1 \input texinfo @c -*-Texinfo-*-
2 @c Copyright (C) 1991-2018 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
50 @set abnormal-separator
54 @settitle Using @value{AS}
57 @settitle Using @value{AS} (@value{TARGET})
59 @setchapternewpage odd
64 @c WARE! Some of the machine-dependent sections contain tables of machine
65 @c instructions. Except in multi-column format, these tables look silly.
66 @c Unfortunately, Texinfo doesn't have a general-purpose multi-col format, so
67 @c the multi-col format is faked within @example sections.
69 @c Again unfortunately, the natural size that fits on a page, for these tables,
70 @c is different depending on whether or not smallbook is turned on.
71 @c This matters, because of order: text flow switches columns at each page
74 @c The format faked in this source works reasonably well for smallbook,
75 @c not well for the default large-page format. This manual expects that if you
76 @c turn on @smallbook, you will also uncomment the "@set SMALL" to enable the
77 @c tables in question. You can turn on one without the other at your
78 @c discretion, of course.
81 @c the insn tables look just as silly in info files regardless of smallbook,
82 @c might as well show 'em anyways.
86 @dircategory Software development
88 * As: (as). The GNU assembler.
89 * Gas: (as). The GNU assembler.
97 This file documents the GNU Assembler "@value{AS}".
99 @c man begin COPYRIGHT
100 Copyright @copyright{} 1991-2018 Free Software Foundation, Inc.
102 Permission is granted to copy, distribute and/or modify this document
103 under the terms of the GNU Free Documentation License, Version 1.3
104 or any later version published by the Free Software Foundation;
105 with no Invariant Sections, with no Front-Cover Texts, and with no
106 Back-Cover Texts. A copy of the license is included in the
107 section entitled ``GNU Free Documentation License''.
113 @title Using @value{AS}
114 @subtitle The @sc{gnu} Assembler
116 @subtitle for the @value{TARGET} family
118 @ifset VERSION_PACKAGE
120 @subtitle @value{VERSION_PACKAGE}
123 @subtitle Version @value{VERSION}
126 The Free Software Foundation Inc.@: thanks The Nice Computer
127 Company of Australia for loaning Dean Elsner to write the
128 first (Vax) version of @command{as} for Project @sc{gnu}.
129 The proprietors, management and staff of TNCCA thank FSF for
130 distracting the boss while they got some work
133 @author Dean Elsner, Jay Fenlason & friends
137 \hfill {\it Using {\tt @value{AS}}}\par
138 \hfill Edited by Cygnus Support\par
140 %"boxit" macro for figures:
141 %Modified from Knuth's ``boxit'' macro from TeXbook (answer to exercise 21.3)
142 \gdef\boxit#1#2{\vbox{\hrule\hbox{\vrule\kern3pt
143 \vbox{\parindent=0pt\parskip=0pt\hsize=#1\kern3pt\strut\hfil
144 #2\hfil\strut\kern3pt}\kern3pt\vrule}\hrule}}%box with visible outline
145 \gdef\ibox#1#2{\hbox to #1{#2\hfil}\kern8pt}% invisible box
148 @vskip 0pt plus 1filll
149 Copyright @copyright{} 1991-2018 Free Software Foundation, Inc.
151 Permission is granted to copy, distribute and/or modify this document
152 under the terms of the GNU Free Documentation License, Version 1.3
153 or any later version published by the Free Software Foundation;
154 with no Invariant Sections, with no Front-Cover Texts, and with no
155 Back-Cover Texts. A copy of the license is included in the
156 section entitled ``GNU Free Documentation License''.
163 @top Using @value{AS}
165 This file is a user guide to the @sc{gnu} assembler @command{@value{AS}}
166 @ifset VERSION_PACKAGE
167 @value{VERSION_PACKAGE}
169 version @value{VERSION}.
171 This version of the file describes @command{@value{AS}} configured to generate
172 code for @value{TARGET} architectures.
175 This document is distributed under the terms of the GNU Free
176 Documentation License. A copy of the license is included in the
177 section entitled ``GNU Free Documentation License''.
180 * Overview:: Overview
181 * Invoking:: Command-Line Options
183 * Sections:: Sections and Relocation
185 * Expressions:: Expressions
186 * Pseudo Ops:: Assembler Directives
188 * Object Attributes:: Object Attributes
190 * Machine Dependencies:: Machine Dependent Features
191 * Reporting Bugs:: Reporting Bugs
192 * Acknowledgements:: Who Did What
193 * GNU Free Documentation License:: GNU Free Documentation License
194 * AS Index:: AS Index
201 This manual is a user guide to the @sc{gnu} assembler @command{@value{AS}}.
203 This version of the manual describes @command{@value{AS}} configured to generate
204 code for @value{TARGET} architectures.
208 @cindex invocation summary
209 @cindex option summary
210 @cindex summary of options
211 Here is a brief summary of how to invoke @command{@value{AS}}. For details,
212 see @ref{Invoking,,Command-Line Options}.
214 @c man title AS the portable GNU assembler.
218 gcc(1), ld(1), and the Info entries for @file{binutils} and @file{ld}.
222 @c We don't use deffn and friends for the following because they seem
223 @c to be limited to one line for the header.
225 @c man begin SYNOPSIS
226 @value{AS} [@b{-a}[@b{cdghlns}][=@var{file}]] [@b{--alternate}] [@b{-D}]
227 [@b{--compress-debug-sections}] [@b{--nocompress-debug-sections}]
228 [@b{--debug-prefix-map} @var{old}=@var{new}]
229 [@b{--defsym} @var{sym}=@var{val}] [@b{-f}] [@b{-g}] [@b{--gstabs}]
230 [@b{--gstabs+}] [@b{--gdwarf-2}] [@b{--gdwarf-sections}]
231 [@b{--help}] [@b{-I} @var{dir}] [@b{-J}]
232 [@b{-K}] [@b{-L}] [@b{--listing-lhs-width}=@var{NUM}]
233 [@b{--listing-lhs-width2}=@var{NUM}] [@b{--listing-rhs-width}=@var{NUM}]
234 [@b{--listing-cont-lines}=@var{NUM}] [@b{--keep-locals}]
235 [@b{--no-pad-sections}]
236 [@b{-o} @var{objfile}] [@b{-R}]
237 [@b{--hash-size}=@var{NUM}] [@b{--reduce-memory-overheads}]
239 [@b{-v}] [@b{-version}] [@b{--version}]
240 [@b{-W}] [@b{--warn}] [@b{--fatal-warnings}] [@b{-w}] [@b{-x}]
241 [@b{-Z}] [@b{@@@var{FILE}}]
242 [@b{--sectname-subst}] [@b{--size-check=[error|warning]}]
243 [@b{--elf-stt-common=[no|yes]}]
244 [@b{--generate-missing-build-notes=[no|yes]}]
245 [@b{--target-help}] [@var{target-options}]
246 [@b{--}|@var{files} @dots{}]
249 @c Target dependent options are listed below. Keep the list sorted.
250 @c Add an empty line for separation.
254 @emph{Target AArch64 options:}
256 [@b{-mabi}=@var{ABI}]
260 @emph{Target Alpha options:}
262 [@b{-mdebug} | @b{-no-mdebug}]
263 [@b{-replace} | @b{-noreplace}]
264 [@b{-relax}] [@b{-g}] [@b{-G@var{size}}]
265 [@b{-F}] [@b{-32addr}]
269 @emph{Target ARC options:}
270 [@b{-mcpu=@var{cpu}}]
271 [@b{-mA6}|@b{-mARC600}|@b{-mARC601}|@b{-mA7}|@b{-mARC700}|@b{-mEM}|@b{-mHS}]
278 @emph{Target ARM options:}
279 @c Don't document the deprecated options
280 [@b{-mcpu}=@var{processor}[+@var{extension}@dots{}]]
281 [@b{-march}=@var{architecture}[+@var{extension}@dots{}]]
282 [@b{-mfpu}=@var{floating-point-format}]
283 [@b{-mfloat-abi}=@var{abi}]
284 [@b{-meabi}=@var{ver}]
287 [@b{-mapcs-32}|@b{-mapcs-26}|@b{-mapcs-float}|
288 @b{-mapcs-reentrant}]
289 [@b{-mthumb-interwork}] [@b{-k}]
293 @emph{Target Blackfin options:}
294 [@b{-mcpu}=@var{processor}[-@var{sirevision}]]
301 @emph{Target CRIS options:}
302 [@b{--underscore} | @b{--no-underscore}]
304 [@b{--emulation=criself} | @b{--emulation=crisaout}]
305 [@b{--march=v0_v10} | @b{--march=v10} | @b{--march=v32} | @b{--march=common_v10_v32}]
306 @c Deprecated -- deliberately not documented.
311 @emph{Target D10V options:}
316 @emph{Target D30V options:}
317 [@b{-O}|@b{-n}|@b{-N}]
321 @emph{Target EPIPHANY options:}
322 [@b{-mepiphany}|@b{-mepiphany16}]
326 @emph{Target H8/300 options:}
330 @c HPPA has no machine-dependent assembler options (yet).
334 @emph{Target i386 options:}
335 [@b{--32}|@b{--x32}|@b{--64}] [@b{-n}]
336 [@b{-march}=@var{CPU}[+@var{EXTENSION}@dots{}]] [@b{-mtune}=@var{CPU}]
340 @emph{Target IA-64 options:}
341 [@b{-mconstant-gp}|@b{-mauto-pic}]
342 [@b{-milp32}|@b{-milp64}|@b{-mlp64}|@b{-mp64}]
344 [@b{-mtune=itanium1}|@b{-mtune=itanium2}]
345 [@b{-munwind-check=warning}|@b{-munwind-check=error}]
346 [@b{-mhint.b=ok}|@b{-mhint.b=warning}|@b{-mhint.b=error}]
347 [@b{-x}|@b{-xexplicit}] [@b{-xauto}] [@b{-xdebug}]
351 @emph{Target IP2K options:}
352 [@b{-mip2022}|@b{-mip2022ext}]
356 @emph{Target M32C options:}
357 [@b{-m32c}|@b{-m16c}] [-relax] [-h-tick-hex]
361 @emph{Target M32R options:}
362 [@b{--m32rx}|@b{--[no-]warn-explicit-parallel-conflicts}|
367 @emph{Target M680X0 options:}
368 [@b{-l}] [@b{-m68000}|@b{-m68010}|@b{-m68020}|@dots{}]
372 @emph{Target M68HC11 options:}
373 [@b{-m68hc11}|@b{-m68hc12}|@b{-m68hcs12}|@b{-mm9s12x}|@b{-mm9s12xg}]
374 [@b{-mshort}|@b{-mlong}]
375 [@b{-mshort-double}|@b{-mlong-double}]
376 [@b{--force-long-branches}] [@b{--short-branches}]
377 [@b{--strict-direct-mode}] [@b{--print-insn-syntax}]
378 [@b{--print-opcodes}] [@b{--generate-example}]
382 @emph{Target MCORE options:}
383 [@b{-jsri2bsr}] [@b{-sifilter}] [@b{-relax}]
384 [@b{-mcpu=[210|340]}]
388 @emph{Target Meta options:}
389 [@b{-mcpu=@var{cpu}}] [@b{-mfpu=@var{cpu}}] [@b{-mdsp=@var{cpu}}]
392 @emph{Target MICROBLAZE options:}
393 @c MicroBlaze has no machine-dependent assembler options.
397 @emph{Target MIPS options:}
398 [@b{-nocpp}] [@b{-EL}] [@b{-EB}] [@b{-O}[@var{optimization level}]]
399 [@b{-g}[@var{debug level}]] [@b{-G} @var{num}] [@b{-KPIC}] [@b{-call_shared}]
400 [@b{-non_shared}] [@b{-xgot} [@b{-mvxworks-pic}]
401 [@b{-mabi}=@var{ABI}] [@b{-32}] [@b{-n32}] [@b{-64}] [@b{-mfp32}] [@b{-mgp32}]
402 [@b{-mfp64}] [@b{-mgp64}] [@b{-mfpxx}]
403 [@b{-modd-spreg}] [@b{-mno-odd-spreg}]
404 [@b{-march}=@var{CPU}] [@b{-mtune}=@var{CPU}] [@b{-mips1}] [@b{-mips2}]
405 [@b{-mips3}] [@b{-mips4}] [@b{-mips5}] [@b{-mips32}] [@b{-mips32r2}]
406 [@b{-mips32r3}] [@b{-mips32r5}] [@b{-mips32r6}] [@b{-mips64}] [@b{-mips64r2}]
407 [@b{-mips64r3}] [@b{-mips64r5}] [@b{-mips64r6}]
408 [@b{-construct-floats}] [@b{-no-construct-floats}]
409 [@b{-mignore-branch-isa}] [@b{-mno-ignore-branch-isa}]
410 [@b{-mnan=@var{encoding}}]
411 [@b{-trap}] [@b{-no-break}] [@b{-break}] [@b{-no-trap}]
412 [@b{-mips16}] [@b{-no-mips16}]
413 [@b{-mmips16e2}] [@b{-mno-mips16e2}]
414 [@b{-mmicromips}] [@b{-mno-micromips}]
415 [@b{-msmartmips}] [@b{-mno-smartmips}]
416 [@b{-mips3d}] [@b{-no-mips3d}]
417 [@b{-mdmx}] [@b{-no-mdmx}]
418 [@b{-mdsp}] [@b{-mno-dsp}]
419 [@b{-mdspr2}] [@b{-mno-dspr2}]
420 [@b{-mdspr3}] [@b{-mno-dspr3}]
421 [@b{-mmsa}] [@b{-mno-msa}]
422 [@b{-mxpa}] [@b{-mno-xpa}]
423 [@b{-mmt}] [@b{-mno-mt}]
424 [@b{-mmcu}] [@b{-mno-mcu}]
425 [@b{-minsn32}] [@b{-mno-insn32}]
426 [@b{-mfix7000}] [@b{-mno-fix7000}]
427 [@b{-mfix-rm7000}] [@b{-mno-fix-rm7000}]
428 [@b{-mfix-vr4120}] [@b{-mno-fix-vr4120}]
429 [@b{-mfix-vr4130}] [@b{-mno-fix-vr4130}]
430 [@b{-mdebug}] [@b{-no-mdebug}]
431 [@b{-mpdr}] [@b{-mno-pdr}]
435 @emph{Target MMIX options:}
436 [@b{--fixed-special-register-names}] [@b{--globalize-symbols}]
437 [@b{--gnu-syntax}] [@b{--relax}] [@b{--no-predefined-symbols}]
438 [@b{--no-expand}] [@b{--no-merge-gregs}] [@b{-x}]
439 [@b{--linker-allocated-gregs}]
443 @emph{Target Nios II options:}
444 [@b{-relax-all}] [@b{-relax-section}] [@b{-no-relax}]
449 @emph{Target NDS32 options:}
450 [@b{-EL}] [@b{-EB}] [@b{-O}] [@b{-Os}] [@b{-mcpu=@var{cpu}}]
451 [@b{-misa=@var{isa}}] [@b{-mabi=@var{abi}}] [@b{-mall-ext}]
452 [@b{-m[no-]16-bit}] [@b{-m[no-]perf-ext}] [@b{-m[no-]perf2-ext}]
453 [@b{-m[no-]string-ext}] [@b{-m[no-]dsp-ext}] [@b{-m[no-]mac}] [@b{-m[no-]div}]
454 [@b{-m[no-]audio-isa-ext}] [@b{-m[no-]fpu-sp-ext}] [@b{-m[no-]fpu-dp-ext}]
455 [@b{-m[no-]fpu-fma}] [@b{-mfpu-freg=@var{FREG}}] [@b{-mreduced-regs}]
456 [@b{-mfull-regs}] [@b{-m[no-]dx-regs}] [@b{-mpic}] [@b{-mno-relax}]
461 @emph{Target PDP11 options:}
462 [@b{-mpic}|@b{-mno-pic}] [@b{-mall}] [@b{-mno-extensions}]
463 [@b{-m}@var{extension}|@b{-mno-}@var{extension}]
464 [@b{-m}@var{cpu}] [@b{-m}@var{machine}]
468 @emph{Target picoJava options:}
473 @emph{Target PowerPC options:}
475 [@b{-mpwrx}|@b{-mpwr2}|@b{-mpwr}|@b{-m601}|@b{-mppc}|@b{-mppc32}|@b{-m603}|@b{-m604}|@b{-m403}|@b{-m405}|
476 @b{-m440}|@b{-m464}|@b{-m476}|@b{-m7400}|@b{-m7410}|@b{-m7450}|@b{-m7455}|@b{-m750cl}|@b{-mppc64}|
477 @b{-m620}|@b{-me500}|@b{-e500x2}|@b{-me500mc}|@b{-me500mc64}|@b{-me5500}|@b{-me6500}|@b{-mppc64bridge}|
478 @b{-mbooke}|@b{-mpower4}|@b{-mpwr4}|@b{-mpower5}|@b{-mpwr5}|@b{-mpwr5x}|@b{-mpower6}|@b{-mpwr6}|
479 @b{-mpower7}|@b{-mpwr7}|@b{-mpower8}|@b{-mpwr8}|@b{-mpower9}|@b{-mpwr9}@b{-ma2}|
480 @b{-mcell}|@b{-mspe}|@b{-mspe2}|@b{-mtitan}|@b{-me300}|@b{-mcom}]
481 [@b{-many}] [@b{-maltivec}|@b{-mvsx}|@b{-mhtm}|@b{-mvle}]
482 [@b{-mregnames}|@b{-mno-regnames}]
483 [@b{-mrelocatable}|@b{-mrelocatable-lib}|@b{-K PIC}] [@b{-memb}]
484 [@b{-mlittle}|@b{-mlittle-endian}|@b{-le}|@b{-mbig}|@b{-mbig-endian}|@b{-be}]
485 [@b{-msolaris}|@b{-mno-solaris}]
486 [@b{-nops=@var{count}}]
490 @emph{Target PRU options:}
493 [@b{-mno-warn-regname-label}]
497 @emph{Target RISC-V options:}
498 [@b{-fpic}|@b{-fPIC}|@b{-fno-pic}]
499 [@b{-march}=@var{ISA}]
500 [@b{-mabi}=@var{ABI}]
504 @emph{Target RL78 options:}
506 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
510 @emph{Target RX options:}
511 [@b{-mlittle-endian}|@b{-mbig-endian}]
512 [@b{-m32bit-doubles}|@b{-m64bit-doubles}]
513 [@b{-muse-conventional-section-names}]
514 [@b{-msmall-data-limit}]
517 [@b{-mint-register=@var{number}}]
518 [@b{-mgcc-abi}|@b{-mrx-abi}]
522 @emph{Target s390 options:}
523 [@b{-m31}|@b{-m64}] [@b{-mesa}|@b{-mzarch}] [@b{-march}=@var{CPU}]
524 [@b{-mregnames}|@b{-mno-regnames}]
525 [@b{-mwarn-areg-zero}]
529 @emph{Target SCORE options:}
530 [@b{-EB}][@b{-EL}][@b{-FIXDD}][@b{-NWARN}]
531 [@b{-SCORE5}][@b{-SCORE5U}][@b{-SCORE7}][@b{-SCORE3}]
532 [@b{-march=score7}][@b{-march=score3}]
533 [@b{-USE_R1}][@b{-KPIC}][@b{-O0}][@b{-G} @var{num}][@b{-V}]
537 @emph{Target SPARC options:}
538 @c The order here is important. See c-sparc.texi.
539 [@b{-Av6}|@b{-Av7}|@b{-Av8}|@b{-Aleon}|@b{-Asparclet}|@b{-Asparclite}
540 @b{-Av8plus}|@b{-Av8plusa}|@b{-Av8plusb}|@b{-Av8plusc}|@b{-Av8plusd}
541 @b{-Av8plusv}|@b{-Av8plusm}|@b{-Av9}|@b{-Av9a}|@b{-Av9b}|@b{-Av9c}
542 @b{-Av9d}|@b{-Av9e}|@b{-Av9v}|@b{-Av9m}|@b{-Asparc}|@b{-Asparcvis}
543 @b{-Asparcvis2}|@b{-Asparcfmaf}|@b{-Asparcima}|@b{-Asparcvis3}
544 @b{-Asparcvisr}|@b{-Asparc5}]
545 [@b{-xarch=v8plus}|@b{-xarch=v8plusa}]|@b{-xarch=v8plusb}|@b{-xarch=v8plusc}
546 @b{-xarch=v8plusd}|@b{-xarch=v8plusv}|@b{-xarch=v8plusm}|@b{-xarch=v9}
547 @b{-xarch=v9a}|@b{-xarch=v9b}|@b{-xarch=v9c}|@b{-xarch=v9d}|@b{-xarch=v9e}
548 @b{-xarch=v9v}|@b{-xarch=v9m}|@b{-xarch=sparc}|@b{-xarch=sparcvis}
549 @b{-xarch=sparcvis2}|@b{-xarch=sparcfmaf}|@b{-xarch=sparcima}
550 @b{-xarch=sparcvis3}|@b{-xarch=sparcvisr}|@b{-xarch=sparc5}
553 [@b{--enforce-aligned-data}][@b{--dcti-couples-detect}]
557 @emph{Target TIC54X options:}
558 [@b{-mcpu=54[123589]}|@b{-mcpu=54[56]lp}] [@b{-mfar-mode}|@b{-mf}]
559 [@b{-merrors-to-file} @var{<filename>}|@b{-me} @var{<filename>}]
563 @emph{Target TIC6X options:}
564 [@b{-march=@var{arch}}] [@b{-mbig-endian}|@b{-mlittle-endian}]
565 [@b{-mdsbt}|@b{-mno-dsbt}] [@b{-mpid=no}|@b{-mpid=near}|@b{-mpid=far}]
566 [@b{-mpic}|@b{-mno-pic}]
570 @emph{Target TILE-Gx options:}
571 [@b{-m32}|@b{-m64}][@b{-EB}][@b{-EL}]
574 @c TILEPro has no machine-dependent assembler options
578 @emph{Target Visium options:}
579 [@b{-mtune=@var{arch}}]
583 @emph{Target Xtensa options:}
584 [@b{--[no-]text-section-literals}] [@b{--[no-]auto-litpools}]
585 [@b{--[no-]absolute-literals}]
586 [@b{--[no-]target-align}] [@b{--[no-]longcalls}]
587 [@b{--[no-]transform}]
588 [@b{--rename-section} @var{oldname}=@var{newname}]
589 [@b{--[no-]trampolines}]
593 @emph{Target Z80 options:}
594 [@b{-z80}] [@b{-r800}]
595 [@b{ -ignore-undocumented-instructions}] [@b{-Wnud}]
596 [@b{ -ignore-unportable-instructions}] [@b{-Wnup}]
597 [@b{ -warn-undocumented-instructions}] [@b{-Wud}]
598 [@b{ -warn-unportable-instructions}] [@b{-Wup}]
599 [@b{ -forbid-undocumented-instructions}] [@b{-Fud}]
600 [@b{ -forbid-unportable-instructions}] [@b{-Fup}]
604 @c Z8000 has no machine-dependent assembler options
613 @include at-file.texi
616 Turn on listings, in any of a variety of ways:
620 omit false conditionals
623 omit debugging directives
626 include general information, like @value{AS} version and options passed
629 include high-level source
635 include macro expansions
638 omit forms processing
644 set the name of the listing file
647 You may combine these options; for example, use @samp{-aln} for assembly
648 listing without forms processing. The @samp{=file} option, if used, must be
649 the last one. By itself, @samp{-a} defaults to @samp{-ahls}.
652 Begin in alternate macro mode.
654 @xref{Altmacro,,@code{.altmacro}}.
657 @item --compress-debug-sections
658 Compress DWARF debug sections using zlib with SHF_COMPRESSED from the
659 ELF ABI. The resulting object file may not be compatible with older
660 linkers and object file utilities. Note if compression would make a
661 given section @emph{larger} then it is not compressed.
664 @cindex @samp{--compress-debug-sections=} option
665 @item --compress-debug-sections=none
666 @itemx --compress-debug-sections=zlib
667 @itemx --compress-debug-sections=zlib-gnu
668 @itemx --compress-debug-sections=zlib-gabi
669 These options control how DWARF debug sections are compressed.
670 @option{--compress-debug-sections=none} is equivalent to
671 @option{--nocompress-debug-sections}.
672 @option{--compress-debug-sections=zlib} and
673 @option{--compress-debug-sections=zlib-gabi} are equivalent to
674 @option{--compress-debug-sections}.
675 @option{--compress-debug-sections=zlib-gnu} compresses DWARF debug
676 sections using zlib. The debug sections are renamed to begin with
677 @samp{.zdebug}. Note if compression would make a given section
678 @emph{larger} then it is not compressed nor renamed.
682 @item --nocompress-debug-sections
683 Do not compress DWARF debug sections. This is usually the default for all
684 targets except the x86/x86_64, but a configure time option can be used to
688 Ignored. This option is accepted for script compatibility with calls to
691 @item --debug-prefix-map @var{old}=@var{new}
692 When assembling files in directory @file{@var{old}}, record debugging
693 information describing them as in @file{@var{new}} instead.
695 @item --defsym @var{sym}=@var{value}
696 Define the symbol @var{sym} to be @var{value} before assembling the input file.
697 @var{value} must be an integer constant. As in C, a leading @samp{0x}
698 indicates a hexadecimal value, and a leading @samp{0} indicates an octal
699 value. The value of the symbol can be overridden inside a source file via the
700 use of a @code{.set} pseudo-op.
703 ``fast''---skip whitespace and comment preprocessing (assume source is
708 Generate debugging information for each assembler source line using whichever
709 debug format is preferred by the target. This currently means either STABS,
713 Generate stabs debugging information for each assembler line. This
714 may help debugging assembler code, if the debugger can handle it.
717 Generate stabs debugging information for each assembler line, with GNU
718 extensions that probably only gdb can handle, and that could make other
719 debuggers crash or refuse to read your program. This
720 may help debugging assembler code. Currently the only GNU extension is
721 the location of the current working directory at assembling time.
724 Generate DWARF2 debugging information for each assembler line. This
725 may help debugging assembler code, if the debugger can handle it. Note---this
726 option is only supported by some targets, not all of them.
728 @item --gdwarf-sections
729 Instead of creating a .debug_line section, create a series of
730 .debug_line.@var{foo} sections where @var{foo} is the name of the
731 corresponding code section. For example a code section called @var{.text.func}
732 will have its dwarf line number information placed into a section called
733 @var{.debug_line.text.func}. If the code section is just called @var{.text}
734 then debug line section will still be called just @var{.debug_line} without any
738 @item --size-check=error
739 @itemx --size-check=warning
740 Issue an error or warning for invalid ELF .size directive.
742 @item --elf-stt-common=no
743 @itemx --elf-stt-common=yes
744 These options control whether the ELF assembler should generate common
745 symbols with the @code{STT_COMMON} type. The default can be controlled
746 by a configure option @option{--enable-elf-stt-common}.
748 @item --generate-missing-build-notes=yes
749 @itemx --generate-missing-build-notes=no
750 These options control whether the ELF assembler should generate GNU Build
751 attribute notes if none are present in the input sources.
752 The default can be controlled by the @option{--enable-generate-build-notes}
758 Print a summary of the command line options and exit.
761 Print a summary of all target specific options and exit.
764 Add directory @var{dir} to the search list for @code{.include} directives.
767 Don't warn about signed overflow.
770 @ifclear DIFF-TBL-KLUGE
771 This option is accepted but has no effect on the @value{TARGET} family.
773 @ifset DIFF-TBL-KLUGE
774 Issue warnings when difference tables altered for long displacements.
779 Keep (in the symbol table) local symbols. These symbols start with
780 system-specific local label prefixes, typically @samp{.L} for ELF systems
781 or @samp{L} for traditional a.out systems.
786 @item --listing-lhs-width=@var{number}
787 Set the maximum width, in words, of the output data column for an assembler
788 listing to @var{number}.
790 @item --listing-lhs-width2=@var{number}
791 Set the maximum width, in words, of the output data column for continuation
792 lines in an assembler listing to @var{number}.
794 @item --listing-rhs-width=@var{number}
795 Set the maximum width of an input source line, as displayed in a listing, to
798 @item --listing-cont-lines=@var{number}
799 Set the maximum number of lines printed in a listing for a single line of input
802 @item --no-pad-sections
803 Stop the assembler for padding the ends of output sections to the alignment
804 of that section. The default is to pad the sections, but this can waste space
805 which might be needed on targets which have tight memory constraints.
807 @item -o @var{objfile}
808 Name the object-file output from @command{@value{AS}} @var{objfile}.
811 Fold the data section into the text section.
813 @item --hash-size=@var{number}
814 Set the default size of GAS's hash tables to a prime number close to
815 @var{number}. Increasing this value can reduce the length of time it takes the
816 assembler to perform its tasks, at the expense of increasing the assembler's
817 memory requirements. Similarly reducing this value can reduce the memory
818 requirements at the expense of speed.
820 @item --reduce-memory-overheads
821 This option reduces GAS's memory requirements, at the expense of making the
822 assembly processes slower. Currently this switch is a synonym for
823 @samp{--hash-size=4051}, but in the future it may have other effects as well.
826 @item --sectname-subst
827 Honor substitution sequences in section names.
829 @xref{Section Name Substitutions,,@code{.section @var{name}}}.
834 Print the maximum space (in bytes) and total time (in seconds) used by
837 @item --strip-local-absolute
838 Remove local absolute symbols from the outgoing symbol table.
842 Print the @command{as} version.
845 Print the @command{as} version and exit.
849 Suppress warning messages.
851 @item --fatal-warnings
852 Treat warnings as errors.
855 Don't suppress warning messages or treat them as errors.
864 Generate an object file even after errors.
866 @item -- | @var{files} @dots{}
867 Standard input, or source files to assemble.
875 @xref{AArch64 Options}, for the options available when @value{AS} is configured
876 for the 64-bit mode of the ARM Architecture (AArch64).
881 The following options are available when @value{AS} is configured for the
882 64-bit mode of the ARM Architecture (AArch64).
885 @include c-aarch64.texi
886 @c ended inside the included file
894 @xref{Alpha Options}, for the options available when @value{AS} is configured
895 for an Alpha processor.
900 The following options are available when @value{AS} is configured for an Alpha
904 @include c-alpha.texi
905 @c ended inside the included file
912 The following options are available when @value{AS} is configured for an ARC
916 @item -mcpu=@var{cpu}
917 This option selects the core processor variant.
919 Select either big-endian (-EB) or little-endian (-EL) output.
921 Enable Code Density extenssion instructions.
926 The following options are available when @value{AS} is configured for the ARM
930 @item -mcpu=@var{processor}[+@var{extension}@dots{}]
931 Specify which ARM processor variant is the target.
932 @item -march=@var{architecture}[+@var{extension}@dots{}]
933 Specify which ARM architecture variant is used by the target.
934 @item -mfpu=@var{floating-point-format}
935 Select which Floating Point architecture is the target.
936 @item -mfloat-abi=@var{abi}
937 Select which floating point ABI is in use.
939 Enable Thumb only instruction decoding.
940 @item -mapcs-32 | -mapcs-26 | -mapcs-float | -mapcs-reentrant
941 Select which procedure calling convention is in use.
943 Select either big-endian (-EB) or little-endian (-EL) output.
944 @item -mthumb-interwork
945 Specify that the code has been generated with interworking between Thumb and
948 Turns on CodeComposer Studio assembly syntax compatibility mode.
950 Specify that PIC code has been generated.
958 @xref{Blackfin Options}, for the options available when @value{AS} is
959 configured for the Blackfin processor family.
964 The following options are available when @value{AS} is configured for
965 the Blackfin processor family.
969 @c ended inside the included file
976 See the info pages for documentation of the CRIS-specific options.
980 The following options are available when @value{AS} is configured for
983 @cindex D10V optimization
984 @cindex optimization, D10V
986 Optimize output by parallelizing instructions.
991 The following options are available when @value{AS} is configured for a D30V
994 @cindex D30V optimization
995 @cindex optimization, D30V
997 Optimize output by parallelizing instructions.
1001 Warn when nops are generated.
1003 @cindex D30V nops after 32-bit multiply
1005 Warn when a nop after a 32-bit multiply instruction is generated.
1011 The following options are available when @value{AS} is configured for the
1012 Adapteva EPIPHANY series.
1015 @xref{Epiphany Options}, for the options available when @value{AS} is
1016 configured for an Epiphany processor.
