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{-minsn32}] [@b{-mno-insn32}]
449 [@b{-mfix7000}] [@b{-mno-fix7000}]
450 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
451 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
452 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
453 [@b{-mdebug}] [@b{-no-mdebug}]
454 [@b{-mpdr}] [@b{-mno-pdr}]
458 @emph{Target MMIX options:}
459 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
460 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
461 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
462 [@b{--linker-allocated-gregs}]
466 @emph{Target Nios II options:}
467 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
472 @emph{Target NDS32 options:}
473 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
474 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
475 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
476 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
477 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
478 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
479 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
484 @emph{Target PDP11 options:}
485 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
486 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
487 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
491 @emph{Target picoJava options:}
496 @emph{Target PowerPC options:}
498 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
499 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mgekko}|
500 @b{-mbroadway}|@b{-mppc64}|@b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|
501 @b{-me6500}|@b{-mppc64bridge}|@b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|
502 @b{-mpower6}|@b{-mpwr6}|@b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-mpower9}|@b{-mpwr9}@b{-ma2}|
503 @b{-mcell}|@b{-mspe}|@b{-mspe2}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
504 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
505 [@b{-mregnames}|@b{-mno-regnames}]
506 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
507 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
508 [@b{-msolaris}|@b{-mno-solaris}]
509 [@b{-nops=@var{count}}]
513 @emph{Target PRU options:}
516 [@b{-mno-warn-regname-label}]
520 @emph{Target RISC-V options:}
521 [@b{-fpic}|@b{-fPIC}|@b{-fno-pic}]
522 [@b{-march}=@var{ISA}]
523 [@b{-mabi}=@var{ABI}]
527 @emph{Target RL78 options:}
529 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
533 @emph{Target RX options:}
534 [@b{-mlittle-endian}|@b{-mbig-endian}]
535 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
536 [@b{-muse-conventional-section-names}]
537 [@b{-msmall-data-limit}]
540 [@b{-mint-register=@var{number}}]
541 [@b{-mgcc-abi}|@b{-mrx-abi}]
545 @emph{Target s390 options:}
546 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
547 [@b{-mregnames}|@b{-mno-regnames}]
548 [@b{-mwarn-areg-zero}]
552 @emph{Target SCORE options:}
553 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
554 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
555 [@b{-march=score7}][@b{-march=score3}]
556 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
560 @emph{Target SPARC options:}
561 @c The order here is important. See c-sparc.texi.
562 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Aleon}|@b{-Asparclet}|@b{-Asparclite}
563 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av8plusb}|@b{-Av8plusc}|@b{-Av8plusd}
564 @b{-Av8plusv}|@b{-Av8plusm}|@b{-Av9}|@b{-Av9a}|@b{-Av9b}|@b{-Av9c}
565 @b{-Av9d}|@b{-Av9e}|@b{-Av9v}|@b{-Av9m}|@b{-Asparc}|@b{-Asparcvis}
566 @b{-Asparcvis2}|@b{-Asparcfmaf}|@b{-Asparcima}|@b{-Asparcvis3}
567 @b{-Asparcvisr}|@b{-Asparc5}]
568 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}]|@b{-xarch=v8plusb}|@b{-xarch=v8plusc}
569 @b{-xarch=v8plusd}|@b{-xarch=v8plusv}|@b{-xarch=v8plusm}|@b{-xarch=v9}
570 @b{-xarch=v9a}|@b{-xarch=v9b}|@b{-xarch=v9c}|@b{-xarch=v9d}|@b{-xarch=v9e}
571 @b{-xarch=v9v}|@b{-xarch=v9m}|@b{-xarch=sparc}|@b{-xarch=sparcvis}
572 @b{-xarch=sparcvis2}|@b{-xarch=sparcfmaf}|@b{-xarch=sparcima}
573 @b{-xarch=sparcvis3}|@b{-xarch=sparcvisr}|@b{-xarch=sparc5}
576 [@b{--enforce-aligned-data}][@b{--dcti-couples-detect}]
580 @emph{Target TIC54X options:}
581 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
582 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
586 @emph{Target TIC6X options:}
587 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
588 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
589 [@b{-mpic}|@b{-mno-pic}]
593 @emph{Target TILE-Gx options:}
594 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
597 @c TILEPro has no machine-dependent assembler options
601 @emph{Target Visium options:}
602 [@b{-mtune=@var{arch}}]
606 @emph{Target Xtensa options:}
607 [@b{--[no-]text-section-literals}] [@b{--[no-]auto-litpools}]
608 [@b{--[no-]absolute-literals}]
609 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
610 [@b{--[no-]transform}]
611 [@b{--rename-section} @var{oldname}=@var{newname}]
612 [@b{--[no-]trampolines}]
616 @emph{Target Z80 options:}
617 [@b{-z80}] [@b{-r800}]
618 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
619 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
620 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
621 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
622 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
623 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
627 @c Z8000 has no machine-dependent assembler options
636 @include at-file.texi
639 Turn on listings, in any of a variety of ways:
643 omit false conditionals
646 omit debugging directives
649 include general information, like @value{AS} version and options passed
652 include high-level source
658 include macro expansions
661 omit forms processing
667 set the name of the listing file
670 You may combine these options; for example, use @samp{-aln} for assembly
671 listing without forms processing. The @samp{=file} option, if used, must be
672 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
675 Begin in alternate macro mode.
677 @xref{Altmacro,,@code{.altmacro}}.
680 @item --compress-debug-sections
681 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
682 ELF ABI. The resulting object file may not be compatible with older
683 linkers and object file utilities. Note if compression would make a
684 given section @emph{larger} then it is not compressed.
687 @cindex @samp{--compress-debug-sections=} option
688 @item --compress-debug-sections=none
689 @itemx --compress-debug-sections=zlib
690 @itemx --compress-debug-sections=zlib-gnu
691 @itemx --compress-debug-sections=zlib-gabi
692 These options control how DWARF debug sections are compressed.
693 @option{--compress-debug-sections=none} is equivalent to
694 @option{--nocompress-debug-sections}.
695 @option{--compress-debug-sections=zlib} and
696 @option{--compress-debug-sections=zlib-gabi} are equivalent to
697 @option{--compress-debug-sections}.
698 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
699 sections using zlib. The debug sections are renamed to begin with
700 @samp{.zdebug}. Note if compression would make a given section
701 @emph{larger} then it is not compressed nor renamed.
705 @item --nocompress-debug-sections
706 Do not compress DWARF debug sections. This is usually the default for all
707 targets except the x86/x86_64, but a configure time option can be used to
711 Ignored. This option is accepted for script compatibility with calls to
714 @item --debug-prefix-map @var{old}=@var{new}
715 When assembling files in directory @file{@var{old}}, record debugging
716 information describing them as in @file{@var{new}} instead.
718 @item --defsym @var{sym}=@var{value}
719 Define the symbol @var{sym} to be @var{value} before assembling the input file.
720 @var{value} must be an integer constant. As in C, a leading @samp{0x}
721 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
722 value. The value of the symbol can be overridden inside a source file via the
723 use of a @code{.set} pseudo-op.
726 ``fast''---skip whitespace and comment preprocessing (assume source is
731 Generate debugging information for each assembler source line using whichever
732 debug format is preferred by the target. This currently means either STABS,
736 Generate stabs debugging information for each assembler line. This
737 may help debugging assembler code, if the debugger can handle it.
740 Generate stabs debugging information for each assembler line, with GNU
741 extensions that probably only gdb can handle, and that could make other
742 debuggers crash or refuse to read your program. This
743 may help debugging assembler code. Currently the only GNU extension is
744 the location of the current working directory at assembling time.
747 Generate DWARF2 debugging information for each assembler line. This
748 may help debugging assembler code, if the debugger can handle it. Note---this
749 option is only supported by some targets, not all of them.
751 @item --gdwarf-sections
752 Instead of creating a .debug_line section, create a series of
753 .debug_line.@var{foo} sections where @var{foo} is the name of the
754 corresponding code section. For example a code section called @var{.text.func}
755 will have its dwarf line number information placed into a section called
756 @var{.debug_line.text.func}. If the code section is just called @var{.text}
757 then debug line section will still be called just @var{.debug_line} without any
761 @item --size-check=error
762 @itemx --size-check=warning
763 Issue an error or warning for invalid ELF .size directive.
765 @item --elf-stt-common=no
766 @itemx --elf-stt-common=yes
767 These options control whether the ELF assembler should generate common
768 symbols with the @code{STT_COMMON} type. The default can be controlled
769 by a configure option @option{--enable-elf-stt-common}.
771 @item --generate-missing-build-notes=yes
772 @itemx --generate-missing-build-notes=no
773 These options control whether the ELF assembler should generate GNU Build
774 attribute notes if none are present in the input sources.
775 The default can be controlled by the @option{--enable-generate-build-notes}
781 Print a summary of the command-line options and exit.
784 Print a summary of all target specific options and exit.
787 Add directory @var{dir} to the search list for @code{.include} directives.
790 Don't warn about signed overflow.
793 @ifclear DIFF-TBL-KLUGE
794 This option is accepted but has no effect on the @value{TARGET} family.
796 @ifset DIFF-TBL-KLUGE
797 Issue warnings when difference tables altered for long displacements.
802 Keep (in the symbol table) local symbols. These symbols start with
803 system-specific local label prefixes, typically @samp{.L} for ELF systems
804 or @samp{L} for traditional a.out systems.
809 @item --listing-lhs-width=@var{number}
810 Set the maximum width, in words, of the output data column for an assembler
811 listing to @var{number}.
813 @item --listing-lhs-width2=@var{number}
814 Set the maximum width, in words, of the output data column for continuation
815 lines in an assembler listing to @var{number}.
817 @item --listing-rhs-width=@var{number}
818 Set the maximum width of an input source line, as displayed in a listing, to
821 @item --listing-cont-lines=@var{number}
822 Set the maximum number of lines printed in a listing for a single line of input
825 @item --no-pad-sections
826 Stop the assembler for padding the ends of output sections to the alignment
827 of that section. The default is to pad the sections, but this can waste space
828 which might be needed on targets which have tight memory constraints.
830 @item -o @var{objfile}
831 Name the object-file output from @command{@value{AS}} @var{objfile}.
834 Fold the data section into the text section.
836 @item --hash-size=@var{number}
837 Set the default size of GAS's hash tables to a prime number close to
838 @var{number}. Increasing this value can reduce the length of time it takes the
839 assembler to perform its tasks, at the expense of increasing the assembler's
840 memory requirements. Similarly reducing this value can reduce the memory
841 requirements at the expense of speed.
843 @item --reduce-memory-overheads
844 This option reduces GAS's memory requirements, at the expense of making the
845 assembly processes slower. Currently this switch is a synonym for
846 @samp{--hash-size=4051}, but in the future it may have other effects as well.
849 @item --sectname-subst
850 Honor substitution sequences in section names.
852 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
857 Print the maximum space (in bytes) and total time (in seconds) used by
860 @item --strip-local-absolute
861 Remove local absolute symbols from the outgoing symbol table.
865 Print the @command{as} version.
868 Print the @command{as} version and exit.
872 Suppress warning messages.
874 @item --fatal-warnings
875 Treat warnings as errors.
878 Don't suppress warning messages or treat them as errors.
887 Generate an object file even after errors.
889 @item -- | @var{files} @dots{}
890 Standard input, or source files to assemble.
898 @xref{AArch64 Options}, for the options available when @value{AS} is configured
899 for the 64-bit mode of the ARM Architecture (AArch64).
904 The following options are available when @value{AS} is configured for the
905 64-bit mode of the ARM Architecture (AArch64).
908 @include c-aarch64.texi
909 @c ended inside the included file
917 @xref{Alpha Options}, for the options available when @value{AS} is configured
918 for an Alpha processor.
923 The following options are available when @value{AS} is configured for an Alpha
927 @include c-alpha.texi
928 @c ended inside the included file
935 The following options are available when @value{AS} is configured for an ARC
939 @item -mcpu=@var{cpu}
940 This option selects the core processor variant.
942 Select either big-endian (-EB) or little-endian (-EL) output.
944 Enable Code Density extenssion instructions.
949 The following options are available when @value{AS} is configured for the ARM
953 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
954 Specify which ARM processor variant is the target.
955 @item -march=@var{architecture}[+@var{extension}@dots{}]
956 Specify which ARM architecture variant is used by the target.
957 @item -mfpu=@var{floating-point-format}
958 Select which Floating Point architecture is the target.
959 @item -mfloat-abi=@var{abi}
960 Select which floating point ABI is in use.
962 Enable Thumb only instruction decoding.
963 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
964 Select which procedure calling convention is in use.
966 Select either big-endian (-EB) or little-endian (-EL) output.
967 @item -mthumb-interwork
968 Specify that the code has been generated with interworking between Thumb and
971 Turns on CodeComposer Studio assembly syntax compatibility mode.
973 Specify that PIC code has been generated.
981 @xref{Blackfin Options}, for the options available when @value{AS} is
982 configured for the Blackfin processor family.
987 The following options are available when @value{AS} is configured for
988 the Blackfin processor family.
992 @c ended inside the included file
999 See the info pages for documentation of the CRIS-specific options.
1005 @xref{C-SKY Options}, for the options available when @value{AS} is
1006 configured for the C-SKY processor family.
1010 @c man begin OPTIONS
1011 The following options are available when @value{AS} is configured for
1012 the C-SKY processor family.
1014 @c man begin INCLUDE
1015 @include c-csky.texi
1016 @c ended inside the included file
1022 The following options are available when @value{AS} is configured for
1025 @cindex D10V optimization
1026 @cindex optimization, D10V
1028 Optimize output by parallelizing instructions.
1033 The following options are available when @value{AS} is configured for a D30V
1036 @cindex D30V optimization
1037 @cindex optimization, D30V
1039 Optimize output by parallelizing instructions.
1043 Warn when nops are generated.
1045 @cindex D30V nops after 32-bit multiply
1047 Warn when a nop after a 32-bit multiply instruction is generated.
1053 The following options are available when @value{AS} is configured for the
1054 Adapteva EPIPHANY series.
1057 @xref{Epiphany Options}, for the options available when @value{AS} is
1058 configured for an Epiphany processor.
1062 @c man begin OPTIONS
1063 The following options are available when @value{AS} is configured for
1064 an Epiphany processor.
1066 @c man begin INCLUDE
1067 @include c-epiphany.texi
1068 @c ended inside the included file
1076 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1077 for an H8/300 processor.
1081 @c man begin OPTIONS
1082 The following options are available when @value{AS} is configured for an H8/300
1085 @c man begin INCLUDE
1086 @include c-h8300.texi
1087 @c ended inside the included file
1095 @xref{i386-Options}, for the options available when @value{AS} is
1096 configured for an i386 processor.
1100 @c man begin OPTIONS
1101 The following options are available when @value{AS} is configured for
1104 @c man begin INCLUDE
1105 @include c-i386.texi
1106 @c ended inside the included file
1111 @c man begin OPTIONS
1113 The following options are available when @value{AS} is configured for the
1119 Specifies that the extended IP2022 instructions are allowed.
1122 Restores the default behaviour, which restricts the permitted instructions to
1123 just the basic IP2022 ones.
1129 The following options are available when @value{AS} is configured for the
1130 Renesas M32C and M16C processors.
1135 Assemble M32C instructions.
1138 Assemble M16C instructions (the default).
1141 Enable support for link-time relaxations.
1144 Support H'00 style hex constants in addition to 0x00 style.
1150 The following options are available when @value{AS} is configured for the
1151 Renesas M32R (formerly Mitsubishi M32R) series.
1156 Specify which processor in the M32R family is the target. The default
1157 is normally the M32R, but this option changes it to the M32RX.
1159 @item --warn-explicit-parallel-conflicts or --Wp
1160 Produce warning messages when questionable parallel constructs are
1163 @item --no-warn-explicit-parallel-conflicts or --Wnp
1164 Do not produce warning messages when questionable parallel constructs are
1171 The following options are available when @value{AS} is configured for the
1172 Motorola 68000 series.
1177 Shorten references to undefined symbols, to one word instead of two.
1179 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1180 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1181 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1182 Specify what processor in the 68000 family is the target. The default
1183 is normally the 68020, but this can be changed at configuration time.
1185 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1186 The target machine does (or does not) have a floating-point coprocessor.
1187 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1188 the basic 68000 is not compatible with the 68881, a combination of the
1189 two can be specified, since it's possible to do emulation of the
1190 coprocessor instructions with the main processor.
1192 @item -m68851 | -mno-68851
1193 The target machine does (or does not) have a memory-management
1194 unit coprocessor. The default is to assume an MMU for 68020 and up.
1202 @xref{Nios II Options}, for the options available when @value{AS} is configured
1203 for an Altera Nios II processor.
1207 @c man begin OPTIONS
1208 The following options are available when @value{AS} is configured for an
1209 Altera Nios II processor.
1211 @c man begin INCLUDE
1212 @include c-nios2.texi
1213 @c ended inside the included file
1219 For details about the PDP-11 machine dependent features options,
1220 see @ref{PDP-11-Options}.
1223 @item -mpic | -mno-pic
1224 Generate position-independent (or position-dependent) code. The
1225 default is @option{-mpic}.
1228 @itemx -mall-extensions
1229 Enable all instruction set extensions. This is the default.
1231 @item -mno-extensions
1232 Disable all instruction set extensions.
1234 @item -m@var{extension} | -mno-@var{extension}
1235 Enable (or disable) a particular instruction set extension.
1238 Enable the instruction set extensions supported by a particular CPU, and
1239 disable all other extensions.
1241 @item -m@var{machine}
1242 Enable the instruction set extensions supported by a particular machine
1243 model, and disable all other extensions.
1249 The following options are available when @value{AS} is configured for
1250 a picoJava processor.
1254 @cindex PJ endianness
1255 @cindex endianness, PJ
1256 @cindex big endian output, PJ
1258 Generate ``big endian'' format output.
1260 @cindex little endian output, PJ
1262 Generate ``little endian'' format output.
1270 @xref{PRU Options}, for the options available when @value{AS} is configured
1271 for a PRU processor.
1275 @c man begin OPTIONS
1276 The following options are available when @value{AS} is configured for a
1279 @c man begin INCLUDE
1281 @c ended inside the included file
1286 The following options are available when @value{AS} is configured for the
1287 Motorola 68HC11 or 68HC12 series.
1291 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1292 Specify what processor is the target. The default is
1293 defined by the configuration option when building the assembler.
1295 @item --xgate-ramoffset
1296 Instruct the linker to offset RAM addresses from S12X address space into
1297 XGATE address space.
1300 Specify to use the 16-bit integer ABI.
1303 Specify to use the 32-bit integer ABI.
1305 @item -mshort-double
1306 Specify to use the 32-bit double ABI.
1309 Specify to use the 64-bit double ABI.
1311 @item --force-long-branches
1312 Relative branches are turned into absolute ones. This concerns
1313 conditional branches, unconditional branches and branches to a
1316 @item -S | --short-branches
1317 Do not turn relative branches into absolute ones
1318 when the offset is out of range.
1320 @item --strict-direct-mode
1321 Do not turn the direct addressing mode into extended addressing mode
1322 when the instruction does not support direct addressing mode.
1324 @item --print-insn-syntax
1325 Print the syntax of instruction in case of error.
1327 @item --print-opcodes
1328 Print the list of instructions with syntax and then exit.
1330 @item --generate-example
1331 Print an example of instruction for each possible instruction and then exit.
1332 This option is only useful for testing @command{@value{AS}}.
1338 The following options are available when @command{@value{AS}} is configured
1339 for the SPARC architecture:
1342 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1343 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1344 Explicitly select a variant of the SPARC architecture.
1346 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1347 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1349 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1350 UltraSPARC extensions.
1352 @item -xarch=v8plus | -xarch=v8plusa
1353 For compatibility with the Solaris v9 assembler. These options are
1354 equivalent to -Av8plus and -Av8plusa, respectively.
1357 Warn when the assembler switches to another architecture.
1362 The following options are available when @value{AS} is configured for the 'c54x
1367 Enable extended addressing mode. All addresses and relocations will assume
1368 extended addressing (usually 23 bits).
1369 @item -mcpu=@var{CPU_VERSION}
1370 Sets the CPU version being compiled for.
1371 @item -merrors-to-file @var{FILENAME}
1372 Redirect error output to a file, for broken systems which don't support such
1373 behaviour in the shell.
1378 @c man begin OPTIONS
1379 The following options are available when @value{AS} is configured for
1384 This option sets the largest size of an object that can be referenced
1385 implicitly with the @code{gp} register. It is only accepted for targets that
1386 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1388 @cindex MIPS endianness
1389 @cindex endianness, MIPS
1390 @cindex big endian output, MIPS
1392 Generate ``big endian'' format output.
1394 @cindex little endian output, MIPS
1396 Generate ``little endian'' format output.
1414 Generate code for a particular MIPS Instruction Set Architecture level.
1415 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1416 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1417 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1418 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1419 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1420 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1421 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1422 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1423 MIPS64 Release 6 ISA processors, respectively.
1425 @item -march=@var{cpu}
1426 Generate code for a particular MIPS CPU.
1428 @item -mtune=@var{cpu}
1429 Schedule and tune for a particular MIPS CPU.
1433 Cause nops to be inserted if the read of the destination register
1434 of an mfhi or mflo instruction occurs in the following two instructions.
1437 @itemx -mno-fix-rm7000
1438 Cause nops to be inserted if a dmult or dmultu instruction is
1439 followed by a load instruction.
1443 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1444 section instead of the standard ELF .stabs sections.
1448 Control generation of @code{.pdr} sections.
1452 The register sizes are normally inferred from the ISA and ABI, but these
1453 flags force a certain group of registers to be treated as 32 bits wide at
1454 all times. @samp{-mgp32} controls the size of general-purpose registers
1455 and @samp{-mfp32} controls the size of floating-point registers.
1459 The register sizes are normally inferred from the ISA and ABI, but these
1460 flags force a certain group of registers to be treated as 64 bits wide at
1461 all times. @samp{-mgp64} controls the size of general-purpose registers
1462 and @samp{-mfp64} controls the size of floating-point registers.
1465 The register sizes are normally inferred from the ISA and ABI, but using
1466 this flag in combination with @samp{-mabi=32} enables an ABI variant
1467 which will operate correctly with floating-point registers which are
1471 @itemx -mno-odd-spreg
1472 Enable use of floating-point operations on odd-numbered single-precision
1473 registers when supported by the ISA. @samp{-mfpxx} implies
1474 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1478 Generate code for the MIPS 16 processor. This is equivalent to putting
1479 @code{.module mips16} at the start of the assembly file. @samp{-no-mips16}
1480 turns off this option.
1483 @itemx -mno-mips16e2
1484 Enable the use of MIPS16e2 instructions in MIPS16 mode. This is equivalent
1485 to putting @code{.module mips16e2} at the start of the assembly file.
1486 @samp{-mno-mips16e2} turns off this option.
1489 @itemx -mno-micromips
1490 Generate code for the microMIPS processor. This is equivalent to putting
1491 @code{.module micromips} at the start of the assembly file.
1492 @samp{-mno-micromips} turns off this option. This is equivalent to putting
1493 @code{.module nomicromips} at the start of the assembly file.
1496 @itemx -mno-smartmips
1497 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1498 equivalent to putting @code{.module smartmips} at the start of the assembly
1499 file. @samp{-mno-smartmips} turns off this option.
1503 Generate code for the MIPS-3D Application Specific Extension.
1504 This tells the assembler to accept MIPS-3D instructions.
1505 @samp{-no-mips3d} turns off this option.
1509 Generate code for the MDMX Application Specific Extension.
1510 This tells the assembler to accept MDMX instructions.
1511 @samp{-no-mdmx} turns off this option.
1515 Generate code for the DSP Release 1 Application Specific Extension.
1516 This tells the assembler to accept DSP Release 1 instructions.
1517 @samp{-mno-dsp} turns off this option.
1521 Generate code for the DSP Release 2 Application Specific Extension.
1522 This option implies @samp{-mdsp}.
1523 This tells the assembler to accept DSP Release 2 instructions.
1524 @samp{-mno-dspr2} turns off this option.
1528 Generate code for the DSP Release 3 Application Specific Extension.
1529 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1530 This tells the assembler to accept DSP Release 3 instructions.
1531 @samp{-mno-dspr3} turns off this option.
1535 Generate code for the MIPS SIMD Architecture Extension.
1536 This tells the assembler to accept MSA instructions.
1537 @samp{-mno-msa} turns off this option.
1541 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1542 This tells the assembler to accept XPA instructions.
1543 @samp{-mno-xpa} turns off this option.
1547 Generate code for the MT Application Specific Extension.
1548 This tells the assembler to accept MT instructions.
1549 @samp{-mno-mt} turns off this option.
1553 Generate code for the MCU Application Specific Extension.
1554 This tells the assembler to accept MCU instructions.
1555 @samp{-mno-mcu} turns off this option.
