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 RISC-V options:}
519 [@b{-mhard-float}|@b{-msoft-float}]
523 @emph{Target s390 options:}
524 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
525 [@b{-mregnames}|@b{-mno-regnames}]
526 [@b{-mwarn-areg-zero}]
530 @emph{Target SCORE options:}
531 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
532 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
533 [@b{-march=score7}][@b{-march=score3}]
534 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
538 @emph{Target SPARC options:}
539 @c The order here is important. See c-sparc.texi.
540 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Aleon}|@b{-Asparclet}|@b{-Asparclite}
541 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av8plusb}|@b{-Av8plusc}|@b{-Av8plusd}
542 @b{-Av8plusv}|@b{-Av8plusm}|@b{-Av9}|@b{-Av9a}|@b{-Av9b}|@b{-Av9c}
543 @b{-Av9d}|@b{-Av9e}|@b{-Av9v}|@b{-Av9m}|@b{-Asparc}|@b{-Asparcvis}
544 @b{-Asparcvis2}|@b{-Asparcfmaf}|@b{-Asparcima}|@b{-Asparcvis3}
545 @b{-Asparcvisr}|@b{-Asparc5}]
546 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}]|@b{-xarch=v8plusb}|@b{-xarch=v8plusc}
547 @b{-xarch=v8plusd}|@b{-xarch=v8plusv}|@b{-xarch=v8plusm}|@b{-xarch=v9}
548 @b{-xarch=v9a}|@b{-xarch=v9b}|@b{-xarch=v9c}|@b{-xarch=v9d}|@b{-xarch=v9e}
549 @b{-xarch=v9v}|@b{-xarch=v9m}|@b{-xarch=sparc}|@b{-xarch=sparcvis}
550 @b{-xarch=sparcvis2}|@b{-xarch=sparcfmaf}|@b{-xarch=sparcima}
551 @b{-xarch=sparcvis3}|@b{-xarch=sparcvisr}|@b{-xarch=sparc5}
554 [@b{--enforce-aligned-data}][@b{--dcti-couples-detect}]
558 @emph{Target TIC54X options:}
559 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
560 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
564 @emph{Target TIC6X options:}
565 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
566 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
567 [@b{-mpic}|@b{-mno-pic}]
571 @emph{Target TILE-Gx options:}
572 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
575 @c TILEPro has no machine-dependent assembler options
579 @emph{Target Visium options:}
580 [@b{-mtune=@var{arch}}]
584 @emph{Target Xtensa options:}
585 [@b{--[no-]text-section-literals}] [@b{--[no-]auto-litpools}]
586 [@b{--[no-]absolute-literals}]
587 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
588 [@b{--[no-]transform}]
589 [@b{--rename-section} @var{oldname}=@var{newname}]
590 [@b{--[no-]trampolines}]
594 @emph{Target Z80 options:}
595 [@b{-z80}] [@b{-r800}]
596 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
597 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
598 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
599 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
600 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
601 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
605 @c Z8000 has no machine-dependent assembler options
614 @include at-file.texi
617 Turn on listings, in any of a variety of ways:
621 omit false conditionals
624 omit debugging directives
627 include general information, like @value{AS} version and options passed
630 include high-level source
636 include macro expansions
639 omit forms processing
645 set the name of the listing file
648 You may combine these options; for example, use @samp{-aln} for assembly
649 listing without forms processing. The @samp{=file} option, if used, must be
650 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
653 Begin in alternate macro mode.
655 @xref{Altmacro,,@code{.altmacro}}.
658 @item --compress-debug-sections
659 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
660 ELF ABI. The resulting object file may not be compatible with older
661 linkers and object file utilities. Note if compression would make a
662 given section @emph{larger} then it is not compressed.
665 @cindex @samp{--compress-debug-sections=} option
666 @item --compress-debug-sections=none
667 @itemx --compress-debug-sections=zlib
668 @itemx --compress-debug-sections=zlib-gnu
669 @itemx --compress-debug-sections=zlib-gabi
670 These options control how DWARF debug sections are compressed.
671 @option{--compress-debug-sections=none} is equivalent to
672 @option{--nocompress-debug-sections}.
673 @option{--compress-debug-sections=zlib} and
674 @option{--compress-debug-sections=zlib-gabi} are equivalent to
675 @option{--compress-debug-sections}.
676 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
677 sections using zlib. The debug sections are renamed to begin with
678 @samp{.zdebug}. Note if compression would make a given section
679 @emph{larger} then it is not compressed nor renamed.
683 @item --nocompress-debug-sections
684 Do not compress DWARF debug sections. This is usually the default for all
685 targets except the x86/x86_64, but a configure time option can be used to
689 Ignored. This option is accepted for script compatibility with calls to
692 @item --debug-prefix-map @var{old}=@var{new}
693 When assembling files in directory @file{@var{old}}, record debugging
694 information describing them as in @file{@var{new}} instead.
696 @item --defsym @var{sym}=@var{value}
697 Define the symbol @var{sym} to be @var{value} before assembling the input file.
698 @var{value} must be an integer constant. As in C, a leading @samp{0x}
699 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
700 value. The value of the symbol can be overridden inside a source file via the
701 use of a @code{.set} pseudo-op.
704 ``fast''---skip whitespace and comment preprocessing (assume source is
709 Generate debugging information for each assembler source line using whichever
710 debug format is preferred by the target. This currently means either STABS,
714 Generate stabs debugging information for each assembler line. This
715 may help debugging assembler code, if the debugger can handle it.
718 Generate stabs debugging information for each assembler line, with GNU
719 extensions that probably only gdb can handle, and that could make other
720 debuggers crash or refuse to read your program. This
721 may help debugging assembler code. Currently the only GNU extension is
722 the location of the current working directory at assembling time.
725 Generate DWARF2 debugging information for each assembler line. This
726 may help debugging assembler code, if the debugger can handle it. Note---this
727 option is only supported by some targets, not all of them.
729 @item --gdwarf-sections
730 Instead of creating a .debug_line section, create a series of
731 .debug_line.@var{foo} sections where @var{foo} is the name of the
732 corresponding code section. For example a code section called @var{.text.func}
733 will have its dwarf line number information placed into a section called
734 @var{.debug_line.text.func}. If the code section is just called @var{.text}
735 then debug line section will still be called just @var{.debug_line} without any
739 @item --size-check=error
740 @itemx --size-check=warning
741 Issue an error or warning for invalid ELF .size directive.
743 @item --elf-stt-common=no
744 @itemx --elf-stt-common=yes
745 These options control whether the ELF assembler should generate common
746 symbols with the @code{STT_COMMON} type. The default can be controlled
747 by a configure option @option{--enable-elf-stt-common}.
751 Print a summary of the command line options and exit.
754 Print a summary of all target specific options and exit.
757 Add directory @var{dir} to the search list for @code{.include} directives.
760 Don't warn about signed overflow.
763 @ifclear DIFF-TBL-KLUGE
764 This option is accepted but has no effect on the @value{TARGET} family.
766 @ifset DIFF-TBL-KLUGE
767 Issue warnings when difference tables altered for long displacements.
772 Keep (in the symbol table) local symbols. These symbols start with
773 system-specific local label prefixes, typically @samp{.L} for ELF systems
774 or @samp{L} for traditional a.out systems.
779 @item --listing-lhs-width=@var{number}
780 Set the maximum width, in words, of the output data column for an assembler
781 listing to @var{number}.
783 @item --listing-lhs-width2=@var{number}
784 Set the maximum width, in words, of the output data column for continuation
785 lines in an assembler listing to @var{number}.
787 @item --listing-rhs-width=@var{number}
788 Set the maximum width of an input source line, as displayed in a listing, to
791 @item --listing-cont-lines=@var{number}
792 Set the maximum number of lines printed in a listing for a single line of input
795 @item --no-pad-sections
796 Stop the assembler for padding the ends of output sections to the alignment
797 of that section. The default is to pad the sections, but this can waste space
798 which might be needed on targets which have tight memory constraints.
800 @item -o @var{objfile}
801 Name the object-file output from @command{@value{AS}} @var{objfile}.
804 Fold the data section into the text section.
806 @item --hash-size=@var{number}
807 Set the default size of GAS's hash tables to a prime number close to
808 @var{number}. Increasing this value can reduce the length of time it takes the
809 assembler to perform its tasks, at the expense of increasing the assembler's
810 memory requirements. Similarly reducing this value can reduce the memory
811 requirements at the expense of speed.
813 @item --reduce-memory-overheads
814 This option reduces GAS's memory requirements, at the expense of making the
815 assembly processes slower. Currently this switch is a synonym for
816 @samp{--hash-size=4051}, but in the future it may have other effects as well.
819 @item --sectname-subst
820 Honor substitution sequences in section names.
822 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
827 Print the maximum space (in bytes) and total time (in seconds) used by
830 @item --strip-local-absolute
831 Remove local absolute symbols from the outgoing symbol table.
835 Print the @command{as} version.
838 Print the @command{as} version and exit.
842 Suppress warning messages.
844 @item --fatal-warnings
845 Treat warnings as errors.
848 Don't suppress warning messages or treat them as errors.
857 Generate an object file even after errors.
859 @item -- | @var{files} @dots{}
860 Standard input, or source files to assemble.
868 @xref{AArch64 Options}, for the options available when @value{AS} is configured
869 for the 64-bit mode of the ARM Architecture (AArch64).
874 The following options are available when @value{AS} is configured for the
875 64-bit mode of the ARM Architecture (AArch64).
878 @include c-aarch64.texi
879 @c ended inside the included file
887 @xref{Alpha Options}, for the options available when @value{AS} is configured
888 for an Alpha processor.
893 The following options are available when @value{AS} is configured for an Alpha
897 @include c-alpha.texi
898 @c ended inside the included file
905 The following options are available when @value{AS} is configured for an ARC
909 @item -mcpu=@var{cpu}
910 This option selects the core processor variant.
912 Select either big-endian (-EB) or little-endian (-EL) output.
914 Enable Code Density extenssion instructions.
919 The following options are available when @value{AS} is configured for the ARM
923 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
924 Specify which ARM processor variant is the target.
925 @item -march=@var{architecture}[+@var{extension}@dots{}]
926 Specify which ARM architecture variant is used by the target.
927 @item -mfpu=@var{floating-point-format}
928 Select which Floating Point architecture is the target.
929 @item -mfloat-abi=@var{abi}
930 Select which floating point ABI is in use.
932 Enable Thumb only instruction decoding.
933 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
934 Select which procedure calling convention is in use.
936 Select either big-endian (-EB) or little-endian (-EL) output.
937 @item -mthumb-interwork
938 Specify that the code has been generated with interworking between Thumb and
941 Turns on CodeComposer Studio assembly syntax compatibility mode.
943 Specify that PIC code has been generated.
951 @xref{Blackfin Options}, for the options available when @value{AS} is
952 configured for the Blackfin processor family.
957 The following options are available when @value{AS} is configured for
958 the Blackfin processor family.
962 @c ended inside the included file
969 See the info pages for documentation of the CRIS-specific options.
973 The following options are available when @value{AS} is configured for
976 @cindex D10V optimization
977 @cindex optimization, D10V
979 Optimize output by parallelizing instructions.
984 The following options are available when @value{AS} is configured for a D30V
987 @cindex D30V optimization
988 @cindex optimization, D30V
990 Optimize output by parallelizing instructions.
994 Warn when nops are generated.
996 @cindex D30V nops after 32-bit multiply
998 Warn when a nop after a 32-bit multiply instruction is generated.
1004 The following options are available when @value{AS} is configured for the
1005 Adapteva EPIPHANY series.
1008 @xref{Epiphany Options}, for the options available when @value{AS} is
1009 configured for an Epiphany processor.
1013 @c man begin OPTIONS
1014 The following options are available when @value{AS} is configured for
1015 an Epiphany processor.
1017 @c man begin INCLUDE
1018 @include c-epiphany.texi
1019 @c ended inside the included file
1027 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1028 for an H8/300 processor.
1032 @c man begin OPTIONS
1033 The following options are available when @value{AS} is configured for an H8/300
1036 @c man begin INCLUDE
1037 @include c-h8300.texi
1038 @c ended inside the included file
1046 @xref{i386-Options}, for the options available when @value{AS} is
1047 configured for an i386 processor.
1051 @c man begin OPTIONS
1052 The following options are available when @value{AS} is configured for
1055 @c man begin INCLUDE
1056 @include c-i386.texi
1057 @c ended inside the included file
1062 @c man begin OPTIONS
1064 The following options are available when @value{AS} is configured for the
1065 Intel 80960 processor.
1068 @item -ACA | -ACA_A | -ACB | -ACC | -AKA | -AKB | -AKC | -AMC
1069 Specify which variant of the 960 architecture is the target.
1072 Add code to collect statistics about branches taken.
1075 Do not alter compare-and-branch instructions for long displacements;
1082 The following options are available when @value{AS} is configured for the
1088 Specifies that the extended IP2022 instructions are allowed.
1091 Restores the default behaviour, which restricts the permitted instructions to
1092 just the basic IP2022 ones.
1098 The following options are available when @value{AS} is configured for the
1099 Renesas M32C and M16C processors.
1104 Assemble M32C instructions.
1107 Assemble M16C instructions (the default).
1110 Enable support for link-time relaxations.
1113 Support H'00 style hex constants in addition to 0x00 style.
1119 The following options are available when @value{AS} is configured for the
1120 Renesas M32R (formerly Mitsubishi M32R) series.
1125 Specify which processor in the M32R family is the target. The default
1126 is normally the M32R, but this option changes it to the M32RX.
1128 @item --warn-explicit-parallel-conflicts or --Wp
1129 Produce warning messages when questionable parallel constructs are
1132 @item --no-warn-explicit-parallel-conflicts or --Wnp
1133 Do not produce warning messages when questionable parallel constructs are
1140 The following options are available when @value{AS} is configured for the
1141 Motorola 68000 series.
1146 Shorten references to undefined symbols, to one word instead of two.
1148 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1149 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1150 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1151 Specify what processor in the 68000 family is the target. The default
1152 is normally the 68020, but this can be changed at configuration time.
1154 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1155 The target machine does (or does not) have a floating-point coprocessor.
1156 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1157 the basic 68000 is not compatible with the 68881, a combination of the
1158 two can be specified, since it's possible to do emulation of the
1159 coprocessor instructions with the main processor.
1161 @item -m68851 | -mno-68851
1162 The target machine does (or does not) have a memory-management
1163 unit coprocessor. The default is to assume an MMU for 68020 and up.
1171 @xref{Nios II Options}, for the options available when @value{AS} is configured
1172 for an Altera Nios II processor.
1176 @c man begin OPTIONS
1177 The following options are available when @value{AS} is configured for an
1178 Altera Nios II processor.
1180 @c man begin INCLUDE
1181 @include c-nios2.texi
1182 @c ended inside the included file
1188 For details about the PDP-11 machine dependent features options,
1189 see @ref{PDP-11-Options}.
1192 @item -mpic | -mno-pic
1193 Generate position-independent (or position-dependent) code. The
1194 default is @option{-mpic}.
1197 @itemx -mall-extensions
1198 Enable all instruction set extensions. This is the default.
1200 @item -mno-extensions
1201 Disable all instruction set extensions.
1203 @item -m@var{extension} | -mno-@var{extension}
1204 Enable (or disable) a particular instruction set extension.
1207 Enable the instruction set extensions supported by a particular CPU, and
1208 disable all other extensions.
1210 @item -m@var{machine}
1211 Enable the instruction set extensions supported by a particular machine
1212 model, and disable all other extensions.
1218 The following options are available when @value{AS} is configured for
1219 a picoJava processor.
1223 @cindex PJ endianness
1224 @cindex endianness, PJ
1225 @cindex big endian output, PJ
1227 Generate ``big endian'' format output.
1229 @cindex little endian output, PJ
1231 Generate ``little endian'' format output.
1237 The following options are available when @value{AS} is configured for the
1238 Motorola 68HC11 or 68HC12 series.
1242 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1243 Specify what processor is the target. The default is
1244 defined by the configuration option when building the assembler.
1246 @item --xgate-ramoffset
1247 Instruct the linker to offset RAM addresses from S12X address space into
1248 XGATE address space.
1251 Specify to use the 16-bit integer ABI.
1254 Specify to use the 32-bit integer ABI.
1256 @item -mshort-double
1257 Specify to use the 32-bit double ABI.
1260 Specify to use the 64-bit double ABI.
1262 @item --force-long-branches
1263 Relative branches are turned into absolute ones. This concerns
1264 conditional branches, unconditional branches and branches to a
1267 @item -S | --short-branches
1268 Do not turn relative branches into absolute ones
1269 when the offset is out of range.
1271 @item --strict-direct-mode
1272 Do not turn the direct addressing mode into extended addressing mode
1273 when the instruction does not support direct addressing mode.
1275 @item --print-insn-syntax
1276 Print the syntax of instruction in case of error.
1278 @item --print-opcodes
1279 Print the list of instructions with syntax and then exit.
1281 @item --generate-example
1282 Print an example of instruction for each possible instruction and then exit.
1283 This option is only useful for testing @command{@value{AS}}.
1289 The following options are available when @command{@value{AS}} is configured
1290 for the SPARC architecture:
1293 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1294 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1295 Explicitly select a variant of the SPARC architecture.
1297 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1298 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1300 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1301 UltraSPARC extensions.
1303 @item -xarch=v8plus | -xarch=v8plusa
1304 For compatibility with the Solaris v9 assembler. These options are
1305 equivalent to -Av8plus and -Av8plusa, respectively.
1308 Warn when the assembler switches to another architecture.
1313 The following options are available when @value{AS} is configured for the 'c54x
1318 Enable extended addressing mode. All addresses and relocations will assume
1319 extended addressing (usually 23 bits).
1320 @item -mcpu=@var{CPU_VERSION}
1321 Sets the CPU version being compiled for.
1322 @item -merrors-to-file @var{FILENAME}
1323 Redirect error output to a file, for broken systems which don't support such
1324 behaviour in the shell.
1329 The following options are available when @value{AS} is configured for
1334 This option sets the largest size of an object that can be referenced
1335 implicitly with the @code{gp} register. It is only accepted for targets that
1336 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1338 @cindex MIPS endianness
1339 @cindex endianness, MIPS
1340 @cindex big endian output, MIPS
1342 Generate ``big endian'' format output.
1344 @cindex little endian output, MIPS
1346 Generate ``little endian'' format output.
1364 Generate code for a particular MIPS Instruction Set Architecture level.
1365 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1366 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1367 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1368 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1369 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1370 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1371 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1372 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1373 MIPS64 Release 6 ISA processors, respectively.
1375 @item -march=@var{cpu}
1376 Generate code for a particular MIPS CPU.
1378 @item -mtune=@var{cpu}
1379 Schedule and tune for a particular MIPS CPU.
1383 Cause nops to be inserted if the read of the destination register
1384 of an mfhi or mflo instruction occurs in the following two instructions.
1387 @itemx -mno-fix-rm7000
1388 Cause nops to be inserted if a dmult or dmultu instruction is
1389 followed by a load instruction.
1393 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1394 section instead of the standard ELF .stabs sections.
1398 Control generation of @code{.pdr} sections.
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 32 bits wide at
1404 all times. @samp{-mgp32} controls the size of general-purpose registers
1405 and @samp{-mfp32} controls the size of floating-point registers.
1409 The register sizes are normally inferred from the ISA and ABI, but these
1410 flags force a certain group of registers to be treated as 64 bits wide at
1411 all times. @samp{-mgp64} controls the size of general-purpose registers
1412 and @samp{-mfp64} controls the size of floating-point registers.
1415 The register sizes are normally inferred from the ISA and ABI, but using
1416 this flag in combination with @samp{-mabi=32} enables an ABI variant
1417 which will operate correctly with floating-point registers which are
1421 @itemx -mno-odd-spreg
1422 Enable use of floating-point operations on odd-numbered single-precision
1423 registers when supported by the ISA. @samp{-mfpxx} implies
1424 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1428 Generate code for the MIPS 16 processor. This is equivalent to putting
1429 @code{.set mips16} at the start of the assembly file. @samp{-no-mips16}
1430 turns off this option.
1433 @itemx -mno-micromips
1434 Generate code for the microMIPS processor. This is equivalent to putting
1435 @code{.set micromips} at the start of the assembly file. @samp{-mno-micromips}
1436 turns off this option. This is equivalent to putting @code{.set nomicromips}
1437 at the start of the assembly file.
1440 @itemx -mno-smartmips
1441 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1442 equivalent to putting @code{.set smartmips} at the start of the assembly file.
1443 @samp{-mno-smartmips} turns off this option.
1447 Generate code for the MIPS-3D Application Specific Extension.
1448 This tells the assembler to accept MIPS-3D instructions.
1449 @samp{-no-mips3d} turns off this option.
1453 Generate code for the MDMX Application Specific Extension.
1454 This tells the assembler to accept MDMX instructions.
1455 @samp{-no-mdmx} turns off this option.
1459 Generate code for the DSP Release 1 Application Specific Extension.
1460 This tells the assembler to accept DSP Release 1 instructions.
1461 @samp{-mno-dsp} turns off this option.
1465 Generate code for the DSP Release 2 Application Specific Extension.
1466 This option implies @samp{-mdsp}.
1467 This tells the assembler to accept DSP Release 2 instructions.
1468 @samp{-mno-dspr2} turns off this option.
1472 Generate code for the DSP Release 3 Application Specific Extension.
1473 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1474 This tells the assembler to accept DSP Release 3 instructions.
1475 @samp{-mno-dspr3} turns off this option.
1479 Generate code for the MIPS SIMD Architecture Extension.
1480 This tells the assembler to accept MSA instructions.
1481 @samp{-mno-msa} turns off this option.
1485 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1486 This tells the assembler to accept XPA instructions.
1487 @samp{-mno-xpa} turns off this option.
1491 Generate code for the MT Application Specific Extension.
1492 This tells the assembler to accept MT instructions.
1493 @samp{-mno-mt} turns off this option.
1497 Generate code for the MCU Application Specific Extension.
1498 This tells the assembler to accept MCU instructions.
1499 @samp{-mno-mcu} turns off this option.
1503 Only use 32-bit instruction encodings when generating code for the
1504 microMIPS processor. This option inhibits the use of any 16-bit
1505 instructions. This is equivalent to putting @code{.set insn32} at
1506 the start of the assembly file. @samp{-mno-insn32} turns off this
1507 option. This is equivalent to putting @code{.set noinsn32} at the
1508 start of the assembly file. By default @samp{-mno-insn32} is
1509 selected, allowing all instructions to be used.
1511 @item --construct-floats
1512 @itemx --no-construct-floats
1513 The @samp{--no-construct-floats} option disables the construction of
1514 double width floating point constants by loading the two halves of the
1515 value into the two single width floating point registers that make up
1516 the double width register. By default @samp{--construct-floats} is
1517 selected, allowing construction of these floating point constants.
1519 @item --relax-branch
1520 @itemx --no-relax-branch
1521 The @samp{--relax-branch} option enables the relaxation of out-of-range
1522 branches. By default @samp{--no-relax-branch} is selected, causing any
1523 out-of-range branches to produce an error.
1525 @item -mnan=@var{encoding}
1526 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1527 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1530 @item --emulation=@var{name}
1531 This option was formerly used to switch between ELF and ECOFF output
1532 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1533 removed in GAS 2.24, so the option now serves little purpose.
1534 It is retained for backwards compatibility.
1536 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1537 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1538 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1539 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1540 preferred options instead.
1543 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1550 Control how to deal with multiplication overflow and division by zero.
1551 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1552 (and only work for Instruction Set Architecture level 2 and higher);
1553 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1557 When this option is used, @command{@value{AS}} will issue a warning every
1558 time it generates a nop instruction from a macro.
1563 The following options are available when @value{AS} is configured for
1569 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1570 The command line option @samp{-nojsri2bsr} can be used to disable it.
1574 Enable or disable the silicon filter behaviour. By default this is disabled.
1575 The default can be overridden by the @samp{-sifilter} command line option.
1578 Alter jump instructions for long displacements.
1580 @item -mcpu=[210|340]
1581 Select the cpu type on the target hardware. This controls which instructions
1585 Assemble for a big endian target.
1588 Assemble for a little endian target.
1597 @xref{Meta Options}, for the options available when @value{AS} is configured
1598 for a Meta processor.
