treats @samp{.line} as though it
5124 were the COFF @samp{.ln} @emph{if} it is found outside a
5125 @code{.def}/@code{.endef} pair.
5127 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5128 used by compilers to generate auxiliary symbol information for
5133 @section @code{.linkonce [@var{type}]}
5135 @cindex @code{linkonce} directive
5136 @cindex common sections
5137 Mark the current section so that the linker only includes a single copy of it.
5138 This may be used to include the same section in several different object files,
5139 but ensure that the linker will only include it once in the final output file.
5140 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5141 Duplicate sections are detected based on the section name, so it should be
5144 This directive is only supported by a few object file formats; as of this
5145 writing, the only object file format which supports it is the Portable
5146 Executable format used on Windows NT.
5148 The @var{type} argument is optional. If specified, it must be one of the
5149 following strings. For example:
5153 Not all types may be supported on all object file formats.
5157 Silently discard duplicate sections. This is the default.
5160 Warn if there are duplicate sections, but still keep only one copy.
5163 Warn if any of the duplicates have different sizes.
5166 Warn if any of the duplicates do not have exactly the same contents.
5170 @section @code{.list}
5172 @cindex @code{list} directive
5173 @cindex listing control, turning on
5174 Control (in conjunction with the @code{.nolist} directive) whether or
5175 not assembly listings are generated. These two directives maintain an
5176 internal counter (which is zero initially). @code{.list} increments the
5177 counter, and @code{.nolist} decrements it. Assembly listings are
5178 generated whenever the counter is greater than zero.
5180 By default, listings are disabled. When you enable them (with the
5181 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5182 the initial value of the listing counter is one.
5185 @section @code{.ln @var{line-number}}
5187 @cindex @code{ln} directive
5188 @ifclear no-line-dir
5189 @samp{.ln} is a synonym for @samp{.line}.
5192 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5193 must be an absolute expression. The next line has that logical
5194 line number, so any other statements on the current line (after a
5195 statement separator character @code{;}) are reported as on logical
5196 line number @var{line-number} @minus{} 1.
5199 This directive is accepted, but ignored, when @command{@value{AS}} is
5200 configured for @code{b.out}; its effect is only associated with COFF
5206 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5207 @cindex @code{loc} directive
5208 When emitting DWARF2 line number information,
5209 the @code{.loc} directive will add a row to the @code{.debug_line} line
5210 number matrix corresponding to the immediately following assembly
5211 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5212 arguments will be applied to the @code{.debug_line} state machine before
5215 The @var{options} are a sequence of the following tokens in any order:
5219 This option will set the @code{basic_block} register in the
5220 @code{.debug_line} state machine to @code{true}.
5223 This option will set the @code{prologue_end} register in the
5224 @code{.debug_line} state machine to @code{true}.
5226 @item epilogue_begin
5227 This option will set the @code{epilogue_begin} register in the
5228 @code{.debug_line} state machine to @code{true}.
5230 @item is_stmt @var{value}
5231 This option will set the @code{is_stmt} register in the
5232 @code{.debug_line} state machine to @code{value}, which must be
5235 @item isa @var{value}
5236 This directive will set the @code{isa} register in the @code{.debug_line}
5237 state machine to @var{value}, which must be an unsigned integer.
5239 @item discriminator @var{value}
5240 This directive will set the @code{discriminator} register in the @code{.debug_line}
5241 state machine to @var{value}, which must be an unsigned integer.
5245 @node Loc_mark_labels
5246 @section @code{.loc_mark_labels @var{enable}}
5247 @cindex @code{loc_mark_labels} directive
5248 When emitting DWARF2 line number information,
5249 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5250 to the @code{.debug_line} line number matrix with the @code{basic_block}
5251 register in the state machine set whenever a code label is seen.
5252 The @var{enable} argument should be either 1 or 0, to enable or disable
5253 this function respectively.
5257 @section @code{.local @var{names}}
5259 @cindex @code{local} directive
5260 This directive, which is available for ELF targets, marks each symbol in
5261 the comma-separated list of @code{names} as a local symbol so that it
5262 will not be externally visible. If the symbols do not already exist,
5263 they will be created.
5265 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5266 accept an alignment argument, which is the case for most ELF targets,
5267 the @code{.local} directive can be used in combination with @code{.comm}
5268 (@pxref{Comm}) to define aligned local common data.
5272 @section @code{.long @var{expressions}}
5274 @cindex @code{long} directive
5275 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5278 @c no one seems to know what this is for or whether this description is
5279 @c what it really ought to do
5281 @section @code{.lsym @var{symbol}, @var{expression}}
5283 @cindex @code{lsym} directive
5284 @cindex symbol, not referenced in assembly
5285 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5286 the hash table, ensuring it cannot be referenced by name during the
5287 rest of the assembly. This sets the attributes of the symbol to be
5288 the same as the expression value:
5290 @var{other} = @var{descriptor} = 0
5291 @var{type} = @r{(section of @var{expression})}
5292 @var{value} = @var{expression}
5295 The new symbol is not flagged as external.
5299 @section @code{.macro}
5302 The commands @code{.macro} and @code{.endm} allow you to define macros that
5303 generate assembly output. For example, this definition specifies a macro
5304 @code{sum} that puts a sequence of numbers into memory:
5307 .macro sum from=0, to=5
5316 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5328 @item .macro @var{macname}
5329 @itemx .macro @var{macname} @var{macargs} @dots{}
5330 @cindex @code{macro} directive
5331 Begin the definition of a macro called @var{macname}. If your macro
5332 definition requires arguments, specify their names after the macro name,
5333 separated by commas or spaces. You can qualify the macro argument to
5334 indicate whether all invocations must specify a non-blank value (through
5335 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5336 (through @samp{:@code{vararg}}). You can supply a default value for any
5337 macro argument by following the name with @samp{=@var{deflt}}. You
5338 cannot define two macros with the same @var{macname} unless it has been
5339 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5340 definitions. For example, these are all valid @code{.macro} statements:
5344 Begin the definition of a macro called @code{comm}, which takes no
5347 @item .macro plus1 p, p1
5348 @itemx .macro plus1 p p1
5349 Either statement begins the definition of a macro called @code{plus1},
5350 which takes two arguments; within the macro definition, write
5351 @samp{\p} or @samp{\p1} to evaluate the arguments.
5353 @item .macro reserve_str p1=0 p2
5354 Begin the definition of a macro called @code{reserve_str}, with two
5355 arguments. The first argument has a default value, but not the second.
5356 After the definition is complete, you can call the macro either as
5357 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5358 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5359 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5360 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5362 @item .macro m p1:req, p2=0, p3:vararg
5363 Begin the definition of a macro called @code{m}, with at least three
5364 arguments. The first argument must always have a value specified, but
5365 not the second, which instead has a default value. The third formal
5366 will get assigned all remaining arguments specified at invocation time.
5368 When you call a macro, you can specify the argument values either by
5369 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5370 @samp{sum to=17, from=9}.