1559 Generate code for the MIPS cyclic redundancy check (CRC) Application
1560 Specific Extension. This tells the assembler to accept CRC instructions.
1561 @samp{-mno-crc} turns off this option.
1565 Generate code for the Global INValidate (GINV) Application Specific
1566 Extension. This tells the assembler to accept GINV instructions.
1567 @samp{-mno-ginv} turns off this option.
1569 @item -mloongson-mmi
1570 @itemx -mno-loongson-mmi
1571 Generate code for the Loongson MultiMedia extensions Instructions (MMI)
1572 Application Specific Extension. This tells the assembler to accept MMI
1574 @samp{-mno-loongson-mmi} turns off this option.
1578 Only use 32-bit instruction encodings when generating code for the
1579 microMIPS processor. This option inhibits the use of any 16-bit
1580 instructions. This is equivalent to putting @code{.set insn32} at
1581 the start of the assembly file. @samp{-mno-insn32} turns off this
1582 option. This is equivalent to putting @code{.set noinsn32} at the
1583 start of the assembly file. By default @samp{-mno-insn32} is
1584 selected, allowing all instructions to be used.
1586 @item --construct-floats
1587 @itemx --no-construct-floats
1588 The @samp{--no-construct-floats} option disables the construction of
1589 double width floating point constants by loading the two halves of the
1590 value into the two single width floating point registers that make up
1591 the double width register. By default @samp{--construct-floats} is
1592 selected, allowing construction of these floating point constants.
1594 @item --relax-branch
1595 @itemx --no-relax-branch
1596 The @samp{--relax-branch} option enables the relaxation of out-of-range
1597 branches. By default @samp{--no-relax-branch} is selected, causing any
1598 out-of-range branches to produce an error.
1600 @item -mignore-branch-isa
1601 @itemx -mno-ignore-branch-isa
1602 Ignore branch checks for invalid transitions between ISA modes. The
1603 semantics of branches does not provide for an ISA mode switch, so in
1604 most cases the ISA mode a branch has been encoded for has to be the
1605 same as the ISA mode of the branch's target label. Therefore GAS has
1606 checks implemented that verify in branch assembly that the two ISA
1607 modes match. @samp{-mignore-branch-isa} disables these checks. By
1608 default @samp{-mno-ignore-branch-isa} is selected, causing any invalid
1609 branch requiring a transition between ISA modes to produce an error.
1611 @item -mnan=@var{encoding}
1612 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1613 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1616 @item --emulation=@var{name}
1617 This option was formerly used to switch between ELF and ECOFF output
1618 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1619 removed in GAS 2.24, so the option now serves little purpose.
1620 It is retained for backwards compatibility.
1622 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1623 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1624 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1625 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1626 preferred options instead.
1629 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1636 Control how to deal with multiplication overflow and division by zero.
1637 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1638 (and only work for Instruction Set Architecture level 2 and higher);
1639 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1643 When this option is used, @command{@value{AS}} will issue a warning every
1644 time it generates a nop instruction from a macro.
1650 The following options are available when @value{AS} is configured for
1656 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1657 The command-line option @samp{-nojsri2bsr} can be used to disable it.
1661 Enable or disable the silicon filter behaviour. By default this is disabled.
1662 The default can be overridden by the @samp{-sifilter} command-line option.
1665 Alter jump instructions for long displacements.
1667 @item -mcpu=[210|340]
1668 Select the cpu type on the target hardware. This controls which instructions
1672 Assemble for a big endian target.
1675 Assemble for a little endian target.
1684 @xref{Meta Options}, for the options available when @value{AS} is configured
1685 for a Meta processor.
1689 @c man begin OPTIONS
1690 The following options are available when @value{AS} is configured for a
1693 @c man begin INCLUDE
1694 @include c-metag.texi
1695 @c ended inside the included file
1700 @c man begin OPTIONS
1702 See the info pages for documentation of the MMIX-specific options.
1708 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1709 for a NDS32 processor.
1711 @c ended inside the included file
1715 @c man begin OPTIONS
1716 The following options are available when @value{AS} is configured for a
1719 @c man begin INCLUDE
1720 @include c-nds32.texi
1721 @c ended inside the included file
1728 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1729 for a PowerPC processor.
1733 @c man begin OPTIONS
1734 The following options are available when @value{AS} is configured for a
1737 @c man begin INCLUDE
1739 @c ended inside the included file
1747 @xref{RISC-V-Options}, for the options available when @value{AS} is configured
1748 for a RISC-V processor.
1752 @c man begin OPTIONS
1753 The following options are available when @value{AS} is configured for a
1756 @c man begin INCLUDE
1757 @include c-riscv.texi
1758 @c ended inside the included file
1763 @c man begin OPTIONS
1765 See the info pages for documentation of the RX-specific options.
1769 The following options are available when @value{AS} is configured for the s390
1775 Select the word size, either 31/32 bits or 64 bits.
1778 Select the architecture mode, either the Enterprise System
1779 Architecture (esa) or the z/Architecture mode (zarch).
1780 @item -march=@var{processor}
1781 Specify which s390 processor variant is the target, @samp{g5} (or
1782 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1783 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1784 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1785 @samp{z13} (or @samp{arch11}), or @samp{z14} (or @samp{arch12}).
1787 @itemx -mno-regnames
1788 Allow or disallow symbolic names for registers.
1789 @item -mwarn-areg-zero
1790 Warn whenever the operand for a base or index register has been specified
1791 but evaluates to zero.
1799 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1800 for a TMS320C6000 processor.
1804 @c man begin OPTIONS
1805 The following options are available when @value{AS} is configured for a
1806 TMS320C6000 processor.
1808 @c man begin INCLUDE
1809 @include c-tic6x.texi
1810 @c ended inside the included file
1818 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1819 for a TILE-Gx processor.
1823 @c man begin OPTIONS
1824 The following options are available when @value{AS} is configured for a TILE-Gx
1827 @c man begin INCLUDE
1828 @include c-tilegx.texi
1829 @c ended inside the included file
1837 @xref{Visium Options}, for the options available when @value{AS} is configured
1838 for a Visium processor.
1842 @c man begin OPTIONS
1843 The following option is available when @value{AS} is configured for a Visium
1846 @c man begin INCLUDE
1847 @include c-visium.texi
1848 @c ended inside the included file
1856 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1857 for an Xtensa processor.
1861 @c man begin OPTIONS
1862 The following options are available when @value{AS} is configured for an
1865 @c man begin INCLUDE
1866 @include c-xtensa.texi
1867 @c ended inside the included file
1872 @c man begin OPTIONS
1875 The following options are available when @value{AS} is configured for
1876 a Z80 family processor.
1879 Assemble for Z80 processor.
1881 Assemble for R800 processor.
1882 @item -ignore-undocumented-instructions
1884 Assemble undocumented Z80 instructions that also work on R800 without warning.
1885 @item -ignore-unportable-instructions
1887 Assemble all undocumented Z80 instructions without warning.
1888 @item -warn-undocumented-instructions
1890 Issue a warning for undocumented Z80 instructions that also work on R800.
1891 @item -warn-unportable-instructions
1893 Issue a warning for undocumented Z80 instructions that do not work on R800.
1894 @item -forbid-undocumented-instructions
1896 Treat all undocumented instructions as errors.
1897 @item -forbid-unportable-instructions
1899 Treat undocumented Z80 instructions that do not work on R800 as errors.
1906 * Manual:: Structure of this Manual
1907 * GNU Assembler:: The GNU Assembler
1908 * Object Formats:: Object File Formats
1909 * Command Line:: Command Line
1910 * Input Files:: Input Files
1911 * Object:: Output (Object) File
1912 * Errors:: Error and Warning Messages
1916 @section Structure of this Manual
1918 @cindex manual, structure and purpose
1919 This manual is intended to describe what you need to know to use
1920 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1921 notation for symbols, constants, and expressions; the directives that
1922 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1925 We also cover special features in the @value{TARGET}
1926 configuration of @command{@value{AS}}, including assembler directives.
1929 This manual also describes some of the machine-dependent features of
1930 various flavors of the assembler.
1933 @cindex machine instructions (not covered)
1934 On the other hand, this manual is @emph{not} intended as an introduction
1935 to programming in assembly language---let alone programming in general!
1936 In a similar vein, we make no attempt to introduce the machine
1937 architecture; we do @emph{not} describe the instruction set, standard
1938 mnemonics, registers or addressing modes that are standard to a
1939 particular architecture.
1941 You may want to consult the manufacturer's
1942 machine architecture manual for this information.
1946 For information on the H8/300 machine instruction set, see @cite{H8/300
1947 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1948 Programming Manual} (Renesas).
1951 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1952 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1953 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1954 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1957 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1961 @c I think this is premature---doc@cygnus.com, 17jan1991
1963 Throughout this manual, we assume that you are running @dfn{GNU},
1964 the portable operating system from the @dfn{Free Software
1965 Foundation, Inc.}. This restricts our attention to certain kinds of
1966 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1967 once this assumption is granted examples and definitions need less
1970 @command{@value{AS}} is part of a team of programs that turn a high-level
1971 human-readable series of instructions into a low-level
1972 computer-readable series of instructions. Different versions of
1973 @command{@value{AS}} are used for different kinds of computer.
1976 @c There used to be a section "Terminology" here, which defined
1977 @c "contents", "byte", "word", and "long". Defining "word" to any
1978 @c particular size is confusing when the .word directive may generate 16
1979 @c bits on one machine and 32 bits on another; in general, for the user
1980 @c version of this manual, none of these terms seem essential to define.
1981 @c They were used very little even in the former draft of the manual;
1982 @c this draft makes an effort to avoid them (except in names of
1986 @section The GNU Assembler
1988 @c man begin DESCRIPTION
1990 @sc{gnu} @command{as} is really a family of assemblers.
1992 This manual describes @command{@value{AS}}, a member of that family which is
1993 configured for the @value{TARGET} architectures.
1995 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1996 should find a fairly similar environment when you use it on another
1997 architecture. Each version has much in common with the others,
1998 including object file formats, most assembler directives (often called
1999 @dfn{pseudo-ops}) and assembler syntax.@refill
2001 @cindex purpose of @sc{gnu} assembler
2002 @command{@value{AS}} is primarily intended to assemble the output of the
2003 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
2004 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
2005 assemble correctly everything that other assemblers for the same
2006 machine would assemble.
2008 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
2011 @c This remark should appear in generic version of manual; assumption
2012 @c here is that generic version sets M680x0.
2013 This doesn't mean @command{@value{AS}} always uses the same syntax as another
2014 assembler for the same architecture; for example, we know of several
2015 incompatible versions of 680x0 assembly language syntax.
2020 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
2021 program in one pass of the source file. This has a subtle impact on the
2022 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
2024 @node Object Formats
2025 @section Object File Formats
2027 @cindex object file format
2028 The @sc{gnu} assembler can be configured to produce several alternative
2029 object file formats. For the most part, this does not affect how you
2030 write assembly language programs; but directives for debugging symbols
2031 are typically different in different file formats. @xref{Symbol
2032 Attributes,,Symbol Attributes}.
2035 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
2036 @value{OBJ-NAME} format object files.
2038 @c The following should exhaust all configs that set MULTI-OBJ, ideally
2040 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
2041 SOM or ELF format object files.
2046 @section Command Line
2048 @cindex command line conventions
2050 After the program name @command{@value{AS}}, the command line may contain
2051 options and file names. Options may appear in any order, and may be
2052 before, after, or between file names. The order of file names is
2055 @cindex standard input, as input file
2057 @file{--} (two hyphens) by itself names the standard input file
2058 explicitly, as one of the files for @command{@value{AS}} to assemble.
2060 @cindex options, command line
2061 Except for @samp{--} any command-line argument that begins with a
2062 hyphen (@samp{-}) is an option. Each option changes the behavior of
2063 @command{@value{AS}}. No option changes the way another option works. An
2064 option is a @samp{-} followed by one or more letters; the case of
2065 the letter is important. All options are optional.
2067 Some options expect exactly one file name to follow them. The file
2068 name may either immediately follow the option's letter (compatible
2069 with older assemblers) or it may be the next command argument (@sc{gnu}
2070 standard). These two command lines are equivalent:
2073 @value{AS} -o my-object-file.o mumble.s
2074 @value{AS} -omy-object-file.o mumble.s
2078 @section Input Files
2081 @cindex source program
2082 @cindex files, input
2083 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2084 describe the program input to one run of @command{@value{AS}}. The program may
2085 be in one or more files; how the source is partitioned into files
2086 doesn't change the meaning of the source.
2088 @c I added "con" prefix to "catenation" just to prove I can overcome my
2089 @c APL training... doc@cygnus.com
2090 The source program is a concatenation of the text in all the files, in the
2093 @c man begin DESCRIPTION
2094 Each time you run @command{@value{AS}} it assembles exactly one source
2095 program. The source program is made up of one or more files.
2096 (The standard input is also a file.)
2098 You give @command{@value{AS}} a command line that has zero or more input file
2099 names. The input files are read (from left file name to right). A
2100 command-line argument (in any position) that has no special meaning
2101 is taken to be an input file name.
2103 If you give @command{@value{AS}} no file names it attempts to read one input file
2104 from the @command{@value{AS}} standard input, which is normally your terminal. You
2105 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2108 Use @samp{--} if you need to explicitly name the standard input file
2109 in your command line.
2111 If the source is empty, @command{@value{AS}} produces a small, empty object
2116 @subheading Filenames and Line-numbers
2118 @cindex input file linenumbers
2119 @cindex line numbers, in input files
2120 There are two ways of locating a line in the input file (or files) and
2121 either may be used in reporting error messages. One way refers to a line
2122 number in a physical file; the other refers to a line number in a
2123 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2125 @dfn{Physical files} are those files named in the command line given
2126 to @command{@value{AS}}.
2128 @dfn{Logical files} are simply names declared explicitly by assembler
2129 directives; they bear no relation to physical files. Logical file names help
2130 error messages reflect the original source file, when @command{@value{AS}} source
2131 is itself synthesized from other files. @command{@value{AS}} understands the
2132 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2133 @ref{File,,@code{.file}}.
2136 @section Output (Object) File
2142 Every time you run @command{@value{AS}} it produces an output file, which is
2143 your assembly language program translated into numbers. This file
2144 is the object file. Its default name is @code{a.out}.
2145 You can give it another name by using the @option{-o} option. Conventionally,
2146 object file names end with @file{.o}. The default name is used for historical
2147 reasons: older assemblers were capable of assembling self-contained programs
2148 directly into a runnable program. (For some formats, this isn't currently
2149 possible, but it can be done for the @code{a.out} format.)
2153 The object file is meant for input to the linker @code{@value{LD}}. It contains
2154 assembled program code, information to help @code{@value{LD}} integrate
2155 the assembled program into a runnable file, and (optionally) symbolic
2156 information for the debugger.
2158 @c link above to some info file(s) like the description of a.out.
2159 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2162 @section Error and Warning Messages
2164 @c man begin DESCRIPTION
2166 @cindex error messages
2167 @cindex warning messages
2168 @cindex messages from assembler
2169 @command{@value{AS}} may write warnings and error messages to the standard error
2170 file (usually your terminal). This should not happen when a compiler
2171 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2172 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2173 grave problem that stops the assembly.
2177 @cindex format of warning messages
2178 Warning messages have the format
2181 file_name:@b{NNN}:Warning Message Text
2185 @cindex file names and line numbers, in warnings/errors
2186 (where @b{NNN} is a line number). If both a logical file name
2187 (@pxref{File,,@code{.file}}) and a logical line number
2189 (@pxref{Line,,@code{.line}})
2191 have been given then they will be used, otherwise the file name and line number
2192 in the current assembler source file will be used. The message text is
2193 intended to be self explanatory (in the grand Unix tradition).
2195 Note the file name must be set via the logical version of the @code{.file}
2196 directive, not the DWARF2 version of the @code{.file} directive. For example:
2200 error_assembler_source
2206 produces this output:
2210 asm.s:2: Error: no such instruction: `error_assembler_source'
2211 foo.c:31: Error: no such instruction: `error_c_source'
2214 @cindex format of error messages
2215 Error messages have the format
2218 file_name:@b{NNN}:FATAL:Error Message Text
2221 The file name and line number are derived as for warning
2222 messages. The actual message text may be rather less explanatory
2223 because many of them aren't supposed to happen.
2226 @chapter Command-Line Options
2228 @cindex options, all versions of assembler
2229 This chapter describes command-line options available in @emph{all}
2230 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2231 for options specific
2233 to the @value{TARGET} target.
2236 to particular machine architectures.
2239 @c man begin DESCRIPTION
2241 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2242 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2243 The assembler arguments must be separated from each other (and the @samp{-Wa})
2244 by commas. For example:
2247 gcc -c -g -O -Wa,-alh,-L file.c
2251 This passes two options to the assembler: @samp{-alh} (emit a listing to
2252 standard output with high-level and assembly source) and @samp{-L} (retain
2253 local symbols in the symbol table).
2255 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2256 command-line options are automatically passed to the assembler by the compiler.
2257 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2258 precisely what options it passes to each compilation pass, including the
2264 * a:: -a[cdghlns] enable listings
2265 * alternate:: --alternate enable alternate macro syntax
2266 * D:: -D for compatibility
2267 * f:: -f to work faster
2268 * I:: -I for .include search path
2269 @ifclear DIFF-TBL-KLUGE
2270 * K:: -K for compatibility
2272 @ifset DIFF-TBL-KLUGE
2273 * K:: -K for difference tables
2276 * L:: -L to retain local symbols
2277 * listing:: --listing-XXX to configure listing output
2278 * M:: -M or --mri to assemble in MRI compatibility mode
2279 * MD:: --MD for dependency tracking
2280 * no-pad-sections:: --no-pad-sections to stop section padding
2281 * o:: -o to name the object file
2282 * R:: -R to join data and text sections
2283 * statistics:: --statistics to see statistics about assembly
2284 * traditional-format:: --traditional-format for compatible output
2285 * v:: -v to announce version
2286 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2287 * Z:: -Z to make object file even after errors
2291 @section Enable Listings: @option{-a[cdghlns]}
2301 @cindex listings, enabling
2302 @cindex assembly listings, enabling
2304 These options enable listing output from the assembler. By itself,
2305 @samp{-a} requests high-level, assembly, and symbols listing.
2306 You can use other letters to select specific options for the list:
2307 @samp{-ah} requests a high-level language listing,
2308 @samp{-al} requests an output-program assembly listing, and
2309 @samp{-as} requests a symbol table listing.
2310 High-level listings require that a compiler debugging option like
2311 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2314 Use the @samp{-ag} option to print a first section with general assembly
2315 information, like @value{AS} version, switches passed, or time stamp.
2317 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2318 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2319 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2320 omitted from the listing.
2322 Use the @samp{-ad} option to omit debugging directives from the
2325 Once you have specified one of these options, you can further control
2326 listing output and its appearance using the directives @code{.list},
2327 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2329 The @samp{-an} option turns off all forms processing.
2330 If you do not request listing output with one of the @samp{-a} options, the
2331 listing-control directives have no effect.
2333 The letters after @samp{-a} may be combined into one option,
2334 @emph{e.g.}, @samp{-aln}.
2336 Note if the assembler source is coming from the standard input (e.g.,
2338 is being created by @code{@value{GCC}} and the @samp{-pipe} command-line switch
2339 is being used) then the listing will not contain any comments or preprocessor
2340 directives. This is because the listing code buffers input source lines from
2341 stdin only after they have been preprocessed by the assembler. This reduces
2342 memory usage and makes the code more efficient.
2345 @section @option{--alternate}
2348 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2351 @section @option{-D}
2354 This option has no effect whatsoever, but it is accepted to make it more
2355 likely that scripts written for other assemblers also work with
2356 @command{@value{AS}}.
2359 @section Work Faster: @option{-f}
2362 @cindex trusted compiler
2363 @cindex faster processing (@option{-f})
2364 @samp{-f} should only be used when assembling programs written by a
2365 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2366 and comment preprocessing on
2367 the input file(s) before assembling them. @xref{Preprocessing,
2371 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2372 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2377 @section @code{.include} Search Path: @option{-I} @var{path}
2379 @kindex -I @var{path}
2380 @cindex paths for @code{.include}
2381 @cindex search path for @code{.include}
2382 @cindex @code{include} directive search path
2383 Use this option to add a @var{path} to the list of directories
2384 @command{@value{AS}} searches for files specified in @code{.include}
2385 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2386 many times as necessary to include a variety of paths. The current
2387 working directory is always searched first; after that, @command{@value{AS}}
2388 searches any @samp{-I} directories in the same order as they were
2389 specified (left to right) on the command line.
2392 @section Difference Tables: @option{-K}
2395 @ifclear DIFF-TBL-KLUGE
2396 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2397 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2398 where it can be used to warn when the assembler alters the machine code
2399 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2400 family does not have the addressing limitations that sometimes lead to this
2401 alteration on other platforms.
2404 @ifset DIFF-TBL-KLUGE
2405 @cindex difference tables, warning
2406 @cindex warning for altered difference tables
2407 @command{@value{AS}} sometimes alters the code emitted for directives of the
2408 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2409 You can use the @samp{-K} option if you want a warning issued when this
2414 @section Include Local Symbols: @option{-L}
2417 @cindex local symbols, retaining in output
2418 Symbols beginning with system-specific local label prefixes, typically
2419 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2420 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2421 such symbols when debugging, because they are intended for the use of
2422 programs (like compilers) that compose assembler programs, not for your
2423 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2424 such symbols, so you do not normally debug with them.
2426 This option tells @command{@value{AS}} to retain those local symbols
2427 in the object file. Usually if you do this you also tell the linker
2428 @code{@value{LD}} to preserve those symbols.
2431 @section Configuring listing output: @option{--listing}
2433 The listing feature of the assembler can be enabled via the command-line switch
2434 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2435 hex dump of the corresponding locations in the output object file, and displays
2436 them as a listing file. The format of this listing can be controlled by
2437 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2438 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2439 @code{.psize} (@pxref{Psize}), and
2440 @code{.eject} (@pxref{Eject}) and also by the following switches:
2443 @item --listing-lhs-width=@samp{number}
2444 @kindex --listing-lhs-width
2445 @cindex Width of first line disassembly output
2446 Sets the maximum width, in words, of the first line of the hex byte dump. This
2447 dump appears on the left hand side of the listing output.
2449 @item --listing-lhs-width2=@samp{number}
2450 @kindex --listing-lhs-width2
2451 @cindex Width of continuation lines of disassembly output
2452 Sets the maximum width, in words, of any further lines of the hex byte dump for
2453 a given input source line. If this value is not specified, it defaults to being
2454 the same as the value specified for @samp{--listing-lhs-width}. If neither
2455 switch is used the default is to one.
2457 @item --listing-rhs-width=@samp{number}
2458 @kindex --listing-rhs-width
2459 @cindex Width of source line output
2460 Sets the maximum width, in characters, of the source line that is displayed
2461 alongside the hex dump. The default value for this parameter is 100. The
2462 source line is displayed on the right hand side of the listing output.
2464 @item --listing-cont-lines=@samp{number}
2465 @kindex --listing-cont-lines
2466 @cindex Maximum number of continuation lines
2467 Sets the maximum number of continuation lines of hex dump that will be
2468 displayed for a given single line of source input. The default value is 4.
2472 @section Assemble in MRI Compatibility Mode: @option{-M}
2475 @cindex MRI compatibility mode
2476 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2477 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2478 compatible with the @code{ASM68K} assembler from Microtec Research.
2479 The exact nature of the
2480 MRI syntax will not be documented here; see the MRI manuals for more
2481 information. Note in particular that the handling of macros and macro
2482 arguments is somewhat different. The purpose of this option is to permit
2483 assembling existing MRI assembler code using @command{@value{AS}}.
2485 The MRI compatibility is not complete. Certain operations of the MRI assembler
2486 depend upon its object file format, and can not be supported using other object
2487 file formats. Supporting these would require enhancing each object file format
2488 individually. These are:
2491 @item global symbols in common section
2493 The m68k MRI assembler supports common sections which are merged by the linker.
2494 Other object file formats do not support this. @command{@value{AS}} handles
2495 common sections by treating them as a single common symbol. It permits local
2496 symbols to be defined within a common section, but it can not support global
2497 symbols, since it has no way to describe them.
2499 @item complex relocations
2501 The MRI assemblers support relocations against a negated section address, and
2502 relocations which combine the start addresses of two or more sections. These
2503 are not support by other object file formats.
2505 @item @code{END} pseudo-op specifying start address
2507 The MRI @code{END} pseudo-op permits the specification of a start address.
2508 This is not supported by other object file formats. The start address may
2509 instead be specified using the @option{-e} option to the linker, or in a linker
2512 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2514 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2515 name to the output file. This is not supported by other object file formats.
2517 @item @code{ORG} pseudo-op
2519 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2520 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2521 which changes the location within the current section. Absolute sections are
2522 not supported by other object file formats. The address of a section may be
2523 assigned within a linker script.
2526 There are some other features of the MRI assembler which are not supported by
2527 @command{@value{AS}}, typically either because they are difficult or because they
2528 seem of little consequence. Some of these may be supported in future releases.