1602 @c man begin OPTIONS
1603 The following options are available when @value{AS} is configured for a
1606 @c man begin INCLUDE
1607 @include c-metag.texi
1608 @c ended inside the included file
1613 @c man begin OPTIONS
1615 See the info pages for documentation of the MMIX-specific options.
1621 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1622 for a NDS32 processor.
1624 @c ended inside the included file
1628 @c man begin OPTIONS
1629 The following options are available when @value{AS} is configured for a
1632 @c man begin INCLUDE
1633 @include c-nds32.texi
1634 @c ended inside the included file
1641 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1642 for a PowerPC processor.
1646 @c man begin OPTIONS
1647 The following options are available when @value{AS} is configured for a
1650 @c man begin INCLUDE
1652 @c ended inside the included file
1660 @xref{RISC-V-Opts}, for the options available when @value{AS} is configured
1661 for a RISC-V processor.
1665 @c man begin OPTIONS
1666 The following options are available when @value{AS} is configured for a
1669 @c man begin INCLUDE
1670 @include c-riscv.texi
1671 @c ended inside the included file
1676 @c man begin OPTIONS
1678 See the info pages for documentation of the RX-specific options.
1682 The following options are available when @value{AS} is configured for the s390
1688 Select the word size, either 31/32 bits or 64 bits.
1691 Select the architecture mode, either the Enterprise System
1692 Architecture (esa) or the z/Architecture mode (zarch).
1693 @item -march=@var{processor}
1694 Specify which s390 processor variant is the target, @samp{g5} (or
1695 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1696 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1697 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1698 or @samp{z13} (or @samp{arch11}).
1700 @itemx -mno-regnames
1701 Allow or disallow symbolic names for registers.
1702 @item -mwarn-areg-zero
1703 Warn whenever the operand for a base or index register has been specified
1704 but evaluates to zero.
1712 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1713 for a TMS320C6000 processor.
1717 @c man begin OPTIONS
1718 The following options are available when @value{AS} is configured for a
1719 TMS320C6000 processor.
1721 @c man begin INCLUDE
1722 @include c-tic6x.texi
1723 @c ended inside the included file
1731 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1732 for a TILE-Gx processor.
1736 @c man begin OPTIONS
1737 The following options are available when @value{AS} is configured for a TILE-Gx
1740 @c man begin INCLUDE
1741 @include c-tilegx.texi
1742 @c ended inside the included file
1750 @xref{Visium Options}, for the options available when @value{AS} is configured
1751 for a Visium processor.
1755 @c man begin OPTIONS
1756 The following option is available when @value{AS} is configured for a Visium
1759 @c man begin INCLUDE
1760 @include c-visium.texi
1761 @c ended inside the included file
1769 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1770 for an Xtensa processor.
1774 @c man begin OPTIONS
1775 The following options are available when @value{AS} is configured for an
1778 @c man begin INCLUDE
1779 @include c-xtensa.texi
1780 @c ended inside the included file
1785 @c man begin OPTIONS
1788 The following options are available when @value{AS} is configured for
1789 a Z80 family processor.
1792 Assemble for Z80 processor.
1794 Assemble for R800 processor.
1795 @item -ignore-undocumented-instructions
1797 Assemble undocumented Z80 instructions that also work on R800 without warning.
1798 @item -ignore-unportable-instructions
1800 Assemble all undocumented Z80 instructions without warning.
1801 @item -warn-undocumented-instructions
1803 Issue a warning for undocumented Z80 instructions that also work on R800.
1804 @item -warn-unportable-instructions
1806 Issue a warning for undocumented Z80 instructions that do not work on R800.
1807 @item -forbid-undocumented-instructions
1809 Treat all undocumented instructions as errors.
1810 @item -forbid-unportable-instructions
1812 Treat undocumented Z80 instructions that do not work on R800 as errors.
1819 * Manual:: Structure of this Manual
1820 * GNU Assembler:: The GNU Assembler
1821 * Object Formats:: Object File Formats
1822 * Command Line:: Command Line
1823 * Input Files:: Input Files
1824 * Object:: Output (Object) File
1825 * Errors:: Error and Warning Messages
1829 @section Structure of this Manual
1831 @cindex manual, structure and purpose
1832 This manual is intended to describe what you need to know to use
1833 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1834 notation for symbols, constants, and expressions; the directives that
1835 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1838 We also cover special features in the @value{TARGET}
1839 configuration of @command{@value{AS}}, including assembler directives.
1842 This manual also describes some of the machine-dependent features of
1843 various flavors of the assembler.
1846 @cindex machine instructions (not covered)
1847 On the other hand, this manual is @emph{not} intended as an introduction
1848 to programming in assembly language---let alone programming in general!
1849 In a similar vein, we make no attempt to introduce the machine
1850 architecture; we do @emph{not} describe the instruction set, standard
1851 mnemonics, registers or addressing modes that are standard to a
1852 particular architecture.
1854 You may want to consult the manufacturer's
1855 machine architecture manual for this information.
1859 For information on the H8/300 machine instruction set, see @cite{H8/300
1860 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1861 Programming Manual} (Renesas).
1864 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1865 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1866 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1867 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1870 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1874 @c I think this is premature---doc@cygnus.com, 17jan1991
1876 Throughout this manual, we assume that you are running @dfn{GNU},
1877 the portable operating system from the @dfn{Free Software
1878 Foundation, Inc.}. This restricts our attention to certain kinds of
1879 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1880 once this assumption is granted examples and definitions need less
1883 @command{@value{AS}} is part of a team of programs that turn a high-level
1884 human-readable series of instructions into a low-level
1885 computer-readable series of instructions. Different versions of
1886 @command{@value{AS}} are used for different kinds of computer.
1889 @c There used to be a section "Terminology" here, which defined
1890 @c "contents", "byte", "word", and "long". Defining "word" to any
1891 @c particular size is confusing when the .word directive may generate 16
1892 @c bits on one machine and 32 bits on another; in general, for the user
1893 @c version of this manual, none of these terms seem essential to define.
1894 @c They were used very little even in the former draft of the manual;
1895 @c this draft makes an effort to avoid them (except in names of
1899 @section The GNU Assembler
1901 @c man begin DESCRIPTION
1903 @sc{gnu} @command{as} is really a family of assemblers.
1905 This manual describes @command{@value{AS}}, a member of that family which is
1906 configured for the @value{TARGET} architectures.
1908 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1909 should find a fairly similar environment when you use it on another
1910 architecture. Each version has much in common with the others,
1911 including object file formats, most assembler directives (often called
1912 @dfn{pseudo-ops}) and assembler syntax.@refill
1914 @cindex purpose of @sc{gnu} assembler
1915 @command{@value{AS}} is primarily intended to assemble the output of the
1916 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1917 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1918 assemble correctly everything that other assemblers for the same
1919 machine would assemble.
1921 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1924 @c This remark should appear in generic version of manual; assumption
1925 @c here is that generic version sets M680x0.
1926 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1927 assembler for the same architecture; for example, we know of several
1928 incompatible versions of 680x0 assembly language syntax.
1933 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1934 program in one pass of the source file. This has a subtle impact on the
1935 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1937 @node Object Formats
1938 @section Object File Formats
1940 @cindex object file format
1941 The @sc{gnu} assembler can be configured to produce several alternative
1942 object file formats. For the most part, this does not affect how you
1943 write assembly language programs; but directives for debugging symbols
1944 are typically different in different file formats. @xref{Symbol
1945 Attributes,,Symbol Attributes}.
1948 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1949 @value{OBJ-NAME} format object files.
1951 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1953 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1954 @code{b.out} or COFF format object files.
1957 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1958 SOM or ELF format object files.
1963 @section Command Line
1965 @cindex command line conventions
1967 After the program name @command{@value{AS}}, the command line may contain
1968 options and file names. Options may appear in any order, and may be
1969 before, after, or between file names. The order of file names is
1972 @cindex standard input, as input file
1974 @file{--} (two hyphens) by itself names the standard input file
1975 explicitly, as one of the files for @command{@value{AS}} to assemble.
1977 @cindex options, command line
1978 Except for @samp{--} any command line argument that begins with a
1979 hyphen (@samp{-}) is an option. Each option changes the behavior of
1980 @command{@value{AS}}. No option changes the way another option works. An
1981 option is a @samp{-} followed by one or more letters; the case of
1982 the letter is important. All options are optional.
1984 Some options expect exactly one file name to follow them. The file
1985 name may either immediately follow the option's letter (compatible
1986 with older assemblers) or it may be the next command argument (@sc{gnu}
1987 standard). These two command lines are equivalent:
1990 @value{AS} -o my-object-file.o mumble.s
1991 @value{AS} -omy-object-file.o mumble.s
1995 @section Input Files
1998 @cindex source program
1999 @cindex files, input
2000 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2001 describe the program input to one run of @command{@value{AS}}. The program may
2002 be in one or more files; how the source is partitioned into files
2003 doesn't change the meaning of the source.
2005 @c I added "con" prefix to "catenation" just to prove I can overcome my
2006 @c APL training... doc@cygnus.com
2007 The source program is a concatenation of the text in all the files, in the
2010 @c man begin DESCRIPTION
2011 Each time you run @command{@value{AS}} it assembles exactly one source
2012 program. The source program is made up of one or more files.
2013 (The standard input is also a file.)
2015 You give @command{@value{AS}} a command line that has zero or more input file
2016 names. The input files are read (from left file name to right). A
2017 command line argument (in any position) that has no special meaning
2018 is taken to be an input file name.
2020 If you give @command{@value{AS}} no file names it attempts to read one input file
2021 from the @command{@value{AS}} standard input, which is normally your terminal. You
2022 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2025 Use @samp{--} if you need to explicitly name the standard input file
2026 in your command line.
2028 If the source is empty, @command{@value{AS}} produces a small, empty object
2033 @subheading Filenames and Line-numbers
2035 @cindex input file linenumbers
2036 @cindex line numbers, in input files
2037 There are two ways of locating a line in the input file (or files) and
2038 either may be used in reporting error messages. One way refers to a line
2039 number in a physical file; the other refers to a line number in a
2040 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2042 @dfn{Physical files} are those files named in the command line given
2043 to @command{@value{AS}}.
2045 @dfn{Logical files} are simply names declared explicitly by assembler
2046 directives; they bear no relation to physical files. Logical file names help
2047 error messages reflect the original source file, when @command{@value{AS}} source
2048 is itself synthesized from other files. @command{@value{AS}} understands the
2049 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2050 @ref{File,,@code{.file}}.
2053 @section Output (Object) File
2059 Every time you run @command{@value{AS}} it produces an output file, which is
2060 your assembly language program translated into numbers. This file
2061 is the object file. Its default name is
2069 @code{b.out} when @command{@value{AS}} is configured for the Intel 80960.
2071 You can give it another name by using the @option{-o} option. Conventionally,
2072 object file names end with @file{.o}. The default name is used for historical
2073 reasons: older assemblers were capable of assembling self-contained programs
2074 directly into a runnable program. (For some formats, this isn't currently
2075 possible, but it can be done for the @code{a.out} format.)
2079 The object file is meant for input to the linker @code{@value{LD}}. It contains
2080 assembled program code, information to help @code{@value{LD}} integrate
2081 the assembled program into a runnable file, and (optionally) symbolic
2082 information for the debugger.
2084 @c link above to some info file(s) like the description of a.out.
2085 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2088 @section Error and Warning Messages
2090 @c man begin DESCRIPTION
2092 @cindex error messages
2093 @cindex warning messages
2094 @cindex messages from assembler
2095 @command{@value{AS}} may write warnings and error messages to the standard error
2096 file (usually your terminal). This should not happen when a compiler
2097 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2098 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2099 grave problem that stops the assembly.
2103 @cindex format of warning messages
2104 Warning messages have the format
2107 file_name:@b{NNN}:Warning Message Text
2111 @cindex file names and line numbers, in warnings/errors
2112 (where @b{NNN} is a line number). If both a logical file name
2113 (@pxref{File,,@code{.file}}) and a logical line number
2115 (@pxref{Line,,@code{.line}})
2117 have been given then they will be used, otherwise the file name and line number
2118 in the current assembler source file will be used. The message text is
2119 intended to be self explanatory (in the grand Unix tradition).
2121 Note the file name must be set via the logical version of the @code{.file}
2122 directive, not the DWARF2 version of the @code{.file} directive. For example:
2126 error_assembler_source
2132 produces this output:
2136 asm.s:2: Error: no such instruction: `error_assembler_source'
2137 foo.c:31: Error: no such instruction: `error_c_source'
2140 @cindex format of error messages
2141 Error messages have the format
2144 file_name:@b{NNN}:FATAL:Error Message Text
2147 The file name and line number are derived as for warning
2148 messages. The actual message text may be rather less explanatory
2149 because many of them aren't supposed to happen.
2152 @chapter Command-Line Options
2154 @cindex options, all versions of assembler
2155 This chapter describes command-line options available in @emph{all}
2156 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2157 for options specific
2159 to the @value{TARGET} target.
2162 to particular machine architectures.
2165 @c man begin DESCRIPTION
2167 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2168 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2169 The assembler arguments must be separated from each other (and the @samp{-Wa})
2170 by commas. For example:
2173 gcc -c -g -O -Wa,-alh,-L file.c
2177 This passes two options to the assembler: @samp{-alh} (emit a listing to
2178 standard output with high-level and assembly source) and @samp{-L} (retain
2179 local symbols in the symbol table).
2181 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2182 command-line options are automatically passed to the assembler by the compiler.
2183 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2184 precisely what options it passes to each compilation pass, including the
2190 * a:: -a[cdghlns] enable listings
2191 * alternate:: --alternate enable alternate macro syntax
2192 * D:: -D for compatibility
2193 * f:: -f to work faster
2194 * I:: -I for .include search path
2195 @ifclear DIFF-TBL-KLUGE
2196 * K:: -K for compatibility
2198 @ifset DIFF-TBL-KLUGE
2199 * K:: -K for difference tables
2202 * L:: -L to retain local symbols
2203 * listing:: --listing-XXX to configure listing output
2204 * M:: -M or --mri to assemble in MRI compatibility mode
2205 * MD:: --MD for dependency tracking
2206 * no-pad-sections:: --no-pad-sections to stop section padding
2207 * o:: -o to name the object file
2208 * R:: -R to join data and text sections
2209 * statistics:: --statistics to see statistics about assembly
2210 * traditional-format:: --traditional-format for compatible output
2211 * v:: -v to announce version
2212 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2213 * Z:: -Z to make object file even after errors
2217 @section Enable Listings: @option{-a[cdghlns]}
2227 @cindex listings, enabling
2228 @cindex assembly listings, enabling
2230 These options enable listing output from the assembler. By itself,
2231 @samp{-a} requests high-level, assembly, and symbols listing.
2232 You can use other letters to select specific options for the list:
2233 @samp{-ah} requests a high-level language listing,
2234 @samp{-al} requests an output-program assembly listing, and
2235 @samp{-as} requests a symbol table listing.
2236 High-level listings require that a compiler debugging option like
2237 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2240 Use the @samp{-ag} option to print a first section with general assembly
2241 information, like @value{AS} version, switches passed, or time stamp.
2243 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2244 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2245 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2246 omitted from the listing.
2248 Use the @samp{-ad} option to omit debugging directives from the
2251 Once you have specified one of these options, you can further control
2252 listing output and its appearance using the directives @code{.list},
2253 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2255 The @samp{-an} option turns off all forms processing.
2256 If you do not request listing output with one of the @samp{-a} options, the
2257 listing-control directives have no effect.
2259 The letters after @samp{-a} may be combined into one option,
2260 @emph{e.g.}, @samp{-aln}.
2262 Note if the assembler source is coming from the standard input (e.g.,
2264 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
2265 is being used) then the listing will not contain any comments or preprocessor
2266 directives. This is because the listing code buffers input source lines from
2267 stdin only after they have been preprocessed by the assembler. This reduces
2268 memory usage and makes the code more efficient.
2271 @section @option{--alternate}
2274 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2277 @section @option{-D}
2280 This option has no effect whatsoever, but it is accepted to make it more
2281 likely that scripts written for other assemblers also work with
2282 @command{@value{AS}}.
2285 @section Work Faster: @option{-f}
2288 @cindex trusted compiler
2289 @cindex faster processing (@option{-f})
2290 @samp{-f} should only be used when assembling programs written by a
2291 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2292 and comment preprocessing on
2293 the input file(s) before assembling them. @xref{Preprocessing,
2297 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2298 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2303 @section @code{.include} Search Path: @option{-I} @var{path}
2305 @kindex -I @var{path}
2306 @cindex paths for @code{.include}
2307 @cindex search path for @code{.include}
2308 @cindex @code{include} directive search path
2309 Use this option to add a @var{path} to the list of directories
2310 @command{@value{AS}} searches for files specified in @code{.include}
2311 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2312 many times as necessary to include a variety of paths. The current
2313 working directory is always searched first; after that, @command{@value{AS}}
2314 searches any @samp{-I} directories in the same order as they were
2315 specified (left to right) on the command line.
2318 @section Difference Tables: @option{-K}
2321 @ifclear DIFF-TBL-KLUGE
2322 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2323 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2324 where it can be used to warn when the assembler alters the machine code
2325 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2326 family does not have the addressing limitations that sometimes lead to this
2327 alteration on other platforms.
2330 @ifset DIFF-TBL-KLUGE
2331 @cindex difference tables, warning
2332 @cindex warning for altered difference tables
2333 @command{@value{AS}} sometimes alters the code emitted for directives of the
2334 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2335 You can use the @samp{-K} option if you want a warning issued when this
2340 @section Include Local Symbols: @option{-L}
2343 @cindex local symbols, retaining in output
2344 Symbols beginning with system-specific local label prefixes, typically
2345 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2346 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2347 such symbols when debugging, because they are intended for the use of
2348 programs (like compilers) that compose assembler programs, not for your
2349 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2350 such symbols, so you do not normally debug with them.
2352 This option tells @command{@value{AS}} to retain those local symbols
2353 in the object file. Usually if you do this you also tell the linker
2354 @code{@value{LD}} to preserve those symbols.
2357 @section Configuring listing output: @option{--listing}
2359 The listing feature of the assembler can be enabled via the command line switch
2360 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2361 hex dump of the corresponding locations in the output object file, and displays
2362 them as a listing file. The format of this listing can be controlled by
2363 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2364 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2365 @code{.psize} (@pxref{Psize}), and
2366 @code{.eject} (@pxref{Eject}) and also by the following switches:
2369 @item --listing-lhs-width=@samp{number}
2370 @kindex --listing-lhs-width
2371 @cindex Width of first line disassembly output
2372 Sets the maximum width, in words, of the first line of the hex byte dump. This
2373 dump appears on the left hand side of the listing output.
2375 @item --listing-lhs-width2=@samp{number}
2376 @kindex --listing-lhs-width2
2377 @cindex Width of continuation lines of disassembly output
2378 Sets the maximum width, in words, of any further lines of the hex byte dump for
2379 a given input source line. If this value is not specified, it defaults to being
2380 the same as the value specified for @samp{--listing-lhs-width}. If neither
2381 switch is used the default is to one.
2383 @item --listing-rhs-width=@samp{number}
2384 @kindex --listing-rhs-width
2385 @cindex Width of source line output
2386 Sets the maximum width, in characters, of the source line that is displayed
2387 alongside the hex dump. The default value for this parameter is 100. The
2388 source line is displayed on the right hand side of the listing output.
2390 @item --listing-cont-lines=@samp{number}
2391 @kindex --listing-cont-lines
2392 @cindex Maximum number of continuation lines
2393 Sets the maximum number of continuation lines of hex dump that will be
2394 displayed for a given single line of source input. The default value is 4.
2398 @section Assemble in MRI Compatibility Mode: @option{-M}
2401 @cindex MRI compatibility mode
2402 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2403 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2404 compatible with the @code{ASM68K} or the @code{ASM960} (depending upon the
2405 configured target) assembler from Microtec Research. The exact nature of the
2406 MRI syntax will not be documented here; see the MRI manuals for more
2407 information. Note in particular that the handling of macros and macro
2408 arguments is somewhat different. The purpose of this option is to permit
2409 assembling existing MRI assembler code using @command{@value{AS}}.
2411 The MRI compatibility is not complete. Certain operations of the MRI assembler
2412 depend upon its object file format, and can not be supported using other object
2413 file formats. Supporting these would require enhancing each object file format
2414 individually. These are:
2417 @item global symbols in common section
2419 The m68k MRI assembler supports common sections which are merged by the linker.
2420 Other object file formats do not support this. @command{@value{AS}} handles
2421 common sections by treating them as a single common symbol. It permits local
2422 symbols to be defined within a common section, but it can not support global
2423 symbols, since it has no way to describe them.
2425 @item complex relocations
2427 The MRI assemblers support relocations against a negated section address, and
2428 relocations which combine the start addresses of two or more sections. These
2429 are not support by other object file formats.
2431 @item @code{END} pseudo-op specifying start address
2433 The MRI @code{END} pseudo-op permits the specification of a start address.
2434 This is not supported by other object file formats. The start address may
2435 instead be specified using the @option{-e} option to the linker, or in a linker
2438 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2440 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2441 name to the output file. This is not supported by other object file formats.
2443 @item @code{ORG} pseudo-op
2445 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2446 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2447 which changes the location within the current section. Absolute sections are
2448 not supported by other object file formats. The address of a section may be
2449 assigned within a linker script.
2452 There are some other features of the MRI assembler which are not supported by
2453 @command{@value{AS}}, typically either because they are difficult or because they
2454 seem of little consequence. Some of these may be supported in future releases.
2458 @item EBCDIC strings
2460 EBCDIC strings are not supported.
2462 @item packed binary coded decimal
2464 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2465 and @code{DCB.P} pseudo-ops are not supported.
2467 @item @code{FEQU} pseudo-op
2469 The m68k @code{FEQU} pseudo-op is not supported.
2471 @item @code{NOOBJ} pseudo-op
2473 The m68k @code{NOOBJ} pseudo-op is not supported.
2475 @item @code{OPT} branch control options
2477 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2478 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2479 relaxes all branches, whether forward or backward, to an appropriate size, so
2480 these options serve no purpose.
2482 @item @code{OPT} list control options
2484 The following m68k @code{OPT} list control options are ignored: @code{C},
2485 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2486 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2488 @item other @code{OPT} options
2490 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2491 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2493 @item @code{OPT} @code{D} option is default
2495 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2496 @code{OPT NOD} may be used to turn it off.
2498 @item @code{XREF} pseudo-op.
2500 The m68k @code{XREF} pseudo-op is ignored.
2502 @item @code{.debug} pseudo-op
2504 The i960 @code{.debug} pseudo-op is not supported.
2506 @item @code{.extended} pseudo-op
2508 The i960 @code{.extended} pseudo-op is not supported.
2510 @item @code{.list} pseudo-op.
2512 The various options of the i960 @code{.list} pseudo-op are not supported.
2514 @item @code{.optimize} pseudo-op
2516 The i960 @code{.optimize} pseudo-op is not supported.
2518 @item @code{.output} pseudo-op
2520 The i960 @code{.output} pseudo-op is not supported.
2522 @item @code{.setreal} pseudo-op
2524 The i960 @code{.setreal} pseudo-op is not supported.
2529 @section Dependency Tracking: @option{--MD}
2532 @cindex dependency tracking
2535 @command{@value{AS}} can generate a dependency file for the file it creates. This
2536 file consists of a single rule suitable for @code{make} describing the
2537 dependencies of the main source file.
2539 The rule is written to the file named in its argument.
2541 This feature is used in the automatic updating of makefiles.
2543 @node no-pad-sections
2544 @section Output Section Padding
2545 @kindex --no-pad-sections
2546 @cindex output section padding
2547 Normally the assembler will pad the end of each output section up to its
2548 alignment boundary. But this can waste space, which can be significant on
2549 memory constrained targets. So the @option{--no-pad-sections} option will
2550 disable this behaviour.
2553 @section Name the Object File: @option{-o}
2556 @cindex naming object file
2557 @cindex object file name
2558 There is always one object file output when you run @command{@value{AS}}. By
2559 default it has the name
2562 @file{a.out} (or @file{b.out}, for Intel 960 targets only).
2576 You use this option (which takes exactly one filename) to give the
2577 object file a different name.
