1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (C) 1991-2020 Free Software Foundation, Inc.
3 @c UPDATE!! On future updates--
4 @c (1) check for new machine-dep cmdline options in
5 @c md_parse_option definitions in config/tc-*.c
6 @c (2) for platform-specific directives, examine md_pseudo_op
8 @c (3) for object-format specific directives, examine obj_pseudo_op
10 @c (4) portable directives in potable[] in read.c
14 @macro gcctabopt{body}
17 @c defaults, config file may override:
22 @include asconfig.texi
27 @c common OR combinations of conditions
53 @set abnormal-separator
57 @settitle Using @value{AS}
60 @settitle Using @value{AS} (@value{TARGET})
62 @setchapternewpage odd
67 @c WARE! Some of the machine-dependent sections contain tables of machine
68 @c instructions. Except in multi-column format, these tables look silly.
69 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
70 @c the multi-col format is faked within @example sections.
72 @c Again unfortunately, the natural size that fits on a page, for these tables,
73 @c is different depending on whether or not smallbook is turned on.
74 @c This matters, because of order: text flow switches columns at each page
77 @c The format faked in this source works reasonably well for smallbook,
78 @c not well for the default large-page format. This manual expects that if you
79 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
80 @c tables in question. You can turn on one without the other at your
81 @c discretion, of course.
84 @c the insn tables look just as silly in info files regardless of smallbook,
85 @c might as well show 'em anyways.
89 @dircategory Software development
91 * As: (as). The GNU assembler.
92 * Gas: (as). The GNU assembler.
100 This file documents the GNU Assembler "@value{AS}".
102 @c man begin COPYRIGHT
103 Copyright @copyright{} 1991-2020 Free Software Foundation, Inc.
105 Permission is granted to copy, distribute and/or modify this document
106 under the terms of the GNU Free Documentation License, Version 1.3
107 or any later version published by the Free Software Foundation;
108 with no Invariant Sections, with no Front-Cover Texts, and with no
109 Back-Cover Texts. A copy of the license is included in the
110 section entitled ``GNU Free Documentation License''.
116 @title Using @value{AS}
117 @subtitle The @sc{gnu} Assembler
119 @subtitle for the @value{TARGET} family
121 @ifset VERSION_PACKAGE
123 @subtitle @value{VERSION_PACKAGE}
126 @subtitle Version @value{VERSION}
129 The Free Software Foundation Inc.@: thanks The Nice Computer
130 Company of Australia for loaning Dean Elsner to write the
131 first (Vax) version of @command{as} for Project @sc{gnu}.
132 The proprietors, management and staff of TNCCA thank FSF for
133 distracting the boss while they got some work
136 @author Dean Elsner, Jay Fenlason & friends
140 \hfill {\it Using {\tt @value{AS}}}\par
141 \hfill Edited by Cygnus Support\par
143 %"boxit" macro for figures:
144 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
145 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
146 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
147 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
148 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
151 @vskip 0pt plus 1filll
152 Copyright @copyright{} 1991-2020 Free Software Foundation, Inc.
154 Permission is granted to copy, distribute and/or modify this document
155 under the terms of the GNU Free Documentation License, Version 1.3
156 or any later version published by the Free Software Foundation;
157 with no Invariant Sections, with no Front-Cover Texts, and with no
158 Back-Cover Texts. A copy of the license is included in the
159 section entitled ``GNU Free Documentation License''.
166 @top Using @value{AS}
168 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
169 @ifset VERSION_PACKAGE
170 @value{VERSION_PACKAGE}
172 version @value{VERSION}.
174 This version of the file describes @command{@value{AS}} configured to generate
175 code for @value{TARGET} architectures.
178 This document is distributed under the terms of the GNU Free
179 Documentation License. A copy of the license is included in the
180 section entitled ``GNU Free Documentation License''.
183 * Overview:: Overview
184 * Invoking:: Command-Line Options
186 * Sections:: Sections and Relocation
188 * Expressions:: Expressions
189 * Pseudo Ops:: Assembler Directives
191 * Object Attributes:: Object Attributes
193 * Machine Dependencies:: Machine Dependent Features
194 * Reporting Bugs:: Reporting Bugs
195 * Acknowledgements:: Who Did What
196 * GNU Free Documentation License:: GNU Free Documentation License
197 * AS Index:: AS Index
204 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
206 This version of the manual describes @command{@value{AS}} configured to generate
207 code for @value{TARGET} architectures.
211 @cindex invocation summary
212 @cindex option summary
213 @cindex summary of options
214 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
215 see @ref{Invoking,,Command-Line Options}.
217 @c man title AS the portable GNU assembler.
221 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
225 @c We don't use deffn and friends for the following because they seem
226 @c to be limited to one line for the header.
228 @c man begin SYNOPSIS
229 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
230 [@b{--compress-debug-sections}] [@b{--nocompress-debug-sections}]
231 [@b{--debug-prefix-map} @var{old}=@var{new}]
232 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
233 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--gdwarf-sections}]
234 [@b{--gdwarf-cie-version}=@var{VERSION}]
235 [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
236 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
237 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
238 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}]
239 [@b{--no-pad-sections}]
240 [@b{-o} @var{objfile}] [@b{-R}]
241 [@b{--hash-size}=@var{NUM}] [@b{--reduce-memory-overheads}]
243 [@b{-v}] [@b{-version}] [@b{--version}]
244 [@b{-W}] [@b{--warn}] [@b{--fatal-warnings}] [@b{-w}] [@b{-x}]
245 [@b{-Z}] [@b{@@@var{FILE}}]
246 [@b{--sectname-subst}] [@b{--size-check=[error|warning]}]
247 [@b{--elf-stt-common=[no|yes]}]
248 [@b{--generate-missing-build-notes=[no|yes]}]
249 [@b{--target-help}] [@var{target-options}]
250 [@b{--}|@var{files} @dots{}]
253 @c Target dependent options are listed below. Keep the list sorted.
254 @c Add an empty line for separation.
258 @emph{Target AArch64 options:}
260 [@b{-mabi}=@var{ABI}]
264 @emph{Target Alpha options:}
266 [@b{-mdebug} | @b{-no-mdebug}]
267 [@b{-replace} | @b{-noreplace}]
268 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
269 [@b{-F}] [@b{-32addr}]
273 @emph{Target ARC options:}
274 [@b{-mcpu=@var{cpu}}]
275 [@b{-mA6}|@b{-mARC600}|@b{-mARC601}|@b{-mA7}|@b{-mARC700}|@b{-mEM}|@b{-mHS}]
282 @emph{Target ARM options:}
283 @c Don't document the deprecated options
284 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
285 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
286 [@b{-mfpu}=@var{floating-point-format}]
287 [@b{-mfloat-abi}=@var{abi}]
288 [@b{-meabi}=@var{ver}]
291 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
292 @b{-mapcs-reentrant}]
293 [@b{-mthumb-interwork}] [@b{-k}]
297 @emph{Target Blackfin options:}
298 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
305 @emph{Target BPF options:}
310 @emph{Target CRIS options:}
311 [@b{--underscore} | @b{--no-underscore}]
313 [@b{--emulation=criself} | @b{--emulation=crisaout}]
314 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
315 @c Deprecated -- deliberately not documented.
320 @emph{Target C-SKY options:}
321 [@b{-march=@var{arch}}] [@b{-mcpu=@var{cpu}}]
322 [@b{-EL}] [@b{-mlittle-endian}] [@b{-EB}] [@b{-mbig-endian}]
323 [@b{-fpic}] [@b{-pic}]
324 [@b{-mljump}] [@b{-mno-ljump}]
325 [@b{-force2bsr}] [@b{-mforce2bsr}] [@b{-no-force2bsr}] [@b{-mno-force2bsr}]
326 [@b{-jsri2bsr}] [@b{-mjsri2bsr}] [@b{-no-jsri2bsr }] [@b{-mno-jsri2bsr}]
327 [@b{-mnolrw }] [@b{-mno-lrw}]
328 [@b{-melrw}] [@b{-mno-elrw}]
329 [@b{-mlaf }] [@b{-mliterals-after-func}]
330 [@b{-mno-laf}] [@b{-mno-literals-after-func}]
331 [@b{-mlabr}] [@b{-mliterals-after-br}]
332 [@b{-mno-labr}] [@b{-mnoliterals-after-br}]
333 [@b{-mistack}] [@b{-mno-istack}]
334 [@b{-mhard-float}] [@b{-mmp}] [@b{-mcp}] [@b{-mcache}]
335 [@b{-msecurity}] [@b{-mtrust}]
336 [@b{-mdsp}] [@b{-medsp}] [@b{-mvdsp}]
340 @emph{Target D10V options:}
345 @emph{Target D30V options:}
346 [@b{-O}|@b{-n}|@b{-N}]
350 @emph{Target EPIPHANY options:}
351 [@b{-mepiphany}|@b{-mepiphany16}]
355 @emph{Target H8/300 options:}
359 @c HPPA has no machine-dependent assembler options (yet).
363 @emph{Target i386 options:}
364 [@b{--32}|@b{--x32}|@b{--64}] [@b{-n}]
365 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
369 @emph{Target IA-64 options:}
370 [@b{-mconstant-gp}|@b{-mauto-pic}]
371 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
373 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
374 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
375 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
376 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
380 @emph{Target IP2K options:}
381 [@b{-mip2022}|@b{-mip2022ext}]
385 @emph{Target M32C options:}
386 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
390 @emph{Target M32R options:}
391 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
396 @emph{Target M680X0 options:}
397 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
401 @emph{Target M68HC11 options:}
402 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}|@b{-mm9s12x}|@b{-mm9s12xg}]
403 [@b{-mshort}|@b{-mlong}]
404 [@b{-mshort-double}|@b{-mlong-double}]
405 [@b{--force-long-branches}] [@b{--short-branches}]
406 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
407 [@b{--print-opcodes}] [@b{--generate-example}]
411 @emph{Target MCORE options:}
412 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
413 [@b{-mcpu=[210|340]}]
417 @emph{Target Meta options:}
418 [@b{-mcpu=@var{cpu}}] [@b{-mfpu=@var{cpu}}] [@b{-mdsp=@var{cpu}}]
421 @emph{Target MICROBLAZE options:}
422 @c MicroBlaze has no machine-dependent assembler options.
426 @emph{Target MIPS options:}
427 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
428 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
429 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
430 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
431 [@b{-mfp64}] [@b{-mgp64}] [@b{-mfpxx}]
432 [@b{-modd-spreg}] [@b{-mno-odd-spreg}]
433 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
434 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
435 [@b{-mips32r3}] [@b{-mips32r5}] [@b{-mips32r6}] [@b{-mips64}] [@b{-mips64r2}]
436 [@b{-mips64r3}] [@b{-mips64r5}] [@b{-mips64r6}]
437 [@b{-construct-floats}] [@b{-no-construct-floats}]
438 [@b{-mignore-branch-isa}] [@b{-mno-ignore-branch-isa}]
439 [@b{-mnan=@var{encoding}}]
440 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
441 [@b{-mips16}] [@b{-no-mips16}]
442 [@b{-mmips16e2}] [@b{-mno-mips16e2}]
443 [@b{-mmicromips}] [@b{-mno-micromips}]
444 [@b{-msmartmips}] [@b{-mno-smartmips}]
445 [@b{-mips3d}] [@b{-no-mips3d}]
446 [@b{-mdmx}] [@b{-no-mdmx}]
447 [@b{-mdsp}] [@b{-mno-dsp}]
448 [@b{-mdspr2}] [@b{-mno-dspr2}]
449 [@b{-mdspr3}] [@b{-mno-dspr3}]
450 [@b{-mmsa}] [@b{-mno-msa}]
451 [@b{-mxpa}] [@b{-mno-xpa}]
452 [@b{-mmt}] [@b{-mno-mt}]
453 [@b{-mmcu}] [@b{-mno-mcu}]
454 [@b{-mcrc}] [@b{-mno-crc}]
455 [@b{-mginv}] [@b{-mno-ginv}]
456 [@b{-mloongson-mmi}] [@b{-mno-loongson-mmi}]
457 [@b{-mloongson-cam}] [@b{-mno-loongson-cam}]
458 [@b{-mloongson-ext}] [@b{-mno-loongson-ext}]
459 [@b{-mloongson-ext2}] [@b{-mno-loongson-ext2}]
460 [@b{-minsn32}] [@b{-mno-insn32}]
461 [@b{-mfix7000}] [@b{-mno-fix7000}]
462 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
463 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
464 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
465 [@b{-mfix-r5900}] [@b{-mno-fix-r5900}]
466 [@b{-mdebug}] [@b{-no-mdebug}]
467 [@b{-mpdr}] [@b{-mno-pdr}]
471 @emph{Target MMIX options:}
472 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
473 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
474 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
475 [@b{--linker-allocated-gregs}]
479 @emph{Target Nios II options:}
480 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
485 @emph{Target NDS32 options:}
486 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
487 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
488 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
489 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
490 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
491 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
492 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
496 @c OpenRISC has no machine-dependent assembler options.
500 @emph{Target PDP11 options:}
501 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
502 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
503 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
507 @emph{Target picoJava options:}
512 @emph{Target PowerPC options:}
514 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
515 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mgekko}|
516 @b{-mbroadway}|@b{-mppc64}|@b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|
517 @b{-me6500}|@b{-mppc64bridge}|@b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|
518 @b{-mpower6}|@b{-mpwr6}|@b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-mpower9}|@b{-mpwr9}@b{-ma2}|
519 @b{-mcell}|@b{-mspe}|@b{-mspe2}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
520 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
521 [@b{-mregnames}|@b{-mno-regnames}]
522 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
523 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
524 [@b{-msolaris}|@b{-mno-solaris}]
525 [@b{-nops=@var{count}}]
529 @emph{Target PRU options:}
532 [@b{-mno-warn-regname-label}]
536 @emph{Target RISC-V options:}
537 [@b{-fpic}|@b{-fPIC}|@b{-fno-pic}]
538 [@b{-march}=@var{ISA}]
539 [@b{-mabi}=@var{ABI}]
543 @emph{Target RL78 options:}
545 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
549 @emph{Target RX options:}
550 [@b{-mlittle-endian}|@b{-mbig-endian}]
551 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
552 [@b{-muse-conventional-section-names}]
553 [@b{-msmall-data-limit}]
556 [@b{-mint-register=@var{number}}]
557 [@b{-mgcc-abi}|@b{-mrx-abi}]
561 @emph{Target s390 options:}
562 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
563 [@b{-mregnames}|@b{-mno-regnames}]
564 [@b{-mwarn-areg-zero}]
568 @emph{Target SCORE options:}
569 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
570 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
571 [@b{-march=score7}][@b{-march=score3}]
572 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
576 @emph{Target SPARC options:}
577 @c The order here is important. See c-sparc.texi.
578 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Aleon}|@b{-Asparclet}|@b{-Asparclite}
579 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av8plusb}|@b{-Av8plusc}|@b{-Av8plusd}
580 @b{-Av8plusv}|@b{-Av8plusm}|@b{-Av9}|@b{-Av9a}|@b{-Av9b}|@b{-Av9c}
581 @b{-Av9d}|@b{-Av9e}|@b{-Av9v}|@b{-Av9m}|@b{-Asparc}|@b{-Asparcvis}
582 @b{-Asparcvis2}|@b{-Asparcfmaf}|@b{-Asparcima}|@b{-Asparcvis3}
583 @b{-Asparcvisr}|@b{-Asparc5}]
584 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}]|@b{-xarch=v8plusb}|@b{-xarch=v8plusc}
585 @b{-xarch=v8plusd}|@b{-xarch=v8plusv}|@b{-xarch=v8plusm}|@b{-xarch=v9}
586 @b{-xarch=v9a}|@b{-xarch=v9b}|@b{-xarch=v9c}|@b{-xarch=v9d}|@b{-xarch=v9e}
587 @b{-xarch=v9v}|@b{-xarch=v9m}|@b{-xarch=sparc}|@b{-xarch=sparcvis}
588 @b{-xarch=sparcvis2}|@b{-xarch=sparcfmaf}|@b{-xarch=sparcima}
589 @b{-xarch=sparcvis3}|@b{-xarch=sparcvisr}|@b{-xarch=sparc5}
592 [@b{--enforce-aligned-data}][@b{--dcti-couples-detect}]
596 @emph{Target TIC54X options:}
597 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
598 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
602 @emph{Target TIC6X options:}
603 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
604 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
605 [@b{-mpic}|@b{-mno-pic}]
609 @emph{Target TILE-Gx options:}
610 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
613 @c TILEPro has no machine-dependent assembler options
617 @emph{Target Visium options:}
618 [@b{-mtune=@var{arch}}]
622 @emph{Target Xtensa options:}
623 [@b{--[no-]text-section-literals}] [@b{--[no-]auto-litpools}]
624 [@b{--[no-]absolute-literals}]
625 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
626 [@b{--[no-]transform}]
627 [@b{--rename-section} @var{oldname}=@var{newname}]
628 [@b{--[no-]trampolines}]
632 @emph{Target Z80 options:}
633 [@b{-z80}]|[@b{-z180}]|[@b{-r800}]|[@b{-ez80}]|[@b{-ez80-adl}]
634 [@b{-local-prefix=}@var{PREFIX}]
637 [@b{-fp-s=}@var{FORMAT}]
638 [@b{-fp-d=}@var{FORMAT}]
639 [@b{-strict}]|[@b{-full}]
640 [@b{-with-inst=@var{INST}[,...]}] [@b{-Wnins @var{INST}[,...]}]
641 [@b{-without-inst=@var{INST}[,...]}] [@b{-Fins @var{INST}[,...]}]
642 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
643 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
644 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
645 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
646 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
647 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
651 @c Z8000 has no machine-dependent assembler options
660 @include at-file.texi
663 Turn on listings, in any of a variety of ways:
667 omit false conditionals
670 omit debugging directives
673 include general information, like @value{AS} version and options passed
676 include high-level source
682 include macro expansions
685 omit forms processing
691 set the name of the listing file
694 You may combine these options; for example, use @samp{-aln} for assembly
695 listing without forms processing. The @samp{=file} option, if used, must be
696 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
699 Begin in alternate macro mode.
701 @xref{Altmacro,,@code{.altmacro}}.
704 @item --compress-debug-sections
705 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
706 ELF ABI. The resulting object file may not be compatible with older
707 linkers and object file utilities. Note if compression would make a
708 given section @emph{larger} then it is not compressed.
711 @cindex @samp{--compress-debug-sections=} option
712 @item --compress-debug-sections=none
713 @itemx --compress-debug-sections=zlib
714 @itemx --compress-debug-sections=zlib-gnu
715 @itemx --compress-debug-sections=zlib-gabi
716 These options control how DWARF debug sections are compressed.
717 @option{--compress-debug-sections=none} is equivalent to
718 @option{--nocompress-debug-sections}.
719 @option{--compress-debug-sections=zlib} and
720 @option{--compress-debug-sections=zlib-gabi} are equivalent to
721 @option{--compress-debug-sections}.
722 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
723 sections using zlib. The debug sections are renamed to begin with
724 @samp{.zdebug}. Note if compression would make a given section
725 @emph{larger} then it is not compressed nor renamed.
729 @item --nocompress-debug-sections
730 Do not compress DWARF debug sections. This is usually the default for all
731 targets except the x86/x86_64, but a configure time option can be used to
735 Ignored. This option is accepted for script compatibility with calls to
738 @item --debug-prefix-map @var{old}=@var{new}
739 When assembling files in directory @file{@var{old}}, record debugging
740 information describing them as in @file{@var{new}} instead.
742 @item --defsym @var{sym}=@var{value}
743 Define the symbol @var{sym} to be @var{value} before assembling the input file.
744 @var{value} must be an integer constant. As in C, a leading @samp{0x}
745 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
746 value. The value of the symbol can be overridden inside a source file via the
747 use of a @code{.set} pseudo-op.
750 ``fast''---skip whitespace and comment preprocessing (assume source is
755 Generate debugging information for each assembler source line using whichever
756 debug format is preferred by the target. This currently means either STABS,
760 Generate stabs debugging information for each assembler line. This
761 may help debugging assembler code, if the debugger can handle it.
764 Generate stabs debugging information for each assembler line, with GNU
765 extensions that probably only gdb can handle, and that could make other
766 debuggers crash or refuse to read your program. This
767 may help debugging assembler code. Currently the only GNU extension is
768 the location of the current working directory at assembling time.
771 Generate DWARF2 debugging information for each assembler line. This
772 may help debugging assembler code, if the debugger can handle it. Note---this
773 option is only supported by some targets, not all of them.
775 @item --gdwarf-sections
776 Instead of creating a .debug_line section, create a series of
777 .debug_line.@var{foo} sections where @var{foo} is the name of the
778 corresponding code section. For example a code section called @var{.text.func}
779 will have its dwarf line number information placed into a section called
780 @var{.debug_line.text.func}. If the code section is just called @var{.text}
781 then debug line section will still be called just @var{.debug_line} without any
784 @item --gdwarf-cie-version=@var{version}
785 Control which version of DWARF Common Information Entries (CIEs) are produced.
786 When this flag is not specificed the default is version 1, though some targets
787 can modify this default. Other possible values for @var{version} are 3 or 4.
790 @item --size-check=error
791 @itemx --size-check=warning
792 Issue an error or warning for invalid ELF .size directive.
794 @item --elf-stt-common=no
795 @itemx --elf-stt-common=yes
796 These options control whether the ELF assembler should generate common
797 symbols with the @code{STT_COMMON} type. The default can be controlled
798 by a configure option @option{--enable-elf-stt-common}.
800 @item --generate-missing-build-notes=yes
801 @itemx --generate-missing-build-notes=no
802 These options control whether the ELF assembler should generate GNU Build
803 attribute notes if none are present in the input sources.
804 The default can be controlled by the @option{--enable-generate-build-notes}
810 Print a summary of the command-line options and exit.
813 Print a summary of all target specific options and exit.
816 Add directory @var{dir} to the search list for @code{.include} directives.
819 Don't warn about signed overflow.
822 @ifclear DIFF-TBL-KLUGE
823 This option is accepted but has no effect on the @value{TARGET} family.
825 @ifset DIFF-TBL-KLUGE
826 Issue warnings when difference tables altered for long displacements.
831 Keep (in the symbol table) local symbols. These symbols start with
832 system-specific local label prefixes, typically @samp{.L} for ELF systems
833 or @samp{L} for traditional a.out systems.
838 @item --listing-lhs-width=@var{number}
839 Set the maximum width, in words, of the output data column for an assembler
840 listing to @var{number}.
842 @item --listing-lhs-width2=@var{number}
843 Set the maximum width, in words, of the output data column for continuation
844 lines in an assembler listing to @var{number}.
846 @item --listing-rhs-width=@var{number}
847 Set the maximum width of an input source line, as displayed in a listing, to
850 @item --listing-cont-lines=@var{number}
851 Set the maximum number of lines printed in a listing for a single line of input
854 @item --no-pad-sections
855 Stop the assembler for padding the ends of output sections to the alignment
856 of that section. The default is to pad the sections, but this can waste space
857 which might be needed on targets which have tight memory constraints.
859 @item -o @var{objfile}
860 Name the object-file output from @command{@value{AS}} @var{objfile}.
863 Fold the data section into the text section.
865 @item --hash-size=@var{number}
866 Set the default size of GAS's hash tables to a prime number close to
867 @var{number}. Increasing this value can reduce the length of time it takes the
868 assembler to perform its tasks, at the expense of increasing the assembler's
869 memory requirements. Similarly reducing this value can reduce the memory
870 requirements at the expense of speed.
872 @item --reduce-memory-overheads
873 This option reduces GAS's memory requirements, at the expense of making the
874 assembly processes slower. Currently this switch is a synonym for
875 @samp{--hash-size=4051}, but in the future it may have other effects as well.
878 @item --sectname-subst
879 Honor substitution sequences in section names.
881 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
886 Print the maximum space (in bytes) and total time (in seconds) used by
889 @item --strip-local-absolute
890 Remove local absolute symbols from the outgoing symbol table.
894 Print the @command{as} version.
897 Print the @command{as} version and exit.
901 Suppress warning messages.
903 @item --fatal-warnings
904 Treat warnings as errors.
907 Don't suppress warning messages or treat them as errors.
916 Generate an object file even after errors.
918 @item -- | @var{files} @dots{}
919 Standard input, or source files to assemble.
927 @xref{AArch64 Options}, for the options available when @value{AS} is configured
928 for the 64-bit mode of the ARM Architecture (AArch64).
933 The following options are available when @value{AS} is configured for the
934 64-bit mode of the ARM Architecture (AArch64).
937 @include c-aarch64.texi
938 @c ended inside the included file
946 @xref{Alpha Options}, for the options available when @value{AS} is configured
947 for an Alpha processor.
952 The following options are available when @value{AS} is configured for an Alpha
956 @include c-alpha.texi
957 @c ended inside the included file
964 The following options are available when @value{AS} is configured for an ARC
968 @item -mcpu=@var{cpu}
969 This option selects the core processor variant.
971 Select either big-endian (-EB) or little-endian (-EL) output.
973 Enable Code Density extenssion instructions.
978 The following options are available when @value{AS} is configured for the ARM
982 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
983 Specify which ARM processor variant is the target.
984 @item -march=@var{architecture}[+@var{extension}@dots{}]
985 Specify which ARM architecture variant is used by the target.
986 @item -mfpu=@var{floating-point-format}
987 Select which Floating Point architecture is the target.
988 @item -mfloat-abi=@var{abi}
989 Select which floating point ABI is in use.
991 Enable Thumb only instruction decoding.
992 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
993 Select which procedure calling convention is in use.
995 Select either big-endian (-EB) or little-endian (-EL) output.
996 @item -mthumb-interwork
997 Specify that the code has been generated with interworking between Thumb and
1000 Turns on CodeComposer Studio assembly syntax compatibility mode.
1002 Specify that PIC code has been generated.
1010 @xref{Blackfin Options}, for the options available when @value{AS} is
1011 configured for the Blackfin processor family.
1015 @c man begin OPTIONS
1016 The following options are available when @value{AS} is configured for
1017 the Blackfin processor family.
1019 @c man begin INCLUDE
1020 @include c-bfin.texi
1021 @c ended inside the included file
1029 @xref{BPF Options}, for the options available when @value{AS} is
1030 configured for the Linux kernel BPF processor family.
1034 @c man begin OPTIONS
1035 The following options are available when @value{AS} is configured for
1036 the Linux kernel BPF processor family.
1038 @c man begin INCLUDE
1040 @c ended inside the included file
1045 @c man begin OPTIONS
1047 See the info pages for documentation of the CRIS-specific options.
1053 @xref{C-SKY Options}, for the options available when @value{AS} is
1054 configured for the C-SKY processor family.
1058 @c man begin OPTIONS
1059 The following options are available when @value{AS} is configured for
1060 the C-SKY processor family.
