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{-march=@var{CPU}@var{[-EXT]}@var{[+EXT]}}]
634 [@b{-local-prefix=}@var{PREFIX}]
637 [@b{-fp-s=}@var{FORMAT}]
638 [@b{-fp-d=}@var{FORMAT}]
642 @c Z8000 has no machine-dependent assembler options
651 @include at-file.texi
654 Turn on listings, in any of a variety of ways:
658 omit false conditionals
661 omit debugging directives
664 include general information, like @value{AS} version and options passed
667 include high-level source
673 include macro expansions
676 omit forms processing
682 set the name of the listing file
685 You may combine these options; for example, use @samp{-aln} for assembly
686 listing without forms processing. The @samp{=file} option, if used, must be
687 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
690 Begin in alternate macro mode.
692 @xref{Altmacro,,@code{.altmacro}}.
695 @item --compress-debug-sections
696 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
697 ELF ABI. The resulting object file may not be compatible with older
698 linkers and object file utilities. Note if compression would make a
699 given section @emph{larger} then it is not compressed.
702 @cindex @samp{--compress-debug-sections=} option
703 @item --compress-debug-sections=none
704 @itemx --compress-debug-sections=zlib
705 @itemx --compress-debug-sections=zlib-gnu
706 @itemx --compress-debug-sections=zlib-gabi
707 These options control how DWARF debug sections are compressed.
708 @option{--compress-debug-sections=none} is equivalent to
709 @option{--nocompress-debug-sections}.
710 @option{--compress-debug-sections=zlib} and
711 @option{--compress-debug-sections=zlib-gabi} are equivalent to
712 @option{--compress-debug-sections}.
713 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
714 sections using zlib. The debug sections are renamed to begin with
715 @samp{.zdebug}. Note if compression would make a given section
716 @emph{larger} then it is not compressed nor renamed.
720 @item --nocompress-debug-sections
721 Do not compress DWARF debug sections. This is usually the default for all
722 targets except the x86/x86_64, but a configure time option can be used to
726 Ignored. This option is accepted for script compatibility with calls to
729 @item --debug-prefix-map @var{old}=@var{new}
730 When assembling files in directory @file{@var{old}}, record debugging
731 information describing them as in @file{@var{new}} instead.
733 @item --defsym @var{sym}=@var{value}
734 Define the symbol @var{sym} to be @var{value} before assembling the input file.
735 @var{value} must be an integer constant. As in C, a leading @samp{0x}
736 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
737 value. The value of the symbol can be overridden inside a source file via the
738 use of a @code{.set} pseudo-op.
741 ``fast''---skip whitespace and comment preprocessing (assume source is
746 Generate debugging information for each assembler source line using whichever
747 debug format is preferred by the target. This currently means either STABS,
751 Generate stabs debugging information for each assembler line. This
752 may help debugging assembler code, if the debugger can handle it.
755 Generate stabs debugging information for each assembler line, with GNU
756 extensions that probably only gdb can handle, and that could make other
757 debuggers crash or refuse to read your program. This
758 may help debugging assembler code. Currently the only GNU extension is
759 the location of the current working directory at assembling time.
762 Generate DWARF2 debugging information for each assembler line. This
763 may help debugging assembler code, if the debugger can handle it. Note---this
764 option is only supported by some targets, not all of them.
766 @item --gdwarf-sections
767 Instead of creating a .debug_line section, create a series of
768 .debug_line.@var{foo} sections where @var{foo} is the name of the
769 corresponding code section. For example a code section called @var{.text.func}
770 will have its dwarf line number information placed into a section called
771 @var{.debug_line.text.func}. If the code section is just called @var{.text}
772 then debug line section will still be called just @var{.debug_line} without any
775 @item --gdwarf-cie-version=@var{version}
776 Control which version of DWARF Common Information Entries (CIEs) are produced.
777 When this flag is not specificed the default is version 1, though some targets
778 can modify this default. Other possible values for @var{version} are 3 or 4.
781 @item --size-check=error
782 @itemx --size-check=warning
783 Issue an error or warning for invalid ELF .size directive.
785 @item --elf-stt-common=no
786 @itemx --elf-stt-common=yes
787 These options control whether the ELF assembler should generate common
788 symbols with the @code{STT_COMMON} type. The default can be controlled
789 by a configure option @option{--enable-elf-stt-common}.
791 @item --generate-missing-build-notes=yes
792 @itemx --generate-missing-build-notes=no
793 These options control whether the ELF assembler should generate GNU Build
794 attribute notes if none are present in the input sources.
795 The default can be controlled by the @option{--enable-generate-build-notes}
801 Print a summary of the command-line options and exit.
804 Print a summary of all target specific options and exit.
807 Add directory @var{dir} to the search list for @code{.include} directives.
810 Don't warn about signed overflow.
813 @ifclear DIFF-TBL-KLUGE
814 This option is accepted but has no effect on the @value{TARGET} family.
816 @ifset DIFF-TBL-KLUGE
817 Issue warnings when difference tables altered for long displacements.
822 Keep (in the symbol table) local symbols. These symbols start with
823 system-specific local label prefixes, typically @samp{.L} for ELF systems
824 or @samp{L} for traditional a.out systems.
829 @item --listing-lhs-width=@var{number}
830 Set the maximum width, in words, of the output data column for an assembler
831 listing to @var{number}.
833 @item --listing-lhs-width2=@var{number}
834 Set the maximum width, in words, of the output data column for continuation
835 lines in an assembler listing to @var{number}.
837 @item --listing-rhs-width=@var{number}
838 Set the maximum width of an input source line, as displayed in a listing, to
841 @item --listing-cont-lines=@var{number}
842 Set the maximum number of lines printed in a listing for a single line of input
845 @item --no-pad-sections
846 Stop the assembler for padding the ends of output sections to the alignment
847 of that section. The default is to pad the sections, but this can waste space
848 which might be needed on targets which have tight memory constraints.
850 @item -o @var{objfile}
851 Name the object-file output from @command{@value{AS}} @var{objfile}.
854 Fold the data section into the text section.
856 @item --hash-size=@var{number}
857 Set the default size of GAS's hash tables to a prime number close to
858 @var{number}. Increasing this value can reduce the length of time it takes the
859 assembler to perform its tasks, at the expense of increasing the assembler's
860 memory requirements. Similarly reducing this value can reduce the memory
861 requirements at the expense of speed.
863 @item --reduce-memory-overheads
864 This option reduces GAS's memory requirements, at the expense of making the
865 assembly processes slower. Currently this switch is a synonym for
866 @samp{--hash-size=4051}, but in the future it may have other effects as well.
869 @item --sectname-subst
870 Honor substitution sequences in section names.
872 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
877 Print the maximum space (in bytes) and total time (in seconds) used by
880 @item --strip-local-absolute
881 Remove local absolute symbols from the outgoing symbol table.
885 Print the @command{as} version.
888 Print the @command{as} version and exit.
892 Suppress warning messages.
894 @item --fatal-warnings
895 Treat warnings as errors.
898 Don't suppress warning messages or treat them as errors.
907 Generate an object file even after errors.
909 @item -- | @var{files} @dots{}
910 Standard input, or source files to assemble.
918 @xref{AArch64 Options}, for the options available when @value{AS} is configured
919 for the 64-bit mode of the ARM Architecture (AArch64).
924 The following options are available when @value{AS} is configured for the
925 64-bit mode of the ARM Architecture (AArch64).
928 @include c-aarch64.texi
929 @c ended inside the included file
937 @xref{Alpha Options}, for the options available when @value{AS} is configured
938 for an Alpha processor.
943 The following options are available when @value{AS} is configured for an Alpha
947 @include c-alpha.texi
948 @c ended inside the included file
955 The following options are available when @value{AS} is configured for an ARC
959 @item -mcpu=@var{cpu}
960 This option selects the core processor variant.
962 Select either big-endian (-EB) or little-endian (-EL) output.
964 Enable Code Density extenssion instructions.
969 The following options are available when @value{AS} is configured for the ARM
973 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
974 Specify which ARM processor variant is the target.
975 @item -march=@var{architecture}[+@var{extension}@dots{}]
976 Specify which ARM architecture variant is used by the target.
977 @item -mfpu=@var{floating-point-format}
978 Select which Floating Point architecture is the target.
979 @item -mfloat-abi=@var{abi}
980 Select which floating point ABI is in use.
982 Enable Thumb only instruction decoding.
983 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
984 Select which procedure calling convention is in use.
986 Select either big-endian (-EB) or little-endian (-EL) output.
987 @item -mthumb-interwork
988 Specify that the code has been generated with interworking between Thumb and
991 Turns on CodeComposer Studio assembly syntax compatibility mode.
993 Specify that PIC code has been generated.
1001 @xref{Blackfin Options}, for the options available when @value{AS} is
1002 configured for the Blackfin processor family.
1006 @c man begin OPTIONS
1007 The following options are available when @value{AS} is configured for
1008 the Blackfin processor family.
1010 @c man begin INCLUDE
1011 @include c-bfin.texi
1012 @c ended inside the included file
1020 @xref{BPF Options}, for the options available when @value{AS} is
1021 configured for the Linux kernel BPF processor family.
1025 @c man begin OPTIONS
1026 The following options are available when @value{AS} is configured for
1027 the Linux kernel BPF processor family.
1029 @c man begin INCLUDE
1031 @c ended inside the included file
1036 @c man begin OPTIONS
1038 See the info pages for documentation of the CRIS-specific options.
1044 @xref{C-SKY Options}, for the options available when @value{AS} is
1045 configured for the C-SKY processor family.
1049 @c man begin OPTIONS
1050 The following options are available when @value{AS} is configured for
1051 the C-SKY processor family.
1053 @c man begin INCLUDE
1054 @include c-csky.texi
1055 @c ended inside the included file
1061 The following options are available when @value{AS} is configured for
1064 @cindex D10V optimization
1065 @cindex optimization, D10V
1067 Optimize output by parallelizing instructions.
1072 The following options are available when @value{AS} is configured for a D30V
1075 @cindex D30V optimization
1076 @cindex optimization, D30V
1078 Optimize output by parallelizing instructions.
1082 Warn when nops are generated.
1084 @cindex D30V nops after 32-bit multiply
1086 Warn when a nop after a 32-bit multiply instruction is generated.
1092 The following options are available when @value{AS} is configured for the
1093 Adapteva EPIPHANY series.
1096 @xref{Epiphany Options}, for the options available when @value{AS} is
1097 configured for an Epiphany processor.
1101 @c man begin OPTIONS
1102 The following options are available when @value{AS} is configured for
1103 an Epiphany processor.
1105 @c man begin INCLUDE
1106 @include c-epiphany.texi
1107 @c ended inside the included file
1115 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1116 for an H8/300 processor.
1120 @c man begin OPTIONS
1121 The following options are available when @value{AS} is configured for an H8/300
1124 @c man begin INCLUDE
1125 @include c-h8300.texi
1126 @c ended inside the included file
1134 @xref{i386-Options}, for the options available when @value{AS} is
1135 configured for an i386 processor.
1139 @c man begin OPTIONS
1140 The following options are available when @value{AS} is configured for
1143 @c man begin INCLUDE
1144 @include c-i386.texi
1145 @c ended inside the included file
1150 @c man begin OPTIONS
1152 The following options are available when @value{AS} is configured for the
1158 Specifies that the extended IP2022 instructions are allowed.
1161 Restores the default behaviour, which restricts the permitted instructions to
1162 just the basic IP2022 ones.
1168 The following options are available when @value{AS} is configured for the
1169 Renesas M32C and M16C processors.
1174 Assemble M32C instructions.
1177 Assemble M16C instructions (the default).
1180 Enable support for link-time relaxations.
1183 Support H'00 style hex constants in addition to 0x00 style.
1189 The following options are available when @value{AS} is configured for the
1190 Renesas M32R (formerly Mitsubishi M32R) series.
1195 Specify which processor in the M32R family is the target. The default
1196 is normally the M32R, but this option changes it to the M32RX.
1198 @item --warn-explicit-parallel-conflicts or --Wp
1199 Produce warning messages when questionable parallel constructs are
1202 @item --no-warn-explicit-parallel-conflicts or --Wnp
1203 Do not produce warning messages when questionable parallel constructs are
1210 The following options are available when @value{AS} is configured for the
1211 Motorola 68000 series.
1216 Shorten references to undefined symbols, to one word instead of two.
1218 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1219 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1220 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1221 Specify what processor in the 68000 family is the target. The default
1222 is normally the 68020, but this can be changed at configuration time.
1224 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1225 The target machine does (or does not) have a floating-point coprocessor.
1226 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1227 the basic 68000 is not compatible with the 68881, a combination of the
1228 two can be specified, since it's possible to do emulation of the
1229 coprocessor instructions with the main processor.
1231 @item -m68851 | -mno-68851
1232 The target machine does (or does not) have a memory-management
1233 unit coprocessor. The default is to assume an MMU for 68020 and up.
1241 @xref{Nios II Options}, for the options available when @value{AS} is configured
1242 for an Altera Nios II processor.
1246 @c man begin OPTIONS
1247 The following options are available when @value{AS} is configured for an
1248 Altera Nios II processor.
1250 @c man begin INCLUDE
1251 @include c-nios2.texi
1252 @c ended inside the included file
1258 For details about the PDP-11 machine dependent features options,
1259 see @ref{PDP-11-Options}.
1262 @item -mpic | -mno-pic
1263 Generate position-independent (or position-dependent) code. The
1264 default is @option{-mpic}.
1267 @itemx -mall-extensions
1268 Enable all instruction set extensions. This is the default.
1270 @item -mno-extensions
1271 Disable all instruction set extensions.
1273 @item -m@var{extension} | -mno-@var{extension}
1274 Enable (or disable) a particular instruction set extension.
1277 Enable the instruction set extensions supported by a particular CPU, and
1278 disable all other extensions.
1280 @item -m@var{machine}
1281 Enable the instruction set extensions supported by a particular machine
1282 model, and disable all other extensions.
1288 The following options are available when @value{AS} is configured for
1289 a picoJava processor.
1293 @cindex PJ endianness
1294 @cindex endianness, PJ
1295 @cindex big endian output, PJ
1297 Generate ``big endian'' format output.
1299 @cindex little endian output, PJ
1301 Generate ``little endian'' format output.
1309 @xref{PRU Options}, for the options available when @value{AS} is configured
1310 for a PRU processor.
1314 @c man begin OPTIONS
1315 The following options are available when @value{AS} is configured for a
1318 @c man begin INCLUDE
1320 @c ended inside the included file
1325 The following options are available when @value{AS} is configured for the
1326 Motorola 68HC11 or 68HC12 series.
1330 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1331 Specify what processor is the target. The default is
1332 defined by the configuration option when building the assembler.
1334 @item --xgate-ramoffset
1335 Instruct the linker to offset RAM addresses from S12X address space into
1336 XGATE address space.
1339 Specify to use the 16-bit integer ABI.
1342 Specify to use the 32-bit integer ABI.
1344 @item -mshort-double
1345 Specify to use the 32-bit double ABI.
1348 Specify to use the 64-bit double ABI.
1350 @item --force-long-branches
1351 Relative branches are turned into absolute ones. This concerns
1352 conditional branches, unconditional branches and branches to a
1355 @item -S | --short-branches
1356 Do not turn relative branches into absolute ones
1357 when the offset is out of range.
1359 @item --strict-direct-mode
1360 Do not turn the direct addressing mode into extended addressing mode
1361 when the instruction does not support direct addressing mode.
1363 @item --print-insn-syntax
1364 Print the syntax of instruction in case of error.
1366 @item --print-opcodes
1367 Print the list of instructions with syntax and then exit.
1369 @item --generate-example
1370 Print an example of instruction for each possible instruction and then exit.
1371 This option is only useful for testing @command{@value{AS}}.
1377 The following options are available when @command{@value{AS}} is configured
1378 for the SPARC architecture:
1381 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1382 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1383 Explicitly select a variant of the SPARC architecture.
1385 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1386 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1388 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1389 UltraSPARC extensions.
1391 @item -xarch=v8plus | -xarch=v8plusa
1392 For compatibility with the Solaris v9 assembler. These options are
1393 equivalent to -Av8plus and -Av8plusa, respectively.
1396 Warn when the assembler switches to another architecture.
1401 The following options are available when @value{AS} is configured for the 'c54x
1406 Enable extended addressing mode. All addresses and relocations will assume
1407 extended addressing (usually 23 bits).
1408 @item -mcpu=@var{CPU_VERSION}
1409 Sets the CPU version being compiled for.
1410 @item -merrors-to-file @var{FILENAME}
1411 Redirect error output to a file, for broken systems which don't support such
1412 behaviour in the shell.
1417 @c man begin OPTIONS
1418 The following options are available when @value{AS} is configured for
1423 This option sets the largest size of an object that can be referenced
1424 implicitly with the @code{gp} register. It is only accepted for targets that
1425 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1427 @cindex MIPS endianness
1428 @cindex endianness, MIPS
1429 @cindex big endian output, MIPS
1431 Generate ``big endian'' format output.
1433 @cindex little endian output, MIPS
1435 Generate ``little endian'' format output.
1453 Generate code for a particular MIPS Instruction Set Architecture level.
1454 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1455 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1456 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1457 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1458 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1459 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1460 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1461 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1462 MIPS64 Release 6 ISA processors, respectively.
1464 @item -march=@var{cpu}
1465 Generate code for a particular MIPS CPU.
1467 @item -mtune=@var{cpu}
1468 Schedule and tune for a particular MIPS CPU.
1472 Cause nops to be inserted if the read of the destination register
1473 of an mfhi or mflo instruction occurs in the following two instructions.
1476 @itemx -mno-fix-rm7000
1477 Cause nops to be inserted if a dmult or dmultu instruction is
1478 followed by a load instruction.
1481 @itemx -mno-fix-r5900
1482 Do not attempt to schedule the preceding instruction into the delay slot
1483 of a branch instruction placed at the end of a short loop of six
1484 instructions or fewer and always schedule a @code{nop} instruction there
1485 instead. The short loop bug under certain conditions causes loops to
1486 execute only once or twice, due to a hardware bug in the R5900 chip.
1490 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1491 section instead of the standard ELF .stabs sections.
1495 Control generation of @code{.pdr} sections.
1499 The register sizes are normally inferred from the ISA and ABI, but these
1500 flags force a certain group of registers to be treated as 32 bits wide at
1501 all times. @samp{-mgp32} controls the size of general-purpose registers
1502 and @samp{-mfp32} controls the size of floating-point registers.
1506 The register sizes are normally inferred from the ISA and ABI, but these
1507 flags force a certain group of registers to be treated as 64 bits wide at
1508 all times. @samp{-mgp64} controls the size of general-purpose registers
1509 and @samp{-mfp64} controls the size of floating-point registers.
1512 The register sizes are normally inferred from the ISA and ABI, but using
1513 this flag in combination with @samp{-mabi=32} enables an ABI variant
1514 which will operate correctly with floating-point registers which are
1518 @itemx -mno-odd-spreg
1519 Enable use of floating-point operations on odd-numbered single-precision
1520 registers when supported by the ISA. @samp{-mfpxx} implies
1521 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1525 Generate code for the MIPS 16 processor. This is equivalent to putting
1526 @code{.module mips16} at the start of the assembly file. @samp{-no-mips16}
1527 turns off this option.
1530 @itemx -mno-mips16e2
1531 Enable the use of MIPS16e2 instructions in MIPS16 mode. This is equivalent
1532 to putting @code{.module mips16e2} at the start of the assembly file.
1533 @samp{-mno-mips16e2} turns off this option.
1536 @itemx -mno-micromips
1537 Generate code for the microMIPS processor. This is equivalent to putting
1538 @code{.module micromips} at the start of the assembly file.
1539 @samp{-mno-micromips} turns off this option. This is equivalent to putting
1540 @code{.module nomicromips} at the start of the assembly file.
1543 @itemx -mno-smartmips
1544 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1545 equivalent to putting @code{.module smartmips} at the start of the assembly
1546 file. @samp{-mno-smartmips} turns off this option.
1550 Generate code for the MIPS-3D Application Specific Extension.
1551 This tells the assembler to accept MIPS-3D instructions.
1552 @samp{-no-mips3d} turns off this option.
1556 Generate code for the MDMX Application Specific Extension.
1557 This tells the assembler to accept MDMX instructions.
1558 @samp{-no-mdmx} turns off this option.
1562 Generate code for the DSP Release 1 Application Specific Extension.
1563 This tells the assembler to accept DSP Release 1 instructions.
1564 @samp{-mno-dsp} turns off this option.
1568 Generate code for the DSP Release 2 Application Specific Extension.
1569 This option implies @samp{-mdsp}.
1570 This tells the assembler to accept DSP Release 2 instructions.
1571 @samp{-mno-dspr2} turns off this option.
1575 Generate code for the DSP Release 3 Application Specific Extension.
1576 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1577 This tells the assembler to accept DSP Release 3 instructions.
1578 @samp{-mno-dspr3} turns off this option.
1582 Generate code for the MIPS SIMD Architecture Extension.
1583 This tells the assembler to accept MSA instructions.
1584 @samp{-mno-msa} turns off this option.
1588 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1589 This tells the assembler to accept XPA instructions.
1590 @samp{-mno-xpa} turns off this option.
1594 Generate code for the MT Application Specific Extension.
1595 This tells the assembler to accept MT instructions.
1596 @samp{-mno-mt} turns off this option.
1600 Generate code for the MCU Application Specific Extension.
1601 This tells the assembler to accept MCU instructions.
1602 @samp{-mno-mcu} turns off this option.
1606 Generate code for the MIPS cyclic redundancy check (CRC) Application
1607 Specific Extension. This tells the assembler to accept CRC instructions.
1608 @samp{-mno-crc} turns off this option.
1612 Generate code for the Global INValidate (GINV) Application Specific
1613 Extension. This tells the assembler to accept GINV instructions.
1614 @samp{-mno-ginv} turns off this option.
1616 @item -mloongson-mmi
1617 @itemx -mno-loongson-mmi
1618 Generate code for the Loongson MultiMedia extensions Instructions (MMI)
1619 Application Specific Extension. This tells the assembler to accept MMI
1621 @samp{-mno-loongson-mmi} turns off this option.
1623 @item -mloongson-cam
1624 @itemx -mno-loongson-cam
1625 Generate code for the Loongson Content Address Memory (CAM) instructions.
1626 This tells the assembler to accept Loongson CAM instructions.
1627 @samp{-mno-loongson-cam} turns off this option.
1629 @item -mloongson-ext
1630 @itemx -mno-loongson-ext
1631 Generate code for the Loongson EXTensions (EXT) instructions.
1632 This tells the assembler to accept Loongson EXT instructions.
1633 @samp{-mno-loongson-ext} turns off this option.
1635 @item -mloongson-ext2
1636 @itemx -mno-loongson-ext2
1637 Generate code for the Loongson EXTensions R2 (EXT2) instructions.
1638 This option implies @samp{-mloongson-ext}.
1639 This tells the assembler to accept Loongson EXT2 instructions.
1640 @samp{-mno-loongson-ext2} turns off this option.
1644 Only use 32-bit instruction encodings when generating code for the
1645 microMIPS processor. This option inhibits the use of any 16-bit
1646 instructions. This is equivalent to putting @code{.set insn32} at
1647 the start of the assembly file. @samp{-mno-insn32} turns off this
1648 option. This is equivalent to putting @code{.set noinsn32} at the
1649 start of the assembly file. By default @samp{-mno-insn32} is
1650 selected, allowing all instructions to be used.
1652 @item --construct-floats
1653 @itemx --no-construct-floats
1654 The @samp{--no-construct-floats} option disables the construction of
1655 double width floating point constants by loading the two halves of the
1656 value into the two single width floating point registers that make up
1657 the double width register. By default @samp{--construct-floats} is
1658 selected, allowing construction of these floating point constants.
1660 @item --relax-branch
1661 @itemx --no-relax-branch
1662 The @samp{--relax-branch} option enables the relaxation of out-of-range
1663 branches. By default @samp{--no-relax-branch} is selected, causing any
1664 out-of-range branches to produce an error.
1666 @item -mignore-branch-isa
1667 @itemx -mno-ignore-branch-isa
1668 Ignore branch checks for invalid transitions between ISA modes. The
1669 semantics of branches does not provide for an ISA mode switch, so in
1670 most cases the ISA mode a branch has been encoded for has to be the
1671 same as the ISA mode of the branch's target label. Therefore GAS has
1672 checks implemented that verify in branch assembly that the two ISA
1673 modes match. @samp{-mignore-branch-isa} disables these checks. By
1674 default @samp{-mno-ignore-branch-isa} is selected, causing any invalid
1675 branch requiring a transition between ISA modes to produce an error.
1677 @item -mnan=@var{encoding}
1678 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1679 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1682 @item --emulation=@var{name}
1683 This option was formerly used to switch between ELF and ECOFF output
1684 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1685 removed in GAS 2.24, so the option now serves little purpose.
1686 It is retained for backwards compatibility.
1688 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1689 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1690 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1691 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1692 preferred options instead.
