1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (C) 1991-2018 Free Software Foundation, Inc.
3 @c UPDATE!! On future updates--
4 @c (1) check for new machine-dep cmdline options in
5 @c md_parse_option definitions in config/tc-*.c
6 @c (2) for platform-specific directives, examine md_pseudo_op
8 @c (3) for object-format specific directives, examine obj_pseudo_op
10 @c (4) portable directives in potable[] in read.c
14 @macro gcctabopt{body}
17 @c defaults, config file may override:
22 @include asconfig.texi
27 @c common OR combinations of conditions
50 @set abnormal-separator
54 @settitle Using @value{AS}
57 @settitle Using @value{AS} (@value{TARGET})
59 @setchapternewpage odd
64 @c WARE! Some of the machine-dependent sections contain tables of machine
65 @c instructions. Except in multi-column format, these tables look silly.
66 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
67 @c the multi-col format is faked within @example sections.
69 @c Again unfortunately, the natural size that fits on a page, for these tables,
70 @c is different depending on whether or not smallbook is turned on.
71 @c This matters, because of order: text flow switches columns at each page
74 @c The format faked in this source works reasonably well for smallbook,
75 @c not well for the default large-page format. This manual expects that if you
76 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
77 @c tables in question. You can turn on one without the other at your
78 @c discretion, of course.
81 @c the insn tables look just as silly in info files regardless of smallbook,
82 @c might as well show 'em anyways.
86 @dircategory Software development
88 * As: (as). The GNU assembler.
89 * Gas: (as). The GNU assembler.
97 This file documents the GNU Assembler "@value{AS}".
99 @c man begin COPYRIGHT
100 Copyright @copyright{} 1991-2018 Free Software Foundation, Inc.
102 Permission is granted to copy, distribute and/or modify this document
103 under the terms of the GNU Free Documentation License, Version 1.3
104 or any later version published by the Free Software Foundation;
105 with no Invariant Sections, with no Front-Cover Texts, and with no
106 Back-Cover Texts. A copy of the license is included in the
107 section entitled ``GNU Free Documentation License''.
113 @title Using @value{AS}
114 @subtitle The @sc{gnu} Assembler
116 @subtitle for the @value{TARGET} family
118 @ifset VERSION_PACKAGE
120 @subtitle @value{VERSION_PACKAGE}
123 @subtitle Version @value{VERSION}
126 The Free Software Foundation Inc.@: thanks The Nice Computer
127 Company of Australia for loaning Dean Elsner to write the
128 first (Vax) version of @command{as} for Project @sc{gnu}.
129 The proprietors, management and staff of TNCCA thank FSF for
130 distracting the boss while they got some work
133 @author Dean Elsner, Jay Fenlason & friends
137 \hfill {\it Using {\tt @value{AS}}}\par
138 \hfill Edited by Cygnus Support\par
140 %"boxit" macro for figures:
141 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
142 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
143 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
144 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
145 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
148 @vskip 0pt plus 1filll
149 Copyright @copyright{} 1991-2018 Free Software Foundation, Inc.
151 Permission is granted to copy, distribute and/or modify this document
152 under the terms of the GNU Free Documentation License, Version 1.3
153 or any later version published by the Free Software Foundation;
154 with no Invariant Sections, with no Front-Cover Texts, and with no
155 Back-Cover Texts. A copy of the license is included in the
156 section entitled ``GNU Free Documentation License''.
163 @top Using @value{AS}
165 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
166 @ifset VERSION_PACKAGE
167 @value{VERSION_PACKAGE}
169 version @value{VERSION}.
171 This version of the file describes @command{@value{AS}} configured to generate
172 code for @value{TARGET} architectures.
175 This document is distributed under the terms of the GNU Free
176 Documentation License. A copy of the license is included in the
177 section entitled ``GNU Free Documentation License''.
180 * Overview:: Overview
181 * Invoking:: Command-Line Options
183 * Sections:: Sections and Relocation
185 * Expressions:: Expressions
186 * Pseudo Ops:: Assembler Directives
188 * Object Attributes:: Object Attributes
190 * Machine Dependencies:: Machine Dependent Features
191 * Reporting Bugs:: Reporting Bugs
192 * Acknowledgements:: Who Did What
193 * GNU Free Documentation License:: GNU Free Documentation License
194 * AS Index:: AS Index
201 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
203 This version of the manual describes @command{@value{AS}} configured to generate
204 code for @value{TARGET} architectures.
208 @cindex invocation summary
209 @cindex option summary
210 @cindex summary of options
211 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
212 see @ref{Invoking,,Command-Line Options}.
214 @c man title AS the portable GNU assembler.
218 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
222 @c We don't use deffn and friends for the following because they seem
223 @c to be limited to one line for the header.
225 @c man begin SYNOPSIS
226 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
227 [@b{--compress-debug-sections}] [@b{--nocompress-debug-sections}]
228 [@b{--debug-prefix-map} @var{old}=@var{new}]
229 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
230 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--gdwarf-sections}]
231 [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
232 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
233 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
234 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}]
235 [@b{--no-pad-sections}]
236 [@b{-o} @var{objfile}] [@b{-R}]
237 [@b{--hash-size}=@var{NUM}] [@b{--reduce-memory-overheads}]
239 [@b{-v}] [@b{-version}] [@b{--version}]
240 [@b{-W}] [@b{--warn}] [@b{--fatal-warnings}] [@b{-w}] [@b{-x}]
241 [@b{-Z}] [@b{@@@var{FILE}}]
242 [@b{--sectname-subst}] [@b{--size-check=[error|warning]}]
243 [@b{--elf-stt-common=[no|yes]}]
244 [@b{--generate-missing-build-notes=[no|yes]}]
245 [@b{--target-help}] [@var{target-options}]
246 [@b{--}|@var{files} @dots{}]
249 @c Target dependent options are listed below. Keep the list sorted.
250 @c Add an empty line for separation.
254 @emph{Target AArch64 options:}
256 [@b{-mabi}=@var{ABI}]
260 @emph{Target Alpha options:}
262 [@b{-mdebug} | @b{-no-mdebug}]
263 [@b{-replace} | @b{-noreplace}]
264 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
265 [@b{-F}] [@b{-32addr}]
269 @emph{Target ARC options:}
270 [@b{-mcpu=@var{cpu}}]
271 [@b{-mA6}|@b{-mARC600}|@b{-mARC601}|@b{-mA7}|@b{-mARC700}|@b{-mEM}|@b{-mHS}]
278 @emph{Target ARM options:}
279 @c Don't document the deprecated options
280 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
281 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
282 [@b{-mfpu}=@var{floating-point-format}]
283 [@b{-mfloat-abi}=@var{abi}]
284 [@b{-meabi}=@var{ver}]
287 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
288 @b{-mapcs-reentrant}]
289 [@b{-mthumb-interwork}] [@b{-k}]
293 @emph{Target Blackfin options:}
294 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
301 @emph{Target CRIS options:}
302 [@b{--underscore} | @b{--no-underscore}]
304 [@b{--emulation=criself} | @b{--emulation=crisaout}]
305 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
306 @c Deprecated -- deliberately not documented.
311 @emph{Target C-SKY options:}
312 [@b{-march=@var{arch}}] [@b{-mcpu=@var{cpu}}]
313 [@b{-EL}] [@b{-mlittle-endian}] [@b{-EB}] [@b{-mbig-endian}]
314 [@b{-fpic}] [@b{-pic}]
315 [@b{-mljump}] [@b{-mno-ljump}]
316 [@b{-force2bsr}] [@b{-mforce2bsr}] [@b{-no-force2bsr}] [@b{-mno-force2bsr}]
317 [@b{-jsri2bsr}] [@b{-mjsri2bsr}] [@b{-no-jsri2bsr }] [@b{-mno-jsri2bsr}]
318 [@b{-mnolrw }] [@b{-mno-lrw}]
319 [@b{-melrw}] [@b{-mno-elrw}]
320 [@b{-mlaf }] [@b{-mliterals-after-func}]
321 [@b{-mno-laf}] [@b{-mno-literals-after-func}]
322 [@b{-mlabr}] [@b{-mliterals-after-br}]
323 [@b{-mno-labr}] [@b{-mnoliterals-after-br}]
324 [@b{-mistack}] [@b{-mno-istack}]
325 [@b{-mhard-float}] [@b{-mmp}] [@b{-mcp}] [@b{-mcache}]
326 [@b{-msecurity}] [@b{-mtrust}]
327 [@b{-mdsp}] [@b{-medsp}] [@b{-mvdsp}]
331 @emph{Target D10V options:}
336 @emph{Target D30V options:}
337 [@b{-O}|@b{-n}|@b{-N}]
341 @emph{Target EPIPHANY options:}
342 [@b{-mepiphany}|@b{-mepiphany16}]
346 @emph{Target H8/300 options:}
350 @c HPPA has no machine-dependent assembler options (yet).
354 @emph{Target i386 options:}
355 [@b{--32}|@b{--x32}|@b{--64}] [@b{-n}]
356 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
360 @emph{Target IA-64 options:}
361 [@b{-mconstant-gp}|@b{-mauto-pic}]
362 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
364 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
365 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
366 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
367 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
371 @emph{Target IP2K options:}
372 [@b{-mip2022}|@b{-mip2022ext}]
376 @emph{Target M32C options:}
377 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
381 @emph{Target M32R options:}
382 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
387 @emph{Target M680X0 options:}
388 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
392 @emph{Target M68HC11 options:}
393 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}|@b{-mm9s12x}|@b{-mm9s12xg}]
394 [@b{-mshort}|@b{-mlong}]
395 [@b{-mshort-double}|@b{-mlong-double}]
396 [@b{--force-long-branches}] [@b{--short-branches}]
397 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
398 [@b{--print-opcodes}] [@b{--generate-example}]
402 @emph{Target MCORE options:}
403 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
404 [@b{-mcpu=[210|340]}]
408 @emph{Target Meta options:}
409 [@b{-mcpu=@var{cpu}}] [@b{-mfpu=@var{cpu}}] [@b{-mdsp=@var{cpu}}]
412 @emph{Target MICROBLAZE options:}
413 @c MicroBlaze has no machine-dependent assembler options.
417 @emph{Target MIPS options:}
418 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
419 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
420 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
421 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
422 [@b{-mfp64}] [@b{-mgp64}] [@b{-mfpxx}]
423 [@b{-modd-spreg}] [@b{-mno-odd-spreg}]
424 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
425 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
426 [@b{-mips32r3}] [@b{-mips32r5}] [@b{-mips32r6}] [@b{-mips64}] [@b{-mips64r2}]
427 [@b{-mips64r3}] [@b{-mips64r5}] [@b{-mips64r6}]
428 [@b{-construct-floats}] [@b{-no-construct-floats}]
429 [@b{-mignore-branch-isa}] [@b{-mno-ignore-branch-isa}]
430 [@b{-mnan=@var{encoding}}]
431 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
432 [@b{-mips16}] [@b{-no-mips16}]
433 [@b{-mmips16e2}] [@b{-mno-mips16e2}]
434 [@b{-mmicromips}] [@b{-mno-micromips}]
435 [@b{-msmartmips}] [@b{-mno-smartmips}]
436 [@b{-mips3d}] [@b{-no-mips3d}]
437 [@b{-mdmx}] [@b{-no-mdmx}]
438 [@b{-mdsp}] [@b{-mno-dsp}]
439 [@b{-mdspr2}] [@b{-mno-dspr2}]
440 [@b{-mdspr3}] [@b{-mno-dspr3}]
441 [@b{-mmsa}] [@b{-mno-msa}]
442 [@b{-mxpa}] [@b{-mno-xpa}]
443 [@b{-mmt}] [@b{-mno-mt}]
444 [@b{-mmcu}] [@b{-mno-mcu}]
445 [@b{-mcrc}] [@b{-mno-crc}]
446 [@b{-mginv}] [@b{-mno-ginv}]
447 [@b{-mloongson-mmi}] [@b{-mno-loongson-mmi}]
448 [@b{-mloongson-cam}] [@b{-mno-loongson-cam}]
449 [@b{-mloongson-ext}] [@b{-mno-loongson-ext}]
450 [@b{-minsn32}] [@b{-mno-insn32}]
451 [@b{-mfix7000}] [@b{-mno-fix7000}]
452 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
453 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
454 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
455 [@b{-mdebug}] [@b{-no-mdebug}]
456 [@b{-mpdr}] [@b{-mno-pdr}]
460 @emph{Target MMIX options:}
461 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
462 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
463 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
464 [@b{--linker-allocated-gregs}]
468 @emph{Target Nios II options:}
469 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
474 @emph{Target NDS32 options:}
475 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
476 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
477 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
478 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
479 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
480 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
481 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
486 @emph{Target PDP11 options:}
487 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
488 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
489 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
493 @emph{Target picoJava options:}
498 @emph{Target PowerPC options:}
500 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
501 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mgekko}|
502 @b{-mbroadway}|@b{-mppc64}|@b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|
503 @b{-me6500}|@b{-mppc64bridge}|@b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|
504 @b{-mpower6}|@b{-mpwr6}|@b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-mpower9}|@b{-mpwr9}@b{-ma2}|
505 @b{-mcell}|@b{-mspe}|@b{-mspe2}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
506 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
507 [@b{-mregnames}|@b{-mno-regnames}]
508 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
509 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
510 [@b{-msolaris}|@b{-mno-solaris}]
511 [@b{-nops=@var{count}}]
515 @emph{Target PRU options:}
518 [@b{-mno-warn-regname-label}]
522 @emph{Target RISC-V options:}
523 [@b{-fpic}|@b{-fPIC}|@b{-fno-pic}]
524 [@b{-march}=@var{ISA}]
525 [@b{-mabi}=@var{ABI}]
529 @emph{Target RL78 options:}
531 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
535 @emph{Target RX options:}
536 [@b{-mlittle-endian}|@b{-mbig-endian}]
537 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
538 [@b{-muse-conventional-section-names}]
539 [@b{-msmall-data-limit}]
542 [@b{-mint-register=@var{number}}]
543 [@b{-mgcc-abi}|@b{-mrx-abi}]
547 @emph{Target s390 options:}
548 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
549 [@b{-mregnames}|@b{-mno-regnames}]
550 [@b{-mwarn-areg-zero}]
554 @emph{Target SCORE options:}
555 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
556 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
557 [@b{-march=score7}][@b{-march=score3}]
558 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
562 @emph{Target SPARC options:}
563 @c The order here is important. See c-sparc.texi.
564 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Aleon}|@b{-Asparclet}|@b{-Asparclite}
565 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av8plusb}|@b{-Av8plusc}|@b{-Av8plusd}
566 @b{-Av8plusv}|@b{-Av8plusm}|@b{-Av9}|@b{-Av9a}|@b{-Av9b}|@b{-Av9c}
567 @b{-Av9d}|@b{-Av9e}|@b{-Av9v}|@b{-Av9m}|@b{-Asparc}|@b{-Asparcvis}
568 @b{-Asparcvis2}|@b{-Asparcfmaf}|@b{-Asparcima}|@b{-Asparcvis3}
569 @b{-Asparcvisr}|@b{-Asparc5}]
570 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}]|@b{-xarch=v8plusb}|@b{-xarch=v8plusc}
571 @b{-xarch=v8plusd}|@b{-xarch=v8plusv}|@b{-xarch=v8plusm}|@b{-xarch=v9}
572 @b{-xarch=v9a}|@b{-xarch=v9b}|@b{-xarch=v9c}|@b{-xarch=v9d}|@b{-xarch=v9e}
573 @b{-xarch=v9v}|@b{-xarch=v9m}|@b{-xarch=sparc}|@b{-xarch=sparcvis}
574 @b{-xarch=sparcvis2}|@b{-xarch=sparcfmaf}|@b{-xarch=sparcima}
575 @b{-xarch=sparcvis3}|@b{-xarch=sparcvisr}|@b{-xarch=sparc5}
578 [@b{--enforce-aligned-data}][@b{--dcti-couples-detect}]
582 @emph{Target TIC54X options:}
583 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
584 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
588 @emph{Target TIC6X options:}
589 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
590 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
591 [@b{-mpic}|@b{-mno-pic}]
595 @emph{Target TILE-Gx options:}
596 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
599 @c TILEPro has no machine-dependent assembler options
603 @emph{Target Visium options:}
604 [@b{-mtune=@var{arch}}]
608 @emph{Target Xtensa options:}
609 [@b{--[no-]text-section-literals}] [@b{--[no-]auto-litpools}]
610 [@b{--[no-]absolute-literals}]
611 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
612 [@b{--[no-]transform}]
613 [@b{--rename-section} @var{oldname}=@var{newname}]
614 [@b{--[no-]trampolines}]
618 @emph{Target Z80 options:}
619 [@b{-z80}] [@b{-r800}]
620 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
621 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
622 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
623 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
624 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
625 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
629 @c Z8000 has no machine-dependent assembler options
638 @include at-file.texi
641 Turn on listings, in any of a variety of ways:
645 omit false conditionals
648 omit debugging directives
651 include general information, like @value{AS} version and options passed
654 include high-level source
660 include macro expansions
663 omit forms processing
669 set the name of the listing file
672 You may combine these options; for example, use @samp{-aln} for assembly
673 listing without forms processing. The @samp{=file} option, if used, must be
674 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
677 Begin in alternate macro mode.
679 @xref{Altmacro,,@code{.altmacro}}.
682 @item --compress-debug-sections
683 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
684 ELF ABI. The resulting object file may not be compatible with older
685 linkers and object file utilities. Note if compression would make a
686 given section @emph{larger} then it is not compressed.
689 @cindex @samp{--compress-debug-sections=} option
690 @item --compress-debug-sections=none
691 @itemx --compress-debug-sections=zlib
692 @itemx --compress-debug-sections=zlib-gnu
693 @itemx --compress-debug-sections=zlib-gabi
694 These options control how DWARF debug sections are compressed.
695 @option{--compress-debug-sections=none} is equivalent to
696 @option{--nocompress-debug-sections}.
697 @option{--compress-debug-sections=zlib} and
698 @option{--compress-debug-sections=zlib-gabi} are equivalent to
699 @option{--compress-debug-sections}.
700 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
701 sections using zlib. The debug sections are renamed to begin with
702 @samp{.zdebug}. Note if compression would make a given section
703 @emph{larger} then it is not compressed nor renamed.
707 @item --nocompress-debug-sections
708 Do not compress DWARF debug sections. This is usually the default for all
709 targets except the x86/x86_64, but a configure time option can be used to
713 Ignored. This option is accepted for script compatibility with calls to
716 @item --debug-prefix-map @var{old}=@var{new}
717 When assembling files in directory @file{@var{old}}, record debugging
718 information describing them as in @file{@var{new}} instead.
720 @item --defsym @var{sym}=@var{value}
721 Define the symbol @var{sym} to be @var{value} before assembling the input file.
722 @var{value} must be an integer constant. As in C, a leading @samp{0x}
723 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
724 value. The value of the symbol can be overridden inside a source file via the
725 use of a @code{.set} pseudo-op.
728 ``fast''---skip whitespace and comment preprocessing (assume source is
733 Generate debugging information for each assembler source line using whichever
734 debug format is preferred by the target. This currently means either STABS,
738 Generate stabs debugging information for each assembler line. This
739 may help debugging assembler code, if the debugger can handle it.
742 Generate stabs debugging information for each assembler line, with GNU
743 extensions that probably only gdb can handle, and that could make other
744 debuggers crash or refuse to read your program. This
745 may help debugging assembler code. Currently the only GNU extension is
746 the location of the current working directory at assembling time.
749 Generate DWARF2 debugging information for each assembler line. This
750 may help debugging assembler code, if the debugger can handle it. Note---this
751 option is only supported by some targets, not all of them.
753 @item --gdwarf-sections
754 Instead of creating a .debug_line section, create a series of
755 .debug_line.@var{foo} sections where @var{foo} is the name of the
756 corresponding code section. For example a code section called @var{.text.func}
757 will have its dwarf line number information placed into a section called
758 @var{.debug_line.text.func}. If the code section is just called @var{.text}
759 then debug line section will still be called just @var{.debug_line} without any
763 @item --size-check=error
764 @itemx --size-check=warning
765 Issue an error or warning for invalid ELF .size directive.
767 @item --elf-stt-common=no
768 @itemx --elf-stt-common=yes
769 These options control whether the ELF assembler should generate common
770 symbols with the @code{STT_COMMON} type. The default can be controlled
771 by a configure option @option{--enable-elf-stt-common}.
773 @item --generate-missing-build-notes=yes
774 @itemx --generate-missing-build-notes=no
775 These options control whether the ELF assembler should generate GNU Build
776 attribute notes if none are present in the input sources.
777 The default can be controlled by the @option{--enable-generate-build-notes}
783 Print a summary of the command-line options and exit.
786 Print a summary of all target specific options and exit.
789 Add directory @var{dir} to the search list for @code{.include} directives.
792 Don't warn about signed overflow.
795 @ifclear DIFF-TBL-KLUGE
796 This option is accepted but has no effect on the @value{TARGET} family.
798 @ifset DIFF-TBL-KLUGE
799 Issue warnings when difference tables altered for long displacements.
804 Keep (in the symbol table) local symbols. These symbols start with
805 system-specific local label prefixes, typically @samp{.L} for ELF systems
806 or @samp{L} for traditional a.out systems.
811 @item --listing-lhs-width=@var{number}
812 Set the maximum width, in words, of the output data column for an assembler
813 listing to @var{number}.
815 @item --listing-lhs-width2=@var{number}
816 Set the maximum width, in words, of the output data column for continuation
817 lines in an assembler listing to @var{number}.
819 @item --listing-rhs-width=@var{number}
820 Set the maximum width of an input source line, as displayed in a listing, to
823 @item --listing-cont-lines=@var{number}
824 Set the maximum number of lines printed in a listing for a single line of input
827 @item --no-pad-sections
828 Stop the assembler for padding the ends of output sections to the alignment
829 of that section. The default is to pad the sections, but this can waste space
830 which might be needed on targets which have tight memory constraints.
832 @item -o @var{objfile}
833 Name the object-file output from @command{@value{AS}} @var{objfile}.
836 Fold the data section into the text section.
838 @item --hash-size=@var{number}
839 Set the default size of GAS's hash tables to a prime number close to
840 @var{number}. Increasing this value can reduce the length of time it takes the
841 assembler to perform its tasks, at the expense of increasing the assembler's
842 memory requirements. Similarly reducing this value can reduce the memory
843 requirements at the expense of speed.
845 @item --reduce-memory-overheads
846 This option reduces GAS's memory requirements, at the expense of making the
847 assembly processes slower. Currently this switch is a synonym for
848 @samp{--hash-size=4051}, but in the future it may have other effects as well.
851 @item --sectname-subst
852 Honor substitution sequences in section names.
854 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
859 Print the maximum space (in bytes) and total time (in seconds) used by
862 @item --strip-local-absolute
863 Remove local absolute symbols from the outgoing symbol table.
867 Print the @command{as} version.
870 Print the @command{as} version and exit.
874 Suppress warning messages.
876 @item --fatal-warnings
877 Treat warnings as errors.
880 Don't suppress warning messages or treat them as errors.
889 Generate an object file even after errors.
891 @item -- | @var{files} @dots{}
892 Standard input, or source files to assemble.
900 @xref{AArch64 Options}, for the options available when @value{AS} is configured
901 for the 64-bit mode of the ARM Architecture (AArch64).
906 The following options are available when @value{AS} is configured for the
907 64-bit mode of the ARM Architecture (AArch64).
910 @include c-aarch64.texi
911 @c ended inside the included file
919 @xref{Alpha Options}, for the options available when @value{AS} is configured
920 for an Alpha processor.
925 The following options are available when @value{AS} is configured for an Alpha
929 @include c-alpha.texi
930 @c ended inside the included file
937 The following options are available when @value{AS} is configured for an ARC
941 @item -mcpu=@var{cpu}
942 This option selects the core processor variant.
944 Select either big-endian (-EB) or little-endian (-EL) output.
946 Enable Code Density extenssion instructions.
951 The following options are available when @value{AS} is configured for the ARM
955 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
956 Specify which ARM processor variant is the target.
957 @item -march=@var{architecture}[+@var{extension}@dots{}]
958 Specify which ARM architecture variant is used by the target.
959 @item -mfpu=@var{floating-point-format}
960 Select which Floating Point architecture is the target.
961 @item -mfloat-abi=@var{abi}
962 Select which floating point ABI is in use.
964 Enable Thumb only instruction decoding.
965 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
966 Select which procedure calling convention is in use.
968 Select either big-endian (-EB) or little-endian (-EL) output.
969 @item -mthumb-interwork
970 Specify that the code has been generated with interworking between Thumb and
973 Turns on CodeComposer Studio assembly syntax compatibility mode.
975 Specify that PIC code has been generated.
983 @xref{Blackfin Options}, for the options available when @value{AS} is
984 configured for the Blackfin processor family.
989 The following options are available when @value{AS} is configured for
990 the Blackfin processor family.
994 @c ended inside the included file
1001 See the info pages for documentation of the CRIS-specific options.
1007 @xref{C-SKY Options}, for the options available when @value{AS} is
1008 configured for the C-SKY processor family.
1012 @c man begin OPTIONS
1013 The following options are available when @value{AS} is configured for
1014 the C-SKY processor family.
1016 @c man begin INCLUDE
1017 @include c-csky.texi
1018 @c ended inside the included file
1024 The following options are available when @value{AS} is configured for
1027 @cindex D10V optimization
1028 @cindex optimization, D10V
1030 Optimize output by parallelizing instructions.
1035 The following options are available when @value{AS} is configured for a D30V
1038 @cindex D30V optimization
1039 @cindex optimization, D30V
1041 Optimize output by parallelizing instructions.
1045 Warn when nops are generated.
1047 @cindex D30V nops after 32-bit multiply
1049 Warn when a nop after a 32-bit multiply instruction is generated.
1055 The following options are available when @value{AS} is configured for the
1056 Adapteva EPIPHANY series.
1059 @xref{Epiphany Options}, for the options available when @value{AS} is
1060 configured for an Epiphany processor.
1064 @c man begin OPTIONS
1065 The following options are available when @value{AS} is configured for
1066 an Epiphany processor.
1068 @c man begin INCLUDE
1069 @include c-epiphany.texi
1070 @c ended inside the included file
1078 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1079 for an H8/300 processor.
1083 @c man begin OPTIONS
1084 The following options are available when @value{AS} is configured for an H8/300
1087 @c man begin INCLUDE
1088 @include c-h8300.texi
1089 @c ended inside the included file
1097 @xref{i386-Options}, for the options available when @value{AS} is
1098 configured for an i386 processor.
