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, 2011, 2012
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, 2011 Free Software Foundation,
109 Permission is granted to copy, distribute and/or modify this document
110 under the terms of the GNU Free Documentation License, Version 1.3
111 or any later version published by the Free Software Foundation;
112 with no Invariant Sections, with no Front-Cover Texts, and with no
113 Back-Cover Texts. A copy of the license is included in the
114 section entitled ``GNU Free Documentation License''.
120 @title Using @value{AS}
121 @subtitle The @sc{gnu} Assembler
123 @subtitle for the @value{TARGET} family
125 @ifset VERSION_PACKAGE
127 @subtitle @value{VERSION_PACKAGE}
130 @subtitle Version @value{VERSION}
133 The Free Software Foundation Inc.@: thanks The Nice Computer
134 Company of Australia for loaning Dean Elsner to write the
135 first (Vax) version of @command{as} for Project @sc{gnu}.
136 The proprietors, management and staff of TNCCA thank FSF for
137 distracting the boss while they got some work
140 @author Dean Elsner, Jay Fenlason & friends
144 \hfill {\it Using {\tt @value{AS}}}\par
145 \hfill Edited by Cygnus Support\par
147 %"boxit" macro for figures:
148 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
149 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
150 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
151 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
152 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
155 @vskip 0pt plus 1filll
156 Copyright @copyright{} 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
157 2000, 2001, 2002, 2006, 2007, 2008, 2009, 2010, 2011 Free Software Foundation,
160 Permission is granted to copy, distribute and/or modify this document
161 under the terms of the GNU Free Documentation License, Version 1.3
162 or any later version published by the Free Software Foundation;
163 with no Invariant Sections, with no Front-Cover Texts, and with no
164 Back-Cover Texts. A copy of the license is included in the
165 section entitled ``GNU Free Documentation License''.
172 @top Using @value{AS}
174 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
175 @ifset VERSION_PACKAGE
176 @value{VERSION_PACKAGE}
178 version @value{VERSION}.
180 This version of the file describes @command{@value{AS}} configured to generate
181 code for @value{TARGET} architectures.
184 This document is distributed under the terms of the GNU Free
185 Documentation License. A copy of the license is included in the
186 section entitled ``GNU Free Documentation License''.
189 * Overview:: Overview
190 * Invoking:: Command-Line Options
192 * Sections:: Sections and Relocation
194 * Expressions:: Expressions
195 * Pseudo Ops:: Assembler Directives
197 * Object Attributes:: Object Attributes
199 * Machine Dependencies:: Machine Dependent Features
200 * Reporting Bugs:: Reporting Bugs
201 * Acknowledgements:: Who Did What
202 * GNU Free Documentation License:: GNU Free Documentation License
203 * AS Index:: AS Index
210 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
212 This version of the manual describes @command{@value{AS}} configured to generate
213 code for @value{TARGET} architectures.
217 @cindex invocation summary
218 @cindex option summary
219 @cindex summary of options
220 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
221 see @ref{Invoking,,Command-Line Options}.
223 @c man title AS the portable GNU assembler.
227 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
231 @c We don't use deffn and friends for the following because they seem
232 @c to be limited to one line for the header.
234 @c man begin SYNOPSIS
235 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
236 [@b{--compress-debug-sections}] [@b{--nocompress-debug-sections}]
237 [@b{--debug-prefix-map} @var{old}=@var{new}]
238 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
239 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
240 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
241 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
242 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}] [@b{-o}
243 @var{objfile}] [@b{-R}] [@b{--reduce-memory-overheads}] [@b{--statistics}]
244 [@b{-v}] [@b{-version}] [@b{--version}] [@b{-W}] [@b{--warn}]
245 [@b{--fatal-warnings}] [@b{-w}] [@b{-x}] [@b{-Z}] [@b{@@@var{FILE}}]
246 [@b{--size-check=[error|warning]}]
247 [@b{--target-help}] [@var{target-options}]
248 [@b{--}|@var{files} @dots{}]
250 @c Target dependent options are listed below. Keep the list sorted.
251 @c Add an empty line for separation.
254 @emph{Target AArch64 options:}
259 @emph{Target Alpha options:}
261 [@b{-mdebug} | @b{-no-mdebug}]
262 [@b{-replace} | @b{-noreplace}]
263 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
264 [@b{-F}] [@b{-32addr}]
268 @emph{Target ARC options:}
274 @emph{Target ARM options:}
275 @c Don't document the deprecated options
276 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
277 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
278 [@b{-mfpu}=@var{floating-point-format}]
279 [@b{-mfloat-abi}=@var{abi}]
280 [@b{-meabi}=@var{ver}]
283 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
284 @b{-mapcs-reentrant}]
285 [@b{-mthumb-interwork}] [@b{-k}]
289 @emph{Target Blackfin options:}
290 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
297 @emph{Target CRIS options:}
298 [@b{--underscore} | @b{--no-underscore}]
300 [@b{--emulation=criself} | @b{--emulation=crisaout}]
301 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
302 @c Deprecated -- deliberately not documented.
307 @emph{Target D10V options:}
312 @emph{Target D30V options:}
313 [@b{-O}|@b{-n}|@b{-N}]
317 @emph{Target EPIPHANY options:}
318 [@b{-mepiphany}|@b{-mepiphany16}]
322 @emph{Target H8/300 options:}
326 @c HPPA has no machine-dependent assembler options (yet).
330 @emph{Target i386 options:}
331 [@b{--32}|@b{--x32}|@b{--64}] [@b{-n}]
332 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
336 @emph{Target i960 options:}
337 @c see md_parse_option in tc-i960.c
338 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
340 [@b{-b}] [@b{-no-relax}]
344 @emph{Target IA-64 options:}
345 [@b{-mconstant-gp}|@b{-mauto-pic}]
346 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
348 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
349 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
350 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
351 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
355 @emph{Target IP2K options:}
356 [@b{-mip2022}|@b{-mip2022ext}]
360 @emph{Target M32C options:}
361 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
365 @emph{Target M32R options:}
366 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
371 @emph{Target M680X0 options:}
372 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
376 @emph{Target M68HC11 options:}
377 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}|@b{-mm9s12x}|@b{-mm9s12xg}]
378 [@b{-mshort}|@b{-mlong}]
379 [@b{-mshort-double}|@b{-mlong-double}]
380 [@b{--force-long-branches}] [@b{--short-branches}]
381 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
382 [@b{--print-opcodes}] [@b{--generate-example}]
386 @emph{Target MCORE options:}
387 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
388 [@b{-mcpu=[210|340]}]
392 @emph{Target Meta options:}
393 [@b{-mcpu=@var{cpu}}] [@b{-mfpu=@var{cpu}}] [@b{-mdsp=@var{cpu}}]
396 @emph{Target MICROBLAZE options:}
397 @c MicroBlaze has no machine-dependent assembler options.
401 @emph{Target MIPS options:}
402 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
403 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
404 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
405 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
406 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
407 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
408 [@b{-mips64}] [@b{-mips64r2}]
409 [@b{-construct-floats}] [@b{-no-construct-floats}]
410 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
411 [@b{-mips16}] [@b{-no-mips16}]
412 [@b{-mmicromips}] [@b{-mno-micromips}]
413 [@b{-msmartmips}] [@b{-mno-smartmips}]
414 [@b{-mips3d}] [@b{-no-mips3d}]
415 [@b{-mdmx}] [@b{-no-mdmx}]
416 [@b{-mdsp}] [@b{-mno-dsp}]
417 [@b{-mdspr2}] [@b{-mno-dspr2}]
418 [@b{-mmt}] [@b{-mno-mt}]
419 [@b{-mmcu}] [@b{-mno-mcu}]
420 [@b{-mfix7000}] [@b{-mno-fix7000}]
421 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
422 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
423 [@b{-mdebug}] [@b{-no-mdebug}]
424 [@b{-mpdr}] [@b{-mno-pdr}]
428 @emph{Target MMIX options:}
429 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
430 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
431 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
432 [@b{--linker-allocated-gregs}]
436 @emph{Target Nios II options:}
437 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
442 @emph{Target PDP11 options:}
443 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
444 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
445 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
449 @emph{Target picoJava options:}
454 @emph{Target PowerPC options:}
456 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
457 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mppc64}|
458 @b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|@b{-me6500}|@b{-mppc64bridge}|
459 @b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|@b{-mpower6}|@b{-mpwr6}|
460 @b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-ma2}|@b{-mcell}|@b{-mspe}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
461 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
462 [@b{-mregnames}|@b{-mno-regnames}]
463 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
464 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
465 [@b{-msolaris}|@b{-mno-solaris}]
466 [@b{-nops=@var{count}}]
470 @emph{Target RX options:}
471 [@b{-mlittle-endian}|@b{-mbig-endian}]
472 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
473 [@b{-muse-conventional-section-names}]
474 [@b{-msmall-data-limit}]
477 [@b{-mint-register=@var{number}}]
478 [@b{-mgcc-abi}|@b{-mrx-abi}]
482 @emph{Target s390 options:}
483 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
484 [@b{-mregnames}|@b{-mno-regnames}]
485 [@b{-mwarn-areg-zero}]
489 @emph{Target SCORE options:}
490 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
491 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
492 [@b{-march=score7}][@b{-march=score3}]
493 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
497 @emph{Target SPARC options:}
498 @c The order here is important. See c-sparc.texi.
499 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Asparclet}|@b{-Asparclite}
500 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av9}|@b{-Av9a}]
501 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}] [@b{-bump}]
506 @emph{Target TIC54X options:}
507 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
508 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
513 @emph{Target TIC6X options:}
514 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
515 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
516 [@b{-mpic}|@b{-mno-pic}]
520 @emph{Target TILE-Gx options:}
521 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
524 @c TILEPro has no machine-dependent assembler options
529 @emph{Target Xtensa options:}
530 [@b{--[no-]text-section-literals}] [@b{--[no-]absolute-literals}]
531 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
532 [@b{--[no-]transform}]
533 [@b{--rename-section} @var{oldname}=@var{newname}]
538 @emph{Target Z80 options:}
539 [@b{-z80}] [@b{-r800}]
540 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
541 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
542 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
543 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
544 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
545 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
549 @c Z8000 has no machine-dependent assembler options
558 @include at-file.texi
561 Turn on listings, in any of a variety of ways:
565 omit false conditionals
568 omit debugging directives
571 include general information, like @value{AS} version and options passed
574 include high-level source
580 include macro expansions
583 omit forms processing
589 set the name of the listing file
592 You may combine these options; for example, use @samp{-aln} for assembly
593 listing without forms processing. The @samp{=file} option, if used, must be
594 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
597 Begin in alternate macro mode.
599 @xref{Altmacro,,@code{.altmacro}}.
602 @item --compress-debug-sections
603 Compress DWARF debug sections using zlib. The debug sections are renamed
604 to begin with @samp{.zdebug}, and the resulting object file may not be
605 compatible with older linkers and object file utilities.
607 @item --nocompress-debug-sections
608 Do not compress DWARF debug sections. This is the default.
611 Ignored. This option is accepted for script compatibility with calls to
614 @item --debug-prefix-map @var{old}=@var{new}
615 When assembling files in directory @file{@var{old}}, record debugging
616 information describing them as in @file{@var{new}} instead.
618 @item --defsym @var{sym}=@var{value}
619 Define the symbol @var{sym} to be @var{value} before assembling the input file.
620 @var{value} must be an integer constant. As in C, a leading @samp{0x}
621 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
622 value. The value of the symbol can be overridden inside a source file via the
623 use of a @code{.set} pseudo-op.
626 ``fast''---skip whitespace and comment preprocessing (assume source is
631 Generate debugging information for each assembler source line using whichever
632 debug format is preferred by the target. This currently means either STABS,
636 Generate stabs debugging information for each assembler line. This
637 may help debugging assembler code, if the debugger can handle it.
640 Generate stabs debugging information for each assembler line, with GNU
641 extensions that probably only gdb can handle, and that could make other
642 debuggers crash or refuse to read your program. This
643 may help debugging assembler code. Currently the only GNU extension is
644 the location of the current working directory at assembling time.
647 Generate DWARF2 debugging information for each assembler line. This
648 may help debugging assembler code, if the debugger can handle it. Note---this
649 option is only supported by some targets, not all of them.
651 @item --size-check=error
652 @itemx --size-check=warning
653 Issue an error or warning for invalid ELF .size directive.
656 Print a summary of the command line options and exit.
659 Print a summary of all target specific options and exit.
662 Add directory @var{dir} to the search list for @code{.include} directives.
665 Don't warn about signed overflow.
668 @ifclear DIFF-TBL-KLUGE
669 This option is accepted but has no effect on the @value{TARGET} family.
671 @ifset DIFF-TBL-KLUGE
672 Issue warnings when difference tables altered for long displacements.
677 Keep (in the symbol table) local symbols. These symbols start with
678 system-specific local label prefixes, typically @samp{.L} for ELF systems
679 or @samp{L} for traditional a.out systems.
684 @item --listing-lhs-width=@var{number}
685 Set the maximum width, in words, of the output data column for an assembler
686 listing to @var{number}.
688 @item --listing-lhs-width2=@var{number}
689 Set the maximum width, in words, of the output data column for continuation
690 lines in an assembler listing to @var{number}.
692 @item --listing-rhs-width=@var{number}
693 Set the maximum width of an input source line, as displayed in a listing, to
696 @item --listing-cont-lines=@var{number}
697 Set the maximum number of lines printed in a listing for a single line of input
700 @item -o @var{objfile}
701 Name the object-file output from @command{@value{AS}} @var{objfile}.
704 Fold the data section into the text section.
706 @kindex --hash-size=@var{number}
707 Set the default size of GAS's hash tables to a prime number close to
708 @var{number}. Increasing this value can reduce the length of time it takes the
709 assembler to perform its tasks, at the expense of increasing the assembler's
710 memory requirements. Similarly reducing this value can reduce the memory
711 requirements at the expense of speed.
713 @item --reduce-memory-overheads
714 This option reduces GAS's memory requirements, at the expense of making the
715 assembly processes slower. Currently this switch is a synonym for
716 @samp{--hash-size=4051}, but in the future it may have other effects as well.
719 Print the maximum space (in bytes) and total time (in seconds) used by
722 @item --strip-local-absolute
723 Remove local absolute symbols from the outgoing symbol table.
727 Print the @command{as} version.
730 Print the @command{as} version and exit.
734 Suppress warning messages.
736 @item --fatal-warnings
737 Treat warnings as errors.
740 Don't suppress warning messages or treat them as errors.
749 Generate an object file even after errors.
751 @item -- | @var{files} @dots{}
752 Standard input, or source files to assemble.
760 @xref{AArch64 Options}, for the options available when @value{AS} is configured
761 for the 64-bit mode of the ARM Architecture (AArch64).
766 The following options are available when @value{AS} is configured for the
767 64-bit mode of the ARM Architecture (AArch64).
770 @include c-aarch64.texi
771 @c ended inside the included file
779 @xref{Alpha Options}, for the options available when @value{AS} is configured
780 for an Alpha processor.
785 The following options are available when @value{AS} is configured for an Alpha
789 @include c-alpha.texi
790 @c ended inside the included file
797 The following options are available when @value{AS} is configured for
802 This option selects the core processor variant.
804 Select either big-endian (-EB) or little-endian (-EL) output.
809 The following options are available when @value{AS} is configured for the ARM
813 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
814 Specify which ARM processor variant is the target.
815 @item -march=@var{architecture}[+@var{extension}@dots{}]
816 Specify which ARM architecture variant is used by the target.
817 @item -mfpu=@var{floating-point-format}
818 Select which Floating Point architecture is the target.
819 @item -mfloat-abi=@var{abi}
820 Select which floating point ABI is in use.
822 Enable Thumb only instruction decoding.
823 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
824 Select which procedure calling convention is in use.
826 Select either big-endian (-EB) or little-endian (-EL) output.
827 @item -mthumb-interwork
828 Specify that the code has been generated with interworking between Thumb and
831 Specify that PIC code has been generated.
839 @xref{Blackfin Options}, for the options available when @value{AS} is
840 configured for the Blackfin processor family.
845 The following options are available when @value{AS} is configured for
846 the Blackfin processor family.
850 @c ended inside the included file
857 See the info pages for documentation of the CRIS-specific options.
861 The following options are available when @value{AS} is configured for
864 @cindex D10V optimization
865 @cindex optimization, D10V
867 Optimize output by parallelizing instructions.
872 The following options are available when @value{AS} is configured for a D30V
875 @cindex D30V optimization
876 @cindex optimization, D30V
878 Optimize output by parallelizing instructions.
882 Warn when nops are generated.
884 @cindex D30V nops after 32-bit multiply
886 Warn when a nop after a 32-bit multiply instruction is generated.
892 The following options are available when @value{AS} is configured for the
893 Adapteva EPIPHANY series.
896 @xref{Epiphany Options}, for the options available when @value{AS} is
897 configured for an Epiphany processor.
902 The following options are available when @value{AS} is configured for
903 an Epiphany processor.
906 @include c-epiphany.texi
907 @c ended inside the included file
915 @xref{i386-Options}, for the options available when @value{AS} is
916 configured for an i386 processor.
921 The following options are available when @value{AS} is configured for
926 @c ended inside the included file
933 The following options are available when @value{AS} is configured for the
934 Intel 80960 processor.
937 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
938 Specify which variant of the 960 architecture is the target.
941 Add code to collect statistics about branches taken.
944 Do not alter compare-and-branch instructions for long displacements;
951 The following options are available when @value{AS} is configured for the
957 Specifies that the extended IP2022 instructions are allowed.
960 Restores the default behaviour, which restricts the permitted instructions to
961 just the basic IP2022 ones.
967 The following options are available when @value{AS} is configured for the
968 Renesas M32C and M16C processors.
973 Assemble M32C instructions.
976 Assemble M16C instructions (the default).
979 Enable support for link-time relaxations.
982 Support H'00 style hex constants in addition to 0x00 style.
988 The following options are available when @value{AS} is configured for the
989 Renesas M32R (formerly Mitsubishi M32R) series.
994 Specify which processor in the M32R family is the target. The default
995 is normally the M32R, but this option changes it to the M32RX.
997 @item --warn-explicit-parallel-conflicts or --Wp
998 Produce warning messages when questionable parallel constructs are
1001 @item --no-warn-explicit-parallel-conflicts or --Wnp
1002 Do not produce warning messages when questionable parallel constructs are
1009 The following options are available when @value{AS} is configured for the
1010 Motorola 68000 series.
1015 Shorten references to undefined symbols, to one word instead of two.
1017 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1018 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1019 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1020 Specify what processor in the 68000 family is the target. The default
1021 is normally the 68020, but this can be changed at configuration time.
1023 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1024 The target machine does (or does not) have a floating-point coprocessor.
1025 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1026 the basic 68000 is not compatible with the 68881, a combination of the
1027 two can be specified, since it's possible to do emulation of the
1028 coprocessor instructions with the main processor.
1030 @item -m68851 | -mno-68851
1031 The target machine does (or does not) have a memory-management
1032 unit coprocessor. The default is to assume an MMU for 68020 and up.
1040 @xref{Nios II Options}, for the options available when @value{AS} is configured
1041 for an Altera Nios II processor.
1045 @c man begin OPTIONS
1046 The following options are available when @value{AS} is configured for an
1047 Altera Nios II processor.
1049 @c man begin INCLUDE
1050 @include c-nios2.texi
1051 @c ended inside the included file
1057 For details about the PDP-11 machine dependent features options,
1058 see @ref{PDP-11-Options}.
1061 @item -mpic | -mno-pic
1062 Generate position-independent (or position-dependent) code. The
1063 default is @option{-mpic}.
1066 @itemx -mall-extensions
1067 Enable all instruction set extensions. This is the default.
1069 @item -mno-extensions
1070 Disable all instruction set extensions.
1072 @item -m@var{extension} | -mno-@var{extension}
1073 Enable (or disable) a particular instruction set extension.
1076 Enable the instruction set extensions supported by a particular CPU, and
1077 disable all other extensions.
1079 @item -m@var{machine}
1080 Enable the instruction set extensions supported by a particular machine
1081 model, and disable all other extensions.
1087 The following options are available when @value{AS} is configured for
1088 a picoJava processor.
1092 @cindex PJ endianness
1093 @cindex endianness, PJ
1094 @cindex big endian output, PJ
1096 Generate ``big endian'' format output.
1098 @cindex little endian output, PJ
1100 Generate ``little endian'' format output.
1106 The following options are available when @value{AS} is configured for the
1107 Motorola 68HC11 or 68HC12 series.
1111 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1112 Specify what processor is the target. The default is
1113 defined by the configuration option when building the assembler.
1115 @item --xgate-ramoffset
1116 Instruct the linker to offset RAM addresses from S12X address space into
1117 XGATE address space.
1120 Specify to use the 16-bit integer ABI.
1123 Specify to use the 32-bit integer ABI.
1125 @item -mshort-double
1126 Specify to use the 32-bit double ABI.
1129 Specify to use the 64-bit double ABI.
1131 @item --force-long-branches
1132 Relative branches are turned into absolute ones. This concerns
1133 conditional branches, unconditional branches and branches to a
1136 @item -S | --short-branches
1137 Do not turn relative branches into absolute ones
1138 when the offset is out of range.
1140 @item --strict-direct-mode
1141 Do not turn the direct addressing mode into extended addressing mode
1142 when the instruction does not support direct addressing mode.
1144 @item --print-insn-syntax
1145 Print the syntax of instruction in case of error.
1147 @item --print-opcodes
1148 Print the list of instructions with syntax and then exit.
1150 @item --generate-example
1151 Print an example of instruction for each possible instruction and then exit.
1152 This option is only useful for testing @command{@value{AS}}.
1158 The following options are available when @command{@value{AS}} is configured
1159 for the SPARC architecture:
1162 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1163 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1164 Explicitly select a variant of the SPARC architecture.
1166 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1167 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1169 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1170 UltraSPARC extensions.
1172 @item -xarch=v8plus | -xarch=v8plusa
1173 For compatibility with the Solaris v9 assembler. These options are
1174 equivalent to -Av8plus and -Av8plusa, respectively.
1177 Warn when the assembler switches to another architecture.
1182 The following options are available when @value{AS} is configured for the 'c54x
1187 Enable extended addressing mode. All addresses and relocations will assume
1188 extended addressing (usually 23 bits).
1189 @item -mcpu=@var{CPU_VERSION}
1190 Sets the CPU version being compiled for.
1191 @item -merrors-to-file @var{FILENAME}
1192 Redirect error output to a file, for broken systems which don't support such
1193 behaviour in the shell.
1198 The following options are available when @value{AS} is configured for
1199 a @sc{mips} processor.
1203 This option sets the largest size of an object that can be referenced
1204 implicitly with the @code{gp} register. It is only accepted for targets that
1205 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1207 @cindex MIPS endianness
1208 @cindex endianness, MIPS
1209 @cindex big endian output, MIPS
1211 Generate ``big endian'' format output.
1213 @cindex little endian output, MIPS
1215 Generate ``little endian'' format output.
1227 Generate code for a particular @sc{mips} Instruction Set Architecture level.
1228 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1229 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1230 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1231 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips64}, and
1233 correspond to generic
1234 @samp{MIPS V}, @samp{MIPS32}, @samp{MIPS32 Release 2}, @samp{MIPS64},
1235 and @samp{MIPS64 Release 2}
1236 ISA processors, respectively.
1238 @item -march=@var{CPU}
1239 Generate code for a particular @sc{mips} cpu.
1241 @item -mtune=@var{cpu}
1242 Schedule and tune for a particular @sc{mips} cpu.
1246 Cause nops to be inserted if the read of the destination register
1247 of an mfhi or mflo instruction occurs in the following two instructions.
1251 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1252 section instead of the standard ELF .stabs sections.
1256 Control generation of @code{.pdr} sections.
1260 The register sizes are normally inferred from the ISA and ABI, but these
1261 flags force a certain group of registers to be treated as 32 bits wide at
1262 all times. @samp{-mgp32} controls the size of general-purpose registers
1263 and @samp{-mfp32} controls the size of floating-point registers.