1695 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1702 Control how to deal with multiplication overflow and division by zero.
1703 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1704 (and only work for Instruction Set Architecture level 2 and higher);
1705 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1709 When this option is used, @command{@value{AS}} will issue a warning every
1710 time it generates a nop instruction from a macro.
1716 The following options are available when @value{AS} is configured for
1722 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1723 The command-line option @samp{-nojsri2bsr} can be used to disable it.
1727 Enable or disable the silicon filter behaviour. By default this is disabled.
1728 The default can be overridden by the @samp{-sifilter} command-line option.
1731 Alter jump instructions for long displacements.
1733 @item -mcpu=[210|340]
1734 Select the cpu type on the target hardware. This controls which instructions
1738 Assemble for a big endian target.
1741 Assemble for a little endian target.
1750 @xref{Meta Options}, for the options available when @value{AS} is configured
1751 for a Meta processor.
1755 @c man begin OPTIONS
1756 The following options are available when @value{AS} is configured for a
1759 @c man begin INCLUDE
1760 @include c-metag.texi
1761 @c ended inside the included file
1766 @c man begin OPTIONS
1768 See the info pages for documentation of the MMIX-specific options.
1774 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1775 for a NDS32 processor.
1777 @c ended inside the included file
1781 @c man begin OPTIONS
1782 The following options are available when @value{AS} is configured for a
1785 @c man begin INCLUDE
1786 @include c-nds32.texi
1787 @c ended inside the included file
1794 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1795 for a PowerPC processor.
1799 @c man begin OPTIONS
1800 The following options are available when @value{AS} is configured for a
1803 @c man begin INCLUDE
1805 @c ended inside the included file
1813 @xref{RISC-V-Options}, for the options available when @value{AS} is configured
1814 for a RISC-V processor.
1818 @c man begin OPTIONS
1819 The following options are available when @value{AS} is configured for a
1822 @c man begin INCLUDE
1823 @include c-riscv.texi
1824 @c ended inside the included file
1829 @c man begin OPTIONS
1831 See the info pages for documentation of the RX-specific options.
1835 The following options are available when @value{AS} is configured for the s390
1841 Select the word size, either 31/32 bits or 64 bits.
1844 Select the architecture mode, either the Enterprise System
1845 Architecture (esa) or the z/Architecture mode (zarch).
1846 @item -march=@var{processor}
1847 Specify which s390 processor variant is the target, @samp{g5} (or
1848 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1849 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1850 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1851 @samp{z13} (or @samp{arch11}), @samp{z14} (or @samp{arch12}), or @samp{z15}
1854 @itemx -mno-regnames
1855 Allow or disallow symbolic names for registers.
1856 @item -mwarn-areg-zero
1857 Warn whenever the operand for a base or index register has been specified
1858 but evaluates to zero.
1866 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1867 for a TMS320C6000 processor.
1871 @c man begin OPTIONS
1872 The following options are available when @value{AS} is configured for a
1873 TMS320C6000 processor.
1875 @c man begin INCLUDE
1876 @include c-tic6x.texi
1877 @c ended inside the included file
1885 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1886 for a TILE-Gx processor.
1890 @c man begin OPTIONS
1891 The following options are available when @value{AS} is configured for a TILE-Gx
1894 @c man begin INCLUDE
1895 @include c-tilegx.texi
1896 @c ended inside the included file
1904 @xref{Visium Options}, for the options available when @value{AS} is configured
1905 for a Visium processor.
1909 @c man begin OPTIONS
1910 The following option is available when @value{AS} is configured for a Visium
1913 @c man begin INCLUDE
1914 @include c-visium.texi
1915 @c ended inside the included file
1923 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1924 for an Xtensa processor.
1928 @c man begin OPTIONS
1929 The following options are available when @value{AS} is configured for an
1932 @c man begin INCLUDE
1933 @include c-xtensa.texi
1934 @c ended inside the included file
1942 @xref{Z80 Options}, for the options available when @value{AS} is configured
1943 for an Z80 processor.
1947 @c man begin OPTIONS
1948 The following options are available when @value{AS} is configured for an
1951 @c man begin INCLUDE
1953 @c ended inside the included file
1959 * Manual:: Structure of this Manual
1960 * GNU Assembler:: The GNU Assembler
1961 * Object Formats:: Object File Formats
1962 * Command Line:: Command Line
1963 * Input Files:: Input Files
1964 * Object:: Output (Object) File
1965 * Errors:: Error and Warning Messages
1969 @section Structure of this Manual
1971 @cindex manual, structure and purpose
1972 This manual is intended to describe what you need to know to use
1973 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1974 notation for symbols, constants, and expressions; the directives that
1975 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1978 We also cover special features in the @value{TARGET}
1979 configuration of @command{@value{AS}}, including assembler directives.
1982 This manual also describes some of the machine-dependent features of
1983 various flavors of the assembler.
1986 @cindex machine instructions (not covered)
1987 On the other hand, this manual is @emph{not} intended as an introduction
1988 to programming in assembly language---let alone programming in general!
1989 In a similar vein, we make no attempt to introduce the machine
1990 architecture; we do @emph{not} describe the instruction set, standard
1991 mnemonics, registers or addressing modes that are standard to a
1992 particular architecture.
1994 You may want to consult the manufacturer's
1995 machine architecture manual for this information.
1999 For information on the H8/300 machine instruction set, see @cite{H8/300
2000 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
2001 Programming Manual} (Renesas).
2004 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
2005 see @cite{SH-Microcomputer User's Manual} (Renesas) or
2006 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
2007 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
2010 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
2014 @c I think this is premature---doc@cygnus.com, 17jan1991
2016 Throughout this manual, we assume that you are running @dfn{GNU},
2017 the portable operating system from the @dfn{Free Software
2018 Foundation, Inc.}. This restricts our attention to certain kinds of
2019 computer (in particular, the kinds of computers that @sc{gnu} can run on);
2020 once this assumption is granted examples and definitions need less
2023 @command{@value{AS}} is part of a team of programs that turn a high-level
2024 human-readable series of instructions into a low-level
2025 computer-readable series of instructions. Different versions of
2026 @command{@value{AS}} are used for different kinds of computer.
2029 @c There used to be a section "Terminology" here, which defined
2030 @c "contents", "byte", "word", and "long". Defining "word" to any
2031 @c particular size is confusing when the .word directive may generate 16
2032 @c bits on one machine and 32 bits on another; in general, for the user
2033 @c version of this manual, none of these terms seem essential to define.
2034 @c They were used very little even in the former draft of the manual;
2035 @c this draft makes an effort to avoid them (except in names of
2039 @section The GNU Assembler
2041 @c man begin DESCRIPTION
2043 @sc{gnu} @command{as} is really a family of assemblers.
2045 This manual describes @command{@value{AS}}, a member of that family which is
2046 configured for the @value{TARGET} architectures.
2048 If you use (or have used) the @sc{gnu} assembler on one architecture, you
2049 should find a fairly similar environment when you use it on another
2050 architecture. Each version has much in common with the others,
2051 including object file formats, most assembler directives (often called
2052 @dfn{pseudo-ops}) and assembler syntax.@refill
2054 @cindex purpose of @sc{gnu} assembler
2055 @command{@value{AS}} is primarily intended to assemble the output of the
2056 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
2057 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
2058 assemble correctly everything that other assemblers for the same
2059 machine would assemble.
2061 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
2064 @c This remark should appear in generic version of manual; assumption
2065 @c here is that generic version sets M680x0.
2066 This doesn't mean @command{@value{AS}} always uses the same syntax as another
2067 assembler for the same architecture; for example, we know of several
2068 incompatible versions of 680x0 assembly language syntax.
2073 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
2074 program in one pass of the source file. This has a subtle impact on the
2075 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
2077 @node Object Formats
2078 @section Object File Formats
2080 @cindex object file format
2081 The @sc{gnu} assembler can be configured to produce several alternative
2082 object file formats. For the most part, this does not affect how you
2083 write assembly language programs; but directives for debugging symbols
2084 are typically different in different file formats. @xref{Symbol
2085 Attributes,,Symbol Attributes}.
2088 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
2089 @value{OBJ-NAME} format object files.
2091 @c The following should exhaust all configs that set MULTI-OBJ, ideally
2093 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
2094 SOM or ELF format object files.
2099 @section Command Line
2101 @cindex command line conventions
2103 After the program name @command{@value{AS}}, the command line may contain
2104 options and file names. Options may appear in any order, and may be
2105 before, after, or between file names. The order of file names is
2108 @cindex standard input, as input file
2110 @file{--} (two hyphens) by itself names the standard input file
2111 explicitly, as one of the files for @command{@value{AS}} to assemble.
2113 @cindex options, command line
2114 Except for @samp{--} any command-line argument that begins with a
2115 hyphen (@samp{-}) is an option. Each option changes the behavior of
2116 @command{@value{AS}}. No option changes the way another option works. An
2117 option is a @samp{-} followed by one or more letters; the case of
2118 the letter is important. All options are optional.
2120 Some options expect exactly one file name to follow them. The file
2121 name may either immediately follow the option's letter (compatible
2122 with older assemblers) or it may be the next command argument (@sc{gnu}
2123 standard). These two command lines are equivalent:
2126 @value{AS} -o my-object-file.o mumble.s
2127 @value{AS} -omy-object-file.o mumble.s
2131 @section Input Files
2134 @cindex source program
2135 @cindex files, input
2136 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2137 describe the program input to one run of @command{@value{AS}}. The program may
2138 be in one or more files; how the source is partitioned into files
2139 doesn't change the meaning of the source.
2141 @c I added "con" prefix to "catenation" just to prove I can overcome my
2142 @c APL training... doc@cygnus.com
2143 The source program is a concatenation of the text in all the files, in the
2146 @c man begin DESCRIPTION
2147 Each time you run @command{@value{AS}} it assembles exactly one source
2148 program. The source program is made up of one or more files.
2149 (The standard input is also a file.)
2151 You give @command{@value{AS}} a command line that has zero or more input file
2152 names. The input files are read (from left file name to right). A
2153 command-line argument (in any position) that has no special meaning
2154 is taken to be an input file name.
2156 If you give @command{@value{AS}} no file names it attempts to read one input file
2157 from the @command{@value{AS}} standard input, which is normally your terminal. You
2158 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2161 Use @samp{--} if you need to explicitly name the standard input file
2162 in your command line.
2164 If the source is empty, @command{@value{AS}} produces a small, empty object
2169 @subheading Filenames and Line-numbers
2171 @cindex input file linenumbers
2172 @cindex line numbers, in input files
2173 There are two ways of locating a line in the input file (or files) and
2174 either may be used in reporting error messages. One way refers to a line
2175 number in a physical file; the other refers to a line number in a
2176 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2178 @dfn{Physical files} are those files named in the command line given
2179 to @command{@value{AS}}.
2181 @dfn{Logical files} are simply names declared explicitly by assembler
2182 directives; they bear no relation to physical files. Logical file names help
2183 error messages reflect the original source file, when @command{@value{AS}} source
2184 is itself synthesized from other files. @command{@value{AS}} understands the
2185 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2186 @ref{File,,@code{.file}}.
2189 @section Output (Object) File
2195 Every time you run @command{@value{AS}} it produces an output file, which is
2196 your assembly language program translated into numbers. This file
2197 is the object file. Its default name is @code{a.out}.
2198 You can give it another name by using the @option{-o} option. Conventionally,
2199 object file names end with @file{.o}. The default name is used for historical
2200 reasons: older assemblers were capable of assembling self-contained programs
2201 directly into a runnable program. (For some formats, this isn't currently
2202 possible, but it can be done for the @code{a.out} format.)
2206 The object file is meant for input to the linker @code{@value{LD}}. It contains
2207 assembled program code, information to help @code{@value{LD}} integrate
2208 the assembled program into a runnable file, and (optionally) symbolic
2209 information for the debugger.
2211 @c link above to some info file(s) like the description of a.out.
2212 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2215 @section Error and Warning Messages
2217 @c man begin DESCRIPTION
2219 @cindex error messages
2220 @cindex warning messages
2221 @cindex messages from assembler
2222 @command{@value{AS}} may write warnings and error messages to the standard error
2223 file (usually your terminal). This should not happen when a compiler
2224 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2225 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2226 grave problem that stops the assembly.
2230 @cindex format of warning messages
2231 Warning messages have the format
2234 file_name:@b{NNN}:Warning Message Text
2238 @cindex file names and line numbers, in warnings/errors
2239 (where @b{NNN} is a line number). If both a logical file name
2240 (@pxref{File,,@code{.file}}) and a logical line number
2242 (@pxref{Line,,@code{.line}})
2244 have been given then they will be used, otherwise the file name and line number
2245 in the current assembler source file will be used. The message text is
2246 intended to be self explanatory (in the grand Unix tradition).
2248 Note the file name must be set via the logical version of the @code{.file}
2249 directive, not the DWARF2 version of the @code{.file} directive. For example:
2253 error_assembler_source
2259 produces this output:
2263 asm.s:2: Error: no such instruction: `error_assembler_source'
2264 foo.c:31: Error: no such instruction: `error_c_source'
2267 @cindex format of error messages
2268 Error messages have the format
2271 file_name:@b{NNN}:FATAL:Error Message Text
2274 The file name and line number are derived as for warning
2275 messages. The actual message text may be rather less explanatory
2276 because many of them aren't supposed to happen.
2279 @chapter Command-Line Options
2281 @cindex options, all versions of assembler
2282 This chapter describes command-line options available in @emph{all}
2283 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2284 for options specific
2286 to the @value{TARGET} target.
2289 to particular machine architectures.
2292 @c man begin DESCRIPTION
2294 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2295 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2296 The assembler arguments must be separated from each other (and the @samp{-Wa})
2297 by commas. For example:
2300 gcc -c -g -O -Wa,-alh,-L file.c
2304 This passes two options to the assembler: @samp{-alh} (emit a listing to
2305 standard output with high-level and assembly source) and @samp{-L} (retain
2306 local symbols in the symbol table).
2308 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2309 command-line options are automatically passed to the assembler by the compiler.
2310 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2311 precisely what options it passes to each compilation pass, including the
2317 * a:: -a[cdghlns] enable listings
2318 * alternate:: --alternate enable alternate macro syntax
2319 * D:: -D for compatibility
2320 * f:: -f to work faster
2321 * I:: -I for .include search path
2322 @ifclear DIFF-TBL-KLUGE
2323 * K:: -K for compatibility
2325 @ifset DIFF-TBL-KLUGE
2326 * K:: -K for difference tables
2329 * L:: -L to retain local symbols
2330 * listing:: --listing-XXX to configure listing output
2331 * M:: -M or --mri to assemble in MRI compatibility mode
2332 * MD:: --MD for dependency tracking
2333 * no-pad-sections:: --no-pad-sections to stop section padding
2334 * o:: -o to name the object file
2335 * R:: -R to join data and text sections
2336 * statistics:: --statistics to see statistics about assembly
2337 * traditional-format:: --traditional-format for compatible output
2338 * v:: -v to announce version
2339 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2340 * Z:: -Z to make object file even after errors
2344 @section Enable Listings: @option{-a[cdghlns]}
2354 @cindex listings, enabling
2355 @cindex assembly listings, enabling
2357 These options enable listing output from the assembler. By itself,
2358 @samp{-a} requests high-level, assembly, and symbols listing.
2359 You can use other letters to select specific options for the list:
2360 @samp{-ah} requests a high-level language listing,
2361 @samp{-al} requests an output-program assembly listing, and
2362 @samp{-as} requests a symbol table listing.
2363 High-level listings require that a compiler debugging option like
2364 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2367 Use the @samp{-ag} option to print a first section with general assembly
2368 information, like @value{AS} version, switches passed, or time stamp.
2370 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2371 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2372 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2373 omitted from the listing.
2375 Use the @samp{-ad} option to omit debugging directives from the
2378 Once you have specified one of these options, you can further control
2379 listing output and its appearance using the directives @code{.list},
2380 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2382 The @samp{-an} option turns off all forms processing.
2383 If you do not request listing output with one of the @samp{-a} options, the
2384 listing-control directives have no effect.
2386 The letters after @samp{-a} may be combined into one option,
2387 @emph{e.g.}, @samp{-aln}.
2389 Note if the assembler source is coming from the standard input (e.g.,
2391 is being created by @code{@value{GCC}} and the @samp{-pipe} command-line switch
2392 is being used) then the listing will not contain any comments or preprocessor
2393 directives. This is because the listing code buffers input source lines from
2394 stdin only after they have been preprocessed by the assembler. This reduces
2395 memory usage and makes the code more efficient.
2398 @section @option{--alternate}
2401 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2404 @section @option{-D}
2407 This option has no effect whatsoever, but it is accepted to make it more
2408 likely that scripts written for other assemblers also work with
2409 @command{@value{AS}}.
2412 @section Work Faster: @option{-f}
2415 @cindex trusted compiler
2416 @cindex faster processing (@option{-f})
2417 @samp{-f} should only be used when assembling programs written by a
2418 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2419 and comment preprocessing on
2420 the input file(s) before assembling them. @xref{Preprocessing,
2424 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2425 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2430 @section @code{.include} Search Path: @option{-I} @var{path}
2432 @kindex -I @var{path}
2433 @cindex paths for @code{.include}
2434 @cindex search path for @code{.include}
2435 @cindex @code{include} directive search path
2436 Use this option to add a @var{path} to the list of directories
2437 @command{@value{AS}} searches for files specified in @code{.include}
2438 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2439 many times as necessary to include a variety of paths. The current
2440 working directory is always searched first; after that, @command{@value{AS}}
2441 searches any @samp{-I} directories in the same order as they were
2442 specified (left to right) on the command line.
2445 @section Difference Tables: @option{-K}
2448 @ifclear DIFF-TBL-KLUGE
2449 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2450 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2451 where it can be used to warn when the assembler alters the machine code
2452 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2453 family does not have the addressing limitations that sometimes lead to this
2454 alteration on other platforms.
2457 @ifset DIFF-TBL-KLUGE
2458 @cindex difference tables, warning
2459 @cindex warning for altered difference tables
2460 @command{@value{AS}} sometimes alters the code emitted for directives of the
2461 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2462 You can use the @samp{-K} option if you want a warning issued when this
2467 @section Include Local Symbols: @option{-L}
2470 @cindex local symbols, retaining in output
2471 Symbols beginning with system-specific local label prefixes, typically
2472 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2473 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2474 such symbols when debugging, because they are intended for the use of
2475 programs (like compilers) that compose assembler programs, not for your
2476 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2477 such symbols, so you do not normally debug with them.
2479 This option tells @command{@value{AS}} to retain those local symbols
2480 in the object file. Usually if you do this you also tell the linker
2481 @code{@value{LD}} to preserve those symbols.
2484 @section Configuring listing output: @option{--listing}
2486 The listing feature of the assembler can be enabled via the command-line switch
2487 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2488 hex dump of the corresponding locations in the output object file, and displays
2489 them as a listing file. The format of this listing can be controlled by
2490 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2491 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2492 @code{.psize} (@pxref{Psize}), and
2493 @code{.eject} (@pxref{Eject}) and also by the following switches:
2496 @item --listing-lhs-width=@samp{number}
2497 @kindex --listing-lhs-width
2498 @cindex Width of first line disassembly output
2499 Sets the maximum width, in words, of the first line of the hex byte dump. This
2500 dump appears on the left hand side of the listing output.
2502 @item --listing-lhs-width2=@samp{number}
2503 @kindex --listing-lhs-width2
2504 @cindex Width of continuation lines of disassembly output
2505 Sets the maximum width, in words, of any further lines of the hex byte dump for
2506 a given input source line. If this value is not specified, it defaults to being
2507 the same as the value specified for @samp{--listing-lhs-width}. If neither
2508 switch is used the default is to one.
2510 @item --listing-rhs-width=@samp{number}
2511 @kindex --listing-rhs-width
2512 @cindex Width of source line output
2513 Sets the maximum width, in characters, of the source line that is displayed
2514 alongside the hex dump. The default value for this parameter is 100. The
2515 source line is displayed on the right hand side of the listing output.
2517 @item --listing-cont-lines=@samp{number}
2518 @kindex --listing-cont-lines
2519 @cindex Maximum number of continuation lines
2520 Sets the maximum number of continuation lines of hex dump that will be
2521 displayed for a given single line of source input. The default value is 4.
2525 @section Assemble in MRI Compatibility Mode: @option{-M}
2528 @cindex MRI compatibility mode
2529 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2530 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2531 compatible with the @code{ASM68K} assembler from Microtec Research.
2532 The exact nature of the
2533 MRI syntax will not be documented here; see the MRI manuals for more
2534 information. Note in particular that the handling of macros and macro
2535 arguments is somewhat different. The purpose of this option is to permit
2536 assembling existing MRI assembler code using @command{@value{AS}}.
2538 The MRI compatibility is not complete. Certain operations of the MRI assembler
2539 depend upon its object file format, and can not be supported using other object
2540 file formats. Supporting these would require enhancing each object file format
2541 individually. These are:
2544 @item global symbols in common section
2546 The m68k MRI assembler supports common sections which are merged by the linker.
2547 Other object file formats do not support this. @command{@value{AS}} handles
2548 common sections by treating them as a single common symbol. It permits local
2549 symbols to be defined within a common section, but it can not support global
2550 symbols, since it has no way to describe them.
2552 @item complex relocations
2554 The MRI assemblers support relocations against a negated section address, and
2555 relocations which combine the start addresses of two or more sections. These
2556 are not support by other object file formats.
2558 @item @code{END} pseudo-op specifying start address
2560 The MRI @code{END} pseudo-op permits the specification of a start address.
2561 This is not supported by other object file formats. The start address may
2562 instead be specified using the @option{-e} option to the linker, or in a linker
2565 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2567 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2568 name to the output file. This is not supported by other object file formats.
2570 @item @code{ORG} pseudo-op
2572 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2573 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2574 which changes the location within the current section. Absolute sections are
2575 not supported by other object file formats. The address of a section may be
2576 assigned within a linker script.
2579 There are some other features of the MRI assembler which are not supported by
2580 @command{@value{AS}}, typically either because they are difficult or because they
2581 seem of little consequence. Some of these may be supported in future releases.
2585 @item EBCDIC strings
2587 EBCDIC strings are not supported.
2589 @item packed binary coded decimal
2591 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2592 and @code{DCB.P} pseudo-ops are not supported.
2594 @item @code{FEQU} pseudo-op
2596 The m68k @code{FEQU} pseudo-op is not supported.
2598 @item @code{NOOBJ} pseudo-op
2600 The m68k @code{NOOBJ} pseudo-op is not supported.
2602 @item @code{OPT} branch control options
2604 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2605 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2606 relaxes all branches, whether forward or backward, to an appropriate size, so
2607 these options serve no purpose.
2609 @item @code{OPT} list control options
2611 The following m68k @code{OPT} list control options are ignored: @code{C},
2612 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2613 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2615 @item other @code{OPT} options
2617 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2618 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2620 @item @code{OPT} @code{D} option is default
2622 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2623 @code{OPT NOD} may be used to turn it off.
2625 @item @code{XREF} pseudo-op.
2627 The m68k @code{XREF} pseudo-op is ignored.
2632 @section Dependency Tracking: @option{--MD}
2635 @cindex dependency tracking
2638 @command{@value{AS}} can generate a dependency file for the file it creates. This
2639 file consists of a single rule suitable for @code{make} describing the
2640 dependencies of the main source file.
2642 The rule is written to the file named in its argument.
2644 This feature is used in the automatic updating of makefiles.
2646 @node no-pad-sections
2647 @section Output Section Padding
2648 @kindex --no-pad-sections
2649 @cindex output section padding
2650 Normally the assembler will pad the end of each output section up to its
2651 alignment boundary. But this can waste space, which can be significant on
2652 memory constrained targets. So the @option{--no-pad-sections} option will
2653 disable this behaviour.
2656 @section Name the Object File: @option{-o}
2659 @cindex naming object file
2660 @cindex object file name
2661 There is always one object file output when you run @command{@value{AS}}. By
2662 default it has the name @file{a.out}.
2663 You use this option (which takes exactly one filename) to give the
2664 object file a different name.
2666 Whatever the object file is called, @command{@value{AS}} overwrites any
2667 existing file of the same name.
2670 @section Join Data and Text Sections: @option{-R}
2673 @cindex data and text sections, joining
2674 @cindex text and data sections, joining
2675 @cindex joining text and data sections
2676 @cindex merging text and data sections
2677 @option{-R} tells @command{@value{AS}} to write the object file as if all
2678 data-section data lives in the text section. This is only done at
2679 the very last moment: your binary data are the same, but data
2680 section parts are relocated differently. The data section part of
2681 your object file is zero bytes long because all its bytes are
2682 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2684 When you specify @option{-R} it would be possible to generate shorter
2685 address displacements (because we do not have to cross between text and
2686 data section). We refrain from doing this simply for compatibility with
2687 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2690 When @command{@value{AS}} is configured for COFF or ELF output,
2691 this option is only useful if you use sections named @samp{.text} and
2696 @option{-R} is not supported for any of the HPPA targets. Using
2697 @option{-R} generates a warning from @command{@value{AS}}.