1102 @c man begin OPTIONS
1103 The following options are available when @value{AS} is configured for
1106 @c man begin INCLUDE
1107 @include c-i386.texi
1108 @c ended inside the included file
1113 @c man begin OPTIONS
1115 The following options are available when @value{AS} is configured for the
1121 Specifies that the extended IP2022 instructions are allowed.
1124 Restores the default behaviour, which restricts the permitted instructions to
1125 just the basic IP2022 ones.
1131 The following options are available when @value{AS} is configured for the
1132 Renesas M32C and M16C processors.
1137 Assemble M32C instructions.
1140 Assemble M16C instructions (the default).
1143 Enable support for link-time relaxations.
1146 Support H'00 style hex constants in addition to 0x00 style.
1152 The following options are available when @value{AS} is configured for the
1153 Renesas M32R (formerly Mitsubishi M32R) series.
1158 Specify which processor in the M32R family is the target. The default
1159 is normally the M32R, but this option changes it to the M32RX.
1161 @item --warn-explicit-parallel-conflicts or --Wp
1162 Produce warning messages when questionable parallel constructs are
1165 @item --no-warn-explicit-parallel-conflicts or --Wnp
1166 Do not produce warning messages when questionable parallel constructs are
1173 The following options are available when @value{AS} is configured for the
1174 Motorola 68000 series.
1179 Shorten references to undefined symbols, to one word instead of two.
1181 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1182 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1183 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1184 Specify what processor in the 68000 family is the target. The default
1185 is normally the 68020, but this can be changed at configuration time.
1187 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1188 The target machine does (or does not) have a floating-point coprocessor.
1189 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1190 the basic 68000 is not compatible with the 68881, a combination of the
1191 two can be specified, since it's possible to do emulation of the
1192 coprocessor instructions with the main processor.
1194 @item -m68851 | -mno-68851
1195 The target machine does (or does not) have a memory-management
1196 unit coprocessor. The default is to assume an MMU for 68020 and up.
1204 @xref{Nios II Options}, for the options available when @value{AS} is configured
1205 for an Altera Nios II processor.
1209 @c man begin OPTIONS
1210 The following options are available when @value{AS} is configured for an
1211 Altera Nios II processor.
1213 @c man begin INCLUDE
1214 @include c-nios2.texi
1215 @c ended inside the included file
1221 For details about the PDP-11 machine dependent features options,
1222 see @ref{PDP-11-Options}.
1225 @item -mpic | -mno-pic
1226 Generate position-independent (or position-dependent) code. The
1227 default is @option{-mpic}.
1230 @itemx -mall-extensions
1231 Enable all instruction set extensions. This is the default.
1233 @item -mno-extensions
1234 Disable all instruction set extensions.
1236 @item -m@var{extension} | -mno-@var{extension}
1237 Enable (or disable) a particular instruction set extension.
1240 Enable the instruction set extensions supported by a particular CPU, and
1241 disable all other extensions.
1243 @item -m@var{machine}
1244 Enable the instruction set extensions supported by a particular machine
1245 model, and disable all other extensions.
1251 The following options are available when @value{AS} is configured for
1252 a picoJava processor.
1256 @cindex PJ endianness
1257 @cindex endianness, PJ
1258 @cindex big endian output, PJ
1260 Generate ``big endian'' format output.
1262 @cindex little endian output, PJ
1264 Generate ``little endian'' format output.
1272 @xref{PRU Options}, for the options available when @value{AS} is configured
1273 for a PRU processor.
1277 @c man begin OPTIONS
1278 The following options are available when @value{AS} is configured for a
1281 @c man begin INCLUDE
1283 @c ended inside the included file
1288 The following options are available when @value{AS} is configured for the
1289 Motorola 68HC11 or 68HC12 series.
1293 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1294 Specify what processor is the target. The default is
1295 defined by the configuration option when building the assembler.
1297 @item --xgate-ramoffset
1298 Instruct the linker to offset RAM addresses from S12X address space into
1299 XGATE address space.
1302 Specify to use the 16-bit integer ABI.
1305 Specify to use the 32-bit integer ABI.
1307 @item -mshort-double
1308 Specify to use the 32-bit double ABI.
1311 Specify to use the 64-bit double ABI.
1313 @item --force-long-branches
1314 Relative branches are turned into absolute ones. This concerns
1315 conditional branches, unconditional branches and branches to a
1318 @item -S | --short-branches
1319 Do not turn relative branches into absolute ones
1320 when the offset is out of range.
1322 @item --strict-direct-mode
1323 Do not turn the direct addressing mode into extended addressing mode
1324 when the instruction does not support direct addressing mode.
1326 @item --print-insn-syntax
1327 Print the syntax of instruction in case of error.
1329 @item --print-opcodes
1330 Print the list of instructions with syntax and then exit.
1332 @item --generate-example
1333 Print an example of instruction for each possible instruction and then exit.
1334 This option is only useful for testing @command{@value{AS}}.
1340 The following options are available when @command{@value{AS}} is configured
1341 for the SPARC architecture:
1344 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1345 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1346 Explicitly select a variant of the SPARC architecture.
1348 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1349 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1351 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1352 UltraSPARC extensions.
1354 @item -xarch=v8plus | -xarch=v8plusa
1355 For compatibility with the Solaris v9 assembler. These options are
1356 equivalent to -Av8plus and -Av8plusa, respectively.
1359 Warn when the assembler switches to another architecture.
1364 The following options are available when @value{AS} is configured for the 'c54x
1369 Enable extended addressing mode. All addresses and relocations will assume
1370 extended addressing (usually 23 bits).
1371 @item -mcpu=@var{CPU_VERSION}
1372 Sets the CPU version being compiled for.
1373 @item -merrors-to-file @var{FILENAME}
1374 Redirect error output to a file, for broken systems which don't support such
1375 behaviour in the shell.
1380 @c man begin OPTIONS
1381 The following options are available when @value{AS} is configured for
1386 This option sets the largest size of an object that can be referenced
1387 implicitly with the @code{gp} register. It is only accepted for targets that
1388 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1390 @cindex MIPS endianness
1391 @cindex endianness, MIPS
1392 @cindex big endian output, MIPS
1394 Generate ``big endian'' format output.
1396 @cindex little endian output, MIPS
1398 Generate ``little endian'' format output.
1416 Generate code for a particular MIPS Instruction Set Architecture level.
1417 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1418 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1419 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1420 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1421 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1422 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1423 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1424 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1425 MIPS64 Release 6 ISA processors, respectively.
1427 @item -march=@var{cpu}
1428 Generate code for a particular MIPS CPU.
1430 @item -mtune=@var{cpu}
1431 Schedule and tune for a particular MIPS CPU.
1435 Cause nops to be inserted if the read of the destination register
1436 of an mfhi or mflo instruction occurs in the following two instructions.
1439 @itemx -mno-fix-rm7000
1440 Cause nops to be inserted if a dmult or dmultu instruction is
1441 followed by a load instruction.
1445 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1446 section instead of the standard ELF .stabs sections.
1450 Control generation of @code{.pdr} sections.
1454 The register sizes are normally inferred from the ISA and ABI, but these
1455 flags force a certain group of registers to be treated as 32 bits wide at
1456 all times. @samp{-mgp32} controls the size of general-purpose registers
1457 and @samp{-mfp32} controls the size of floating-point registers.
1461 The register sizes are normally inferred from the ISA and ABI, but these
1462 flags force a certain group of registers to be treated as 64 bits wide at
1463 all times. @samp{-mgp64} controls the size of general-purpose registers
1464 and @samp{-mfp64} controls the size of floating-point registers.
1467 The register sizes are normally inferred from the ISA and ABI, but using
1468 this flag in combination with @samp{-mabi=32} enables an ABI variant
1469 which will operate correctly with floating-point registers which are
1473 @itemx -mno-odd-spreg
1474 Enable use of floating-point operations on odd-numbered single-precision
1475 registers when supported by the ISA. @samp{-mfpxx} implies
1476 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1480 Generate code for the MIPS 16 processor. This is equivalent to putting
1481 @code{.module mips16} at the start of the assembly file. @samp{-no-mips16}
1482 turns off this option.
1485 @itemx -mno-mips16e2
1486 Enable the use of MIPS16e2 instructions in MIPS16 mode. This is equivalent
1487 to putting @code{.module mips16e2} at the start of the assembly file.
1488 @samp{-mno-mips16e2} turns off this option.
1491 @itemx -mno-micromips
1492 Generate code for the microMIPS processor. This is equivalent to putting
1493 @code{.module micromips} at the start of the assembly file.
1494 @samp{-mno-micromips} turns off this option. This is equivalent to putting
1495 @code{.module nomicromips} at the start of the assembly file.
1498 @itemx -mno-smartmips
1499 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1500 equivalent to putting @code{.module smartmips} at the start of the assembly
1501 file. @samp{-mno-smartmips} turns off this option.
1505 Generate code for the MIPS-3D Application Specific Extension.
1506 This tells the assembler to accept MIPS-3D instructions.
1507 @samp{-no-mips3d} turns off this option.
1511 Generate code for the MDMX Application Specific Extension.
1512 This tells the assembler to accept MDMX instructions.
1513 @samp{-no-mdmx} turns off this option.
1517 Generate code for the DSP Release 1 Application Specific Extension.
1518 This tells the assembler to accept DSP Release 1 instructions.
1519 @samp{-mno-dsp} turns off this option.
1523 Generate code for the DSP Release 2 Application Specific Extension.
1524 This option implies @samp{-mdsp}.
1525 This tells the assembler to accept DSP Release 2 instructions.
1526 @samp{-mno-dspr2} turns off this option.
1530 Generate code for the DSP Release 3 Application Specific Extension.
1531 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1532 This tells the assembler to accept DSP Release 3 instructions.
1533 @samp{-mno-dspr3} turns off this option.
1537 Generate code for the MIPS SIMD Architecture Extension.
1538 This tells the assembler to accept MSA instructions.
1539 @samp{-mno-msa} turns off this option.
1543 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1544 This tells the assembler to accept XPA instructions.
1545 @samp{-mno-xpa} turns off this option.
1549 Generate code for the MT Application Specific Extension.
1550 This tells the assembler to accept MT instructions.
1551 @samp{-mno-mt} turns off this option.
1555 Generate code for the MCU Application Specific Extension.
1556 This tells the assembler to accept MCU instructions.
1557 @samp{-mno-mcu} turns off this option.
1561 Generate code for the MIPS cyclic redundancy check (CRC) Application
1562 Specific Extension. This tells the assembler to accept CRC instructions.
1563 @samp{-mno-crc} turns off this option.
1567 Generate code for the Global INValidate (GINV) Application Specific
1568 Extension. This tells the assembler to accept GINV instructions.
1569 @samp{-mno-ginv} turns off this option.
1571 @item -mloongson-mmi
1572 @itemx -mno-loongson-mmi
1573 Generate code for the Loongson MultiMedia extensions Instructions (MMI)
1574 Application Specific Extension. This tells the assembler to accept MMI
1576 @samp{-mno-loongson-mmi} turns off this option.
1578 @item -mloongson-cam
1579 @itemx -mno-loongson-cam
1580 Generate code for the Loongson Content Address Memory (CAM) instructions.
1581 This tells the assembler to accept Loongson CAM instructions.
1582 @samp{-mno-loongson-cam} turns off this option.
1584 @item -mloongson-ext
1585 @itemx -mno-loongson-ext
1586 Generate code for the Loongson EXTensions (EXT) instructions.
1587 This tells the assembler to accept Loongson EXT instructions.
1588 @samp{-mno-loongson-ext} turns off this option.
1592 Only use 32-bit instruction encodings when generating code for the
1593 microMIPS processor. This option inhibits the use of any 16-bit
1594 instructions. This is equivalent to putting @code{.set insn32} at
1595 the start of the assembly file. @samp{-mno-insn32} turns off this
1596 option. This is equivalent to putting @code{.set noinsn32} at the
1597 start of the assembly file. By default @samp{-mno-insn32} is
1598 selected, allowing all instructions to be used.
1600 @item --construct-floats
1601 @itemx --no-construct-floats
1602 The @samp{--no-construct-floats} option disables the construction of
1603 double width floating point constants by loading the two halves of the
1604 value into the two single width floating point registers that make up
1605 the double width register. By default @samp{--construct-floats} is
1606 selected, allowing construction of these floating point constants.
1608 @item --relax-branch
1609 @itemx --no-relax-branch
1610 The @samp{--relax-branch} option enables the relaxation of out-of-range
1611 branches. By default @samp{--no-relax-branch} is selected, causing any
1612 out-of-range branches to produce an error.
1614 @item -mignore-branch-isa
1615 @itemx -mno-ignore-branch-isa
1616 Ignore branch checks for invalid transitions between ISA modes. The
1617 semantics of branches does not provide for an ISA mode switch, so in
1618 most cases the ISA mode a branch has been encoded for has to be the
1619 same as the ISA mode of the branch's target label. Therefore GAS has
1620 checks implemented that verify in branch assembly that the two ISA
1621 modes match. @samp{-mignore-branch-isa} disables these checks. By
1622 default @samp{-mno-ignore-branch-isa} is selected, causing any invalid
1623 branch requiring a transition between ISA modes to produce an error.
1625 @item -mnan=@var{encoding}
1626 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1627 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1630 @item --emulation=@var{name}
1631 This option was formerly used to switch between ELF and ECOFF output
1632 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1633 removed in GAS 2.24, so the option now serves little purpose.
1634 It is retained for backwards compatibility.
1636 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1637 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1638 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1639 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1640 preferred options instead.
1643 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1650 Control how to deal with multiplication overflow and division by zero.
1651 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1652 (and only work for Instruction Set Architecture level 2 and higher);
1653 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1657 When this option is used, @command{@value{AS}} will issue a warning every
1658 time it generates a nop instruction from a macro.
1664 The following options are available when @value{AS} is configured for
1670 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1671 The command-line option @samp{-nojsri2bsr} can be used to disable it.
1675 Enable or disable the silicon filter behaviour. By default this is disabled.
1676 The default can be overridden by the @samp{-sifilter} command-line option.
1679 Alter jump instructions for long displacements.
1681 @item -mcpu=[210|340]
1682 Select the cpu type on the target hardware. This controls which instructions
1686 Assemble for a big endian target.
1689 Assemble for a little endian target.
1698 @xref{Meta Options}, for the options available when @value{AS} is configured
1699 for a Meta processor.
1703 @c man begin OPTIONS
1704 The following options are available when @value{AS} is configured for a
1707 @c man begin INCLUDE
1708 @include c-metag.texi
1709 @c ended inside the included file
1714 @c man begin OPTIONS
1716 See the info pages for documentation of the MMIX-specific options.
1722 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1723 for a NDS32 processor.
1725 @c ended inside the included file
1729 @c man begin OPTIONS
1730 The following options are available when @value{AS} is configured for a
1733 @c man begin INCLUDE
1734 @include c-nds32.texi
1735 @c ended inside the included file
1742 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1743 for a PowerPC processor.
1747 @c man begin OPTIONS
1748 The following options are available when @value{AS} is configured for a
1751 @c man begin INCLUDE
1753 @c ended inside the included file
1761 @xref{RISC-V-Options}, for the options available when @value{AS} is configured
1762 for a RISC-V processor.
1766 @c man begin OPTIONS
1767 The following options are available when @value{AS} is configured for a
1770 @c man begin INCLUDE
1771 @include c-riscv.texi
1772 @c ended inside the included file
1777 @c man begin OPTIONS
1779 See the info pages for documentation of the RX-specific options.
1783 The following options are available when @value{AS} is configured for the s390
1789 Select the word size, either 31/32 bits or 64 bits.
1792 Select the architecture mode, either the Enterprise System
1793 Architecture (esa) or the z/Architecture mode (zarch).
1794 @item -march=@var{processor}
1795 Specify which s390 processor variant is the target, @samp{g5} (or
1796 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1797 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1798 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1799 @samp{z13} (or @samp{arch11}), or @samp{z14} (or @samp{arch12}).
1801 @itemx -mno-regnames
1802 Allow or disallow symbolic names for registers.
1803 @item -mwarn-areg-zero
1804 Warn whenever the operand for a base or index register has been specified
1805 but evaluates to zero.
1813 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1814 for a TMS320C6000 processor.
1818 @c man begin OPTIONS
1819 The following options are available when @value{AS} is configured for a
1820 TMS320C6000 processor.
1822 @c man begin INCLUDE
1823 @include c-tic6x.texi
1824 @c ended inside the included file
1832 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1833 for a TILE-Gx processor.
1837 @c man begin OPTIONS
1838 The following options are available when @value{AS} is configured for a TILE-Gx
1841 @c man begin INCLUDE
1842 @include c-tilegx.texi
1843 @c ended inside the included file
1851 @xref{Visium Options}, for the options available when @value{AS} is configured
1852 for a Visium processor.
1856 @c man begin OPTIONS
1857 The following option is available when @value{AS} is configured for a Visium
1860 @c man begin INCLUDE
1861 @include c-visium.texi
1862 @c ended inside the included file
1870 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1871 for an Xtensa processor.
1875 @c man begin OPTIONS
1876 The following options are available when @value{AS} is configured for an
1879 @c man begin INCLUDE
1880 @include c-xtensa.texi
1881 @c ended inside the included file
1886 @c man begin OPTIONS
1889 The following options are available when @value{AS} is configured for
1890 a Z80 family processor.
1893 Assemble for Z80 processor.
1895 Assemble for R800 processor.
1896 @item -ignore-undocumented-instructions
1898 Assemble undocumented Z80 instructions that also work on R800 without warning.
1899 @item -ignore-unportable-instructions
1901 Assemble all undocumented Z80 instructions without warning.
1902 @item -warn-undocumented-instructions
1904 Issue a warning for undocumented Z80 instructions that also work on R800.
1905 @item -warn-unportable-instructions
1907 Issue a warning for undocumented Z80 instructions that do not work on R800.
1908 @item -forbid-undocumented-instructions
1910 Treat all undocumented instructions as errors.
1911 @item -forbid-unportable-instructions
1913 Treat undocumented Z80 instructions that do not work on R800 as errors.
1920 * Manual:: Structure of this Manual
1921 * GNU Assembler:: The GNU Assembler
1922 * Object Formats:: Object File Formats
1923 * Command Line:: Command Line
1924 * Input Files:: Input Files
1925 * Object:: Output (Object) File
1926 * Errors:: Error and Warning Messages
1930 @section Structure of this Manual
1932 @cindex manual, structure and purpose
1933 This manual is intended to describe what you need to know to use
1934 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1935 notation for symbols, constants, and expressions; the directives that
1936 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1939 We also cover special features in the @value{TARGET}
1940 configuration of @command{@value{AS}}, including assembler directives.
1943 This manual also describes some of the machine-dependent features of
1944 various flavors of the assembler.
1947 @cindex machine instructions (not covered)
1948 On the other hand, this manual is @emph{not} intended as an introduction
1949 to programming in assembly language---let alone programming in general!
1950 In a similar vein, we make no attempt to introduce the machine
1951 architecture; we do @emph{not} describe the instruction set, standard
1952 mnemonics, registers or addressing modes that are standard to a
1953 particular architecture.
1955 You may want to consult the manufacturer's
1956 machine architecture manual for this information.
1960 For information on the H8/300 machine instruction set, see @cite{H8/300
1961 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1962 Programming Manual} (Renesas).
1965 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1966 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1967 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1968 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1971 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1975 @c I think this is premature---doc@cygnus.com, 17jan1991
1977 Throughout this manual, we assume that you are running @dfn{GNU},
1978 the portable operating system from the @dfn{Free Software
1979 Foundation, Inc.}. This restricts our attention to certain kinds of
1980 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1981 once this assumption is granted examples and definitions need less
1984 @command{@value{AS}} is part of a team of programs that turn a high-level
1985 human-readable series of instructions into a low-level
1986 computer-readable series of instructions. Different versions of
1987 @command{@value{AS}} are used for different kinds of computer.
1990 @c There used to be a section "Terminology" here, which defined
1991 @c "contents", "byte", "word", and "long". Defining "word" to any
1992 @c particular size is confusing when the .word directive may generate 16
1993 @c bits on one machine and 32 bits on another; in general, for the user
1994 @c version of this manual, none of these terms seem essential to define.
1995 @c They were used very little even in the former draft of the manual;
1996 @c this draft makes an effort to avoid them (except in names of
2000 @section The GNU Assembler
2002 @c man begin DESCRIPTION
2004 @sc{gnu} @command{as} is really a family of assemblers.
2006 This manual describes @command{@value{AS}}, a member of that family which is
2007 configured for the @value{TARGET} architectures.
2009 If you use (or have used) the @sc{gnu} assembler on one architecture, you
2010 should find a fairly similar environment when you use it on another
2011 architecture. Each version has much in common with the others,
2012 including object file formats, most assembler directives (often called
2013 @dfn{pseudo-ops}) and assembler syntax.@refill
2015 @cindex purpose of @sc{gnu} assembler
2016 @command{@value{AS}} is primarily intended to assemble the output of the
2017 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
2018 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
2019 assemble correctly everything that other assemblers for the same
2020 machine would assemble.
2022 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
2025 @c This remark should appear in generic version of manual; assumption
2026 @c here is that generic version sets M680x0.
2027 This doesn't mean @command{@value{AS}} always uses the same syntax as another
2028 assembler for the same architecture; for example, we know of several
2029 incompatible versions of 680x0 assembly language syntax.
2034 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
2035 program in one pass of the source file. This has a subtle impact on the
2036 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
2038 @node Object Formats
2039 @section Object File Formats
2041 @cindex object file format
2042 The @sc{gnu} assembler can be configured to produce several alternative
2043 object file formats. For the most part, this does not affect how you
2044 write assembly language programs; but directives for debugging symbols
2045 are typically different in different file formats. @xref{Symbol
2046 Attributes,,Symbol Attributes}.
2049 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
2050 @value{OBJ-NAME} format object files.
2052 @c The following should exhaust all configs that set MULTI-OBJ, ideally
2054 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
2055 SOM or ELF format object files.
2060 @section Command Line
2062 @cindex command line conventions
2064 After the program name @command{@value{AS}}, the command line may contain
2065 options and file names. Options may appear in any order, and may be
2066 before, after, or between file names. The order of file names is
2069 @cindex standard input, as input file
2071 @file{--} (two hyphens) by itself names the standard input file
2072 explicitly, as one of the files for @command{@value{AS}} to assemble.
2074 @cindex options, command line
2075 Except for @samp{--} any command-line argument that begins with a
2076 hyphen (@samp{-}) is an option. Each option changes the behavior of
2077 @command{@value{AS}}. No option changes the way another option works. An
2078 option is a @samp{-} followed by one or more letters; the case of
2079 the letter is important. All options are optional.
2081 Some options expect exactly one file name to follow them. The file
2082 name may either immediately follow the option's letter (compatible
2083 with older assemblers) or it may be the next command argument (@sc{gnu}
2084 standard). These two command lines are equivalent:
2087 @value{AS} -o my-object-file.o mumble.s
2088 @value{AS} -omy-object-file.o mumble.s
2092 @section Input Files
2095 @cindex source program
2096 @cindex files, input
2097 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2098 describe the program input to one run of @command{@value{AS}}. The program may
2099 be in one or more files; how the source is partitioned into files
2100 doesn't change the meaning of the source.
2102 @c I added "con" prefix to "catenation" just to prove I can overcome my
2103 @c APL training... doc@cygnus.com
2104 The source program is a concatenation of the text in all the files, in the
2107 @c man begin DESCRIPTION
2108 Each time you run @command{@value{AS}} it assembles exactly one source
2109 program. The source program is made up of one or more files.
2110 (The standard input is also a file.)
2112 You give @command{@value{AS}} a command line that has zero or more input file
2113 names. The input files are read (from left file name to right). A
2114 command-line argument (in any position) that has no special meaning
2115 is taken to be an input file name.
2117 If you give @command{@value{AS}} no file names it attempts to read one input file
2118 from the @command{@value{AS}} standard input, which is normally your terminal. You
2119 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2122 Use @samp{--} if you need to explicitly name the standard input file
2123 in your command line.
2125 If the source is empty, @command{@value{AS}} produces a small, empty object
2130 @subheading Filenames and Line-numbers
2132 @cindex input file linenumbers
2133 @cindex line numbers, in input files
2134 There are two ways of locating a line in the input file (or files) and
2135 either may be used in reporting error messages. One way refers to a line
2136 number in a physical file; the other refers to a line number in a
2137 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2139 @dfn{Physical files} are those files named in the command line given
2140 to @command{@value{AS}}.
2142 @dfn{Logical files} are simply names declared explicitly by assembler
2143 directives; they bear no relation to physical files. Logical file names help
2144 error messages reflect the original source file, when @command{@value{AS}} source
2145 is itself synthesized from other files. @command{@value{AS}} understands the
2146 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2147 @ref{File,,@code{.file}}.
2150 @section Output (Object) File
2156 Every time you run @command{@value{AS}} it produces an output file, which is
2157 your assembly language program translated into numbers. This file
2158 is the object file. Its default name is @code{a.out}.
2159 You can give it another name by using the @option{-o} option. Conventionally,
2160 object file names end with @file{.o}. The default name is used for historical
2161 reasons: older assemblers were capable of assembling self-contained programs
2162 directly into a runnable program. (For some formats, this isn't currently
2163 possible, but it can be done for the @code{a.out} format.)
2167 The object file is meant for input to the linker @code{@value{LD}}. It contains
2168 assembled program code, information to help @code{@value{LD}} integrate
2169 the assembled program into a runnable file, and (optionally) symbolic
2170 information for the debugger.
2172 @c link above to some info file(s) like the description of a.out.
2173 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2176 @section Error and Warning Messages
2178 @c man begin DESCRIPTION
2180 @cindex error messages
2181 @cindex warning messages
2182 @cindex messages from assembler
2183 @command{@value{AS}} may write warnings and error messages to the standard error
2184 file (usually your terminal). This should not happen when a compiler
2185 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2186 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2187 grave problem that stops the assembly.
2191 @cindex format of warning messages
2192 Warning messages have the format
2195 file_name:@b{NNN}:Warning Message Text
2199 @cindex file names and line numbers, in warnings/errors
2200 (where @b{NNN} is a line number). If both a logical file name
2201 (@pxref{File,,@code{.file}}) and a logical line number
2203 (@pxref{Line,,@code{.line}})
2205 have been given then they will be used, otherwise the file name and line number
2206 in the current assembler source file will be used. The message text is
2207 intended to be self explanatory (in the grand Unix tradition).