1267 Generate code for the MIPS 16 processor. This is equivalent to putting
1268 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1269 turns off this option.
1272 @itemx -mno-micromips
1273 Generate code for the microMIPS processor. This is equivalent to putting
1274 @code{.set micromips} at the start of the assembly file. @samp{-mno-micromips}
1275 turns off this option. This is equivalent to putting @code{.set nomicromips}
1276 at the start of the assembly file.
1279 @itemx -mno-smartmips
1280 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1281 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1282 @samp{-mno-smartmips} turns off this option.
1286 Generate code for the MIPS-3D Application Specific Extension.
1287 This tells the assembler to accept MIPS-3D instructions.
1288 @samp{-no-mips3d} turns off this option.
1292 Generate code for the MDMX Application Specific Extension.
1293 This tells the assembler to accept MDMX instructions.
1294 @samp{-no-mdmx} turns off this option.
1298 Generate code for the DSP Release 1 Application Specific Extension.
1299 This tells the assembler to accept DSP Release 1 instructions.
1300 @samp{-mno-dsp} turns off this option.
1304 Generate code for the DSP Release 2 Application Specific Extension.
1305 This option implies -mdsp.
1306 This tells the assembler to accept DSP Release 2 instructions.
1307 @samp{-mno-dspr2} turns off this option.
1311 Generate code for the MT Application Specific Extension.
1312 This tells the assembler to accept MT instructions.
1313 @samp{-mno-mt} turns off this option.
1317 Generate code for the MCU Application Specific Extension.
1318 This tells the assembler to accept MCU instructions.
1319 @samp{-mno-mcu} turns off this option.
1321 @item --construct-floats
1322 @itemx --no-construct-floats
1323 The @samp{--no-construct-floats} option disables the construction of
1324 double width floating point constants by loading the two halves of the
1325 value into the two single width floating point registers that make up
1326 the double width register. By default @samp{--construct-floats} is
1327 selected, allowing construction of these floating point constants.
1330 @item --emulation=@var{name}
1331 This option causes @command{@value{AS}} to emulate @command{@value{AS}} configured
1332 for some other target, in all respects, including output format (choosing
1333 between ELF and ECOFF only), handling of pseudo-opcodes which may generate
1334 debugging information or store symbol table information, and default
1335 endianness. The available configuration names are: @samp{mipsecoff},
1336 @samp{mipself}, @samp{mipslecoff}, @samp{mipsbecoff}, @samp{mipslelf},
1337 @samp{mipsbelf}. The first two do not alter the default endianness from that
1338 of the primary target for which the assembler was configured; the others change
1339 the default to little- or big-endian as indicated by the @samp{b} or @samp{l}
1340 in the name. Using @samp{-EB} or @samp{-EL} will override the endianness
1341 selection in any case.
1343 This option is currently supported only when the primary target
1344 @command{@value{AS}} is configured for is a @sc{mips} ELF or ECOFF target.
1345 Furthermore, the primary target or others specified with
1346 @samp{--enable-targets=@dots{}} at configuration time must include support for
1347 the other format, if both are to be available. For example, the Irix 5
1348 configuration includes support for both.
1350 Eventually, this option will support more configurations, with more
1351 fine-grained control over the assembler's behavior, and will be supported for
1355 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1362 Control how to deal with multiplication overflow and division by zero.
1363 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1364 (and only work for Instruction Set Architecture level 2 and higher);
1365 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1369 When this option is used, @command{@value{AS}} will issue a warning every
1370 time it generates a nop instruction from a macro.
1375 The following options are available when @value{AS} is configured for
1381 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1382 The command line option @samp{-nojsri2bsr} can be used to disable it.
1386 Enable or disable the silicon filter behaviour. By default this is disabled.
1387 The default can be overridden by the @samp{-sifilter} command line option.
1390 Alter jump instructions for long displacements.
1392 @item -mcpu=[210|340]
1393 Select the cpu type on the target hardware. This controls which instructions
1397 Assemble for a big endian target.
1400 Assemble for a little endian target.
1409 @xref{Meta Options}, for the options available when @value{AS} is configured
1410 for a Meta processor.
1414 @c man begin OPTIONS
1415 The following options are available when @value{AS} is configured for a
1418 @c man begin INCLUDE
1419 @include c-metag.texi
1420 @c ended inside the included file
1425 @c man begin OPTIONS
1427 See the info pages for documentation of the MMIX-specific options.
1434 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1435 for a PowerPC processor.
1439 @c man begin OPTIONS
1440 The following options are available when @value{AS} is configured for a
1443 @c man begin INCLUDE
1445 @c ended inside the included file
1450 @c man begin OPTIONS
1452 See the info pages for documentation of the RX-specific options.
1456 The following options are available when @value{AS} is configured for the s390
1462 Select the word size, either 31/32 bits or 64 bits.
1465 Select the architecture mode, either the Enterprise System
1466 Architecture (esa) or the z/Architecture mode (zarch).
1467 @item -march=@var{processor}
1468 Specify which s390 processor variant is the target, @samp{g6}, @samp{g6},
1469 @samp{z900}, @samp{z990}, @samp{z9-109}, @samp{z9-ec}, @samp{z10},
1470 @samp{z196}, or @samp{zEC12}.
1472 @itemx -mno-regnames
1473 Allow or disallow symbolic names for registers.
1474 @item -mwarn-areg-zero
1475 Warn whenever the operand for a base or index register has been specified
1476 but evaluates to zero.
1484 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1485 for a TMS320C6000 processor.
1489 @c man begin OPTIONS
1490 The following options are available when @value{AS} is configured for a
1491 TMS320C6000 processor.
1493 @c man begin INCLUDE
1494 @include c-tic6x.texi
1495 @c ended inside the included file
1503 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1504 for a TILE-Gx processor.
1508 @c man begin OPTIONS
1509 The following options are available when @value{AS} is configured for a TILE-Gx
1512 @c man begin INCLUDE
1513 @include c-tilegx.texi
1514 @c ended inside the included file
1522 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1523 for an Xtensa processor.
1527 @c man begin OPTIONS
1528 The following options are available when @value{AS} is configured for an
1531 @c man begin INCLUDE
1532 @include c-xtensa.texi
1533 @c ended inside the included file
1538 @c man begin OPTIONS
1541 The following options are available when @value{AS} is configured for
1542 a Z80 family processor.
1545 Assemble for Z80 processor.
1547 Assemble for R800 processor.
1548 @item -ignore-undocumented-instructions
1550 Assemble undocumented Z80 instructions that also work on R800 without warning.
1551 @item -ignore-unportable-instructions
1553 Assemble all undocumented Z80 instructions without warning.
1554 @item -warn-undocumented-instructions
1556 Issue a warning for undocumented Z80 instructions that also work on R800.
1557 @item -warn-unportable-instructions
1559 Issue a warning for undocumented Z80 instructions that do not work on R800.
1560 @item -forbid-undocumented-instructions
1562 Treat all undocumented instructions as errors.
1563 @item -forbid-unportable-instructions
1565 Treat undocumented Z80 instructions that do not work on R800 as errors.
1572 * Manual:: Structure of this Manual
1573 * GNU Assembler:: The GNU Assembler
1574 * Object Formats:: Object File Formats
1575 * Command Line:: Command Line
1576 * Input Files:: Input Files
1577 * Object:: Output (Object) File
1578 * Errors:: Error and Warning Messages
1582 @section Structure of this Manual
1584 @cindex manual, structure and purpose
1585 This manual is intended to describe what you need to know to use
1586 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1587 notation for symbols, constants, and expressions; the directives that
1588 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1591 We also cover special features in the @value{TARGET}
1592 configuration of @command{@value{AS}}, including assembler directives.
1595 This manual also describes some of the machine-dependent features of
1596 various flavors of the assembler.
1599 @cindex machine instructions (not covered)
1600 On the other hand, this manual is @emph{not} intended as an introduction
1601 to programming in assembly language---let alone programming in general!
1602 In a similar vein, we make no attempt to introduce the machine
1603 architecture; we do @emph{not} describe the instruction set, standard
1604 mnemonics, registers or addressing modes that are standard to a
1605 particular architecture.
1607 You may want to consult the manufacturer's
1608 machine architecture manual for this information.
1612 For information on the H8/300 machine instruction set, see @cite{H8/300
1613 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1614 Programming Manual} (Renesas).
1617 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1618 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1619 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1620 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1623 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1627 @c I think this is premature---doc@cygnus.com, 17jan1991
1629 Throughout this manual, we assume that you are running @dfn{GNU},
1630 the portable operating system from the @dfn{Free Software
1631 Foundation, Inc.}. This restricts our attention to certain kinds of
1632 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1633 once this assumption is granted examples and definitions need less
1636 @command{@value{AS}} is part of a team of programs that turn a high-level
1637 human-readable series of instructions into a low-level
1638 computer-readable series of instructions. Different versions of
1639 @command{@value{AS}} are used for different kinds of computer.
1642 @c There used to be a section "Terminology" here, which defined
1643 @c "contents", "byte", "word", and "long". Defining "word" to any
1644 @c particular size is confusing when the .word directive may generate 16
1645 @c bits on one machine and 32 bits on another; in general, for the user
1646 @c version of this manual, none of these terms seem essential to define.
1647 @c They were used very little even in the former draft of the manual;
1648 @c this draft makes an effort to avoid them (except in names of
1652 @section The GNU Assembler
1654 @c man begin DESCRIPTION
1656 @sc{gnu} @command{as} is really a family of assemblers.
1658 This manual describes @command{@value{AS}}, a member of that family which is
1659 configured for the @value{TARGET} architectures.
1661 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1662 should find a fairly similar environment when you use it on another
1663 architecture. Each version has much in common with the others,
1664 including object file formats, most assembler directives (often called
1665 @dfn{pseudo-ops}) and assembler syntax.@refill
1667 @cindex purpose of @sc{gnu} assembler
1668 @command{@value{AS}} is primarily intended to assemble the output of the
1669 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1670 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1671 assemble correctly everything that other assemblers for the same
1672 machine would assemble.
1674 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1677 @c This remark should appear in generic version of manual; assumption
1678 @c here is that generic version sets M680x0.
1679 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1680 assembler for the same architecture; for example, we know of several
1681 incompatible versions of 680x0 assembly language syntax.
1686 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1687 program in one pass of the source file. This has a subtle impact on the
1688 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1690 @node Object Formats
1691 @section Object File Formats
1693 @cindex object file format
1694 The @sc{gnu} assembler can be configured to produce several alternative
1695 object file formats. For the most part, this does not affect how you
1696 write assembly language programs; but directives for debugging symbols
1697 are typically different in different file formats. @xref{Symbol
1698 Attributes,,Symbol Attributes}.
1701 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1702 @value{OBJ-NAME} format object files.
1704 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1706 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1707 @code{b.out} or COFF format object files.
1710 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1711 SOM or ELF format object files.
1716 @section Command Line
1718 @cindex command line conventions
1720 After the program name @command{@value{AS}}, the command line may contain
1721 options and file names. Options may appear in any order, and may be
1722 before, after, or between file names. The order of file names is
1725 @cindex standard input, as input file
1727 @file{--} (two hyphens) by itself names the standard input file
1728 explicitly, as one of the files for @command{@value{AS}} to assemble.
1730 @cindex options, command line
1731 Except for @samp{--} any command line argument that begins with a
1732 hyphen (@samp{-}) is an option. Each option changes the behavior of
1733 @command{@value{AS}}. No option changes the way another option works. An
1734 option is a @samp{-} followed by one or more letters; the case of
1735 the letter is important. All options are optional.
1737 Some options expect exactly one file name to follow them. The file
1738 name may either immediately follow the option's letter (compatible
1739 with older assemblers) or it may be the next command argument (@sc{gnu}
1740 standard). These two command lines are equivalent:
1743 @value{AS} -o my-object-file.o mumble.s
1744 @value{AS} -omy-object-file.o mumble.s
1748 @section Input Files
1751 @cindex source program
1752 @cindex files, input
1753 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1754 describe the program input to one run of @command{@value{AS}}. The program may
1755 be in one or more files; how the source is partitioned into files
1756 doesn't change the meaning of the source.
1758 @c I added "con" prefix to "catenation" just to prove I can overcome my
1759 @c APL training... doc@cygnus.com
1760 The source program is a concatenation of the text in all the files, in the
1763 @c man begin DESCRIPTION
1764 Each time you run @command{@value{AS}} it assembles exactly one source
1765 program. The source program is made up of one or more files.
1766 (The standard input is also a file.)
1768 You give @command{@value{AS}} a command line that has zero or more input file
1769 names. The input files are read (from left file name to right). A
1770 command line argument (in any position) that has no special meaning
1771 is taken to be an input file name.
1773 If you give @command{@value{AS}} no file names it attempts to read one input file
1774 from the @command{@value{AS}} standard input, which is normally your terminal. You
1775 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1778 Use @samp{--} if you need to explicitly name the standard input file
1779 in your command line.
1781 If the source is empty, @command{@value{AS}} produces a small, empty object
1786 @subheading Filenames and Line-numbers
1788 @cindex input file linenumbers
1789 @cindex line numbers, in input files
1790 There are two ways of locating a line in the input file (or files) and
1791 either may be used in reporting error messages. One way refers to a line
1792 number in a physical file; the other refers to a line number in a
1793 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
1795 @dfn{Physical files} are those files named in the command line given
1796 to @command{@value{AS}}.
1798 @dfn{Logical files} are simply names declared explicitly by assembler
1799 directives; they bear no relation to physical files. Logical file names help
1800 error messages reflect the original source file, when @command{@value{AS}} source
1801 is itself synthesized from other files. @command{@value{AS}} understands the
1802 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
1803 @ref{File,,@code{.file}}.
1806 @section Output (Object) File
1812 Every time you run @command{@value{AS}} it produces an output file, which is
1813 your assembly language program translated into numbers. This file
1814 is the object file. Its default name is
1822 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
1824 You can give it another name by using the @option{-o} option. Conventionally,
1825 object file names end with @file{.o}. The default name is used for historical
1826 reasons: older assemblers were capable of assembling self-contained programs
1827 directly into a runnable program. (For some formats, this isn't currently
1828 possible, but it can be done for the @code{a.out} format.)
1832 The object file is meant for input to the linker @code{@value{LD}}. It contains
1833 assembled program code, information to help @code{@value{LD}} integrate
1834 the assembled program into a runnable file, and (optionally) symbolic
1835 information for the debugger.
1837 @c link above to some info file(s) like the description of a.out.
1838 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
1841 @section Error and Warning Messages
1843 @c man begin DESCRIPTION
1845 @cindex error messages
1846 @cindex warning messages
1847 @cindex messages from assembler
1848 @command{@value{AS}} may write warnings and error messages to the standard error
1849 file (usually your terminal). This should not happen when a compiler
1850 runs @command{@value{AS}} automatically. Warnings report an assumption made so
1851 that @command{@value{AS}} could keep assembling a flawed program; errors report a
1852 grave problem that stops the assembly.
1856 @cindex format of warning messages
1857 Warning messages have the format
1860 file_name:@b{NNN}:Warning Message Text
1864 @cindex line numbers, in warnings/errors
1865 (where @b{NNN} is a line number). If a logical file name has been given
1866 (@pxref{File,,@code{.file}}) it is used for the filename, otherwise the name of
1867 the current input file is used. If a logical line number was given
1869 (@pxref{Line,,@code{.line}})
1871 then it is used to calculate the number printed,
1872 otherwise the actual line in the current source file is printed. The
1873 message text is intended to be self explanatory (in the grand Unix
1876 @cindex format of error messages
1877 Error messages have the format
1879 file_name:@b{NNN}:FATAL:Error Message Text
1881 The file name and line number are derived as for warning
1882 messages. The actual message text may be rather less explanatory
1883 because many of them aren't supposed to happen.
1886 @chapter Command-Line Options
1888 @cindex options, all versions of assembler
1889 This chapter describes command-line options available in @emph{all}
1890 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
1891 for options specific
1893 to the @value{TARGET} target.
1896 to particular machine architectures.
1899 @c man begin DESCRIPTION
1901 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
1902 you can use the @samp{-Wa} option to pass arguments through to the assembler.
1903 The assembler arguments must be separated from each other (and the @samp{-Wa})
1904 by commas. For example:
1907 gcc -c -g -O -Wa,-alh,-L file.c
1911 This passes two options to the assembler: @samp{-alh} (emit a listing to
1912 standard output with high-level and assembly source) and @samp{-L} (retain
1913 local symbols in the symbol table).
1915 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
1916 command-line options are automatically passed to the assembler by the compiler.
1917 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
1918 precisely what options it passes to each compilation pass, including the
1924 * a:: -a[cdghlns] enable listings
1925 * alternate:: --alternate enable alternate macro syntax
1926 * D:: -D for compatibility
1927 * f:: -f to work faster
1928 * I:: -I for .include search path
1929 @ifclear DIFF-TBL-KLUGE
1930 * K:: -K for compatibility
1932 @ifset DIFF-TBL-KLUGE
1933 * K:: -K for difference tables
1936 * L:: -L to retain local symbols
1937 * listing:: --listing-XXX to configure listing output
1938 * M:: -M or --mri to assemble in MRI compatibility mode
1939 * MD:: --MD for dependency tracking
1940 * o:: -o to name the object file
1941 * R:: -R to join data and text sections
1942 * statistics:: --statistics to see statistics about assembly
1943 * traditional-format:: --traditional-format for compatible output
1944 * v:: -v to announce version
1945 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
1946 * Z:: -Z to make object file even after errors
1950 @section Enable Listings: @option{-a[cdghlns]}
1960 @cindex listings, enabling
1961 @cindex assembly listings, enabling
1963 These options enable listing output from the assembler. By itself,
1964 @samp{-a} requests high-level, assembly, and symbols listing.
1965 You can use other letters to select specific options for the list:
1966 @samp{-ah} requests a high-level language listing,
1967 @samp{-al} requests an output-program assembly listing, and
1968 @samp{-as} requests a symbol table listing.
1969 High-level listings require that a compiler debugging option like
1970 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
1973 Use the @samp{-ag} option to print a first section with general assembly
1974 information, like @value{AS} version, switches passed, or time stamp.
1976 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
1977 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
1978 other conditional), or a true @code{.if} followed by an @code{.else}, will be
1979 omitted from the listing.
1981 Use the @samp{-ad} option to omit debugging directives from the
1984 Once you have specified one of these options, you can further control
1985 listing output and its appearance using the directives @code{.list},
1986 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
1988 The @samp{-an} option turns off all forms processing.
1989 If you do not request listing output with one of the @samp{-a} options, the
1990 listing-control directives have no effect.
1992 The letters after @samp{-a} may be combined into one option,
1993 @emph{e.g.}, @samp{-aln}.
1995 Note if the assembler source is coming from the standard input (e.g.,
1997 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
1998 is being used) then the listing will not contain any comments or preprocessor
1999 directives. This is because the listing code buffers input source lines from
2000 stdin only after they have been preprocessed by the assembler. This reduces
2001 memory usage and makes the code more efficient.
2004 @section @option{--alternate}
2007 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2010 @section @option{-D}
2013 This option has no effect whatsoever, but it is accepted to make it more
2014 likely that scripts written for other assemblers also work with
2015 @command{@value{AS}}.
2018 @section Work Faster: @option{-f}
2021 @cindex trusted compiler
2022 @cindex faster processing (@option{-f})
2023 @samp{-f} should only be used when assembling programs written by a
2024 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2025 and comment preprocessing on
2026 the input file(s) before assembling them. @xref{Preprocessing,
2030 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2031 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2036 @section @code{.include} Search Path: @option{-I} @var{path}
2038 @kindex -I @var{path}
2039 @cindex paths for @code{.include}
2040 @cindex search path for @code{.include}
2041 @cindex @code{include} directive search path
2042 Use this option to add a @var{path} to the list of directories
2043 @command{@value{AS}} searches for files specified in @code{.include}
2044 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2045 many times as necessary to include a variety of paths. The current
2046 working directory is always searched first; after that, @command{@value{AS}}
2047 searches any @samp{-I} directories in the same order as they were
2048 specified (left to right) on the command line.
2051 @section Difference Tables: @option{-K}
2054 @ifclear DIFF-TBL-KLUGE
2055 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2056 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2057 where it can be used to warn when the assembler alters the machine code
2058 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2059 family does not have the addressing limitations that sometimes lead to this
2060 alteration on other platforms.
2063 @ifset DIFF-TBL-KLUGE
2064 @cindex difference tables, warning
2065 @cindex warning for altered difference tables
2066 @command{@value{AS}} sometimes alters the code emitted for directives of the
2067 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2068 You can use the @samp{-K} option if you want a warning issued when this
2073 @section Include Local Symbols: @option{-L}
2076 @cindex local symbols, retaining in output
2077 Symbols beginning with system-specific local label prefixes, typically
2078 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2079 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2080 such symbols when debugging, because they are intended for the use of
2081 programs (like compilers) that compose assembler programs, not for your
2082 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2083 such symbols, so you do not normally debug with them.
2085 This option tells @command{@value{AS}} to retain those local symbols
2086 in the object file. Usually if you do this you also tell the linker
2087 @code{@value{LD}} to preserve those symbols.
2090 @section Configuring listing output: @option{--listing}
2092 The listing feature of the assembler can be enabled via the command line switch
2093 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2094 hex dump of the corresponding locations in the output object file, and displays
2095 them as a listing file. The format of this listing can be controlled by
2096 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2097 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2098 @code{.psize} (@pxref{Psize}), and
2099 @code{.eject} (@pxref{Eject}) and also by the following switches:
2102 @item --listing-lhs-width=@samp{number}
2103 @kindex --listing-lhs-width
2104 @cindex Width of first line disassembly output
2105 Sets the maximum width, in words, of the first line of the hex byte dump. This
2106 dump appears on the left hand side of the listing output.
2108 @item --listing-lhs-width2=@samp{number}
2109 @kindex --listing-lhs-width2
2110 @cindex Width of continuation lines of disassembly output
2111 Sets the maximum width, in words, of any further lines of the hex byte dump for
2112 a given input source line. If this value is not specified, it defaults to being
2113 the same as the value specified for @samp{--listing-lhs-width}. If neither
2114 switch is used the default is to one.
2116 @item --listing-rhs-width=@samp{number}
2117 @kindex --listing-rhs-width
2118 @cindex Width of source line output
2119 Sets the maximum width, in characters, of the source line that is displayed
2120 alongside the hex dump. The default value for this parameter is 100. The
2121 source line is displayed on the right hand side of the listing output.
2123 @item --listing-cont-lines=@samp{number}
2124 @kindex --listing-cont-lines
2125 @cindex Maximum number of continuation lines
2126 Sets the maximum number of continuation lines of hex dump that will be
2127 displayed for a given single line of source input. The default value is 4.
2131 @section Assemble in MRI Compatibility Mode: @option{-M}
2134 @cindex MRI compatibility mode
2135 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2136 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2137 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
2138 configured target) assembler from Microtec Research. The exact nature of the
2139 MRI syntax will not be documented here; see the MRI manuals for more
2140 information. Note in particular that the handling of macros and macro
2141 arguments is somewhat different. The purpose of this option is to permit
2142 assembling existing MRI assembler code using @command{@value{AS}}.
2144 The MRI compatibility is not complete. Certain operations of the MRI assembler
2145 depend upon its object file format, and can not be supported using other object
2146 file formats. Supporting these would require enhancing each object file format
2147 individually. These are:
2150 @item global symbols in common section
2152 The m68k MRI assembler supports common sections which are merged by the linker.
2153 Other object file formats do not support this. @command{@value{AS}} handles
2154 common sections by treating them as a single common symbol. It permits local
2155 symbols to be defined within a common section, but it can not support global
2156 symbols, since it has no way to describe them.
2158 @item complex relocations
2160 The MRI assemblers support relocations against a negated section address, and
2161 relocations which combine the start addresses of two or more sections. These
2162 are not support by other object file formats.