5374 Note that since each of the @var{macargs} can be an identifier exactly
5375 as any other one permitted by the target architecture, there may be
5376 occasional problems if the target hand-crafts special meanings to certain
5377 characters when they occur in a special position. For example, if the colon
5378 (@code{:}) is generally permitted to be part of a symbol name, but the
5379 architecture specific code special-cases it when occurring as the final
5380 character of a symbol (to denote a label), then the macro parameter
5381 replacement code will have no way of knowing that and consider the whole
5382 construct (including the colon) an identifier, and check only this
5383 identifier for being the subject to parameter substitution. So for example
5384 this macro definition:
5392 might not work as expected. Invoking @samp{label foo} might not create a label
5393 called @samp{foo} but instead just insert the text @samp{\l:} into the
5394 assembler source, probably generating an error about an unrecognised
5397 Similarly problems might occur with the period character (@samp{.})
5398 which is often allowed inside opcode names (and hence identifier names). So
5399 for example constructing a macro to build an opcode from a base name and a
5400 length specifier like this:
5403 .macro opcode base length
5408 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5409 instruction but instead generate some kind of error as the assembler tries to
5410 interpret the text @samp{\base.\length}.
5412 There are several possible ways around this problem:
5415 @item Insert white space
5416 If it is possible to use white space characters then this is the simplest
5425 @item Use @samp{\()}
5426 The string @samp{\()} can be used to separate the end of a macro argument from
5427 the following text. eg:
5430 .macro opcode base length
5435 @item Use the alternate macro syntax mode
5436 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5437 used as a separator. eg:
5447 Note: this problem of correctly identifying string parameters to pseudo ops
5448 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5449 and @code{.irpc} (@pxref{Irpc}) as well.
5452 @cindex @code{endm} directive
5453 Mark the end of a macro definition.
5456 @cindex @code{exitm} directive
5457 Exit early from the current macro definition.
5459 @cindex number of macros executed
5460 @cindex macros, count executed
5462 @command{@value{AS}} maintains a counter of how many macros it has
5463 executed in this pseudo-variable; you can copy that number to your
5464 output with @samp{\@@}, but @emph{only within a macro definition}.
5466 @item LOCAL @var{name} [ , @dots{} ]
5467 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5468 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5469 @xref{Altmacro,,@code{.altmacro}}.
5473 @section @code{.mri @var{val}}
5475 @cindex @code{mri} directive
5476 @cindex MRI mode, temporarily
5477 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5478 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5479 affects code assembled until the next @code{.mri} directive, or until the end
5480 of the file. @xref{M, MRI mode, MRI mode}.
5483 @section @code{.noaltmacro}
5484 Disable alternate macro mode. @xref{Altmacro}.
5487 @section @code{.nolist}
5489 @cindex @code{nolist} directive
5490 @cindex listing control, turning off
5491 Control (in conjunction with the @code{.list} directive) whether or
5492 not assembly listings are generated. These two directives maintain an
5493 internal counter (which is zero initially). @code{.list} increments the
5494 counter, and @code{.nolist} decrements it. Assembly listings are
5495 generated whenever the counter is greater than zero.
5498 @section @code{.octa @var{bignums}}
5500 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5501 @cindex @code{octa} directive
5502 @cindex integer, 16-byte
5503 @cindex sixteen byte integer
5504 This directive expects zero or more bignums, separated by commas. For each
5505 bignum, it emits a 16-byte integer.
5507 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5508 hence @emph{octa}-word for 16 bytes.
5511 @section @code{.org @var{new-lc} , @var{fill}}
5513 @cindex @code{org} directive
5514 @cindex location counter, advancing
5515 @cindex advancing location counter
5516 @cindex current address, advancing
5517 Advance the location counter of the current section to
5518 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5519 expression with the same section as the current subsection. That is,
5520 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5521 wrong section, the @code{.org} directive is ignored. To be compatible
5522 with former assemblers, if the section of @var{new-lc} is absolute,
5523 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5524 is the same as the current subsection.
5526 @code{.org} may only increase the location counter, or leave it
5527 unchanged; you cannot use @code{.org} to move the location counter
5530 @c double negative used below "not undefined" because this is a specific
5531 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5533 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5534 may not be undefined. If you really detest this restriction we eagerly await
5535 a chance to share your improved assembler.
5537 Beware that the origin is relative to the start of the section, not
5538 to the start of the subsection. This is compatible with other
5539 people's assemblers.
5541 When the location counter (of the current subsection) is advanced, the
5542 intervening bytes are filled with @var{fill} which should be an
5543 absolute expression. If the comma and @var{fill} are omitted,
5544 @var{fill} defaults to zero.
5547 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5549 @cindex padding the location counter given a power of two
5550 @cindex @code{p2align} directive
5551 Pad the location counter (in the current subsection) to a particular
5552 storage boundary. The first expression (which must be absolute) is the
5553 number of low-order zero bits the location counter must have after
5554 advancement. For example @samp{.p2align 3} advances the location
5555 counter until it a multiple of 8. If the location counter is already a
5556 multiple of 8, no change is needed.
5558 The second expression (also absolute) gives the fill value to be stored in the
5559 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5560 padding bytes are normally zero. However, on some systems, if the section is
5561 marked as containing code and the fill value is omitted, the space is filled
5562 with no-op instructions.
5564 The third expression is also absolute, and is also optional. If it is present,
5565 it is the maximum number of bytes that should be skipped by this alignment
5566 directive. If doing the alignment would require skipping more bytes than the
5567 specified maximum, then the alignment is not done at all. You can omit the
5568 fill value (the second argument) entirely by simply using two commas after the
5569 required alignment; this can be useful if you want the alignment to be filled
5570 with no-op instructions when appropriate.
5572 @cindex @code{p2alignw} directive
5573 @cindex @code{p2alignl} directive
5574 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5575 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5576 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5577 fill pattern as a four byte longword value. For example, @code{.p2alignw
5578 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5579 filled in with the value 0x368d (the exact placement of the bytes depends upon
5580 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5585 @section @code{.popsection}
5587 @cindex @code{popsection} directive
5588 @cindex Section Stack
5589 This is one of the ELF section stack manipulation directives. The others are
5590 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5591 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5594 This directive replaces the current section (and subsection) with the top
5595 section (and subsection) on the section stack. This section is popped off the
5601 @section @code{.previous}
5603 @cindex @code{previous} directive
5604 @cindex Section Stack
5605 This is one of the ELF section stack manipulation directives. The others are
5606 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5607 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5608 (@pxref{PopSection}).
5610 This directive swaps the current section (and subsection) with most recently
5611 referenced section/subsection pair prior to this one. Multiple
5612 @code{.previous} directives in a row will flip between two sections (and their
5613 subsections). For example:
5625 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5631 # Now in section A subsection 1
5635 # Now in section B subsection 0
5638 # Now in section B subsection 1
5641 # Now in section B subsection 0
5645 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5646 section B and 0x9abc into subsection 1 of section B.
5648 In terms of the section stack, this directive swaps the current section with
5649 the top section on the section stack.
5653 @section @code{.print @var{string}}
5655 @cindex @code{print} directive
5656 @command{@value{AS}} will print @var{string} on the standard output during
5657 assembly. You must put @var{string} in double quotes.
