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-<N>}] [@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{-march=@var{CPU}@var{[-EXT]}@var{[+EXT]}}]
634 [@b{-local-prefix=}@var{PREFIX}]
637 [@b{-fp-s=}@var{FORMAT}]
638 [@b{-fp-d=}@var{FORMAT}]
642 @c Z8000 has no machine-dependent assembler options
651 @include at-file.texi
654 Turn on listings, in any of a variety of ways:
658 omit false conditionals
661 omit debugging directives
664 include general information, like @value{AS} version and options passed
667 include high-level source
673 include macro expansions
676 omit forms processing
682 set the name of the listing file
685 You may combine these options; for example, use @samp{-aln} for assembly
686 listing without forms processing. The @samp{=file} option, if used, must be
687 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
690 Begin in alternate macro mode.
692 @xref{Altmacro,,@code{.altmacro}}.
695 @item --compress-debug-sections
696 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
697 ELF ABI. The resulting object file may not be compatible with older
698 linkers and object file utilities. Note if compression would make a
699 given section @emph{larger} then it is not compressed.
702 @cindex @samp{--compress-debug-sections=} option
703 @item --compress-debug-sections=none
704 @itemx --compress-debug-sections=zlib
705 @itemx --compress-debug-sections=zlib-gnu
706 @itemx --compress-debug-sections=zlib-gabi
707 These options control how DWARF debug sections are compressed.
708 @option{--compress-debug-sections=none} is equivalent to
709 @option{--nocompress-debug-sections}.
710 @option{--compress-debug-sections=zlib} and
711 @option{--compress-debug-sections=zlib-gabi} are equivalent to
712 @option{--compress-debug-sections}.
713 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
714 sections using zlib. The debug sections are renamed to begin with
715 @samp{.zdebug}. Note if compression would make a given section
716 @emph{larger} then it is not compressed nor renamed.
720 @item --nocompress-debug-sections
721 Do not compress DWARF debug sections. This is usually the default for all
722 targets except the x86/x86_64, but a configure time option can be used to
726 Ignored. This option is accepted for script compatibility with calls to
729 @item --debug-prefix-map @var{old}=@var{new}
730 When assembling files in directory @file{@var{old}}, record debugging
731 information describing them as in @file{@var{new}} instead.
733 @item --defsym @var{sym}=@var{value}
734 Define the symbol @var{sym} to be @var{value} before assembling the input file.
735 @var{value} must be an integer constant. As in C, a leading @samp{0x}
736 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
737 value. The value of the symbol can be overridden inside a source file via the
738 use of a @code{.set} pseudo-op.
741 ``fast''---skip whitespace and comment preprocessing (assume source is
746 Generate debugging information for each assembler source line using whichever
747 debug format is preferred by the target. This currently means either STABS,
751 Generate stabs debugging information for each assembler line. This
752 may help debugging assembler code, if the debugger can handle it.
755 Generate stabs debugging information for each assembler line, with GNU
756 extensions that probably only gdb can handle, and that could make other
757 debuggers crash or refuse to read your program. This
758 may help debugging assembler code. Currently the only GNU extension is
759 the location of the current working directory at assembling time.
762 Generate DWARF2 debugging information for each assembler line. This
763 may help debugging assembler code, if the debugger can handle it. Note---this
764 option is only supported by some targets, not all of them.
767 This option is the same as the @option{--gdwarf-2} option, except that it
768 allows for the possibility of the generation of extra debug information as per
769 version 3 of the DWARF specification. Note - enabling this option does not
770 guarantee the generation of any extra infortmation, the choice to do so is on a
774 This option is the same as the @option{--gdwarf-2} option, except that it
775 allows for the possibility of the generation of extra debug information as per
776 version 4 of the DWARF specification. Note - enabling this option does not
777 guarantee the generation of any extra infortmation, the choice to do so is on a
781 This option is the same as the @option{--gdwarf-2} option, except that it
782 allows for the possibility of the generation of extra debug information as per
783 version 5 of the DWARF specification. Note - enabling this option does not
784 guarantee the generation of any extra infortmation, the choice to do so is on a
787 @item --gdwarf-sections
788 Instead of creating a .debug_line section, create a series of
789 .debug_line.@var{foo} sections where @var{foo} is the name of the
790 corresponding code section. For example a code section called @var{.text.func}
791 will have its dwarf line number information placed into a section called
792 @var{.debug_line.text.func}. If the code section is just called @var{.text}
793 then debug line section will still be called just @var{.debug_line} without any
796 @item --gdwarf-cie-version=@var{version}
797 Control which version of DWARF Common Information Entries (CIEs) are produced.
798 When this flag is not specificed the default is version 1, though some targets
799 can modify this default. Other possible values for @var{version} are 3 or 4.
802 @item --size-check=error
803 @itemx --size-check=warning
804 Issue an error or warning for invalid ELF .size directive.
806 @item --elf-stt-common=no
807 @itemx --elf-stt-common=yes
808 These options control whether the ELF assembler should generate common
809 symbols with the @code{STT_COMMON} type. The default can be controlled
810 by a configure option @option{--enable-elf-stt-common}.
812 @item --generate-missing-build-notes=yes
813 @itemx --generate-missing-build-notes=no
814 These options control whether the ELF assembler should generate GNU Build
815 attribute notes if none are present in the input sources.
816 The default can be controlled by the @option{--enable-generate-build-notes}
822 Print a summary of the command-line options and exit.
825 Print a summary of all target specific options and exit.
828 Add directory @var{dir} to the search list for @code{.include} directives.
831 Don't warn about signed overflow.
834 @ifclear DIFF-TBL-KLUGE
835 This option is accepted but has no effect on the @value{TARGET} family.
837 @ifset DIFF-TBL-KLUGE
838 Issue warnings when difference tables altered for long displacements.
843 Keep (in the symbol table) local symbols. These symbols start with
844 system-specific local label prefixes, typically @samp{.L} for ELF systems
845 or @samp{L} for traditional a.out systems.
850 @item --listing-lhs-width=@var{number}
851 Set the maximum width, in words, of the output data column for an assembler
852 listing to @var{number}.
854 @item --listing-lhs-width2=@var{number}
855 Set the maximum width, in words, of the output data column for continuation
856 lines in an assembler listing to @var{number}.
858 @item --listing-rhs-width=@var{number}
859 Set the maximum width of an input source line, as displayed in a listing, to
862 @item --listing-cont-lines=@var{number}
863 Set the maximum number of lines printed in a listing for a single line of input
866 @item --no-pad-sections
867 Stop the assembler for padding the ends of output sections to the alignment
868 of that section. The default is to pad the sections, but this can waste space
869 which might be needed on targets which have tight memory constraints.
871 @item -o @var{objfile}
872 Name the object-file output from @command{@value{AS}} @var{objfile}.
875 Fold the data section into the text section.
877 @item --hash-size=@var{number}
878 Set the default size of GAS's hash tables to a prime number close to
879 @var{number}. Increasing this value can reduce the length of time it takes the
880 assembler to perform its tasks, at the expense of increasing the assembler's
881 memory requirements. Similarly reducing this value can reduce the memory
882 requirements at the expense of speed.
884 @item --reduce-memory-overheads
885 This option reduces GAS's memory requirements, at the expense of making the
886 assembly processes slower. Currently this switch is a synonym for
887 @samp{--hash-size=4051}, but in the future it may have other effects as well.
890 @item --sectname-subst
891 Honor substitution sequences in section names.
893 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
898 Print the maximum space (in bytes) and total time (in seconds) used by
901 @item --strip-local-absolute
902 Remove local absolute symbols from the outgoing symbol table.
906 Print the @command{as} version.
909 Print the @command{as} version and exit.
913 Suppress warning messages.
915 @item --fatal-warnings
916 Treat warnings as errors.
919 Don't suppress warning messages or treat them as errors.
928 Generate an object file even after errors.
930 @item -- | @var{files} @dots{}
931 Standard input, or source files to assemble.
939 @xref{AArch64 Options}, for the options available when @value{AS} is configured
940 for the 64-bit mode of the ARM Architecture (AArch64).
945 The following options are available when @value{AS} is configured for the
946 64-bit mode of the ARM Architecture (AArch64).
949 @include c-aarch64.texi
950 @c ended inside the included file
958 @xref{Alpha Options}, for the options available when @value{AS} is configured
959 for an Alpha processor.
964 The following options are available when @value{AS} is configured for an Alpha
968 @include c-alpha.texi
969 @c ended inside the included file
976 The following options are available when @value{AS} is configured for an ARC
980 @item -mcpu=@var{cpu}
981 This option selects the core processor variant.
983 Select either big-endian (-EB) or little-endian (-EL) output.
985 Enable Code Density extenssion instructions.
990 The following options are available when @value{AS} is configured for the ARM
994 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
995 Specify which ARM processor variant is the target.
996 @item -march=@var{architecture}[+@var{extension}@dots{}]
997 Specify which ARM architecture variant is used by the target.
998 @item -mfpu=@var{floating-point-format}
999 Select which Floating Point architecture is the target.
1000 @item -mfloat-abi=@var{abi}
1001 Select which floating point ABI is in use.
1003 Enable Thumb only instruction decoding.
1004 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
1005 Select which procedure calling convention is in use.
1007 Select either big-endian (-EB) or little-endian (-EL) output.
1008 @item -mthumb-interwork
1009 Specify that the code has been generated with interworking between Thumb and
1012 Turns on CodeComposer Studio assembly syntax compatibility mode.
1014 Specify that PIC code has been generated.
1022 @xref{Blackfin Options}, for the options available when @value{AS} is
1023 configured for the Blackfin processor family.
1027 @c man begin OPTIONS
1028 The following options are available when @value{AS} is configured for
1029 the Blackfin processor family.
1031 @c man begin INCLUDE
1032 @include c-bfin.texi
1033 @c ended inside the included file
1041 @xref{BPF Options}, for the options available when @value{AS} is
1042 configured for the Linux kernel BPF processor family.
1046 @c man begin OPTIONS
1047 The following options are available when @value{AS} is configured for
1048 the Linux kernel BPF processor family.
1050 @c man begin INCLUDE
1052 @c ended inside the included file
1057 @c man begin OPTIONS
1059 See the info pages for documentation of the CRIS-specific options.
1065 @xref{C-SKY Options}, for the options available when @value{AS} is
1066 configured for the C-SKY processor family.
1070 @c man begin OPTIONS
1071 The following options are available when @value{AS} is configured for
1072 the C-SKY processor family.
1074 @c man begin INCLUDE
1075 @include c-csky.texi
1076 @c ended inside the included file
1082 The following options are available when @value{AS} is configured for
1085 @cindex D10V optimization
1086 @cindex optimization, D10V
1088 Optimize output by parallelizing instructions.
1093 The following options are available when @value{AS} is configured for a D30V
1096 @cindex D30V optimization
1097 @cindex optimization, D30V
1099 Optimize output by parallelizing instructions.
1103 Warn when nops are generated.
1105 @cindex D30V nops after 32-bit multiply
1107 Warn when a nop after a 32-bit multiply instruction is generated.
1113 The following options are available when @value{AS} is configured for the
1114 Adapteva EPIPHANY series.
1117 @xref{Epiphany Options}, for the options available when @value{AS} is
1118 configured for an Epiphany processor.
1122 @c man begin OPTIONS
1123 The following options are available when @value{AS} is configured for
1124 an Epiphany processor.
1126 @c man begin INCLUDE
1127 @include c-epiphany.texi
1128 @c ended inside the included file
1136 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1137 for an H8/300 processor.
1141 @c man begin OPTIONS
1142 The following options are available when @value{AS} is configured for an H8/300
1145 @c man begin INCLUDE
1146 @include c-h8300.texi
1147 @c ended inside the included file
1155 @xref{i386-Options}, for the options available when @value{AS} is
1156 configured for an i386 processor.
1160 @c man begin OPTIONS
1161 The following options are available when @value{AS} is configured for
1164 @c man begin INCLUDE
1165 @include c-i386.texi
1166 @c ended inside the included file
1171 @c man begin OPTIONS
1173 The following options are available when @value{AS} is configured for the
1179 Specifies that the extended IP2022 instructions are allowed.
1182 Restores the default behaviour, which restricts the permitted instructions to
1183 just the basic IP2022 ones.
1189 The following options are available when @value{AS} is configured for the
1190 Renesas M32C and M16C processors.
1195 Assemble M32C instructions.
1198 Assemble M16C instructions (the default).
1201 Enable support for link-time relaxations.
1204 Support H'00 style hex constants in addition to 0x00 style.
1210 The following options are available when @value{AS} is configured for the
1211 Renesas M32R (formerly Mitsubishi M32R) series.
1216 Specify which processor in the M32R family is the target. The default
1217 is normally the M32R, but this option changes it to the M32RX.
1219 @item --warn-explicit-parallel-conflicts or --Wp
1220 Produce warning messages when questionable parallel constructs are
1223 @item --no-warn-explicit-parallel-conflicts or --Wnp
1224 Do not produce warning messages when questionable parallel constructs are
1231 The following options are available when @value{AS} is configured for the
1232 Motorola 68000 series.
1237 Shorten references to undefined symbols, to one word instead of two.
1239 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1240 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1241 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1242 Specify what processor in the 68000 family is the target. The default
1243 is normally the 68020, but this can be changed at configuration time.
1245 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1246 The target machine does (or does not) have a floating-point coprocessor.
1247 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1248 the basic 68000 is not compatible with the 68881, a combination of the
1249 two can be specified, since it's possible to do emulation of the
1250 coprocessor instructions with the main processor.
1252 @item -m68851 | -mno-68851
1253 The target machine does (or does not) have a memory-management
1254 unit coprocessor. The default is to assume an MMU for 68020 and up.
1262 @xref{Nios II Options}, for the options available when @value{AS} is configured
1263 for an Altera Nios II processor.
1267 @c man begin OPTIONS
1268 The following options are available when @value{AS} is configured for an
1269 Altera Nios II processor.
1271 @c man begin INCLUDE
1272 @include c-nios2.texi
1273 @c ended inside the included file
1279 For details about the PDP-11 machine dependent features options,
1280 see @ref{PDP-11-Options}.
1283 @item -mpic | -mno-pic
1284 Generate position-independent (or position-dependent) code. The
1285 default is @option{-mpic}.
1288 @itemx -mall-extensions
1289 Enable all instruction set extensions. This is the default.
1291 @item -mno-extensions
1292 Disable all instruction set extensions.
1294 @item -m@var{extension} | -mno-@var{extension}
1295 Enable (or disable) a particular instruction set extension.
1298 Enable the instruction set extensions supported by a particular CPU, and
1299 disable all other extensions.
1301 @item -m@var{machine}
1302 Enable the instruction set extensions supported by a particular machine
1303 model, and disable all other extensions.
1309 The following options are available when @value{AS} is configured for
1310 a picoJava processor.
1314 @cindex PJ endianness
1315 @cindex endianness, PJ
1316 @cindex big endian output, PJ
1318 Generate ``big endian'' format output.
1320 @cindex little endian output, PJ
1322 Generate ``little endian'' format output.
1330 @xref{PRU Options}, for the options available when @value{AS} is configured
1331 for a PRU processor.
1335 @c man begin OPTIONS
1336 The following options are available when @value{AS} is configured for a
1339 @c man begin INCLUDE
1341 @c ended inside the included file
1346 The following options are available when @value{AS} is configured for the
1347 Motorola 68HC11 or 68HC12 series.
1351 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1352 Specify what processor is the target. The default is
1353 defined by the configuration option when building the assembler.
1355 @item --xgate-ramoffset
1356 Instruct the linker to offset RAM addresses from S12X address space into
1357 XGATE address space.
1360 Specify to use the 16-bit integer ABI.
1363 Specify to use the 32-bit integer ABI.
1365 @item -mshort-double
1366 Specify to use the 32-bit double ABI.
1369 Specify to use the 64-bit double ABI.
1371 @item --force-long-branches
1372 Relative branches are turned into absolute ones. This concerns
1373 conditional branches, unconditional branches and branches to a
1376 @item -S | --short-branches
1377 Do not turn relative branches into absolute ones
1378 when the offset is out of range.
1380 @item --strict-direct-mode
1381 Do not turn the direct addressing mode into extended addressing mode
1382 when the instruction does not support direct addressing mode.
1384 @item --print-insn-syntax
1385 Print the syntax of instruction in case of error.
1387 @item --print-opcodes
1388 Print the list of instructions with syntax and then exit.
1390 @item --generate-example
1391 Print an example of instruction for each possible instruction and then exit.
1392 This option is only useful for testing @command{@value{AS}}.
1398 The following options are available when @command{@value{AS}} is configured
1399 for the SPARC architecture:
1402 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1403 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1404 Explicitly select a variant of the SPARC architecture.
1406 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1407 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1409 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1410 UltraSPARC extensions.
1412 @item -xarch=v8plus | -xarch=v8plusa
1413 For compatibility with the Solaris v9 assembler. These options are
1414 equivalent to -Av8plus and -Av8plusa, respectively.
1417 Warn when the assembler switches to another architecture.
1422 The following options are available when @value{AS} is configured for the 'c54x
1427 Enable extended addressing mode. All addresses and relocations will assume
1428 extended addressing (usually 23 bits).
1429 @item -mcpu=@var{CPU_VERSION}
1430 Sets the CPU version being compiled for.
1431 @item -merrors-to-file @var{FILENAME}
1432 Redirect error output to a file, for broken systems which don't support such
1433 behaviour in the shell.
1438 @c man begin OPTIONS
1439 The following options are available when @value{AS} is configured for
1444 This option sets the largest size of an object that can be referenced
1445 implicitly with the @code{gp} register. It is only accepted for targets that
1446 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1448 @cindex MIPS endianness
1449 @cindex endianness, MIPS
1450 @cindex big endian output, MIPS
1452 Generate ``big endian'' format output.
1454 @cindex little endian output, MIPS
1456 Generate ``little endian'' format output.
1474 Generate code for a particular MIPS Instruction Set Architecture level.
1475 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1476 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1477 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1478 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1479 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1480 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1481 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1482 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1483 MIPS64 Release 6 ISA processors, respectively.
1485 @item -march=@var{cpu}
1486 Generate code for a particular MIPS CPU.
1488 @item -mtune=@var{cpu}
1489 Schedule and tune for a particular MIPS CPU.
1493 Cause nops to be inserted if the read of the destination register
1494 of an mfhi or mflo instruction occurs in the following two instructions.
1497 @itemx -mno-fix-rm7000
1498 Cause nops to be inserted if a dmult or dmultu instruction is
1499 followed by a load instruction.
1502 @itemx -mno-fix-r5900
1503 Do not attempt to schedule the preceding instruction into the delay slot
1504 of a branch instruction placed at the end of a short loop of six
1505 instructions or fewer and always schedule a @code{nop} instruction there
1506 instead. The short loop bug under certain conditions causes loops to
1507 execute only once or twice, due to a hardware bug in the R5900 chip.
1511 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1512 section instead of the standard ELF .stabs sections.
1516 Control generation of @code{.pdr} sections.
1520 The register sizes are normally inferred from the ISA and ABI, but these
1521 flags force a certain group of registers to be treated as 32 bits wide at
1522 all times. @samp{-mgp32} controls the size of general-purpose registers
1523 and @samp{-mfp32} controls the size of floating-point registers.
1527 The register sizes are normally inferred from the ISA and ABI, but these
1528 flags force a certain group of registers to be treated as 64 bits wide at
1529 all times. @samp{-mgp64} controls the size of general-purpose registers
1530 and @samp{-mfp64} controls the size of floating-point registers.
1533 The register sizes are normally inferred from the ISA and ABI, but using
1534 this flag in combination with @samp{-mabi=32} enables an ABI variant
1535 which will operate correctly with floating-point registers which are
1539 @itemx -mno-odd-spreg
1540 Enable use of floating-point operations on odd-numbered single-precision
1541 registers when supported by the ISA. @samp{-mfpxx} implies
1542 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1546 Generate code for the MIPS 16 processor. This is equivalent to putting
1547 @code{.module mips16} at the start of the assembly file. @samp{-no-mips16}
1548 turns off this option.
1551 @itemx -mno-mips16e2
1552 Enable the use of MIPS16e2 instructions in MIPS16 mode. This is equivalent
1553 to putting @code{.module mips16e2} at the start of the assembly file.
1554 @samp{-mno-mips16e2} turns off this option.
1557 @itemx -mno-micromips
1558 Generate code for the microMIPS processor. This is equivalent to putting
1559 @code{.module micromips} at the start of the assembly file.
1560 @samp{-mno-micromips} turns off this option. This is equivalent to putting
1561 @code{.module nomicromips} at the start of the assembly file.
1564 @itemx -mno-smartmips
1565 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1566 equivalent to putting @code{.module smartmips} at the start of the assembly
1567 file. @samp{-mno-smartmips} turns off this option.
1571 Generate code for the MIPS-3D Application Specific Extension.
1572 This tells the assembler to accept MIPS-3D instructions.
1573 @samp{-no-mips3d} turns off this option.
1577 Generate code for the MDMX Application Specific Extension.
1578 This tells the assembler to accept MDMX instructions.
1579 @samp{-no-mdmx} turns off this option.
1583 Generate code for the DSP Release 1 Application Specific Extension.
1584 This tells the assembler to accept DSP Release 1 instructions.
1585 @samp{-mno-dsp} turns off this option.
1589 Generate code for the DSP Release 2 Application Specific Extension.
1590 This option implies @samp{-mdsp}.
1591 This tells the assembler to accept DSP Release 2 instructions.
1592 @samp{-mno-dspr2} turns off this option.
1596 Generate code for the DSP Release 3 Application Specific Extension.
1597 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1598 This tells the assembler to accept DSP Release 3 instructions.
1599 @samp{-mno-dspr3} turns off this option.
1603 Generate code for the MIPS SIMD Architecture Extension.
1604 This tells the assembler to accept MSA instructions.
1605 @samp{-mno-msa} turns off this option.
1609 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1610 This tells the assembler to accept XPA instructions.
1611 @samp{-mno-xpa} turns off this option.
1615 Generate code for the MT Application Specific Extension.
1616 This tells the assembler to accept MT instructions.
1617 @samp{-mno-mt} turns off this option.
1621 Generate code for the MCU Application Specific Extension.
1622 This tells the assembler to accept MCU instructions.
1623 @samp{-mno-mcu} turns off this option.
1627 Generate code for the MIPS cyclic redundancy check (CRC) Application
1628 Specific Extension. This tells the assembler to accept CRC instructions.
1629 @samp{-mno-crc} turns off this option.
1633 Generate code for the Global INValidate (GINV) Application Specific
1634 Extension. This tells the assembler to accept GINV instructions.
1635 @samp{-mno-ginv} turns off this option.
1637 @item -mloongson-mmi
1638 @itemx -mno-loongson-mmi
1639 Generate code for the Loongson MultiMedia extensions Instructions (MMI)
1640 Application Specific Extension. This tells the assembler to accept MMI
1642 @samp{-mno-loongson-mmi} turns off this option.
1644 @item -mloongson-cam
1645 @itemx -mno-loongson-cam
1646 Generate code for the Loongson Content Address Memory (CAM) instructions.
1647 This tells the assembler to accept Loongson CAM instructions.
1648 @samp{-mno-loongson-cam} turns off this option.
1650 @item -mloongson-ext
1651 @itemx -mno-loongson-ext
1652 Generate code for the Loongson EXTensions (EXT) instructions.
1653 This tells the assembler to accept Loongson EXT instructions.
1654 @samp{-mno-loongson-ext} turns off this option.
1656 @item -mloongson-ext2
1657 @itemx -mno-loongson-ext2
1658 Generate code for the Loongson EXTensions R2 (EXT2) instructions.
1659 This option implies @samp{-mloongson-ext}.
1660 This tells the assembler to accept Loongson EXT2 instructions.
1661 @samp{-mno-loongson-ext2} turns off this option.
1665 Only use 32-bit instruction encodings when generating code for the
1666 microMIPS processor. This option inhibits the use of any 16-bit
1667 instructions. This is equivalent to putting @code{.set insn32} at
1668 the start of the assembly file. @samp{-mno-insn32} turns off this
1669 option. This is equivalent to putting @code{.set noinsn32} at the
1670 start of the assembly file. By default @samp{-mno-insn32} is
1671 selected, allowing all instructions to be used.
1673 @item --construct-floats
1674 @itemx --no-construct-floats
1675 The @samp{--no-construct-floats} option disables the construction of
1676 double width floating point constants by loading the two halves of the
1677 value into the two single width floating point registers that make up
1678 the double width register. By default @samp{--construct-floats} is
1679 selected, allowing construction of these floating point constants.
1681 @item --relax-branch
1682 @itemx --no-relax-branch
1683 The @samp{--relax-branch} option enables the relaxation of out-of-range
1684 branches. By default @samp{--no-relax-branch} is selected, causing any
1685 out-of-range branches to produce an error.
1687 @item -mignore-branch-isa
1688 @itemx -mno-ignore-branch-isa
1689 Ignore branch checks for invalid transitions between ISA modes. The
1690 semantics of branches does not provide for an ISA mode switch, so in
1691 most cases the ISA mode a branch has been encoded for has to be the
1692 same as the ISA mode of the branch's target label. Therefore GAS has
1693 checks implemented that verify in branch assembly that the two ISA
1694 modes match. @samp{-mignore-branch-isa} disables these checks. By
1695 default @samp{-mno-ignore-branch-isa} is selected, causing any invalid
1696 branch requiring a transition between ISA modes to produce an error.