1020 @c man begin OPTIONS
1021 The following options are available when @value{AS} is configured for
1022 an Epiphany processor.
1024 @c man begin INCLUDE
1025 @include c-epiphany.texi
1026 @c ended inside the included file
1034 @xref{H8/300 Options}, for the options available when @value{AS} is configured
1035 for an H8/300 processor.
1039 @c man begin OPTIONS
1040 The following options are available when @value{AS} is configured for an H8/300
1043 @c man begin INCLUDE
1044 @include c-h8300.texi
1045 @c ended inside the included file
1053 @xref{i386-Options}, for the options available when @value{AS} is
1054 configured for an i386 processor.
1058 @c man begin OPTIONS
1059 The following options are available when @value{AS} is configured for
1062 @c man begin INCLUDE
1063 @include c-i386.texi
1064 @c ended inside the included file
1069 @c man begin OPTIONS
1071 The following options are available when @value{AS} is configured for the
1077 Specifies that the extended IP2022 instructions are allowed.
1080 Restores the default behaviour, which restricts the permitted instructions to
1081 just the basic IP2022 ones.
1087 The following options are available when @value{AS} is configured for the
1088 Renesas M32C and M16C processors.
1093 Assemble M32C instructions.
1096 Assemble M16C instructions (the default).
1099 Enable support for link-time relaxations.
1102 Support H'00 style hex constants in addition to 0x00 style.
1108 The following options are available when @value{AS} is configured for the
1109 Renesas M32R (formerly Mitsubishi M32R) series.
1114 Specify which processor in the M32R family is the target. The default
1115 is normally the M32R, but this option changes it to the M32RX.
1117 @item --warn-explicit-parallel-conflicts or --Wp
1118 Produce warning messages when questionable parallel constructs are
1121 @item --no-warn-explicit-parallel-conflicts or --Wnp
1122 Do not produce warning messages when questionable parallel constructs are
1129 The following options are available when @value{AS} is configured for the
1130 Motorola 68000 series.
1135 Shorten references to undefined symbols, to one word instead of two.
1137 @item -m68000 | -m68008 | -m68010 | -m68020 | -m68030
1138 @itemx | -m68040 | -m68060 | -m68302 | -m68331 | -m68332
1139 @itemx | -m68333 | -m68340 | -mcpu32 | -m5200
1140 Specify what processor in the 68000 family is the target. The default
1141 is normally the 68020, but this can be changed at configuration time.
1143 @item -m68881 | -m68882 | -mno-68881 | -mno-68882
1144 The target machine does (or does not) have a floating-point coprocessor.
1145 The default is to assume a coprocessor for 68020, 68030, and cpu32. Although
1146 the basic 68000 is not compatible with the 68881, a combination of the
1147 two can be specified, since it's possible to do emulation of the
1148 coprocessor instructions with the main processor.
1150 @item -m68851 | -mno-68851
1151 The target machine does (or does not) have a memory-management
1152 unit coprocessor. The default is to assume an MMU for 68020 and up.
1160 @xref{Nios II Options}, for the options available when @value{AS} is configured
1161 for an Altera Nios II processor.
1165 @c man begin OPTIONS
1166 The following options are available when @value{AS} is configured for an
1167 Altera Nios II processor.
1169 @c man begin INCLUDE
1170 @include c-nios2.texi
1171 @c ended inside the included file
1177 For details about the PDP-11 machine dependent features options,
1178 see @ref{PDP-11-Options}.
1181 @item -mpic | -mno-pic
1182 Generate position-independent (or position-dependent) code. The
1183 default is @option{-mpic}.
1186 @itemx -mall-extensions
1187 Enable all instruction set extensions. This is the default.
1189 @item -mno-extensions
1190 Disable all instruction set extensions.
1192 @item -m@var{extension} | -mno-@var{extension}
1193 Enable (or disable) a particular instruction set extension.
1196 Enable the instruction set extensions supported by a particular CPU, and
1197 disable all other extensions.
1199 @item -m@var{machine}
1200 Enable the instruction set extensions supported by a particular machine
1201 model, and disable all other extensions.
1207 The following options are available when @value{AS} is configured for
1208 a picoJava processor.
1212 @cindex PJ endianness
1213 @cindex endianness, PJ
1214 @cindex big endian output, PJ
1216 Generate ``big endian'' format output.
1218 @cindex little endian output, PJ
1220 Generate ``little endian'' format output.
1228 @xref{PRU Options}, for the options available when @value{AS} is configured
1229 for a PRU processor.
1233 @c man begin OPTIONS
1234 The following options are available when @value{AS} is configured for a
1237 @c man begin INCLUDE
1239 @c ended inside the included file
1244 The following options are available when @value{AS} is configured for the
1245 Motorola 68HC11 or 68HC12 series.
1249 @item -m68hc11 | -m68hc12 | -m68hcs12 | -mm9s12x | -mm9s12xg
1250 Specify what processor is the target. The default is
1251 defined by the configuration option when building the assembler.
1253 @item --xgate-ramoffset
1254 Instruct the linker to offset RAM addresses from S12X address space into
1255 XGATE address space.
1258 Specify to use the 16-bit integer ABI.
1261 Specify to use the 32-bit integer ABI.
1263 @item -mshort-double
1264 Specify to use the 32-bit double ABI.
1267 Specify to use the 64-bit double ABI.
1269 @item --force-long-branches
1270 Relative branches are turned into absolute ones. This concerns
1271 conditional branches, unconditional branches and branches to a
1274 @item -S | --short-branches
1275 Do not turn relative branches into absolute ones
1276 when the offset is out of range.
1278 @item --strict-direct-mode
1279 Do not turn the direct addressing mode into extended addressing mode
1280 when the instruction does not support direct addressing mode.
1282 @item --print-insn-syntax
1283 Print the syntax of instruction in case of error.
1285 @item --print-opcodes
1286 Print the list of instructions with syntax and then exit.
1288 @item --generate-example
1289 Print an example of instruction for each possible instruction and then exit.
1290 This option is only useful for testing @command{@value{AS}}.
1296 The following options are available when @command{@value{AS}} is configured
1297 for the SPARC architecture:
1300 @item -Av6 | -Av7 | -Av8 | -Asparclet | -Asparclite
1301 @itemx -Av8plus | -Av8plusa | -Av9 | -Av9a
1302 Explicitly select a variant of the SPARC architecture.
1304 @samp{-Av8plus} and @samp{-Av8plusa} select a 32 bit environment.
1305 @samp{-Av9} and @samp{-Av9a} select a 64 bit environment.
1307 @samp{-Av8plusa} and @samp{-Av9a} enable the SPARC V9 instruction set with
1308 UltraSPARC extensions.
1310 @item -xarch=v8plus | -xarch=v8plusa
1311 For compatibility with the Solaris v9 assembler. These options are
1312 equivalent to -Av8plus and -Av8plusa, respectively.
1315 Warn when the assembler switches to another architecture.
1320 The following options are available when @value{AS} is configured for the 'c54x
1325 Enable extended addressing mode. All addresses and relocations will assume
1326 extended addressing (usually 23 bits).
1327 @item -mcpu=@var{CPU_VERSION}
1328 Sets the CPU version being compiled for.
1329 @item -merrors-to-file @var{FILENAME}
1330 Redirect error output to a file, for broken systems which don't support such
1331 behaviour in the shell.
1336 @c man begin OPTIONS
1337 The following options are available when @value{AS} is configured for
1342 This option sets the largest size of an object that can be referenced
1343 implicitly with the @code{gp} register. It is only accepted for targets that
1344 use ECOFF format, such as a DECstation running Ultrix. The default value is 8.
1346 @cindex MIPS endianness
1347 @cindex endianness, MIPS
1348 @cindex big endian output, MIPS
1350 Generate ``big endian'' format output.
1352 @cindex little endian output, MIPS
1354 Generate ``little endian'' format output.
1372 Generate code for a particular MIPS Instruction Set Architecture level.
1373 @samp{-mips1} is an alias for @samp{-march=r3000}, @samp{-mips2} is an
1374 alias for @samp{-march=r6000}, @samp{-mips3} is an alias for
1375 @samp{-march=r4000} and @samp{-mips4} is an alias for @samp{-march=r8000}.
1376 @samp{-mips5}, @samp{-mips32}, @samp{-mips32r2}, @samp{-mips32r3},
1377 @samp{-mips32r5}, @samp{-mips32r6}, @samp{-mips64}, @samp{-mips64r2},
1378 @samp{-mips64r3}, @samp{-mips64r5}, and @samp{-mips64r6} correspond to generic
1379 MIPS V, MIPS32, MIPS32 Release 2, MIPS32 Release 3, MIPS32 Release 5, MIPS32
1380 Release 6, MIPS64, MIPS64 Release 2, MIPS64 Release 3, MIPS64 Release 5, and
1381 MIPS64 Release 6 ISA processors, respectively.
1383 @item -march=@var{cpu}
1384 Generate code for a particular MIPS CPU.
1386 @item -mtune=@var{cpu}
1387 Schedule and tune for a particular MIPS CPU.
1391 Cause nops to be inserted if the read of the destination register
1392 of an mfhi or mflo instruction occurs in the following two instructions.
1395 @itemx -mno-fix-rm7000
1396 Cause nops to be inserted if a dmult or dmultu instruction is
1397 followed by a load instruction.
1401 Cause stabs-style debugging output to go into an ECOFF-style .mdebug
1402 section instead of the standard ELF .stabs sections.
1406 Control generation of @code{.pdr} sections.
1410 The register sizes are normally inferred from the ISA and ABI, but these
1411 flags force a certain group of registers to be treated as 32 bits wide at
1412 all times. @samp{-mgp32} controls the size of general-purpose registers
1413 and @samp{-mfp32} controls the size of floating-point registers.
1417 The register sizes are normally inferred from the ISA and ABI, but these
1418 flags force a certain group of registers to be treated as 64 bits wide at
1419 all times. @samp{-mgp64} controls the size of general-purpose registers
1420 and @samp{-mfp64} controls the size of floating-point registers.
1423 The register sizes are normally inferred from the ISA and ABI, but using
1424 this flag in combination with @samp{-mabi=32} enables an ABI variant
1425 which will operate correctly with floating-point registers which are
1429 @itemx -mno-odd-spreg
1430 Enable use of floating-point operations on odd-numbered single-precision
1431 registers when supported by the ISA. @samp{-mfpxx} implies
1432 @samp{-mno-odd-spreg}, otherwise the default is @samp{-modd-spreg}.
1436 Generate code for the MIPS 16 processor. This is equivalent to putting
1437 @code{.module mips16} at the start of the assembly file. @samp{-no-mips16}
1438 turns off this option.
1441 @itemx -mno-mips16e2
1442 Enable the use of MIPS16e2 instructions in MIPS16 mode. This is equivalent
1443 to putting @code{.module mips16e2} at the start of the assembly file.
1444 @samp{-mno-mips16e2} turns off this option.
1447 @itemx -mno-micromips
1448 Generate code for the microMIPS processor. This is equivalent to putting
1449 @code{.module micromips} at the start of the assembly file.
1450 @samp{-mno-micromips} turns off this option. This is equivalent to putting
1451 @code{.module nomicromips} at the start of the assembly file.
1454 @itemx -mno-smartmips
1455 Enables the SmartMIPS extension to the MIPS32 instruction set. This is
1456 equivalent to putting @code{.module smartmips} at the start of the assembly
1457 file. @samp{-mno-smartmips} turns off this option.
1461 Generate code for the MIPS-3D Application Specific Extension.
1462 This tells the assembler to accept MIPS-3D instructions.
1463 @samp{-no-mips3d} turns off this option.
1467 Generate code for the MDMX Application Specific Extension.
1468 This tells the assembler to accept MDMX instructions.
1469 @samp{-no-mdmx} turns off this option.
1473 Generate code for the DSP Release 1 Application Specific Extension.
1474 This tells the assembler to accept DSP Release 1 instructions.
1475 @samp{-mno-dsp} turns off this option.
1479 Generate code for the DSP Release 2 Application Specific Extension.
1480 This option implies @samp{-mdsp}.
1481 This tells the assembler to accept DSP Release 2 instructions.
1482 @samp{-mno-dspr2} turns off this option.
1486 Generate code for the DSP Release 3 Application Specific Extension.
1487 This option implies @samp{-mdsp} and @samp{-mdspr2}.
1488 This tells the assembler to accept DSP Release 3 instructions.
1489 @samp{-mno-dspr3} turns off this option.
1493 Generate code for the MIPS SIMD Architecture Extension.
1494 This tells the assembler to accept MSA instructions.
1495 @samp{-mno-msa} turns off this option.
1499 Generate code for the MIPS eXtended Physical Address (XPA) Extension.
1500 This tells the assembler to accept XPA instructions.
1501 @samp{-mno-xpa} turns off this option.
1505 Generate code for the MT Application Specific Extension.
1506 This tells the assembler to accept MT instructions.
1507 @samp{-mno-mt} turns off this option.
1511 Generate code for the MCU Application Specific Extension.
1512 This tells the assembler to accept MCU instructions.
1513 @samp{-mno-mcu} turns off this option.
1517 Only use 32-bit instruction encodings when generating code for the
1518 microMIPS processor. This option inhibits the use of any 16-bit
1519 instructions. This is equivalent to putting @code{.set insn32} at
1520 the start of the assembly file. @samp{-mno-insn32} turns off this
1521 option. This is equivalent to putting @code{.set noinsn32} at the
1522 start of the assembly file. By default @samp{-mno-insn32} is
1523 selected, allowing all instructions to be used.
1525 @item --construct-floats
1526 @itemx --no-construct-floats
1527 The @samp{--no-construct-floats} option disables the construction of
1528 double width floating point constants by loading the two halves of the
1529 value into the two single width floating point registers that make up
1530 the double width register. By default @samp{--construct-floats} is
1531 selected, allowing construction of these floating point constants.
1533 @item --relax-branch
1534 @itemx --no-relax-branch
1535 The @samp{--relax-branch} option enables the relaxation of out-of-range
1536 branches. By default @samp{--no-relax-branch} is selected, causing any
1537 out-of-range branches to produce an error.
1539 @item -mignore-branch-isa
1540 @itemx -mno-ignore-branch-isa
1541 Ignore branch checks for invalid transitions between ISA modes. The
1542 semantics of branches does not provide for an ISA mode switch, so in
1543 most cases the ISA mode a branch has been encoded for has to be the
1544 same as the ISA mode of the branch's target label. Therefore GAS has
1545 checks implemented that verify in branch assembly that the two ISA
1546 modes match. @samp{-mignore-branch-isa} disables these checks. By
1547 default @samp{-mno-ignore-branch-isa} is selected, causing any invalid
1548 branch requiring a transition between ISA modes to produce an error.
1550 @item -mnan=@var{encoding}
1551 Select between the IEEE 754-2008 (@option{-mnan=2008}) or the legacy
1552 (@option{-mnan=legacy}) NaN encoding format. The latter is the default.
1555 @item --emulation=@var{name}
1556 This option was formerly used to switch between ELF and ECOFF output
1557 on targets like IRIX 5 that supported both. MIPS ECOFF support was
1558 removed in GAS 2.24, so the option now serves little purpose.
1559 It is retained for backwards compatibility.
1561 The available configuration names are: @samp{mipself}, @samp{mipslelf} and
1562 @samp{mipsbelf}. Choosing @samp{mipself} now has no effect, since the output
1563 is always ELF. @samp{mipslelf} and @samp{mipsbelf} select little- and
1564 big-endian output respectively, but @samp{-EL} and @samp{-EB} are now the
1565 preferred options instead.
1568 @command{@value{AS}} ignores this option. It is accepted for compatibility with
1575 Control how to deal with multiplication overflow and division by zero.
1576 @samp{--trap} or @samp{--no-break} (which are synonyms) take a trap exception
1577 (and only work for Instruction Set Architecture level 2 and higher);
1578 @samp{--break} or @samp{--no-trap} (also synonyms, and the default) take a
1582 When this option is used, @command{@value{AS}} will issue a warning every
1583 time it generates a nop instruction from a macro.
1589 The following options are available when @value{AS} is configured for
1595 Enable or disable the JSRI to BSR transformation. By default this is enabled.
1596 The command line option @samp{-nojsri2bsr} can be used to disable it.
1600 Enable or disable the silicon filter behaviour. By default this is disabled.
1601 The default can be overridden by the @samp{-sifilter} command line option.
1604 Alter jump instructions for long displacements.
1606 @item -mcpu=[210|340]
1607 Select the cpu type on the target hardware. This controls which instructions
1611 Assemble for a big endian target.
1614 Assemble for a little endian target.
1623 @xref{Meta Options}, for the options available when @value{AS} is configured
1624 for a Meta processor.
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-metag.texi
1634 @c ended inside the included file
1639 @c man begin OPTIONS
1641 See the info pages for documentation of the MMIX-specific options.
1647 @xref{NDS32 Options}, for the options available when @value{AS} is configured
1648 for a NDS32 processor.
1650 @c ended inside the included file
1654 @c man begin OPTIONS
1655 The following options are available when @value{AS} is configured for a
1658 @c man begin INCLUDE
1659 @include c-nds32.texi
1660 @c ended inside the included file
1667 @xref{PowerPC-Opts}, for the options available when @value{AS} is configured
1668 for a PowerPC processor.
1672 @c man begin OPTIONS
1673 The following options are available when @value{AS} is configured for a
1676 @c man begin INCLUDE
1678 @c ended inside the included file
1686 @xref{RISC-V-Options}, for the options available when @value{AS} is configured
1687 for a RISC-V processor.
1691 @c man begin OPTIONS
1692 The following options are available when @value{AS} is configured for a
1695 @c man begin INCLUDE
1696 @include c-riscv.texi
1697 @c ended inside the included file
1702 @c man begin OPTIONS
1704 See the info pages for documentation of the RX-specific options.
1708 The following options are available when @value{AS} is configured for the s390
1714 Select the word size, either 31/32 bits or 64 bits.
1717 Select the architecture mode, either the Enterprise System
1718 Architecture (esa) or the z/Architecture mode (zarch).
1719 @item -march=@var{processor}
1720 Specify which s390 processor variant is the target, @samp{g5} (or
1721 @samp{arch3}), @samp{g6}, @samp{z900} (or @samp{arch5}), @samp{z990} (or
1722 @samp{arch6}), @samp{z9-109}, @samp{z9-ec} (or @samp{arch7}), @samp{z10} (or
1723 @samp{arch8}), @samp{z196} (or @samp{arch9}), @samp{zEC12} (or @samp{arch10}),
1724 @samp{z13} (or @samp{arch11}), or @samp{z14} (or @samp{arch12}).
1726 @itemx -mno-regnames
1727 Allow or disallow symbolic names for registers.
1728 @item -mwarn-areg-zero
1729 Warn whenever the operand for a base or index register has been specified
1730 but evaluates to zero.
1738 @xref{TIC6X Options}, for the options available when @value{AS} is configured
1739 for a TMS320C6000 processor.
1743 @c man begin OPTIONS
1744 The following options are available when @value{AS} is configured for a
1745 TMS320C6000 processor.
1747 @c man begin INCLUDE
1748 @include c-tic6x.texi
1749 @c ended inside the included file
1757 @xref{TILE-Gx Options}, for the options available when @value{AS} is configured
1758 for a TILE-Gx processor.
1762 @c man begin OPTIONS
1763 The following options are available when @value{AS} is configured for a TILE-Gx
1766 @c man begin INCLUDE
1767 @include c-tilegx.texi
1768 @c ended inside the included file
1776 @xref{Visium Options}, for the options available when @value{AS} is configured
1777 for a Visium processor.
1781 @c man begin OPTIONS
1782 The following option is available when @value{AS} is configured for a Visium
1785 @c man begin INCLUDE
1786 @include c-visium.texi
1787 @c ended inside the included file
1795 @xref{Xtensa Options}, for the options available when @value{AS} is configured
1796 for an Xtensa processor.
1800 @c man begin OPTIONS
1801 The following options are available when @value{AS} is configured for an
1804 @c man begin INCLUDE
1805 @include c-xtensa.texi
1806 @c ended inside the included file
1811 @c man begin OPTIONS
1814 The following options are available when @value{AS} is configured for
1815 a Z80 family processor.
1818 Assemble for Z80 processor.
1820 Assemble for R800 processor.
1821 @item -ignore-undocumented-instructions
1823 Assemble undocumented Z80 instructions that also work on R800 without warning.
1824 @item -ignore-unportable-instructions
1826 Assemble all undocumented Z80 instructions without warning.
1827 @item -warn-undocumented-instructions
1829 Issue a warning for undocumented Z80 instructions that also work on R800.
1830 @item -warn-unportable-instructions
1832 Issue a warning for undocumented Z80 instructions that do not work on R800.
1833 @item -forbid-undocumented-instructions
1835 Treat all undocumented instructions as errors.
1836 @item -forbid-unportable-instructions
1838 Treat undocumented Z80 instructions that do not work on R800 as errors.
1845 * Manual:: Structure of this Manual
1846 * GNU Assembler:: The GNU Assembler
1847 * Object Formats:: Object File Formats
1848 * Command Line:: Command Line
1849 * Input Files:: Input Files
1850 * Object:: Output (Object) File
1851 * Errors:: Error and Warning Messages
1855 @section Structure of this Manual
1857 @cindex manual, structure and purpose
1858 This manual is intended to describe what you need to know to use
1859 @sc{gnu} @command{@value{AS}}. We cover the syntax expected in source files, including
1860 notation for symbols, constants, and expressions; the directives that
1861 @command{@value{AS}} understands; and of course how to invoke @command{@value{AS}}.
1864 We also cover special features in the @value{TARGET}
1865 configuration of @command{@value{AS}}, including assembler directives.
1868 This manual also describes some of the machine-dependent features of
1869 various flavors of the assembler.
1872 @cindex machine instructions (not covered)
1873 On the other hand, this manual is @emph{not} intended as an introduction
1874 to programming in assembly language---let alone programming in general!
1875 In a similar vein, we make no attempt to introduce the machine
1876 architecture; we do @emph{not} describe the instruction set, standard
1877 mnemonics, registers or addressing modes that are standard to a
1878 particular architecture.
1880 You may want to consult the manufacturer's
1881 machine architecture manual for this information.
1885 For information on the H8/300 machine instruction set, see @cite{H8/300
1886 Series Programming Manual}. For the H8/300H, see @cite{H8/300H Series
1887 Programming Manual} (Renesas).
1890 For information on the Renesas (formerly Hitachi) / SuperH SH machine instruction set,
1891 see @cite{SH-Microcomputer User's Manual} (Renesas) or
1892 @cite{SH-4 32-bit CPU Core Architecture} (SuperH) and
1893 @cite{SuperH (SH) 64-Bit RISC Series} (SuperH).
1896 For information on the Z8000 machine instruction set, see @cite{Z8000 CPU Technical Manual}
1900 @c I think this is premature---doc@cygnus.com, 17jan1991
1902 Throughout this manual, we assume that you are running @dfn{GNU},
1903 the portable operating system from the @dfn{Free Software
1904 Foundation, Inc.}. This restricts our attention to certain kinds of
1905 computer (in particular, the kinds of computers that @sc{gnu} can run on);
1906 once this assumption is granted examples and definitions need less
1909 @command{@value{AS}} is part of a team of programs that turn a high-level
1910 human-readable series of instructions into a low-level
1911 computer-readable series of instructions. Different versions of
1912 @command{@value{AS}} are used for different kinds of computer.
1915 @c There used to be a section "Terminology" here, which defined
1916 @c "contents", "byte", "word", and "long". Defining "word" to any
1917 @c particular size is confusing when the .word directive may generate 16
1918 @c bits on one machine and 32 bits on another; in general, for the user
1919 @c version of this manual, none of these terms seem essential to define.
1920 @c They were used very little even in the former draft of the manual;
1921 @c this draft makes an effort to avoid them (except in names of
1925 @section The GNU Assembler
1927 @c man begin DESCRIPTION
1929 @sc{gnu} @command{as} is really a family of assemblers.
1931 This manual describes @command{@value{AS}}, a member of that family which is
1932 configured for the @value{TARGET} architectures.
1934 If you use (or have used) the @sc{gnu} assembler on one architecture, you
1935 should find a fairly similar environment when you use it on another
1936 architecture. Each version has much in common with the others,
1937 including object file formats, most assembler directives (often called
1938 @dfn{pseudo-ops}) and assembler syntax.@refill
1940 @cindex purpose of @sc{gnu} assembler
1941 @command{@value{AS}} is primarily intended to assemble the output of the
1942 @sc{gnu} C compiler @code{@value{GCC}} for use by the linker
1943 @code{@value{LD}}. Nevertheless, we've tried to make @command{@value{AS}}
1944 assemble correctly everything that other assemblers for the same
1945 machine would assemble.
1947 Any exceptions are documented explicitly (@pxref{Machine Dependencies}).
1950 @c This remark should appear in generic version of manual; assumption
1951 @c here is that generic version sets M680x0.
1952 This doesn't mean @command{@value{AS}} always uses the same syntax as another
1953 assembler for the same architecture; for example, we know of several
1954 incompatible versions of 680x0 assembly language syntax.
1959 Unlike older assemblers, @command{@value{AS}} is designed to assemble a source
1960 program in one pass of the source file. This has a subtle impact on the
1961 @kbd{.org} directive (@pxref{Org,,@code{.org}}).
1963 @node Object Formats
1964 @section Object File Formats
1966 @cindex object file format
1967 The @sc{gnu} assembler can be configured to produce several alternative
1968 object file formats. For the most part, this does not affect how you
1969 write assembly language programs; but directives for debugging symbols
1970 are typically different in different file formats. @xref{Symbol
1971 Attributes,,Symbol Attributes}.
1974 For the @value{TARGET} target, @command{@value{AS}} is configured to produce
1975 @value{OBJ-NAME} format object files.
1977 @c The following should exhaust all configs that set MULTI-OBJ, ideally
1979 On the @value{TARGET}, @command{@value{AS}} can be configured to produce either
1980 SOM or ELF format object files.
1985 @section Command Line
1987 @cindex command line conventions
1989 After the program name @command{@value{AS}}, the command line may contain
1990 options and file names. Options may appear in any order, and may be
1991 before, after, or between file names. The order of file names is
1994 @cindex standard input, as input file
1996 @file{--} (two hyphens) by itself names the standard input file
1997 explicitly, as one of the files for @command{@value{AS}} to assemble.
1999 @cindex options, command line
2000 Except for @samp{--} any command line argument that begins with a
2001 hyphen (@samp{-}) is an option. Each option changes the behavior of
2002 @command{@value{AS}}. No option changes the way another option works. An
2003 option is a @samp{-} followed by one or more letters; the case of
2004 the letter is important. All options are optional.
2006 Some options expect exactly one file name to follow them. The file
2007 name may either immediately follow the option's letter (compatible
2008 with older assemblers) or it may be the next command argument (@sc{gnu}
2009 standard). These two command lines are equivalent:
2012 @value{AS} -o my-object-file.o mumble.s
2013 @value{AS} -omy-object-file.o mumble.s
2017 @section Input Files
2020 @cindex source program
2021 @cindex files, input
2022 We use the phrase @dfn{source program}, abbreviated @dfn{source}, to
2023 describe the program input to one run of @command{@value{AS}}. The program may
2024 be in one or more files; how the source is partitioned into files
2025 doesn't change the meaning of the source.
2027 @c I added "con" prefix to "catenation" just to prove I can overcome my
2028 @c APL training... doc@cygnus.com
2029 The source program is a concatenation of the text in all the files, in the
2032 @c man begin DESCRIPTION
2033 Each time you run @command{@value{AS}} it assembles exactly one source
2034 program. The source program is made up of one or more files.
2035 (The standard input is also a file.)
2037 You give @command{@value{AS}} a command line that has zero or more input file
2038 names. The input files are read (from left file name to right). A
2039 command line argument (in any position) that has no special meaning
2040 is taken to be an input file name.
2042 If you give @command{@value{AS}} no file names it attempts to read one input file
2043 from the @command{@value{AS}} standard input, which is normally your terminal. You
2044 may have to type @key{ctl-D} to tell @command{@value{AS}} there is no more program
2047 Use @samp{--} if you need to explicitly name the standard input file
2048 in your command line.
2050 If the source is empty, @command{@value{AS}} produces a small, empty object
2055 @subheading Filenames and Line-numbers
2057 @cindex input file linenumbers
2058 @cindex line numbers, in input files
2059 There are two ways of locating a line in the input file (or files) and
2060 either may be used in reporting error messages. One way refers to a line
2061 number in a physical file; the other refers to a line number in a
2062 ``logical'' file. @xref{Errors, ,Error and Warning Messages}.
2064 @dfn{Physical files} are those files named in the command line given
2065 to @command{@value{AS}}.
2067 @dfn{Logical files} are simply names declared explicitly by assembler
2068 directives; they bear no relation to physical files. Logical file names help
2069 error messages reflect the original source file, when @command{@value{AS}} source
2070 is itself synthesized from other files. @command{@value{AS}} understands the
2071 @samp{#} directives emitted by the @code{@value{GCC}} preprocessor. See also
2072 @ref{File,,@code{.file}}.
2075 @section Output (Object) File
2081 Every time you run @command{@value{AS}} it produces an output file, which is
2082 your assembly language program translated into numbers. This file
2083 is the object file. Its default name is @code{a.out}.
2084 You can give it another name by using the @option{-o} option. Conventionally,
2085 object file names end with @file{.o}. The default name is used for historical
2086 reasons: older assemblers were capable of assembling self-contained programs
2087 directly into a runnable program. (For some formats, this isn't currently
2088 possible, but it can be done for the @code{a.out} format.)
2092 The object file is meant for input to the linker @code{@value{LD}}. It contains
2093 assembled program code, information to help @code{@value{LD}} integrate
2094 the assembled program into a runnable file, and (optionally) symbolic
2095 information for the debugger.
2097 @c link above to some info file(s) like the description of a.out.
2098 @c don't forget to describe @sc{gnu} info as well as Unix lossage.
2101 @section Error and Warning Messages
2103 @c man begin DESCRIPTION
2105 @cindex error messages
2106 @cindex warning messages
2107 @cindex messages from assembler
2108 @command{@value{AS}} may write warnings and error messages to the standard error
2109 file (usually your terminal). This should not happen when a compiler
2110 runs @command{@value{AS}} automatically. Warnings report an assumption made so
2111 that @command{@value{AS}} could keep assembling a flawed program; errors report a
2112 grave problem that stops the assembly.
2116 @cindex format of warning messages
2117 Warning messages have the format
2120 file_name:@b{NNN}:Warning Message Text
2124 @cindex file names and line numbers, in warnings/errors
2125 (where @b{NNN} is a line number). If both a logical file name
2126 (@pxref{File,,@code{.file}}) and a logical line number
2128 (@pxref{Line,,@code{.line}})
2130 have been given then they will be used, otherwise the file name and line number
2131 in the current assembler source file will be used. The message text is
2132 intended to be self explanatory (in the grand Unix tradition).
2134 Note the file name must be set via the logical version of the @code{.file}
2135 directive, not the DWARF2 version of the @code{.file} directive. For example:
2139 error_assembler_source
2145 produces this output:
2149 asm.s:2: Error: no such instruction: `error_assembler_source'
2150 foo.c:31: Error: no such instruction: `error_c_source'
2153 @cindex format of error messages
2154 Error messages have the format
2157 file_name:@b{NNN}:FATAL:Error Message Text
2160 The file name and line number are derived as for warning
2161 messages. The actual message text may be rather less explanatory
2162 because many of them aren't supposed to happen.
2165 @chapter Command-Line Options
2167 @cindex options, all versions of assembler
2168 This chapter describes command-line options available in @emph{all}
2169 versions of the @sc{gnu} assembler; see @ref{Machine Dependencies},
2170 for options specific
2172 to the @value{TARGET} target.
2175 to particular machine architectures.
2178 @c man begin DESCRIPTION
2180 If you are invoking @command{@value{AS}} via the @sc{gnu} C compiler,
2181 you can use the @samp{-Wa} option to pass arguments through to the assembler.
