1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (C) 1991-2020 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
53 @set abnormal-separator
57 @settitle Using @value{AS}
60 @settitle Using @value{AS} (@value{TARGET})
62 @setchapternewpage odd
67 @c WARE! Some of the machine-dependent sections contain tables of machine
68 @c instructions. Except in multi-column format, these tables look silly.
69 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
70 @c the multi-col format is faked within @example sections.
72 @c Again unfortunately, the natural size that fits on a page, for these tables,
73 @c is different depending on whether or not smallbook is turned on.
74 @c This matters, because of order: text flow switches columns at each page
77 @c The format faked in this source works reasonably well for smallbook,
78 @c not well for the default large-page format. This manual expects that if you
79 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
80 @c tables in question. You can turn on one without the other at your
81 @c discretion, of course.
84 @c the insn tables look just as silly in info files regardless of smallbook,
85 @c might as well show 'em anyways.
89 @dircategory Software development
91 * As: (as). The GNU assembler.
92 * Gas: (as). The GNU assembler.
100 This file documents the GNU Assembler "@value{AS}".
102 @c man begin COPYRIGHT
103 Copyright @copyright{} 1991-2020 Free Software Foundation, Inc.
105 Permission is granted to copy, distribute and/or modify this document
106 under the terms of the GNU Free Documentation License, Version 1.3
107 or any later version published by the Free Software Foundation;
108 with no Invariant Sections, with no Front-Cover Texts, and with no
109 Back-Cover Texts. A copy of the license is included in the
110 section entitled ``GNU Free Documentation License''.
116 @title Using @value{AS}
117 @subtitle The @sc{gnu} Assembler
119 @subtitle for the @value{TARGET} family
121 @ifset VERSION_PACKAGE
123 @subtitle @value{VERSION_PACKAGE}
126 @subtitle Version @value{VERSION}
129 The Free Software Foundation Inc.@: thanks The Nice Computer
130 Company of Australia for loaning Dean Elsner to write the
131 first (Vax) version of @command{as} for Project @sc{gnu}.
132 The proprietors, management and staff of TNCCA thank FSF for
133 distracting the boss while they got some work
136 @author Dean Elsner, Jay Fenlason & friends
140 \hfill {\it Using {\tt @value{AS}}}\par
141 \hfill Edited by Cygnus Support\par
143 %"boxit" macro for figures:
144 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
145 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
146 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
147 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
148 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
151 @vskip 0pt plus 1filll
152 Copyright @copyright{} 1991-2020 Free Software Foundation, Inc.
154 Permission is granted to copy, distribute and/or modify this document
155 under the terms of the GNU Free Documentation License, Version 1.3
156 or any later version published by the Free Software Foundation;
157 with no Invariant Sections, with no Front-Cover Texts, and with no
158 Back-Cover Texts. A copy of the license is included in the
159 section entitled ``GNU Free Documentation License''.
166 @top Using @value{AS}
168 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
169 @ifset VERSION_PACKAGE
170 @value{VERSION_PACKAGE}
172 version @value{VERSION}.
174 This version of the file describes @command{@value{AS}} configured to generate
175 code for @value{TARGET} architectures.
178 This document is distributed under the terms of the GNU Free
179 Documentation License. A copy of the license is included in the
180 section entitled ``GNU Free Documentation License''.
183 * Overview:: Overview
184 * Invoking:: Command-Line Options
186 * Sections:: Sections and Relocation
188 * Expressions:: Expressions
189 * Pseudo Ops:: Assembler Directives
191 * Object Attributes:: Object Attributes
193 * Machine Dependencies:: Machine Dependent Features
194 * Reporting Bugs:: Reporting Bugs
195 * Acknowledgements:: Who Did What
196 * GNU Free Documentation License:: GNU Free Documentation License
197 * AS Index:: AS Index
204 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
206 This version of the manual describes @command{@value{AS}} configured to generate
207 code for @value{TARGET} architectures.
211 @cindex invocation summary
212 @cindex option summary
213 @cindex summary of options
214 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
215 see @ref{Invoking,,Command-Line Options}.
217 @c man title AS the portable GNU assembler.
221 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
225 @c We don't use deffn and friends for the following because they seem
226 @c to be limited to one line for the header.
228 @c man begin SYNOPSIS
229 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
230 [@b{--compress-debug-sections}] [@b{--nocompress-debug-sections}]
231 [@b{--debug-prefix-map} @var{old}=@var{new}]
232 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
233 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--gdwarf-sections}]
234 [@b{--gdwarf-cie-version}=@var{VERSION}]
235 [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
236 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
237 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
238 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}]
239 [@b{--no-pad-sections}]
240 [@b{-o} @var{objfile}] [@b{-R}]
241 [@b{--hash-size}=@var{NUM}] [@b{--reduce-memory-overheads}]
243 [@b{-v}] [@b{-version}] [@b{--version}]
244 [@b{-W}] [@b{--warn}] [@b{--fatal-warnings}] [@b{-w}] [@b{-x}]
245 [@b{-Z}] [@b{@@@var{FILE}}]
246 [@b{--sectname-subst}] [@b{--size-check=[error|warning]}]
247 [@b{--elf-stt-common=[no|yes]}]
248 [@b{--generate-missing-build-notes=[no|yes]}]
249 [@b{--target-help}] [@var{target-options}]
250 [@b{--}|@var{files} @dots{}]
253 @c Target dependent options are listed below. Keep the list sorted.
254 @c Add an empty line for separation.
258 @emph{Target AArch64 options:}
260 [@b{-mabi}=@var{ABI}]
264 @emph{Target Alpha options:}
266 [@b{-mdebug} | @b{-no-mdebug}]
267 [@b{-replace} | @b{-noreplace}]
268 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
269 [@b{-F}] [@b{-32addr}]
273 @emph{Target ARC options:}
274 [@b{-mcpu=@var{cpu}}]
275 [@b{-mA6}|@b{-mARC600}|@b{-mARC601}|@b{-mA7}|@b{-mARC700}|@b{-mEM}|@b{-mHS}]
282 @emph{Target ARM options:}
283 @c Don't document the deprecated options
284 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
285 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
286 [@b{-mfpu}=@var{floating-point-format}]
287 [@b{-mfloat-abi}=@var{abi}]
288 [@b{-meabi}=@var{ver}]
291 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
292 @b{-mapcs-reentrant}]
293 [@b{-mthumb-interwork}] [@b{-k}]
297 @emph{Target Blackfin options:}
298 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
305 @emph{Target BPF options:}
310 @emph{Target CRIS options:}
311 [@b{--underscore} | @b{--no-underscore}]
313 [@b{--emulation=criself} | @b{--emulation=crisaout}]
314 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
315 @c Deprecated -- deliberately not documented.
320 @emph{Target C-SKY options:}
321 [@b{-march=@var{arch}}] [@b{-mcpu=@var{cpu}}]
322 [@b{-EL}] [@b{-mlittle-endian}] [@b{-EB}] [@b{-mbig-endian}]
323 [@b{-fpic}] [@b{-pic}]
324 [@b{-mljump}] [@b{-mno-ljump}]
325 [@b{-force2bsr}] [@b{-mforce2bsr}] [@b{-no-force2bsr}] [@b{-mno-force2bsr}]
326 [@b{-jsri2bsr}] [@b{-mjsri2bsr}] [@b{-no-jsri2bsr }] [@b{-mno-jsri2bsr}]
327 [@b{-mnolrw }] [@b{-mno-lrw}]
328 [@b{-melrw}] [@b{-mno-elrw}]
329 [@b{-mlaf }] [@b{-mliterals-after-func}]
330 [@b{-mno-laf}] [@b{-mno-literals-after-func}]
331 [@b{-mlabr}] [@b{-mliterals-after-br}]
332 [@b{-mno-labr}] [@b{-mnoliterals-after-br}]
333 [@b{-mistack}] [@b{-mno-istack}]
334 [@b{-mhard-float}] [@b{-mmp}] [@b{-mcp}] [@b{-mcache}]
335 [@b{-msecurity}] [@b{-mtrust}]
336 [@b{-mdsp}] [@b{-medsp}] [@b{-mvdsp}]
340 @emph{Target D10V options:}
345 @emph{Target D30V options:}
346 [@b{-O}|@b{-n}|@b{-N}]
350 @emph{Target EPIPHANY options:}
351 [@b{-mepiphany}|@b{-mepiphany16}]
355 @emph{Target H8/300 options:}
359 @c HPPA has no machine-dependent assembler options (yet).
363 @emph{Target i386 options:}
364 [@b{--32}|@b{--x32}|@b{--64}] [@b{-n}]
365 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
369 @emph{Target IA-64 options:}
370 [@b{-mconstant-gp}|@b{-mauto-pic}]
371 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
373 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
374 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
375 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
376 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
380 @emph{Target IP2K options:}
381 [@b{-mip2022}|@b{-mip2022ext}]
385 @emph{Target M32C options:}
386 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
390 @emph{Target M32R options:}
391 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
396 @emph{Target M680X0 options:}
397 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
401 @emph{Target M68HC11 options:}
402 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}|@b{-mm9s12x}|@b{-mm9s12xg}]
403 [@b{-mshort}|@b{-mlong}]
404 [@b{-mshort-double}|@b{-mlong-double}]
405 [@b{--force-long-branches}] [@b{--short-branches}]
406 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
407 [@b{--print-opcodes}] [@b{--generate-example}]
411 @emph{Target MCORE options:}
412 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
413 [@b{-mcpu=[210|340]}]
417 @emph{Target Meta options:}
418 [@b{-mcpu=@var{cpu}}] [@b{-mfpu=@var{cpu}}] [@b{-mdsp=@var{cpu}}]
421 @emph{Target MICROBLAZE options:}
422 @c MicroBlaze has no machine-dependent assembler options.
426 @emph{Target MIPS options:}
427 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
428 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
429 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
430 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
431 [@b{-mfp64}] [@b{-mgp64}] [@b{-mfpxx}]
432 [@b{-modd-spreg}] [@b{-mno-odd-spreg}]
433 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
434 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
435 [@b{-mips32r3}] [@b{-mips32r5}] [@b{-mips32r6}] [@b{-mips64}] [@b{-mips64r2}]
436 [@b{-mips64r3}] [@b{-mips64r5}] [@b{-mips64r6}]
437 [@b{-construct-floats}] [@b{-no-construct-floats}]
438 [@b{-mignore-branch-isa}] [@b{-mno-ignore-branch-isa}]
439 [@b{-mnan=@var{encoding}}]
440 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
441 [@b{-mips16}] [@b{-no-mips16}]
442 [@b{-mmips16e2}] [@b{-mno-mips16e2}]
443 [@b{-mmicromips}] [@b{-mno-micromips}]
444 [@b{-msmartmips}] [@b{-mno-smartmips}]
445 [@b{-mips3d}] [@b{-no-mips3d}]
446 [@b{-mdmx}] [@b{-no-mdmx}]
447 [@b{-mdsp}] [@b{-mno-dsp}]
448 [@b{-mdspr2}] [@b{-mno-dspr2}]
449 [@b{-mdspr3}] [@b{-mno-dspr3}]
450 [@b{-mmsa}] [@b{-mno-msa}]
451 [@b{-mxpa}] [@b{-mno-xpa}]
452 [@b{-mmt}] [@b{-mno-mt}]
453 [@b{-mmcu}] [@b{-mno-mcu}]
454 [@b{-mcrc}] [@b{-mno-crc}]
455 [@b{-mginv}] [@b{-mno-ginv}]
456 [@b{-mloongson-mmi}] [@b{-mno-loongson-mmi}]
457 [@b{-mloongson-cam}] [@b{-mno-loongson-cam}]
458 [@b{-mloongson-ext}] [@b{-mno-loongson-ext}]
459 [@b{-mloongson-ext2}] [@b{-mno-loongson-ext2}]
460 [@b{-minsn32}] [@b{-mno-insn32}]
461 [@b{-mfix7000}] [@b{-mno-fix7000}]
462 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
463 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
464 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
465 [@b{-mfix-r5900}] [@b{-mno-fix-r5900}]
466 [@b{-mdebug}] [@b{-no-mdebug}]
467 [@b{-mpdr}] [@b{-mno-pdr}]
471 @emph{Target MMIX options:}
472 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
473 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
474 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
475 [@b{--linker-allocated-gregs}]
479 @emph{Target Nios II options:}
480 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
485 @emph{Target NDS32 options:}
486 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
487 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
488 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
489 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
490 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
491 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
492 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
496 @c OpenRISC has no machine-dependent assembler options.
500 @emph{Target PDP11 options:}
501 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
502 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
503 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
507 @emph{Target picoJava options:}
512 @emph{Target PowerPC options:}
514 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
515 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mgekko}|
516 @b{-mbroadway}|@b{-mppc64}|@b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|
517 @b{-me6500}|@b{-mppc64bridge}|@b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|
518 @b{-mpower6}|@b{-mpwr6}|@b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-mpower9}|@b{-mpwr9}@b{-ma2}|
519 @b{-mcell}|@b{-mspe}|@b{-mspe2}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
520 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
521 [@b{-mregnames}|@b{-mno-regnames}]
522 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
523 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
524 [@b{-msolaris}|@b{-mno-solaris}]
525 [@b{-nops=@var{count}}]
529 @emph{Target PRU options:}
532 [@b{-mno-warn-regname-label}]
536 @emph{Target RISC-V options:}
537 [@b{-fpic}|@b{-fPIC}|@b{-fno-pic}]
538 [@b{-march}=@var{ISA}]
539 [@b{-mabi}=@var{ABI}]
543 @emph{Target RL78 options:}
545 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
549 @emph{Target RX options:}
550 [@b{-mlittle-endian}|@b{-mbig-endian}]
551 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
552 [@b{-muse-conventional-section-names}]
553 [@b{-msmall-data-limit}]
556 [@b{-mint-register=@var{number}}]
557 [@b{-mgcc-abi}|@b{-mrx-abi}]
561 @emph{Target s390 options:}
562 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
563 [@b{-mregnames}|@b{-mno-regnames}]
564 [@b{-mwarn-areg-zero}]
568 @emph{Target SCORE options:}
569 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
570 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
571 [@b{-march=score7}][@b{-march=score3}]
572 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
576 @emph{Target SPARC options:}
577 @c The order here is important. See c-sparc.texi.
578 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Aleon}|@b{-Asparclet}|@b{-Asparclite}
579 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av8plusb}|@b{-Av8plusc}|@b{-Av8plusd}
580 @b{-Av8plusv}|@b{-Av8plusm}|@b{-Av9}|@b{-Av9a}|@b{-Av9b}|@b{-Av9c}
581 @b{-Av9d}|@b{-Av9e}|@b{-Av9v}|@b{-Av9m}|@b{-Asparc}|@b{-Asparcvis}
582 @b{-Asparcvis2}|@b{-Asparcfmaf}|@b{-Asparcima}|@b{-Asparcvis3}
583 @b{-Asparcvisr}|@b{-Asparc5}]
584 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}]|@b{-xarch=v8plusb}|@b{-xarch=v8plusc}
585 @b{-xarch=v8plusd}|@b{-xarch=v8plusv}|@b{-xarch=v8plusm}|@b{-xarch=v9}
586 @b{-xarch=v9a}|@b{-xarch=v9b}|@b{-xarch=v9c}|@b{-xarch=v9d}|@b{-xarch=v9e}
587 @b{-xarch=v9v}|@b{-xarch=v9m}|@b{-xarch=sparc}|@b{-xarch=sparcvis}
588 @b{-xarch=sparcvis2}|@b{-xarch=sparcfmaf}|@b{-xarch=sparcima}
589 @b{-xarch=sparcvis3}|@b{-xarch=sparcvisr}|@b{-xarch=sparc5}
592 [@b{--enforce-aligned-data}][@b{--dcti-couples-detect}]
596 @emph{Target TIC54X options:}
597 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
598 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
602 @emph{Target TIC6X options:}
603 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
604 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
605 [@b{-mpic}|@b{-mno-pic}]
609 @emph{Target TILE-Gx options:}
610 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
613 @c TILEPro has no machine-dependent assembler options
617 @emph{Target Visium options:}
618 [@b{-mtune=@var{arch}}]
622 @emph{Target Xtensa options:}
623 [@b{--[no-]text-section-literals}] [@b{--[no-]auto-litpools}]
624 [@b{--[no-]absolute-literals}]
625 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
626 [@b{--[no-]transform}]
627 [@b{--rename-section} @var{oldname}=@var{newname}]
628 [@b{--[no-]trampolines}]
632 @emph{Target Z80 options:}
633 [@b{-z80}]|[@b{-z180}]|[@b{-r800}]|[@b{-ez80}]|[@b{-ez80-adl}]
634 [@b{-strict}]|[@b{-full}]
635 [@b{-with-inst=@var{INST}[,...]}] [@b{-Wnins @var{INST}[,...]}]
636 [@b{-without-inst=@var{INST}[,...]}] [@b{-Fins @var{INST}[,...]}]
637 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
638 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
639 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
640 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
641 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
642 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
646 @c Z8000 has no machine-dependent assembler options
655 @include at-file.texi
658 Turn on listings, in any of a variety of ways:
662 omit false conditionals
665 omit debugging directives
668 include general information, like @value{AS} version and options passed
671 include high-level source
677 include macro expansions
680 omit forms processing
686 set the name of the listing file
689 You may combine these options; for example, use @samp{-aln} for assembly
690 listing without forms processing. The @samp{=file} option, if used, must be
691 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
694 Begin in alternate macro mode.
696 @xref{Altmacro,,@code{.altmacro}}.
699 @item --compress-debug-sections
700 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
701 ELF ABI. The resulting object file may not be compatible with older
702 linkers and object file utilities. Note if compression would make a
703 given section @emph{larger} then it is not compressed.
706 @cindex @samp{--compress-debug-sections=} option
707 @item --compress-debug-sections=none
708 @itemx --compress-debug-sections=zlib
709 @itemx --compress-debug-sections=zlib-gnu
710 @itemx --compress-debug-sections=zlib-gabi
711 These options control how DWARF debug sections are compressed.
712 @option{--compress-debug-sections=none} is equivalent to
713 @option{--nocompress-debug-sections}.
714 @option{--compress-debug-sections=zlib} and
715 @option{--compress-debug-sections=zlib-gabi} are equivalent to
716 @option{--compress-debug-sections}.
717 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
718 sections using zlib. The debug sections are renamed to begin with
719 @samp{.zdebug}. Note if compression would make a given section
720 @emph{larger} then it is not compressed nor renamed.
724 @item --nocompress-debug-sections
725 Do not compress DWARF debug sections. This is usually the default for all
726 targets except the x86/x86_64, but a configure time option can be used to
730 Ignored. This option is accepted for script compatibility with calls to
733 @item --debug-prefix-map @var{old}=@var{new}
734 When assembling files in directory @file{@var{old}}, record debugging
735 information describing them as in @file{@var{new}} instead.
737 @item --defsym @var{sym}=@var{value}
738 Define the symbol @var{sym} to be @var{value} before assembling the input file.
739 @var{value} must be an integer constant. As in C, a leading @samp{0x}
740 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
741 value. The value of the symbol can be overridden inside a source file via the
742 use of a @code{.set} pseudo-op.
745 ``fast''---skip whitespace and comment preprocessing (assume source is
750 Generate debugging information for each assembler source line using whichever
751 debug format is preferred by the target. This currently means either STABS,
755 Generate stabs debugging information for each assembler line. This
756 may help debugging assembler code, if the debugger can handle it.
759 Generate stabs debugging information for each assembler line, with GNU
760 extensions that probably only gdb can handle, and that could make other
761 debuggers crash or refuse to read your program. This
762 may help debugging assembler code. Currently the only GNU extension is
763 the location of the current working directory at assembling time.
766 Generate DWARF2 debugging information for each assembler line. This
767 may help debugging assembler code, if the debugger can handle it. Note---this
768 option is only supported by some targets, not all of them.
770 @item --gdwarf-sections
771 Instead of creating a .debug_line section, create a series of
772 .debug_line.@var{foo} sections where @var{foo} is the name of the
773 corresponding code section. For example a code section called @var{.text.func}
774 will have its dwarf line number information placed into a section called
775 @var{.debug_line.text.func}. If the code section is just called @var{.text}
776 then debug line section will still be called just @var{.debug_line} without any
779 @item --gdwarf-cie-version=@var{version}
780 Control which version of DWARF Common Information Entries (CIEs) are produced.
781 When this flag is not specificed the default is version 1, though some targets
782 can modify this default. Other possible values for @var{version} are 3 or 4.
785 @item --size-check=error
786 @itemx --size-check=warning
787 Issue an error or warning for invalid ELF .size directive.
789 @item --elf-stt-common=no
790 @itemx --elf-stt-common=yes
791 These options control whether the ELF assembler should generate common
792 symbols with the @code{STT_COMMON} type. The default can be controlled
793 by a configure option @option{--enable-elf-stt-common}.
795 @item --generate-missing-build-notes=yes
796 @itemx --generate-missing-build-notes=no
797 These options control whether the ELF assembler should generate GNU Build
798 attribute notes if none are present in the input sources.
799 The default can be controlled by the @option{--enable-generate-build-notes}
805 Print a summary of the command-line options and exit.
808 Print a summary of all target specific options and exit.
811 Add directory @var{dir} to the search list for @code{.include} directives.
814 Don't warn about signed overflow.
817 @ifclear DIFF-TBL-KLUGE
818 This option is accepted but has no effect on the @value{TARGET} family.
820 @ifset DIFF-TBL-KLUGE
821 Issue warnings when difference tables altered for long displacements.
826 Keep (in the symbol table) local symbols. These symbols start with
827 system-specific local label prefixes, typically @samp{.L} for ELF systems
828 or @samp{L} for traditional a.out systems.
833 @item --listing-lhs-width=@var{number}
834 Set the maximum width, in words, of the output data column for an assembler
835 listing to @var{number}.
837 @item --listing-lhs-width2=@var{number}
838 Set the maximum width, in words, of the output data column for continuation
839 lines in an assembler listing to @var{number}.
841 @item --listing-rhs-width=@var{number}
842 Set the maximum width of an input source line, as displayed in a listing, to
845 @item --listing-cont-lines=@var{number}
846 Set the maximum number of lines printed in a listing for a single line of input
849 @item --no-pad-sections
850 Stop the assembler for padding the ends of output sections to the alignment
851 of that section. The default is to pad the sections, but this can waste space
852 which might be needed on targets which have tight memory constraints.
854 @item -o @var{objfile}
855 Name the object-file output from @command{@value{AS}} @var{objfile}.
858 Fold the data section into the text section.
860 @item --hash-size=@var{number}
861 Set the default size of GAS's hash tables to a prime number close to
862 @var{number}. Increasing this value can reduce the length of time it takes the
863 assembler to perform its tasks, at the expense of increasing the assembler's
864 memory requirements. Similarly reducing this value can reduce the memory
865 requirements at the expense of speed.
867 @item --reduce-memory-overheads
868 This option reduces GAS's memory requirements, at the expense of making the
869 assembly processes slower. Currently this switch is a synonym for
870 @samp{--hash-size=4051}, but in the future it may have other effects as well.
873 @item --sectname-subst
874 Honor substitution sequences in section names.
876 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
881 Print the maximum space (in bytes) and total time (in seconds) used by
884 @item --strip-local-absolute
885 Remove local absolute symbols from the outgoing symbol table.
889 Print the @command{as} version.
892 Print the @command{as} version and exit.
896 Suppress warning messages.
898 @item --fatal-warnings
899 Treat warnings as errors.
902 Don't suppress warning messages or treat them as errors.
911 Generate an object file even after errors.
913 @item -- | @var{files} @dots{}
914 Standard input, or source files to assemble.
922 @xref{AArch64 Options}, for the options available when @value{AS} is configured
923 for the 64-bit mode of the ARM Architecture (AArch64).
928 The following options are available when @value{AS} is configured for the
929 64-bit mode of the ARM Architecture (AArch64).
932 @include c-aarch64.texi
933 @c ended inside the included file
941 @xref{Alpha Options}, for the options available when @value{AS} is configured
942 for an Alpha processor.
947 The following options are available when @value{AS} is configured for an Alpha
951 @include c-alpha.texi
952 @c ended inside the included file
959 The following options are available when @value{AS} is configured for an ARC
963 @item -mcpu=@var{cpu}
964 This option selects the core processor variant.
966 Select either big-endian (-EB) or little-endian (-EL) output.
968 Enable Code Density extenssion instructions.
973 The following options are available when @value{AS} is configured for the ARM
977 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
978 Specify which ARM processor variant is the target.
979 @item -march=@var{architecture}[+@var{extension}@dots{}]
980 Specify which ARM architecture variant is used by the target.
981 @item -mfpu=@var{floating-point-format}
982 Select which Floating Point architecture is the target.
983 @item -mfloat-abi=@var{abi}
984 Select which floating point ABI is in use.
986 Enable Thumb only instruction decoding.
987 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
988 Select which procedure calling convention is in use.
990 Select either big-endian (-EB) or little-endian (-EL) output.
991 @item -mthumb-interwork
992 Specify that the code has been generated with interworking between Thumb and
995 Turns on CodeComposer Studio assembly syntax compatibility mode.
997 Specify that PIC code has been generated.
1005 @xref{Blackfin Options}, for the options available when @value{AS} is
1006 configured for the Blackfin processor family.
1010 @c man begin OPTIONS
1011 The following options are available when @value{AS} is configured for
1012 the Blackfin processor family.
1014 @c man begin INCLUDE
1015 @include c-bfin.texi
1016 @c ended inside the included file
1024 @xref{BPF Options}, for the options available when @value{AS} is
1025 configured for the Linux kernel BPF processor family.
1029 @c man begin OPTIONS
1030 The following options are available when @value{AS} is configured for
1031 the Linux kernel BPF processor family.
1033 @c man begin INCLUDE
1035 @c ended inside the included file
1040 @c man begin OPTIONS
1042 See the info pages for documentation of the CRIS-specific options.
1048 @xref{C-SKY Options}, for the options available when @value{AS} is
1049 configured for the C-SKY processor family.
1053 @c man begin OPTIONS
1054 The following options are available when @value{AS} is configured for
1055 the C-SKY processor family.
1057 @c man begin INCLUDE
1058 @include c-csky.texi
1059 @c ended inside the included file
1065 The following options are available when @value{AS} is configured for
1068 @cindex D10V optimization
1069 @cindex optimization, D10V
1071 Optimize output by parallelizing instructions.
1076 The following options are available when @value{AS} is configured for a D30V
1079 @cindex D30V optimization
1080 @cindex optimization, D30V
1082 Optimize output by parallelizing instructions.
1086 Warn when nops are generated.
1088 @cindex D30V nops after 32-bit multiply
1090 Warn when a nop after a 32-bit multiply instruction is generated.
1096 The following options are available when @value{AS} is configured for the
1097 Adapteva EPIPHANY series.
1100 @xref{Epiphany Options}, for the options available when @value{AS} is
1101 configured for an Epiphany processor.
1105 @c man begin OPTIONS
1106 The following options are available when @value{AS} is configured for
1107 an Epiphany processor.
