1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (C) 1991-2019 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{} 1991-2019 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{} 1991-2019 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{--help}] [@b{-I} @var{dir}] [@b{-J}]
235 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
236 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
237 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}]
238 [@b{--no-pad-sections}]
239 [@b{-o} @var{objfile}] [@b{-R}]
240 [@b{--hash-size}=@var{NUM}] [@b{--reduce-memory-overheads}]
242 [@b{-v}] [@b{-version}] [@b{--version}]
243 [@b{-W}] [@b{--warn}] [@b{--fatal-warnings}] [@b{-w}] [@b{-x}]
244 [@b{-Z}] [@b{@@@var{FILE}}]
245 [@b{--sectname-subst}] [@b{--size-check=[error|warning]}]
246 [@b{--elf-stt-common=[no|yes]}]
247 [@b{--generate-missing-build-notes=[no|yes]}]
248 [@b{--target-help}] [@var{target-options}]
249 [@b{--}|@var{files} @dots{}]
252 @c Target dependent options are listed below. Keep the list sorted.
253 @c Add an empty line for separation.
257 @emph{Target AArch64 options:}
259 [@b{-mabi}=@var{ABI}]
263 @emph{Target Alpha options:}
265 [@b{-mdebug} | @b{-no-mdebug}]
266 [@b{-replace} | @b{-noreplace}]
267 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
268 [@b{-F}] [@b{-32addr}]
272 @emph{Target ARC options:}
273 [@b{-mcpu=@var{cpu}}]
274 [@b{-mA6}|@b{-mARC600}|@b{-mARC601}|@b{-mA7}|@b{-mARC700}|@b{-mEM}|@b{-mHS}]
281 @emph{Target ARM options:}
282 @c Don't document the deprecated options
283 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
284 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
285 [@b{-mfpu}=@var{floating-point-format}]
286 [@b{-mfloat-abi}=@var{abi}]
287 [@b{-meabi}=@var{ver}]
290 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
291 @b{-mapcs-reentrant}]
292 [@b{-mthumb-interwork}] [@b{-k}]
296 @emph{Target Blackfin options:}
297 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
304 @emph{Target BPF options:}
309 @emph{Target CRIS options:}
310 [@b{--underscore} | @b{--no-underscore}]
312 [@b{--emulation=criself} | @b{--emulation=crisaout}]
313 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
314 @c Deprecated -- deliberately not documented.
319 @emph{Target C-SKY options:}
320 [@b{-march=@var{arch}}] [@b{-mcpu=@var{cpu}}]
321 [@b{-EL}] [@b{-mlittle-endian}] [@b{-EB}] [@b{-mbig-endian}]
322 [@b{-fpic}] [@b{-pic}]
323 [@b{-mljump}] [@b{-mno-ljump}]
324 [@b{-force2bsr}] [@b{-mforce2bsr}] [@b{-no-force2bsr}] [@b{-mno-force2bsr}]
325 [@b{-jsri2bsr}] [@b{-mjsri2bsr}] [@b{-no-jsri2bsr }] [@b{-mno-jsri2bsr}]
326 [@b{-mnolrw }] [@b{-mno-lrw}]
327 [@b{-melrw}] [@b{-mno-elrw}]
328 [@b{-mlaf }] [@b{-mliterals-after-func}]
329 [@b{-mno-laf}] [@b{-mno-literals-after-func}]
330 [@b{-mlabr}] [@b{-mliterals-after-br}]
331 [@b{-mno-labr}] [@b{-mnoliterals-after-br}]
332 [@b{-mistack}] [@b{-mno-istack}]
333 [@b{-mhard-float}] [@b{-mmp}] [@b{-mcp}] [@b{-mcache}]
334 [@b{-msecurity}] [@b{-mtrust}]
335 [@b{-mdsp}] [@b{-medsp}] [@b{-mvdsp}]
339 @emph{Target D10V options:}
344 @emph{Target D30V options:}
345 [@b{-O}|@b{-n}|@b{-N}]
349 @emph{Target EPIPHANY options:}
350 [@b{-mepiphany}|@b{-mepiphany16}]
354 @emph{Target H8/300 options:}
358 @c HPPA has no machine-dependent assembler options (yet).
362 @emph{Target i386 options:}
363 [@b{--32}|@b{--x32}|@b{--64}] [@b{-n}]
364 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
368 @emph{Target IA-64 options:}
369 [@b{-mconstant-gp}|@b{-mauto-pic}]
370 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
372 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
373 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
374 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
375 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
379 @emph{Target IP2K options:}
380 [@b{-mip2022}|@b{-mip2022ext}]
384 @emph{Target M32C options:}
385 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
389 @emph{Target M32R options:}
390 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
395 @emph{Target M680X0 options:}
396 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
400 @emph{Target M68HC11 options:}
401 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}|@b{-mm9s12x}|@b{-mm9s12xg}]
402 [@b{-mshort}|@b{-mlong}]
403 [@b{-mshort-double}|@b{-mlong-double}]
404 [@b{--force-long-branches}] [@b{--short-branches}]
405 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
406 [@b{--print-opcodes}] [@b{--generate-example}]
410 @emph{Target MCORE options:}
411 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
412 [@b{-mcpu=[210|340]}]
416 @emph{Target Meta options:}
417 [@b{-mcpu=@var{cpu}}] [@b{-mfpu=@var{cpu}}] [@b{-mdsp=@var{cpu}}]
420 @emph{Target MICROBLAZE options:}
421 @c MicroBlaze has no machine-dependent assembler options.
425 @emph{Target MIPS options:}
426 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
427 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
428 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
429 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
430 [@b{-mfp64}] [@b{-mgp64}] [@b{-mfpxx}]
431 [@b{-modd-spreg}] [@b{-mno-odd-spreg}]
432 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
433 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
434 [@b{-mips32r3}] [@b{-mips32r5}] [@b{-mips32r6}] [@b{-mips64}] [@b{-mips64r2}]
435 [@b{-mips64r3}] [@b{-mips64r5}] [@b{-mips64r6}]
436 [@b{-construct-floats}] [@b{-no-construct-floats}]
437 [@b{-mignore-branch-isa}] [@b{-mno-ignore-branch-isa}]
438 [@b{-mnan=@var{encoding}}]
439 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
440 [@b{-mips16}] [@b{-no-mips16}]
441 [@b{-mmips16e2}] [@b{-mno-mips16e2}]
442 [@b{-mmicromips}] [@b{-mno-micromips}]
443 [@b{-msmartmips}] [@b{-mno-smartmips}]
444 [@b{-mips3d}] [@b{-no-mips3d}]
445 [@b{-mdmx}] [@b{-no-mdmx}]
446 [@b{-mdsp}] [@b{-mno-dsp}]
447 [@b{-mdspr2}] [@b{-mno-dspr2}]
448 [@b{-mdspr3}] [@b{-mno-dspr3}]
449 [@b{-mmsa}] [@b{-mno-msa}]
450 [@b{-mxpa}] [@b{-mno-xpa}]
451 [@b{-mmt}] [@b{-mno-mt}]
452 [@b{-mmcu}] [@b{-mno-mcu}]
453 [@b{-mcrc}] [@b{-mno-crc}]
454 [@b{-mginv}] [@b{-mno-ginv}]
455 [@b{-mloongson-mmi}] [@b{-mno-loongson-mmi}]
456 [@b{-mloongson-cam}] [@b{-mno-loongson-cam}]
457 [@b{-mloongson-ext}] [@b{-mno-loongson-ext}]
458 [@b{-mloongson-ext2}] [@b{-mno-loongson-ext2}]
459 [@b{-minsn32}] [@b{-mno-insn32}]
460 [@b{-mfix7000}] [@b{-mno-fix7000}]
461 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
462 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
463 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
464 [@b{-mfix-r5900}] [@b{-mno-fix-r5900}]
465 [@b{-mdebug}] [@b{-no-mdebug}]
466 [@b{-mpdr}] [@b{-mno-pdr}]
470 @emph{Target MMIX options:}
471 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
472 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
473 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
474 [@b{--linker-allocated-gregs}]
478 @emph{Target Nios II options:}
479 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
484 @emph{Target NDS32 options:}
485 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
486 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
487 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
488 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
489 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
490 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
491 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
495 @c OpenRISC has no machine-dependent assembler options.
499 @emph{Target PDP11 options:}
500 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
501 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
502 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
506 @emph{Target picoJava options:}
511 @emph{Target PowerPC options:}
513 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
514 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mgekko}|
515 @b{-mbroadway}|@b{-mppc64}|@b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|
516 @b{-me6500}|@b{-mppc64bridge}|@b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|
517 @b{-mpower6}|@b{-mpwr6}|@b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-mpower9}|@b{-mpwr9}@b{-ma2}|
518 @b{-mcell}|@b{-mspe}|@b{-mspe2}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
519 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
520 [@b{-mregnames}|@b{-mno-regnames}]
521 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
522 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
523 [@b{-msolaris}|@b{-mno-solaris}]
524 [@b{-nops=@var{count}}]
528 @emph{Target PRU options:}
531 [@b{-mno-warn-regname-label}]
535 @emph{Target RISC-V options:}
536 [@b{-fpic}|@b{-fPIC}|@b{-fno-pic}]
537 [@b{-march}=@var{ISA}]
538 [@b{-mabi}=@var{ABI}]
542 @emph{Target RL78 options:}
544 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
548 @emph{Target RX options:}
549 [@b{-mlittle-endian}|@b{-mbig-endian}]
550 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
551 [@b{-muse-conventional-section-names}]
552 [@b{-msmall-data-limit}]
555 [@b{-mint-register=@var{number}}]
556 [@b{-mgcc-abi}|@b{-mrx-abi}]
560 @emph{Target s390 options:}
561 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
562 [@b{-mregnames}|@b{-mno-regnames}]
563 [@b{-mwarn-areg-zero}]
567 @emph{Target SCORE options:}
568 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
569 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
570 [@b{-march=score7}][@b{-march=score3}]
571 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
575 @emph{Target SPARC options:}
576 @c The order here is important. See c-sparc.texi.
577 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Aleon}|@b{-Asparclet}|@b{-Asparclite}
578 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av8plusb}|@b{-Av8plusc}|@b{-Av8plusd}
579 @b{-Av8plusv}|@b{-Av8plusm}|@b{-Av9}|@b{-Av9a}|@b{-Av9b}|@b{-Av9c}
580 @b{-Av9d}|@b{-Av9e}|@b{-Av9v}|@b{-Av9m}|@b{-Asparc}|@b{-Asparcvis}
581 @b{-Asparcvis2}|@b{-Asparcfmaf}|@b{-Asparcima}|@b{-Asparcvis3}
582 @b{-Asparcvisr}|@b{-Asparc5}]
583 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}]|@b{-xarch=v8plusb}|@b{-xarch=v8plusc}
584 @b{-xarch=v8plusd}|@b{-xarch=v8plusv}|@b{-xarch=v8plusm}|@b{-xarch=v9}
585 @b{-xarch=v9a}|@b{-xarch=v9b}|@b{-xarch=v9c}|@b{-xarch=v9d}|@b{-xarch=v9e}
586 @b{-xarch=v9v}|@b{-xarch=v9m}|@b{-xarch=sparc}|@b{-xarch=sparcvis}
587 @b{-xarch=sparcvis2}|@b{-xarch=sparcfmaf}|@b{-xarch=sparcima}
588 @b{-xarch=sparcvis3}|@b{-xarch=sparcvisr}|@b{-xarch=sparc5}
591 [@b{--enforce-aligned-data}][@b{--dcti-couples-detect}]
595 @emph{Target TIC54X options:}
596 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
597 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
601 @emph{Target TIC6X options:}
602 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
603 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
604 [@b{-mpic}|@b{-mno-pic}]
608 @emph{Target TILE-Gx options:}
609 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
612 @c TILEPro has no machine-dependent assembler options
616 @emph{Target Visium options:}
617 [@b{-mtune=@var{arch}}]
621 @emph{Target Xtensa options:}
622 [@b{--[no-]text-section-literals}] [@b{--[no-]auto-litpools}]
623 [@b{--[no-]absolute-literals}]
624 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
625 [@b{--[no-]transform}]
626 [@b{--rename-section} @var{oldname}=@var{newname}]
627 [@b{--[no-]trampolines}]
631 @emph{Target Z80 options:}
632 [@b{-z80}] [@b{-r800}]
633 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
634 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
635 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
636 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
637 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
638 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
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
776 @item --size-check=error
777 @itemx --size-check=warning
778 Issue an error or warning for invalid ELF .size directive.
780 @item --elf-stt-common=no
781 @itemx --elf-stt-common=yes
782 These options control whether the ELF assembler should generate common
783 symbols with the @code{STT_COMMON} type. The default can be controlled
784 by a configure option @option{--enable-elf-stt-common}.
786 @item --generate-missing-build-notes=yes
787 @itemx --generate-missing-build-notes=no
788 These options control whether the ELF assembler should generate GNU Build
789 attribute notes if none are present in the input sources.
790 The default can be controlled by the @option{--enable-generate-build-notes}
796 Print a summary of the command-line options and exit.
799 Print a summary of all target specific options and exit.
802 Add directory @var{dir} to the search list for @code{.include} directives.
805 Don't warn about signed overflow.
808 @ifclear DIFF-TBL-KLUGE
809 This option is accepted but has no effect on the @value{TARGET} family.
811 @ifset DIFF-TBL-KLUGE
812 Issue warnings when difference tables altered for long displacements.
817 Keep (in the symbol table) local symbols. These symbols start with
818 system-specific local label prefixes, typically @samp{.L} for ELF systems
819 or @samp{L} for traditional a.out systems.
824 @item --listing-lhs-width=@var{number}
825 Set the maximum width, in words, of the output data column for an assembler
826 listing to @var{number}.
828 @item --listing-lhs-width2=@var{number}
829 Set the maximum width, in words, of the output data column for continuation
830 lines in an assembler listing to @var{number}.
832 @item --listing-rhs-width=@var{number}
833 Set the maximum width of an input source line, as displayed in a listing, to
836 @item --listing-cont-lines=@var{number}
837 Set the maximum number of lines printed in a listing for a single line of input
840 @item --no-pad-sections
841 Stop the assembler for padding the ends of output sections to the alignment
842 of that section. The default is to pad the sections, but this can waste space
843 which might be needed on targets which have tight memory constraints.
845 @item -o @var{objfile}
846 Name the object-file output from @command{@value{AS}} @var{objfile}.
849 Fold the data section into the text section.
851 @item --hash-size=@var{number}
852 Set the default size of GAS's hash tables to a prime number close to
853 @var{number}. Increasing this value can reduce the length of time it takes the
854 assembler to perform its tasks, at the expense of increasing the assembler's
855 memory requirements. Similarly reducing this value can reduce the memory
856 requirements at the expense of speed.
858 @item --reduce-memory-overheads
859 This option reduces GAS's memory requirements, at the expense of making the
860 assembly processes slower. Currently this switch is a synonym for
861 @samp{--hash-size=4051}, but in the future it may have other effects as well.
864 @item --sectname-subst
865 Honor substitution sequences in section names.
867 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
872 Print the maximum space (in bytes) and total time (in seconds) used by
875 @item --strip-local-absolute
876 Remove local absolute symbols from the outgoing symbol table.
880 Print the @command{as} version.
883 Print the @command{as} version and exit.
887 Suppress warning messages.
889 @item --fatal-warnings
890 Treat warnings as errors.
893 Don't suppress warning messages or treat them as errors.
902 Generate an object file even after errors.
904 @item -- | @var{files} @dots{}
905 Standard input, or source files to assemble.
913 @xref{AArch64 Options}, for the options available when @value{AS} is configured
914 for the 64-bit mode of the ARM Architecture (AArch64).
919 The following options are available when @value{AS} is configured for the
920 64-bit mode of the ARM Architecture (AArch64).
923 @include c-aarch64.texi
924 @c ended inside the included file
932 @xref{Alpha Options}, for the options available when @value{AS} is configured
933 for an Alpha processor.
938 The following options are available when @value{AS} is configured for an Alpha
942 @include c-alpha.texi
943 @c ended inside the included file
950 The following options are available when @value{AS} is configured for an ARC
954 @item -mcpu=@var{cpu}
955 This option selects the core processor variant.
957 Select either big-endian (-EB) or little-endian (-EL) output.
959 Enable Code Density extenssion instructions.
964 The following options are available when @value{AS} is configured for the ARM
968 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
969 Specify which ARM processor variant is the target.
970 @item -march=@var{architecture}[+@var{extension}@dots{}]
971 Specify which ARM architecture variant is used by the target.
972 @item -mfpu=@var{floating-point-format}
973 Select which Floating Point architecture is the target.
974 @item -mfloat-abi=@var{abi}
975 Select which floating point ABI is in use.
977 Enable Thumb only instruction decoding.
978 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
979 Select which procedure calling convention is in use.
981 Select either big-endian (-EB) or little-endian (-EL) output.
982 @item -mthumb-interwork
983 Specify that the code has been generated with interworking between Thumb and
986 Turns on CodeComposer Studio assembly syntax compatibility mode.
988 Specify that PIC code has been generated.
996 @xref{Blackfin Options}, for the options available when @value{AS} is
997 configured for the Blackfin processor family.
1001 @c man begin OPTIONS
1002 The following options are available when @value{AS} is configured for
1003 the Blackfin processor family.
1005 @c man begin INCLUDE
1006 @include c-bfin.texi
1007 @c ended inside the included file
1015 @xref{BPF Options}, for the options available when @value{AS} is
1016 configured for the Linux kernel BPF processor family.
1020 @c man begin OPTIONS
1021 The following options are available when @value{AS} is configured for
1022 the Linux kernel BPF processor family.
1024 @c man begin INCLUDE
1026 @c ended inside the included file
1031 @c man begin OPTIONS
1033 See the info pages for documentation of the CRIS-specific options.
1039 @xref{C-SKY Options}, for the options available when @value{AS} is
1040 configured for the C-SKY processor family.
1044 @c man begin OPTIONS
1045 The following options are available when @value{AS} is configured for
1046 the C-SKY processor family.
1048 @c man begin INCLUDE
1049 @include c-csky.texi
1050 @c ended inside the included file
1056 The following options are available when @value{AS} is configured for
1059 @cindex D10V optimization
1060 @cindex optimization, D10V
1062 Optimize output by parallelizing instructions.
1067 The following options are available when @value{AS} is configured for a D30V
1070 @cindex D30V optimization
1071 @cindex optimization, D30V
1073 Optimize output by parallelizing instructions.
1077 Warn when nops are generated.
1079 @cindex D30V nops after 32-bit multiply
1081 Warn when a nop after a 32-bit multiply instruction is generated.
1087 The following options are available when @value{AS} is configured for the
1088 Adapteva EPIPHANY series.
1091 @xref{Epiphany Options}, for the options available when @value{AS} is
1092 configured for an Epiphany processor.
1096 @c man begin OPTIONS
1097 The following options are available when @value{AS} is configured for
1098 an Epiphany processor.
1100 @c man begin INCLUDE
1101 @include c-epiphany.texi
1102 @c ended inside the included file
1110 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1111 for an H8/300 processor.
1115 @c man begin OPTIONS
1116 The following options are available when @value{AS} is configured for an H8/300
1119 @c man begin INCLUDE
1120 @include c-h8300.texi
1121 @c ended inside the included file
1129 @xref{i386-Options}, for the options available when @value{AS} is
1130 configured for an i386 processor.
1134 @c man begin OPTIONS
1135 The following options are available when @value{AS} is configured for
1138 @c man begin INCLUDE
1139 @include c-i386.texi
1140 @c ended inside the included file
1145 @c man begin OPTIONS
1147 The following options are available when @value{AS} is configured for the
1153 Specifies that the extended IP2022 instructions are allowed.
1156 Restores the default behaviour, which restricts the permitted instructions to
1157 just the basic IP2022 ones.
1163 The following options are available when @value{AS} is configured for the
1164 Renesas M32C and M16C processors.
1169 Assemble M32C instructions.
1172 Assemble M16C instructions (the default).
1175 Enable support for link-time relaxations.
1178 Support H'00 style hex constants in addition to 0x00 style.
1184 The following options are available when @value{AS} is configured for the
1185 Renesas M32R (formerly Mitsubishi M32R) series.
1190 Specify which processor in the M32R family is the target. The default
1191 is normally the M32R, but this option changes it to the M32RX.
1193 @item --warn-explicit-parallel-conflicts or --Wp
1194 Produce warning messages when questionable parallel constructs are
1197 @item --no-warn-explicit-parallel-conflicts or --Wnp
1198 Do not produce warning messages when questionable parallel constructs are
1205 The following options are available when @value{AS} is configured for the
1206 Motorola 68000 series.
1211 Shorten references to undefined symbols, to one word instead of two.
1213 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1214 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1215 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1216 Specify what processor in the 68000 family is the target. The default
1217 is normally the 68020, but this can be changed at configuration time.
1219 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1220 The target machine does (or does not) have a floating-point coprocessor.
1221 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1222 the basic 68000 is not compatible with the 68881, a combination of the
1223 two can be specified, since it's possible to do emulation of the
1224 coprocessor instructions with the main processor.
1226 @item -m68851 | -mno-68851
1227 The target machine does (or does not) have a memory-management
1228 unit coprocessor. The default is to assume an MMU for 68020 and up.
1236 @xref{Nios II Options}, for the options available when @value{AS} is configured
1237 for an Altera Nios II processor.
1241 @c man begin OPTIONS
1242 The following options are available when @value{AS} is configured for an
1243 Altera Nios II processor.
1245 @c man begin INCLUDE
1246 @include c-nios2.texi
1247 @c ended inside the included file
1253 For details about the PDP-11 machine dependent features options,
1254 see @ref{PDP-11-Options}.
1257 @item -mpic | -mno-pic
1258 Generate position-independent (or position-dependent) code. The
1259 default is @option{-mpic}.
1262 @itemx -mall-extensions
1263 Enable all instruction set extensions. This is the default.
1265 @item -mno-extensions
1266 Disable all instruction set extensions.
1268 @item -m@var{extension} | -mno-@var{extension}
1269 Enable (or disable) a particular instruction set extension.
1272 Enable the instruction set extensions supported by a particular CPU, and
1273 disable all other extensions.
1275 @item -m@var{machine}
1276 Enable the instruction set extensions supported by a particular machine
1277 model, and disable all other extensions.
1283 The following options are available when @value{AS} is configured for
1284 a picoJava processor.
1288 @cindex PJ endianness
1289 @cindex endianness, PJ
1290 @cindex big endian output, PJ
1292 Generate ``big endian'' format output.
1294 @cindex little endian output, PJ
1296 Generate ``little endian'' format output.
1304 @xref{PRU Options}, for the options available when @value{AS} is configured
1305 for a PRU processor.
1309 @c man begin OPTIONS
1310 The following options are available when @value{AS} is configured for a
1313 @c man begin INCLUDE
1315 @c ended inside the included file
1320 The following options are available when @value{AS} is configured for the
1321 Motorola 68HC11 or 68HC12 series.
1325 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1326 Specify what processor is the target. The default is
1327 defined by the configuration option when building the assembler.
1329 @item --xgate-ramoffset
1330 Instruct the linker to offset RAM addresses from S12X address space into
1331 XGATE address space.
1334 Specify to use the 16-bit integer ABI.
1337 Specify to use the 32-bit integer ABI.
1339 @item -mshort-double
1340 Specify to use the 32-bit double ABI.
1343 Specify to use the 64-bit double ABI.
1345 @item --force-long-branches
1346 Relative branches are turned into absolute ones. This concerns
1347 conditional branches, unconditional branches and branches to a
1350 @item -S | --short-branches
1351 Do not turn relative branches into absolute ones
1352 when the offset is out of range.
1354 @item --strict-direct-mode
1355 Do not turn the direct addressing mode into extended addressing mode
1356 when the instruction does not support direct addressing mode.
1358 @item --print-insn-syntax
1359 Print the syntax of instruction in case of error.
1361 @item --print-opcodes
1362 Print the list of instructions with syntax and then exit.
1364 @item --generate-example
1365 Print an example of instruction for each possible instruction and then exit.
1366 This option is only useful for testing @command{@value{AS}}.
1372 The following options are available when @command{@value{AS}} is configured
1373 for the SPARC architecture:
1376 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1377 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1378 Explicitly select a variant of the SPARC architecture.
1380 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1381 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1383 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1384 UltraSPARC extensions.
1386 @item -xarch=v8plus | -xarch=v8plusa
1387 For compatibility with the Solaris v9 assembler. These options are
1388 equivalent to -Av8plus and -Av8plusa, respectively.
1391 Warn when the assembler switches to another architecture.
1396 The following options are available when @value{AS} is configured for the 'c54x
1401 Enable extended addressing mode. All addresses and relocations will assume
1402 extended addressing (usually 23 bits).
1403 @item -mcpu=@var{CPU_VERSION}
1404 Sets the CPU version being compiled for.
1405 @item -merrors-to-file @var{FILENAME}
1406 Redirect error output to a file, for broken systems which don't support such
1407 behaviour in the shell.
1412 @c man begin OPTIONS
1413 The following options are available when @value{AS} is configured for
1418 This option sets the largest size of an object that can be referenced
1419 implicitly with the @code{gp} register. It is only accepted for targets that
1420 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1422 @cindex MIPS endianness
1423 @cindex endianness, MIPS
1424 @cindex big endian output, MIPS
1426 Generate ``big endian'' format output.
1428 @cindex little endian output, MIPS
1430 Generate ``little endian'' format output.
1448 Generate code for a particular MIPS Instruction Set Architecture level.
1449 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1450 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1451 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1452 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1453 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1454 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1455 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1456 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1457 MIPS64 Release 6 ISA processors, respectively.