1698 @item -mnan=@var{encoding}
1699 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1700 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1703 @item --emulation=@var{name}
1704 This option was formerly used to switch between ELF and ECOFF output
1705 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1706 removed in GAS 2.24, so the option now serves little purpose.
1707 It is retained for backwards compatibility.
1709 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1710 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1711 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1712 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1713 preferred options instead.
1716 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1723 Control how to deal with multiplication overflow and division by zero.
1724 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1725 (and only work for Instruction Set Architecture level 2 and higher);
1726 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1730 When this option is used, @command{@value{AS}} will issue a warning every
1731 time it generates a nop instruction from a macro.
1737 The following options are available when @value{AS} is configured for
1743 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1744 The command-line option @samp{-nojsri2bsr} can be used to disable it.
1748 Enable or disable the silicon filter behaviour. By default this is disabled.
1749 The default can be overridden by the @samp{-sifilter} command-line option.
1752 Alter jump instructions for long displacements.
1754 @item -mcpu=[210|340]
1755 Select the cpu type on the target hardware. This controls which instructions
1759 Assemble for a big endian target.
1762 Assemble for a little endian target.
1771 @xref{Meta Options}, for the options available when @value{AS} is configured
1772 for a Meta processor.
1776 @c man begin OPTIONS
1777 The following options are available when @value{AS} is configured for a
1780 @c man begin INCLUDE
1781 @include c-metag.texi
1782 @c ended inside the included file
1787 @c man begin OPTIONS
1789 See the info pages for documentation of the MMIX-specific options.
1795 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1796 for a NDS32 processor.
1798 @c ended inside the included file
1802 @c man begin OPTIONS
1803 The following options are available when @value{AS} is configured for a
1806 @c man begin INCLUDE
1807 @include c-nds32.texi
1808 @c ended inside the included file
1815 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1816 for a PowerPC processor.
1820 @c man begin OPTIONS
1821 The following options are available when @value{AS} is configured for a
1824 @c man begin INCLUDE
1826 @c ended inside the included file
1834 @xref{RISC-V-Options}, for the options available when @value{AS} is configured
1835 for a RISC-V processor.
1839 @c man begin OPTIONS
1840 The following options are available when @value{AS} is configured for a
1843 @c man begin INCLUDE
1844 @include c-riscv.texi
1845 @c ended inside the included file
1850 @c man begin OPTIONS
1852 See the info pages for documentation of the RX-specific options.
1856 The following options are available when @value{AS} is configured for the s390
1862 Select the word size, either 31/32 bits or 64 bits.
1865 Select the architecture mode, either the Enterprise System
1866 Architecture (esa) or the z/Architecture mode (zarch).
1867 @item -march=@var{processor}
1868 Specify which s390 processor variant is the target, @samp{g5} (or
1869 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1870 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1871 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1872 @samp{z13} (or @samp{arch11}), @samp{z14} (or @samp{arch12}), or @samp{z15}
1875 @itemx -mno-regnames
1876 Allow or disallow symbolic names for registers.
1877 @item -mwarn-areg-zero
1878 Warn whenever the operand for a base or index register has been specified
1879 but evaluates to zero.
1887 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1888 for a TMS320C6000 processor.
1892 @c man begin OPTIONS
1893 The following options are available when @value{AS} is configured for a
1894 TMS320C6000 processor.
1896 @c man begin INCLUDE
1897 @include c-tic6x.texi
1898 @c ended inside the included file
1906 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1907 for a TILE-Gx processor.
1911 @c man begin OPTIONS
1912 The following options are available when @value{AS} is configured for a TILE-Gx
1915 @c man begin INCLUDE
1916 @include c-tilegx.texi
1917 @c ended inside the included file
1925 @xref{Visium Options}, for the options available when @value{AS} is configured
1926 for a Visium processor.
1930 @c man begin OPTIONS
1931 The following option is available when @value{AS} is configured for a Visium
1934 @c man begin INCLUDE
1935 @include c-visium.texi
1936 @c ended inside the included file
1944 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1945 for an Xtensa processor.
1949 @c man begin OPTIONS
1950 The following options are available when @value{AS} is configured for an
1953 @c man begin INCLUDE
1954 @include c-xtensa.texi
1955 @c ended inside the included file
1963 @xref{Z80 Options}, for the options available when @value{AS} is configured
1964 for an Z80 processor.
1968 @c man begin OPTIONS
1969 The following options are available when @value{AS} is configured for an
1972 @c man begin INCLUDE
1974 @c ended inside the included file
1980 * Manual:: Structure of this Manual
1981 * GNU Assembler:: The GNU Assembler
1982 * Object Formats:: Object File Formats
1983 * Command Line:: Command Line
1984 * Input Files:: Input Files
1985 * Object:: Output (Object) File
1986 * Errors:: Error and Warning Messages
1990 @section Structure of this Manual
1992 @cindex manual, structure and purpose
1993 This manual is intended to describe what you need to know to use
1994 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1995 notation for symbols, constants, and expressions; the directives that
1996 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1999 We also cover special features in the @value{TARGET}
2000 configuration of @command{@value{AS}}, including assembler directives.
2003 This manual also describes some of the machine-dependent features of
2004 various flavors of the assembler.
2007 @cindex machine instructions (not covered)
2008 On the other hand, this manual is @emph{not} intended as an introduction
2009 to programming in assembly language---let alone programming in general!
2010 In a similar vein, we make no attempt to introduce the machine
2011 architecture; we do @emph{not} describe the instruction set, standard
2012 mnemonics, registers or addressing modes that are standard to a
2013 particular architecture.
2015 You may want to consult the manufacturer's
2016 machine architecture manual for this information.
2020 For information on the H8/300 machine instruction set, see @cite{H8/300
2021 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
2022 Programming Manual} (Renesas).
2025 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
2026 see @cite{SH-Microcomputer User's Manual} (Renesas) or
2027 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
2028 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
2031 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
2035 @c I think this is premature---doc@cygnus.com, 17jan1991
2037 Throughout this manual, we assume that you are running @dfn{GNU},
2038 the portable operating system from the @dfn{Free Software
2039 Foundation, Inc.}. This restricts our attention to certain kinds of
2040 computer (in particular, the kinds of computers that @sc{gnu} can run on);
2041 once this assumption is granted examples and definitions need less
2044 @command{@value{AS}} is part of a team of programs that turn a high-level
2045 human-readable series of instructions into a low-level
2046 computer-readable series of instructions. Different versions of
2047 @command{@value{AS}} are used for different kinds of computer.
2050 @c There used to be a section "Terminology" here, which defined
2051 @c "contents", "byte", "word", and "long". Defining "word" to any
2052 @c particular size is confusing when the .word directive may generate 16
2053 @c bits on one machine and 32 bits on another; in general, for the user
2054 @c version of this manual, none of these terms seem essential to define.
2055 @c They were used very little even in the former draft of the manual;
2056 @c this draft makes an effort to avoid them (except in names of
2060 @section The GNU Assembler
2062 @c man begin DESCRIPTION
2064 @sc{gnu} @command{as} is really a family of assemblers.
2066 This manual describes @command{@value{AS}}, a member of that family which is
2067 configured for the @value{TARGET} architectures.
2069 If you use (or have used) the @sc{gnu} assembler on one architecture, you
2070 should find a fairly similar environment when you use it on another
2071 architecture. Each version has much in common with the others,
2072 including object file formats, most assembler directives (often called
2073 @dfn{pseudo-ops}) and assembler syntax.@refill
2075 @cindex purpose of @sc{gnu} assembler
2076 @command{@value{AS}} is primarily intended to assemble the output of the
2077 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
2078 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
2079 assemble correctly everything that other assemblers for the same
2080 machine would assemble.
2082 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
2085 @c This remark should appear in generic version of manual; assumption
2086 @c here is that generic version sets M680x0.
2087 This doesn't mean @command{@value{AS}} always uses the same syntax as another
2088 assembler for the same architecture; for example, we know of several
2089 incompatible versions of 680x0 assembly language syntax.
2094 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
2095 program in one pass of the source file. This has a subtle impact on the
2096 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
2098 @node Object Formats
2099 @section Object File Formats
2101 @cindex object file format
2102 The @sc{gnu} assembler can be configured to produce several alternative
2103 object file formats. For the most part, this does not affect how you
2104 write assembly language programs; but directives for debugging symbols
2105 are typically different in different file formats. @xref{Symbol
2106 Attributes,,Symbol Attributes}.
2109 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
2110 @value{OBJ-NAME} format object files.
2112 @c The following should exhaust all configs that set MULTI-OBJ, ideally
2114 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
2115 SOM or ELF format object files.
2120 @section Command Line
2122 @cindex command line conventions
2124 After the program name @command{@value{AS}}, the command line may contain
2125 options and file names. Options may appear in any order, and may be
2126 before, after, or between file names. The order of file names is
2129 @cindex standard input, as input file
2131 @file{--} (two hyphens) by itself names the standard input file
2132 explicitly, as one of the files for @command{@value{AS}} to assemble.
2134 @cindex options, command line
2135 Except for @samp{--} any command-line argument that begins with a
2136 hyphen (@samp{-}) is an option. Each option changes the behavior of
2137 @command{@value{AS}}. No option changes the way another option works. An
2138 option is a @samp{-} followed by one or more letters; the case of
2139 the letter is important. All options are optional.
2141 Some options expect exactly one file name to follow them. The file
2142 name may either immediately follow the option's letter (compatible
2143 with older assemblers) or it may be the next command argument (@sc{gnu}
2144 standard). These two command lines are equivalent:
2147 @value{AS} -o my-object-file.o mumble.s
2148 @value{AS} -omy-object-file.o mumble.s
2152 @section Input Files
2155 @cindex source program
2156 @cindex files, input
2157 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2158 describe the program input to one run of @command{@value{AS}}. The program may
2159 be in one or more files; how the source is partitioned into files
2160 doesn't change the meaning of the source.
2162 @c I added "con" prefix to "catenation" just to prove I can overcome my
2163 @c APL training... doc@cygnus.com
2164 The source program is a concatenation of the text in all the files, in the
2167 @c man begin DESCRIPTION
2168 Each time you run @command{@value{AS}} it assembles exactly one source
2169 program. The source program is made up of one or more files.
2170 (The standard input is also a file.)
2172 You give @command{@value{AS}} a command line that has zero or more input file
2173 names. The input files are read (from left file name to right). A
2174 command-line argument (in any position) that has no special meaning
2175 is taken to be an input file name.
2177 If you give @command{@value{AS}} no file names it attempts to read one input file
2178 from the @command{@value{AS}} standard input, which is normally your terminal. You
2179 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2182 Use @samp{--} if you need to explicitly name the standard input file
2183 in your command line.
2185 If the source is empty, @command{@value{AS}} produces a small, empty object
2190 @subheading Filenames and Line-numbers
2192 @cindex input file linenumbers
2193 @cindex line numbers, in input files
2194 There are two ways of locating a line in the input file (or files) and
2195 either may be used in reporting error messages. One way refers to a line
2196 number in a physical file; the other refers to a line number in a
2197 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2199 @dfn{Physical files} are those files named in the command line given
2200 to @command{@value{AS}}.
2202 @dfn{Logical files} are simply names declared explicitly by assembler
2203 directives; they bear no relation to physical files. Logical file names help
2204 error messages reflect the original source file, when @command{@value{AS}} source
2205 is itself synthesized from other files. @command{@value{AS}} understands the
2206 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2207 @ref{File,,@code{.file}}.
2210 @section Output (Object) File
2216 Every time you run @command{@value{AS}} it produces an output file, which is
2217 your assembly language program translated into numbers. This file
2218 is the object file. Its default name is @code{a.out}.
2219 You can give it another name by using the @option{-o} option. Conventionally,
2220 object file names end with @file{.o}. The default name is used for historical
2221 reasons: older assemblers were capable of assembling self-contained programs
2222 directly into a runnable program. (For some formats, this isn't currently
2223 possible, but it can be done for the @code{a.out} format.)
2227 The object file is meant for input to the linker @code{@value{LD}}. It contains
2228 assembled program code, information to help @code{@value{LD}} integrate
2229 the assembled program into a runnable file, and (optionally) symbolic
2230 information for the debugger.
2232 @c link above to some info file(s) like the description of a.out.
2233 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2236 @section Error and Warning Messages
2238 @c man begin DESCRIPTION
2240 @cindex error messages
2241 @cindex warning messages
2242 @cindex messages from assembler
2243 @command{@value{AS}} may write warnings and error messages to the standard error
2244 file (usually your terminal). This should not happen when a compiler
2245 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2246 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2247 grave problem that stops the assembly.
2251 @cindex format of warning messages
2252 Warning messages have the format
2255 file_name:@b{NNN}:Warning Message Text
2259 @cindex file names and line numbers, in warnings/errors
2260 (where @b{NNN} is a line number). If both a logical file name
2261 (@pxref{File,,@code{.file}}) and a logical line number
2263 (@pxref{Line,,@code{.line}})
2265 have been given then they will be used, otherwise the file name and line number
2266 in the current assembler source file will be used. The message text is
2267 intended to be self explanatory (in the grand Unix tradition).
2269 Note the file name must be set via the logical version of the @code{.file}
2270 directive, not the DWARF2 version of the @code{.file} directive. For example:
2274 error_assembler_source
2280 produces this output:
2284 asm.s:2: Error: no such instruction: `error_assembler_source'
2285 foo.c:31: Error: no such instruction: `error_c_source'
2288 @cindex format of error messages
2289 Error messages have the format
2292 file_name:@b{NNN}:FATAL:Error Message Text
2295 The file name and line number are derived as for warning
2296 messages. The actual message text may be rather less explanatory
2297 because many of them aren't supposed to happen.
2300 @chapter Command-Line Options
2302 @cindex options, all versions of assembler
2303 This chapter describes command-line options available in @emph{all}
2304 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2305 for options specific
2307 to the @value{TARGET} target.
2310 to particular machine architectures.
2313 @c man begin DESCRIPTION
2315 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2316 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2317 The assembler arguments must be separated from each other (and the @samp{-Wa})
2318 by commas. For example:
2321 gcc -c -g -O -Wa,-alh,-L file.c
2325 This passes two options to the assembler: @samp{-alh} (emit a listing to
2326 standard output with high-level and assembly source) and @samp{-L} (retain
2327 local symbols in the symbol table).
2329 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2330 command-line options are automatically passed to the assembler by the compiler.
2331 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2332 precisely what options it passes to each compilation pass, including the
2338 * a:: -a[cdghlns] enable listings
2339 * alternate:: --alternate enable alternate macro syntax
2340 * D:: -D for compatibility
2341 * f:: -f to work faster
2342 * I:: -I for .include search path
2343 @ifclear DIFF-TBL-KLUGE
2344 * K:: -K for compatibility
2346 @ifset DIFF-TBL-KLUGE
2347 * K:: -K for difference tables
2350 * L:: -L to retain local symbols
2351 * listing:: --listing-XXX to configure listing output
2352 * M:: -M or --mri to assemble in MRI compatibility mode
2353 * MD:: --MD for dependency tracking
2354 * no-pad-sections:: --no-pad-sections to stop section padding
2355 * o:: -o to name the object file
2356 * R:: -R to join data and text sections
2357 * statistics:: --statistics to see statistics about assembly
2358 * traditional-format:: --traditional-format for compatible output
2359 * v:: -v to announce version
2360 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2361 * Z:: -Z to make object file even after errors
2365 @section Enable Listings: @option{-a[cdghlns]}
2375 @cindex listings, enabling
2376 @cindex assembly listings, enabling
2378 These options enable listing output from the assembler. By itself,
2379 @samp{-a} requests high-level, assembly, and symbols listing.
2380 You can use other letters to select specific options for the list:
2381 @samp{-ah} requests a high-level language listing,
2382 @samp{-al} requests an output-program assembly listing, and
2383 @samp{-as} requests a symbol table listing.
2384 High-level listings require that a compiler debugging option like
2385 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2388 Use the @samp{-ag} option to print a first section with general assembly
2389 information, like @value{AS} version, switches passed, or time stamp.
2391 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2392 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2393 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2394 omitted from the listing.
2396 Use the @samp{-ad} option to omit debugging directives from the
2399 Once you have specified one of these options, you can further control
2400 listing output and its appearance using the directives @code{.list},
2401 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2403 The @samp{-an} option turns off all forms processing.
2404 If you do not request listing output with one of the @samp{-a} options, the
2405 listing-control directives have no effect.
2407 The letters after @samp{-a} may be combined into one option,
2408 @emph{e.g.}, @samp{-aln}.
2410 Note if the assembler source is coming from the standard input (e.g.,
2412 is being created by @code{@value{GCC}} and the @samp{-pipe} command-line switch
2413 is being used) then the listing will not contain any comments or preprocessor
2414 directives. This is because the listing code buffers input source lines from
2415 stdin only after they have been preprocessed by the assembler. This reduces
2416 memory usage and makes the code more efficient.
2419 @section @option{--alternate}
2422 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2425 @section @option{-D}
2428 This option has no effect whatsoever, but it is accepted to make it more
2429 likely that scripts written for other assemblers also work with
2430 @command{@value{AS}}.
2433 @section Work Faster: @option{-f}
2436 @cindex trusted compiler
2437 @cindex faster processing (@option{-f})
2438 @samp{-f} should only be used when assembling programs written by a
2439 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2440 and comment preprocessing on
2441 the input file(s) before assembling them. @xref{Preprocessing,
2445 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2446 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2451 @section @code{.include} Search Path: @option{-I} @var{path}
2453 @kindex -I @var{path}
2454 @cindex paths for @code{.include}
2455 @cindex search path for @code{.include}
2456 @cindex @code{include} directive search path
2457 Use this option to add a @var{path} to the list of directories
2458 @command{@value{AS}} searches for files specified in @code{.include}
2459 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2460 many times as necessary to include a variety of paths. The current
2461 working directory is always searched first; after that, @command{@value{AS}}
2462 searches any @samp{-I} directories in the same order as they were
2463 specified (left to right) on the command line.
2466 @section Difference Tables: @option{-K}
2469 @ifclear DIFF-TBL-KLUGE
2470 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2471 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2472 where it can be used to warn when the assembler alters the machine code
2473 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2474 family does not have the addressing limitations that sometimes lead to this
2475 alteration on other platforms.
2478 @ifset DIFF-TBL-KLUGE
2479 @cindex difference tables, warning
2480 @cindex warning for altered difference tables
2481 @command{@value{AS}} sometimes alters the code emitted for directives of the
2482 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2483 You can use the @samp{-K} option if you want a warning issued when this
2488 @section Include Local Symbols: @option{-L}
2491 @cindex local symbols, retaining in output
2492 Symbols beginning with system-specific local label prefixes, typically
2493 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2494 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2495 such symbols when debugging, because they are intended for the use of
2496 programs (like compilers) that compose assembler programs, not for your
2497 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2498 such symbols, so you do not normally debug with them.
2500 This option tells @command{@value{AS}} to retain those local symbols
2501 in the object file. Usually if you do this you also tell the linker
2502 @code{@value{LD}} to preserve those symbols.
2505 @section Configuring listing output: @option{--listing}
2507 The listing feature of the assembler can be enabled via the command-line switch
2508 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2509 hex dump of the corresponding locations in the output object file, and displays
2510 them as a listing file. The format of this listing can be controlled by
2511 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2512 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2513 @code{.psize} (@pxref{Psize}), and
2514 @code{.eject} (@pxref{Eject}) and also by the following switches:
2517 @item --listing-lhs-width=@samp{number}
2518 @kindex --listing-lhs-width
2519 @cindex Width of first line disassembly output
2520 Sets the maximum width, in words, of the first line of the hex byte dump. This
2521 dump appears on the left hand side of the listing output.
2523 @item --listing-lhs-width2=@samp{number}
2524 @kindex --listing-lhs-width2
2525 @cindex Width of continuation lines of disassembly output
2526 Sets the maximum width, in words, of any further lines of the hex byte dump for
2527 a given input source line. If this value is not specified, it defaults to being
2528 the same as the value specified for @samp{--listing-lhs-width}. If neither
2529 switch is used the default is to one.
2531 @item --listing-rhs-width=@samp{number}
2532 @kindex --listing-rhs-width
2533 @cindex Width of source line output
2534 Sets the maximum width, in characters, of the source line that is displayed
2535 alongside the hex dump. The default value for this parameter is 100. The
2536 source line is displayed on the right hand side of the listing output.
2538 @item --listing-cont-lines=@samp{number}
2539 @kindex --listing-cont-lines
2540 @cindex Maximum number of continuation lines
2541 Sets the maximum number of continuation lines of hex dump that will be
2542 displayed for a given single line of source input. The default value is 4.
2546 @section Assemble in MRI Compatibility Mode: @option{-M}
2549 @cindex MRI compatibility mode
2550 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2551 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2552 compatible with the @code{ASM68K} assembler from Microtec Research.
2553 The exact nature of the
2554 MRI syntax will not be documented here; see the MRI manuals for more
2555 information. Note in particular that the handling of macros and macro
2556 arguments is somewhat different. The purpose of this option is to permit
2557 assembling existing MRI assembler code using @command{@value{AS}}.
2559 The MRI compatibility is not complete. Certain operations of the MRI assembler
2560 depend upon its object file format, and can not be supported using other object
2561 file formats. Supporting these would require enhancing each object file format
2562 individually. These are:
2565 @item global symbols in common section
2567 The m68k MRI assembler supports common sections which are merged by the linker.
2568 Other object file formats do not support this. @command{@value{AS}} handles
2569 common sections by treating them as a single common symbol. It permits local
2570 symbols to be defined within a common section, but it can not support global
2571 symbols, since it has no way to describe them.
2573 @item complex relocations
2575 The MRI assemblers support relocations against a negated section address, and
2576 relocations which combine the start addresses of two or more sections. These
2577 are not support by other object file formats.
2579 @item @code{END} pseudo-op specifying start address
2581 The MRI @code{END} pseudo-op permits the specification of a start address.
2582 This is not supported by other object file formats. The start address may
2583 instead be specified using the @option{-e} option to the linker, or in a linker
2586 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2588 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2589 name to the output file. This is not supported by other object file formats.
2591 @item @code{ORG} pseudo-op
2593 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2594 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2595 which changes the location within the current section. Absolute sections are
2596 not supported by other object file formats. The address of a section may be
2597 assigned within a linker script.
2600 There are some other features of the MRI assembler which are not supported by
2601 @command{@value{AS}}, typically either because they are difficult or because they
2602 seem of little consequence. Some of these may be supported in future releases.
2606 @item EBCDIC strings
2608 EBCDIC strings are not supported.
2610 @item packed binary coded decimal
2612 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2613 and @code{DCB.P} pseudo-ops are not supported.
2615 @item @code{FEQU} pseudo-op
2617 The m68k @code{FEQU} pseudo-op is not supported.
2619 @item @code{NOOBJ} pseudo-op
2621 The m68k @code{NOOBJ} pseudo-op is not supported.
2623 @item @code{OPT} branch control options
2625 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2626 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2627 relaxes all branches, whether forward or backward, to an appropriate size, so
2628 these options serve no purpose.
2630 @item @code{OPT} list control options
2632 The following m68k @code{OPT} list control options are ignored: @code{C},
2633 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2634 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2636 @item other @code{OPT} options
2638 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2639 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2641 @item @code{OPT} @code{D} option is default
2643 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2644 @code{OPT NOD} may be used to turn it off.
2646 @item @code{XREF} pseudo-op.
2648 The m68k @code{XREF} pseudo-op is ignored.
2653 @section Dependency Tracking: @option{--MD}
2656 @cindex dependency tracking
2659 @command{@value{AS}} can generate a dependency file for the file it creates. This
2660 file consists of a single rule suitable for @code{make} describing the
2661 dependencies of the main source file.
2663 The rule is written to the file named in its argument.
2665 This feature is used in the automatic updating of makefiles.
2667 @node no-pad-sections
2668 @section Output Section Padding
2669 @kindex --no-pad-sections
2670 @cindex output section padding
2671 Normally the assembler will pad the end of each output section up to its
2672 alignment boundary. But this can waste space, which can be significant on
2673 memory constrained targets. So the @option{--no-pad-sections} option will
2674 disable this behaviour.
2677 @section Name the Object File: @option{-o}
2680 @cindex naming object file
2681 @cindex object file name
2682 There is always one object file output when you run @command{@value{AS}}. By
2683 default it has the name @file{a.out}.
2684 You use this option (which takes exactly one filename) to give the
2685 object file a different name.
2687 Whatever the object file is called, @command{@value{AS}} overwrites any
2688 existing file of the same name.
2691 @section Join Data and Text Sections: @option{-R}
2694 @cindex data and text sections, joining
2695 @cindex text and data sections, joining
2696 @cindex joining text and data sections
2697 @cindex merging text and data sections
2698 @option{-R} tells @command{@value{AS}} to write the object file as if all
2699 data-section data lives in the text section. This is only done at
2700 the very last moment: your binary data are the same, but data
2701 section parts are relocated differently. The data section part of
2702 your object file is zero bytes long because all its bytes are
2703 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2705 When you specify @option{-R} it would be possible to generate shorter
2706 address displacements (because we do not have to cross between text and
2707 data section). We refrain from doing this simply for compatibility with
2708 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2711 When @command{@value{AS}} is configured for COFF or ELF output,
2712 this option is only useful if you use sections named @samp{.text} and
2717 @option{-R} is not supported for any of the HPPA targets. Using
2718 @option{-R} generates a warning from @command{@value{AS}}.