2209 Note the file name must be set via the logical version of the @code{.file}
2210 directive, not the DWARF2 version of the @code{.file} directive. For example:
2214 error_assembler_source
2220 produces this output:
2224 asm.s:2: Error: no such instruction: `error_assembler_source'
2225 foo.c:31: Error: no such instruction: `error_c_source'
2228 @cindex format of error messages
2229 Error messages have the format
2232 file_name:@b{NNN}:FATAL:Error Message Text
2235 The file name and line number are derived as for warning
2236 messages. The actual message text may be rather less explanatory
2237 because many of them aren't supposed to happen.
2240 @chapter Command-Line Options
2242 @cindex options, all versions of assembler
2243 This chapter describes command-line options available in @emph{all}
2244 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2245 for options specific
2247 to the @value{TARGET} target.
2250 to particular machine architectures.
2253 @c man begin DESCRIPTION
2255 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2256 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2257 The assembler arguments must be separated from each other (and the @samp{-Wa})
2258 by commas. For example:
2261 gcc -c -g -O -Wa,-alh,-L file.c
2265 This passes two options to the assembler: @samp{-alh} (emit a listing to
2266 standard output with high-level and assembly source) and @samp{-L} (retain
2267 local symbols in the symbol table).
2269 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2270 command-line options are automatically passed to the assembler by the compiler.
2271 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2272 precisely what options it passes to each compilation pass, including the
2278 * a:: -a[cdghlns] enable listings
2279 * alternate:: --alternate enable alternate macro syntax
2280 * D:: -D for compatibility
2281 * f:: -f to work faster
2282 * I:: -I for .include search path
2283 @ifclear DIFF-TBL-KLUGE
2284 * K:: -K for compatibility
2286 @ifset DIFF-TBL-KLUGE
2287 * K:: -K for difference tables
2290 * L:: -L to retain local symbols
2291 * listing:: --listing-XXX to configure listing output
2292 * M:: -M or --mri to assemble in MRI compatibility mode
2293 * MD:: --MD for dependency tracking
2294 * no-pad-sections:: --no-pad-sections to stop section padding
2295 * o:: -o to name the object file
2296 * R:: -R to join data and text sections
2297 * statistics:: --statistics to see statistics about assembly
2298 * traditional-format:: --traditional-format for compatible output
2299 * v:: -v to announce version
2300 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2301 * Z:: -Z to make object file even after errors
2305 @section Enable Listings: @option{-a[cdghlns]}
2315 @cindex listings, enabling
2316 @cindex assembly listings, enabling
2318 These options enable listing output from the assembler. By itself,
2319 @samp{-a} requests high-level, assembly, and symbols listing.
2320 You can use other letters to select specific options for the list:
2321 @samp{-ah} requests a high-level language listing,
2322 @samp{-al} requests an output-program assembly listing, and
2323 @samp{-as} requests a symbol table listing.
2324 High-level listings require that a compiler debugging option like
2325 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2328 Use the @samp{-ag} option to print a first section with general assembly
2329 information, like @value{AS} version, switches passed, or time stamp.
2331 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2332 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2333 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2334 omitted from the listing.
2336 Use the @samp{-ad} option to omit debugging directives from the
2339 Once you have specified one of these options, you can further control
2340 listing output and its appearance using the directives @code{.list},
2341 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2343 The @samp{-an} option turns off all forms processing.
2344 If you do not request listing output with one of the @samp{-a} options, the
2345 listing-control directives have no effect.
2347 The letters after @samp{-a} may be combined into one option,
2348 @emph{e.g.}, @samp{-aln}.
2350 Note if the assembler source is coming from the standard input (e.g.,
2352 is being created by @code{@value{GCC}} and the @samp{-pipe} command-line switch
2353 is being used) then the listing will not contain any comments or preprocessor
2354 directives. This is because the listing code buffers input source lines from
2355 stdin only after they have been preprocessed by the assembler. This reduces
2356 memory usage and makes the code more efficient.
2359 @section @option{--alternate}
2362 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2365 @section @option{-D}
2368 This option has no effect whatsoever, but it is accepted to make it more
2369 likely that scripts written for other assemblers also work with
2370 @command{@value{AS}}.
2373 @section Work Faster: @option{-f}
2376 @cindex trusted compiler
2377 @cindex faster processing (@option{-f})
2378 @samp{-f} should only be used when assembling programs written by a
2379 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2380 and comment preprocessing on
2381 the input file(s) before assembling them. @xref{Preprocessing,
2385 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2386 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2391 @section @code{.include} Search Path: @option{-I} @var{path}
2393 @kindex -I @var{path}
2394 @cindex paths for @code{.include}
2395 @cindex search path for @code{.include}
2396 @cindex @code{include} directive search path
2397 Use this option to add a @var{path} to the list of directories
2398 @command{@value{AS}} searches for files specified in @code{.include}
2399 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2400 many times as necessary to include a variety of paths. The current
2401 working directory is always searched first; after that, @command{@value{AS}}
2402 searches any @samp{-I} directories in the same order as they were
2403 specified (left to right) on the command line.
2406 @section Difference Tables: @option{-K}
2409 @ifclear DIFF-TBL-KLUGE
2410 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2411 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2412 where it can be used to warn when the assembler alters the machine code
2413 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2414 family does not have the addressing limitations that sometimes lead to this
2415 alteration on other platforms.
2418 @ifset DIFF-TBL-KLUGE
2419 @cindex difference tables, warning
2420 @cindex warning for altered difference tables
2421 @command{@value{AS}} sometimes alters the code emitted for directives of the
2422 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2423 You can use the @samp{-K} option if you want a warning issued when this
2428 @section Include Local Symbols: @option{-L}
2431 @cindex local symbols, retaining in output
2432 Symbols beginning with system-specific local label prefixes, typically
2433 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2434 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2435 such symbols when debugging, because they are intended for the use of
2436 programs (like compilers) that compose assembler programs, not for your
2437 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2438 such symbols, so you do not normally debug with them.
2440 This option tells @command{@value{AS}} to retain those local symbols
2441 in the object file. Usually if you do this you also tell the linker
2442 @code{@value{LD}} to preserve those symbols.
2445 @section Configuring listing output: @option{--listing}
2447 The listing feature of the assembler can be enabled via the command-line switch
2448 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2449 hex dump of the corresponding locations in the output object file, and displays
2450 them as a listing file. The format of this listing can be controlled by
2451 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2452 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2453 @code{.psize} (@pxref{Psize}), and
2454 @code{.eject} (@pxref{Eject}) and also by the following switches:
2457 @item --listing-lhs-width=@samp{number}
2458 @kindex --listing-lhs-width
2459 @cindex Width of first line disassembly output
2460 Sets the maximum width, in words, of the first line of the hex byte dump. This
2461 dump appears on the left hand side of the listing output.
2463 @item --listing-lhs-width2=@samp{number}
2464 @kindex --listing-lhs-width2
2465 @cindex Width of continuation lines of disassembly output
2466 Sets the maximum width, in words, of any further lines of the hex byte dump for
2467 a given input source line. If this value is not specified, it defaults to being
2468 the same as the value specified for @samp{--listing-lhs-width}. If neither
2469 switch is used the default is to one.
2471 @item --listing-rhs-width=@samp{number}
2472 @kindex --listing-rhs-width
2473 @cindex Width of source line output
2474 Sets the maximum width, in characters, of the source line that is displayed
2475 alongside the hex dump. The default value for this parameter is 100. The
2476 source line is displayed on the right hand side of the listing output.
2478 @item --listing-cont-lines=@samp{number}
2479 @kindex --listing-cont-lines
2480 @cindex Maximum number of continuation lines
2481 Sets the maximum number of continuation lines of hex dump that will be
2482 displayed for a given single line of source input. The default value is 4.
2486 @section Assemble in MRI Compatibility Mode: @option{-M}
2489 @cindex MRI compatibility mode
2490 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2491 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2492 compatible with the @code{ASM68K} assembler from Microtec Research.
2493 The exact nature of the
2494 MRI syntax will not be documented here; see the MRI manuals for more
2495 information. Note in particular that the handling of macros and macro
2496 arguments is somewhat different. The purpose of this option is to permit
2497 assembling existing MRI assembler code using @command{@value{AS}}.
2499 The MRI compatibility is not complete. Certain operations of the MRI assembler
2500 depend upon its object file format, and can not be supported using other object
2501 file formats. Supporting these would require enhancing each object file format
2502 individually. These are:
2505 @item global symbols in common section
2507 The m68k MRI assembler supports common sections which are merged by the linker.
2508 Other object file formats do not support this. @command{@value{AS}} handles
2509 common sections by treating them as a single common symbol. It permits local
2510 symbols to be defined within a common section, but it can not support global
2511 symbols, since it has no way to describe them.
2513 @item complex relocations
2515 The MRI assemblers support relocations against a negated section address, and
2516 relocations which combine the start addresses of two or more sections. These
2517 are not support by other object file formats.
2519 @item @code{END} pseudo-op specifying start address
2521 The MRI @code{END} pseudo-op permits the specification of a start address.
2522 This is not supported by other object file formats. The start address may
2523 instead be specified using the @option{-e} option to the linker, or in a linker
2526 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2528 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2529 name to the output file. This is not supported by other object file formats.
2531 @item @code{ORG} pseudo-op
2533 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2534 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2535 which changes the location within the current section. Absolute sections are
2536 not supported by other object file formats. The address of a section may be
2537 assigned within a linker script.
2540 There are some other features of the MRI assembler which are not supported by
2541 @command{@value{AS}}, typically either because they are difficult or because they
2542 seem of little consequence. Some of these may be supported in future releases.
2546 @item EBCDIC strings
2548 EBCDIC strings are not supported.
2550 @item packed binary coded decimal
2552 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2553 and @code{DCB.P} pseudo-ops are not supported.
2555 @item @code{FEQU} pseudo-op
2557 The m68k @code{FEQU} pseudo-op is not supported.
2559 @item @code{NOOBJ} pseudo-op
2561 The m68k @code{NOOBJ} pseudo-op is not supported.
2563 @item @code{OPT} branch control options
2565 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2566 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2567 relaxes all branches, whether forward or backward, to an appropriate size, so
2568 these options serve no purpose.
2570 @item @code{OPT} list control options
2572 The following m68k @code{OPT} list control options are ignored: @code{C},
2573 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2574 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2576 @item other @code{OPT} options
2578 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2579 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2581 @item @code{OPT} @code{D} option is default
2583 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2584 @code{OPT NOD} may be used to turn it off.
2586 @item @code{XREF} pseudo-op.
2588 The m68k @code{XREF} pseudo-op is ignored.
2593 @section Dependency Tracking: @option{--MD}
2596 @cindex dependency tracking
2599 @command{@value{AS}} can generate a dependency file for the file it creates. This
2600 file consists of a single rule suitable for @code{make} describing the
2601 dependencies of the main source file.
2603 The rule is written to the file named in its argument.
2605 This feature is used in the automatic updating of makefiles.
2607 @node no-pad-sections
2608 @section Output Section Padding
2609 @kindex --no-pad-sections
2610 @cindex output section padding
2611 Normally the assembler will pad the end of each output section up to its
2612 alignment boundary. But this can waste space, which can be significant on
2613 memory constrained targets. So the @option{--no-pad-sections} option will
2614 disable this behaviour.
2617 @section Name the Object File: @option{-o}
2620 @cindex naming object file
2621 @cindex object file name
2622 There is always one object file output when you run @command{@value{AS}}. By
2623 default it has the name @file{a.out}.
2624 You use this option (which takes exactly one filename) to give the
2625 object file a different name.
2627 Whatever the object file is called, @command{@value{AS}} overwrites any
2628 existing file of the same name.
2631 @section Join Data and Text Sections: @option{-R}
2634 @cindex data and text sections, joining
2635 @cindex text and data sections, joining
2636 @cindex joining text and data sections
2637 @cindex merging text and data sections
2638 @option{-R} tells @command{@value{AS}} to write the object file as if all
2639 data-section data lives in the text section. This is only done at
2640 the very last moment: your binary data are the same, but data
2641 section parts are relocated differently. The data section part of
2642 your object file is zero bytes long because all its bytes are
2643 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2645 When you specify @option{-R} it would be possible to generate shorter
2646 address displacements (because we do not have to cross between text and
2647 data section). We refrain from doing this simply for compatibility with
2648 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2651 When @command{@value{AS}} is configured for COFF or ELF output,
2652 this option is only useful if you use sections named @samp{.text} and
2657 @option{-R} is not supported for any of the HPPA targets. Using
2658 @option{-R} generates a warning from @command{@value{AS}}.
2662 @section Display Assembly Statistics: @option{--statistics}
2664 @kindex --statistics
2665 @cindex statistics, about assembly
2666 @cindex time, total for assembly
2667 @cindex space used, maximum for assembly
2668 Use @samp{--statistics} to display two statistics about the resources used by
2669 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2670 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2673 @node traditional-format
2674 @section Compatible Output: @option{--traditional-format}
2676 @kindex --traditional-format
2677 For some targets, the output of @command{@value{AS}} is different in some ways
2678 from the output of some existing assembler. This switch requests
2679 @command{@value{AS}} to use the traditional format instead.
2681 For example, it disables the exception frame optimizations which
2682 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2685 @section Announce Version: @option{-v}
2689 @cindex assembler version
2690 @cindex version of assembler
2691 You can find out what version of as is running by including the
2692 option @samp{-v} (which you can also spell as @samp{-version}) on the
2696 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2698 @command{@value{AS}} should never give a warning or error message when
2699 assembling compiler output. But programs written by people often
2700 cause @command{@value{AS}} to give a warning that a particular assumption was
2701 made. All such warnings are directed to the standard error file.
2705 @cindex suppressing warnings
2706 @cindex warnings, suppressing
2707 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2708 This only affects the warning messages: it does not change any particular of
2709 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2712 @kindex --fatal-warnings
2713 @cindex errors, caused by warnings
2714 @cindex warnings, causing error
2715 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2716 files that generate warnings to be in error.
2719 @cindex warnings, switching on
2720 You can switch these options off again by specifying @option{--warn}, which
2721 causes warnings to be output as usual.
2724 @section Generate Object File in Spite of Errors: @option{-Z}
2725 @cindex object file, after errors
2726 @cindex errors, continuing after
2727 After an error message, @command{@value{AS}} normally produces no output. If for
2728 some reason you are interested in object file output even after
2729 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2730 option. If there are any errors, @command{@value{AS}} continues anyways, and
2731 writes an object file after a final warning message of the form @samp{@var{n}
2732 errors, @var{m} warnings, generating bad object file.}
2737 @cindex machine-independent syntax
2738 @cindex syntax, machine-independent
2739 This chapter describes the machine-independent syntax allowed in a
2740 source file. @command{@value{AS}} syntax is similar to what many other
2741 assemblers use; it is inspired by the BSD 4.2
2746 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2750 * Preprocessing:: Preprocessing
2751 * Whitespace:: Whitespace
2752 * Comments:: Comments
2753 * Symbol Intro:: Symbols
2754 * Statements:: Statements
2755 * Constants:: Constants
2759 @section Preprocessing
2761 @cindex preprocessing
2762 The @command{@value{AS}} internal preprocessor:
2764 @cindex whitespace, removed by preprocessor
2766 adjusts and removes extra whitespace. It leaves one space or tab before
2767 the keywords on a line, and turns any other whitespace on the line into
2770 @cindex comments, removed by preprocessor
2772 removes all comments, replacing them with a single space, or an
2773 appropriate number of newlines.
2775 @cindex constants, converted by preprocessor
2777 converts character constants into the appropriate numeric values.
2780 It does not do macro processing, include file handling, or
2781 anything else you may get from your C compiler's preprocessor. You can
2782 do include file processing with the @code{.include} directive
2783 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2784 to get other ``CPP'' style preprocessing by giving the input file a
2785 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2786 Output, gcc info, Using GNU CC}.
2788 Excess whitespace, comments, and character constants
2789 cannot be used in the portions of the input text that are not
2792 @cindex turning preprocessing on and off
2793 @cindex preprocessing, turning on and off
2796 If the first line of an input file is @code{#NO_APP} or if you use the
2797 @samp{-f} option, whitespace and comments are not removed from the input file.
2798 Within an input file, you can ask for whitespace and comment removal in
2799 specific portions of the by putting a line that says @code{#APP} before the
2800 text that may contain whitespace or comments, and putting a line that says
2801 @code{#NO_APP} after this text. This feature is mainly intend to support
2802 @code{asm} statements in compilers whose output is otherwise free of comments
2809 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2810 Whitespace is used to separate symbols, and to make programs neater for
2811 people to read. Unless within character constants
2812 (@pxref{Characters,,Character Constants}), any whitespace means the same
2813 as exactly one space.
2819 There are two ways of rendering comments to @command{@value{AS}}. In both
2820 cases the comment is equivalent to one space.
2822 Anything from @samp{/*} through the next @samp{*/} is a comment.
2823 This means you may not nest these comments.
2827 The only way to include a newline ('\n') in a comment
2828 is to use this sort of comment.
2831 /* This sort of comment does not nest. */
2834 @cindex line comment character
2835 Anything from a @dfn{line comment} character up to the next newline is
2836 considered a comment and is ignored. The line comment character is target
2837 specific, and some targets multiple comment characters. Some targets also have
2838 line comment characters that only work if they are the first character on a
2839 line. Some targets use a sequence of two characters to introduce a line
2840 comment. Some targets can also change their line comment characters depending
2841 upon command-line options that have been used. For more details see the
2842 @emph{Syntax} section in the documentation for individual targets.
2844 If the line comment character is the hash sign (@samp{#}) then it still has the
2845 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2846 to specify logical line numbers:
2849 @cindex lines starting with @code{#}
2850 @cindex logical line numbers
2851 To be compatible with past assemblers, lines that begin with @samp{#} have a
2852 special interpretation. Following the @samp{#} should be an absolute
2853 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2854 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2855 new logical file name. The rest of the line, if any, should be whitespace.
2857 If the first non-whitespace characters on the line are not numeric,
2858 the line is ignored. (Just like a comment.)
2861 # This is an ordinary comment.
2862 # 42-6 "new_file_name" # New logical file name
2863 # This is logical line # 36.
2865 This feature is deprecated, and may disappear from future versions
2866 of @command{@value{AS}}.
2871 @cindex characters used in symbols
2872 @ifclear SPECIAL-SYMS
2873 A @dfn{symbol} is one or more characters chosen from the set of all
2874 letters (both upper and lower case), digits and the three characters
2880 A @dfn{symbol} is one or more characters chosen from the set of all
2881 letters (both upper and lower case), digits and the three characters
2882 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2888 On most machines, you can also use @code{$} in symbol names; exceptions
2889 are noted in @ref{Machine Dependencies}.
2891 No symbol may begin with a digit. Case is significant.
2892 There is no length limit; all characters are significant. Multibyte characters
2893 are supported. Symbols are delimited by characters not in that set, or by the
2894 beginning of a file (since the source program must end with a newline, the end
2895 of a file is not a possible symbol delimiter). @xref{Symbols}.
2897 Symbol names may also be enclosed in double quote @code{"} characters. In such
2898 cases any characters are allowed, except for the NUL character. If a double
2899 quote character is to be included in the symbol name it must be preceeded by a
2900 backslash @code{\} character.
2901 @cindex length of symbols
2906 @cindex statements, structure of
2907 @cindex line separator character
2908 @cindex statement separator character
2910 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2911 @dfn{line separator character}. The line separator character is target
2912 specific and described in the @emph{Syntax} section of each
2913 target's documentation. Not all targets support a line separator character.
2914 The newline or line separator character is considered to be part of the
2915 preceding statement. Newlines and separators within character constants are an
2916 exception: they do not end statements.
2918 @cindex newline, required at file end
2919 @cindex EOF, newline must precede
2920 It is an error to end any statement with end-of-file: the last
2921 character of any input file should be a newline.@refill
2923 An empty statement is allowed, and may include whitespace. It is ignored.
2925 @cindex instructions and directives
2926 @cindex directives and instructions
2927 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2928 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2930 A statement begins with zero or more labels, optionally followed by a
2931 key symbol which determines what kind of statement it is. The key
2932 symbol determines the syntax of the rest of the statement. If the
2933 symbol begins with a dot @samp{.} then the statement is an assembler
2934 directive: typically valid for any computer. If the symbol begins with
2935 a letter the statement is an assembly language @dfn{instruction}: it
2936 assembles into a machine language instruction.
2938 Different versions of @command{@value{AS}} for different computers
2939 recognize different instructions. In fact, the same symbol may
2940 represent a different instruction in a different computer's assembly
2944 @cindex @code{:} (label)
2945 @cindex label (@code{:})
2946 A label is a symbol immediately followed by a colon (@code{:}).
2947 Whitespace before a label or after a colon is permitted, but you may not
2948 have whitespace between a label's symbol and its colon. @xref{Labels}.
2951 For HPPA targets, labels need not be immediately followed by a colon, but
2952 the definition of a label must begin in column zero. This also implies that
2953 only one label may be defined on each line.
2957 label: .directive followed by something
2958 another_label: # This is an empty statement.
2959 instruction operand_1, operand_2, @dots{}
2966 A constant is a number, written so that its value is known by
2967 inspection, without knowing any context. Like this:
2970 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2971 .ascii "Ring the bell\7" # A string constant.
2972 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2973 .float 0f-314159265358979323846264338327\
2974 95028841971.693993751E-40 # - pi, a flonum.
2979 * Characters:: Character Constants
2980 * Numbers:: Number Constants
2984 @subsection Character Constants
2986 @cindex character constants
2987 @cindex constants, character
2988 There are two kinds of character constants. A @dfn{character} stands
2989 for one character in one byte and its value may be used in
2990 numeric expressions. String constants (properly called string
2991 @emph{literals}) are potentially many bytes and their values may not be
2992 used in arithmetic expressions.
2996 * Chars:: Characters
3000 @subsubsection Strings
3002 @cindex string constants
3003 @cindex constants, string
3004 A @dfn{string} is written between double-quotes. It may contain
3005 double-quotes or null characters. The way to get special characters
3006 into a string is to @dfn{escape} these characters: precede them with
3007 a backslash @samp{\} character. For example @samp{\\} represents
3008 one backslash: the first @code{\} is an escape which tells
3009 @command{@value{AS}} to interpret the second character literally as a backslash
3010 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
3011 escape character). The complete list of escapes follows.
3013 @cindex escape codes, character
3014 @cindex character escape codes
3015 @c NOTE: Cindex entries must not start with a backlash character.
3016 @c NOTE: This confuses the pdf2texi script when it is creating the
3017 @c NOTE: index based upon the first character and so it generates:
3018 @c NOTE: \initial {\\}
3019 @c NOTE: which then results in the error message:
3020 @c NOTE: Argument of \\ has an extra }.
3021 @c NOTE: So in the index entries below a space character has been
3022 @c NOTE: prepended to avoid this problem.
3025 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
3027 @cindex @code{ \b} (backspace character)
3028 @cindex backspace (@code{\b})
3030 Mnemonic for backspace; for ASCII this is octal code 010.
3033 @c Mnemonic for EOText; for ASCII this is octal code 004.
3035 @cindex @code{ \f} (formfeed character)
3036 @cindex formfeed (@code{\f})
3038 Mnemonic for FormFeed; for ASCII this is octal code 014.
3040 @cindex @code{ \n} (newline character)
3041 @cindex newline (@code{\n})
3043 Mnemonic for newline; for ASCII this is octal code 012.
3046 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
3048 @cindex @code{ \r} (carriage return character)
3049 @cindex carriage return (@code{backslash-r})
3051 Mnemonic for carriage-Return; for ASCII this is octal code 015.
3054 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
3055 @c other assemblers.
3057 @cindex @code{ \t} (tab)
3058 @cindex tab (@code{\t})
3060 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
3063 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
3064 @c @item \x @var{digit} @var{digit} @var{digit}
3065 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
3067 @cindex @code{ \@var{ddd}} (octal character code)
3068 @cindex octal character code (@code{\@var{ddd}})
3069 @item \ @var{digit} @var{digit} @var{digit}
3070 An octal character code. The numeric code is 3 octal digits.
3071 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
3072 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3074 @cindex @code{ \@var{xd...}} (hex character code)
3075 @cindex hex character code (@code{\@var{xd...}})
3076 @item \@code{x} @var{hex-digits...}
3077 A hex character code. All trailing hex digits are combined. Either upper or
3078 lower case @code{x} works.
3080 @cindex @code{ \\} (@samp{\} character)
3081 @cindex backslash (@code{\\})
3083 Represents one @samp{\} character.
3086 @c Represents one @samp{'} (accent acute) character.
3087 @c This is needed in single character literals
3088 @c (@xref{Characters,,Character Constants}.) to represent
3091 @cindex @code{ \"} (doublequote character)
3092 @cindex doublequote (@code{\"})
3094 Represents one @samp{"} character. Needed in strings to represent
3095 this character, because an unescaped @samp{"} would end the string.
3097 @item \ @var{anything-else}
3098 Any other character when escaped by @kbd{\} gives a warning, but
3099 assembles as if the @samp{\} was not present. The idea is that if
3100 you used an escape sequence you clearly didn't want the literal
3101 interpretation of the following character. However @command{@value{AS}} has no
3102 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3103 code and warns you of the fact.
3106 Which characters are escapable, and what those escapes represent,
3107 varies widely among assemblers. The current set is what we think
3108 the BSD 4.2 assembler recognizes, and is a subset of what most C
3109 compilers recognize. If you are in doubt, do not use an escape
3113 @subsubsection Characters
3115 @cindex single character constant
3116 @cindex character, single
3117 @cindex constant, single character
3118 A single character may be written as a single quote immediately followed by
3119 that character. Some backslash escapes apply to characters, @code{\b},
3120 @code{\f}, @code{\n}, @code{\r}, @code{\t}, and @code{\"} with the same meaning
3121 as for strings, plus @code{\'} for a single quote. So if you want to write the
3122 character backslash, you must write @kbd{'\\} where the first @code{\} escapes
3123 the second @code{\}. As you can see, the quote is an acute accent, not a grave
3126 @ifclear abnormal-separator
3127 (or semicolon @samp{;})
3129 @ifset abnormal-separator
3131 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3136 immediately following an acute accent is taken as a literal character
3137 and does not count as the end of a statement. The value of a character
3138 constant in a numeric expression is the machine's byte-wide code for
3139 that character. @command{@value{AS}} assumes your character code is ASCII:
3140 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3143 @subsection Number Constants
3145 @cindex constants, number
3146 @cindex number constants
3147 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3148 are stored in the target machine. @emph{Integers} are numbers that
3149 would fit into an @code{int} in the C language. @emph{Bignums} are
3150 integers, but they are stored in more than 32 bits. @emph{Flonums}
3151 are floating point numbers, described below.