1459 @item -march=@var{cpu}
1460 Generate code for a particular MIPS CPU.
1462 @item -mtune=@var{cpu}
1463 Schedule and tune for a particular MIPS CPU.
1467 Cause nops to be inserted if the read of the destination register
1468 of an mfhi or mflo instruction occurs in the following two instructions.
1471 @itemx -mno-fix-rm7000
1472 Cause nops to be inserted if a dmult or dmultu instruction is
1473 followed by a load instruction.
1476 @itemx -mno-fix-r5900
1477 Do not attempt to schedule the preceding instruction into the delay slot
1478 of a branch instruction placed at the end of a short loop of six
1479 instructions or fewer and always schedule a @code{nop} instruction there
1480 instead. The short loop bug under certain conditions causes loops to
1481 execute only once or twice, due to a hardware bug in the R5900 chip.
1485 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1486 section instead of the standard ELF .stabs sections.
1490 Control generation of @code{.pdr} sections.
1494 The register sizes are normally inferred from the ISA and ABI, but these
1495 flags force a certain group of registers to be treated as 32 bits wide at
1496 all times. @samp{-mgp32} controls the size of general-purpose registers
1497 and @samp{-mfp32} controls the size of floating-point registers.
1501 The register sizes are normally inferred from the ISA and ABI, but these
1502 flags force a certain group of registers to be treated as 64 bits wide at
1503 all times. @samp{-mgp64} controls the size of general-purpose registers
1504 and @samp{-mfp64} controls the size of floating-point registers.
1507 The register sizes are normally inferred from the ISA and ABI, but using
1508 this flag in combination with @samp{-mabi=32} enables an ABI variant
1509 which will operate correctly with floating-point registers which are
1513 @itemx -mno-odd-spreg
1514 Enable use of floating-point operations on odd-numbered single-precision
1515 registers when supported by the ISA. @samp{-mfpxx} implies
1516 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1520 Generate code for the MIPS 16 processor. This is equivalent to putting
1521 @code{.module mips16} at the start of the assembly file. @samp{-no-mips16}
1522 turns off this option.
1525 @itemx -mno-mips16e2
1526 Enable the use of MIPS16e2 instructions in MIPS16 mode. This is equivalent
1527 to putting @code{.module mips16e2} at the start of the assembly file.
1528 @samp{-mno-mips16e2} turns off this option.
1531 @itemx -mno-micromips
1532 Generate code for the microMIPS processor. This is equivalent to putting
1533 @code{.module micromips} at the start of the assembly file.
1534 @samp{-mno-micromips} turns off this option. This is equivalent to putting
1535 @code{.module nomicromips} at the start of the assembly file.
1538 @itemx -mno-smartmips
1539 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1540 equivalent to putting @code{.module smartmips} at the start of the assembly
1541 file. @samp{-mno-smartmips} turns off this option.
1545 Generate code for the MIPS-3D Application Specific Extension.
1546 This tells the assembler to accept MIPS-3D instructions.
1547 @samp{-no-mips3d} turns off this option.
1551 Generate code for the MDMX Application Specific Extension.
1552 This tells the assembler to accept MDMX instructions.
1553 @samp{-no-mdmx} turns off this option.
1557 Generate code for the DSP Release 1 Application Specific Extension.
1558 This tells the assembler to accept DSP Release 1 instructions.
1559 @samp{-mno-dsp} turns off this option.
1563 Generate code for the DSP Release 2 Application Specific Extension.
1564 This option implies @samp{-mdsp}.
1565 This tells the assembler to accept DSP Release 2 instructions.
1566 @samp{-mno-dspr2} turns off this option.
1570 Generate code for the DSP Release 3 Application Specific Extension.
1571 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1572 This tells the assembler to accept DSP Release 3 instructions.
1573 @samp{-mno-dspr3} turns off this option.
1577 Generate code for the MIPS SIMD Architecture Extension.
1578 This tells the assembler to accept MSA instructions.
1579 @samp{-mno-msa} turns off this option.
1583 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1584 This tells the assembler to accept XPA instructions.
1585 @samp{-mno-xpa} turns off this option.
1589 Generate code for the MT Application Specific Extension.
1590 This tells the assembler to accept MT instructions.
1591 @samp{-mno-mt} turns off this option.
1595 Generate code for the MCU Application Specific Extension.
1596 This tells the assembler to accept MCU instructions.
1597 @samp{-mno-mcu} turns off this option.
1601 Generate code for the MIPS cyclic redundancy check (CRC) Application
1602 Specific Extension. This tells the assembler to accept CRC instructions.
1603 @samp{-mno-crc} turns off this option.
1607 Generate code for the Global INValidate (GINV) Application Specific
1608 Extension. This tells the assembler to accept GINV instructions.
1609 @samp{-mno-ginv} turns off this option.
1611 @item -mloongson-mmi
1612 @itemx -mno-loongson-mmi
1613 Generate code for the Loongson MultiMedia extensions Instructions (MMI)
1614 Application Specific Extension. This tells the assembler to accept MMI
1616 @samp{-mno-loongson-mmi} turns off this option.
1618 @item -mloongson-cam
1619 @itemx -mno-loongson-cam
1620 Generate code for the Loongson Content Address Memory (CAM) instructions.
1621 This tells the assembler to accept Loongson CAM instructions.
1622 @samp{-mno-loongson-cam} turns off this option.
1624 @item -mloongson-ext
1625 @itemx -mno-loongson-ext
1626 Generate code for the Loongson EXTensions (EXT) instructions.
1627 This tells the assembler to accept Loongson EXT instructions.
1628 @samp{-mno-loongson-ext} turns off this option.
1630 @item -mloongson-ext2
1631 @itemx -mno-loongson-ext2
1632 Generate code for the Loongson EXTensions R2 (EXT2) instructions.
1633 This option implies @samp{-mloongson-ext}.
1634 This tells the assembler to accept Loongson EXT2 instructions.
1635 @samp{-mno-loongson-ext2} turns off this option.
1639 Only use 32-bit instruction encodings when generating code for the
1640 microMIPS processor. This option inhibits the use of any 16-bit
1641 instructions. This is equivalent to putting @code{.set insn32} at
1642 the start of the assembly file. @samp{-mno-insn32} turns off this
1643 option. This is equivalent to putting @code{.set noinsn32} at the
1644 start of the assembly file. By default @samp{-mno-insn32} is
1645 selected, allowing all instructions to be used.
1647 @item --construct-floats
1648 @itemx --no-construct-floats
1649 The @samp{--no-construct-floats} option disables the construction of
1650 double width floating point constants by loading the two halves of the
1651 value into the two single width floating point registers that make up
1652 the double width register. By default @samp{--construct-floats} is
1653 selected, allowing construction of these floating point constants.
1655 @item --relax-branch
1656 @itemx --no-relax-branch
1657 The @samp{--relax-branch} option enables the relaxation of out-of-range
1658 branches. By default @samp{--no-relax-branch} is selected, causing any
1659 out-of-range branches to produce an error.
1661 @item -mignore-branch-isa
1662 @itemx -mno-ignore-branch-isa
1663 Ignore branch checks for invalid transitions between ISA modes. The
1664 semantics of branches does not provide for an ISA mode switch, so in
1665 most cases the ISA mode a branch has been encoded for has to be the
1666 same as the ISA mode of the branch's target label. Therefore GAS has
1667 checks implemented that verify in branch assembly that the two ISA
1668 modes match. @samp{-mignore-branch-isa} disables these checks. By
1669 default @samp{-mno-ignore-branch-isa} is selected, causing any invalid
1670 branch requiring a transition between ISA modes to produce an error.
1672 @item -mnan=@var{encoding}
1673 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1674 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1677 @item --emulation=@var{name}
1678 This option was formerly used to switch between ELF and ECOFF output
1679 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1680 removed in GAS 2.24, so the option now serves little purpose.
1681 It is retained for backwards compatibility.
1683 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1684 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1685 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1686 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1687 preferred options instead.
1690 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1697 Control how to deal with multiplication overflow and division by zero.
1698 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1699 (and only work for Instruction Set Architecture level 2 and higher);
1700 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1704 When this option is used, @command{@value{AS}} will issue a warning every
1705 time it generates a nop instruction from a macro.
1711 The following options are available when @value{AS} is configured for
1717 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1718 The command-line option @samp{-nojsri2bsr} can be used to disable it.
1722 Enable or disable the silicon filter behaviour. By default this is disabled.
1723 The default can be overridden by the @samp{-sifilter} command-line option.
1726 Alter jump instructions for long displacements.
1728 @item -mcpu=[210|340]
1729 Select the cpu type on the target hardware. This controls which instructions
1733 Assemble for a big endian target.
1736 Assemble for a little endian target.
1745 @xref{Meta Options}, for the options available when @value{AS} is configured
1746 for a Meta processor.
1750 @c man begin OPTIONS
1751 The following options are available when @value{AS} is configured for a
1754 @c man begin INCLUDE
1755 @include c-metag.texi
1756 @c ended inside the included file
1761 @c man begin OPTIONS
1763 See the info pages for documentation of the MMIX-specific options.
1769 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1770 for a NDS32 processor.
1772 @c ended inside the included file
1776 @c man begin OPTIONS
1777 The following options are available when @value{AS} is configured for a
1780 @c man begin INCLUDE
1781 @include c-nds32.texi
1782 @c ended inside the included file
1789 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1790 for a PowerPC processor.
1794 @c man begin OPTIONS
1795 The following options are available when @value{AS} is configured for a
1798 @c man begin INCLUDE
1800 @c ended inside the included file
1808 @xref{RISC-V-Options}, for the options available when @value{AS} is configured
1809 for a RISC-V processor.
1813 @c man begin OPTIONS
1814 The following options are available when @value{AS} is configured for a
1817 @c man begin INCLUDE
1818 @include c-riscv.texi
1819 @c ended inside the included file
1824 @c man begin OPTIONS
1826 See the info pages for documentation of the RX-specific options.
1830 The following options are available when @value{AS} is configured for the s390
1836 Select the word size, either 31/32 bits or 64 bits.
1839 Select the architecture mode, either the Enterprise System
1840 Architecture (esa) or the z/Architecture mode (zarch).
1841 @item -march=@var{processor}
1842 Specify which s390 processor variant is the target, @samp{g5} (or
1843 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1844 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1845 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1846 @samp{z13} (or @samp{arch11}), or @samp{z14} (or @samp{arch12}).
1848 @itemx -mno-regnames
1849 Allow or disallow symbolic names for registers.
1850 @item -mwarn-areg-zero
1851 Warn whenever the operand for a base or index register has been specified
1852 but evaluates to zero.
1860 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1861 for a TMS320C6000 processor.
1865 @c man begin OPTIONS
1866 The following options are available when @value{AS} is configured for a
1867 TMS320C6000 processor.
1869 @c man begin INCLUDE
1870 @include c-tic6x.texi
1871 @c ended inside the included file
1879 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1880 for a TILE-Gx processor.
1884 @c man begin OPTIONS
1885 The following options are available when @value{AS} is configured for a TILE-Gx
1888 @c man begin INCLUDE
1889 @include c-tilegx.texi
1890 @c ended inside the included file
1898 @xref{Visium Options}, for the options available when @value{AS} is configured
1899 for a Visium processor.
1903 @c man begin OPTIONS
1904 The following option is available when @value{AS} is configured for a Visium
1907 @c man begin INCLUDE
1908 @include c-visium.texi
1909 @c ended inside the included file
1917 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1918 for an Xtensa processor.
1922 @c man begin OPTIONS
1923 The following options are available when @value{AS} is configured for an
1926 @c man begin INCLUDE
1927 @include c-xtensa.texi
1928 @c ended inside the included file
1933 @c man begin OPTIONS
1936 The following options are available when @value{AS} is configured for
1937 a Z80 family processor.
1940 Assemble for Z80 processor.
1942 Assemble for R800 processor.
1943 @item -ignore-undocumented-instructions
1945 Assemble undocumented Z80 instructions that also work on R800 without warning.
1946 @item -ignore-unportable-instructions
1948 Assemble all undocumented Z80 instructions without warning.
1949 @item -warn-undocumented-instructions
1951 Issue a warning for undocumented Z80 instructions that also work on R800.
1952 @item -warn-unportable-instructions
1954 Issue a warning for undocumented Z80 instructions that do not work on R800.
1955 @item -forbid-undocumented-instructions
1957 Treat all undocumented instructions as errors.
1958 @item -forbid-unportable-instructions
1960 Treat undocumented Z80 instructions that do not work on R800 as errors.
1967 * Manual:: Structure of this Manual
1968 * GNU Assembler:: The GNU Assembler
1969 * Object Formats:: Object File Formats
1970 * Command Line:: Command Line
1971 * Input Files:: Input Files
1972 * Object:: Output (Object) File
1973 * Errors:: Error and Warning Messages
1977 @section Structure of this Manual
1979 @cindex manual, structure and purpose
1980 This manual is intended to describe what you need to know to use
1981 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1982 notation for symbols, constants, and expressions; the directives that
1983 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1986 We also cover special features in the @value{TARGET}
1987 configuration of @command{@value{AS}}, including assembler directives.
1990 This manual also describes some of the machine-dependent features of
1991 various flavors of the assembler.
1994 @cindex machine instructions (not covered)
1995 On the other hand, this manual is @emph{not} intended as an introduction
1996 to programming in assembly language---let alone programming in general!
1997 In a similar vein, we make no attempt to introduce the machine
1998 architecture; we do @emph{not} describe the instruction set, standard
1999 mnemonics, registers or addressing modes that are standard to a
2000 particular architecture.
2002 You may want to consult the manufacturer's
2003 machine architecture manual for this information.
2007 For information on the H8/300 machine instruction set, see @cite{H8/300
2008 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
2009 Programming Manual} (Renesas).
2012 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
2013 see @cite{SH-Microcomputer User's Manual} (Renesas) or
2014 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
2015 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
2018 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
2022 @c I think this is premature---doc@cygnus.com, 17jan1991
2024 Throughout this manual, we assume that you are running @dfn{GNU},
2025 the portable operating system from the @dfn{Free Software
2026 Foundation, Inc.}. This restricts our attention to certain kinds of
2027 computer (in particular, the kinds of computers that @sc{gnu} can run on);
2028 once this assumption is granted examples and definitions need less
2031 @command{@value{AS}} is part of a team of programs that turn a high-level
2032 human-readable series of instructions into a low-level
2033 computer-readable series of instructions. Different versions of
2034 @command{@value{AS}} are used for different kinds of computer.
2037 @c There used to be a section "Terminology" here, which defined
2038 @c "contents", "byte", "word", and "long". Defining "word" to any
2039 @c particular size is confusing when the .word directive may generate 16
2040 @c bits on one machine and 32 bits on another; in general, for the user
2041 @c version of this manual, none of these terms seem essential to define.
2042 @c They were used very little even in the former draft of the manual;
2043 @c this draft makes an effort to avoid them (except in names of
2047 @section The GNU Assembler
2049 @c man begin DESCRIPTION
2051 @sc{gnu} @command{as} is really a family of assemblers.
2053 This manual describes @command{@value{AS}}, a member of that family which is
2054 configured for the @value{TARGET} architectures.
2056 If you use (or have used) the @sc{gnu} assembler on one architecture, you
2057 should find a fairly similar environment when you use it on another
2058 architecture. Each version has much in common with the others,
2059 including object file formats, most assembler directives (often called
2060 @dfn{pseudo-ops}) and assembler syntax.@refill
2062 @cindex purpose of @sc{gnu} assembler
2063 @command{@value{AS}} is primarily intended to assemble the output of the
2064 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
2065 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
2066 assemble correctly everything that other assemblers for the same
2067 machine would assemble.
2069 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
2072 @c This remark should appear in generic version of manual; assumption
2073 @c here is that generic version sets M680x0.
2074 This doesn't mean @command{@value{AS}} always uses the same syntax as another
2075 assembler for the same architecture; for example, we know of several
2076 incompatible versions of 680x0 assembly language syntax.
2081 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
2082 program in one pass of the source file. This has a subtle impact on the
2083 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
2085 @node Object Formats
2086 @section Object File Formats
2088 @cindex object file format
2089 The @sc{gnu} assembler can be configured to produce several alternative
2090 object file formats. For the most part, this does not affect how you
2091 write assembly language programs; but directives for debugging symbols
2092 are typically different in different file formats. @xref{Symbol
2093 Attributes,,Symbol Attributes}.
2096 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
2097 @value{OBJ-NAME} format object files.
2099 @c The following should exhaust all configs that set MULTI-OBJ, ideally
2101 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
2102 SOM or ELF format object files.
2107 @section Command Line
2109 @cindex command line conventions
2111 After the program name @command{@value{AS}}, the command line may contain
2112 options and file names. Options may appear in any order, and may be
2113 before, after, or between file names. The order of file names is
2116 @cindex standard input, as input file
2118 @file{--} (two hyphens) by itself names the standard input file
2119 explicitly, as one of the files for @command{@value{AS}} to assemble.
2121 @cindex options, command line
2122 Except for @samp{--} any command-line argument that begins with a
2123 hyphen (@samp{-}) is an option. Each option changes the behavior of
2124 @command{@value{AS}}. No option changes the way another option works. An
2125 option is a @samp{-} followed by one or more letters; the case of
2126 the letter is important. All options are optional.
2128 Some options expect exactly one file name to follow them. The file
2129 name may either immediately follow the option's letter (compatible
2130 with older assemblers) or it may be the next command argument (@sc{gnu}
2131 standard). These two command lines are equivalent:
2134 @value{AS} -o my-object-file.o mumble.s
2135 @value{AS} -omy-object-file.o mumble.s
2139 @section Input Files
2142 @cindex source program
2143 @cindex files, input
2144 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2145 describe the program input to one run of @command{@value{AS}}. The program may
2146 be in one or more files; how the source is partitioned into files
2147 doesn't change the meaning of the source.
2149 @c I added "con" prefix to "catenation" just to prove I can overcome my
2150 @c APL training... doc@cygnus.com
2151 The source program is a concatenation of the text in all the files, in the
2154 @c man begin DESCRIPTION
2155 Each time you run @command{@value{AS}} it assembles exactly one source
2156 program. The source program is made up of one or more files.
2157 (The standard input is also a file.)
2159 You give @command{@value{AS}} a command line that has zero or more input file
2160 names. The input files are read (from left file name to right). A
2161 command-line argument (in any position) that has no special meaning
2162 is taken to be an input file name.
2164 If you give @command{@value{AS}} no file names it attempts to read one input file
2165 from the @command{@value{AS}} standard input, which is normally your terminal. You
2166 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2169 Use @samp{--} if you need to explicitly name the standard input file
2170 in your command line.
2172 If the source is empty, @command{@value{AS}} produces a small, empty object
2177 @subheading Filenames and Line-numbers
2179 @cindex input file linenumbers
2180 @cindex line numbers, in input files
2181 There are two ways of locating a line in the input file (or files) and
2182 either may be used in reporting error messages. One way refers to a line
2183 number in a physical file; the other refers to a line number in a
2184 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2186 @dfn{Physical files} are those files named in the command line given
2187 to @command{@value{AS}}.
2189 @dfn{Logical files} are simply names declared explicitly by assembler
2190 directives; they bear no relation to physical files. Logical file names help
2191 error messages reflect the original source file, when @command{@value{AS}} source
2192 is itself synthesized from other files. @command{@value{AS}} understands the
2193 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2194 @ref{File,,@code{.file}}.
2197 @section Output (Object) File
2203 Every time you run @command{@value{AS}} it produces an output file, which is
2204 your assembly language program translated into numbers. This file
2205 is the object file. Its default name is @code{a.out}.
2206 You can give it another name by using the @option{-o} option. Conventionally,
2207 object file names end with @file{.o}. The default name is used for historical
2208 reasons: older assemblers were capable of assembling self-contained programs
2209 directly into a runnable program. (For some formats, this isn't currently
2210 possible, but it can be done for the @code{a.out} format.)
2214 The object file is meant for input to the linker @code{@value{LD}}. It contains
2215 assembled program code, information to help @code{@value{LD}} integrate
2216 the assembled program into a runnable file, and (optionally) symbolic
2217 information for the debugger.
2219 @c link above to some info file(s) like the description of a.out.
2220 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2223 @section Error and Warning Messages
2225 @c man begin DESCRIPTION
2227 @cindex error messages
2228 @cindex warning messages
2229 @cindex messages from assembler
2230 @command{@value{AS}} may write warnings and error messages to the standard error
2231 file (usually your terminal). This should not happen when a compiler
2232 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2233 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2234 grave problem that stops the assembly.
2238 @cindex format of warning messages
2239 Warning messages have the format
2242 file_name:@b{NNN}:Warning Message Text
2246 @cindex file names and line numbers, in warnings/errors
2247 (where @b{NNN} is a line number). If both a logical file name
2248 (@pxref{File,,@code{.file}}) and a logical line number
2250 (@pxref{Line,,@code{.line}})
2252 have been given then they will be used, otherwise the file name and line number
2253 in the current assembler source file will be used. The message text is
2254 intended to be self explanatory (in the grand Unix tradition).
2256 Note the file name must be set via the logical version of the @code{.file}
2257 directive, not the DWARF2 version of the @code{.file} directive. For example:
2261 error_assembler_source
2267 produces this output:
2271 asm.s:2: Error: no such instruction: `error_assembler_source'
2272 foo.c:31: Error: no such instruction: `error_c_source'
2275 @cindex format of error messages
2276 Error messages have the format
2279 file_name:@b{NNN}:FATAL:Error Message Text
2282 The file name and line number are derived as for warning
2283 messages. The actual message text may be rather less explanatory
2284 because many of them aren't supposed to happen.
2287 @chapter Command-Line Options
2289 @cindex options, all versions of assembler
2290 This chapter describes command-line options available in @emph{all}
2291 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2292 for options specific
2294 to the @value{TARGET} target.
2297 to particular machine architectures.
2300 @c man begin DESCRIPTION
2302 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2303 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2304 The assembler arguments must be separated from each other (and the @samp{-Wa})
2305 by commas. For example:
2308 gcc -c -g -O -Wa,-alh,-L file.c
2312 This passes two options to the assembler: @samp{-alh} (emit a listing to
2313 standard output with high-level and assembly source) and @samp{-L} (retain
2314 local symbols in the symbol table).
2316 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2317 command-line options are automatically passed to the assembler by the compiler.
2318 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2319 precisely what options it passes to each compilation pass, including the
2325 * a:: -a[cdghlns] enable listings
2326 * alternate:: --alternate enable alternate macro syntax
2327 * D:: -D for compatibility
2328 * f:: -f to work faster
2329 * I:: -I for .include search path
2330 @ifclear DIFF-TBL-KLUGE
2331 * K:: -K for compatibility
2333 @ifset DIFF-TBL-KLUGE
2334 * K:: -K for difference tables
2337 * L:: -L to retain local symbols
2338 * listing:: --listing-XXX to configure listing output
2339 * M:: -M or --mri to assemble in MRI compatibility mode
2340 * MD:: --MD for dependency tracking
2341 * no-pad-sections:: --no-pad-sections to stop section padding
2342 * o:: -o to name the object file
2343 * R:: -R to join data and text sections
2344 * statistics:: --statistics to see statistics about assembly
2345 * traditional-format:: --traditional-format for compatible output
2346 * v:: -v to announce version
2347 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2348 * Z:: -Z to make object file even after errors
2352 @section Enable Listings: @option{-a[cdghlns]}
2362 @cindex listings, enabling
2363 @cindex assembly listings, enabling
2365 These options enable listing output from the assembler. By itself,
2366 @samp{-a} requests high-level, assembly, and symbols listing.
2367 You can use other letters to select specific options for the list:
2368 @samp{-ah} requests a high-level language listing,
2369 @samp{-al} requests an output-program assembly listing, and
2370 @samp{-as} requests a symbol table listing.
2371 High-level listings require that a compiler debugging option like
2372 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2375 Use the @samp{-ag} option to print a first section with general assembly
2376 information, like @value{AS} version, switches passed, or time stamp.
2378 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2379 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2380 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2381 omitted from the listing.
2383 Use the @samp{-ad} option to omit debugging directives from the
2386 Once you have specified one of these options, you can further control
2387 listing output and its appearance using the directives @code{.list},
2388 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2390 The @samp{-an} option turns off all forms processing.
2391 If you do not request listing output with one of the @samp{-a} options, the
2392 listing-control directives have no effect.
2394 The letters after @samp{-a} may be combined into one option,
2395 @emph{e.g.}, @samp{-aln}.
2397 Note if the assembler source is coming from the standard input (e.g.,
2399 is being created by @code{@value{GCC}} and the @samp{-pipe} command-line switch
2400 is being used) then the listing will not contain any comments or preprocessor
2401 directives. This is because the listing code buffers input source lines from
2402 stdin only after they have been preprocessed by the assembler. This reduces
2403 memory usage and makes the code more efficient.
2406 @section @option{--alternate}
2409 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2412 @section @option{-D}
2415 This option has no effect whatsoever, but it is accepted to make it more
2416 likely that scripts written for other assemblers also work with
2417 @command{@value{AS}}.
2420 @section Work Faster: @option{-f}
2423 @cindex trusted compiler
2424 @cindex faster processing (@option{-f})
2425 @samp{-f} should only be used when assembling programs written by a
2426 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2427 and comment preprocessing on
2428 the input file(s) before assembling them. @xref{Preprocessing,
2432 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2433 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2438 @section @code{.include} Search Path: @option{-I} @var{path}
2440 @kindex -I @var{path}
2441 @cindex paths for @code{.include}
2442 @cindex search path for @code{.include}
2443 @cindex @code{include} directive search path
2444 Use this option to add a @var{path} to the list of directories
2445 @command{@value{AS}} searches for files specified in @code{.include}
2446 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2447 many times as necessary to include a variety of paths. The current
2448 working directory is always searched first; after that, @command{@value{AS}}
2449 searches any @samp{-I} directories in the same order as they were
2450 specified (left to right) on the command line.
