1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (C) 1991-2016 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{} 1991-2016 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{} 1991-2016 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{--target-help}] [@var{target-options}]
248 [@b{--}|@var{files} @dots{}]
251 @c Target dependent options are listed below. Keep the list sorted.
252 @c Add an empty line for separation.
256 @emph{Target AArch64 options:}
258 [@b{-mabi}=@var{ABI}]
262 @emph{Target Alpha options:}
264 [@b{-mdebug} | @b{-no-mdebug}]
265 [@b{-replace} | @b{-noreplace}]
266 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
267 [@b{-F}] [@b{-32addr}]
271 @emph{Target ARC options:}
272 [@b{-mcpu=@var{cpu}}]
273 [@b{-mA6}|@b{-mARC600}|@b{-mARC601}|@b{-mA7}|@b{-mARC700}|@b{-mEM}|@b{-mHS}]
280 @emph{Target ARM options:}
281 @c Don't document the deprecated options
282 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
283 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
284 [@b{-mfpu}=@var{floating-point-format}]
285 [@b{-mfloat-abi}=@var{abi}]
286 [@b{-meabi}=@var{ver}]
289 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
290 @b{-mapcs-reentrant}]
291 [@b{-mthumb-interwork}] [@b{-k}]
295 @emph{Target Blackfin options:}
296 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
303 @emph{Target CRIS options:}
304 [@b{--underscore} | @b{--no-underscore}]
306 [@b{--emulation=criself} | @b{--emulation=crisaout}]
307 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
308 @c Deprecated -- deliberately not documented.
313 @emph{Target D10V options:}
318 @emph{Target D30V options:}
319 [@b{-O}|@b{-n}|@b{-N}]
323 @emph{Target EPIPHANY options:}
324 [@b{-mepiphany}|@b{-mepiphany16}]
328 @emph{Target H8/300 options:}
332 @c HPPA has no machine-dependent assembler options (yet).
336 @emph{Target i386 options:}
337 [@b{--32}|@b{--x32}|@b{--64}] [@b{-n}]
338 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
342 @emph{Target i960 options:}
343 @c see md_parse_option in tc-i960.c
344 [@b{-ACA}|@b{-ACA_A}|@b{-ACB}|@b{-ACC}|@b{-AKA}|@b{-AKB}|
346 [@b{-b}] [@b{-no-relax}]
350 @emph{Target IA-64 options:}
351 [@b{-mconstant-gp}|@b{-mauto-pic}]
352 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
354 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
355 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
356 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
357 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
361 @emph{Target IP2K options:}
362 [@b{-mip2022}|@b{-mip2022ext}]
366 @emph{Target M32C options:}
367 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
371 @emph{Target M32R options:}
372 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
377 @emph{Target M680X0 options:}
378 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
382 @emph{Target M68HC11 options:}
383 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}|@b{-mm9s12x}|@b{-mm9s12xg}]
384 [@b{-mshort}|@b{-mlong}]
385 [@b{-mshort-double}|@b{-mlong-double}]
386 [@b{--force-long-branches}] [@b{--short-branches}]
387 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
388 [@b{--print-opcodes}] [@b{--generate-example}]
392 @emph{Target MCORE options:}
393 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
394 [@b{-mcpu=[210|340]}]
398 @emph{Target Meta options:}
399 [@b{-mcpu=@var{cpu}}] [@b{-mfpu=@var{cpu}}] [@b{-mdsp=@var{cpu}}]
402 @emph{Target MICROBLAZE options:}
403 @c MicroBlaze has no machine-dependent assembler options.
407 @emph{Target MIPS options:}
408 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
409 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
410 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
411 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
412 [@b{-mfp64}] [@b{-mgp64}] [@b{-mfpxx}]
413 [@b{-modd-spreg}] [@b{-mno-odd-spreg}]
414 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
415 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
416 [@b{-mips32r3}] [@b{-mips32r5}] [@b{-mips32r6}] [@b{-mips64}] [@b{-mips64r2}]
417 [@b{-mips64r3}] [@b{-mips64r5}] [@b{-mips64r6}]
418 [@b{-construct-floats}] [@b{-no-construct-floats}]
419 [@b{-mnan=@var{encoding}}]
420 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
421 [@b{-mips16}] [@b{-no-mips16}]
422 [@b{-mmicromips}] [@b{-mno-micromips}]
423 [@b{-msmartmips}] [@b{-mno-smartmips}]
424 [@b{-mips3d}] [@b{-no-mips3d}]
425 [@b{-mdmx}] [@b{-no-mdmx}]
426 [@b{-mdsp}] [@b{-mno-dsp}]
427 [@b{-mdspr2}] [@b{-mno-dspr2}]
428 [@b{-mdspr3}] [@b{-mno-dspr3}]
429 [@b{-mmsa}] [@b{-mno-msa}]
430 [@b{-mxpa}] [@b{-mno-xpa}]
431 [@b{-mmt}] [@b{-mno-mt}]
432 [@b{-mmcu}] [@b{-mno-mcu}]
433 [@b{-minsn32}] [@b{-mno-insn32}]
434 [@b{-mfix7000}] [@b{-mno-fix7000}]
435 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
436 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
437 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
438 [@b{-mdebug}] [@b{-no-mdebug}]
439 [@b{-mpdr}] [@b{-mno-pdr}]
443 @emph{Target MMIX options:}
444 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
445 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
446 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
447 [@b{--linker-allocated-gregs}]
451 @emph{Target Nios II options:}
452 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
457 @emph{Target NDS32 options:}
458 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
459 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
460 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
461 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
462 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
463 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
464 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
469 @emph{Target PDP11 options:}
470 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
471 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
472 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
476 @emph{Target picoJava options:}
481 @emph{Target PowerPC options:}
483 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
484 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mppc64}|
485 @b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|@b{-me6500}|@b{-mppc64bridge}|
486 @b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|@b{-mpower6}|@b{-mpwr6}|
487 @b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-mpower9}|@b{-mpwr9}@b{-ma2}|
488 @b{-mcell}|@b{-mspe}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
489 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
490 [@b{-mregnames}|@b{-mno-regnames}]
491 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
492 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
493 [@b{-msolaris}|@b{-mno-solaris}]
494 [@b{-nops=@var{count}}]
498 @emph{Target RL78 options:}
500 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
504 @emph{Target RX options:}
505 [@b{-mlittle-endian}|@b{-mbig-endian}]
506 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
507 [@b{-muse-conventional-section-names}]
508 [@b{-msmall-data-limit}]
511 [@b{-mint-register=@var{number}}]
512 [@b{-mgcc-abi}|@b{-mrx-abi}]
516 @emph{Target s390 options:}
517 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
518 [@b{-mregnames}|@b{-mno-regnames}]
519 [@b{-mwarn-areg-zero}]
523 @emph{Target SCORE options:}
524 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
525 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
526 [@b{-march=score7}][@b{-march=score3}]
527 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
531 @emph{Target SPARC options:}
532 @c The order here is important. See c-sparc.texi.
533 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Aleon}|@b{-Asparclet}|@b{-Asparclite}
534 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av8plusb}|@b{-Av8plusc}|@b{-Av8plusd}
535 @b{-Av8plusv}|@b{-Av8plusm}|@b{-Av9}|@b{-Av9a}|@b{-Av9b}|@b{-Av9c}
536 @b{-Av9d}|@b{-Av9e}|@b{-Av9v}|@b{-Av9m}|@b{-Asparc}|@b{-Asparcvis}
537 @b{-Asparcvis2}|@b{-Asparcfmaf}|@b{-Asparcima}|@b{-Asparcvis3}
538 @b{-Asparcvisr}|@b{-Asparc5}]
539 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}]|@b{-xarch=v8plusb}|@b{-xarch=v8plusc}
540 @b{-xarch=v8plusd}|@b{-xarch=v8plusv}|@b{-xarch=v8plusm}|@b{-xarch=v9}
541 @b{-xarch=v9a}|@b{-xarch=v9b}|@b{-xarch=v9c}|@b{-xarch=v9d}|@b{-xarch=v9e}
542 @b{-xarch=v9v}|@b{-xarch=v9m}|@b{-xarch=sparc}|@b{-xarch=sparcvis}
543 @b{-xarch=sparcvis2}|@b{-xarch=sparcfmaf}|@b{-xarch=sparcima}
544 @b{-xarch=sparcvis3}|@b{-xarch=sparcvisr}|@b{-xarch=sparc5}
547 [@b{--enforce-aligned-data}][@b{--dcti-couples-detect}]
551 @emph{Target TIC54X options:}
552 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
553 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
557 @emph{Target TIC6X options:}
558 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
559 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
560 [@b{-mpic}|@b{-mno-pic}]
564 @emph{Target TILE-Gx options:}
565 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
568 @c TILEPro has no machine-dependent assembler options
572 @emph{Target Visium options:}
573 [@b{-mtune=@var{arch}}]
577 @emph{Target Xtensa options:}
578 [@b{--[no-]text-section-literals}] [@b{--[no-]auto-litpools}]
579 [@b{--[no-]absolute-literals}]
580 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
581 [@b{--[no-]transform}]
582 [@b{--rename-section} @var{oldname}=@var{newname}]
583 [@b{--[no-]trampolines}]
587 @emph{Target Z80 options:}
588 [@b{-z80}] [@b{-r800}]
589 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
590 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
591 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
592 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
593 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
594 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
598 @c Z8000 has no machine-dependent assembler options
607 @include at-file.texi
610 Turn on listings, in any of a variety of ways:
614 omit false conditionals
617 omit debugging directives
620 include general information, like @value{AS} version and options passed
623 include high-level source
629 include macro expansions
632 omit forms processing
638 set the name of the listing file
641 You may combine these options; for example, use @samp{-aln} for assembly
642 listing without forms processing. The @samp{=file} option, if used, must be
643 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
646 Begin in alternate macro mode.
648 @xref{Altmacro,,@code{.altmacro}}.
651 @item --compress-debug-sections
652 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
653 ELF ABI. The resulting object file may not be compatible with older
654 linkers and object file utilities. Note if compression would make a
655 given section @emph{larger} then it is not compressed.
658 @cindex @samp{--compress-debug-sections=} option
659 @item --compress-debug-sections=none
660 @itemx --compress-debug-sections=zlib
661 @itemx --compress-debug-sections=zlib-gnu
662 @itemx --compress-debug-sections=zlib-gabi
663 These options control how DWARF debug sections are compressed.
664 @option{--compress-debug-sections=none} is equivalent to
665 @option{--nocompress-debug-sections}.
666 @option{--compress-debug-sections=zlib} and
667 @option{--compress-debug-sections=zlib-gabi} are equivalent to
668 @option{--compress-debug-sections}.
669 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
670 sections using zlib. The debug sections are renamed to begin with
671 @samp{.zdebug}. Note if compression would make a given section
672 @emph{larger} then it is not compressed nor renamed.
676 @item --nocompress-debug-sections
677 Do not compress DWARF debug sections. This is usually the default for all
678 targets except the x86/x86_64, but a configure time option can be used to
682 Ignored. This option is accepted for script compatibility with calls to
685 @item --debug-prefix-map @var{old}=@var{new}
686 When assembling files in directory @file{@var{old}}, record debugging
687 information describing them as in @file{@var{new}} instead.
689 @item --defsym @var{sym}=@var{value}
690 Define the symbol @var{sym} to be @var{value} before assembling the input file.
691 @var{value} must be an integer constant. As in C, a leading @samp{0x}
692 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
693 value. The value of the symbol can be overridden inside a source file via the
694 use of a @code{.set} pseudo-op.
697 ``fast''---skip whitespace and comment preprocessing (assume source is
702 Generate debugging information for each assembler source line using whichever
703 debug format is preferred by the target. This currently means either STABS,
707 Generate stabs debugging information for each assembler line. This
708 may help debugging assembler code, if the debugger can handle it.
711 Generate stabs debugging information for each assembler line, with GNU
712 extensions that probably only gdb can handle, and that could make other
713 debuggers crash or refuse to read your program. This
714 may help debugging assembler code. Currently the only GNU extension is
715 the location of the current working directory at assembling time.
718 Generate DWARF2 debugging information for each assembler line. This
719 may help debugging assembler code, if the debugger can handle it. Note---this
720 option is only supported by some targets, not all of them.
722 @item --gdwarf-sections
723 Instead of creating a .debug_line section, create a series of
724 .debug_line.@var{foo} sections where @var{foo} is the name of the
725 corresponding code section. For example a code section called @var{.text.func}
726 will have its dwarf line number information placed into a section called
727 @var{.debug_line.text.func}. If the code section is just called @var{.text}
728 then debug line section will still be called just @var{.debug_line} without any
732 @item --size-check=error
733 @itemx --size-check=warning
734 Issue an error or warning for invalid ELF .size directive.
736 @item --elf-stt-common=no
737 @itemx --elf-stt-common=yes
738 These options control whether the ELF assembler should generate common
739 symbols with the @code{STT_COMMON} type. The default can be controlled
740 by a configure option @option{--enable-elf-stt-common}.
744 Print a summary of the command line options and exit.
747 Print a summary of all target specific options and exit.
750 Add directory @var{dir} to the search list for @code{.include} directives.
753 Don't warn about signed overflow.
756 @ifclear DIFF-TBL-KLUGE
757 This option is accepted but has no effect on the @value{TARGET} family.
759 @ifset DIFF-TBL-KLUGE
760 Issue warnings when difference tables altered for long displacements.
765 Keep (in the symbol table) local symbols. These symbols start with
766 system-specific local label prefixes, typically @samp{.L} for ELF systems
767 or @samp{L} for traditional a.out systems.
772 @item --listing-lhs-width=@var{number}
773 Set the maximum width, in words, of the output data column for an assembler
774 listing to @var{number}.
776 @item --listing-lhs-width2=@var{number}
777 Set the maximum width, in words, of the output data column for continuation
778 lines in an assembler listing to @var{number}.
780 @item --listing-rhs-width=@var{number}
781 Set the maximum width of an input source line, as displayed in a listing, to
784 @item --listing-cont-lines=@var{number}
785 Set the maximum number of lines printed in a listing for a single line of input
788 @item --no-pad-sections
789 Stop the assembler for padding the ends of output sections to the alignment
790 of that section. The default is to pad the sections, but this can waste space
791 which might be needed on targets which have tight memory constraints.
793 @item -o @var{objfile}
794 Name the object-file output from @command{@value{AS}} @var{objfile}.
797 Fold the data section into the text section.
799 @item --hash-size=@var{number}
800 Set the default size of GAS's hash tables to a prime number close to
801 @var{number}. Increasing this value can reduce the length of time it takes the
802 assembler to perform its tasks, at the expense of increasing the assembler's
803 memory requirements. Similarly reducing this value can reduce the memory
804 requirements at the expense of speed.
806 @item --reduce-memory-overheads
807 This option reduces GAS's memory requirements, at the expense of making the
808 assembly processes slower. Currently this switch is a synonym for
809 @samp{--hash-size=4051}, but in the future it may have other effects as well.
812 @item --sectname-subst
813 Honor substitution sequences in section names.
815 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
820 Print the maximum space (in bytes) and total time (in seconds) used by
823 @item --strip-local-absolute
824 Remove local absolute symbols from the outgoing symbol table.
828 Print the @command{as} version.
831 Print the @command{as} version and exit.
835 Suppress warning messages.
837 @item --fatal-warnings
838 Treat warnings as errors.
841 Don't suppress warning messages or treat them as errors.
850 Generate an object file even after errors.
852 @item -- | @var{files} @dots{}
853 Standard input, or source files to assemble.
861 @xref{AArch64 Options}, for the options available when @value{AS} is configured
862 for the 64-bit mode of the ARM Architecture (AArch64).
867 The following options are available when @value{AS} is configured for the
868 64-bit mode of the ARM Architecture (AArch64).
871 @include c-aarch64.texi
872 @c ended inside the included file
880 @xref{Alpha Options}, for the options available when @value{AS} is configured
881 for an Alpha processor.
886 The following options are available when @value{AS} is configured for an Alpha
890 @include c-alpha.texi
891 @c ended inside the included file
898 The following options are available when @value{AS} is configured for an ARC
902 @item -mcpu=@var{cpu}
903 This option selects the core processor variant.
905 Select either big-endian (-EB) or little-endian (-EL) output.
907 Enable Code Density extenssion instructions.
912 The following options are available when @value{AS} is configured for the ARM
916 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
917 Specify which ARM processor variant is the target.
918 @item -march=@var{architecture}[+@var{extension}@dots{}]
919 Specify which ARM architecture variant is used by the target.
920 @item -mfpu=@var{floating-point-format}
921 Select which Floating Point architecture is the target.
922 @item -mfloat-abi=@var{abi}
923 Select which floating point ABI is in use.
925 Enable Thumb only instruction decoding.
926 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
927 Select which procedure calling convention is in use.
929 Select either big-endian (-EB) or little-endian (-EL) output.
930 @item -mthumb-interwork
931 Specify that the code has been generated with interworking between Thumb and
934 Turns on CodeComposer Studio assembly syntax compatibility mode.
936 Specify that PIC code has been generated.
944 @xref{Blackfin Options}, for the options available when @value{AS} is
945 configured for the Blackfin processor family.
950 The following options are available when @value{AS} is configured for
951 the Blackfin processor family.
955 @c ended inside the included file
962 See the info pages for documentation of the CRIS-specific options.
966 The following options are available when @value{AS} is configured for
969 @cindex D10V optimization
970 @cindex optimization, D10V
972 Optimize output by parallelizing instructions.
977 The following options are available when @value{AS} is configured for a D30V
980 @cindex D30V optimization
981 @cindex optimization, D30V
983 Optimize output by parallelizing instructions.
987 Warn when nops are generated.
989 @cindex D30V nops after 32-bit multiply
991 Warn when a nop after a 32-bit multiply instruction is generated.
997 The following options are available when @value{AS} is configured for the
998 Adapteva EPIPHANY series.
1001 @xref{Epiphany Options}, for the options available when @value{AS} is
1002 configured for an Epiphany processor.
1006 @c man begin OPTIONS
1007 The following options are available when @value{AS} is configured for
1008 an Epiphany processor.
1010 @c man begin INCLUDE
1011 @include c-epiphany.texi
1012 @c ended inside the included file
1020 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1021 for an H8/300 processor.
1025 @c man begin OPTIONS
1026 The following options are available when @value{AS} is configured for an H8/300
1029 @c man begin INCLUDE
1030 @include c-h8300.texi
1031 @c ended inside the included file
1039 @xref{i386-Options}, for the options available when @value{AS} is
1040 configured for an i386 processor.
1044 @c man begin OPTIONS
1045 The following options are available when @value{AS} is configured for
1048 @c man begin INCLUDE
1049 @include c-i386.texi
1050 @c ended inside the included file
1055 @c man begin OPTIONS
1057 The following options are available when @value{AS} is configured for the
1058 Intel 80960 processor.
1061 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
1062 Specify which variant of the 960 architecture is the target.
1065 Add code to collect statistics about branches taken.
1068 Do not alter compare-and-branch instructions for long displacements;
1075 The following options are available when @value{AS} is configured for the
1081 Specifies that the extended IP2022 instructions are allowed.
1084 Restores the default behaviour, which restricts the permitted instructions to
1085 just the basic IP2022 ones.
1091 The following options are available when @value{AS} is configured for the
1092 Renesas M32C and M16C processors.
1097 Assemble M32C instructions.
1100 Assemble M16C instructions (the default).
1103 Enable support for link-time relaxations.
1106 Support H'00 style hex constants in addition to 0x00 style.
1112 The following options are available when @value{AS} is configured for the
1113 Renesas M32R (formerly Mitsubishi M32R) series.
1118 Specify which processor in the M32R family is the target. The default
1119 is normally the M32R, but this option changes it to the M32RX.
1121 @item --warn-explicit-parallel-conflicts or --Wp
1122 Produce warning messages when questionable parallel constructs are
1125 @item --no-warn-explicit-parallel-conflicts or --Wnp
1126 Do not produce warning messages when questionable parallel constructs are
1133 The following options are available when @value{AS} is configured for the
1134 Motorola 68000 series.
1139 Shorten references to undefined symbols, to one word instead of two.
1141 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1142 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1143 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1144 Specify what processor in the 68000 family is the target. The default
1145 is normally the 68020, but this can be changed at configuration time.
1147 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1148 The target machine does (or does not) have a floating-point coprocessor.
1149 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1150 the basic 68000 is not compatible with the 68881, a combination of the
1151 two can be specified, since it's possible to do emulation of the
1152 coprocessor instructions with the main processor.
1154 @item -m68851 | -mno-68851
1155 The target machine does (or does not) have a memory-management
1156 unit coprocessor. The default is to assume an MMU for 68020 and up.
1164 @xref{Nios II Options}, for the options available when @value{AS} is configured
1165 for an Altera Nios II processor.
1169 @c man begin OPTIONS
1170 The following options are available when @value{AS} is configured for an
1171 Altera Nios II processor.
1173 @c man begin INCLUDE
1174 @include c-nios2.texi
1175 @c ended inside the included file
1181 For details about the PDP-11 machine dependent features options,
1182 see @ref{PDP-11-Options}.
1185 @item -mpic | -mno-pic
1186 Generate position-independent (or position-dependent) code. The
1187 default is @option{-mpic}.
1190 @itemx -mall-extensions
1191 Enable all instruction set extensions. This is the default.
1193 @item -mno-extensions
1194 Disable all instruction set extensions.
1196 @item -m@var{extension} | -mno-@var{extension}
1197 Enable (or disable) a particular instruction set extension.
1200 Enable the instruction set extensions supported by a particular CPU, and
1201 disable all other extensions.
1203 @item -m@var{machine}
1204 Enable the instruction set extensions supported by a particular machine
1205 model, and disable all other extensions.
1211 The following options are available when @value{AS} is configured for
1212 a picoJava processor.
1216 @cindex PJ endianness
1217 @cindex endianness, PJ
1218 @cindex big endian output, PJ
1220 Generate ``big endian'' format output.
1222 @cindex little endian output, PJ
1224 Generate ``little endian'' format output.
1230 The following options are available when @value{AS} is configured for the
1231 Motorola 68HC11 or 68HC12 series.
1235 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1236 Specify what processor is the target. The default is
1237 defined by the configuration option when building the assembler.
1239 @item --xgate-ramoffset
1240 Instruct the linker to offset RAM addresses from S12X address space into
1241 XGATE address space.
1244 Specify to use the 16-bit integer ABI.
1247 Specify to use the 32-bit integer ABI.
1249 @item -mshort-double
1250 Specify to use the 32-bit double ABI.
1253 Specify to use the 64-bit double ABI.
1255 @item --force-long-branches
1256 Relative branches are turned into absolute ones. This concerns
1257 conditional branches, unconditional branches and branches to a
1260 @item -S | --short-branches
1261 Do not turn relative branches into absolute ones
1262 when the offset is out of range.
1264 @item --strict-direct-mode
1265 Do not turn the direct addressing mode into extended addressing mode
1266 when the instruction does not support direct addressing mode.
1268 @item --print-insn-syntax
1269 Print the syntax of instruction in case of error.
1271 @item --print-opcodes
1272 Print the list of instructions with syntax and then exit.
1274 @item --generate-example
1275 Print an example of instruction for each possible instruction and then exit.
1276 This option is only useful for testing @command{@value{AS}}.
1282 The following options are available when @command{@value{AS}} is configured
1283 for the SPARC architecture:
1286 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1287 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1288 Explicitly select a variant of the SPARC architecture.
1290 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1291 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1293 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1294 UltraSPARC extensions.
1296 @item -xarch=v8plus | -xarch=v8plusa
1297 For compatibility with the Solaris v9 assembler. These options are
1298 equivalent to -Av8plus and -Av8plusa, respectively.
1301 Warn when the assembler switches to another architecture.
1306 The following options are available when @value{AS} is configured for the 'c54x
1311 Enable extended addressing mode. All addresses and relocations will assume
1312 extended addressing (usually 23 bits).
1313 @item -mcpu=@var{CPU_VERSION}
1314 Sets the CPU version being compiled for.
1315 @item -merrors-to-file @var{FILENAME}
1316 Redirect error output to a file, for broken systems which don't support such
1317 behaviour in the shell.
1322 The following options are available when @value{AS} is configured for
1327 This option sets the largest size of an object that can be referenced
1328 implicitly with the @code{gp} register. It is only accepted for targets that
1329 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1331 @cindex MIPS endianness
1332 @cindex endianness, MIPS
1333 @cindex big endian output, MIPS
1335 Generate ``big endian'' format output.
1337 @cindex little endian output, MIPS
1339 Generate ``little endian'' format output.
1357 Generate code for a particular MIPS Instruction Set Architecture level.
1358 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1359 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1360 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1361 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1362 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1363 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1364 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1365 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1366 MIPS64 Release 6 ISA processors, respectively.