2579 Whatever the object file is called, @command{@value{AS}} overwrites any
2580 existing file of the same name.
2583 @section Join Data and Text Sections: @option{-R}
2586 @cindex data and text sections, joining
2587 @cindex text and data sections, joining
2588 @cindex joining text and data sections
2589 @cindex merging text and data sections
2590 @option{-R} tells @command{@value{AS}} to write the object file as if all
2591 data-section data lives in the text section. This is only done at
2592 the very last moment: your binary data are the same, but data
2593 section parts are relocated differently. The data section part of
2594 your object file is zero bytes long because all its bytes are
2595 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2597 When you specify @option{-R} it would be possible to generate shorter
2598 address displacements (because we do not have to cross between text and
2599 data section). We refrain from doing this simply for compatibility with
2600 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2603 When @command{@value{AS}} is configured for COFF or ELF output,
2604 this option is only useful if you use sections named @samp{.text} and
2609 @option{-R} is not supported for any of the HPPA targets. Using
2610 @option{-R} generates a warning from @command{@value{AS}}.
2614 @section Display Assembly Statistics: @option{--statistics}
2616 @kindex --statistics
2617 @cindex statistics, about assembly
2618 @cindex time, total for assembly
2619 @cindex space used, maximum for assembly
2620 Use @samp{--statistics} to display two statistics about the resources used by
2621 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2622 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2625 @node traditional-format
2626 @section Compatible Output: @option{--traditional-format}
2628 @kindex --traditional-format
2629 For some targets, the output of @command{@value{AS}} is different in some ways
2630 from the output of some existing assembler. This switch requests
2631 @command{@value{AS}} to use the traditional format instead.
2633 For example, it disables the exception frame optimizations which
2634 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2637 @section Announce Version: @option{-v}
2641 @cindex assembler version
2642 @cindex version of assembler
2643 You can find out what version of as is running by including the
2644 option @samp{-v} (which you can also spell as @samp{-version}) on the
2648 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2650 @command{@value{AS}} should never give a warning or error message when
2651 assembling compiler output. But programs written by people often
2652 cause @command{@value{AS}} to give a warning that a particular assumption was
2653 made. All such warnings are directed to the standard error file.
2657 @cindex suppressing warnings
2658 @cindex warnings, suppressing
2659 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2660 This only affects the warning messages: it does not change any particular of
2661 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2664 @kindex --fatal-warnings
2665 @cindex errors, caused by warnings
2666 @cindex warnings, causing error
2667 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2668 files that generate warnings to be in error.
2671 @cindex warnings, switching on
2672 You can switch these options off again by specifying @option{--warn}, which
2673 causes warnings to be output as usual.
2676 @section Generate Object File in Spite of Errors: @option{-Z}
2677 @cindex object file, after errors
2678 @cindex errors, continuing after
2679 After an error message, @command{@value{AS}} normally produces no output. If for
2680 some reason you are interested in object file output even after
2681 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2682 option. If there are any errors, @command{@value{AS}} continues anyways, and
2683 writes an object file after a final warning message of the form @samp{@var{n}
2684 errors, @var{m} warnings, generating bad object file.}
2689 @cindex machine-independent syntax
2690 @cindex syntax, machine-independent
2691 This chapter describes the machine-independent syntax allowed in a
2692 source file. @command{@value{AS}} syntax is similar to what many other
2693 assemblers use; it is inspired by the BSD 4.2
2698 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2702 * Preprocessing:: Preprocessing
2703 * Whitespace:: Whitespace
2704 * Comments:: Comments
2705 * Symbol Intro:: Symbols
2706 * Statements:: Statements
2707 * Constants:: Constants
2711 @section Preprocessing
2713 @cindex preprocessing
2714 The @command{@value{AS}} internal preprocessor:
2716 @cindex whitespace, removed by preprocessor
2718 adjusts and removes extra whitespace. It leaves one space or tab before
2719 the keywords on a line, and turns any other whitespace on the line into
2722 @cindex comments, removed by preprocessor
2724 removes all comments, replacing them with a single space, or an
2725 appropriate number of newlines.
2727 @cindex constants, converted by preprocessor
2729 converts character constants into the appropriate numeric values.
2732 It does not do macro processing, include file handling, or
2733 anything else you may get from your C compiler's preprocessor. You can
2734 do include file processing with the @code{.include} directive
2735 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2736 to get other ``CPP'' style preprocessing by giving the input file a
2737 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2738 Output, gcc info, Using GNU CC}.
2740 Excess whitespace, comments, and character constants
2741 cannot be used in the portions of the input text that are not
2744 @cindex turning preprocessing on and off
2745 @cindex preprocessing, turning on and off
2748 If the first line of an input file is @code{#NO_APP} or if you use the
2749 @samp{-f} option, whitespace and comments are not removed from the input file.
2750 Within an input file, you can ask for whitespace and comment removal in
2751 specific portions of the by putting a line that says @code{#APP} before the
2752 text that may contain whitespace or comments, and putting a line that says
2753 @code{#NO_APP} after this text. This feature is mainly intend to support
2754 @code{asm} statements in compilers whose output is otherwise free of comments
2761 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2762 Whitespace is used to separate symbols, and to make programs neater for
2763 people to read. Unless within character constants
2764 (@pxref{Characters,,Character Constants}), any whitespace means the same
2765 as exactly one space.
2771 There are two ways of rendering comments to @command{@value{AS}}. In both
2772 cases the comment is equivalent to one space.
2774 Anything from @samp{/*} through the next @samp{*/} is a comment.
2775 This means you may not nest these comments.
2779 The only way to include a newline ('\n') in a comment
2780 is to use this sort of comment.
2783 /* This sort of comment does not nest. */
2786 @cindex line comment character
2787 Anything from a @dfn{line comment} character up to the next newline is
2788 considered a comment and is ignored. The line comment character is target
2789 specific, and some targets multiple comment characters. Some targets also have
2790 line comment characters that only work if they are the first character on a
2791 line. Some targets use a sequence of two characters to introduce a line
2792 comment. Some targets can also change their line comment characters depending
2793 upon command line options that have been used. For more details see the
2794 @emph{Syntax} section in the documentation for individual targets.
2796 If the line comment character is the hash sign (@samp{#}) then it still has the
2797 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2798 to specify logical line numbers:
2801 @cindex lines starting with @code{#}
2802 @cindex logical line numbers
2803 To be compatible with past assemblers, lines that begin with @samp{#} have a
2804 special interpretation. Following the @samp{#} should be an absolute
2805 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2806 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2807 new logical file name. The rest of the line, if any, should be whitespace.
2809 If the first non-whitespace characters on the line are not numeric,
2810 the line is ignored. (Just like a comment.)
2813 # This is an ordinary comment.
2814 # 42-6 "new_file_name" # New logical file name
2815 # This is logical line # 36.
2817 This feature is deprecated, and may disappear from future versions
2818 of @command{@value{AS}}.
2823 @cindex characters used in symbols
2824 @ifclear SPECIAL-SYMS
2825 A @dfn{symbol} is one or more characters chosen from the set of all
2826 letters (both upper and lower case), digits and the three characters
2832 A @dfn{symbol} is one or more characters chosen from the set of all
2833 letters (both upper and lower case), digits and the three characters
2834 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2840 On most machines, you can also use @code{$} in symbol names; exceptions
2841 are noted in @ref{Machine Dependencies}.
2843 No symbol may begin with a digit. Case is significant.
2844 There is no length limit; all characters are significant. Multibyte characters
2845 are supported. Symbols are delimited by characters not in that set, or by the
2846 beginning of a file (since the source program must end with a newline, the end
2847 of a file is not a possible symbol delimiter). @xref{Symbols}.
2849 Symbol names may also be enclosed in double quote @code{"} characters. In such
2850 cases any characters are allowed, except for the NUL character. If a double
2851 quote character is to be included in the symbol name it must be preceeded by a
2852 backslash @code{\} character.
2853 @cindex length of symbols
2858 @cindex statements, structure of
2859 @cindex line separator character
2860 @cindex statement separator character
2862 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2863 @dfn{line separator character}. The line separator character is target
2864 specific and described in the @emph{Syntax} section of each
2865 target's documentation. Not all targets support a line separator character.
2866 The newline or line separator character is considered to be part of the
2867 preceding statement. Newlines and separators within character constants are an
2868 exception: they do not end statements.
2870 @cindex newline, required at file end
2871 @cindex EOF, newline must precede
2872 It is an error to end any statement with end-of-file: the last
2873 character of any input file should be a newline.@refill
2875 An empty statement is allowed, and may include whitespace. It is ignored.
2877 @cindex instructions and directives
2878 @cindex directives and instructions
2879 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2880 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2882 A statement begins with zero or more labels, optionally followed by a
2883 key symbol which determines what kind of statement it is. The key
2884 symbol determines the syntax of the rest of the statement. If the
2885 symbol begins with a dot @samp{.} then the statement is an assembler
2886 directive: typically valid for any computer. If the symbol begins with
2887 a letter the statement is an assembly language @dfn{instruction}: it
2888 assembles into a machine language instruction.
2890 Different versions of @command{@value{AS}} for different computers
2891 recognize different instructions. In fact, the same symbol may
2892 represent a different instruction in a different computer's assembly
2896 @cindex @code{:} (label)
2897 @cindex label (@code{:})
2898 A label is a symbol immediately followed by a colon (@code{:}).
2899 Whitespace before a label or after a colon is permitted, but you may not
2900 have whitespace between a label's symbol and its colon. @xref{Labels}.
2903 For HPPA targets, labels need not be immediately followed by a colon, but
2904 the definition of a label must begin in column zero. This also implies that
2905 only one label may be defined on each line.
2909 label: .directive followed by something
2910 another_label: # This is an empty statement.
2911 instruction operand_1, operand_2, @dots{}
2918 A constant is a number, written so that its value is known by
2919 inspection, without knowing any context. Like this:
2922 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2923 .ascii "Ring the bell\7" # A string constant.
2924 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2925 .float 0f-314159265358979323846264338327\
2926 95028841971.693993751E-40 # - pi, a flonum.
2931 * Characters:: Character Constants
2932 * Numbers:: Number Constants
2936 @subsection Character Constants
2938 @cindex character constants
2939 @cindex constants, character
2940 There are two kinds of character constants. A @dfn{character} stands
2941 for one character in one byte and its value may be used in
2942 numeric expressions. String constants (properly called string
2943 @emph{literals}) are potentially many bytes and their values may not be
2944 used in arithmetic expressions.
2948 * Chars:: Characters
2952 @subsubsection Strings
2954 @cindex string constants
2955 @cindex constants, string
2956 A @dfn{string} is written between double-quotes. It may contain
2957 double-quotes or null characters. The way to get special characters
2958 into a string is to @dfn{escape} these characters: precede them with
2959 a backslash @samp{\} character. For example @samp{\\} represents
2960 one backslash: the first @code{\} is an escape which tells
2961 @command{@value{AS}} to interpret the second character literally as a backslash
2962 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2963 escape character). The complete list of escapes follows.
2965 @cindex escape codes, character
2966 @cindex character escape codes
2967 @c NOTE: Cindex entries must not start with a backlash character.
2968 @c NOTE: This confuses the pdf2texi script when it is creating the
2969 @c NOTE: index based upon the first character and so it generates:
2970 @c NOTE: \initial {\\}
2971 @c NOTE: which then results in the error message:
2972 @c NOTE: Argument of \\ has an extra }.
2973 @c NOTE: So in the index entries below a space character has been
2974 @c NOTE: prepended to avoid this problem.
2977 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2979 @cindex @code{ \b} (backspace character)
2980 @cindex backspace (@code{\b})
2982 Mnemonic for backspace; for ASCII this is octal code 010.
2985 @c Mnemonic for EOText; for ASCII this is octal code 004.
2987 @cindex @code{ \f} (formfeed character)
2988 @cindex formfeed (@code{\f})
2990 Mnemonic for FormFeed; for ASCII this is octal code 014.
2992 @cindex @code{ \n} (newline character)
2993 @cindex newline (@code{\n})
2995 Mnemonic for newline; for ASCII this is octal code 012.
2998 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
3000 @cindex @code{ \r} (carriage return character)
3001 @cindex carriage return (@code{backslash-r})
3003 Mnemonic for carriage-Return; for ASCII this is octal code 015.
3006 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
3007 @c other assemblers.
3009 @cindex @code{ \t} (tab)
3010 @cindex tab (@code{\t})
3012 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
3015 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
3016 @c @item \x @var{digit} @var{digit} @var{digit}
3017 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
3019 @cindex @code{ \@var{ddd}} (octal character code)
3020 @cindex octal character code (@code{\@var{ddd}})
3021 @item \ @var{digit} @var{digit} @var{digit}
3022 An octal character code. The numeric code is 3 octal digits.
3023 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
3024 for example, @code{\008} has the value 010, and @code{\009} the value 011.
3026 @cindex @code{ \@var{xd...}} (hex character code)
3027 @cindex hex character code (@code{\@var{xd...}})
3028 @item \@code{x} @var{hex-digits...}
3029 A hex character code. All trailing hex digits are combined. Either upper or
3030 lower case @code{x} works.
3032 @cindex @code{ \\} (@samp{\} character)
3033 @cindex backslash (@code{\\})
3035 Represents one @samp{\} character.
3038 @c Represents one @samp{'} (accent acute) character.
3039 @c This is needed in single character literals
3040 @c (@xref{Characters,,Character Constants}.) to represent
3043 @cindex @code{ \"} (doublequote character)
3044 @cindex doublequote (@code{\"})
3046 Represents one @samp{"} character. Needed in strings to represent
3047 this character, because an unescaped @samp{"} would end the string.
3049 @item \ @var{anything-else}
3050 Any other character when escaped by @kbd{\} gives a warning, but
3051 assembles as if the @samp{\} was not present. The idea is that if
3052 you used an escape sequence you clearly didn't want the literal
3053 interpretation of the following character. However @command{@value{AS}} has no
3054 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3055 code and warns you of the fact.
3058 Which characters are escapable, and what those escapes represent,
3059 varies widely among assemblers. The current set is what we think
3060 the BSD 4.2 assembler recognizes, and is a subset of what most C
3061 compilers recognize. If you are in doubt, do not use an escape
3065 @subsubsection Characters
3067 @cindex single character constant
3068 @cindex character, single
3069 @cindex constant, single character
3070 A single character may be written as a single quote immediately
3071 followed by that character. The same escapes apply to characters as
3072 to strings. So if you want to write the character backslash, you
3073 must write @kbd{'\\} where the first @code{\} escapes the second
3074 @code{\}. As you can see, the quote is an acute accent, not a
3075 grave accent. A newline
3077 @ifclear abnormal-separator
3078 (or semicolon @samp{;})
3080 @ifset abnormal-separator
3082 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3087 immediately following an acute accent is taken as a literal character
3088 and does not count as the end of a statement. The value of a character
3089 constant in a numeric expression is the machine's byte-wide code for
3090 that character. @command{@value{AS}} assumes your character code is ASCII:
3091 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3094 @subsection Number Constants
3096 @cindex constants, number
3097 @cindex number constants
3098 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3099 are stored in the target machine. @emph{Integers} are numbers that
3100 would fit into an @code{int} in the C language. @emph{Bignums} are
3101 integers, but they are stored in more than 32 bits. @emph{Flonums}
3102 are floating point numbers, described below.
3105 * Integers:: Integers
3110 * Bit Fields:: Bit Fields
3116 @subsubsection Integers
3118 @cindex constants, integer
3120 @cindex binary integers
3121 @cindex integers, binary
3122 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3123 the binary digits @samp{01}.
3125 @cindex octal integers
3126 @cindex integers, octal
3127 An octal integer is @samp{0} followed by zero or more of the octal
3128 digits (@samp{01234567}).
3130 @cindex decimal integers
3131 @cindex integers, decimal
3132 A decimal integer starts with a non-zero digit followed by zero or
3133 more digits (@samp{0123456789}).
3135 @cindex hexadecimal integers
3136 @cindex integers, hexadecimal
3137 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3138 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3140 Integers have the usual values. To denote a negative integer, use
3141 the prefix operator @samp{-} discussed under expressions
3142 (@pxref{Prefix Ops,,Prefix Operators}).
3145 @subsubsection Bignums
3148 @cindex constants, bignum
3149 A @dfn{bignum} has the same syntax and semantics as an integer
3150 except that the number (or its negative) takes more than 32 bits to
3151 represent in binary. The distinction is made because in some places
3152 integers are permitted while bignums are not.
3155 @subsubsection Flonums
3157 @cindex floating point numbers
3158 @cindex constants, floating point
3160 @cindex precision, floating point
3161 A @dfn{flonum} represents a floating point number. The translation is
3162 indirect: a decimal floating point number from the text is converted by
3163 @command{@value{AS}} to a generic binary floating point number of more than
3164 sufficient precision. This generic floating point number is converted
3165 to a particular computer's floating point format (or formats) by a
3166 portion of @command{@value{AS}} specialized to that computer.
3168 A flonum is written by writing (in order)
3173 (@samp{0} is optional on the HPPA.)
3177 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3179 @kbd{e} is recommended. Case is not important.
3181 @c FIXME: verify if flonum syntax really this vague for most cases
3182 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3183 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3186 On the H8/300, Renesas / SuperH SH,
3187 and AMD 29K architectures, the letter must be
3188 one of the letters @samp{DFPRSX} (in upper or lower case).
3190 On the ARC, the letter must be one of the letters @samp{DFRS}
3191 (in upper or lower case).
3193 On the Intel 960 architecture, the letter must be
3194 one of the letters @samp{DFT} (in upper or lower case).
3196 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3200 One of the letters @samp{DFRS} (in upper or lower case).
3203 One of the letters @samp{DFPRSX} (in upper or lower case).
3206 The letter @samp{E} (upper case only).
3209 One of the letters @samp{DFT} (in upper or lower case).
3214 An optional sign: either @samp{+} or @samp{-}.
3217 An optional @dfn{integer part}: zero or more decimal digits.
3220 An optional @dfn{fractional part}: @samp{.} followed by zero
3221 or more decimal digits.
3224 An optional exponent, consisting of:
3228 An @samp{E} or @samp{e}.
3229 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3230 @c principle this can perfectly well be different on different targets.
3232 Optional sign: either @samp{+} or @samp{-}.
3234 One or more decimal digits.
3239 At least one of the integer part or the fractional part must be
3240 present. The floating point number has the usual base-10 value.
3242 @command{@value{AS}} does all processing using integers. Flonums are computed
3243 independently of any floating point hardware in the computer running
3244 @command{@value{AS}}.
3248 @c Bit fields are written as a general facility but are also controlled
3249 @c by a conditional-compilation flag---which is as of now (21mar91)
3250 @c turned on only by the i960 config of GAS.
3252 @subsubsection Bit Fields
3255 @cindex constants, bit field
3256 You can also define numeric constants as @dfn{bit fields}.
3257 Specify two numbers separated by a colon---
3259 @var{mask}:@var{value}
3262 @command{@value{AS}} applies a bitwise @sc{and} between @var{mask} and
3265 The resulting number is then packed
3267 @c this conditional paren in case bit fields turned on elsewhere than 960
3268 (in host-dependent byte order)
3270 into a field whose width depends on which assembler directive has the
3271 bit-field as its argument. Overflow (a result from the bitwise and
3272 requiring more binary digits to represent) is not an error; instead,
3273 more constants are generated, of the specified width, beginning with the
3274 least significant digits.@refill
3276 The directives @code{.byte}, @code{.hword}, @code{.int}, @code{.long},
3277 @code{.short}, and @code{.word} accept bit-field arguments.
3282 @chapter Sections and Relocation
3287 * Secs Background:: Background
3288 * Ld Sections:: Linker Sections
3289 * As Sections:: Assembler Internal Sections
3290 * Sub-Sections:: Sub-Sections
3294 @node Secs Background
3297 Roughly, a section is a range of addresses, with no gaps; all data
3298 ``in'' those addresses is treated the same for some particular purpose.
3299 For example there may be a ``read only'' section.
3301 @cindex linker, and assembler
3302 @cindex assembler, and linker
3303 The linker @code{@value{LD}} reads many object files (partial programs) and
3304 combines their contents to form a runnable program. When @command{@value{AS}}
3305 emits an object file, the partial program is assumed to start at address 0.
3306 @code{@value{LD}} assigns the final addresses for the partial program, so that
3307 different partial programs do not overlap. This is actually an
3308 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3311 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3312 addresses. These blocks slide to their run-time addresses as rigid
3313 units; their length does not change and neither does the order of bytes
3314 within them. Such a rigid unit is called a @emph{section}. Assigning
3315 run-time addresses to sections is called @dfn{relocation}. It includes
3316 the task of adjusting mentions of object-file addresses so they refer to
3317 the proper run-time addresses.
3319 For the H8/300, and for the Renesas / SuperH SH,
3320 @command{@value{AS}} pads sections if needed to
3321 ensure they end on a word (sixteen bit) boundary.
3324 @cindex standard assembler sections
3325 An object file written by @command{@value{AS}} has at least three sections, any
3326 of which may be empty. These are named @dfn{text}, @dfn{data} and
3331 When it generates COFF or ELF output,
3333 @command{@value{AS}} can also generate whatever other named sections you specify
3334 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3335 If you do not use any directives that place output in the @samp{.text}
3336 or @samp{.data} sections, these sections still exist, but are empty.
3341 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3343 @command{@value{AS}} can also generate whatever other named sections you
3344 specify using the @samp{.space} and @samp{.subspace} directives. See
3345 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3346 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3347 assembler directives.
3350 Additionally, @command{@value{AS}} uses different names for the standard
3351 text, data, and bss sections when generating SOM output. Program text
3352 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3353 BSS into @samp{$BSS$}.
3357 Within the object file, the text section starts at address @code{0}, the
3358 data section follows, and the bss section follows the data section.
3361 When generating either SOM or ELF output files on the HPPA, the text
3362 section starts at address @code{0}, the data section at address
3363 @code{0x4000000}, and the bss section follows the data section.
3366 To let @code{@value{LD}} know which data changes when the sections are
3367 relocated, and how to change that data, @command{@value{AS}} also writes to the
3368 object file details of the relocation needed. To perform relocation
3369 @code{@value{LD}} must know, each time an address in the object
3373 Where in the object file is the beginning of this reference to
3376 How long (in bytes) is this reference?
3378 Which section does the address refer to? What is the numeric value of
3380 (@var{address}) @minus{} (@var{start-address of section})?
3383 Is the reference to an address ``Program-Counter relative''?
3386 @cindex addresses, format of
3387 @cindex section-relative addressing
3388 In fact, every address @command{@value{AS}} ever uses is expressed as
3390 (@var{section}) + (@var{offset into section})
3393 Further, most expressions @command{@value{AS}} computes have this section-relative
3396 (For some object formats, such as SOM for the HPPA, some expressions are
3397 symbol-relative instead.)
3400 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3401 @var{N} into section @var{secname}.''
3403 Apart from text, data and bss sections you need to know about the
3404 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3405 addresses in the absolute section remain unchanged. For example, address
3406 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3407 @code{@value{LD}}. Although the linker never arranges two partial programs'
3408 data sections with overlapping addresses after linking, @emph{by definition}
3409 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3410 part of a program is always the same address when the program is running as
3411 address @code{@{absolute@ 239@}} in any other part of the program.
3413 The idea of sections is extended to the @dfn{undefined} section. Any
3414 address whose section is unknown at assembly time is by definition
3415 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3416 Since numbers are always defined, the only way to generate an undefined
3417 address is to mention an undefined symbol. A reference to a named
3418 common block would be such a symbol: its value is unknown at assembly
3419 time so it has section @emph{undefined}.
3421 By analogy the word @emph{section} is used to describe groups of sections in
3422 the linked program. @code{@value{LD}} puts all partial programs' text
3423 sections in contiguous addresses in the linked program. It is
3424 customary to refer to the @emph{text section} of a program, meaning all
3425 the addresses of all partial programs' text sections. Likewise for
3426 data and bss sections.
3428 Some sections are manipulated by @code{@value{LD}}; others are invented for
3429 use of @command{@value{AS}} and have no meaning except during assembly.