1062 @c man begin INCLUDE
1063 @include c-csky.texi
1064 @c ended inside the included file
1070 The following options are available when @value{AS} is configured for
1073 @cindex D10V optimization
1074 @cindex optimization, D10V
1076 Optimize output by parallelizing instructions.
1081 The following options are available when @value{AS} is configured for a D30V
1084 @cindex D30V optimization
1085 @cindex optimization, D30V
1087 Optimize output by parallelizing instructions.
1091 Warn when nops are generated.
1093 @cindex D30V nops after 32-bit multiply
1095 Warn when a nop after a 32-bit multiply instruction is generated.
1101 The following options are available when @value{AS} is configured for the
1102 Adapteva EPIPHANY series.
1105 @xref{Epiphany Options}, for the options available when @value{AS} is
1106 configured for an Epiphany processor.
1110 @c man begin OPTIONS
1111 The following options are available when @value{AS} is configured for
1112 an Epiphany processor.
1114 @c man begin INCLUDE
1115 @include c-epiphany.texi
1116 @c ended inside the included file
1124 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1125 for an H8/300 processor.
1129 @c man begin OPTIONS
1130 The following options are available when @value{AS} is configured for an H8/300
1133 @c man begin INCLUDE
1134 @include c-h8300.texi
1135 @c ended inside the included file
1143 @xref{i386-Options}, for the options available when @value{AS} is
1144 configured for an i386 processor.
1148 @c man begin OPTIONS
1149 The following options are available when @value{AS} is configured for
1152 @c man begin INCLUDE
1153 @include c-i386.texi
1154 @c ended inside the included file
1159 @c man begin OPTIONS
1161 The following options are available when @value{AS} is configured for the
1167 Specifies that the extended IP2022 instructions are allowed.
1170 Restores the default behaviour, which restricts the permitted instructions to
1171 just the basic IP2022 ones.
1177 The following options are available when @value{AS} is configured for the
1178 Renesas M32C and M16C processors.
1183 Assemble M32C instructions.
1186 Assemble M16C instructions (the default).
1189 Enable support for link-time relaxations.
1192 Support H'00 style hex constants in addition to 0x00 style.
1198 The following options are available when @value{AS} is configured for the
1199 Renesas M32R (formerly Mitsubishi M32R) series.
1204 Specify which processor in the M32R family is the target. The default
1205 is normally the M32R, but this option changes it to the M32RX.
1207 @item --warn-explicit-parallel-conflicts or --Wp
1208 Produce warning messages when questionable parallel constructs are
1211 @item --no-warn-explicit-parallel-conflicts or --Wnp
1212 Do not produce warning messages when questionable parallel constructs are
1219 The following options are available when @value{AS} is configured for the
1220 Motorola 68000 series.
1225 Shorten references to undefined symbols, to one word instead of two.
1227 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1228 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1229 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1230 Specify what processor in the 68000 family is the target. The default
1231 is normally the 68020, but this can be changed at configuration time.
1233 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1234 The target machine does (or does not) have a floating-point coprocessor.
1235 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1236 the basic 68000 is not compatible with the 68881, a combination of the
1237 two can be specified, since it's possible to do emulation of the
1238 coprocessor instructions with the main processor.
1240 @item -m68851 | -mno-68851
1241 The target machine does (or does not) have a memory-management
1242 unit coprocessor. The default is to assume an MMU for 68020 and up.
1250 @xref{Nios II Options}, for the options available when @value{AS} is configured
1251 for an Altera Nios II processor.
1255 @c man begin OPTIONS
1256 The following options are available when @value{AS} is configured for an
1257 Altera Nios II processor.
1259 @c man begin INCLUDE
1260 @include c-nios2.texi
1261 @c ended inside the included file
1267 For details about the PDP-11 machine dependent features options,
1268 see @ref{PDP-11-Options}.
1271 @item -mpic | -mno-pic
1272 Generate position-independent (or position-dependent) code. The
1273 default is @option{-mpic}.
1276 @itemx -mall-extensions
1277 Enable all instruction set extensions. This is the default.
1279 @item -mno-extensions
1280 Disable all instruction set extensions.
1282 @item -m@var{extension} | -mno-@var{extension}
1283 Enable (or disable) a particular instruction set extension.
1286 Enable the instruction set extensions supported by a particular CPU, and
1287 disable all other extensions.
1289 @item -m@var{machine}
1290 Enable the instruction set extensions supported by a particular machine
1291 model, and disable all other extensions.
1297 The following options are available when @value{AS} is configured for
1298 a picoJava processor.
1302 @cindex PJ endianness
1303 @cindex endianness, PJ
1304 @cindex big endian output, PJ
1306 Generate ``big endian'' format output.
1308 @cindex little endian output, PJ
1310 Generate ``little endian'' format output.
1318 @xref{PRU Options}, for the options available when @value{AS} is configured
1319 for a PRU processor.
1323 @c man begin OPTIONS
1324 The following options are available when @value{AS} is configured for a
1327 @c man begin INCLUDE
1329 @c ended inside the included file
1334 The following options are available when @value{AS} is configured for the
1335 Motorola 68HC11 or 68HC12 series.
1339 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1340 Specify what processor is the target. The default is
1341 defined by the configuration option when building the assembler.
1343 @item --xgate-ramoffset
1344 Instruct the linker to offset RAM addresses from S12X address space into
1345 XGATE address space.
1348 Specify to use the 16-bit integer ABI.
1351 Specify to use the 32-bit integer ABI.
1353 @item -mshort-double
1354 Specify to use the 32-bit double ABI.
1357 Specify to use the 64-bit double ABI.
1359 @item --force-long-branches
1360 Relative branches are turned into absolute ones. This concerns
1361 conditional branches, unconditional branches and branches to a
1364 @item -S | --short-branches
1365 Do not turn relative branches into absolute ones
1366 when the offset is out of range.
1368 @item --strict-direct-mode
1369 Do not turn the direct addressing mode into extended addressing mode
1370 when the instruction does not support direct addressing mode.
1372 @item --print-insn-syntax
1373 Print the syntax of instruction in case of error.
1375 @item --print-opcodes
1376 Print the list of instructions with syntax and then exit.
1378 @item --generate-example
1379 Print an example of instruction for each possible instruction and then exit.
1380 This option is only useful for testing @command{@value{AS}}.
1386 The following options are available when @command{@value{AS}} is configured
1387 for the SPARC architecture:
1390 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1391 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1392 Explicitly select a variant of the SPARC architecture.
1394 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1395 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1397 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1398 UltraSPARC extensions.
1400 @item -xarch=v8plus | -xarch=v8plusa
1401 For compatibility with the Solaris v9 assembler. These options are
1402 equivalent to -Av8plus and -Av8plusa, respectively.
1405 Warn when the assembler switches to another architecture.
1410 The following options are available when @value{AS} is configured for the 'c54x
1415 Enable extended addressing mode. All addresses and relocations will assume
1416 extended addressing (usually 23 bits).
1417 @item -mcpu=@var{CPU_VERSION}
1418 Sets the CPU version being compiled for.
1419 @item -merrors-to-file @var{FILENAME}
1420 Redirect error output to a file, for broken systems which don't support such
1421 behaviour in the shell.
1426 @c man begin OPTIONS
1427 The following options are available when @value{AS} is configured for
1432 This option sets the largest size of an object that can be referenced
1433 implicitly with the @code{gp} register. It is only accepted for targets that
1434 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1436 @cindex MIPS endianness
1437 @cindex endianness, MIPS
1438 @cindex big endian output, MIPS
1440 Generate ``big endian'' format output.
1442 @cindex little endian output, MIPS
1444 Generate ``little endian'' format output.
1462 Generate code for a particular MIPS Instruction Set Architecture level.
1463 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1464 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1465 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1466 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1467 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1468 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1469 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1470 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1471 MIPS64 Release 6 ISA processors, respectively.
1473 @item -march=@var{cpu}
1474 Generate code for a particular MIPS CPU.
1476 @item -mtune=@var{cpu}
1477 Schedule and tune for a particular MIPS CPU.
1481 Cause nops to be inserted if the read of the destination register
1482 of an mfhi or mflo instruction occurs in the following two instructions.
1485 @itemx -mno-fix-rm7000
1486 Cause nops to be inserted if a dmult or dmultu instruction is
1487 followed by a load instruction.
1490 @itemx -mno-fix-r5900
1491 Do not attempt to schedule the preceding instruction into the delay slot
1492 of a branch instruction placed at the end of a short loop of six
1493 instructions or fewer and always schedule a @code{nop} instruction there
1494 instead. The short loop bug under certain conditions causes loops to
1495 execute only once or twice, due to a hardware bug in the R5900 chip.
1499 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1500 section instead of the standard ELF .stabs sections.
1504 Control generation of @code{.pdr} sections.
1508 The register sizes are normally inferred from the ISA and ABI, but these
1509 flags force a certain group of registers to be treated as 32 bits wide at
1510 all times. @samp{-mgp32} controls the size of general-purpose registers
1511 and @samp{-mfp32} controls the size of floating-point registers.
1515 The register sizes are normally inferred from the ISA and ABI, but these
1516 flags force a certain group of registers to be treated as 64 bits wide at
1517 all times. @samp{-mgp64} controls the size of general-purpose registers
1518 and @samp{-mfp64} controls the size of floating-point registers.
1521 The register sizes are normally inferred from the ISA and ABI, but using
1522 this flag in combination with @samp{-mabi=32} enables an ABI variant
1523 which will operate correctly with floating-point registers which are
1527 @itemx -mno-odd-spreg
1528 Enable use of floating-point operations on odd-numbered single-precision
1529 registers when supported by the ISA. @samp{-mfpxx} implies
1530 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1534 Generate code for the MIPS 16 processor. This is equivalent to putting
1535 @code{.module mips16} at the start of the assembly file. @samp{-no-mips16}
1536 turns off this option.
1539 @itemx -mno-mips16e2
1540 Enable the use of MIPS16e2 instructions in MIPS16 mode. This is equivalent
1541 to putting @code{.module mips16e2} at the start of the assembly file.
1542 @samp{-mno-mips16e2} turns off this option.
1545 @itemx -mno-micromips
1546 Generate code for the microMIPS processor. This is equivalent to putting
1547 @code{.module micromips} at the start of the assembly file.
1548 @samp{-mno-micromips} turns off this option. This is equivalent to putting
1549 @code{.module nomicromips} at the start of the assembly file.
1552 @itemx -mno-smartmips
1553 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1554 equivalent to putting @code{.module smartmips} at the start of the assembly
1555 file. @samp{-mno-smartmips} turns off this option.
1559 Generate code for the MIPS-3D Application Specific Extension.
1560 This tells the assembler to accept MIPS-3D instructions.
1561 @samp{-no-mips3d} turns off this option.
1565 Generate code for the MDMX Application Specific Extension.
1566 This tells the assembler to accept MDMX instructions.
1567 @samp{-no-mdmx} turns off this option.
1571 Generate code for the DSP Release 1 Application Specific Extension.
1572 This tells the assembler to accept DSP Release 1 instructions.
1573 @samp{-mno-dsp} turns off this option.
1577 Generate code for the DSP Release 2 Application Specific Extension.
1578 This option implies @samp{-mdsp}.
1579 This tells the assembler to accept DSP Release 2 instructions.
1580 @samp{-mno-dspr2} turns off this option.
1584 Generate code for the DSP Release 3 Application Specific Extension.
1585 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1586 This tells the assembler to accept DSP Release 3 instructions.
1587 @samp{-mno-dspr3} turns off this option.
1591 Generate code for the MIPS SIMD Architecture Extension.
1592 This tells the assembler to accept MSA instructions.
1593 @samp{-mno-msa} turns off this option.
1597 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1598 This tells the assembler to accept XPA instructions.
1599 @samp{-mno-xpa} turns off this option.
1603 Generate code for the MT Application Specific Extension.
1604 This tells the assembler to accept MT instructions.
1605 @samp{-mno-mt} turns off this option.
1609 Generate code for the MCU Application Specific Extension.
1610 This tells the assembler to accept MCU instructions.
1611 @samp{-mno-mcu} turns off this option.
1615 Generate code for the MIPS cyclic redundancy check (CRC) Application
1616 Specific Extension. This tells the assembler to accept CRC instructions.
1617 @samp{-mno-crc} turns off this option.
1621 Generate code for the Global INValidate (GINV) Application Specific
1622 Extension. This tells the assembler to accept GINV instructions.
1623 @samp{-mno-ginv} turns off this option.
1625 @item -mloongson-mmi
1626 @itemx -mno-loongson-mmi
1627 Generate code for the Loongson MultiMedia extensions Instructions (MMI)
1628 Application Specific Extension. This tells the assembler to accept MMI
1630 @samp{-mno-loongson-mmi} turns off this option.
1632 @item -mloongson-cam
1633 @itemx -mno-loongson-cam
1634 Generate code for the Loongson Content Address Memory (CAM) instructions.
1635 This tells the assembler to accept Loongson CAM instructions.
1636 @samp{-mno-loongson-cam} turns off this option.
1638 @item -mloongson-ext
1639 @itemx -mno-loongson-ext
1640 Generate code for the Loongson EXTensions (EXT) instructions.
1641 This tells the assembler to accept Loongson EXT instructions.
1642 @samp{-mno-loongson-ext} turns off this option.
1644 @item -mloongson-ext2
1645 @itemx -mno-loongson-ext2
1646 Generate code for the Loongson EXTensions R2 (EXT2) instructions.
1647 This option implies @samp{-mloongson-ext}.
1648 This tells the assembler to accept Loongson EXT2 instructions.
1649 @samp{-mno-loongson-ext2} turns off this option.
1653 Only use 32-bit instruction encodings when generating code for the
1654 microMIPS processor. This option inhibits the use of any 16-bit
1655 instructions. This is equivalent to putting @code{.set insn32} at
1656 the start of the assembly file. @samp{-mno-insn32} turns off this
1657 option. This is equivalent to putting @code{.set noinsn32} at the
1658 start of the assembly file. By default @samp{-mno-insn32} is
1659 selected, allowing all instructions to be used.
1661 @item --construct-floats
1662 @itemx --no-construct-floats
1663 The @samp{--no-construct-floats} option disables the construction of
1664 double width floating point constants by loading the two halves of the
1665 value into the two single width floating point registers that make up
1666 the double width register. By default @samp{--construct-floats} is
1667 selected, allowing construction of these floating point constants.
1669 @item --relax-branch
1670 @itemx --no-relax-branch
1671 The @samp{--relax-branch} option enables the relaxation of out-of-range
1672 branches. By default @samp{--no-relax-branch} is selected, causing any
1673 out-of-range branches to produce an error.
1675 @item -mignore-branch-isa
1676 @itemx -mno-ignore-branch-isa
1677 Ignore branch checks for invalid transitions between ISA modes. The
1678 semantics of branches does not provide for an ISA mode switch, so in
1679 most cases the ISA mode a branch has been encoded for has to be the
1680 same as the ISA mode of the branch's target label. Therefore GAS has
1681 checks implemented that verify in branch assembly that the two ISA
1682 modes match. @samp{-mignore-branch-isa} disables these checks. By
1683 default @samp{-mno-ignore-branch-isa} is selected, causing any invalid
1684 branch requiring a transition between ISA modes to produce an error.
1686 @item -mnan=@var{encoding}
1687 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1688 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1691 @item --emulation=@var{name}
1692 This option was formerly used to switch between ELF and ECOFF output
1693 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1694 removed in GAS 2.24, so the option now serves little purpose.
1695 It is retained for backwards compatibility.
1697 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1698 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1699 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1700 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1701 preferred options instead.
1704 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1711 Control how to deal with multiplication overflow and division by zero.
1712 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1713 (and only work for Instruction Set Architecture level 2 and higher);
1714 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1718 When this option is used, @command{@value{AS}} will issue a warning every
1719 time it generates a nop instruction from a macro.
1725 The following options are available when @value{AS} is configured for
1731 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1732 The command-line option @samp{-nojsri2bsr} can be used to disable it.
1736 Enable or disable the silicon filter behaviour. By default this is disabled.
1737 The default can be overridden by the @samp{-sifilter} command-line option.
1740 Alter jump instructions for long displacements.
1742 @item -mcpu=[210|340]
1743 Select the cpu type on the target hardware. This controls which instructions
1747 Assemble for a big endian target.
1750 Assemble for a little endian target.
1759 @xref{Meta Options}, for the options available when @value{AS} is configured
1760 for a Meta processor.
1764 @c man begin OPTIONS
1765 The following options are available when @value{AS} is configured for a
1768 @c man begin INCLUDE
1769 @include c-metag.texi
1770 @c ended inside the included file
1775 @c man begin OPTIONS
1777 See the info pages for documentation of the MMIX-specific options.
1783 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1784 for a NDS32 processor.
1786 @c ended inside the included file
1790 @c man begin OPTIONS
1791 The following options are available when @value{AS} is configured for a
1794 @c man begin INCLUDE
1795 @include c-nds32.texi
1796 @c ended inside the included file
1803 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1804 for a PowerPC processor.
1808 @c man begin OPTIONS
1809 The following options are available when @value{AS} is configured for a
1812 @c man begin INCLUDE
1814 @c ended inside the included file
1822 @xref{RISC-V-Options}, for the options available when @value{AS} is configured
1823 for a RISC-V processor.
1827 @c man begin OPTIONS
1828 The following options are available when @value{AS} is configured for a
1831 @c man begin INCLUDE
1832 @include c-riscv.texi
1833 @c ended inside the included file
1838 @c man begin OPTIONS
1840 See the info pages for documentation of the RX-specific options.
1844 The following options are available when @value{AS} is configured for the s390
1850 Select the word size, either 31/32 bits or 64 bits.
1853 Select the architecture mode, either the Enterprise System
1854 Architecture (esa) or the z/Architecture mode (zarch).
1855 @item -march=@var{processor}
1856 Specify which s390 processor variant is the target, @samp{g5} (or
1857 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1858 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1859 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1860 @samp{z13} (or @samp{arch11}), @samp{z14} (or @samp{arch12}), or @samp{z15}
1863 @itemx -mno-regnames
1864 Allow or disallow symbolic names for registers.
1865 @item -mwarn-areg-zero
1866 Warn whenever the operand for a base or index register has been specified
1867 but evaluates to zero.
1875 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1876 for a TMS320C6000 processor.
1880 @c man begin OPTIONS
1881 The following options are available when @value{AS} is configured for a
1882 TMS320C6000 processor.
1884 @c man begin INCLUDE
1885 @include c-tic6x.texi
1886 @c ended inside the included file
1894 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1895 for a TILE-Gx processor.
1899 @c man begin OPTIONS
1900 The following options are available when @value{AS} is configured for a TILE-Gx
1903 @c man begin INCLUDE
1904 @include c-tilegx.texi
1905 @c ended inside the included file
1913 @xref{Visium Options}, for the options available when @value{AS} is configured
1914 for a Visium processor.
1918 @c man begin OPTIONS
1919 The following option is available when @value{AS} is configured for a Visium
1922 @c man begin INCLUDE
1923 @include c-visium.texi
1924 @c ended inside the included file
1932 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1933 for an Xtensa processor.
1937 @c man begin OPTIONS
1938 The following options are available when @value{AS} is configured for an
1941 @c man begin INCLUDE
1942 @include c-xtensa.texi
1943 @c ended inside the included file
1951 @xref{Z80 Options}, for the options available when @value{AS} is configured
1952 for an Z80 processor.
1956 @c man begin OPTIONS
1957 The following options are available when @value{AS} is configured for an
1960 @c man begin INCLUDE
1962 @c ended inside the included file
1968 * Manual:: Structure of this Manual
1969 * GNU Assembler:: The GNU Assembler
1970 * Object Formats:: Object File Formats
1971 * Command Line:: Command Line
1972 * Input Files:: Input Files
1973 * Object:: Output (Object) File
1974 * Errors:: Error and Warning Messages
1978 @section Structure of this Manual
1980 @cindex manual, structure and purpose
1981 This manual is intended to describe what you need to know to use
1982 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1983 notation for symbols, constants, and expressions; the directives that
1984 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1987 We also cover special features in the @value{TARGET}
1988 configuration of @command{@value{AS}}, including assembler directives.
1991 This manual also describes some of the machine-dependent features of
1992 various flavors of the assembler.
1995 @cindex machine instructions (not covered)
1996 On the other hand, this manual is @emph{not} intended as an introduction
1997 to programming in assembly language---let alone programming in general!
1998 In a similar vein, we make no attempt to introduce the machine
1999 architecture; we do @emph{not} describe the instruction set, standard
2000 mnemonics, registers or addressing modes that are standard to a
2001 particular architecture.
2003 You may want to consult the manufacturer's
2004 machine architecture manual for this information.
2008 For information on the H8/300 machine instruction set, see @cite{H8/300
2009 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
2010 Programming Manual} (Renesas).
2013 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
2014 see @cite{SH-Microcomputer User's Manual} (Renesas) or
2015 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
2016 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
2019 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
2023 @c I think this is premature---doc@cygnus.com, 17jan1991
2025 Throughout this manual, we assume that you are running @dfn{GNU},
2026 the portable operating system from the @dfn{Free Software
2027 Foundation, Inc.}. This restricts our attention to certain kinds of
2028 computer (in particular, the kinds of computers that @sc{gnu} can run on);
2029 once this assumption is granted examples and definitions need less
2032 @command{@value{AS}} is part of a team of programs that turn a high-level
2033 human-readable series of instructions into a low-level
2034 computer-readable series of instructions. Different versions of
2035 @command{@value{AS}} are used for different kinds of computer.
2038 @c There used to be a section "Terminology" here, which defined
2039 @c "contents", "byte", "word", and "long". Defining "word" to any
2040 @c particular size is confusing when the .word directive may generate 16
2041 @c bits on one machine and 32 bits on another; in general, for the user
2042 @c version of this manual, none of these terms seem essential to define.
2043 @c They were used very little even in the former draft of the manual;
2044 @c this draft makes an effort to avoid them (except in names of
2048 @section The GNU Assembler
2050 @c man begin DESCRIPTION
2052 @sc{gnu} @command{as} is really a family of assemblers.
2054 This manual describes @command{@value{AS}}, a member of that family which is
2055 configured for the @value{TARGET} architectures.
2057 If you use (or have used) the @sc{gnu} assembler on one architecture, you
2058 should find a fairly similar environment when you use it on another
2059 architecture. Each version has much in common with the others,
2060 including object file formats, most assembler directives (often called
2061 @dfn{pseudo-ops}) and assembler syntax.@refill
2063 @cindex purpose of @sc{gnu} assembler
2064 @command{@value{AS}} is primarily intended to assemble the output of the
2065 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
2066 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
2067 assemble correctly everything that other assemblers for the same
2068 machine would assemble.
2070 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
2073 @c This remark should appear in generic version of manual; assumption
2074 @c here is that generic version sets M680x0.
2075 This doesn't mean @command{@value{AS}} always uses the same syntax as another
2076 assembler for the same architecture; for example, we know of several
2077 incompatible versions of 680x0 assembly language syntax.
2082 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
2083 program in one pass of the source file. This has a subtle impact on the
2084 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
2086 @node Object Formats
2087 @section Object File Formats
2089 @cindex object file format
2090 The @sc{gnu} assembler can be configured to produce several alternative
2091 object file formats. For the most part, this does not affect how you
2092 write assembly language programs; but directives for debugging symbols
2093 are typically different in different file formats. @xref{Symbol
2094 Attributes,,Symbol Attributes}.
2097 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
2098 @value{OBJ-NAME} format object files.
2100 @c The following should exhaust all configs that set MULTI-OBJ, ideally
2102 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
2103 SOM or ELF format object files.
2108 @section Command Line
2110 @cindex command line conventions
2112 After the program name @command{@value{AS}}, the command line may contain
2113 options and file names. Options may appear in any order, and may be
2114 before, after, or between file names. The order of file names is
2117 @cindex standard input, as input file
2119 @file{--} (two hyphens) by itself names the standard input file
2120 explicitly, as one of the files for @command{@value{AS}} to assemble.
2122 @cindex options, command line
2123 Except for @samp{--} any command-line argument that begins with a
2124 hyphen (@samp{-}) is an option. Each option changes the behavior of
2125 @command{@value{AS}}. No option changes the way another option works. An
2126 option is a @samp{-} followed by one or more letters; the case of
2127 the letter is important. All options are optional.
2129 Some options expect exactly one file name to follow them. The file
2130 name may either immediately follow the option's letter (compatible
2131 with older assemblers) or it may be the next command argument (@sc{gnu}
2132 standard). These two command lines are equivalent:
2135 @value{AS} -o my-object-file.o mumble.s
2136 @value{AS} -omy-object-file.o mumble.s
2140 @section Input Files
2143 @cindex source program
2144 @cindex files, input
2145 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2146 describe the program input to one run of @command{@value{AS}}. The program may
2147 be in one or more files; how the source is partitioned into files
2148 doesn't change the meaning of the source.
2150 @c I added "con" prefix to "catenation" just to prove I can overcome my
2151 @c APL training... doc@cygnus.com
2152 The source program is a concatenation of the text in all the files, in the
2155 @c man begin DESCRIPTION
2156 Each time you run @command{@value{AS}} it assembles exactly one source
2157 program. The source program is made up of one or more files.
2158 (The standard input is also a file.)
2160 You give @command{@value{AS}} a command line that has zero or more input file
2161 names. The input files are read (from left file name to right). A
2162 command-line argument (in any position) that has no special meaning
2163 is taken to be an input file name.
2165 If you give @command{@value{AS}} no file names it attempts to read one input file
2166 from the @command{@value{AS}} standard input, which is normally your terminal. You
2167 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2170 Use @samp{--} if you need to explicitly name the standard input file
2171 in your command line.
2173 If the source is empty, @command{@value{AS}} produces a small, empty object
2178 @subheading Filenames and Line-numbers
2180 @cindex input file linenumbers
2181 @cindex line numbers, in input files
2182 There are two ways of locating a line in the input file (or files) and
2183 either may be used in reporting error messages. One way refers to a line
2184 number in a physical file; the other refers to a line number in a
2185 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2187 @dfn{Physical files} are those files named in the command line given
2188 to @command{@value{AS}}.
2190 @dfn{Logical files} are simply names declared explicitly by assembler
2191 directives; they bear no relation to physical files. Logical file names help
2192 error messages reflect the original source file, when @command{@value{AS}} source
2193 is itself synthesized from other files. @command{@value{AS}} understands the
2194 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2195 @ref{File,,@code{.file}}.
2198 @section Output (Object) File
2204 Every time you run @command{@value{AS}} it produces an output file, which is
2205 your assembly language program translated into numbers. This file
2206 is the object file. Its default name is @code{a.out}.
2207 You can give it another name by using the @option{-o} option. Conventionally,
2208 object file names end with @file{.o}. The default name is used for historical
2209 reasons: older assemblers were capable of assembling self-contained programs
2210 directly into a runnable program. (For some formats, this isn't currently
2211 possible, but it can be done for the @code{a.out} format.)
2215 The object file is meant for input to the linker @code{@value{LD}}. It contains
2216 assembled program code, information to help @code{@value{LD}} integrate
2217 the assembled program into a runnable file, and (optionally) symbolic
2218 information for the debugger.
2220 @c link above to some info file(s) like the description of a.out.