5661 @section @code{.protected @var{names}}
5663 @cindex @code{protected} directive
5665 This is one of the ELF visibility directives. The other two are
5666 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5668 This directive overrides the named symbols default visibility (which is set by
5669 their binding: local, global or weak). The directive sets the visibility to
5670 @code{protected} which means that any references to the symbols from within the
5671 components that defines them must be resolved to the definition in that
5672 component, even if a definition in another component would normally preempt
5677 @section @code{.psize @var{lines} , @var{columns}}
5679 @cindex @code{psize} directive
5680 @cindex listing control: paper size
5681 @cindex paper size, for listings
5682 Use this directive to declare the number of lines---and, optionally, the
5683 number of columns---to use for each page, when generating listings.
5685 If you do not use @code{.psize}, listings use a default line-count
5686 of 60. You may omit the comma and @var{columns} specification; the
5687 default width is 200 columns.
5689 @command{@value{AS}} generates formfeeds whenever the specified number of
5690 lines is exceeded (or whenever you explicitly request one, using
5693 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5694 those explicitly specified with @code{.eject}.
5697 @section @code{.purgem @var{name}}
5699 @cindex @code{purgem} directive
5700 Undefine the macro @var{name}, so that later uses of the string will not be
5701 expanded. @xref{Macro}.
5705 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5707 @cindex @code{pushsection} directive
5708 @cindex Section Stack
5709 This is one of the ELF section stack manipulation directives. The others are
5710 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5711 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5714 This directive pushes the current section (and subsection) onto the
5715 top of the section stack, and then replaces the current section and
5716 subsection with @code{name} and @code{subsection}. The optional
5717 @code{flags}, @code{type} and @code{arguments} are treated the same
5718 as in the @code{.section} (@pxref{Section}) directive.
5722 @section @code{.quad @var{bignums}}
5724 @cindex @code{quad} directive
5725 @code{.quad} expects zero or more bignums, separated by commas. For
5726 each bignum, it emits
5728 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5729 warning message; and just takes the lowest order 8 bytes of the bignum.
5730 @cindex eight-byte integer
5731 @cindex integer, 8-byte
5733 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5734 hence @emph{quad}-word for 8 bytes.
5737 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5738 warning message; and just takes the lowest order 16 bytes of the bignum.
5739 @cindex sixteen-byte integer
5740 @cindex integer, 16-byte
5744 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5746 @cindex @code{reloc} directive
5747 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5748 @var{expression}. If @var{offset} is a number, the relocation is generated in
5749 the current section. If @var{offset} is an expression that resolves to a
5750 symbol plus offset, the relocation is generated in the given symbol's section.
5751 @var{expression}, if present, must resolve to a symbol plus addend or to an
5752 absolute value, but note that not all targets support an addend. e.g. ELF REL
5753 targets such as i386 store an addend in the section contents rather than in the
5754 relocation. This low level interface does not support addends stored in the
5758 @section @code{.rept @var{count}}
5760 @cindex @code{rept} directive
5761 Repeat the sequence of lines between the @code{.rept} directive and the next
5762 @code{.endr} directive @var{count} times.
5764 For example, assembling
5772 is equivalent to assembling
5781 @section @code{.sbttl "@var{subheading}"}
5783 @cindex @code{sbttl} directive
5784 @cindex subtitles for listings
5785 @cindex listing control: subtitle
5786 Use @var{subheading} as the title (third line, immediately after the
5787 title line) when generating assembly listings.
5789 This directive affects subsequent pages, as well as the current page if
5790 it appears within ten lines of the top of a page.
5794 @section @code{.scl @var{class}}
5796 @cindex @code{scl} directive
5797 @cindex symbol storage class (COFF)
5798 @cindex COFF symbol storage class
5799 Set the storage-class value for a symbol. This directive may only be
5800 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5801 whether a symbol is static or external, or it may record further
5802 symbolic debugging information.
5805 The @samp{.scl} directive is primarily associated with COFF output; when
5806 configured to generate @code{b.out} output format, @command{@value{AS}}
5807 accepts this directive but ignores it.
5813 @section @code{.section @var{name}}
5815 @cindex named section
5816 Use the @code{.section} directive to assemble the following code into a section
5819 This directive is only supported for targets that actually support arbitrarily
5820 named sections; on @code{a.out} targets, for example, it is not accepted, even
5821 with a standard @code{a.out} section name.
5825 @c only print the extra heading if both COFF and ELF are set
5826 @subheading COFF Version
5829 @cindex @code{section} directive (COFF version)
5830 For COFF targets, the @code{.section} directive is used in one of the following
5834 .section @var{name}[, "@var{flags}"]
5835 .section @var{name}[, @var{subsection}]
5838 If the optional argument is quoted, it is taken as flags to use for the
5839 section. Each flag is a single character. The following flags are recognized:
5842 bss section (uninitialized data)
5844 section is not loaded
5854 shared section (meaningful for PE targets)
5856 ignored. (For compatibility with the ELF version)
5858 section is not readable (meaningful for PE targets)
5860 single-digit power-of-two section alignment (GNU extension)
5863 If no flags are specified, the default flags depend upon the section name. If
5864 the section name is not recognized, the default will be for the section to be
5865 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
5866 from the section, rather than adding them, so if they are used on their own it
5867 will be as if no flags had been specified at all.
5869 If the optional argument to the @code{.section} directive is not quoted, it is
5870 taken as a subsection number (@pxref{Sub-Sections}).
5875 @c only print the extra heading if both COFF and ELF are set
5876 @subheading ELF Version
5879 @cindex Section Stack
5880 This is one of the ELF section stack manipulation directives. The others are
5881 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
5882 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
5883 @code{.previous} (@pxref{Previous}).
5885 @cindex @code{section} directive (ELF version)
5886 For ELF targets, the @code{.section} directive is used like this:
5889 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
5892 The optional @var{flags} argument is a quoted string which may contain any
5893 combination of the following characters:
5896 section is allocatable
5898 section is excluded from executable and shared library.
5902 section is executable
5904 section is mergeable
5906 section contains zero terminated strings
5908 section is a member of a section group
5910 section is used for thread-local-storage
5912 section is a member of the previously-current section's group, if any
5915 The optional @var{type} argument may contain one of the following constants:
5918 section contains data
5920 section does not contain data (i.e., section only occupies space)
5922 section contains data which is used by things other than the program
5924 section contains an array of pointers to init functions
5926 section contains an array of pointers to finish functions
5927 @item @@preinit_array
5928 section contains an array of pointers to pre-init functions
5931 Many targets only support the first three section types.
5933 Note on targets where the @code{@@} character is the start of a comment (eg
5934 ARM) then another character is used instead. For example the ARM port uses the
5937 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
5938 be specified as well as an extra argument---@var{entsize}---like this:
5941 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
5944 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
5945 constants, each @var{entsize} octets long. Sections with both @code{M} and
5946 @code{S} must contain zero terminated strings where each character is
5947 @var{entsize} bytes long. The linker may remove duplicates within sections with
5948 the same name, same entity size and same flags. @var{entsize} must be an
5949 absolute expression. For sections with both @code{M} and @code{S}, a string
5950 which is a suffix of a larger string is considered a duplicate. Thus
5951 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
5952 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
5954 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
5955 be present along with an additional field like this:
5958 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
5961 The @var{GroupName} field specifies the name of the section group to which this
5962 particular section belongs. The optional linkage field can contain:
5965 indicates that only one copy of this section should be retained
5970 Note: if both the @var{M} and @var{G} flags are present then the fields for
5971 the Merge flag should come first, like this:
5974 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
5977 If @var{flags} contains the @code{?} symbol then it may not also contain the
5978 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
5979 present. Instead, @code{?} says to consider the section that's current before
5980 this directive. If that section used @code{G}, then the new section will use
5981 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
5982 If not, then the @code{?} symbol has no effect.
5984 If no flags are specified, the default flags depend upon the section name. If
5985 the section name is not recognized, the default will be for the section to have
5986 none of the above flags: it will not be allocated in memory, nor writable, nor
5987 executable. The section will contain data.