2722 @section Display Assembly Statistics: @option{--statistics}
2724 @kindex --statistics
2725 @cindex statistics, about assembly
2726 @cindex time, total for assembly
2727 @cindex space used, maximum for assembly
2728 Use @samp{--statistics} to display two statistics about the resources used by
2729 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2730 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2733 @node traditional-format
2734 @section Compatible Output: @option{--traditional-format}
2736 @kindex --traditional-format
2737 For some targets, the output of @command{@value{AS}} is different in some ways
2738 from the output of some existing assembler. This switch requests
2739 @command{@value{AS}} to use the traditional format instead.
2741 For example, it disables the exception frame optimizations which
2742 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2745 @section Announce Version: @option{-v}
2749 @cindex assembler version
2750 @cindex version of assembler
2751 You can find out what version of as is running by including the
2752 option @samp{-v} (which you can also spell as @samp{-version}) on the
2756 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2758 @command{@value{AS}} should never give a warning or error message when
2759 assembling compiler output. But programs written by people often
2760 cause @command{@value{AS}} to give a warning that a particular assumption was
2761 made. All such warnings are directed to the standard error file.
2765 @cindex suppressing warnings
2766 @cindex warnings, suppressing
2767 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2768 This only affects the warning messages: it does not change any particular of
2769 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2772 @kindex --fatal-warnings
2773 @cindex errors, caused by warnings
2774 @cindex warnings, causing error
2775 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2776 files that generate warnings to be in error.
2779 @cindex warnings, switching on
2780 You can switch these options off again by specifying @option{--warn}, which
2781 causes warnings to be output as usual.
2784 @section Generate Object File in Spite of Errors: @option{-Z}
2785 @cindex object file, after errors
2786 @cindex errors, continuing after
2787 After an error message, @command{@value{AS}} normally produces no output. If for
2788 some reason you are interested in object file output even after
2789 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2790 option. If there are any errors, @command{@value{AS}} continues anyways, and
2791 writes an object file after a final warning message of the form @samp{@var{n}
2792 errors, @var{m} warnings, generating bad object file.}
2797 @cindex machine-independent syntax
2798 @cindex syntax, machine-independent
2799 This chapter describes the machine-independent syntax allowed in a
2800 source file. @command{@value{AS}} syntax is similar to what many other
2801 assemblers use; it is inspired by the BSD 4.2
2806 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2810 * Preprocessing:: Preprocessing
2811 * Whitespace:: Whitespace
2812 * Comments:: Comments
2813 * Symbol Intro:: Symbols
2814 * Statements:: Statements
2815 * Constants:: Constants
2819 @section Preprocessing
2821 @cindex preprocessing
2822 The @command{@value{AS}} internal preprocessor:
2824 @cindex whitespace, removed by preprocessor
2826 adjusts and removes extra whitespace. It leaves one space or tab before
2827 the keywords on a line, and turns any other whitespace on the line into
2830 @cindex comments, removed by preprocessor
2832 removes all comments, replacing them with a single space, or an
2833 appropriate number of newlines.
2835 @cindex constants, converted by preprocessor
2837 converts character constants into the appropriate numeric values.
2840 It does not do macro processing, include file handling, or
2841 anything else you may get from your C compiler's preprocessor. You can
2842 do include file processing with the @code{.include} directive
2843 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2844 to get other ``CPP'' style preprocessing by giving the input file a
2845 @samp{.S} suffix. @url{https://gcc.gnu.org/onlinedocs/gcc/Overall-Options.html#Overall-Options,
2846 See the 'Options Controlling the Kind of Output' section of the GCC manual for
2849 Excess whitespace, comments, and character constants
2850 cannot be used in the portions of the input text that are not
2853 @cindex turning preprocessing on and off
2854 @cindex preprocessing, turning on and off
2857 If the first line of an input file is @code{#NO_APP} or if you use the
2858 @samp{-f} option, whitespace and comments are not removed from the input file.
2859 Within an input file, you can ask for whitespace and comment removal in
2860 specific portions of the by putting a line that says @code{#APP} before the
2861 text that may contain whitespace or comments, and putting a line that says
2862 @code{#NO_APP} after this text. This feature is mainly intend to support
2863 @code{asm} statements in compilers whose output is otherwise free of comments
2870 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2871 Whitespace is used to separate symbols, and to make programs neater for
2872 people to read. Unless within character constants
2873 (@pxref{Characters,,Character Constants}), any whitespace means the same
2874 as exactly one space.
2880 There are two ways of rendering comments to @command{@value{AS}}. In both
2881 cases the comment is equivalent to one space.
2883 Anything from @samp{/*} through the next @samp{*/} is a comment.
2884 This means you may not nest these comments.
2888 The only way to include a newline ('\n') in a comment
2889 is to use this sort of comment.
2892 /* This sort of comment does not nest. */
2895 @cindex line comment character
2896 Anything from a @dfn{line comment} character up to the next newline is
2897 considered a comment and is ignored. The line comment character is target
2898 specific, and some targets multiple comment characters. Some targets also have
2899 line comment characters that only work if they are the first character on a
2900 line. Some targets use a sequence of two characters to introduce a line
2901 comment. Some targets can also change their line comment characters depending
2902 upon command-line options that have been used. For more details see the
2903 @emph{Syntax} section in the documentation for individual targets.
2905 If the line comment character is the hash sign (@samp{#}) then it still has the
2906 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2907 to specify logical line numbers:
2910 @cindex lines starting with @code{#}
2911 @cindex logical line numbers
2912 To be compatible with past assemblers, lines that begin with @samp{#} have a
2913 special interpretation. Following the @samp{#} should be an absolute
2914 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2915 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2916 new logical file name. The rest of the line, if any, should be whitespace.
2918 If the first non-whitespace characters on the line are not numeric,
2919 the line is ignored. (Just like a comment.)
2922 # This is an ordinary comment.
2923 # 42-6 "new_file_name" # New logical file name
2924 # This is logical line # 36.
2926 This feature is deprecated, and may disappear from future versions
2927 of @command{@value{AS}}.
2932 @cindex characters used in symbols
2933 @ifclear SPECIAL-SYMS
2934 A @dfn{symbol} is one or more characters chosen from the set of all
2935 letters (both upper and lower case), digits and the three characters
2941 A @dfn{symbol} is one or more characters chosen from the set of all
2942 letters (both upper and lower case), digits and the three characters
2943 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2949 On most machines, you can also use @code{$} in symbol names; exceptions
2950 are noted in @ref{Machine Dependencies}.
2952 No symbol may begin with a digit. Case is significant.
2953 There is no length limit; all characters are significant. Multibyte characters
2954 are supported. Symbols are delimited by characters not in that set, or by the
2955 beginning of a file (since the source program must end with a newline, the end
2956 of a file is not a possible symbol delimiter). @xref{Symbols}.
2958 Symbol names may also be enclosed in double quote @code{"} characters. In such
2959 cases any characters are allowed, except for the NUL character. If a double
2960 quote character is to be included in the symbol name it must be preceeded by a
2961 backslash @code{\} character.
2962 @cindex length of symbols
2967 @cindex statements, structure of
2968 @cindex line separator character
2969 @cindex statement separator character
2971 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2972 @dfn{line separator character}. The line separator character is target
2973 specific and described in the @emph{Syntax} section of each
2974 target's documentation. Not all targets support a line separator character.
2975 The newline or line separator character is considered to be part of the
2976 preceding statement. Newlines and separators within character constants are an
2977 exception: they do not end statements.
2979 @cindex newline, required at file end
2980 @cindex EOF, newline must precede
2981 It is an error to end any statement with end-of-file: the last
2982 character of any input file should be a newline.@refill
2984 An empty statement is allowed, and may include whitespace. It is ignored.
2986 @cindex instructions and directives
2987 @cindex directives and instructions
2988 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2989 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2991 A statement begins with zero or more labels, optionally followed by a
2992 key symbol which determines what kind of statement it is. The key
2993 symbol determines the syntax of the rest of the statement. If the
2994 symbol begins with a dot @samp{.} then the statement is an assembler
2995 directive: typically valid for any computer. If the symbol begins with
2996 a letter the statement is an assembly language @dfn{instruction}: it
2997 assembles into a machine language instruction.
2999 Different versions of @command{@value{AS}} for different computers
3000 recognize different instructions. In fact, the same symbol may
3001 represent a different instruction in a different computer's assembly
3005 @cindex @code{:} (label)
3006 @cindex label (@code{:})
3007 A label is a symbol immediately followed by a colon (@code{:}).
3008 Whitespace before a label or after a colon is permitted, but you may not
3009 have whitespace between a label's symbol and its colon. @xref{Labels}.
3012 For HPPA targets, labels need not be immediately followed by a colon, but
3013 the definition of a label must begin in column zero. This also implies that
3014 only one label may be defined on each line.
3018 label: .directive followed by something
3019 another_label: # This is an empty statement.
3020 instruction operand_1, operand_2, @dots{}
3027 A constant is a number, written so that its value is known by
3028 inspection, without knowing any context. Like this:
3031 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
3032 .ascii "Ring the bell\7" # A string constant.
3033 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
3034 .float 0f-314159265358979323846264338327\
3035 95028841971.693993751E-40 # - pi, a flonum.
3040 * Characters:: Character Constants
3041 * Numbers:: Number Constants
3045 @subsection Character Constants
3047 @cindex character constants
3048 @cindex constants, character
3049 There are two kinds of character constants. A @dfn{character} stands
3050 for one character in one byte and its value may be used in
3051 numeric expressions. String constants (properly called string
3052 @emph{literals}) are potentially many bytes and their values may not be
3053 used in arithmetic expressions.
3057 * Chars:: Characters
3061 @subsubsection Strings
3063 @cindex string constants
3064 @cindex constants, string
3065 A @dfn{string} is written between double-quotes. It may contain
3066 double-quotes or null characters. The way to get special characters
3067 into a string is to @dfn{escape} these characters: precede them with
3068 a backslash @samp{\} character. For example @samp{\\} represents
3069 one backslash: the first @code{\} is an escape which tells
3070 @command{@value{AS}} to interpret the second character literally as a backslash
3071 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
3072 escape character). The complete list of escapes follows.
3074 @cindex escape codes, character
3075 @cindex character escape codes
3076 @c NOTE: Cindex entries must not start with a backlash character.
3077 @c NOTE: This confuses the pdf2texi script when it is creating the
3078 @c NOTE: index based upon the first character and so it generates:
3079 @c NOTE: \initial {\\}
3080 @c NOTE: which then results in the error message:
3081 @c NOTE: Argument of \\ has an extra }.
3082 @c NOTE: So in the index entries below a space character has been
3083 @c NOTE: prepended to avoid this problem.
3086 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
3088 @cindex @code{ \b} (backspace character)
3089 @cindex backspace (@code{\b})
3091 Mnemonic for backspace; for ASCII this is octal code 010.
3094 @c Mnemonic for EOText; for ASCII this is octal code 004.
3096 @cindex @code{ \f} (formfeed character)
3097 @cindex formfeed (@code{\f})
3099 Mnemonic for FormFeed; for ASCII this is octal code 014.
3101 @cindex @code{ \n} (newline character)
3102 @cindex newline (@code{\n})
3104 Mnemonic for newline; for ASCII this is octal code 012.
3107 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
3109 @cindex @code{ \r} (carriage return character)
3110 @cindex carriage return (@code{backslash-r})
3112 Mnemonic for carriage-Return; for ASCII this is octal code 015.
3115 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
3116 @c other assemblers.
3118 @cindex @code{ \t} (tab)
3119 @cindex tab (@code{\t})
3121 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
3124 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
3125 @c @item \x @var{digit} @var{digit} @var{digit}
3126 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
3128 @cindex @code{ \@var{ddd}} (octal character code)
3129 @cindex octal character code (@code{\@var{ddd}})
3130 @item \ @var{digit} @var{digit} @var{digit}
3131 An octal character code. The numeric code is 3 octal digits.
3132 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
3133 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3135 @cindex @code{ \@var{xd...}} (hex character code)
3136 @cindex hex character code (@code{\@var{xd...}})
3137 @item \@code{x} @var{hex-digits...}
3138 A hex character code. All trailing hex digits are combined. Either upper or
3139 lower case @code{x} works.
3141 @cindex @code{ \\} (@samp{\} character)
3142 @cindex backslash (@code{\\})
3144 Represents one @samp{\} character.
3147 @c Represents one @samp{'} (accent acute) character.
3148 @c This is needed in single character literals
3149 @c (@xref{Characters,,Character Constants}.) to represent
3152 @cindex @code{ \"} (doublequote character)
3153 @cindex doublequote (@code{\"})
3155 Represents one @samp{"} character. Needed in strings to represent
3156 this character, because an unescaped @samp{"} would end the string.
3158 @item \ @var{anything-else}
3159 Any other character when escaped by @kbd{\} gives a warning, but
3160 assembles as if the @samp{\} was not present. The idea is that if
3161 you used an escape sequence you clearly didn't want the literal
3162 interpretation of the following character. However @command{@value{AS}} has no
3163 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3164 code and warns you of the fact.
3167 Which characters are escapable, and what those escapes represent,
3168 varies widely among assemblers. The current set is what we think
3169 the BSD 4.2 assembler recognizes, and is a subset of what most C
3170 compilers recognize. If you are in doubt, do not use an escape
3174 @subsubsection Characters
3176 @cindex single character constant
3177 @cindex character, single
3178 @cindex constant, single character
3179 A single character may be written as a single quote immediately followed by
3180 that character. Some backslash escapes apply to characters, @code{\b},
3181 @code{\f}, @code{\n}, @code{\r}, @code{\t}, and @code{\"} with the same meaning
3182 as for strings, plus @code{\'} for a single quote. So if you want to write the
3183 character backslash, you must write @kbd{'\\} where the first @code{\} escapes
3184 the second @code{\}. As you can see, the quote is an acute accent, not a grave
3187 @ifclear abnormal-separator
3188 (or semicolon @samp{;})
3190 @ifset abnormal-separator
3192 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3197 immediately following an acute accent is taken as a literal character
3198 and does not count as the end of a statement. The value of a character
3199 constant in a numeric expression is the machine's byte-wide code for
3200 that character. @command{@value{AS}} assumes your character code is ASCII:
3201 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3204 @subsection Number Constants
3206 @cindex constants, number
3207 @cindex number constants
3208 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3209 are stored in the target machine. @emph{Integers} are numbers that
3210 would fit into an @code{int} in the C language. @emph{Bignums} are
3211 integers, but they are stored in more than 32 bits. @emph{Flonums}
3212 are floating point numbers, described below.
3215 * Integers:: Integers
3223 @subsubsection Integers
3225 @cindex constants, integer
3227 @cindex binary integers
3228 @cindex integers, binary
3229 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3230 the binary digits @samp{01}.
3232 @cindex octal integers
3233 @cindex integers, octal
3234 An octal integer is @samp{0} followed by zero or more of the octal
3235 digits (@samp{01234567}).
3237 @cindex decimal integers
3238 @cindex integers, decimal
3239 A decimal integer starts with a non-zero digit followed by zero or
3240 more digits (@samp{0123456789}).
3242 @cindex hexadecimal integers
3243 @cindex integers, hexadecimal
3244 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3245 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3247 Integers have the usual values. To denote a negative integer, use
3248 the prefix operator @samp{-} discussed under expressions
3249 (@pxref{Prefix Ops,,Prefix Operators}).
3252 @subsubsection Bignums
3255 @cindex constants, bignum
3256 A @dfn{bignum} has the same syntax and semantics as an integer
3257 except that the number (or its negative) takes more than 32 bits to
3258 represent in binary. The distinction is made because in some places
3259 integers are permitted while bignums are not.
3262 @subsubsection Flonums
3264 @cindex floating point numbers
3265 @cindex constants, floating point
3267 @cindex precision, floating point
3268 A @dfn{flonum} represents a floating point number. The translation is
3269 indirect: a decimal floating point number from the text is converted by
3270 @command{@value{AS}} to a generic binary floating point number of more than
3271 sufficient precision. This generic floating point number is converted
3272 to a particular computer's floating point format (or formats) by a
3273 portion of @command{@value{AS}} specialized to that computer.
3275 A flonum is written by writing (in order)
3280 (@samp{0} is optional on the HPPA.)
3284 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3286 @kbd{e} is recommended. Case is not important.
3288 @c FIXME: verify if flonum syntax really this vague for most cases
3289 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3290 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3293 On the H8/300 and Renesas / SuperH SH architectures, the letter must be
3294 one of the letters @samp{DFPRSX} (in upper or lower case).
3296 On the ARC, the letter must be one of the letters @samp{DFRS}
3297 (in upper or lower case).
3299 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3303 One of the letters @samp{DFRS} (in upper or lower case).
3306 One of the letters @samp{DFPRSX} (in upper or lower case).
3309 The letter @samp{E} (upper case only).
3314 An optional sign: either @samp{+} or @samp{-}.
3317 An optional @dfn{integer part}: zero or more decimal digits.
3320 An optional @dfn{fractional part}: @samp{.} followed by zero
3321 or more decimal digits.
3324 An optional exponent, consisting of:
3328 An @samp{E} or @samp{e}.
3329 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3330 @c principle this can perfectly well be different on different targets.
3332 Optional sign: either @samp{+} or @samp{-}.
3334 One or more decimal digits.
3339 At least one of the integer part or the fractional part must be
3340 present. The floating point number has the usual base-10 value.
3342 @command{@value{AS}} does all processing using integers. Flonums are computed
3343 independently of any floating point hardware in the computer running
3344 @command{@value{AS}}.
3347 @chapter Sections and Relocation
3352 * Secs Background:: Background
3353 * Ld Sections:: Linker Sections
3354 * As Sections:: Assembler Internal Sections
3355 * Sub-Sections:: Sub-Sections
3359 @node Secs Background
3362 Roughly, a section is a range of addresses, with no gaps; all data
3363 ``in'' those addresses is treated the same for some particular purpose.
3364 For example there may be a ``read only'' section.
3366 @cindex linker, and assembler
3367 @cindex assembler, and linker
3368 The linker @code{@value{LD}} reads many object files (partial programs) and
3369 combines their contents to form a runnable program. When @command{@value{AS}}
3370 emits an object file, the partial program is assumed to start at address 0.
3371 @code{@value{LD}} assigns the final addresses for the partial program, so that
3372 different partial programs do not overlap. This is actually an
3373 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3376 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3377 addresses. These blocks slide to their run-time addresses as rigid
3378 units; their length does not change and neither does the order of bytes
3379 within them. Such a rigid unit is called a @emph{section}. Assigning
3380 run-time addresses to sections is called @dfn{relocation}. It includes
3381 the task of adjusting mentions of object-file addresses so they refer to
3382 the proper run-time addresses.
3384 For the H8/300, and for the Renesas / SuperH SH,
3385 @command{@value{AS}} pads sections if needed to
3386 ensure they end on a word (sixteen bit) boundary.
3389 @cindex standard assembler sections
3390 An object file written by @command{@value{AS}} has at least three sections, any
3391 of which may be empty. These are named @dfn{text}, @dfn{data} and
3396 When it generates COFF or ELF output,
3398 @command{@value{AS}} can also generate whatever other named sections you specify
3399 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3400 If you do not use any directives that place output in the @samp{.text}
3401 or @samp{.data} sections, these sections still exist, but are empty.
3406 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3408 @command{@value{AS}} can also generate whatever other named sections you
3409 specify using the @samp{.space} and @samp{.subspace} directives. See
3410 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3411 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3412 assembler directives.
3415 Additionally, @command{@value{AS}} uses different names for the standard
3416 text, data, and bss sections when generating SOM output. Program text
3417 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3418 BSS into @samp{$BSS$}.
3422 Within the object file, the text section starts at address @code{0}, the
3423 data section follows, and the bss section follows the data section.
3426 When generating either SOM or ELF output files on the HPPA, the text
3427 section starts at address @code{0}, the data section at address
3428 @code{0x4000000}, and the bss section follows the data section.
3431 To let @code{@value{LD}} know which data changes when the sections are
3432 relocated, and how to change that data, @command{@value{AS}} also writes to the
3433 object file details of the relocation needed. To perform relocation
3434 @code{@value{LD}} must know, each time an address in the object
3438 Where in the object file is the beginning of this reference to
3441 How long (in bytes) is this reference?
3443 Which section does the address refer to? What is the numeric value of
3445 (@var{address}) @minus{} (@var{start-address of section})?
3448 Is the reference to an address ``Program-Counter relative''?
3451 @cindex addresses, format of
3452 @cindex section-relative addressing
3453 In fact, every address @command{@value{AS}} ever uses is expressed as
3455 (@var{section}) + (@var{offset into section})
3458 Further, most expressions @command{@value{AS}} computes have this section-relative
3461 (For some object formats, such as SOM for the HPPA, some expressions are
3462 symbol-relative instead.)
3465 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3466 @var{N} into section @var{secname}.''
3468 Apart from text, data and bss sections you need to know about the
3469 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3470 addresses in the absolute section remain unchanged. For example, address
3471 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3472 @code{@value{LD}}. Although the linker never arranges two partial programs'
3473 data sections with overlapping addresses after linking, @emph{by definition}
3474 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3475 part of a program is always the same address when the program is running as
3476 address @code{@{absolute@ 239@}} in any other part of the program.
3478 The idea of sections is extended to the @dfn{undefined} section. Any
3479 address whose section is unknown at assembly time is by definition
3480 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3481 Since numbers are always defined, the only way to generate an undefined
3482 address is to mention an undefined symbol. A reference to a named
3483 common block would be such a symbol: its value is unknown at assembly
3484 time so it has section @emph{undefined}.
3486 By analogy the word @emph{section} is used to describe groups of sections in
3487 the linked program. @code{@value{LD}} puts all partial programs' text
3488 sections in contiguous addresses in the linked program. It is
3489 customary to refer to the @emph{text section} of a program, meaning all
3490 the addresses of all partial programs' text sections. Likewise for
3491 data and bss sections.
3493 Some sections are manipulated by @code{@value{LD}}; others are invented for
3494 use of @command{@value{AS}} and have no meaning except during assembly.
3497 @section Linker Sections
3498 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3503 @cindex named sections
3504 @cindex sections, named
3505 @item named sections
3508 @cindex text section
3509 @cindex data section
3513 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3514 separate but equal sections. Anything you can say of one section is
3517 When the program is running, however, it is
3518 customary for the text section to be unalterable. The
3519 text section is often shared among processes: it contains
3520 instructions, constants and the like. The data section of a running
3521 program is usually alterable: for example, C variables would be stored
3522 in the data section.
3527 This section contains zeroed bytes when your program begins running. It
3528 is used to hold uninitialized variables or common storage. The length of
3529 each partial program's bss section is important, but because it starts
3530 out containing zeroed bytes there is no need to store explicit zero
3531 bytes in the object file. The bss section was invented to eliminate
3532 those explicit zeros from object files.
3534 @cindex absolute section
3535 @item absolute section
3536 Address 0 of this section is always ``relocated'' to runtime address 0.
3537 This is useful if you want to refer to an address that @code{@value{LD}} must
3538 not change when relocating. In this sense we speak of absolute
3539 addresses being ``unrelocatable'': they do not change during relocation.
3541 @cindex undefined section
3542 @item undefined section
3543 This ``section'' is a catch-all for address references to objects not in
3544 the preceding sections.
3545 @c FIXME: ref to some other doc on obj-file formats could go here.
3548 @cindex relocation example
3549 An idealized example of three relocatable sections follows.
3551 The example uses the traditional section names @samp{.text} and @samp{.data}.
3553 Memory addresses are on the horizontal axis.
3557 @c END TEXI2ROFF-KILL
3560 partial program # 1: |ttttt|dddd|00|
3567 partial program # 2: |TTT|DDD|000|
3570 +--+---+-----+--+----+---+-----+~~
3571 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3572 +--+---+-----+--+----+---+-----+~~
3574 addresses: 0 @dots{}
3581 \line{\it Partial program \#1: \hfil}
3582 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3583 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3585 \line{\it Partial program \#2: \hfil}
3586 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3587 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3589 \line{\it linked program: \hfil}
3590 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3591 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3592 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3593 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3595 \line{\it addresses: \hfil}
3599 @c END TEXI2ROFF-KILL
3602 @section Assembler Internal Sections
3604 @cindex internal assembler sections
3605 @cindex sections in messages, internal
3606 These sections are meant only for the internal use of @command{@value{AS}}. They
3607 have no meaning at run-time. You do not really need to know about these
3608 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3609 warning messages, so it might be helpful to have an idea of their
3610 meanings to @command{@value{AS}}. These sections are used to permit the
3611 value of every expression in your assembly language program to be a
3612 section-relative address.
3615 @cindex assembler internal logic error
3616 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3617 An internal assembler logic error has been found. This means there is a
3618 bug in the assembler.
3620 @cindex expr (internal section)
3622 The assembler stores complex expression internally as combinations of
3623 symbols. When it needs to represent an expression as a symbol, it puts
3624 it in the expr section.