2221 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2224 @section Error and Warning Messages
2226 @c man begin DESCRIPTION
2228 @cindex error messages
2229 @cindex warning messages
2230 @cindex messages from assembler
2231 @command{@value{AS}} may write warnings and error messages to the standard error
2232 file (usually your terminal). This should not happen when a compiler
2233 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2234 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2235 grave problem that stops the assembly.
2239 @cindex format of warning messages
2240 Warning messages have the format
2243 file_name:@b{NNN}:Warning Message Text
2247 @cindex file names and line numbers, in warnings/errors
2248 (where @b{NNN} is a line number). If both a logical file name
2249 (@pxref{File,,@code{.file}}) and a logical line number
2251 (@pxref{Line,,@code{.line}})
2253 have been given then they will be used, otherwise the file name and line number
2254 in the current assembler source file will be used. The message text is
2255 intended to be self explanatory (in the grand Unix tradition).
2257 Note the file name must be set via the logical version of the @code{.file}
2258 directive, not the DWARF2 version of the @code{.file} directive. For example:
2262 error_assembler_source
2268 produces this output:
2272 asm.s:2: Error: no such instruction: `error_assembler_source'
2273 foo.c:31: Error: no such instruction: `error_c_source'
2276 @cindex format of error messages
2277 Error messages have the format
2280 file_name:@b{NNN}:FATAL:Error Message Text
2283 The file name and line number are derived as for warning
2284 messages. The actual message text may be rather less explanatory
2285 because many of them aren't supposed to happen.
2288 @chapter Command-Line Options
2290 @cindex options, all versions of assembler
2291 This chapter describes command-line options available in @emph{all}
2292 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2293 for options specific
2295 to the @value{TARGET} target.
2298 to particular machine architectures.
2301 @c man begin DESCRIPTION
2303 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2304 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2305 The assembler arguments must be separated from each other (and the @samp{-Wa})
2306 by commas. For example:
2309 gcc -c -g -O -Wa,-alh,-L file.c
2313 This passes two options to the assembler: @samp{-alh} (emit a listing to
2314 standard output with high-level and assembly source) and @samp{-L} (retain
2315 local symbols in the symbol table).
2317 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2318 command-line options are automatically passed to the assembler by the compiler.
2319 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2320 precisely what options it passes to each compilation pass, including the
2326 * a:: -a[cdghlns] enable listings
2327 * alternate:: --alternate enable alternate macro syntax
2328 * D:: -D for compatibility
2329 * f:: -f to work faster
2330 * I:: -I for .include search path
2331 @ifclear DIFF-TBL-KLUGE
2332 * K:: -K for compatibility
2334 @ifset DIFF-TBL-KLUGE
2335 * K:: -K for difference tables
2338 * L:: -L to retain local symbols
2339 * listing:: --listing-XXX to configure listing output
2340 * M:: -M or --mri to assemble in MRI compatibility mode
2341 * MD:: --MD for dependency tracking
2342 * no-pad-sections:: --no-pad-sections to stop section padding
2343 * o:: -o to name the object file
2344 * R:: -R to join data and text sections
2345 * statistics:: --statistics to see statistics about assembly
2346 * traditional-format:: --traditional-format for compatible output
2347 * v:: -v to announce version
2348 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2349 * Z:: -Z to make object file even after errors
2353 @section Enable Listings: @option{-a[cdghlns]}
2363 @cindex listings, enabling
2364 @cindex assembly listings, enabling
2366 These options enable listing output from the assembler. By itself,
2367 @samp{-a} requests high-level, assembly, and symbols listing.
2368 You can use other letters to select specific options for the list:
2369 @samp{-ah} requests a high-level language listing,
2370 @samp{-al} requests an output-program assembly listing, and
2371 @samp{-as} requests a symbol table listing.
2372 High-level listings require that a compiler debugging option like
2373 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2376 Use the @samp{-ag} option to print a first section with general assembly
2377 information, like @value{AS} version, switches passed, or time stamp.
2379 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2380 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2381 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2382 omitted from the listing.
2384 Use the @samp{-ad} option to omit debugging directives from the
2387 Once you have specified one of these options, you can further control
2388 listing output and its appearance using the directives @code{.list},
2389 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2391 The @samp{-an} option turns off all forms processing.
2392 If you do not request listing output with one of the @samp{-a} options, the
2393 listing-control directives have no effect.
2395 The letters after @samp{-a} may be combined into one option,
2396 @emph{e.g.}, @samp{-aln}.
2398 Note if the assembler source is coming from the standard input (e.g.,
2400 is being created by @code{@value{GCC}} and the @samp{-pipe} command-line switch
2401 is being used) then the listing will not contain any comments or preprocessor
2402 directives. This is because the listing code buffers input source lines from
2403 stdin only after they have been preprocessed by the assembler. This reduces
2404 memory usage and makes the code more efficient.
2407 @section @option{--alternate}
2410 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2413 @section @option{-D}
2416 This option has no effect whatsoever, but it is accepted to make it more
2417 likely that scripts written for other assemblers also work with
2418 @command{@value{AS}}.
2421 @section Work Faster: @option{-f}
2424 @cindex trusted compiler
2425 @cindex faster processing (@option{-f})
2426 @samp{-f} should only be used when assembling programs written by a
2427 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2428 and comment preprocessing on
2429 the input file(s) before assembling them. @xref{Preprocessing,
2433 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2434 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2439 @section @code{.include} Search Path: @option{-I} @var{path}
2441 @kindex -I @var{path}
2442 @cindex paths for @code{.include}
2443 @cindex search path for @code{.include}
2444 @cindex @code{include} directive search path
2445 Use this option to add a @var{path} to the list of directories
2446 @command{@value{AS}} searches for files specified in @code{.include}
2447 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2448 many times as necessary to include a variety of paths. The current
2449 working directory is always searched first; after that, @command{@value{AS}}
2450 searches any @samp{-I} directories in the same order as they were
2451 specified (left to right) on the command line.
2454 @section Difference Tables: @option{-K}
2457 @ifclear DIFF-TBL-KLUGE
2458 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2459 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2460 where it can be used to warn when the assembler alters the machine code
2461 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2462 family does not have the addressing limitations that sometimes lead to this
2463 alteration on other platforms.
2466 @ifset DIFF-TBL-KLUGE
2467 @cindex difference tables, warning
2468 @cindex warning for altered difference tables
2469 @command{@value{AS}} sometimes alters the code emitted for directives of the
2470 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2471 You can use the @samp{-K} option if you want a warning issued when this
2476 @section Include Local Symbols: @option{-L}
2479 @cindex local symbols, retaining in output
2480 Symbols beginning with system-specific local label prefixes, typically
2481 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2482 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2483 such symbols when debugging, because they are intended for the use of
2484 programs (like compilers) that compose assembler programs, not for your
2485 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2486 such symbols, so you do not normally debug with them.
2488 This option tells @command{@value{AS}} to retain those local symbols
2489 in the object file. Usually if you do this you also tell the linker
2490 @code{@value{LD}} to preserve those symbols.
2493 @section Configuring listing output: @option{--listing}
2495 The listing feature of the assembler can be enabled via the command-line switch
2496 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2497 hex dump of the corresponding locations in the output object file, and displays
2498 them as a listing file. The format of this listing can be controlled by
2499 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2500 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2501 @code{.psize} (@pxref{Psize}), and
2502 @code{.eject} (@pxref{Eject}) and also by the following switches:
2505 @item --listing-lhs-width=@samp{number}
2506 @kindex --listing-lhs-width
2507 @cindex Width of first line disassembly output
2508 Sets the maximum width, in words, of the first line of the hex byte dump. This
2509 dump appears on the left hand side of the listing output.
2511 @item --listing-lhs-width2=@samp{number}
2512 @kindex --listing-lhs-width2
2513 @cindex Width of continuation lines of disassembly output
2514 Sets the maximum width, in words, of any further lines of the hex byte dump for
2515 a given input source line. If this value is not specified, it defaults to being
2516 the same as the value specified for @samp{--listing-lhs-width}. If neither
2517 switch is used the default is to one.
2519 @item --listing-rhs-width=@samp{number}
2520 @kindex --listing-rhs-width
2521 @cindex Width of source line output
2522 Sets the maximum width, in characters, of the source line that is displayed
2523 alongside the hex dump. The default value for this parameter is 100. The
2524 source line is displayed on the right hand side of the listing output.
2526 @item --listing-cont-lines=@samp{number}
2527 @kindex --listing-cont-lines
2528 @cindex Maximum number of continuation lines
2529 Sets the maximum number of continuation lines of hex dump that will be
2530 displayed for a given single line of source input. The default value is 4.
2534 @section Assemble in MRI Compatibility Mode: @option{-M}
2537 @cindex MRI compatibility mode
2538 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2539 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2540 compatible with the @code{ASM68K} assembler from Microtec Research.
2541 The exact nature of the
2542 MRI syntax will not be documented here; see the MRI manuals for more
2543 information. Note in particular that the handling of macros and macro
2544 arguments is somewhat different. The purpose of this option is to permit
2545 assembling existing MRI assembler code using @command{@value{AS}}.
2547 The MRI compatibility is not complete. Certain operations of the MRI assembler
2548 depend upon its object file format, and can not be supported using other object
2549 file formats. Supporting these would require enhancing each object file format
2550 individually. These are:
2553 @item global symbols in common section
2555 The m68k MRI assembler supports common sections which are merged by the linker.
2556 Other object file formats do not support this. @command{@value{AS}} handles
2557 common sections by treating them as a single common symbol. It permits local
2558 symbols to be defined within a common section, but it can not support global
2559 symbols, since it has no way to describe them.
2561 @item complex relocations
2563 The MRI assemblers support relocations against a negated section address, and
2564 relocations which combine the start addresses of two or more sections. These
2565 are not support by other object file formats.
2567 @item @code{END} pseudo-op specifying start address
2569 The MRI @code{END} pseudo-op permits the specification of a start address.
2570 This is not supported by other object file formats. The start address may
2571 instead be specified using the @option{-e} option to the linker, or in a linker
2574 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2576 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2577 name to the output file. This is not supported by other object file formats.
2579 @item @code{ORG} pseudo-op
2581 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2582 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2583 which changes the location within the current section. Absolute sections are
2584 not supported by other object file formats. The address of a section may be
2585 assigned within a linker script.
2588 There are some other features of the MRI assembler which are not supported by
2589 @command{@value{AS}}, typically either because they are difficult or because they
2590 seem of little consequence. Some of these may be supported in future releases.
2594 @item EBCDIC strings
2596 EBCDIC strings are not supported.
2598 @item packed binary coded decimal
2600 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2601 and @code{DCB.P} pseudo-ops are not supported.
2603 @item @code{FEQU} pseudo-op
2605 The m68k @code{FEQU} pseudo-op is not supported.
2607 @item @code{NOOBJ} pseudo-op
2609 The m68k @code{NOOBJ} pseudo-op is not supported.
2611 @item @code{OPT} branch control options
2613 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2614 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2615 relaxes all branches, whether forward or backward, to an appropriate size, so
2616 these options serve no purpose.
2618 @item @code{OPT} list control options
2620 The following m68k @code{OPT} list control options are ignored: @code{C},
2621 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2622 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2624 @item other @code{OPT} options
2626 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2627 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2629 @item @code{OPT} @code{D} option is default
2631 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2632 @code{OPT NOD} may be used to turn it off.
2634 @item @code{XREF} pseudo-op.
2636 The m68k @code{XREF} pseudo-op is ignored.
2641 @section Dependency Tracking: @option{--MD}
2644 @cindex dependency tracking
2647 @command{@value{AS}} can generate a dependency file for the file it creates. This
2648 file consists of a single rule suitable for @code{make} describing the
2649 dependencies of the main source file.
2651 The rule is written to the file named in its argument.
2653 This feature is used in the automatic updating of makefiles.
2655 @node no-pad-sections
2656 @section Output Section Padding
2657 @kindex --no-pad-sections
2658 @cindex output section padding
2659 Normally the assembler will pad the end of each output section up to its
2660 alignment boundary. But this can waste space, which can be significant on
2661 memory constrained targets. So the @option{--no-pad-sections} option will
2662 disable this behaviour.
2665 @section Name the Object File: @option{-o}
2668 @cindex naming object file
2669 @cindex object file name
2670 There is always one object file output when you run @command{@value{AS}}. By
2671 default it has the name @file{a.out}.
2672 You use this option (which takes exactly one filename) to give the
2673 object file a different name.
2675 Whatever the object file is called, @command{@value{AS}} overwrites any
2676 existing file of the same name.
2679 @section Join Data and Text Sections: @option{-R}
2682 @cindex data and text sections, joining
2683 @cindex text and data sections, joining
2684 @cindex joining text and data sections
2685 @cindex merging text and data sections
2686 @option{-R} tells @command{@value{AS}} to write the object file as if all
2687 data-section data lives in the text section. This is only done at
2688 the very last moment: your binary data are the same, but data
2689 section parts are relocated differently. The data section part of
2690 your object file is zero bytes long because all its bytes are
2691 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2693 When you specify @option{-R} it would be possible to generate shorter
2694 address displacements (because we do not have to cross between text and
2695 data section). We refrain from doing this simply for compatibility with
2696 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2699 When @command{@value{AS}} is configured for COFF or ELF output,
2700 this option is only useful if you use sections named @samp{.text} and
2705 @option{-R} is not supported for any of the HPPA targets. Using
2706 @option{-R} generates a warning from @command{@value{AS}}.
2710 @section Display Assembly Statistics: @option{--statistics}
2712 @kindex --statistics
2713 @cindex statistics, about assembly
2714 @cindex time, total for assembly
2715 @cindex space used, maximum for assembly
2716 Use @samp{--statistics} to display two statistics about the resources used by
2717 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2718 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2721 @node traditional-format
2722 @section Compatible Output: @option{--traditional-format}
2724 @kindex --traditional-format
2725 For some targets, the output of @command{@value{AS}} is different in some ways
2726 from the output of some existing assembler. This switch requests
2727 @command{@value{AS}} to use the traditional format instead.
2729 For example, it disables the exception frame optimizations which
2730 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2733 @section Announce Version: @option{-v}
2737 @cindex assembler version
2738 @cindex version of assembler
2739 You can find out what version of as is running by including the
2740 option @samp{-v} (which you can also spell as @samp{-version}) on the
2744 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2746 @command{@value{AS}} should never give a warning or error message when
2747 assembling compiler output. But programs written by people often
2748 cause @command{@value{AS}} to give a warning that a particular assumption was
2749 made. All such warnings are directed to the standard error file.
2753 @cindex suppressing warnings
2754 @cindex warnings, suppressing
2755 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2756 This only affects the warning messages: it does not change any particular of
2757 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2760 @kindex --fatal-warnings
2761 @cindex errors, caused by warnings
2762 @cindex warnings, causing error
2763 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2764 files that generate warnings to be in error.
2767 @cindex warnings, switching on
2768 You can switch these options off again by specifying @option{--warn}, which
2769 causes warnings to be output as usual.
2772 @section Generate Object File in Spite of Errors: @option{-Z}
2773 @cindex object file, after errors
2774 @cindex errors, continuing after
2775 After an error message, @command{@value{AS}} normally produces no output. If for
2776 some reason you are interested in object file output even after
2777 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2778 option. If there are any errors, @command{@value{AS}} continues anyways, and
2779 writes an object file after a final warning message of the form @samp{@var{n}
2780 errors, @var{m} warnings, generating bad object file.}
2785 @cindex machine-independent syntax
2786 @cindex syntax, machine-independent
2787 This chapter describes the machine-independent syntax allowed in a
2788 source file. @command{@value{AS}} syntax is similar to what many other
2789 assemblers use; it is inspired by the BSD 4.2
2794 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2798 * Preprocessing:: Preprocessing
2799 * Whitespace:: Whitespace
2800 * Comments:: Comments
2801 * Symbol Intro:: Symbols
2802 * Statements:: Statements
2803 * Constants:: Constants
2807 @section Preprocessing
2809 @cindex preprocessing
2810 The @command{@value{AS}} internal preprocessor:
2812 @cindex whitespace, removed by preprocessor
2814 adjusts and removes extra whitespace. It leaves one space or tab before
2815 the keywords on a line, and turns any other whitespace on the line into
2818 @cindex comments, removed by preprocessor
2820 removes all comments, replacing them with a single space, or an
2821 appropriate number of newlines.
2823 @cindex constants, converted by preprocessor
2825 converts character constants into the appropriate numeric values.
2828 It does not do macro processing, include file handling, or
2829 anything else you may get from your C compiler's preprocessor. You can
2830 do include file processing with the @code{.include} directive
2831 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2832 to get other ``CPP'' style preprocessing by giving the input file a
2833 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2834 Output, gcc info, Using GNU CC}.
2836 Excess whitespace, comments, and character constants
2837 cannot be used in the portions of the input text that are not
2840 @cindex turning preprocessing on and off
2841 @cindex preprocessing, turning on and off
2844 If the first line of an input file is @code{#NO_APP} or if you use the
2845 @samp{-f} option, whitespace and comments are not removed from the input file.
2846 Within an input file, you can ask for whitespace and comment removal in
2847 specific portions of the by putting a line that says @code{#APP} before the
2848 text that may contain whitespace or comments, and putting a line that says
2849 @code{#NO_APP} after this text. This feature is mainly intend to support
2850 @code{asm} statements in compilers whose output is otherwise free of comments
2857 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2858 Whitespace is used to separate symbols, and to make programs neater for
2859 people to read. Unless within character constants
2860 (@pxref{Characters,,Character Constants}), any whitespace means the same
2861 as exactly one space.
2867 There are two ways of rendering comments to @command{@value{AS}}. In both
2868 cases the comment is equivalent to one space.
2870 Anything from @samp{/*} through the next @samp{*/} is a comment.
2871 This means you may not nest these comments.
2875 The only way to include a newline ('\n') in a comment
2876 is to use this sort of comment.
2879 /* This sort of comment does not nest. */
2882 @cindex line comment character
2883 Anything from a @dfn{line comment} character up to the next newline is
2884 considered a comment and is ignored. The line comment character is target
2885 specific, and some targets multiple comment characters. Some targets also have
2886 line comment characters that only work if they are the first character on a
2887 line. Some targets use a sequence of two characters to introduce a line
2888 comment. Some targets can also change their line comment characters depending
2889 upon command-line options that have been used. For more details see the
2890 @emph{Syntax} section in the documentation for individual targets.
2892 If the line comment character is the hash sign (@samp{#}) then it still has the
2893 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2894 to specify logical line numbers:
2897 @cindex lines starting with @code{#}
2898 @cindex logical line numbers
2899 To be compatible with past assemblers, lines that begin with @samp{#} have a
2900 special interpretation. Following the @samp{#} should be an absolute
2901 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2902 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2903 new logical file name. The rest of the line, if any, should be whitespace.
2905 If the first non-whitespace characters on the line are not numeric,
2906 the line is ignored. (Just like a comment.)
2909 # This is an ordinary comment.
2910 # 42-6 "new_file_name" # New logical file name
2911 # This is logical line # 36.
2913 This feature is deprecated, and may disappear from future versions
2914 of @command{@value{AS}}.
2919 @cindex characters used in symbols
2920 @ifclear SPECIAL-SYMS
2921 A @dfn{symbol} is one or more characters chosen from the set of all
2922 letters (both upper and lower case), digits and the three characters
2928 A @dfn{symbol} is one or more characters chosen from the set of all
2929 letters (both upper and lower case), digits and the three characters
2930 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2936 On most machines, you can also use @code{$} in symbol names; exceptions
2937 are noted in @ref{Machine Dependencies}.
2939 No symbol may begin with a digit. Case is significant.
2940 There is no length limit; all characters are significant. Multibyte characters
2941 are supported. Symbols are delimited by characters not in that set, or by the
2942 beginning of a file (since the source program must end with a newline, the end
2943 of a file is not a possible symbol delimiter). @xref{Symbols}.
2945 Symbol names may also be enclosed in double quote @code{"} characters. In such
2946 cases any characters are allowed, except for the NUL character. If a double
2947 quote character is to be included in the symbol name it must be preceeded by a
2948 backslash @code{\} character.
2949 @cindex length of symbols
2954 @cindex statements, structure of
2955 @cindex line separator character
2956 @cindex statement separator character
2958 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2959 @dfn{line separator character}. The line separator character is target
2960 specific and described in the @emph{Syntax} section of each
2961 target's documentation. Not all targets support a line separator character.
2962 The newline or line separator character is considered to be part of the
2963 preceding statement. Newlines and separators within character constants are an
2964 exception: they do not end statements.
2966 @cindex newline, required at file end
2967 @cindex EOF, newline must precede
2968 It is an error to end any statement with end-of-file: the last
2969 character of any input file should be a newline.@refill
2971 An empty statement is allowed, and may include whitespace. It is ignored.
2973 @cindex instructions and directives
2974 @cindex directives and instructions
2975 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2976 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2978 A statement begins with zero or more labels, optionally followed by a
2979 key symbol which determines what kind of statement it is. The key
2980 symbol determines the syntax of the rest of the statement. If the
2981 symbol begins with a dot @samp{.} then the statement is an assembler
2982 directive: typically valid for any computer. If the symbol begins with
2983 a letter the statement is an assembly language @dfn{instruction}: it
2984 assembles into a machine language instruction.
2986 Different versions of @command{@value{AS}} for different computers
2987 recognize different instructions. In fact, the same symbol may
2988 represent a different instruction in a different computer's assembly
2992 @cindex @code{:} (label)
2993 @cindex label (@code{:})
2994 A label is a symbol immediately followed by a colon (@code{:}).
2995 Whitespace before a label or after a colon is permitted, but you may not
2996 have whitespace between a label's symbol and its colon. @xref{Labels}.
2999 For HPPA targets, labels need not be immediately followed by a colon, but
3000 the definition of a label must begin in column zero. This also implies that
3001 only one label may be defined on each line.
3005 label: .directive followed by something
3006 another_label: # This is an empty statement.
3007 instruction operand_1, operand_2, @dots{}
3014 A constant is a number, written so that its value is known by
3015 inspection, without knowing any context. Like this:
3018 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
3019 .ascii "Ring the bell\7" # A string constant.
3020 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
3021 .float 0f-314159265358979323846264338327\
3022 95028841971.693993751E-40 # - pi, a flonum.
3027 * Characters:: Character Constants
3028 * Numbers:: Number Constants
3032 @subsection Character Constants
3034 @cindex character constants
3035 @cindex constants, character
3036 There are two kinds of character constants. A @dfn{character} stands
3037 for one character in one byte and its value may be used in
3038 numeric expressions. String constants (properly called string
3039 @emph{literals}) are potentially many bytes and their values may not be
3040 used in arithmetic expressions.
3044 * Chars:: Characters
3048 @subsubsection Strings
3050 @cindex string constants
3051 @cindex constants, string
3052 A @dfn{string} is written between double-quotes. It may contain
3053 double-quotes or null characters. The way to get special characters
3054 into a string is to @dfn{escape} these characters: precede them with
3055 a backslash @samp{\} character. For example @samp{\\} represents
3056 one backslash: the first @code{\} is an escape which tells
3057 @command{@value{AS}} to interpret the second character literally as a backslash
3058 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
3059 escape character). The complete list of escapes follows.
3061 @cindex escape codes, character
3062 @cindex character escape codes
3063 @c NOTE: Cindex entries must not start with a backlash character.
3064 @c NOTE: This confuses the pdf2texi script when it is creating the
3065 @c NOTE: index based upon the first character and so it generates:
3066 @c NOTE: \initial {\\}
3067 @c NOTE: which then results in the error message:
3068 @c NOTE: Argument of \\ has an extra }.
3069 @c NOTE: So in the index entries below a space character has been
3070 @c NOTE: prepended to avoid this problem.
3073 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
3075 @cindex @code{ \b} (backspace character)
3076 @cindex backspace (@code{\b})
3078 Mnemonic for backspace; for ASCII this is octal code 010.
3081 @c Mnemonic for EOText; for ASCII this is octal code 004.
3083 @cindex @code{ \f} (formfeed character)
3084 @cindex formfeed (@code{\f})
3086 Mnemonic for FormFeed; for ASCII this is octal code 014.
3088 @cindex @code{ \n} (newline character)
3089 @cindex newline (@code{\n})
3091 Mnemonic for newline; for ASCII this is octal code 012.
3094 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
3096 @cindex @code{ \r} (carriage return character)
3097 @cindex carriage return (@code{backslash-r})
3099 Mnemonic for carriage-Return; for ASCII this is octal code 015.
3102 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
3103 @c other assemblers.
3105 @cindex @code{ \t} (tab)
3106 @cindex tab (@code{\t})
3108 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
3111 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
3112 @c @item \x @var{digit} @var{digit} @var{digit}
3113 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
3115 @cindex @code{ \@var{ddd}} (octal character code)
3116 @cindex octal character code (@code{\@var{ddd}})
3117 @item \ @var{digit} @var{digit} @var{digit}
3118 An octal character code. The numeric code is 3 octal digits.
3119 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
3120 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3122 @cindex @code{ \@var{xd...}} (hex character code)
3123 @cindex hex character code (@code{\@var{xd...}})
3124 @item \@code{x} @var{hex-digits...}
3125 A hex character code. All trailing hex digits are combined. Either upper or
3126 lower case @code{x} works.
3128 @cindex @code{ \\} (@samp{\} character)
3129 @cindex backslash (@code{\\})
3131 Represents one @samp{\} character.
3134 @c Represents one @samp{'} (accent acute) character.
3135 @c This is needed in single character literals
3136 @c (@xref{Characters,,Character Constants}.) to represent
3139 @cindex @code{ \"} (doublequote character)
3140 @cindex doublequote (@code{\"})
3142 Represents one @samp{"} character. Needed in strings to represent
3143 this character, because an unescaped @samp{"} would end the string.
3145 @item \ @var{anything-else}
3146 Any other character when escaped by @kbd{\} gives a warning, but
3147 assembles as if the @samp{\} was not present. The idea is that if
3148 you used an escape sequence you clearly didn't want the literal
3149 interpretation of the following character. However @command{@value{AS}} has no
3150 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3151 code and warns you of the fact.
3154 Which characters are escapable, and what those escapes represent,
3155 varies widely among assemblers. The current set is what we think
3156 the BSD 4.2 assembler recognizes, and is a subset of what most C
3157 compilers recognize. If you are in doubt, do not use an escape
3161 @subsubsection Characters
3163 @cindex single character constant
3164 @cindex character, single
3165 @cindex constant, single character
3166 A single character may be written as a single quote immediately followed by
3167 that character. Some backslash escapes apply to characters, @code{\b},
3168 @code{\f}, @code{\n}, @code{\r}, @code{\t}, and @code{\"} with the same meaning
3169 as for strings, plus @code{\'} for a single quote. So if you want to write the
3170 character backslash, you must write @kbd{'\\} where the first @code{\} escapes
3171 the second @code{\}. As you can see, the quote is an acute accent, not a grave
3174 @ifclear abnormal-separator
3175 (or semicolon @samp{;})
3177 @ifset abnormal-separator
3179 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3184 immediately following an acute accent is taken as a literal character
3185 and does not count as the end of a statement. The value of a character
3186 constant in a numeric expression is the machine's byte-wide code for
3187 that character. @command{@value{AS}} assumes your character code is ASCII:
3188 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3191 @subsection Number Constants
3193 @cindex constants, number
3194 @cindex number constants
3195 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3196 are stored in the target machine. @emph{Integers} are numbers that
3197 would fit into an @code{int} in the C language. @emph{Bignums} are
3198 integers, but they are stored in more than 32 bits. @emph{Flonums}
3199 are floating point numbers, described below.