2182 The assembler arguments must be separated from each other (and the @samp{-Wa})
2183 by commas. For example:
2186 gcc -c -g -O -Wa,-alh,-L file.c
2190 This passes two options to the assembler: @samp{-alh} (emit a listing to
2191 standard output with high-level and assembly source) and @samp{-L} (retain
2192 local symbols in the symbol table).
2194 Usually you do not need to use this @samp{-Wa} mechanism, since many compiler
2195 command-line options are automatically passed to the assembler by the compiler.
2196 (You can call the @sc{gnu} compiler driver with the @samp{-v} option to see
2197 precisely what options it passes to each compilation pass, including the
2203 * a:: -a[cdghlns] enable listings
2204 * alternate:: --alternate enable alternate macro syntax
2205 * D:: -D for compatibility
2206 * f:: -f to work faster
2207 * I:: -I for .include search path
2208 @ifclear DIFF-TBL-KLUGE
2209 * K:: -K for compatibility
2211 @ifset DIFF-TBL-KLUGE
2212 * K:: -K for difference tables
2215 * L:: -L to retain local symbols
2216 * listing:: --listing-XXX to configure listing output
2217 * M:: -M or --mri to assemble in MRI compatibility mode
2218 * MD:: --MD for dependency tracking
2219 * no-pad-sections:: --no-pad-sections to stop section padding
2220 * o:: -o to name the object file
2221 * R:: -R to join data and text sections
2222 * statistics:: --statistics to see statistics about assembly
2223 * traditional-format:: --traditional-format for compatible output
2224 * v:: -v to announce version
2225 * W:: -W, --no-warn, --warn, --fatal-warnings to control warnings
2226 * Z:: -Z to make object file even after errors
2230 @section Enable Listings: @option{-a[cdghlns]}
2240 @cindex listings, enabling
2241 @cindex assembly listings, enabling
2243 These options enable listing output from the assembler. By itself,
2244 @samp{-a} requests high-level, assembly, and symbols listing.
2245 You can use other letters to select specific options for the list:
2246 @samp{-ah} requests a high-level language listing,
2247 @samp{-al} requests an output-program assembly listing, and
2248 @samp{-as} requests a symbol table listing.
2249 High-level listings require that a compiler debugging option like
2250 @samp{-g} be used, and that assembly listings (@samp{-al}) be requested
2253 Use the @samp{-ag} option to print a first section with general assembly
2254 information, like @value{AS} version, switches passed, or time stamp.
2256 Use the @samp{-ac} option to omit false conditionals from a listing. Any lines
2257 which are not assembled because of a false @code{.if} (or @code{.ifdef}, or any
2258 other conditional), or a true @code{.if} followed by an @code{.else}, will be
2259 omitted from the listing.
2261 Use the @samp{-ad} option to omit debugging directives from the
2264 Once you have specified one of these options, you can further control
2265 listing output and its appearance using the directives @code{.list},
2266 @code{.nolist}, @code{.psize}, @code{.eject}, @code{.title}, and
2268 The @samp{-an} option turns off all forms processing.
2269 If you do not request listing output with one of the @samp{-a} options, the
2270 listing-control directives have no effect.
2272 The letters after @samp{-a} may be combined into one option,
2273 @emph{e.g.}, @samp{-aln}.
2275 Note if the assembler source is coming from the standard input (e.g.,
2277 is being created by @code{@value{GCC}} and the @samp{-pipe} command line switch
2278 is being used) then the listing will not contain any comments or preprocessor
2279 directives. This is because the listing code buffers input source lines from
2280 stdin only after they have been preprocessed by the assembler. This reduces
2281 memory usage and makes the code more efficient.
2284 @section @option{--alternate}
2287 Begin in alternate macro mode, see @ref{Altmacro,,@code{.altmacro}}.
2290 @section @option{-D}
2293 This option has no effect whatsoever, but it is accepted to make it more
2294 likely that scripts written for other assemblers also work with
2295 @command{@value{AS}}.
2298 @section Work Faster: @option{-f}
2301 @cindex trusted compiler
2302 @cindex faster processing (@option{-f})
2303 @samp{-f} should only be used when assembling programs written by a
2304 (trusted) compiler. @samp{-f} stops the assembler from doing whitespace
2305 and comment preprocessing on
2306 the input file(s) before assembling them. @xref{Preprocessing,
2310 @emph{Warning:} if you use @samp{-f} when the files actually need to be
2311 preprocessed (if they contain comments, for example), @command{@value{AS}} does
2316 @section @code{.include} Search Path: @option{-I} @var{path}
2318 @kindex -I @var{path}
2319 @cindex paths for @code{.include}
2320 @cindex search path for @code{.include}
2321 @cindex @code{include} directive search path
2322 Use this option to add a @var{path} to the list of directories
2323 @command{@value{AS}} searches for files specified in @code{.include}
2324 directives (@pxref{Include,,@code{.include}}). You may use @option{-I} as
2325 many times as necessary to include a variety of paths. The current
2326 working directory is always searched first; after that, @command{@value{AS}}
2327 searches any @samp{-I} directories in the same order as they were
2328 specified (left to right) on the command line.
2331 @section Difference Tables: @option{-K}
2334 @ifclear DIFF-TBL-KLUGE
2335 On the @value{TARGET} family, this option is allowed, but has no effect. It is
2336 permitted for compatibility with the @sc{gnu} assembler on other platforms,
2337 where it can be used to warn when the assembler alters the machine code
2338 generated for @samp{.word} directives in difference tables. The @value{TARGET}
2339 family does not have the addressing limitations that sometimes lead to this
2340 alteration on other platforms.
2343 @ifset DIFF-TBL-KLUGE
2344 @cindex difference tables, warning
2345 @cindex warning for altered difference tables
2346 @command{@value{AS}} sometimes alters the code emitted for directives of the
2347 form @samp{.word @var{sym1}-@var{sym2}}. @xref{Word,,@code{.word}}.
2348 You can use the @samp{-K} option if you want a warning issued when this
2353 @section Include Local Symbols: @option{-L}
2356 @cindex local symbols, retaining in output
2357 Symbols beginning with system-specific local label prefixes, typically
2358 @samp{.L} for ELF systems or @samp{L} for traditional a.out systems, are
2359 called @dfn{local symbols}. @xref{Symbol Names}. Normally you do not see
2360 such symbols when debugging, because they are intended for the use of
2361 programs (like compilers) that compose assembler programs, not for your
2362 notice. Normally both @command{@value{AS}} and @code{@value{LD}} discard
2363 such symbols, so you do not normally debug with them.
2365 This option tells @command{@value{AS}} to retain those local symbols
2366 in the object file. Usually if you do this you also tell the linker
2367 @code{@value{LD}} to preserve those symbols.
2370 @section Configuring listing output: @option{--listing}
2372 The listing feature of the assembler can be enabled via the command line switch
2373 @samp{-a} (@pxref{a}). This feature combines the input source file(s) with a
2374 hex dump of the corresponding locations in the output object file, and displays
2375 them as a listing file. The format of this listing can be controlled by
2376 directives inside the assembler source (i.e., @code{.list} (@pxref{List}),
2377 @code{.title} (@pxref{Title}), @code{.sbttl} (@pxref{Sbttl}),
2378 @code{.psize} (@pxref{Psize}), and
2379 @code{.eject} (@pxref{Eject}) and also by the following switches:
2382 @item --listing-lhs-width=@samp{number}
2383 @kindex --listing-lhs-width
2384 @cindex Width of first line disassembly output
2385 Sets the maximum width, in words, of the first line of the hex byte dump. This
2386 dump appears on the left hand side of the listing output.
2388 @item --listing-lhs-width2=@samp{number}
2389 @kindex --listing-lhs-width2
2390 @cindex Width of continuation lines of disassembly output
2391 Sets the maximum width, in words, of any further lines of the hex byte dump for
2392 a given input source line. If this value is not specified, it defaults to being
2393 the same as the value specified for @samp{--listing-lhs-width}. If neither
2394 switch is used the default is to one.
2396 @item --listing-rhs-width=@samp{number}
2397 @kindex --listing-rhs-width
2398 @cindex Width of source line output
2399 Sets the maximum width, in characters, of the source line that is displayed
2400 alongside the hex dump. The default value for this parameter is 100. The
2401 source line is displayed on the right hand side of the listing output.
2403 @item --listing-cont-lines=@samp{number}
2404 @kindex --listing-cont-lines
2405 @cindex Maximum number of continuation lines
2406 Sets the maximum number of continuation lines of hex dump that will be
2407 displayed for a given single line of source input. The default value is 4.
2411 @section Assemble in MRI Compatibility Mode: @option{-M}
2414 @cindex MRI compatibility mode
2415 The @option{-M} or @option{--mri} option selects MRI compatibility mode. This
2416 changes the syntax and pseudo-op handling of @command{@value{AS}} to make it
2417 compatible with the @code{ASM68K} assembler from Microtec Research.
2418 The exact nature of the
2419 MRI syntax will not be documented here; see the MRI manuals for more
2420 information. Note in particular that the handling of macros and macro
2421 arguments is somewhat different. The purpose of this option is to permit
2422 assembling existing MRI assembler code using @command{@value{AS}}.
2424 The MRI compatibility is not complete. Certain operations of the MRI assembler
2425 depend upon its object file format, and can not be supported using other object
2426 file formats. Supporting these would require enhancing each object file format
2427 individually. These are:
2430 @item global symbols in common section
2432 The m68k MRI assembler supports common sections which are merged by the linker.
2433 Other object file formats do not support this. @command{@value{AS}} handles
2434 common sections by treating them as a single common symbol. It permits local
2435 symbols to be defined within a common section, but it can not support global
2436 symbols, since it has no way to describe them.
2438 @item complex relocations
2440 The MRI assemblers support relocations against a negated section address, and
2441 relocations which combine the start addresses of two or more sections. These
2442 are not support by other object file formats.
2444 @item @code{END} pseudo-op specifying start address
2446 The MRI @code{END} pseudo-op permits the specification of a start address.
2447 This is not supported by other object file formats. The start address may
2448 instead be specified using the @option{-e} option to the linker, or in a linker
2451 @item @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops
2453 The MRI @code{IDNT}, @code{.ident} and @code{NAME} pseudo-ops assign a module
2454 name to the output file. This is not supported by other object file formats.
2456 @item @code{ORG} pseudo-op
2458 The m68k MRI @code{ORG} pseudo-op begins an absolute section at a given
2459 address. This differs from the usual @command{@value{AS}} @code{.org} pseudo-op,
2460 which changes the location within the current section. Absolute sections are
2461 not supported by other object file formats. The address of a section may be
2462 assigned within a linker script.
2465 There are some other features of the MRI assembler which are not supported by
2466 @command{@value{AS}}, typically either because they are difficult or because they
2467 seem of little consequence. Some of these may be supported in future releases.
2471 @item EBCDIC strings
2473 EBCDIC strings are not supported.
2475 @item packed binary coded decimal
2477 Packed binary coded decimal is not supported. This means that the @code{DC.P}
2478 and @code{DCB.P} pseudo-ops are not supported.
2480 @item @code{FEQU} pseudo-op
2482 The m68k @code{FEQU} pseudo-op is not supported.
2484 @item @code{NOOBJ} pseudo-op
2486 The m68k @code{NOOBJ} pseudo-op is not supported.
2488 @item @code{OPT} branch control options
2490 The m68k @code{OPT} branch control options---@code{B}, @code{BRS}, @code{BRB},
2491 @code{BRL}, and @code{BRW}---are ignored. @command{@value{AS}} automatically
2492 relaxes all branches, whether forward or backward, to an appropriate size, so
2493 these options serve no purpose.
2495 @item @code{OPT} list control options
2497 The following m68k @code{OPT} list control options are ignored: @code{C},
2498 @code{CEX}, @code{CL}, @code{CRE}, @code{E}, @code{G}, @code{I}, @code{M},
2499 @code{MEX}, @code{MC}, @code{MD}, @code{X}.
2501 @item other @code{OPT} options
2503 The following m68k @code{OPT} options are ignored: @code{NEST}, @code{O},
2504 @code{OLD}, @code{OP}, @code{P}, @code{PCO}, @code{PCR}, @code{PCS}, @code{R}.
2506 @item @code{OPT} @code{D} option is default
2508 The m68k @code{OPT} @code{D} option is the default, unlike the MRI assembler.
2509 @code{OPT NOD} may be used to turn it off.
2511 @item @code{XREF} pseudo-op.
2513 The m68k @code{XREF} pseudo-op is ignored.
2518 @section Dependency Tracking: @option{--MD}
2521 @cindex dependency tracking
2524 @command{@value{AS}} can generate a dependency file for the file it creates. This
2525 file consists of a single rule suitable for @code{make} describing the
2526 dependencies of the main source file.
2528 The rule is written to the file named in its argument.
2530 This feature is used in the automatic updating of makefiles.
2532 @node no-pad-sections
2533 @section Output Section Padding
2534 @kindex --no-pad-sections
2535 @cindex output section padding
2536 Normally the assembler will pad the end of each output section up to its
2537 alignment boundary. But this can waste space, which can be significant on
2538 memory constrained targets. So the @option{--no-pad-sections} option will
2539 disable this behaviour.
2542 @section Name the Object File: @option{-o}
2545 @cindex naming object file
2546 @cindex object file name
2547 There is always one object file output when you run @command{@value{AS}}. By
2548 default it has the name @file{a.out}.
2549 You use this option (which takes exactly one filename) to give the
2550 object file a different name.
2552 Whatever the object file is called, @command{@value{AS}} overwrites any
2553 existing file of the same name.
2556 @section Join Data and Text Sections: @option{-R}
2559 @cindex data and text sections, joining
2560 @cindex text and data sections, joining
2561 @cindex joining text and data sections
2562 @cindex merging text and data sections
2563 @option{-R} tells @command{@value{AS}} to write the object file as if all
2564 data-section data lives in the text section. This is only done at
2565 the very last moment: your binary data are the same, but data
2566 section parts are relocated differently. The data section part of
2567 your object file is zero bytes long because all its bytes are
2568 appended to the text section. (@xref{Sections,,Sections and Relocation}.)
2570 When you specify @option{-R} it would be possible to generate shorter
2571 address displacements (because we do not have to cross between text and
2572 data section). We refrain from doing this simply for compatibility with
2573 older versions of @command{@value{AS}}. In future, @option{-R} may work this way.
2576 When @command{@value{AS}} is configured for COFF or ELF output,
2577 this option is only useful if you use sections named @samp{.text} and
2582 @option{-R} is not supported for any of the HPPA targets. Using
2583 @option{-R} generates a warning from @command{@value{AS}}.
2587 @section Display Assembly Statistics: @option{--statistics}
2589 @kindex --statistics
2590 @cindex statistics, about assembly
2591 @cindex time, total for assembly
2592 @cindex space used, maximum for assembly
2593 Use @samp{--statistics} to display two statistics about the resources used by
2594 @command{@value{AS}}: the maximum amount of space allocated during the assembly
2595 (in bytes), and the total execution time taken for the assembly (in @sc{cpu}
2598 @node traditional-format
2599 @section Compatible Output: @option{--traditional-format}
2601 @kindex --traditional-format
2602 For some targets, the output of @command{@value{AS}} is different in some ways
2603 from the output of some existing assembler. This switch requests
2604 @command{@value{AS}} to use the traditional format instead.
2606 For example, it disables the exception frame optimizations which
2607 @command{@value{AS}} normally does by default on @code{@value{GCC}} output.
2610 @section Announce Version: @option{-v}
2614 @cindex assembler version
2615 @cindex version of assembler
2616 You can find out what version of as is running by including the
2617 option @samp{-v} (which you can also spell as @samp{-version}) on the
2621 @section Control Warnings: @option{-W}, @option{--warn}, @option{--no-warn}, @option{--fatal-warnings}
2623 @command{@value{AS}} should never give a warning or error message when
2624 assembling compiler output. But programs written by people often
2625 cause @command{@value{AS}} to give a warning that a particular assumption was
2626 made. All such warnings are directed to the standard error file.
2630 @cindex suppressing warnings
2631 @cindex warnings, suppressing
2632 If you use the @option{-W} and @option{--no-warn} options, no warnings are issued.
2633 This only affects the warning messages: it does not change any particular of
2634 how @command{@value{AS}} assembles your file. Errors, which stop the assembly,
2637 @kindex --fatal-warnings
2638 @cindex errors, caused by warnings
2639 @cindex warnings, causing error
2640 If you use the @option{--fatal-warnings} option, @command{@value{AS}} considers
2641 files that generate warnings to be in error.
2644 @cindex warnings, switching on
2645 You can switch these options off again by specifying @option{--warn}, which
2646 causes warnings to be output as usual.
2649 @section Generate Object File in Spite of Errors: @option{-Z}
2650 @cindex object file, after errors
2651 @cindex errors, continuing after
2652 After an error message, @command{@value{AS}} normally produces no output. If for
2653 some reason you are interested in object file output even after
2654 @command{@value{AS}} gives an error message on your program, use the @samp{-Z}
2655 option. If there are any errors, @command{@value{AS}} continues anyways, and
2656 writes an object file after a final warning message of the form @samp{@var{n}
2657 errors, @var{m} warnings, generating bad object file.}
2662 @cindex machine-independent syntax
2663 @cindex syntax, machine-independent
2664 This chapter describes the machine-independent syntax allowed in a
2665 source file. @command{@value{AS}} syntax is similar to what many other
2666 assemblers use; it is inspired by the BSD 4.2
2671 assembler, except that @command{@value{AS}} does not assemble Vax bit-fields.
2675 * Preprocessing:: Preprocessing
2676 * Whitespace:: Whitespace
2677 * Comments:: Comments
2678 * Symbol Intro:: Symbols
2679 * Statements:: Statements
2680 * Constants:: Constants
2684 @section Preprocessing
2686 @cindex preprocessing
2687 The @command{@value{AS}} internal preprocessor:
2689 @cindex whitespace, removed by preprocessor
2691 adjusts and removes extra whitespace. It leaves one space or tab before
2692 the keywords on a line, and turns any other whitespace on the line into
2695 @cindex comments, removed by preprocessor
2697 removes all comments, replacing them with a single space, or an
2698 appropriate number of newlines.
2700 @cindex constants, converted by preprocessor
2702 converts character constants into the appropriate numeric values.
2705 It does not do macro processing, include file handling, or
2706 anything else you may get from your C compiler's preprocessor. You can
2707 do include file processing with the @code{.include} directive
2708 (@pxref{Include,,@code{.include}}). You can use the @sc{gnu} C compiler driver
2709 to get other ``CPP'' style preprocessing by giving the input file a
2710 @samp{.S} suffix. @xref{Overall Options, ,Options Controlling the Kind of
2711 Output, gcc info, Using GNU CC}.
2713 Excess whitespace, comments, and character constants
2714 cannot be used in the portions of the input text that are not
2717 @cindex turning preprocessing on and off
2718 @cindex preprocessing, turning on and off
2721 If the first line of an input file is @code{#NO_APP} or if you use the
2722 @samp{-f} option, whitespace and comments are not removed from the input file.
2723 Within an input file, you can ask for whitespace and comment removal in
2724 specific portions of the by putting a line that says @code{#APP} before the
2725 text that may contain whitespace or comments, and putting a line that says
2726 @code{#NO_APP} after this text. This feature is mainly intend to support
2727 @code{asm} statements in compilers whose output is otherwise free of comments
2734 @dfn{Whitespace} is one or more blanks or tabs, in any order.
2735 Whitespace is used to separate symbols, and to make programs neater for
2736 people to read. Unless within character constants
2737 (@pxref{Characters,,Character Constants}), any whitespace means the same
2738 as exactly one space.
2744 There are two ways of rendering comments to @command{@value{AS}}. In both
2745 cases the comment is equivalent to one space.
2747 Anything from @samp{/*} through the next @samp{*/} is a comment.
2748 This means you may not nest these comments.
2752 The only way to include a newline ('\n') in a comment
2753 is to use this sort of comment.
2756 /* This sort of comment does not nest. */
2759 @cindex line comment character
2760 Anything from a @dfn{line comment} character up to the next newline is
2761 considered a comment and is ignored. The line comment character is target
2762 specific, and some targets multiple comment characters. Some targets also have
2763 line comment characters that only work if they are the first character on a
2764 line. Some targets use a sequence of two characters to introduce a line
2765 comment. Some targets can also change their line comment characters depending
2766 upon command line options that have been used. For more details see the
2767 @emph{Syntax} section in the documentation for individual targets.
2769 If the line comment character is the hash sign (@samp{#}) then it still has the
2770 special ability to enable and disable preprocessing (@pxref{Preprocessing}) and
2771 to specify logical line numbers:
2774 @cindex lines starting with @code{#}
2775 @cindex logical line numbers
2776 To be compatible with past assemblers, lines that begin with @samp{#} have a
2777 special interpretation. Following the @samp{#} should be an absolute
2778 expression (@pxref{Expressions}): the logical line number of the @emph{next}
2779 line. Then a string (@pxref{Strings, ,Strings}) is allowed: if present it is a
2780 new logical file name. The rest of the line, if any, should be whitespace.
2782 If the first non-whitespace characters on the line are not numeric,
2783 the line is ignored. (Just like a comment.)
2786 # This is an ordinary comment.
2787 # 42-6 "new_file_name" # New logical file name
2788 # This is logical line # 36.
2790 This feature is deprecated, and may disappear from future versions
2791 of @command{@value{AS}}.
2796 @cindex characters used in symbols
2797 @ifclear SPECIAL-SYMS
2798 A @dfn{symbol} is one or more characters chosen from the set of all
2799 letters (both upper and lower case), digits and the three characters
2805 A @dfn{symbol} is one or more characters chosen from the set of all
2806 letters (both upper and lower case), digits and the three characters
2807 @samp{._$}. (Save that, on the H8/300 only, you may not use @samp{$} in
2813 On most machines, you can also use @code{$} in symbol names; exceptions
2814 are noted in @ref{Machine Dependencies}.
2816 No symbol may begin with a digit. Case is significant.
2817 There is no length limit; all characters are significant. Multibyte characters
2818 are supported. Symbols are delimited by characters not in that set, or by the
2819 beginning of a file (since the source program must end with a newline, the end
2820 of a file is not a possible symbol delimiter). @xref{Symbols}.
2822 Symbol names may also be enclosed in double quote @code{"} characters. In such
2823 cases any characters are allowed, except for the NUL character. If a double
2824 quote character is to be included in the symbol name it must be preceeded by a
2825 backslash @code{\} character.
2826 @cindex length of symbols
2831 @cindex statements, structure of
2832 @cindex line separator character
2833 @cindex statement separator character
2835 A @dfn{statement} ends at a newline character (@samp{\n}) or a
2836 @dfn{line separator character}. The line separator character is target
2837 specific and described in the @emph{Syntax} section of each
2838 target's documentation. Not all targets support a line separator character.
2839 The newline or line separator character is considered to be part of the
2840 preceding statement. Newlines and separators within character constants are an
2841 exception: they do not end statements.
2843 @cindex newline, required at file end
2844 @cindex EOF, newline must precede
2845 It is an error to end any statement with end-of-file: the last
2846 character of any input file should be a newline.@refill
2848 An empty statement is allowed, and may include whitespace. It is ignored.
2850 @cindex instructions and directives
2851 @cindex directives and instructions
2852 @c "key symbol" is not used elsewhere in the document; seems pedantic to
2853 @c @defn{} it in that case, as was done previously... doc@cygnus.com,
2855 A statement begins with zero or more labels, optionally followed by a
2856 key symbol which determines what kind of statement it is. The key
2857 symbol determines the syntax of the rest of the statement. If the
2858 symbol begins with a dot @samp{.} then the statement is an assembler
2859 directive: typically valid for any computer. If the symbol begins with
2860 a letter the statement is an assembly language @dfn{instruction}: it
2861 assembles into a machine language instruction.
2863 Different versions of @command{@value{AS}} for different computers
2864 recognize different instructions. In fact, the same symbol may
2865 represent a different instruction in a different computer's assembly
2869 @cindex @code{:} (label)
2870 @cindex label (@code{:})
2871 A label is a symbol immediately followed by a colon (@code{:}).
2872 Whitespace before a label or after a colon is permitted, but you may not
2873 have whitespace between a label's symbol and its colon. @xref{Labels}.
2876 For HPPA targets, labels need not be immediately followed by a colon, but
2877 the definition of a label must begin in column zero. This also implies that
2878 only one label may be defined on each line.
2882 label: .directive followed by something
2883 another_label: # This is an empty statement.
2884 instruction operand_1, operand_2, @dots{}
2891 A constant is a number, written so that its value is known by
2892 inspection, without knowing any context. Like this:
2895 .byte 74, 0112, 092, 0x4A, 0X4a, 'J, '\J # All the same value.
2896 .ascii "Ring the bell\7" # A string constant.
2897 .octa 0x123456789abcdef0123456789ABCDEF0 # A bignum.
2898 .float 0f-314159265358979323846264338327\
2899 95028841971.693993751E-40 # - pi, a flonum.
2904 * Characters:: Character Constants
2905 * Numbers:: Number Constants
2909 @subsection Character Constants
2911 @cindex character constants
2912 @cindex constants, character
2913 There are two kinds of character constants. A @dfn{character} stands
2914 for one character in one byte and its value may be used in
2915 numeric expressions. String constants (properly called string
2916 @emph{literals}) are potentially many bytes and their values may not be
2917 used in arithmetic expressions.
2921 * Chars:: Characters
2925 @subsubsection Strings
2927 @cindex string constants
2928 @cindex constants, string
2929 A @dfn{string} is written between double-quotes. It may contain
2930 double-quotes or null characters. The way to get special characters
2931 into a string is to @dfn{escape} these characters: precede them with
2932 a backslash @samp{\} character. For example @samp{\\} represents
2933 one backslash: the first @code{\} is an escape which tells
2934 @command{@value{AS}} to interpret the second character literally as a backslash
2935 (which prevents @command{@value{AS}} from recognizing the second @code{\} as an
2936 escape character). The complete list of escapes follows.
2938 @cindex escape codes, character
2939 @cindex character escape codes
2940 @c NOTE: Cindex entries must not start with a backlash character.
2941 @c NOTE: This confuses the pdf2texi script when it is creating the
2942 @c NOTE: index based upon the first character and so it generates:
2943 @c NOTE: \initial {\\}
2944 @c NOTE: which then results in the error message:
2945 @c NOTE: Argument of \\ has an extra }.
2946 @c NOTE: So in the index entries below a space character has been
2947 @c NOTE: prepended to avoid this problem.
2950 @c Mnemonic for ACKnowledge; for ASCII this is octal code 007.
2952 @cindex @code{ \b} (backspace character)
2953 @cindex backspace (@code{\b})
2955 Mnemonic for backspace; for ASCII this is octal code 010.
2958 @c Mnemonic for EOText; for ASCII this is octal code 004.
2960 @cindex @code{ \f} (formfeed character)
2961 @cindex formfeed (@code{\f})
2963 Mnemonic for FormFeed; for ASCII this is octal code 014.
2965 @cindex @code{ \n} (newline character)
2966 @cindex newline (@code{\n})
2968 Mnemonic for newline; for ASCII this is octal code 012.
2971 @c Mnemonic for prefix; for ASCII this is octal code 033, usually known as @code{escape}.
2973 @cindex @code{ \r} (carriage return character)
2974 @cindex carriage return (@code{backslash-r})
2976 Mnemonic for carriage-Return; for ASCII this is octal code 015.
2979 @c Mnemonic for space; for ASCII this is octal code 040. Included for compliance with
2980 @c other assemblers.
2982 @cindex @code{ \t} (tab)
2983 @cindex tab (@code{\t})
2985 Mnemonic for horizontal Tab; for ASCII this is octal code 011.
2988 @c Mnemonic for Vertical tab; for ASCII this is octal code 013.
2989 @c @item \x @var{digit} @var{digit} @var{digit}
2990 @c A hexadecimal character code. The numeric code is 3 hexadecimal digits.
2992 @cindex @code{ \@var{ddd}} (octal character code)
2993 @cindex octal character code (@code{\@var{ddd}})
2994 @item \ @var{digit} @var{digit} @var{digit}
2995 An octal character code. The numeric code is 3 octal digits.
2996 For compatibility with other Unix systems, 8 and 9 are accepted as digits:
2997 for example, @code{\008} has the value 010, and @code{\009} the value 011.
2999 @cindex @code{ \@var{xd...}} (hex character code)
3000 @cindex hex character code (@code{\@var{xd...}})
3001 @item \@code{x} @var{hex-digits...}
3002 A hex character code. All trailing hex digits are combined. Either upper or
3003 lower case @code{x} works.
3005 @cindex @code{ \\} (@samp{\} character)
3006 @cindex backslash (@code{\\})
3008 Represents one @samp{\} character.
3011 @c Represents one @samp{'} (accent acute) character.
3012 @c This is needed in single character literals
3013 @c (@xref{Characters,,Character Constants}.) to represent
3016 @cindex @code{ \"} (doublequote character)
3017 @cindex doublequote (@code{\"})
3019 Represents one @samp{"} character. Needed in strings to represent
3020 this character, because an unescaped @samp{"} would end the string.
3022 @item \ @var{anything-else}
3023 Any other character when escaped by @kbd{\} gives a warning, but
3024 assembles as if the @samp{\} was not present. The idea is that if
3025 you used an escape sequence you clearly didn't want the literal
3026 interpretation of the following character. However @command{@value{AS}} has no
3027 other interpretation, so @command{@value{AS}} knows it is giving you the wrong
3028 code and warns you of the fact.
3031 Which characters are escapable, and what those escapes represent,
3032 varies widely among assemblers. The current set is what we think
3033 the BSD 4.2 assembler recognizes, and is a subset of what most C
3034 compilers recognize. If you are in doubt, do not use an escape
3038 @subsubsection Characters
3040 @cindex single character constant
3041 @cindex character, single
3042 @cindex constant, single character
3043 A single character may be written as a single quote immediately followed by
3044 that character. Some backslash escapes apply to characters, @code{\b},
3045 @code{\f}, @code{\n}, @code{\r}, @code{\t}, and @code{\"} with the same meaning
3046 as for strings, plus @code{\'} for a single quote. So if you want to write the
3047 character backslash, you must write @kbd{'\\} where the first @code{\} escapes
3048 the second @code{\}. As you can see, the quote is an acute accent, not a grave
3051 @ifclear abnormal-separator
3052 (or semicolon @samp{;})
3054 @ifset abnormal-separator
3056 (or dollar sign @samp{$}, for the H8/300; or semicolon @samp{;} for the
3061 immediately following an acute accent is taken as a literal character
3062 and does not count as the end of a statement. The value of a character
3063 constant in a numeric expression is the machine's byte-wide code for
3064 that character. @command{@value{AS}} assumes your character code is ASCII:
3065 @kbd{'A} means 65, @kbd{'B} means 66, and so on. @refill
3068 @subsection Number Constants
3070 @cindex constants, number
3071 @cindex number constants
3072 @command{@value{AS}} distinguishes three kinds of numbers according to how they
3073 are stored in the target machine. @emph{Integers} are numbers that
3074 would fit into an @code{int} in the C language. @emph{Bignums} are
3075 integers, but they are stored in more than 32 bits. @emph{Flonums}
3076 are floating point numbers, described below.
3079 * Integers:: Integers
3087 @subsubsection Integers
3089 @cindex constants, integer
3091 @cindex binary integers
3092 @cindex integers, binary
3093 A binary integer is @samp{0b} or @samp{0B} followed by zero or more of
3094 the binary digits @samp{01}.
3096 @cindex octal integers
3097 @cindex integers, octal
3098 An octal integer is @samp{0} followed by zero or more of the octal
3099 digits (@samp{01234567}).
3101 @cindex decimal integers
3102 @cindex integers, decimal
3103 A decimal integer starts with a non-zero digit followed by zero or
3104 more digits (@samp{0123456789}).
3106 @cindex hexadecimal integers
3107 @cindex integers, hexadecimal
3108 A hexadecimal integer is @samp{0x} or @samp{0X} followed by one or
3109 more hexadecimal digits chosen from @samp{0123456789abcdefABCDEF}.
3111 Integers have the usual values. To denote a negative integer, use
3112 the prefix operator @samp{-} discussed under expressions
3113 (@pxref{Prefix Ops,,Prefix Operators}).