1109 @c man begin INCLUDE
1110 @include c-epiphany.texi
1111 @c ended inside the included file
1119 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1120 for an H8/300 processor.
1124 @c man begin OPTIONS
1125 The following options are available when @value{AS} is configured for an H8/300
1128 @c man begin INCLUDE
1129 @include c-h8300.texi
1130 @c ended inside the included file
1138 @xref{i386-Options}, for the options available when @value{AS} is
1139 configured for an i386 processor.
1143 @c man begin OPTIONS
1144 The following options are available when @value{AS} is configured for
1147 @c man begin INCLUDE
1148 @include c-i386.texi
1149 @c ended inside the included file
1154 @c man begin OPTIONS
1156 The following options are available when @value{AS} is configured for the
1162 Specifies that the extended IP2022 instructions are allowed.
1165 Restores the default behaviour, which restricts the permitted instructions to
1166 just the basic IP2022 ones.
1172 The following options are available when @value{AS} is configured for the
1173 Renesas M32C and M16C processors.
1178 Assemble M32C instructions.
1181 Assemble M16C instructions (the default).
1184 Enable support for link-time relaxations.
1187 Support H'00 style hex constants in addition to 0x00 style.
1193 The following options are available when @value{AS} is configured for the
1194 Renesas M32R (formerly Mitsubishi M32R) series.
1199 Specify which processor in the M32R family is the target. The default
1200 is normally the M32R, but this option changes it to the M32RX.
1202 @item --warn-explicit-parallel-conflicts or --Wp
1203 Produce warning messages when questionable parallel constructs are
1206 @item --no-warn-explicit-parallel-conflicts or --Wnp
1207 Do not produce warning messages when questionable parallel constructs are
1214 The following options are available when @value{AS} is configured for the
1215 Motorola 68000 series.
1220 Shorten references to undefined symbols, to one word instead of two.
1222 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1223 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1224 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1225 Specify what processor in the 68000 family is the target. The default
1226 is normally the 68020, but this can be changed at configuration time.
1228 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1229 The target machine does (or does not) have a floating-point coprocessor.
1230 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1231 the basic 68000 is not compatible with the 68881, a combination of the
1232 two can be specified, since it's possible to do emulation of the
1233 coprocessor instructions with the main processor.
1235 @item -m68851 | -mno-68851
1236 The target machine does (or does not) have a memory-management
1237 unit coprocessor. The default is to assume an MMU for 68020 and up.
1245 @xref{Nios II Options}, for the options available when @value{AS} is configured
1246 for an Altera Nios II processor.
1250 @c man begin OPTIONS
1251 The following options are available when @value{AS} is configured for an
1252 Altera Nios II processor.
1254 @c man begin INCLUDE
1255 @include c-nios2.texi
1256 @c ended inside the included file
1262 For details about the PDP-11 machine dependent features options,
1263 see @ref{PDP-11-Options}.
1266 @item -mpic | -mno-pic
1267 Generate position-independent (or position-dependent) code. The
1268 default is @option{-mpic}.
1271 @itemx -mall-extensions
1272 Enable all instruction set extensions. This is the default.
1274 @item -mno-extensions
1275 Disable all instruction set extensions.
1277 @item -m@var{extension} | -mno-@var{extension}
1278 Enable (or disable) a particular instruction set extension.
1281 Enable the instruction set extensions supported by a particular CPU, and
1282 disable all other extensions.
1284 @item -m@var{machine}
1285 Enable the instruction set extensions supported by a particular machine
1286 model, and disable all other extensions.
1292 The following options are available when @value{AS} is configured for
1293 a picoJava processor.
1297 @cindex PJ endianness
1298 @cindex endianness, PJ
1299 @cindex big endian output, PJ
1301 Generate ``big endian'' format output.
1303 @cindex little endian output, PJ
1305 Generate ``little endian'' format output.
1313 @xref{PRU Options}, for the options available when @value{AS} is configured
1314 for a PRU processor.
1318 @c man begin OPTIONS
1319 The following options are available when @value{AS} is configured for a
1322 @c man begin INCLUDE
1324 @c ended inside the included file
1329 The following options are available when @value{AS} is configured for the
1330 Motorola 68HC11 or 68HC12 series.
1334 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1335 Specify what processor is the target. The default is
1336 defined by the configuration option when building the assembler.
1338 @item --xgate-ramoffset
1339 Instruct the linker to offset RAM addresses from S12X address space into
1340 XGATE address space.
1343 Specify to use the 16-bit integer ABI.
1346 Specify to use the 32-bit integer ABI.
1348 @item -mshort-double
1349 Specify to use the 32-bit double ABI.
1352 Specify to use the 64-bit double ABI.
1354 @item --force-long-branches
1355 Relative branches are turned into absolute ones. This concerns
1356 conditional branches, unconditional branches and branches to a
1359 @item -S | --short-branches
1360 Do not turn relative branches into absolute ones
1361 when the offset is out of range.
1363 @item --strict-direct-mode
1364 Do not turn the direct addressing mode into extended addressing mode
1365 when the instruction does not support direct addressing mode.
1367 @item --print-insn-syntax
1368 Print the syntax of instruction in case of error.
1370 @item --print-opcodes
1371 Print the list of instructions with syntax and then exit.
1373 @item --generate-example
1374 Print an example of instruction for each possible instruction and then exit.
1375 This option is only useful for testing @command{@value{AS}}.
1381 The following options are available when @command{@value{AS}} is configured
1382 for the SPARC architecture:
1385 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1386 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1387 Explicitly select a variant of the SPARC architecture.
1389 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1390 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1392 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1393 UltraSPARC extensions.
1395 @item -xarch=v8plus | -xarch=v8plusa
1396 For compatibility with the Solaris v9 assembler. These options are
1397 equivalent to -Av8plus and -Av8plusa, respectively.
1400 Warn when the assembler switches to another architecture.
1405 The following options are available when @value{AS} is configured for the 'c54x
1410 Enable extended addressing mode. All addresses and relocations will assume
1411 extended addressing (usually 23 bits).
1412 @item -mcpu=@var{CPU_VERSION}
1413 Sets the CPU version being compiled for.
1414 @item -merrors-to-file @var{FILENAME}
1415 Redirect error output to a file, for broken systems which don't support such
1416 behaviour in the shell.
1421 @c man begin OPTIONS
1422 The following options are available when @value{AS} is configured for
1427 This option sets the largest size of an object that can be referenced
1428 implicitly with the @code{gp} register. It is only accepted for targets that
1429 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1431 @cindex MIPS endianness
1432 @cindex endianness, MIPS
1433 @cindex big endian output, MIPS
1435 Generate ``big endian'' format output.
1437 @cindex little endian output, MIPS
1439 Generate ``little endian'' format output.
1457 Generate code for a particular MIPS Instruction Set Architecture level.
1458 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1459 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1460 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1461 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1462 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1463 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1464 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1465 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1466 MIPS64 Release 6 ISA processors, respectively.
1468 @item -march=@var{cpu}
1469 Generate code for a particular MIPS CPU.
1471 @item -mtune=@var{cpu}
1472 Schedule and tune for a particular MIPS CPU.
1476 Cause nops to be inserted if the read of the destination register
1477 of an mfhi or mflo instruction occurs in the following two instructions.
1480 @itemx -mno-fix-rm7000
1481 Cause nops to be inserted if a dmult or dmultu instruction is
1482 followed by a load instruction.
1485 @itemx -mno-fix-r5900
1486 Do not attempt to schedule the preceding instruction into the delay slot
1487 of a branch instruction placed at the end of a short loop of six
1488 instructions or fewer and always schedule a @code{nop} instruction there
1489 instead. The short loop bug under certain conditions causes loops to
1490 execute only once or twice, due to a hardware bug in the R5900 chip.
1494 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1495 section instead of the standard ELF .stabs sections.
1499 Control generation of @code{.pdr} sections.
1503 The register sizes are normally inferred from the ISA and ABI, but these
1504 flags force a certain group of registers to be treated as 32 bits wide at
1505 all times. @samp{-mgp32} controls the size of general-purpose registers
1506 and @samp{-mfp32} controls the size of floating-point registers.
1510 The register sizes are normally inferred from the ISA and ABI, but these
1511 flags force a certain group of registers to be treated as 64 bits wide at
1512 all times. @samp{-mgp64} controls the size of general-purpose registers
1513 and @samp{-mfp64} controls the size of floating-point registers.
1516 The register sizes are normally inferred from the ISA and ABI, but using
1517 this flag in combination with @samp{-mabi=32} enables an ABI variant
1518 which will operate correctly with floating-point registers which are
1522 @itemx -mno-odd-spreg
1523 Enable use of floating-point operations on odd-numbered single-precision
1524 registers when supported by the ISA. @samp{-mfpxx} implies
1525 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1529 Generate code for the MIPS 16 processor. This is equivalent to putting
1530 @code{.module mips16} at the start of the assembly file. @samp{-no-mips16}
1531 turns off this option.
1534 @itemx -mno-mips16e2
1535 Enable the use of MIPS16e2 instructions in MIPS16 mode. This is equivalent
1536 to putting @code{.module mips16e2} at the start of the assembly file.
1537 @samp{-mno-mips16e2} turns off this option.
1540 @itemx -mno-micromips
1541 Generate code for the microMIPS processor. This is equivalent to putting
1542 @code{.module micromips} at the start of the assembly file.
1543 @samp{-mno-micromips} turns off this option. This is equivalent to putting
1544 @code{.module nomicromips} at the start of the assembly file.
1547 @itemx -mno-smartmips
1548 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1549 equivalent to putting @code{.module smartmips} at the start of the assembly
1550 file. @samp{-mno-smartmips} turns off this option.
1554 Generate code for the MIPS-3D Application Specific Extension.
1555 This tells the assembler to accept MIPS-3D instructions.
1556 @samp{-no-mips3d} turns off this option.
1560 Generate code for the MDMX Application Specific Extension.
1561 This tells the assembler to accept MDMX instructions.
1562 @samp{-no-mdmx} turns off this option.
1566 Generate code for the DSP Release 1 Application Specific Extension.
1567 This tells the assembler to accept DSP Release 1 instructions.
1568 @samp{-mno-dsp} turns off this option.
1572 Generate code for the DSP Release 2 Application Specific Extension.
1573 This option implies @samp{-mdsp}.
1574 This tells the assembler to accept DSP Release 2 instructions.
1575 @samp{-mno-dspr2} turns off this option.
1579 Generate code for the DSP Release 3 Application Specific Extension.
1580 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1581 This tells the assembler to accept DSP Release 3 instructions.
1582 @samp{-mno-dspr3} turns off this option.
1586 Generate code for the MIPS SIMD Architecture Extension.
1587 This tells the assembler to accept MSA instructions.
1588 @samp{-mno-msa} turns off this option.
1592 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1593 This tells the assembler to accept XPA instructions.
1594 @samp{-mno-xpa} turns off this option.
1598 Generate code for the MT Application Specific Extension.
1599 This tells the assembler to accept MT instructions.
1600 @samp{-mno-mt} turns off this option.
1604 Generate code for the MCU Application Specific Extension.
1605 This tells the assembler to accept MCU instructions.
1606 @samp{-mno-mcu} turns off this option.
1610 Generate code for the MIPS cyclic redundancy check (CRC) Application
1611 Specific Extension. This tells the assembler to accept CRC instructions.
1612 @samp{-mno-crc} turns off this option.
1616 Generate code for the Global INValidate (GINV) Application Specific
1617 Extension. This tells the assembler to accept GINV instructions.
1618 @samp{-mno-ginv} turns off this option.
1620 @item -mloongson-mmi
1621 @itemx -mno-loongson-mmi
1622 Generate code for the Loongson MultiMedia extensions Instructions (MMI)
1623 Application Specific Extension. This tells the assembler to accept MMI
1625 @samp{-mno-loongson-mmi} turns off this option.
1627 @item -mloongson-cam
1628 @itemx -mno-loongson-cam
1629 Generate code for the Loongson Content Address Memory (CAM) instructions.
1630 This tells the assembler to accept Loongson CAM instructions.
1631 @samp{-mno-loongson-cam} turns off this option.
1633 @item -mloongson-ext
1634 @itemx -mno-loongson-ext
1635 Generate code for the Loongson EXTensions (EXT) instructions.
1636 This tells the assembler to accept Loongson EXT instructions.
1637 @samp{-mno-loongson-ext} turns off this option.
1639 @item -mloongson-ext2
1640 @itemx -mno-loongson-ext2
1641 Generate code for the Loongson EXTensions R2 (EXT2) instructions.
1642 This option implies @samp{-mloongson-ext}.
1643 This tells the assembler to accept Loongson EXT2 instructions.
1644 @samp{-mno-loongson-ext2} turns off this option.
1648 Only use 32-bit instruction encodings when generating code for the
1649 microMIPS processor. This option inhibits the use of any 16-bit
1650 instructions. This is equivalent to putting @code{.set insn32} at
1651 the start of the assembly file. @samp{-mno-insn32} turns off this
1652 option. This is equivalent to putting @code{.set noinsn32} at the
1653 start of the assembly file. By default @samp{-mno-insn32} is
1654 selected, allowing all instructions to be used.
1656 @item --construct-floats
1657 @itemx --no-construct-floats
1658 The @samp{--no-construct-floats} option disables the construction of
1659 double width floating point constants by loading the two halves of the
1660 value into the two single width floating point registers that make up
1661 the double width register. By default @samp{--construct-floats} is
1662 selected, allowing construction of these floating point constants.
1664 @item --relax-branch
1665 @itemx --no-relax-branch
1666 The @samp{--relax-branch} option enables the relaxation of out-of-range
1667 branches. By default @samp{--no-relax-branch} is selected, causing any
1668 out-of-range branches to produce an error.
1670 @item -mignore-branch-isa
1671 @itemx -mno-ignore-branch-isa
1672 Ignore branch checks for invalid transitions between ISA modes. The
1673 semantics of branches does not provide for an ISA mode switch, so in
1674 most cases the ISA mode a branch has been encoded for has to be the
1675 same as the ISA mode of the branch's target label. Therefore GAS has
1676 checks implemented that verify in branch assembly that the two ISA
1677 modes match. @samp{-mignore-branch-isa} disables these checks. By
1678 default @samp{-mno-ignore-branch-isa} is selected, causing any invalid
1679 branch requiring a transition between ISA modes to produce an error.
1681 @item -mnan=@var{encoding}
1682 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1683 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1686 @item --emulation=@var{name}
1687 This option was formerly used to switch between ELF and ECOFF output
1688 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1689 removed in GAS 2.24, so the option now serves little purpose.
1690 It is retained for backwards compatibility.
1692 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1693 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1694 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1695 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1696 preferred options instead.
1699 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1706 Control how to deal with multiplication overflow and division by zero.
1707 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1708 (and only work for Instruction Set Architecture level 2 and higher);
1709 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1713 When this option is used, @command{@value{AS}} will issue a warning every
1714 time it generates a nop instruction from a macro.
1720 The following options are available when @value{AS} is configured for
1726 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1727 The command-line option @samp{-nojsri2bsr} can be used to disable it.
1731 Enable or disable the silicon filter behaviour. By default this is disabled.
1732 The default can be overridden by the @samp{-sifilter} command-line option.
1735 Alter jump instructions for long displacements.
1737 @item -mcpu=[210|340]
1738 Select the cpu type on the target hardware. This controls which instructions
1742 Assemble for a big endian target.
1745 Assemble for a little endian target.
1754 @xref{Meta Options}, for the options available when @value{AS} is configured
1755 for a Meta processor.
1759 @c man begin OPTIONS
1760 The following options are available when @value{AS} is configured for a
1763 @c man begin INCLUDE
1764 @include c-metag.texi
1765 @c ended inside the included file
1770 @c man begin OPTIONS
1772 See the info pages for documentation of the MMIX-specific options.
1778 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1779 for a NDS32 processor.
1781 @c ended inside the included file
1785 @c man begin OPTIONS
1786 The following options are available when @value{AS} is configured for a
1789 @c man begin INCLUDE
1790 @include c-nds32.texi
1791 @c ended inside the included file
1798 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1799 for a PowerPC processor.
1803 @c man begin OPTIONS
1804 The following options are available when @value{AS} is configured for a
1807 @c man begin INCLUDE
1809 @c ended inside the included file
1817 @xref{RISC-V-Options}, for the options available when @value{AS} is configured
1818 for a RISC-V processor.
1822 @c man begin OPTIONS
1823 The following options are available when @value{AS} is configured for a
1826 @c man begin INCLUDE
1827 @include c-riscv.texi
1828 @c ended inside the included file
1833 @c man begin OPTIONS
1835 See the info pages for documentation of the RX-specific options.
1839 The following options are available when @value{AS} is configured for the s390
1845 Select the word size, either 31/32 bits or 64 bits.
1848 Select the architecture mode, either the Enterprise System
1849 Architecture (esa) or the z/Architecture mode (zarch).
1850 @item -march=@var{processor}
1851 Specify which s390 processor variant is the target, @samp{g5} (or
1852 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1853 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1854 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1855 @samp{z13} (or @samp{arch11}), @samp{z14} (or @samp{arch12}), or @samp{z15}
1858 @itemx -mno-regnames
1859 Allow or disallow symbolic names for registers.
1860 @item -mwarn-areg-zero
1861 Warn whenever the operand for a base or index register has been specified
1862 but evaluates to zero.
1870 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1871 for a TMS320C6000 processor.
1875 @c man begin OPTIONS
1876 The following options are available when @value{AS} is configured for a
1877 TMS320C6000 processor.
1879 @c man begin INCLUDE
1880 @include c-tic6x.texi
1881 @c ended inside the included file
1889 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1890 for a TILE-Gx processor.
1894 @c man begin OPTIONS
1895 The following options are available when @value{AS} is configured for a TILE-Gx
1898 @c man begin INCLUDE
1899 @include c-tilegx.texi
1900 @c ended inside the included file
1908 @xref{Visium Options}, for the options available when @value{AS} is configured
1909 for a Visium processor.
1913 @c man begin OPTIONS
1914 The following option is available when @value{AS} is configured for a Visium
1917 @c man begin INCLUDE
1918 @include c-visium.texi
1919 @c ended inside the included file
1927 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1928 for an Xtensa processor.
1932 @c man begin OPTIONS
1933 The following options are available when @value{AS} is configured for an
1936 @c man begin INCLUDE
1937 @include c-xtensa.texi
1938 @c ended inside the included file
1943 @c man begin OPTIONS
1946 The following options are available when @value{AS} is configured for
1947 a Z80 family processor.
1951 Assemble for Z80 processor.
1953 Assemble for R800 processor.
1955 Assemble for Z180 processor.
1957 Assemble for eZ80 processor in Z80 memory mode by default.
1959 Assemble for eZ80 processor in ADL memory mode by default.
1961 @item @code{-colonless}
1962 Accept colonless labels. All names at line begin are treated as labels.
1964 Accept assembler code produces by SDCC.
1966 @item @code{-strict}
1967 Accept documented instructions only.
1969 Accept all known Z80 instructions.
1970 @item @code{-with-inst=INST[,...]}
1971 @itemx @code{-Wnins INST[,...]}
1972 Enable specified undocumented instruction(s).
1973 @item @code{-without-inst=INST[,...]}
1974 @itemx @code{-Fins INST[,...]}
1975 Disable specified undocumented instruction(s).
1977 @item -ignore-undocumented-instructions
1979 Assemble undocumented Z80 instructions that also work on R800 without warning.
1980 @item -ignore-unportable-instructions
1982 Assemble all undocumented Z80 instructions without warning.
1983 @item -warn-undocumented-instructions
1985 Issue a warning for undocumented Z80 instructions that also work on R800.
1986 @item -warn-unportable-instructions
1988 Issue a warning for undocumented Z80 instructions that do not work on R800.
1989 @item -forbid-undocumented-instructions
1991 Treat all undocumented instructions as errors.
1992 @item -forbid-unportable-instructions
1994 Treat undocumented Z80 instructions that do not work on R800 as errors.
1997 Folowing undocumented instructions may be enabled/disabled by
1998 @code{-with-inst}/@code{-without-inst}:
2000 @item @code{idx-reg-halves}
2001 All operations with halves of index registers (IXL, IXH, IYL, IYH).
2003 SLI or SLL instruction.
2004 @item @code{op-ii-ld}
2005 Istructions like @code{<op> (<ii>+<d>),<r>}, where @code{<op>}
2006 is shift or bit manipulation instruction (RLC, SLA, SET, RES...).
2008 Instruction @code{IN F,(C)}.
2009 @item @code{out-c-0}
2010 Instruction @code{OUT (C),0}
2017 * Manual:: Structure of this Manual
2018 * GNU Assembler:: The GNU Assembler
2019 * Object Formats:: Object File Formats
2020 * Command Line:: Command Line
2021 * Input Files:: Input Files
2022 * Object:: Output (Object) File
2023 * Errors:: Error and Warning Messages
2027 @section Structure of this Manual
2029 @cindex manual, structure and purpose
2030 This manual is intended to describe what you need to know to use
2031 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
2032 notation for symbols, constants, and expressions; the directives that
2033 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
2036 We also cover special features in the @value{TARGET}
2037 configuration of @command{@value{AS}}, including assembler directives.
2040 This manual also describes some of the machine-dependent features of
2041 various flavors of the assembler.
2044 @cindex machine instructions (not covered)
2045 On the other hand, this manual is @emph{not} intended as an introduction
2046 to programming in assembly language---let alone programming in general!
2047 In a similar vein, we make no attempt to introduce the machine
2048 architecture; we do @emph{not} describe the instruction set, standard
2049 mnemonics, registers or addressing modes that are standard to a
2050 particular architecture.
2052 You may want to consult the manufacturer's
2053 machine architecture manual for this information.
2057 For information on the H8/300 machine instruction set, see @cite{H8/300
2058 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
2059 Programming Manual} (Renesas).
2062 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
2063 see @cite{SH-Microcomputer User's Manual} (Renesas) or
2064 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
2065 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
2068 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
2072 @c I think this is premature---doc@cygnus.com, 17jan1991
2074 Throughout this manual, we assume that you are running @dfn{GNU},
2075 the portable operating system from the @dfn{Free Software
2076 Foundation, Inc.}. This restricts our attention to certain kinds of
2077 computer (in particular, the kinds of computers that @sc{gnu} can run on);
2078 once this assumption is granted examples and definitions need less
2081 @command{@value{AS}} is part of a team of programs that turn a high-level
2082 human-readable series of instructions into a low-level
2083 computer-readable series of instructions. Different versions of
2084 @command{@value{AS}} are used for different kinds of computer.
2087 @c There used to be a section "Terminology" here, which defined
2088 @c "contents", "byte", "word", and "long". Defining "word" to any
2089 @c particular size is confusing when the .word directive may generate 16
2090 @c bits on one machine and 32 bits on another; in general, for the user
2091 @c version of this manual, none of these terms seem essential to define.
2092 @c They were used very little even in the former draft of the manual;
2093 @c this draft makes an effort to avoid them (except in names of
2097 @section The GNU Assembler
2099 @c man begin DESCRIPTION
2101 @sc{gnu} @command{as} is really a family of assemblers.
2103 This manual describes @command{@value{AS}}, a member of that family which is
2104 configured for the @value{TARGET} architectures.
2106 If you use (or have used) the @sc{gnu} assembler on one architecture, you
2107 should find a fairly similar environment when you use it on another
2108 architecture. Each version has much in common with the others,
2109 including object file formats, most assembler directives (often called
2110 @dfn{pseudo-ops}) and assembler syntax.@refill
2112 @cindex purpose of @sc{gnu} assembler
2113 @command{@value{AS}} is primarily intended to assemble the output of the
2114 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
2115 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
2116 assemble correctly everything that other assemblers for the same
2117 machine would assemble.
2119 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
2122 @c This remark should appear in generic version of manual; assumption
2123 @c here is that generic version sets M680x0.
2124 This doesn't mean @command{@value{AS}} always uses the same syntax as another
2125 assembler for the same architecture; for example, we know of several
2126 incompatible versions of 680x0 assembly language syntax.
2131 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
2132 program in one pass of the source file. This has a subtle impact on the
2133 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
2135 @node Object Formats
2136 @section Object File Formats
2138 @cindex object file format
2139 The @sc{gnu} assembler can be configured to produce several alternative
2140 object file formats. For the most part, this does not affect how you
2141 write assembly language programs; but directives for debugging symbols
2142 are typically different in different file formats. @xref{Symbol
2143 Attributes,,Symbol Attributes}.
2146 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
2147 @value{OBJ-NAME} format object files.
2149 @c The following should exhaust all configs that set MULTI-OBJ, ideally
2151 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
2152 SOM or ELF format object files.
2157 @section Command Line
2159 @cindex command line conventions
2161 After the program name @command{@value{AS}}, the command line may contain
2162 options and file names. Options may appear in any order, and may be
2163 before, after, or between file names. The order of file names is
2166 @cindex standard input, as input file
2168 @file{--} (two hyphens) by itself names the standard input file
2169 explicitly, as one of the files for @command{@value{AS}} to assemble.
2171 @cindex options, command line
2172 Except for @samp{--} any command-line argument that begins with a
2173 hyphen (@samp{-}) is an option. Each option changes the behavior of
2174 @command{@value{AS}}. No option changes the way another option works. An
2175 option is a @samp{-} followed by one or more letters; the case of
2176 the letter is important. All options are optional.
2178 Some options expect exactly one file name to follow them. The file
2179 name may either immediately follow the option's letter (compatible
2180 with older assemblers) or it may be the next command argument (@sc{gnu}
2181 standard). These two command lines are equivalent:
2184 @value{AS} -o my-object-file.o mumble.s
2185 @value{AS} -omy-object-file.o mumble.s
2189 @section Input Files
2192 @cindex source program
2193 @cindex files, input
2194 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2195 describe the program input to one run of @command{@value{AS}}. The program may
2196 be in one or more files; how the source is partitioned into files
2197 doesn't change the meaning of the source.
2199 @c I added "con" prefix to "catenation" just to prove I can overcome my
2200 @c APL training... doc@cygnus.com
2201 The source program is a concatenation of the text in all the files, in the
2204 @c man begin DESCRIPTION
2205 Each time you run @command{@value{AS}} it assembles exactly one source
2206 program. The source program is made up of one or more files.
2207 (The standard input is also a file.)
2209 You give @command{@value{AS}} a command line that has zero or more input file
2210 names. The input files are read (from left file name to right). A
2211 command-line argument (in any position) that has no special meaning
2212 is taken to be an input file name.
2214 If you give @command{@value{AS}} no file names it attempts to read one input file
2215 from the @command{@value{AS}} standard input, which is normally your terminal. You
2216 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2219 Use @samp{--} if you need to explicitly name the standard input file
2220 in your command line.
2222 If the source is empty, @command{@value{AS}} produces a small, empty object
2227 @subheading Filenames and Line-numbers
2229 @cindex input file linenumbers
2230 @cindex line numbers, in input files
2231 There are two ways of locating a line in the input file (or files) and
2232 either may be used in reporting error messages. One way refers to a line
2233 number in a physical file; the other refers to a line number in a
2234 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2236 @dfn{Physical files} are those files named in the command line given
2237 to @command{@value{AS}}.