3202 * Integers:: Integers
3210 @subsubsection Integers
3212 @cindex constants, integer
3214 @cindex binary integers
3215 @cindex integers, binary
3216 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3217 the binary digits @samp{01}.
3219 @cindex octal integers
3220 @cindex integers, octal
3221 An octal integer is @samp{0} followed by zero or more of the octal
3222 digits (@samp{01234567}).
3224 @cindex decimal integers
3225 @cindex integers, decimal
3226 A decimal integer starts with a non-zero digit followed by zero or
3227 more digits (@samp{0123456789}).
3229 @cindex hexadecimal integers
3230 @cindex integers, hexadecimal
3231 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3232 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3234 Integers have the usual values. To denote a negative integer, use
3235 the prefix operator @samp{-} discussed under expressions
3236 (@pxref{Prefix Ops,,Prefix Operators}).
3239 @subsubsection Bignums
3242 @cindex constants, bignum
3243 A @dfn{bignum} has the same syntax and semantics as an integer
3244 except that the number (or its negative) takes more than 32 bits to
3245 represent in binary. The distinction is made because in some places
3246 integers are permitted while bignums are not.
3249 @subsubsection Flonums
3251 @cindex floating point numbers
3252 @cindex constants, floating point
3254 @cindex precision, floating point
3255 A @dfn{flonum} represents a floating point number. The translation is
3256 indirect: a decimal floating point number from the text is converted by
3257 @command{@value{AS}} to a generic binary floating point number of more than
3258 sufficient precision. This generic floating point number is converted
3259 to a particular computer's floating point format (or formats) by a
3260 portion of @command{@value{AS}} specialized to that computer.
3262 A flonum is written by writing (in order)
3267 (@samp{0} is optional on the HPPA.)
3271 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3273 @kbd{e} is recommended. Case is not important.
3275 @c FIXME: verify if flonum syntax really this vague for most cases
3276 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3277 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3280 On the H8/300 and Renesas / SuperH SH architectures, the letter must be
3281 one of the letters @samp{DFPRSX} (in upper or lower case).
3283 On the ARC, the letter must be one of the letters @samp{DFRS}
3284 (in upper or lower case).
3286 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3290 One of the letters @samp{DFRS} (in upper or lower case).
3293 One of the letters @samp{DFPRSX} (in upper or lower case).
3296 The letter @samp{E} (upper case only).
3301 An optional sign: either @samp{+} or @samp{-}.
3304 An optional @dfn{integer part}: zero or more decimal digits.
3307 An optional @dfn{fractional part}: @samp{.} followed by zero
3308 or more decimal digits.
3311 An optional exponent, consisting of:
3315 An @samp{E} or @samp{e}.
3316 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3317 @c principle this can perfectly well be different on different targets.
3319 Optional sign: either @samp{+} or @samp{-}.
3321 One or more decimal digits.
3326 At least one of the integer part or the fractional part must be
3327 present. The floating point number has the usual base-10 value.
3329 @command{@value{AS}} does all processing using integers. Flonums are computed
3330 independently of any floating point hardware in the computer running
3331 @command{@value{AS}}.
3334 @chapter Sections and Relocation
3339 * Secs Background:: Background
3340 * Ld Sections:: Linker Sections
3341 * As Sections:: Assembler Internal Sections
3342 * Sub-Sections:: Sub-Sections
3346 @node Secs Background
3349 Roughly, a section is a range of addresses, with no gaps; all data
3350 ``in'' those addresses is treated the same for some particular purpose.
3351 For example there may be a ``read only'' section.
3353 @cindex linker, and assembler
3354 @cindex assembler, and linker
3355 The linker @code{@value{LD}} reads many object files (partial programs) and
3356 combines their contents to form a runnable program. When @command{@value{AS}}
3357 emits an object file, the partial program is assumed to start at address 0.
3358 @code{@value{LD}} assigns the final addresses for the partial program, so that
3359 different partial programs do not overlap. This is actually an
3360 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3363 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3364 addresses. These blocks slide to their run-time addresses as rigid
3365 units; their length does not change and neither does the order of bytes
3366 within them. Such a rigid unit is called a @emph{section}. Assigning
3367 run-time addresses to sections is called @dfn{relocation}. It includes
3368 the task of adjusting mentions of object-file addresses so they refer to
3369 the proper run-time addresses.
3371 For the H8/300, and for the Renesas / SuperH SH,
3372 @command{@value{AS}} pads sections if needed to
3373 ensure they end on a word (sixteen bit) boundary.
3376 @cindex standard assembler sections
3377 An object file written by @command{@value{AS}} has at least three sections, any
3378 of which may be empty. These are named @dfn{text}, @dfn{data} and
3383 When it generates COFF or ELF output,
3385 @command{@value{AS}} can also generate whatever other named sections you specify
3386 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3387 If you do not use any directives that place output in the @samp{.text}
3388 or @samp{.data} sections, these sections still exist, but are empty.
3393 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3395 @command{@value{AS}} can also generate whatever other named sections you
3396 specify using the @samp{.space} and @samp{.subspace} directives. See
3397 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3398 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3399 assembler directives.
3402 Additionally, @command{@value{AS}} uses different names for the standard
3403 text, data, and bss sections when generating SOM output. Program text
3404 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3405 BSS into @samp{$BSS$}.
3409 Within the object file, the text section starts at address @code{0}, the
3410 data section follows, and the bss section follows the data section.
3413 When generating either SOM or ELF output files on the HPPA, the text
3414 section starts at address @code{0}, the data section at address
3415 @code{0x4000000}, and the bss section follows the data section.
3418 To let @code{@value{LD}} know which data changes when the sections are
3419 relocated, and how to change that data, @command{@value{AS}} also writes to the
3420 object file details of the relocation needed. To perform relocation
3421 @code{@value{LD}} must know, each time an address in the object
3425 Where in the object file is the beginning of this reference to
3428 How long (in bytes) is this reference?
3430 Which section does the address refer to? What is the numeric value of
3432 (@var{address}) @minus{} (@var{start-address of section})?
3435 Is the reference to an address ``Program-Counter relative''?
3438 @cindex addresses, format of
3439 @cindex section-relative addressing
3440 In fact, every address @command{@value{AS}} ever uses is expressed as
3442 (@var{section}) + (@var{offset into section})
3445 Further, most expressions @command{@value{AS}} computes have this section-relative
3448 (For some object formats, such as SOM for the HPPA, some expressions are
3449 symbol-relative instead.)
3452 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3453 @var{N} into section @var{secname}.''
3455 Apart from text, data and bss sections you need to know about the
3456 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3457 addresses in the absolute section remain unchanged. For example, address
3458 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3459 @code{@value{LD}}. Although the linker never arranges two partial programs'
3460 data sections with overlapping addresses after linking, @emph{by definition}
3461 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3462 part of a program is always the same address when the program is running as
3463 address @code{@{absolute@ 239@}} in any other part of the program.
3465 The idea of sections is extended to the @dfn{undefined} section. Any
3466 address whose section is unknown at assembly time is by definition
3467 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3468 Since numbers are always defined, the only way to generate an undefined
3469 address is to mention an undefined symbol. A reference to a named
3470 common block would be such a symbol: its value is unknown at assembly
3471 time so it has section @emph{undefined}.
3473 By analogy the word @emph{section} is used to describe groups of sections in
3474 the linked program. @code{@value{LD}} puts all partial programs' text
3475 sections in contiguous addresses in the linked program. It is
3476 customary to refer to the @emph{text section} of a program, meaning all
3477 the addresses of all partial programs' text sections. Likewise for
3478 data and bss sections.
3480 Some sections are manipulated by @code{@value{LD}}; others are invented for
3481 use of @command{@value{AS}} and have no meaning except during assembly.
3484 @section Linker Sections
3485 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3490 @cindex named sections
3491 @cindex sections, named
3492 @item named sections
3495 @cindex text section
3496 @cindex data section
3500 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3501 separate but equal sections. Anything you can say of one section is
3504 When the program is running, however, it is
3505 customary for the text section to be unalterable. The
3506 text section is often shared among processes: it contains
3507 instructions, constants and the like. The data section of a running
3508 program is usually alterable: for example, C variables would be stored
3509 in the data section.
3514 This section contains zeroed bytes when your program begins running. It
3515 is used to hold uninitialized variables or common storage. The length of
3516 each partial program's bss section is important, but because it starts
3517 out containing zeroed bytes there is no need to store explicit zero
3518 bytes in the object file. The bss section was invented to eliminate
3519 those explicit zeros from object files.
3521 @cindex absolute section
3522 @item absolute section
3523 Address 0 of this section is always ``relocated'' to runtime address 0.
3524 This is useful if you want to refer to an address that @code{@value{LD}} must
3525 not change when relocating. In this sense we speak of absolute
3526 addresses being ``unrelocatable'': they do not change during relocation.
3528 @cindex undefined section
3529 @item undefined section
3530 This ``section'' is a catch-all for address references to objects not in
3531 the preceding sections.
3532 @c FIXME: ref to some other doc on obj-file formats could go here.
3535 @cindex relocation example
3536 An idealized example of three relocatable sections follows.
3538 The example uses the traditional section names @samp{.text} and @samp{.data}.
3540 Memory addresses are on the horizontal axis.
3544 @c END TEXI2ROFF-KILL
3547 partial program # 1: |ttttt|dddd|00|
3554 partial program # 2: |TTT|DDD|000|
3557 +--+---+-----+--+----+---+-----+~~
3558 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3559 +--+---+-----+--+----+---+-----+~~
3561 addresses: 0 @dots{}
3568 \line{\it Partial program \#1: \hfil}
3569 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3570 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3572 \line{\it Partial program \#2: \hfil}
3573 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3574 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3576 \line{\it linked program: \hfil}
3577 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3578 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3579 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3580 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3582 \line{\it addresses: \hfil}
3586 @c END TEXI2ROFF-KILL
3589 @section Assembler Internal Sections
3591 @cindex internal assembler sections
3592 @cindex sections in messages, internal
3593 These sections are meant only for the internal use of @command{@value{AS}}. They
3594 have no meaning at run-time. You do not really need to know about these
3595 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3596 warning messages, so it might be helpful to have an idea of their
3597 meanings to @command{@value{AS}}. These sections are used to permit the
3598 value of every expression in your assembly language program to be a
3599 section-relative address.
3602 @cindex assembler internal logic error
3603 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3604 An internal assembler logic error has been found. This means there is a
3605 bug in the assembler.
3607 @cindex expr (internal section)
3609 The assembler stores complex expression internally as combinations of
3610 symbols. When it needs to represent an expression as a symbol, it puts
3611 it in the expr section.
3613 @c FIXME item transfer[t] vector preload
3614 @c FIXME item transfer[t] vector postload
3615 @c FIXME item register
3619 @section Sub-Sections
3621 @cindex numbered subsections
3622 @cindex grouping data
3628 fall into two sections: text and data.
3630 You may have separate groups of
3632 data in named sections
3636 data in named sections
3642 that you want to end up near to each other in the object file, even though they
3643 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3644 use @dfn{subsections} for this purpose. Within each section, there can be
3645 numbered subsections with values from 0 to 8192. Objects assembled into the
3646 same subsection go into the object file together with other objects in the same
3647 subsection. For example, a compiler might want to store constants in the text
3648 section, but might not want to have them interspersed with the program being
3649 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3650 section of code being output, and a @samp{.text 1} before each group of
3651 constants being output.
3653 Subsections are optional. If you do not use subsections, everything
3654 goes in subsection number zero.
3657 Each subsection is zero-padded up to a multiple of four bytes.
3658 (Subsections may be padded a different amount on different flavors
3659 of @command{@value{AS}}.)
3663 On the H8/300 platform, each subsection is zero-padded to a word
3664 boundary (two bytes).
3665 The same is true on the Renesas SH.
3669 Subsections appear in your object file in numeric order, lowest numbered
3670 to highest. (All this to be compatible with other people's assemblers.)
3671 The object file contains no representation of subsections; @code{@value{LD}} and
3672 other programs that manipulate object files see no trace of them.
3673 They just see all your text subsections as a text section, and all your
3674 data subsections as a data section.
3676 To specify which subsection you want subsequent statements assembled
3677 into, use a numeric argument to specify it, in a @samp{.text
3678 @var{expression}} or a @samp{.data @var{expression}} statement.
3681 When generating COFF output, you
3686 can also use an extra subsection
3687 argument with arbitrary named sections: @samp{.section @var{name},
3692 When generating ELF output, you
3697 can also use the @code{.subsection} directive (@pxref{SubSection})
3698 to specify a subsection: @samp{.subsection @var{expression}}.
3700 @var{Expression} should be an absolute expression
3701 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3702 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3703 begins in @code{text 0}. For instance:
3705 .text 0 # The default subsection is text 0 anyway.
3706 .ascii "This lives in the first text subsection. *"
3708 .ascii "But this lives in the second text subsection."
3710 .ascii "This lives in the data section,"
3711 .ascii "in the first data subsection."
3713 .ascii "This lives in the first text section,"
3714 .ascii "immediately following the asterisk (*)."
3717 Each section has a @dfn{location counter} incremented by one for every byte
3718 assembled into that section. Because subsections are merely a convenience
3719 restricted to @command{@value{AS}} there is no concept of a subsection location
3720 counter. There is no way to directly manipulate a location counter---but the
3721 @code{.align} directive changes it, and any label definition captures its
3722 current value. The location counter of the section where statements are being
3723 assembled is said to be the @dfn{active} location counter.
3726 @section bss Section
3729 @cindex common variable storage
3730 The bss section is used for local common variable storage.
3731 You may allocate address space in the bss section, but you may
3732 not dictate data to load into it before your program executes. When
3733 your program starts running, all the contents of the bss
3734 section are zeroed bytes.
3736 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3737 @ref{Lcomm,,@code{.lcomm}}.
3739 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3740 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3743 When assembling for a target which supports multiple sections, such as ELF or
3744 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3745 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3746 section. Typically the section will only contain symbol definitions and
3747 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3754 Symbols are a central concept: the programmer uses symbols to name
3755 things, the linker uses symbols to link, and the debugger uses symbols
3759 @cindex debuggers, and symbol order
3760 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3761 the same order they were declared. This may break some debuggers.
3766 * Setting Symbols:: Giving Symbols Other Values
3767 * Symbol Names:: Symbol Names
3768 * Dot:: The Special Dot Symbol
3769 * Symbol Attributes:: Symbol Attributes
3776 A @dfn{label} is written as a symbol immediately followed by a colon
3777 @samp{:}. The symbol then represents the current value of the
3778 active location counter, and is, for example, a suitable instruction
3779 operand. You are warned if you use the same symbol to represent two
3780 different locations: the first definition overrides any other
3784 On the HPPA, the usual form for a label need not be immediately followed by a
3785 colon, but instead must start in column zero. Only one label may be defined on
3786 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3787 provides a special directive @code{.label} for defining labels more flexibly.
3790 @node Setting Symbols
3791 @section Giving Symbols Other Values
3793 @cindex assigning values to symbols
3794 @cindex symbol values, assigning
3795 A symbol can be given an arbitrary value by writing a symbol, followed
3796 by an equals sign @samp{=}, followed by an expression
3797 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3798 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3799 equals sign @samp{=}@samp{=} here represents an equivalent of the
3800 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3803 Blackfin does not support symbol assignment with @samp{=}.
3807 @section Symbol Names
3809 @cindex symbol names
3810 @cindex names, symbol
3811 @ifclear SPECIAL-SYMS
3812 Symbol names begin with a letter or with one of @samp{._}. On most
3813 machines, you can also use @code{$} in symbol names; exceptions are
3814 noted in @ref{Machine Dependencies}. That character may be followed by any
3815 string of digits, letters, dollar signs (unless otherwise noted for a
3816 particular target machine), and underscores.
3820 Symbol names begin with a letter or with one of @samp{._}. On the
3821 Renesas SH you can also use @code{$} in symbol names. That
3822 character may be followed by any string of digits, letters, dollar signs (save
3823 on the H8/300), and underscores.
3827 Case of letters is significant: @code{foo} is a different symbol name
3830 Symbol names do not start with a digit. An exception to this rule is made for
3831 Local Labels. See below.
3833 Multibyte characters are supported. To generate a symbol name containing
3834 multibyte characters enclose it within double quotes and use escape codes. cf
3835 @xref{Strings}. Generating a multibyte symbol name from a label is not
3836 currently supported.
3838 Each symbol has exactly one name. Each name in an assembly language program
3839 refers to exactly one symbol. You may use that symbol name any number of times
3842 @subheading Local Symbol Names
3844 @cindex local symbol names
3845 @cindex symbol names, local
3846 A local symbol is any symbol beginning with certain local label prefixes.
3847 By default, the local label prefix is @samp{.L} for ELF systems or
3848 @samp{L} for traditional a.out systems, but each target may have its own
3849 set of local label prefixes.
3851 On the HPPA local symbols begin with @samp{L$}.
3854 Local symbols are defined and used within the assembler, but they are
3855 normally not saved in object files. Thus, they are not visible when debugging.
3856 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3857 to retain the local symbols in the object files.
3859 @subheading Local Labels
3861 @cindex local labels
3862 @cindex temporary symbol names
3863 @cindex symbol names, temporary
3864 Local labels are different from local symbols. Local labels help compilers and
3865 programmers use names temporarily. They create symbols which are guaranteed to
3866 be unique over the entire scope of the input source code and which can be
3867 referred to by a simple notation. To define a local label, write a label of
3868 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3869 To refer to the most recent previous definition of that label write
3870 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3871 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3872 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3874 There is no restriction on how you can use these labels, and you can reuse them
3875 too. So that it is possible to repeatedly define the same local label (using
3876 the same number @samp{@b{N}}), although you can only refer to the most recently
3877 defined local label of that number (for a backwards reference) or the next
3878 definition of a specific local label for a forward reference. It is also worth
3879 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3880 implemented in a slightly more efficient manner than the others.
3891 Which is the equivalent of:
3894 label_1: branch label_3
3895 label_2: branch label_1
3896 label_3: branch label_4
3897 label_4: branch label_3
3900 Local label names are only a notational device. They are immediately
3901 transformed into more conventional symbol names before the assembler uses them.
3902 The symbol names are stored in the symbol table, appear in error messages, and
3903 are optionally emitted to the object file. The names are constructed using
3907 @item @emph{local label prefix}
3908 All local symbols begin with the system-specific local label prefix.
3909 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3910 that start with the local label prefix. These labels are
3911 used for symbols you are never intended to see. If you use the
3912 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3913 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3914 you may use them in debugging.
3917 This is the number that was used in the local label definition. So if the
3918 label is written @samp{55:} then the number is @samp{55}.
3921 This unusual character is included so you do not accidentally invent a symbol
3922 of the same name. The character has ASCII value of @samp{\002} (control-B).
3924 @item @emph{ordinal number}
3925 This is a serial number to keep the labels distinct. The first definition of
3926 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3927 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3928 the number @samp{1} and its 15th definition gets @samp{15} as well.
3931 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3932 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3934 @subheading Dollar Local Labels
3935 @cindex dollar local symbols
3937 On some targets @code{@value{AS}} also supports an even more local form of
3938 local labels called dollar labels. These labels go out of scope (i.e., they
3939 become undefined) as soon as a non-local label is defined. Thus they remain
3940 valid for only a small region of the input source code. Normal local labels,
3941 by contrast, remain in scope for the entire file, or until they are redefined
3942 by another occurrence of the same local label.
3944 Dollar labels are defined in exactly the same way as ordinary local labels,
3945 except that they have a dollar sign suffix to their numeric value, e.g.,
3948 They can also be distinguished from ordinary local labels by their transformed
3949 names which use ASCII character @samp{\001} (control-A) as the magic character
3950 to distinguish them from ordinary labels. For example, the fifth definition of
3951 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3954 @section The Special Dot Symbol
3956 @cindex dot (symbol)
3957 @cindex @code{.} (symbol)
3958 @cindex current address
3959 @cindex location counter
3960 The special symbol @samp{.} refers to the current address that
3961 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3962 .long .} defines @code{melvin} to contain its own address.
3963 Assigning a value to @code{.} is treated the same as a @code{.org}
3965 @ifclear no-space-dir
3966 Thus, the expression @samp{.=.+4} is the same as saying
3970 @node Symbol Attributes
3971 @section Symbol Attributes
3973 @cindex symbol attributes
3974 @cindex attributes, symbol
3975 Every symbol has, as well as its name, the attributes ``Value'' and
3976 ``Type''. Depending on output format, symbols can also have auxiliary
3979 The detailed definitions are in @file{a.out.h}.
3982 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3983 all these attributes, and probably won't warn you. This makes the
3984 symbol an externally defined symbol, which is generally what you
3988 * Symbol Value:: Value
3989 * Symbol Type:: Type
3991 * a.out Symbols:: Symbol Attributes: @code{a.out}
3994 * COFF Symbols:: Symbol Attributes for COFF
3997 * SOM Symbols:: Symbol Attributes for SOM
4004 @cindex value of a symbol
4005 @cindex symbol value
4006 The value of a symbol is (usually) 32 bits. For a symbol which labels a
4007 location in the text, data, bss or absolute sections the value is the
4008 number of addresses from the start of that section to the label.
4009 Naturally for text, data and bss sections the value of a symbol changes
4010 as @code{@value{LD}} changes section base addresses during linking. Absolute
4011 symbols' values do not change during linking: that is why they are
4014 The value of an undefined symbol is treated in a special way. If it is
4015 0 then the symbol is not defined in this assembler source file, and
4016 @code{@value{LD}} tries to determine its value from other files linked into the
4017 same program. You make this kind of symbol simply by mentioning a symbol
4018 name without defining it. A non-zero value represents a @code{.comm}
4019 common declaration. The value is how much common storage to reserve, in
4020 bytes (addresses). The symbol refers to the first address of the
4026 @cindex type of a symbol
4028 The type attribute of a symbol contains relocation (section)
4029 information, any flag settings indicating that a symbol is external, and
4030 (optionally), other information for linkers and debuggers. The exact
4031 format depends on the object-code output format in use.
4035 @subsection Symbol Attributes: @code{a.out}
4037 @cindex @code{a.out} symbol attributes
4038 @cindex symbol attributes, @code{a.out}
4041 * Symbol Desc:: Descriptor
4042 * Symbol Other:: Other
4046 @subsubsection Descriptor
4048 @cindex descriptor, of @code{a.out} symbol
4049 This is an arbitrary 16-bit value. You may establish a symbol's
4050 descriptor value by using a @code{.desc} statement
4051 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
4052 @command{@value{AS}}.
4055 @subsubsection Other
4057 @cindex other attribute, of @code{a.out} symbol
4058 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
4063 @subsection Symbol Attributes for COFF
4065 @cindex COFF symbol attributes
4066 @cindex symbol attributes, COFF
4068 The COFF format supports a multitude of auxiliary symbol attributes;
4069 like the primary symbol attributes, they are set between @code{.def} and
4070 @code{.endef} directives.
4072 @subsubsection Primary Attributes
4074 @cindex primary attributes, COFF symbols
4075 The symbol name is set with @code{.def}; the value and type,
4076 respectively, with @code{.val} and @code{.type}.
4078 @subsubsection Auxiliary Attributes
4080 @cindex auxiliary attributes, COFF symbols
4081 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
4082 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
4083 table information for COFF.
4088 @subsection Symbol Attributes for SOM
4090 @cindex SOM symbol attributes
4091 @cindex symbol attributes, SOM
4093 The SOM format for the HPPA supports a multitude of symbol attributes set with
4094 the @code{.EXPORT} and @code{.IMPORT} directives.
4096 The attributes are described in @cite{HP9000 Series 800 Assembly
4097 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
4098 @code{EXPORT} assembler directive documentation.
4102 @chapter Expressions
4106 @cindex numeric values
4107 An @dfn{expression} specifies an address or numeric value.
4108 Whitespace may precede and/or follow an expression.
4110 The result of an expression must be an absolute number, or else an offset into
4111 a particular section. If an expression is not absolute, and there is not
4112 enough information when @command{@value{AS}} sees the expression to know its
4113 section, a second pass over the source program might be necessary to interpret
4114 the expression---but the second pass is currently not implemented.
4115 @command{@value{AS}} aborts with an error message in this situation.
4118 * Empty Exprs:: Empty Expressions
4119 * Integer Exprs:: Integer Expressions
4123 @section Empty Expressions
4125 @cindex empty expressions
4126 @cindex expressions, empty
4127 An empty expression has no value: it is just whitespace or null.
4128 Wherever an absolute expression is required, you may omit the
4129 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4130 is compatible with other assemblers.
4133 @section Integer Expressions
4135 @cindex integer expressions
4136 @cindex expressions, integer
4137 An @dfn{integer expression} is one or more @emph{arguments} delimited
4138 by @emph{operators}.
4141 * Arguments:: Arguments
4142 * Operators:: Operators
4143 * Prefix Ops:: Prefix Operators
4144 * Infix Ops:: Infix Operators
4148 @subsection Arguments
4150 @cindex expression arguments
4151 @cindex arguments in expressions
4152 @cindex operands in expressions
4153 @cindex arithmetic operands
4154 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4155 contexts arguments are sometimes called ``arithmetic operands''. In
4156 this manual, to avoid confusing them with the ``instruction operands'' of
4157 the machine language, we use the term ``argument'' to refer to parts of
4158 expressions only, reserving the word ``operand'' to refer only to machine
4159 instruction operands.
4161 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4162 @var{section} is one of text, data, bss, absolute,
4163 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4166 Numbers are usually integers.
4168 A number can be a flonum or bignum. In this case, you are warned
4169 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4170 these 32 bits are an integer. You may write integer-manipulating
4171 instructions that act on exotic constants, compatible with other
4174 @cindex subexpressions
4175 Subexpressions are a left parenthesis @samp{(} followed by an integer
4176 expression, followed by a right parenthesis @samp{)}; or a prefix
4177 operator followed by an argument.
4180 @subsection Operators
4182 @cindex operators, in expressions
4183 @cindex arithmetic functions
4184 @cindex functions, in expressions
4185 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4186 operators are followed by an argument. Infix operators appear
4187 between their arguments. Operators may be preceded and/or followed by
4191 @subsection Prefix Operator
4193 @cindex prefix operators
4194 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4195 one argument, which must be absolute.
4197 @c the tex/end tex stuff surrounding this small table is meant to make
4198 @c it align, on the printed page, with the similar table in the next
4199 @c section (which is inside an enumerate).
4201 \global\advance\leftskip by \itemindent
4206 @dfn{Negation}. Two's complement negation.