2532 @item EBCDIC strings
2534 EBCDIC strings are not supported.
2536 @item packed binary coded decimal
2538 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2539 and @code{DCB.P} pseudo-ops are not supported.
2541 @item @code{FEQU} pseudo-op
2543 The m68k @code{FEQU} pseudo-op is not supported.
2545 @item @code{NOOBJ} pseudo-op
2547 The m68k @code{NOOBJ} pseudo-op is not supported.
2549 @item @code{OPT} branch control options
2551 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2552 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2553 relaxes all branches, whether forward or backward, to an appropriate size, so
2554 these options serve no purpose.
2556 @item @code{OPT} list control options
2558 The following m68k @code{OPT} list control options are ignored: @code{C},
2559 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2560 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2562 @item other @code{OPT} options
2564 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2565 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2567 @item @code{OPT} @code{D} option is default
2569 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2570 @code{OPT NOD} may be used to turn it off.
2572 @item @code{XREF} pseudo-op.
2574 The m68k @code{XREF} pseudo-op is ignored.
2579 @section Dependency Tracking: @option{--MD}
2582 @cindex dependency tracking
2585 @command{@value{AS}} can generate a dependency file for the file it creates. This
2586 file consists of a single rule suitable for @code{make} describing the
2587 dependencies of the main source file.
2589 The rule is written to the file named in its argument.
2591 This feature is used in the automatic updating of makefiles.
2593 @node no-pad-sections
2594 @section Output Section Padding
2595 @kindex --no-pad-sections
2596 @cindex output section padding
2597 Normally the assembler will pad the end of each output section up to its
2598 alignment boundary. But this can waste space, which can be significant on
2599 memory constrained targets. So the @option{--no-pad-sections} option will
2600 disable this behaviour.
2603 @section Name the Object File: @option{-o}
2606 @cindex naming object file
2607 @cindex object file name
2608 There is always one object file output when you run @command{@value{AS}}. By
2609 default it has the name @file{a.out}.
2610 You use this option (which takes exactly one filename) to give the
2611 object file a different name.
2613 Whatever the object file is called, @command{@value{AS}} overwrites any
2614 existing file of the same name.
2617 @section Join Data and Text Sections: @option{-R}
2620 @cindex data and text sections, joining
2621 @cindex text and data sections, joining
2622 @cindex joining text and data sections
2623 @cindex merging text and data sections
2624 @option{-R} tells @command{@value{AS}} to write the object file as if all
2625 data-section data lives in the text section. This is only done at
2626 the very last moment: your binary data are the same, but data
2627 section parts are relocated differently. The data section part of
2628 your object file is zero bytes long because all its bytes are
2629 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2631 When you specify @option{-R} it would be possible to generate shorter
2632 address displacements (because we do not have to cross between text and
2633 data section). We refrain from doing this simply for compatibility with
2634 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2637 When @command{@value{AS}} is configured for COFF or ELF output,
2638 this option is only useful if you use sections named @samp{.text} and
2643 @option{-R} is not supported for any of the HPPA targets. Using
2644 @option{-R} generates a warning from @command{@value{AS}}.
2648 @section Display Assembly Statistics: @option{--statistics}
2650 @kindex --statistics
2651 @cindex statistics, about assembly
2652 @cindex time, total for assembly
2653 @cindex space used, maximum for assembly
2654 Use @samp{--statistics} to display two statistics about the resources used by
2655 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2656 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2659 @node traditional-format
2660 @section Compatible Output: @option{--traditional-format}
2662 @kindex --traditional-format
2663 For some targets, the output of @command{@value{AS}} is different in some ways
2664 from the output of some existing assembler. This switch requests
2665 @command{@value{AS}} to use the traditional format instead.
2667 For example, it disables the exception frame optimizations which
2668 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2671 @section Announce Version: @option{-v}
2675 @cindex assembler version
2676 @cindex version of assembler
2677 You can find out what version of as is running by including the
2678 option @samp{-v} (which you can also spell as @samp{-version}) on the
2682 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2684 @command{@value{AS}} should never give a warning or error message when
2685 assembling compiler output. But programs written by people often
2686 cause @command{@value{AS}} to give a warning that a particular assumption was
2687 made. All such warnings are directed to the standard error file.
2691 @cindex suppressing warnings
2692 @cindex warnings, suppressing
2693 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2694 This only affects the warning messages: it does not change any particular of
2695 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2698 @kindex --fatal-warnings
2699 @cindex errors, caused by warnings
2700 @cindex warnings, causing error
2701 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2702 files that generate warnings to be in error.
2705 @cindex warnings, switching on
2706 You can switch these options off again by specifying @option{--warn}, which
2707 causes warnings to be output as usual.
2710 @section Generate Object File in Spite of Errors: @option{-Z}
2711 @cindex object file, after errors
2712 @cindex errors, continuing after
2713 After an error message, @command{@value{AS}} normally produces no output. If for
2714 some reason you are interested in object file output even after
2715 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2716 option. If there are any errors, @command{@value{AS}} continues anyways, and
2717 writes an object file after a final warning message of the form @samp{@var{n}
2718 errors, @var{m} warnings, generating bad object file.}
2723 @cindex machine-independent syntax
2724 @cindex syntax, machine-independent
2725 This chapter describes the machine-independent syntax allowed in a
2726 source file. @command{@value{AS}} syntax is similar to what many other
2727 assemblers use; it is inspired by the BSD 4.2
2732 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2736 * Preprocessing:: Preprocessing
2737 * Whitespace:: Whitespace
2738 * Comments:: Comments
2739 * Symbol Intro:: Symbols
2740 * Statements:: Statements
2741 * Constants:: Constants
2745 @section Preprocessing
2747 @cindex preprocessing
2748 The @command{@value{AS}} internal preprocessor:
2750 @cindex whitespace, removed by preprocessor
2752 adjusts and removes extra whitespace. It leaves one space or tab before
2753 the keywords on a line, and turns any other whitespace on the line into
2756 @cindex comments, removed by preprocessor
2758 removes all comments, replacing them with a single space, or an
2759 appropriate number of newlines.
2761 @cindex constants, converted by preprocessor
2763 converts character constants into the appropriate numeric values.
2766 It does not do macro processing, include file handling, or
2767 anything else you may get from your C compiler's preprocessor. You can
2768 do include file processing with the @code{.include} directive
2769 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2770 to get other ``CPP'' style preprocessing by giving the input file a
2771 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2772 Output, gcc info, Using GNU CC}.
2774 Excess whitespace, comments, and character constants
2775 cannot be used in the portions of the input text that are not
2778 @cindex turning preprocessing on and off
2779 @cindex preprocessing, turning on and off
2782 If the first line of an input file is @code{#NO_APP} or if you use the
2783 @samp{-f} option, whitespace and comments are not removed from the input file.
2784 Within an input file, you can ask for whitespace and comment removal in
2785 specific portions of the by putting a line that says @code{#APP} before the
2786 text that may contain whitespace or comments, and putting a line that says
2787 @code{#NO_APP} after this text. This feature is mainly intend to support
2788 @code{asm} statements in compilers whose output is otherwise free of comments
2795 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2796 Whitespace is used to separate symbols, and to make programs neater for
2797 people to read. Unless within character constants
2798 (@pxref{Characters,,Character Constants}), any whitespace means the same
2799 as exactly one space.
2805 There are two ways of rendering comments to @command{@value{AS}}. In both
2806 cases the comment is equivalent to one space.
2808 Anything from @samp{/*} through the next @samp{*/} is a comment.
2809 This means you may not nest these comments.
2813 The only way to include a newline ('\n') in a comment
2814 is to use this sort of comment.
2817 /* This sort of comment does not nest. */
2820 @cindex line comment character
2821 Anything from a @dfn{line comment} character up to the next newline is
2822 considered a comment and is ignored. The line comment character is target
2823 specific, and some targets multiple comment characters. Some targets also have
2824 line comment characters that only work if they are the first character on a
2825 line. Some targets use a sequence of two characters to introduce a line
2826 comment. Some targets can also change their line comment characters depending
2827 upon command-line options that have been used. For more details see the
2828 @emph{Syntax} section in the documentation for individual targets.
2830 If the line comment character is the hash sign (@samp{#}) then it still has the
2831 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2832 to specify logical line numbers:
2835 @cindex lines starting with @code{#}
2836 @cindex logical line numbers
2837 To be compatible with past assemblers, lines that begin with @samp{#} have a
2838 special interpretation. Following the @samp{#} should be an absolute
2839 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2840 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2841 new logical file name. The rest of the line, if any, should be whitespace.
2843 If the first non-whitespace characters on the line are not numeric,
2844 the line is ignored. (Just like a comment.)
2847 # This is an ordinary comment.
2848 # 42-6 "new_file_name" # New logical file name
2849 # This is logical line # 36.
2851 This feature is deprecated, and may disappear from future versions
2852 of @command{@value{AS}}.
2857 @cindex characters used in symbols
2858 @ifclear SPECIAL-SYMS
2859 A @dfn{symbol} is one or more characters chosen from the set of all
2860 letters (both upper and lower case), digits and the three characters
2866 A @dfn{symbol} is one or more characters chosen from the set of all
2867 letters (both upper and lower case), digits and the three characters
2868 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2874 On most machines, you can also use @code{$} in symbol names; exceptions
2875 are noted in @ref{Machine Dependencies}.
2877 No symbol may begin with a digit. Case is significant.
2878 There is no length limit; all characters are significant. Multibyte characters
2879 are supported. Symbols are delimited by characters not in that set, or by the
2880 beginning of a file (since the source program must end with a newline, the end
2881 of a file is not a possible symbol delimiter). @xref{Symbols}.
2883 Symbol names may also be enclosed in double quote @code{"} characters. In such
2884 cases any characters are allowed, except for the NUL character. If a double
2885 quote character is to be included in the symbol name it must be preceeded by a
2886 backslash @code{\} character.
2887 @cindex length of symbols
2892 @cindex statements, structure of
2893 @cindex line separator character
2894 @cindex statement separator character
2896 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2897 @dfn{line separator character}. The line separator character is target
2898 specific and described in the @emph{Syntax} section of each
2899 target's documentation. Not all targets support a line separator character.
2900 The newline or line separator character is considered to be part of the
2901 preceding statement. Newlines and separators within character constants are an
2902 exception: they do not end statements.
2904 @cindex newline, required at file end
2905 @cindex EOF, newline must precede
2906 It is an error to end any statement with end-of-file: the last
2907 character of any input file should be a newline.@refill
2909 An empty statement is allowed, and may include whitespace. It is ignored.
2911 @cindex instructions and directives
2912 @cindex directives and instructions
2913 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2914 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2916 A statement begins with zero or more labels, optionally followed by a
2917 key symbol which determines what kind of statement it is. The key
2918 symbol determines the syntax of the rest of the statement. If the
2919 symbol begins with a dot @samp{.} then the statement is an assembler
2920 directive: typically valid for any computer. If the symbol begins with
2921 a letter the statement is an assembly language @dfn{instruction}: it
2922 assembles into a machine language instruction.
2924 Different versions of @command{@value{AS}} for different computers
2925 recognize different instructions. In fact, the same symbol may
2926 represent a different instruction in a different computer's assembly
2930 @cindex @code{:} (label)
2931 @cindex label (@code{:})
2932 A label is a symbol immediately followed by a colon (@code{:}).
2933 Whitespace before a label or after a colon is permitted, but you may not
2934 have whitespace between a label's symbol and its colon. @xref{Labels}.
2937 For HPPA targets, labels need not be immediately followed by a colon, but
2938 the definition of a label must begin in column zero. This also implies that
2939 only one label may be defined on each line.
2943 label: .directive followed by something
2944 another_label: # This is an empty statement.
2945 instruction operand_1, operand_2, @dots{}
2952 A constant is a number, written so that its value is known by
2953 inspection, without knowing any context. Like this:
2956 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2957 .ascii "Ring the bell\7" # A string constant.
2958 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2959 .float 0f-314159265358979323846264338327\
2960 95028841971.693993751E-40 # - pi, a flonum.
2965 * Characters:: Character Constants
2966 * Numbers:: Number Constants
2970 @subsection Character Constants
2972 @cindex character constants
2973 @cindex constants, character
2974 There are two kinds of character constants. A @dfn{character} stands
2975 for one character in one byte and its value may be used in
2976 numeric expressions. String constants (properly called string
2977 @emph{literals}) are potentially many bytes and their values may not be
2978 used in arithmetic expressions.
2982 * Chars:: Characters
2986 @subsubsection Strings
2988 @cindex string constants
2989 @cindex constants, string
2990 A @dfn{string} is written between double-quotes. It may contain
2991 double-quotes or null characters. The way to get special characters
2992 into a string is to @dfn{escape} these characters: precede them with
2993 a backslash @samp{\} character. For example @samp{\\} represents
2994 one backslash: the first @code{\} is an escape which tells
2995 @command{@value{AS}} to interpret the second character literally as a backslash
2996 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2997 escape character). The complete list of escapes follows.
2999 @cindex escape codes, character
3000 @cindex character escape codes
3001 @c NOTE: Cindex entries must not start with a backlash character.
3002 @c NOTE: This confuses the pdf2texi script when it is creating the
3003 @c NOTE: index based upon the first character and so it generates:
3004 @c NOTE: \initial {\\}
3005 @c NOTE: which then results in the error message:
3006 @c NOTE: Argument of \\ has an extra }.
3007 @c NOTE: So in the index entries below a space character has been
3008 @c NOTE: prepended to avoid this problem.
3011 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
3013 @cindex @code{ \b} (backspace character)
3014 @cindex backspace (@code{\b})
3016 Mnemonic for backspace; for ASCII this is octal code 010.
3019 @c Mnemonic for EOText; for ASCII this is octal code 004.
3021 @cindex @code{ \f} (formfeed character)
3022 @cindex formfeed (@code{\f})
3024 Mnemonic for FormFeed; for ASCII this is octal code 014.
3026 @cindex @code{ \n} (newline character)
3027 @cindex newline (@code{\n})
3029 Mnemonic for newline; for ASCII this is octal code 012.
3032 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
3034 @cindex @code{ \r} (carriage return character)
3035 @cindex carriage return (@code{backslash-r})
3037 Mnemonic for carriage-Return; for ASCII this is octal code 015.
3040 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
3041 @c other assemblers.
3043 @cindex @code{ \t} (tab)
3044 @cindex tab (@code{\t})
3046 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
3049 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
3050 @c @item \x @var{digit} @var{digit} @var{digit}
3051 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
3053 @cindex @code{ \@var{ddd}} (octal character code)
3054 @cindex octal character code (@code{\@var{ddd}})
3055 @item \ @var{digit} @var{digit} @var{digit}
3056 An octal character code. The numeric code is 3 octal digits.
3057 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
3058 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3060 @cindex @code{ \@var{xd...}} (hex character code)
3061 @cindex hex character code (@code{\@var{xd...}})
3062 @item \@code{x} @var{hex-digits...}
3063 A hex character code. All trailing hex digits are combined. Either upper or
3064 lower case @code{x} works.
3066 @cindex @code{ \\} (@samp{\} character)
3067 @cindex backslash (@code{\\})
3069 Represents one @samp{\} character.
3072 @c Represents one @samp{'} (accent acute) character.
3073 @c This is needed in single character literals
3074 @c (@xref{Characters,,Character Constants}.) to represent
3077 @cindex @code{ \"} (doublequote character)
3078 @cindex doublequote (@code{\"})
3080 Represents one @samp{"} character. Needed in strings to represent
3081 this character, because an unescaped @samp{"} would end the string.
3083 @item \ @var{anything-else}
3084 Any other character when escaped by @kbd{\} gives a warning, but
3085 assembles as if the @samp{\} was not present. The idea is that if
3086 you used an escape sequence you clearly didn't want the literal
3087 interpretation of the following character. However @command{@value{AS}} has no
3088 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3089 code and warns you of the fact.
3092 Which characters are escapable, and what those escapes represent,
3093 varies widely among assemblers. The current set is what we think
3094 the BSD 4.2 assembler recognizes, and is a subset of what most C
3095 compilers recognize. If you are in doubt, do not use an escape
3099 @subsubsection Characters
3101 @cindex single character constant
3102 @cindex character, single
3103 @cindex constant, single character
3104 A single character may be written as a single quote immediately followed by
3105 that character. Some backslash escapes apply to characters, @code{\b},
3106 @code{\f}, @code{\n}, @code{\r}, @code{\t}, and @code{\"} with the same meaning
3107 as for strings, plus @code{\'} for a single quote. So if you want to write the
3108 character backslash, you must write @kbd{'\\} where the first @code{\} escapes
3109 the second @code{\}. As you can see, the quote is an acute accent, not a grave
3112 @ifclear abnormal-separator
3113 (or semicolon @samp{;})
3115 @ifset abnormal-separator
3117 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3122 immediately following an acute accent is taken as a literal character
3123 and does not count as the end of a statement. The value of a character
3124 constant in a numeric expression is the machine's byte-wide code for
3125 that character. @command{@value{AS}} assumes your character code is ASCII:
3126 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3129 @subsection Number Constants
3131 @cindex constants, number
3132 @cindex number constants
3133 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3134 are stored in the target machine. @emph{Integers} are numbers that
3135 would fit into an @code{int} in the C language. @emph{Bignums} are
3136 integers, but they are stored in more than 32 bits. @emph{Flonums}
3137 are floating point numbers, described below.
3140 * Integers:: Integers
3148 @subsubsection Integers
3150 @cindex constants, integer
3152 @cindex binary integers
3153 @cindex integers, binary
3154 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3155 the binary digits @samp{01}.
3157 @cindex octal integers
3158 @cindex integers, octal
3159 An octal integer is @samp{0} followed by zero or more of the octal
3160 digits (@samp{01234567}).
3162 @cindex decimal integers
3163 @cindex integers, decimal
3164 A decimal integer starts with a non-zero digit followed by zero or
3165 more digits (@samp{0123456789}).
3167 @cindex hexadecimal integers
3168 @cindex integers, hexadecimal
3169 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3170 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3172 Integers have the usual values. To denote a negative integer, use
3173 the prefix operator @samp{-} discussed under expressions
3174 (@pxref{Prefix Ops,,Prefix Operators}).
3177 @subsubsection Bignums
3180 @cindex constants, bignum
3181 A @dfn{bignum} has the same syntax and semantics as an integer
3182 except that the number (or its negative) takes more than 32 bits to
3183 represent in binary. The distinction is made because in some places
3184 integers are permitted while bignums are not.
3187 @subsubsection Flonums
3189 @cindex floating point numbers
3190 @cindex constants, floating point
3192 @cindex precision, floating point
3193 A @dfn{flonum} represents a floating point number. The translation is
3194 indirect: a decimal floating point number from the text is converted by
3195 @command{@value{AS}} to a generic binary floating point number of more than
3196 sufficient precision. This generic floating point number is converted
3197 to a particular computer's floating point format (or formats) by a
3198 portion of @command{@value{AS}} specialized to that computer.
3200 A flonum is written by writing (in order)
3205 (@samp{0} is optional on the HPPA.)
3209 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3211 @kbd{e} is recommended. Case is not important.
3213 @c FIXME: verify if flonum syntax really this vague for most cases
3214 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3215 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3218 On the H8/300 and Renesas / SuperH SH architectures, the letter must be
3219 one of the letters @samp{DFPRSX} (in upper or lower case).
3221 On the ARC, the letter must be one of the letters @samp{DFRS}
3222 (in upper or lower case).
3224 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3228 One of the letters @samp{DFRS} (in upper or lower case).
3231 One of the letters @samp{DFPRSX} (in upper or lower case).
3234 The letter @samp{E} (upper case only).
3239 An optional sign: either @samp{+} or @samp{-}.
3242 An optional @dfn{integer part}: zero or more decimal digits.
3245 An optional @dfn{fractional part}: @samp{.} followed by zero
3246 or more decimal digits.
3249 An optional exponent, consisting of:
3253 An @samp{E} or @samp{e}.
3254 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3255 @c principle this can perfectly well be different on different targets.
3257 Optional sign: either @samp{+} or @samp{-}.
3259 One or more decimal digits.
3264 At least one of the integer part or the fractional part must be
3265 present. The floating point number has the usual base-10 value.
3267 @command{@value{AS}} does all processing using integers. Flonums are computed
3268 independently of any floating point hardware in the computer running
3269 @command{@value{AS}}.
3272 @chapter Sections and Relocation
3277 * Secs Background:: Background
3278 * Ld Sections:: Linker Sections
3279 * As Sections:: Assembler Internal Sections
3280 * Sub-Sections:: Sub-Sections
3284 @node Secs Background
3287 Roughly, a section is a range of addresses, with no gaps; all data
3288 ``in'' those addresses is treated the same for some particular purpose.
3289 For example there may be a ``read only'' section.
3291 @cindex linker, and assembler
3292 @cindex assembler, and linker
3293 The linker @code{@value{LD}} reads many object files (partial programs) and
3294 combines their contents to form a runnable program. When @command{@value{AS}}
3295 emits an object file, the partial program is assumed to start at address 0.
3296 @code{@value{LD}} assigns the final addresses for the partial program, so that
3297 different partial programs do not overlap. This is actually an
3298 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3301 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3302 addresses. These blocks slide to their run-time addresses as rigid
3303 units; their length does not change and neither does the order of bytes
3304 within them. Such a rigid unit is called a @emph{section}. Assigning
3305 run-time addresses to sections is called @dfn{relocation}. It includes
3306 the task of adjusting mentions of object-file addresses so they refer to
3307 the proper run-time addresses.
3309 For the H8/300, and for the Renesas / SuperH SH,
3310 @command{@value{AS}} pads sections if needed to
3311 ensure they end on a word (sixteen bit) boundary.
3314 @cindex standard assembler sections
3315 An object file written by @command{@value{AS}} has at least three sections, any
3316 of which may be empty. These are named @dfn{text}, @dfn{data} and
3321 When it generates COFF or ELF output,
3323 @command{@value{AS}} can also generate whatever other named sections you specify
3324 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3325 If you do not use any directives that place output in the @samp{.text}
3326 or @samp{.data} sections, these sections still exist, but are empty.
3331 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3333 @command{@value{AS}} can also generate whatever other named sections you
3334 specify using the @samp{.space} and @samp{.subspace} directives. See
3335 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3336 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3337 assembler directives.
3340 Additionally, @command{@value{AS}} uses different names for the standard
3341 text, data, and bss sections when generating SOM output. Program text
3342 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3343 BSS into @samp{$BSS$}.
3347 Within the object file, the text section starts at address @code{0}, the
3348 data section follows, and the bss section follows the data section.
3351 When generating either SOM or ELF output files on the HPPA, the text
3352 section starts at address @code{0}, the data section at address
3353 @code{0x4000000}, and the bss section follows the data section.
3356 To let @code{@value{LD}} know which data changes when the sections are
3357 relocated, and how to change that data, @command{@value{AS}} also writes to the
3358 object file details of the relocation needed. To perform relocation
3359 @code{@value{LD}} must know, each time an address in the object
3363 Where in the object file is the beginning of this reference to
3366 How long (in bytes) is this reference?
3368 Which section does the address refer to? What is the numeric value of
3370 (@var{address}) @minus{} (@var{start-address of section})?
3373 Is the reference to an address ``Program-Counter relative''?
3376 @cindex addresses, format of
3377 @cindex section-relative addressing
3378 In fact, every address @command{@value{AS}} ever uses is expressed as
3380 (@var{section}) + (@var{offset into section})
3383 Further, most expressions @command{@value{AS}} computes have this section-relative
3386 (For some object formats, such as SOM for the HPPA, some expressions are
3387 symbol-relative instead.)
3390 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3391 @var{N} into section @var{secname}.''
3393 Apart from text, data and bss sections you need to know about the
3394 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3395 addresses in the absolute section remain unchanged. For example, address
3396 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3397 @code{@value{LD}}. Although the linker never arranges two partial programs'
3398 data sections with overlapping addresses after linking, @emph{by definition}
3399 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3400 part of a program is always the same address when the program is running as
3401 address @code{@{absolute@ 239@}} in any other part of the program.
3403 The idea of sections is extended to the @dfn{undefined} section. Any
3404 address whose section is unknown at assembly time is by definition
3405 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3406 Since numbers are always defined, the only way to generate an undefined
3407 address is to mention an undefined symbol. A reference to a named
3408 common block would be such a symbol: its value is unknown at assembly
3409 time so it has section @emph{undefined}.
3411 By analogy the word @emph{section} is used to describe groups of sections in
3412 the linked program. @code{@value{LD}} puts all partial programs' text
3413 sections in contiguous addresses in the linked program. It is
3414 customary to refer to the @emph{text section} of a program, meaning all
3415 the addresses of all partial programs' text sections. Likewise for
3416 data and bss sections.
3418 Some sections are manipulated by @code{@value{LD}}; others are invented for
3419 use of @command{@value{AS}} and have no meaning except during assembly.