2239 @dfn{Logical files} are simply names declared explicitly by assembler
2240 directives; they bear no relation to physical files. Logical file names help
2241 error messages reflect the original source file, when @command{@value{AS}} source
2242 is itself synthesized from other files. @command{@value{AS}} understands the
2243 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2244 @ref{File,,@code{.file}}.
2247 @section Output (Object) File
2253 Every time you run @command{@value{AS}} it produces an output file, which is
2254 your assembly language program translated into numbers. This file
2255 is the object file. Its default name is @code{a.out}.
2256 You can give it another name by using the @option{-o} option. Conventionally,
2257 object file names end with @file{.o}. The default name is used for historical
2258 reasons: older assemblers were capable of assembling self-contained programs
2259 directly into a runnable program. (For some formats, this isn't currently
2260 possible, but it can be done for the @code{a.out} format.)
2264 The object file is meant for input to the linker @code{@value{LD}}. It contains
2265 assembled program code, information to help @code{@value{LD}} integrate
2266 the assembled program into a runnable file, and (optionally) symbolic
2267 information for the debugger.
2269 @c link above to some info file(s) like the description of a.out.
2270 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2273 @section Error and Warning Messages
2275 @c man begin DESCRIPTION
2277 @cindex error messages
2278 @cindex warning messages
2279 @cindex messages from assembler
2280 @command{@value{AS}} may write warnings and error messages to the standard error
2281 file (usually your terminal). This should not happen when a compiler
2282 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2283 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2284 grave problem that stops the assembly.
2288 @cindex format of warning messages
2289 Warning messages have the format
2292 file_name:@b{NNN}:Warning Message Text
2296 @cindex file names and line numbers, in warnings/errors
2297 (where @b{NNN} is a line number). If both a logical file name
2298 (@pxref{File,,@code{.file}}) and a logical line number
2300 (@pxref{Line,,@code{.line}})
2302 have been given then they will be used, otherwise the file name and line number
2303 in the current assembler source file will be used. The message text is
2304 intended to be self explanatory (in the grand Unix tradition).
2306 Note the file name must be set via the logical version of the @code{.file}
2307 directive, not the DWARF2 version of the @code{.file} directive. For example:
2311 error_assembler_source
2317 produces this output:
2321 asm.s:2: Error: no such instruction: `error_assembler_source'
2322 foo.c:31: Error: no such instruction: `error_c_source'
2325 @cindex format of error messages
2326 Error messages have the format
2329 file_name:@b{NNN}:FATAL:Error Message Text
2332 The file name and line number are derived as for warning
2333 messages. The actual message text may be rather less explanatory
2334 because many of them aren't supposed to happen.
2337 @chapter Command-Line Options
2339 @cindex options, all versions of assembler
2340 This chapter describes command-line options available in @emph{all}
2341 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2342 for options specific
2344 to the @value{TARGET} target.
2347 to particular machine architectures.
2350 @c man begin DESCRIPTION
2352 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2353 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2354 The assembler arguments must be separated from each other (and the @samp{-Wa})
2355 by commas. For example:
2358 gcc -c -g -O -Wa,-alh,-L file.c
2362 This passes two options to the assembler: @samp{-alh} (emit a listing to
2363 standard output with high-level and assembly source) and @samp{-L} (retain
2364 local symbols in the symbol table).
2366 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2367 command-line options are automatically passed to the assembler by the compiler.
2368 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2369 precisely what options it passes to each compilation pass, including the
2375 * a:: -a[cdghlns] enable listings
2376 * alternate:: --alternate enable alternate macro syntax
2377 * D:: -D for compatibility
2378 * f:: -f to work faster
2379 * I:: -I for .include search path
2380 @ifclear DIFF-TBL-KLUGE
2381 * K:: -K for compatibility
2383 @ifset DIFF-TBL-KLUGE
2384 * K:: -K for difference tables
2387 * L:: -L to retain local symbols
2388 * listing:: --listing-XXX to configure listing output
2389 * M:: -M or --mri to assemble in MRI compatibility mode
2390 * MD:: --MD for dependency tracking
2391 * no-pad-sections:: --no-pad-sections to stop section padding
2392 * o:: -o to name the object file
2393 * R:: -R to join data and text sections
2394 * statistics:: --statistics to see statistics about assembly
2395 * traditional-format:: --traditional-format for compatible output
2396 * v:: -v to announce version
2397 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2398 * Z:: -Z to make object file even after errors
2402 @section Enable Listings: @option{-a[cdghlns]}
2412 @cindex listings, enabling
2413 @cindex assembly listings, enabling
2415 These options enable listing output from the assembler. By itself,
2416 @samp{-a} requests high-level, assembly, and symbols listing.
2417 You can use other letters to select specific options for the list:
2418 @samp{-ah} requests a high-level language listing,
2419 @samp{-al} requests an output-program assembly listing, and
2420 @samp{-as} requests a symbol table listing.
2421 High-level listings require that a compiler debugging option like
2422 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2425 Use the @samp{-ag} option to print a first section with general assembly
2426 information, like @value{AS} version, switches passed, or time stamp.
2428 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2429 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2430 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2431 omitted from the listing.
2433 Use the @samp{-ad} option to omit debugging directives from the
2436 Once you have specified one of these options, you can further control
2437 listing output and its appearance using the directives @code{.list},
2438 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2440 The @samp{-an} option turns off all forms processing.
2441 If you do not request listing output with one of the @samp{-a} options, the
2442 listing-control directives have no effect.
2444 The letters after @samp{-a} may be combined into one option,
2445 @emph{e.g.}, @samp{-aln}.
2447 Note if the assembler source is coming from the standard input (e.g.,
2449 is being created by @code{@value{GCC}} and the @samp{-pipe} command-line switch
2450 is being used) then the listing will not contain any comments or preprocessor
2451 directives. This is because the listing code buffers input source lines from
2452 stdin only after they have been preprocessed by the assembler. This reduces
2453 memory usage and makes the code more efficient.
2456 @section @option{--alternate}
2459 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2462 @section @option{-D}
2465 This option has no effect whatsoever, but it is accepted to make it more
2466 likely that scripts written for other assemblers also work with
2467 @command{@value{AS}}.
2470 @section Work Faster: @option{-f}
2473 @cindex trusted compiler
2474 @cindex faster processing (@option{-f})
2475 @samp{-f} should only be used when assembling programs written by a
2476 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2477 and comment preprocessing on
2478 the input file(s) before assembling them. @xref{Preprocessing,
2482 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2483 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2488 @section @code{.include} Search Path: @option{-I} @var{path}
2490 @kindex -I @var{path}
2491 @cindex paths for @code{.include}
2492 @cindex search path for @code{.include}
2493 @cindex @code{include} directive search path
2494 Use this option to add a @var{path} to the list of directories
2495 @command{@value{AS}} searches for files specified in @code{.include}
2496 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2497 many times as necessary to include a variety of paths. The current
2498 working directory is always searched first; after that, @command{@value{AS}}
2499 searches any @samp{-I} directories in the same order as they were
2500 specified (left to right) on the command line.
2503 @section Difference Tables: @option{-K}
2506 @ifclear DIFF-TBL-KLUGE
2507 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2508 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2509 where it can be used to warn when the assembler alters the machine code
2510 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2511 family does not have the addressing limitations that sometimes lead to this
2512 alteration on other platforms.
2515 @ifset DIFF-TBL-KLUGE
2516 @cindex difference tables, warning
2517 @cindex warning for altered difference tables
2518 @command{@value{AS}} sometimes alters the code emitted for directives of the
2519 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2520 You can use the @samp{-K} option if you want a warning issued when this
2525 @section Include Local Symbols: @option{-L}
2528 @cindex local symbols, retaining in output
2529 Symbols beginning with system-specific local label prefixes, typically
2530 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2531 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2532 such symbols when debugging, because they are intended for the use of
2533 programs (like compilers) that compose assembler programs, not for your
2534 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2535 such symbols, so you do not normally debug with them.
2537 This option tells @command{@value{AS}} to retain those local symbols
2538 in the object file. Usually if you do this you also tell the linker
2539 @code{@value{LD}} to preserve those symbols.
2542 @section Configuring listing output: @option{--listing}
2544 The listing feature of the assembler can be enabled via the command-line switch
2545 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2546 hex dump of the corresponding locations in the output object file, and displays
2547 them as a listing file. The format of this listing can be controlled by
2548 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2549 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2550 @code{.psize} (@pxref{Psize}), and
2551 @code{.eject} (@pxref{Eject}) and also by the following switches:
2554 @item --listing-lhs-width=@samp{number}
2555 @kindex --listing-lhs-width
2556 @cindex Width of first line disassembly output
2557 Sets the maximum width, in words, of the first line of the hex byte dump. This
2558 dump appears on the left hand side of the listing output.
2560 @item --listing-lhs-width2=@samp{number}
2561 @kindex --listing-lhs-width2
2562 @cindex Width of continuation lines of disassembly output
2563 Sets the maximum width, in words, of any further lines of the hex byte dump for
2564 a given input source line. If this value is not specified, it defaults to being
2565 the same as the value specified for @samp{--listing-lhs-width}. If neither
2566 switch is used the default is to one.
2568 @item --listing-rhs-width=@samp{number}
2569 @kindex --listing-rhs-width
2570 @cindex Width of source line output
2571 Sets the maximum width, in characters, of the source line that is displayed
2572 alongside the hex dump. The default value for this parameter is 100. The
2573 source line is displayed on the right hand side of the listing output.
2575 @item --listing-cont-lines=@samp{number}
2576 @kindex --listing-cont-lines
2577 @cindex Maximum number of continuation lines
2578 Sets the maximum number of continuation lines of hex dump that will be
2579 displayed for a given single line of source input. The default value is 4.
2583 @section Assemble in MRI Compatibility Mode: @option{-M}
2586 @cindex MRI compatibility mode
2587 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2588 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2589 compatible with the @code{ASM68K} assembler from Microtec Research.
2590 The exact nature of the
2591 MRI syntax will not be documented here; see the MRI manuals for more
2592 information. Note in particular that the handling of macros and macro
2593 arguments is somewhat different. The purpose of this option is to permit
2594 assembling existing MRI assembler code using @command{@value{AS}}.
2596 The MRI compatibility is not complete. Certain operations of the MRI assembler
2597 depend upon its object file format, and can not be supported using other object
2598 file formats. Supporting these would require enhancing each object file format
2599 individually. These are:
2602 @item global symbols in common section
2604 The m68k MRI assembler supports common sections which are merged by the linker.
2605 Other object file formats do not support this. @command{@value{AS}} handles
2606 common sections by treating them as a single common symbol. It permits local
2607 symbols to be defined within a common section, but it can not support global
2608 symbols, since it has no way to describe them.
2610 @item complex relocations
2612 The MRI assemblers support relocations against a negated section address, and
2613 relocations which combine the start addresses of two or more sections. These
2614 are not support by other object file formats.
2616 @item @code{END} pseudo-op specifying start address
2618 The MRI @code{END} pseudo-op permits the specification of a start address.
2619 This is not supported by other object file formats. The start address may
2620 instead be specified using the @option{-e} option to the linker, or in a linker
2623 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2625 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2626 name to the output file. This is not supported by other object file formats.
2628 @item @code{ORG} pseudo-op
2630 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2631 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2632 which changes the location within the current section. Absolute sections are
2633 not supported by other object file formats. The address of a section may be
2634 assigned within a linker script.
2637 There are some other features of the MRI assembler which are not supported by
2638 @command{@value{AS}}, typically either because they are difficult or because they
2639 seem of little consequence. Some of these may be supported in future releases.
2643 @item EBCDIC strings
2645 EBCDIC strings are not supported.
2647 @item packed binary coded decimal
2649 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2650 and @code{DCB.P} pseudo-ops are not supported.
2652 @item @code{FEQU} pseudo-op
2654 The m68k @code{FEQU} pseudo-op is not supported.
2656 @item @code{NOOBJ} pseudo-op
2658 The m68k @code{NOOBJ} pseudo-op is not supported.
2660 @item @code{OPT} branch control options
2662 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2663 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2664 relaxes all branches, whether forward or backward, to an appropriate size, so
2665 these options serve no purpose.
2667 @item @code{OPT} list control options
2669 The following m68k @code{OPT} list control options are ignored: @code{C},
2670 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2671 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2673 @item other @code{OPT} options
2675 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2676 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2678 @item @code{OPT} @code{D} option is default
2680 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2681 @code{OPT NOD} may be used to turn it off.
2683 @item @code{XREF} pseudo-op.
2685 The m68k @code{XREF} pseudo-op is ignored.
2690 @section Dependency Tracking: @option{--MD}
2693 @cindex dependency tracking
2696 @command{@value{AS}} can generate a dependency file for the file it creates. This
2697 file consists of a single rule suitable for @code{make} describing the
2698 dependencies of the main source file.
2700 The rule is written to the file named in its argument.
2702 This feature is used in the automatic updating of makefiles.
2704 @node no-pad-sections
2705 @section Output Section Padding
2706 @kindex --no-pad-sections
2707 @cindex output section padding
2708 Normally the assembler will pad the end of each output section up to its
2709 alignment boundary. But this can waste space, which can be significant on
2710 memory constrained targets. So the @option{--no-pad-sections} option will
2711 disable this behaviour.
2714 @section Name the Object File: @option{-o}
2717 @cindex naming object file
2718 @cindex object file name
2719 There is always one object file output when you run @command{@value{AS}}. By
2720 default it has the name @file{a.out}.
2721 You use this option (which takes exactly one filename) to give the
2722 object file a different name.
2724 Whatever the object file is called, @command{@value{AS}} overwrites any
2725 existing file of the same name.
2728 @section Join Data and Text Sections: @option{-R}
2731 @cindex data and text sections, joining
2732 @cindex text and data sections, joining
2733 @cindex joining text and data sections
2734 @cindex merging text and data sections
2735 @option{-R} tells @command{@value{AS}} to write the object file as if all
2736 data-section data lives in the text section. This is only done at
2737 the very last moment: your binary data are the same, but data
2738 section parts are relocated differently. The data section part of
2739 your object file is zero bytes long because all its bytes are
2740 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2742 When you specify @option{-R} it would be possible to generate shorter
2743 address displacements (because we do not have to cross between text and
2744 data section). We refrain from doing this simply for compatibility with
2745 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2748 When @command{@value{AS}} is configured for COFF or ELF output,
2749 this option is only useful if you use sections named @samp{.text} and
2754 @option{-R} is not supported for any of the HPPA targets. Using
2755 @option{-R} generates a warning from @command{@value{AS}}.
2759 @section Display Assembly Statistics: @option{--statistics}
2761 @kindex --statistics
2762 @cindex statistics, about assembly
2763 @cindex time, total for assembly
2764 @cindex space used, maximum for assembly
2765 Use @samp{--statistics} to display two statistics about the resources used by
2766 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2767 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2770 @node traditional-format
2771 @section Compatible Output: @option{--traditional-format}
2773 @kindex --traditional-format
2774 For some targets, the output of @command{@value{AS}} is different in some ways
2775 from the output of some existing assembler. This switch requests
2776 @command{@value{AS}} to use the traditional format instead.
2778 For example, it disables the exception frame optimizations which
2779 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2782 @section Announce Version: @option{-v}
2786 @cindex assembler version
2787 @cindex version of assembler
2788 You can find out what version of as is running by including the
2789 option @samp{-v} (which you can also spell as @samp{-version}) on the
2793 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2795 @command{@value{AS}} should never give a warning or error message when
2796 assembling compiler output. But programs written by people often
2797 cause @command{@value{AS}} to give a warning that a particular assumption was
2798 made. All such warnings are directed to the standard error file.
2802 @cindex suppressing warnings
2803 @cindex warnings, suppressing
2804 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2805 This only affects the warning messages: it does not change any particular of
2806 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2809 @kindex --fatal-warnings
2810 @cindex errors, caused by warnings
2811 @cindex warnings, causing error
2812 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2813 files that generate warnings to be in error.
2816 @cindex warnings, switching on
2817 You can switch these options off again by specifying @option{--warn}, which
2818 causes warnings to be output as usual.
2821 @section Generate Object File in Spite of Errors: @option{-Z}
2822 @cindex object file, after errors
2823 @cindex errors, continuing after
2824 After an error message, @command{@value{AS}} normally produces no output. If for
2825 some reason you are interested in object file output even after
2826 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2827 option. If there are any errors, @command{@value{AS}} continues anyways, and
2828 writes an object file after a final warning message of the form @samp{@var{n}
2829 errors, @var{m} warnings, generating bad object file.}
2834 @cindex machine-independent syntax
2835 @cindex syntax, machine-independent
2836 This chapter describes the machine-independent syntax allowed in a
2837 source file. @command{@value{AS}} syntax is similar to what many other
2838 assemblers use; it is inspired by the BSD 4.2
2843 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2847 * Preprocessing:: Preprocessing
2848 * Whitespace:: Whitespace
2849 * Comments:: Comments
2850 * Symbol Intro:: Symbols
2851 * Statements:: Statements
2852 * Constants:: Constants
2856 @section Preprocessing
2858 @cindex preprocessing
2859 The @command{@value{AS}} internal preprocessor:
2861 @cindex whitespace, removed by preprocessor
2863 adjusts and removes extra whitespace. It leaves one space or tab before
2864 the keywords on a line, and turns any other whitespace on the line into
2867 @cindex comments, removed by preprocessor
2869 removes all comments, replacing them with a single space, or an
2870 appropriate number of newlines.
2872 @cindex constants, converted by preprocessor
2874 converts character constants into the appropriate numeric values.
2877 It does not do macro processing, include file handling, or
2878 anything else you may get from your C compiler's preprocessor. You can
2879 do include file processing with the @code{.include} directive
2880 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2881 to get other ``CPP'' style preprocessing by giving the input file a
2882 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2883 Output, gcc info, Using GNU CC}.
2885 Excess whitespace, comments, and character constants
2886 cannot be used in the portions of the input text that are not
2889 @cindex turning preprocessing on and off
2890 @cindex preprocessing, turning on and off
2893 If the first line of an input file is @code{#NO_APP} or if you use the
2894 @samp{-f} option, whitespace and comments are not removed from the input file.
2895 Within an input file, you can ask for whitespace and comment removal in
2896 specific portions of the by putting a line that says @code{#APP} before the
2897 text that may contain whitespace or comments, and putting a line that says
2898 @code{#NO_APP} after this text. This feature is mainly intend to support
2899 @code{asm} statements in compilers whose output is otherwise free of comments
2906 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2907 Whitespace is used to separate symbols, and to make programs neater for
2908 people to read. Unless within character constants
2909 (@pxref{Characters,,Character Constants}), any whitespace means the same
2910 as exactly one space.
2916 There are two ways of rendering comments to @command{@value{AS}}. In both
2917 cases the comment is equivalent to one space.
2919 Anything from @samp{/*} through the next @samp{*/} is a comment.
2920 This means you may not nest these comments.
2924 The only way to include a newline ('\n') in a comment
2925 is to use this sort of comment.
2928 /* This sort of comment does not nest. */
2931 @cindex line comment character
2932 Anything from a @dfn{line comment} character up to the next newline is
2933 considered a comment and is ignored. The line comment character is target
2934 specific, and some targets multiple comment characters. Some targets also have
2935 line comment characters that only work if they are the first character on a
2936 line. Some targets use a sequence of two characters to introduce a line
2937 comment. Some targets can also change their line comment characters depending
2938 upon command-line options that have been used. For more details see the
2939 @emph{Syntax} section in the documentation for individual targets.
2941 If the line comment character is the hash sign (@samp{#}) then it still has the
2942 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2943 to specify logical line numbers:
2946 @cindex lines starting with @code{#}
2947 @cindex logical line numbers
2948 To be compatible with past assemblers, lines that begin with @samp{#} have a
2949 special interpretation. Following the @samp{#} should be an absolute
2950 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2951 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2952 new logical file name. The rest of the line, if any, should be whitespace.
2954 If the first non-whitespace characters on the line are not numeric,
2955 the line is ignored. (Just like a comment.)
2958 # This is an ordinary comment.
2959 # 42-6 "new_file_name" # New logical file name
2960 # This is logical line # 36.
2962 This feature is deprecated, and may disappear from future versions
2963 of @command{@value{AS}}.
2968 @cindex characters used in symbols
2969 @ifclear SPECIAL-SYMS
2970 A @dfn{symbol} is one or more characters chosen from the set of all
2971 letters (both upper and lower case), digits and the three characters
2977 A @dfn{symbol} is one or more characters chosen from the set of all
2978 letters (both upper and lower case), digits and the three characters
2979 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2985 On most machines, you can also use @code{$} in symbol names; exceptions
2986 are noted in @ref{Machine Dependencies}.
2988 No symbol may begin with a digit. Case is significant.
2989 There is no length limit; all characters are significant. Multibyte characters
2990 are supported. Symbols are delimited by characters not in that set, or by the
2991 beginning of a file (since the source program must end with a newline, the end
2992 of a file is not a possible symbol delimiter). @xref{Symbols}.
2994 Symbol names may also be enclosed in double quote @code{"} characters. In such
2995 cases any characters are allowed, except for the NUL character. If a double
2996 quote character is to be included in the symbol name it must be preceeded by a
2997 backslash @code{\} character.
2998 @cindex length of symbols
3003 @cindex statements, structure of
3004 @cindex line separator character
3005 @cindex statement separator character
3007 A @dfn{statement} ends at a newline character (@samp{\n}) or a
3008 @dfn{line separator character}. The line separator character is target
3009 specific and described in the @emph{Syntax} section of each
3010 target's documentation. Not all targets support a line separator character.
3011 The newline or line separator character is considered to be part of the
3012 preceding statement. Newlines and separators within character constants are an
3013 exception: they do not end statements.
3015 @cindex newline, required at file end
3016 @cindex EOF, newline must precede
3017 It is an error to end any statement with end-of-file: the last
3018 character of any input file should be a newline.@refill
3020 An empty statement is allowed, and may include whitespace. It is ignored.
3022 @cindex instructions and directives
3023 @cindex directives and instructions
3024 @c "key symbol" is not used elsewhere in the document; seems pedantic to
3025 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
3027 A statement begins with zero or more labels, optionally followed by a
3028 key symbol which determines what kind of statement it is. The key
3029 symbol determines the syntax of the rest of the statement. If the
3030 symbol begins with a dot @samp{.} then the statement is an assembler
3031 directive: typically valid for any computer. If the symbol begins with
3032 a letter the statement is an assembly language @dfn{instruction}: it
3033 assembles into a machine language instruction.
3035 Different versions of @command{@value{AS}} for different computers
3036 recognize different instructions. In fact, the same symbol may
3037 represent a different instruction in a different computer's assembly
3041 @cindex @code{:} (label)
3042 @cindex label (@code{:})
3043 A label is a symbol immediately followed by a colon (@code{:}).
3044 Whitespace before a label or after a colon is permitted, but you may not
3045 have whitespace between a label's symbol and its colon. @xref{Labels}.
3048 For HPPA targets, labels need not be immediately followed by a colon, but
3049 the definition of a label must begin in column zero. This also implies that
3050 only one label may be defined on each line.
3054 label: .directive followed by something
3055 another_label: # This is an empty statement.
3056 instruction operand_1, operand_2, @dots{}
3063 A constant is a number, written so that its value is known by
3064 inspection, without knowing any context. Like this:
3067 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
3068 .ascii "Ring the bell\7" # A string constant.
3069 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
3070 .float 0f-314159265358979323846264338327\
3071 95028841971.693993751E-40 # - pi, a flonum.
3076 * Characters:: Character Constants
3077 * Numbers:: Number Constants
3081 @subsection Character Constants
3083 @cindex character constants
3084 @cindex constants, character
3085 There are two kinds of character constants. A @dfn{character} stands
3086 for one character in one byte and its value may be used in
3087 numeric expressions. String constants (properly called string
3088 @emph{literals}) are potentially many bytes and their values may not be
3089 used in arithmetic expressions.
3093 * Chars:: Characters
3097 @subsubsection Strings
3099 @cindex string constants
3100 @cindex constants, string
3101 A @dfn{string} is written between double-quotes. It may contain
3102 double-quotes or null characters. The way to get special characters
3103 into a string is to @dfn{escape} these characters: precede them with
3104 a backslash @samp{\} character. For example @samp{\\} represents
3105 one backslash: the first @code{\} is an escape which tells
3106 @command{@value{AS}} to interpret the second character literally as a backslash
3107 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
3108 escape character). The complete list of escapes follows.
3110 @cindex escape codes, character
3111 @cindex character escape codes
3112 @c NOTE: Cindex entries must not start with a backlash character.
3113 @c NOTE: This confuses the pdf2texi script when it is creating the
3114 @c NOTE: index based upon the first character and so it generates:
3115 @c NOTE: \initial {\\}
3116 @c NOTE: which then results in the error message:
3117 @c NOTE: Argument of \\ has an extra }.
3118 @c NOTE: So in the index entries below a space character has been
3119 @c NOTE: prepended to avoid this problem.
3122 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
3124 @cindex @code{ \b} (backspace character)
3125 @cindex backspace (@code{\b})
3127 Mnemonic for backspace; for ASCII this is octal code 010.
3130 @c Mnemonic for EOText; for ASCII this is octal code 004.
3132 @cindex @code{ \f} (formfeed character)
3133 @cindex formfeed (@code{\f})
3135 Mnemonic for FormFeed; for ASCII this is octal code 014.
3137 @cindex @code{ \n} (newline character)
3138 @cindex newline (@code{\n})
3140 Mnemonic for newline; for ASCII this is octal code 012.
3143 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
3145 @cindex @code{ \r} (carriage return character)
3146 @cindex carriage return (@code{backslash-r})
3148 Mnemonic for carriage-Return; for ASCII this is octal code 015.
3151 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
3152 @c other assemblers.
3154 @cindex @code{ \t} (tab)
3155 @cindex tab (@code{\t})
3157 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
3160 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
3161 @c @item \x @var{digit} @var{digit} @var{digit}
3162 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
3164 @cindex @code{ \@var{ddd}} (octal character code)
3165 @cindex octal character code (@code{\@var{ddd}})
3166 @item \ @var{digit} @var{digit} @var{digit}
3167 An octal character code. The numeric code is 3 octal digits.
3168 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
3169 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3171 @cindex @code{ \@var{xd...}} (hex character code)
3172 @cindex hex character code (@code{\@var{xd...}})
3173 @item \@code{x} @var{hex-digits...}
3174 A hex character code. All trailing hex digits are combined. Either upper or
3175 lower case @code{x} works.
3177 @cindex @code{ \\} (@samp{\} character)
3178 @cindex backslash (@code{\\})
3180 Represents one @samp{\} character.
3183 @c Represents one @samp{'} (accent acute) character.
3184 @c This is needed in single character literals
3185 @c (@xref{Characters,,Character Constants}.) to represent
3188 @cindex @code{ \"} (doublequote character)
3189 @cindex doublequote (@code{\"})
3191 Represents one @samp{"} character. Needed in strings to represent
3192 this character, because an unescaped @samp{"} would end the string.