3626 @c FIXME item transfer[t] vector preload
3627 @c FIXME item transfer[t] vector postload
3628 @c FIXME item register
3632 @section Sub-Sections
3634 @cindex numbered subsections
3635 @cindex grouping data
3641 fall into two sections: text and data.
3643 You may have separate groups of
3645 data in named sections
3649 data in named sections
3655 that you want to end up near to each other in the object file, even though they
3656 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3657 use @dfn{subsections} for this purpose. Within each section, there can be
3658 numbered subsections with values from 0 to 8192. Objects assembled into the
3659 same subsection go into the object file together with other objects in the same
3660 subsection. For example, a compiler might want to store constants in the text
3661 section, but might not want to have them interspersed with the program being
3662 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3663 section of code being output, and a @samp{.text 1} before each group of
3664 constants being output.
3666 Subsections are optional. If you do not use subsections, everything
3667 goes in subsection number zero.
3670 Each subsection is zero-padded up to a multiple of four bytes.
3671 (Subsections may be padded a different amount on different flavors
3672 of @command{@value{AS}}.)
3676 On the H8/300 platform, each subsection is zero-padded to a word
3677 boundary (two bytes).
3678 The same is true on the Renesas SH.
3682 Subsections appear in your object file in numeric order, lowest numbered
3683 to highest. (All this to be compatible with other people's assemblers.)
3684 The object file contains no representation of subsections; @code{@value{LD}} and
3685 other programs that manipulate object files see no trace of them.
3686 They just see all your text subsections as a text section, and all your
3687 data subsections as a data section.
3689 To specify which subsection you want subsequent statements assembled
3690 into, use a numeric argument to specify it, in a @samp{.text
3691 @var{expression}} or a @samp{.data @var{expression}} statement.
3694 When generating COFF output, you
3699 can also use an extra subsection
3700 argument with arbitrary named sections: @samp{.section @var{name},
3705 When generating ELF output, you
3710 can also use the @code{.subsection} directive (@pxref{SubSection})
3711 to specify a subsection: @samp{.subsection @var{expression}}.
3713 @var{Expression} should be an absolute expression
3714 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3715 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3716 begins in @code{text 0}. For instance:
3718 .text 0 # The default subsection is text 0 anyway.
3719 .ascii "This lives in the first text subsection. *"
3721 .ascii "But this lives in the second text subsection."
3723 .ascii "This lives in the data section,"
3724 .ascii "in the first data subsection."
3726 .ascii "This lives in the first text section,"
3727 .ascii "immediately following the asterisk (*)."
3730 Each section has a @dfn{location counter} incremented by one for every byte
3731 assembled into that section. Because subsections are merely a convenience
3732 restricted to @command{@value{AS}} there is no concept of a subsection location
3733 counter. There is no way to directly manipulate a location counter---but the
3734 @code{.align} directive changes it, and any label definition captures its
3735 current value. The location counter of the section where statements are being
3736 assembled is said to be the @dfn{active} location counter.
3739 @section bss Section
3742 @cindex common variable storage
3743 The bss section is used for local common variable storage.
3744 You may allocate address space in the bss section, but you may
3745 not dictate data to load into it before your program executes. When
3746 your program starts running, all the contents of the bss
3747 section are zeroed bytes.
3749 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3750 @ref{Lcomm,,@code{.lcomm}}.
3752 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3753 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3756 When assembling for a target which supports multiple sections, such as ELF or
3757 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3758 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3759 section. Typically the section will only contain symbol definitions and
3760 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3767 Symbols are a central concept: the programmer uses symbols to name
3768 things, the linker uses symbols to link, and the debugger uses symbols
3772 @cindex debuggers, and symbol order
3773 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3774 the same order they were declared. This may break some debuggers.
3779 * Setting Symbols:: Giving Symbols Other Values
3780 * Symbol Names:: Symbol Names
3781 * Dot:: The Special Dot Symbol
3782 * Symbol Attributes:: Symbol Attributes
3789 A @dfn{label} is written as a symbol immediately followed by a colon
3790 @samp{:}. The symbol then represents the current value of the
3791 active location counter, and is, for example, a suitable instruction
3792 operand. You are warned if you use the same symbol to represent two
3793 different locations: the first definition overrides any other
3797 On the HPPA, the usual form for a label need not be immediately followed by a
3798 colon, but instead must start in column zero. Only one label may be defined on
3799 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3800 provides a special directive @code{.label} for defining labels more flexibly.
3803 @node Setting Symbols
3804 @section Giving Symbols Other Values
3806 @cindex assigning values to symbols
3807 @cindex symbol values, assigning
3808 A symbol can be given an arbitrary value by writing a symbol, followed
3809 by an equals sign @samp{=}, followed by an expression
3810 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3811 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3812 equals sign @samp{=}@samp{=} here represents an equivalent of the
3813 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3816 Blackfin does not support symbol assignment with @samp{=}.
3820 @section Symbol Names
3822 @cindex symbol names
3823 @cindex names, symbol
3824 @ifclear SPECIAL-SYMS
3825 Symbol names begin with a letter or with one of @samp{._}. On most
3826 machines, you can also use @code{$} in symbol names; exceptions are
3827 noted in @ref{Machine Dependencies}. That character may be followed by any
3828 string of digits, letters, dollar signs (unless otherwise noted for a
3829 particular target machine), and underscores.
3833 Symbol names begin with a letter or with one of @samp{._}. On the
3834 Renesas SH you can also use @code{$} in symbol names. That
3835 character may be followed by any string of digits, letters, dollar signs (save
3836 on the H8/300), and underscores.
3840 Case of letters is significant: @code{foo} is a different symbol name
3843 Symbol names do not start with a digit. An exception to this rule is made for
3844 Local Labels. See below.
3846 Multibyte characters are supported. To generate a symbol name containing
3847 multibyte characters enclose it within double quotes and use escape codes. cf
3848 @xref{Strings}. Generating a multibyte symbol name from a label is not
3849 currently supported.
3851 Each symbol has exactly one name. Each name in an assembly language program
3852 refers to exactly one symbol. You may use that symbol name any number of times
3855 @subheading Local Symbol Names
3857 @cindex local symbol names
3858 @cindex symbol names, local
3859 A local symbol is any symbol beginning with certain local label prefixes.
3860 By default, the local label prefix is @samp{.L} for ELF systems or
3861 @samp{L} for traditional a.out systems, but each target may have its own
3862 set of local label prefixes.
3864 On the HPPA local symbols begin with @samp{L$}.
3867 Local symbols are defined and used within the assembler, but they are
3868 normally not saved in object files. Thus, they are not visible when debugging.
3869 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3870 to retain the local symbols in the object files.
3872 @subheading Local Labels
3874 @cindex local labels
3875 @cindex temporary symbol names
3876 @cindex symbol names, temporary
3877 Local labels are different from local symbols. Local labels help compilers and
3878 programmers use names temporarily. They create symbols which are guaranteed to
3879 be unique over the entire scope of the input source code and which can be
3880 referred to by a simple notation. To define a local label, write a label of
3881 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3882 To refer to the most recent previous definition of that label write
3883 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3884 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3885 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3887 There is no restriction on how you can use these labels, and you can reuse them
3888 too. So that it is possible to repeatedly define the same local label (using
3889 the same number @samp{@b{N}}), although you can only refer to the most recently
3890 defined local label of that number (for a backwards reference) or the next
3891 definition of a specific local label for a forward reference. It is also worth
3892 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3893 implemented in a slightly more efficient manner than the others.
3904 Which is the equivalent of:
3907 label_1: branch label_3
3908 label_2: branch label_1
3909 label_3: branch label_4
3910 label_4: branch label_3
3913 Local label names are only a notational device. They are immediately
3914 transformed into more conventional symbol names before the assembler uses them.
3915 The symbol names are stored in the symbol table, appear in error messages, and
3916 are optionally emitted to the object file. The names are constructed using
3920 @item @emph{local label prefix}
3921 All local symbols begin with the system-specific local label prefix.
3922 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3923 that start with the local label prefix. These labels are
3924 used for symbols you are never intended to see. If you use the
3925 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3926 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3927 you may use them in debugging.
3930 This is the number that was used in the local label definition. So if the
3931 label is written @samp{55:} then the number is @samp{55}.
3934 This unusual character is included so you do not accidentally invent a symbol
3935 of the same name. The character has ASCII value of @samp{\002} (control-B).
3937 @item @emph{ordinal number}
3938 This is a serial number to keep the labels distinct. The first definition of
3939 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3940 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3941 the number @samp{1} and its 15th definition gets @samp{15} as well.
3944 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3945 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3947 @subheading Dollar Local Labels
3948 @cindex dollar local symbols
3950 On some targets @code{@value{AS}} also supports an even more local form of
3951 local labels called dollar labels. These labels go out of scope (i.e., they
3952 become undefined) as soon as a non-local label is defined. Thus they remain
3953 valid for only a small region of the input source code. Normal local labels,
3954 by contrast, remain in scope for the entire file, or until they are redefined
3955 by another occurrence of the same local label.
3957 Dollar labels are defined in exactly the same way as ordinary local labels,
3958 except that they have a dollar sign suffix to their numeric value, e.g.,
3961 They can also be distinguished from ordinary local labels by their transformed
3962 names which use ASCII character @samp{\001} (control-A) as the magic character
3963 to distinguish them from ordinary labels. For example, the fifth definition of
3964 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3967 @section The Special Dot Symbol
3969 @cindex dot (symbol)
3970 @cindex @code{.} (symbol)
3971 @cindex current address
3972 @cindex location counter
3973 The special symbol @samp{.} refers to the current address that
3974 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3975 .long .} defines @code{melvin} to contain its own address.
3976 Assigning a value to @code{.} is treated the same as a @code{.org}
3978 @ifclear no-space-dir
3979 Thus, the expression @samp{.=.+4} is the same as saying
3983 @node Symbol Attributes
3984 @section Symbol Attributes
3986 @cindex symbol attributes
3987 @cindex attributes, symbol
3988 Every symbol has, as well as its name, the attributes ``Value'' and
3989 ``Type''. Depending on output format, symbols can also have auxiliary
3992 The detailed definitions are in @file{a.out.h}.
3995 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3996 all these attributes, and probably won't warn you. This makes the
3997 symbol an externally defined symbol, which is generally what you
4001 * Symbol Value:: Value
4002 * Symbol Type:: Type
4004 * a.out Symbols:: Symbol Attributes: @code{a.out}
4007 * COFF Symbols:: Symbol Attributes for COFF
4010 * SOM Symbols:: Symbol Attributes for SOM
4017 @cindex value of a symbol
4018 @cindex symbol value
4019 The value of a symbol is (usually) 32 bits. For a symbol which labels a
4020 location in the text, data, bss or absolute sections the value is the
4021 number of addresses from the start of that section to the label.
4022 Naturally for text, data and bss sections the value of a symbol changes
4023 as @code{@value{LD}} changes section base addresses during linking. Absolute
4024 symbols' values do not change during linking: that is why they are
4027 The value of an undefined symbol is treated in a special way. If it is
4028 0 then the symbol is not defined in this assembler source file, and
4029 @code{@value{LD}} tries to determine its value from other files linked into the
4030 same program. You make this kind of symbol simply by mentioning a symbol
4031 name without defining it. A non-zero value represents a @code{.comm}
4032 common declaration. The value is how much common storage to reserve, in
4033 bytes (addresses). The symbol refers to the first address of the
4039 @cindex type of a symbol
4041 The type attribute of a symbol contains relocation (section)
4042 information, any flag settings indicating that a symbol is external, and
4043 (optionally), other information for linkers and debuggers. The exact
4044 format depends on the object-code output format in use.
4048 @subsection Symbol Attributes: @code{a.out}
4050 @cindex @code{a.out} symbol attributes
4051 @cindex symbol attributes, @code{a.out}
4054 * Symbol Desc:: Descriptor
4055 * Symbol Other:: Other
4059 @subsubsection Descriptor
4061 @cindex descriptor, of @code{a.out} symbol
4062 This is an arbitrary 16-bit value. You may establish a symbol's
4063 descriptor value by using a @code{.desc} statement
4064 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
4065 @command{@value{AS}}.
4068 @subsubsection Other
4070 @cindex other attribute, of @code{a.out} symbol
4071 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
4076 @subsection Symbol Attributes for COFF
4078 @cindex COFF symbol attributes
4079 @cindex symbol attributes, COFF
4081 The COFF format supports a multitude of auxiliary symbol attributes;
4082 like the primary symbol attributes, they are set between @code{.def} and
4083 @code{.endef} directives.
4085 @subsubsection Primary Attributes
4087 @cindex primary attributes, COFF symbols
4088 The symbol name is set with @code{.def}; the value and type,
4089 respectively, with @code{.val} and @code{.type}.
4091 @subsubsection Auxiliary Attributes
4093 @cindex auxiliary attributes, COFF symbols
4094 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
4095 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
4096 table information for COFF.
4101 @subsection Symbol Attributes for SOM
4103 @cindex SOM symbol attributes
4104 @cindex symbol attributes, SOM
4106 The SOM format for the HPPA supports a multitude of symbol attributes set with
4107 the @code{.EXPORT} and @code{.IMPORT} directives.
4109 The attributes are described in @cite{HP9000 Series 800 Assembly
4110 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
4111 @code{EXPORT} assembler directive documentation.
4115 @chapter Expressions
4119 @cindex numeric values
4120 An @dfn{expression} specifies an address or numeric value.
4121 Whitespace may precede and/or follow an expression.
4123 The result of an expression must be an absolute number, or else an offset into
4124 a particular section. If an expression is not absolute, and there is not
4125 enough information when @command{@value{AS}} sees the expression to know its
4126 section, a second pass over the source program might be necessary to interpret
4127 the expression---but the second pass is currently not implemented.
4128 @command{@value{AS}} aborts with an error message in this situation.
4131 * Empty Exprs:: Empty Expressions
4132 * Integer Exprs:: Integer Expressions
4136 @section Empty Expressions
4138 @cindex empty expressions
4139 @cindex expressions, empty
4140 An empty expression has no value: it is just whitespace or null.
4141 Wherever an absolute expression is required, you may omit the
4142 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4143 is compatible with other assemblers.
4146 @section Integer Expressions
4148 @cindex integer expressions
4149 @cindex expressions, integer
4150 An @dfn{integer expression} is one or more @emph{arguments} delimited
4151 by @emph{operators}.
4154 * Arguments:: Arguments
4155 * Operators:: Operators
4156 * Prefix Ops:: Prefix Operators
4157 * Infix Ops:: Infix Operators
4161 @subsection Arguments
4163 @cindex expression arguments
4164 @cindex arguments in expressions
4165 @cindex operands in expressions
4166 @cindex arithmetic operands
4167 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4168 contexts arguments are sometimes called ``arithmetic operands''. In
4169 this manual, to avoid confusing them with the ``instruction operands'' of
4170 the machine language, we use the term ``argument'' to refer to parts of
4171 expressions only, reserving the word ``operand'' to refer only to machine
4172 instruction operands.
4174 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4175 @var{section} is one of text, data, bss, absolute,
4176 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4179 Numbers are usually integers.
4181 A number can be a flonum or bignum. In this case, you are warned
4182 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4183 these 32 bits are an integer. You may write integer-manipulating
4184 instructions that act on exotic constants, compatible with other
4187 @cindex subexpressions
4188 Subexpressions are a left parenthesis @samp{(} followed by an integer
4189 expression, followed by a right parenthesis @samp{)}; or a prefix
4190 operator followed by an argument.
4193 @subsection Operators
4195 @cindex operators, in expressions
4196 @cindex arithmetic functions
4197 @cindex functions, in expressions
4198 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4199 operators are followed by an argument. Infix operators appear
4200 between their arguments. Operators may be preceded and/or followed by
4204 @subsection Prefix Operator
4206 @cindex prefix operators
4207 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4208 one argument, which must be absolute.
4210 @c the tex/end tex stuff surrounding this small table is meant to make
4211 @c it align, on the printed page, with the similar table in the next
4212 @c section (which is inside an enumerate).
4214 \global\advance\leftskip by \itemindent
4219 @dfn{Negation}. Two's complement negation.
4221 @dfn{Complementation}. Bitwise not.
4225 \global\advance\leftskip by -\itemindent
4229 @subsection Infix Operators
4231 @cindex infix operators
4232 @cindex operators, permitted arguments
4233 @dfn{Infix operators} take two arguments, one on either side. Operators
4234 have precedence, but operations with equal precedence are performed left
4235 to right. Apart from @code{+} or @option{-}, both arguments must be
4236 absolute, and the result is absolute.
4239 @cindex operator precedence
4240 @cindex precedence of operators
4247 @dfn{Multiplication}.
4250 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4256 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4259 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4263 Intermediate precedence
4268 @dfn{Bitwise Inclusive Or}.
4274 @dfn{Bitwise Exclusive Or}.
4277 @dfn{Bitwise Or Not}.
4284 @cindex addition, permitted arguments
4285 @cindex plus, permitted arguments
4286 @cindex arguments for addition
4288 @dfn{Addition}. If either argument is absolute, the result has the section of
4289 the other argument. You may not add together arguments from different
4292 @cindex subtraction, permitted arguments
4293 @cindex minus, permitted arguments
4294 @cindex arguments for subtraction
4296 @dfn{Subtraction}. If the right argument is absolute, the
4297 result has the section of the left argument.
4298 If both arguments are in the same section, the result is absolute.
4299 You may not subtract arguments from different sections.
4300 @c FIXME is there still something useful to say about undefined - undefined ?
4302 @cindex comparison expressions
4303 @cindex expressions, comparison
4308 @dfn{Is Not Equal To}
4312 @dfn{Is Greater Than}
4314 @dfn{Is Greater Than Or Equal To}
4316 @dfn{Is Less Than Or Equal To}
4318 The comparison operators can be used as infix operators. A true results has a
4319 value of -1 whereas a false result has a value of 0. Note, these operators
4320 perform signed comparisons.
4323 @item Lowest Precedence
4332 These two logical operations can be used to combine the results of sub
4333 expressions. Note, unlike the comparison operators a true result returns a
4334 value of 1 but a false results does still return 0. Also note that the logical
4335 or operator has a slightly lower precedence than logical and.
4340 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4341 address; you can only have a defined section in one of the two arguments.
4344 @chapter Assembler Directives
4346 @cindex directives, machine independent
4347 @cindex pseudo-ops, machine independent
4348 @cindex machine independent directives
4349 All assembler directives have names that begin with a period (@samp{.}).
4350 The names are case insensitive for most targets, and usually written
4353 This chapter discusses directives that are available regardless of the
4354 target machine configuration for the @sc{gnu} assembler.
4356 Some machine configurations provide additional directives.
4357 @xref{Machine Dependencies}.
4360 @ifset machine-directives
4361 @xref{Machine Dependencies}, for additional directives.
4366 * Abort:: @code{.abort}
4368 * ABORT (COFF):: @code{.ABORT}
4371 * Align:: @code{.align [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4372 * Altmacro:: @code{.altmacro}
4373 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4374 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4375 * Balign:: @code{.balign [@var{abs-expr}[, @var{abs-expr}]]}
4376 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4377 * Byte:: @code{.byte @var{expressions}}
4378 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4379 * Comm:: @code{.comm @var{symbol} , @var{length} }
4380 * Data:: @code{.data @var{subsection}}
4381 * Dc:: @code{.dc[@var{size}] @var{expressions}}
4382 * Dcb:: @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4383 * Ds:: @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4385 * Def:: @code{.def @var{name}}
4388 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4394 * Double:: @code{.double @var{flonums}}
4395 * Eject:: @code{.eject}
4396 * Else:: @code{.else}
4397 * Elseif:: @code{.elseif}
4400 * Endef:: @code{.endef}
4403 * Endfunc:: @code{.endfunc}
4404 * Endif:: @code{.endif}
4405 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4406 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4407 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4409 * Error:: @code{.error @var{string}}
4410 * Exitm:: @code{.exitm}
4411 * Extern:: @code{.extern}
4412 * Fail:: @code{.fail}
4413 * File:: @code{.file}
4414 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4415 * Float:: @code{.float @var{flonums}}
4416 * Func:: @code{.func}
4417 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4419 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4420 * Hidden:: @code{.hidden @var{names}}
4423 * hword:: @code{.hword @var{expressions}}
4424 * Ident:: @code{.ident}
4425 * If:: @code{.if @var{absolute expression}}
4426 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4427 * Include:: @code{.include "@var{file}"}
4428 * Int:: @code{.int @var{expressions}}
4430 * Internal:: @code{.internal @var{names}}
4433 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4434 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4435 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4436 * Lflags:: @code{.lflags}
4437 @ifclear no-line-dir
4438 * Line:: @code{.line @var{line-number}}
4441 * Linkonce:: @code{.linkonce [@var{type}]}
4442 * List:: @code{.list}
4443 * Ln:: @code{.ln @var{line-number}}
4444 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4445 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4447 * Local:: @code{.local @var{names}}
4450 * Long:: @code{.long @var{expressions}}
4452 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4455 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4456 * MRI:: @code{.mri @var{val}}
4457 * Noaltmacro:: @code{.noaltmacro}
4458 * Nolist:: @code{.nolist}
4459 * Nops:: @code{.nops @var{size}[, @var{control}]}
4460 * Octa:: @code{.octa @var{bignums}}
4461 * Offset:: @code{.offset @var{loc}}
4462 * Org:: @code{.org @var{new-lc}, @var{fill}}
4463 * P2align:: @code{.p2align [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4465 * PopSection:: @code{.popsection}
4466 * Previous:: @code{.previous}
4469 * Print:: @code{.print @var{string}}
4471 * Protected:: @code{.protected @var{names}}
4474 * Psize:: @code{.psize @var{lines}, @var{columns}}
4475 * Purgem:: @code{.purgem @var{name}}
4477 * PushSection:: @code{.pushsection @var{name}}
4480 * Quad:: @code{.quad @var{bignums}}
4481 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4482 * Rept:: @code{.rept @var{count}}
4483 * Sbttl:: @code{.sbttl "@var{subheading}"}
4485 * Scl:: @code{.scl @var{class}}
4488 * Section:: @code{.section @var{name}[, @var{flags}]}
4491 * Set:: @code{.set @var{symbol}, @var{expression}}
4492 * Short:: @code{.short @var{expressions}}
4493 * Single:: @code{.single @var{flonums}}
4495 * Size:: @code{.size [@var{name} , @var{expression}]}
4497 @ifclear no-space-dir
4498 * Skip:: @code{.skip @var{size} [,@var{fill}]}
4501 * Sleb128:: @code{.sleb128 @var{expressions}}
4502 @ifclear no-space-dir
4503 * Space:: @code{.space @var{size} [,@var{fill}]}
4506 * Stab:: @code{.stabd, .stabn, .stabs}
4509 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4510 * Struct:: @code{.struct @var{expression}}
4512 * SubSection:: @code{.subsection}
4513 * Symver:: @code{.symver @var{name},@var{name2@@nodename}[,@var{visibility}]}
4517 * Tag:: @code{.tag @var{structname}}
4520 * Text:: @code{.text @var{subsection}}
4521 * Title:: @code{.title "@var{heading}"}
4523 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4526 * Uleb128:: @code{.uleb128 @var{expressions}}
4528 * Val:: @code{.val @var{addr}}
4532 * Version:: @code{.version "@var{string}"}
4533 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4534 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4537 * Warning:: @code{.warning @var{string}}
4538 * Weak:: @code{.weak @var{names}}
4539 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4540 * Word:: @code{.word @var{expressions}}
4541 @ifclear no-space-dir
4542 * Zero:: @code{.zero @var{size}}
4545 * 2byte:: @code{.2byte @var{expressions}}
4546 * 4byte:: @code{.4byte @var{expressions}}
4547 * 8byte:: @code{.8byte @var{bignums}}
4549 * Deprecated:: Deprecated Directives
4553 @section @code{.abort}
4555 @cindex @code{abort} directive
4556 @cindex stopping the assembly
4557 This directive stops the assembly immediately. It is for
4558 compatibility with other assemblers. The original idea was that the
4559 assembly language source would be piped into the assembler. If the sender
4560 of the source quit, it could use this directive tells @command{@value{AS}} to
4561 quit also. One day @code{.abort} will not be supported.
4565 @section @code{.ABORT} (COFF)
4567 @cindex @code{ABORT} directive
4568 When producing COFF output, @command{@value{AS}} accepts this directive as a
4569 synonym for @samp{.abort}.
4574 @section @code{.align [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4576 @cindex padding the location counter
4577 @cindex @code{align} directive
4578 Pad the location counter (in the current subsection) to a particular storage
4579 boundary. The first expression (which must be absolute) is the alignment
4580 required, as described below. If this expression is omitted then a default
4581 value of 0 is used, effectively disabling alignment requirements.
4583 The second expression (also absolute) gives the fill value to be stored in the
4584 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4585 padding bytes are normally zero. However, on most systems, if the section is
4586 marked as containing code and the fill value is omitted, the space is filled
4587 with no-op instructions.
4589 The third expression is also absolute, and is also optional. If it is present,
4590 it is the maximum number of bytes that should be skipped by this alignment
4591 directive. If doing the alignment would require skipping more bytes than the
4592 specified maximum, then the alignment is not done at all. You can omit the
4593 fill value (the second argument) entirely by simply using two commas after the
4594 required alignment; this can be useful if you want the alignment to be filled
4595 with no-op instructions when appropriate.