5989 For ELF targets, the assembler supports another type of @code{.section}
5990 directive for compatibility with the Solaris assembler:
5993 .section "@var{name}"[, @var{flags}...]
5996 Note that the section name is quoted. There may be a sequence of comma
6000 section is allocatable
6004 section is executable
6006 section is excluded from executable and shared library.
6008 section is used for thread local storage
6011 This directive replaces the current section and subsection. See the
6012 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6013 some examples of how this directive and the other section stack directives
6019 @section @code{.set @var{symbol}, @var{expression}}
6021 @cindex @code{set} directive
6022 @cindex symbol value, setting
6023 Set the value of @var{symbol} to @var{expression}. This
6024 changes @var{symbol}'s value and type to conform to
6025 @var{expression}. If @var{symbol} was flagged as external, it remains
6026 flagged (@pxref{Symbol Attributes}).
6028 You may @code{.set} a symbol many times in the same assembly.
6030 If you @code{.set} a global symbol, the value stored in the object
6031 file is the last value stored into it.
6034 On Z80 @code{set} is a real instruction, use
6035 @samp{@var{symbol} defl @var{expression}} instead.
6039 @section @code{.short @var{expressions}}
6041 @cindex @code{short} directive
6043 @code{.short} is normally the same as @samp{.word}.
6044 @xref{Word,,@code{.word}}.
6046 In some configurations, however, @code{.short} and @code{.word} generate
6047 numbers of different lengths. @xref{Machine Dependencies}.
6051 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6054 This expects zero or more @var{expressions}, and emits
6055 a 16 bit number for each.
6060 @section @code{.single @var{flonums}}
6062 @cindex @code{single} directive
6063 @cindex floating point numbers (single)
6064 This directive assembles zero or more flonums, separated by commas. It
6065 has the same effect as @code{.float}.
6067 The exact kind of floating point numbers emitted depends on how
6068 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6072 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6073 numbers in @sc{ieee} format.
6079 @section @code{.size}
6081 This directive is used to set the size associated with a symbol.
6085 @c only print the extra heading if both COFF and ELF are set
6086 @subheading COFF Version
6089 @cindex @code{size} directive (COFF version)
6090 For COFF targets, the @code{.size} directive is only permitted inside
6091 @code{.def}/@code{.endef} pairs. It is used like this:
6094 .size @var{expression}
6098 @samp{.size} is only meaningful when generating COFF format output; when
6099 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6106 @c only print the extra heading if both COFF and ELF are set
6107 @subheading ELF Version
6110 @cindex @code{size} directive (ELF version)
6111 For ELF targets, the @code{.size} directive is used like this:
6114 .size @var{name} , @var{expression}
6117 This directive sets the size associated with a symbol @var{name}.
6118 The size in bytes is computed from @var{expression} which can make use of label
6119 arithmetic. This directive is typically used to set the size of function
6124 @ifclear no-space-dir
6126 @section @code{.skip @var{size} , @var{fill}}
6128 @cindex @code{skip} directive
6129 @cindex filling memory
6130 This directive emits @var{size} bytes, each of value @var{fill}. Both
6131 @var{size} and @var{fill} are absolute expressions. If the comma and
6132 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6137 @section @code{.sleb128 @var{expressions}}
6139 @cindex @code{sleb128} directive
6140 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6141 compact, variable length representation of numbers used by the DWARF
6142 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6144 @ifclear no-space-dir
6146 @section @code{.space @var{size} , @var{fill}}
6148 @cindex @code{space} directive
6149 @cindex filling memory
6150 This directive emits @var{size} bytes, each of value @var{fill}. Both
6151 @var{size} and @var{fill} are absolute expressions. If the comma
6152 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6157 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6158 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6159 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6160 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6168 @section @code{.stabd, .stabn, .stabs}
6170 @cindex symbolic debuggers, information for
6171 @cindex @code{stab@var{x}} directives
6172 There are three directives that begin @samp{.stab}.
6173 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6174 The symbols are not entered in the @command{@value{AS}} hash table: they
6175 cannot be referenced elsewhere in the source file.
6176 Up to five fields are required:
6180 This is the symbol's name. It may contain any character except
6181 @samp{\000}, so is more general than ordinary symbol names. Some
6182 debuggers used to code arbitrarily complex structures into symbol names
6186 An absolute expression. The symbol's type is set to the low 8 bits of
6187 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6188 and debuggers choke on silly bit patterns.
6191 An absolute expression. The symbol's ``other'' attribute is set to the
6192 low 8 bits of this expression.
6195 An absolute expression. The symbol's descriptor is set to the low 16
6196 bits of this expression.
6199 An absolute expression which becomes the symbol's value.
6202 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6203 or @code{.stabs} statement, the symbol has probably already been created;
6204 you get a half-formed symbol in your object file. This is
6205 compatible with earlier assemblers!
6208 @cindex @code{stabd} directive
6209 @item .stabd @var{type} , @var{other} , @var{desc}
6211 The ``name'' of the symbol generated is not even an empty string.
6212 It is a null pointer, for compatibility. Older assemblers used a
6213 null pointer so they didn't waste space in object files with empty
6216 The symbol's value is set to the location counter,
6217 relocatably. When your program is linked, the value of this symbol
6218 is the address of the location counter when the @code{.stabd} was
6221 @cindex @code{stabn} directive
6222 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6223 The name of the symbol is set to the empty string @code{""}.
6225 @cindex @code{stabs} directive
6226 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6227 All five fields are specified.
6233 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6234 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6236 @cindex string, copying to object file
6237 @cindex string8, copying to object file
6238 @cindex string16, copying to object file
6239 @cindex string32, copying to object file
6240 @cindex string64, copying to object file
6241 @cindex @code{string} directive
6242 @cindex @code{string8} directive
6243 @cindex @code{string16} directive
6244 @cindex @code{string32} directive
6245 @cindex @code{string64} directive
6247 Copy the characters in @var{str} to the object file. You may specify more than
6248 one string to copy, separated by commas. Unless otherwise specified for a
6249 particular machine, the assembler marks the end of each string with a 0 byte.
6250 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6252 The variants @code{string16}, @code{string32} and @code{string64} differ from
6253 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6254 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6255 are stored in target endianness byte order.