3194 @item \ @var{anything-else}
3195 Any other character when escaped by @kbd{\} gives a warning, but
3196 assembles as if the @samp{\} was not present. The idea is that if
3197 you used an escape sequence you clearly didn't want the literal
3198 interpretation of the following character. However @command{@value{AS}} has no
3199 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3200 code and warns you of the fact.
3203 Which characters are escapable, and what those escapes represent,
3204 varies widely among assemblers. The current set is what we think
3205 the BSD 4.2 assembler recognizes, and is a subset of what most C
3206 compilers recognize. If you are in doubt, do not use an escape
3210 @subsubsection Characters
3212 @cindex single character constant
3213 @cindex character, single
3214 @cindex constant, single character
3215 A single character may be written as a single quote immediately followed by
3216 that character. Some backslash escapes apply to characters, @code{\b},
3217 @code{\f}, @code{\n}, @code{\r}, @code{\t}, and @code{\"} with the same meaning
3218 as for strings, plus @code{\'} for a single quote. So if you want to write the
3219 character backslash, you must write @kbd{'\\} where the first @code{\} escapes
3220 the second @code{\}. As you can see, the quote is an acute accent, not a grave
3223 @ifclear abnormal-separator
3224 (or semicolon @samp{;})
3226 @ifset abnormal-separator
3228 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3233 immediately following an acute accent is taken as a literal character
3234 and does not count as the end of a statement. The value of a character
3235 constant in a numeric expression is the machine's byte-wide code for
3236 that character. @command{@value{AS}} assumes your character code is ASCII:
3237 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3240 @subsection Number Constants
3242 @cindex constants, number
3243 @cindex number constants
3244 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3245 are stored in the target machine. @emph{Integers} are numbers that
3246 would fit into an @code{int} in the C language. @emph{Bignums} are
3247 integers, but they are stored in more than 32 bits. @emph{Flonums}
3248 are floating point numbers, described below.
3251 * Integers:: Integers
3259 @subsubsection Integers
3261 @cindex constants, integer
3263 @cindex binary integers
3264 @cindex integers, binary
3265 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3266 the binary digits @samp{01}.
3268 @cindex octal integers
3269 @cindex integers, octal
3270 An octal integer is @samp{0} followed by zero or more of the octal
3271 digits (@samp{01234567}).
3273 @cindex decimal integers
3274 @cindex integers, decimal
3275 A decimal integer starts with a non-zero digit followed by zero or
3276 more digits (@samp{0123456789}).
3278 @cindex hexadecimal integers
3279 @cindex integers, hexadecimal
3280 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3281 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3283 Integers have the usual values. To denote a negative integer, use
3284 the prefix operator @samp{-} discussed under expressions
3285 (@pxref{Prefix Ops,,Prefix Operators}).
3288 @subsubsection Bignums
3291 @cindex constants, bignum
3292 A @dfn{bignum} has the same syntax and semantics as an integer
3293 except that the number (or its negative) takes more than 32 bits to
3294 represent in binary. The distinction is made because in some places
3295 integers are permitted while bignums are not.
3298 @subsubsection Flonums
3300 @cindex floating point numbers
3301 @cindex constants, floating point
3303 @cindex precision, floating point
3304 A @dfn{flonum} represents a floating point number. The translation is
3305 indirect: a decimal floating point number from the text is converted by
3306 @command{@value{AS}} to a generic binary floating point number of more than
3307 sufficient precision. This generic floating point number is converted
3308 to a particular computer's floating point format (or formats) by a
3309 portion of @command{@value{AS}} specialized to that computer.
3311 A flonum is written by writing (in order)
3316 (@samp{0} is optional on the HPPA.)
3320 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3322 @kbd{e} is recommended. Case is not important.
3324 @c FIXME: verify if flonum syntax really this vague for most cases
3325 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3326 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3329 On the H8/300 and Renesas / SuperH SH architectures, the letter must be
3330 one of the letters @samp{DFPRSX} (in upper or lower case).
3332 On the ARC, the letter must be one of the letters @samp{DFRS}
3333 (in upper or lower case).
3335 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3339 One of the letters @samp{DFRS} (in upper or lower case).
3342 One of the letters @samp{DFPRSX} (in upper or lower case).
3345 The letter @samp{E} (upper case only).
3350 An optional sign: either @samp{+} or @samp{-}.
3353 An optional @dfn{integer part}: zero or more decimal digits.
3356 An optional @dfn{fractional part}: @samp{.} followed by zero
3357 or more decimal digits.
3360 An optional exponent, consisting of:
3364 An @samp{E} or @samp{e}.
3365 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3366 @c principle this can perfectly well be different on different targets.
3368 Optional sign: either @samp{+} or @samp{-}.
3370 One or more decimal digits.
3375 At least one of the integer part or the fractional part must be
3376 present. The floating point number has the usual base-10 value.
3378 @command{@value{AS}} does all processing using integers. Flonums are computed
3379 independently of any floating point hardware in the computer running
3380 @command{@value{AS}}.
3383 @chapter Sections and Relocation
3388 * Secs Background:: Background
3389 * Ld Sections:: Linker Sections
3390 * As Sections:: Assembler Internal Sections
3391 * Sub-Sections:: Sub-Sections
3395 @node Secs Background
3398 Roughly, a section is a range of addresses, with no gaps; all data
3399 ``in'' those addresses is treated the same for some particular purpose.
3400 For example there may be a ``read only'' section.
3402 @cindex linker, and assembler
3403 @cindex assembler, and linker
3404 The linker @code{@value{LD}} reads many object files (partial programs) and
3405 combines their contents to form a runnable program. When @command{@value{AS}}
3406 emits an object file, the partial program is assumed to start at address 0.
3407 @code{@value{LD}} assigns the final addresses for the partial program, so that
3408 different partial programs do not overlap. This is actually an
3409 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3412 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3413 addresses. These blocks slide to their run-time addresses as rigid
3414 units; their length does not change and neither does the order of bytes
3415 within them. Such a rigid unit is called a @emph{section}. Assigning
3416 run-time addresses to sections is called @dfn{relocation}. It includes
3417 the task of adjusting mentions of object-file addresses so they refer to
3418 the proper run-time addresses.
3420 For the H8/300, and for the Renesas / SuperH SH,
3421 @command{@value{AS}} pads sections if needed to
3422 ensure they end on a word (sixteen bit) boundary.
3425 @cindex standard assembler sections
3426 An object file written by @command{@value{AS}} has at least three sections, any
3427 of which may be empty. These are named @dfn{text}, @dfn{data} and
3432 When it generates COFF or ELF output,
3434 @command{@value{AS}} can also generate whatever other named sections you specify
3435 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3436 If you do not use any directives that place output in the @samp{.text}
3437 or @samp{.data} sections, these sections still exist, but are empty.
3442 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3444 @command{@value{AS}} can also generate whatever other named sections you
3445 specify using the @samp{.space} and @samp{.subspace} directives. See
3446 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3447 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3448 assembler directives.
3451 Additionally, @command{@value{AS}} uses different names for the standard
3452 text, data, and bss sections when generating SOM output. Program text
3453 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3454 BSS into @samp{$BSS$}.
3458 Within the object file, the text section starts at address @code{0}, the
3459 data section follows, and the bss section follows the data section.
3462 When generating either SOM or ELF output files on the HPPA, the text
3463 section starts at address @code{0}, the data section at address
3464 @code{0x4000000}, and the bss section follows the data section.
3467 To let @code{@value{LD}} know which data changes when the sections are
3468 relocated, and how to change that data, @command{@value{AS}} also writes to the
3469 object file details of the relocation needed. To perform relocation
3470 @code{@value{LD}} must know, each time an address in the object
3474 Where in the object file is the beginning of this reference to
3477 How long (in bytes) is this reference?
3479 Which section does the address refer to? What is the numeric value of
3481 (@var{address}) @minus{} (@var{start-address of section})?
3484 Is the reference to an address ``Program-Counter relative''?
3487 @cindex addresses, format of
3488 @cindex section-relative addressing
3489 In fact, every address @command{@value{AS}} ever uses is expressed as
3491 (@var{section}) + (@var{offset into section})
3494 Further, most expressions @command{@value{AS}} computes have this section-relative
3497 (For some object formats, such as SOM for the HPPA, some expressions are
3498 symbol-relative instead.)
3501 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3502 @var{N} into section @var{secname}.''
3504 Apart from text, data and bss sections you need to know about the
3505 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3506 addresses in the absolute section remain unchanged. For example, address
3507 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3508 @code{@value{LD}}. Although the linker never arranges two partial programs'
3509 data sections with overlapping addresses after linking, @emph{by definition}
3510 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3511 part of a program is always the same address when the program is running as
3512 address @code{@{absolute@ 239@}} in any other part of the program.
3514 The idea of sections is extended to the @dfn{undefined} section. Any
3515 address whose section is unknown at assembly time is by definition
3516 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3517 Since numbers are always defined, the only way to generate an undefined
3518 address is to mention an undefined symbol. A reference to a named
3519 common block would be such a symbol: its value is unknown at assembly
3520 time so it has section @emph{undefined}.
3522 By analogy the word @emph{section} is used to describe groups of sections in
3523 the linked program. @code{@value{LD}} puts all partial programs' text
3524 sections in contiguous addresses in the linked program. It is
3525 customary to refer to the @emph{text section} of a program, meaning all
3526 the addresses of all partial programs' text sections. Likewise for
3527 data and bss sections.
3529 Some sections are manipulated by @code{@value{LD}}; others are invented for
3530 use of @command{@value{AS}} and have no meaning except during assembly.
3533 @section Linker Sections
3534 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3539 @cindex named sections
3540 @cindex sections, named
3541 @item named sections
3544 @cindex text section
3545 @cindex data section
3549 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3550 separate but equal sections. Anything you can say of one section is
3553 When the program is running, however, it is
3554 customary for the text section to be unalterable. The
3555 text section is often shared among processes: it contains
3556 instructions, constants and the like. The data section of a running
3557 program is usually alterable: for example, C variables would be stored
3558 in the data section.
3563 This section contains zeroed bytes when your program begins running. It
3564 is used to hold uninitialized variables or common storage. The length of
3565 each partial program's bss section is important, but because it starts
3566 out containing zeroed bytes there is no need to store explicit zero
3567 bytes in the object file. The bss section was invented to eliminate
3568 those explicit zeros from object files.
3570 @cindex absolute section
3571 @item absolute section
3572 Address 0 of this section is always ``relocated'' to runtime address 0.
3573 This is useful if you want to refer to an address that @code{@value{LD}} must
3574 not change when relocating. In this sense we speak of absolute
3575 addresses being ``unrelocatable'': they do not change during relocation.
3577 @cindex undefined section
3578 @item undefined section
3579 This ``section'' is a catch-all for address references to objects not in
3580 the preceding sections.
3581 @c FIXME: ref to some other doc on obj-file formats could go here.
3584 @cindex relocation example
3585 An idealized example of three relocatable sections follows.
3587 The example uses the traditional section names @samp{.text} and @samp{.data}.
3589 Memory addresses are on the horizontal axis.
3593 @c END TEXI2ROFF-KILL
3596 partial program # 1: |ttttt|dddd|00|
3603 partial program # 2: |TTT|DDD|000|
3606 +--+---+-----+--+----+---+-----+~~
3607 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3608 +--+---+-----+--+----+---+-----+~~
3610 addresses: 0 @dots{}
3617 \line{\it Partial program \#1: \hfil}
3618 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3619 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3621 \line{\it Partial program \#2: \hfil}
3622 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3623 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3625 \line{\it linked program: \hfil}
3626 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3627 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3628 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3629 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3631 \line{\it addresses: \hfil}
3635 @c END TEXI2ROFF-KILL
3638 @section Assembler Internal Sections
3640 @cindex internal assembler sections
3641 @cindex sections in messages, internal
3642 These sections are meant only for the internal use of @command{@value{AS}}. They
3643 have no meaning at run-time. You do not really need to know about these
3644 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3645 warning messages, so it might be helpful to have an idea of their
3646 meanings to @command{@value{AS}}. These sections are used to permit the
3647 value of every expression in your assembly language program to be a
3648 section-relative address.
3651 @cindex assembler internal logic error
3652 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3653 An internal assembler logic error has been found. This means there is a
3654 bug in the assembler.
3656 @cindex expr (internal section)
3658 The assembler stores complex expression internally as combinations of
3659 symbols. When it needs to represent an expression as a symbol, it puts
3660 it in the expr section.
3662 @c FIXME item transfer[t] vector preload
3663 @c FIXME item transfer[t] vector postload
3664 @c FIXME item register
3668 @section Sub-Sections
3670 @cindex numbered subsections
3671 @cindex grouping data
3677 fall into two sections: text and data.
3679 You may have separate groups of
3681 data in named sections
3685 data in named sections
3691 that you want to end up near to each other in the object file, even though they
3692 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3693 use @dfn{subsections} for this purpose. Within each section, there can be
3694 numbered subsections with values from 0 to 8192. Objects assembled into the
3695 same subsection go into the object file together with other objects in the same
3696 subsection. For example, a compiler might want to store constants in the text
3697 section, but might not want to have them interspersed with the program being
3698 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3699 section of code being output, and a @samp{.text 1} before each group of
3700 constants being output.
3702 Subsections are optional. If you do not use subsections, everything
3703 goes in subsection number zero.
3706 Each subsection is zero-padded up to a multiple of four bytes.
3707 (Subsections may be padded a different amount on different flavors
3708 of @command{@value{AS}}.)
3712 On the H8/300 platform, each subsection is zero-padded to a word
3713 boundary (two bytes).
3714 The same is true on the Renesas SH.
3718 Subsections appear in your object file in numeric order, lowest numbered
3719 to highest. (All this to be compatible with other people's assemblers.)
3720 The object file contains no representation of subsections; @code{@value{LD}} and
3721 other programs that manipulate object files see no trace of them.
3722 They just see all your text subsections as a text section, and all your
3723 data subsections as a data section.
3725 To specify which subsection you want subsequent statements assembled
3726 into, use a numeric argument to specify it, in a @samp{.text
3727 @var{expression}} or a @samp{.data @var{expression}} statement.
3730 When generating COFF output, you
3735 can also use an extra subsection
3736 argument with arbitrary named sections: @samp{.section @var{name},
3741 When generating ELF output, you
3746 can also use the @code{.subsection} directive (@pxref{SubSection})
3747 to specify a subsection: @samp{.subsection @var{expression}}.
3749 @var{Expression} should be an absolute expression
3750 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3751 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3752 begins in @code{text 0}. For instance:
3754 .text 0 # The default subsection is text 0 anyway.
3755 .ascii "This lives in the first text subsection. *"
3757 .ascii "But this lives in the second text subsection."
3759 .ascii "This lives in the data section,"
3760 .ascii "in the first data subsection."
3762 .ascii "This lives in the first text section,"
3763 .ascii "immediately following the asterisk (*)."
3766 Each section has a @dfn{location counter} incremented by one for every byte
3767 assembled into that section. Because subsections are merely a convenience
3768 restricted to @command{@value{AS}} there is no concept of a subsection location
3769 counter. There is no way to directly manipulate a location counter---but the
3770 @code{.align} directive changes it, and any label definition captures its
3771 current value. The location counter of the section where statements are being
3772 assembled is said to be the @dfn{active} location counter.
3775 @section bss Section
3778 @cindex common variable storage
3779 The bss section is used for local common variable storage.
3780 You may allocate address space in the bss section, but you may
3781 not dictate data to load into it before your program executes. When
3782 your program starts running, all the contents of the bss
3783 section are zeroed bytes.
3785 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3786 @ref{Lcomm,,@code{.lcomm}}.
3788 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3789 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3792 When assembling for a target which supports multiple sections, such as ELF or
3793 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3794 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3795 section. Typically the section will only contain symbol definitions and
3796 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3803 Symbols are a central concept: the programmer uses symbols to name
3804 things, the linker uses symbols to link, and the debugger uses symbols
3808 @cindex debuggers, and symbol order
3809 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3810 the same order they were declared. This may break some debuggers.
3815 * Setting Symbols:: Giving Symbols Other Values
3816 * Symbol Names:: Symbol Names
3817 * Dot:: The Special Dot Symbol
3818 * Symbol Attributes:: Symbol Attributes
3825 A @dfn{label} is written as a symbol immediately followed by a colon
3826 @samp{:}. The symbol then represents the current value of the
3827 active location counter, and is, for example, a suitable instruction
3828 operand. You are warned if you use the same symbol to represent two
3829 different locations: the first definition overrides any other
3833 On the HPPA, the usual form for a label need not be immediately followed by a
3834 colon, but instead must start in column zero. Only one label may be defined on
3835 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3836 provides a special directive @code{.label} for defining labels more flexibly.
3839 @node Setting Symbols
3840 @section Giving Symbols Other Values
3842 @cindex assigning values to symbols
3843 @cindex symbol values, assigning
3844 A symbol can be given an arbitrary value by writing a symbol, followed
3845 by an equals sign @samp{=}, followed by an expression
3846 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3847 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3848 equals sign @samp{=}@samp{=} here represents an equivalent of the
3849 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3852 Blackfin does not support symbol assignment with @samp{=}.
3856 @section Symbol Names
3858 @cindex symbol names
3859 @cindex names, symbol
3860 @ifclear SPECIAL-SYMS
3861 Symbol names begin with a letter or with one of @samp{._}. On most
3862 machines, you can also use @code{$} in symbol names; exceptions are
3863 noted in @ref{Machine Dependencies}. That character may be followed by any
3864 string of digits, letters, dollar signs (unless otherwise noted for a
3865 particular target machine), and underscores.
3869 Symbol names begin with a letter or with one of @samp{._}. On the
3870 Renesas SH you can also use @code{$} in symbol names. That
3871 character may be followed by any string of digits, letters, dollar signs (save
3872 on the H8/300), and underscores.
3876 Case of letters is significant: @code{foo} is a different symbol name
3879 Symbol names do not start with a digit. An exception to this rule is made for
3880 Local Labels. See below.
3882 Multibyte characters are supported. To generate a symbol name containing
3883 multibyte characters enclose it within double quotes and use escape codes. cf
3884 @xref{Strings}. Generating a multibyte symbol name from a label is not
3885 currently supported.
3887 Each symbol has exactly one name. Each name in an assembly language program
3888 refers to exactly one symbol. You may use that symbol name any number of times
3891 @subheading Local Symbol Names
3893 @cindex local symbol names
3894 @cindex symbol names, local
3895 A local symbol is any symbol beginning with certain local label prefixes.
3896 By default, the local label prefix is @samp{.L} for ELF systems or
3897 @samp{L} for traditional a.out systems, but each target may have its own
3898 set of local label prefixes.
3900 On the HPPA local symbols begin with @samp{L$}.
3903 Local symbols are defined and used within the assembler, but they are
3904 normally not saved in object files. Thus, they are not visible when debugging.
3905 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3906 to retain the local symbols in the object files.
3908 @subheading Local Labels
3910 @cindex local labels
3911 @cindex temporary symbol names
3912 @cindex symbol names, temporary
3913 Local labels are different from local symbols. Local labels help compilers and
3914 programmers use names temporarily. They create symbols which are guaranteed to
3915 be unique over the entire scope of the input source code and which can be
3916 referred to by a simple notation. To define a local label, write a label of
3917 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3918 To refer to the most recent previous definition of that label write
3919 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3920 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3921 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3923 There is no restriction on how you can use these labels, and you can reuse them
3924 too. So that it is possible to repeatedly define the same local label (using
3925 the same number @samp{@b{N}}), although you can only refer to the most recently
3926 defined local label of that number (for a backwards reference) or the next
3927 definition of a specific local label for a forward reference. It is also worth
3928 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3929 implemented in a slightly more efficient manner than the others.
3940 Which is the equivalent of:
3943 label_1: branch label_3
3944 label_2: branch label_1
3945 label_3: branch label_4
3946 label_4: branch label_3
3949 Local label names are only a notational device. They are immediately
3950 transformed into more conventional symbol names before the assembler uses them.
3951 The symbol names are stored in the symbol table, appear in error messages, and
3952 are optionally emitted to the object file. The names are constructed using
3956 @item @emph{local label prefix}
3957 All local symbols begin with the system-specific local label prefix.
3958 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3959 that start with the local label prefix. These labels are
3960 used for symbols you are never intended to see. If you use the
3961 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3962 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3963 you may use them in debugging.
3966 This is the number that was used in the local label definition. So if the
3967 label is written @samp{55:} then the number is @samp{55}.
3970 This unusual character is included so you do not accidentally invent a symbol
3971 of the same name. The character has ASCII value of @samp{\002} (control-B).
3973 @item @emph{ordinal number}
3974 This is a serial number to keep the labels distinct. The first definition of
3975 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3976 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3977 the number @samp{1} and its 15th definition gets @samp{15} as well.
3980 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3981 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3983 @subheading Dollar Local Labels
3984 @cindex dollar local symbols
3986 On some targets @code{@value{AS}} also supports an even more local form of
3987 local labels called dollar labels. These labels go out of scope (i.e., they
3988 become undefined) as soon as a non-local label is defined. Thus they remain
3989 valid for only a small region of the input source code. Normal local labels,
3990 by contrast, remain in scope for the entire file, or until they are redefined
3991 by another occurrence of the same local label.
3993 Dollar labels are defined in exactly the same way as ordinary local labels,
3994 except that they have a dollar sign suffix to their numeric value, e.g.,
3997 They can also be distinguished from ordinary local labels by their transformed
3998 names which use ASCII character @samp{\001} (control-A) as the magic character
3999 to distinguish them from ordinary labels. For example, the fifth definition of
4000 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
4003 @section The Special Dot Symbol
4005 @cindex dot (symbol)
4006 @cindex @code{.} (symbol)
4007 @cindex current address
4008 @cindex location counter
4009 The special symbol @samp{.} refers to the current address that
4010 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
4011 .long .} defines @code{melvin} to contain its own address.
4012 Assigning a value to @code{.} is treated the same as a @code{.org}
4014 @ifclear no-space-dir
4015 Thus, the expression @samp{.=.+4} is the same as saying
4019 @node Symbol Attributes
4020 @section Symbol Attributes
4022 @cindex symbol attributes
4023 @cindex attributes, symbol
4024 Every symbol has, as well as its name, the attributes ``Value'' and
4025 ``Type''. Depending on output format, symbols can also have auxiliary
4028 The detailed definitions are in @file{a.out.h}.
4031 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
4032 all these attributes, and probably won't warn you. This makes the
4033 symbol an externally defined symbol, which is generally what you
4037 * Symbol Value:: Value
4038 * Symbol Type:: Type
4040 * a.out Symbols:: Symbol Attributes: @code{a.out}
4043 * COFF Symbols:: Symbol Attributes for COFF
4046 * SOM Symbols:: Symbol Attributes for SOM
4053 @cindex value of a symbol
4054 @cindex symbol value
4055 The value of a symbol is (usually) 32 bits. For a symbol which labels a
4056 location in the text, data, bss or absolute sections the value is the
4057 number of addresses from the start of that section to the label.
4058 Naturally for text, data and bss sections the value of a symbol changes
4059 as @code{@value{LD}} changes section base addresses during linking. Absolute
4060 symbols' values do not change during linking: that is why they are
4063 The value of an undefined symbol is treated in a special way. If it is
4064 0 then the symbol is not defined in this assembler source file, and
4065 @code{@value{LD}} tries to determine its value from other files linked into the
4066 same program. You make this kind of symbol simply by mentioning a symbol
4067 name without defining it. A non-zero value represents a @code{.comm}
4068 common declaration. The value is how much common storage to reserve, in
4069 bytes (addresses). The symbol refers to the first address of the
4075 @cindex type of a symbol
4077 The type attribute of a symbol contains relocation (section)
4078 information, any flag settings indicating that a symbol is external, and
4079 (optionally), other information for linkers and debuggers. The exact
4080 format depends on the object-code output format in use.
4084 @subsection Symbol Attributes: @code{a.out}
4086 @cindex @code{a.out} symbol attributes
4087 @cindex symbol attributes, @code{a.out}
4090 * Symbol Desc:: Descriptor
4091 * Symbol Other:: Other
4095 @subsubsection Descriptor
4097 @cindex descriptor, of @code{a.out} symbol
4098 This is an arbitrary 16-bit value. You may establish a symbol's
4099 descriptor value by using a @code{.desc} statement
4100 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
4101 @command{@value{AS}}.
4104 @subsubsection Other
4106 @cindex other attribute, of @code{a.out} symbol
4107 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
4112 @subsection Symbol Attributes for COFF
4114 @cindex COFF symbol attributes
4115 @cindex symbol attributes, COFF
4117 The COFF format supports a multitude of auxiliary symbol attributes;
4118 like the primary symbol attributes, they are set between @code{.def} and
4119 @code{.endef} directives.
4121 @subsubsection Primary Attributes
4123 @cindex primary attributes, COFF symbols
4124 The symbol name is set with @code{.def}; the value and type,
4125 respectively, with @code{.val} and @code{.type}.
4127 @subsubsection Auxiliary Attributes
4129 @cindex auxiliary attributes, COFF symbols
4130 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
4131 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
4132 table information for COFF.
4137 @subsection Symbol Attributes for SOM
4139 @cindex SOM symbol attributes
4140 @cindex symbol attributes, SOM
4142 The SOM format for the HPPA supports a multitude of symbol attributes set with
4143 the @code{.EXPORT} and @code{.IMPORT} directives.
4145 The attributes are described in @cite{HP9000 Series 800 Assembly
4146 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
4147 @code{EXPORT} assembler directive documentation.
4151 @chapter Expressions
4155 @cindex numeric values
4156 An @dfn{expression} specifies an address or numeric value.
4157 Whitespace may precede and/or follow an expression.
4159 The result of an expression must be an absolute number, or else an offset into
4160 a particular section. If an expression is not absolute, and there is not
4161 enough information when @command{@value{AS}} sees the expression to know its
4162 section, a second pass over the source program might be necessary to interpret
4163 the expression---but the second pass is currently not implemented.
4164 @command{@value{AS}} aborts with an error message in this situation.
4167 * Empty Exprs:: Empty Expressions
4168 * Integer Exprs:: Integer Expressions
4172 @section Empty Expressions
4174 @cindex empty expressions
4175 @cindex expressions, empty
4176 An empty expression has no value: it is just whitespace or null.
4177 Wherever an absolute expression is required, you may omit the
4178 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4179 is compatible with other assemblers.
4182 @section Integer Expressions
4184 @cindex integer expressions
4185 @cindex expressions, integer
4186 An @dfn{integer expression} is one or more @emph{arguments} delimited
4187 by @emph{operators}.
4190 * Arguments:: Arguments
4191 * Operators:: Operators
4192 * Prefix Ops:: Prefix Operators
4193 * Infix Ops:: Infix Operators
4197 @subsection Arguments
4199 @cindex expression arguments
4200 @cindex arguments in expressions
4201 @cindex operands in expressions
4202 @cindex arithmetic operands
4203 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4204 contexts arguments are sometimes called ``arithmetic operands''. In
4205 this manual, to avoid confusing them with the ``instruction operands'' of
4206 the machine language, we use the term ``argument'' to refer to parts of
4207 expressions only, reserving the word ``operand'' to refer only to machine
4208 instruction operands.