4597 The way the required alignment is specified varies from system to system.
4598 For the arc, hppa, i386 using ELF, iq2000, m68k, or1k,
4599 s390, sparc, tic4x and xtensa, the first expression is the
4600 alignment request in bytes. For example @samp{.align 8} advances
4601 the location counter until it is a multiple of 8. If the location counter
4602 is already a multiple of 8, no change is needed. For the tic54x, the
4603 first expression is the alignment request in words.
4605 For other systems, including ppc, i386 using a.out format, arm and
4606 strongarm, it is the
4607 number of low-order zero bits the location counter must have after
4608 advancement. For example @samp{.align 3} advances the location
4609 counter until it is a multiple of 8. If the location counter is already a
4610 multiple of 8, no change is needed.
4612 This inconsistency is due to the different behaviors of the various
4613 native assemblers for these systems which GAS must emulate.
4614 GAS also provides @code{.balign} and @code{.p2align} directives,
4615 described later, which have a consistent behavior across all
4616 architectures (but are specific to GAS).
4619 @section @code{.altmacro}
4620 Enable alternate macro mode, enabling:
4623 @item LOCAL @var{name} [ , @dots{} ]
4624 One additional directive, @code{LOCAL}, is available. It is used to
4625 generate a string replacement for each of the @var{name} arguments, and
4626 replace any instances of @var{name} in each macro expansion. The
4627 replacement string is unique in the assembly, and different for each
4628 separate macro expansion. @code{LOCAL} allows you to write macros that
4629 define symbols, without fear of conflict between separate macro expansions.
4631 @item String delimiters
4632 You can write strings delimited in these other ways besides
4633 @code{"@var{string}"}:
4636 @item '@var{string}'
4637 You can delimit strings with single-quote characters.
4639 @item <@var{string}>
4640 You can delimit strings with matching angle brackets.
4643 @item single-character string escape
4644 To include any single character literally in a string (even if the
4645 character would otherwise have some special meaning), you can prefix the
4646 character with @samp{!} (an exclamation mark). For example, you can
4647 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4649 @item Expression results as strings
4650 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4651 and use the result as a string.
4655 @section @code{.ascii "@var{string}"}@dots{}
4657 @cindex @code{ascii} directive
4658 @cindex string literals
4659 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4660 separated by commas. It assembles each string (with no automatic
4661 trailing zero byte) into consecutive addresses.
4664 @section @code{.asciz "@var{string}"}@dots{}
4666 @cindex @code{asciz} directive
4667 @cindex zero-terminated strings
4668 @cindex null-terminated strings
4669 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4670 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4673 @section @code{.balign[wl] [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4675 @cindex padding the location counter given number of bytes
4676 @cindex @code{balign} directive
4677 Pad the location counter (in the current subsection) to a particular
4678 storage boundary. The first expression (which must be absolute) is the
4679 alignment request in bytes. For example @samp{.balign 8} advances
4680 the location counter until it is a multiple of 8. If the location counter
4681 is already a multiple of 8, no change is needed. If the expression is omitted
4682 then a default value of 0 is used, effectively disabling alignment requirements.
4684 The second expression (also absolute) gives the fill value to be stored in the
4685 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4686 padding bytes are normally zero. However, on most systems, if the section is
4687 marked as containing code and the fill value is omitted, the space is filled
4688 with no-op instructions.
4690 The third expression is also absolute, and is also optional. If it is present,
4691 it is the maximum number of bytes that should be skipped by this alignment
4692 directive. If doing the alignment would require skipping more bytes than the
4693 specified maximum, then the alignment is not done at all. You can omit the
4694 fill value (the second argument) entirely by simply using two commas after the
4695 required alignment; this can be useful if you want the alignment to be filled
4696 with no-op instructions when appropriate.
4698 @cindex @code{balignw} directive
4699 @cindex @code{balignl} directive
4700 The @code{.balignw} and @code{.balignl} directives are variants of the
4701 @code{.balign} directive. The @code{.balignw} directive treats the fill
4702 pattern as a two byte word value. The @code{.balignl} directives treats the
4703 fill pattern as a four byte longword value. For example, @code{.balignw
4704 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4705 filled in with the value 0x368d (the exact placement of the bytes depends upon
4706 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4709 @node Bundle directives
4710 @section Bundle directives
4711 @subsection @code{.bundle_align_mode @var{abs-expr}}
4712 @cindex @code{bundle_align_mode} directive
4714 @cindex instruction bundle
4715 @cindex aligned instruction bundle
4716 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4717 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4718 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4719 disabled (which is the default state). If the argument it not zero, it
4720 gives the size of an instruction bundle as a power of two (as for the
4721 @code{.p2align} directive, @pxref{P2align}).
4723 For some targets, it's an ABI requirement that no instruction may span a
4724 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4725 instructions that starts on an aligned boundary. For example, if
4726 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4727 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4728 effect, no single instruction may span a boundary between bundles. If an
4729 instruction would start too close to the end of a bundle for the length of
4730 that particular instruction to fit within the bundle, then the space at the
4731 end of that bundle is filled with no-op instructions so the instruction
4732 starts in the next bundle. As a corollary, it's an error if any single
4733 instruction's encoding is longer than the bundle size.
4735 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4736 @cindex @code{bundle_lock} directive
4737 @cindex @code{bundle_unlock} directive
4738 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4739 allow explicit control over instruction bundle padding. These directives
4740 are only valid when @code{.bundle_align_mode} has been used to enable
4741 aligned instruction bundle mode. It's an error if they appear when
4742 @code{.bundle_align_mode} has not been used at all, or when the last
4743 directive was @w{@code{.bundle_align_mode 0}}.
4745 @cindex bundle-locked
4746 For some targets, it's an ABI requirement that certain instructions may
4747 appear only as part of specified permissible sequences of multiple
4748 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4749 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4750 instruction sequence. For purposes of aligned instruction bundle mode, a
4751 sequence starting with @code{.bundle_lock} and ending with
4752 @code{.bundle_unlock} is treated as a single instruction. That is, the
4753 entire sequence must fit into a single bundle and may not span a bundle
4754 boundary. If necessary, no-op instructions will be inserted before the
4755 first instruction of the sequence so that the whole sequence starts on an
4756 aligned bundle boundary. It's an error if the sequence is longer than the
4759 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4760 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4761 nested. That is, a second @code{.bundle_lock} directive before the next
4762 @code{.bundle_unlock} directive has no effect except that it must be
4763 matched by another closing @code{.bundle_unlock} so that there is the
4764 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4767 @section @code{.byte @var{expressions}}
4769 @cindex @code{byte} directive
4770 @cindex integers, one byte
4771 @code{.byte} expects zero or more expressions, separated by commas.
4772 Each expression is assembled into the next byte.
4774 @node CFI directives
4775 @section CFI directives
4776 @subsection @code{.cfi_sections @var{section_list}}
4777 @cindex @code{cfi_sections} directive
4778 @code{.cfi_sections} may be used to specify whether CFI directives
4779 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4780 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4781 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4782 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4783 directive is not used is @code{.cfi_sections .eh_frame}.
4785 On targets that support compact unwinding tables these can be generated
4786 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4788 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4789 which is used by the @value{TIC6X} target.
4791 The @code{.cfi_sections} directive can be repeated, with the same or different
4792 arguments, provided that CFI generation has not yet started. Once CFI
4793 generation has started however the section list is fixed and any attempts to
4794 redefine it will result in an error.
4796 @subsection @code{.cfi_startproc [simple]}
4797 @cindex @code{cfi_startproc} directive
4798 @code{.cfi_startproc} is used at the beginning of each function that
4799 should have an entry in @code{.eh_frame}. It initializes some internal
4800 data structures. Don't forget to close the function by
4801 @code{.cfi_endproc}.
4803 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4804 it also emits some architecture dependent initial CFI instructions.
4806 @subsection @code{.cfi_endproc}
4807 @cindex @code{cfi_endproc} directive
4808 @code{.cfi_endproc} is used at the end of a function where it closes its
4809 unwind entry previously opened by
4810 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4812 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4813 @cindex @code{cfi_personality} directive
4814 @code{.cfi_personality} defines personality routine and its encoding.
4815 @var{encoding} must be a constant determining how the personality
4816 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4817 argument is not present, otherwise second argument should be
4818 a constant or a symbol name. When using indirect encodings,
4819 the symbol provided should be the location where personality
4820 can be loaded from, not the personality routine itself.
4821 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4822 no personality routine.
4824 @subsection @code{.cfi_personality_id @var{id}}
4825 @cindex @code{cfi_personality_id} directive
4826 @code{cfi_personality_id} defines a personality routine by its index as
4827 defined in a compact unwinding format.
4828 Only valid when generating compact EH frames (i.e.
4829 with @code{.cfi_sections eh_frame_entry}.
4831 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4832 @cindex @code{cfi_fde_data} directive
4833 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4834 used for the current function. These are emitted inline in the
4835 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4836 in the @code{.gnu.extab} section otherwise.
4837 Only valid when generating compact EH frames (i.e.
4838 with @code{.cfi_sections eh_frame_entry}.
4840 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4841 @code{.cfi_lsda} defines LSDA and its encoding.
4842 @var{encoding} must be a constant determining how the LSDA
4843 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4844 argument is not present, otherwise the second argument should be a constant
4845 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4846 meaning that no LSDA is present.
4848 @subsection @code{.cfi_inline_lsda} [@var{align}]
4849 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4850 switches to the corresponding @code{.gnu.extab} section.
4851 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4852 Only valid when generating compact EH frames (i.e.
4853 with @code{.cfi_sections eh_frame_entry}.
4855 The table header and unwinding opcodes will be generated at this point,
4856 so that they are immediately followed by the LSDA data. The symbol
4857 referenced by the @code{.cfi_lsda} directive should still be defined
4858 in case a fallback FDE based encoding is used. The LSDA data is terminated
4859 by a section directive.
4861 The optional @var{align} argument specifies the alignment required.
4862 The alignment is specified as a power of two, as with the
4863 @code{.p2align} directive.
4865 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4866 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4867 address from @var{register} and add @var{offset} to it}.
4869 @subsection @code{.cfi_def_cfa_register @var{register}}
4870 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4871 now on @var{register} will be used instead of the old one. Offset
4874 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4875 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4876 remains the same, but @var{offset} is new. Note that it is the
4877 absolute offset that will be added to a defined register to compute
4880 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4881 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4882 value that is added/subtracted from the previous offset.
4884 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4885 Previous value of @var{register} is saved at offset @var{offset} from
4888 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4889 Previous value of @var{register} is CFA + @var{offset}.
4891 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4892 Previous value of @var{register} is saved at offset @var{offset} from
4893 the current CFA register. This is transformed to @code{.cfi_offset}
4894 using the known displacement of the CFA register from the CFA.
4895 This is often easier to use, because the number will match the
4896 code it's annotating.
4898 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4899 Previous value of @var{register1} is saved in register @var{register2}.
4901 @subsection @code{.cfi_restore @var{register}}
4902 @code{.cfi_restore} says that the rule for @var{register} is now the
4903 same as it was at the beginning of the function, after all initial
4904 instruction added by @code{.cfi_startproc} were executed.
4906 @subsection @code{.cfi_undefined @var{register}}
4907 From now on the previous value of @var{register} can't be restored anymore.
4909 @subsection @code{.cfi_same_value @var{register}}
4910 Current value of @var{register} is the same like in the previous frame,
4911 i.e. no restoration needed.
4913 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4914 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4915 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4916 places them in the current row. This is useful for situations where you have
4917 multiple @code{.cfi_*} directives that need to be undone due to the control
4918 flow of the program. For example, we could have something like this (assuming
4919 the CFA is the value of @code{rbp}):
4929 .cfi_def_cfa %rsp, 8
4932 /* Do something else */
4935 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4936 to the instructions before @code{label}. This means we'd have to add multiple
4937 @code{.cfi} directives after @code{label} to recreate the original save
4938 locations of the registers, as well as setting the CFA back to the value of
4939 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4951 .cfi_def_cfa %rsp, 8
4955 /* Do something else */
4958 That way, the rules for the instructions after @code{label} will be the same
4959 as before the first @code{.cfi_restore} without having to use multiple
4960 @code{.cfi} directives.
4962 @subsection @code{.cfi_return_column @var{register}}
4963 Change return column @var{register}, i.e. the return address is either
4964 directly in @var{register} or can be accessed by rules for @var{register}.
4966 @subsection @code{.cfi_signal_frame}
4967 Mark current function as signal trampoline.
4969 @subsection @code{.cfi_window_save}
4970 SPARC register window has been saved.
4972 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4973 Allows the user to add arbitrary bytes to the unwind info. One
4974 might use this to add OS-specific CFI opcodes, or generic CFI
4975 opcodes that GAS does not yet support.
4977 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4978 The current value of @var{register} is @var{label}. The value of @var{label}
4979 will be encoded in the output file according to @var{encoding}; see the
4980 description of @code{.cfi_personality} for details on this encoding.
4982 The usefulness of equating a register to a fixed label is probably
4983 limited to the return address register. Here, it can be useful to
4984 mark a code segment that has only one return address which is reached
4985 by a direct branch and no copy of the return address exists in memory
4986 or another register.
4989 @section @code{.comm @var{symbol} , @var{length} }
4991 @cindex @code{comm} directive
4992 @cindex symbol, common
4993 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4994 common symbol in one object file may be merged with a defined or common symbol
4995 of the same name in another object file. If @code{@value{LD}} does not see a
4996 definition for the symbol--just one or more common symbols--then it will
4997 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4998 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4999 the same name, and they do not all have the same size, it will allocate space
5000 using the largest size.
5003 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
5004 an optional third argument. This is the desired alignment of the symbol,
5005 specified for ELF as a byte boundary (for example, an alignment of 16 means
5006 that the least significant 4 bits of the address should be zero), and for PE
5007 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
5008 boundary). The alignment must be an absolute expression, and it must be a
5009 power of two. If @code{@value{LD}} allocates uninitialized memory for the
5010 common symbol, it will use the alignment when placing the symbol. If no
5011 alignment is specified, @command{@value{AS}} will set the alignment to the
5012 largest power of two less than or equal to the size of the symbol, up to a
5013 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
5014 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
5015 @samp{--section-alignment} option; image file sections in PE are aligned to
5016 multiples of 4096, which is far too large an alignment for ordinary variables.
5017 It is rather the default alignment for (non-debug) sections within object
5018 (@samp{*.o}) files, which are less strictly aligned.}.
5022 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
5023 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
5027 @section @code{.data @var{subsection}}
5028 @cindex @code{data} directive
5030 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
5031 end of the data subsection numbered @var{subsection} (which is an
5032 absolute expression). If @var{subsection} is omitted, it defaults
5036 @section @code{.dc[@var{size}] @var{expressions}}
5037 @cindex @code{dc} directive
5039 The @code{.dc} directive expects zero or more @var{expressions} separated by
5040 commas. These expressions are evaluated and their values inserted into the
5041 current section. The size of the emitted value depends upon the suffix to the
5042 @code{.dc} directive:
5046 Emits N-bit values, where N is the size of an address on the target system.
5050 Emits double precision floating-point values.
5052 Emits 32-bit values.
5054 Emits single precision floating-point values.
5056 Emits 16-bit values.
5057 Note - this is true even on targets where the @code{.word} directive would emit
5060 Emits long double precision floating-point values.
5063 If no suffix is used then @samp{.w} is assumed.
5065 The byte ordering is target dependent, as is the size and format of floating
5069 @section @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
5070 @cindex @code{dcb} directive
5071 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5072 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5073 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5074 @var{size} suffix, if present, must be one of:
5078 Emits single byte values.
5080 Emits double-precision floating point values.
5082 Emits 4-byte values.
5084 Emits single-precision floating point values.
5086 Emits 2-byte values.
5088 Emits long double-precision floating point values.
5091 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5093 The byte ordering is target dependent, as is the size and format of floating
5097 @section @code{.ds[@var{size}] @var{number} [,@var{fill}]}
5098 @cindex @code{ds} directive
5099 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5100 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5101 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5102 @var{size} suffix, if present, must be one of:
5106 Emits single byte values.
5108 Emits 8-byte values.
5110 Emits 4-byte values.
5112 Emits 12-byte values.
5114 Emits 4-byte values.
5116 Emits 2-byte values.
5118 Emits 12-byte values.
5121 Note - unlike the @code{.dcb} directive the @samp{.d}, @samp{.s} and @samp{.x}
5122 suffixes do not indicate that floating-point values are to be inserted.
5124 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5126 The byte ordering is target dependent.
5131 @section @code{.def @var{name}}
5133 @cindex @code{def} directive
5134 @cindex COFF symbols, debugging
5135 @cindex debugging COFF symbols
5136 Begin defining debugging information for a symbol @var{name}; the
5137 definition extends until the @code{.endef} directive is encountered.
5142 @section @code{.desc @var{symbol}, @var{abs-expression}}
5144 @cindex @code{desc} directive
5145 @cindex COFF symbol descriptor
5146 @cindex symbol descriptor, COFF
5147 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5148 to the low 16 bits of an absolute expression.
5151 The @samp{.desc} directive is not available when @command{@value{AS}} is
5152 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5153 object format. For the sake of compatibility, @command{@value{AS}} accepts
5154 it, but produces no output, when configured for COFF.
5160 @section @code{.dim}
5162 @cindex @code{dim} directive
5163 @cindex COFF auxiliary symbol information
5164 @cindex auxiliary symbol information, COFF
5165 This directive is generated by compilers to include auxiliary debugging
5166 information in the symbol table. It is only permitted inside
5167 @code{.def}/@code{.endef} pairs.
5171 @section @code{.double @var{flonums}}
5173 @cindex @code{double} directive
5174 @cindex floating point numbers (double)
5175 @code{.double} expects zero or more flonums, separated by commas. It
5176 assembles floating point numbers.
5178 The exact kind of floating point numbers emitted depends on how
5179 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5183 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5184 in @sc{ieee} format.
5189 @section @code{.eject}
5191 @cindex @code{eject} directive
5192 @cindex new page, in listings
5193 @cindex page, in listings
5194 @cindex listing control: new page
5195 Force a page break at this point, when generating assembly listings.
5198 @section @code{.else}
5200 @cindex @code{else} directive
5201 @code{.else} is part of the @command{@value{AS}} support for conditional
5202 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5203 of code to be assembled if the condition for the preceding @code{.if}
5207 @section @code{.elseif}
5209 @cindex @code{elseif} directive
5210 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5211 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5212 @code{.if} block that would otherwise fill the entire @code{.else} section.
5215 @section @code{.end}
5217 @cindex @code{end} directive
5218 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5219 process anything in the file past the @code{.end} directive.
5223 @section @code{.endef}
5225 @cindex @code{endef} directive
5226 This directive flags the end of a symbol definition begun with
5231 @section @code{.endfunc}
5232 @cindex @code{endfunc} directive
5233 @code{.endfunc} marks the end of a function specified with @code{.func}.
5236 @section @code{.endif}
5238 @cindex @code{endif} directive
5239 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5240 it marks the end of a block of code that is only assembled
5241 conditionally. @xref{If,,@code{.if}}.
5244 @section @code{.equ @var{symbol}, @var{expression}}
5246 @cindex @code{equ} directive
5247 @cindex assigning values to symbols
5248 @cindex symbols, assigning values to
5249 This directive sets the value of @var{symbol} to @var{expression}.
5250 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5253 The syntax for @code{equ} on the HPPA is
5254 @samp{@var{symbol} .equ @var{expression}}.
5258 The syntax for @code{equ} on the Z80 is
5259 @samp{@var{symbol} equ @var{expression}}.
5260 On the Z80 it is an error if @var{symbol} is already defined,
5261 but the symbol is not protected from later redefinition.
5262 Compare @ref{Equiv}.
5266 @section @code{.equiv @var{symbol}, @var{expression}}
5267 @cindex @code{equiv} directive
5268 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5269 the assembler will signal an error if @var{symbol} is already defined. Note a
5270 symbol which has been referenced but not actually defined is considered to be
5273 Except for the contents of the error message, this is roughly equivalent to
5280 plus it protects the symbol from later redefinition.
5283 @section @code{.eqv @var{symbol}, @var{expression}}
5284 @cindex @code{eqv} directive
5285 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5286 evaluate the expression or any part of it immediately. Instead each time
5287 the resulting symbol is used in an expression, a snapshot of its current
5291 @section @code{.err}
5292 @cindex @code{err} directive
5293 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5294 message and, unless the @option{-Z} option was used, it will not generate an
5295 object file. This can be used to signal an error in conditionally compiled code.
5298 @section @code{.error "@var{string}"}
5299 @cindex error directive
5301 Similarly to @code{.err}, this directive emits an error, but you can specify a
5302 string that will be emitted as the error message. If you don't specify the
5303 message, it defaults to @code{".error directive invoked in source file"}.
5304 @xref{Errors, ,Error and Warning Messages}.
5307 .error "This code has not been assembled and tested."
5311 @section @code{.exitm}
5312 Exit early from the current macro definition. @xref{Macro}.
5315 @section @code{.extern}
5317 @cindex @code{extern} directive
5318 @code{.extern} is accepted in the source program---for compatibility
5319 with other assemblers---but it is ignored. @command{@value{AS}} treats
5320 all undefined symbols as external.
5323 @section @code{.fail @var{expression}}
5325 @cindex @code{fail} directive
5326 Generates an error or a warning. If the value of the @var{expression} is 500
5327 or more, @command{@value{AS}} will print a warning message. If the value is less
5328 than 500, @command{@value{AS}} will print an error message. The message will
5329 include the value of @var{expression}. This can occasionally be useful inside
5330 complex nested macros or conditional assembly.
5333 @section @code{.file}
5334 @cindex @code{file} directive
5336 @ifclear no-file-dir
5337 There are two different versions of the @code{.file} directive. Targets
5338 that support DWARF2 line number information use the DWARF2 version of
5339 @code{.file}. Other targets use the default version.
5341 @subheading Default Version
5343 @cindex logical file name
5344 @cindex file name, logical
5345 This version of the @code{.file} directive tells @command{@value{AS}} that we
5346 are about to start a new logical file. The syntax is:
5352 @var{string} is the new file name. In general, the filename is
5353 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5354 to specify an empty file name, you must give the quotes--@code{""}. This
5355 statement may go away in future: it is only recognized to be compatible with
5356 old @command{@value{AS}} programs.
5358 @subheading DWARF2 Version
5361 When emitting DWARF2 line number information, @code{.file} assigns filenames
5362 to the @code{.debug_line} file name table. The syntax is:
5365 .file @var{fileno} @var{filename}
5368 The @var{fileno} operand should be a unique positive integer to use as the
5369 index of the entry in the table. The @var{filename} operand is a C string
5370 literal enclosed in double quotes. The @var{filename} can include directory
5371 elements. If it does, then the directory will be added to the directory table
5372 and the basename will be added to the file table.
5374 The detail of filename indices is exposed to the user because the filename
5375 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5376 information, and thus the user must know the exact indices that table
5379 If DWARF-5 support has been enabled via the @option{-gdwarf-5} option then
5380 an extended version of the @code{file} is also allowed:
5383 .file @var{fileno} [@var{dirname}] @var{filename} [md5 @var{value}]
5386 With this version a separate directory name is allowed, although if this is
5387 used then @var{filename} should not contain any directory components. In
5388 addtion an md5 hash value of the contents of @var{filename} can be provided.
5389 This will be stored in the the file table as well, and can be used by tools
5390 reading the debug information to verify that the contents of the source file
5391 match the contents of the compiled file.
5394 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5396 @cindex @code{fill} directive
5397 @cindex writing patterns in memory
5398 @cindex patterns, writing in memory
5399 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5400 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5401 may be zero or more. @var{Size} may be zero or more, but if it is
5402 more than 8, then it is deemed to have the value 8, compatible with
5403 other people's assemblers. The contents of each @var{repeat} bytes
5404 is taken from an 8-byte number. The highest order 4 bytes are
5405 zero. The lowest order 4 bytes are @var{value} rendered in the
5406 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5407 Each @var{size} bytes in a repetition is taken from the lowest order
5408 @var{size} bytes of this number. Again, this bizarre behavior is
5409 compatible with other people's assemblers.
5411 @var{size} and @var{value} are optional.
5412 If the second comma and @var{value} are absent, @var{value} is
5413 assumed zero. If the first comma and following tokens are absent,
5414 @var{size} is assumed to be 1.
5417 @section @code{.float @var{flonums}}
5419 @cindex floating point numbers (single)
5420 @cindex @code{float} directive
5421 This directive assembles zero or more flonums, separated by commas. It
5422 has the same effect as @code{.single}.
5424 The exact kind of floating point numbers emitted depends on how
5425 @command{@value{AS}} is configured.
5426 @xref{Machine Dependencies}.
5430 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5431 in @sc{ieee} format.
5436 @section @code{.func @var{name}[,@var{label}]}
5437 @cindex @code{func} directive
5438 @code{.func} emits debugging information to denote function @var{name}, and
5439 is ignored unless the file is assembled with debugging enabled.
5440 Only @samp{--gstabs[+]} is currently supported.
5441 @var{label} is the entry point of the function and if omitted @var{name}
5442 prepended with the @samp{leading char} is used.
5443 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5444 All functions are currently defined to have @code{void} return type.