3154 * Integers:: Integers
3162 @subsubsection Integers
3164 @cindex constants, integer
3166 @cindex binary integers
3167 @cindex integers, binary
3168 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3169 the binary digits @samp{01}.
3171 @cindex octal integers
3172 @cindex integers, octal
3173 An octal integer is @samp{0} followed by zero or more of the octal
3174 digits (@samp{01234567}).
3176 @cindex decimal integers
3177 @cindex integers, decimal
3178 A decimal integer starts with a non-zero digit followed by zero or
3179 more digits (@samp{0123456789}).
3181 @cindex hexadecimal integers
3182 @cindex integers, hexadecimal
3183 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3184 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3186 Integers have the usual values. To denote a negative integer, use
3187 the prefix operator @samp{-} discussed under expressions
3188 (@pxref{Prefix Ops,,Prefix Operators}).
3191 @subsubsection Bignums
3194 @cindex constants, bignum
3195 A @dfn{bignum} has the same syntax and semantics as an integer
3196 except that the number (or its negative) takes more than 32 bits to
3197 represent in binary. The distinction is made because in some places
3198 integers are permitted while bignums are not.
3201 @subsubsection Flonums
3203 @cindex floating point numbers
3204 @cindex constants, floating point
3206 @cindex precision, floating point
3207 A @dfn{flonum} represents a floating point number. The translation is
3208 indirect: a decimal floating point number from the text is converted by
3209 @command{@value{AS}} to a generic binary floating point number of more than
3210 sufficient precision. This generic floating point number is converted
3211 to a particular computer's floating point format (or formats) by a
3212 portion of @command{@value{AS}} specialized to that computer.
3214 A flonum is written by writing (in order)
3219 (@samp{0} is optional on the HPPA.)
3223 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3225 @kbd{e} is recommended. Case is not important.
3227 @c FIXME: verify if flonum syntax really this vague for most cases
3228 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3229 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3232 On the H8/300 and Renesas / SuperH SH architectures, the letter must be
3233 one of the letters @samp{DFPRSX} (in upper or lower case).
3235 On the ARC, the letter must be one of the letters @samp{DFRS}
3236 (in upper or lower case).
3238 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3242 One of the letters @samp{DFRS} (in upper or lower case).
3245 One of the letters @samp{DFPRSX} (in upper or lower case).
3248 The letter @samp{E} (upper case only).
3253 An optional sign: either @samp{+} or @samp{-}.
3256 An optional @dfn{integer part}: zero or more decimal digits.
3259 An optional @dfn{fractional part}: @samp{.} followed by zero
3260 or more decimal digits.
3263 An optional exponent, consisting of:
3267 An @samp{E} or @samp{e}.
3268 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3269 @c principle this can perfectly well be different on different targets.
3271 Optional sign: either @samp{+} or @samp{-}.
3273 One or more decimal digits.
3278 At least one of the integer part or the fractional part must be
3279 present. The floating point number has the usual base-10 value.
3281 @command{@value{AS}} does all processing using integers. Flonums are computed
3282 independently of any floating point hardware in the computer running
3283 @command{@value{AS}}.
3286 @chapter Sections and Relocation
3291 * Secs Background:: Background
3292 * Ld Sections:: Linker Sections
3293 * As Sections:: Assembler Internal Sections
3294 * Sub-Sections:: Sub-Sections
3298 @node Secs Background
3301 Roughly, a section is a range of addresses, with no gaps; all data
3302 ``in'' those addresses is treated the same for some particular purpose.
3303 For example there may be a ``read only'' section.
3305 @cindex linker, and assembler
3306 @cindex assembler, and linker
3307 The linker @code{@value{LD}} reads many object files (partial programs) and
3308 combines their contents to form a runnable program. When @command{@value{AS}}
3309 emits an object file, the partial program is assumed to start at address 0.
3310 @code{@value{LD}} assigns the final addresses for the partial program, so that
3311 different partial programs do not overlap. This is actually an
3312 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3315 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3316 addresses. These blocks slide to their run-time addresses as rigid
3317 units; their length does not change and neither does the order of bytes
3318 within them. Such a rigid unit is called a @emph{section}. Assigning
3319 run-time addresses to sections is called @dfn{relocation}. It includes
3320 the task of adjusting mentions of object-file addresses so they refer to
3321 the proper run-time addresses.
3323 For the H8/300, and for the Renesas / SuperH SH,
3324 @command{@value{AS}} pads sections if needed to
3325 ensure they end on a word (sixteen bit) boundary.
3328 @cindex standard assembler sections
3329 An object file written by @command{@value{AS}} has at least three sections, any
3330 of which may be empty. These are named @dfn{text}, @dfn{data} and
3335 When it generates COFF or ELF output,
3337 @command{@value{AS}} can also generate whatever other named sections you specify
3338 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3339 If you do not use any directives that place output in the @samp{.text}
3340 or @samp{.data} sections, these sections still exist, but are empty.
3345 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3347 @command{@value{AS}} can also generate whatever other named sections you
3348 specify using the @samp{.space} and @samp{.subspace} directives. See
3349 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3350 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3351 assembler directives.
3354 Additionally, @command{@value{AS}} uses different names for the standard
3355 text, data, and bss sections when generating SOM output. Program text
3356 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3357 BSS into @samp{$BSS$}.
3361 Within the object file, the text section starts at address @code{0}, the
3362 data section follows, and the bss section follows the data section.
3365 When generating either SOM or ELF output files on the HPPA, the text
3366 section starts at address @code{0}, the data section at address
3367 @code{0x4000000}, and the bss section follows the data section.
3370 To let @code{@value{LD}} know which data changes when the sections are
3371 relocated, and how to change that data, @command{@value{AS}} also writes to the
3372 object file details of the relocation needed. To perform relocation
3373 @code{@value{LD}} must know, each time an address in the object
3377 Where in the object file is the beginning of this reference to
3380 How long (in bytes) is this reference?
3382 Which section does the address refer to? What is the numeric value of
3384 (@var{address}) @minus{} (@var{start-address of section})?
3387 Is the reference to an address ``Program-Counter relative''?
3390 @cindex addresses, format of
3391 @cindex section-relative addressing
3392 In fact, every address @command{@value{AS}} ever uses is expressed as
3394 (@var{section}) + (@var{offset into section})
3397 Further, most expressions @command{@value{AS}} computes have this section-relative
3400 (For some object formats, such as SOM for the HPPA, some expressions are
3401 symbol-relative instead.)
3404 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3405 @var{N} into section @var{secname}.''
3407 Apart from text, data and bss sections you need to know about the
3408 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3409 addresses in the absolute section remain unchanged. For example, address
3410 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3411 @code{@value{LD}}. Although the linker never arranges two partial programs'
3412 data sections with overlapping addresses after linking, @emph{by definition}
3413 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3414 part of a program is always the same address when the program is running as
3415 address @code{@{absolute@ 239@}} in any other part of the program.
3417 The idea of sections is extended to the @dfn{undefined} section. Any
3418 address whose section is unknown at assembly time is by definition
3419 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3420 Since numbers are always defined, the only way to generate an undefined
3421 address is to mention an undefined symbol. A reference to a named
3422 common block would be such a symbol: its value is unknown at assembly
3423 time so it has section @emph{undefined}.
3425 By analogy the word @emph{section} is used to describe groups of sections in
3426 the linked program. @code{@value{LD}} puts all partial programs' text
3427 sections in contiguous addresses in the linked program. It is
3428 customary to refer to the @emph{text section} of a program, meaning all
3429 the addresses of all partial programs' text sections. Likewise for
3430 data and bss sections.
3432 Some sections are manipulated by @code{@value{LD}}; others are invented for
3433 use of @command{@value{AS}} and have no meaning except during assembly.
3436 @section Linker Sections
3437 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3442 @cindex named sections
3443 @cindex sections, named
3444 @item named sections
3447 @cindex text section
3448 @cindex data section
3452 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3453 separate but equal sections. Anything you can say of one section is
3456 When the program is running, however, it is
3457 customary for the text section to be unalterable. The
3458 text section is often shared among processes: it contains
3459 instructions, constants and the like. The data section of a running
3460 program is usually alterable: for example, C variables would be stored
3461 in the data section.
3466 This section contains zeroed bytes when your program begins running. It
3467 is used to hold uninitialized variables or common storage. The length of
3468 each partial program's bss section is important, but because it starts
3469 out containing zeroed bytes there is no need to store explicit zero
3470 bytes in the object file. The bss section was invented to eliminate
3471 those explicit zeros from object files.
3473 @cindex absolute section
3474 @item absolute section
3475 Address 0 of this section is always ``relocated'' to runtime address 0.
3476 This is useful if you want to refer to an address that @code{@value{LD}} must
3477 not change when relocating. In this sense we speak of absolute
3478 addresses being ``unrelocatable'': they do not change during relocation.
3480 @cindex undefined section
3481 @item undefined section
3482 This ``section'' is a catch-all for address references to objects not in
3483 the preceding sections.
3484 @c FIXME: ref to some other doc on obj-file formats could go here.
3487 @cindex relocation example
3488 An idealized example of three relocatable sections follows.
3490 The example uses the traditional section names @samp{.text} and @samp{.data}.
3492 Memory addresses are on the horizontal axis.
3496 @c END TEXI2ROFF-KILL
3499 partial program # 1: |ttttt|dddd|00|
3506 partial program # 2: |TTT|DDD|000|
3509 +--+---+-----+--+----+---+-----+~~
3510 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3511 +--+---+-----+--+----+---+-----+~~
3513 addresses: 0 @dots{}
3520 \line{\it Partial program \#1: \hfil}
3521 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3522 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3524 \line{\it Partial program \#2: \hfil}
3525 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3526 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3528 \line{\it linked program: \hfil}
3529 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3530 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3531 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3532 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3534 \line{\it addresses: \hfil}
3538 @c END TEXI2ROFF-KILL
3541 @section Assembler Internal Sections
3543 @cindex internal assembler sections
3544 @cindex sections in messages, internal
3545 These sections are meant only for the internal use of @command{@value{AS}}. They
3546 have no meaning at run-time. You do not really need to know about these
3547 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3548 warning messages, so it might be helpful to have an idea of their
3549 meanings to @command{@value{AS}}. These sections are used to permit the
3550 value of every expression in your assembly language program to be a
3551 section-relative address.
3554 @cindex assembler internal logic error
3555 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3556 An internal assembler logic error has been found. This means there is a
3557 bug in the assembler.
3559 @cindex expr (internal section)
3561 The assembler stores complex expression internally as combinations of
3562 symbols. When it needs to represent an expression as a symbol, it puts
3563 it in the expr section.
3565 @c FIXME item transfer[t] vector preload
3566 @c FIXME item transfer[t] vector postload
3567 @c FIXME item register
3571 @section Sub-Sections
3573 @cindex numbered subsections
3574 @cindex grouping data
3580 fall into two sections: text and data.
3582 You may have separate groups of
3584 data in named sections
3588 data in named sections
3594 that you want to end up near to each other in the object file, even though they
3595 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3596 use @dfn{subsections} for this purpose. Within each section, there can be
3597 numbered subsections with values from 0 to 8192. Objects assembled into the
3598 same subsection go into the object file together with other objects in the same
3599 subsection. For example, a compiler might want to store constants in the text
3600 section, but might not want to have them interspersed with the program being
3601 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3602 section of code being output, and a @samp{.text 1} before each group of
3603 constants being output.
3605 Subsections are optional. If you do not use subsections, everything
3606 goes in subsection number zero.
3609 Each subsection is zero-padded up to a multiple of four bytes.
3610 (Subsections may be padded a different amount on different flavors
3611 of @command{@value{AS}}.)
3615 On the H8/300 platform, each subsection is zero-padded to a word
3616 boundary (two bytes).
3617 The same is true on the Renesas SH.
3621 Subsections appear in your object file in numeric order, lowest numbered
3622 to highest. (All this to be compatible with other people's assemblers.)
3623 The object file contains no representation of subsections; @code{@value{LD}} and
3624 other programs that manipulate object files see no trace of them.
3625 They just see all your text subsections as a text section, and all your
3626 data subsections as a data section.
3628 To specify which subsection you want subsequent statements assembled
3629 into, use a numeric argument to specify it, in a @samp{.text
3630 @var{expression}} or a @samp{.data @var{expression}} statement.
3633 When generating COFF output, you
3638 can also use an extra subsection
3639 argument with arbitrary named sections: @samp{.section @var{name},
3644 When generating ELF output, you
3649 can also use the @code{.subsection} directive (@pxref{SubSection})
3650 to specify a subsection: @samp{.subsection @var{expression}}.
3652 @var{Expression} should be an absolute expression
3653 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3654 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3655 begins in @code{text 0}. For instance:
3657 .text 0 # The default subsection is text 0 anyway.
3658 .ascii "This lives in the first text subsection. *"
3660 .ascii "But this lives in the second text subsection."
3662 .ascii "This lives in the data section,"
3663 .ascii "in the first data subsection."
3665 .ascii "This lives in the first text section,"
3666 .ascii "immediately following the asterisk (*)."
3669 Each section has a @dfn{location counter} incremented by one for every byte
3670 assembled into that section. Because subsections are merely a convenience
3671 restricted to @command{@value{AS}} there is no concept of a subsection location
3672 counter. There is no way to directly manipulate a location counter---but the
3673 @code{.align} directive changes it, and any label definition captures its
3674 current value. The location counter of the section where statements are being
3675 assembled is said to be the @dfn{active} location counter.
3678 @section bss Section
3681 @cindex common variable storage
3682 The bss section is used for local common variable storage.
3683 You may allocate address space in the bss section, but you may
3684 not dictate data to load into it before your program executes. When
3685 your program starts running, all the contents of the bss
3686 section are zeroed bytes.
3688 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3689 @ref{Lcomm,,@code{.lcomm}}.
3691 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3692 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3695 When assembling for a target which supports multiple sections, such as ELF or
3696 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3697 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3698 section. Typically the section will only contain symbol definitions and
3699 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3706 Symbols are a central concept: the programmer uses symbols to name
3707 things, the linker uses symbols to link, and the debugger uses symbols
3711 @cindex debuggers, and symbol order
3712 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3713 the same order they were declared. This may break some debuggers.
3718 * Setting Symbols:: Giving Symbols Other Values
3719 * Symbol Names:: Symbol Names
3720 * Dot:: The Special Dot Symbol
3721 * Symbol Attributes:: Symbol Attributes
3728 A @dfn{label} is written as a symbol immediately followed by a colon
3729 @samp{:}. The symbol then represents the current value of the
3730 active location counter, and is, for example, a suitable instruction
3731 operand. You are warned if you use the same symbol to represent two
3732 different locations: the first definition overrides any other
3736 On the HPPA, the usual form for a label need not be immediately followed by a
3737 colon, but instead must start in column zero. Only one label may be defined on
3738 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3739 provides a special directive @code{.label} for defining labels more flexibly.
3742 @node Setting Symbols
3743 @section Giving Symbols Other Values
3745 @cindex assigning values to symbols
3746 @cindex symbol values, assigning
3747 A symbol can be given an arbitrary value by writing a symbol, followed
3748 by an equals sign @samp{=}, followed by an expression
3749 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3750 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3751 equals sign @samp{=}@samp{=} here represents an equivalent of the
3752 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3755 Blackfin does not support symbol assignment with @samp{=}.
3759 @section Symbol Names
3761 @cindex symbol names
3762 @cindex names, symbol
3763 @ifclear SPECIAL-SYMS
3764 Symbol names begin with a letter or with one of @samp{._}. On most
3765 machines, you can also use @code{$} in symbol names; exceptions are
3766 noted in @ref{Machine Dependencies}. That character may be followed by any
3767 string of digits, letters, dollar signs (unless otherwise noted for a
3768 particular target machine), and underscores.
3772 Symbol names begin with a letter or with one of @samp{._}. On the
3773 Renesas SH you can also use @code{$} in symbol names. That
3774 character may be followed by any string of digits, letters, dollar signs (save
3775 on the H8/300), and underscores.
3779 Case of letters is significant: @code{foo} is a different symbol name
3782 Symbol names do not start with a digit. An exception to this rule is made for
3783 Local Labels. See below.
3785 Multibyte characters are supported. To generate a symbol name containing
3786 multibyte characters enclose it within double quotes and use escape codes. cf
3787 @xref{Strings}. Generating a multibyte symbol name from a label is not
3788 currently supported.
3790 Each symbol has exactly one name. Each name in an assembly language program
3791 refers to exactly one symbol. You may use that symbol name any number of times
3794 @subheading Local Symbol Names
3796 @cindex local symbol names
3797 @cindex symbol names, local
3798 A local symbol is any symbol beginning with certain local label prefixes.
3799 By default, the local label prefix is @samp{.L} for ELF systems or
3800 @samp{L} for traditional a.out systems, but each target may have its own
3801 set of local label prefixes.
3803 On the HPPA local symbols begin with @samp{L$}.
3806 Local symbols are defined and used within the assembler, but they are
3807 normally not saved in object files. Thus, they are not visible when debugging.
3808 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3809 to retain the local symbols in the object files.
3811 @subheading Local Labels
3813 @cindex local labels
3814 @cindex temporary symbol names
3815 @cindex symbol names, temporary
3816 Local labels are different from local symbols. Local labels help compilers and
3817 programmers use names temporarily. They create symbols which are guaranteed to
3818 be unique over the entire scope of the input source code and which can be
3819 referred to by a simple notation. To define a local label, write a label of
3820 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3821 To refer to the most recent previous definition of that label write
3822 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3823 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3824 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3826 There is no restriction on how you can use these labels, and you can reuse them
3827 too. So that it is possible to repeatedly define the same local label (using
3828 the same number @samp{@b{N}}), although you can only refer to the most recently
3829 defined local label of that number (for a backwards reference) or the next
3830 definition of a specific local label for a forward reference. It is also worth
3831 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3832 implemented in a slightly more efficient manner than the others.
3843 Which is the equivalent of:
3846 label_1: branch label_3
3847 label_2: branch label_1
3848 label_3: branch label_4
3849 label_4: branch label_3
3852 Local label names are only a notational device. They are immediately
3853 transformed into more conventional symbol names before the assembler uses them.
3854 The symbol names are stored in the symbol table, appear in error messages, and
3855 are optionally emitted to the object file. The names are constructed using
3859 @item @emph{local label prefix}
3860 All local symbols begin with the system-specific local label prefix.
3861 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3862 that start with the local label prefix. These labels are
3863 used for symbols you are never intended to see. If you use the
3864 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3865 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3866 you may use them in debugging.
3869 This is the number that was used in the local label definition. So if the
3870 label is written @samp{55:} then the number is @samp{55}.
3873 This unusual character is included so you do not accidentally invent a symbol
3874 of the same name. The character has ASCII value of @samp{\002} (control-B).
3876 @item @emph{ordinal number}
3877 This is a serial number to keep the labels distinct. The first definition of
3878 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3879 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3880 the number @samp{1} and its 15th definition gets @samp{15} as well.
3883 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3884 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3886 @subheading Dollar Local Labels
3887 @cindex dollar local symbols
3889 On some targets @code{@value{AS}} also supports an even more local form of
3890 local labels called dollar labels. These labels go out of scope (i.e., they
3891 become undefined) as soon as a non-local label is defined. Thus they remain
3892 valid for only a small region of the input source code. Normal local labels,
3893 by contrast, remain in scope for the entire file, or until they are redefined
3894 by another occurrence of the same local label.
3896 Dollar labels are defined in exactly the same way as ordinary local labels,
3897 except that they have a dollar sign suffix to their numeric value, e.g.,
3900 They can also be distinguished from ordinary local labels by their transformed
3901 names which use ASCII character @samp{\001} (control-A) as the magic character
3902 to distinguish them from ordinary labels. For example, the fifth definition of
3903 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3906 @section The Special Dot Symbol
3908 @cindex dot (symbol)
3909 @cindex @code{.} (symbol)
3910 @cindex current address
3911 @cindex location counter
3912 The special symbol @samp{.} refers to the current address that
3913 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3914 .long .} defines @code{melvin} to contain its own address.
3915 Assigning a value to @code{.} is treated the same as a @code{.org}
3917 @ifclear no-space-dir
3918 Thus, the expression @samp{.=.+4} is the same as saying
3922 @node Symbol Attributes
3923 @section Symbol Attributes
3925 @cindex symbol attributes
3926 @cindex attributes, symbol
3927 Every symbol has, as well as its name, the attributes ``Value'' and
3928 ``Type''. Depending on output format, symbols can also have auxiliary
3931 The detailed definitions are in @file{a.out.h}.
3934 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3935 all these attributes, and probably won't warn you. This makes the
3936 symbol an externally defined symbol, which is generally what you
3940 * Symbol Value:: Value
3941 * Symbol Type:: Type
3943 * a.out Symbols:: Symbol Attributes: @code{a.out}
3946 * COFF Symbols:: Symbol Attributes for COFF
3949 * SOM Symbols:: Symbol Attributes for SOM
3956 @cindex value of a symbol
3957 @cindex symbol value
3958 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3959 location in the text, data, bss or absolute sections the value is the
3960 number of addresses from the start of that section to the label.
3961 Naturally for text, data and bss sections the value of a symbol changes
3962 as @code{@value{LD}} changes section base addresses during linking. Absolute
3963 symbols' values do not change during linking: that is why they are
3966 The value of an undefined symbol is treated in a special way. If it is
3967 0 then the symbol is not defined in this assembler source file, and
3968 @code{@value{LD}} tries to determine its value from other files linked into the
3969 same program. You make this kind of symbol simply by mentioning a symbol
3970 name without defining it. A non-zero value represents a @code{.comm}
3971 common declaration. The value is how much common storage to reserve, in
3972 bytes (addresses). The symbol refers to the first address of the
3978 @cindex type of a symbol
3980 The type attribute of a symbol contains relocation (section)
3981 information, any flag settings indicating that a symbol is external, and
3982 (optionally), other information for linkers and debuggers. The exact
3983 format depends on the object-code output format in use.
3987 @subsection Symbol Attributes: @code{a.out}
3989 @cindex @code{a.out} symbol attributes
3990 @cindex symbol attributes, @code{a.out}
3993 * Symbol Desc:: Descriptor
3994 * Symbol Other:: Other
3998 @subsubsection Descriptor
4000 @cindex descriptor, of @code{a.out} symbol
4001 This is an arbitrary 16-bit value. You may establish a symbol's
4002 descriptor value by using a @code{.desc} statement
4003 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
4004 @command{@value{AS}}.
4007 @subsubsection Other
4009 @cindex other attribute, of @code{a.out} symbol
4010 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
4015 @subsection Symbol Attributes for COFF
4017 @cindex COFF symbol attributes
4018 @cindex symbol attributes, COFF
4020 The COFF format supports a multitude of auxiliary symbol attributes;
4021 like the primary symbol attributes, they are set between @code{.def} and
4022 @code{.endef} directives.
4024 @subsubsection Primary Attributes
4026 @cindex primary attributes, COFF symbols
4027 The symbol name is set with @code{.def}; the value and type,
4028 respectively, with @code{.val} and @code{.type}.
4030 @subsubsection Auxiliary Attributes
4032 @cindex auxiliary attributes, COFF symbols
4033 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
4034 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
4035 table information for COFF.
4040 @subsection Symbol Attributes for SOM
4042 @cindex SOM symbol attributes
4043 @cindex symbol attributes, SOM
4045 The SOM format for the HPPA supports a multitude of symbol attributes set with
4046 the @code{.EXPORT} and @code{.IMPORT} directives.
4048 The attributes are described in @cite{HP9000 Series 800 Assembly
4049 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
4050 @code{EXPORT} assembler directive documentation.
4054 @chapter Expressions
4058 @cindex numeric values
4059 An @dfn{expression} specifies an address or numeric value.
4060 Whitespace may precede and/or follow an expression.
4062 The result of an expression must be an absolute number, or else an offset into
4063 a particular section. If an expression is not absolute, and there is not
4064 enough information when @command{@value{AS}} sees the expression to know its
4065 section, a second pass over the source program might be necessary to interpret
4066 the expression---but the second pass is currently not implemented.
4067 @command{@value{AS}} aborts with an error message in this situation.
4070 * Empty Exprs:: Empty Expressions
4071 * Integer Exprs:: Integer Expressions
4075 @section Empty Expressions
4077 @cindex empty expressions
4078 @cindex expressions, empty
4079 An empty expression has no value: it is just whitespace or null.
4080 Wherever an absolute expression is required, you may omit the
4081 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4082 is compatible with other assemblers.
4085 @section Integer Expressions
4087 @cindex integer expressions
4088 @cindex expressions, integer
4089 An @dfn{integer expression} is one or more @emph{arguments} delimited
4090 by @emph{operators}.
4093 * Arguments:: Arguments
4094 * Operators:: Operators
4095 * Prefix Ops:: Prefix Operators
4096 * Infix Ops:: Infix Operators
4100 @subsection Arguments
4102 @cindex expression arguments
4103 @cindex arguments in expressions
4104 @cindex operands in expressions
4105 @cindex arithmetic operands
4106 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4107 contexts arguments are sometimes called ``arithmetic operands''. In
4108 this manual, to avoid confusing them with the ``instruction operands'' of
4109 the machine language, we use the term ``argument'' to refer to parts of
4110 expressions only, reserving the word ``operand'' to refer only to machine
4111 instruction operands.
4113 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4114 @var{section} is one of text, data, bss, absolute,
4115 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4118 Numbers are usually integers.
4120 A number can be a flonum or bignum. In this case, you are warned
4121 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4122 these 32 bits are an integer. You may write integer-manipulating
4123 instructions that act on exotic constants, compatible with other
4126 @cindex subexpressions
4127 Subexpressions are a left parenthesis @samp{(} followed by an integer
4128 expression, followed by a right parenthesis @samp{)}; or a prefix
4129 operator followed by an argument.
4132 @subsection Operators
4134 @cindex operators, in expressions
4135 @cindex arithmetic functions
4136 @cindex functions, in expressions
4137 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4138 operators are followed by an argument. Infix operators appear
4139 between their arguments. Operators may be preceded and/or followed by
4143 @subsection Prefix Operator
4145 @cindex prefix operators
4146 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4147 one argument, which must be absolute.
4149 @c the tex/end tex stuff surrounding this small table is meant to make
4150 @c it align, on the printed page, with the similar table in the next
4151 @c section (which is inside an enumerate).