4210 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4211 @var{section} is one of text, data, bss, absolute,
4212 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4215 Numbers are usually integers.
4217 A number can be a flonum or bignum. In this case, you are warned
4218 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4219 these 32 bits are an integer. You may write integer-manipulating
4220 instructions that act on exotic constants, compatible with other
4223 @cindex subexpressions
4224 Subexpressions are a left parenthesis @samp{(} followed by an integer
4225 expression, followed by a right parenthesis @samp{)}; or a prefix
4226 operator followed by an argument.
4229 @subsection Operators
4231 @cindex operators, in expressions
4232 @cindex arithmetic functions
4233 @cindex functions, in expressions
4234 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4235 operators are followed by an argument. Infix operators appear
4236 between their arguments. Operators may be preceded and/or followed by
4240 @subsection Prefix Operator
4242 @cindex prefix operators
4243 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4244 one argument, which must be absolute.
4246 @c the tex/end tex stuff surrounding this small table is meant to make
4247 @c it align, on the printed page, with the similar table in the next
4248 @c section (which is inside an enumerate).
4250 \global\advance\leftskip by \itemindent
4255 @dfn{Negation}. Two's complement negation.
4257 @dfn{Complementation}. Bitwise not.
4261 \global\advance\leftskip by -\itemindent
4265 @subsection Infix Operators
4267 @cindex infix operators
4268 @cindex operators, permitted arguments
4269 @dfn{Infix operators} take two arguments, one on either side. Operators
4270 have precedence, but operations with equal precedence are performed left
4271 to right. Apart from @code{+} or @option{-}, both arguments must be
4272 absolute, and the result is absolute.
4275 @cindex operator precedence
4276 @cindex precedence of operators
4283 @dfn{Multiplication}.
4286 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4292 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4295 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4299 Intermediate precedence
4304 @dfn{Bitwise Inclusive Or}.
4310 @dfn{Bitwise Exclusive Or}.
4313 @dfn{Bitwise Or Not}.
4320 @cindex addition, permitted arguments
4321 @cindex plus, permitted arguments
4322 @cindex arguments for addition
4324 @dfn{Addition}. If either argument is absolute, the result has the section of
4325 the other argument. You may not add together arguments from different
4328 @cindex subtraction, permitted arguments
4329 @cindex minus, permitted arguments
4330 @cindex arguments for subtraction
4332 @dfn{Subtraction}. If the right argument is absolute, the
4333 result has the section of the left argument.
4334 If both arguments are in the same section, the result is absolute.
4335 You may not subtract arguments from different sections.
4336 @c FIXME is there still something useful to say about undefined - undefined ?
4338 @cindex comparison expressions
4339 @cindex expressions, comparison
4344 @dfn{Is Not Equal To}
4348 @dfn{Is Greater Than}
4350 @dfn{Is Greater Than Or Equal To}
4352 @dfn{Is Less Than Or Equal To}
4354 The comparison operators can be used as infix operators. A true results has a
4355 value of -1 whereas a false result has a value of 0. Note, these operators
4356 perform signed comparisons.
4359 @item Lowest Precedence
4368 These two logical operations can be used to combine the results of sub
4369 expressions. Note, unlike the comparison operators a true result returns a
4370 value of 1 but a false results does still return 0. Also note that the logical
4371 or operator has a slightly lower precedence than logical and.
4376 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4377 address; you can only have a defined section in one of the two arguments.
4380 @chapter Assembler Directives
4382 @cindex directives, machine independent
4383 @cindex pseudo-ops, machine independent
4384 @cindex machine independent directives
4385 All assembler directives have names that begin with a period (@samp{.}).
4386 The names are case insensitive for most targets, and usually written
4389 This chapter discusses directives that are available regardless of the
4390 target machine configuration for the @sc{gnu} assembler.
4392 Some machine configurations provide additional directives.
4393 @xref{Machine Dependencies}.
4396 @ifset machine-directives
4397 @xref{Machine Dependencies}, for additional directives.
4402 * Abort:: @code{.abort}
4404 * ABORT (COFF):: @code{.ABORT}
4407 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4408 * Altmacro:: @code{.altmacro}
4409 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4410 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4411 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4412 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4413 * Byte:: @code{.byte @var{expressions}}
4414 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4415 * Comm:: @code{.comm @var{symbol} , @var{length} }
4416 * Data:: @code{.data @var{subsection}}
4417 * Dc:: @code{.dc[@var{size}] @var{expressions}}
4418 * Dcb:: @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4419 * Ds:: @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4421 * Def:: @code{.def @var{name}}
4424 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4430 * Double:: @code{.double @var{flonums}}
4431 * Eject:: @code{.eject}
4432 * Else:: @code{.else}
4433 * Elseif:: @code{.elseif}
4436 * Endef:: @code{.endef}
4439 * Endfunc:: @code{.endfunc}
4440 * Endif:: @code{.endif}
4441 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4442 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4443 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4445 * Error:: @code{.error @var{string}}
4446 * Exitm:: @code{.exitm}
4447 * Extern:: @code{.extern}
4448 * Fail:: @code{.fail}
4449 * File:: @code{.file}
4450 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4451 * Float:: @code{.float @var{flonums}}
4452 * Func:: @code{.func}
4453 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4455 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4456 * Hidden:: @code{.hidden @var{names}}
4459 * hword:: @code{.hword @var{expressions}}
4460 * Ident:: @code{.ident}
4461 * If:: @code{.if @var{absolute expression}}
4462 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4463 * Include:: @code{.include "@var{file}"}
4464 * Int:: @code{.int @var{expressions}}
4466 * Internal:: @code{.internal @var{names}}
4469 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4470 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4471 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4472 * Lflags:: @code{.lflags}
4473 @ifclear no-line-dir
4474 * Line:: @code{.line @var{line-number}}
4477 * Linkonce:: @code{.linkonce [@var{type}]}
4478 * List:: @code{.list}
4479 * Ln:: @code{.ln @var{line-number}}
4480 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4481 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4483 * Local:: @code{.local @var{names}}
4486 * Long:: @code{.long @var{expressions}}
4488 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4491 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4492 * MRI:: @code{.mri @var{val}}
4493 * Noaltmacro:: @code{.noaltmacro}
4494 * Nolist:: @code{.nolist}
4495 * Nops:: @code{.nops @var{size}[, @var{control}]}
4496 * Octa:: @code{.octa @var{bignums}}
4497 * Offset:: @code{.offset @var{loc}}
4498 * Org:: @code{.org @var{new-lc}, @var{fill}}
4499 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4501 * PopSection:: @code{.popsection}
4502 * Previous:: @code{.previous}
4505 * Print:: @code{.print @var{string}}
4507 * Protected:: @code{.protected @var{names}}
4510 * Psize:: @code{.psize @var{lines}, @var{columns}}
4511 * Purgem:: @code{.purgem @var{name}}
4513 * PushSection:: @code{.pushsection @var{name}}
4516 * Quad:: @code{.quad @var{bignums}}
4517 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4518 * Rept:: @code{.rept @var{count}}
4519 * Sbttl:: @code{.sbttl "@var{subheading}"}
4521 * Scl:: @code{.scl @var{class}}
4524 * Section:: @code{.section @var{name}[, @var{flags}]}
4527 * Set:: @code{.set @var{symbol}, @var{expression}}
4528 * Short:: @code{.short @var{expressions}}
4529 * Single:: @code{.single @var{flonums}}
4531 * Size:: @code{.size [@var{name} , @var{expression}]}
4533 @ifclear no-space-dir
4534 * Skip:: @code{.skip @var{size} [,@var{fill}]}
4537 * Sleb128:: @code{.sleb128 @var{expressions}}
4538 @ifclear no-space-dir
4539 * Space:: @code{.space @var{size} [,@var{fill}]}
4542 * Stab:: @code{.stabd, .stabn, .stabs}
4545 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4546 * Struct:: @code{.struct @var{expression}}
4548 * SubSection:: @code{.subsection}
4549 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4553 * Tag:: @code{.tag @var{structname}}
4556 * Text:: @code{.text @var{subsection}}
4557 * Title:: @code{.title "@var{heading}"}
4559 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4562 * Uleb128:: @code{.uleb128 @var{expressions}}
4564 * Val:: @code{.val @var{addr}}
4568 * Version:: @code{.version "@var{string}"}
4569 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4570 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4573 * Warning:: @code{.warning @var{string}}
4574 * Weak:: @code{.weak @var{names}}
4575 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4576 * Word:: @code{.word @var{expressions}}
4577 @ifclear no-space-dir
4578 * Zero:: @code{.zero @var{size}}
4581 * 2byte:: @code{.2byte @var{expressions}}
4582 * 4byte:: @code{.4byte @var{expressions}}
4583 * 8byte:: @code{.8byte @var{bignums}}
4585 * Deprecated:: Deprecated Directives
4589 @section @code{.abort}
4591 @cindex @code{abort} directive
4592 @cindex stopping the assembly
4593 This directive stops the assembly immediately. It is for
4594 compatibility with other assemblers. The original idea was that the
4595 assembly language source would be piped into the assembler. If the sender
4596 of the source quit, it could use this directive tells @command{@value{AS}} to
4597 quit also. One day @code{.abort} will not be supported.
4601 @section @code{.ABORT} (COFF)
4603 @cindex @code{ABORT} directive
4604 When producing COFF output, @command{@value{AS}} accepts this directive as a
4605 synonym for @samp{.abort}.
4610 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4612 @cindex padding the location counter
4613 @cindex @code{align} directive
4614 Pad the location counter (in the current subsection) to a particular storage
4615 boundary. The first expression (which must be absolute) is the alignment
4616 required, as described below.
4618 The second expression (also absolute) gives the fill value to be stored in the
4619 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4620 padding bytes are normally zero. However, on most systems, if the section is
4621 marked as containing code and the fill value is omitted, the space is filled
4622 with no-op instructions.
4624 The third expression is also absolute, and is also optional. If it is present,
4625 it is the maximum number of bytes that should be skipped by this alignment
4626 directive. If doing the alignment would require skipping more bytes than the
4627 specified maximum, then the alignment is not done at all. You can omit the
4628 fill value (the second argument) entirely by simply using two commas after the
4629 required alignment; this can be useful if you want the alignment to be filled
4630 with no-op instructions when appropriate.
4632 The way the required alignment is specified varies from system to system.
4633 For the arc, hppa, i386 using ELF, iq2000, m68k, or1k,
4634 s390, sparc, tic4x and xtensa, the first expression is the
4635 alignment request in bytes. For example @samp{.align 8} advances
4636 the location counter until it is a multiple of 8. If the location counter
4637 is already a multiple of 8, no change is needed. For the tic54x, the
4638 first expression is the alignment request in words.
4640 For other systems, including ppc, i386 using a.out format, arm and
4641 strongarm, it is the
4642 number of low-order zero bits the location counter must have after
4643 advancement. For example @samp{.align 3} advances the location
4644 counter until it is a multiple of 8. If the location counter is already a
4645 multiple of 8, no change is needed.
4647 This inconsistency is due to the different behaviors of the various
4648 native assemblers for these systems which GAS must emulate.
4649 GAS also provides @code{.balign} and @code{.p2align} directives,
4650 described later, which have a consistent behavior across all
4651 architectures (but are specific to GAS).
4654 @section @code{.altmacro}
4655 Enable alternate macro mode, enabling:
4658 @item LOCAL @var{name} [ , @dots{} ]
4659 One additional directive, @code{LOCAL}, is available. It is used to
4660 generate a string replacement for each of the @var{name} arguments, and
4661 replace any instances of @var{name} in each macro expansion. The
4662 replacement string is unique in the assembly, and different for each
4663 separate macro expansion. @code{LOCAL} allows you to write macros that
4664 define symbols, without fear of conflict between separate macro expansions.
4666 @item String delimiters
4667 You can write strings delimited in these other ways besides
4668 @code{"@var{string}"}:
4671 @item '@var{string}'
4672 You can delimit strings with single-quote characters.
4674 @item <@var{string}>
4675 You can delimit strings with matching angle brackets.
4678 @item single-character string escape
4679 To include any single character literally in a string (even if the
4680 character would otherwise have some special meaning), you can prefix the
4681 character with @samp{!} (an exclamation mark). For example, you can
4682 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4684 @item Expression results as strings
4685 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4686 and use the result as a string.
4690 @section @code{.ascii "@var{string}"}@dots{}
4692 @cindex @code{ascii} directive
4693 @cindex string literals
4694 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4695 separated by commas. It assembles each string (with no automatic
4696 trailing zero byte) into consecutive addresses.
4699 @section @code{.asciz "@var{string}"}@dots{}
4701 @cindex @code{asciz} directive
4702 @cindex zero-terminated strings
4703 @cindex null-terminated strings
4704 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4705 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4708 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4710 @cindex padding the location counter given number of bytes
4711 @cindex @code{balign} directive
4712 Pad the location counter (in the current subsection) to a particular
4713 storage boundary. The first expression (which must be absolute) is the
4714 alignment request in bytes. For example @samp{.balign 8} advances
4715 the location counter until it is a multiple of 8. If the location counter
4716 is already a multiple of 8, no change is needed.
4718 The second expression (also absolute) gives the fill value to be stored in the
4719 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4720 padding bytes are normally zero. However, on most systems, if the section is
4721 marked as containing code and the fill value is omitted, the space is filled
4722 with no-op instructions.
4724 The third expression is also absolute, and is also optional. If it is present,
4725 it is the maximum number of bytes that should be skipped by this alignment
4726 directive. If doing the alignment would require skipping more bytes than the
4727 specified maximum, then the alignment is not done at all. You can omit the
4728 fill value (the second argument) entirely by simply using two commas after the
4729 required alignment; this can be useful if you want the alignment to be filled
4730 with no-op instructions when appropriate.
4732 @cindex @code{balignw} directive
4733 @cindex @code{balignl} directive
4734 The @code{.balignw} and @code{.balignl} directives are variants of the
4735 @code{.balign} directive. The @code{.balignw} directive treats the fill
4736 pattern as a two byte word value. The @code{.balignl} directives treats the
4737 fill pattern as a four byte longword value. For example, @code{.balignw
4738 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4739 filled in with the value 0x368d (the exact placement of the bytes depends upon
4740 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4743 @node Bundle directives
4744 @section Bundle directives
4745 @subsection @code{.bundle_align_mode @var{abs-expr}}
4746 @cindex @code{bundle_align_mode} directive
4748 @cindex instruction bundle
4749 @cindex aligned instruction bundle
4750 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4751 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4752 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4753 disabled (which is the default state). If the argument it not zero, it
4754 gives the size of an instruction bundle as a power of two (as for the
4755 @code{.p2align} directive, @pxref{P2align}).
4757 For some targets, it's an ABI requirement that no instruction may span a
4758 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4759 instructions that starts on an aligned boundary. For example, if
4760 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4761 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4762 effect, no single instruction may span a boundary between bundles. If an
4763 instruction would start too close to the end of a bundle for the length of
4764 that particular instruction to fit within the bundle, then the space at the
4765 end of that bundle is filled with no-op instructions so the instruction
4766 starts in the next bundle. As a corollary, it's an error if any single
4767 instruction's encoding is longer than the bundle size.
4769 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4770 @cindex @code{bundle_lock} directive
4771 @cindex @code{bundle_unlock} directive
4772 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4773 allow explicit control over instruction bundle padding. These directives
4774 are only valid when @code{.bundle_align_mode} has been used to enable
4775 aligned instruction bundle mode. It's an error if they appear when
4776 @code{.bundle_align_mode} has not been used at all, or when the last
4777 directive was @w{@code{.bundle_align_mode 0}}.
4779 @cindex bundle-locked
4780 For some targets, it's an ABI requirement that certain instructions may
4781 appear only as part of specified permissible sequences of multiple
4782 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4783 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4784 instruction sequence. For purposes of aligned instruction bundle mode, a
4785 sequence starting with @code{.bundle_lock} and ending with
4786 @code{.bundle_unlock} is treated as a single instruction. That is, the
4787 entire sequence must fit into a single bundle and may not span a bundle
4788 boundary. If necessary, no-op instructions will be inserted before the
4789 first instruction of the sequence so that the whole sequence starts on an
4790 aligned bundle boundary. It's an error if the sequence is longer than the
4793 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4794 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4795 nested. That is, a second @code{.bundle_lock} directive before the next
4796 @code{.bundle_unlock} directive has no effect except that it must be
4797 matched by another closing @code{.bundle_unlock} so that there is the
4798 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4801 @section @code{.byte @var{expressions}}
4803 @cindex @code{byte} directive
4804 @cindex integers, one byte
4805 @code{.byte} expects zero or more expressions, separated by commas.
4806 Each expression is assembled into the next byte.
4808 @node CFI directives
4809 @section CFI directives
4810 @subsection @code{.cfi_sections @var{section_list}}
4811 @cindex @code{cfi_sections} directive
4812 @code{.cfi_sections} may be used to specify whether CFI directives
4813 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4814 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4815 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4816 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4817 directive is not used is @code{.cfi_sections .eh_frame}.
4819 On targets that support compact unwinding tables these can be generated
4820 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4822 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4823 which is used by the @value{TIC6X} target.
4825 The @code{.cfi_sections} directive can be repeated, with the same or different
4826 arguments, provided that CFI generation has not yet started. Once CFI
4827 generation has started however the section list is fixed and any attempts to
4828 redefine it will result in an error.
4830 @subsection @code{.cfi_startproc [simple]}
4831 @cindex @code{cfi_startproc} directive
4832 @code{.cfi_startproc} is used at the beginning of each function that
4833 should have an entry in @code{.eh_frame}. It initializes some internal
4834 data structures. Don't forget to close the function by
4835 @code{.cfi_endproc}.
4837 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4838 it also emits some architecture dependent initial CFI instructions.
4840 @subsection @code{.cfi_endproc}
4841 @cindex @code{cfi_endproc} directive
4842 @code{.cfi_endproc} is used at the end of a function where it closes its
4843 unwind entry previously opened by
4844 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4846 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4847 @cindex @code{cfi_personality} directive
4848 @code{.cfi_personality} defines personality routine and its encoding.
4849 @var{encoding} must be a constant determining how the personality
4850 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4851 argument is not present, otherwise second argument should be
4852 a constant or a symbol name. When using indirect encodings,
4853 the symbol provided should be the location where personality
4854 can be loaded from, not the personality routine itself.
4855 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4856 no personality routine.
4858 @subsection @code{.cfi_personality_id @var{id}}
4859 @cindex @code{cfi_personality_id} directive
4860 @code{cfi_personality_id} defines a personality routine by its index as
4861 defined in a compact unwinding format.
4862 Only valid when generating compact EH frames (i.e.
4863 with @code{.cfi_sections eh_frame_entry}.
4865 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4866 @cindex @code{cfi_fde_data} directive
4867 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4868 used for the current function. These are emitted inline in the
4869 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4870 in the @code{.gnu.extab} section otherwise.
4871 Only valid when generating compact EH frames (i.e.
4872 with @code{.cfi_sections eh_frame_entry}.
4874 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4875 @code{.cfi_lsda} defines LSDA and its encoding.
4876 @var{encoding} must be a constant determining how the LSDA
4877 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4878 argument is not present, otherwise the second argument should be a constant
4879 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4880 meaning that no LSDA is present.
4882 @subsection @code{.cfi_inline_lsda} [@var{align}]
4883 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4884 switches to the corresponding @code{.gnu.extab} section.
4885 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4886 Only valid when generating compact EH frames (i.e.
4887 with @code{.cfi_sections eh_frame_entry}.
4889 The table header and unwinding opcodes will be generated at this point,
4890 so that they are immediately followed by the LSDA data. The symbol
4891 referenced by the @code{.cfi_lsda} directive should still be defined
4892 in case a fallback FDE based encoding is used. The LSDA data is terminated
4893 by a section directive.
4895 The optional @var{align} argument specifies the alignment required.
4896 The alignment is specified as a power of two, as with the
4897 @code{.p2align} directive.
4899 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4900 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4901 address from @var{register} and add @var{offset} to it}.
4903 @subsection @code{.cfi_def_cfa_register @var{register}}
4904 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4905 now on @var{register} will be used instead of the old one. Offset
4908 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4909 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4910 remains the same, but @var{offset} is new. Note that it is the
4911 absolute offset that will be added to a defined register to compute
4914 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4915 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4916 value that is added/subtracted from the previous offset.
4918 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4919 Previous value of @var{register} is saved at offset @var{offset} from
4922 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4923 Previous value of @var{register} is CFA + @var{offset}.
4925 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4926 Previous value of @var{register} is saved at offset @var{offset} from
4927 the current CFA register. This is transformed to @code{.cfi_offset}
4928 using the known displacement of the CFA register from the CFA.
4929 This is often easier to use, because the number will match the
4930 code it's annotating.
4932 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4933 Previous value of @var{register1} is saved in register @var{register2}.
4935 @subsection @code{.cfi_restore @var{register}}
4936 @code{.cfi_restore} says that the rule for @var{register} is now the
4937 same as it was at the beginning of the function, after all initial
4938 instruction added by @code{.cfi_startproc} were executed.
4940 @subsection @code{.cfi_undefined @var{register}}
4941 From now on the previous value of @var{register} can't be restored anymore.
4943 @subsection @code{.cfi_same_value @var{register}}
4944 Current value of @var{register} is the same like in the previous frame,
4945 i.e. no restoration needed.
4947 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4948 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4949 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4950 places them in the current row. This is useful for situations where you have
4951 multiple @code{.cfi_*} directives that need to be undone due to the control
4952 flow of the program. For example, we could have something like this (assuming
4953 the CFA is the value of @code{rbp}):
4963 .cfi_def_cfa %rsp, 8
4966 /* Do something else */
4969 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4970 to the instructions before @code{label}. This means we'd have to add multiple
4971 @code{.cfi} directives after @code{label} to recreate the original save
4972 locations of the registers, as well as setting the CFA back to the value of
4973 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4985 .cfi_def_cfa %rsp, 8
4989 /* Do something else */
4992 That way, the rules for the instructions after @code{label} will be the same
4993 as before the first @code{.cfi_restore} without having to use multiple
4994 @code{.cfi} directives.
4996 @subsection @code{.cfi_return_column @var{register}}
4997 Change return column @var{register}, i.e. the return address is either
4998 directly in @var{register} or can be accessed by rules for @var{register}.
5000 @subsection @code{.cfi_signal_frame}
5001 Mark current function as signal trampoline.
5003 @subsection @code{.cfi_window_save}
5004 SPARC register window has been saved.
5006 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
5007 Allows the user to add arbitrary bytes to the unwind info. One
5008 might use this to add OS-specific CFI opcodes, or generic CFI
5009 opcodes that GAS does not yet support.
5011 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
5012 The current value of @var{register} is @var{label}. The value of @var{label}
5013 will be encoded in the output file according to @var{encoding}; see the
5014 description of @code{.cfi_personality} for details on this encoding.
5016 The usefulness of equating a register to a fixed label is probably
5017 limited to the return address register. Here, it can be useful to
5018 mark a code segment that has only one return address which is reached
5019 by a direct branch and no copy of the return address exists in memory
5020 or another register.
5023 @section @code{.comm @var{symbol} , @var{length} }
5025 @cindex @code{comm} directive
5026 @cindex symbol, common
5027 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
5028 common symbol in one object file may be merged with a defined or common symbol
5029 of the same name in another object file. If @code{@value{LD}} does not see a
5030 definition for the symbol--just one or more common symbols--then it will
5031 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
5032 absolute expression. If @code{@value{LD}} sees multiple common symbols with
5033 the same name, and they do not all have the same size, it will allocate space
5034 using the largest size.
5037 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
5038 an optional third argument. This is the desired alignment of the symbol,
5039 specified for ELF as a byte boundary (for example, an alignment of 16 means
5040 that the least significant 4 bits of the address should be zero), and for PE
5041 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
5042 boundary). The alignment must be an absolute expression, and it must be a
5043 power of two. If @code{@value{LD}} allocates uninitialized memory for the
5044 common symbol, it will use the alignment when placing the symbol. If no
5045 alignment is specified, @command{@value{AS}} will set the alignment to the
5046 largest power of two less than or equal to the size of the symbol, up to a
5047 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
5048 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
5049 @samp{--section-alignment} option; image file sections in PE are aligned to
5050 multiples of 4096, which is far too large an alignment for ordinary variables.
5051 It is rather the default alignment for (non-debug) sections within object
5052 (@samp{*.o}) files, which are less strictly aligned.}.
5056 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
5057 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
5061 @section @code{.data @var{subsection}}
5062 @cindex @code{data} directive
5064 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
5065 end of the data subsection numbered @var{subsection} (which is an
5066 absolute expression). If @var{subsection} is omitted, it defaults
5070 @section @code{.dc[@var{size}] @var{expressions}}
5071 @cindex @code{dc} directive
5073 The @code{.dc} directive expects zero or more @var{expressions} separated by
5074 commas. These expressions are evaluated and their values inserted into the
5075 current section. The size of the emitted value depends upon the suffix to the
5076 @code{.dc} directive:
5080 Emits N-bit values, where N is the size of an address on the target system.
5084 Emits double precision floating-point values.
5086 Emits 32-bit values.
5088 Emits single precision floating-point values.
5090 Emits 16-bit values.
5091 Note - this is true even on targets where the @code{.word} directive would emit
5094 Emits long double precision floating-point values.
5097 If no suffix is used then @samp{.w} is assumed.
5099 The byte ordering is target dependent, as is the size and format of floating
5103 @section @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
5104 @cindex @code{dcb} directive
5105 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5106 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5107 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5108 @var{size} suffix, if present, must be one of:
5112 Emits single byte values.
5114 Emits double-precision floating point values.
5116 Emits 4-byte values.
5118 Emits single-precision floating point values.
5120 Emits 2-byte values.
5122 Emits long double-precision floating point values.
5125 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5127 The byte ordering is target dependent, as is the size and format of floating
5131 @section @code{.ds[@var{size}] @var{number} [,@var{fill}]}
5132 @cindex @code{ds} directive
5133 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5134 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5135 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5136 @var{size} suffix, if present, must be one of:
5140 Emits single byte values.
5142 Emits 8-byte values.
5144 Emits 4-byte values.
5146 Emits 12-byte values.
5148 Emits 4-byte values.
5150 Emits 2-byte values.
5152 Emits 12-byte values.
5155 Note - unlike the @code{.dcb} directive the @samp{.d}, @samp{.s} and @samp{.x}
5156 suffixes do not indicate that floating-point values are to be inserted.
5158 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5160 The byte ordering is target dependent.
5165 @section @code{.def @var{name}}
5167 @cindex @code{def} directive
5168 @cindex COFF symbols, debugging
5169 @cindex debugging COFF symbols
5170 Begin defining debugging information for a symbol @var{name}; the
5171 definition extends until the @code{.endef} directive is encountered.
5176 @section @code{.desc @var{symbol}, @var{abs-expression}}
5178 @cindex @code{desc} directive
5179 @cindex COFF symbol descriptor
5180 @cindex symbol descriptor, COFF
5181 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5182 to the low 16 bits of an absolute expression.