3116 @subsubsection Bignums
3119 @cindex constants, bignum
3120 A @dfn{bignum} has the same syntax and semantics as an integer
3121 except that the number (or its negative) takes more than 32 bits to
3122 represent in binary. The distinction is made because in some places
3123 integers are permitted while bignums are not.
3126 @subsubsection Flonums
3128 @cindex floating point numbers
3129 @cindex constants, floating point
3131 @cindex precision, floating point
3132 A @dfn{flonum} represents a floating point number. The translation is
3133 indirect: a decimal floating point number from the text is converted by
3134 @command{@value{AS}} to a generic binary floating point number of more than
3135 sufficient precision. This generic floating point number is converted
3136 to a particular computer's floating point format (or formats) by a
3137 portion of @command{@value{AS}} specialized to that computer.
3139 A flonum is written by writing (in order)
3144 (@samp{0} is optional on the HPPA.)
3148 A letter, to tell @command{@value{AS}} the rest of the number is a flonum.
3150 @kbd{e} is recommended. Case is not important.
3152 @c FIXME: verify if flonum syntax really this vague for most cases
3153 (Any otherwise illegal letter works here, but that might be changed. Vax BSD
3154 4.2 assembler seems to allow any of @samp{defghDEFGH}.)
3157 On the H8/300 and Renesas / SuperH SH architectures, the letter must be
3158 one of the letters @samp{DFPRSX} (in upper or lower case).
3160 On the ARC, the letter must be one of the letters @samp{DFRS}
3161 (in upper or lower case).
3163 On the HPPA architecture, the letter must be @samp{E} (upper case only).
3167 One of the letters @samp{DFRS} (in upper or lower case).
3170 One of the letters @samp{DFPRSX} (in upper or lower case).
3173 The letter @samp{E} (upper case only).
3178 An optional sign: either @samp{+} or @samp{-}.
3181 An optional @dfn{integer part}: zero or more decimal digits.
3184 An optional @dfn{fractional part}: @samp{.} followed by zero
3185 or more decimal digits.
3188 An optional exponent, consisting of:
3192 An @samp{E} or @samp{e}.
3193 @c I can't find a config where "EXP_CHARS" is other than 'eE', but in
3194 @c principle this can perfectly well be different on different targets.
3196 Optional sign: either @samp{+} or @samp{-}.
3198 One or more decimal digits.
3203 At least one of the integer part or the fractional part must be
3204 present. The floating point number has the usual base-10 value.
3206 @command{@value{AS}} does all processing using integers. Flonums are computed
3207 independently of any floating point hardware in the computer running
3208 @command{@value{AS}}.
3211 @chapter Sections and Relocation
3216 * Secs Background:: Background
3217 * Ld Sections:: Linker Sections
3218 * As Sections:: Assembler Internal Sections
3219 * Sub-Sections:: Sub-Sections
3223 @node Secs Background
3226 Roughly, a section is a range of addresses, with no gaps; all data
3227 ``in'' those addresses is treated the same for some particular purpose.
3228 For example there may be a ``read only'' section.
3230 @cindex linker, and assembler
3231 @cindex assembler, and linker
3232 The linker @code{@value{LD}} reads many object files (partial programs) and
3233 combines their contents to form a runnable program. When @command{@value{AS}}
3234 emits an object file, the partial program is assumed to start at address 0.
3235 @code{@value{LD}} assigns the final addresses for the partial program, so that
3236 different partial programs do not overlap. This is actually an
3237 oversimplification, but it suffices to explain how @command{@value{AS}} uses
3240 @code{@value{LD}} moves blocks of bytes of your program to their run-time
3241 addresses. These blocks slide to their run-time addresses as rigid
3242 units; their length does not change and neither does the order of bytes
3243 within them. Such a rigid unit is called a @emph{section}. Assigning
3244 run-time addresses to sections is called @dfn{relocation}. It includes
3245 the task of adjusting mentions of object-file addresses so they refer to
3246 the proper run-time addresses.
3248 For the H8/300, and for the Renesas / SuperH SH,
3249 @command{@value{AS}} pads sections if needed to
3250 ensure they end on a word (sixteen bit) boundary.
3253 @cindex standard assembler sections
3254 An object file written by @command{@value{AS}} has at least three sections, any
3255 of which may be empty. These are named @dfn{text}, @dfn{data} and
3260 When it generates COFF or ELF output,
3262 @command{@value{AS}} can also generate whatever other named sections you specify
3263 using the @samp{.section} directive (@pxref{Section,,@code{.section}}).
3264 If you do not use any directives that place output in the @samp{.text}
3265 or @samp{.data} sections, these sections still exist, but are empty.
3270 When @command{@value{AS}} generates SOM or ELF output for the HPPA,
3272 @command{@value{AS}} can also generate whatever other named sections you
3273 specify using the @samp{.space} and @samp{.subspace} directives. See
3274 @cite{HP9000 Series 800 Assembly Language Reference Manual}
3275 (HP 92432-90001) for details on the @samp{.space} and @samp{.subspace}
3276 assembler directives.
3279 Additionally, @command{@value{AS}} uses different names for the standard
3280 text, data, and bss sections when generating SOM output. Program text
3281 is placed into the @samp{$CODE$} section, data into @samp{$DATA$}, and
3282 BSS into @samp{$BSS$}.
3286 Within the object file, the text section starts at address @code{0}, the
3287 data section follows, and the bss section follows the data section.
3290 When generating either SOM or ELF output files on the HPPA, the text
3291 section starts at address @code{0}, the data section at address
3292 @code{0x4000000}, and the bss section follows the data section.
3295 To let @code{@value{LD}} know which data changes when the sections are
3296 relocated, and how to change that data, @command{@value{AS}} also writes to the
3297 object file details of the relocation needed. To perform relocation
3298 @code{@value{LD}} must know, each time an address in the object
3302 Where in the object file is the beginning of this reference to
3305 How long (in bytes) is this reference?
3307 Which section does the address refer to? What is the numeric value of
3309 (@var{address}) @minus{} (@var{start-address of section})?
3312 Is the reference to an address ``Program-Counter relative''?
3315 @cindex addresses, format of
3316 @cindex section-relative addressing
3317 In fact, every address @command{@value{AS}} ever uses is expressed as
3319 (@var{section}) + (@var{offset into section})
3322 Further, most expressions @command{@value{AS}} computes have this section-relative
3325 (For some object formats, such as SOM for the HPPA, some expressions are
3326 symbol-relative instead.)
3329 In this manual we use the notation @{@var{secname} @var{N}@} to mean ``offset
3330 @var{N} into section @var{secname}.''
3332 Apart from text, data and bss sections you need to know about the
3333 @dfn{absolute} section. When @code{@value{LD}} mixes partial programs,
3334 addresses in the absolute section remain unchanged. For example, address
3335 @code{@{absolute 0@}} is ``relocated'' to run-time address 0 by
3336 @code{@value{LD}}. Although the linker never arranges two partial programs'
3337 data sections with overlapping addresses after linking, @emph{by definition}
3338 their absolute sections must overlap. Address @code{@{absolute@ 239@}} in one
3339 part of a program is always the same address when the program is running as
3340 address @code{@{absolute@ 239@}} in any other part of the program.
3342 The idea of sections is extended to the @dfn{undefined} section. Any
3343 address whose section is unknown at assembly time is by definition
3344 rendered @{undefined @var{U}@}---where @var{U} is filled in later.
3345 Since numbers are always defined, the only way to generate an undefined
3346 address is to mention an undefined symbol. A reference to a named
3347 common block would be such a symbol: its value is unknown at assembly
3348 time so it has section @emph{undefined}.
3350 By analogy the word @emph{section} is used to describe groups of sections in
3351 the linked program. @code{@value{LD}} puts all partial programs' text
3352 sections in contiguous addresses in the linked program. It is
3353 customary to refer to the @emph{text section} of a program, meaning all
3354 the addresses of all partial programs' text sections. Likewise for
3355 data and bss sections.
3357 Some sections are manipulated by @code{@value{LD}}; others are invented for
3358 use of @command{@value{AS}} and have no meaning except during assembly.
3361 @section Linker Sections
3362 @code{@value{LD}} deals with just four kinds of sections, summarized below.
3367 @cindex named sections
3368 @cindex sections, named
3369 @item named sections
3372 @cindex text section
3373 @cindex data section
3377 These sections hold your program. @command{@value{AS}} and @code{@value{LD}} treat them as
3378 separate but equal sections. Anything you can say of one section is
3381 When the program is running, however, it is
3382 customary for the text section to be unalterable. The
3383 text section is often shared among processes: it contains
3384 instructions, constants and the like. The data section of a running
3385 program is usually alterable: for example, C variables would be stored
3386 in the data section.
3391 This section contains zeroed bytes when your program begins running. It
3392 is used to hold uninitialized variables or common storage. The length of
3393 each partial program's bss section is important, but because it starts
3394 out containing zeroed bytes there is no need to store explicit zero
3395 bytes in the object file. The bss section was invented to eliminate
3396 those explicit zeros from object files.
3398 @cindex absolute section
3399 @item absolute section
3400 Address 0 of this section is always ``relocated'' to runtime address 0.
3401 This is useful if you want to refer to an address that @code{@value{LD}} must
3402 not change when relocating. In this sense we speak of absolute
3403 addresses being ``unrelocatable'': they do not change during relocation.
3405 @cindex undefined section
3406 @item undefined section
3407 This ``section'' is a catch-all for address references to objects not in
3408 the preceding sections.
3409 @c FIXME: ref to some other doc on obj-file formats could go here.
3412 @cindex relocation example
3413 An idealized example of three relocatable sections follows.
3415 The example uses the traditional section names @samp{.text} and @samp{.data}.
3417 Memory addresses are on the horizontal axis.
3421 @c END TEXI2ROFF-KILL
3424 partial program # 1: |ttttt|dddd|00|
3431 partial program # 2: |TTT|DDD|000|
3434 +--+---+-----+--+----+---+-----+~~
3435 linked program: | |TTT|ttttt| |dddd|DDD|00000|
3436 +--+---+-----+--+----+---+-----+~~
3438 addresses: 0 @dots{}
3445 \line{\it Partial program \#1: \hfil}
3446 \line{\ibox{2.5cm}{\tt text}\ibox{2cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3447 \line{\boxit{2.5cm}{\tt ttttt}\boxit{2cm}{\tt dddd}\boxit{1cm}{\tt 00}\hfil}
3449 \line{\it Partial program \#2: \hfil}
3450 \line{\ibox{1cm}{\tt text}\ibox{1.5cm}{\tt data}\ibox{1cm}{\tt bss}\hfil}
3451 \line{\boxit{1cm}{\tt TTT}\boxit{1.5cm}{\tt DDDD}\boxit{1cm}{\tt 000}\hfil}
3453 \line{\it linked program: \hfil}
3454 \line{\ibox{.5cm}{}\ibox{1cm}{\tt text}\ibox{2.5cm}{}\ibox{.75cm}{}\ibox{2cm}{\tt data}\ibox{1.5cm}{}\ibox{2cm}{\tt bss}\hfil}
3455 \line{\boxit{.5cm}{}\boxit{1cm}{\tt TTT}\boxit{2.5cm}{\tt
3456 ttttt}\boxit{.75cm}{}\boxit{2cm}{\tt dddd}\boxit{1.5cm}{\tt
3457 DDDD}\boxit{2cm}{\tt 00000}\ \dots\hfil}
3459 \line{\it addresses: \hfil}
3463 @c END TEXI2ROFF-KILL
3466 @section Assembler Internal Sections
3468 @cindex internal assembler sections
3469 @cindex sections in messages, internal
3470 These sections are meant only for the internal use of @command{@value{AS}}. They
3471 have no meaning at run-time. You do not really need to know about these
3472 sections for most purposes; but they can be mentioned in @command{@value{AS}}
3473 warning messages, so it might be helpful to have an idea of their
3474 meanings to @command{@value{AS}}. These sections are used to permit the
3475 value of every expression in your assembly language program to be a
3476 section-relative address.
3479 @cindex assembler internal logic error
3480 @item ASSEMBLER-INTERNAL-LOGIC-ERROR!
3481 An internal assembler logic error has been found. This means there is a
3482 bug in the assembler.
3484 @cindex expr (internal section)
3486 The assembler stores complex expression internally as combinations of
3487 symbols. When it needs to represent an expression as a symbol, it puts
3488 it in the expr section.
3490 @c FIXME item transfer[t] vector preload
3491 @c FIXME item transfer[t] vector postload
3492 @c FIXME item register
3496 @section Sub-Sections
3498 @cindex numbered subsections
3499 @cindex grouping data
3505 fall into two sections: text and data.
3507 You may have separate groups of
3509 data in named sections
3513 data in named sections
3519 that you want to end up near to each other in the object file, even though they
3520 are not contiguous in the assembler source. @command{@value{AS}} allows you to
3521 use @dfn{subsections} for this purpose. Within each section, there can be
3522 numbered subsections with values from 0 to 8192. Objects assembled into the
3523 same subsection go into the object file together with other objects in the same
3524 subsection. For example, a compiler might want to store constants in the text
3525 section, but might not want to have them interspersed with the program being
3526 assembled. In this case, the compiler could issue a @samp{.text 0} before each
3527 section of code being output, and a @samp{.text 1} before each group of
3528 constants being output.
3530 Subsections are optional. If you do not use subsections, everything
3531 goes in subsection number zero.
3534 Each subsection is zero-padded up to a multiple of four bytes.
3535 (Subsections may be padded a different amount on different flavors
3536 of @command{@value{AS}}.)
3540 On the H8/300 platform, each subsection is zero-padded to a word
3541 boundary (two bytes).
3542 The same is true on the Renesas SH.
3546 Subsections appear in your object file in numeric order, lowest numbered
3547 to highest. (All this to be compatible with other people's assemblers.)
3548 The object file contains no representation of subsections; @code{@value{LD}} and
3549 other programs that manipulate object files see no trace of them.
3550 They just see all your text subsections as a text section, and all your
3551 data subsections as a data section.
3553 To specify which subsection you want subsequent statements assembled
3554 into, use a numeric argument to specify it, in a @samp{.text
3555 @var{expression}} or a @samp{.data @var{expression}} statement.
3558 When generating COFF output, you
3563 can also use an extra subsection
3564 argument with arbitrary named sections: @samp{.section @var{name},
3569 When generating ELF output, you
3574 can also use the @code{.subsection} directive (@pxref{SubSection})
3575 to specify a subsection: @samp{.subsection @var{expression}}.
3577 @var{Expression} should be an absolute expression
3578 (@pxref{Expressions}). If you just say @samp{.text} then @samp{.text 0}
3579 is assumed. Likewise @samp{.data} means @samp{.data 0}. Assembly
3580 begins in @code{text 0}. For instance:
3582 .text 0 # The default subsection is text 0 anyway.
3583 .ascii "This lives in the first text subsection. *"
3585 .ascii "But this lives in the second text subsection."
3587 .ascii "This lives in the data section,"
3588 .ascii "in the first data subsection."
3590 .ascii "This lives in the first text section,"
3591 .ascii "immediately following the asterisk (*)."
3594 Each section has a @dfn{location counter} incremented by one for every byte
3595 assembled into that section. Because subsections are merely a convenience
3596 restricted to @command{@value{AS}} there is no concept of a subsection location
3597 counter. There is no way to directly manipulate a location counter---but the
3598 @code{.align} directive changes it, and any label definition captures its
3599 current value. The location counter of the section where statements are being
3600 assembled is said to be the @dfn{active} location counter.
3603 @section bss Section
3606 @cindex common variable storage
3607 The bss section is used for local common variable storage.
3608 You may allocate address space in the bss section, but you may
3609 not dictate data to load into it before your program executes. When
3610 your program starts running, all the contents of the bss
3611 section are zeroed bytes.
3613 The @code{.lcomm} pseudo-op defines a symbol in the bss section; see
3614 @ref{Lcomm,,@code{.lcomm}}.
3616 The @code{.comm} pseudo-op may be used to declare a common symbol, which is
3617 another form of uninitialized symbol; see @ref{Comm,,@code{.comm}}.
3620 When assembling for a target which supports multiple sections, such as ELF or
3621 COFF, you may switch into the @code{.bss} section and define symbols as usual;
3622 see @ref{Section,,@code{.section}}. You may only assemble zero values into the
3623 section. Typically the section will only contain symbol definitions and
3624 @code{.skip} directives (@pxref{Skip,,@code{.skip}}).
3631 Symbols are a central concept: the programmer uses symbols to name
3632 things, the linker uses symbols to link, and the debugger uses symbols
3636 @cindex debuggers, and symbol order
3637 @emph{Warning:} @command{@value{AS}} does not place symbols in the object file in
3638 the same order they were declared. This may break some debuggers.
3643 * Setting Symbols:: Giving Symbols Other Values
3644 * Symbol Names:: Symbol Names
3645 * Dot:: The Special Dot Symbol
3646 * Symbol Attributes:: Symbol Attributes
3653 A @dfn{label} is written as a symbol immediately followed by a colon
3654 @samp{:}. The symbol then represents the current value of the
3655 active location counter, and is, for example, a suitable instruction
3656 operand. You are warned if you use the same symbol to represent two
3657 different locations: the first definition overrides any other
3661 On the HPPA, the usual form for a label need not be immediately followed by a
3662 colon, but instead must start in column zero. Only one label may be defined on
3663 a single line. To work around this, the HPPA version of @command{@value{AS}} also
3664 provides a special directive @code{.label} for defining labels more flexibly.
3667 @node Setting Symbols
3668 @section Giving Symbols Other Values
3670 @cindex assigning values to symbols
3671 @cindex symbol values, assigning
3672 A symbol can be given an arbitrary value by writing a symbol, followed
3673 by an equals sign @samp{=}, followed by an expression
3674 (@pxref{Expressions}). This is equivalent to using the @code{.set}
3675 directive. @xref{Set,,@code{.set}}. In the same way, using a double
3676 equals sign @samp{=}@samp{=} here represents an equivalent of the
3677 @code{.eqv} directive. @xref{Eqv,,@code{.eqv}}.
3680 Blackfin does not support symbol assignment with @samp{=}.
3684 @section Symbol Names
3686 @cindex symbol names
3687 @cindex names, symbol
3688 @ifclear SPECIAL-SYMS
3689 Symbol names begin with a letter or with one of @samp{._}. On most
3690 machines, you can also use @code{$} in symbol names; exceptions are
3691 noted in @ref{Machine Dependencies}. That character may be followed by any
3692 string of digits, letters, dollar signs (unless otherwise noted for a
3693 particular target machine), and underscores.
3697 Symbol names begin with a letter or with one of @samp{._}. On the
3698 Renesas SH you can also use @code{$} in symbol names. That
3699 character may be followed by any string of digits, letters, dollar signs (save
3700 on the H8/300), and underscores.
3704 Case of letters is significant: @code{foo} is a different symbol name
3707 Symbol names do not start with a digit. An exception to this rule is made for
3708 Local Labels. See below.
3710 Multibyte characters are supported. To generate a symbol name containing
3711 multibyte characters enclose it within double quotes and use escape codes. cf
3712 @xref{Strings}. Generating a multibyte symbol name from a label is not
3713 currently supported.
3715 Each symbol has exactly one name. Each name in an assembly language program
3716 refers to exactly one symbol. You may use that symbol name any number of times
3719 @subheading Local Symbol Names
3721 @cindex local symbol names
3722 @cindex symbol names, local
3723 A local symbol is any symbol beginning with certain local label prefixes.
3724 By default, the local label prefix is @samp{.L} for ELF systems or
3725 @samp{L} for traditional a.out systems, but each target may have its own
3726 set of local label prefixes.
3728 On the HPPA local symbols begin with @samp{L$}.
3731 Local symbols are defined and used within the assembler, but they are
3732 normally not saved in object files. Thus, they are not visible when debugging.
3733 You may use the @samp{-L} option (@pxref{L, ,Include Local Symbols})
3734 to retain the local symbols in the object files.
3736 @subheading Local Labels
3738 @cindex local labels
3739 @cindex temporary symbol names
3740 @cindex symbol names, temporary
3741 Local labels are different from local symbols. Local labels help compilers and
3742 programmers use names temporarily. They create symbols which are guaranteed to
3743 be unique over the entire scope of the input source code and which can be
3744 referred to by a simple notation. To define a local label, write a label of
3745 the form @samp{@b{N}:} (where @b{N} represents any non-negative integer).
3746 To refer to the most recent previous definition of that label write
3747 @samp{@b{N}b}, using the same number as when you defined the label. To refer
3748 to the next definition of a local label, write @samp{@b{N}f}. The @samp{b}
3749 stands for ``backwards'' and the @samp{f} stands for ``forwards''.
3751 There is no restriction on how you can use these labels, and you can reuse them
3752 too. So that it is possible to repeatedly define the same local label (using
3753 the same number @samp{@b{N}}), although you can only refer to the most recently
3754 defined local label of that number (for a backwards reference) or the next
3755 definition of a specific local label for a forward reference. It is also worth
3756 noting that the first 10 local labels (@samp{@b{0:}}@dots{}@samp{@b{9:}}) are
3757 implemented in a slightly more efficient manner than the others.
3768 Which is the equivalent of:
3771 label_1: branch label_3
3772 label_2: branch label_1
3773 label_3: branch label_4
3774 label_4: branch label_3
3777 Local label names are only a notational device. They are immediately
3778 transformed into more conventional symbol names before the assembler uses them.
3779 The symbol names are stored in the symbol table, appear in error messages, and
3780 are optionally emitted to the object file. The names are constructed using
3784 @item @emph{local label prefix}
3785 All local symbols begin with the system-specific local label prefix.
3786 Normally both @command{@value{AS}} and @code{@value{LD}} forget symbols
3787 that start with the local label prefix. These labels are
3788 used for symbols you are never intended to see. If you use the
3789 @samp{-L} option then @command{@value{AS}} retains these symbols in the
3790 object file. If you also instruct @code{@value{LD}} to retain these symbols,
3791 you may use them in debugging.
3794 This is the number that was used in the local label definition. So if the
3795 label is written @samp{55:} then the number is @samp{55}.
3798 This unusual character is included so you do not accidentally invent a symbol
3799 of the same name. The character has ASCII value of @samp{\002} (control-B).
3801 @item @emph{ordinal number}
3802 This is a serial number to keep the labels distinct. The first definition of
3803 @samp{0:} gets the number @samp{1}. The 15th definition of @samp{0:} gets the
3804 number @samp{15}, and so on. Likewise the first definition of @samp{1:} gets
3805 the number @samp{1} and its 15th definition gets @samp{15} as well.
3808 So for example, the first @code{1:} may be named @code{.L1@kbd{C-B}1}, and
3809 the 44th @code{3:} may be named @code{.L3@kbd{C-B}44}.
3811 @subheading Dollar Local Labels
3812 @cindex dollar local symbols
3814 On some targets @code{@value{AS}} also supports an even more local form of
3815 local labels called dollar labels. These labels go out of scope (i.e., they
3816 become undefined) as soon as a non-local label is defined. Thus they remain
3817 valid for only a small region of the input source code. Normal local labels,
3818 by contrast, remain in scope for the entire file, or until they are redefined
3819 by another occurrence of the same local label.
3821 Dollar labels are defined in exactly the same way as ordinary local labels,
3822 except that they have a dollar sign suffix to their numeric value, e.g.,
3825 They can also be distinguished from ordinary local labels by their transformed
3826 names which use ASCII character @samp{\001} (control-A) as the magic character
3827 to distinguish them from ordinary labels. For example, the fifth definition of
3828 @samp{6$} may be named @samp{.L6@kbd{C-A}5}.
3831 @section The Special Dot Symbol
3833 @cindex dot (symbol)
3834 @cindex @code{.} (symbol)
3835 @cindex current address
3836 @cindex location counter
3837 The special symbol @samp{.} refers to the current address that
3838 @command{@value{AS}} is assembling into. Thus, the expression @samp{melvin:
3839 .long .} defines @code{melvin} to contain its own address.
3840 Assigning a value to @code{.} is treated the same as a @code{.org}
3842 @ifclear no-space-dir
3843 Thus, the expression @samp{.=.+4} is the same as saying
3847 @node Symbol Attributes
3848 @section Symbol Attributes
3850 @cindex symbol attributes
3851 @cindex attributes, symbol
3852 Every symbol has, as well as its name, the attributes ``Value'' and
3853 ``Type''. Depending on output format, symbols can also have auxiliary
3856 The detailed definitions are in @file{a.out.h}.
3859 If you use a symbol without defining it, @command{@value{AS}} assumes zero for
3860 all these attributes, and probably won't warn you. This makes the
3861 symbol an externally defined symbol, which is generally what you
3865 * Symbol Value:: Value
3866 * Symbol Type:: Type
3868 * a.out Symbols:: Symbol Attributes: @code{a.out}
3871 * COFF Symbols:: Symbol Attributes for COFF
3874 * SOM Symbols:: Symbol Attributes for SOM
3881 @cindex value of a symbol
3882 @cindex symbol value
3883 The value of a symbol is (usually) 32 bits. For a symbol which labels a
3884 location in the text, data, bss or absolute sections the value is the
3885 number of addresses from the start of that section to the label.
3886 Naturally for text, data and bss sections the value of a symbol changes
3887 as @code{@value{LD}} changes section base addresses during linking. Absolute
3888 symbols' values do not change during linking: that is why they are
3891 The value of an undefined symbol is treated in a special way. If it is
3892 0 then the symbol is not defined in this assembler source file, and
3893 @code{@value{LD}} tries to determine its value from other files linked into the
3894 same program. You make this kind of symbol simply by mentioning a symbol
3895 name without defining it. A non-zero value represents a @code{.comm}
3896 common declaration. The value is how much common storage to reserve, in
3897 bytes (addresses). The symbol refers to the first address of the
3903 @cindex type of a symbol
3905 The type attribute of a symbol contains relocation (section)
3906 information, any flag settings indicating that a symbol is external, and
3907 (optionally), other information for linkers and debuggers. The exact
3908 format depends on the object-code output format in use.
3912 @subsection Symbol Attributes: @code{a.out}
3914 @cindex @code{a.out} symbol attributes
3915 @cindex symbol attributes, @code{a.out}
3918 * Symbol Desc:: Descriptor
3919 * Symbol Other:: Other
3923 @subsubsection Descriptor
3925 @cindex descriptor, of @code{a.out} symbol
3926 This is an arbitrary 16-bit value. You may establish a symbol's
3927 descriptor value by using a @code{.desc} statement
3928 (@pxref{Desc,,@code{.desc}}). A descriptor value means nothing to
3929 @command{@value{AS}}.
3932 @subsubsection Other
3934 @cindex other attribute, of @code{a.out} symbol
3935 This is an arbitrary 8-bit value. It means nothing to @command{@value{AS}}.
3940 @subsection Symbol Attributes for COFF
3942 @cindex COFF symbol attributes
3943 @cindex symbol attributes, COFF
3945 The COFF format supports a multitude of auxiliary symbol attributes;
3946 like the primary symbol attributes, they are set between @code{.def} and
3947 @code{.endef} directives.
3949 @subsubsection Primary Attributes
3951 @cindex primary attributes, COFF symbols
3952 The symbol name is set with @code{.def}; the value and type,
3953 respectively, with @code{.val} and @code{.type}.
3955 @subsubsection Auxiliary Attributes
3957 @cindex auxiliary attributes, COFF symbols
3958 The @command{@value{AS}} directives @code{.dim}, @code{.line}, @code{.scl},
3959 @code{.size}, @code{.tag}, and @code{.weak} can generate auxiliary symbol
3960 table information for COFF.
3965 @subsection Symbol Attributes for SOM
3967 @cindex SOM symbol attributes
3968 @cindex symbol attributes, SOM
3970 The SOM format for the HPPA supports a multitude of symbol attributes set with
3971 the @code{.EXPORT} and @code{.IMPORT} directives.
3973 The attributes are described in @cite{HP9000 Series 800 Assembly
3974 Language Reference Manual} (HP 92432-90001) under the @code{IMPORT} and
3975 @code{EXPORT} assembler directive documentation.
3979 @chapter Expressions
3983 @cindex numeric values
3984 An @dfn{expression} specifies an address or numeric value.
3985 Whitespace may precede and/or follow an expression.
3987 The result of an expression must be an absolute number, or else an offset into
3988 a particular section. If an expression is not absolute, and there is not
3989 enough information when @command{@value{AS}} sees the expression to know its
3990 section, a second pass over the source program might be necessary to interpret
3991 the expression---but the second pass is currently not implemented.
3992 @command{@value{AS}} aborts with an error message in this situation.
3995 * Empty Exprs:: Empty Expressions
3996 * Integer Exprs:: Integer Expressions
4000 @section Empty Expressions
4002 @cindex empty expressions
4003 @cindex expressions, empty
4004 An empty expression has no value: it is just whitespace or null.
4005 Wherever an absolute expression is required, you may omit the
4006 expression, and @command{@value{AS}} assumes a value of (absolute) 0. This
4007 is compatible with other assemblers.
4010 @section Integer Expressions
4012 @cindex integer expressions
4013 @cindex expressions, integer
4014 An @dfn{integer expression} is one or more @emph{arguments} delimited
4015 by @emph{operators}.
4018 * Arguments:: Arguments
4019 * Operators:: Operators
4020 * Prefix Ops:: Prefix Operators
4021 * Infix Ops:: Infix Operators
4025 @subsection Arguments
4027 @cindex expression arguments
4028 @cindex arguments in expressions
4029 @cindex operands in expressions
4030 @cindex arithmetic operands
4031 @dfn{Arguments} are symbols, numbers or subexpressions. In other
4032 contexts arguments are sometimes called ``arithmetic operands''. In
4033 this manual, to avoid confusing them with the ``instruction operands'' of
4034 the machine language, we use the term ``argument'' to refer to parts of
4035 expressions only, reserving the word ``operand'' to refer only to machine
4036 instruction operands.
4038 Symbols are evaluated to yield @{@var{section} @var{NNN}@} where
4039 @var{section} is one of text, data, bss, absolute,
4040 or undefined. @var{NNN} is a signed, 2's complement 32 bit
4043 Numbers are usually integers.
4045 A number can be a flonum or bignum. In this case, you are warned
4046 that only the low order 32 bits are used, and @command{@value{AS}} pretends
4047 these 32 bits are an integer. You may write integer-manipulating
4048 instructions that act on exotic constants, compatible with other
4051 @cindex subexpressions
4052 Subexpressions are a left parenthesis @samp{(} followed by an integer
4053 expression, followed by a right parenthesis @samp{)}; or a prefix
4054 operator followed by an argument.
4057 @subsection Operators
4059 @cindex operators, in expressions
4060 @cindex arithmetic functions
4061 @cindex functions, in expressions
4062 @dfn{Operators} are arithmetic functions, like @code{+} or @code{%}. Prefix
4063 operators are followed by an argument. Infix operators appear
4064 between their arguments. Operators may be preceded and/or followed by
4068 @subsection Prefix Operator
4070 @cindex prefix operators
4071 @command{@value{AS}} has the following @dfn{prefix operators}. They each take
4072 one argument, which must be absolute.
4074 @c the tex/end tex stuff surrounding this small table is meant to make
4075 @c it align, on the printed page, with the similar table in the next
4076 @c section (which is inside an enumerate).
4078 \global\advance\leftskip by \itemindent
4083 @dfn{Negation}. Two's complement negation.
4085 @dfn{Complementation}. Bitwise not.
4089 \global\advance\leftskip by -\itemindent
4093 @subsection Infix Operators
4095 @cindex infix operators
4096 @cindex operators, permitted arguments
4097 @dfn{Infix operators} take two arguments, one on either side. Operators
4098 have precedence, but operations with equal precedence are performed left
4099 to right. Apart from @code{+} or @option{-}, both arguments must be
4100 absolute, and the result is absolute.
4103 @cindex operator precedence
4104 @cindex precedence of operators
4111 @dfn{Multiplication}.
4114 @dfn{Division}. Truncation is the same as the C operator @samp{/}
4120 @dfn{Shift Left}. Same as the C operator @samp{<<}.
4123 @dfn{Shift Right}. Same as the C operator @samp{>>}.
4127 Intermediate precedence
4132 @dfn{Bitwise Inclusive Or}.
4138 @dfn{Bitwise Exclusive Or}.