4153 \global\advance\leftskip by \itemindent
4158 @dfn{Negation}. Two's complement negation.
4160 @dfn{Complementation}. Bitwise not.
4164 \global\advance\leftskip by -\itemindent
4168 @subsection Infix Operators
4170 @cindex infix operators
4171 @cindex operators, permitted arguments
4172 @dfn{Infix operators} take two arguments, one on either side. Operators
4173 have precedence, but operations with equal precedence are performed left
4174 to right. Apart from @code{+} or @option{-}, both arguments must be
4175 absolute, and the result is absolute.
4178 @cindex operator precedence
4179 @cindex precedence of operators
4186 @dfn{Multiplication}.
4189 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4195 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4198 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4202 Intermediate precedence
4207 @dfn{Bitwise Inclusive Or}.
4213 @dfn{Bitwise Exclusive Or}.
4216 @dfn{Bitwise Or Not}.
4223 @cindex addition, permitted arguments
4224 @cindex plus, permitted arguments
4225 @cindex arguments for addition
4227 @dfn{Addition}. If either argument is absolute, the result has the section of
4228 the other argument. You may not add together arguments from different
4231 @cindex subtraction, permitted arguments
4232 @cindex minus, permitted arguments
4233 @cindex arguments for subtraction
4235 @dfn{Subtraction}. If the right argument is absolute, the
4236 result has the section of the left argument.
4237 If both arguments are in the same section, the result is absolute.
4238 You may not subtract arguments from different sections.
4239 @c FIXME is there still something useful to say about undefined - undefined ?
4241 @cindex comparison expressions
4242 @cindex expressions, comparison
4247 @dfn{Is Not Equal To}
4251 @dfn{Is Greater Than}
4253 @dfn{Is Greater Than Or Equal To}
4255 @dfn{Is Less Than Or Equal To}
4257 The comparison operators can be used as infix operators. A true results has a
4258 value of -1 whereas a false result has a value of 0. Note, these operators
4259 perform signed comparisons.
4262 @item Lowest Precedence
4271 These two logical operations can be used to combine the results of sub
4272 expressions. Note, unlike the comparison operators a true result returns a
4273 value of 1 but a false results does still return 0. Also note that the logical
4274 or operator has a slightly lower precedence than logical and.
4279 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4280 address; you can only have a defined section in one of the two arguments.
4283 @chapter Assembler Directives
4285 @cindex directives, machine independent
4286 @cindex pseudo-ops, machine independent
4287 @cindex machine independent directives
4288 All assembler directives have names that begin with a period (@samp{.}).
4289 The names are case insensitive for most targets, and usually written
4292 This chapter discusses directives that are available regardless of the
4293 target machine configuration for the @sc{gnu} assembler.
4295 Some machine configurations provide additional directives.
4296 @xref{Machine Dependencies}.
4299 @ifset machine-directives
4300 @xref{Machine Dependencies}, for additional directives.
4305 * Abort:: @code{.abort}
4307 * ABORT (COFF):: @code{.ABORT}
4310 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4311 * Altmacro:: @code{.altmacro}
4312 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4313 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4314 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4315 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4316 * Byte:: @code{.byte @var{expressions}}
4317 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4318 * Comm:: @code{.comm @var{symbol} , @var{length} }
4319 * Data:: @code{.data @var{subsection}}
4320 * Dc:: @code{.dc[@var{size}] @var{expressions}}
4321 * Dcb:: @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4322 * Ds:: @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4324 * Def:: @code{.def @var{name}}
4327 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4333 * Double:: @code{.double @var{flonums}}
4334 * Eject:: @code{.eject}
4335 * Else:: @code{.else}
4336 * Elseif:: @code{.elseif}
4339 * Endef:: @code{.endef}
4342 * Endfunc:: @code{.endfunc}
4343 * Endif:: @code{.endif}
4344 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4345 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4346 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4348 * Error:: @code{.error @var{string}}
4349 * Exitm:: @code{.exitm}
4350 * Extern:: @code{.extern}
4351 * Fail:: @code{.fail}
4352 * File:: @code{.file}
4353 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4354 * Float:: @code{.float @var{flonums}}
4355 * Func:: @code{.func}
4356 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4358 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4359 * Hidden:: @code{.hidden @var{names}}
4362 * hword:: @code{.hword @var{expressions}}
4363 * Ident:: @code{.ident}
4364 * If:: @code{.if @var{absolute expression}}
4365 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4366 * Include:: @code{.include "@var{file}"}
4367 * Int:: @code{.int @var{expressions}}
4369 * Internal:: @code{.internal @var{names}}
4372 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4373 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4374 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4375 * Lflags:: @code{.lflags}
4376 @ifclear no-line-dir
4377 * Line:: @code{.line @var{line-number}}
4380 * Linkonce:: @code{.linkonce [@var{type}]}
4381 * List:: @code{.list}
4382 * Ln:: @code{.ln @var{line-number}}
4383 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4384 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4386 * Local:: @code{.local @var{names}}
4389 * Long:: @code{.long @var{expressions}}
4391 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4394 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4395 * MRI:: @code{.mri @var{val}}
4396 * Noaltmacro:: @code{.noaltmacro}
4397 * Nolist:: @code{.nolist}
4398 * Nops:: @code{.nops @var{size}[, @var{control}]}
4399 * Octa:: @code{.octa @var{bignums}}
4400 * Offset:: @code{.offset @var{loc}}
4401 * Org:: @code{.org @var{new-lc}, @var{fill}}
4402 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4404 * PopSection:: @code{.popsection}
4405 * Previous:: @code{.previous}
4408 * Print:: @code{.print @var{string}}
4410 * Protected:: @code{.protected @var{names}}
4413 * Psize:: @code{.psize @var{lines}, @var{columns}}
4414 * Purgem:: @code{.purgem @var{name}}
4416 * PushSection:: @code{.pushsection @var{name}}
4419 * Quad:: @code{.quad @var{bignums}}
4420 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4421 * Rept:: @code{.rept @var{count}}
4422 * Sbttl:: @code{.sbttl "@var{subheading}"}
4424 * Scl:: @code{.scl @var{class}}
4427 * Section:: @code{.section @var{name}[, @var{flags}]}
4430 * Set:: @code{.set @var{symbol}, @var{expression}}
4431 * Short:: @code{.short @var{expressions}}
4432 * Single:: @code{.single @var{flonums}}
4434 * Size:: @code{.size [@var{name} , @var{expression}]}
4436 @ifclear no-space-dir
4437 * Skip:: @code{.skip @var{size} [,@var{fill}]}
4440 * Sleb128:: @code{.sleb128 @var{expressions}}
4441 @ifclear no-space-dir
4442 * Space:: @code{.space @var{size} [,@var{fill}]}
4445 * Stab:: @code{.stabd, .stabn, .stabs}
4448 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4449 * Struct:: @code{.struct @var{expression}}
4451 * SubSection:: @code{.subsection}
4452 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4456 * Tag:: @code{.tag @var{structname}}
4459 * Text:: @code{.text @var{subsection}}
4460 * Title:: @code{.title "@var{heading}"}
4462 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4465 * Uleb128:: @code{.uleb128 @var{expressions}}
4467 * Val:: @code{.val @var{addr}}
4471 * Version:: @code{.version "@var{string}"}
4472 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4473 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4476 * Warning:: @code{.warning @var{string}}
4477 * Weak:: @code{.weak @var{names}}
4478 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4479 * Word:: @code{.word @var{expressions}}
4480 @ifclear no-space-dir
4481 * Zero:: @code{.zero @var{size}}
4484 * 2byte:: @code{.2byte @var{expressions}}
4485 * 4byte:: @code{.4byte @var{expressions}}
4486 * 8byte:: @code{.8byte @var{bignums}}
4488 * Deprecated:: Deprecated Directives
4492 @section @code{.abort}
4494 @cindex @code{abort} directive
4495 @cindex stopping the assembly
4496 This directive stops the assembly immediately. It is for
4497 compatibility with other assemblers. The original idea was that the
4498 assembly language source would be piped into the assembler. If the sender
4499 of the source quit, it could use this directive tells @command{@value{AS}} to
4500 quit also. One day @code{.abort} will not be supported.
4504 @section @code{.ABORT} (COFF)
4506 @cindex @code{ABORT} directive
4507 When producing COFF output, @command{@value{AS}} accepts this directive as a
4508 synonym for @samp{.abort}.
4513 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4515 @cindex padding the location counter
4516 @cindex @code{align} directive
4517 Pad the location counter (in the current subsection) to a particular storage
4518 boundary. The first expression (which must be absolute) is the alignment
4519 required, as described below.
4521 The second expression (also absolute) gives the fill value to be stored in the
4522 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4523 padding bytes are normally zero. However, on most systems, if the section is
4524 marked as containing code and the fill value is omitted, the space is filled
4525 with no-op instructions.
4527 The third expression is also absolute, and is also optional. If it is present,
4528 it is the maximum number of bytes that should be skipped by this alignment
4529 directive. If doing the alignment would require skipping more bytes than the
4530 specified maximum, then the alignment is not done at all. You can omit the
4531 fill value (the second argument) entirely by simply using two commas after the
4532 required alignment; this can be useful if you want the alignment to be filled
4533 with no-op instructions when appropriate.
4535 The way the required alignment is specified varies from system to system.
4536 For the arc, hppa, i386 using ELF, iq2000, m68k, or1k,
4537 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4538 alignment request in bytes. For example @samp{.align 8} advances
4539 the location counter until it is a multiple of 8. If the location counter
4540 is already a multiple of 8, no change is needed. For the tic54x, the
4541 first expression is the alignment request in words.
4543 For other systems, including ppc, i386 using a.out format, arm and
4544 strongarm, it is the
4545 number of low-order zero bits the location counter must have after
4546 advancement. For example @samp{.align 3} advances the location
4547 counter until it a multiple of 8. If the location counter is already a
4548 multiple of 8, no change is needed.
4550 This inconsistency is due to the different behaviors of the various
4551 native assemblers for these systems which GAS must emulate.
4552 GAS also provides @code{.balign} and @code{.p2align} directives,
4553 described later, which have a consistent behavior across all
4554 architectures (but are specific to GAS).
4557 @section @code{.altmacro}
4558 Enable alternate macro mode, enabling:
4561 @item LOCAL @var{name} [ , @dots{} ]
4562 One additional directive, @code{LOCAL}, is available. It is used to
4563 generate a string replacement for each of the @var{name} arguments, and
4564 replace any instances of @var{name} in each macro expansion. The
4565 replacement string is unique in the assembly, and different for each
4566 separate macro expansion. @code{LOCAL} allows you to write macros that
4567 define symbols, without fear of conflict between separate macro expansions.
4569 @item String delimiters
4570 You can write strings delimited in these other ways besides
4571 @code{"@var{string}"}:
4574 @item '@var{string}'
4575 You can delimit strings with single-quote characters.
4577 @item <@var{string}>
4578 You can delimit strings with matching angle brackets.
4581 @item single-character string escape
4582 To include any single character literally in a string (even if the
4583 character would otherwise have some special meaning), you can prefix the
4584 character with @samp{!} (an exclamation mark). For example, you can
4585 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4587 @item Expression results as strings
4588 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4589 and use the result as a string.
4593 @section @code{.ascii "@var{string}"}@dots{}
4595 @cindex @code{ascii} directive
4596 @cindex string literals
4597 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4598 separated by commas. It assembles each string (with no automatic
4599 trailing zero byte) into consecutive addresses.
4602 @section @code{.asciz "@var{string}"}@dots{}
4604 @cindex @code{asciz} directive
4605 @cindex zero-terminated strings
4606 @cindex null-terminated strings
4607 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4608 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4611 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4613 @cindex padding the location counter given number of bytes
4614 @cindex @code{balign} directive
4615 Pad the location counter (in the current subsection) to a particular
4616 storage boundary. The first expression (which must be absolute) is the
4617 alignment request in bytes. For example @samp{.balign 8} advances
4618 the location counter until it is a multiple of 8. If the location counter
4619 is already a multiple of 8, no change is needed.
4621 The second expression (also absolute) gives the fill value to be stored in the
4622 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4623 padding bytes are normally zero. However, on most systems, if the section is
4624 marked as containing code and the fill value is omitted, the space is filled
4625 with no-op instructions.
4627 The third expression is also absolute, and is also optional. If it is present,
4628 it is the maximum number of bytes that should be skipped by this alignment
4629 directive. If doing the alignment would require skipping more bytes than the
4630 specified maximum, then the alignment is not done at all. You can omit the
4631 fill value (the second argument) entirely by simply using two commas after the
4632 required alignment; this can be useful if you want the alignment to be filled
4633 with no-op instructions when appropriate.
4635 @cindex @code{balignw} directive
4636 @cindex @code{balignl} directive
4637 The @code{.balignw} and @code{.balignl} directives are variants of the
4638 @code{.balign} directive. The @code{.balignw} directive treats the fill
4639 pattern as a two byte word value. The @code{.balignl} directives treats the
4640 fill pattern as a four byte longword value. For example, @code{.balignw
4641 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4642 filled in with the value 0x368d (the exact placement of the bytes depends upon
4643 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4646 @node Bundle directives
4647 @section Bundle directives
4648 @subsection @code{.bundle_align_mode @var{abs-expr}}
4649 @cindex @code{bundle_align_mode} directive
4651 @cindex instruction bundle
4652 @cindex aligned instruction bundle
4653 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4654 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4655 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4656 disabled (which is the default state). If the argument it not zero, it
4657 gives the size of an instruction bundle as a power of two (as for the
4658 @code{.p2align} directive, @pxref{P2align}).
4660 For some targets, it's an ABI requirement that no instruction may span a
4661 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4662 instructions that starts on an aligned boundary. For example, if
4663 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4664 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4665 effect, no single instruction may span a boundary between bundles. If an
4666 instruction would start too close to the end of a bundle for the length of
4667 that particular instruction to fit within the bundle, then the space at the
4668 end of that bundle is filled with no-op instructions so the instruction
4669 starts in the next bundle. As a corollary, it's an error if any single
4670 instruction's encoding is longer than the bundle size.
4672 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4673 @cindex @code{bundle_lock} directive
4674 @cindex @code{bundle_unlock} directive
4675 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4676 allow explicit control over instruction bundle padding. These directives
4677 are only valid when @code{.bundle_align_mode} has been used to enable
4678 aligned instruction bundle mode. It's an error if they appear when
4679 @code{.bundle_align_mode} has not been used at all, or when the last
4680 directive was @w{@code{.bundle_align_mode 0}}.
4682 @cindex bundle-locked
4683 For some targets, it's an ABI requirement that certain instructions may
4684 appear only as part of specified permissible sequences of multiple
4685 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4686 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4687 instruction sequence. For purposes of aligned instruction bundle mode, a
4688 sequence starting with @code{.bundle_lock} and ending with
4689 @code{.bundle_unlock} is treated as a single instruction. That is, the
4690 entire sequence must fit into a single bundle and may not span a bundle
4691 boundary. If necessary, no-op instructions will be inserted before the
4692 first instruction of the sequence so that the whole sequence starts on an
4693 aligned bundle boundary. It's an error if the sequence is longer than the
4696 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4697 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4698 nested. That is, a second @code{.bundle_lock} directive before the next
4699 @code{.bundle_unlock} directive has no effect except that it must be
4700 matched by another closing @code{.bundle_unlock} so that there is the
4701 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4704 @section @code{.byte @var{expressions}}
4706 @cindex @code{byte} directive
4707 @cindex integers, one byte
4708 @code{.byte} expects zero or more expressions, separated by commas.
4709 Each expression is assembled into the next byte.
4711 @node CFI directives
4712 @section CFI directives
4713 @subsection @code{.cfi_sections @var{section_list}}
4714 @cindex @code{cfi_sections} directive
4715 @code{.cfi_sections} may be used to specify whether CFI directives
4716 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4717 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4718 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4719 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4720 directive is not used is @code{.cfi_sections .eh_frame}.
4722 On targets that support compact unwinding tables these can be generated
4723 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4725 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4726 which is used by the @value{TIC6X} target.
4728 The @code{.cfi_sections} directive can be repeated, with the same or different
4729 arguments, provided that CFI generation has not yet started. Once CFI
4730 generation has started however the section list is fixed and any attempts to
4731 redefine it will result in an error.
4733 @subsection @code{.cfi_startproc [simple]}
4734 @cindex @code{cfi_startproc} directive
4735 @code{.cfi_startproc} is used at the beginning of each function that
4736 should have an entry in @code{.eh_frame}. It initializes some internal
4737 data structures. Don't forget to close the function by
4738 @code{.cfi_endproc}.
4740 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4741 it also emits some architecture dependent initial CFI instructions.
4743 @subsection @code{.cfi_endproc}
4744 @cindex @code{cfi_endproc} directive
4745 @code{.cfi_endproc} is used at the end of a function where it closes its
4746 unwind entry previously opened by
4747 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4749 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4750 @cindex @code{cfi_personality} directive
4751 @code{.cfi_personality} defines personality routine and its encoding.
4752 @var{encoding} must be a constant determining how the personality
4753 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4754 argument is not present, otherwise second argument should be
4755 a constant or a symbol name. When using indirect encodings,
4756 the symbol provided should be the location where personality
4757 can be loaded from, not the personality routine itself.
4758 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4759 no personality routine.
4761 @subsection @code{.cfi_personality_id @var{id}}
4762 @cindex @code{cfi_personality_id} directive
4763 @code{cfi_personality_id} defines a personality routine by its index as
4764 defined in a compact unwinding format.
4765 Only valid when generating compact EH frames (i.e.
4766 with @code{.cfi_sections eh_frame_entry}.
4768 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4769 @cindex @code{cfi_fde_data} directive
4770 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4771 used for the current function. These are emitted inline in the
4772 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4773 in the @code{.gnu.extab} section otherwise.
4774 Only valid when generating compact EH frames (i.e.
4775 with @code{.cfi_sections eh_frame_entry}.
4777 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4778 @code{.cfi_lsda} defines LSDA and its encoding.
4779 @var{encoding} must be a constant determining how the LSDA
4780 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4781 argument is not present, otherwise the second argument should be a constant
4782 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4783 meaning that no LSDA is present.
4785 @subsection @code{.cfi_inline_lsda} [@var{align}]
4786 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4787 switches to the corresponding @code{.gnu.extab} section.
4788 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4789 Only valid when generating compact EH frames (i.e.
4790 with @code{.cfi_sections eh_frame_entry}.
4792 The table header and unwinding opcodes will be generated at this point,
4793 so that they are immediately followed by the LSDA data. The symbol
4794 referenced by the @code{.cfi_lsda} directive should still be defined
4795 in case a fallback FDE based encoding is used. The LSDA data is terminated
4796 by a section directive.
4798 The optional @var{align} argument specifies the alignment required.
4799 The alignment is specified as a power of two, as with the
4800 @code{.p2align} directive.
4802 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4803 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4804 address from @var{register} and add @var{offset} to it}.
4806 @subsection @code{.cfi_def_cfa_register @var{register}}
4807 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4808 now on @var{register} will be used instead of the old one. Offset
4811 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4812 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4813 remains the same, but @var{offset} is new. Note that it is the
4814 absolute offset that will be added to a defined register to compute
4817 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4818 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4819 value that is added/subtracted from the previous offset.
4821 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4822 Previous value of @var{register} is saved at offset @var{offset} from
4825 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4826 Previous value of @var{register} is CFA + @var{offset}.
4828 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4829 Previous value of @var{register} is saved at offset @var{offset} from
4830 the current CFA register. This is transformed to @code{.cfi_offset}
4831 using the known displacement of the CFA register from the CFA.
4832 This is often easier to use, because the number will match the
4833 code it's annotating.
4835 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4836 Previous value of @var{register1} is saved in register @var{register2}.
4838 @subsection @code{.cfi_restore @var{register}}
4839 @code{.cfi_restore} says that the rule for @var{register} is now the
4840 same as it was at the beginning of the function, after all initial
4841 instruction added by @code{.cfi_startproc} were executed.
4843 @subsection @code{.cfi_undefined @var{register}}
4844 From now on the previous value of @var{register} can't be restored anymore.
4846 @subsection @code{.cfi_same_value @var{register}}
4847 Current value of @var{register} is the same like in the previous frame,
4848 i.e. no restoration needed.
4850 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4851 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4852 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4853 places them in the current row. This is useful for situations where you have
4854 multiple @code{.cfi_*} directives that need to be undone due to the control
4855 flow of the program. For example, we could have something like this (assuming
4856 the CFA is the value of @code{rbp}):
4866 .cfi_def_cfa %rsp, 8
4869 /* Do something else */
4872 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4873 to the instructions before @code{label}. This means we'd have to add multiple
4874 @code{.cfi} directives after @code{label} to recreate the original save
4875 locations of the registers, as well as setting the CFA back to the value of
4876 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4888 .cfi_def_cfa %rsp, 8
4892 /* Do something else */
4895 That way, the rules for the instructions after @code{label} will be the same
4896 as before the first @code{.cfi_restore} without having to use multiple
4897 @code{.cfi} directives.
4899 @subsection @code{.cfi_return_column @var{register}}
4900 Change return column @var{register}, i.e. the return address is either
4901 directly in @var{register} or can be accessed by rules for @var{register}.
4903 @subsection @code{.cfi_signal_frame}
4904 Mark current function as signal trampoline.
4906 @subsection @code{.cfi_window_save}
4907 SPARC register window has been saved.
4909 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4910 Allows the user to add arbitrary bytes to the unwind info. One
4911 might use this to add OS-specific CFI opcodes, or generic CFI
4912 opcodes that GAS does not yet support.
4914 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4915 The current value of @var{register} is @var{label}. The value of @var{label}
4916 will be encoded in the output file according to @var{encoding}; see the
4917 description of @code{.cfi_personality} for details on this encoding.
4919 The usefulness of equating a register to a fixed label is probably
4920 limited to the return address register. Here, it can be useful to
4921 mark a code segment that has only one return address which is reached
4922 by a direct branch and no copy of the return address exists in memory
4923 or another register.
4926 @section @code{.comm @var{symbol} , @var{length} }
4928 @cindex @code{comm} directive
4929 @cindex symbol, common
4930 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4931 common symbol in one object file may be merged with a defined or common symbol
4932 of the same name in another object file. If @code{@value{LD}} does not see a
4933 definition for the symbol--just one or more common symbols--then it will
4934 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4935 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4936 the same name, and they do not all have the same size, it will allocate space
4937 using the largest size.
4940 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4941 an optional third argument. This is the desired alignment of the symbol,
4942 specified for ELF as a byte boundary (for example, an alignment of 16 means
4943 that the least significant 4 bits of the address should be zero), and for PE
4944 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4945 boundary). The alignment must be an absolute expression, and it must be a
4946 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4947 common symbol, it will use the alignment when placing the symbol. If no
4948 alignment is specified, @command{@value{AS}} will set the alignment to the
4949 largest power of two less than or equal to the size of the symbol, up to a
4950 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4951 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4952 @samp{--section-alignment} option; image file sections in PE are aligned to
4953 multiples of 4096, which is far too large an alignment for ordinary variables.
4954 It is rather the default alignment for (non-debug) sections within object
4955 (@samp{*.o}) files, which are less strictly aligned.}.
4959 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4960 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4964 @section @code{.data @var{subsection}}
4965 @cindex @code{data} directive
4967 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4968 end of the data subsection numbered @var{subsection} (which is an
4969 absolute expression). If @var{subsection} is omitted, it defaults
4973 @section @code{.dc[@var{size}] @var{expressions}}
4974 @cindex @code{dc} directive
4976 The @code{.dc} directive expects zero or more @var{expressions} separated by
4977 commas. These expressions are evaluated and their values inserted into the
4978 current section. The size of the emitted value depends upon the suffix to the
4979 @code{.dc} directive:
4983 Emits N-bit values, where N is the size of an address on the target system.
4987 Emits double precision floating-point values.
4989 Emits 32-bit values.
4991 Emits single precision floating-point values.
4993 Emits 16-bit values.
4994 Note - this is true even on targets where the @code{.word} directive would emit
4997 Emits long double precision floating-point values.
5000 If no suffix is used then @samp{.w} is assumed.
5002 The byte ordering is target dependent, as is the size and format of floating
5006 @section @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
5007 @cindex @code{dcb} directive
5008 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5009 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5010 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5011 @var{size} suffix, if present, must be one of:
5015 Emits single byte values.
5017 Emits double-precision floating point values.
5019 Emits 4-byte values.
5021 Emits single-precision floating point values.
5023 Emits 2-byte values.
5025 Emits long double-precision floating point values.
5028 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5030 The byte ordering is target dependent, as is the size and format of floating
5034 @section @code{.ds[@var{size}] @var{number} [,@var{fill}]}
5035 @cindex @code{ds} directive
5036 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5037 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5038 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5039 @var{size} suffix, if present, must be one of:
5043 Emits single byte values.
5045 Emits 8-byte values.
5047 Emits 4-byte values.
5049 Emits 12-byte values.
5051 Emits 4-byte values.
5053 Emits 2-byte values.
5055 Emits 12-byte values.
5058 Note - unlike the @code{.dcb} directive the @samp{.d}, @samp{.s} and @samp{.x}
5059 suffixes do not indicate that floating-point values are to be inserted.
5061 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5063 The byte ordering is target dependent.
5068 @section @code{.def @var{name}}
5070 @cindex @code{def} directive
5071 @cindex COFF symbols, debugging
5072 @cindex debugging COFF symbols
5073 Begin defining debugging information for a symbol @var{name}; the
5074 definition extends until the @code{.endef} directive is encountered.
5079 @section @code{.desc @var{symbol}, @var{abs-expression}}
5081 @cindex @code{desc} directive
5082 @cindex COFF symbol descriptor
5083 @cindex symbol descriptor, COFF
5084 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5085 to the low 16 bits of an absolute expression.
5088 The @samp{.desc} directive is not available when @command{@value{AS}} is
5089 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5090 object format. For the sake of compatibility, @command{@value{AS}} accepts
5091 it, but produces no output, when configured for COFF.
5097 @section @code{.dim}
5099 @cindex @code{dim} directive
5100 @cindex COFF auxiliary symbol information
5101 @cindex auxiliary symbol information, COFF
5102 This directive is generated by compilers to include auxiliary debugging
5103 information in the symbol table. It is only permitted inside
5104 @code{.def}/@code{.endef} pairs.
5108 @section @code{.double @var{flonums}}
5110 @cindex @code{double} directive
5111 @cindex floating point numbers (double)
5112 @code{.double} expects zero or more flonums, separated by commas. It
5113 assembles floating point numbers.