2164 @item @code{END} pseudo-op specifying start address
2166 The MRI @code{END} pseudo-op permits the specification of a start address.
2167 This is not supported by other object file formats. The start address may
2168 instead be specified using the @option{-e} option to the linker, or in a linker
2171 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2173 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2174 name to the output file. This is not supported by other object file formats.
2176 @item @code{ORG} pseudo-op
2178 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2179 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2180 which changes the location within the current section. Absolute sections are
2181 not supported by other object file formats. The address of a section may be
2182 assigned within a linker script.
2185 There are some other features of the MRI assembler which are not supported by
2186 @command{@value{AS}}, typically either because they are difficult or because they
2187 seem of little consequence. Some of these may be supported in future releases.
2191 @item EBCDIC strings
2193 EBCDIC strings are not supported.
2195 @item packed binary coded decimal
2197 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2198 and @code{DCB.P} pseudo-ops are not supported.
2200 @item @code{FEQU} pseudo-op
2202 The m68k @code{FEQU} pseudo-op is not supported.
2204 @item @code{NOOBJ} pseudo-op
2206 The m68k @code{NOOBJ} pseudo-op is not supported.
2208 @item @code{OPT} branch control options
2210 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2211 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2212 relaxes all branches, whether forward or backward, to an appropriate size, so
2213 these options serve no purpose.
2215 @item @code{OPT} list control options
2217 The following m68k @code{OPT} list control options are ignored: @code{C},
2218 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2219 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2221 @item other @code{OPT} options
2223 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2224 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2226 @item @code{OPT} @code{D} option is default
2228 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2229 @code{OPT NOD} may be used to turn it off.
2231 @item @code{XREF} pseudo-op.
2233 The m68k @code{XREF} pseudo-op is ignored.
2235 @item @code{.debug} pseudo-op
2237 The i960 @code{.debug} pseudo-op is not supported.
2239 @item @code{.extended} pseudo-op
2241 The i960 @code{.extended} pseudo-op is not supported.
2243 @item @code{.list} pseudo-op.
2245 The various options of the i960 @code{.list} pseudo-op are not supported.
2247 @item @code{.optimize} pseudo-op
2249 The i960 @code{.optimize} pseudo-op is not supported.
2251 @item @code{.output} pseudo-op
2253 The i960 @code{.output} pseudo-op is not supported.
2255 @item @code{.setreal} pseudo-op
2257 The i960 @code{.setreal} pseudo-op is not supported.
2262 @section Dependency Tracking: @option{--MD}
2265 @cindex dependency tracking
2268 @command{@value{AS}} can generate a dependency file for the file it creates. This
2269 file consists of a single rule suitable for @code{make} describing the
2270 dependencies of the main source file.
2272 The rule is written to the file named in its argument.
2274 This feature is used in the automatic updating of makefiles.
2277 @section Name the Object File: @option{-o}
2280 @cindex naming object file
2281 @cindex object file name
2282 There is always one object file output when you run @command{@value{AS}}. By
2283 default it has the name
2286 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2300 You use this option (which takes exactly one filename) to give the
2301 object file a different name.
2303 Whatever the object file is called, @command{@value{AS}} overwrites any
2304 existing file of the same name.
2307 @section Join Data and Text Sections: @option{-R}
2310 @cindex data and text sections, joining
2311 @cindex text and data sections, joining
2312 @cindex joining text and data sections
2313 @cindex merging text and data sections
2314 @option{-R} tells @command{@value{AS}} to write the object file as if all
2315 data-section data lives in the text section. This is only done at
2316 the very last moment: your binary data are the same, but data
2317 section parts are relocated differently. The data section part of
2318 your object file is zero bytes long because all its bytes are
2319 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2321 When you specify @option{-R} it would be possible to generate shorter
2322 address displacements (because we do not have to cross between text and
2323 data section). We refrain from doing this simply for compatibility with
2324 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2327 When @command{@value{AS}} is configured for COFF or ELF output,
2328 this option is only useful if you use sections named @samp{.text} and
2333 @option{-R} is not supported for any of the HPPA targets. Using
2334 @option{-R} generates a warning from @command{@value{AS}}.
2338 @section Display Assembly Statistics: @option{--statistics}
2340 @kindex --statistics
2341 @cindex statistics, about assembly
2342 @cindex time, total for assembly
2343 @cindex space used, maximum for assembly
2344 Use @samp{--statistics} to display two statistics about the resources used by
2345 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2346 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2349 @node traditional-format
2350 @section Compatible Output: @option{--traditional-format}
2352 @kindex --traditional-format
2353 For some targets, the output of @command{@value{AS}} is different in some ways
2354 from the output of some existing assembler. This switch requests
2355 @command{@value{AS}} to use the traditional format instead.
2357 For example, it disables the exception frame optimizations which
2358 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2361 @section Announce Version: @option{-v}
2365 @cindex assembler version
2366 @cindex version of assembler
2367 You can find out what version of as is running by including the
2368 option @samp{-v} (which you can also spell as @samp{-version}) on the
2372 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2374 @command{@value{AS}} should never give a warning or error message when
2375 assembling compiler output. But programs written by people often
2376 cause @command{@value{AS}} to give a warning that a particular assumption was
2377 made. All such warnings are directed to the standard error file.
2381 @cindex suppressing warnings
2382 @cindex warnings, suppressing
2383 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2384 This only affects the warning messages: it does not change any particular of
2385 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2388 @kindex --fatal-warnings
2389 @cindex errors, caused by warnings
2390 @cindex warnings, causing error
2391 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2392 files that generate warnings to be in error.
2395 @cindex warnings, switching on
2396 You can switch these options off again by specifying @option{--warn}, which
2397 causes warnings to be output as usual.
2400 @section Generate Object File in Spite of Errors: @option{-Z}
2401 @cindex object file, after errors
2402 @cindex errors, continuing after
2403 After an error message, @command{@value{AS}} normally produces no output. If for
2404 some reason you are interested in object file output even after
2405 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2406 option. If there are any errors, @command{@value{AS}} continues anyways, and
2407 writes an object file after a final warning message of the form @samp{@var{n}
2408 errors, @var{m} warnings, generating bad object file.}
2413 @cindex machine-independent syntax
2414 @cindex syntax, machine-independent
2415 This chapter describes the machine-independent syntax allowed in a
2416 source file. @command{@value{AS}} syntax is similar to what many other
2417 assemblers use; it is inspired by the BSD 4.2
2422 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2426 * Preprocessing:: Preprocessing
2427 * Whitespace:: Whitespace
2428 * Comments:: Comments
2429 * Symbol Intro:: Symbols
2430 * Statements:: Statements
2431 * Constants:: Constants
2435 @section Preprocessing
2437 @cindex preprocessing
2438 The @command{@value{AS}} internal preprocessor:
2440 @cindex whitespace, removed by preprocessor
2442 adjusts and removes extra whitespace. It leaves one space or tab before
2443 the keywords on a line, and turns any other whitespace on the line into
2446 @cindex comments, removed by preprocessor
2448 removes all comments, replacing them with a single space, or an
2449 appropriate number of newlines.
2451 @cindex constants, converted by preprocessor
2453 converts character constants into the appropriate numeric values.
2456 It does not do macro processing, include file handling, or
2457 anything else you may get from your C compiler's preprocessor. You can
2458 do include file processing with the @code{.include} directive
2459 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2460 to get other ``CPP'' style preprocessing by giving the input file a
2461 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2462 Output, gcc.info, Using GNU CC}.
2464 Excess whitespace, comments, and character constants
2465 cannot be used in the portions of the input text that are not
2468 @cindex turning preprocessing on and off
2469 @cindex preprocessing, turning on and off
2472 If the first line of an input file is @code{#NO_APP} or if you use the
2473 @samp{-f} option, whitespace and comments are not removed from the input file.
2474 Within an input file, you can ask for whitespace and comment removal in
2475 specific portions of the by putting a line that says @code{#APP} before the
2476 text that may contain whitespace or comments, and putting a line that says
2477 @code{#NO_APP} after this text. This feature is mainly intend to support
2478 @code{asm} statements in compilers whose output is otherwise free of comments
2485 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2486 Whitespace is used to separate symbols, and to make programs neater for
2487 people to read. Unless within character constants
2488 (@pxref{Characters,,Character Constants}), any whitespace means the same
2489 as exactly one space.
2495 There are two ways of rendering comments to @command{@value{AS}}. In both
2496 cases the comment is equivalent to one space.
2498 Anything from @samp{/*} through the next @samp{*/} is a comment.
2499 This means you may not nest these comments.
2503 The only way to include a newline ('\n') in a comment
2504 is to use this sort of comment.
2507 /* This sort of comment does not nest. */
2510 @cindex line comment character
2511 Anything from a @dfn{line comment} character up to the next newline is
2512 considered a comment and is ignored. The line comment character is target
2513 specific, and some targets multiple comment characters. Some targets also have
2514 line comment characters that only work if they are the first character on a
2515 line. Some targets use a sequence of two characters to introduce a line
2516 comment. Some targets can also change their line comment characters depending
2517 upon command line options that have been used. For more details see the
2518 @emph{Syntax} section in the documentation for individual targets.
2520 If the line comment character is the hash sign (@samp{#}) then it still has the
2521 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2522 to specify logical line numbers:
2525 @cindex lines starting with @code{#}
2526 @cindex logical line numbers
2527 To be compatible with past assemblers, lines that begin with @samp{#} have a
2528 special interpretation. Following the @samp{#} should be an absolute
2529 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2530 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2531 new logical file name. The rest of the line, if any, should be whitespace.
2533 If the first non-whitespace characters on the line are not numeric,
2534 the line is ignored. (Just like a comment.)
2537 # This is an ordinary comment.
2538 # 42-6 "new_file_name" # New logical file name
2539 # This is logical line # 36.
2541 This feature is deprecated, and may disappear from future versions
2542 of @command{@value{AS}}.
2547 @cindex characters used in symbols
2548 @ifclear SPECIAL-SYMS
2549 A @dfn{symbol} is one or more characters chosen from the set of all
2550 letters (both upper and lower case), digits and the three characters
2556 A @dfn{symbol} is one or more characters chosen from the set of all
2557 letters (both upper and lower case), digits and the three characters
2558 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2564 On most machines, you can also use @code{$} in symbol names; exceptions
2565 are noted in @ref{Machine Dependencies}.
2567 No symbol may begin with a digit. Case is significant.
2568 There is no length limit: all characters are significant. Multibyte characters
2569 are supported. Symbols are delimited by characters not in that set, or by the
2570 beginning of a file (since the source program must end with a newline, the end
2571 of a file is not a possible symbol delimiter). @xref{Symbols}.
2572 @cindex length of symbols
2577 @cindex statements, structure of
2578 @cindex line separator character
2579 @cindex statement separator character
2581 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2582 @dfn{line separator character}. The line separator character is target
2583 specific and described in the @emph{Syntax} section of each
2584 target's documentation. Not all targets support a line separator character.
2585 The newline or line separator character is considered to be part of the
2586 preceding statement. Newlines and separators within character constants are an
2587 exception: they do not end statements.
2589 @cindex newline, required at file end
2590 @cindex EOF, newline must precede
2591 It is an error to end any statement with end-of-file: the last
2592 character of any input file should be a newline.@refill
2594 An empty statement is allowed, and may include whitespace. It is ignored.
2596 @cindex instructions and directives
2597 @cindex directives and instructions
2598 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2599 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2601 A statement begins with zero or more labels, optionally followed by a
2602 key symbol which determines what kind of statement it is. The key
2603 symbol determines the syntax of the rest of the statement. If the
2604 symbol begins with a dot @samp{.} then the statement is an assembler
2605 directive: typically valid for any computer. If the symbol begins with
2606 a letter the statement is an assembly language @dfn{instruction}: it
2607 assembles into a machine language instruction.
2609 Different versions of @command{@value{AS}} for different computers
2610 recognize different instructions. In fact, the same symbol may
2611 represent a different instruction in a different computer's assembly
2615 @cindex @code{:} (label)
2616 @cindex label (@code{:})
2617 A label is a symbol immediately followed by a colon (@code{:}).
2618 Whitespace before a label or after a colon is permitted, but you may not
2619 have whitespace between a label's symbol and its colon. @xref{Labels}.
2622 For HPPA targets, labels need not be immediately followed by a colon, but
2623 the definition of a label must begin in column zero. This also implies that
2624 only one label may be defined on each line.
2628 label: .directive followed by something
2629 another_label: # This is an empty statement.
2630 instruction operand_1, operand_2, @dots{}
2637 A constant is a number, written so that its value is known by
2638 inspection, without knowing any context. Like this:
2641 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2642 .ascii "Ring the bell\7" # A string constant.
2643 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2644 .float 0f-314159265358979323846264338327\
2645 95028841971.693993751E-40 # - pi, a flonum.
2650 * Characters:: Character Constants
2651 * Numbers:: Number Constants
2655 @subsection Character Constants
2657 @cindex character constants
2658 @cindex constants, character
2659 There are two kinds of character constants. A @dfn{character} stands
2660 for one character in one byte and its value may be used in
2661 numeric expressions. String constants (properly called string
2662 @emph{literals}) are potentially many bytes and their values may not be
2663 used in arithmetic expressions.
2667 * Chars:: Characters
2671 @subsubsection Strings
2673 @cindex string constants
2674 @cindex constants, string
2675 A @dfn{string} is written between double-quotes. It may contain
2676 double-quotes or null characters. The way to get special characters
2677 into a string is to @dfn{escape} these characters: precede them with
2678 a backslash @samp{\} character. For example @samp{\\} represents
2679 one backslash: the first @code{\} is an escape which tells
2680 @command{@value{AS}} to interpret the second character literally as a backslash
2681 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2682 escape character). The complete list of escapes follows.
2684 @cindex escape codes, character
2685 @cindex character escape codes
2688 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2690 @cindex @code{\b} (backspace character)
2691 @cindex backspace (@code{\b})
2693 Mnemonic for backspace; for ASCII this is octal code 010.
2696 @c Mnemonic for EOText; for ASCII this is octal code 004.
2698 @cindex @code{\f} (formfeed character)
2699 @cindex formfeed (@code{\f})
2701 Mnemonic for FormFeed; for ASCII this is octal code 014.
2703 @cindex @code{\n} (newline character)
2704 @cindex newline (@code{\n})
2706 Mnemonic for newline; for ASCII this is octal code 012.
2709 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2711 @cindex @code{\r} (carriage return character)
2712 @cindex carriage return (@code{\r})
2714 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2717 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2718 @c other assemblers.
2720 @cindex @code{\t} (tab)
2721 @cindex tab (@code{\t})
2723 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2726 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2727 @c @item \x @var{digit} @var{digit} @var{digit}
2728 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2730 @cindex @code{\@var{ddd}} (octal character code)
2731 @cindex octal character code (@code{\@var{ddd}})
2732 @item \ @var{digit} @var{digit} @var{digit}
2733 An octal character code. The numeric code is 3 octal digits.
2734 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2735 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2737 @cindex @code{\@var{xd...}} (hex character code)
2738 @cindex hex character code (@code{\@var{xd...}})
2739 @item \@code{x} @var{hex-digits...}
2740 A hex character code. All trailing hex digits are combined. Either upper or
2741 lower case @code{x} works.
2743 @cindex @code{\\} (@samp{\} character)
2744 @cindex backslash (@code{\\})
2746 Represents one @samp{\} character.
2749 @c Represents one @samp{'} (accent acute) character.
2750 @c This is needed in single character literals
2751 @c (@xref{Characters,,Character Constants}.) to represent
2754 @cindex @code{\"} (doublequote character)
2755 @cindex doublequote (@code{\"})
2757 Represents one @samp{"} character. Needed in strings to represent
2758 this character, because an unescaped @samp{"} would end the string.
2760 @item \ @var{anything-else}
2761 Any other character when escaped by @kbd{\} gives a warning, but
2762 assembles as if the @samp{\} was not present. The idea is that if
2763 you used an escape sequence you clearly didn't want the literal
2764 interpretation of the following character. However @command{@value{AS}} has no
2765 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
2766 code and warns you of the fact.
2769 Which characters are escapable, and what those escapes represent,
2770 varies widely among assemblers. The current set is what we think
2771 the BSD 4.2 assembler recognizes, and is a subset of what most C
2772 compilers recognize. If you are in doubt, do not use an escape
2776 @subsubsection Characters
2778 @cindex single character constant
2779 @cindex character, single
2780 @cindex constant, single character
2781 A single character may be written as a single quote immediately
2782 followed by that character. The same escapes apply to characters as
2783 to strings. So if you want to write the character backslash, you
2784 must write @kbd{'\\} where the first @code{\} escapes the second
2785 @code{\}. As you can see, the quote is an acute accent, not a
2786 grave accent. A newline
2788 @ifclear abnormal-separator
2789 (or semicolon @samp{;})
2791 @ifset abnormal-separator
2793 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
2798 immediately following an acute accent is taken as a literal character
2799 and does not count as the end of a statement. The value of a character
2800 constant in a numeric expression is the machine's byte-wide code for
2801 that character. @command{@value{AS}} assumes your character code is ASCII:
2802 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
2805 @subsection Number Constants
2807 @cindex constants, number
2808 @cindex number constants
2809 @command{@value{AS}} distinguishes three kinds of numbers according to how they
2810 are stored in the target machine. @emph{Integers} are numbers that
2811 would fit into an @code{int} in the C language. @emph{Bignums} are
2812 integers, but they are stored in more than 32 bits. @emph{Flonums}
2813 are floating point numbers, described below.
2816 * Integers:: Integers
2821 * Bit Fields:: Bit Fields
2827 @subsubsection Integers
2829 @cindex constants, integer
2831 @cindex binary integers
2832 @cindex integers, binary
2833 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
2834 the binary digits @samp{01}.
2836 @cindex octal integers
2837 @cindex integers, octal
2838 An octal integer is @samp{0} followed by zero or more of the octal
2839 digits (@samp{01234567}).
2841 @cindex decimal integers
2842 @cindex integers, decimal
2843 A decimal integer starts with a non-zero digit followed by zero or
2844 more digits (@samp{0123456789}).
2846 @cindex hexadecimal integers
2847 @cindex integers, hexadecimal
2848 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
2849 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
2851 Integers have the usual values. To denote a negative integer, use
2852 the prefix operator @samp{-} discussed under expressions
2853 (@pxref{Prefix Ops,,Prefix Operators}).
2856 @subsubsection Bignums
2859 @cindex constants, bignum
2860 A @dfn{bignum} has the same syntax and semantics as an integer
2861 except that the number (or its negative) takes more than 32 bits to
2862 represent in binary. The distinction is made because in some places
2863 integers are permitted while bignums are not.
2866 @subsubsection Flonums
2868 @cindex floating point numbers
2869 @cindex constants, floating point
2871 @cindex precision, floating point
2872 A @dfn{flonum} represents a floating point number. The translation is
2873 indirect: a decimal floating point number from the text is converted by
2874 @command{@value{AS}} to a generic binary floating point number of more than
2875 sufficient precision. This generic floating point number is converted
2876 to a particular computer's floating point format (or formats) by a
2877 portion of @command{@value{AS}} specialized to that computer.
2879 A flonum is written by writing (in order)
2884 (@samp{0} is optional on the HPPA.)
2888 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
2890 @kbd{e} is recommended. Case is not important.
2892 @c FIXME: verify if flonum syntax really this vague for most cases
2893 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
2894 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
2897 On the H8/300, Renesas / SuperH SH,
2898 and AMD 29K architectures, the letter must be
2899 one of the letters @samp{DFPRSX} (in upper or lower case).
2901 On the ARC, the letter must be one of the letters @samp{DFRS}
2902 (in upper or lower case).
2904 On the Intel 960 architecture, the letter must be
2905 one of the letters @samp{DFT} (in upper or lower case).
2907 On the HPPA architecture, the letter must be @samp{E} (upper case only).
2911 One of the letters @samp{DFRS} (in upper or lower case).
2914 One of the letters @samp{DFPRSX} (in upper or lower case).
2917 The letter @samp{E} (upper case only).
2920 One of the letters @samp{DFT} (in upper or lower case).
2925 An optional sign: either @samp{+} or @samp{-}.
2928 An optional @dfn{integer part}: zero or more decimal digits.
2931 An optional @dfn{fractional part}: @samp{.} followed by zero
2932 or more decimal digits.
2935 An optional exponent, consisting of:
2939 An @samp{E} or @samp{e}.
2940 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
2941 @c principle this can perfectly well be different on different targets.
2943 Optional sign: either @samp{+} or @samp{-}.
2945 One or more decimal digits.
2950 At least one of the integer part or the fractional part must be
2951 present. The floating point number has the usual base-10 value.
2953 @command{@value{AS}} does all processing using integers. Flonums are computed
2954 independently of any floating point hardware in the computer running
2955 @command{@value{AS}}.
2959 @c Bit fields are written as a general facility but are also controlled
2960 @c by a conditional-compilation flag---which is as of now (21mar91)
2961 @c turned on only by the i960 config of GAS.
2963 @subsubsection Bit Fields
2966 @cindex constants, bit field
2967 You can also define numeric constants as @dfn{bit fields}.
2968 Specify two numbers separated by a colon---
2970 @var{mask}:@var{value}
2973 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
2976 The resulting number is then packed
2978 @c this conditional paren in case bit fields turned on elsewhere than 960
2979 (in host-dependent byte order)
2981 into a field whose width depends on which assembler directive has the
2982 bit-field as its argument. Overflow (a result from the bitwise and
2983 requiring more binary digits to represent) is not an error; instead,
2984 more constants are generated, of the specified width, beginning with the
2985 least significant digits.@refill
2987 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
2988 @code{.short}, and @code{.word} accept bit-field arguments.
2993 @chapter Sections and Relocation
2998 * Secs Background:: Background
2999 * Ld Sections:: Linker Sections
3000 * As Sections:: Assembler Internal Sections
3001 * Sub-Sections:: Sub-Sections
3005 @node Secs Background
3008 Roughly, a section is a range of addresses, with no gaps; all data
3009 ``in'' those addresses is treated the same for some particular purpose.
3010 For example there may be a ``read only'' section.
3012 @cindex linker, and assembler
3013 @cindex assembler, and linker
3014 The linker @code{@value{LD}} reads many object files (partial programs) and
3015 combines their contents to form a runnable program. When @command{@value{AS}}
3016 emits an object file, the partial program is assumed to start at address 0.
3017 @code{@value{LD}} assigns the final addresses for the partial program, so that
3018 different partial programs do not overlap. This is actually an
3019 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3022 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3023 addresses. These blocks slide to their run-time addresses as rigid
3024 units; their length does not change and neither does the order of bytes
3025 within them. Such a rigid unit is called a @emph{section}. Assigning
3026 run-time addresses to sections is called @dfn{relocation}. It includes
3027 the task of adjusting mentions of object-file addresses so they refer to
3028 the proper run-time addresses.
3030 For the H8/300, and for the Renesas / SuperH SH,
3031 @command{@value{AS}} pads sections if needed to
3032 ensure they end on a word (sixteen bit) boundary.
3035 @cindex standard assembler sections
3036 An object file written by @command{@value{AS}} has at least three sections, any
3037 of which may be empty. These are named @dfn{text}, @dfn{data} and
3042 When it generates COFF or ELF output,
3044 @command{@value{AS}} can also generate whatever other named sections you specify
3045 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3046 If you do not use any directives that place output in the @samp{.text}
3047 or @samp{.data} sections, these sections still exist, but are empty.
3052 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3054 @command{@value{AS}} can also generate whatever other named sections you
3055 specify using the @samp{.space} and @samp{.subspace} directives. See
3056 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3057 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3058 assembler directives.
3061 Additionally, @command{@value{AS}} uses different names for the standard
3062 text, data, and bss sections when generating SOM output. Program text
3063 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3064 BSS into @samp{$BSS$}.
3068 Within the object file, the text section starts at address @code{0}, the
3069 data section follows, and the bss section follows the data section.
3072 When generating either SOM or ELF output files on the HPPA, the text
3073 section starts at address @code{0}, the data section at address
3074 @code{0x4000000}, and the bss section follows the data section.