2701 @section Display Assembly Statistics: @option{--statistics}
2703 @kindex --statistics
2704 @cindex statistics, about assembly
2705 @cindex time, total for assembly
2706 @cindex space used, maximum for assembly
2707 Use @samp{--statistics} to display two statistics about the resources used by
2708 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2709 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2712 @node traditional-format
2713 @section Compatible Output: @option{--traditional-format}
2715 @kindex --traditional-format
2716 For some targets, the output of @command{@value{AS}} is different in some ways
2717 from the output of some existing assembler. This switch requests
2718 @command{@value{AS}} to use the traditional format instead.
2720 For example, it disables the exception frame optimizations which
2721 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2724 @section Announce Version: @option{-v}
2728 @cindex assembler version
2729 @cindex version of assembler
2730 You can find out what version of as is running by including the
2731 option @samp{-v} (which you can also spell as @samp{-version}) on the
2735 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2737 @command{@value{AS}} should never give a warning or error message when
2738 assembling compiler output. But programs written by people often
2739 cause @command{@value{AS}} to give a warning that a particular assumption was
2740 made. All such warnings are directed to the standard error file.
2744 @cindex suppressing warnings
2745 @cindex warnings, suppressing
2746 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2747 This only affects the warning messages: it does not change any particular of
2748 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2751 @kindex --fatal-warnings
2752 @cindex errors, caused by warnings
2753 @cindex warnings, causing error
2754 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2755 files that generate warnings to be in error.
2758 @cindex warnings, switching on
2759 You can switch these options off again by specifying @option{--warn}, which
2760 causes warnings to be output as usual.
2763 @section Generate Object File in Spite of Errors: @option{-Z}
2764 @cindex object file, after errors
2765 @cindex errors, continuing after
2766 After an error message, @command{@value{AS}} normally produces no output. If for
2767 some reason you are interested in object file output even after
2768 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2769 option. If there are any errors, @command{@value{AS}} continues anyways, and
2770 writes an object file after a final warning message of the form @samp{@var{n}
2771 errors, @var{m} warnings, generating bad object file.}
2776 @cindex machine-independent syntax
2777 @cindex syntax, machine-independent
2778 This chapter describes the machine-independent syntax allowed in a
2779 source file. @command{@value{AS}} syntax is similar to what many other
2780 assemblers use; it is inspired by the BSD 4.2
2785 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2789 * Preprocessing:: Preprocessing
2790 * Whitespace:: Whitespace
2791 * Comments:: Comments
2792 * Symbol Intro:: Symbols
2793 * Statements:: Statements
2794 * Constants:: Constants
2798 @section Preprocessing
2800 @cindex preprocessing
2801 The @command{@value{AS}} internal preprocessor:
2803 @cindex whitespace, removed by preprocessor
2805 adjusts and removes extra whitespace. It leaves one space or tab before
2806 the keywords on a line, and turns any other whitespace on the line into
2809 @cindex comments, removed by preprocessor
2811 removes all comments, replacing them with a single space, or an
2812 appropriate number of newlines.
2814 @cindex constants, converted by preprocessor
2816 converts character constants into the appropriate numeric values.
2819 It does not do macro processing, include file handling, or
2820 anything else you may get from your C compiler's preprocessor. You can
2821 do include file processing with the @code{.include} directive
2822 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2823 to get other ``CPP'' style preprocessing by giving the input file a
2824 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2825 Output, gcc info, Using GNU CC}.
2827 Excess whitespace, comments, and character constants
2828 cannot be used in the portions of the input text that are not
2831 @cindex turning preprocessing on and off
2832 @cindex preprocessing, turning on and off
2835 If the first line of an input file is @code{#NO_APP} or if you use the
2836 @samp{-f} option, whitespace and comments are not removed from the input file.
2837 Within an input file, you can ask for whitespace and comment removal in
2838 specific portions of the by putting a line that says @code{#APP} before the
2839 text that may contain whitespace or comments, and putting a line that says
2840 @code{#NO_APP} after this text. This feature is mainly intend to support
2841 @code{asm} statements in compilers whose output is otherwise free of comments
2848 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2849 Whitespace is used to separate symbols, and to make programs neater for
2850 people to read. Unless within character constants
2851 (@pxref{Characters,,Character Constants}), any whitespace means the same
2852 as exactly one space.
2858 There are two ways of rendering comments to @command{@value{AS}}. In both
2859 cases the comment is equivalent to one space.
2861 Anything from @samp{/*} through the next @samp{*/} is a comment.
2862 This means you may not nest these comments.
2866 The only way to include a newline ('\n') in a comment
2867 is to use this sort of comment.
2870 /* This sort of comment does not nest. */
2873 @cindex line comment character
2874 Anything from a @dfn{line comment} character up to the next newline is
2875 considered a comment and is ignored. The line comment character is target
2876 specific, and some targets multiple comment characters. Some targets also have
2877 line comment characters that only work if they are the first character on a
2878 line. Some targets use a sequence of two characters to introduce a line
2879 comment. Some targets can also change their line comment characters depending
2880 upon command-line options that have been used. For more details see the
2881 @emph{Syntax} section in the documentation for individual targets.
2883 If the line comment character is the hash sign (@samp{#}) then it still has the
2884 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2885 to specify logical line numbers:
2888 @cindex lines starting with @code{#}
2889 @cindex logical line numbers
2890 To be compatible with past assemblers, lines that begin with @samp{#} have a
2891 special interpretation. Following the @samp{#} should be an absolute
2892 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2893 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2894 new logical file name. The rest of the line, if any, should be whitespace.
2896 If the first non-whitespace characters on the line are not numeric,
2897 the line is ignored. (Just like a comment.)
2900 # This is an ordinary comment.
2901 # 42-6 "new_file_name" # New logical file name
2902 # This is logical line # 36.
2904 This feature is deprecated, and may disappear from future versions
2905 of @command{@value{AS}}.
2910 @cindex characters used in symbols
2911 @ifclear SPECIAL-SYMS
2912 A @dfn{symbol} is one or more characters chosen from the set of all
2913 letters (both upper and lower case), digits and the three characters
2919 A @dfn{symbol} is one or more characters chosen from the set of all
2920 letters (both upper and lower case), digits and the three characters
2921 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2927 On most machines, you can also use @code{$} in symbol names; exceptions
2928 are noted in @ref{Machine Dependencies}.
2930 No symbol may begin with a digit. Case is significant.
2931 There is no length limit; all characters are significant. Multibyte characters
2932 are supported. Symbols are delimited by characters not in that set, or by the
2933 beginning of a file (since the source program must end with a newline, the end
2934 of a file is not a possible symbol delimiter). @xref{Symbols}.
2936 Symbol names may also be enclosed in double quote @code{"} characters. In such
2937 cases any characters are allowed, except for the NUL character. If a double
2938 quote character is to be included in the symbol name it must be preceeded by a
2939 backslash @code{\} character.
2940 @cindex length of symbols
2945 @cindex statements, structure of
2946 @cindex line separator character
2947 @cindex statement separator character
2949 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2950 @dfn{line separator character}. The line separator character is target
2951 specific and described in the @emph{Syntax} section of each
2952 target's documentation. Not all targets support a line separator character.
2953 The newline or line separator character is considered to be part of the
2954 preceding statement. Newlines and separators within character constants are an
2955 exception: they do not end statements.
2957 @cindex newline, required at file end
2958 @cindex EOF, newline must precede
2959 It is an error to end any statement with end-of-file: the last
2960 character of any input file should be a newline.@refill
2962 An empty statement is allowed, and may include whitespace. It is ignored.
2964 @cindex instructions and directives
2965 @cindex directives and instructions
2966 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2967 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2969 A statement begins with zero or more labels, optionally followed by a
2970 key symbol which determines what kind of statement it is. The key
2971 symbol determines the syntax of the rest of the statement. If the
2972 symbol begins with a dot @samp{.} then the statement is an assembler
2973 directive: typically valid for any computer. If the symbol begins with
2974 a letter the statement is an assembly language @dfn{instruction}: it
2975 assembles into a machine language instruction.
2977 Different versions of @command{@value{AS}} for different computers
2978 recognize different instructions. In fact, the same symbol may
2979 represent a different instruction in a different computer's assembly
2983 @cindex @code{:} (label)
2984 @cindex label (@code{:})
2985 A label is a symbol immediately followed by a colon (@code{:}).
2986 Whitespace before a label or after a colon is permitted, but you may not
2987 have whitespace between a label's symbol and its colon. @xref{Labels}.
2990 For HPPA targets, labels need not be immediately followed by a colon, but
2991 the definition of a label must begin in column zero. This also implies that
2992 only one label may be defined on each line.
2996 label: .directive followed by something
2997 another_label: # This is an empty statement.
2998 instruction operand_1, operand_2, @dots{}
3005 A constant is a number, written so that its value is known by
3006 inspection, without knowing any context. Like this:
3009 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
3010 .ascii "Ring the bell\7" # A string constant.
3011 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
3012 .float 0f-314159265358979323846264338327\
3013 95028841971.693993751E-40 # - pi, a flonum.
3018 * Characters:: Character Constants
3019 * Numbers:: Number Constants
3023 @subsection Character Constants
3025 @cindex character constants
3026 @cindex constants, character
3027 There are two kinds of character constants. A @dfn{character} stands
3028 for one character in one byte and its value may be used in
3029 numeric expressions. String constants (properly called string
3030 @emph{literals}) are potentially many bytes and their values may not be
3031 used in arithmetic expressions.
3035 * Chars:: Characters
3039 @subsubsection Strings
3041 @cindex string constants
3042 @cindex constants, string
3043 A @dfn{string} is written between double-quotes. It may contain
3044 double-quotes or null characters. The way to get special characters
3045 into a string is to @dfn{escape} these characters: precede them with
3046 a backslash @samp{\} character. For example @samp{\\} represents
3047 one backslash: the first @code{\} is an escape which tells
3048 @command{@value{AS}} to interpret the second character literally as a backslash
3049 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
3050 escape character). The complete list of escapes follows.
3052 @cindex escape codes, character
3053 @cindex character escape codes
3054 @c NOTE: Cindex entries must not start with a backlash character.
3055 @c NOTE: This confuses the pdf2texi script when it is creating the
3056 @c NOTE: index based upon the first character and so it generates:
3057 @c NOTE: \initial {\\}
3058 @c NOTE: which then results in the error message:
3059 @c NOTE: Argument of \\ has an extra }.
3060 @c NOTE: So in the index entries below a space character has been
3061 @c NOTE: prepended to avoid this problem.
3064 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
3066 @cindex @code{ \b} (backspace character)
3067 @cindex backspace (@code{\b})
3069 Mnemonic for backspace; for ASCII this is octal code 010.
3072 @c Mnemonic for EOText; for ASCII this is octal code 004.
3074 @cindex @code{ \f} (formfeed character)
3075 @cindex formfeed (@code{\f})
3077 Mnemonic for FormFeed; for ASCII this is octal code 014.
3079 @cindex @code{ \n} (newline character)
3080 @cindex newline (@code{\n})
3082 Mnemonic for newline; for ASCII this is octal code 012.
3085 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
3087 @cindex @code{ \r} (carriage return character)
3088 @cindex carriage return (@code{backslash-r})
3090 Mnemonic for carriage-Return; for ASCII this is octal code 015.
3093 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
3094 @c other assemblers.
3096 @cindex @code{ \t} (tab)
3097 @cindex tab (@code{\t})
3099 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
3102 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
3103 @c @item \x @var{digit} @var{digit} @var{digit}
3104 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
3106 @cindex @code{ \@var{ddd}} (octal character code)
3107 @cindex octal character code (@code{\@var{ddd}})
3108 @item \ @var{digit} @var{digit} @var{digit}
3109 An octal character code. The numeric code is 3 octal digits.
3110 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
3111 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3113 @cindex @code{ \@var{xd...}} (hex character code)
3114 @cindex hex character code (@code{\@var{xd...}})
3115 @item \@code{x} @var{hex-digits...}
3116 A hex character code. All trailing hex digits are combined. Either upper or
3117 lower case @code{x} works.
3119 @cindex @code{ \\} (@samp{\} character)
3120 @cindex backslash (@code{\\})
3122 Represents one @samp{\} character.
3125 @c Represents one @samp{'} (accent acute) character.
3126 @c This is needed in single character literals
3127 @c (@xref{Characters,,Character Constants}.) to represent
3130 @cindex @code{ \"} (doublequote character)
3131 @cindex doublequote (@code{\"})
3133 Represents one @samp{"} character. Needed in strings to represent
3134 this character, because an unescaped @samp{"} would end the string.
3136 @item \ @var{anything-else}
3137 Any other character when escaped by @kbd{\} gives a warning, but
3138 assembles as if the @samp{\} was not present. The idea is that if
3139 you used an escape sequence you clearly didn't want the literal
3140 interpretation of the following character. However @command{@value{AS}} has no
3141 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3142 code and warns you of the fact.
3145 Which characters are escapable, and what those escapes represent,
3146 varies widely among assemblers. The current set is what we think
3147 the BSD 4.2 assembler recognizes, and is a subset of what most C
3148 compilers recognize. If you are in doubt, do not use an escape
3152 @subsubsection Characters
3154 @cindex single character constant
3155 @cindex character, single
3156 @cindex constant, single character
3157 A single character may be written as a single quote immediately followed by
3158 that character. Some backslash escapes apply to characters, @code{\b},
3159 @code{\f}, @code{\n}, @code{\r}, @code{\t}, and @code{\"} with the same meaning
3160 as for strings, plus @code{\'} for a single quote. So if you want to write the
3161 character backslash, you must write @kbd{'\\} where the first @code{\} escapes
3162 the second @code{\}. As you can see, the quote is an acute accent, not a grave
3165 @ifclear abnormal-separator
3166 (or semicolon @samp{;})
3168 @ifset abnormal-separator
3170 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3175 immediately following an acute accent is taken as a literal character
3176 and does not count as the end of a statement. The value of a character
3177 constant in a numeric expression is the machine's byte-wide code for
3178 that character. @command{@value{AS}} assumes your character code is ASCII:
3179 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3182 @subsection Number Constants
3184 @cindex constants, number
3185 @cindex number constants
3186 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3187 are stored in the target machine. @emph{Integers} are numbers that
3188 would fit into an @code{int} in the C language. @emph{Bignums} are
3189 integers, but they are stored in more than 32 bits. @emph{Flonums}
3190 are floating point numbers, described below.
3193 * Integers:: Integers
3201 @subsubsection Integers
3203 @cindex constants, integer
3205 @cindex binary integers
3206 @cindex integers, binary
3207 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3208 the binary digits @samp{01}.
3210 @cindex octal integers
3211 @cindex integers, octal
3212 An octal integer is @samp{0} followed by zero or more of the octal
3213 digits (@samp{01234567}).
3215 @cindex decimal integers
3216 @cindex integers, decimal
3217 A decimal integer starts with a non-zero digit followed by zero or
3218 more digits (@samp{0123456789}).
3220 @cindex hexadecimal integers
3221 @cindex integers, hexadecimal
3222 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3223 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3225 Integers have the usual values. To denote a negative integer, use
3226 the prefix operator @samp{-} discussed under expressions
3227 (@pxref{Prefix Ops,,Prefix Operators}).
3230 @subsubsection Bignums
3233 @cindex constants, bignum
3234 A @dfn{bignum} has the same syntax and semantics as an integer
3235 except that the number (or its negative) takes more than 32 bits to
3236 represent in binary. The distinction is made because in some places
3237 integers are permitted while bignums are not.
3240 @subsubsection Flonums
3242 @cindex floating point numbers
3243 @cindex constants, floating point
3245 @cindex precision, floating point
3246 A @dfn{flonum} represents a floating point number. The translation is
3247 indirect: a decimal floating point number from the text is converted by
3248 @command{@value{AS}} to a generic binary floating point number of more than
3249 sufficient precision. This generic floating point number is converted
3250 to a particular computer's floating point format (or formats) by a
3251 portion of @command{@value{AS}} specialized to that computer.
3253 A flonum is written by writing (in order)
3258 (@samp{0} is optional on the HPPA.)
3262 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3264 @kbd{e} is recommended. Case is not important.
3266 @c FIXME: verify if flonum syntax really this vague for most cases
3267 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3268 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3271 On the H8/300 and Renesas / SuperH SH architectures, the letter must be
3272 one of the letters @samp{DFPRSX} (in upper or lower case).
3274 On the ARC, the letter must be one of the letters @samp{DFRS}
3275 (in upper or lower case).
3277 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3281 One of the letters @samp{DFRS} (in upper or lower case).
3284 One of the letters @samp{DFPRSX} (in upper or lower case).
3287 The letter @samp{E} (upper case only).
3292 An optional sign: either @samp{+} or @samp{-}.
3295 An optional @dfn{integer part}: zero or more decimal digits.
3298 An optional @dfn{fractional part}: @samp{.} followed by zero
3299 or more decimal digits.
3302 An optional exponent, consisting of:
3306 An @samp{E} or @samp{e}.
3307 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3308 @c principle this can perfectly well be different on different targets.
3310 Optional sign: either @samp{+} or @samp{-}.
3312 One or more decimal digits.
3317 At least one of the integer part or the fractional part must be
3318 present. The floating point number has the usual base-10 value.
3320 @command{@value{AS}} does all processing using integers. Flonums are computed
3321 independently of any floating point hardware in the computer running
3322 @command{@value{AS}}.
3325 @chapter Sections and Relocation
3330 * Secs Background:: Background
3331 * Ld Sections:: Linker Sections
3332 * As Sections:: Assembler Internal Sections
3333 * Sub-Sections:: Sub-Sections
3337 @node Secs Background
3340 Roughly, a section is a range of addresses, with no gaps; all data
3341 ``in'' those addresses is treated the same for some particular purpose.
3342 For example there may be a ``read only'' section.
3344 @cindex linker, and assembler
3345 @cindex assembler, and linker
3346 The linker @code{@value{LD}} reads many object files (partial programs) and
3347 combines their contents to form a runnable program. When @command{@value{AS}}
3348 emits an object file, the partial program is assumed to start at address 0.
3349 @code{@value{LD}} assigns the final addresses for the partial program, so that
3350 different partial programs do not overlap. This is actually an
3351 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3354 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3355 addresses. These blocks slide to their run-time addresses as rigid
3356 units; their length does not change and neither does the order of bytes
3357 within them. Such a rigid unit is called a @emph{section}. Assigning
3358 run-time addresses to sections is called @dfn{relocation}. It includes
3359 the task of adjusting mentions of object-file addresses so they refer to
3360 the proper run-time addresses.
3362 For the H8/300, and for the Renesas / SuperH SH,
3363 @command{@value{AS}} pads sections if needed to
3364 ensure they end on a word (sixteen bit) boundary.
3367 @cindex standard assembler sections
3368 An object file written by @command{@value{AS}} has at least three sections, any
3369 of which may be empty. These are named @dfn{text}, @dfn{data} and
3374 When it generates COFF or ELF output,
3376 @command{@value{AS}} can also generate whatever other named sections you specify
3377 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3378 If you do not use any directives that place output in the @samp{.text}
3379 or @samp{.data} sections, these sections still exist, but are empty.
3384 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3386 @command{@value{AS}} can also generate whatever other named sections you
3387 specify using the @samp{.space} and @samp{.subspace} directives. See
3388 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3389 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3390 assembler directives.
3393 Additionally, @command{@value{AS}} uses different names for the standard
3394 text, data, and bss sections when generating SOM output. Program text
3395 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3396 BSS into @samp{$BSS$}.
3400 Within the object file, the text section starts at address @code{0}, the
3401 data section follows, and the bss section follows the data section.
3404 When generating either SOM or ELF output files on the HPPA, the text
3405 section starts at address @code{0}, the data section at address
3406 @code{0x4000000}, and the bss section follows the data section.
3409 To let @code{@value{LD}} know which data changes when the sections are
3410 relocated, and how to change that data, @command{@value{AS}} also writes to the
3411 object file details of the relocation needed. To perform relocation
3412 @code{@value{LD}} must know, each time an address in the object
3416 Where in the object file is the beginning of this reference to
3419 How long (in bytes) is this reference?
3421 Which section does the address refer to? What is the numeric value of
3423 (@var{address}) @minus{} (@var{start-address of section})?
3426 Is the reference to an address ``Program-Counter relative''?
3429 @cindex addresses, format of
3430 @cindex section-relative addressing
3431 In fact, every address @command{@value{AS}} ever uses is expressed as
3433 (@var{section}) + (@var{offset into section})
3436 Further, most expressions @command{@value{AS}} computes have this section-relative
3439 (For some object formats, such as SOM for the HPPA, some expressions are
3440 symbol-relative instead.)
3443 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3444 @var{N} into section @var{secname}.''
3446 Apart from text, data and bss sections you need to know about the
3447 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3448 addresses in the absolute section remain unchanged. For example, address
3449 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3450 @code{@value{LD}}. Although the linker never arranges two partial programs'
3451 data sections with overlapping addresses after linking, @emph{by definition}
3452 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3453 part of a program is always the same address when the program is running as
3454 address @code{@{absolute@ 239@}} in any other part of the program.
3456 The idea of sections is extended to the @dfn{undefined} section. Any
3457 address whose section is unknown at assembly time is by definition
3458 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3459 Since numbers are always defined, the only way to generate an undefined
3460 address is to mention an undefined symbol. A reference to a named
3461 common block would be such a symbol: its value is unknown at assembly
3462 time so it has section @emph{undefined}.
3464 By analogy the word @emph{section} is used to describe groups of sections in
3465 the linked program. @code{@value{LD}} puts all partial programs' text
3466 sections in contiguous addresses in the linked program. It is
3467 customary to refer to the @emph{text section} of a program, meaning all
3468 the addresses of all partial programs' text sections. Likewise for
3469 data and bss sections.
3471 Some sections are manipulated by @code{@value{LD}}; others are invented for
3472 use of @command{@value{AS}} and have no meaning except during assembly.
3475 @section Linker Sections
3476 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3481 @cindex named sections
3482 @cindex sections, named
3483 @item named sections
3486 @cindex text section
3487 @cindex data section
3491 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3492 separate but equal sections. Anything you can say of one section is
3495 When the program is running, however, it is
3496 customary for the text section to be unalterable. The
3497 text section is often shared among processes: it contains
3498 instructions, constants and the like. The data section of a running
3499 program is usually alterable: for example, C variables would be stored
3500 in the data section.
3505 This section contains zeroed bytes when your program begins running. It
3506 is used to hold uninitialized variables or common storage. The length of
3507 each partial program's bss section is important, but because it starts
3508 out containing zeroed bytes there is no need to store explicit zero
3509 bytes in the object file. The bss section was invented to eliminate
3510 those explicit zeros from object files.
3512 @cindex absolute section
3513 @item absolute section
3514 Address 0 of this section is always ``relocated'' to runtime address 0.
3515 This is useful if you want to refer to an address that @code{@value{LD}} must
3516 not change when relocating. In this sense we speak of absolute
3517 addresses being ``unrelocatable'': they do not change during relocation.
3519 @cindex undefined section
3520 @item undefined section
3521 This ``section'' is a catch-all for address references to objects not in
3522 the preceding sections.
3523 @c FIXME: ref to some other doc on obj-file formats could go here.
3526 @cindex relocation example
3527 An idealized example of three relocatable sections follows.
3529 The example uses the traditional section names @samp{.text} and @samp{.data}.
3531 Memory addresses are on the horizontal axis.
3535 @c END TEXI2ROFF-KILL
3538 partial program # 1: |ttttt|dddd|00|
3545 partial program # 2: |TTT|DDD|000|
3548 +--+---+-----+--+----+---+-----+~~
3549 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3550 +--+---+-----+--+----+---+-----+~~
3552 addresses: 0 @dots{}
3559 \line{\it Partial program \#1: \hfil}
3560 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3561 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3563 \line{\it Partial program \#2: \hfil}
3564 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3565 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3567 \line{\it linked program: \hfil}
3568 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3569 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3570 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3571 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3573 \line{\it addresses: \hfil}
3577 @c END TEXI2ROFF-KILL
3580 @section Assembler Internal Sections
3582 @cindex internal assembler sections
3583 @cindex sections in messages, internal
3584 These sections are meant only for the internal use of @command{@value{AS}}. They
3585 have no meaning at run-time. You do not really need to know about these
3586 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3587 warning messages, so it might be helpful to have an idea of their
3588 meanings to @command{@value{AS}}. These sections are used to permit the
3589 value of every expression in your assembly language program to be a
3590 section-relative address.
3593 @cindex assembler internal logic error
3594 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3595 An internal assembler logic error has been found. This means there is a
3596 bug in the assembler.
3598 @cindex expr (internal section)
3600 The assembler stores complex expression internally as combinations of
3601 symbols. When it needs to represent an expression as a symbol, it puts
3602 it in the expr section.