3432 @section Linker Sections
3433 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3438 @cindex named sections
3439 @cindex sections, named
3440 @item named sections
3443 @cindex text section
3444 @cindex data section
3448 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3449 separate but equal sections. Anything you can say of one section is
3452 When the program is running, however, it is
3453 customary for the text section to be unalterable. The
3454 text section is often shared among processes: it contains
3455 instructions, constants and the like. The data section of a running
3456 program is usually alterable: for example, C variables would be stored
3457 in the data section.
3462 This section contains zeroed bytes when your program begins running. It
3463 is used to hold uninitialized variables or common storage. The length of
3464 each partial program's bss section is important, but because it starts
3465 out containing zeroed bytes there is no need to store explicit zero
3466 bytes in the object file. The bss section was invented to eliminate
3467 those explicit zeros from object files.
3469 @cindex absolute section
3470 @item absolute section
3471 Address 0 of this section is always ``relocated'' to runtime address 0.
3472 This is useful if you want to refer to an address that @code{@value{LD}} must
3473 not change when relocating. In this sense we speak of absolute
3474 addresses being ``unrelocatable'': they do not change during relocation.
3476 @cindex undefined section
3477 @item undefined section
3478 This ``section'' is a catch-all for address references to objects not in
3479 the preceding sections.
3480 @c FIXME: ref to some other doc on obj-file formats could go here.
3483 @cindex relocation example
3484 An idealized example of three relocatable sections follows.
3486 The example uses the traditional section names @samp{.text} and @samp{.data}.
3488 Memory addresses are on the horizontal axis.
3492 @c END TEXI2ROFF-KILL
3495 partial program # 1: |ttttt|dddd|00|
3502 partial program # 2: |TTT|DDD|000|
3505 +--+---+-----+--+----+---+-----+~~
3506 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3507 +--+---+-----+--+----+---+-----+~~
3509 addresses: 0 @dots{}
3516 \line{\it Partial program \#1: \hfil}
3517 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3518 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3520 \line{\it Partial program \#2: \hfil}
3521 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3522 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3524 \line{\it linked program: \hfil}
3525 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3526 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3527 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3528 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3530 \line{\it addresses: \hfil}
3534 @c END TEXI2ROFF-KILL
3537 @section Assembler Internal Sections
3539 @cindex internal assembler sections
3540 @cindex sections in messages, internal
3541 These sections are meant only for the internal use of @command{@value{AS}}. They
3542 have no meaning at run-time. You do not really need to know about these
3543 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3544 warning messages, so it might be helpful to have an idea of their
3545 meanings to @command{@value{AS}}. These sections are used to permit the
3546 value of every expression in your assembly language program to be a
3547 section-relative address.
3550 @cindex assembler internal logic error
3551 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3552 An internal assembler logic error has been found. This means there is a
3553 bug in the assembler.
3555 @cindex expr (internal section)
3557 The assembler stores complex expression internally as combinations of
3558 symbols. When it needs to represent an expression as a symbol, it puts
3559 it in the expr section.
3561 @c FIXME item transfer[t] vector preload
3562 @c FIXME item transfer[t] vector postload
3563 @c FIXME item register
3567 @section Sub-Sections
3569 @cindex numbered subsections
3570 @cindex grouping data
3576 fall into two sections: text and data.
3578 You may have separate groups of
3580 data in named sections
3584 data in named sections
3590 that you want to end up near to each other in the object file, even though they
3591 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3592 use @dfn{subsections} for this purpose. Within each section, there can be
3593 numbered subsections with values from 0 to 8192. Objects assembled into the
3594 same subsection go into the object file together with other objects in the same
3595 subsection. For example, a compiler might want to store constants in the text
3596 section, but might not want to have them interspersed with the program being
3597 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3598 section of code being output, and a @samp{.text 1} before each group of
3599 constants being output.
3601 Subsections are optional. If you do not use subsections, everything
3602 goes in subsection number zero.
3605 Each subsection is zero-padded up to a multiple of four bytes.
3606 (Subsections may be padded a different amount on different flavors
3607 of @command{@value{AS}}.)
3611 On the H8/300 platform, each subsection is zero-padded to a word
3612 boundary (two bytes).
3613 The same is true on the Renesas SH.
3616 @c FIXME section padding (alignment)?
3617 @c Rich Pixley says padding here depends on target obj code format; that
3618 @c doesn't seem particularly useful to say without further elaboration,
3619 @c so for now I say nothing about it. If this is a generic BFD issue,
3620 @c these paragraphs might need to vanish from this manual, and be
3621 @c discussed in BFD chapter of binutils (or some such).
3625 Subsections appear in your object file in numeric order, lowest numbered
3626 to highest. (All this to be compatible with other people's assemblers.)
3627 The object file contains no representation of subsections; @code{@value{LD}} and
3628 other programs that manipulate object files see no trace of them.
3629 They just see all your text subsections as a text section, and all your
3630 data subsections as a data section.
3632 To specify which subsection you want subsequent statements assembled
3633 into, use a numeric argument to specify it, in a @samp{.text
3634 @var{expression}} or a @samp{.data @var{expression}} statement.
3637 When generating COFF output, you
3642 can also use an extra subsection
3643 argument with arbitrary named sections: @samp{.section @var{name},
3648 When generating ELF output, you
3653 can also use the @code{.subsection} directive (@pxref{SubSection})
3654 to specify a subsection: @samp{.subsection @var{expression}}.
3656 @var{Expression} should be an absolute expression
3657 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3658 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3659 begins in @code{text 0}. For instance:
3661 .text 0 # The default subsection is text 0 anyway.
3662 .ascii "This lives in the first text subsection. *"
3664 .ascii "But this lives in the second text subsection."
3666 .ascii "This lives in the data section,"
3667 .ascii "in the first data subsection."
3669 .ascii "This lives in the first text section,"
3670 .ascii "immediately following the asterisk (*)."
3673 Each section has a @dfn{location counter} incremented by one for every byte
3674 assembled into that section. Because subsections are merely a convenience
3675 restricted to @command{@value{AS}} there is no concept of a subsection location
3676 counter. There is no way to directly manipulate a location counter---but the
3677 @code{.align} directive changes it, and any label definition captures its
3678 current value. The location counter of the section where statements are being
3679 assembled is said to be the @dfn{active} location counter.
3682 @section bss Section
3685 @cindex common variable storage
3686 The bss section is used for local common variable storage.
3687 You may allocate address space in the bss section, but you may
3688 not dictate data to load into it before your program executes. When
3689 your program starts running, all the contents of the bss
3690 section are zeroed bytes.
3692 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3693 @ref{Lcomm,,@code{.lcomm}}.
3695 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3696 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3699 When assembling for a target which supports multiple sections, such as ELF or
3700 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3701 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3702 section. Typically the section will only contain symbol definitions and
3703 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3710 Symbols are a central concept: the programmer uses symbols to name
3711 things, the linker uses symbols to link, and the debugger uses symbols
3715 @cindex debuggers, and symbol order
3716 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3717 the same order they were declared. This may break some debuggers.
3722 * Setting Symbols:: Giving Symbols Other Values
3723 * Symbol Names:: Symbol Names
3724 * Dot:: The Special Dot Symbol
3725 * Symbol Attributes:: Symbol Attributes
3732 A @dfn{label} is written as a symbol immediately followed by a colon
3733 @samp{:}. The symbol then represents the current value of the
3734 active location counter, and is, for example, a suitable instruction
3735 operand. You are warned if you use the same symbol to represent two
3736 different locations: the first definition overrides any other
3740 On the HPPA, the usual form for a label need not be immediately followed by a
3741 colon, but instead must start in column zero. Only one label may be defined on
3742 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3743 provides a special directive @code{.label} for defining labels more flexibly.
3746 @node Setting Symbols
3747 @section Giving Symbols Other Values
3749 @cindex assigning values to symbols
3750 @cindex symbol values, assigning
3751 A symbol can be given an arbitrary value by writing a symbol, followed
3752 by an equals sign @samp{=}, followed by an expression
3753 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3754 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3755 equals sign @samp{=}@samp{=} here represents an equivalent of the
3756 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3759 Blackfin does not support symbol assignment with @samp{=}.
3763 @section Symbol Names
3765 @cindex symbol names
3766 @cindex names, symbol
3767 @ifclear SPECIAL-SYMS
3768 Symbol names begin with a letter or with one of @samp{._}. On most
3769 machines, you can also use @code{$} in symbol names; exceptions are
3770 noted in @ref{Machine Dependencies}. That character may be followed by any
3771 string of digits, letters, dollar signs (unless otherwise noted for a
3772 particular target machine), and underscores.
3776 Symbol names begin with a letter or with one of @samp{._}. On the
3777 Renesas SH you can also use @code{$} in symbol names. That
3778 character may be followed by any string of digits, letters, dollar signs (save
3779 on the H8/300), and underscores.
3783 Case of letters is significant: @code{foo} is a different symbol name
3786 Symbol names do not start with a digit. An exception to this rule is made for
3787 Local Labels. See below.
3789 Multibyte characters are supported. To generate a symbol name containing
3790 multibyte characters enclose it within double quotes and use escape codes. cf
3791 @xref{Strings}. Generating a multibyte symbol name from a label is not
3792 currently supported.
3794 Each symbol has exactly one name. Each name in an assembly language program
3795 refers to exactly one symbol. You may use that symbol name any number of times
3798 @subheading Local Symbol Names
3800 @cindex local symbol names
3801 @cindex symbol names, local
3802 A local symbol is any symbol beginning with certain local label prefixes.
3803 By default, the local label prefix is @samp{.L} for ELF systems or
3804 @samp{L} for traditional a.out systems, but each target may have its own
3805 set of local label prefixes.
3807 On the HPPA local symbols begin with @samp{L$}.
3810 Local symbols are defined and used within the assembler, but they are
3811 normally not saved in object files. Thus, they are not visible when debugging.
3812 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3813 to retain the local symbols in the object files.
3815 @subheading Local Labels
3817 @cindex local labels
3818 @cindex temporary symbol names
3819 @cindex symbol names, temporary
3820 Local labels are different from local symbols. Local labels help compilers and
3821 programmers use names temporarily. They create symbols which are guaranteed to
3822 be unique over the entire scope of the input source code and which can be
3823 referred to by a simple notation. To define a local label, write a label of
3824 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3825 To refer to the most recent previous definition of that label write
3826 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3827 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3828 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3830 There is no restriction on how you can use these labels, and you can reuse them
3831 too. So that it is possible to repeatedly define the same local label (using
3832 the same number @samp{@b{N}}), although you can only refer to the most recently
3833 defined local label of that number (for a backwards reference) or the next
3834 definition of a specific local label for a forward reference. It is also worth
3835 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3836 implemented in a slightly more efficient manner than the others.
3847 Which is the equivalent of:
3850 label_1: branch label_3
3851 label_2: branch label_1
3852 label_3: branch label_4
3853 label_4: branch label_3
3856 Local label names are only a notational device. They are immediately
3857 transformed into more conventional symbol names before the assembler uses them.
3858 The symbol names are stored in the symbol table, appear in error messages, and
3859 are optionally emitted to the object file. The names are constructed using
3863 @item @emph{local label prefix}
3864 All local symbols begin with the system-specific local label prefix.
3865 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3866 that start with the local label prefix. These labels are
3867 used for symbols you are never intended to see. If you use the
3868 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3869 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3870 you may use them in debugging.
3873 This is the number that was used in the local label definition. So if the
3874 label is written @samp{55:} then the number is @samp{55}.
3877 This unusual character is included so you do not accidentally invent a symbol
3878 of the same name. The character has ASCII value of @samp{\002} (control-B).
3880 @item @emph{ordinal number}
3881 This is a serial number to keep the labels distinct. The first definition of
3882 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3883 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3884 the number @samp{1} and its 15th definition gets @samp{15} as well.
3887 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3888 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3890 @subheading Dollar Local Labels
3891 @cindex dollar local symbols
3893 On some targets @code{@value{AS}} also supports an even more local form of
3894 local labels called dollar labels. These labels go out of scope (i.e., they
3895 become undefined) as soon as a non-local label is defined. Thus they remain
3896 valid for only a small region of the input source code. Normal local labels,
3897 by contrast, remain in scope for the entire file, or until they are redefined
3898 by another occurrence of the same local label.
3900 Dollar labels are defined in exactly the same way as ordinary local labels,
3901 except that they have a dollar sign suffix to their numeric value, e.g.,
3904 They can also be distinguished from ordinary local labels by their transformed
3905 names which use ASCII character @samp{\001} (control-A) as the magic character
3906 to distinguish them from ordinary labels. For example, the fifth definition of
3907 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3910 @section The Special Dot Symbol
3912 @cindex dot (symbol)
3913 @cindex @code{.} (symbol)
3914 @cindex current address
3915 @cindex location counter
3916 The special symbol @samp{.} refers to the current address that
3917 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3918 .long .} defines @code{melvin} to contain its own address.
3919 Assigning a value to @code{.} is treated the same as a @code{.org}
3921 @ifclear no-space-dir
3922 Thus, the expression @samp{.=.+4} is the same as saying
3926 @node Symbol Attributes
3927 @section Symbol Attributes
3929 @cindex symbol attributes
3930 @cindex attributes, symbol
3931 Every symbol has, as well as its name, the attributes ``Value'' and
3932 ``Type''. Depending on output format, symbols can also have auxiliary
3935 The detailed definitions are in @file{a.out.h}.
3938 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3939 all these attributes, and probably won't warn you. This makes the
3940 symbol an externally defined symbol, which is generally what you
3944 * Symbol Value:: Value
3945 * Symbol Type:: Type
3948 * a.out Symbols:: Symbol Attributes: @code{a.out}
3952 * a.out Symbols:: Symbol Attributes: @code{a.out}
3955 * a.out Symbols:: Symbol Attributes: @code{a.out}, @code{b.out}
3960 * COFF Symbols:: Symbol Attributes for COFF
3963 * SOM Symbols:: Symbol Attributes for SOM
3970 @cindex value of a symbol
3971 @cindex symbol value
3972 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3973 location in the text, data, bss or absolute sections the value is the
3974 number of addresses from the start of that section to the label.
3975 Naturally for text, data and bss sections the value of a symbol changes
3976 as @code{@value{LD}} changes section base addresses during linking. Absolute
3977 symbols' values do not change during linking: that is why they are
3980 The value of an undefined symbol is treated in a special way. If it is
3981 0 then the symbol is not defined in this assembler source file, and
3982 @code{@value{LD}} tries to determine its value from other files linked into the
3983 same program. You make this kind of symbol simply by mentioning a symbol
3984 name without defining it. A non-zero value represents a @code{.comm}
3985 common declaration. The value is how much common storage to reserve, in
3986 bytes (addresses). The symbol refers to the first address of the
3992 @cindex type of a symbol
3994 The type attribute of a symbol contains relocation (section)
3995 information, any flag settings indicating that a symbol is external, and
3996 (optionally), other information for linkers and debuggers. The exact
3997 format depends on the object-code output format in use.
4002 @c The following avoids a "widow" subsection title. @group would be
4003 @c better if it were available outside examples.
4006 @subsection Symbol Attributes: @code{a.out}, @code{b.out}
4008 @cindex @code{b.out} symbol attributes
4009 @cindex symbol attributes, @code{b.out}
4010 These symbol attributes appear only when @command{@value{AS}} is configured for
4011 one of the Berkeley-descended object output formats---@code{a.out} or
4017 @subsection Symbol Attributes: @code{a.out}
4019 @cindex @code{a.out} symbol attributes
4020 @cindex symbol attributes, @code{a.out}
4026 @subsection Symbol Attributes: @code{a.out}
4028 @cindex @code{a.out} symbol attributes
4029 @cindex symbol attributes, @code{a.out}
4033 * Symbol Desc:: Descriptor
4034 * Symbol Other:: Other
4038 @subsubsection Descriptor
4040 @cindex descriptor, of @code{a.out} symbol
4041 This is an arbitrary 16-bit value. You may establish a symbol's
4042 descriptor value by using a @code{.desc} statement
4043 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
4044 @command{@value{AS}}.
4047 @subsubsection Other
4049 @cindex other attribute, of @code{a.out} symbol
4050 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
4055 @subsection Symbol Attributes for COFF
4057 @cindex COFF symbol attributes
4058 @cindex symbol attributes, COFF
4060 The COFF format supports a multitude of auxiliary symbol attributes;
4061 like the primary symbol attributes, they are set between @code{.def} and
4062 @code{.endef} directives.
4064 @subsubsection Primary Attributes
4066 @cindex primary attributes, COFF symbols
4067 The symbol name is set with @code{.def}; the value and type,
4068 respectively, with @code{.val} and @code{.type}.
4070 @subsubsection Auxiliary Attributes
4072 @cindex auxiliary attributes, COFF symbols
4073 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
4074 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
4075 table information for COFF.
4080 @subsection Symbol Attributes for SOM
4082 @cindex SOM symbol attributes
4083 @cindex symbol attributes, SOM
4085 The SOM format for the HPPA supports a multitude of symbol attributes set with
4086 the @code{.EXPORT} and @code{.IMPORT} directives.
4088 The attributes are described in @cite{HP9000 Series 800 Assembly
4089 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
4090 @code{EXPORT} assembler directive documentation.
4094 @chapter Expressions
4098 @cindex numeric values
4099 An @dfn{expression} specifies an address or numeric value.
4100 Whitespace may precede and/or follow an expression.
4102 The result of an expression must be an absolute number, or else an offset into
4103 a particular section. If an expression is not absolute, and there is not
4104 enough information when @command{@value{AS}} sees the expression to know its
4105 section, a second pass over the source program might be necessary to interpret
4106 the expression---but the second pass is currently not implemented.
4107 @command{@value{AS}} aborts with an error message in this situation.
4110 * Empty Exprs:: Empty Expressions
4111 * Integer Exprs:: Integer Expressions
4115 @section Empty Expressions
4117 @cindex empty expressions
4118 @cindex expressions, empty
4119 An empty expression has no value: it is just whitespace or null.
4120 Wherever an absolute expression is required, you may omit the
4121 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4122 is compatible with other assemblers.
4125 @section Integer Expressions
4127 @cindex integer expressions
4128 @cindex expressions, integer
4129 An @dfn{integer expression} is one or more @emph{arguments} delimited
4130 by @emph{operators}.
4133 * Arguments:: Arguments
4134 * Operators:: Operators
4135 * Prefix Ops:: Prefix Operators
4136 * Infix Ops:: Infix Operators
4140 @subsection Arguments
4142 @cindex expression arguments
4143 @cindex arguments in expressions
4144 @cindex operands in expressions
4145 @cindex arithmetic operands
4146 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4147 contexts arguments are sometimes called ``arithmetic operands''. In
4148 this manual, to avoid confusing them with the ``instruction operands'' of
4149 the machine language, we use the term ``argument'' to refer to parts of
4150 expressions only, reserving the word ``operand'' to refer only to machine
4151 instruction operands.
4153 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4154 @var{section} is one of text, data, bss, absolute,
4155 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4158 Numbers are usually integers.
4160 A number can be a flonum or bignum. In this case, you are warned
4161 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4162 these 32 bits are an integer. You may write integer-manipulating
4163 instructions that act on exotic constants, compatible with other
4166 @cindex subexpressions
4167 Subexpressions are a left parenthesis @samp{(} followed by an integer
4168 expression, followed by a right parenthesis @samp{)}; or a prefix
4169 operator followed by an argument.
4172 @subsection Operators
4174 @cindex operators, in expressions
4175 @cindex arithmetic functions
4176 @cindex functions, in expressions
4177 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4178 operators are followed by an argument. Infix operators appear
4179 between their arguments. Operators may be preceded and/or followed by
4183 @subsection Prefix Operator
4185 @cindex prefix operators
4186 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4187 one argument, which must be absolute.
4189 @c the tex/end tex stuff surrounding this small table is meant to make
4190 @c it align, on the printed page, with the similar table in the next
4191 @c section (which is inside an enumerate).
4193 \global\advance\leftskip by \itemindent
4198 @dfn{Negation}. Two's complement negation.
4200 @dfn{Complementation}. Bitwise not.
4204 \global\advance\leftskip by -\itemindent
4208 @subsection Infix Operators
4210 @cindex infix operators
4211 @cindex operators, permitted arguments
4212 @dfn{Infix operators} take two arguments, one on either side. Operators
4213 have precedence, but operations with equal precedence are performed left
4214 to right. Apart from @code{+} or @option{-}, both arguments must be
4215 absolute, and the result is absolute.
4218 @cindex operator precedence
4219 @cindex precedence of operators
4226 @dfn{Multiplication}.
4229 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4235 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4238 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4242 Intermediate precedence
4247 @dfn{Bitwise Inclusive Or}.
4253 @dfn{Bitwise Exclusive Or}.
4256 @dfn{Bitwise Or Not}.
4263 @cindex addition, permitted arguments
4264 @cindex plus, permitted arguments
4265 @cindex arguments for addition
4267 @dfn{Addition}. If either argument is absolute, the result has the section of
4268 the other argument. You may not add together arguments from different
4271 @cindex subtraction, permitted arguments
4272 @cindex minus, permitted arguments
4273 @cindex arguments for subtraction
4275 @dfn{Subtraction}. If the right argument is absolute, the
4276 result has the section of the left argument.
4277 If both arguments are in the same section, the result is absolute.
4278 You may not subtract arguments from different sections.
4279 @c FIXME is there still something useful to say about undefined - undefined ?
4281 @cindex comparison expressions
4282 @cindex expressions, comparison
4287 @dfn{Is Not Equal To}
4291 @dfn{Is Greater Than}
4293 @dfn{Is Greater Than Or Equal To}
4295 @dfn{Is Less Than Or Equal To}
4297 The comparison operators can be used as infix operators. A true results has a
4298 value of -1 whereas a false result has a value of 0. Note, these operators
4299 perform signed comparisons.
4302 @item Lowest Precedence
4311 These two logical operations can be used to combine the results of sub
4312 expressions. Note, unlike the comparison operators a true result returns a
4313 value of 1 but a false results does still return 0. Also note that the logical
4314 or operator has a slightly lower precedence than logical and.
4319 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4320 address; you can only have a defined section in one of the two arguments.
4323 @chapter Assembler Directives
4325 @cindex directives, machine independent
4326 @cindex pseudo-ops, machine independent
4327 @cindex machine independent directives
4328 All assembler directives have names that begin with a period (@samp{.}).
4329 The names are case insensitive for most targets, and usually written
4332 This chapter discusses directives that are available regardless of the
4333 target machine configuration for the @sc{gnu} assembler.
4335 Some machine configurations provide additional directives.
4336 @xref{Machine Dependencies}.
4339 @ifset machine-directives
4340 @xref{Machine Dependencies}, for additional directives.