4208 @dfn{Complementation}. Bitwise not.
4212 \global\advance\leftskip by -\itemindent
4216 @subsection Infix Operators
4218 @cindex infix operators
4219 @cindex operators, permitted arguments
4220 @dfn{Infix operators} take two arguments, one on either side. Operators
4221 have precedence, but operations with equal precedence are performed left
4222 to right. Apart from @code{+} or @option{-}, both arguments must be
4223 absolute, and the result is absolute.
4226 @cindex operator precedence
4227 @cindex precedence of operators
4234 @dfn{Multiplication}.
4237 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4243 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4246 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4250 Intermediate precedence
4255 @dfn{Bitwise Inclusive Or}.
4261 @dfn{Bitwise Exclusive Or}.
4264 @dfn{Bitwise Or Not}.
4271 @cindex addition, permitted arguments
4272 @cindex plus, permitted arguments
4273 @cindex arguments for addition
4275 @dfn{Addition}. If either argument is absolute, the result has the section of
4276 the other argument. You may not add together arguments from different
4279 @cindex subtraction, permitted arguments
4280 @cindex minus, permitted arguments
4281 @cindex arguments for subtraction
4283 @dfn{Subtraction}. If the right argument is absolute, the
4284 result has the section of the left argument.
4285 If both arguments are in the same section, the result is absolute.
4286 You may not subtract arguments from different sections.
4287 @c FIXME is there still something useful to say about undefined - undefined ?
4289 @cindex comparison expressions
4290 @cindex expressions, comparison
4295 @dfn{Is Not Equal To}
4299 @dfn{Is Greater Than}
4301 @dfn{Is Greater Than Or Equal To}
4303 @dfn{Is Less Than Or Equal To}
4305 The comparison operators can be used as infix operators. A true results has a
4306 value of -1 whereas a false result has a value of 0. Note, these operators
4307 perform signed comparisons.
4310 @item Lowest Precedence
4319 These two logical operations can be used to combine the results of sub
4320 expressions. Note, unlike the comparison operators a true result returns a
4321 value of 1 but a false results does still return 0. Also note that the logical
4322 or operator has a slightly lower precedence than logical and.
4327 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4328 address; you can only have a defined section in one of the two arguments.
4331 @chapter Assembler Directives
4333 @cindex directives, machine independent
4334 @cindex pseudo-ops, machine independent
4335 @cindex machine independent directives
4336 All assembler directives have names that begin with a period (@samp{.}).
4337 The names are case insensitive for most targets, and usually written
4340 This chapter discusses directives that are available regardless of the
4341 target machine configuration for the @sc{gnu} assembler.
4343 Some machine configurations provide additional directives.
4344 @xref{Machine Dependencies}.
4347 @ifset machine-directives
4348 @xref{Machine Dependencies}, for additional directives.
4353 * Abort:: @code{.abort}
4355 * ABORT (COFF):: @code{.ABORT}
4358 * Align:: @code{.align [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4359 * Altmacro:: @code{.altmacro}
4360 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4361 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4362 * Balign:: @code{.balign [@var{abs-expr}[, @var{abs-expr}]]}
4363 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4364 * Byte:: @code{.byte @var{expressions}}
4365 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4366 * Comm:: @code{.comm @var{symbol} , @var{length} }
4367 * Data:: @code{.data @var{subsection}}
4368 * Dc:: @code{.dc[@var{size}] @var{expressions}}
4369 * Dcb:: @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4370 * Ds:: @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4372 * Def:: @code{.def @var{name}}
4375 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4381 * Double:: @code{.double @var{flonums}}
4382 * Eject:: @code{.eject}
4383 * Else:: @code{.else}
4384 * Elseif:: @code{.elseif}
4387 * Endef:: @code{.endef}
4390 * Endfunc:: @code{.endfunc}
4391 * Endif:: @code{.endif}
4392 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4393 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4394 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4396 * Error:: @code{.error @var{string}}
4397 * Exitm:: @code{.exitm}
4398 * Extern:: @code{.extern}
4399 * Fail:: @code{.fail}
4400 * File:: @code{.file}
4401 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4402 * Float:: @code{.float @var{flonums}}
4403 * Func:: @code{.func}
4404 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4406 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4407 * Hidden:: @code{.hidden @var{names}}
4410 * hword:: @code{.hword @var{expressions}}
4411 * Ident:: @code{.ident}
4412 * If:: @code{.if @var{absolute expression}}
4413 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4414 * Include:: @code{.include "@var{file}"}
4415 * Int:: @code{.int @var{expressions}}
4417 * Internal:: @code{.internal @var{names}}
4420 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4421 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4422 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4423 * Lflags:: @code{.lflags}
4424 @ifclear no-line-dir
4425 * Line:: @code{.line @var{line-number}}
4428 * Linkonce:: @code{.linkonce [@var{type}]}
4429 * List:: @code{.list}
4430 * Ln:: @code{.ln @var{line-number}}
4431 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4432 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4434 * Local:: @code{.local @var{names}}
4437 * Long:: @code{.long @var{expressions}}
4439 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4442 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4443 * MRI:: @code{.mri @var{val}}
4444 * Noaltmacro:: @code{.noaltmacro}
4445 * Nolist:: @code{.nolist}
4446 * Nops:: @code{.nops @var{size}[, @var{control}]}
4447 * Octa:: @code{.octa @var{bignums}}
4448 * Offset:: @code{.offset @var{loc}}
4449 * Org:: @code{.org @var{new-lc}, @var{fill}}
4450 * P2align:: @code{.p2align [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4452 * PopSection:: @code{.popsection}
4453 * Previous:: @code{.previous}
4456 * Print:: @code{.print @var{string}}
4458 * Protected:: @code{.protected @var{names}}
4461 * Psize:: @code{.psize @var{lines}, @var{columns}}
4462 * Purgem:: @code{.purgem @var{name}}
4464 * PushSection:: @code{.pushsection @var{name}}
4467 * Quad:: @code{.quad @var{bignums}}
4468 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4469 * Rept:: @code{.rept @var{count}}
4470 * Sbttl:: @code{.sbttl "@var{subheading}"}
4472 * Scl:: @code{.scl @var{class}}
4475 * Section:: @code{.section @var{name}[, @var{flags}]}
4478 * Set:: @code{.set @var{symbol}, @var{expression}}
4479 * Short:: @code{.short @var{expressions}}
4480 * Single:: @code{.single @var{flonums}}
4482 * Size:: @code{.size [@var{name} , @var{expression}]}
4484 @ifclear no-space-dir
4485 * Skip:: @code{.skip @var{size} [,@var{fill}]}
4488 * Sleb128:: @code{.sleb128 @var{expressions}}
4489 @ifclear no-space-dir
4490 * Space:: @code{.space @var{size} [,@var{fill}]}
4493 * Stab:: @code{.stabd, .stabn, .stabs}
4496 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4497 * Struct:: @code{.struct @var{expression}}
4499 * SubSection:: @code{.subsection}
4500 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4504 * Tag:: @code{.tag @var{structname}}
4507 * Text:: @code{.text @var{subsection}}
4508 * Title:: @code{.title "@var{heading}"}
4510 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4513 * Uleb128:: @code{.uleb128 @var{expressions}}
4515 * Val:: @code{.val @var{addr}}
4519 * Version:: @code{.version "@var{string}"}
4520 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4521 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4524 * Warning:: @code{.warning @var{string}}
4525 * Weak:: @code{.weak @var{names}}
4526 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4527 * Word:: @code{.word @var{expressions}}
4528 @ifclear no-space-dir
4529 * Zero:: @code{.zero @var{size}}
4532 * 2byte:: @code{.2byte @var{expressions}}
4533 * 4byte:: @code{.4byte @var{expressions}}
4534 * 8byte:: @code{.8byte @var{bignums}}
4536 * Deprecated:: Deprecated Directives
4540 @section @code{.abort}
4542 @cindex @code{abort} directive
4543 @cindex stopping the assembly
4544 This directive stops the assembly immediately. It is for
4545 compatibility with other assemblers. The original idea was that the
4546 assembly language source would be piped into the assembler. If the sender
4547 of the source quit, it could use this directive tells @command{@value{AS}} to
4548 quit also. One day @code{.abort} will not be supported.
4552 @section @code{.ABORT} (COFF)
4554 @cindex @code{ABORT} directive
4555 When producing COFF output, @command{@value{AS}} accepts this directive as a
4556 synonym for @samp{.abort}.
4561 @section @code{.align [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4563 @cindex padding the location counter
4564 @cindex @code{align} directive
4565 Pad the location counter (in the current subsection) to a particular storage
4566 boundary. The first expression (which must be absolute) is the alignment
4567 required, as described below. If this expression is omitted then a default
4568 value of 0 is used, effectively disabling alignment requirements.
4570 The second expression (also absolute) gives the fill value to be stored in the
4571 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4572 padding bytes are normally zero. However, on most systems, if the section is
4573 marked as containing code and the fill value is omitted, the space is filled
4574 with no-op instructions.
4576 The third expression is also absolute, and is also optional. If it is present,
4577 it is the maximum number of bytes that should be skipped by this alignment
4578 directive. If doing the alignment would require skipping more bytes than the
4579 specified maximum, then the alignment is not done at all. You can omit the
4580 fill value (the second argument) entirely by simply using two commas after the
4581 required alignment; this can be useful if you want the alignment to be filled
4582 with no-op instructions when appropriate.
4584 The way the required alignment is specified varies from system to system.
4585 For the arc, hppa, i386 using ELF, iq2000, m68k, or1k,
4586 s390, sparc, tic4x and xtensa, the first expression is the
4587 alignment request in bytes. For example @samp{.align 8} advances
4588 the location counter until it is a multiple of 8. If the location counter
4589 is already a multiple of 8, no change is needed. For the tic54x, the
4590 first expression is the alignment request in words.
4592 For other systems, including ppc, i386 using a.out format, arm and
4593 strongarm, it is the
4594 number of low-order zero bits the location counter must have after
4595 advancement. For example @samp{.align 3} advances the location
4596 counter until it is a multiple of 8. If the location counter is already a
4597 multiple of 8, no change is needed.
4599 This inconsistency is due to the different behaviors of the various
4600 native assemblers for these systems which GAS must emulate.
4601 GAS also provides @code{.balign} and @code{.p2align} directives,
4602 described later, which have a consistent behavior across all
4603 architectures (but are specific to GAS).
4606 @section @code{.altmacro}
4607 Enable alternate macro mode, enabling:
4610 @item LOCAL @var{name} [ , @dots{} ]
4611 One additional directive, @code{LOCAL}, is available. It is used to
4612 generate a string replacement for each of the @var{name} arguments, and
4613 replace any instances of @var{name} in each macro expansion. The
4614 replacement string is unique in the assembly, and different for each
4615 separate macro expansion. @code{LOCAL} allows you to write macros that
4616 define symbols, without fear of conflict between separate macro expansions.
4618 @item String delimiters
4619 You can write strings delimited in these other ways besides
4620 @code{"@var{string}"}:
4623 @item '@var{string}'
4624 You can delimit strings with single-quote characters.
4626 @item <@var{string}>
4627 You can delimit strings with matching angle brackets.
4630 @item single-character string escape
4631 To include any single character literally in a string (even if the
4632 character would otherwise have some special meaning), you can prefix the
4633 character with @samp{!} (an exclamation mark). For example, you can
4634 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4636 @item Expression results as strings
4637 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4638 and use the result as a string.
4642 @section @code{.ascii "@var{string}"}@dots{}
4644 @cindex @code{ascii} directive
4645 @cindex string literals
4646 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4647 separated by commas. It assembles each string (with no automatic
4648 trailing zero byte) into consecutive addresses.
4651 @section @code{.asciz "@var{string}"}@dots{}
4653 @cindex @code{asciz} directive
4654 @cindex zero-terminated strings
4655 @cindex null-terminated strings
4656 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4657 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4660 @section @code{.balign[wl] [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4662 @cindex padding the location counter given number of bytes
4663 @cindex @code{balign} directive
4664 Pad the location counter (in the current subsection) to a particular
4665 storage boundary. The first expression (which must be absolute) is the
4666 alignment request in bytes. For example @samp{.balign 8} advances
4667 the location counter until it is a multiple of 8. If the location counter
4668 is already a multiple of 8, no change is needed. If the expression is omitted
4669 then a default value of 0 is used, effectively disabling alignment requirements.
4671 The second expression (also absolute) gives the fill value to be stored in the
4672 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4673 padding bytes are normally zero. However, on most systems, if the section is
4674 marked as containing code and the fill value is omitted, the space is filled
4675 with no-op instructions.
4677 The third expression is also absolute, and is also optional. If it is present,
4678 it is the maximum number of bytes that should be skipped by this alignment
4679 directive. If doing the alignment would require skipping more bytes than the
4680 specified maximum, then the alignment is not done at all. You can omit the
4681 fill value (the second argument) entirely by simply using two commas after the
4682 required alignment; this can be useful if you want the alignment to be filled
4683 with no-op instructions when appropriate.
4685 @cindex @code{balignw} directive
4686 @cindex @code{balignl} directive
4687 The @code{.balignw} and @code{.balignl} directives are variants of the
4688 @code{.balign} directive. The @code{.balignw} directive treats the fill
4689 pattern as a two byte word value. The @code{.balignl} directives treats the
4690 fill pattern as a four byte longword value. For example, @code{.balignw
4691 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4692 filled in with the value 0x368d (the exact placement of the bytes depends upon
4693 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4696 @node Bundle directives
4697 @section Bundle directives
4698 @subsection @code{.bundle_align_mode @var{abs-expr}}
4699 @cindex @code{bundle_align_mode} directive
4701 @cindex instruction bundle
4702 @cindex aligned instruction bundle
4703 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4704 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4705 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4706 disabled (which is the default state). If the argument it not zero, it
4707 gives the size of an instruction bundle as a power of two (as for the
4708 @code{.p2align} directive, @pxref{P2align}).
4710 For some targets, it's an ABI requirement that no instruction may span a
4711 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4712 instructions that starts on an aligned boundary. For example, if
4713 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4714 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4715 effect, no single instruction may span a boundary between bundles. If an
4716 instruction would start too close to the end of a bundle for the length of
4717 that particular instruction to fit within the bundle, then the space at the
4718 end of that bundle is filled with no-op instructions so the instruction
4719 starts in the next bundle. As a corollary, it's an error if any single
4720 instruction's encoding is longer than the bundle size.
4722 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4723 @cindex @code{bundle_lock} directive
4724 @cindex @code{bundle_unlock} directive
4725 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4726 allow explicit control over instruction bundle padding. These directives
4727 are only valid when @code{.bundle_align_mode} has been used to enable
4728 aligned instruction bundle mode. It's an error if they appear when
4729 @code{.bundle_align_mode} has not been used at all, or when the last
4730 directive was @w{@code{.bundle_align_mode 0}}.
4732 @cindex bundle-locked
4733 For some targets, it's an ABI requirement that certain instructions may
4734 appear only as part of specified permissible sequences of multiple
4735 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4736 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4737 instruction sequence. For purposes of aligned instruction bundle mode, a
4738 sequence starting with @code{.bundle_lock} and ending with
4739 @code{.bundle_unlock} is treated as a single instruction. That is, the
4740 entire sequence must fit into a single bundle and may not span a bundle
4741 boundary. If necessary, no-op instructions will be inserted before the
4742 first instruction of the sequence so that the whole sequence starts on an
4743 aligned bundle boundary. It's an error if the sequence is longer than the
4746 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4747 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4748 nested. That is, a second @code{.bundle_lock} directive before the next
4749 @code{.bundle_unlock} directive has no effect except that it must be
4750 matched by another closing @code{.bundle_unlock} so that there is the
4751 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4754 @section @code{.byte @var{expressions}}
4756 @cindex @code{byte} directive
4757 @cindex integers, one byte
4758 @code{.byte} expects zero or more expressions, separated by commas.
4759 Each expression is assembled into the next byte.
4761 @node CFI directives
4762 @section CFI directives
4763 @subsection @code{.cfi_sections @var{section_list}}
4764 @cindex @code{cfi_sections} directive
4765 @code{.cfi_sections} may be used to specify whether CFI directives
4766 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4767 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4768 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4769 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4770 directive is not used is @code{.cfi_sections .eh_frame}.
4772 On targets that support compact unwinding tables these can be generated
4773 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4775 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4776 which is used by the @value{TIC6X} target.
4778 The @code{.cfi_sections} directive can be repeated, with the same or different
4779 arguments, provided that CFI generation has not yet started. Once CFI
4780 generation has started however the section list is fixed and any attempts to
4781 redefine it will result in an error.
4783 @subsection @code{.cfi_startproc [simple]}
4784 @cindex @code{cfi_startproc} directive
4785 @code{.cfi_startproc} is used at the beginning of each function that
4786 should have an entry in @code{.eh_frame}. It initializes some internal
4787 data structures. Don't forget to close the function by
4788 @code{.cfi_endproc}.
4790 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4791 it also emits some architecture dependent initial CFI instructions.
4793 @subsection @code{.cfi_endproc}
4794 @cindex @code{cfi_endproc} directive
4795 @code{.cfi_endproc} is used at the end of a function where it closes its
4796 unwind entry previously opened by
4797 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4799 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4800 @cindex @code{cfi_personality} directive
4801 @code{.cfi_personality} defines personality routine and its encoding.
4802 @var{encoding} must be a constant determining how the personality
4803 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4804 argument is not present, otherwise second argument should be
4805 a constant or a symbol name. When using indirect encodings,
4806 the symbol provided should be the location where personality
4807 can be loaded from, not the personality routine itself.
4808 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4809 no personality routine.
4811 @subsection @code{.cfi_personality_id @var{id}}
4812 @cindex @code{cfi_personality_id} directive
4813 @code{cfi_personality_id} defines a personality routine by its index as
4814 defined in a compact unwinding format.
4815 Only valid when generating compact EH frames (i.e.
4816 with @code{.cfi_sections eh_frame_entry}.
4818 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4819 @cindex @code{cfi_fde_data} directive
4820 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4821 used for the current function. These are emitted inline in the
4822 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4823 in the @code{.gnu.extab} section otherwise.
4824 Only valid when generating compact EH frames (i.e.
4825 with @code{.cfi_sections eh_frame_entry}.
4827 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4828 @code{.cfi_lsda} defines LSDA and its encoding.
4829 @var{encoding} must be a constant determining how the LSDA
4830 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4831 argument is not present, otherwise the second argument should be a constant
4832 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4833 meaning that no LSDA is present.
4835 @subsection @code{.cfi_inline_lsda} [@var{align}]
4836 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4837 switches to the corresponding @code{.gnu.extab} section.
4838 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4839 Only valid when generating compact EH frames (i.e.
4840 with @code{.cfi_sections eh_frame_entry}.
4842 The table header and unwinding opcodes will be generated at this point,
4843 so that they are immediately followed by the LSDA data. The symbol
4844 referenced by the @code{.cfi_lsda} directive should still be defined
4845 in case a fallback FDE based encoding is used. The LSDA data is terminated
4846 by a section directive.
4848 The optional @var{align} argument specifies the alignment required.
4849 The alignment is specified as a power of two, as with the
4850 @code{.p2align} directive.
4852 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4853 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4854 address from @var{register} and add @var{offset} to it}.
4856 @subsection @code{.cfi_def_cfa_register @var{register}}
4857 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4858 now on @var{register} will be used instead of the old one. Offset
4861 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4862 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4863 remains the same, but @var{offset} is new. Note that it is the
4864 absolute offset that will be added to a defined register to compute
4867 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4868 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4869 value that is added/subtracted from the previous offset.
4871 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4872 Previous value of @var{register} is saved at offset @var{offset} from
4875 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4876 Previous value of @var{register} is CFA + @var{offset}.
4878 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4879 Previous value of @var{register} is saved at offset @var{offset} from
4880 the current CFA register. This is transformed to @code{.cfi_offset}
4881 using the known displacement of the CFA register from the CFA.
4882 This is often easier to use, because the number will match the
4883 code it's annotating.
4885 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4886 Previous value of @var{register1} is saved in register @var{register2}.
4888 @subsection @code{.cfi_restore @var{register}}
4889 @code{.cfi_restore} says that the rule for @var{register} is now the
4890 same as it was at the beginning of the function, after all initial
4891 instruction added by @code{.cfi_startproc} were executed.
4893 @subsection @code{.cfi_undefined @var{register}}
4894 From now on the previous value of @var{register} can't be restored anymore.
4896 @subsection @code{.cfi_same_value @var{register}}
4897 Current value of @var{register} is the same like in the previous frame,
4898 i.e. no restoration needed.
4900 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4901 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4902 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4903 places them in the current row. This is useful for situations where you have
4904 multiple @code{.cfi_*} directives that need to be undone due to the control
4905 flow of the program. For example, we could have something like this (assuming
4906 the CFA is the value of @code{rbp}):
4916 .cfi_def_cfa %rsp, 8
4919 /* Do something else */
4922 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4923 to the instructions before @code{label}. This means we'd have to add multiple
4924 @code{.cfi} directives after @code{label} to recreate the original save
4925 locations of the registers, as well as setting the CFA back to the value of
4926 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4938 .cfi_def_cfa %rsp, 8
4942 /* Do something else */
4945 That way, the rules for the instructions after @code{label} will be the same
4946 as before the first @code{.cfi_restore} without having to use multiple
4947 @code{.cfi} directives.
4949 @subsection @code{.cfi_return_column @var{register}}
4950 Change return column @var{register}, i.e. the return address is either
4951 directly in @var{register} or can be accessed by rules for @var{register}.
4953 @subsection @code{.cfi_signal_frame}
4954 Mark current function as signal trampoline.
4956 @subsection @code{.cfi_window_save}
4957 SPARC register window has been saved.
4959 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4960 Allows the user to add arbitrary bytes to the unwind info. One
4961 might use this to add OS-specific CFI opcodes, or generic CFI
4962 opcodes that GAS does not yet support.
4964 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4965 The current value of @var{register} is @var{label}. The value of @var{label}
4966 will be encoded in the output file according to @var{encoding}; see the
4967 description of @code{.cfi_personality} for details on this encoding.
4969 The usefulness of equating a register to a fixed label is probably
4970 limited to the return address register. Here, it can be useful to
4971 mark a code segment that has only one return address which is reached
4972 by a direct branch and no copy of the return address exists in memory
4973 or another register.
4976 @section @code{.comm @var{symbol} , @var{length} }
4978 @cindex @code{comm} directive
4979 @cindex symbol, common
4980 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4981 common symbol in one object file may be merged with a defined or common symbol
4982 of the same name in another object file. If @code{@value{LD}} does not see a
4983 definition for the symbol--just one or more common symbols--then it will
4984 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4985 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4986 the same name, and they do not all have the same size, it will allocate space
4987 using the largest size.
4990 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4991 an optional third argument. This is the desired alignment of the symbol,
4992 specified for ELF as a byte boundary (for example, an alignment of 16 means
4993 that the least significant 4 bits of the address should be zero), and for PE
4994 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4995 boundary). The alignment must be an absolute expression, and it must be a
4996 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4997 common symbol, it will use the alignment when placing the symbol. If no
4998 alignment is specified, @command{@value{AS}} will set the alignment to the
4999 largest power of two less than or equal to the size of the symbol, up to a
5000 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
5001 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
5002 @samp{--section-alignment} option; image file sections in PE are aligned to
5003 multiples of 4096, which is far too large an alignment for ordinary variables.
5004 It is rather the default alignment for (non-debug) sections within object
5005 (@samp{*.o}) files, which are less strictly aligned.}.
5009 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
5010 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
5014 @section @code{.data @var{subsection}}
5015 @cindex @code{data} directive
5017 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
5018 end of the data subsection numbered @var{subsection} (which is an
5019 absolute expression). If @var{subsection} is omitted, it defaults
5023 @section @code{.dc[@var{size}] @var{expressions}}
5024 @cindex @code{dc} directive
5026 The @code{.dc} directive expects zero or more @var{expressions} separated by
5027 commas. These expressions are evaluated and their values inserted into the
5028 current section. The size of the emitted value depends upon the suffix to the
5029 @code{.dc} directive:
5033 Emits N-bit values, where N is the size of an address on the target system.
5037 Emits double precision floating-point values.
5039 Emits 32-bit values.
5041 Emits single precision floating-point values.
5043 Emits 16-bit values.
5044 Note - this is true even on targets where the @code{.word} directive would emit
5047 Emits long double precision floating-point values.
5050 If no suffix is used then @samp{.w} is assumed.
5052 The byte ordering is target dependent, as is the size and format of floating
5056 @section @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
5057 @cindex @code{dcb} directive
5058 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5059 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5060 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5061 @var{size} suffix, if present, must be one of:
5065 Emits single byte values.
5067 Emits double-precision floating point values.
5069 Emits 4-byte values.
5071 Emits single-precision floating point values.
5073 Emits 2-byte values.
5075 Emits long double-precision floating point values.
5078 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5080 The byte ordering is target dependent, as is the size and format of floating
5084 @section @code{.ds[@var{size}] @var{number} [,@var{fill}]}
5085 @cindex @code{ds} directive
5086 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5087 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5088 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5089 @var{size} suffix, if present, must be one of:
5093 Emits single byte values.
5095 Emits 8-byte values.
5097 Emits 4-byte values.
5099 Emits 12-byte values.
5101 Emits 4-byte values.
5103 Emits 2-byte values.
5105 Emits 12-byte values.
5108 Note - unlike the @code{.dcb} directive the @samp{.d}, @samp{.s} and @samp{.x}
5109 suffixes do not indicate that floating-point values are to be inserted.
5111 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5113 The byte ordering is target dependent.
5118 @section @code{.def @var{name}}
5120 @cindex @code{def} directive
5121 @cindex COFF symbols, debugging
5122 @cindex debugging COFF symbols
5123 Begin defining debugging information for a symbol @var{name}; the
5124 definition extends until the @code{.endef} directive is encountered.
5129 @section @code{.desc @var{symbol}, @var{abs-expression}}
5131 @cindex @code{desc} directive
5132 @cindex COFF symbol descriptor
5133 @cindex symbol descriptor, COFF
5134 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5135 to the low 16 bits of an absolute expression.
5138 The @samp{.desc} directive is not available when @command{@value{AS}} is
5139 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5140 object format. For the sake of compatibility, @command{@value{AS}} accepts
5141 it, but produces no output, when configured for COFF.
5147 @section @code{.dim}
5149 @cindex @code{dim} directive
5150 @cindex COFF auxiliary symbol information
5151 @cindex auxiliary symbol information, COFF
5152 This directive is generated by compilers to include auxiliary debugging
5153 information in the symbol table. It is only permitted inside
5154 @code{.def}/@code{.endef} pairs.
5158 @section @code{.double @var{flonums}}
5160 @cindex @code{double} directive
5161 @cindex floating point numbers (double)
5162 @code{.double} expects zero or more flonums, separated by commas. It
5163 assembles floating point numbers.
5165 The exact kind of floating point numbers emitted depends on how
5166 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5170 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5171 in @sc{ieee} format.