2453 @section Difference Tables: @option{-K}
2456 @ifclear DIFF-TBL-KLUGE
2457 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2458 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2459 where it can be used to warn when the assembler alters the machine code
2460 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2461 family does not have the addressing limitations that sometimes lead to this
2462 alteration on other platforms.
2465 @ifset DIFF-TBL-KLUGE
2466 @cindex difference tables, warning
2467 @cindex warning for altered difference tables
2468 @command{@value{AS}} sometimes alters the code emitted for directives of the
2469 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2470 You can use the @samp{-K} option if you want a warning issued when this
2475 @section Include Local Symbols: @option{-L}
2478 @cindex local symbols, retaining in output
2479 Symbols beginning with system-specific local label prefixes, typically
2480 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2481 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2482 such symbols when debugging, because they are intended for the use of
2483 programs (like compilers) that compose assembler programs, not for your
2484 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2485 such symbols, so you do not normally debug with them.
2487 This option tells @command{@value{AS}} to retain those local symbols
2488 in the object file. Usually if you do this you also tell the linker
2489 @code{@value{LD}} to preserve those symbols.
2492 @section Configuring listing output: @option{--listing}
2494 The listing feature of the assembler can be enabled via the command-line switch
2495 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2496 hex dump of the corresponding locations in the output object file, and displays
2497 them as a listing file. The format of this listing can be controlled by
2498 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2499 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2500 @code{.psize} (@pxref{Psize}), and
2501 @code{.eject} (@pxref{Eject}) and also by the following switches:
2504 @item --listing-lhs-width=@samp{number}
2505 @kindex --listing-lhs-width
2506 @cindex Width of first line disassembly output
2507 Sets the maximum width, in words, of the first line of the hex byte dump. This
2508 dump appears on the left hand side of the listing output.
2510 @item --listing-lhs-width2=@samp{number}
2511 @kindex --listing-lhs-width2
2512 @cindex Width of continuation lines of disassembly output
2513 Sets the maximum width, in words, of any further lines of the hex byte dump for
2514 a given input source line. If this value is not specified, it defaults to being
2515 the same as the value specified for @samp{--listing-lhs-width}. If neither
2516 switch is used the default is to one.
2518 @item --listing-rhs-width=@samp{number}
2519 @kindex --listing-rhs-width
2520 @cindex Width of source line output
2521 Sets the maximum width, in characters, of the source line that is displayed
2522 alongside the hex dump. The default value for this parameter is 100. The
2523 source line is displayed on the right hand side of the listing output.
2525 @item --listing-cont-lines=@samp{number}
2526 @kindex --listing-cont-lines
2527 @cindex Maximum number of continuation lines
2528 Sets the maximum number of continuation lines of hex dump that will be
2529 displayed for a given single line of source input. The default value is 4.
2533 @section Assemble in MRI Compatibility Mode: @option{-M}
2536 @cindex MRI compatibility mode
2537 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2538 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2539 compatible with the @code{ASM68K} assembler from Microtec Research.
2540 The exact nature of the
2541 MRI syntax will not be documented here; see the MRI manuals for more
2542 information. Note in particular that the handling of macros and macro
2543 arguments is somewhat different. The purpose of this option is to permit
2544 assembling existing MRI assembler code using @command{@value{AS}}.
2546 The MRI compatibility is not complete. Certain operations of the MRI assembler
2547 depend upon its object file format, and can not be supported using other object
2548 file formats. Supporting these would require enhancing each object file format
2549 individually. These are:
2552 @item global symbols in common section
2554 The m68k MRI assembler supports common sections which are merged by the linker.
2555 Other object file formats do not support this. @command{@value{AS}} handles
2556 common sections by treating them as a single common symbol. It permits local
2557 symbols to be defined within a common section, but it can not support global
2558 symbols, since it has no way to describe them.
2560 @item complex relocations
2562 The MRI assemblers support relocations against a negated section address, and
2563 relocations which combine the start addresses of two or more sections. These
2564 are not support by other object file formats.
2566 @item @code{END} pseudo-op specifying start address
2568 The MRI @code{END} pseudo-op permits the specification of a start address.
2569 This is not supported by other object file formats. The start address may
2570 instead be specified using the @option{-e} option to the linker, or in a linker
2573 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2575 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2576 name to the output file. This is not supported by other object file formats.
2578 @item @code{ORG} pseudo-op
2580 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2581 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2582 which changes the location within the current section. Absolute sections are
2583 not supported by other object file formats. The address of a section may be
2584 assigned within a linker script.
2587 There are some other features of the MRI assembler which are not supported by
2588 @command{@value{AS}}, typically either because they are difficult or because they
2589 seem of little consequence. Some of these may be supported in future releases.
2593 @item EBCDIC strings
2595 EBCDIC strings are not supported.
2597 @item packed binary coded decimal
2599 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2600 and @code{DCB.P} pseudo-ops are not supported.
2602 @item @code{FEQU} pseudo-op
2604 The m68k @code{FEQU} pseudo-op is not supported.
2606 @item @code{NOOBJ} pseudo-op
2608 The m68k @code{NOOBJ} pseudo-op is not supported.
2610 @item @code{OPT} branch control options
2612 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2613 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2614 relaxes all branches, whether forward or backward, to an appropriate size, so
2615 these options serve no purpose.
2617 @item @code{OPT} list control options
2619 The following m68k @code{OPT} list control options are ignored: @code{C},
2620 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2621 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2623 @item other @code{OPT} options
2625 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2626 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2628 @item @code{OPT} @code{D} option is default
2630 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2631 @code{OPT NOD} may be used to turn it off.
2633 @item @code{XREF} pseudo-op.
2635 The m68k @code{XREF} pseudo-op is ignored.
2640 @section Dependency Tracking: @option{--MD}
2643 @cindex dependency tracking
2646 @command{@value{AS}} can generate a dependency file for the file it creates. This
2647 file consists of a single rule suitable for @code{make} describing the
2648 dependencies of the main source file.
2650 The rule is written to the file named in its argument.
2652 This feature is used in the automatic updating of makefiles.
2654 @node no-pad-sections
2655 @section Output Section Padding
2656 @kindex --no-pad-sections
2657 @cindex output section padding
2658 Normally the assembler will pad the end of each output section up to its
2659 alignment boundary. But this can waste space, which can be significant on
2660 memory constrained targets. So the @option{--no-pad-sections} option will
2661 disable this behaviour.
2664 @section Name the Object File: @option{-o}
2667 @cindex naming object file
2668 @cindex object file name
2669 There is always one object file output when you run @command{@value{AS}}. By
2670 default it has the name @file{a.out}.
2671 You use this option (which takes exactly one filename) to give the
2672 object file a different name.
2674 Whatever the object file is called, @command{@value{AS}} overwrites any
2675 existing file of the same name.
2678 @section Join Data and Text Sections: @option{-R}
2681 @cindex data and text sections, joining
2682 @cindex text and data sections, joining
2683 @cindex joining text and data sections
2684 @cindex merging text and data sections
2685 @option{-R} tells @command{@value{AS}} to write the object file as if all
2686 data-section data lives in the text section. This is only done at
2687 the very last moment: your binary data are the same, but data
2688 section parts are relocated differently. The data section part of
2689 your object file is zero bytes long because all its bytes are
2690 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2692 When you specify @option{-R} it would be possible to generate shorter
2693 address displacements (because we do not have to cross between text and
2694 data section). We refrain from doing this simply for compatibility with
2695 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2698 When @command{@value{AS}} is configured for COFF or ELF output,
2699 this option is only useful if you use sections named @samp{.text} and
2704 @option{-R} is not supported for any of the HPPA targets. Using
2705 @option{-R} generates a warning from @command{@value{AS}}.
2709 @section Display Assembly Statistics: @option{--statistics}
2711 @kindex --statistics
2712 @cindex statistics, about assembly
2713 @cindex time, total for assembly
2714 @cindex space used, maximum for assembly
2715 Use @samp{--statistics} to display two statistics about the resources used by
2716 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2717 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2720 @node traditional-format
2721 @section Compatible Output: @option{--traditional-format}
2723 @kindex --traditional-format
2724 For some targets, the output of @command{@value{AS}} is different in some ways
2725 from the output of some existing assembler. This switch requests
2726 @command{@value{AS}} to use the traditional format instead.
2728 For example, it disables the exception frame optimizations which
2729 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2732 @section Announce Version: @option{-v}
2736 @cindex assembler version
2737 @cindex version of assembler
2738 You can find out what version of as is running by including the
2739 option @samp{-v} (which you can also spell as @samp{-version}) on the
2743 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2745 @command{@value{AS}} should never give a warning or error message when
2746 assembling compiler output. But programs written by people often
2747 cause @command{@value{AS}} to give a warning that a particular assumption was
2748 made. All such warnings are directed to the standard error file.
2752 @cindex suppressing warnings
2753 @cindex warnings, suppressing
2754 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2755 This only affects the warning messages: it does not change any particular of
2756 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2759 @kindex --fatal-warnings
2760 @cindex errors, caused by warnings
2761 @cindex warnings, causing error
2762 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2763 files that generate warnings to be in error.
2766 @cindex warnings, switching on
2767 You can switch these options off again by specifying @option{--warn}, which
2768 causes warnings to be output as usual.
2771 @section Generate Object File in Spite of Errors: @option{-Z}
2772 @cindex object file, after errors
2773 @cindex errors, continuing after
2774 After an error message, @command{@value{AS}} normally produces no output. If for
2775 some reason you are interested in object file output even after
2776 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2777 option. If there are any errors, @command{@value{AS}} continues anyways, and
2778 writes an object file after a final warning message of the form @samp{@var{n}
2779 errors, @var{m} warnings, generating bad object file.}
2784 @cindex machine-independent syntax
2785 @cindex syntax, machine-independent
2786 This chapter describes the machine-independent syntax allowed in a
2787 source file. @command{@value{AS}} syntax is similar to what many other
2788 assemblers use; it is inspired by the BSD 4.2
2793 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2797 * Preprocessing:: Preprocessing
2798 * Whitespace:: Whitespace
2799 * Comments:: Comments
2800 * Symbol Intro:: Symbols
2801 * Statements:: Statements
2802 * Constants:: Constants
2806 @section Preprocessing
2808 @cindex preprocessing
2809 The @command{@value{AS}} internal preprocessor:
2811 @cindex whitespace, removed by preprocessor
2813 adjusts and removes extra whitespace. It leaves one space or tab before
2814 the keywords on a line, and turns any other whitespace on the line into
2817 @cindex comments, removed by preprocessor
2819 removes all comments, replacing them with a single space, or an
2820 appropriate number of newlines.
2822 @cindex constants, converted by preprocessor
2824 converts character constants into the appropriate numeric values.
2827 It does not do macro processing, include file handling, or
2828 anything else you may get from your C compiler's preprocessor. You can
2829 do include file processing with the @code{.include} directive
2830 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2831 to get other ``CPP'' style preprocessing by giving the input file a
2832 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2833 Output, gcc info, Using GNU CC}.
2835 Excess whitespace, comments, and character constants
2836 cannot be used in the portions of the input text that are not
2839 @cindex turning preprocessing on and off
2840 @cindex preprocessing, turning on and off
2843 If the first line of an input file is @code{#NO_APP} or if you use the
2844 @samp{-f} option, whitespace and comments are not removed from the input file.
2845 Within an input file, you can ask for whitespace and comment removal in
2846 specific portions of the by putting a line that says @code{#APP} before the
2847 text that may contain whitespace or comments, and putting a line that says
2848 @code{#NO_APP} after this text. This feature is mainly intend to support
2849 @code{asm} statements in compilers whose output is otherwise free of comments
2856 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2857 Whitespace is used to separate symbols, and to make programs neater for
2858 people to read. Unless within character constants
2859 (@pxref{Characters,,Character Constants}), any whitespace means the same
2860 as exactly one space.
2866 There are two ways of rendering comments to @command{@value{AS}}. In both
2867 cases the comment is equivalent to one space.
2869 Anything from @samp{/*} through the next @samp{*/} is a comment.
2870 This means you may not nest these comments.
2874 The only way to include a newline ('\n') in a comment
2875 is to use this sort of comment.
2878 /* This sort of comment does not nest. */
2881 @cindex line comment character
2882 Anything from a @dfn{line comment} character up to the next newline is
2883 considered a comment and is ignored. The line comment character is target
2884 specific, and some targets multiple comment characters. Some targets also have
2885 line comment characters that only work if they are the first character on a
2886 line. Some targets use a sequence of two characters to introduce a line
2887 comment. Some targets can also change their line comment characters depending
2888 upon command-line options that have been used. For more details see the
2889 @emph{Syntax} section in the documentation for individual targets.
2891 If the line comment character is the hash sign (@samp{#}) then it still has the
2892 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2893 to specify logical line numbers:
2896 @cindex lines starting with @code{#}
2897 @cindex logical line numbers
2898 To be compatible with past assemblers, lines that begin with @samp{#} have a
2899 special interpretation. Following the @samp{#} should be an absolute
2900 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2901 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2902 new logical file name. The rest of the line, if any, should be whitespace.
2904 If the first non-whitespace characters on the line are not numeric,
2905 the line is ignored. (Just like a comment.)
2908 # This is an ordinary comment.
2909 # 42-6 "new_file_name" # New logical file name
2910 # This is logical line # 36.
2912 This feature is deprecated, and may disappear from future versions
2913 of @command{@value{AS}}.
2918 @cindex characters used in symbols
2919 @ifclear SPECIAL-SYMS
2920 A @dfn{symbol} is one or more characters chosen from the set of all
2921 letters (both upper and lower case), digits and the three characters
2927 A @dfn{symbol} is one or more characters chosen from the set of all
2928 letters (both upper and lower case), digits and the three characters
2929 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2935 On most machines, you can also use @code{$} in symbol names; exceptions
2936 are noted in @ref{Machine Dependencies}.
2938 No symbol may begin with a digit. Case is significant.
2939 There is no length limit; all characters are significant. Multibyte characters
2940 are supported. Symbols are delimited by characters not in that set, or by the
2941 beginning of a file (since the source program must end with a newline, the end
2942 of a file is not a possible symbol delimiter). @xref{Symbols}.
2944 Symbol names may also be enclosed in double quote @code{"} characters. In such
2945 cases any characters are allowed, except for the NUL character. If a double
2946 quote character is to be included in the symbol name it must be preceeded by a
2947 backslash @code{\} character.
2948 @cindex length of symbols
2953 @cindex statements, structure of
2954 @cindex line separator character
2955 @cindex statement separator character
2957 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2958 @dfn{line separator character}. The line separator character is target
2959 specific and described in the @emph{Syntax} section of each
2960 target's documentation. Not all targets support a line separator character.
2961 The newline or line separator character is considered to be part of the
2962 preceding statement. Newlines and separators within character constants are an
2963 exception: they do not end statements.
2965 @cindex newline, required at file end
2966 @cindex EOF, newline must precede
2967 It is an error to end any statement with end-of-file: the last
2968 character of any input file should be a newline.@refill
2970 An empty statement is allowed, and may include whitespace. It is ignored.
2972 @cindex instructions and directives
2973 @cindex directives and instructions
2974 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2975 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2977 A statement begins with zero or more labels, optionally followed by a
2978 key symbol which determines what kind of statement it is. The key
2979 symbol determines the syntax of the rest of the statement. If the
2980 symbol begins with a dot @samp{.} then the statement is an assembler
2981 directive: typically valid for any computer. If the symbol begins with
2982 a letter the statement is an assembly language @dfn{instruction}: it
2983 assembles into a machine language instruction.
2985 Different versions of @command{@value{AS}} for different computers
2986 recognize different instructions. In fact, the same symbol may
2987 represent a different instruction in a different computer's assembly
2991 @cindex @code{:} (label)
2992 @cindex label (@code{:})
2993 A label is a symbol immediately followed by a colon (@code{:}).
2994 Whitespace before a label or after a colon is permitted, but you may not
2995 have whitespace between a label's symbol and its colon. @xref{Labels}.
2998 For HPPA targets, labels need not be immediately followed by a colon, but
2999 the definition of a label must begin in column zero. This also implies that
3000 only one label may be defined on each line.
3004 label: .directive followed by something
3005 another_label: # This is an empty statement.
3006 instruction operand_1, operand_2, @dots{}
3013 A constant is a number, written so that its value is known by
3014 inspection, without knowing any context. Like this:
3017 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
3018 .ascii "Ring the bell\7" # A string constant.
3019 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
3020 .float 0f-314159265358979323846264338327\
3021 95028841971.693993751E-40 # - pi, a flonum.
3026 * Characters:: Character Constants
3027 * Numbers:: Number Constants
3031 @subsection Character Constants
3033 @cindex character constants
3034 @cindex constants, character
3035 There are two kinds of character constants. A @dfn{character} stands
3036 for one character in one byte and its value may be used in
3037 numeric expressions. String constants (properly called string
3038 @emph{literals}) are potentially many bytes and their values may not be
3039 used in arithmetic expressions.
3043 * Chars:: Characters
3047 @subsubsection Strings
3049 @cindex string constants
3050 @cindex constants, string
3051 A @dfn{string} is written between double-quotes. It may contain
3052 double-quotes or null characters. The way to get special characters
3053 into a string is to @dfn{escape} these characters: precede them with
3054 a backslash @samp{\} character. For example @samp{\\} represents
3055 one backslash: the first @code{\} is an escape which tells
3056 @command{@value{AS}} to interpret the second character literally as a backslash
3057 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
3058 escape character). The complete list of escapes follows.
3060 @cindex escape codes, character
3061 @cindex character escape codes
3062 @c NOTE: Cindex entries must not start with a backlash character.
3063 @c NOTE: This confuses the pdf2texi script when it is creating the
3064 @c NOTE: index based upon the first character and so it generates:
3065 @c NOTE: \initial {\\}
3066 @c NOTE: which then results in the error message:
3067 @c NOTE: Argument of \\ has an extra }.
3068 @c NOTE: So in the index entries below a space character has been
3069 @c NOTE: prepended to avoid this problem.
3072 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
3074 @cindex @code{ \b} (backspace character)
3075 @cindex backspace (@code{\b})
3077 Mnemonic for backspace; for ASCII this is octal code 010.
3080 @c Mnemonic for EOText; for ASCII this is octal code 004.
3082 @cindex @code{ \f} (formfeed character)
3083 @cindex formfeed (@code{\f})
3085 Mnemonic for FormFeed; for ASCII this is octal code 014.
3087 @cindex @code{ \n} (newline character)
3088 @cindex newline (@code{\n})
3090 Mnemonic for newline; for ASCII this is octal code 012.
3093 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
3095 @cindex @code{ \r} (carriage return character)
3096 @cindex carriage return (@code{backslash-r})
3098 Mnemonic for carriage-Return; for ASCII this is octal code 015.
3101 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
3102 @c other assemblers.
3104 @cindex @code{ \t} (tab)
3105 @cindex tab (@code{\t})
3107 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
3110 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
3111 @c @item \x @var{digit} @var{digit} @var{digit}
3112 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
3114 @cindex @code{ \@var{ddd}} (octal character code)
3115 @cindex octal character code (@code{\@var{ddd}})
3116 @item \ @var{digit} @var{digit} @var{digit}
3117 An octal character code. The numeric code is 3 octal digits.
3118 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
3119 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3121 @cindex @code{ \@var{xd...}} (hex character code)
3122 @cindex hex character code (@code{\@var{xd...}})
3123 @item \@code{x} @var{hex-digits...}
3124 A hex character code. All trailing hex digits are combined. Either upper or
3125 lower case @code{x} works.
3127 @cindex @code{ \\} (@samp{\} character)
3128 @cindex backslash (@code{\\})
3130 Represents one @samp{\} character.
3133 @c Represents one @samp{'} (accent acute) character.
3134 @c This is needed in single character literals
3135 @c (@xref{Characters,,Character Constants}.) to represent
3138 @cindex @code{ \"} (doublequote character)
3139 @cindex doublequote (@code{\"})
3141 Represents one @samp{"} character. Needed in strings to represent
3142 this character, because an unescaped @samp{"} would end the string.
3144 @item \ @var{anything-else}
3145 Any other character when escaped by @kbd{\} gives a warning, but
3146 assembles as if the @samp{\} was not present. The idea is that if
3147 you used an escape sequence you clearly didn't want the literal
3148 interpretation of the following character. However @command{@value{AS}} has no
3149 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3150 code and warns you of the fact.
3153 Which characters are escapable, and what those escapes represent,
3154 varies widely among assemblers. The current set is what we think
3155 the BSD 4.2 assembler recognizes, and is a subset of what most C
3156 compilers recognize. If you are in doubt, do not use an escape
3160 @subsubsection Characters
3162 @cindex single character constant
3163 @cindex character, single
3164 @cindex constant, single character
3165 A single character may be written as a single quote immediately followed by
3166 that character. Some backslash escapes apply to characters, @code{\b},
3167 @code{\f}, @code{\n}, @code{\r}, @code{\t}, and @code{\"} with the same meaning
3168 as for strings, plus @code{\'} for a single quote. So if you want to write the
3169 character backslash, you must write @kbd{'\\} where the first @code{\} escapes
3170 the second @code{\}. As you can see, the quote is an acute accent, not a grave
3173 @ifclear abnormal-separator
3174 (or semicolon @samp{;})
3176 @ifset abnormal-separator
3178 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3183 immediately following an acute accent is taken as a literal character
3184 and does not count as the end of a statement. The value of a character
3185 constant in a numeric expression is the machine's byte-wide code for
3186 that character. @command{@value{AS}} assumes your character code is ASCII:
3187 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3190 @subsection Number Constants
3192 @cindex constants, number
3193 @cindex number constants
3194 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3195 are stored in the target machine. @emph{Integers} are numbers that
3196 would fit into an @code{int} in the C language. @emph{Bignums} are
3197 integers, but they are stored in more than 32 bits. @emph{Flonums}
3198 are floating point numbers, described below.
3201 * Integers:: Integers
3209 @subsubsection Integers
3211 @cindex constants, integer
3213 @cindex binary integers
3214 @cindex integers, binary
3215 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3216 the binary digits @samp{01}.
3218 @cindex octal integers
3219 @cindex integers, octal
3220 An octal integer is @samp{0} followed by zero or more of the octal
3221 digits (@samp{01234567}).
3223 @cindex decimal integers
3224 @cindex integers, decimal
3225 A decimal integer starts with a non-zero digit followed by zero or
3226 more digits (@samp{0123456789}).
3228 @cindex hexadecimal integers
3229 @cindex integers, hexadecimal
3230 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3231 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3233 Integers have the usual values. To denote a negative integer, use
3234 the prefix operator @samp{-} discussed under expressions
3235 (@pxref{Prefix Ops,,Prefix Operators}).
3238 @subsubsection Bignums
3241 @cindex constants, bignum
3242 A @dfn{bignum} has the same syntax and semantics as an integer
3243 except that the number (or its negative) takes more than 32 bits to
3244 represent in binary. The distinction is made because in some places
3245 integers are permitted while bignums are not.
3248 @subsubsection Flonums
3250 @cindex floating point numbers
3251 @cindex constants, floating point
3253 @cindex precision, floating point
3254 A @dfn{flonum} represents a floating point number. The translation is
3255 indirect: a decimal floating point number from the text is converted by
3256 @command{@value{AS}} to a generic binary floating point number of more than
3257 sufficient precision. This generic floating point number is converted
3258 to a particular computer's floating point format (or formats) by a
3259 portion of @command{@value{AS}} specialized to that computer.
3261 A flonum is written by writing (in order)
3266 (@samp{0} is optional on the HPPA.)
3270 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3272 @kbd{e} is recommended. Case is not important.
3274 @c FIXME: verify if flonum syntax really this vague for most cases
3275 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3276 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3279 On the H8/300 and Renesas / SuperH SH architectures, the letter must be
3280 one of the letters @samp{DFPRSX} (in upper or lower case).
3282 On the ARC, the letter must be one of the letters @samp{DFRS}
3283 (in upper or lower case).
3285 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3289 One of the letters @samp{DFRS} (in upper or lower case).
3292 One of the letters @samp{DFPRSX} (in upper or lower case).
3295 The letter @samp{E} (upper case only).
3300 An optional sign: either @samp{+} or @samp{-}.
3303 An optional @dfn{integer part}: zero or more decimal digits.
3306 An optional @dfn{fractional part}: @samp{.} followed by zero
3307 or more decimal digits.
3310 An optional exponent, consisting of:
3314 An @samp{E} or @samp{e}.
3315 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3316 @c principle this can perfectly well be different on different targets.
3318 Optional sign: either @samp{+} or @samp{-}.
3320 One or more decimal digits.
3325 At least one of the integer part or the fractional part must be
3326 present. The floating point number has the usual base-10 value.
3328 @command{@value{AS}} does all processing using integers. Flonums are computed
3329 independently of any floating point hardware in the computer running
3330 @command{@value{AS}}.
3333 @chapter Sections and Relocation
3338 * Secs Background:: Background
3339 * Ld Sections:: Linker Sections
3340 * As Sections:: Assembler Internal Sections
3341 * Sub-Sections:: Sub-Sections
3345 @node Secs Background
3348 Roughly, a section is a range of addresses, with no gaps; all data
3349 ``in'' those addresses is treated the same for some particular purpose.
3350 For example there may be a ``read only'' section.
3352 @cindex linker, and assembler
3353 @cindex assembler, and linker
3354 The linker @code{@value{LD}} reads many object files (partial programs) and
3355 combines their contents to form a runnable program. When @command{@value{AS}}
3356 emits an object file, the partial program is assumed to start at address 0.
3357 @code{@value{LD}} assigns the final addresses for the partial program, so that
3358 different partial programs do not overlap. This is actually an
3359 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3362 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3363 addresses. These blocks slide to their run-time addresses as rigid
3364 units; their length does not change and neither does the order of bytes
3365 within them. Such a rigid unit is called a @emph{section}. Assigning
3366 run-time addresses to sections is called @dfn{relocation}. It includes
3367 the task of adjusting mentions of object-file addresses so they refer to
3368 the proper run-time addresses.