3077 To let @code{@value{LD}} know which data changes when the sections are
3078 relocated, and how to change that data, @command{@value{AS}} also writes to the
3079 object file details of the relocation needed. To perform relocation
3080 @code{@value{LD}} must know, each time an address in the object
3084 Where in the object file is the beginning of this reference to
3087 How long (in bytes) is this reference?
3089 Which section does the address refer to? What is the numeric value of
3091 (@var{address}) @minus{} (@var{start-address of section})?
3094 Is the reference to an address ``Program-Counter relative''?
3097 @cindex addresses, format of
3098 @cindex section-relative addressing
3099 In fact, every address @command{@value{AS}} ever uses is expressed as
3101 (@var{section}) + (@var{offset into section})
3104 Further, most expressions @command{@value{AS}} computes have this section-relative
3107 (For some object formats, such as SOM for the HPPA, some expressions are
3108 symbol-relative instead.)
3111 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3112 @var{N} into section @var{secname}.''
3114 Apart from text, data and bss sections you need to know about the
3115 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3116 addresses in the absolute section remain unchanged. For example, address
3117 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3118 @code{@value{LD}}. Although the linker never arranges two partial programs'
3119 data sections with overlapping addresses after linking, @emph{by definition}
3120 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3121 part of a program is always the same address when the program is running as
3122 address @code{@{absolute@ 239@}} in any other part of the program.
3124 The idea of sections is extended to the @dfn{undefined} section. Any
3125 address whose section is unknown at assembly time is by definition
3126 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3127 Since numbers are always defined, the only way to generate an undefined
3128 address is to mention an undefined symbol. A reference to a named
3129 common block would be such a symbol: its value is unknown at assembly
3130 time so it has section @emph{undefined}.
3132 By analogy the word @emph{section} is used to describe groups of sections in
3133 the linked program. @code{@value{LD}} puts all partial programs' text
3134 sections in contiguous addresses in the linked program. It is
3135 customary to refer to the @emph{text section} of a program, meaning all
3136 the addresses of all partial programs' text sections. Likewise for
3137 data and bss sections.
3139 Some sections are manipulated by @code{@value{LD}}; others are invented for
3140 use of @command{@value{AS}} and have no meaning except during assembly.
3143 @section Linker Sections
3144 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3149 @cindex named sections
3150 @cindex sections, named
3151 @item named sections
3154 @cindex text section
3155 @cindex data section
3159 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3160 separate but equal sections. Anything you can say of one section is
3163 When the program is running, however, it is
3164 customary for the text section to be unalterable. The
3165 text section is often shared among processes: it contains
3166 instructions, constants and the like. The data section of a running
3167 program is usually alterable: for example, C variables would be stored
3168 in the data section.
3173 This section contains zeroed bytes when your program begins running. It
3174 is used to hold uninitialized variables or common storage. The length of
3175 each partial program's bss section is important, but because it starts
3176 out containing zeroed bytes there is no need to store explicit zero
3177 bytes in the object file. The bss section was invented to eliminate
3178 those explicit zeros from object files.
3180 @cindex absolute section
3181 @item absolute section
3182 Address 0 of this section is always ``relocated'' to runtime address 0.
3183 This is useful if you want to refer to an address that @code{@value{LD}} must
3184 not change when relocating. In this sense we speak of absolute
3185 addresses being ``unrelocatable'': they do not change during relocation.
3187 @cindex undefined section
3188 @item undefined section
3189 This ``section'' is a catch-all for address references to objects not in
3190 the preceding sections.
3191 @c FIXME: ref to some other doc on obj-file formats could go here.
3194 @cindex relocation example
3195 An idealized example of three relocatable sections follows.
3197 The example uses the traditional section names @samp{.text} and @samp{.data}.
3199 Memory addresses are on the horizontal axis.
3203 @c END TEXI2ROFF-KILL
3206 partial program # 1: |ttttt|dddd|00|
3213 partial program # 2: |TTT|DDD|000|
3216 +--+---+-----+--+----+---+-----+~~
3217 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3218 +--+---+-----+--+----+---+-----+~~
3220 addresses: 0 @dots{}
3227 \line{\it Partial program \#1: \hfil}
3228 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3229 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3231 \line{\it Partial program \#2: \hfil}
3232 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3233 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3235 \line{\it linked program: \hfil}
3236 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3237 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3238 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3239 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3241 \line{\it addresses: \hfil}
3245 @c END TEXI2ROFF-KILL
3248 @section Assembler Internal Sections
3250 @cindex internal assembler sections
3251 @cindex sections in messages, internal
3252 These sections are meant only for the internal use of @command{@value{AS}}. They
3253 have no meaning at run-time. You do not really need to know about these
3254 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3255 warning messages, so it might be helpful to have an idea of their
3256 meanings to @command{@value{AS}}. These sections are used to permit the
3257 value of every expression in your assembly language program to be a
3258 section-relative address.
3261 @cindex assembler internal logic error
3262 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3263 An internal assembler logic error has been found. This means there is a
3264 bug in the assembler.
3266 @cindex expr (internal section)
3268 The assembler stores complex expression internally as combinations of
3269 symbols. When it needs to represent an expression as a symbol, it puts
3270 it in the expr section.
3272 @c FIXME item transfer[t] vector preload
3273 @c FIXME item transfer[t] vector postload
3274 @c FIXME item register
3278 @section Sub-Sections
3280 @cindex numbered subsections
3281 @cindex grouping data
3287 fall into two sections: text and data.
3289 You may have separate groups of
3291 data in named sections
3295 data in named sections
3301 that you want to end up near to each other in the object file, even though they
3302 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3303 use @dfn{subsections} for this purpose. Within each section, there can be
3304 numbered subsections with values from 0 to 8192. Objects assembled into the
3305 same subsection go into the object file together with other objects in the same
3306 subsection. For example, a compiler might want to store constants in the text
3307 section, but might not want to have them interspersed with the program being
3308 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3309 section of code being output, and a @samp{.text 1} before each group of
3310 constants being output.
3312 Subsections are optional. If you do not use subsections, everything
3313 goes in subsection number zero.
3316 Each subsection is zero-padded up to a multiple of four bytes.
3317 (Subsections may be padded a different amount on different flavors
3318 of @command{@value{AS}}.)
3322 On the H8/300 platform, each subsection is zero-padded to a word
3323 boundary (two bytes).
3324 The same is true on the Renesas SH.
3327 @c FIXME section padding (alignment)?
3328 @c Rich Pixley says padding here depends on target obj code format; that
3329 @c doesn't seem particularly useful to say without further elaboration,
3330 @c so for now I say nothing about it. If this is a generic BFD issue,
3331 @c these paragraphs might need to vanish from this manual, and be
3332 @c discussed in BFD chapter of binutils (or some such).
3336 Subsections appear in your object file in numeric order, lowest numbered
3337 to highest. (All this to be compatible with other people's assemblers.)
3338 The object file contains no representation of subsections; @code{@value{LD}} and
3339 other programs that manipulate object files see no trace of them.
3340 They just see all your text subsections as a text section, and all your
3341 data subsections as a data section.
3343 To specify which subsection you want subsequent statements assembled
3344 into, use a numeric argument to specify it, in a @samp{.text
3345 @var{expression}} or a @samp{.data @var{expression}} statement.
3348 When generating COFF output, you
3353 can also use an extra subsection
3354 argument with arbitrary named sections: @samp{.section @var{name},
3359 When generating ELF output, you
3364 can also use the @code{.subsection} directive (@pxref{SubSection})
3365 to specify a subsection: @samp{.subsection @var{expression}}.
3367 @var{Expression} should be an absolute expression
3368 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3369 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3370 begins in @code{text 0}. For instance:
3372 .text 0 # The default subsection is text 0 anyway.
3373 .ascii "This lives in the first text subsection. *"
3375 .ascii "But this lives in the second text subsection."
3377 .ascii "This lives in the data section,"
3378 .ascii "in the first data subsection."
3380 .ascii "This lives in the first text section,"
3381 .ascii "immediately following the asterisk (*)."
3384 Each section has a @dfn{location counter} incremented by one for every byte
3385 assembled into that section. Because subsections are merely a convenience
3386 restricted to @command{@value{AS}} there is no concept of a subsection location
3387 counter. There is no way to directly manipulate a location counter---but the
3388 @code{.align} directive changes it, and any label definition captures its
3389 current value. The location counter of the section where statements are being
3390 assembled is said to be the @dfn{active} location counter.
3393 @section bss Section
3396 @cindex common variable storage
3397 The bss section is used for local common variable storage.
3398 You may allocate address space in the bss section, but you may
3399 not dictate data to load into it before your program executes. When
3400 your program starts running, all the contents of the bss
3401 section are zeroed bytes.
3403 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3404 @ref{Lcomm,,@code{.lcomm}}.
3406 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3407 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3410 When assembling for a target which supports multiple sections, such as ELF or
3411 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3412 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3413 section. Typically the section will only contain symbol definitions and
3414 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3421 Symbols are a central concept: the programmer uses symbols to name
3422 things, the linker uses symbols to link, and the debugger uses symbols
3426 @cindex debuggers, and symbol order
3427 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3428 the same order they were declared. This may break some debuggers.
3433 * Setting Symbols:: Giving Symbols Other Values
3434 * Symbol Names:: Symbol Names
3435 * Dot:: The Special Dot Symbol
3436 * Symbol Attributes:: Symbol Attributes
3443 A @dfn{label} is written as a symbol immediately followed by a colon
3444 @samp{:}. The symbol then represents the current value of the
3445 active location counter, and is, for example, a suitable instruction
3446 operand. You are warned if you use the same symbol to represent two
3447 different locations: the first definition overrides any other
3451 On the HPPA, the usual form for a label need not be immediately followed by a
3452 colon, but instead must start in column zero. Only one label may be defined on
3453 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3454 provides a special directive @code{.label} for defining labels more flexibly.
3457 @node Setting Symbols
3458 @section Giving Symbols Other Values
3460 @cindex assigning values to symbols
3461 @cindex symbol values, assigning
3462 A symbol can be given an arbitrary value by writing a symbol, followed
3463 by an equals sign @samp{=}, followed by an expression
3464 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3465 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3466 equals sign @samp{=}@samp{=} here represents an equivalent of the
3467 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3470 Blackfin does not support symbol assignment with @samp{=}.
3474 @section Symbol Names
3476 @cindex symbol names
3477 @cindex names, symbol
3478 @ifclear SPECIAL-SYMS
3479 Symbol names begin with a letter or with one of @samp{._}. On most
3480 machines, you can also use @code{$} in symbol names; exceptions are
3481 noted in @ref{Machine Dependencies}. That character may be followed by any
3482 string of digits, letters, dollar signs (unless otherwise noted for a
3483 particular target machine), and underscores.
3487 Symbol names begin with a letter or with one of @samp{._}. On the
3488 Renesas SH you can also use @code{$} in symbol names. That
3489 character may be followed by any string of digits, letters, dollar signs (save
3490 on the H8/300), and underscores.
3494 Case of letters is significant: @code{foo} is a different symbol name
3497 Multibyte characters are supported. To generate a symbol name containing
3498 multibyte characters enclose it within double quotes and use escape codes. cf
3499 @xref{Strings}. Generating a multibyte symbol name from a label is not
3500 currently supported.
3502 Each symbol has exactly one name. Each name in an assembly language program
3503 refers to exactly one symbol. You may use that symbol name any number of times
3506 @subheading Local Symbol Names
3508 @cindex local symbol names
3509 @cindex symbol names, local
3510 A local symbol is any symbol beginning with certain local label prefixes.
3511 By default, the local label prefix is @samp{.L} for ELF systems or
3512 @samp{L} for traditional a.out systems, but each target may have its own
3513 set of local label prefixes.
3515 On the HPPA local symbols begin with @samp{L$}.
3518 Local symbols are defined and used within the assembler, but they are
3519 normally not saved in object files. Thus, they are not visible when debugging.
3520 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols:
3521 @option{-L}}) to retain the local symbols in the object files.
3523 @subheading Local Labels
3525 @cindex local labels
3526 @cindex temporary symbol names
3527 @cindex symbol names, temporary
3528 Local labels help compilers and programmers use names temporarily.
3529 They create symbols which are guaranteed to be unique over the entire scope of
3530 the input source code and which can be referred to by a simple notation.
3531 To define a local label, write a label of the form @samp{@b{N}:} (where @b{N}
3532 represents any positive integer). To refer to the most recent previous
3533 definition of that label write @samp{@b{N}b}, using the same number as when
3534 you defined the label. To refer to the next definition of a local label, write
3535 @samp{@b{N}f}---the @samp{b} stands for ``backwards'' and the @samp{f} stands
3538 There is no restriction on how you can use these labels, and you can reuse them
3539 too. So that it is possible to repeatedly define the same local label (using
3540 the same number @samp{@b{N}}), although you can only refer to the most recently
3541 defined local label of that number (for a backwards reference) or the next
3542 definition of a specific local label for a forward reference. It is also worth
3543 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3544 implemented in a slightly more efficient manner than the others.
3555 Which is the equivalent of:
3558 label_1: branch label_3
3559 label_2: branch label_1
3560 label_3: branch label_4
3561 label_4: branch label_3
3564 Local label names are only a notational device. They are immediately
3565 transformed into more conventional symbol names before the assembler uses them.
3566 The symbol names are stored in the symbol table, appear in error messages, and
3567 are optionally emitted to the object file. The names are constructed using
3571 @item @emph{local label prefix}
3572 All local symbols begin with the system-specific local label prefix.
3573 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3574 that start with the local label prefix. These labels are
3575 used for symbols you are never intended to see. If you use the
3576 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3577 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3578 you may use them in debugging.
3581 This is the number that was used in the local label definition. So if the
3582 label is written @samp{55:} then the number is @samp{55}.
3585 This unusual character is included so you do not accidentally invent a symbol
3586 of the same name. The character has ASCII value of @samp{\002} (control-B).
3588 @item @emph{ordinal number}
3589 This is a serial number to keep the labels distinct. The first definition of
3590 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3591 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3592 the number @samp{1} and its 15th definition gets @samp{15} as well.
3595 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3596 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3598 @subheading Dollar Local Labels
3599 @cindex dollar local symbols
3601 @code{@value{AS}} also supports an even more local form of local labels called
3602 dollar labels. These labels go out of scope (i.e., they become undefined) as
3603 soon as a non-local label is defined. Thus they remain valid for only a small
3604 region of the input source code. Normal local labels, by contrast, remain in
3605 scope for the entire file, or until they are redefined by another occurrence of
3606 the same local label.
3608 Dollar labels are defined in exactly the same way as ordinary local labels,
3609 except that they have a dollar sign suffix to their numeric value, e.g.,
3612 They can also be distinguished from ordinary local labels by their transformed
3613 names which use ASCII character @samp{\001} (control-A) as the magic character
3614 to distinguish them from ordinary labels. For example, the fifth definition of
3615 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3618 @section The Special Dot Symbol
3620 @cindex dot (symbol)
3621 @cindex @code{.} (symbol)
3622 @cindex current address
3623 @cindex location counter
3624 The special symbol @samp{.} refers to the current address that
3625 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3626 .long .} defines @code{melvin} to contain its own address.
3627 Assigning a value to @code{.} is treated the same as a @code{.org}
3629 @ifclear no-space-dir
3630 Thus, the expression @samp{.=.+4} is the same as saying
3634 @node Symbol Attributes
3635 @section Symbol Attributes
3637 @cindex symbol attributes
3638 @cindex attributes, symbol
3639 Every symbol has, as well as its name, the attributes ``Value'' and
3640 ``Type''. Depending on output format, symbols can also have auxiliary
3643 The detailed definitions are in @file{a.out.h}.
3646 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3647 all these attributes, and probably won't warn you. This makes the
3648 symbol an externally defined symbol, which is generally what you
3652 * Symbol Value:: Value
3653 * Symbol Type:: Type
3656 * a.out Symbols:: Symbol Attributes: @code{a.out}
3660 * a.out Symbols:: Symbol Attributes: @code{a.out}
3663 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3668 * COFF Symbols:: Symbol Attributes for COFF
3671 * SOM Symbols:: Symbol Attributes for SOM
3678 @cindex value of a symbol
3679 @cindex symbol value
3680 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3681 location in the text, data, bss or absolute sections the value is the
3682 number of addresses from the start of that section to the label.
3683 Naturally for text, data and bss sections the value of a symbol changes
3684 as @code{@value{LD}} changes section base addresses during linking. Absolute
3685 symbols' values do not change during linking: that is why they are
3688 The value of an undefined symbol is treated in a special way. If it is
3689 0 then the symbol is not defined in this assembler source file, and
3690 @code{@value{LD}} tries to determine its value from other files linked into the
3691 same program. You make this kind of symbol simply by mentioning a symbol
3692 name without defining it. A non-zero value represents a @code{.comm}
3693 common declaration. The value is how much common storage to reserve, in
3694 bytes (addresses). The symbol refers to the first address of the
3700 @cindex type of a symbol
3702 The type attribute of a symbol contains relocation (section)
3703 information, any flag settings indicating that a symbol is external, and
3704 (optionally), other information for linkers and debuggers. The exact
3705 format depends on the object-code output format in use.
3710 @c The following avoids a "widow" subsection title. @group would be
3711 @c better if it were available outside examples.
3714 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3716 @cindex @code{b.out} symbol attributes
3717 @cindex symbol attributes, @code{b.out}
3718 These symbol attributes appear only when @command{@value{AS}} is configured for
3719 one of the Berkeley-descended object output formats---@code{a.out} or
3725 @subsection Symbol Attributes: @code{a.out}
3727 @cindex @code{a.out} symbol attributes
3728 @cindex symbol attributes, @code{a.out}
3734 @subsection Symbol Attributes: @code{a.out}
3736 @cindex @code{a.out} symbol attributes
3737 @cindex symbol attributes, @code{a.out}
3741 * Symbol Desc:: Descriptor
3742 * Symbol Other:: Other
3746 @subsubsection Descriptor
3748 @cindex descriptor, of @code{a.out} symbol
3749 This is an arbitrary 16-bit value. You may establish a symbol's
3750 descriptor value by using a @code{.desc} statement
3751 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3752 @command{@value{AS}}.
3755 @subsubsection Other
3757 @cindex other attribute, of @code{a.out} symbol
3758 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3763 @subsection Symbol Attributes for COFF
3765 @cindex COFF symbol attributes
3766 @cindex symbol attributes, COFF
3768 The COFF format supports a multitude of auxiliary symbol attributes;
3769 like the primary symbol attributes, they are set between @code{.def} and
3770 @code{.endef} directives.
3772 @subsubsection Primary Attributes
3774 @cindex primary attributes, COFF symbols
3775 The symbol name is set with @code{.def}; the value and type,
3776 respectively, with @code{.val} and @code{.type}.
3778 @subsubsection Auxiliary Attributes
3780 @cindex auxiliary attributes, COFF symbols
3781 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3782 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3783 table information for COFF.
3788 @subsection Symbol Attributes for SOM
3790 @cindex SOM symbol attributes
3791 @cindex symbol attributes, SOM
3793 The SOM format for the HPPA supports a multitude of symbol attributes set with
3794 the @code{.EXPORT} and @code{.IMPORT} directives.
3796 The attributes are described in @cite{HP9000 Series 800 Assembly
3797 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3798 @code{EXPORT} assembler directive documentation.
3802 @chapter Expressions
3806 @cindex numeric values
3807 An @dfn{expression} specifies an address or numeric value.
3808 Whitespace may precede and/or follow an expression.
3810 The result of an expression must be an absolute number, or else an offset into
3811 a particular section. If an expression is not absolute, and there is not
3812 enough information when @command{@value{AS}} sees the expression to know its
3813 section, a second pass over the source program might be necessary to interpret
3814 the expression---but the second pass is currently not implemented.
3815 @command{@value{AS}} aborts with an error message in this situation.
3818 * Empty Exprs:: Empty Expressions
3819 * Integer Exprs:: Integer Expressions
3823 @section Empty Expressions
3825 @cindex empty expressions
3826 @cindex expressions, empty
3827 An empty expression has no value: it is just whitespace or null.
3828 Wherever an absolute expression is required, you may omit the
3829 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
3830 is compatible with other assemblers.
3833 @section Integer Expressions
3835 @cindex integer expressions
3836 @cindex expressions, integer
3837 An @dfn{integer expression} is one or more @emph{arguments} delimited
3838 by @emph{operators}.
3841 * Arguments:: Arguments
3842 * Operators:: Operators
3843 * Prefix Ops:: Prefix Operators
3844 * Infix Ops:: Infix Operators
3848 @subsection Arguments
3850 @cindex expression arguments
3851 @cindex arguments in expressions
3852 @cindex operands in expressions
3853 @cindex arithmetic operands
3854 @dfn{Arguments} are symbols, numbers or subexpressions. In other
3855 contexts arguments are sometimes called ``arithmetic operands''. In
3856 this manual, to avoid confusing them with the ``instruction operands'' of
3857 the machine language, we use the term ``argument'' to refer to parts of
3858 expressions only, reserving the word ``operand'' to refer only to machine
3859 instruction operands.
3861 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
3862 @var{section} is one of text, data, bss, absolute,
3863 or undefined. @var{NNN} is a signed, 2's complement 32 bit
3866 Numbers are usually integers.
3868 A number can be a flonum or bignum. In this case, you are warned
3869 that only the low order 32 bits are used, and @command{@value{AS}} pretends
3870 these 32 bits are an integer. You may write integer-manipulating
3871 instructions that act on exotic constants, compatible with other
3874 @cindex subexpressions
3875 Subexpressions are a left parenthesis @samp{(} followed by an integer
3876 expression, followed by a right parenthesis @samp{)}; or a prefix
3877 operator followed by an argument.
3880 @subsection Operators
3882 @cindex operators, in expressions
3883 @cindex arithmetic functions
3884 @cindex functions, in expressions
3885 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
3886 operators are followed by an argument. Infix operators appear
3887 between their arguments. Operators may be preceded and/or followed by
3891 @subsection Prefix Operator
3893 @cindex prefix operators
3894 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
3895 one argument, which must be absolute.
3897 @c the tex/end tex stuff surrounding this small table is meant to make
3898 @c it align, on the printed page, with the similar table in the next
3899 @c section (which is inside an enumerate).
3901 \global\advance\leftskip by \itemindent
3906 @dfn{Negation}. Two's complement negation.
3908 @dfn{Complementation}. Bitwise not.
3912 \global\advance\leftskip by -\itemindent
3916 @subsection Infix Operators
3918 @cindex infix operators
3919 @cindex operators, permitted arguments
3920 @dfn{Infix operators} take two arguments, one on either side. Operators
3921 have precedence, but operations with equal precedence are performed left
3922 to right. Apart from @code{+} or @option{-}, both arguments must be
3923 absolute, and the result is absolute.
3926 @cindex operator precedence
3927 @cindex precedence of operators
3934 @dfn{Multiplication}.
3937 @dfn{Division}. Truncation is the same as the C operator @samp{/}
3943 @dfn{Shift Left}. Same as the C operator @samp{<<}.
3946 @dfn{Shift Right}. Same as the C operator @samp{>>}.
3950 Intermediate precedence
3955 @dfn{Bitwise Inclusive Or}.
3961 @dfn{Bitwise Exclusive Or}.
3964 @dfn{Bitwise Or Not}.
3971 @cindex addition, permitted arguments
3972 @cindex plus, permitted arguments
3973 @cindex arguments for addition
3975 @dfn{Addition}. If either argument is absolute, the result has the section of
3976 the other argument. You may not add together arguments from different
3979 @cindex subtraction, permitted arguments
3980 @cindex minus, permitted arguments
3981 @cindex arguments for subtraction
3983 @dfn{Subtraction}. If the right argument is absolute, the
3984 result has the section of the left argument.
3985 If both arguments are in the same section, the result is absolute.
3986 You may not subtract arguments from different sections.
3987 @c FIXME is there still something useful to say about undefined - undefined ?