5176 @section @code{.eject}
5178 @cindex @code{eject} directive
5179 @cindex new page, in listings
5180 @cindex page, in listings
5181 @cindex listing control: new page
5182 Force a page break at this point, when generating assembly listings.
5185 @section @code{.else}
5187 @cindex @code{else} directive
5188 @code{.else} is part of the @command{@value{AS}} support for conditional
5189 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5190 of code to be assembled if the condition for the preceding @code{.if}
5194 @section @code{.elseif}
5196 @cindex @code{elseif} directive
5197 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5198 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5199 @code{.if} block that would otherwise fill the entire @code{.else} section.
5202 @section @code{.end}
5204 @cindex @code{end} directive
5205 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5206 process anything in the file past the @code{.end} directive.
5210 @section @code{.endef}
5212 @cindex @code{endef} directive
5213 This directive flags the end of a symbol definition begun with
5218 @section @code{.endfunc}
5219 @cindex @code{endfunc} directive
5220 @code{.endfunc} marks the end of a function specified with @code{.func}.
5223 @section @code{.endif}
5225 @cindex @code{endif} directive
5226 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5227 it marks the end of a block of code that is only assembled
5228 conditionally. @xref{If,,@code{.if}}.
5231 @section @code{.equ @var{symbol}, @var{expression}}
5233 @cindex @code{equ} directive
5234 @cindex assigning values to symbols
5235 @cindex symbols, assigning values to
5236 This directive sets the value of @var{symbol} to @var{expression}.
5237 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5240 The syntax for @code{equ} on the HPPA is
5241 @samp{@var{symbol} .equ @var{expression}}.
5245 The syntax for @code{equ} on the Z80 is
5246 @samp{@var{symbol} equ @var{expression}}.
5247 On the Z80 it is an error if @var{symbol} is already defined,
5248 but the symbol is not protected from later redefinition.
5249 Compare @ref{Equiv}.
5253 @section @code{.equiv @var{symbol}, @var{expression}}
5254 @cindex @code{equiv} directive
5255 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5256 the assembler will signal an error if @var{symbol} is already defined. Note a
5257 symbol which has been referenced but not actually defined is considered to be
5260 Except for the contents of the error message, this is roughly equivalent to
5267 plus it protects the symbol from later redefinition.
5270 @section @code{.eqv @var{symbol}, @var{expression}}
5271 @cindex @code{eqv} directive
5272 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5273 evaluate the expression or any part of it immediately. Instead each time
5274 the resulting symbol is used in an expression, a snapshot of its current
5278 @section @code{.err}
5279 @cindex @code{err} directive
5280 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5281 message and, unless the @option{-Z} option was used, it will not generate an
5282 object file. This can be used to signal an error in conditionally compiled code.
5285 @section @code{.error "@var{string}"}
5286 @cindex error directive
5288 Similarly to @code{.err}, this directive emits an error, but you can specify a
5289 string that will be emitted as the error message. If you don't specify the
5290 message, it defaults to @code{".error directive invoked in source file"}.
5291 @xref{Errors, ,Error and Warning Messages}.
5294 .error "This code has not been assembled and tested."
5298 @section @code{.exitm}
5299 Exit early from the current macro definition. @xref{Macro}.
5302 @section @code{.extern}
5304 @cindex @code{extern} directive
5305 @code{.extern} is accepted in the source program---for compatibility
5306 with other assemblers---but it is ignored. @command{@value{AS}} treats
5307 all undefined symbols as external.
5310 @section @code{.fail @var{expression}}
5312 @cindex @code{fail} directive
5313 Generates an error or a warning. If the value of the @var{expression} is 500
5314 or more, @command{@value{AS}} will print a warning message. If the value is less
5315 than 500, @command{@value{AS}} will print an error message. The message will
5316 include the value of @var{expression}. This can occasionally be useful inside
5317 complex nested macros or conditional assembly.
5320 @section @code{.file}
5321 @cindex @code{file} directive
5323 @ifclear no-file-dir
5324 There are two different versions of the @code{.file} directive. Targets
5325 that support DWARF2 line number information use the DWARF2 version of
5326 @code{.file}. Other targets use the default version.
5328 @subheading Default Version
5330 @cindex logical file name
5331 @cindex file name, logical
5332 This version of the @code{.file} directive tells @command{@value{AS}} that we
5333 are about to start a new logical file. The syntax is:
5339 @var{string} is the new file name. In general, the filename is
5340 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5341 to specify an empty file name, you must give the quotes--@code{""}. This
5342 statement may go away in future: it is only recognized to be compatible with
5343 old @command{@value{AS}} programs.
5345 @subheading DWARF2 Version
5348 When emitting DWARF2 line number information, @code{.file} assigns filenames
5349 to the @code{.debug_line} file name table. The syntax is:
5352 .file @var{fileno} @var{filename}
5355 The @var{fileno} operand should be a unique positive integer to use as the
5356 index of the entry in the table. The @var{filename} operand is a C string
5359 The detail of filename indices is exposed to the user because the filename
5360 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5361 information, and thus the user must know the exact indices that table
5365 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5367 @cindex @code{fill} directive
5368 @cindex writing patterns in memory
5369 @cindex patterns, writing in memory
5370 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5371 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5372 may be zero or more. @var{Size} may be zero or more, but if it is
5373 more than 8, then it is deemed to have the value 8, compatible with
5374 other people's assemblers. The contents of each @var{repeat} bytes
5375 is taken from an 8-byte number. The highest order 4 bytes are
5376 zero. The lowest order 4 bytes are @var{value} rendered in the
5377 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5378 Each @var{size} bytes in a repetition is taken from the lowest order
5379 @var{size} bytes of this number. Again, this bizarre behavior is
5380 compatible with other people's assemblers.
5382 @var{size} and @var{value} are optional.
5383 If the second comma and @var{value} are absent, @var{value} is
5384 assumed zero. If the first comma and following tokens are absent,
5385 @var{size} is assumed to be 1.
5388 @section @code{.float @var{flonums}}
5390 @cindex floating point numbers (single)
5391 @cindex @code{float} directive
5392 This directive assembles zero or more flonums, separated by commas. It
5393 has the same effect as @code{.single}.
5395 The exact kind of floating point numbers emitted depends on how
5396 @command{@value{AS}} is configured.
5397 @xref{Machine Dependencies}.
5401 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5402 in @sc{ieee} format.
5407 @section @code{.func @var{name}[,@var{label}]}
5408 @cindex @code{func} directive
5409 @code{.func} emits debugging information to denote function @var{name}, and
5410 is ignored unless the file is assembled with debugging enabled.
5411 Only @samp{--gstabs[+]} is currently supported.
5412 @var{label} is the entry point of the function and if omitted @var{name}
5413 prepended with the @samp{leading char} is used.
5414 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5415 All functions are currently defined to have @code{void} return type.
5416 The function must be terminated with @code{.endfunc}.
5419 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5421 @cindex @code{global} directive
5422 @cindex symbol, making visible to linker
5423 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5424 @var{symbol} in your partial program, its value is made available to
5425 other partial programs that are linked with it. Otherwise,
5426 @var{symbol} takes its attributes from a symbol of the same name
5427 from another file linked into the same program.
5429 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5430 compatibility with other assemblers.
5433 On the HPPA, @code{.global} is not always enough to make it accessible to other
5434 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5435 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5440 @section @code{.gnu_attribute @var{tag},@var{value}}
5441 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5444 @section @code{.hidden @var{names}}
5446 @cindex @code{hidden} directive
5448 This is one of the ELF visibility directives. The other two are
5449 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5450 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5452 This directive overrides the named symbols default visibility (which is set by
5453 their binding: local, global or weak). The directive sets the visibility to
5454 @code{hidden} which means that the symbols are not visible to other components.
5455 Such symbols are always considered to be @code{protected} as well.
5459 @section @code{.hword @var{expressions}}
5461 @cindex @code{hword} directive
5462 @cindex integers, 16-bit
5463 @cindex numbers, 16-bit
5464 @cindex sixteen bit integers
5465 This expects zero or more @var{expressions}, and emits
5466 a 16 bit number for each.
5469 This directive is a synonym for @samp{.short}; depending on the target
5470 architecture, it may also be a synonym for @samp{.word}.
5474 This directive is a synonym for @samp{.short}.
5477 This directive is a synonym for both @samp{.short} and @samp{.word}.
5482 @section @code{.ident}
5484 @cindex @code{ident} directive
5486 This directive is used by some assemblers to place tags in object files. The
5487 behavior of this directive varies depending on the target. When using the
5488 a.out object file format, @command{@value{AS}} simply accepts the directive for
5489 source-file compatibility with existing assemblers, but does not emit anything
5490 for it. When using COFF, comments are emitted to the @code{.comment} or
5491 @code{.rdata} section, depending on the target. When using ELF, comments are
5492 emitted to the @code{.comment} section.
5495 @section @code{.if @var{absolute expression}}
5497 @cindex conditional assembly
5498 @cindex @code{if} directive
5499 @code{.if} marks the beginning of a section of code which is only
5500 considered part of the source program being assembled if the argument
5501 (which must be an @var{absolute expression}) is non-zero. The end of
5502 the conditional section of code must be marked by @code{.endif}
5503 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5504 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5505 If you have several conditions to check, @code{.elseif} may be used to avoid
5506 nesting blocks if/else within each subsequent @code{.else} block.
5508 The following variants of @code{.if} are also supported:
5510 @cindex @code{ifdef} directive
5511 @item .ifdef @var{symbol}
5512 Assembles the following section of code if the specified @var{symbol}
5513 has been defined. Note a symbol which has been referenced but not yet defined
5514 is considered to be undefined.
5516 @cindex @code{ifb} directive
5517 @item .ifb @var{text}
5518 Assembles the following section of code if the operand is blank (empty).
5520 @cindex @code{ifc} directive
5521 @item .ifc @var{string1},@var{string2}
5522 Assembles the following section of code if the two strings are the same. The
5523 strings may be optionally quoted with single quotes. If they are not quoted,
5524 the first string stops at the first comma, and the second string stops at the
5525 end of the line. Strings which contain whitespace should be quoted. The
5526 string comparison is case sensitive.
5528 @cindex @code{ifeq} directive
5529 @item .ifeq @var{absolute expression}
5530 Assembles the following section of code if the argument is zero.
5532 @cindex @code{ifeqs} directive
5533 @item .ifeqs @var{string1},@var{string2}
5534 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5536 @cindex @code{ifge} directive
5537 @item .ifge @var{absolute expression}
5538 Assembles the following section of code if the argument is greater than or
5541 @cindex @code{ifgt} directive
5542 @item .ifgt @var{absolute expression}
5543 Assembles the following section of code if the argument is greater than zero.
5545 @cindex @code{ifle} directive
5546 @item .ifle @var{absolute expression}
5547 Assembles the following section of code if the argument is less than or equal
5550 @cindex @code{iflt} directive
5551 @item .iflt @var{absolute expression}
5552 Assembles the following section of code if the argument is less than zero.
5554 @cindex @code{ifnb} directive
5555 @item .ifnb @var{text}
5556 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5557 following section of code if the operand is non-blank (non-empty).
5559 @cindex @code{ifnc} directive
5560 @item .ifnc @var{string1},@var{string2}.
5561 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5562 following section of code if the two strings are not the same.
5564 @cindex @code{ifndef} directive
5565 @cindex @code{ifnotdef} directive
5566 @item .ifndef @var{symbol}
5567 @itemx .ifnotdef @var{symbol}
5568 Assembles the following section of code if the specified @var{symbol}
5569 has not been defined. Both spelling variants are equivalent. Note a symbol
5570 which has been referenced but not yet defined is considered to be undefined.
5572 @cindex @code{ifne} directive
5573 @item .ifne @var{absolute expression}
5574 Assembles the following section of code if the argument is not equal to zero
5575 (in other words, this is equivalent to @code{.if}).
5577 @cindex @code{ifnes} directive
5578 @item .ifnes @var{string1},@var{string2}
5579 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5580 following section of code if the two strings are not the same.
5584 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5586 @cindex @code{incbin} directive
5587 @cindex binary files, including
5588 The @code{incbin} directive includes @var{file} verbatim at the current
5589 location. You can control the search paths used with the @samp{-I} command-line
5590 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5593 The @var{skip} argument skips a number of bytes from the start of the
5594 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5595 read. Note that the data is not aligned in any way, so it is the user's
5596 responsibility to make sure that proper alignment is provided both before and
5597 after the @code{incbin} directive.
5600 @section @code{.include "@var{file}"}
5602 @cindex @code{include} directive
5603 @cindex supporting files, including
5604 @cindex files, including
5605 This directive provides a way to include supporting files at specified
5606 points in your source program. The code from @var{file} is assembled as
5607 if it followed the point of the @code{.include}; when the end of the
5608 included file is reached, assembly of the original file continues. You
5609 can control the search paths used with the @samp{-I} command-line option
5610 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5614 @section @code{.int @var{expressions}}
5616 @cindex @code{int} directive
5617 @cindex integers, 32-bit
5618 Expect zero or more @var{expressions}, of any section, separated by commas.
5619 For each expression, emit a number that, at run time, is the value of that
5620 expression. The byte order and bit size of the number depends on what kind
5621 of target the assembly is for.
5625 On most forms of the H8/300, @code{.int} emits 16-bit
5626 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5633 @section @code{.internal @var{names}}
5635 @cindex @code{internal} directive
5637 This is one of the ELF visibility directives. The other two are
5638 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5639 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5641 This directive overrides the named symbols default visibility (which is set by
5642 their binding: local, global or weak). The directive sets the visibility to
5643 @code{internal} which means that the symbols are considered to be @code{hidden}
5644 (i.e., not visible to other components), and that some extra, processor specific
5645 processing must also be performed upon the symbols as well.
5649 @section @code{.irp @var{symbol},@var{values}}@dots{}
5651 @cindex @code{irp} directive
5652 Evaluate a sequence of statements assigning different values to @var{symbol}.
5653 The sequence of statements starts at the @code{.irp} directive, and is
5654 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5655 set to @var{value}, and the sequence of statements is assembled. If no
5656 @var{value} is listed, the sequence of statements is assembled once, with
5657 @var{symbol} set to the null string. To refer to @var{symbol} within the
5658 sequence of statements, use @var{\symbol}.
5660 For example, assembling
5668 is equivalent to assembling
5676 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5679 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5681 @cindex @code{irpc} directive
5682 Evaluate a sequence of statements assigning different values to @var{symbol}.
5683 The sequence of statements starts at the @code{.irpc} directive, and is
5684 terminated by an @code{.endr} directive. For each character in @var{value},
5685 @var{symbol} is set to the character, and the sequence of statements is
5686 assembled. If no @var{value} is listed, the sequence of statements is
5687 assembled once, with @var{symbol} set to the null string. To refer to
5688 @var{symbol} within the sequence of statements, use @var{\symbol}.
5690 For example, assembling
5698 is equivalent to assembling
5706 For some caveats with the spelling of @var{symbol}, see also the discussion
5710 @section @code{.lcomm @var{symbol} , @var{length}}
5712 @cindex @code{lcomm} directive
5713 @cindex local common symbols
5714 @cindex symbols, local common
5715 Reserve @var{length} (an absolute expression) bytes for a local common
5716 denoted by @var{symbol}. The section and value of @var{symbol} are
5717 those of the new local common. The addresses are allocated in the bss
5718 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5719 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5720 not visible to @code{@value{LD}}.
5723 Some targets permit a third argument to be used with @code{.lcomm}. This
5724 argument specifies the desired alignment of the symbol in the bss section.
5728 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5729 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5733 @section @code{.lflags}
5735 @cindex @code{lflags} directive (ignored)
5736 @command{@value{AS}} accepts this directive, for compatibility with other
5737 assemblers, but ignores it.
5739 @ifclear no-line-dir
5741 @section @code{.line @var{line-number}}
5743 @cindex @code{line} directive
5744 @cindex logical line number
5746 Change the logical line number. @var{line-number} must be an absolute
5747 expression. The next line has that logical line number. Therefore any other
5748 statements on the current line (after a statement separator character) are
5749 reported as on logical line number @var{line-number} @minus{} 1. One day
5750 @command{@value{AS}} will no longer support this directive: it is recognized only
5751 for compatibility with existing assembler programs.
5754 Even though this is a directive associated with the @code{a.out} or
5755 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5756 when producing COFF output, and treats @samp{.line} as though it
5757 were the COFF @samp{.ln} @emph{if} it is found outside a
5758 @code{.def}/@code{.endef} pair.
5760 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5761 used by compilers to generate auxiliary symbol information for
5766 @section @code{.linkonce [@var{type}]}
5768 @cindex @code{linkonce} directive
5769 @cindex common sections
5770 Mark the current section so that the linker only includes a single copy of it.
5771 This may be used to include the same section in several different object files,
5772 but ensure that the linker will only include it once in the final output file.
5773 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5774 Duplicate sections are detected based on the section name, so it should be
5777 This directive is only supported by a few object file formats; as of this
5778 writing, the only object file format which supports it is the Portable
5779 Executable format used on Windows NT.
5781 The @var{type} argument is optional. If specified, it must be one of the
5782 following strings. For example:
5786 Not all types may be supported on all object file formats.
5790 Silently discard duplicate sections. This is the default.
5793 Warn if there are duplicate sections, but still keep only one copy.
5796 Warn if any of the duplicates have different sizes.
5799 Warn if any of the duplicates do not have exactly the same contents.
5803 @section @code{.list}
5805 @cindex @code{list} directive
5806 @cindex listing control, turning on
5807 Control (in conjunction with the @code{.nolist} directive) whether or
5808 not assembly listings are generated. These two directives maintain an
5809 internal counter (which is zero initially). @code{.list} increments the
5810 counter, and @code{.nolist} decrements it. Assembly listings are
5811 generated whenever the counter is greater than zero.
5813 By default, listings are disabled. When you enable them (with the
5814 @samp{-a} command-line option; @pxref{Invoking,,Command-Line Options}),
5815 the initial value of the listing counter is one.
5818 @section @code{.ln @var{line-number}}
5820 @cindex @code{ln} directive
5821 @ifclear no-line-dir
5822 @samp{.ln} is a synonym for @samp{.line}.
5825 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5826 must be an absolute expression. The next line has that logical
5827 line number, so any other statements on the current line (after a
5828 statement separator character @code{;}) are reported as on logical
5829 line number @var{line-number} @minus{} 1.
5833 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5834 @cindex @code{loc} directive
5835 When emitting DWARF2 line number information,
5836 the @code{.loc} directive will add a row to the @code{.debug_line} line
5837 number matrix corresponding to the immediately following assembly
5838 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5839 arguments will be applied to the @code{.debug_line} state machine before
5842 The @var{options} are a sequence of the following tokens in any order:
5846 This option will set the @code{basic_block} register in the
5847 @code{.debug_line} state machine to @code{true}.
5850 This option will set the @code{prologue_end} register in the
5851 @code{.debug_line} state machine to @code{true}.
5853 @item epilogue_begin
5854 This option will set the @code{epilogue_begin} register in the
5855 @code{.debug_line} state machine to @code{true}.
5857 @item is_stmt @var{value}
5858 This option will set the @code{is_stmt} register in the
5859 @code{.debug_line} state machine to @code{value}, which must be
5862 @item isa @var{value}
5863 This directive will set the @code{isa} register in the @code{.debug_line}
5864 state machine to @var{value}, which must be an unsigned integer.
5866 @item discriminator @var{value}
5867 This directive will set the @code{discriminator} register in the @code{.debug_line}
5868 state machine to @var{value}, which must be an unsigned integer.
5870 @item view @var{value}
5871 This option causes a row to be added to @code{.debug_line} in reference to the
5872 current address (which might not be the same as that of the following assembly
5873 instruction), and to associate @var{value} with the @code{view} register in the
5874 @code{.debug_line} state machine. If @var{value} is a label, both the
5875 @code{view} register and the label are set to the number of prior @code{.loc}
5876 directives at the same program location. If @var{value} is the literal
5877 @code{0}, the @code{view} register is set to zero, and the assembler asserts
5878 that there aren't any prior @code{.loc} directives at the same program
5879 location. If @var{value} is the literal @code{-0}, the assembler arrange for
5880 the @code{view} register to be reset in this row, even if there are prior
5881 @code{.loc} directives at the same program location.
5885 @node Loc_mark_labels
5886 @section @code{.loc_mark_labels @var{enable}}
5887 @cindex @code{loc_mark_labels} directive
5888 When emitting DWARF2 line number information,
5889 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5890 to the @code{.debug_line} line number matrix with the @code{basic_block}
5891 register in the state machine set whenever a code label is seen.
5892 The @var{enable} argument should be either 1 or 0, to enable or disable
5893 this function respectively.
5897 @section @code{.local @var{names}}
5899 @cindex @code{local} directive
5900 This directive, which is available for ELF targets, marks each symbol in
5901 the comma-separated list of @code{names} as a local symbol so that it
5902 will not be externally visible. If the symbols do not already exist,
5903 they will be created.
5905 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5906 accept an alignment argument, which is the case for most ELF targets,
5907 the @code{.local} directive can be used in combination with @code{.comm}
5908 (@pxref{Comm}) to define aligned local common data.
5912 @section @code{.long @var{expressions}}
5914 @cindex @code{long} directive
5915 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5918 @c no one seems to know what this is for or whether this description is
5919 @c what it really ought to do
5921 @section @code{.lsym @var{symbol}, @var{expression}}
5923 @cindex @code{lsym} directive
5924 @cindex symbol, not referenced in assembly
5925 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5926 the hash table, ensuring it cannot be referenced by name during the
5927 rest of the assembly. This sets the attributes of the symbol to be
5928 the same as the expression value:
5930 @var{other} = @var{descriptor} = 0
5931 @var{type} = @r{(section of @var{expression})}
5932 @var{value} = @var{expression}
5935 The new symbol is not flagged as external.
5939 @section @code{.macro}
5942 The commands @code{.macro} and @code{.endm} allow you to define macros that
5943 generate assembly output. For example, this definition specifies a macro
5944 @code{sum} that puts a sequence of numbers into memory:
5947 .macro sum from=0, to=5
5956 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5968 @item .macro @var{macname}
5969 @itemx .macro @var{macname} @var{macargs} @dots{}
5970 @cindex @code{macro} directive
5971 Begin the definition of a macro called @var{macname}. If your macro
5972 definition requires arguments, specify their names after the macro name,
5973 separated by commas or spaces. You can qualify the macro argument to
5974 indicate whether all invocations must specify a non-blank value (through
5975 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5976 (through @samp{:@code{vararg}}). You can supply a default value for any
5977 macro argument by following the name with @samp{=@var{deflt}}. You
5978 cannot define two macros with the same @var{macname} unless it has been
5979 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5980 definitions. For example, these are all valid @code{.macro} statements:
5984 Begin the definition of a macro called @code{comm}, which takes no
5987 @item .macro plus1 p, p1
5988 @itemx .macro plus1 p p1
5989 Either statement begins the definition of a macro called @code{plus1},
5990 which takes two arguments; within the macro definition, write
5991 @samp{\p} or @samp{\p1} to evaluate the arguments.
5993 @item .macro reserve_str p1=0 p2
5994 Begin the definition of a macro called @code{reserve_str}, with two
5995 arguments. The first argument has a default value, but not the second.
5996 After the definition is complete, you can call the macro either as
5997 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5998 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5999 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
6000 @samp{0}, and @samp{\p2} evaluating to @var{b}).
6002 @item .macro m p1:req, p2=0, p3:vararg
6003 Begin the definition of a macro called @code{m}, with at least three
6004 arguments. The first argument must always have a value specified, but
6005 not the second, which instead has a default value. The third formal
6006 will get assigned all remaining arguments specified at invocation time.
6008 When you call a macro, you can specify the argument values either by
6009 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
6010 @samp{sum to=17, from=9}.
6014 Note that since each of the @var{macargs} can be an identifier exactly
6015 as any other one permitted by the target architecture, there may be
6016 occasional problems if the target hand-crafts special meanings to certain
6017 characters when they occur in a special position. For example, if the colon
6018 (@code{:}) is generally permitted to be part of a symbol name, but the
6019 architecture specific code special-cases it when occurring as the final
6020 character of a symbol (to denote a label), then the macro parameter
6021 replacement code will have no way of knowing that and consider the whole
6022 construct (including the colon) an identifier, and check only this
6023 identifier for being the subject to parameter substitution. So for example
6024 this macro definition:
6032 might not work as expected. Invoking @samp{label foo} might not create a label
6033 called @samp{foo} but instead just insert the text @samp{\l:} into the
6034 assembler source, probably generating an error about an unrecognised
6037 Similarly problems might occur with the period character (@samp{.})
6038 which is often allowed inside opcode names (and hence identifier names). So
6039 for example constructing a macro to build an opcode from a base name and a
6040 length specifier like this:
6043 .macro opcode base length
6048 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
6049 instruction but instead generate some kind of error as the assembler tries to
6050 interpret the text @samp{\base.\length}.
6052 There are several possible ways around this problem:
6055 @item Insert white space
6056 If it is possible to use white space characters then this is the simplest
6065 @item Use @samp{\()}
6066 The string @samp{\()} can be used to separate the end of a macro argument from
6067 the following text. eg:
6070 .macro opcode base length
6075 @item Use the alternate macro syntax mode
6076 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
6077 used as a separator. eg:
6087 Note: this problem of correctly identifying string parameters to pseudo ops
6088 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
6089 and @code{.irpc} (@pxref{Irpc}) as well.
6092 @cindex @code{endm} directive
6093 Mark the end of a macro definition.
6096 @cindex @code{exitm} directive
6097 Exit early from the current macro definition.
6099 @cindex number of macros executed
6100 @cindex macros, count executed
6102 @command{@value{AS}} maintains a counter of how many macros it has
6103 executed in this pseudo-variable; you can copy that number to your
6104 output with @samp{\@@}, but @emph{only within a macro definition}.
6106 @item LOCAL @var{name} [ , @dots{} ]
6107 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
6108 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
6109 @xref{Altmacro,,@code{.altmacro}}.
6113 @section @code{.mri @var{val}}
6115 @cindex @code{mri} directive
6116 @cindex MRI mode, temporarily
6117 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
6118 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
6119 affects code assembled until the next @code{.mri} directive, or until the end
6120 of the file. @xref{M, MRI mode, MRI mode}.
6123 @section @code{.noaltmacro}
6124 Disable alternate macro mode. @xref{Altmacro}.
6127 @section @code{.nolist}
6129 @cindex @code{nolist} directive
6130 @cindex listing control, turning off
6131 Control (in conjunction with the @code{.list} directive) whether or
6132 not assembly listings are generated. These two directives maintain an
6133 internal counter (which is zero initially). @code{.list} increments the
6134 counter, and @code{.nolist} decrements it. Assembly listings are
6135 generated whenever the counter is greater than zero.
6138 @section @code{.nops @var{size}[, @var{control}]}
6140 @cindex @code{nops} directive
6141 @cindex filling memory with no-op instructions
6142 This directive emits @var{size} bytes filled with no-op instructions.
6143 @var{size} is absolute expression, which must be a positve value.