1368 @item -march=@var{cpu}
1369 Generate code for a particular MIPS CPU.
1371 @item -mtune=@var{cpu}
1372 Schedule and tune for a particular MIPS CPU.
1376 Cause nops to be inserted if the read of the destination register
1377 of an mfhi or mflo instruction occurs in the following two instructions.
1380 @itemx -mno-fix-rm7000
1381 Cause nops to be inserted if a dmult or dmultu instruction is
1382 followed by a load instruction.
1386 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1387 section instead of the standard ELF .stabs sections.
1391 Control generation of @code{.pdr} sections.
1395 The register sizes are normally inferred from the ISA and ABI, but these
1396 flags force a certain group of registers to be treated as 32 bits wide at
1397 all times. @samp{-mgp32} controls the size of general-purpose registers
1398 and @samp{-mfp32} controls the size of floating-point registers.
1402 The register sizes are normally inferred from the ISA and ABI, but these
1403 flags force a certain group of registers to be treated as 64 bits wide at
1404 all times. @samp{-mgp64} controls the size of general-purpose registers
1405 and @samp{-mfp64} controls the size of floating-point registers.
1408 The register sizes are normally inferred from the ISA and ABI, but using
1409 this flag in combination with @samp{-mabi=32} enables an ABI variant
1410 which will operate correctly with floating-point registers which are
1414 @itemx -mno-odd-spreg
1415 Enable use of floating-point operations on odd-numbered single-precision
1416 registers when supported by the ISA. @samp{-mfpxx} implies
1417 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1421 Generate code for the MIPS 16 processor. This is equivalent to putting
1422 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1423 turns off this option.
1426 @itemx -mno-micromips
1427 Generate code for the microMIPS processor. This is equivalent to putting
1428 @code{.set micromips} at the start of the assembly file. @samp{-mno-micromips}
1429 turns off this option. This is equivalent to putting @code{.set nomicromips}
1430 at the start of the assembly file.
1433 @itemx -mno-smartmips
1434 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1435 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1436 @samp{-mno-smartmips} turns off this option.
1440 Generate code for the MIPS-3D Application Specific Extension.
1441 This tells the assembler to accept MIPS-3D instructions.
1442 @samp{-no-mips3d} turns off this option.
1446 Generate code for the MDMX Application Specific Extension.
1447 This tells the assembler to accept MDMX instructions.
1448 @samp{-no-mdmx} turns off this option.
1452 Generate code for the DSP Release 1 Application Specific Extension.
1453 This tells the assembler to accept DSP Release 1 instructions.
1454 @samp{-mno-dsp} turns off this option.
1458 Generate code for the DSP Release 2 Application Specific Extension.
1459 This option implies @samp{-mdsp}.
1460 This tells the assembler to accept DSP Release 2 instructions.
1461 @samp{-mno-dspr2} turns off this option.
1465 Generate code for the DSP Release 3 Application Specific Extension.
1466 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1467 This tells the assembler to accept DSP Release 3 instructions.
1468 @samp{-mno-dspr3} turns off this option.
1472 Generate code for the MIPS SIMD Architecture Extension.
1473 This tells the assembler to accept MSA instructions.
1474 @samp{-mno-msa} turns off this option.
1478 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1479 This tells the assembler to accept XPA instructions.
1480 @samp{-mno-xpa} turns off this option.
1484 Generate code for the MT Application Specific Extension.
1485 This tells the assembler to accept MT instructions.
1486 @samp{-mno-mt} turns off this option.
1490 Generate code for the MCU Application Specific Extension.
1491 This tells the assembler to accept MCU instructions.
1492 @samp{-mno-mcu} turns off this option.
1496 Only use 32-bit instruction encodings when generating code for the
1497 microMIPS processor. This option inhibits the use of any 16-bit
1498 instructions. This is equivalent to putting @code{.set insn32} at
1499 the start of the assembly file. @samp{-mno-insn32} turns off this
1500 option. This is equivalent to putting @code{.set noinsn32} at the
1501 start of the assembly file. By default @samp{-mno-insn32} is
1502 selected, allowing all instructions to be used.
1504 @item --construct-floats
1505 @itemx --no-construct-floats
1506 The @samp{--no-construct-floats} option disables the construction of
1507 double width floating point constants by loading the two halves of the
1508 value into the two single width floating point registers that make up
1509 the double width register. By default @samp{--construct-floats} is
1510 selected, allowing construction of these floating point constants.
1512 @item --relax-branch
1513 @itemx --no-relax-branch
1514 The @samp{--relax-branch} option enables the relaxation of out-of-range
1515 branches. By default @samp{--no-relax-branch} is selected, causing any
1516 out-of-range branches to produce an error.
1518 @item -mnan=@var{encoding}
1519 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1520 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1523 @item --emulation=@var{name}
1524 This option was formerly used to switch between ELF and ECOFF output
1525 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1526 removed in GAS 2.24, so the option now serves little purpose.
1527 It is retained for backwards compatibility.
1529 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1530 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1531 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1532 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1533 preferred options instead.
1536 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1543 Control how to deal with multiplication overflow and division by zero.
1544 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1545 (and only work for Instruction Set Architecture level 2 and higher);
1546 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1550 When this option is used, @command{@value{AS}} will issue a warning every
1551 time it generates a nop instruction from a macro.
1556 The following options are available when @value{AS} is configured for
1562 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1563 The command line option @samp{-nojsri2bsr} can be used to disable it.
1567 Enable or disable the silicon filter behaviour. By default this is disabled.
1568 The default can be overridden by the @samp{-sifilter} command line option.
1571 Alter jump instructions for long displacements.
1573 @item -mcpu=[210|340]
1574 Select the cpu type on the target hardware. This controls which instructions
1578 Assemble for a big endian target.
1581 Assemble for a little endian target.
1590 @xref{Meta Options}, for the options available when @value{AS} is configured
1591 for a Meta processor.
1595 @c man begin OPTIONS
1596 The following options are available when @value{AS} is configured for a
1599 @c man begin INCLUDE
1600 @include c-metag.texi
1601 @c ended inside the included file
1606 @c man begin OPTIONS
1608 See the info pages for documentation of the MMIX-specific options.
1614 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1615 for a NDS32 processor.
1617 @c ended inside the included file
1621 @c man begin OPTIONS
1622 The following options are available when @value{AS} is configured for a
1625 @c man begin INCLUDE
1626 @include c-nds32.texi
1627 @c ended inside the included file
1634 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1635 for a PowerPC processor.
1639 @c man begin OPTIONS
1640 The following options are available when @value{AS} is configured for a
1643 @c man begin INCLUDE
1645 @c ended inside the included file
1650 @c man begin OPTIONS
1652 See the info pages for documentation of the RX-specific options.
1656 The following options are available when @value{AS} is configured for the s390
1662 Select the word size, either 31/32 bits or 64 bits.
1665 Select the architecture mode, either the Enterprise System
1666 Architecture (esa) or the z/Architecture mode (zarch).
1667 @item -march=@var{processor}
1668 Specify which s390 processor variant is the target, @samp{g5} (or
1669 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1670 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1671 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1672 or @samp{z13} (or @samp{arch11}).
1674 @itemx -mno-regnames
1675 Allow or disallow symbolic names for registers.
1676 @item -mwarn-areg-zero
1677 Warn whenever the operand for a base or index register has been specified
1678 but evaluates to zero.
1686 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1687 for a TMS320C6000 processor.
1691 @c man begin OPTIONS
1692 The following options are available when @value{AS} is configured for a
1693 TMS320C6000 processor.
1695 @c man begin INCLUDE
1696 @include c-tic6x.texi
1697 @c ended inside the included file
1705 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1706 for a TILE-Gx processor.
1710 @c man begin OPTIONS
1711 The following options are available when @value{AS} is configured for a TILE-Gx
1714 @c man begin INCLUDE
1715 @include c-tilegx.texi
1716 @c ended inside the included file
1724 @xref{Visium Options}, for the options available when @value{AS} is configured
1725 for a Visium processor.
1729 @c man begin OPTIONS
1730 The following option is available when @value{AS} is configured for a Visium
1733 @c man begin INCLUDE
1734 @include c-visium.texi
1735 @c ended inside the included file
1743 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1744 for an Xtensa processor.
1748 @c man begin OPTIONS
1749 The following options are available when @value{AS} is configured for an
1752 @c man begin INCLUDE
1753 @include c-xtensa.texi
1754 @c ended inside the included file
1759 @c man begin OPTIONS
1762 The following options are available when @value{AS} is configured for
1763 a Z80 family processor.
1766 Assemble for Z80 processor.
1768 Assemble for R800 processor.
1769 @item -ignore-undocumented-instructions
1771 Assemble undocumented Z80 instructions that also work on R800 without warning.
1772 @item -ignore-unportable-instructions
1774 Assemble all undocumented Z80 instructions without warning.
1775 @item -warn-undocumented-instructions
1777 Issue a warning for undocumented Z80 instructions that also work on R800.
1778 @item -warn-unportable-instructions
1780 Issue a warning for undocumented Z80 instructions that do not work on R800.
1781 @item -forbid-undocumented-instructions
1783 Treat all undocumented instructions as errors.
1784 @item -forbid-unportable-instructions
1786 Treat undocumented Z80 instructions that do not work on R800 as errors.
1793 * Manual:: Structure of this Manual
1794 * GNU Assembler:: The GNU Assembler
1795 * Object Formats:: Object File Formats
1796 * Command Line:: Command Line
1797 * Input Files:: Input Files
1798 * Object:: Output (Object) File
1799 * Errors:: Error and Warning Messages
1803 @section Structure of this Manual
1805 @cindex manual, structure and purpose
1806 This manual is intended to describe what you need to know to use
1807 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1808 notation for symbols, constants, and expressions; the directives that
1809 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1812 We also cover special features in the @value{TARGET}
1813 configuration of @command{@value{AS}}, including assembler directives.
1816 This manual also describes some of the machine-dependent features of
1817 various flavors of the assembler.
1820 @cindex machine instructions (not covered)
1821 On the other hand, this manual is @emph{not} intended as an introduction
1822 to programming in assembly language---let alone programming in general!
1823 In a similar vein, we make no attempt to introduce the machine
1824 architecture; we do @emph{not} describe the instruction set, standard
1825 mnemonics, registers or addressing modes that are standard to a
1826 particular architecture.
1828 You may want to consult the manufacturer's
1829 machine architecture manual for this information.
1833 For information on the H8/300 machine instruction set, see @cite{H8/300
1834 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1835 Programming Manual} (Renesas).
1838 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1839 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1840 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1841 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1844 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1848 @c I think this is premature---doc@cygnus.com, 17jan1991
1850 Throughout this manual, we assume that you are running @dfn{GNU},
1851 the portable operating system from the @dfn{Free Software
1852 Foundation, Inc.}. This restricts our attention to certain kinds of
1853 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1854 once this assumption is granted examples and definitions need less
1857 @command{@value{AS}} is part of a team of programs that turn a high-level
1858 human-readable series of instructions into a low-level
1859 computer-readable series of instructions. Different versions of
1860 @command{@value{AS}} are used for different kinds of computer.
1863 @c There used to be a section "Terminology" here, which defined
1864 @c "contents", "byte", "word", and "long". Defining "word" to any
1865 @c particular size is confusing when the .word directive may generate 16
1866 @c bits on one machine and 32 bits on another; in general, for the user
1867 @c version of this manual, none of these terms seem essential to define.
1868 @c They were used very little even in the former draft of the manual;
1869 @c this draft makes an effort to avoid them (except in names of
1873 @section The GNU Assembler
1875 @c man begin DESCRIPTION
1877 @sc{gnu} @command{as} is really a family of assemblers.
1879 This manual describes @command{@value{AS}}, a member of that family which is
1880 configured for the @value{TARGET} architectures.
1882 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1883 should find a fairly similar environment when you use it on another
1884 architecture. Each version has much in common with the others,
1885 including object file formats, most assembler directives (often called
1886 @dfn{pseudo-ops}) and assembler syntax.@refill
1888 @cindex purpose of @sc{gnu} assembler
1889 @command{@value{AS}} is primarily intended to assemble the output of the
1890 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1891 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1892 assemble correctly everything that other assemblers for the same
1893 machine would assemble.
1895 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1898 @c This remark should appear in generic version of manual; assumption
1899 @c here is that generic version sets M680x0.
1900 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1901 assembler for the same architecture; for example, we know of several
1902 incompatible versions of 680x0 assembly language syntax.
1907 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1908 program in one pass of the source file. This has a subtle impact on the
1909 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1911 @node Object Formats
1912 @section Object File Formats
1914 @cindex object file format
1915 The @sc{gnu} assembler can be configured to produce several alternative
1916 object file formats. For the most part, this does not affect how you
1917 write assembly language programs; but directives for debugging symbols
1918 are typically different in different file formats. @xref{Symbol
1919 Attributes,,Symbol Attributes}.
1922 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1923 @value{OBJ-NAME} format object files.
1925 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1927 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1928 @code{b.out} or COFF format object files.
1931 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1932 SOM or ELF format object files.
1937 @section Command Line
1939 @cindex command line conventions
1941 After the program name @command{@value{AS}}, the command line may contain
1942 options and file names. Options may appear in any order, and may be
1943 before, after, or between file names. The order of file names is
1946 @cindex standard input, as input file
1948 @file{--} (two hyphens) by itself names the standard input file
1949 explicitly, as one of the files for @command{@value{AS}} to assemble.
1951 @cindex options, command line
1952 Except for @samp{--} any command line argument that begins with a
1953 hyphen (@samp{-}) is an option. Each option changes the behavior of
1954 @command{@value{AS}}. No option changes the way another option works. An
1955 option is a @samp{-} followed by one or more letters; the case of
1956 the letter is important. All options are optional.
1958 Some options expect exactly one file name to follow them. The file
1959 name may either immediately follow the option's letter (compatible
1960 with older assemblers) or it may be the next command argument (@sc{gnu}
1961 standard). These two command lines are equivalent:
1964 @value{AS} -o my-object-file.o mumble.s
1965 @value{AS} -omy-object-file.o mumble.s
1969 @section Input Files
1972 @cindex source program
1973 @cindex files, input
1974 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
1975 describe the program input to one run of @command{@value{AS}}. The program may
1976 be in one or more files; how the source is partitioned into files
1977 doesn't change the meaning of the source.
1979 @c I added "con" prefix to "catenation" just to prove I can overcome my
1980 @c APL training... doc@cygnus.com
1981 The source program is a concatenation of the text in all the files, in the
1984 @c man begin DESCRIPTION
1985 Each time you run @command{@value{AS}} it assembles exactly one source
1986 program. The source program is made up of one or more files.
1987 (The standard input is also a file.)
1989 You give @command{@value{AS}} a command line that has zero or more input file
1990 names. The input files are read (from left file name to right). A
1991 command line argument (in any position) that has no special meaning
1992 is taken to be an input file name.
1994 If you give @command{@value{AS}} no file names it attempts to read one input file
1995 from the @command{@value{AS}} standard input, which is normally your terminal. You
1996 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
1999 Use @samp{--} if you need to explicitly name the standard input file
2000 in your command line.
2002 If the source is empty, @command{@value{AS}} produces a small, empty object
2007 @subheading Filenames and Line-numbers
2009 @cindex input file linenumbers
2010 @cindex line numbers, in input files
2011 There are two ways of locating a line in the input file (or files) and
2012 either may be used in reporting error messages. One way refers to a line
2013 number in a physical file; the other refers to a line number in a
2014 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2016 @dfn{Physical files} are those files named in the command line given
2017 to @command{@value{AS}}.
2019 @dfn{Logical files} are simply names declared explicitly by assembler
2020 directives; they bear no relation to physical files. Logical file names help
2021 error messages reflect the original source file, when @command{@value{AS}} source
2022 is itself synthesized from other files. @command{@value{AS}} understands the
2023 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2024 @ref{File,,@code{.file}}.
2027 @section Output (Object) File
2033 Every time you run @command{@value{AS}} it produces an output file, which is
2034 your assembly language program translated into numbers. This file
2035 is the object file. Its default name is
2043 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
2045 You can give it another name by using the @option{-o} option. Conventionally,
2046 object file names end with @file{.o}. The default name is used for historical
2047 reasons: older assemblers were capable of assembling self-contained programs
2048 directly into a runnable program. (For some formats, this isn't currently
2049 possible, but it can be done for the @code{a.out} format.)
2053 The object file is meant for input to the linker @code{@value{LD}}. It contains
2054 assembled program code, information to help @code{@value{LD}} integrate
2055 the assembled program into a runnable file, and (optionally) symbolic
2056 information for the debugger.
2058 @c link above to some info file(s) like the description of a.out.
2059 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2062 @section Error and Warning Messages
2064 @c man begin DESCRIPTION
2066 @cindex error messages
2067 @cindex warning messages
2068 @cindex messages from assembler
2069 @command{@value{AS}} may write warnings and error messages to the standard error
2070 file (usually your terminal). This should not happen when a compiler
2071 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2072 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2073 grave problem that stops the assembly.
2077 @cindex format of warning messages
2078 Warning messages have the format
2081 file_name:@b{NNN}:Warning Message Text
2085 @cindex file names and line numbers, in warnings/errors
2086 (where @b{NNN} is a line number). If both a logical file name
2087 (@pxref{File,,@code{.file}}) and a logical line number
2089 (@pxref{Line,,@code{.line}})
2091 have been given then they will be used, otherwise the file name and line number
2092 in the current assembler source file will be used. The message text is
2093 intended to be self explanatory (in the grand Unix tradition).
2095 Note the file name must be set via the logical version of the @code{.file}
2096 directive, not the DWARF2 version of the @code{.file} directive. For example:
2100 error_assembler_source
2106 produces this output:
2110 asm.s:2: Error: no such instruction: `error_assembler_source'
2111 foo.c:31: Error: no such instruction: `error_c_source'
2114 @cindex format of error messages
2115 Error messages have the format
2118 file_name:@b{NNN}:FATAL:Error Message Text
2121 The file name and line number are derived as for warning
2122 messages. The actual message text may be rather less explanatory
2123 because many of them aren't supposed to happen.
2126 @chapter Command-Line Options
2128 @cindex options, all versions of assembler
2129 This chapter describes command-line options available in @emph{all}
2130 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2131 for options specific
2133 to the @value{TARGET} target.
2136 to particular machine architectures.
2139 @c man begin DESCRIPTION
2141 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2142 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2143 The assembler arguments must be separated from each other (and the @samp{-Wa})
2144 by commas. For example:
2147 gcc -c -g -O -Wa,-alh,-L file.c
2151 This passes two options to the assembler: @samp{-alh} (emit a listing to
2152 standard output with high-level and assembly source) and @samp{-L} (retain
2153 local symbols in the symbol table).
2155 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2156 command-line options are automatically passed to the assembler by the compiler.
2157 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2158 precisely what options it passes to each compilation pass, including the
2164 * a:: -a[cdghlns] enable listings
2165 * alternate:: --alternate enable alternate macro syntax
2166 * D:: -D for compatibility
2167 * f:: -f to work faster
2168 * I:: -I for .include search path
2169 @ifclear DIFF-TBL-KLUGE
2170 * K:: -K for compatibility
2172 @ifset DIFF-TBL-KLUGE
2173 * K:: -K for difference tables
2176 * L:: -L to retain local symbols
2177 * listing:: --listing-XXX to configure listing output
2178 * M:: -M or --mri to assemble in MRI compatibility mode
2179 * MD:: --MD for dependency tracking
2180 * no-pad-sections:: --no-pad-sections to stop section padding
2181 * o:: -o to name the object file
2182 * R:: -R to join data and text sections
2183 * statistics:: --statistics to see statistics about assembly
2184 * traditional-format:: --traditional-format for compatible output
2185 * v:: -v to announce version
2186 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2187 * Z:: -Z to make object file even after errors
2191 @section Enable Listings: @option{-a[cdghlns]}
2201 @cindex listings, enabling
2202 @cindex assembly listings, enabling
2204 These options enable listing output from the assembler. By itself,
2205 @samp{-a} requests high-level, assembly, and symbols listing.
2206 You can use other letters to select specific options for the list:
2207 @samp{-ah} requests a high-level language listing,
2208 @samp{-al} requests an output-program assembly listing, and
2209 @samp{-as} requests a symbol table listing.
2210 High-level listings require that a compiler debugging option like
2211 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2214 Use the @samp{-ag} option to print a first section with general assembly
2215 information, like @value{AS} version, switches passed, or time stamp.
2217 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2218 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2219 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2220 omitted from the listing.
2222 Use the @samp{-ad} option to omit debugging directives from the
2225 Once you have specified one of these options, you can further control
2226 listing output and its appearance using the directives @code{.list},
2227 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2229 The @samp{-an} option turns off all forms processing.
2230 If you do not request listing output with one of the @samp{-a} options, the
2231 listing-control directives have no effect.
2233 The letters after @samp{-a} may be combined into one option,
2234 @emph{e.g.}, @samp{-aln}.
2236 Note if the assembler source is coming from the standard input (e.g.,
2238 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
2239 is being used) then the listing will not contain any comments or preprocessor
2240 directives. This is because the listing code buffers input source lines from
2241 stdin only after they have been preprocessed by the assembler. This reduces
2242 memory usage and makes the code more efficient.
2245 @section @option{--alternate}
2248 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2251 @section @option{-D}
2254 This option has no effect whatsoever, but it is accepted to make it more
2255 likely that scripts written for other assemblers also work with
2256 @command{@value{AS}}.
2259 @section Work Faster: @option{-f}
2262 @cindex trusted compiler
2263 @cindex faster processing (@option{-f})
2264 @samp{-f} should only be used when assembling programs written by a
2265 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2266 and comment preprocessing on
2267 the input file(s) before assembling them. @xref{Preprocessing,
2271 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2272 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2277 @section @code{.include} Search Path: @option{-I} @var{path}
2279 @kindex -I @var{path}
2280 @cindex paths for @code{.include}
2281 @cindex search path for @code{.include}
2282 @cindex @code{include} directive search path
2283 Use this option to add a @var{path} to the list of directories
2284 @command{@value{AS}} searches for files specified in @code{.include}
2285 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2286 many times as necessary to include a variety of paths. The current
2287 working directory is always searched first; after that, @command{@value{AS}}
2288 searches any @samp{-I} directories in the same order as they were
2289 specified (left to right) on the command line.
2292 @section Difference Tables: @option{-K}
2295 @ifclear DIFF-TBL-KLUGE
2296 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2297 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2298 where it can be used to warn when the assembler alters the machine code
2299 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2300 family does not have the addressing limitations that sometimes lead to this
2301 alteration on other platforms.
2304 @ifset DIFF-TBL-KLUGE
2305 @cindex difference tables, warning
2306 @cindex warning for altered difference tables
2307 @command{@value{AS}} sometimes alters the code emitted for directives of the
2308 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2309 You can use the @samp{-K} option if you want a warning issued when this
2314 @section Include Local Symbols: @option{-L}
2317 @cindex local symbols, retaining in output
2318 Symbols beginning with system-specific local label prefixes, typically
2319 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2320 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2321 such symbols when debugging, because they are intended for the use of
2322 programs (like compilers) that compose assembler programs, not for your
2323 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2324 such symbols, so you do not normally debug with them.
2326 This option tells @command{@value{AS}} to retain those local symbols
2327 in the object file. Usually if you do this you also tell the linker
2328 @code{@value{LD}} to preserve those symbols.
2331 @section Configuring listing output: @option{--listing}
2333 The listing feature of the assembler can be enabled via the command line switch
2334 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2335 hex dump of the corresponding locations in the output object file, and displays
2336 them as a listing file. The format of this listing can be controlled by
2337 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2338 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2339 @code{.psize} (@pxref{Psize}), and
2340 @code{.eject} (@pxref{Eject}) and also by the following switches:
2343 @item --listing-lhs-width=@samp{number}
2344 @kindex --listing-lhs-width
2345 @cindex Width of first line disassembly output
2346 Sets the maximum width, in words, of the first line of the hex byte dump. This
2347 dump appears on the left hand side of the listing output.
2349 @item --listing-lhs-width2=@samp{number}
2350 @kindex --listing-lhs-width2
2351 @cindex Width of continuation lines of disassembly output
2352 Sets the maximum width, in words, of any further lines of the hex byte dump for
2353 a given input source line. If this value is not specified, it defaults to being
2354 the same as the value specified for @samp{--listing-lhs-width}. If neither
2355 switch is used the default is to one.
2357 @item --listing-rhs-width=@samp{number}
2358 @kindex --listing-rhs-width
2359 @cindex Width of source line output
2360 Sets the maximum width, in characters, of the source line that is displayed
2361 alongside the hex dump. The default value for this parameter is 100. The
2362 source line is displayed on the right hand side of the listing output.
2364 @item --listing-cont-lines=@samp{number}
2365 @kindex --listing-cont-lines
2366 @cindex Maximum number of continuation lines
2367 Sets the maximum number of continuation lines of hex dump that will be
2368 displayed for a given single line of source input. The default value is 4.