3989 @cindex comparison expressions
3990 @cindex expressions, comparison
3995 @dfn{Is Not Equal To}
3999 @dfn{Is Greater Than}
4001 @dfn{Is Greater Than Or Equal To}
4003 @dfn{Is Less Than Or Equal To}
4005 The comparison operators can be used as infix operators. A true results has a
4006 value of -1 whereas a false result has a value of 0. Note, these operators
4007 perform signed comparisons.
4010 @item Lowest Precedence
4019 These two logical operations can be used to combine the results of sub
4020 expressions. Note, unlike the comparison operators a true result returns a
4021 value of 1 but a false results does still return 0. Also note that the logical
4022 or operator has a slightly lower precedence than logical and.
4027 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4028 address; you can only have a defined section in one of the two arguments.
4031 @chapter Assembler Directives
4033 @cindex directives, machine independent
4034 @cindex pseudo-ops, machine independent
4035 @cindex machine independent directives
4036 All assembler directives have names that begin with a period (@samp{.}).
4037 The rest of the name is letters, usually in lower case.
4039 This chapter discusses directives that are available regardless of the
4040 target machine configuration for the @sc{gnu} assembler.
4042 Some machine configurations provide additional directives.
4043 @xref{Machine Dependencies}.
4046 @ifset machine-directives
4047 @xref{Machine Dependencies}, for additional directives.
4052 * Abort:: @code{.abort}
4054 * ABORT (COFF):: @code{.ABORT}
4057 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4058 * Altmacro:: @code{.altmacro}
4059 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4060 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4061 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4062 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, @code{.bundle_lock}, @code{.bundle_unlock}
4063 * Byte:: @code{.byte @var{expressions}}
4064 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4065 * Comm:: @code{.comm @var{symbol} , @var{length} }
4066 * Data:: @code{.data @var{subsection}}
4068 * Def:: @code{.def @var{name}}
4071 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4077 * Double:: @code{.double @var{flonums}}
4078 * Eject:: @code{.eject}
4079 * Else:: @code{.else}
4080 * Elseif:: @code{.elseif}
4083 * Endef:: @code{.endef}
4086 * Endfunc:: @code{.endfunc}
4087 * Endif:: @code{.endif}
4088 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4089 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4090 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4092 * Error:: @code{.error @var{string}}
4093 * Exitm:: @code{.exitm}
4094 * Extern:: @code{.extern}
4095 * Fail:: @code{.fail}
4096 * File:: @code{.file}
4097 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4098 * Float:: @code{.float @var{flonums}}
4099 * Func:: @code{.func}
4100 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4102 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4103 * Hidden:: @code{.hidden @var{names}}
4106 * hword:: @code{.hword @var{expressions}}
4107 * Ident:: @code{.ident}
4108 * If:: @code{.if @var{absolute expression}}
4109 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4110 * Include:: @code{.include "@var{file}"}
4111 * Int:: @code{.int @var{expressions}}
4113 * Internal:: @code{.internal @var{names}}
4116 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4117 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4118 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4119 * Lflags:: @code{.lflags}
4120 @ifclear no-line-dir
4121 * Line:: @code{.line @var{line-number}}
4124 * Linkonce:: @code{.linkonce [@var{type}]}
4125 * List:: @code{.list}
4126 * Ln:: @code{.ln @var{line-number}}
4127 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4128 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4130 * Local:: @code{.local @var{names}}
4133 * Long:: @code{.long @var{expressions}}
4135 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4138 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4139 * MRI:: @code{.mri @var{val}}
4140 * Noaltmacro:: @code{.noaltmacro}
4141 * Nolist:: @code{.nolist}
4142 * Octa:: @code{.octa @var{bignums}}
4143 * Offset:: @code{.offset @var{loc}}
4144 * Org:: @code{.org @var{new-lc}, @var{fill}}
4145 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4147 * PopSection:: @code{.popsection}
4148 * Previous:: @code{.previous}
4151 * Print:: @code{.print @var{string}}
4153 * Protected:: @code{.protected @var{names}}
4156 * Psize:: @code{.psize @var{lines}, @var{columns}}
4157 * Purgem:: @code{.purgem @var{name}}
4159 * PushSection:: @code{.pushsection @var{name}}
4162 * Quad:: @code{.quad @var{bignums}}
4163 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4164 * Rept:: @code{.rept @var{count}}
4165 * Sbttl:: @code{.sbttl "@var{subheading}"}
4167 * Scl:: @code{.scl @var{class}}
4170 * Section:: @code{.section @var{name}[, @var{flags}]}
4173 * Set:: @code{.set @var{symbol}, @var{expression}}
4174 * Short:: @code{.short @var{expressions}}
4175 * Single:: @code{.single @var{flonums}}
4177 * Size:: @code{.size [@var{name} , @var{expression}]}
4179 @ifclear no-space-dir
4180 * Skip:: @code{.skip @var{size} , @var{fill}}
4183 * Sleb128:: @code{.sleb128 @var{expressions}}
4184 @ifclear no-space-dir
4185 * Space:: @code{.space @var{size} , @var{fill}}
4188 * Stab:: @code{.stabd, .stabn, .stabs}
4191 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4192 * Struct:: @code{.struct @var{expression}}
4194 * SubSection:: @code{.subsection}
4195 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4199 * Tag:: @code{.tag @var{structname}}
4202 * Text:: @code{.text @var{subsection}}
4203 * Title:: @code{.title "@var{heading}"}
4205 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4208 * Uleb128:: @code{.uleb128 @var{expressions}}
4210 * Val:: @code{.val @var{addr}}
4214 * Version:: @code{.version "@var{string}"}
4215 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4216 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4219 * Warning:: @code{.warning @var{string}}
4220 * Weak:: @code{.weak @var{names}}
4221 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4222 * Word:: @code{.word @var{expressions}}
4223 * Deprecated:: Deprecated Directives
4227 @section @code{.abort}
4229 @cindex @code{abort} directive
4230 @cindex stopping the assembly
4231 This directive stops the assembly immediately. It is for
4232 compatibility with other assemblers. The original idea was that the
4233 assembly language source would be piped into the assembler. If the sender
4234 of the source quit, it could use this directive tells @command{@value{AS}} to
4235 quit also. One day @code{.abort} will not be supported.
4239 @section @code{.ABORT} (COFF)
4241 @cindex @code{ABORT} directive
4242 When producing COFF output, @command{@value{AS}} accepts this directive as a
4243 synonym for @samp{.abort}.
4246 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4252 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4254 @cindex padding the location counter
4255 @cindex @code{align} directive
4256 Pad the location counter (in the current subsection) to a particular storage
4257 boundary. The first expression (which must be absolute) is the alignment
4258 required, as described below.
4260 The second expression (also absolute) gives the fill value to be stored in the
4261 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4262 padding bytes are normally zero. However, on some systems, if the section is
4263 marked as containing code and the fill value is omitted, the space is filled
4264 with no-op instructions.
4266 The third expression is also absolute, and is also optional. If it is present,
4267 it is the maximum number of bytes that should be skipped by this alignment
4268 directive. If doing the alignment would require skipping more bytes than the
4269 specified maximum, then the alignment is not done at all. You can omit the
4270 fill value (the second argument) entirely by simply using two commas after the
4271 required alignment; this can be useful if you want the alignment to be filled
4272 with no-op instructions when appropriate.
4274 The way the required alignment is specified varies from system to system.
4275 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or32,
4276 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4277 alignment request in bytes. For example @samp{.align 8} advances
4278 the location counter until it is a multiple of 8. If the location counter
4279 is already a multiple of 8, no change is needed. For the tic54x, the
4280 first expression is the alignment request in words.
4282 For other systems, including ppc, i386 using a.out format, arm and
4283 strongarm, it is the
4284 number of low-order zero bits the location counter must have after
4285 advancement. For example @samp{.align 3} advances the location
4286 counter until it a multiple of 8. If the location counter is already a
4287 multiple of 8, no change is needed.
4289 This inconsistency is due to the different behaviors of the various
4290 native assemblers for these systems which GAS must emulate.
4291 GAS also provides @code{.balign} and @code{.p2align} directives,
4292 described later, which have a consistent behavior across all
4293 architectures (but are specific to GAS).
4296 @section @code{.altmacro}
4297 Enable alternate macro mode, enabling:
4300 @item LOCAL @var{name} [ , @dots{} ]
4301 One additional directive, @code{LOCAL}, is available. It is used to
4302 generate a string replacement for each of the @var{name} arguments, and
4303 replace any instances of @var{name} in each macro expansion. The
4304 replacement string is unique in the assembly, and different for each
4305 separate macro expansion. @code{LOCAL} allows you to write macros that
4306 define symbols, without fear of conflict between separate macro expansions.
4308 @item String delimiters
4309 You can write strings delimited in these other ways besides
4310 @code{"@var{string}"}:
4313 @item '@var{string}'
4314 You can delimit strings with single-quote characters.
4316 @item <@var{string}>
4317 You can delimit strings with matching angle brackets.
4320 @item single-character string escape
4321 To include any single character literally in a string (even if the
4322 character would otherwise have some special meaning), you can prefix the
4323 character with @samp{!} (an exclamation mark). For example, you can
4324 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4326 @item Expression results as strings
4327 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4328 and use the result as a string.
4332 @section @code{.ascii "@var{string}"}@dots{}
4334 @cindex @code{ascii} directive
4335 @cindex string literals
4336 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4337 separated by commas. It assembles each string (with no automatic
4338 trailing zero byte) into consecutive addresses.
4341 @section @code{.asciz "@var{string}"}@dots{}
4343 @cindex @code{asciz} directive
4344 @cindex zero-terminated strings
4345 @cindex null-terminated strings
4346 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4347 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4350 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4352 @cindex padding the location counter given number of bytes
4353 @cindex @code{balign} directive
4354 Pad the location counter (in the current subsection) to a particular
4355 storage boundary. The first expression (which must be absolute) is the
4356 alignment request in bytes. For example @samp{.balign 8} advances
4357 the location counter until it is a multiple of 8. If the location counter
4358 is already a multiple of 8, no change is needed.
4360 The second expression (also absolute) gives the fill value to be stored in the
4361 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4362 padding bytes are normally zero. However, on some systems, if the section is
4363 marked as containing code and the fill value is omitted, the space is filled
4364 with no-op instructions.
4366 The third expression is also absolute, and is also optional. If it is present,
4367 it is the maximum number of bytes that should be skipped by this alignment
4368 directive. If doing the alignment would require skipping more bytes than the
4369 specified maximum, then the alignment is not done at all. You can omit the
4370 fill value (the second argument) entirely by simply using two commas after the
4371 required alignment; this can be useful if you want the alignment to be filled
4372 with no-op instructions when appropriate.
4374 @cindex @code{balignw} directive
4375 @cindex @code{balignl} directive
4376 The @code{.balignw} and @code{.balignl} directives are variants of the
4377 @code{.balign} directive. The @code{.balignw} directive treats the fill
4378 pattern as a two byte word value. The @code{.balignl} directives treats the
4379 fill pattern as a four byte longword value. For example, @code{.balignw
4380 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4381 filled in with the value 0x368d (the exact placement of the bytes depends upon
4382 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4385 @node Bundle directives
4386 @section @code{.bundle_align_mode @var{abs-expr}}
4387 @cindex @code{bundle_align_mode} directive
4389 @cindex instruction bundle
4390 @cindex aligned instruction bundle
4391 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4392 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4393 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4394 disabled (which is the default state). If the argument it not zero, it
4395 gives the size of an instruction bundle as a power of two (as for the
4396 @code{.p2align} directive, @pxref{P2align}).
4398 For some targets, it's an ABI requirement that no instruction may span a
4399 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4400 instructions that starts on an aligned boundary. For example, if
4401 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4402 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4403 effect, no single instruction may span a boundary between bundles. If an
4404 instruction would start too close to the end of a bundle for the length of
4405 that particular instruction to fit within the bundle, then the space at the
4406 end of that bundle is filled with no-op instructions so the instruction
4407 starts in the next bundle. As a corollary, it's an error if any single
4408 instruction's encoding is longer than the bundle size.
4410 @section @code{.bundle_lock} and @code{.bundle_unlock}
4411 @cindex @code{bundle_lock} directive
4412 @cindex @code{bundle_unlock} directive
4413 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4414 allow explicit control over instruction bundle padding. These directives
4415 are only valid when @code{.bundle_align_mode} has been used to enable
4416 aligned instruction bundle mode. It's an error if they appear when
4417 @code{.bundle_align_mode} has not been used at all, or when the last
4418 directive was @w{@code{.bundle_align_mode 0}}.
4420 @cindex bundle-locked
4421 For some targets, it's an ABI requirement that certain instructions may
4422 appear only as part of specified permissible sequences of multiple
4423 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4424 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4425 instruction sequence. For purposes of aligned instruction bundle mode, a
4426 sequence starting with @code{.bundle_lock} and ending with
4427 @code{.bundle_unlock} is treated as a single instruction. That is, the
4428 entire sequence must fit into a single bundle and may not span a bundle
4429 boundary. If necessary, no-op instructions will be inserted before the
4430 first instruction of the sequence so that the whole sequence starts on an
4431 aligned bundle boundary. It's an error if the sequence is longer than the
4434 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4435 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4436 nested. That is, a second @code{.bundle_lock} directive before the next
4437 @code{.bundle_unlock} directive has no effect except that it must be
4438 matched by another closing @code{.bundle_unlock} so that there is the
4439 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4442 @section @code{.byte @var{expressions}}
4444 @cindex @code{byte} directive
4445 @cindex integers, one byte
4446 @code{.byte} expects zero or more expressions, separated by commas.
4447 Each expression is assembled into the next byte.
4449 @node CFI directives
4450 @section @code{.cfi_sections @var{section_list}}
4451 @cindex @code{cfi_sections} directive
4452 @code{.cfi_sections} may be used to specify whether CFI directives
4453 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4454 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4455 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4456 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4457 directive is not used is @code{.cfi_sections .eh_frame}.
4459 @section @code{.cfi_startproc [simple]}
4460 @cindex @code{cfi_startproc} directive
4461 @code{.cfi_startproc} is used at the beginning of each function that
4462 should have an entry in @code{.eh_frame}. It initializes some internal
4463 data structures. Don't forget to close the function by
4464 @code{.cfi_endproc}.
4466 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4467 it also emits some architecture dependent initial CFI instructions.
4469 @section @code{.cfi_endproc}
4470 @cindex @code{cfi_endproc} directive
4471 @code{.cfi_endproc} is used at the end of a function where it closes its
4472 unwind entry previously opened by
4473 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4475 @section @code{.cfi_personality @var{encoding} [, @var{exp}]}
4476 @code{.cfi_personality} defines personality routine and its encoding.
4477 @var{encoding} must be a constant determining how the personality
4478 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4479 argument is not present, otherwise second argument should be
4480 a constant or a symbol name. When using indirect encodings,
4481 the symbol provided should be the location where personality
4482 can be loaded from, not the personality routine itself.
4483 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4484 no personality routine.
4486 @section @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4487 @code{.cfi_lsda} defines LSDA and its encoding.
4488 @var{encoding} must be a constant determining how the LSDA
4489 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4490 argument is not present, otherwise second argument should be a constant
4491 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4494 @section @code{.cfi_def_cfa @var{register}, @var{offset}}
4495 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4496 address from @var{register} and add @var{offset} to it}.
4498 @section @code{.cfi_def_cfa_register @var{register}}
4499 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4500 now on @var{register} will be used instead of the old one. Offset
4503 @section @code{.cfi_def_cfa_offset @var{offset}}
4504 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4505 remains the same, but @var{offset} is new. Note that it is the
4506 absolute offset that will be added to a defined register to compute
4509 @section @code{.cfi_adjust_cfa_offset @var{offset}}
4510 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4511 value that is added/substracted from the previous offset.
4513 @section @code{.cfi_offset @var{register}, @var{offset}}
4514 Previous value of @var{register} is saved at offset @var{offset} from
4517 @section @code{.cfi_rel_offset @var{register}, @var{offset}}
4518 Previous value of @var{register} is saved at offset @var{offset} from
4519 the current CFA register. This is transformed to @code{.cfi_offset}
4520 using the known displacement of the CFA register from the CFA.
4521 This is often easier to use, because the number will match the
4522 code it's annotating.
4524 @section @code{.cfi_register @var{register1}, @var{register2}}
4525 Previous value of @var{register1} is saved in register @var{register2}.
4527 @section @code{.cfi_restore @var{register}}
4528 @code{.cfi_restore} says that the rule for @var{register} is now the
4529 same as it was at the beginning of the function, after all initial
4530 instruction added by @code{.cfi_startproc} were executed.
4532 @section @code{.cfi_undefined @var{register}}
4533 From now on the previous value of @var{register} can't be restored anymore.
4535 @section @code{.cfi_same_value @var{register}}
4536 Current value of @var{register} is the same like in the previous frame,
4537 i.e. no restoration needed.
4539 @section @code{.cfi_remember_state},
4540 First save all current rules for all registers by @code{.cfi_remember_state},
4541 then totally screw them up by subsequent @code{.cfi_*} directives and when
4542 everything is hopelessly bad, use @code{.cfi_restore_state} to restore
4543 the previous saved state.
4545 @section @code{.cfi_return_column @var{register}}
4546 Change return column @var{register}, i.e. the return address is either
4547 directly in @var{register} or can be accessed by rules for @var{register}.
4549 @section @code{.cfi_signal_frame}
4550 Mark current function as signal trampoline.
4552 @section @code{.cfi_window_save}
4553 SPARC register window has been saved.
4555 @section @code{.cfi_escape} @var{expression}[, @dots{}]
4556 Allows the user to add arbitrary bytes to the unwind info. One
4557 might use this to add OS-specific CFI opcodes, or generic CFI
4558 opcodes that GAS does not yet support.
4560 @section @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4561 The current value of @var{register} is @var{label}. The value of @var{label}
4562 will be encoded in the output file according to @var{encoding}; see the
4563 description of @code{.cfi_personality} for details on this encoding.
4565 The usefulness of equating a register to a fixed label is probably
4566 limited to the return address register. Here, it can be useful to
4567 mark a code segment that has only one return address which is reached
4568 by a direct branch and no copy of the return address exists in memory
4569 or another register.
4572 @section @code{.comm @var{symbol} , @var{length} }
4574 @cindex @code{comm} directive
4575 @cindex symbol, common
4576 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4577 common symbol in one object file may be merged with a defined or common symbol
4578 of the same name in another object file. If @code{@value{LD}} does not see a
4579 definition for the symbol--just one or more common symbols--then it will
4580 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4581 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4582 the same name, and they do not all have the same size, it will allocate space
4583 using the largest size.
4586 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4587 an optional third argument. This is the desired alignment of the symbol,
4588 specified for ELF as a byte boundary (for example, an alignment of 16 means
4589 that the least significant 4 bits of the address should be zero), and for PE
4590 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4591 boundary). The alignment must be an absolute expression, and it must be a
4592 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4593 common symbol, it will use the alignment when placing the symbol. If no
4594 alignment is specified, @command{@value{AS}} will set the alignment to the
4595 largest power of two less than or equal to the size of the symbol, up to a
4596 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4597 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4598 @samp{--section-alignment} option; image file sections in PE are aligned to
4599 multiples of 4096, which is far too large an alignment for ordinary variables.
4600 It is rather the default alignment for (non-debug) sections within object
4601 (@samp{*.o}) files, which are less strictly aligned.}.
4605 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4606 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4610 @section @code{.data @var{subsection}}
4612 @cindex @code{data} directive
4613 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4614 end of the data subsection numbered @var{subsection} (which is an
4615 absolute expression). If @var{subsection} is omitted, it defaults
4620 @section @code{.def @var{name}}
4622 @cindex @code{def} directive
4623 @cindex COFF symbols, debugging
4624 @cindex debugging COFF symbols
4625 Begin defining debugging information for a symbol @var{name}; the
4626 definition extends until the @code{.endef} directive is encountered.
4629 This directive is only observed when @command{@value{AS}} is configured for COFF
4630 format output; when producing @code{b.out}, @samp{.def} is recognized,
4637 @section @code{.desc @var{symbol}, @var{abs-expression}}
4639 @cindex @code{desc} directive
4640 @cindex COFF symbol descriptor
4641 @cindex symbol descriptor, COFF
4642 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
4643 to the low 16 bits of an absolute expression.
4646 The @samp{.desc} directive is not available when @command{@value{AS}} is
4647 configured for COFF output; it is only for @code{a.out} or @code{b.out}
4648 object format. For the sake of compatibility, @command{@value{AS}} accepts
4649 it, but produces no output, when configured for COFF.
4655 @section @code{.dim}
4657 @cindex @code{dim} directive
4658 @cindex COFF auxiliary symbol information
4659 @cindex auxiliary symbol information, COFF
4660 This directive is generated by compilers to include auxiliary debugging
4661 information in the symbol table. It is only permitted inside
4662 @code{.def}/@code{.endef} pairs.
4665 @samp{.dim} is only meaningful when generating COFF format output; when
4666 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
4672 @section @code{.double @var{flonums}}
4674 @cindex @code{double} directive
4675 @cindex floating point numbers (double)
4676 @code{.double} expects zero or more flonums, separated by commas. It
4677 assembles floating point numbers.
4679 The exact kind of floating point numbers emitted depends on how
4680 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
4684 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
4685 in @sc{ieee} format.
4690 @section @code{.eject}
4692 @cindex @code{eject} directive
4693 @cindex new page, in listings
4694 @cindex page, in listings
4695 @cindex listing control: new page
4696 Force a page break at this point, when generating assembly listings.
4699 @section @code{.else}
4701 @cindex @code{else} directive
4702 @code{.else} is part of the @command{@value{AS}} support for conditional
4703 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
4704 of code to be assembled if the condition for the preceding @code{.if}
4708 @section @code{.elseif}
4710 @cindex @code{elseif} directive
4711 @code{.elseif} is part of the @command{@value{AS}} support for conditional
4712 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
4713 @code{.if} block that would otherwise fill the entire @code{.else} section.
4716 @section @code{.end}
4718 @cindex @code{end} directive
4719 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
4720 process anything in the file past the @code{.end} directive.
4724 @section @code{.endef}
4726 @cindex @code{endef} directive
4727 This directive flags the end of a symbol definition begun with
4731 @samp{.endef} is only meaningful when generating COFF format output; if
4732 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
4733 directive but ignores it.
4738 @section @code{.endfunc}
4739 @cindex @code{endfunc} directive
4740 @code{.endfunc} marks the end of a function specified with @code{.func}.
4743 @section @code{.endif}
4745 @cindex @code{endif} directive
4746 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
4747 it marks the end of a block of code that is only assembled
4748 conditionally. @xref{If,,@code{.if}}.
4751 @section @code{.equ @var{symbol}, @var{expression}}
4753 @cindex @code{equ} directive
4754 @cindex assigning values to symbols
4755 @cindex symbols, assigning values to
4756 This directive sets the value of @var{symbol} to @var{expression}.
4757 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
4760 The syntax for @code{equ} on the HPPA is
4761 @samp{@var{symbol} .equ @var{expression}}.
4765 The syntax for @code{equ} on the Z80 is
4766 @samp{@var{symbol} equ @var{expression}}.
4767 On the Z80 it is an eror if @var{symbol} is already defined,
4768 but the symbol is not protected from later redefinition.
4769 Compare @ref{Equiv}.
4773 @section @code{.equiv @var{symbol}, @var{expression}}
4774 @cindex @code{equiv} directive
4775 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
4776 the assembler will signal an error if @var{symbol} is already defined. Note a
4777 symbol which has been referenced but not actually defined is considered to be
4780 Except for the contents of the error message, this is roughly equivalent to
4787 plus it protects the symbol from later redefinition.
4790 @section @code{.eqv @var{symbol}, @var{expression}}
4791 @cindex @code{eqv} directive
4792 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
4793 evaluate the expression or any part of it immediately. Instead each time
4794 the resulting symbol is used in an expression, a snapshot of its current
4798 @section @code{.err}
4799 @cindex @code{err} directive
4800 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
4801 message and, unless the @option{-Z} option was used, it will not generate an
4802 object file. This can be used to signal an error in conditionally compiled code.
