1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (C) 1991-2018 Free Software Foundation, Inc.
3 @c UPDATE!! On future updates--
4 @c (1) check for new machine-dep cmdline options in
5 @c md_parse_option definitions in config/tc-*.c
6 @c (2) for platform-specific directives, examine md_pseudo_op
8 @c (3) for object-format specific directives, examine obj_pseudo_op
10 @c (4) portable directives in potable[] in read.c
14 @macro gcctabopt{body}
17 @c defaults, config file may override:
22 @include asconfig.texi
27 @c common OR combinations of conditions
50 @set abnormal-separator
54 @settitle Using @value{AS}
57 @settitle Using @value{AS} (@value{TARGET})
59 @setchapternewpage odd
64 @c WARE! Some of the machine-dependent sections contain tables of machine
65 @c instructions. Except in multi-column format, these tables look silly.
66 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
67 @c the multi-col format is faked within @example sections.
69 @c Again unfortunately, the natural size that fits on a page, for these tables,
70 @c is different depending on whether or not smallbook is turned on.
71 @c This matters, because of order: text flow switches columns at each page
74 @c The format faked in this source works reasonably well for smallbook,
75 @c not well for the default large-page format. This manual expects that if you
76 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
77 @c tables in question. You can turn on one without the other at your
78 @c discretion, of course.
81 @c the insn tables look just as silly in info files regardless of smallbook,
82 @c might as well show 'em anyways.
86 @dircategory Software development
88 * As: (as). The GNU assembler.
89 * Gas: (as). The GNU assembler.
97 This file documents the GNU Assembler "@value{AS}".
99 @c man begin COPYRIGHT
100 Copyright @copyright{} 1991-2018 Free Software Foundation, Inc.
102 Permission is granted to copy, distribute and/or modify this document
103 under the terms of the GNU Free Documentation License, Version 1.3
104 or any later version published by the Free Software Foundation;
105 with no Invariant Sections, with no Front-Cover Texts, and with no
106 Back-Cover Texts. A copy of the license is included in the
107 section entitled ``GNU Free Documentation License''.
113 @title Using @value{AS}
114 @subtitle The @sc{gnu} Assembler
116 @subtitle for the @value{TARGET} family
118 @ifset VERSION_PACKAGE
120 @subtitle @value{VERSION_PACKAGE}
123 @subtitle Version @value{VERSION}
126 The Free Software Foundation Inc.@: thanks The Nice Computer
127 Company of Australia for loaning Dean Elsner to write the
128 first (Vax) version of @command{as} for Project @sc{gnu}.
129 The proprietors, management and staff of TNCCA thank FSF for
130 distracting the boss while they got some work
133 @author Dean Elsner, Jay Fenlason & friends
137 \hfill {\it Using {\tt @value{AS}}}\par
138 \hfill Edited by Cygnus Support\par
140 %"boxit" macro for figures:
141 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
142 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
143 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
144 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
145 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
148 @vskip 0pt plus 1filll
149 Copyright @copyright{} 1991-2018 Free Software Foundation, Inc.
151 Permission is granted to copy, distribute and/or modify this document
152 under the terms of the GNU Free Documentation License, Version 1.3
153 or any later version published by the Free Software Foundation;
154 with no Invariant Sections, with no Front-Cover Texts, and with no
155 Back-Cover Texts. A copy of the license is included in the
156 section entitled ``GNU Free Documentation License''.
163 @top Using @value{AS}
165 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
166 @ifset VERSION_PACKAGE
167 @value{VERSION_PACKAGE}
169 version @value{VERSION}.
171 This version of the file describes @command{@value{AS}} configured to generate
172 code for @value{TARGET} architectures.
175 This document is distributed under the terms of the GNU Free
176 Documentation License. A copy of the license is included in the
177 section entitled ``GNU Free Documentation License''.
180 * Overview:: Overview
181 * Invoking:: Command-Line Options
183 * Sections:: Sections and Relocation
185 * Expressions:: Expressions
186 * Pseudo Ops:: Assembler Directives
188 * Object Attributes:: Object Attributes
190 * Machine Dependencies:: Machine Dependent Features
191 * Reporting Bugs:: Reporting Bugs
192 * Acknowledgements:: Who Did What
193 * GNU Free Documentation License:: GNU Free Documentation License
194 * AS Index:: AS Index
201 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
203 This version of the manual describes @command{@value{AS}} configured to generate
204 code for @value{TARGET} architectures.
208 @cindex invocation summary
209 @cindex option summary
210 @cindex summary of options
211 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
212 see @ref{Invoking,,Command-Line Options}.
214 @c man title AS the portable GNU assembler.
218 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
222 @c We don't use deffn and friends for the following because they seem
223 @c to be limited to one line for the header.
225 @c man begin SYNOPSIS
226 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
227 [@b{--compress-debug-sections}] [@b{--nocompress-debug-sections}]
228 [@b{--debug-prefix-map} @var{old}=@var{new}]
229 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
230 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--gdwarf-sections}]
231 [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
232 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
233 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
234 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}]
235 [@b{--no-pad-sections}]
236 [@b{-o} @var{objfile}] [@b{-R}]
237 [@b{--hash-size}=@var{NUM}] [@b{--reduce-memory-overheads}]
239 [@b{-v}] [@b{-version}] [@b{--version}]
240 [@b{-W}] [@b{--warn}] [@b{--fatal-warnings}] [@b{-w}] [@b{-x}]
241 [@b{-Z}] [@b{@@@var{FILE}}]
242 [@b{--sectname-subst}] [@b{--size-check=[error|warning]}]
243 [@b{--elf-stt-common=[no|yes]}]
244 [@b{--generate-missing-build-notes=[no|yes]}]
245 [@b{--target-help}] [@var{target-options}]
246 [@b{--}|@var{files} @dots{}]
249 @c Target dependent options are listed below. Keep the list sorted.
250 @c Add an empty line for separation.
254 @emph{Target AArch64 options:}
256 [@b{-mabi}=@var{ABI}]
260 @emph{Target Alpha options:}
262 [@b{-mdebug} | @b{-no-mdebug}]
263 [@b{-replace} | @b{-noreplace}]
264 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
265 [@b{-F}] [@b{-32addr}]
269 @emph{Target ARC options:}
270 [@b{-mcpu=@var{cpu}}]
271 [@b{-mA6}|@b{-mARC600}|@b{-mARC601}|@b{-mA7}|@b{-mARC700}|@b{-mEM}|@b{-mHS}]
278 @emph{Target ARM options:}
279 @c Don't document the deprecated options
280 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
281 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
282 [@b{-mfpu}=@var{floating-point-format}]
283 [@b{-mfloat-abi}=@var{abi}]
284 [@b{-meabi}=@var{ver}]
287 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
288 @b{-mapcs-reentrant}]
289 [@b{-mthumb-interwork}] [@b{-k}]
293 @emph{Target Blackfin options:}
294 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
301 @emph{Target CRIS options:}
302 [@b{--underscore} | @b{--no-underscore}]
304 [@b{--emulation=criself} | @b{--emulation=crisaout}]
305 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
306 @c Deprecated -- deliberately not documented.
311 @emph{Target C-SKY options:}
312 [@b{-march=@var{arch}}] [@b{-mcpu=@var{cpu}}]
313 [@b{-EL}] [@b{-mlittle-endian}] [@b{-EB}] [@b{-mbig-endian}]
314 [@b{-fpic}] [@b{-pic}]
315 [@b{-mljump}] [@b{-mno-ljump}]
316 [@b{-force2bsr}] [@b{-mforce2bsr}] [@b{-no-force2bsr}] [@b{-mno-force2bsr}]
317 [@b{-jsri2bsr}] [@b{-mjsri2bsr}] [@b{-no-jsri2bsr }] [@b{-mno-jsri2bsr}]
318 [@b{-mnolrw }] [@b{-mno-lrw}]
319 [@b{-melrw}] [@b{-mno-elrw}]
320 [@b{-mlaf }] [@b{-mliterals-after-func}]
321 [@b{-mno-laf}] [@b{-mno-literals-after-func}]
322 [@b{-mlabr}] [@b{-mliterals-after-br}]
323 [@b{-mno-labr}] [@b{-mnoliterals-after-br}]
324 [@b{-mistack}] [@b{-mno-istack}]
325 [@b{-mhard-float}] [@b{-mmp}] [@b{-mcp}] [@b{-mcache}]
326 [@b{-msecurity}] [@b{-mtrust}]
327 [@b{-mdsp}] [@b{-medsp}] [@b{-mvdsp}]
331 @emph{Target D10V options:}
336 @emph{Target D30V options:}
337 [@b{-O}|@b{-n}|@b{-N}]
341 @emph{Target EPIPHANY options:}
342 [@b{-mepiphany}|@b{-mepiphany16}]
346 @emph{Target H8/300 options:}
350 @c HPPA has no machine-dependent assembler options (yet).
354 @emph{Target i386 options:}
355 [@b{--32}|@b{--x32}|@b{--64}] [@b{-n}]
356 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
360 @emph{Target IA-64 options:}
361 [@b{-mconstant-gp}|@b{-mauto-pic}]
362 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
364 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
365 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
366 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
367 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
371 @emph{Target IP2K options:}
372 [@b{-mip2022}|@b{-mip2022ext}]
376 @emph{Target M32C options:}
377 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
381 @emph{Target M32R options:}
382 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
387 @emph{Target M680X0 options:}
388 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
392 @emph{Target M68HC11 options:}
393 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}|@b{-mm9s12x}|@b{-mm9s12xg}]
394 [@b{-mshort}|@b{-mlong}]
395 [@b{-mshort-double}|@b{-mlong-double}]
396 [@b{--force-long-branches}] [@b{--short-branches}]
397 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
398 [@b{--print-opcodes}] [@b{--generate-example}]
402 @emph{Target MCORE options:}
403 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
404 [@b{-mcpu=[210|340]}]
408 @emph{Target Meta options:}
409 [@b{-mcpu=@var{cpu}}] [@b{-mfpu=@var{cpu}}] [@b{-mdsp=@var{cpu}}]
412 @emph{Target MICROBLAZE options:}
413 @c MicroBlaze has no machine-dependent assembler options.
417 @emph{Target MIPS options:}
418 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
419 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
420 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
421 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
422 [@b{-mfp64}] [@b{-mgp64}] [@b{-mfpxx}]
423 [@b{-modd-spreg}] [@b{-mno-odd-spreg}]
424 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
425 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
426 [@b{-mips32r3}] [@b{-mips32r5}] [@b{-mips32r6}] [@b{-mips64}] [@b{-mips64r2}]
427 [@b{-mips64r3}] [@b{-mips64r5}] [@b{-mips64r6}]
428 [@b{-construct-floats}] [@b{-no-construct-floats}]
429 [@b{-mignore-branch-isa}] [@b{-mno-ignore-branch-isa}]
430 [@b{-mnan=@var{encoding}}]
431 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
432 [@b{-mips16}] [@b{-no-mips16}]
433 [@b{-mmips16e2}] [@b{-mno-mips16e2}]
434 [@b{-mmicromips}] [@b{-mno-micromips}]
435 [@b{-msmartmips}] [@b{-mno-smartmips}]
436 [@b{-mips3d}] [@b{-no-mips3d}]
437 [@b{-mdmx}] [@b{-no-mdmx}]
438 [@b{-mdsp}] [@b{-mno-dsp}]
439 [@b{-mdspr2}] [@b{-mno-dspr2}]
440 [@b{-mdspr3}] [@b{-mno-dspr3}]
441 [@b{-mmsa}] [@b{-mno-msa}]
442 [@b{-mxpa}] [@b{-mno-xpa}]
443 [@b{-mmt}] [@b{-mno-mt}]
444 [@b{-mmcu}] [@b{-mno-mcu}]
445 [@b{-mcrc}] [@b{-mno-crc}]
446 [@b{-mginv}] [@b{-mno-ginv}]
447 [@b{-mloongson-mmi}] [@b{-mno-loongson-mmi}]
448 [@b{-mloongson-cam}] [@b{-mno-loongson-cam}]
449 [@b{-mloongson-ext}] [@b{-mno-loongson-ext}]
450 [@b{-mloongson-ext2}] [@b{-mno-loongson-ext2}]
451 [@b{-minsn32}] [@b{-mno-insn32}]
452 [@b{-mfix7000}] [@b{-mno-fix7000}]
453 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
454 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
455 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
456 [@b{-mdebug}] [@b{-no-mdebug}]
457 [@b{-mpdr}] [@b{-mno-pdr}]
461 @emph{Target MMIX options:}
462 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
463 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
464 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
465 [@b{--linker-allocated-gregs}]
469 @emph{Target Nios II options:}
470 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
475 @emph{Target NDS32 options:}
476 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
477 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
478 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
479 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
480 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
481 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
482 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
486 @c OpenRISC has no machine-dependent assembler options.
490 @emph{Target PDP11 options:}
491 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
492 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
493 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
497 @emph{Target picoJava options:}
502 @emph{Target PowerPC options:}
504 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
505 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mgekko}|
506 @b{-mbroadway}|@b{-mppc64}|@b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|
507 @b{-me6500}|@b{-mppc64bridge}|@b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|
508 @b{-mpower6}|@b{-mpwr6}|@b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-mpower9}|@b{-mpwr9}@b{-ma2}|
509 @b{-mcell}|@b{-mspe}|@b{-mspe2}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
510 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
511 [@b{-mregnames}|@b{-mno-regnames}]
512 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
513 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
514 [@b{-msolaris}|@b{-mno-solaris}]
515 [@b{-nops=@var{count}}]
519 @emph{Target PRU options:}
522 [@b{-mno-warn-regname-label}]
526 @emph{Target RISC-V options:}
527 [@b{-fpic}|@b{-fPIC}|@b{-fno-pic}]
528 [@b{-march}=@var{ISA}]
529 [@b{-mabi}=@var{ABI}]
533 @emph{Target RL78 options:}
535 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
539 @emph{Target RX options:}
540 [@b{-mlittle-endian}|@b{-mbig-endian}]
541 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
542 [@b{-muse-conventional-section-names}]
543 [@b{-msmall-data-limit}]
546 [@b{-mint-register=@var{number}}]
547 [@b{-mgcc-abi}|@b{-mrx-abi}]
551 @emph{Target s390 options:}
552 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
553 [@b{-mregnames}|@b{-mno-regnames}]
554 [@b{-mwarn-areg-zero}]
558 @emph{Target SCORE options:}
559 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
560 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
561 [@b{-march=score7}][@b{-march=score3}]
562 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
566 @emph{Target SPARC options:}
567 @c The order here is important. See c-sparc.texi.
568 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Aleon}|@b{-Asparclet}|@b{-Asparclite}
569 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av8plusb}|@b{-Av8plusc}|@b{-Av8plusd}
570 @b{-Av8plusv}|@b{-Av8plusm}|@b{-Av9}|@b{-Av9a}|@b{-Av9b}|@b{-Av9c}
571 @b{-Av9d}|@b{-Av9e}|@b{-Av9v}|@b{-Av9m}|@b{-Asparc}|@b{-Asparcvis}
572 @b{-Asparcvis2}|@b{-Asparcfmaf}|@b{-Asparcima}|@b{-Asparcvis3}
573 @b{-Asparcvisr}|@b{-Asparc5}]
574 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}]|@b{-xarch=v8plusb}|@b{-xarch=v8plusc}
575 @b{-xarch=v8plusd}|@b{-xarch=v8plusv}|@b{-xarch=v8plusm}|@b{-xarch=v9}
576 @b{-xarch=v9a}|@b{-xarch=v9b}|@b{-xarch=v9c}|@b{-xarch=v9d}|@b{-xarch=v9e}
577 @b{-xarch=v9v}|@b{-xarch=v9m}|@b{-xarch=sparc}|@b{-xarch=sparcvis}
578 @b{-xarch=sparcvis2}|@b{-xarch=sparcfmaf}|@b{-xarch=sparcima}
579 @b{-xarch=sparcvis3}|@b{-xarch=sparcvisr}|@b{-xarch=sparc5}
582 [@b{--enforce-aligned-data}][@b{--dcti-couples-detect}]
586 @emph{Target TIC54X options:}
587 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
588 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
592 @emph{Target TIC6X options:}
593 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
594 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
595 [@b{-mpic}|@b{-mno-pic}]
599 @emph{Target TILE-Gx options:}
600 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
603 @c TILEPro has no machine-dependent assembler options
607 @emph{Target Visium options:}
608 [@b{-mtune=@var{arch}}]
612 @emph{Target Xtensa options:}
613 [@b{--[no-]text-section-literals}] [@b{--[no-]auto-litpools}]
614 [@b{--[no-]absolute-literals}]
615 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
616 [@b{--[no-]transform}]
617 [@b{--rename-section} @var{oldname}=@var{newname}]
618 [@b{--[no-]trampolines}]
622 @emph{Target Z80 options:}
623 [@b{-z80}] [@b{-r800}]
624 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
625 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
626 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
627 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
628 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
629 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
633 @c Z8000 has no machine-dependent assembler options
642 @include at-file.texi
645 Turn on listings, in any of a variety of ways:
649 omit false conditionals
652 omit debugging directives
655 include general information, like @value{AS} version and options passed
658 include high-level source
664 include macro expansions
667 omit forms processing
673 set the name of the listing file
676 You may combine these options; for example, use @samp{-aln} for assembly
677 listing without forms processing. The @samp{=file} option, if used, must be
678 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
681 Begin in alternate macro mode.
683 @xref{Altmacro,,@code{.altmacro}}.
686 @item --compress-debug-sections
687 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
688 ELF ABI. The resulting object file may not be compatible with older
689 linkers and object file utilities. Note if compression would make a
690 given section @emph{larger} then it is not compressed.
693 @cindex @samp{--compress-debug-sections=} option
694 @item --compress-debug-sections=none
695 @itemx --compress-debug-sections=zlib
696 @itemx --compress-debug-sections=zlib-gnu
697 @itemx --compress-debug-sections=zlib-gabi
698 These options control how DWARF debug sections are compressed.
699 @option{--compress-debug-sections=none} is equivalent to
700 @option{--nocompress-debug-sections}.
701 @option{--compress-debug-sections=zlib} and
702 @option{--compress-debug-sections=zlib-gabi} are equivalent to
703 @option{--compress-debug-sections}.
704 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
705 sections using zlib. The debug sections are renamed to begin with
706 @samp{.zdebug}. Note if compression would make a given section
707 @emph{larger} then it is not compressed nor renamed.
711 @item --nocompress-debug-sections
712 Do not compress DWARF debug sections. This is usually the default for all
713 targets except the x86/x86_64, but a configure time option can be used to
717 Ignored. This option is accepted for script compatibility with calls to
720 @item --debug-prefix-map @var{old}=@var{new}
721 When assembling files in directory @file{@var{old}}, record debugging
722 information describing them as in @file{@var{new}} instead.
724 @item --defsym @var{sym}=@var{value}
725 Define the symbol @var{sym} to be @var{value} before assembling the input file.
726 @var{value} must be an integer constant. As in C, a leading @samp{0x}
727 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
728 value. The value of the symbol can be overridden inside a source file via the
729 use of a @code{.set} pseudo-op.
732 ``fast''---skip whitespace and comment preprocessing (assume source is
737 Generate debugging information for each assembler source line using whichever
738 debug format is preferred by the target. This currently means either STABS,
742 Generate stabs debugging information for each assembler line. This
743 may help debugging assembler code, if the debugger can handle it.
746 Generate stabs debugging information for each assembler line, with GNU
747 extensions that probably only gdb can handle, and that could make other
748 debuggers crash or refuse to read your program. This
749 may help debugging assembler code. Currently the only GNU extension is
750 the location of the current working directory at assembling time.
753 Generate DWARF2 debugging information for each assembler line. This
754 may help debugging assembler code, if the debugger can handle it. Note---this
755 option is only supported by some targets, not all of them.
757 @item --gdwarf-sections
758 Instead of creating a .debug_line section, create a series of
759 .debug_line.@var{foo} sections where @var{foo} is the name of the
760 corresponding code section. For example a code section called @var{.text.func}
761 will have its dwarf line number information placed into a section called
762 @var{.debug_line.text.func}. If the code section is just called @var{.text}
763 then debug line section will still be called just @var{.debug_line} without any
767 @item --size-check=error
768 @itemx --size-check=warning
769 Issue an error or warning for invalid ELF .size directive.
771 @item --elf-stt-common=no
772 @itemx --elf-stt-common=yes
773 These options control whether the ELF assembler should generate common
774 symbols with the @code{STT_COMMON} type. The default can be controlled
775 by a configure option @option{--enable-elf-stt-common}.
777 @item --generate-missing-build-notes=yes
778 @itemx --generate-missing-build-notes=no
779 These options control whether the ELF assembler should generate GNU Build
780 attribute notes if none are present in the input sources.
781 The default can be controlled by the @option{--enable-generate-build-notes}
787 Print a summary of the command-line options and exit.
790 Print a summary of all target specific options and exit.
793 Add directory @var{dir} to the search list for @code{.include} directives.
796 Don't warn about signed overflow.
799 @ifclear DIFF-TBL-KLUGE
800 This option is accepted but has no effect on the @value{TARGET} family.
802 @ifset DIFF-TBL-KLUGE
803 Issue warnings when difference tables altered for long displacements.
808 Keep (in the symbol table) local symbols. These symbols start with
809 system-specific local label prefixes, typically @samp{.L} for ELF systems
810 or @samp{L} for traditional a.out systems.
815 @item --listing-lhs-width=@var{number}
816 Set the maximum width, in words, of the output data column for an assembler
817 listing to @var{number}.
819 @item --listing-lhs-width2=@var{number}
820 Set the maximum width, in words, of the output data column for continuation
821 lines in an assembler listing to @var{number}.
823 @item --listing-rhs-width=@var{number}
824 Set the maximum width of an input source line, as displayed in a listing, to
827 @item --listing-cont-lines=@var{number}
828 Set the maximum number of lines printed in a listing for a single line of input
831 @item --no-pad-sections
832 Stop the assembler for padding the ends of output sections to the alignment
833 of that section. The default is to pad the sections, but this can waste space
834 which might be needed on targets which have tight memory constraints.
836 @item -o @var{objfile}
837 Name the object-file output from @command{@value{AS}} @var{objfile}.
840 Fold the data section into the text section.
842 @item --hash-size=@var{number}
843 Set the default size of GAS's hash tables to a prime number close to
844 @var{number}. Increasing this value can reduce the length of time it takes the
845 assembler to perform its tasks, at the expense of increasing the assembler's
846 memory requirements. Similarly reducing this value can reduce the memory
847 requirements at the expense of speed.
849 @item --reduce-memory-overheads
850 This option reduces GAS's memory requirements, at the expense of making the
851 assembly processes slower. Currently this switch is a synonym for
852 @samp{--hash-size=4051}, but in the future it may have other effects as well.
855 @item --sectname-subst
856 Honor substitution sequences in section names.
858 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
863 Print the maximum space (in bytes) and total time (in seconds) used by
866 @item --strip-local-absolute
867 Remove local absolute symbols from the outgoing symbol table.
871 Print the @command{as} version.
874 Print the @command{as} version and exit.
878 Suppress warning messages.
880 @item --fatal-warnings
881 Treat warnings as errors.
884 Don't suppress warning messages or treat them as errors.
893 Generate an object file even after errors.
895 @item -- | @var{files} @dots{}
896 Standard input, or source files to assemble.
904 @xref{AArch64 Options}, for the options available when @value{AS} is configured
905 for the 64-bit mode of the ARM Architecture (AArch64).
910 The following options are available when @value{AS} is configured for the
911 64-bit mode of the ARM Architecture (AArch64).
914 @include c-aarch64.texi
915 @c ended inside the included file
923 @xref{Alpha Options}, for the options available when @value{AS} is configured
924 for an Alpha processor.
929 The following options are available when @value{AS} is configured for an Alpha
933 @include c-alpha.texi
934 @c ended inside the included file
941 The following options are available when @value{AS} is configured for an ARC
945 @item -mcpu=@var{cpu}
946 This option selects the core processor variant.
948 Select either big-endian (-EB) or little-endian (-EL) output.
950 Enable Code Density extenssion instructions.
955 The following options are available when @value{AS} is configured for the ARM
959 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
960 Specify which ARM processor variant is the target.
961 @item -march=@var{architecture}[+@var{extension}@dots{}]
962 Specify which ARM architecture variant is used by the target.
963 @item -mfpu=@var{floating-point-format}
964 Select which Floating Point architecture is the target.
965 @item -mfloat-abi=@var{abi}
966 Select which floating point ABI is in use.
968 Enable Thumb only instruction decoding.
969 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
970 Select which procedure calling convention is in use.
972 Select either big-endian (-EB) or little-endian (-EL) output.
973 @item -mthumb-interwork
974 Specify that the code has been generated with interworking between Thumb and
977 Turns on CodeComposer Studio assembly syntax compatibility mode.
979 Specify that PIC code has been generated.
987 @xref{Blackfin Options}, for the options available when @value{AS} is
988 configured for the Blackfin processor family.
993 The following options are available when @value{AS} is configured for
994 the Blackfin processor family.
998 @c ended inside the included file
1003 @c man begin OPTIONS
1005 See the info pages for documentation of the CRIS-specific options.
1011 @xref{C-SKY Options}, for the options available when @value{AS} is
1012 configured for the C-SKY processor family.
1016 @c man begin OPTIONS
1017 The following options are available when @value{AS} is configured for
1018 the C-SKY processor family.
1020 @c man begin INCLUDE
1021 @include c-csky.texi
1022 @c ended inside the included file
1028 The following options are available when @value{AS} is configured for
1031 @cindex D10V optimization
1032 @cindex optimization, D10V
1034 Optimize output by parallelizing instructions.
1039 The following options are available when @value{AS} is configured for a D30V
1042 @cindex D30V optimization
1043 @cindex optimization, D30V
1045 Optimize output by parallelizing instructions.
1049 Warn when nops are generated.
1051 @cindex D30V nops after 32-bit multiply
1053 Warn when a nop after a 32-bit multiply instruction is generated.
1059 The following options are available when @value{AS} is configured for the
1060 Adapteva EPIPHANY series.
1063 @xref{Epiphany Options}, for the options available when @value{AS} is
1064 configured for an Epiphany processor.
1068 @c man begin OPTIONS
1069 The following options are available when @value{AS} is configured for
1070 an Epiphany processor.
1072 @c man begin INCLUDE
1073 @include c-epiphany.texi
1074 @c ended inside the included file
1082 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1083 for an H8/300 processor.