3422 @section Linker Sections
3423 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3428 @cindex named sections
3429 @cindex sections, named
3430 @item named sections
3433 @cindex text section
3434 @cindex data section
3438 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3439 separate but equal sections. Anything you can say of one section is
3442 When the program is running, however, it is
3443 customary for the text section to be unalterable. The
3444 text section is often shared among processes: it contains
3445 instructions, constants and the like. The data section of a running
3446 program is usually alterable: for example, C variables would be stored
3447 in the data section.
3452 This section contains zeroed bytes when your program begins running. It
3453 is used to hold uninitialized variables or common storage. The length of
3454 each partial program's bss section is important, but because it starts
3455 out containing zeroed bytes there is no need to store explicit zero
3456 bytes in the object file. The bss section was invented to eliminate
3457 those explicit zeros from object files.
3459 @cindex absolute section
3460 @item absolute section
3461 Address 0 of this section is always ``relocated'' to runtime address 0.
3462 This is useful if you want to refer to an address that @code{@value{LD}} must
3463 not change when relocating. In this sense we speak of absolute
3464 addresses being ``unrelocatable'': they do not change during relocation.
3466 @cindex undefined section
3467 @item undefined section
3468 This ``section'' is a catch-all for address references to objects not in
3469 the preceding sections.
3470 @c FIXME: ref to some other doc on obj-file formats could go here.
3473 @cindex relocation example
3474 An idealized example of three relocatable sections follows.
3476 The example uses the traditional section names @samp{.text} and @samp{.data}.
3478 Memory addresses are on the horizontal axis.
3482 @c END TEXI2ROFF-KILL
3485 partial program # 1: |ttttt|dddd|00|
3492 partial program # 2: |TTT|DDD|000|
3495 +--+---+-----+--+----+---+-----+~~
3496 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3497 +--+---+-----+--+----+---+-----+~~
3499 addresses: 0 @dots{}
3506 \line{\it Partial program \#1: \hfil}
3507 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3508 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3510 \line{\it Partial program \#2: \hfil}
3511 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3512 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3514 \line{\it linked program: \hfil}
3515 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3516 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3517 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3518 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3520 \line{\it addresses: \hfil}
3524 @c END TEXI2ROFF-KILL
3527 @section Assembler Internal Sections
3529 @cindex internal assembler sections
3530 @cindex sections in messages, internal
3531 These sections are meant only for the internal use of @command{@value{AS}}. They
3532 have no meaning at run-time. You do not really need to know about these
3533 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3534 warning messages, so it might be helpful to have an idea of their
3535 meanings to @command{@value{AS}}. These sections are used to permit the
3536 value of every expression in your assembly language program to be a
3537 section-relative address.
3540 @cindex assembler internal logic error
3541 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3542 An internal assembler logic error has been found. This means there is a
3543 bug in the assembler.
3545 @cindex expr (internal section)
3547 The assembler stores complex expression internally as combinations of
3548 symbols. When it needs to represent an expression as a symbol, it puts
3549 it in the expr section.
3551 @c FIXME item transfer[t] vector preload
3552 @c FIXME item transfer[t] vector postload
3553 @c FIXME item register
3557 @section Sub-Sections
3559 @cindex numbered subsections
3560 @cindex grouping data
3566 fall into two sections: text and data.
3568 You may have separate groups of
3570 data in named sections
3574 data in named sections
3580 that you want to end up near to each other in the object file, even though they
3581 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3582 use @dfn{subsections} for this purpose. Within each section, there can be
3583 numbered subsections with values from 0 to 8192. Objects assembled into the
3584 same subsection go into the object file together with other objects in the same
3585 subsection. For example, a compiler might want to store constants in the text
3586 section, but might not want to have them interspersed with the program being
3587 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3588 section of code being output, and a @samp{.text 1} before each group of
3589 constants being output.
3591 Subsections are optional. If you do not use subsections, everything
3592 goes in subsection number zero.
3595 Each subsection is zero-padded up to a multiple of four bytes.
3596 (Subsections may be padded a different amount on different flavors
3597 of @command{@value{AS}}.)
3601 On the H8/300 platform, each subsection is zero-padded to a word
3602 boundary (two bytes).
3603 The same is true on the Renesas SH.
3607 Subsections appear in your object file in numeric order, lowest numbered
3608 to highest. (All this to be compatible with other people's assemblers.)
3609 The object file contains no representation of subsections; @code{@value{LD}} and
3610 other programs that manipulate object files see no trace of them.
3611 They just see all your text subsections as a text section, and all your
3612 data subsections as a data section.
3614 To specify which subsection you want subsequent statements assembled
3615 into, use a numeric argument to specify it, in a @samp{.text
3616 @var{expression}} or a @samp{.data @var{expression}} statement.
3619 When generating COFF output, you
3624 can also use an extra subsection
3625 argument with arbitrary named sections: @samp{.section @var{name},
3630 When generating ELF output, you
3635 can also use the @code{.subsection} directive (@pxref{SubSection})
3636 to specify a subsection: @samp{.subsection @var{expression}}.
3638 @var{Expression} should be an absolute expression
3639 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3640 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3641 begins in @code{text 0}. For instance:
3643 .text 0 # The default subsection is text 0 anyway.
3644 .ascii "This lives in the first text subsection. *"
3646 .ascii "But this lives in the second text subsection."
3648 .ascii "This lives in the data section,"
3649 .ascii "in the first data subsection."
3651 .ascii "This lives in the first text section,"
3652 .ascii "immediately following the asterisk (*)."
3655 Each section has a @dfn{location counter} incremented by one for every byte
3656 assembled into that section. Because subsections are merely a convenience
3657 restricted to @command{@value{AS}} there is no concept of a subsection location
3658 counter. There is no way to directly manipulate a location counter---but the
3659 @code{.align} directive changes it, and any label definition captures its
3660 current value. The location counter of the section where statements are being
3661 assembled is said to be the @dfn{active} location counter.
3664 @section bss Section
3667 @cindex common variable storage
3668 The bss section is used for local common variable storage.
3669 You may allocate address space in the bss section, but you may
3670 not dictate data to load into it before your program executes. When
3671 your program starts running, all the contents of the bss
3672 section are zeroed bytes.
3674 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3675 @ref{Lcomm,,@code{.lcomm}}.
3677 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3678 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3681 When assembling for a target which supports multiple sections, such as ELF or
3682 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3683 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3684 section. Typically the section will only contain symbol definitions and
3685 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3692 Symbols are a central concept: the programmer uses symbols to name
3693 things, the linker uses symbols to link, and the debugger uses symbols
3697 @cindex debuggers, and symbol order
3698 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3699 the same order they were declared. This may break some debuggers.
3704 * Setting Symbols:: Giving Symbols Other Values
3705 * Symbol Names:: Symbol Names
3706 * Dot:: The Special Dot Symbol
3707 * Symbol Attributes:: Symbol Attributes
3714 A @dfn{label} is written as a symbol immediately followed by a colon
3715 @samp{:}. The symbol then represents the current value of the
3716 active location counter, and is, for example, a suitable instruction
3717 operand. You are warned if you use the same symbol to represent two
3718 different locations: the first definition overrides any other
3722 On the HPPA, the usual form for a label need not be immediately followed by a
3723 colon, but instead must start in column zero. Only one label may be defined on
3724 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3725 provides a special directive @code{.label} for defining labels more flexibly.
3728 @node Setting Symbols
3729 @section Giving Symbols Other Values
3731 @cindex assigning values to symbols
3732 @cindex symbol values, assigning
3733 A symbol can be given an arbitrary value by writing a symbol, followed
3734 by an equals sign @samp{=}, followed by an expression
3735 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3736 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3737 equals sign @samp{=}@samp{=} here represents an equivalent of the
3738 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3741 Blackfin does not support symbol assignment with @samp{=}.
3745 @section Symbol Names
3747 @cindex symbol names
3748 @cindex names, symbol
3749 @ifclear SPECIAL-SYMS
3750 Symbol names begin with a letter or with one of @samp{._}. On most
3751 machines, you can also use @code{$} in symbol names; exceptions are
3752 noted in @ref{Machine Dependencies}. That character may be followed by any
3753 string of digits, letters, dollar signs (unless otherwise noted for a
3754 particular target machine), and underscores.
3758 Symbol names begin with a letter or with one of @samp{._}. On the
3759 Renesas SH you can also use @code{$} in symbol names. That
3760 character may be followed by any string of digits, letters, dollar signs (save
3761 on the H8/300), and underscores.
3765 Case of letters is significant: @code{foo} is a different symbol name
3768 Symbol names do not start with a digit. An exception to this rule is made for
3769 Local Labels. See below.
3771 Multibyte characters are supported. To generate a symbol name containing
3772 multibyte characters enclose it within double quotes and use escape codes. cf
3773 @xref{Strings}. Generating a multibyte symbol name from a label is not
3774 currently supported.
3776 Each symbol has exactly one name. Each name in an assembly language program
3777 refers to exactly one symbol. You may use that symbol name any number of times
3780 @subheading Local Symbol Names
3782 @cindex local symbol names
3783 @cindex symbol names, local
3784 A local symbol is any symbol beginning with certain local label prefixes.
3785 By default, the local label prefix is @samp{.L} for ELF systems or
3786 @samp{L} for traditional a.out systems, but each target may have its own
3787 set of local label prefixes.
3789 On the HPPA local symbols begin with @samp{L$}.
3792 Local symbols are defined and used within the assembler, but they are
3793 normally not saved in object files. Thus, they are not visible when debugging.
3794 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3795 to retain the local symbols in the object files.
3797 @subheading Local Labels
3799 @cindex local labels
3800 @cindex temporary symbol names
3801 @cindex symbol names, temporary
3802 Local labels are different from local symbols. Local labels help compilers and
3803 programmers use names temporarily. They create symbols which are guaranteed to
3804 be unique over the entire scope of the input source code and which can be
3805 referred to by a simple notation. To define a local label, write a label of
3806 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3807 To refer to the most recent previous definition of that label write
3808 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3809 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3810 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3812 There is no restriction on how you can use these labels, and you can reuse them
3813 too. So that it is possible to repeatedly define the same local label (using
3814 the same number @samp{@b{N}}), although you can only refer to the most recently
3815 defined local label of that number (for a backwards reference) or the next
3816 definition of a specific local label for a forward reference. It is also worth
3817 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3818 implemented in a slightly more efficient manner than the others.
3829 Which is the equivalent of:
3832 label_1: branch label_3
3833 label_2: branch label_1
3834 label_3: branch label_4
3835 label_4: branch label_3
3838 Local label names are only a notational device. They are immediately
3839 transformed into more conventional symbol names before the assembler uses them.
3840 The symbol names are stored in the symbol table, appear in error messages, and
3841 are optionally emitted to the object file. The names are constructed using
3845 @item @emph{local label prefix}
3846 All local symbols begin with the system-specific local label prefix.
3847 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3848 that start with the local label prefix. These labels are
3849 used for symbols you are never intended to see. If you use the
3850 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3851 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3852 you may use them in debugging.
3855 This is the number that was used in the local label definition. So if the
3856 label is written @samp{55:} then the number is @samp{55}.
3859 This unusual character is included so you do not accidentally invent a symbol
3860 of the same name. The character has ASCII value of @samp{\002} (control-B).
3862 @item @emph{ordinal number}
3863 This is a serial number to keep the labels distinct. The first definition of
3864 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3865 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3866 the number @samp{1} and its 15th definition gets @samp{15} as well.
3869 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3870 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3872 @subheading Dollar Local Labels
3873 @cindex dollar local symbols
3875 On some targets @code{@value{AS}} also supports an even more local form of
3876 local labels called dollar labels. These labels go out of scope (i.e., they
3877 become undefined) as soon as a non-local label is defined. Thus they remain
3878 valid for only a small region of the input source code. Normal local labels,
3879 by contrast, remain in scope for the entire file, or until they are redefined
3880 by another occurrence of the same local label.
3882 Dollar labels are defined in exactly the same way as ordinary local labels,
3883 except that they have a dollar sign suffix to their numeric value, e.g.,
3886 They can also be distinguished from ordinary local labels by their transformed
3887 names which use ASCII character @samp{\001} (control-A) as the magic character
3888 to distinguish them from ordinary labels. For example, the fifth definition of
3889 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3892 @section The Special Dot Symbol
3894 @cindex dot (symbol)
3895 @cindex @code{.} (symbol)
3896 @cindex current address
3897 @cindex location counter
3898 The special symbol @samp{.} refers to the current address that
3899 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3900 .long .} defines @code{melvin} to contain its own address.
3901 Assigning a value to @code{.} is treated the same as a @code{.org}
3903 @ifclear no-space-dir
3904 Thus, the expression @samp{.=.+4} is the same as saying
3908 @node Symbol Attributes
3909 @section Symbol Attributes
3911 @cindex symbol attributes
3912 @cindex attributes, symbol
3913 Every symbol has, as well as its name, the attributes ``Value'' and
3914 ``Type''. Depending on output format, symbols can also have auxiliary
3917 The detailed definitions are in @file{a.out.h}.
3920 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3921 all these attributes, and probably won't warn you. This makes the
3922 symbol an externally defined symbol, which is generally what you
3926 * Symbol Value:: Value
3927 * Symbol Type:: Type
3929 * a.out Symbols:: Symbol Attributes: @code{a.out}
3932 * COFF Symbols:: Symbol Attributes for COFF
3935 * SOM Symbols:: Symbol Attributes for SOM
3942 @cindex value of a symbol
3943 @cindex symbol value
3944 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3945 location in the text, data, bss or absolute sections the value is the
3946 number of addresses from the start of that section to the label.
3947 Naturally for text, data and bss sections the value of a symbol changes
3948 as @code{@value{LD}} changes section base addresses during linking. Absolute
3949 symbols' values do not change during linking: that is why they are
3952 The value of an undefined symbol is treated in a special way. If it is
3953 0 then the symbol is not defined in this assembler source file, and
3954 @code{@value{LD}} tries to determine its value from other files linked into the
3955 same program. You make this kind of symbol simply by mentioning a symbol
3956 name without defining it. A non-zero value represents a @code{.comm}
3957 common declaration. The value is how much common storage to reserve, in
3958 bytes (addresses). The symbol refers to the first address of the
3964 @cindex type of a symbol
3966 The type attribute of a symbol contains relocation (section)
3967 information, any flag settings indicating that a symbol is external, and
3968 (optionally), other information for linkers and debuggers. The exact
3969 format depends on the object-code output format in use.
3973 @subsection Symbol Attributes: @code{a.out}
3975 @cindex @code{a.out} symbol attributes
3976 @cindex symbol attributes, @code{a.out}
3979 * Symbol Desc:: Descriptor
3980 * Symbol Other:: Other
3984 @subsubsection Descriptor
3986 @cindex descriptor, of @code{a.out} symbol
3987 This is an arbitrary 16-bit value. You may establish a symbol's
3988 descriptor value by using a @code{.desc} statement
3989 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3990 @command{@value{AS}}.
3993 @subsubsection Other
3995 @cindex other attribute, of @code{a.out} symbol
3996 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
4001 @subsection Symbol Attributes for COFF
4003 @cindex COFF symbol attributes
4004 @cindex symbol attributes, COFF
4006 The COFF format supports a multitude of auxiliary symbol attributes;
4007 like the primary symbol attributes, they are set between @code{.def} and
4008 @code{.endef} directives.
4010 @subsubsection Primary Attributes
4012 @cindex primary attributes, COFF symbols
4013 The symbol name is set with @code{.def}; the value and type,
4014 respectively, with @code{.val} and @code{.type}.
4016 @subsubsection Auxiliary Attributes
4018 @cindex auxiliary attributes, COFF symbols
4019 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
4020 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
4021 table information for COFF.
4026 @subsection Symbol Attributes for SOM
4028 @cindex SOM symbol attributes
4029 @cindex symbol attributes, SOM
4031 The SOM format for the HPPA supports a multitude of symbol attributes set with
4032 the @code{.EXPORT} and @code{.IMPORT} directives.
4034 The attributes are described in @cite{HP9000 Series 800 Assembly
4035 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
4036 @code{EXPORT} assembler directive documentation.
4040 @chapter Expressions
4044 @cindex numeric values
4045 An @dfn{expression} specifies an address or numeric value.
4046 Whitespace may precede and/or follow an expression.
4048 The result of an expression must be an absolute number, or else an offset into
4049 a particular section. If an expression is not absolute, and there is not
4050 enough information when @command{@value{AS}} sees the expression to know its
4051 section, a second pass over the source program might be necessary to interpret
4052 the expression---but the second pass is currently not implemented.
4053 @command{@value{AS}} aborts with an error message in this situation.
4056 * Empty Exprs:: Empty Expressions
4057 * Integer Exprs:: Integer Expressions
4061 @section Empty Expressions
4063 @cindex empty expressions
4064 @cindex expressions, empty
4065 An empty expression has no value: it is just whitespace or null.
4066 Wherever an absolute expression is required, you may omit the
4067 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4068 is compatible with other assemblers.
4071 @section Integer Expressions
4073 @cindex integer expressions
4074 @cindex expressions, integer
4075 An @dfn{integer expression} is one or more @emph{arguments} delimited
4076 by @emph{operators}.
4079 * Arguments:: Arguments
4080 * Operators:: Operators
4081 * Prefix Ops:: Prefix Operators
4082 * Infix Ops:: Infix Operators
4086 @subsection Arguments
4088 @cindex expression arguments
4089 @cindex arguments in expressions
4090 @cindex operands in expressions
4091 @cindex arithmetic operands
4092 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4093 contexts arguments are sometimes called ``arithmetic operands''. In
4094 this manual, to avoid confusing them with the ``instruction operands'' of
4095 the machine language, we use the term ``argument'' to refer to parts of
4096 expressions only, reserving the word ``operand'' to refer only to machine
4097 instruction operands.
4099 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4100 @var{section} is one of text, data, bss, absolute,
4101 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4104 Numbers are usually integers.
4106 A number can be a flonum or bignum. In this case, you are warned
4107 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4108 these 32 bits are an integer. You may write integer-manipulating
4109 instructions that act on exotic constants, compatible with other
4112 @cindex subexpressions
4113 Subexpressions are a left parenthesis @samp{(} followed by an integer
4114 expression, followed by a right parenthesis @samp{)}; or a prefix
4115 operator followed by an argument.
4118 @subsection Operators
4120 @cindex operators, in expressions
4121 @cindex arithmetic functions
4122 @cindex functions, in expressions
4123 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4124 operators are followed by an argument. Infix operators appear
4125 between their arguments. Operators may be preceded and/or followed by
4129 @subsection Prefix Operator
4131 @cindex prefix operators
4132 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4133 one argument, which must be absolute.
4135 @c the tex/end tex stuff surrounding this small table is meant to make
4136 @c it align, on the printed page, with the similar table in the next
4137 @c section (which is inside an enumerate).
4139 \global\advance\leftskip by \itemindent
4144 @dfn{Negation}. Two's complement negation.
4146 @dfn{Complementation}. Bitwise not.
4150 \global\advance\leftskip by -\itemindent
4154 @subsection Infix Operators
4156 @cindex infix operators
4157 @cindex operators, permitted arguments
4158 @dfn{Infix operators} take two arguments, one on either side. Operators
4159 have precedence, but operations with equal precedence are performed left
4160 to right. Apart from @code{+} or @option{-}, both arguments must be
4161 absolute, and the result is absolute.
4164 @cindex operator precedence
4165 @cindex precedence of operators
4172 @dfn{Multiplication}.
4175 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4181 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4184 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4188 Intermediate precedence
4193 @dfn{Bitwise Inclusive Or}.
4199 @dfn{Bitwise Exclusive Or}.
4202 @dfn{Bitwise Or Not}.
4209 @cindex addition, permitted arguments
4210 @cindex plus, permitted arguments
4211 @cindex arguments for addition
4213 @dfn{Addition}. If either argument is absolute, the result has the section of
4214 the other argument. You may not add together arguments from different
4217 @cindex subtraction, permitted arguments
4218 @cindex minus, permitted arguments
4219 @cindex arguments for subtraction
4221 @dfn{Subtraction}. If the right argument is absolute, the
4222 result has the section of the left argument.
4223 If both arguments are in the same section, the result is absolute.
4224 You may not subtract arguments from different sections.
4225 @c FIXME is there still something useful to say about undefined - undefined ?
4227 @cindex comparison expressions
4228 @cindex expressions, comparison
4233 @dfn{Is Not Equal To}
4237 @dfn{Is Greater Than}
4239 @dfn{Is Greater Than Or Equal To}
4241 @dfn{Is Less Than Or Equal To}
4243 The comparison operators can be used as infix operators. A true results has a
4244 value of -1 whereas a false result has a value of 0. Note, these operators
4245 perform signed comparisons.
4248 @item Lowest Precedence
4257 These two logical operations can be used to combine the results of sub
4258 expressions. Note, unlike the comparison operators a true result returns a
4259 value of 1 but a false results does still return 0. Also note that the logical
4260 or operator has a slightly lower precedence than logical and.
4265 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4266 address; you can only have a defined section in one of the two arguments.
4269 @chapter Assembler Directives
4271 @cindex directives, machine independent
4272 @cindex pseudo-ops, machine independent
4273 @cindex machine independent directives
4274 All assembler directives have names that begin with a period (@samp{.}).
4275 The names are case insensitive for most targets, and usually written
4278 This chapter discusses directives that are available regardless of the
4279 target machine configuration for the @sc{gnu} assembler.
4281 Some machine configurations provide additional directives.
4282 @xref{Machine Dependencies}.
4285 @ifset machine-directives
4286 @xref{Machine Dependencies}, for additional directives.
4291 * Abort:: @code{.abort}
4293 * ABORT (COFF):: @code{.ABORT}
4296 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4297 * Altmacro:: @code{.altmacro}
4298 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4299 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4300 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4301 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4302 * Byte:: @code{.byte @var{expressions}}
4303 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4304 * Comm:: @code{.comm @var{symbol} , @var{length} }
4305 * Data:: @code{.data @var{subsection}}
4306 * Dc:: @code{.dc[@var{size}] @var{expressions}}
4307 * Dcb:: @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4308 * Ds:: @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4310 * Def:: @code{.def @var{name}}
4313 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4319 * Double:: @code{.double @var{flonums}}
4320 * Eject:: @code{.eject}
4321 * Else:: @code{.else}
4322 * Elseif:: @code{.elseif}
4325 * Endef:: @code{.endef}
4328 * Endfunc:: @code{.endfunc}
4329 * Endif:: @code{.endif}
4330 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4331 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4332 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4334 * Error:: @code{.error @var{string}}
4335 * Exitm:: @code{.exitm}
4336 * Extern:: @code{.extern}
4337 * Fail:: @code{.fail}
4338 * File:: @code{.file}
4339 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4340 * Float:: @code{.float @var{flonums}}
4341 * Func:: @code{.func}
4342 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4344 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4345 * Hidden:: @code{.hidden @var{names}}
4348 * hword:: @code{.hword @var{expressions}}
4349 * Ident:: @code{.ident}
4350 * If:: @code{.if @var{absolute expression}}
4351 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4352 * Include:: @code{.include "@var{file}"}
4353 * Int:: @code{.int @var{expressions}}
4355 * Internal:: @code{.internal @var{names}}
4358 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4359 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4360 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4361 * Lflags:: @code{.lflags}
4362 @ifclear no-line-dir
4363 * Line:: @code{.line @var{line-number}}
4366 * Linkonce:: @code{.linkonce [@var{type}]}
4367 * List:: @code{.list}
4368 * Ln:: @code{.ln @var{line-number}}
4369 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4370 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4372 * Local:: @code{.local @var{names}}
4375 * Long:: @code{.long @var{expressions}}
4377 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4380 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4381 * MRI:: @code{.mri @var{val}}
4382 * Noaltmacro:: @code{.noaltmacro}
4383 * Nolist:: @code{.nolist}
4384 * Nops:: @code{.nops @var{size}[, @var{control}]}
4385 * Octa:: @code{.octa @var{bignums}}
4386 * Offset:: @code{.offset @var{loc}}
4387 * Org:: @code{.org @var{new-lc}, @var{fill}}
4388 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4390 * PopSection:: @code{.popsection}
4391 * Previous:: @code{.previous}
4394 * Print:: @code{.print @var{string}}
4396 * Protected:: @code{.protected @var{names}}
4399 * Psize:: @code{.psize @var{lines}, @var{columns}}
4400 * Purgem:: @code{.purgem @var{name}}
4402 * PushSection:: @code{.pushsection @var{name}}
4405 * Quad:: @code{.quad @var{bignums}}
4406 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4407 * Rept:: @code{.rept @var{count}}
4408 * Sbttl:: @code{.sbttl "@var{subheading}"}
4410 * Scl:: @code{.scl @var{class}}
4413 * Section:: @code{.section @var{name}[, @var{flags}]}
4416 * Set:: @code{.set @var{symbol}, @var{expression}}
4417 * Short:: @code{.short @var{expressions}}
4418 * Single:: @code{.single @var{flonums}}
4420 * Size:: @code{.size [@var{name} , @var{expression}]}
4422 @ifclear no-space-dir
4423 * Skip:: @code{.skip @var{size} [,@var{fill}]}
4426 * Sleb128:: @code{.sleb128 @var{expressions}}
4427 @ifclear no-space-dir
4428 * Space:: @code{.space @var{size} [,@var{fill}]}
4431 * Stab:: @code{.stabd, .stabn, .stabs}
4434 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4435 * Struct:: @code{.struct @var{expression}}
4437 * SubSection:: @code{.subsection}
4438 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4442 * Tag:: @code{.tag @var{structname}}
4445 * Text:: @code{.text @var{subsection}}
4446 * Title:: @code{.title "@var{heading}"}
4448 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4451 * Uleb128:: @code{.uleb128 @var{expressions}}
4453 * Val:: @code{.val @var{addr}}
4457 * Version:: @code{.version "@var{string}"}
4458 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4459 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4462 * Warning:: @code{.warning @var{string}}
4463 * Weak:: @code{.weak @var{names}}
4464 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4465 * Word:: @code{.word @var{expressions}}
4466 @ifclear no-space-dir
4467 * Zero:: @code{.zero @var{size}}
4470 * 2byte:: @code{.2byte @var{expressions}}
4471 * 4byte:: @code{.4byte @var{expressions}}
4472 * 8byte:: @code{.8byte @var{bignums}}
4474 * Deprecated:: Deprecated Directives
4478 @section @code{.abort}
4480 @cindex @code{abort} directive
4481 @cindex stopping the assembly
4482 This directive stops the assembly immediately. It is for
4483 compatibility with other assemblers. The original idea was that the
4484 assembly language source would be piped into the assembler. If the sender
4485 of the source quit, it could use this directive tells @command{@value{AS}} to
4486 quit also. One day @code{.abort} will not be supported.