2372 @section Assemble in MRI Compatibility Mode: @option{-M}
2375 @cindex MRI compatibility mode
2376 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2377 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2378 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
2379 configured target) assembler from Microtec Research. The exact nature of the
2380 MRI syntax will not be documented here; see the MRI manuals for more
2381 information. Note in particular that the handling of macros and macro
2382 arguments is somewhat different. The purpose of this option is to permit
2383 assembling existing MRI assembler code using @command{@value{AS}}.
2385 The MRI compatibility is not complete. Certain operations of the MRI assembler
2386 depend upon its object file format, and can not be supported using other object
2387 file formats. Supporting these would require enhancing each object file format
2388 individually. These are:
2391 @item global symbols in common section
2393 The m68k MRI assembler supports common sections which are merged by the linker.
2394 Other object file formats do not support this. @command{@value{AS}} handles
2395 common sections by treating them as a single common symbol. It permits local
2396 symbols to be defined within a common section, but it can not support global
2397 symbols, since it has no way to describe them.
2399 @item complex relocations
2401 The MRI assemblers support relocations against a negated section address, and
2402 relocations which combine the start addresses of two or more sections. These
2403 are not support by other object file formats.
2405 @item @code{END} pseudo-op specifying start address
2407 The MRI @code{END} pseudo-op permits the specification of a start address.
2408 This is not supported by other object file formats. The start address may
2409 instead be specified using the @option{-e} option to the linker, or in a linker
2412 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2414 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2415 name to the output file. This is not supported by other object file formats.
2417 @item @code{ORG} pseudo-op
2419 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2420 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2421 which changes the location within the current section. Absolute sections are
2422 not supported by other object file formats. The address of a section may be
2423 assigned within a linker script.
2426 There are some other features of the MRI assembler which are not supported by
2427 @command{@value{AS}}, typically either because they are difficult or because they
2428 seem of little consequence. Some of these may be supported in future releases.
2432 @item EBCDIC strings
2434 EBCDIC strings are not supported.
2436 @item packed binary coded decimal
2438 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2439 and @code{DCB.P} pseudo-ops are not supported.
2441 @item @code{FEQU} pseudo-op
2443 The m68k @code{FEQU} pseudo-op is not supported.
2445 @item @code{NOOBJ} pseudo-op
2447 The m68k @code{NOOBJ} pseudo-op is not supported.
2449 @item @code{OPT} branch control options
2451 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2452 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2453 relaxes all branches, whether forward or backward, to an appropriate size, so
2454 these options serve no purpose.
2456 @item @code{OPT} list control options
2458 The following m68k @code{OPT} list control options are ignored: @code{C},
2459 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2460 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2462 @item other @code{OPT} options
2464 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2465 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2467 @item @code{OPT} @code{D} option is default
2469 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2470 @code{OPT NOD} may be used to turn it off.
2472 @item @code{XREF} pseudo-op.
2474 The m68k @code{XREF} pseudo-op is ignored.
2476 @item @code{.debug} pseudo-op
2478 The i960 @code{.debug} pseudo-op is not supported.
2480 @item @code{.extended} pseudo-op
2482 The i960 @code{.extended} pseudo-op is not supported.
2484 @item @code{.list} pseudo-op.
2486 The various options of the i960 @code{.list} pseudo-op are not supported.
2488 @item @code{.optimize} pseudo-op
2490 The i960 @code{.optimize} pseudo-op is not supported.
2492 @item @code{.output} pseudo-op
2494 The i960 @code{.output} pseudo-op is not supported.
2496 @item @code{.setreal} pseudo-op
2498 The i960 @code{.setreal} pseudo-op is not supported.
2503 @section Dependency Tracking: @option{--MD}
2506 @cindex dependency tracking
2509 @command{@value{AS}} can generate a dependency file for the file it creates. This
2510 file consists of a single rule suitable for @code{make} describing the
2511 dependencies of the main source file.
2513 The rule is written to the file named in its argument.
2515 This feature is used in the automatic updating of makefiles.
2517 @node no-pad-sections
2518 @section Output Section Padding
2519 @kindex --no-pad-sections
2520 @cindex output section padding
2521 Normally the assembler will pad the end of each output section up to its
2522 alignment boundary. But this can waste space, which can be significant on
2523 memory constrained targets. So the @option{--no-pad-sections} option will
2524 disable this behaviour.
2527 @section Name the Object File: @option{-o}
2530 @cindex naming object file
2531 @cindex object file name
2532 There is always one object file output when you run @command{@value{AS}}. By
2533 default it has the name
2536 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2550 You use this option (which takes exactly one filename) to give the
2551 object file a different name.
2553 Whatever the object file is called, @command{@value{AS}} overwrites any
2554 existing file of the same name.
2557 @section Join Data and Text Sections: @option{-R}
2560 @cindex data and text sections, joining
2561 @cindex text and data sections, joining
2562 @cindex joining text and data sections
2563 @cindex merging text and data sections
2564 @option{-R} tells @command{@value{AS}} to write the object file as if all
2565 data-section data lives in the text section. This is only done at
2566 the very last moment: your binary data are the same, but data
2567 section parts are relocated differently. The data section part of
2568 your object file is zero bytes long because all its bytes are
2569 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2571 When you specify @option{-R} it would be possible to generate shorter
2572 address displacements (because we do not have to cross between text and
2573 data section). We refrain from doing this simply for compatibility with
2574 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2577 When @command{@value{AS}} is configured for COFF or ELF output,
2578 this option is only useful if you use sections named @samp{.text} and
2583 @option{-R} is not supported for any of the HPPA targets. Using
2584 @option{-R} generates a warning from @command{@value{AS}}.
2588 @section Display Assembly Statistics: @option{--statistics}
2590 @kindex --statistics
2591 @cindex statistics, about assembly
2592 @cindex time, total for assembly
2593 @cindex space used, maximum for assembly
2594 Use @samp{--statistics} to display two statistics about the resources used by
2595 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2596 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2599 @node traditional-format
2600 @section Compatible Output: @option{--traditional-format}
2602 @kindex --traditional-format
2603 For some targets, the output of @command{@value{AS}} is different in some ways
2604 from the output of some existing assembler. This switch requests
2605 @command{@value{AS}} to use the traditional format instead.
2607 For example, it disables the exception frame optimizations which
2608 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2611 @section Announce Version: @option{-v}
2615 @cindex assembler version
2616 @cindex version of assembler
2617 You can find out what version of as is running by including the
2618 option @samp{-v} (which you can also spell as @samp{-version}) on the
2622 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2624 @command{@value{AS}} should never give a warning or error message when
2625 assembling compiler output. But programs written by people often
2626 cause @command{@value{AS}} to give a warning that a particular assumption was
2627 made. All such warnings are directed to the standard error file.
2631 @cindex suppressing warnings
2632 @cindex warnings, suppressing
2633 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2634 This only affects the warning messages: it does not change any particular of
2635 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2638 @kindex --fatal-warnings
2639 @cindex errors, caused by warnings
2640 @cindex warnings, causing error
2641 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2642 files that generate warnings to be in error.
2645 @cindex warnings, switching on
2646 You can switch these options off again by specifying @option{--warn}, which
2647 causes warnings to be output as usual.
2650 @section Generate Object File in Spite of Errors: @option{-Z}
2651 @cindex object file, after errors
2652 @cindex errors, continuing after
2653 After an error message, @command{@value{AS}} normally produces no output. If for
2654 some reason you are interested in object file output even after
2655 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2656 option. If there are any errors, @command{@value{AS}} continues anyways, and
2657 writes an object file after a final warning message of the form @samp{@var{n}
2658 errors, @var{m} warnings, generating bad object file.}
2663 @cindex machine-independent syntax
2664 @cindex syntax, machine-independent
2665 This chapter describes the machine-independent syntax allowed in a
2666 source file. @command{@value{AS}} syntax is similar to what many other
2667 assemblers use; it is inspired by the BSD 4.2
2672 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2676 * Preprocessing:: Preprocessing
2677 * Whitespace:: Whitespace
2678 * Comments:: Comments
2679 * Symbol Intro:: Symbols
2680 * Statements:: Statements
2681 * Constants:: Constants
2685 @section Preprocessing
2687 @cindex preprocessing
2688 The @command{@value{AS}} internal preprocessor:
2690 @cindex whitespace, removed by preprocessor
2692 adjusts and removes extra whitespace. It leaves one space or tab before
2693 the keywords on a line, and turns any other whitespace on the line into
2696 @cindex comments, removed by preprocessor
2698 removes all comments, replacing them with a single space, or an
2699 appropriate number of newlines.
2701 @cindex constants, converted by preprocessor
2703 converts character constants into the appropriate numeric values.
2706 It does not do macro processing, include file handling, or
2707 anything else you may get from your C compiler's preprocessor. You can
2708 do include file processing with the @code{.include} directive
2709 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2710 to get other ``CPP'' style preprocessing by giving the input file a
2711 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2712 Output, gcc info, Using GNU CC}.
2714 Excess whitespace, comments, and character constants
2715 cannot be used in the portions of the input text that are not
2718 @cindex turning preprocessing on and off
2719 @cindex preprocessing, turning on and off
2722 If the first line of an input file is @code{#NO_APP} or if you use the
2723 @samp{-f} option, whitespace and comments are not removed from the input file.
2724 Within an input file, you can ask for whitespace and comment removal in
2725 specific portions of the by putting a line that says @code{#APP} before the
2726 text that may contain whitespace or comments, and putting a line that says
2727 @code{#NO_APP} after this text. This feature is mainly intend to support
2728 @code{asm} statements in compilers whose output is otherwise free of comments
2735 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2736 Whitespace is used to separate symbols, and to make programs neater for
2737 people to read. Unless within character constants
2738 (@pxref{Characters,,Character Constants}), any whitespace means the same
2739 as exactly one space.
2745 There are two ways of rendering comments to @command{@value{AS}}. In both
2746 cases the comment is equivalent to one space.
2748 Anything from @samp{/*} through the next @samp{*/} is a comment.
2749 This means you may not nest these comments.
2753 The only way to include a newline ('\n') in a comment
2754 is to use this sort of comment.
2757 /* This sort of comment does not nest. */
2760 @cindex line comment character
2761 Anything from a @dfn{line comment} character up to the next newline is
2762 considered a comment and is ignored. The line comment character is target
2763 specific, and some targets multiple comment characters. Some targets also have
2764 line comment characters that only work if they are the first character on a
2765 line. Some targets use a sequence of two characters to introduce a line
2766 comment. Some targets can also change their line comment characters depending
2767 upon command line options that have been used. For more details see the
2768 @emph{Syntax} section in the documentation for individual targets.
2770 If the line comment character is the hash sign (@samp{#}) then it still has the
2771 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2772 to specify logical line numbers:
2775 @cindex lines starting with @code{#}
2776 @cindex logical line numbers
2777 To be compatible with past assemblers, lines that begin with @samp{#} have a
2778 special interpretation. Following the @samp{#} should be an absolute
2779 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2780 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2781 new logical file name. The rest of the line, if any, should be whitespace.
2783 If the first non-whitespace characters on the line are not numeric,
2784 the line is ignored. (Just like a comment.)
2787 # This is an ordinary comment.
2788 # 42-6 "new_file_name" # New logical file name
2789 # This is logical line # 36.
2791 This feature is deprecated, and may disappear from future versions
2792 of @command{@value{AS}}.
2797 @cindex characters used in symbols
2798 @ifclear SPECIAL-SYMS
2799 A @dfn{symbol} is one or more characters chosen from the set of all
2800 letters (both upper and lower case), digits and the three characters
2806 A @dfn{symbol} is one or more characters chosen from the set of all
2807 letters (both upper and lower case), digits and the three characters
2808 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2814 On most machines, you can also use @code{$} in symbol names; exceptions
2815 are noted in @ref{Machine Dependencies}.
2817 No symbol may begin with a digit. Case is significant.
2818 There is no length limit; all characters are significant. Multibyte characters
2819 are supported. Symbols are delimited by characters not in that set, or by the
2820 beginning of a file (since the source program must end with a newline, the end
2821 of a file is not a possible symbol delimiter). @xref{Symbols}.
2823 Symbol names may also be enclosed in double quote @code{"} characters. In such
2824 cases any characters are allowed, except for the NUL character. If a double
2825 quote character is to be included in the symbol name it must be preceeded by a
2826 backslash @code{\} character.
2827 @cindex length of symbols
2832 @cindex statements, structure of
2833 @cindex line separator character
2834 @cindex statement separator character
2836 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2837 @dfn{line separator character}. The line separator character is target
2838 specific and described in the @emph{Syntax} section of each
2839 target's documentation. Not all targets support a line separator character.
2840 The newline or line separator character is considered to be part of the
2841 preceding statement. Newlines and separators within character constants are an
2842 exception: they do not end statements.
2844 @cindex newline, required at file end
2845 @cindex EOF, newline must precede
2846 It is an error to end any statement with end-of-file: the last
2847 character of any input file should be a newline.@refill
2849 An empty statement is allowed, and may include whitespace. It is ignored.
2851 @cindex instructions and directives
2852 @cindex directives and instructions
2853 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2854 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2856 A statement begins with zero or more labels, optionally followed by a
2857 key symbol which determines what kind of statement it is. The key
2858 symbol determines the syntax of the rest of the statement. If the
2859 symbol begins with a dot @samp{.} then the statement is an assembler
2860 directive: typically valid for any computer. If the symbol begins with
2861 a letter the statement is an assembly language @dfn{instruction}: it
2862 assembles into a machine language instruction.
2864 Different versions of @command{@value{AS}} for different computers
2865 recognize different instructions. In fact, the same symbol may
2866 represent a different instruction in a different computer's assembly
2870 @cindex @code{:} (label)
2871 @cindex label (@code{:})
2872 A label is a symbol immediately followed by a colon (@code{:}).
2873 Whitespace before a label or after a colon is permitted, but you may not
2874 have whitespace between a label's symbol and its colon. @xref{Labels}.
2877 For HPPA targets, labels need not be immediately followed by a colon, but
2878 the definition of a label must begin in column zero. This also implies that
2879 only one label may be defined on each line.
2883 label: .directive followed by something
2884 another_label: # This is an empty statement.
2885 instruction operand_1, operand_2, @dots{}
2892 A constant is a number, written so that its value is known by
2893 inspection, without knowing any context. Like this:
2896 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2897 .ascii "Ring the bell\7" # A string constant.
2898 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2899 .float 0f-314159265358979323846264338327\
2900 95028841971.693993751E-40 # - pi, a flonum.
2905 * Characters:: Character Constants
2906 * Numbers:: Number Constants
2910 @subsection Character Constants
2912 @cindex character constants
2913 @cindex constants, character
2914 There are two kinds of character constants. A @dfn{character} stands
2915 for one character in one byte and its value may be used in
2916 numeric expressions. String constants (properly called string
2917 @emph{literals}) are potentially many bytes and their values may not be
2918 used in arithmetic expressions.
2922 * Chars:: Characters
2926 @subsubsection Strings
2928 @cindex string constants
2929 @cindex constants, string
2930 A @dfn{string} is written between double-quotes. It may contain
2931 double-quotes or null characters. The way to get special characters
2932 into a string is to @dfn{escape} these characters: precede them with
2933 a backslash @samp{\} character. For example @samp{\\} represents
2934 one backslash: the first @code{\} is an escape which tells
2935 @command{@value{AS}} to interpret the second character literally as a backslash
2936 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2937 escape character). The complete list of escapes follows.
2939 @cindex escape codes, character
2940 @cindex character escape codes
2941 @c NOTE: Cindex entries must not start with a backlash character.
2942 @c NOTE: This confuses the pdf2texi script when it is creating the
2943 @c NOTE: index based upon the first character and so it generates:
2944 @c NOTE: \initial {\\}
2945 @c NOTE: which then results in the error message:
2946 @c NOTE: Argument of \\ has an extra }.
2947 @c NOTE: So in the index entries below a space character has been
2948 @c NOTE: prepended to avoid this problem.
2951 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2953 @cindex @code{ \b} (backspace character)
2954 @cindex backspace (@code{\b})
2956 Mnemonic for backspace; for ASCII this is octal code 010.
2959 @c Mnemonic for EOText; for ASCII this is octal code 004.
2961 @cindex @code{ \f} (formfeed character)
2962 @cindex formfeed (@code{\f})
2964 Mnemonic for FormFeed; for ASCII this is octal code 014.
2966 @cindex @code{ \n} (newline character)
2967 @cindex newline (@code{\n})
2969 Mnemonic for newline; for ASCII this is octal code 012.
2972 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2974 @cindex @code{ \r} (carriage return character)
2975 @cindex carriage return (@code{backslash-r})
2977 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2980 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2981 @c other assemblers.
2983 @cindex @code{ \t} (tab)
2984 @cindex tab (@code{\t})
2986 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2989 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2990 @c @item \x @var{digit} @var{digit} @var{digit}
2991 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2993 @cindex @code{ \@var{ddd}} (octal character code)
2994 @cindex octal character code (@code{\@var{ddd}})
2995 @item \ @var{digit} @var{digit} @var{digit}
2996 An octal character code. The numeric code is 3 octal digits.
2997 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2998 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3000 @cindex @code{ \@var{xd...}} (hex character code)
3001 @cindex hex character code (@code{\@var{xd...}})
3002 @item \@code{x} @var{hex-digits...}
3003 A hex character code. All trailing hex digits are combined. Either upper or
3004 lower case @code{x} works.
3006 @cindex @code{ \\} (@samp{\} character)
3007 @cindex backslash (@code{\\})
3009 Represents one @samp{\} character.
3012 @c Represents one @samp{'} (accent acute) character.
3013 @c This is needed in single character literals
3014 @c (@xref{Characters,,Character Constants}.) to represent
3017 @cindex @code{ \"} (doublequote character)
3018 @cindex doublequote (@code{\"})
3020 Represents one @samp{"} character. Needed in strings to represent
3021 this character, because an unescaped @samp{"} would end the string.
3023 @item \ @var{anything-else}
3024 Any other character when escaped by @kbd{\} gives a warning, but
3025 assembles as if the @samp{\} was not present. The idea is that if
3026 you used an escape sequence you clearly didn't want the literal
3027 interpretation of the following character. However @command{@value{AS}} has no
3028 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3029 code and warns you of the fact.
3032 Which characters are escapable, and what those escapes represent,
3033 varies widely among assemblers. The current set is what we think
3034 the BSD 4.2 assembler recognizes, and is a subset of what most C
3035 compilers recognize. If you are in doubt, do not use an escape
3039 @subsubsection Characters
3041 @cindex single character constant
3042 @cindex character, single
3043 @cindex constant, single character
3044 A single character may be written as a single quote immediately
3045 followed by that character. The same escapes apply to characters as
3046 to strings. So if you want to write the character backslash, you
3047 must write @kbd{'\\} where the first @code{\} escapes the second
3048 @code{\}. As you can see, the quote is an acute accent, not a
3049 grave accent. A newline
3051 @ifclear abnormal-separator
3052 (or semicolon @samp{;})
3054 @ifset abnormal-separator
3056 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3061 immediately following an acute accent is taken as a literal character
3062 and does not count as the end of a statement. The value of a character
3063 constant in a numeric expression is the machine's byte-wide code for
3064 that character. @command{@value{AS}} assumes your character code is ASCII:
3065 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3068 @subsection Number Constants
3070 @cindex constants, number
3071 @cindex number constants
3072 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3073 are stored in the target machine. @emph{Integers} are numbers that
3074 would fit into an @code{int} in the C language. @emph{Bignums} are
3075 integers, but they are stored in more than 32 bits. @emph{Flonums}
3076 are floating point numbers, described below.
3079 * Integers:: Integers
3084 * Bit Fields:: Bit Fields
3090 @subsubsection Integers
3092 @cindex constants, integer
3094 @cindex binary integers
3095 @cindex integers, binary
3096 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3097 the binary digits @samp{01}.
3099 @cindex octal integers
3100 @cindex integers, octal
3101 An octal integer is @samp{0} followed by zero or more of the octal
3102 digits (@samp{01234567}).
3104 @cindex decimal integers
3105 @cindex integers, decimal
3106 A decimal integer starts with a non-zero digit followed by zero or
3107 more digits (@samp{0123456789}).
3109 @cindex hexadecimal integers
3110 @cindex integers, hexadecimal
3111 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3112 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3114 Integers have the usual values. To denote a negative integer, use
3115 the prefix operator @samp{-} discussed under expressions
3116 (@pxref{Prefix Ops,,Prefix Operators}).
3119 @subsubsection Bignums
3122 @cindex constants, bignum
3123 A @dfn{bignum} has the same syntax and semantics as an integer
3124 except that the number (or its negative) takes more than 32 bits to
3125 represent in binary. The distinction is made because in some places
3126 integers are permitted while bignums are not.
3129 @subsubsection Flonums
3131 @cindex floating point numbers
3132 @cindex constants, floating point
3134 @cindex precision, floating point
3135 A @dfn{flonum} represents a floating point number. The translation is
3136 indirect: a decimal floating point number from the text is converted by
3137 @command{@value{AS}} to a generic binary floating point number of more than
3138 sufficient precision. This generic floating point number is converted
3139 to a particular computer's floating point format (or formats) by a
3140 portion of @command{@value{AS}} specialized to that computer.
3142 A flonum is written by writing (in order)
3147 (@samp{0} is optional on the HPPA.)
3151 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3153 @kbd{e} is recommended. Case is not important.
3155 @c FIXME: verify if flonum syntax really this vague for most cases
3156 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3157 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3160 On the H8/300, Renesas / SuperH SH,
3161 and AMD 29K architectures, the letter must be
3162 one of the letters @samp{DFPRSX} (in upper or lower case).
3164 On the ARC, the letter must be one of the letters @samp{DFRS}
3165 (in upper or lower case).
3167 On the Intel 960 architecture, the letter must be
3168 one of the letters @samp{DFT} (in upper or lower case).
3170 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3174 One of the letters @samp{DFRS} (in upper or lower case).
3177 One of the letters @samp{DFPRSX} (in upper or lower case).
3180 The letter @samp{E} (upper case only).
3183 One of the letters @samp{DFT} (in upper or lower case).
3188 An optional sign: either @samp{+} or @samp{-}.
3191 An optional @dfn{integer part}: zero or more decimal digits.
3194 An optional @dfn{fractional part}: @samp{.} followed by zero
3195 or more decimal digits.
3198 An optional exponent, consisting of:
3202 An @samp{E} or @samp{e}.
3203 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3204 @c principle this can perfectly well be different on different targets.
3206 Optional sign: either @samp{+} or @samp{-}.
3208 One or more decimal digits.
3213 At least one of the integer part or the fractional part must be
3214 present. The floating point number has the usual base-10 value.
3216 @command{@value{AS}} does all processing using integers. Flonums are computed
3217 independently of any floating point hardware in the computer running
3218 @command{@value{AS}}.
3222 @c Bit fields are written as a general facility but are also controlled
3223 @c by a conditional-compilation flag---which is as of now (21mar91)
3224 @c turned on only by the i960 config of GAS.
3226 @subsubsection Bit Fields
3229 @cindex constants, bit field
3230 You can also define numeric constants as @dfn{bit fields}.
3231 Specify two numbers separated by a colon---
3233 @var{mask}:@var{value}
3236 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
3239 The resulting number is then packed
3241 @c this conditional paren in case bit fields turned on elsewhere than 960
3242 (in host-dependent byte order)
3244 into a field whose width depends on which assembler directive has the
3245 bit-field as its argument. Overflow (a result from the bitwise and
3246 requiring more binary digits to represent) is not an error; instead,
3247 more constants are generated, of the specified width, beginning with the
3248 least significant digits.@refill
3250 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
3251 @code{.short}, and @code{.word} accept bit-field arguments.
3256 @chapter Sections and Relocation
3261 * Secs Background:: Background
3262 * Ld Sections:: Linker Sections
3263 * As Sections:: Assembler Internal Sections
3264 * Sub-Sections:: Sub-Sections
3268 @node Secs Background
3271 Roughly, a section is a range of addresses, with no gaps; all data
3272 ``in'' those addresses is treated the same for some particular purpose.
3273 For example there may be a ``read only'' section.
3275 @cindex linker, and assembler
3276 @cindex assembler, and linker
3277 The linker @code{@value{LD}} reads many object files (partial programs) and
3278 combines their contents to form a runnable program. When @command{@value{AS}}
3279 emits an object file, the partial program is assumed to start at address 0.
3280 @code{@value{LD}} assigns the final addresses for the partial program, so that
3281 different partial programs do not overlap. This is actually an
3282 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3285 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3286 addresses. These blocks slide to their run-time addresses as rigid
3287 units; their length does not change and neither does the order of bytes
3288 within them. Such a rigid unit is called a @emph{section}. Assigning
3289 run-time addresses to sections is called @dfn{relocation}. It includes
3290 the task of adjusting mentions of object-file addresses so they refer to
3291 the proper run-time addresses.