1087 @c man begin OPTIONS
1088 The following options are available when @value{AS} is configured for an H8/300
1091 @c man begin INCLUDE
1092 @include c-h8300.texi
1093 @c ended inside the included file
1101 @xref{i386-Options}, for the options available when @value{AS} is
1102 configured for an i386 processor.
1106 @c man begin OPTIONS
1107 The following options are available when @value{AS} is configured for
1110 @c man begin INCLUDE
1111 @include c-i386.texi
1112 @c ended inside the included file
1117 @c man begin OPTIONS
1119 The following options are available when @value{AS} is configured for the
1125 Specifies that the extended IP2022 instructions are allowed.
1128 Restores the default behaviour, which restricts the permitted instructions to
1129 just the basic IP2022 ones.
1135 The following options are available when @value{AS} is configured for the
1136 Renesas M32C and M16C processors.
1141 Assemble M32C instructions.
1144 Assemble M16C instructions (the default).
1147 Enable support for link-time relaxations.
1150 Support H'00 style hex constants in addition to 0x00 style.
1156 The following options are available when @value{AS} is configured for the
1157 Renesas M32R (formerly Mitsubishi M32R) series.
1162 Specify which processor in the M32R family is the target. The default
1163 is normally the M32R, but this option changes it to the M32RX.
1165 @item --warn-explicit-parallel-conflicts or --Wp
1166 Produce warning messages when questionable parallel constructs are
1169 @item --no-warn-explicit-parallel-conflicts or --Wnp
1170 Do not produce warning messages when questionable parallel constructs are
1177 The following options are available when @value{AS} is configured for the
1178 Motorola 68000 series.
1183 Shorten references to undefined symbols, to one word instead of two.
1185 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1186 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1187 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1188 Specify what processor in the 68000 family is the target. The default
1189 is normally the 68020, but this can be changed at configuration time.
1191 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1192 The target machine does (or does not) have a floating-point coprocessor.
1193 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1194 the basic 68000 is not compatible with the 68881, a combination of the
1195 two can be specified, since it's possible to do emulation of the
1196 coprocessor instructions with the main processor.
1198 @item -m68851 | -mno-68851
1199 The target machine does (or does not) have a memory-management
1200 unit coprocessor. The default is to assume an MMU for 68020 and up.
1208 @xref{Nios II Options}, for the options available when @value{AS} is configured
1209 for an Altera Nios II processor.
1213 @c man begin OPTIONS
1214 The following options are available when @value{AS} is configured for an
1215 Altera Nios II processor.
1217 @c man begin INCLUDE
1218 @include c-nios2.texi
1219 @c ended inside the included file
1225 For details about the PDP-11 machine dependent features options,
1226 see @ref{PDP-11-Options}.
1229 @item -mpic | -mno-pic
1230 Generate position-independent (or position-dependent) code. The
1231 default is @option{-mpic}.
1234 @itemx -mall-extensions
1235 Enable all instruction set extensions. This is the default.
1237 @item -mno-extensions
1238 Disable all instruction set extensions.
1240 @item -m@var{extension} | -mno-@var{extension}
1241 Enable (or disable) a particular instruction set extension.
1244 Enable the instruction set extensions supported by a particular CPU, and
1245 disable all other extensions.
1247 @item -m@var{machine}
1248 Enable the instruction set extensions supported by a particular machine
1249 model, and disable all other extensions.
1255 The following options are available when @value{AS} is configured for
1256 a picoJava processor.
1260 @cindex PJ endianness
1261 @cindex endianness, PJ
1262 @cindex big endian output, PJ
1264 Generate ``big endian'' format output.
1266 @cindex little endian output, PJ
1268 Generate ``little endian'' format output.
1276 @xref{PRU Options}, for the options available when @value{AS} is configured
1277 for a PRU processor.
1281 @c man begin OPTIONS
1282 The following options are available when @value{AS} is configured for a
1285 @c man begin INCLUDE
1287 @c ended inside the included file
1292 The following options are available when @value{AS} is configured for the
1293 Motorola 68HC11 or 68HC12 series.
1297 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1298 Specify what processor is the target. The default is
1299 defined by the configuration option when building the assembler.
1301 @item --xgate-ramoffset
1302 Instruct the linker to offset RAM addresses from S12X address space into
1303 XGATE address space.
1306 Specify to use the 16-bit integer ABI.
1309 Specify to use the 32-bit integer ABI.
1311 @item -mshort-double
1312 Specify to use the 32-bit double ABI.
1315 Specify to use the 64-bit double ABI.
1317 @item --force-long-branches
1318 Relative branches are turned into absolute ones. This concerns
1319 conditional branches, unconditional branches and branches to a
1322 @item -S | --short-branches
1323 Do not turn relative branches into absolute ones
1324 when the offset is out of range.
1326 @item --strict-direct-mode
1327 Do not turn the direct addressing mode into extended addressing mode
1328 when the instruction does not support direct addressing mode.
1330 @item --print-insn-syntax
1331 Print the syntax of instruction in case of error.
1333 @item --print-opcodes
1334 Print the list of instructions with syntax and then exit.
1336 @item --generate-example
1337 Print an example of instruction for each possible instruction and then exit.
1338 This option is only useful for testing @command{@value{AS}}.
1344 The following options are available when @command{@value{AS}} is configured
1345 for the SPARC architecture:
1348 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1349 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1350 Explicitly select a variant of the SPARC architecture.
1352 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1353 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1355 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1356 UltraSPARC extensions.
1358 @item -xarch=v8plus | -xarch=v8plusa
1359 For compatibility with the Solaris v9 assembler. These options are
1360 equivalent to -Av8plus and -Av8plusa, respectively.
1363 Warn when the assembler switches to another architecture.
1368 The following options are available when @value{AS} is configured for the 'c54x
1373 Enable extended addressing mode. All addresses and relocations will assume
1374 extended addressing (usually 23 bits).
1375 @item -mcpu=@var{CPU_VERSION}
1376 Sets the CPU version being compiled for.
1377 @item -merrors-to-file @var{FILENAME}
1378 Redirect error output to a file, for broken systems which don't support such
1379 behaviour in the shell.
1384 @c man begin OPTIONS
1385 The following options are available when @value{AS} is configured for
1390 This option sets the largest size of an object that can be referenced
1391 implicitly with the @code{gp} register. It is only accepted for targets that
1392 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1394 @cindex MIPS endianness
1395 @cindex endianness, MIPS
1396 @cindex big endian output, MIPS
1398 Generate ``big endian'' format output.
1400 @cindex little endian output, MIPS
1402 Generate ``little endian'' format output.
1420 Generate code for a particular MIPS Instruction Set Architecture level.
1421 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1422 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1423 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1424 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1425 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1426 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1427 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1428 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1429 MIPS64 Release 6 ISA processors, respectively.
1431 @item -march=@var{cpu}
1432 Generate code for a particular MIPS CPU.
1434 @item -mtune=@var{cpu}
1435 Schedule and tune for a particular MIPS CPU.
1439 Cause nops to be inserted if the read of the destination register
1440 of an mfhi or mflo instruction occurs in the following two instructions.
1443 @itemx -mno-fix-rm7000
1444 Cause nops to be inserted if a dmult or dmultu instruction is
1445 followed by a load instruction.
1449 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1450 section instead of the standard ELF .stabs sections.
1454 Control generation of @code{.pdr} sections.
1458 The register sizes are normally inferred from the ISA and ABI, but these
1459 flags force a certain group of registers to be treated as 32 bits wide at
1460 all times. @samp{-mgp32} controls the size of general-purpose registers
1461 and @samp{-mfp32} controls the size of floating-point registers.
1465 The register sizes are normally inferred from the ISA and ABI, but these
1466 flags force a certain group of registers to be treated as 64 bits wide at
1467 all times. @samp{-mgp64} controls the size of general-purpose registers
1468 and @samp{-mfp64} controls the size of floating-point registers.
1471 The register sizes are normally inferred from the ISA and ABI, but using
1472 this flag in combination with @samp{-mabi=32} enables an ABI variant
1473 which will operate correctly with floating-point registers which are
1477 @itemx -mno-odd-spreg
1478 Enable use of floating-point operations on odd-numbered single-precision
1479 registers when supported by the ISA. @samp{-mfpxx} implies
1480 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1484 Generate code for the MIPS 16 processor. This is equivalent to putting
1485 @code{.module mips16} at the start of the assembly file. @samp{-no-mips16}
1486 turns off this option.
1489 @itemx -mno-mips16e2
1490 Enable the use of MIPS16e2 instructions in MIPS16 mode. This is equivalent
1491 to putting @code{.module mips16e2} at the start of the assembly file.
1492 @samp{-mno-mips16e2} turns off this option.
1495 @itemx -mno-micromips
1496 Generate code for the microMIPS processor. This is equivalent to putting
1497 @code{.module micromips} at the start of the assembly file.
1498 @samp{-mno-micromips} turns off this option. This is equivalent to putting
1499 @code{.module nomicromips} at the start of the assembly file.
1502 @itemx -mno-smartmips
1503 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1504 equivalent to putting @code{.module smartmips} at the start of the assembly
1505 file. @samp{-mno-smartmips} turns off this option.
1509 Generate code for the MIPS-3D Application Specific Extension.
1510 This tells the assembler to accept MIPS-3D instructions.
1511 @samp{-no-mips3d} turns off this option.
1515 Generate code for the MDMX Application Specific Extension.
1516 This tells the assembler to accept MDMX instructions.
1517 @samp{-no-mdmx} turns off this option.
1521 Generate code for the DSP Release 1 Application Specific Extension.
1522 This tells the assembler to accept DSP Release 1 instructions.
1523 @samp{-mno-dsp} turns off this option.
1527 Generate code for the DSP Release 2 Application Specific Extension.
1528 This option implies @samp{-mdsp}.
1529 This tells the assembler to accept DSP Release 2 instructions.
1530 @samp{-mno-dspr2} turns off this option.
1534 Generate code for the DSP Release 3 Application Specific Extension.
1535 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1536 This tells the assembler to accept DSP Release 3 instructions.
1537 @samp{-mno-dspr3} turns off this option.
1541 Generate code for the MIPS SIMD Architecture Extension.
1542 This tells the assembler to accept MSA instructions.
1543 @samp{-mno-msa} turns off this option.
1547 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1548 This tells the assembler to accept XPA instructions.
1549 @samp{-mno-xpa} turns off this option.
1553 Generate code for the MT Application Specific Extension.
1554 This tells the assembler to accept MT instructions.
1555 @samp{-mno-mt} turns off this option.
1559 Generate code for the MCU Application Specific Extension.
1560 This tells the assembler to accept MCU instructions.
1561 @samp{-mno-mcu} turns off this option.
1565 Generate code for the MIPS cyclic redundancy check (CRC) Application
1566 Specific Extension. This tells the assembler to accept CRC instructions.
1567 @samp{-mno-crc} turns off this option.
1571 Generate code for the Global INValidate (GINV) Application Specific
1572 Extension. This tells the assembler to accept GINV instructions.
1573 @samp{-mno-ginv} turns off this option.
1575 @item -mloongson-mmi
1576 @itemx -mno-loongson-mmi
1577 Generate code for the Loongson MultiMedia extensions Instructions (MMI)
1578 Application Specific Extension. This tells the assembler to accept MMI
1580 @samp{-mno-loongson-mmi} turns off this option.
1582 @item -mloongson-cam
1583 @itemx -mno-loongson-cam
1584 Generate code for the Loongson Content Address Memory (CAM) instructions.
1585 This tells the assembler to accept Loongson CAM instructions.
1586 @samp{-mno-loongson-cam} turns off this option.
1588 @item -mloongson-ext
1589 @itemx -mno-loongson-ext
1590 Generate code for the Loongson EXTensions (EXT) instructions.
1591 This tells the assembler to accept Loongson EXT instructions.
1592 @samp{-mno-loongson-ext} turns off this option.
1594 @item -mloongson-ext2
1595 @itemx -mno-loongson-ext2
1596 Generate code for the Loongson EXTensions R2 (EXT2) instructions.
1597 This option implies @samp{-mloongson-ext}.
1598 This tells the assembler to accept Loongson EXT2 instructions.
1599 @samp{-mno-loongson-ext2} turns off this option.
1603 Only use 32-bit instruction encodings when generating code for the
1604 microMIPS processor. This option inhibits the use of any 16-bit
1605 instructions. This is equivalent to putting @code{.set insn32} at
1606 the start of the assembly file. @samp{-mno-insn32} turns off this
1607 option. This is equivalent to putting @code{.set noinsn32} at the
1608 start of the assembly file. By default @samp{-mno-insn32} is
1609 selected, allowing all instructions to be used.
1611 @item --construct-floats
1612 @itemx --no-construct-floats
1613 The @samp{--no-construct-floats} option disables the construction of
1614 double width floating point constants by loading the two halves of the
1615 value into the two single width floating point registers that make up
1616 the double width register. By default @samp{--construct-floats} is
1617 selected, allowing construction of these floating point constants.
1619 @item --relax-branch
1620 @itemx --no-relax-branch
1621 The @samp{--relax-branch} option enables the relaxation of out-of-range
1622 branches. By default @samp{--no-relax-branch} is selected, causing any
1623 out-of-range branches to produce an error.
1625 @item -mignore-branch-isa
1626 @itemx -mno-ignore-branch-isa
1627 Ignore branch checks for invalid transitions between ISA modes. The
1628 semantics of branches does not provide for an ISA mode switch, so in
1629 most cases the ISA mode a branch has been encoded for has to be the
1630 same as the ISA mode of the branch's target label. Therefore GAS has
1631 checks implemented that verify in branch assembly that the two ISA
1632 modes match. @samp{-mignore-branch-isa} disables these checks. By
1633 default @samp{-mno-ignore-branch-isa} is selected, causing any invalid
1634 branch requiring a transition between ISA modes to produce an error.
1636 @item -mnan=@var{encoding}
1637 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1638 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1641 @item --emulation=@var{name}
1642 This option was formerly used to switch between ELF and ECOFF output
1643 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1644 removed in GAS 2.24, so the option now serves little purpose.
1645 It is retained for backwards compatibility.
1647 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1648 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1649 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1650 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1651 preferred options instead.
1654 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1661 Control how to deal with multiplication overflow and division by zero.
1662 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1663 (and only work for Instruction Set Architecture level 2 and higher);
1664 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1668 When this option is used, @command{@value{AS}} will issue a warning every
1669 time it generates a nop instruction from a macro.
1675 The following options are available when @value{AS} is configured for
1681 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1682 The command-line option @samp{-nojsri2bsr} can be used to disable it.
1686 Enable or disable the silicon filter behaviour. By default this is disabled.
1687 The default can be overridden by the @samp{-sifilter} command-line option.
1690 Alter jump instructions for long displacements.
1692 @item -mcpu=[210|340]
1693 Select the cpu type on the target hardware. This controls which instructions
1697 Assemble for a big endian target.
1700 Assemble for a little endian target.
1709 @xref{Meta Options}, for the options available when @value{AS} is configured
1710 for a Meta processor.
1714 @c man begin OPTIONS
1715 The following options are available when @value{AS} is configured for a
1718 @c man begin INCLUDE
1719 @include c-metag.texi
1720 @c ended inside the included file
1725 @c man begin OPTIONS
1727 See the info pages for documentation of the MMIX-specific options.
1733 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1734 for a NDS32 processor.
1736 @c ended inside the included file
1740 @c man begin OPTIONS
1741 The following options are available when @value{AS} is configured for a
1744 @c man begin INCLUDE
1745 @include c-nds32.texi
1746 @c ended inside the included file
1753 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1754 for a PowerPC processor.
1758 @c man begin OPTIONS
1759 The following options are available when @value{AS} is configured for a
1762 @c man begin INCLUDE
1764 @c ended inside the included file
1772 @xref{RISC-V-Options}, for the options available when @value{AS} is configured
1773 for a RISC-V processor.
1777 @c man begin OPTIONS
1778 The following options are available when @value{AS} is configured for a
1781 @c man begin INCLUDE
1782 @include c-riscv.texi
1783 @c ended inside the included file
1788 @c man begin OPTIONS
1790 See the info pages for documentation of the RX-specific options.
1794 The following options are available when @value{AS} is configured for the s390
1800 Select the word size, either 31/32 bits or 64 bits.
1803 Select the architecture mode, either the Enterprise System
1804 Architecture (esa) or the z/Architecture mode (zarch).
1805 @item -march=@var{processor}
1806 Specify which s390 processor variant is the target, @samp{g5} (or
1807 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1808 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1809 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1810 @samp{z13} (or @samp{arch11}), or @samp{z14} (or @samp{arch12}).
1812 @itemx -mno-regnames
1813 Allow or disallow symbolic names for registers.
1814 @item -mwarn-areg-zero
1815 Warn whenever the operand for a base or index register has been specified
1816 but evaluates to zero.
1824 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1825 for a TMS320C6000 processor.
1829 @c man begin OPTIONS
1830 The following options are available when @value{AS} is configured for a
1831 TMS320C6000 processor.
1833 @c man begin INCLUDE
1834 @include c-tic6x.texi
1835 @c ended inside the included file
1843 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1844 for a TILE-Gx processor.
1848 @c man begin OPTIONS
1849 The following options are available when @value{AS} is configured for a TILE-Gx
1852 @c man begin INCLUDE
1853 @include c-tilegx.texi
1854 @c ended inside the included file
1862 @xref{Visium Options}, for the options available when @value{AS} is configured
1863 for a Visium processor.
1867 @c man begin OPTIONS
1868 The following option is available when @value{AS} is configured for a Visium
1871 @c man begin INCLUDE
1872 @include c-visium.texi
1873 @c ended inside the included file
1881 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1882 for an Xtensa processor.
1886 @c man begin OPTIONS
1887 The following options are available when @value{AS} is configured for an
1890 @c man begin INCLUDE
1891 @include c-xtensa.texi
1892 @c ended inside the included file
1897 @c man begin OPTIONS
1900 The following options are available when @value{AS} is configured for
1901 a Z80 family processor.
1904 Assemble for Z80 processor.
1906 Assemble for R800 processor.
1907 @item -ignore-undocumented-instructions
1909 Assemble undocumented Z80 instructions that also work on R800 without warning.
1910 @item -ignore-unportable-instructions
1912 Assemble all undocumented Z80 instructions without warning.
1913 @item -warn-undocumented-instructions
1915 Issue a warning for undocumented Z80 instructions that also work on R800.
1916 @item -warn-unportable-instructions
1918 Issue a warning for undocumented Z80 instructions that do not work on R800.
1919 @item -forbid-undocumented-instructions
1921 Treat all undocumented instructions as errors.
1922 @item -forbid-unportable-instructions
1924 Treat undocumented Z80 instructions that do not work on R800 as errors.
1931 * Manual:: Structure of this Manual
1932 * GNU Assembler:: The GNU Assembler
1933 * Object Formats:: Object File Formats
1934 * Command Line:: Command Line
1935 * Input Files:: Input Files
1936 * Object:: Output (Object) File
1937 * Errors:: Error and Warning Messages
1941 @section Structure of this Manual
1943 @cindex manual, structure and purpose
1944 This manual is intended to describe what you need to know to use
1945 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1946 notation for symbols, constants, and expressions; the directives that
1947 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1950 We also cover special features in the @value{TARGET}
1951 configuration of @command{@value{AS}}, including assembler directives.
1954 This manual also describes some of the machine-dependent features of
1955 various flavors of the assembler.
1958 @cindex machine instructions (not covered)
1959 On the other hand, this manual is @emph{not} intended as an introduction
1960 to programming in assembly language---let alone programming in general!
1961 In a similar vein, we make no attempt to introduce the machine
1962 architecture; we do @emph{not} describe the instruction set, standard
1963 mnemonics, registers or addressing modes that are standard to a
1964 particular architecture.
1966 You may want to consult the manufacturer's
1967 machine architecture manual for this information.
1971 For information on the H8/300 machine instruction set, see @cite{H8/300
1972 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1973 Programming Manual} (Renesas).
1976 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1977 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1978 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1979 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1982 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1986 @c I think this is premature---doc@cygnus.com, 17jan1991
1988 Throughout this manual, we assume that you are running @dfn{GNU},
1989 the portable operating system from the @dfn{Free Software
1990 Foundation, Inc.}. This restricts our attention to certain kinds of
1991 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1992 once this assumption is granted examples and definitions need less
1995 @command{@value{AS}} is part of a team of programs that turn a high-level
1996 human-readable series of instructions into a low-level
1997 computer-readable series of instructions. Different versions of
1998 @command{@value{AS}} are used for different kinds of computer.
2001 @c There used to be a section "Terminology" here, which defined
2002 @c "contents", "byte", "word", and "long". Defining "word" to any
2003 @c particular size is confusing when the .word directive may generate 16
2004 @c bits on one machine and 32 bits on another; in general, for the user
2005 @c version of this manual, none of these terms seem essential to define.
2006 @c They were used very little even in the former draft of the manual;
2007 @c this draft makes an effort to avoid them (except in names of
2011 @section The GNU Assembler
2013 @c man begin DESCRIPTION
2015 @sc{gnu} @command{as} is really a family of assemblers.
2017 This manual describes @command{@value{AS}}, a member of that family which is
2018 configured for the @value{TARGET} architectures.
2020 If you use (or have used) the @sc{gnu} assembler on one architecture, you
2021 should find a fairly similar environment when you use it on another
2022 architecture. Each version has much in common with the others,
2023 including object file formats, most assembler directives (often called
2024 @dfn{pseudo-ops}) and assembler syntax.@refill
2026 @cindex purpose of @sc{gnu} assembler
2027 @command{@value{AS}} is primarily intended to assemble the output of the
2028 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
2029 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
2030 assemble correctly everything that other assemblers for the same
2031 machine would assemble.
2033 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
2036 @c This remark should appear in generic version of manual; assumption
2037 @c here is that generic version sets M680x0.
2038 This doesn't mean @command{@value{AS}} always uses the same syntax as another
2039 assembler for the same architecture; for example, we know of several
2040 incompatible versions of 680x0 assembly language syntax.
2045 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
2046 program in one pass of the source file. This has a subtle impact on the
2047 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
2049 @node Object Formats
2050 @section Object File Formats
2052 @cindex object file format
2053 The @sc{gnu} assembler can be configured to produce several alternative
2054 object file formats. For the most part, this does not affect how you
2055 write assembly language programs; but directives for debugging symbols
2056 are typically different in different file formats. @xref{Symbol
2057 Attributes,,Symbol Attributes}.
2060 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
2061 @value{OBJ-NAME} format object files.
2063 @c The following should exhaust all configs that set MULTI-OBJ, ideally
2065 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
2066 SOM or ELF format object files.
2071 @section Command Line
2073 @cindex command line conventions
2075 After the program name @command{@value{AS}}, the command line may contain
2076 options and file names. Options may appear in any order, and may be
2077 before, after, or between file names. The order of file names is
2080 @cindex standard input, as input file
2082 @file{--} (two hyphens) by itself names the standard input file
2083 explicitly, as one of the files for @command{@value{AS}} to assemble.
2085 @cindex options, command line
2086 Except for @samp{--} any command-line argument that begins with a
2087 hyphen (@samp{-}) is an option. Each option changes the behavior of
2088 @command{@value{AS}}. No option changes the way another option works. An
2089 option is a @samp{-} followed by one or more letters; the case of
2090 the letter is important. All options are optional.
2092 Some options expect exactly one file name to follow them. The file
2093 name may either immediately follow the option's letter (compatible
2094 with older assemblers) or it may be the next command argument (@sc{gnu}
2095 standard). These two command lines are equivalent:
2098 @value{AS} -o my-object-file.o mumble.s
2099 @value{AS} -omy-object-file.o mumble.s
2103 @section Input Files
2106 @cindex source program
2107 @cindex files, input
2108 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2109 describe the program input to one run of @command{@value{AS}}. The program may
2110 be in one or more files; how the source is partitioned into files
2111 doesn't change the meaning of the source.
2113 @c I added "con" prefix to "catenation" just to prove I can overcome my
2114 @c APL training... doc@cygnus.com
2115 The source program is a concatenation of the text in all the files, in the
2118 @c man begin DESCRIPTION
2119 Each time you run @command{@value{AS}} it assembles exactly one source
2120 program. The source program is made up of one or more files.
2121 (The standard input is also a file.)
2123 You give @command{@value{AS}} a command line that has zero or more input file
2124 names. The input files are read (from left file name to right). A
2125 command-line argument (in any position) that has no special meaning
2126 is taken to be an input file name.
2128 If you give @command{@value{AS}} no file names it attempts to read one input file
2129 from the @command{@value{AS}} standard input, which is normally your terminal. You
2130 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2133 Use @samp{--} if you need to explicitly name the standard input file
2134 in your command line.
2136 If the source is empty, @command{@value{AS}} produces a small, empty object
2141 @subheading Filenames and Line-numbers
2143 @cindex input file linenumbers
2144 @cindex line numbers, in input files
2145 There are two ways of locating a line in the input file (or files) and
2146 either may be used in reporting error messages. One way refers to a line
2147 number in a physical file; the other refers to a line number in a
2148 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2150 @dfn{Physical files} are those files named in the command line given
2151 to @command{@value{AS}}.
2153 @dfn{Logical files} are simply names declared explicitly by assembler
2154 directives; they bear no relation to physical files. Logical file names help
2155 error messages reflect the original source file, when @command{@value{AS}} source
2156 is itself synthesized from other files. @command{@value{AS}} understands the
2157 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2158 @ref{File,,@code{.file}}.
2161 @section Output (Object) File
2167 Every time you run @command{@value{AS}} it produces an output file, which is
2168 your assembly language program translated into numbers. This file
2169 is the object file. Its default name is @code{a.out}.
2170 You can give it another name by using the @option{-o} option. Conventionally,
2171 object file names end with @file{.o}. The default name is used for historical
2172 reasons: older assemblers were capable of assembling self-contained programs
2173 directly into a runnable program. (For some formats, this isn't currently
2174 possible, but it can be done for the @code{a.out} format.)
2178 The object file is meant for input to the linker @code{@value{LD}}. It contains
2179 assembled program code, information to help @code{@value{LD}} integrate
2180 the assembled program into a runnable file, and (optionally) symbolic
2181 information for the debugger.
2183 @c link above to some info file(s) like the description of a.out.
2184 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2187 @section Error and Warning Messages
2189 @c man begin DESCRIPTION
2191 @cindex error messages
2192 @cindex warning messages
2193 @cindex messages from assembler
2194 @command{@value{AS}} may write warnings and error messages to the standard error
2195 file (usually your terminal). This should not happen when a compiler
2196 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2197 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2198 grave problem that stops the assembly.