4490 @section @code{.ABORT} (COFF)
4492 @cindex @code{ABORT} directive
4493 When producing COFF output, @command{@value{AS}} accepts this directive as a
4494 synonym for @samp{.abort}.
4499 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4501 @cindex padding the location counter
4502 @cindex @code{align} directive
4503 Pad the location counter (in the current subsection) to a particular storage
4504 boundary. The first expression (which must be absolute) is the alignment
4505 required, as described below.
4507 The second expression (also absolute) gives the fill value to be stored in the
4508 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4509 padding bytes are normally zero. However, on most systems, if the section is
4510 marked as containing code and the fill value is omitted, the space is filled
4511 with no-op instructions.
4513 The third expression is also absolute, and is also optional. If it is present,
4514 it is the maximum number of bytes that should be skipped by this alignment
4515 directive. If doing the alignment would require skipping more bytes than the
4516 specified maximum, then the alignment is not done at all. You can omit the
4517 fill value (the second argument) entirely by simply using two commas after the
4518 required alignment; this can be useful if you want the alignment to be filled
4519 with no-op instructions when appropriate.
4521 The way the required alignment is specified varies from system to system.
4522 For the arc, hppa, i386 using ELF, iq2000, m68k, or1k,
4523 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4524 alignment request in bytes. For example @samp{.align 8} advances
4525 the location counter until it is a multiple of 8. If the location counter
4526 is already a multiple of 8, no change is needed. For the tic54x, the
4527 first expression is the alignment request in words.
4529 For other systems, including ppc, i386 using a.out format, arm and
4530 strongarm, it is the
4531 number of low-order zero bits the location counter must have after
4532 advancement. For example @samp{.align 3} advances the location
4533 counter until it a multiple of 8. If the location counter is already a
4534 multiple of 8, no change is needed.
4536 This inconsistency is due to the different behaviors of the various
4537 native assemblers for these systems which GAS must emulate.
4538 GAS also provides @code{.balign} and @code{.p2align} directives,
4539 described later, which have a consistent behavior across all
4540 architectures (but are specific to GAS).
4543 @section @code{.altmacro}
4544 Enable alternate macro mode, enabling:
4547 @item LOCAL @var{name} [ , @dots{} ]
4548 One additional directive, @code{LOCAL}, is available. It is used to
4549 generate a string replacement for each of the @var{name} arguments, and
4550 replace any instances of @var{name} in each macro expansion. The
4551 replacement string is unique in the assembly, and different for each
4552 separate macro expansion. @code{LOCAL} allows you to write macros that
4553 define symbols, without fear of conflict between separate macro expansions.
4555 @item String delimiters
4556 You can write strings delimited in these other ways besides
4557 @code{"@var{string}"}:
4560 @item '@var{string}'
4561 You can delimit strings with single-quote characters.
4563 @item <@var{string}>
4564 You can delimit strings with matching angle brackets.
4567 @item single-character string escape
4568 To include any single character literally in a string (even if the
4569 character would otherwise have some special meaning), you can prefix the
4570 character with @samp{!} (an exclamation mark). For example, you can
4571 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4573 @item Expression results as strings
4574 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4575 and use the result as a string.
4579 @section @code{.ascii "@var{string}"}@dots{}
4581 @cindex @code{ascii} directive
4582 @cindex string literals
4583 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4584 separated by commas. It assembles each string (with no automatic
4585 trailing zero byte) into consecutive addresses.
4588 @section @code{.asciz "@var{string}"}@dots{}
4590 @cindex @code{asciz} directive
4591 @cindex zero-terminated strings
4592 @cindex null-terminated strings
4593 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4594 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4597 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4599 @cindex padding the location counter given number of bytes
4600 @cindex @code{balign} directive
4601 Pad the location counter (in the current subsection) to a particular
4602 storage boundary. The first expression (which must be absolute) is the
4603 alignment request in bytes. For example @samp{.balign 8} advances
4604 the location counter until it is a multiple of 8. If the location counter
4605 is already a multiple of 8, no change is needed.
4607 The second expression (also absolute) gives the fill value to be stored in the
4608 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4609 padding bytes are normally zero. However, on most systems, if the section is
4610 marked as containing code and the fill value is omitted, the space is filled
4611 with no-op instructions.
4613 The third expression is also absolute, and is also optional. If it is present,
4614 it is the maximum number of bytes that should be skipped by this alignment
4615 directive. If doing the alignment would require skipping more bytes than the
4616 specified maximum, then the alignment is not done at all. You can omit the
4617 fill value (the second argument) entirely by simply using two commas after the
4618 required alignment; this can be useful if you want the alignment to be filled
4619 with no-op instructions when appropriate.
4621 @cindex @code{balignw} directive
4622 @cindex @code{balignl} directive
4623 The @code{.balignw} and @code{.balignl} directives are variants of the
4624 @code{.balign} directive. The @code{.balignw} directive treats the fill
4625 pattern as a two byte word value. The @code{.balignl} directives treats the
4626 fill pattern as a four byte longword value. For example, @code{.balignw
4627 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4628 filled in with the value 0x368d (the exact placement of the bytes depends upon
4629 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4632 @node Bundle directives
4633 @section Bundle directives
4634 @subsection @code{.bundle_align_mode @var{abs-expr}}
4635 @cindex @code{bundle_align_mode} directive
4637 @cindex instruction bundle
4638 @cindex aligned instruction bundle
4639 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4640 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4641 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4642 disabled (which is the default state). If the argument it not zero, it
4643 gives the size of an instruction bundle as a power of two (as for the
4644 @code{.p2align} directive, @pxref{P2align}).
4646 For some targets, it's an ABI requirement that no instruction may span a
4647 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4648 instructions that starts on an aligned boundary. For example, if
4649 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4650 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4651 effect, no single instruction may span a boundary between bundles. If an
4652 instruction would start too close to the end of a bundle for the length of
4653 that particular instruction to fit within the bundle, then the space at the
4654 end of that bundle is filled with no-op instructions so the instruction
4655 starts in the next bundle. As a corollary, it's an error if any single
4656 instruction's encoding is longer than the bundle size.
4658 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4659 @cindex @code{bundle_lock} directive
4660 @cindex @code{bundle_unlock} directive
4661 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4662 allow explicit control over instruction bundle padding. These directives
4663 are only valid when @code{.bundle_align_mode} has been used to enable
4664 aligned instruction bundle mode. It's an error if they appear when
4665 @code{.bundle_align_mode} has not been used at all, or when the last
4666 directive was @w{@code{.bundle_align_mode 0}}.
4668 @cindex bundle-locked
4669 For some targets, it's an ABI requirement that certain instructions may
4670 appear only as part of specified permissible sequences of multiple
4671 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4672 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4673 instruction sequence. For purposes of aligned instruction bundle mode, a
4674 sequence starting with @code{.bundle_lock} and ending with
4675 @code{.bundle_unlock} is treated as a single instruction. That is, the
4676 entire sequence must fit into a single bundle and may not span a bundle
4677 boundary. If necessary, no-op instructions will be inserted before the
4678 first instruction of the sequence so that the whole sequence starts on an
4679 aligned bundle boundary. It's an error if the sequence is longer than the
4682 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4683 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4684 nested. That is, a second @code{.bundle_lock} directive before the next
4685 @code{.bundle_unlock} directive has no effect except that it must be
4686 matched by another closing @code{.bundle_unlock} so that there is the
4687 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4690 @section @code{.byte @var{expressions}}
4692 @cindex @code{byte} directive
4693 @cindex integers, one byte
4694 @code{.byte} expects zero or more expressions, separated by commas.
4695 Each expression is assembled into the next byte.
4697 @node CFI directives
4698 @section CFI directives
4699 @subsection @code{.cfi_sections @var{section_list}}
4700 @cindex @code{cfi_sections} directive
4701 @code{.cfi_sections} may be used to specify whether CFI directives
4702 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4703 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4704 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4705 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4706 directive is not used is @code{.cfi_sections .eh_frame}.
4708 On targets that support compact unwinding tables these can be generated
4709 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4711 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4712 which is used by the @value{TIC6X} target.
4714 The @code{.cfi_sections} directive can be repeated, with the same or different
4715 arguments, provided that CFI generation has not yet started. Once CFI
4716 generation has started however the section list is fixed and any attempts to
4717 redefine it will result in an error.
4719 @subsection @code{.cfi_startproc [simple]}
4720 @cindex @code{cfi_startproc} directive
4721 @code{.cfi_startproc} is used at the beginning of each function that
4722 should have an entry in @code{.eh_frame}. It initializes some internal
4723 data structures. Don't forget to close the function by
4724 @code{.cfi_endproc}.
4726 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4727 it also emits some architecture dependent initial CFI instructions.
4729 @subsection @code{.cfi_endproc}
4730 @cindex @code{cfi_endproc} directive
4731 @code{.cfi_endproc} is used at the end of a function where it closes its
4732 unwind entry previously opened by
4733 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4735 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4736 @cindex @code{cfi_personality} directive
4737 @code{.cfi_personality} defines personality routine and its encoding.
4738 @var{encoding} must be a constant determining how the personality
4739 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4740 argument is not present, otherwise second argument should be
4741 a constant or a symbol name. When using indirect encodings,
4742 the symbol provided should be the location where personality
4743 can be loaded from, not the personality routine itself.
4744 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4745 no personality routine.
4747 @subsection @code{.cfi_personality_id @var{id}}
4748 @cindex @code{cfi_personality_id} directive
4749 @code{cfi_personality_id} defines a personality routine by its index as
4750 defined in a compact unwinding format.
4751 Only valid when generating compact EH frames (i.e.
4752 with @code{.cfi_sections eh_frame_entry}.
4754 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4755 @cindex @code{cfi_fde_data} directive
4756 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4757 used for the current function. These are emitted inline in the
4758 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4759 in the @code{.gnu.extab} section otherwise.
4760 Only valid when generating compact EH frames (i.e.
4761 with @code{.cfi_sections eh_frame_entry}.
4763 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4764 @code{.cfi_lsda} defines LSDA and its encoding.
4765 @var{encoding} must be a constant determining how the LSDA
4766 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4767 argument is not present, otherwise the second argument should be a constant
4768 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4769 meaning that no LSDA is present.
4771 @subsection @code{.cfi_inline_lsda} [@var{align}]
4772 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4773 switches to the corresponding @code{.gnu.extab} section.
4774 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4775 Only valid when generating compact EH frames (i.e.
4776 with @code{.cfi_sections eh_frame_entry}.
4778 The table header and unwinding opcodes will be generated at this point,
4779 so that they are immediately followed by the LSDA data. The symbol
4780 referenced by the @code{.cfi_lsda} directive should still be defined
4781 in case a fallback FDE based encoding is used. The LSDA data is terminated
4782 by a section directive.
4784 The optional @var{align} argument specifies the alignment required.
4785 The alignment is specified as a power of two, as with the
4786 @code{.p2align} directive.
4788 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4789 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4790 address from @var{register} and add @var{offset} to it}.
4792 @subsection @code{.cfi_def_cfa_register @var{register}}
4793 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4794 now on @var{register} will be used instead of the old one. Offset
4797 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4798 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4799 remains the same, but @var{offset} is new. Note that it is the
4800 absolute offset that will be added to a defined register to compute
4803 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4804 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4805 value that is added/subtracted from the previous offset.
4807 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4808 Previous value of @var{register} is saved at offset @var{offset} from
4811 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4812 Previous value of @var{register} is CFA + @var{offset}.
4814 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4815 Previous value of @var{register} is saved at offset @var{offset} from
4816 the current CFA register. This is transformed to @code{.cfi_offset}
4817 using the known displacement of the CFA register from the CFA.
4818 This is often easier to use, because the number will match the
4819 code it's annotating.
4821 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4822 Previous value of @var{register1} is saved in register @var{register2}.
4824 @subsection @code{.cfi_restore @var{register}}
4825 @code{.cfi_restore} says that the rule for @var{register} is now the
4826 same as it was at the beginning of the function, after all initial
4827 instruction added by @code{.cfi_startproc} were executed.
4829 @subsection @code{.cfi_undefined @var{register}}
4830 From now on the previous value of @var{register} can't be restored anymore.
4832 @subsection @code{.cfi_same_value @var{register}}
4833 Current value of @var{register} is the same like in the previous frame,
4834 i.e. no restoration needed.
4836 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4837 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4838 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4839 places them in the current row. This is useful for situations where you have
4840 multiple @code{.cfi_*} directives that need to be undone due to the control
4841 flow of the program. For example, we could have something like this (assuming
4842 the CFA is the value of @code{rbp}):
4852 .cfi_def_cfa %rsp, 8
4855 /* Do something else */
4858 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4859 to the instructions before @code{label}. This means we'd have to add multiple
4860 @code{.cfi} directives after @code{label} to recreate the original save
4861 locations of the registers, as well as setting the CFA back to the value of
4862 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4874 .cfi_def_cfa %rsp, 8
4878 /* Do something else */
4881 That way, the rules for the instructions after @code{label} will be the same
4882 as before the first @code{.cfi_restore} without having to use multiple
4883 @code{.cfi} directives.
4885 @subsection @code{.cfi_return_column @var{register}}
4886 Change return column @var{register}, i.e. the return address is either
4887 directly in @var{register} or can be accessed by rules for @var{register}.
4889 @subsection @code{.cfi_signal_frame}
4890 Mark current function as signal trampoline.
4892 @subsection @code{.cfi_window_save}
4893 SPARC register window has been saved.
4895 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4896 Allows the user to add arbitrary bytes to the unwind info. One
4897 might use this to add OS-specific CFI opcodes, or generic CFI
4898 opcodes that GAS does not yet support.
4900 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4901 The current value of @var{register} is @var{label}. The value of @var{label}
4902 will be encoded in the output file according to @var{encoding}; see the
4903 description of @code{.cfi_personality} for details on this encoding.
4905 The usefulness of equating a register to a fixed label is probably
4906 limited to the return address register. Here, it can be useful to
4907 mark a code segment that has only one return address which is reached
4908 by a direct branch and no copy of the return address exists in memory
4909 or another register.
4912 @section @code{.comm @var{symbol} , @var{length} }
4914 @cindex @code{comm} directive
4915 @cindex symbol, common
4916 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4917 common symbol in one object file may be merged with a defined or common symbol
4918 of the same name in another object file. If @code{@value{LD}} does not see a
4919 definition for the symbol--just one or more common symbols--then it will
4920 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4921 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4922 the same name, and they do not all have the same size, it will allocate space
4923 using the largest size.
4926 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4927 an optional third argument. This is the desired alignment of the symbol,
4928 specified for ELF as a byte boundary (for example, an alignment of 16 means
4929 that the least significant 4 bits of the address should be zero), and for PE
4930 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4931 boundary). The alignment must be an absolute expression, and it must be a
4932 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4933 common symbol, it will use the alignment when placing the symbol. If no
4934 alignment is specified, @command{@value{AS}} will set the alignment to the
4935 largest power of two less than or equal to the size of the symbol, up to a
4936 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4937 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4938 @samp{--section-alignment} option; image file sections in PE are aligned to
4939 multiples of 4096, which is far too large an alignment for ordinary variables.
4940 It is rather the default alignment for (non-debug) sections within object
4941 (@samp{*.o}) files, which are less strictly aligned.}.
4945 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4946 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4950 @section @code{.data @var{subsection}}
4951 @cindex @code{data} directive
4953 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4954 end of the data subsection numbered @var{subsection} (which is an
4955 absolute expression). If @var{subsection} is omitted, it defaults
4959 @section @code{.dc[@var{size}] @var{expressions}}
4960 @cindex @code{dc} directive
4962 The @code{.dc} directive expects zero or more @var{expressions} separated by
4963 commas. These expressions are evaluated and their values inserted into the
4964 current section. The size of the emitted value depends upon the suffix to the
4965 @code{.dc} directive:
4969 Emits N-bit values, where N is the size of an address on the target system.
4973 Emits double precision floating-point values.
4975 Emits 32-bit values.
4977 Emits single precision floating-point values.
4979 Emits 16-bit values.
4980 Note - this is true even on targets where the @code{.word} directive would emit
4983 Emits long double precision floating-point values.
4986 If no suffix is used then @samp{.w} is assumed.
4988 The byte ordering is target dependent, as is the size and format of floating
4992 @section @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4993 @cindex @code{dcb} directive
4994 This directive emits @var{number} copies of @var{fill}, each of @var{size}
4995 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
4996 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
4997 @var{size} suffix, if present, must be one of:
5001 Emits single byte values.
5003 Emits double-precision floating point values.
5005 Emits 4-byte values.
5007 Emits single-precision floating point values.
5009 Emits 2-byte values.
5011 Emits long double-precision floating point values.
5014 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5016 The byte ordering is target dependent, as is the size and format of floating
5020 @section @code{.ds[@var{size}] @var{number} [,@var{fill}]}
5021 @cindex @code{ds} directive
5022 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5023 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5024 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5025 @var{size} suffix, if present, must be one of:
5029 Emits single byte values.
5031 Emits 8-byte values.
5033 Emits 4-byte values.
5035 Emits 12-byte values.
5037 Emits 4-byte values.
5039 Emits 2-byte values.
5041 Emits 12-byte values.
5044 Note - unlike the @code{.dcb} directive the @samp{.d}, @samp{.s} and @samp{.x}
5045 suffixes do not indicate that floating-point values are to be inserted.
5047 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5049 The byte ordering is target dependent.
5054 @section @code{.def @var{name}}
5056 @cindex @code{def} directive
5057 @cindex COFF symbols, debugging
5058 @cindex debugging COFF symbols
5059 Begin defining debugging information for a symbol @var{name}; the
5060 definition extends until the @code{.endef} directive is encountered.
5065 @section @code{.desc @var{symbol}, @var{abs-expression}}
5067 @cindex @code{desc} directive
5068 @cindex COFF symbol descriptor
5069 @cindex symbol descriptor, COFF
5070 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5071 to the low 16 bits of an absolute expression.
5074 The @samp{.desc} directive is not available when @command{@value{AS}} is
5075 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5076 object format. For the sake of compatibility, @command{@value{AS}} accepts
5077 it, but produces no output, when configured for COFF.
5083 @section @code{.dim}
5085 @cindex @code{dim} directive
5086 @cindex COFF auxiliary symbol information
5087 @cindex auxiliary symbol information, COFF
5088 This directive is generated by compilers to include auxiliary debugging
5089 information in the symbol table. It is only permitted inside
5090 @code{.def}/@code{.endef} pairs.
5094 @section @code{.double @var{flonums}}
5096 @cindex @code{double} directive
5097 @cindex floating point numbers (double)
5098 @code{.double} expects zero or more flonums, separated by commas. It
5099 assembles floating point numbers.
5101 The exact kind of floating point numbers emitted depends on how
5102 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5106 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5107 in @sc{ieee} format.
5112 @section @code{.eject}
5114 @cindex @code{eject} directive
5115 @cindex new page, in listings
5116 @cindex page, in listings
5117 @cindex listing control: new page
5118 Force a page break at this point, when generating assembly listings.
5121 @section @code{.else}
5123 @cindex @code{else} directive
5124 @code{.else} is part of the @command{@value{AS}} support for conditional
5125 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5126 of code to be assembled if the condition for the preceding @code{.if}
5130 @section @code{.elseif}
5132 @cindex @code{elseif} directive
5133 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5134 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5135 @code{.if} block that would otherwise fill the entire @code{.else} section.
5138 @section @code{.end}
5140 @cindex @code{end} directive
5141 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5142 process anything in the file past the @code{.end} directive.
5146 @section @code{.endef}
5148 @cindex @code{endef} directive
5149 This directive flags the end of a symbol definition begun with
5154 @section @code{.endfunc}
5155 @cindex @code{endfunc} directive
5156 @code{.endfunc} marks the end of a function specified with @code{.func}.
5159 @section @code{.endif}
5161 @cindex @code{endif} directive
5162 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5163 it marks the end of a block of code that is only assembled
5164 conditionally. @xref{If,,@code{.if}}.
5167 @section @code{.equ @var{symbol}, @var{expression}}
5169 @cindex @code{equ} directive
5170 @cindex assigning values to symbols
5171 @cindex symbols, assigning values to
5172 This directive sets the value of @var{symbol} to @var{expression}.
5173 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5176 The syntax for @code{equ} on the HPPA is
5177 @samp{@var{symbol} .equ @var{expression}}.
5181 The syntax for @code{equ} on the Z80 is
5182 @samp{@var{symbol} equ @var{expression}}.
5183 On the Z80 it is an error if @var{symbol} is already defined,
5184 but the symbol is not protected from later redefinition.
5185 Compare @ref{Equiv}.
5189 @section @code{.equiv @var{symbol}, @var{expression}}
5190 @cindex @code{equiv} directive
5191 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5192 the assembler will signal an error if @var{symbol} is already defined. Note a
5193 symbol which has been referenced but not actually defined is considered to be
5196 Except for the contents of the error message, this is roughly equivalent to
5203 plus it protects the symbol from later redefinition.
5206 @section @code{.eqv @var{symbol}, @var{expression}}
5207 @cindex @code{eqv} directive
5208 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5209 evaluate the expression or any part of it immediately. Instead each time
5210 the resulting symbol is used in an expression, a snapshot of its current
5214 @section @code{.err}
5215 @cindex @code{err} directive
5216 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5217 message and, unless the @option{-Z} option was used, it will not generate an
5218 object file. This can be used to signal an error in conditionally compiled code.
5221 @section @code{.error "@var{string}"}
5222 @cindex error directive
5224 Similarly to @code{.err}, this directive emits an error, but you can specify a
5225 string that will be emitted as the error message. If you don't specify the
5226 message, it defaults to @code{".error directive invoked in source file"}.
5227 @xref{Errors, ,Error and Warning Messages}.
5230 .error "This code has not been assembled and tested."
5234 @section @code{.exitm}
5235 Exit early from the current macro definition. @xref{Macro}.
5238 @section @code{.extern}
5240 @cindex @code{extern} directive
5241 @code{.extern} is accepted in the source program---for compatibility
5242 with other assemblers---but it is ignored. @command{@value{AS}} treats
5243 all undefined symbols as external.
5246 @section @code{.fail @var{expression}}
5248 @cindex @code{fail} directive
5249 Generates an error or a warning. If the value of the @var{expression} is 500
5250 or more, @command{@value{AS}} will print a warning message. If the value is less
5251 than 500, @command{@value{AS}} will print an error message. The message will
5252 include the value of @var{expression}. This can occasionally be useful inside
5253 complex nested macros or conditional assembly.
5256 @section @code{.file}
5257 @cindex @code{file} directive
5259 @ifclear no-file-dir
5260 There are two different versions of the @code{.file} directive. Targets
5261 that support DWARF2 line number information use the DWARF2 version of
5262 @code{.file}. Other targets use the default version.
5264 @subheading Default Version
5266 @cindex logical file name
5267 @cindex file name, logical
5268 This version of the @code{.file} directive tells @command{@value{AS}} that we
5269 are about to start a new logical file. The syntax is:
5275 @var{string} is the new file name. In general, the filename is
5276 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5277 to specify an empty file name, you must give the quotes--@code{""}. This
5278 statement may go away in future: it is only recognized to be compatible with
5279 old @command{@value{AS}} programs.