5115 The exact kind of floating point numbers emitted depends on how
5116 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5120 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5121 in @sc{ieee} format.
5126 @section @code{.eject}
5128 @cindex @code{eject} directive
5129 @cindex new page, in listings
5130 @cindex page, in listings
5131 @cindex listing control: new page
5132 Force a page break at this point, when generating assembly listings.
5135 @section @code{.else}
5137 @cindex @code{else} directive
5138 @code{.else} is part of the @command{@value{AS}} support for conditional
5139 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5140 of code to be assembled if the condition for the preceding @code{.if}
5144 @section @code{.elseif}
5146 @cindex @code{elseif} directive
5147 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5148 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5149 @code{.if} block that would otherwise fill the entire @code{.else} section.
5152 @section @code{.end}
5154 @cindex @code{end} directive
5155 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5156 process anything in the file past the @code{.end} directive.
5160 @section @code{.endef}
5162 @cindex @code{endef} directive
5163 This directive flags the end of a symbol definition begun with
5168 @section @code{.endfunc}
5169 @cindex @code{endfunc} directive
5170 @code{.endfunc} marks the end of a function specified with @code{.func}.
5173 @section @code{.endif}
5175 @cindex @code{endif} directive
5176 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5177 it marks the end of a block of code that is only assembled
5178 conditionally. @xref{If,,@code{.if}}.
5181 @section @code{.equ @var{symbol}, @var{expression}}
5183 @cindex @code{equ} directive
5184 @cindex assigning values to symbols
5185 @cindex symbols, assigning values to
5186 This directive sets the value of @var{symbol} to @var{expression}.
5187 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5190 The syntax for @code{equ} on the HPPA is
5191 @samp{@var{symbol} .equ @var{expression}}.
5195 The syntax for @code{equ} on the Z80 is
5196 @samp{@var{symbol} equ @var{expression}}.
5197 On the Z80 it is an error if @var{symbol} is already defined,
5198 but the symbol is not protected from later redefinition.
5199 Compare @ref{Equiv}.
5203 @section @code{.equiv @var{symbol}, @var{expression}}
5204 @cindex @code{equiv} directive
5205 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5206 the assembler will signal an error if @var{symbol} is already defined. Note a
5207 symbol which has been referenced but not actually defined is considered to be
5210 Except for the contents of the error message, this is roughly equivalent to
5217 plus it protects the symbol from later redefinition.
5220 @section @code{.eqv @var{symbol}, @var{expression}}
5221 @cindex @code{eqv} directive
5222 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5223 evaluate the expression or any part of it immediately. Instead each time
5224 the resulting symbol is used in an expression, a snapshot of its current
5228 @section @code{.err}
5229 @cindex @code{err} directive
5230 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5231 message and, unless the @option{-Z} option was used, it will not generate an
5232 object file. This can be used to signal an error in conditionally compiled code.
5235 @section @code{.error "@var{string}"}
5236 @cindex error directive
5238 Similarly to @code{.err}, this directive emits an error, but you can specify a
5239 string that will be emitted as the error message. If you don't specify the
5240 message, it defaults to @code{".error directive invoked in source file"}.
5241 @xref{Errors, ,Error and Warning Messages}.
5244 .error "This code has not been assembled and tested."
5248 @section @code{.exitm}
5249 Exit early from the current macro definition. @xref{Macro}.
5252 @section @code{.extern}
5254 @cindex @code{extern} directive
5255 @code{.extern} is accepted in the source program---for compatibility
5256 with other assemblers---but it is ignored. @command{@value{AS}} treats
5257 all undefined symbols as external.
5260 @section @code{.fail @var{expression}}
5262 @cindex @code{fail} directive
5263 Generates an error or a warning. If the value of the @var{expression} is 500
5264 or more, @command{@value{AS}} will print a warning message. If the value is less
5265 than 500, @command{@value{AS}} will print an error message. The message will
5266 include the value of @var{expression}. This can occasionally be useful inside
5267 complex nested macros or conditional assembly.
5270 @section @code{.file}
5271 @cindex @code{file} directive
5273 @ifclear no-file-dir
5274 There are two different versions of the @code{.file} directive. Targets
5275 that support DWARF2 line number information use the DWARF2 version of
5276 @code{.file}. Other targets use the default version.
5278 @subheading Default Version
5280 @cindex logical file name
5281 @cindex file name, logical
5282 This version of the @code{.file} directive tells @command{@value{AS}} that we
5283 are about to start a new logical file. The syntax is:
5289 @var{string} is the new file name. In general, the filename is
5290 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5291 to specify an empty file name, you must give the quotes--@code{""}. This
5292 statement may go away in future: it is only recognized to be compatible with
5293 old @command{@value{AS}} programs.
5295 @subheading DWARF2 Version
5298 When emitting DWARF2 line number information, @code{.file} assigns filenames
5299 to the @code{.debug_line} file name table. The syntax is:
5302 .file @var{fileno} @var{filename}
5305 The @var{fileno} operand should be a unique positive integer to use as the
5306 index of the entry in the table. The @var{filename} operand is a C string
5309 The detail of filename indices is exposed to the user because the filename
5310 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5311 information, and thus the user must know the exact indices that table
5315 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5317 @cindex @code{fill} directive
5318 @cindex writing patterns in memory
5319 @cindex patterns, writing in memory
5320 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5321 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5322 may be zero or more. @var{Size} may be zero or more, but if it is
5323 more than 8, then it is deemed to have the value 8, compatible with
5324 other people's assemblers. The contents of each @var{repeat} bytes
5325 is taken from an 8-byte number. The highest order 4 bytes are
5326 zero. The lowest order 4 bytes are @var{value} rendered in the
5327 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5328 Each @var{size} bytes in a repetition is taken from the lowest order
5329 @var{size} bytes of this number. Again, this bizarre behavior is
5330 compatible with other people's assemblers.
5332 @var{size} and @var{value} are optional.
5333 If the second comma and @var{value} are absent, @var{value} is
5334 assumed zero. If the first comma and following tokens are absent,
5335 @var{size} is assumed to be 1.
5338 @section @code{.float @var{flonums}}
5340 @cindex floating point numbers (single)
5341 @cindex @code{float} directive
5342 This directive assembles zero or more flonums, separated by commas. It
5343 has the same effect as @code{.single}.
5345 The exact kind of floating point numbers emitted depends on how
5346 @command{@value{AS}} is configured.
5347 @xref{Machine Dependencies}.
5351 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5352 in @sc{ieee} format.
5357 @section @code{.func @var{name}[,@var{label}]}
5358 @cindex @code{func} directive
5359 @code{.func} emits debugging information to denote function @var{name}, and
5360 is ignored unless the file is assembled with debugging enabled.
5361 Only @samp{--gstabs[+]} is currently supported.
5362 @var{label} is the entry point of the function and if omitted @var{name}
5363 prepended with the @samp{leading char} is used.
5364 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5365 All functions are currently defined to have @code{void} return type.
5366 The function must be terminated with @code{.endfunc}.
5369 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5371 @cindex @code{global} directive
5372 @cindex symbol, making visible to linker
5373 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5374 @var{symbol} in your partial program, its value is made available to
5375 other partial programs that are linked with it. Otherwise,
5376 @var{symbol} takes its attributes from a symbol of the same name
5377 from another file linked into the same program.
5379 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5380 compatibility with other assemblers.
5383 On the HPPA, @code{.global} is not always enough to make it accessible to other
5384 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5385 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5390 @section @code{.gnu_attribute @var{tag},@var{value}}
5391 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5394 @section @code{.hidden @var{names}}
5396 @cindex @code{hidden} directive
5398 This is one of the ELF visibility directives. The other two are
5399 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5400 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5402 This directive overrides the named symbols default visibility (which is set by
5403 their binding: local, global or weak). The directive sets the visibility to
5404 @code{hidden} which means that the symbols are not visible to other components.
5405 Such symbols are always considered to be @code{protected} as well.
5409 @section @code{.hword @var{expressions}}
5411 @cindex @code{hword} directive
5412 @cindex integers, 16-bit
5413 @cindex numbers, 16-bit
5414 @cindex sixteen bit integers
5415 This expects zero or more @var{expressions}, and emits
5416 a 16 bit number for each.
5419 This directive is a synonym for @samp{.short}; depending on the target
5420 architecture, it may also be a synonym for @samp{.word}.
5424 This directive is a synonym for @samp{.short}.
5427 This directive is a synonym for both @samp{.short} and @samp{.word}.
5432 @section @code{.ident}
5434 @cindex @code{ident} directive
5436 This directive is used by some assemblers to place tags in object files. The
5437 behavior of this directive varies depending on the target. When using the
5438 a.out object file format, @command{@value{AS}} simply accepts the directive for
5439 source-file compatibility with existing assemblers, but does not emit anything
5440 for it. When using COFF, comments are emitted to the @code{.comment} or
5441 @code{.rdata} section, depending on the target. When using ELF, comments are
5442 emitted to the @code{.comment} section.
5445 @section @code{.if @var{absolute expression}}
5447 @cindex conditional assembly
5448 @cindex @code{if} directive
5449 @code{.if} marks the beginning of a section of code which is only
5450 considered part of the source program being assembled if the argument
5451 (which must be an @var{absolute expression}) is non-zero. The end of
5452 the conditional section of code must be marked by @code{.endif}
5453 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5454 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5455 If you have several conditions to check, @code{.elseif} may be used to avoid
5456 nesting blocks if/else within each subsequent @code{.else} block.
5458 The following variants of @code{.if} are also supported:
5460 @cindex @code{ifdef} directive
5461 @item .ifdef @var{symbol}
5462 Assembles the following section of code if the specified @var{symbol}
5463 has been defined. Note a symbol which has been referenced but not yet defined
5464 is considered to be undefined.
5466 @cindex @code{ifb} directive
5467 @item .ifb @var{text}
5468 Assembles the following section of code if the operand is blank (empty).
5470 @cindex @code{ifc} directive
5471 @item .ifc @var{string1},@var{string2}
5472 Assembles the following section of code if the two strings are the same. The
5473 strings may be optionally quoted with single quotes. If they are not quoted,
5474 the first string stops at the first comma, and the second string stops at the
5475 end of the line. Strings which contain whitespace should be quoted. The
5476 string comparison is case sensitive.
5478 @cindex @code{ifeq} directive
5479 @item .ifeq @var{absolute expression}
5480 Assembles the following section of code if the argument is zero.
5482 @cindex @code{ifeqs} directive
5483 @item .ifeqs @var{string1},@var{string2}
5484 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5486 @cindex @code{ifge} directive
5487 @item .ifge @var{absolute expression}
5488 Assembles the following section of code if the argument is greater than or
5491 @cindex @code{ifgt} directive
5492 @item .ifgt @var{absolute expression}
5493 Assembles the following section of code if the argument is greater than zero.
5495 @cindex @code{ifle} directive
5496 @item .ifle @var{absolute expression}
5497 Assembles the following section of code if the argument is less than or equal
5500 @cindex @code{iflt} directive
5501 @item .iflt @var{absolute expression}
5502 Assembles the following section of code if the argument is less than zero.
5504 @cindex @code{ifnb} directive
5505 @item .ifnb @var{text}
5506 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5507 following section of code if the operand is non-blank (non-empty).
5509 @cindex @code{ifnc} directive
5510 @item .ifnc @var{string1},@var{string2}.
5511 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5512 following section of code if the two strings are not the same.
5514 @cindex @code{ifndef} directive
5515 @cindex @code{ifnotdef} directive
5516 @item .ifndef @var{symbol}
5517 @itemx .ifnotdef @var{symbol}
5518 Assembles the following section of code if the specified @var{symbol}
5519 has not been defined. Both spelling variants are equivalent. Note a symbol
5520 which has been referenced but not yet defined is considered to be undefined.
5522 @cindex @code{ifne} directive
5523 @item .ifne @var{absolute expression}
5524 Assembles the following section of code if the argument is not equal to zero
5525 (in other words, this is equivalent to @code{.if}).
5527 @cindex @code{ifnes} directive
5528 @item .ifnes @var{string1},@var{string2}
5529 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5530 following section of code if the two strings are not the same.
5534 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5536 @cindex @code{incbin} directive
5537 @cindex binary files, including
5538 The @code{incbin} directive includes @var{file} verbatim at the current
5539 location. You can control the search paths used with the @samp{-I} command-line
5540 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5543 The @var{skip} argument skips a number of bytes from the start of the
5544 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5545 read. Note that the data is not aligned in any way, so it is the user's
5546 responsibility to make sure that proper alignment is provided both before and
5547 after the @code{incbin} directive.
5550 @section @code{.include "@var{file}"}
5552 @cindex @code{include} directive
5553 @cindex supporting files, including
5554 @cindex files, including
5555 This directive provides a way to include supporting files at specified
5556 points in your source program. The code from @var{file} is assembled as
5557 if it followed the point of the @code{.include}; when the end of the
5558 included file is reached, assembly of the original file continues. You
5559 can control the search paths used with the @samp{-I} command-line option
5560 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5564 @section @code{.int @var{expressions}}
5566 @cindex @code{int} directive
5567 @cindex integers, 32-bit
5568 Expect zero or more @var{expressions}, of any section, separated by commas.
5569 For each expression, emit a number that, at run time, is the value of that
5570 expression. The byte order and bit size of the number depends on what kind
5571 of target the assembly is for.
5575 On most forms of the H8/300, @code{.int} emits 16-bit
5576 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5583 @section @code{.internal @var{names}}
5585 @cindex @code{internal} directive
5587 This is one of the ELF visibility directives. The other two are
5588 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5589 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5591 This directive overrides the named symbols default visibility (which is set by
5592 their binding: local, global or weak). The directive sets the visibility to
5593 @code{internal} which means that the symbols are considered to be @code{hidden}
5594 (i.e., not visible to other components), and that some extra, processor specific
5595 processing must also be performed upon the symbols as well.
5599 @section @code{.irp @var{symbol},@var{values}}@dots{}
5601 @cindex @code{irp} directive
5602 Evaluate a sequence of statements assigning different values to @var{symbol}.
5603 The sequence of statements starts at the @code{.irp} directive, and is
5604 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5605 set to @var{value}, and the sequence of statements is assembled. If no
5606 @var{value} is listed, the sequence of statements is assembled once, with
5607 @var{symbol} set to the null string. To refer to @var{symbol} within the
5608 sequence of statements, use @var{\symbol}.
5610 For example, assembling
5618 is equivalent to assembling
5626 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5629 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5631 @cindex @code{irpc} directive
5632 Evaluate a sequence of statements assigning different values to @var{symbol}.
5633 The sequence of statements starts at the @code{.irpc} directive, and is
5634 terminated by an @code{.endr} directive. For each character in @var{value},
5635 @var{symbol} is set to the character, and the sequence of statements is
5636 assembled. If no @var{value} is listed, the sequence of statements is
5637 assembled once, with @var{symbol} set to the null string. To refer to
5638 @var{symbol} within the sequence of statements, use @var{\symbol}.
5640 For example, assembling
5648 is equivalent to assembling
5656 For some caveats with the spelling of @var{symbol}, see also the discussion
5660 @section @code{.lcomm @var{symbol} , @var{length}}
5662 @cindex @code{lcomm} directive
5663 @cindex local common symbols
5664 @cindex symbols, local common
5665 Reserve @var{length} (an absolute expression) bytes for a local common
5666 denoted by @var{symbol}. The section and value of @var{symbol} are
5667 those of the new local common. The addresses are allocated in the bss
5668 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5669 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5670 not visible to @code{@value{LD}}.
5673 Some targets permit a third argument to be used with @code{.lcomm}. This
5674 argument specifies the desired alignment of the symbol in the bss section.
5678 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5679 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5683 @section @code{.lflags}
5685 @cindex @code{lflags} directive (ignored)
5686 @command{@value{AS}} accepts this directive, for compatibility with other
5687 assemblers, but ignores it.
5689 @ifclear no-line-dir
5691 @section @code{.line @var{line-number}}
5693 @cindex @code{line} directive
5694 @cindex logical line number
5696 Change the logical line number. @var{line-number} must be an absolute
5697 expression. The next line has that logical line number. Therefore any other
5698 statements on the current line (after a statement separator character) are
5699 reported as on logical line number @var{line-number} @minus{} 1. One day
5700 @command{@value{AS}} will no longer support this directive: it is recognized only
5701 for compatibility with existing assembler programs.
5704 Even though this is a directive associated with the @code{a.out} or
5705 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5706 when producing COFF output, and treats @samp{.line} as though it
5707 were the COFF @samp{.ln} @emph{if} it is found outside a
5708 @code{.def}/@code{.endef} pair.
5710 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5711 used by compilers to generate auxiliary symbol information for
5716 @section @code{.linkonce [@var{type}]}
5718 @cindex @code{linkonce} directive
5719 @cindex common sections
5720 Mark the current section so that the linker only includes a single copy of it.
5721 This may be used to include the same section in several different object files,
5722 but ensure that the linker will only include it once in the final output file.
5723 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5724 Duplicate sections are detected based on the section name, so it should be
5727 This directive is only supported by a few object file formats; as of this
5728 writing, the only object file format which supports it is the Portable
5729 Executable format used on Windows NT.
5731 The @var{type} argument is optional. If specified, it must be one of the
5732 following strings. For example:
5736 Not all types may be supported on all object file formats.
5740 Silently discard duplicate sections. This is the default.
5743 Warn if there are duplicate sections, but still keep only one copy.
5746 Warn if any of the duplicates have different sizes.
5749 Warn if any of the duplicates do not have exactly the same contents.
5753 @section @code{.list}
5755 @cindex @code{list} directive
5756 @cindex listing control, turning on
5757 Control (in conjunction with the @code{.nolist} directive) whether or
5758 not assembly listings are generated. These two directives maintain an
5759 internal counter (which is zero initially). @code{.list} increments the
5760 counter, and @code{.nolist} decrements it. Assembly listings are
5761 generated whenever the counter is greater than zero.
5763 By default, listings are disabled. When you enable them (with the
5764 @samp{-a} command-line option; @pxref{Invoking,,Command-Line Options}),
5765 the initial value of the listing counter is one.
5768 @section @code{.ln @var{line-number}}
5770 @cindex @code{ln} directive
5771 @ifclear no-line-dir
5772 @samp{.ln} is a synonym for @samp{.line}.
5775 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5776 must be an absolute expression. The next line has that logical
5777 line number, so any other statements on the current line (after a
5778 statement separator character @code{;}) are reported as on logical
5779 line number @var{line-number} @minus{} 1.
5783 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5784 @cindex @code{loc} directive
5785 When emitting DWARF2 line number information,
5786 the @code{.loc} directive will add a row to the @code{.debug_line} line
5787 number matrix corresponding to the immediately following assembly
5788 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5789 arguments will be applied to the @code{.debug_line} state machine before
5792 The @var{options} are a sequence of the following tokens in any order:
5796 This option will set the @code{basic_block} register in the
5797 @code{.debug_line} state machine to @code{true}.
5800 This option will set the @code{prologue_end} register in the
5801 @code{.debug_line} state machine to @code{true}.
5803 @item epilogue_begin
5804 This option will set the @code{epilogue_begin} register in the
5805 @code{.debug_line} state machine to @code{true}.
5807 @item is_stmt @var{value}
5808 This option will set the @code{is_stmt} register in the
5809 @code{.debug_line} state machine to @code{value}, which must be
5812 @item isa @var{value}
5813 This directive will set the @code{isa} register in the @code{.debug_line}
5814 state machine to @var{value}, which must be an unsigned integer.
5816 @item discriminator @var{value}
5817 This directive will set the @code{discriminator} register in the @code{.debug_line}
5818 state machine to @var{value}, which must be an unsigned integer.
5820 @item view @var{value}
5821 This option causes a row to be added to @code{.debug_line} in reference to the
5822 current address (which might not be the same as that of the following assembly
5823 instruction), and to associate @var{value} with the @code{view} register in the
5824 @code{.debug_line} state machine. If @var{value} is a label, both the
5825 @code{view} register and the label are set to the number of prior @code{.loc}
5826 directives at the same program location. If @var{value} is the literal
5827 @code{0}, the @code{view} register is set to zero, and the assembler asserts
5828 that there aren't any prior @code{.loc} directives at the same program
5829 location. If @var{value} is the literal @code{-0}, the assembler arrange for
5830 the @code{view} register to be reset in this row, even if there are prior
5831 @code{.loc} directives at the same program location.
5835 @node Loc_mark_labels
5836 @section @code{.loc_mark_labels @var{enable}}
5837 @cindex @code{loc_mark_labels} directive
5838 When emitting DWARF2 line number information,
5839 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5840 to the @code{.debug_line} line number matrix with the @code{basic_block}
5841 register in the state machine set whenever a code label is seen.
5842 The @var{enable} argument should be either 1 or 0, to enable or disable
5843 this function respectively.
5847 @section @code{.local @var{names}}
5849 @cindex @code{local} directive
5850 This directive, which is available for ELF targets, marks each symbol in
5851 the comma-separated list of @code{names} as a local symbol so that it
5852 will not be externally visible. If the symbols do not already exist,
5853 they will be created.
5855 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5856 accept an alignment argument, which is the case for most ELF targets,
5857 the @code{.local} directive can be used in combination with @code{.comm}
5858 (@pxref{Comm}) to define aligned local common data.
5862 @section @code{.long @var{expressions}}
5864 @cindex @code{long} directive
5865 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5868 @c no one seems to know what this is for or whether this description is
5869 @c what it really ought to do
5871 @section @code{.lsym @var{symbol}, @var{expression}}
5873 @cindex @code{lsym} directive
5874 @cindex symbol, not referenced in assembly
5875 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5876 the hash table, ensuring it cannot be referenced by name during the
5877 rest of the assembly. This sets the attributes of the symbol to be
5878 the same as the expression value:
5880 @var{other} = @var{descriptor} = 0
5881 @var{type} = @r{(section of @var{expression})}
5882 @var{value} = @var{expression}
5885 The new symbol is not flagged as external.
5889 @section @code{.macro}
5892 The commands @code{.macro} and @code{.endm} allow you to define macros that
5893 generate assembly output. For example, this definition specifies a macro
5894 @code{sum} that puts a sequence of numbers into memory:
5897 .macro sum from=0, to=5
5906 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5918 @item .macro @var{macname}
5919 @itemx .macro @var{macname} @var{macargs} @dots{}
5920 @cindex @code{macro} directive
5921 Begin the definition of a macro called @var{macname}. If your macro
5922 definition requires arguments, specify their names after the macro name,
5923 separated by commas or spaces. You can qualify the macro argument to
5924 indicate whether all invocations must specify a non-blank value (through
5925 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5926 (through @samp{:@code{vararg}}). You can supply a default value for any
5927 macro argument by following the name with @samp{=@var{deflt}}. You
5928 cannot define two macros with the same @var{macname} unless it has been
5929 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5930 definitions. For example, these are all valid @code{.macro} statements:
5934 Begin the definition of a macro called @code{comm}, which takes no
5937 @item .macro plus1 p, p1
5938 @itemx .macro plus1 p p1
5939 Either statement begins the definition of a macro called @code{plus1},
5940 which takes two arguments; within the macro definition, write
5941 @samp{\p} or @samp{\p1} to evaluate the arguments.
5943 @item .macro reserve_str p1=0 p2
5944 Begin the definition of a macro called @code{reserve_str}, with two
5945 arguments. The first argument has a default value, but not the second.
5946 After the definition is complete, you can call the macro either as
5947 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5948 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5949 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5950 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5952 @item .macro m p1:req, p2=0, p3:vararg
5953 Begin the definition of a macro called @code{m}, with at least three
5954 arguments. The first argument must always have a value specified, but
5955 not the second, which instead has a default value. The third formal
5956 will get assigned all remaining arguments specified at invocation time.
5958 When you call a macro, you can specify the argument values either by
5959 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5960 @samp{sum to=17, from=9}.
5964 Note that since each of the @var{macargs} can be an identifier exactly
5965 as any other one permitted by the target architecture, there may be
5966 occasional problems if the target hand-crafts special meanings to certain
5967 characters when they occur in a special position. For example, if the colon
5968 (@code{:}) is generally permitted to be part of a symbol name, but the
5969 architecture specific code special-cases it when occurring as the final
5970 character of a symbol (to denote a label), then the macro parameter
5971 replacement code will have no way of knowing that and consider the whole
5972 construct (including the colon) an identifier, and check only this
5973 identifier for being the subject to parameter substitution. So for example
5974 this macro definition:
5982 might not work as expected. Invoking @samp{label foo} might not create a label
5983 called @samp{foo} but instead just insert the text @samp{\l:} into the
5984 assembler source, probably generating an error about an unrecognised
5987 Similarly problems might occur with the period character (@samp{.})
5988 which is often allowed inside opcode names (and hence identifier names). So
5989 for example constructing a macro to build an opcode from a base name and a
5990 length specifier like this:
5993 .macro opcode base length
5998 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5999 instruction but instead generate some kind of error as the assembler tries to
6000 interpret the text @samp{\base.\length}.
6002 There are several possible ways around this problem:
6005 @item Insert white space
6006 If it is possible to use white space characters then this is the simplest
6015 @item Use @samp{\()}
6016 The string @samp{\()} can be used to separate the end of a macro argument from
6017 the following text. eg:
6020 .macro opcode base length
6025 @item Use the alternate macro syntax mode
6026 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
6027 used as a separator. eg:
6037 Note: this problem of correctly identifying string parameters to pseudo ops
6038 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
6039 and @code{.irpc} (@pxref{Irpc}) as well.
6042 @cindex @code{endm} directive
6043 Mark the end of a macro definition.
6046 @cindex @code{exitm} directive
6047 Exit early from the current macro definition.
6049 @cindex number of macros executed
6050 @cindex macros, count executed
6052 @command{@value{AS}} maintains a counter of how many macros it has
6053 executed in this pseudo-variable; you can copy that number to your
6054 output with @samp{\@@}, but @emph{only within a macro definition}.
6056 @item LOCAL @var{name} [ , @dots{} ]
6057 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
6058 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
6059 @xref{Altmacro,,@code{.altmacro}}.
6063 @section @code{.mri @var{val}}
6065 @cindex @code{mri} directive
6066 @cindex MRI mode, temporarily
6067 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
6068 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
6069 affects code assembled until the next @code{.mri} directive, or until the end
6070 of the file. @xref{M, MRI mode, MRI mode}.