2202 @cindex format of warning messages
2203 Warning messages have the format
2206 file_name:@b{NNN}:Warning Message Text
2210 @cindex file names and line numbers, in warnings/errors
2211 (where @b{NNN} is a line number). If both a logical file name
2212 (@pxref{File,,@code{.file}}) and a logical line number
2214 (@pxref{Line,,@code{.line}})
2216 have been given then they will be used, otherwise the file name and line number
2217 in the current assembler source file will be used. The message text is
2218 intended to be self explanatory (in the grand Unix tradition).
2220 Note the file name must be set via the logical version of the @code{.file}
2221 directive, not the DWARF2 version of the @code{.file} directive. For example:
2225 error_assembler_source
2231 produces this output:
2235 asm.s:2: Error: no such instruction: `error_assembler_source'
2236 foo.c:31: Error: no such instruction: `error_c_source'
2239 @cindex format of error messages
2240 Error messages have the format
2243 file_name:@b{NNN}:FATAL:Error Message Text
2246 The file name and line number are derived as for warning
2247 messages. The actual message text may be rather less explanatory
2248 because many of them aren't supposed to happen.
2251 @chapter Command-Line Options
2253 @cindex options, all versions of assembler
2254 This chapter describes command-line options available in @emph{all}
2255 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2256 for options specific
2258 to the @value{TARGET} target.
2261 to particular machine architectures.
2264 @c man begin DESCRIPTION
2266 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2267 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2268 The assembler arguments must be separated from each other (and the @samp{-Wa})
2269 by commas. For example:
2272 gcc -c -g -O -Wa,-alh,-L file.c
2276 This passes two options to the assembler: @samp{-alh} (emit a listing to
2277 standard output with high-level and assembly source) and @samp{-L} (retain
2278 local symbols in the symbol table).
2280 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2281 command-line options are automatically passed to the assembler by the compiler.
2282 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2283 precisely what options it passes to each compilation pass, including the
2289 * a:: -a[cdghlns] enable listings
2290 * alternate:: --alternate enable alternate macro syntax
2291 * D:: -D for compatibility
2292 * f:: -f to work faster
2293 * I:: -I for .include search path
2294 @ifclear DIFF-TBL-KLUGE
2295 * K:: -K for compatibility
2297 @ifset DIFF-TBL-KLUGE
2298 * K:: -K for difference tables
2301 * L:: -L to retain local symbols
2302 * listing:: --listing-XXX to configure listing output
2303 * M:: -M or --mri to assemble in MRI compatibility mode
2304 * MD:: --MD for dependency tracking
2305 * no-pad-sections:: --no-pad-sections to stop section padding
2306 * o:: -o to name the object file
2307 * R:: -R to join data and text sections
2308 * statistics:: --statistics to see statistics about assembly
2309 * traditional-format:: --traditional-format for compatible output
2310 * v:: -v to announce version
2311 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2312 * Z:: -Z to make object file even after errors
2316 @section Enable Listings: @option{-a[cdghlns]}
2326 @cindex listings, enabling
2327 @cindex assembly listings, enabling
2329 These options enable listing output from the assembler. By itself,
2330 @samp{-a} requests high-level, assembly, and symbols listing.
2331 You can use other letters to select specific options for the list:
2332 @samp{-ah} requests a high-level language listing,
2333 @samp{-al} requests an output-program assembly listing, and
2334 @samp{-as} requests a symbol table listing.
2335 High-level listings require that a compiler debugging option like
2336 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2339 Use the @samp{-ag} option to print a first section with general assembly
2340 information, like @value{AS} version, switches passed, or time stamp.
2342 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2343 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2344 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2345 omitted from the listing.
2347 Use the @samp{-ad} option to omit debugging directives from the
2350 Once you have specified one of these options, you can further control
2351 listing output and its appearance using the directives @code{.list},
2352 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2354 The @samp{-an} option turns off all forms processing.
2355 If you do not request listing output with one of the @samp{-a} options, the
2356 listing-control directives have no effect.
2358 The letters after @samp{-a} may be combined into one option,
2359 @emph{e.g.}, @samp{-aln}.
2361 Note if the assembler source is coming from the standard input (e.g.,
2363 is being created by @code{@value{GCC}} and the @samp{-pipe} command-line switch
2364 is being used) then the listing will not contain any comments or preprocessor
2365 directives. This is because the listing code buffers input source lines from
2366 stdin only after they have been preprocessed by the assembler. This reduces
2367 memory usage and makes the code more efficient.
2370 @section @option{--alternate}
2373 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2376 @section @option{-D}
2379 This option has no effect whatsoever, but it is accepted to make it more
2380 likely that scripts written for other assemblers also work with
2381 @command{@value{AS}}.
2384 @section Work Faster: @option{-f}
2387 @cindex trusted compiler
2388 @cindex faster processing (@option{-f})
2389 @samp{-f} should only be used when assembling programs written by a
2390 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2391 and comment preprocessing on
2392 the input file(s) before assembling them. @xref{Preprocessing,
2396 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2397 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2402 @section @code{.include} Search Path: @option{-I} @var{path}
2404 @kindex -I @var{path}
2405 @cindex paths for @code{.include}
2406 @cindex search path for @code{.include}
2407 @cindex @code{include} directive search path
2408 Use this option to add a @var{path} to the list of directories
2409 @command{@value{AS}} searches for files specified in @code{.include}
2410 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2411 many times as necessary to include a variety of paths. The current
2412 working directory is always searched first; after that, @command{@value{AS}}
2413 searches any @samp{-I} directories in the same order as they were
2414 specified (left to right) on the command line.
2417 @section Difference Tables: @option{-K}
2420 @ifclear DIFF-TBL-KLUGE
2421 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2422 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2423 where it can be used to warn when the assembler alters the machine code
2424 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2425 family does not have the addressing limitations that sometimes lead to this
2426 alteration on other platforms.
2429 @ifset DIFF-TBL-KLUGE
2430 @cindex difference tables, warning
2431 @cindex warning for altered difference tables
2432 @command{@value{AS}} sometimes alters the code emitted for directives of the
2433 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2434 You can use the @samp{-K} option if you want a warning issued when this
2439 @section Include Local Symbols: @option{-L}
2442 @cindex local symbols, retaining in output
2443 Symbols beginning with system-specific local label prefixes, typically
2444 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2445 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2446 such symbols when debugging, because they are intended for the use of
2447 programs (like compilers) that compose assembler programs, not for your
2448 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2449 such symbols, so you do not normally debug with them.
2451 This option tells @command{@value{AS}} to retain those local symbols
2452 in the object file. Usually if you do this you also tell the linker
2453 @code{@value{LD}} to preserve those symbols.
2456 @section Configuring listing output: @option{--listing}
2458 The listing feature of the assembler can be enabled via the command-line switch
2459 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2460 hex dump of the corresponding locations in the output object file, and displays
2461 them as a listing file. The format of this listing can be controlled by
2462 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2463 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2464 @code{.psize} (@pxref{Psize}), and
2465 @code{.eject} (@pxref{Eject}) and also by the following switches:
2468 @item --listing-lhs-width=@samp{number}
2469 @kindex --listing-lhs-width
2470 @cindex Width of first line disassembly output
2471 Sets the maximum width, in words, of the first line of the hex byte dump. This
2472 dump appears on the left hand side of the listing output.
2474 @item --listing-lhs-width2=@samp{number}
2475 @kindex --listing-lhs-width2
2476 @cindex Width of continuation lines of disassembly output
2477 Sets the maximum width, in words, of any further lines of the hex byte dump for
2478 a given input source line. If this value is not specified, it defaults to being
2479 the same as the value specified for @samp{--listing-lhs-width}. If neither
2480 switch is used the default is to one.
2482 @item --listing-rhs-width=@samp{number}
2483 @kindex --listing-rhs-width
2484 @cindex Width of source line output
2485 Sets the maximum width, in characters, of the source line that is displayed
2486 alongside the hex dump. The default value for this parameter is 100. The
2487 source line is displayed on the right hand side of the listing output.
2489 @item --listing-cont-lines=@samp{number}
2490 @kindex --listing-cont-lines
2491 @cindex Maximum number of continuation lines
2492 Sets the maximum number of continuation lines of hex dump that will be
2493 displayed for a given single line of source input. The default value is 4.
2497 @section Assemble in MRI Compatibility Mode: @option{-M}
2500 @cindex MRI compatibility mode
2501 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2502 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2503 compatible with the @code{ASM68K} assembler from Microtec Research.
2504 The exact nature of the
2505 MRI syntax will not be documented here; see the MRI manuals for more
2506 information. Note in particular that the handling of macros and macro
2507 arguments is somewhat different. The purpose of this option is to permit
2508 assembling existing MRI assembler code using @command{@value{AS}}.
2510 The MRI compatibility is not complete. Certain operations of the MRI assembler
2511 depend upon its object file format, and can not be supported using other object
2512 file formats. Supporting these would require enhancing each object file format
2513 individually. These are:
2516 @item global symbols in common section
2518 The m68k MRI assembler supports common sections which are merged by the linker.
2519 Other object file formats do not support this. @command{@value{AS}} handles
2520 common sections by treating them as a single common symbol. It permits local
2521 symbols to be defined within a common section, but it can not support global
2522 symbols, since it has no way to describe them.
2524 @item complex relocations
2526 The MRI assemblers support relocations against a negated section address, and
2527 relocations which combine the start addresses of two or more sections. These
2528 are not support by other object file formats.
2530 @item @code{END} pseudo-op specifying start address
2532 The MRI @code{END} pseudo-op permits the specification of a start address.
2533 This is not supported by other object file formats. The start address may
2534 instead be specified using the @option{-e} option to the linker, or in a linker
2537 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2539 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2540 name to the output file. This is not supported by other object file formats.
2542 @item @code{ORG} pseudo-op
2544 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2545 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2546 which changes the location within the current section. Absolute sections are
2547 not supported by other object file formats. The address of a section may be
2548 assigned within a linker script.
2551 There are some other features of the MRI assembler which are not supported by
2552 @command{@value{AS}}, typically either because they are difficult or because they
2553 seem of little consequence. Some of these may be supported in future releases.
2557 @item EBCDIC strings
2559 EBCDIC strings are not supported.
2561 @item packed binary coded decimal
2563 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2564 and @code{DCB.P} pseudo-ops are not supported.
2566 @item @code{FEQU} pseudo-op
2568 The m68k @code{FEQU} pseudo-op is not supported.
2570 @item @code{NOOBJ} pseudo-op
2572 The m68k @code{NOOBJ} pseudo-op is not supported.
2574 @item @code{OPT} branch control options
2576 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2577 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2578 relaxes all branches, whether forward or backward, to an appropriate size, so
2579 these options serve no purpose.
2581 @item @code{OPT} list control options
2583 The following m68k @code{OPT} list control options are ignored: @code{C},
2584 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2585 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2587 @item other @code{OPT} options
2589 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2590 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2592 @item @code{OPT} @code{D} option is default
2594 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2595 @code{OPT NOD} may be used to turn it off.
2597 @item @code{XREF} pseudo-op.
2599 The m68k @code{XREF} pseudo-op is ignored.
2604 @section Dependency Tracking: @option{--MD}
2607 @cindex dependency tracking
2610 @command{@value{AS}} can generate a dependency file for the file it creates. This
2611 file consists of a single rule suitable for @code{make} describing the
2612 dependencies of the main source file.
2614 The rule is written to the file named in its argument.
2616 This feature is used in the automatic updating of makefiles.
2618 @node no-pad-sections
2619 @section Output Section Padding
2620 @kindex --no-pad-sections
2621 @cindex output section padding
2622 Normally the assembler will pad the end of each output section up to its
2623 alignment boundary. But this can waste space, which can be significant on
2624 memory constrained targets. So the @option{--no-pad-sections} option will
2625 disable this behaviour.
2628 @section Name the Object File: @option{-o}
2631 @cindex naming object file
2632 @cindex object file name
2633 There is always one object file output when you run @command{@value{AS}}. By
2634 default it has the name @file{a.out}.
2635 You use this option (which takes exactly one filename) to give the
2636 object file a different name.
2638 Whatever the object file is called, @command{@value{AS}} overwrites any
2639 existing file of the same name.
2642 @section Join Data and Text Sections: @option{-R}
2645 @cindex data and text sections, joining
2646 @cindex text and data sections, joining
2647 @cindex joining text and data sections
2648 @cindex merging text and data sections
2649 @option{-R} tells @command{@value{AS}} to write the object file as if all
2650 data-section data lives in the text section. This is only done at
2651 the very last moment: your binary data are the same, but data
2652 section parts are relocated differently. The data section part of
2653 your object file is zero bytes long because all its bytes are
2654 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2656 When you specify @option{-R} it would be possible to generate shorter
2657 address displacements (because we do not have to cross between text and
2658 data section). We refrain from doing this simply for compatibility with
2659 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2662 When @command{@value{AS}} is configured for COFF or ELF output,
2663 this option is only useful if you use sections named @samp{.text} and
2668 @option{-R} is not supported for any of the HPPA targets. Using
2669 @option{-R} generates a warning from @command{@value{AS}}.
2673 @section Display Assembly Statistics: @option{--statistics}
2675 @kindex --statistics
2676 @cindex statistics, about assembly
2677 @cindex time, total for assembly
2678 @cindex space used, maximum for assembly
2679 Use @samp{--statistics} to display two statistics about the resources used by
2680 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2681 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2684 @node traditional-format
2685 @section Compatible Output: @option{--traditional-format}
2687 @kindex --traditional-format
2688 For some targets, the output of @command{@value{AS}} is different in some ways
2689 from the output of some existing assembler. This switch requests
2690 @command{@value{AS}} to use the traditional format instead.
2692 For example, it disables the exception frame optimizations which
2693 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2696 @section Announce Version: @option{-v}
2700 @cindex assembler version
2701 @cindex version of assembler
2702 You can find out what version of as is running by including the
2703 option @samp{-v} (which you can also spell as @samp{-version}) on the
2707 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2709 @command{@value{AS}} should never give a warning or error message when
2710 assembling compiler output. But programs written by people often
2711 cause @command{@value{AS}} to give a warning that a particular assumption was
2712 made. All such warnings are directed to the standard error file.
2716 @cindex suppressing warnings
2717 @cindex warnings, suppressing
2718 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2719 This only affects the warning messages: it does not change any particular of
2720 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2723 @kindex --fatal-warnings
2724 @cindex errors, caused by warnings
2725 @cindex warnings, causing error
2726 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2727 files that generate warnings to be in error.
2730 @cindex warnings, switching on
2731 You can switch these options off again by specifying @option{--warn}, which
2732 causes warnings to be output as usual.
2735 @section Generate Object File in Spite of Errors: @option{-Z}
2736 @cindex object file, after errors
2737 @cindex errors, continuing after
2738 After an error message, @command{@value{AS}} normally produces no output. If for
2739 some reason you are interested in object file output even after
2740 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2741 option. If there are any errors, @command{@value{AS}} continues anyways, and
2742 writes an object file after a final warning message of the form @samp{@var{n}
2743 errors, @var{m} warnings, generating bad object file.}
2748 @cindex machine-independent syntax
2749 @cindex syntax, machine-independent
2750 This chapter describes the machine-independent syntax allowed in a
2751 source file. @command{@value{AS}} syntax is similar to what many other
2752 assemblers use; it is inspired by the BSD 4.2
2757 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2761 * Preprocessing:: Preprocessing
2762 * Whitespace:: Whitespace
2763 * Comments:: Comments
2764 * Symbol Intro:: Symbols
2765 * Statements:: Statements
2766 * Constants:: Constants
2770 @section Preprocessing
2772 @cindex preprocessing
2773 The @command{@value{AS}} internal preprocessor:
2775 @cindex whitespace, removed by preprocessor
2777 adjusts and removes extra whitespace. It leaves one space or tab before
2778 the keywords on a line, and turns any other whitespace on the line into
2781 @cindex comments, removed by preprocessor
2783 removes all comments, replacing them with a single space, or an
2784 appropriate number of newlines.
2786 @cindex constants, converted by preprocessor
2788 converts character constants into the appropriate numeric values.
2791 It does not do macro processing, include file handling, or
2792 anything else you may get from your C compiler's preprocessor. You can
2793 do include file processing with the @code{.include} directive
2794 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2795 to get other ``CPP'' style preprocessing by giving the input file a
2796 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2797 Output, gcc info, Using GNU CC}.
2799 Excess whitespace, comments, and character constants
2800 cannot be used in the portions of the input text that are not
2803 @cindex turning preprocessing on and off
2804 @cindex preprocessing, turning on and off
2807 If the first line of an input file is @code{#NO_APP} or if you use the
2808 @samp{-f} option, whitespace and comments are not removed from the input file.
2809 Within an input file, you can ask for whitespace and comment removal in
2810 specific portions of the by putting a line that says @code{#APP} before the
2811 text that may contain whitespace or comments, and putting a line that says
2812 @code{#NO_APP} after this text. This feature is mainly intend to support
2813 @code{asm} statements in compilers whose output is otherwise free of comments
2820 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2821 Whitespace is used to separate symbols, and to make programs neater for
2822 people to read. Unless within character constants
2823 (@pxref{Characters,,Character Constants}), any whitespace means the same
2824 as exactly one space.
2830 There are two ways of rendering comments to @command{@value{AS}}. In both
2831 cases the comment is equivalent to one space.
2833 Anything from @samp{/*} through the next @samp{*/} is a comment.
2834 This means you may not nest these comments.
2838 The only way to include a newline ('\n') in a comment
2839 is to use this sort of comment.
2842 /* This sort of comment does not nest. */
2845 @cindex line comment character
2846 Anything from a @dfn{line comment} character up to the next newline is
2847 considered a comment and is ignored. The line comment character is target
2848 specific, and some targets multiple comment characters. Some targets also have
2849 line comment characters that only work if they are the first character on a
2850 line. Some targets use a sequence of two characters to introduce a line
2851 comment. Some targets can also change their line comment characters depending
2852 upon command-line options that have been used. For more details see the
2853 @emph{Syntax} section in the documentation for individual targets.
2855 If the line comment character is the hash sign (@samp{#}) then it still has the
2856 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2857 to specify logical line numbers:
2860 @cindex lines starting with @code{#}
2861 @cindex logical line numbers
2862 To be compatible with past assemblers, lines that begin with @samp{#} have a
2863 special interpretation. Following the @samp{#} should be an absolute
2864 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2865 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2866 new logical file name. The rest of the line, if any, should be whitespace.
2868 If the first non-whitespace characters on the line are not numeric,
2869 the line is ignored. (Just like a comment.)
2872 # This is an ordinary comment.
2873 # 42-6 "new_file_name" # New logical file name
2874 # This is logical line # 36.
2876 This feature is deprecated, and may disappear from future versions
2877 of @command{@value{AS}}.
2882 @cindex characters used in symbols
2883 @ifclear SPECIAL-SYMS
2884 A @dfn{symbol} is one or more characters chosen from the set of all
2885 letters (both upper and lower case), digits and the three characters
2891 A @dfn{symbol} is one or more characters chosen from the set of all
2892 letters (both upper and lower case), digits and the three characters
2893 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2899 On most machines, you can also use @code{$} in symbol names; exceptions
2900 are noted in @ref{Machine Dependencies}.
2902 No symbol may begin with a digit. Case is significant.
2903 There is no length limit; all characters are significant. Multibyte characters
2904 are supported. Symbols are delimited by characters not in that set, or by the
2905 beginning of a file (since the source program must end with a newline, the end
2906 of a file is not a possible symbol delimiter). @xref{Symbols}.
2908 Symbol names may also be enclosed in double quote @code{"} characters. In such
2909 cases any characters are allowed, except for the NUL character. If a double
2910 quote character is to be included in the symbol name it must be preceeded by a
2911 backslash @code{\} character.
2912 @cindex length of symbols
2917 @cindex statements, structure of
2918 @cindex line separator character
2919 @cindex statement separator character
2921 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2922 @dfn{line separator character}. The line separator character is target
2923 specific and described in the @emph{Syntax} section of each
2924 target's documentation. Not all targets support a line separator character.
2925 The newline or line separator character is considered to be part of the
2926 preceding statement. Newlines and separators within character constants are an
2927 exception: they do not end statements.
2929 @cindex newline, required at file end
2930 @cindex EOF, newline must precede
2931 It is an error to end any statement with end-of-file: the last
2932 character of any input file should be a newline.@refill
2934 An empty statement is allowed, and may include whitespace. It is ignored.
2936 @cindex instructions and directives
2937 @cindex directives and instructions
2938 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2939 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2941 A statement begins with zero or more labels, optionally followed by a
2942 key symbol which determines what kind of statement it is. The key
2943 symbol determines the syntax of the rest of the statement. If the
2944 symbol begins with a dot @samp{.} then the statement is an assembler
2945 directive: typically valid for any computer. If the symbol begins with
2946 a letter the statement is an assembly language @dfn{instruction}: it
2947 assembles into a machine language instruction.
2949 Different versions of @command{@value{AS}} for different computers
2950 recognize different instructions. In fact, the same symbol may
2951 represent a different instruction in a different computer's assembly
2955 @cindex @code{:} (label)
2956 @cindex label (@code{:})
2957 A label is a symbol immediately followed by a colon (@code{:}).
2958 Whitespace before a label or after a colon is permitted, but you may not
2959 have whitespace between a label's symbol and its colon. @xref{Labels}.
2962 For HPPA targets, labels need not be immediately followed by a colon, but
2963 the definition of a label must begin in column zero. This also implies that
2964 only one label may be defined on each line.
2968 label: .directive followed by something
2969 another_label: # This is an empty statement.
2970 instruction operand_1, operand_2, @dots{}
2977 A constant is a number, written so that its value is known by
2978 inspection, without knowing any context. Like this:
2981 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2982 .ascii "Ring the bell\7" # A string constant.
2983 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2984 .float 0f-314159265358979323846264338327\
2985 95028841971.693993751E-40 # - pi, a flonum.
2990 * Characters:: Character Constants
2991 * Numbers:: Number Constants
2995 @subsection Character Constants
2997 @cindex character constants
2998 @cindex constants, character
2999 There are two kinds of character constants. A @dfn{character} stands
3000 for one character in one byte and its value may be used in
3001 numeric expressions. String constants (properly called string
3002 @emph{literals}) are potentially many bytes and their values may not be
3003 used in arithmetic expressions.
3007 * Chars:: Characters
3011 @subsubsection Strings
3013 @cindex string constants
3014 @cindex constants, string
3015 A @dfn{string} is written between double-quotes. It may contain
3016 double-quotes or null characters. The way to get special characters
3017 into a string is to @dfn{escape} these characters: precede them with
3018 a backslash @samp{\} character. For example @samp{\\} represents
3019 one backslash: the first @code{\} is an escape which tells
3020 @command{@value{AS}} to interpret the second character literally as a backslash
3021 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
3022 escape character). The complete list of escapes follows.
3024 @cindex escape codes, character
3025 @cindex character escape codes
3026 @c NOTE: Cindex entries must not start with a backlash character.
3027 @c NOTE: This confuses the pdf2texi script when it is creating the
3028 @c NOTE: index based upon the first character and so it generates:
3029 @c NOTE: \initial {\\}
3030 @c NOTE: which then results in the error message:
3031 @c NOTE: Argument of \\ has an extra }.
3032 @c NOTE: So in the index entries below a space character has been
3033 @c NOTE: prepended to avoid this problem.
3036 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
3038 @cindex @code{ \b} (backspace character)
3039 @cindex backspace (@code{\b})
3041 Mnemonic for backspace; for ASCII this is octal code 010.
3044 @c Mnemonic for EOText; for ASCII this is octal code 004.
3046 @cindex @code{ \f} (formfeed character)
3047 @cindex formfeed (@code{\f})
3049 Mnemonic for FormFeed; for ASCII this is octal code 014.
3051 @cindex @code{ \n} (newline character)
3052 @cindex newline (@code{\n})
3054 Mnemonic for newline; for ASCII this is octal code 012.
3057 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
3059 @cindex @code{ \r} (carriage return character)
3060 @cindex carriage return (@code{backslash-r})
3062 Mnemonic for carriage-Return; for ASCII this is octal code 015.
3065 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
3066 @c other assemblers.
3068 @cindex @code{ \t} (tab)
3069 @cindex tab (@code{\t})
3071 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
3074 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
3075 @c @item \x @var{digit} @var{digit} @var{digit}
3076 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
3078 @cindex @code{ \@var{ddd}} (octal character code)
3079 @cindex octal character code (@code{\@var{ddd}})
3080 @item \ @var{digit} @var{digit} @var{digit}
3081 An octal character code. The numeric code is 3 octal digits.
3082 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
3083 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3085 @cindex @code{ \@var{xd...}} (hex character code)
3086 @cindex hex character code (@code{\@var{xd...}})
3087 @item \@code{x} @var{hex-digits...}
3088 A hex character code. All trailing hex digits are combined. Either upper or
3089 lower case @code{x} works.
3091 @cindex @code{ \\} (@samp{\} character)
3092 @cindex backslash (@code{\\})
3094 Represents one @samp{\} character.
3097 @c Represents one @samp{'} (accent acute) character.
3098 @c This is needed in single character literals
3099 @c (@xref{Characters,,Character Constants}.) to represent
3102 @cindex @code{ \"} (doublequote character)
3103 @cindex doublequote (@code{\"})
3105 Represents one @samp{"} character. Needed in strings to represent
3106 this character, because an unescaped @samp{"} would end the string.
3108 @item \ @var{anything-else}
3109 Any other character when escaped by @kbd{\} gives a warning, but
3110 assembles as if the @samp{\} was not present. The idea is that if
3111 you used an escape sequence you clearly didn't want the literal
3112 interpretation of the following character. However @command{@value{AS}} has no
3113 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3114 code and warns you of the fact.
3117 Which characters are escapable, and what those escapes represent,
3118 varies widely among assemblers. The current set is what we think
3119 the BSD 4.2 assembler recognizes, and is a subset of what most C
3120 compilers recognize. If you are in doubt, do not use an escape
3124 @subsubsection Characters
3126 @cindex single character constant
3127 @cindex character, single
3128 @cindex constant, single character
3129 A single character may be written as a single quote immediately followed by
3130 that character. Some backslash escapes apply to characters, @code{\b},
3131 @code{\f}, @code{\n}, @code{\r}, @code{\t}, and @code{\"} with the same meaning
3132 as for strings, plus @code{\'} for a single quote. So if you want to write the
3133 character backslash, you must write @kbd{'\\} where the first @code{\} escapes
3134 the second @code{\}. As you can see, the quote is an acute accent, not a grave
3137 @ifclear abnormal-separator
3138 (or semicolon @samp{;})
3140 @ifset abnormal-separator
3142 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3147 immediately following an acute accent is taken as a literal character
3148 and does not count as the end of a statement. The value of a character
3149 constant in a numeric expression is the machine's byte-wide code for
3150 that character. @command{@value{AS}} assumes your character code is ASCII:
3151 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3154 @subsection Number Constants
3156 @cindex constants, number
3157 @cindex number constants
3158 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3159 are stored in the target machine. @emph{Integers} are numbers that
3160 would fit into an @code{int} in the C language. @emph{Bignums} are
3161 integers, but they are stored in more than 32 bits. @emph{Flonums}
3162 are floating point numbers, described below.