3293 For the H8/300, and for the Renesas / SuperH SH,
3294 @command{@value{AS}} pads sections if needed to
3295 ensure they end on a word (sixteen bit) boundary.
3298 @cindex standard assembler sections
3299 An object file written by @command{@value{AS}} has at least three sections, any
3300 of which may be empty. These are named @dfn{text}, @dfn{data} and
3305 When it generates COFF or ELF output,
3307 @command{@value{AS}} can also generate whatever other named sections you specify
3308 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3309 If you do not use any directives that place output in the @samp{.text}
3310 or @samp{.data} sections, these sections still exist, but are empty.
3315 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3317 @command{@value{AS}} can also generate whatever other named sections you
3318 specify using the @samp{.space} and @samp{.subspace} directives. See
3319 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3320 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3321 assembler directives.
3324 Additionally, @command{@value{AS}} uses different names for the standard
3325 text, data, and bss sections when generating SOM output. Program text
3326 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3327 BSS into @samp{$BSS$}.
3331 Within the object file, the text section starts at address @code{0}, the
3332 data section follows, and the bss section follows the data section.
3335 When generating either SOM or ELF output files on the HPPA, the text
3336 section starts at address @code{0}, the data section at address
3337 @code{0x4000000}, and the bss section follows the data section.
3340 To let @code{@value{LD}} know which data changes when the sections are
3341 relocated, and how to change that data, @command{@value{AS}} also writes to the
3342 object file details of the relocation needed. To perform relocation
3343 @code{@value{LD}} must know, each time an address in the object
3347 Where in the object file is the beginning of this reference to
3350 How long (in bytes) is this reference?
3352 Which section does the address refer to? What is the numeric value of
3354 (@var{address}) @minus{} (@var{start-address of section})?
3357 Is the reference to an address ``Program-Counter relative''?
3360 @cindex addresses, format of
3361 @cindex section-relative addressing
3362 In fact, every address @command{@value{AS}} ever uses is expressed as
3364 (@var{section}) + (@var{offset into section})
3367 Further, most expressions @command{@value{AS}} computes have this section-relative
3370 (For some object formats, such as SOM for the HPPA, some expressions are
3371 symbol-relative instead.)
3374 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3375 @var{N} into section @var{secname}.''
3377 Apart from text, data and bss sections you need to know about the
3378 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3379 addresses in the absolute section remain unchanged. For example, address
3380 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3381 @code{@value{LD}}. Although the linker never arranges two partial programs'
3382 data sections with overlapping addresses after linking, @emph{by definition}
3383 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3384 part of a program is always the same address when the program is running as
3385 address @code{@{absolute@ 239@}} in any other part of the program.
3387 The idea of sections is extended to the @dfn{undefined} section. Any
3388 address whose section is unknown at assembly time is by definition
3389 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3390 Since numbers are always defined, the only way to generate an undefined
3391 address is to mention an undefined symbol. A reference to a named
3392 common block would be such a symbol: its value is unknown at assembly
3393 time so it has section @emph{undefined}.
3395 By analogy the word @emph{section} is used to describe groups of sections in
3396 the linked program. @code{@value{LD}} puts all partial programs' text
3397 sections in contiguous addresses in the linked program. It is
3398 customary to refer to the @emph{text section} of a program, meaning all
3399 the addresses of all partial programs' text sections. Likewise for
3400 data and bss sections.
3402 Some sections are manipulated by @code{@value{LD}}; others are invented for
3403 use of @command{@value{AS}} and have no meaning except during assembly.
3406 @section Linker Sections
3407 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3412 @cindex named sections
3413 @cindex sections, named
3414 @item named sections
3417 @cindex text section
3418 @cindex data section
3422 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3423 separate but equal sections. Anything you can say of one section is
3426 When the program is running, however, it is
3427 customary for the text section to be unalterable. The
3428 text section is often shared among processes: it contains
3429 instructions, constants and the like. The data section of a running
3430 program is usually alterable: for example, C variables would be stored
3431 in the data section.
3436 This section contains zeroed bytes when your program begins running. It
3437 is used to hold uninitialized variables or common storage. The length of
3438 each partial program's bss section is important, but because it starts
3439 out containing zeroed bytes there is no need to store explicit zero
3440 bytes in the object file. The bss section was invented to eliminate
3441 those explicit zeros from object files.
3443 @cindex absolute section
3444 @item absolute section
3445 Address 0 of this section is always ``relocated'' to runtime address 0.
3446 This is useful if you want to refer to an address that @code{@value{LD}} must
3447 not change when relocating. In this sense we speak of absolute
3448 addresses being ``unrelocatable'': they do not change during relocation.
3450 @cindex undefined section
3451 @item undefined section
3452 This ``section'' is a catch-all for address references to objects not in
3453 the preceding sections.
3454 @c FIXME: ref to some other doc on obj-file formats could go here.
3457 @cindex relocation example
3458 An idealized example of three relocatable sections follows.
3460 The example uses the traditional section names @samp{.text} and @samp{.data}.
3462 Memory addresses are on the horizontal axis.
3466 @c END TEXI2ROFF-KILL
3469 partial program # 1: |ttttt|dddd|00|
3476 partial program # 2: |TTT|DDD|000|
3479 +--+---+-----+--+----+---+-----+~~
3480 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3481 +--+---+-----+--+----+---+-----+~~
3483 addresses: 0 @dots{}
3490 \line{\it Partial program \#1: \hfil}
3491 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3492 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3494 \line{\it Partial program \#2: \hfil}
3495 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3496 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3498 \line{\it linked program: \hfil}
3499 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3500 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3501 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3502 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3504 \line{\it addresses: \hfil}
3508 @c END TEXI2ROFF-KILL
3511 @section Assembler Internal Sections
3513 @cindex internal assembler sections
3514 @cindex sections in messages, internal
3515 These sections are meant only for the internal use of @command{@value{AS}}. They
3516 have no meaning at run-time. You do not really need to know about these
3517 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3518 warning messages, so it might be helpful to have an idea of their
3519 meanings to @command{@value{AS}}. These sections are used to permit the
3520 value of every expression in your assembly language program to be a
3521 section-relative address.
3524 @cindex assembler internal logic error
3525 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3526 An internal assembler logic error has been found. This means there is a
3527 bug in the assembler.
3529 @cindex expr (internal section)
3531 The assembler stores complex expression internally as combinations of
3532 symbols. When it needs to represent an expression as a symbol, it puts
3533 it in the expr section.
3535 @c FIXME item transfer[t] vector preload
3536 @c FIXME item transfer[t] vector postload
3537 @c FIXME item register
3541 @section Sub-Sections
3543 @cindex numbered subsections
3544 @cindex grouping data
3550 fall into two sections: text and data.
3552 You may have separate groups of
3554 data in named sections
3558 data in named sections
3564 that you want to end up near to each other in the object file, even though they
3565 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3566 use @dfn{subsections} for this purpose. Within each section, there can be
3567 numbered subsections with values from 0 to 8192. Objects assembled into the
3568 same subsection go into the object file together with other objects in the same
3569 subsection. For example, a compiler might want to store constants in the text
3570 section, but might not want to have them interspersed with the program being
3571 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3572 section of code being output, and a @samp{.text 1} before each group of
3573 constants being output.
3575 Subsections are optional. If you do not use subsections, everything
3576 goes in subsection number zero.
3579 Each subsection is zero-padded up to a multiple of four bytes.
3580 (Subsections may be padded a different amount on different flavors
3581 of @command{@value{AS}}.)
3585 On the H8/300 platform, each subsection is zero-padded to a word
3586 boundary (two bytes).
3587 The same is true on the Renesas SH.
3590 @c FIXME section padding (alignment)?
3591 @c Rich Pixley says padding here depends on target obj code format; that
3592 @c doesn't seem particularly useful to say without further elaboration,
3593 @c so for now I say nothing about it. If this is a generic BFD issue,
3594 @c these paragraphs might need to vanish from this manual, and be
3595 @c discussed in BFD chapter of binutils (or some such).
3599 Subsections appear in your object file in numeric order, lowest numbered
3600 to highest. (All this to be compatible with other people's assemblers.)
3601 The object file contains no representation of subsections; @code{@value{LD}} and
3602 other programs that manipulate object files see no trace of them.
3603 They just see all your text subsections as a text section, and all your
3604 data subsections as a data section.
3606 To specify which subsection you want subsequent statements assembled
3607 into, use a numeric argument to specify it, in a @samp{.text
3608 @var{expression}} or a @samp{.data @var{expression}} statement.
3611 When generating COFF output, you
3616 can also use an extra subsection
3617 argument with arbitrary named sections: @samp{.section @var{name},
3622 When generating ELF output, you
3627 can also use the @code{.subsection} directive (@pxref{SubSection})
3628 to specify a subsection: @samp{.subsection @var{expression}}.
3630 @var{Expression} should be an absolute expression
3631 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3632 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3633 begins in @code{text 0}. For instance:
3635 .text 0 # The default subsection is text 0 anyway.
3636 .ascii "This lives in the first text subsection. *"
3638 .ascii "But this lives in the second text subsection."
3640 .ascii "This lives in the data section,"
3641 .ascii "in the first data subsection."
3643 .ascii "This lives in the first text section,"
3644 .ascii "immediately following the asterisk (*)."
3647 Each section has a @dfn{location counter} incremented by one for every byte
3648 assembled into that section. Because subsections are merely a convenience
3649 restricted to @command{@value{AS}} there is no concept of a subsection location
3650 counter. There is no way to directly manipulate a location counter---but the
3651 @code{.align} directive changes it, and any label definition captures its
3652 current value. The location counter of the section where statements are being
3653 assembled is said to be the @dfn{active} location counter.
3656 @section bss Section
3659 @cindex common variable storage
3660 The bss section is used for local common variable storage.
3661 You may allocate address space in the bss section, but you may
3662 not dictate data to load into it before your program executes. When
3663 your program starts running, all the contents of the bss
3664 section are zeroed bytes.
3666 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3667 @ref{Lcomm,,@code{.lcomm}}.
3669 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3670 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3673 When assembling for a target which supports multiple sections, such as ELF or
3674 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3675 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3676 section. Typically the section will only contain symbol definitions and
3677 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3684 Symbols are a central concept: the programmer uses symbols to name
3685 things, the linker uses symbols to link, and the debugger uses symbols
3689 @cindex debuggers, and symbol order
3690 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3691 the same order they were declared. This may break some debuggers.
3696 * Setting Symbols:: Giving Symbols Other Values
3697 * Symbol Names:: Symbol Names
3698 * Dot:: The Special Dot Symbol
3699 * Symbol Attributes:: Symbol Attributes
3706 A @dfn{label} is written as a symbol immediately followed by a colon
3707 @samp{:}. The symbol then represents the current value of the
3708 active location counter, and is, for example, a suitable instruction
3709 operand. You are warned if you use the same symbol to represent two
3710 different locations: the first definition overrides any other
3714 On the HPPA, the usual form for a label need not be immediately followed by a
3715 colon, but instead must start in column zero. Only one label may be defined on
3716 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3717 provides a special directive @code{.label} for defining labels more flexibly.
3720 @node Setting Symbols
3721 @section Giving Symbols Other Values
3723 @cindex assigning values to symbols
3724 @cindex symbol values, assigning
3725 A symbol can be given an arbitrary value by writing a symbol, followed
3726 by an equals sign @samp{=}, followed by an expression
3727 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3728 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3729 equals sign @samp{=}@samp{=} here represents an equivalent of the
3730 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3733 Blackfin does not support symbol assignment with @samp{=}.
3737 @section Symbol Names
3739 @cindex symbol names
3740 @cindex names, symbol
3741 @ifclear SPECIAL-SYMS
3742 Symbol names begin with a letter or with one of @samp{._}. On most
3743 machines, you can also use @code{$} in symbol names; exceptions are
3744 noted in @ref{Machine Dependencies}. That character may be followed by any
3745 string of digits, letters, dollar signs (unless otherwise noted for a
3746 particular target machine), and underscores.
3750 Symbol names begin with a letter or with one of @samp{._}. On the
3751 Renesas SH you can also use @code{$} in symbol names. That
3752 character may be followed by any string of digits, letters, dollar signs (save
3753 on the H8/300), and underscores.
3757 Case of letters is significant: @code{foo} is a different symbol name
3760 Symbol names do not start with a digit. An exception to this rule is made for
3761 Local Labels. See below.
3763 Multibyte characters are supported. To generate a symbol name containing
3764 multibyte characters enclose it within double quotes and use escape codes. cf
3765 @xref{Strings}. Generating a multibyte symbol name from a label is not
3766 currently supported.
3768 Each symbol has exactly one name. Each name in an assembly language program
3769 refers to exactly one symbol. You may use that symbol name any number of times
3772 @subheading Local Symbol Names
3774 @cindex local symbol names
3775 @cindex symbol names, local
3776 A local symbol is any symbol beginning with certain local label prefixes.
3777 By default, the local label prefix is @samp{.L} for ELF systems or
3778 @samp{L} for traditional a.out systems, but each target may have its own
3779 set of local label prefixes.
3781 On the HPPA local symbols begin with @samp{L$}.
3784 Local symbols are defined and used within the assembler, but they are
3785 normally not saved in object files. Thus, they are not visible when debugging.
3786 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3787 to retain the local symbols in the object files.
3789 @subheading Local Labels
3791 @cindex local labels
3792 @cindex temporary symbol names
3793 @cindex symbol names, temporary
3794 Local labels are different from local symbols. Local labels help compilers and
3795 programmers use names temporarily. They create symbols which are guaranteed to
3796 be unique over the entire scope of the input source code and which can be
3797 referred to by a simple notation. To define a local label, write a label of
3798 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3799 To refer to the most recent previous definition of that label write
3800 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3801 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3802 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3804 There is no restriction on how you can use these labels, and you can reuse them
3805 too. So that it is possible to repeatedly define the same local label (using
3806 the same number @samp{@b{N}}), although you can only refer to the most recently
3807 defined local label of that number (for a backwards reference) or the next
3808 definition of a specific local label for a forward reference. It is also worth
3809 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3810 implemented in a slightly more efficient manner than the others.
3821 Which is the equivalent of:
3824 label_1: branch label_3
3825 label_2: branch label_1
3826 label_3: branch label_4
3827 label_4: branch label_3
3830 Local label names are only a notational device. They are immediately
3831 transformed into more conventional symbol names before the assembler uses them.
3832 The symbol names are stored in the symbol table, appear in error messages, and
3833 are optionally emitted to the object file. The names are constructed using
3837 @item @emph{local label prefix}
3838 All local symbols begin with the system-specific local label prefix.
3839 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3840 that start with the local label prefix. These labels are
3841 used for symbols you are never intended to see. If you use the
3842 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3843 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3844 you may use them in debugging.
3847 This is the number that was used in the local label definition. So if the
3848 label is written @samp{55:} then the number is @samp{55}.
3851 This unusual character is included so you do not accidentally invent a symbol
3852 of the same name. The character has ASCII value of @samp{\002} (control-B).
3854 @item @emph{ordinal number}
3855 This is a serial number to keep the labels distinct. The first definition of
3856 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3857 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3858 the number @samp{1} and its 15th definition gets @samp{15} as well.
3861 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3862 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3864 @subheading Dollar Local Labels
3865 @cindex dollar local symbols
3867 On some targets @code{@value{AS}} also supports an even more local form of
3868 local labels called dollar labels. These labels go out of scope (i.e., they
3869 become undefined) as soon as a non-local label is defined. Thus they remain
3870 valid for only a small region of the input source code. Normal local labels,
3871 by contrast, remain in scope for the entire file, or until they are redefined
3872 by another occurrence of the same local label.
3874 Dollar labels are defined in exactly the same way as ordinary local labels,
3875 except that they have a dollar sign suffix to their numeric value, e.g.,
3878 They can also be distinguished from ordinary local labels by their transformed
3879 names which use ASCII character @samp{\001} (control-A) as the magic character
3880 to distinguish them from ordinary labels. For example, the fifth definition of
3881 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3884 @section The Special Dot Symbol
3886 @cindex dot (symbol)
3887 @cindex @code{.} (symbol)
3888 @cindex current address
3889 @cindex location counter
3890 The special symbol @samp{.} refers to the current address that
3891 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3892 .long .} defines @code{melvin} to contain its own address.
3893 Assigning a value to @code{.} is treated the same as a @code{.org}
3895 @ifclear no-space-dir
3896 Thus, the expression @samp{.=.+4} is the same as saying
3900 @node Symbol Attributes
3901 @section Symbol Attributes
3903 @cindex symbol attributes
3904 @cindex attributes, symbol
3905 Every symbol has, as well as its name, the attributes ``Value'' and
3906 ``Type''. Depending on output format, symbols can also have auxiliary
3909 The detailed definitions are in @file{a.out.h}.
3912 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3913 all these attributes, and probably won't warn you. This makes the
3914 symbol an externally defined symbol, which is generally what you
3918 * Symbol Value:: Value
3919 * Symbol Type:: Type
3922 * a.out Symbols:: Symbol Attributes: @code{a.out}
3926 * a.out Symbols:: Symbol Attributes: @code{a.out}
3929 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3934 * COFF Symbols:: Symbol Attributes for COFF
3937 * SOM Symbols:: Symbol Attributes for SOM
3944 @cindex value of a symbol
3945 @cindex symbol value
3946 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3947 location in the text, data, bss or absolute sections the value is the
3948 number of addresses from the start of that section to the label.
3949 Naturally for text, data and bss sections the value of a symbol changes
3950 as @code{@value{LD}} changes section base addresses during linking. Absolute
3951 symbols' values do not change during linking: that is why they are
3954 The value of an undefined symbol is treated in a special way. If it is
3955 0 then the symbol is not defined in this assembler source file, and
3956 @code{@value{LD}} tries to determine its value from other files linked into the
3957 same program. You make this kind of symbol simply by mentioning a symbol
3958 name without defining it. A non-zero value represents a @code{.comm}
3959 common declaration. The value is how much common storage to reserve, in
3960 bytes (addresses). The symbol refers to the first address of the
3966 @cindex type of a symbol
3968 The type attribute of a symbol contains relocation (section)
3969 information, any flag settings indicating that a symbol is external, and
3970 (optionally), other information for linkers and debuggers. The exact
3971 format depends on the object-code output format in use.
3976 @c The following avoids a "widow" subsection title. @group would be
3977 @c better if it were available outside examples.
3980 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
3982 @cindex @code{b.out} symbol attributes
3983 @cindex symbol attributes, @code{b.out}
3984 These symbol attributes appear only when @command{@value{AS}} is configured for
3985 one of the Berkeley-descended object output formats---@code{a.out} or
3991 @subsection Symbol Attributes: @code{a.out}
3993 @cindex @code{a.out} symbol attributes
3994 @cindex symbol attributes, @code{a.out}
4000 @subsection Symbol Attributes: @code{a.out}
4002 @cindex @code{a.out} symbol attributes
4003 @cindex symbol attributes, @code{a.out}
4007 * Symbol Desc:: Descriptor
4008 * Symbol Other:: Other
4012 @subsubsection Descriptor
4014 @cindex descriptor, of @code{a.out} symbol
4015 This is an arbitrary 16-bit value. You may establish a symbol's
4016 descriptor value by using a @code{.desc} statement
4017 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
4018 @command{@value{AS}}.
4021 @subsubsection Other
4023 @cindex other attribute, of @code{a.out} symbol
4024 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
4029 @subsection Symbol Attributes for COFF
4031 @cindex COFF symbol attributes
4032 @cindex symbol attributes, COFF
4034 The COFF format supports a multitude of auxiliary symbol attributes;
4035 like the primary symbol attributes, they are set between @code{.def} and
4036 @code{.endef} directives.
4038 @subsubsection Primary Attributes
4040 @cindex primary attributes, COFF symbols
4041 The symbol name is set with @code{.def}; the value and type,
4042 respectively, with @code{.val} and @code{.type}.
4044 @subsubsection Auxiliary Attributes
4046 @cindex auxiliary attributes, COFF symbols
4047 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
4048 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
4049 table information for COFF.
4054 @subsection Symbol Attributes for SOM
4056 @cindex SOM symbol attributes
4057 @cindex symbol attributes, SOM
4059 The SOM format for the HPPA supports a multitude of symbol attributes set with
4060 the @code{.EXPORT} and @code{.IMPORT} directives.
4062 The attributes are described in @cite{HP9000 Series 800 Assembly
4063 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
4064 @code{EXPORT} assembler directive documentation.
4068 @chapter Expressions
4072 @cindex numeric values
4073 An @dfn{expression} specifies an address or numeric value.
4074 Whitespace may precede and/or follow an expression.
4076 The result of an expression must be an absolute number, or else an offset into
4077 a particular section. If an expression is not absolute, and there is not
4078 enough information when @command{@value{AS}} sees the expression to know its
4079 section, a second pass over the source program might be necessary to interpret
4080 the expression---but the second pass is currently not implemented.
4081 @command{@value{AS}} aborts with an error message in this situation.
4084 * Empty Exprs:: Empty Expressions
4085 * Integer Exprs:: Integer Expressions
4089 @section Empty Expressions
4091 @cindex empty expressions
4092 @cindex expressions, empty
4093 An empty expression has no value: it is just whitespace or null.
4094 Wherever an absolute expression is required, you may omit the
4095 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4096 is compatible with other assemblers.
4099 @section Integer Expressions
4101 @cindex integer expressions
4102 @cindex expressions, integer
4103 An @dfn{integer expression} is one or more @emph{arguments} delimited
4104 by @emph{operators}.
4107 * Arguments:: Arguments
4108 * Operators:: Operators
4109 * Prefix Ops:: Prefix Operators
4110 * Infix Ops:: Infix Operators
4114 @subsection Arguments
4116 @cindex expression arguments
4117 @cindex arguments in expressions
4118 @cindex operands in expressions
4119 @cindex arithmetic operands
4120 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4121 contexts arguments are sometimes called ``arithmetic operands''. In
4122 this manual, to avoid confusing them with the ``instruction operands'' of
4123 the machine language, we use the term ``argument'' to refer to parts of
4124 expressions only, reserving the word ``operand'' to refer only to machine
4125 instruction operands.
4127 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4128 @var{section} is one of text, data, bss, absolute,
4129 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4132 Numbers are usually integers.
4134 A number can be a flonum or bignum. In this case, you are warned
4135 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4136 these 32 bits are an integer. You may write integer-manipulating
4137 instructions that act on exotic constants, compatible with other
4140 @cindex subexpressions
4141 Subexpressions are a left parenthesis @samp{(} followed by an integer
4142 expression, followed by a right parenthesis @samp{)}; or a prefix
4143 operator followed by an argument.
4146 @subsection Operators
4148 @cindex operators, in expressions
4149 @cindex arithmetic functions
4150 @cindex functions, in expressions
4151 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4152 operators are followed by an argument. Infix operators appear
4153 between their arguments. Operators may be preceded and/or followed by
4157 @subsection Prefix Operator
4159 @cindex prefix operators
4160 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4161 one argument, which must be absolute.
4163 @c the tex/end tex stuff surrounding this small table is meant to make
4164 @c it align, on the printed page, with the similar table in the next
4165 @c section (which is inside an enumerate).
4167 \global\advance\leftskip by \itemindent
4172 @dfn{Negation}. Two's complement negation.
4174 @dfn{Complementation}. Bitwise not.
4178 \global\advance\leftskip by -\itemindent
4182 @subsection Infix Operators
4184 @cindex infix operators
4185 @cindex operators, permitted arguments
4186 @dfn{Infix operators} take two arguments, one on either side. Operators
4187 have precedence, but operations with equal precedence are performed left
4188 to right. Apart from @code{+} or @option{-}, both arguments must be
4189 absolute, and the result is absolute.
4192 @cindex operator precedence
4193 @cindex precedence of operators
4200 @dfn{Multiplication}.
4203 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4209 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4212 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4216 Intermediate precedence
4221 @dfn{Bitwise Inclusive Or}.
4227 @dfn{Bitwise Exclusive Or}.
4230 @dfn{Bitwise Or Not}.
4237 @cindex addition, permitted arguments
4238 @cindex plus, permitted arguments
4239 @cindex arguments for addition
4241 @dfn{Addition}. If either argument is absolute, the result has the section of
4242 the other argument. You may not add together arguments from different
4245 @cindex subtraction, permitted arguments
4246 @cindex minus, permitted arguments
4247 @cindex arguments for subtraction
4249 @dfn{Subtraction}. If the right argument is absolute, the
4250 result has the section of the left argument.
4251 If both arguments are in the same section, the result is absolute.
4252 You may not subtract arguments from different sections.
4253 @c FIXME is there still something useful to say about undefined - undefined ?