6073 @section @code{.noaltmacro}
6074 Disable alternate macro mode. @xref{Altmacro}.
6077 @section @code{.nolist}
6079 @cindex @code{nolist} directive
6080 @cindex listing control, turning off
6081 Control (in conjunction with the @code{.list} directive) whether or
6082 not assembly listings are generated. These two directives maintain an
6083 internal counter (which is zero initially). @code{.list} increments the
6084 counter, and @code{.nolist} decrements it. Assembly listings are
6085 generated whenever the counter is greater than zero.
6088 @section @code{.nops @var{size}[, @var{control}]}
6090 @cindex @code{nops} directive
6091 @cindex filling memory with no-op instructions
6092 This directive emits @var{size} bytes filled with no-op instructions.
6093 @var{size} is absolute expression, which must be a positve value.
6094 @var{control} controls how no-op instructions should be generated. If
6095 the comma and @var{control} are omitted, @var{control} is assumed to be
6098 Note: For Intel 80386 and AMD x86-64 targets, @var{control} specifies
6099 the size limit of a no-op instruction. The valid values of @var{control}
6100 are between 0 and 4 in 16-bit mode, between 0 and 7 when tuning for
6101 older processors in 32-bit mode, between 0 and 11 in 64-bit mode or when
6102 tuning for newer processors in 32-bit mode. When 0 is used, the no-op
6103 instruction size limit is set to the maximum supported size.
6106 @section @code{.octa @var{bignums}}
6108 @c FIXME: double size emitted for "octa" on some? Or warn?
6109 @cindex @code{octa} directive
6110 @cindex integer, 16-byte
6111 @cindex sixteen byte integer
6112 This directive expects zero or more bignums, separated by commas. For each
6113 bignum, it emits a 16-byte integer.
6115 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6116 hence @emph{octa}-word for 16 bytes.
6119 @section @code{.offset @var{loc}}
6121 @cindex @code{offset} directive
6122 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6123 be an absolute expression. This directive may be useful for defining
6124 symbols with absolute values. Do not confuse it with the @code{.org}
6128 @section @code{.org @var{new-lc} , @var{fill}}
6130 @cindex @code{org} directive
6131 @cindex location counter, advancing
6132 @cindex advancing location counter
6133 @cindex current address, advancing
6134 Advance the location counter of the current section to
6135 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6136 expression with the same section as the current subsection. That is,
6137 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6138 wrong section, the @code{.org} directive is ignored. To be compatible
6139 with former assemblers, if the section of @var{new-lc} is absolute,
6140 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6141 is the same as the current subsection.
6143 @code{.org} may only increase the location counter, or leave it
6144 unchanged; you cannot use @code{.org} to move the location counter
6147 @c double negative used below "not undefined" because this is a specific
6148 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6149 @c section. doc@cygnus.com 18feb91
6150 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6151 may not be undefined. If you really detest this restriction we eagerly await
6152 a chance to share your improved assembler.
6154 Beware that the origin is relative to the start of the section, not
6155 to the start of the subsection. This is compatible with other
6156 people's assemblers.
6158 When the location counter (of the current subsection) is advanced, the
6159 intervening bytes are filled with @var{fill} which should be an
6160 absolute expression. If the comma and @var{fill} are omitted,
6161 @var{fill} defaults to zero.
6164 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
6166 @cindex padding the location counter given a power of two
6167 @cindex @code{p2align} directive
6168 Pad the location counter (in the current subsection) to a particular
6169 storage boundary. The first expression (which must be absolute) is the
6170 number of low-order zero bits the location counter must have after
6171 advancement. For example @samp{.p2align 3} advances the location
6172 counter until it a multiple of 8. If the location counter is already a
6173 multiple of 8, no change is needed.
6175 The second expression (also absolute) gives the fill value to be stored in the
6176 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6177 padding bytes are normally zero. However, on most systems, if the section is
6178 marked as containing code and the fill value is omitted, the space is filled
6179 with no-op instructions.
6181 The third expression is also absolute, and is also optional. If it is present,
6182 it is the maximum number of bytes that should be skipped by this alignment
6183 directive. If doing the alignment would require skipping more bytes than the
6184 specified maximum, then the alignment is not done at all. You can omit the
6185 fill value (the second argument) entirely by simply using two commas after the
6186 required alignment; this can be useful if you want the alignment to be filled
6187 with no-op instructions when appropriate.
6189 @cindex @code{p2alignw} directive
6190 @cindex @code{p2alignl} directive
6191 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6192 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6193 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6194 fill pattern as a four byte longword value. For example, @code{.p2alignw
6195 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6196 filled in with the value 0x368d (the exact placement of the bytes depends upon
6197 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6202 @section @code{.popsection}
6204 @cindex @code{popsection} directive
6205 @cindex Section Stack
6206 This is one of the ELF section stack manipulation directives. The others are
6207 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6208 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6211 This directive replaces the current section (and subsection) with the top
6212 section (and subsection) on the section stack. This section is popped off the
6218 @section @code{.previous}
6220 @cindex @code{previous} directive
6221 @cindex Section Stack
6222 This is one of the ELF section stack manipulation directives. The others are
6223 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6224 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6225 (@pxref{PopSection}).
6227 This directive swaps the current section (and subsection) with most recently
6228 referenced section/subsection pair prior to this one. Multiple
6229 @code{.previous} directives in a row will flip between two sections (and their
6230 subsections). For example:
6242 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6248 # Now in section A subsection 1
6252 # Now in section B subsection 0
6255 # Now in section B subsection 1
6258 # Now in section B subsection 0
6262 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6263 section B and 0x9abc into subsection 1 of section B.
6265 In terms of the section stack, this directive swaps the current section with
6266 the top section on the section stack.
6270 @section @code{.print @var{string}}
6272 @cindex @code{print} directive
6273 @command{@value{AS}} will print @var{string} on the standard output during
6274 assembly. You must put @var{string} in double quotes.
6278 @section @code{.protected @var{names}}
6280 @cindex @code{protected} directive
6282 This is one of the ELF visibility directives. The other two are
6283 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6285 This directive overrides the named symbols default visibility (which is set by
6286 their binding: local, global or weak). The directive sets the visibility to
6287 @code{protected} which means that any references to the symbols from within the
6288 components that defines them must be resolved to the definition in that
6289 component, even if a definition in another component would normally preempt
6294 @section @code{.psize @var{lines} , @var{columns}}
6296 @cindex @code{psize} directive
6297 @cindex listing control: paper size
6298 @cindex paper size, for listings
6299 Use this directive to declare the number of lines---and, optionally, the
6300 number of columns---to use for each page, when generating listings.
6302 If you do not use @code{.psize}, listings use a default line-count
6303 of 60. You may omit the comma and @var{columns} specification; the
6304 default width is 200 columns.
6306 @command{@value{AS}} generates formfeeds whenever the specified number of
6307 lines is exceeded (or whenever you explicitly request one, using
6310 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6311 those explicitly specified with @code{.eject}.
6314 @section @code{.purgem @var{name}}
6316 @cindex @code{purgem} directive
6317 Undefine the macro @var{name}, so that later uses of the string will not be
6318 expanded. @xref{Macro}.
6322 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6324 @cindex @code{pushsection} directive
6325 @cindex Section Stack
6326 This is one of the ELF section stack manipulation directives. The others are
6327 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6328 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6331 This directive pushes the current section (and subsection) onto the
6332 top of the section stack, and then replaces the current section and
6333 subsection with @code{name} and @code{subsection}. The optional
6334 @code{flags}, @code{type} and @code{arguments} are treated the same
6335 as in the @code{.section} (@pxref{Section}) directive.
6339 @section @code{.quad @var{bignums}}
6341 @cindex @code{quad} directive
6342 @code{.quad} expects zero or more bignums, separated by commas. For
6343 each bignum, it emits
6345 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6346 warning message; and just takes the lowest order 8 bytes of the bignum.
6347 @cindex eight-byte integer
6348 @cindex integer, 8-byte
6350 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6351 hence @emph{quad}-word for 8 bytes.
6354 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6355 warning message; and just takes the lowest order 16 bytes of the bignum.
6356 @cindex sixteen-byte integer
6357 @cindex integer, 16-byte
6361 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6363 @cindex @code{reloc} directive
6364 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6365 @var{expression}. If @var{offset} is a number, the relocation is generated in
6366 the current section. If @var{offset} is an expression that resolves to a
6367 symbol plus offset, the relocation is generated in the given symbol's section.
6368 @var{expression}, if present, must resolve to a symbol plus addend or to an
6369 absolute value, but note that not all targets support an addend. e.g. ELF REL
6370 targets such as i386 store an addend in the section contents rather than in the
6371 relocation. This low level interface does not support addends stored in the
6375 @section @code{.rept @var{count}}
6377 @cindex @code{rept} directive
6378 Repeat the sequence of lines between the @code{.rept} directive and the next
6379 @code{.endr} directive @var{count} times.
6381 For example, assembling
6389 is equivalent to assembling
6397 A count of zero is allowed, but nothing is generated. Negative counts are not
6398 allowed and if encountered will be treated as if they were zero.
6401 @section @code{.sbttl "@var{subheading}"}
6403 @cindex @code{sbttl} directive
6404 @cindex subtitles for listings
6405 @cindex listing control: subtitle
6406 Use @var{subheading} as the title (third line, immediately after the
6407 title line) when generating assembly listings.
6409 This directive affects subsequent pages, as well as the current page if
6410 it appears within ten lines of the top of a page.
6414 @section @code{.scl @var{class}}
6416 @cindex @code{scl} directive
6417 @cindex symbol storage class (COFF)
6418 @cindex COFF symbol storage class
6419 Set the storage-class value for a symbol. This directive may only be
6420 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6421 whether a symbol is static or external, or it may record further
6422 symbolic debugging information.
6427 @section @code{.section @var{name}}
6429 @cindex named section
6430 Use the @code{.section} directive to assemble the following code into a section
6433 This directive is only supported for targets that actually support arbitrarily
6434 named sections; on @code{a.out} targets, for example, it is not accepted, even
6435 with a standard @code{a.out} section name.
6439 @c only print the extra heading if both COFF and ELF are set
6440 @subheading COFF Version
6443 @cindex @code{section} directive (COFF version)
6444 For COFF targets, the @code{.section} directive is used in one of the following
6448 .section @var{name}[, "@var{flags}"]
6449 .section @var{name}[, @var{subsection}]
6452 If the optional argument is quoted, it is taken as flags to use for the
6453 section. Each flag is a single character. The following flags are recognized:
6457 bss section (uninitialized data)
6459 section is not loaded
6465 exclude section from linking
6471 shared section (meaningful for PE targets)
6473 ignored. (For compatibility with the ELF version)
6475 section is not readable (meaningful for PE targets)
6477 single-digit power-of-two section alignment (GNU extension)
6480 If no flags are specified, the default flags depend upon the section name. If
6481 the section name is not recognized, the default will be for the section to be
6482 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6483 from the section, rather than adding them, so if they are used on their own it
6484 will be as if no flags had been specified at all.
6486 If the optional argument to the @code{.section} directive is not quoted, it is
6487 taken as a subsection number (@pxref{Sub-Sections}).
6492 @c only print the extra heading if both COFF and ELF are set
6493 @subheading ELF Version
6496 @cindex Section Stack
6497 This is one of the ELF section stack manipulation directives. The others are
6498 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6499 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6500 @code{.previous} (@pxref{Previous}).
6502 @cindex @code{section} directive (ELF version)
6503 For ELF targets, the @code{.section} directive is used like this:
6506 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6509 @anchor{Section Name Substitutions}
6510 @kindex --sectname-subst
6511 @cindex section name substitution
6512 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6513 argument may contain a substitution sequence. Only @code{%S} is supported
6514 at the moment, and substitutes the current section name. For example:
6517 .macro exception_code
6518 .section %S.exception
6519 [exception code here]
6534 The two @code{exception_code} invocations above would create the
6535 @code{.text.exception} and @code{.init.exception} sections respectively.
6536 This is useful e.g. to discriminate between ancillary sections that are
6537 tied to setup code to be discarded after use from ancillary sections that
6538 need to stay resident without having to define multiple @code{exception_code}
6539 macros just for that purpose.
6541 The optional @var{flags} argument is a quoted string which may contain any
6542 combination of the following characters:
6546 section is allocatable
6548 section is a GNU_MBIND section
6550 section is excluded from executable and shared library.
6554 section is executable
6556 section is mergeable
6558 section contains zero terminated strings
6560 section is a member of a section group
6562 section is used for thread-local-storage
6564 section is a member of the previously-current section's group, if any
6565 @item @code{<number>}
6566 a numeric value indicating the bits to be set in the ELF section header's flags
6567 field. Note - if one or more of the alphabetic characters described above is
6568 also included in the flags field, their bit values will be ORed into the
6570 @item @code{<target specific>}
6571 some targets extend this list with their own flag characters
6574 Note - once a section's flags have been set they cannot be changed. There are
6575 a few exceptions to this rule however. Processor and application specific
6576 flags can be added to an already defined section. The @code{.interp},
6577 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6578 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6579 section may have the executable (@code{x}) flag added.
6581 The optional @var{type} argument may contain one of the following constants:
6585 section contains data
6587 section does not contain data (i.e., section only occupies space)
6589 section contains data which is used by things other than the program
6591 section contains an array of pointers to init functions
6593 section contains an array of pointers to finish functions
6594 @item @@preinit_array
6595 section contains an array of pointers to pre-init functions
6596 @item @@@code{<number>}
6597 a numeric value to be set as the ELF section header's type field.
6598 @item @@@code{<target specific>}
6599 some targets extend this list with their own types
6602 Many targets only support the first three section types. The type may be
6603 enclosed in double quotes if necessary.
6605 Note on targets where the @code{@@} character is the start of a comment (eg
6606 ARM) then another character is used instead. For example the ARM port uses the
6609 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6610 special and have fixed types. Any attempt to declare them with a different
6611 type will generate an error from the assembler.
6613 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6614 be specified as well as an extra argument---@var{entsize}---like this:
6617 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6620 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6621 constants, each @var{entsize} octets long. Sections with both @code{M} and
6622 @code{S} must contain zero terminated strings where each character is
6623 @var{entsize} bytes long. The linker may remove duplicates within sections with
6624 the same name, same entity size and same flags. @var{entsize} must be an
6625 absolute expression. For sections with both @code{M} and @code{S}, a string
6626 which is a suffix of a larger string is considered a duplicate. Thus
6627 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6628 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6630 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6631 be present along with an additional field like this:
6634 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6637 The @var{GroupName} field specifies the name of the section group to which this
6638 particular section belongs. The optional linkage field can contain:
6642 indicates that only one copy of this section should be retained
6647 Note: if both the @var{M} and @var{G} flags are present then the fields for
6648 the Merge flag should come first, like this:
6651 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6654 If @var{flags} contains the @code{?} symbol then it may not also contain the
6655 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6656 present. Instead, @code{?} says to consider the section that's current before
6657 this directive. If that section used @code{G}, then the new section will use
6658 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6659 If not, then the @code{?} symbol has no effect.
6661 If no flags are specified, the default flags depend upon the section name. If
6662 the section name is not recognized, the default will be for the section to have
6663 none of the above flags: it will not be allocated in memory, nor writable, nor
6664 executable. The section will contain data.
6666 For ELF targets, the assembler supports another type of @code{.section}
6667 directive for compatibility with the Solaris assembler:
6670 .section "@var{name}"[, @var{flags}...]
6673 Note that the section name is quoted. There may be a sequence of comma
6678 section is allocatable
6682 section is executable
6684 section is excluded from executable and shared library.
6686 section is used for thread local storage
6689 This directive replaces the current section and subsection. See the
6690 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6691 some examples of how this directive and the other section stack directives
6697 @section @code{.set @var{symbol}, @var{expression}}
6699 @cindex @code{set} directive
6700 @cindex symbol value, setting
6701 Set the value of @var{symbol} to @var{expression}. This
6702 changes @var{symbol}'s value and type to conform to
6703 @var{expression}. If @var{symbol} was flagged as external, it remains
6704 flagged (@pxref{Symbol Attributes}).
6706 You may @code{.set} a symbol many times in the same assembly provided that the
6707 values given to the symbol are constants. Values that are based on expressions
6708 involving other symbols are allowed, but some targets may restrict this to only
6709 being done once per assembly. This is because those targets do not set the
6710 addresses of symbols at assembly time, but rather delay the assignment until a
6711 final link is performed. This allows the linker a chance to change the code in
6712 the files, changing the location of, and the relative distance between, various
6715 If you @code{.set} a global symbol, the value stored in the object
6716 file is the last value stored into it.
6719 On Z80 @code{set} is a real instruction, use
6720 @samp{@var{symbol} defl @var{expression}} instead.
6724 @section @code{.short @var{expressions}}
6726 @cindex @code{short} directive
6728 @code{.short} is normally the same as @samp{.word}.
6729 @xref{Word,,@code{.word}}.
6731 In some configurations, however, @code{.short} and @code{.word} generate
6732 numbers of different lengths. @xref{Machine Dependencies}.
6736 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6739 This expects zero or more @var{expressions}, and emits
6740 a 16 bit number for each.
6745 @section @code{.single @var{flonums}}
6747 @cindex @code{single} directive
6748 @cindex floating point numbers (single)
6749 This directive assembles zero or more flonums, separated by commas. It
6750 has the same effect as @code{.float}.
6752 The exact kind of floating point numbers emitted depends on how
6753 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6757 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6758 numbers in @sc{ieee} format.
6764 @section @code{.size}
6766 This directive is used to set the size associated with a symbol.
6770 @c only print the extra heading if both COFF and ELF are set
6771 @subheading COFF Version
6774 @cindex @code{size} directive (COFF version)
6775 For COFF targets, the @code{.size} directive is only permitted inside
6776 @code{.def}/@code{.endef} pairs. It is used like this:
6779 .size @var{expression}
6786 @c only print the extra heading if both COFF and ELF are set
6787 @subheading ELF Version
6790 @cindex @code{size} directive (ELF version)
6791 For ELF targets, the @code{.size} directive is used like this:
6794 .size @var{name} , @var{expression}
6797 This directive sets the size associated with a symbol @var{name}.
6798 The size in bytes is computed from @var{expression} which can make use of label
6799 arithmetic. This directive is typically used to set the size of function
6804 @ifclear no-space-dir
6806 @section @code{.skip @var{size} [,@var{fill}]}
6808 @cindex @code{skip} directive
6809 @cindex filling memory
6810 This directive emits @var{size} bytes, each of value @var{fill}. Both
6811 @var{size} and @var{fill} are absolute expressions. If the comma and
6812 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6817 @section @code{.sleb128 @var{expressions}}
6819 @cindex @code{sleb128} directive
6820 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6821 compact, variable length representation of numbers used by the DWARF
6822 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6824 @ifclear no-space-dir
6826 @section @code{.space @var{size} [,@var{fill}]}
6828 @cindex @code{space} directive
6829 @cindex filling memory
6830 This directive emits @var{size} bytes, each of value @var{fill}. Both
6831 @var{size} and @var{fill} are absolute expressions. If the comma
6832 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6837 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6838 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6839 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6840 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6848 @section @code{.stabd, .stabn, .stabs}
6850 @cindex symbolic debuggers, information for
6851 @cindex @code{stab@var{x}} directives
6852 There are three directives that begin @samp{.stab}.
6853 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6854 The symbols are not entered in the @command{@value{AS}} hash table: they
6855 cannot be referenced elsewhere in the source file.
6856 Up to five fields are required:
6860 This is the symbol's name. It may contain any character except
6861 @samp{\000}, so is more general than ordinary symbol names. Some
6862 debuggers used to code arbitrarily complex structures into symbol names
6866 An absolute expression. The symbol's type is set to the low 8 bits of
6867 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6868 and debuggers choke on silly bit patterns.
6871 An absolute expression. The symbol's ``other'' attribute is set to the
6872 low 8 bits of this expression.
6875 An absolute expression. The symbol's descriptor is set to the low 16
6876 bits of this expression.
6879 An absolute expression which becomes the symbol's value.
6882 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6883 or @code{.stabs} statement, the symbol has probably already been created;
6884 you get a half-formed symbol in your object file. This is
6885 compatible with earlier assemblers!
6888 @cindex @code{stabd} directive
6889 @item .stabd @var{type} , @var{other} , @var{desc}
6891 The ``name'' of the symbol generated is not even an empty string.
6892 It is a null pointer, for compatibility. Older assemblers used a
6893 null pointer so they didn't waste space in object files with empty
6896 The symbol's value is set to the location counter,
6897 relocatably. When your program is linked, the value of this symbol
6898 is the address of the location counter when the @code{.stabd} was
6901 @cindex @code{stabn} directive
6902 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6903 The name of the symbol is set to the empty string @code{""}.
6905 @cindex @code{stabs} directive
6906 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6907 All five fields are specified.
6913 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6914 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6916 @cindex string, copying to object file
6917 @cindex string8, copying to object file
6918 @cindex string16, copying to object file
6919 @cindex string32, copying to object file
6920 @cindex string64, copying to object file
6921 @cindex @code{string} directive
6922 @cindex @code{string8} directive
6923 @cindex @code{string16} directive
6924 @cindex @code{string32} directive
6925 @cindex @code{string64} directive
6927 Copy the characters in @var{str} to the object file. You may specify more than
6928 one string to copy, separated by commas. Unless otherwise specified for a
6929 particular machine, the assembler marks the end of each string with a 0 byte.
6930 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6932 The variants @code{string16}, @code{string32} and @code{string64} differ from
6933 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6934 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6935 are stored in target endianness byte order.
6941 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6942 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6947 @section @code{.struct @var{expression}}
6949 @cindex @code{struct} directive
6950 Switch to the absolute section, and set the section offset to @var{expression},
6951 which must be an absolute expression. You might use this as follows:
6960 This would define the symbol @code{field1} to have the value 0, the symbol
6961 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6962 value 8. Assembly would be left in the absolute section, and you would need to
6963 use a @code{.section} directive of some sort to change to some other section
6964 before further assembly.
6968 @section @code{.subsection @var{name}}
6970 @cindex @code{subsection} directive
6971 @cindex Section Stack
6972 This is one of the ELF section stack manipulation directives. The others are
6973 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6974 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6977 This directive replaces the current subsection with @code{name}. The current
6978 section is not changed. The replaced subsection is put onto the section stack
6979 in place of the then current top of stack subsection.
6984 @section @code{.symver}
6985 @cindex @code{symver} directive
6986 @cindex symbol versioning
6987 @cindex versions of symbols
6988 Use the @code{.symver} directive to bind symbols to specific version nodes
6989 within a source file. This is only supported on ELF platforms, and is
6990 typically used when assembling files to be linked into a shared library.
6991 There are cases where it may make sense to use this in objects to be bound
6992 into an application itself so as to override a versioned symbol from a
6995 For ELF targets, the @code{.symver} directive can be used like this:
6997 .symver @var{name}, @var{name2@@nodename}
6999 If the symbol @var{name} is defined within the file
7000 being assembled, the @code{.symver} directive effectively creates a symbol
7001 alias with the name @var{name2@@nodename}, and in fact the main reason that we
7002 just don't try and create a regular alias is that the @var{@@} character isn't
7003 permitted in symbol names. The @var{name2} part of the name is the actual name
7004 of the symbol by which it will be externally referenced. The name @var{name}
7005 itself is merely a name of convenience that is used so that it is possible to
7006 have definitions for multiple versions of a function within a single source
7007 file, and so that the compiler can unambiguously know which version of a
7008 function is being mentioned. The @var{nodename} portion of the alias should be
7009 the name of a node specified in the version script supplied to the linker when
7010 building a shared library. If you are attempting to override a versioned
7011 symbol from a shared library, then @var{nodename} should correspond to the
7012 nodename of the symbol you are trying to override.
7014 If the symbol @var{name} is not defined within the file being assembled, all
7015 references to @var{name} will be changed to @var{name2@@nodename}. If no
7016 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
7019 Another usage of the @code{.symver} directive is:
7021 .symver @var{name}, @var{name2@@@@nodename}
7023 In this case, the symbol @var{name} must exist and be defined within
7024 the file being assembled. It is similar to @var{name2@@nodename}. The
7025 difference is @var{name2@@@@nodename} will also be used to resolve
7026 references to @var{name2} by the linker.
7028 The third usage of the @code{.symver} directive is:
7030 .symver @var{name}, @var{name2@@@@@@nodename}
7032 When @var{name} is not defined within the
7033 file being assembled, it is treated as @var{name2@@nodename}. When
7034 @var{name} is defined within the file being assembled, the symbol
7035 name, @var{name}, will be changed to @var{name2@@@@nodename}.
7040 @section @code{.tag @var{structname}}
7042 @cindex COFF structure debugging
7043 @cindex structure debugging, COFF
7044 @cindex @code{tag} directive
7045 This directive is generated by compilers to include auxiliary debugging
7046 information in the symbol table. It is only permitted inside
7047 @code{.def}/@code{.endef} pairs. Tags are used to link structure
7048 definitions in the symbol table with instances of those structures.
7052 @section @code{.text @var{subsection}}
7054 @cindex @code{text} directive
7055 Tells @command{@value{AS}} to assemble the following statements onto the end of
7056 the text subsection numbered @var{subsection}, which is an absolute
7057 expression. If @var{subsection} is omitted, subsection number zero
7061 @section @code{.title "@var{heading}"}
7063 @cindex @code{title} directive
7064 @cindex listing control: title line
7065 Use @var{heading} as the title (second line, immediately after the
7066 source file name and pagenumber) when generating assembly listings.
7068 This directive affects subsequent pages, as well as the current page if
7069 it appears within ten lines of the top of a page.
7073 @section @code{.type}
7075 This directive is used to set the type of a symbol.
7079 @c only print the extra heading if both COFF and ELF are set
7080 @subheading COFF Version
7083 @cindex COFF symbol type
7084 @cindex symbol type, COFF
7085 @cindex @code{type} directive (COFF version)
7086 For COFF targets, this directive is permitted only within
7087 @code{.def}/@code{.endef} pairs. It is used like this:
7093 This records the integer @var{int} as the type attribute of a symbol table
7100 @c only print the extra heading if both COFF and ELF are set
7101 @subheading ELF Version
7104 @cindex ELF symbol type
7105 @cindex symbol type, ELF
7106 @cindex @code{type} directive (ELF version)
7107 For ELF targets, the @code{.type} directive is used like this:
7110 .type @var{name} , @var{type description}
7113 This sets the type of symbol @var{name} to be either a
7114 function symbol or an object symbol. There are five different syntaxes
7115 supported for the @var{type description} field, in order to provide
7116 compatibility with various other assemblers.