3165 * Integers:: Integers
3173 @subsubsection Integers
3175 @cindex constants, integer
3177 @cindex binary integers
3178 @cindex integers, binary
3179 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3180 the binary digits @samp{01}.
3182 @cindex octal integers
3183 @cindex integers, octal
3184 An octal integer is @samp{0} followed by zero or more of the octal
3185 digits (@samp{01234567}).
3187 @cindex decimal integers
3188 @cindex integers, decimal
3189 A decimal integer starts with a non-zero digit followed by zero or
3190 more digits (@samp{0123456789}).
3192 @cindex hexadecimal integers
3193 @cindex integers, hexadecimal
3194 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3195 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3197 Integers have the usual values. To denote a negative integer, use
3198 the prefix operator @samp{-} discussed under expressions
3199 (@pxref{Prefix Ops,,Prefix Operators}).
3202 @subsubsection Bignums
3205 @cindex constants, bignum
3206 A @dfn{bignum} has the same syntax and semantics as an integer
3207 except that the number (or its negative) takes more than 32 bits to
3208 represent in binary. The distinction is made because in some places
3209 integers are permitted while bignums are not.
3212 @subsubsection Flonums
3214 @cindex floating point numbers
3215 @cindex constants, floating point
3217 @cindex precision, floating point
3218 A @dfn{flonum} represents a floating point number. The translation is
3219 indirect: a decimal floating point number from the text is converted by
3220 @command{@value{AS}} to a generic binary floating point number of more than
3221 sufficient precision. This generic floating point number is converted
3222 to a particular computer's floating point format (or formats) by a
3223 portion of @command{@value{AS}} specialized to that computer.
3225 A flonum is written by writing (in order)
3230 (@samp{0} is optional on the HPPA.)
3234 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3236 @kbd{e} is recommended. Case is not important.
3238 @c FIXME: verify if flonum syntax really this vague for most cases
3239 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3240 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3243 On the H8/300 and Renesas / SuperH SH architectures, the letter must be
3244 one of the letters @samp{DFPRSX} (in upper or lower case).
3246 On the ARC, the letter must be one of the letters @samp{DFRS}
3247 (in upper or lower case).
3249 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3253 One of the letters @samp{DFRS} (in upper or lower case).
3256 One of the letters @samp{DFPRSX} (in upper or lower case).
3259 The letter @samp{E} (upper case only).
3264 An optional sign: either @samp{+} or @samp{-}.
3267 An optional @dfn{integer part}: zero or more decimal digits.
3270 An optional @dfn{fractional part}: @samp{.} followed by zero
3271 or more decimal digits.
3274 An optional exponent, consisting of:
3278 An @samp{E} or @samp{e}.
3279 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3280 @c principle this can perfectly well be different on different targets.
3282 Optional sign: either @samp{+} or @samp{-}.
3284 One or more decimal digits.
3289 At least one of the integer part or the fractional part must be
3290 present. The floating point number has the usual base-10 value.
3292 @command{@value{AS}} does all processing using integers. Flonums are computed
3293 independently of any floating point hardware in the computer running
3294 @command{@value{AS}}.
3297 @chapter Sections and Relocation
3302 * Secs Background:: Background
3303 * Ld Sections:: Linker Sections
3304 * As Sections:: Assembler Internal Sections
3305 * Sub-Sections:: Sub-Sections
3309 @node Secs Background
3312 Roughly, a section is a range of addresses, with no gaps; all data
3313 ``in'' those addresses is treated the same for some particular purpose.
3314 For example there may be a ``read only'' section.
3316 @cindex linker, and assembler
3317 @cindex assembler, and linker
3318 The linker @code{@value{LD}} reads many object files (partial programs) and
3319 combines their contents to form a runnable program. When @command{@value{AS}}
3320 emits an object file, the partial program is assumed to start at address 0.
3321 @code{@value{LD}} assigns the final addresses for the partial program, so that
3322 different partial programs do not overlap. This is actually an
3323 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3326 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3327 addresses. These blocks slide to their run-time addresses as rigid
3328 units; their length does not change and neither does the order of bytes
3329 within them. Such a rigid unit is called a @emph{section}. Assigning
3330 run-time addresses to sections is called @dfn{relocation}. It includes
3331 the task of adjusting mentions of object-file addresses so they refer to
3332 the proper run-time addresses.
3334 For the H8/300, and for the Renesas / SuperH SH,
3335 @command{@value{AS}} pads sections if needed to
3336 ensure they end on a word (sixteen bit) boundary.
3339 @cindex standard assembler sections
3340 An object file written by @command{@value{AS}} has at least three sections, any
3341 of which may be empty. These are named @dfn{text}, @dfn{data} and
3346 When it generates COFF or ELF output,
3348 @command{@value{AS}} can also generate whatever other named sections you specify
3349 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3350 If you do not use any directives that place output in the @samp{.text}
3351 or @samp{.data} sections, these sections still exist, but are empty.
3356 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3358 @command{@value{AS}} can also generate whatever other named sections you
3359 specify using the @samp{.space} and @samp{.subspace} directives. See
3360 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3361 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3362 assembler directives.
3365 Additionally, @command{@value{AS}} uses different names for the standard
3366 text, data, and bss sections when generating SOM output. Program text
3367 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3368 BSS into @samp{$BSS$}.
3372 Within the object file, the text section starts at address @code{0}, the
3373 data section follows, and the bss section follows the data section.
3376 When generating either SOM or ELF output files on the HPPA, the text
3377 section starts at address @code{0}, the data section at address
3378 @code{0x4000000}, and the bss section follows the data section.
3381 To let @code{@value{LD}} know which data changes when the sections are
3382 relocated, and how to change that data, @command{@value{AS}} also writes to the
3383 object file details of the relocation needed. To perform relocation
3384 @code{@value{LD}} must know, each time an address in the object
3388 Where in the object file is the beginning of this reference to
3391 How long (in bytes) is this reference?
3393 Which section does the address refer to? What is the numeric value of
3395 (@var{address}) @minus{} (@var{start-address of section})?
3398 Is the reference to an address ``Program-Counter relative''?
3401 @cindex addresses, format of
3402 @cindex section-relative addressing
3403 In fact, every address @command{@value{AS}} ever uses is expressed as
3405 (@var{section}) + (@var{offset into section})
3408 Further, most expressions @command{@value{AS}} computes have this section-relative
3411 (For some object formats, such as SOM for the HPPA, some expressions are
3412 symbol-relative instead.)
3415 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3416 @var{N} into section @var{secname}.''
3418 Apart from text, data and bss sections you need to know about the
3419 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3420 addresses in the absolute section remain unchanged. For example, address
3421 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3422 @code{@value{LD}}. Although the linker never arranges two partial programs'
3423 data sections with overlapping addresses after linking, @emph{by definition}
3424 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3425 part of a program is always the same address when the program is running as
3426 address @code{@{absolute@ 239@}} in any other part of the program.
3428 The idea of sections is extended to the @dfn{undefined} section. Any
3429 address whose section is unknown at assembly time is by definition
3430 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3431 Since numbers are always defined, the only way to generate an undefined
3432 address is to mention an undefined symbol. A reference to a named
3433 common block would be such a symbol: its value is unknown at assembly
3434 time so it has section @emph{undefined}.
3436 By analogy the word @emph{section} is used to describe groups of sections in
3437 the linked program. @code{@value{LD}} puts all partial programs' text
3438 sections in contiguous addresses in the linked program. It is
3439 customary to refer to the @emph{text section} of a program, meaning all
3440 the addresses of all partial programs' text sections. Likewise for
3441 data and bss sections.
3443 Some sections are manipulated by @code{@value{LD}}; others are invented for
3444 use of @command{@value{AS}} and have no meaning except during assembly.
3447 @section Linker Sections
3448 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3453 @cindex named sections
3454 @cindex sections, named
3455 @item named sections
3458 @cindex text section
3459 @cindex data section
3463 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3464 separate but equal sections. Anything you can say of one section is
3467 When the program is running, however, it is
3468 customary for the text section to be unalterable. The
3469 text section is often shared among processes: it contains
3470 instructions, constants and the like. The data section of a running
3471 program is usually alterable: for example, C variables would be stored
3472 in the data section.
3477 This section contains zeroed bytes when your program begins running. It
3478 is used to hold uninitialized variables or common storage. The length of
3479 each partial program's bss section is important, but because it starts
3480 out containing zeroed bytes there is no need to store explicit zero
3481 bytes in the object file. The bss section was invented to eliminate
3482 those explicit zeros from object files.
3484 @cindex absolute section
3485 @item absolute section
3486 Address 0 of this section is always ``relocated'' to runtime address 0.
3487 This is useful if you want to refer to an address that @code{@value{LD}} must
3488 not change when relocating. In this sense we speak of absolute
3489 addresses being ``unrelocatable'': they do not change during relocation.
3491 @cindex undefined section
3492 @item undefined section
3493 This ``section'' is a catch-all for address references to objects not in
3494 the preceding sections.
3495 @c FIXME: ref to some other doc on obj-file formats could go here.
3498 @cindex relocation example
3499 An idealized example of three relocatable sections follows.
3501 The example uses the traditional section names @samp{.text} and @samp{.data}.
3503 Memory addresses are on the horizontal axis.
3507 @c END TEXI2ROFF-KILL
3510 partial program # 1: |ttttt|dddd|00|
3517 partial program # 2: |TTT|DDD|000|
3520 +--+---+-----+--+----+---+-----+~~
3521 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3522 +--+---+-----+--+----+---+-----+~~
3524 addresses: 0 @dots{}
3531 \line{\it Partial program \#1: \hfil}
3532 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3533 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3535 \line{\it Partial program \#2: \hfil}
3536 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3537 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3539 \line{\it linked program: \hfil}
3540 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3541 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3542 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3543 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3545 \line{\it addresses: \hfil}
3549 @c END TEXI2ROFF-KILL
3552 @section Assembler Internal Sections
3554 @cindex internal assembler sections
3555 @cindex sections in messages, internal
3556 These sections are meant only for the internal use of @command{@value{AS}}. They
3557 have no meaning at run-time. You do not really need to know about these
3558 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3559 warning messages, so it might be helpful to have an idea of their
3560 meanings to @command{@value{AS}}. These sections are used to permit the
3561 value of every expression in your assembly language program to be a
3562 section-relative address.
3565 @cindex assembler internal logic error
3566 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3567 An internal assembler logic error has been found. This means there is a
3568 bug in the assembler.
3570 @cindex expr (internal section)
3572 The assembler stores complex expression internally as combinations of
3573 symbols. When it needs to represent an expression as a symbol, it puts
3574 it in the expr section.
3576 @c FIXME item transfer[t] vector preload
3577 @c FIXME item transfer[t] vector postload
3578 @c FIXME item register
3582 @section Sub-Sections
3584 @cindex numbered subsections
3585 @cindex grouping data
3591 fall into two sections: text and data.
3593 You may have separate groups of
3595 data in named sections
3599 data in named sections
3605 that you want to end up near to each other in the object file, even though they
3606 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3607 use @dfn{subsections} for this purpose. Within each section, there can be
3608 numbered subsections with values from 0 to 8192. Objects assembled into the
3609 same subsection go into the object file together with other objects in the same
3610 subsection. For example, a compiler might want to store constants in the text
3611 section, but might not want to have them interspersed with the program being
3612 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3613 section of code being output, and a @samp{.text 1} before each group of
3614 constants being output.
3616 Subsections are optional. If you do not use subsections, everything
3617 goes in subsection number zero.
3620 Each subsection is zero-padded up to a multiple of four bytes.
3621 (Subsections may be padded a different amount on different flavors
3622 of @command{@value{AS}}.)
3626 On the H8/300 platform, each subsection is zero-padded to a word
3627 boundary (two bytes).
3628 The same is true on the Renesas SH.
3632 Subsections appear in your object file in numeric order, lowest numbered
3633 to highest. (All this to be compatible with other people's assemblers.)
3634 The object file contains no representation of subsections; @code{@value{LD}} and
3635 other programs that manipulate object files see no trace of them.
3636 They just see all your text subsections as a text section, and all your
3637 data subsections as a data section.
3639 To specify which subsection you want subsequent statements assembled
3640 into, use a numeric argument to specify it, in a @samp{.text
3641 @var{expression}} or a @samp{.data @var{expression}} statement.
3644 When generating COFF output, you
3649 can also use an extra subsection
3650 argument with arbitrary named sections: @samp{.section @var{name},
3655 When generating ELF output, you
3660 can also use the @code{.subsection} directive (@pxref{SubSection})
3661 to specify a subsection: @samp{.subsection @var{expression}}.
3663 @var{Expression} should be an absolute expression
3664 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3665 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3666 begins in @code{text 0}. For instance:
3668 .text 0 # The default subsection is text 0 anyway.
3669 .ascii "This lives in the first text subsection. *"
3671 .ascii "But this lives in the second text subsection."
3673 .ascii "This lives in the data section,"
3674 .ascii "in the first data subsection."
3676 .ascii "This lives in the first text section,"
3677 .ascii "immediately following the asterisk (*)."
3680 Each section has a @dfn{location counter} incremented by one for every byte
3681 assembled into that section. Because subsections are merely a convenience
3682 restricted to @command{@value{AS}} there is no concept of a subsection location
3683 counter. There is no way to directly manipulate a location counter---but the
3684 @code{.align} directive changes it, and any label definition captures its
3685 current value. The location counter of the section where statements are being
3686 assembled is said to be the @dfn{active} location counter.
3689 @section bss Section
3692 @cindex common variable storage
3693 The bss section is used for local common variable storage.
3694 You may allocate address space in the bss section, but you may
3695 not dictate data to load into it before your program executes. When
3696 your program starts running, all the contents of the bss
3697 section are zeroed bytes.
3699 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3700 @ref{Lcomm,,@code{.lcomm}}.
3702 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3703 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3706 When assembling for a target which supports multiple sections, such as ELF or
3707 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3708 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3709 section. Typically the section will only contain symbol definitions and
3710 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3717 Symbols are a central concept: the programmer uses symbols to name
3718 things, the linker uses symbols to link, and the debugger uses symbols
3722 @cindex debuggers, and symbol order
3723 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3724 the same order they were declared. This may break some debuggers.
3729 * Setting Symbols:: Giving Symbols Other Values
3730 * Symbol Names:: Symbol Names
3731 * Dot:: The Special Dot Symbol
3732 * Symbol Attributes:: Symbol Attributes
3739 A @dfn{label} is written as a symbol immediately followed by a colon
3740 @samp{:}. The symbol then represents the current value of the
3741 active location counter, and is, for example, a suitable instruction
3742 operand. You are warned if you use the same symbol to represent two
3743 different locations: the first definition overrides any other
3747 On the HPPA, the usual form for a label need not be immediately followed by a
3748 colon, but instead must start in column zero. Only one label may be defined on
3749 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3750 provides a special directive @code{.label} for defining labels more flexibly.
3753 @node Setting Symbols
3754 @section Giving Symbols Other Values
3756 @cindex assigning values to symbols
3757 @cindex symbol values, assigning
3758 A symbol can be given an arbitrary value by writing a symbol, followed
3759 by an equals sign @samp{=}, followed by an expression
3760 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3761 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3762 equals sign @samp{=}@samp{=} here represents an equivalent of the
3763 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3766 Blackfin does not support symbol assignment with @samp{=}.
3770 @section Symbol Names
3772 @cindex symbol names
3773 @cindex names, symbol
3774 @ifclear SPECIAL-SYMS
3775 Symbol names begin with a letter or with one of @samp{._}. On most
3776 machines, you can also use @code{$} in symbol names; exceptions are
3777 noted in @ref{Machine Dependencies}. That character may be followed by any
3778 string of digits, letters, dollar signs (unless otherwise noted for a
3779 particular target machine), and underscores.
3783 Symbol names begin with a letter or with one of @samp{._}. On the
3784 Renesas SH you can also use @code{$} in symbol names. That
3785 character may be followed by any string of digits, letters, dollar signs (save
3786 on the H8/300), and underscores.
3790 Case of letters is significant: @code{foo} is a different symbol name
3793 Symbol names do not start with a digit. An exception to this rule is made for
3794 Local Labels. See below.
3796 Multibyte characters are supported. To generate a symbol name containing
3797 multibyte characters enclose it within double quotes and use escape codes. cf
3798 @xref{Strings}. Generating a multibyte symbol name from a label is not
3799 currently supported.
3801 Each symbol has exactly one name. Each name in an assembly language program
3802 refers to exactly one symbol. You may use that symbol name any number of times
3805 @subheading Local Symbol Names
3807 @cindex local symbol names
3808 @cindex symbol names, local
3809 A local symbol is any symbol beginning with certain local label prefixes.
3810 By default, the local label prefix is @samp{.L} for ELF systems or
3811 @samp{L} for traditional a.out systems, but each target may have its own
3812 set of local label prefixes.
3814 On the HPPA local symbols begin with @samp{L$}.
3817 Local symbols are defined and used within the assembler, but they are
3818 normally not saved in object files. Thus, they are not visible when debugging.
3819 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3820 to retain the local symbols in the object files.
3822 @subheading Local Labels
3824 @cindex local labels
3825 @cindex temporary symbol names
3826 @cindex symbol names, temporary
3827 Local labels are different from local symbols. Local labels help compilers and
3828 programmers use names temporarily. They create symbols which are guaranteed to
3829 be unique over the entire scope of the input source code and which can be
3830 referred to by a simple notation. To define a local label, write a label of
3831 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3832 To refer to the most recent previous definition of that label write
3833 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3834 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3835 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3837 There is no restriction on how you can use these labels, and you can reuse them
3838 too. So that it is possible to repeatedly define the same local label (using
3839 the same number @samp{@b{N}}), although you can only refer to the most recently
3840 defined local label of that number (for a backwards reference) or the next
3841 definition of a specific local label for a forward reference. It is also worth
3842 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3843 implemented in a slightly more efficient manner than the others.
3854 Which is the equivalent of:
3857 label_1: branch label_3
3858 label_2: branch label_1
3859 label_3: branch label_4
3860 label_4: branch label_3
3863 Local label names are only a notational device. They are immediately
3864 transformed into more conventional symbol names before the assembler uses them.
3865 The symbol names are stored in the symbol table, appear in error messages, and
3866 are optionally emitted to the object file. The names are constructed using
3870 @item @emph{local label prefix}
3871 All local symbols begin with the system-specific local label prefix.
3872 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3873 that start with the local label prefix. These labels are
3874 used for symbols you are never intended to see. If you use the
3875 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3876 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3877 you may use them in debugging.
3880 This is the number that was used in the local label definition. So if the
3881 label is written @samp{55:} then the number is @samp{55}.
3884 This unusual character is included so you do not accidentally invent a symbol
3885 of the same name. The character has ASCII value of @samp{\002} (control-B).
3887 @item @emph{ordinal number}
3888 This is a serial number to keep the labels distinct. The first definition of
3889 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3890 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3891 the number @samp{1} and its 15th definition gets @samp{15} as well.
3894 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3895 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3897 @subheading Dollar Local Labels
3898 @cindex dollar local symbols
3900 On some targets @code{@value{AS}} also supports an even more local form of
3901 local labels called dollar labels. These labels go out of scope (i.e., they
3902 become undefined) as soon as a non-local label is defined. Thus they remain
3903 valid for only a small region of the input source code. Normal local labels,
3904 by contrast, remain in scope for the entire file, or until they are redefined
3905 by another occurrence of the same local label.
3907 Dollar labels are defined in exactly the same way as ordinary local labels,
3908 except that they have a dollar sign suffix to their numeric value, e.g.,
3911 They can also be distinguished from ordinary local labels by their transformed
3912 names which use ASCII character @samp{\001} (control-A) as the magic character
3913 to distinguish them from ordinary labels. For example, the fifth definition of
3914 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3917 @section The Special Dot Symbol
3919 @cindex dot (symbol)
3920 @cindex @code{.} (symbol)
3921 @cindex current address
3922 @cindex location counter
3923 The special symbol @samp{.} refers to the current address that
3924 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3925 .long .} defines @code{melvin} to contain its own address.
3926 Assigning a value to @code{.} is treated the same as a @code{.org}
3928 @ifclear no-space-dir
3929 Thus, the expression @samp{.=.+4} is the same as saying
3933 @node Symbol Attributes
3934 @section Symbol Attributes
3936 @cindex symbol attributes
3937 @cindex attributes, symbol
3938 Every symbol has, as well as its name, the attributes ``Value'' and
3939 ``Type''. Depending on output format, symbols can also have auxiliary
3942 The detailed definitions are in @file{a.out.h}.
3945 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3946 all these attributes, and probably won't warn you. This makes the
3947 symbol an externally defined symbol, which is generally what you
3951 * Symbol Value:: Value
3952 * Symbol Type:: Type
3954 * a.out Symbols:: Symbol Attributes: @code{a.out}
3957 * COFF Symbols:: Symbol Attributes for COFF
3960 * SOM Symbols:: Symbol Attributes for SOM
3967 @cindex value of a symbol
3968 @cindex symbol value
3969 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3970 location in the text, data, bss or absolute sections the value is the
3971 number of addresses from the start of that section to the label.
3972 Naturally for text, data and bss sections the value of a symbol changes
3973 as @code{@value{LD}} changes section base addresses during linking. Absolute
3974 symbols' values do not change during linking: that is why they are
3977 The value of an undefined symbol is treated in a special way. If it is
3978 0 then the symbol is not defined in this assembler source file, and
3979 @code{@value{LD}} tries to determine its value from other files linked into the
3980 same program. You make this kind of symbol simply by mentioning a symbol
3981 name without defining it. A non-zero value represents a @code{.comm}
3982 common declaration. The value is how much common storage to reserve, in
3983 bytes (addresses). The symbol refers to the first address of the
3989 @cindex type of a symbol
3991 The type attribute of a symbol contains relocation (section)
3992 information, any flag settings indicating that a symbol is external, and
3993 (optionally), other information for linkers and debuggers. The exact
3994 format depends on the object-code output format in use.
3998 @subsection Symbol Attributes: @code{a.out}
4000 @cindex @code{a.out} symbol attributes
4001 @cindex symbol attributes, @code{a.out}
4004 * Symbol Desc:: Descriptor
4005 * Symbol Other:: Other
4009 @subsubsection Descriptor
4011 @cindex descriptor, of @code{a.out} symbol
4012 This is an arbitrary 16-bit value. You may establish a symbol's
4013 descriptor value by using a @code{.desc} statement
4014 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
4015 @command{@value{AS}}.
4018 @subsubsection Other
4020 @cindex other attribute, of @code{a.out} symbol
4021 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
4026 @subsection Symbol Attributes for COFF
4028 @cindex COFF symbol attributes
4029 @cindex symbol attributes, COFF
4031 The COFF format supports a multitude of auxiliary symbol attributes;
4032 like the primary symbol attributes, they are set between @code{.def} and
4033 @code{.endef} directives.
4035 @subsubsection Primary Attributes
4037 @cindex primary attributes, COFF symbols
4038 The symbol name is set with @code{.def}; the value and type,
4039 respectively, with @code{.val} and @code{.type}.
4041 @subsubsection Auxiliary Attributes
4043 @cindex auxiliary attributes, COFF symbols
4044 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
4045 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
4046 table information for COFF.
4051 @subsection Symbol Attributes for SOM
4053 @cindex SOM symbol attributes
4054 @cindex symbol attributes, SOM
4056 The SOM format for the HPPA supports a multitude of symbol attributes set with
4057 the @code{.EXPORT} and @code{.IMPORT} directives.
4059 The attributes are described in @cite{HP9000 Series 800 Assembly
4060 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
4061 @code{EXPORT} assembler directive documentation.
4065 @chapter Expressions
4069 @cindex numeric values
4070 An @dfn{expression} specifies an address or numeric value.
4071 Whitespace may precede and/or follow an expression.
4073 The result of an expression must be an absolute number, or else an offset into
4074 a particular section. If an expression is not absolute, and there is not
4075 enough information when @command{@value{AS}} sees the expression to know its
4076 section, a second pass over the source program might be necessary to interpret
4077 the expression---but the second pass is currently not implemented.
4078 @command{@value{AS}} aborts with an error message in this situation.
4081 * Empty Exprs:: Empty Expressions
4082 * Integer Exprs:: Integer Expressions
4086 @section Empty Expressions
4088 @cindex empty expressions
4089 @cindex expressions, empty
4090 An empty expression has no value: it is just whitespace or null.
4091 Wherever an absolute expression is required, you may omit the
4092 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4093 is compatible with other assemblers.
4096 @section Integer Expressions
4098 @cindex integer expressions
4099 @cindex expressions, integer
4100 An @dfn{integer expression} is one or more @emph{arguments} delimited
4101 by @emph{operators}.
4104 * Arguments:: Arguments
4105 * Operators:: Operators
4106 * Prefix Ops:: Prefix Operators
4107 * Infix Ops:: Infix Operators
4111 @subsection Arguments
4113 @cindex expression arguments
4114 @cindex arguments in expressions
4115 @cindex operands in expressions
4116 @cindex arithmetic operands
4117 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4118 contexts arguments are sometimes called ``arithmetic operands''. In
4119 this manual, to avoid confusing them with the ``instruction operands'' of
4120 the machine language, we use the term ``argument'' to refer to parts of
4121 expressions only, reserving the word ``operand'' to refer only to machine
4122 instruction operands.
4124 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4125 @var{section} is one of text, data, bss, absolute,
4126 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4129 Numbers are usually integers.
4131 A number can be a flonum or bignum. In this case, you are warned
4132 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4133 these 32 bits are an integer. You may write integer-manipulating
4134 instructions that act on exotic constants, compatible with other
4137 @cindex subexpressions
4138 Subexpressions are a left parenthesis @samp{(} followed by an integer
4139 expression, followed by a right parenthesis @samp{)}; or a prefix
4140 operator followed by an argument.
4143 @subsection Operators
4145 @cindex operators, in expressions
4146 @cindex arithmetic functions
4147 @cindex functions, in expressions
4148 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4149 operators are followed by an argument. Infix operators appear
4150 between their arguments. Operators may be preceded and/or followed by
4154 @subsection Prefix Operator
4156 @cindex prefix operators
4157 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4158 one argument, which must be absolute.