4141 @dfn{Bitwise Or Not}.
4148 @cindex addition, permitted arguments
4149 @cindex plus, permitted arguments
4150 @cindex arguments for addition
4152 @dfn{Addition}. If either argument is absolute, the result has the section of
4153 the other argument. You may not add together arguments from different
4156 @cindex subtraction, permitted arguments
4157 @cindex minus, permitted arguments
4158 @cindex arguments for subtraction
4160 @dfn{Subtraction}. If the right argument is absolute, the
4161 result has the section of the left argument.
4162 If both arguments are in the same section, the result is absolute.
4163 You may not subtract arguments from different sections.
4164 @c FIXME is there still something useful to say about undefined - undefined ?
4166 @cindex comparison expressions
4167 @cindex expressions, comparison
4172 @dfn{Is Not Equal To}
4176 @dfn{Is Greater Than}
4178 @dfn{Is Greater Than Or Equal To}
4180 @dfn{Is Less Than Or Equal To}
4182 The comparison operators can be used as infix operators. A true results has a
4183 value of -1 whereas a false result has a value of 0. Note, these operators
4184 perform signed comparisons.
4187 @item Lowest Precedence
4196 These two logical operations can be used to combine the results of sub
4197 expressions. Note, unlike the comparison operators a true result returns a
4198 value of 1 but a false results does still return 0. Also note that the logical
4199 or operator has a slightly lower precedence than logical and.
4204 In short, it's only meaningful to add or subtract the @emph{offsets} in an
4205 address; you can only have a defined section in one of the two arguments.
4208 @chapter Assembler Directives
4210 @cindex directives, machine independent
4211 @cindex pseudo-ops, machine independent
4212 @cindex machine independent directives
4213 All assembler directives have names that begin with a period (@samp{.}).
4214 The names are case insensitive for most targets, and usually written
4217 This chapter discusses directives that are available regardless of the
4218 target machine configuration for the @sc{gnu} assembler.
4220 Some machine configurations provide additional directives.
4221 @xref{Machine Dependencies}.
4224 @ifset machine-directives
4225 @xref{Machine Dependencies}, for additional directives.
4230 * Abort:: @code{.abort}
4232 * ABORT (COFF):: @code{.ABORT}
4235 * Align:: @code{.align @var{abs-expr} , @var{abs-expr}}
4236 * Altmacro:: @code{.altmacro}
4237 * Ascii:: @code{.ascii "@var{string}"}@dots{}
4238 * Asciz:: @code{.asciz "@var{string}"}@dots{}
4239 * Balign:: @code{.balign @var{abs-expr} , @var{abs-expr}}
4240 * Bundle directives:: @code{.bundle_align_mode @var{abs-expr}}, etc
4241 * Byte:: @code{.byte @var{expressions}}
4242 * CFI directives:: @code{.cfi_startproc [simple]}, @code{.cfi_endproc}, etc.
4243 * Comm:: @code{.comm @var{symbol} , @var{length} }
4244 * Data:: @code{.data @var{subsection}}
4245 * Dc:: @code{.dc[@var{size}] @var{expressions}}
4246 * Dcb:: @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4247 * Ds:: @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4249 * Def:: @code{.def @var{name}}
4252 * Desc:: @code{.desc @var{symbol}, @var{abs-expression}}
4258 * Double:: @code{.double @var{flonums}}
4259 * Eject:: @code{.eject}
4260 * Else:: @code{.else}
4261 * Elseif:: @code{.elseif}
4264 * Endef:: @code{.endef}
4267 * Endfunc:: @code{.endfunc}
4268 * Endif:: @code{.endif}
4269 * Equ:: @code{.equ @var{symbol}, @var{expression}}
4270 * Equiv:: @code{.equiv @var{symbol}, @var{expression}}
4271 * Eqv:: @code{.eqv @var{symbol}, @var{expression}}
4273 * Error:: @code{.error @var{string}}
4274 * Exitm:: @code{.exitm}
4275 * Extern:: @code{.extern}
4276 * Fail:: @code{.fail}
4277 * File:: @code{.file}
4278 * Fill:: @code{.fill @var{repeat} , @var{size} , @var{value}}
4279 * Float:: @code{.float @var{flonums}}
4280 * Func:: @code{.func}
4281 * Global:: @code{.global @var{symbol}}, @code{.globl @var{symbol}}
4283 * Gnu_attribute:: @code{.gnu_attribute @var{tag},@var{value}}
4284 * Hidden:: @code{.hidden @var{names}}
4287 * hword:: @code{.hword @var{expressions}}
4288 * Ident:: @code{.ident}
4289 * If:: @code{.if @var{absolute expression}}
4290 * Incbin:: @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
4291 * Include:: @code{.include "@var{file}"}
4292 * Int:: @code{.int @var{expressions}}
4294 * Internal:: @code{.internal @var{names}}
4297 * Irp:: @code{.irp @var{symbol},@var{values}}@dots{}
4298 * Irpc:: @code{.irpc @var{symbol},@var{values}}@dots{}
4299 * Lcomm:: @code{.lcomm @var{symbol} , @var{length}}
4300 * Lflags:: @code{.lflags}
4301 @ifclear no-line-dir
4302 * Line:: @code{.line @var{line-number}}
4305 * Linkonce:: @code{.linkonce [@var{type}]}
4306 * List:: @code{.list}
4307 * Ln:: @code{.ln @var{line-number}}
4308 * Loc:: @code{.loc @var{fileno} @var{lineno}}
4309 * Loc_mark_labels:: @code{.loc_mark_labels @var{enable}}
4311 * Local:: @code{.local @var{names}}
4314 * Long:: @code{.long @var{expressions}}
4316 * Lsym:: @code{.lsym @var{symbol}, @var{expression}}
4319 * Macro:: @code{.macro @var{name} @var{args}}@dots{}
4320 * MRI:: @code{.mri @var{val}}
4321 * Noaltmacro:: @code{.noaltmacro}
4322 * Nolist:: @code{.nolist}
4323 * Nops:: @code{.nops @var{size}[, @var{control}]}
4324 * Octa:: @code{.octa @var{bignums}}
4325 * Offset:: @code{.offset @var{loc}}
4326 * Org:: @code{.org @var{new-lc}, @var{fill}}
4327 * P2align:: @code{.p2align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4329 * PopSection:: @code{.popsection}
4330 * Previous:: @code{.previous}
4333 * Print:: @code{.print @var{string}}
4335 * Protected:: @code{.protected @var{names}}
4338 * Psize:: @code{.psize @var{lines}, @var{columns}}
4339 * Purgem:: @code{.purgem @var{name}}
4341 * PushSection:: @code{.pushsection @var{name}}
4344 * Quad:: @code{.quad @var{bignums}}
4345 * Reloc:: @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
4346 * Rept:: @code{.rept @var{count}}
4347 * Sbttl:: @code{.sbttl "@var{subheading}"}
4349 * Scl:: @code{.scl @var{class}}
4352 * Section:: @code{.section @var{name}[, @var{flags}]}
4355 * Set:: @code{.set @var{symbol}, @var{expression}}
4356 * Short:: @code{.short @var{expressions}}
4357 * Single:: @code{.single @var{flonums}}
4359 * Size:: @code{.size [@var{name} , @var{expression}]}
4361 @ifclear no-space-dir
4362 * Skip:: @code{.skip @var{size} [,@var{fill}]}
4365 * Sleb128:: @code{.sleb128 @var{expressions}}
4366 @ifclear no-space-dir
4367 * Space:: @code{.space @var{size} [,@var{fill}]}
4370 * Stab:: @code{.stabd, .stabn, .stabs}
4373 * String:: @code{.string "@var{str}"}, @code{.string8 "@var{str}"}, @code{.string16 "@var{str}"}, @code{.string32 "@var{str}"}, @code{.string64 "@var{str}"}
4374 * Struct:: @code{.struct @var{expression}}
4376 * SubSection:: @code{.subsection}
4377 * Symver:: @code{.symver @var{name},@var{name2@@nodename}}
4381 * Tag:: @code{.tag @var{structname}}
4384 * Text:: @code{.text @var{subsection}}
4385 * Title:: @code{.title "@var{heading}"}
4387 * Type:: @code{.type <@var{int} | @var{name} , @var{type description}>}
4390 * Uleb128:: @code{.uleb128 @var{expressions}}
4392 * Val:: @code{.val @var{addr}}
4396 * Version:: @code{.version "@var{string}"}
4397 * VTableEntry:: @code{.vtable_entry @var{table}, @var{offset}}
4398 * VTableInherit:: @code{.vtable_inherit @var{child}, @var{parent}}
4401 * Warning:: @code{.warning @var{string}}
4402 * Weak:: @code{.weak @var{names}}
4403 * Weakref:: @code{.weakref @var{alias}, @var{symbol}}
4404 * Word:: @code{.word @var{expressions}}
4405 @ifclear no-space-dir
4406 * Zero:: @code{.zero @var{size}}
4409 * 2byte:: @code{.2byte @var{expressions}}
4410 * 4byte:: @code{.4byte @var{expressions}}
4411 * 8byte:: @code{.8byte @var{bignums}}
4413 * Deprecated:: Deprecated Directives
4417 @section @code{.abort}
4419 @cindex @code{abort} directive
4420 @cindex stopping the assembly
4421 This directive stops the assembly immediately. It is for
4422 compatibility with other assemblers. The original idea was that the
4423 assembly language source would be piped into the assembler. If the sender
4424 of the source quit, it could use this directive tells @command{@value{AS}} to
4425 quit also. One day @code{.abort} will not be supported.
4429 @section @code{.ABORT} (COFF)
4431 @cindex @code{ABORT} directive
4432 When producing COFF output, @command{@value{AS}} accepts this directive as a
4433 synonym for @samp{.abort}.
4438 @section @code{.align @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4440 @cindex padding the location counter
4441 @cindex @code{align} directive
4442 Pad the location counter (in the current subsection) to a particular storage
4443 boundary. The first expression (which must be absolute) is the alignment
4444 required, as described below.
4446 The second expression (also absolute) gives the fill value to be stored in the
4447 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4448 padding bytes are normally zero. However, on most systems, if the section is
4449 marked as containing code and the fill value is omitted, the space is filled
4450 with no-op instructions.
4452 The third expression is also absolute, and is also optional. If it is present,
4453 it is the maximum number of bytes that should be skipped by this alignment
4454 directive. If doing the alignment would require skipping more bytes than the
4455 specified maximum, then the alignment is not done at all. You can omit the
4456 fill value (the second argument) entirely by simply using two commas after the
4457 required alignment; this can be useful if you want the alignment to be filled
4458 with no-op instructions when appropriate.
4460 The way the required alignment is specified varies from system to system.
4461 For the arc, hppa, i386 using ELF, iq2000, m68k, or1k,
4462 s390, sparc, tic4x, tic80 and xtensa, the first expression is the
4463 alignment request in bytes. For example @samp{.align 8} advances
4464 the location counter until it is a multiple of 8. If the location counter
4465 is already a multiple of 8, no change is needed. For the tic54x, the
4466 first expression is the alignment request in words.
4468 For other systems, including ppc, i386 using a.out format, arm and
4469 strongarm, it is the
4470 number of low-order zero bits the location counter must have after
4471 advancement. For example @samp{.align 3} advances the location
4472 counter until it a multiple of 8. If the location counter is already a
4473 multiple of 8, no change is needed.
4475 This inconsistency is due to the different behaviors of the various
4476 native assemblers for these systems which GAS must emulate.
4477 GAS also provides @code{.balign} and @code{.p2align} directives,
4478 described later, which have a consistent behavior across all
4479 architectures (but are specific to GAS).
4482 @section @code{.altmacro}
4483 Enable alternate macro mode, enabling:
4486 @item LOCAL @var{name} [ , @dots{} ]
4487 One additional directive, @code{LOCAL}, is available. It is used to
4488 generate a string replacement for each of the @var{name} arguments, and
4489 replace any instances of @var{name} in each macro expansion. The
4490 replacement string is unique in the assembly, and different for each
4491 separate macro expansion. @code{LOCAL} allows you to write macros that
4492 define symbols, without fear of conflict between separate macro expansions.
4494 @item String delimiters
4495 You can write strings delimited in these other ways besides
4496 @code{"@var{string}"}:
4499 @item '@var{string}'
4500 You can delimit strings with single-quote characters.
4502 @item <@var{string}>
4503 You can delimit strings with matching angle brackets.
4506 @item single-character string escape
4507 To include any single character literally in a string (even if the
4508 character would otherwise have some special meaning), you can prefix the
4509 character with @samp{!} (an exclamation mark). For example, you can
4510 write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
4512 @item Expression results as strings
4513 You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
4514 and use the result as a string.
4518 @section @code{.ascii "@var{string}"}@dots{}
4520 @cindex @code{ascii} directive
4521 @cindex string literals
4522 @code{.ascii} expects zero or more string literals (@pxref{Strings})
4523 separated by commas. It assembles each string (with no automatic
4524 trailing zero byte) into consecutive addresses.
4527 @section @code{.asciz "@var{string}"}@dots{}
4529 @cindex @code{asciz} directive
4530 @cindex zero-terminated strings
4531 @cindex null-terminated strings
4532 @code{.asciz} is just like @code{.ascii}, but each string is followed by
4533 a zero byte. The ``z'' in @samp{.asciz} stands for ``zero''.
4536 @section @code{.balign[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
4538 @cindex padding the location counter given number of bytes
4539 @cindex @code{balign} directive
4540 Pad the location counter (in the current subsection) to a particular
4541 storage boundary. The first expression (which must be absolute) is the
4542 alignment request in bytes. For example @samp{.balign 8} advances
4543 the location counter until it is a multiple of 8. If the location counter
4544 is already a multiple of 8, no change is needed.
4546 The second expression (also absolute) gives the fill value to be stored in the
4547 padding bytes. It (and the comma) may be omitted. If it is omitted, the
4548 padding bytes are normally zero. However, on most systems, if the section is
4549 marked as containing code and the fill value is omitted, the space is filled
4550 with no-op instructions.
4552 The third expression is also absolute, and is also optional. If it is present,
4553 it is the maximum number of bytes that should be skipped by this alignment
4554 directive. If doing the alignment would require skipping more bytes than the
4555 specified maximum, then the alignment is not done at all. You can omit the
4556 fill value (the second argument) entirely by simply using two commas after the
4557 required alignment; this can be useful if you want the alignment to be filled
4558 with no-op instructions when appropriate.
4560 @cindex @code{balignw} directive
4561 @cindex @code{balignl} directive
4562 The @code{.balignw} and @code{.balignl} directives are variants of the
4563 @code{.balign} directive. The @code{.balignw} directive treats the fill
4564 pattern as a two byte word value. The @code{.balignl} directives treats the
4565 fill pattern as a four byte longword value. For example, @code{.balignw
4566 4,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
4567 filled in with the value 0x368d (the exact placement of the bytes depends upon
4568 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
4571 @node Bundle directives
4572 @section Bundle directives
4573 @subsection @code{.bundle_align_mode @var{abs-expr}}
4574 @cindex @code{bundle_align_mode} directive
4576 @cindex instruction bundle
4577 @cindex aligned instruction bundle
4578 @code{.bundle_align_mode} enables or disables @dfn{aligned instruction
4579 bundle} mode. In this mode, sequences of adjacent instructions are grouped
4580 into fixed-sized @dfn{bundles}. If the argument is zero, this mode is
4581 disabled (which is the default state). If the argument it not zero, it
4582 gives the size of an instruction bundle as a power of two (as for the
4583 @code{.p2align} directive, @pxref{P2align}).
4585 For some targets, it's an ABI requirement that no instruction may span a
4586 certain aligned boundary. A @dfn{bundle} is simply a sequence of
4587 instructions that starts on an aligned boundary. For example, if
4588 @var{abs-expr} is @code{5} then the bundle size is 32, so each aligned
4589 chunk of 32 bytes is a bundle. When aligned instruction bundle mode is in
4590 effect, no single instruction may span a boundary between bundles. If an
4591 instruction would start too close to the end of a bundle for the length of
4592 that particular instruction to fit within the bundle, then the space at the
4593 end of that bundle is filled with no-op instructions so the instruction
4594 starts in the next bundle. As a corollary, it's an error if any single
4595 instruction's encoding is longer than the bundle size.
4597 @subsection @code{.bundle_lock} and @code{.bundle_unlock}
4598 @cindex @code{bundle_lock} directive
4599 @cindex @code{bundle_unlock} directive
4600 The @code{.bundle_lock} and directive @code{.bundle_unlock} directives
4601 allow explicit control over instruction bundle padding. These directives
4602 are only valid when @code{.bundle_align_mode} has been used to enable
4603 aligned instruction bundle mode. It's an error if they appear when
4604 @code{.bundle_align_mode} has not been used at all, or when the last
4605 directive was @w{@code{.bundle_align_mode 0}}.
4607 @cindex bundle-locked
4608 For some targets, it's an ABI requirement that certain instructions may
4609 appear only as part of specified permissible sequences of multiple
4610 instructions, all within the same bundle. A pair of @code{.bundle_lock}
4611 and @code{.bundle_unlock} directives define a @dfn{bundle-locked}
4612 instruction sequence. For purposes of aligned instruction bundle mode, a
4613 sequence starting with @code{.bundle_lock} and ending with
4614 @code{.bundle_unlock} is treated as a single instruction. That is, the
4615 entire sequence must fit into a single bundle and may not span a bundle
4616 boundary. If necessary, no-op instructions will be inserted before the
4617 first instruction of the sequence so that the whole sequence starts on an
4618 aligned bundle boundary. It's an error if the sequence is longer than the
4621 For convenience when using @code{.bundle_lock} and @code{.bundle_unlock}
4622 inside assembler macros (@pxref{Macro}), bundle-locked sequences may be
4623 nested. That is, a second @code{.bundle_lock} directive before the next
4624 @code{.bundle_unlock} directive has no effect except that it must be
4625 matched by another closing @code{.bundle_unlock} so that there is the
4626 same number of @code{.bundle_lock} and @code{.bundle_unlock} directives.
4629 @section @code{.byte @var{expressions}}
4631 @cindex @code{byte} directive
4632 @cindex integers, one byte
4633 @code{.byte} expects zero or more expressions, separated by commas.
4634 Each expression is assembled into the next byte.
4636 @node CFI directives
4637 @section CFI directives
4638 @subsection @code{.cfi_sections @var{section_list}}
4639 @cindex @code{cfi_sections} directive
4640 @code{.cfi_sections} may be used to specify whether CFI directives
4641 should emit @code{.eh_frame} section and/or @code{.debug_frame} section.
4642 If @var{section_list} is @code{.eh_frame}, @code{.eh_frame} is emitted,
4643 if @var{section_list} is @code{.debug_frame}, @code{.debug_frame} is emitted.
4644 To emit both use @code{.eh_frame, .debug_frame}. The default if this
4645 directive is not used is @code{.cfi_sections .eh_frame}.
4647 On targets that support compact unwinding tables these can be generated
4648 by specifying @code{.eh_frame_entry} instead of @code{.eh_frame}.
4650 Some targets may support an additional name, such as @code{.c6xabi.exidx}
4651 which is used by the @value{TIC6X} target.
4653 The @code{.cfi_sections} directive can be repeated, with the same or different
4654 arguments, provided that CFI generation has not yet started. Once CFI
4655 generation has started however the section list is fixed and any attempts to
4656 redefine it will result in an error.
4658 @subsection @code{.cfi_startproc [simple]}
4659 @cindex @code{cfi_startproc} directive
4660 @code{.cfi_startproc} is used at the beginning of each function that
4661 should have an entry in @code{.eh_frame}. It initializes some internal
4662 data structures. Don't forget to close the function by
4663 @code{.cfi_endproc}.
4665 Unless @code{.cfi_startproc} is used along with parameter @code{simple}
4666 it also emits some architecture dependent initial CFI instructions.
4668 @subsection @code{.cfi_endproc}
4669 @cindex @code{cfi_endproc} directive
4670 @code{.cfi_endproc} is used at the end of a function where it closes its
4671 unwind entry previously opened by
4672 @code{.cfi_startproc}, and emits it to @code{.eh_frame}.
4674 @subsection @code{.cfi_personality @var{encoding} [, @var{exp}]}
4675 @cindex @code{cfi_personality} directive
4676 @code{.cfi_personality} defines personality routine and its encoding.
4677 @var{encoding} must be a constant determining how the personality
4678 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), second
4679 argument is not present, otherwise second argument should be
4680 a constant or a symbol name. When using indirect encodings,
4681 the symbol provided should be the location where personality
4682 can be loaded from, not the personality routine itself.
4683 The default after @code{.cfi_startproc} is @code{.cfi_personality 0xff},
4684 no personality routine.
4686 @subsection @code{.cfi_personality_id @var{id}}
4687 @cindex @code{cfi_personality_id} directive
4688 @code{cfi_personality_id} defines a personality routine by its index as
4689 defined in a compact unwinding format.
4690 Only valid when generating compact EH frames (i.e.
4691 with @code{.cfi_sections eh_frame_entry}.
4693 @subsection @code{.cfi_fde_data [@var{opcode1} [, @dots{}]]}
4694 @cindex @code{cfi_fde_data} directive
4695 @code{cfi_fde_data} is used to describe the compact unwind opcodes to be
4696 used for the current function. These are emitted inline in the
4697 @code{.eh_frame_entry} section if small enough and there is no LSDA, or
4698 in the @code{.gnu.extab} section otherwise.
4699 Only valid when generating compact EH frames (i.e.
4700 with @code{.cfi_sections eh_frame_entry}.
4702 @subsection @code{.cfi_lsda @var{encoding} [, @var{exp}]}
4703 @code{.cfi_lsda} defines LSDA and its encoding.
4704 @var{encoding} must be a constant determining how the LSDA
4705 should be encoded. If it is 255 (@code{DW_EH_PE_omit}), the second
4706 argument is not present, otherwise the second argument should be a constant
4707 or a symbol name. The default after @code{.cfi_startproc} is @code{.cfi_lsda 0xff},
4708 meaning that no LSDA is present.
4710 @subsection @code{.cfi_inline_lsda} [@var{align}]
4711 @code{.cfi_inline_lsda} marks the start of a LSDA data section and
4712 switches to the corresponding @code{.gnu.extab} section.
4713 Must be preceded by a CFI block containing a @code{.cfi_lsda} directive.
4714 Only valid when generating compact EH frames (i.e.
4715 with @code{.cfi_sections eh_frame_entry}.
4717 The table header and unwinding opcodes will be generated at this point,
4718 so that they are immediately followed by the LSDA data. The symbol
4719 referenced by the @code{.cfi_lsda} directive should still be defined
4720 in case a fallback FDE based encoding is used. The LSDA data is terminated
4721 by a section directive.
4723 The optional @var{align} argument specifies the alignment required.
4724 The alignment is specified as a power of two, as with the
4725 @code{.p2align} directive.
4727 @subsection @code{.cfi_def_cfa @var{register}, @var{offset}}
4728 @code{.cfi_def_cfa} defines a rule for computing CFA as: @i{take
4729 address from @var{register} and add @var{offset} to it}.
4731 @subsection @code{.cfi_def_cfa_register @var{register}}
4732 @code{.cfi_def_cfa_register} modifies a rule for computing CFA. From
4733 now on @var{register} will be used instead of the old one. Offset
4736 @subsection @code{.cfi_def_cfa_offset @var{offset}}
4737 @code{.cfi_def_cfa_offset} modifies a rule for computing CFA. Register
4738 remains the same, but @var{offset} is new. Note that it is the
4739 absolute offset that will be added to a defined register to compute
4742 @subsection @code{.cfi_adjust_cfa_offset @var{offset}}
4743 Same as @code{.cfi_def_cfa_offset} but @var{offset} is a relative
4744 value that is added/subtracted from the previous offset.
4746 @subsection @code{.cfi_offset @var{register}, @var{offset}}
4747 Previous value of @var{register} is saved at offset @var{offset} from
4750 @subsection @code{.cfi_val_offset @var{register}, @var{offset}}
4751 Previous value of @var{register} is CFA + @var{offset}.
4753 @subsection @code{.cfi_rel_offset @var{register}, @var{offset}}
4754 Previous value of @var{register} is saved at offset @var{offset} from
4755 the current CFA register. This is transformed to @code{.cfi_offset}
4756 using the known displacement of the CFA register from the CFA.
4757 This is often easier to use, because the number will match the
4758 code it's annotating.
4760 @subsection @code{.cfi_register @var{register1}, @var{register2}}
4761 Previous value of @var{register1} is saved in register @var{register2}.
4763 @subsection @code{.cfi_restore @var{register}}
4764 @code{.cfi_restore} says that the rule for @var{register} is now the
4765 same as it was at the beginning of the function, after all initial
4766 instruction added by @code{.cfi_startproc} were executed.
4768 @subsection @code{.cfi_undefined @var{register}}
4769 From now on the previous value of @var{register} can't be restored anymore.
4771 @subsection @code{.cfi_same_value @var{register}}
4772 Current value of @var{register} is the same like in the previous frame,
4773 i.e. no restoration needed.
4775 @subsection @code{.cfi_remember_state} and @code{.cfi_restore_state}
4776 @code{.cfi_remember_state} pushes the set of rules for every register onto an
4777 implicit stack, while @code{.cfi_restore_state} pops them off the stack and
4778 places them in the current row. This is useful for situations where you have
4779 multiple @code{.cfi_*} directives that need to be undone due to the control
4780 flow of the program. For example, we could have something like this (assuming
4781 the CFA is the value of @code{rbp}):
4791 .cfi_def_cfa %rsp, 8
4794 /* Do something else */
4797 Here, we want the @code{.cfi} directives to affect only the rows corresponding
4798 to the instructions before @code{label}. This means we'd have to add multiple
4799 @code{.cfi} directives after @code{label} to recreate the original save
4800 locations of the registers, as well as setting the CFA back to the value of
4801 @code{rbp}. This would be clumsy, and result in a larger binary size. Instead,
4813 .cfi_def_cfa %rsp, 8
4817 /* Do something else */
4820 That way, the rules for the instructions after @code{label} will be the same
4821 as before the first @code{.cfi_restore} without having to use multiple
4822 @code{.cfi} directives.
4824 @subsection @code{.cfi_return_column @var{register}}
4825 Change return column @var{register}, i.e. the return address is either
4826 directly in @var{register} or can be accessed by rules for @var{register}.
4828 @subsection @code{.cfi_signal_frame}
4829 Mark current function as signal trampoline.
4831 @subsection @code{.cfi_window_save}
4832 SPARC register window has been saved.
4834 @subsection @code{.cfi_escape} @var{expression}[, @dots{}]
4835 Allows the user to add arbitrary bytes to the unwind info. One
4836 might use this to add OS-specific CFI opcodes, or generic CFI
4837 opcodes that GAS does not yet support.
4839 @subsection @code{.cfi_val_encoded_addr @var{register}, @var{encoding}, @var{label}}
4840 The current value of @var{register} is @var{label}. The value of @var{label}
4841 will be encoded in the output file according to @var{encoding}; see the
4842 description of @code{.cfi_personality} for details on this encoding.
4844 The usefulness of equating a register to a fixed label is probably
4845 limited to the return address register. Here, it can be useful to
4846 mark a code segment that has only one return address which is reached
4847 by a direct branch and no copy of the return address exists in memory
4848 or another register.
4851 @section @code{.comm @var{symbol} , @var{length} }
4853 @cindex @code{comm} directive
4854 @cindex symbol, common
4855 @code{.comm} declares a common symbol named @var{symbol}. When linking, a
4856 common symbol in one object file may be merged with a defined or common symbol
4857 of the same name in another object file. If @code{@value{LD}} does not see a
4858 definition for the symbol--just one or more common symbols--then it will
4859 allocate @var{length} bytes of uninitialized memory. @var{length} must be an
4860 absolute expression. If @code{@value{LD}} sees multiple common symbols with
4861 the same name, and they do not all have the same size, it will allocate space
4862 using the largest size.
4865 When using ELF or (as a GNU extension) PE, the @code{.comm} directive takes
4866 an optional third argument. This is the desired alignment of the symbol,
4867 specified for ELF as a byte boundary (for example, an alignment of 16 means
4868 that the least significant 4 bits of the address should be zero), and for PE
4869 as a power of two (for example, an alignment of 5 means aligned to a 32-byte
4870 boundary). The alignment must be an absolute expression, and it must be a
4871 power of two. If @code{@value{LD}} allocates uninitialized memory for the
4872 common symbol, it will use the alignment when placing the symbol. If no
4873 alignment is specified, @command{@value{AS}} will set the alignment to the
4874 largest power of two less than or equal to the size of the symbol, up to a
4875 maximum of 16 on ELF, or the default section alignment of 4 on PE@footnote{This
4876 is not the same as the executable image file alignment controlled by @code{@value{LD}}'s
4877 @samp{--section-alignment} option; image file sections in PE are aligned to
4878 multiples of 4096, which is far too large an alignment for ordinary variables.
4879 It is rather the default alignment for (non-debug) sections within object
4880 (@samp{*.o}) files, which are less strictly aligned.}.
4884 The syntax for @code{.comm} differs slightly on the HPPA. The syntax is
4885 @samp{@var{symbol} .comm, @var{length}}; @var{symbol} is optional.
4889 @section @code{.data @var{subsection}}
4890 @cindex @code{data} directive
4892 @code{.data} tells @command{@value{AS}} to assemble the following statements onto the
4893 end of the data subsection numbered @var{subsection} (which is an
4894 absolute expression). If @var{subsection} is omitted, it defaults
4898 @section @code{.dc[@var{size}] @var{expressions}}
4899 @cindex @code{dc} directive
4901 The @code{.dc} directive expects zero or more @var{expressions} separated by
4902 commas. These expressions are evaluated and their values inserted into the
4903 current section. The size of the emitted value depends upon the suffix to the
4904 @code{.dc} directive:
4908 Emits N-bit values, where N is the size of an address on the target system.
4912 Emits double precision floating-point values.
4914 Emits 32-bit values.
4916 Emits single precision floating-point values.
4918 Emits 16-bit values.
4919 Note - this is true even on targets where the @code{.word} directive would emit
4922 Emits long double precision floating-point values.
4925 If no suffix is used then @samp{.w} is assumed.
4927 The byte ordering is target dependent, as is the size and format of floating
4931 @section @code{.dcb[@var{size}] @var{number} [,@var{fill}]}
4932 @cindex @code{dcb} directive
4933 This directive emits @var{number} copies of @var{fill}, each of @var{size}
4934 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
4935 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
4936 @var{size} suffix, if present, must be one of:
4940 Emits single byte values.
4942 Emits double-precision floating point values.
4944 Emits 4-byte values.
4946 Emits single-precision floating point values.
4948 Emits 2-byte values.
4950 Emits long double-precision floating point values.
4953 If the @var{size} suffix is omitted then @samp{.w} is assumed.
4955 The byte ordering is target dependent, as is the size and format of floating
4959 @section @code{.ds[@var{size}] @var{number} [,@var{fill}]}
4960 @cindex @code{ds} directive
4961 This directive emits @var{number} copies of @var{fill}, each of @var{size}
4962 bytes. Both @var{number} and @var{fill} are absolute expressions. If the
4963 comma and @var{fill} are omitted, @var{fill} is assumed to be zero. The
4964 @var{size} suffix, if present, must be one of:
4968 Emits single byte values.
4970 Emits 8-byte values.
4972 Emits 4-byte values.
4974 Emits 12-byte values.
4976 Emits 4-byte values.
4978 Emits 2-byte values.
4980 Emits 12-byte values.
4983 Note - unlike the @code{.dcb} directive the @samp{.d}, @samp{.s} and @samp{.x}
4984 suffixes do not indicate that floating-point values are to be inserted.
4986 If the @var{size} suffix is omitted then @samp{.w} is assumed.
4988 The byte ordering is target dependent.
4993 @section @code{.def @var{name}}
4995 @cindex @code{def} directive
4996 @cindex COFF symbols, debugging
4997 @cindex debugging COFF symbols
4998 Begin defining debugging information for a symbol @var{name}; the
4999 definition extends until the @code{.endef} directive is encountered.
5004 @section @code{.desc @var{symbol}, @var{abs-expression}}
5006 @cindex @code{desc} directive
5007 @cindex COFF symbol descriptor
5008 @cindex symbol descriptor, COFF
5009 This directive sets the descriptor of the symbol (@pxref{Symbol Attributes})
5010 to the low 16 bits of an absolute expression.