3604 @c FIXME item transfer[t] vector preload
3605 @c FIXME item transfer[t] vector postload
3606 @c FIXME item register
3610 @section Sub-Sections
3612 @cindex numbered subsections
3613 @cindex grouping data
3619 fall into two sections: text and data.
3621 You may have separate groups of
3623 data in named sections
3627 data in named sections
3633 that you want to end up near to each other in the object file, even though they
3634 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3635 use @dfn{subsections} for this purpose. Within each section, there can be
3636 numbered subsections with values from 0 to 8192. Objects assembled into the
3637 same subsection go into the object file together with other objects in the same
3638 subsection. For example, a compiler might want to store constants in the text
3639 section, but might not want to have them interspersed with the program being
3640 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3641 section of code being output, and a @samp{.text 1} before each group of
3642 constants being output.
3644 Subsections are optional. If you do not use subsections, everything
3645 goes in subsection number zero.
3648 Each subsection is zero-padded up to a multiple of four bytes.
3649 (Subsections may be padded a different amount on different flavors
3650 of @command{@value{AS}}.)
3654 On the H8/300 platform, each subsection is zero-padded to a word
3655 boundary (two bytes).
3656 The same is true on the Renesas SH.
3660 Subsections appear in your object file in numeric order, lowest numbered
3661 to highest. (All this to be compatible with other people's assemblers.)
3662 The object file contains no representation of subsections; @code{@value{LD}} and
3663 other programs that manipulate object files see no trace of them.
3664 They just see all your text subsections as a text section, and all your
3665 data subsections as a data section.
3667 To specify which subsection you want subsequent statements assembled
3668 into, use a numeric argument to specify it, in a @samp{.text
3669 @var{expression}} or a @samp{.data @var{expression}} statement.
3672 When generating COFF output, you
3677 can also use an extra subsection
3678 argument with arbitrary named sections: @samp{.section @var{name},
3683 When generating ELF output, you
3688 can also use the @code{.subsection} directive (@pxref{SubSection})
3689 to specify a subsection: @samp{.subsection @var{expression}}.
3691 @var{Expression} should be an absolute expression
3692 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3693 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3694 begins in @code{text 0}. For instance:
3696 .text 0 # The default subsection is text 0 anyway.
3697 .ascii "This lives in the first text subsection. *"
3699 .ascii "But this lives in the second text subsection."
3701 .ascii "This lives in the data section,"
3702 .ascii "in the first data subsection."
3704 .ascii "This lives in the first text section,"
3705 .ascii "immediately following the asterisk (*)."
3708 Each section has a @dfn{location counter} incremented by one for every byte
3709 assembled into that section. Because subsections are merely a convenience
3710 restricted to @command{@value{AS}} there is no concept of a subsection location
3711 counter. There is no way to directly manipulate a location counter---but the
3712 @code{.align} directive changes it, and any label definition captures its
3713 current value. The location counter of the section where statements are being
3714 assembled is said to be the @dfn{active} location counter.
3717 @section bss Section
3720 @cindex common variable storage
3721 The bss section is used for local common variable storage.
3722 You may allocate address space in the bss section, but you may
3723 not dictate data to load into it before your program executes. When
3724 your program starts running, all the contents of the bss
3725 section are zeroed bytes.
3727 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3728 @ref{Lcomm,,@code{.lcomm}}.
3730 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3731 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3734 When assembling for a target which supports multiple sections, such as ELF or
3735 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3736 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3737 section. Typically the section will only contain symbol definitions and
3738 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3745 Symbols are a central concept: the programmer uses symbols to name
3746 things, the linker uses symbols to link, and the debugger uses symbols
3750 @cindex debuggers, and symbol order
3751 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3752 the same order they were declared. This may break some debuggers.
3757 * Setting Symbols:: Giving Symbols Other Values
3758 * Symbol Names:: Symbol Names
3759 * Dot:: The Special Dot Symbol
3760 * Symbol Attributes:: Symbol Attributes
3767 A @dfn{label} is written as a symbol immediately followed by a colon
3768 @samp{:}. The symbol then represents the current value of the
3769 active location counter, and is, for example, a suitable instruction
3770 operand. You are warned if you use the same symbol to represent two
3771 different locations: the first definition overrides any other
3775 On the HPPA, the usual form for a label need not be immediately followed by a
3776 colon, but instead must start in column zero. Only one label may be defined on
3777 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3778 provides a special directive @code{.label} for defining labels more flexibly.
3781 @node Setting Symbols
3782 @section Giving Symbols Other Values
3784 @cindex assigning values to symbols
3785 @cindex symbol values, assigning
3786 A symbol can be given an arbitrary value by writing a symbol, followed
3787 by an equals sign @samp{=}, followed by an expression
3788 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3789 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3790 equals sign @samp{=}@samp{=} here represents an equivalent of the
3791 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3794 Blackfin does not support symbol assignment with @samp{=}.
3798 @section Symbol Names
3800 @cindex symbol names
3801 @cindex names, symbol
3802 @ifclear SPECIAL-SYMS
3803 Symbol names begin with a letter or with one of @samp{._}. On most
3804 machines, you can also use @code{$} in symbol names; exceptions are
3805 noted in @ref{Machine Dependencies}. That character may be followed by any
3806 string of digits, letters, dollar signs (unless otherwise noted for a
3807 particular target machine), and underscores.
3811 Symbol names begin with a letter or with one of @samp{._}. On the
3812 Renesas SH you can also use @code{$} in symbol names. That
3813 character may be followed by any string of digits, letters, dollar signs (save
3814 on the H8/300), and underscores.
3818 Case of letters is significant: @code{foo} is a different symbol name
3821 Symbol names do not start with a digit. An exception to this rule is made for
3822 Local Labels. See below.
3824 Multibyte characters are supported. To generate a symbol name containing
3825 multibyte characters enclose it within double quotes and use escape codes. cf
3826 @xref{Strings}. Generating a multibyte symbol name from a label is not
3827 currently supported.
3829 Each symbol has exactly one name. Each name in an assembly language program
3830 refers to exactly one symbol. You may use that symbol name any number of times
3833 @subheading Local Symbol Names
3835 @cindex local symbol names
3836 @cindex symbol names, local
3837 A local symbol is any symbol beginning with certain local label prefixes.
3838 By default, the local label prefix is @samp{.L} for ELF systems or
3839 @samp{L} for traditional a.out systems, but each target may have its own
3840 set of local label prefixes.
3842 On the HPPA local symbols begin with @samp{L$}.
3845 Local symbols are defined and used within the assembler, but they are
3846 normally not saved in object files. Thus, they are not visible when debugging.
3847 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3848 to retain the local symbols in the object files.
3850 @subheading Local Labels
3852 @cindex local labels
3853 @cindex temporary symbol names
3854 @cindex symbol names, temporary
3855 Local labels are different from local symbols. Local labels help compilers and
3856 programmers use names temporarily. They create symbols which are guaranteed to
3857 be unique over the entire scope of the input source code and which can be
3858 referred to by a simple notation. To define a local label, write a label of
3859 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3860 To refer to the most recent previous definition of that label write
3861 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3862 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3863 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3865 There is no restriction on how you can use these labels, and you can reuse them
3866 too. So that it is possible to repeatedly define the same local label (using
3867 the same number @samp{@b{N}}), although you can only refer to the most recently
3868 defined local label of that number (for a backwards reference) or the next
3869 definition of a specific local label for a forward reference. It is also worth
3870 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3871 implemented in a slightly more efficient manner than the others.
3882 Which is the equivalent of:
3885 label_1: branch label_3
3886 label_2: branch label_1
3887 label_3: branch label_4
3888 label_4: branch label_3
3891 Local label names are only a notational device. They are immediately
3892 transformed into more conventional symbol names before the assembler uses them.
3893 The symbol names are stored in the symbol table, appear in error messages, and
3894 are optionally emitted to the object file. The names are constructed using
3898 @item @emph{local label prefix}
3899 All local symbols begin with the system-specific local label prefix.
3900 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3901 that start with the local label prefix. These labels are
3902 used for symbols you are never intended to see. If you use the
3903 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3904 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3905 you may use them in debugging.
3908 This is the number that was used in the local label definition. So if the
3909 label is written @samp{55:} then the number is @samp{55}.
3912 This unusual character is included so you do not accidentally invent a symbol
3913 of the same name. The character has ASCII value of @samp{\002} (control-B).
3915 @item @emph{ordinal number}
3916 This is a serial number to keep the labels distinct. The first definition of
3917 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3918 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3919 the number @samp{1} and its 15th definition gets @samp{15} as well.
3922 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3923 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3925 @subheading Dollar Local Labels
3926 @cindex dollar local symbols
3928 On some targets @code{@value{AS}} also supports an even more local form of
3929 local labels called dollar labels. These labels go out of scope (i.e., they
3930 become undefined) as soon as a non-local label is defined. Thus they remain
3931 valid for only a small region of the input source code. Normal local labels,
3932 by contrast, remain in scope for the entire file, or until they are redefined
3933 by another occurrence of the same local label.
3935 Dollar labels are defined in exactly the same way as ordinary local labels,
3936 except that they have a dollar sign suffix to their numeric value, e.g.,
3939 They can also be distinguished from ordinary local labels by their transformed
3940 names which use ASCII character @samp{\001} (control-A) as the magic character
3941 to distinguish them from ordinary labels. For example, the fifth definition of
3942 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3945 @section The Special Dot Symbol
3947 @cindex dot (symbol)
3948 @cindex @code{.} (symbol)
3949 @cindex current address
3950 @cindex location counter
3951 The special symbol @samp{.} refers to the current address that
3952 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3953 .long .} defines @code{melvin} to contain its own address.
3954 Assigning a value to @code{.} is treated the same as a @code{.org}
3956 @ifclear no-space-dir
3957 Thus, the expression @samp{.=.+4} is the same as saying
3961 @node Symbol Attributes
3962 @section Symbol Attributes
3964 @cindex symbol attributes
3965 @cindex attributes, symbol
3966 Every symbol has, as well as its name, the attributes ``Value'' and
3967 ``Type''. Depending on output format, symbols can also have auxiliary
3970 The detailed definitions are in @file{a.out.h}.
3973 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3974 all these attributes, and probably won't warn you. This makes the
3975 symbol an externally defined symbol, which is generally what you
3979 * Symbol Value:: Value
3980 * Symbol Type:: Type
3982 * a.out Symbols:: Symbol Attributes: @code{a.out}
3985 * COFF Symbols:: Symbol Attributes for COFF
3988 * SOM Symbols:: Symbol Attributes for SOM
3995 @cindex value of a symbol
3996 @cindex symbol value
3997 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3998 location in the text, data, bss or absolute sections the value is the
3999 number of addresses from the start of that section to the label.
4000 Naturally for text, data and bss sections the value of a symbol changes
4001 as @code{@value{LD}} changes section base addresses during linking. Absolute
4002 symbols' values do not change during linking: that is why they are
4005 The value of an undefined symbol is treated in a special way. If it is
4006 0 then the symbol is not defined in this assembler source file, and
4007 @code{@value{LD}} tries to determine its value from other files linked into the
4008 same program. You make this kind of symbol simply by mentioning a symbol
4009 name without defining it. A non-zero value represents a @code{.comm}
4010 common declaration. The value is how much common storage to reserve, in
4011 bytes (addresses). The symbol refers to the first address of the
4017 @cindex type of a symbol
4019 The type attribute of a symbol contains relocation (section)
4020 information, any flag settings indicating that a symbol is external, and
4021 (optionally), other information for linkers and debuggers. The exact
4022 format depends on the object-code output format in use.
4026 @subsection Symbol Attributes: @code{a.out}
4028 @cindex @code{a.out} symbol attributes
4029 @cindex symbol attributes, @code{a.out}
4032 * Symbol Desc:: Descriptor
4033 * Symbol Other:: Other
4037 @subsubsection Descriptor
4039 @cindex descriptor, of @code{a.out} symbol
4040 This is an arbitrary 16-bit value. You may establish a symbol's
4041 descriptor value by using a @code{.desc} statement
4042 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
4043 @command{@value{AS}}.
4046 @subsubsection Other
4048 @cindex other attribute, of @code{a.out} symbol
4049 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
4054 @subsection Symbol Attributes for COFF
4056 @cindex COFF symbol attributes
4057 @cindex symbol attributes, COFF
4059 The COFF format supports a multitude of auxiliary symbol attributes;
4060 like the primary symbol attributes, they are set between @code{.def} and
4061 @code{.endef} directives.
4063 @subsubsection Primary Attributes
4065 @cindex primary attributes, COFF symbols
4066 The symbol name is set with @code{.def}; the value and type,
4067 respectively, with @code{.val} and @code{.type}.
4069 @subsubsection Auxiliary Attributes
4071 @cindex auxiliary attributes, COFF symbols
4072 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
4073 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
4074 table information for COFF.
4079 @subsection Symbol Attributes for SOM
4081 @cindex SOM symbol attributes
4082 @cindex symbol attributes, SOM
4084 The SOM format for the HPPA supports a multitude of symbol attributes set with
4085 the @code{.EXPORT} and @code{.IMPORT} directives.
4087 The attributes are described in @cite{HP9000 Series 800 Assembly
4088 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
4089 @code{EXPORT} assembler directive documentation.
4093 @chapter Expressions
4097 @cindex numeric values
4098 An @dfn{expression} specifies an address or numeric value.
4099 Whitespace may precede and/or follow an expression.
4101 The result of an expression must be an absolute number, or else an offset into
4102 a particular section. If an expression is not absolute, and there is not
4103 enough information when @command{@value{AS}} sees the expression to know its
4104 section, a second pass over the source program might be necessary to interpret
4105 the expression---but the second pass is currently not implemented.
4106 @command{@value{AS}} aborts with an error message in this situation.
4109 * Empty Exprs:: Empty Expressions
4110 * Integer Exprs:: Integer Expressions
4114 @section Empty Expressions
4116 @cindex empty expressions
4117 @cindex expressions, empty
4118 An empty expression has no value: it is just whitespace or null.
4119 Wherever an absolute expression is required, you may omit the
4120 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4121 is compatible with other assemblers.
4124 @section Integer Expressions
4126 @cindex integer expressions
4127 @cindex expressions, integer
4128 An @dfn{integer expression} is one or more @emph{arguments} delimited
4129 by @emph{operators}.
4132 * Arguments:: Arguments
4133 * Operators:: Operators
4134 * Prefix Ops:: Prefix Operators
4135 * Infix Ops:: Infix Operators
4139 @subsection Arguments
4141 @cindex expression arguments
4142 @cindex arguments in expressions
4143 @cindex operands in expressions
4144 @cindex arithmetic operands
4145 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4146 contexts arguments are sometimes called ``arithmetic operands''. In
4147 this manual, to avoid confusing them with the ``instruction operands'' of
4148 the machine language, we use the term ``argument'' to refer to parts of
4149 expressions only, reserving the word ``operand'' to refer only to machine
4150 instruction operands.
4152 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4153 @var{section} is one of text, data, bss, absolute,
4154 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4157 Numbers are usually integers.
4159 A number can be a flonum or bignum. In this case, you are warned
4160 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4161 these 32 bits are an integer. You may write integer-manipulating
4162 instructions that act on exotic constants, compatible with other
4165 @cindex subexpressions
4166 Subexpressions are a left parenthesis @samp{(} followed by an integer
4167 expression, followed by a right parenthesis @samp{)}; or a prefix
4168 operator followed by an argument.
4171 @subsection Operators
4173 @cindex operators, in expressions
4174 @cindex arithmetic functions
4175 @cindex functions, in expressions
4176 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4177 operators are followed by an argument. Infix operators appear
4178 between their arguments. Operators may be preceded and/or followed by
4182 @subsection Prefix Operator
4184 @cindex prefix operators
4185 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4186 one argument, which must be absolute.
4188 @c the tex/end tex stuff surrounding this small table is meant to make
4189 @c it align, on the printed page, with the similar table in the next
4190 @c section (which is inside an enumerate).
4192 \global\advance\leftskip by \itemindent
4197 @dfn{Negation}. Two's complement negation.
4199 @dfn{Complementation}. Bitwise not.
4203 \global\advance\leftskip by -\itemindent
4207 @subsection Infix Operators
4209 @cindex infix operators
4210 @cindex operators, permitted arguments
4211 @dfn{Infix operators} take two arguments, one on either side. Operators
4212 have precedence, but operations with equal precedence are performed left
4213 to right. Apart from @code{+} or @option{-}, both arguments must be
4214 absolute, and the result is absolute.
4217 @cindex operator precedence
4218 @cindex precedence of operators
4225 @dfn{Multiplication}.
4228 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4234 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4237 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4241 Intermediate precedence
4246 @dfn{Bitwise Inclusive Or}.
4252 @dfn{Bitwise Exclusive Or}.
4255 @dfn{Bitwise Or Not}.
4262 @cindex addition, permitted arguments
4263 @cindex plus, permitted arguments
4264 @cindex arguments for addition
4266 @dfn{Addition}. If either argument is absolute, the result has the section of
4267 the other argument. You may not add together arguments from different
4270 @cindex subtraction, permitted arguments
4271 @cindex minus, permitted arguments
4272 @cindex arguments for subtraction
4274 @dfn{Subtraction}. If the right argument is absolute, the
4275 result has the section of the left argument.
4276 If both arguments are in the same section, the result is absolute.
4277 You may not subtract arguments from different sections.
4278 @c FIXME is there still something useful to say about undefined - undefined ?
4280 @cindex comparison expressions
4281 @cindex expressions, comparison
4286 @dfn{Is Not Equal To}
4290 @dfn{Is Greater Than}
4292 @dfn{Is Greater Than Or Equal To}
4294 @dfn{Is Less Than Or Equal To}
4296 The comparison operators can be used as infix operators. A true results has a
4297 value of -1 whereas a false result has a value of 0. Note, these operators
4298 perform signed comparisons.
4301 @item Lowest Precedence
4310 These two logical operations can be used to combine the results of sub
4311 expressions. Note, unlike the comparison operators a true result returns a
4312 value of 1 but a false results does still return 0. Also note that the logical
4313 or operator has a slightly lower precedence than logical and.
4318 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4319 address; you can only have a defined section in one of the two arguments.
4322 @chapter Assembler Directives
4324 @cindex directives, machine independent
4325 @cindex pseudo-ops, machine independent
4326 @cindex machine independent directives
4327 All assembler directives have names that begin with a period (@samp{.}).
4328 The names are case insensitive for most targets, and usually written
4331 This chapter discusses directives that are available regardless of the
4332 target machine configuration for the @sc{gnu} assembler.
4334 Some machine configurations provide additional directives.
4335 @xref{Machine Dependencies}.
4338 @ifset machine-directives
4339 @xref{Machine Dependencies}, for additional directives.
4344 * Abort:: @code{.abort}
4346 * ABORT (COFF):: @code{.ABORT}
4349 * Align:: @code{.align [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4350 * Altmacro:: @code{.altmacro}
4351 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4352 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4353 * Balign:: @code{.balign [@var{abs-expr}[, @var{abs-expr}]]}
4354 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4355 * Byte:: @code{.byte @var{expressions}}
4356 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4357 * Comm:: @code{.comm @var{symbol} , @var{length} }
4358 * Data:: @code{.data @var{subsection}}
4359 * Dc:: @code{.dc[@var{size}] @var{expressions}}
4360 * Dcb:: @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4361 * Ds:: @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4363 * Def:: @code{.def @var{name}}
4366 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4372 * Double:: @code{.double @var{flonums}}
4373 * Eject:: @code{.eject}
4374 * Else:: @code{.else}
4375 * Elseif:: @code{.elseif}
4378 * Endef:: @code{.endef}
4381 * Endfunc:: @code{.endfunc}
4382 * Endif:: @code{.endif}
4383 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4384 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4385 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4387 * Error:: @code{.error @var{string}}
4388 * Exitm:: @code{.exitm}
4389 * Extern:: @code{.extern}
4390 * Fail:: @code{.fail}
4391 * File:: @code{.file}
4392 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4393 * Float:: @code{.float @var{flonums}}
4394 * Func:: @code{.func}
4395 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4397 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4398 * Hidden:: @code{.hidden @var{names}}
4401 * hword:: @code{.hword @var{expressions}}
4402 * Ident:: @code{.ident}
4403 * If:: @code{.if @var{absolute expression}}
4404 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4405 * Include:: @code{.include "@var{file}"}
4406 * Int:: @code{.int @var{expressions}}
4408 * Internal:: @code{.internal @var{names}}
4411 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4412 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4413 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4414 * Lflags:: @code{.lflags}
4415 @ifclear no-line-dir
4416 * Line:: @code{.line @var{line-number}}
4419 * Linkonce:: @code{.linkonce [@var{type}]}
4420 * List:: @code{.list}
4421 * Ln:: @code{.ln @var{line-number}}
4422 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4423 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4425 * Local:: @code{.local @var{names}}
4428 * Long:: @code{.long @var{expressions}}
4430 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4433 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4434 * MRI:: @code{.mri @var{val}}
4435 * Noaltmacro:: @code{.noaltmacro}
4436 * Nolist:: @code{.nolist}
4437 * Nops:: @code{.nops @var{size}[, @var{control}]}
4438 * Octa:: @code{.octa @var{bignums}}
4439 * Offset:: @code{.offset @var{loc}}
4440 * Org:: @code{.org @var{new-lc}, @var{fill}}
4441 * P2align:: @code{.p2align [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4443 * PopSection:: @code{.popsection}
4444 * Previous:: @code{.previous}
4447 * Print:: @code{.print @var{string}}
4449 * Protected:: @code{.protected @var{names}}
4452 * Psize:: @code{.psize @var{lines}, @var{columns}}
4453 * Purgem:: @code{.purgem @var{name}}
4455 * PushSection:: @code{.pushsection @var{name}}
4458 * Quad:: @code{.quad @var{bignums}}
4459 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4460 * Rept:: @code{.rept @var{count}}
4461 * Sbttl:: @code{.sbttl "@var{subheading}"}
4463 * Scl:: @code{.scl @var{class}}
4466 * Section:: @code{.section @var{name}[, @var{flags}]}
4469 * Set:: @code{.set @var{symbol}, @var{expression}}
4470 * Short:: @code{.short @var{expressions}}
4471 * Single:: @code{.single @var{flonums}}
4473 * Size:: @code{.size [@var{name} , @var{expression}]}
4475 @ifclear no-space-dir
4476 * Skip:: @code{.skip @var{size} [,@var{fill}]}
4479 * Sleb128:: @code{.sleb128 @var{expressions}}
4480 @ifclear no-space-dir
4481 * Space:: @code{.space @var{size} [,@var{fill}]}
4484 * Stab:: @code{.stabd, .stabn, .stabs}
4487 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4488 * Struct:: @code{.struct @var{expression}}
4490 * SubSection:: @code{.subsection}
4491 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4495 * Tag:: @code{.tag @var{structname}}
4498 * Text:: @code{.text @var{subsection}}
4499 * Title:: @code{.title "@var{heading}"}
4501 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4504 * Uleb128:: @code{.uleb128 @var{expressions}}
4506 * Val:: @code{.val @var{addr}}
4510 * Version:: @code{.version "@var{string}"}
4511 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4512 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4515 * Warning:: @code{.warning @var{string}}
4516 * Weak:: @code{.weak @var{names}}
4517 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4518 * Word:: @code{.word @var{expressions}}
4519 @ifclear no-space-dir
4520 * Zero:: @code{.zero @var{size}}
4523 * 2byte:: @code{.2byte @var{expressions}}
4524 * 4byte:: @code{.4byte @var{expressions}}
4525 * 8byte:: @code{.8byte @var{bignums}}
4527 * Deprecated:: Deprecated Directives
4531 @section @code{.abort}
4533 @cindex @code{abort} directive
4534 @cindex stopping the assembly
4535 This directive stops the assembly immediately. It is for
4536 compatibility with other assemblers. The original idea was that the
4537 assembly language source would be piped into the assembler. If the sender
4538 of the source quit, it could use this directive tells @command{@value{AS}} to
4539 quit also. One day @code{.abort} will not be supported.
4543 @section @code{.ABORT} (COFF)
4545 @cindex @code{ABORT} directive
4546 When producing COFF output, @command{@value{AS}} accepts this directive as a
4547 synonym for @samp{.abort}.
4552 @section @code{.align [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4554 @cindex padding the location counter
4555 @cindex @code{align} directive
4556 Pad the location counter (in the current subsection) to a particular storage
4557 boundary. The first expression (which must be absolute) is the alignment
4558 required, as described below. If this expression is omitted then a default
4559 value of 0 is used, effectively disabling alignment requirements.
4561 The second expression (also absolute) gives the fill value to be stored in the
4562 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4563 padding bytes are normally zero. However, on most systems, if the section is
4564 marked as containing code and the fill value is omitted, the space is filled
4565 with no-op instructions.
4567 The third expression is also absolute, and is also optional. If it is present,
4568 it is the maximum number of bytes that should be skipped by this alignment
4569 directive. If doing the alignment would require skipping more bytes than the
4570 specified maximum, then the alignment is not done at all. You can omit the
4571 fill value (the second argument) entirely by simply using two commas after the
4572 required alignment; this can be useful if you want the alignment to be filled
4573 with no-op instructions when appropriate.