4160 @c the tex/end tex stuff surrounding this small table is meant to make
4161 @c it align, on the printed page, with the similar table in the next
4162 @c section (which is inside an enumerate).
4164 \global\advance\leftskip by \itemindent
4169 @dfn{Negation}. Two's complement negation.
4171 @dfn{Complementation}. Bitwise not.
4175 \global\advance\leftskip by -\itemindent
4179 @subsection Infix Operators
4181 @cindex infix operators
4182 @cindex operators, permitted arguments
4183 @dfn{Infix operators} take two arguments, one on either side. Operators
4184 have precedence, but operations with equal precedence are performed left
4185 to right. Apart from @code{+} or @option{-}, both arguments must be
4186 absolute, and the result is absolute.
4189 @cindex operator precedence
4190 @cindex precedence of operators
4197 @dfn{Multiplication}.
4200 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4206 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4209 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4213 Intermediate precedence
4218 @dfn{Bitwise Inclusive Or}.
4224 @dfn{Bitwise Exclusive Or}.
4227 @dfn{Bitwise Or Not}.
4234 @cindex addition, permitted arguments
4235 @cindex plus, permitted arguments
4236 @cindex arguments for addition
4238 @dfn{Addition}. If either argument is absolute, the result has the section of
4239 the other argument. You may not add together arguments from different
4242 @cindex subtraction, permitted arguments
4243 @cindex minus, permitted arguments
4244 @cindex arguments for subtraction
4246 @dfn{Subtraction}. If the right argument is absolute, the
4247 result has the section of the left argument.
4248 If both arguments are in the same section, the result is absolute.
4249 You may not subtract arguments from different sections.
4250 @c FIXME is there still something useful to say about undefined - undefined ?
4252 @cindex comparison expressions
4253 @cindex expressions, comparison
4258 @dfn{Is Not Equal To}
4262 @dfn{Is Greater Than}
4264 @dfn{Is Greater Than Or Equal To}
4266 @dfn{Is Less Than Or Equal To}
4268 The comparison operators can be used as infix operators. A true results has a
4269 value of -1 whereas a false result has a value of 0. Note, these operators
4270 perform signed comparisons.
4273 @item Lowest Precedence
4282 These two logical operations can be used to combine the results of sub
4283 expressions. Note, unlike the comparison operators a true result returns a
4284 value of 1 but a false results does still return 0. Also note that the logical
4285 or operator has a slightly lower precedence than logical and.
4290 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4291 address; you can only have a defined section in one of the two arguments.
4294 @chapter Assembler Directives
4296 @cindex directives, machine independent
4297 @cindex pseudo-ops, machine independent
4298 @cindex machine independent directives
4299 All assembler directives have names that begin with a period (@samp{.}).
4300 The names are case insensitive for most targets, and usually written
4303 This chapter discusses directives that are available regardless of the
4304 target machine configuration for the @sc{gnu} assembler.
4306 Some machine configurations provide additional directives.
4307 @xref{Machine Dependencies}.
4310 @ifset machine-directives
4311 @xref{Machine Dependencies}, for additional directives.
4316 * Abort:: @code{.abort}
4318 * ABORT (COFF):: @code{.ABORT}
4321 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4322 * Altmacro:: @code{.altmacro}
4323 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4324 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4325 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4326 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4327 * Byte:: @code{.byte @var{expressions}}
4328 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4329 * Comm:: @code{.comm @var{symbol} , @var{length} }
4330 * Data:: @code{.data @var{subsection}}
4331 * Dc:: @code{.dc[@var{size}] @var{expressions}}
4332 * Dcb:: @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4333 * Ds:: @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4335 * Def:: @code{.def @var{name}}
4338 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4344 * Double:: @code{.double @var{flonums}}
4345 * Eject:: @code{.eject}
4346 * Else:: @code{.else}
4347 * Elseif:: @code{.elseif}
4350 * Endef:: @code{.endef}
4353 * Endfunc:: @code{.endfunc}
4354 * Endif:: @code{.endif}
4355 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4356 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4357 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4359 * Error:: @code{.error @var{string}}
4360 * Exitm:: @code{.exitm}
4361 * Extern:: @code{.extern}
4362 * Fail:: @code{.fail}
4363 * File:: @code{.file}
4364 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4365 * Float:: @code{.float @var{flonums}}
4366 * Func:: @code{.func}
4367 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4369 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4370 * Hidden:: @code{.hidden @var{names}}
4373 * hword:: @code{.hword @var{expressions}}
4374 * Ident:: @code{.ident}
4375 * If:: @code{.if @var{absolute expression}}
4376 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4377 * Include:: @code{.include "@var{file}"}
4378 * Int:: @code{.int @var{expressions}}
4380 * Internal:: @code{.internal @var{names}}
4383 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4384 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4385 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4386 * Lflags:: @code{.lflags}
4387 @ifclear no-line-dir
4388 * Line:: @code{.line @var{line-number}}
4391 * Linkonce:: @code{.linkonce [@var{type}]}
4392 * List:: @code{.list}
4393 * Ln:: @code{.ln @var{line-number}}
4394 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4395 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4397 * Local:: @code{.local @var{names}}
4400 * Long:: @code{.long @var{expressions}}
4402 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4405 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4406 * MRI:: @code{.mri @var{val}}
4407 * Noaltmacro:: @code{.noaltmacro}
4408 * Nolist:: @code{.nolist}
4409 * Nops:: @code{.nops @var{size}[, @var{control}]}
4410 * Octa:: @code{.octa @var{bignums}}
4411 * Offset:: @code{.offset @var{loc}}
4412 * Org:: @code{.org @var{new-lc}, @var{fill}}
4413 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4415 * PopSection:: @code{.popsection}
4416 * Previous:: @code{.previous}
4419 * Print:: @code{.print @var{string}}
4421 * Protected:: @code{.protected @var{names}}
4424 * Psize:: @code{.psize @var{lines}, @var{columns}}
4425 * Purgem:: @code{.purgem @var{name}}
4427 * PushSection:: @code{.pushsection @var{name}}
4430 * Quad:: @code{.quad @var{bignums}}
4431 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4432 * Rept:: @code{.rept @var{count}}
4433 * Sbttl:: @code{.sbttl "@var{subheading}"}
4435 * Scl:: @code{.scl @var{class}}
4438 * Section:: @code{.section @var{name}[, @var{flags}]}
4441 * Set:: @code{.set @var{symbol}, @var{expression}}
4442 * Short:: @code{.short @var{expressions}}
4443 * Single:: @code{.single @var{flonums}}
4445 * Size:: @code{.size [@var{name} , @var{expression}]}
4447 @ifclear no-space-dir
4448 * Skip:: @code{.skip @var{size} [,@var{fill}]}
4451 * Sleb128:: @code{.sleb128 @var{expressions}}
4452 @ifclear no-space-dir
4453 * Space:: @code{.space @var{size} [,@var{fill}]}
4456 * Stab:: @code{.stabd, .stabn, .stabs}
4459 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4460 * Struct:: @code{.struct @var{expression}}
4462 * SubSection:: @code{.subsection}
4463 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4467 * Tag:: @code{.tag @var{structname}}
4470 * Text:: @code{.text @var{subsection}}
4471 * Title:: @code{.title "@var{heading}"}
4473 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4476 * Uleb128:: @code{.uleb128 @var{expressions}}
4478 * Val:: @code{.val @var{addr}}
4482 * Version:: @code{.version "@var{string}"}
4483 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4484 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4487 * Warning:: @code{.warning @var{string}}
4488 * Weak:: @code{.weak @var{names}}
4489 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4490 * Word:: @code{.word @var{expressions}}
4491 @ifclear no-space-dir
4492 * Zero:: @code{.zero @var{size}}
4495 * 2byte:: @code{.2byte @var{expressions}}
4496 * 4byte:: @code{.4byte @var{expressions}}
4497 * 8byte:: @code{.8byte @var{bignums}}
4499 * Deprecated:: Deprecated Directives
4503 @section @code{.abort}
4505 @cindex @code{abort} directive
4506 @cindex stopping the assembly
4507 This directive stops the assembly immediately. It is for
4508 compatibility with other assemblers. The original idea was that the
4509 assembly language source would be piped into the assembler. If the sender
4510 of the source quit, it could use this directive tells @command{@value{AS}} to
4511 quit also. One day @code{.abort} will not be supported.
4515 @section @code{.ABORT} (COFF)
4517 @cindex @code{ABORT} directive
4518 When producing COFF output, @command{@value{AS}} accepts this directive as a
4519 synonym for @samp{.abort}.
4524 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4526 @cindex padding the location counter
4527 @cindex @code{align} directive
4528 Pad the location counter (in the current subsection) to a particular storage
4529 boundary. The first expression (which must be absolute) is the alignment
4530 required, as described below.
4532 The second expression (also absolute) gives the fill value to be stored in the
4533 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4534 padding bytes are normally zero. However, on most systems, if the section is
4535 marked as containing code and the fill value is omitted, the space is filled
4536 with no-op instructions.
4538 The third expression is also absolute, and is also optional. If it is present,
4539 it is the maximum number of bytes that should be skipped by this alignment
4540 directive. If doing the alignment would require skipping more bytes than the
4541 specified maximum, then the alignment is not done at all. You can omit the
4542 fill value (the second argument) entirely by simply using two commas after the
4543 required alignment; this can be useful if you want the alignment to be filled
4544 with no-op instructions when appropriate.
4546 The way the required alignment is specified varies from system to system.
4547 For the arc, hppa, i386 using ELF, iq2000, m68k, or1k,
4548 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4549 alignment request in bytes. For example @samp{.align 8} advances
4550 the location counter until it is a multiple of 8. If the location counter
4551 is already a multiple of 8, no change is needed. For the tic54x, the
4552 first expression is the alignment request in words.
4554 For other systems, including ppc, i386 using a.out format, arm and
4555 strongarm, it is the
4556 number of low-order zero bits the location counter must have after
4557 advancement. For example @samp{.align 3} advances the location
4558 counter until it a multiple of 8. If the location counter is already a
4559 multiple of 8, no change is needed.
4561 This inconsistency is due to the different behaviors of the various
4562 native assemblers for these systems which GAS must emulate.
4563 GAS also provides @code{.balign} and @code{.p2align} directives,
4564 described later, which have a consistent behavior across all
4565 architectures (but are specific to GAS).
4568 @section @code{.altmacro}
4569 Enable alternate macro mode, enabling:
4572 @item LOCAL @var{name} [ , @dots{} ]
4573 One additional directive, @code{LOCAL}, is available. It is used to
4574 generate a string replacement for each of the @var{name} arguments, and
4575 replace any instances of @var{name} in each macro expansion. The
4576 replacement string is unique in the assembly, and different for each
4577 separate macro expansion. @code{LOCAL} allows you to write macros that
4578 define symbols, without fear of conflict between separate macro expansions.
4580 @item String delimiters
4581 You can write strings delimited in these other ways besides
4582 @code{"@var{string}"}:
4585 @item '@var{string}'
4586 You can delimit strings with single-quote characters.
4588 @item <@var{string}>
4589 You can delimit strings with matching angle brackets.
4592 @item single-character string escape
4593 To include any single character literally in a string (even if the
4594 character would otherwise have some special meaning), you can prefix the
4595 character with @samp{!} (an exclamation mark). For example, you can
4596 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4598 @item Expression results as strings
4599 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4600 and use the result as a string.
4604 @section @code{.ascii "@var{string}"}@dots{}
4606 @cindex @code{ascii} directive
4607 @cindex string literals
4608 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4609 separated by commas. It assembles each string (with no automatic
4610 trailing zero byte) into consecutive addresses.
4613 @section @code{.asciz "@var{string}"}@dots{}
4615 @cindex @code{asciz} directive
4616 @cindex zero-terminated strings
4617 @cindex null-terminated strings
4618 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4619 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4622 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4624 @cindex padding the location counter given number of bytes
4625 @cindex @code{balign} directive
4626 Pad the location counter (in the current subsection) to a particular
4627 storage boundary. The first expression (which must be absolute) is the
4628 alignment request in bytes. For example @samp{.balign 8} advances
4629 the location counter until it is a multiple of 8. If the location counter
4630 is already a multiple of 8, no change is needed.
4632 The second expression (also absolute) gives the fill value to be stored in the
4633 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4634 padding bytes are normally zero. However, on most systems, if the section is
4635 marked as containing code and the fill value is omitted, the space is filled
4636 with no-op instructions.
4638 The third expression is also absolute, and is also optional. If it is present,
4639 it is the maximum number of bytes that should be skipped by this alignment
4640 directive. If doing the alignment would require skipping more bytes than the
4641 specified maximum, then the alignment is not done at all. You can omit the
4642 fill value (the second argument) entirely by simply using two commas after the
4643 required alignment; this can be useful if you want the alignment to be filled
4644 with no-op instructions when appropriate.
4646 @cindex @code{balignw} directive
4647 @cindex @code{balignl} directive
4648 The @code{.balignw} and @code{.balignl} directives are variants of the
4649 @code{.balign} directive. The @code{.balignw} directive treats the fill
4650 pattern as a two byte word value. The @code{.balignl} directives treats the
4651 fill pattern as a four byte longword value. For example, @code{.balignw
4652 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4653 filled in with the value 0x368d (the exact placement of the bytes depends upon
4654 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4657 @node Bundle directives
4658 @section Bundle directives
4659 @subsection @code{.bundle_align_mode @var{abs-expr}}
4660 @cindex @code{bundle_align_mode} directive
4662 @cindex instruction bundle
4663 @cindex aligned instruction bundle
4664 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4665 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4666 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4667 disabled (which is the default state). If the argument it not zero, it
4668 gives the size of an instruction bundle as a power of two (as for the
4669 @code{.p2align} directive, @pxref{P2align}).
4671 For some targets, it's an ABI requirement that no instruction may span a
4672 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4673 instructions that starts on an aligned boundary. For example, if
4674 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4675 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4676 effect, no single instruction may span a boundary between bundles. If an
4677 instruction would start too close to the end of a bundle for the length of
4678 that particular instruction to fit within the bundle, then the space at the
4679 end of that bundle is filled with no-op instructions so the instruction
4680 starts in the next bundle. As a corollary, it's an error if any single
4681 instruction's encoding is longer than the bundle size.
4683 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4684 @cindex @code{bundle_lock} directive
4685 @cindex @code{bundle_unlock} directive
4686 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4687 allow explicit control over instruction bundle padding. These directives
4688 are only valid when @code{.bundle_align_mode} has been used to enable
4689 aligned instruction bundle mode. It's an error if they appear when
4690 @code{.bundle_align_mode} has not been used at all, or when the last
4691 directive was @w{@code{.bundle_align_mode 0}}.
4693 @cindex bundle-locked
4694 For some targets, it's an ABI requirement that certain instructions may
4695 appear only as part of specified permissible sequences of multiple
4696 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4697 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4698 instruction sequence. For purposes of aligned instruction bundle mode, a
4699 sequence starting with @code{.bundle_lock} and ending with
4700 @code{.bundle_unlock} is treated as a single instruction. That is, the
4701 entire sequence must fit into a single bundle and may not span a bundle
4702 boundary. If necessary, no-op instructions will be inserted before the
4703 first instruction of the sequence so that the whole sequence starts on an
4704 aligned bundle boundary. It's an error if the sequence is longer than the
4707 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4708 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4709 nested. That is, a second @code{.bundle_lock} directive before the next
4710 @code{.bundle_unlock} directive has no effect except that it must be
4711 matched by another closing @code{.bundle_unlock} so that there is the
4712 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4715 @section @code{.byte @var{expressions}}
4717 @cindex @code{byte} directive
4718 @cindex integers, one byte
4719 @code{.byte} expects zero or more expressions, separated by commas.
4720 Each expression is assembled into the next byte.
4722 @node CFI directives
4723 @section CFI directives
4724 @subsection @code{.cfi_sections @var{section_list}}
4725 @cindex @code{cfi_sections} directive
4726 @code{.cfi_sections} may be used to specify whether CFI directives
4727 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4728 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4729 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4730 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4731 directive is not used is @code{.cfi_sections .eh_frame}.
4733 On targets that support compact unwinding tables these can be generated
4734 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4736 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4737 which is used by the @value{TIC6X} target.
4739 The @code{.cfi_sections} directive can be repeated, with the same or different
4740 arguments, provided that CFI generation has not yet started. Once CFI
4741 generation has started however the section list is fixed and any attempts to
4742 redefine it will result in an error.
4744 @subsection @code{.cfi_startproc [simple]}
4745 @cindex @code{cfi_startproc} directive
4746 @code{.cfi_startproc} is used at the beginning of each function that
4747 should have an entry in @code{.eh_frame}. It initializes some internal
4748 data structures. Don't forget to close the function by
4749 @code{.cfi_endproc}.
4751 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4752 it also emits some architecture dependent initial CFI instructions.
4754 @subsection @code{.cfi_endproc}
4755 @cindex @code{cfi_endproc} directive
4756 @code{.cfi_endproc} is used at the end of a function where it closes its
4757 unwind entry previously opened by
4758 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4760 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4761 @cindex @code{cfi_personality} directive
4762 @code{.cfi_personality} defines personality routine and its encoding.
4763 @var{encoding} must be a constant determining how the personality
4764 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4765 argument is not present, otherwise second argument should be
4766 a constant or a symbol name. When using indirect encodings,
4767 the symbol provided should be the location where personality
4768 can be loaded from, not the personality routine itself.
4769 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4770 no personality routine.
4772 @subsection @code{.cfi_personality_id @var{id}}
4773 @cindex @code{cfi_personality_id} directive
4774 @code{cfi_personality_id} defines a personality routine by its index as
4775 defined in a compact unwinding format.
4776 Only valid when generating compact EH frames (i.e.
4777 with @code{.cfi_sections eh_frame_entry}.
4779 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4780 @cindex @code{cfi_fde_data} directive
4781 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4782 used for the current function. These are emitted inline in the
4783 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4784 in the @code{.gnu.extab} section otherwise.
4785 Only valid when generating compact EH frames (i.e.
4786 with @code{.cfi_sections eh_frame_entry}.
4788 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4789 @code{.cfi_lsda} defines LSDA and its encoding.
4790 @var{encoding} must be a constant determining how the LSDA
4791 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4792 argument is not present, otherwise the second argument should be a constant
4793 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4794 meaning that no LSDA is present.
4796 @subsection @code{.cfi_inline_lsda} [@var{align}]
4797 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4798 switches to the corresponding @code{.gnu.extab} section.
4799 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4800 Only valid when generating compact EH frames (i.e.
4801 with @code{.cfi_sections eh_frame_entry}.
4803 The table header and unwinding opcodes will be generated at this point,
4804 so that they are immediately followed by the LSDA data. The symbol
4805 referenced by the @code{.cfi_lsda} directive should still be defined
4806 in case a fallback FDE based encoding is used. The LSDA data is terminated
4807 by a section directive.
4809 The optional @var{align} argument specifies the alignment required.
4810 The alignment is specified as a power of two, as with the
4811 @code{.p2align} directive.
4813 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4814 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4815 address from @var{register} and add @var{offset} to it}.
4817 @subsection @code{.cfi_def_cfa_register @var{register}}
4818 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4819 now on @var{register} will be used instead of the old one. Offset
4822 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4823 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4824 remains the same, but @var{offset} is new. Note that it is the
4825 absolute offset that will be added to a defined register to compute
4828 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4829 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4830 value that is added/subtracted from the previous offset.
4832 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4833 Previous value of @var{register} is saved at offset @var{offset} from
4836 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4837 Previous value of @var{register} is CFA + @var{offset}.
4839 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4840 Previous value of @var{register} is saved at offset @var{offset} from
4841 the current CFA register. This is transformed to @code{.cfi_offset}
4842 using the known displacement of the CFA register from the CFA.
4843 This is often easier to use, because the number will match the
4844 code it's annotating.
4846 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4847 Previous value of @var{register1} is saved in register @var{register2}.
4849 @subsection @code{.cfi_restore @var{register}}
4850 @code{.cfi_restore} says that the rule for @var{register} is now the
4851 same as it was at the beginning of the function, after all initial
4852 instruction added by @code{.cfi_startproc} were executed.
4854 @subsection @code{.cfi_undefined @var{register}}
4855 From now on the previous value of @var{register} can't be restored anymore.
4857 @subsection @code{.cfi_same_value @var{register}}
4858 Current value of @var{register} is the same like in the previous frame,
4859 i.e. no restoration needed.
4861 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4862 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4863 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4864 places them in the current row. This is useful for situations where you have
4865 multiple @code{.cfi_*} directives that need to be undone due to the control
4866 flow of the program. For example, we could have something like this (assuming
4867 the CFA is the value of @code{rbp}):
4877 .cfi_def_cfa %rsp, 8
4880 /* Do something else */
4883 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4884 to the instructions before @code{label}. This means we'd have to add multiple
4885 @code{.cfi} directives after @code{label} to recreate the original save
4886 locations of the registers, as well as setting the CFA back to the value of
4887 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4899 .cfi_def_cfa %rsp, 8
4903 /* Do something else */
4906 That way, the rules for the instructions after @code{label} will be the same
4907 as before the first @code{.cfi_restore} without having to use multiple
4908 @code{.cfi} directives.
4910 @subsection @code{.cfi_return_column @var{register}}
4911 Change return column @var{register}, i.e. the return address is either
4912 directly in @var{register} or can be accessed by rules for @var{register}.
4914 @subsection @code{.cfi_signal_frame}
4915 Mark current function as signal trampoline.
4917 @subsection @code{.cfi_window_save}
4918 SPARC register window has been saved.
4920 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4921 Allows the user to add arbitrary bytes to the unwind info. One
4922 might use this to add OS-specific CFI opcodes, or generic CFI
4923 opcodes that GAS does not yet support.
4925 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4926 The current value of @var{register} is @var{label}. The value of @var{label}
4927 will be encoded in the output file according to @var{encoding}; see the
4928 description of @code{.cfi_personality} for details on this encoding.
4930 The usefulness of equating a register to a fixed label is probably
4931 limited to the return address register. Here, it can be useful to
4932 mark a code segment that has only one return address which is reached
4933 by a direct branch and no copy of the return address exists in memory
4934 or another register.
4937 @section @code{.comm @var{symbol} , @var{length} }
4939 @cindex @code{comm} directive
4940 @cindex symbol, common
4941 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4942 common symbol in one object file may be merged with a defined or common symbol
4943 of the same name in another object file. If @code{@value{LD}} does not see a
4944 definition for the symbol--just one or more common symbols--then it will
4945 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4946 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4947 the same name, and they do not all have the same size, it will allocate space
4948 using the largest size.
4951 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4952 an optional third argument. This is the desired alignment of the symbol,
4953 specified for ELF as a byte boundary (for example, an alignment of 16 means
4954 that the least significant 4 bits of the address should be zero), and for PE
4955 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4956 boundary). The alignment must be an absolute expression, and it must be a
4957 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4958 common symbol, it will use the alignment when placing the symbol. If no
4959 alignment is specified, @command{@value{AS}} will set the alignment to the
4960 largest power of two less than or equal to the size of the symbol, up to a
4961 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4962 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4963 @samp{--section-alignment} option; image file sections in PE are aligned to
4964 multiples of 4096, which is far too large an alignment for ordinary variables.
4965 It is rather the default alignment for (non-debug) sections within object
4966 (@samp{*.o}) files, which are less strictly aligned.}.
4970 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4971 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4975 @section @code{.data @var{subsection}}
4976 @cindex @code{data} directive
4978 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4979 end of the data subsection numbered @var{subsection} (which is an
4980 absolute expression). If @var{subsection} is omitted, it defaults
4984 @section @code{.dc[@var{size}] @var{expressions}}
4985 @cindex @code{dc} directive
4987 The @code{.dc} directive expects zero or more @var{expressions} separated by
4988 commas. These expressions are evaluated and their values inserted into the
4989 current section. The size of the emitted value depends upon the suffix to the
4990 @code{.dc} directive:
4994 Emits N-bit values, where N is the size of an address on the target system.
4998 Emits double precision floating-point values.
5000 Emits 32-bit values.
5002 Emits single precision floating-point values.
5004 Emits 16-bit values.
5005 Note - this is true even on targets where the @code{.word} directive would emit
5008 Emits long double precision floating-point values.
5011 If no suffix is used then @samp{.w} is assumed.
5013 The byte ordering is target dependent, as is the size and format of floating
5017 @section @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
5018 @cindex @code{dcb} directive
5019 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5020 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5021 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5022 @var{size} suffix, if present, must be one of:
5026 Emits single byte values.
5028 Emits double-precision floating point values.
5030 Emits 4-byte values.
5032 Emits single-precision floating point values.
5034 Emits 2-byte values.
5036 Emits long double-precision floating point values.
5039 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5041 The byte ordering is target dependent, as is the size and format of floating
5045 @section @code{.ds[@var{size}] @var{number} [,@var{fill}]}
5046 @cindex @code{ds} directive
5047 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5048 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5049 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5050 @var{size} suffix, if present, must be one of:
5054 Emits single byte values.
5056 Emits 8-byte values.
5058 Emits 4-byte values.
5060 Emits 12-byte values.
5062 Emits 4-byte values.
5064 Emits 2-byte values.
5066 Emits 12-byte values.
5069 Note - unlike the @code{.dcb} directive the @samp{.d}, @samp{.s} and @samp{.x}
5070 suffixes do not indicate that floating-point values are to be inserted.
5072 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5074 The byte ordering is target dependent.
5079 @section @code{.def @var{name}}
5081 @cindex @code{def} directive
5082 @cindex COFF symbols, debugging
5083 @cindex debugging COFF symbols
5084 Begin defining debugging information for a symbol @var{name}; the
5085 definition extends until the @code{.endef} directive is encountered.
5090 @section @code{.desc @var{symbol}, @var{abs-expression}}
5092 @cindex @code{desc} directive
5093 @cindex COFF symbol descriptor
5094 @cindex symbol descriptor, COFF
5095 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5096 to the low 16 bits of an absolute expression.
5099 The @samp{.desc} directive is not available when @command{@value{AS}} is
5100 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5101 object format. For the sake of compatibility, @command{@value{AS}} accepts
5102 it, but produces no output, when configured for COFF.
5108 @section @code{.dim}
5110 @cindex @code{dim} directive
5111 @cindex COFF auxiliary symbol information
5112 @cindex auxiliary symbol information, COFF
5113 This directive is generated by compilers to include auxiliary debugging
5114 information in the symbol table. It is only permitted inside
5115 @code{.def}/@code{.endef} pairs.