6144 @var{control} controls how no-op instructions should be generated. If
6145 the comma and @var{control} are omitted, @var{control} is assumed to be
6148 Note: For Intel 80386 and AMD x86-64 targets, @var{control} specifies
6149 the size limit of a no-op instruction. The valid values of @var{control}
6150 are between 0 and 4 in 16-bit mode, between 0 and 7 when tuning for
6151 older processors in 32-bit mode, between 0 and 11 in 64-bit mode or when
6152 tuning for newer processors in 32-bit mode. When 0 is used, the no-op
6153 instruction size limit is set to the maximum supported size.
6156 @section @code{.octa @var{bignums}}
6158 @c FIXME: double size emitted for "octa" on some? Or warn?
6159 @cindex @code{octa} directive
6160 @cindex integer, 16-byte
6161 @cindex sixteen byte integer
6162 This directive expects zero or more bignums, separated by commas. For each
6163 bignum, it emits a 16-byte integer.
6165 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6166 hence @emph{octa}-word for 16 bytes.
6169 @section @code{.offset @var{loc}}
6171 @cindex @code{offset} directive
6172 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6173 be an absolute expression. This directive may be useful for defining
6174 symbols with absolute values. Do not confuse it with the @code{.org}
6178 @section @code{.org @var{new-lc} , @var{fill}}
6180 @cindex @code{org} directive
6181 @cindex location counter, advancing
6182 @cindex advancing location counter
6183 @cindex current address, advancing
6184 Advance the location counter of the current section to
6185 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6186 expression with the same section as the current subsection. That is,
6187 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6188 wrong section, the @code{.org} directive is ignored. To be compatible
6189 with former assemblers, if the section of @var{new-lc} is absolute,
6190 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6191 is the same as the current subsection.
6193 @code{.org} may only increase the location counter, or leave it
6194 unchanged; you cannot use @code{.org} to move the location counter
6197 @c double negative used below "not undefined" because this is a specific
6198 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6199 @c section. doc@cygnus.com 18feb91
6200 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6201 may not be undefined. If you really detest this restriction we eagerly await
6202 a chance to share your improved assembler.
6204 Beware that the origin is relative to the start of the section, not
6205 to the start of the subsection. This is compatible with other
6206 people's assemblers.
6208 When the location counter (of the current subsection) is advanced, the
6209 intervening bytes are filled with @var{fill} which should be an
6210 absolute expression. If the comma and @var{fill} are omitted,
6211 @var{fill} defaults to zero.
6214 @section @code{.p2align[wl] [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
6216 @cindex padding the location counter given a power of two
6217 @cindex @code{p2align} directive
6218 Pad the location counter (in the current subsection) to a particular
6219 storage boundary. The first expression (which must be absolute) is the
6220 number of low-order zero bits the location counter must have after
6221 advancement. For example @samp{.p2align 3} advances the location
6222 counter until it is a multiple of 8. If the location counter is already a
6223 multiple of 8, no change is needed. If the expression is omitted then a
6224 default value of 0 is used, effectively disabling alignment requirements.
6226 The second expression (also absolute) gives the fill value to be stored in the
6227 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6228 padding bytes are normally zero. However, on most systems, if the section is
6229 marked as containing code and the fill value is omitted, the space is filled
6230 with no-op instructions.
6232 The third expression is also absolute, and is also optional. If it is present,
6233 it is the maximum number of bytes that should be skipped by this alignment
6234 directive. If doing the alignment would require skipping more bytes than the
6235 specified maximum, then the alignment is not done at all. You can omit the
6236 fill value (the second argument) entirely by simply using two commas after the
6237 required alignment; this can be useful if you want the alignment to be filled
6238 with no-op instructions when appropriate.
6240 @cindex @code{p2alignw} directive
6241 @cindex @code{p2alignl} directive
6242 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6243 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6244 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6245 fill pattern as a four byte longword value. For example, @code{.p2alignw
6246 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6247 filled in with the value 0x368d (the exact placement of the bytes depends upon
6248 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6253 @section @code{.popsection}
6255 @cindex @code{popsection} directive
6256 @cindex Section Stack
6257 This is one of the ELF section stack manipulation directives. The others are
6258 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6259 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6262 This directive replaces the current section (and subsection) with the top
6263 section (and subsection) on the section stack. This section is popped off the
6269 @section @code{.previous}
6271 @cindex @code{previous} directive
6272 @cindex Section Stack
6273 This is one of the ELF section stack manipulation directives. The others are
6274 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6275 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6276 (@pxref{PopSection}).
6278 This directive swaps the current section (and subsection) with most recently
6279 referenced section/subsection pair prior to this one. Multiple
6280 @code{.previous} directives in a row will flip between two sections (and their
6281 subsections). For example:
6293 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6299 # Now in section A subsection 1
6303 # Now in section B subsection 0
6306 # Now in section B subsection 1
6309 # Now in section B subsection 0
6313 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6314 section B and 0x9abc into subsection 1 of section B.
6316 In terms of the section stack, this directive swaps the current section with
6317 the top section on the section stack.
6321 @section @code{.print @var{string}}
6323 @cindex @code{print} directive
6324 @command{@value{AS}} will print @var{string} on the standard output during
6325 assembly. You must put @var{string} in double quotes.
6329 @section @code{.protected @var{names}}
6331 @cindex @code{protected} directive
6333 This is one of the ELF visibility directives. The other two are
6334 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6336 This directive overrides the named symbols default visibility (which is set by
6337 their binding: local, global or weak). The directive sets the visibility to
6338 @code{protected} which means that any references to the symbols from within the
6339 components that defines them must be resolved to the definition in that
6340 component, even if a definition in another component would normally preempt
6345 @section @code{.psize @var{lines} , @var{columns}}
6347 @cindex @code{psize} directive
6348 @cindex listing control: paper size
6349 @cindex paper size, for listings
6350 Use this directive to declare the number of lines---and, optionally, the
6351 number of columns---to use for each page, when generating listings.
6353 If you do not use @code{.psize}, listings use a default line-count
6354 of 60. You may omit the comma and @var{columns} specification; the
6355 default width is 200 columns.
6357 @command{@value{AS}} generates formfeeds whenever the specified number of
6358 lines is exceeded (or whenever you explicitly request one, using
6361 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6362 those explicitly specified with @code{.eject}.
6365 @section @code{.purgem @var{name}}
6367 @cindex @code{purgem} directive
6368 Undefine the macro @var{name}, so that later uses of the string will not be
6369 expanded. @xref{Macro}.
6373 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6375 @cindex @code{pushsection} directive
6376 @cindex Section Stack
6377 This is one of the ELF section stack manipulation directives. The others are
6378 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6379 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6382 This directive pushes the current section (and subsection) onto the
6383 top of the section stack, and then replaces the current section and
6384 subsection with @code{name} and @code{subsection}. The optional
6385 @code{flags}, @code{type} and @code{arguments} are treated the same
6386 as in the @code{.section} (@pxref{Section}) directive.
6390 @section @code{.quad @var{bignums}}
6392 @cindex @code{quad} directive
6393 @code{.quad} expects zero or more bignums, separated by commas. For
6394 each bignum, it emits
6396 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6397 warning message; and just takes the lowest order 8 bytes of the bignum.
6398 @cindex eight-byte integer
6399 @cindex integer, 8-byte
6401 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6402 hence @emph{quad}-word for 8 bytes.
6405 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6406 warning message; and just takes the lowest order 16 bytes of the bignum.
6407 @cindex sixteen-byte integer
6408 @cindex integer, 16-byte
6412 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6414 @cindex @code{reloc} directive
6415 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6416 @var{expression}. If @var{offset} is a number, the relocation is generated in
6417 the current section. If @var{offset} is an expression that resolves to a
6418 symbol plus offset, the relocation is generated in the given symbol's section.
6419 @var{expression}, if present, must resolve to a symbol plus addend or to an
6420 absolute value, but note that not all targets support an addend. e.g. ELF REL
6421 targets such as i386 store an addend in the section contents rather than in the
6422 relocation. This low level interface does not support addends stored in the
6426 @section @code{.rept @var{count}}
6428 @cindex @code{rept} directive
6429 Repeat the sequence of lines between the @code{.rept} directive and the next
6430 @code{.endr} directive @var{count} times.
6432 For example, assembling
6440 is equivalent to assembling
6448 A count of zero is allowed, but nothing is generated. Negative counts are not
6449 allowed and if encountered will be treated as if they were zero.
6452 @section @code{.sbttl "@var{subheading}"}
6454 @cindex @code{sbttl} directive
6455 @cindex subtitles for listings
6456 @cindex listing control: subtitle
6457 Use @var{subheading} as the title (third line, immediately after the
6458 title line) when generating assembly listings.
6460 This directive affects subsequent pages, as well as the current page if
6461 it appears within ten lines of the top of a page.
6465 @section @code{.scl @var{class}}
6467 @cindex @code{scl} directive
6468 @cindex symbol storage class (COFF)
6469 @cindex COFF symbol storage class
6470 Set the storage-class value for a symbol. This directive may only be
6471 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6472 whether a symbol is static or external, or it may record further
6473 symbolic debugging information.
6478 @section @code{.section @var{name}}
6480 @cindex named section
6481 Use the @code{.section} directive to assemble the following code into a section
6484 This directive is only supported for targets that actually support arbitrarily
6485 named sections; on @code{a.out} targets, for example, it is not accepted, even
6486 with a standard @code{a.out} section name.
6490 @c only print the extra heading if both COFF and ELF are set
6491 @subheading COFF Version
6494 @cindex @code{section} directive (COFF version)
6495 For COFF targets, the @code{.section} directive is used in one of the following
6499 .section @var{name}[, "@var{flags}"]
6500 .section @var{name}[, @var{subsection}]
6503 If the optional argument is quoted, it is taken as flags to use for the
6504 section. Each flag is a single character. The following flags are recognized:
6508 bss section (uninitialized data)
6510 section is not loaded
6516 exclude section from linking
6522 shared section (meaningful for PE targets)
6524 ignored. (For compatibility with the ELF version)
6526 section is not readable (meaningful for PE targets)
6528 single-digit power-of-two section alignment (GNU extension)
6531 If no flags are specified, the default flags depend upon the section name. If
6532 the section name is not recognized, the default will be for the section to be
6533 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6534 from the section, rather than adding them, so if they are used on their own it
6535 will be as if no flags had been specified at all.
6537 If the optional argument to the @code{.section} directive is not quoted, it is
6538 taken as a subsection number (@pxref{Sub-Sections}).
6543 @c only print the extra heading if both COFF and ELF are set
6544 @subheading ELF Version
6547 @cindex Section Stack
6548 This is one of the ELF section stack manipulation directives. The others are
6549 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6550 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6551 @code{.previous} (@pxref{Previous}).
6553 @cindex @code{section} directive (ELF version)
6554 For ELF targets, the @code{.section} directive is used like this:
6557 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6560 @anchor{Section Name Substitutions}
6561 @kindex --sectname-subst
6562 @cindex section name substitution
6563 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6564 argument may contain a substitution sequence. Only @code{%S} is supported
6565 at the moment, and substitutes the current section name. For example:
6568 .macro exception_code
6569 .section %S.exception
6570 [exception code here]
6585 The two @code{exception_code} invocations above would create the
6586 @code{.text.exception} and @code{.init.exception} sections respectively.
6587 This is useful e.g. to discriminate between ancillary sections that are
6588 tied to setup code to be discarded after use from ancillary sections that
6589 need to stay resident without having to define multiple @code{exception_code}
6590 macros just for that purpose.
6592 The optional @var{flags} argument is a quoted string which may contain any
6593 combination of the following characters:
6597 section is allocatable
6599 section is a GNU_MBIND section
6601 section is excluded from executable and shared library.
6605 section is executable
6607 section is mergeable
6609 section contains zero terminated strings
6611 section is a member of a section group
6613 section is used for thread-local-storage
6615 section is a member of the previously-current section's group, if any
6616 @item @code{<number>}
6617 a numeric value indicating the bits to be set in the ELF section header's flags
6618 field. Note - if one or more of the alphabetic characters described above is
6619 also included in the flags field, their bit values will be ORed into the
6621 @item @code{<target specific>}
6622 some targets extend this list with their own flag characters
6625 Note - once a section's flags have been set they cannot be changed. There are
6626 a few exceptions to this rule however. Processor and application specific
6627 flags can be added to an already defined section. The @code{.interp},
6628 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6629 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6630 section may have the executable (@code{x}) flag added.
6632 The optional @var{type} argument may contain one of the following constants:
6636 section contains data
6638 section does not contain data (i.e., section only occupies space)
6640 section contains data which is used by things other than the program
6642 section contains an array of pointers to init functions
6644 section contains an array of pointers to finish functions
6645 @item @@preinit_array
6646 section contains an array of pointers to pre-init functions
6647 @item @@@code{<number>}
6648 a numeric value to be set as the ELF section header's type field.
6649 @item @@@code{<target specific>}
6650 some targets extend this list with their own types
6653 Many targets only support the first three section types. The type may be
6654 enclosed in double quotes if necessary.
6656 Note on targets where the @code{@@} character is the start of a comment (eg
6657 ARM) then another character is used instead. For example the ARM port uses the
6660 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6661 special and have fixed types. Any attempt to declare them with a different
6662 type will generate an error from the assembler.
6664 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6665 be specified as well as an extra argument---@var{entsize}---like this:
6668 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6671 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6672 constants, each @var{entsize} octets long. Sections with both @code{M} and
6673 @code{S} must contain zero terminated strings where each character is
6674 @var{entsize} bytes long. The linker may remove duplicates within sections with
6675 the same name, same entity size and same flags. @var{entsize} must be an
6676 absolute expression. For sections with both @code{M} and @code{S}, a string
6677 which is a suffix of a larger string is considered a duplicate. Thus
6678 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6679 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6681 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6682 be present along with an additional field like this:
6685 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6688 The @var{GroupName} field specifies the name of the section group to which this
6689 particular section belongs. The optional linkage field can contain:
6693 indicates that only one copy of this section should be retained
6698 Note: if both the @var{M} and @var{G} flags are present then the fields for
6699 the Merge flag should come first, like this:
6702 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6705 If @var{flags} contains the @code{?} symbol then it may not also contain the
6706 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6707 present. Instead, @code{?} says to consider the section that's current before
6708 this directive. If that section used @code{G}, then the new section will use
6709 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6710 If not, then the @code{?} symbol has no effect.
6712 The optional @var{unique,@code{<number>}} argument must come last. It
6713 assigns @var{@code{<number>}} as a unique section ID to distinguish
6714 different sections with the same section name like these:
6717 .section @var{name},"@var{flags}",@@@var{type},@var{unique,@code{<number>}}
6718 .section @var{name},"@var{flags}"G,@@@var{type},@var{GroupName},[@var{linkage}],@var{unique,@code{<number>}}
6719 .section @var{name},"@var{flags}"MG,@@@var{type},@var{entsize},@var{GroupName}[,@var{linkage}],@var{unique,@code{<number>}}
6722 The valid values of @var{@code{<number>}} are between 0 and 4294967295.
6724 If no flags are specified, the default flags depend upon the section name. If
6725 the section name is not recognized, the default will be for the section to have
6726 none of the above flags: it will not be allocated in memory, nor writable, nor
6727 executable. The section will contain data.
6729 For ELF targets, the assembler supports another type of @code{.section}
6730 directive for compatibility with the Solaris assembler:
6733 .section "@var{name}"[, @var{flags}...]
6736 Note that the section name is quoted. There may be a sequence of comma
6741 section is allocatable
6745 section is executable
6747 section is excluded from executable and shared library.
6749 section is used for thread local storage
6752 This directive replaces the current section and subsection. See the
6753 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6754 some examples of how this directive and the other section stack directives
6760 @section @code{.set @var{symbol}, @var{expression}}
6762 @cindex @code{set} directive
6763 @cindex symbol value, setting
6764 Set the value of @var{symbol} to @var{expression}. This
6765 changes @var{symbol}'s value and type to conform to
6766 @var{expression}. If @var{symbol} was flagged as external, it remains
6767 flagged (@pxref{Symbol Attributes}).
6769 You may @code{.set} a symbol many times in the same assembly provided that the
6770 values given to the symbol are constants. Values that are based on expressions
6771 involving other symbols are allowed, but some targets may restrict this to only
6772 being done once per assembly. This is because those targets do not set the
6773 addresses of symbols at assembly time, but rather delay the assignment until a
6774 final link is performed. This allows the linker a chance to change the code in
6775 the files, changing the location of, and the relative distance between, various
6778 If you @code{.set} a global symbol, the value stored in the object
6779 file is the last value stored into it.
6782 On Z80 @code{set} is a real instruction, use @code{.set} or
6783 @samp{@var{symbol} defl @var{expression}} instead.
6787 @section @code{.short @var{expressions}}
6789 @cindex @code{short} directive
6791 @code{.short} is normally the same as @samp{.word}.
6792 @xref{Word,,@code{.word}}.
6794 In some configurations, however, @code{.short} and @code{.word} generate
6795 numbers of different lengths. @xref{Machine Dependencies}.
6799 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6802 This expects zero or more @var{expressions}, and emits
6803 a 16 bit number for each.
6808 @section @code{.single @var{flonums}}
6810 @cindex @code{single} directive
6811 @cindex floating point numbers (single)
6812 This directive assembles zero or more flonums, separated by commas. It
6813 has the same effect as @code{.float}.
6815 The exact kind of floating point numbers emitted depends on how
6816 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6820 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6821 numbers in @sc{ieee} format.
6827 @section @code{.size}
6829 This directive is used to set the size associated with a symbol.
6833 @c only print the extra heading if both COFF and ELF are set
6834 @subheading COFF Version
6837 @cindex @code{size} directive (COFF version)
6838 For COFF targets, the @code{.size} directive is only permitted inside
6839 @code{.def}/@code{.endef} pairs. It is used like this:
6842 .size @var{expression}
6849 @c only print the extra heading if both COFF and ELF are set
6850 @subheading ELF Version
6853 @cindex @code{size} directive (ELF version)
6854 For ELF targets, the @code{.size} directive is used like this:
6857 .size @var{name} , @var{expression}
6860 This directive sets the size associated with a symbol @var{name}.
6861 The size in bytes is computed from @var{expression} which can make use of label
6862 arithmetic. This directive is typically used to set the size of function
6867 @ifclear no-space-dir
6869 @section @code{.skip @var{size} [,@var{fill}]}
6871 @cindex @code{skip} directive
6872 @cindex filling memory
6873 This directive emits @var{size} bytes, each of value @var{fill}. Both
6874 @var{size} and @var{fill} are absolute expressions. If the comma and
6875 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6880 @section @code{.sleb128 @var{expressions}}
6882 @cindex @code{sleb128} directive
6883 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6884 compact, variable length representation of numbers used by the DWARF
6885 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6887 @ifclear no-space-dir
6889 @section @code{.space @var{size} [,@var{fill}]}
6891 @cindex @code{space} directive
6892 @cindex filling memory
6893 This directive emits @var{size} bytes, each of value @var{fill}. Both
6894 @var{size} and @var{fill} are absolute expressions. If the comma
6895 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6900 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6901 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6902 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6903 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6911 @section @code{.stabd, .stabn, .stabs}
6913 @cindex symbolic debuggers, information for
6914 @cindex @code{stab@var{x}} directives
6915 There are three directives that begin @samp{.stab}.
6916 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6917 The symbols are not entered in the @command{@value{AS}} hash table: they
6918 cannot be referenced elsewhere in the source file.
6919 Up to five fields are required:
6923 This is the symbol's name. It may contain any character except
6924 @samp{\000}, so is more general than ordinary symbol names. Some
6925 debuggers used to code arbitrarily complex structures into symbol names
6929 An absolute expression. The symbol's type is set to the low 8 bits of
6930 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6931 and debuggers choke on silly bit patterns.
6934 An absolute expression. The symbol's ``other'' attribute is set to the
6935 low 8 bits of this expression.
6938 An absolute expression. The symbol's descriptor is set to the low 16
6939 bits of this expression.
6942 An absolute expression which becomes the symbol's value.
6945 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6946 or @code{.stabs} statement, the symbol has probably already been created;
6947 you get a half-formed symbol in your object file. This is
6948 compatible with earlier assemblers!
6951 @cindex @code{stabd} directive
6952 @item .stabd @var{type} , @var{other} , @var{desc}
6954 The ``name'' of the symbol generated is not even an empty string.
6955 It is a null pointer, for compatibility. Older assemblers used a
6956 null pointer so they didn't waste space in object files with empty
6959 The symbol's value is set to the location counter,
6960 relocatably. When your program is linked, the value of this symbol
6961 is the address of the location counter when the @code{.stabd} was
6964 @cindex @code{stabn} directive
6965 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6966 The name of the symbol is set to the empty string @code{""}.
6968 @cindex @code{stabs} directive
6969 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6970 All five fields are specified.
6976 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6977 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6979 @cindex string, copying to object file
6980 @cindex string8, copying to object file
6981 @cindex string16, copying to object file
6982 @cindex string32, copying to object file
6983 @cindex string64, copying to object file
6984 @cindex @code{string} directive
6985 @cindex @code{string8} directive
6986 @cindex @code{string16} directive
6987 @cindex @code{string32} directive
6988 @cindex @code{string64} directive
6990 Copy the characters in @var{str} to the object file. You may specify more than
6991 one string to copy, separated by commas. Unless otherwise specified for a
6992 particular machine, the assembler marks the end of each string with a 0 byte.
6993 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6995 The variants @code{string16}, @code{string32} and @code{string64} differ from
6996 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6997 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6998 are stored in target endianness byte order.
7004 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
7005 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
7010 @section @code{.struct @var{expression}}
7012 @cindex @code{struct} directive
7013 Switch to the absolute section, and set the section offset to @var{expression},
7014 which must be an absolute expression. You might use this as follows:
7023 This would define the symbol @code{field1} to have the value 0, the symbol
7024 @code{field2} to have the value 4, and the symbol @code{field3} to have the
7025 value 8. Assembly would be left in the absolute section, and you would need to
7026 use a @code{.section} directive of some sort to change to some other section
7027 before further assembly.
7031 @section @code{.subsection @var{name}}
7033 @cindex @code{subsection} directive
7034 @cindex Section Stack
7035 This is one of the ELF section stack manipulation directives. The others are
7036 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
7037 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
7040 This directive replaces the current subsection with @code{name}. The current
7041 section is not changed. The replaced subsection is put onto the section stack
7042 in place of the then current top of stack subsection.
7047 @section @code{.symver}
7048 @cindex @code{symver} directive
7049 @cindex symbol versioning
7050 @cindex versions of symbols
7051 Use the @code{.symver} directive to bind symbols to specific version nodes
7052 within a source file. This is only supported on ELF platforms, and is
7053 typically used when assembling files to be linked into a shared library.
7054 There are cases where it may make sense to use this in objects to be bound
7055 into an application itself so as to override a versioned symbol from a
7058 For ELF targets, the @code{.symver} directive can be used like this:
7060 .symver @var{name}, @var{name2@@nodename}
7062 If the symbol @var{name} is defined within the file
7063 being assembled, the @code{.symver} directive effectively creates a symbol
7064 alias with the name @var{name2@@nodename}, and in fact the main reason that we
7065 just don't try and create a regular alias is that the @var{@@} character isn't
7066 permitted in symbol names. The @var{name2} part of the name is the actual name
7067 of the symbol by which it will be externally referenced. The name @var{name}
7068 itself is merely a name of convenience that is used so that it is possible to
7069 have definitions for multiple versions of a function within a single source
7070 file, and so that the compiler can unambiguously know which version of a
7071 function is being mentioned. The @var{nodename} portion of the alias should be
7072 the name of a node specified in the version script supplied to the linker when
7073 building a shared library. If you are attempting to override a versioned
7074 symbol from a shared library, then @var{nodename} should correspond to the
7075 nodename of the symbol you are trying to override.
7077 If the symbol @var{name} is not defined within the file being assembled, all
7078 references to @var{name} will be changed to @var{name2@@nodename}. If no
7079 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
7082 Another usage of the @code{.symver} directive is:
7084 .symver @var{name}, @var{name2@@@@nodename}
7086 In this case, the symbol @var{name} must exist and be defined within
7087 the file being assembled. It is similar to @var{name2@@nodename}. The
7088 difference is @var{name2@@@@nodename} will also be used to resolve
7089 references to @var{name2} by the linker.
7091 The third usage of the @code{.symver} directive is:
7093 .symver @var{name}, @var{name2@@@@@@nodename}
7095 When @var{name} is not defined within the
7096 file being assembled, it is treated as @var{name2@@nodename}. When
7097 @var{name} is defined within the file being assembled, the symbol
7098 name, @var{name}, will be changed to @var{name2@@@@nodename}.
7103 @section @code{.tag @var{structname}}
7105 @cindex COFF structure debugging
7106 @cindex structure debugging, COFF
7107 @cindex @code{tag} directive
7108 This directive is generated by compilers to include auxiliary debugging
7109 information in the symbol table. It is only permitted inside
7110 @code{.def}/@code{.endef} pairs. Tags are used to link structure
7111 definitions in the symbol table with instances of those structures.
7115 @section @code{.text @var{subsection}}
7117 @cindex @code{text} directive
7118 Tells @command{@value{AS}} to assemble the following statements onto the end of
7119 the text subsection numbered @var{subsection}, which is an absolute
7120 expression. If @var{subsection} is omitted, subsection number zero
7124 @section @code{.title "@var{heading}"}
7126 @cindex @code{title} directive
7127 @cindex listing control: title line
7128 Use @var{heading} as the title (second line, immediately after the
7129 source file name and pagenumber) when generating assembly listings.
7131 This directive affects subsequent pages, as well as the current page if
7132 it appears within ten lines of the top of a page.
7136 @section @code{.type}
7138 This directive is used to set the type of a symbol.
7142 @c only print the extra heading if both COFF and ELF are set
7143 @subheading COFF Version
7146 @cindex COFF symbol type
7147 @cindex symbol type, COFF
7148 @cindex @code{type} directive (COFF version)
7149 For COFF targets, this directive is permitted only within
7150 @code{.def}/@code{.endef} pairs. It is used like this:
7156 This records the integer @var{int} as the type attribute of a symbol table
7163 @c only print the extra heading if both COFF and ELF are set
7164 @subheading ELF Version
7167 @cindex ELF symbol type
7168 @cindex symbol type, ELF
7169 @cindex @code{type} directive (ELF version)
7170 For ELF targets, the @code{.type} directive is used like this:
7173 .type @var{name} , @var{type description}
7176 This sets the type of symbol @var{name} to be either a
7177 function symbol or an object symbol. There are five different syntaxes
7178 supported for the @var{type description} field, in order to provide
7179 compatibility with various other assemblers.