3370 For the H8/300, and for the Renesas / SuperH SH,
3371 @command{@value{AS}} pads sections if needed to
3372 ensure they end on a word (sixteen bit) boundary.
3375 @cindex standard assembler sections
3376 An object file written by @command{@value{AS}} has at least three sections, any
3377 of which may be empty. These are named @dfn{text}, @dfn{data} and
3382 When it generates COFF or ELF output,
3384 @command{@value{AS}} can also generate whatever other named sections you specify
3385 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3386 If you do not use any directives that place output in the @samp{.text}
3387 or @samp{.data} sections, these sections still exist, but are empty.
3392 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3394 @command{@value{AS}} can also generate whatever other named sections you
3395 specify using the @samp{.space} and @samp{.subspace} directives. See
3396 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3397 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3398 assembler directives.
3401 Additionally, @command{@value{AS}} uses different names for the standard
3402 text, data, and bss sections when generating SOM output. Program text
3403 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3404 BSS into @samp{$BSS$}.
3408 Within the object file, the text section starts at address @code{0}, the
3409 data section follows, and the bss section follows the data section.
3412 When generating either SOM or ELF output files on the HPPA, the text
3413 section starts at address @code{0}, the data section at address
3414 @code{0x4000000}, and the bss section follows the data section.
3417 To let @code{@value{LD}} know which data changes when the sections are
3418 relocated, and how to change that data, @command{@value{AS}} also writes to the
3419 object file details of the relocation needed. To perform relocation
3420 @code{@value{LD}} must know, each time an address in the object
3424 Where in the object file is the beginning of this reference to
3427 How long (in bytes) is this reference?
3429 Which section does the address refer to? What is the numeric value of
3431 (@var{address}) @minus{} (@var{start-address of section})?
3434 Is the reference to an address ``Program-Counter relative''?
3437 @cindex addresses, format of
3438 @cindex section-relative addressing
3439 In fact, every address @command{@value{AS}} ever uses is expressed as
3441 (@var{section}) + (@var{offset into section})
3444 Further, most expressions @command{@value{AS}} computes have this section-relative
3447 (For some object formats, such as SOM for the HPPA, some expressions are
3448 symbol-relative instead.)
3451 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3452 @var{N} into section @var{secname}.''
3454 Apart from text, data and bss sections you need to know about the
3455 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3456 addresses in the absolute section remain unchanged. For example, address
3457 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3458 @code{@value{LD}}. Although the linker never arranges two partial programs'
3459 data sections with overlapping addresses after linking, @emph{by definition}
3460 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3461 part of a program is always the same address when the program is running as
3462 address @code{@{absolute@ 239@}} in any other part of the program.
3464 The idea of sections is extended to the @dfn{undefined} section. Any
3465 address whose section is unknown at assembly time is by definition
3466 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3467 Since numbers are always defined, the only way to generate an undefined
3468 address is to mention an undefined symbol. A reference to a named
3469 common block would be such a symbol: its value is unknown at assembly
3470 time so it has section @emph{undefined}.
3472 By analogy the word @emph{section} is used to describe groups of sections in
3473 the linked program. @code{@value{LD}} puts all partial programs' text
3474 sections in contiguous addresses in the linked program. It is
3475 customary to refer to the @emph{text section} of a program, meaning all
3476 the addresses of all partial programs' text sections. Likewise for
3477 data and bss sections.
3479 Some sections are manipulated by @code{@value{LD}}; others are invented for
3480 use of @command{@value{AS}} and have no meaning except during assembly.
3483 @section Linker Sections
3484 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3489 @cindex named sections
3490 @cindex sections, named
3491 @item named sections
3494 @cindex text section
3495 @cindex data section
3499 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3500 separate but equal sections. Anything you can say of one section is
3503 When the program is running, however, it is
3504 customary for the text section to be unalterable. The
3505 text section is often shared among processes: it contains
3506 instructions, constants and the like. The data section of a running
3507 program is usually alterable: for example, C variables would be stored
3508 in the data section.
3513 This section contains zeroed bytes when your program begins running. It
3514 is used to hold uninitialized variables or common storage. The length of
3515 each partial program's bss section is important, but because it starts
3516 out containing zeroed bytes there is no need to store explicit zero
3517 bytes in the object file. The bss section was invented to eliminate
3518 those explicit zeros from object files.
3520 @cindex absolute section
3521 @item absolute section
3522 Address 0 of this section is always ``relocated'' to runtime address 0.
3523 This is useful if you want to refer to an address that @code{@value{LD}} must
3524 not change when relocating. In this sense we speak of absolute
3525 addresses being ``unrelocatable'': they do not change during relocation.
3527 @cindex undefined section
3528 @item undefined section
3529 This ``section'' is a catch-all for address references to objects not in
3530 the preceding sections.
3531 @c FIXME: ref to some other doc on obj-file formats could go here.
3534 @cindex relocation example
3535 An idealized example of three relocatable sections follows.
3537 The example uses the traditional section names @samp{.text} and @samp{.data}.
3539 Memory addresses are on the horizontal axis.
3543 @c END TEXI2ROFF-KILL
3546 partial program # 1: |ttttt|dddd|00|
3553 partial program # 2: |TTT|DDD|000|
3556 +--+---+-----+--+----+---+-----+~~
3557 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3558 +--+---+-----+--+----+---+-----+~~
3560 addresses: 0 @dots{}
3567 \line{\it Partial program \#1: \hfil}
3568 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3569 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3571 \line{\it Partial program \#2: \hfil}
3572 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3573 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3575 \line{\it linked program: \hfil}
3576 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3577 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3578 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3579 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3581 \line{\it addresses: \hfil}
3585 @c END TEXI2ROFF-KILL
3588 @section Assembler Internal Sections
3590 @cindex internal assembler sections
3591 @cindex sections in messages, internal
3592 These sections are meant only for the internal use of @command{@value{AS}}. They
3593 have no meaning at run-time. You do not really need to know about these
3594 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3595 warning messages, so it might be helpful to have an idea of their
3596 meanings to @command{@value{AS}}. These sections are used to permit the
3597 value of every expression in your assembly language program to be a
3598 section-relative address.
3601 @cindex assembler internal logic error
3602 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3603 An internal assembler logic error has been found. This means there is a
3604 bug in the assembler.
3606 @cindex expr (internal section)
3608 The assembler stores complex expression internally as combinations of
3609 symbols. When it needs to represent an expression as a symbol, it puts
3610 it in the expr section.
3612 @c FIXME item transfer[t] vector preload
3613 @c FIXME item transfer[t] vector postload
3614 @c FIXME item register
3618 @section Sub-Sections
3620 @cindex numbered subsections
3621 @cindex grouping data
3627 fall into two sections: text and data.
3629 You may have separate groups of
3631 data in named sections
3635 data in named sections
3641 that you want to end up near to each other in the object file, even though they
3642 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3643 use @dfn{subsections} for this purpose. Within each section, there can be
3644 numbered subsections with values from 0 to 8192. Objects assembled into the
3645 same subsection go into the object file together with other objects in the same
3646 subsection. For example, a compiler might want to store constants in the text
3647 section, but might not want to have them interspersed with the program being
3648 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3649 section of code being output, and a @samp{.text 1} before each group of
3650 constants being output.
3652 Subsections are optional. If you do not use subsections, everything
3653 goes in subsection number zero.
3656 Each subsection is zero-padded up to a multiple of four bytes.
3657 (Subsections may be padded a different amount on different flavors
3658 of @command{@value{AS}}.)
3662 On the H8/300 platform, each subsection is zero-padded to a word
3663 boundary (two bytes).
3664 The same is true on the Renesas SH.
3668 Subsections appear in your object file in numeric order, lowest numbered
3669 to highest. (All this to be compatible with other people's assemblers.)
3670 The object file contains no representation of subsections; @code{@value{LD}} and
3671 other programs that manipulate object files see no trace of them.
3672 They just see all your text subsections as a text section, and all your
3673 data subsections as a data section.
3675 To specify which subsection you want subsequent statements assembled
3676 into, use a numeric argument to specify it, in a @samp{.text
3677 @var{expression}} or a @samp{.data @var{expression}} statement.
3680 When generating COFF output, you
3685 can also use an extra subsection
3686 argument with arbitrary named sections: @samp{.section @var{name},
3691 When generating ELF output, you
3696 can also use the @code{.subsection} directive (@pxref{SubSection})
3697 to specify a subsection: @samp{.subsection @var{expression}}.
3699 @var{Expression} should be an absolute expression
3700 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3701 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3702 begins in @code{text 0}. For instance:
3704 .text 0 # The default subsection is text 0 anyway.
3705 .ascii "This lives in the first text subsection. *"
3707 .ascii "But this lives in the second text subsection."
3709 .ascii "This lives in the data section,"
3710 .ascii "in the first data subsection."
3712 .ascii "This lives in the first text section,"
3713 .ascii "immediately following the asterisk (*)."
3716 Each section has a @dfn{location counter} incremented by one for every byte
3717 assembled into that section. Because subsections are merely a convenience
3718 restricted to @command{@value{AS}} there is no concept of a subsection location
3719 counter. There is no way to directly manipulate a location counter---but the
3720 @code{.align} directive changes it, and any label definition captures its
3721 current value. The location counter of the section where statements are being
3722 assembled is said to be the @dfn{active} location counter.
3725 @section bss Section
3728 @cindex common variable storage
3729 The bss section is used for local common variable storage.
3730 You may allocate address space in the bss section, but you may
3731 not dictate data to load into it before your program executes. When
3732 your program starts running, all the contents of the bss
3733 section are zeroed bytes.
3735 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3736 @ref{Lcomm,,@code{.lcomm}}.
3738 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3739 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3742 When assembling for a target which supports multiple sections, such as ELF or
3743 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3744 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3745 section. Typically the section will only contain symbol definitions and
3746 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3753 Symbols are a central concept: the programmer uses symbols to name
3754 things, the linker uses symbols to link, and the debugger uses symbols
3758 @cindex debuggers, and symbol order
3759 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3760 the same order they were declared. This may break some debuggers.
3765 * Setting Symbols:: Giving Symbols Other Values
3766 * Symbol Names:: Symbol Names
3767 * Dot:: The Special Dot Symbol
3768 * Symbol Attributes:: Symbol Attributes
3775 A @dfn{label} is written as a symbol immediately followed by a colon
3776 @samp{:}. The symbol then represents the current value of the
3777 active location counter, and is, for example, a suitable instruction
3778 operand. You are warned if you use the same symbol to represent two
3779 different locations: the first definition overrides any other
3783 On the HPPA, the usual form for a label need not be immediately followed by a
3784 colon, but instead must start in column zero. Only one label may be defined on
3785 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3786 provides a special directive @code{.label} for defining labels more flexibly.
3789 @node Setting Symbols
3790 @section Giving Symbols Other Values
3792 @cindex assigning values to symbols
3793 @cindex symbol values, assigning
3794 A symbol can be given an arbitrary value by writing a symbol, followed
3795 by an equals sign @samp{=}, followed by an expression
3796 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3797 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3798 equals sign @samp{=}@samp{=} here represents an equivalent of the
3799 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3802 Blackfin does not support symbol assignment with @samp{=}.
3806 @section Symbol Names
3808 @cindex symbol names
3809 @cindex names, symbol
3810 @ifclear SPECIAL-SYMS
3811 Symbol names begin with a letter or with one of @samp{._}. On most
3812 machines, you can also use @code{$} in symbol names; exceptions are
3813 noted in @ref{Machine Dependencies}. That character may be followed by any
3814 string of digits, letters, dollar signs (unless otherwise noted for a
3815 particular target machine), and underscores.
3819 Symbol names begin with a letter or with one of @samp{._}. On the
3820 Renesas SH you can also use @code{$} in symbol names. That
3821 character may be followed by any string of digits, letters, dollar signs (save
3822 on the H8/300), and underscores.
3826 Case of letters is significant: @code{foo} is a different symbol name
3829 Symbol names do not start with a digit. An exception to this rule is made for
3830 Local Labels. See below.
3832 Multibyte characters are supported. To generate a symbol name containing
3833 multibyte characters enclose it within double quotes and use escape codes. cf
3834 @xref{Strings}. Generating a multibyte symbol name from a label is not
3835 currently supported.
3837 Each symbol has exactly one name. Each name in an assembly language program
3838 refers to exactly one symbol. You may use that symbol name any number of times
3841 @subheading Local Symbol Names
3843 @cindex local symbol names
3844 @cindex symbol names, local
3845 A local symbol is any symbol beginning with certain local label prefixes.
3846 By default, the local label prefix is @samp{.L} for ELF systems or
3847 @samp{L} for traditional a.out systems, but each target may have its own
3848 set of local label prefixes.
3850 On the HPPA local symbols begin with @samp{L$}.
3853 Local symbols are defined and used within the assembler, but they are
3854 normally not saved in object files. Thus, they are not visible when debugging.
3855 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3856 to retain the local symbols in the object files.
3858 @subheading Local Labels
3860 @cindex local labels
3861 @cindex temporary symbol names
3862 @cindex symbol names, temporary
3863 Local labels are different from local symbols. Local labels help compilers and
3864 programmers use names temporarily. They create symbols which are guaranteed to
3865 be unique over the entire scope of the input source code and which can be
3866 referred to by a simple notation. To define a local label, write a label of
3867 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3868 To refer to the most recent previous definition of that label write
3869 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3870 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3871 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3873 There is no restriction on how you can use these labels, and you can reuse them
3874 too. So that it is possible to repeatedly define the same local label (using
3875 the same number @samp{@b{N}}), although you can only refer to the most recently
3876 defined local label of that number (for a backwards reference) or the next
3877 definition of a specific local label for a forward reference. It is also worth
3878 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3879 implemented in a slightly more efficient manner than the others.
3890 Which is the equivalent of:
3893 label_1: branch label_3
3894 label_2: branch label_1
3895 label_3: branch label_4
3896 label_4: branch label_3
3899 Local label names are only a notational device. They are immediately
3900 transformed into more conventional symbol names before the assembler uses them.
3901 The symbol names are stored in the symbol table, appear in error messages, and
3902 are optionally emitted to the object file. The names are constructed using
3906 @item @emph{local label prefix}
3907 All local symbols begin with the system-specific local label prefix.
3908 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3909 that start with the local label prefix. These labels are
3910 used for symbols you are never intended to see. If you use the
3911 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3912 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3913 you may use them in debugging.
3916 This is the number that was used in the local label definition. So if the
3917 label is written @samp{55:} then the number is @samp{55}.
3920 This unusual character is included so you do not accidentally invent a symbol
3921 of the same name. The character has ASCII value of @samp{\002} (control-B).
3923 @item @emph{ordinal number}
3924 This is a serial number to keep the labels distinct. The first definition of
3925 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3926 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3927 the number @samp{1} and its 15th definition gets @samp{15} as well.
3930 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3931 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3933 @subheading Dollar Local Labels
3934 @cindex dollar local symbols
3936 On some targets @code{@value{AS}} also supports an even more local form of
3937 local labels called dollar labels. These labels go out of scope (i.e., they
3938 become undefined) as soon as a non-local label is defined. Thus they remain
3939 valid for only a small region of the input source code. Normal local labels,
3940 by contrast, remain in scope for the entire file, or until they are redefined
3941 by another occurrence of the same local label.
3943 Dollar labels are defined in exactly the same way as ordinary local labels,
3944 except that they have a dollar sign suffix to their numeric value, e.g.,
3947 They can also be distinguished from ordinary local labels by their transformed
3948 names which use ASCII character @samp{\001} (control-A) as the magic character
3949 to distinguish them from ordinary labels. For example, the fifth definition of
3950 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3953 @section The Special Dot Symbol
3955 @cindex dot (symbol)
3956 @cindex @code{.} (symbol)
3957 @cindex current address
3958 @cindex location counter
3959 The special symbol @samp{.} refers to the current address that
3960 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3961 .long .} defines @code{melvin} to contain its own address.
3962 Assigning a value to @code{.} is treated the same as a @code{.org}
3964 @ifclear no-space-dir
3965 Thus, the expression @samp{.=.+4} is the same as saying
3969 @node Symbol Attributes
3970 @section Symbol Attributes
3972 @cindex symbol attributes
3973 @cindex attributes, symbol
3974 Every symbol has, as well as its name, the attributes ``Value'' and
3975 ``Type''. Depending on output format, symbols can also have auxiliary
3978 The detailed definitions are in @file{a.out.h}.
3981 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3982 all these attributes, and probably won't warn you. This makes the
3983 symbol an externally defined symbol, which is generally what you
3987 * Symbol Value:: Value
3988 * Symbol Type:: Type
3990 * a.out Symbols:: Symbol Attributes: @code{a.out}
3993 * COFF Symbols:: Symbol Attributes for COFF
3996 * SOM Symbols:: Symbol Attributes for SOM
4003 @cindex value of a symbol
4004 @cindex symbol value
4005 The value of a symbol is (usually) 32 bits. For a symbol which labels a
4006 location in the text, data, bss or absolute sections the value is the
4007 number of addresses from the start of that section to the label.
4008 Naturally for text, data and bss sections the value of a symbol changes
4009 as @code{@value{LD}} changes section base addresses during linking. Absolute
4010 symbols' values do not change during linking: that is why they are
4013 The value of an undefined symbol is treated in a special way. If it is
4014 0 then the symbol is not defined in this assembler source file, and
4015 @code{@value{LD}} tries to determine its value from other files linked into the
4016 same program. You make this kind of symbol simply by mentioning a symbol
4017 name without defining it. A non-zero value represents a @code{.comm}
4018 common declaration. The value is how much common storage to reserve, in
4019 bytes (addresses). The symbol refers to the first address of the
4025 @cindex type of a symbol
4027 The type attribute of a symbol contains relocation (section)
4028 information, any flag settings indicating that a symbol is external, and
4029 (optionally), other information for linkers and debuggers. The exact
4030 format depends on the object-code output format in use.
4034 @subsection Symbol Attributes: @code{a.out}
4036 @cindex @code{a.out} symbol attributes
4037 @cindex symbol attributes, @code{a.out}
4040 * Symbol Desc:: Descriptor
4041 * Symbol Other:: Other
4045 @subsubsection Descriptor
4047 @cindex descriptor, of @code{a.out} symbol
4048 This is an arbitrary 16-bit value. You may establish a symbol's
4049 descriptor value by using a @code{.desc} statement
4050 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
4051 @command{@value{AS}}.
4054 @subsubsection Other
4056 @cindex other attribute, of @code{a.out} symbol
4057 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
4062 @subsection Symbol Attributes for COFF
4064 @cindex COFF symbol attributes
4065 @cindex symbol attributes, COFF
4067 The COFF format supports a multitude of auxiliary symbol attributes;
4068 like the primary symbol attributes, they are set between @code{.def} and
4069 @code{.endef} directives.
4071 @subsubsection Primary Attributes
4073 @cindex primary attributes, COFF symbols
4074 The symbol name is set with @code{.def}; the value and type,
4075 respectively, with @code{.val} and @code{.type}.
4077 @subsubsection Auxiliary Attributes
4079 @cindex auxiliary attributes, COFF symbols
4080 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
4081 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
4082 table information for COFF.
4087 @subsection Symbol Attributes for SOM
4089 @cindex SOM symbol attributes
4090 @cindex symbol attributes, SOM
4092 The SOM format for the HPPA supports a multitude of symbol attributes set with
4093 the @code{.EXPORT} and @code{.IMPORT} directives.
4095 The attributes are described in @cite{HP9000 Series 800 Assembly
4096 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
4097 @code{EXPORT} assembler directive documentation.
4101 @chapter Expressions
4105 @cindex numeric values
4106 An @dfn{expression} specifies an address or numeric value.
4107 Whitespace may precede and/or follow an expression.
4109 The result of an expression must be an absolute number, or else an offset into
4110 a particular section. If an expression is not absolute, and there is not
4111 enough information when @command{@value{AS}} sees the expression to know its
4112 section, a second pass over the source program might be necessary to interpret
4113 the expression---but the second pass is currently not implemented.
4114 @command{@value{AS}} aborts with an error message in this situation.
4117 * Empty Exprs:: Empty Expressions
4118 * Integer Exprs:: Integer Expressions
4122 @section Empty Expressions
4124 @cindex empty expressions
4125 @cindex expressions, empty
4126 An empty expression has no value: it is just whitespace or null.
4127 Wherever an absolute expression is required, you may omit the
4128 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4129 is compatible with other assemblers.
4132 @section Integer Expressions
4134 @cindex integer expressions
4135 @cindex expressions, integer
4136 An @dfn{integer expression} is one or more @emph{arguments} delimited
4137 by @emph{operators}.
4140 * Arguments:: Arguments
4141 * Operators:: Operators
4142 * Prefix Ops:: Prefix Operators
4143 * Infix Ops:: Infix Operators
4147 @subsection Arguments
4149 @cindex expression arguments
4150 @cindex arguments in expressions
4151 @cindex operands in expressions
4152 @cindex arithmetic operands
4153 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4154 contexts arguments are sometimes called ``arithmetic operands''. In
4155 this manual, to avoid confusing them with the ``instruction operands'' of
4156 the machine language, we use the term ``argument'' to refer to parts of
4157 expressions only, reserving the word ``operand'' to refer only to machine
4158 instruction operands.
4160 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4161 @var{section} is one of text, data, bss, absolute,
4162 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4165 Numbers are usually integers.
4167 A number can be a flonum or bignum. In this case, you are warned
4168 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4169 these 32 bits are an integer. You may write integer-manipulating
4170 instructions that act on exotic constants, compatible with other
4173 @cindex subexpressions
4174 Subexpressions are a left parenthesis @samp{(} followed by an integer
4175 expression, followed by a right parenthesis @samp{)}; or a prefix
4176 operator followed by an argument.
4179 @subsection Operators
4181 @cindex operators, in expressions
4182 @cindex arithmetic functions
4183 @cindex functions, in expressions
4184 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4185 operators are followed by an argument. Infix operators appear
4186 between their arguments. Operators may be preceded and/or followed by
4190 @subsection Prefix Operator
4192 @cindex prefix operators
4193 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4194 one argument, which must be absolute.
4196 @c the tex/end tex stuff surrounding this small table is meant to make
4197 @c it align, on the printed page, with the similar table in the next
4198 @c section (which is inside an enumerate).
4200 \global\advance\leftskip by \itemindent
4205 @dfn{Negation}. Two's complement negation.
4207 @dfn{Complementation}. Bitwise not.
4211 \global\advance\leftskip by -\itemindent
4215 @subsection Infix Operators
4217 @cindex infix operators
4218 @cindex operators, permitted arguments
4219 @dfn{Infix operators} take two arguments, one on either side. Operators
4220 have precedence, but operations with equal precedence are performed left
4221 to right. Apart from @code{+} or @option{-}, both arguments must be
4222 absolute, and the result is absolute.
4225 @cindex operator precedence
4226 @cindex precedence of operators
4233 @dfn{Multiplication}.
4236 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4242 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4245 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4249 Intermediate precedence
4254 @dfn{Bitwise Inclusive Or}.
4260 @dfn{Bitwise Exclusive Or}.
4263 @dfn{Bitwise Or Not}.
4270 @cindex addition, permitted arguments
4271 @cindex plus, permitted arguments
4272 @cindex arguments for addition
4274 @dfn{Addition}. If either argument is absolute, the result has the section of
4275 the other argument. You may not add together arguments from different
4278 @cindex subtraction, permitted arguments
4279 @cindex minus, permitted arguments
4280 @cindex arguments for subtraction
4282 @dfn{Subtraction}. If the right argument is absolute, the
4283 result has the section of the left argument.
4284 If both arguments are in the same section, the result is absolute.
4285 You may not subtract arguments from different sections.
4286 @c FIXME is there still something useful to say about undefined - undefined ?
4288 @cindex comparison expressions
4289 @cindex expressions, comparison
4294 @dfn{Is Not Equal To}
4298 @dfn{Is Greater Than}
4300 @dfn{Is Greater Than Or Equal To}
4302 @dfn{Is Less Than Or Equal To}
4304 The comparison operators can be used as infix operators. A true results has a
4305 value of -1 whereas a false result has a value of 0. Note, these operators
4306 perform signed comparisons.
4309 @item Lowest Precedence
4318 These two logical operations can be used to combine the results of sub
4319 expressions. Note, unlike the comparison operators a true result returns a
4320 value of 1 but a false results does still return 0. Also note that the logical
4321 or operator has a slightly lower precedence than logical and.
4326 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4327 address; you can only have a defined section in one of the two arguments.
4330 @chapter Assembler Directives
4332 @cindex directives, machine independent
4333 @cindex pseudo-ops, machine independent
4334 @cindex machine independent directives
4335 All assembler directives have names that begin with a period (@samp{.}).
4336 The names are case insensitive for most targets, and usually written
4339 This chapter discusses directives that are available regardless of the
4340 target machine configuration for the @sc{gnu} assembler.
4342 Some machine configurations provide additional directives.
4343 @xref{Machine Dependencies}.
4346 @ifset machine-directives
4347 @xref{Machine Dependencies}, for additional directives.