4575 The way the required alignment is specified varies from system to system.
4576 For the arc, hppa, i386 using ELF, iq2000, m68k, or1k,
4577 s390, sparc, tic4x and xtensa, the first expression is the
4578 alignment request in bytes. For example @samp{.align 8} advances
4579 the location counter until it is a multiple of 8. If the location counter
4580 is already a multiple of 8, no change is needed. For the tic54x, the
4581 first expression is the alignment request in words.
4583 For other systems, including ppc, i386 using a.out format, arm and
4584 strongarm, it is the
4585 number of low-order zero bits the location counter must have after
4586 advancement. For example @samp{.align 3} advances the location
4587 counter until it is a multiple of 8. If the location counter is already a
4588 multiple of 8, no change is needed.
4590 This inconsistency is due to the different behaviors of the various
4591 native assemblers for these systems which GAS must emulate.
4592 GAS also provides @code{.balign} and @code{.p2align} directives,
4593 described later, which have a consistent behavior across all
4594 architectures (but are specific to GAS).
4597 @section @code{.altmacro}
4598 Enable alternate macro mode, enabling:
4601 @item LOCAL @var{name} [ , @dots{} ]
4602 One additional directive, @code{LOCAL}, is available. It is used to
4603 generate a string replacement for each of the @var{name} arguments, and
4604 replace any instances of @var{name} in each macro expansion. The
4605 replacement string is unique in the assembly, and different for each
4606 separate macro expansion. @code{LOCAL} allows you to write macros that
4607 define symbols, without fear of conflict between separate macro expansions.
4609 @item String delimiters
4610 You can write strings delimited in these other ways besides
4611 @code{"@var{string}"}:
4614 @item '@var{string}'
4615 You can delimit strings with single-quote characters.
4617 @item <@var{string}>
4618 You can delimit strings with matching angle brackets.
4621 @item single-character string escape
4622 To include any single character literally in a string (even if the
4623 character would otherwise have some special meaning), you can prefix the
4624 character with @samp{!} (an exclamation mark). For example, you can
4625 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4627 @item Expression results as strings
4628 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4629 and use the result as a string.
4633 @section @code{.ascii "@var{string}"}@dots{}
4635 @cindex @code{ascii} directive
4636 @cindex string literals
4637 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4638 separated by commas. It assembles each string (with no automatic
4639 trailing zero byte) into consecutive addresses.
4642 @section @code{.asciz "@var{string}"}@dots{}
4644 @cindex @code{asciz} directive
4645 @cindex zero-terminated strings
4646 @cindex null-terminated strings
4647 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4648 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4651 @section @code{.balign[wl] [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
4653 @cindex padding the location counter given number of bytes
4654 @cindex @code{balign} directive
4655 Pad the location counter (in the current subsection) to a particular
4656 storage boundary. The first expression (which must be absolute) is the
4657 alignment request in bytes. For example @samp{.balign 8} advances
4658 the location counter until it is a multiple of 8. If the location counter
4659 is already a multiple of 8, no change is needed. If the expression is omitted
4660 then a default value of 0 is used, effectively disabling alignment requirements.
4662 The second expression (also absolute) gives the fill value to be stored in the
4663 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4664 padding bytes are normally zero. However, on most systems, if the section is
4665 marked as containing code and the fill value is omitted, the space is filled
4666 with no-op instructions.
4668 The third expression is also absolute, and is also optional. If it is present,
4669 it is the maximum number of bytes that should be skipped by this alignment
4670 directive. If doing the alignment would require skipping more bytes than the
4671 specified maximum, then the alignment is not done at all. You can omit the
4672 fill value (the second argument) entirely by simply using two commas after the
4673 required alignment; this can be useful if you want the alignment to be filled
4674 with no-op instructions when appropriate.
4676 @cindex @code{balignw} directive
4677 @cindex @code{balignl} directive
4678 The @code{.balignw} and @code{.balignl} directives are variants of the
4679 @code{.balign} directive. The @code{.balignw} directive treats the fill
4680 pattern as a two byte word value. The @code{.balignl} directives treats the
4681 fill pattern as a four byte longword value. For example, @code{.balignw
4682 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4683 filled in with the value 0x368d (the exact placement of the bytes depends upon
4684 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4687 @node Bundle directives
4688 @section Bundle directives
4689 @subsection @code{.bundle_align_mode @var{abs-expr}}
4690 @cindex @code{bundle_align_mode} directive
4692 @cindex instruction bundle
4693 @cindex aligned instruction bundle
4694 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4695 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4696 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4697 disabled (which is the default state). If the argument it not zero, it
4698 gives the size of an instruction bundle as a power of two (as for the
4699 @code{.p2align} directive, @pxref{P2align}).
4701 For some targets, it's an ABI requirement that no instruction may span a
4702 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4703 instructions that starts on an aligned boundary. For example, if
4704 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4705 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4706 effect, no single instruction may span a boundary between bundles. If an
4707 instruction would start too close to the end of a bundle for the length of
4708 that particular instruction to fit within the bundle, then the space at the
4709 end of that bundle is filled with no-op instructions so the instruction
4710 starts in the next bundle. As a corollary, it's an error if any single
4711 instruction's encoding is longer than the bundle size.
4713 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4714 @cindex @code{bundle_lock} directive
4715 @cindex @code{bundle_unlock} directive
4716 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4717 allow explicit control over instruction bundle padding. These directives
4718 are only valid when @code{.bundle_align_mode} has been used to enable
4719 aligned instruction bundle mode. It's an error if they appear when
4720 @code{.bundle_align_mode} has not been used at all, or when the last
4721 directive was @w{@code{.bundle_align_mode 0}}.
4723 @cindex bundle-locked
4724 For some targets, it's an ABI requirement that certain instructions may
4725 appear only as part of specified permissible sequences of multiple
4726 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4727 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4728 instruction sequence. For purposes of aligned instruction bundle mode, a
4729 sequence starting with @code{.bundle_lock} and ending with
4730 @code{.bundle_unlock} is treated as a single instruction. That is, the
4731 entire sequence must fit into a single bundle and may not span a bundle
4732 boundary. If necessary, no-op instructions will be inserted before the
4733 first instruction of the sequence so that the whole sequence starts on an
4734 aligned bundle boundary. It's an error if the sequence is longer than the
4737 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4738 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4739 nested. That is, a second @code{.bundle_lock} directive before the next
4740 @code{.bundle_unlock} directive has no effect except that it must be
4741 matched by another closing @code{.bundle_unlock} so that there is the
4742 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4745 @section @code{.byte @var{expressions}}
4747 @cindex @code{byte} directive
4748 @cindex integers, one byte
4749 @code{.byte} expects zero or more expressions, separated by commas.
4750 Each expression is assembled into the next byte.
4752 @node CFI directives
4753 @section CFI directives
4754 @subsection @code{.cfi_sections @var{section_list}}
4755 @cindex @code{cfi_sections} directive
4756 @code{.cfi_sections} may be used to specify whether CFI directives
4757 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4758 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4759 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4760 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4761 directive is not used is @code{.cfi_sections .eh_frame}.
4763 On targets that support compact unwinding tables these can be generated
4764 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4766 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4767 which is used by the @value{TIC6X} target.
4769 The @code{.cfi_sections} directive can be repeated, with the same or different
4770 arguments, provided that CFI generation has not yet started. Once CFI
4771 generation has started however the section list is fixed and any attempts to
4772 redefine it will result in an error.
4774 @subsection @code{.cfi_startproc [simple]}
4775 @cindex @code{cfi_startproc} directive
4776 @code{.cfi_startproc} is used at the beginning of each function that
4777 should have an entry in @code{.eh_frame}. It initializes some internal
4778 data structures. Don't forget to close the function by
4779 @code{.cfi_endproc}.
4781 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4782 it also emits some architecture dependent initial CFI instructions.
4784 @subsection @code{.cfi_endproc}
4785 @cindex @code{cfi_endproc} directive
4786 @code{.cfi_endproc} is used at the end of a function where it closes its
4787 unwind entry previously opened by
4788 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4790 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4791 @cindex @code{cfi_personality} directive
4792 @code{.cfi_personality} defines personality routine and its encoding.
4793 @var{encoding} must be a constant determining how the personality
4794 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4795 argument is not present, otherwise second argument should be
4796 a constant or a symbol name. When using indirect encodings,
4797 the symbol provided should be the location where personality
4798 can be loaded from, not the personality routine itself.
4799 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4800 no personality routine.
4802 @subsection @code{.cfi_personality_id @var{id}}
4803 @cindex @code{cfi_personality_id} directive
4804 @code{cfi_personality_id} defines a personality routine by its index as
4805 defined in a compact unwinding format.
4806 Only valid when generating compact EH frames (i.e.
4807 with @code{.cfi_sections eh_frame_entry}.
4809 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4810 @cindex @code{cfi_fde_data} directive
4811 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4812 used for the current function. These are emitted inline in the
4813 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4814 in the @code{.gnu.extab} section otherwise.
4815 Only valid when generating compact EH frames (i.e.
4816 with @code{.cfi_sections eh_frame_entry}.
4818 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4819 @code{.cfi_lsda} defines LSDA and its encoding.
4820 @var{encoding} must be a constant determining how the LSDA
4821 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4822 argument is not present, otherwise the second argument should be a constant
4823 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4824 meaning that no LSDA is present.
4826 @subsection @code{.cfi_inline_lsda} [@var{align}]
4827 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4828 switches to the corresponding @code{.gnu.extab} section.
4829 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4830 Only valid when generating compact EH frames (i.e.
4831 with @code{.cfi_sections eh_frame_entry}.
4833 The table header and unwinding opcodes will be generated at this point,
4834 so that they are immediately followed by the LSDA data. The symbol
4835 referenced by the @code{.cfi_lsda} directive should still be defined
4836 in case a fallback FDE based encoding is used. The LSDA data is terminated
4837 by a section directive.
4839 The optional @var{align} argument specifies the alignment required.
4840 The alignment is specified as a power of two, as with the
4841 @code{.p2align} directive.
4843 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4844 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4845 address from @var{register} and add @var{offset} to it}.
4847 @subsection @code{.cfi_def_cfa_register @var{register}}
4848 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4849 now on @var{register} will be used instead of the old one. Offset
4852 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4853 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4854 remains the same, but @var{offset} is new. Note that it is the
4855 absolute offset that will be added to a defined register to compute
4858 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4859 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4860 value that is added/subtracted from the previous offset.
4862 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4863 Previous value of @var{register} is saved at offset @var{offset} from
4866 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4867 Previous value of @var{register} is CFA + @var{offset}.
4869 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4870 Previous value of @var{register} is saved at offset @var{offset} from
4871 the current CFA register. This is transformed to @code{.cfi_offset}
4872 using the known displacement of the CFA register from the CFA.
4873 This is often easier to use, because the number will match the
4874 code it's annotating.
4876 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4877 Previous value of @var{register1} is saved in register @var{register2}.
4879 @subsection @code{.cfi_restore @var{register}}
4880 @code{.cfi_restore} says that the rule for @var{register} is now the
4881 same as it was at the beginning of the function, after all initial
4882 instruction added by @code{.cfi_startproc} were executed.
4884 @subsection @code{.cfi_undefined @var{register}}
4885 From now on the previous value of @var{register} can't be restored anymore.
4887 @subsection @code{.cfi_same_value @var{register}}
4888 Current value of @var{register} is the same like in the previous frame,
4889 i.e. no restoration needed.
4891 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4892 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4893 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4894 places them in the current row. This is useful for situations where you have
4895 multiple @code{.cfi_*} directives that need to be undone due to the control
4896 flow of the program. For example, we could have something like this (assuming
4897 the CFA is the value of @code{rbp}):
4907 .cfi_def_cfa %rsp, 8
4910 /* Do something else */
4913 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4914 to the instructions before @code{label}. This means we'd have to add multiple
4915 @code{.cfi} directives after @code{label} to recreate the original save
4916 locations of the registers, as well as setting the CFA back to the value of
4917 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4929 .cfi_def_cfa %rsp, 8
4933 /* Do something else */
4936 That way, the rules for the instructions after @code{label} will be the same
4937 as before the first @code{.cfi_restore} without having to use multiple
4938 @code{.cfi} directives.
4940 @subsection @code{.cfi_return_column @var{register}}
4941 Change return column @var{register}, i.e. the return address is either
4942 directly in @var{register} or can be accessed by rules for @var{register}.
4944 @subsection @code{.cfi_signal_frame}
4945 Mark current function as signal trampoline.
4947 @subsection @code{.cfi_window_save}
4948 SPARC register window has been saved.
4950 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4951 Allows the user to add arbitrary bytes to the unwind info. One
4952 might use this to add OS-specific CFI opcodes, or generic CFI
4953 opcodes that GAS does not yet support.
4955 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4956 The current value of @var{register} is @var{label}. The value of @var{label}
4957 will be encoded in the output file according to @var{encoding}; see the
4958 description of @code{.cfi_personality} for details on this encoding.
4960 The usefulness of equating a register to a fixed label is probably
4961 limited to the return address register. Here, it can be useful to
4962 mark a code segment that has only one return address which is reached
4963 by a direct branch and no copy of the return address exists in memory
4964 or another register.
4967 @section @code{.comm @var{symbol} , @var{length} }
4969 @cindex @code{comm} directive
4970 @cindex symbol, common
4971 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4972 common symbol in one object file may be merged with a defined or common symbol
4973 of the same name in another object file. If @code{@value{LD}} does not see a
4974 definition for the symbol--just one or more common symbols--then it will
4975 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4976 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4977 the same name, and they do not all have the same size, it will allocate space
4978 using the largest size.
4981 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4982 an optional third argument. This is the desired alignment of the symbol,
4983 specified for ELF as a byte boundary (for example, an alignment of 16 means
4984 that the least significant 4 bits of the address should be zero), and for PE
4985 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4986 boundary). The alignment must be an absolute expression, and it must be a
4987 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4988 common symbol, it will use the alignment when placing the symbol. If no
4989 alignment is specified, @command{@value{AS}} will set the alignment to the
4990 largest power of two less than or equal to the size of the symbol, up to a
4991 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4992 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4993 @samp{--section-alignment} option; image file sections in PE are aligned to
4994 multiples of 4096, which is far too large an alignment for ordinary variables.
4995 It is rather the default alignment for (non-debug) sections within object
4996 (@samp{*.o}) files, which are less strictly aligned.}.
5000 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
5001 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
5005 @section @code{.data @var{subsection}}
5006 @cindex @code{data} directive
5008 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
5009 end of the data subsection numbered @var{subsection} (which is an
5010 absolute expression). If @var{subsection} is omitted, it defaults
5014 @section @code{.dc[@var{size}] @var{expressions}}
5015 @cindex @code{dc} directive
5017 The @code{.dc} directive expects zero or more @var{expressions} separated by
5018 commas. These expressions are evaluated and their values inserted into the
5019 current section. The size of the emitted value depends upon the suffix to the
5020 @code{.dc} directive:
5024 Emits N-bit values, where N is the size of an address on the target system.
5028 Emits double precision floating-point values.
5030 Emits 32-bit values.
5032 Emits single precision floating-point values.
5034 Emits 16-bit values.
5035 Note - this is true even on targets where the @code{.word} directive would emit
5038 Emits long double precision floating-point values.
5041 If no suffix is used then @samp{.w} is assumed.
5043 The byte ordering is target dependent, as is the size and format of floating
5047 @section @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
5048 @cindex @code{dcb} directive
5049 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5050 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5051 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5052 @var{size} suffix, if present, must be one of:
5056 Emits single byte values.
5058 Emits double-precision floating point values.
5060 Emits 4-byte values.
5062 Emits single-precision floating point values.
5064 Emits 2-byte values.
5066 Emits long double-precision floating point values.
5069 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5071 The byte ordering is target dependent, as is the size and format of floating
5075 @section @code{.ds[@var{size}] @var{number} [,@var{fill}]}
5076 @cindex @code{ds} directive
5077 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5078 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5079 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5080 @var{size} suffix, if present, must be one of:
5084 Emits single byte values.
5086 Emits 8-byte values.
5088 Emits 4-byte values.
5090 Emits 12-byte values.
5092 Emits 4-byte values.
5094 Emits 2-byte values.
5096 Emits 12-byte values.
5099 Note - unlike the @code{.dcb} directive the @samp{.d}, @samp{.s} and @samp{.x}
5100 suffixes do not indicate that floating-point values are to be inserted.
5102 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5104 The byte ordering is target dependent.
5109 @section @code{.def @var{name}}
5111 @cindex @code{def} directive
5112 @cindex COFF symbols, debugging
5113 @cindex debugging COFF symbols
5114 Begin defining debugging information for a symbol @var{name}; the
5115 definition extends until the @code{.endef} directive is encountered.
5120 @section @code{.desc @var{symbol}, @var{abs-expression}}
5122 @cindex @code{desc} directive
5123 @cindex COFF symbol descriptor
5124 @cindex symbol descriptor, COFF
5125 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5126 to the low 16 bits of an absolute expression.
5129 The @samp{.desc} directive is not available when @command{@value{AS}} is
5130 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5131 object format. For the sake of compatibility, @command{@value{AS}} accepts
5132 it, but produces no output, when configured for COFF.
5138 @section @code{.dim}
5140 @cindex @code{dim} directive
5141 @cindex COFF auxiliary symbol information
5142 @cindex auxiliary symbol information, COFF
5143 This directive is generated by compilers to include auxiliary debugging
5144 information in the symbol table. It is only permitted inside
5145 @code{.def}/@code{.endef} pairs.
5149 @section @code{.double @var{flonums}}
5151 @cindex @code{double} directive
5152 @cindex floating point numbers (double)
5153 @code{.double} expects zero or more flonums, separated by commas. It
5154 assembles floating point numbers.
5156 The exact kind of floating point numbers emitted depends on how
5157 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5161 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5162 in @sc{ieee} format.
5167 @section @code{.eject}
5169 @cindex @code{eject} directive
5170 @cindex new page, in listings
5171 @cindex page, in listings
5172 @cindex listing control: new page
5173 Force a page break at this point, when generating assembly listings.
5176 @section @code{.else}
5178 @cindex @code{else} directive
5179 @code{.else} is part of the @command{@value{AS}} support for conditional
5180 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5181 of code to be assembled if the condition for the preceding @code{.if}
5185 @section @code{.elseif}
5187 @cindex @code{elseif} directive
5188 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5189 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5190 @code{.if} block that would otherwise fill the entire @code{.else} section.
5193 @section @code{.end}
5195 @cindex @code{end} directive
5196 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5197 process anything in the file past the @code{.end} directive.
5201 @section @code{.endef}
5203 @cindex @code{endef} directive
5204 This directive flags the end of a symbol definition begun with
5209 @section @code{.endfunc}
5210 @cindex @code{endfunc} directive
5211 @code{.endfunc} marks the end of a function specified with @code{.func}.
5214 @section @code{.endif}
5216 @cindex @code{endif} directive
5217 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5218 it marks the end of a block of code that is only assembled
5219 conditionally. @xref{If,,@code{.if}}.
5222 @section @code{.equ @var{symbol}, @var{expression}}
5224 @cindex @code{equ} directive
5225 @cindex assigning values to symbols
5226 @cindex symbols, assigning values to
5227 This directive sets the value of @var{symbol} to @var{expression}.
5228 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5231 The syntax for @code{equ} on the HPPA is
5232 @samp{@var{symbol} .equ @var{expression}}.
5236 The syntax for @code{equ} on the Z80 is
5237 @samp{@var{symbol} equ @var{expression}}.
5238 On the Z80 it is an error if @var{symbol} is already defined,
5239 but the symbol is not protected from later redefinition.
5240 Compare @ref{Equiv}.
5244 @section @code{.equiv @var{symbol}, @var{expression}}
5245 @cindex @code{equiv} directive
5246 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5247 the assembler will signal an error if @var{symbol} is already defined. Note a
5248 symbol which has been referenced but not actually defined is considered to be
5251 Except for the contents of the error message, this is roughly equivalent to
5258 plus it protects the symbol from later redefinition.
5261 @section @code{.eqv @var{symbol}, @var{expression}}
5262 @cindex @code{eqv} directive
5263 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5264 evaluate the expression or any part of it immediately. Instead each time
5265 the resulting symbol is used in an expression, a snapshot of its current
5269 @section @code{.err}
5270 @cindex @code{err} directive
5271 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5272 message and, unless the @option{-Z} option was used, it will not generate an
5273 object file. This can be used to signal an error in conditionally compiled code.
5276 @section @code{.error "@var{string}"}
5277 @cindex error directive
5279 Similarly to @code{.err}, this directive emits an error, but you can specify a
5280 string that will be emitted as the error message. If you don't specify the
5281 message, it defaults to @code{".error directive invoked in source file"}.
5282 @xref{Errors, ,Error and Warning Messages}.
5285 .error "This code has not been assembled and tested."
5289 @section @code{.exitm}
5290 Exit early from the current macro definition. @xref{Macro}.
5293 @section @code{.extern}
5295 @cindex @code{extern} directive
5296 @code{.extern} is accepted in the source program---for compatibility
5297 with other assemblers---but it is ignored. @command{@value{AS}} treats
5298 all undefined symbols as external.
5301 @section @code{.fail @var{expression}}
5303 @cindex @code{fail} directive
5304 Generates an error or a warning. If the value of the @var{expression} is 500
5305 or more, @command{@value{AS}} will print a warning message. If the value is less
5306 than 500, @command{@value{AS}} will print an error message. The message will
5307 include the value of @var{expression}. This can occasionally be useful inside
5308 complex nested macros or conditional assembly.
5311 @section @code{.file}
5312 @cindex @code{file} directive
5314 @ifclear no-file-dir
5315 There are two different versions of the @code{.file} directive. Targets
5316 that support DWARF2 line number information use the DWARF2 version of
5317 @code{.file}. Other targets use the default version.
5319 @subheading Default Version
5321 @cindex logical file name
5322 @cindex file name, logical
5323 This version of the @code{.file} directive tells @command{@value{AS}} that we
5324 are about to start a new logical file. The syntax is:
5330 @var{string} is the new file name. In general, the filename is
5331 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5332 to specify an empty file name, you must give the quotes--@code{""}. This
5333 statement may go away in future: it is only recognized to be compatible with
5334 old @command{@value{AS}} programs.
5336 @subheading DWARF2 Version
5339 When emitting DWARF2 line number information, @code{.file} assigns filenames
5340 to the @code{.debug_line} file name table. The syntax is:
5343 .file @var{fileno} @var{filename}
5346 The @var{fileno} operand should be a unique positive integer to use as the
5347 index of the entry in the table. The @var{filename} operand is a C string
5350 The detail of filename indices is exposed to the user because the filename
5351 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5352 information, and thus the user must know the exact indices that table
5356 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5358 @cindex @code{fill} directive
5359 @cindex writing patterns in memory
5360 @cindex patterns, writing in memory
5361 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5362 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5363 may be zero or more. @var{Size} may be zero or more, but if it is
5364 more than 8, then it is deemed to have the value 8, compatible with
5365 other people's assemblers. The contents of each @var{repeat} bytes
5366 is taken from an 8-byte number. The highest order 4 bytes are
5367 zero. The lowest order 4 bytes are @var{value} rendered in the
5368 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5369 Each @var{size} bytes in a repetition is taken from the lowest order
5370 @var{size} bytes of this number. Again, this bizarre behavior is
5371 compatible with other people's assemblers.
5373 @var{size} and @var{value} are optional.
5374 If the second comma and @var{value} are absent, @var{value} is
5375 assumed zero. If the first comma and following tokens are absent,
5376 @var{size} is assumed to be 1.
5379 @section @code{.float @var{flonums}}
5381 @cindex floating point numbers (single)
5382 @cindex @code{float} directive
5383 This directive assembles zero or more flonums, separated by commas. It
5384 has the same effect as @code{.single}.
5386 The exact kind of floating point numbers emitted depends on how
5387 @command{@value{AS}} is configured.
5388 @xref{Machine Dependencies}.
5392 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5393 in @sc{ieee} format.
5398 @section @code{.func @var{name}[,@var{label}]}
5399 @cindex @code{func} directive
5400 @code{.func} emits debugging information to denote function @var{name}, and
5401 is ignored unless the file is assembled with debugging enabled.
5402 Only @samp{--gstabs[+]} is currently supported.
5403 @var{label} is the entry point of the function and if omitted @var{name}
5404 prepended with the @samp{leading char} is used.
5405 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5406 All functions are currently defined to have @code{void} return type.
5407 The function must be terminated with @code{.endfunc}.
5410 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5412 @cindex @code{global} directive
5413 @cindex symbol, making visible to linker
5414 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5415 @var{symbol} in your partial program, its value is made available to
5416 other partial programs that are linked with it. Otherwise,
5417 @var{symbol} takes its attributes from a symbol of the same name
5418 from another file linked into the same program.