4345 * Abort:: @code{.abort}
4347 * ABORT (COFF):: @code{.ABORT}
4350 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4351 * Altmacro:: @code{.altmacro}
4352 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4353 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4354 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4355 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4356 * Byte:: @code{.byte @var{expressions}}
4357 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4358 * Comm:: @code{.comm @var{symbol} , @var{length} }
4359 * Data:: @code{.data @var{subsection}}
4361 * Def:: @code{.def @var{name}}
4364 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4370 * Double:: @code{.double @var{flonums}}
4371 * Eject:: @code{.eject}
4372 * Else:: @code{.else}
4373 * Elseif:: @code{.elseif}
4376 * Endef:: @code{.endef}
4379 * Endfunc:: @code{.endfunc}
4380 * Endif:: @code{.endif}
4381 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4382 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4383 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4385 * Error:: @code{.error @var{string}}
4386 * Exitm:: @code{.exitm}
4387 * Extern:: @code{.extern}
4388 * Fail:: @code{.fail}
4389 * File:: @code{.file}
4390 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4391 * Float:: @code{.float @var{flonums}}
4392 * Func:: @code{.func}
4393 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4395 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4396 * Hidden:: @code{.hidden @var{names}}
4399 * hword:: @code{.hword @var{expressions}}
4400 * Ident:: @code{.ident}
4401 * If:: @code{.if @var{absolute expression}}
4402 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4403 * Include:: @code{.include "@var{file}"}
4404 * Int:: @code{.int @var{expressions}}
4406 * Internal:: @code{.internal @var{names}}
4409 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4410 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4411 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4412 * Lflags:: @code{.lflags}
4413 @ifclear no-line-dir
4414 * Line:: @code{.line @var{line-number}}
4417 * Linkonce:: @code{.linkonce [@var{type}]}
4418 * List:: @code{.list}
4419 * Ln:: @code{.ln @var{line-number}}
4420 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4421 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4423 * Local:: @code{.local @var{names}}
4426 * Long:: @code{.long @var{expressions}}
4428 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4431 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4432 * MRI:: @code{.mri @var{val}}
4433 * Noaltmacro:: @code{.noaltmacro}
4434 * Nolist:: @code{.nolist}
4435 * Octa:: @code{.octa @var{bignums}}
4436 * Offset:: @code{.offset @var{loc}}
4437 * Org:: @code{.org @var{new-lc}, @var{fill}}
4438 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4440 * PopSection:: @code{.popsection}
4441 * Previous:: @code{.previous}
4444 * Print:: @code{.print @var{string}}
4446 * Protected:: @code{.protected @var{names}}
4449 * Psize:: @code{.psize @var{lines}, @var{columns}}
4450 * Purgem:: @code{.purgem @var{name}}
4452 * PushSection:: @code{.pushsection @var{name}}
4455 * Quad:: @code{.quad @var{bignums}}
4456 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4457 * Rept:: @code{.rept @var{count}}
4458 * Sbttl:: @code{.sbttl "@var{subheading}"}
4460 * Scl:: @code{.scl @var{class}}
4463 * Section:: @code{.section @var{name}[, @var{flags}]}
4466 * Set:: @code{.set @var{symbol}, @var{expression}}
4467 * Short:: @code{.short @var{expressions}}
4468 * Single:: @code{.single @var{flonums}}
4470 * Size:: @code{.size [@var{name} , @var{expression}]}
4472 @ifclear no-space-dir
4473 * Skip:: @code{.skip @var{size} , @var{fill}}
4476 * Sleb128:: @code{.sleb128 @var{expressions}}
4477 @ifclear no-space-dir
4478 * Space:: @code{.space @var{size} , @var{fill}}
4481 * Stab:: @code{.stabd, .stabn, .stabs}
4484 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4485 * Struct:: @code{.struct @var{expression}}
4487 * SubSection:: @code{.subsection}
4488 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4492 * Tag:: @code{.tag @var{structname}}
4495 * Text:: @code{.text @var{subsection}}
4496 * Title:: @code{.title "@var{heading}"}
4498 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4501 * Uleb128:: @code{.uleb128 @var{expressions}}
4503 * Val:: @code{.val @var{addr}}
4507 * Version:: @code{.version "@var{string}"}
4508 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4509 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4512 * Warning:: @code{.warning @var{string}}
4513 * Weak:: @code{.weak @var{names}}
4514 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4515 * Word:: @code{.word @var{expressions}}
4516 @ifclear no-space-dir
4517 * Zero:: @code{.zero @var{size}}
4519 * Deprecated:: Deprecated Directives
4523 @section @code{.abort}
4525 @cindex @code{abort} directive
4526 @cindex stopping the assembly
4527 This directive stops the assembly immediately. It is for
4528 compatibility with other assemblers. The original idea was that the
4529 assembly language source would be piped into the assembler. If the sender
4530 of the source quit, it could use this directive tells @command{@value{AS}} to
4531 quit also. One day @code{.abort} will not be supported.
4535 @section @code{.ABORT} (COFF)
4537 @cindex @code{ABORT} directive
4538 When producing COFF output, @command{@value{AS}} accepts this directive as a
4539 synonym for @samp{.abort}.
4542 When producing @code{b.out} output, @command{@value{AS}} accepts this directive,
4548 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4550 @cindex padding the location counter
4551 @cindex @code{align} directive
4552 Pad the location counter (in the current subsection) to a particular storage
4553 boundary. The first expression (which must be absolute) is the alignment
4554 required, as described below.
4556 The second expression (also absolute) gives the fill value to be stored in the
4557 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4558 padding bytes are normally zero. However, on some systems, if the section is
4559 marked as containing code and the fill value is omitted, the space is filled
4560 with no-op instructions.
4562 The third expression is also absolute, and is also optional. If it is present,
4563 it is the maximum number of bytes that should be skipped by this alignment
4564 directive. If doing the alignment would require skipping more bytes than the
4565 specified maximum, then the alignment is not done at all. You can omit the
4566 fill value (the second argument) entirely by simply using two commas after the
4567 required alignment; this can be useful if you want the alignment to be filled
4568 with no-op instructions when appropriate.
4570 The way the required alignment is specified varies from system to system.
4571 For the arc, hppa, i386 using ELF, i860, iq2000, m68k, or1k,
4572 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4573 alignment request in bytes. For example @samp{.align 8} advances
4574 the location counter until it is a multiple of 8. If the location counter
4575 is already a multiple of 8, no change is needed. For the tic54x, the
4576 first expression is the alignment request in words.
4578 For other systems, including ppc, i386 using a.out format, arm and
4579 strongarm, it is the
4580 number of low-order zero bits the location counter must have after
4581 advancement. For example @samp{.align 3} advances the location
4582 counter until it a multiple of 8. If the location counter is already a
4583 multiple of 8, no change is needed.
4585 This inconsistency is due to the different behaviors of the various
4586 native assemblers for these systems which GAS must emulate.
4587 GAS also provides @code{.balign} and @code{.p2align} directives,
4588 described later, which have a consistent behavior across all
4589 architectures (but are specific to GAS).
4592 @section @code{.altmacro}
4593 Enable alternate macro mode, enabling:
4596 @item LOCAL @var{name} [ , @dots{} ]
4597 One additional directive, @code{LOCAL}, is available. It is used to
4598 generate a string replacement for each of the @var{name} arguments, and
4599 replace any instances of @var{name} in each macro expansion. The
4600 replacement string is unique in the assembly, and different for each
4601 separate macro expansion. @code{LOCAL} allows you to write macros that
4602 define symbols, without fear of conflict between separate macro expansions.
4604 @item String delimiters
4605 You can write strings delimited in these other ways besides
4606 @code{"@var{string}"}:
4609 @item '@var{string}'
4610 You can delimit strings with single-quote characters.
4612 @item <@var{string}>
4613 You can delimit strings with matching angle brackets.
4616 @item single-character string escape
4617 To include any single character literally in a string (even if the
4618 character would otherwise have some special meaning), you can prefix the
4619 character with @samp{!} (an exclamation mark). For example, you can
4620 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4622 @item Expression results as strings
4623 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4624 and use the result as a string.
4628 @section @code{.ascii "@var{string}"}@dots{}
4630 @cindex @code{ascii} directive
4631 @cindex string literals
4632 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4633 separated by commas. It assembles each string (with no automatic
4634 trailing zero byte) into consecutive addresses.
4637 @section @code{.asciz "@var{string}"}@dots{}
4639 @cindex @code{asciz} directive
4640 @cindex zero-terminated strings
4641 @cindex null-terminated strings
4642 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4643 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4646 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4648 @cindex padding the location counter given number of bytes
4649 @cindex @code{balign} directive
4650 Pad the location counter (in the current subsection) to a particular
4651 storage boundary. The first expression (which must be absolute) is the
4652 alignment request in bytes. For example @samp{.balign 8} advances
4653 the location counter until it is a multiple of 8. If the location counter
4654 is already a multiple of 8, no change is needed.
4656 The second expression (also absolute) gives the fill value to be stored in the
4657 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4658 padding bytes are normally zero. However, on some systems, if the section is
4659 marked as containing code and the fill value is omitted, the space is filled
4660 with no-op instructions.
4662 The third expression is also absolute, and is also optional. If it is present,
4663 it is the maximum number of bytes that should be skipped by this alignment
4664 directive. If doing the alignment would require skipping more bytes than the
4665 specified maximum, then the alignment is not done at all. You can omit the
4666 fill value (the second argument) entirely by simply using two commas after the
4667 required alignment; this can be useful if you want the alignment to be filled
4668 with no-op instructions when appropriate.
4670 @cindex @code{balignw} directive
4671 @cindex @code{balignl} directive
4672 The @code{.balignw} and @code{.balignl} directives are variants of the
4673 @code{.balign} directive. The @code{.balignw} directive treats the fill
4674 pattern as a two byte word value. The @code{.balignl} directives treats the
4675 fill pattern as a four byte longword value. For example, @code{.balignw
4676 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4677 filled in with the value 0x368d (the exact placement of the bytes depends upon
4678 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4681 @node Bundle directives
4682 @section Bundle directives
4683 @subsection @code{.bundle_align_mode @var{abs-expr}}
4684 @cindex @code{bundle_align_mode} directive
4686 @cindex instruction bundle
4687 @cindex aligned instruction bundle
4688 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4689 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4690 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4691 disabled (which is the default state). If the argument it not zero, it
4692 gives the size of an instruction bundle as a power of two (as for the
4693 @code{.p2align} directive, @pxref{P2align}).
4695 For some targets, it's an ABI requirement that no instruction may span a
4696 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4697 instructions that starts on an aligned boundary. For example, if
4698 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4699 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4700 effect, no single instruction may span a boundary between bundles. If an
4701 instruction would start too close to the end of a bundle for the length of
4702 that particular instruction to fit within the bundle, then the space at the
4703 end of that bundle is filled with no-op instructions so the instruction
4704 starts in the next bundle. As a corollary, it's an error if any single
4705 instruction's encoding is longer than the bundle size.
4707 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4708 @cindex @code{bundle_lock} directive
4709 @cindex @code{bundle_unlock} directive
4710 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4711 allow explicit control over instruction bundle padding. These directives
4712 are only valid when @code{.bundle_align_mode} has been used to enable
4713 aligned instruction bundle mode. It's an error if they appear when
4714 @code{.bundle_align_mode} has not been used at all, or when the last
4715 directive was @w{@code{.bundle_align_mode 0}}.
4717 @cindex bundle-locked
4718 For some targets, it's an ABI requirement that certain instructions may
4719 appear only as part of specified permissible sequences of multiple
4720 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4721 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4722 instruction sequence. For purposes of aligned instruction bundle mode, a
4723 sequence starting with @code{.bundle_lock} and ending with
4724 @code{.bundle_unlock} is treated as a single instruction. That is, the
4725 entire sequence must fit into a single bundle and may not span a bundle
4726 boundary. If necessary, no-op instructions will be inserted before the
4727 first instruction of the sequence so that the whole sequence starts on an
4728 aligned bundle boundary. It's an error if the sequence is longer than the
4731 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4732 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4733 nested. That is, a second @code{.bundle_lock} directive before the next
4734 @code{.bundle_unlock} directive has no effect except that it must be
4735 matched by another closing @code{.bundle_unlock} so that there is the
4736 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4739 @section @code{.byte @var{expressions}}
4741 @cindex @code{byte} directive
4742 @cindex integers, one byte
4743 @code{.byte} expects zero or more expressions, separated by commas.
4744 Each expression is assembled into the next byte.
4746 @node CFI directives
4747 @section CFI directives
4748 @subsection @code{.cfi_sections @var{section_list}}
4749 @cindex @code{cfi_sections} directive
4750 @code{.cfi_sections} may be used to specify whether CFI directives
4751 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4752 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4753 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4754 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4755 directive is not used is @code{.cfi_sections .eh_frame}.
4757 On targets that support compact unwinding tables these can be generated
4758 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4760 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4761 which is used by the @value{TIC6X} target.
4763 The @code{.cfi_sections} directive can be repeated, with the same or different
4764 arguments, provided that CFI generation has not yet started. Once CFI
4765 generation has started however the section list is fixed and any attempts to
4766 redefine it will result in an error.
4768 @subsection @code{.cfi_startproc [simple]}
4769 @cindex @code{cfi_startproc} directive
4770 @code{.cfi_startproc} is used at the beginning of each function that
4771 should have an entry in @code{.eh_frame}. It initializes some internal
4772 data structures. Don't forget to close the function by
4773 @code{.cfi_endproc}.
4775 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4776 it also emits some architecture dependent initial CFI instructions.
4778 @subsection @code{.cfi_endproc}
4779 @cindex @code{cfi_endproc} directive
4780 @code{.cfi_endproc} is used at the end of a function where it closes its
4781 unwind entry previously opened by
4782 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4784 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4785 @cindex @code{cfi_personality} directive
4786 @code{.cfi_personality} defines personality routine and its encoding.
4787 @var{encoding} must be a constant determining how the personality
4788 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4789 argument is not present, otherwise second argument should be
4790 a constant or a symbol name. When using indirect encodings,
4791 the symbol provided should be the location where personality
4792 can be loaded from, not the personality routine itself.
4793 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4794 no personality routine.
4796 @subsection @code{.cfi_personality_id @var{id}}
4797 @cindex @code{cfi_personality_id} directive
4798 @code{cfi_personality_id} defines a personality routine by its index as
4799 defined in a compact unwinding format.
4800 Only valid when generating compact EH frames (i.e.
4801 with @code{.cfi_sections eh_frame_entry}.
4803 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4804 @cindex @code{cfi_fde_data} directive
4805 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4806 used for the current function. These are emitted inline in the
4807 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4808 in the @code{.gnu.extab} section otherwise.
4809 Only valid when generating compact EH frames (i.e.
4810 with @code{.cfi_sections eh_frame_entry}.
4812 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4813 @code{.cfi_lsda} defines LSDA and its encoding.
4814 @var{encoding} must be a constant determining how the LSDA
4815 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4816 argument is not present, otherwise the second argument should be a constant
4817 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4818 meaning that no LSDA is present.
4820 @subsection @code{.cfi_inline_lsda} [@var{align}]
4821 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4822 switches to the corresponding @code{.gnu.extab} section.
4823 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4824 Only valid when generating compact EH frames (i.e.
4825 with @code{.cfi_sections eh_frame_entry}.
4827 The table header and unwinding opcodes will be generated at this point,
4828 so that they are immediately followed by the LSDA data. The symbol
4829 referenced by the @code{.cfi_lsda} directive should still be defined
4830 in case a fallback FDE based encoding is used. The LSDA data is terminated
4831 by a section directive.
4833 The optional @var{align} argument specifies the alignment required.
4834 The alignment is specified as a power of two, as with the
4835 @code{.p2align} directive.
4837 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4838 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4839 address from @var{register} and add @var{offset} to it}.
4841 @subsection @code{.cfi_def_cfa_register @var{register}}
4842 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4843 now on @var{register} will be used instead of the old one. Offset
4846 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4847 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4848 remains the same, but @var{offset} is new. Note that it is the
4849 absolute offset that will be added to a defined register to compute
4852 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4853 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4854 value that is added/substracted from the previous offset.
4856 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4857 Previous value of @var{register} is saved at offset @var{offset} from
4860 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4861 Previous value of @var{register} is CFA + @var{offset}.
4863 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4864 Previous value of @var{register} is saved at offset @var{offset} from
4865 the current CFA register. This is transformed to @code{.cfi_offset}
4866 using the known displacement of the CFA register from the CFA.
4867 This is often easier to use, because the number will match the
4868 code it's annotating.
4870 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4871 Previous value of @var{register1} is saved in register @var{register2}.
4873 @subsection @code{.cfi_restore @var{register}}
4874 @code{.cfi_restore} says that the rule for @var{register} is now the
4875 same as it was at the beginning of the function, after all initial
4876 instruction added by @code{.cfi_startproc} were executed.
4878 @subsection @code{.cfi_undefined @var{register}}
4879 From now on the previous value of @var{register} can't be restored anymore.
4881 @subsection @code{.cfi_same_value @var{register}}
4882 Current value of @var{register} is the same like in the previous frame,
4883 i.e. no restoration needed.
4885 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4886 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4887 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4888 places them in the current row. This is useful for situations where you have
4889 multiple @code{.cfi_*} directives that need to be undone due to the control
4890 flow of the program. For example, we could have something like this (assuming
4891 the CFA is the value of @code{rbp}):
4901 .cfi_def_cfa %rsp, 8
4904 /* Do something else */
4907 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4908 to the instructions before @code{label}. This means we'd have to add multiple
4909 @code{.cfi} directives after @code{label} to recreate the original save
4910 locations of the registers, as well as setting the CFA back to the value of
4911 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4923 .cfi_def_cfa %rsp, 8
4927 /* Do something else */
4930 That way, the rules for the instructions after @code{label} will be the same
4931 as before the first @code{.cfi_restore} without having to use multiple
4932 @code{.cfi} directives.
4934 @subsection @code{.cfi_return_column @var{register}}
4935 Change return column @var{register}, i.e. the return address is either
4936 directly in @var{register} or can be accessed by rules for @var{register}.
4938 @subsection @code{.cfi_signal_frame}
4939 Mark current function as signal trampoline.
4941 @subsection @code{.cfi_window_save}
4942 SPARC register window has been saved.
4944 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4945 Allows the user to add arbitrary bytes to the unwind info. One
4946 might use this to add OS-specific CFI opcodes, or generic CFI
4947 opcodes that GAS does not yet support.
4949 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4950 The current value of @var{register} is @var{label}. The value of @var{label}
4951 will be encoded in the output file according to @var{encoding}; see the
4952 description of @code{.cfi_personality} for details on this encoding.
4954 The usefulness of equating a register to a fixed label is probably
4955 limited to the return address register. Here, it can be useful to
4956 mark a code segment that has only one return address which is reached
4957 by a direct branch and no copy of the return address exists in memory
4958 or another register.
4961 @section @code{.comm @var{symbol} , @var{length} }
4963 @cindex @code{comm} directive
4964 @cindex symbol, common
4965 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4966 common symbol in one object file may be merged with a defined or common symbol
4967 of the same name in another object file. If @code{@value{LD}} does not see a
4968 definition for the symbol--just one or more common symbols--then it will
4969 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4970 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4971 the same name, and they do not all have the same size, it will allocate space
4972 using the largest size.
4975 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4976 an optional third argument. This is the desired alignment of the symbol,
4977 specified for ELF as a byte boundary (for example, an alignment of 16 means
4978 that the least significant 4 bits of the address should be zero), and for PE
4979 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4980 boundary). The alignment must be an absolute expression, and it must be a
4981 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4982 common symbol, it will use the alignment when placing the symbol. If no
4983 alignment is specified, @command{@value{AS}} will set the alignment to the
4984 largest power of two less than or equal to the size of the symbol, up to a
4985 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4986 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4987 @samp{--section-alignment} option; image file sections in PE are aligned to
4988 multiples of 4096, which is far too large an alignment for ordinary variables.
4989 It is rather the default alignment for (non-debug) sections within object
4990 (@samp{*.o}) files, which are less strictly aligned.}.
4994 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4995 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4999 @section @code{.data @var{subsection}}
5001 @cindex @code{data} directive
5002 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
5003 end of the data subsection numbered @var{subsection} (which is an
5004 absolute expression). If @var{subsection} is omitted, it defaults
5009 @section @code{.def @var{name}}
5011 @cindex @code{def} directive
5012 @cindex COFF symbols, debugging
5013 @cindex debugging COFF symbols
5014 Begin defining debugging information for a symbol @var{name}; the
5015 definition extends until the @code{.endef} directive is encountered.
5018 This directive is only observed when @command{@value{AS}} is configured for COFF
5019 format output; when producing @code{b.out}, @samp{.def} is recognized,
5026 @section @code{.desc @var{symbol}, @var{abs-expression}}
5028 @cindex @code{desc} directive
5029 @cindex COFF symbol descriptor
5030 @cindex symbol descriptor, COFF
5031 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5032 to the low 16 bits of an absolute expression.
5035 The @samp{.desc} directive is not available when @command{@value{AS}} is
5036 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5037 object format. For the sake of compatibility, @command{@value{AS}} accepts
5038 it, but produces no output, when configured for COFF.
5044 @section @code{.dim}
5046 @cindex @code{dim} directive
5047 @cindex COFF auxiliary symbol information
5048 @cindex auxiliary symbol information, COFF
5049 This directive is generated by compilers to include auxiliary debugging
5050 information in the symbol table. It is only permitted inside
5051 @code{.def}/@code{.endef} pairs.
5054 @samp{.dim} is only meaningful when generating COFF format output; when
5055 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
5061 @section @code{.double @var{flonums}}
5063 @cindex @code{double} directive
5064 @cindex floating point numbers (double)
5065 @code{.double} expects zero or more flonums, separated by commas. It
5066 assembles floating point numbers.
5068 The exact kind of floating point numbers emitted depends on how
5069 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5073 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5074 in @sc{ieee} format.
5079 @section @code{.eject}
5081 @cindex @code{eject} directive
5082 @cindex new page, in listings
5083 @cindex page, in listings
5084 @cindex listing control: new page
5085 Force a page break at this point, when generating assembly listings.
5088 @section @code{.else}
5090 @cindex @code{else} directive
5091 @code{.else} is part of the @command{@value{AS}} support for conditional
5092 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5093 of code to be assembled if the condition for the preceding @code{.if}
5097 @section @code{.elseif}
5099 @cindex @code{elseif} directive
5100 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5101 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5102 @code{.if} block that would otherwise fill the entire @code{.else} section.
5105 @section @code{.end}
5107 @cindex @code{end} directive
5108 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5109 process anything in the file past the @code{.end} directive.
5113 @section @code{.endef}
5115 @cindex @code{endef} directive
5116 This directive flags the end of a symbol definition begun with
5120 @samp{.endef} is only meaningful when generating COFF format output; if
5121 @command{@value{AS}} is configured to generate @code{b.out}, it accepts this
5122 directive but ignores it.
5127 @section @code{.endfunc}
5128 @cindex @code{endfunc} directive
5129 @code{.endfunc} marks the end of a function specified with @code{.func}.
5132 @section @code{.endif}
5134 @cindex @code{endif} directive
5135 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5136 it marks the end of a block of code that is only assembled
5137 conditionally. @xref{If,,@code{.if}}.
5140 @section @code{.equ @var{symbol}, @var{expression}}
5142 @cindex @code{equ} directive
5143 @cindex assigning values to symbols
5144 @cindex symbols, assigning values to
5145 This directive sets the value of @var{symbol} to @var{expression}.
5146 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5149 The syntax for @code{equ} on the HPPA is
5150 @samp{@var{symbol} .equ @var{expression}}.
5154 The syntax for @code{equ} on the Z80 is
5155 @samp{@var{symbol} equ @var{expression}}.
5156 On the Z80 it is an eror if @var{symbol} is already defined,
5157 but the symbol is not protected from later redefinition.
5158 Compare @ref{Equiv}.
5162 @section @code{.equiv @var{symbol}, @var{expression}}
5163 @cindex @code{equiv} directive
5164 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5165 the assembler will signal an error if @var{symbol} is already defined. Note a
5166 symbol which has been referenced but not actually defined is considered to be
5169 Except for the contents of the error message, this is roughly equivalent to
5176 plus it protects the symbol from later redefinition.
5179 @section @code{.eqv @var{symbol}, @var{expression}}
5180 @cindex @code{eqv} directive
5181 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5182 evaluate the expression or any part of it immediately. Instead each time
5183 the resulting symbol is used in an expression, a snapshot of its current
5187 @section @code{.err}
5188 @cindex @code{err} directive
5189 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5190 message and, unless the @option{-Z} option was used, it will not generate an
5191 object file. This can be used to signal an error in conditionally compiled code.
5194 @section @code{.error "@var{string}"}
5195 @cindex error directive
5197 Similarly to @code{.err}, this directive emits an error, but you can specify a
5198 string that will be emitted as the error message. If you don't specify the
5199 message, it defaults to @code{".error directive invoked in source file"}.
5200 @xref{Errors, ,Error and Warning Messages}.
5203 .error "This code has not been assembled and tested."
5207 @section @code{.exitm}
5208 Exit early from the current macro definition. @xref{Macro}.
5211 @section @code{.extern}
5213 @cindex @code{extern} directive
5214 @code{.extern} is accepted in the source program---for compatibility
5215 with other assemblers---but it is ignored. @command{@value{AS}} treats
5216 all undefined symbols as external.
5219 @section @code{.fail @var{expression}}
5221 @cindex @code{fail} directive
5222 Generates an error or a warning. If the value of the @var{expression} is 500
5223 or more, @command{@value{AS}} will print a warning message. If the value is less
5224 than 500, @command{@value{AS}} will print an error message. The message will
5225 include the value of @var{expression}. This can occasionally be useful inside
5226 complex nested macros or conditional assembly.
5229 @section @code{.file}
5230 @cindex @code{file} directive
5232 @ifclear no-file-dir
5233 There are two different versions of the @code{.file} directive. Targets
5234 that support DWARF2 line number information use the DWARF2 version of
5235 @code{.file}. Other targets use the default version.