5185 The @samp{.desc} directive is not available when @command{@value{AS}} is
5186 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5187 object format. For the sake of compatibility, @command{@value{AS}} accepts
5188 it, but produces no output, when configured for COFF.
5194 @section @code{.dim}
5196 @cindex @code{dim} directive
5197 @cindex COFF auxiliary symbol information
5198 @cindex auxiliary symbol information, COFF
5199 This directive is generated by compilers to include auxiliary debugging
5200 information in the symbol table. It is only permitted inside
5201 @code{.def}/@code{.endef} pairs.
5205 @section @code{.double @var{flonums}}
5207 @cindex @code{double} directive
5208 @cindex floating point numbers (double)
5209 @code{.double} expects zero or more flonums, separated by commas. It
5210 assembles floating point numbers.
5212 The exact kind of floating point numbers emitted depends on how
5213 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5217 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5218 in @sc{ieee} format.
5223 @section @code{.eject}
5225 @cindex @code{eject} directive
5226 @cindex new page, in listings
5227 @cindex page, in listings
5228 @cindex listing control: new page
5229 Force a page break at this point, when generating assembly listings.
5232 @section @code{.else}
5234 @cindex @code{else} directive
5235 @code{.else} is part of the @command{@value{AS}} support for conditional
5236 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5237 of code to be assembled if the condition for the preceding @code{.if}
5241 @section @code{.elseif}
5243 @cindex @code{elseif} directive
5244 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5245 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5246 @code{.if} block that would otherwise fill the entire @code{.else} section.
5249 @section @code{.end}
5251 @cindex @code{end} directive
5252 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5253 process anything in the file past the @code{.end} directive.
5257 @section @code{.endef}
5259 @cindex @code{endef} directive
5260 This directive flags the end of a symbol definition begun with
5265 @section @code{.endfunc}
5266 @cindex @code{endfunc} directive
5267 @code{.endfunc} marks the end of a function specified with @code{.func}.
5270 @section @code{.endif}
5272 @cindex @code{endif} directive
5273 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5274 it marks the end of a block of code that is only assembled
5275 conditionally. @xref{If,,@code{.if}}.
5278 @section @code{.equ @var{symbol}, @var{expression}}
5280 @cindex @code{equ} directive
5281 @cindex assigning values to symbols
5282 @cindex symbols, assigning values to
5283 This directive sets the value of @var{symbol} to @var{expression}.
5284 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5287 The syntax for @code{equ} on the HPPA is
5288 @samp{@var{symbol} .equ @var{expression}}.
5292 The syntax for @code{equ} on the Z80 is
5293 @samp{@var{symbol} equ @var{expression}}.
5294 On the Z80 it is an error if @var{symbol} is already defined,
5295 but the symbol is not protected from later redefinition.
5296 Compare @ref{Equiv}.
5300 @section @code{.equiv @var{symbol}, @var{expression}}
5301 @cindex @code{equiv} directive
5302 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5303 the assembler will signal an error if @var{symbol} is already defined. Note a
5304 symbol which has been referenced but not actually defined is considered to be
5307 Except for the contents of the error message, this is roughly equivalent to
5314 plus it protects the symbol from later redefinition.
5317 @section @code{.eqv @var{symbol}, @var{expression}}
5318 @cindex @code{eqv} directive
5319 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5320 evaluate the expression or any part of it immediately. Instead each time
5321 the resulting symbol is used in an expression, a snapshot of its current
5325 @section @code{.err}
5326 @cindex @code{err} directive
5327 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5328 message and, unless the @option{-Z} option was used, it will not generate an
5329 object file. This can be used to signal an error in conditionally compiled code.
5332 @section @code{.error "@var{string}"}
5333 @cindex error directive
5335 Similarly to @code{.err}, this directive emits an error, but you can specify a
5336 string that will be emitted as the error message. If you don't specify the
5337 message, it defaults to @code{".error directive invoked in source file"}.
5338 @xref{Errors, ,Error and Warning Messages}.
5341 .error "This code has not been assembled and tested."
5345 @section @code{.exitm}
5346 Exit early from the current macro definition. @xref{Macro}.
5349 @section @code{.extern}
5351 @cindex @code{extern} directive
5352 @code{.extern} is accepted in the source program---for compatibility
5353 with other assemblers---but it is ignored. @command{@value{AS}} treats
5354 all undefined symbols as external.
5357 @section @code{.fail @var{expression}}
5359 @cindex @code{fail} directive
5360 Generates an error or a warning. If the value of the @var{expression} is 500
5361 or more, @command{@value{AS}} will print a warning message. If the value is less
5362 than 500, @command{@value{AS}} will print an error message. The message will
5363 include the value of @var{expression}. This can occasionally be useful inside
5364 complex nested macros or conditional assembly.
5367 @section @code{.file}
5368 @cindex @code{file} directive
5370 @ifclear no-file-dir
5371 There are two different versions of the @code{.file} directive. Targets
5372 that support DWARF2 line number information use the DWARF2 version of
5373 @code{.file}. Other targets use the default version.
5375 @subheading Default Version
5377 @cindex logical file name
5378 @cindex file name, logical
5379 This version of the @code{.file} directive tells @command{@value{AS}} that we
5380 are about to start a new logical file. The syntax is:
5386 @var{string} is the new file name. In general, the filename is
5387 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5388 to specify an empty file name, you must give the quotes--@code{""}. This
5389 statement may go away in future: it is only recognized to be compatible with
5390 old @command{@value{AS}} programs.
5392 @subheading DWARF2 Version
5395 When emitting DWARF2 line number information, @code{.file} assigns filenames
5396 to the @code{.debug_line} file name table. The syntax is:
5399 .file @var{fileno} @var{filename}
5402 The @var{fileno} operand should be a unique positive integer to use as the
5403 index of the entry in the table. The @var{filename} operand is a C string
5406 The detail of filename indices is exposed to the user because the filename
5407 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5408 information, and thus the user must know the exact indices that table
5412 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5414 @cindex @code{fill} directive
5415 @cindex writing patterns in memory
5416 @cindex patterns, writing in memory
5417 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5418 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5419 may be zero or more. @var{Size} may be zero or more, but if it is
5420 more than 8, then it is deemed to have the value 8, compatible with
5421 other people's assemblers. The contents of each @var{repeat} bytes
5422 is taken from an 8-byte number. The highest order 4 bytes are
5423 zero. The lowest order 4 bytes are @var{value} rendered in the
5424 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5425 Each @var{size} bytes in a repetition is taken from the lowest order
5426 @var{size} bytes of this number. Again, this bizarre behavior is
5427 compatible with other people's assemblers.
5429 @var{size} and @var{value} are optional.
5430 If the second comma and @var{value} are absent, @var{value} is
5431 assumed zero. If the first comma and following tokens are absent,
5432 @var{size} is assumed to be 1.
5435 @section @code{.float @var{flonums}}
5437 @cindex floating point numbers (single)
5438 @cindex @code{float} directive
5439 This directive assembles zero or more flonums, separated by commas. It
5440 has the same effect as @code{.single}.
5442 The exact kind of floating point numbers emitted depends on how
5443 @command{@value{AS}} is configured.
5444 @xref{Machine Dependencies}.
5448 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5449 in @sc{ieee} format.
5454 @section @code{.func @var{name}[,@var{label}]}
5455 @cindex @code{func} directive
5456 @code{.func} emits debugging information to denote function @var{name}, and
5457 is ignored unless the file is assembled with debugging enabled.
5458 Only @samp{--gstabs[+]} is currently supported.
5459 @var{label} is the entry point of the function and if omitted @var{name}
5460 prepended with the @samp{leading char} is used.
5461 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5462 All functions are currently defined to have @code{void} return type.
5463 The function must be terminated with @code{.endfunc}.
5466 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5468 @cindex @code{global} directive
5469 @cindex symbol, making visible to linker
5470 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5471 @var{symbol} in your partial program, its value is made available to
5472 other partial programs that are linked with it. Otherwise,
5473 @var{symbol} takes its attributes from a symbol of the same name
5474 from another file linked into the same program.
5476 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5477 compatibility with other assemblers.
5480 On the HPPA, @code{.global} is not always enough to make it accessible to other
5481 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5482 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5487 @section @code{.gnu_attribute @var{tag},@var{value}}
5488 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5491 @section @code{.hidden @var{names}}
5493 @cindex @code{hidden} directive
5495 This is one of the ELF visibility directives. The other two are
5496 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5497 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5499 This directive overrides the named symbols default visibility (which is set by
5500 their binding: local, global or weak). The directive sets the visibility to
5501 @code{hidden} which means that the symbols are not visible to other components.
5502 Such symbols are always considered to be @code{protected} as well.
5506 @section @code{.hword @var{expressions}}
5508 @cindex @code{hword} directive
5509 @cindex integers, 16-bit
5510 @cindex numbers, 16-bit
5511 @cindex sixteen bit integers
5512 This expects zero or more @var{expressions}, and emits
5513 a 16 bit number for each.
5516 This directive is a synonym for @samp{.short}; depending on the target
5517 architecture, it may also be a synonym for @samp{.word}.
5521 This directive is a synonym for @samp{.short}.
5524 This directive is a synonym for both @samp{.short} and @samp{.word}.
5529 @section @code{.ident}
5531 @cindex @code{ident} directive
5533 This directive is used by some assemblers to place tags in object files. The
5534 behavior of this directive varies depending on the target. When using the
5535 a.out object file format, @command{@value{AS}} simply accepts the directive for
5536 source-file compatibility with existing assemblers, but does not emit anything
5537 for it. When using COFF, comments are emitted to the @code{.comment} or
5538 @code{.rdata} section, depending on the target. When using ELF, comments are
5539 emitted to the @code{.comment} section.
5542 @section @code{.if @var{absolute expression}}
5544 @cindex conditional assembly
5545 @cindex @code{if} directive
5546 @code{.if} marks the beginning of a section of code which is only
5547 considered part of the source program being assembled if the argument
5548 (which must be an @var{absolute expression}) is non-zero. The end of
5549 the conditional section of code must be marked by @code{.endif}
5550 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5551 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5552 If you have several conditions to check, @code{.elseif} may be used to avoid
5553 nesting blocks if/else within each subsequent @code{.else} block.
5555 The following variants of @code{.if} are also supported:
5557 @cindex @code{ifdef} directive
5558 @item .ifdef @var{symbol}
5559 Assembles the following section of code if the specified @var{symbol}
5560 has been defined. Note a symbol which has been referenced but not yet defined
5561 is considered to be undefined.
5563 @cindex @code{ifb} directive
5564 @item .ifb @var{text}
5565 Assembles the following section of code if the operand is blank (empty).
5567 @cindex @code{ifc} directive
5568 @item .ifc @var{string1},@var{string2}
5569 Assembles the following section of code if the two strings are the same. The
5570 strings may be optionally quoted with single quotes. If they are not quoted,
5571 the first string stops at the first comma, and the second string stops at the
5572 end of the line. Strings which contain whitespace should be quoted. The
5573 string comparison is case sensitive.
5575 @cindex @code{ifeq} directive
5576 @item .ifeq @var{absolute expression}
5577 Assembles the following section of code if the argument is zero.
5579 @cindex @code{ifeqs} directive
5580 @item .ifeqs @var{string1},@var{string2}
5581 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5583 @cindex @code{ifge} directive
5584 @item .ifge @var{absolute expression}
5585 Assembles the following section of code if the argument is greater than or
5588 @cindex @code{ifgt} directive
5589 @item .ifgt @var{absolute expression}
5590 Assembles the following section of code if the argument is greater than zero.
5592 @cindex @code{ifle} directive
5593 @item .ifle @var{absolute expression}
5594 Assembles the following section of code if the argument is less than or equal
5597 @cindex @code{iflt} directive
5598 @item .iflt @var{absolute expression}
5599 Assembles the following section of code if the argument is less than zero.
5601 @cindex @code{ifnb} directive
5602 @item .ifnb @var{text}
5603 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5604 following section of code if the operand is non-blank (non-empty).
5606 @cindex @code{ifnc} directive
5607 @item .ifnc @var{string1},@var{string2}.
5608 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5609 following section of code if the two strings are not the same.
5611 @cindex @code{ifndef} directive
5612 @cindex @code{ifnotdef} directive
5613 @item .ifndef @var{symbol}
5614 @itemx .ifnotdef @var{symbol}
5615 Assembles the following section of code if the specified @var{symbol}
5616 has not been defined. Both spelling variants are equivalent. Note a symbol
5617 which has been referenced but not yet defined is considered to be undefined.
5619 @cindex @code{ifne} directive
5620 @item .ifne @var{absolute expression}
5621 Assembles the following section of code if the argument is not equal to zero
5622 (in other words, this is equivalent to @code{.if}).
5624 @cindex @code{ifnes} directive
5625 @item .ifnes @var{string1},@var{string2}
5626 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5627 following section of code if the two strings are not the same.
5631 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5633 @cindex @code{incbin} directive
5634 @cindex binary files, including
5635 The @code{incbin} directive includes @var{file} verbatim at the current
5636 location. You can control the search paths used with the @samp{-I} command-line
5637 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5640 The @var{skip} argument skips a number of bytes from the start of the
5641 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5642 read. Note that the data is not aligned in any way, so it is the user's
5643 responsibility to make sure that proper alignment is provided both before and
5644 after the @code{incbin} directive.
5647 @section @code{.include "@var{file}"}
5649 @cindex @code{include} directive
5650 @cindex supporting files, including
5651 @cindex files, including
5652 This directive provides a way to include supporting files at specified
5653 points in your source program. The code from @var{file} is assembled as
5654 if it followed the point of the @code{.include}; when the end of the
5655 included file is reached, assembly of the original file continues. You
5656 can control the search paths used with the @samp{-I} command-line option
5657 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5661 @section @code{.int @var{expressions}}
5663 @cindex @code{int} directive
5664 @cindex integers, 32-bit
5665 Expect zero or more @var{expressions}, of any section, separated by commas.
5666 For each expression, emit a number that, at run time, is the value of that
5667 expression. The byte order and bit size of the number depends on what kind
5668 of target the assembly is for.
5672 On most forms of the H8/300, @code{.int} emits 16-bit
5673 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5680 @section @code{.internal @var{names}}
5682 @cindex @code{internal} directive
5684 This is one of the ELF visibility directives. The other two are
5685 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5686 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5688 This directive overrides the named symbols default visibility (which is set by
5689 their binding: local, global or weak). The directive sets the visibility to
5690 @code{internal} which means that the symbols are considered to be @code{hidden}
5691 (i.e., not visible to other components), and that some extra, processor specific
5692 processing must also be performed upon the symbols as well.
5696 @section @code{.irp @var{symbol},@var{values}}@dots{}
5698 @cindex @code{irp} directive
5699 Evaluate a sequence of statements assigning different values to @var{symbol}.
5700 The sequence of statements starts at the @code{.irp} directive, and is
5701 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5702 set to @var{value}, and the sequence of statements is assembled. If no
5703 @var{value} is listed, the sequence of statements is assembled once, with
5704 @var{symbol} set to the null string. To refer to @var{symbol} within the
5705 sequence of statements, use @var{\symbol}.
5707 For example, assembling
5715 is equivalent to assembling
5723 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5726 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5728 @cindex @code{irpc} directive
5729 Evaluate a sequence of statements assigning different values to @var{symbol}.
5730 The sequence of statements starts at the @code{.irpc} directive, and is
5731 terminated by an @code{.endr} directive. For each character in @var{value},
5732 @var{symbol} is set to the character, and the sequence of statements is
5733 assembled. If no @var{value} is listed, the sequence of statements is
5734 assembled once, with @var{symbol} set to the null string. To refer to
5735 @var{symbol} within the sequence of statements, use @var{\symbol}.
5737 For example, assembling
5745 is equivalent to assembling
5753 For some caveats with the spelling of @var{symbol}, see also the discussion
5757 @section @code{.lcomm @var{symbol} , @var{length}}
5759 @cindex @code{lcomm} directive
5760 @cindex local common symbols
5761 @cindex symbols, local common
5762 Reserve @var{length} (an absolute expression) bytes for a local common
5763 denoted by @var{symbol}. The section and value of @var{symbol} are
5764 those of the new local common. The addresses are allocated in the bss
5765 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5766 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5767 not visible to @code{@value{LD}}.
5770 Some targets permit a third argument to be used with @code{.lcomm}. This
5771 argument specifies the desired alignment of the symbol in the bss section.
5775 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5776 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5780 @section @code{.lflags}
5782 @cindex @code{lflags} directive (ignored)
5783 @command{@value{AS}} accepts this directive, for compatibility with other
5784 assemblers, but ignores it.
5786 @ifclear no-line-dir
5788 @section @code{.line @var{line-number}}
5790 @cindex @code{line} directive
5791 @cindex logical line number
5793 Change the logical line number. @var{line-number} must be an absolute
5794 expression. The next line has that logical line number. Therefore any other
5795 statements on the current line (after a statement separator character) are
5796 reported as on logical line number @var{line-number} @minus{} 1. One day
5797 @command{@value{AS}} will no longer support this directive: it is recognized only
5798 for compatibility with existing assembler programs.
5801 Even though this is a directive associated with the @code{a.out} or
5802 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5803 when producing COFF output, and treats @samp{.line} as though it
5804 were the COFF @samp{.ln} @emph{if} it is found outside a
5805 @code{.def}/@code{.endef} pair.
5807 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5808 used by compilers to generate auxiliary symbol information for
5813 @section @code{.linkonce [@var{type}]}
5815 @cindex @code{linkonce} directive
5816 @cindex common sections
5817 Mark the current section so that the linker only includes a single copy of it.
5818 This may be used to include the same section in several different object files,
5819 but ensure that the linker will only include it once in the final output file.
5820 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5821 Duplicate sections are detected based on the section name, so it should be
5824 This directive is only supported by a few object file formats; as of this
5825 writing, the only object file format which supports it is the Portable
5826 Executable format used on Windows NT.
5828 The @var{type} argument is optional. If specified, it must be one of the
5829 following strings. For example:
5833 Not all types may be supported on all object file formats.
5837 Silently discard duplicate sections. This is the default.
5840 Warn if there are duplicate sections, but still keep only one copy.
5843 Warn if any of the duplicates have different sizes.
5846 Warn if any of the duplicates do not have exactly the same contents.
5850 @section @code{.list}
5852 @cindex @code{list} directive
5853 @cindex listing control, turning on
5854 Control (in conjunction with the @code{.nolist} directive) whether or
5855 not assembly listings are generated. These two directives maintain an
5856 internal counter (which is zero initially). @code{.list} increments the
5857 counter, and @code{.nolist} decrements it. Assembly listings are
5858 generated whenever the counter is greater than zero.
5860 By default, listings are disabled. When you enable them (with the
5861 @samp{-a} command-line option; @pxref{Invoking,,Command-Line Options}),
5862 the initial value of the listing counter is one.
5865 @section @code{.ln @var{line-number}}
5867 @cindex @code{ln} directive
5868 @ifclear no-line-dir
5869 @samp{.ln} is a synonym for @samp{.line}.
5872 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5873 must be an absolute expression. The next line has that logical
5874 line number, so any other statements on the current line (after a
5875 statement separator character @code{;}) are reported as on logical
5876 line number @var{line-number} @minus{} 1.
5880 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5881 @cindex @code{loc} directive
5882 When emitting DWARF2 line number information,
5883 the @code{.loc} directive will add a row to the @code{.debug_line} line
5884 number matrix corresponding to the immediately following assembly
5885 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5886 arguments will be applied to the @code{.debug_line} state machine before
5889 The @var{options} are a sequence of the following tokens in any order:
5893 This option will set the @code{basic_block} register in the
5894 @code{.debug_line} state machine to @code{true}.
5897 This option will set the @code{prologue_end} register in the
5898 @code{.debug_line} state machine to @code{true}.
5900 @item epilogue_begin
5901 This option will set the @code{epilogue_begin} register in the
5902 @code{.debug_line} state machine to @code{true}.
5904 @item is_stmt @var{value}
5905 This option will set the @code{is_stmt} register in the
5906 @code{.debug_line} state machine to @code{value}, which must be
5909 @item isa @var{value}
5910 This directive will set the @code{isa} register in the @code{.debug_line}
5911 state machine to @var{value}, which must be an unsigned integer.
5913 @item discriminator @var{value}
5914 This directive will set the @code{discriminator} register in the @code{.debug_line}
5915 state machine to @var{value}, which must be an unsigned integer.
5917 @item view @var{value}
5918 This option causes a row to be added to @code{.debug_line} in reference to the
5919 current address (which might not be the same as that of the following assembly
5920 instruction), and to associate @var{value} with the @code{view} register in the
5921 @code{.debug_line} state machine. If @var{value} is a label, both the
5922 @code{view} register and the label are set to the number of prior @code{.loc}
5923 directives at the same program location. If @var{value} is the literal
5924 @code{0}, the @code{view} register is set to zero, and the assembler asserts
5925 that there aren't any prior @code{.loc} directives at the same program
5926 location. If @var{value} is the literal @code{-0}, the assembler arrange for
5927 the @code{view} register to be reset in this row, even if there are prior
5928 @code{.loc} directives at the same program location.
5932 @node Loc_mark_labels
5933 @section @code{.loc_mark_labels @var{enable}}
5934 @cindex @code{loc_mark_labels} directive
5935 When emitting DWARF2 line number information,
5936 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5937 to the @code{.debug_line} line number matrix with the @code{basic_block}
5938 register in the state machine set whenever a code label is seen.
5939 The @var{enable} argument should be either 1 or 0, to enable or disable
5940 this function respectively.
5944 @section @code{.local @var{names}}
5946 @cindex @code{local} directive
5947 This directive, which is available for ELF targets, marks each symbol in
5948 the comma-separated list of @code{names} as a local symbol so that it
5949 will not be externally visible. If the symbols do not already exist,
5950 they will be created.
5952 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5953 accept an alignment argument, which is the case for most ELF targets,
5954 the @code{.local} directive can be used in combination with @code{.comm}
5955 (@pxref{Comm}) to define aligned local common data.
5959 @section @code{.long @var{expressions}}
5961 @cindex @code{long} directive
5962 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5965 @c no one seems to know what this is for or whether this description is
5966 @c what it really ought to do
5968 @section @code{.lsym @var{symbol}, @var{expression}}
5970 @cindex @code{lsym} directive
5971 @cindex symbol, not referenced in assembly
5972 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5973 the hash table, ensuring it cannot be referenced by name during the
5974 rest of the assembly. This sets the attributes of the symbol to be
5975 the same as the expression value:
5977 @var{other} = @var{descriptor} = 0
5978 @var{type} = @r{(section of @var{expression})}
5979 @var{value} = @var{expression}
5982 The new symbol is not flagged as external.
5986 @section @code{.macro}
5989 The commands @code{.macro} and @code{.endm} allow you to define macros that
5990 generate assembly output. For example, this definition specifies a macro
5991 @code{sum} that puts a sequence of numbers into memory:
5994 .macro sum from=0, to=5
6003 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
6015 @item .macro @var{macname}
6016 @itemx .macro @var{macname} @var{macargs} @dots{}
6017 @cindex @code{macro} directive
6018 Begin the definition of a macro called @var{macname}. If your macro
6019 definition requires arguments, specify their names after the macro name,
6020 separated by commas or spaces. You can qualify the macro argument to
6021 indicate whether all invocations must specify a non-blank value (through
6022 @samp{:@code{req}}), or whether it takes all of the remaining arguments
6023 (through @samp{:@code{vararg}}). You can supply a default value for any
6024 macro argument by following the name with @samp{=@var{deflt}}. You
6025 cannot define two macros with the same @var{macname} unless it has been
6026 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
6027 definitions. For example, these are all valid @code{.macro} statements:
6031 Begin the definition of a macro called @code{comm}, which takes no
6034 @item .macro plus1 p, p1
6035 @itemx .macro plus1 p p1
6036 Either statement begins the definition of a macro called @code{plus1},
6037 which takes two arguments; within the macro definition, write
6038 @samp{\p} or @samp{\p1} to evaluate the arguments.
6040 @item .macro reserve_str p1=0 p2
6041 Begin the definition of a macro called @code{reserve_str}, with two
6042 arguments. The first argument has a default value, but not the second.
6043 After the definition is complete, you can call the macro either as
6044 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
6045 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
6046 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
6047 @samp{0}, and @samp{\p2} evaluating to @var{b}).
6049 @item .macro m p1:req, p2=0, p3:vararg
6050 Begin the definition of a macro called @code{m}, with at least three
6051 arguments. The first argument must always have a value specified, but
6052 not the second, which instead has a default value. The third formal
6053 will get assigned all remaining arguments specified at invocation time.
6055 When you call a macro, you can specify the argument values either by
6056 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
6057 @samp{sum to=17, from=9}.
6061 Note that since each of the @var{macargs} can be an identifier exactly
6062 as any other one permitted by the target architecture, there may be
6063 occasional problems if the target hand-crafts special meanings to certain
6064 characters when they occur in a special position. For example, if the colon
6065 (@code{:}) is generally permitted to be part of a symbol name, but the
6066 architecture specific code special-cases it when occurring as the final
6067 character of a symbol (to denote a label), then the macro parameter
6068 replacement code will have no way of knowing that and consider the whole
6069 construct (including the colon) an identifier, and check only this
6070 identifier for being the subject to parameter substitution. So for example
6071 this macro definition:
6079 might not work as expected. Invoking @samp{label foo} might not create a label
6080 called @samp{foo} but instead just insert the text @samp{\l:} into the
6081 assembler source, probably generating an error about an unrecognised
6084 Similarly problems might occur with the period character (@samp{.})
6085 which is often allowed inside opcode names (and hence identifier names). So
6086 for example constructing a macro to build an opcode from a base name and a
6087 length specifier like this:
6090 .macro opcode base length
6095 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
6096 instruction but instead generate some kind of error as the assembler tries to
6097 interpret the text @samp{\base.\length}.
6099 There are several possible ways around this problem:
6102 @item Insert white space
6103 If it is possible to use white space characters then this is the simplest
6112 @item Use @samp{\()}
6113 The string @samp{\()} can be used to separate the end of a macro argument from
6114 the following text. eg:
6117 .macro opcode base length
6122 @item Use the alternate macro syntax mode
6123 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
6124 used as a separator. eg:
6134 Note: this problem of correctly identifying string parameters to pseudo ops
6135 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
6136 and @code{.irpc} (@pxref{Irpc}) as well.
6139 @cindex @code{endm} directive
6140 Mark the end of a macro definition.
6143 @cindex @code{exitm} directive
6144 Exit early from the current macro definition.
6146 @cindex number of macros executed
6147 @cindex macros, count executed
6149 @command{@value{AS}} maintains a counter of how many macros it has
6150 executed in this pseudo-variable; you can copy that number to your
6151 output with @samp{\@@}, but @emph{only within a macro definition}.
6153 @item LOCAL @var{name} [ , @dots{} ]
6154 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
6155 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
6156 @xref{Altmacro,,@code{.altmacro}}.
6160 @section @code{.mri @var{val}}
6162 @cindex @code{mri} directive
6163 @cindex MRI mode, temporarily
6164 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
6165 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
6166 affects code assembled until the next @code{.mri} directive, or until the end
6167 of the file. @xref{M, MRI mode, MRI mode}.