5119 @section @code{.double @var{flonums}}
5121 @cindex @code{double} directive
5122 @cindex floating point numbers (double)
5123 @code{.double} expects zero or more flonums, separated by commas. It
5124 assembles floating point numbers.
5126 The exact kind of floating point numbers emitted depends on how
5127 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5131 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5132 in @sc{ieee} format.
5137 @section @code{.eject}
5139 @cindex @code{eject} directive
5140 @cindex new page, in listings
5141 @cindex page, in listings
5142 @cindex listing control: new page
5143 Force a page break at this point, when generating assembly listings.
5146 @section @code{.else}
5148 @cindex @code{else} directive
5149 @code{.else} is part of the @command{@value{AS}} support for conditional
5150 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5151 of code to be assembled if the condition for the preceding @code{.if}
5155 @section @code{.elseif}
5157 @cindex @code{elseif} directive
5158 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5159 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5160 @code{.if} block that would otherwise fill the entire @code{.else} section.
5163 @section @code{.end}
5165 @cindex @code{end} directive
5166 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5167 process anything in the file past the @code{.end} directive.
5171 @section @code{.endef}
5173 @cindex @code{endef} directive
5174 This directive flags the end of a symbol definition begun with
5179 @section @code{.endfunc}
5180 @cindex @code{endfunc} directive
5181 @code{.endfunc} marks the end of a function specified with @code{.func}.
5184 @section @code{.endif}
5186 @cindex @code{endif} directive
5187 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5188 it marks the end of a block of code that is only assembled
5189 conditionally. @xref{If,,@code{.if}}.
5192 @section @code{.equ @var{symbol}, @var{expression}}
5194 @cindex @code{equ} directive
5195 @cindex assigning values to symbols
5196 @cindex symbols, assigning values to
5197 This directive sets the value of @var{symbol} to @var{expression}.
5198 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5201 The syntax for @code{equ} on the HPPA is
5202 @samp{@var{symbol} .equ @var{expression}}.
5206 The syntax for @code{equ} on the Z80 is
5207 @samp{@var{symbol} equ @var{expression}}.
5208 On the Z80 it is an error if @var{symbol} is already defined,
5209 but the symbol is not protected from later redefinition.
5210 Compare @ref{Equiv}.
5214 @section @code{.equiv @var{symbol}, @var{expression}}
5215 @cindex @code{equiv} directive
5216 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5217 the assembler will signal an error if @var{symbol} is already defined. Note a
5218 symbol which has been referenced but not actually defined is considered to be
5221 Except for the contents of the error message, this is roughly equivalent to
5228 plus it protects the symbol from later redefinition.
5231 @section @code{.eqv @var{symbol}, @var{expression}}
5232 @cindex @code{eqv} directive
5233 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5234 evaluate the expression or any part of it immediately. Instead each time
5235 the resulting symbol is used in an expression, a snapshot of its current
5239 @section @code{.err}
5240 @cindex @code{err} directive
5241 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5242 message and, unless the @option{-Z} option was used, it will not generate an
5243 object file. This can be used to signal an error in conditionally compiled code.
5246 @section @code{.error "@var{string}"}
5247 @cindex error directive
5249 Similarly to @code{.err}, this directive emits an error, but you can specify a
5250 string that will be emitted as the error message. If you don't specify the
5251 message, it defaults to @code{".error directive invoked in source file"}.
5252 @xref{Errors, ,Error and Warning Messages}.
5255 .error "This code has not been assembled and tested."
5259 @section @code{.exitm}
5260 Exit early from the current macro definition. @xref{Macro}.
5263 @section @code{.extern}
5265 @cindex @code{extern} directive
5266 @code{.extern} is accepted in the source program---for compatibility
5267 with other assemblers---but it is ignored. @command{@value{AS}} treats
5268 all undefined symbols as external.
5271 @section @code{.fail @var{expression}}
5273 @cindex @code{fail} directive
5274 Generates an error or a warning. If the value of the @var{expression} is 500
5275 or more, @command{@value{AS}} will print a warning message. If the value is less
5276 than 500, @command{@value{AS}} will print an error message. The message will
5277 include the value of @var{expression}. This can occasionally be useful inside
5278 complex nested macros or conditional assembly.
5281 @section @code{.file}
5282 @cindex @code{file} directive
5284 @ifclear no-file-dir
5285 There are two different versions of the @code{.file} directive. Targets
5286 that support DWARF2 line number information use the DWARF2 version of
5287 @code{.file}. Other targets use the default version.
5289 @subheading Default Version
5291 @cindex logical file name
5292 @cindex file name, logical
5293 This version of the @code{.file} directive tells @command{@value{AS}} that we
5294 are about to start a new logical file. The syntax is:
5300 @var{string} is the new file name. In general, the filename is
5301 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5302 to specify an empty file name, you must give the quotes--@code{""}. This
5303 statement may go away in future: it is only recognized to be compatible with
5304 old @command{@value{AS}} programs.
5306 @subheading DWARF2 Version
5309 When emitting DWARF2 line number information, @code{.file} assigns filenames
5310 to the @code{.debug_line} file name table. The syntax is:
5313 .file @var{fileno} @var{filename}
5316 The @var{fileno} operand should be a unique positive integer to use as the
5317 index of the entry in the table. The @var{filename} operand is a C string
5320 The detail of filename indices is exposed to the user because the filename
5321 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5322 information, and thus the user must know the exact indices that table
5326 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5328 @cindex @code{fill} directive
5329 @cindex writing patterns in memory
5330 @cindex patterns, writing in memory
5331 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5332 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5333 may be zero or more. @var{Size} may be zero or more, but if it is
5334 more than 8, then it is deemed to have the value 8, compatible with
5335 other people's assemblers. The contents of each @var{repeat} bytes
5336 is taken from an 8-byte number. The highest order 4 bytes are
5337 zero. The lowest order 4 bytes are @var{value} rendered in the
5338 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5339 Each @var{size} bytes in a repetition is taken from the lowest order
5340 @var{size} bytes of this number. Again, this bizarre behavior is
5341 compatible with other people's assemblers.
5343 @var{size} and @var{value} are optional.
5344 If the second comma and @var{value} are absent, @var{value} is
5345 assumed zero. If the first comma and following tokens are absent,
5346 @var{size} is assumed to be 1.
5349 @section @code{.float @var{flonums}}
5351 @cindex floating point numbers (single)
5352 @cindex @code{float} directive
5353 This directive assembles zero or more flonums, separated by commas. It
5354 has the same effect as @code{.single}.
5356 The exact kind of floating point numbers emitted depends on how
5357 @command{@value{AS}} is configured.
5358 @xref{Machine Dependencies}.
5362 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5363 in @sc{ieee} format.
5368 @section @code{.func @var{name}[,@var{label}]}
5369 @cindex @code{func} directive
5370 @code{.func} emits debugging information to denote function @var{name}, and
5371 is ignored unless the file is assembled with debugging enabled.
5372 Only @samp{--gstabs[+]} is currently supported.
5373 @var{label} is the entry point of the function and if omitted @var{name}
5374 prepended with the @samp{leading char} is used.
5375 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5376 All functions are currently defined to have @code{void} return type.
5377 The function must be terminated with @code{.endfunc}.
5380 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5382 @cindex @code{global} directive
5383 @cindex symbol, making visible to linker
5384 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5385 @var{symbol} in your partial program, its value is made available to
5386 other partial programs that are linked with it. Otherwise,
5387 @var{symbol} takes its attributes from a symbol of the same name
5388 from another file linked into the same program.
5390 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5391 compatibility with other assemblers.
5394 On the HPPA, @code{.global} is not always enough to make it accessible to other
5395 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5396 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5401 @section @code{.gnu_attribute @var{tag},@var{value}}
5402 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5405 @section @code{.hidden @var{names}}
5407 @cindex @code{hidden} directive
5409 This is one of the ELF visibility directives. The other two are
5410 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5411 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5413 This directive overrides the named symbols default visibility (which is set by
5414 their binding: local, global or weak). The directive sets the visibility to
5415 @code{hidden} which means that the symbols are not visible to other components.
5416 Such symbols are always considered to be @code{protected} as well.
5420 @section @code{.hword @var{expressions}}
5422 @cindex @code{hword} directive
5423 @cindex integers, 16-bit
5424 @cindex numbers, 16-bit
5425 @cindex sixteen bit integers
5426 This expects zero or more @var{expressions}, and emits
5427 a 16 bit number for each.
5430 This directive is a synonym for @samp{.short}; depending on the target
5431 architecture, it may also be a synonym for @samp{.word}.
5435 This directive is a synonym for @samp{.short}.
5438 This directive is a synonym for both @samp{.short} and @samp{.word}.
5443 @section @code{.ident}
5445 @cindex @code{ident} directive
5447 This directive is used by some assemblers to place tags in object files. The
5448 behavior of this directive varies depending on the target. When using the
5449 a.out object file format, @command{@value{AS}} simply accepts the directive for
5450 source-file compatibility with existing assemblers, but does not emit anything
5451 for it. When using COFF, comments are emitted to the @code{.comment} or
5452 @code{.rdata} section, depending on the target. When using ELF, comments are
5453 emitted to the @code{.comment} section.
5456 @section @code{.if @var{absolute expression}}
5458 @cindex conditional assembly
5459 @cindex @code{if} directive
5460 @code{.if} marks the beginning of a section of code which is only
5461 considered part of the source program being assembled if the argument
5462 (which must be an @var{absolute expression}) is non-zero. The end of
5463 the conditional section of code must be marked by @code{.endif}
5464 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5465 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5466 If you have several conditions to check, @code{.elseif} may be used to avoid
5467 nesting blocks if/else within each subsequent @code{.else} block.
5469 The following variants of @code{.if} are also supported:
5471 @cindex @code{ifdef} directive
5472 @item .ifdef @var{symbol}
5473 Assembles the following section of code if the specified @var{symbol}
5474 has been defined. Note a symbol which has been referenced but not yet defined
5475 is considered to be undefined.
5477 @cindex @code{ifb} directive
5478 @item .ifb @var{text}
5479 Assembles the following section of code if the operand is blank (empty).
5481 @cindex @code{ifc} directive
5482 @item .ifc @var{string1},@var{string2}
5483 Assembles the following section of code if the two strings are the same. The
5484 strings may be optionally quoted with single quotes. If they are not quoted,
5485 the first string stops at the first comma, and the second string stops at the
5486 end of the line. Strings which contain whitespace should be quoted. The
5487 string comparison is case sensitive.
5489 @cindex @code{ifeq} directive
5490 @item .ifeq @var{absolute expression}
5491 Assembles the following section of code if the argument is zero.
5493 @cindex @code{ifeqs} directive
5494 @item .ifeqs @var{string1},@var{string2}
5495 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5497 @cindex @code{ifge} directive
5498 @item .ifge @var{absolute expression}
5499 Assembles the following section of code if the argument is greater than or
5502 @cindex @code{ifgt} directive
5503 @item .ifgt @var{absolute expression}
5504 Assembles the following section of code if the argument is greater than zero.
5506 @cindex @code{ifle} directive
5507 @item .ifle @var{absolute expression}
5508 Assembles the following section of code if the argument is less than or equal
5511 @cindex @code{iflt} directive
5512 @item .iflt @var{absolute expression}
5513 Assembles the following section of code if the argument is less than zero.
5515 @cindex @code{ifnb} directive
5516 @item .ifnb @var{text}
5517 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5518 following section of code if the operand is non-blank (non-empty).
5520 @cindex @code{ifnc} directive
5521 @item .ifnc @var{string1},@var{string2}.
5522 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5523 following section of code if the two strings are not the same.
5525 @cindex @code{ifndef} directive
5526 @cindex @code{ifnotdef} directive
5527 @item .ifndef @var{symbol}
5528 @itemx .ifnotdef @var{symbol}
5529 Assembles the following section of code if the specified @var{symbol}
5530 has not been defined. Both spelling variants are equivalent. Note a symbol
5531 which has been referenced but not yet defined is considered to be undefined.
5533 @cindex @code{ifne} directive
5534 @item .ifne @var{absolute expression}
5535 Assembles the following section of code if the argument is not equal to zero
5536 (in other words, this is equivalent to @code{.if}).
5538 @cindex @code{ifnes} directive
5539 @item .ifnes @var{string1},@var{string2}
5540 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5541 following section of code if the two strings are not the same.
5545 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5547 @cindex @code{incbin} directive
5548 @cindex binary files, including
5549 The @code{incbin} directive includes @var{file} verbatim at the current
5550 location. You can control the search paths used with the @samp{-I} command-line
5551 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5554 The @var{skip} argument skips a number of bytes from the start of the
5555 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5556 read. Note that the data is not aligned in any way, so it is the user's
5557 responsibility to make sure that proper alignment is provided both before and
5558 after the @code{incbin} directive.
5561 @section @code{.include "@var{file}"}
5563 @cindex @code{include} directive
5564 @cindex supporting files, including
5565 @cindex files, including
5566 This directive provides a way to include supporting files at specified
5567 points in your source program. The code from @var{file} is assembled as
5568 if it followed the point of the @code{.include}; when the end of the
5569 included file is reached, assembly of the original file continues. You
5570 can control the search paths used with the @samp{-I} command-line option
5571 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5575 @section @code{.int @var{expressions}}
5577 @cindex @code{int} directive
5578 @cindex integers, 32-bit
5579 Expect zero or more @var{expressions}, of any section, separated by commas.
5580 For each expression, emit a number that, at run time, is the value of that
5581 expression. The byte order and bit size of the number depends on what kind
5582 of target the assembly is for.
5586 On most forms of the H8/300, @code{.int} emits 16-bit
5587 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5594 @section @code{.internal @var{names}}
5596 @cindex @code{internal} directive
5598 This is one of the ELF visibility directives. The other two are
5599 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5600 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5602 This directive overrides the named symbols default visibility (which is set by
5603 their binding: local, global or weak). The directive sets the visibility to
5604 @code{internal} which means that the symbols are considered to be @code{hidden}
5605 (i.e., not visible to other components), and that some extra, processor specific
5606 processing must also be performed upon the symbols as well.
5610 @section @code{.irp @var{symbol},@var{values}}@dots{}
5612 @cindex @code{irp} directive
5613 Evaluate a sequence of statements assigning different values to @var{symbol}.
5614 The sequence of statements starts at the @code{.irp} directive, and is
5615 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5616 set to @var{value}, and the sequence of statements is assembled. If no
5617 @var{value} is listed, the sequence of statements is assembled once, with
5618 @var{symbol} set to the null string. To refer to @var{symbol} within the
5619 sequence of statements, use @var{\symbol}.
5621 For example, assembling
5629 is equivalent to assembling
5637 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5640 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5642 @cindex @code{irpc} directive
5643 Evaluate a sequence of statements assigning different values to @var{symbol}.
5644 The sequence of statements starts at the @code{.irpc} directive, and is
5645 terminated by an @code{.endr} directive. For each character in @var{value},
5646 @var{symbol} is set to the character, and the sequence of statements is
5647 assembled. If no @var{value} is listed, the sequence of statements is
5648 assembled once, with @var{symbol} set to the null string. To refer to
5649 @var{symbol} within the sequence of statements, use @var{\symbol}.
5651 For example, assembling
5659 is equivalent to assembling
5667 For some caveats with the spelling of @var{symbol}, see also the discussion
5671 @section @code{.lcomm @var{symbol} , @var{length}}
5673 @cindex @code{lcomm} directive
5674 @cindex local common symbols
5675 @cindex symbols, local common
5676 Reserve @var{length} (an absolute expression) bytes for a local common
5677 denoted by @var{symbol}. The section and value of @var{symbol} are
5678 those of the new local common. The addresses are allocated in the bss
5679 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5680 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5681 not visible to @code{@value{LD}}.
5684 Some targets permit a third argument to be used with @code{.lcomm}. This
5685 argument specifies the desired alignment of the symbol in the bss section.
5689 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5690 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5694 @section @code{.lflags}
5696 @cindex @code{lflags} directive (ignored)
5697 @command{@value{AS}} accepts this directive, for compatibility with other
5698 assemblers, but ignores it.
5700 @ifclear no-line-dir
5702 @section @code{.line @var{line-number}}
5704 @cindex @code{line} directive
5705 @cindex logical line number
5707 Change the logical line number. @var{line-number} must be an absolute
5708 expression. The next line has that logical line number. Therefore any other
5709 statements on the current line (after a statement separator character) are
5710 reported as on logical line number @var{line-number} @minus{} 1. One day
5711 @command{@value{AS}} will no longer support this directive: it is recognized only
5712 for compatibility with existing assembler programs.
5715 Even though this is a directive associated with the @code{a.out} or
5716 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5717 when producing COFF output, and treats @samp{.line} as though it
5718 were the COFF @samp{.ln} @emph{if} it is found outside a
5719 @code{.def}/@code{.endef} pair.
5721 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5722 used by compilers to generate auxiliary symbol information for
5727 @section @code{.linkonce [@var{type}]}
5729 @cindex @code{linkonce} directive
5730 @cindex common sections
5731 Mark the current section so that the linker only includes a single copy of it.
5732 This may be used to include the same section in several different object files,
5733 but ensure that the linker will only include it once in the final output file.
5734 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5735 Duplicate sections are detected based on the section name, so it should be
5738 This directive is only supported by a few object file formats; as of this
5739 writing, the only object file format which supports it is the Portable
5740 Executable format used on Windows NT.
5742 The @var{type} argument is optional. If specified, it must be one of the
5743 following strings. For example:
5747 Not all types may be supported on all object file formats.
5751 Silently discard duplicate sections. This is the default.
5754 Warn if there are duplicate sections, but still keep only one copy.
5757 Warn if any of the duplicates have different sizes.
5760 Warn if any of the duplicates do not have exactly the same contents.
5764 @section @code{.list}
5766 @cindex @code{list} directive
5767 @cindex listing control, turning on
5768 Control (in conjunction with the @code{.nolist} directive) whether or
5769 not assembly listings are generated. These two directives maintain an
5770 internal counter (which is zero initially). @code{.list} increments the
5771 counter, and @code{.nolist} decrements it. Assembly listings are
5772 generated whenever the counter is greater than zero.
5774 By default, listings are disabled. When you enable them (with the
5775 @samp{-a} command-line option; @pxref{Invoking,,Command-Line Options}),
5776 the initial value of the listing counter is one.
5779 @section @code{.ln @var{line-number}}
5781 @cindex @code{ln} directive
5782 @ifclear no-line-dir
5783 @samp{.ln} is a synonym for @samp{.line}.
5786 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5787 must be an absolute expression. The next line has that logical
5788 line number, so any other statements on the current line (after a
5789 statement separator character @code{;}) are reported as on logical
5790 line number @var{line-number} @minus{} 1.
5794 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5795 @cindex @code{loc} directive
5796 When emitting DWARF2 line number information,
5797 the @code{.loc} directive will add a row to the @code{.debug_line} line
5798 number matrix corresponding to the immediately following assembly
5799 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5800 arguments will be applied to the @code{.debug_line} state machine before
5803 The @var{options} are a sequence of the following tokens in any order:
5807 This option will set the @code{basic_block} register in the
5808 @code{.debug_line} state machine to @code{true}.
5811 This option will set the @code{prologue_end} register in the
5812 @code{.debug_line} state machine to @code{true}.
5814 @item epilogue_begin
5815 This option will set the @code{epilogue_begin} register in the
5816 @code{.debug_line} state machine to @code{true}.
5818 @item is_stmt @var{value}
5819 This option will set the @code{is_stmt} register in the
5820 @code{.debug_line} state machine to @code{value}, which must be
5823 @item isa @var{value}
5824 This directive will set the @code{isa} register in the @code{.debug_line}
5825 state machine to @var{value}, which must be an unsigned integer.
5827 @item discriminator @var{value}
5828 This directive will set the @code{discriminator} register in the @code{.debug_line}
5829 state machine to @var{value}, which must be an unsigned integer.
5831 @item view @var{value}
5832 This option causes a row to be added to @code{.debug_line} in reference to the
5833 current address (which might not be the same as that of the following assembly
5834 instruction), and to associate @var{value} with the @code{view} register in the
5835 @code{.debug_line} state machine. If @var{value} is a label, both the
5836 @code{view} register and the label are set to the number of prior @code{.loc}
5837 directives at the same program location. If @var{value} is the literal
5838 @code{0}, the @code{view} register is set to zero, and the assembler asserts
5839 that there aren't any prior @code{.loc} directives at the same program
5840 location. If @var{value} is the literal @code{-0}, the assembler arrange for
5841 the @code{view} register to be reset in this row, even if there are prior
5842 @code{.loc} directives at the same program location.
5846 @node Loc_mark_labels
5847 @section @code{.loc_mark_labels @var{enable}}
5848 @cindex @code{loc_mark_labels} directive
5849 When emitting DWARF2 line number information,
5850 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5851 to the @code{.debug_line} line number matrix with the @code{basic_block}
5852 register in the state machine set whenever a code label is seen.
5853 The @var{enable} argument should be either 1 or 0, to enable or disable
5854 this function respectively.
5858 @section @code{.local @var{names}}
5860 @cindex @code{local} directive
5861 This directive, which is available for ELF targets, marks each symbol in
5862 the comma-separated list of @code{names} as a local symbol so that it
5863 will not be externally visible. If the symbols do not already exist,
5864 they will be created.
5866 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5867 accept an alignment argument, which is the case for most ELF targets,
5868 the @code{.local} directive can be used in combination with @code{.comm}
5869 (@pxref{Comm}) to define aligned local common data.
5873 @section @code{.long @var{expressions}}
5875 @cindex @code{long} directive
5876 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5879 @c no one seems to know what this is for or whether this description is
5880 @c what it really ought to do
5882 @section @code{.lsym @var{symbol}, @var{expression}}
5884 @cindex @code{lsym} directive
5885 @cindex symbol, not referenced in assembly
5886 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5887 the hash table, ensuring it cannot be referenced by name during the
5888 rest of the assembly. This sets the attributes of the symbol to be
5889 the same as the expression value:
5891 @var{other} = @var{descriptor} = 0
5892 @var{type} = @r{(section of @var{expression})}
5893 @var{value} = @var{expression}
5896 The new symbol is not flagged as external.
5900 @section @code{.macro}
5903 The commands @code{.macro} and @code{.endm} allow you to define macros that
5904 generate assembly output. For example, this definition specifies a macro
5905 @code{sum} that puts a sequence of numbers into memory:
5908 .macro sum from=0, to=5
5917 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5929 @item .macro @var{macname}
5930 @itemx .macro @var{macname} @var{macargs} @dots{}
5931 @cindex @code{macro} directive
5932 Begin the definition of a macro called @var{macname}. If your macro
5933 definition requires arguments, specify their names after the macro name,
5934 separated by commas or spaces. You can qualify the macro argument to
5935 indicate whether all invocations must specify a non-blank value (through
5936 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5937 (through @samp{:@code{vararg}}). You can supply a default value for any
5938 macro argument by following the name with @samp{=@var{deflt}}. You
5939 cannot define two macros with the same @var{macname} unless it has been
5940 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5941 definitions. For example, these are all valid @code{.macro} statements:
5945 Begin the definition of a macro called @code{comm}, which takes no
5948 @item .macro plus1 p, p1
5949 @itemx .macro plus1 p p1
5950 Either statement begins the definition of a macro called @code{plus1},
5951 which takes two arguments; within the macro definition, write
5952 @samp{\p} or @samp{\p1} to evaluate the arguments.
5954 @item .macro reserve_str p1=0 p2
5955 Begin the definition of a macro called @code{reserve_str}, with two
5956 arguments. The first argument has a default value, but not the second.
5957 After the definition is complete, you can call the macro either as
5958 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5959 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5960 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5961 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5963 @item .macro m p1:req, p2=0, p3:vararg
5964 Begin the definition of a macro called @code{m}, with at least three
5965 arguments. The first argument must always have a value specified, but
5966 not the second, which instead has a default value. The third formal
5967 will get assigned all remaining arguments specified at invocation time.
5969 When you call a macro, you can specify the argument values either by
5970 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5971 @samp{sum to=17, from=9}.
5975 Note that since each of the @var{macargs} can be an identifier exactly
5976 as any other one permitted by the target architecture, there may be
5977 occasional problems if the target hand-crafts special meanings to certain
5978 characters when they occur in a special position. For example, if the colon
5979 (@code{:}) is generally permitted to be part of a symbol name, but the
5980 architecture specific code special-cases it when occurring as the final
5981 character of a symbol (to denote a label), then the macro parameter
5982 replacement code will have no way of knowing that and consider the whole
5983 construct (including the colon) an identifier, and check only this
5984 identifier for being the subject to parameter substitution. So for example
5985 this macro definition:
5993 might not work as expected. Invoking @samp{label foo} might not create a label
5994 called @samp{foo} but instead just insert the text @samp{\l:} into the
5995 assembler source, probably generating an error about an unrecognised
5998 Similarly problems might occur with the period character (@samp{.})
5999 which is often allowed inside opcode names (and hence identifier names). So
6000 for example constructing a macro to build an opcode from a base name and a
6001 length specifier like this:
6004 .macro opcode base length
6009 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
6010 instruction but instead generate some kind of error as the assembler tries to
6011 interpret the text @samp{\base.\length}.
6013 There are several possible ways around this problem:
6016 @item Insert white space
6017 If it is possible to use white space characters then this is the simplest
6026 @item Use @samp{\()}
6027 The string @samp{\()} can be used to separate the end of a macro argument from
6028 the following text. eg:
6031 .macro opcode base length
6036 @item Use the alternate macro syntax mode
6037 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
6038 used as a separator. eg:
6048 Note: this problem of correctly identifying string parameters to pseudo ops
6049 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
6050 and @code{.irpc} (@pxref{Irpc}) as well.
6053 @cindex @code{endm} directive
6054 Mark the end of a macro definition.
6057 @cindex @code{exitm} directive
6058 Exit early from the current macro definition.
6060 @cindex number of macros executed
6061 @cindex macros, count executed
6063 @command{@value{AS}} maintains a counter of how many macros it has
6064 executed in this pseudo-variable; you can copy that number to your
6065 output with @samp{\@@}, but @emph{only within a macro definition}.
6067 @item LOCAL @var{name} [ , @dots{} ]
6068 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
6069 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
6070 @xref{Altmacro,,@code{.altmacro}}.
6074 @section @code{.mri @var{val}}
6076 @cindex @code{mri} directive
6077 @cindex MRI mode, temporarily
6078 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
6079 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
6080 affects code assembled until the next @code{.mri} directive, or until the end
6081 of the file. @xref{M, MRI mode, MRI mode}.