4805 @section @code{.error "@var{string}"}
4806 @cindex error directive
4808 Similarly to @code{.err}, this directive emits an error, but you can specify a
4809 string that will be emitted as the error message. If you don't specify the
4810 message, it defaults to @code{".error directive invoked in source file"}.
4811 @xref{Errors, ,Error and Warning Messages}.
4814 .error "This code has not been assembled and tested."
4818 @section @code{.exitm}
4819 Exit early from the current macro definition. @xref{Macro}.
4822 @section @code{.extern}
4824 @cindex @code{extern} directive
4825 @code{.extern} is accepted in the source program---for compatibility
4826 with other assemblers---but it is ignored. @command{@value{AS}} treats
4827 all undefined symbols as external.
4830 @section @code{.fail @var{expression}}
4832 @cindex @code{fail} directive
4833 Generates an error or a warning. If the value of the @var{expression} is 500
4834 or more, @command{@value{AS}} will print a warning message. If the value is less
4835 than 500, @command{@value{AS}} will print an error message. The message will
4836 include the value of @var{expression}. This can occasionally be useful inside
4837 complex nested macros or conditional assembly.
4840 @section @code{.file}
4841 @cindex @code{file} directive
4843 @ifclear no-file-dir
4844 There are two different versions of the @code{.file} directive. Targets
4845 that support DWARF2 line number information use the DWARF2 version of
4846 @code{.file}. Other targets use the default version.
4848 @subheading Default Version
4850 @cindex logical file name
4851 @cindex file name, logical
4852 This version of the @code{.file} directive tells @command{@value{AS}} that we
4853 are about to start a new logical file. The syntax is:
4859 @var{string} is the new file name. In general, the filename is
4860 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
4861 to specify an empty file name, you must give the quotes--@code{""}. This
4862 statement may go away in future: it is only recognized to be compatible with
4863 old @command{@value{AS}} programs.
4865 @subheading DWARF2 Version
4868 When emitting DWARF2 line number information, @code{.file} assigns filenames
4869 to the @code{.debug_line} file name table. The syntax is:
4872 .file @var{fileno} @var{filename}
4875 The @var{fileno} operand should be a unique positive integer to use as the
4876 index of the entry in the table. The @var{filename} operand is a C string
4879 The detail of filename indices is exposed to the user because the filename
4880 table is shared with the @code{.debug_info} section of the DWARF2 debugging
4881 information, and thus the user must know the exact indices that table
4885 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
4887 @cindex @code{fill} directive
4888 @cindex writing patterns in memory
4889 @cindex patterns, writing in memory
4890 @var{repeat}, @var{size} and @var{value} are absolute expressions.
4891 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
4892 may be zero or more. @var{Size} may be zero or more, but if it is
4893 more than 8, then it is deemed to have the value 8, compatible with
4894 other people's assemblers. The contents of each @var{repeat} bytes
4895 is taken from an 8-byte number. The highest order 4 bytes are
4896 zero. The lowest order 4 bytes are @var{value} rendered in the
4897 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
4898 Each @var{size} bytes in a repetition is taken from the lowest order
4899 @var{size} bytes of this number. Again, this bizarre behavior is
4900 compatible with other people's assemblers.
4902 @var{size} and @var{value} are optional.
4903 If the second comma and @var{value} are absent, @var{value} is
4904 assumed zero. If the first comma and following tokens are absent,
4905 @var{size} is assumed to be 1.
4908 @section @code{.float @var{flonums}}
4910 @cindex floating point numbers (single)
4911 @cindex @code{float} directive
4912 This directive assembles zero or more flonums, separated by commas. It
4913 has the same effect as @code{.single}.
4915 The exact kind of floating point numbers emitted depends on how
4916 @command{@value{AS}} is configured.
4917 @xref{Machine Dependencies}.
4921 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
4922 in @sc{ieee} format.
4927 @section @code{.func @var{name}[,@var{label}]}
4928 @cindex @code{func} directive
4929 @code{.func} emits debugging information to denote function @var{name}, and
4930 is ignored unless the file is assembled with debugging enabled.
4931 Only @samp{--gstabs[+]} is currently supported.
4932 @var{label} is the entry point of the function and if omitted @var{name}
4933 prepended with the @samp{leading char} is used.
4934 @samp{leading char} is usually @code{_} or nothing, depending on the target.
4935 All functions are currently defined to have @code{void} return type.
4936 The function must be terminated with @code{.endfunc}.
4939 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4941 @cindex @code{global} directive
4942 @cindex symbol, making visible to linker
4943 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
4944 @var{symbol} in your partial program, its value is made available to
4945 other partial programs that are linked with it. Otherwise,
4946 @var{symbol} takes its attributes from a symbol of the same name
4947 from another file linked into the same program.
4949 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
4950 compatibility with other assemblers.
4953 On the HPPA, @code{.global} is not always enough to make it accessible to other
4954 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
4955 @xref{HPPA Directives, ,HPPA Assembler Directives}.
4960 @section @code{.gnu_attribute @var{tag},@var{value}}
4961 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
4964 @section @code{.hidden @var{names}}
4966 @cindex @code{hidden} directive
4968 This is one of the ELF visibility directives. The other two are
4969 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
4970 @code{.protected} (@pxref{Protected,,@code{.protected}}).
4972 This directive overrides the named symbols default visibility (which is set by
4973 their binding: local, global or weak). The directive sets the visibility to
4974 @code{hidden} which means that the symbols are not visible to other components.
4975 Such symbols are always considered to be @code{protected} as well.
4979 @section @code{.hword @var{expressions}}
4981 @cindex @code{hword} directive
4982 @cindex integers, 16-bit
4983 @cindex numbers, 16-bit
4984 @cindex sixteen bit integers
4985 This expects zero or more @var{expressions}, and emits
4986 a 16 bit number for each.
4989 This directive is a synonym for @samp{.short}; depending on the target
4990 architecture, it may also be a synonym for @samp{.word}.
4994 This directive is a synonym for @samp{.short}.
4997 This directive is a synonym for both @samp{.short} and @samp{.word}.
5002 @section @code{.ident}
5004 @cindex @code{ident} directive
5006 This directive is used by some assemblers to place tags in object files. The
5007 behavior of this directive varies depending on the target. When using the
5008 a.out object file format, @command{@value{AS}} simply accepts the directive for
5009 source-file compatibility with existing assemblers, but does not emit anything
5010 for it. When using COFF, comments are emitted to the @code{.comment} or
5011 @code{.rdata} section, depending on the target. When using ELF, comments are
5012 emitted to the @code{.comment} section.
5015 @section @code{.if @var{absolute expression}}
5017 @cindex conditional assembly
5018 @cindex @code{if} directive
5019 @code{.if} marks the beginning of a section of code which is only
5020 considered part of the source program being assembled if the argument
5021 (which must be an @var{absolute expression}) is non-zero. The end of
5022 the conditional section of code must be marked by @code{.endif}
5023 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5024 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5025 If you have several conditions to check, @code{.elseif} may be used to avoid
5026 nesting blocks if/else within each subsequent @code{.else} block.
5028 The following variants of @code{.if} are also supported:
5030 @cindex @code{ifdef} directive
5031 @item .ifdef @var{symbol}
5032 Assembles the following section of code if the specified @var{symbol}
5033 has been defined. Note a symbol which has been referenced but not yet defined
5034 is considered to be undefined.
5036 @cindex @code{ifb} directive
5037 @item .ifb @var{text}
5038 Assembles the following section of code if the operand is blank (empty).
5040 @cindex @code{ifc} directive
5041 @item .ifc @var{string1},@var{string2}
5042 Assembles the following section of code if the two strings are the same. The
5043 strings may be optionally quoted with single quotes. If they are not quoted,
5044 the first string stops at the first comma, and the second string stops at the
5045 end of the line. Strings which contain whitespace should be quoted. The
5046 string comparison is case sensitive.
5048 @cindex @code{ifeq} directive
5049 @item .ifeq @var{absolute expression}
5050 Assembles the following section of code if the argument is zero.
5052 @cindex @code{ifeqs} directive
5053 @item .ifeqs @var{string1},@var{string2}
5054 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5056 @cindex @code{ifge} directive
5057 @item .ifge @var{absolute expression}
5058 Assembles the following section of code if the argument is greater than or
5061 @cindex @code{ifgt} directive
5062 @item .ifgt @var{absolute expression}
5063 Assembles the following section of code if the argument is greater than zero.
5065 @cindex @code{ifle} directive
5066 @item .ifle @var{absolute expression}
5067 Assembles the following section of code if the argument is less than or equal
5070 @cindex @code{iflt} directive
5071 @item .iflt @var{absolute expression}
5072 Assembles the following section of code if the argument is less than zero.
5074 @cindex @code{ifnb} directive
5075 @item .ifnb @var{text}
5076 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5077 following section of code if the operand is non-blank (non-empty).
5079 @cindex @code{ifnc} directive
5080 @item .ifnc @var{string1},@var{string2}.
5081 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5082 following section of code if the two strings are not the same.
5084 @cindex @code{ifndef} directive
5085 @cindex @code{ifnotdef} directive
5086 @item .ifndef @var{symbol}
5087 @itemx .ifnotdef @var{symbol}
5088 Assembles the following section of code if the specified @var{symbol}
5089 has not been defined. Both spelling variants are equivalent. Note a symbol
5090 which has been referenced but not yet defined is considered to be undefined.
5092 @cindex @code{ifne} directive
5093 @item .ifne @var{absolute expression}
5094 Assembles the following section of code if the argument is not equal to zero
5095 (in other words, this is equivalent to @code{.if}).
5097 @cindex @code{ifnes} directive
5098 @item .ifnes @var{string1},@var{string2}
5099 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5100 following section of code if the two strings are not the same.
5104 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5106 @cindex @code{incbin} directive
5107 @cindex binary files, including
5108 The @code{incbin} directive includes @var{file} verbatim at the current
5109 location. You can control the search paths used with the @samp{-I} command-line
5110 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5113 The @var{skip} argument skips a number of bytes from the start of the
5114 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5115 read. Note that the data is not aligned in any way, so it is the user's
5116 responsibility to make sure that proper alignment is provided both before and
5117 after the @code{incbin} directive.
5120 @section @code{.include "@var{file}"}
5122 @cindex @code{include} directive
5123 @cindex supporting files, including
5124 @cindex files, including
5125 This directive provides a way to include supporting files at specified
5126 points in your source program. The code from @var{file} is assembled as
5127 if it followed the point of the @code{.include}; when the end of the
5128 included file is reached, assembly of the original file continues. You
5129 can control the search paths used with the @samp{-I} command-line option
5130 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5134 @section @code{.int @var{expressions}}
5136 @cindex @code{int} directive
5137 @cindex integers, 32-bit
5138 Expect zero or more @var{expressions}, of any section, separated by commas.
5139 For each expression, emit a number that, at run time, is the value of that
5140 expression. The byte order and bit size of the number depends on what kind
5141 of target the assembly is for.
5145 On most forms of the H8/300, @code{.int} emits 16-bit
5146 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5153 @section @code{.internal @var{names}}
5155 @cindex @code{internal} directive
5157 This is one of the ELF visibility directives. The other two are
5158 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5159 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5161 This directive overrides the named symbols default visibility (which is set by
5162 their binding: local, global or weak). The directive sets the visibility to
5163 @code{internal} which means that the symbols are considered to be @code{hidden}
5164 (i.e., not visible to other components), and that some extra, processor specific
5165 processing must also be performed upon the symbols as well.
5169 @section @code{.irp @var{symbol},@var{values}}@dots{}
5171 @cindex @code{irp} directive
5172 Evaluate a sequence of statements assigning different values to @var{symbol}.
5173 The sequence of statements starts at the @code{.irp} directive, and is
5174 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5175 set to @var{value}, and the sequence of statements is assembled. If no
5176 @var{value} is listed, the sequence of statements is assembled once, with
5177 @var{symbol} set to the null string. To refer to @var{symbol} within the
5178 sequence of statements, use @var{\symbol}.
5180 For example, assembling
5188 is equivalent to assembling
5196 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5199 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5201 @cindex @code{irpc} directive
5202 Evaluate a sequence of statements assigning different values to @var{symbol}.
5203 The sequence of statements starts at the @code{.irpc} directive, and is
5204 terminated by an @code{.endr} directive. For each character in @var{value},
5205 @var{symbol} is set to the character, and the sequence of statements is
5206 assembled. If no @var{value} is listed, the sequence of statements is
5207 assembled once, with @var{symbol} set to the null string. To refer to
5208 @var{symbol} within the sequence of statements, use @var{\symbol}.
5210 For example, assembling
5218 is equivalent to assembling
5226 For some caveats with the spelling of @var{symbol}, see also the discussion
5230 @section @code{.lcomm @var{symbol} , @var{length}}
5232 @cindex @code{lcomm} directive
5233 @cindex local common symbols
5234 @cindex symbols, local common
5235 Reserve @var{length} (an absolute expression) bytes for a local common
5236 denoted by @var{symbol}. The section and value of @var{symbol} are
5237 those of the new local common. The addresses are allocated in the bss
5238 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5239 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5240 not visible to @code{@value{LD}}.
5243 Some targets permit a third argument to be used with @code{.lcomm}. This
5244 argument specifies the desired alignment of the symbol in the bss section.
5248 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5249 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5253 @section @code{.lflags}
5255 @cindex @code{lflags} directive (ignored)
5256 @command{@value{AS}} accepts this directive, for compatibility with other
5257 assemblers, but ignores it.
5259 @ifclear no-line-dir
5261 @section @code{.line @var{line-number}}
5263 @cindex @code{line} directive
5264 @cindex logical line number
5266 Change the logical line number. @var{line-number} must be an absolute
5267 expression. The next line has that logical line number. Therefore any other
5268 statements on the current line (after a statement separator character) are
5269 reported as on logical line number @var{line-number} @minus{} 1. One day
5270 @command{@value{AS}} will no longer support this directive: it is recognized only
5271 for compatibility with existing assembler programs.
5274 Even though this is a directive associated with the @code{a.out} or
5275 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5276 when producing COFF output, and treats @samp{.line} as though it
5277 were the COFF @samp{.ln} @emph{if} it is found outside a
5278 @code{.def}/@code{.endef} pair.
5280 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5281 used by compilers to generate auxiliary symbol information for
5286 @section @code{.linkonce [@var{type}]}
5288 @cindex @code{linkonce} directive
5289 @cindex common sections
5290 Mark the current section so that the linker only includes a single copy of it.
5291 This may be used to include the same section in several different object files,
5292 but ensure that the linker will only include it once in the final output file.
5293 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5294 Duplicate sections are detected based on the section name, so it should be
5297 This directive is only supported by a few object file formats; as of this
5298 writing, the only object file format which supports it is the Portable
5299 Executable format used on Windows NT.
5301 The @var{type} argument is optional. If specified, it must be one of the
5302 following strings. For example:
5306 Not all types may be supported on all object file formats.
5310 Silently discard duplicate sections. This is the default.
5313 Warn if there are duplicate sections, but still keep only one copy.
5316 Warn if any of the duplicates have different sizes.
5319 Warn if any of the duplicates do not have exactly the same contents.
5323 @section @code{.list}
5325 @cindex @code{list} directive
5326 @cindex listing control, turning on
5327 Control (in conjunction with the @code{.nolist} directive) whether or
5328 not assembly listings are generated. These two directives maintain an
5329 internal counter (which is zero initially). @code{.list} increments the
5330 counter, and @code{.nolist} decrements it. Assembly listings are
5331 generated whenever the counter is greater than zero.
5333 By default, listings are disabled. When you enable them (with the
5334 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5335 the initial value of the listing counter is one.
5338 @section @code{.ln @var{line-number}}
5340 @cindex @code{ln} directive
5341 @ifclear no-line-dir
5342 @samp{.ln} is a synonym for @samp{.line}.
5345 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5346 must be an absolute expression. The next line has that logical
5347 line number, so any other statements on the current line (after a
5348 statement separator character @code{;}) are reported as on logical
5349 line number @var{line-number} @minus{} 1.
5352 This directive is accepted, but ignored, when @command{@value{AS}} is
5353 configured for @code{b.out}; its effect is only associated with COFF
5359 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5360 @cindex @code{loc} directive
5361 When emitting DWARF2 line number information,
5362 the @code{.loc} directive will add a row to the @code{.debug_line} line
5363 number matrix corresponding to the immediately following assembly
5364 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5365 arguments will be applied to the @code{.debug_line} state machine before
5368 The @var{options} are a sequence of the following tokens in any order:
5372 This option will set the @code{basic_block} register in the
5373 @code{.debug_line} state machine to @code{true}.
5376 This option will set the @code{prologue_end} register in the
5377 @code{.debug_line} state machine to @code{true}.
5379 @item epilogue_begin
5380 This option will set the @code{epilogue_begin} register in the
5381 @code{.debug_line} state machine to @code{true}.
5383 @item is_stmt @var{value}
5384 This option will set the @code{is_stmt} register in the
5385 @code{.debug_line} state machine to @code{value}, which must be
5388 @item isa @var{value}
5389 This directive will set the @code{isa} register in the @code{.debug_line}
5390 state machine to @var{value}, which must be an unsigned integer.
5392 @item discriminator @var{value}
5393 This directive will set the @code{discriminator} register in the @code{.debug_line}
5394 state machine to @var{value}, which must be an unsigned integer.
5398 @node Loc_mark_labels
5399 @section @code{.loc_mark_labels @var{enable}}
5400 @cindex @code{loc_mark_labels} directive
5401 When emitting DWARF2 line number information,
5402 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5403 to the @code{.debug_line} line number matrix with the @code{basic_block}
5404 register in the state machine set whenever a code label is seen.
5405 The @var{enable} argument should be either 1 or 0, to enable or disable
5406 this function respectively.
5410 @section @code{.local @var{names}}
5412 @cindex @code{local} directive
5413 This directive, which is available for ELF targets, marks each symbol in
5414 the comma-separated list of @code{names} as a local symbol so that it
5415 will not be externally visible. If the symbols do not already exist,
5416 they will be created.
5418 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5419 accept an alignment argument, which is the case for most ELF targets,
5420 the @code{.local} directive can be used in combination with @code{.comm}
5421 (@pxref{Comm}) to define aligned local common data.
5425 @section @code{.long @var{expressions}}
5427 @cindex @code{long} directive
5428 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5431 @c no one seems to know what this is for or whether this description is
5432 @c what it really ought to do
5434 @section @code{.lsym @var{symbol}, @var{expression}}
5436 @cindex @code{lsym} directive
5437 @cindex symbol, not referenced in assembly
5438 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5439 the hash table, ensuring it cannot be referenced by name during the
5440 rest of the assembly. This sets the attributes of the symbol to be
5441 the same as the expression value:
5443 @var{other} = @var{descriptor} = 0
5444 @var{type} = @r{(section of @var{expression})}
5445 @var{value} = @var{expression}
5448 The new symbol is not flagged as external.
5452 @section @code{.macro}
5455 The commands @code{.macro} and @code{.endm} allow you to define macros that
5456 generate assembly output. For example, this definition specifies a macro
5457 @code{sum} that puts a sequence of numbers into memory:
5460 .macro sum from=0, to=5
5469 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5481 @item .macro @var{macname}
5482 @itemx .macro @var{macname} @var{macargs} @dots{}
5483 @cindex @code{macro} directive
5484 Begin the definition of a macro called @var{macname}. If your macro
5485 definition requires arguments, specify their names after the macro name,
5486 separated by commas or spaces. You can qualify the macro argument to
5487 indicate whether all invocations must specify a non-blank value (through
5488 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5489 (through @samp{:@code{vararg}}). You can supply a default value for any
5490 macro argument by following the name with @samp{=@var{deflt}}. You
5491 cannot define two macros with the same @var{macname} unless it has been
5492 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5493 definitions. For example, these are all valid @code{.macro} statements:
5497 Begin the definition of a macro called @code{comm}, which takes no
5500 @item .macro plus1 p, p1
5501 @itemx .macro plus1 p p1
5502 Either statement begins the definition of a macro called @code{plus1},
5503 which takes two arguments; within the macro definition, write
5504 @samp{\p} or @samp{\p1} to evaluate the arguments.
5506 @item .macro reserve_str p1=0 p2
5507 Begin the definition of a macro called @code{reserve_str}, with two
5508 arguments. The first argument has a default value, but not the second.
5509 After the definition is complete, you can call the macro either as
5510 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5511 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5512 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5513 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5515 @item .macro m p1:req, p2=0, p3:vararg
5516 Begin the definition of a macro called @code{m}, with at least three
5517 arguments. The first argument must always have a value specified, but
5518 not the second, which instead has a default value. The third formal
5519 will get assigned all remaining arguments specified at invocation time.
5521 When you call a macro, you can specify the argument values either by
5522 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5523 @samp{sum to=17, from=9}.
5527 Note that since each of the @var{macargs} can be an identifier exactly
5528 as any other one permitted by the target architecture, there may be
5529 occasional problems if the target hand-crafts special meanings to certain
5530 characters when they occur in a special position. For example, if the colon
5531 (@code{:}) is generally permitted to be part of a symbol name, but the
5532 architecture specific code special-cases it when occurring as the final
5533 character of a symbol (to denote a label), then the macro parameter
5534 replacement code will have no way of knowing that and consider the whole
5535 construct (including the colon) an identifier, and check only this
5536 identifier for being the subject to parameter substitution. So for example
5537 this macro definition:
5545 might not work as expected. Invoking @samp{label foo} might not create a label
5546 called @samp{foo} but instead just insert the text @samp{\l:} into the
5547 assembler source, probably generating an error about an unrecognised
5550 Similarly problems might occur with the period character (@samp{.})
5551 which is often allowed inside opcode names (and hence identifier names). So
5552 for example constructing a macro to build an opcode from a base name and a
5553 length specifier like this:
5556 .macro opcode base length
5561 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5562 instruction but instead generate some kind of error as the assembler tries to
5563 interpret the text @samp{\base.\length}.
5565 There are several possible ways around this problem:
5568 @item Insert white space
5569 If it is possible to use white space characters then this is the simplest
5578 @item Use @samp{\()}
5579 The string @samp{\()} can be used to separate the end of a macro argument from
5580 the following text. eg:
5583 .macro opcode base length
5588 @item Use the alternate macro syntax mode
5589 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5590 used as a separator. eg:
5600 Note: this problem of correctly identifying string parameters to pseudo ops
5601 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5602 and @code{.irpc} (@pxref{Irpc}) as well.
5605 @cindex @code{endm} directive
5606 Mark the end of a macro definition.
5609 @cindex @code{exitm} directive
5610 Exit early from the current macro definition.
5612 @cindex number of macros executed
5613 @cindex macros, count executed
5615 @command{@value{AS}} maintains a counter of how many macros it has
5616 executed in this pseudo-variable; you can copy that number to your
5617 output with @samp{\@@}, but @emph{only within a macro definition}.
5619 @item LOCAL @var{name} [ , @dots{} ]
5620 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5621 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5622 @xref{Altmacro,,@code{.altmacro}}.
5626 @section @code{.mri @var{val}}
5628 @cindex @code{mri} directive
5629 @cindex MRI mode, temporarily
5630 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5631 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5632 affects code assembled until the next @code{.mri} directive, or until the end
5633 of the file. @xref{M, MRI mode, MRI mode}.
5636 @section @code{.noaltmacro}
5637 Disable alternate macro mode. @xref{Altmacro}.
5640 @section @code{.nolist}
5642 @cindex @code{nolist} directive
5643 @cindex listing control, turning off
5644 Control (in conjunction with the @code{.list} directive) whether or
5645 not assembly listings are generated. These two directives maintain an
5646 internal counter (which is zero initially). @code{.list} increments the
5647 counter, and @code{.nolist} decrements it. Assembly listings are
5648 generated whenever the counter is greater than zero.
5651 @section @code{.octa @var{bignums}}
5653 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
5654 @cindex @code{octa} directive
5655 @cindex integer, 16-byte
5656 @cindex sixteen byte integer
5657 This directive expects zero or more bignums, separated by commas. For each
5658 bignum, it emits a 16-byte integer.