6261 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6262 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6267 @section @code{.struct @var{expression}}
6269 @cindex @code{struct} directive
6270 Switch to the absolute section, and set the section offset to @var{expression},
6271 which must be an absolute expression. You might use this as follows:
6280 This would define the symbol @code{field1} to have the value 0, the symbol
6281 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6282 value 8. Assembly would be left in the absolute section, and you would need to
6283 use a @code{.section} directive of some sort to change to some other section
6284 before further assembly.
6288 @section @code{.subsection @var{name}}
6290 @cindex @code{subsection} directive
6291 @cindex Section Stack
6292 This is one of the ELF section stack manipulation directives. The others are
6293 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6294 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6297 This directive replaces the current subsection with @code{name}. The current
6298 section is not changed. The replaced subsection is put onto the section stack
6299 in place of the then current top of stack subsection.
6304 @section @code{.symver}
6305 @cindex @code{symver} directive
6306 @cindex symbol versioning
6307 @cindex versions of symbols
6308 Use the @code{.symver} directive to bind symbols to specific version nodes
6309 within a source file. This is only supported on ELF platforms, and is
6310 typically used when assembling files to be linked into a shared library.
6311 There are cases where it may make sense to use this in objects to be bound
6312 into an application itself so as to override a versioned symbol from a
6315 For ELF targets, the @code{.symver} directive can be used like this:
6317 .symver @var{name}, @var{name2@@nodename}
6319 If the symbol @var{name} is defined within the file
6320 being assembled, the @code{.symver} directive effectively creates a symbol
6321 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6322 just don't try and create a regular alias is that the @var{@@} character isn't
6323 permitted in symbol names. The @var{name2} part of the name is the actual name
6324 of the symbol by which it will be externally referenced. The name @var{name}
6325 itself is merely a name of convenience that is used so that it is possible to
6326 have definitions for multiple versions of a function within a single source
6327 file, and so that the compiler can unambiguously know which version of a
6328 function is being mentioned. The @var{nodename} portion of the alias should be
6329 the name of a node specified in the version script supplied to the linker when
6330 building a shared library. If you are attempting to override a versioned
6331 symbol from a shared library, then @var{nodename} should correspond to the
6332 nodename of the symbol you are trying to override.
6334 If the symbol @var{name} is not defined within the file being assembled, all
6335 references to @var{name} will be changed to @var{name2@@nodename}. If no
6336 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6339 Another usage of the @code{.symver} directive is:
6341 .symver @var{name}, @var{name2@@@@nodename}
6343 In this case, the symbol @var{name} must exist and be defined within
6344 the file being assembled. It is similar to @var{name2@@nodename}. The
6345 difference is @var{name2@@@@nodename} will also be used to resolve
6346 references to @var{name2} by the linker.
6348 The third usage of the @code{.symver} directive is:
6350 .symver @var{name}, @var{name2@@@@@@nodename}
6352 When @var{name} is not defined within the
6353 file being assembled, it is treated as @var{name2@@nodename}. When
6354 @var{name} is defined within the file being assembled, the symbol
6355 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6360 @section @code{.tag @var{structname}}
6362 @cindex COFF structure debugging
6363 @cindex structure debugging, COFF
6364 @cindex @code{tag} directive
6365 This directive is generated by compilers to include auxiliary debugging
6366 information in the symbol table. It is only permitted inside
6367 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6368 definitions in the symbol table with instances of those structures.
6371 @samp{.tag} is only used when generating COFF format output; when
6372 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6378 @section @code{.text @var{subsection}}
6380 @cindex @code{text} directive
6381 Tells @command{@value{AS}} to assemble the following statements onto the end of
6382 the text subsection numbered @var{subsection}, which is an absolute
6383 expression. If @var{subsection} is omitted, subsection number zero
6387 @section @code{.title "@var{heading}"}
6389 @cindex @code{title} directive
6390 @cindex listing control: title line
6391 Use @var{heading} as the title (second line, immediately after the
6392 source file name and pagenumber) when generating assembly listings.
6394 This directive affects subsequent pages, as well as the current page if
6395 it appears within ten lines of the top of a page.
6399 @section @code{.type}
6401 This directive is used to set the type of a symbol.
6405 @c only print the extra heading if both COFF and ELF are set
6406 @subheading COFF Version
6409 @cindex COFF symbol type
6410 @cindex symbol type, COFF
6411 @cindex @code{type} directive (COFF version)
6412 For COFF targets, this directive is permitted only within
6413 @code{.def}/@code{.endef} pairs. It is used like this:
6419 This records the integer @var{int} as the type attribute of a symbol table
6423 @samp{.type} is associated only with COFF format output; when
6424 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6425 directive but ignores it.
6431 @c only print the extra heading if both COFF and ELF are set
6432 @subheading ELF Version
6435 @cindex ELF symbol type
6436 @cindex symbol type, ELF
6437 @cindex @code{type} directive (ELF version)
6438 For ELF targets, the @code{.type} directive is used like this:
6441 .type @var{name} , @var{type description}
6444 This sets the type of symbol @var{name} to be either a
6445 function symbol or an object symbol. There are five different syntaxes
6446 supported for the @var{type description} field, in order to provide
6447 compatibility with various other assemblers.
6449 Because some of the characters used in these syntaxes (such as @samp{@@} and
6450 @samp{#}) are comment characters for some architectures, some of the syntaxes
6451 below do not work on all architectures. The first variant will be accepted by
6452 the GNU assembler on all architectures so that variant should be used for
6453 maximum portability, if you do not need to assemble your code with other
6456 The syntaxes supported are:
6459 .type <name> STT_<TYPE_IN_UPPER_CASE>
6460 .type <name>,#<type>
6461 .type <name>,@@<type>
6462 .type <name>,%<type>
6463 .type <name>,"<type>"
6466 The types supported are:
6471 Mark the symbol as being a function name.
6474 @itemx gnu_indirect_function
6475 Mark the symbol as an indirect function when evaluated during reloc
6476 processing. (This is only supported on Linux targeted assemblers).
6480 Mark the symbol as being a data object.
6484 Mark the symbol as being a thead-local data object.
6488 Mark the symbol as being a common data object.
6492 Does not mark the symbol in any way. It is supported just for completeness.
6494 @item gnu_unique_object
6495 Marks the symbol as being a globally unique data object. The dynamic linker
6496 will make sure that in the entire process there is just one symbol with this
6497 name and type in use. (This is only supported on Linux targeted assemblers).
6501 Note: Some targets support extra types in addition to those listed above.
6507 @section @code{.uleb128 @var{expressions}}
6509 @cindex @code{uleb128} directive
6510 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6511 compact, variable length representation of numbers used by the DWARF
6512 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6516 @section @code{.val @var{addr}}
6518 @cindex @code{val} directive
6519 @cindex COFF value attribute
6520 @cindex value attribute, COFF
6521 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6522 records the address @var{addr} as the value attribute of a symbol table
6526 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6527 configured for @code{b.out}, it accepts this directive but ignores it.