5420 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5421 compatibility with other assemblers.
5424 On the HPPA, @code{.global} is not always enough to make it accessible to other
5425 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5426 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5431 @section @code{.gnu_attribute @var{tag},@var{value}}
5432 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5435 @section @code{.hidden @var{names}}
5437 @cindex @code{hidden} directive
5439 This is one of the ELF visibility directives. The other two are
5440 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5441 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5443 This directive overrides the named symbols default visibility (which is set by
5444 their binding: local, global or weak). The directive sets the visibility to
5445 @code{hidden} which means that the symbols are not visible to other components.
5446 Such symbols are always considered to be @code{protected} as well.
5450 @section @code{.hword @var{expressions}}
5452 @cindex @code{hword} directive
5453 @cindex integers, 16-bit
5454 @cindex numbers, 16-bit
5455 @cindex sixteen bit integers
5456 This expects zero or more @var{expressions}, and emits
5457 a 16 bit number for each.
5460 This directive is a synonym for @samp{.short}; depending on the target
5461 architecture, it may also be a synonym for @samp{.word}.
5465 This directive is a synonym for @samp{.short}.
5468 This directive is a synonym for both @samp{.short} and @samp{.word}.
5473 @section @code{.ident}
5475 @cindex @code{ident} directive
5477 This directive is used by some assemblers to place tags in object files. The
5478 behavior of this directive varies depending on the target. When using the
5479 a.out object file format, @command{@value{AS}} simply accepts the directive for
5480 source-file compatibility with existing assemblers, but does not emit anything
5481 for it. When using COFF, comments are emitted to the @code{.comment} or
5482 @code{.rdata} section, depending on the target. When using ELF, comments are
5483 emitted to the @code{.comment} section.
5486 @section @code{.if @var{absolute expression}}
5488 @cindex conditional assembly
5489 @cindex @code{if} directive
5490 @code{.if} marks the beginning of a section of code which is only
5491 considered part of the source program being assembled if the argument
5492 (which must be an @var{absolute expression}) is non-zero. The end of
5493 the conditional section of code must be marked by @code{.endif}
5494 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5495 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5496 If you have several conditions to check, @code{.elseif} may be used to avoid
5497 nesting blocks if/else within each subsequent @code{.else} block.
5499 The following variants of @code{.if} are also supported:
5501 @cindex @code{ifdef} directive
5502 @item .ifdef @var{symbol}
5503 Assembles the following section of code if the specified @var{symbol}
5504 has been defined. Note a symbol which has been referenced but not yet defined
5505 is considered to be undefined.
5507 @cindex @code{ifb} directive
5508 @item .ifb @var{text}
5509 Assembles the following section of code if the operand is blank (empty).
5511 @cindex @code{ifc} directive
5512 @item .ifc @var{string1},@var{string2}
5513 Assembles the following section of code if the two strings are the same. The
5514 strings may be optionally quoted with single quotes. If they are not quoted,
5515 the first string stops at the first comma, and the second string stops at the
5516 end of the line. Strings which contain whitespace should be quoted. The
5517 string comparison is case sensitive.
5519 @cindex @code{ifeq} directive
5520 @item .ifeq @var{absolute expression}
5521 Assembles the following section of code if the argument is zero.
5523 @cindex @code{ifeqs} directive
5524 @item .ifeqs @var{string1},@var{string2}
5525 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5527 @cindex @code{ifge} directive
5528 @item .ifge @var{absolute expression}
5529 Assembles the following section of code if the argument is greater than or
5532 @cindex @code{ifgt} directive
5533 @item .ifgt @var{absolute expression}
5534 Assembles the following section of code if the argument is greater than zero.
5536 @cindex @code{ifle} directive
5537 @item .ifle @var{absolute expression}
5538 Assembles the following section of code if the argument is less than or equal
5541 @cindex @code{iflt} directive
5542 @item .iflt @var{absolute expression}
5543 Assembles the following section of code if the argument is less than zero.
5545 @cindex @code{ifnb} directive
5546 @item .ifnb @var{text}
5547 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5548 following section of code if the operand is non-blank (non-empty).
5550 @cindex @code{ifnc} directive
5551 @item .ifnc @var{string1},@var{string2}.
5552 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5553 following section of code if the two strings are not the same.
5555 @cindex @code{ifndef} directive
5556 @cindex @code{ifnotdef} directive
5557 @item .ifndef @var{symbol}
5558 @itemx .ifnotdef @var{symbol}
5559 Assembles the following section of code if the specified @var{symbol}
5560 has not been defined. Both spelling variants are equivalent. Note a symbol
5561 which has been referenced but not yet defined is considered to be undefined.
5563 @cindex @code{ifne} directive
5564 @item .ifne @var{absolute expression}
5565 Assembles the following section of code if the argument is not equal to zero
5566 (in other words, this is equivalent to @code{.if}).
5568 @cindex @code{ifnes} directive
5569 @item .ifnes @var{string1},@var{string2}
5570 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5571 following section of code if the two strings are not the same.
5575 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5577 @cindex @code{incbin} directive
5578 @cindex binary files, including
5579 The @code{incbin} directive includes @var{file} verbatim at the current
5580 location. You can control the search paths used with the @samp{-I} command-line
5581 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5584 The @var{skip} argument skips a number of bytes from the start of the
5585 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5586 read. Note that the data is not aligned in any way, so it is the user's
5587 responsibility to make sure that proper alignment is provided both before and
5588 after the @code{incbin} directive.
5591 @section @code{.include "@var{file}"}
5593 @cindex @code{include} directive
5594 @cindex supporting files, including
5595 @cindex files, including
5596 This directive provides a way to include supporting files at specified
5597 points in your source program. The code from @var{file} is assembled as
5598 if it followed the point of the @code{.include}; when the end of the
5599 included file is reached, assembly of the original file continues. You
5600 can control the search paths used with the @samp{-I} command-line option
5601 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5605 @section @code{.int @var{expressions}}
5607 @cindex @code{int} directive
5608 @cindex integers, 32-bit
5609 Expect zero or more @var{expressions}, of any section, separated by commas.
5610 For each expression, emit a number that, at run time, is the value of that
5611 expression. The byte order and bit size of the number depends on what kind
5612 of target the assembly is for.
5616 On most forms of the H8/300, @code{.int} emits 16-bit
5617 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5624 @section @code{.internal @var{names}}
5626 @cindex @code{internal} directive
5628 This is one of the ELF visibility directives. The other two are
5629 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5630 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5632 This directive overrides the named symbols default visibility (which is set by
5633 their binding: local, global or weak). The directive sets the visibility to
5634 @code{internal} which means that the symbols are considered to be @code{hidden}
5635 (i.e., not visible to other components), and that some extra, processor specific
5636 processing must also be performed upon the symbols as well.
5640 @section @code{.irp @var{symbol},@var{values}}@dots{}
5642 @cindex @code{irp} directive
5643 Evaluate a sequence of statements assigning different values to @var{symbol}.
5644 The sequence of statements starts at the @code{.irp} directive, and is
5645 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5646 set to @var{value}, and the sequence of statements is assembled. If no
5647 @var{value} is listed, the sequence of statements is assembled once, with
5648 @var{symbol} set to the null string. To refer to @var{symbol} within the
5649 sequence of statements, use @var{\symbol}.
5651 For example, assembling
5659 is equivalent to assembling
5667 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5670 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5672 @cindex @code{irpc} directive
5673 Evaluate a sequence of statements assigning different values to @var{symbol}.
5674 The sequence of statements starts at the @code{.irpc} directive, and is
5675 terminated by an @code{.endr} directive. For each character in @var{value},
5676 @var{symbol} is set to the character, and the sequence of statements is
5677 assembled. If no @var{value} is listed, the sequence of statements is
5678 assembled once, with @var{symbol} set to the null string. To refer to
5679 @var{symbol} within the sequence of statements, use @var{\symbol}.
5681 For example, assembling
5689 is equivalent to assembling
5697 For some caveats with the spelling of @var{symbol}, see also the discussion
5701 @section @code{.lcomm @var{symbol} , @var{length}}
5703 @cindex @code{lcomm} directive
5704 @cindex local common symbols
5705 @cindex symbols, local common
5706 Reserve @var{length} (an absolute expression) bytes for a local common
5707 denoted by @var{symbol}. The section and value of @var{symbol} are
5708 those of the new local common. The addresses are allocated in the bss
5709 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5710 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5711 not visible to @code{@value{LD}}.
5714 Some targets permit a third argument to be used with @code{.lcomm}. This
5715 argument specifies the desired alignment of the symbol in the bss section.
5719 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5720 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5724 @section @code{.lflags}
5726 @cindex @code{lflags} directive (ignored)
5727 @command{@value{AS}} accepts this directive, for compatibility with other
5728 assemblers, but ignores it.
5730 @ifclear no-line-dir
5732 @section @code{.line @var{line-number}}
5734 @cindex @code{line} directive
5735 @cindex logical line number
5737 Change the logical line number. @var{line-number} must be an absolute
5738 expression. The next line has that logical line number. Therefore any other
5739 statements on the current line (after a statement separator character) are
5740 reported as on logical line number @var{line-number} @minus{} 1. One day
5741 @command{@value{AS}} will no longer support this directive: it is recognized only
5742 for compatibility with existing assembler programs.
5745 Even though this is a directive associated with the @code{a.out} or
5746 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5747 when producing COFF output, and treats @samp{.line} as though it
5748 were the COFF @samp{.ln} @emph{if} it is found outside a
5749 @code{.def}/@code{.endef} pair.
5751 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5752 used by compilers to generate auxiliary symbol information for
5757 @section @code{.linkonce [@var{type}]}
5759 @cindex @code{linkonce} directive
5760 @cindex common sections
5761 Mark the current section so that the linker only includes a single copy of it.
5762 This may be used to include the same section in several different object files,
5763 but ensure that the linker will only include it once in the final output file.
5764 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5765 Duplicate sections are detected based on the section name, so it should be
5768 This directive is only supported by a few object file formats; as of this
5769 writing, the only object file format which supports it is the Portable
5770 Executable format used on Windows NT.
5772 The @var{type} argument is optional. If specified, it must be one of the
5773 following strings. For example:
5777 Not all types may be supported on all object file formats.
5781 Silently discard duplicate sections. This is the default.
5784 Warn if there are duplicate sections, but still keep only one copy.
5787 Warn if any of the duplicates have different sizes.
5790 Warn if any of the duplicates do not have exactly the same contents.
5794 @section @code{.list}
5796 @cindex @code{list} directive
5797 @cindex listing control, turning on
5798 Control (in conjunction with the @code{.nolist} directive) whether or
5799 not assembly listings are generated. These two directives maintain an
5800 internal counter (which is zero initially). @code{.list} increments the
5801 counter, and @code{.nolist} decrements it. Assembly listings are
5802 generated whenever the counter is greater than zero.
5804 By default, listings are disabled. When you enable them (with the
5805 @samp{-a} command-line option; @pxref{Invoking,,Command-Line Options}),
5806 the initial value of the listing counter is one.
5809 @section @code{.ln @var{line-number}}
5811 @cindex @code{ln} directive
5812 @ifclear no-line-dir
5813 @samp{.ln} is a synonym for @samp{.line}.
5816 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5817 must be an absolute expression. The next line has that logical
5818 line number, so any other statements on the current line (after a
5819 statement separator character @code{;}) are reported as on logical
5820 line number @var{line-number} @minus{} 1.
5824 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5825 @cindex @code{loc} directive
5826 When emitting DWARF2 line number information,
5827 the @code{.loc} directive will add a row to the @code{.debug_line} line
5828 number matrix corresponding to the immediately following assembly
5829 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5830 arguments will be applied to the @code{.debug_line} state machine before
5833 The @var{options} are a sequence of the following tokens in any order:
5837 This option will set the @code{basic_block} register in the
5838 @code{.debug_line} state machine to @code{true}.
5841 This option will set the @code{prologue_end} register in the
5842 @code{.debug_line} state machine to @code{true}.
5844 @item epilogue_begin
5845 This option will set the @code{epilogue_begin} register in the
5846 @code{.debug_line} state machine to @code{true}.
5848 @item is_stmt @var{value}
5849 This option will set the @code{is_stmt} register in the
5850 @code{.debug_line} state machine to @code{value}, which must be
5853 @item isa @var{value}
5854 This directive will set the @code{isa} register in the @code{.debug_line}
5855 state machine to @var{value}, which must be an unsigned integer.
5857 @item discriminator @var{value}
5858 This directive will set the @code{discriminator} register in the @code{.debug_line}
5859 state machine to @var{value}, which must be an unsigned integer.
5861 @item view @var{value}
5862 This option causes a row to be added to @code{.debug_line} in reference to the
5863 current address (which might not be the same as that of the following assembly
5864 instruction), and to associate @var{value} with the @code{view} register in the
5865 @code{.debug_line} state machine. If @var{value} is a label, both the
5866 @code{view} register and the label are set to the number of prior @code{.loc}
5867 directives at the same program location. If @var{value} is the literal
5868 @code{0}, the @code{view} register is set to zero, and the assembler asserts
5869 that there aren't any prior @code{.loc} directives at the same program
5870 location. If @var{value} is the literal @code{-0}, the assembler arrange for
5871 the @code{view} register to be reset in this row, even if there are prior
5872 @code{.loc} directives at the same program location.
5876 @node Loc_mark_labels
5877 @section @code{.loc_mark_labels @var{enable}}
5878 @cindex @code{loc_mark_labels} directive
5879 When emitting DWARF2 line number information,
5880 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5881 to the @code{.debug_line} line number matrix with the @code{basic_block}
5882 register in the state machine set whenever a code label is seen.
5883 The @var{enable} argument should be either 1 or 0, to enable or disable
5884 this function respectively.
5888 @section @code{.local @var{names}}
5890 @cindex @code{local} directive
5891 This directive, which is available for ELF targets, marks each symbol in
5892 the comma-separated list of @code{names} as a local symbol so that it
5893 will not be externally visible. If the symbols do not already exist,
5894 they will be created.
5896 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5897 accept an alignment argument, which is the case for most ELF targets,
5898 the @code{.local} directive can be used in combination with @code{.comm}
5899 (@pxref{Comm}) to define aligned local common data.
5903 @section @code{.long @var{expressions}}
5905 @cindex @code{long} directive
5906 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5909 @c no one seems to know what this is for or whether this description is
5910 @c what it really ought to do
5912 @section @code{.lsym @var{symbol}, @var{expression}}
5914 @cindex @code{lsym} directive
5915 @cindex symbol, not referenced in assembly
5916 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5917 the hash table, ensuring it cannot be referenced by name during the
5918 rest of the assembly. This sets the attributes of the symbol to be
5919 the same as the expression value:
5921 @var{other} = @var{descriptor} = 0
5922 @var{type} = @r{(section of @var{expression})}
5923 @var{value} = @var{expression}
5926 The new symbol is not flagged as external.
5930 @section @code{.macro}
5933 The commands @code{.macro} and @code{.endm} allow you to define macros that
5934 generate assembly output. For example, this definition specifies a macro
5935 @code{sum} that puts a sequence of numbers into memory:
5938 .macro sum from=0, to=5
5947 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5959 @item .macro @var{macname}
5960 @itemx .macro @var{macname} @var{macargs} @dots{}
5961 @cindex @code{macro} directive
5962 Begin the definition of a macro called @var{macname}. If your macro
5963 definition requires arguments, specify their names after the macro name,
5964 separated by commas or spaces. You can qualify the macro argument to
5965 indicate whether all invocations must specify a non-blank value (through
5966 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5967 (through @samp{:@code{vararg}}). You can supply a default value for any
5968 macro argument by following the name with @samp{=@var{deflt}}. You
5969 cannot define two macros with the same @var{macname} unless it has been
5970 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5971 definitions. For example, these are all valid @code{.macro} statements:
5975 Begin the definition of a macro called @code{comm}, which takes no
5978 @item .macro plus1 p, p1
5979 @itemx .macro plus1 p p1
5980 Either statement begins the definition of a macro called @code{plus1},
5981 which takes two arguments; within the macro definition, write
5982 @samp{\p} or @samp{\p1} to evaluate the arguments.
5984 @item .macro reserve_str p1=0 p2
5985 Begin the definition of a macro called @code{reserve_str}, with two
5986 arguments. The first argument has a default value, but not the second.
5987 After the definition is complete, you can call the macro either as
5988 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5989 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5990 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5991 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5993 @item .macro m p1:req, p2=0, p3:vararg
5994 Begin the definition of a macro called @code{m}, with at least three
5995 arguments. The first argument must always have a value specified, but
5996 not the second, which instead has a default value. The third formal
5997 will get assigned all remaining arguments specified at invocation time.
5999 When you call a macro, you can specify the argument values either by
6000 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
6001 @samp{sum to=17, from=9}.
6005 Note that since each of the @var{macargs} can be an identifier exactly
6006 as any other one permitted by the target architecture, there may be
6007 occasional problems if the target hand-crafts special meanings to certain
6008 characters when they occur in a special position. For example, if the colon
6009 (@code{:}) is generally permitted to be part of a symbol name, but the
6010 architecture specific code special-cases it when occurring as the final
6011 character of a symbol (to denote a label), then the macro parameter
6012 replacement code will have no way of knowing that and consider the whole
6013 construct (including the colon) an identifier, and check only this
6014 identifier for being the subject to parameter substitution. So for example
6015 this macro definition:
6023 might not work as expected. Invoking @samp{label foo} might not create a label
6024 called @samp{foo} but instead just insert the text @samp{\l:} into the
6025 assembler source, probably generating an error about an unrecognised
6028 Similarly problems might occur with the period character (@samp{.})
6029 which is often allowed inside opcode names (and hence identifier names). So
6030 for example constructing a macro to build an opcode from a base name and a
6031 length specifier like this:
6034 .macro opcode base length
6039 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
6040 instruction but instead generate some kind of error as the assembler tries to
6041 interpret the text @samp{\base.\length}.
6043 There are several possible ways around this problem:
6046 @item Insert white space
6047 If it is possible to use white space characters then this is the simplest
6056 @item Use @samp{\()}
6057 The string @samp{\()} can be used to separate the end of a macro argument from
6058 the following text. eg:
6061 .macro opcode base length
6066 @item Use the alternate macro syntax mode
6067 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
6068 used as a separator. eg:
6078 Note: this problem of correctly identifying string parameters to pseudo ops
6079 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
6080 and @code{.irpc} (@pxref{Irpc}) as well.
6083 @cindex @code{endm} directive
6084 Mark the end of a macro definition.
6087 @cindex @code{exitm} directive
6088 Exit early from the current macro definition.
6090 @cindex number of macros executed
6091 @cindex macros, count executed
6093 @command{@value{AS}} maintains a counter of how many macros it has
6094 executed in this pseudo-variable; you can copy that number to your
6095 output with @samp{\@@}, but @emph{only within a macro definition}.
6097 @item LOCAL @var{name} [ , @dots{} ]
6098 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
6099 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
6100 @xref{Altmacro,,@code{.altmacro}}.
6104 @section @code{.mri @var{val}}
6106 @cindex @code{mri} directive
6107 @cindex MRI mode, temporarily
6108 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
6109 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
6110 affects code assembled until the next @code{.mri} directive, or until the end
6111 of the file. @xref{M, MRI mode, MRI mode}.
6114 @section @code{.noaltmacro}
6115 Disable alternate macro mode. @xref{Altmacro}.
6118 @section @code{.nolist}
6120 @cindex @code{nolist} directive
6121 @cindex listing control, turning off
6122 Control (in conjunction with the @code{.list} directive) whether or
6123 not assembly listings are generated. These two directives maintain an
6124 internal counter (which is zero initially). @code{.list} increments the
6125 counter, and @code{.nolist} decrements it. Assembly listings are
6126 generated whenever the counter is greater than zero.
6129 @section @code{.nops @var{size}[, @var{control}]}
6131 @cindex @code{nops} directive
6132 @cindex filling memory with no-op instructions
6133 This directive emits @var{size} bytes filled with no-op instructions.
6134 @var{size} is absolute expression, which must be a positve value.
6135 @var{control} controls how no-op instructions should be generated. If
6136 the comma and @var{control} are omitted, @var{control} is assumed to be
6139 Note: For Intel 80386 and AMD x86-64 targets, @var{control} specifies
6140 the size limit of a no-op instruction. The valid values of @var{control}
6141 are between 0 and 4 in 16-bit mode, between 0 and 7 when tuning for
6142 older processors in 32-bit mode, between 0 and 11 in 64-bit mode or when
6143 tuning for newer processors in 32-bit mode. When 0 is used, the no-op
6144 instruction size limit is set to the maximum supported size.
6147 @section @code{.octa @var{bignums}}
6149 @c FIXME: double size emitted for "octa" on some? Or warn?
6150 @cindex @code{octa} directive
6151 @cindex integer, 16-byte
6152 @cindex sixteen byte integer
6153 This directive expects zero or more bignums, separated by commas. For each
6154 bignum, it emits a 16-byte integer.
6156 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6157 hence @emph{octa}-word for 16 bytes.
6160 @section @code{.offset @var{loc}}
6162 @cindex @code{offset} directive
6163 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6164 be an absolute expression. This directive may be useful for defining
6165 symbols with absolute values. Do not confuse it with the @code{.org}
6169 @section @code{.org @var{new-lc} , @var{fill}}
6171 @cindex @code{org} directive
6172 @cindex location counter, advancing
6173 @cindex advancing location counter
6174 @cindex current address, advancing
6175 Advance the location counter of the current section to
6176 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6177 expression with the same section as the current subsection. That is,
6178 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6179 wrong section, the @code{.org} directive is ignored. To be compatible
6180 with former assemblers, if the section of @var{new-lc} is absolute,
6181 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6182 is the same as the current subsection.
6184 @code{.org} may only increase the location counter, or leave it
6185 unchanged; you cannot use @code{.org} to move the location counter
6188 @c double negative used below "not undefined" because this is a specific
6189 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6190 @c section. doc@cygnus.com 18feb91
6191 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6192 may not be undefined. If you really detest this restriction we eagerly await
6193 a chance to share your improved assembler.
6195 Beware that the origin is relative to the start of the section, not
6196 to the start of the subsection. This is compatible with other
6197 people's assemblers.
6199 When the location counter (of the current subsection) is advanced, the
6200 intervening bytes are filled with @var{fill} which should be an
6201 absolute expression. If the comma and @var{fill} are omitted,
6202 @var{fill} defaults to zero.
6205 @section @code{.p2align[wl] [@var{abs-expr}[, @var{abs-expr}[, @var{abs-expr}]]]}
6207 @cindex padding the location counter given a power of two
6208 @cindex @code{p2align} directive
6209 Pad the location counter (in the current subsection) to a particular
6210 storage boundary. The first expression (which must be absolute) is the
6211 number of low-order zero bits the location counter must have after
6212 advancement. For example @samp{.p2align 3} advances the location
6213 counter until it is a multiple of 8. If the location counter is already a
6214 multiple of 8, no change is needed. If the expression is omitted then a
6215 default value of 0 is used, effectively disabling alignment requirements.
6217 The second expression (also absolute) gives the fill value to be stored in the
6218 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6219 padding bytes are normally zero. However, on most systems, if the section is
6220 marked as containing code and the fill value is omitted, the space is filled
6221 with no-op instructions.
6223 The third expression is also absolute, and is also optional. If it is present,
6224 it is the maximum number of bytes that should be skipped by this alignment
6225 directive. If doing the alignment would require skipping more bytes than the
6226 specified maximum, then the alignment is not done at all. You can omit the
6227 fill value (the second argument) entirely by simply using two commas after the
6228 required alignment; this can be useful if you want the alignment to be filled
6229 with no-op instructions when appropriate.
6231 @cindex @code{p2alignw} directive
6232 @cindex @code{p2alignl} directive
6233 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6234 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6235 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6236 fill pattern as a four byte longword value. For example, @code{.p2alignw
6237 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6238 filled in with the value 0x368d (the exact placement of the bytes depends upon
6239 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6244 @section @code{.popsection}
6246 @cindex @code{popsection} directive
6247 @cindex Section Stack
6248 This is one of the ELF section stack manipulation directives. The others are
6249 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6250 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6253 This directive replaces the current section (and subsection) with the top
6254 section (and subsection) on the section stack. This section is popped off the
6260 @section @code{.previous}
6262 @cindex @code{previous} directive
6263 @cindex Section Stack
6264 This is one of the ELF section stack manipulation directives. The others are
6265 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6266 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6267 (@pxref{PopSection}).
6269 This directive swaps the current section (and subsection) with most recently
6270 referenced section/subsection pair prior to this one. Multiple
6271 @code{.previous} directives in a row will flip between two sections (and their
6272 subsections). For example:
6284 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6290 # Now in section A subsection 1
6294 # Now in section B subsection 0
6297 # Now in section B subsection 1
6300 # Now in section B subsection 0
6304 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6305 section B and 0x9abc into subsection 1 of section B.
6307 In terms of the section stack, this directive swaps the current section with
6308 the top section on the section stack.