5445 The function must be terminated with @code{.endfunc}.
5448 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5450 @cindex @code{global} directive
5451 @cindex symbol, making visible to linker
5452 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5453 @var{symbol} in your partial program, its value is made available to
5454 other partial programs that are linked with it. Otherwise,
5455 @var{symbol} takes its attributes from a symbol of the same name
5456 from another file linked into the same program.
5458 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5459 compatibility with other assemblers.
5462 On the HPPA, @code{.global} is not always enough to make it accessible to other
5463 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5464 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5469 @section @code{.gnu_attribute @var{tag},@var{value}}
5470 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5473 @section @code{.hidden @var{names}}
5475 @cindex @code{hidden} directive
5477 This is one of the ELF visibility directives. The other two are
5478 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5479 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5481 This directive overrides the named symbols default visibility (which is set by
5482 their binding: local, global or weak). The directive sets the visibility to
5483 @code{hidden} which means that the symbols are not visible to other components.
5484 Such symbols are always considered to be @code{protected} as well.
5488 @section @code{.hword @var{expressions}}
5490 @cindex @code{hword} directive
5491 @cindex integers, 16-bit
5492 @cindex numbers, 16-bit
5493 @cindex sixteen bit integers
5494 This expects zero or more @var{expressions}, and emits
5495 a 16 bit number for each.
5498 This directive is a synonym for @samp{.short}; depending on the target
5499 architecture, it may also be a synonym for @samp{.word}.
5503 This directive is a synonym for @samp{.short}.
5506 This directive is a synonym for both @samp{.short} and @samp{.word}.
5511 @section @code{.ident}
5513 @cindex @code{ident} directive
5515 This directive is used by some assemblers to place tags in object files. The
5516 behavior of this directive varies depending on the target. When using the
5517 a.out object file format, @command{@value{AS}} simply accepts the directive for
5518 source-file compatibility with existing assemblers, but does not emit anything
5519 for it. When using COFF, comments are emitted to the @code{.comment} or
5520 @code{.rdata} section, depending on the target. When using ELF, comments are
5521 emitted to the @code{.comment} section.
5524 @section @code{.if @var{absolute expression}}
5526 @cindex conditional assembly
5527 @cindex @code{if} directive
5528 @code{.if} marks the beginning of a section of code which is only
5529 considered part of the source program being assembled if the argument
5530 (which must be an @var{absolute expression}) is non-zero. The end of
5531 the conditional section of code must be marked by @code{.endif}
5532 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5533 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5534 If you have several conditions to check, @code{.elseif} may be used to avoid
5535 nesting blocks if/else within each subsequent @code{.else} block.
5537 The following variants of @code{.if} are also supported:
5539 @cindex @code{ifdef} directive
5540 @item .ifdef @var{symbol}
5541 Assembles the following section of code if the specified @var{symbol}
5542 has been defined. Note a symbol which has been referenced but not yet defined
5543 is considered to be undefined.
5545 @cindex @code{ifb} directive
5546 @item .ifb @var{text}
5547 Assembles the following section of code if the operand is blank (empty).
5549 @cindex @code{ifc} directive
5550 @item .ifc @var{string1},@var{string2}
5551 Assembles the following section of code if the two strings are the same. The
5552 strings may be optionally quoted with single quotes. If they are not quoted,
5553 the first string stops at the first comma, and the second string stops at the
5554 end of the line. Strings which contain whitespace should be quoted. The
5555 string comparison is case sensitive.
5557 @cindex @code{ifeq} directive
5558 @item .ifeq @var{absolute expression}
5559 Assembles the following section of code if the argument is zero.
5561 @cindex @code{ifeqs} directive
5562 @item .ifeqs @var{string1},@var{string2}
5563 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5565 @cindex @code{ifge} directive
5566 @item .ifge @var{absolute expression}
5567 Assembles the following section of code if the argument is greater than or
5570 @cindex @code{ifgt} directive
5571 @item .ifgt @var{absolute expression}
5572 Assembles the following section of code if the argument is greater than zero.
5574 @cindex @code{ifle} directive
5575 @item .ifle @var{absolute expression}
5576 Assembles the following section of code if the argument is less than or equal
5579 @cindex @code{iflt} directive
5580 @item .iflt @var{absolute expression}
5581 Assembles the following section of code if the argument is less than zero.
5583 @cindex @code{ifnb} directive
5584 @item .ifnb @var{text}
5585 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5586 following section of code if the operand is non-blank (non-empty).
5588 @cindex @code{ifnc} directive
5589 @item .ifnc @var{string1},@var{string2}.
5590 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5591 following section of code if the two strings are not the same.
5593 @cindex @code{ifndef} directive
5594 @cindex @code{ifnotdef} directive
5595 @item .ifndef @var{symbol}
5596 @itemx .ifnotdef @var{symbol}
5597 Assembles the following section of code if the specified @var{symbol}
5598 has not been defined. Both spelling variants are equivalent. Note a symbol
5599 which has been referenced but not yet defined is considered to be undefined.
5601 @cindex @code{ifne} directive
5602 @item .ifne @var{absolute expression}
5603 Assembles the following section of code if the argument is not equal to zero
5604 (in other words, this is equivalent to @code{.if}).
5606 @cindex @code{ifnes} directive
5607 @item .ifnes @var{string1},@var{string2}
5608 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5609 following section of code if the two strings are not the same.
5613 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5615 @cindex @code{incbin} directive
5616 @cindex binary files, including
5617 The @code{incbin} directive includes @var{file} verbatim at the current
5618 location. You can control the search paths used with the @samp{-I} command-line
5619 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5622 The @var{skip} argument skips a number of bytes from the start of the
5623 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5624 read. Note that the data is not aligned in any way, so it is the user's
5625 responsibility to make sure that proper alignment is provided both before and
5626 after the @code{incbin} directive.
5629 @section @code{.include "@var{file}"}
5631 @cindex @code{include} directive
5632 @cindex supporting files, including
5633 @cindex files, including
5634 This directive provides a way to include supporting files at specified
5635 points in your source program. The code from @var{file} is assembled as
5636 if it followed the point of the @code{.include}; when the end of the
5637 included file is reached, assembly of the original file continues. You
5638 can control the search paths used with the @samp{-I} command-line option
5639 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5643 @section @code{.int @var{expressions}}
5645 @cindex @code{int} directive
5646 @cindex integers, 32-bit
5647 Expect zero or more @var{expressions}, of any section, separated by commas.
5648 For each expression, emit a number that, at run time, is the value of that
5649 expression. The byte order and bit size of the number depends on what kind
5650 of target the assembly is for.
5654 On most forms of the H8/300, @code{.int} emits 16-bit
5655 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5662 @section @code{.internal @var{names}}
5664 @cindex @code{internal} directive
5666 This is one of the ELF visibility directives. The other two are
5667 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5668 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5670 This directive overrides the named symbols default visibility (which is set by
5671 their binding: local, global or weak). The directive sets the visibility to
5672 @code{internal} which means that the symbols are considered to be @code{hidden}
5673 (i.e., not visible to other components), and that some extra, processor specific
5674 processing must also be performed upon the symbols as well.
5678 @section @code{.irp @var{symbol},@var{values}}@dots{}
5680 @cindex @code{irp} directive
5681 Evaluate a sequence of statements assigning different values to @var{symbol}.
5682 The sequence of statements starts at the @code{.irp} directive, and is
5683 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5684 set to @var{value}, and the sequence of statements is assembled. If no
5685 @var{value} is listed, the sequence of statements is assembled once, with
5686 @var{symbol} set to the null string. To refer to @var{symbol} within the
5687 sequence of statements, use @var{\symbol}.
5689 For example, assembling
5697 is equivalent to assembling
5705 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5708 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5710 @cindex @code{irpc} directive
5711 Evaluate a sequence of statements assigning different values to @var{symbol}.
5712 The sequence of statements starts at the @code{.irpc} directive, and is
5713 terminated by an @code{.endr} directive. For each character in @var{value},
5714 @var{symbol} is set to the character, and the sequence of statements is
5715 assembled. If no @var{value} is listed, the sequence of statements is
5716 assembled once, with @var{symbol} set to the null string. To refer to
5717 @var{symbol} within the sequence of statements, use @var{\symbol}.
5719 For example, assembling
5727 is equivalent to assembling
5735 For some caveats with the spelling of @var{symbol}, see also the discussion
5739 @section @code{.lcomm @var{symbol} , @var{length}}
5741 @cindex @code{lcomm} directive
5742 @cindex local common symbols
5743 @cindex symbols, local common
5744 Reserve @var{length} (an absolute expression) bytes for a local common
5745 denoted by @var{symbol}. The section and value of @var{symbol} are
5746 those of the new local common. The addresses are allocated in the bss
5747 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5748 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5749 not visible to @code{@value{LD}}.
5752 Some targets permit a third argument to be used with @code{.lcomm}. This
5753 argument specifies the desired alignment of the symbol in the bss section.
5757 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5758 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5762 @section @code{.lflags}
5764 @cindex @code{lflags} directive (ignored)
5765 @command{@value{AS}} accepts this directive, for compatibility with other
5766 assemblers, but ignores it.
5768 @ifclear no-line-dir
5770 @section @code{.line @var{line-number}}
5772 @cindex @code{line} directive
5773 @cindex logical line number
5775 Change the logical line number. @var{line-number} must be an absolute
5776 expression. The next line has that logical line number. Therefore any other
5777 statements on the current line (after a statement separator character) are
5778 reported as on logical line number @var{line-number} @minus{} 1. One day
5779 @command{@value{AS}} will no longer support this directive: it is recognized only
5780 for compatibility with existing assembler programs.
5783 Even though this is a directive associated with the @code{a.out} or
5784 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5785 when producing COFF output, and treats @samp{.line} as though it
5786 were the COFF @samp{.ln} @emph{if} it is found outside a
5787 @code{.def}/@code{.endef} pair.
5789 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5790 used by compilers to generate auxiliary symbol information for
5795 @section @code{.linkonce [@var{type}]}
5797 @cindex @code{linkonce} directive
5798 @cindex common sections
5799 Mark the current section so that the linker only includes a single copy of it.
5800 This may be used to include the same section in several different object files,
5801 but ensure that the linker will only include it once in the final output file.
5802 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5803 Duplicate sections are detected based on the section name, so it should be
5806 This directive is only supported by a few object file formats; as of this
5807 writing, the only object file format which supports it is the Portable
5808 Executable format used on Windows NT.
5810 The @var{type} argument is optional. If specified, it must be one of the
5811 following strings. For example:
5815 Not all types may be supported on all object file formats.
5819 Silently discard duplicate sections. This is the default.
5822 Warn if there are duplicate sections, but still keep only one copy.
5825 Warn if any of the duplicates have different sizes.
5828 Warn if any of the duplicates do not have exactly the same contents.
5832 @section @code{.list}
5834 @cindex @code{list} directive
5835 @cindex listing control, turning on
5836 Control (in conjunction with the @code{.nolist} directive) whether or
5837 not assembly listings are generated. These two directives maintain an
5838 internal counter (which is zero initially). @code{.list} increments the
5839 counter, and @code{.nolist} decrements it. Assembly listings are
5840 generated whenever the counter is greater than zero.
5842 By default, listings are disabled. When you enable them (with the
5843 @samp{-a} command-line option; @pxref{Invoking,,Command-Line Options}),
5844 the initial value of the listing counter is one.
5847 @section @code{.ln @var{line-number}}
5849 @cindex @code{ln} directive
5850 @ifclear no-line-dir
5851 @samp{.ln} is a synonym for @samp{.line}.
5854 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5855 must be an absolute expression. The next line has that logical
5856 line number, so any other statements on the current line (after a
5857 statement separator character @code{;}) are reported as on logical
5858 line number @var{line-number} @minus{} 1.
5862 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5863 @cindex @code{loc} directive
5864 When emitting DWARF2 line number information,
5865 the @code{.loc} directive will add a row to the @code{.debug_line} line
5866 number matrix corresponding to the immediately following assembly
5867 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5868 arguments will be applied to the @code{.debug_line} state machine before
5871 The @var{options} are a sequence of the following tokens in any order:
5875 This option will set the @code{basic_block} register in the
5876 @code{.debug_line} state machine to @code{true}.
5879 This option will set the @code{prologue_end} register in the
5880 @code{.debug_line} state machine to @code{true}.
5882 @item epilogue_begin
5883 This option will set the @code{epilogue_begin} register in the
5884 @code{.debug_line} state machine to @code{true}.
5886 @item is_stmt @var{value}
5887 This option will set the @code{is_stmt} register in the
5888 @code{.debug_line} state machine to @code{value}, which must be
5891 @item isa @var{value}
5892 This directive will set the @code{isa} register in the @code{.debug_line}
5893 state machine to @var{value}, which must be an unsigned integer.
5895 @item discriminator @var{value}
5896 This directive will set the @code{discriminator} register in the @code{.debug_line}
5897 state machine to @var{value}, which must be an unsigned integer.
5899 @item view @var{value}
5900 This option causes a row to be added to @code{.debug_line} in reference to the
5901 current address (which might not be the same as that of the following assembly
5902 instruction), and to associate @var{value} with the @code{view} register in the
5903 @code{.debug_line} state machine. If @var{value} is a label, both the
5904 @code{view} register and the label are set to the number of prior @code{.loc}
5905 directives at the same program location. If @var{value} is the literal
5906 @code{0}, the @code{view} register is set to zero, and the assembler asserts
5907 that there aren't any prior @code{.loc} directives at the same program
5908 location. If @var{value} is the literal @code{-0}, the assembler arrange for
5909 the @code{view} register to be reset in this row, even if there are prior
5910 @code{.loc} directives at the same program location.
5914 @node Loc_mark_labels
5915 @section @code{.loc_mark_labels @var{enable}}
5916 @cindex @code{loc_mark_labels} directive
5917 When emitting DWARF2 line number information,
5918 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5919 to the @code{.debug_line} line number matrix with the @code{basic_block}
5920 register in the state machine set whenever a code label is seen.
5921 The @var{enable} argument should be either 1 or 0, to enable or disable
5922 this function respectively.
5926 @section @code{.local @var{names}}
5928 @cindex @code{local} directive
5929 This directive, which is available for ELF targets, marks each symbol in
5930 the comma-separated list of @code{names} as a local symbol so that it
5931 will not be externally visible. If the symbols do not already exist,
5932 they will be created.
5934 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5935 accept an alignment argument, which is the case for most ELF targets,
5936 the @code{.local} directive can be used in combination with @code{.comm}
5937 (@pxref{Comm}) to define aligned local common data.
5941 @section @code{.long @var{expressions}}
5943 @cindex @code{long} directive
5944 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5947 @c no one seems to know what this is for or whether this description is
5948 @c what it really ought to do
5950 @section @code{.lsym @var{symbol}, @var{expression}}
5952 @cindex @code{lsym} directive
5953 @cindex symbol, not referenced in assembly
5954 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5955 the hash table, ensuring it cannot be referenced by name during the
5956 rest of the assembly. This sets the attributes of the symbol to be
5957 the same as the expression value:
5959 @var{other} = @var{descriptor} = 0
5960 @var{type} = @r{(section of @var{expression})}
5961 @var{value} = @var{expression}
5964 The new symbol is not flagged as external.
5968 @section @code{.macro}
5971 The commands @code{.macro} and @code{.endm} allow you to define macros that
5972 generate assembly output. For example, this definition specifies a macro
5973 @code{sum} that puts a sequence of numbers into memory:
5976 .macro sum from=0, to=5
5985 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5997 @item .macro @var{macname}
5998 @itemx .macro @var{macname} @var{macargs} @dots{}
5999 @cindex @code{macro} directive
6000 Begin the definition of a macro called @var{macname}. If your macro
6001 definition requires arguments, specify their names after the macro name,
6002 separated by commas or spaces. You can qualify the macro argument to
6003 indicate whether all invocations must specify a non-blank value (through
6004 @samp{:@code{req}}), or whether it takes all of the remaining arguments
6005 (through @samp{:@code{vararg}}). You can supply a default value for any
6006 macro argument by following the name with @samp{=@var{deflt}}. You
6007 cannot define two macros with the same @var{macname} unless it has been
6008 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
6009 definitions. For example, these are all valid @code{.macro} statements:
6013 Begin the definition of a macro called @code{comm}, which takes no
6016 @item .macro plus1 p, p1
6017 @itemx .macro plus1 p p1
6018 Either statement begins the definition of a macro called @code{plus1},
6019 which takes two arguments; within the macro definition, write
6020 @samp{\p} or @samp{\p1} to evaluate the arguments.
6022 @item .macro reserve_str p1=0 p2
6023 Begin the definition of a macro called @code{reserve_str}, with two
6024 arguments. The first argument has a default value, but not the second.
6025 After the definition is complete, you can call the macro either as
6026 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
6027 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
6028 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
6029 @samp{0}, and @samp{\p2} evaluating to @var{b}).
6031 @item .macro m p1:req, p2=0, p3:vararg
6032 Begin the definition of a macro called @code{m}, with at least three
6033 arguments. The first argument must always have a value specified, but
6034 not the second, which instead has a default value. The third formal
6035 will get assigned all remaining arguments specified at invocation time.
6037 When you call a macro, you can specify the argument values either by
6038 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
6039 @samp{sum to=17, from=9}.
6043 Note that since each of the @var{macargs} can be an identifier exactly
6044 as any other one permitted by the target architecture, there may be
6045 occasional problems if the target hand-crafts special meanings to certain
6046 characters when they occur in a special position. For example, if the colon
6047 (@code{:}) is generally permitted to be part of a symbol name, but the
6048 architecture specific code special-cases it when occurring as the final
6049 character of a symbol (to denote a label), then the macro parameter
6050 replacement code will have no way of knowing that and consider the whole
6051 construct (including the colon) an identifier, and check only this
6052 identifier for being the subject to parameter substitution. So for example
6053 this macro definition:
6061 might not work as expected. Invoking @samp{label foo} might not create a label
6062 called @samp{foo} but instead just insert the text @samp{\l:} into the
6063 assembler source, probably generating an error about an unrecognised
6066 Similarly problems might occur with the period character (@samp{.})
6067 which is often allowed inside opcode names (and hence identifier names). So
6068 for example constructing a macro to build an opcode from a base name and a
6069 length specifier like this:
6072 .macro opcode base length
6077 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
6078 instruction but instead generate some kind of error as the assembler tries to
6079 interpret the text @samp{\base.\length}.
6081 There are several possible ways around this problem:
6084 @item Insert white space
6085 If it is possible to use white space characters then this is the simplest
6094 @item Use @samp{\()}
6095 The string @samp{\()} can be used to separate the end of a macro argument from
6096 the following text. eg:
6099 .macro opcode base length
6104 @item Use the alternate macro syntax mode
6105 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
6106 used as a separator. eg:
6116 Note: this problem of correctly identifying string parameters to pseudo ops
6117 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
6118 and @code{.irpc} (@pxref{Irpc}) as well.
6121 @cindex @code{endm} directive
6122 Mark the end of a macro definition.
6125 @cindex @code{exitm} directive
6126 Exit early from the current macro definition.
6128 @cindex number of macros executed
6129 @cindex macros, count executed
6131 @command{@value{AS}} maintains a counter of how many macros it has
6132 executed in this pseudo-variable; you can copy that number to your
6133 output with @samp{\@@}, but @emph{only within a macro definition}.
6135 @item LOCAL @var{name} [ , @dots{} ]
6136 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
6137 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
6138 @xref{Altmacro,,@code{.altmacro}}.
6142 @section @code{.mri @var{val}}
6144 @cindex @code{mri} directive
6145 @cindex MRI mode, temporarily
6146 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
6147 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
6148 affects code assembled until the next @code{.mri} directive, or until the end
6149 of the file. @xref{M, MRI mode, MRI mode}.
6152 @section @code{.noaltmacro}
6153 Disable alternate macro mode. @xref{Altmacro}.
6156 @section @code{.nolist}
6158 @cindex @code{nolist} directive
6159 @cindex listing control, turning off
6160 Control (in conjunction with the @code{.list} directive) whether or
6161 not assembly listings are generated. These two directives maintain an
6162 internal counter (which is zero initially). @code{.list} increments the
6163 counter, and @code{.nolist} decrements it. Assembly listings are
6164 generated whenever the counter is greater than zero.
6167 @section @code{.nops @var{size}[, @var{control}]}
6169 @cindex @code{nops} directive
6170 @cindex filling memory with no-op instructions
6171 This directive emits @var{size} bytes filled with no-op instructions.
6172 @var{size} is absolute expression, which must be a positve value.
6173 @var{control} controls how no-op instructions should be generated. If
6174 the comma and @var{control} are omitted, @var{control} is assumed to be
6177 Note: For Intel 80386 and AMD x86-64 targets, @var{control} specifies
6178 the size limit of a no-op instruction. The valid values of @var{control}
6179 are between 0 and 4 in 16-bit mode, between 0 and 7 when tuning for
6180 older processors in 32-bit mode, between 0 and 11 in 64-bit mode or when
6181 tuning for newer processors in 32-bit mode. When 0 is used, the no-op
6182 instruction size limit is set to the maximum supported size.
6185 @section @code{.octa @var{bignums}}
6187 @c FIXME: double size emitted for "octa" on some? Or warn?
6188 @cindex @code{octa} directive
6189 @cindex integer, 16-byte
6190 @cindex sixteen byte integer
6191 This directive expects zero or more bignums, separated by commas. For each
6192 bignum, it emits a 16-byte integer.
6194 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6195 hence @emph{octa}-word for 16 bytes.
6198 @section @code{.offset @var{loc}}
6200 @cindex @code{offset} directive
6201 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6202 be an absolute expression. This directive may be useful for defining
6203 symbols with absolute values. Do not confuse it with the @code{.org}
6207 @section @code{.org @var{new-lc} , @var{fill}}
6209 @cindex @code{org} directive
6210 @cindex location counter, advancing
6211 @cindex advancing location counter
6212 @cindex current address, advancing
6213 Advance the location counter of the current section to
6214 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6215 expression with the same section as the current subsection. That is,
6216 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6217 wrong section, the @code{.org} directive is ignored. To be compatible
6218 with former assemblers, if the section of @var{new-lc} is absolute,
6219 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6220 is the same as the current subsection.
6222 @code{.org} may only increase the location counter, or leave it
6223 unchanged; you cannot use @code{.org} to move the location counter
6226 @c double negative used below "not undefined" because this is a specific
6227 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6228 @c section. doc@cygnus.com 18feb91
6229 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6230 may not be undefined. If you really detest this restriction we eagerly await
6231 a chance to share your improved assembler.
6233 Beware that the origin is relative to the start of the section, not
6234 to the start of the subsection. This is compatible with other
6235 people's assemblers.
6237 When the location counter (of the current subsection) is advanced, the
6238 intervening bytes are filled with @var{fill} which should be an
6239 absolute expression. If the comma and @var{fill} are omitted,
6240 @var{fill} defaults to zero.
6243 @section @code{.p2align[wl] [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
6245 @cindex padding the location counter given a power of two
6246 @cindex @code{p2align} directive
6247 Pad the location counter (in the current subsection) to a particular
6248 storage boundary. The first expression (which must be absolute) is the
6249 number of low-order zero bits the location counter must have after
6250 advancement. For example @samp{.p2align 3} advances the location
6251 counter until it is a multiple of 8. If the location counter is already a
6252 multiple of 8, no change is needed. If the expression is omitted then a
6253 default value of 0 is used, effectively disabling alignment requirements.
6255 The second expression (also absolute) gives the fill value to be stored in the
6256 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6257 padding bytes are normally zero. However, on most systems, if the section is
6258 marked as containing code and the fill value is omitted, the space is filled
6259 with no-op instructions.
6261 The third expression is also absolute, and is also optional. If it is present,
6262 it is the maximum number of bytes that should be skipped by this alignment
6263 directive. If doing the alignment would require skipping more bytes than the
6264 specified maximum, then the alignment is not done at all. You can omit the
6265 fill value (the second argument) entirely by simply using two commas after the
6266 required alignment; this can be useful if you want the alignment to be filled
6267 with no-op instructions when appropriate.
6269 @cindex @code{p2alignw} directive
6270 @cindex @code{p2alignl} directive
6271 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6272 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6273 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6274 fill pattern as a four byte longword value. For example, @code{.p2alignw
6275 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6276 filled in with the value 0x368d (the exact placement of the bytes depends upon
6277 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6282 @section @code{.popsection}
6284 @cindex @code{popsection} directive
6285 @cindex Section Stack
6286 This is one of the ELF section stack manipulation directives. The others are
6287 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6288 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6291 This directive replaces the current section (and subsection) with the top
6292 section (and subsection) on the section stack. This section is popped off the
6298 @section @code{.previous}
6300 @cindex @code{previous} directive
6301 @cindex Section Stack
6302 This is one of the ELF section stack manipulation directives. The others are
6303 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6304 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6305 (@pxref{PopSection}).
6307 This directive swaps the current section (and subsection) with most recently
6308 referenced section/subsection pair prior to this one. Multiple
6309 @code{.previous} directives in a row will flip between two sections (and their
6310 subsections). For example:
6322 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6328 # Now in section A subsection 1
6332 # Now in section B subsection 0
6335 # Now in section B subsection 1
6338 # Now in section B subsection 0
6342 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6343 section B and 0x9abc into subsection 1 of section B.
6345 In terms of the section stack, this directive swaps the current section with
6346 the top section on the section stack.