6533 @section @code{.version "@var{string}"}
6535 @cindex @code{version} directive
6536 This directive creates a @code{.note} section and places into it an ELF
6537 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6542 @section @code{.vtable_entry @var{table}, @var{offset}}
6544 @cindex @code{vtable_entry} directive
6545 This directive finds or creates a symbol @code{table} and creates a
6546 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6549 @section @code{.vtable_inherit @var{child}, @var{parent}}
6551 @cindex @code{vtable_inherit} directive
6552 This directive finds the symbol @code{child} and finds or creates the symbol
6553 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6554 parent whose addend is the value of the child symbol. As a special case the
6555 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6559 @section @code{.warning "@var{string}"}
6560 @cindex warning directive
6561 Similar to the directive @code{.error}
6562 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6565 @section @code{.weak @var{names}}
6567 @cindex @code{weak} directive
6568 This directive sets the weak attribute on the comma separated list of symbol
6569 @code{names}. If the symbols do not already exist, they will be created.
6571 On COFF targets other than PE, weak symbols are a GNU extension. This
6572 directive sets the weak attribute on the comma separated list of symbol
6573 @code{names}. If the symbols do not already exist, they will be created.
6575 On the PE target, weak symbols are supported natively as weak aliases.
6576 When a weak symbol is created that is not an alias, GAS creates an
6577 alternate symbol to hold the default value.
6580 @section @code{.weakref @var{alias}, @var{target}}
6582 @cindex @code{weakref} directive
6583 This directive creates an alias to the target symbol that enables the symbol to
6584 be referenced with weak-symbol semantics, but without actually making it weak.
6585 If direct references or definitions of the symbol are present, then the symbol
6586 will not be weak, but if all references to it are through weak references, the
6587 symbol will be marked as weak in the symbol table.
6589 The effect is equivalent to moving all references to the alias to a separate
6590 assembly source file, renaming the alias to the symbol in it, declaring the
6591 symbol as weak there, and running a reloadable link to merge the object files
6592 resulting from the assembly of the new source file and the old source file that
6593 had the references to the alias removed.
6595 The alias itself never makes to the symbol table, and is entirely handled
6596 within the assembler.
6599 @section @code{.word @var{expressions}}
6601 @cindex @code{word} directive
6602 This directive expects zero or more @var{expressions}, of any section,
6603 separated by commas.
6606 For each expression, @command{@value{AS}} emits a 32-bit number.
6609 For each expression, @command{@value{AS}} emits a 16-bit number.
6614 The size of the number emitted, and its byte order,
6615 depend on what target computer the assembly is for.
6618 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6619 @c happen---32-bit addressability, period; no long/short jumps.
6620 @ifset DIFF-TBL-KLUGE
6621 @cindex difference tables altered
6622 @cindex altered difference tables
6624 @emph{Warning: Special Treatment to support Compilers}
6628 Machines with a 32-bit address space, but that do less than 32-bit
6629 addressing, require the following special treatment. If the machine of
6630 interest to you does 32-bit addressing (or doesn't require it;
6631 @pxref{Machine Dependencies}), you can ignore this issue.
6634 In order to assemble compiler output into something that works,
6635 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6636 Directives of the form @samp{.word sym1-sym2} are often emitted by
6637 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6638 directive of the form @samp{.word sym1-sym2}, and the difference between
6639 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6640 creates a @dfn{secondary jump table}, immediately before the next label.
6641 This secondary jump table is preceded by a short-jump to the
6642 first byte after the secondary table. This short-jump prevents the flow
6643 of control from accidentally falling into the new table. Inside the
6644 table is a long-jump to @code{sym2}. The original @samp{.word}
6645 contains @code{sym1} minus the address of the long-jump to
6648 If there were several occurrences of @samp{.word sym1-sym2} before the
6649 secondary jump table, all of them are adjusted. If there was a
6650 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6651 long-jump to @code{sym4} is included in the secondary jump table,
6652 and the @code{.word} directives are adjusted to contain @code{sym3}
6653 minus the address of the long-jump to @code{sym4}; and so on, for as many
6654 entries in the original jump table as necessary.
6657 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6658 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6659 assembly language programmers.
6662 @c end DIFF-TBL-KLUGE
6665 @section Deprecated Directives
6667 @cindex deprecated directives
6668 @cindex obsolescent directives
6669 One day these directives won't work.
6670 They are included for compatibility with older assemblers.
6677 @node Object Attributes
6678 @chapter Object Attributes
6679 @cindex object attributes
6681 @command{@value{AS}} assembles source files written for a specific architecture
6682 into object files for that architecture. But not all object files are alike.
6683 Many architectures support incompatible variations. For instance, floating
6684 point arguments might be passed in floating point registers if the object file
6685 requires hardware floating point support---or floating point arguments might be
6686 passed in integer registers if the object file supports processors with no
6687 hardware floating point unit. Or, if two objects are built for different
6688 generations of the same architecture, the combination may require the
6689 newer generation at run-time.
6691 This information is useful during and after linking. At link time,
6692 @command{@value{LD}} can warn about incompatible object files. After link
6693 time, tools like @command{gdb} can use it to process the linked file
6696 Compatibility information is recorded as a series of object attributes. Each
6697 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6698 string, and indicates who sets the meaning of the tag. The tag is an integer,
6699 and indicates what property the attribute describes. The value may be a string
6700 or an integer, and indicates how the property affects this object. Missing
6701 attributes are the same as attributes with a zero value or empty string value.
6703 Object attributes were developed as part of the ABI for the ARM Architecture.
6704 The file format is documented in @cite{ELF for the ARM Architecture}.
6707 * GNU Object Attributes:: @sc{gnu} Object Attributes
6708 * Defining New Object Attributes:: Defining New Object Attributes
6711 @node GNU Object Attributes
6712 @section @sc{gnu} Object Attributes
6714 The @code{.gnu_attribute} directive records an object attribute
6715 with vendor @samp{gnu}.
6717 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6718 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6719 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6720 2} is set for architecture-independent attributes and clear for
6721 architecture-dependent ones.
6723 @subsection Common @sc{gnu} attributes
6725 These attributes are valid on all architectures.
6728 @item Tag_compatibility (32)
6729 The compatibility attribute takes an integer flag value and a vendor name. If
6730 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6731 then the file is only compatible with the named toolchain. If it is greater
6732 than 1, the file can only be processed by other toolchains under some private
6733 arrangement indicated by the flag value and the vendor name.
6736 @subsection MIPS Attributes
6739 @item Tag_GNU_MIPS_ABI_FP (4)
6740 The floating-point ABI used by this object file. The value will be:
6744 0 for files not affected by the floating-point ABI.
6746 1 for files using the hardware floating-point with a standard double-precision
6749 2 for files using the hardware floating-point ABI with a single-precision FPU.
6751 3 for files using the software floating-point ABI.
6753 4 for files using the hardware floating-point ABI with 64-bit wide
6754 double-precision floating-point registers and 32-bit wide general
6759 @subsection PowerPC Attributes
6762 @item Tag_GNU_Power_ABI_FP (4)
6763 The floating-point ABI used by this object file. The value will be:
6767 0 for files not affected by the floating-point ABI.
6769 1 for files using double-precision hardware floating-point ABI.
6771 2 for files using the software floating-point ABI.
6773 3 for files using single-precision hardware floating-point ABI.
6776 @item Tag_GNU_Power_ABI_Vector (8)
6777 The vector ABI used by this object file. The value will be:
6781 0 for files not affected by the vector ABI.