5237 @subheading Default Version
5239 @cindex logical file name
5240 @cindex file name, logical
5241 This version of the @code{.file} directive tells @command{@value{AS}} that we
5242 are about to start a new logical file. The syntax is:
5248 @var{string} is the new file name. In general, the filename is
5249 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5250 to specify an empty file name, you must give the quotes--@code{""}. This
5251 statement may go away in future: it is only recognized to be compatible with
5252 old @command{@value{AS}} programs.
5254 @subheading DWARF2 Version
5257 When emitting DWARF2 line number information, @code{.file} assigns filenames
5258 to the @code{.debug_line} file name table. The syntax is:
5261 .file @var{fileno} @var{filename}
5264 The @var{fileno} operand should be a unique positive integer to use as the
5265 index of the entry in the table. The @var{filename} operand is a C string
5268 The detail of filename indices is exposed to the user because the filename
5269 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5270 information, and thus the user must know the exact indices that table
5274 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5276 @cindex @code{fill} directive
5277 @cindex writing patterns in memory
5278 @cindex patterns, writing in memory
5279 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5280 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5281 may be zero or more. @var{Size} may be zero or more, but if it is
5282 more than 8, then it is deemed to have the value 8, compatible with
5283 other people's assemblers. The contents of each @var{repeat} bytes
5284 is taken from an 8-byte number. The highest order 4 bytes are
5285 zero. The lowest order 4 bytes are @var{value} rendered in the
5286 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5287 Each @var{size} bytes in a repetition is taken from the lowest order
5288 @var{size} bytes of this number. Again, this bizarre behavior is
5289 compatible with other people's assemblers.
5291 @var{size} and @var{value} are optional.
5292 If the second comma and @var{value} are absent, @var{value} is
5293 assumed zero. If the first comma and following tokens are absent,
5294 @var{size} is assumed to be 1.
5297 @section @code{.float @var{flonums}}
5299 @cindex floating point numbers (single)
5300 @cindex @code{float} directive
5301 This directive assembles zero or more flonums, separated by commas. It
5302 has the same effect as @code{.single}.
5304 The exact kind of floating point numbers emitted depends on how
5305 @command{@value{AS}} is configured.
5306 @xref{Machine Dependencies}.
5310 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5311 in @sc{ieee} format.
5316 @section @code{.func @var{name}[,@var{label}]}
5317 @cindex @code{func} directive
5318 @code{.func} emits debugging information to denote function @var{name}, and
5319 is ignored unless the file is assembled with debugging enabled.
5320 Only @samp{--gstabs[+]} is currently supported.
5321 @var{label} is the entry point of the function and if omitted @var{name}
5322 prepended with the @samp{leading char} is used.
5323 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5324 All functions are currently defined to have @code{void} return type.
5325 The function must be terminated with @code{.endfunc}.
5328 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5330 @cindex @code{global} directive
5331 @cindex symbol, making visible to linker
5332 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5333 @var{symbol} in your partial program, its value is made available to
5334 other partial programs that are linked with it. Otherwise,
5335 @var{symbol} takes its attributes from a symbol of the same name
5336 from another file linked into the same program.
5338 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5339 compatibility with other assemblers.
5342 On the HPPA, @code{.global} is not always enough to make it accessible to other
5343 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5344 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5349 @section @code{.gnu_attribute @var{tag},@var{value}}
5350 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5353 @section @code{.hidden @var{names}}
5355 @cindex @code{hidden} directive
5357 This is one of the ELF visibility directives. The other two are
5358 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5359 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5361 This directive overrides the named symbols default visibility (which is set by
5362 their binding: local, global or weak). The directive sets the visibility to
5363 @code{hidden} which means that the symbols are not visible to other components.
5364 Such symbols are always considered to be @code{protected} as well.
5368 @section @code{.hword @var{expressions}}
5370 @cindex @code{hword} directive
5371 @cindex integers, 16-bit
5372 @cindex numbers, 16-bit
5373 @cindex sixteen bit integers
5374 This expects zero or more @var{expressions}, and emits
5375 a 16 bit number for each.
5378 This directive is a synonym for @samp{.short}; depending on the target
5379 architecture, it may also be a synonym for @samp{.word}.
5383 This directive is a synonym for @samp{.short}.
5386 This directive is a synonym for both @samp{.short} and @samp{.word}.
5391 @section @code{.ident}
5393 @cindex @code{ident} directive
5395 This directive is used by some assemblers to place tags in object files. The
5396 behavior of this directive varies depending on the target. When using the
5397 a.out object file format, @command{@value{AS}} simply accepts the directive for
5398 source-file compatibility with existing assemblers, but does not emit anything
5399 for it. When using COFF, comments are emitted to the @code{.comment} or
5400 @code{.rdata} section, depending on the target. When using ELF, comments are
5401 emitted to the @code{.comment} section.
5404 @section @code{.if @var{absolute expression}}
5406 @cindex conditional assembly
5407 @cindex @code{if} directive
5408 @code{.if} marks the beginning of a section of code which is only
5409 considered part of the source program being assembled if the argument
5410 (which must be an @var{absolute expression}) is non-zero. The end of
5411 the conditional section of code must be marked by @code{.endif}
5412 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5413 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5414 If you have several conditions to check, @code{.elseif} may be used to avoid
5415 nesting blocks if/else within each subsequent @code{.else} block.
5417 The following variants of @code{.if} are also supported:
5419 @cindex @code{ifdef} directive
5420 @item .ifdef @var{symbol}
5421 Assembles the following section of code if the specified @var{symbol}
5422 has been defined. Note a symbol which has been referenced but not yet defined
5423 is considered to be undefined.
5425 @cindex @code{ifb} directive
5426 @item .ifb @var{text}
5427 Assembles the following section of code if the operand is blank (empty).
5429 @cindex @code{ifc} directive
5430 @item .ifc @var{string1},@var{string2}
5431 Assembles the following section of code if the two strings are the same. The
5432 strings may be optionally quoted with single quotes. If they are not quoted,
5433 the first string stops at the first comma, and the second string stops at the
5434 end of the line. Strings which contain whitespace should be quoted. The
5435 string comparison is case sensitive.
5437 @cindex @code{ifeq} directive
5438 @item .ifeq @var{absolute expression}
5439 Assembles the following section of code if the argument is zero.
5441 @cindex @code{ifeqs} directive
5442 @item .ifeqs @var{string1},@var{string2}
5443 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5445 @cindex @code{ifge} directive
5446 @item .ifge @var{absolute expression}
5447 Assembles the following section of code if the argument is greater than or
5450 @cindex @code{ifgt} directive
5451 @item .ifgt @var{absolute expression}
5452 Assembles the following section of code if the argument is greater than zero.
5454 @cindex @code{ifle} directive
5455 @item .ifle @var{absolute expression}
5456 Assembles the following section of code if the argument is less than or equal
5459 @cindex @code{iflt} directive
5460 @item .iflt @var{absolute expression}
5461 Assembles the following section of code if the argument is less than zero.
5463 @cindex @code{ifnb} directive
5464 @item .ifnb @var{text}
5465 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5466 following section of code if the operand is non-blank (non-empty).
5468 @cindex @code{ifnc} directive
5469 @item .ifnc @var{string1},@var{string2}.
5470 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5471 following section of code if the two strings are not the same.
5473 @cindex @code{ifndef} directive
5474 @cindex @code{ifnotdef} directive
5475 @item .ifndef @var{symbol}
5476 @itemx .ifnotdef @var{symbol}
5477 Assembles the following section of code if the specified @var{symbol}
5478 has not been defined. Both spelling variants are equivalent. Note a symbol
5479 which has been referenced but not yet defined is considered to be undefined.
5481 @cindex @code{ifne} directive
5482 @item .ifne @var{absolute expression}
5483 Assembles the following section of code if the argument is not equal to zero
5484 (in other words, this is equivalent to @code{.if}).
5486 @cindex @code{ifnes} directive
5487 @item .ifnes @var{string1},@var{string2}
5488 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5489 following section of code if the two strings are not the same.
5493 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5495 @cindex @code{incbin} directive
5496 @cindex binary files, including
5497 The @code{incbin} directive includes @var{file} verbatim at the current
5498 location. You can control the search paths used with the @samp{-I} command-line
5499 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5502 The @var{skip} argument skips a number of bytes from the start of the
5503 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5504 read. Note that the data is not aligned in any way, so it is the user's
5505 responsibility to make sure that proper alignment is provided both before and
5506 after the @code{incbin} directive.
5509 @section @code{.include "@var{file}"}
5511 @cindex @code{include} directive
5512 @cindex supporting files, including
5513 @cindex files, including
5514 This directive provides a way to include supporting files at specified
5515 points in your source program. The code from @var{file} is assembled as
5516 if it followed the point of the @code{.include}; when the end of the
5517 included file is reached, assembly of the original file continues. You
5518 can control the search paths used with the @samp{-I} command-line option
5519 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5523 @section @code{.int @var{expressions}}
5525 @cindex @code{int} directive
5526 @cindex integers, 32-bit
5527 Expect zero or more @var{expressions}, of any section, separated by commas.
5528 For each expression, emit a number that, at run time, is the value of that
5529 expression. The byte order and bit size of the number depends on what kind
5530 of target the assembly is for.
5534 On most forms of the H8/300, @code{.int} emits 16-bit
5535 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5542 @section @code{.internal @var{names}}
5544 @cindex @code{internal} directive
5546 This is one of the ELF visibility directives. The other two are
5547 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5548 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5550 This directive overrides the named symbols default visibility (which is set by
5551 their binding: local, global or weak). The directive sets the visibility to
5552 @code{internal} which means that the symbols are considered to be @code{hidden}
5553 (i.e., not visible to other components), and that some extra, processor specific
5554 processing must also be performed upon the symbols as well.
5558 @section @code{.irp @var{symbol},@var{values}}@dots{}
5560 @cindex @code{irp} directive
5561 Evaluate a sequence of statements assigning different values to @var{symbol}.
5562 The sequence of statements starts at the @code{.irp} directive, and is
5563 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5564 set to @var{value}, and the sequence of statements is assembled. If no
5565 @var{value} is listed, the sequence of statements is assembled once, with
5566 @var{symbol} set to the null string. To refer to @var{symbol} within the
5567 sequence of statements, use @var{\symbol}.
5569 For example, assembling
5577 is equivalent to assembling
5585 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5588 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5590 @cindex @code{irpc} directive
5591 Evaluate a sequence of statements assigning different values to @var{symbol}.
5592 The sequence of statements starts at the @code{.irpc} directive, and is
5593 terminated by an @code{.endr} directive. For each character in @var{value},
5594 @var{symbol} is set to the character, and the sequence of statements is
5595 assembled. If no @var{value} is listed, the sequence of statements is
5596 assembled once, with @var{symbol} set to the null string. To refer to
5597 @var{symbol} within the sequence of statements, use @var{\symbol}.
5599 For example, assembling
5607 is equivalent to assembling
5615 For some caveats with the spelling of @var{symbol}, see also the discussion
5619 @section @code{.lcomm @var{symbol} , @var{length}}
5621 @cindex @code{lcomm} directive
5622 @cindex local common symbols
5623 @cindex symbols, local common
5624 Reserve @var{length} (an absolute expression) bytes for a local common
5625 denoted by @var{symbol}. The section and value of @var{symbol} are
5626 those of the new local common. The addresses are allocated in the bss
5627 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5628 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5629 not visible to @code{@value{LD}}.
5632 Some targets permit a third argument to be used with @code{.lcomm}. This
5633 argument specifies the desired alignment of the symbol in the bss section.
5637 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5638 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5642 @section @code{.lflags}
5644 @cindex @code{lflags} directive (ignored)
5645 @command{@value{AS}} accepts this directive, for compatibility with other
5646 assemblers, but ignores it.
5648 @ifclear no-line-dir
5650 @section @code{.line @var{line-number}}
5652 @cindex @code{line} directive
5653 @cindex logical line number
5655 Change the logical line number. @var{line-number} must be an absolute
5656 expression. The next line has that logical line number. Therefore any other
5657 statements on the current line (after a statement separator character) are
5658 reported as on logical line number @var{line-number} @minus{} 1. One day
5659 @command{@value{AS}} will no longer support this directive: it is recognized only
5660 for compatibility with existing assembler programs.
5663 Even though this is a directive associated with the @code{a.out} or
5664 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5665 when producing COFF output, and treats @samp{.line} as though it
5666 were the COFF @samp{.ln} @emph{if} it is found outside a
5667 @code{.def}/@code{.endef} pair.
5669 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5670 used by compilers to generate auxiliary symbol information for
5675 @section @code{.linkonce [@var{type}]}
5677 @cindex @code{linkonce} directive
5678 @cindex common sections
5679 Mark the current section so that the linker only includes a single copy of it.
5680 This may be used to include the same section in several different object files,
5681 but ensure that the linker will only include it once in the final output file.
5682 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5683 Duplicate sections are detected based on the section name, so it should be
5686 This directive is only supported by a few object file formats; as of this
5687 writing, the only object file format which supports it is the Portable
5688 Executable format used on Windows NT.
5690 The @var{type} argument is optional. If specified, it must be one of the
5691 following strings. For example:
5695 Not all types may be supported on all object file formats.
5699 Silently discard duplicate sections. This is the default.
5702 Warn if there are duplicate sections, but still keep only one copy.
5705 Warn if any of the duplicates have different sizes.
5708 Warn if any of the duplicates do not have exactly the same contents.
5712 @section @code{.list}
5714 @cindex @code{list} directive
5715 @cindex listing control, turning on
5716 Control (in conjunction with the @code{.nolist} directive) whether or
5717 not assembly listings are generated. These two directives maintain an
5718 internal counter (which is zero initially). @code{.list} increments the
5719 counter, and @code{.nolist} decrements it. Assembly listings are
5720 generated whenever the counter is greater than zero.
5722 By default, listings are disabled. When you enable them (with the
5723 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5724 the initial value of the listing counter is one.
5727 @section @code{.ln @var{line-number}}
5729 @cindex @code{ln} directive
5730 @ifclear no-line-dir
5731 @samp{.ln} is a synonym for @samp{.line}.
5734 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5735 must be an absolute expression. The next line has that logical
5736 line number, so any other statements on the current line (after a
5737 statement separator character @code{;}) are reported as on logical
5738 line number @var{line-number} @minus{} 1.
5741 This directive is accepted, but ignored, when @command{@value{AS}} is
5742 configured for @code{b.out}; its effect is only associated with COFF
5748 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5749 @cindex @code{loc} directive
5750 When emitting DWARF2 line number information,
5751 the @code{.loc} directive will add a row to the @code{.debug_line} line
5752 number matrix corresponding to the immediately following assembly
5753 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5754 arguments will be applied to the @code{.debug_line} state machine before
5757 The @var{options} are a sequence of the following tokens in any order:
5761 This option will set the @code{basic_block} register in the
5762 @code{.debug_line} state machine to @code{true}.
5765 This option will set the @code{prologue_end} register in the
5766 @code{.debug_line} state machine to @code{true}.
5768 @item epilogue_begin
5769 This option will set the @code{epilogue_begin} register in the
5770 @code{.debug_line} state machine to @code{true}.
5772 @item is_stmt @var{value}
5773 This option will set the @code{is_stmt} register in the
5774 @code{.debug_line} state machine to @code{value}, which must be
5777 @item isa @var{value}
5778 This directive will set the @code{isa} register in the @code{.debug_line}
5779 state machine to @var{value}, which must be an unsigned integer.
5781 @item discriminator @var{value}
5782 This directive will set the @code{discriminator} register in the @code{.debug_line}
5783 state machine to @var{value}, which must be an unsigned integer.
5787 @node Loc_mark_labels
5788 @section @code{.loc_mark_labels @var{enable}}
5789 @cindex @code{loc_mark_labels} directive
5790 When emitting DWARF2 line number information,
5791 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5792 to the @code{.debug_line} line number matrix with the @code{basic_block}
5793 register in the state machine set whenever a code label is seen.
5794 The @var{enable} argument should be either 1 or 0, to enable or disable
5795 this function respectively.
5799 @section @code{.local @var{names}}
5801 @cindex @code{local} directive
5802 This directive, which is available for ELF targets, marks each symbol in
5803 the comma-separated list of @code{names} as a local symbol so that it
5804 will not be externally visible. If the symbols do not already exist,
5805 they will be created.
5807 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5808 accept an alignment argument, which is the case for most ELF targets,
5809 the @code{.local} directive can be used in combination with @code{.comm}
5810 (@pxref{Comm}) to define aligned local common data.
5814 @section @code{.long @var{expressions}}
5816 @cindex @code{long} directive
5817 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5820 @c no one seems to know what this is for or whether this description is
5821 @c what it really ought to do
5823 @section @code{.lsym @var{symbol}, @var{expression}}
5825 @cindex @code{lsym} directive
5826 @cindex symbol, not referenced in assembly
5827 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5828 the hash table, ensuring it cannot be referenced by name during the
5829 rest of the assembly. This sets the attributes of the symbol to be
5830 the same as the expression value:
5832 @var{other} = @var{descriptor} = 0
5833 @var{type} = @r{(section of @var{expression})}
5834 @var{value} = @var{expression}
5837 The new symbol is not flagged as external.
5841 @section @code{.macro}
5844 The commands @code{.macro} and @code{.endm} allow you to define macros that
5845 generate assembly output. For example, this definition specifies a macro
5846 @code{sum} that puts a sequence of numbers into memory:
5849 .macro sum from=0, to=5
5858 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5870 @item .macro @var{macname}
5871 @itemx .macro @var{macname} @var{macargs} @dots{}
5872 @cindex @code{macro} directive
5873 Begin the definition of a macro called @var{macname}. If your macro
5874 definition requires arguments, specify their names after the macro name,
5875 separated by commas or spaces. You can qualify the macro argument to
5876 indicate whether all invocations must specify a non-blank value (through
5877 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5878 (through @samp{:@code{vararg}}). You can supply a default value for any
5879 macro argument by following the name with @samp{=@var{deflt}}. You
5880 cannot define two macros with the same @var{macname} unless it has been
5881 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5882 definitions. For example, these are all valid @code{.macro} statements:
5886 Begin the definition of a macro called @code{comm}, which takes no
5889 @item .macro plus1 p, p1
5890 @itemx .macro plus1 p p1
5891 Either statement begins the definition of a macro called @code{plus1},
5892 which takes two arguments; within the macro definition, write
5893 @samp{\p} or @samp{\p1} to evaluate the arguments.
5895 @item .macro reserve_str p1=0 p2
5896 Begin the definition of a macro called @code{reserve_str}, with two
5897 arguments. The first argument has a default value, but not the second.
5898 After the definition is complete, you can call the macro either as
5899 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5900 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5901 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5902 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5904 @item .macro m p1:req, p2=0, p3:vararg
5905 Begin the definition of a macro called @code{m}, with at least three
5906 arguments. The first argument must always have a value specified, but
5907 not the second, which instead has a default value. The third formal
5908 will get assigned all remaining arguments specified at invocation time.
5910 When you call a macro, you can specify the argument values either by
5911 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5912 @samp{sum to=17, from=9}.
5916 Note that since each of the @var{macargs} can be an identifier exactly
5917 as any other one permitted by the target architecture, there may be
5918 occasional problems if the target hand-crafts special meanings to certain
5919 characters when they occur in a special position. For example, if the colon
5920 (@code{:}) is generally permitted to be part of a symbol name, but the
5921 architecture specific code special-cases it when occurring as the final
5922 character of a symbol (to denote a label), then the macro parameter
5923 replacement code will have no way of knowing that and consider the whole
5924 construct (including the colon) an identifier, and check only this
5925 identifier for being the subject to parameter substitution. So for example
5926 this macro definition:
5934 might not work as expected. Invoking @samp{label foo} might not create a label
5935 called @samp{foo} but instead just insert the text @samp{\l:} into the
5936 assembler source, probably generating an error about an unrecognised
5939 Similarly problems might occur with the period character (@samp{.})
5940 which is often allowed inside opcode names (and hence identifier names). So
5941 for example constructing a macro to build an opcode from a base name and a
5942 length specifier like this:
5945 .macro opcode base length
5950 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5951 instruction but instead generate some kind of error as the assembler tries to
5952 interpret the text @samp{\base.\length}.
5954 There are several possible ways around this problem:
5957 @item Insert white space
5958 If it is possible to use white space characters then this is the simplest
5967 @item Use @samp{\()}
5968 The string @samp{\()} can be used to separate the end of a macro argument from
5969 the following text. eg:
5972 .macro opcode base length
5977 @item Use the alternate macro syntax mode
5978 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5979 used as a separator. eg:
5989 Note: this problem of correctly identifying string parameters to pseudo ops
5990 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5991 and @code{.irpc} (@pxref{Irpc}) as well.
5994 @cindex @code{endm} directive
5995 Mark the end of a macro definition.
5998 @cindex @code{exitm} directive
5999 Exit early from the current macro definition.
6001 @cindex number of macros executed
6002 @cindex macros, count executed
6004 @command{@value{AS}} maintains a counter of how many macros it has
6005 executed in this pseudo-variable; you can copy that number to your
6006 output with @samp{\@@}, but @emph{only within a macro definition}.
6008 @item LOCAL @var{name} [ , @dots{} ]
6009 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
6010 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
6011 @xref{Altmacro,,@code{.altmacro}}.
6015 @section @code{.mri @var{val}}
6017 @cindex @code{mri} directive
6018 @cindex MRI mode, temporarily
6019 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
6020 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
6021 affects code assembled until the next @code{.mri} directive, or until the end
6022 of the file. @xref{M, MRI mode, MRI mode}.
6025 @section @code{.noaltmacro}
6026 Disable alternate macro mode. @xref{Altmacro}.
6029 @section @code{.nolist}
6031 @cindex @code{nolist} directive
6032 @cindex listing control, turning off
6033 Control (in conjunction with the @code{.list} directive) whether or
6034 not assembly listings are generated. These two directives maintain an
6035 internal counter (which is zero initially). @code{.list} increments the
6036 counter, and @code{.nolist} decrements it. Assembly listings are
6037 generated whenever the counter is greater than zero.
6040 @section @code{.octa @var{bignums}}
6042 @c FIXME: double size emitted for "octa" on i960, others? Or warn?
6043 @cindex @code{octa} directive
6044 @cindex integer, 16-byte
6045 @cindex sixteen byte integer
6046 This directive expects zero or more bignums, separated by commas. For each
6047 bignum, it emits a 16-byte integer.
6049 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6050 hence @emph{octa}-word for 16 bytes.
6053 @section @code{.offset @var{loc}}
6055 @cindex @code{offset} directive
6056 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6057 be an absolute expression. This directive may be useful for defining
6058 symbols with absolute values. Do not confuse it with the @code{.org}
6062 @section @code{.org @var{new-lc} , @var{fill}}
6064 @cindex @code{org} directive
6065 @cindex location counter, advancing
6066 @cindex advancing location counter
6067 @cindex current address, advancing
6068 Advance the location counter of the current section to
6069 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6070 expression with the same section as the current subsection. That is,
6071 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6072 wrong section, the @code{.org} directive is ignored. To be compatible
6073 with former assemblers, if the section of @var{new-lc} is absolute,
6074 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6075 is the same as the current subsection.
6077 @code{.org} may only increase the location counter, or leave it
6078 unchanged; you cannot use @code{.org} to move the location counter
6081 @c double negative used below "not undefined" because this is a specific
6082 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6083 @c section. doc@cygnus.com 18feb91
6084 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6085 may not be undefined. If you really detest this restriction we eagerly await
6086 a chance to share your improved assembler.
6088 Beware that the origin is relative to the start of the section, not
6089 to the start of the subsection. This is compatible with other
6090 people's assemblers.
6092 When the location counter (of the current subsection) is advanced, the
6093 intervening bytes are filled with @var{fill} which should be an
6094 absolute expression. If the comma and @var{fill} are omitted,
6095 @var{fill} defaults to zero.
6098 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
6100 @cindex padding the location counter given a power of two
6101 @cindex @code{p2align} directive
6102 Pad the location counter (in the current subsection) to a particular
6103 storage boundary. The first expression (which must be absolute) is the
6104 number of low-order zero bits the location counter must have after
6105 advancement. For example @samp{.p2align 3} advances the location
6106 counter until it a multiple of 8. If the location counter is already a
6107 multiple of 8, no change is needed.