4255 @cindex comparison expressions
4256 @cindex expressions, comparison
4261 @dfn{Is Not Equal To}
4265 @dfn{Is Greater Than}
4267 @dfn{Is Greater Than Or Equal To}
4269 @dfn{Is Less Than Or Equal To}
4271 The comparison operators can be used as infix operators. A true results has a
4272 value of -1 whereas a false result has a value of 0. Note, these operators
4273 perform signed comparisons.
4276 @item Lowest Precedence
4285 These two logical operations can be used to combine the results of sub
4286 expressions. Note, unlike the comparison operators a true result returns a
4287 value of 1 but a false results does still return 0. Also note that the logical
4288 or operator has a slightly lower precedence than logical and.
4293 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4294 address; you can only have a defined section in one of the two arguments.
4297 @chapter Assembler Directives
4299 @cindex directives, machine independent
4300 @cindex pseudo-ops, machine independent
4301 @cindex machine independent directives
4302 All assembler directives have names that begin with a period (@samp{.}).
4303 The names are case insensitive for most targets, and usually written
4306 This chapter discusses directives that are available regardless of the
4307 target machine configuration for the @sc{gnu} assembler.
4309 Some machine configurations provide additional directives.
4310 @xref{Machine Dependencies}.
4313 @ifset machine-directives
4314 @xref{Machine Dependencies}, for additional directives.
4319 * Abort:: @code{.abort}
4321 * ABORT (COFF):: @code{.ABORT}
4324 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4325 * Altmacro:: @code{.altmacro}
4326 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4327 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4328 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4329 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4330 * Byte:: @code{.byte @var{expressions}}
4331 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4332 * Comm:: @code{.comm @var{symbol} , @var{length} }
4333 * Data:: @code{.data @var{subsection}}
4335 * Def:: @code{.def @var{name}}
4338 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4344 * Double:: @code{.double @var{flonums}}
4345 * Eject:: @code{.eject}
4346 * Else:: @code{.else}
4347 * Elseif:: @code{.elseif}
4350 * Endef:: @code{.endef}
4353 * Endfunc:: @code{.endfunc}
4354 * Endif:: @code{.endif}
4355 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4356 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4357 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4359 * Error:: @code{.error @var{string}}
4360 * Exitm:: @code{.exitm}
4361 * Extern:: @code{.extern}
4362 * Fail:: @code{.fail}
4363 * File:: @code{.file}
4364 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4365 * Float:: @code{.float @var{flonums}}
4366 * Func:: @code{.func}
4367 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4369 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4370 * Hidden:: @code{.hidden @var{names}}
4373 * hword:: @code{.hword @var{expressions}}
4374 * Ident:: @code{.ident}
4375 * If:: @code{.if @var{absolute expression}}
4376 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4377 * Include:: @code{.include "@var{file}"}
4378 * Int:: @code{.int @var{expressions}}
4380 * Internal:: @code{.internal @var{names}}
4383 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4384 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4385 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4386 * Lflags:: @code{.lflags}
4387 @ifclear no-line-dir
4388 * Line:: @code{.line @var{line-number}}
4391 * Linkonce:: @code{.linkonce [@var{type}]}
4392 * List:: @code{.list}
4393 * Ln:: @code{.ln @var{line-number}}
4394 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4395 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4397 * Local:: @code{.local @var{names}}
4400 * Long:: @code{.long @var{expressions}}
4402 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4405 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4406 * MRI:: @code{.mri @var{val}}
4407 * Noaltmacro:: @code{.noaltmacro}
4408 * Nolist:: @code{.nolist}
4409 * Octa:: @code{.octa @var{bignums}}
4410 * Offset:: @code{.offset @var{loc}}
4411 * Org:: @code{.org @var{new-lc}, @var{fill}}
4412 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4414 * PopSection:: @code{.popsection}
4415 * Previous:: @code{.previous}
4418 * Print:: @code{.print @var{string}}
4420 * Protected:: @code{.protected @var{names}}
4423 * Psize:: @code{.psize @var{lines}, @var{columns}}
4424 * Purgem:: @code{.purgem @var{name}}
4426 * PushSection:: @code{.pushsection @var{name}}
4429 * Quad:: @code{.quad @var{bignums}}
4430 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4431 * Rept:: @code{.rept @var{count}}
4432 * Sbttl:: @code{.sbttl "@var{subheading}"}
4434 * Scl:: @code{.scl @var{class}}
4437 * Section:: @code{.section @var{name}[, @var{flags}]}
4440 * Set:: @code{.set @var{symbol}, @var{expression}}
4441 * Short:: @code{.short @var{expressions}}
4442 * Single:: @code{.single @var{flonums}}
4444 * Size:: @code{.size [@var{name} , @var{expression}]}
4446 @ifclear no-space-dir
4447 * Skip:: @code{.skip @var{size} , @var{fill}}
4450 * Sleb128:: @code{.sleb128 @var{expressions}}
4451 @ifclear no-space-dir
4452 * Space:: @code{.space @var{size} , @var{fill}}
4455 * Stab:: @code{.stabd, .stabn, .stabs}
4458 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4459 * Struct:: @code{.struct @var{expression}}
4461 * SubSection:: @code{.subsection}
4462 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4466 * Tag:: @code{.tag @var{structname}}
4469 * Text:: @code{.text @var{subsection}}
4470 * Title:: @code{.title "@var{heading}"}
4472 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4475 * Uleb128:: @code{.uleb128 @var{expressions}}
4477 * Val:: @code{.val @var{addr}}
4481 * Version:: @code{.version "@var{string}"}
4482 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4483 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4486 * Warning:: @code{.warning @var{string}}
4487 * Weak:: @code{.weak @var{names}}
4488 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4489 * Word:: @code{.word @var{expressions}}
4490 @ifclear no-space-dir
4491 * Zero:: @code{.zero @var{size}}
4493 * Deprecated:: Deprecated Directives
4497 @section @code{.abort}
4499 @cindex @code{abort} directive
4500 @cindex stopping the assembly
4501 This directive stops the assembly immediately. It is for
4502 compatibility with other assemblers. The original idea was that the
4503 assembly language source would be piped into the assembler. If the sender
4504 of the source quit, it could use this directive tells @command{@value{AS}} to
4505 quit also. One day @code{.abort} will not be supported.
4509 @section @code{.ABORT} (COFF)
4511 @cindex @code{ABORT} directive
4512 When producing COFF output, @command{@value{AS}} accepts this directive as a
4513 synonym for @samp{.abort}.
4516 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4522 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4524 @cindex padding the location counter
4525 @cindex @code{align} directive
4526 Pad the location counter (in the current subsection) to a particular storage
4527 boundary. The first expression (which must be absolute) is the alignment
4528 required, as described below.
4530 The second expression (also absolute) gives the fill value to be stored in the
4531 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4532 padding bytes are normally zero. However, on some systems, if the section is
4533 marked as containing code and the fill value is omitted, the space is filled
4534 with no-op instructions.
4536 The third expression is also absolute, and is also optional. If it is present,
4537 it is the maximum number of bytes that should be skipped by this alignment
4538 directive. If doing the alignment would require skipping more bytes than the
4539 specified maximum, then the alignment is not done at all. You can omit the
4540 fill value (the second argument) entirely by simply using two commas after the
4541 required alignment; this can be useful if you want the alignment to be filled
4542 with no-op instructions when appropriate.
4544 The way the required alignment is specified varies from system to system.
4545 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or1k,
4546 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4547 alignment request in bytes. For example @samp{.align 8} advances
4548 the location counter until it is a multiple of 8. If the location counter
4549 is already a multiple of 8, no change is needed. For the tic54x, the
4550 first expression is the alignment request in words.
4552 For other systems, including ppc, i386 using a.out format, arm and
4553 strongarm, it is the
4554 number of low-order zero bits the location counter must have after
4555 advancement. For example @samp{.align 3} advances the location
4556 counter until it a multiple of 8. If the location counter is already a
4557 multiple of 8, no change is needed.
4559 This inconsistency is due to the different behaviors of the various
4560 native assemblers for these systems which GAS must emulate.
4561 GAS also provides @code{.balign} and @code{.p2align} directives,
4562 described later, which have a consistent behavior across all
4563 architectures (but are specific to GAS).
4566 @section @code{.altmacro}
4567 Enable alternate macro mode, enabling:
4570 @item LOCAL @var{name} [ , @dots{} ]
4571 One additional directive, @code{LOCAL}, is available. It is used to
4572 generate a string replacement for each of the @var{name} arguments, and
4573 replace any instances of @var{name} in each macro expansion. The
4574 replacement string is unique in the assembly, and different for each
4575 separate macro expansion. @code{LOCAL} allows you to write macros that
4576 define symbols, without fear of conflict between separate macro expansions.
4578 @item String delimiters
4579 You can write strings delimited in these other ways besides
4580 @code{"@var{string}"}:
4583 @item '@var{string}'
4584 You can delimit strings with single-quote characters.
4586 @item <@var{string}>
4587 You can delimit strings with matching angle brackets.
4590 @item single-character string escape
4591 To include any single character literally in a string (even if the
4592 character would otherwise have some special meaning), you can prefix the
4593 character with @samp{!} (an exclamation mark). For example, you can
4594 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4596 @item Expression results as strings
4597 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4598 and use the result as a string.
4602 @section @code{.ascii "@var{string}"}@dots{}
4604 @cindex @code{ascii} directive
4605 @cindex string literals
4606 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4607 separated by commas. It assembles each string (with no automatic
4608 trailing zero byte) into consecutive addresses.
4611 @section @code{.asciz "@var{string}"}@dots{}
4613 @cindex @code{asciz} directive
4614 @cindex zero-terminated strings
4615 @cindex null-terminated strings
4616 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4617 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4620 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4622 @cindex padding the location counter given number of bytes
4623 @cindex @code{balign} directive
4624 Pad the location counter (in the current subsection) to a particular
4625 storage boundary. The first expression (which must be absolute) is the
4626 alignment request in bytes. For example @samp{.balign 8} advances
4627 the location counter until it is a multiple of 8. If the location counter
4628 is already a multiple of 8, no change is needed.
4630 The second expression (also absolute) gives the fill value to be stored in the
4631 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4632 padding bytes are normally zero. However, on some systems, if the section is
4633 marked as containing code and the fill value is omitted, the space is filled
4634 with no-op instructions.
4636 The third expression is also absolute, and is also optional. If it is present,
4637 it is the maximum number of bytes that should be skipped by this alignment
4638 directive. If doing the alignment would require skipping more bytes than the
4639 specified maximum, then the alignment is not done at all. You can omit the
4640 fill value (the second argument) entirely by simply using two commas after the
4641 required alignment; this can be useful if you want the alignment to be filled
4642 with no-op instructions when appropriate.
4644 @cindex @code{balignw} directive
4645 @cindex @code{balignl} directive
4646 The @code{.balignw} and @code{.balignl} directives are variants of the
4647 @code{.balign} directive. The @code{.balignw} directive treats the fill
4648 pattern as a two byte word value. The @code{.balignl} directives treats the
4649 fill pattern as a four byte longword value. For example, @code{.balignw
4650 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4651 filled in with the value 0x368d (the exact placement of the bytes depends upon
4652 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4655 @node Bundle directives
4656 @section Bundle directives
4657 @subsection @code{.bundle_align_mode @var{abs-expr}}
4658 @cindex @code{bundle_align_mode} directive
4660 @cindex instruction bundle
4661 @cindex aligned instruction bundle
4662 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4663 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4664 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4665 disabled (which is the default state). If the argument it not zero, it
4666 gives the size of an instruction bundle as a power of two (as for the
4667 @code{.p2align} directive, @pxref{P2align}).
4669 For some targets, it's an ABI requirement that no instruction may span a
4670 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4671 instructions that starts on an aligned boundary. For example, if
4672 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4673 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4674 effect, no single instruction may span a boundary between bundles. If an
4675 instruction would start too close to the end of a bundle for the length of
4676 that particular instruction to fit within the bundle, then the space at the
4677 end of that bundle is filled with no-op instructions so the instruction
4678 starts in the next bundle. As a corollary, it's an error if any single
4679 instruction's encoding is longer than the bundle size.
4681 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4682 @cindex @code{bundle_lock} directive
4683 @cindex @code{bundle_unlock} directive
4684 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4685 allow explicit control over instruction bundle padding. These directives
4686 are only valid when @code{.bundle_align_mode} has been used to enable
4687 aligned instruction bundle mode. It's an error if they appear when
4688 @code{.bundle_align_mode} has not been used at all, or when the last
4689 directive was @w{@code{.bundle_align_mode 0}}.
4691 @cindex bundle-locked
4692 For some targets, it's an ABI requirement that certain instructions may
4693 appear only as part of specified permissible sequences of multiple
4694 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4695 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4696 instruction sequence. For purposes of aligned instruction bundle mode, a
4697 sequence starting with @code{.bundle_lock} and ending with
4698 @code{.bundle_unlock} is treated as a single instruction. That is, the
4699 entire sequence must fit into a single bundle and may not span a bundle
4700 boundary. If necessary, no-op instructions will be inserted before the
4701 first instruction of the sequence so that the whole sequence starts on an
4702 aligned bundle boundary. It's an error if the sequence is longer than the
4705 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4706 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4707 nested. That is, a second @code{.bundle_lock} directive before the next
4708 @code{.bundle_unlock} directive has no effect except that it must be
4709 matched by another closing @code{.bundle_unlock} so that there is the
4710 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4713 @section @code{.byte @var{expressions}}
4715 @cindex @code{byte} directive
4716 @cindex integers, one byte
4717 @code{.byte} expects zero or more expressions, separated by commas.
4718 Each expression is assembled into the next byte.
4720 @node CFI directives
4721 @section CFI directives
4722 @subsection @code{.cfi_sections @var{section_list}}
4723 @cindex @code{cfi_sections} directive
4724 @code{.cfi_sections} may be used to specify whether CFI directives
4725 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4726 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4727 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4728 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4729 directive is not used is @code{.cfi_sections .eh_frame}.
4731 On targets that support compact unwinding tables these can be generated
4732 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4734 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4735 which is used by the @value{TIC6X} target.
4737 The @code{.cfi_sections} directive can be repeated, with the same or different
4738 arguments, provided that CFI generation has not yet started. Once CFI
4739 generation has started however the section list is fixed and any attempts to
4740 redefine it will result in an error.
4742 @subsection @code{.cfi_startproc [simple]}
4743 @cindex @code{cfi_startproc} directive
4744 @code{.cfi_startproc} is used at the beginning of each function that
4745 should have an entry in @code{.eh_frame}. It initializes some internal
4746 data structures. Don't forget to close the function by
4747 @code{.cfi_endproc}.
4749 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4750 it also emits some architecture dependent initial CFI instructions.
4752 @subsection @code{.cfi_endproc}
4753 @cindex @code{cfi_endproc} directive
4754 @code{.cfi_endproc} is used at the end of a function where it closes its
4755 unwind entry previously opened by
4756 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4758 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4759 @cindex @code{cfi_personality} directive
4760 @code{.cfi_personality} defines personality routine and its encoding.
4761 @var{encoding} must be a constant determining how the personality
4762 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4763 argument is not present, otherwise second argument should be
4764 a constant or a symbol name. When using indirect encodings,
4765 the symbol provided should be the location where personality
4766 can be loaded from, not the personality routine itself.
4767 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4768 no personality routine.
4770 @subsection @code{.cfi_personality_id @var{id}}
4771 @cindex @code{cfi_personality_id} directive
4772 @code{cfi_personality_id} defines a personality routine by its index as
4773 defined in a compact unwinding format.
4774 Only valid when generating compact EH frames (i.e.
4775 with @code{.cfi_sections eh_frame_entry}.
4777 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4778 @cindex @code{cfi_fde_data} directive
4779 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4780 used for the current function. These are emitted inline in the
4781 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4782 in the @code{.gnu.extab} section otherwise.
4783 Only valid when generating compact EH frames (i.e.
4784 with @code{.cfi_sections eh_frame_entry}.
4786 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4787 @code{.cfi_lsda} defines LSDA and its encoding.
4788 @var{encoding} must be a constant determining how the LSDA
4789 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4790 argument is not present, otherwise the second argument should be a constant
4791 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4792 meaning that no LSDA is present.
4794 @subsection @code{.cfi_inline_lsda} [@var{align}]
4795 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4796 switches to the corresponding @code{.gnu.extab} section.
4797 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4798 Only valid when generating compact EH frames (i.e.
4799 with @code{.cfi_sections eh_frame_entry}.
4801 The table header and unwinding opcodes will be generated at this point,
4802 so that they are immediately followed by the LSDA data. The symbol
4803 referenced by the @code{.cfi_lsda} directive should still be defined
4804 in case a fallback FDE based encoding is used. The LSDA data is terminated
4805 by a section directive.
4807 The optional @var{align} argument specifies the alignment required.
4808 The alignment is specified as a power of two, as with the
4809 @code{.p2align} directive.
4811 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4812 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4813 address from @var{register} and add @var{offset} to it}.
4815 @subsection @code{.cfi_def_cfa_register @var{register}}
4816 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4817 now on @var{register} will be used instead of the old one. Offset
4820 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4821 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4822 remains the same, but @var{offset} is new. Note that it is the
4823 absolute offset that will be added to a defined register to compute
4826 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4827 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4828 value that is added/substracted from the previous offset.
4830 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4831 Previous value of @var{register} is saved at offset @var{offset} from
4834 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4835 Previous value of @var{register} is saved at offset @var{offset} from
4836 the current CFA register. This is transformed to @code{.cfi_offset}
4837 using the known displacement of the CFA register from the CFA.
4838 This is often easier to use, because the number will match the
4839 code it's annotating.
4841 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4842 Previous value of @var{register1} is saved in register @var{register2}.
4844 @subsection @code{.cfi_restore @var{register}}
4845 @code{.cfi_restore} says that the rule for @var{register} is now the
4846 same as it was at the beginning of the function, after all initial
4847 instruction added by @code{.cfi_startproc} were executed.
4849 @subsection @code{.cfi_undefined @var{register}}
4850 From now on the previous value of @var{register} can't be restored anymore.
4852 @subsection @code{.cfi_same_value @var{register}}
4853 Current value of @var{register} is the same like in the previous frame,
4854 i.e. no restoration needed.
4856 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4857 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4858 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4859 places them in the current row. This is useful for situations where you have
4860 multiple @code{.cfi_*} directives that need to be undone due to the control
4861 flow of the program. For example, we could have something like this (assuming
4862 the CFA is the value of @code{rbp}):
4872 .cfi_def_cfa %rsp, 8
4875 /* Do something else */
4878 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4879 to the instructions before @code{label}. This means we'd have to add multiple
4880 @code{.cfi} directives after @code{label} to recreate the original save
4881 locations of the registers, as well as setting the CFA back to the value of
4882 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4894 .cfi_def_cfa %rsp, 8
4898 /* Do something else */
4901 That way, the rules for the instructions after @code{label} will be the same
4902 as before the first @code{.cfi_restore} without having to use multiple
4903 @code{.cfi} directives.
4905 @subsection @code{.cfi_return_column @var{register}}
4906 Change return column @var{register}, i.e. the return address is either
4907 directly in @var{register} or can be accessed by rules for @var{register}.
4909 @subsection @code{.cfi_signal_frame}
4910 Mark current function as signal trampoline.
4912 @subsection @code{.cfi_window_save}
4913 SPARC register window has been saved.
4915 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4916 Allows the user to add arbitrary bytes to the unwind info. One
4917 might use this to add OS-specific CFI opcodes, or generic CFI
4918 opcodes that GAS does not yet support.
4920 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4921 The current value of @var{register} is @var{label}. The value of @var{label}
4922 will be encoded in the output file according to @var{encoding}; see the
4923 description of @code{.cfi_personality} for details on this encoding.
4925 The usefulness of equating a register to a fixed label is probably
4926 limited to the return address register. Here, it can be useful to
4927 mark a code segment that has only one return address which is reached
4928 by a direct branch and no copy of the return address exists in memory
4929 or another register.
4932 @section @code{.comm @var{symbol} , @var{length} }
4934 @cindex @code{comm} directive
4935 @cindex symbol, common
4936 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4937 common symbol in one object file may be merged with a defined or common symbol
4938 of the same name in another object file. If @code{@value{LD}} does not see a
4939 definition for the symbol--just one or more common symbols--then it will
4940 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4941 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4942 the same name, and they do not all have the same size, it will allocate space
4943 using the largest size.
4946 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4947 an optional third argument. This is the desired alignment of the symbol,
4948 specified for ELF as a byte boundary (for example, an alignment of 16 means
4949 that the least significant 4 bits of the address should be zero), and for PE
4950 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4951 boundary). The alignment must be an absolute expression, and it must be a
4952 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4953 common symbol, it will use the alignment when placing the symbol. If no
4954 alignment is specified, @command{@value{AS}} will set the alignment to the
4955 largest power of two less than or equal to the size of the symbol, up to a
4956 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4957 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4958 @samp{--section-alignment} option; image file sections in PE are aligned to
4959 multiples of 4096, which is far too large an alignment for ordinary variables.
4960 It is rather the default alignment for (non-debug) sections within object
4961 (@samp{*.o}) files, which are less strictly aligned.}.
4965 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4966 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4970 @section @code{.data @var{subsection}}
4972 @cindex @code{data} directive
4973 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4974 end of the data subsection numbered @var{subsection} (which is an
4975 absolute expression). If @var{subsection} is omitted, it defaults
4980 @section @code{.def @var{name}}
4982 @cindex @code{def} directive
4983 @cindex COFF symbols, debugging
4984 @cindex debugging COFF symbols
4985 Begin defining debugging information for a symbol @var{name}; the
4986 definition extends until the @code{.endef} directive is encountered.
4989 This directive is only observed when @command{@value{AS}} is configured for COFF
4990 format output; when producing @code{b.out}, @samp{.def} is recognized,
4997 @section @code{.desc @var{symbol}, @var{abs-expression}}
4999 @cindex @code{desc} directive
5000 @cindex COFF symbol descriptor
5001 @cindex symbol descriptor, COFF
5002 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5003 to the low 16 bits of an absolute expression.
5006 The @samp{.desc} directive is not available when @command{@value{AS}} is
5007 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5008 object format. For the sake of compatibility, @command{@value{AS}} accepts
5009 it, but produces no output, when configured for COFF.
5015 @section @code{.dim}
5017 @cindex @code{dim} directive
5018 @cindex COFF auxiliary symbol information
5019 @cindex auxiliary symbol information, COFF
5020 This directive is generated by compilers to include auxiliary debugging
5021 information in the symbol table. It is only permitted inside
5022 @code{.def}/@code{.endef} pairs.
5025 @samp{.dim} is only meaningful when generating COFF format output; when
5026 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5032 @section @code{.double @var{flonums}}
5034 @cindex @code{double} directive
5035 @cindex floating point numbers (double)
5036 @code{.double} expects zero or more flonums, separated by commas. It
5037 assembles floating point numbers.
5039 The exact kind of floating point numbers emitted depends on how
5040 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5044 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5045 in @sc{ieee} format.
5050 @section @code{.eject}
5052 @cindex @code{eject} directive
5053 @cindex new page, in listings
5054 @cindex page, in listings
5055 @cindex listing control: new page
5056 Force a page break at this point, when generating assembly listings.
5059 @section @code{.else}
5061 @cindex @code{else} directive
5062 @code{.else} is part of the @command{@value{AS}} support for conditional
5063 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5064 of code to be assembled if the condition for the preceding @code{.if}
5068 @section @code{.elseif}
5070 @cindex @code{elseif} directive
5071 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5072 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5073 @code{.if} block that would otherwise fill the entire @code{.else} section.
5076 @section @code{.end}
5078 @cindex @code{end} directive
5079 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5080 process anything in the file past the @code{.end} directive.
5084 @section @code{.endef}
5086 @cindex @code{endef} directive
5087 This directive flags the end of a symbol definition begun with
5091 @samp{.endef} is only meaningful when generating COFF format output; if
5092 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
5093 directive but ignores it.
5098 @section @code{.endfunc}
5099 @cindex @code{endfunc} directive
5100 @code{.endfunc} marks the end of a function specified with @code{.func}.
5103 @section @code{.endif}
5105 @cindex @code{endif} directive
5106 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5107 it marks the end of a block of code that is only assembled
5108 conditionally. @xref{If,,@code{.if}}.
5111 @section @code{.equ @var{symbol}, @var{expression}}
5113 @cindex @code{equ} directive
5114 @cindex assigning values to symbols
5115 @cindex symbols, assigning values to
5116 This directive sets the value of @var{symbol} to @var{expression}.
5117 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5120 The syntax for @code{equ} on the HPPA is
5121 @samp{@var{symbol} .equ @var{expression}}.
5125 The syntax for @code{equ} on the Z80 is
5126 @samp{@var{symbol} equ @var{expression}}.
5127 On the Z80 it is an eror if @var{symbol} is already defined,
5128 but the symbol is not protected from later redefinition.