5660 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
5661 hence @emph{octa}-word for 16 bytes.
5664 @section @code{.offset @var{loc}}
5666 @cindex @code{offset} directive
5667 Set the location counter to @var{loc} in the absolute section. @var{loc} must
5668 be an absolute expression. This directive may be useful for defining
5669 symbols with absolute values. Do not confuse it with the @code{.org}
5673 @section @code{.org @var{new-lc} , @var{fill}}
5675 @cindex @code{org} directive
5676 @cindex location counter, advancing
5677 @cindex advancing location counter
5678 @cindex current address, advancing
5679 Advance the location counter of the current section to
5680 @var{new-lc}. @var{new-lc} is either an absolute expression or an
5681 expression with the same section as the current subsection. That is,
5682 you can't use @code{.org} to cross sections: if @var{new-lc} has the
5683 wrong section, the @code{.org} directive is ignored. To be compatible
5684 with former assemblers, if the section of @var{new-lc} is absolute,
5685 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
5686 is the same as the current subsection.
5688 @code{.org} may only increase the location counter, or leave it
5689 unchanged; you cannot use @code{.org} to move the location counter
5692 @c double negative used below "not undefined" because this is a specific
5693 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
5694 @c section. doc@cygnus.com 18feb91
5695 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
5696 may not be undefined. If you really detest this restriction we eagerly await
5697 a chance to share your improved assembler.
5699 Beware that the origin is relative to the start of the section, not
5700 to the start of the subsection. This is compatible with other
5701 people's assemblers.
5703 When the location counter (of the current subsection) is advanced, the
5704 intervening bytes are filled with @var{fill} which should be an
5705 absolute expression. If the comma and @var{fill} are omitted,
5706 @var{fill} defaults to zero.
5709 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
5711 @cindex padding the location counter given a power of two
5712 @cindex @code{p2align} directive
5713 Pad the location counter (in the current subsection) to a particular
5714 storage boundary. The first expression (which must be absolute) is the
5715 number of low-order zero bits the location counter must have after
5716 advancement. For example @samp{.p2align 3} advances the location
5717 counter until it a multiple of 8. If the location counter is already a
5718 multiple of 8, no change is needed.
5720 The second expression (also absolute) gives the fill value to be stored in the
5721 padding bytes. It (and the comma) may be omitted. If it is omitted, the
5722 padding bytes are normally zero. However, on some systems, if the section is
5723 marked as containing code and the fill value is omitted, the space is filled
5724 with no-op instructions.
5726 The third expression is also absolute, and is also optional. If it is present,
5727 it is the maximum number of bytes that should be skipped by this alignment
5728 directive. If doing the alignment would require skipping more bytes than the
5729 specified maximum, then the alignment is not done at all. You can omit the
5730 fill value (the second argument) entirely by simply using two commas after the
5731 required alignment; this can be useful if you want the alignment to be filled
5732 with no-op instructions when appropriate.
5734 @cindex @code{p2alignw} directive
5735 @cindex @code{p2alignl} directive
5736 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
5737 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
5738 pattern as a two byte word value. The @code{.p2alignl} directives treats the
5739 fill pattern as a four byte longword value. For example, @code{.p2alignw
5740 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
5741 filled in with the value 0x368d (the exact placement of the bytes depends upon
5742 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
5747 @section @code{.popsection}
5749 @cindex @code{popsection} directive
5750 @cindex Section Stack
5751 This is one of the ELF section stack manipulation directives. The others are
5752 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5753 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
5756 This directive replaces the current section (and subsection) with the top
5757 section (and subsection) on the section stack. This section is popped off the
5763 @section @code{.previous}
5765 @cindex @code{previous} directive
5766 @cindex Section Stack
5767 This is one of the ELF section stack manipulation directives. The others are
5768 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5769 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
5770 (@pxref{PopSection}).
5772 This directive swaps the current section (and subsection) with most recently
5773 referenced section/subsection pair prior to this one. Multiple
5774 @code{.previous} directives in a row will flip between two sections (and their
5775 subsections). For example:
5787 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
5793 # Now in section A subsection 1
5797 # Now in section B subsection 0
5800 # Now in section B subsection 1
5803 # Now in section B subsection 0
5807 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
5808 section B and 0x9abc into subsection 1 of section B.
5810 In terms of the section stack, this directive swaps the current section with
5811 the top section on the section stack.
5815 @section @code{.print @var{string}}
5817 @cindex @code{print} directive
5818 @command{@value{AS}} will print @var{string} on the standard output during
5819 assembly. You must put @var{string} in double quotes.
5823 @section @code{.protected @var{names}}
5825 @cindex @code{protected} directive
5827 This is one of the ELF visibility directives. The other two are
5828 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
5830 This directive overrides the named symbols default visibility (which is set by
5831 their binding: local, global or weak). The directive sets the visibility to
5832 @code{protected} which means that any references to the symbols from within the
5833 components that defines them must be resolved to the definition in that
5834 component, even if a definition in another component would normally preempt
5839 @section @code{.psize @var{lines} , @var{columns}}
5841 @cindex @code{psize} directive
5842 @cindex listing control: paper size
5843 @cindex paper size, for listings
5844 Use this directive to declare the number of lines---and, optionally, the
5845 number of columns---to use for each page, when generating listings.
5847 If you do not use @code{.psize}, listings use a default line-count
5848 of 60. You may omit the comma and @var{columns} specification; the
5849 default width is 200 columns.
5851 @command{@value{AS}} generates formfeeds whenever the specified number of
5852 lines is exceeded (or whenever you explicitly request one, using
5855 If you specify @var{lines} as @code{0}, no formfeeds are generated save
5856 those explicitly specified with @code{.eject}.
5859 @section @code{.purgem @var{name}}
5861 @cindex @code{purgem} directive
5862 Undefine the macro @var{name}, so that later uses of the string will not be
5863 expanded. @xref{Macro}.
5867 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
5869 @cindex @code{pushsection} directive
5870 @cindex Section Stack
5871 This is one of the ELF section stack manipulation directives. The others are
5872 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
5873 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
5876 This directive pushes the current section (and subsection) onto the
5877 top of the section stack, and then replaces the current section and
5878 subsection with @code{name} and @code{subsection}. The optional
5879 @code{flags}, @code{type} and @code{arguments} are treated the same
5880 as in the @code{.section} (@pxref{Section}) directive.
5884 @section @code{.quad @var{bignums}}
5886 @cindex @code{quad} directive
5887 @code{.quad} expects zero or more bignums, separated by commas. For
5888 each bignum, it emits
5890 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
5891 warning message; and just takes the lowest order 8 bytes of the bignum.
5892 @cindex eight-byte integer
5893 @cindex integer, 8-byte
5895 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
5896 hence @emph{quad}-word for 8 bytes.
5899 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
5900 warning message; and just takes the lowest order 16 bytes of the bignum.
5901 @cindex sixteen-byte integer
5902 @cindex integer, 16-byte
5906 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
5908 @cindex @code{reloc} directive
5909 Generate a relocation at @var{offset} of type @var{reloc_name} with value
5910 @var{expression}. If @var{offset} is a number, the relocation is generated in
5911 the current section. If @var{offset} is an expression that resolves to a
5912 symbol plus offset, the relocation is generated in the given symbol's section.
5913 @var{expression}, if present, must resolve to a symbol plus addend or to an
5914 absolute value, but note that not all targets support an addend. e.g. ELF REL
5915 targets such as i386 store an addend in the section contents rather than in the
5916 relocation. This low level interface does not support addends stored in the
5920 @section @code{.rept @var{count}}
5922 @cindex @code{rept} directive
5923 Repeat the sequence of lines between the @code{.rept} directive and the next
5924 @code{.endr} directive @var{count} times.
5926 For example, assembling
5934 is equivalent to assembling
5943 @section @code{.sbttl "@var{subheading}"}
5945 @cindex @code{sbttl} directive
5946 @cindex subtitles for listings
5947 @cindex listing control: subtitle
5948 Use @var{subheading} as the title (third line, immediately after the
5949 title line) when generating assembly listings.
5951 This directive affects subsequent pages, as well as the current page if
5952 it appears within ten lines of the top of a page.
5956 @section @code{.scl @var{class}}
5958 @cindex @code{scl} directive
5959 @cindex symbol storage class (COFF)
5960 @cindex COFF symbol storage class
5961 Set the storage-class value for a symbol. This directive may only be
5962 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
5963 whether a symbol is static or external, or it may record further
5964 symbolic debugging information.
5967 The @samp{.scl} directive is primarily associated with COFF output; when
5968 configured to generate @code{b.out} output format, @command{@value{AS}}
5969 accepts this directive but ignores it.
5975 @section @code{.section @var{name}}
5977 @cindex named section
5978 Use the @code{.section} directive to assemble the following code into a section
5981 This directive is only supported for targets that actually support arbitrarily
5982 named sections; on @code{a.out} targets, for example, it is not accepted, even
5983 with a standard @code{a.out} section name.
5987 @c only print the extra heading if both COFF and ELF are set
5988 @subheading COFF Version
5991 @cindex @code{section} directive (COFF version)
5992 For COFF targets, the @code{.section} directive is used in one of the following
5996 .section @var{name}[, "@var{flags}"]
5997 .section @var{name}[, @var{subsection}]
6000 If the optional argument is quoted, it is taken as flags to use for the
6001 section. Each flag is a single character. The following flags are recognized:
6004 bss section (uninitialized data)
6006 section is not loaded
6012 exclude section from linking
6018 shared section (meaningful for PE targets)
6020 ignored. (For compatibility with the ELF version)
6022 section is not readable (meaningful for PE targets)
6024 single-digit power-of-two section alignment (GNU extension)
6027 If no flags are specified, the default flags depend upon the section name. If
6028 the section name is not recognized, the default will be for the section to be
6029 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6030 from the section, rather than adding them, so if they are used on their own it
6031 will be as if no flags had been specified at all.
6033 If the optional argument to the @code{.section} directive is not quoted, it is
6034 taken as a subsection number (@pxref{Sub-Sections}).
6039 @c only print the extra heading if both COFF and ELF are set
6040 @subheading ELF Version
6043 @cindex Section Stack
6044 This is one of the ELF section stack manipulation directives. The others are
6045 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6046 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6047 @code{.previous} (@pxref{Previous}).
6049 @cindex @code{section} directive (ELF version)
6050 For ELF targets, the @code{.section} directive is used like this:
6053 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6056 The optional @var{flags} argument is a quoted string which may contain any
6057 combination of the following characters:
6060 section is allocatable
6062 section is excluded from executable and shared library.
6066 section is executable
6068 section is mergeable
6070 section contains zero terminated strings
6072 section is a member of a section group
6074 section is used for thread-local-storage
6076 section is a member of the previously-current section's group, if any
6079 The optional @var{type} argument may contain one of the following constants:
6082 section contains data
6084 section does not contain data (i.e., section only occupies space)
6086 section contains data which is used by things other than the program
6088 section contains an array of pointers to init functions
6090 section contains an array of pointers to finish functions
6091 @item @@preinit_array
6092 section contains an array of pointers to pre-init functions
6095 Many targets only support the first three section types.
6097 Note on targets where the @code{@@} character is the start of a comment (eg
6098 ARM) then another character is used instead. For example the ARM port uses the
6101 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6102 be specified as well as an extra argument---@var{entsize}---like this:
6105 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6108 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6109 constants, each @var{entsize} octets long. Sections with both @code{M} and
6110 @code{S} must contain zero terminated strings where each character is
6111 @var{entsize} bytes long. The linker may remove duplicates within sections with
6112 the same name, same entity size and same flags. @var{entsize} must be an
6113 absolute expression. For sections with both @code{M} and @code{S}, a string
6114 which is a suffix of a larger string is considered a duplicate. Thus
6115 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6116 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6118 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6119 be present along with an additional field like this:
6122 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6125 The @var{GroupName} field specifies the name of the section group to which this
6126 particular section belongs. The optional linkage field can contain:
6129 indicates that only one copy of this section should be retained
6134 Note: if both the @var{M} and @var{G} flags are present then the fields for
6135 the Merge flag should come first, like this:
6138 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6141 If @var{flags} contains the @code{?} symbol then it may not also contain the
6142 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6143 present. Instead, @code{?} says to consider the section that's current before
6144 this directive. If that section used @code{G}, then the new section will use
6145 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6146 If not, then the @code{?} symbol has no effect.
6148 If no flags are specified, the default flags depend upon the section name. If
6149 the section name is not recognized, the default will be for the section to have
6150 none of the above flags: it will not be allocated in memory, nor writable, nor
6151 executable. The section will contain data.
6153 For ELF targets, the assembler supports another type of @code{.section}
6154 directive for compatibility with the Solaris assembler:
6157 .section "@var{name}"[, @var{flags}...]
6160 Note that the section name is quoted. There may be a sequence of comma
6164 section is allocatable
6168 section is executable
6170 section is excluded from executable and shared library.
6172 section is used for thread local storage
6175 This directive replaces the current section and subsection. See the
6176 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6177 some examples of how this directive and the other section stack directives
6183 @section @code{.set @var{symbol}, @var{expression}}
6185 @cindex @code{set} directive
6186 @cindex symbol value, setting
6187 Set the value of @var{symbol} to @var{expression}. This
6188 changes @var{symbol}'s value and type to conform to
6189 @var{expression}. If @var{symbol} was flagged as external, it remains
6190 flagged (@pxref{Symbol Attributes}).
6192 You may @code{.set} a symbol many times in the same assembly.
6194 If you @code{.set} a global symbol, the value stored in the object
6195 file is the last value stored into it.
6198 On Z80 @code{set} is a real instruction, use
6199 @samp{@var{symbol} defl @var{expression}} instead.
6203 @section @code{.short @var{expressions}}
6205 @cindex @code{short} directive
6207 @code{.short} is normally the same as @samp{.word}.
6208 @xref{Word,,@code{.word}}.
6210 In some configurations, however, @code{.short} and @code{.word} generate
6211 numbers of different lengths. @xref{Machine Dependencies}.
6215 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6218 This expects zero or more @var{expressions}, and emits
6219 a 16 bit number for each.
6224 @section @code{.single @var{flonums}}
6226 @cindex @code{single} directive
6227 @cindex floating point numbers (single)
6228 This directive assembles zero or more flonums, separated by commas. It
6229 has the same effect as @code{.float}.
6231 The exact kind of floating point numbers emitted depends on how
6232 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6236 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6237 numbers in @sc{ieee} format.
6243 @section @code{.size}
6245 This directive is used to set the size associated with a symbol.
6249 @c only print the extra heading if both COFF and ELF are set
6250 @subheading COFF Version
6253 @cindex @code{size} directive (COFF version)
6254 For COFF targets, the @code{.size} directive is only permitted inside
6255 @code{.def}/@code{.endef} pairs. It is used like this:
6258 .size @var{expression}
6262 @samp{.size} is only meaningful when generating COFF format output; when
6263 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6270 @c only print the extra heading if both COFF and ELF are set
6271 @subheading ELF Version
6274 @cindex @code{size} directive (ELF version)
6275 For ELF targets, the @code{.size} directive is used like this:
6278 .size @var{name} , @var{expression}
6281 This directive sets the size associated with a symbol @var{name}.
6282 The size in bytes is computed from @var{expression} which can make use of label
6283 arithmetic. This directive is typically used to set the size of function
6288 @ifclear no-space-dir
6290 @section @code{.skip @var{size} , @var{fill}}
6292 @cindex @code{skip} directive
6293 @cindex filling memory
6294 This directive emits @var{size} bytes, each of value @var{fill}. Both
6295 @var{size} and @var{fill} are absolute expressions. If the comma and
6296 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6301 @section @code{.sleb128 @var{expressions}}
6303 @cindex @code{sleb128} directive
6304 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6305 compact, variable length representation of numbers used by the DWARF
6306 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6308 @ifclear no-space-dir
6310 @section @code{.space @var{size} , @var{fill}}
6312 @cindex @code{space} directive
6313 @cindex filling memory
6314 This directive emits @var{size} bytes, each of value @var{fill}. Both
6315 @var{size} and @var{fill} are absolute expressions. If the comma
6316 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6321 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6322 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6323 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6324 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6332 @section @code{.stabd, .stabn, .stabs}
6334 @cindex symbolic debuggers, information for
6335 @cindex @code{stab@var{x}} directives
6336 There are three directives that begin @samp{.stab}.
6337 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6338 The symbols are not entered in the @command{@value{AS}} hash table: they
6339 cannot be referenced elsewhere in the source file.
6340 Up to five fields are required:
6344 This is the symbol's name. It may contain any character except
6345 @samp{\000}, so is more general than ordinary symbol names. Some
6346 debuggers used to code arbitrarily complex structures into symbol names
6350 An absolute expression. The symbol's type is set to the low 8 bits of
6351 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6352 and debuggers choke on silly bit patterns.
6355 An absolute expression. The symbol's ``other'' attribute is set to the
6356 low 8 bits of this expression.
6359 An absolute expression. The symbol's descriptor is set to the low 16
6360 bits of this expression.
6363 An absolute expression which becomes the symbol's value.
6366 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6367 or @code{.stabs} statement, the symbol has probably already been created;
6368 you get a half-formed symbol in your object file. This is
6369 compatible with earlier assemblers!
6372 @cindex @code{stabd} directive
6373 @item .stabd @var{type} , @var{other} , @var{desc}
6375 The ``name'' of the symbol generated is not even an empty string.
6376 It is a null pointer, for compatibility. Older assemblers used a
6377 null pointer so they didn't waste space in object files with empty
6380 The symbol's value is set to the location counter,
6381 relocatably. When your program is linked, the value of this symbol
6382 is the address of the location counter when the @code{.stabd} was
6385 @cindex @code{stabn} directive
6386 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6387 The name of the symbol is set to the empty string @code{""}.
6389 @cindex @code{stabs} directive
6390 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6391 All five fields are specified.
6397 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6398 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6400 @cindex string, copying to object file
6401 @cindex string8, copying to object file
6402 @cindex string16, copying to object file
6403 @cindex string32, copying to object file
6404 @cindex string64, copying to object file
6405 @cindex @code{string} directive
6406 @cindex @code{string8} directive
6407 @cindex @code{string16} directive
6408 @cindex @code{string32} directive
6409 @cindex @code{string64} directive
6411 Copy the characters in @var{str} to the object file. You may specify more than
6412 one string to copy, separated by commas. Unless otherwise specified for a
6413 particular machine, the assembler marks the end of each string with a 0 byte.
6414 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6416 The variants @code{string16}, @code{string32} and @code{string64} differ from
6417 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6418 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6419 are stored in target endianness byte order.
6425 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6426 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6431 @section @code{.struct @var{expression}}
6433 @cindex @code{struct} directive
6434 Switch to the absolute section, and set the section offset to @var{expression},
6435 which must be an absolute expression. You might use this as follows:
6444 This would define the symbol @code{field1} to have the value 0, the symbol
6445 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6446 value 8. Assembly would be left in the absolute section, and you would need to
6447 use a @code{.section} directive of some sort to change to some other section
6448 before further assembly.
6452 @section @code{.subsection @var{name}}
6454 @cindex @code{subsection} directive
6455 @cindex Section Stack
6456 This is one of the ELF section stack manipulation directives. The others are
6457 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6458 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6461 This directive replaces the current subsection with @code{name}. The current
6462 section is not changed. The replaced subsection is put onto the section stack
6463 in place of the then current top of stack subsection.
6468 @section @code{.symver}
6469 @cindex @code{symver} directive
6470 @cindex symbol versioning
6471 @cindex versions of symbols
6472 Use the @code{.symver} directive to bind symbols to specific version nodes
6473 within a source file. This is only supported on ELF platforms, and is
6474 typically used when assembling files to be linked into a shared library.
6475 There are cases where it may make sense to use this in objects to be bound
6476 into an application itself so as to override a versioned symbol from a
6479 For ELF targets, the @code{.symver} directive can be used like this:
6481 .symver @var{name}, @var{name2@@nodename}
6483 If the symbol @var{name} is defined within the file
6484 being assembled, the @code{.symver} directive effectively creates a symbol
6485 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6486 just don't try and create a regular alias is that the @var{@@} character isn't
6487 permitted in symbol names. The @var{name2} part of the name is the actual name
6488 of the symbol by which it will be externally referenced. The name @var{name}
6489 itself is merely a name of convenience that is used so that it is possible to
6490 have definitions for multiple versions of a function within a single source
6491 file, and so that the compiler can unambiguously know which version of a
6492 function is being mentioned. The @var{nodename} portion of the alias should be
6493 the name of a node specified in the version script supplied to the linker when
6494 building a shared library. If you are attempting to override a versioned
6495 symbol from a shared library, then @var{nodename} should correspond to the
6496 nodename of the symbol you are trying to override.
6498 If the symbol @var{name} is not defined within the file being assembled, all
6499 references to @var{name} will be changed to @var{name2@@nodename}. If no
6500 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6503 Another usage of the @code{.symver} directive is:
6505 .symver @var{name}, @var{name2@@@@nodename}
6507 In this case, the symbol @var{name} must exist and be defined within
6508 the file being assembled. It is similar to @var{name2@@nodename}. The
6509 difference is @var{name2@@@@nodename} will also be used to resolve
6510 references to @var{name2} by the linker.
6512 The third usage of the @code{.symver} directive is:
6514 .symver @var{name}, @var{name2@@@@@@nodename}
6516 When @var{name} is not defined within the
6517 file being assembled, it is treated as @var{name2@@nodename}. When
6518 @var{name} is defined within the file being assembled, the symbol
6519 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6524 @section @code{.tag @var{structname}}
6526 @cindex COFF structure debugging
6527 @cindex structure debugging, COFF
6528 @cindex @code{tag} directive
6529 This directive is generated by compilers to include auxiliary debugging
6530 information in the symbol table. It is only permitted inside
6531 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6532 definitions in the symbol table with instances of those structures.
6535 @samp{.tag} is only used when generating COFF format output; when
6536 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6542 @section @code{.text @var{subsection}}
6544 @cindex @code{text} directive
6545 Tells @command{@value{AS}} to assemble the following statements onto the end of
6546 the text subsection numbered @var{subsection}, which is an absolute
6547 expression. If @var{subsection} is omitted, subsection number zero
6551 @section @code{.title "@var{heading}"}
6553 @cindex @code{title} directive
6554 @cindex listing control: title line
6555 Use @var{heading} as the title (second line, immediately after the
6556 source file name and pagenumber) when generating assembly listings.
6558 This directive affects subsequent pages, as well as the current page if
6559 it appears within ten lines of the top of a page.
6563 @section @code{.type}
6565 This directive is used to set the type of a symbol.
6569 @c only print the extra heading if both COFF and ELF are set
6570 @subheading COFF Version
6573 @cindex COFF symbol type
6574 @cindex symbol type, COFF
6575 @cindex @code{type} directive (COFF version)
6576 For COFF targets, this directive is permitted only within
6577 @code{.def}/@code{.endef} pairs. It is used like this:
6583 This records the integer @var{int} as the type attribute of a symbol table
6587 @samp{.type} is associated only with COFF format output; when
6588 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
6589 directive but ignores it.
6595 @c only print the extra heading if both COFF and ELF are set
6596 @subheading ELF Version
6599 @cindex ELF symbol type
6600 @cindex symbol type, ELF
6601 @cindex @code{type} directive (ELF version)
6602 For ELF targets, the @code{.type} directive is used like this:
6605 .type @var{name} , @var{type description}
6608 This sets the type of symbol @var{name} to be either a
6609 function symbol or an object symbol. There are five different syntaxes
6610 supported for the @var{type description} field, in order to provide
6611 compatibility with various other assemblers.
6613 Because some of the characters used in these syntaxes (such as @samp{@@} and
6614 @samp{#}) are comment characters for some architectures, some of the syntaxes
6615 below do not work on all architectures. The first variant will be accepted by
6616 the GNU assembler on all architectures so that variant should be used for
6617 maximum portability, if you do not need to assemble your code with other
6620 The syntaxes supported are:
6623 .type <name> STT_<TYPE_IN_UPPER_CASE>
6624 .type <name>,#<type>
6625 .type <name>,@@<type>
6626 .type <name>,%<type>
6627 .type <name>,"<type>"
6630 The types supported are:
6635 Mark the symbol as being a function name.
6638 @itemx gnu_indirect_function
6639 Mark the symbol as an indirect function when evaluated during reloc
6640 processing. (This is only supported on assemblers targeting GNU systems).