6109 The second expression (also absolute) gives the fill value to be stored in the
6110 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6111 padding bytes are normally zero. However, on some systems, if the section is
6112 marked as containing code and the fill value is omitted, the space is filled
6113 with no-op instructions.
6115 The third expression is also absolute, and is also optional. If it is present,
6116 it is the maximum number of bytes that should be skipped by this alignment
6117 directive. If doing the alignment would require skipping more bytes than the
6118 specified maximum, then the alignment is not done at all. You can omit the
6119 fill value (the second argument) entirely by simply using two commas after the
6120 required alignment; this can be useful if you want the alignment to be filled
6121 with no-op instructions when appropriate.
6123 @cindex @code{p2alignw} directive
6124 @cindex @code{p2alignl} directive
6125 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6126 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6127 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6128 fill pattern as a four byte longword value. For example, @code{.p2alignw
6129 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6130 filled in with the value 0x368d (the exact placement of the bytes depends upon
6131 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6136 @section @code{.popsection}
6138 @cindex @code{popsection} directive
6139 @cindex Section Stack
6140 This is one of the ELF section stack manipulation directives. The others are
6141 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6142 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6145 This directive replaces the current section (and subsection) with the top
6146 section (and subsection) on the section stack. This section is popped off the
6152 @section @code{.previous}
6154 @cindex @code{previous} directive
6155 @cindex Section Stack
6156 This is one of the ELF section stack manipulation directives. The others are
6157 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6158 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6159 (@pxref{PopSection}).
6161 This directive swaps the current section (and subsection) with most recently
6162 referenced section/subsection pair prior to this one. Multiple
6163 @code{.previous} directives in a row will flip between two sections (and their
6164 subsections). For example:
6176 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6182 # Now in section A subsection 1
6186 # Now in section B subsection 0
6189 # Now in section B subsection 1
6192 # Now in section B subsection 0
6196 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6197 section B and 0x9abc into subsection 1 of section B.
6199 In terms of the section stack, this directive swaps the current section with
6200 the top section on the section stack.
6204 @section @code{.print @var{string}}
6206 @cindex @code{print} directive
6207 @command{@value{AS}} will print @var{string} on the standard output during
6208 assembly. You must put @var{string} in double quotes.
6212 @section @code{.protected @var{names}}
6214 @cindex @code{protected} directive
6216 This is one of the ELF visibility directives. The other two are
6217 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6219 This directive overrides the named symbols default visibility (which is set by
6220 their binding: local, global or weak). The directive sets the visibility to
6221 @code{protected} which means that any references to the symbols from within the
6222 components that defines them must be resolved to the definition in that
6223 component, even if a definition in another component would normally preempt
6228 @section @code{.psize @var{lines} , @var{columns}}
6230 @cindex @code{psize} directive
6231 @cindex listing control: paper size
6232 @cindex paper size, for listings
6233 Use this directive to declare the number of lines---and, optionally, the
6234 number of columns---to use for each page, when generating listings.
6236 If you do not use @code{.psize}, listings use a default line-count
6237 of 60. You may omit the comma and @var{columns} specification; the
6238 default width is 200 columns.
6240 @command{@value{AS}} generates formfeeds whenever the specified number of
6241 lines is exceeded (or whenever you explicitly request one, using
6244 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6245 those explicitly specified with @code{.eject}.
6248 @section @code{.purgem @var{name}}
6250 @cindex @code{purgem} directive
6251 Undefine the macro @var{name}, so that later uses of the string will not be
6252 expanded. @xref{Macro}.
6256 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6258 @cindex @code{pushsection} directive
6259 @cindex Section Stack
6260 This is one of the ELF section stack manipulation directives. The others are
6261 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6262 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6265 This directive pushes the current section (and subsection) onto the
6266 top of the section stack, and then replaces the current section and
6267 subsection with @code{name} and @code{subsection}. The optional
6268 @code{flags}, @code{type} and @code{arguments} are treated the same
6269 as in the @code{.section} (@pxref{Section}) directive.
6273 @section @code{.quad @var{bignums}}
6275 @cindex @code{quad} directive
6276 @code{.quad} expects zero or more bignums, separated by commas. For
6277 each bignum, it emits
6279 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6280 warning message; and just takes the lowest order 8 bytes of the bignum.
6281 @cindex eight-byte integer
6282 @cindex integer, 8-byte
6284 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6285 hence @emph{quad}-word for 8 bytes.
6288 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6289 warning message; and just takes the lowest order 16 bytes of the bignum.
6290 @cindex sixteen-byte integer
6291 @cindex integer, 16-byte
6295 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6297 @cindex @code{reloc} directive
6298 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6299 @var{expression}. If @var{offset} is a number, the relocation is generated in
6300 the current section. If @var{offset} is an expression that resolves to a
6301 symbol plus offset, the relocation is generated in the given symbol's section.
6302 @var{expression}, if present, must resolve to a symbol plus addend or to an
6303 absolute value, but note that not all targets support an addend. e.g. ELF REL
6304 targets such as i386 store an addend in the section contents rather than in the
6305 relocation. This low level interface does not support addends stored in the
6309 @section @code{.rept @var{count}}
6311 @cindex @code{rept} directive
6312 Repeat the sequence of lines between the @code{.rept} directive and the next
6313 @code{.endr} directive @var{count} times.
6315 For example, assembling
6323 is equivalent to assembling
6332 @section @code{.sbttl "@var{subheading}"}
6334 @cindex @code{sbttl} directive
6335 @cindex subtitles for listings
6336 @cindex listing control: subtitle
6337 Use @var{subheading} as the title (third line, immediately after the
6338 title line) when generating assembly listings.
6340 This directive affects subsequent pages, as well as the current page if
6341 it appears within ten lines of the top of a page.
6345 @section @code{.scl @var{class}}
6347 @cindex @code{scl} directive
6348 @cindex symbol storage class (COFF)
6349 @cindex COFF symbol storage class
6350 Set the storage-class value for a symbol. This directive may only be
6351 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6352 whether a symbol is static or external, or it may record further
6353 symbolic debugging information.
6356 The @samp{.scl} directive is primarily associated with COFF output; when
6357 configured to generate @code{b.out} output format, @command{@value{AS}}
6358 accepts this directive but ignores it.
6364 @section @code{.section @var{name}}
6366 @cindex named section
6367 Use the @code{.section} directive to assemble the following code into a section
6370 This directive is only supported for targets that actually support arbitrarily
6371 named sections; on @code{a.out} targets, for example, it is not accepted, even
6372 with a standard @code{a.out} section name.
6376 @c only print the extra heading if both COFF and ELF are set
6377 @subheading COFF Version
6380 @cindex @code{section} directive (COFF version)
6381 For COFF targets, the @code{.section} directive is used in one of the following
6385 .section @var{name}[, "@var{flags}"]
6386 .section @var{name}[, @var{subsection}]
6389 If the optional argument is quoted, it is taken as flags to use for the
6390 section. Each flag is a single character. The following flags are recognized:
6394 bss section (uninitialized data)
6396 section is not loaded
6402 exclude section from linking
6408 shared section (meaningful for PE targets)
6410 ignored. (For compatibility with the ELF version)
6412 section is not readable (meaningful for PE targets)
6414 single-digit power-of-two section alignment (GNU extension)
6417 If no flags are specified, the default flags depend upon the section name. If
6418 the section name is not recognized, the default will be for the section to be
6419 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6420 from the section, rather than adding them, so if they are used on their own it
6421 will be as if no flags had been specified at all.
6423 If the optional argument to the @code{.section} directive is not quoted, it is
6424 taken as a subsection number (@pxref{Sub-Sections}).
6429 @c only print the extra heading if both COFF and ELF are set
6430 @subheading ELF Version
6433 @cindex Section Stack
6434 This is one of the ELF section stack manipulation directives. The others are
6435 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6436 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6437 @code{.previous} (@pxref{Previous}).
6439 @cindex @code{section} directive (ELF version)
6440 For ELF targets, the @code{.section} directive is used like this:
6443 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6446 @anchor{Section Name Substitutions}
6447 @kindex --sectname-subst
6448 @cindex section name substitution
6449 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6450 argument may contain a substitution sequence. Only @code{%S} is supported
6451 at the moment, and substitutes the current section name. For example:
6454 .macro exception_code
6455 .section %S.exception
6456 [exception code here]
6471 The two @code{exception_code} invocations above would create the
6472 @code{.text.exception} and @code{.init.exception} sections respectively.
6473 This is useful e.g. to discriminate between anciliary sections that are
6474 tied to setup code to be discarded after use from anciliary sections that
6475 need to stay resident without having to define multiple @code{exception_code}
6476 macros just for that purpose.
6478 The optional @var{flags} argument is a quoted string which may contain any
6479 combination of the following characters:
6483 section is allocatable
6485 section is excluded from executable and shared library.
6489 section is executable
6491 section is mergeable
6493 section contains zero terminated strings
6495 section is a member of a section group
6497 section is used for thread-local-storage
6499 section is a member of the previously-current section's group, if any
6500 @item @code{<number>}
6501 a numeric value indicating the bits to be set in the ELF section header's flags
6502 field. Note - if one or more of the alphabetic characters described above is
6503 also included in the flags field, their bit values will be ORed into the
6505 @item @code{<target specific>}
6506 some targets extend this list with their own flag characters
6509 Note - once a section's flags have been set they cannot be changed. There are
6510 a few exceptions to this rule however. Processor and application specific
6511 flags can be added to an already defined section. The @code{.interp},
6512 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6513 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6514 section may have the executable (@code{x}) flag added.
6516 The optional @var{type} argument may contain one of the following constants:
6520 section contains data
6522 section does not contain data (i.e., section only occupies space)
6524 section contains data which is used by things other than the program
6526 section contains an array of pointers to init functions
6528 section contains an array of pointers to finish functions
6529 @item @@preinit_array
6530 section contains an array of pointers to pre-init functions
6531 @item @@@code{<number>}
6532 a numeric value to be set as the ELF section header's type field.
6533 @item @@@code{<target specific>}
6534 some targets extend this list with their own types
6537 Many targets only support the first three section types. The type may be
6538 enclosed in double quotes if necessary.
6540 Note on targets where the @code{@@} character is the start of a comment (eg
6541 ARM) then another character is used instead. For example the ARM port uses the
6544 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6545 special and have fixed types. Any attempt to declare them with a different
6546 type will generate an error from the assembler.
6548 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6549 be specified as well as an extra argument---@var{entsize}---like this:
6552 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6555 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6556 constants, each @var{entsize} octets long. Sections with both @code{M} and
6557 @code{S} must contain zero terminated strings where each character is
6558 @var{entsize} bytes long. The linker may remove duplicates within sections with
6559 the same name, same entity size and same flags. @var{entsize} must be an
6560 absolute expression. For sections with both @code{M} and @code{S}, a string
6561 which is a suffix of a larger string is considered a duplicate. Thus
6562 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6563 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6565 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6566 be present along with an additional field like this:
6569 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6572 The @var{GroupName} field specifies the name of the section group to which this
6573 particular section belongs. The optional linkage field can contain:
6577 indicates that only one copy of this section should be retained
6582 Note: if both the @var{M} and @var{G} flags are present then the fields for
6583 the Merge flag should come first, like this:
6586 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6589 If @var{flags} contains the @code{?} symbol then it may not also contain the
6590 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6591 present. Instead, @code{?} says to consider the section that's current before
6592 this directive. If that section used @code{G}, then the new section will use
6593 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6594 If not, then the @code{?} symbol has no effect.
6596 If no flags are specified, the default flags depend upon the section name. If
6597 the section name is not recognized, the default will be for the section to have
6598 none of the above flags: it will not be allocated in memory, nor writable, nor
6599 executable. The section will contain data.
6601 For ELF targets, the assembler supports another type of @code{.section}
6602 directive for compatibility with the Solaris assembler:
6605 .section "@var{name}"[, @var{flags}...]
6608 Note that the section name is quoted. There may be a sequence of comma
6613 section is allocatable
6617 section is executable
6619 section is excluded from executable and shared library.
6621 section is used for thread local storage
6624 This directive replaces the current section and subsection. See the
6625 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6626 some examples of how this directive and the other section stack directives
6632 @section @code{.set @var{symbol}, @var{expression}}
6634 @cindex @code{set} directive
6635 @cindex symbol value, setting
6636 Set the value of @var{symbol} to @var{expression}. This
6637 changes @var{symbol}'s value and type to conform to
6638 @var{expression}. If @var{symbol} was flagged as external, it remains
6639 flagged (@pxref{Symbol Attributes}).
6641 You may @code{.set} a symbol many times in the same assembly provided that the
6642 values given to the symbol are constants. Values that are based on expressions
6643 involving other symbols are allowed, but some targets may restrict this to only
6644 being done once per assembly. This is because those targets do not set the
6645 addresses of symbols at assembly time, but rather delay the assignment until a
6646 final link is performed. This allows the linker a chance to change the code in
6647 the files, changing the location of, and the relative distance between, various
6650 If you @code{.set} a global symbol, the value stored in the object
6651 file is the last value stored into it.
6654 On Z80 @code{set} is a real instruction, use
6655 @samp{@var{symbol} defl @var{expression}} instead.
6659 @section @code{.short @var{expressions}}
6661 @cindex @code{short} directive
6663 @code{.short} is normally the same as @samp{.word}.
6664 @xref{Word,,@code{.word}}.
6666 In some configurations, however, @code{.short} and @code{.word} generate
6667 numbers of different lengths. @xref{Machine Dependencies}.
6671 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6674 This expects zero or more @var{expressions}, and emits
6675 a 16 bit number for each.
6680 @section @code{.single @var{flonums}}
6682 @cindex @code{single} directive
6683 @cindex floating point numbers (single)
6684 This directive assembles zero or more flonums, separated by commas. It
6685 has the same effect as @code{.float}.
6687 The exact kind of floating point numbers emitted depends on how
6688 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6692 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6693 numbers in @sc{ieee} format.
6699 @section @code{.size}
6701 This directive is used to set the size associated with a symbol.
6705 @c only print the extra heading if both COFF and ELF are set
6706 @subheading COFF Version
6709 @cindex @code{size} directive (COFF version)
6710 For COFF targets, the @code{.size} directive is only permitted inside
6711 @code{.def}/@code{.endef} pairs. It is used like this:
6714 .size @var{expression}
6718 @samp{.size} is only meaningful when generating COFF format output; when
6719 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6726 @c only print the extra heading if both COFF and ELF are set
6727 @subheading ELF Version
6730 @cindex @code{size} directive (ELF version)
6731 For ELF targets, the @code{.size} directive is used like this:
6734 .size @var{name} , @var{expression}
6737 This directive sets the size associated with a symbol @var{name}.
6738 The size in bytes is computed from @var{expression} which can make use of label
6739 arithmetic. This directive is typically used to set the size of function
6744 @ifclear no-space-dir
6746 @section @code{.skip @var{size} , @var{fill}}
6748 @cindex @code{skip} directive
6749 @cindex filling memory
6750 This directive emits @var{size} bytes, each of value @var{fill}. Both
6751 @var{size} and @var{fill} are absolute expressions. If the comma and
6752 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6757 @section @code{.sleb128 @var{expressions}}
6759 @cindex @code{sleb128} directive
6760 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6761 compact, variable length representation of numbers used by the DWARF
6762 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6764 @ifclear no-space-dir
6766 @section @code{.space @var{size} , @var{fill}}
6768 @cindex @code{space} directive
6769 @cindex filling memory
6770 This directive emits @var{size} bytes, each of value @var{fill}. Both
6771 @var{size} and @var{fill} are absolute expressions. If the comma
6772 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6777 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6778 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6779 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6780 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6788 @section @code{.stabd, .stabn, .stabs}
6790 @cindex symbolic debuggers, information for
6791 @cindex @code{stab@var{x}} directives
6792 There are three directives that begin @samp{.stab}.
6793 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6794 The symbols are not entered in the @command{@value{AS}} hash table: they
6795 cannot be referenced elsewhere in the source file.
6796 Up to five fields are required:
6800 This is the symbol's name. It may contain any character except
6801 @samp{\000}, so is more general than ordinary symbol names. Some
6802 debuggers used to code arbitrarily complex structures into symbol names
6806 An absolute expression. The symbol's type is set to the low 8 bits of
6807 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6808 and debuggers choke on silly bit patterns.
6811 An absolute expression. The symbol's ``other'' attribute is set to the
6812 low 8 bits of this expression.
6815 An absolute expression. The symbol's descriptor is set to the low 16
6816 bits of this expression.
6819 An absolute expression which becomes the symbol's value.
6822 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6823 or @code{.stabs} statement, the symbol has probably already been created;
6824 you get a half-formed symbol in your object file. This is
6825 compatible with earlier assemblers!
6828 @cindex @code{stabd} directive
6829 @item .stabd @var{type} , @var{other} , @var{desc}
6831 The ``name'' of the symbol generated is not even an empty string.
6832 It is a null pointer, for compatibility. Older assemblers used a
6833 null pointer so they didn't waste space in object files with empty
6836 The symbol's value is set to the location counter,
6837 relocatably. When your program is linked, the value of this symbol
6838 is the address of the location counter when the @code{.stabd} was
6841 @cindex @code{stabn} directive
6842 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6843 The name of the symbol is set to the empty string @code{""}.
6845 @cindex @code{stabs} directive
6846 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6847 All five fields are specified.
6853 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6854 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6856 @cindex string, copying to object file
6857 @cindex string8, copying to object file
6858 @cindex string16, copying to object file
6859 @cindex string32, copying to object file
6860 @cindex string64, copying to object file
6861 @cindex @code{string} directive
6862 @cindex @code{string8} directive
6863 @cindex @code{string16} directive
6864 @cindex @code{string32} directive
6865 @cindex @code{string64} directive
6867 Copy the characters in @var{str} to the object file. You may specify more than
6868 one string to copy, separated by commas. Unless otherwise specified for a
6869 particular machine, the assembler marks the end of each string with a 0 byte.
6870 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6872 The variants @code{string16}, @code{string32} and @code{string64} differ from
6873 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6874 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6875 are stored in target endianness byte order.
6881 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6882 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6887 @section @code{.struct @var{expression}}
6889 @cindex @code{struct} directive
6890 Switch to the absolute section, and set the section offset to @var{expression},
6891 which must be an absolute expression. You might use this as follows:
6900 This would define the symbol @code{field1} to have the value 0, the symbol
6901 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6902 value 8. Assembly would be left in the absolute section, and you would need to
6903 use a @code{.section} directive of some sort to change to some other section
6904 before further assembly.
6908 @section @code{.subsection @var{name}}
6910 @cindex @code{subsection} directive
6911 @cindex Section Stack
6912 This is one of the ELF section stack manipulation directives. The others are
6913 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6914 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6917 This directive replaces the current subsection with @code{name}. The current
6918 section is not changed. The replaced subsection is put onto the section stack
6919 in place of the then current top of stack subsection.
6924 @section @code{.symver}
6925 @cindex @code{symver} directive
6926 @cindex symbol versioning
6927 @cindex versions of symbols
6928 Use the @code{.symver} directive to bind symbols to specific version nodes
6929 within a source file. This is only supported on ELF platforms, and is
6930 typically used when assembling files to be linked into a shared library.
6931 There are cases where it may make sense to use this in objects to be bound
6932 into an application itself so as to override a versioned symbol from a
6935 For ELF targets, the @code{.symver} directive can be used like this:
6937 .symver @var{name}, @var{name2@@nodename}
6939 If the symbol @var{name} is defined within the file
6940 being assembled, the @code{.symver} directive effectively creates a symbol
6941 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6942 just don't try and create a regular alias is that the @var{@@} character isn't
6943 permitted in symbol names. The @var{name2} part of the name is the actual name
6944 of the symbol by which it will be externally referenced. The name @var{name}
6945 itself is merely a name of convenience that is used so that it is possible to
6946 have definitions for multiple versions of a function within a single source
6947 file, and so that the compiler can unambiguously know which version of a
6948 function is being mentioned. The @var{nodename} portion of the alias should be
6949 the name of a node specified in the version script supplied to the linker when
6950 building a shared library. If you are attempting to override a versioned
6951 symbol from a shared library, then @var{nodename} should correspond to the
6952 nodename of the symbol you are trying to override.
6954 If the symbol @var{name} is not defined within the file being assembled, all
6955 references to @var{name} will be changed to @var{name2@@nodename}. If no
6956 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6959 Another usage of the @code{.symver} directive is:
6961 .symver @var{name}, @var{name2@@@@nodename}
6963 In this case, the symbol @var{name} must exist and be defined within
6964 the file being assembled. It is similar to @var{name2@@nodename}. The
6965 difference is @var{name2@@@@nodename} will also be used to resolve
6966 references to @var{name2} by the linker.
6968 The third usage of the @code{.symver} directive is:
6970 .symver @var{name}, @var{name2@@@@@@nodename}
6972 When @var{name} is not defined within the
6973 file being assembled, it is treated as @var{name2@@nodename}. When
6974 @var{name} is defined within the file being assembled, the symbol
6975 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6980 @section @code{.tag @var{structname}}
6982 @cindex COFF structure debugging
6983 @cindex structure debugging, COFF
6984 @cindex @code{tag} directive
6985 This directive is generated by compilers to include auxiliary debugging
6986 information in the symbol table. It is only permitted inside
6987 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6988 definitions in the symbol table with instances of those structures.
6991 @samp{.tag} is only used when generating COFF format output; when
6992 @command{@value{AS}} is generating @code{b.out}, it accepts this directive but
6998 @section @code{.text @var{subsection}}
7000 @cindex @code{text} directive
7001 Tells @command{@value{AS}} to assemble the following statements onto the end of
7002 the text subsection numbered @var{subsection}, which is an absolute
7003 expression. If @var{subsection} is omitted, subsection number zero
7007 @section @code{.title "@var{heading}"}
7009 @cindex @code{title} directive
7010 @cindex listing control: title line
7011 Use @var{heading} as the title (second line, immediately after the
7012 source file name and pagenumber) when generating assembly listings.
7014 This directive affects subsequent pages, as well as the current page if
7015 it appears within ten lines of the top of a page.
7019 @section @code{.type}
7021 This directive is used to set the type of a symbol.
7025 @c only print the extra heading if both COFF and ELF are set
7026 @subheading COFF Version
7029 @cindex COFF symbol type
7030 @cindex symbol type, COFF
7031 @cindex @code{type} directive (COFF version)
7032 For COFF targets, this directive is permitted only within
7033 @code{.def}/@code{.endef} pairs. It is used like this:
7039 This records the integer @var{int} as the type attribute of a symbol table
7043 @samp{.type} is associated only with COFF format output; when
7044 @command{@value{AS}} is configured for @code{b.out} output, it accepts this
7045 directive but ignores it.
7051 @c only print the extra heading if both COFF and ELF are set
7052 @subheading ELF Version
7055 @cindex ELF symbol type
7056 @cindex symbol type, ELF
7057 @cindex @code{type} directive (ELF version)
7058 For ELF targets, the @code{.type} directive is used like this:
7061 .type @var{name} , @var{type description}
7064 This sets the type of symbol @var{name} to be either a
7065 function symbol or an object symbol. There are five different syntaxes
7066 supported for the @var{type description} field, in order to provide
7067 compatibility with various other assemblers.
7069 Because some of the characters used in these syntaxes (such as @samp{@@} and
7070 @samp{#}) are comment characters for some architectures, some of the syntaxes
7071 below do not work on all architectures. The first variant will be accepted by
7072 the GNU assembler on all architectures so that variant should be used for
7073 maximum portability, if you do not need to assemble your code with other
7076 The syntaxes supported are:
7079 .type <name> STT_<TYPE_IN_UPPER_CASE>
7080 .type <name>,#<type>
7081 .type <name>,@@<type>
7082 .type <name>,%<type>
7083 .type <name>,"<type>"
7086 The types supported are:
7091 Mark the symbol as being a function name.
7094 @itemx gnu_indirect_function
7095 Mark the symbol as an indirect function when evaluated during reloc
7096 processing. (This is only supported on assemblers targeting GNU systems).