6350 @section @code{.print @var{string}}
6352 @cindex @code{print} directive
6353 @command{@value{AS}} will print @var{string} on the standard output during
6354 assembly. You must put @var{string} in double quotes.
6358 @section @code{.protected @var{names}}
6360 @cindex @code{protected} directive
6362 This is one of the ELF visibility directives. The other two are
6363 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6365 This directive overrides the named symbols default visibility (which is set by
6366 their binding: local, global or weak). The directive sets the visibility to
6367 @code{protected} which means that any references to the symbols from within the
6368 components that defines them must be resolved to the definition in that
6369 component, even if a definition in another component would normally preempt
6374 @section @code{.psize @var{lines} , @var{columns}}
6376 @cindex @code{psize} directive
6377 @cindex listing control: paper size
6378 @cindex paper size, for listings
6379 Use this directive to declare the number of lines---and, optionally, the
6380 number of columns---to use for each page, when generating listings.
6382 If you do not use @code{.psize}, listings use a default line-count
6383 of 60. You may omit the comma and @var{columns} specification; the
6384 default width is 200 columns.
6386 @command{@value{AS}} generates formfeeds whenever the specified number of
6387 lines is exceeded (or whenever you explicitly request one, using
6390 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6391 those explicitly specified with @code{.eject}.
6394 @section @code{.purgem @var{name}}
6396 @cindex @code{purgem} directive
6397 Undefine the macro @var{name}, so that later uses of the string will not be
6398 expanded. @xref{Macro}.
6402 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6404 @cindex @code{pushsection} directive
6405 @cindex Section Stack
6406 This is one of the ELF section stack manipulation directives. The others are
6407 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6408 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6411 This directive pushes the current section (and subsection) onto the
6412 top of the section stack, and then replaces the current section and
6413 subsection with @code{name} and @code{subsection}. The optional
6414 @code{flags}, @code{type} and @code{arguments} are treated the same
6415 as in the @code{.section} (@pxref{Section}) directive.
6419 @section @code{.quad @var{bignums}}
6421 @cindex @code{quad} directive
6422 @code{.quad} expects zero or more bignums, separated by commas. For
6423 each bignum, it emits
6425 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6426 warning message; and just takes the lowest order 8 bytes of the bignum.
6427 @cindex eight-byte integer
6428 @cindex integer, 8-byte
6430 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6431 hence @emph{quad}-word for 8 bytes.
6434 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6435 warning message; and just takes the lowest order 16 bytes of the bignum.
6436 @cindex sixteen-byte integer
6437 @cindex integer, 16-byte
6441 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6443 @cindex @code{reloc} directive
6444 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6445 @var{expression}. If @var{offset} is a number, the relocation is generated in
6446 the current section. If @var{offset} is an expression that resolves to a
6447 symbol plus offset, the relocation is generated in the given symbol's section.
6448 @var{expression}, if present, must resolve to a symbol plus addend or to an
6449 absolute value, but note that not all targets support an addend. e.g. ELF REL
6450 targets such as i386 store an addend in the section contents rather than in the
6451 relocation. This low level interface does not support addends stored in the
6455 @section @code{.rept @var{count}}
6457 @cindex @code{rept} directive
6458 Repeat the sequence of lines between the @code{.rept} directive and the next
6459 @code{.endr} directive @var{count} times.
6461 For example, assembling
6469 is equivalent to assembling
6477 A count of zero is allowed, but nothing is generated. Negative counts are not
6478 allowed and if encountered will be treated as if they were zero.
6481 @section @code{.sbttl "@var{subheading}"}
6483 @cindex @code{sbttl} directive
6484 @cindex subtitles for listings
6485 @cindex listing control: subtitle
6486 Use @var{subheading} as the title (third line, immediately after the
6487 title line) when generating assembly listings.
6489 This directive affects subsequent pages, as well as the current page if
6490 it appears within ten lines of the top of a page.
6494 @section @code{.scl @var{class}}
6496 @cindex @code{scl} directive
6497 @cindex symbol storage class (COFF)
6498 @cindex COFF symbol storage class
6499 Set the storage-class value for a symbol. This directive may only be
6500 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6501 whether a symbol is static or external, or it may record further
6502 symbolic debugging information.
6507 @section @code{.section @var{name}}
6509 @cindex named section
6510 Use the @code{.section} directive to assemble the following code into a section
6513 This directive is only supported for targets that actually support arbitrarily
6514 named sections; on @code{a.out} targets, for example, it is not accepted, even
6515 with a standard @code{a.out} section name.
6519 @c only print the extra heading if both COFF and ELF are set
6520 @subheading COFF Version
6523 @cindex @code{section} directive (COFF version)
6524 For COFF targets, the @code{.section} directive is used in one of the following
6528 .section @var{name}[, "@var{flags}"]
6529 .section @var{name}[, @var{subsection}]
6532 If the optional argument is quoted, it is taken as flags to use for the
6533 section. Each flag is a single character. The following flags are recognized:
6537 bss section (uninitialized data)
6539 section is not loaded
6545 exclude section from linking
6551 shared section (meaningful for PE targets)
6553 ignored. (For compatibility with the ELF version)
6555 section is not readable (meaningful for PE targets)
6557 single-digit power-of-two section alignment (GNU extension)
6560 If no flags are specified, the default flags depend upon the section name. If
6561 the section name is not recognized, the default will be for the section to be
6562 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6563 from the section, rather than adding them, so if they are used on their own it
6564 will be as if no flags had been specified at all.
6566 If the optional argument to the @code{.section} directive is not quoted, it is
6567 taken as a subsection number (@pxref{Sub-Sections}).
6572 @c only print the extra heading if both COFF and ELF are set
6573 @subheading ELF Version
6576 @cindex Section Stack
6577 This is one of the ELF section stack manipulation directives. The others are
6578 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6579 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6580 @code{.previous} (@pxref{Previous}).
6582 @cindex @code{section} directive (ELF version)
6583 For ELF targets, the @code{.section} directive is used like this:
6586 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6589 @anchor{Section Name Substitutions}
6590 @kindex --sectname-subst
6591 @cindex section name substitution
6592 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6593 argument may contain a substitution sequence. Only @code{%S} is supported
6594 at the moment, and substitutes the current section name. For example:
6597 .macro exception_code
6598 .section %S.exception
6599 [exception code here]
6614 The two @code{exception_code} invocations above would create the
6615 @code{.text.exception} and @code{.init.exception} sections respectively.
6616 This is useful e.g. to discriminate between ancillary sections that are
6617 tied to setup code to be discarded after use from ancillary sections that
6618 need to stay resident without having to define multiple @code{exception_code}
6619 macros just for that purpose.
6621 The optional @var{flags} argument is a quoted string which may contain any
6622 combination of the following characters:
6626 section is allocatable
6628 section is a GNU_MBIND section
6630 section is excluded from executable and shared library.
6632 section references a symbol defined in another section (the linked-to
6633 section) in the same file.
6637 section is executable
6639 section is mergeable
6641 section contains zero terminated strings
6643 section is a member of a section group
6645 section is used for thread-local-storage
6647 section is a member of the previously-current section's group, if any
6648 @item @code{<number>}
6649 a numeric value indicating the bits to be set in the ELF section header's flags
6650 field. Note - if one or more of the alphabetic characters described above is
6651 also included in the flags field, their bit values will be ORed into the
6653 @item @code{<target specific>}
6654 some targets extend this list with their own flag characters
6657 Note - once a section's flags have been set they cannot be changed. There are
6658 a few exceptions to this rule however. Processor and application specific
6659 flags can be added to an already defined section. The @code{.interp},
6660 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6661 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6662 section may have the executable (@code{x}) flag added.
6664 The optional @var{type} argument may contain one of the following constants:
6668 section contains data
6670 section does not contain data (i.e., section only occupies space)
6672 section contains data which is used by things other than the program
6674 section contains an array of pointers to init functions
6676 section contains an array of pointers to finish functions
6677 @item @@preinit_array
6678 section contains an array of pointers to pre-init functions
6679 @item @@@code{<number>}
6680 a numeric value to be set as the ELF section header's type field.
6681 @item @@@code{<target specific>}
6682 some targets extend this list with their own types
6685 Many targets only support the first three section types. The type may be
6686 enclosed in double quotes if necessary.
6688 Note on targets where the @code{@@} character is the start of a comment (eg
6689 ARM) then another character is used instead. For example the ARM port uses the
6692 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6693 special and have fixed types. Any attempt to declare them with a different
6694 type will generate an error from the assembler.
6696 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6697 be specified as well as an extra argument---@var{entsize}---like this:
6700 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6703 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6704 constants, each @var{entsize} octets long. Sections with both @code{M} and
6705 @code{S} must contain zero terminated strings where each character is
6706 @var{entsize} bytes long. The linker may remove duplicates within sections with
6707 the same name, same entity size and same flags. @var{entsize} must be an
6708 absolute expression. For sections with both @code{M} and @code{S}, a string
6709 which is a suffix of a larger string is considered a duplicate. Thus
6710 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6711 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6713 If @var{flags} contains the @code{o} flag, then the @var{type} argument
6714 must be present along with an additional field like this:
6717 .section @var{name},"@var{flags}"o,@@@var{type},@var{SymbolName}
6720 The @var{SymbolName} field specifies the symbol name which the section
6723 Note: If both the @var{M} and @var{o} flags are present, then the fields
6724 for the Merge flag should come first, like this:
6727 .section @var{name},"@var{flags}"Mo,@@@var{type},@var{entsize},@var{SymbolName}
6730 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6731 be present along with an additional field like this:
6734 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6737 The @var{GroupName} field specifies the name of the section group to which this
6738 particular section belongs. The optional linkage field can contain:
6742 indicates that only one copy of this section should be retained
6747 Note: if both the @var{M} and @var{G} flags are present then the fields for
6748 the Merge flag should come first, like this:
6751 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6754 If both @code{o} flag and @code{G} flag are present, then the
6755 @var{SymbolName} field for @code{o} comes first, like this:
6758 .section @var{name},"@var{flags}"oG,@@@var{type},@var{SymbolName},@var{GroupName}[,@var{linkage}]
6761 If @var{flags} contains the @code{?} symbol then it may not also contain the
6762 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6763 present. Instead, @code{?} says to consider the section that's current before
6764 this directive. If that section used @code{G}, then the new section will use
6765 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6766 If not, then the @code{?} symbol has no effect.
6768 The optional @var{unique,@code{<number>}} argument must come last. It
6769 assigns @var{@code{<number>}} as a unique section ID to distinguish
6770 different sections with the same section name like these:
6773 .section @var{name},"@var{flags}",@@@var{type},@var{unique,@code{<number>}}
6774 .section @var{name},"@var{flags}"G,@@@var{type},@var{GroupName},[@var{linkage}],@var{unique,@code{<number>}}
6775 .section @var{name},"@var{flags}"MG,@@@var{type},@var{entsize},@var{GroupName}[,@var{linkage}],@var{unique,@code{<number>}}
6778 The valid values of @var{@code{<number>}} are between 0 and 4294967295.
6780 If no flags are specified, the default flags depend upon the section name. If
6781 the section name is not recognized, the default will be for the section to have
6782 none of the above flags: it will not be allocated in memory, nor writable, nor
6783 executable. The section will contain data.
6785 For ELF targets, the assembler supports another type of @code{.section}
6786 directive for compatibility with the Solaris assembler:
6789 .section "@var{name}"[, @var{flags}...]
6792 Note that the section name is quoted. There may be a sequence of comma
6797 section is allocatable
6801 section is executable
6803 section is excluded from executable and shared library.
6805 section is used for thread local storage
6808 This directive replaces the current section and subsection. See the
6809 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6810 some examples of how this directive and the other section stack directives
6816 @section @code{.set @var{symbol}, @var{expression}}
6818 @cindex @code{set} directive
6819 @cindex symbol value, setting
6820 Set the value of @var{symbol} to @var{expression}. This
6821 changes @var{symbol}'s value and type to conform to
6822 @var{expression}. If @var{symbol} was flagged as external, it remains
6823 flagged (@pxref{Symbol Attributes}).
6825 You may @code{.set} a symbol many times in the same assembly provided that the
6826 values given to the symbol are constants. Values that are based on expressions
6827 involving other symbols are allowed, but some targets may restrict this to only
6828 being done once per assembly. This is because those targets do not set the
6829 addresses of symbols at assembly time, but rather delay the assignment until a
6830 final link is performed. This allows the linker a chance to change the code in
6831 the files, changing the location of, and the relative distance between, various
6834 If you @code{.set} a global symbol, the value stored in the object
6835 file is the last value stored into it.
6838 On Z80 @code{set} is a real instruction, use @code{.set} or
6839 @samp{@var{symbol} defl @var{expression}} instead.
6843 @section @code{.short @var{expressions}}
6845 @cindex @code{short} directive
6847 @code{.short} is normally the same as @samp{.word}.
6848 @xref{Word,,@code{.word}}.
6850 In some configurations, however, @code{.short} and @code{.word} generate
6851 numbers of different lengths. @xref{Machine Dependencies}.
6855 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6858 This expects zero or more @var{expressions}, and emits
6859 a 16 bit number for each.
6864 @section @code{.single @var{flonums}}
6866 @cindex @code{single} directive
6867 @cindex floating point numbers (single)
6868 This directive assembles zero or more flonums, separated by commas. It
6869 has the same effect as @code{.float}.
6871 The exact kind of floating point numbers emitted depends on how
6872 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6876 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6877 numbers in @sc{ieee} format.
6883 @section @code{.size}
6885 This directive is used to set the size associated with a symbol.
6889 @c only print the extra heading if both COFF and ELF are set
6890 @subheading COFF Version
6893 @cindex @code{size} directive (COFF version)
6894 For COFF targets, the @code{.size} directive is only permitted inside
6895 @code{.def}/@code{.endef} pairs. It is used like this:
6898 .size @var{expression}
6905 @c only print the extra heading if both COFF and ELF are set
6906 @subheading ELF Version
6909 @cindex @code{size} directive (ELF version)
6910 For ELF targets, the @code{.size} directive is used like this:
6913 .size @var{name} , @var{expression}
6916 This directive sets the size associated with a symbol @var{name}.
6917 The size in bytes is computed from @var{expression} which can make use of label
6918 arithmetic. This directive is typically used to set the size of function
6923 @ifclear no-space-dir
6925 @section @code{.skip @var{size} [,@var{fill}]}
6927 @cindex @code{skip} directive
6928 @cindex filling memory
6929 This directive emits @var{size} bytes, each of value @var{fill}. Both
6930 @var{size} and @var{fill} are absolute expressions. If the comma and
6931 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6936 @section @code{.sleb128 @var{expressions}}
6938 @cindex @code{sleb128} directive
6939 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6940 compact, variable length representation of numbers used by the DWARF
6941 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6943 @ifclear no-space-dir
6945 @section @code{.space @var{size} [,@var{fill}]}
6947 @cindex @code{space} directive
6948 @cindex filling memory
6949 This directive emits @var{size} bytes, each of value @var{fill}. Both
6950 @var{size} and @var{fill} are absolute expressions. If the comma
6951 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6956 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6957 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6958 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6959 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6967 @section @code{.stabd, .stabn, .stabs}
6969 @cindex symbolic debuggers, information for
6970 @cindex @code{stab@var{x}} directives
6971 There are three directives that begin @samp{.stab}.
6972 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6973 The symbols are not entered in the @command{@value{AS}} hash table: they
6974 cannot be referenced elsewhere in the source file.
6975 Up to five fields are required:
6979 This is the symbol's name. It may contain any character except
6980 @samp{\000}, so is more general than ordinary symbol names. Some
6981 debuggers used to code arbitrarily complex structures into symbol names
6985 An absolute expression. The symbol's type is set to the low 8 bits of
6986 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6987 and debuggers choke on silly bit patterns.
6990 An absolute expression. The symbol's ``other'' attribute is set to the
6991 low 8 bits of this expression.
6994 An absolute expression. The symbol's descriptor is set to the low 16
6995 bits of this expression.
6998 An absolute expression which becomes the symbol's value.
7001 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
7002 or @code{.stabs} statement, the symbol has probably already been created;
7003 you get a half-formed symbol in your object file. This is
7004 compatible with earlier assemblers!
7007 @cindex @code{stabd} directive
7008 @item .stabd @var{type} , @var{other} , @var{desc}
7010 The ``name'' of the symbol generated is not even an empty string.
7011 It is a null pointer, for compatibility. Older assemblers used a
7012 null pointer so they didn't waste space in object files with empty
7015 The symbol's value is set to the location counter,
7016 relocatably. When your program is linked, the value of this symbol
7017 is the address of the location counter when the @code{.stabd} was
7020 @cindex @code{stabn} directive
7021 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
7022 The name of the symbol is set to the empty string @code{""}.
7024 @cindex @code{stabs} directive
7025 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
7026 All five fields are specified.
7032 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
7033 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
7035 @cindex string, copying to object file
7036 @cindex string8, copying to object file
7037 @cindex string16, copying to object file
7038 @cindex string32, copying to object file
7039 @cindex string64, copying to object file
7040 @cindex @code{string} directive
7041 @cindex @code{string8} directive
7042 @cindex @code{string16} directive
7043 @cindex @code{string32} directive
7044 @cindex @code{string64} directive
7046 Copy the characters in @var{str} to the object file. You may specify more than
7047 one string to copy, separated by commas. Unless otherwise specified for a
7048 particular machine, the assembler marks the end of each string with a 0 byte.
7049 You can use any of the escape sequences described in @ref{Strings,,Strings}.
7051 The variants @code{string16}, @code{string32} and @code{string64} differ from
7052 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
7053 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
7054 are stored in target endianness byte order.
7060 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
7061 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
7066 @section @code{.struct @var{expression}}
7068 @cindex @code{struct} directive
7069 Switch to the absolute section, and set the section offset to @var{expression},
7070 which must be an absolute expression. You might use this as follows:
7079 This would define the symbol @code{field1} to have the value 0, the symbol
7080 @code{field2} to have the value 4, and the symbol @code{field3} to have the
7081 value 8. Assembly would be left in the absolute section, and you would need to
7082 use a @code{.section} directive of some sort to change to some other section
7083 before further assembly.
7087 @section @code{.subsection @var{name}}
7089 @cindex @code{subsection} directive
7090 @cindex Section Stack
7091 This is one of the ELF section stack manipulation directives. The others are
7092 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
7093 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
7096 This directive replaces the current subsection with @code{name}. The current
7097 section is not changed. The replaced subsection is put onto the section stack
7098 in place of the then current top of stack subsection.
7103 @section @code{.symver}
7104 @cindex @code{symver} directive
7105 @cindex symbol versioning
7106 @cindex versions of symbols
7107 Use the @code{.symver} directive to bind symbols to specific version nodes
7108 within a source file. This is only supported on ELF platforms, and is
7109 typically used when assembling files to be linked into a shared library.
7110 There are cases where it may make sense to use this in objects to be bound
7111 into an application itself so as to override a versioned symbol from a
7114 For ELF targets, the @code{.symver} directive can be used like this:
7116 .symver @var{name}, @var{name2@@nodename}[ ,@var{visibility}]
7118 If the original symbol @var{name} is defined within the file
7119 being assembled, the @code{.symver} directive effectively creates a symbol
7120 alias with the name @var{name2@@nodename}, and in fact the main reason that we
7121 just don't try and create a regular alias is that the @var{@@} character isn't
7122 permitted in symbol names. The @var{name2} part of the name is the actual name
7123 of the symbol by which it will be externally referenced. The name @var{name}
7124 itself is merely a name of convenience that is used so that it is possible to
7125 have definitions for multiple versions of a function within a single source
7126 file, and so that the compiler can unambiguously know which version of a
7127 function is being mentioned. The @var{nodename} portion of the alias should be
7128 the name of a node specified in the version script supplied to the linker when
7129 building a shared library. If you are attempting to override a versioned
7130 symbol from a shared library, then @var{nodename} should correspond to the
7131 nodename of the symbol you are trying to override. The optional argument
7132 @var{visibility} updates the visibility of the original symbol. The valid
7133 visibilities are @code{local}, @code{hidden}, and @code{remove}. The
7134 @code{local} visibility makes the original symbol a local symbol
7135 (@pxref{Local}). The @code{hidden} visibility sets the visibility of the
7136 original symbol to @code{hidden} (@pxref{Hidden}). The @code{remove}
7137 visibility removes the original symbol from the symbol table. If visibility
7138 isn't specified, the original symbol is unchanged.
7140 If the symbol @var{name} is not defined within the file being assembled, all
7141 references to @var{name} will be changed to @var{name2@@nodename}. If no
7142 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
7145 Another usage of the @code{.symver} directive is:
7147 .symver @var{name}, @var{name2@@@@nodename}
7149 In this case, the symbol @var{name} must exist and be defined within
7150 the file being assembled. It is similar to @var{name2@@nodename}. The
7151 difference is @var{name2@@@@nodename} will also be used to resolve
7152 references to @var{name2} by the linker.
7154 The third usage of the @code{.symver} directive is:
7156 .symver @var{name}, @var{name2@@@@@@nodename}
7158 When @var{name} is not defined within the
7159 file being assembled, it is treated as @var{name2@@nodename}. When
7160 @var{name} is defined within the file being assembled, the symbol
7161 name, @var{name}, will be changed to @var{name2@@@@nodename}.
7166 @section @code{.tag @var{structname}}
7168 @cindex COFF structure debugging
7169 @cindex structure debugging, COFF
7170 @cindex @code{tag} directive
7171 This directive is generated by compilers to include auxiliary debugging
7172 information in the symbol table. It is only permitted inside
7173 @code{.def}/@code{.endef} pairs. Tags are used to link structure
7174 definitions in the symbol table with instances of those structures.
7178 @section @code{.text @var{subsection}}
7180 @cindex @code{text} directive
7181 Tells @command{@value{AS}} to assemble the following statements onto the end of
7182 the text subsection numbered @var{subsection}, which is an absolute
7183 expression. If @var{subsection} is omitted, subsection number zero
7187 @section @code{.title "@var{heading}"}
7189 @cindex @code{title} directive
7190 @cindex listing control: title line
7191 Use @var{heading} as the title (second line, immediately after the
7192 source file name and pagenumber) when generating assembly listings.
7194 This directive affects subsequent pages, as well as the current page if
7195 it appears within ten lines of the top of a page.
7199 @section @code{.type}
7201 This directive is used to set the type of a symbol.
7205 @c only print the extra heading if both COFF and ELF are set
7206 @subheading COFF Version
7209 @cindex COFF symbol type
7210 @cindex symbol type, COFF
7211 @cindex @code{type} directive (COFF version)
7212 For COFF targets, this directive is permitted only within
7213 @code{.def}/@code{.endef} pairs. It is used like this:
7219 This records the integer @var{int} as the type attribute of a symbol table
7226 @c only print the extra heading if both COFF and ELF are set
7227 @subheading ELF Version
7230 @cindex ELF symbol type
7231 @cindex symbol type, ELF
7232 @cindex @code{type} directive (ELF version)
7233 For ELF targets, the @code{.type} directive is used like this:
7236 .type @var{name} , @var{type description}
7239 This sets the type of symbol @var{name} to be either a
7240 function symbol or an object symbol. There are five different syntaxes
7241 supported for the @var{type description} field, in order to provide
7242 compatibility with various other assemblers.
7244 Because some of the characters used in these syntaxes (such as @samp{@@} and
7245 @samp{#}) are comment characters for some architectures, some of the syntaxes
7246 below do not work on all architectures. The first variant will be accepted by
7247 the GNU assembler on all architectures so that variant should be used for
7248 maximum portability, if you do not need to assemble your code with other
7251 The syntaxes supported are:
7254 .type <name> STT_<TYPE_IN_UPPER_CASE>
7255 .type <name>,#<type>
7256 .type <name>,@@<type>
7257 .type <name>,%<type>
7258 .type <name>,"<type>"
7261 The types supported are:
7266 Mark the symbol as being a function name.
7269 @itemx gnu_indirect_function
7270 Mark the symbol as an indirect function when evaluated during reloc
7271 processing. (This is only supported on assemblers targeting GNU systems).
7275 Mark the symbol as being a data object.
7279 Mark the symbol as being a thread-local data object.
7283 Mark the symbol as being a common data object.
7287 Does not mark the symbol in any way. It is supported just for completeness.
7289 @item gnu_unique_object
7290 Marks the symbol as being a globally unique data object. The dynamic linker
7291 will make sure that in the entire process there is just one symbol with this
7292 name and type in use. (This is only supported on assemblers targeting GNU
7297 Changing between incompatible types other than from/to STT_NOTYPE will
7298 result in a diagnostic. An intermediate change to STT_NOTYPE will silence
7301 Note: Some targets support extra types in addition to those listed above.
7307 @section @code{.uleb128 @var{expressions}}
7309 @cindex @code{uleb128} directive
7310 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7311 compact, variable length representation of numbers used by the DWARF
7312 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7316 @section @code{.val @var{addr}}
7318 @cindex @code{val} directive
7319 @cindex COFF value attribute
7320 @cindex value attribute, COFF
7321 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7322 records the address @var{addr} as the value attribute of a symbol table
7328 @section @code{.version "@var{string}"}
7330 @cindex @code{version} directive
7331 This directive creates a @code{.note} section and places into it an ELF
7332 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7337 @section @code{.vtable_entry @var{table}, @var{offset}}
7339 @cindex @code{vtable_entry} directive
7340 This directive finds or creates a symbol @code{table} and creates a
7341 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7344 @section @code{.vtable_inherit @var{child}, @var{parent}}
7346 @cindex @code{vtable_inherit} directive
7347 This directive finds the symbol @code{child} and finds or creates the symbol
7348 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7349 parent whose addend is the value of the child symbol. As a special case the
7350 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7354 @section @code{.warning "@var{string}"}
7355 @cindex warning directive
7356 Similar to the directive @code{.error}
7357 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7360 @section @code{.weak @var{names}}
7362 @cindex @code{weak} directive
7363 This directive sets the weak attribute on the comma separated list of symbol
7364 @code{names}. If the symbols do not already exist, they will be created.