5281 @subheading DWARF2 Version
5284 When emitting DWARF2 line number information, @code{.file} assigns filenames
5285 to the @code{.debug_line} file name table. The syntax is:
5288 .file @var{fileno} @var{filename}
5291 The @var{fileno} operand should be a unique positive integer to use as the
5292 index of the entry in the table. The @var{filename} operand is a C string
5295 The detail of filename indices is exposed to the user because the filename
5296 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5297 information, and thus the user must know the exact indices that table
5301 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5303 @cindex @code{fill} directive
5304 @cindex writing patterns in memory
5305 @cindex patterns, writing in memory
5306 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5307 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5308 may be zero or more. @var{Size} may be zero or more, but if it is
5309 more than 8, then it is deemed to have the value 8, compatible with
5310 other people's assemblers. The contents of each @var{repeat} bytes
5311 is taken from an 8-byte number. The highest order 4 bytes are
5312 zero. The lowest order 4 bytes are @var{value} rendered in the
5313 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5314 Each @var{size} bytes in a repetition is taken from the lowest order
5315 @var{size} bytes of this number. Again, this bizarre behavior is
5316 compatible with other people's assemblers.
5318 @var{size} and @var{value} are optional.
5319 If the second comma and @var{value} are absent, @var{value} is
5320 assumed zero. If the first comma and following tokens are absent,
5321 @var{size} is assumed to be 1.
5324 @section @code{.float @var{flonums}}
5326 @cindex floating point numbers (single)
5327 @cindex @code{float} directive
5328 This directive assembles zero or more flonums, separated by commas. It
5329 has the same effect as @code{.single}.
5331 The exact kind of floating point numbers emitted depends on how
5332 @command{@value{AS}} is configured.
5333 @xref{Machine Dependencies}.
5337 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5338 in @sc{ieee} format.
5343 @section @code{.func @var{name}[,@var{label}]}
5344 @cindex @code{func} directive
5345 @code{.func} emits debugging information to denote function @var{name}, and
5346 is ignored unless the file is assembled with debugging enabled.
5347 Only @samp{--gstabs[+]} is currently supported.
5348 @var{label} is the entry point of the function and if omitted @var{name}
5349 prepended with the @samp{leading char} is used.
5350 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5351 All functions are currently defined to have @code{void} return type.
5352 The function must be terminated with @code{.endfunc}.
5355 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5357 @cindex @code{global} directive
5358 @cindex symbol, making visible to linker
5359 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5360 @var{symbol} in your partial program, its value is made available to
5361 other partial programs that are linked with it. Otherwise,
5362 @var{symbol} takes its attributes from a symbol of the same name
5363 from another file linked into the same program.
5365 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5366 compatibility with other assemblers.
5369 On the HPPA, @code{.global} is not always enough to make it accessible to other
5370 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5371 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5376 @section @code{.gnu_attribute @var{tag},@var{value}}
5377 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5380 @section @code{.hidden @var{names}}
5382 @cindex @code{hidden} directive
5384 This is one of the ELF visibility directives. The other two are
5385 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5386 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5388 This directive overrides the named symbols default visibility (which is set by
5389 their binding: local, global or weak). The directive sets the visibility to
5390 @code{hidden} which means that the symbols are not visible to other components.
5391 Such symbols are always considered to be @code{protected} as well.
5395 @section @code{.hword @var{expressions}}
5397 @cindex @code{hword} directive
5398 @cindex integers, 16-bit
5399 @cindex numbers, 16-bit
5400 @cindex sixteen bit integers
5401 This expects zero or more @var{expressions}, and emits
5402 a 16 bit number for each.
5405 This directive is a synonym for @samp{.short}; depending on the target
5406 architecture, it may also be a synonym for @samp{.word}.
5410 This directive is a synonym for @samp{.short}.
5413 This directive is a synonym for both @samp{.short} and @samp{.word}.
5418 @section @code{.ident}
5420 @cindex @code{ident} directive
5422 This directive is used by some assemblers to place tags in object files. The
5423 behavior of this directive varies depending on the target. When using the
5424 a.out object file format, @command{@value{AS}} simply accepts the directive for
5425 source-file compatibility with existing assemblers, but does not emit anything
5426 for it. When using COFF, comments are emitted to the @code{.comment} or
5427 @code{.rdata} section, depending on the target. When using ELF, comments are
5428 emitted to the @code{.comment} section.
5431 @section @code{.if @var{absolute expression}}
5433 @cindex conditional assembly
5434 @cindex @code{if} directive
5435 @code{.if} marks the beginning of a section of code which is only
5436 considered part of the source program being assembled if the argument
5437 (which must be an @var{absolute expression}) is non-zero. The end of
5438 the conditional section of code must be marked by @code{.endif}
5439 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5440 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5441 If you have several conditions to check, @code{.elseif} may be used to avoid
5442 nesting blocks if/else within each subsequent @code{.else} block.
5444 The following variants of @code{.if} are also supported:
5446 @cindex @code{ifdef} directive
5447 @item .ifdef @var{symbol}
5448 Assembles the following section of code if the specified @var{symbol}
5449 has been defined. Note a symbol which has been referenced but not yet defined
5450 is considered to be undefined.
5452 @cindex @code{ifb} directive
5453 @item .ifb @var{text}
5454 Assembles the following section of code if the operand is blank (empty).
5456 @cindex @code{ifc} directive
5457 @item .ifc @var{string1},@var{string2}
5458 Assembles the following section of code if the two strings are the same. The
5459 strings may be optionally quoted with single quotes. If they are not quoted,
5460 the first string stops at the first comma, and the second string stops at the
5461 end of the line. Strings which contain whitespace should be quoted. The
5462 string comparison is case sensitive.
5464 @cindex @code{ifeq} directive
5465 @item .ifeq @var{absolute expression}
5466 Assembles the following section of code if the argument is zero.
5468 @cindex @code{ifeqs} directive
5469 @item .ifeqs @var{string1},@var{string2}
5470 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5472 @cindex @code{ifge} directive
5473 @item .ifge @var{absolute expression}
5474 Assembles the following section of code if the argument is greater than or
5477 @cindex @code{ifgt} directive
5478 @item .ifgt @var{absolute expression}
5479 Assembles the following section of code if the argument is greater than zero.
5481 @cindex @code{ifle} directive
5482 @item .ifle @var{absolute expression}
5483 Assembles the following section of code if the argument is less than or equal
5486 @cindex @code{iflt} directive
5487 @item .iflt @var{absolute expression}
5488 Assembles the following section of code if the argument is less than zero.
5490 @cindex @code{ifnb} directive
5491 @item .ifnb @var{text}
5492 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5493 following section of code if the operand is non-blank (non-empty).
5495 @cindex @code{ifnc} directive
5496 @item .ifnc @var{string1},@var{string2}.
5497 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5498 following section of code if the two strings are not the same.
5500 @cindex @code{ifndef} directive
5501 @cindex @code{ifnotdef} directive
5502 @item .ifndef @var{symbol}
5503 @itemx .ifnotdef @var{symbol}
5504 Assembles the following section of code if the specified @var{symbol}
5505 has not been defined. Both spelling variants are equivalent. Note a symbol
5506 which has been referenced but not yet defined is considered to be undefined.
5508 @cindex @code{ifne} directive
5509 @item .ifne @var{absolute expression}
5510 Assembles the following section of code if the argument is not equal to zero
5511 (in other words, this is equivalent to @code{.if}).
5513 @cindex @code{ifnes} directive
5514 @item .ifnes @var{string1},@var{string2}
5515 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5516 following section of code if the two strings are not the same.
5520 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5522 @cindex @code{incbin} directive
5523 @cindex binary files, including
5524 The @code{incbin} directive includes @var{file} verbatim at the current
5525 location. You can control the search paths used with the @samp{-I} command-line
5526 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5529 The @var{skip} argument skips a number of bytes from the start of the
5530 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5531 read. Note that the data is not aligned in any way, so it is the user's
5532 responsibility to make sure that proper alignment is provided both before and
5533 after the @code{incbin} directive.
5536 @section @code{.include "@var{file}"}
5538 @cindex @code{include} directive
5539 @cindex supporting files, including
5540 @cindex files, including
5541 This directive provides a way to include supporting files at specified
5542 points in your source program. The code from @var{file} is assembled as
5543 if it followed the point of the @code{.include}; when the end of the
5544 included file is reached, assembly of the original file continues. You
5545 can control the search paths used with the @samp{-I} command-line option
5546 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5550 @section @code{.int @var{expressions}}
5552 @cindex @code{int} directive
5553 @cindex integers, 32-bit
5554 Expect zero or more @var{expressions}, of any section, separated by commas.
5555 For each expression, emit a number that, at run time, is the value of that
5556 expression. The byte order and bit size of the number depends on what kind
5557 of target the assembly is for.
5561 On most forms of the H8/300, @code{.int} emits 16-bit
5562 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5569 @section @code{.internal @var{names}}
5571 @cindex @code{internal} directive
5573 This is one of the ELF visibility directives. The other two are
5574 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5575 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5577 This directive overrides the named symbols default visibility (which is set by
5578 their binding: local, global or weak). The directive sets the visibility to
5579 @code{internal} which means that the symbols are considered to be @code{hidden}
5580 (i.e., not visible to other components), and that some extra, processor specific
5581 processing must also be performed upon the symbols as well.
5585 @section @code{.irp @var{symbol},@var{values}}@dots{}
5587 @cindex @code{irp} directive
5588 Evaluate a sequence of statements assigning different values to @var{symbol}.
5589 The sequence of statements starts at the @code{.irp} directive, and is
5590 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5591 set to @var{value}, and the sequence of statements is assembled. If no
5592 @var{value} is listed, the sequence of statements is assembled once, with
5593 @var{symbol} set to the null string. To refer to @var{symbol} within the
5594 sequence of statements, use @var{\symbol}.
5596 For example, assembling
5604 is equivalent to assembling
5612 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5615 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5617 @cindex @code{irpc} directive
5618 Evaluate a sequence of statements assigning different values to @var{symbol}.
5619 The sequence of statements starts at the @code{.irpc} directive, and is
5620 terminated by an @code{.endr} directive. For each character in @var{value},
5621 @var{symbol} is set to the character, and the sequence of statements is
5622 assembled. If no @var{value} is listed, the sequence of statements is
5623 assembled once, with @var{symbol} set to the null string. To refer to
5624 @var{symbol} within the sequence of statements, use @var{\symbol}.
5626 For example, assembling
5634 is equivalent to assembling
5642 For some caveats with the spelling of @var{symbol}, see also the discussion
5646 @section @code{.lcomm @var{symbol} , @var{length}}
5648 @cindex @code{lcomm} directive
5649 @cindex local common symbols
5650 @cindex symbols, local common
5651 Reserve @var{length} (an absolute expression) bytes for a local common
5652 denoted by @var{symbol}. The section and value of @var{symbol} are
5653 those of the new local common. The addresses are allocated in the bss
5654 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5655 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5656 not visible to @code{@value{LD}}.
5659 Some targets permit a third argument to be used with @code{.lcomm}. This
5660 argument specifies the desired alignment of the symbol in the bss section.
5664 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5665 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5669 @section @code{.lflags}
5671 @cindex @code{lflags} directive (ignored)
5672 @command{@value{AS}} accepts this directive, for compatibility with other
5673 assemblers, but ignores it.
5675 @ifclear no-line-dir
5677 @section @code{.line @var{line-number}}
5679 @cindex @code{line} directive
5680 @cindex logical line number
5682 Change the logical line number. @var{line-number} must be an absolute
5683 expression. The next line has that logical line number. Therefore any other
5684 statements on the current line (after a statement separator character) are
5685 reported as on logical line number @var{line-number} @minus{} 1. One day
5686 @command{@value{AS}} will no longer support this directive: it is recognized only
5687 for compatibility with existing assembler programs.
5690 Even though this is a directive associated with the @code{a.out} or
5691 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5692 when producing COFF output, and treats @samp{.line} as though it
5693 were the COFF @samp{.ln} @emph{if} it is found outside a
5694 @code{.def}/@code{.endef} pair.
5696 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5697 used by compilers to generate auxiliary symbol information for
5702 @section @code{.linkonce [@var{type}]}
5704 @cindex @code{linkonce} directive
5705 @cindex common sections
5706 Mark the current section so that the linker only includes a single copy of it.
5707 This may be used to include the same section in several different object files,
5708 but ensure that the linker will only include it once in the final output file.
5709 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5710 Duplicate sections are detected based on the section name, so it should be
5713 This directive is only supported by a few object file formats; as of this
5714 writing, the only object file format which supports it is the Portable
5715 Executable format used on Windows NT.
5717 The @var{type} argument is optional. If specified, it must be one of the
5718 following strings. For example:
5722 Not all types may be supported on all object file formats.
5726 Silently discard duplicate sections. This is the default.
5729 Warn if there are duplicate sections, but still keep only one copy.
5732 Warn if any of the duplicates have different sizes.
5735 Warn if any of the duplicates do not have exactly the same contents.
5739 @section @code{.list}
5741 @cindex @code{list} directive
5742 @cindex listing control, turning on
5743 Control (in conjunction with the @code{.nolist} directive) whether or
5744 not assembly listings are generated. These two directives maintain an
5745 internal counter (which is zero initially). @code{.list} increments the
5746 counter, and @code{.nolist} decrements it. Assembly listings are
5747 generated whenever the counter is greater than zero.
5749 By default, listings are disabled. When you enable them (with the
5750 @samp{-a} command-line option; @pxref{Invoking,,Command-Line Options}),
5751 the initial value of the listing counter is one.
5754 @section @code{.ln @var{line-number}}
5756 @cindex @code{ln} directive
5757 @ifclear no-line-dir
5758 @samp{.ln} is a synonym for @samp{.line}.
5761 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5762 must be an absolute expression. The next line has that logical
5763 line number, so any other statements on the current line (after a
5764 statement separator character @code{;}) are reported as on logical
5765 line number @var{line-number} @minus{} 1.
5769 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5770 @cindex @code{loc} directive
5771 When emitting DWARF2 line number information,
5772 the @code{.loc} directive will add a row to the @code{.debug_line} line
5773 number matrix corresponding to the immediately following assembly
5774 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5775 arguments will be applied to the @code{.debug_line} state machine before
5778 The @var{options} are a sequence of the following tokens in any order:
5782 This option will set the @code{basic_block} register in the
5783 @code{.debug_line} state machine to @code{true}.
5786 This option will set the @code{prologue_end} register in the
5787 @code{.debug_line} state machine to @code{true}.
5789 @item epilogue_begin
5790 This option will set the @code{epilogue_begin} register in the
5791 @code{.debug_line} state machine to @code{true}.
5793 @item is_stmt @var{value}
5794 This option will set the @code{is_stmt} register in the
5795 @code{.debug_line} state machine to @code{value}, which must be
5798 @item isa @var{value}
5799 This directive will set the @code{isa} register in the @code{.debug_line}
5800 state machine to @var{value}, which must be an unsigned integer.
5802 @item discriminator @var{value}
5803 This directive will set the @code{discriminator} register in the @code{.debug_line}
5804 state machine to @var{value}, which must be an unsigned integer.
5806 @item view @var{value}
5807 This option causes a row to be added to @code{.debug_line} in reference to the
5808 current address (which might not be the same as that of the following assembly
5809 instruction), and to associate @var{value} with the @code{view} register in the
5810 @code{.debug_line} state machine. If @var{value} is a label, both the
5811 @code{view} register and the label are set to the number of prior @code{.loc}
5812 directives at the same program location. If @var{value} is the literal
5813 @code{0}, the @code{view} register is set to zero, and the assembler asserts
5814 that there aren't any prior @code{.loc} directives at the same program
5815 location. If @var{value} is the literal @code{-0}, the assembler arrange for
5816 the @code{view} register to be reset in this row, even if there are prior
5817 @code{.loc} directives at the same program location.
5821 @node Loc_mark_labels
5822 @section @code{.loc_mark_labels @var{enable}}
5823 @cindex @code{loc_mark_labels} directive
5824 When emitting DWARF2 line number information,
5825 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5826 to the @code{.debug_line} line number matrix with the @code{basic_block}
5827 register in the state machine set whenever a code label is seen.
5828 The @var{enable} argument should be either 1 or 0, to enable or disable
5829 this function respectively.
5833 @section @code{.local @var{names}}
5835 @cindex @code{local} directive
5836 This directive, which is available for ELF targets, marks each symbol in
5837 the comma-separated list of @code{names} as a local symbol so that it
5838 will not be externally visible. If the symbols do not already exist,
5839 they will be created.
5841 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5842 accept an alignment argument, which is the case for most ELF targets,
5843 the @code{.local} directive can be used in combination with @code{.comm}
5844 (@pxref{Comm}) to define aligned local common data.
5848 @section @code{.long @var{expressions}}
5850 @cindex @code{long} directive
5851 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5854 @c no one seems to know what this is for or whether this description is
5855 @c what it really ought to do
5857 @section @code{.lsym @var{symbol}, @var{expression}}
5859 @cindex @code{lsym} directive
5860 @cindex symbol, not referenced in assembly
5861 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5862 the hash table, ensuring it cannot be referenced by name during the
5863 rest of the assembly. This sets the attributes of the symbol to be
5864 the same as the expression value:
5866 @var{other} = @var{descriptor} = 0
5867 @var{type} = @r{(section of @var{expression})}
5868 @var{value} = @var{expression}
5871 The new symbol is not flagged as external.
5875 @section @code{.macro}
5878 The commands @code{.macro} and @code{.endm} allow you to define macros that
5879 generate assembly output. For example, this definition specifies a macro
5880 @code{sum} that puts a sequence of numbers into memory:
5883 .macro sum from=0, to=5
5892 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5904 @item .macro @var{macname}
5905 @itemx .macro @var{macname} @var{macargs} @dots{}
5906 @cindex @code{macro} directive
5907 Begin the definition of a macro called @var{macname}. If your macro
5908 definition requires arguments, specify their names after the macro name,
5909 separated by commas or spaces. You can qualify the macro argument to
5910 indicate whether all invocations must specify a non-blank value (through
5911 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5912 (through @samp{:@code{vararg}}). You can supply a default value for any
5913 macro argument by following the name with @samp{=@var{deflt}}. You
5914 cannot define two macros with the same @var{macname} unless it has been
5915 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5916 definitions. For example, these are all valid @code{.macro} statements:
5920 Begin the definition of a macro called @code{comm}, which takes no
5923 @item .macro plus1 p, p1
5924 @itemx .macro plus1 p p1
5925 Either statement begins the definition of a macro called @code{plus1},
5926 which takes two arguments; within the macro definition, write
5927 @samp{\p} or @samp{\p1} to evaluate the arguments.
5929 @item .macro reserve_str p1=0 p2
5930 Begin the definition of a macro called @code{reserve_str}, with two
5931 arguments. The first argument has a default value, but not the second.
5932 After the definition is complete, you can call the macro either as
5933 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5934 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5935 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5936 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5938 @item .macro m p1:req, p2=0, p3:vararg
5939 Begin the definition of a macro called @code{m}, with at least three
5940 arguments. The first argument must always have a value specified, but
5941 not the second, which instead has a default value. The third formal
5942 will get assigned all remaining arguments specified at invocation time.
5944 When you call a macro, you can specify the argument values either by
5945 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5946 @samp{sum to=17, from=9}.
5950 Note that since each of the @var{macargs} can be an identifier exactly
5951 as any other one permitted by the target architecture, there may be
5952 occasional problems if the target hand-crafts special meanings to certain
5953 characters when they occur in a special position. For example, if the colon
5954 (@code{:}) is generally permitted to be part of a symbol name, but the
5955 architecture specific code special-cases it when occurring as the final
5956 character of a symbol (to denote a label), then the macro parameter
5957 replacement code will have no way of knowing that and consider the whole
5958 construct (including the colon) an identifier, and check only this
5959 identifier for being the subject to parameter substitution. So for example
5960 this macro definition:
5968 might not work as expected. Invoking @samp{label foo} might not create a label
5969 called @samp{foo} but instead just insert the text @samp{\l:} into the
5970 assembler source, probably generating an error about an unrecognised
5973 Similarly problems might occur with the period character (@samp{.})
5974 which is often allowed inside opcode names (and hence identifier names). So
5975 for example constructing a macro to build an opcode from a base name and a
5976 length specifier like this:
5979 .macro opcode base length
5984 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5985 instruction but instead generate some kind of error as the assembler tries to
5986 interpret the text @samp{\base.\length}.
5988 There are several possible ways around this problem:
5991 @item Insert white space
5992 If it is possible to use white space characters then this is the simplest
6001 @item Use @samp{\()}
6002 The string @samp{\()} can be used to separate the end of a macro argument from
6003 the following text. eg:
6006 .macro opcode base length
6011 @item Use the alternate macro syntax mode
6012 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
6013 used as a separator. eg:
6023 Note: this problem of correctly identifying string parameters to pseudo ops
6024 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
6025 and @code{.irpc} (@pxref{Irpc}) as well.
6028 @cindex @code{endm} directive
6029 Mark the end of a macro definition.
6032 @cindex @code{exitm} directive
6033 Exit early from the current macro definition.
6035 @cindex number of macros executed
6036 @cindex macros, count executed
6038 @command{@value{AS}} maintains a counter of how many macros it has
6039 executed in this pseudo-variable; you can copy that number to your
6040 output with @samp{\@@}, but @emph{only within a macro definition}.
6042 @item LOCAL @var{name} [ , @dots{} ]
6043 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
6044 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
6045 @xref{Altmacro,,@code{.altmacro}}.
6049 @section @code{.mri @var{val}}
6051 @cindex @code{mri} directive
6052 @cindex MRI mode, temporarily
6053 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
6054 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
6055 affects code assembled until the next @code{.mri} directive, or until the end
6056 of the file. @xref{M, MRI mode, MRI mode}.
6059 @section @code{.noaltmacro}
6060 Disable alternate macro mode. @xref{Altmacro}.
6063 @section @code{.nolist}
6065 @cindex @code{nolist} directive
6066 @cindex listing control, turning off
6067 Control (in conjunction with the @code{.list} directive) whether or
6068 not assembly listings are generated. These two directives maintain an
6069 internal counter (which is zero initially). @code{.list} increments the
6070 counter, and @code{.nolist} decrements it. Assembly listings are
6071 generated whenever the counter is greater than zero.
6074 @section @code{.nops @var{size}[, @var{control}]}
6076 @cindex @code{nops} directive
6077 @cindex filling memory with no-op instructions
6078 This directive emits @var{size} bytes filled with no-op instructions.
6079 @var{size} is absolute expression, which must be a positve value.
6080 @var{control} controls how no-op instructions should be generated. If
6081 the comma and @var{control} are omitted, @var{control} is assumed to be
6084 Note: For Intel 80386 and AMD x86-64 targets, @var{control} specifies
6085 the size limit of a no-op instruction. The valid values of @var{control}
6086 are between 0 and 4 in 16-bit mode, between 0 and 7 when tuning for
6087 older processors in 32-bit mode, between 0 and 11 in 64-bit mode or when
6088 tuning for newer processors in 32-bit mode. When 0 is used, the no-op
6089 instruction size limit is set to the maximum supported size.
6092 @section @code{.octa @var{bignums}}
6094 @c FIXME: double size emitted for "octa" on some? Or warn?
6095 @cindex @code{octa} directive
6096 @cindex integer, 16-byte
6097 @cindex sixteen byte integer
6098 This directive expects zero or more bignums, separated by commas. For each
6099 bignum, it emits a 16-byte integer.
6101 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6102 hence @emph{octa}-word for 16 bytes.
6105 @section @code{.offset @var{loc}}
6107 @cindex @code{offset} directive
6108 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6109 be an absolute expression. This directive may be useful for defining
6110 symbols with absolute values. Do not confuse it with the @code{.org}
6114 @section @code{.org @var{new-lc} , @var{fill}}
6116 @cindex @code{org} directive
6117 @cindex location counter, advancing
6118 @cindex advancing location counter
6119 @cindex current address, advancing
6120 Advance the location counter of the current section to
6121 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6122 expression with the same section as the current subsection. That is,
6123 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6124 wrong section, the @code{.org} directive is ignored. To be compatible
6125 with former assemblers, if the section of @var{new-lc} is absolute,
6126 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6127 is the same as the current subsection.
6129 @code{.org} may only increase the location counter, or leave it
6130 unchanged; you cannot use @code{.org} to move the location counter
6133 @c double negative used below "not undefined" because this is a specific
6134 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6135 @c section. doc@cygnus.com 18feb91
6136 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6137 may not be undefined. If you really detest this restriction we eagerly await
6138 a chance to share your improved assembler.
6140 Beware that the origin is relative to the start of the section, not
6141 to the start of the subsection. This is compatible with other
6142 people's assemblers.
6144 When the location counter (of the current subsection) is advanced, the
6145 intervening bytes are filled with @var{fill} which should be an
6146 absolute expression. If the comma and @var{fill} are omitted,
6147 @var{fill} defaults to zero.