5129 Compare @ref{Equiv}.
5133 @section @code{.equiv @var{symbol}, @var{expression}}
5134 @cindex @code{equiv} directive
5135 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5136 the assembler will signal an error if @var{symbol} is already defined. Note a
5137 symbol which has been referenced but not actually defined is considered to be
5140 Except for the contents of the error message, this is roughly equivalent to
5147 plus it protects the symbol from later redefinition.
5150 @section @code{.eqv @var{symbol}, @var{expression}}
5151 @cindex @code{eqv} directive
5152 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5153 evaluate the expression or any part of it immediately. Instead each time
5154 the resulting symbol is used in an expression, a snapshot of its current
5158 @section @code{.err}
5159 @cindex @code{err} directive
5160 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5161 message and, unless the @option{-Z} option was used, it will not generate an
5162 object file. This can be used to signal an error in conditionally compiled code.
5165 @section @code{.error "@var{string}"}
5166 @cindex error directive
5168 Similarly to @code{.err}, this directive emits an error, but you can specify a
5169 string that will be emitted as the error message. If you don't specify the
5170 message, it defaults to @code{".error directive invoked in source file"}.
5171 @xref{Errors, ,Error and Warning Messages}.
5174 .error "This code has not been assembled and tested."
5178 @section @code{.exitm}
5179 Exit early from the current macro definition. @xref{Macro}.
5182 @section @code{.extern}
5184 @cindex @code{extern} directive
5185 @code{.extern} is accepted in the source program---for compatibility
5186 with other assemblers---but it is ignored. @command{@value{AS}} treats
5187 all undefined symbols as external.
5190 @section @code{.fail @var{expression}}
5192 @cindex @code{fail} directive
5193 Generates an error or a warning. If the value of the @var{expression} is 500
5194 or more, @command{@value{AS}} will print a warning message. If the value is less
5195 than 500, @command{@value{AS}} will print an error message. The message will
5196 include the value of @var{expression}. This can occasionally be useful inside
5197 complex nested macros or conditional assembly.
5200 @section @code{.file}
5201 @cindex @code{file} directive
5203 @ifclear no-file-dir
5204 There are two different versions of the @code{.file} directive. Targets
5205 that support DWARF2 line number information use the DWARF2 version of
5206 @code{.file}. Other targets use the default version.
5208 @subheading Default Version
5210 @cindex logical file name
5211 @cindex file name, logical
5212 This version of the @code{.file} directive tells @command{@value{AS}} that we
5213 are about to start a new logical file. The syntax is:
5219 @var{string} is the new file name. In general, the filename is
5220 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5221 to specify an empty file name, you must give the quotes--@code{""}. This
5222 statement may go away in future: it is only recognized to be compatible with
5223 old @command{@value{AS}} programs.
5225 @subheading DWARF2 Version
5228 When emitting DWARF2 line number information, @code{.file} assigns filenames
5229 to the @code{.debug_line} file name table. The syntax is:
5232 .file @var{fileno} @var{filename}
5235 The @var{fileno} operand should be a unique positive integer to use as the
5236 index of the entry in the table. The @var{filename} operand is a C string
5239 The detail of filename indices is exposed to the user because the filename
5240 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5241 information, and thus the user must know the exact indices that table
5245 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5247 @cindex @code{fill} directive
5248 @cindex writing patterns in memory
5249 @cindex patterns, writing in memory
5250 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5251 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5252 may be zero or more. @var{Size} may be zero or more, but if it is
5253 more than 8, then it is deemed to have the value 8, compatible with
5254 other people's assemblers. The contents of each @var{repeat} bytes
5255 is taken from an 8-byte number. The highest order 4 bytes are
5256 zero. The lowest order 4 bytes are @var{value} rendered in the
5257 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5258 Each @var{size} bytes in a repetition is taken from the lowest order
5259 @var{size} bytes of this number. Again, this bizarre behavior is
5260 compatible with other people's assemblers.
5262 @var{size} and @var{value} are optional.
5263 If the second comma and @var{value} are absent, @var{value} is
5264 assumed zero. If the first comma and following tokens are absent,
5265 @var{size} is assumed to be 1.
5268 @section @code{.float @var{flonums}}
5270 @cindex floating point numbers (single)
5271 @cindex @code{float} directive
5272 This directive assembles zero or more flonums, separated by commas. It
5273 has the same effect as @code{.single}.
5275 The exact kind of floating point numbers emitted depends on how
5276 @command{@value{AS}} is configured.
5277 @xref{Machine Dependencies}.
5281 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5282 in @sc{ieee} format.
5287 @section @code{.func @var{name}[,@var{label}]}
5288 @cindex @code{func} directive
5289 @code{.func} emits debugging information to denote function @var{name}, and
5290 is ignored unless the file is assembled with debugging enabled.
5291 Only @samp{--gstabs[+]} is currently supported.
5292 @var{label} is the entry point of the function and if omitted @var{name}
5293 prepended with the @samp{leading char} is used.
5294 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5295 All functions are currently defined to have @code{void} return type.
5296 The function must be terminated with @code{.endfunc}.
5299 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5301 @cindex @code{global} directive
5302 @cindex symbol, making visible to linker
5303 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5304 @var{symbol} in your partial program, its value is made available to
5305 other partial programs that are linked with it. Otherwise,
5306 @var{symbol} takes its attributes from a symbol of the same name
5307 from another file linked into the same program.
5309 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5310 compatibility with other assemblers.
5313 On the HPPA, @code{.global} is not always enough to make it accessible to other
5314 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5315 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5320 @section @code{.gnu_attribute @var{tag},@var{value}}
5321 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5324 @section @code{.hidden @var{names}}
5326 @cindex @code{hidden} directive
5328 This is one of the ELF visibility directives. The other two are
5329 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5330 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5332 This directive overrides the named symbols default visibility (which is set by
5333 their binding: local, global or weak). The directive sets the visibility to
5334 @code{hidden} which means that the symbols are not visible to other components.
5335 Such symbols are always considered to be @code{protected} as well.
5339 @section @code{.hword @var{expressions}}
5341 @cindex @code{hword} directive
5342 @cindex integers, 16-bit
5343 @cindex numbers, 16-bit
5344 @cindex sixteen bit integers
5345 This expects zero or more @var{expressions}, and emits
5346 a 16 bit number for each.
5349 This directive is a synonym for @samp{.short}; depending on the target
5350 architecture, it may also be a synonym for @samp{.word}.
5354 This directive is a synonym for @samp{.short}.
5357 This directive is a synonym for both @samp{.short} and @samp{.word}.
5362 @section @code{.ident}
5364 @cindex @code{ident} directive
5366 This directive is used by some assemblers to place tags in object files. The
5367 behavior of this directive varies depending on the target. When using the
5368 a.out object file format, @command{@value{AS}} simply accepts the directive for
5369 source-file compatibility with existing assemblers, but does not emit anything
5370 for it. When using COFF, comments are emitted to the @code{.comment} or
5371 @code{.rdata} section, depending on the target. When using ELF, comments are
5372 emitted to the @code{.comment} section.
5375 @section @code{.if @var{absolute expression}}
5377 @cindex conditional assembly
5378 @cindex @code{if} directive
5379 @code{.if} marks the beginning of a section of code which is only
5380 considered part of the source program being assembled if the argument
5381 (which must be an @var{absolute expression}) is non-zero. The end of
5382 the conditional section of code must be marked by @code{.endif}
5383 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5384 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5385 If you have several conditions to check, @code{.elseif} may be used to avoid
5386 nesting blocks if/else within each subsequent @code{.else} block.
5388 The following variants of @code{.if} are also supported:
5390 @cindex @code{ifdef} directive
5391 @item .ifdef @var{symbol}
5392 Assembles the following section of code if the specified @var{symbol}
5393 has been defined. Note a symbol which has been referenced but not yet defined
5394 is considered to be undefined.
5396 @cindex @code{ifb} directive
5397 @item .ifb @var{text}
5398 Assembles the following section of code if the operand is blank (empty).
5400 @cindex @code{ifc} directive
5401 @item .ifc @var{string1},@var{string2}
5402 Assembles the following section of code if the two strings are the same. The
5403 strings may be optionally quoted with single quotes. If they are not quoted,
5404 the first string stops at the first comma, and the second string stops at the
5405 end of the line. Strings which contain whitespace should be quoted. The
5406 string comparison is case sensitive.
5408 @cindex @code{ifeq} directive
5409 @item .ifeq @var{absolute expression}
5410 Assembles the following section of code if the argument is zero.
5412 @cindex @code{ifeqs} directive
5413 @item .ifeqs @var{string1},@var{string2}
5414 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5416 @cindex @code{ifge} directive
5417 @item .ifge @var{absolute expression}
5418 Assembles the following section of code if the argument is greater than or
5421 @cindex @code{ifgt} directive
5422 @item .ifgt @var{absolute expression}
5423 Assembles the following section of code if the argument is greater than zero.
5425 @cindex @code{ifle} directive
5426 @item .ifle @var{absolute expression}
5427 Assembles the following section of code if the argument is less than or equal
5430 @cindex @code{iflt} directive
5431 @item .iflt @var{absolute expression}
5432 Assembles the following section of code if the argument is less than zero.
5434 @cindex @code{ifnb} directive
5435 @item .ifnb @var{text}
5436 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5437 following section of code if the operand is non-blank (non-empty).
5439 @cindex @code{ifnc} directive
5440 @item .ifnc @var{string1},@var{string2}.
5441 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5442 following section of code if the two strings are not the same.
5444 @cindex @code{ifndef} directive
5445 @cindex @code{ifnotdef} directive
5446 @item .ifndef @var{symbol}
5447 @itemx .ifnotdef @var{symbol}
5448 Assembles the following section of code if the specified @var{symbol}
5449 has not been defined. Both spelling variants are equivalent. Note a symbol
5450 which has been referenced but not yet defined is considered to be undefined.
5452 @cindex @code{ifne} directive
5453 @item .ifne @var{absolute expression}
5454 Assembles the following section of code if the argument is not equal to zero
5455 (in other words, this is equivalent to @code{.if}).
5457 @cindex @code{ifnes} directive
5458 @item .ifnes @var{string1},@var{string2}
5459 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5460 following section of code if the two strings are not the same.
5464 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5466 @cindex @code{incbin} directive
5467 @cindex binary files, including
5468 The @code{incbin} directive includes @var{file} verbatim at the current
5469 location. You can control the search paths used with the @samp{-I} command-line
5470 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5473 The @var{skip} argument skips a number of bytes from the start of the
5474 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5475 read. Note that the data is not aligned in any way, so it is the user's
5476 responsibility to make sure that proper alignment is provided both before and
5477 after the @code{incbin} directive.
5480 @section @code{.include "@var{file}"}
5482 @cindex @code{include} directive
5483 @cindex supporting files, including
5484 @cindex files, including
5485 This directive provides a way to include supporting files at specified
5486 points in your source program. The code from @var{file} is assembled as
5487 if it followed the point of the @code{.include}; when the end of the
5488 included file is reached, assembly of the original file continues. You
5489 can control the search paths used with the @samp{-I} command-line option
5490 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5494 @section @code{.int @var{expressions}}
5496 @cindex @code{int} directive
5497 @cindex integers, 32-bit
5498 Expect zero or more @var{expressions}, of any section, separated by commas.
5499 For each expression, emit a number that, at run time, is the value of that
5500 expression. The byte order and bit size of the number depends on what kind
5501 of target the assembly is for.
5505 On most forms of the H8/300, @code{.int} emits 16-bit
5506 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5513 @section @code{.internal @var{names}}
5515 @cindex @code{internal} directive
5517 This is one of the ELF visibility directives. The other two are
5518 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5519 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5521 This directive overrides the named symbols default visibility (which is set by
5522 their binding: local, global or weak). The directive sets the visibility to
5523 @code{internal} which means that the symbols are considered to be @code{hidden}
5524 (i.e., not visible to other components), and that some extra, processor specific
5525 processing must also be performed upon the symbols as well.
5529 @section @code{.irp @var{symbol},@var{values}}@dots{}
5531 @cindex @code{irp} directive
5532 Evaluate a sequence of statements assigning different values to @var{symbol}.
5533 The sequence of statements starts at the @code{.irp} directive, and is
5534 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5535 set to @var{value}, and the sequence of statements is assembled. If no
5536 @var{value} is listed, the sequence of statements is assembled once, with
5537 @var{symbol} set to the null string. To refer to @var{symbol} within the
5538 sequence of statements, use @var{\symbol}.
5540 For example, assembling
5548 is equivalent to assembling
5556 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5559 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5561 @cindex @code{irpc} directive
5562 Evaluate a sequence of statements assigning different values to @var{symbol}.
5563 The sequence of statements starts at the @code{.irpc} directive, and is
5564 terminated by an @code{.endr} directive. For each character in @var{value},
5565 @var{symbol} is set to the character, and the sequence of statements is
5566 assembled. If no @var{value} is listed, the sequence of statements is
5567 assembled once, with @var{symbol} set to the null string. To refer to
5568 @var{symbol} within the sequence of statements, use @var{\symbol}.
5570 For example, assembling
5578 is equivalent to assembling
5586 For some caveats with the spelling of @var{symbol}, see also the discussion
5590 @section @code{.lcomm @var{symbol} , @var{length}}
5592 @cindex @code{lcomm} directive
5593 @cindex local common symbols
5594 @cindex symbols, local common
5595 Reserve @var{length} (an absolute expression) bytes for a local common
5596 denoted by @var{symbol}. The section and value of @var{symbol} are
5597 those of the new local common. The addresses are allocated in the bss
5598 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5599 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5600 not visible to @code{@value{LD}}.
5603 Some targets permit a third argument to be used with @code{.lcomm}. This
5604 argument specifies the desired alignment of the symbol in the bss section.
5608 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5609 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5613 @section @code{.lflags}
5615 @cindex @code{lflags} directive (ignored)
5616 @command{@value{AS}} accepts this directive, for compatibility with other
5617 assemblers, but ignores it.
5619 @ifclear no-line-dir
5621 @section @code{.line @var{line-number}}
5623 @cindex @code{line} directive
5624 @cindex logical line number
5626 Change the logical line number. @var{line-number} must be an absolute
5627 expression. The next line has that logical line number. Therefore any other
5628 statements on the current line (after a statement separator character) are
5629 reported as on logical line number @var{line-number} @minus{} 1. One day
5630 @command{@value{AS}} will no longer support this directive: it is recognized only
5631 for compatibility with existing assembler programs.
5634 Even though this is a directive associated with the @code{a.out} or
5635 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5636 when producing COFF output, and treats @samp{.line} as though it
5637 were the COFF @samp{.ln} @emph{if} it is found outside a
5638 @code{.def}/@code{.endef} pair.
5640 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5641 used by compilers to generate auxiliary symbol information for
5646 @section @code{.linkonce [@var{type}]}
5648 @cindex @code{linkonce} directive
5649 @cindex common sections
5650 Mark the current section so that the linker only includes a single copy of it.
5651 This may be used to include the same section in several different object files,
5652 but ensure that the linker will only include it once in the final output file.
5653 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5654 Duplicate sections are detected based on the section name, so it should be
5657 This directive is only supported by a few object file formats; as of this
5658 writing, the only object file format which supports it is the Portable
5659 Executable format used on Windows NT.
5661 The @var{type} argument is optional. If specified, it must be one of the
5662 following strings. For example:
5666 Not all types may be supported on all object file formats.
5670 Silently discard duplicate sections. This is the default.
5673 Warn if there are duplicate sections, but still keep only one copy.
5676 Warn if any of the duplicates have different sizes.
5679 Warn if any of the duplicates do not have exactly the same contents.
5683 @section @code{.list}
5685 @cindex @code{list} directive
5686 @cindex listing control, turning on
5687 Control (in conjunction with the @code{.nolist} directive) whether or
5688 not assembly listings are generated. These two directives maintain an
5689 internal counter (which is zero initially). @code{.list} increments the
5690 counter, and @code{.nolist} decrements it. Assembly listings are
5691 generated whenever the counter is greater than zero.
5693 By default, listings are disabled. When you enable them (with the
5694 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5695 the initial value of the listing counter is one.
5698 @section @code{.ln @var{line-number}}
5700 @cindex @code{ln} directive
5701 @ifclear no-line-dir
5702 @samp{.ln} is a synonym for @samp{.line}.
5705 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5706 must be an absolute expression. The next line has that logical
5707 line number, so any other statements on the current line (after a
5708 statement separator character @code{;}) are reported as on logical
5709 line number @var{line-number} @minus{} 1.
5712 This directive is accepted, but ignored, when @command{@value{AS}} is
5713 configured for @code{b.out}; its effect is only associated with COFF
5719 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5720 @cindex @code{loc} directive
5721 When emitting DWARF2 line number information,
5722 the @code{.loc} directive will add a row to the @code{.debug_line} line
5723 number matrix corresponding to the immediately following assembly
5724 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5725 arguments will be applied to the @code{.debug_line} state machine before
5728 The @var{options} are a sequence of the following tokens in any order:
5732 This option will set the @code{basic_block} register in the
5733 @code{.debug_line} state machine to @code{true}.
5736 This option will set the @code{prologue_end} register in the
5737 @code{.debug_line} state machine to @code{true}.
5739 @item epilogue_begin
5740 This option will set the @code{epilogue_begin} register in the
5741 @code{.debug_line} state machine to @code{true}.
5743 @item is_stmt @var{value}
5744 This option will set the @code{is_stmt} register in the
5745 @code{.debug_line} state machine to @code{value}, which must be
5748 @item isa @var{value}
5749 This directive will set the @code{isa} register in the @code{.debug_line}
5750 state machine to @var{value}, which must be an unsigned integer.
5752 @item discriminator @var{value}
5753 This directive will set the @code{discriminator} register in the @code{.debug_line}
5754 state machine to @var{value}, which must be an unsigned integer.
5758 @node Loc_mark_labels
5759 @section @code{.loc_mark_labels @var{enable}}
5760 @cindex @code{loc_mark_labels} directive
5761 When emitting DWARF2 line number information,
5762 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5763 to the @code{.debug_line} line number matrix with the @code{basic_block}
5764 register in the state machine set whenever a code label is seen.
5765 The @var{enable} argument should be either 1 or 0, to enable or disable
5766 this function respectively.
5770 @section @code{.local @var{names}}
5772 @cindex @code{local} directive
5773 This directive, which is available for ELF targets, marks each symbol in
5774 the comma-separated list of @code{names} as a local symbol so that it
5775 will not be externally visible. If the symbols do not already exist,
5776 they will be created.
5778 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5779 accept an alignment argument, which is the case for most ELF targets,
5780 the @code{.local} directive can be used in combination with @code{.comm}
5781 (@pxref{Comm}) to define aligned local common data.
5785 @section @code{.long @var{expressions}}
5787 @cindex @code{long} directive
5788 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5791 @c no one seems to know what this is for or whether this description is
5792 @c what it really ought to do
5794 @section @code{.lsym @var{symbol}, @var{expression}}
5796 @cindex @code{lsym} directive
5797 @cindex symbol, not referenced in assembly
5798 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5799 the hash table, ensuring it cannot be referenced by name during the
5800 rest of the assembly. This sets the attributes of the symbol to be
5801 the same as the expression value:
5803 @var{other} = @var{descriptor} = 0
5804 @var{type} = @r{(section of @var{expression})}
5805 @var{value} = @var{expression}
5808 The new symbol is not flagged as external.
5812 @section @code{.macro}
5815 The commands @code{.macro} and @code{.endm} allow you to define macros that
5816 generate assembly output. For example, this definition specifies a macro
5817 @code{sum} that puts a sequence of numbers into memory:
5820 .macro sum from=0, to=5
5829 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5841 @item .macro @var{macname}
5842 @itemx .macro @var{macname} @var{macargs} @dots{}
5843 @cindex @code{macro} directive
5844 Begin the definition of a macro called @var{macname}. If your macro
5845 definition requires arguments, specify their names after the macro name,
5846 separated by commas or spaces. You can qualify the macro argument to
5847 indicate whether all invocations must specify a non-blank value (through
5848 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5849 (through @samp{:@code{vararg}}). You can supply a default value for any
5850 macro argument by following the name with @samp{=@var{deflt}}. You
5851 cannot define two macros with the same @var{macname} unless it has been
5852 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5853 definitions. For example, these are all valid @code{.macro} statements:
5857 Begin the definition of a macro called @code{comm}, which takes no
5860 @item .macro plus1 p, p1
5861 @itemx .macro plus1 p p1
5862 Either statement begins the definition of a macro called @code{plus1},
5863 which takes two arguments; within the macro definition, write
5864 @samp{\p} or @samp{\p1} to evaluate the arguments.
5866 @item .macro reserve_str p1=0 p2
5867 Begin the definition of a macro called @code{reserve_str}, with two
5868 arguments. The first argument has a default value, but not the second.
5869 After the definition is complete, you can call the macro either as
5870 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5871 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5872 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5873 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5875 @item .macro m p1:req, p2=0, p3:vararg
5876 Begin the definition of a macro called @code{m}, with at least three
5877 arguments. The first argument must always have a value specified, but
5878 not the second, which instead has a default value. The third formal
5879 will get assigned all remaining arguments specified at invocation time.
5881 When you call a macro, you can specify the argument values either by
5882 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5883 @samp{sum to=17, from=9}.
5887 Note that since each of the @var{macargs} can be an identifier exactly
5888 as any other one permitted by the target architecture, there may be
5889 occasional problems if the target hand-crafts special meanings to certain
5890 characters when they occur in a special position. For example, if the colon
5891 (@code{:}) is generally permitted to be part of a symbol name, but the
5892 architecture specific code special-cases it when occurring as the final
5893 character of a symbol (to denote a label), then the macro parameter
5894 replacement code will have no way of knowing that and consider the whole
5895 construct (including the colon) an identifier, and check only this
5896 identifier for being the subject to parameter substitution. So for example
5897 this macro definition:
5905 might not work as expected. Invoking @samp{label foo} might not create a label
5906 called @samp{foo} but instead just insert the text @samp{\l:} into the
5907 assembler source, probably generating an error about an unrecognised
5910 Similarly problems might occur with the period character (@samp{.})
5911 which is often allowed inside opcode names (and hence identifier names). So
5912 for example constructing a macro to build an opcode from a base name and a
5913 length specifier like this:
5916 .macro opcode base length
5921 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5922 instruction but instead generate some kind of error as the assembler tries to
5923 interpret the text @samp{\base.\length}.
5925 There are several possible ways around this problem:
5928 @item Insert white space
5929 If it is possible to use white space characters then this is the simplest
5938 @item Use @samp{\()}
5939 The string @samp{\()} can be used to separate the end of a macro argument from
5940 the following text. eg:
5943 .macro opcode base length
5948 @item Use the alternate macro syntax mode
5949 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5950 used as a separator. eg:
5960 Note: this problem of correctly identifying string parameters to pseudo ops
5961 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5962 and @code{.irpc} (@pxref{Irpc}) as well.
5965 @cindex @code{endm} directive
5966 Mark the end of a macro definition.
5969 @cindex @code{exitm} directive
5970 Exit early from the current macro definition.
5972 @cindex number of macros executed
5973 @cindex macros, count executed
5975 @command{@value{AS}} maintains a counter of how many macros it has
5976 executed in this pseudo-variable; you can copy that number to your
5977 output with @samp{\@@}, but @emph{only within a macro definition}.
5979 @item LOCAL @var{name} [ , @dots{} ]
5980 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5981 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5982 @xref{Altmacro,,@code{.altmacro}}.
5986 @section @code{.mri @var{val}}
5988 @cindex @code{mri} directive
5989 @cindex MRI mode, temporarily
5990 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5991 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5992 affects code assembled until the next @code{.mri} directive, or until the end
5993 of the file. @xref{M, MRI mode, MRI mode}.
5996 @section @code{.noaltmacro}
5997 Disable alternate macro mode. @xref{Altmacro}.
6000 @section @code{.nolist}
6002 @cindex @code{nolist} directive
6003 @cindex listing control, turning off
6004 Control (in conjunction with the @code{.list} directive) whether or
6005 not assembly listings are generated. These two directives maintain an
6006 internal counter (which is zero initially). @code{.list} increments the
6007 counter, and @code{.nolist} decrements it. Assembly listings are
6008 generated whenever the counter is greater than zero.
6011 @section @code{.octa @var{bignums}}
6013 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
6014 @cindex @code{octa} directive
6015 @cindex integer, 16-byte
6016 @cindex sixteen byte integer
6017 This directive expects zero or more bignums, separated by commas. For each
6018 bignum, it emits a 16-byte integer.
6020 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6021 hence @emph{octa}-word for 16 bytes.