6084 @section @code{.noaltmacro}
6085 Disable alternate macro mode. @xref{Altmacro}.
6088 @section @code{.nolist}
6090 @cindex @code{nolist} directive
6091 @cindex listing control, turning off
6092 Control (in conjunction with the @code{.list} directive) whether or
6093 not assembly listings are generated. These two directives maintain an
6094 internal counter (which is zero initially). @code{.list} increments the
6095 counter, and @code{.nolist} decrements it. Assembly listings are
6096 generated whenever the counter is greater than zero.
6099 @section @code{.nops @var{size}[, @var{control}]}
6101 @cindex @code{nops} directive
6102 @cindex filling memory with no-op instructions
6103 This directive emits @var{size} bytes filled with no-op instructions.
6104 @var{size} is absolute expression, which must be a positve value.
6105 @var{control} controls how no-op instructions should be generated. If
6106 the comma and @var{control} are omitted, @var{control} is assumed to be
6109 Note: For Intel 80386 and AMD x86-64 targets, @var{control} specifies
6110 the size limit of a no-op instruction. The valid values of @var{control}
6111 are between 0 and 4 in 16-bit mode, between 0 and 7 when tuning for
6112 older processors in 32-bit mode, between 0 and 11 in 64-bit mode or when
6113 tuning for newer processors in 32-bit mode. When 0 is used, the no-op
6114 instruction size limit is set to the maximum supported size.
6117 @section @code{.octa @var{bignums}}
6119 @c FIXME: double size emitted for "octa" on some? Or warn?
6120 @cindex @code{octa} directive
6121 @cindex integer, 16-byte
6122 @cindex sixteen byte integer
6123 This directive expects zero or more bignums, separated by commas. For each
6124 bignum, it emits a 16-byte integer.
6126 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6127 hence @emph{octa}-word for 16 bytes.
6130 @section @code{.offset @var{loc}}
6132 @cindex @code{offset} directive
6133 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6134 be an absolute expression. This directive may be useful for defining
6135 symbols with absolute values. Do not confuse it with the @code{.org}
6139 @section @code{.org @var{new-lc} , @var{fill}}
6141 @cindex @code{org} directive
6142 @cindex location counter, advancing
6143 @cindex advancing location counter
6144 @cindex current address, advancing
6145 Advance the location counter of the current section to
6146 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6147 expression with the same section as the current subsection. That is,
6148 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6149 wrong section, the @code{.org} directive is ignored. To be compatible
6150 with former assemblers, if the section of @var{new-lc} is absolute,
6151 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6152 is the same as the current subsection.
6154 @code{.org} may only increase the location counter, or leave it
6155 unchanged; you cannot use @code{.org} to move the location counter
6158 @c double negative used below "not undefined" because this is a specific
6159 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6160 @c section. doc@cygnus.com 18feb91
6161 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6162 may not be undefined. If you really detest this restriction we eagerly await
6163 a chance to share your improved assembler.
6165 Beware that the origin is relative to the start of the section, not
6166 to the start of the subsection. This is compatible with other
6167 people's assemblers.
6169 When the location counter (of the current subsection) is advanced, the
6170 intervening bytes are filled with @var{fill} which should be an
6171 absolute expression. If the comma and @var{fill} are omitted,
6172 @var{fill} defaults to zero.
6175 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
6177 @cindex padding the location counter given a power of two
6178 @cindex @code{p2align} directive
6179 Pad the location counter (in the current subsection) to a particular
6180 storage boundary. The first expression (which must be absolute) is the
6181 number of low-order zero bits the location counter must have after
6182 advancement. For example @samp{.p2align 3} advances the location
6183 counter until it a multiple of 8. If the location counter is already a
6184 multiple of 8, no change is needed.
6186 The second expression (also absolute) gives the fill value to be stored in the
6187 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6188 padding bytes are normally zero. However, on most systems, if the section is
6189 marked as containing code and the fill value is omitted, the space is filled
6190 with no-op instructions.
6192 The third expression is also absolute, and is also optional. If it is present,
6193 it is the maximum number of bytes that should be skipped by this alignment
6194 directive. If doing the alignment would require skipping more bytes than the
6195 specified maximum, then the alignment is not done at all. You can omit the
6196 fill value (the second argument) entirely by simply using two commas after the
6197 required alignment; this can be useful if you want the alignment to be filled
6198 with no-op instructions when appropriate.
6200 @cindex @code{p2alignw} directive
6201 @cindex @code{p2alignl} directive
6202 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6203 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6204 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6205 fill pattern as a four byte longword value. For example, @code{.p2alignw
6206 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6207 filled in with the value 0x368d (the exact placement of the bytes depends upon
6208 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6213 @section @code{.popsection}
6215 @cindex @code{popsection} directive
6216 @cindex Section Stack
6217 This is one of the ELF section stack manipulation directives. The others are
6218 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6219 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6222 This directive replaces the current section (and subsection) with the top
6223 section (and subsection) on the section stack. This section is popped off the
6229 @section @code{.previous}
6231 @cindex @code{previous} directive
6232 @cindex Section Stack
6233 This is one of the ELF section stack manipulation directives. The others are
6234 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6235 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6236 (@pxref{PopSection}).
6238 This directive swaps the current section (and subsection) with most recently
6239 referenced section/subsection pair prior to this one. Multiple
6240 @code{.previous} directives in a row will flip between two sections (and their
6241 subsections). For example:
6253 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6259 # Now in section A subsection 1
6263 # Now in section B subsection 0
6266 # Now in section B subsection 1
6269 # Now in section B subsection 0
6273 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6274 section B and 0x9abc into subsection 1 of section B.
6276 In terms of the section stack, this directive swaps the current section with
6277 the top section on the section stack.
6281 @section @code{.print @var{string}}
6283 @cindex @code{print} directive
6284 @command{@value{AS}} will print @var{string} on the standard output during
6285 assembly. You must put @var{string} in double quotes.
6289 @section @code{.protected @var{names}}
6291 @cindex @code{protected} directive
6293 This is one of the ELF visibility directives. The other two are
6294 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6296 This directive overrides the named symbols default visibility (which is set by
6297 their binding: local, global or weak). The directive sets the visibility to
6298 @code{protected} which means that any references to the symbols from within the
6299 components that defines them must be resolved to the definition in that
6300 component, even if a definition in another component would normally preempt
6305 @section @code{.psize @var{lines} , @var{columns}}
6307 @cindex @code{psize} directive
6308 @cindex listing control: paper size
6309 @cindex paper size, for listings
6310 Use this directive to declare the number of lines---and, optionally, the
6311 number of columns---to use for each page, when generating listings.
6313 If you do not use @code{.psize}, listings use a default line-count
6314 of 60. You may omit the comma and @var{columns} specification; the
6315 default width is 200 columns.
6317 @command{@value{AS}} generates formfeeds whenever the specified number of
6318 lines is exceeded (or whenever you explicitly request one, using
6321 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6322 those explicitly specified with @code{.eject}.
6325 @section @code{.purgem @var{name}}
6327 @cindex @code{purgem} directive
6328 Undefine the macro @var{name}, so that later uses of the string will not be
6329 expanded. @xref{Macro}.
6333 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6335 @cindex @code{pushsection} directive
6336 @cindex Section Stack
6337 This is one of the ELF section stack manipulation directives. The others are
6338 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6339 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6342 This directive pushes the current section (and subsection) onto the
6343 top of the section stack, and then replaces the current section and
6344 subsection with @code{name} and @code{subsection}. The optional
6345 @code{flags}, @code{type} and @code{arguments} are treated the same
6346 as in the @code{.section} (@pxref{Section}) directive.
6350 @section @code{.quad @var{bignums}}
6352 @cindex @code{quad} directive
6353 @code{.quad} expects zero or more bignums, separated by commas. For
6354 each bignum, it emits
6356 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6357 warning message; and just takes the lowest order 8 bytes of the bignum.
6358 @cindex eight-byte integer
6359 @cindex integer, 8-byte
6361 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6362 hence @emph{quad}-word for 8 bytes.
6365 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6366 warning message; and just takes the lowest order 16 bytes of the bignum.
6367 @cindex sixteen-byte integer
6368 @cindex integer, 16-byte
6372 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6374 @cindex @code{reloc} directive
6375 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6376 @var{expression}. If @var{offset} is a number, the relocation is generated in
6377 the current section. If @var{offset} is an expression that resolves to a
6378 symbol plus offset, the relocation is generated in the given symbol's section.
6379 @var{expression}, if present, must resolve to a symbol plus addend or to an
6380 absolute value, but note that not all targets support an addend. e.g. ELF REL
6381 targets such as i386 store an addend in the section contents rather than in the
6382 relocation. This low level interface does not support addends stored in the
6386 @section @code{.rept @var{count}}
6388 @cindex @code{rept} directive
6389 Repeat the sequence of lines between the @code{.rept} directive and the next
6390 @code{.endr} directive @var{count} times.
6392 For example, assembling
6400 is equivalent to assembling
6408 A count of zero is allowed, but nothing is generated. Negative counts are not
6409 allowed and if encountered will be treated as if they were zero.
6412 @section @code{.sbttl "@var{subheading}"}
6414 @cindex @code{sbttl} directive
6415 @cindex subtitles for listings
6416 @cindex listing control: subtitle
6417 Use @var{subheading} as the title (third line, immediately after the
6418 title line) when generating assembly listings.
6420 This directive affects subsequent pages, as well as the current page if
6421 it appears within ten lines of the top of a page.
6425 @section @code{.scl @var{class}}
6427 @cindex @code{scl} directive
6428 @cindex symbol storage class (COFF)
6429 @cindex COFF symbol storage class
6430 Set the storage-class value for a symbol. This directive may only be
6431 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6432 whether a symbol is static or external, or it may record further
6433 symbolic debugging information.
6438 @section @code{.section @var{name}}
6440 @cindex named section
6441 Use the @code{.section} directive to assemble the following code into a section
6444 This directive is only supported for targets that actually support arbitrarily
6445 named sections; on @code{a.out} targets, for example, it is not accepted, even
6446 with a standard @code{a.out} section name.
6450 @c only print the extra heading if both COFF and ELF are set
6451 @subheading COFF Version
6454 @cindex @code{section} directive (COFF version)
6455 For COFF targets, the @code{.section} directive is used in one of the following
6459 .section @var{name}[, "@var{flags}"]
6460 .section @var{name}[, @var{subsection}]
6463 If the optional argument is quoted, it is taken as flags to use for the
6464 section. Each flag is a single character. The following flags are recognized:
6468 bss section (uninitialized data)
6470 section is not loaded
6476 exclude section from linking
6482 shared section (meaningful for PE targets)
6484 ignored. (For compatibility with the ELF version)
6486 section is not readable (meaningful for PE targets)
6488 single-digit power-of-two section alignment (GNU extension)
6491 If no flags are specified, the default flags depend upon the section name. If
6492 the section name is not recognized, the default will be for the section to be
6493 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6494 from the section, rather than adding them, so if they are used on their own it
6495 will be as if no flags had been specified at all.
6497 If the optional argument to the @code{.section} directive is not quoted, it is
6498 taken as a subsection number (@pxref{Sub-Sections}).
6503 @c only print the extra heading if both COFF and ELF are set
6504 @subheading ELF Version
6507 @cindex Section Stack
6508 This is one of the ELF section stack manipulation directives. The others are
6509 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6510 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6511 @code{.previous} (@pxref{Previous}).
6513 @cindex @code{section} directive (ELF version)
6514 For ELF targets, the @code{.section} directive is used like this:
6517 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6520 @anchor{Section Name Substitutions}
6521 @kindex --sectname-subst
6522 @cindex section name substitution
6523 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6524 argument may contain a substitution sequence. Only @code{%S} is supported
6525 at the moment, and substitutes the current section name. For example:
6528 .macro exception_code
6529 .section %S.exception
6530 [exception code here]
6545 The two @code{exception_code} invocations above would create the
6546 @code{.text.exception} and @code{.init.exception} sections respectively.
6547 This is useful e.g. to discriminate between ancillary sections that are
6548 tied to setup code to be discarded after use from ancillary sections that
6549 need to stay resident without having to define multiple @code{exception_code}
6550 macros just for that purpose.
6552 The optional @var{flags} argument is a quoted string which may contain any
6553 combination of the following characters:
6557 section is allocatable
6559 section is a GNU_MBIND section
6561 section is excluded from executable and shared library.
6565 section is executable
6567 section is mergeable
6569 section contains zero terminated strings
6571 section is a member of a section group
6573 section is used for thread-local-storage
6575 section is a member of the previously-current section's group, if any
6576 @item @code{<number>}
6577 a numeric value indicating the bits to be set in the ELF section header's flags
6578 field. Note - if one or more of the alphabetic characters described above is
6579 also included in the flags field, their bit values will be ORed into the
6581 @item @code{<target specific>}
6582 some targets extend this list with their own flag characters
6585 Note - once a section's flags have been set they cannot be changed. There are
6586 a few exceptions to this rule however. Processor and application specific
6587 flags can be added to an already defined section. The @code{.interp},
6588 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6589 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6590 section may have the executable (@code{x}) flag added.
6592 The optional @var{type} argument may contain one of the following constants:
6596 section contains data
6598 section does not contain data (i.e., section only occupies space)
6600 section contains data which is used by things other than the program
6602 section contains an array of pointers to init functions
6604 section contains an array of pointers to finish functions
6605 @item @@preinit_array
6606 section contains an array of pointers to pre-init functions
6607 @item @@@code{<number>}
6608 a numeric value to be set as the ELF section header's type field.
6609 @item @@@code{<target specific>}
6610 some targets extend this list with their own types
6613 Many targets only support the first three section types. The type may be
6614 enclosed in double quotes if necessary.
6616 Note on targets where the @code{@@} character is the start of a comment (eg
6617 ARM) then another character is used instead. For example the ARM port uses the
6620 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6621 special and have fixed types. Any attempt to declare them with a different
6622 type will generate an error from the assembler.
6624 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6625 be specified as well as an extra argument---@var{entsize}---like this:
6628 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6631 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6632 constants, each @var{entsize} octets long. Sections with both @code{M} and
6633 @code{S} must contain zero terminated strings where each character is
6634 @var{entsize} bytes long. The linker may remove duplicates within sections with
6635 the same name, same entity size and same flags. @var{entsize} must be an
6636 absolute expression. For sections with both @code{M} and @code{S}, a string
6637 which is a suffix of a larger string is considered a duplicate. Thus
6638 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6639 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6641 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6642 be present along with an additional field like this:
6645 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6648 The @var{GroupName} field specifies the name of the section group to which this
6649 particular section belongs. The optional linkage field can contain:
6653 indicates that only one copy of this section should be retained
6658 Note: if both the @var{M} and @var{G} flags are present then the fields for
6659 the Merge flag should come first, like this:
6662 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6665 If @var{flags} contains the @code{?} symbol then it may not also contain the
6666 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6667 present. Instead, @code{?} says to consider the section that's current before
6668 this directive. If that section used @code{G}, then the new section will use
6669 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6670 If not, then the @code{?} symbol has no effect.
6672 If no flags are specified, the default flags depend upon the section name. If
6673 the section name is not recognized, the default will be for the section to have
6674 none of the above flags: it will not be allocated in memory, nor writable, nor
6675 executable. The section will contain data.
6677 For ELF targets, the assembler supports another type of @code{.section}
6678 directive for compatibility with the Solaris assembler:
6681 .section "@var{name}"[, @var{flags}...]
6684 Note that the section name is quoted. There may be a sequence of comma
6689 section is allocatable
6693 section is executable
6695 section is excluded from executable and shared library.
6697 section is used for thread local storage
6700 This directive replaces the current section and subsection. See the
6701 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6702 some examples of how this directive and the other section stack directives
6708 @section @code{.set @var{symbol}, @var{expression}}
6710 @cindex @code{set} directive
6711 @cindex symbol value, setting
6712 Set the value of @var{symbol} to @var{expression}. This
6713 changes @var{symbol}'s value and type to conform to
6714 @var{expression}. If @var{symbol} was flagged as external, it remains
6715 flagged (@pxref{Symbol Attributes}).
6717 You may @code{.set} a symbol many times in the same assembly provided that the
6718 values given to the symbol are constants. Values that are based on expressions
6719 involving other symbols are allowed, but some targets may restrict this to only
6720 being done once per assembly. This is because those targets do not set the
6721 addresses of symbols at assembly time, but rather delay the assignment until a
6722 final link is performed. This allows the linker a chance to change the code in
6723 the files, changing the location of, and the relative distance between, various
6726 If you @code{.set} a global symbol, the value stored in the object
6727 file is the last value stored into it.
6730 On Z80 @code{set} is a real instruction, use
6731 @samp{@var{symbol} defl @var{expression}} instead.
6735 @section @code{.short @var{expressions}}
6737 @cindex @code{short} directive
6739 @code{.short} is normally the same as @samp{.word}.
6740 @xref{Word,,@code{.word}}.
6742 In some configurations, however, @code{.short} and @code{.word} generate
6743 numbers of different lengths. @xref{Machine Dependencies}.
6747 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6750 This expects zero or more @var{expressions}, and emits
6751 a 16 bit number for each.
6756 @section @code{.single @var{flonums}}
6758 @cindex @code{single} directive
6759 @cindex floating point numbers (single)
6760 This directive assembles zero or more flonums, separated by commas. It
6761 has the same effect as @code{.float}.
6763 The exact kind of floating point numbers emitted depends on how
6764 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6768 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6769 numbers in @sc{ieee} format.
6775 @section @code{.size}
6777 This directive is used to set the size associated with a symbol.
6781 @c only print the extra heading if both COFF and ELF are set
6782 @subheading COFF Version
6785 @cindex @code{size} directive (COFF version)
6786 For COFF targets, the @code{.size} directive is only permitted inside
6787 @code{.def}/@code{.endef} pairs. It is used like this:
6790 .size @var{expression}
6797 @c only print the extra heading if both COFF and ELF are set
6798 @subheading ELF Version
6801 @cindex @code{size} directive (ELF version)
6802 For ELF targets, the @code{.size} directive is used like this:
6805 .size @var{name} , @var{expression}
6808 This directive sets the size associated with a symbol @var{name}.
6809 The size in bytes is computed from @var{expression} which can make use of label
6810 arithmetic. This directive is typically used to set the size of function
6815 @ifclear no-space-dir
6817 @section @code{.skip @var{size} [,@var{fill}]}
6819 @cindex @code{skip} directive
6820 @cindex filling memory
6821 This directive emits @var{size} bytes, each of value @var{fill}. Both
6822 @var{size} and @var{fill} are absolute expressions. If the comma and
6823 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6828 @section @code{.sleb128 @var{expressions}}
6830 @cindex @code{sleb128} directive
6831 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6832 compact, variable length representation of numbers used by the DWARF
6833 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6835 @ifclear no-space-dir
6837 @section @code{.space @var{size} [,@var{fill}]}
6839 @cindex @code{space} directive
6840 @cindex filling memory
6841 This directive emits @var{size} bytes, each of value @var{fill}. Both
6842 @var{size} and @var{fill} are absolute expressions. If the comma
6843 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6848 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6849 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6850 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6851 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6859 @section @code{.stabd, .stabn, .stabs}
6861 @cindex symbolic debuggers, information for
6862 @cindex @code{stab@var{x}} directives
6863 There are three directives that begin @samp{.stab}.
6864 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6865 The symbols are not entered in the @command{@value{AS}} hash table: they
6866 cannot be referenced elsewhere in the source file.
6867 Up to five fields are required:
6871 This is the symbol's name. It may contain any character except
6872 @samp{\000}, so is more general than ordinary symbol names. Some
6873 debuggers used to code arbitrarily complex structures into symbol names
6877 An absolute expression. The symbol's type is set to the low 8 bits of
6878 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6879 and debuggers choke on silly bit patterns.
6882 An absolute expression. The symbol's ``other'' attribute is set to the
6883 low 8 bits of this expression.
6886 An absolute expression. The symbol's descriptor is set to the low 16
6887 bits of this expression.
6890 An absolute expression which becomes the symbol's value.
6893 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6894 or @code{.stabs} statement, the symbol has probably already been created;
6895 you get a half-formed symbol in your object file. This is
6896 compatible with earlier assemblers!
6899 @cindex @code{stabd} directive
6900 @item .stabd @var{type} , @var{other} , @var{desc}
6902 The ``name'' of the symbol generated is not even an empty string.
6903 It is a null pointer, for compatibility. Older assemblers used a
6904 null pointer so they didn't waste space in object files with empty
6907 The symbol's value is set to the location counter,
6908 relocatably. When your program is linked, the value of this symbol
6909 is the address of the location counter when the @code{.stabd} was
6912 @cindex @code{stabn} directive
6913 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6914 The name of the symbol is set to the empty string @code{""}.
6916 @cindex @code{stabs} directive
6917 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6918 All five fields are specified.
6924 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6925 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6927 @cindex string, copying to object file
6928 @cindex string8, copying to object file
6929 @cindex string16, copying to object file
6930 @cindex string32, copying to object file
6931 @cindex string64, copying to object file
6932 @cindex @code{string} directive
6933 @cindex @code{string8} directive
6934 @cindex @code{string16} directive
6935 @cindex @code{string32} directive
6936 @cindex @code{string64} directive
6938 Copy the characters in @var{str} to the object file. You may specify more than
6939 one string to copy, separated by commas. Unless otherwise specified for a
6940 particular machine, the assembler marks the end of each string with a 0 byte.
6941 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6943 The variants @code{string16}, @code{string32} and @code{string64} differ from
6944 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6945 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6946 are stored in target endianness byte order.
6952 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6953 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6958 @section @code{.struct @var{expression}}
6960 @cindex @code{struct} directive
6961 Switch to the absolute section, and set the section offset to @var{expression},
6962 which must be an absolute expression. You might use this as follows:
6971 This would define the symbol @code{field1} to have the value 0, the symbol
6972 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6973 value 8. Assembly would be left in the absolute section, and you would need to
6974 use a @code{.section} directive of some sort to change to some other section
6975 before further assembly.
6979 @section @code{.subsection @var{name}}
6981 @cindex @code{subsection} directive
6982 @cindex Section Stack
6983 This is one of the ELF section stack manipulation directives. The others are
6984 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6985 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6988 This directive replaces the current subsection with @code{name}. The current
6989 section is not changed. The replaced subsection is put onto the section stack
6990 in place of the then current top of stack subsection.
6995 @section @code{.symver}
6996 @cindex @code{symver} directive
6997 @cindex symbol versioning
6998 @cindex versions of symbols
6999 Use the @code{.symver} directive to bind symbols to specific version nodes
7000 within a source file. This is only supported on ELF platforms, and is
7001 typically used when assembling files to be linked into a shared library.
7002 There are cases where it may make sense to use this in objects to be bound
7003 into an application itself so as to override a versioned symbol from a
7006 For ELF targets, the @code{.symver} directive can be used like this:
7008 .symver @var{name}, @var{name2@@nodename}
7010 If the symbol @var{name} is defined within the file
7011 being assembled, the @code{.symver} directive effectively creates a symbol
7012 alias with the name @var{name2@@nodename}, and in fact the main reason that we
7013 just don't try and create a regular alias is that the @var{@@} character isn't
7014 permitted in symbol names. The @var{name2} part of the name is the actual name
7015 of the symbol by which it will be externally referenced. The name @var{name}
7016 itself is merely a name of convenience that is used so that it is possible to
7017 have definitions for multiple versions of a function within a single source
7018 file, and so that the compiler can unambiguously know which version of a
7019 function is being mentioned. The @var{nodename} portion of the alias should be
7020 the name of a node specified in the version script supplied to the linker when
7021 building a shared library. If you are attempting to override a versioned
7022 symbol from a shared library, then @var{nodename} should correspond to the
7023 nodename of the symbol you are trying to override.
7025 If the symbol @var{name} is not defined within the file being assembled, all
7026 references to @var{name} will be changed to @var{name2@@nodename}. If no
7027 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
7030 Another usage of the @code{.symver} directive is:
7032 .symver @var{name}, @var{name2@@@@nodename}
7034 In this case, the symbol @var{name} must exist and be defined within
7035 the file being assembled. It is similar to @var{name2@@nodename}. The
7036 difference is @var{name2@@@@nodename} will also be used to resolve
7037 references to @var{name2} by the linker.
7039 The third usage of the @code{.symver} directive is:
7041 .symver @var{name}, @var{name2@@@@@@nodename}
7043 When @var{name} is not defined within the
7044 file being assembled, it is treated as @var{name2@@nodename}. When
7045 @var{name} is defined within the file being assembled, the symbol
7046 name, @var{name}, will be changed to @var{name2@@@@nodename}.
7051 @section @code{.tag @var{structname}}
7053 @cindex COFF structure debugging
7054 @cindex structure debugging, COFF
7055 @cindex @code{tag} directive
7056 This directive is generated by compilers to include auxiliary debugging
7057 information in the symbol table. It is only permitted inside
7058 @code{.def}/@code{.endef} pairs. Tags are used to link structure
7059 definitions in the symbol table with instances of those structures.
7063 @section @code{.text @var{subsection}}
7065 @cindex @code{text} directive
7066 Tells @command{@value{AS}} to assemble the following statements onto the end of
7067 the text subsection numbered @var{subsection}, which is an absolute
7068 expression. If @var{subsection} is omitted, subsection number zero
7072 @section @code{.title "@var{heading}"}
7074 @cindex @code{title} directive
7075 @cindex listing control: title line
7076 Use @var{heading} as the title (second line, immediately after the
7077 source file name and pagenumber) when generating assembly listings.
7079 This directive affects subsequent pages, as well as the current page if
7080 it appears within ten lines of the top of a page.
7084 @section @code{.type}
7086 This directive is used to set the type of a symbol.
7090 @c only print the extra heading if both COFF and ELF are set
7091 @subheading COFF Version
7094 @cindex COFF symbol type
7095 @cindex symbol type, COFF
7096 @cindex @code{type} directive (COFF version)
7097 For COFF targets, this directive is permitted only within
7098 @code{.def}/@code{.endef} pairs. It is used like this:
7104 This records the integer @var{int} as the type attribute of a symbol table
7111 @c only print the extra heading if both COFF and ELF are set
7112 @subheading ELF Version
7115 @cindex ELF symbol type
7116 @cindex symbol type, ELF
7117 @cindex @code{type} directive (ELF version)
7118 For ELF targets, the @code{.type} directive is used like this:
7121 .type @var{name} , @var{type description}
7124 This sets the type of symbol @var{name} to be either a
7125 function symbol or an object symbol. There are five different syntaxes
7126 supported for the @var{type description} field, in order to provide
7127 compatibility with various other assemblers.