4352 * Abort:: @code{.abort}
4354 * ABORT (COFF):: @code{.ABORT}
4357 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4358 * Altmacro:: @code{.altmacro}
4359 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4360 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4361 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4362 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4363 * Byte:: @code{.byte @var{expressions}}
4364 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4365 * Comm:: @code{.comm @var{symbol} , @var{length} }
4366 * Data:: @code{.data @var{subsection}}
4367 * Dc:: @code{.dc[@var{size}] @var{expressions}}
4368 * Dcb:: @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4369 * Ds:: @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4371 * Def:: @code{.def @var{name}}
4374 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4380 * Double:: @code{.double @var{flonums}}
4381 * Eject:: @code{.eject}
4382 * Else:: @code{.else}
4383 * Elseif:: @code{.elseif}
4386 * Endef:: @code{.endef}
4389 * Endfunc:: @code{.endfunc}
4390 * Endif:: @code{.endif}
4391 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4392 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4393 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4395 * Error:: @code{.error @var{string}}
4396 * Exitm:: @code{.exitm}
4397 * Extern:: @code{.extern}
4398 * Fail:: @code{.fail}
4399 * File:: @code{.file}
4400 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4401 * Float:: @code{.float @var{flonums}}
4402 * Func:: @code{.func}
4403 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4405 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4406 * Hidden:: @code{.hidden @var{names}}
4409 * hword:: @code{.hword @var{expressions}}
4410 * Ident:: @code{.ident}
4411 * If:: @code{.if @var{absolute expression}}
4412 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4413 * Include:: @code{.include "@var{file}"}
4414 * Int:: @code{.int @var{expressions}}
4416 * Internal:: @code{.internal @var{names}}
4419 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4420 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4421 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4422 * Lflags:: @code{.lflags}
4423 @ifclear no-line-dir
4424 * Line:: @code{.line @var{line-number}}
4427 * Linkonce:: @code{.linkonce [@var{type}]}
4428 * List:: @code{.list}
4429 * Ln:: @code{.ln @var{line-number}}
4430 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4431 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4433 * Local:: @code{.local @var{names}}
4436 * Long:: @code{.long @var{expressions}}
4438 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4441 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4442 * MRI:: @code{.mri @var{val}}
4443 * Noaltmacro:: @code{.noaltmacro}
4444 * Nolist:: @code{.nolist}
4445 * Nops:: @code{.nops @var{size}[, @var{control}]}
4446 * Octa:: @code{.octa @var{bignums}}
4447 * Offset:: @code{.offset @var{loc}}
4448 * Org:: @code{.org @var{new-lc}, @var{fill}}
4449 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4451 * PopSection:: @code{.popsection}
4452 * Previous:: @code{.previous}
4455 * Print:: @code{.print @var{string}}
4457 * Protected:: @code{.protected @var{names}}
4460 * Psize:: @code{.psize @var{lines}, @var{columns}}
4461 * Purgem:: @code{.purgem @var{name}}
4463 * PushSection:: @code{.pushsection @var{name}}
4466 * Quad:: @code{.quad @var{bignums}}
4467 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4468 * Rept:: @code{.rept @var{count}}
4469 * Sbttl:: @code{.sbttl "@var{subheading}"}
4471 * Scl:: @code{.scl @var{class}}
4474 * Section:: @code{.section @var{name}[, @var{flags}]}
4477 * Set:: @code{.set @var{symbol}, @var{expression}}
4478 * Short:: @code{.short @var{expressions}}
4479 * Single:: @code{.single @var{flonums}}
4481 * Size:: @code{.size [@var{name} , @var{expression}]}
4483 @ifclear no-space-dir
4484 * Skip:: @code{.skip @var{size} [,@var{fill}]}
4487 * Sleb128:: @code{.sleb128 @var{expressions}}
4488 @ifclear no-space-dir
4489 * Space:: @code{.space @var{size} [,@var{fill}]}
4492 * Stab:: @code{.stabd, .stabn, .stabs}
4495 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4496 * Struct:: @code{.struct @var{expression}}
4498 * SubSection:: @code{.subsection}
4499 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4503 * Tag:: @code{.tag @var{structname}}
4506 * Text:: @code{.text @var{subsection}}
4507 * Title:: @code{.title "@var{heading}"}
4509 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4512 * Uleb128:: @code{.uleb128 @var{expressions}}
4514 * Val:: @code{.val @var{addr}}
4518 * Version:: @code{.version "@var{string}"}
4519 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4520 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4523 * Warning:: @code{.warning @var{string}}
4524 * Weak:: @code{.weak @var{names}}
4525 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4526 * Word:: @code{.word @var{expressions}}
4527 @ifclear no-space-dir
4528 * Zero:: @code{.zero @var{size}}
4531 * 2byte:: @code{.2byte @var{expressions}}
4532 * 4byte:: @code{.4byte @var{expressions}}
4533 * 8byte:: @code{.8byte @var{bignums}}
4535 * Deprecated:: Deprecated Directives
4539 @section @code{.abort}
4541 @cindex @code{abort} directive
4542 @cindex stopping the assembly
4543 This directive stops the assembly immediately. It is for
4544 compatibility with other assemblers. The original idea was that the
4545 assembly language source would be piped into the assembler. If the sender
4546 of the source quit, it could use this directive tells @command{@value{AS}} to
4547 quit also. One day @code{.abort} will not be supported.
4551 @section @code{.ABORT} (COFF)
4553 @cindex @code{ABORT} directive
4554 When producing COFF output, @command{@value{AS}} accepts this directive as a
4555 synonym for @samp{.abort}.
4560 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4562 @cindex padding the location counter
4563 @cindex @code{align} directive
4564 Pad the location counter (in the current subsection) to a particular storage
4565 boundary. The first expression (which must be absolute) is the alignment
4566 required, as described below.
4568 The second expression (also absolute) gives the fill value to be stored in the
4569 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4570 padding bytes are normally zero. However, on most systems, if the section is
4571 marked as containing code and the fill value is omitted, the space is filled
4572 with no-op instructions.
4574 The third expression is also absolute, and is also optional. If it is present,
4575 it is the maximum number of bytes that should be skipped by this alignment
4576 directive. If doing the alignment would require skipping more bytes than the
4577 specified maximum, then the alignment is not done at all. You can omit the
4578 fill value (the second argument) entirely by simply using two commas after the
4579 required alignment; this can be useful if you want the alignment to be filled
4580 with no-op instructions when appropriate.
4582 The way the required alignment is specified varies from system to system.
4583 For the arc, hppa, i386 using ELF, iq2000, m68k, or1k,
4584 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4585 alignment request in bytes. For example @samp{.align 8} advances
4586 the location counter until it is a multiple of 8. If the location counter
4587 is already a multiple of 8, no change is needed. For the tic54x, the
4588 first expression is the alignment request in words.
4590 For other systems, including ppc, i386 using a.out format, arm and
4591 strongarm, it is the
4592 number of low-order zero bits the location counter must have after
4593 advancement. For example @samp{.align 3} advances the location
4594 counter until it is a multiple of 8. If the location counter is already a
4595 multiple of 8, no change is needed.
4597 This inconsistency is due to the different behaviors of the various
4598 native assemblers for these systems which GAS must emulate.
4599 GAS also provides @code{.balign} and @code{.p2align} directives,
4600 described later, which have a consistent behavior across all
4601 architectures (but are specific to GAS).
4604 @section @code{.altmacro}
4605 Enable alternate macro mode, enabling:
4608 @item LOCAL @var{name} [ , @dots{} ]
4609 One additional directive, @code{LOCAL}, is available. It is used to
4610 generate a string replacement for each of the @var{name} arguments, and
4611 replace any instances of @var{name} in each macro expansion. The
4612 replacement string is unique in the assembly, and different for each
4613 separate macro expansion. @code{LOCAL} allows you to write macros that
4614 define symbols, without fear of conflict between separate macro expansions.
4616 @item String delimiters
4617 You can write strings delimited in these other ways besides
4618 @code{"@var{string}"}:
4621 @item '@var{string}'
4622 You can delimit strings with single-quote characters.
4624 @item <@var{string}>
4625 You can delimit strings with matching angle brackets.
4628 @item single-character string escape
4629 To include any single character literally in a string (even if the
4630 character would otherwise have some special meaning), you can prefix the
4631 character with @samp{!} (an exclamation mark). For example, you can
4632 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4634 @item Expression results as strings
4635 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4636 and use the result as a string.
4640 @section @code{.ascii "@var{string}"}@dots{}
4642 @cindex @code{ascii} directive
4643 @cindex string literals
4644 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4645 separated by commas. It assembles each string (with no automatic
4646 trailing zero byte) into consecutive addresses.
4649 @section @code{.asciz "@var{string}"}@dots{}
4651 @cindex @code{asciz} directive
4652 @cindex zero-terminated strings
4653 @cindex null-terminated strings
4654 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4655 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4658 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4660 @cindex padding the location counter given number of bytes
4661 @cindex @code{balign} directive
4662 Pad the location counter (in the current subsection) to a particular
4663 storage boundary. The first expression (which must be absolute) is the
4664 alignment request in bytes. For example @samp{.balign 8} advances
4665 the location counter until it is a multiple of 8. If the location counter
4666 is already a multiple of 8, no change is needed.
4668 The second expression (also absolute) gives the fill value to be stored in the
4669 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4670 padding bytes are normally zero. However, on most systems, if the section is
4671 marked as containing code and the fill value is omitted, the space is filled
4672 with no-op instructions.
4674 The third expression is also absolute, and is also optional. If it is present,
4675 it is the maximum number of bytes that should be skipped by this alignment
4676 directive. If doing the alignment would require skipping more bytes than the
4677 specified maximum, then the alignment is not done at all. You can omit the
4678 fill value (the second argument) entirely by simply using two commas after the
4679 required alignment; this can be useful if you want the alignment to be filled
4680 with no-op instructions when appropriate.
4682 @cindex @code{balignw} directive
4683 @cindex @code{balignl} directive
4684 The @code{.balignw} and @code{.balignl} directives are variants of the
4685 @code{.balign} directive. The @code{.balignw} directive treats the fill
4686 pattern as a two byte word value. The @code{.balignl} directives treats the
4687 fill pattern as a four byte longword value. For example, @code{.balignw
4688 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4689 filled in with the value 0x368d (the exact placement of the bytes depends upon
4690 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4693 @node Bundle directives
4694 @section Bundle directives
4695 @subsection @code{.bundle_align_mode @var{abs-expr}}
4696 @cindex @code{bundle_align_mode} directive
4698 @cindex instruction bundle
4699 @cindex aligned instruction bundle
4700 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4701 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4702 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4703 disabled (which is the default state). If the argument it not zero, it
4704 gives the size of an instruction bundle as a power of two (as for the
4705 @code{.p2align} directive, @pxref{P2align}).
4707 For some targets, it's an ABI requirement that no instruction may span a
4708 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4709 instructions that starts on an aligned boundary. For example, if
4710 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4711 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4712 effect, no single instruction may span a boundary between bundles. If an
4713 instruction would start too close to the end of a bundle for the length of
4714 that particular instruction to fit within the bundle, then the space at the
4715 end of that bundle is filled with no-op instructions so the instruction
4716 starts in the next bundle. As a corollary, it's an error if any single
4717 instruction's encoding is longer than the bundle size.
4719 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4720 @cindex @code{bundle_lock} directive
4721 @cindex @code{bundle_unlock} directive
4722 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4723 allow explicit control over instruction bundle padding. These directives
4724 are only valid when @code{.bundle_align_mode} has been used to enable
4725 aligned instruction bundle mode. It's an error if they appear when
4726 @code{.bundle_align_mode} has not been used at all, or when the last
4727 directive was @w{@code{.bundle_align_mode 0}}.
4729 @cindex bundle-locked
4730 For some targets, it's an ABI requirement that certain instructions may
4731 appear only as part of specified permissible sequences of multiple
4732 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4733 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4734 instruction sequence. For purposes of aligned instruction bundle mode, a
4735 sequence starting with @code{.bundle_lock} and ending with
4736 @code{.bundle_unlock} is treated as a single instruction. That is, the
4737 entire sequence must fit into a single bundle and may not span a bundle
4738 boundary. If necessary, no-op instructions will be inserted before the
4739 first instruction of the sequence so that the whole sequence starts on an
4740 aligned bundle boundary. It's an error if the sequence is longer than the
4743 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4744 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4745 nested. That is, a second @code{.bundle_lock} directive before the next
4746 @code{.bundle_unlock} directive has no effect except that it must be
4747 matched by another closing @code{.bundle_unlock} so that there is the
4748 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4751 @section @code{.byte @var{expressions}}
4753 @cindex @code{byte} directive
4754 @cindex integers, one byte
4755 @code{.byte} expects zero or more expressions, separated by commas.
4756 Each expression is assembled into the next byte.
4758 @node CFI directives
4759 @section CFI directives
4760 @subsection @code{.cfi_sections @var{section_list}}
4761 @cindex @code{cfi_sections} directive
4762 @code{.cfi_sections} may be used to specify whether CFI directives
4763 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4764 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4765 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4766 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4767 directive is not used is @code{.cfi_sections .eh_frame}.
4769 On targets that support compact unwinding tables these can be generated
4770 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4772 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4773 which is used by the @value{TIC6X} target.
4775 The @code{.cfi_sections} directive can be repeated, with the same or different
4776 arguments, provided that CFI generation has not yet started. Once CFI
4777 generation has started however the section list is fixed and any attempts to
4778 redefine it will result in an error.
4780 @subsection @code{.cfi_startproc [simple]}
4781 @cindex @code{cfi_startproc} directive
4782 @code{.cfi_startproc} is used at the beginning of each function that
4783 should have an entry in @code{.eh_frame}. It initializes some internal
4784 data structures. Don't forget to close the function by
4785 @code{.cfi_endproc}.
4787 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4788 it also emits some architecture dependent initial CFI instructions.
4790 @subsection @code{.cfi_endproc}
4791 @cindex @code{cfi_endproc} directive
4792 @code{.cfi_endproc} is used at the end of a function where it closes its
4793 unwind entry previously opened by
4794 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4796 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4797 @cindex @code{cfi_personality} directive
4798 @code{.cfi_personality} defines personality routine and its encoding.
4799 @var{encoding} must be a constant determining how the personality
4800 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4801 argument is not present, otherwise second argument should be
4802 a constant or a symbol name. When using indirect encodings,
4803 the symbol provided should be the location where personality
4804 can be loaded from, not the personality routine itself.
4805 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4806 no personality routine.
4808 @subsection @code{.cfi_personality_id @var{id}}
4809 @cindex @code{cfi_personality_id} directive
4810 @code{cfi_personality_id} defines a personality routine by its index as
4811 defined in a compact unwinding format.
4812 Only valid when generating compact EH frames (i.e.
4813 with @code{.cfi_sections eh_frame_entry}.
4815 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4816 @cindex @code{cfi_fde_data} directive
4817 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4818 used for the current function. These are emitted inline in the
4819 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4820 in the @code{.gnu.extab} section otherwise.
4821 Only valid when generating compact EH frames (i.e.
4822 with @code{.cfi_sections eh_frame_entry}.
4824 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4825 @code{.cfi_lsda} defines LSDA and its encoding.
4826 @var{encoding} must be a constant determining how the LSDA
4827 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4828 argument is not present, otherwise the second argument should be a constant
4829 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4830 meaning that no LSDA is present.
4832 @subsection @code{.cfi_inline_lsda} [@var{align}]
4833 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4834 switches to the corresponding @code{.gnu.extab} section.
4835 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4836 Only valid when generating compact EH frames (i.e.
4837 with @code{.cfi_sections eh_frame_entry}.
4839 The table header and unwinding opcodes will be generated at this point,
4840 so that they are immediately followed by the LSDA data. The symbol
4841 referenced by the @code{.cfi_lsda} directive should still be defined
4842 in case a fallback FDE based encoding is used. The LSDA data is terminated
4843 by a section directive.
4845 The optional @var{align} argument specifies the alignment required.
4846 The alignment is specified as a power of two, as with the
4847 @code{.p2align} directive.
4849 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4850 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4851 address from @var{register} and add @var{offset} to it}.
4853 @subsection @code{.cfi_def_cfa_register @var{register}}
4854 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4855 now on @var{register} will be used instead of the old one. Offset
4858 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4859 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4860 remains the same, but @var{offset} is new. Note that it is the
4861 absolute offset that will be added to a defined register to compute
4864 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4865 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4866 value that is added/subtracted from the previous offset.
4868 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4869 Previous value of @var{register} is saved at offset @var{offset} from
4872 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4873 Previous value of @var{register} is CFA + @var{offset}.
4875 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4876 Previous value of @var{register} is saved at offset @var{offset} from
4877 the current CFA register. This is transformed to @code{.cfi_offset}
4878 using the known displacement of the CFA register from the CFA.
4879 This is often easier to use, because the number will match the
4880 code it's annotating.
4882 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4883 Previous value of @var{register1} is saved in register @var{register2}.
4885 @subsection @code{.cfi_restore @var{register}}
4886 @code{.cfi_restore} says that the rule for @var{register} is now the
4887 same as it was at the beginning of the function, after all initial
4888 instruction added by @code{.cfi_startproc} were executed.
4890 @subsection @code{.cfi_undefined @var{register}}
4891 From now on the previous value of @var{register} can't be restored anymore.
4893 @subsection @code{.cfi_same_value @var{register}}
4894 Current value of @var{register} is the same like in the previous frame,
4895 i.e. no restoration needed.
4897 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4898 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4899 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4900 places them in the current row. This is useful for situations where you have
4901 multiple @code{.cfi_*} directives that need to be undone due to the control
4902 flow of the program. For example, we could have something like this (assuming
4903 the CFA is the value of @code{rbp}):
4913 .cfi_def_cfa %rsp, 8
4916 /* Do something else */
4919 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4920 to the instructions before @code{label}. This means we'd have to add multiple
4921 @code{.cfi} directives after @code{label} to recreate the original save
4922 locations of the registers, as well as setting the CFA back to the value of
4923 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4935 .cfi_def_cfa %rsp, 8
4939 /* Do something else */
4942 That way, the rules for the instructions after @code{label} will be the same
4943 as before the first @code{.cfi_restore} without having to use multiple
4944 @code{.cfi} directives.
4946 @subsection @code{.cfi_return_column @var{register}}
4947 Change return column @var{register}, i.e. the return address is either
4948 directly in @var{register} or can be accessed by rules for @var{register}.
4950 @subsection @code{.cfi_signal_frame}
4951 Mark current function as signal trampoline.
4953 @subsection @code{.cfi_window_save}
4954 SPARC register window has been saved.
4956 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4957 Allows the user to add arbitrary bytes to the unwind info. One
4958 might use this to add OS-specific CFI opcodes, or generic CFI
4959 opcodes that GAS does not yet support.
4961 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4962 The current value of @var{register} is @var{label}. The value of @var{label}
4963 will be encoded in the output file according to @var{encoding}; see the
4964 description of @code{.cfi_personality} for details on this encoding.
4966 The usefulness of equating a register to a fixed label is probably
4967 limited to the return address register. Here, it can be useful to
4968 mark a code segment that has only one return address which is reached
4969 by a direct branch and no copy of the return address exists in memory
4970 or another register.
4973 @section @code{.comm @var{symbol} , @var{length} }
4975 @cindex @code{comm} directive
4976 @cindex symbol, common
4977 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4978 common symbol in one object file may be merged with a defined or common symbol
4979 of the same name in another object file. If @code{@value{LD}} does not see a
4980 definition for the symbol--just one or more common symbols--then it will
4981 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4982 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4983 the same name, and they do not all have the same size, it will allocate space
4984 using the largest size.
4987 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4988 an optional third argument. This is the desired alignment of the symbol,
4989 specified for ELF as a byte boundary (for example, an alignment of 16 means
4990 that the least significant 4 bits of the address should be zero), and for PE
4991 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4992 boundary). The alignment must be an absolute expression, and it must be a
4993 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4994 common symbol, it will use the alignment when placing the symbol. If no
4995 alignment is specified, @command{@value{AS}} will set the alignment to the
4996 largest power of two less than or equal to the size of the symbol, up to a
4997 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4998 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4999 @samp{--section-alignment} option; image file sections in PE are aligned to
5000 multiples of 4096, which is far too large an alignment for ordinary variables.
5001 It is rather the default alignment for (non-debug) sections within object
5002 (@samp{*.o}) files, which are less strictly aligned.}.
5006 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
5007 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
5011 @section @code{.data @var{subsection}}
5012 @cindex @code{data} directive
5014 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
5015 end of the data subsection numbered @var{subsection} (which is an
5016 absolute expression). If @var{subsection} is omitted, it defaults
5020 @section @code{.dc[@var{size}] @var{expressions}}
5021 @cindex @code{dc} directive
5023 The @code{.dc} directive expects zero or more @var{expressions} separated by
5024 commas. These expressions are evaluated and their values inserted into the
5025 current section. The size of the emitted value depends upon the suffix to the
5026 @code{.dc} directive:
5030 Emits N-bit values, where N is the size of an address on the target system.
5034 Emits double precision floating-point values.
5036 Emits 32-bit values.
5038 Emits single precision floating-point values.
5040 Emits 16-bit values.
5041 Note - this is true even on targets where the @code{.word} directive would emit
5044 Emits long double precision floating-point values.
5047 If no suffix is used then @samp{.w} is assumed.
5049 The byte ordering is target dependent, as is the size and format of floating
5053 @section @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
5054 @cindex @code{dcb} directive
5055 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5056 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5057 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5058 @var{size} suffix, if present, must be one of:
5062 Emits single byte values.
5064 Emits double-precision floating point values.
5066 Emits 4-byte values.
5068 Emits single-precision floating point values.
5070 Emits 2-byte values.
5072 Emits long double-precision floating point values.
5075 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5077 The byte ordering is target dependent, as is the size and format of floating
5081 @section @code{.ds[@var{size}] @var{number} [,@var{fill}]}
5082 @cindex @code{ds} directive
5083 This directive emits @var{number} copies of @var{fill}, each of @var{size}
5084 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
5085 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
5086 @var{size} suffix, if present, must be one of:
5090 Emits single byte values.
5092 Emits 8-byte values.
5094 Emits 4-byte values.
5096 Emits 12-byte values.
5098 Emits 4-byte values.
5100 Emits 2-byte values.
5102 Emits 12-byte values.
5105 Note - unlike the @code{.dcb} directive the @samp{.d}, @samp{.s} and @samp{.x}
5106 suffixes do not indicate that floating-point values are to be inserted.
5108 If the @var{size} suffix is omitted then @samp{.w} is assumed.
5110 The byte ordering is target dependent.
5115 @section @code{.def @var{name}}
5117 @cindex @code{def} directive
5118 @cindex COFF symbols, debugging
5119 @cindex debugging COFF symbols
5120 Begin defining debugging information for a symbol @var{name}; the
5121 definition extends until the @code{.endef} directive is encountered.
5126 @section @code{.desc @var{symbol}, @var{abs-expression}}
5128 @cindex @code{desc} directive
5129 @cindex COFF symbol descriptor
5130 @cindex symbol descriptor, COFF
5131 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5132 to the low 16 bits of an absolute expression.
5135 The @samp{.desc} directive is not available when @command{@value{AS}} is
5136 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5137 object format. For the sake of compatibility, @command{@value{AS}} accepts
5138 it, but produces no output, when configured for COFF.
5144 @section @code{.dim}
5146 @cindex @code{dim} directive
5147 @cindex COFF auxiliary symbol information
5148 @cindex auxiliary symbol information, COFF
5149 This directive is generated by compilers to include auxiliary debugging
5150 information in the symbol table. It is only permitted inside
5151 @code{.def}/@code{.endef} pairs.
5155 @section @code{.double @var{flonums}}
5157 @cindex @code{double} directive
5158 @cindex floating point numbers (double)
5159 @code{.double} expects zero or more flonums, separated by commas. It
5160 assembles floating point numbers.
5162 The exact kind of floating point numbers emitted depends on how
5163 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5167 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5168 in @sc{ieee} format.
5173 @section @code{.eject}
5175 @cindex @code{eject} directive
5176 @cindex new page, in listings
5177 @cindex page, in listings
5178 @cindex listing control: new page
5179 Force a page break at this point, when generating assembly listings.
5182 @section @code{.else}
5184 @cindex @code{else} directive
5185 @code{.else} is part of the @command{@value{AS}} support for conditional
5186 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5187 of code to be assembled if the condition for the preceding @code{.if}
5191 @section @code{.elseif}
5193 @cindex @code{elseif} directive
5194 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5195 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5196 @code{.if} block that would otherwise fill the entire @code{.else} section.
5199 @section @code{.end}
5201 @cindex @code{end} directive
5202 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5203 process anything in the file past the @code{.end} directive.
5207 @section @code{.endef}
5209 @cindex @code{endef} directive
5210 This directive flags the end of a symbol definition begun with
5215 @section @code{.endfunc}
5216 @cindex @code{endfunc} directive
5217 @code{.endfunc} marks the end of a function specified with @code{.func}.
5220 @section @code{.endif}
5222 @cindex @code{endif} directive
5223 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5224 it marks the end of a block of code that is only assembled
5225 conditionally. @xref{If,,@code{.if}}.
5228 @section @code{.equ @var{symbol}, @var{expression}}
5230 @cindex @code{equ} directive
5231 @cindex assigning values to symbols
5232 @cindex symbols, assigning values to
5233 This directive sets the value of @var{symbol} to @var{expression}.
5234 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5237 The syntax for @code{equ} on the HPPA is
5238 @samp{@var{symbol} .equ @var{expression}}.
5242 The syntax for @code{equ} on the Z80 is
5243 @samp{@var{symbol} equ @var{expression}}.
5244 On the Z80 it is an error if @var{symbol} is already defined,
5245 but the symbol is not protected from later redefinition.
5246 Compare @ref{Equiv}.
5250 @section @code{.equiv @var{symbol}, @var{expression}}
5251 @cindex @code{equiv} directive
5252 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5253 the assembler will signal an error if @var{symbol} is already defined. Note a
5254 symbol which has been referenced but not actually defined is considered to be
5257 Except for the contents of the error message, this is roughly equivalent to
5264 plus it protects the symbol from later redefinition.