5013 The @samp{.desc} directive is not available when @command{@value{AS}} is
5014 configured for COFF output; it is only for @code{a.out} or @code{b.out}
5015 object format. For the sake of compatibility, @command{@value{AS}} accepts
5016 it, but produces no output, when configured for COFF.
5022 @section @code{.dim}
5024 @cindex @code{dim} directive
5025 @cindex COFF auxiliary symbol information
5026 @cindex auxiliary symbol information, COFF
5027 This directive is generated by compilers to include auxiliary debugging
5028 information in the symbol table. It is only permitted inside
5029 @code{.def}/@code{.endef} pairs.
5033 @section @code{.double @var{flonums}}
5035 @cindex @code{double} directive
5036 @cindex floating point numbers (double)
5037 @code{.double} expects zero or more flonums, separated by commas. It
5038 assembles floating point numbers.
5040 The exact kind of floating point numbers emitted depends on how
5041 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
5045 On the @value{TARGET} family @samp{.double} emits 64-bit floating-point numbers
5046 in @sc{ieee} format.
5051 @section @code{.eject}
5053 @cindex @code{eject} directive
5054 @cindex new page, in listings
5055 @cindex page, in listings
5056 @cindex listing control: new page
5057 Force a page break at this point, when generating assembly listings.
5060 @section @code{.else}
5062 @cindex @code{else} directive
5063 @code{.else} is part of the @command{@value{AS}} support for conditional
5064 assembly; see @ref{If,,@code{.if}}. It marks the beginning of a section
5065 of code to be assembled if the condition for the preceding @code{.if}
5069 @section @code{.elseif}
5071 @cindex @code{elseif} directive
5072 @code{.elseif} is part of the @command{@value{AS}} support for conditional
5073 assembly; see @ref{If,,@code{.if}}. It is shorthand for beginning a new
5074 @code{.if} block that would otherwise fill the entire @code{.else} section.
5077 @section @code{.end}
5079 @cindex @code{end} directive
5080 @code{.end} marks the end of the assembly file. @command{@value{AS}} does not
5081 process anything in the file past the @code{.end} directive.
5085 @section @code{.endef}
5087 @cindex @code{endef} directive
5088 This directive flags the end of a symbol definition begun with
5093 @section @code{.endfunc}
5094 @cindex @code{endfunc} directive
5095 @code{.endfunc} marks the end of a function specified with @code{.func}.
5098 @section @code{.endif}
5100 @cindex @code{endif} directive
5101 @code{.endif} is part of the @command{@value{AS}} support for conditional assembly;
5102 it marks the end of a block of code that is only assembled
5103 conditionally. @xref{If,,@code{.if}}.
5106 @section @code{.equ @var{symbol}, @var{expression}}
5108 @cindex @code{equ} directive
5109 @cindex assigning values to symbols
5110 @cindex symbols, assigning values to
5111 This directive sets the value of @var{symbol} to @var{expression}.
5112 It is synonymous with @samp{.set}; see @ref{Set,,@code{.set}}.
5115 The syntax for @code{equ} on the HPPA is
5116 @samp{@var{symbol} .equ @var{expression}}.
5120 The syntax for @code{equ} on the Z80 is
5121 @samp{@var{symbol} equ @var{expression}}.
5122 On the Z80 it is an error if @var{symbol} is already defined,
5123 but the symbol is not protected from later redefinition.
5124 Compare @ref{Equiv}.
5128 @section @code{.equiv @var{symbol}, @var{expression}}
5129 @cindex @code{equiv} directive
5130 The @code{.equiv} directive is like @code{.equ} and @code{.set}, except that
5131 the assembler will signal an error if @var{symbol} is already defined. Note a
5132 symbol which has been referenced but not actually defined is considered to be
5135 Except for the contents of the error message, this is roughly equivalent to
5142 plus it protects the symbol from later redefinition.
5145 @section @code{.eqv @var{symbol}, @var{expression}}
5146 @cindex @code{eqv} directive
5147 The @code{.eqv} directive is like @code{.equiv}, but no attempt is made to
5148 evaluate the expression or any part of it immediately. Instead each time
5149 the resulting symbol is used in an expression, a snapshot of its current
5153 @section @code{.err}
5154 @cindex @code{err} directive
5155 If @command{@value{AS}} assembles a @code{.err} directive, it will print an error
5156 message and, unless the @option{-Z} option was used, it will not generate an
5157 object file. This can be used to signal an error in conditionally compiled code.
5160 @section @code{.error "@var{string}"}
5161 @cindex error directive
5163 Similarly to @code{.err}, this directive emits an error, but you can specify a
5164 string that will be emitted as the error message. If you don't specify the
5165 message, it defaults to @code{".error directive invoked in source file"}.
5166 @xref{Errors, ,Error and Warning Messages}.
5169 .error "This code has not been assembled and tested."
5173 @section @code{.exitm}
5174 Exit early from the current macro definition. @xref{Macro}.
5177 @section @code{.extern}
5179 @cindex @code{extern} directive
5180 @code{.extern} is accepted in the source program---for compatibility
5181 with other assemblers---but it is ignored. @command{@value{AS}} treats
5182 all undefined symbols as external.
5185 @section @code{.fail @var{expression}}
5187 @cindex @code{fail} directive
5188 Generates an error or a warning. If the value of the @var{expression} is 500
5189 or more, @command{@value{AS}} will print a warning message. If the value is less
5190 than 500, @command{@value{AS}} will print an error message. The message will
5191 include the value of @var{expression}. This can occasionally be useful inside
5192 complex nested macros or conditional assembly.
5195 @section @code{.file}
5196 @cindex @code{file} directive
5198 @ifclear no-file-dir
5199 There are two different versions of the @code{.file} directive. Targets
5200 that support DWARF2 line number information use the DWARF2 version of
5201 @code{.file}. Other targets use the default version.
5203 @subheading Default Version
5205 @cindex logical file name
5206 @cindex file name, logical
5207 This version of the @code{.file} directive tells @command{@value{AS}} that we
5208 are about to start a new logical file. The syntax is:
5214 @var{string} is the new file name. In general, the filename is
5215 recognized whether or not it is surrounded by quotes @samp{"}; but if you wish
5216 to specify an empty file name, you must give the quotes--@code{""}. This
5217 statement may go away in future: it is only recognized to be compatible with
5218 old @command{@value{AS}} programs.
5220 @subheading DWARF2 Version
5223 When emitting DWARF2 line number information, @code{.file} assigns filenames
5224 to the @code{.debug_line} file name table. The syntax is:
5227 .file @var{fileno} @var{filename}
5230 The @var{fileno} operand should be a unique positive integer to use as the
5231 index of the entry in the table. The @var{filename} operand is a C string
5234 The detail of filename indices is exposed to the user because the filename
5235 table is shared with the @code{.debug_info} section of the DWARF2 debugging
5236 information, and thus the user must know the exact indices that table
5240 @section @code{.fill @var{repeat} , @var{size} , @var{value}}
5242 @cindex @code{fill} directive
5243 @cindex writing patterns in memory
5244 @cindex patterns, writing in memory
5245 @var{repeat}, @var{size} and @var{value} are absolute expressions.
5246 This emits @var{repeat} copies of @var{size} bytes. @var{Repeat}
5247 may be zero or more. @var{Size} may be zero or more, but if it is
5248 more than 8, then it is deemed to have the value 8, compatible with
5249 other people's assemblers. The contents of each @var{repeat} bytes
5250 is taken from an 8-byte number. The highest order 4 bytes are
5251 zero. The lowest order 4 bytes are @var{value} rendered in the
5252 byte-order of an integer on the computer @command{@value{AS}} is assembling for.
5253 Each @var{size} bytes in a repetition is taken from the lowest order
5254 @var{size} bytes of this number. Again, this bizarre behavior is
5255 compatible with other people's assemblers.
5257 @var{size} and @var{value} are optional.
5258 If the second comma and @var{value} are absent, @var{value} is
5259 assumed zero. If the first comma and following tokens are absent,
5260 @var{size} is assumed to be 1.
5263 @section @code{.float @var{flonums}}
5265 @cindex floating point numbers (single)
5266 @cindex @code{float} directive
5267 This directive assembles zero or more flonums, separated by commas. It
5268 has the same effect as @code{.single}.
5270 The exact kind of floating point numbers emitted depends on how
5271 @command{@value{AS}} is configured.
5272 @xref{Machine Dependencies}.
5276 On the @value{TARGET} family, @code{.float} emits 32-bit floating point numbers
5277 in @sc{ieee} format.
5282 @section @code{.func @var{name}[,@var{label}]}
5283 @cindex @code{func} directive
5284 @code{.func} emits debugging information to denote function @var{name}, and
5285 is ignored unless the file is assembled with debugging enabled.
5286 Only @samp{--gstabs[+]} is currently supported.
5287 @var{label} is the entry point of the function and if omitted @var{name}
5288 prepended with the @samp{leading char} is used.
5289 @samp{leading char} is usually @code{_} or nothing, depending on the target.
5290 All functions are currently defined to have @code{void} return type.
5291 The function must be terminated with @code{.endfunc}.
5294 @section @code{.global @var{symbol}}, @code{.globl @var{symbol}}
5296 @cindex @code{global} directive
5297 @cindex symbol, making visible to linker
5298 @code{.global} makes the symbol visible to @code{@value{LD}}. If you define
5299 @var{symbol} in your partial program, its value is made available to
5300 other partial programs that are linked with it. Otherwise,
5301 @var{symbol} takes its attributes from a symbol of the same name
5302 from another file linked into the same program.
5304 Both spellings (@samp{.globl} and @samp{.global}) are accepted, for
5305 compatibility with other assemblers.
5308 On the HPPA, @code{.global} is not always enough to make it accessible to other
5309 partial programs. You may need the HPPA-only @code{.EXPORT} directive as well.
5310 @xref{HPPA Directives, ,HPPA Assembler Directives}.
5315 @section @code{.gnu_attribute @var{tag},@var{value}}
5316 Record a @sc{gnu} object attribute for this file. @xref{Object Attributes}.
5319 @section @code{.hidden @var{names}}
5321 @cindex @code{hidden} directive
5323 This is one of the ELF visibility directives. The other two are
5324 @code{.internal} (@pxref{Internal,,@code{.internal}}) and
5325 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5327 This directive overrides the named symbols default visibility (which is set by
5328 their binding: local, global or weak). The directive sets the visibility to
5329 @code{hidden} which means that the symbols are not visible to other components.
5330 Such symbols are always considered to be @code{protected} as well.
5334 @section @code{.hword @var{expressions}}
5336 @cindex @code{hword} directive
5337 @cindex integers, 16-bit
5338 @cindex numbers, 16-bit
5339 @cindex sixteen bit integers
5340 This expects zero or more @var{expressions}, and emits
5341 a 16 bit number for each.
5344 This directive is a synonym for @samp{.short}; depending on the target
5345 architecture, it may also be a synonym for @samp{.word}.
5349 This directive is a synonym for @samp{.short}.
5352 This directive is a synonym for both @samp{.short} and @samp{.word}.
5357 @section @code{.ident}
5359 @cindex @code{ident} directive
5361 This directive is used by some assemblers to place tags in object files. The
5362 behavior of this directive varies depending on the target. When using the
5363 a.out object file format, @command{@value{AS}} simply accepts the directive for
5364 source-file compatibility with existing assemblers, but does not emit anything
5365 for it. When using COFF, comments are emitted to the @code{.comment} or
5366 @code{.rdata} section, depending on the target. When using ELF, comments are
5367 emitted to the @code{.comment} section.
5370 @section @code{.if @var{absolute expression}}
5372 @cindex conditional assembly
5373 @cindex @code{if} directive
5374 @code{.if} marks the beginning of a section of code which is only
5375 considered part of the source program being assembled if the argument
5376 (which must be an @var{absolute expression}) is non-zero. The end of
5377 the conditional section of code must be marked by @code{.endif}
5378 (@pxref{Endif,,@code{.endif}}); optionally, you may include code for the
5379 alternative condition, flagged by @code{.else} (@pxref{Else,,@code{.else}}).
5380 If you have several conditions to check, @code{.elseif} may be used to avoid
5381 nesting blocks if/else within each subsequent @code{.else} block.
5383 The following variants of @code{.if} are also supported:
5385 @cindex @code{ifdef} directive
5386 @item .ifdef @var{symbol}
5387 Assembles the following section of code if the specified @var{symbol}
5388 has been defined. Note a symbol which has been referenced but not yet defined
5389 is considered to be undefined.
5391 @cindex @code{ifb} directive
5392 @item .ifb @var{text}
5393 Assembles the following section of code if the operand is blank (empty).
5395 @cindex @code{ifc} directive
5396 @item .ifc @var{string1},@var{string2}
5397 Assembles the following section of code if the two strings are the same. The
5398 strings may be optionally quoted with single quotes. If they are not quoted,
5399 the first string stops at the first comma, and the second string stops at the
5400 end of the line. Strings which contain whitespace should be quoted. The
5401 string comparison is case sensitive.
5403 @cindex @code{ifeq} directive
5404 @item .ifeq @var{absolute expression}
5405 Assembles the following section of code if the argument is zero.
5407 @cindex @code{ifeqs} directive
5408 @item .ifeqs @var{string1},@var{string2}
5409 Another form of @code{.ifc}. The strings must be quoted using double quotes.
5411 @cindex @code{ifge} directive
5412 @item .ifge @var{absolute expression}
5413 Assembles the following section of code if the argument is greater than or
5416 @cindex @code{ifgt} directive
5417 @item .ifgt @var{absolute expression}
5418 Assembles the following section of code if the argument is greater than zero.
5420 @cindex @code{ifle} directive
5421 @item .ifle @var{absolute expression}
5422 Assembles the following section of code if the argument is less than or equal
5425 @cindex @code{iflt} directive
5426 @item .iflt @var{absolute expression}
5427 Assembles the following section of code if the argument is less than zero.
5429 @cindex @code{ifnb} directive
5430 @item .ifnb @var{text}
5431 Like @code{.ifb}, but the sense of the test is reversed: this assembles the
5432 following section of code if the operand is non-blank (non-empty).
5434 @cindex @code{ifnc} directive
5435 @item .ifnc @var{string1},@var{string2}.
5436 Like @code{.ifc}, but the sense of the test is reversed: this assembles the
5437 following section of code if the two strings are not the same.
5439 @cindex @code{ifndef} directive
5440 @cindex @code{ifnotdef} directive
5441 @item .ifndef @var{symbol}
5442 @itemx .ifnotdef @var{symbol}
5443 Assembles the following section of code if the specified @var{symbol}
5444 has not been defined. Both spelling variants are equivalent. Note a symbol
5445 which has been referenced but not yet defined is considered to be undefined.
5447 @cindex @code{ifne} directive
5448 @item .ifne @var{absolute expression}
5449 Assembles the following section of code if the argument is not equal to zero
5450 (in other words, this is equivalent to @code{.if}).
5452 @cindex @code{ifnes} directive
5453 @item .ifnes @var{string1},@var{string2}
5454 Like @code{.ifeqs}, but the sense of the test is reversed: this assembles the
5455 following section of code if the two strings are not the same.
5459 @section @code{.incbin "@var{file}"[,@var{skip}[,@var{count}]]}
5461 @cindex @code{incbin} directive
5462 @cindex binary files, including
5463 The @code{incbin} directive includes @var{file} verbatim at the current
5464 location. You can control the search paths used with the @samp{-I} command-line
5465 option (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5468 The @var{skip} argument skips a number of bytes from the start of the
5469 @var{file}. The @var{count} argument indicates the maximum number of bytes to
5470 read. Note that the data is not aligned in any way, so it is the user's
5471 responsibility to make sure that proper alignment is provided both before and
5472 after the @code{incbin} directive.
5475 @section @code{.include "@var{file}"}
5477 @cindex @code{include} directive
5478 @cindex supporting files, including
5479 @cindex files, including
5480 This directive provides a way to include supporting files at specified
5481 points in your source program. The code from @var{file} is assembled as
5482 if it followed the point of the @code{.include}; when the end of the
5483 included file is reached, assembly of the original file continues. You
5484 can control the search paths used with the @samp{-I} command-line option
5485 (@pxref{Invoking,,Command-Line Options}). Quotation marks are required
5489 @section @code{.int @var{expressions}}
5491 @cindex @code{int} directive
5492 @cindex integers, 32-bit
5493 Expect zero or more @var{expressions}, of any section, separated by commas.
5494 For each expression, emit a number that, at run time, is the value of that
5495 expression. The byte order and bit size of the number depends on what kind
5496 of target the assembly is for.
5500 On most forms of the H8/300, @code{.int} emits 16-bit
5501 integers. On the H8/300H and the Renesas SH, however, @code{.int} emits
5508 @section @code{.internal @var{names}}
5510 @cindex @code{internal} directive
5512 This is one of the ELF visibility directives. The other two are
5513 @code{.hidden} (@pxref{Hidden,,@code{.hidden}}) and
5514 @code{.protected} (@pxref{Protected,,@code{.protected}}).
5516 This directive overrides the named symbols default visibility (which is set by
5517 their binding: local, global or weak). The directive sets the visibility to
5518 @code{internal} which means that the symbols are considered to be @code{hidden}
5519 (i.e., not visible to other components), and that some extra, processor specific
5520 processing must also be performed upon the symbols as well.
5524 @section @code{.irp @var{symbol},@var{values}}@dots{}
5526 @cindex @code{irp} directive
5527 Evaluate a sequence of statements assigning different values to @var{symbol}.
5528 The sequence of statements starts at the @code{.irp} directive, and is
5529 terminated by an @code{.endr} directive. For each @var{value}, @var{symbol} is
5530 set to @var{value}, and the sequence of statements is assembled. If no
5531 @var{value} is listed, the sequence of statements is assembled once, with
5532 @var{symbol} set to the null string. To refer to @var{symbol} within the
5533 sequence of statements, use @var{\symbol}.
5535 For example, assembling
5543 is equivalent to assembling
5551 For some caveats with the spelling of @var{symbol}, see also @ref{Macro}.
5554 @section @code{.irpc @var{symbol},@var{values}}@dots{}
5556 @cindex @code{irpc} directive
5557 Evaluate a sequence of statements assigning different values to @var{symbol}.
5558 The sequence of statements starts at the @code{.irpc} directive, and is
5559 terminated by an @code{.endr} directive. For each character in @var{value},
5560 @var{symbol} is set to the character, and the sequence of statements is
5561 assembled. If no @var{value} is listed, the sequence of statements is
5562 assembled once, with @var{symbol} set to the null string. To refer to
5563 @var{symbol} within the sequence of statements, use @var{\symbol}.
5565 For example, assembling
5573 is equivalent to assembling
5581 For some caveats with the spelling of @var{symbol}, see also the discussion
5585 @section @code{.lcomm @var{symbol} , @var{length}}
5587 @cindex @code{lcomm} directive
5588 @cindex local common symbols
5589 @cindex symbols, local common
5590 Reserve @var{length} (an absolute expression) bytes for a local common
5591 denoted by @var{symbol}. The section and value of @var{symbol} are
5592 those of the new local common. The addresses are allocated in the bss
5593 section, so that at run-time the bytes start off zeroed. @var{Symbol}
5594 is not declared global (@pxref{Global,,@code{.global}}), so is normally
5595 not visible to @code{@value{LD}}.
5598 Some targets permit a third argument to be used with @code{.lcomm}. This
5599 argument specifies the desired alignment of the symbol in the bss section.
5603 The syntax for @code{.lcomm} differs slightly on the HPPA. The syntax is
5604 @samp{@var{symbol} .lcomm, @var{length}}; @var{symbol} is optional.
5608 @section @code{.lflags}
5610 @cindex @code{lflags} directive (ignored)
5611 @command{@value{AS}} accepts this directive, for compatibility with other
5612 assemblers, but ignores it.
5614 @ifclear no-line-dir
5616 @section @code{.line @var{line-number}}
5618 @cindex @code{line} directive
5619 @cindex logical line number
5621 Change the logical line number. @var{line-number} must be an absolute
5622 expression. The next line has that logical line number. Therefore any other
5623 statements on the current line (after a statement separator character) are
5624 reported as on logical line number @var{line-number} @minus{} 1. One day
5625 @command{@value{AS}} will no longer support this directive: it is recognized only
5626 for compatibility with existing assembler programs.
5629 Even though this is a directive associated with the @code{a.out} or
5630 @code{b.out} object-code formats, @command{@value{AS}} still recognizes it
5631 when producing COFF output, and treats @samp{.line} as though it
5632 were the COFF @samp{.ln} @emph{if} it is found outside a
5633 @code{.def}/@code{.endef} pair.
5635 Inside a @code{.def}, @samp{.line} is, instead, one of the directives
5636 used by compilers to generate auxiliary symbol information for
5641 @section @code{.linkonce [@var{type}]}
5643 @cindex @code{linkonce} directive
5644 @cindex common sections
5645 Mark the current section so that the linker only includes a single copy of it.
5646 This may be used to include the same section in several different object files,
5647 but ensure that the linker will only include it once in the final output file.
5648 The @code{.linkonce} pseudo-op must be used for each instance of the section.
5649 Duplicate sections are detected based on the section name, so it should be
5652 This directive is only supported by a few object file formats; as of this
5653 writing, the only object file format which supports it is the Portable
5654 Executable format used on Windows NT.
5656 The @var{type} argument is optional. If specified, it must be one of the
5657 following strings. For example:
5661 Not all types may be supported on all object file formats.
5665 Silently discard duplicate sections. This is the default.
5668 Warn if there are duplicate sections, but still keep only one copy.
5671 Warn if any of the duplicates have different sizes.
5674 Warn if any of the duplicates do not have exactly the same contents.
5678 @section @code{.list}
5680 @cindex @code{list} directive
5681 @cindex listing control, turning on
5682 Control (in conjunction with the @code{.nolist} directive) whether or
5683 not assembly listings are generated. These two directives maintain an
5684 internal counter (which is zero initially). @code{.list} increments the
5685 counter, and @code{.nolist} decrements it. Assembly listings are
5686 generated whenever the counter is greater than zero.
5688 By default, listings are disabled. When you enable them (with the
5689 @samp{-a} command line option; @pxref{Invoking,,Command-Line Options}),
5690 the initial value of the listing counter is one.
5693 @section @code{.ln @var{line-number}}
5695 @cindex @code{ln} directive
5696 @ifclear no-line-dir
5697 @samp{.ln} is a synonym for @samp{.line}.
5700 Tell @command{@value{AS}} to change the logical line number. @var{line-number}
5701 must be an absolute expression. The next line has that logical
5702 line number, so any other statements on the current line (after a
5703 statement separator character @code{;}) are reported as on logical
5704 line number @var{line-number} @minus{} 1.
5708 @section @code{.loc @var{fileno} @var{lineno} [@var{column}] [@var{options}]}
5709 @cindex @code{loc} directive
5710 When emitting DWARF2 line number information,
5711 the @code{.loc} directive will add a row to the @code{.debug_line} line
5712 number matrix corresponding to the immediately following assembly
5713 instruction. The @var{fileno}, @var{lineno}, and optional @var{column}
5714 arguments will be applied to the @code{.debug_line} state machine before
5717 The @var{options} are a sequence of the following tokens in any order:
5721 This option will set the @code{basic_block} register in the
5722 @code{.debug_line} state machine to @code{true}.
5725 This option will set the @code{prologue_end} register in the
5726 @code{.debug_line} state machine to @code{true}.
5728 @item epilogue_begin
5729 This option will set the @code{epilogue_begin} register in the
5730 @code{.debug_line} state machine to @code{true}.
5732 @item is_stmt @var{value}
5733 This option will set the @code{is_stmt} register in the
5734 @code{.debug_line} state machine to @code{value}, which must be
5737 @item isa @var{value}
5738 This directive will set the @code{isa} register in the @code{.debug_line}
5739 state machine to @var{value}, which must be an unsigned integer.
5741 @item discriminator @var{value}
5742 This directive will set the @code{discriminator} register in the @code{.debug_line}
5743 state machine to @var{value}, which must be an unsigned integer.
5745 @item view @var{value}
5746 This option causes a row to be added to @code{.debug_line} in reference to the
5747 current address (which might not be the same as that of the following assembly
5748 instruction), and to associate @var{value} with the @code{view} register in the
5749 @code{.debug_line} state machine. If @var{value} is a label, both the
5750 @code{view} register and the label are set to the number of prior @code{.loc}
5751 directives at the same program location. If @var{value} is the literal
5752 @code{0}, the @code{view} register is set to zero, and the assembler asserts
5753 that there aren't any prior @code{.loc} directives at the same program
5754 location. If @var{value} is the literal @code{-0}, the assembler arrange for
5755 the @code{view} register to be reset in this row, even if there are prior
5756 @code{.loc} directives at the same program location.
5760 @node Loc_mark_labels
5761 @section @code{.loc_mark_labels @var{enable}}
5762 @cindex @code{loc_mark_labels} directive
5763 When emitting DWARF2 line number information,
5764 the @code{.loc_mark_labels} directive makes the assembler emit an entry
5765 to the @code{.debug_line} line number matrix with the @code{basic_block}
5766 register in the state machine set whenever a code label is seen.
5767 The @var{enable} argument should be either 1 or 0, to enable or disable
5768 this function respectively.
5772 @section @code{.local @var{names}}
5774 @cindex @code{local} directive
5775 This directive, which is available for ELF targets, marks each symbol in
5776 the comma-separated list of @code{names} as a local symbol so that it
5777 will not be externally visible. If the symbols do not already exist,
5778 they will be created.
5780 For targets where the @code{.lcomm} directive (@pxref{Lcomm}) does not
5781 accept an alignment argument, which is the case for most ELF targets,
5782 the @code{.local} directive can be used in combination with @code{.comm}
5783 (@pxref{Comm}) to define aligned local common data.
5787 @section @code{.long @var{expressions}}
5789 @cindex @code{long} directive
5790 @code{.long} is the same as @samp{.int}. @xref{Int,,@code{.int}}.
5793 @c no one seems to know what this is for or whether this description is
5794 @c what it really ought to do
5796 @section @code{.lsym @var{symbol}, @var{expression}}
5798 @cindex @code{lsym} directive
5799 @cindex symbol, not referenced in assembly
5800 @code{.lsym} creates a new symbol named @var{symbol}, but does not put it in
5801 the hash table, ensuring it cannot be referenced by name during the
5802 rest of the assembly. This sets the attributes of the symbol to be
5803 the same as the expression value:
5805 @var{other} = @var{descriptor} = 0
5806 @var{type} = @r{(section of @var{expression})}
5807 @var{value} = @var{expression}
5810 The new symbol is not flagged as external.
5814 @section @code{.macro}
5817 The commands @code{.macro} and @code{.endm} allow you to define macros that
5818 generate assembly output. For example, this definition specifies a macro
5819 @code{sum} that puts a sequence of numbers into memory:
5822 .macro sum from=0, to=5
5831 With that definition, @samp{SUM 0,5} is equivalent to this assembly input:
5843 @item .macro @var{macname}
5844 @itemx .macro @var{macname} @var{macargs} @dots{}
5845 @cindex @code{macro} directive
5846 Begin the definition of a macro called @var{macname}. If your macro
5847 definition requires arguments, specify their names after the macro name,
5848 separated by commas or spaces. You can qualify the macro argument to
5849 indicate whether all invocations must specify a non-blank value (through
5850 @samp{:@code{req}}), or whether it takes all of the remaining arguments
5851 (through @samp{:@code{vararg}}). You can supply a default value for any
5852 macro argument by following the name with @samp{=@var{deflt}}. You
5853 cannot define two macros with the same @var{macname} unless it has been
5854 subject to the @code{.purgem} directive (@pxref{Purgem}) between the two
5855 definitions. For example, these are all valid @code{.macro} statements:
5859 Begin the definition of a macro called @code{comm}, which takes no
5862 @item .macro plus1 p, p1
5863 @itemx .macro plus1 p p1
5864 Either statement begins the definition of a macro called @code{plus1},
5865 which takes two arguments; within the macro definition, write
5866 @samp{\p} or @samp{\p1} to evaluate the arguments.
5868 @item .macro reserve_str p1=0 p2
5869 Begin the definition of a macro called @code{reserve_str}, with two
5870 arguments. The first argument has a default value, but not the second.
5871 After the definition is complete, you can call the macro either as
5872 @samp{reserve_str @var{a},@var{b}} (with @samp{\p1} evaluating to
5873 @var{a} and @samp{\p2} evaluating to @var{b}), or as @samp{reserve_str
5874 ,@var{b}} (with @samp{\p1} evaluating as the default, in this case
5875 @samp{0}, and @samp{\p2} evaluating to @var{b}).
5877 @item .macro m p1:req, p2=0, p3:vararg
5878 Begin the definition of a macro called @code{m}, with at least three
5879 arguments. The first argument must always have a value specified, but
5880 not the second, which instead has a default value. The third formal
5881 will get assigned all remaining arguments specified at invocation time.
5883 When you call a macro, you can specify the argument values either by
5884 position, or by keyword. For example, @samp{sum 9,17} is equivalent to
5885 @samp{sum to=17, from=9}.
5889 Note that since each of the @var{macargs} can be an identifier exactly
5890 as any other one permitted by the target architecture, there may be
5891 occasional problems if the target hand-crafts special meanings to certain
5892 characters when they occur in a special position. For example, if the colon
5893 (@code{:}) is generally permitted to be part of a symbol name, but the
5894 architecture specific code special-cases it when occurring as the final
5895 character of a symbol (to denote a label), then the macro parameter
5896 replacement code will have no way of knowing that and consider the whole
5897 construct (including the colon) an identifier, and check only this
5898 identifier for being the subject to parameter substitution. So for example
5899 this macro definition:
5907 might not work as expected. Invoking @samp{label foo} might not create a label
5908 called @samp{foo} but instead just insert the text @samp{\l:} into the
5909 assembler source, probably generating an error about an unrecognised
5912 Similarly problems might occur with the period character (@samp{.})
5913 which is often allowed inside opcode names (and hence identifier names). So
5914 for example constructing a macro to build an opcode from a base name and a
5915 length specifier like this:
5918 .macro opcode base length
5923 and invoking it as @samp{opcode store l} will not create a @samp{store.l}
5924 instruction but instead generate some kind of error as the assembler tries to
5925 interpret the text @samp{\base.\length}.
5927 There are several possible ways around this problem:
5930 @item Insert white space
5931 If it is possible to use white space characters then this is the simplest
5940 @item Use @samp{\()}
5941 The string @samp{\()} can be used to separate the end of a macro argument from
5942 the following text. eg:
5945 .macro opcode base length
5950 @item Use the alternate macro syntax mode
5951 In the alternative macro syntax mode the ampersand character (@samp{&}) can be
5952 used as a separator. eg:
5962 Note: this problem of correctly identifying string parameters to pseudo ops
5963 also applies to the identifiers used in @code{.irp} (@pxref{Irp})
5964 and @code{.irpc} (@pxref{Irpc}) as well.
5967 @cindex @code{endm} directive
5968 Mark the end of a macro definition.
5971 @cindex @code{exitm} directive
5972 Exit early from the current macro definition.
5974 @cindex number of macros executed
5975 @cindex macros, count executed
5977 @command{@value{AS}} maintains a counter of how many macros it has
5978 executed in this pseudo-variable; you can copy that number to your
5979 output with @samp{\@@}, but @emph{only within a macro definition}.
5981 @item LOCAL @var{name} [ , @dots{} ]
5982 @emph{Warning: @code{LOCAL} is only available if you select ``alternate
5983 macro syntax'' with @samp{--alternate} or @code{.altmacro}.}
5984 @xref{Altmacro,,@code{.altmacro}}.
5988 @section @code{.mri @var{val}}
5990 @cindex @code{mri} directive
5991 @cindex MRI mode, temporarily
5992 If @var{val} is non-zero, this tells @command{@value{AS}} to enter MRI mode. If
5993 @var{val} is zero, this tells @command{@value{AS}} to exit MRI mode. This change
5994 affects code assembled until the next @code{.mri} directive, or until the end
5995 of the file. @xref{M, MRI mode, MRI mode}.