7129 Because some of the characters used in these syntaxes (such as @samp{@@} and
7130 @samp{#}) are comment characters for some architectures, some of the syntaxes
7131 below do not work on all architectures. The first variant will be accepted by
7132 the GNU assembler on all architectures so that variant should be used for
7133 maximum portability, if you do not need to assemble your code with other
7136 The syntaxes supported are:
7139 .type <name> STT_<TYPE_IN_UPPER_CASE>
7140 .type <name>,#<type>
7141 .type <name>,@@<type>
7142 .type <name>,%<type>
7143 .type <name>,"<type>"
7146 The types supported are:
7151 Mark the symbol as being a function name.
7154 @itemx gnu_indirect_function
7155 Mark the symbol as an indirect function when evaluated during reloc
7156 processing. (This is only supported on assemblers targeting GNU systems).
7160 Mark the symbol as being a data object.
7164 Mark the symbol as being a thread-local data object.
7168 Mark the symbol as being a common data object.
7172 Does not mark the symbol in any way. It is supported just for completeness.
7174 @item gnu_unique_object
7175 Marks the symbol as being a globally unique data object. The dynamic linker
7176 will make sure that in the entire process there is just one symbol with this
7177 name and type in use. (This is only supported on assemblers targeting GNU
7182 Note: Some targets support extra types in addition to those listed above.
7188 @section @code{.uleb128 @var{expressions}}
7190 @cindex @code{uleb128} directive
7191 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7192 compact, variable length representation of numbers used by the DWARF
7193 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7197 @section @code{.val @var{addr}}
7199 @cindex @code{val} directive
7200 @cindex COFF value attribute
7201 @cindex value attribute, COFF
7202 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7203 records the address @var{addr} as the value attribute of a symbol table
7209 @section @code{.version "@var{string}"}
7211 @cindex @code{version} directive
7212 This directive creates a @code{.note} section and places into it an ELF
7213 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7218 @section @code{.vtable_entry @var{table}, @var{offset}}
7220 @cindex @code{vtable_entry} directive
7221 This directive finds or creates a symbol @code{table} and creates a
7222 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7225 @section @code{.vtable_inherit @var{child}, @var{parent}}
7227 @cindex @code{vtable_inherit} directive
7228 This directive finds the symbol @code{child} and finds or creates the symbol
7229 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7230 parent whose addend is the value of the child symbol. As a special case the
7231 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7235 @section @code{.warning "@var{string}"}
7236 @cindex warning directive
7237 Similar to the directive @code{.error}
7238 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7241 @section @code{.weak @var{names}}
7243 @cindex @code{weak} directive
7244 This directive sets the weak attribute on the comma separated list of symbol
7245 @code{names}. If the symbols do not already exist, they will be created.
7247 On COFF targets other than PE, weak symbols are a GNU extension. This
7248 directive sets the weak attribute on the comma separated list of symbol
7249 @code{names}. If the symbols do not already exist, they will be created.
7251 On the PE target, weak symbols are supported natively as weak aliases.
7252 When a weak symbol is created that is not an alias, GAS creates an
7253 alternate symbol to hold the default value.
7256 @section @code{.weakref @var{alias}, @var{target}}
7258 @cindex @code{weakref} directive
7259 This directive creates an alias to the target symbol that enables the symbol to
7260 be referenced with weak-symbol semantics, but without actually making it weak.
7261 If direct references or definitions of the symbol are present, then the symbol
7262 will not be weak, but if all references to it are through weak references, the
7263 symbol will be marked as weak in the symbol table.
7265 The effect is equivalent to moving all references to the alias to a separate
7266 assembly source file, renaming the alias to the symbol in it, declaring the
7267 symbol as weak there, and running a reloadable link to merge the object files
7268 resulting from the assembly of the new source file and the old source file that
7269 had the references to the alias removed.
7271 The alias itself never makes to the symbol table, and is entirely handled
7272 within the assembler.
7275 @section @code{.word @var{expressions}}
7277 @cindex @code{word} directive
7278 This directive expects zero or more @var{expressions}, of any section,
7279 separated by commas.
7282 For each expression, @command{@value{AS}} emits a 32-bit number.
7285 For each expression, @command{@value{AS}} emits a 16-bit number.
7290 The size of the number emitted, and its byte order,
7291 depend on what target computer the assembly is for.
7294 @c on sparc the "special treatment to support compilers" doesn't
7295 @c happen---32-bit addressability, period; no long/short jumps.
7296 @ifset DIFF-TBL-KLUGE
7297 @cindex difference tables altered
7298 @cindex altered difference tables
7300 @emph{Warning: Special Treatment to support Compilers}
7304 Machines with a 32-bit address space, but that do less than 32-bit
7305 addressing, require the following special treatment. If the machine of
7306 interest to you does 32-bit addressing (or doesn't require it;
7307 @pxref{Machine Dependencies}), you can ignore this issue.
7310 In order to assemble compiler output into something that works,
7311 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7312 Directives of the form @samp{.word sym1-sym2} are often emitted by
7313 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7314 directive of the form @samp{.word sym1-sym2}, and the difference between
7315 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7316 creates a @dfn{secondary jump table}, immediately before the next label.
7317 This secondary jump table is preceded by a short-jump to the
7318 first byte after the secondary table. This short-jump prevents the flow
7319 of control from accidentally falling into the new table. Inside the
7320 table is a long-jump to @code{sym2}. The original @samp{.word}
7321 contains @code{sym1} minus the address of the long-jump to
7324 If there were several occurrences of @samp{.word sym1-sym2} before the
7325 secondary jump table, all of them are adjusted. If there was a
7326 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7327 long-jump to @code{sym4} is included in the secondary jump table,
7328 and the @code{.word} directives are adjusted to contain @code{sym3}
7329 minus the address of the long-jump to @code{sym4}; and so on, for as many
7330 entries in the original jump table as necessary.
7333 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7334 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7335 assembly language programmers.
7338 @c end DIFF-TBL-KLUGE
7340 @ifclear no-space-dir
7342 @section @code{.zero @var{size}}
7344 @cindex @code{zero} directive
7345 @cindex filling memory with zero bytes
7346 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7347 expression. This directive is actually an alias for the @samp{.skip} directive
7348 so in can take an optional second argument of the value to store in the bytes
7349 instead of zero. Using @samp{.zero} in this way would be confusing however.
7354 @section @code{.2byte @var{expression} [, @var{expression}]*}
7355 @cindex @code{2byte} directive
7356 @cindex two-byte integer
7357 @cindex integer, 2-byte
7359 This directive expects zero or more expressions, separated by commas. If there
7360 are no expressions then the directive does nothing. Otherwise each expression
7361 is evaluated in turn and placed in the next two bytes of the current output
7362 section, using the endian model of the target. If an expression will not fit
7363 in two bytes, a warning message is displayed and the least significant two
7364 bytes of the expression's value are used. If an expression cannot be evaluated
7365 at assembly time then relocations will be generated in order to compute the
7368 This directive does not apply any alignment before or after inserting the
7369 values. As a result of this, if relocations are generated, they may be
7370 different from those used for inserting values with a guaranteed alignment.
7372 This directive is only available for ELF targets,
7375 @section @code{.4byte @var{expression} [, @var{expression}]*}
7376 @cindex @code{4byte} directive
7377 @cindex four-byte integer
7378 @cindex integer, 4-byte
7380 Like the @option{.2byte} directive, except that it inserts unaligned, four byte
7381 long values into the output.
7384 @section @code{.8byte @var{expression} [, @var{expression}]*}
7385 @cindex @code{8byte} directive
7386 @cindex eight-byte integer
7387 @cindex integer, 8-byte
7389 Like the @option{.2byte} directive, except that it inserts unaligned, eight
7390 byte long bignum values into the output.
7395 @section Deprecated Directives
7397 @cindex deprecated directives
7398 @cindex obsolescent directives
7399 One day these directives won't work.
7400 They are included for compatibility with older assemblers.
7407 @node Object Attributes
7408 @chapter Object Attributes
7409 @cindex object attributes
7411 @command{@value{AS}} assembles source files written for a specific architecture
7412 into object files for that architecture. But not all object files are alike.
7413 Many architectures support incompatible variations. For instance, floating
7414 point arguments might be passed in floating point registers if the object file
7415 requires hardware floating point support---or floating point arguments might be
7416 passed in integer registers if the object file supports processors with no
7417 hardware floating point unit. Or, if two objects are built for different
7418 generations of the same architecture, the combination may require the
7419 newer generation at run-time.
7421 This information is useful during and after linking. At link time,
7422 @command{@value{LD}} can warn about incompatible object files. After link
7423 time, tools like @command{gdb} can use it to process the linked file
7426 Compatibility information is recorded as a series of object attributes. Each
7427 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7428 string, and indicates who sets the meaning of the tag. The tag is an integer,
7429 and indicates what property the attribute describes. The value may be a string
7430 or an integer, and indicates how the property affects this object. Missing
7431 attributes are the same as attributes with a zero value or empty string value.
7433 Object attributes were developed as part of the ABI for the ARM Architecture.
7434 The file format is documented in @cite{ELF for the ARM Architecture}.
7437 * GNU Object Attributes:: @sc{gnu} Object Attributes
7438 * Defining New Object Attributes:: Defining New Object Attributes
7441 @node GNU Object Attributes
7442 @section @sc{gnu} Object Attributes
7444 The @code{.gnu_attribute} directive records an object attribute
7445 with vendor @samp{gnu}.
7447 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7448 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7449 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7450 2} is set for architecture-independent attributes and clear for
7451 architecture-dependent ones.
7453 @subsection Common @sc{gnu} attributes
7455 These attributes are valid on all architectures.
7458 @item Tag_compatibility (32)
7459 The compatibility attribute takes an integer flag value and a vendor name. If
7460 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7461 then the file is only compatible with the named toolchain. If it is greater
7462 than 1, the file can only be processed by other toolchains under some private
7463 arrangement indicated by the flag value and the vendor name.
7466 @subsection MIPS Attributes
7469 @item Tag_GNU_MIPS_ABI_FP (4)
7470 The floating-point ABI used by this object file. The value will be:
7474 0 for files not affected by the floating-point ABI.
7476 1 for files using the hardware floating-point ABI with a standard
7477 double-precision FPU.
7479 2 for files using the hardware floating-point ABI with a single-precision FPU.
7481 3 for files using the software floating-point ABI.
7483 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7484 floating-point registers, 32-bit general-purpose registers and increased the
7485 number of callee-saved floating-point registers.
7487 5 for files using the hardware floating-point ABI with a double-precision FPU
7488 with either 32-bit or 64-bit floating-point registers and 32-bit
7489 general-purpose registers.
7491 6 for files using the hardware floating-point ABI with 64-bit floating-point
7492 registers and 32-bit general-purpose registers.
7494 7 for files using the hardware floating-point ABI with 64-bit floating-point
7495 registers, 32-bit general-purpose registers and a rule that forbids the
7496 direct use of odd-numbered single-precision floating-point registers.
7500 @subsection PowerPC Attributes
7503 @item Tag_GNU_Power_ABI_FP (4)
7504 The floating-point ABI used by this object file. The value will be:
7508 0 for files not affected by the floating-point ABI.
7510 1 for files using double-precision hardware floating-point ABI.
7512 2 for files using the software floating-point ABI.
7514 3 for files using single-precision hardware floating-point ABI.
7517 @item Tag_GNU_Power_ABI_Vector (8)
7518 The vector ABI used by this object file. The value will be:
7522 0 for files not affected by the vector ABI.
7524 1 for files using general purpose registers to pass vectors.
7526 2 for files using AltiVec registers to pass vectors.
7528 3 for files using SPE registers to pass vectors.
7532 @subsection IBM z Systems Attributes
7535 @item Tag_GNU_S390_ABI_Vector (8)
7536 The vector ABI used by this object file. The value will be:
7540 0 for files not affected by the vector ABI.
7542 1 for files using software vector ABI.
7544 2 for files using hardware vector ABI.
7548 @node Defining New Object Attributes
7549 @section Defining New Object Attributes
7551 If you want to define a new @sc{gnu} object attribute, here are the places you
7552 will need to modify. New attributes should be discussed on the @samp{binutils}
7557 This manual, which is the official register of attributes.
7559 The header for your architecture @file{include/elf}, to define the tag.
7561 The @file{bfd} support file for your architecture, to merge the attribute
7562 and issue any appropriate link warnings.
7564 Test cases in @file{ld/testsuite} for merging and link warnings.
7566 @file{binutils/readelf.c} to display your attribute.
7568 GCC, if you want the compiler to mark the attribute automatically.
7574 @node Machine Dependencies
7575 @chapter Machine Dependent Features
7577 @cindex machine dependencies
7578 The machine instruction sets are (almost by definition) different on
7579 each machine where @command{@value{AS}} runs. Floating point representations
7580 vary as well, and @command{@value{AS}} often supports a few additional
7581 directives or command-line options for compatibility with other
7582 assemblers on a particular platform. Finally, some versions of
7583 @command{@value{AS}} support special pseudo-instructions for branch
7586 This chapter discusses most of these differences, though it does not
7587 include details on any machine's instruction set. For details on that
7588 subject, see the hardware manufacturer's manual.
7592 * AArch64-Dependent:: AArch64 Dependent Features
7595 * Alpha-Dependent:: Alpha Dependent Features
7598 * ARC-Dependent:: ARC Dependent Features
7601 * ARM-Dependent:: ARM Dependent Features
7604 * AVR-Dependent:: AVR Dependent Features
7607 * Blackfin-Dependent:: Blackfin Dependent Features
7610 * CR16-Dependent:: CR16 Dependent Features
7613 * CRIS-Dependent:: CRIS Dependent Features
7616 * C-SKY-Dependent:: C-SKY Dependent Features
7619 * D10V-Dependent:: D10V Dependent Features
7622 * D30V-Dependent:: D30V Dependent Features
7625 * Epiphany-Dependent:: EPIPHANY Dependent Features
7628 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7631 * HPPA-Dependent:: HPPA Dependent Features
7634 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7637 * IA-64-Dependent:: Intel IA-64 Dependent Features
7640 * IP2K-Dependent:: IP2K Dependent Features
7643 * LM32-Dependent:: LM32 Dependent Features
7646 * M32C-Dependent:: M32C Dependent Features
7649 * M32R-Dependent:: M32R Dependent Features
7652 * M68K-Dependent:: M680x0 Dependent Features
7655 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7658 * S12Z-Dependent:: S12Z Dependent Features
7661 * Meta-Dependent :: Meta Dependent Features
7664 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7667 * MIPS-Dependent:: MIPS Dependent Features
7670 * MMIX-Dependent:: MMIX Dependent Features
7673 * MSP430-Dependent:: MSP430 Dependent Features
7676 * NDS32-Dependent:: Andes NDS32 Dependent Features
7679 * NiosII-Dependent:: Altera Nios II Dependent Features
7682 * NS32K-Dependent:: NS32K Dependent Features
7685 * OpenRISC-Dependent:: OpenRISC 1000 Features
7688 * PDP-11-Dependent:: PDP-11 Dependent Features
7691 * PJ-Dependent:: picoJava Dependent Features
7694 * PPC-Dependent:: PowerPC Dependent Features
7697 * PRU-Dependent:: PRU Dependent Features
7700 * RISC-V-Dependent:: RISC-V Dependent Features
7703 * RL78-Dependent:: RL78 Dependent Features
7706 * RX-Dependent:: RX Dependent Features
7709 * S/390-Dependent:: IBM S/390 Dependent Features
7712 * SCORE-Dependent:: SCORE Dependent Features
7715 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7718 * Sparc-Dependent:: SPARC Dependent Features
7721 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7724 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7727 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7730 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7733 * V850-Dependent:: V850 Dependent Features
7736 * Vax-Dependent:: VAX Dependent Features
7739 * Visium-Dependent:: Visium Dependent Features
7742 * WebAssembly-Dependent:: WebAssembly Dependent Features
7745 * XGATE-Dependent:: XGATE Dependent Features
7748 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7751 * Xtensa-Dependent:: Xtensa Dependent Features
7754 * Z80-Dependent:: Z80 Dependent Features
7757 * Z8000-Dependent:: Z8000 Dependent Features
7764 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7765 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7766 @c peculiarity: to preserve cross-references, there must be a node called
7767 @c "Machine Dependencies". Hence the conditional nodenames in each
7768 @c major node below. Node defaulting in makeinfo requires adjacency of
7769 @c node and sectioning commands; hence the repetition of @chapter BLAH
7770 @c in both conditional blocks.
7773 @include c-aarch64.texi
7777 @include c-alpha.texi
7793 @include c-bfin.texi
7797 @include c-cr16.texi
7801 @include c-cris.texi
7805 @include c-csky.texi
7810 @node Machine Dependencies
7811 @chapter Machine Dependent Features
7813 The machine instruction sets are different on each Renesas chip family,
7814 and there are also some syntax differences among the families. This
7815 chapter describes the specific @command{@value{AS}} features for each
7819 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7820 * SH-Dependent:: Renesas SH Dependent Features
7827 @include c-d10v.texi
7831 @include c-d30v.texi
7835 @include c-epiphany.texi
7839 @include c-h8300.texi
7843 @include c-hppa.texi
7847 @include c-i386.texi
7851 @include c-ia64.texi
7855 @include c-ip2k.texi
7859 @include c-lm32.texi
7863 @include c-m32c.texi
7867 @include c-m32r.texi
7871 @include c-m68k.texi
7875 @include c-m68hc11.texi
7879 @include c-s12z.texi
7883 @include c-metag.texi
7887 @include c-microblaze.texi
7891 @include c-mips.texi
7895 @include c-mmix.texi
7899 @include c-msp430.texi
7903 @include c-nds32.texi
7907 @include c-nios2.texi
7911 @include c-ns32k.texi
7915 @include c-or1k.texi
7919 @include c-pdp11.texi
7935 @include c-riscv.texi
7939 @include c-rl78.texi
7947 @include c-s390.texi
7951 @include c-score.texi
7959 @include c-sparc.texi
7963 @include c-tic54x.texi
7967 @include c-tic6x.texi
7971 @include c-tilegx.texi
7975 @include c-tilepro.texi
7979 @include c-v850.texi
7987 @include c-visium.texi
7991 @include c-wasm32.texi
7995 @include c-xgate.texi
7999 @include c-xstormy16.texi
8003 @include c-xtensa.texi
8015 @c reverse effect of @down at top of generic Machine-Dep chapter
8019 @node Reporting Bugs
8020 @chapter Reporting Bugs
8021 @cindex bugs in assembler
8022 @cindex reporting bugs in assembler
8024 Your bug reports play an essential role in making @command{@value{AS}} reliable.
8026 Reporting a bug may help you by bringing a solution to your problem, or it may
8027 not. But in any case the principal function of a bug report is to help the
8028 entire community by making the next version of @command{@value{AS}} work better.
8029 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
8031 In order for a bug report to serve its purpose, you must include the
8032 information that enables us to fix the bug.
8035 * Bug Criteria:: Have you found a bug?
8036 * Bug Reporting:: How to report bugs
8040 @section Have You Found a Bug?
8041 @cindex bug criteria
8043 If you are not sure whether you have found a bug, here are some guidelines:
8046 @cindex fatal signal
8047 @cindex assembler crash
8048 @cindex crash of assembler
8050 If the assembler gets a fatal signal, for any input whatever, that is a
8051 @command{@value{AS}} bug. Reliable assemblers never crash.
8053 @cindex error on valid input
8055 If @command{@value{AS}} produces an error message for valid input, that is a bug.
8057 @cindex invalid input
8059 If @command{@value{AS}} does not produce an error message for invalid input, that
8060 is a bug. However, you should note that your idea of ``invalid input'' might
8061 be our idea of ``an extension'' or ``support for traditional practice''.
8064 If you are an experienced user of assemblers, your suggestions for improvement
8065 of @command{@value{AS}} are welcome in any case.
8069 @section How to Report Bugs
8071 @cindex assembler bugs, reporting
8073 A number of companies and individuals offer support for @sc{gnu} products. If
8074 you obtained @command{@value{AS}} from a support organization, we recommend you
8075 contact that organization first.
8077 You can find contact information for many support companies and
8078 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
8082 In any event, we also recommend that you send bug reports for @command{@value{AS}}
8086 The fundamental principle of reporting bugs usefully is this:
8087 @strong{report all the facts}. If you are not sure whether to state a
8088 fact or leave it out, state it!
8090 Often people omit facts because they think they know what causes the problem
8091 and assume that some details do not matter. Thus, you might assume that the
8092 name of a symbol you use in an example does not matter. Well, probably it does
8093 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
8094 happens to fetch from the location where that name is stored in memory;
8095 perhaps, if the name were different, the contents of that location would fool
8096 the assembler into doing the right thing despite the bug. Play it safe and
8097 give a specific, complete example. That is the easiest thing for you to do,
8098 and the most helpful.
8100 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
8101 it is new to us. Therefore, always write your bug reports on the assumption
8102 that the bug has not been reported previously.
8104 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8105 bell?'' This cannot help us fix a bug, so it is basically useless. We
8106 respond by asking for enough details to enable us to investigate.
8107 You might as well expedite matters by sending them to begin with.
8109 To enable us to fix the bug, you should include all these things:
8113 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8114 it with the @samp{--version} argument.
8116 Without this, we will not know whether there is any point in looking for
8117 the bug in the current version of @command{@value{AS}}.
8120 Any patches you may have applied to the @command{@value{AS}} source.
8123 The type of machine you are using, and the operating system name and
8127 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8131 The command arguments you gave the assembler to assemble your example and
8132 observe the bug. To guarantee you will not omit something important, list them
8133 all. A copy of the Makefile (or the output from make) is sufficient.
8135 If we were to try to guess the arguments, we would probably guess wrong
8136 and then we might not encounter the bug.
8139 A complete input file that will reproduce the bug. If the bug is observed when
8140 the assembler is invoked via a compiler, send the assembler source, not the
8141 high level language source. Most compilers will produce the assembler source
8142 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8143 the options @samp{-v --save-temps}; this will save the assembler source in a
8144 file with an extension of @file{.s}, and also show you exactly how
8145 @command{@value{AS}} is being run.
8148 A description of what behavior you observe that you believe is
8149 incorrect. For example, ``It gets a fatal signal.''
8151 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8152 will certainly notice it. But if the bug is incorrect output, we might not
8153 notice unless it is glaringly wrong. You might as well not give us a chance to
8156 Even if the problem you experience is a fatal signal, you should still say so
8157 explicitly. Suppose something strange is going on, such as, your copy of
8158 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8159 library on your system. (This has happened!) Your copy might crash and ours
8160 would not. If you told us to expect a crash, then when ours fails to crash, we
8161 would know that the bug was not happening for us. If you had not told us to
8162 expect a crash, then we would not be able to draw any conclusion from our
8166 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8167 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8168 option. Always send diffs from the old file to the new file. If you even
8169 discuss something in the @command{@value{AS}} source, refer to it by context, not
8172 The line numbers in our development sources will not match those in your
8173 sources. Your line numbers would convey no useful information to us.
8176 Here are some things that are not necessary:
8180 A description of the envelope of the bug.
8182 Often people who encounter a bug spend a lot of time investigating
8183 which changes to the input file will make the bug go away and which
8184 changes will not affect it.
8186 This is often time consuming and not very useful, because the way we
8187 will find the bug is by running a single example under the debugger
8188 with breakpoints, not by pure deduction from a series of examples.
8189 We recommend that you save your time for something else.
8191 Of course, if you can find a simpler example to report @emph{instead}
8192 of the original one, that is a convenience for us. Errors in the
8193 output will be easier to spot, running under the debugger will take
8194 less time, and so on.
8196 However, simplification is not vital; if you do not want to do this,
8197 report the bug anyway and send us the entire test case you used.
8200 A patch for the bug.
8202 A patch for the bug does help us if it is a good one. But do not omit
8203 the necessary information, such as the test case, on the assumption that
8204 a patch is all we need. We might see problems with your patch and decide
8205 to fix the problem another way, or we might not understand it at all.
8207 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8208 construct an example that will make the program follow a certain path through
8209 the code. If you do not send us the example, we will not be able to construct
8210 one, so we will not be able to verify that the bug is fixed.
8212 And if we cannot understand what bug you are trying to fix, or why your
8213 patch should be an improvement, we will not install it. A test case will
8214 help us to understand.
8217 A guess about what the bug is or what it depends on.
8219 Such guesses are usually wrong. Even we cannot guess right about such
8220 things without first using the debugger to find the facts.
8223 @node Acknowledgements
8224 @chapter Acknowledgements
8226 If you have contributed to GAS and your name isn't listed here,
8227 it is not meant as a slight. We just don't know about it. Send mail to the
8228 maintainer, and we'll correct the situation. Currently
8230 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8232 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8235 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8236 information and the 68k series machines, most of the preprocessing pass, and
8237 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8239 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8240 many bug fixes, including merging support for several processors, breaking GAS
8241 up to handle multiple object file format back ends (including heavy rewrite,
8242 testing, an integration of the coff and b.out back ends), adding configuration
8243 including heavy testing and verification of cross assemblers and file splits
8244 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8245 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8246 port (including considerable amounts of reverse engineering), a SPARC opcode
8247 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8248 assertions and made them work, much other reorganization, cleanup, and lint.
8250 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8251 in format-specific I/O modules.
8253 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8254 has done much work with it since.
8256 The Intel 80386 machine description was written by Eliot Dresselhaus.
8258 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8260 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8261 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8263 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8264 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8265 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8266 support a.out format.
8268 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8269 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8270 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8271 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8274 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8275 simplified the configuration of which versions accept which directives. He
8276 updated the 68k machine description so that Motorola's opcodes always produced
8277 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8278 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8279 cross-compilation support, and one bug in relaxation that took a week and
8280 required the proverbial one-bit fix.
8282 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8283 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8284 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8285 PowerPC assembler, and made a few other minor patches.
8287 Steve Chamberlain made GAS able to generate listings.
8289 Hewlett-Packard contributed support for the HP9000/300.
8291 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8292 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8293 formats). This work was supported by both the Center for Software Science at
8294 the University of Utah and Cygnus Support.
8296 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8297 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8298 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8299 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8300 and some initial 64-bit support).
8302 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8304 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8305 support for openVMS/Alpha.
8307 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8310 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8311 Inc.@: added support for Xtensa processors.
8313 Several engineers at Cygnus Support have also provided many small bug fixes and
8314 configuration enhancements.
8316 Jon Beniston added support for the Lattice Mico32 architecture.
8318 Many others have contributed large or small bugfixes and enhancements. If
8319 you have contributed significant work and are not mentioned on this list, and
8320 want to be, let us know. Some of the history has been lost; we are not
8321 intentionally leaving anyone out.
8323 @node GNU Free Documentation License
8324 @appendix GNU Free Documentation License
8328 @unnumbered AS Index