6783 1 for files using general purpose registers to pass vectors.
6785 2 for files using AltiVec registers to pass vectors.
6787 3 for files using SPE registers to pass vectors.
6791 @node Defining New Object Attributes
6792 @section Defining New Object Attributes
6794 If you want to define a new @sc{gnu} object attribute, here are the places you
6795 will need to modify. New attributes should be discussed on the @samp{binutils}
6800 This manual, which is the official register of attributes.
6802 The header for your architecture @file{include/elf}, to define the tag.
6804 The @file{bfd} support file for your architecture, to merge the attribute
6805 and issue any appropriate link warnings.
6807 Test cases in @file{ld/testsuite} for merging and link warnings.
6809 @file{binutils/readelf.c} to display your attribute.
6811 GCC, if you want the compiler to mark the attribute automatically.
6817 @node Machine Dependencies
6818 @chapter Machine Dependent Features
6820 @cindex machine dependencies
6821 The machine instruction sets are (almost by definition) different on
6822 each machine where @command{@value{AS}} runs. Floating point representations
6823 vary as well, and @command{@value{AS}} often supports a few additional
6824 directives or command-line options for compatibility with other
6825 assemblers on a particular platform. Finally, some versions of
6826 @command{@value{AS}} support special pseudo-instructions for branch
6829 This chapter discusses most of these differences, though it does not
6830 include details on any machine's instruction set. For details on that
6831 subject, see the hardware manufacturer's manual.
6835 * Alpha-Dependent:: Alpha Dependent Features
6838 * ARC-Dependent:: ARC Dependent Features
6841 * ARM-Dependent:: ARM Dependent Features
6844 * AVR-Dependent:: AVR Dependent Features
6847 * Blackfin-Dependent:: Blackfin Dependent Features
6850 * CR16-Dependent:: CR16 Dependent Features
6853 * CRIS-Dependent:: CRIS Dependent Features
6856 * D10V-Dependent:: D10V Dependent Features
6859 * D30V-Dependent:: D30V Dependent Features
6862 * H8/300-Dependent:: Renesas H8/300 Dependent Features
6865 * HPPA-Dependent:: HPPA Dependent Features
6868 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
6871 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
6874 * i860-Dependent:: Intel 80860 Dependent Features
6877 * i960-Dependent:: Intel 80960 Dependent Features
6880 * IA-64-Dependent:: Intel IA-64 Dependent Features
6883 * IP2K-Dependent:: IP2K Dependent Features
6886 * LM32-Dependent:: LM32 Dependent Features
6889 * M32C-Dependent:: M32C Dependent Features
6892 * M32R-Dependent:: M32R Dependent Features
6895 * M68K-Dependent:: M680x0 Dependent Features
6898 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
6901 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
6904 * MIPS-Dependent:: MIPS Dependent Features
6907 * MMIX-Dependent:: MMIX Dependent Features
6910 * MSP430-Dependent:: MSP430 Dependent Features
6913 * SH-Dependent:: Renesas / SuperH SH Dependent Features
6914 * SH64-Dependent:: SuperH SH64 Dependent Features
6917 * PDP-11-Dependent:: PDP-11 Dependent Features
6920 * PJ-Dependent:: picoJava Dependent Features
6923 * PPC-Dependent:: PowerPC Dependent Features
6926 * RX-Dependent:: RX Dependent Features
6929 * S/390-Dependent:: IBM S/390 Dependent Features
6932 * SCORE-Dependent:: SCORE Dependent Features
6935 * Sparc-Dependent:: SPARC Dependent Features
6938 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
6941 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
6944 * V850-Dependent:: V850 Dependent Features
6947 * Xtensa-Dependent:: Xtensa Dependent Features
6950 * Z80-Dependent:: Z80 Dependent Features
6953 * Z8000-Dependent:: Z8000 Dependent Features
6956 * Vax-Dependent:: VAX Dependent Features
6963 @c The following major nodes are *sections* in the GENERIC version, *chapters*
6964 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
6965 @c peculiarity: to preserve cross-references, there must be a node called
6966 @c "Machine Dependencies". Hence the conditional nodenames in each
6967 @c major node below. Node defaulting in makeinfo requires adjacency of
6968 @c node and sectioning commands; hence the repetition of @chapter BLAH
6969 @c in both conditional blocks.
6972 @include c-alpha.texi
6988 @include c-bfin.texi
6992 @include c-cr16.texi
6996 @include c-cris.texi
7001 @node Machine Dependencies
7002 @chapter Machine Dependent Features
7004 The machine instruction sets are different on each Renesas chip family,
7005 and there are also some syntax differences among the families. This
7006 chapter describes the specific @command{@value{AS}} features for each
7010 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7011 * SH-Dependent:: Renesas SH Dependent Features
7018 @include c-d10v.texi
7022 @include c-d30v.texi
7026 @include c-h8300.texi
7030 @include c-hppa.texi
7034 @include c-i370.texi
7038 @include c-i386.texi
7042 @include c-i860.texi
7046 @include c-i960.texi
7050 @include c-ia64.texi
7054 @include c-ip2k.texi
7058 @include c-lm32.texi
7062 @include c-m32c.texi
7066 @include c-m32r.texi
7070 @include c-m68k.texi
7074 @include c-m68hc11.texi
7078 @include c-microblaze.texi
7082 @include c-mips.texi
7086 @include c-mmix.texi
7090 @include c-msp430.texi
7094 @include c-ns32k.texi
7098 @include c-pdp11.texi
7114 @include c-s390.texi
7118 @include c-score.texi
7123 @include c-sh64.texi
7127 @include c-sparc.texi
7131 @include c-tic54x.texi
7135 @include c-tic6x.texi
7151 @include c-v850.texi
7155 @include c-xtensa.texi
7159 @c reverse effect of @down at top of generic Machine-Dep chapter
7163 @node Reporting Bugs
7164 @chapter Reporting Bugs
7165 @cindex bugs in assembler
7166 @cindex reporting bugs in assembler
7168 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7170 Reporting a bug may help you by bringing a solution to your problem, or it may
7171 not. But in any case the principal function of a bug report is to help the
7172 entire community by making the next version of @command{@value{AS}} work better.
7173 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7175 In order for a bug report to serve its purpose, you must include the
7176 information that enables us to fix the bug.
7179 * Bug Criteria:: Have you found a bug?
7180 * Bug Reporting:: How to report bugs
7184 @section Have You Found a Bug?
7185 @cindex bug criteria
7187 If you are not sure whether you have found a bug, here are some guidelines:
7190 @cindex fatal signal
7191 @cindex assembler crash
7192 @cindex crash of assembler
7194 If the assembler gets a fatal signal, for any input whatever, that is a
7195 @command{@value{AS}} bug. Reliable assemblers never crash.
7197 @cindex error on valid input
7199 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7201 @cindex invalid input
7203 If @command{@value{AS}} does not produce an error message for invalid input, that
7204 is a bug. However, you should note that your idea of ``invalid input'' might
7205 be our idea of ``an extension'' or ``support for traditional practice''.
7208 If you are an experienced user of assemblers, your suggestions for improvement
7209 of @command{@value{AS}} are welcome in any case.