6170 @section @code{.noaltmacro}
6171 Disable alternate macro mode. @xref{Altmacro}.
6174 @section @code{.nolist}
6176 @cindex @code{nolist} directive
6177 @cindex listing control, turning off
6178 Control (in conjunction with the @code{.list} directive) whether or
6179 not assembly listings are generated. These two directives maintain an
6180 internal counter (which is zero initially). @code{.list} increments the
6181 counter, and @code{.nolist} decrements it. Assembly listings are
6182 generated whenever the counter is greater than zero.
6185 @section @code{.nops @var{size}[, @var{control}]}
6187 @cindex @code{nops} directive
6188 @cindex filling memory with no-op instructions
6189 This directive emits @var{size} bytes filled with no-op instructions.
6190 @var{size} is absolute expression, which must be a positve value.
6191 @var{control} controls how no-op instructions should be generated. If
6192 the comma and @var{control} are omitted, @var{control} is assumed to be
6195 Note: For Intel 80386 and AMD x86-64 targets, @var{control} specifies
6196 the size limit of a no-op instruction. The valid values of @var{control}
6197 are between 0 and 4 in 16-bit mode, between 0 and 7 when tuning for
6198 older processors in 32-bit mode, between 0 and 11 in 64-bit mode or when
6199 tuning for newer processors in 32-bit mode. When 0 is used, the no-op
6200 instruction size limit is set to the maximum supported size.
6203 @section @code{.octa @var{bignums}}
6205 @c FIXME: double size emitted for "octa" on some? Or warn?
6206 @cindex @code{octa} directive
6207 @cindex integer, 16-byte
6208 @cindex sixteen byte integer
6209 This directive expects zero or more bignums, separated by commas. For each
6210 bignum, it emits a 16-byte integer.
6212 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6213 hence @emph{octa}-word for 16 bytes.
6216 @section @code{.offset @var{loc}}
6218 @cindex @code{offset} directive
6219 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6220 be an absolute expression. This directive may be useful for defining
6221 symbols with absolute values. Do not confuse it with the @code{.org}
6225 @section @code{.org @var{new-lc} , @var{fill}}
6227 @cindex @code{org} directive
6228 @cindex location counter, advancing
6229 @cindex advancing location counter
6230 @cindex current address, advancing
6231 Advance the location counter of the current section to
6232 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6233 expression with the same section as the current subsection. That is,
6234 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6235 wrong section, the @code{.org} directive is ignored. To be compatible
6236 with former assemblers, if the section of @var{new-lc} is absolute,
6237 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6238 is the same as the current subsection.
6240 @code{.org} may only increase the location counter, or leave it
6241 unchanged; you cannot use @code{.org} to move the location counter
6244 @c double negative used below "not undefined" because this is a specific
6245 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6246 @c section. doc@cygnus.com 18feb91
6247 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6248 may not be undefined. If you really detest this restriction we eagerly await
6249 a chance to share your improved assembler.
6251 Beware that the origin is relative to the start of the section, not
6252 to the start of the subsection. This is compatible with other
6253 people's assemblers.
6255 When the location counter (of the current subsection) is advanced, the
6256 intervening bytes are filled with @var{fill} which should be an
6257 absolute expression. If the comma and @var{fill} are omitted,
6258 @var{fill} defaults to zero.
6261 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
6263 @cindex padding the location counter given a power of two
6264 @cindex @code{p2align} directive
6265 Pad the location counter (in the current subsection) to a particular
6266 storage boundary. The first expression (which must be absolute) is the
6267 number of low-order zero bits the location counter must have after
6268 advancement. For example @samp{.p2align 3} advances the location
6269 counter until it is a multiple of 8. If the location counter is already a
6270 multiple of 8, no change is needed.
6272 The second expression (also absolute) gives the fill value to be stored in the
6273 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6274 padding bytes are normally zero. However, on most systems, if the section is
6275 marked as containing code and the fill value is omitted, the space is filled
6276 with no-op instructions.
6278 The third expression is also absolute, and is also optional. If it is present,
6279 it is the maximum number of bytes that should be skipped by this alignment
6280 directive. If doing the alignment would require skipping more bytes than the
6281 specified maximum, then the alignment is not done at all. You can omit the
6282 fill value (the second argument) entirely by simply using two commas after the
6283 required alignment; this can be useful if you want the alignment to be filled
6284 with no-op instructions when appropriate.
6286 @cindex @code{p2alignw} directive
6287 @cindex @code{p2alignl} directive
6288 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6289 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6290 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6291 fill pattern as a four byte longword value. For example, @code{.p2alignw
6292 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6293 filled in with the value 0x368d (the exact placement of the bytes depends upon
6294 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6299 @section @code{.popsection}
6301 @cindex @code{popsection} directive
6302 @cindex Section Stack
6303 This is one of the ELF section stack manipulation directives. The others are
6304 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6305 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6308 This directive replaces the current section (and subsection) with the top
6309 section (and subsection) on the section stack. This section is popped off the
6315 @section @code{.previous}
6317 @cindex @code{previous} directive
6318 @cindex Section Stack
6319 This is one of the ELF section stack manipulation directives. The others are
6320 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6321 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6322 (@pxref{PopSection}).
6324 This directive swaps the current section (and subsection) with most recently
6325 referenced section/subsection pair prior to this one. Multiple
6326 @code{.previous} directives in a row will flip between two sections (and their
6327 subsections). For example:
6339 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6345 # Now in section A subsection 1
6349 # Now in section B subsection 0
6352 # Now in section B subsection 1
6355 # Now in section B subsection 0
6359 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6360 section B and 0x9abc into subsection 1 of section B.
6362 In terms of the section stack, this directive swaps the current section with
6363 the top section on the section stack.
6367 @section @code{.print @var{string}}
6369 @cindex @code{print} directive
6370 @command{@value{AS}} will print @var{string} on the standard output during
6371 assembly. You must put @var{string} in double quotes.
6375 @section @code{.protected @var{names}}
6377 @cindex @code{protected} directive
6379 This is one of the ELF visibility directives. The other two are
6380 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6382 This directive overrides the named symbols default visibility (which is set by
6383 their binding: local, global or weak). The directive sets the visibility to
6384 @code{protected} which means that any references to the symbols from within the
6385 components that defines them must be resolved to the definition in that
6386 component, even if a definition in another component would normally preempt
6391 @section @code{.psize @var{lines} , @var{columns}}
6393 @cindex @code{psize} directive
6394 @cindex listing control: paper size
6395 @cindex paper size, for listings
6396 Use this directive to declare the number of lines---and, optionally, the
6397 number of columns---to use for each page, when generating listings.
6399 If you do not use @code{.psize}, listings use a default line-count
6400 of 60. You may omit the comma and @var{columns} specification; the
6401 default width is 200 columns.
6403 @command{@value{AS}} generates formfeeds whenever the specified number of
6404 lines is exceeded (or whenever you explicitly request one, using
6407 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6408 those explicitly specified with @code{.eject}.
6411 @section @code{.purgem @var{name}}
6413 @cindex @code{purgem} directive
6414 Undefine the macro @var{name}, so that later uses of the string will not be
6415 expanded. @xref{Macro}.
6419 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6421 @cindex @code{pushsection} directive
6422 @cindex Section Stack
6423 This is one of the ELF section stack manipulation directives. The others are
6424 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6425 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6428 This directive pushes the current section (and subsection) onto the
6429 top of the section stack, and then replaces the current section and
6430 subsection with @code{name} and @code{subsection}. The optional
6431 @code{flags}, @code{type} and @code{arguments} are treated the same
6432 as in the @code{.section} (@pxref{Section}) directive.
6436 @section @code{.quad @var{bignums}}
6438 @cindex @code{quad} directive
6439 @code{.quad} expects zero or more bignums, separated by commas. For
6440 each bignum, it emits
6442 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6443 warning message; and just takes the lowest order 8 bytes of the bignum.
6444 @cindex eight-byte integer
6445 @cindex integer, 8-byte
6447 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6448 hence @emph{quad}-word for 8 bytes.
6451 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6452 warning message; and just takes the lowest order 16 bytes of the bignum.
6453 @cindex sixteen-byte integer
6454 @cindex integer, 16-byte
6458 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6460 @cindex @code{reloc} directive
6461 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6462 @var{expression}. If @var{offset} is a number, the relocation is generated in
6463 the current section. If @var{offset} is an expression that resolves to a
6464 symbol plus offset, the relocation is generated in the given symbol's section.
6465 @var{expression}, if present, must resolve to a symbol plus addend or to an
6466 absolute value, but note that not all targets support an addend. e.g. ELF REL
6467 targets such as i386 store an addend in the section contents rather than in the
6468 relocation. This low level interface does not support addends stored in the
6472 @section @code{.rept @var{count}}
6474 @cindex @code{rept} directive
6475 Repeat the sequence of lines between the @code{.rept} directive and the next
6476 @code{.endr} directive @var{count} times.
6478 For example, assembling
6486 is equivalent to assembling
6494 A count of zero is allowed, but nothing is generated. Negative counts are not
6495 allowed and if encountered will be treated as if they were zero.
6498 @section @code{.sbttl "@var{subheading}"}
6500 @cindex @code{sbttl} directive
6501 @cindex subtitles for listings
6502 @cindex listing control: subtitle
6503 Use @var{subheading} as the title (third line, immediately after the
6504 title line) when generating assembly listings.
6506 This directive affects subsequent pages, as well as the current page if
6507 it appears within ten lines of the top of a page.
6511 @section @code{.scl @var{class}}
6513 @cindex @code{scl} directive
6514 @cindex symbol storage class (COFF)
6515 @cindex COFF symbol storage class
6516 Set the storage-class value for a symbol. This directive may only be
6517 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6518 whether a symbol is static or external, or it may record further
6519 symbolic debugging information.
6524 @section @code{.section @var{name}}
6526 @cindex named section
6527 Use the @code{.section} directive to assemble the following code into a section
6530 This directive is only supported for targets that actually support arbitrarily
6531 named sections; on @code{a.out} targets, for example, it is not accepted, even
6532 with a standard @code{a.out} section name.
6536 @c only print the extra heading if both COFF and ELF are set
6537 @subheading COFF Version
6540 @cindex @code{section} directive (COFF version)
6541 For COFF targets, the @code{.section} directive is used in one of the following
6545 .section @var{name}[, "@var{flags}"]
6546 .section @var{name}[, @var{subsection}]
6549 If the optional argument is quoted, it is taken as flags to use for the
6550 section. Each flag is a single character. The following flags are recognized:
6554 bss section (uninitialized data)
6556 section is not loaded
6562 exclude section from linking
6568 shared section (meaningful for PE targets)
6570 ignored. (For compatibility with the ELF version)
6572 section is not readable (meaningful for PE targets)
6574 single-digit power-of-two section alignment (GNU extension)
6577 If no flags are specified, the default flags depend upon the section name. If
6578 the section name is not recognized, the default will be for the section to be
6579 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6580 from the section, rather than adding them, so if they are used on their own it
6581 will be as if no flags had been specified at all.
6583 If the optional argument to the @code{.section} directive is not quoted, it is
6584 taken as a subsection number (@pxref{Sub-Sections}).
6589 @c only print the extra heading if both COFF and ELF are set
6590 @subheading ELF Version
6593 @cindex Section Stack
6594 This is one of the ELF section stack manipulation directives. The others are
6595 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6596 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6597 @code{.previous} (@pxref{Previous}).
6599 @cindex @code{section} directive (ELF version)
6600 For ELF targets, the @code{.section} directive is used like this:
6603 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6606 @anchor{Section Name Substitutions}
6607 @kindex --sectname-subst
6608 @cindex section name substitution
6609 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6610 argument may contain a substitution sequence. Only @code{%S} is supported
6611 at the moment, and substitutes the current section name. For example:
6614 .macro exception_code
6615 .section %S.exception
6616 [exception code here]
6631 The two @code{exception_code} invocations above would create the
6632 @code{.text.exception} and @code{.init.exception} sections respectively.
6633 This is useful e.g. to discriminate between ancillary sections that are
6634 tied to setup code to be discarded after use from ancillary sections that
6635 need to stay resident without having to define multiple @code{exception_code}
6636 macros just for that purpose.
6638 The optional @var{flags} argument is a quoted string which may contain any
6639 combination of the following characters:
6643 section is allocatable
6645 section is a GNU_MBIND section
6647 section is excluded from executable and shared library.
6651 section is executable
6653 section is mergeable
6655 section contains zero terminated strings
6657 section is a member of a section group
6659 section is used for thread-local-storage
6661 section is a member of the previously-current section's group, if any
6662 @item @code{<number>}
6663 a numeric value indicating the bits to be set in the ELF section header's flags
6664 field. Note - if one or more of the alphabetic characters described above is
6665 also included in the flags field, their bit values will be ORed into the
6667 @item @code{<target specific>}
6668 some targets extend this list with their own flag characters
6671 Note - once a section's flags have been set they cannot be changed. There are
6672 a few exceptions to this rule however. Processor and application specific
6673 flags can be added to an already defined section. The @code{.interp},
6674 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6675 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6676 section may have the executable (@code{x}) flag added.
6678 The optional @var{type} argument may contain one of the following constants:
6682 section contains data
6684 section does not contain data (i.e., section only occupies space)
6686 section contains data which is used by things other than the program
6688 section contains an array of pointers to init functions
6690 section contains an array of pointers to finish functions
6691 @item @@preinit_array
6692 section contains an array of pointers to pre-init functions
6693 @item @@@code{<number>}
6694 a numeric value to be set as the ELF section header's type field.
6695 @item @@@code{<target specific>}
6696 some targets extend this list with their own types
6699 Many targets only support the first three section types. The type may be
6700 enclosed in double quotes if necessary.
6702 Note on targets where the @code{@@} character is the start of a comment (eg
6703 ARM) then another character is used instead. For example the ARM port uses the
6706 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6707 special and have fixed types. Any attempt to declare them with a different
6708 type will generate an error from the assembler.
6710 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6711 be specified as well as an extra argument---@var{entsize}---like this:
6714 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6717 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6718 constants, each @var{entsize} octets long. Sections with both @code{M} and
6719 @code{S} must contain zero terminated strings where each character is
6720 @var{entsize} bytes long. The linker may remove duplicates within sections with
6721 the same name, same entity size and same flags. @var{entsize} must be an
6722 absolute expression. For sections with both @code{M} and @code{S}, a string
6723 which is a suffix of a larger string is considered a duplicate. Thus
6724 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6725 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6727 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6728 be present along with an additional field like this:
6731 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6734 The @var{GroupName} field specifies the name of the section group to which this
6735 particular section belongs. The optional linkage field can contain:
6739 indicates that only one copy of this section should be retained
6744 Note: if both the @var{M} and @var{G} flags are present then the fields for
6745 the Merge flag should come first, like this:
6748 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6751 If @var{flags} contains the @code{?} symbol then it may not also contain the
6752 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6753 present. Instead, @code{?} says to consider the section that's current before
6754 this directive. If that section used @code{G}, then the new section will use
6755 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6756 If not, then the @code{?} symbol has no effect.
6758 If no flags are specified, the default flags depend upon the section name. If
6759 the section name is not recognized, the default will be for the section to have
6760 none of the above flags: it will not be allocated in memory, nor writable, nor
6761 executable. The section will contain data.
6763 For ELF targets, the assembler supports another type of @code{.section}
6764 directive for compatibility with the Solaris assembler:
6767 .section "@var{name}"[, @var{flags}...]
6770 Note that the section name is quoted. There may be a sequence of comma
6775 section is allocatable
6779 section is executable
6781 section is excluded from executable and shared library.
6783 section is used for thread local storage
6786 This directive replaces the current section and subsection. See the
6787 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6788 some examples of how this directive and the other section stack directives
6794 @section @code{.set @var{symbol}, @var{expression}}
6796 @cindex @code{set} directive
6797 @cindex symbol value, setting
6798 Set the value of @var{symbol} to @var{expression}. This
6799 changes @var{symbol}'s value and type to conform to
6800 @var{expression}. If @var{symbol} was flagged as external, it remains
6801 flagged (@pxref{Symbol Attributes}).
6803 You may @code{.set} a symbol many times in the same assembly provided that the
6804 values given to the symbol are constants. Values that are based on expressions
6805 involving other symbols are allowed, but some targets may restrict this to only
6806 being done once per assembly. This is because those targets do not set the
6807 addresses of symbols at assembly time, but rather delay the assignment until a
6808 final link is performed. This allows the linker a chance to change the code in
6809 the files, changing the location of, and the relative distance between, various
6812 If you @code{.set} a global symbol, the value stored in the object
6813 file is the last value stored into it.
6816 On Z80 @code{set} is a real instruction, use @code{.set} or
6817 @samp{@var{symbol} defl @var{expression}} instead.
6821 @section @code{.short @var{expressions}}
6823 @cindex @code{short} directive
6825 @code{.short} is normally the same as @samp{.word}.
6826 @xref{Word,,@code{.word}}.
6828 In some configurations, however, @code{.short} and @code{.word} generate
6829 numbers of different lengths. @xref{Machine Dependencies}.
6833 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6836 This expects zero or more @var{expressions}, and emits
6837 a 16 bit number for each.
6842 @section @code{.single @var{flonums}}
6844 @cindex @code{single} directive
6845 @cindex floating point numbers (single)
6846 This directive assembles zero or more flonums, separated by commas. It
6847 has the same effect as @code{.float}.
6849 The exact kind of floating point numbers emitted depends on how
6850 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6854 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6855 numbers in @sc{ieee} format.
6861 @section @code{.size}
6863 This directive is used to set the size associated with a symbol.
6867 @c only print the extra heading if both COFF and ELF are set
6868 @subheading COFF Version
6871 @cindex @code{size} directive (COFF version)
6872 For COFF targets, the @code{.size} directive is only permitted inside
6873 @code{.def}/@code{.endef} pairs. It is used like this:
6876 .size @var{expression}
6883 @c only print the extra heading if both COFF and ELF are set
6884 @subheading ELF Version
6887 @cindex @code{size} directive (ELF version)
6888 For ELF targets, the @code{.size} directive is used like this:
6891 .size @var{name} , @var{expression}
6894 This directive sets the size associated with a symbol @var{name}.
6895 The size in bytes is computed from @var{expression} which can make use of label
6896 arithmetic. This directive is typically used to set the size of function
6901 @ifclear no-space-dir
6903 @section @code{.skip @var{size} [,@var{fill}]}
6905 @cindex @code{skip} directive
6906 @cindex filling memory
6907 This directive emits @var{size} bytes, each of value @var{fill}. Both
6908 @var{size} and @var{fill} are absolute expressions. If the comma and
6909 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6914 @section @code{.sleb128 @var{expressions}}
6916 @cindex @code{sleb128} directive
6917 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6918 compact, variable length representation of numbers used by the DWARF
6919 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6921 @ifclear no-space-dir
6923 @section @code{.space @var{size} [,@var{fill}]}
6925 @cindex @code{space} directive
6926 @cindex filling memory
6927 This directive emits @var{size} bytes, each of value @var{fill}. Both
6928 @var{size} and @var{fill} are absolute expressions. If the comma
6929 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6934 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6935 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6936 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6937 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6945 @section @code{.stabd, .stabn, .stabs}
6947 @cindex symbolic debuggers, information for
6948 @cindex @code{stab@var{x}} directives
6949 There are three directives that begin @samp{.stab}.
6950 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6951 The symbols are not entered in the @command{@value{AS}} hash table: they
6952 cannot be referenced elsewhere in the source file.
6953 Up to five fields are required:
6957 This is the symbol's name. It may contain any character except
6958 @samp{\000}, so is more general than ordinary symbol names. Some
6959 debuggers used to code arbitrarily complex structures into symbol names
6963 An absolute expression. The symbol's type is set to the low 8 bits of
6964 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6965 and debuggers choke on silly bit patterns.
6968 An absolute expression. The symbol's ``other'' attribute is set to the
6969 low 8 bits of this expression.
6972 An absolute expression. The symbol's descriptor is set to the low 16
6973 bits of this expression.
6976 An absolute expression which becomes the symbol's value.
6979 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6980 or @code{.stabs} statement, the symbol has probably already been created;
6981 you get a half-formed symbol in your object file. This is
6982 compatible with earlier assemblers!
6985 @cindex @code{stabd} directive
6986 @item .stabd @var{type} , @var{other} , @var{desc}
6988 The ``name'' of the symbol generated is not even an empty string.
6989 It is a null pointer, for compatibility. Older assemblers used a
6990 null pointer so they didn't waste space in object files with empty
6993 The symbol's value is set to the location counter,
6994 relocatably. When your program is linked, the value of this symbol
6995 is the address of the location counter when the @code{.stabd} was
6998 @cindex @code{stabn} directive
6999 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
7000 The name of the symbol is set to the empty string @code{""}.
7002 @cindex @code{stabs} directive
7003 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
7004 All five fields are specified.
7010 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
7011 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
7013 @cindex string, copying to object file
7014 @cindex string8, copying to object file
7015 @cindex string16, copying to object file
7016 @cindex string32, copying to object file
7017 @cindex string64, copying to object file
7018 @cindex @code{string} directive
7019 @cindex @code{string8} directive
7020 @cindex @code{string16} directive
7021 @cindex @code{string32} directive
7022 @cindex @code{string64} directive
7024 Copy the characters in @var{str} to the object file. You may specify more than
7025 one string to copy, separated by commas. Unless otherwise specified for a
7026 particular machine, the assembler marks the end of each string with a 0 byte.
7027 You can use any of the escape sequences described in @ref{Strings,,Strings}.
7029 The variants @code{string16}, @code{string32} and @code{string64} differ from
7030 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
7031 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
7032 are stored in target endianness byte order.
7038 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
7039 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
7044 @section @code{.struct @var{expression}}
7046 @cindex @code{struct} directive
7047 Switch to the absolute section, and set the section offset to @var{expression},
7048 which must be an absolute expression. You might use this as follows:
7057 This would define the symbol @code{field1} to have the value 0, the symbol
7058 @code{field2} to have the value 4, and the symbol @code{field3} to have the
7059 value 8. Assembly would be left in the absolute section, and you would need to
7060 use a @code{.section} directive of some sort to change to some other section
7061 before further assembly.
7065 @section @code{.subsection @var{name}}
7067 @cindex @code{subsection} directive
7068 @cindex Section Stack
7069 This is one of the ELF section stack manipulation directives. The others are
7070 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
7071 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
7074 This directive replaces the current subsection with @code{name}. The current
7075 section is not changed. The replaced subsection is put onto the section stack
7076 in place of the then current top of stack subsection.
7081 @section @code{.symver}
7082 @cindex @code{symver} directive
7083 @cindex symbol versioning
7084 @cindex versions of symbols
7085 Use the @code{.symver} directive to bind symbols to specific version nodes
7086 within a source file. This is only supported on ELF platforms, and is
7087 typically used when assembling files to be linked into a shared library.
7088 There are cases where it may make sense to use this in objects to be bound
7089 into an application itself so as to override a versioned symbol from a
7092 For ELF targets, the @code{.symver} directive can be used like this:
7094 .symver @var{name}, @var{name2@@nodename}
7096 If the symbol @var{name} is defined within the file
7097 being assembled, the @code{.symver} directive effectively creates a symbol
7098 alias with the name @var{name2@@nodename}, and in fact the main reason that we
7099 just don't try and create a regular alias is that the @var{@@} character isn't
7100 permitted in symbol names. The @var{name2} part of the name is the actual name
7101 of the symbol by which it will be externally referenced. The name @var{name}
7102 itself is merely a name of convenience that is used so that it is possible to
7103 have definitions for multiple versions of a function within a single source
7104 file, and so that the compiler can unambiguously know which version of a
7105 function is being mentioned. The @var{nodename} portion of the alias should be
7106 the name of a node specified in the version script supplied to the linker when
7107 building a shared library. If you are attempting to override a versioned
7108 symbol from a shared library, then @var{nodename} should correspond to the
7109 nodename of the symbol you are trying to override.
7111 If the symbol @var{name} is not defined within the file being assembled, all
7112 references to @var{name} will be changed to @var{name2@@nodename}. If no
7113 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
7116 Another usage of the @code{.symver} directive is:
7118 .symver @var{name}, @var{name2@@@@nodename}
7120 In this case, the symbol @var{name} must exist and be defined within
7121 the file being assembled. It is similar to @var{name2@@nodename}. The
7122 difference is @var{name2@@@@nodename} will also be used to resolve
7123 references to @var{name2} by the linker.
7125 The third usage of the @code{.symver} directive is:
7127 .symver @var{name}, @var{name2@@@@@@nodename}
7129 When @var{name} is not defined within the
7130 file being assembled, it is treated as @var{name2@@nodename}. When
7131 @var{name} is defined within the file being assembled, the symbol
7132 name, @var{name}, will be changed to @var{name2@@@@nodename}.
7137 @section @code{.tag @var{structname}}
7139 @cindex COFF structure debugging
7140 @cindex structure debugging, COFF
7141 @cindex @code{tag} directive
7142 This directive is generated by compilers to include auxiliary debugging
7143 information in the symbol table. It is only permitted inside
7144 @code{.def}/@code{.endef} pairs. Tags are used to link structure
7145 definitions in the symbol table with instances of those structures.
7149 @section @code{.text @var{subsection}}
7151 @cindex @code{text} directive
7152 Tells @command{@value{AS}} to assemble the following statements onto the end of
7153 the text subsection numbered @var{subsection}, which is an absolute
7154 expression. If @var{subsection} is omitted, subsection number zero
7158 @section @code{.title "@var{heading}"}
7160 @cindex @code{title} directive
7161 @cindex listing control: title line
7162 Use @var{heading} as the title (second line, immediately after the
7163 source file name and pagenumber) when generating assembly listings.
7165 This directive affects subsequent pages, as well as the current page if
7166 it appears within ten lines of the top of a page.
7170 @section @code{.type}
7172 This directive is used to set the type of a symbol.
7176 @c only print the extra heading if both COFF and ELF are set
7177 @subheading COFF Version
7180 @cindex COFF symbol type
7181 @cindex symbol type, COFF
7182 @cindex @code{type} directive (COFF version)
7183 For COFF targets, this directive is permitted only within
7184 @code{.def}/@code{.endef} pairs. It is used like this:
7190 This records the integer @var{int} as the type attribute of a symbol table
7197 @c only print the extra heading if both COFF and ELF are set
7198 @subheading ELF Version
7201 @cindex ELF symbol type
7202 @cindex symbol type, ELF
7203 @cindex @code{type} directive (ELF version)
7204 For ELF targets, the @code{.type} directive is used like this:
7207 .type @var{name} , @var{type description}
7210 This sets the type of symbol @var{name} to be either a
7211 function symbol or an object symbol. There are five different syntaxes
7212 supported for the @var{type description} field, in order to provide
7213 compatibility with various other assemblers.