6024 @section @code{.offset @var{loc}}
6026 @cindex @code{offset} directive
6027 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6028 be an absolute expression. This directive may be useful for defining
6029 symbols with absolute values. Do not confuse it with the @code{.org}
6033 @section @code{.org @var{new-lc} , @var{fill}}
6035 @cindex @code{org} directive
6036 @cindex location counter, advancing
6037 @cindex advancing location counter
6038 @cindex current address, advancing
6039 Advance the location counter of the current section to
6040 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6041 expression with the same section as the current subsection. That is,
6042 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6043 wrong section, the @code{.org} directive is ignored. To be compatible
6044 with former assemblers, if the section of @var{new-lc} is absolute,
6045 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6046 is the same as the current subsection.
6048 @code{.org} may only increase the location counter, or leave it
6049 unchanged; you cannot use @code{.org} to move the location counter
6052 @c double negative used below "not undefined" because this is a specific
6053 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6054 @c section. doc@cygnus.com 18feb91
6055 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6056 may not be undefined. If you really detest this restriction we eagerly await
6057 a chance to share your improved assembler.
6059 Beware that the origin is relative to the start of the section, not
6060 to the start of the subsection. This is compatible with other
6061 people's assemblers.
6063 When the location counter (of the current subsection) is advanced, the
6064 intervening bytes are filled with @var{fill} which should be an
6065 absolute expression. If the comma and @var{fill} are omitted,
6066 @var{fill} defaults to zero.
6069 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
6071 @cindex padding the location counter given a power of two
6072 @cindex @code{p2align} directive
6073 Pad the location counter (in the current subsection) to a particular
6074 storage boundary. The first expression (which must be absolute) is the
6075 number of low-order zero bits the location counter must have after
6076 advancement. For example @samp{.p2align 3} advances the location
6077 counter until it a multiple of 8. If the location counter is already a
6078 multiple of 8, no change is needed.
6080 The second expression (also absolute) gives the fill value to be stored in the
6081 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6082 padding bytes are normally zero. However, on some systems, if the section is
6083 marked as containing code and the fill value is omitted, the space is filled
6084 with no-op instructions.
6086 The third expression is also absolute, and is also optional. If it is present,
6087 it is the maximum number of bytes that should be skipped by this alignment
6088 directive. If doing the alignment would require skipping more bytes than the
6089 specified maximum, then the alignment is not done at all. You can omit the
6090 fill value (the second argument) entirely by simply using two commas after the
6091 required alignment; this can be useful if you want the alignment to be filled
6092 with no-op instructions when appropriate.
6094 @cindex @code{p2alignw} directive
6095 @cindex @code{p2alignl} directive
6096 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6097 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6098 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6099 fill pattern as a four byte longword value. For example, @code{.p2alignw
6100 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6101 filled in with the value 0x368d (the exact placement of the bytes depends upon
6102 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6107 @section @code{.popsection}
6109 @cindex @code{popsection} directive
6110 @cindex Section Stack
6111 This is one of the ELF section stack manipulation directives. The others are
6112 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6113 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6116 This directive replaces the current section (and subsection) with the top
6117 section (and subsection) on the section stack. This section is popped off the
6123 @section @code{.previous}
6125 @cindex @code{previous} directive
6126 @cindex Section Stack
6127 This is one of the ELF section stack manipulation directives. The others are
6128 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6129 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6130 (@pxref{PopSection}).
6132 This directive swaps the current section (and subsection) with most recently
6133 referenced section/subsection pair prior to this one. Multiple
6134 @code{.previous} directives in a row will flip between two sections (and their
6135 subsections). For example:
6147 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6153 # Now in section A subsection 1
6157 # Now in section B subsection 0
6160 # Now in section B subsection 1
6163 # Now in section B subsection 0
6167 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6168 section B and 0x9abc into subsection 1 of section B.
6170 In terms of the section stack, this directive swaps the current section with
6171 the top section on the section stack.
6175 @section @code{.print @var{string}}
6177 @cindex @code{print} directive
6178 @command{@value{AS}} will print @var{string} on the standard output during
6179 assembly. You must put @var{string} in double quotes.
6183 @section @code{.protected @var{names}}
6185 @cindex @code{protected} directive
6187 This is one of the ELF visibility directives. The other two are
6188 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6190 This directive overrides the named symbols default visibility (which is set by
6191 their binding: local, global or weak). The directive sets the visibility to
6192 @code{protected} which means that any references to the symbols from within the
6193 components that defines them must be resolved to the definition in that
6194 component, even if a definition in another component would normally preempt
6199 @section @code{.psize @var{lines} , @var{columns}}
6201 @cindex @code{psize} directive
6202 @cindex listing control: paper size
6203 @cindex paper size, for listings
6204 Use this directive to declare the number of lines---and, optionally, the
6205 number of columns---to use for each page, when generating listings.
6207 If you do not use @code{.psize}, listings use a default line-count
6208 of 60. You may omit the comma and @var{columns} specification; the
6209 default width is 200 columns.
6211 @command{@value{AS}} generates formfeeds whenever the specified number of
6212 lines is exceeded (or whenever you explicitly request one, using
6215 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6216 those explicitly specified with @code{.eject}.
6219 @section @code{.purgem @var{name}}
6221 @cindex @code{purgem} directive
6222 Undefine the macro @var{name}, so that later uses of the string will not be
6223 expanded. @xref{Macro}.
6227 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6229 @cindex @code{pushsection} directive
6230 @cindex Section Stack
6231 This is one of the ELF section stack manipulation directives. The others are
6232 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6233 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6236 This directive pushes the current section (and subsection) onto the
6237 top of the section stack, and then replaces the current section and
6238 subsection with @code{name} and @code{subsection}. The optional
6239 @code{flags}, @code{type} and @code{arguments} are treated the same
6240 as in the @code{.section} (@pxref{Section}) directive.
6244 @section @code{.quad @var{bignums}}
6246 @cindex @code{quad} directive
6247 @code{.quad} expects zero or more bignums, separated by commas. For
6248 each bignum, it emits
6250 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6251 warning message; and just takes the lowest order 8 bytes of the bignum.
6252 @cindex eight-byte integer
6253 @cindex integer, 8-byte
6255 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6256 hence @emph{quad}-word for 8 bytes.
6259 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6260 warning message; and just takes the lowest order 16 bytes of the bignum.
6261 @cindex sixteen-byte integer
6262 @cindex integer, 16-byte
6266 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6268 @cindex @code{reloc} directive
6269 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6270 @var{expression}. If @var{offset} is a number, the relocation is generated in
6271 the current section. If @var{offset} is an expression that resolves to a
6272 symbol plus offset, the relocation is generated in the given symbol's section.
6273 @var{expression}, if present, must resolve to a symbol plus addend or to an
6274 absolute value, but note that not all targets support an addend. e.g. ELF REL
6275 targets such as i386 store an addend in the section contents rather than in the
6276 relocation. This low level interface does not support addends stored in the
6280 @section @code{.rept @var{count}}
6282 @cindex @code{rept} directive
6283 Repeat the sequence of lines between the @code{.rept} directive and the next
6284 @code{.endr} directive @var{count} times.
6286 For example, assembling
6294 is equivalent to assembling
6303 @section @code{.sbttl "@var{subheading}"}
6305 @cindex @code{sbttl} directive
6306 @cindex subtitles for listings
6307 @cindex listing control: subtitle
6308 Use @var{subheading} as the title (third line, immediately after the
6309 title line) when generating assembly listings.
6311 This directive affects subsequent pages, as well as the current page if
6312 it appears within ten lines of the top of a page.
6316 @section @code{.scl @var{class}}
6318 @cindex @code{scl} directive
6319 @cindex symbol storage class (COFF)
6320 @cindex COFF symbol storage class
6321 Set the storage-class value for a symbol. This directive may only be
6322 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6323 whether a symbol is static or external, or it may record further
6324 symbolic debugging information.
6327 The @samp{.scl} directive is primarily associated with COFF output; when
6328 configured to generate @code{b.out} output format, @command{@value{AS}}
6329 accepts this directive but ignores it.
6335 @section @code{.section @var{name}}
6337 @cindex named section
6338 Use the @code{.section} directive to assemble the following code into a section
6341 This directive is only supported for targets that actually support arbitrarily
6342 named sections; on @code{a.out} targets, for example, it is not accepted, even
6343 with a standard @code{a.out} section name.
6347 @c only print the extra heading if both COFF and ELF are set
6348 @subheading COFF Version
6351 @cindex @code{section} directive (COFF version)
6352 For COFF targets, the @code{.section} directive is used in one of the following
6356 .section @var{name}[, "@var{flags}"]
6357 .section @var{name}[, @var{subsection}]
6360 If the optional argument is quoted, it is taken as flags to use for the
6361 section. Each flag is a single character. The following flags are recognized:
6365 bss section (uninitialized data)
6367 section is not loaded
6373 exclude section from linking
6379 shared section (meaningful for PE targets)
6381 ignored. (For compatibility with the ELF version)
6383 section is not readable (meaningful for PE targets)
6385 single-digit power-of-two section alignment (GNU extension)
6388 If no flags are specified, the default flags depend upon the section name. If
6389 the section name is not recognized, the default will be for the section to be
6390 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6391 from the section, rather than adding them, so if they are used on their own it
6392 will be as if no flags had been specified at all.
6394 If the optional argument to the @code{.section} directive is not quoted, it is
6395 taken as a subsection number (@pxref{Sub-Sections}).
6400 @c only print the extra heading if both COFF and ELF are set
6401 @subheading ELF Version
6404 @cindex Section Stack
6405 This is one of the ELF section stack manipulation directives. The others are
6406 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6407 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6408 @code{.previous} (@pxref{Previous}).
6410 @cindex @code{section} directive (ELF version)
6411 For ELF targets, the @code{.section} directive is used like this:
6414 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6417 @anchor{Section Name Substitutions}
6418 @kindex --sectname-subst
6419 @cindex section name substitution
6420 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6421 argument may contain a substitution sequence. Only @code{%S} is supported
6422 at the moment, and substitutes the current section name. For example:
6425 .macro exception_code
6426 .section %S.exception
6427 [exception code here]
6442 The two @code{exception_code} invocations above would create the
6443 @code{.text.exception} and @code{.init.exception} sections respectively.
6444 This is useful e.g. to discriminate between anciliary sections that are
6445 tied to setup code to be discarded after use from anciliary sections that
6446 need to stay resident without having to define multiple @code{exception_code}
6447 macros just for that purpose.
6449 The optional @var{flags} argument is a quoted string which may contain any
6450 combination of the following characters:
6454 section is allocatable
6456 section is excluded from executable and shared library.
6460 section is executable
6462 section is mergeable
6464 section contains zero terminated strings
6466 section is a member of a section group
6468 section is used for thread-local-storage
6470 section is a member of the previously-current section's group, if any
6471 @item @code{<number>}
6472 a numeric value indicating the bits to be set in the ELF section header's flags
6473 field. Note - if one or more of the alphabetic characters described above is
6474 also included in the flags field, their bit values will be ORed into the
6476 @item @code{<target specific>}
6477 some targets extend this list with their own flag characters
6480 Note - once a section's flags have been set they cannot be changed. There are
6481 a few exceptions to this rule however. Processor and application specific
6482 flags can be added to an already defined section. The @code{.interp},
6483 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6484 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6485 section may have the executable (@code{x}) flag added.
6487 The optional @var{type} argument may contain one of the following constants:
6491 section contains data
6493 section does not contain data (i.e., section only occupies space)
6495 section contains data which is used by things other than the program
6497 section contains an array of pointers to init functions
6499 section contains an array of pointers to finish functions
6500 @item @@preinit_array
6501 section contains an array of pointers to pre-init functions
6502 @item @@@code{<number>}
6503 a numeric value to be set as the ELF section header's type field.
6504 @item @@@code{<target specific>}
6505 some targets extend this list with their own types
6508 Many targets only support the first three section types. The type may be
6509 enclosed in double quotes if necessary.
6511 Note on targets where the @code{@@} character is the start of a comment (eg
6512 ARM) then another character is used instead. For example the ARM port uses the
6515 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6516 special and have fixed types. Any attempt to declare them with a different
6517 type will generate an error from the assembler.
6519 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6520 be specified as well as an extra argument---@var{entsize}---like this:
6523 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6526 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6527 constants, each @var{entsize} octets long. Sections with both @code{M} and
6528 @code{S} must contain zero terminated strings where each character is
6529 @var{entsize} bytes long. The linker may remove duplicates within sections with
6530 the same name, same entity size and same flags. @var{entsize} must be an
6531 absolute expression. For sections with both @code{M} and @code{S}, a string
6532 which is a suffix of a larger string is considered a duplicate. Thus
6533 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6534 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6536 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6537 be present along with an additional field like this:
6540 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6543 The @var{GroupName} field specifies the name of the section group to which this
6544 particular section belongs. The optional linkage field can contain:
6548 indicates that only one copy of this section should be retained
6553 Note: if both the @var{M} and @var{G} flags are present then the fields for
6554 the Merge flag should come first, like this:
6557 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6560 If @var{flags} contains the @code{?} symbol then it may not also contain the
6561 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6562 present. Instead, @code{?} says to consider the section that's current before
6563 this directive. If that section used @code{G}, then the new section will use
6564 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6565 If not, then the @code{?} symbol has no effect.
6567 If no flags are specified, the default flags depend upon the section name. If
6568 the section name is not recognized, the default will be for the section to have
6569 none of the above flags: it will not be allocated in memory, nor writable, nor
6570 executable. The section will contain data.
6572 For ELF targets, the assembler supports another type of @code{.section}
6573 directive for compatibility with the Solaris assembler:
6576 .section "@var{name}"[, @var{flags}...]
6579 Note that the section name is quoted. There may be a sequence of comma
6584 section is allocatable
6588 section is executable
6590 section is excluded from executable and shared library.
6592 section is used for thread local storage
6595 This directive replaces the current section and subsection. See the
6596 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6597 some examples of how this directive and the other section stack directives
6603 @section @code{.set @var{symbol}, @var{expression}}
6605 @cindex @code{set} directive
6606 @cindex symbol value, setting
6607 Set the value of @var{symbol} to @var{expression}. This
6608 changes @var{symbol}'s value and type to conform to
6609 @var{expression}. If @var{symbol} was flagged as external, it remains
6610 flagged (@pxref{Symbol Attributes}).
6612 You may @code{.set} a symbol many times in the same assembly provided that the
6613 values given to the symbol are constants. Values that are based on expressions
6614 involving other symbols are allowed, but some targets may restrict this to only
6615 being done once per assembly. This is because those targets do not set the
6616 addresses of symbols at assembly time, but rather delay the assignment until a
6617 final link is performed. This allows the linker a chance to change the code in
6618 the files, changing the location of, and the relative distance between, various
6621 If you @code{.set} a global symbol, the value stored in the object
6622 file is the last value stored into it.
6625 On Z80 @code{set} is a real instruction, use
6626 @samp{@var{symbol} defl @var{expression}} instead.
6630 @section @code{.short @var{expressions}}
6632 @cindex @code{short} directive
6634 @code{.short} is normally the same as @samp{.word}.
6635 @xref{Word,,@code{.word}}.
6637 In some configurations, however, @code{.short} and @code{.word} generate
6638 numbers of different lengths. @xref{Machine Dependencies}.
6642 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6645 This expects zero or more @var{expressions}, and emits
6646 a 16 bit number for each.
6651 @section @code{.single @var{flonums}}
6653 @cindex @code{single} directive
6654 @cindex floating point numbers (single)
6655 This directive assembles zero or more flonums, separated by commas. It
6656 has the same effect as @code{.float}.
6658 The exact kind of floating point numbers emitted depends on how
6659 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6663 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6664 numbers in @sc{ieee} format.
6670 @section @code{.size}
6672 This directive is used to set the size associated with a symbol.
6676 @c only print the extra heading if both COFF and ELF are set
6677 @subheading COFF Version
6680 @cindex @code{size} directive (COFF version)
6681 For COFF targets, the @code{.size} directive is only permitted inside
6682 @code{.def}/@code{.endef} pairs. It is used like this:
6685 .size @var{expression}
6689 @samp{.size} is only meaningful when generating COFF format output; when
6690 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6697 @c only print the extra heading if both COFF and ELF are set
6698 @subheading ELF Version
6701 @cindex @code{size} directive (ELF version)
6702 For ELF targets, the @code{.size} directive is used like this:
6705 .size @var{name} , @var{expression}
6708 This directive sets the size associated with a symbol @var{name}.
6709 The size in bytes is computed from @var{expression} which can make use of label
6710 arithmetic. This directive is typically used to set the size of function
6715 @ifclear no-space-dir
6717 @section @code{.skip @var{size} , @var{fill}}
6719 @cindex @code{skip} directive
6720 @cindex filling memory
6721 This directive emits @var{size} bytes, each of value @var{fill}. Both
6722 @var{size} and @var{fill} are absolute expressions. If the comma and
6723 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6728 @section @code{.sleb128 @var{expressions}}
6730 @cindex @code{sleb128} directive
6731 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6732 compact, variable length representation of numbers used by the DWARF
6733 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6735 @ifclear no-space-dir
6737 @section @code{.space @var{size} , @var{fill}}
6739 @cindex @code{space} directive
6740 @cindex filling memory
6741 This directive emits @var{size} bytes, each of value @var{fill}. Both
6742 @var{size} and @var{fill} are absolute expressions. If the comma
6743 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6748 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6749 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6750 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6751 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6759 @section @code{.stabd, .stabn, .stabs}
6761 @cindex symbolic debuggers, information for
6762 @cindex @code{stab@var{x}} directives
6763 There are three directives that begin @samp{.stab}.
6764 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6765 The symbols are not entered in the @command{@value{AS}} hash table: they
6766 cannot be referenced elsewhere in the source file.
6767 Up to five fields are required:
6771 This is the symbol's name. It may contain any character except
6772 @samp{\000}, so is more general than ordinary symbol names. Some
6773 debuggers used to code arbitrarily complex structures into symbol names
6777 An absolute expression. The symbol's type is set to the low 8 bits of
6778 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6779 and debuggers choke on silly bit patterns.
6782 An absolute expression. The symbol's ``other'' attribute is set to the
6783 low 8 bits of this expression.
6786 An absolute expression. The symbol's descriptor is set to the low 16
6787 bits of this expression.
6790 An absolute expression which becomes the symbol's value.
6793 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6794 or @code{.stabs} statement, the symbol has probably already been created;
6795 you get a half-formed symbol in your object file. This is
6796 compatible with earlier assemblers!
6799 @cindex @code{stabd} directive
6800 @item .stabd @var{type} , @var{other} , @var{desc}
6802 The ``name'' of the symbol generated is not even an empty string.
6803 It is a null pointer, for compatibility. Older assemblers used a
6804 null pointer so they didn't waste space in object files with empty
6807 The symbol's value is set to the location counter,
6808 relocatably. When your program is linked, the value of this symbol
6809 is the address of the location counter when the @code{.stabd} was
6812 @cindex @code{stabn} directive
6813 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6814 The name of the symbol is set to the empty string @code{""}.
6816 @cindex @code{stabs} directive
6817 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6818 All five fields are specified.
6824 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6825 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6827 @cindex string, copying to object file
6828 @cindex string8, copying to object file
6829 @cindex string16, copying to object file
6830 @cindex string32, copying to object file
6831 @cindex string64, copying to object file
6832 @cindex @code{string} directive
6833 @cindex @code{string8} directive
6834 @cindex @code{string16} directive
6835 @cindex @code{string32} directive
6836 @cindex @code{string64} directive
6838 Copy the characters in @var{str} to the object file. You may specify more than
6839 one string to copy, separated by commas. Unless otherwise specified for a
6840 particular machine, the assembler marks the end of each string with a 0 byte.
6841 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6843 The variants @code{string16}, @code{string32} and @code{string64} differ from
6844 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6845 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6846 are stored in target endianness byte order.
6852 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6853 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6858 @section @code{.struct @var{expression}}
6860 @cindex @code{struct} directive
6861 Switch to the absolute section, and set the section offset to @var{expression},
6862 which must be an absolute expression. You might use this as follows:
6871 This would define the symbol @code{field1} to have the value 0, the symbol
6872 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6873 value 8. Assembly would be left in the absolute section, and you would need to
6874 use a @code{.section} directive of some sort to change to some other section
6875 before further assembly.
6879 @section @code{.subsection @var{name}}
6881 @cindex @code{subsection} directive
6882 @cindex Section Stack
6883 This is one of the ELF section stack manipulation directives. The others are
6884 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6885 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6888 This directive replaces the current subsection with @code{name}. The current
6889 section is not changed. The replaced subsection is put onto the section stack
6890 in place of the then current top of stack subsection.
6895 @section @code{.symver}
6896 @cindex @code{symver} directive
6897 @cindex symbol versioning
6898 @cindex versions of symbols
6899 Use the @code{.symver} directive to bind symbols to specific version nodes
6900 within a source file. This is only supported on ELF platforms, and is
6901 typically used when assembling files to be linked into a shared library.
6902 There are cases where it may make sense to use this in objects to be bound
6903 into an application itself so as to override a versioned symbol from a
6906 For ELF targets, the @code{.symver} directive can be used like this:
6908 .symver @var{name}, @var{name2@@nodename}
6910 If the symbol @var{name} is defined within the file
6911 being assembled, the @code{.symver} directive effectively creates a symbol
6912 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6913 just don't try and create a regular alias is that the @var{@@} character isn't
6914 permitted in symbol names. The @var{name2} part of the name is the actual name
6915 of the symbol by which it will be externally referenced. The name @var{name}
6916 itself is merely a name of convenience that is used so that it is possible to
6917 have definitions for multiple versions of a function within a single source
6918 file, and so that the compiler can unambiguously know which version of a
6919 function is being mentioned. The @var{nodename} portion of the alias should be
6920 the name of a node specified in the version script supplied to the linker when
6921 building a shared library. If you are attempting to override a versioned
6922 symbol from a shared library, then @var{nodename} should correspond to the
6923 nodename of the symbol you are trying to override.
6925 If the symbol @var{name} is not defined within the file being assembled, all
6926 references to @var{name} will be changed to @var{name2@@nodename}. If no
6927 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6930 Another usage of the @code{.symver} directive is:
6932 .symver @var{name}, @var{name2@@@@nodename}
6934 In this case, the symbol @var{name} must exist and be defined within
6935 the file being assembled. It is similar to @var{name2@@nodename}. The
6936 difference is @var{name2@@@@nodename} will also be used to resolve
6937 references to @var{name2} by the linker.
6939 The third usage of the @code{.symver} directive is:
6941 .symver @var{name}, @var{name2@@@@@@nodename}
6943 When @var{name} is not defined within the
6944 file being assembled, it is treated as @var{name2@@nodename}. When
6945 @var{name} is defined within the file being assembled, the symbol
6946 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6951 @section @code{.tag @var{structname}}
6953 @cindex COFF structure debugging
6954 @cindex structure debugging, COFF
6955 @cindex @code{tag} directive
6956 This directive is generated by compilers to include auxiliary debugging
6957 information in the symbol table. It is only permitted inside
6958 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6959 definitions in the symbol table with instances of those structures.
6962 @samp{.tag} is only used when generating COFF format output; when
6963 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6969 @section @code{.text @var{subsection}}
6971 @cindex @code{text} directive
6972 Tells @command{@value{AS}} to assemble the following statements onto the end of
6973 the text subsection numbered @var{subsection}, which is an absolute
6974 expression. If @var{subsection} is omitted, subsection number zero
6978 @section @code{.title "@var{heading}"}
6980 @cindex @code{title} directive
6981 @cindex listing control: title line
6982 Use @var{heading} as the title (second line, immediately after the
6983 source file name and pagenumber) when generating assembly listings.
6985 This directive affects subsequent pages, as well as the current page if
6986 it appears within ten lines of the top of a page.
6990 @section @code{.type}
6992 This directive is used to set the type of a symbol.
6996 @c only print the extra heading if both COFF and ELF are set
6997 @subheading COFF Version
7000 @cindex COFF symbol type
7001 @cindex symbol type, COFF
7002 @cindex @code{type} directive (COFF version)
7003 For COFF targets, this directive is permitted only within
7004 @code{.def}/@code{.endef} pairs. It is used like this:
7010 This records the integer @var{int} as the type attribute of a symbol table
7014 @samp{.type} is associated only with COFF format output; when
7015 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
7016 directive but ignores it.
7022 @c only print the extra heading if both COFF and ELF are set
7023 @subheading ELF Version
7026 @cindex ELF symbol type
7027 @cindex symbol type, ELF
7028 @cindex @code{type} directive (ELF version)
7029 For ELF targets, the @code{.type} directive is used like this:
7032 .type @var{name} , @var{type description}
7035 This sets the type of symbol @var{name} to be either a
7036 function symbol or an object symbol. There are five different syntaxes
7037 supported for the @var{type description} field, in order to provide
7038 compatibility with various other assemblers.