7213 @section How to Report Bugs
7215 @cindex assembler bugs, reporting
7217 A number of companies and individuals offer support for @sc{gnu} products. If
7218 you obtained @command{@value{AS}} from a support organization, we recommend you
7219 contact that organization first.
7221 You can find contact information for many support companies and
7222 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7226 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7230 The fundamental principle of reporting bugs usefully is this:
7231 @strong{report all the facts}. If you are not sure whether to state a
7232 fact or leave it out, state it!
7234 Often people omit facts because they think they know what causes the problem
7235 and assume that some details do not matter. Thus, you might assume that the
7236 name of a symbol you use in an example does not matter. Well, probably it does
7237 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7238 happens to fetch from the location where that name is stored in memory;
7239 perhaps, if the name were different, the contents of that location would fool
7240 the assembler into doing the right thing despite the bug. Play it safe and
7241 give a specific, complete example. That is the easiest thing for you to do,
7242 and the most helpful.
7244 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7245 it is new to us. Therefore, always write your bug reports on the assumption
7246 that the bug has not been reported previously.
7248 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7249 bell?'' This cannot help us fix a bug, so it is basically useless. We
7250 respond by asking for enough details to enable us to investigate.
7251 You might as well expedite matters by sending them to begin with.
7253 To enable us to fix the bug, you should include all these things:
7257 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7258 it with the @samp{--version} argument.
7260 Without this, we will not know whether there is any point in looking for
7261 the bug in the current version of @command{@value{AS}}.
7264 Any patches you may have applied to the @command{@value{AS}} source.
7267 The type of machine you are using, and the operating system name and
7271 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7275 The command arguments you gave the assembler to assemble your example and
7276 observe the bug. To guarantee you will not omit something important, list them
7277 all. A copy of the Makefile (or the output from make) is sufficient.
7279 If we were to try to guess the arguments, we would probably guess wrong
7280 and then we might not encounter the bug.
7283 A complete input file that will reproduce the bug. If the bug is observed when
7284 the assembler is invoked via a compiler, send the assembler source, not the
7285 high level language source. Most compilers will produce the assembler source
7286 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7287 the options @samp{-v --save-temps}; this will save the assembler source in a
7288 file with an extension of @file{.s}, and also show you exactly how
7289 @command{@value{AS}} is being run.
7292 A description of what behavior you observe that you believe is
7293 incorrect. For example, ``It gets a fatal signal.''
7295 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7296 will certainly notice it. But if the bug is incorrect output, we might not
7297 notice unless it is glaringly wrong. You might as well not give us a chance to
7300 Even if the problem you experience is a fatal signal, you should still say so
7301 explicitly. Suppose something strange is going on, such as, your copy of
7302 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7303 library on your system. (This has happened!) Your copy might crash and ours
7304 would not. If you told us to expect a crash, then when ours fails to crash, we
7305 would know that the bug was not happening for us. If you had not told us to
7306 expect a crash, then we would not be able to draw any conclusion from our
7310 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7311 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7312 option. Always send diffs from the old file to the new file. If you even
7313 discuss something in the @command{@value{AS}} source, refer to it by context, not
7316 The line numbers in our development sources will not match those in your
7317 sources. Your line numbers would convey no useful information to us.
7320 Here are some things that are not necessary:
7324 A description of the envelope of the bug.
7326 Often people who encounter a bug spend a lot of time investigating
7327 which changes to the input file will make the bug go away and which
7328 changes will not affect it.
7330 This is often time consuming and not very useful, because the way we
7331 will find the bug is by running a single example under the debugger
7332 with breakpoints, not by pure deduction from a series of examples.
7333 We recommend that you save your time for something else.
7335 Of course, if you can find a simpler example to report @emph{instead}
7336 of the original one, that is a convenience for us. Errors in the
7337 output will be easier to spot, running under the debugger will take
7338 less time, and so on.
7340 However, simplification is not vital; if you do not want to do this,
7341 report the bug anyway and send us the entire test case you used.
7344 A patch for the bug.
7346 A patch for the bug does help us if it is a good one. But do not omit
7347 the necessary information, such as the test case, on the assumption that
7348 a patch is all we need. We might see problems with your patch and decide
7349 to fix the problem another way, or we might not understand it at all.
7351 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7352 construct an example that will make the program follow a certain path through
7353 the code. If you do not send us the example, we will not be able to construct
7354 one, so we will not be able to verify that the bug is fixed.
7356 And if we cannot understand what bug you are trying to fix, or why your
7357 patch should be an improvement, we will not install it. A test case will
7358 help us to understand.
7361 A guess about what the bug is or what it depends on.
7363 Such guesses are usually wrong. Even we cannot guess right about such
7364 things without first using the debugger to find the facts.
7367 @node Acknowledgements
7368 @chapter Acknowledgements
7370 If you have contributed to GAS and your name isn't listed here,
7371 it is not meant as a slight. We just don't know about it. Send mail to the
7372 maintainer, and we'll correct the situation. Currently
7374 the maintainer is Ken Raeburn (email address @code{raeburn@@cygnus.com}).
7376 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7379 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7380 information and the 68k series machines, most of the preprocessing pass, and
7381 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7383 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7384 many bug fixes, including merging support for several processors, breaking GAS
7385 up to handle multiple object file format back ends (including heavy rewrite,
7386 testing, an integration of the coff and b.out back ends), adding configuration
7387 including heavy testing and verification of cross assemblers and file splits
7388 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7389 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7390 port (including considerable amounts of reverse engineering), a SPARC opcode
7391 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7392 assertions and made them work, much other reorganization, cleanup, and lint.
7394 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7395 in format-specific I/O modules.
7397 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7398 has done much work with it since.
7400 The Intel 80386 machine description was written by Eliot Dresselhaus.
7402 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7404 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7405 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7407 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7408 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7409 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7410 support a.out format.
7412 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7413 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7414 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7415 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7418 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7419 simplified the configuration of which versions accept which directives. He
7420 updated the 68k machine description so that Motorola's opcodes always produced
7421 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7422 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7423 cross-compilation support, and one bug in relaxation that took a week and
7424 required the proverbial one-bit fix.
7426 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7427 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7428 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7429 PowerPC assembler, and made a few other minor patches.
7431 Steve Chamberlain made GAS able to generate listings.
7433 Hewlett-Packard contributed support for the HP9000/300.
7435 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7436 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7437 formats). This work was supported by both the Center for Software Science at
7438 the University of Utah and Cygnus Support.
7440 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7441 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7442 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7443 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7444 and some initial 64-bit support).
7446 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7448 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7449 support for openVMS/Alpha.
7451 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7454 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7455 Inc.@: added support for Xtensa processors.
7457 Several engineers at Cygnus Support have also provided many small bug fixes and
7458 configuration enhancements.
7460 Jon Beniston added support for the Lattice Mico32 architecture.
7462 Many others have contributed large or small bugfixes and enhancements. If
7463 you have contributed significant work and are not mentioned on this list, and
7464 want to be, let us know. Some of the history has been lost; we are not
7465 intentionally leaving anyone out.
7467 @node GNU Free Documentation License
7468 @appendix GNU Free Documentation License
7472 @unnumbered AS Index
projects / binutils.git / blob