7215 Because some of the characters used in these syntaxes (such as @samp{@@} and
7216 @samp{#}) are comment characters for some architectures, some of the syntaxes
7217 below do not work on all architectures. The first variant will be accepted by
7218 the GNU assembler on all architectures so that variant should be used for
7219 maximum portability, if you do not need to assemble your code with other
7222 The syntaxes supported are:
7225 .type <name> STT_<TYPE_IN_UPPER_CASE>
7226 .type <name>,#<type>
7227 .type <name>,@@<type>
7228 .type <name>,%<type>
7229 .type <name>,"<type>"
7232 The types supported are:
7237 Mark the symbol as being a function name.
7240 @itemx gnu_indirect_function
7241 Mark the symbol as an indirect function when evaluated during reloc
7242 processing. (This is only supported on assemblers targeting GNU systems).
7246 Mark the symbol as being a data object.
7250 Mark the symbol as being a thread-local data object.
7254 Mark the symbol as being a common data object.
7258 Does not mark the symbol in any way. It is supported just for completeness.
7260 @item gnu_unique_object
7261 Marks the symbol as being a globally unique data object. The dynamic linker
7262 will make sure that in the entire process there is just one symbol with this
7263 name and type in use. (This is only supported on assemblers targeting GNU
7268 Changing between incompatible types other than from/to STT_NOTYPE will
7269 result in a diagnostic. An intermediate change to STT_NOTYPE will silence
7272 Note: Some targets support extra types in addition to those listed above.
7278 @section @code{.uleb128 @var{expressions}}
7280 @cindex @code{uleb128} directive
7281 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7282 compact, variable length representation of numbers used by the DWARF
7283 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7287 @section @code{.val @var{addr}}
7289 @cindex @code{val} directive
7290 @cindex COFF value attribute
7291 @cindex value attribute, COFF
7292 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7293 records the address @var{addr} as the value attribute of a symbol table
7299 @section @code{.version "@var{string}"}
7301 @cindex @code{version} directive
7302 This directive creates a @code{.note} section and places into it an ELF
7303 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7308 @section @code{.vtable_entry @var{table}, @var{offset}}
7310 @cindex @code{vtable_entry} directive
7311 This directive finds or creates a symbol @code{table} and creates a
7312 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7315 @section @code{.vtable_inherit @var{child}, @var{parent}}
7317 @cindex @code{vtable_inherit} directive
7318 This directive finds the symbol @code{child} and finds or creates the symbol
7319 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7320 parent whose addend is the value of the child symbol. As a special case the
7321 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7325 @section @code{.warning "@var{string}"}
7326 @cindex warning directive
7327 Similar to the directive @code{.error}
7328 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7331 @section @code{.weak @var{names}}
7333 @cindex @code{weak} directive
7334 This directive sets the weak attribute on the comma separated list of symbol
7335 @code{names}. If the symbols do not already exist, they will be created.
7337 On COFF targets other than PE, weak symbols are a GNU extension. This
7338 directive sets the weak attribute on the comma separated list of symbol
7339 @code{names}. If the symbols do not already exist, they will be created.
7341 On the PE target, weak symbols are supported natively as weak aliases.
7342 When a weak symbol is created that is not an alias, GAS creates an
7343 alternate symbol to hold the default value.
7346 @section @code{.weakref @var{alias}, @var{target}}
7348 @cindex @code{weakref} directive
7349 This directive creates an alias to the target symbol that enables the symbol to
7350 be referenced with weak-symbol semantics, but without actually making it weak.
7351 If direct references or definitions of the symbol are present, then the symbol
7352 will not be weak, but if all references to it are through weak references, the
7353 symbol will be marked as weak in the symbol table.
7355 The effect is equivalent to moving all references to the alias to a separate
7356 assembly source file, renaming the alias to the symbol in it, declaring the
7357 symbol as weak there, and running a reloadable link to merge the object files
7358 resulting from the assembly of the new source file and the old source file that
7359 had the references to the alias removed.
7361 The alias itself never makes to the symbol table, and is entirely handled
7362 within the assembler.
7365 @section @code{.word @var{expressions}}
7367 @cindex @code{word} directive
7368 This directive expects zero or more @var{expressions}, of any section,
7369 separated by commas.
7372 For each expression, @command{@value{AS}} emits a 32-bit number.
7375 For each expression, @command{@value{AS}} emits a 16-bit number.
7380 The size of the number emitted, and its byte order,
7381 depend on what target computer the assembly is for.
7384 @c on sparc the "special treatment to support compilers" doesn't
7385 @c happen---32-bit addressability, period; no long/short jumps.
7386 @ifset DIFF-TBL-KLUGE
7387 @cindex difference tables altered
7388 @cindex altered difference tables
7390 @emph{Warning: Special Treatment to support Compilers}
7394 Machines with a 32-bit address space, but that do less than 32-bit
7395 addressing, require the following special treatment. If the machine of
7396 interest to you does 32-bit addressing (or doesn't require it;
7397 @pxref{Machine Dependencies}), you can ignore this issue.
7400 In order to assemble compiler output into something that works,
7401 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7402 Directives of the form @samp{.word sym1-sym2} are often emitted by
7403 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7404 directive of the form @samp{.word sym1-sym2}, and the difference between
7405 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7406 creates a @dfn{secondary jump table}, immediately before the next label.
7407 This secondary jump table is preceded by a short-jump to the
7408 first byte after the secondary table. This short-jump prevents the flow
7409 of control from accidentally falling into the new table. Inside the
7410 table is a long-jump to @code{sym2}. The original @samp{.word}
7411 contains @code{sym1} minus the address of the long-jump to
7414 If there were several occurrences of @samp{.word sym1-sym2} before the
7415 secondary jump table, all of them are adjusted. If there was a
7416 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7417 long-jump to @code{sym4} is included in the secondary jump table,
7418 and the @code{.word} directives are adjusted to contain @code{sym3}
7419 minus the address of the long-jump to @code{sym4}; and so on, for as many
7420 entries in the original jump table as necessary.
7423 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7424 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7425 assembly language programmers.
7428 @c end DIFF-TBL-KLUGE
7430 @ifclear no-space-dir
7432 @section @code{.zero @var{size}}
7434 @cindex @code{zero} directive
7435 @cindex filling memory with zero bytes
7436 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7437 expression. This directive is actually an alias for the @samp{.skip} directive
7438 so it can take an optional second argument of the value to store in the bytes
7439 instead of zero. Using @samp{.zero} in this way would be confusing however.
7444 @section @code{.2byte @var{expression} [, @var{expression}]*}
7445 @cindex @code{2byte} directive
7446 @cindex two-byte integer
7447 @cindex integer, 2-byte
7449 This directive expects zero or more expressions, separated by commas. If there
7450 are no expressions then the directive does nothing. Otherwise each expression
7451 is evaluated in turn and placed in the next two bytes of the current output
7452 section, using the endian model of the target. If an expression will not fit
7453 in two bytes, a warning message is displayed and the least significant two
7454 bytes of the expression's value are used. If an expression cannot be evaluated
7455 at assembly time then relocations will be generated in order to compute the
7458 This directive does not apply any alignment before or after inserting the
7459 values. As a result of this, if relocations are generated, they may be
7460 different from those used for inserting values with a guaranteed alignment.
7462 This directive is only available for ELF targets,
7465 @section @code{.4byte @var{expression} [, @var{expression}]*}
7466 @cindex @code{4byte} directive
7467 @cindex four-byte integer
7468 @cindex integer, 4-byte
7470 Like the @option{.2byte} directive, except that it inserts unaligned, four byte
7471 long values into the output.
7474 @section @code{.8byte @var{expression} [, @var{expression}]*}
7475 @cindex @code{8byte} directive
7476 @cindex eight-byte integer
7477 @cindex integer, 8-byte
7479 Like the @option{.2byte} directive, except that it inserts unaligned, eight
7480 byte long bignum values into the output.
7485 @section Deprecated Directives
7487 @cindex deprecated directives
7488 @cindex obsolescent directives
7489 One day these directives won't work.
7490 They are included for compatibility with older assemblers.
7497 @node Object Attributes
7498 @chapter Object Attributes
7499 @cindex object attributes
7501 @command{@value{AS}} assembles source files written for a specific architecture
7502 into object files for that architecture. But not all object files are alike.
7503 Many architectures support incompatible variations. For instance, floating
7504 point arguments might be passed in floating point registers if the object file
7505 requires hardware floating point support---or floating point arguments might be
7506 passed in integer registers if the object file supports processors with no
7507 hardware floating point unit. Or, if two objects are built for different
7508 generations of the same architecture, the combination may require the
7509 newer generation at run-time.
7511 This information is useful during and after linking. At link time,
7512 @command{@value{LD}} can warn about incompatible object files. After link
7513 time, tools like @command{gdb} can use it to process the linked file
7516 Compatibility information is recorded as a series of object attributes. Each
7517 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7518 string, and indicates who sets the meaning of the tag. The tag is an integer,
7519 and indicates what property the attribute describes. The value may be a string
7520 or an integer, and indicates how the property affects this object. Missing
7521 attributes are the same as attributes with a zero value or empty string value.
7523 Object attributes were developed as part of the ABI for the ARM Architecture.
7524 The file format is documented in @cite{ELF for the ARM Architecture}.
7527 * GNU Object Attributes:: @sc{gnu} Object Attributes
7528 * Defining New Object Attributes:: Defining New Object Attributes
7531 @node GNU Object Attributes
7532 @section @sc{gnu} Object Attributes
7534 The @code{.gnu_attribute} directive records an object attribute
7535 with vendor @samp{gnu}.
7537 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7538 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7539 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7540 2} is set for architecture-independent attributes and clear for
7541 architecture-dependent ones.
7543 @subsection Common @sc{gnu} attributes
7545 These attributes are valid on all architectures.
7548 @item Tag_compatibility (32)
7549 The compatibility attribute takes an integer flag value and a vendor name. If
7550 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7551 then the file is only compatible with the named toolchain. If it is greater
7552 than 1, the file can only be processed by other toolchains under some private
7553 arrangement indicated by the flag value and the vendor name.
7556 @subsection MIPS Attributes
7559 @item Tag_GNU_MIPS_ABI_FP (4)
7560 The floating-point ABI used by this object file. The value will be:
7564 0 for files not affected by the floating-point ABI.
7566 1 for files using the hardware floating-point ABI with a standard
7567 double-precision FPU.
7569 2 for files using the hardware floating-point ABI with a single-precision FPU.
7571 3 for files using the software floating-point ABI.
7573 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7574 floating-point registers, 32-bit general-purpose registers and increased the
7575 number of callee-saved floating-point registers.
7577 5 for files using the hardware floating-point ABI with a double-precision FPU
7578 with either 32-bit or 64-bit floating-point registers and 32-bit
7579 general-purpose registers.
7581 6 for files using the hardware floating-point ABI with 64-bit floating-point
7582 registers and 32-bit general-purpose registers.
7584 7 for files using the hardware floating-point ABI with 64-bit floating-point
7585 registers, 32-bit general-purpose registers and a rule that forbids the
7586 direct use of odd-numbered single-precision floating-point registers.
7590 @subsection PowerPC Attributes
7593 @item Tag_GNU_Power_ABI_FP (4)
7594 The floating-point ABI used by this object file. The value will be:
7598 0 for files not affected by the floating-point ABI.
7600 1 for files using double-precision hardware floating-point ABI.
7602 2 for files using the software floating-point ABI.
7604 3 for files using single-precision hardware floating-point ABI.
7607 @item Tag_GNU_Power_ABI_Vector (8)
7608 The vector ABI used by this object file. The value will be:
7612 0 for files not affected by the vector ABI.
7614 1 for files using general purpose registers to pass vectors.
7616 2 for files using AltiVec registers to pass vectors.
7618 3 for files using SPE registers to pass vectors.
7622 @subsection IBM z Systems Attributes
7625 @item Tag_GNU_S390_ABI_Vector (8)
7626 The vector ABI used by this object file. The value will be:
7630 0 for files not affected by the vector ABI.
7632 1 for files using software vector ABI.
7634 2 for files using hardware vector ABI.
7638 @subsection MSP430 Attributes
7641 @item Tag_GNU_MSP430_Data_Region (4)
7642 The data region used by this object file. The value will be:
7646 0 for files not using the large memory model.
7648 1 for files which have been compiled with the condition that all
7649 data is in the lower memory region, i.e. below address 0x10000.
7651 2 for files which allow data to be placed in the full 20-bit memory range.
7655 @node Defining New Object Attributes
7656 @section Defining New Object Attributes
7658 If you want to define a new @sc{gnu} object attribute, here are the places you
7659 will need to modify. New attributes should be discussed on the @samp{binutils}
7664 This manual, which is the official register of attributes.
7666 The header for your architecture @file{include/elf}, to define the tag.
7668 The @file{bfd} support file for your architecture, to merge the attribute
7669 and issue any appropriate link warnings.
7671 Test cases in @file{ld/testsuite} for merging and link warnings.
7673 @file{binutils/readelf.c} to display your attribute.
7675 GCC, if you want the compiler to mark the attribute automatically.
7681 @node Machine Dependencies
7682 @chapter Machine Dependent Features
7684 @cindex machine dependencies
7685 The machine instruction sets are (almost by definition) different on
7686 each machine where @command{@value{AS}} runs. Floating point representations
7687 vary as well, and @command{@value{AS}} often supports a few additional
7688 directives or command-line options for compatibility with other
7689 assemblers on a particular platform. Finally, some versions of
7690 @command{@value{AS}} support special pseudo-instructions for branch
7693 This chapter discusses most of these differences, though it does not
7694 include details on any machine's instruction set. For details on that
7695 subject, see the hardware manufacturer's manual.
7699 * AArch64-Dependent:: AArch64 Dependent Features
7702 * Alpha-Dependent:: Alpha Dependent Features
7705 * ARC-Dependent:: ARC Dependent Features
7708 * ARM-Dependent:: ARM Dependent Features
7711 * AVR-Dependent:: AVR Dependent Features
7714 * Blackfin-Dependent:: Blackfin Dependent Features
7717 * BPF-Dependent:: BPF Dependent Features
7720 * CR16-Dependent:: CR16 Dependent Features
7723 * CRIS-Dependent:: CRIS Dependent Features
7726 * C-SKY-Dependent:: C-SKY Dependent Features
7729 * D10V-Dependent:: D10V Dependent Features
7732 * D30V-Dependent:: D30V Dependent Features
7735 * Epiphany-Dependent:: EPIPHANY Dependent Features
7738 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7741 * HPPA-Dependent:: HPPA Dependent Features
7744 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7747 * IA-64-Dependent:: Intel IA-64 Dependent Features
7750 * IP2K-Dependent:: IP2K Dependent Features
7753 * LM32-Dependent:: LM32 Dependent Features
7756 * M32C-Dependent:: M32C Dependent Features
7759 * M32R-Dependent:: M32R Dependent Features
7762 * M68K-Dependent:: M680x0 Dependent Features
7765 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7768 * S12Z-Dependent:: S12Z Dependent Features
7771 * Meta-Dependent :: Meta Dependent Features
7774 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7777 * MIPS-Dependent:: MIPS Dependent Features
7780 * MMIX-Dependent:: MMIX Dependent Features
7783 * MSP430-Dependent:: MSP430 Dependent Features
7786 * NDS32-Dependent:: Andes NDS32 Dependent Features
7789 * NiosII-Dependent:: Altera Nios II Dependent Features
7792 * NS32K-Dependent:: NS32K Dependent Features
7795 * OpenRISC-Dependent:: OpenRISC 1000 Features
7798 * PDP-11-Dependent:: PDP-11 Dependent Features
7801 * PJ-Dependent:: picoJava Dependent Features
7804 * PPC-Dependent:: PowerPC Dependent Features
7807 * PRU-Dependent:: PRU Dependent Features
7810 * RISC-V-Dependent:: RISC-V Dependent Features
7813 * RL78-Dependent:: RL78 Dependent Features
7816 * RX-Dependent:: RX Dependent Features
7819 * S/390-Dependent:: IBM S/390 Dependent Features
7822 * SCORE-Dependent:: SCORE Dependent Features
7825 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7828 * Sparc-Dependent:: SPARC Dependent Features
7831 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7834 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7837 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7840 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7843 * V850-Dependent:: V850 Dependent Features
7846 * Vax-Dependent:: VAX Dependent Features
7849 * Visium-Dependent:: Visium Dependent Features
7852 * WebAssembly-Dependent:: WebAssembly Dependent Features
7855 * XGATE-Dependent:: XGATE Dependent Features
7858 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7861 * Xtensa-Dependent:: Xtensa Dependent Features
7864 * Z80-Dependent:: Z80 Dependent Features
7867 * Z8000-Dependent:: Z8000 Dependent Features
7874 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7875 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7876 @c peculiarity: to preserve cross-references, there must be a node called
7877 @c "Machine Dependencies". Hence the conditional nodenames in each
7878 @c major node below. Node defaulting in makeinfo requires adjacency of
7879 @c node and sectioning commands; hence the repetition of @chapter BLAH
7880 @c in both conditional blocks.
7883 @include c-aarch64.texi
7887 @include c-alpha.texi
7903 @include c-bfin.texi
7911 @include c-cr16.texi
7915 @include c-cris.texi
7919 @include c-csky.texi
7924 @node Machine Dependencies
7925 @chapter Machine Dependent Features
7927 The machine instruction sets are different on each Renesas chip family,
7928 and there are also some syntax differences among the families. This
7929 chapter describes the specific @command{@value{AS}} features for each
7933 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7934 * SH-Dependent:: Renesas SH Dependent Features
7941 @include c-d10v.texi
7945 @include c-d30v.texi
7949 @include c-epiphany.texi
7953 @include c-h8300.texi
7957 @include c-hppa.texi
7961 @include c-i386.texi
7965 @include c-ia64.texi
7969 @include c-ip2k.texi
7973 @include c-lm32.texi
7977 @include c-m32c.texi
7981 @include c-m32r.texi
7985 @include c-m68k.texi
7989 @include c-m68hc11.texi
7993 @include c-s12z.texi
7997 @include c-metag.texi
8001 @include c-microblaze.texi
8005 @include c-mips.texi
8009 @include c-mmix.texi
8013 @include c-msp430.texi
8017 @include c-nds32.texi
8021 @include c-nios2.texi
8025 @include c-ns32k.texi
8029 @include c-or1k.texi
8033 @include c-pdp11.texi
8049 @include c-riscv.texi
8053 @include c-rl78.texi
8061 @include c-s390.texi
8065 @include c-score.texi
8073 @include c-sparc.texi
8077 @include c-tic54x.texi
8081 @include c-tic6x.texi
8085 @include c-tilegx.texi
8089 @include c-tilepro.texi
8093 @include c-v850.texi
8101 @include c-visium.texi
8105 @include c-wasm32.texi
8109 @include c-xgate.texi
8113 @include c-xstormy16.texi
8117 @include c-xtensa.texi
8129 @c reverse effect of @down at top of generic Machine-Dep chapter
8133 @node Reporting Bugs
8134 @chapter Reporting Bugs
8135 @cindex bugs in assembler
8136 @cindex reporting bugs in assembler
8138 Your bug reports play an essential role in making @command{@value{AS}} reliable.
8140 Reporting a bug may help you by bringing a solution to your problem, or it may
8141 not. But in any case the principal function of a bug report is to help the
8142 entire community by making the next version of @command{@value{AS}} work better.
8143 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
8145 In order for a bug report to serve its purpose, you must include the
8146 information that enables us to fix the bug.
8149 * Bug Criteria:: Have you found a bug?
8150 * Bug Reporting:: How to report bugs
8154 @section Have You Found a Bug?
8155 @cindex bug criteria
8157 If you are not sure whether you have found a bug, here are some guidelines:
8160 @cindex fatal signal
8161 @cindex assembler crash
8162 @cindex crash of assembler
8164 If the assembler gets a fatal signal, for any input whatever, that is a
8165 @command{@value{AS}} bug. Reliable assemblers never crash.
8167 @cindex error on valid input
8169 If @command{@value{AS}} produces an error message for valid input, that is a bug.
8171 @cindex invalid input
8173 If @command{@value{AS}} does not produce an error message for invalid input, that
8174 is a bug. However, you should note that your idea of ``invalid input'' might
8175 be our idea of ``an extension'' or ``support for traditional practice''.
8178 If you are an experienced user of assemblers, your suggestions for improvement
8179 of @command{@value{AS}} are welcome in any case.
8183 @section How to Report Bugs
8185 @cindex assembler bugs, reporting
8187 A number of companies and individuals offer support for @sc{gnu} products. If
8188 you obtained @command{@value{AS}} from a support organization, we recommend you
8189 contact that organization first.
8191 You can find contact information for many support companies and
8192 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
8196 In any event, we also recommend that you send bug reports for @command{@value{AS}}
8200 The fundamental principle of reporting bugs usefully is this:
8201 @strong{report all the facts}. If you are not sure whether to state a
8202 fact or leave it out, state it!
8204 Often people omit facts because they think they know what causes the problem
8205 and assume that some details do not matter. Thus, you might assume that the
8206 name of a symbol you use in an example does not matter. Well, probably it does
8207 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
8208 happens to fetch from the location where that name is stored in memory;
8209 perhaps, if the name were different, the contents of that location would fool
8210 the assembler into doing the right thing despite the bug. Play it safe and
8211 give a specific, complete example. That is the easiest thing for you to do,
8212 and the most helpful.
8214 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
8215 it is new to us. Therefore, always write your bug reports on the assumption
8216 that the bug has not been reported previously.
8218 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8219 bell?'' This cannot help us fix a bug, so it is basically useless. We
8220 respond by asking for enough details to enable us to investigate.
8221 You might as well expedite matters by sending them to begin with.
8223 To enable us to fix the bug, you should include all these things:
8227 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8228 it with the @samp{--version} argument.
8230 Without this, we will not know whether there is any point in looking for
8231 the bug in the current version of @command{@value{AS}}.
8234 Any patches you may have applied to the @command{@value{AS}} source.
8237 The type of machine you are using, and the operating system name and
8241 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8245 The command arguments you gave the assembler to assemble your example and
8246 observe the bug. To guarantee you will not omit something important, list them
8247 all. A copy of the Makefile (or the output from make) is sufficient.
8249 If we were to try to guess the arguments, we would probably guess wrong
8250 and then we might not encounter the bug.
8253 A complete input file that will reproduce the bug. If the bug is observed when
8254 the assembler is invoked via a compiler, send the assembler source, not the
8255 high level language source. Most compilers will produce the assembler source
8256 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8257 the options @samp{-v --save-temps}; this will save the assembler source in a
8258 file with an extension of @file{.s}, and also show you exactly how
8259 @command{@value{AS}} is being run.
8262 A description of what behavior you observe that you believe is
8263 incorrect. For example, ``It gets a fatal signal.''
8265 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8266 will certainly notice it. But if the bug is incorrect output, we might not
8267 notice unless it is glaringly wrong. You might as well not give us a chance to
8270 Even if the problem you experience is a fatal signal, you should still say so
8271 explicitly. Suppose something strange is going on, such as, your copy of
8272 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8273 library on your system. (This has happened!) Your copy might crash and ours
8274 would not. If you told us to expect a crash, then when ours fails to crash, we
8275 would know that the bug was not happening for us. If you had not told us to
8276 expect a crash, then we would not be able to draw any conclusion from our
8280 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8281 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8282 option. Always send diffs from the old file to the new file. If you even
8283 discuss something in the @command{@value{AS}} source, refer to it by context, not
8286 The line numbers in our development sources will not match those in your
8287 sources. Your line numbers would convey no useful information to us.
8290 Here are some things that are not necessary:
8294 A description of the envelope of the bug.
8296 Often people who encounter a bug spend a lot of time investigating
8297 which changes to the input file will make the bug go away and which
8298 changes will not affect it.
8300 This is often time consuming and not very useful, because the way we
8301 will find the bug is by running a single example under the debugger
8302 with breakpoints, not by pure deduction from a series of examples.
8303 We recommend that you save your time for something else.
8305 Of course, if you can find a simpler example to report @emph{instead}
8306 of the original one, that is a convenience for us. Errors in the
8307 output will be easier to spot, running under the debugger will take
8308 less time, and so on.
8310 However, simplification is not vital; if you do not want to do this,
8311 report the bug anyway and send us the entire test case you used.
8314 A patch for the bug.
8316 A patch for the bug does help us if it is a good one. But do not omit
8317 the necessary information, such as the test case, on the assumption that
8318 a patch is all we need. We might see problems with your patch and decide
8319 to fix the problem another way, or we might not understand it at all.
8321 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8322 construct an example that will make the program follow a certain path through
8323 the code. If you do not send us the example, we will not be able to construct
8324 one, so we will not be able to verify that the bug is fixed.
8326 And if we cannot understand what bug you are trying to fix, or why your
8327 patch should be an improvement, we will not install it. A test case will
8328 help us to understand.
8331 A guess about what the bug is or what it depends on.
8333 Such guesses are usually wrong. Even we cannot guess right about such
8334 things without first using the debugger to find the facts.
8337 @node Acknowledgements
8338 @chapter Acknowledgements
8340 If you have contributed to GAS and your name isn't listed here,
8341 it is not meant as a slight. We just don't know about it. Send mail to the
8342 maintainer, and we'll correct the situation. Currently
8344 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8346 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8349 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8350 information and the 68k series machines, most of the preprocessing pass, and
8351 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8353 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8354 many bug fixes, including merging support for several processors, breaking GAS
8355 up to handle multiple object file format back ends (including heavy rewrite,
8356 testing, an integration of the coff and b.out back ends), adding configuration
8357 including heavy testing and verification of cross assemblers and file splits
8358 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8359 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8360 port (including considerable amounts of reverse engineering), a SPARC opcode
8361 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8362 assertions and made them work, much other reorganization, cleanup, and lint.
8364 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8365 in format-specific I/O modules.
8367 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8368 has done much work with it since.
8370 The Intel 80386 machine description was written by Eliot Dresselhaus.
8372 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8374 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8375 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8377 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8378 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8379 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8380 support a.out format.
8382 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8383 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8384 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8385 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8388 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8389 simplified the configuration of which versions accept which directives. He
8390 updated the 68k machine description so that Motorola's opcodes always produced
8391 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8392 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8393 cross-compilation support, and one bug in relaxation that took a week and
8394 required the proverbial one-bit fix.
8396 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8397 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8398 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8399 PowerPC assembler, and made a few other minor patches.
8401 Steve Chamberlain made GAS able to generate listings.
8403 Hewlett-Packard contributed support for the HP9000/300.
8405 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8406 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8407 formats). This work was supported by both the Center for Software Science at
8408 the University of Utah and Cygnus Support.
8410 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8411 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8412 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8413 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8414 and some initial 64-bit support).
8416 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8418 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8419 support for openVMS/Alpha.
8421 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8424 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8425 Inc.@: added support for Xtensa processors.
8427 Several engineers at Cygnus Support have also provided many small bug fixes and
8428 configuration enhancements.
8430 Jon Beniston added support for the Lattice Mico32 architecture.
8432 Many others have contributed large or small bugfixes and enhancements. If
8433 you have contributed significant work and are not mentioned on this list, and
8434 want to be, let us know. Some of the history has been lost; we are not
8435 intentionally leaving anyone out.
8437 @node GNU Free Documentation License
8438 @appendix GNU Free Documentation License
8442 @unnumbered AS Index