5267 @section @code{.eqv @var{symbol}, @var{expression}}
5268 @cindex @code{eqv} directive
5269 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5270 evaluate the expression or any part of it immediately. Instead each time
5271 the resulting symbol is used in an expression, a snapshot of its current
5275 @section @code{.err}
5276 @cindex @code{err} directive
5277 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5278 message and, unless the @option{-Z} option was used, it will not generate an
5279 object file. This can be used to signal an error in conditionally compiled code.
5282 @section @code{.error "@var{string}"}
5283 @cindex error directive
5285 Similarly to @code{.err}, this directive emits an error, but you can specify a
5286 string that will be emitted as the error message. If you don't specify the
5287 message, it defaults to @code{".error directive invoked in source file"}.
5288 @xref{Errors, ,Error and Warning Messages}.
5291 .error "This code has not been assembled and tested."
5295 @section @code{.exitm}
5296 Exit early from the current macro definition. @xref{Macro}.
5299 @section @code{.extern}
5301 @cindex @code{extern} directive
5302 @code{.extern} is accepted in the source program---for compatibility
5303 with other assemblers---but it is ignored. @command{@value{AS}} treats
5304 all undefined symbols as external.
5307 @section @code{.fail @var{expression}}
5309 @cindex @code{fail} directive
5310 Generates an error or a warning. If the value of the @var{expression} is 500
5311 or more, @command{@value{AS}} will print a warning message. If the value is less
5312 than 500, @command{@value{AS}} will print an error message. The message will
5313 include the value of @var{expression}. This can occasionally be useful inside
5314 complex nested macros or conditional assembly.
5317 @section @code{.file}
5318 @cindex @code{file} directive
5320 @ifclear no-file-dir
5321 There are two different versions of the @code{.file} directive. Targets
5322 that support DWARF2 line number information use the DWARF2 version of
5323 @code{.file}. Other targets use the default version.
5325 @subheading Default Version
5327 @cindex logical file name
5328 @cindex file name, logical
5329 This version of the @code{.file} directive tells @command{@value{AS}} that we
5330 are about to start a new logical file. The syntax is:
5336 @var{string} is the new file name. In general, the filename is
5337 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5338 to specify an empty file name, you must give the quotes--@code{""}. This
5339 statement may go away in future: it is only recognized to be compatible with
5340 old @command{@value{AS}} programs.
5342 @subheading DWARF2 Version
5345 When emitting DWARF2 line number information, @code{.file} assigns filenames
5346 to the @code{.debug_line} file name table. The syntax is:
5349 .file @var{fileno} @var{filename}
5352 The @var{fileno} operand should be a unique positive integer to use as the
5353 index of the entry in the table. The @var{filename} operand is a C string
5356 The detail of filename indices is exposed to the user because the filename
5357 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5358 information, and thus the user must know the exact indices that table
5362 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5364 @cindex @code{fill} directive
5365 @cindex writing patterns in memory
5366 @cindex patterns, writing in memory
5367 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5368 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5369 may be zero or more. @var{Size} may be zero or more, but if it is
5370 more than 8, then it is deemed to have the value 8, compatible with
5371 other people's assemblers. The contents of each @var{repeat} bytes
5372 is taken from an 8-byte number. The highest order 4 bytes are
5373 zero. The lowest order 4 bytes are @var{value} rendered in the
5374 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5375 Each @var{size} bytes in a repetition is taken from the lowest order
5376 @var{size} bytes of this number. Again, this bizarre behavior is
5377 compatible with other people's assemblers.
5379 @var{size} and @var{value} are optional.
5380 If the second comma and @var{value} are absent, @var{value} is
5381 assumed zero. If the first comma and following tokens are absent,
5382 @var{size} is assumed to be 1.
5385 @section @code{.float @var{flonums}}
5387 @cindex floating point numbers (single)
5388 @cindex @code{float} directive
5389 This directive assembles zero or more flonums, separated by commas. It
5390 has the same effect as @code{.single}.
5392 The exact kind of floating point numbers emitted depends on how
5393 @command{@value{AS}} is configured.
5394 @xref{Machine Dependencies}.
5398 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5399 in @sc{ieee} format.
5404 @section @code{.func @var{name}[,@var{label}]}
5405 @cindex @code{func} directive
5406 @code{.func} emits debugging information to denote function @var{name}, and
5407 is ignored unless the file is assembled with debugging enabled.
5408 Only @samp{--gstabs[+]} is currently supported.
5409 @var{label} is the entry point of the function and if omitted @var{name}
5410 prepended with the @samp{leading char} is used.
5411 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5412 All functions are currently defined to have @code{void} return type.
5413 The function must be terminated with @code{.endfunc}.
5416 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5418 @cindex @code{global} directive
5419 @cindex symbol, making visible to linker
5420 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5421 @var{symbol} in your partial program, its value is made available to
5422 other partial programs that are linked with it. Otherwise,
5423 @var{symbol} takes its attributes from a symbol of the same name
5424 from another file linked into the same program.
5426 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5427 compatibility with other assemblers.
5430 On the HPPA, @code{.global} is not always enough to make it accessible to other
5431 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5432 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5437 @section @code{.gnu_attribute @var{tag},@var{value}}
5438 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5441 @section @code{.hidden @var{names}}
5443 @cindex @code{hidden} directive
5445 This is one of the ELF visibility directives. The other two are
5446 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5447 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5449 This directive overrides the named symbols default visibility (which is set by
5450 their binding: local, global or weak). The directive sets the visibility to
5451 @code{hidden} which means that the symbols are not visible to other components.
5452 Such symbols are always considered to be @code{protected} as well.
5456 @section @code{.hword @var{expressions}}
5458 @cindex @code{hword} directive
5459 @cindex integers, 16-bit
5460 @cindex numbers, 16-bit
5461 @cindex sixteen bit integers
5462 This expects zero or more @var{expressions}, and emits
5463 a 16 bit number for each.
5466 This directive is a synonym for @samp{.short}; depending on the target
5467 architecture, it may also be a synonym for @samp{.word}.
5471 This directive is a synonym for @samp{.short}.
5474 This directive is a synonym for both @samp{.short} and @samp{.word}.
5479 @section @code{.ident}
5481 @cindex @code{ident} directive
5483 This directive is used by some assemblers to place tags in object files. The
5484 behavior of this directive varies depending on the target. When using the
5485 a.out object file format, @command{@value{AS}} simply accepts the directive for
5486 source-file compatibility with existing assemblers, but does not emit anything
5487 for it. When using COFF, comments are emitted to the @code{.comment} or
5488 @code{.rdata} section, depending on the target. When using ELF, comments are
5489 emitted to the @code{.comment} section.
5492 @section @code{.if @var{absolute expression}}
5494 @cindex conditional assembly
5495 @cindex @code{if} directive
5496 @code{.if} marks the beginning of a section of code which is only
5497 considered part of the source program being assembled if the argument
5498 (which must be an @var{absolute expression}) is non-zero. The end of
5499 the conditional section of code must be marked by @code{.endif}
5500 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5501 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5502 If you have several conditions to check, @code{.elseif} may be used to avoid
5503 nesting blocks if/else within each subsequent @code{.else} block.
5505 The following variants of @code{.if} are also supported:
5507 @cindex @code{ifdef} directive
5508 @item .ifdef @var{symbol}
5509 Assembles the following section of code if the specified @var{symbol}
5510 has been defined. Note a symbol which has been referenced but not yet defined
5511 is considered to be undefined.
5513 @cindex @code{ifb} directive
5514 @item .ifb @var{text}
5515 Assembles the following section of code if the operand is blank (empty).
5517 @cindex @code{ifc} directive
5518 @item .ifc @var{string1},@var{string2}
5519 Assembles the following section of code if the two strings are the same. The
5520 strings may be optionally quoted with single quotes. If they are not quoted,
5521 the first string stops at the first comma, and the second string stops at the
5522 end of the line. Strings which contain whitespace should be quoted. The
5523 string comparison is case sensitive.
5525 @cindex @code{ifeq} directive
5526 @item .ifeq @var{absolute expression}
5527 Assembles the following section of code if the argument is zero.
5529 @cindex @code{ifeqs} directive
5530 @item .ifeqs @var{string1},@var{string2}
5531 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5533 @cindex @code{ifge} directive
5534 @item .ifge @var{absolute expression}
5535 Assembles the following section of code if the argument is greater than or
5538 @cindex @code{ifgt} directive
5539 @item .ifgt @var{absolute expression}
5540 Assembles the following section of code if the argument is greater than zero.
5542 @cindex @code{ifle} directive
5543 @item .ifle @var{absolute expression}
5544 Assembles the following section of code if the argument is less than or equal
5547 @cindex @code{iflt} directive
5548 @item .iflt @var{absolute expression}
5549 Assembles the following section of code if the argument is less than zero.
5551 @cindex @code{ifnb} directive
5552 @item .ifnb @var{text}
5553 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5554 following section of code if the operand is non-blank (non-empty).
5556 @cindex @code{ifnc} directive
5557 @item .ifnc @var{string1},@var{string2}.
5558 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5559 following section of code if the two strings are not the same.
5561 @cindex @code{ifndef} directive
5562 @cindex @code{ifnotdef} directive
5563 @item .ifndef @var{symbol}
5564 @itemx .ifnotdef @var{symbol}
5565 Assembles the following section of code if the specified @var{symbol}
5566 has not been defined. Both spelling variants are equivalent. Note a symbol
5567 which has been referenced but not yet defined is considered to be undefined.
5569 @cindex @code{ifne} directive
5570 @item .ifne @var{absolute expression}
5571 Assembles the following section of code if the argument is not equal to zero
5572 (in other words, this is equivalent to @code{.if}).
5574 @cindex @code{ifnes} directive
5575 @item .ifnes @var{string1},@var{string2}
5576 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5577 following section of code if the two strings are not the same.
5581 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5583 @cindex @code{incbin} directive
5584 @cindex binary files, including
5585 The @code{incbin} directive includes @var{file} verbatim at the current
5586 location. You can control the search paths used with the @samp{-I} command-line
5587 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5590 The @var{skip} argument skips a number of bytes from the start of the
5591 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5592 read. Note that the data is not aligned in any way, so it is the user's
5593 responsibility to make sure that proper alignment is provided both before and
5594 after the @code{incbin} directive.
5597 @section @code{.include "@var{file}"}
5599 @cindex @code{include} directive
5600 @cindex supporting files, including
5601 @cindex files, including
5602 This directive provides a way to include supporting files at specified
5603 points in your source program. The code from @var{file} is assembled as
5604 if it followed the point of the @code{.include}; when the end of the
5605 included file is reached, assembly of the original file continues. You
5606 can control the search paths used with the @samp{-I} command-line option
5607 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5611 @section @code{.int @var{expressions}}
5613 @cindex @code{int} directive
5614 @cindex integers, 32-bit
5615 Expect zero or more @var{expressions}, of any section, separated by commas.
5616 For each expression, emit a number that, at run time, is the value of that
5617 expression. The byte order and bit size of the number depends on what kind
5618 of target the assembly is for.
5622 On most forms of the H8/300, @code{.int} emits 16-bit
5623 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5630 @section @code{.internal @var{names}}
5632 @cindex @code{internal} directive
5634 This is one of the ELF visibility directives. The other two are
5635 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5636 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5638 This directive overrides the named symbols default visibility (which is set by
5639 their binding: local, global or weak). The directive sets the visibility to
5640 @code{internal} which means that the symbols are considered to be @code{hidden}
5641 (i.e., not visible to other components), and that some extra, processor specific
5642 processing must also be performed upon the symbols as well.
5646 @section @code{.irp @var{symbol},@var{values}}@dots{}
5648 @cindex @code{irp} directive
5649 Evaluate a sequence of statements assigning different values to @var{symbol}.
5650 The sequence of statements starts at the @code{.irp} directive, and is
5651 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5652 set to @var{value}, and the sequence of statements is assembled. If no
5653 @var{value} is listed, the sequence of statements is assembled once, with
5654 @var{symbol} set to the null string. To refer to @var{symbol} within the
5655 sequence of statements, use @var{\symbol}.
5657 For example, assembling
5665 is equivalent to assembling
5673 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5676 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5678 @cindex @code{irpc} directive
5679 Evaluate a sequence of statements assigning different values to @var{symbol}.
5680 The sequence of statements starts at the @code{.irpc} directive, and is
5681 terminated by an @code{.endr} directive. For each character in @var{value},
5682 @var{symbol} is set to the character, and the sequence of statements is
5683 assembled. If no @var{value} is listed, the sequence of statements is
5684 assembled once, with @var{symbol} set to the null string. To refer to
5685 @var{symbol} within the sequence of statements, use @var{\symbol}.
5687 For example, assembling
5695 is equivalent to assembling
5703 For some caveats with the spelling of @var{symbol}, see also the discussion
5707 @section @code{.lcomm @var{symbol} , @var{length}}
5709 @cindex @code{lcomm} directive
5710 @cindex local common symbols
5711 @cindex symbols, local common
5712 Reserve @var{length} (an absolute expression) bytes for a local common
5713 denoted by @var{symbol}. The section and value of @var{symbol} are
5714 those of the new local common. The addresses are allocated in the bss
5715 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5716 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5717 not visible to @code{@value{LD}}.
5720 Some targets permit a third argument to be used with @code{.lcomm}. This
5721 argument specifies the desired alignment of the symbol in the bss section.
5725 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5726 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5730 @section @code{.lflags}
5732 @cindex @code{lflags} directive (ignored)
5733 @command{@value{AS}} accepts this directive, for compatibility with other
5734 assemblers, but ignores it.
5736 @ifclear no-line-dir
5738 @section @code{.line @var{line-number}}
5740 @cindex @code{line} directive
5741 @cindex logical line number
5743 Change the logical line number. @var{line-number} must be an absolute
5744 expression. The next line has that logical line number. Therefore any other
5745 statements on the current line (after a statement separator character) are
5746 reported as on logical line number @var{line-number} @minus{} 1. One day
5747 @command{@value{AS}} will no longer support this directive: it is recognized only
5748 for compatibility with existing assembler programs.
5751 Even though this is a directive associated with the @code{a.out} or
5752 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5753 when producing COFF output, and treats @samp{.line} as though it
5754 were the COFF @samp{.ln} @emph{if} it is found outside a
5755 @code{.def}/@code{.endef} pair.
5757 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5758 used by compilers to generate auxiliary symbol information for
5763 @section @code{.linkonce [@var{type}]}
5765 @cindex @code{linkonce} directive
5766 @cindex common sections
5767 Mark the current section so that the linker only includes a single copy of it.
5768 This may be used to include the same section in several different object files,
5769 but ensure that the linker will only include it once in the final output file.
5770 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5771 Duplicate sections are detected based on the section name, so it should be
5774 This directive is only supported by a few object file formats; as of this
5775 writing, the only object file format which supports it is the Portable
5776 Executable format used on Windows NT.
5778 The @var{type} argument is optional. If specified, it must be one of the
5779 following strings. For example:
5783 Not all types may be supported on all object file formats.
5787 Silently discard duplicate sections. This is the default.
5790 Warn if there are duplicate sections, but still keep only one copy.
5793 Warn if any of the duplicates have different sizes.
5796 Warn if any of the duplicates do not have exactly the same contents.
5800 @section @code{.list}
5802 @cindex @code{list} directive
5803 @cindex listing control, turning on
5804 Control (in conjunction with the @code{.nolist} directive) whether or
5805 not assembly listings are generated. These two directives maintain an
5806 internal counter (which is zero initially). @code{.list} increments the
5807 counter, and @code{.nolist} decrements it. Assembly listings are
5808 generated whenever the counter is greater than zero.
5810 By default, listings are disabled. When you enable them (with the
5811 @samp{-a} command-line option; @pxref{Invoking,,Command-Line Options}),
5812 the initial value of the listing counter is one.
5815 @section @code{.ln @var{line-number}}
5817 @cindex @code{ln} directive
5818 @ifclear no-line-dir
5819 @samp{.ln} is a synonym for @samp{.line}.
5822 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5823 must be an absolute expression. The next line has that logical
5824 line number, so any other statements on the current line (after a
5825 statement separator character @code{;}) are reported as on logical
5826 line number @var{line-number} @minus{} 1.
5830 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5831 @cindex @code{loc} directive
5832 When emitting DWARF2 line number information,
5833 the @code{.loc} directive will add a row to the @code{.debug_line} line
5834 number matrix corresponding to the immediately following assembly
5835 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5836 arguments will be applied to the @code{.debug_line} state machine before
5839 The @var{options} are a sequence of the following tokens in any order:
5843 This option will set the @code{basic_block} register in the
5844 @code{.debug_line} state machine to @code{true}.
5847 This option will set the @code{prologue_end} register in the
5848 @code{.debug_line} state machine to @code{true}.
5850 @item epilogue_begin
5851 This option will set the @code{epilogue_begin} register in the
5852 @code{.debug_line} state machine to @code{true}.
5854 @item is_stmt @var{value}
5855 This option will set the @code{is_stmt} register in the
5856 @code{.debug_line} state machine to @code{value}, which must be
5859 @item isa @var{value}
5860 This directive will set the @code{isa} register in the @code{.debug_line}
5861 state machine to @var{value}, which must be an unsigned integer.
5863 @item discriminator @var{value}
5864 This directive will set the @code{discriminator} register in the @code{.debug_line}
5865 state machine to @var{value}, which must be an unsigned integer.
5867 @item view @var{value}
5868 This option causes a row to be added to @code{.debug_line} in reference to the
5869 current address (which might not be the same as that of the following assembly
5870 instruction), and to associate @var{value} with the @code{view} register in the
5871 @code{.debug_line} state machine. If @var{value} is a label, both the
5872 @code{view} register and the label are set to the number of prior @code{.loc}
5873 directives at the same program location. If @var{value} is the literal
5874 @code{0}, the @code{view} register is set to zero, and the assembler asserts
5875 that there aren't any prior @code{.loc} directives at the same program
5876 location. If @var{value} is the literal @code{-0}, the assembler arrange for
5877 the @code{view} register to be reset in this row, even if there are prior
5878 @code{.loc} directives at the same program location.
5882 @node Loc_mark_labels
5883 @section @code{.loc_mark_labels @var{enable}}
5884 @cindex @code{loc_mark_labels} directive
5885 When emitting DWARF2 line number information,
5886 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5887 to the @code{.debug_line} line number matrix with the @code{basic_block}
5888 register in the state machine set whenever a code label is seen.
5889 The @var{enable} argument should be either 1 or 0, to enable or disable
5890 this function respectively.
5894 @section @code{.local @var{names}}
5896 @cindex @code{local} directive
5897 This directive, which is available for ELF targets, marks each symbol in
5898 the comma-separated list of @code{names} as a local symbol so that it
5899 will not be externally visible. If the symbols do not already exist,
5900 they will be created.
5902 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5903 accept an alignment argument, which is the case for most ELF targets,
5904 the @code{.local} directive can be used in combination with @code{.comm}
5905 (@pxref{Comm}) to define aligned local common data.
5909 @section @code{.long @var{expressions}}
5911 @cindex @code{long} directive
5912 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5915 @c no one seems to know what this is for or whether this description is
5916 @c what it really ought to do
5918 @section @code{.lsym @var{symbol}, @var{expression}}
5920 @cindex @code{lsym} directive
5921 @cindex symbol, not referenced in assembly
5922 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5923 the hash table, ensuring it cannot be referenced by name during the
5924 rest of the assembly. This sets the attributes of the symbol to be
5925 the same as the expression value:
5927 @var{other} = @var{descriptor} = 0
5928 @var{type} = @r{(section of @var{expression})}
5929 @var{value} = @var{expression}
5932 The new symbol is not flagged as external.
5936 @section @code{.macro}
5939 The commands @code{.macro} and @code{.endm} allow you to define macros that
5940 generate assembly output. For example, this definition specifies a macro
5941 @code{sum} that puts a sequence of numbers into memory:
5944 .macro sum from=0, to=5
5953 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5965 @item .macro @var{macname}
5966 @itemx .macro @var{macname} @var{macargs} @dots{}
5967 @cindex @code{macro} directive
5968 Begin the definition of a macro called @var{macname}. If your macro
5969 definition requires arguments, specify their names after the macro name,
5970 separated by commas or spaces. You can qualify the macro argument to
5971 indicate whether all invocations must specify a non-blank value (through
5972 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5973 (through @samp{:@code{vararg}}). You can supply a default value for any
5974 macro argument by following the name with @samp{=@var{deflt}}. You
5975 cannot define two macros with the same @var{macname} unless it has been
5976 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5977 definitions. For example, these are all valid @code{.macro} statements:
5981 Begin the definition of a macro called @code{comm}, which takes no
5984 @item .macro plus1 p, p1
5985 @itemx .macro plus1 p p1
5986 Either statement begins the definition of a macro called @code{plus1},
5987 which takes two arguments; within the macro definition, write
5988 @samp{\p} or @samp{\p1} to evaluate the arguments.
5990 @item .macro reserve_str p1=0 p2
5991 Begin the definition of a macro called @code{reserve_str}, with two
5992 arguments. The first argument has a default value, but not the second.
5993 After the definition is complete, you can call the macro either as
5994 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5995 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5996 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5997 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5999 @item .macro m p1:req, p2=0, p3:vararg
6000 Begin the definition of a macro called @code{m}, with at least three
6001 arguments. The first argument must always have a value specified, but
6002 not the second, which instead has a default value. The third formal
6003 will get assigned all remaining arguments specified at invocation time.
6005 When you call a macro, you can specify the argument values either by
6006 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
6007 @samp{sum to=17, from=9}.
6011 Note that since each of the @var{macargs} can be an identifier exactly
6012 as any other one permitted by the target architecture, there may be
6013 occasional problems if the target hand-crafts special meanings to certain
6014 characters when they occur in a special position. For example, if the colon
6015 (@code{:}) is generally permitted to be part of a symbol name, but the
6016 architecture specific code special-cases it when occurring as the final
6017 character of a symbol (to denote a label), then the macro parameter
6018 replacement code will have no way of knowing that and consider the whole
6019 construct (including the colon) an identifier, and check only this
6020 identifier for being the subject to parameter substitution. So for example
6021 this macro definition:
6029 might not work as expected. Invoking @samp{label foo} might not create a label
6030 called @samp{foo} but instead just insert the text @samp{\l:} into the
6031 assembler source, probably generating an error about an unrecognised
6034 Similarly problems might occur with the period character (@samp{.})
6035 which is often allowed inside opcode names (and hence identifier names). So
6036 for example constructing a macro to build an opcode from a base name and a
6037 length specifier like this:
6040 .macro opcode base length
6045 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
6046 instruction but instead generate some kind of error as the assembler tries to
6047 interpret the text @samp{\base.\length}.
6049 There are several possible ways around this problem:
6052 @item Insert white space
6053 If it is possible to use white space characters then this is the simplest
6062 @item Use @samp{\()}
6063 The string @samp{\()} can be used to separate the end of a macro argument from
6064 the following text. eg:
6067 .macro opcode base length
6072 @item Use the alternate macro syntax mode
6073 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
6074 used as a separator. eg:
6084 Note: this problem of correctly identifying string parameters to pseudo ops
6085 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
6086 and @code{.irpc} (@pxref{Irpc}) as well.
6089 @cindex @code{endm} directive
6090 Mark the end of a macro definition.
6093 @cindex @code{exitm} directive
6094 Exit early from the current macro definition.
6096 @cindex number of macros executed
6097 @cindex macros, count executed
6099 @command{@value{AS}} maintains a counter of how many macros it has
6100 executed in this pseudo-variable; you can copy that number to your
6101 output with @samp{\@@}, but @emph{only within a macro definition}.
6103 @item LOCAL @var{name} [ , @dots{} ]
6104 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
6105 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
6106 @xref{Altmacro,,@code{.altmacro}}.
6110 @section @code{.mri @var{val}}
6112 @cindex @code{mri} directive
6113 @cindex MRI mode, temporarily
6114 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
6115 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
6116 affects code assembled until the next @code{.mri} directive, or until the end
6117 of the file. @xref{M, MRI mode, MRI mode}.
6120 @section @code{.noaltmacro}
6121 Disable alternate macro mode. @xref{Altmacro}.
6124 @section @code{.nolist}
6126 @cindex @code{nolist} directive
6127 @cindex listing control, turning off
6128 Control (in conjunction with the @code{.list} directive) whether or
6129 not assembly listings are generated. These two directives maintain an
6130 internal counter (which is zero initially). @code{.list} increments the
6131 counter, and @code{.nolist} decrements it. Assembly listings are
6132 generated whenever the counter is greater than zero.
6135 @section @code{.nops @var{size}[, @var{control}]}
6137 @cindex @code{nops} directive
6138 @cindex filling memory with no-op instructions
6139 This directive emits @var{size} bytes filled with no-op instructions.
6140 @var{size} is absolute expression, which must be a positve value.
6141 @var{control} controls how no-op instructions should be generated. If
6142 the comma and @var{control} are omitted, @var{control} is assumed to be
6145 Note: For Intel 80386 and AMD x86-64 targets, @var{control} specifies
6146 the size limit of a no-op instruction. The valid values of @var{control}
6147 are between 0 and 4 in 16-bit mode, between 0 and 7 when tuning for
6148 older processors in 32-bit mode, between 0 and 11 in 64-bit mode or when
6149 tuning for newer processors in 32-bit mode. When 0 is used, the no-op
6150 instruction size limit is set to the maximum supported size.
6153 @section @code{.octa @var{bignums}}
6155 @c FIXME: double size emitted for "octa" on some? Or warn?
6156 @cindex @code{octa} directive
6157 @cindex integer, 16-byte
6158 @cindex sixteen byte integer
6159 This directive expects zero or more bignums, separated by commas. For each
6160 bignum, it emits a 16-byte integer.
6162 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6163 hence @emph{octa}-word for 16 bytes.