6312 @section @code{.print @var{string}}
6314 @cindex @code{print} directive
6315 @command{@value{AS}} will print @var{string} on the standard output during
6316 assembly. You must put @var{string} in double quotes.
6320 @section @code{.protected @var{names}}
6322 @cindex @code{protected} directive
6324 This is one of the ELF visibility directives. The other two are
6325 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6327 This directive overrides the named symbols default visibility (which is set by
6328 their binding: local, global or weak). The directive sets the visibility to
6329 @code{protected} which means that any references to the symbols from within the
6330 components that defines them must be resolved to the definition in that
6331 component, even if a definition in another component would normally preempt
6336 @section @code{.psize @var{lines} , @var{columns}}
6338 @cindex @code{psize} directive
6339 @cindex listing control: paper size
6340 @cindex paper size, for listings
6341 Use this directive to declare the number of lines---and, optionally, the
6342 number of columns---to use for each page, when generating listings.
6344 If you do not use @code{.psize}, listings use a default line-count
6345 of 60. You may omit the comma and @var{columns} specification; the
6346 default width is 200 columns.
6348 @command{@value{AS}} generates formfeeds whenever the specified number of
6349 lines is exceeded (or whenever you explicitly request one, using
6352 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6353 those explicitly specified with @code{.eject}.
6356 @section @code{.purgem @var{name}}
6358 @cindex @code{purgem} directive
6359 Undefine the macro @var{name}, so that later uses of the string will not be
6360 expanded. @xref{Macro}.
6364 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6366 @cindex @code{pushsection} directive
6367 @cindex Section Stack
6368 This is one of the ELF section stack manipulation directives. The others are
6369 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6370 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6373 This directive pushes the current section (and subsection) onto the
6374 top of the section stack, and then replaces the current section and
6375 subsection with @code{name} and @code{subsection}. The optional
6376 @code{flags}, @code{type} and @code{arguments} are treated the same
6377 as in the @code{.section} (@pxref{Section}) directive.
6381 @section @code{.quad @var{bignums}}
6383 @cindex @code{quad} directive
6384 @code{.quad} expects zero or more bignums, separated by commas. For
6385 each bignum, it emits
6387 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6388 warning message; and just takes the lowest order 8 bytes of the bignum.
6389 @cindex eight-byte integer
6390 @cindex integer, 8-byte
6392 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6393 hence @emph{quad}-word for 8 bytes.
6396 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6397 warning message; and just takes the lowest order 16 bytes of the bignum.
6398 @cindex sixteen-byte integer
6399 @cindex integer, 16-byte
6403 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6405 @cindex @code{reloc} directive
6406 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6407 @var{expression}. If @var{offset} is a number, the relocation is generated in
6408 the current section. If @var{offset} is an expression that resolves to a
6409 symbol plus offset, the relocation is generated in the given symbol's section.
6410 @var{expression}, if present, must resolve to a symbol plus addend or to an
6411 absolute value, but note that not all targets support an addend. e.g. ELF REL
6412 targets such as i386 store an addend in the section contents rather than in the
6413 relocation. This low level interface does not support addends stored in the
6417 @section @code{.rept @var{count}}
6419 @cindex @code{rept} directive
6420 Repeat the sequence of lines between the @code{.rept} directive and the next
6421 @code{.endr} directive @var{count} times.
6423 For example, assembling
6431 is equivalent to assembling
6439 A count of zero is allowed, but nothing is generated. Negative counts are not
6440 allowed and if encountered will be treated as if they were zero.
6443 @section @code{.sbttl "@var{subheading}"}
6445 @cindex @code{sbttl} directive
6446 @cindex subtitles for listings
6447 @cindex listing control: subtitle
6448 Use @var{subheading} as the title (third line, immediately after the
6449 title line) when generating assembly listings.
6451 This directive affects subsequent pages, as well as the current page if
6452 it appears within ten lines of the top of a page.
6456 @section @code{.scl @var{class}}
6458 @cindex @code{scl} directive
6459 @cindex symbol storage class (COFF)
6460 @cindex COFF symbol storage class
6461 Set the storage-class value for a symbol. This directive may only be
6462 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6463 whether a symbol is static or external, or it may record further
6464 symbolic debugging information.
6469 @section @code{.section @var{name}}
6471 @cindex named section
6472 Use the @code{.section} directive to assemble the following code into a section
6475 This directive is only supported for targets that actually support arbitrarily
6476 named sections; on @code{a.out} targets, for example, it is not accepted, even
6477 with a standard @code{a.out} section name.
6481 @c only print the extra heading if both COFF and ELF are set
6482 @subheading COFF Version
6485 @cindex @code{section} directive (COFF version)
6486 For COFF targets, the @code{.section} directive is used in one of the following
6490 .section @var{name}[, "@var{flags}"]
6491 .section @var{name}[, @var{subsection}]
6494 If the optional argument is quoted, it is taken as flags to use for the
6495 section. Each flag is a single character. The following flags are recognized:
6499 bss section (uninitialized data)
6501 section is not loaded
6507 exclude section from linking
6513 shared section (meaningful for PE targets)
6515 ignored. (For compatibility with the ELF version)
6517 section is not readable (meaningful for PE targets)
6519 single-digit power-of-two section alignment (GNU extension)
6522 If no flags are specified, the default flags depend upon the section name. If
6523 the section name is not recognized, the default will be for the section to be
6524 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6525 from the section, rather than adding them, so if they are used on their own it
6526 will be as if no flags had been specified at all.
6528 If the optional argument to the @code{.section} directive is not quoted, it is
6529 taken as a subsection number (@pxref{Sub-Sections}).
6534 @c only print the extra heading if both COFF and ELF are set
6535 @subheading ELF Version
6538 @cindex Section Stack
6539 This is one of the ELF section stack manipulation directives. The others are
6540 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6541 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6542 @code{.previous} (@pxref{Previous}).
6544 @cindex @code{section} directive (ELF version)
6545 For ELF targets, the @code{.section} directive is used like this:
6548 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6551 @anchor{Section Name Substitutions}
6552 @kindex --sectname-subst
6553 @cindex section name substitution
6554 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6555 argument may contain a substitution sequence. Only @code{%S} is supported
6556 at the moment, and substitutes the current section name. For example:
6559 .macro exception_code
6560 .section %S.exception
6561 [exception code here]
6576 The two @code{exception_code} invocations above would create the
6577 @code{.text.exception} and @code{.init.exception} sections respectively.
6578 This is useful e.g. to discriminate between ancillary sections that are
6579 tied to setup code to be discarded after use from ancillary sections that
6580 need to stay resident without having to define multiple @code{exception_code}
6581 macros just for that purpose.
6583 The optional @var{flags} argument is a quoted string which may contain any
6584 combination of the following characters:
6588 section is allocatable
6590 section is a GNU_MBIND section
6592 section is excluded from executable and shared library.
6594 section references a symbol defined in another section (the linked-to
6595 section) in the same file.
6599 section is executable
6601 section is mergeable
6603 section contains zero terminated strings
6605 section is a member of a section group
6607 section is used for thread-local-storage
6609 section is a member of the previously-current section's group, if any
6610 @item @code{<number>}
6611 a numeric value indicating the bits to be set in the ELF section header's flags
6612 field. Note - if one or more of the alphabetic characters described above is
6613 also included in the flags field, their bit values will be ORed into the
6615 @item @code{<target specific>}
6616 some targets extend this list with their own flag characters
6619 Note - once a section's flags have been set they cannot be changed. There are
6620 a few exceptions to this rule however. Processor and application specific
6621 flags can be added to an already defined section. The @code{.interp},
6622 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6623 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6624 section may have the executable (@code{x}) flag added.
6626 The optional @var{type} argument may contain one of the following constants:
6630 section contains data
6632 section does not contain data (i.e., section only occupies space)
6634 section contains data which is used by things other than the program
6636 section contains an array of pointers to init functions
6638 section contains an array of pointers to finish functions
6639 @item @@preinit_array
6640 section contains an array of pointers to pre-init functions
6641 @item @@@code{<number>}
6642 a numeric value to be set as the ELF section header's type field.
6643 @item @@@code{<target specific>}
6644 some targets extend this list with their own types
6647 Many targets only support the first three section types. The type may be
6648 enclosed in double quotes if necessary.
6650 Note on targets where the @code{@@} character is the start of a comment (eg
6651 ARM) then another character is used instead. For example the ARM port uses the
6654 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6655 special and have fixed types. Any attempt to declare them with a different
6656 type will generate an error from the assembler.
6658 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6659 be specified as well as an extra argument---@var{entsize}---like this:
6662 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6665 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6666 constants, each @var{entsize} octets long. Sections with both @code{M} and
6667 @code{S} must contain zero terminated strings where each character is
6668 @var{entsize} bytes long. The linker may remove duplicates within sections with
6669 the same name, same entity size and same flags. @var{entsize} must be an
6670 absolute expression. For sections with both @code{M} and @code{S}, a string
6671 which is a suffix of a larger string is considered a duplicate. Thus
6672 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6673 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6675 If @var{flags} contains the @code{o} flag, then the @var{type} argument
6676 must be present along with an additional field like this:
6679 .section @var{name},"@var{flags}"o,@@@var{type},@var{SymbolName}
6682 The @var{SymbolName} field specifies the symbol name which the section
6685 Note: If both the @var{M} and @var{o} flags are present, then the fields
6686 for the Merge flag should come first, like this:
6689 .section @var{name},"@var{flags}"Mo,@@@var{type},@var{entsize},@var{SymbolName}
6692 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6693 be present along with an additional field like this:
6696 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6699 The @var{GroupName} field specifies the name of the section group to which this
6700 particular section belongs. The optional linkage field can contain:
6704 indicates that only one copy of this section should be retained
6709 Note: if both the @var{M} and @var{G} flags are present then the fields for
6710 the Merge flag should come first, like this:
6713 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6716 If both @code{o} flag and @code{G} flag are present, then the
6717 @var{SymbolName} field for @code{o} comes first, like this:
6720 .section @var{name},"@var{flags}"oG,@@@var{type},@var{SymbolName},@var{GroupName}[,@var{linkage}]
6723 If @var{flags} contains the @code{?} symbol then it may not also contain the
6724 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6725 present. Instead, @code{?} says to consider the section that's current before
6726 this directive. If that section used @code{G}, then the new section will use
6727 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6728 If not, then the @code{?} symbol has no effect.
6730 The optional @var{unique,@code{<number>}} argument must come last. It
6731 assigns @var{@code{<number>}} as a unique section ID to distinguish
6732 different sections with the same section name like these:
6735 .section @var{name},"@var{flags}",@@@var{type},@var{unique,@code{<number>}}
6736 .section @var{name},"@var{flags}"G,@@@var{type},@var{GroupName},[@var{linkage}],@var{unique,@code{<number>}}
6737 .section @var{name},"@var{flags}"MG,@@@var{type},@var{entsize},@var{GroupName}[,@var{linkage}],@var{unique,@code{<number>}}
6740 The valid values of @var{@code{<number>}} are between 0 and 4294967295.
6742 If no flags are specified, the default flags depend upon the section name. If
6743 the section name is not recognized, the default will be for the section to have
6744 none of the above flags: it will not be allocated in memory, nor writable, nor
6745 executable. The section will contain data.
6747 For ELF targets, the assembler supports another type of @code{.section}
6748 directive for compatibility with the Solaris assembler:
6751 .section "@var{name}"[, @var{flags}...]
6754 Note that the section name is quoted. There may be a sequence of comma
6759 section is allocatable
6763 section is executable
6765 section is excluded from executable and shared library.
6767 section is used for thread local storage
6770 This directive replaces the current section and subsection. See the
6771 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6772 some examples of how this directive and the other section stack directives
6778 @section @code{.set @var{symbol}, @var{expression}}
6780 @cindex @code{set} directive
6781 @cindex symbol value, setting
6782 Set the value of @var{symbol} to @var{expression}. This
6783 changes @var{symbol}'s value and type to conform to
6784 @var{expression}. If @var{symbol} was flagged as external, it remains
6785 flagged (@pxref{Symbol Attributes}).
6787 You may @code{.set} a symbol many times in the same assembly provided that the
6788 values given to the symbol are constants. Values that are based on expressions
6789 involving other symbols are allowed, but some targets may restrict this to only
6790 being done once per assembly. This is because those targets do not set the
6791 addresses of symbols at assembly time, but rather delay the assignment until a
6792 final link is performed. This allows the linker a chance to change the code in
6793 the files, changing the location of, and the relative distance between, various
6796 If you @code{.set} a global symbol, the value stored in the object
6797 file is the last value stored into it.
6800 On Z80 @code{set} is a real instruction, use @code{.set} or
6801 @samp{@var{symbol} defl @var{expression}} instead.
6805 @section @code{.short @var{expressions}}
6807 @cindex @code{short} directive
6809 @code{.short} is normally the same as @samp{.word}.
6810 @xref{Word,,@code{.word}}.
6812 In some configurations, however, @code{.short} and @code{.word} generate
6813 numbers of different lengths. @xref{Machine Dependencies}.
6817 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6820 This expects zero or more @var{expressions}, and emits
6821 a 16 bit number for each.
6826 @section @code{.single @var{flonums}}
6828 @cindex @code{single} directive
6829 @cindex floating point numbers (single)
6830 This directive assembles zero or more flonums, separated by commas. It
6831 has the same effect as @code{.float}.
6833 The exact kind of floating point numbers emitted depends on how
6834 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6838 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6839 numbers in @sc{ieee} format.
6845 @section @code{.size}
6847 This directive is used to set the size associated with a symbol.
6851 @c only print the extra heading if both COFF and ELF are set
6852 @subheading COFF Version
6855 @cindex @code{size} directive (COFF version)
6856 For COFF targets, the @code{.size} directive is only permitted inside
6857 @code{.def}/@code{.endef} pairs. It is used like this:
6860 .size @var{expression}
6867 @c only print the extra heading if both COFF and ELF are set
6868 @subheading ELF Version
6871 @cindex @code{size} directive (ELF version)
6872 For ELF targets, the @code{.size} directive is used like this:
6875 .size @var{name} , @var{expression}
6878 This directive sets the size associated with a symbol @var{name}.
6879 The size in bytes is computed from @var{expression} which can make use of label
6880 arithmetic. This directive is typically used to set the size of function
6885 @ifclear no-space-dir
6887 @section @code{.skip @var{size} [,@var{fill}]}
6889 @cindex @code{skip} directive
6890 @cindex filling memory
6891 This directive emits @var{size} bytes, each of value @var{fill}. Both
6892 @var{size} and @var{fill} are absolute expressions. If the comma and
6893 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6898 @section @code{.sleb128 @var{expressions}}
6900 @cindex @code{sleb128} directive
6901 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6902 compact, variable length representation of numbers used by the DWARF
6903 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6905 @ifclear no-space-dir
6907 @section @code{.space @var{size} [,@var{fill}]}
6909 @cindex @code{space} directive
6910 @cindex filling memory
6911 This directive emits @var{size} bytes, each of value @var{fill}. Both
6912 @var{size} and @var{fill} are absolute expressions. If the comma
6913 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6918 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6919 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6920 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6921 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6929 @section @code{.stabd, .stabn, .stabs}
6931 @cindex symbolic debuggers, information for
6932 @cindex @code{stab@var{x}} directives
6933 There are three directives that begin @samp{.stab}.
6934 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6935 The symbols are not entered in the @command{@value{AS}} hash table: they
6936 cannot be referenced elsewhere in the source file.
6937 Up to five fields are required:
6941 This is the symbol's name. It may contain any character except
6942 @samp{\000}, so is more general than ordinary symbol names. Some
6943 debuggers used to code arbitrarily complex structures into symbol names
6947 An absolute expression. The symbol's type is set to the low 8 bits of
6948 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6949 and debuggers choke on silly bit patterns.
6952 An absolute expression. The symbol's ``other'' attribute is set to the
6953 low 8 bits of this expression.
6956 An absolute expression. The symbol's descriptor is set to the low 16
6957 bits of this expression.
6960 An absolute expression which becomes the symbol's value.
6963 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6964 or @code{.stabs} statement, the symbol has probably already been created;
6965 you get a half-formed symbol in your object file. This is
6966 compatible with earlier assemblers!
6969 @cindex @code{stabd} directive
6970 @item .stabd @var{type} , @var{other} , @var{desc}
6972 The ``name'' of the symbol generated is not even an empty string.
6973 It is a null pointer, for compatibility. Older assemblers used a
6974 null pointer so they didn't waste space in object files with empty
6977 The symbol's value is set to the location counter,
6978 relocatably. When your program is linked, the value of this symbol
6979 is the address of the location counter when the @code{.stabd} was
6982 @cindex @code{stabn} directive
6983 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6984 The name of the symbol is set to the empty string @code{""}.
6986 @cindex @code{stabs} directive
6987 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6988 All five fields are specified.
6994 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6995 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6997 @cindex string, copying to object file
6998 @cindex string8, copying to object file
6999 @cindex string16, copying to object file
7000 @cindex string32, copying to object file
7001 @cindex string64, copying to object file
7002 @cindex @code{string} directive
7003 @cindex @code{string8} directive
7004 @cindex @code{string16} directive
7005 @cindex @code{string32} directive
7006 @cindex @code{string64} directive
7008 Copy the characters in @var{str} to the object file. You may specify more than
7009 one string to copy, separated by commas. Unless otherwise specified for a
7010 particular machine, the assembler marks the end of each string with a 0 byte.
7011 You can use any of the escape sequences described in @ref{Strings,,Strings}.
7013 The variants @code{string16}, @code{string32} and @code{string64} differ from
7014 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
7015 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
7016 are stored in target endianness byte order.
7022 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
7023 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
7028 @section @code{.struct @var{expression}}
7030 @cindex @code{struct} directive
7031 Switch to the absolute section, and set the section offset to @var{expression},
7032 which must be an absolute expression. You might use this as follows:
7041 This would define the symbol @code{field1} to have the value 0, the symbol
7042 @code{field2} to have the value 4, and the symbol @code{field3} to have the
7043 value 8. Assembly would be left in the absolute section, and you would need to
7044 use a @code{.section} directive of some sort to change to some other section
7045 before further assembly.
7049 @section @code{.subsection @var{name}}
7051 @cindex @code{subsection} directive
7052 @cindex Section Stack
7053 This is one of the ELF section stack manipulation directives. The others are
7054 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
7055 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
7058 This directive replaces the current subsection with @code{name}. The current
7059 section is not changed. The replaced subsection is put onto the section stack
7060 in place of the then current top of stack subsection.
7065 @section @code{.symver}
7066 @cindex @code{symver} directive
7067 @cindex symbol versioning
7068 @cindex versions of symbols
7069 Use the @code{.symver} directive to bind symbols to specific version nodes
7070 within a source file. This is only supported on ELF platforms, and is
7071 typically used when assembling files to be linked into a shared library.
7072 There are cases where it may make sense to use this in objects to be bound
7073 into an application itself so as to override a versioned symbol from a
7076 For ELF targets, the @code{.symver} directive can be used like this:
7078 .symver @var{name}, @var{name2@@nodename}
7080 If the symbol @var{name} is defined within the file
7081 being assembled, the @code{.symver} directive effectively creates a symbol
7082 alias with the name @var{name2@@nodename}, and in fact the main reason that we
7083 just don't try and create a regular alias is that the @var{@@} character isn't
7084 permitted in symbol names. The @var{name2} part of the name is the actual name
7085 of the symbol by which it will be externally referenced. The name @var{name}
7086 itself is merely a name of convenience that is used so that it is possible to
7087 have definitions for multiple versions of a function within a single source
7088 file, and so that the compiler can unambiguously know which version of a
7089 function is being mentioned. The @var{nodename} portion of the alias should be
7090 the name of a node specified in the version script supplied to the linker when
7091 building a shared library. If you are attempting to override a versioned
7092 symbol from a shared library, then @var{nodename} should correspond to the
7093 nodename of the symbol you are trying to override.
7095 If the symbol @var{name} is not defined within the file being assembled, all
7096 references to @var{name} will be changed to @var{name2@@nodename}. If no
7097 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
7100 Another usage of the @code{.symver} directive is:
7102 .symver @var{name}, @var{name2@@@@nodename}
7104 In this case, the symbol @var{name} must exist and be defined within
7105 the file being assembled. It is similar to @var{name2@@nodename}. The
7106 difference is @var{name2@@@@nodename} will also be used to resolve
7107 references to @var{name2} by the linker.
7109 The third usage of the @code{.symver} directive is:
7111 .symver @var{name}, @var{name2@@@@@@nodename}
7113 When @var{name} is not defined within the
7114 file being assembled, it is treated as @var{name2@@nodename}. When
7115 @var{name} is defined within the file being assembled, the symbol
7116 name, @var{name}, will be changed to @var{name2@@@@nodename}.
7121 @section @code{.tag @var{structname}}
7123 @cindex COFF structure debugging
7124 @cindex structure debugging, COFF
7125 @cindex @code{tag} directive
7126 This directive is generated by compilers to include auxiliary debugging
7127 information in the symbol table. It is only permitted inside
7128 @code{.def}/@code{.endef} pairs. Tags are used to link structure
7129 definitions in the symbol table with instances of those structures.
7133 @section @code{.text @var{subsection}}
7135 @cindex @code{text} directive
7136 Tells @command{@value{AS}} to assemble the following statements onto the end of
7137 the text subsection numbered @var{subsection}, which is an absolute
7138 expression. If @var{subsection} is omitted, subsection number zero
7142 @section @code{.title "@var{heading}"}
7144 @cindex @code{title} directive
7145 @cindex listing control: title line
7146 Use @var{heading} as the title (second line, immediately after the
7147 source file name and pagenumber) when generating assembly listings.
7149 This directive affects subsequent pages, as well as the current page if
7150 it appears within ten lines of the top of a page.
7154 @section @code{.type}
7156 This directive is used to set the type of a symbol.
7160 @c only print the extra heading if both COFF and ELF are set
7161 @subheading COFF Version
7164 @cindex COFF symbol type
7165 @cindex symbol type, COFF
7166 @cindex @code{type} directive (COFF version)
7167 For COFF targets, this directive is permitted only within
7168 @code{.def}/@code{.endef} pairs. It is used like this:
7174 This records the integer @var{int} as the type attribute of a symbol table
7181 @c only print the extra heading if both COFF and ELF are set
7182 @subheading ELF Version
7185 @cindex ELF symbol type
7186 @cindex symbol type, ELF
7187 @cindex @code{type} directive (ELF version)
7188 For ELF targets, the @code{.type} directive is used like this:
7191 .type @var{name} , @var{type description}
7194 This sets the type of symbol @var{name} to be either a
7195 function symbol or an object symbol. There are five different syntaxes
7196 supported for the @var{type description} field, in order to provide
7197 compatibility with various other assemblers.
7199 Because some of the characters used in these syntaxes (such as @samp{@@} and
7200 @samp{#}) are comment characters for some architectures, some of the syntaxes
7201 below do not work on all architectures. The first variant will be accepted by
7202 the GNU assembler on all architectures so that variant should be used for
7203 maximum portability, if you do not need to assemble your code with other
7206 The syntaxes supported are:
7209 .type <name> STT_<TYPE_IN_UPPER_CASE>
7210 .type <name>,#<type>
7211 .type <name>,@@<type>
7212 .type <name>,%<type>
7213 .type <name>,"<type>"
7216 The types supported are:
7221 Mark the symbol as being a function name.
7224 @itemx gnu_indirect_function
7225 Mark the symbol as an indirect function when evaluated during reloc
7226 processing. (This is only supported on assemblers targeting GNU systems).
7230 Mark the symbol as being a data object.
7234 Mark the symbol as being a thread-local data object.
7238 Mark the symbol as being a common data object.
7242 Does not mark the symbol in any way. It is supported just for completeness.
7244 @item gnu_unique_object
7245 Marks the symbol as being a globally unique data object. The dynamic linker
7246 will make sure that in the entire process there is just one symbol with this
7247 name and type in use. (This is only supported on assemblers targeting GNU
7252 Changing between incompatible types other than from/to STT_NOTYPE will
7253 result in a diagnostic. An intermediate change to STT_NOTYPE will silence
7256 Note: Some targets support extra types in addition to those listed above.
7262 @section @code{.uleb128 @var{expressions}}
7264 @cindex @code{uleb128} directive
7265 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7266 compact, variable length representation of numbers used by the DWARF
7267 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7271 @section @code{.val @var{addr}}
7273 @cindex @code{val} directive
7274 @cindex COFF value attribute
7275 @cindex value attribute, COFF
7276 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7277 records the address @var{addr} as the value attribute of a symbol table
7283 @section @code{.version "@var{string}"}
7285 @cindex @code{version} directive
7286 This directive creates a @code{.note} section and places into it an ELF
7287 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7292 @section @code{.vtable_entry @var{table}, @var{offset}}
7294 @cindex @code{vtable_entry} directive
7295 This directive finds or creates a symbol @code{table} and creates a
7296 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7299 @section @code{.vtable_inherit @var{child}, @var{parent}}
7301 @cindex @code{vtable_inherit} directive
7302 This directive finds the symbol @code{child} and finds or creates the symbol
7303 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7304 parent whose addend is the value of the child symbol. As a special case the
7305 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7309 @section @code{.warning "@var{string}"}
7310 @cindex warning directive
7311 Similar to the directive @code{.error}
7312 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7315 @section @code{.weak @var{names}}
7317 @cindex @code{weak} directive
7318 This directive sets the weak attribute on the comma separated list of symbol
7319 @code{names}. If the symbols do not already exist, they will be created.