7100 Mark the symbol as being a data object.
7104 Mark the symbol as being a thead-local data object.
7108 Mark the symbol as being a common data object.
7112 Does not mark the symbol in any way. It is supported just for completeness.
7114 @item gnu_unique_object
7115 Marks the symbol as being a globally unique data object. The dynamic linker
7116 will make sure that in the entire process there is just one symbol with this
7117 name and type in use. (This is only supported on assemblers targeting GNU
7122 Note: Some targets support extra types in addition to those listed above.
7128 @section @code{.uleb128 @var{expressions}}
7130 @cindex @code{uleb128} directive
7131 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7132 compact, variable length representation of numbers used by the DWARF
7133 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7137 @section @code{.val @var{addr}}
7139 @cindex @code{val} directive
7140 @cindex COFF value attribute
7141 @cindex value attribute, COFF
7142 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7143 records the address @var{addr} as the value attribute of a symbol table
7147 @samp{.val} is used only for COFF output; when @command{@value{AS}} is
7148 configured for @code{b.out}, it accepts this directive but ignores it.
7154 @section @code{.version "@var{string}"}
7156 @cindex @code{version} directive
7157 This directive creates a @code{.note} section and places into it an ELF
7158 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7163 @section @code{.vtable_entry @var{table}, @var{offset}}
7165 @cindex @code{vtable_entry} directive
7166 This directive finds or creates a symbol @code{table} and creates a
7167 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7170 @section @code{.vtable_inherit @var{child}, @var{parent}}
7172 @cindex @code{vtable_inherit} directive
7173 This directive finds the symbol @code{child} and finds or creates the symbol
7174 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7175 parent whose addend is the value of the child symbol. As a special case the
7176 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7180 @section @code{.warning "@var{string}"}
7181 @cindex warning directive
7182 Similar to the directive @code{.error}
7183 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7186 @section @code{.weak @var{names}}
7188 @cindex @code{weak} directive
7189 This directive sets the weak attribute on the comma separated list of symbol
7190 @code{names}. If the symbols do not already exist, they will be created.
7192 On COFF targets other than PE, weak symbols are a GNU extension. This
7193 directive sets the weak attribute on the comma separated list of symbol
7194 @code{names}. If the symbols do not already exist, they will be created.
7196 On the PE target, weak symbols are supported natively as weak aliases.
7197 When a weak symbol is created that is not an alias, GAS creates an
7198 alternate symbol to hold the default value.
7201 @section @code{.weakref @var{alias}, @var{target}}
7203 @cindex @code{weakref} directive
7204 This directive creates an alias to the target symbol that enables the symbol to
7205 be referenced with weak-symbol semantics, but without actually making it weak.
7206 If direct references or definitions of the symbol are present, then the symbol
7207 will not be weak, but if all references to it are through weak references, the
7208 symbol will be marked as weak in the symbol table.
7210 The effect is equivalent to moving all references to the alias to a separate
7211 assembly source file, renaming the alias to the symbol in it, declaring the
7212 symbol as weak there, and running a reloadable link to merge the object files
7213 resulting from the assembly of the new source file and the old source file that
7214 had the references to the alias removed.
7216 The alias itself never makes to the symbol table, and is entirely handled
7217 within the assembler.
7220 @section @code{.word @var{expressions}}
7222 @cindex @code{word} directive
7223 This directive expects zero or more @var{expressions}, of any section,
7224 separated by commas.
7227 For each expression, @command{@value{AS}} emits a 32-bit number.
7230 For each expression, @command{@value{AS}} emits a 16-bit number.
7235 The size of the number emitted, and its byte order,
7236 depend on what target computer the assembly is for.
7239 @c on amd29k, i960, sparc the "special treatment to support compilers" doesn't
7240 @c happen---32-bit addressability, period; no long/short jumps.
7241 @ifset DIFF-TBL-KLUGE
7242 @cindex difference tables altered
7243 @cindex altered difference tables
7245 @emph{Warning: Special Treatment to support Compilers}
7249 Machines with a 32-bit address space, but that do less than 32-bit
7250 addressing, require the following special treatment. If the machine of
7251 interest to you does 32-bit addressing (or doesn't require it;
7252 @pxref{Machine Dependencies}), you can ignore this issue.
7255 In order to assemble compiler output into something that works,
7256 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7257 Directives of the form @samp{.word sym1-sym2} are often emitted by
7258 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7259 directive of the form @samp{.word sym1-sym2}, and the difference between
7260 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7261 creates a @dfn{secondary jump table}, immediately before the next label.
7262 This secondary jump table is preceded by a short-jump to the
7263 first byte after the secondary table. This short-jump prevents the flow
7264 of control from accidentally falling into the new table. Inside the
7265 table is a long-jump to @code{sym2}. The original @samp{.word}
7266 contains @code{sym1} minus the address of the long-jump to
7269 If there were several occurrences of @samp{.word sym1-sym2} before the
7270 secondary jump table, all of them are adjusted. If there was a
7271 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7272 long-jump to @code{sym4} is included in the secondary jump table,
7273 and the @code{.word} directives are adjusted to contain @code{sym3}
7274 minus the address of the long-jump to @code{sym4}; and so on, for as many
7275 entries in the original jump table as necessary.
7278 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7279 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7280 assembly language programmers.
7283 @c end DIFF-TBL-KLUGE
7285 @ifclear no-space-dir
7287 @section @code{.zero @var{size}}
7289 @cindex @code{zero} directive
7290 @cindex filling memory with zero bytes
7291 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7292 expression. This directive is actually an alias for the @samp{.skip} directive
7293 so in can take an optional second argument of the value to store in the bytes
7294 instead of zero. Using @samp{.zero} in this way would be confusing however.
7298 @section Deprecated Directives
7300 @cindex deprecated directives
7301 @cindex obsolescent directives
7302 One day these directives won't work.
7303 They are included for compatibility with older assemblers.
7310 @node Object Attributes
7311 @chapter Object Attributes
7312 @cindex object attributes
7314 @command{@value{AS}} assembles source files written for a specific architecture
7315 into object files for that architecture. But not all object files are alike.
7316 Many architectures support incompatible variations. For instance, floating
7317 point arguments might be passed in floating point registers if the object file
7318 requires hardware floating point support---or floating point arguments might be
7319 passed in integer registers if the object file supports processors with no
7320 hardware floating point unit. Or, if two objects are built for different
7321 generations of the same architecture, the combination may require the
7322 newer generation at run-time.
7324 This information is useful during and after linking. At link time,
7325 @command{@value{LD}} can warn about incompatible object files. After link
7326 time, tools like @command{gdb} can use it to process the linked file
7329 Compatibility information is recorded as a series of object attributes. Each
7330 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7331 string, and indicates who sets the meaning of the tag. The tag is an integer,
7332 and indicates what property the attribute describes. The value may be a string
7333 or an integer, and indicates how the property affects this object. Missing
7334 attributes are the same as attributes with a zero value or empty string value.
7336 Object attributes were developed as part of the ABI for the ARM Architecture.
7337 The file format is documented in @cite{ELF for the ARM Architecture}.
7340 * GNU Object Attributes:: @sc{gnu} Object Attributes
7341 * Defining New Object Attributes:: Defining New Object Attributes
7344 @node GNU Object Attributes
7345 @section @sc{gnu} Object Attributes
7347 The @code{.gnu_attribute} directive records an object attribute
7348 with vendor @samp{gnu}.
7350 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7351 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7352 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7353 2} is set for architecture-independent attributes and clear for
7354 architecture-dependent ones.
7356 @subsection Common @sc{gnu} attributes
7358 These attributes are valid on all architectures.
7361 @item Tag_compatibility (32)
7362 The compatibility attribute takes an integer flag value and a vendor name. If
7363 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7364 then the file is only compatible with the named toolchain. If it is greater
7365 than 1, the file can only be processed by other toolchains under some private
7366 arrangement indicated by the flag value and the vendor name.
7369 @subsection MIPS Attributes
7372 @item Tag_GNU_MIPS_ABI_FP (4)
7373 The floating-point ABI used by this object file. The value will be:
7377 0 for files not affected by the floating-point ABI.
7379 1 for files using the hardware floating-point ABI with a standard
7380 double-precision FPU.
7382 2 for files using the hardware floating-point ABI with a single-precision FPU.
7384 3 for files using the software floating-point ABI.
7386 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7387 floating-point registers, 32-bit general-purpose registers and increased the
7388 number of callee-saved floating-point registers.
7390 5 for files using the hardware floating-point ABI with a double-precision FPU
7391 with either 32-bit or 64-bit floating-point registers and 32-bit
7392 general-purpose registers.
7394 6 for files using the hardware floating-point ABI with 64-bit floating-point
7395 registers and 32-bit general-purpose registers.
7397 7 for files using the hardware floating-point ABI with 64-bit floating-point
7398 registers, 32-bit general-purpose registers and a rule that forbids the
7399 direct use of odd-numbered single-precision floating-point registers.
7403 @subsection PowerPC Attributes
7406 @item Tag_GNU_Power_ABI_FP (4)
7407 The floating-point ABI used by this object file. The value will be:
7411 0 for files not affected by the floating-point ABI.
7413 1 for files using double-precision hardware floating-point ABI.
7415 2 for files using the software floating-point ABI.
7417 3 for files using single-precision hardware floating-point ABI.
7420 @item Tag_GNU_Power_ABI_Vector (8)
7421 The vector ABI used by this object file. The value will be:
7425 0 for files not affected by the vector ABI.
7427 1 for files using general purpose registers to pass vectors.
7429 2 for files using AltiVec registers to pass vectors.
7431 3 for files using SPE registers to pass vectors.
7435 @subsection IBM z Systems Attributes
7438 @item Tag_GNU_S390_ABI_Vector (8)
7439 The vector ABI used by this object file. The value will be:
7443 0 for files not affected by the vector ABI.
7445 1 for files using software vector ABI.
7447 2 for files using hardware vector ABI.
7451 @node Defining New Object Attributes
7452 @section Defining New Object Attributes
7454 If you want to define a new @sc{gnu} object attribute, here are the places you
7455 will need to modify. New attributes should be discussed on the @samp{binutils}
7460 This manual, which is the official register of attributes.
7462 The header for your architecture @file{include/elf}, to define the tag.
7464 The @file{bfd} support file for your architecture, to merge the attribute
7465 and issue any appropriate link warnings.
7467 Test cases in @file{ld/testsuite} for merging and link warnings.
7469 @file{binutils/readelf.c} to display your attribute.
7471 GCC, if you want the compiler to mark the attribute automatically.
7477 @node Machine Dependencies
7478 @chapter Machine Dependent Features
7480 @cindex machine dependencies
7481 The machine instruction sets are (almost by definition) different on
7482 each machine where @command{@value{AS}} runs. Floating point representations
7483 vary as well, and @command{@value{AS}} often supports a few additional
7484 directives or command-line options for compatibility with other
7485 assemblers on a particular platform. Finally, some versions of
7486 @command{@value{AS}} support special pseudo-instructions for branch
7489 This chapter discusses most of these differences, though it does not
7490 include details on any machine's instruction set. For details on that
7491 subject, see the hardware manufacturer's manual.
7495 * AArch64-Dependent:: AArch64 Dependent Features
7498 * Alpha-Dependent:: Alpha Dependent Features
7501 * ARC-Dependent:: ARC Dependent Features
7504 * ARM-Dependent:: ARM Dependent Features
7507 * AVR-Dependent:: AVR Dependent Features
7510 * Blackfin-Dependent:: Blackfin Dependent Features
7513 * CR16-Dependent:: CR16 Dependent Features
7516 * CRIS-Dependent:: CRIS Dependent Features
7519 * D10V-Dependent:: D10V Dependent Features
7522 * D30V-Dependent:: D30V Dependent Features
7525 * Epiphany-Dependent:: EPIPHANY Dependent Features
7528 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7531 * HPPA-Dependent:: HPPA Dependent Features
7534 * ESA/390-Dependent:: IBM ESA/390 Dependent Features
7537 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7540 * i860-Dependent:: Intel 80860 Dependent Features
7543 * i960-Dependent:: Intel 80960 Dependent Features
7546 * IA-64-Dependent:: Intel IA-64 Dependent Features
7549 * IP2K-Dependent:: IP2K Dependent Features
7552 * LM32-Dependent:: LM32 Dependent Features
7555 * M32C-Dependent:: M32C Dependent Features
7558 * M32R-Dependent:: M32R Dependent Features
7561 * M68K-Dependent:: M680x0 Dependent Features
7564 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7567 * Meta-Dependent :: Meta Dependent Features
7570 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7573 * MIPS-Dependent:: MIPS Dependent Features
7576 * MMIX-Dependent:: MMIX Dependent Features
7579 * MSP430-Dependent:: MSP430 Dependent Features
7582 * NDS32-Dependent:: Andes NDS32 Dependent Features
7585 * NiosII-Dependent:: Altera Nios II Dependent Features
7588 * NS32K-Dependent:: NS32K Dependent Features
7591 * PDP-11-Dependent:: PDP-11 Dependent Features
7594 * PJ-Dependent:: picoJava Dependent Features
7597 * PPC-Dependent:: PowerPC Dependent Features
7600 * RL78-Dependent:: RL78 Dependent Features
7603 * RISC-V-Dependent:: RISC-V Dependent Features
7606 * RX-Dependent:: RX Dependent Features
7609 * S/390-Dependent:: IBM S/390 Dependent Features
7612 * SCORE-Dependent:: SCORE Dependent Features
7615 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7616 * SH64-Dependent:: SuperH SH64 Dependent Features
7619 * Sparc-Dependent:: SPARC Dependent Features
7622 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7625 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7628 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7631 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7634 * V850-Dependent:: V850 Dependent Features
7637 * Vax-Dependent:: VAX Dependent Features
7640 * Visium-Dependent:: Visium Dependent Features
7643 * XGATE-Dependent:: XGATE Features
7646 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7649 * Xtensa-Dependent:: Xtensa Dependent Features
7652 * Z80-Dependent:: Z80 Dependent Features
7655 * Z8000-Dependent:: Z8000 Dependent Features
7662 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7663 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7664 @c peculiarity: to preserve cross-references, there must be a node called
7665 @c "Machine Dependencies". Hence the conditional nodenames in each
7666 @c major node below. Node defaulting in makeinfo requires adjacency of
7667 @c node and sectioning commands; hence the repetition of @chapter BLAH
7668 @c in both conditional blocks.
7671 @include c-aarch64.texi
7675 @include c-alpha.texi
7691 @include c-bfin.texi
7695 @include c-cr16.texi
7699 @include c-cris.texi
7704 @node Machine Dependencies
7705 @chapter Machine Dependent Features
7707 The machine instruction sets are different on each Renesas chip family,
7708 and there are also some syntax differences among the families. This
7709 chapter describes the specific @command{@value{AS}} features for each
7713 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7714 * SH-Dependent:: Renesas SH Dependent Features
7721 @include c-d10v.texi
7725 @include c-d30v.texi
7729 @include c-epiphany.texi
7733 @include c-h8300.texi
7737 @include c-hppa.texi
7741 @include c-i370.texi
7745 @include c-i386.texi
7749 @include c-i860.texi
7753 @include c-i960.texi
7757 @include c-ia64.texi
7761 @include c-ip2k.texi
7765 @include c-lm32.texi
7769 @include c-m32c.texi
7773 @include c-m32r.texi
7777 @include c-m68k.texi
7781 @include c-m68hc11.texi
7785 @include c-metag.texi
7789 @include c-microblaze.texi
7793 @include c-mips.texi
7797 @include c-mmix.texi
7801 @include c-msp430.texi
7805 @include c-nds32.texi
7809 @include c-nios2.texi
7813 @include c-ns32k.texi
7817 @include c-pdp11.texi
7829 @include c-rl78.texi
7833 @include c-riscv.texi
7841 @include c-s390.texi
7845 @include c-score.texi
7850 @include c-sh64.texi
7854 @include c-sparc.texi
7858 @include c-tic54x.texi
7862 @include c-tic6x.texi
7866 @include c-tilegx.texi
7870 @include c-tilepro.texi
7874 @include c-v850.texi
7882 @include c-visium.texi
7886 @include c-xgate.texi
7890 @include c-xstormy16.texi
7894 @include c-xtensa.texi
7906 @c reverse effect of @down at top of generic Machine-Dep chapter
7910 @node Reporting Bugs
7911 @chapter Reporting Bugs
7912 @cindex bugs in assembler
7913 @cindex reporting bugs in assembler
7915 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7917 Reporting a bug may help you by bringing a solution to your problem, or it may
7918 not. But in any case the principal function of a bug report is to help the
7919 entire community by making the next version of @command{@value{AS}} work better.
7920 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7922 In order for a bug report to serve its purpose, you must include the
7923 information that enables us to fix the bug.
7926 * Bug Criteria:: Have you found a bug?
7927 * Bug Reporting:: How to report bugs
7931 @section Have You Found a Bug?
7932 @cindex bug criteria
7934 If you are not sure whether you have found a bug, here are some guidelines:
7937 @cindex fatal signal
7938 @cindex assembler crash
7939 @cindex crash of assembler
7941 If the assembler gets a fatal signal, for any input whatever, that is a
7942 @command{@value{AS}} bug. Reliable assemblers never crash.
7944 @cindex error on valid input
7946 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7948 @cindex invalid input
7950 If @command{@value{AS}} does not produce an error message for invalid input, that
7951 is a bug. However, you should note that your idea of ``invalid input'' might
7952 be our idea of ``an extension'' or ``support for traditional practice''.
7955 If you are an experienced user of assemblers, your suggestions for improvement
7956 of @command{@value{AS}} are welcome in any case.
7960 @section How to Report Bugs
7962 @cindex assembler bugs, reporting
7964 A number of companies and individuals offer support for @sc{gnu} products. If
7965 you obtained @command{@value{AS}} from a support organization, we recommend you
7966 contact that organization first.
7968 You can find contact information for many support companies and
7969 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7973 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7977 The fundamental principle of reporting bugs usefully is this:
7978 @strong{report all the facts}. If you are not sure whether to state a
7979 fact or leave it out, state it!
7981 Often people omit facts because they think they know what causes the problem
7982 and assume that some details do not matter. Thus, you might assume that the
7983 name of a symbol you use in an example does not matter. Well, probably it does
7984 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7985 happens to fetch from the location where that name is stored in memory;
7986 perhaps, if the name were different, the contents of that location would fool
7987 the assembler into doing the right thing despite the bug. Play it safe and
7988 give a specific, complete example. That is the easiest thing for you to do,
7989 and the most helpful.
7991 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7992 it is new to us. Therefore, always write your bug reports on the assumption
7993 that the bug has not been reported previously.
7995 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7996 bell?'' This cannot help us fix a bug, so it is basically useless. We
7997 respond by asking for enough details to enable us to investigate.
7998 You might as well expedite matters by sending them to begin with.
8000 To enable us to fix the bug, you should include all these things:
8004 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8005 it with the @samp{--version} argument.
8007 Without this, we will not know whether there is any point in looking for
8008 the bug in the current version of @command{@value{AS}}.
8011 Any patches you may have applied to the @command{@value{AS}} source.
8014 The type of machine you are using, and the operating system name and
8018 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8022 The command arguments you gave the assembler to assemble your example and
8023 observe the bug. To guarantee you will not omit something important, list them
8024 all. A copy of the Makefile (or the output from make) is sufficient.
8026 If we were to try to guess the arguments, we would probably guess wrong
8027 and then we might not encounter the bug.
8030 A complete input file that will reproduce the bug. If the bug is observed when
8031 the assembler is invoked via a compiler, send the assembler source, not the
8032 high level language source. Most compilers will produce the assembler source
8033 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8034 the options @samp{-v --save-temps}; this will save the assembler source in a
8035 file with an extension of @file{.s}, and also show you exactly how
8036 @command{@value{AS}} is being run.
8039 A description of what behavior you observe that you believe is
8040 incorrect. For example, ``It gets a fatal signal.''
8042 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8043 will certainly notice it. But if the bug is incorrect output, we might not
8044 notice unless it is glaringly wrong. You might as well not give us a chance to
8047 Even if the problem you experience is a fatal signal, you should still say so
8048 explicitly. Suppose something strange is going on, such as, your copy of
8049 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8050 library on your system. (This has happened!) Your copy might crash and ours
8051 would not. If you told us to expect a crash, then when ours fails to crash, we
8052 would know that the bug was not happening for us. If you had not told us to
8053 expect a crash, then we would not be able to draw any conclusion from our
8057 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8058 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8059 option. Always send diffs from the old file to the new file. If you even
8060 discuss something in the @command{@value{AS}} source, refer to it by context, not
8063 The line numbers in our development sources will not match those in your
8064 sources. Your line numbers would convey no useful information to us.
8067 Here are some things that are not necessary:
8071 A description of the envelope of the bug.
8073 Often people who encounter a bug spend a lot of time investigating
8074 which changes to the input file will make the bug go away and which
8075 changes will not affect it.
8077 This is often time consuming and not very useful, because the way we
8078 will find the bug is by running a single example under the debugger
8079 with breakpoints, not by pure deduction from a series of examples.
8080 We recommend that you save your time for something else.
8082 Of course, if you can find a simpler example to report @emph{instead}
8083 of the original one, that is a convenience for us. Errors in the
8084 output will be easier to spot, running under the debugger will take
8085 less time, and so on.
8087 However, simplification is not vital; if you do not want to do this,
8088 report the bug anyway and send us the entire test case you used.
8091 A patch for the bug.
8093 A patch for the bug does help us if it is a good one. But do not omit
8094 the necessary information, such as the test case, on the assumption that
8095 a patch is all we need. We might see problems with your patch and decide
8096 to fix the problem another way, or we might not understand it at all.
8098 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8099 construct an example that will make the program follow a certain path through
8100 the code. If you do not send us the example, we will not be able to construct
8101 one, so we will not be able to verify that the bug is fixed.
8103 And if we cannot understand what bug you are trying to fix, or why your
8104 patch should be an improvement, we will not install it. A test case will
8105 help us to understand.
8108 A guess about what the bug is or what it depends on.
8110 Such guesses are usually wrong. Even we cannot guess right about such
8111 things without first using the debugger to find the facts.
8114 @node Acknowledgements
8115 @chapter Acknowledgements
8117 If you have contributed to GAS and your name isn't listed here,
8118 it is not meant as a slight. We just don't know about it. Send mail to the
8119 maintainer, and we'll correct the situation. Currently
8121 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8123 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8126 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8127 information and the 68k series machines, most of the preprocessing pass, and
8128 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8130 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8131 many bug fixes, including merging support for several processors, breaking GAS
8132 up to handle multiple object file format back ends (including heavy rewrite,
8133 testing, an integration of the coff and b.out back ends), adding configuration
8134 including heavy testing and verification of cross assemblers and file splits
8135 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8136 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8137 port (including considerable amounts of reverse engineering), a SPARC opcode
8138 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8139 assertions and made them work, much other reorganization, cleanup, and lint.
8141 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8142 in format-specific I/O modules.
8144 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8145 has done much work with it since.
8147 The Intel 80386 machine description was written by Eliot Dresselhaus.
8149 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8151 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8152 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8154 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8155 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8156 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8157 support a.out format.
8159 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8160 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8161 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8162 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8165 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8166 simplified the configuration of which versions accept which directives. He
8167 updated the 68k machine description so that Motorola's opcodes always produced
8168 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8169 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8170 cross-compilation support, and one bug in relaxation that took a week and
8171 required the proverbial one-bit fix.
8173 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8174 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8175 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8176 PowerPC assembler, and made a few other minor patches.
8178 Steve Chamberlain made GAS able to generate listings.
8180 Hewlett-Packard contributed support for the HP9000/300.
8182 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8183 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8184 formats). This work was supported by both the Center for Software Science at
8185 the University of Utah and Cygnus Support.
8187 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8188 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8189 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8190 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8191 and some initial 64-bit support).
8193 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8195 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8196 support for openVMS/Alpha.
8198 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8201 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8202 Inc.@: added support for Xtensa processors.
8204 Several engineers at Cygnus Support have also provided many small bug fixes and
8205 configuration enhancements.
8207 Jon Beniston added support for the Lattice Mico32 architecture.
8209 Many others have contributed large or small bugfixes and enhancements. If
8210 you have contributed significant work and are not mentioned on this list, and
8211 want to be, let us know. Some of the history has been lost; we are not
8212 intentionally leaving anyone out.
8214 @node GNU Free Documentation License
8215 @appendix GNU Free Documentation License
8219 @unnumbered AS Index