6166 @section @code{.offset @var{loc}}
6168 @cindex @code{offset} directive
6169 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6170 be an absolute expression. This directive may be useful for defining
6171 symbols with absolute values. Do not confuse it with the @code{.org}
6175 @section @code{.org @var{new-lc} , @var{fill}}
6177 @cindex @code{org} directive
6178 @cindex location counter, advancing
6179 @cindex advancing location counter
6180 @cindex current address, advancing
6181 Advance the location counter of the current section to
6182 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6183 expression with the same section as the current subsection. That is,
6184 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6185 wrong section, the @code{.org} directive is ignored. To be compatible
6186 with former assemblers, if the section of @var{new-lc} is absolute,
6187 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6188 is the same as the current subsection.
6190 @code{.org} may only increase the location counter, or leave it
6191 unchanged; you cannot use @code{.org} to move the location counter
6194 @c double negative used below "not undefined" because this is a specific
6195 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6196 @c section. doc@cygnus.com 18feb91
6197 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6198 may not be undefined. If you really detest this restriction we eagerly await
6199 a chance to share your improved assembler.
6201 Beware that the origin is relative to the start of the section, not
6202 to the start of the subsection. This is compatible with other
6203 people's assemblers.
6205 When the location counter (of the current subsection) is advanced, the
6206 intervening bytes are filled with @var{fill} which should be an
6207 absolute expression. If the comma and @var{fill} are omitted,
6208 @var{fill} defaults to zero.
6211 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
6213 @cindex padding the location counter given a power of two
6214 @cindex @code{p2align} directive
6215 Pad the location counter (in the current subsection) to a particular
6216 storage boundary. The first expression (which must be absolute) is the
6217 number of low-order zero bits the location counter must have after
6218 advancement. For example @samp{.p2align 3} advances the location
6219 counter until it is a multiple of 8. If the location counter is already a
6220 multiple of 8, no change is needed.
6222 The second expression (also absolute) gives the fill value to be stored in the
6223 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6224 padding bytes are normally zero. However, on most systems, if the section is
6225 marked as containing code and the fill value is omitted, the space is filled
6226 with no-op instructions.
6228 The third expression is also absolute, and is also optional. If it is present,
6229 it is the maximum number of bytes that should be skipped by this alignment
6230 directive. If doing the alignment would require skipping more bytes than the
6231 specified maximum, then the alignment is not done at all. You can omit the
6232 fill value (the second argument) entirely by simply using two commas after the
6233 required alignment; this can be useful if you want the alignment to be filled
6234 with no-op instructions when appropriate.
6236 @cindex @code{p2alignw} directive
6237 @cindex @code{p2alignl} directive
6238 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6239 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6240 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6241 fill pattern as a four byte longword value. For example, @code{.p2alignw
6242 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6243 filled in with the value 0x368d (the exact placement of the bytes depends upon
6244 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6249 @section @code{.popsection}
6251 @cindex @code{popsection} directive
6252 @cindex Section Stack
6253 This is one of the ELF section stack manipulation directives. The others are
6254 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6255 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6258 This directive replaces the current section (and subsection) with the top
6259 section (and subsection) on the section stack. This section is popped off the
6265 @section @code{.previous}
6267 @cindex @code{previous} directive
6268 @cindex Section Stack
6269 This is one of the ELF section stack manipulation directives. The others are
6270 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6271 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6272 (@pxref{PopSection}).
6274 This directive swaps the current section (and subsection) with most recently
6275 referenced section/subsection pair prior to this one. Multiple
6276 @code{.previous} directives in a row will flip between two sections (and their
6277 subsections). For example:
6289 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6295 # Now in section A subsection 1
6299 # Now in section B subsection 0
6302 # Now in section B subsection 1
6305 # Now in section B subsection 0
6309 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6310 section B and 0x9abc into subsection 1 of section B.
6312 In terms of the section stack, this directive swaps the current section with
6313 the top section on the section stack.
6317 @section @code{.print @var{string}}
6319 @cindex @code{print} directive
6320 @command{@value{AS}} will print @var{string} on the standard output during
6321 assembly. You must put @var{string} in double quotes.
6325 @section @code{.protected @var{names}}
6327 @cindex @code{protected} directive
6329 This is one of the ELF visibility directives. The other two are
6330 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6332 This directive overrides the named symbols default visibility (which is set by
6333 their binding: local, global or weak). The directive sets the visibility to
6334 @code{protected} which means that any references to the symbols from within the
6335 components that defines them must be resolved to the definition in that
6336 component, even if a definition in another component would normally preempt
6341 @section @code{.psize @var{lines} , @var{columns}}
6343 @cindex @code{psize} directive
6344 @cindex listing control: paper size
6345 @cindex paper size, for listings
6346 Use this directive to declare the number of lines---and, optionally, the
6347 number of columns---to use for each page, when generating listings.
6349 If you do not use @code{.psize}, listings use a default line-count
6350 of 60. You may omit the comma and @var{columns} specification; the
6351 default width is 200 columns.
6353 @command{@value{AS}} generates formfeeds whenever the specified number of
6354 lines is exceeded (or whenever you explicitly request one, using
6357 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6358 those explicitly specified with @code{.eject}.
6361 @section @code{.purgem @var{name}}
6363 @cindex @code{purgem} directive
6364 Undefine the macro @var{name}, so that later uses of the string will not be
6365 expanded. @xref{Macro}.
6369 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6371 @cindex @code{pushsection} directive
6372 @cindex Section Stack
6373 This is one of the ELF section stack manipulation directives. The others are
6374 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6375 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6378 This directive pushes the current section (and subsection) onto the
6379 top of the section stack, and then replaces the current section and
6380 subsection with @code{name} and @code{subsection}. The optional
6381 @code{flags}, @code{type} and @code{arguments} are treated the same
6382 as in the @code{.section} (@pxref{Section}) directive.
6386 @section @code{.quad @var{bignums}}
6388 @cindex @code{quad} directive
6389 @code{.quad} expects zero or more bignums, separated by commas. For
6390 each bignum, it emits
6392 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6393 warning message; and just takes the lowest order 8 bytes of the bignum.
6394 @cindex eight-byte integer
6395 @cindex integer, 8-byte
6397 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6398 hence @emph{quad}-word for 8 bytes.
6401 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6402 warning message; and just takes the lowest order 16 bytes of the bignum.
6403 @cindex sixteen-byte integer
6404 @cindex integer, 16-byte
6408 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6410 @cindex @code{reloc} directive
6411 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6412 @var{expression}. If @var{offset} is a number, the relocation is generated in
6413 the current section. If @var{offset} is an expression that resolves to a
6414 symbol plus offset, the relocation is generated in the given symbol's section.
6415 @var{expression}, if present, must resolve to a symbol plus addend or to an
6416 absolute value, but note that not all targets support an addend. e.g. ELF REL
6417 targets such as i386 store an addend in the section contents rather than in the
6418 relocation. This low level interface does not support addends stored in the
6422 @section @code{.rept @var{count}}
6424 @cindex @code{rept} directive
6425 Repeat the sequence of lines between the @code{.rept} directive and the next
6426 @code{.endr} directive @var{count} times.
6428 For example, assembling
6436 is equivalent to assembling
6444 A count of zero is allowed, but nothing is generated. Negative counts are not
6445 allowed and if encountered will be treated as if they were zero.
6448 @section @code{.sbttl "@var{subheading}"}
6450 @cindex @code{sbttl} directive
6451 @cindex subtitles for listings
6452 @cindex listing control: subtitle
6453 Use @var{subheading} as the title (third line, immediately after the
6454 title line) when generating assembly listings.
6456 This directive affects subsequent pages, as well as the current page if
6457 it appears within ten lines of the top of a page.
6461 @section @code{.scl @var{class}}
6463 @cindex @code{scl} directive
6464 @cindex symbol storage class (COFF)
6465 @cindex COFF symbol storage class
6466 Set the storage-class value for a symbol. This directive may only be
6467 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6468 whether a symbol is static or external, or it may record further
6469 symbolic debugging information.
6474 @section @code{.section @var{name}}
6476 @cindex named section
6477 Use the @code{.section} directive to assemble the following code into a section
6480 This directive is only supported for targets that actually support arbitrarily
6481 named sections; on @code{a.out} targets, for example, it is not accepted, even
6482 with a standard @code{a.out} section name.
6486 @c only print the extra heading if both COFF and ELF are set
6487 @subheading COFF Version
6490 @cindex @code{section} directive (COFF version)
6491 For COFF targets, the @code{.section} directive is used in one of the following
6495 .section @var{name}[, "@var{flags}"]
6496 .section @var{name}[, @var{subsection}]
6499 If the optional argument is quoted, it is taken as flags to use for the
6500 section. Each flag is a single character. The following flags are recognized:
6504 bss section (uninitialized data)
6506 section is not loaded
6512 exclude section from linking
6518 shared section (meaningful for PE targets)
6520 ignored. (For compatibility with the ELF version)
6522 section is not readable (meaningful for PE targets)
6524 single-digit power-of-two section alignment (GNU extension)
6527 If no flags are specified, the default flags depend upon the section name. If
6528 the section name is not recognized, the default will be for the section to be
6529 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6530 from the section, rather than adding them, so if they are used on their own it
6531 will be as if no flags had been specified at all.
6533 If the optional argument to the @code{.section} directive is not quoted, it is
6534 taken as a subsection number (@pxref{Sub-Sections}).
6539 @c only print the extra heading if both COFF and ELF are set
6540 @subheading ELF Version
6543 @cindex Section Stack
6544 This is one of the ELF section stack manipulation directives. The others are
6545 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6546 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6547 @code{.previous} (@pxref{Previous}).
6549 @cindex @code{section} directive (ELF version)
6550 For ELF targets, the @code{.section} directive is used like this:
6553 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6556 @anchor{Section Name Substitutions}
6557 @kindex --sectname-subst
6558 @cindex section name substitution
6559 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6560 argument may contain a substitution sequence. Only @code{%S} is supported
6561 at the moment, and substitutes the current section name. For example:
6564 .macro exception_code
6565 .section %S.exception
6566 [exception code here]
6581 The two @code{exception_code} invocations above would create the
6582 @code{.text.exception} and @code{.init.exception} sections respectively.
6583 This is useful e.g. to discriminate between ancillary sections that are
6584 tied to setup code to be discarded after use from ancillary sections that
6585 need to stay resident without having to define multiple @code{exception_code}
6586 macros just for that purpose.
6588 The optional @var{flags} argument is a quoted string which may contain any
6589 combination of the following characters:
6593 section is allocatable
6595 section is a GNU_MBIND section
6597 section is excluded from executable and shared library.
6601 section is executable
6603 section is mergeable
6605 section contains zero terminated strings
6607 section is a member of a section group
6609 section is used for thread-local-storage
6611 section is a member of the previously-current section's group, if any
6612 @item @code{<number>}
6613 a numeric value indicating the bits to be set in the ELF section header's flags
6614 field. Note - if one or more of the alphabetic characters described above is
6615 also included in the flags field, their bit values will be ORed into the
6617 @item @code{<target specific>}
6618 some targets extend this list with their own flag characters
6621 Note - once a section's flags have been set they cannot be changed. There are
6622 a few exceptions to this rule however. Processor and application specific
6623 flags can be added to an already defined section. The @code{.interp},
6624 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6625 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6626 section may have the executable (@code{x}) flag added.
6628 The optional @var{type} argument may contain one of the following constants:
6632 section contains data
6634 section does not contain data (i.e., section only occupies space)
6636 section contains data which is used by things other than the program
6638 section contains an array of pointers to init functions
6640 section contains an array of pointers to finish functions
6641 @item @@preinit_array
6642 section contains an array of pointers to pre-init functions
6643 @item @@@code{<number>}
6644 a numeric value to be set as the ELF section header's type field.
6645 @item @@@code{<target specific>}
6646 some targets extend this list with their own types
6649 Many targets only support the first three section types. The type may be
6650 enclosed in double quotes if necessary.
6652 Note on targets where the @code{@@} character is the start of a comment (eg
6653 ARM) then another character is used instead. For example the ARM port uses the
6656 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6657 special and have fixed types. Any attempt to declare them with a different
6658 type will generate an error from the assembler.
6660 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6661 be specified as well as an extra argument---@var{entsize}---like this:
6664 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6667 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6668 constants, each @var{entsize} octets long. Sections with both @code{M} and
6669 @code{S} must contain zero terminated strings where each character is
6670 @var{entsize} bytes long. The linker may remove duplicates within sections with
6671 the same name, same entity size and same flags. @var{entsize} must be an
6672 absolute expression. For sections with both @code{M} and @code{S}, a string
6673 which is a suffix of a larger string is considered a duplicate. Thus
6674 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6675 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6677 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6678 be present along with an additional field like this:
6681 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6684 The @var{GroupName} field specifies the name of the section group to which this
6685 particular section belongs. The optional linkage field can contain:
6689 indicates that only one copy of this section should be retained
6694 Note: if both the @var{M} and @var{G} flags are present then the fields for
6695 the Merge flag should come first, like this:
6698 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6701 If @var{flags} contains the @code{?} symbol then it may not also contain the
6702 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6703 present. Instead, @code{?} says to consider the section that's current before
6704 this directive. If that section used @code{G}, then the new section will use
6705 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6706 If not, then the @code{?} symbol has no effect.
6708 If no flags are specified, the default flags depend upon the section name. If
6709 the section name is not recognized, the default will be for the section to have
6710 none of the above flags: it will not be allocated in memory, nor writable, nor
6711 executable. The section will contain data.
6713 For ELF targets, the assembler supports another type of @code{.section}
6714 directive for compatibility with the Solaris assembler:
6717 .section "@var{name}"[, @var{flags}...]
6720 Note that the section name is quoted. There may be a sequence of comma
6725 section is allocatable
6729 section is executable
6731 section is excluded from executable and shared library.
6733 section is used for thread local storage
6736 This directive replaces the current section and subsection. See the
6737 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6738 some examples of how this directive and the other section stack directives
6744 @section @code{.set @var{symbol}, @var{expression}}
6746 @cindex @code{set} directive
6747 @cindex symbol value, setting
6748 Set the value of @var{symbol} to @var{expression}. This
6749 changes @var{symbol}'s value and type to conform to
6750 @var{expression}. If @var{symbol} was flagged as external, it remains
6751 flagged (@pxref{Symbol Attributes}).
6753 You may @code{.set} a symbol many times in the same assembly provided that the
6754 values given to the symbol are constants. Values that are based on expressions
6755 involving other symbols are allowed, but some targets may restrict this to only
6756 being done once per assembly. This is because those targets do not set the
6757 addresses of symbols at assembly time, but rather delay the assignment until a
6758 final link is performed. This allows the linker a chance to change the code in
6759 the files, changing the location of, and the relative distance between, various
6762 If you @code{.set} a global symbol, the value stored in the object
6763 file is the last value stored into it.
6766 On Z80 @code{set} is a real instruction, use
6767 @samp{@var{symbol} defl @var{expression}} instead.
6771 @section @code{.short @var{expressions}}
6773 @cindex @code{short} directive
6775 @code{.short} is normally the same as @samp{.word}.
6776 @xref{Word,,@code{.word}}.
6778 In some configurations, however, @code{.short} and @code{.word} generate
6779 numbers of different lengths. @xref{Machine Dependencies}.
6783 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6786 This expects zero or more @var{expressions}, and emits
6787 a 16 bit number for each.
6792 @section @code{.single @var{flonums}}
6794 @cindex @code{single} directive
6795 @cindex floating point numbers (single)
6796 This directive assembles zero or more flonums, separated by commas. It
6797 has the same effect as @code{.float}.
6799 The exact kind of floating point numbers emitted depends on how
6800 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6804 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6805 numbers in @sc{ieee} format.
6811 @section @code{.size}
6813 This directive is used to set the size associated with a symbol.
6817 @c only print the extra heading if both COFF and ELF are set
6818 @subheading COFF Version
6821 @cindex @code{size} directive (COFF version)
6822 For COFF targets, the @code{.size} directive is only permitted inside
6823 @code{.def}/@code{.endef} pairs. It is used like this:
6826 .size @var{expression}
6833 @c only print the extra heading if both COFF and ELF are set
6834 @subheading ELF Version
6837 @cindex @code{size} directive (ELF version)
6838 For ELF targets, the @code{.size} directive is used like this:
6841 .size @var{name} , @var{expression}
6844 This directive sets the size associated with a symbol @var{name}.
6845 The size in bytes is computed from @var{expression} which can make use of label
6846 arithmetic. This directive is typically used to set the size of function
6851 @ifclear no-space-dir
6853 @section @code{.skip @var{size} [,@var{fill}]}
6855 @cindex @code{skip} directive
6856 @cindex filling memory
6857 This directive emits @var{size} bytes, each of value @var{fill}. Both
6858 @var{size} and @var{fill} are absolute expressions. If the comma and
6859 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6864 @section @code{.sleb128 @var{expressions}}
6866 @cindex @code{sleb128} directive
6867 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6868 compact, variable length representation of numbers used by the DWARF
6869 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6871 @ifclear no-space-dir
6873 @section @code{.space @var{size} [,@var{fill}]}
6875 @cindex @code{space} directive
6876 @cindex filling memory
6877 This directive emits @var{size} bytes, each of value @var{fill}. Both
6878 @var{size} and @var{fill} are absolute expressions. If the comma
6879 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6884 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6885 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6886 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6887 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6895 @section @code{.stabd, .stabn, .stabs}
6897 @cindex symbolic debuggers, information for
6898 @cindex @code{stab@var{x}} directives
6899 There are three directives that begin @samp{.stab}.
6900 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6901 The symbols are not entered in the @command{@value{AS}} hash table: they
6902 cannot be referenced elsewhere in the source file.
6903 Up to five fields are required:
6907 This is the symbol's name. It may contain any character except
6908 @samp{\000}, so is more general than ordinary symbol names. Some
6909 debuggers used to code arbitrarily complex structures into symbol names
6913 An absolute expression. The symbol's type is set to the low 8 bits of
6914 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6915 and debuggers choke on silly bit patterns.
6918 An absolute expression. The symbol's ``other'' attribute is set to the
6919 low 8 bits of this expression.
6922 An absolute expression. The symbol's descriptor is set to the low 16
6923 bits of this expression.
6926 An absolute expression which becomes the symbol's value.
6929 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6930 or @code{.stabs} statement, the symbol has probably already been created;
6931 you get a half-formed symbol in your object file. This is
6932 compatible with earlier assemblers!
6935 @cindex @code{stabd} directive
6936 @item .stabd @var{type} , @var{other} , @var{desc}
6938 The ``name'' of the symbol generated is not even an empty string.
6939 It is a null pointer, for compatibility. Older assemblers used a
6940 null pointer so they didn't waste space in object files with empty
6943 The symbol's value is set to the location counter,
6944 relocatably. When your program is linked, the value of this symbol
6945 is the address of the location counter when the @code{.stabd} was
6948 @cindex @code{stabn} directive
6949 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6950 The name of the symbol is set to the empty string @code{""}.
6952 @cindex @code{stabs} directive
6953 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6954 All five fields are specified.
6960 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6961 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6963 @cindex string, copying to object file
6964 @cindex string8, copying to object file
6965 @cindex string16, copying to object file
6966 @cindex string32, copying to object file
6967 @cindex string64, copying to object file
6968 @cindex @code{string} directive
6969 @cindex @code{string8} directive
6970 @cindex @code{string16} directive
6971 @cindex @code{string32} directive
6972 @cindex @code{string64} directive
6974 Copy the characters in @var{str} to the object file. You may specify more than
6975 one string to copy, separated by commas. Unless otherwise specified for a
6976 particular machine, the assembler marks the end of each string with a 0 byte.
6977 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6979 The variants @code{string16}, @code{string32} and @code{string64} differ from
6980 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6981 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6982 are stored in target endianness byte order.
6988 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6989 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6994 @section @code{.struct @var{expression}}
6996 @cindex @code{struct} directive
6997 Switch to the absolute section, and set the section offset to @var{expression},
6998 which must be an absolute expression. You might use this as follows:
7007 This would define the symbol @code{field1} to have the value 0, the symbol
7008 @code{field2} to have the value 4, and the symbol @code{field3} to have the
7009 value 8. Assembly would be left in the absolute section, and you would need to
7010 use a @code{.section} directive of some sort to change to some other section
7011 before further assembly.
7015 @section @code{.subsection @var{name}}
7017 @cindex @code{subsection} directive
7018 @cindex Section Stack
7019 This is one of the ELF section stack manipulation directives. The others are
7020 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
7021 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
7024 This directive replaces the current subsection with @code{name}. The current
7025 section is not changed. The replaced subsection is put onto the section stack
7026 in place of the then current top of stack subsection.
7031 @section @code{.symver}
7032 @cindex @code{symver} directive
7033 @cindex symbol versioning
7034 @cindex versions of symbols
7035 Use the @code{.symver} directive to bind symbols to specific version nodes
7036 within a source file. This is only supported on ELF platforms, and is
7037 typically used when assembling files to be linked into a shared library.
7038 There are cases where it may make sense to use this in objects to be bound
7039 into an application itself so as to override a versioned symbol from a
7042 For ELF targets, the @code{.symver} directive can be used like this:
7044 .symver @var{name}, @var{name2@@nodename}
7046 If the symbol @var{name} is defined within the file
7047 being assembled, the @code{.symver} directive effectively creates a symbol
7048 alias with the name @var{name2@@nodename}, and in fact the main reason that we
7049 just don't try and create a regular alias is that the @var{@@} character isn't
7050 permitted in symbol names. The @var{name2} part of the name is the actual name
7051 of the symbol by which it will be externally referenced. The name @var{name}
7052 itself is merely a name of convenience that is used so that it is possible to
7053 have definitions for multiple versions of a function within a single source
7054 file, and so that the compiler can unambiguously know which version of a
7055 function is being mentioned. The @var{nodename} portion of the alias should be
7056 the name of a node specified in the version script supplied to the linker when
7057 building a shared library. If you are attempting to override a versioned
7058 symbol from a shared library, then @var{nodename} should correspond to the
7059 nodename of the symbol you are trying to override.
7061 If the symbol @var{name} is not defined within the file being assembled, all
7062 references to @var{name} will be changed to @var{name2@@nodename}. If no
7063 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
7066 Another usage of the @code{.symver} directive is:
7068 .symver @var{name}, @var{name2@@@@nodename}
7070 In this case, the symbol @var{name} must exist and be defined within
7071 the file being assembled. It is similar to @var{name2@@nodename}. The
7072 difference is @var{name2@@@@nodename} will also be used to resolve
7073 references to @var{name2} by the linker.
7075 The third usage of the @code{.symver} directive is:
7077 .symver @var{name}, @var{name2@@@@@@nodename}
7079 When @var{name} is not defined within the
7080 file being assembled, it is treated as @var{name2@@nodename}. When
7081 @var{name} is defined within the file being assembled, the symbol
7082 name, @var{name}, will be changed to @var{name2@@@@nodename}.
7087 @section @code{.tag @var{structname}}
7089 @cindex COFF structure debugging
7090 @cindex structure debugging, COFF
7091 @cindex @code{tag} directive
7092 This directive is generated by compilers to include auxiliary debugging
7093 information in the symbol table. It is only permitted inside
7094 @code{.def}/@code{.endef} pairs. Tags are used to link structure
7095 definitions in the symbol table with instances of those structures.
7099 @section @code{.text @var{subsection}}
7101 @cindex @code{text} directive
7102 Tells @command{@value{AS}} to assemble the following statements onto the end of
7103 the text subsection numbered @var{subsection}, which is an absolute
7104 expression. If @var{subsection} is omitted, subsection number zero
7108 @section @code{.title "@var{heading}"}
7110 @cindex @code{title} directive
7111 @cindex listing control: title line
7112 Use @var{heading} as the title (second line, immediately after the
7113 source file name and pagenumber) when generating assembly listings.
7115 This directive affects subsequent pages, as well as the current page if
7116 it appears within ten lines of the top of a page.
7120 @section @code{.type}
7122 This directive is used to set the type of a symbol.
7126 @c only print the extra heading if both COFF and ELF are set
7127 @subheading COFF Version
7130 @cindex COFF symbol type
7131 @cindex symbol type, COFF
7132 @cindex @code{type} directive (COFF version)
7133 For COFF targets, this directive is permitted only within
7134 @code{.def}/@code{.endef} pairs. It is used like this:
7140 This records the integer @var{int} as the type attribute of a symbol table
7147 @c only print the extra heading if both COFF and ELF are set
7148 @subheading ELF Version
7151 @cindex ELF symbol type
7152 @cindex symbol type, ELF
7153 @cindex @code{type} directive (ELF version)
7154 For ELF targets, the @code{.type} directive is used like this:
7157 .type @var{name} , @var{type description}
7160 This sets the type of symbol @var{name} to be either a
7161 function symbol or an object symbol. There are five different syntaxes
7162 supported for the @var{type description} field, in order to provide
7163 compatibility with various other assemblers.
7165 Because some of the characters used in these syntaxes (such as @samp{@@} and
7166 @samp{#}) are comment characters for some architectures, some of the syntaxes
7167 below do not work on all architectures. The first variant will be accepted by
7168 the GNU assembler on all architectures so that variant should be used for
7169 maximum portability, if you do not need to assemble your code with other
7172 The syntaxes supported are:
7175 .type <name> STT_<TYPE_IN_UPPER_CASE>
7176 .type <name>,#<type>
7177 .type <name>,@@<type>
7178 .type <name>,%<type>
7179 .type <name>,"<type>"
7182 The types supported are:
7187 Mark the symbol as being a function name.
7190 @itemx gnu_indirect_function
7191 Mark the symbol as an indirect function when evaluated during reloc
7192 processing. (This is only supported on assemblers targeting GNU systems).