7040 Because some of the characters used in these syntaxes (such as @samp{@@} and
7041 @samp{#}) are comment characters for some architectures, some of the syntaxes
7042 below do not work on all architectures. The first variant will be accepted by
7043 the GNU assembler on all architectures so that variant should be used for
7044 maximum portability, if you do not need to assemble your code with other
7047 The syntaxes supported are:
7050 .type <name> STT_<TYPE_IN_UPPER_CASE>
7051 .type <name>,#<type>
7052 .type <name>,@@<type>
7053 .type <name>,%<type>
7054 .type <name>,"<type>"
7057 The types supported are:
7062 Mark the symbol as being a function name.
7065 @itemx gnu_indirect_function
7066 Mark the symbol as an indirect function when evaluated during reloc
7067 processing. (This is only supported on assemblers targeting GNU systems).
7071 Mark the symbol as being a data object.
7075 Mark the symbol as being a thead-local data object.
7079 Mark the symbol as being a common data object.
7083 Does not mark the symbol in any way. It is supported just for completeness.
7085 @item gnu_unique_object
7086 Marks the symbol as being a globally unique data object. The dynamic linker
7087 will make sure that in the entire process there is just one symbol with this
7088 name and type in use. (This is only supported on assemblers targeting GNU
7093 Note: Some targets support extra types in addition to those listed above.
7099 @section @code{.uleb128 @var{expressions}}
7101 @cindex @code{uleb128} directive
7102 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7103 compact, variable length representation of numbers used by the DWARF
7104 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7108 @section @code{.val @var{addr}}
7110 @cindex @code{val} directive
7111 @cindex COFF value attribute
7112 @cindex value attribute, COFF
7113 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7114 records the address @var{addr} as the value attribute of a symbol table
7118 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
7119 configured for @code{b.out}, it accepts this directive but ignores it.
7125 @section @code{.version "@var{string}"}
7127 @cindex @code{version} directive
7128 This directive creates a @code{.note} section and places into it an ELF
7129 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7134 @section @code{.vtable_entry @var{table}, @var{offset}}
7136 @cindex @code{vtable_entry} directive
7137 This directive finds or creates a symbol @code{table} and creates a
7138 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7141 @section @code{.vtable_inherit @var{child}, @var{parent}}
7143 @cindex @code{vtable_inherit} directive
7144 This directive finds the symbol @code{child} and finds or creates the symbol
7145 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7146 parent whose addend is the value of the child symbol. As a special case the
7147 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7151 @section @code{.warning "@var{string}"}
7152 @cindex warning directive
7153 Similar to the directive @code{.error}
7154 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7157 @section @code{.weak @var{names}}
7159 @cindex @code{weak} directive
7160 This directive sets the weak attribute on the comma separated list of symbol
7161 @code{names}. If the symbols do not already exist, they will be created.
7163 On COFF targets other than PE, weak symbols are a GNU extension. This
7164 directive sets the weak attribute on the comma separated list of symbol
7165 @code{names}. If the symbols do not already exist, they will be created.
7167 On the PE target, weak symbols are supported natively as weak aliases.
7168 When a weak symbol is created that is not an alias, GAS creates an
7169 alternate symbol to hold the default value.
7172 @section @code{.weakref @var{alias}, @var{target}}
7174 @cindex @code{weakref} directive
7175 This directive creates an alias to the target symbol that enables the symbol to
7176 be referenced with weak-symbol semantics, but without actually making it weak.
7177 If direct references or definitions of the symbol are present, then the symbol
7178 will not be weak, but if all references to it are through weak references, the
7179 symbol will be marked as weak in the symbol table.
7181 The effect is equivalent to moving all references to the alias to a separate
7182 assembly source file, renaming the alias to the symbol in it, declaring the
7183 symbol as weak there, and running a reloadable link to merge the object files
7184 resulting from the assembly of the new source file and the old source file that
7185 had the references to the alias removed.
7187 The alias itself never makes to the symbol table, and is entirely handled
7188 within the assembler.
7191 @section @code{.word @var{expressions}}
7193 @cindex @code{word} directive
7194 This directive expects zero or more @var{expressions}, of any section,
7195 separated by commas.
7198 For each expression, @command{@value{AS}} emits a 32-bit number.
7201 For each expression, @command{@value{AS}} emits a 16-bit number.
7206 The size of the number emitted, and its byte order,
7207 depend on what target computer the assembly is for.
7210 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
7211 @c happen---32-bit addressability, period; no long/short jumps.
7212 @ifset DIFF-TBL-KLUGE
7213 @cindex difference tables altered
7214 @cindex altered difference tables
7216 @emph{Warning: Special Treatment to support Compilers}
7220 Machines with a 32-bit address space, but that do less than 32-bit
7221 addressing, require the following special treatment. If the machine of
7222 interest to you does 32-bit addressing (or doesn't require it;
7223 @pxref{Machine Dependencies}), you can ignore this issue.
7226 In order to assemble compiler output into something that works,
7227 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7228 Directives of the form @samp{.word sym1-sym2} are often emitted by
7229 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7230 directive of the form @samp{.word sym1-sym2}, and the difference between
7231 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7232 creates a @dfn{secondary jump table}, immediately before the next label.
7233 This secondary jump table is preceded by a short-jump to the
7234 first byte after the secondary table. This short-jump prevents the flow
7235 of control from accidentally falling into the new table. Inside the
7236 table is a long-jump to @code{sym2}. The original @samp{.word}
7237 contains @code{sym1} minus the address of the long-jump to
7240 If there were several occurrences of @samp{.word sym1-sym2} before the
7241 secondary jump table, all of them are adjusted. If there was a
7242 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7243 long-jump to @code{sym4} is included in the secondary jump table,
7244 and the @code{.word} directives are adjusted to contain @code{sym3}
7245 minus the address of the long-jump to @code{sym4}; and so on, for as many
7246 entries in the original jump table as necessary.
7249 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7250 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7251 assembly language programmers.
7254 @c end DIFF-TBL-KLUGE
7256 @ifclear no-space-dir
7258 @section @code{.zero @var{size}}
7260 @cindex @code{zero} directive
7261 @cindex filling memory with zero bytes
7262 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7263 expression. This directive is actually an alias for the @samp{.skip} directive
7264 so in can take an optional second argument of the value to store in the bytes
7265 instead of zero. Using @samp{.zero} in this way would be confusing however.
7269 @section Deprecated Directives
7271 @cindex deprecated directives
7272 @cindex obsolescent directives
7273 One day these directives won't work.
7274 They are included for compatibility with older assemblers.
7281 @node Object Attributes
7282 @chapter Object Attributes
7283 @cindex object attributes
7285 @command{@value{AS}} assembles source files written for a specific architecture
7286 into object files for that architecture. But not all object files are alike.
7287 Many architectures support incompatible variations. For instance, floating
7288 point arguments might be passed in floating point registers if the object file
7289 requires hardware floating point support---or floating point arguments might be
7290 passed in integer registers if the object file supports processors with no
7291 hardware floating point unit. Or, if two objects are built for different
7292 generations of the same architecture, the combination may require the
7293 newer generation at run-time.
7295 This information is useful during and after linking. At link time,
7296 @command{@value{LD}} can warn about incompatible object files. After link
7297 time, tools like @command{gdb} can use it to process the linked file
7300 Compatibility information is recorded as a series of object attributes. Each
7301 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7302 string, and indicates who sets the meaning of the tag. The tag is an integer,
7303 and indicates what property the attribute describes. The value may be a string
7304 or an integer, and indicates how the property affects this object. Missing
7305 attributes are the same as attributes with a zero value or empty string value.
7307 Object attributes were developed as part of the ABI for the ARM Architecture.
7308 The file format is documented in @cite{ELF for the ARM Architecture}.
7311 * GNU Object Attributes:: @sc{gnu} Object Attributes
7312 * Defining New Object Attributes:: Defining New Object Attributes
7315 @node GNU Object Attributes
7316 @section @sc{gnu} Object Attributes
7318 The @code{.gnu_attribute} directive records an object attribute
7319 with vendor @samp{gnu}.
7321 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7322 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7323 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7324 2} is set for architecture-independent attributes and clear for
7325 architecture-dependent ones.
7327 @subsection Common @sc{gnu} attributes
7329 These attributes are valid on all architectures.
7332 @item Tag_compatibility (32)
7333 The compatibility attribute takes an integer flag value and a vendor name. If
7334 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7335 then the file is only compatible with the named toolchain. If it is greater
7336 than 1, the file can only be processed by other toolchains under some private
7337 arrangement indicated by the flag value and the vendor name.
7340 @subsection MIPS Attributes
7343 @item Tag_GNU_MIPS_ABI_FP (4)
7344 The floating-point ABI used by this object file. The value will be:
7348 0 for files not affected by the floating-point ABI.
7350 1 for files using the hardware floating-point ABI with a standard
7351 double-precision FPU.
7353 2 for files using the hardware floating-point ABI with a single-precision FPU.
7355 3 for files using the software floating-point ABI.
7357 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7358 floating-point registers, 32-bit general-purpose registers and increased the
7359 number of callee-saved floating-point registers.
7361 5 for files using the hardware floating-point ABI with a double-precision FPU
7362 with either 32-bit or 64-bit floating-point registers and 32-bit
7363 general-purpose registers.
7365 6 for files using the hardware floating-point ABI with 64-bit floating-point
7366 registers and 32-bit general-purpose registers.
7368 7 for files using the hardware floating-point ABI with 64-bit floating-point
7369 registers, 32-bit general-purpose registers and a rule that forbids the
7370 direct use of odd-numbered single-precision floating-point registers.
7374 @subsection PowerPC Attributes
7377 @item Tag_GNU_Power_ABI_FP (4)
7378 The floating-point ABI used by this object file. The value will be:
7382 0 for files not affected by the floating-point ABI.
7384 1 for files using double-precision hardware floating-point ABI.
7386 2 for files using the software floating-point ABI.
7388 3 for files using single-precision hardware floating-point ABI.
7391 @item Tag_GNU_Power_ABI_Vector (8)
7392 The vector ABI used by this object file. The value will be:
7396 0 for files not affected by the vector ABI.
7398 1 for files using general purpose registers to pass vectors.
7400 2 for files using AltiVec registers to pass vectors.
7402 3 for files using SPE registers to pass vectors.
7406 @subsection IBM z Systems Attributes
7409 @item Tag_GNU_S390_ABI_Vector (8)
7410 The vector ABI used by this object file. The value will be:
7414 0 for files not affected by the vector ABI.
7416 1 for files using software vector ABI.
7418 2 for files using hardware vector ABI.
7422 @node Defining New Object Attributes
7423 @section Defining New Object Attributes
7425 If you want to define a new @sc{gnu} object attribute, here are the places you
7426 will need to modify. New attributes should be discussed on the @samp{binutils}
7431 This manual, which is the official register of attributes.
7433 The header for your architecture @file{include/elf}, to define the tag.
7435 The @file{bfd} support file for your architecture, to merge the attribute
7436 and issue any appropriate link warnings.
7438 Test cases in @file{ld/testsuite} for merging and link warnings.
7440 @file{binutils/readelf.c} to display your attribute.
7442 GCC, if you want the compiler to mark the attribute automatically.
7448 @node Machine Dependencies
7449 @chapter Machine Dependent Features
7451 @cindex machine dependencies
7452 The machine instruction sets are (almost by definition) different on
7453 each machine where @command{@value{AS}} runs. Floating point representations
7454 vary as well, and @command{@value{AS}} often supports a few additional
7455 directives or command-line options for compatibility with other
7456 assemblers on a particular platform. Finally, some versions of
7457 @command{@value{AS}} support special pseudo-instructions for branch
7460 This chapter discusses most of these differences, though it does not
7461 include details on any machine's instruction set. For details on that
7462 subject, see the hardware manufacturer's manual.
7466 * AArch64-Dependent:: AArch64 Dependent Features
7469 * Alpha-Dependent:: Alpha Dependent Features
7472 * ARC-Dependent:: ARC Dependent Features
7475 * ARM-Dependent:: ARM Dependent Features
7478 * AVR-Dependent:: AVR Dependent Features
7481 * Blackfin-Dependent:: Blackfin Dependent Features
7484 * CR16-Dependent:: CR16 Dependent Features
7487 * CRIS-Dependent:: CRIS Dependent Features
7490 * D10V-Dependent:: D10V Dependent Features
7493 * D30V-Dependent:: D30V Dependent Features
7496 * Epiphany-Dependent:: EPIPHANY Dependent Features
7499 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7502 * HPPA-Dependent:: HPPA Dependent Features
7505 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
7508 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7511 * i860-Dependent:: Intel 80860 Dependent Features
7514 * i960-Dependent:: Intel 80960 Dependent Features
7517 * IA-64-Dependent:: Intel IA-64 Dependent Features
7520 * IP2K-Dependent:: IP2K Dependent Features
7523 * LM32-Dependent:: LM32 Dependent Features
7526 * M32C-Dependent:: M32C Dependent Features
7529 * M32R-Dependent:: M32R Dependent Features
7532 * M68K-Dependent:: M680x0 Dependent Features
7535 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7538 * Meta-Dependent :: Meta Dependent Features
7541 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7544 * MIPS-Dependent:: MIPS Dependent Features
7547 * MMIX-Dependent:: MMIX Dependent Features
7550 * MSP430-Dependent:: MSP430 Dependent Features
7553 * NDS32-Dependent:: Andes NDS32 Dependent Features
7556 * NiosII-Dependent:: Altera Nios II Dependent Features
7559 * NS32K-Dependent:: NS32K Dependent Features
7562 * PDP-11-Dependent:: PDP-11 Dependent Features
7565 * PJ-Dependent:: picoJava Dependent Features
7568 * PPC-Dependent:: PowerPC Dependent Features
7571 * RL78-Dependent:: RL78 Dependent Features
7574 * RX-Dependent:: RX Dependent Features
7577 * S/390-Dependent:: IBM S/390 Dependent Features
7580 * SCORE-Dependent:: SCORE Dependent Features
7583 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7584 * SH64-Dependent:: SuperH SH64 Dependent Features
7587 * Sparc-Dependent:: SPARC Dependent Features
7590 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7593 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7596 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7599 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7602 * V850-Dependent:: V850 Dependent Features
7605 * Vax-Dependent:: VAX Dependent Features
7608 * Visium-Dependent:: Visium Dependent Features
7611 * XGATE-Dependent:: XGATE Features
7614 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7617 * Xtensa-Dependent:: Xtensa Dependent Features
7620 * Z80-Dependent:: Z80 Dependent Features
7623 * Z8000-Dependent:: Z8000 Dependent Features
7630 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7631 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7632 @c peculiarity: to preserve cross-references, there must be a node called
7633 @c "Machine Dependencies". Hence the conditional nodenames in each
7634 @c major node below. Node defaulting in makeinfo requires adjacency of
7635 @c node and sectioning commands; hence the repetition of @chapter BLAH
7636 @c in both conditional blocks.
7639 @include c-aarch64.texi
7643 @include c-alpha.texi
7659 @include c-bfin.texi
7663 @include c-cr16.texi
7667 @include c-cris.texi
7672 @node Machine Dependencies
7673 @chapter Machine Dependent Features
7675 The machine instruction sets are different on each Renesas chip family,
7676 and there are also some syntax differences among the families. This
7677 chapter describes the specific @command{@value{AS}} features for each
7681 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7682 * SH-Dependent:: Renesas SH Dependent Features
7689 @include c-d10v.texi
7693 @include c-d30v.texi
7697 @include c-epiphany.texi
7701 @include c-h8300.texi
7705 @include c-hppa.texi
7709 @include c-i370.texi
7713 @include c-i386.texi
7717 @include c-i860.texi
7721 @include c-i960.texi
7725 @include c-ia64.texi
7729 @include c-ip2k.texi
7733 @include c-lm32.texi
7737 @include c-m32c.texi
7741 @include c-m32r.texi
7745 @include c-m68k.texi
7749 @include c-m68hc11.texi
7753 @include c-metag.texi
7757 @include c-microblaze.texi
7761 @include c-mips.texi
7765 @include c-mmix.texi
7769 @include c-msp430.texi
7773 @include c-nds32.texi
7777 @include c-nios2.texi
7781 @include c-ns32k.texi
7785 @include c-pdp11.texi
7797 @include c-rl78.texi
7805 @include c-s390.texi
7809 @include c-score.texi
7814 @include c-sh64.texi
7818 @include c-sparc.texi
7822 @include c-tic54x.texi
7826 @include c-tic6x.texi
7830 @include c-tilegx.texi
7834 @include c-tilepro.texi
7838 @include c-v850.texi
7846 @include c-visium.texi
7850 @include c-xgate.texi
7854 @include c-xstormy16.texi
7858 @include c-xtensa.texi
7870 @c reverse effect of @down at top of generic Machine-Dep chapter
7874 @node Reporting Bugs
7875 @chapter Reporting Bugs
7876 @cindex bugs in assembler
7877 @cindex reporting bugs in assembler
7879 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7881 Reporting a bug may help you by bringing a solution to your problem, or it may
7882 not. But in any case the principal function of a bug report is to help the
7883 entire community by making the next version of @command{@value{AS}} work better.
7884 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7886 In order for a bug report to serve its purpose, you must include the
7887 information that enables us to fix the bug.
7890 * Bug Criteria:: Have you found a bug?
7891 * Bug Reporting:: How to report bugs
7895 @section Have You Found a Bug?
7896 @cindex bug criteria
7898 If you are not sure whether you have found a bug, here are some guidelines:
7901 @cindex fatal signal
7902 @cindex assembler crash
7903 @cindex crash of assembler
7905 If the assembler gets a fatal signal, for any input whatever, that is a
7906 @command{@value{AS}} bug. Reliable assemblers never crash.
7908 @cindex error on valid input
7910 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7912 @cindex invalid input
7914 If @command{@value{AS}} does not produce an error message for invalid input, that
7915 is a bug. However, you should note that your idea of ``invalid input'' might
7916 be our idea of ``an extension'' or ``support for traditional practice''.
7919 If you are an experienced user of assemblers, your suggestions for improvement
7920 of @command{@value{AS}} are welcome in any case.
7924 @section How to Report Bugs
7926 @cindex assembler bugs, reporting
7928 A number of companies and individuals offer support for @sc{gnu} products. If
7929 you obtained @command{@value{AS}} from a support organization, we recommend you
7930 contact that organization first.
7932 You can find contact information for many support companies and
7933 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7937 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7941 The fundamental principle of reporting bugs usefully is this:
7942 @strong{report all the facts}. If you are not sure whether to state a
7943 fact or leave it out, state it!
7945 Often people omit facts because they think they know what causes the problem
7946 and assume that some details do not matter. Thus, you might assume that the
7947 name of a symbol you use in an example does not matter. Well, probably it does
7948 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7949 happens to fetch from the location where that name is stored in memory;
7950 perhaps, if the name were different, the contents of that location would fool
7951 the assembler into doing the right thing despite the bug. Play it safe and
7952 give a specific, complete example. That is the easiest thing for you to do,
7953 and the most helpful.
7955 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7956 it is new to us. Therefore, always write your bug reports on the assumption
7957 that the bug has not been reported previously.
7959 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7960 bell?'' This cannot help us fix a bug, so it is basically useless. We
7961 respond by asking for enough details to enable us to investigate.
7962 You might as well expedite matters by sending them to begin with.
7964 To enable us to fix the bug, you should include all these things:
7968 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
7969 it with the @samp{--version} argument.
7971 Without this, we will not know whether there is any point in looking for
7972 the bug in the current version of @command{@value{AS}}.
7975 Any patches you may have applied to the @command{@value{AS}} source.
7978 The type of machine you are using, and the operating system name and
7982 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
7986 The command arguments you gave the assembler to assemble your example and
7987 observe the bug. To guarantee you will not omit something important, list them
7988 all. A copy of the Makefile (or the output from make) is sufficient.
7990 If we were to try to guess the arguments, we would probably guess wrong
7991 and then we might not encounter the bug.
7994 A complete input file that will reproduce the bug. If the bug is observed when
7995 the assembler is invoked via a compiler, send the assembler source, not the
7996 high level language source. Most compilers will produce the assembler source
7997 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
7998 the options @samp{-v --save-temps}; this will save the assembler source in a
7999 file with an extension of @file{.s}, and also show you exactly how
8000 @command{@value{AS}} is being run.
8003 A description of what behavior you observe that you believe is
8004 incorrect. For example, ``It gets a fatal signal.''
8006 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8007 will certainly notice it. But if the bug is incorrect output, we might not
8008 notice unless it is glaringly wrong. You might as well not give us a chance to
8011 Even if the problem you experience is a fatal signal, you should still say so
8012 explicitly. Suppose something strange is going on, such as, your copy of
8013 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8014 library on your system. (This has happened!) Your copy might crash and ours
8015 would not. If you told us to expect a crash, then when ours fails to crash, we
8016 would know that the bug was not happening for us. If you had not told us to
8017 expect a crash, then we would not be able to draw any conclusion from our
8021 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8022 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8023 option. Always send diffs from the old file to the new file. If you even
8024 discuss something in the @command{@value{AS}} source, refer to it by context, not
8027 The line numbers in our development sources will not match those in your
8028 sources. Your line numbers would convey no useful information to us.
8031 Here are some things that are not necessary:
8035 A description of the envelope of the bug.
8037 Often people who encounter a bug spend a lot of time investigating
8038 which changes to the input file will make the bug go away and which
8039 changes will not affect it.
8041 This is often time consuming and not very useful, because the way we
8042 will find the bug is by running a single example under the debugger
8043 with breakpoints, not by pure deduction from a series of examples.
8044 We recommend that you save your time for something else.
8046 Of course, if you can find a simpler example to report @emph{instead}
8047 of the original one, that is a convenience for us. Errors in the
8048 output will be easier to spot, running under the debugger will take
8049 less time, and so on.
8051 However, simplification is not vital; if you do not want to do this,
8052 report the bug anyway and send us the entire test case you used.
8055 A patch for the bug.
8057 A patch for the bug does help us if it is a good one. But do not omit
8058 the necessary information, such as the test case, on the assumption that
8059 a patch is all we need. We might see problems with your patch and decide
8060 to fix the problem another way, or we might not understand it at all.
8062 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8063 construct an example that will make the program follow a certain path through
8064 the code. If you do not send us the example, we will not be able to construct
8065 one, so we will not be able to verify that the bug is fixed.
8067 And if we cannot understand what bug you are trying to fix, or why your
8068 patch should be an improvement, we will not install it. A test case will
8069 help us to understand.
8072 A guess about what the bug is or what it depends on.
8074 Such guesses are usually wrong. Even we cannot guess right about such
8075 things without first using the debugger to find the facts.
8078 @node Acknowledgements
8079 @chapter Acknowledgements
8081 If you have contributed to GAS and your name isn't listed here,
8082 it is not meant as a slight. We just don't know about it. Send mail to the
8083 maintainer, and we'll correct the situation. Currently
8085 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8087 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8090 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8091 information and the 68k series machines, most of the preprocessing pass, and
8092 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8094 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8095 many bug fixes, including merging support for several processors, breaking GAS
8096 up to handle multiple object file format back ends (including heavy rewrite,
8097 testing, an integration of the coff and b.out back ends), adding configuration
8098 including heavy testing and verification of cross assemblers and file splits
8099 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8100 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8101 port (including considerable amounts of reverse engineering), a SPARC opcode
8102 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8103 assertions and made them work, much other reorganization, cleanup, and lint.
8105 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8106 in format-specific I/O modules.
8108 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8109 has done much work with it since.
8111 The Intel 80386 machine description was written by Eliot Dresselhaus.
8113 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8115 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8116 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8118 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8119 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8120 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8121 support a.out format.
8123 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8124 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8125 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8126 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8129 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8130 simplified the configuration of which versions accept which directives. He
8131 updated the 68k machine description so that Motorola's opcodes always produced
8132 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8133 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8134 cross-compilation support, and one bug in relaxation that took a week and
8135 required the proverbial one-bit fix.
8137 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8138 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8139 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8140 PowerPC assembler, and made a few other minor patches.
8142 Steve Chamberlain made GAS able to generate listings.
8144 Hewlett-Packard contributed support for the HP9000/300.
8146 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8147 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8148 formats). This work was supported by both the Center for Software Science at
8149 the University of Utah and Cygnus Support.
8151 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8152 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8153 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8154 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8155 and some initial 64-bit support).
8157 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8159 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8160 support for openVMS/Alpha.
8162 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8165 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8166 Inc.@: added support for Xtensa processors.
8168 Several engineers at Cygnus Support have also provided many small bug fixes and
8169 configuration enhancements.
8171 Jon Beniston added support for the Lattice Mico32 architecture.
8173 Many others have contributed large or small bugfixes and enhancements. If
8174 you have contributed significant work and are not mentioned on this list, and
8175 want to be, let us know. Some of the history has been lost; we are not
8176 intentionally leaving anyone out.
8178 @node GNU Free Documentation License
8179 @appendix GNU Free Documentation License
8183 @unnumbered AS Index