7366 On COFF targets other than PE, weak symbols are a GNU extension. This
7367 directive sets the weak attribute on the comma separated list of symbol
7368 @code{names}. If the symbols do not already exist, they will be created.
7370 On the PE target, weak symbols are supported natively as weak aliases.
7371 When a weak symbol is created that is not an alias, GAS creates an
7372 alternate symbol to hold the default value.
7375 @section @code{.weakref @var{alias}, @var{target}}
7377 @cindex @code{weakref} directive
7378 This directive creates an alias to the target symbol that enables the symbol to
7379 be referenced with weak-symbol semantics, but without actually making it weak.
7380 If direct references or definitions of the symbol are present, then the symbol
7381 will not be weak, but if all references to it are through weak references, the
7382 symbol will be marked as weak in the symbol table.
7384 The effect is equivalent to moving all references to the alias to a separate
7385 assembly source file, renaming the alias to the symbol in it, declaring the
7386 symbol as weak there, and running a reloadable link to merge the object files
7387 resulting from the assembly of the new source file and the old source file that
7388 had the references to the alias removed.
7390 The alias itself never makes to the symbol table, and is entirely handled
7391 within the assembler.
7394 @section @code{.word @var{expressions}}
7396 @cindex @code{word} directive
7397 This directive expects zero or more @var{expressions}, of any section,
7398 separated by commas.
7401 For each expression, @command{@value{AS}} emits a 32-bit number.
7404 For each expression, @command{@value{AS}} emits a 16-bit number.
7409 The size of the number emitted, and its byte order,
7410 depend on what target computer the assembly is for.
7413 @c on sparc the "special treatment to support compilers" doesn't
7414 @c happen---32-bit addressability, period; no long/short jumps.
7415 @ifset DIFF-TBL-KLUGE
7416 @cindex difference tables altered
7417 @cindex altered difference tables
7419 @emph{Warning: Special Treatment to support Compilers}
7423 Machines with a 32-bit address space, but that do less than 32-bit
7424 addressing, require the following special treatment. If the machine of
7425 interest to you does 32-bit addressing (or doesn't require it;
7426 @pxref{Machine Dependencies}), you can ignore this issue.
7429 In order to assemble compiler output into something that works,
7430 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7431 Directives of the form @samp{.word sym1-sym2} are often emitted by
7432 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7433 directive of the form @samp{.word sym1-sym2}, and the difference between
7434 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7435 creates a @dfn{secondary jump table}, immediately before the next label.
7436 This secondary jump table is preceded by a short-jump to the
7437 first byte after the secondary table. This short-jump prevents the flow
7438 of control from accidentally falling into the new table. Inside the
7439 table is a long-jump to @code{sym2}. The original @samp{.word}
7440 contains @code{sym1} minus the address of the long-jump to
7443 If there were several occurrences of @samp{.word sym1-sym2} before the
7444 secondary jump table, all of them are adjusted. If there was a
7445 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7446 long-jump to @code{sym4} is included in the secondary jump table,
7447 and the @code{.word} directives are adjusted to contain @code{sym3}
7448 minus the address of the long-jump to @code{sym4}; and so on, for as many
7449 entries in the original jump table as necessary.
7452 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7453 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7454 assembly language programmers.
7457 @c end DIFF-TBL-KLUGE
7459 @ifclear no-space-dir
7461 @section @code{.zero @var{size}}
7463 @cindex @code{zero} directive
7464 @cindex filling memory with zero bytes
7465 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7466 expression. This directive is actually an alias for the @samp{.skip} directive
7467 so it can take an optional second argument of the value to store in the bytes
7468 instead of zero. Using @samp{.zero} in this way would be confusing however.
7473 @section @code{.2byte @var{expression} [, @var{expression}]*}
7474 @cindex @code{2byte} directive
7475 @cindex two-byte integer
7476 @cindex integer, 2-byte
7478 This directive expects zero or more expressions, separated by commas. If there
7479 are no expressions then the directive does nothing. Otherwise each expression
7480 is evaluated in turn and placed in the next two bytes of the current output
7481 section, using the endian model of the target. If an expression will not fit
7482 in two bytes, a warning message is displayed and the least significant two
7483 bytes of the expression's value are used. If an expression cannot be evaluated
7484 at assembly time then relocations will be generated in order to compute the
7487 This directive does not apply any alignment before or after inserting the
7488 values. As a result of this, if relocations are generated, they may be
7489 different from those used for inserting values with a guaranteed alignment.
7491 This directive is only available for ELF targets,
7494 @section @code{.4byte @var{expression} [, @var{expression}]*}
7495 @cindex @code{4byte} directive
7496 @cindex four-byte integer
7497 @cindex integer, 4-byte
7499 Like the @option{.2byte} directive, except that it inserts unaligned, four byte
7500 long values into the output.
7503 @section @code{.8byte @var{expression} [, @var{expression}]*}
7504 @cindex @code{8byte} directive
7505 @cindex eight-byte integer
7506 @cindex integer, 8-byte
7508 Like the @option{.2byte} directive, except that it inserts unaligned, eight
7509 byte long bignum values into the output.
7514 @section Deprecated Directives
7516 @cindex deprecated directives
7517 @cindex obsolescent directives
7518 One day these directives won't work.
7519 They are included for compatibility with older assemblers.
7526 @node Object Attributes
7527 @chapter Object Attributes
7528 @cindex object attributes
7530 @command{@value{AS}} assembles source files written for a specific architecture
7531 into object files for that architecture. But not all object files are alike.
7532 Many architectures support incompatible variations. For instance, floating
7533 point arguments might be passed in floating point registers if the object file
7534 requires hardware floating point support---or floating point arguments might be
7535 passed in integer registers if the object file supports processors with no
7536 hardware floating point unit. Or, if two objects are built for different
7537 generations of the same architecture, the combination may require the
7538 newer generation at run-time.
7540 This information is useful during and after linking. At link time,
7541 @command{@value{LD}} can warn about incompatible object files. After link
7542 time, tools like @command{gdb} can use it to process the linked file
7545 Compatibility information is recorded as a series of object attributes. Each
7546 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7547 string, and indicates who sets the meaning of the tag. The tag is an integer,
7548 and indicates what property the attribute describes. The value may be a string
7549 or an integer, and indicates how the property affects this object. Missing
7550 attributes are the same as attributes with a zero value or empty string value.
7552 Object attributes were developed as part of the ABI for the ARM Architecture.
7553 The file format is documented in @cite{ELF for the ARM Architecture}.
7556 * GNU Object Attributes:: @sc{gnu} Object Attributes
7557 * Defining New Object Attributes:: Defining New Object Attributes
7560 @node GNU Object Attributes
7561 @section @sc{gnu} Object Attributes
7563 The @code{.gnu_attribute} directive records an object attribute
7564 with vendor @samp{gnu}.
7566 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7567 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7568 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7569 2} is set for architecture-independent attributes and clear for
7570 architecture-dependent ones.
7572 @subsection Common @sc{gnu} attributes
7574 These attributes are valid on all architectures.
7577 @item Tag_compatibility (32)
7578 The compatibility attribute takes an integer flag value and a vendor name. If
7579 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7580 then the file is only compatible with the named toolchain. If it is greater
7581 than 1, the file can only be processed by other toolchains under some private
7582 arrangement indicated by the flag value and the vendor name.
7585 @subsection MIPS Attributes
7588 @item Tag_GNU_MIPS_ABI_FP (4)
7589 The floating-point ABI used by this object file. The value will be:
7593 0 for files not affected by the floating-point ABI.
7595 1 for files using the hardware floating-point ABI with a standard
7596 double-precision FPU.
7598 2 for files using the hardware floating-point ABI with a single-precision FPU.
7600 3 for files using the software floating-point ABI.
7602 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7603 floating-point registers, 32-bit general-purpose registers and increased the
7604 number of callee-saved floating-point registers.
7606 5 for files using the hardware floating-point ABI with a double-precision FPU
7607 with either 32-bit or 64-bit floating-point registers and 32-bit
7608 general-purpose registers.
7610 6 for files using the hardware floating-point ABI with 64-bit floating-point
7611 registers and 32-bit general-purpose registers.
7613 7 for files using the hardware floating-point ABI with 64-bit floating-point
7614 registers, 32-bit general-purpose registers and a rule that forbids the
7615 direct use of odd-numbered single-precision floating-point registers.
7619 @subsection PowerPC Attributes
7622 @item Tag_GNU_Power_ABI_FP (4)
7623 The floating-point ABI used by this object file. The value will be:
7627 0 for files not affected by the floating-point ABI.
7629 1 for files using double-precision hardware floating-point ABI.
7631 2 for files using the software floating-point ABI.
7633 3 for files using single-precision hardware floating-point ABI.
7636 @item Tag_GNU_Power_ABI_Vector (8)
7637 The vector ABI used by this object file. The value will be:
7641 0 for files not affected by the vector ABI.
7643 1 for files using general purpose registers to pass vectors.
7645 2 for files using AltiVec registers to pass vectors.
7647 3 for files using SPE registers to pass vectors.
7651 @subsection IBM z Systems Attributes
7654 @item Tag_GNU_S390_ABI_Vector (8)
7655 The vector ABI used by this object file. The value will be:
7659 0 for files not affected by the vector ABI.
7661 1 for files using software vector ABI.
7663 2 for files using hardware vector ABI.
7667 @subsection MSP430 Attributes
7670 @item Tag_GNU_MSP430_Data_Region (4)
7671 The data region used by this object file. The value will be:
7675 0 for files not using the large memory model.
7677 1 for files which have been compiled with the condition that all
7678 data is in the lower memory region, i.e. below address 0x10000.
7680 2 for files which allow data to be placed in the full 20-bit memory range.
7684 @node Defining New Object Attributes
7685 @section Defining New Object Attributes
7687 If you want to define a new @sc{gnu} object attribute, here are the places you
7688 will need to modify. New attributes should be discussed on the @samp{binutils}
7693 This manual, which is the official register of attributes.
7695 The header for your architecture @file{include/elf}, to define the tag.
7697 The @file{bfd} support file for your architecture, to merge the attribute
7698 and issue any appropriate link warnings.
7700 Test cases in @file{ld/testsuite} for merging and link warnings.
7702 @file{binutils/readelf.c} to display your attribute.
7704 GCC, if you want the compiler to mark the attribute automatically.
7710 @node Machine Dependencies
7711 @chapter Machine Dependent Features
7713 @cindex machine dependencies
7714 The machine instruction sets are (almost by definition) different on
7715 each machine where @command{@value{AS}} runs. Floating point representations
7716 vary as well, and @command{@value{AS}} often supports a few additional
7717 directives or command-line options for compatibility with other
7718 assemblers on a particular platform. Finally, some versions of
7719 @command{@value{AS}} support special pseudo-instructions for branch
7722 This chapter discusses most of these differences, though it does not
7723 include details on any machine's instruction set. For details on that
7724 subject, see the hardware manufacturer's manual.
7728 * AArch64-Dependent:: AArch64 Dependent Features
7731 * Alpha-Dependent:: Alpha Dependent Features
7734 * ARC-Dependent:: ARC Dependent Features
7737 * ARM-Dependent:: ARM Dependent Features
7740 * AVR-Dependent:: AVR Dependent Features
7743 * Blackfin-Dependent:: Blackfin Dependent Features
7746 * BPF-Dependent:: BPF Dependent Features
7749 * CR16-Dependent:: CR16 Dependent Features
7752 * CRIS-Dependent:: CRIS Dependent Features
7755 * C-SKY-Dependent:: C-SKY Dependent Features
7758 * D10V-Dependent:: D10V Dependent Features
7761 * D30V-Dependent:: D30V Dependent Features
7764 * Epiphany-Dependent:: EPIPHANY Dependent Features
7767 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7770 * HPPA-Dependent:: HPPA Dependent Features
7773 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7776 * IA-64-Dependent:: Intel IA-64 Dependent Features
7779 * IP2K-Dependent:: IP2K Dependent Features
7782 * LM32-Dependent:: LM32 Dependent Features
7785 * M32C-Dependent:: M32C Dependent Features
7788 * M32R-Dependent:: M32R Dependent Features
7791 * M68K-Dependent:: M680x0 Dependent Features
7794 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7797 * S12Z-Dependent:: S12Z Dependent Features
7800 * Meta-Dependent :: Meta Dependent Features
7803 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7806 * MIPS-Dependent:: MIPS Dependent Features
7809 * MMIX-Dependent:: MMIX Dependent Features
7812 * MSP430-Dependent:: MSP430 Dependent Features
7815 * NDS32-Dependent:: Andes NDS32 Dependent Features
7818 * NiosII-Dependent:: Altera Nios II Dependent Features
7821 * NS32K-Dependent:: NS32K Dependent Features
7824 * OpenRISC-Dependent:: OpenRISC 1000 Features
7827 * PDP-11-Dependent:: PDP-11 Dependent Features
7830 * PJ-Dependent:: picoJava Dependent Features
7833 * PPC-Dependent:: PowerPC Dependent Features
7836 * PRU-Dependent:: PRU Dependent Features
7839 * RISC-V-Dependent:: RISC-V Dependent Features
7842 * RL78-Dependent:: RL78 Dependent Features
7845 * RX-Dependent:: RX Dependent Features
7848 * S/390-Dependent:: IBM S/390 Dependent Features
7851 * SCORE-Dependent:: SCORE Dependent Features
7854 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7857 * Sparc-Dependent:: SPARC Dependent Features
7860 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7863 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7866 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7869 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7872 * V850-Dependent:: V850 Dependent Features
7875 * Vax-Dependent:: VAX Dependent Features
7878 * Visium-Dependent:: Visium Dependent Features
7881 * WebAssembly-Dependent:: WebAssembly Dependent Features
7884 * XGATE-Dependent:: XGATE Dependent Features
7887 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7890 * Xtensa-Dependent:: Xtensa Dependent Features
7893 * Z80-Dependent:: Z80 Dependent Features
7896 * Z8000-Dependent:: Z8000 Dependent Features
7903 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7904 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7905 @c peculiarity: to preserve cross-references, there must be a node called
7906 @c "Machine Dependencies". Hence the conditional nodenames in each
7907 @c major node below. Node defaulting in makeinfo requires adjacency of
7908 @c node and sectioning commands; hence the repetition of @chapter BLAH
7909 @c in both conditional blocks.
7912 @include c-aarch64.texi
7916 @include c-alpha.texi
7932 @include c-bfin.texi
7940 @include c-cr16.texi
7944 @include c-cris.texi
7948 @include c-csky.texi
7953 @node Machine Dependencies
7954 @chapter Machine Dependent Features
7956 The machine instruction sets are different on each Renesas chip family,
7957 and there are also some syntax differences among the families. This
7958 chapter describes the specific @command{@value{AS}} features for each
7962 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7963 * SH-Dependent:: Renesas SH Dependent Features
7970 @include c-d10v.texi
7974 @include c-d30v.texi
7978 @include c-epiphany.texi
7982 @include c-h8300.texi
7986 @include c-hppa.texi
7990 @include c-i386.texi
7994 @include c-ia64.texi
7998 @include c-ip2k.texi
8002 @include c-lm32.texi
8006 @include c-m32c.texi
8010 @include c-m32r.texi
8014 @include c-m68k.texi
8018 @include c-m68hc11.texi
8022 @include c-s12z.texi
8026 @include c-metag.texi
8030 @include c-microblaze.texi
8034 @include c-mips.texi
8038 @include c-mmix.texi
8042 @include c-msp430.texi
8046 @include c-nds32.texi
8050 @include c-nios2.texi
8054 @include c-ns32k.texi
8058 @include c-or1k.texi
8062 @include c-pdp11.texi
8078 @include c-riscv.texi
8082 @include c-rl78.texi
8090 @include c-s390.texi
8094 @include c-score.texi
8102 @include c-sparc.texi
8106 @include c-tic54x.texi
8110 @include c-tic6x.texi
8114 @include c-tilegx.texi
8118 @include c-tilepro.texi
8122 @include c-v850.texi
8130 @include c-visium.texi
8134 @include c-wasm32.texi
8138 @include c-xgate.texi
8142 @include c-xstormy16.texi
8146 @include c-xtensa.texi
8158 @c reverse effect of @down at top of generic Machine-Dep chapter
8162 @node Reporting Bugs
8163 @chapter Reporting Bugs
8164 @cindex bugs in assembler
8165 @cindex reporting bugs in assembler
8167 Your bug reports play an essential role in making @command{@value{AS}} reliable.
8169 Reporting a bug may help you by bringing a solution to your problem, or it may
8170 not. But in any case the principal function of a bug report is to help the
8171 entire community by making the next version of @command{@value{AS}} work better.
8172 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
8174 In order for a bug report to serve its purpose, you must include the
8175 information that enables us to fix the bug.
8178 * Bug Criteria:: Have you found a bug?
8179 * Bug Reporting:: How to report bugs
8183 @section Have You Found a Bug?
8184 @cindex bug criteria
8186 If you are not sure whether you have found a bug, here are some guidelines:
8189 @cindex fatal signal
8190 @cindex assembler crash
8191 @cindex crash of assembler
8193 If the assembler gets a fatal signal, for any input whatever, that is a
8194 @command{@value{AS}} bug. Reliable assemblers never crash.
8196 @cindex error on valid input
8198 If @command{@value{AS}} produces an error message for valid input, that is a bug.
8200 @cindex invalid input
8202 If @command{@value{AS}} does not produce an error message for invalid input, that
8203 is a bug. However, you should note that your idea of ``invalid input'' might
8204 be our idea of ``an extension'' or ``support for traditional practice''.
8207 If you are an experienced user of assemblers, your suggestions for improvement
8208 of @command{@value{AS}} are welcome in any case.
8212 @section How to Report Bugs
8214 @cindex assembler bugs, reporting
8216 A number of companies and individuals offer support for @sc{gnu} products. If
8217 you obtained @command{@value{AS}} from a support organization, we recommend you
8218 contact that organization first.
8220 You can find contact information for many support companies and
8221 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
8225 In any event, we also recommend that you send bug reports for @command{@value{AS}}
8229 The fundamental principle of reporting bugs usefully is this:
8230 @strong{report all the facts}. If you are not sure whether to state a
8231 fact or leave it out, state it!
8233 Often people omit facts because they think they know what causes the problem
8234 and assume that some details do not matter. Thus, you might assume that the
8235 name of a symbol you use in an example does not matter. Well, probably it does
8236 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
8237 happens to fetch from the location where that name is stored in memory;
8238 perhaps, if the name were different, the contents of that location would fool
8239 the assembler into doing the right thing despite the bug. Play it safe and
8240 give a specific, complete example. That is the easiest thing for you to do,
8241 and the most helpful.
8243 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
8244 it is new to us. Therefore, always write your bug reports on the assumption
8245 that the bug has not been reported previously.
8247 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8248 bell?'' This cannot help us fix a bug, so it is basically useless. We
8249 respond by asking for enough details to enable us to investigate.
8250 You might as well expedite matters by sending them to begin with.
8252 To enable us to fix the bug, you should include all these things:
8256 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8257 it with the @samp{--version} argument.
8259 Without this, we will not know whether there is any point in looking for
8260 the bug in the current version of @command{@value{AS}}.
8263 Any patches you may have applied to the @command{@value{AS}} source.
8266 The type of machine you are using, and the operating system name and
8270 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8274 The command arguments you gave the assembler to assemble your example and
8275 observe the bug. To guarantee you will not omit something important, list them
8276 all. A copy of the Makefile (or the output from make) is sufficient.
8278 If we were to try to guess the arguments, we would probably guess wrong
8279 and then we might not encounter the bug.
8282 A complete input file that will reproduce the bug. If the bug is observed when
8283 the assembler is invoked via a compiler, send the assembler source, not the
8284 high level language source. Most compilers will produce the assembler source
8285 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8286 the options @samp{-v --save-temps}; this will save the assembler source in a
8287 file with an extension of @file{.s}, and also show you exactly how
8288 @command{@value{AS}} is being run.
8291 A description of what behavior you observe that you believe is
8292 incorrect. For example, ``It gets a fatal signal.''
8294 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8295 will certainly notice it. But if the bug is incorrect output, we might not
8296 notice unless it is glaringly wrong. You might as well not give us a chance to
8299 Even if the problem you experience is a fatal signal, you should still say so
8300 explicitly. Suppose something strange is going on, such as, your copy of
8301 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8302 library on your system. (This has happened!) Your copy might crash and ours
8303 would not. If you told us to expect a crash, then when ours fails to crash, we
8304 would know that the bug was not happening for us. If you had not told us to
8305 expect a crash, then we would not be able to draw any conclusion from our
8309 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8310 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8311 option. Always send diffs from the old file to the new file. If you even
8312 discuss something in the @command{@value{AS}} source, refer to it by context, not
8315 The line numbers in our development sources will not match those in your
8316 sources. Your line numbers would convey no useful information to us.
8319 Here are some things that are not necessary:
8323 A description of the envelope of the bug.
8325 Often people who encounter a bug spend a lot of time investigating
8326 which changes to the input file will make the bug go away and which
8327 changes will not affect it.
8329 This is often time consuming and not very useful, because the way we
8330 will find the bug is by running a single example under the debugger
8331 with breakpoints, not by pure deduction from a series of examples.
8332 We recommend that you save your time for something else.
8334 Of course, if you can find a simpler example to report @emph{instead}
8335 of the original one, that is a convenience for us. Errors in the
8336 output will be easier to spot, running under the debugger will take
8337 less time, and so on.
8339 However, simplification is not vital; if you do not want to do this,
8340 report the bug anyway and send us the entire test case you used.
8343 A patch for the bug.
8345 A patch for the bug does help us if it is a good one. But do not omit
8346 the necessary information, such as the test case, on the assumption that
8347 a patch is all we need. We might see problems with your patch and decide
8348 to fix the problem another way, or we might not understand it at all.
8350 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8351 construct an example that will make the program follow a certain path through
8352 the code. If you do not send us the example, we will not be able to construct
8353 one, so we will not be able to verify that the bug is fixed.
8355 And if we cannot understand what bug you are trying to fix, or why your
8356 patch should be an improvement, we will not install it. A test case will
8357 help us to understand.
8360 A guess about what the bug is or what it depends on.
8362 Such guesses are usually wrong. Even we cannot guess right about such
8363 things without first using the debugger to find the facts.
8366 @node Acknowledgements
8367 @chapter Acknowledgements
8369 If you have contributed to GAS and your name isn't listed here,
8370 it is not meant as a slight. We just don't know about it. Send mail to the
8371 maintainer, and we'll correct the situation. Currently
8373 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8375 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8378 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8379 information and the 68k series machines, most of the preprocessing pass, and
8380 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8382 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8383 many bug fixes, including merging support for several processors, breaking GAS
8384 up to handle multiple object file format back ends (including heavy rewrite,
8385 testing, an integration of the coff and b.out back ends), adding configuration
8386 including heavy testing and verification of cross assemblers and file splits
8387 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8388 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8389 port (including considerable amounts of reverse engineering), a SPARC opcode
8390 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8391 assertions and made them work, much other reorganization, cleanup, and lint.
8393 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8394 in format-specific I/O modules.
8396 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8397 has done much work with it since.
8399 The Intel 80386 machine description was written by Eliot Dresselhaus.
8401 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8403 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8404 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8406 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8407 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8408 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8409 support a.out format.
8411 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8412 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8413 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8414 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8417 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8418 simplified the configuration of which versions accept which directives. He
8419 updated the 68k machine description so that Motorola's opcodes always produced
8420 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8421 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8422 cross-compilation support, and one bug in relaxation that took a week and
8423 required the proverbial one-bit fix.
8425 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8426 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8427 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8428 PowerPC assembler, and made a few other minor patches.
8430 Steve Chamberlain made GAS able to generate listings.
8432 Hewlett-Packard contributed support for the HP9000/300.
8434 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8435 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8436 formats). This work was supported by both the Center for Software Science at
8437 the University of Utah and Cygnus Support.
8439 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8440 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8441 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8442 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8443 and some initial 64-bit support).
8445 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8447 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8448 support for openVMS/Alpha.
8450 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8453 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8454 Inc.@: added support for Xtensa processors.
8456 Several engineers at Cygnus Support have also provided many small bug fixes and
8457 configuration enhancements.
8459 Jon Beniston added support for the Lattice Mico32 architecture.
8461 Many others have contributed large or small bugfixes and enhancements. If
8462 you have contributed significant work and are not mentioned on this list, and
8463 want to be, let us know. Some of the history has been lost; we are not
8464 intentionally leaving anyone out.
8466 @node GNU Free Documentation License
8467 @appendix GNU Free Documentation License
8471 @unnumbered AS Index