6150 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
6152 @cindex padding the location counter given a power of two
6153 @cindex @code{p2align} directive
6154 Pad the location counter (in the current subsection) to a particular
6155 storage boundary. The first expression (which must be absolute) is the
6156 number of low-order zero bits the location counter must have after
6157 advancement. For example @samp{.p2align 3} advances the location
6158 counter until it a multiple of 8. If the location counter is already a
6159 multiple of 8, no change is needed.
6161 The second expression (also absolute) gives the fill value to be stored in the
6162 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6163 padding bytes are normally zero. However, on most systems, if the section is
6164 marked as containing code and the fill value is omitted, the space is filled
6165 with no-op instructions.
6167 The third expression is also absolute, and is also optional. If it is present,
6168 it is the maximum number of bytes that should be skipped by this alignment
6169 directive. If doing the alignment would require skipping more bytes than the
6170 specified maximum, then the alignment is not done at all. You can omit the
6171 fill value (the second argument) entirely by simply using two commas after the
6172 required alignment; this can be useful if you want the alignment to be filled
6173 with no-op instructions when appropriate.
6175 @cindex @code{p2alignw} directive
6176 @cindex @code{p2alignl} directive
6177 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6178 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6179 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6180 fill pattern as a four byte longword value. For example, @code{.p2alignw
6181 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6182 filled in with the value 0x368d (the exact placement of the bytes depends upon
6183 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6188 @section @code{.popsection}
6190 @cindex @code{popsection} directive
6191 @cindex Section Stack
6192 This is one of the ELF section stack manipulation directives. The others are
6193 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6194 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6197 This directive replaces the current section (and subsection) with the top
6198 section (and subsection) on the section stack. This section is popped off the
6204 @section @code{.previous}
6206 @cindex @code{previous} directive
6207 @cindex Section Stack
6208 This is one of the ELF section stack manipulation directives. The others are
6209 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6210 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6211 (@pxref{PopSection}).
6213 This directive swaps the current section (and subsection) with most recently
6214 referenced section/subsection pair prior to this one. Multiple
6215 @code{.previous} directives in a row will flip between two sections (and their
6216 subsections). For example:
6228 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6234 # Now in section A subsection 1
6238 # Now in section B subsection 0
6241 # Now in section B subsection 1
6244 # Now in section B subsection 0
6248 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6249 section B and 0x9abc into subsection 1 of section B.
6251 In terms of the section stack, this directive swaps the current section with
6252 the top section on the section stack.
6256 @section @code{.print @var{string}}
6258 @cindex @code{print} directive
6259 @command{@value{AS}} will print @var{string} on the standard output during
6260 assembly. You must put @var{string} in double quotes.
6264 @section @code{.protected @var{names}}
6266 @cindex @code{protected} directive
6268 This is one of the ELF visibility directives. The other two are
6269 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6271 This directive overrides the named symbols default visibility (which is set by
6272 their binding: local, global or weak). The directive sets the visibility to
6273 @code{protected} which means that any references to the symbols from within the
6274 components that defines them must be resolved to the definition in that
6275 component, even if a definition in another component would normally preempt
6280 @section @code{.psize @var{lines} , @var{columns}}
6282 @cindex @code{psize} directive
6283 @cindex listing control: paper size
6284 @cindex paper size, for listings
6285 Use this directive to declare the number of lines---and, optionally, the
6286 number of columns---to use for each page, when generating listings.
6288 If you do not use @code{.psize}, listings use a default line-count
6289 of 60. You may omit the comma and @var{columns} specification; the
6290 default width is 200 columns.
6292 @command{@value{AS}} generates formfeeds whenever the specified number of
6293 lines is exceeded (or whenever you explicitly request one, using
6296 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6297 those explicitly specified with @code{.eject}.
6300 @section @code{.purgem @var{name}}
6302 @cindex @code{purgem} directive
6303 Undefine the macro @var{name}, so that later uses of the string will not be
6304 expanded. @xref{Macro}.
6308 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6310 @cindex @code{pushsection} directive
6311 @cindex Section Stack
6312 This is one of the ELF section stack manipulation directives. The others are
6313 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6314 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6317 This directive pushes the current section (and subsection) onto the
6318 top of the section stack, and then replaces the current section and
6319 subsection with @code{name} and @code{subsection}. The optional
6320 @code{flags}, @code{type} and @code{arguments} are treated the same
6321 as in the @code{.section} (@pxref{Section}) directive.
6325 @section @code{.quad @var{bignums}}
6327 @cindex @code{quad} directive
6328 @code{.quad} expects zero or more bignums, separated by commas. For
6329 each bignum, it emits
6331 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6332 warning message; and just takes the lowest order 8 bytes of the bignum.
6333 @cindex eight-byte integer
6334 @cindex integer, 8-byte
6336 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6337 hence @emph{quad}-word for 8 bytes.
6340 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6341 warning message; and just takes the lowest order 16 bytes of the bignum.
6342 @cindex sixteen-byte integer
6343 @cindex integer, 16-byte
6347 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6349 @cindex @code{reloc} directive
6350 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6351 @var{expression}. If @var{offset} is a number, the relocation is generated in
6352 the current section. If @var{offset} is an expression that resolves to a
6353 symbol plus offset, the relocation is generated in the given symbol's section.
6354 @var{expression}, if present, must resolve to a symbol plus addend or to an
6355 absolute value, but note that not all targets support an addend. e.g. ELF REL
6356 targets such as i386 store an addend in the section contents rather than in the
6357 relocation. This low level interface does not support addends stored in the
6361 @section @code{.rept @var{count}}
6363 @cindex @code{rept} directive
6364 Repeat the sequence of lines between the @code{.rept} directive and the next
6365 @code{.endr} directive @var{count} times.
6367 For example, assembling
6375 is equivalent to assembling
6383 A count of zero is allowed, but nothing is generated. Negative counts are not
6384 allowed and if encountered will be treated as if they were zero.
6387 @section @code{.sbttl "@var{subheading}"}
6389 @cindex @code{sbttl} directive
6390 @cindex subtitles for listings
6391 @cindex listing control: subtitle
6392 Use @var{subheading} as the title (third line, immediately after the
6393 title line) when generating assembly listings.
6395 This directive affects subsequent pages, as well as the current page if
6396 it appears within ten lines of the top of a page.
6400 @section @code{.scl @var{class}}
6402 @cindex @code{scl} directive
6403 @cindex symbol storage class (COFF)
6404 @cindex COFF symbol storage class
6405 Set the storage-class value for a symbol. This directive may only be
6406 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6407 whether a symbol is static or external, or it may record further
6408 symbolic debugging information.
6413 @section @code{.section @var{name}}
6415 @cindex named section
6416 Use the @code{.section} directive to assemble the following code into a section
6419 This directive is only supported for targets that actually support arbitrarily
6420 named sections; on @code{a.out} targets, for example, it is not accepted, even
6421 with a standard @code{a.out} section name.
6425 @c only print the extra heading if both COFF and ELF are set
6426 @subheading COFF Version
6429 @cindex @code{section} directive (COFF version)
6430 For COFF targets, the @code{.section} directive is used in one of the following
6434 .section @var{name}[, "@var{flags}"]
6435 .section @var{name}[, @var{subsection}]
6438 If the optional argument is quoted, it is taken as flags to use for the
6439 section. Each flag is a single character. The following flags are recognized:
6443 bss section (uninitialized data)
6445 section is not loaded
6451 exclude section from linking
6457 shared section (meaningful for PE targets)
6459 ignored. (For compatibility with the ELF version)
6461 section is not readable (meaningful for PE targets)
6463 single-digit power-of-two section alignment (GNU extension)
6466 If no flags are specified, the default flags depend upon the section name. If
6467 the section name is not recognized, the default will be for the section to be
6468 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6469 from the section, rather than adding them, so if they are used on their own it
6470 will be as if no flags had been specified at all.
6472 If the optional argument to the @code{.section} directive is not quoted, it is
6473 taken as a subsection number (@pxref{Sub-Sections}).
6478 @c only print the extra heading if both COFF and ELF are set
6479 @subheading ELF Version
6482 @cindex Section Stack
6483 This is one of the ELF section stack manipulation directives. The others are
6484 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6485 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6486 @code{.previous} (@pxref{Previous}).
6488 @cindex @code{section} directive (ELF version)
6489 For ELF targets, the @code{.section} directive is used like this:
6492 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6495 @anchor{Section Name Substitutions}
6496 @kindex --sectname-subst
6497 @cindex section name substitution
6498 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6499 argument may contain a substitution sequence. Only @code{%S} is supported
6500 at the moment, and substitutes the current section name. For example:
6503 .macro exception_code
6504 .section %S.exception
6505 [exception code here]
6520 The two @code{exception_code} invocations above would create the
6521 @code{.text.exception} and @code{.init.exception} sections respectively.
6522 This is useful e.g. to discriminate between ancillary sections that are
6523 tied to setup code to be discarded after use from ancillary sections that
6524 need to stay resident without having to define multiple @code{exception_code}
6525 macros just for that purpose.
6527 The optional @var{flags} argument is a quoted string which may contain any
6528 combination of the following characters:
6532 section is allocatable
6534 section is a GNU_MBIND section
6536 section is excluded from executable and shared library.
6540 section is executable
6542 section is mergeable
6544 section contains zero terminated strings
6546 section is a member of a section group
6548 section is used for thread-local-storage
6550 section is a member of the previously-current section's group, if any
6551 @item @code{<number>}
6552 a numeric value indicating the bits to be set in the ELF section header's flags
6553 field. Note - if one or more of the alphabetic characters described above is
6554 also included in the flags field, their bit values will be ORed into the
6556 @item @code{<target specific>}
6557 some targets extend this list with their own flag characters
6560 Note - once a section's flags have been set they cannot be changed. There are
6561 a few exceptions to this rule however. Processor and application specific
6562 flags can be added to an already defined section. The @code{.interp},
6563 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6564 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6565 section may have the executable (@code{x}) flag added.
6567 The optional @var{type} argument may contain one of the following constants:
6571 section contains data
6573 section does not contain data (i.e., section only occupies space)
6575 section contains data which is used by things other than the program
6577 section contains an array of pointers to init functions
6579 section contains an array of pointers to finish functions
6580 @item @@preinit_array
6581 section contains an array of pointers to pre-init functions
6582 @item @@@code{<number>}
6583 a numeric value to be set as the ELF section header's type field.
6584 @item @@@code{<target specific>}
6585 some targets extend this list with their own types
6588 Many targets only support the first three section types. The type may be
6589 enclosed in double quotes if necessary.
6591 Note on targets where the @code{@@} character is the start of a comment (eg
6592 ARM) then another character is used instead. For example the ARM port uses the
6595 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6596 special and have fixed types. Any attempt to declare them with a different
6597 type will generate an error from the assembler.
6599 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6600 be specified as well as an extra argument---@var{entsize}---like this:
6603 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6606 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6607 constants, each @var{entsize} octets long. Sections with both @code{M} and
6608 @code{S} must contain zero terminated strings where each character is
6609 @var{entsize} bytes long. The linker may remove duplicates within sections with
6610 the same name, same entity size and same flags. @var{entsize} must be an
6611 absolute expression. For sections with both @code{M} and @code{S}, a string
6612 which is a suffix of a larger string is considered a duplicate. Thus
6613 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6614 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6616 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6617 be present along with an additional field like this:
6620 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6623 The @var{GroupName} field specifies the name of the section group to which this
6624 particular section belongs. The optional linkage field can contain:
6628 indicates that only one copy of this section should be retained
6633 Note: if both the @var{M} and @var{G} flags are present then the fields for
6634 the Merge flag should come first, like this:
6637 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6640 If @var{flags} contains the @code{?} symbol then it may not also contain the
6641 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6642 present. Instead, @code{?} says to consider the section that's current before
6643 this directive. If that section used @code{G}, then the new section will use
6644 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6645 If not, then the @code{?} symbol has no effect.
6647 If no flags are specified, the default flags depend upon the section name. If
6648 the section name is not recognized, the default will be for the section to have
6649 none of the above flags: it will not be allocated in memory, nor writable, nor
6650 executable. The section will contain data.
6652 For ELF targets, the assembler supports another type of @code{.section}
6653 directive for compatibility with the Solaris assembler:
6656 .section "@var{name}"[, @var{flags}...]
6659 Note that the section name is quoted. There may be a sequence of comma
6664 section is allocatable
6668 section is executable
6670 section is excluded from executable and shared library.
6672 section is used for thread local storage
6675 This directive replaces the current section and subsection. See the
6676 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6677 some examples of how this directive and the other section stack directives
6683 @section @code{.set @var{symbol}, @var{expression}}
6685 @cindex @code{set} directive
6686 @cindex symbol value, setting
6687 Set the value of @var{symbol} to @var{expression}. This
6688 changes @var{symbol}'s value and type to conform to
6689 @var{expression}. If @var{symbol} was flagged as external, it remains
6690 flagged (@pxref{Symbol Attributes}).
6692 You may @code{.set} a symbol many times in the same assembly provided that the
6693 values given to the symbol are constants. Values that are based on expressions
6694 involving other symbols are allowed, but some targets may restrict this to only
6695 being done once per assembly. This is because those targets do not set the
6696 addresses of symbols at assembly time, but rather delay the assignment until a
6697 final link is performed. This allows the linker a chance to change the code in
6698 the files, changing the location of, and the relative distance between, various
6701 If you @code{.set} a global symbol, the value stored in the object
6702 file is the last value stored into it.
6705 On Z80 @code{set} is a real instruction, use
6706 @samp{@var{symbol} defl @var{expression}} instead.
6710 @section @code{.short @var{expressions}}
6712 @cindex @code{short} directive
6714 @code{.short} is normally the same as @samp{.word}.
6715 @xref{Word,,@code{.word}}.
6717 In some configurations, however, @code{.short} and @code{.word} generate
6718 numbers of different lengths. @xref{Machine Dependencies}.
6722 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6725 This expects zero or more @var{expressions}, and emits
6726 a 16 bit number for each.
6731 @section @code{.single @var{flonums}}
6733 @cindex @code{single} directive
6734 @cindex floating point numbers (single)
6735 This directive assembles zero or more flonums, separated by commas. It
6736 has the same effect as @code{.float}.
6738 The exact kind of floating point numbers emitted depends on how
6739 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6743 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6744 numbers in @sc{ieee} format.
6750 @section @code{.size}
6752 This directive is used to set the size associated with a symbol.
6756 @c only print the extra heading if both COFF and ELF are set
6757 @subheading COFF Version
6760 @cindex @code{size} directive (COFF version)
6761 For COFF targets, the @code{.size} directive is only permitted inside
6762 @code{.def}/@code{.endef} pairs. It is used like this:
6765 .size @var{expression}
6772 @c only print the extra heading if both COFF and ELF are set
6773 @subheading ELF Version
6776 @cindex @code{size} directive (ELF version)
6777 For ELF targets, the @code{.size} directive is used like this:
6780 .size @var{name} , @var{expression}
6783 This directive sets the size associated with a symbol @var{name}.
6784 The size in bytes is computed from @var{expression} which can make use of label
6785 arithmetic. This directive is typically used to set the size of function
6790 @ifclear no-space-dir
6792 @section @code{.skip @var{size} [,@var{fill}]}
6794 @cindex @code{skip} directive
6795 @cindex filling memory
6796 This directive emits @var{size} bytes, each of value @var{fill}. Both
6797 @var{size} and @var{fill} are absolute expressions. If the comma and
6798 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6803 @section @code{.sleb128 @var{expressions}}
6805 @cindex @code{sleb128} directive
6806 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6807 compact, variable length representation of numbers used by the DWARF
6808 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6810 @ifclear no-space-dir
6812 @section @code{.space @var{size} [,@var{fill}]}
6814 @cindex @code{space} directive
6815 @cindex filling memory
6816 This directive emits @var{size} bytes, each of value @var{fill}. Both
6817 @var{size} and @var{fill} are absolute expressions. If the comma
6818 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6823 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6824 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6825 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6826 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6834 @section @code{.stabd, .stabn, .stabs}
6836 @cindex symbolic debuggers, information for
6837 @cindex @code{stab@var{x}} directives
6838 There are three directives that begin @samp{.stab}.
6839 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6840 The symbols are not entered in the @command{@value{AS}} hash table: they
6841 cannot be referenced elsewhere in the source file.
6842 Up to five fields are required:
6846 This is the symbol's name. It may contain any character except
6847 @samp{\000}, so is more general than ordinary symbol names. Some
6848 debuggers used to code arbitrarily complex structures into symbol names
6852 An absolute expression. The symbol's type is set to the low 8 bits of
6853 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6854 and debuggers choke on silly bit patterns.
6857 An absolute expression. The symbol's ``other'' attribute is set to the
6858 low 8 bits of this expression.
6861 An absolute expression. The symbol's descriptor is set to the low 16
6862 bits of this expression.
6865 An absolute expression which becomes the symbol's value.
6868 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6869 or @code{.stabs} statement, the symbol has probably already been created;
6870 you get a half-formed symbol in your object file. This is
6871 compatible with earlier assemblers!
6874 @cindex @code{stabd} directive
6875 @item .stabd @var{type} , @var{other} , @var{desc}
6877 The ``name'' of the symbol generated is not even an empty string.
6878 It is a null pointer, for compatibility. Older assemblers used a
6879 null pointer so they didn't waste space in object files with empty
6882 The symbol's value is set to the location counter,
6883 relocatably. When your program is linked, the value of this symbol
6884 is the address of the location counter when the @code{.stabd} was
6887 @cindex @code{stabn} directive
6888 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6889 The name of the symbol is set to the empty string @code{""}.
6891 @cindex @code{stabs} directive
6892 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6893 All five fields are specified.
6899 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6900 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6902 @cindex string, copying to object file
6903 @cindex string8, copying to object file
6904 @cindex string16, copying to object file
6905 @cindex string32, copying to object file
6906 @cindex string64, copying to object file
6907 @cindex @code{string} directive
6908 @cindex @code{string8} directive
6909 @cindex @code{string16} directive
6910 @cindex @code{string32} directive
6911 @cindex @code{string64} directive
6913 Copy the characters in @var{str} to the object file. You may specify more than
6914 one string to copy, separated by commas. Unless otherwise specified for a
6915 particular machine, the assembler marks the end of each string with a 0 byte.
6916 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6918 The variants @code{string16}, @code{string32} and @code{string64} differ from
6919 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6920 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6921 are stored in target endianness byte order.
6927 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6928 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6933 @section @code{.struct @var{expression}}
6935 @cindex @code{struct} directive
6936 Switch to the absolute section, and set the section offset to @var{expression},
6937 which must be an absolute expression. You might use this as follows:
6946 This would define the symbol @code{field1} to have the value 0, the symbol
6947 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6948 value 8. Assembly would be left in the absolute section, and you would need to
6949 use a @code{.section} directive of some sort to change to some other section
6950 before further assembly.
6954 @section @code{.subsection @var{name}}
6956 @cindex @code{subsection} directive
6957 @cindex Section Stack
6958 This is one of the ELF section stack manipulation directives. The others are
6959 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6960 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6963 This directive replaces the current subsection with @code{name}. The current
6964 section is not changed. The replaced subsection is put onto the section stack
6965 in place of the then current top of stack subsection.
6970 @section @code{.symver}
6971 @cindex @code{symver} directive
6972 @cindex symbol versioning
6973 @cindex versions of symbols
6974 Use the @code{.symver} directive to bind symbols to specific version nodes
6975 within a source file. This is only supported on ELF platforms, and is
6976 typically used when assembling files to be linked into a shared library.
6977 There are cases where it may make sense to use this in objects to be bound
6978 into an application itself so as to override a versioned symbol from a
6981 For ELF targets, the @code{.symver} directive can be used like this:
6983 .symver @var{name}, @var{name2@@nodename}
6985 If the symbol @var{name} is defined within the file
6986 being assembled, the @code{.symver} directive effectively creates a symbol
6987 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6988 just don't try and create a regular alias is that the @var{@@} character isn't
6989 permitted in symbol names. The @var{name2} part of the name is the actual name
6990 of the symbol by which it will be externally referenced. The name @var{name}
6991 itself is merely a name of convenience that is used so that it is possible to
6992 have definitions for multiple versions of a function within a single source
6993 file, and so that the compiler can unambiguously know which version of a
6994 function is being mentioned. The @var{nodename} portion of the alias should be
6995 the name of a node specified in the version script supplied to the linker when
6996 building a shared library. If you are attempting to override a versioned
6997 symbol from a shared library, then @var{nodename} should correspond to the
6998 nodename of the symbol you are trying to override.
7000 If the symbol @var{name} is not defined within the file being assembled, all
7001 references to @var{name} will be changed to @var{name2@@nodename}. If no
7002 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
7005 Another usage of the @code{.symver} directive is:
7007 .symver @var{name}, @var{name2@@@@nodename}
7009 In this case, the symbol @var{name} must exist and be defined within
7010 the file being assembled. It is similar to @var{name2@@nodename}. The
7011 difference is @var{name2@@@@nodename} will also be used to resolve
7012 references to @var{name2} by the linker.
7014 The third usage of the @code{.symver} directive is:
7016 .symver @var{name}, @var{name2@@@@@@nodename}
7018 When @var{name} is not defined within the
7019 file being assembled, it is treated as @var{name2@@nodename}. When
7020 @var{name} is defined within the file being assembled, the symbol
7021 name, @var{name}, will be changed to @var{name2@@@@nodename}.
7026 @section @code{.tag @var{structname}}
7028 @cindex COFF structure debugging
7029 @cindex structure debugging, COFF
7030 @cindex @code{tag} directive
7031 This directive is generated by compilers to include auxiliary debugging
7032 information in the symbol table. It is only permitted inside
7033 @code{.def}/@code{.endef} pairs. Tags are used to link structure
7034 definitions in the symbol table with instances of those structures.
7038 @section @code{.text @var{subsection}}
7040 @cindex @code{text} directive
7041 Tells @command{@value{AS}} to assemble the following statements onto the end of
7042 the text subsection numbered @var{subsection}, which is an absolute
7043 expression. If @var{subsection} is omitted, subsection number zero
7047 @section @code{.title "@var{heading}"}
7049 @cindex @code{title} directive
7050 @cindex listing control: title line
7051 Use @var{heading} as the title (second line, immediately after the
7052 source file name and pagenumber) when generating assembly listings.
7054 This directive affects subsequent pages, as well as the current page if
7055 it appears within ten lines of the top of a page.
7059 @section @code{.type}
7061 This directive is used to set the type of a symbol.
7065 @c only print the extra heading if both COFF and ELF are set
7066 @subheading COFF Version
7069 @cindex COFF symbol type
7070 @cindex symbol type, COFF
7071 @cindex @code{type} directive (COFF version)
7072 For COFF targets, this directive is permitted only within
7073 @code{.def}/@code{.endef} pairs. It is used like this:
7079 This records the integer @var{int} as the type attribute of a symbol table
7086 @c only print the extra heading if both COFF and ELF are set
7087 @subheading ELF Version
7090 @cindex ELF symbol type
7091 @cindex symbol type, ELF
7092 @cindex @code{type} directive (ELF version)
7093 For ELF targets, the @code{.type} directive is used like this:
7096 .type @var{name} , @var{type description}
7099 This sets the type of symbol @var{name} to be either a
7100 function symbol or an object symbol. There are five different syntaxes
7101 supported for the @var{type description} field, in order to provide
7102 compatibility with various other assemblers.
7104 Because some of the characters used in these syntaxes (such as @samp{@@} and
7105 @samp{#}) are comment characters for some architectures, some of the syntaxes
7106 below do not work on all architectures. The first variant will be accepted by
7107 the GNU assembler on all architectures so that variant should be used for
7108 maximum portability, if you do not need to assemble your code with other
7111 The syntaxes supported are:
7114 .type <name> STT_<TYPE_IN_UPPER_CASE>
7115 .type <name>,#<type>
7116 .type <name>,@@<type>
7117 .type <name>,%<type>
7118 .type <name>,"<type>"
7121 The types supported are:
7126 Mark the symbol as being a function name.
7129 @itemx gnu_indirect_function
7130 Mark the symbol as an indirect function when evaluated during reloc
7131 processing. (This is only supported on assemblers targeting GNU systems).