5998 @section @code{.noaltmacro}
5999 Disable alternate macro mode. @xref{Altmacro}.
6002 @section @code{.nolist}
6004 @cindex @code{nolist} directive
6005 @cindex listing control, turning off
6006 Control (in conjunction with the @code{.list} directive) whether or
6007 not assembly listings are generated. These two directives maintain an
6008 internal counter (which is zero initially). @code{.list} increments the
6009 counter, and @code{.nolist} decrements it. Assembly listings are
6010 generated whenever the counter is greater than zero.
6013 @section @code{.nops @var{size}[, @var{control}]}
6015 @cindex @code{nops} directive
6016 @cindex filling memory with no-op instructions
6017 This directive emits @var{size} bytes filled with no-op instructions.
6018 @var{size} is absolute expression, which must be a positve value.
6019 @var{control} controls how no-op instructions should be generated. If
6020 the comma and @var{control} are omitted, @var{control} is assumed to be
6023 Note: For Intel 80386 and AMD x86-64 targets, @var{control} specifies
6024 the size limit of a no-op instruction. The valid values of @var{control}
6025 are between 0 and 4 in 16-bit mode, between 0 and 7 when tuning for
6026 older processors in 32-bit mode, between 0 and 11 in 64-bit mode or when
6027 tuning for newer processors in 32-bit mode. When 0 is used, the no-op
6028 instruction size limit is set to the maximum supported size.
6031 @section @code{.octa @var{bignums}}
6033 @c FIXME: double size emitted for "octa" on some? Or warn?
6034 @cindex @code{octa} directive
6035 @cindex integer, 16-byte
6036 @cindex sixteen byte integer
6037 This directive expects zero or more bignums, separated by commas. For each
6038 bignum, it emits a 16-byte integer.
6040 The term ``octa'' comes from contexts in which a ``word'' is two bytes;
6041 hence @emph{octa}-word for 16 bytes.
6044 @section @code{.offset @var{loc}}
6046 @cindex @code{offset} directive
6047 Set the location counter to @var{loc} in the absolute section. @var{loc} must
6048 be an absolute expression. This directive may be useful for defining
6049 symbols with absolute values. Do not confuse it with the @code{.org}
6053 @section @code{.org @var{new-lc} , @var{fill}}
6055 @cindex @code{org} directive
6056 @cindex location counter, advancing
6057 @cindex advancing location counter
6058 @cindex current address, advancing
6059 Advance the location counter of the current section to
6060 @var{new-lc}. @var{new-lc} is either an absolute expression or an
6061 expression with the same section as the current subsection. That is,
6062 you can't use @code{.org} to cross sections: if @var{new-lc} has the
6063 wrong section, the @code{.org} directive is ignored. To be compatible
6064 with former assemblers, if the section of @var{new-lc} is absolute,
6065 @command{@value{AS}} issues a warning, then pretends the section of @var{new-lc}
6066 is the same as the current subsection.
6068 @code{.org} may only increase the location counter, or leave it
6069 unchanged; you cannot use @code{.org} to move the location counter
6072 @c double negative used below "not undefined" because this is a specific
6073 @c reference to "undefined" (as SEG_UNKNOWN is called in this manual)
6074 @c section. doc@cygnus.com 18feb91
6075 Because @command{@value{AS}} tries to assemble programs in one pass, @var{new-lc}
6076 may not be undefined. If you really detest this restriction we eagerly await
6077 a chance to share your improved assembler.
6079 Beware that the origin is relative to the start of the section, not
6080 to the start of the subsection. This is compatible with other
6081 people's assemblers.
6083 When the location counter (of the current subsection) is advanced, the
6084 intervening bytes are filled with @var{fill} which should be an
6085 absolute expression. If the comma and @var{fill} are omitted,
6086 @var{fill} defaults to zero.
6089 @section @code{.p2align[wl] @var{abs-expr}, @var{abs-expr}, @var{abs-expr}}
6091 @cindex padding the location counter given a power of two
6092 @cindex @code{p2align} directive
6093 Pad the location counter (in the current subsection) to a particular
6094 storage boundary. The first expression (which must be absolute) is the
6095 number of low-order zero bits the location counter must have after
6096 advancement. For example @samp{.p2align 3} advances the location
6097 counter until it a multiple of 8. If the location counter is already a
6098 multiple of 8, no change is needed.
6100 The second expression (also absolute) gives the fill value to be stored in the
6101 padding bytes. It (and the comma) may be omitted. If it is omitted, the
6102 padding bytes are normally zero. However, on most systems, if the section is
6103 marked as containing code and the fill value is omitted, the space is filled
6104 with no-op instructions.
6106 The third expression is also absolute, and is also optional. If it is present,
6107 it is the maximum number of bytes that should be skipped by this alignment
6108 directive. If doing the alignment would require skipping more bytes than the
6109 specified maximum, then the alignment is not done at all. You can omit the
6110 fill value (the second argument) entirely by simply using two commas after the
6111 required alignment; this can be useful if you want the alignment to be filled
6112 with no-op instructions when appropriate.
6114 @cindex @code{p2alignw} directive
6115 @cindex @code{p2alignl} directive
6116 The @code{.p2alignw} and @code{.p2alignl} directives are variants of the
6117 @code{.p2align} directive. The @code{.p2alignw} directive treats the fill
6118 pattern as a two byte word value. The @code{.p2alignl} directives treats the
6119 fill pattern as a four byte longword value. For example, @code{.p2alignw
6120 2,0x368d} will align to a multiple of 4. If it skips two bytes, they will be
6121 filled in with the value 0x368d (the exact placement of the bytes depends upon
6122 the endianness of the processor). If it skips 1 or 3 bytes, the fill value is
6127 @section @code{.popsection}
6129 @cindex @code{popsection} directive
6130 @cindex Section Stack
6131 This is one of the ELF section stack manipulation directives. The others are
6132 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6133 @code{.pushsection} (@pxref{PushSection}), and @code{.previous}
6136 This directive replaces the current section (and subsection) with the top
6137 section (and subsection) on the section stack. This section is popped off the
6143 @section @code{.previous}
6145 @cindex @code{previous} directive
6146 @cindex Section Stack
6147 This is one of the ELF section stack manipulation directives. The others are
6148 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6149 @code{.pushsection} (@pxref{PushSection}), and @code{.popsection}
6150 (@pxref{PopSection}).
6152 This directive swaps the current section (and subsection) with most recently
6153 referenced section/subsection pair prior to this one. Multiple
6154 @code{.previous} directives in a row will flip between two sections (and their
6155 subsections). For example:
6167 Will place 0x1234 and 0x9abc into subsection 1 and 0x5678 into subsection 2 of
6173 # Now in section A subsection 1
6177 # Now in section B subsection 0
6180 # Now in section B subsection 1
6183 # Now in section B subsection 0
6187 Will place 0x1234 into section A, 0x5678 and 0xdef0 into subsection 0 of
6188 section B and 0x9abc into subsection 1 of section B.
6190 In terms of the section stack, this directive swaps the current section with
6191 the top section on the section stack.
6195 @section @code{.print @var{string}}
6197 @cindex @code{print} directive
6198 @command{@value{AS}} will print @var{string} on the standard output during
6199 assembly. You must put @var{string} in double quotes.
6203 @section @code{.protected @var{names}}
6205 @cindex @code{protected} directive
6207 This is one of the ELF visibility directives. The other two are
6208 @code{.hidden} (@pxref{Hidden}) and @code{.internal} (@pxref{Internal}).
6210 This directive overrides the named symbols default visibility (which is set by
6211 their binding: local, global or weak). The directive sets the visibility to
6212 @code{protected} which means that any references to the symbols from within the
6213 components that defines them must be resolved to the definition in that
6214 component, even if a definition in another component would normally preempt
6219 @section @code{.psize @var{lines} , @var{columns}}
6221 @cindex @code{psize} directive
6222 @cindex listing control: paper size
6223 @cindex paper size, for listings
6224 Use this directive to declare the number of lines---and, optionally, the
6225 number of columns---to use for each page, when generating listings.
6227 If you do not use @code{.psize}, listings use a default line-count
6228 of 60. You may omit the comma and @var{columns} specification; the
6229 default width is 200 columns.
6231 @command{@value{AS}} generates formfeeds whenever the specified number of
6232 lines is exceeded (or whenever you explicitly request one, using
6235 If you specify @var{lines} as @code{0}, no formfeeds are generated save
6236 those explicitly specified with @code{.eject}.
6239 @section @code{.purgem @var{name}}
6241 @cindex @code{purgem} directive
6242 Undefine the macro @var{name}, so that later uses of the string will not be
6243 expanded. @xref{Macro}.
6247 @section @code{.pushsection @var{name} [, @var{subsection}] [, "@var{flags}"[, @@@var{type}[,@var{arguments}]]]}
6249 @cindex @code{pushsection} directive
6250 @cindex Section Stack
6251 This is one of the ELF section stack manipulation directives. The others are
6252 @code{.section} (@pxref{Section}), @code{.subsection} (@pxref{SubSection}),
6253 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6256 This directive pushes the current section (and subsection) onto the
6257 top of the section stack, and then replaces the current section and
6258 subsection with @code{name} and @code{subsection}. The optional
6259 @code{flags}, @code{type} and @code{arguments} are treated the same
6260 as in the @code{.section} (@pxref{Section}) directive.
6264 @section @code{.quad @var{bignums}}
6266 @cindex @code{quad} directive
6267 @code{.quad} expects zero or more bignums, separated by commas. For
6268 each bignum, it emits
6270 an 8-byte integer. If the bignum won't fit in 8 bytes, it prints a
6271 warning message; and just takes the lowest order 8 bytes of the bignum.
6272 @cindex eight-byte integer
6273 @cindex integer, 8-byte
6275 The term ``quad'' comes from contexts in which a ``word'' is two bytes;
6276 hence @emph{quad}-word for 8 bytes.
6279 a 16-byte integer. If the bignum won't fit in 16 bytes, it prints a
6280 warning message; and just takes the lowest order 16 bytes of the bignum.
6281 @cindex sixteen-byte integer
6282 @cindex integer, 16-byte
6286 @section @code{.reloc @var{offset}, @var{reloc_name}[, @var{expression}]}
6288 @cindex @code{reloc} directive
6289 Generate a relocation at @var{offset} of type @var{reloc_name} with value
6290 @var{expression}. If @var{offset} is a number, the relocation is generated in
6291 the current section. If @var{offset} is an expression that resolves to a
6292 symbol plus offset, the relocation is generated in the given symbol's section.
6293 @var{expression}, if present, must resolve to a symbol plus addend or to an
6294 absolute value, but note that not all targets support an addend. e.g. ELF REL
6295 targets such as i386 store an addend in the section contents rather than in the
6296 relocation. This low level interface does not support addends stored in the
6300 @section @code{.rept @var{count}}
6302 @cindex @code{rept} directive
6303 Repeat the sequence of lines between the @code{.rept} directive and the next
6304 @code{.endr} directive @var{count} times.
6306 For example, assembling
6314 is equivalent to assembling
6322 A count of zero is allowed, but nothing is generated. Negative counts are not
6323 allowed and if encountered will be treated as if they were zero.
6326 @section @code{.sbttl "@var{subheading}"}
6328 @cindex @code{sbttl} directive
6329 @cindex subtitles for listings
6330 @cindex listing control: subtitle
6331 Use @var{subheading} as the title (third line, immediately after the
6332 title line) when generating assembly listings.
6334 This directive affects subsequent pages, as well as the current page if
6335 it appears within ten lines of the top of a page.
6339 @section @code{.scl @var{class}}
6341 @cindex @code{scl} directive
6342 @cindex symbol storage class (COFF)
6343 @cindex COFF symbol storage class
6344 Set the storage-class value for a symbol. This directive may only be
6345 used inside a @code{.def}/@code{.endef} pair. Storage class may flag
6346 whether a symbol is static or external, or it may record further
6347 symbolic debugging information.
6352 @section @code{.section @var{name}}
6354 @cindex named section
6355 Use the @code{.section} directive to assemble the following code into a section
6358 This directive is only supported for targets that actually support arbitrarily
6359 named sections; on @code{a.out} targets, for example, it is not accepted, even
6360 with a standard @code{a.out} section name.
6364 @c only print the extra heading if both COFF and ELF are set
6365 @subheading COFF Version
6368 @cindex @code{section} directive (COFF version)
6369 For COFF targets, the @code{.section} directive is used in one of the following
6373 .section @var{name}[, "@var{flags}"]
6374 .section @var{name}[, @var{subsection}]
6377 If the optional argument is quoted, it is taken as flags to use for the
6378 section. Each flag is a single character. The following flags are recognized:
6382 bss section (uninitialized data)
6384 section is not loaded
6390 exclude section from linking
6396 shared section (meaningful for PE targets)
6398 ignored. (For compatibility with the ELF version)
6400 section is not readable (meaningful for PE targets)
6402 single-digit power-of-two section alignment (GNU extension)
6405 If no flags are specified, the default flags depend upon the section name. If
6406 the section name is not recognized, the default will be for the section to be
6407 loaded and writable. Note the @code{n} and @code{w} flags remove attributes
6408 from the section, rather than adding them, so if they are used on their own it
6409 will be as if no flags had been specified at all.
6411 If the optional argument to the @code{.section} directive is not quoted, it is
6412 taken as a subsection number (@pxref{Sub-Sections}).
6417 @c only print the extra heading if both COFF and ELF are set
6418 @subheading ELF Version
6421 @cindex Section Stack
6422 This is one of the ELF section stack manipulation directives. The others are
6423 @code{.subsection} (@pxref{SubSection}), @code{.pushsection}
6424 (@pxref{PushSection}), @code{.popsection} (@pxref{PopSection}), and
6425 @code{.previous} (@pxref{Previous}).
6427 @cindex @code{section} directive (ELF version)
6428 For ELF targets, the @code{.section} directive is used like this:
6431 .section @var{name} [, "@var{flags}"[, @@@var{type}[,@var{flag_specific_arguments}]]]
6434 @anchor{Section Name Substitutions}
6435 @kindex --sectname-subst
6436 @cindex section name substitution
6437 If the @samp{--sectname-subst} command-line option is provided, the @var{name}
6438 argument may contain a substitution sequence. Only @code{%S} is supported
6439 at the moment, and substitutes the current section name. For example:
6442 .macro exception_code
6443 .section %S.exception
6444 [exception code here]
6459 The two @code{exception_code} invocations above would create the
6460 @code{.text.exception} and @code{.init.exception} sections respectively.
6461 This is useful e.g. to discriminate between ancillary sections that are
6462 tied to setup code to be discarded after use from ancillary sections that
6463 need to stay resident without having to define multiple @code{exception_code}
6464 macros just for that purpose.
6466 The optional @var{flags} argument is a quoted string which may contain any
6467 combination of the following characters:
6471 section is allocatable
6473 section is a GNU_MBIND section
6475 section is excluded from executable and shared library.
6479 section is executable
6481 section is mergeable
6483 section contains zero terminated strings
6485 section is a member of a section group
6487 section is used for thread-local-storage
6489 section is a member of the previously-current section's group, if any
6490 @item @code{<number>}
6491 a numeric value indicating the bits to be set in the ELF section header's flags
6492 field. Note - if one or more of the alphabetic characters described above is
6493 also included in the flags field, their bit values will be ORed into the
6495 @item @code{<target specific>}
6496 some targets extend this list with their own flag characters
6499 Note - once a section's flags have been set they cannot be changed. There are
6500 a few exceptions to this rule however. Processor and application specific
6501 flags can be added to an already defined section. The @code{.interp},
6502 @code{.strtab} and @code{.symtab} sections can have the allocate flag
6503 (@code{a}) set after they are initially defined, and the @code{.note-GNU-stack}
6504 section may have the executable (@code{x}) flag added.
6506 The optional @var{type} argument may contain one of the following constants:
6510 section contains data
6512 section does not contain data (i.e., section only occupies space)
6514 section contains data which is used by things other than the program
6516 section contains an array of pointers to init functions
6518 section contains an array of pointers to finish functions
6519 @item @@preinit_array
6520 section contains an array of pointers to pre-init functions
6521 @item @@@code{<number>}
6522 a numeric value to be set as the ELF section header's type field.
6523 @item @@@code{<target specific>}
6524 some targets extend this list with their own types
6527 Many targets only support the first three section types. The type may be
6528 enclosed in double quotes if necessary.
6530 Note on targets where the @code{@@} character is the start of a comment (eg
6531 ARM) then another character is used instead. For example the ARM port uses the
6534 Note - some sections, eg @code{.text} and @code{.data} are considered to be
6535 special and have fixed types. Any attempt to declare them with a different
6536 type will generate an error from the assembler.
6538 If @var{flags} contains the @code{M} symbol then the @var{type} argument must
6539 be specified as well as an extra argument---@var{entsize}---like this:
6542 .section @var{name} , "@var{flags}"M, @@@var{type}, @var{entsize}
6545 Sections with the @code{M} flag but not @code{S} flag must contain fixed size
6546 constants, each @var{entsize} octets long. Sections with both @code{M} and
6547 @code{S} must contain zero terminated strings where each character is
6548 @var{entsize} bytes long. The linker may remove duplicates within sections with
6549 the same name, same entity size and same flags. @var{entsize} must be an
6550 absolute expression. For sections with both @code{M} and @code{S}, a string
6551 which is a suffix of a larger string is considered a duplicate. Thus
6552 @code{"def"} will be merged with @code{"abcdef"}; A reference to the first
6553 @code{"def"} will be changed to a reference to @code{"abcdef"+3}.
6555 If @var{flags} contains the @code{G} symbol then the @var{type} argument must
6556 be present along with an additional field like this:
6559 .section @var{name} , "@var{flags}"G, @@@var{type}, @var{GroupName}[, @var{linkage}]
6562 The @var{GroupName} field specifies the name of the section group to which this
6563 particular section belongs. The optional linkage field can contain:
6567 indicates that only one copy of this section should be retained
6572 Note: if both the @var{M} and @var{G} flags are present then the fields for
6573 the Merge flag should come first, like this:
6576 .section @var{name} , "@var{flags}"MG, @@@var{type}, @var{entsize}, @var{GroupName}[, @var{linkage}]
6579 If @var{flags} contains the @code{?} symbol then it may not also contain the
6580 @code{G} symbol and the @var{GroupName} or @var{linkage} fields should not be
6581 present. Instead, @code{?} says to consider the section that's current before
6582 this directive. If that section used @code{G}, then the new section will use
6583 @code{G} with those same @var{GroupName} and @var{linkage} fields implicitly.
6584 If not, then the @code{?} symbol has no effect.
6586 If no flags are specified, the default flags depend upon the section name. If
6587 the section name is not recognized, the default will be for the section to have
6588 none of the above flags: it will not be allocated in memory, nor writable, nor
6589 executable. The section will contain data.
6591 For ELF targets, the assembler supports another type of @code{.section}
6592 directive for compatibility with the Solaris assembler:
6595 .section "@var{name}"[, @var{flags}...]
6598 Note that the section name is quoted. There may be a sequence of comma
6603 section is allocatable
6607 section is executable
6609 section is excluded from executable and shared library.
6611 section is used for thread local storage
6614 This directive replaces the current section and subsection. See the
6615 contents of the gas testsuite directory @code{gas/testsuite/gas/elf} for
6616 some examples of how this directive and the other section stack directives
6622 @section @code{.set @var{symbol}, @var{expression}}
6624 @cindex @code{set} directive
6625 @cindex symbol value, setting
6626 Set the value of @var{symbol} to @var{expression}. This
6627 changes @var{symbol}'s value and type to conform to
6628 @var{expression}. If @var{symbol} was flagged as external, it remains
6629 flagged (@pxref{Symbol Attributes}).
6631 You may @code{.set} a symbol many times in the same assembly provided that the
6632 values given to the symbol are constants. Values that are based on expressions
6633 involving other symbols are allowed, but some targets may restrict this to only
6634 being done once per assembly. This is because those targets do not set the
6635 addresses of symbols at assembly time, but rather delay the assignment until a
6636 final link is performed. This allows the linker a chance to change the code in
6637 the files, changing the location of, and the relative distance between, various
6640 If you @code{.set} a global symbol, the value stored in the object
6641 file is the last value stored into it.
6644 On Z80 @code{set} is a real instruction, use
6645 @samp{@var{symbol} defl @var{expression}} instead.
6649 @section @code{.short @var{expressions}}
6651 @cindex @code{short} directive
6653 @code{.short} is normally the same as @samp{.word}.
6654 @xref{Word,,@code{.word}}.
6656 In some configurations, however, @code{.short} and @code{.word} generate
6657 numbers of different lengths. @xref{Machine Dependencies}.
6661 @code{.short} is the same as @samp{.word}. @xref{Word,,@code{.word}}.
6664 This expects zero or more @var{expressions}, and emits
6665 a 16 bit number for each.
6670 @section @code{.single @var{flonums}}
6672 @cindex @code{single} directive
6673 @cindex floating point numbers (single)
6674 This directive assembles zero or more flonums, separated by commas. It
6675 has the same effect as @code{.float}.
6677 The exact kind of floating point numbers emitted depends on how
6678 @command{@value{AS}} is configured. @xref{Machine Dependencies}.
6682 On the @value{TARGET} family, @code{.single} emits 32-bit floating point
6683 numbers in @sc{ieee} format.
6689 @section @code{.size}
6691 This directive is used to set the size associated with a symbol.
6695 @c only print the extra heading if both COFF and ELF are set
6696 @subheading COFF Version
6699 @cindex @code{size} directive (COFF version)
6700 For COFF targets, the @code{.size} directive is only permitted inside
6701 @code{.def}/@code{.endef} pairs. It is used like this:
6704 .size @var{expression}
6711 @c only print the extra heading if both COFF and ELF are set
6712 @subheading ELF Version
6715 @cindex @code{size} directive (ELF version)
6716 For ELF targets, the @code{.size} directive is used like this:
6719 .size @var{name} , @var{expression}
6722 This directive sets the size associated with a symbol @var{name}.
6723 The size in bytes is computed from @var{expression} which can make use of label
6724 arithmetic. This directive is typically used to set the size of function
6729 @ifclear no-space-dir
6731 @section @code{.skip @var{size} [,@var{fill}]}
6733 @cindex @code{skip} directive
6734 @cindex filling memory
6735 This directive emits @var{size} bytes, each of value @var{fill}. Both
6736 @var{size} and @var{fill} are absolute expressions. If the comma and
6737 @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same as
6742 @section @code{.sleb128 @var{expressions}}
6744 @cindex @code{sleb128} directive
6745 @var{sleb128} stands for ``signed little endian base 128.'' This is a
6746 compact, variable length representation of numbers used by the DWARF
6747 symbolic debugging format. @xref{Uleb128, ,@code{.uleb128}}.
6749 @ifclear no-space-dir
6751 @section @code{.space @var{size} [,@var{fill}]}
6753 @cindex @code{space} directive
6754 @cindex filling memory
6755 This directive emits @var{size} bytes, each of value @var{fill}. Both
6756 @var{size} and @var{fill} are absolute expressions. If the comma
6757 and @var{fill} are omitted, @var{fill} is assumed to be zero. This is the same
6762 @emph{Warning:} @code{.space} has a completely different meaning for HPPA
6763 targets; use @code{.block} as a substitute. See @cite{HP9000 Series 800
6764 Assembly Language Reference Manual} (HP 92432-90001) for the meaning of the
6765 @code{.space} directive. @xref{HPPA Directives,,HPPA Assembler Directives},
6773 @section @code{.stabd, .stabn, .stabs}
6775 @cindex symbolic debuggers, information for
6776 @cindex @code{stab@var{x}} directives
6777 There are three directives that begin @samp{.stab}.
6778 All emit symbols (@pxref{Symbols}), for use by symbolic debuggers.
6779 The symbols are not entered in the @command{@value{AS}} hash table: they
6780 cannot be referenced elsewhere in the source file.
6781 Up to five fields are required:
6785 This is the symbol's name. It may contain any character except
6786 @samp{\000}, so is more general than ordinary symbol names. Some
6787 debuggers used to code arbitrarily complex structures into symbol names
6791 An absolute expression. The symbol's type is set to the low 8 bits of
6792 this expression. Any bit pattern is permitted, but @code{@value{LD}}
6793 and debuggers choke on silly bit patterns.
6796 An absolute expression. The symbol's ``other'' attribute is set to the
6797 low 8 bits of this expression.
6800 An absolute expression. The symbol's descriptor is set to the low 16
6801 bits of this expression.
6804 An absolute expression which becomes the symbol's value.
6807 If a warning is detected while reading a @code{.stabd}, @code{.stabn},
6808 or @code{.stabs} statement, the symbol has probably already been created;
6809 you get a half-formed symbol in your object file. This is
6810 compatible with earlier assemblers!
6813 @cindex @code{stabd} directive
6814 @item .stabd @var{type} , @var{other} , @var{desc}
6816 The ``name'' of the symbol generated is not even an empty string.
6817 It is a null pointer, for compatibility. Older assemblers used a
6818 null pointer so they didn't waste space in object files with empty
6821 The symbol's value is set to the location counter,
6822 relocatably. When your program is linked, the value of this symbol
6823 is the address of the location counter when the @code{.stabd} was
6826 @cindex @code{stabn} directive
6827 @item .stabn @var{type} , @var{other} , @var{desc} , @var{value}
6828 The name of the symbol is set to the empty string @code{""}.
6830 @cindex @code{stabs} directive
6831 @item .stabs @var{string} , @var{type} , @var{other} , @var{desc} , @var{value}
6832 All five fields are specified.
6838 @section @code{.string} "@var{str}", @code{.string8} "@var{str}", @code{.string16}
6839 "@var{str}", @code{.string32} "@var{str}", @code{.string64} "@var{str}"
6841 @cindex string, copying to object file
6842 @cindex string8, copying to object file
6843 @cindex string16, copying to object file
6844 @cindex string32, copying to object file
6845 @cindex string64, copying to object file
6846 @cindex @code{string} directive
6847 @cindex @code{string8} directive
6848 @cindex @code{string16} directive
6849 @cindex @code{string32} directive
6850 @cindex @code{string64} directive
6852 Copy the characters in @var{str} to the object file. You may specify more than
6853 one string to copy, separated by commas. Unless otherwise specified for a
6854 particular machine, the assembler marks the end of each string with a 0 byte.
6855 You can use any of the escape sequences described in @ref{Strings,,Strings}.
6857 The variants @code{string16}, @code{string32} and @code{string64} differ from
6858 the @code{string} pseudo opcode in that each 8-bit character from @var{str} is
6859 copied and expanded to 16, 32 or 64 bits respectively. The expanded characters
6860 are stored in target endianness byte order.
6866 .string "B\0\0\0Y\0\0\0E\0\0\0" /* On little endian targets. */
6867 .string "\0\0\0B\0\0\0Y\0\0\0E" /* On big endian targets. */
6872 @section @code{.struct @var{expression}}
6874 @cindex @code{struct} directive
6875 Switch to the absolute section, and set the section offset to @var{expression},
6876 which must be an absolute expression. You might use this as follows:
6885 This would define the symbol @code{field1} to have the value 0, the symbol
6886 @code{field2} to have the value 4, and the symbol @code{field3} to have the
6887 value 8. Assembly would be left in the absolute section, and you would need to
6888 use a @code{.section} directive of some sort to change to some other section
6889 before further assembly.
6893 @section @code{.subsection @var{name}}
6895 @cindex @code{subsection} directive
6896 @cindex Section Stack
6897 This is one of the ELF section stack manipulation directives. The others are
6898 @code{.section} (@pxref{Section}), @code{.pushsection} (@pxref{PushSection}),
6899 @code{.popsection} (@pxref{PopSection}), and @code{.previous}
6902 This directive replaces the current subsection with @code{name}. The current
6903 section is not changed. The replaced subsection is put onto the section stack
6904 in place of the then current top of stack subsection.
6909 @section @code{.symver}
6910 @cindex @code{symver} directive
6911 @cindex symbol versioning
6912 @cindex versions of symbols
6913 Use the @code{.symver} directive to bind symbols to specific version nodes
6914 within a source file. This is only supported on ELF platforms, and is
6915 typically used when assembling files to be linked into a shared library.
6916 There are cases where it may make sense to use this in objects to be bound
6917 into an application itself so as to override a versioned symbol from a
6920 For ELF targets, the @code{.symver} directive can be used like this:
6922 .symver @var{name}, @var{name2@@nodename}
6924 If the symbol @var{name} is defined within the file
6925 being assembled, the @code{.symver} directive effectively creates a symbol
6926 alias with the name @var{name2@@nodename}, and in fact the main reason that we
6927 just don't try and create a regular alias is that the @var{@@} character isn't
6928 permitted in symbol names. The @var{name2} part of the name is the actual name
6929 of the symbol by which it will be externally referenced. The name @var{name}
6930 itself is merely a name of convenience that is used so that it is possible to
6931 have definitions for multiple versions of a function within a single source
6932 file, and so that the compiler can unambiguously know which version of a
6933 function is being mentioned. The @var{nodename} portion of the alias should be
6934 the name of a node specified in the version script supplied to the linker when
6935 building a shared library. If you are attempting to override a versioned
6936 symbol from a shared library, then @var{nodename} should correspond to the
6937 nodename of the symbol you are trying to override.
6939 If the symbol @var{name} is not defined within the file being assembled, all
6940 references to @var{name} will be changed to @var{name2@@nodename}. If no
6941 reference to @var{name} is made, @var{name2@@nodename} will be removed from the
6944 Another usage of the @code{.symver} directive is:
6946 .symver @var{name}, @var{name2@@@@nodename}
6948 In this case, the symbol @var{name} must exist and be defined within
6949 the file being assembled. It is similar to @var{name2@@nodename}. The
6950 difference is @var{name2@@@@nodename} will also be used to resolve
6951 references to @var{name2} by the linker.
6953 The third usage of the @code{.symver} directive is:
6955 .symver @var{name}, @var{name2@@@@@@nodename}
6957 When @var{name} is not defined within the
6958 file being assembled, it is treated as @var{name2@@nodename}. When
6959 @var{name} is defined within the file being assembled, the symbol
6960 name, @var{name}, will be changed to @var{name2@@@@nodename}.
6965 @section @code{.tag @var{structname}}
6967 @cindex COFF structure debugging
6968 @cindex structure debugging, COFF
6969 @cindex @code{tag} directive
6970 This directive is generated by compilers to include auxiliary debugging
6971 information in the symbol table. It is only permitted inside
6972 @code{.def}/@code{.endef} pairs. Tags are used to link structure
6973 definitions in the symbol table with instances of those structures.
6977 @section @code{.text @var{subsection}}
6979 @cindex @code{text} directive
6980 Tells @command{@value{AS}} to assemble the following statements onto the end of
6981 the text subsection numbered @var{subsection}, which is an absolute
6982 expression. If @var{subsection} is omitted, subsection number zero
6986 @section @code{.title "@var{heading}"}
6988 @cindex @code{title} directive
6989 @cindex listing control: title line
6990 Use @var{heading} as the title (second line, immediately after the
6991 source file name and pagenumber) when generating assembly listings.
6993 This directive affects subsequent pages, as well as the current page if
6994 it appears within ten lines of the top of a page.
6998 @section @code{.type}
7000 This directive is used to set the type of a symbol.
7004 @c only print the extra heading if both COFF and ELF are set
7005 @subheading COFF Version
7008 @cindex COFF symbol type
7009 @cindex symbol type, COFF
7010 @cindex @code{type} directive (COFF version)
7011 For COFF targets, this directive is permitted only within
7012 @code{.def}/@code{.endef} pairs. It is used like this:
7018 This records the integer @var{int} as the type attribute of a symbol table
7025 @c only print the extra heading if both COFF and ELF are set
7026 @subheading ELF Version
7029 @cindex ELF symbol type
7030 @cindex symbol type, ELF
7031 @cindex @code{type} directive (ELF version)
7032 For ELF targets, the @code{.type} directive is used like this:
7035 .type @var{name} , @var{type description}
7038 This sets the type of symbol @var{name} to be either a
7039 function symbol or an object symbol. There are five different syntaxes
7040 supported for the @var{type description} field, in order to provide
7041 compatibility with various other assemblers.