6644 Mark the symbol as being a data object.
6648 Mark the symbol as being a thead-local data object.
6652 Mark the symbol as being a common data object.
6656 Does not mark the symbol in any way. It is supported just for completeness.
6658 @item gnu_unique_object
6659 Marks the symbol as being a globally unique data object. The dynamic linker
6660 will make sure that in the entire process there is just one symbol with this
6661 name and type in use. (This is only supported on assemblers targeting GNU
6666 Note: Some targets support extra types in addition to those listed above.
6672 @section @code{.uleb128 @var{expressions}}
6674 @cindex @code{uleb128} directive
6675 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
6676 compact, variable length representation of numbers used by the DWARF
6677 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
6681 @section @code{.val @var{addr}}
6683 @cindex @code{val} directive
6684 @cindex COFF value attribute
6685 @cindex value attribute, COFF
6686 This directive, permitted only within @code{.def}/@code{.endef} pairs,
6687 records the address @var{addr} as the value attribute of a symbol table
6691 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
6692 configured for @code{b.out}, it accepts this directive but ignores it.
6698 @section @code{.version "@var{string}"}
6700 @cindex @code{version} directive
6701 This directive creates a @code{.note} section and places into it an ELF
6702 formatted note of type NT_VERSION. The note's name is set to @code{string}.
6707 @section @code{.vtable_entry @var{table}, @var{offset}}
6709 @cindex @code{vtable_entry} directive
6710 This directive finds or creates a symbol @code{table} and creates a
6711 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
6714 @section @code{.vtable_inherit @var{child}, @var{parent}}
6716 @cindex @code{vtable_inherit} directive
6717 This directive finds the symbol @code{child} and finds or creates the symbol
6718 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
6719 parent whose addend is the value of the child symbol. As a special case the
6720 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
6724 @section @code{.warning "@var{string}"}
6725 @cindex warning directive
6726 Similar to the directive @code{.error}
6727 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
6730 @section @code{.weak @var{names}}
6732 @cindex @code{weak} directive
6733 This directive sets the weak attribute on the comma separated list of symbol
6734 @code{names}. If the symbols do not already exist, they will be created.
6736 On COFF targets other than PE, weak symbols are a GNU extension. This
6737 directive sets the weak attribute on the comma separated list of symbol
6738 @code{names}. If the symbols do not already exist, they will be created.
6740 On the PE target, weak symbols are supported natively as weak aliases.
6741 When a weak symbol is created that is not an alias, GAS creates an
6742 alternate symbol to hold the default value.
6745 @section @code{.weakref @var{alias}, @var{target}}
6747 @cindex @code{weakref} directive
6748 This directive creates an alias to the target symbol that enables the symbol to
6749 be referenced with weak-symbol semantics, but without actually making it weak.
6750 If direct references or definitions of the symbol are present, then the symbol
6751 will not be weak, but if all references to it are through weak references, the
6752 symbol will be marked as weak in the symbol table.
6754 The effect is equivalent to moving all references to the alias to a separate
6755 assembly source file, renaming the alias to the symbol in it, declaring the
6756 symbol as weak there, and running a reloadable link to merge the object files
6757 resulting from the assembly of the new source file and the old source file that
6758 had the references to the alias removed.
6760 The alias itself never makes to the symbol table, and is entirely handled
6761 within the assembler.
6764 @section @code{.word @var{expressions}}
6766 @cindex @code{word} directive
6767 This directive expects zero or more @var{expressions}, of any section,
6768 separated by commas.
6771 For each expression, @command{@value{AS}} emits a 32-bit number.
6774 For each expression, @command{@value{AS}} emits a 16-bit number.
6779 The size of the number emitted, and its byte order,
6780 depend on what target computer the assembly is for.
6783 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
6784 @c happen---32-bit addressability, period; no long/short jumps.
6785 @ifset DIFF-TBL-KLUGE
6786 @cindex difference tables altered
6787 @cindex altered difference tables
6789 @emph{Warning: Special Treatment to support Compilers}
6793 Machines with a 32-bit address space, but that do less than 32-bit
6794 addressing, require the following special treatment. If the machine of
6795 interest to you does 32-bit addressing (or doesn't require it;
6796 @pxref{Machine Dependencies}), you can ignore this issue.
6799 In order to assemble compiler output into something that works,
6800 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
6801 Directives of the form @samp{.word sym1-sym2} are often emitted by
6802 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
6803 directive of the form @samp{.word sym1-sym2}, and the difference between
6804 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
6805 creates a @dfn{secondary jump table}, immediately before the next label.
6806 This secondary jump table is preceded by a short-jump to the
6807 first byte after the secondary table. This short-jump prevents the flow
6808 of control from accidentally falling into the new table. Inside the
6809 table is a long-jump to @code{sym2}. The original @samp{.word}
6810 contains @code{sym1} minus the address of the long-jump to
6813 If there were several occurrences of @samp{.word sym1-sym2} before the
6814 secondary jump table, all of them are adjusted. If there was a
6815 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
6816 long-jump to @code{sym4} is included in the secondary jump table,
6817 and the @code{.word} directives are adjusted to contain @code{sym3}
6818 minus the address of the long-jump to @code{sym4}; and so on, for as many
6819 entries in the original jump table as necessary.
6822 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
6823 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
6824 assembly language programmers.
6827 @c end DIFF-TBL-KLUGE
6830 @section Deprecated Directives
6832 @cindex deprecated directives
6833 @cindex obsolescent directives
6834 One day these directives won't work.
6835 They are included for compatibility with older assemblers.
6842 @node Object Attributes
6843 @chapter Object Attributes
6844 @cindex object attributes
6846 @command{@value{AS}} assembles source files written for a specific architecture
6847 into object files for that architecture. But not all object files are alike.
6848 Many architectures support incompatible variations. For instance, floating
6849 point arguments might be passed in floating point registers if the object file
6850 requires hardware floating point support---or floating point arguments might be
6851 passed in integer registers if the object file supports processors with no
6852 hardware floating point unit. Or, if two objects are built for different
6853 generations of the same architecture, the combination may require the
6854 newer generation at run-time.
6856 This information is useful during and after linking. At link time,
6857 @command{@value{LD}} can warn about incompatible object files. After link
6858 time, tools like @command{gdb} can use it to process the linked file
6861 Compatibility information is recorded as a series of object attributes. Each
6862 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
6863 string, and indicates who sets the meaning of the tag. The tag is an integer,
6864 and indicates what property the attribute describes. The value may be a string
6865 or an integer, and indicates how the property affects this object. Missing
6866 attributes are the same as attributes with a zero value or empty string value.
6868 Object attributes were developed as part of the ABI for the ARM Architecture.
6869 The file format is documented in @cite{ELF for the ARM Architecture}.
6872 * GNU Object Attributes:: @sc{gnu} Object Attributes
6873 * Defining New Object Attributes:: Defining New Object Attributes
6876 @node GNU Object Attributes
6877 @section @sc{gnu} Object Attributes
6879 The @code{.gnu_attribute} directive records an object attribute
6880 with vendor @samp{gnu}.
6882 Except for @samp{Tag_compatibility}, which has both an integer and a string for
6883 its value, @sc{gnu} attributes have a string value if the tag number is odd and
6884 an integer value if the tag number is even. The second bit (@code{@var{tag} &
6885 2} is set for architecture-independent attributes and clear for
6886 architecture-dependent ones.
6888 @subsection Common @sc{gnu} attributes
6890 These attributes are valid on all architectures.
6893 @item Tag_compatibility (32)
6894 The compatibility attribute takes an integer flag value and a vendor name. If
6895 the flag value is 0, the file is compatible with other toolchains. If it is 1,
6896 then the file is only compatible with the named toolchain. If it is greater
6897 than 1, the file can only be processed by other toolchains under some private
6898 arrangement indicated by the flag value and the vendor name.
6901 @subsection MIPS Attributes
6904 @item Tag_GNU_MIPS_ABI_FP (4)
6905 The floating-point ABI used by this object file. The value will be:
6909 0 for files not affected by the floating-point ABI.
6911 1 for files using the hardware floating-point with a standard double-precision
6914 2 for files using the hardware floating-point ABI with a single-precision FPU.
6916 3 for files using the software floating-point ABI.
6918 4 for files using the hardware floating-point ABI with 64-bit wide
6919 double-precision floating-point registers and 32-bit wide general
6924 @subsection PowerPC Attributes
6927 @item Tag_GNU_Power_ABI_FP (4)
6928 The floating-point ABI used by this object file. The value will be:
6932 0 for files not affected by the floating-point ABI.
6934 1 for files using double-precision hardware floating-point ABI.
6936 2 for files using the software floating-point ABI.
6938 3 for files using single-precision hardware floating-point ABI.
6941 @item Tag_GNU_Power_ABI_Vector (8)
6942 The vector ABI used by this object file. The value will be:
6946 0 for files not affected by the vector ABI.
6948 1 for files using general purpose registers to pass vectors.
6950 2 for files using AltiVec registers to pass vectors.
6952 3 for files using SPE registers to pass vectors.
6956 @node Defining New Object Attributes
6957 @section Defining New Object Attributes
6959 If you want to define a new @sc{gnu} object attribute, here are the places you
6960 will need to modify. New attributes should be discussed on the @samp{binutils}
6965 This manual, which is the official register of attributes.
6967 The header for your architecture @file{include/elf}, to define the tag.
6969 The @file{bfd} support file for your architecture, to merge the attribute
6970 and issue any appropriate link warnings.
6972 Test cases in @file{ld/testsuite} for merging and link warnings.
6974 @file{binutils/readelf.c} to display your attribute.
6976 GCC, if you want the compiler to mark the attribute automatically.
6982 @node Machine Dependencies
6983 @chapter Machine Dependent Features
6985 @cindex machine dependencies
6986 The machine instruction sets are (almost by definition) different on
6987 each machine where @command{@value{AS}} runs. Floating point representations
6988 vary as well, and @command{@value{AS}} often supports a few additional
6989 directives or command-line options for compatibility with other
6990 assemblers on a particular platform. Finally, some versions of
6991 @command{@value{AS}} support special pseudo-instructions for branch
6994 This chapter discusses most of these differences, though it does not
6995 include details on any machine's instruction set. For details on that
6996 subject, see the hardware manufacturer's manual.
7000 * AArch64-Dependent:: AArch64 Dependent Features
7003 * Alpha-Dependent:: Alpha Dependent Features
7006 * ARC-Dependent:: ARC Dependent Features
7009 * ARM-Dependent:: ARM Dependent Features
7012 * AVR-Dependent:: AVR Dependent Features
7015 * Blackfin-Dependent:: Blackfin Dependent Features
7018 * CR16-Dependent:: CR16 Dependent Features
7021 * CRIS-Dependent:: CRIS Dependent Features
7024 * D10V-Dependent:: D10V Dependent Features
7027 * D30V-Dependent:: D30V Dependent Features
7030 * Epiphany-Dependent:: EPIPHANY Dependent Features
7033 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7036 * HPPA-Dependent:: HPPA Dependent Features
7039 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
7042 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7045 * i860-Dependent:: Intel 80860 Dependent Features
7048 * i960-Dependent:: Intel 80960 Dependent Features
7051 * IA-64-Dependent:: Intel IA-64 Dependent Features
7054 * IP2K-Dependent:: IP2K Dependent Features
7057 * LM32-Dependent:: LM32 Dependent Features
7060 * M32C-Dependent:: M32C Dependent Features
7063 * M32R-Dependent:: M32R Dependent Features
7066 * M68K-Dependent:: M680x0 Dependent Features
7069 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7072 * Meta-Dependent :: Meta Dependent Features
7075 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7078 * MIPS-Dependent:: MIPS Dependent Features
7081 * MMIX-Dependent:: MMIX Dependent Features
7084 * MSP430-Dependent:: MSP430 Dependent Features
7087 * NiosII-Dependent:: Altera Nios II Dependent Features
7090 * NS32K-Dependent:: NS32K Dependent Features
7093 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7094 * SH64-Dependent:: SuperH SH64 Dependent Features
7097 * PDP-11-Dependent:: PDP-11 Dependent Features
7100 * PJ-Dependent:: picoJava Dependent Features
7103 * PPC-Dependent:: PowerPC Dependent Features
7106 * RL78-Dependent:: RL78 Dependent Features
7109 * RX-Dependent:: RX Dependent Features
7112 * S/390-Dependent:: IBM S/390 Dependent Features
7115 * SCORE-Dependent:: SCORE Dependent Features
7118 * Sparc-Dependent:: SPARC Dependent Features
7121 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7124 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7127 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7130 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7133 * V850-Dependent:: V850 Dependent Features
7136 * XGATE-Dependent:: XGATE Features
7139 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7142 * Xtensa-Dependent:: Xtensa Dependent Features
7145 * Z80-Dependent:: Z80 Dependent Features
7148 * Z8000-Dependent:: Z8000 Dependent Features
7151 * Vax-Dependent:: VAX Dependent Features
7158 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7159 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7160 @c peculiarity: to preserve cross-references, there must be a node called
7161 @c "Machine Dependencies". Hence the conditional nodenames in each
7162 @c major node below. Node defaulting in makeinfo requires adjacency of
7163 @c node and sectioning commands; hence the repetition of @chapter BLAH
7164 @c in both conditional blocks.
7167 @include c-aarch64.texi
7171 @include c-alpha.texi
7187 @include c-bfin.texi
7191 @include c-cr16.texi
7195 @include c-cris.texi
7200 @node Machine Dependencies
7201 @chapter Machine Dependent Features
7203 The machine instruction sets are different on each Renesas chip family,
7204 and there are also some syntax differences among the families. This
7205 chapter describes the specific @command{@value{AS}} features for each
7209 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7210 * SH-Dependent:: Renesas SH Dependent Features
7217 @include c-d10v.texi
7221 @include c-d30v.texi
7225 @include c-epiphany.texi
7229 @include c-h8300.texi
7233 @include c-hppa.texi
7237 @include c-i370.texi
7241 @include c-i386.texi
7245 @include c-i860.texi
7249 @include c-i960.texi
7253 @include c-ia64.texi
7257 @include c-ip2k.texi
7261 @include c-lm32.texi
7265 @include c-m32c.texi
7269 @include c-m32r.texi
7273 @include c-m68k.texi
7277 @include c-m68hc11.texi
7281 @include c-metag.texi
7285 @include c-microblaze.texi
7289 @include c-mips.texi
7293 @include c-mmix.texi
7297 @include c-msp430.texi
7301 @include c-nios2.texi
7305 @include c-ns32k.texi
7309 @include c-pdp11.texi
7321 @include c-rl78.texi
7329 @include c-s390.texi
7333 @include c-score.texi
7338 @include c-sh64.texi
7342 @include c-sparc.texi
7346 @include c-tic54x.texi
7350 @include c-tic6x.texi
7354 @include c-tilegx.texi
7358 @include c-tilepro.texi
7374 @include c-v850.texi
7378 @include c-xgate.texi
7382 @include c-xstormy16.texi
7386 @include c-xtensa.texi
7390 @c reverse effect of @down at top of generic Machine-Dep chapter
7394 @node Reporting Bugs
7395 @chapter Reporting Bugs
7396 @cindex bugs in assembler
7397 @cindex reporting bugs in assembler
7399 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7401 Reporting a bug may help you by bringing a solution to your problem, or it may
7402 not. But in any case the principal function of a bug report is to help the
7403 entire community by making the next version of @command{@value{AS}} work better.
7404 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7406 In order for a bug report to serve its purpose, you must include the
7407 information that enables us to fix the bug.
7410 * Bug Criteria:: Have you found a bug?
7411 * Bug Reporting:: How to report bugs
7415 @section Have You Found a Bug?
7416 @cindex bug criteria
7418 If you are not sure whether you have found a bug, here are some guidelines:
7421 @cindex fatal signal
7422 @cindex assembler crash
7423 @cindex crash of assembler
7425 If the assembler gets a fatal signal, for any input whatever, that is a
7426 @command{@value{AS}} bug. Reliable assemblers never crash.
7428 @cindex error on valid input
7430 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7432 @cindex invalid input
7434 If @command{@value{AS}} does not produce an error message for invalid input, that
7435 is a bug. However, you should note that your idea of ``invalid input'' might
7436 be our idea of ``an extension'' or ``support for traditional practice''.
7439 If you are an experienced user of assemblers, your suggestions for improvement
7440 of @command{@value{AS}} are welcome in any case.
7444 @section How to Report Bugs
7446 @cindex assembler bugs, reporting
7448 A number of companies and individuals offer support for @sc{gnu} products. If
7449 you obtained @command{@value{AS}} from a support organization, we recommend you
7450 contact that organization first.
7452 You can find contact information for many support companies and
7453 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7457 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7461 The fundamental principle of reporting bugs usefully is this:
7462 @strong{report all the facts}. If you are not sure whether to state a
7463 fact or leave it out, state it!
7465 Often people omit facts because they think they know what causes the problem
7466 and assume that some details do not matter. Thus, you might assume that the
7467 name of a symbol you use in an example does not matter. Well, probably it does
7468 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7469 happens to fetch from the location where that name is stored in memory;
7470 perhaps, if the name were different, the contents of that location would fool
7471 the assembler into doing the right thing despite the bug. Play it safe and
7472 give a specific, complete example. That is the easiest thing for you to do,
7473 and the most helpful.
7475 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7476 it is new to us. Therefore, always write your bug reports on the assumption
7477 that the bug has not been reported previously.
7479 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7480 bell?'' This cannot help us fix a bug, so it is basically useless. We
7481 respond by asking for enough details to enable us to investigate.
7482 You might as well expedite matters by sending them to begin with.
7484 To enable us to fix the bug, you should include all these things:
7488 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7489 it with the @samp{--version} argument.
7491 Without this, we will not know whether there is any point in looking for
7492 the bug in the current version of @command{@value{AS}}.
7495 Any patches you may have applied to the @command{@value{AS}} source.
7498 The type of machine you are using, and the operating system name and
7502 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7506 The command arguments you gave the assembler to assemble your example and
7507 observe the bug. To guarantee you will not omit something important, list them
7508 all. A copy of the Makefile (or the output from make) is sufficient.
7510 If we were to try to guess the arguments, we would probably guess wrong
7511 and then we might not encounter the bug.
7514 A complete input file that will reproduce the bug. If the bug is observed when
7515 the assembler is invoked via a compiler, send the assembler source, not the
7516 high level language source. Most compilers will produce the assembler source
7517 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7518 the options @samp{-v --save-temps}; this will save the assembler source in a
7519 file with an extension of @file{.s}, and also show you exactly how
7520 @command{@value{AS}} is being run.
7523 A description of what behavior you observe that you believe is
7524 incorrect. For example, ``It gets a fatal signal.''
7526 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
7527 will certainly notice it. But if the bug is incorrect output, we might not
7528 notice unless it is glaringly wrong. You might as well not give us a chance to
7531 Even if the problem you experience is a fatal signal, you should still say so
7532 explicitly. Suppose something strange is going on, such as, your copy of
7533 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
7534 library on your system. (This has happened!) Your copy might crash and ours
7535 would not. If you told us to expect a crash, then when ours fails to crash, we
7536 would know that the bug was not happening for us. If you had not told us to
7537 expect a crash, then we would not be able to draw any conclusion from our
7541 If you wish to suggest changes to the @command{@value{AS}} source, send us context
7542 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
7543 option. Always send diffs from the old file to the new file. If you even
7544 discuss something in the @command{@value{AS}} source, refer to it by context, not
7547 The line numbers in our development sources will not match those in your
7548 sources. Your line numbers would convey no useful information to us.
7551 Here are some things that are not necessary:
7555 A description of the envelope of the bug.
7557 Often people who encounter a bug spend a lot of time investigating
7558 which changes to the input file will make the bug go away and which
7559 changes will not affect it.
7561 This is often time consuming and not very useful, because the way we
7562 will find the bug is by running a single example under the debugger
7563 with breakpoints, not by pure deduction from a series of examples.
7564 We recommend that you save your time for something else.
7566 Of course, if you can find a simpler example to report @emph{instead}
7567 of the original one, that is a convenience for us. Errors in the
7568 output will be easier to spot, running under the debugger will take
7569 less time, and so on.
7571 However, simplification is not vital; if you do not want to do this,
7572 report the bug anyway and send us the entire test case you used.
7575 A patch for the bug.
7577 A patch for the bug does help us if it is a good one. But do not omit
7578 the necessary information, such as the test case, on the assumption that
7579 a patch is all we need. We might see problems with your patch and decide
7580 to fix the problem another way, or we might not understand it at all.
7582 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
7583 construct an example that will make the program follow a certain path through
7584 the code. If you do not send us the example, we will not be able to construct
7585 one, so we will not be able to verify that the bug is fixed.
7587 And if we cannot understand what bug you are trying to fix, or why your
7588 patch should be an improvement, we will not install it. A test case will
7589 help us to understand.
7592 A guess about what the bug is or what it depends on.
7594 Such guesses are usually wrong. Even we cannot guess right about such
7595 things without first using the debugger to find the facts.
7598 @node Acknowledgements
7599 @chapter Acknowledgements
7601 If you have contributed to GAS and your name isn't listed here,
7602 it is not meant as a slight. We just don't know about it. Send mail to the
7603 maintainer, and we'll correct the situation. Currently
7605 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
7607 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
7610 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
7611 information and the 68k series machines, most of the preprocessing pass, and
7612 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
7614 K. Richard Pixley maintained GAS for a while, adding various enhancements and
7615 many bug fixes, including merging support for several processors, breaking GAS
7616 up to handle multiple object file format back ends (including heavy rewrite,
7617 testing, an integration of the coff and b.out back ends), adding configuration
7618 including heavy testing and verification of cross assemblers and file splits
7619 and renaming, converted GAS to strictly ANSI C including full prototypes, added
7620 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
7621 port (including considerable amounts of reverse engineering), a SPARC opcode
7622 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
7623 assertions and made them work, much other reorganization, cleanup, and lint.
7625 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
7626 in format-specific I/O modules.
7628 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
7629 has done much work with it since.
7631 The Intel 80386 machine description was written by Eliot Dresselhaus.
7633 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
7635 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
7636 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
7638 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
7639 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
7640 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
7641 support a.out format.
7643 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
7644 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
7645 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
7646 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
7649 John Gilmore built the AMD 29000 support, added @code{.include} support, and
7650 simplified the configuration of which versions accept which directives. He
7651 updated the 68k machine description so that Motorola's opcodes always produced
7652 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
7653 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
7654 cross-compilation support, and one bug in relaxation that took a week and
7655 required the proverbial one-bit fix.
7657 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
7658 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
7659 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
7660 PowerPC assembler, and made a few other minor patches.
7662 Steve Chamberlain made GAS able to generate listings.
7664 Hewlett-Packard contributed support for the HP9000/300.
7666 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
7667 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
7668 formats). This work was supported by both the Center for Software Science at
7669 the University of Utah and Cygnus Support.
7671 Support for ELF format files has been worked on by Mark Eichin of Cygnus
7672 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
7673 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
7674 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
7675 and some initial 64-bit support).
7677 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
7679 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
7680 support for openVMS/Alpha.
7682 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
7685 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
7686 Inc.@: added support for Xtensa processors.
7688 Several engineers at Cygnus Support have also provided many small bug fixes and
7689 configuration enhancements.
7691 Jon Beniston added support for the Lattice Mico32 architecture.
7693 Many others have contributed large or small bugfixes and enhancements. If
7694 you have contributed significant work and are not mentioned on this list, and
7695 want to be, let us know. Some of the history has been lost; we are not
7696 intentionally leaving anyone out.
7698 @node GNU Free Documentation License
7699 @appendix GNU Free Documentation License
7703 @unnumbered AS Index
projects / platform / upstream / binutils.git / blob