7135 Mark the symbol as being a data object.
7139 Mark the symbol as being a thread-local data object.
7143 Mark the symbol as being a common data object.
7147 Does not mark the symbol in any way. It is supported just for completeness.
7149 @item gnu_unique_object
7150 Marks the symbol as being a globally unique data object. The dynamic linker
7151 will make sure that in the entire process there is just one symbol with this
7152 name and type in use. (This is only supported on assemblers targeting GNU
7157 Note: Some targets support extra types in addition to those listed above.
7163 @section @code{.uleb128 @var{expressions}}
7165 @cindex @code{uleb128} directive
7166 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7167 compact, variable length representation of numbers used by the DWARF
7168 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7172 @section @code{.val @var{addr}}
7174 @cindex @code{val} directive
7175 @cindex COFF value attribute
7176 @cindex value attribute, COFF
7177 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7178 records the address @var{addr} as the value attribute of a symbol table
7184 @section @code{.version "@var{string}"}
7186 @cindex @code{version} directive
7187 This directive creates a @code{.note} section and places into it an ELF
7188 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7193 @section @code{.vtable_entry @var{table}, @var{offset}}
7195 @cindex @code{vtable_entry} directive
7196 This directive finds or creates a symbol @code{table} and creates a
7197 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7200 @section @code{.vtable_inherit @var{child}, @var{parent}}
7202 @cindex @code{vtable_inherit} directive
7203 This directive finds the symbol @code{child} and finds or creates the symbol
7204 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7205 parent whose addend is the value of the child symbol. As a special case the
7206 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7210 @section @code{.warning "@var{string}"}
7211 @cindex warning directive
7212 Similar to the directive @code{.error}
7213 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7216 @section @code{.weak @var{names}}
7218 @cindex @code{weak} directive
7219 This directive sets the weak attribute on the comma separated list of symbol
7220 @code{names}. If the symbols do not already exist, they will be created.
7222 On COFF targets other than PE, weak symbols are a GNU extension. This
7223 directive sets the weak attribute on the comma separated list of symbol
7224 @code{names}. If the symbols do not already exist, they will be created.
7226 On the PE target, weak symbols are supported natively as weak aliases.
7227 When a weak symbol is created that is not an alias, GAS creates an
7228 alternate symbol to hold the default value.
7231 @section @code{.weakref @var{alias}, @var{target}}
7233 @cindex @code{weakref} directive
7234 This directive creates an alias to the target symbol that enables the symbol to
7235 be referenced with weak-symbol semantics, but without actually making it weak.
7236 If direct references or definitions of the symbol are present, then the symbol
7237 will not be weak, but if all references to it are through weak references, the
7238 symbol will be marked as weak in the symbol table.
7240 The effect is equivalent to moving all references to the alias to a separate
7241 assembly source file, renaming the alias to the symbol in it, declaring the
7242 symbol as weak there, and running a reloadable link to merge the object files
7243 resulting from the assembly of the new source file and the old source file that
7244 had the references to the alias removed.
7246 The alias itself never makes to the symbol table, and is entirely handled
7247 within the assembler.
7250 @section @code{.word @var{expressions}}
7252 @cindex @code{word} directive
7253 This directive expects zero or more @var{expressions}, of any section,
7254 separated by commas.
7257 For each expression, @command{@value{AS}} emits a 32-bit number.
7260 For each expression, @command{@value{AS}} emits a 16-bit number.
7265 The size of the number emitted, and its byte order,
7266 depend on what target computer the assembly is for.
7269 @c on sparc the "special treatment to support compilers" doesn't
7270 @c happen---32-bit addressability, period; no long/short jumps.
7271 @ifset DIFF-TBL-KLUGE
7272 @cindex difference tables altered
7273 @cindex altered difference tables
7275 @emph{Warning: Special Treatment to support Compilers}
7279 Machines with a 32-bit address space, but that do less than 32-bit
7280 addressing, require the following special treatment. If the machine of
7281 interest to you does 32-bit addressing (or doesn't require it;
7282 @pxref{Machine Dependencies}), you can ignore this issue.
7285 In order to assemble compiler output into something that works,
7286 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7287 Directives of the form @samp{.word sym1-sym2} are often emitted by
7288 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7289 directive of the form @samp{.word sym1-sym2}, and the difference between
7290 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7291 creates a @dfn{secondary jump table}, immediately before the next label.
7292 This secondary jump table is preceded by a short-jump to the
7293 first byte after the secondary table. This short-jump prevents the flow
7294 of control from accidentally falling into the new table. Inside the
7295 table is a long-jump to @code{sym2}. The original @samp{.word}
7296 contains @code{sym1} minus the address of the long-jump to
7299 If there were several occurrences of @samp{.word sym1-sym2} before the
7300 secondary jump table, all of them are adjusted. If there was a
7301 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7302 long-jump to @code{sym4} is included in the secondary jump table,
7303 and the @code{.word} directives are adjusted to contain @code{sym3}
7304 minus the address of the long-jump to @code{sym4}; and so on, for as many
7305 entries in the original jump table as necessary.
7308 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7309 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7310 assembly language programmers.
7313 @c end DIFF-TBL-KLUGE
7315 @ifclear no-space-dir
7317 @section @code{.zero @var{size}}
7319 @cindex @code{zero} directive
7320 @cindex filling memory with zero bytes
7321 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7322 expression. This directive is actually an alias for the @samp{.skip} directive
7323 so in can take an optional second argument of the value to store in the bytes
7324 instead of zero. Using @samp{.zero} in this way would be confusing however.
7329 @section @code{.2byte @var{expression} [, @var{expression}]*}
7330 @cindex @code{2byte} directive
7331 @cindex two-byte integer
7332 @cindex integer, 2-byte
7334 This directive expects zero or more expressions, separated by commas. If there
7335 are no expressions then the directive does nothing. Otherwise each expression
7336 is evaluated in turn and placed in the next two bytes of the current output
7337 section, using the endian model of the target. If an expression will not fit
7338 in two bytes, a warning message is displayed and the least significant two
7339 bytes of the expression's value are used. If an expression cannot be evaluated
7340 at assembly time then relocations will be generated in order to compute the
7343 This directive does not apply any alignment before or after inserting the
7344 values. As a result of this, if relocations are generated, they may be
7345 different from those used for inserting values with a guaranteed alignment.
7347 This directive is only available for ELF targets,
7350 @section @code{.4byte @var{expression} [, @var{expression}]*}
7351 @cindex @code{4byte} directive
7352 @cindex four-byte integer
7353 @cindex integer, 4-byte
7355 Like the @option{.2byte} directive, except that it inserts unaligned, four byte
7356 long values into the output.
7359 @section @code{.8byte @var{expression} [, @var{expression}]*}
7360 @cindex @code{8byte} directive
7361 @cindex eight-byte integer
7362 @cindex integer, 8-byte
7364 Like the @option{.2byte} directive, except that it inserts unaligned, eight
7365 byte long bignum values into the output.
7370 @section Deprecated Directives
7372 @cindex deprecated directives
7373 @cindex obsolescent directives
7374 One day these directives won't work.
7375 They are included for compatibility with older assemblers.
7382 @node Object Attributes
7383 @chapter Object Attributes
7384 @cindex object attributes
7386 @command{@value{AS}} assembles source files written for a specific architecture
7387 into object files for that architecture. But not all object files are alike.
7388 Many architectures support incompatible variations. For instance, floating
7389 point arguments might be passed in floating point registers if the object file
7390 requires hardware floating point support---or floating point arguments might be
7391 passed in integer registers if the object file supports processors with no
7392 hardware floating point unit. Or, if two objects are built for different
7393 generations of the same architecture, the combination may require the
7394 newer generation at run-time.
7396 This information is useful during and after linking. At link time,
7397 @command{@value{LD}} can warn about incompatible object files. After link
7398 time, tools like @command{gdb} can use it to process the linked file
7401 Compatibility information is recorded as a series of object attributes. Each
7402 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7403 string, and indicates who sets the meaning of the tag. The tag is an integer,
7404 and indicates what property the attribute describes. The value may be a string
7405 or an integer, and indicates how the property affects this object. Missing
7406 attributes are the same as attributes with a zero value or empty string value.
7408 Object attributes were developed as part of the ABI for the ARM Architecture.
7409 The file format is documented in @cite{ELF for the ARM Architecture}.
7412 * GNU Object Attributes:: @sc{gnu} Object Attributes
7413 * Defining New Object Attributes:: Defining New Object Attributes
7416 @node GNU Object Attributes
7417 @section @sc{gnu} Object Attributes
7419 The @code{.gnu_attribute} directive records an object attribute
7420 with vendor @samp{gnu}.
7422 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7423 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7424 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7425 2} is set for architecture-independent attributes and clear for
7426 architecture-dependent ones.
7428 @subsection Common @sc{gnu} attributes
7430 These attributes are valid on all architectures.
7433 @item Tag_compatibility (32)
7434 The compatibility attribute takes an integer flag value and a vendor name. If
7435 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7436 then the file is only compatible with the named toolchain. If it is greater
7437 than 1, the file can only be processed by other toolchains under some private
7438 arrangement indicated by the flag value and the vendor name.
7441 @subsection MIPS Attributes
7444 @item Tag_GNU_MIPS_ABI_FP (4)
7445 The floating-point ABI used by this object file. The value will be:
7449 0 for files not affected by the floating-point ABI.
7451 1 for files using the hardware floating-point ABI with a standard
7452 double-precision FPU.
7454 2 for files using the hardware floating-point ABI with a single-precision FPU.
7456 3 for files using the software floating-point ABI.
7458 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7459 floating-point registers, 32-bit general-purpose registers and increased the
7460 number of callee-saved floating-point registers.
7462 5 for files using the hardware floating-point ABI with a double-precision FPU
7463 with either 32-bit or 64-bit floating-point registers and 32-bit
7464 general-purpose registers.
7466 6 for files using the hardware floating-point ABI with 64-bit floating-point
7467 registers and 32-bit general-purpose registers.
7469 7 for files using the hardware floating-point ABI with 64-bit floating-point
7470 registers, 32-bit general-purpose registers and a rule that forbids the
7471 direct use of odd-numbered single-precision floating-point registers.
7475 @subsection PowerPC Attributes
7478 @item Tag_GNU_Power_ABI_FP (4)
7479 The floating-point ABI used by this object file. The value will be:
7483 0 for files not affected by the floating-point ABI.
7485 1 for files using double-precision hardware floating-point ABI.
7487 2 for files using the software floating-point ABI.
7489 3 for files using single-precision hardware floating-point ABI.
7492 @item Tag_GNU_Power_ABI_Vector (8)
7493 The vector ABI used by this object file. The value will be:
7497 0 for files not affected by the vector ABI.
7499 1 for files using general purpose registers to pass vectors.
7501 2 for files using AltiVec registers to pass vectors.
7503 3 for files using SPE registers to pass vectors.
7507 @subsection IBM z Systems Attributes
7510 @item Tag_GNU_S390_ABI_Vector (8)
7511 The vector ABI used by this object file. The value will be:
7515 0 for files not affected by the vector ABI.
7517 1 for files using software vector ABI.
7519 2 for files using hardware vector ABI.
7523 @node Defining New Object Attributes
7524 @section Defining New Object Attributes
7526 If you want to define a new @sc{gnu} object attribute, here are the places you
7527 will need to modify. New attributes should be discussed on the @samp{binutils}
7532 This manual, which is the official register of attributes.
7534 The header for your architecture @file{include/elf}, to define the tag.
7536 The @file{bfd} support file for your architecture, to merge the attribute
7537 and issue any appropriate link warnings.
7539 Test cases in @file{ld/testsuite} for merging and link warnings.
7541 @file{binutils/readelf.c} to display your attribute.
7543 GCC, if you want the compiler to mark the attribute automatically.
7549 @node Machine Dependencies
7550 @chapter Machine Dependent Features
7552 @cindex machine dependencies
7553 The machine instruction sets are (almost by definition) different on
7554 each machine where @command{@value{AS}} runs. Floating point representations
7555 vary as well, and @command{@value{AS}} often supports a few additional
7556 directives or command-line options for compatibility with other
7557 assemblers on a particular platform. Finally, some versions of
7558 @command{@value{AS}} support special pseudo-instructions for branch
7561 This chapter discusses most of these differences, though it does not
7562 include details on any machine's instruction set. For details on that
7563 subject, see the hardware manufacturer's manual.
7567 * AArch64-Dependent:: AArch64 Dependent Features
7570 * Alpha-Dependent:: Alpha Dependent Features
7573 * ARC-Dependent:: ARC Dependent Features
7576 * ARM-Dependent:: ARM Dependent Features
7579 * AVR-Dependent:: AVR Dependent Features
7582 * Blackfin-Dependent:: Blackfin Dependent Features
7585 * CR16-Dependent:: CR16 Dependent Features
7588 * CRIS-Dependent:: CRIS Dependent Features
7591 * C-SKY-Dependent:: C-SKY Dependent Features
7594 * D10V-Dependent:: D10V Dependent Features
7597 * D30V-Dependent:: D30V Dependent Features
7600 * Epiphany-Dependent:: EPIPHANY Dependent Features
7603 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7606 * HPPA-Dependent:: HPPA Dependent Features
7609 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7612 * IA-64-Dependent:: Intel IA-64 Dependent Features
7615 * IP2K-Dependent:: IP2K Dependent Features
7618 * LM32-Dependent:: LM32 Dependent Features
7621 * M32C-Dependent:: M32C Dependent Features
7624 * M32R-Dependent:: M32R Dependent Features
7627 * M68K-Dependent:: M680x0 Dependent Features
7630 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7633 * S12Z-Dependent:: S12Z Dependent Features
7636 * Meta-Dependent :: Meta Dependent Features
7639 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7642 * MIPS-Dependent:: MIPS Dependent Features
7645 * MMIX-Dependent:: MMIX Dependent Features
7648 * MSP430-Dependent:: MSP430 Dependent Features
7651 * NDS32-Dependent:: Andes NDS32 Dependent Features
7654 * NiosII-Dependent:: Altera Nios II Dependent Features
7657 * NS32K-Dependent:: NS32K Dependent Features
7660 * PDP-11-Dependent:: PDP-11 Dependent Features
7663 * PJ-Dependent:: picoJava Dependent Features
7666 * PPC-Dependent:: PowerPC Dependent Features
7669 * PRU-Dependent:: PRU Dependent Features
7672 * RISC-V-Dependent:: RISC-V Dependent Features
7675 * RL78-Dependent:: RL78 Dependent Features
7678 * RX-Dependent:: RX Dependent Features
7681 * S/390-Dependent:: IBM S/390 Dependent Features
7684 * SCORE-Dependent:: SCORE Dependent Features
7687 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7690 * Sparc-Dependent:: SPARC Dependent Features
7693 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7696 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7699 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7702 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7705 * V850-Dependent:: V850 Dependent Features
7708 * Vax-Dependent:: VAX Dependent Features
7711 * Visium-Dependent:: Visium Dependent Features
7714 * WebAssembly-Dependent:: WebAssembly Dependent Features
7717 * XGATE-Dependent:: XGATE Dependent Features
7720 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7723 * Xtensa-Dependent:: Xtensa Dependent Features
7726 * Z80-Dependent:: Z80 Dependent Features
7729 * Z8000-Dependent:: Z8000 Dependent Features
7736 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7737 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7738 @c peculiarity: to preserve cross-references, there must be a node called
7739 @c "Machine Dependencies". Hence the conditional nodenames in each
7740 @c major node below. Node defaulting in makeinfo requires adjacency of
7741 @c node and sectioning commands; hence the repetition of @chapter BLAH
7742 @c in both conditional blocks.
7745 @include c-aarch64.texi
7749 @include c-alpha.texi
7765 @include c-bfin.texi
7769 @include c-cr16.texi
7773 @include c-cris.texi
7777 @include c-csky.texi
7782 @node Machine Dependencies
7783 @chapter Machine Dependent Features
7785 The machine instruction sets are different on each Renesas chip family,
7786 and there are also some syntax differences among the families. This
7787 chapter describes the specific @command{@value{AS}} features for each
7791 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7792 * SH-Dependent:: Renesas SH Dependent Features
7799 @include c-d10v.texi
7803 @include c-d30v.texi
7807 @include c-epiphany.texi
7811 @include c-h8300.texi
7815 @include c-hppa.texi
7819 @include c-i386.texi
7823 @include c-ia64.texi
7827 @include c-ip2k.texi
7831 @include c-lm32.texi
7835 @include c-m32c.texi
7839 @include c-m32r.texi
7843 @include c-m68k.texi
7847 @include c-m68hc11.texi
7851 @include c-s12z.texi
7855 @include c-metag.texi
7859 @include c-microblaze.texi
7863 @include c-mips.texi
7867 @include c-mmix.texi
7871 @include c-msp430.texi
7875 @include c-nds32.texi
7879 @include c-nios2.texi
7883 @include c-ns32k.texi
7887 @include c-pdp11.texi
7903 @include c-riscv.texi
7907 @include c-rl78.texi
7915 @include c-s390.texi
7919 @include c-score.texi
7927 @include c-sparc.texi
7931 @include c-tic54x.texi
7935 @include c-tic6x.texi
7939 @include c-tilegx.texi
7943 @include c-tilepro.texi
7947 @include c-v850.texi
7955 @include c-visium.texi
7959 @include c-wasm32.texi
7963 @include c-xgate.texi
7967 @include c-xstormy16.texi
7971 @include c-xtensa.texi
7983 @c reverse effect of @down at top of generic Machine-Dep chapter
7987 @node Reporting Bugs
7988 @chapter Reporting Bugs
7989 @cindex bugs in assembler
7990 @cindex reporting bugs in assembler
7992 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7994 Reporting a bug may help you by bringing a solution to your problem, or it may
7995 not. But in any case the principal function of a bug report is to help the
7996 entire community by making the next version of @command{@value{AS}} work better.
7997 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7999 In order for a bug report to serve its purpose, you must include the
8000 information that enables us to fix the bug.
8003 * Bug Criteria:: Have you found a bug?
8004 * Bug Reporting:: How to report bugs
8008 @section Have You Found a Bug?
8009 @cindex bug criteria
8011 If you are not sure whether you have found a bug, here are some guidelines:
8014 @cindex fatal signal
8015 @cindex assembler crash
8016 @cindex crash of assembler
8018 If the assembler gets a fatal signal, for any input whatever, that is a
8019 @command{@value{AS}} bug. Reliable assemblers never crash.
8021 @cindex error on valid input
8023 If @command{@value{AS}} produces an error message for valid input, that is a bug.
8025 @cindex invalid input
8027 If @command{@value{AS}} does not produce an error message for invalid input, that
8028 is a bug. However, you should note that your idea of ``invalid input'' might
8029 be our idea of ``an extension'' or ``support for traditional practice''.
8032 If you are an experienced user of assemblers, your suggestions for improvement
8033 of @command{@value{AS}} are welcome in any case.
8037 @section How to Report Bugs
8039 @cindex assembler bugs, reporting
8041 A number of companies and individuals offer support for @sc{gnu} products. If
8042 you obtained @command{@value{AS}} from a support organization, we recommend you
8043 contact that organization first.
8045 You can find contact information for many support companies and
8046 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
8050 In any event, we also recommend that you send bug reports for @command{@value{AS}}
8054 The fundamental principle of reporting bugs usefully is this:
8055 @strong{report all the facts}. If you are not sure whether to state a
8056 fact or leave it out, state it!
8058 Often people omit facts because they think they know what causes the problem
8059 and assume that some details do not matter. Thus, you might assume that the
8060 name of a symbol you use in an example does not matter. Well, probably it does
8061 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
8062 happens to fetch from the location where that name is stored in memory;
8063 perhaps, if the name were different, the contents of that location would fool
8064 the assembler into doing the right thing despite the bug. Play it safe and
8065 give a specific, complete example. That is the easiest thing for you to do,
8066 and the most helpful.
8068 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
8069 it is new to us. Therefore, always write your bug reports on the assumption
8070 that the bug has not been reported previously.
8072 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8073 bell?'' This cannot help us fix a bug, so it is basically useless. We
8074 respond by asking for enough details to enable us to investigate.
8075 You might as well expedite matters by sending them to begin with.
8077 To enable us to fix the bug, you should include all these things:
8081 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8082 it with the @samp{--version} argument.
8084 Without this, we will not know whether there is any point in looking for
8085 the bug in the current version of @command{@value{AS}}.
8088 Any patches you may have applied to the @command{@value{AS}} source.
8091 The type of machine you are using, and the operating system name and
8095 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8099 The command arguments you gave the assembler to assemble your example and
8100 observe the bug. To guarantee you will not omit something important, list them
8101 all. A copy of the Makefile (or the output from make) is sufficient.
8103 If we were to try to guess the arguments, we would probably guess wrong
8104 and then we might not encounter the bug.
8107 A complete input file that will reproduce the bug. If the bug is observed when
8108 the assembler is invoked via a compiler, send the assembler source, not the
8109 high level language source. Most compilers will produce the assembler source
8110 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8111 the options @samp{-v --save-temps}; this will save the assembler source in a
8112 file with an extension of @file{.s}, and also show you exactly how
8113 @command{@value{AS}} is being run.
8116 A description of what behavior you observe that you believe is
8117 incorrect. For example, ``It gets a fatal signal.''
8119 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8120 will certainly notice it. But if the bug is incorrect output, we might not
8121 notice unless it is glaringly wrong. You might as well not give us a chance to
8124 Even if the problem you experience is a fatal signal, you should still say so
8125 explicitly. Suppose something strange is going on, such as, your copy of
8126 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8127 library on your system. (This has happened!) Your copy might crash and ours
8128 would not. If you told us to expect a crash, then when ours fails to crash, we
8129 would know that the bug was not happening for us. If you had not told us to
8130 expect a crash, then we would not be able to draw any conclusion from our
8134 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8135 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8136 option. Always send diffs from the old file to the new file. If you even
8137 discuss something in the @command{@value{AS}} source, refer to it by context, not
8140 The line numbers in our development sources will not match those in your
8141 sources. Your line numbers would convey no useful information to us.
8144 Here are some things that are not necessary:
8148 A description of the envelope of the bug.
8150 Often people who encounter a bug spend a lot of time investigating
8151 which changes to the input file will make the bug go away and which
8152 changes will not affect it.
8154 This is often time consuming and not very useful, because the way we
8155 will find the bug is by running a single example under the debugger
8156 with breakpoints, not by pure deduction from a series of examples.
8157 We recommend that you save your time for something else.
8159 Of course, if you can find a simpler example to report @emph{instead}
8160 of the original one, that is a convenience for us. Errors in the
8161 output will be easier to spot, running under the debugger will take
8162 less time, and so on.
8164 However, simplification is not vital; if you do not want to do this,
8165 report the bug anyway and send us the entire test case you used.
8168 A patch for the bug.
8170 A patch for the bug does help us if it is a good one. But do not omit
8171 the necessary information, such as the test case, on the assumption that
8172 a patch is all we need. We might see problems with your patch and decide
8173 to fix the problem another way, or we might not understand it at all.
8175 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8176 construct an example that will make the program follow a certain path through
8177 the code. If you do not send us the example, we will not be able to construct
8178 one, so we will not be able to verify that the bug is fixed.
8180 And if we cannot understand what bug you are trying to fix, or why your
8181 patch should be an improvement, we will not install it. A test case will
8182 help us to understand.
8185 A guess about what the bug is or what it depends on.
8187 Such guesses are usually wrong. Even we cannot guess right about such
8188 things without first using the debugger to find the facts.
8191 @node Acknowledgements
8192 @chapter Acknowledgements
8194 If you have contributed to GAS and your name isn't listed here,
8195 it is not meant as a slight. We just don't know about it. Send mail to the
8196 maintainer, and we'll correct the situation. Currently
8198 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8200 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8203 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8204 information and the 68k series machines, most of the preprocessing pass, and
8205 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8207 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8208 many bug fixes, including merging support for several processors, breaking GAS
8209 up to handle multiple object file format back ends (including heavy rewrite,
8210 testing, an integration of the coff and b.out back ends), adding configuration
8211 including heavy testing and verification of cross assemblers and file splits
8212 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8213 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8214 port (including considerable amounts of reverse engineering), a SPARC opcode
8215 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8216 assertions and made them work, much other reorganization, cleanup, and lint.
8218 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8219 in format-specific I/O modules.
8221 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8222 has done much work with it since.
8224 The Intel 80386 machine description was written by Eliot Dresselhaus.
8226 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8228 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8229 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8231 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8232 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8233 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8234 support a.out format.
8236 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8237 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8238 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8239 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8242 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8243 simplified the configuration of which versions accept which directives. He
8244 updated the 68k machine description so that Motorola's opcodes always produced
8245 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8246 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8247 cross-compilation support, and one bug in relaxation that took a week and
8248 required the proverbial one-bit fix.
8250 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8251 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8252 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8253 PowerPC assembler, and made a few other minor patches.
8255 Steve Chamberlain made GAS able to generate listings.
8257 Hewlett-Packard contributed support for the HP9000/300.
8259 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8260 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8261 formats). This work was supported by both the Center for Software Science at
8262 the University of Utah and Cygnus Support.
8264 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8265 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8266 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8267 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8268 and some initial 64-bit support).
8270 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8272 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8273 support for openVMS/Alpha.
8275 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8278 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8279 Inc.@: added support for Xtensa processors.
8281 Several engineers at Cygnus Support have also provided many small bug fixes and
8282 configuration enhancements.
8284 Jon Beniston added support for the Lattice Mico32 architecture.
8286 Many others have contributed large or small bugfixes and enhancements. If
8287 you have contributed significant work and are not mentioned on this list, and
8288 want to be, let us know. Some of the history has been lost; we are not
8289 intentionally leaving anyone out.
8291 @node GNU Free Documentation License
8292 @appendix GNU Free Documentation License
8296 @unnumbered AS Index