7043 Because some of the characters used in these syntaxes (such as @samp{@@} and
7044 @samp{#}) are comment characters for some architectures, some of the syntaxes
7045 below do not work on all architectures. The first variant will be accepted by
7046 the GNU assembler on all architectures so that variant should be used for
7047 maximum portability, if you do not need to assemble your code with other
7050 The syntaxes supported are:
7053 .type <name> STT_<TYPE_IN_UPPER_CASE>
7054 .type <name>,#<type>
7055 .type <name>,@@<type>
7056 .type <name>,%<type>
7057 .type <name>,"<type>"
7060 The types supported are:
7065 Mark the symbol as being a function name.
7068 @itemx gnu_indirect_function
7069 Mark the symbol as an indirect function when evaluated during reloc
7070 processing. (This is only supported on assemblers targeting GNU systems).
7074 Mark the symbol as being a data object.
7078 Mark the symbol as being a thread-local data object.
7082 Mark the symbol as being a common data object.
7086 Does not mark the symbol in any way. It is supported just for completeness.
7088 @item gnu_unique_object
7089 Marks the symbol as being a globally unique data object. The dynamic linker
7090 will make sure that in the entire process there is just one symbol with this
7091 name and type in use. (This is only supported on assemblers targeting GNU
7096 Note: Some targets support extra types in addition to those listed above.
7102 @section @code{.uleb128 @var{expressions}}
7104 @cindex @code{uleb128} directive
7105 @var{uleb128} stands for ``unsigned little endian base 128.'' This is a
7106 compact, variable length representation of numbers used by the DWARF
7107 symbolic debugging format. @xref{Sleb128, ,@code{.sleb128}}.
7111 @section @code{.val @var{addr}}
7113 @cindex @code{val} directive
7114 @cindex COFF value attribute
7115 @cindex value attribute, COFF
7116 This directive, permitted only within @code{.def}/@code{.endef} pairs,
7117 records the address @var{addr} as the value attribute of a symbol table
7123 @section @code{.version "@var{string}"}
7125 @cindex @code{version} directive
7126 This directive creates a @code{.note} section and places into it an ELF
7127 formatted note of type NT_VERSION. The note's name is set to @code{string}.
7132 @section @code{.vtable_entry @var{table}, @var{offset}}
7134 @cindex @code{vtable_entry} directive
7135 This directive finds or creates a symbol @code{table} and creates a
7136 @code{VTABLE_ENTRY} relocation for it with an addend of @code{offset}.
7139 @section @code{.vtable_inherit @var{child}, @var{parent}}
7141 @cindex @code{vtable_inherit} directive
7142 This directive finds the symbol @code{child} and finds or creates the symbol
7143 @code{parent} and then creates a @code{VTABLE_INHERIT} relocation for the
7144 parent whose addend is the value of the child symbol. As a special case the
7145 parent name of @code{0} is treated as referring to the @code{*ABS*} section.
7149 @section @code{.warning "@var{string}"}
7150 @cindex warning directive
7151 Similar to the directive @code{.error}
7152 (@pxref{Error,,@code{.error "@var{string}"}}), but just emits a warning.
7155 @section @code{.weak @var{names}}
7157 @cindex @code{weak} directive
7158 This directive sets the weak attribute on the comma separated list of symbol
7159 @code{names}. If the symbols do not already exist, they will be created.
7161 On COFF targets other than PE, weak symbols are a GNU extension. This
7162 directive sets the weak attribute on the comma separated list of symbol
7163 @code{names}. If the symbols do not already exist, they will be created.
7165 On the PE target, weak symbols are supported natively as weak aliases.
7166 When a weak symbol is created that is not an alias, GAS creates an
7167 alternate symbol to hold the default value.
7170 @section @code{.weakref @var{alias}, @var{target}}
7172 @cindex @code{weakref} directive
7173 This directive creates an alias to the target symbol that enables the symbol to
7174 be referenced with weak-symbol semantics, but without actually making it weak.
7175 If direct references or definitions of the symbol are present, then the symbol
7176 will not be weak, but if all references to it are through weak references, the
7177 symbol will be marked as weak in the symbol table.
7179 The effect is equivalent to moving all references to the alias to a separate
7180 assembly source file, renaming the alias to the symbol in it, declaring the
7181 symbol as weak there, and running a reloadable link to merge the object files
7182 resulting from the assembly of the new source file and the old source file that
7183 had the references to the alias removed.
7185 The alias itself never makes to the symbol table, and is entirely handled
7186 within the assembler.
7189 @section @code{.word @var{expressions}}
7191 @cindex @code{word} directive
7192 This directive expects zero or more @var{expressions}, of any section,
7193 separated by commas.
7196 For each expression, @command{@value{AS}} emits a 32-bit number.
7199 For each expression, @command{@value{AS}} emits a 16-bit number.
7204 The size of the number emitted, and its byte order,
7205 depend on what target computer the assembly is for.
7208 @c on sparc the "special treatment to support compilers" doesn't
7209 @c happen---32-bit addressability, period; no long/short jumps.
7210 @ifset DIFF-TBL-KLUGE
7211 @cindex difference tables altered
7212 @cindex altered difference tables
7214 @emph{Warning: Special Treatment to support Compilers}
7218 Machines with a 32-bit address space, but that do less than 32-bit
7219 addressing, require the following special treatment. If the machine of
7220 interest to you does 32-bit addressing (or doesn't require it;
7221 @pxref{Machine Dependencies}), you can ignore this issue.
7224 In order to assemble compiler output into something that works,
7225 @command{@value{AS}} occasionally does strange things to @samp{.word} directives.
7226 Directives of the form @samp{.word sym1-sym2} are often emitted by
7227 compilers as part of jump tables. Therefore, when @command{@value{AS}} assembles a
7228 directive of the form @samp{.word sym1-sym2}, and the difference between
7229 @code{sym1} and @code{sym2} does not fit in 16 bits, @command{@value{AS}}
7230 creates a @dfn{secondary jump table}, immediately before the next label.
7231 This secondary jump table is preceded by a short-jump to the
7232 first byte after the secondary table. This short-jump prevents the flow
7233 of control from accidentally falling into the new table. Inside the
7234 table is a long-jump to @code{sym2}. The original @samp{.word}
7235 contains @code{sym1} minus the address of the long-jump to
7238 If there were several occurrences of @samp{.word sym1-sym2} before the
7239 secondary jump table, all of them are adjusted. If there was a
7240 @samp{.word sym3-sym4}, that also did not fit in sixteen bits, a
7241 long-jump to @code{sym4} is included in the secondary jump table,
7242 and the @code{.word} directives are adjusted to contain @code{sym3}
7243 minus the address of the long-jump to @code{sym4}; and so on, for as many
7244 entries in the original jump table as necessary.
7247 @emph{This feature may be disabled by compiling @command{@value{AS}} with the
7248 @samp{-DWORKING_DOT_WORD} option.} This feature is likely to confuse
7249 assembly language programmers.
7252 @c end DIFF-TBL-KLUGE
7254 @ifclear no-space-dir
7256 @section @code{.zero @var{size}}
7258 @cindex @code{zero} directive
7259 @cindex filling memory with zero bytes
7260 This directive emits @var{size} 0-valued bytes. @var{size} must be an absolute
7261 expression. This directive is actually an alias for the @samp{.skip} directive
7262 so in can take an optional second argument of the value to store in the bytes
7263 instead of zero. Using @samp{.zero} in this way would be confusing however.
7268 @section @code{.2byte @var{expression} [, @var{expression}]*}
7269 @cindex @code{2byte} directive
7270 @cindex two-byte integer
7271 @cindex integer, 2-byte
7273 This directive expects zero or more expressions, separated by commas. If there
7274 are no expressions then the directive does nothing. Otherwise each expression
7275 is evaluated in turn and placed in the next two bytes of the current output
7276 section, using the endian model of the target. If an expression will not fit
7277 in two bytes, a warning message is displayed and the least significant two
7278 bytes of the expression's value are used. If an expression cannot be evaluated
7279 at assembly time then relocations will be generated in order to compute the
7282 This directive does not apply any alignment before or after inserting the
7283 values. As a result of this, if relocations are generated, they may be
7284 different from those used for inserting values with a guaranteed alignment.
7286 This directive is only available for ELF targets,
7289 @section @code{.4byte @var{expression} [, @var{expression}]*}
7290 @cindex @code{4byte} directive
7291 @cindex four-byte integer
7292 @cindex integer, 4-byte
7294 Like the @option{.2byte} directive, except that it inserts unaligned, four byte
7295 long values into the output.
7298 @section @code{.8byte @var{expression} [, @var{expression}]*}
7299 @cindex @code{8byte} directive
7300 @cindex eight-byte integer
7301 @cindex integer, 8-byte
7303 Like the @option{.2byte} directive, except that it inserts unaligned, eight
7304 byte long bignum values into the output.
7309 @section Deprecated Directives
7311 @cindex deprecated directives
7312 @cindex obsolescent directives
7313 One day these directives won't work.
7314 They are included for compatibility with older assemblers.
7321 @node Object Attributes
7322 @chapter Object Attributes
7323 @cindex object attributes
7325 @command{@value{AS}} assembles source files written for a specific architecture
7326 into object files for that architecture. But not all object files are alike.
7327 Many architectures support incompatible variations. For instance, floating
7328 point arguments might be passed in floating point registers if the object file
7329 requires hardware floating point support---or floating point arguments might be
7330 passed in integer registers if the object file supports processors with no
7331 hardware floating point unit. Or, if two objects are built for different
7332 generations of the same architecture, the combination may require the
7333 newer generation at run-time.
7335 This information is useful during and after linking. At link time,
7336 @command{@value{LD}} can warn about incompatible object files. After link
7337 time, tools like @command{gdb} can use it to process the linked file
7340 Compatibility information is recorded as a series of object attributes. Each
7341 attribute has a @dfn{vendor}, @dfn{tag}, and @dfn{value}. The vendor is a
7342 string, and indicates who sets the meaning of the tag. The tag is an integer,
7343 and indicates what property the attribute describes. The value may be a string
7344 or an integer, and indicates how the property affects this object. Missing
7345 attributes are the same as attributes with a zero value or empty string value.
7347 Object attributes were developed as part of the ABI for the ARM Architecture.
7348 The file format is documented in @cite{ELF for the ARM Architecture}.
7351 * GNU Object Attributes:: @sc{gnu} Object Attributes
7352 * Defining New Object Attributes:: Defining New Object Attributes
7355 @node GNU Object Attributes
7356 @section @sc{gnu} Object Attributes
7358 The @code{.gnu_attribute} directive records an object attribute
7359 with vendor @samp{gnu}.
7361 Except for @samp{Tag_compatibility}, which has both an integer and a string for
7362 its value, @sc{gnu} attributes have a string value if the tag number is odd and
7363 an integer value if the tag number is even. The second bit (@code{@var{tag} &
7364 2} is set for architecture-independent attributes and clear for
7365 architecture-dependent ones.
7367 @subsection Common @sc{gnu} attributes
7369 These attributes are valid on all architectures.
7372 @item Tag_compatibility (32)
7373 The compatibility attribute takes an integer flag value and a vendor name. If
7374 the flag value is 0, the file is compatible with other toolchains. If it is 1,
7375 then the file is only compatible with the named toolchain. If it is greater
7376 than 1, the file can only be processed by other toolchains under some private
7377 arrangement indicated by the flag value and the vendor name.
7380 @subsection MIPS Attributes
7383 @item Tag_GNU_MIPS_ABI_FP (4)
7384 The floating-point ABI used by this object file. The value will be:
7388 0 for files not affected by the floating-point ABI.
7390 1 for files using the hardware floating-point ABI with a standard
7391 double-precision FPU.
7393 2 for files using the hardware floating-point ABI with a single-precision FPU.
7395 3 for files using the software floating-point ABI.
7397 4 for files using the deprecated hardware floating-point ABI which used 64-bit
7398 floating-point registers, 32-bit general-purpose registers and increased the
7399 number of callee-saved floating-point registers.
7401 5 for files using the hardware floating-point ABI with a double-precision FPU
7402 with either 32-bit or 64-bit floating-point registers and 32-bit
7403 general-purpose registers.
7405 6 for files using the hardware floating-point ABI with 64-bit floating-point
7406 registers and 32-bit general-purpose registers.
7408 7 for files using the hardware floating-point ABI with 64-bit floating-point
7409 registers, 32-bit general-purpose registers and a rule that forbids the
7410 direct use of odd-numbered single-precision floating-point registers.
7414 @subsection PowerPC Attributes
7417 @item Tag_GNU_Power_ABI_FP (4)
7418 The floating-point ABI used by this object file. The value will be:
7422 0 for files not affected by the floating-point ABI.
7424 1 for files using double-precision hardware floating-point ABI.
7426 2 for files using the software floating-point ABI.
7428 3 for files using single-precision hardware floating-point ABI.
7431 @item Tag_GNU_Power_ABI_Vector (8)
7432 The vector ABI used by this object file. The value will be:
7436 0 for files not affected by the vector ABI.
7438 1 for files using general purpose registers to pass vectors.
7440 2 for files using AltiVec registers to pass vectors.
7442 3 for files using SPE registers to pass vectors.
7446 @subsection IBM z Systems Attributes
7449 @item Tag_GNU_S390_ABI_Vector (8)
7450 The vector ABI used by this object file. The value will be:
7454 0 for files not affected by the vector ABI.
7456 1 for files using software vector ABI.
7458 2 for files using hardware vector ABI.
7462 @node Defining New Object Attributes
7463 @section Defining New Object Attributes
7465 If you want to define a new @sc{gnu} object attribute, here are the places you
7466 will need to modify. New attributes should be discussed on the @samp{binutils}
7471 This manual, which is the official register of attributes.
7473 The header for your architecture @file{include/elf}, to define the tag.
7475 The @file{bfd} support file for your architecture, to merge the attribute
7476 and issue any appropriate link warnings.
7478 Test cases in @file{ld/testsuite} for merging and link warnings.
7480 @file{binutils/readelf.c} to display your attribute.
7482 GCC, if you want the compiler to mark the attribute automatically.
7488 @node Machine Dependencies
7489 @chapter Machine Dependent Features
7491 @cindex machine dependencies
7492 The machine instruction sets are (almost by definition) different on
7493 each machine where @command{@value{AS}} runs. Floating point representations
7494 vary as well, and @command{@value{AS}} often supports a few additional
7495 directives or command-line options for compatibility with other
7496 assemblers on a particular platform. Finally, some versions of
7497 @command{@value{AS}} support special pseudo-instructions for branch
7500 This chapter discusses most of these differences, though it does not
7501 include details on any machine's instruction set. For details on that
7502 subject, see the hardware manufacturer's manual.
7506 * AArch64-Dependent:: AArch64 Dependent Features
7509 * Alpha-Dependent:: Alpha Dependent Features
7512 * ARC-Dependent:: ARC Dependent Features
7515 * ARM-Dependent:: ARM Dependent Features
7518 * AVR-Dependent:: AVR Dependent Features
7521 * Blackfin-Dependent:: Blackfin Dependent Features
7524 * CR16-Dependent:: CR16 Dependent Features
7527 * CRIS-Dependent:: CRIS Dependent Features
7530 * D10V-Dependent:: D10V Dependent Features
7533 * D30V-Dependent:: D30V Dependent Features
7536 * Epiphany-Dependent:: EPIPHANY Dependent Features
7539 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7542 * HPPA-Dependent:: HPPA Dependent Features
7545 * i386-Dependent:: Intel 80386 and AMD x86-64 Dependent Features
7548 * IA-64-Dependent:: Intel IA-64 Dependent Features
7551 * IP2K-Dependent:: IP2K Dependent Features
7554 * LM32-Dependent:: LM32 Dependent Features
7557 * M32C-Dependent:: M32C Dependent Features
7560 * M32R-Dependent:: M32R Dependent Features
7563 * M68K-Dependent:: M680x0 Dependent Features
7566 * M68HC11-Dependent:: M68HC11 and 68HC12 Dependent Features
7569 * Meta-Dependent :: Meta Dependent Features
7572 * MicroBlaze-Dependent:: MICROBLAZE Dependent Features
7575 * MIPS-Dependent:: MIPS Dependent Features
7578 * MMIX-Dependent:: MMIX Dependent Features
7581 * MSP430-Dependent:: MSP430 Dependent Features
7584 * NDS32-Dependent:: Andes NDS32 Dependent Features
7587 * NiosII-Dependent:: Altera Nios II Dependent Features
7590 * NS32K-Dependent:: NS32K Dependent Features
7593 * PDP-11-Dependent:: PDP-11 Dependent Features
7596 * PJ-Dependent:: picoJava Dependent Features
7599 * PPC-Dependent:: PowerPC Dependent Features
7602 * PRU-Dependent:: PRU Dependent Features
7605 * RISC-V-Dependent:: RISC-V Dependent Features
7608 * RL78-Dependent:: RL78 Dependent Features
7611 * RX-Dependent:: RX Dependent Features
7614 * S/390-Dependent:: IBM S/390 Dependent Features
7617 * SCORE-Dependent:: SCORE Dependent Features
7620 * SH-Dependent:: Renesas / SuperH SH Dependent Features
7623 * Sparc-Dependent:: SPARC Dependent Features
7626 * TIC54X-Dependent:: TI TMS320C54x Dependent Features
7629 * TIC6X-Dependent :: TI TMS320C6x Dependent Features
7632 * TILE-Gx-Dependent :: Tilera TILE-Gx Dependent Features
7635 * TILEPro-Dependent :: Tilera TILEPro Dependent Features
7638 * V850-Dependent:: V850 Dependent Features
7641 * Vax-Dependent:: VAX Dependent Features
7644 * Visium-Dependent:: Visium Dependent Features
7647 * WebAssembly-Dependent:: WebAssembly Dependent Features
7650 * XGATE-Dependent:: XGATE Dependent Features
7653 * XSTORMY16-Dependent:: XStormy16 Dependent Features
7656 * Xtensa-Dependent:: Xtensa Dependent Features
7659 * Z80-Dependent:: Z80 Dependent Features
7662 * Z8000-Dependent:: Z8000 Dependent Features
7669 @c The following major nodes are *sections* in the GENERIC version, *chapters*
7670 @c in single-cpu versions. This is mainly achieved by @lowersections. There is a
7671 @c peculiarity: to preserve cross-references, there must be a node called
7672 @c "Machine Dependencies". Hence the conditional nodenames in each
7673 @c major node below. Node defaulting in makeinfo requires adjacency of
7674 @c node and sectioning commands; hence the repetition of @chapter BLAH
7675 @c in both conditional blocks.
7678 @include c-aarch64.texi
7682 @include c-alpha.texi
7698 @include c-bfin.texi
7702 @include c-cr16.texi
7706 @include c-cris.texi
7711 @node Machine Dependencies
7712 @chapter Machine Dependent Features
7714 The machine instruction sets are different on each Renesas chip family,
7715 and there are also some syntax differences among the families. This
7716 chapter describes the specific @command{@value{AS}} features for each
7720 * H8/300-Dependent:: Renesas H8/300 Dependent Features
7721 * SH-Dependent:: Renesas SH Dependent Features
7728 @include c-d10v.texi
7732 @include c-d30v.texi
7736 @include c-epiphany.texi
7740 @include c-h8300.texi
7744 @include c-hppa.texi
7748 @include c-i386.texi
7752 @include c-ia64.texi
7756 @include c-ip2k.texi
7760 @include c-lm32.texi
7764 @include c-m32c.texi
7768 @include c-m32r.texi
7772 @include c-m68k.texi
7776 @include c-m68hc11.texi
7780 @include c-metag.texi
7784 @include c-microblaze.texi
7788 @include c-mips.texi
7792 @include c-mmix.texi
7796 @include c-msp430.texi
7800 @include c-nds32.texi
7804 @include c-nios2.texi
7808 @include c-ns32k.texi
7812 @include c-pdp11.texi
7828 @include c-riscv.texi
7832 @include c-rl78.texi
7840 @include c-s390.texi
7844 @include c-score.texi
7852 @include c-sparc.texi
7856 @include c-tic54x.texi
7860 @include c-tic6x.texi
7864 @include c-tilegx.texi
7868 @include c-tilepro.texi
7872 @include c-v850.texi
7880 @include c-visium.texi
7884 @include c-wasm32.texi
7888 @include c-xgate.texi
7892 @include c-xstormy16.texi
7896 @include c-xtensa.texi
7908 @c reverse effect of @down at top of generic Machine-Dep chapter
7912 @node Reporting Bugs
7913 @chapter Reporting Bugs
7914 @cindex bugs in assembler
7915 @cindex reporting bugs in assembler
7917 Your bug reports play an essential role in making @command{@value{AS}} reliable.
7919 Reporting a bug may help you by bringing a solution to your problem, or it may
7920 not. But in any case the principal function of a bug report is to help the
7921 entire community by making the next version of @command{@value{AS}} work better.
7922 Bug reports are your contribution to the maintenance of @command{@value{AS}}.
7924 In order for a bug report to serve its purpose, you must include the
7925 information that enables us to fix the bug.
7928 * Bug Criteria:: Have you found a bug?
7929 * Bug Reporting:: How to report bugs
7933 @section Have You Found a Bug?
7934 @cindex bug criteria
7936 If you are not sure whether you have found a bug, here are some guidelines:
7939 @cindex fatal signal
7940 @cindex assembler crash
7941 @cindex crash of assembler
7943 If the assembler gets a fatal signal, for any input whatever, that is a
7944 @command{@value{AS}} bug. Reliable assemblers never crash.
7946 @cindex error on valid input
7948 If @command{@value{AS}} produces an error message for valid input, that is a bug.
7950 @cindex invalid input
7952 If @command{@value{AS}} does not produce an error message for invalid input, that
7953 is a bug. However, you should note that your idea of ``invalid input'' might
7954 be our idea of ``an extension'' or ``support for traditional practice''.
7957 If you are an experienced user of assemblers, your suggestions for improvement
7958 of @command{@value{AS}} are welcome in any case.
7962 @section How to Report Bugs
7964 @cindex assembler bugs, reporting
7966 A number of companies and individuals offer support for @sc{gnu} products. If
7967 you obtained @command{@value{AS}} from a support organization, we recommend you
7968 contact that organization first.
7970 You can find contact information for many support companies and
7971 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
7975 In any event, we also recommend that you send bug reports for @command{@value{AS}}
7979 The fundamental principle of reporting bugs usefully is this:
7980 @strong{report all the facts}. If you are not sure whether to state a
7981 fact or leave it out, state it!
7983 Often people omit facts because they think they know what causes the problem
7984 and assume that some details do not matter. Thus, you might assume that the
7985 name of a symbol you use in an example does not matter. Well, probably it does
7986 not, but one cannot be sure. Perhaps the bug is a stray memory reference which
7987 happens to fetch from the location where that name is stored in memory;
7988 perhaps, if the name were different, the contents of that location would fool
7989 the assembler into doing the right thing despite the bug. Play it safe and
7990 give a specific, complete example. That is the easiest thing for you to do,
7991 and the most helpful.
7993 Keep in mind that the purpose of a bug report is to enable us to fix the bug if
7994 it is new to us. Therefore, always write your bug reports on the assumption
7995 that the bug has not been reported previously.
7997 Sometimes people give a few sketchy facts and ask, ``Does this ring a
7998 bell?'' This cannot help us fix a bug, so it is basically useless. We
7999 respond by asking for enough details to enable us to investigate.
8000 You might as well expedite matters by sending them to begin with.
8002 To enable us to fix the bug, you should include all these things:
8006 The version of @command{@value{AS}}. @command{@value{AS}} announces it if you start
8007 it with the @samp{--version} argument.
8009 Without this, we will not know whether there is any point in looking for
8010 the bug in the current version of @command{@value{AS}}.
8013 Any patches you may have applied to the @command{@value{AS}} source.
8016 The type of machine you are using, and the operating system name and
8020 What compiler (and its version) was used to compile @command{@value{AS}}---e.g.
8024 The command arguments you gave the assembler to assemble your example and
8025 observe the bug. To guarantee you will not omit something important, list them
8026 all. A copy of the Makefile (or the output from make) is sufficient.
8028 If we were to try to guess the arguments, we would probably guess wrong
8029 and then we might not encounter the bug.
8032 A complete input file that will reproduce the bug. If the bug is observed when
8033 the assembler is invoked via a compiler, send the assembler source, not the
8034 high level language source. Most compilers will produce the assembler source
8035 when run with the @samp{-S} option. If you are using @code{@value{GCC}}, use
8036 the options @samp{-v --save-temps}; this will save the assembler source in a
8037 file with an extension of @file{.s}, and also show you exactly how
8038 @command{@value{AS}} is being run.
8041 A description of what behavior you observe that you believe is
8042 incorrect. For example, ``It gets a fatal signal.''
8044 Of course, if the bug is that @command{@value{AS}} gets a fatal signal, then we
8045 will certainly notice it. But if the bug is incorrect output, we might not
8046 notice unless it is glaringly wrong. You might as well not give us a chance to
8049 Even if the problem you experience is a fatal signal, you should still say so
8050 explicitly. Suppose something strange is going on, such as, your copy of
8051 @command{@value{AS}} is out of sync, or you have encountered a bug in the C
8052 library on your system. (This has happened!) Your copy might crash and ours
8053 would not. If you told us to expect a crash, then when ours fails to crash, we
8054 would know that the bug was not happening for us. If you had not told us to
8055 expect a crash, then we would not be able to draw any conclusion from our
8059 If you wish to suggest changes to the @command{@value{AS}} source, send us context
8060 diffs, as generated by @code{diff} with the @samp{-u}, @samp{-c}, or @samp{-p}
8061 option. Always send diffs from the old file to the new file. If you even
8062 discuss something in the @command{@value{AS}} source, refer to it by context, not
8065 The line numbers in our development sources will not match those in your
8066 sources. Your line numbers would convey no useful information to us.
8069 Here are some things that are not necessary:
8073 A description of the envelope of the bug.
8075 Often people who encounter a bug spend a lot of time investigating
8076 which changes to the input file will make the bug go away and which
8077 changes will not affect it.
8079 This is often time consuming and not very useful, because the way we
8080 will find the bug is by running a single example under the debugger
8081 with breakpoints, not by pure deduction from a series of examples.
8082 We recommend that you save your time for something else.
8084 Of course, if you can find a simpler example to report @emph{instead}
8085 of the original one, that is a convenience for us. Errors in the
8086 output will be easier to spot, running under the debugger will take
8087 less time, and so on.
8089 However, simplification is not vital; if you do not want to do this,
8090 report the bug anyway and send us the entire test case you used.
8093 A patch for the bug.
8095 A patch for the bug does help us if it is a good one. But do not omit
8096 the necessary information, such as the test case, on the assumption that
8097 a patch is all we need. We might see problems with your patch and decide
8098 to fix the problem another way, or we might not understand it at all.
8100 Sometimes with a program as complicated as @command{@value{AS}} it is very hard to
8101 construct an example that will make the program follow a certain path through
8102 the code. If you do not send us the example, we will not be able to construct
8103 one, so we will not be able to verify that the bug is fixed.
8105 And if we cannot understand what bug you are trying to fix, or why your
8106 patch should be an improvement, we will not install it. A test case will
8107 help us to understand.
8110 A guess about what the bug is or what it depends on.
8112 Such guesses are usually wrong. Even we cannot guess right about such
8113 things without first using the debugger to find the facts.
8116 @node Acknowledgements
8117 @chapter Acknowledgements
8119 If you have contributed to GAS and your name isn't listed here,
8120 it is not meant as a slight. We just don't know about it. Send mail to the
8121 maintainer, and we'll correct the situation. Currently
8123 the maintainer is Nick Clifton (email address @code{nickc@@redhat.com}).
8125 Dean Elsner wrote the original @sc{gnu} assembler for the VAX.@footnote{Any
8128 Jay Fenlason maintained GAS for a while, adding support for GDB-specific debug
8129 information and the 68k series machines, most of the preprocessing pass, and
8130 extensive changes in @file{messages.c}, @file{input-file.c}, @file{write.c}.
8132 K. Richard Pixley maintained GAS for a while, adding various enhancements and
8133 many bug fixes, including merging support for several processors, breaking GAS
8134 up to handle multiple object file format back ends (including heavy rewrite,
8135 testing, an integration of the coff and b.out back ends), adding configuration
8136 including heavy testing and verification of cross assemblers and file splits
8137 and renaming, converted GAS to strictly ANSI C including full prototypes, added
8138 support for m680[34]0 and cpu32, did considerable work on i960 including a COFF
8139 port (including considerable amounts of reverse engineering), a SPARC opcode
8140 file rewrite, DECstation, rs6000, and hp300hpux host ports, updated ``know''
8141 assertions and made them work, much other reorganization, cleanup, and lint.
8143 Ken Raeburn wrote the high-level BFD interface code to replace most of the code
8144 in format-specific I/O modules.
8146 The original VMS support was contributed by David L. Kashtan. Eric Youngdale
8147 has done much work with it since.
8149 The Intel 80386 machine description was written by Eliot Dresselhaus.
8151 Minh Tran-Le at IntelliCorp contributed some AIX 386 support.
8153 The Motorola 88k machine description was contributed by Devon Bowen of Buffalo
8154 University and Torbjorn Granlund of the Swedish Institute of Computer Science.
8156 Keith Knowles at the Open Software Foundation wrote the original MIPS back end
8157 (@file{tc-mips.c}, @file{tc-mips.h}), and contributed Rose format support
8158 (which hasn't been merged in yet). Ralph Campbell worked with the MIPS code to
8159 support a.out format.
8161 Support for the Zilog Z8k and Renesas H8/300 processors (tc-z8k,
8162 tc-h8300), and IEEE 695 object file format (obj-ieee), was written by
8163 Steve Chamberlain of Cygnus Support. Steve also modified the COFF back end to
8164 use BFD for some low-level operations, for use with the H8/300 and AMD 29k
8167 John Gilmore built the AMD 29000 support, added @code{.include} support, and
8168 simplified the configuration of which versions accept which directives. He
8169 updated the 68k machine description so that Motorola's opcodes always produced
8170 fixed-size instructions (e.g., @code{jsr}), while synthetic instructions
8171 remained shrinkable (@code{jbsr}). John fixed many bugs, including true tested
8172 cross-compilation support, and one bug in relaxation that took a week and
8173 required the proverbial one-bit fix.
8175 Ian Lance Taylor of Cygnus Support merged the Motorola and MIT syntax for the
8176 68k, completed support for some COFF targets (68k, i386 SVR3, and SCO Unix),
8177 added support for MIPS ECOFF and ELF targets, wrote the initial RS/6000 and
8178 PowerPC assembler, and made a few other minor patches.
8180 Steve Chamberlain made GAS able to generate listings.
8182 Hewlett-Packard contributed support for the HP9000/300.
8184 Jeff Law wrote GAS and BFD support for the native HPPA object format (SOM)
8185 along with a fairly extensive HPPA testsuite (for both SOM and ELF object
8186 formats). This work was supported by both the Center for Software Science at
8187 the University of Utah and Cygnus Support.
8189 Support for ELF format files has been worked on by Mark Eichin of Cygnus
8190 Support (original, incomplete implementation for SPARC), Pete Hoogenboom and
8191 Jeff Law at the University of Utah (HPPA mainly), Michael Meissner of the Open
8192 Software Foundation (i386 mainly), and Ken Raeburn of Cygnus Support (sparc,
8193 and some initial 64-bit support).
8195 Linas Vepstas added GAS support for the ESA/390 ``IBM 370'' architecture.
8197 Richard Henderson rewrote the Alpha assembler. Klaus Kaempf wrote GAS and BFD
8198 support for openVMS/Alpha.
8200 Timothy Wall, Michael Hayes, and Greg Smart contributed to the various tic*
8203 David Heine, Sterling Augustine, Bob Wilson and John Ruttenberg from Tensilica,
8204 Inc.@: added support for Xtensa processors.
8206 Several engineers at Cygnus Support have also provided many small bug fixes and
8207 configuration enhancements.
8209 Jon Beniston added support for the Lattice Mico32 architecture.
8211 Many others have contributed large or small bugfixes and enhancements. If
8212 you have contributed significant work and are not mentioned on this list, and
8213 want to be, let us know. Some of the history has been lost; we are not
8214 intentionally leaving anyone out.
8216 @node GNU Free Documentation License
8217 @appendix GNU Free Documentation License
8221 @unnumbered AS Index
projects / deliverable / binutils-gdb.git / blob