7181 Because some of the characters used in these syntaxes (such as @samp{@@} and
7182 @samp{#}) are comment characters for some architectures, some of the syntaxes
7183 below do not work on all architectures. The first variant will be accepted by
7184 the GNU assembler on all architectures so that variant should be used for
7185 maximum portability, if you do not need to assemble your code with other
7188 The syntaxes supported are:
7191 .type <name> STT_<TYPE_IN_UPPER_CASE>
7192 .type <name>,#<type>
7193 .type <name>,@@<type>
7194 .type <name>,%<type>
7195 .type <name>,"<type>"
7198 The types supported are:
7203 Mark the symbol as being a function name.
7206 @itemx gnu_indirect_function
7207 Mark the symbol as an indirect function when evaluated during reloc
7208 processing. (This is only supported on assemblers targeting GNU systems).
7212 Mark the symbol as being a data object.
7216 Mark the symbol as being a thread-local data object.
7220 Mark the symbol as being a common data object.
7224 Does not mark the symbol in any way. It is supported just for completeness.
7226 @item gnu_unique_object
7227 Marks the symbol as being a globally unique data object. The dynamic linker
7228 will make sure that in the entire process there is just one symbol with this
7229 name and type in use. (This is only supported on assemblers targeting GNU
7234 Changing between incompatible types other than from/to STT_NOTYPE will
7235 result in a diagnostic. An intermediate change to STT_NOTYPE will silence
7238 Note: Some targets support extra types in addition to those listed above.
7244 @section @code{.uleb128 @var{expressions}}
7246 @cindex @code{uleb128} directive
7247 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7248 compact, variable length representation of numbers used by the DWARF
7249 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7253 @section @code{.val @var{addr}}
7255 @cindex @code{val} directive
7256 @cindex COFF value attribute
7257 @cindex value attribute, COFF
7258 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7259 records the address @var{addr} as the value attribute of a symbol table
7265 @section @code{.version "@var{string}"}
7267 @cindex @code{version} directive
7268 This directive creates a @code{.note} section and places into it an ELF
7269 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7274 @section @code{.vtable_entry @var{table}, @var{offset}}
7276 @cindex @code{vtable_entry} directive
7277 This directive finds or creates a symbol @code{table} and creates a
7278 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7281 @section @code{.vtable_inherit @var{child}, @var{parent}}
7283 @cindex @code{vtable_inherit} directive
7284 This directive finds the symbol @code{child} and finds or creates the symbol
7285 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7286 parent whose addend is the value of the child symbol. As a special case the
7287 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7291 @section @code{.warning "@var{string}"}
7292 @cindex warning directive
7293 Similar to the directive @code{.error}
7294 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7297 @section @code{.weak @var{names}}
7299 @cindex @code{weak} directive
7300 This directive sets the weak attribute on the comma separated list of symbol
7301 @code{names}. If the symbols do not already exist, they will be created.
7303 On COFF targets other than PE, weak symbols are a GNU extension. This
7304 directive sets the weak attribute on the comma separated list of symbol
7305 @code{names}. If the symbols do not already exist, they will be created.
7307 On the PE target, weak symbols are supported natively as weak aliases.
7308 When a weak symbol is created that is not an alias, GAS creates an
7309 alternate symbol to hold the default value.
7312 @section @code{.weakref @var{alias}, @var{target}}
7314 @cindex @code{weakref} directive
7315 This directive creates an alias to the target symbol that enables the symbol to
7316 be referenced with weak-symbol semantics, but without actually making it weak.
7317 If direct references or definitions of the symbol are present, then the symbol
7318 will not be weak, but if all references to it are through weak references, the
7319 symbol will be marked as weak in the symbol table.
7321 The effect is equivalent to moving all references to the alias to a separate
7322 assembly source file, renaming the alias to the symbol in it, declaring the
7323 symbol as weak there, and running a reloadable link to merge the object files
7324 resulting from the assembly of the new source file and the old source file that
7325 had the references to the alias removed.
7327 The alias itself never makes to the symbol table, and is entirely handled
7328 within the assembler.
7331 @section @code{.word @var{expressions}}
7333 @cindex @code{word} directive
7334 This directive expects zero or more @var{expressions}, of any section,
7335 separated by commas.
7338 For each expression, @command{@value{AS}} emits a 32-bit number.
7341 For each expression, @command{@value{AS}} emits a 16-bit number.
7346 The size of the number emitted, and its byte order,
7347 depend on what target computer the assembly is for.
7350 @c on sparc the "special treatment to support compilers" doesn't
7351 @c happen---32-bit addressability, period; no long/short jumps.
7352 @ifset DIFF-TBL-KLUGE
7353 @cindex difference tables altered
7354 @cindex altered difference tables
7356 @emph{Warning: Special Treatment to support Compilers}
7360 Machines with a 32-bit address space, but that do less than 32-bit
7361 addressing, require the following special treatment. If the machine of
7362 interest to you does 32-bit addressing (or doesn't require it;
7363 @pxref{Machine Dependencies}), you can ignore this issue.
7366 In order to assemble compiler output into something that works,
7367 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7368 Directives of the form @samp{.word sym1-sym2} are often emitted by
7369 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7370 directive of the form @samp{.word sym1-sym2}, and the difference between
7371 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7372 creates a @dfn{secondary jump table}, immediately before the next label.
7373 This secondary jump table is preceded by a short-jump to the
7374 first byte after the secondary table. This short-jump prevents the flow
7375 of control from accidentally falling into the new table. Inside the
7376 table is a long-jump to @code{sym2}. The original @samp{.word}
7377 contains @code{sym1} minus the address of the long-jump to
7380 If there were several occurrences of @samp{.word sym1-sym2} before the
7381 secondary jump table, all of them are adjusted. If there was a
7382 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7383 long-jump to @code{sym4} is included in the secondary jump table,
7384 and the @code{.word} directives are adjusted to contain @code{sym3}
7385 minus the address of the long-jump to @code{sym4}; and so on, for as many
7386 entries in the original jump table as necessary.
7389 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7390 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7391 assembly language programmers.
7394 @c end DIFF-TBL-KLUGE
7396 @ifclear no-space-dir
7398 @section @code{.zero @var{size}}
7400 @cindex @code{zero} directive
7401 @cindex filling memory with zero bytes
7402 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7403 expression. This directive is actually an alias for the @samp{.skip} directive
7404 so it can take an optional second argument of the value to store in the bytes
7405 instead of zero. Using @samp{.zero} in this way would be confusing however.
7410 @section @code{.2byte @var{expression} [, @var{expression}]*}
7411 @cindex @code{2byte} directive
7412 @cindex two-byte integer
7413 @cindex integer, 2-byte
7415 This directive expects zero or more expressions, separated by commas. If there
7416 are no expressions then the directive does nothing. Otherwise each expression
7417 is evaluated in turn and placed in the next two bytes of the current output
7418 section, using the endian model of the target. If an expression will not fit
7419 in two bytes, a warning message is displayed and the least significant two
7420 bytes of the expression's value are used. If an expression cannot be evaluated
7421 at assembly time then relocations will be generated in order to compute the
7424 This directive does not apply any alignment before or after inserting the
7425 values. As a result of this, if relocations are generated, they may be
7426 different from those used for inserting values with a guaranteed alignment.
7428 This directive is only available for ELF targets,
7431 @section @code{.4byte @var{expression} [, @var{expression}]*}
7432 @cindex @code{4byte} directive
7433 @cindex four-byte integer
7434 @cindex integer, 4-byte
7436 Like the @option{.2byte} directive, except that it inserts unaligned, four byte
7437 long values into the output.
7440 @section @code{.8byte @var{expression} [, @var{expression}]*}
7441 @cindex @code{8byte} directive
7442 @cindex eight-byte integer
7443 @cindex integer, 8-byte
7445 Like the @option{.2byte} directive, except that it inserts unaligned, eight
7446 byte long bignum values into the output.
7451 @section Deprecated Directives
7453 @cindex deprecated directives
7454 @cindex obsolescent directives
7455 One day these directives won't work.
7456 They are included for compatibility with older assemblers.
7463 @node Object Attributes
7464 @chapter Object Attributes
7465 @cindex object attributes
7467 @command{@value{AS}} assembles source files written for a specific architecture
7468 into object files for that architecture. But not all object files are alike.
7469 Many architectures support incompatible variations. For instance, floating
7470 point arguments might be passed in floating point registers if the object file
7471 requires hardware floating point support---or floating point arguments might be
7472 passed in integer registers if the object file supports processors with no
7473 hardware floating point unit. Or, if two objects are built for different
7474 generations of the same architecture, the combination may require the
7475 newer generation at run-time.
7477 This information is useful during and after linking. At link time,
7478 @command{@value{LD}} can warn about incompatible object files. After link
7479 time, tools like @command{gdb} can use it to process the linked file
7482 Compatibility information is recorded as a series of object attributes. Each
7483 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7484 string, and indicates who sets the meaning of the tag. The tag is an integer,
7485 and indicates what property the attribute describes. The value may be a string
7486 or an integer, and indicates how the property affects this object. Missing
7487 attributes are the same as attributes with a zero value or empty string value.
7489 Object attributes were developed as part of the ABI for the ARM Architecture.
7490 The file format is documented in @cite{ELF for the ARM Architecture}.
7493 * GNU Object Attributes:: @sc{gnu} Object Attributes
7494 * Defining New Object Attributes:: Defining New Object Attributes
7497 @node GNU Object Attributes
7498 @section @sc{gnu} Object Attributes
7500 The @code{.gnu_attribute} directive records an object attribute
7501 with vendor @samp{gnu}.
7503 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7504 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7505 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7506 2} is set for architecture-independent attributes and clear for
7507 architecture-dependent ones.
7509 @subsection Common @sc{gnu} attributes
7511 These attributes are valid on all architectures.
7514 @item Tag_compatibility (32)
7515 The compatibility attribute takes an integer flag value and a vendor name. If
7516 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7517 then the file is only compatible with the named toolchain. If it is greater
7518 than 1, the file can only be processed by other toolchains under some private
7519 arrangement indicated by the flag value and the vendor name.
7522 @subsection MIPS Attributes
7525 @item Tag_GNU_MIPS_ABI_FP (4)
7526 The floating-point ABI used by this object file. The value will be:
7530 0 for files not affected by the floating-point ABI.
7532 1 for files using the hardware floating-point ABI with a standard
7533 double-precision FPU.
7535 2 for files using the hardware floating-point ABI with a single-precision FPU.
7537 3 for files using the software floating-point ABI.
7539 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7540 floating-point registers, 32-bit general-purpose registers and increased the
7541 number of callee-saved floating-point registers.
7543 5 for files using the hardware floating-point ABI with a double-precision FPU
7544 with either 32-bit or 64-bit floating-point registers and 32-bit
7545 general-purpose registers.
7547 6 for files using the hardware floating-point ABI with 64-bit floating-point
7548 registers and 32-bit general-purpose registers.
7550 7 for files using the hardware floating-point ABI with 64-bit floating-point
7551 registers, 32-bit general-purpose registers and a rule that forbids the
7552 direct use of odd-numbered single-precision floating-point registers.
7556 @subsection PowerPC Attributes
7559 @item Tag_GNU_Power_ABI_FP (4)
7560 The floating-point ABI used by this object file. The value will be:
7564 0 for files not affected by the floating-point ABI.
7566 1 for files using double-precision hardware floating-point ABI.
7568 2 for files using the software floating-point ABI.
7570 3 for files using single-precision hardware floating-point ABI.
7573 @item Tag_GNU_Power_ABI_Vector (8)
7574 The vector ABI used by this object file. The value will be:
7578 0 for files not affected by the vector ABI.
7580 1 for files using general purpose registers to pass vectors.
7582 2 for files using AltiVec registers to pass vectors.
7584 3 for files using SPE registers to pass vectors.
7588 @subsection IBM z Systems Attributes
7591 @item Tag_GNU_S390_ABI_Vector (8)
7592 The vector ABI used by this object file. The value will be:
7596 0 for files not affected by the vector ABI.
7598 1 for files using software vector ABI.
7600 2 for files using hardware vector ABI.
7604 @subsection MSP430 Attributes
7607 @item Tag_GNU_MSP430_Data_Region (4)
7608 The data region used by this object file. The value will be:
7612 0 for files not using the large memory model.
7614 1 for files which have been compiled with the condition that all
7615 data is in the lower memory region, i.e. below address 0x10000.
7617 2 for files which allow data to be placed in the full 20-bit memory range.
7621 @node Defining New Object Attributes
7622 @section Defining New Object Attributes
7624 If you want to define a new @sc{gnu} object attribute, here are the places you
7625 will need to modify. New attributes should be discussed on the @samp{binutils}
7630 This manual, which is the official register of attributes.
7632 The header for your architecture @file{include/elf}, to define the tag.
7634 The @file{bfd} support file for your architecture, to merge the attribute
7635 and issue any appropriate link warnings.
7637 Test cases in @file{ld/testsuite} for merging and link warnings.
7639 @file{binutils/readelf.c} to display your attribute.
7641 GCC, if you want the compiler to mark the attribute automatically.
7647 @node Machine Dependencies
7648 @chapter Machine Dependent Features
7650 @cindex machine dependencies
7651 The machine instruction sets are (almost by definition) different on
7652 each machine where @command{@value{AS}} runs. Floating point representations
7653 vary as well, and @command{@value{AS}} often supports a few additional
7654 directives or command-line options for compatibility with other
7655 assemblers on a particular platform. Finally, some versions of
7656 @command{@value{AS}} support special pseudo-instructions for branch
7659 This chapter discusses most of these differences, though it does not
7660 include details on any machine's instruction set. For details on that
7661 subject, see the hardware manufacturer's manual.
7665 * AArch64-Dependent:: AArch64 Dependent Features
7668 * Alpha-Dependent:: Alpha Dependent Features
7671 * ARC-Dependent:: ARC Dependent Features
7674 * ARM-Dependent:: ARM Dependent Features
7677 * AVR-Dependent:: AVR Dependent Features
7680 * Blackfin-Dependent:: Blackfin Dependent Features
7683 * BPF-Dependent:: BPF Dependent Features
7686 * CR16-Dependent:: CR16 Dependent Features
7689 * CRIS-Dependent:: CRIS Dependent Features
7692 * C-SKY-Dependent:: C-SKY Dependent Features
7695 * D10V-Dependent:: D10V Dependent Features
7698 * D30V-Dependent:: D30V Dependent Features
7701 * Epiphany-Dependent:: EPIPHANY Dependent Features
7704 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7707 * HPPA-Dependent:: HPPA Dependent Features
7710 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7713 * IA-64-Dependent:: Intel IA-64 Dependent Features
7716 * IP2K-Dependent:: IP2K Dependent Features
7719 * LM32-Dependent:: LM32 Dependent Features
7722 * M32C-Dependent:: M32C Dependent Features
7725 * M32R-Dependent:: M32R Dependent Features
7728 * M68K-Dependent:: M680x0 Dependent Features
7731 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7734 * S12Z-Dependent:: S12Z Dependent Features
7737 * Meta-Dependent :: Meta Dependent Features
7740 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7743 * MIPS-Dependent:: MIPS Dependent Features
7746 * MMIX-Dependent:: MMIX Dependent Features
7749 * MSP430-Dependent:: MSP430 Dependent Features
7752 * NDS32-Dependent:: Andes NDS32 Dependent Features
7755 * NiosII-Dependent:: Altera Nios II Dependent Features
7758 * NS32K-Dependent:: NS32K Dependent Features
7761 * OpenRISC-Dependent:: OpenRISC 1000 Features
7764 * PDP-11-Dependent:: PDP-11 Dependent Features
7767 * PJ-Dependent:: picoJava Dependent Features
7770 * PPC-Dependent:: PowerPC Dependent Features
7773 * PRU-Dependent:: PRU Dependent Features
7776 * RISC-V-Dependent:: RISC-V Dependent Features
7779 * RL78-Dependent:: RL78 Dependent Features
7782 * RX-Dependent:: RX Dependent Features
7785 * S/390-Dependent:: IBM S/390 Dependent Features
7788 * SCORE-Dependent:: SCORE Dependent Features
7791 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7794 * Sparc-Dependent:: SPARC Dependent Features
7797 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7800 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7803 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7806 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7809 * V850-Dependent:: V850 Dependent Features
7812 * Vax-Dependent:: VAX Dependent Features
7815 * Visium-Dependent:: Visium Dependent Features
7818 * WebAssembly-Dependent:: WebAssembly Dependent Features
7821 * XGATE-Dependent:: XGATE Dependent Features
7824 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7827 * Xtensa-Dependent:: Xtensa Dependent Features
7830 * Z80-Dependent:: Z80 Dependent Features
7833 * Z8000-Dependent:: Z8000 Dependent Features
7840 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7841 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7842 @c peculiarity: to preserve cross-references, there must be a node called
7843 @c "Machine Dependencies". Hence the conditional nodenames in each
7844 @c major node below. Node defaulting in makeinfo requires adjacency of
7845 @c node and sectioning commands; hence the repetition of @chapter BLAH
7846 @c in both conditional blocks.
7849 @include c-aarch64.texi
7853 @include c-alpha.texi
7869 @include c-bfin.texi
7877 @include c-cr16.texi
7881 @include c-cris.texi
7885 @include c-csky.texi
7890 @node Machine Dependencies
7891 @chapter Machine Dependent Features
7893 The machine instruction sets are different on each Renesas chip family,
7894 and there are also some syntax differences among the families. This
7895 chapter describes the specific @command{@value{AS}} features for each
7899 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7900 * SH-Dependent:: Renesas SH Dependent Features
7907 @include c-d10v.texi
7911 @include c-d30v.texi
7915 @include c-epiphany.texi
7919 @include c-h8300.texi
7923 @include c-hppa.texi
7927 @include c-i386.texi
7931 @include c-ia64.texi
7935 @include c-ip2k.texi
7939 @include c-lm32.texi
7943 @include c-m32c.texi
7947 @include c-m32r.texi
7951 @include c-m68k.texi
7955 @include c-m68hc11.texi
7959 @include c-s12z.texi
7963 @include c-metag.texi
7967 @include c-microblaze.texi
7971 @include c-mips.texi
7975 @include c-mmix.texi
7979 @include c-msp430.texi
7983 @include c-nds32.texi
7987 @include c-nios2.texi
7991 @include c-ns32k.texi
7995 @include c-or1k.texi
7999 @include c-pdp11.texi
8015 @include c-riscv.texi
8019 @include c-rl78.texi
8027 @include c-s390.texi
8031 @include c-score.texi
8039 @include c-sparc.texi
8043 @include c-tic54x.texi
8047 @include c-tic6x.texi
8051 @include c-tilegx.texi
8055 @include c-tilepro.texi
8059 @include c-v850.texi
8067 @include c-visium.texi
8071 @include c-wasm32.texi
8075 @include c-xgate.texi
8079 @include c-xstormy16.texi
8083 @include c-xtensa.texi
8095 @c reverse effect of @down at top of generic Machine-Dep chapter
8099 @node Reporting Bugs
8100 @chapter Reporting Bugs
8101 @cindex bugs in assembler
8102 @cindex reporting bugs in assembler
8104 Your bug reports play an essential role in making @command{@value{AS}} reliable.
8106 Reporting a bug may help you by bringing a solution to your problem, or it may
8107 not. But in any case the principal function of a bug report is to help the
8108 entire community by making the next version of @command{@value{AS}} work better.
8109 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
8111 In order for a bug report to serve its purpose, you must include the
8112 information that enables us to fix the bug.
8115 * Bug Criteria:: Have you found a bug?
8116 * Bug Reporting:: How to report bugs
8120 @section Have You Found a Bug?
8121 @cindex bug criteria
8123 If you are not sure whether you have found a bug, here are some guidelines:
8126 @cindex fatal signal
8127 @cindex assembler crash
8128 @cindex crash of assembler
8130 If the assembler gets a fatal signal, for any input whatever, that is a
8131 @command{@value{AS}} bug. Reliable assemblers never crash.
8133 @cindex error on valid input
8135 If @command{@value{AS}} produces an error message for valid input, that is a bug.
8137 @cindex invalid input
8139 If @command{@value{AS}} does not produce an error message for invalid input, that
8140 is a bug. However, you should note that your idea of ``invalid input'' might
8141 be our idea of ``an extension'' or ``support for traditional practice''.
8144 If you are an experienced user of assemblers, your suggestions for improvement
8145 of @command{@value{AS}} are welcome in any case.
8149 @section How to Report Bugs
8151 @cindex assembler bugs, reporting
8153 A number of companies and individuals offer support for @sc{gnu} products. If
8154 you obtained @command{@value{AS}} from a support organization, we recommend you
8155 contact that organization first.
8157 You can find contact information for many support companies and
8158 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
8162 In any event, we also recommend that you send bug reports for @command{@value{AS}}
8166 The fundamental principle of reporting bugs usefully is this:
8167 @strong{report all the facts}. If you are not sure whether to state a
8168 fact or leave it out, state it!
8170 Often people omit facts because they think they know what causes the problem
8171 and assume that some details do not matter. Thus, you might assume that the
8172 name of a symbol you use in an example does not matter. Well, probably it does
8173 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
8174 happens to fetch from the location where that name is stored in memory;
8175 perhaps, if the name were different, the contents of that location would fool
8176 the assembler into doing the right thing despite the bug. Play it safe and
8177 give a specific, complete example. That is the easiest thing for you to do,
8178 and the most helpful.
8180 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
8181 it is new to us. Therefore, always write your bug reports on the assumption
8182 that the bug has not been reported previously.
8184 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8185 bell?'' This cannot help us fix a bug, so it is basically useless. We
8186 respond by asking for enough details to enable us to investigate.
8187 You might as well expedite matters by sending them to begin with.
8189 To enable us to fix the bug, you should include all these things:
8193 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8194 it with the @samp{--version} argument.
8196 Without this, we will not know whether there is any point in looking for
8197 the bug in the current version of @command{@value{AS}}.
8200 Any patches you may have applied to the @command{@value{AS}} source.
8203 The type of machine you are using, and the operating system name and
8207 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8211 The command arguments you gave the assembler to assemble your example and
8212 observe the bug. To guarantee you will not omit something important, list them
8213 all. A copy of the Makefile (or the output from make) is sufficient.
8215 If we were to try to guess the arguments, we would probably guess wrong
8216 and then we might not encounter the bug.
8219 A complete input file that will reproduce the bug. If the bug is observed when
8220 the assembler is invoked via a compiler, send the assembler source, not the
8221 high level language source. Most compilers will produce the assembler source
8222 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8223 the options @samp{-v --save-temps}; this will save the assembler source in a
8224 file with an extension of @file{.s}, and also show you exactly how
8225 @command{@value{AS}} is being run.
8228 A description of what behavior you observe that you believe is
8229 incorrect. For example, ``It gets a fatal signal.''
8231 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8232 will certainly notice it. But if the bug is incorrect output, we might not
8233 notice unless it is glaringly wrong. You might as well not give us a chance to
8236 Even if the problem you experience is a fatal signal, you should still say so
8237 explicitly. Suppose something strange is going on, such as, your copy of
8238 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8239 library on your system. (This has happened!) Your copy might crash and ours
8240 would not. If you told us to expect a crash, then when ours fails to crash, we
8241 would know that the bug was not happening for us. If you had not told us to
8242 expect a crash, then we would not be able to draw any conclusion from our
8246 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8247 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8248 option. Always send diffs from the old file to the new file. If you even
8249 discuss something in the @command{@value{AS}} source, refer to it by context, not
8252 The line numbers in our development sources will not match those in your
8253 sources. Your line numbers would convey no useful information to us.
8256 Here are some things that are not necessary:
8260 A description of the envelope of the bug.
8262 Often people who encounter a bug spend a lot of time investigating
8263 which changes to the input file will make the bug go away and which
8264 changes will not affect it.
8266 This is often time consuming and not very useful, because the way we
8267 will find the bug is by running a single example under the debugger
8268 with breakpoints, not by pure deduction from a series of examples.
8269 We recommend that you save your time for something else.
8271 Of course, if you can find a simpler example to report @emph{instead}
8272 of the original one, that is a convenience for us. Errors in the
8273 output will be easier to spot, running under the debugger will take
8274 less time, and so on.
8276 However, simplification is not vital; if you do not want to do this,
8277 report the bug anyway and send us the entire test case you used.
8280 A patch for the bug.
8282 A patch for the bug does help us if it is a good one. But do not omit
8283 the necessary information, such as the test case, on the assumption that
8284 a patch is all we need. We might see problems with your patch and decide
8285 to fix the problem another way, or we might not understand it at all.
8287 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8288 construct an example that will make the program follow a certain path through
8289 the code. If you do not send us the example, we will not be able to construct
8290 one, so we will not be able to verify that the bug is fixed.
8292 And if we cannot understand what bug you are trying to fix, or why your
8293 patch should be an improvement, we will not install it. A test case will
8294 help us to understand.
8297 A guess about what the bug is or what it depends on.
8299 Such guesses are usually wrong. Even we cannot guess right about such
8300 things without first using the debugger to find the facts.
8303 @node Acknowledgements
8304 @chapter Acknowledgements
8306 If you have contributed to GAS and your name isn't listed here,
8307 it is not meant as a slight. We just don't know about it. Send mail to the
8308 maintainer, and we'll correct the situation. Currently
8310 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8312 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8315 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8316 information and the 68k series machines, most of the preprocessing pass, and
8317 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8319 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8320 many bug fixes, including merging support for several processors, breaking GAS
8321 up to handle multiple object file format back ends (including heavy rewrite,
8322 testing, an integration of the coff and b.out back ends), adding configuration
8323 including heavy testing and verification of cross assemblers and file splits
8324 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8325 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8326 port (including considerable amounts of reverse engineering), a SPARC opcode
8327 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8328 assertions and made them work, much other reorganization, cleanup, and lint.
8330 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8331 in format-specific I/O modules.
8333 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8334 has done much work with it since.
8336 The Intel 80386 machine description was written by Eliot Dresselhaus.
8338 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8340 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8341 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8343 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8344 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8345 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8346 support a.out format.
8348 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8349 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8350 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8351 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8354 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8355 simplified the configuration of which versions accept which directives. He
8356 updated the 68k machine description so that Motorola's opcodes always produced
8357 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8358 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8359 cross-compilation support, and one bug in relaxation that took a week and
8360 required the proverbial one-bit fix.
8362 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8363 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8364 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8365 PowerPC assembler, and made a few other minor patches.
8367 Steve Chamberlain made GAS able to generate listings.
8369 Hewlett-Packard contributed support for the HP9000/300.
8371 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8372 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8373 formats). This work was supported by both the Center for Software Science at
8374 the University of Utah and Cygnus Support.
8376 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8377 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8378 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8379 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8380 and some initial 64-bit support).
8382 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8384 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8385 support for openVMS/Alpha.
8387 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8390 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8391 Inc.@: added support for Xtensa processors.
8393 Several engineers at Cygnus Support have also provided many small bug fixes and
8394 configuration enhancements.
8396 Jon Beniston added support for the Lattice Mico32 architecture.
8398 Many others have contributed large or small bugfixes and enhancements. If
8399 you have contributed significant work and are not mentioned on this list, and
8400 want to be, let us know. Some of the history has been lost; we are not
8401 intentionally leaving anyone out.
8403 @node GNU Free Documentation License
8404 @appendix GNU Free Documentation License
8408 @unnumbered AS Index