7321 On COFF targets other than PE, weak symbols are a GNU extension. This
7322 directive sets the weak attribute on the comma separated list of symbol
7323 @code{names}. If the symbols do not already exist, they will be created.
7325 On the PE target, weak symbols are supported natively as weak aliases.
7326 When a weak symbol is created that is not an alias, GAS creates an
7327 alternate symbol to hold the default value.
7330 @section @code{.weakref @var{alias}, @var{target}}
7332 @cindex @code{weakref} directive
7333 This directive creates an alias to the target symbol that enables the symbol to
7334 be referenced with weak-symbol semantics, but without actually making it weak.
7335 If direct references or definitions of the symbol are present, then the symbol
7336 will not be weak, but if all references to it are through weak references, the
7337 symbol will be marked as weak in the symbol table.
7339 The effect is equivalent to moving all references to the alias to a separate
7340 assembly source file, renaming the alias to the symbol in it, declaring the
7341 symbol as weak there, and running a reloadable link to merge the object files
7342 resulting from the assembly of the new source file and the old source file that
7343 had the references to the alias removed.
7345 The alias itself never makes to the symbol table, and is entirely handled
7346 within the assembler.
7349 @section @code{.word @var{expressions}}
7351 @cindex @code{word} directive
7352 This directive expects zero or more @var{expressions}, of any section,
7353 separated by commas.
7356 For each expression, @command{@value{AS}} emits a 32-bit number.
7359 For each expression, @command{@value{AS}} emits a 16-bit number.
7364 The size of the number emitted, and its byte order,
7365 depend on what target computer the assembly is for.
7368 @c on sparc the "special treatment to support compilers" doesn't
7369 @c happen---32-bit addressability, period; no long/short jumps.
7370 @ifset DIFF-TBL-KLUGE
7371 @cindex difference tables altered
7372 @cindex altered difference tables
7374 @emph{Warning: Special Treatment to support Compilers}
7378 Machines with a 32-bit address space, but that do less than 32-bit
7379 addressing, require the following special treatment. If the machine of
7380 interest to you does 32-bit addressing (or doesn't require it;
7381 @pxref{Machine Dependencies}), you can ignore this issue.
7384 In order to assemble compiler output into something that works,
7385 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7386 Directives of the form @samp{.word sym1-sym2} are often emitted by
7387 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7388 directive of the form @samp{.word sym1-sym2}, and the difference between
7389 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7390 creates a @dfn{secondary jump table}, immediately before the next label.
7391 This secondary jump table is preceded by a short-jump to the
7392 first byte after the secondary table. This short-jump prevents the flow
7393 of control from accidentally falling into the new table. Inside the
7394 table is a long-jump to @code{sym2}. The original @samp{.word}
7395 contains @code{sym1} minus the address of the long-jump to
7398 If there were several occurrences of @samp{.word sym1-sym2} before the
7399 secondary jump table, all of them are adjusted. If there was a
7400 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7401 long-jump to @code{sym4} is included in the secondary jump table,
7402 and the @code{.word} directives are adjusted to contain @code{sym3}
7403 minus the address of the long-jump to @code{sym4}; and so on, for as many
7404 entries in the original jump table as necessary.
7407 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7408 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7409 assembly language programmers.
7412 @c end DIFF-TBL-KLUGE
7414 @ifclear no-space-dir
7416 @section @code{.zero @var{size}}
7418 @cindex @code{zero} directive
7419 @cindex filling memory with zero bytes
7420 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7421 expression. This directive is actually an alias for the @samp{.skip} directive
7422 so it can take an optional second argument of the value to store in the bytes
7423 instead of zero. Using @samp{.zero} in this way would be confusing however.
7428 @section @code{.2byte @var{expression} [, @var{expression}]*}
7429 @cindex @code{2byte} directive
7430 @cindex two-byte integer
7431 @cindex integer, 2-byte
7433 This directive expects zero or more expressions, separated by commas. If there
7434 are no expressions then the directive does nothing. Otherwise each expression
7435 is evaluated in turn and placed in the next two bytes of the current output
7436 section, using the endian model of the target. If an expression will not fit
7437 in two bytes, a warning message is displayed and the least significant two
7438 bytes of the expression's value are used. If an expression cannot be evaluated
7439 at assembly time then relocations will be generated in order to compute the
7442 This directive does not apply any alignment before or after inserting the
7443 values. As a result of this, if relocations are generated, they may be
7444 different from those used for inserting values with a guaranteed alignment.
7446 This directive is only available for ELF targets,
7449 @section @code{.4byte @var{expression} [, @var{expression}]*}
7450 @cindex @code{4byte} directive
7451 @cindex four-byte integer
7452 @cindex integer, 4-byte
7454 Like the @option{.2byte} directive, except that it inserts unaligned, four byte
7455 long values into the output.
7458 @section @code{.8byte @var{expression} [, @var{expression}]*}
7459 @cindex @code{8byte} directive
7460 @cindex eight-byte integer
7461 @cindex integer, 8-byte
7463 Like the @option{.2byte} directive, except that it inserts unaligned, eight
7464 byte long bignum values into the output.
7469 @section Deprecated Directives
7471 @cindex deprecated directives
7472 @cindex obsolescent directives
7473 One day these directives won't work.
7474 They are included for compatibility with older assemblers.
7481 @node Object Attributes
7482 @chapter Object Attributes
7483 @cindex object attributes
7485 @command{@value{AS}} assembles source files written for a specific architecture
7486 into object files for that architecture. But not all object files are alike.
7487 Many architectures support incompatible variations. For instance, floating
7488 point arguments might be passed in floating point registers if the object file
7489 requires hardware floating point support---or floating point arguments might be
7490 passed in integer registers if the object file supports processors with no
7491 hardware floating point unit. Or, if two objects are built for different
7492 generations of the same architecture, the combination may require the
7493 newer generation at run-time.
7495 This information is useful during and after linking. At link time,
7496 @command{@value{LD}} can warn about incompatible object files. After link
7497 time, tools like @command{gdb} can use it to process the linked file
7500 Compatibility information is recorded as a series of object attributes. Each
7501 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7502 string, and indicates who sets the meaning of the tag. The tag is an integer,
7503 and indicates what property the attribute describes. The value may be a string
7504 or an integer, and indicates how the property affects this object. Missing
7505 attributes are the same as attributes with a zero value or empty string value.
7507 Object attributes were developed as part of the ABI for the ARM Architecture.
7508 The file format is documented in @cite{ELF for the ARM Architecture}.
7511 * GNU Object Attributes:: @sc{gnu} Object Attributes
7512 * Defining New Object Attributes:: Defining New Object Attributes
7515 @node GNU Object Attributes
7516 @section @sc{gnu} Object Attributes
7518 The @code{.gnu_attribute} directive records an object attribute
7519 with vendor @samp{gnu}.
7521 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7522 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7523 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7524 2} is set for architecture-independent attributes and clear for
7525 architecture-dependent ones.
7527 @subsection Common @sc{gnu} attributes
7529 These attributes are valid on all architectures.
7532 @item Tag_compatibility (32)
7533 The compatibility attribute takes an integer flag value and a vendor name. If
7534 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7535 then the file is only compatible with the named toolchain. If it is greater
7536 than 1, the file can only be processed by other toolchains under some private
7537 arrangement indicated by the flag value and the vendor name.
7540 @subsection MIPS Attributes
7543 @item Tag_GNU_MIPS_ABI_FP (4)
7544 The floating-point ABI used by this object file. The value will be:
7548 0 for files not affected by the floating-point ABI.
7550 1 for files using the hardware floating-point ABI with a standard
7551 double-precision FPU.
7553 2 for files using the hardware floating-point ABI with a single-precision FPU.
7555 3 for files using the software floating-point ABI.
7557 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7558 floating-point registers, 32-bit general-purpose registers and increased the
7559 number of callee-saved floating-point registers.
7561 5 for files using the hardware floating-point ABI with a double-precision FPU
7562 with either 32-bit or 64-bit floating-point registers and 32-bit
7563 general-purpose registers.
7565 6 for files using the hardware floating-point ABI with 64-bit floating-point
7566 registers and 32-bit general-purpose registers.
7568 7 for files using the hardware floating-point ABI with 64-bit floating-point
7569 registers, 32-bit general-purpose registers and a rule that forbids the
7570 direct use of odd-numbered single-precision floating-point registers.
7574 @subsection PowerPC Attributes
7577 @item Tag_GNU_Power_ABI_FP (4)
7578 The floating-point ABI used by this object file. The value will be:
7582 0 for files not affected by the floating-point ABI.
7584 1 for files using double-precision hardware floating-point ABI.
7586 2 for files using the software floating-point ABI.
7588 3 for files using single-precision hardware floating-point ABI.
7591 @item Tag_GNU_Power_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 general purpose registers to pass vectors.
7600 2 for files using AltiVec registers to pass vectors.
7602 3 for files using SPE registers to pass vectors.
7606 @subsection IBM z Systems Attributes
7609 @item Tag_GNU_S390_ABI_Vector (8)
7610 The vector ABI used by this object file. The value will be:
7614 0 for files not affected by the vector ABI.
7616 1 for files using software vector ABI.
7618 2 for files using hardware vector ABI.
7622 @subsection MSP430 Attributes
7625 @item Tag_GNU_MSP430_Data_Region (4)
7626 The data region used by this object file. The value will be:
7630 0 for files not using the large memory model.
7632 1 for files which have been compiled with the condition that all
7633 data is in the lower memory region, i.e. below address 0x10000.
7635 2 for files which allow data to be placed in the full 20-bit memory range.
7639 @node Defining New Object Attributes
7640 @section Defining New Object Attributes
7642 If you want to define a new @sc{gnu} object attribute, here are the places you
7643 will need to modify. New attributes should be discussed on the @samp{binutils}
7648 This manual, which is the official register of attributes.
7650 The header for your architecture @file{include/elf}, to define the tag.
7652 The @file{bfd} support file for your architecture, to merge the attribute
7653 and issue any appropriate link warnings.
7655 Test cases in @file{ld/testsuite} for merging and link warnings.
7657 @file{binutils/readelf.c} to display your attribute.
7659 GCC, if you want the compiler to mark the attribute automatically.
7665 @node Machine Dependencies
7666 @chapter Machine Dependent Features
7668 @cindex machine dependencies
7669 The machine instruction sets are (almost by definition) different on
7670 each machine where @command{@value{AS}} runs. Floating point representations
7671 vary as well, and @command{@value{AS}} often supports a few additional
7672 directives or command-line options for compatibility with other
7673 assemblers on a particular platform. Finally, some versions of
7674 @command{@value{AS}} support special pseudo-instructions for branch
7677 This chapter discusses most of these differences, though it does not
7678 include details on any machine's instruction set. For details on that
7679 subject, see the hardware manufacturer's manual.
7683 * AArch64-Dependent:: AArch64 Dependent Features
7686 * Alpha-Dependent:: Alpha Dependent Features
7689 * ARC-Dependent:: ARC Dependent Features
7692 * ARM-Dependent:: ARM Dependent Features
7695 * AVR-Dependent:: AVR Dependent Features
7698 * Blackfin-Dependent:: Blackfin Dependent Features
7701 * BPF-Dependent:: BPF Dependent Features
7704 * CR16-Dependent:: CR16 Dependent Features
7707 * CRIS-Dependent:: CRIS Dependent Features
7710 * C-SKY-Dependent:: C-SKY Dependent Features
7713 * D10V-Dependent:: D10V Dependent Features
7716 * D30V-Dependent:: D30V Dependent Features
7719 * Epiphany-Dependent:: EPIPHANY Dependent Features
7722 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7725 * HPPA-Dependent:: HPPA Dependent Features
7728 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7731 * IA-64-Dependent:: Intel IA-64 Dependent Features
7734 * IP2K-Dependent:: IP2K Dependent Features
7737 * LM32-Dependent:: LM32 Dependent Features
7740 * M32C-Dependent:: M32C Dependent Features
7743 * M32R-Dependent:: M32R Dependent Features
7746 * M68K-Dependent:: M680x0 Dependent Features
7749 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7752 * S12Z-Dependent:: S12Z Dependent Features
7755 * Meta-Dependent :: Meta Dependent Features
7758 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7761 * MIPS-Dependent:: MIPS Dependent Features
7764 * MMIX-Dependent:: MMIX Dependent Features
7767 * MSP430-Dependent:: MSP430 Dependent Features
7770 * NDS32-Dependent:: Andes NDS32 Dependent Features
7773 * NiosII-Dependent:: Altera Nios II Dependent Features
7776 * NS32K-Dependent:: NS32K Dependent Features
7779 * OpenRISC-Dependent:: OpenRISC 1000 Features
7782 * PDP-11-Dependent:: PDP-11 Dependent Features
7785 * PJ-Dependent:: picoJava Dependent Features
7788 * PPC-Dependent:: PowerPC Dependent Features
7791 * PRU-Dependent:: PRU Dependent Features
7794 * RISC-V-Dependent:: RISC-V Dependent Features
7797 * RL78-Dependent:: RL78 Dependent Features
7800 * RX-Dependent:: RX Dependent Features
7803 * S/390-Dependent:: IBM S/390 Dependent Features
7806 * SCORE-Dependent:: SCORE Dependent Features
7809 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7812 * Sparc-Dependent:: SPARC Dependent Features
7815 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7818 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7821 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7824 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7827 * V850-Dependent:: V850 Dependent Features
7830 * Vax-Dependent:: VAX Dependent Features
7833 * Visium-Dependent:: Visium Dependent Features
7836 * WebAssembly-Dependent:: WebAssembly Dependent Features
7839 * XGATE-Dependent:: XGATE Dependent Features
7842 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7845 * Xtensa-Dependent:: Xtensa Dependent Features
7848 * Z80-Dependent:: Z80 Dependent Features
7851 * Z8000-Dependent:: Z8000 Dependent Features
7858 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7859 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7860 @c peculiarity: to preserve cross-references, there must be a node called
7861 @c "Machine Dependencies". Hence the conditional nodenames in each
7862 @c major node below. Node defaulting in makeinfo requires adjacency of
7863 @c node and sectioning commands; hence the repetition of @chapter BLAH
7864 @c in both conditional blocks.
7867 @include c-aarch64.texi
7871 @include c-alpha.texi
7887 @include c-bfin.texi
7895 @include c-cr16.texi
7899 @include c-cris.texi
7903 @include c-csky.texi
7908 @node Machine Dependencies
7909 @chapter Machine Dependent Features
7911 The machine instruction sets are different on each Renesas chip family,
7912 and there are also some syntax differences among the families. This
7913 chapter describes the specific @command{@value{AS}} features for each
7917 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7918 * SH-Dependent:: Renesas SH Dependent Features
7925 @include c-d10v.texi
7929 @include c-d30v.texi
7933 @include c-epiphany.texi
7937 @include c-h8300.texi
7941 @include c-hppa.texi
7945 @include c-i386.texi
7949 @include c-ia64.texi
7953 @include c-ip2k.texi
7957 @include c-lm32.texi
7961 @include c-m32c.texi
7965 @include c-m32r.texi
7969 @include c-m68k.texi
7973 @include c-m68hc11.texi
7977 @include c-s12z.texi
7981 @include c-metag.texi
7985 @include c-microblaze.texi
7989 @include c-mips.texi
7993 @include c-mmix.texi
7997 @include c-msp430.texi
8001 @include c-nds32.texi
8005 @include c-nios2.texi
8009 @include c-ns32k.texi
8013 @include c-or1k.texi
8017 @include c-pdp11.texi
8033 @include c-riscv.texi
8037 @include c-rl78.texi
8045 @include c-s390.texi
8049 @include c-score.texi
8057 @include c-sparc.texi
8061 @include c-tic54x.texi
8065 @include c-tic6x.texi
8069 @include c-tilegx.texi
8073 @include c-tilepro.texi
8077 @include c-v850.texi
8085 @include c-visium.texi
8089 @include c-wasm32.texi
8093 @include c-xgate.texi
8097 @include c-xstormy16.texi
8101 @include c-xtensa.texi
8113 @c reverse effect of @down at top of generic Machine-Dep chapter
8117 @node Reporting Bugs
8118 @chapter Reporting Bugs
8119 @cindex bugs in assembler
8120 @cindex reporting bugs in assembler
8122 Your bug reports play an essential role in making @command{@value{AS}} reliable.
8124 Reporting a bug may help you by bringing a solution to your problem, or it may
8125 not. But in any case the principal function of a bug report is to help the
8126 entire community by making the next version of @command{@value{AS}} work better.
8127 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
8129 In order for a bug report to serve its purpose, you must include the
8130 information that enables us to fix the bug.
8133 * Bug Criteria:: Have you found a bug?
8134 * Bug Reporting:: How to report bugs
8138 @section Have You Found a Bug?
8139 @cindex bug criteria
8141 If you are not sure whether you have found a bug, here are some guidelines:
8144 @cindex fatal signal
8145 @cindex assembler crash
8146 @cindex crash of assembler
8148 If the assembler gets a fatal signal, for any input whatever, that is a
8149 @command{@value{AS}} bug. Reliable assemblers never crash.
8151 @cindex error on valid input
8153 If @command{@value{AS}} produces an error message for valid input, that is a bug.
8155 @cindex invalid input
8157 If @command{@value{AS}} does not produce an error message for invalid input, that
8158 is a bug. However, you should note that your idea of ``invalid input'' might
8159 be our idea of ``an extension'' or ``support for traditional practice''.
8162 If you are an experienced user of assemblers, your suggestions for improvement
8163 of @command{@value{AS}} are welcome in any case.
8167 @section How to Report Bugs
8169 @cindex assembler bugs, reporting
8171 A number of companies and individuals offer support for @sc{gnu} products. If
8172 you obtained @command{@value{AS}} from a support organization, we recommend you
8173 contact that organization first.
8175 You can find contact information for many support companies and
8176 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
8180 In any event, we also recommend that you send bug reports for @command{@value{AS}}
8184 The fundamental principle of reporting bugs usefully is this:
8185 @strong{report all the facts}. If you are not sure whether to state a
8186 fact or leave it out, state it!
8188 Often people omit facts because they think they know what causes the problem
8189 and assume that some details do not matter. Thus, you might assume that the
8190 name of a symbol you use in an example does not matter. Well, probably it does
8191 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
8192 happens to fetch from the location where that name is stored in memory;
8193 perhaps, if the name were different, the contents of that location would fool
8194 the assembler into doing the right thing despite the bug. Play it safe and
8195 give a specific, complete example. That is the easiest thing for you to do,
8196 and the most helpful.
8198 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
8199 it is new to us. Therefore, always write your bug reports on the assumption
8200 that the bug has not been reported previously.
8202 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8203 bell?'' This cannot help us fix a bug, so it is basically useless. We
8204 respond by asking for enough details to enable us to investigate.
8205 You might as well expedite matters by sending them to begin with.
8207 To enable us to fix the bug, you should include all these things:
8211 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8212 it with the @samp{--version} argument.
8214 Without this, we will not know whether there is any point in looking for
8215 the bug in the current version of @command{@value{AS}}.
8218 Any patches you may have applied to the @command{@value{AS}} source.
8221 The type of machine you are using, and the operating system name and
8225 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8229 The command arguments you gave the assembler to assemble your example and
8230 observe the bug. To guarantee you will not omit something important, list them
8231 all. A copy of the Makefile (or the output from make) is sufficient.
8233 If we were to try to guess the arguments, we would probably guess wrong
8234 and then we might not encounter the bug.
8237 A complete input file that will reproduce the bug. If the bug is observed when
8238 the assembler is invoked via a compiler, send the assembler source, not the
8239 high level language source. Most compilers will produce the assembler source
8240 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8241 the options @samp{-v --save-temps}; this will save the assembler source in a
8242 file with an extension of @file{.s}, and also show you exactly how
8243 @command{@value{AS}} is being run.
8246 A description of what behavior you observe that you believe is
8247 incorrect. For example, ``It gets a fatal signal.''
8249 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8250 will certainly notice it. But if the bug is incorrect output, we might not
8251 notice unless it is glaringly wrong. You might as well not give us a chance to
8254 Even if the problem you experience is a fatal signal, you should still say so
8255 explicitly. Suppose something strange is going on, such as, your copy of
8256 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8257 library on your system. (This has happened!) Your copy might crash and ours
8258 would not. If you told us to expect a crash, then when ours fails to crash, we
8259 would know that the bug was not happening for us. If you had not told us to
8260 expect a crash, then we would not be able to draw any conclusion from our
8264 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8265 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8266 option. Always send diffs from the old file to the new file. If you even
8267 discuss something in the @command{@value{AS}} source, refer to it by context, not
8270 The line numbers in our development sources will not match those in your
8271 sources. Your line numbers would convey no useful information to us.
8274 Here are some things that are not necessary:
8278 A description of the envelope of the bug.
8280 Often people who encounter a bug spend a lot of time investigating
8281 which changes to the input file will make the bug go away and which
8282 changes will not affect it.
8284 This is often time consuming and not very useful, because the way we
8285 will find the bug is by running a single example under the debugger
8286 with breakpoints, not by pure deduction from a series of examples.
8287 We recommend that you save your time for something else.
8289 Of course, if you can find a simpler example to report @emph{instead}
8290 of the original one, that is a convenience for us. Errors in the
8291 output will be easier to spot, running under the debugger will take
8292 less time, and so on.
8294 However, simplification is not vital; if you do not want to do this,
8295 report the bug anyway and send us the entire test case you used.
8298 A patch for the bug.
8300 A patch for the bug does help us if it is a good one. But do not omit
8301 the necessary information, such as the test case, on the assumption that
8302 a patch is all we need. We might see problems with your patch and decide
8303 to fix the problem another way, or we might not understand it at all.
8305 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8306 construct an example that will make the program follow a certain path through
8307 the code. If you do not send us the example, we will not be able to construct
8308 one, so we will not be able to verify that the bug is fixed.
8310 And if we cannot understand what bug you are trying to fix, or why your
8311 patch should be an improvement, we will not install it. A test case will
8312 help us to understand.
8315 A guess about what the bug is or what it depends on.
8317 Such guesses are usually wrong. Even we cannot guess right about such
8318 things without first using the debugger to find the facts.
8321 @node Acknowledgements
8322 @chapter Acknowledgements
8324 If you have contributed to GAS and your name isn't listed here,
8325 it is not meant as a slight. We just don't know about it. Send mail to the
8326 maintainer, and we'll correct the situation. Currently
8328 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8330 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8333 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8334 information and the 68k series machines, most of the preprocessing pass, and
8335 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8337 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8338 many bug fixes, including merging support for several processors, breaking GAS
8339 up to handle multiple object file format back ends (including heavy rewrite,
8340 testing, an integration of the coff and b.out back ends), adding configuration
8341 including heavy testing and verification of cross assemblers and file splits
8342 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8343 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8344 port (including considerable amounts of reverse engineering), a SPARC opcode
8345 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8346 assertions and made them work, much other reorganization, cleanup, and lint.
8348 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8349 in format-specific I/O modules.
8351 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8352 has done much work with it since.
8354 The Intel 80386 machine description was written by Eliot Dresselhaus.
8356 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8358 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8359 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8361 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8362 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8363 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8364 support a.out format.
8366 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8367 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8368 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8369 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8372 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8373 simplified the configuration of which versions accept which directives. He
8374 updated the 68k machine description so that Motorola's opcodes always produced
8375 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8376 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8377 cross-compilation support, and one bug in relaxation that took a week and
8378 required the proverbial one-bit fix.
8380 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8381 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8382 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8383 PowerPC assembler, and made a few other minor patches.
8385 Steve Chamberlain made GAS able to generate listings.
8387 Hewlett-Packard contributed support for the HP9000/300.
8389 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8390 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8391 formats). This work was supported by both the Center for Software Science at
8392 the University of Utah and Cygnus Support.
8394 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8395 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8396 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8397 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8398 and some initial 64-bit support).
8400 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8402 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8403 support for openVMS/Alpha.
8405 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8408 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8409 Inc.@: added support for Xtensa processors.
8411 Several engineers at Cygnus Support have also provided many small bug fixes and
8412 configuration enhancements.
8414 Jon Beniston added support for the Lattice Mico32 architecture.
8416 Many others have contributed large or small bugfixes and enhancements. If
8417 you have contributed significant work and are not mentioned on this list, and
8418 want to be, let us know. Some of the history has been lost; we are not
8419 intentionally leaving anyone out.
8421 @node GNU Free Documentation License
8422 @appendix GNU Free Documentation License
8426 @unnumbered AS Index