7196 Mark the symbol as being a data object.
7200 Mark the symbol as being a thread-local data object.
7204 Mark the symbol as being a common data object.
7208 Does not mark the symbol in any way. It is supported just for completeness.
7210 @item gnu_unique_object
7211 Marks the symbol as being a globally unique data object. The dynamic linker
7212 will make sure that in the entire process there is just one symbol with this
7213 name and type in use. (This is only supported on assemblers targeting GNU
7218 Note: Some targets support extra types in addition to those listed above.
7224 @section @code{.uleb128 @var{expressions}}
7226 @cindex @code{uleb128} directive
7227 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7228 compact, variable length representation of numbers used by the DWARF
7229 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7233 @section @code{.val @var{addr}}
7235 @cindex @code{val} directive
7236 @cindex COFF value attribute
7237 @cindex value attribute, COFF
7238 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7239 records the address @var{addr} as the value attribute of a symbol table
7245 @section @code{.version "@var{string}"}
7247 @cindex @code{version} directive
7248 This directive creates a @code{.note} section and places into it an ELF
7249 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7254 @section @code{.vtable_entry @var{table}, @var{offset}}
7256 @cindex @code{vtable_entry} directive
7257 This directive finds or creates a symbol @code{table} and creates a
7258 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7261 @section @code{.vtable_inherit @var{child}, @var{parent}}
7263 @cindex @code{vtable_inherit} directive
7264 This directive finds the symbol @code{child} and finds or creates the symbol
7265 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7266 parent whose addend is the value of the child symbol. As a special case the
7267 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7271 @section @code{.warning "@var{string}"}
7272 @cindex warning directive
7273 Similar to the directive @code{.error}
7274 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7277 @section @code{.weak @var{names}}
7279 @cindex @code{weak} directive
7280 This directive sets the weak attribute on the comma separated list of symbol
7281 @code{names}. If the symbols do not already exist, they will be created.
7283 On COFF targets other than PE, weak symbols are a GNU extension. This
7284 directive sets the weak attribute on the comma separated list of symbol
7285 @code{names}. If the symbols do not already exist, they will be created.
7287 On the PE target, weak symbols are supported natively as weak aliases.
7288 When a weak symbol is created that is not an alias, GAS creates an
7289 alternate symbol to hold the default value.
7292 @section @code{.weakref @var{alias}, @var{target}}
7294 @cindex @code{weakref} directive
7295 This directive creates an alias to the target symbol that enables the symbol to
7296 be referenced with weak-symbol semantics, but without actually making it weak.
7297 If direct references or definitions of the symbol are present, then the symbol
7298 will not be weak, but if all references to it are through weak references, the
7299 symbol will be marked as weak in the symbol table.
7301 The effect is equivalent to moving all references to the alias to a separate
7302 assembly source file, renaming the alias to the symbol in it, declaring the
7303 symbol as weak there, and running a reloadable link to merge the object files
7304 resulting from the assembly of the new source file and the old source file that
7305 had the references to the alias removed.
7307 The alias itself never makes to the symbol table, and is entirely handled
7308 within the assembler.
7311 @section @code{.word @var{expressions}}
7313 @cindex @code{word} directive
7314 This directive expects zero or more @var{expressions}, of any section,
7315 separated by commas.
7318 For each expression, @command{@value{AS}} emits a 32-bit number.
7321 For each expression, @command{@value{AS}} emits a 16-bit number.
7326 The size of the number emitted, and its byte order,
7327 depend on what target computer the assembly is for.
7330 @c on sparc the "special treatment to support compilers" doesn't
7331 @c happen---32-bit addressability, period; no long/short jumps.
7332 @ifset DIFF-TBL-KLUGE
7333 @cindex difference tables altered
7334 @cindex altered difference tables
7336 @emph{Warning: Special Treatment to support Compilers}
7340 Machines with a 32-bit address space, but that do less than 32-bit
7341 addressing, require the following special treatment. If the machine of
7342 interest to you does 32-bit addressing (or doesn't require it;
7343 @pxref{Machine Dependencies}), you can ignore this issue.
7346 In order to assemble compiler output into something that works,
7347 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7348 Directives of the form @samp{.word sym1-sym2} are often emitted by
7349 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7350 directive of the form @samp{.word sym1-sym2}, and the difference between
7351 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7352 creates a @dfn{secondary jump table}, immediately before the next label.
7353 This secondary jump table is preceded by a short-jump to the
7354 first byte after the secondary table. This short-jump prevents the flow
7355 of control from accidentally falling into the new table. Inside the
7356 table is a long-jump to @code{sym2}. The original @samp{.word}
7357 contains @code{sym1} minus the address of the long-jump to
7360 If there were several occurrences of @samp{.word sym1-sym2} before the
7361 secondary jump table, all of them are adjusted. If there was a
7362 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7363 long-jump to @code{sym4} is included in the secondary jump table,
7364 and the @code{.word} directives are adjusted to contain @code{sym3}
7365 minus the address of the long-jump to @code{sym4}; and so on, for as many
7366 entries in the original jump table as necessary.
7369 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7370 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7371 assembly language programmers.
7374 @c end DIFF-TBL-KLUGE
7376 @ifclear no-space-dir
7378 @section @code{.zero @var{size}}
7380 @cindex @code{zero} directive
7381 @cindex filling memory with zero bytes
7382 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7383 expression. This directive is actually an alias for the @samp{.skip} directive
7384 so it can take an optional second argument of the value to store in the bytes
7385 instead of zero. Using @samp{.zero} in this way would be confusing however.
7390 @section @code{.2byte @var{expression} [, @var{expression}]*}
7391 @cindex @code{2byte} directive
7392 @cindex two-byte integer
7393 @cindex integer, 2-byte
7395 This directive expects zero or more expressions, separated by commas. If there
7396 are no expressions then the directive does nothing. Otherwise each expression
7397 is evaluated in turn and placed in the next two bytes of the current output
7398 section, using the endian model of the target. If an expression will not fit
7399 in two bytes, a warning message is displayed and the least significant two
7400 bytes of the expression's value are used. If an expression cannot be evaluated
7401 at assembly time then relocations will be generated in order to compute the
7404 This directive does not apply any alignment before or after inserting the
7405 values. As a result of this, if relocations are generated, they may be
7406 different from those used for inserting values with a guaranteed alignment.
7408 This directive is only available for ELF targets,
7411 @section @code{.4byte @var{expression} [, @var{expression}]*}
7412 @cindex @code{4byte} directive
7413 @cindex four-byte integer
7414 @cindex integer, 4-byte
7416 Like the @option{.2byte} directive, except that it inserts unaligned, four byte
7417 long values into the output.
7420 @section @code{.8byte @var{expression} [, @var{expression}]*}
7421 @cindex @code{8byte} directive
7422 @cindex eight-byte integer
7423 @cindex integer, 8-byte
7425 Like the @option{.2byte} directive, except that it inserts unaligned, eight
7426 byte long bignum values into the output.
7431 @section Deprecated Directives
7433 @cindex deprecated directives
7434 @cindex obsolescent directives
7435 One day these directives won't work.
7436 They are included for compatibility with older assemblers.
7443 @node Object Attributes
7444 @chapter Object Attributes
7445 @cindex object attributes
7447 @command{@value{AS}} assembles source files written for a specific architecture
7448 into object files for that architecture. But not all object files are alike.
7449 Many architectures support incompatible variations. For instance, floating
7450 point arguments might be passed in floating point registers if the object file
7451 requires hardware floating point support---or floating point arguments might be
7452 passed in integer registers if the object file supports processors with no
7453 hardware floating point unit. Or, if two objects are built for different
7454 generations of the same architecture, the combination may require the
7455 newer generation at run-time.
7457 This information is useful during and after linking. At link time,
7458 @command{@value{LD}} can warn about incompatible object files. After link
7459 time, tools like @command{gdb} can use it to process the linked file
7462 Compatibility information is recorded as a series of object attributes. Each
7463 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7464 string, and indicates who sets the meaning of the tag. The tag is an integer,
7465 and indicates what property the attribute describes. The value may be a string
7466 or an integer, and indicates how the property affects this object. Missing
7467 attributes are the same as attributes with a zero value or empty string value.
7469 Object attributes were developed as part of the ABI for the ARM Architecture.
7470 The file format is documented in @cite{ELF for the ARM Architecture}.
7473 * GNU Object Attributes:: @sc{gnu} Object Attributes
7474 * Defining New Object Attributes:: Defining New Object Attributes
7477 @node GNU Object Attributes
7478 @section @sc{gnu} Object Attributes
7480 The @code{.gnu_attribute} directive records an object attribute
7481 with vendor @samp{gnu}.
7483 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7484 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7485 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7486 2} is set for architecture-independent attributes and clear for
7487 architecture-dependent ones.
7489 @subsection Common @sc{gnu} attributes
7491 These attributes are valid on all architectures.
7494 @item Tag_compatibility (32)
7495 The compatibility attribute takes an integer flag value and a vendor name. If
7496 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7497 then the file is only compatible with the named toolchain. If it is greater
7498 than 1, the file can only be processed by other toolchains under some private
7499 arrangement indicated by the flag value and the vendor name.
7502 @subsection MIPS Attributes
7505 @item Tag_GNU_MIPS_ABI_FP (4)
7506 The floating-point ABI used by this object file. The value will be:
7510 0 for files not affected by the floating-point ABI.
7512 1 for files using the hardware floating-point ABI with a standard
7513 double-precision FPU.
7515 2 for files using the hardware floating-point ABI with a single-precision FPU.
7517 3 for files using the software floating-point ABI.
7519 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7520 floating-point registers, 32-bit general-purpose registers and increased the
7521 number of callee-saved floating-point registers.
7523 5 for files using the hardware floating-point ABI with a double-precision FPU
7524 with either 32-bit or 64-bit floating-point registers and 32-bit
7525 general-purpose registers.
7527 6 for files using the hardware floating-point ABI with 64-bit floating-point
7528 registers and 32-bit general-purpose registers.
7530 7 for files using the hardware floating-point ABI with 64-bit floating-point
7531 registers, 32-bit general-purpose registers and a rule that forbids the
7532 direct use of odd-numbered single-precision floating-point registers.
7536 @subsection PowerPC Attributes
7539 @item Tag_GNU_Power_ABI_FP (4)
7540 The floating-point ABI used by this object file. The value will be:
7544 0 for files not affected by the floating-point ABI.
7546 1 for files using double-precision hardware floating-point ABI.
7548 2 for files using the software floating-point ABI.
7550 3 for files using single-precision hardware floating-point ABI.
7553 @item Tag_GNU_Power_ABI_Vector (8)
7554 The vector ABI used by this object file. The value will be:
7558 0 for files not affected by the vector ABI.
7560 1 for files using general purpose registers to pass vectors.
7562 2 for files using AltiVec registers to pass vectors.
7564 3 for files using SPE registers to pass vectors.
7568 @subsection IBM z Systems Attributes
7571 @item Tag_GNU_S390_ABI_Vector (8)
7572 The vector ABI used by this object file. The value will be:
7576 0 for files not affected by the vector ABI.
7578 1 for files using software vector ABI.
7580 2 for files using hardware vector ABI.
7584 @node Defining New Object Attributes
7585 @section Defining New Object Attributes
7587 If you want to define a new @sc{gnu} object attribute, here are the places you
7588 will need to modify. New attributes should be discussed on the @samp{binutils}
7593 This manual, which is the official register of attributes.
7595 The header for your architecture @file{include/elf}, to define the tag.
7597 The @file{bfd} support file for your architecture, to merge the attribute
7598 and issue any appropriate link warnings.
7600 Test cases in @file{ld/testsuite} for merging and link warnings.
7602 @file{binutils/readelf.c} to display your attribute.
7604 GCC, if you want the compiler to mark the attribute automatically.
7610 @node Machine Dependencies
7611 @chapter Machine Dependent Features
7613 @cindex machine dependencies
7614 The machine instruction sets are (almost by definition) different on
7615 each machine where @command{@value{AS}} runs. Floating point representations
7616 vary as well, and @command{@value{AS}} often supports a few additional
7617 directives or command-line options for compatibility with other
7618 assemblers on a particular platform. Finally, some versions of
7619 @command{@value{AS}} support special pseudo-instructions for branch
7622 This chapter discusses most of these differences, though it does not
7623 include details on any machine's instruction set. For details on that
7624 subject, see the hardware manufacturer's manual.
7628 * AArch64-Dependent:: AArch64 Dependent Features
7631 * Alpha-Dependent:: Alpha Dependent Features
7634 * ARC-Dependent:: ARC Dependent Features
7637 * ARM-Dependent:: ARM Dependent Features
7640 * AVR-Dependent:: AVR Dependent Features
7643 * Blackfin-Dependent:: Blackfin Dependent Features
7646 * BPF-Dependent:: BPF Dependent Features
7649 * CR16-Dependent:: CR16 Dependent Features
7652 * CRIS-Dependent:: CRIS Dependent Features
7655 * C-SKY-Dependent:: C-SKY Dependent Features
7658 * D10V-Dependent:: D10V Dependent Features
7661 * D30V-Dependent:: D30V Dependent Features
7664 * Epiphany-Dependent:: EPIPHANY Dependent Features
7667 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7670 * HPPA-Dependent:: HPPA Dependent Features
7673 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7676 * IA-64-Dependent:: Intel IA-64 Dependent Features
7679 * IP2K-Dependent:: IP2K Dependent Features
7682 * LM32-Dependent:: LM32 Dependent Features
7685 * M32C-Dependent:: M32C Dependent Features
7688 * M32R-Dependent:: M32R Dependent Features
7691 * M68K-Dependent:: M680x0 Dependent Features
7694 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7697 * S12Z-Dependent:: S12Z Dependent Features
7700 * Meta-Dependent :: Meta Dependent Features
7703 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7706 * MIPS-Dependent:: MIPS Dependent Features
7709 * MMIX-Dependent:: MMIX Dependent Features
7712 * MSP430-Dependent:: MSP430 Dependent Features
7715 * NDS32-Dependent:: Andes NDS32 Dependent Features
7718 * NiosII-Dependent:: Altera Nios II Dependent Features
7721 * NS32K-Dependent:: NS32K Dependent Features
7724 * OpenRISC-Dependent:: OpenRISC 1000 Features
7727 * PDP-11-Dependent:: PDP-11 Dependent Features
7730 * PJ-Dependent:: picoJava Dependent Features
7733 * PPC-Dependent:: PowerPC Dependent Features
7736 * PRU-Dependent:: PRU Dependent Features
7739 * RISC-V-Dependent:: RISC-V Dependent Features
7742 * RL78-Dependent:: RL78 Dependent Features
7745 * RX-Dependent:: RX Dependent Features
7748 * S/390-Dependent:: IBM S/390 Dependent Features
7751 * SCORE-Dependent:: SCORE Dependent Features
7754 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7757 * Sparc-Dependent:: SPARC Dependent Features
7760 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7763 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7766 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7769 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7772 * V850-Dependent:: V850 Dependent Features
7775 * Vax-Dependent:: VAX Dependent Features
7778 * Visium-Dependent:: Visium Dependent Features
7781 * WebAssembly-Dependent:: WebAssembly Dependent Features
7784 * XGATE-Dependent:: XGATE Dependent Features
7787 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7790 * Xtensa-Dependent:: Xtensa Dependent Features
7793 * Z80-Dependent:: Z80 Dependent Features
7796 * Z8000-Dependent:: Z8000 Dependent Features
7803 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7804 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7805 @c peculiarity: to preserve cross-references, there must be a node called
7806 @c "Machine Dependencies". Hence the conditional nodenames in each
7807 @c major node below. Node defaulting in makeinfo requires adjacency of
7808 @c node and sectioning commands; hence the repetition of @chapter BLAH
7809 @c in both conditional blocks.
7812 @include c-aarch64.texi
7816 @include c-alpha.texi
7832 @include c-bfin.texi
7840 @include c-cr16.texi
7844 @include c-cris.texi
7848 @include c-csky.texi
7853 @node Machine Dependencies
7854 @chapter Machine Dependent Features
7856 The machine instruction sets are different on each Renesas chip family,
7857 and there are also some syntax differences among the families. This
7858 chapter describes the specific @command{@value{AS}} features for each
7862 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7863 * SH-Dependent:: Renesas SH Dependent Features
7870 @include c-d10v.texi
7874 @include c-d30v.texi
7878 @include c-epiphany.texi
7882 @include c-h8300.texi
7886 @include c-hppa.texi
7890 @include c-i386.texi
7894 @include c-ia64.texi
7898 @include c-ip2k.texi
7902 @include c-lm32.texi
7906 @include c-m32c.texi
7910 @include c-m32r.texi
7914 @include c-m68k.texi
7918 @include c-m68hc11.texi
7922 @include c-s12z.texi
7926 @include c-metag.texi
7930 @include c-microblaze.texi
7934 @include c-mips.texi
7938 @include c-mmix.texi
7942 @include c-msp430.texi
7946 @include c-nds32.texi
7950 @include c-nios2.texi
7954 @include c-ns32k.texi
7958 @include c-or1k.texi
7962 @include c-pdp11.texi
7978 @include c-riscv.texi
7982 @include c-rl78.texi
7990 @include c-s390.texi
7994 @include c-score.texi
8002 @include c-sparc.texi
8006 @include c-tic54x.texi
8010 @include c-tic6x.texi
8014 @include c-tilegx.texi
8018 @include c-tilepro.texi
8022 @include c-v850.texi
8030 @include c-visium.texi
8034 @include c-wasm32.texi
8038 @include c-xgate.texi
8042 @include c-xstormy16.texi
8046 @include c-xtensa.texi
8058 @c reverse effect of @down at top of generic Machine-Dep chapter
8062 @node Reporting Bugs
8063 @chapter Reporting Bugs
8064 @cindex bugs in assembler
8065 @cindex reporting bugs in assembler
8067 Your bug reports play an essential role in making @command{@value{AS}} reliable.
8069 Reporting a bug may help you by bringing a solution to your problem, or it may
8070 not. But in any case the principal function of a bug report is to help the
8071 entire community by making the next version of @command{@value{AS}} work better.
8072 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
8074 In order for a bug report to serve its purpose, you must include the
8075 information that enables us to fix the bug.
8078 * Bug Criteria:: Have you found a bug?
8079 * Bug Reporting:: How to report bugs
8083 @section Have You Found a Bug?
8084 @cindex bug criteria
8086 If you are not sure whether you have found a bug, here are some guidelines:
8089 @cindex fatal signal
8090 @cindex assembler crash
8091 @cindex crash of assembler
8093 If the assembler gets a fatal signal, for any input whatever, that is a
8094 @command{@value{AS}} bug. Reliable assemblers never crash.
8096 @cindex error on valid input
8098 If @command{@value{AS}} produces an error message for valid input, that is a bug.
8100 @cindex invalid input
8102 If @command{@value{AS}} does not produce an error message for invalid input, that
8103 is a bug. However, you should note that your idea of ``invalid input'' might
8104 be our idea of ``an extension'' or ``support for traditional practice''.
8107 If you are an experienced user of assemblers, your suggestions for improvement
8108 of @command{@value{AS}} are welcome in any case.
8112 @section How to Report Bugs
8114 @cindex assembler bugs, reporting
8116 A number of companies and individuals offer support for @sc{gnu} products. If
8117 you obtained @command{@value{AS}} from a support organization, we recommend you
8118 contact that organization first.
8120 You can find contact information for many support companies and
8121 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
8125 In any event, we also recommend that you send bug reports for @command{@value{AS}}
8129 The fundamental principle of reporting bugs usefully is this:
8130 @strong{report all the facts}. If you are not sure whether to state a
8131 fact or leave it out, state it!
8133 Often people omit facts because they think they know what causes the problem
8134 and assume that some details do not matter. Thus, you might assume that the
8135 name of a symbol you use in an example does not matter. Well, probably it does
8136 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
8137 happens to fetch from the location where that name is stored in memory;
8138 perhaps, if the name were different, the contents of that location would fool
8139 the assembler into doing the right thing despite the bug. Play it safe and
8140 give a specific, complete example. That is the easiest thing for you to do,
8141 and the most helpful.
8143 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
8144 it is new to us. Therefore, always write your bug reports on the assumption
8145 that the bug has not been reported previously.
8147 Sometimes people give a few sketchy facts and ask, ``Does this ring a
8148 bell?'' This cannot help us fix a bug, so it is basically useless. We
8149 respond by asking for enough details to enable us to investigate.
8150 You might as well expedite matters by sending them to begin with.
8152 To enable us to fix the bug, you should include all these things:
8156 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8157 it with the @samp{--version} argument.
8159 Without this, we will not know whether there is any point in looking for
8160 the bug in the current version of @command{@value{AS}}.
8163 Any patches you may have applied to the @command{@value{AS}} source.
8166 The type of machine you are using, and the operating system name and
8170 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8174 The command arguments you gave the assembler to assemble your example and
8175 observe the bug. To guarantee you will not omit something important, list them
8176 all. A copy of the Makefile (or the output from make) is sufficient.
8178 If we were to try to guess the arguments, we would probably guess wrong
8179 and then we might not encounter the bug.
8182 A complete input file that will reproduce the bug. If the bug is observed when
8183 the assembler is invoked via a compiler, send the assembler source, not the
8184 high level language source. Most compilers will produce the assembler source
8185 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8186 the options @samp{-v --save-temps}; this will save the assembler source in a
8187 file with an extension of @file{.s}, and also show you exactly how
8188 @command{@value{AS}} is being run.
8191 A description of what behavior you observe that you believe is
8192 incorrect. For example, ``It gets a fatal signal.''
8194 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8195 will certainly notice it. But if the bug is incorrect output, we might not
8196 notice unless it is glaringly wrong. You might as well not give us a chance to
8199 Even if the problem you experience is a fatal signal, you should still say so
8200 explicitly. Suppose something strange is going on, such as, your copy of
8201 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8202 library on your system. (This has happened!) Your copy might crash and ours
8203 would not. If you told us to expect a crash, then when ours fails to crash, we
8204 would know that the bug was not happening for us. If you had not told us to
8205 expect a crash, then we would not be able to draw any conclusion from our
8209 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8210 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8211 option. Always send diffs from the old file to the new file. If you even
8212 discuss something in the @command{@value{AS}} source, refer to it by context, not
8215 The line numbers in our development sources will not match those in your
8216 sources. Your line numbers would convey no useful information to us.
8219 Here are some things that are not necessary:
8223 A description of the envelope of the bug.
8225 Often people who encounter a bug spend a lot of time investigating
8226 which changes to the input file will make the bug go away and which
8227 changes will not affect it.
8229 This is often time consuming and not very useful, because the way we
8230 will find the bug is by running a single example under the debugger
8231 with breakpoints, not by pure deduction from a series of examples.
8232 We recommend that you save your time for something else.
8234 Of course, if you can find a simpler example to report @emph{instead}
8235 of the original one, that is a convenience for us. Errors in the
8236 output will be easier to spot, running under the debugger will take
8237 less time, and so on.
8239 However, simplification is not vital; if you do not want to do this,
8240 report the bug anyway and send us the entire test case you used.
8243 A patch for the bug.
8245 A patch for the bug does help us if it is a good one. But do not omit
8246 the necessary information, such as the test case, on the assumption that
8247 a patch is all we need. We might see problems with your patch and decide
8248 to fix the problem another way, or we might not understand it at all.
8250 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8251 construct an example that will make the program follow a certain path through
8252 the code. If you do not send us the example, we will not be able to construct
8253 one, so we will not be able to verify that the bug is fixed.
8255 And if we cannot understand what bug you are trying to fix, or why your
8256 patch should be an improvement, we will not install it. A test case will
8257 help us to understand.
8260 A guess about what the bug is or what it depends on.
8262 Such guesses are usually wrong. Even we cannot guess right about such
8263 things without first using the debugger to find the facts.
8266 @node Acknowledgements
8267 @chapter Acknowledgements
8269 If you have contributed to GAS and your name isn't listed here,
8270 it is not meant as a slight. We just don't know about it. Send mail to the
8271 maintainer, and we'll correct the situation. Currently
8273 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8275 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8278 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8279 information and the 68k series machines, most of the preprocessing pass, and
8280 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8282 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8283 many bug fixes, including merging support for several processors, breaking GAS
8284 up to handle multiple object file format back ends (including heavy rewrite,
8285 testing, an integration of the coff and b.out back ends), adding configuration
8286 including heavy testing and verification of cross assemblers and file splits
8287 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8288 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8289 port (including considerable amounts of reverse engineering), a SPARC opcode
8290 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8291 assertions and made them work, much other reorganization, cleanup, and lint.
8293 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8294 in format-specific I/O modules.
8296 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8297 has done much work with it since.
8299 The Intel 80386 machine description was written by Eliot Dresselhaus.
8301 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8303 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8304 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8306 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8307 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8308 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8309 support a.out format.
8311 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8312 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8313 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8314 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8317 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8318 simplified the configuration of which versions accept which directives. He
8319 updated the 68k machine description so that Motorola's opcodes always produced
8320 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8321 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8322 cross-compilation support, and one bug in relaxation that took a week and
8323 required the proverbial one-bit fix.
8325 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8326 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8327 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8328 PowerPC assembler, and made a few other minor patches.
8330 Steve Chamberlain made GAS able to generate listings.
8332 Hewlett-Packard contributed support for the HP9000/300.
8334 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8335 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8336 formats). This work was supported by both the Center for Software Science at
8337 the University of Utah and Cygnus Support.
8339 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8340 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8341 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8342 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8343 and some initial 64-bit support).
8345 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8347 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8348 support for openVMS/Alpha.
8350 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8353 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8354 Inc.@: added support for Xtensa processors.
8356 Several engineers at Cygnus Support have also provided many small bug fixes and
8357 configuration enhancements.
8359 Jon Beniston added support for the Lattice Mico32 architecture.
8361 Many others have contributed large or small bugfixes and enhancements. If
8362 you have contributed significant work and are not mentioned on this list, and
8363 want to be, let us know. Some of the history has been lost; we are not
8364 intentionally leaving anyone out.
8366 @node GNU Free Documentation License
8367 @appendix GNU Free Documentation License
8371 @unnumbered AS Index