7118 Because some of the characters used in these syntaxes (such as @samp{@@} and
7119 @samp{#}) are comment characters for some architectures, some of the syntaxes
7120 below do not work on all architectures. The first variant will be accepted by
7121 the GNU assembler on all architectures so that variant should be used for
7122 maximum portability, if you do not need to assemble your code with other
7125 The syntaxes supported are:
7128 .type <name> STT_<TYPE_IN_UPPER_CASE>
7129 .type <name>,#<type>
7130 .type <name>,@@<type>
7131 .type <name>,%<type>
7132 .type <name>,"<type>"
7135 The types supported are:
7140 Mark the symbol as being a function name.
7143 @itemx gnu_indirect_function
7144 Mark the symbol as an indirect function when evaluated during reloc
7145 processing. (This is only supported on assemblers targeting GNU systems).
7149 Mark the symbol as being a data object.
7153 Mark the symbol as being a thread-local data object.
7157 Mark the symbol as being a common data object.
7161 Does not mark the symbol in any way. It is supported just for completeness.
7163 @item gnu_unique_object
7164 Marks the symbol as being a globally unique data object. The dynamic linker
7165 will make sure that in the entire process there is just one symbol with this
7166 name and type in use. (This is only supported on assemblers targeting GNU
7171 Note: Some targets support extra types in addition to those listed above.
7177 @section @code{.uleb128 @var{expressions}}
7179 @cindex @code{uleb128} directive
7180 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7181 compact, variable length representation of numbers used by the DWARF
7182 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7186 @section @code{.val @var{addr}}
7188 @cindex @code{val} directive
7189 @cindex COFF value attribute
7190 @cindex value attribute, COFF
7191 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7192 records the address @var{addr} as the value attribute of a symbol table
7198 @section @code{.version "@var{string}"}
7200 @cindex @code{version} directive
7201 This directive creates a @code{.note} section and places into it an ELF
7202 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7207 @section @code{.vtable_entry @var{table}, @var{offset}}
7209 @cindex @code{vtable_entry} directive
7210 This directive finds or creates a symbol @code{table} and creates a
7211 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7214 @section @code{.vtable_inherit @var{child}, @var{parent}}
7216 @cindex @code{vtable_inherit} directive
7217 This directive finds the symbol @code{child} and finds or creates the symbol
7218 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7219 parent whose addend is the value of the child symbol. As a special case the
7220 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7224 @section @code{.warning "@var{string}"}
7225 @cindex warning directive
7226 Similar to the directive @code{.error}
7227 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7230 @section @code{.weak @var{names}}
7232 @cindex @code{weak} directive
7233 This directive sets the weak attribute on the comma separated list of symbol
7234 @code{names}. If the symbols do not already exist, they will be created.
7236 On COFF targets other than PE, weak symbols are a GNU extension. This
7237 directive sets the weak attribute on the comma separated list of symbol
7238 @code{names}. If the symbols do not already exist, they will be created.
7240 On the PE target, weak symbols are supported natively as weak aliases.
7241 When a weak symbol is created that is not an alias, GAS creates an
7242 alternate symbol to hold the default value.
7245 @section @code{.weakref @var{alias}, @var{target}}
7247 @cindex @code{weakref} directive
7248 This directive creates an alias to the target symbol that enables the symbol to
7249 be referenced with weak-symbol semantics, but without actually making it weak.
7250 If direct references or definitions of the symbol are present, then the symbol
7251 will not be weak, but if all references to it are through weak references, the
7252 symbol will be marked as weak in the symbol table.
7254 The effect is equivalent to moving all references to the alias to a separate
7255 assembly source file, renaming the alias to the symbol in it, declaring the
7256 symbol as weak there, and running a reloadable link to merge the object files
7257 resulting from the assembly of the new source file and the old source file that
7258 had the references to the alias removed.
7260 The alias itself never makes to the symbol table, and is entirely handled
7261 within the assembler.
7264 @section @code{.word @var{expressions}}
7266 @cindex @code{word} directive
7267 This directive expects zero or more @var{expressions}, of any section,
7268 separated by commas.
7271 For each expression, @command{@value{AS}} emits a 32-bit number.
7274 For each expression, @command{@value{AS}} emits a 16-bit number.
7279 The size of the number emitted, and its byte order,
7280 depend on what target computer the assembly is for.
7283 @c on sparc the "special treatment to support compilers" doesn't
7284 @c happen---32-bit addressability, period; no long/short jumps.
7285 @ifset DIFF-TBL-KLUGE
7286 @cindex difference tables altered
7287 @cindex altered difference tables
7289 @emph{Warning: Special Treatment to support Compilers}
7293 Machines with a 32-bit address space, but that do less than 32-bit
7294 addressing, require the following special treatment. If the machine of
7295 interest to you does 32-bit addressing (or doesn't require it;
7296 @pxref{Machine Dependencies}), you can ignore this issue.
7299 In order to assemble compiler output into something that works,
7300 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7301 Directives of the form @samp{.word sym1-sym2} are often emitted by
7302 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7303 directive of the form @samp{.word sym1-sym2}, and the difference between
7304 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7305 creates a @dfn{secondary jump table}, immediately before the next label.
7306 This secondary jump table is preceded by a short-jump to the
7307 first byte after the secondary table. This short-jump prevents the flow
7308 of control from accidentally falling into the new table. Inside the
7309 table is a long-jump to @code{sym2}. The original @samp{.word}
7310 contains @code{sym1} minus the address of the long-jump to
7313 If there were several occurrences of @samp{.word sym1-sym2} before the
7314 secondary jump table, all of them are adjusted. If there was a
7315 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7316 long-jump to @code{sym4} is included in the secondary jump table,
7317 and the @code{.word} directives are adjusted to contain @code{sym3}
7318 minus the address of the long-jump to @code{sym4}; and so on, for as many
7319 entries in the original jump table as necessary.
7322 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7323 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7324 assembly language programmers.
7327 @c end DIFF-TBL-KLUGE
7329 @ifclear no-space-dir
7331 @section @code{.zero @var{size}}
7333 @cindex @code{zero} directive
7334 @cindex filling memory with zero bytes
7335 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7336 expression. This directive is actually an alias for the @samp{.skip} directive
7337 so in can take an optional second argument of the value to store in the bytes
7338 instead of zero. Using @samp{.zero} in this way would be confusing however.
7343 @section @code{.2byte @var{expression} [, @var{expression}]*}
7344 @cindex @code{2byte} directive
7345 @cindex two-byte integer
7346 @cindex integer, 2-byte
7348 This directive expects zero or more expressions, separated by commas. If there
7349 are no expressions then the directive does nothing. Otherwise each expression
7350 is evaluated in turn and placed in the next two bytes of the current output
7351 section, using the endian model of the target. If an expression will not fit
7352 in two bytes, a warning message is displayed and the least significant two
7353 bytes of the expression's value are used. If an expression cannot be evaluated
7354 at assembly time then relocations will be generated in order to compute the
7357 This directive does not apply any alignment before or after inserting the
7358 values. As a result of this, if relocations are generated, they may be
7359 different from those used for inserting values with a guaranteed alignment.
7361 This directive is only available for ELF targets,
7364 @section @code{.4byte @var{expression} [, @var{expression}]*}
7365 @cindex @code{4byte} directive
7366 @cindex four-byte integer
7367 @cindex integer, 4-byte
7369 Like the @option{.2byte} directive, except that it inserts unaligned, four byte
7370 long values into the output.
7373 @section @code{.8byte @var{expression} [, @var{expression}]*}
7374 @cindex @code{8byte} directive
7375 @cindex eight-byte integer
7376 @cindex integer, 8-byte
7378 Like the @option{.2byte} directive, except that it inserts unaligned, eight
7379 byte long bignum values into the output.
7384 @section Deprecated Directives
7386 @cindex deprecated directives
7387 @cindex obsolescent directives
7388 One day these directives won't work.
7389 They are included for compatibility with older assemblers.
7396 @node Object Attributes
7397 @chapter Object Attributes
7398 @cindex object attributes
7400 @command{@value{AS}} assembles source files written for a specific architecture
7401 into object files for that architecture. But not all object files are alike.
7402 Many architectures support incompatible variations. For instance, floating
7403 point arguments might be passed in floating point registers if the object file
7404 requires hardware floating point support---or floating point arguments might be
7405 passed in integer registers if the object file supports processors with no
7406 hardware floating point unit. Or, if two objects are built for different
7407 generations of the same architecture, the combination may require the
7408 newer generation at run-time.
7410 This information is useful during and after linking. At link time,
7411 @command{@value{LD}} can warn about incompatible object files. After link
7412 time, tools like @command{gdb} can use it to process the linked file
7415 Compatibility information is recorded as a series of object attributes. Each
7416 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7417 string, and indicates who sets the meaning of the tag. The tag is an integer,
7418 and indicates what property the attribute describes. The value may be a string
7419 or an integer, and indicates how the property affects this object. Missing
7420 attributes are the same as attributes with a zero value or empty string value.
7422 Object attributes were developed as part of the ABI for the ARM Architecture.
7423 The file format is documented in @cite{ELF for the ARM Architecture}.
7426 * GNU Object Attributes:: @sc{gnu} Object Attributes
7427 * Defining New Object Attributes:: Defining New Object Attributes
7430 @node GNU Object Attributes
7431 @section @sc{gnu} Object Attributes
7433 The @code{.gnu_attribute} directive records an object attribute
7434 with vendor @samp{gnu}.
7436 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7437 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7438 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7439 2} is set for architecture-independent attributes and clear for
7440 architecture-dependent ones.
7442 @subsection Common @sc{gnu} attributes
7444 These attributes are valid on all architectures.
7447 @item Tag_compatibility (32)
7448 The compatibility attribute takes an integer flag value and a vendor name. If
7449 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7450 then the file is only compatible with the named toolchain. If it is greater
7451 than 1, the file can only be processed by other toolchains under some private
7452 arrangement indicated by the flag value and the vendor name.
7455 @subsection MIPS Attributes
7458 @item Tag_GNU_MIPS_ABI_FP (4)
7459 The floating-point ABI used by this object file. The value will be:
7463 0 for files not affected by the floating-point ABI.
7465 1 for files using the hardware floating-point ABI with a standard
7466 double-precision FPU.
7468 2 for files using the hardware floating-point ABI with a single-precision FPU.
7470 3 for files using the software floating-point ABI.
7472 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7473 floating-point registers, 32-bit general-purpose registers and increased the
7474 number of callee-saved floating-point registers.
7476 5 for files using the hardware floating-point ABI with a double-precision FPU
7477 with either 32-bit or 64-bit floating-point registers and 32-bit
7478 general-purpose registers.
7480 6 for files using the hardware floating-point ABI with 64-bit floating-point
7481 registers and 32-bit general-purpose registers.
7483 7 for files using the hardware floating-point ABI with 64-bit floating-point
7484 registers, 32-bit general-purpose registers and a rule that forbids the
7485 direct use of odd-numbered single-precision floating-point registers.
7489 @subsection PowerPC Attributes
7492 @item Tag_GNU_Power_ABI_FP (4)
7493 The floating-point ABI used by this object file. The value will be:
7497 0 for files not affected by the floating-point ABI.
7499 1 for files using double-precision hardware floating-point ABI.
7501 2 for files using the software floating-point ABI.
7503 3 for files using single-precision hardware floating-point ABI.
7506 @item Tag_GNU_Power_ABI_Vector (8)
7507 The vector ABI used by this object file. The value will be:
7511 0 for files not affected by the vector ABI.
7513 1 for files using general purpose registers to pass vectors.
7515 2 for files using AltiVec registers to pass vectors.
7517 3 for files using SPE registers to pass vectors.
7521 @subsection IBM z Systems Attributes
7524 @item Tag_GNU_S390_ABI_Vector (8)
7525 The vector ABI used by this object file. The value will be:
7529 0 for files not affected by the vector ABI.
7531 1 for files using software vector ABI.
7533 2 for files using hardware vector ABI.
7537 @node Defining New Object Attributes
7538 @section Defining New Object Attributes
7540 If you want to define a new @sc{gnu} object attribute, here are the places you
7541 will need to modify. New attributes should be discussed on the @samp{binutils}
7546 This manual, which is the official register of attributes.
7548 The header for your architecture @file{include/elf}, to define the tag.
7550 The @file{bfd} support file for your architecture, to merge the attribute
7551 and issue any appropriate link warnings.
7553 Test cases in @file{ld/testsuite} for merging and link warnings.
7555 @file{binutils/readelf.c} to display your attribute.
7557 GCC, if you want the compiler to mark the attribute automatically.
7563 @node Machine Dependencies
7564 @chapter Machine Dependent Features
7566 @cindex machine dependencies
7567 The machine instruction sets are (almost by definition) different on
7568 each machine where @command{@value{AS}} runs. Floating point representations
7569 vary as well, and @command{@value{AS}} often supports a few additional
7570 directives or command-line options for compatibility with other
7571 assemblers on a particular platform. Finally, some versions of
7572 @command{@value{AS}} support special pseudo-instructions for branch
7575 This chapter discusses most of these differences, though it does not
7576 include details on any machine's instruction set. For details on that
7577 subject, see the hardware manufacturer's manual.
7581 * AArch64-Dependent:: AArch64 Dependent Features
7584 * Alpha-Dependent:: Alpha Dependent Features
7587 * ARC-Dependent:: ARC Dependent Features
7590 * ARM-Dependent:: ARM Dependent Features
7593 * AVR-Dependent:: AVR Dependent Features
7596 * Blackfin-Dependent:: Blackfin Dependent Features
7599 * CR16-Dependent:: CR16 Dependent Features
7602 * CRIS-Dependent:: CRIS Dependent Features
7605 * C-SKY-Dependent:: C-SKY Dependent Features
7608 * D10V-Dependent:: D10V Dependent Features
7611 * D30V-Dependent:: D30V Dependent Features
7614 * Epiphany-Dependent:: EPIPHANY Dependent Features
7617 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7620 * HPPA-Dependent:: HPPA Dependent Features
7623 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7626 * IA-64-Dependent:: Intel IA-64 Dependent Features
7629 * IP2K-Dependent:: IP2K Dependent Features
7632 * LM32-Dependent:: LM32 Dependent Features
7635 * M32C-Dependent:: M32C Dependent Features
7638 * M32R-Dependent:: M32R Dependent Features
7641 * M68K-Dependent:: M680x0 Dependent Features
7644 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7647 * S12Z-Dependent:: S12Z Dependent Features
7650 * Meta-Dependent :: Meta Dependent Features
7653 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7656 * MIPS-Dependent:: MIPS Dependent Features
7659 * MMIX-Dependent:: MMIX Dependent Features
7662 * MSP430-Dependent:: MSP430 Dependent Features
7665 * NDS32-Dependent:: Andes NDS32 Dependent Features
7668 * NiosII-Dependent:: Altera Nios II Dependent Features
7671 * NS32K-Dependent:: NS32K Dependent Features
7674 * PDP-11-Dependent:: PDP-11 Dependent Features
7677 * PJ-Dependent:: picoJava Dependent Features
7680 * PPC-Dependent:: PowerPC Dependent Features
7683 * PRU-Dependent:: PRU Dependent Features
7686 * RISC-V-Dependent:: RISC-V Dependent Features
7689 * RL78-Dependent:: RL78 Dependent Features
7692 * RX-Dependent:: RX Dependent Features
7695 * S/390-Dependent:: IBM S/390 Dependent Features
7698 * SCORE-Dependent:: SCORE Dependent Features
7701 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7704 * Sparc-Dependent:: SPARC Dependent Features
7707 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7710 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7713 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7716 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7719 * V850-Dependent:: V850 Dependent Features
7722 * Vax-Dependent:: VAX Dependent Features
7725 * Visium-Dependent:: Visium Dependent Features
7728 * WebAssembly-Dependent:: WebAssembly Dependent Features
7731 * XGATE-Dependent:: XGATE Dependent Features
7734 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7737 * Xtensa-Dependent:: Xtensa Dependent Features
7740 * Z80-Dependent:: Z80 Dependent Features
7743 * Z8000-Dependent:: Z8000 Dependent Features
7750 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7751 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7752 @c peculiarity: to preserve cross-references, there must be a node called
7753 @c "Machine Dependencies". Hence the conditional nodenames in each
7754 @c major node below. Node defaulting in makeinfo requires adjacency of
7755 @c node and sectioning commands; hence the repetition of @chapter BLAH
7756 @c in both conditional blocks.
7759 @include c-aarch64.texi
7763 @include c-alpha.texi
7779 @include c-bfin.texi
7783 @include c-cr16.texi
7787 @include c-cris.texi
7791 @include c-csky.texi
7796 @node Machine Dependencies
7797 @chapter Machine Dependent Features
7799 The machine instruction sets are different on each Renesas chip family,
7800 and there are also some syntax differences among the families. This
7801 chapter describes the specific @command{@value{AS}} features for each
7805 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7806 * SH-Dependent:: Renesas SH Dependent Features
7813 @include c-d10v.texi
7817 @include c-d30v.texi
7821 @include c-epiphany.texi
7825 @include c-h8300.texi
7829 @include c-hppa.texi
7833 @include c-i386.texi
7837 @include c-ia64.texi
7841 @include c-ip2k.texi
7845 @include c-lm32.texi
7849 @include c-m32c.texi
7853 @include c-m32r.texi
7857 @include c-m68k.texi
7861 @include c-m68hc11.texi
7865 @include c-s12z.texi
7869 @include c-metag.texi
7873 @include c-microblaze.texi
7877 @include c-mips.texi
7881 @include c-mmix.texi
7885 @include c-msp430.texi
7889 @include c-nds32.texi
7893 @include c-nios2.texi
7897 @include c-ns32k.texi
7901 @include c-pdp11.texi
7917 @include c-riscv.texi
7921 @include c-rl78.texi
7929 @include c-s390.texi
7933 @include c-score.texi
7941 @include c-sparc.texi
7945 @include c-tic54x.texi
7949 @include c-tic6x.texi
7953 @include c-tilegx.texi
7957 @include c-tilepro.texi
7961 @include c-v850.texi
7969 @include c-visium.texi
7973 @include c-wasm32.texi
7977 @include c-xgate.texi
7981 @include c-xstormy16.texi
7985 @include c-xtensa.texi
7997 @c reverse effect of @down at top of generic Machine-Dep chapter
8001 @node Reporting Bugs
8002 @chapter Reporting Bugs
8003 @cindex bugs in assembler
8004 @cindex reporting bugs in assembler
8006 Your bug reports play an essential role in making @command{@value{AS}} reliable.
8008 Reporting a bug may help you by bringing a solution to your problem, or it may
8009 not. But in any case the principal function of a bug report is to help the
8010 entire community by making the next version of @command{@value{AS}} work better.
8011 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
8013 In order for a bug report to serve its purpose, you must include the
8014 information that enables us to fix the bug.
8017 * Bug Criteria:: Have you found a bug?
8018 * Bug Reporting:: How to report bugs
8022 @section Have You Found a Bug?
8023 @cindex bug criteria
8025 If you are not sure whether you have found a bug, here are some guidelines:
8028 @cindex fatal signal
8029 @cindex assembler crash
8030 @cindex crash of assembler
8032 If the assembler gets a fatal signal, for any input whatever, that is a
8033 @command{@value{AS}} bug. Reliable assemblers never crash.
8035 @cindex error on valid input
8037 If @command{@value{AS}} produces an error message for valid input, that is a bug.
8039 @cindex invalid input
8041 If @command{@value{AS}} does not produce an error message for invalid input, that
8042 is a bug. However, you should note that your idea of ``invalid input'' might
8043 be our idea of ``an extension'' or ``support for traditional practice''.
8046 If you are an experienced user of assemblers, your suggestions for improvement
8047 of @command{@value{AS}} are welcome in any case.
8051 @section How to Report Bugs
8053 @cindex assembler bugs, reporting
8055 A number of companies and individuals offer support for @sc{gnu} products. If
8056 you obtained @command{@value{AS}} from a support organization, we recommend you
8057 contact that organization first.
8059 You can find contact information for many support companies and
8060 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
8064 In any event, we also recommend that you send bug reports for @command{@value{AS}}
8068 The fundamental principle of reporting bugs usefully is this:
8069 @strong{report all the facts}. If you are not sure whether to state a
8070 fact or leave it out, state it!
8072 Often people omit facts because they think they know what causes the problem
8073 and assume that some details do not matter. Thus, you might assume that the
8074 name of a symbol you use in an example does not matter. Well, probably it does
8075 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
8076 happens to fetch from the location where that name is stored in memory;
8077 perhaps, if the name were different, the contents of that location would fool
8078 the assembler into doing the right thing despite the bug. Play it safe and
8079 give a specific, complete example. That is the easiest thing for you to do,
8080 and the most helpful.
8082 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
8083 it is new to us. Therefore, always write your bug reports on the assumption
8084 that the bug has not been reported previously.
8086 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8087 bell?'' This cannot help us fix a bug, so it is basically useless. We
8088 respond by asking for enough details to enable us to investigate.
8089 You might as well expedite matters by sending them to begin with.
8091 To enable us to fix the bug, you should include all these things:
8095 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8096 it with the @samp{--version} argument.
8098 Without this, we will not know whether there is any point in looking for
8099 the bug in the current version of @command{@value{AS}}.
8102 Any patches you may have applied to the @command{@value{AS}} source.
8105 The type of machine you are using, and the operating system name and
8109 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8113 The command arguments you gave the assembler to assemble your example and
8114 observe the bug. To guarantee you will not omit something important, list them
8115 all. A copy of the Makefile (or the output from make) is sufficient.
8117 If we were to try to guess the arguments, we would probably guess wrong
8118 and then we might not encounter the bug.
8121 A complete input file that will reproduce the bug. If the bug is observed when
8122 the assembler is invoked via a compiler, send the assembler source, not the
8123 high level language source. Most compilers will produce the assembler source
8124 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8125 the options @samp{-v --save-temps}; this will save the assembler source in a
8126 file with an extension of @file{.s}, and also show you exactly how
8127 @command{@value{AS}} is being run.
8130 A description of what behavior you observe that you believe is
8131 incorrect. For example, ``It gets a fatal signal.''
8133 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8134 will certainly notice it. But if the bug is incorrect output, we might not
8135 notice unless it is glaringly wrong. You might as well not give us a chance to
8138 Even if the problem you experience is a fatal signal, you should still say so
8139 explicitly. Suppose something strange is going on, such as, your copy of
8140 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8141 library on your system. (This has happened!) Your copy might crash and ours
8142 would not. If you told us to expect a crash, then when ours fails to crash, we
8143 would know that the bug was not happening for us. If you had not told us to
8144 expect a crash, then we would not be able to draw any conclusion from our
8148 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8149 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8150 option. Always send diffs from the old file to the new file. If you even
8151 discuss something in the @command{@value{AS}} source, refer to it by context, not
8154 The line numbers in our development sources will not match those in your
8155 sources. Your line numbers would convey no useful information to us.
8158 Here are some things that are not necessary:
8162 A description of the envelope of the bug.
8164 Often people who encounter a bug spend a lot of time investigating
8165 which changes to the input file will make the bug go away and which
8166 changes will not affect it.
8168 This is often time consuming and not very useful, because the way we
8169 will find the bug is by running a single example under the debugger
8170 with breakpoints, not by pure deduction from a series of examples.
8171 We recommend that you save your time for something else.
8173 Of course, if you can find a simpler example to report @emph{instead}
8174 of the original one, that is a convenience for us. Errors in the
8175 output will be easier to spot, running under the debugger will take
8176 less time, and so on.
8178 However, simplification is not vital; if you do not want to do this,
8179 report the bug anyway and send us the entire test case you used.
8182 A patch for the bug.
8184 A patch for the bug does help us if it is a good one. But do not omit
8185 the necessary information, such as the test case, on the assumption that
8186 a patch is all we need. We might see problems with your patch and decide
8187 to fix the problem another way, or we might not understand it at all.
8189 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8190 construct an example that will make the program follow a certain path through
8191 the code. If you do not send us the example, we will not be able to construct
8192 one, so we will not be able to verify that the bug is fixed.
8194 And if we cannot understand what bug you are trying to fix, or why your
8195 patch should be an improvement, we will not install it. A test case will
8196 help us to understand.
8199 A guess about what the bug is or what it depends on.
8201 Such guesses are usually wrong. Even we cannot guess right about such
8202 things without first using the debugger to find the facts.
8205 @node Acknowledgements
8206 @chapter Acknowledgements
8208 If you have contributed to GAS and your name isn't listed here,
8209 it is not meant as a slight. We just don't know about it. Send mail to the
8210 maintainer, and we'll correct the situation. Currently
8212 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8214 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8217 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8218 information and the 68k series machines, most of the preprocessing pass, and
8219 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8221 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8222 many bug fixes, including merging support for several processors, breaking GAS
8223 up to handle multiple object file format back ends (including heavy rewrite,
8224 testing, an integration of the coff and b.out back ends), adding configuration
8225 including heavy testing and verification of cross assemblers and file splits
8226 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8227 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8228 port (including considerable amounts of reverse engineering), a SPARC opcode
8229 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8230 assertions and made them work, much other reorganization, cleanup, and lint.
8232 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8233 in format-specific I/O modules.
8235 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8236 has done much work with it since.
8238 The Intel 80386 machine description was written by Eliot Dresselhaus.
8240 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8242 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8243 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8245 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8246 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8247 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8248 support a.out format.
8250 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8251 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8252 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8253 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8256 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8257 simplified the configuration of which versions accept which directives. He
8258 updated the 68k machine description so that Motorola's opcodes always produced
8259 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8260 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8261 cross-compilation support, and one bug in relaxation that took a week and
8262 required the proverbial one-bit fix.
8264 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8265 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8266 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8267 PowerPC assembler, and made a few other minor patches.
8269 Steve Chamberlain made GAS able to generate listings.
8271 Hewlett-Packard contributed support for the HP9000/300.
8273 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8274 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8275 formats). This work was supported by both the Center for Software Science at
8276 the University of Utah and Cygnus Support.
8278 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8279 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8280 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8281 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8282 and some initial 64-bit support).
8284 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8286 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8287 support for openVMS/Alpha.
8289 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8292 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8293 Inc.@: added support for Xtensa processors.
8295 Several engineers at Cygnus Support have also provided many small bug fixes and
8296 configuration enhancements.
8298 Jon Beniston added support for the Lattice Mico32 architecture.
8300 Many others have contributed large or small bugfixes and enhancements. If
8301 you have contributed significant work and are not mentioned on this list, and
8302 want to be, let us know. Some of the history has been lost; we are not
8303 intentionally leaving anyone out.
8305 @node GNU Free Documentation License
8306 @appendix GNU Free Documentation License
8310 @unnumbered AS Index
projects / thirdparty / binutils-gdb.git / blob